[H-DS06] Tsunami and tsunami forecast

[HDS06-P09] Attenuation process of tsunami caused by the 2011 off the Pacific coast of Tohoku Earthquake

*Naoko Shinmoto¹, Toshitaka Baba² (1.Department of Civil and Environmental Engineering, Division of Science and Technology, Graduate School of Sciences and Technology for Innovation, Tokushima University, 2.Graduate School of Technology, Industrial and Social Sciences, Tokushima University)

Keywords:tsunami, attenuation process

The attenuation process is complex for tsunamis generated by distant and giant earthquakes because of the long-lasting subsequent large waves. In order to cancel tsunami warnings with clear numerical evidence, it is necessary to improve prediction techniques for tsunami attenuation. We, therefore, attempted to numerically simulate the attenuation of the tsunami caused by the 2011 off the Pacific coast of Tohoku earthquake. Our numerical tsunami simulation used the source model of Satake et al. (2012) and a topographic nesting scheme. The first layer of the nesting grids (i.e., most coarse) used the linear long-wave equations, and the second and remaining layers used the nonlinear long-wave equations. We focused on six tide gauges at Hanasaki, Mera, Okada, Chichijima, Awayuki, and Amami, one wave gauge at Kushiro Port (NOWPHAS), and three offshore GPS wave gauges off North Miyagi, Central Miyagi, and Kaiyo-Tokushima. We used the moving root mean square (MRMS) amplitude defined by Hayashi (2010) for comparison.
We repeated 48-hour tsunami simulations by changing the calculation area and resolutions. It showed a significant difference in the reproducibility of MRMS amplitude waveforms in the calculations with an area up to the emperor seamount chain and an area of the northern Pacific Ocean. The calculation for the northern Pacific Ocean showed good accuracy. The high-resolution topographic grids around the observation point improved the reproducibility of the MRMS amplitude waveforms in the calculation using the northern Pacific Ocean. However, for the calculation with the limited area to the emperor seamount chain, high-resolution topographic grids around the observation point did not improve the reproducibility of MRMS amplitude. The topographic resolution required to reproduce MRMS amplitude waveforms differed by the analysis point. While a minimum grid interval of 20 arc-sec was sufficient for Hanasaki, 20/3 arc-sec was required for Mera, 20/9 arc-sec for Okada, and 20/27 arc-sec for Chichijima, Awayuki, and Amami. The minimum grid interval of 20 arc-sec was sufficient for Kushiro Port, North Miyagi, Central Miyagi, and Kaiyo-Tokushima because of offshore gauges. The required resolution for accurate prediction depends on the surrounding topography. We also examined extending the calculation area that covers the entire Pacific Ocean. However, the reproducibility of MRMS amplitude did not improve. Based on our calculations, the calculation of the pacific northern hemisphere was sufficient for predicting tsunami attenuation 48 hours after the earthquake.