Abstract ID: 190
Seasonal prediction of regional Arctic sea ice using the high-resolution climate prediction system CMA-CPSv3
Lead Author: Min Chu
CMA Earth System Modeling and Prediction Centre (CEMC), China
Keywords: climate prediction system, Arctic sea ice, seasonal prediction, sea ice model, sea ice data assimilation
Abstract: The prediction of Arctic sea ice is an extremely challenging problem. Firstly, it requires a better climate system model; secondly, a sea ice data assimilation system is also needed. CMA-CPS is a climate prediction system developed by China Meteorological Administration, in which a climate system model and a data assimilation system are included.
BCC-CSM is a fully coupled climate system model developed by Beijing Climate Center. Based on the historical experiments of CMIP5 and CMIP6, two versions of BCC-CSM are comprehensively evaluated. Results show that CMIP6 historical experiment has a great improvement in both the seasonal cycle of Arctic sea ice extent and spacial distribution of sea ice thickness. Also, a sea ice data assimilation system is established based on the Optimal Interpolation (OI) method, and the satellite-derived sea ice concentration and thickness are assimilated. Experiments show that the average error of the unassimilated ice concentration and thickness has decreased overt 30% while assimilating meanwhile sea ice concentration and thickness. Sea ice data assimilation system has laid a good foundation for the near-term and sub-seasonal to seasonal prediction of Arctic sea ice.
Based on the high-resolution climate prediction system CMA-CPSv3, a series of seasonal prediction experiments had been conducted. With those experiments, this work presents a comprehensive assessment of regional Arctic SIE and Arctic climate prediction skill and analyzes the main factors affecting prediction skills. Compared with BCC-CPSv2, BCC-CPSv3 has a certain improvement in the seasonal prediction skill of sea ice in various regions of the Arctic. Especially for the pan-Arctic region, BCC-CPSv3 has the capability to predict the sea ice evolution in September. As to the physical mechanisms underlying the regional skill, the analysis shows that BCC-CPSv3 has significantly improved the winter sea ice forecast in the marginal seas close to the Atlantic Ocean mainly due to the model’s performance of the SST in the North Atlantic Ocean. And similarly, the improvement of summer sea ice forecasting ability in sea areas close to the Arctic center is related to the improvement of summer sea ice thickness forecasting ability in these areas.
Co-authors:
Panxi Dai (Zhejiang University, China)
Xiangwen Liu （CMA Earth System Modeling and Prediction Centre (CEMC), China）
Tongwen Wu（CMA Earth System Modeling and Prediction Centre (CEMC), China）
Yixiong Lu（CMA Earth System Modeling and Prediction Centre (CEMC), China)
