Development of probabilistic risk assessment method for aquatic environmental disasters based on large ensemble climate forecast data

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K04276
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 22040:Hydroengineering-related
• Research Institution: Fukuoka Institute of Technology (2023); Kyushu University (2021-2022)
• Principal Investigator: Tai Akira 福岡工業大学, 社会環境学部, 准教授 (20585921)
• Co-Investigator(Kenkyū-buntansha): 齋田 倫範 鹿児島大学, 理工学域工学系, 准教授 (80432863)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: アンサンブル気候予測データ / 沿岸海洋環境災害

Outline of Final Research Achievements

In this study, we develop a water environment prediction model using machine learning that takes climate data as input and outputs data indicative of coastal marine water environments, such as density and dissolved oxygen. The developed model is used to perform water environment predictions on ensemble climate prediction data, with the aim of conducting probabilistic environmental disaster risk assessments that consider uncertainties. First, we focused on the changes in the water quality environment at point B3 of the observation tower of the Kyushu Regional Agricultural Administration Office, located in the center of Isahaya Bay, and constructed a support vector regression model. We analyzed long-term observation data in the Ariake Sea and Isahaya Bay to clarify the interannual variations and their relationships of tidal currents, bottom layer dissolved oxygen concentration, river flow, and stratification.

Free Research Field

土木水工学

Academic Significance and Societal Importance of the Research Achievements

本研究では，気候予測データベースd4PDFに基づく九州地方の降雨及び河川流量の将来変化特性について，有明海・諫早湾の潮汐・潮流の経年変化特性，有明海諫早湾における流動構造の経年変化と水質の関係，河川流量を用いた機械学習による有明海の水質予測などを明らかにし，査読付き論文や学会などで発表を行った．