5:15 PM - 6:45 PM
[SSS05-P02] Relationship between the Boso slow slip events and the accompanying earthquake swarm
Keywords:slow slip, earthquake swarm, ΔCFF, fluid movement
Slow-slip events (SSEs) are events in which a plate boundary slips slowly without radiating seismic waves. They are detected by continuous geodetic observations of crustal deformation, and their occurrence has been reported in Japan and other parts of the world. Understanding SSEs is necessary to solve important problems such as revealing the mechanisms of stress accumulation and release at plate boundaries and predicting the occurrence of large earthquakes.
In the off Boso area, GNSS observations indicate that short-term SSEs with a duration of several to several tens of days have occurred a total of six times with recurrence periods of about 2-7 years from 1996 to 2018. In these Boso SSEs, the existence of earthquake swarm activity associated with it has been known.
The occurrence of earthquake swarms and SSEs are thought to be related: about the 2007 and 2011 Boso SSEs, the slip and seismic activity were well correlated in time and space, suggesting that the SSEs may have induced seismic activity (Hirose et al, GRL, 2014).
The purpose of this study was to investigate whether the 2007 Boso SSE triggered the associated earthquake swarm.
We estimated the spatio-temporal distribution of slip by performing ABIC inversion analysis on daily coordinate values (F5 solution) of GNSS data by the Geospatial Information Authority of Japan (GSI). From the obtained slip distribution, we calculated the stress change on the faults of the swarm and obtained ΔCFF, an index of the fault's susceptibility to slip. The stress changes were calculated for 21 faults inferred from the earthquake mechanism solutions published by the F-net of the National Research Institute for Earth Science and Disaster Prevention (NIED). We selected earthquakes of Mw 3.5 or greater that occurred during the SSE period and one month thereafter.
As a result, 43 % (9/21) of the ΔCFFs obtained were above the threshold value of 0.01 MPa, 14 % (3/21) were positive below the threshold value, and 43 % (9/21) were negative. The spatial distribution of ΔCFFs showed no regularity in its positive or negative values. Therefore, it is unreasonable to assume that the entire earthquake swarm was triggered by SSEs. We considered the existence of pore pressure that changes the ΔCFF to above the threshold value for faults where the ΔCFF is below the threshold value.
An increase in pore pressure of about 0.1-0.2 MPa was estimated on the fault of the swarm located in the slip area of the SSE. In the area where the Boso SSE occurs, the estimated effective normal stress (10-50 MPa), which is very small compared to the static rock pressure (350-600 MPa), suggests the existence of high pore pressure (Kobayashi and Sato, GRL, 2021), and it is possible that the slip of the SSE caused fluid migration from its slip area to the fault of the earthquake swarm.
Thus, we consider that the earthquake swarm were triggered by the slip of SSE through an indirect factor of fluid movement from the SSE slip area to the faults of the swarm.
Acknowledgement
GSI daily coordinates (F5 solution) were used in the analysis. Coulomb 3.3 from the U.S. Geological Survey was used to calculate stresses. F-net of NIED was used for the earthquake swarm mechanism solution. This work was supported by JSPS (23K03541).
In the off Boso area, GNSS observations indicate that short-term SSEs with a duration of several to several tens of days have occurred a total of six times with recurrence periods of about 2-7 years from 1996 to 2018. In these Boso SSEs, the existence of earthquake swarm activity associated with it has been known.
The occurrence of earthquake swarms and SSEs are thought to be related: about the 2007 and 2011 Boso SSEs, the slip and seismic activity were well correlated in time and space, suggesting that the SSEs may have induced seismic activity (Hirose et al, GRL, 2014).
The purpose of this study was to investigate whether the 2007 Boso SSE triggered the associated earthquake swarm.
We estimated the spatio-temporal distribution of slip by performing ABIC inversion analysis on daily coordinate values (F5 solution) of GNSS data by the Geospatial Information Authority of Japan (GSI). From the obtained slip distribution, we calculated the stress change on the faults of the swarm and obtained ΔCFF, an index of the fault's susceptibility to slip. The stress changes were calculated for 21 faults inferred from the earthquake mechanism solutions published by the F-net of the National Research Institute for Earth Science and Disaster Prevention (NIED). We selected earthquakes of Mw 3.5 or greater that occurred during the SSE period and one month thereafter.
As a result, 43 % (9/21) of the ΔCFFs obtained were above the threshold value of 0.01 MPa, 14 % (3/21) were positive below the threshold value, and 43 % (9/21) were negative. The spatial distribution of ΔCFFs showed no regularity in its positive or negative values. Therefore, it is unreasonable to assume that the entire earthquake swarm was triggered by SSEs. We considered the existence of pore pressure that changes the ΔCFF to above the threshold value for faults where the ΔCFF is below the threshold value.
An increase in pore pressure of about 0.1-0.2 MPa was estimated on the fault of the swarm located in the slip area of the SSE. In the area where the Boso SSE occurs, the estimated effective normal stress (10-50 MPa), which is very small compared to the static rock pressure (350-600 MPa), suggests the existence of high pore pressure (Kobayashi and Sato, GRL, 2021), and it is possible that the slip of the SSE caused fluid migration from its slip area to the fault of the earthquake swarm.
Thus, we consider that the earthquake swarm were triggered by the slip of SSE through an indirect factor of fluid movement from the SSE slip area to the faults of the swarm.
Acknowledgement
GSI daily coordinates (F5 solution) were used in the analysis. Coulomb 3.3 from the U.S. Geological Survey was used to calculate stresses. F-net of NIED was used for the earthquake swarm mechanism solution. This work was supported by JSPS (23K03541).