2:30 PM - 2:45 PM
[SCG50-16] Spatiotemporal change in the inelastic strain rate due to aftershock activity of the 2016 Kumamoto earthquake sequence, in central Kyushu, Japan
Keywords:inelastic strain rate, stress field, aftershock activity
For a purpose of understanding physical process of earthquake generation, we investigate the stress and strain field around the hypocentral area of the 2016 Kumamoto earthquake sequence. The stress field around the region is normal or strike-slip type, and the magnitude of the stress tensor was estimated to be comparable to the stress change due to the maximum foreshock (Mj 6.5) and mainshock (Mj 7.3) [Mitsuoka et al., 2020]. Aftershock activity following the mainshock has persisted in the region to the present day.
Aftershock activity can be considered as an inelastic response to stress loading by large co-seismic slip of a mainshock. As known the modified Omori’s law [Utsu 1957], earthquake rate of the aftershocks follows the power law of lapse time from occurrence of the mainshock. Similarly, inelastic strain rate by aftershock activity follows a power law (dε/dt∝t-p) [Matsumoto et al., 2020]. We defined the power -p in this relation as the p-value and investigated the spatial distribution of the p-value in the aftershock area of the 2016 Kumamoto earthquake sequence. We estimated the inelastic strain tensor in a spatial bin by summing the moment tensors [Kostrov, 1974] at several lapse time windows, and calculated the inelastic strain rate by the time length of each time window. In addition, we estimated the p-value by the least squares method, assuming that the inelastic strain rate follows the power law (dε/dt∝t-p).
The p-value around large co-seismic slip area [e.g., Asano and Iwata, 2016; Mitsuoka et al., 2020] is high (p>1), therefore the deformation at this area can be interpreted as a response to elastic strain by the co-seismic faults when considering exponentiation fluids [e.g., Nanjo, 2007].
At extensions of the co-seismic fault (i.e. the southern Hinagu fault and the western of the Futagawa fault), we found the low p-value (p<1). The low p-value areas correspond to the area of the low co-seismic stress change (approximately less than 1 MPa) by the maximum foreshock and mainshock. The low p-value indicates the slow decay rate of the strain rate, and p<1 condition is that the inelastic strain increase over time. This suggests that post-earthquake activities such as afterslip cause the low p-value area. We discuss a fault slip model or an inelastic deformation in and around the target area that can explain the inelastic deformation in this low p-value region.
Aftershock activity can be considered as an inelastic response to stress loading by large co-seismic slip of a mainshock. As known the modified Omori’s law [Utsu 1957], earthquake rate of the aftershocks follows the power law of lapse time from occurrence of the mainshock. Similarly, inelastic strain rate by aftershock activity follows a power law (dε/dt∝t-p) [Matsumoto et al., 2020]. We defined the power -p in this relation as the p-value and investigated the spatial distribution of the p-value in the aftershock area of the 2016 Kumamoto earthquake sequence. We estimated the inelastic strain tensor in a spatial bin by summing the moment tensors [Kostrov, 1974] at several lapse time windows, and calculated the inelastic strain rate by the time length of each time window. In addition, we estimated the p-value by the least squares method, assuming that the inelastic strain rate follows the power law (dε/dt∝t-p).
The p-value around large co-seismic slip area [e.g., Asano and Iwata, 2016; Mitsuoka et al., 2020] is high (p>1), therefore the deformation at this area can be interpreted as a response to elastic strain by the co-seismic faults when considering exponentiation fluids [e.g., Nanjo, 2007].
At extensions of the co-seismic fault (i.e. the southern Hinagu fault and the western of the Futagawa fault), we found the low p-value (p<1). The low p-value areas correspond to the area of the low co-seismic stress change (approximately less than 1 MPa) by the maximum foreshock and mainshock. The low p-value indicates the slow decay rate of the strain rate, and p<1 condition is that the inelastic strain increase over time. This suggests that post-earthquake activities such as afterslip cause the low p-value area. We discuss a fault slip model or an inelastic deformation in and around the target area that can explain the inelastic deformation in this low p-value region.