5:15 PM - 6:45 PM
[SSS07-P12] Peak delay of seismic wave in the backarc side of northeast Japan
Keywords:Scattering, Inhomogeneity, Envelope broadening, Northeast Japan, The backarc side
If the Earth's interior is homogeneous, the observed waveforms should consist mainly of only direct P-wave and direct S-wave. In reality, however, an impulsive wave collapses and coda waves are observed. This is mainly due to seismic wave scattering caused by the small-scale inhomogeneous structure of the Earth's interior. In the previous studies, it has known that envelope broadening in the S-wave part (hereinafter referred to as S-wave envelope broadening) well characterizes the short-wavelength inhomogeneity that causes seismic wave scattering. Although there have been many studies analyzing earthquakes on the Pacific side of northeastern Japan, there seems to be no examples of studies on earthquakes that occurred on the Sea of Japan side. However, even for earthquakes occurring in the Sea of Japan, we observed that seismogram waveform collapses significantly, causing S-wave envelope broadening even without crossing the volcanic front. In this study, we analyzed the waveforms of earthquakes that occurred in the Sea of Japan in northeastern Japan to investigate the regionality of small-scale inhomogeneity on the backarc side of northeastern Japan.
In this study, we used the earthquake catalog of the Japan Meteorological Agency, and seismic velocity traces from the high-sensitivity seismic observation network (Hi-net) of the National Research Institute for Earth Science and Disaster Prevention. We analyzed earthquakes with magnitudes between 3.0 and 5.5 and depths equal or less than 70 km that occurred on the Sea of Japan of around the Tohoku region between April 2004 and April 2023. For these earthquakes, we used Hi-net stations with hypocentral distances between 100 km and 250 km. Following the previous studies, we used peak delay time as a parameter to characterize the S-wave envelope broadening. We measured the arrival time of the maximum amplitude of the smoothed envelope waveform for each of four frequency bands: 2-4, 4-8, 8-16, and 16-32 Hz. Then, the peak delay time was estimated by subtracting the S-wave theoretical travel time predicted from the JMA2001 model. In addition, we determined the logarithmic deviation from the regression of peak delay time against hypocentral distance. We investigated the regional nature of the detailed inhomogeneity structure by representing it in a two-dimensional spatial distribution in which the area of interest is divided into blocks with a size of 10 km.
We confirmed that the peak delay time depends on the hypocentral distance and path in all frequency bands. In particular, the peak delay times were larger off Akita and Yamagata prefectures and along paths beyond the volcanic fronts. On the other hand, the dependence of frequency was very weak in our measurement on both the fore-arc and the backarc sides of the volcanic front. The logarithmic deviation of peak delay time was large in some areas off Akita and Yamagata prefectures. It is noteworthy that the locations of large and small values were in good agreement with the locations of volcanoes, and with gaps between the volcanoes on the volcanic front, respectively. Large logarithmic deviation of the peak delay implies the existence of short-wavelength inhomogeneity, and the results around the volcanoes are consistent with the previous studies. However, in some cases, the arrival time of the S-wave, an important indicator in peak delay analysis, deviated from the theoretical travel time considerably. In such cases, it is not possible to distinguish whether the arrival time of the S-wave is delayed, or the peak of the S-wave is delayed in our measurement. Therefore, it is expected that the accuracy of peak delay time can be improved by precise travel time evaluation.
In this study, we used the earthquake catalog of the Japan Meteorological Agency, and seismic velocity traces from the high-sensitivity seismic observation network (Hi-net) of the National Research Institute for Earth Science and Disaster Prevention. We analyzed earthquakes with magnitudes between 3.0 and 5.5 and depths equal or less than 70 km that occurred on the Sea of Japan of around the Tohoku region between April 2004 and April 2023. For these earthquakes, we used Hi-net stations with hypocentral distances between 100 km and 250 km. Following the previous studies, we used peak delay time as a parameter to characterize the S-wave envelope broadening. We measured the arrival time of the maximum amplitude of the smoothed envelope waveform for each of four frequency bands: 2-4, 4-8, 8-16, and 16-32 Hz. Then, the peak delay time was estimated by subtracting the S-wave theoretical travel time predicted from the JMA2001 model. In addition, we determined the logarithmic deviation from the regression of peak delay time against hypocentral distance. We investigated the regional nature of the detailed inhomogeneity structure by representing it in a two-dimensional spatial distribution in which the area of interest is divided into blocks with a size of 10 km.
We confirmed that the peak delay time depends on the hypocentral distance and path in all frequency bands. In particular, the peak delay times were larger off Akita and Yamagata prefectures and along paths beyond the volcanic fronts. On the other hand, the dependence of frequency was very weak in our measurement on both the fore-arc and the backarc sides of the volcanic front. The logarithmic deviation of peak delay time was large in some areas off Akita and Yamagata prefectures. It is noteworthy that the locations of large and small values were in good agreement with the locations of volcanoes, and with gaps between the volcanoes on the volcanic front, respectively. Large logarithmic deviation of the peak delay implies the existence of short-wavelength inhomogeneity, and the results around the volcanoes are consistent with the previous studies. However, in some cases, the arrival time of the S-wave, an important indicator in peak delay analysis, deviated from the theoretical travel time considerably. In such cases, it is not possible to distinguish whether the arrival time of the S-wave is delayed, or the peak of the S-wave is delayed in our measurement. Therefore, it is expected that the accuracy of peak delay time can be improved by precise travel time evaluation.