Development of point process models and prediction methods for non-stationary seismic activity

2023 Fiscal Year Final Research Report

• Project/Area Number: 20K11722
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 60030:Statistical science-related
• Research Institution: Waseda University (2021-2023); The Institute of Statistical Mathematics (2020)
• Principal Investigator: Nomura Shunichi 早稲田大学, 商学学術院(会計研究科), 准教授(任期付) (70719640)
• Co-Investigator(Kenkyū-buntansha): 田中 昌之 気象庁気象研究所, 地震津波研究部, 主任研究官 (10742728)
• Project Period (FY): 2020-04-01 – 2024-03-31
• Keywords: 地震予測 / 繰り返し地震 / 更新過程 / 余震減衰 / スプライン関数 / BPT分布 / 点過程 / 時空間モデル

Outline of Final Research Achievements

A forecasting method has been developed to estimate the future probability of occurrence of moderate recurrent earthquakes that occur quasi-periodically at plate boundaries. The recurrence intervals of moderate-sized recurrent earthquakes at plate boundaries are influenced by the occurrence of large earthquakes in the vicinity and by changes in plate motion. In this study, the Omori-Utsu aftershock decay law and spline functions are fitted to such changes in recurrence intervals to predict future changes and evaluate the probability of earthquake occurrence. The proposed method is applied to a group of medium-sized recurrent earthquakes on the subduction zone of the Pacific plate in the Tohoku region, and the future earthquake occurrence probability is evaluated for each recurrent earthquake series, which is verified with the actual occurrence during the evaluation period.

Free Research Field

統計地震学

Academic Significance and Societal Importance of the Research Achievements

大規模な繰り返し地震活動の予測手法は、周期が数十年から数万年にわたるため、予測の正しさを実績と照らし合わせて検証することが難しい。そこで、本研究では数年周期の中規模な繰り返し地震活動の予測へと焦点を当て、大地震やプレート運動による影響を考慮した予測手法を提案した。提案手法を実際の繰り返し地震のデータへ適用して将来にわたる地震発生確率を定期的に評価し、評価期間終了後に実績と照らし合わせた結果、妥当な予測が行えていることを実証することができた。