15:30 〜 15:45
[SCG40-07] 2D FDM simulation of seismic ground motions and inversion of velocity structure models in the Kathmandu Valley, Nepal
キーワード：Kathmandu Valley, strong ground motion, 2D FDM simulation, inversion, velocity structure model
Great and disastrous earthquakes have repeatedly occurred at the subduction interface along the Himalayan arc in the past and would continue to occur in the future due to the continuous collision between the Indian Plate and Eurasian Plate. And a recent example was the 2015 Mw7.8 Gorkha earthquake, which induced serious seismic disasters in the Kathmandu Valley, Nepal. The Kathmandu Valley sits on an ancient lake bed zone with irregular soft sedimentary layers of several hundred meters depth in the deep part. Previous studies of the tomography of the Kathmandu valley demonstrated that the velocity structure model in the Kathmandu Valley is highly spatially variable, with obvious difference of thickness of soft sedimentary layers within short distances. The gravity survey and seismic reflection experiment conducted in the Kathmandu Valley in latest years by the joint research team between Japan and Nepal has confirmed the high thickness gradients of sediments and obtained a preliminary 3D velocity structure model. Because seismic ground motions especially long-period ground motions that can be significantly amplified and reverberated due to the basin effect within the valley can pose a serious threat to the buildings and infrastructures in the Kathmandu Valley, it’s necessary to investigate the propagation effect and the amplification effect of seismic ground motions in the valley and to refine the preliminary velocity structure model for the strong ground motion simulations for seismic hazard assessment.
A collaboration between Japan and Nepal has begun since 2011 to study the strong motion characteristics in the Kathmandu Valley and several strong motion accelerometers have been installed within the valley since then. Three-component strong motions of several moderate regional and local earthquakes and small local earthquakes have been recorded by these strong motion stations. In this study, we performed 2D FDM simulations of seismic ground motions at the strong motion station sites in the Kathmandu Valley using the preliminary velocity structure model and compared with other researcher’s 3D FDM simulation results and the observation data for a moderate regional earthquake. The results showed that the source size effect of the moderate regional earthquake could not be ignored, and that the 2D FDM simulation could reproduce the early phases of the propagation effect but could not reproduce the later phases of the large surface waves generated in the deep part of the Kathmandu Basin. Due to the limited number of strong motion records accumulated in the Kathmandu Valley, we perform inversion of 2D velocity structure models based on the preliminary 3D model in the valley. To avoid the source size effect of the moderate regional or local earthquakes, we choose to use small local earthquakes which occurred near the Kathmandu Valley to perform inversion of 2D velocity structure models. The 2D velocity structure models along the lines connecting the epicenter of the small local earthquake and different strong motion station sites are inverted. The 2D FDM simulations of strong ground motions using the inverted velocity structure models in the Kathmandu Valley could reproduce the amplitudes of the main phases of the observation data. Our study on the strong motion simulations and inversion of velocity structure models in the Kathmandu Valley provides a basis for the seismic hazard assessment in this area.
A collaboration between Japan and Nepal has begun since 2011 to study the strong motion characteristics in the Kathmandu Valley and several strong motion accelerometers have been installed within the valley since then. Three-component strong motions of several moderate regional and local earthquakes and small local earthquakes have been recorded by these strong motion stations. In this study, we performed 2D FDM simulations of seismic ground motions at the strong motion station sites in the Kathmandu Valley using the preliminary velocity structure model and compared with other researcher’s 3D FDM simulation results and the observation data for a moderate regional earthquake. The results showed that the source size effect of the moderate regional earthquake could not be ignored, and that the 2D FDM simulation could reproduce the early phases of the propagation effect but could not reproduce the later phases of the large surface waves generated in the deep part of the Kathmandu Basin. Due to the limited number of strong motion records accumulated in the Kathmandu Valley, we perform inversion of 2D velocity structure models based on the preliminary 3D model in the valley. To avoid the source size effect of the moderate regional or local earthquakes, we choose to use small local earthquakes which occurred near the Kathmandu Valley to perform inversion of 2D velocity structure models. The 2D velocity structure models along the lines connecting the epicenter of the small local earthquake and different strong motion station sites are inverted. The 2D FDM simulations of strong ground motions using the inverted velocity structure models in the Kathmandu Valley could reproduce the amplitudes of the main phases of the observation data. Our study on the strong motion simulations and inversion of velocity structure models in the Kathmandu Valley provides a basis for the seismic hazard assessment in this area.