14:30 〜 14:45
[MGI29-04] LETKF-based Ocean Research Analysis (LORA) in the Western North Pacific region
キーワード：アンサンブルカルマンフィルタ、海洋データ同化、LORA、西部北太平洋
Various atmosphere and ocean analysis datasets have been produced by operational centers and research institutions, and are used for a variety of geoscience studies. In the Pacific region, the following high-resolution ocean analysis datasets are now available: Japan Coastal Ocean Predictability Experiment 2M (JCOPE2M; Miyazawa et al. 2017) implemented with multi-scale three-dimensional variational assimilation (Li et al. 2015), Four-dimensional variational Ocean Re-analysis for the Western North Pacific over 30 years (FORA-WNP30; Usui et al. 2017) with four-dimensional variational assimilation, and Data assimilation Research of the East Asian Marine System (DREAMS; Hirose et al. 2013) with Kalman filter. To the best of the authors’ knowledge, no ensemble Kalman filter (EnKF)-based high-resolution analysis datasets exist yet for the Pacific region. The TOPAZ4 (Sakov et al. 2012; Xie et al. 2017) is created by an EnKF-based sea-ice coupled system constructed for the North Atlantic and Arctic regions but is released only around the Arctic region on the website of Copernicus Marine Environment Monitoring Service (CMEMS; https://marine.copernicus.eu/).
We have developed an eddy-permitting EnKF-based ocean data assimilation system in the western North Pacific (WNP) region (Ohishi et al. in prep.), in which satellite and in-situ observations are assimilated at a shorter interval of 1 day than 5 days for the TOPAZ4. With the recent enhancement of the high-performance computing resources, we are now able to integrate a sufficiently high-resolution system to resolve fronts and eddies. In the WNP region, the abundance of fronts and eddies exist in association with strong western boundary currents and is a key factor for mid-latitude air-sea interaction (e.g., Minobe et al. 2008; Kida et al. 2015). Therefore, this study aims at creating a high-resolution EnKF-based ocean analysis dataset in the WNP region and comparing the accuracy with the existing analysis and observational datasets, respectively, JCOPE2M and Archiving, Validation, and Interpretation of Satellite Oceanographic Data (AVISO; Ducet et al. 2000). The dataset constructed in this study is referred to as LORA: Local ensemble transform Kalman filter (LETKF)-based Ocean Research Analysis.
The validation results show that the accuracy of surface horizontal velocity in the LORA is lower in the subtropical region but higher in the mid-latitude region especially along the Kuroshio Extension (KE) compared with the JCOPE2M, although the accuracy in the AVISO is the best. The accuracy of sea surface temperatures in the LORA is better than the JCOPE2M over the almost domain. However, the validation results using an independent buoy south of the KE indicate that higher temperature accuracy in the LORA than the JCOPE2M might be limited to a vertical extent shallower than 300 m likely because of the prescribed vertical localization scale of 100 m. In summary, the LORA has sufficient accuracy to utilize for a variety of geoscience research as well as fisheries, marine transports, and marine environment consultants.
We have developed an eddy-permitting EnKF-based ocean data assimilation system in the western North Pacific (WNP) region (Ohishi et al. in prep.), in which satellite and in-situ observations are assimilated at a shorter interval of 1 day than 5 days for the TOPAZ4. With the recent enhancement of the high-performance computing resources, we are now able to integrate a sufficiently high-resolution system to resolve fronts and eddies. In the WNP region, the abundance of fronts and eddies exist in association with strong western boundary currents and is a key factor for mid-latitude air-sea interaction (e.g., Minobe et al. 2008; Kida et al. 2015). Therefore, this study aims at creating a high-resolution EnKF-based ocean analysis dataset in the WNP region and comparing the accuracy with the existing analysis and observational datasets, respectively, JCOPE2M and Archiving, Validation, and Interpretation of Satellite Oceanographic Data (AVISO; Ducet et al. 2000). The dataset constructed in this study is referred to as LORA: Local ensemble transform Kalman filter (LETKF)-based Ocean Research Analysis.
The validation results show that the accuracy of surface horizontal velocity in the LORA is lower in the subtropical region but higher in the mid-latitude region especially along the Kuroshio Extension (KE) compared with the JCOPE2M, although the accuracy in the AVISO is the best. The accuracy of sea surface temperatures in the LORA is better than the JCOPE2M over the almost domain. However, the validation results using an independent buoy south of the KE indicate that higher temperature accuracy in the LORA than the JCOPE2M might be limited to a vertical extent shallower than 300 m likely because of the prescribed vertical localization scale of 100 m. In summary, the LORA has sufficient accuracy to utilize for a variety of geoscience research as well as fisheries, marine transports, and marine environment consultants.