10:00 〜 10:15
[SCG45-17] Postseismicity underneath a slow slip event reported
キーワード：Slow slip events、Afterslip、Slow earthquake cycle
Slow earthquakes are thought to be one form of faulting, similarly to regular earthquakes of fast sliding. Then, why is each phase of the seismic cycle not ubiquitously reported for slow earthquakes, unlike regular ones?
The slow-earthquake cycle itself appears to exist, at least in a classical sense of a seismic cycle since Reid (1910) that a plate boundary fault accumulates and releases the seismic moment recurrently. Bartlow (2020) and Wallace (2020) have reported that the slip zones of repeating SSEs vary their coupling ratios over their recurrence intervals. However, in its modern sense, the seismic cycle consists of preseismic, coseismic, postseismic, and interseismic phases (Avouac, 2015). Coupling-ratio modulations during repeating SSEs demonstrate the presence of coseismic and interseismic phases of slow earthquakes but serve no evidence of the pre/postseismic phase in the sense of regular earthquakes. Preseismicity is under debate even for regular earthquakes (e.g., Kanamori, 1981), arguably outside the established issue. Verifiable yet to be scrutinized is thus the postseismicity of slow earthquakes.
For regular earthquakes, the postseismic phase refers to the period where the surrounding coupled zone of a plate boundary creeps with off-fault ductility to release the slip deficit, lagging behind the failure of a locked zone in a coseismic period (Wang et al., 2012). Aftershocks here occur as per the Omori-Utsu law in accordance with afterslip distance (Perfettini et al., 2018) and coseismically loaded stress (Wetzler et al., 2018). So, returning to our question, suppose the slow earthquake is a type of plate-loading-driven frictional failure of plate boundary faults. Then, it appears unnatural to think that there would be no postseismic deformation accompanying the failure of an asperity that produces a slip rate significantly exceeding the plate loading rate because such is the same coseismic kinematics as the regular earthquake invoking its own postseismicity.
We argue that we might have missed the seismic phases from preseismic to postseismic ones underneath each of the slow earthquakes, especially underneath the slow slip event (SSE), due to the deficiency of slip accelerations. Our argument is motivated by a geodetic observation of the 2009-2010 Bungo-Channel long-term SSE in south-western Japan (Kano et al., 2024). A close look at this observation has allowed us to recognize apparent preseismic signals are followed by short but significant coseismic signals from an asperity, which lead to long-lasting postseismic signals that indicate the slip of the asperity's surrounding proportional to logarithmic time-lapse, consistent with synchronized tremor frequency. One remarkable feature of this SSE is the dominance of the postseismic slip in terms of the interval and magnitude. Inverted slip patterns of the 2009-2010 Bungo-Channel SSE indicate that its nominal slip is contaminated with the 2008-2009 north Hyuga-nada SSE (Ozawa et al., 2024) as apparent preseismicity. Our physics-based simulation suggests that such spatiotemporal proximity of SSEs enhanced their afterslips to an undoubtedly detectable level. Repeating SSEs in the absence of migrations could occur with a predominance of postseismic slips, which begin to slow down right after their onsets, thus being predictable, with potential distinguishability from megathrust earthquake nuclei that keep accelerating to disruptive failure.
The slow-earthquake cycle itself appears to exist, at least in a classical sense of a seismic cycle since Reid (1910) that a plate boundary fault accumulates and releases the seismic moment recurrently. Bartlow (2020) and Wallace (2020) have reported that the slip zones of repeating SSEs vary their coupling ratios over their recurrence intervals. However, in its modern sense, the seismic cycle consists of preseismic, coseismic, postseismic, and interseismic phases (Avouac, 2015). Coupling-ratio modulations during repeating SSEs demonstrate the presence of coseismic and interseismic phases of slow earthquakes but serve no evidence of the pre/postseismic phase in the sense of regular earthquakes. Preseismicity is under debate even for regular earthquakes (e.g., Kanamori, 1981), arguably outside the established issue. Verifiable yet to be scrutinized is thus the postseismicity of slow earthquakes.
For regular earthquakes, the postseismic phase refers to the period where the surrounding coupled zone of a plate boundary creeps with off-fault ductility to release the slip deficit, lagging behind the failure of a locked zone in a coseismic period (Wang et al., 2012). Aftershocks here occur as per the Omori-Utsu law in accordance with afterslip distance (Perfettini et al., 2018) and coseismically loaded stress (Wetzler et al., 2018). So, returning to our question, suppose the slow earthquake is a type of plate-loading-driven frictional failure of plate boundary faults. Then, it appears unnatural to think that there would be no postseismic deformation accompanying the failure of an asperity that produces a slip rate significantly exceeding the plate loading rate because such is the same coseismic kinematics as the regular earthquake invoking its own postseismicity.
We argue that we might have missed the seismic phases from preseismic to postseismic ones underneath each of the slow earthquakes, especially underneath the slow slip event (SSE), due to the deficiency of slip accelerations. Our argument is motivated by a geodetic observation of the 2009-2010 Bungo-Channel long-term SSE in south-western Japan (Kano et al., 2024). A close look at this observation has allowed us to recognize apparent preseismic signals are followed by short but significant coseismic signals from an asperity, which lead to long-lasting postseismic signals that indicate the slip of the asperity's surrounding proportional to logarithmic time-lapse, consistent with synchronized tremor frequency. One remarkable feature of this SSE is the dominance of the postseismic slip in terms of the interval and magnitude. Inverted slip patterns of the 2009-2010 Bungo-Channel SSE indicate that its nominal slip is contaminated with the 2008-2009 north Hyuga-nada SSE (Ozawa et al., 2024) as apparent preseismicity. Our physics-based simulation suggests that such spatiotemporal proximity of SSEs enhanced their afterslips to an undoubtedly detectable level. Repeating SSEs in the absence of migrations could occur with a predominance of postseismic slips, which begin to slow down right after their onsets, thus being predictable, with potential distinguishability from megathrust earthquake nuclei that keep accelerating to disruptive failure.