Local receiver fault dependency of seismicity shut down in the 2011 Tohoku-oki stress shadow

The 11 March 2011 Tohoku-oki earthquake of Mw=9.0 brought a wide spread aftershocks in northern Honshu island, Japan. Either static Coulomb stress, dynamic stress, pore fluid pressure change or a combination of them takes a responsibility for the occurrence of the off-fault aftershocks. In contrast, seismicity decrease after the Tohoku-oki is extremely limited to several regions, the vicinity of the 2011 hypocenter and aftershock zones of a few preceding large inland earthquakes. Such pre-Tohoku-oki high seismicity and/or post-Tohoku-oki longer observation would be needed to detect significance of seismic quiescence. We now focus on the other requirement, sensitivity to local-scale receiver fault mechanism, to see if hidden faults were indeed brought farther from failure by the Tohoku-oki mainshock.

Here we demonstrate how much local fault structure affects the observed aftershock shut down of the 14 June 2008 Iwate-Miyagi inland earthquake (Mw=6.6) after the Tohoku-oki. To perform the detail investigation, we use determined 3668 focal mechanisms (Yoshida et al., 2013) of the Iwate-Miyagi aftershocks occurred until 10 March 2011, as a proxy for hidden active faults. We then compute Coulomb stress change (ΔCFF) on all nodal planes using a coseismic slip model of the Tohoku-oki (Iinuma et al., 2012) in an elastic half-space with a wide range of apparent friction coefficient μ’ of 0.0, 0.4 and 0.8. 80% and 50% of the resolved ΔCFF on all the planes are negative under μ’=0.0 and 0.8, respectively. In contrast, positive ΔCFF are obtained mostly from strike-slip nodal planes occupied in the western 2008 rupture zone. Post-Tohoku-oki seismicity was active in the area while most of the 2008 aftershock zone occupied by reverse faults was in relative quiescence.

To explain overall time-dependent behavior of the observed 2008 aftershock shut down due to the Tohoku-oki, we next employ rate- and state-dependent friction law of Dieterich (1994). A group of the calculated ΔCFF values on all nodal planes is used to simulate the 2008 aftershock response to the 2011 Tohoku-oki stress perturbation following the Monte Carlo method of Toda et al. (2012). Despite a fraction of resolved stress increase onto strike-slip faults, rate/state model well reproduces the rate decrease under the lower μ’ that may imply weak faults.