3:00 PM - 3:15 PM
[SCG53-06] Near real-time estimation of hypocenter distribution from seismic amplitudes
Keywords:seismic amplitude, hypocenter distribution, real-time analysis
Hypocenter distribution is important information for monitoring seismicity. In general, arrival times of seismic waves are used to determine hypocenters, but it is difficult to read arrival times when seismicity becomes active, resulting in a decrease in the numbers of hypocenters.
The amplitude source location (ASL) method uses seismic amplitudes to determine hypocenters. The ASL method does not require arrival times, but the precision of hypocenters is usually worse compared with that of arrival-time based hypocenters. Ogiso and Yomogida (2021, EPS) has proposed a method to estimate relative source locations of seismic events from seismic amplitude ratios. The relative source locations by their method show a good agreement with those estimated using phase arrival times. Hence, if the method of Ogiso and Yomogida (2021) could be applied in near real-time, the hypocenter distribution derived by the method would complement that derived by phase-arrival-time based automated hypocenter location method (e.g., Tamaribuchi, 2018, EPS), and the method would be useful for monitoring seismicities in real-time.
First, we applied the method to the continuous waveforms recorded on 29 September 2008 at Meakandake volcano. A volcanic tremor that occurred on 14:11 had activated seismicity at the volcano. The number of earthquakes that occurred from 14:00 to 16:00 was about 230 according to Sapporo District Meteorological Observatory, in which 34 earthquakes were estimated there hypocenters. We successfully estimated hypocenters of 103 earthquakes in that time window by applying the method.
Next, we conducted a synthetic test to investigate the applicability of the method in a wider region than in the case of volcanoes. The test was conducted in a focal region of the 2016 Kumamoto earthquake. At least an epicenter can be appropriately estimated for the earthquake about 26 km far from the reference event if we calculated amplitude at each station appropriately, that is, amplitude calculation considering the difference of arrival times between the reference and target events. This result implies that the method might be useful for hypocenter determination of aftershocks automatically after a large earthquake with an appropriate correction of arrival times when measuring the amplitude of each subevent at each station.
The amplitude source location (ASL) method uses seismic amplitudes to determine hypocenters. The ASL method does not require arrival times, but the precision of hypocenters is usually worse compared with that of arrival-time based hypocenters. Ogiso and Yomogida (2021, EPS) has proposed a method to estimate relative source locations of seismic events from seismic amplitude ratios. The relative source locations by their method show a good agreement with those estimated using phase arrival times. Hence, if the method of Ogiso and Yomogida (2021) could be applied in near real-time, the hypocenter distribution derived by the method would complement that derived by phase-arrival-time based automated hypocenter location method (e.g., Tamaribuchi, 2018, EPS), and the method would be useful for monitoring seismicities in real-time.
First, we applied the method to the continuous waveforms recorded on 29 September 2008 at Meakandake volcano. A volcanic tremor that occurred on 14:11 had activated seismicity at the volcano. The number of earthquakes that occurred from 14:00 to 16:00 was about 230 according to Sapporo District Meteorological Observatory, in which 34 earthquakes were estimated there hypocenters. We successfully estimated hypocenters of 103 earthquakes in that time window by applying the method.
Next, we conducted a synthetic test to investigate the applicability of the method in a wider region than in the case of volcanoes. The test was conducted in a focal region of the 2016 Kumamoto earthquake. At least an epicenter can be appropriately estimated for the earthquake about 26 km far from the reference event if we calculated amplitude at each station appropriately, that is, amplitude calculation considering the difference of arrival times between the reference and target events. This result implies that the method might be useful for hypocenter determination of aftershocks automatically after a large earthquake with an appropriate correction of arrival times when measuring the amplitude of each subevent at each station.