5:15 PM - 7:15 PM
[SSS11-P12] Autocorrelation analysis of strong-motion records for 3D sedimentary structure in Okayama Prefecture using the generalized inversion technique
Keywords:Autocorrelation analysis, Generalized inversion technique, Strong motion record, Okayama Prefecture
In recent years, much research has been focused on extracting information on the ground structure through autocorrelation analysis of strong-motion records. The autocorrelation function can be interpreted as a pseudo-reflection profile with the ground surface as the epicenter. Signals with large amplitude in the autocorrelation function correspond to reflected waves from surfaces with large velocity contrasts. In order to enhance the signal of the autocorrelation function, the spectral whitening process should be appropriately implemented.
In this study, source and path factors separated by the generalized inversion technique of Kaneshima et al. (2024, JpGU) are applied to autocorrelation analysis. We then estimated the sedimentary structure throughout the Okayama Prefecture.
Fig. 2 shows pseudo-reflection profiles without surface-related multiples at 121 stations in Okayama Prefecture. The waveforms are sorted in the order of categories based on the site amplification factors (Kaneshima, et al., 2024, Fig. 1).
The first large-amplitude blue signal probably corresponds to reflected waves from the alluvium and diluvium in the shallow ground, and the second probably corresponds to reflected waves from the diluvium and tertiary layers.
By reading the two-way travel time of the S-waves from Fig. 2 and using the information on the S-wave velocity, the depths of the bottom of the alluvium (VS: about 150 m/s) and the bottom of the diluvium (VS: about 200 to 400 m/s) are estimated. The depth of the bottom of the diluvium was estimated to be about 60 m around Lake Kojima and about 110 m in the Hiruzen highlands.
From the results obtained, a depth distribution model was constructed using kriging interpolation. In almost all areas of Okayama Prefecture, the depth of the lower surface of the diluvium is estimated to be deeper than the depth of the lower surface of the uppermost layer of the J-SHIS (NIED) subsurface structure model. It is considered necessary to investigate the subsurface structure model as well.
In this study, source and path factors separated by the generalized inversion technique of Kaneshima et al. (2024, JpGU) are applied to autocorrelation analysis. We then estimated the sedimentary structure throughout the Okayama Prefecture.
Fig. 2 shows pseudo-reflection profiles without surface-related multiples at 121 stations in Okayama Prefecture. The waveforms are sorted in the order of categories based on the site amplification factors (Kaneshima, et al., 2024, Fig. 1).
The first large-amplitude blue signal probably corresponds to reflected waves from the alluvium and diluvium in the shallow ground, and the second probably corresponds to reflected waves from the diluvium and tertiary layers.
By reading the two-way travel time of the S-waves from Fig. 2 and using the information on the S-wave velocity, the depths of the bottom of the alluvium (VS: about 150 m/s) and the bottom of the diluvium (VS: about 200 to 400 m/s) are estimated. The depth of the bottom of the diluvium was estimated to be about 60 m around Lake Kojima and about 110 m in the Hiruzen highlands.
From the results obtained, a depth distribution model was constructed using kriging interpolation. In almost all areas of Okayama Prefecture, the depth of the lower surface of the diluvium is estimated to be deeper than the depth of the lower surface of the uppermost layer of the J-SHIS (NIED) subsurface structure model. It is considered necessary to investigate the subsurface structure model as well.