Friday Seminar: (Nov. 29th, 2019) Dr. Naoki Uchida (Tohoku University)

The exploration of the S-net cabled seafloor observation data in Tohoku University and the results on the forearc Shear-wave splitting

Recently, the “Seafloor Observation Network for Earthquakes and Tsunamis along the Japan Trench” (S-net) was established off NE Japan by the National Research Institute for Earth Science and Disaster Resilience (NIED). The deployment of the cable system began in 2013, completed in 2017, and the data were made publicly available from October 2018 onward. The new cabled system covers a subsea area of about 300 ×1000 km with 150 ocean bottom seismometers (OBSs) connected by a 5,800 km long fiberoptic cable. Seismic records from these instruments help expand the understanding of seismicity, megathrust slip and structure into a vast virgin territory. In this presentation, I introduce the efforts to make the best use of this precious data by the colleagues in Tohoku University and my results on the anisotropic structure in the forearc area detailed below.

Shear-wave anisotropy in subduction zones provides important information on crustal structure, tectonic stress, and mantle wedge flow. Because of severe paucity of observations, the anisotropy of the offshore forearc is essentially unknown, and the incomplete knowledge based on onshore observations alone has led to controversial and competing geodynamic models. Here we show that seismic observations made with a cabled ocean bottom seismic network offshore of NE Japan, namely the S-net, has enabled the first systematic mapping of shear-wave anisotropy in the offshore part of a subduction zone forearc. By analyzing waveforms of 606 interplate and 108 upper-plate earthquakes recorded by 84 S-net stations, we find that the fast directions are predominantly trench-parallel, with delay times of ~0.1 s, similar to those in the onshore forearc previously determined from waveforms of intraslab earthquakes recorded by land-based stations. The splitting parameters are insensitive to changes in source depths down to 100 km, suggesting that most of the anisotropy resides in the crust of the upper plate, not associated with trench-parallel mantle flow or the presence of B-type olivine in the mantle wedge. Comparison with regional geology indicates that the anisotropy is due mainly to structural fabrics, although crustal stress may also play a role. The lack of anisotropy in the forearc mantle wedge is in sharp contrast with the strong anisotropy of the rest of the mantle wedge, clearly reflecting a spatial change in mantle wedge dynamics. The transition from no to strong mantle anisotropy occurs where the plate interface is at 70-90 km depth and where the slab begins to be fully coupled with the mantle wedge. The stagnant part of the mantle wedge overlying the decoupled interface features little anisotropy. The flowing part of the mantle wedge overlying the coupled interface features strong anisotropy due to lattice-preferred orientation of anisotropicminerals.