The investigation of dynamic rupture propagation is very important to understand the seismic behavior of mega-thrust earthquakes such as the 2011 Tohoku earthquake. The shallow parts of the fault (near the trench) hosted large slip and long period seismic wave radiation, whereas the deep parts of the rupture (near the coast) hosted smaller slip and strong radiation of short period seismic waves. Understanding such depth-dependent feature of the rupture process of the Tohoku earthquake is necessary as it may occur during future mega-thrust earthquakes in this and other regions. In order to achieve such understanding, dynamic rupture modeling is an important tool (e.g., Galvez et al., 2014). By incorporating the results of laboratory studies of samples of fault materials collected from plate boundary fault zones, such as the Japan Trench, dynamic rupture simulations can be made more realistic.

In this study, we developed dynamic rupture models of the Tohoku earthquake based on initial conditions and fault strength properties constrained by results of experimental studies (Hirono et al., 2016). Our large-scale simulation used the 3D spectral element method on unstructured grids (Galvez et al., 2014).

Our model reproduces the depth-dependency of the rupture process of the Tohoku earthquake. We also examine the sensitivity of the results to model parameters and assumptions, for instance to the value of the slip weakening distance (D_C). We find that the value of Dc does not affect the final slip distribution, as long as it is small enough to allow the rupture to develop and break up to the trench, and that long Dc promotes the generation of long period seismic waves.

This study investigates the building damage in Mashiki Town based on damage survey data of the local government due to the 2016 Kumamoto, Japan, earthquake. The damage ratios of buildings were investigated from the viewpoints of structural material, construction period, roof types and the number of floors. Moreover, the heavy damage ratios of wooden buildings were compared with those from the 1995 Kobe earthquake and the 2007 Niigata-ken Chuetsu-Oki earthquake. As the result, the major damage ratio of wooden houses was found to be much larger than those of other structural materials (RC, steel and light-gauge steel), and it got higher as the construction period becomes older. The results were further compared with the estimated distributions of the peak ground velocity (PGV) and the Instrumental JMA seismic intensity and the fragility curves for different structural materials (wood, S, and LS) and those for wooden buildings for five construction periods were developed.

We proposed a method to extract earthquake motions accurately from continuous observation records in noisy places. A simple method to connect the extracted records with data of an earthquake catalog was also proposed. We installed 4 accelerometers in a downtown area of Tokyo. Earthquake motions were picked based on the Fourier amplitude in the range from 0.5 to 2.0 Hz in which the amplitude level of microtremors was small. This method resulted in higher accuracy rate and higher extraction rate than the general method using STA/LTA. Although this method is difficult to extract for earthquakes smaller than M3, we successfully extracted 120 earthquakes including the earthquake occurred overseas from records of about half a year even in the noisy downtown area of Tokyo.

We explained seismic responses of vertical ground motions of S-wave part by considering obliquely incident S wave at bedrock. Vertical component of surface and borehole ground motions for vertically incident P wave and obliquely incident S wave were theoretically calculated for two different velocity structure models: Contrast model (velocity changes discontinuously between seismic bedrock and sediments), and Mirage model (velocity changes continuously between seismic bedrock and the surface). The result shown that the borehole-to-surface spectral ratios for the P- and S-wave incidents into the Contrast model were similar; however the spectral ratios into the Mirage model were significantly different. These results indicated that the obliquely incident S wave is required to evaluate seismic response for the Mirage model. We tested the S-wave incident approximation to explain seismic responses of vertical component seismograms recorded at KiK-net station SRCH08. The borehole-to-surface spectral ratios of the P- and S-wave parts of the observed vertical component seismograms were clearly different. To make an appropriate velocity structure model for SRCH08, we inverted receiver functions and borehole-to-surface spectral ratios of horizontal components. The observed borehole-to-surface spectral ratios well agree to the spectral ratios calculated for the obliquely incident S wave rather than the ones for the vertical P-wave incident. The synthetic seismograms for the oblique S-wave incident also coincide with the observed records in the time domain.

Demand for warehouses is increasing as logistics are globally expanding. This demand comes with an increasing requirement for warehouses to be cost effective and seismic proof, with the capability of preserving their contents. To meet these requirements effectively, the application of “sliding effect” to the structural design of warehouses is one option that can be used. Sliding effect is a technique that has attracted attention to reduce the seismic response of a building and load by sliding loads during an earthquake. The effectiveness of the sliding effect for structures under fixed base conditions has been confirmed through the experimental and analytical results of previous studies. However, this has yet to be confirmed in the case of considering soil-structure interaction (SSI). In this study, a large three-story warehouse is trial designed, and parametric studies considering SSI using modified Penzien model, which can analyze non-linear behavior, are conducted. Furthermore, additive equivalent damping factor h_ae, which converts the response reduction value into an additional damping factor, is calculated. The results show that although the response of the SSI model does not match that of the base fixed model, the SSI model has a sliding effect equivalent to the base fixed model, and more than 6% high h_ae are calculated.

Reliability evaluation of airport function due to earthquake occurrence is required to check seismic performance of the airports. On the other hand, many airports in Japan are located in the coastal area including the artificial island type. Therefore, it is necessary to evaluate the reliability of the airport function due to composite damage caused by earthquake motions, liquefaction and tsunami. We propose a method to evaluate the reliability of airport functions considering composite damage, in this paper. The proposed method is applied to the coastal area airport in the Kyushu region, and we evaluate the soundness curves in consideration of composite damage in case of inland earthquake and Nankai megathrust earthquake, respectively.

In order to understand spectral characteristics of source spectra in the high frequency band in detail, we estimated acceleration source spectra up to 70 Hz high frequency band of the 2016 Tottori-ken chubu earthquake (foreshock (Mj4.2), mainshock (Mj6.6) and largest aftershock (Mj5.0)) and the 2000 Tottori-ken seibu earthquake (four aftershocks (Mj2.5 to Mj3.0)) by using 1000 m deep borehole array recordings of hard rock site. The source spectra were calculated from the bedrock spectra at each depth of 100 m (Vs = 2.6 km/s), 300 m (Vs = 2.8 km/s) and 1000 m (Vs = 3.2 km/s). Based on a conventional method, the source spectra obtained by removing the propagation characteristics of the observation site from the bedrock spectra of each depth exhibit a tendency to decrease in the high frequency band above the corner frequency f_c and have different spectral shapes. On the other hand, the source spectra obtained by removing "bedrock characteristics" defined by Kobayashi and Mamada (2018) in addition to the propagation characteristics exhibit flat frequency characteristics in the high frequency band above f_c. In this case, there is little difference in the spectral shapes for each depth of seismic bedrock, and the source spectra can be stably estimated. The source spectra estimated by appropriately taking into account the bedrock characteristics and the propagation characteristics of the observation site show that the ω^-2 source spectral model roughly holds up to high frequency band up to 70 Hz which observation accuracy is guaranteed.

It was proposed to compile knowledge related to the broad-based crustal structure boundaries of the eastern margin of the Japan Sea obtained so far and to study uncertainties concerning the rupture zone and earthquake magnitude, and multi-segment earthquake. For that study, the crustal structure boundaries are set as the first branch to determine a logic tree for evaluation of epistemic uncertainties for tsunami estimation in the area. Moreover, the tsunami source models of the crustal structure boundary type were presented as one of the definite models that are necessary to evaluate the uncertainties. These tsunami source models are models exceeding the magnitude of the previous ones. They were developed for the region of which east dipping reverse faults were confirmed continuously by exploring the crustal structures concerning the nascent plate boundary theory, previous knowledge of the strain concentration zone, and latest finding on the crustal structure boundaries were compiled for development of the models. The assumable earthquake magnitudes are Mw 7.7 to 8.6, which exceed the magnitudes (Mw 7.5 to 7.9) of the crustal earthquakes of the tsunami source models of Ministry of Land, Infrastructure, Transport and Tourism (2014). Furthermore, the numerical simulation of tsunami propagation was performed using the tsunami source models of crustal structure boundary type and the crustal earthquake type to confirm the effect of the tsunami water level in the Japan Sea coast area. The results showed that the tsunami water levels obtained by the crustal structure boundary type models were two to three times higher than that by the crustal earthquake type models. Furthermore, the regional differences of the maximum water level were emphasized more by the crustal structure boundary type models than the crustal earthquake type models.

Okinawa Islands, southwestern Japan, have often been damaged by strong ground motion during some M7 class earthquakes, and there are high risks of the disaster in future earthquakes around these islands. We conducted array observation of microtremor survey at 23 sites on the islands and modelled one-dimensional S-wave velocity structure for each site with the inversion of estimated phase velocity. The S-wave velocity structures are modeled with one to three layers overlying a seismic bedrock, velocity layer over 3 km/s. Depths of the bedrock are from 0.2 to 0.4 km at the sites of the northern part of Okinawa island and Kume island, whereas these depths are from 2.0 to 2.5 km of the southern part of Okinawa one. Based on these one-dimensional structures, in case of four layers model, the averaged of S-wave velocities over all sites are 0.69, 1.10, 2.01 and 3.46 km/s, respectively, which are similar to those of the Sakishima islands. It suggests the feasibility of construction of subsurface 3D numerical model over all the Nansei Islands.

We estimated source spectra, Q value and site amplification factors of the source region of the 2018 Hokkaido Eastern Iburi Earthquake by using spectral inversion analysis. Q(f)=54f^0.89 was obtained in the source region. At HKD126, around which the building damage was severe, the site amplification factor from 1 to 3 Hz was about 10 and larger than surrounding sites. The short-period level of the main shock was two times larger than conventional research. The stress drop slightly showed focal depth dependency.