9:45 AM - 10:00 AM
[SSS10-14] Earthquake Horizontal-to-Vertical spectral ratios in Osaka, Kyoto, and Nara basins from waveforms recorded in 1994-2023
Keywords:Site effects, Strong ground motion, Earthquake waveforms, H/V spectral ratio, H/V directionality, Osaka basin
The earthquake ground motion observed at a site is a result of source, path, and site effects. The site effects can be the origin of casualties and significant damage; hence, they should be investigated especially in urban areas such as Osaka, Kyoto, and Nara basins (Kansai area, Japan). A simple method to investigate the site-specific amplification is using empirical Horizontal-to-Vertical (H/V) spectral ratios, because frequencies of empirical H/V spectral peaks are usually correlated to the local spectral amplification of seismic waves. The spectral ratios can be computed from three-component recordings of microtremors (ambient noise) or local earthquakes, in which both of these approaches have some advantages and disadvantages. The processing of microtremors evaluates the properties of the diffuse wave field, but data acquisition is very rapid. Instead, earthquake spectral ratios reflect the properties of upward body waves and basin-induced surface waves, but a long monitoring time is necessary to acquire a representative set of waveforms.
In this study, we evaluate H/V spectral ratios from earthquakes that were recorded by strong-motion monitoring networks at 449 sites within the Kansai area in the period 1994–2023 (up to 30 years of waveform recordings for some sites). We use three-component earthquake waveforms, which were recorded by surface stations operated by local governments (Osaka-fu, Kyoto-fu, Kyoto-shi, Nara-ken, Shiga-ken, Hyogo-ken, Wakayama-ken), the National Research Institute for Earth Science and Disaster Resilience (NIED K-NET, KiK-net), the Japan Meteorological Agency (JMA Shindo-kei), the Committee of Earthquake Observation and Research in the Kansai Area (CEORKA), and stations operated by the Disaster Prevention Research Institute (DPRI) of Kyoto University. For each event-station pair, we computed H/V spectral curves for transversal, radial, E-W, N-S, SE-NW, and NE-SW horizontal components of ground motion, and we evaluated frequency bands, in which earthquake spectra have values at least three times larger than the local noise spectrum. Then, the spectral curves were statistically evaluated to obtain site-specific representative H/V ratios including spatial directionality (predominant azimuths) of H/V peaks. Finally, we created a database of site-specific H/V earthquake spectral curves within the frequency range of 0.08–30 Hz. The frequency range is limited by the sufficient signal-to-noise ratio of observed earthquake spectra and the technical parameters of the recording system. The H/V curves above approximately 10 Hz may, but may not, be influenced by sensor housing effects; still, fundamental H/V peaks have rather low frequencies in deep sedimentary basins, which are retrieved with high quality by our approach.
The inferred frequency and directionality of fundamental H/V peaks can be spatially correlated with the sedimentary layer thickness (bedrock depth below the station) and disposition (position of the station in the sedimentary basin), respectively. The resultant map of empirical long-period fundamental resonance frequency can be useful for the identification of vulnerable high-rise buildings (fundamental frequency 0.13-0.25 Hz in Osaka, and 0.3-0.6 Hz in Kyoto and Nara basins) that may resonate with local strong ground motions caused by future large earthquakes such as the Nankai megathrust earthquake.
Acknowledgements: This research is supported by the Japan Society for the Promotion of Science (JSPS) through Fellowship number P23070 and Grant-in-Aid for JSPS Fellows number 23KF0149. We use waveforms from seismic stations operated by the NIED, JMA, CEORKA, DPRI, and local governments.
In this study, we evaluate H/V spectral ratios from earthquakes that were recorded by strong-motion monitoring networks at 449 sites within the Kansai area in the period 1994–2023 (up to 30 years of waveform recordings for some sites). We use three-component earthquake waveforms, which were recorded by surface stations operated by local governments (Osaka-fu, Kyoto-fu, Kyoto-shi, Nara-ken, Shiga-ken, Hyogo-ken, Wakayama-ken), the National Research Institute for Earth Science and Disaster Resilience (NIED K-NET, KiK-net), the Japan Meteorological Agency (JMA Shindo-kei), the Committee of Earthquake Observation and Research in the Kansai Area (CEORKA), and stations operated by the Disaster Prevention Research Institute (DPRI) of Kyoto University. For each event-station pair, we computed H/V spectral curves for transversal, radial, E-W, N-S, SE-NW, and NE-SW horizontal components of ground motion, and we evaluated frequency bands, in which earthquake spectra have values at least three times larger than the local noise spectrum. Then, the spectral curves were statistically evaluated to obtain site-specific representative H/V ratios including spatial directionality (predominant azimuths) of H/V peaks. Finally, we created a database of site-specific H/V earthquake spectral curves within the frequency range of 0.08–30 Hz. The frequency range is limited by the sufficient signal-to-noise ratio of observed earthquake spectra and the technical parameters of the recording system. The H/V curves above approximately 10 Hz may, but may not, be influenced by sensor housing effects; still, fundamental H/V peaks have rather low frequencies in deep sedimentary basins, which are retrieved with high quality by our approach.
The inferred frequency and directionality of fundamental H/V peaks can be spatially correlated with the sedimentary layer thickness (bedrock depth below the station) and disposition (position of the station in the sedimentary basin), respectively. The resultant map of empirical long-period fundamental resonance frequency can be useful for the identification of vulnerable high-rise buildings (fundamental frequency 0.13-0.25 Hz in Osaka, and 0.3-0.6 Hz in Kyoto and Nara basins) that may resonate with local strong ground motions caused by future large earthquakes such as the Nankai megathrust earthquake.
Acknowledgements: This research is supported by the Japan Society for the Promotion of Science (JSPS) through Fellowship number P23070 and Grant-in-Aid for JSPS Fellows number 23KF0149. We use waveforms from seismic stations operated by the NIED, JMA, CEORKA, DPRI, and local governments.