Crustal Deformation and Interplate Coupling Associated with the 2011 Tohoku-oki Earthquake Based on a Viscoelastic Earthquake Cycle Model

The Pacific coast of northeast Japan has been subsiding at about 5 mm/yr before the 2011 M9.0 Tohoku-oki earthquake. On the other hand, existence of stage 5e marine terraces indicated the same area is slowly uplifting in a long term. Thus significant uplift was expected to occur to balance between the short term and the long term vertical movements. However, the 2011 event caused additional subsidence over 1 m along the Pacific coast. How to reconcile all these observations is still an open question. The question is also related to the mechanical coupling on the plate boundary. There have been several studies on this topic and they unanimously reported that the plate boundary off Miyagi is almost fully locked. Deeper limit of the locked zone has been estimated as 50 – 70 km in most studies while Suwa et al. (2006) estimated the locked region extended down to 100 km depth in order to reproduce interseismic subsidence with their elastic dislocation model. This discrepancy is crucial in evaluating future seismic hazard.

In order to tackle these problems, a simple kinematic model of earthquake deformation cycle is considered. This model, in which viscoelastic relaxation of the asthenosphere is taken into account, shows that postseismic uplift may recover both interseismic and coseismic subsidence if the shallow portion of the plate interface has a much longer recurrence interval than the viscoelastic relaxation time of the asthenosphere. This inference is consistent with the observation that there has been no giant earthquake along the Japan Trench at least after 1611. Afterslip on the plate interface also plays a significant role in postseismic uplift at the coastal area, but the afterslip is considered to last for only several years after the main shock and its total contribution in the whole earthquake cycle is limited. Reduced interseismic slip deficit below 50 km depth and concentrated aftershock of the 2011 event around this depth range indicate that the plate interface shallower than 50 km is the main part of interplate coupling. On the other hand, the deeper portion should be characterized as the velocity-strengthening friction. It is demonstrated that consideration of viscoelastic relaxation effects and cyclic effects from plate boundary earthquakes are essential in interpreting the observed deformation.