5:15 PM - 7:15 PM
[SSS08-P03] Three-dimensional S-wave attenuation structure around the Hida Mountains
Keywords:S-wave attenuation, Tomography, Hida Mountains
It has long been known that seismic waves attenuate significantly around the Hida Mountains. For example, Mizoue et al. (1983) pointed out that the amplitude of seismic waves from the west side of the Hida Mountains attenuated significantly at stations in the east. To understand this anomalous attenuation of seismic waves in and around the Hida Mountains, quantitative and three-dimensional analysis is necessary. However, few previous tomographic studies have focused on the Hida Mountains, and have not clarified high attenuation regions in the crust. Therefore, in this study, Qs values, which represent the degree of attenuation, are estimated by tomographic inversion to clarify the three-dimensional attenuation structure in and around the Hida Mountains.
The data used in this analysis consist of acceleration strong-motion records from the K-NET and KiK-net strong-motion seismographs of the National Research Institute for Earth Science and Disaster Resilience, event waveform data from the Hi-net seismograph, and waveform data from temporarily observation. The latitude, longitude, depth, and moment magnitude from F-net solutions are also used. A total of 436 earthquakes from K-NET and Kik-net, 727 earthquakes from Hi-net, and 288 earthquakes from temporary stations are used for the analysis.
In this study, we estimated Qs values in and around the Hida Mountains using the three-dimensional tomographic inversion method propsed by Nakamura (2009). The size of one block was set to be 0.1° in latitude × 0.1° in longitude × 10 km in depth. The acceleration Fourier spectrum at observation point i due to earthquake j can be expressed as follows :
αij(f) = Saj(f)･Ge･(ρj βj / ρSB βSB)0.5･gi(f)･exp(-πfΣTijk/Qk(f)),
where Saj(f) is the source acceleration Fourier spectrum, Ge is the reciprocal of the source distance, ρSB and βSB are the density and S-wave velocity at the seismic basement, ρj and βj are the density and S-wave velocity at the source respectively, gi is the site amplification effects at the observation point, and Qk(f) is the Qs value. Tijk is the travel time through the block. The LSQR program (Paige and Sunders, 1982) was used to estimate -πf/Q(f)k for each block (Figure). In addition, a checkerboard resolution test was conducted to evaluate the resolution.
At the depth of 0-10 km, anomalous attenuation was observed around the Hida Mountains. In the depth range of 10-20 km, an anomalous attenuation area is observed between Mt. Tateyama, Mt. Yake-dake, and Mt. Norikura-dake. In the 20-40 km depth interval, a high-attenuation region was observed on the dorsal arc side of the volcanic front. In the depth range of 40-60 km, a low attenuation region extends from Sagami Bay to the north, which would be due to the subducting PHS slab.
The Hida Mountains are one of the locations where the anomalous attenuation zone at the depth of 0-10 km coincides with a region of high uplift (Fujiwara et al., 2005). It is possible that the high uplift velocity creates large cracks in the upper crust, and the fluid entering the cracks may cause the anomalous attenuation zone. In fact, Genauchi et al. (2002) pointed out that the ultra-low density area beneath the Hida Mountains cannot explained solely by rock melting.
Acknowledgements
Observation data from K-NET, Kik-net and Hi-net and mechanism solutions from F-net were used in this study.
The data used in this analysis consist of acceleration strong-motion records from the K-NET and KiK-net strong-motion seismographs of the National Research Institute for Earth Science and Disaster Resilience, event waveform data from the Hi-net seismograph, and waveform data from temporarily observation. The latitude, longitude, depth, and moment magnitude from F-net solutions are also used. A total of 436 earthquakes from K-NET and Kik-net, 727 earthquakes from Hi-net, and 288 earthquakes from temporary stations are used for the analysis.
In this study, we estimated Qs values in and around the Hida Mountains using the three-dimensional tomographic inversion method propsed by Nakamura (2009). The size of one block was set to be 0.1° in latitude × 0.1° in longitude × 10 km in depth. The acceleration Fourier spectrum at observation point i due to earthquake j can be expressed as follows :
αij(f) = Saj(f)･Ge･(ρj βj / ρSB βSB)0.5･gi(f)･exp(-πfΣTijk/Qk(f)),
where Saj(f) is the source acceleration Fourier spectrum, Ge is the reciprocal of the source distance, ρSB and βSB are the density and S-wave velocity at the seismic basement, ρj and βj are the density and S-wave velocity at the source respectively, gi is the site amplification effects at the observation point, and Qk(f) is the Qs value. Tijk is the travel time through the block. The LSQR program (Paige and Sunders, 1982) was used to estimate -πf/Q(f)k for each block (Figure). In addition, a checkerboard resolution test was conducted to evaluate the resolution.
At the depth of 0-10 km, anomalous attenuation was observed around the Hida Mountains. In the depth range of 10-20 km, an anomalous attenuation area is observed between Mt. Tateyama, Mt. Yake-dake, and Mt. Norikura-dake. In the 20-40 km depth interval, a high-attenuation region was observed on the dorsal arc side of the volcanic front. In the depth range of 40-60 km, a low attenuation region extends from Sagami Bay to the north, which would be due to the subducting PHS slab.
The Hida Mountains are one of the locations where the anomalous attenuation zone at the depth of 0-10 km coincides with a region of high uplift (Fujiwara et al., 2005). It is possible that the high uplift velocity creates large cracks in the upper crust, and the fluid entering the cracks may cause the anomalous attenuation zone. In fact, Genauchi et al. (2002) pointed out that the ultra-low density area beneath the Hida Mountains cannot explained solely by rock melting.
Acknowledgements
Observation data from K-NET, Kik-net and Hi-net and mechanism solutions from F-net were used in this study.