The tsunami source models of the 2011 Tohoku Earthquake, which reproduce the wave gauge record offshore the Fukushima Daiichi Nuclear Power Plant, were obtained by the joint inversion of tsunami waveform data, 1-second sampling GPS data, and sea bottom crustal deformation data. For improving the reproducibility of the wave gauge record, we used an approach that includes a “hypothetical tsunami waveform” data in the joint inversion. The hypothetical tsunami waveforms were defined at the depth of 50 m offshore the Fukushima Daiichi Nuclear Power Plant,. The tsunami waves at the position of the wave gauge were simulated based on nonlinear long-wave theory using the tsunami source models obtained by the joint inversion including the hypothetical tsunami waveforms. The final modified hypothetical tsunami waveform, which minimized the sum of squared residuals between observed and calculated tsunami waves at the position of wave gauge, was searched among many hypothetical tsunami waveforms. By including the modified hypothetical tsunami waveform in the joint inversion, we obtained the tsunami source models reproducing the wave gauge record well. A similar approach using the hypothetical tsunami waveform data was also applied for improving the fit between the observed and calculated tsunami heights from Hokkaido to Chiba prefectures. It also became clear that the combined use of the 1-second sampling GPS data in the inversion is effective for constraining the characteristics of the fault rupture process.

This study quantitatively verifies the “slide effect” for warehouse buildings with reinforced concrete (RC) column-steel beam (RCS), the usage of which has increased in recent years. The slide effect not only reduces the maximum displacement of the frame and the maximum acceleration of the load by the way of load sliding (both also referred as responses), but it also has the potential to reduce variation in these responses. In this study, analysis models of warehouse buildings with RCS were developed. One hundred artificial seismic waves created based on the provisions of the Japanese Building Code (JBC) were used as input for the models, and parameters such as average value, coefficient of variation, and confidence interval were calculated using loading rate, multiplying factor, and capacity reduction factor. The analysis results confirmed that the RCS structure has a slide effect equivalent to or better than that of the S structure. The slide effect was found to be the highest in the case of a warehouse building with a high loading ratio and a large ultimate lateral strength, under the conditions of a moderate earthquake. The slide effect was quantitatively expressed by converting it into an equivalent additional damping factor. The slide effect not only confirmed reduction in the variation of the maximum response value in many layers, but also that it had the effect of equalizing the maximum story drift of each layer when a warehouse building was subjected to a seismic input motion defined by the JBC as a large earthquake.

Vibration control technology of wooden main hall with flexible horizontal diaphragm was successfully developed. Response displacement of the wooden main hall was demonstrated to be reduced by 10% by employed vibration technology device installed in the horizontal plane. This technique was applied to the main hall of Cho-nenji Temple, in Kawasaki. Earthquake monitoring was performed to verify the effectiveness of equipment installed after completion of construction. Three dimensional analysis using a nonlinear frame model was compared with microtremor measurements and seismic records.

We studied a method of estimating construction years of wooden houses from pictures of exterior appearance. First, we performed estimation tests using wooden houses whose construction years were known to make sure the accuracy of the method. We confirmed that the construction year distribution of wooden houses could be estimated with an error of less than 5% pt by judging 5 or more people. Finally, we estimated construction year distribution by using this method around the strong ground motion observation points.