[H-DS11] 津波とその予測

[HDS11-04] Study on the Tsunami Generation by Sector Collapse of Komagatake Volcano for Future Tsunami Warning System

*Rinda Nita Ratnasari¹、Yuichiro Tanioka¹ (1.Institute of Seismology and Volcanology, Hokkaido University, Sapporo, Japan)

キーワード：Volcanic tsunamis, Komagatake volcano, Tsunami Early Warning

Tsunami from volcanic activity including sector collapse is considered as an important source mechanism. Constructing an effective tsunami warning system for volcanic tsunamis is necessary for disaster mitigation strategies. However, it comes with challenges owing to the different characteristics of each collapse event. In the development of tsunami early warning, it is important to understand the tsunami characteristics from volcanic sector collapse for designing a suitable tsunami forecasting method. In this study, we try to identify the tsunami characteristics and examine the feasibility of monitoring systems for tsunami forecasting by simulating the sector collapse of the Komagatake volcano. Numerical modeling of the 1640 Komagatake tsunami was first conducted to test the reliability of our selected models using an initial collapse volume of 1.2 km³. The simulation was compared to available tsunami data from several field surveys. As a result, tsunami deposits at seven locations from the survey of Nakanishi and Okamura (2019) were explained well by our tsunami inundation simulation. However, the observed tsunami height at a few places from the study of Nishimura and Miyaji (1998) was underestimated by our simulation. Additional simulations were then conducted using a larger initial collapse volume of 1.8 and 2.4 km³ to identify the appropriate collapse volume to explain the observed tsunami data. We also compared the landslide depositional area from three different volumes simulation with the depositional area from the field survey. The result indicates that larger initial collapse volumes of 1.8 and 2.4 km³ explained both the observed tsunami heights and the depositional area from the field survey of Yoshimoto et al. (2003). Further analysis of tsunami characteristics and tsunami hazards was investigated by simulating three hypothetical collapse scenarios of the Komagatake volcano. Three virtual observation stations were intentionally placed in the vicinity of the volcano as a part of our strategy to identify the tsunami characteristics from the sector collapse of Komagatake. Simulation from three hypothetical collapse scenarios indicates that the tsunami from Komagatake is characterized by wide-distributed deposits caused by landslides in multiple directions. From the numerical simulation, the sector collapse from all three scenarios resulted in similar tsunami distribution patterns along Uchiura Bay. Comparing three different scenarios, we found that the tsunami heights along Uchiura Bay were likely influenced by collapse volume. The variations in tsunami height can be identified by looking at amplitude differences from the observed waveforms at virtual stations. These findings suggest that the tsunami heights along Uchiura Bay should be well predicted by using the waveform amplitudes observed at virtual stations. A study on the volcanic sector collapse of Komagatake volcano shows that the existence of observation stations in the vicinity of the volcano can provide essential information for tsunami forecasting.