2:30 PM - 2:45 PM
[SCG55-27] Mechanism prohibiting earthquake growth
Keywords:The 2011 Tohoku-Oki earthquake, Japan Trench, Seafloor observation and exploration, Hadal paleoseismology, Earthquake cycle simulation
The 2011 Tohoku-Oki earthquake (Mw 9.0) (hereafter referred to as the "Tohoku-Oki earthquake") attracted attention not only for its immense magnitude but also because the rupture extended to the shallowest part of the plate boundary near the Japan Trench, an area previously believed incapable of large-displacement seismic slip. We hypothesize that slip to the trench (STT) in the shallow part of the plate boundary played a crucial role in the amplification of this earthquake. Since immediately after the earthquake, we have conducted survey observations in the deep-sea region along the Japan Trench, where shallow plate boundary slip occurred. As a result, we found that STT events of a similar scale to the 2011 earthquake had repeatedly occurred only in the central part of the Japan Trench. In the southern region, where significant postseismic slip was observed after the Tohoku-Oki earthquake, such slip characteristics likely suppressed the occurrence of large-scale STT events. However, in the northern region, where the 1896 Meiji Sanriku earthquake, a tsunami earthquake attributed to STT, occurred, we did not find traces of slip on the same scale as in the central region. Why is this the case? To answer this question, we conducted focused survey observations in the northern Japan Trench from 2019 to 2023.
In this region, it has been shown that slow slip events (SSEs) have repeatedly occurred with cycles of several years before the Tohoku-Oki earthquake, based on temporal changes in the activity of small repeating earthquakes. A large-scale SSE was known to have occurred in 2015, just before this study began, and a smaller SSE was detected in 2019. However, during this SSE, an observation using a dense OBS network revealed that tectonic tremor activity did not extend close to the trench axis. Additionally, seafloor acoustic distance measurements across the trench axis showed no significant baseline shortening. These findings suggest that SSE-induced slip did not reach the trench axis area, where STT occurred during the Meiji Sanriku earthquake, and that slip remained confined to deeper regions. In the shallow part of the plate boundary near the trench axis, the relative motion between plates cannot be accommodated by repeated SSEs alone, leading to the accumulation of slip deficit, which is eventually released during earthquakes like the Meiji Sanriku earthquake.
In parallel with this study, we conducted a seafloor sediment survey to investigate the occurrence history of Meiji Sanriku-type earthquakes. Sediment core analyses from the middle slope terrace along the Japan Trench revealed that, in the northern Japan Trench, smaller-scale events than the 2011 Tohoku-Oki earthquake had occurred more frequently. Event deposits corresponding to the Meiji Sanriku earthquake were identified in the surface core samples, and multiple traces of earthquakes of similar magnitude were found in deeper samples. By estimating the ages of these deposits using paleomagnetic analyses, we determined that large earthquakes have repeatedly occurred approximately once every 100 years. These event deposits are thought to result from the mobilization and redeposition of surface sediments due to strong seismic shaking. However, the source of strong shaking could include not only Meiji Sanriku-type STT earthquakes but also other nearby large earthquakes, such as outer-rise events. Therefore, the actual frequency of STT events may be slightly lower than our estimates.
In the central Japan Trench, large-scale STT events likely occurred when slip deficits in the shallow plate boundary, supported by strong deep locking, were occasionally released. Similarly, in the northern Japan Trench, the accumulation of slip deficits due to STT is expected, suggesting that the frequency of deep ruptures, which act as triggers for STT release, constrains the size of STT events. The higher frequency of deep ruptures that trigger STT in the northern region compared to the central region appears to be related to the repeated occurrence of SSEs in the north. We aim to verify this hypothesis through earthquake cycle simulations.
In this region, it has been shown that slow slip events (SSEs) have repeatedly occurred with cycles of several years before the Tohoku-Oki earthquake, based on temporal changes in the activity of small repeating earthquakes. A large-scale SSE was known to have occurred in 2015, just before this study began, and a smaller SSE was detected in 2019. However, during this SSE, an observation using a dense OBS network revealed that tectonic tremor activity did not extend close to the trench axis. Additionally, seafloor acoustic distance measurements across the trench axis showed no significant baseline shortening. These findings suggest that SSE-induced slip did not reach the trench axis area, where STT occurred during the Meiji Sanriku earthquake, and that slip remained confined to deeper regions. In the shallow part of the plate boundary near the trench axis, the relative motion between plates cannot be accommodated by repeated SSEs alone, leading to the accumulation of slip deficit, which is eventually released during earthquakes like the Meiji Sanriku earthquake.
In parallel with this study, we conducted a seafloor sediment survey to investigate the occurrence history of Meiji Sanriku-type earthquakes. Sediment core analyses from the middle slope terrace along the Japan Trench revealed that, in the northern Japan Trench, smaller-scale events than the 2011 Tohoku-Oki earthquake had occurred more frequently. Event deposits corresponding to the Meiji Sanriku earthquake were identified in the surface core samples, and multiple traces of earthquakes of similar magnitude were found in deeper samples. By estimating the ages of these deposits using paleomagnetic analyses, we determined that large earthquakes have repeatedly occurred approximately once every 100 years. These event deposits are thought to result from the mobilization and redeposition of surface sediments due to strong seismic shaking. However, the source of strong shaking could include not only Meiji Sanriku-type STT earthquakes but also other nearby large earthquakes, such as outer-rise events. Therefore, the actual frequency of STT events may be slightly lower than our estimates.
In the central Japan Trench, large-scale STT events likely occurred when slip deficits in the shallow plate boundary, supported by strong deep locking, were occasionally released. Similarly, in the northern Japan Trench, the accumulation of slip deficits due to STT is expected, suggesting that the frequency of deep ruptures, which act as triggers for STT release, constrains the size of STT events. The higher frequency of deep ruptures that trigger STT in the northern region compared to the central region appears to be related to the repeated occurrence of SSEs in the north. We aim to verify this hypothesis through earthquake cycle simulations.