[S-SS11] Strong Ground Motion and Earthquake Disaster

[SSS11-04] Average S-wave velocity maps based on a microtremor empirical method: Case of a region with pyroclastic flow deposits to the east of the Aso caldera

*Ikuo Cho¹, Keiya Yoneoka¹, Daisaku Kawabata¹, Yoshinori MIYACHI¹, Hideo Hoshizumi¹, Nobuo Geshi² (1.National Institute of Advanced Industrial Science and Technology, 2.Kyushu University)

Keywords:slope disaster, Aso, microtremor, S-wave velocity

We conducted microtremor array observations at 173 points within an area of 17 x 15 km with pyroclastic flow deposits to the east of Aso Caldera. We applied an empirical method to the obtained phase velocity data to construct average S-wave velocity (AVS) maps to the depths of 10, 20, and 30 m. The empirical method is straightforward where we regard the phase velocities corresponding to 13, 25, and 40 m wavelength as AVS10, 20, and 30, respectively. This rule of thumb was proposed about two decades ago, but it does not seem to be widely used in making an AVS map. With this background, we first validated this method based on PS logging data at a single site. Next, we confirmed the AVS maps obtained by adopting this method are in good agreement with the geological distribution of the target site. It was observed, for example, that the AVS values obtained within Aso-3 ignimbrite areas are 30-40 % larger than those obtained within areas of thick tephra/soil deposits above Aso-4 ignimbrite. It was also observed that they are over 10 % larger than those obtained within Aso-4 ignimbrite areas, while they are about 10 % smaller than those obtained within the areas of the older units. Finally, this method was statistically verified based on a comparison with AVS values based on conventional inversion. Although there were systematic discrepancies of about 15 % between these two kinds of AVS values, correlation coefficients of more than 0.9 were obtained. It seems that these discrepancies were caused contributed by both methods. Although we need to pursue clarifying the causes of the systematic discrepancies in future studies, we can adequately say that the AVS identification based on the empirical method has been quantitatively validated in the case of our target site. This empirical method is much more convenient than conducting inversion analysis, in terms of simplicity and certainty of planning, observational efficiency, and simplicity of data processing, and is considered extremely useful for processing a large number of sites, such as constructing an AVS map.