11:00 AM - 1:00 PM
[SCG52-P06] Comparison of non-double-couple earthquake and double-couple earthquake using 0.1 manten hyperdense seismic observation data
Keywords:non-double-couple, the 2000 Western Tottori earthquake, focal mechanism
Currently, the force system that causes fault rupture is considered the double-couple (DC) model which forces along the source fault and forces in the direction orthogonal to the source fault act with the same magnitude. In this case, P-wave radiation patterns with a four-leaf pattern are observed, with the direction of the force acting as a node. It has been reported that this pattern is common at low frequencies, but not at high frequencies, and the rupture process affecting P-wave radiation patterns differ depending on the frequency (Castro et al. 2006). They pointed out that the characteristics of the non-double-couple (NDC) component are expressed at higher frequencies. On the other hand, at higher frequencies, the scattering phenomenon disrupt P-wave radiation patterns with a four-leaf pattern.The purpose of this study is to discuss the causes that cannot be expressed P-wave radiation patterns with the double-couple (DC) model at higher frequencies, comparing of non-double-couple earthquake’s RMS and double-couple earthquake’s RMS using 0.1 manten hyperdense seismic observation data.
In this study, we used the data from 0.1 manten hyperdense seismic observation data located in Tottori, Shimane, and Okayama prefectures. We used 2 non-double-couple earthquakes (magnitude, M = 2.7, 2.2) and 2 double-couple earthquakes (M = 2.9, 2.1) recorded there from April to November 2017.
The analysis method is explained below. We used only those waveform data of the waveform data from all stations within 60 seconds of the earthquake, for which the arrival of P-and S-waves was automatically read. The waveforms were classified into the following frequency bands: 1-2, 2-4, 4-8, 8-16, and 16-32 Hz. We calculated the root-mean-square amplitude (RMS) for 0.5 seconds from the time of P-wave arrival for each station and frequency band. Ts is the time between the occurrence of the earthquake and the slowest arrival of the S-wave among all the stations. To remove site effects at each station, we calculated the root-mean-square-amplitude (S_RMS) for 3 seconds from 2Ts seconds after the earthquake occurred for each station and frequency band. We calculated the A_RMS which considered site effects by dividing RMS by S_RMS for each station and frequency band. Then, we calculated the root-mean-square amplitude (N_RMS) for 3 seconds from 5 seconds to P-wave arrival for each station and frequency band. To remove waveform with high noise, we did not use data when doubled N_RMS was greater than S_RMS. This A_RMS was classified in terms of epicentral distance as 0-3, 3-6, 6-9, 9-12, 12-15, 15-18, and 18-21 km. Also, this A_RMS was classified in terms of take-off angle classified as 90-100, 100-110, 110-120, 120-130, 130-140, 140-150, 150-160, 160-170, and 170-180°. We graphed each classified data in polar graph, with the origin as the epicenter. The distance from the origin showed A_RMS, and the angle showed azimuth.
We calculated ideal data by dividing P-wave radiation orientation dependence by hypocentral distance. After that, we standardized ideal data that the average of A_RMS coincides with the average of ideal data for each station and frequency band.Also, we calculated the correlation coefficient of A_RMS and ideal data at each classified data, and we showed scatter diagram of A_RMS and standardized ideal data.
For results, Comparing the polar graphs of the DC and NDC earthquakes at different frequencies for the same take-off angle, it was confirmed that the pattern was more disrupted at the NDC earthquake than the DC earthquake at high frequencies.
We considered that complexity in the source process provides this result.
In this study, we used the data from 0.1 manten hyperdense seismic observation data located in Tottori, Shimane, and Okayama prefectures. We used 2 non-double-couple earthquakes (magnitude, M = 2.7, 2.2) and 2 double-couple earthquakes (M = 2.9, 2.1) recorded there from April to November 2017.
The analysis method is explained below. We used only those waveform data of the waveform data from all stations within 60 seconds of the earthquake, for which the arrival of P-and S-waves was automatically read. The waveforms were classified into the following frequency bands: 1-2, 2-4, 4-8, 8-16, and 16-32 Hz. We calculated the root-mean-square amplitude (RMS) for 0.5 seconds from the time of P-wave arrival for each station and frequency band. Ts is the time between the occurrence of the earthquake and the slowest arrival of the S-wave among all the stations. To remove site effects at each station, we calculated the root-mean-square-amplitude (S_RMS) for 3 seconds from 2Ts seconds after the earthquake occurred for each station and frequency band. We calculated the A_RMS which considered site effects by dividing RMS by S_RMS for each station and frequency band. Then, we calculated the root-mean-square amplitude (N_RMS) for 3 seconds from 5 seconds to P-wave arrival for each station and frequency band. To remove waveform with high noise, we did not use data when doubled N_RMS was greater than S_RMS. This A_RMS was classified in terms of epicentral distance as 0-3, 3-6, 6-9, 9-12, 12-15, 15-18, and 18-21 km. Also, this A_RMS was classified in terms of take-off angle classified as 90-100, 100-110, 110-120, 120-130, 130-140, 140-150, 150-160, 160-170, and 170-180°. We graphed each classified data in polar graph, with the origin as the epicenter. The distance from the origin showed A_RMS, and the angle showed azimuth.
We calculated ideal data by dividing P-wave radiation orientation dependence by hypocentral distance. After that, we standardized ideal data that the average of A_RMS coincides with the average of ideal data for each station and frequency band.Also, we calculated the correlation coefficient of A_RMS and ideal data at each classified data, and we showed scatter diagram of A_RMS and standardized ideal data.
For results, Comparing the polar graphs of the DC and NDC earthquakes at different frequencies for the same take-off angle, it was confirmed that the pattern was more disrupted at the NDC earthquake than the DC earthquake at high frequencies.
We considered that complexity in the source process provides this result.