• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Next-generation Bayesian seismic source estimation integrated with ensemble underground structure estimation

Research Project

Project/Area Number 21K14024
Research Category

Grant-in-Aid for Early-Career Scientists

Allocation TypeMulti-year Fund
Review Section Basic Section 17040:Solid earth sciences-related
Research InstitutionJapan Agency for Marine-Earth Science and Technology

Principal Investigator

Agata Ryoichiro  国立研究開発法人海洋研究開発機構, 海域地震火山部門(地震発生帯研究センター), 研究員 (80793679)

Project Period (FY) 2021-04-01 – 2024-03-31
Project Status Completed (Fiscal Year 2023)
Budget Amount *help
¥3,770,000 (Direct Cost: ¥2,900,000、Indirect Cost: ¥870,000)
Fiscal Year 2023: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2022: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2021: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Keywordsベイズ推定 / 震源決定 / 地震波トモグラフィ / アンサンブル / 深層学習 / 震源解析 / 地下構造推定
Outline of Research at the Start

本研究では、1.ベイズ推定に基づいた地下物性の構造推定により、地下構造を不確かさの統計的性質ごとアンサンブル(ランダムなサンプルの集まり)として推定し、2.そのアンサンブルを計算用モデルとしてそのまま取り込むことで地下構造を不確かさごと考慮した地震の震源解析を行う、という構造推定と震源解析を一気通貫した枠組みの下に、一つの地下構造モデルを恣意的に選択することによる悪影響を排除した次世代震源解析システムの開発を目指す。

Outline of Final Research Achievements

In this study, we aim to estimate the subsurface seismic wave velocity structure as an ensemble with uncertainties and to determine the earthquake hypocenter considering these uncertainties. First, we realized the quantification of uncertainties in seismic tomography, a method for estimating seismic wave velocity structures, by using deep learning techniques based on physical laws. Next, we applied this method to analyze seismic exploration data near the hypocentral region of an earthquake occurred off the southeastern coast of Mie Prefecture in 2016, obtaining an ensemble model of P-wave velocity structures. Based on this ensemble, we determined the hypocenter of the earthquake while considering the uncertainties in the P-wave velocity structure estimation. Our results demonstrated that the impact of uncertainty propagation from the velocity structure on hypocenter determination cannot be ignored.

Academic Significance and Societal Importance of the Research Achievements

地震波速度構造の推定における不確実性が、その結果を使った震源決定に与える影響を検討したのは、本研究が初めてである。その影響が、2016年三重県南東沖地震という過去の巨大地震の震源域で起きた重要な地震の震源位置の推定において重要な意味を持つことを本研究は指摘している。また、本研究では、物理法則に基づく深層学習を不確実性定量化の組み合わせにより実自然科学データの解析に成功したが、このような例は固体地球科学分野に限らずこれまでにほとんどない、計算科学の側面においても重要な成果といえる。

Report

(4 results)
  • 2023 Annual Research Report   Final Research Report ( PDF )
  • 2022 Research-status Report
  • 2021 Research-status Report
  • Research Products

    (14 results)

All 2023 2022 2021

All Journal Article (2 results) (of which Peer Reviewed: 2 results) Presentation (12 results) (of which Int'l Joint Research: 2 results,  Invited: 2 results)

  • [Journal Article] Bayesian Seismic Tomography Based on Velocity-Space Stein Variational Gradient Descent for Physics-Informed Neural Network2023

    • Author(s)
      Agata Ryoichiro、Shiraishi Kazuya、Fujie Gou
    • Journal Title

      IEEE Transactions on Geoscience and Remote Sensing

      Volume: 61 Pages: 1-17

    • DOI

      10.1109/tgrs.2023.3295414

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Bayesian Multi‐Model Estimation of Fault Slip Distribution for Slow Slip Events in Southwest Japan: Effects of Prior Constraints and Uncertain Underground Structure2022

    • Author(s)
      Agata, R., Nakata, R., Kasahara, A., Yagi, Y., Seshimo, Y., Yoshioka, S., & Iinuma, T
    • Journal Title

      Journal of Geophysical Research: Solid Earth

      Volume: 127 Issue: 8

    • DOI

      10.1029/2021jb023712

    • Related Report
      2022 Research-status Report
    • Peer Reviewed
  • [Presentation] Physics-informed neural networksと粒子ベース変分推論に基づくベイジアン屈折初動走時トモグラフィ手法の開発2023

    • Author(s)
      縣 亮一郎、白石 和也、藤江 剛
    • Organizer
      日本地球惑星科学連合2023年大会
    • Related Report
      2023 Annual Research Report
  • [Presentation] 三重県南東沖における地下構造と震源のPhysics-informed neural networkによるアンサンブル推定2023

    • Author(s)
      縣 亮一郎、白石 和也、藤江 剛
    • Organizer
      日本地震学会2023年度秋季大会
    • Related Report
      2023 Annual Research Report
  • [Presentation] アイコナル方程式に基づくトモグラフィ解析のBayesian physics-informed neural networkを用いた不確実性定量化と地震探査データによる地震波速度構造推定への適用2023

    • Author(s)
      縣 亮一郎、白石 和也、藤江 剛
    • Organizer
      第20回(2023年度)日本応用数理学会研究部会連合発表会
    • Related Report
      2023 Annual Research Report
  • [Presentation] Bayesian seismic tomography based on particle-based variational inference in velocity-space for physics-informed neural network2023

    • Author(s)
      Ryoichiro Agata, Kazuya Shiraishi, Gou Fujie
    • Organizer
      Asia Oceania Geosciences Society 20th Annual Meeting
    • Related Report
      2023 Annual Research Report
  • [Presentation] Bayesian physics-informed neural networks for seismic tomography based on function-space particle-based variational inference2023

    • Author(s)
      Ryoichiro Agata, Kazuya Shiraishi, Gou Fujie
    • Organizer
      10th International Congress on Industrial and Applied Mathematics
    • Related Report
      2023 Annual Research Report
  • [Presentation] Ensemble estimation of seismic velocity and hypocenter based on physics-informed neural network2023

    • Author(s)
      Ryoichiro Agata, Kazuya Shiraishi, Gou Fujie
    • Organizer
      AGU Annual Meeting 2023
    • Related Report
      2023 Annual Research Report
  • [Presentation] 地下構造の不確かさを考慮したベイズマルチモデル断層すべり推定について2022

    • Author(s)
      縣 亮一郎、中田 令子、笠原 天人、八木 勇治、瀬下 幸成、吉岡 祥一、飯沼 卓史
    • Organizer
      日本地球惑星科学連合2022年大会
    • Related Report
      2022 Research-status Report
    • Invited
  • [Presentation] Physics-informed neural networksと粒子ベース変分推論による地震波トモグラフィの不確かさ定量2022

    • Author(s)
      縣 亮一郎
    • Organizer
      日本地震学会2022年度秋季大会
    • Related Report
      2022 Research-status Report
    • Invited
  • [Presentation] Bayesian Multi-model Fault Slip Estimation for Slow Slip Events Considering the Uncertainty of Underground Structure2022

    • Author(s)
      Ryoichiro Agata, Ryoko Nakata, Amato Kasahara, Yuji Yagi,Yukinari Seshimo, Shoichi Yoshioka, Takeshi Iinuma
    • Organizer
      AOGS2022 Virtual
    • Related Report
      2022 Research-status Report
    • Int'l Joint Research
  • [Presentation] 豊後水道長期的スロースリップにおけるすべり分布のベイズマルチモデル推定:先験的拘束条件の推定結果に対する影響2021

    • Author(s)
      縣 亮一郎、中田 令子、笠原 天人、八木 勇治、瀬下 幸成、吉岡 祥一、飯沼 卓史
    • Organizer
      地震学会2021年度秋季大会
    • Related Report
      2021 Research-status Report
  • [Presentation] A type of Bayesian multi-model inference for fault slip distribution: the effect of prior constraints in the estimation for slow slip events beneath the Bungo Channel, southwest Japan2021

    • Author(s)
      Ryoichiro Agata, Ryoko Nakata, Amato Kasahara, Yuji Yagi, Yukinari Seshimo, Shoichi Yoshioka, Takeshi Iinuma
    • Organizer
      AGU Fall Meeting 2021
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] A Bayesian inversion for slip distribution of slow slip events beneath the Bungo Channel based on ensemble modeling of the uncertainty of underground structure2021

    • Author(s)
      Ryoichiro Agata, Ryoko Nakata, Yuji Yagi, Takeshi Iinuma
    • Organizer
      Japan Geoscience Union Meeting 2021
    • Related Report
      2021 Research-status Report

URL: 

Published: 2021-04-28   Modified: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi