2023 Fiscal Year Final Research Report
Understanding the physics of foreshocks based on dense seismic observation and seismicity model
Project/Area Number |
21H01191
|
Research Category |
Grant-in-Aid for Scientific Research (B)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Review Section |
Basic Section 17040:Solid earth sciences-related
|
Research Institution | Hokkaido University (2023) Kyoto University (2021-2022) |
Principal Investigator |
Naoi Makoto 北海道大学, 理学研究院, 准教授 (10734618)
|
Co-Investigator(Kenkyū-buntansha) |
岩田 貴樹 県立広島大学, 公私立大学の部局等(庄原キャンパス), 准教授 (30418991)
飯尾 能久 京都大学, 防災研究所, 名誉教授 (50159547)
平野 史朗 立命館大学, 理工学部, 助教 (60726199)
中谷 正生 東京大学, 地震研究所, 教授 (90345174)
|
Project Period (FY) |
2021-04-01 – 2024-03-31
|
Keywords | 深層学習 / 震源カタログ構築 / 前震活動 |
Outline of Final Research Achievements |
We have achieved the development and improvement of analysis methods to investigate preseismic activity before a major earthquake. In particular, for the earthquake cataloging procedure, we improved the conventional processing pipeline by using deep learning and developed an efficient similar component search technique based on deep learning, which is utilized in image recognition problems, to seismic waveforms. In the actual data analysis, remarkable results were achieved mainly in the analysis of acoustic emission data obtained in laboratory hydraulic fracturing experiments, revealing detailed patterns of acoustic emissions in the preparation process of the macroscopic fracture.
|
Free Research Field |
地震学
|
Academic Significance and Societal Importance of the Research Achievements |
大地震に先行して生じる特定の地震活動パタンが確認できるかは,微小な地震の検出能力で結果が変わっていることが先行研究のメタ分析で示されており,効率よく地震カタログを作成する技術は非常に重要である.本課題で達成した深層学習を用いた従来型処理プロセスの改善や,効率的類似波形探索の実現は,従来手法が抱えていた処理速度や精度などの課題を解決しうるものであり,大きな学術的・社会的意義を持つとともに,今後の研究を促進させると期待できる.
|