5:15 PM - 6:30 PM
[SCG39-P10] Scaled energy of shallow slow earthquakes in Hyuga-nada, southwest Japan
Keywords:Slow earthquake, Low frequency tremor
Slow earthquakes, such as tremors, very low frequency earthquakes (VLFEs), and slow slip events, are fault slips with slower slip speed compared to regular earthquakes. Shallow slow earthquakes, which occur in the shallower extensions of seismogenic zones near trench axes, are very active in Hyuga-nada (Asano et al., 2008). Yamashita et al. (2015; 2017) revealed the spatiotemporal distribution of shallow tremors by using temporal ocean bottom seismometers (OBSs) deployed in Hyuga-nada in 2013 and 2015, respectively. To obtain more physical characteristics of shallow slow earthquakes, scaled energy of shallow slow earthquakes has been discussed by Yabe et al. (2019; 2021) in Nankai, off Tokachi and Tohoku. In this study, to clarify the characteristics of shallow slow earthquakes in Hyuga-nada and the differences with other regions, we estimated scaled energies using tremors of Yamashita et al. (2015; 2017) and accompanying VLFEs.
For the tremor analysis, we used OBS records in 2013 and 2015 (Yamashita et al., 2015; 2017), applied a band-pass filter of 2–8 Hz, and calculated root-mean-square envelope. After estimating site amplifications and attenuations based on the coda-normalization method, we estimated energy rate functions of tremors at each station by the method of Maeda and Obara (2009). The energy rate function of an event is estimated by averaging energy rate functions calculated at each station. The duration of the event is defined as the time window in which the value of the energy rate function exceeded 0.2 times the maximum value. The energy rate of the event is calculated by the ratio of integration of the energy rate function to the duration.
Next, we estimated seismic moments and source durations of VLFEs which correspond to tremors using NIED F-net seismograms in a frequency range of 0.02-0.05 Hz. We calculated Green’s functions from the assumed source grids on the plate boundary by the OpenSWPC; (Maeda et al., 2017) using a three-dimensional velocity structure model, JIVSM (Koketsu et al., 2012). Based on the method of Yabe, Baba et al. (2021), the source duration with the highest cross-correlation coefficients between observed and synthetic waveforms was selected for each event and seismic moments were estimated by the relative amplitudes to synthetic waveforms. Scaled energy is calculated by the ratio of the seismic energy rate of a tremor to the seismic moment rate of accompanying VLFE.
The ranges of energy rates of tremors and moment rates of VLFEs are 102–105 J/s and 1013–1015 Nm/s, respectively. In 2013, tremors occurred from 30.3ºN to 31.7ºN, and events with large energy rates and large moment rates were concentrated in the south of 31ºN. Around 31.0ºN, the trench is bending, and the Kyushu-Palau ridge is subducting. The relationship between the geometrical characteristics and VLFE activity in Hyuga-nada was suggested in Tonegawa et al. (2020). In 2015, tremors occurred in the north of 31.0ºN, and events with large energy rates and large moment rates tended to locate in the east of 132.6ºE, which is the shallower part of the plate boundary. The cumulative energy of tremors is larger in the 2015 episode than in the 2013 episode. The range of scaled energy in Hyuga-nada is 10-12–10-10, which is one to three orders smaller than that of shallow Nankai and shallow Tohoku. Smaller scaled energy may be related to weak interplate locking or complex geometry in Hyuga-nada.
Acknowledgements: The temporal seismological observations in Hyuga-nada were conducted as part of "Research project for compound disaster mitigation on the great earthquakes and tsunamis around the Nankai Trough region," a project of Ministry of Education, Culture, Sports, Science and Technology. We used NIED F-net broadband seismograms (https://doi.org/10.17598/NIED.0005). Numerical simulations were performed on the Fujitsu PRIMERGY CX600M1/CX1640M1 (Oakforest-PACS) at the Information Technology Center, the University of Tokyo.
For the tremor analysis, we used OBS records in 2013 and 2015 (Yamashita et al., 2015; 2017), applied a band-pass filter of 2–8 Hz, and calculated root-mean-square envelope. After estimating site amplifications and attenuations based on the coda-normalization method, we estimated energy rate functions of tremors at each station by the method of Maeda and Obara (2009). The energy rate function of an event is estimated by averaging energy rate functions calculated at each station. The duration of the event is defined as the time window in which the value of the energy rate function exceeded 0.2 times the maximum value. The energy rate of the event is calculated by the ratio of integration of the energy rate function to the duration.
Next, we estimated seismic moments and source durations of VLFEs which correspond to tremors using NIED F-net seismograms in a frequency range of 0.02-0.05 Hz. We calculated Green’s functions from the assumed source grids on the plate boundary by the OpenSWPC; (Maeda et al., 2017) using a three-dimensional velocity structure model, JIVSM (Koketsu et al., 2012). Based on the method of Yabe, Baba et al. (2021), the source duration with the highest cross-correlation coefficients between observed and synthetic waveforms was selected for each event and seismic moments were estimated by the relative amplitudes to synthetic waveforms. Scaled energy is calculated by the ratio of the seismic energy rate of a tremor to the seismic moment rate of accompanying VLFE.
The ranges of energy rates of tremors and moment rates of VLFEs are 102–105 J/s and 1013–1015 Nm/s, respectively. In 2013, tremors occurred from 30.3ºN to 31.7ºN, and events with large energy rates and large moment rates were concentrated in the south of 31ºN. Around 31.0ºN, the trench is bending, and the Kyushu-Palau ridge is subducting. The relationship between the geometrical characteristics and VLFE activity in Hyuga-nada was suggested in Tonegawa et al. (2020). In 2015, tremors occurred in the north of 31.0ºN, and events with large energy rates and large moment rates tended to locate in the east of 132.6ºE, which is the shallower part of the plate boundary. The cumulative energy of tremors is larger in the 2015 episode than in the 2013 episode. The range of scaled energy in Hyuga-nada is 10-12–10-10, which is one to three orders smaller than that of shallow Nankai and shallow Tohoku. Smaller scaled energy may be related to weak interplate locking or complex geometry in Hyuga-nada.
Acknowledgements: The temporal seismological observations in Hyuga-nada were conducted as part of "Research project for compound disaster mitigation on the great earthquakes and tsunamis around the Nankai Trough region," a project of Ministry of Education, Culture, Sports, Science and Technology. We used NIED F-net broadband seismograms (https://doi.org/10.17598/NIED.0005). Numerical simulations were performed on the Fujitsu PRIMERGY CX600M1/CX1640M1 (Oakforest-PACS) at the Information Technology Center, the University of Tokyo.