11:00 AM - 1:00 PM
[SSS03-P04] Seismological Property and Structure beneath the Seafloor at the Hyuga-nada, Estimated Using Ocean Bottom Seismographs
Keywords:the Hyuga-nada, receiver function, slow earthquake, fluid, the Kyushu-Palau Ridge
Seafloor seismic observations can contribute to estimating the subsurface seismological structure around shallow plate boundaries and accretionary complexes hosting slow earthquake events, which are far from onshore stations. Receiver function (RF) analysis, which examines a possible velocity discontinuity beneath a seismic station, has been applied to teleseismic recordings of ocean bottom seismographs (OBS) as well as onshore stations (e.g., Audet, 2016; Akuhara et al., 2019), and succeeded in discussing specific hydrous conditions around shallow plate interfaces (Akuhara et al., 2017).
The Hyuga-nada, located at the western end of the Nankai subduction zone, now attracts our attention in terms of vigorous slow earthquake activities in shallower parts of the subduction zone (Yamashita et al., 2015, 2021; Tonegawa et al., 2020). These findings were derived from broadband or short-period OBS observations for about 10 to 18 months, carried out by several projects. The existence of the Kyushu-Palau Ridge (KPR) at the Hyuga-nada makes the structural features of the overriding plate complicated (e.g., Nishizawa et al., 2009). However, few studies have examined a velocity discontinuity around the KPR by computing receiver functions (Akuhara et al., 2020, SSJ Fall Meeting). By using the widely distributed networks of broadband and short-period OBSs in the projects, we investigate the seismological structure of the shallow subduction zone by multi-band receiver function analysis and aim to clarify the spatial characteristics of the region where slow earthquakes occur frequently.
We first correct the horizontal sensor orientation of OBS and evaluate the orientation uncertainty. We analyze the particle motion of a Rayleigh-wave seismogram originating from teleseismic events with an epicentral distance between 5° and 175°, by rotating horizontal components (Stachnik et al., 2012; Doran and Laske, 2017). The orientation uncertainty is defined as the half-width of a 95% confidence interval. Under this evaluation, the orientation uncertainty is less than 5° and 10° for broadband and short-period OBSs, respectively. The exceptions are short-period OBSs deployed in the shallower seafloor, which exhibit anomalously large uncertainties. Since the uncertainty in horizontal azimuth highly affects RF analysis, we only use the OBS with low orientation uncertainty to compute RFs.
We use the time-domain iterative deconvolution method developed by Ligorría and Ammon (1999) to compute RFs and apply a low-pass Gaussian filter in several corner frequencies in 0.6–2.0 Hz for calculation of multi-band RFs. The earthquake source for the RF analysis consists of the following two types based on Sawaki et al. (2021): regional deep-focus events with an epicentral distance within 11° that occurred in the Pacific Slab and teleseismic events with an epicentral distance between 30° and 90°.
One of the prominent features in OBS RFs is that the largest signal appears around 1–3 seconds, but a significant amplitude can rarely be found at zero seconds. The delay is due to the P-to-S conversions from the base of low-shear-velocity sediments beneath the sea floor and can be explained by synthetic RFs and H -κ stacking method (Akuhara et al., 2017). We will discuss the relationship between the difference in the crustal structure around and outside the KPR indicated by the RF phases and reported slow earthquake activities in the shallow portion.
[Acknowledgements]
We would like to appreciate the support for the seafloor observation projects by the Ministry of Education, Culture, Sports, Science, and Technology: “Research Project for Disaster Prevention on the great Earthquakes along the Nankai Trough” and “Science of Slow Earthquakes” in Grant-in-Aid for Scientific Research on Innovative Area.
The Hyuga-nada, located at the western end of the Nankai subduction zone, now attracts our attention in terms of vigorous slow earthquake activities in shallower parts of the subduction zone (Yamashita et al., 2015, 2021; Tonegawa et al., 2020). These findings were derived from broadband or short-period OBS observations for about 10 to 18 months, carried out by several projects. The existence of the Kyushu-Palau Ridge (KPR) at the Hyuga-nada makes the structural features of the overriding plate complicated (e.g., Nishizawa et al., 2009). However, few studies have examined a velocity discontinuity around the KPR by computing receiver functions (Akuhara et al., 2020, SSJ Fall Meeting). By using the widely distributed networks of broadband and short-period OBSs in the projects, we investigate the seismological structure of the shallow subduction zone by multi-band receiver function analysis and aim to clarify the spatial characteristics of the region where slow earthquakes occur frequently.
We first correct the horizontal sensor orientation of OBS and evaluate the orientation uncertainty. We analyze the particle motion of a Rayleigh-wave seismogram originating from teleseismic events with an epicentral distance between 5° and 175°, by rotating horizontal components (Stachnik et al., 2012; Doran and Laske, 2017). The orientation uncertainty is defined as the half-width of a 95% confidence interval. Under this evaluation, the orientation uncertainty is less than 5° and 10° for broadband and short-period OBSs, respectively. The exceptions are short-period OBSs deployed in the shallower seafloor, which exhibit anomalously large uncertainties. Since the uncertainty in horizontal azimuth highly affects RF analysis, we only use the OBS with low orientation uncertainty to compute RFs.
We use the time-domain iterative deconvolution method developed by Ligorría and Ammon (1999) to compute RFs and apply a low-pass Gaussian filter in several corner frequencies in 0.6–2.0 Hz for calculation of multi-band RFs. The earthquake source for the RF analysis consists of the following two types based on Sawaki et al. (2021): regional deep-focus events with an epicentral distance within 11° that occurred in the Pacific Slab and teleseismic events with an epicentral distance between 30° and 90°.
One of the prominent features in OBS RFs is that the largest signal appears around 1–3 seconds, but a significant amplitude can rarely be found at zero seconds. The delay is due to the P-to-S conversions from the base of low-shear-velocity sediments beneath the sea floor and can be explained by synthetic RFs and H -κ stacking method (Akuhara et al., 2017). We will discuss the relationship between the difference in the crustal structure around and outside the KPR indicated by the RF phases and reported slow earthquake activities in the shallow portion.
[Acknowledgements]
We would like to appreciate the support for the seafloor observation projects by the Ministry of Education, Culture, Sports, Science, and Technology: “Research Project for Disaster Prevention on the great Earthquakes along the Nankai Trough” and “Science of Slow Earthquakes” in Grant-in-Aid for Scientific Research on Innovative Area.