5:15 PM - 7:15 PM
[SSS11-P07] Estimation of the bedrock depth of the southern part of Takamatsu Basin using microtremor exploration
Keywords:Takamatsu basin, microtremors, seismic bedrock depth, H/V, seismic interferometry
Takamatsu City is situated in the Takamatsu basin and faces the risk of the Nankai Trough huge megathrust earthquake. It is important to understand the sedimentary basin structure for the purpose of the ground motion simulation. The city center is located northern part of the basin and has been conducted deep boreholes and estimated the depth to the bedrock (e.g. Hasegawa and Saito, 1989). However, the southern part of the basin has not been well investigated. We apply the microtremor horizontal to vertical spectral ratio (H/V) technique to model the basin structure. The point of this study is the assumption of the velocity of the sedimentary layers because the assumption is necessary to convert the peak frequency of H/V to the depth.
To convert the peak frequency of H/V to the depth of the sediment, it is required to assume velocity value. For the purpose, we performed the seismic interferometry in the Takamatsu basin. Four temporary stations were installed in the basin from 13 to 20th Jun, 2024. We used the three-component sensor JEP-6A3 which is a high-damping velocity meter with a sensitivity of 10V/g connected with a 24-bit data logger LS-8800. The sensor has a flat response from 0.1 to 30Hz.
We computed cross correlation functions (CCFs) between the possible two stations pairs using a week continuous record by following Chimoto and Yamanaka (2014). The zero-crossing method (Ekstrom et al., 2009) was then applied to the CCFs to measure the dispersion curve of the phase velocity of Rayleigh wave. The dispersion curve of Rayleigh wave phase velocity at each station pair shows significant dispersive features in the frequency range of 0.5-2Hz. The phase velocity values at 2-4Hz are approximately 500m/s. We therefore assumed the average Vs of the deep sediment is 500 m/s.
The dominant frequency of H/V was converted to the bedrock depth using quarter wavelength method. We assume a simple two layers model for the deep sediment. Mitoyo Formation, early Pleistocene overlays Ryoke granite which is a bedrock of Sanuki Region (Hasegawa and Saito, 1989). We neglected the alluvial deposite for the estimation of the bedrock depth because the thickness is expected to be not large in the southern part of the basin.
We used the same instruments for the single-station microtremor measurement. For the microtremor measurement, we observed approximately 10-minute recordings at each station over 280 sites. Most of the H/Vs observed at all the sites have a peak at approximately 1Hz and some have a peak at approximately 3Hz. We observe low dominant frequency with a significant amplitude at southern and central part of the basin located at alluvial fan. Dominant frequency at these sites is approximately 0.7 Hz, the lowest in the basin, lower than the coastal region, the north part of the basin. South part shows the peak at 1Hz. The south end of the basin does not show peak in H/V or has very high peak frequency. There is the Nagao Fault Zone in the south part of Takamatsu basin. However, the H/V was not significantly varied across the Nagao Fault in this region. The northwestern part and eastern part of the basin has a slightly higher dominant frequency compared to the central part of the basin.
As it was seen in the peak frequency map, the deepest part of the basin is southern and central part of the basin, with the depth of approximately 180 m. The J-SHIS Version 2 (NIED, 2019) model matched with the estimated depth in this study. However, there is a discrepancy at some site in the central part of the basin. Although other parts are matched well each other, we must remind the J-SHIS model has mainly two layers as sedimentary layers with the Vs of 600 m/s and 1100-1400m/s. Therefore, the peak must appear at higher frequency if we assume the J-SHIS model. This study suggests the average velocity must be modeled appropriately to explain the measured peak frequency of H/V in this region.
To convert the peak frequency of H/V to the depth of the sediment, it is required to assume velocity value. For the purpose, we performed the seismic interferometry in the Takamatsu basin. Four temporary stations were installed in the basin from 13 to 20th Jun, 2024. We used the three-component sensor JEP-6A3 which is a high-damping velocity meter with a sensitivity of 10V/g connected with a 24-bit data logger LS-8800. The sensor has a flat response from 0.1 to 30Hz.
We computed cross correlation functions (CCFs) between the possible two stations pairs using a week continuous record by following Chimoto and Yamanaka (2014). The zero-crossing method (Ekstrom et al., 2009) was then applied to the CCFs to measure the dispersion curve of the phase velocity of Rayleigh wave. The dispersion curve of Rayleigh wave phase velocity at each station pair shows significant dispersive features in the frequency range of 0.5-2Hz. The phase velocity values at 2-4Hz are approximately 500m/s. We therefore assumed the average Vs of the deep sediment is 500 m/s.
The dominant frequency of H/V was converted to the bedrock depth using quarter wavelength method. We assume a simple two layers model for the deep sediment. Mitoyo Formation, early Pleistocene overlays Ryoke granite which is a bedrock of Sanuki Region (Hasegawa and Saito, 1989). We neglected the alluvial deposite for the estimation of the bedrock depth because the thickness is expected to be not large in the southern part of the basin.
We used the same instruments for the single-station microtremor measurement. For the microtremor measurement, we observed approximately 10-minute recordings at each station over 280 sites. Most of the H/Vs observed at all the sites have a peak at approximately 1Hz and some have a peak at approximately 3Hz. We observe low dominant frequency with a significant amplitude at southern and central part of the basin located at alluvial fan. Dominant frequency at these sites is approximately 0.7 Hz, the lowest in the basin, lower than the coastal region, the north part of the basin. South part shows the peak at 1Hz. The south end of the basin does not show peak in H/V or has very high peak frequency. There is the Nagao Fault Zone in the south part of Takamatsu basin. However, the H/V was not significantly varied across the Nagao Fault in this region. The northwestern part and eastern part of the basin has a slightly higher dominant frequency compared to the central part of the basin.
As it was seen in the peak frequency map, the deepest part of the basin is southern and central part of the basin, with the depth of approximately 180 m. The J-SHIS Version 2 (NIED, 2019) model matched with the estimated depth in this study. However, there is a discrepancy at some site in the central part of the basin. Although other parts are matched well each other, we must remind the J-SHIS model has mainly two layers as sedimentary layers with the Vs of 600 m/s and 1100-1400m/s. Therefore, the peak must appear at higher frequency if we assume the J-SHIS model. This study suggests the average velocity must be modeled appropriately to explain the measured peak frequency of H/V in this region.