日本地球惑星科学連合2025年大会

講演情報

[E] 口頭発表

セッション記号 M (領域外・複数領域) » M-GI 地球科学一般・情報地球科学

[M-GI26] Data assimilation: A fundamental approach in geosciences

2025年5月30日(金) 09:00 〜 10:30 展示場特設会場 (6) (幕張メッセ国際展示場 7・8ホール)

コンビーナ:中野 慎也(情報・システム研究機構 統計数理研究所)、堀田 大介(気象研究所)、大石 俊(理化学研究所 計算科学研究センター)、加納 将行(東北大学理学研究科)、座長:中野 慎也(情報・システム研究機構 統計数理研究所)、近藤 圭一(気象庁気象研究所)

10:15 〜 10:30

[MGI26-06] TOPAZ5: Upgraded the Arctic coupled ocean and sea-ice forecasting system with Ensemble Kalman Filter

★Invited Papers

*Jiping Xie1,2、Achref Othmani Othmani1,2、Alfatih Ali3、Annette Samuelsen1,2、Laurent Bertino1,2 (1.Nansen Environmental and Remote Sensing Center, Bergen 5007, Norway、2.Bjerknes Centre for Climate Research, Bergen 5007, Norway、3.Division of Ocean and Ice, Norwegian Meteorological Institute, Bergen 5007, Norway)

キーワード:CICE, HYCOM, EnKF, Ice Thickness Distribution

With the increased changes in the Arctic Ocean, more human activities and industries require a high-resolution forecast of the ocean and sea ice, and more relevant variables are expected as a knowledge base for sustainable management of resources. The upgraded version of the coupled Arctic ocean and sea-ice data assimilation system, TOPAZ5, uses version 2.2.98 of the Hybrid Coordinate Ocean Model, which is coupled with the sea ice model (CICE, the Los Alamos National Laboratory Community Ice CodE). To enhance the eddy dynamic changes in the Pan-Arctic region, the model grids have been refined both horizontally to ~6 km and vertically from 28 to 50 hybrid layers. Instead of climatological boundary conditions, TOPAZ5 uses the nesting boundary from the global NEMO model system by the Copernicus Marine Service. The sea ice model has been upgraded to CICE v5.1, with five categories of sea ice thickness and complex thermodynamics. The TOPAZ5 system has been tested under two assimilation experiments in 2021, using an Ensemble Kalman filter with 100 model members, whose ensemble runs are driven by randomly perturbed atmospheric forcing centered on the ECMWF reanalysis. The results show that multiple types of ocean physical and sea ice observations can be successfully assimilated, and prevent the model forecast bias from increasing. The forecasting based on this system improves upon the previous TOPAZ4 version of the system, and the relevant products are released daily on the Copernicus Marine Services. We will discuss different avenues for the further development of this system.