日本地球惑星科学連合2021年大会

講演情報

[J] 口頭発表

セッション記号 S (固体地球科学) » S-CG 固体地球科学複合領域・一般

[S-CG54] 東北地方太平洋沖地震から10年―固体地球科学の到達点

2021年6月6日(日) 10:45 〜 12:15 Ch.17 (Zoom会場17)

コンビーナ:日野 亮太(東北大学大学院理学研究科)、小平 秀一(海洋研究開発機構 海域地震火山部門)、松澤 暢(東北大学大学院理学研究科附属地震・噴火予知研究観測センター)、飯沼 卓史(国立研究開発法人 海洋研究開発機構)、座長:小平 秀一(海洋研究開発機構 海域地震火山部門)、日野 亮太(東北大学大学院理学研究科)

12:00 〜 12:15

[SCG54-12] Real-time Tsunami Inundation Forecast in Japan - Present and Future

*越村 俊一1,5、日野 亮太2,5、太田 雄策2,5、撫佐 昭裕3,5、鈴木 崇之5、阿部 孝志1、小林 広明4,5 (1.東北大学災害科学国際研究所、2.東北大学大学院理学研究科、3.東北大学サイバーサイエンスセンター、4.東北大学大学院情報科学研究科、5.(株)RTi-cast)

キーワード:津波、リアルタイム予測、災害対応

The 2011 Great East Japan earthquake and tsunami disaster revealed many problems in Japan’s disaster management policies, and these have undergone reforms in the years since to promote initiatives for building national resilience in confronting any future disasters. One of the key challenges in the aftermath of tsunami disaster is identifying its impact and prioritizing disaster response and relief activities. Because of the widespread damage to infrastructure and communication networks, the impacted regions were hampered in addressing the overall damage, sometimes for months. This experience highlighted the need to develop technologies to forecast the regional impact of tsunamis. Recent advances in high-performance computing and large data sets comprising observations of tsunami emergence, propagation, and effects hold out the promise of dramatically improving our understanding of the whole picture of tsunami-affected areas in real-time.

Three approaches to real-time tsunami inundation forecast methods have been proposed in Japan. The first is the “Tsunami Scenario Database” approach: searching tsunami forecast data from pre-computed database connected with offshore tsunami observation. Once a tsunami is observed offshore, the pre-computed database starts searching for the best matching pair of offshore tsunami heights that fits the observations, then projects the tsunami inundation scenarios most likely to correctly forecast coastal tsunami heights.

The second is the “Data Assimilation” approach that assimilates tsunami wave field using dense offshore tsunami observation networks. This method estimates the tsunami wave field (tsunami height and tsunami velocity) in real-time by repeatedly assimilating dense tsunami observation data into a numerical simulation. Both are highly dependent on the configuration of offshore tsunami sensors.

The third is the “Real-time Forward Simulation” approach that runs simulations in real-time with estimation of tsunami source models based on seismic and geodetic observations. The real-time forward approach has the advantages in simulating tsunami inundation on land—if reliable tsunami source model information is obtained. None of the three methods was developed before the 2011 Great East Japan Earthquake. Some “Tsunami Scenario Database” methods are now being operated in some coastal areas, and the “Real-time Forward Simulation” method is now in operation as a part of the Japanese central government’s emergency response effort.

Throughout the case studies of modeling recent and potential tsunami events, we discuss advantages and drawbacks of those approaches with regard to applicabilities towards next generation tsunami inundation forecast system.