IAG-IASPEI 2017

Presentation information

Poster

Joint Symposia » J04. Geohazard early warning systems

[J04-P] Poster

Thu. Aug 3, 2017 3:30 PM - 4:30 PM Shinsho Hall (The KOBE Chamber of Commerce and Industry, 3F)

3:30 PM - 4:30 PM

[J04-P-03] Numerical shake prediction incorporating heterogeneous structure: the 2016 Kumamoto Earthquake

Masashi Ogiso1, Mitsuyuki Hoshiba1, Azusa Shito2, Satoshi Matsumoto2 (1.Meteorological Research Institute, Japan Meteorological Agency, Tsukuba, Japan, 2.Institute of Seismology and Volcanology, Kyushu University)

Needless to say, heterogeneous structure, especially attenuation structure is important for ground motion prediction. In the new concept of real time ground motion prediction, called “Numerical Shake Prediction" proposed by Hoshiba and Aoki (2015), prediction of ground motion prediction is done by the estimation of current wavefield using data assimilation and simulation of wave propagation successively. We expected that the simulation with heterogeneous structure would improve the precision of ground motion prediction in the numerical shake prediction scheme. In this study, first we estimated the heterogeneous attenuation structure of the western part of Japan, then we evaluated the effect of the structure through the case of the 2016 Kumamoto Earthquake sequence.
To estimate intrinsic and scattering attenuation structure simultaneously, we conducted Multiple Lapse Time Window Analysis for the seismic stations distributed in the western part of Japan. Derived attenuation structure shows strong intrinsic and scattering attenuation around active faults and volcanoes in the Kyushu area. We derived tentative 2D velocity structure using the distribution of average apparent velocity of each station used in the MLTWA.
We conducted real time ground motion prediction simulation based on the numerical shake prediction scheme with heterogeneous structure for the Kumamoto Earthquake. In the case of 10 s ahead prediction, RMS of intensity prediction residuals improved by 15 % compared with the case of homogeneous structure. The rate of improvement became high in the 20 s ahead prediction. These fact shows the importance of considering the heterogeneous structure for the longer lead time.

Acknowledgment
We used waveforms observed by K-NET/KiK-net/Hi-net operated by NIED, the seismic network of Kyoto Univ., Kyushu Univ. and JMA. This study was supported by the Joint Usage/Research Center program of Earthquake Research Institute, the University of Tokyo.