10:15 AM - 10:30 AM
[SCG50-06] Effects of a crustal heterogeneity in the postseismic deformation of 2008 Iwate-Miyagi Nairiku Earthquake
Keywords:Intra-plate Earthquake, Viscoelasticity, InSAR, postseismic deformation, crustal heterogeneity
The Iwate-Miyagi Nairiku Earthquake (hereafter IMNE) occurred on June 14, 2008 (Mj 7.2) at the eastern flank of Mt. Kurikoma, an active volcano. The IMNE is characterized by conjugate fault rupture: west-dipping and east-dipping faults brought about complex surface deformation (e.g., Takada et al., 2009). A relatively large amount of post-seismic deformation has been observed by GNSS surveys (e.g., Ohzono et al., 2012) and InSAR analyses. The post-seismic displacement field at high spatial resolution detected by InSAR should reflect multi-scale crustal heterogeneities in the hypocentral area (e.g., Yoshida, 2001), which, however, has not been investigated so far.
In this study, first, we executed InSAR time-series analysis (SBAS by Berardino et al. 2002) to elaborate the post-seismic displacements in the hypocentral area using a series of SAR images taken by ALOS from descending orbit. During 2.2 years after IMNE, we detected the following signals: (1) LOS length increase at the footwall of the west-dipping fault, (2) LOS shortening at the hanging wall of the east-dipping fault, and (3) LOS shortening to the west of Mt. Kurikoma. We also found that these three signals accumulate with time. The first and the second signals would be due to afterslip on the west- and the east-dipping faults, respectively. The third signal reaches more than 10 cm in ~2 years, which indicates the existence of rheological heterogeneity to the west of Mt. Kurikoma. Interestingly, the location of the third signal coincides with the area of localized subsidence triggered by the 2011 Tohoku-Oki Earthquake (Takada and Fukushima, 2013).
To understand the physical effects of such a crustal heterogeneity, we construct a physical model of the postseismic deformation following IMNE using a numerical code Unicycle (Barbot et al., 2017; Moore et al., 2017) with which we consider fault slip, viscoelastic flow, and those interactions through the stress redistribution. We set the coseismic slip distribution as proposed by Iinuma et al. (2009). We also assume that the fault slip obeys the rate-and-state friction law (Dietrich, 1979; Ruina, 1983). First, we modeled crust-mantle structure as an elastic layer of 20 km thickness overlying a Maxwell viscoelastic substratum following Ohozono et al. (2012). Next, we put a localized viscoelastic volume at 8-20 km depth to the west of Mt. Kurikoma where the viscosity is three times lower than that of the substratum. The numerical result indicates that the postseismic slip on the west-dipping fault expands to the shallower and the deeper portion of the coseismic rupture area with time. Furthermore, the deeper extension of the postseismic slip reaches the low viscosity region to the west of Mt. Kurikoma, and it causes volumetric flow which strongly enhances surface uplift to the west of Mt. Kurikoma and explains LOS shortening there. As a result, our model well explains the time-series of postseismic LOS change obtained by the SBAS analysis. Our model also well explains the LOS length increase to the east of the west-dipping fault through the shallow postseismic slip. Thus, we suggested that localized crustal heterogeneities have the potential to cause a large-scale postseismic deformation through physical interaction with fault slips. Combining our result with the instantaneous subsidence in this area due to the 2011 Tohoku-Oki earthquake (Takada and Fukushima, 2013), we may say that the shallower portion of Mt. Kurikoma is characterized by heterogeneously low elasticity, and the deeper portion is characterized by heterogeneously low viscosity. For the east-dipping fault, the coseismic slip area overlapped with the postseismic slip area especially at the shallower portion, which indicates the importance of pre-stress. In the presentation, we mention how the postseismic slip depends on the pre-stress distribution.
In this study, first, we executed InSAR time-series analysis (SBAS by Berardino et al. 2002) to elaborate the post-seismic displacements in the hypocentral area using a series of SAR images taken by ALOS from descending orbit. During 2.2 years after IMNE, we detected the following signals: (1) LOS length increase at the footwall of the west-dipping fault, (2) LOS shortening at the hanging wall of the east-dipping fault, and (3) LOS shortening to the west of Mt. Kurikoma. We also found that these three signals accumulate with time. The first and the second signals would be due to afterslip on the west- and the east-dipping faults, respectively. The third signal reaches more than 10 cm in ~2 years, which indicates the existence of rheological heterogeneity to the west of Mt. Kurikoma. Interestingly, the location of the third signal coincides with the area of localized subsidence triggered by the 2011 Tohoku-Oki Earthquake (Takada and Fukushima, 2013).
To understand the physical effects of such a crustal heterogeneity, we construct a physical model of the postseismic deformation following IMNE using a numerical code Unicycle (Barbot et al., 2017; Moore et al., 2017) with which we consider fault slip, viscoelastic flow, and those interactions through the stress redistribution. We set the coseismic slip distribution as proposed by Iinuma et al. (2009). We also assume that the fault slip obeys the rate-and-state friction law (Dietrich, 1979; Ruina, 1983). First, we modeled crust-mantle structure as an elastic layer of 20 km thickness overlying a Maxwell viscoelastic substratum following Ohozono et al. (2012). Next, we put a localized viscoelastic volume at 8-20 km depth to the west of Mt. Kurikoma where the viscosity is three times lower than that of the substratum. The numerical result indicates that the postseismic slip on the west-dipping fault expands to the shallower and the deeper portion of the coseismic rupture area with time. Furthermore, the deeper extension of the postseismic slip reaches the low viscosity region to the west of Mt. Kurikoma, and it causes volumetric flow which strongly enhances surface uplift to the west of Mt. Kurikoma and explains LOS shortening there. As a result, our model well explains the time-series of postseismic LOS change obtained by the SBAS analysis. Our model also well explains the LOS length increase to the east of the west-dipping fault through the shallow postseismic slip. Thus, we suggested that localized crustal heterogeneities have the potential to cause a large-scale postseismic deformation through physical interaction with fault slips. Combining our result with the instantaneous subsidence in this area due to the 2011 Tohoku-Oki earthquake (Takada and Fukushima, 2013), we may say that the shallower portion of Mt. Kurikoma is characterized by heterogeneously low elasticity, and the deeper portion is characterized by heterogeneously low viscosity. For the east-dipping fault, the coseismic slip area overlapped with the postseismic slip area especially at the shallower portion, which indicates the importance of pre-stress. In the presentation, we mention how the postseismic slip depends on the pre-stress distribution.