Study on improving global atmospheric state prediction through the expansion of observational information

2023 Fiscal Year Final Research Report

• Project/Area Number: 17K05658
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Meteorology/Physical oceanography/Hydrology
• Research Institution: Japan, Meteorological Research Institute
• Principal Investigator: Ishibashi Toshiyuki 気象庁気象研究所, 気象観測研究部, 主任研究官 (30585857)
• Project Period (FY): 2017-04-01 – 2024-03-31
• Keywords: データ同化

Outline of Final Research Achievements

In this study, by improving the accuracy of the error covariance matrix, the most important parameter in data assimilation, we relaxed the restrictions on assimilable observations for global atmospheric state analysis, and made it possible to assimilate a dramatically larger amount of observational information. This resulted in significant improvements in analysis and prediction accuracy and theoretical consistency. We showed that by increasing observational information, high accuracy can be obtained even when the adjoint model of the variational method is replaced by an ensemble forecast. The high-precision background error covariance matrix constructed by objective estimation through ensemble assimilation was analyzed based on network theory, and the basic properties of atmospheric perturbations were clarified.

Free Research Field

データ同化

Academic Significance and Societal Importance of the Research Achievements

大気はカオス系であり、その状態解析や予測の高精度化や理論整合性の向上は重要な科学的知見である。高精度な大気状態解析は、大気科学の発展に不可欠なデータセットの生成を可能にし、これらの発展にも不可欠である。ネットワーク理論による大気摂動の基本構造の解明は大気科学に新しい描像を提供する。また、データ同化を利用する他分野（海洋、固体地球、地球重力圏等）に広範囲に応用可能な普遍的知見となる。大気状態の解析や予測は社会基盤情報であり、その精度や理論整合性の向上は社会的にも重要である。