Geometrical Effects of a Possible Subducted Seamount in the 2011 Mw 9.0 Tohoku Earthquake

Abstract:

A concentrated large slip patch near the hypocenter of the 2011 Mw 9.0 Tohoku earthquake and tens of seconds delay in up-dip rupture propagation in the early stage of the event, revealed by kinematic inversions of seismic, GPS, and/or tsunami data, were explained by a possible subducted seamount just up-dip of the hypocenter in our previous study (Duan, 2012). Bathymetry of the source region (Amante and Eakins, 2009), seismicity studies (e.g., Uchida et al., 2010), and tomographic studies (e.g., Zhao et al., 2011; Yamamoto et al., 2014) support the seamount hypothesis. However, the seamount in our previous study was parameterized by elevated effective normal stress and shear stress on a planar, shallow-dipping subduction plane. In this study, we incorporate seamount geometry into spontaneous rupture models of a non-planar megathrust fault, to examine geometrical effects of a subducted seamount in the 2011 event. In particular, we examine geometrical effects of the seamount with various plausible initial stress conditions on the non-planar fault, including heterogeneous stress fields that may account for effects of previous magnitude 7~8 earthquakes. Furthermore, our non-planar fault is embedded in layered velocity structure, so that combined effects of fault geometrical complexity and surrounding velocity structure on rupture propagation of the 2011 event can be examined. We will compare our synthetics with near-field coseismic GPS and/or seismic recordings to constrain our models. More physical insights into why the rupture propagated in the way it did in the 2011 event will be gained from this study.