[J04-4] Geohazard early warning systems IV

[J04-4-02] Real-Time Tsunami Inundation Forecast System using NIED S-net

^○Shin Aoi, Naotaka Yamamoto, Wataru Suzuki, Kenji Hirata, Hiromitsu Nakamura, Takashi Kunugi, Tomohiro Kubo (NIED, Tsukuba, Japan)

invited

One of the lessons learned from the 2011 Tohoku earthquake is the importance of offshore observations. JMA issued the initial tsunami warning three minutes after the 2011 Tohoku earthquake, however, estimated tsunami heights were underestimated. Although the estimated tsunami heights were raised about half an hours after the earthquake occurrences, the power failures prevented the residence from receiving updated tsunami information. It consequently caused enormous fatalities and significant damages to the eastern Pacific coast of Japan, indicating the importance to deliver prompt and accurate forecast of earthquake and tsunami to the onshore residences. The Japanese government decided to construct the Seafloor Observation Network for Earthquakes and Tsunamis along the Japan Trench (S-net), which is operated by National Research Institute for Earth Science and Disaster Resilience (NIED). NIED S-net is expected to provide additional lead time for earthquake and tsunami early warning. We develop a new realtime tsunami inundation forecast system using NIED S-net. In our system, several appropriate tsunami scenarios that can explain offshore tsunami observations are quickly selected by using multiple indices from the Tsunami Scenario Bank (TSB), which contains offshore tsunami pre-calculated waveforms, coastal tsunami heights, inundation depth maps, and others. One of the key features of our system is that tsunami inundations are estimated explicitly without any source information, which may contain large estimation error. Many calculations are performed in advance to investigate the sensitivities of the source models to coastal tsunami heights and inundation depth along the Pacific coast of Chiba prefecture in the Kanto region. We evaluate and improve our system through the demonstration experiments with local governments. We also validate the performance of the system for the simulated data as well as the observed data of the 2016 Fukushima-oki earthquake.