10:45 AM - 12:15 PM
[SCG45-P33] Toward integrated seismic imaging based on active and passive seismic records for the shallow slow earthquake fault zone
Keywords:tremor, Nankai trough, slow earthquake, Reflection seismic survey
At least two types of earthquakes occur in shallow subduction zones: tremors, a type of slow earthquakes, and ordinary earthquakes. The cause of the difference between the two types of seismic events is a fundamental question in seismology. Recently, Toh et al. (2023, under review) suggested that tremor sources are likely surrounded by a strongly scattering medium, based on observations of shallow tremors by ocean-bottom seismometers. Their seismic wave propagation simulations showed that the observed waveforms could be qualitatively explained by many small, low-S-wave velocity inclusions (scatterers) distributed around the tremor source. In another study based on ocean-bottom drilling, a structure with patchily distributed overpressured aquifers has been proposed for a tremor source region (Hirose et al., 2021, JGR). The inclusions may be the seismic expressions of the aquifers, thus making the two models consistent. Moreover, the scattering medium could be the structure along the slow-earthquake fault zone and could be the main factor that distinguishes the two types of earthquakes.
We aim to further constrain the structure of the tremor source region using seismic reflection survey profiles. Our objective is to constrain the structure of the slow-earthquake fault zone by waveform modeling of both passive and controlled seismic records. The inclusions proposed above have a large impedance contrast to the surrounding medium; thus, the seismic reflection records should be highly sensitive to their presence, if they exist.
The high fluid pressure zone proposed by Hirose et al. (2021), which may also contain the distributed inclusions proposed by Toh et al. (2023), corresponds to the fault zones from the décollement down to the top of the oceanic crust when projected onto a seismic reflection profile (Moore et al., 2001). The décollement is characterized as a low-velocity zone with high pore pressure and is much thicker compared to the seismic wavelength used in the survey, resulting in its observation as a strong reflecting interface. Conversely, the zone below the décollement, where an aquifer was detected by drilling, exhibits only weak reflectors in the seismic reflection profile. At first glance, this feature appears to contradict the distributed inclusion model mentioned above since the inclusions with large impedance contrast to the surrounding would generate strong reflections. However, it is known that the signal amplitude can sometimes become almost invisible when the thickness of the fault zone is smaller than the seismic wavelength due to interference of reflected waves from the upper and lower interfaces. Thus, the weak amplitude of the signals in the tremor source region in the seismic reflection profile provides an important constraint on the structure.
To begin with, we performed waveform modeling on a 1D model where we replaced the inclusion structure of Toh et al. (2023) with a low velocity zone. We searched for a thickness of the layer where the amplitude of the reflected wave gets reduced. For example, we confirmed that the amplitude of the reflected signal gets reduced to half of the original size when a 5 m thick layer with 30 % velocity reduction exists 4km below the seafloor and when the air gun signal is assumed by a Ricker wavelet with a center frequency at 25 Hz. We will extend our waveform modeling to 2D and search for structures that can simultaneously explain both the distributed inclusions proposed from the tremor observations and the features observed in the seismic reflection survey profile.
We aim to further constrain the structure of the tremor source region using seismic reflection survey profiles. Our objective is to constrain the structure of the slow-earthquake fault zone by waveform modeling of both passive and controlled seismic records. The inclusions proposed above have a large impedance contrast to the surrounding medium; thus, the seismic reflection records should be highly sensitive to their presence, if they exist.
The high fluid pressure zone proposed by Hirose et al. (2021), which may also contain the distributed inclusions proposed by Toh et al. (2023), corresponds to the fault zones from the décollement down to the top of the oceanic crust when projected onto a seismic reflection profile (Moore et al., 2001). The décollement is characterized as a low-velocity zone with high pore pressure and is much thicker compared to the seismic wavelength used in the survey, resulting in its observation as a strong reflecting interface. Conversely, the zone below the décollement, where an aquifer was detected by drilling, exhibits only weak reflectors in the seismic reflection profile. At first glance, this feature appears to contradict the distributed inclusion model mentioned above since the inclusions with large impedance contrast to the surrounding would generate strong reflections. However, it is known that the signal amplitude can sometimes become almost invisible when the thickness of the fault zone is smaller than the seismic wavelength due to interference of reflected waves from the upper and lower interfaces. Thus, the weak amplitude of the signals in the tremor source region in the seismic reflection profile provides an important constraint on the structure.
To begin with, we performed waveform modeling on a 1D model where we replaced the inclusion structure of Toh et al. (2023) with a low velocity zone. We searched for a thickness of the layer where the amplitude of the reflected wave gets reduced. For example, we confirmed that the amplitude of the reflected signal gets reduced to half of the original size when a 5 m thick layer with 30 % velocity reduction exists 4km below the seafloor and when the air gun signal is assumed by a Ricker wavelet with a center frequency at 25 Hz. We will extend our waveform modeling to 2D and search for structures that can simultaneously explain both the distributed inclusions proposed from the tremor observations and the features observed in the seismic reflection survey profile.