10:45 AM - 12:15 PM
[SSS10-P04] Detection of immediate aftershocks following the 2011 Tohoku earthquake based on MeSO-net
Keywords:2011 Tohoku earthquake, aftershock
Very significant aftershock activity was driven by the 2011 Tohoku earthquake. It is considered that many small- and medium-events during the immediate aftershocks were missed from existing earthquake catalogs because the wavefields generated by large-events masked their wavefields. In fact, the number of small- and medium-events recorded in JMA’s earthquake catalog is fewer than that expected by the Gutenberg-Richter (GR) law. One of the methods to recover the missed events is template-matching technique, which has been used by many previous studies to detect a lot of events and understand seismicity (e.g., Peng and Zhao, 2009; Kato and Obara, 2014). This method detects events that have similar waveforms to known events from continuous waveform records. Actually, we use waveform records of known events as templates and calculate the correlation coefficient between templates and continuous waveform records. When the coefficients exceeded a predefined threshold, we regard that an event occurred at that time. Lengline et al. (2011) applied the method to the aftershocks of the 2011 Tohoku earthquake, using waveforms of Hi-net stations in the eastern Japan. However, the Hi-net has a problem that its detection capability gradually degraded over time due to an increasing number of stations that stopped operating as a result of power outage or tsunami.
Therefore, we attempt to detect immediate aftershocks following the 2011 Tohoku earthquake by performing the template-matching using MeSO-net records. MeSO-net is a seismic network located in Kanto region, relatively close to the hypocentral area. We expect that MeSO-net has time-stable detection capability, since most of the stations continued observation without stopping their operation. In order to begin this study, we visually checked waveforms of each large aftershock, and initially selected 1-8 Hz as the frequency range for the band-pass filter. Next, we calculated the theoretical S-wave travel time from the epicenter to each station in the network for each known event. The time window of the template was set from 3 seconds before to 7 seconds after the S-wave arrival time to account for the difference between the theoretical and actual travel time. We are now detecting events, and will show the results.
Therefore, we attempt to detect immediate aftershocks following the 2011 Tohoku earthquake by performing the template-matching using MeSO-net records. MeSO-net is a seismic network located in Kanto region, relatively close to the hypocentral area. We expect that MeSO-net has time-stable detection capability, since most of the stations continued observation without stopping their operation. In order to begin this study, we visually checked waveforms of each large aftershock, and initially selected 1-8 Hz as the frequency range for the band-pass filter. Next, we calculated the theoretical S-wave travel time from the epicenter to each station in the network for each known event. The time window of the template was set from 3 seconds before to 7 seconds after the S-wave arrival time to account for the difference between the theoretical and actual travel time. We are now detecting events, and will show the results.