10:45 AM - 12:15 PM
[SCG45-P31] Interaction between ordinary earthquakes and shallow tectonic tremors in the northern Japan Trench before and after the 2011 Tohoku-Oki earthquake
Keywords:tectonic tremor, repeating earthquake, OBS, VLFE, Tohoku-oki earthquake, slow slip
In many subduction zones, slow earthquakes are observed in the shallow and deep sides of the megathrust seismogenic zone, whereas in the Japan Trench subduction zone, tectonic tremors (hereinafter referred to as tremors), a type of slow earthquakes, and ordinary earthquakes (hereinafter referred to as earthquakes) are distributed in the same depth range, and in close proximity. Matsumoto et al. [2022] investigated the spatio-temporal correlation between tremors and earthquakes from October 2007 to June 2008 along the shallow plate boundary in the northern Japan Trench, and found that swarms of earthquakes associated with tremor bursts and both of them had a common migration pattern. In addition, there were cases of VLFEs and repeating earthquakes occurring during the earthquake and tremor activations. Therefore, it is highly likely that aseismic slip triggered these activities. In this study, we analyzed tremors and earthquakes in the area from 2016 to 2022 in order to clarify whether the same kind of interaction is observed after the 2011 Tohoku-Oki earthquake.
We used the event list prepared by Ohyanagi et al. [2022]. They applied the envelope correlation method [Mizuno and Ide, 2019] and one-dimensional S-wave velocity structure model [JMA2001; Ueno et al., 2002] to continuous waveforms from August 2016 to July 2022 at the S-net stations in the northern Japan Trench to detect seismic events. From the detected events,they removed those corresponding to signals from seismic explorations. On the other hand,they applied a deep learning based event detection [Mousavi et al., 2020] to the seismograms and identified earthquakes as the events with durations less than 20 seconds. From the events detected by the envelope correlation method, the detected earthquakes were excluded, and those with a duration of 20 seconds or more were designated as tremors after applying spatio-temporal clustering [e.g. Obara et al., 2010].
Compared to the activities prior to the 2011 Tohoku-Oki earthquake, the frequency of tremor burst increased whereas the characteristics of the spatial distribution of tremors and earthquakes have not changed, indicating that the spatial segregation of tremor and earthquake activities was maintained. We found simultaneous activations of tremors and regular earthquakes, including repeating earthquakes by S-net observation after 2016, as observed in 2007 and 2008. This characteristic indicates that tremor and earthquake associations with aseismic slip events are the persistent nature of the area before and after the Tohoku-Oki earthquake. However, we also recognized that not all of the tremor bursts were accompanied by significant seismic activity after 2016. In this region, the aseismic slip velocity is considered to be faster than that in 2007-2008 due to the afterslip of the Tohoku-Oki earthquake. We consider that tremor and earthquake association are caused by the activation of tremors and earthquakes due to the transient slip acceleration. We speculate that the response of the activity of regular earthquakes to the slip acceleration becomes less prominent than that of tremors, under the larger background slip rate due to the afterslip to explain the difference in the association pattern of tremors and earthquakes before and after the Tohoku-Oki earthquake.
We used the event list prepared by Ohyanagi et al. [2022]. They applied the envelope correlation method [Mizuno and Ide, 2019] and one-dimensional S-wave velocity structure model [JMA2001; Ueno et al., 2002] to continuous waveforms from August 2016 to July 2022 at the S-net stations in the northern Japan Trench to detect seismic events. From the detected events,they removed those corresponding to signals from seismic explorations. On the other hand,they applied a deep learning based event detection [Mousavi et al., 2020] to the seismograms and identified earthquakes as the events with durations less than 20 seconds. From the events detected by the envelope correlation method, the detected earthquakes were excluded, and those with a duration of 20 seconds or more were designated as tremors after applying spatio-temporal clustering [e.g. Obara et al., 2010].
Compared to the activities prior to the 2011 Tohoku-Oki earthquake, the frequency of tremor burst increased whereas the characteristics of the spatial distribution of tremors and earthquakes have not changed, indicating that the spatial segregation of tremor and earthquake activities was maintained. We found simultaneous activations of tremors and regular earthquakes, including repeating earthquakes by S-net observation after 2016, as observed in 2007 and 2008. This characteristic indicates that tremor and earthquake associations with aseismic slip events are the persistent nature of the area before and after the Tohoku-Oki earthquake. However, we also recognized that not all of the tremor bursts were accompanied by significant seismic activity after 2016. In this region, the aseismic slip velocity is considered to be faster than that in 2007-2008 due to the afterslip of the Tohoku-Oki earthquake. We consider that tremor and earthquake association are caused by the activation of tremors and earthquakes due to the transient slip acceleration. We speculate that the response of the activity of regular earthquakes to the slip acceleration becomes less prominent than that of tremors, under the larger background slip rate due to the afterslip to explain the difference in the association pattern of tremors and earthquakes before and after the Tohoku-Oki earthquake.