10:45 AM - 12:15 PM
[SCG45-P28] Detection of shallow tremors and investigation of their waveform characteristics by distributed acoustic sensing using offshore fiber-optic cable at the Nankai Trough, southwest Japan
Keywords:Distributed acoustic sensing (DAS), Slow earthquakes, Shallow tremor, Nankai Trough
Distributed acoustic sensing (DAS) measurement uses a fiber-optic cable as a strain sensor. Taking advantage of its high-density data, DAS measurement has been widely used for seismic observations recently. We conduct DAS measurement with an offshore fiber-optic cable off the Cape Muroto, along the Nankai Trough in southwest Japan in December 2019 and from January 2022. This area is a typical area with adjacent occurrences of megathrust and updip slow earthquakes. Recently, the relationship between slow and megathrust earthquakes has been pointed out (e.g., Obara and Kato, 2016), therefore, monitoring slow earthquakes with a high resolution is necessary to understand tectonic conditions in subduction zones. Although a lot of studies have observed regular earthquakes by DAS measurement recently, not many studies have observed slow earthquakes with DAS. In addition, the differences in waveforms and amplitudes between strain rate observed by DAS and displacement and velocity waveforms observed by seismometers were not discussed well. In this study, we observed shallow tremors, a type of slow earthquake in a frequency range of 2–10 Hz, by DAS measurement and investigated the characteristics of DAS strain waveforms of tremors.
We found 28 shallow tremor signals off Cape Muroto by using DAS from January 30 to February 8, 2022. The signals of these tremors were also observed in broadband seismograms of the Dense Oceanfloor Network system for Earthquake and Tsunami (DONET) data. We manually picked the arrivals of the tremor signals in root-mean-square envelopes of DAS strain and DONET velocity waveforms, and located the tremor events at the point where the residual between synthetic and observed arrival times is the least by the grid search. Synthetic travel times were calculated based on a one-dimensional S-wave velocity structure model representing the area near the Nankai Trough (Nakano et al., 2013). The tremors were located mainly around 135.7ºE and 33.8ºN, which corresponds to a subducted seamount peak indicated by Nakamura et al. (2022).
The amplitude of tremor signals in the frequency range of 2–10 Hz observed in a DAS channel is 1–2 nstrain. Assuming a plane wave, the velocity waveforms can be calculated by multiplying the apparent velocity by the strain waveform (e.g., Daley et al., 2016). The apparent velocity of the tremor signal propagation was estimated to be ~4 km/s in DAS data; therefore, the amplitudes of the velocities were estimated to be 4000–8000 nm/s. This amplitude range is the same as or one order larger than that in broadband seismometers of the nearest DONET stations. In detail, tremor signals in DAS data are composed of several phases with variable apparent velocities, and these phases are coherent within only 50–100 m. Generally, the duration of tremors observed in the DAS channels (40–60 s) is longer than that observed in velocity waveforms of DONET broadband seismometers (30–50 s).
To investigate if the long duration of shallow tremors in DAS is generally observed in regular earthquakes, we examined the DAS data in 2019, when ocean bottom seismometers (OBSs) were installed close to the DAS cable. Generally, the duration of regular earthquake signals is longer in the offshore area both in DAS and seismometer data (OBSs in 2019 and DONET). This long duration in the offshore area can be caused by the thick sediment. Comparing the waveforms of regular earthquakes in DAS and OBSs at the same location, we found that the duration of the regular earthquakes in the DAS strain waveform is also longer than that in the OBS velocity waveform. The amplitudes of coda waves in the DAS strain waveform are generally larger than those in OBSs. The difference in waveform characteristics may be caused by that in the sensitivity of the incident angle between DAS and OBS. To clarify the cause of such differences, numerical simulation of strain waveforms can be required in the future studies.
We found 28 shallow tremor signals off Cape Muroto by using DAS from January 30 to February 8, 2022. The signals of these tremors were also observed in broadband seismograms of the Dense Oceanfloor Network system for Earthquake and Tsunami (DONET) data. We manually picked the arrivals of the tremor signals in root-mean-square envelopes of DAS strain and DONET velocity waveforms, and located the tremor events at the point where the residual between synthetic and observed arrival times is the least by the grid search. Synthetic travel times were calculated based on a one-dimensional S-wave velocity structure model representing the area near the Nankai Trough (Nakano et al., 2013). The tremors were located mainly around 135.7ºE and 33.8ºN, which corresponds to a subducted seamount peak indicated by Nakamura et al. (2022).
The amplitude of tremor signals in the frequency range of 2–10 Hz observed in a DAS channel is 1–2 nstrain. Assuming a plane wave, the velocity waveforms can be calculated by multiplying the apparent velocity by the strain waveform (e.g., Daley et al., 2016). The apparent velocity of the tremor signal propagation was estimated to be ~4 km/s in DAS data; therefore, the amplitudes of the velocities were estimated to be 4000–8000 nm/s. This amplitude range is the same as or one order larger than that in broadband seismometers of the nearest DONET stations. In detail, tremor signals in DAS data are composed of several phases with variable apparent velocities, and these phases are coherent within only 50–100 m. Generally, the duration of tremors observed in the DAS channels (40–60 s) is longer than that observed in velocity waveforms of DONET broadband seismometers (30–50 s).
To investigate if the long duration of shallow tremors in DAS is generally observed in regular earthquakes, we examined the DAS data in 2019, when ocean bottom seismometers (OBSs) were installed close to the DAS cable. Generally, the duration of regular earthquake signals is longer in the offshore area both in DAS and seismometer data (OBSs in 2019 and DONET). This long duration in the offshore area can be caused by the thick sediment. Comparing the waveforms of regular earthquakes in DAS and OBSs at the same location, we found that the duration of the regular earthquakes in the DAS strain waveform is also longer than that in the OBS velocity waveform. The amplitudes of coda waves in the DAS strain waveform are generally larger than those in OBSs. The difference in waveform characteristics may be caused by that in the sensitivity of the incident angle between DAS and OBS. To clarify the cause of such differences, numerical simulation of strain waveforms can be required in the future studies.