*Brook Tozer1, Dan Bassett1, Shuichi Kodaira2, Adrien Arnulf3, Ayako Nakanishi4,2, Seiichi Miura4,2, Gou Fujie2, Kimihiro Mochizuki4
(1.GNS Science, 2.Japan Agency for Marine-Earth Science and Technology, 3.University of Texas Institute for Geophysics, 4.Earthquake Research Institute, the University of Tokyo)
Keywords:Subduction, Tōhoku, Megathrust
Recent studies have shown that crustal structure of the overlying plate may play an important role in modulating the slip behaviour of megathrust faults. Here, we present an integrated active- and passive-source three-dimensional tomographic model of crustal structure in the central part of the NE Japan subduction zone – spanning the rupture area of the 2011 M9 Tōhoku earthquake. This model has been constructed using seismic travel-time data generated by over two decades of offshore seismic surveys undertaken by JAMSTEC, ERI and others, which were recorded passively by ~150 seismographs located both onshore (Hi-net, F-net, Tōhoku University and Japan Meteorological Agency (JMA) networks) and offshore (S-net and temporary deployed Ocean Bottom Seismometers). These data have been augmented by twenty years of earthquake travel-time arrivals recorded on ~350 seismographs from the same networks, as documented by the JMA and relocated in this study.
Notably, our model resolves a ~ 0.4 km s-1 reduction in P-wave velocity of the forearc crust from north to south across a pre-identified forearc segment boundary interpreted as the offshore extension of the Median Tectonic Line. This supports a previous interpretation of Bassett et al. [2016], whereby the differences in P-wave velocities can be attributed to lithologic heterogeneity in the forearc, with higher velocity volcanic rocks in the north and accretionary sediments in the south. It follows that upper-plate structure may have played a key role in modulating both the accumulation and release of elastic strain associated with the 2011 Tōhoku earthquake.