Shear-wave anisotropy beneath the Ryukyu arc

Abstract

We have investigated regional shear-wave splitting beneath the Ryukyu arc in order to clarify the mantle flow beneath the Ryukyu arc and compare it with the extension of the Okinawa Trough. The splitting of regional shear waves (direct S) in the central and southern Ryukyu arc regions was analyzed in terms of fast-polarization direction and delay time using data from 11 short-period stations and two broadband seismic stations. The events occurred at depths ranging between 45 and 217 km, and the magnitude of the events ranged from 2.8 to 5.2. In general, the orientations of fast polarization were found to parallel to the strike of the Ryukyu Trench, with average delay times of 0.25-0.56 s. Specifically, the orientation of fast polarization was parallel to the direction of the extension in the south Ryukyu arc, suggesting that the direction of mantle flow is parallel to the extension in the south Okinawa Trough. However, the orientation of fast polarization was oblique to the direction of extension in the central Ryukyu arc, indicating that the mantle flow beneath the central Ryukyu arc is not parallel to the extension of the central Okinawa Trough. Based on these results, we suggest that the difference between the direction of the mantle flow and that of the lithosphere extension induces extensional strain with a simple shear component, which in turn causes oblique rifting in the central Okinawa Trough.

B‐type olivine fabric and mantle wedge serpentinization beneath the Ryukyu arc

“…Our trial‐and‐error modeling procedure closely follows that used by Wirth and Long []. While our starting isotropic model matched the radial component receiver functions well, we were unable to match features on the transverse component data that arrive between 3–6 s. We added a layer of anisotropy in the wedge mantle with a trench‐parallel fast symmetry axis, as inferred from previous studies [e.g., Long and van der Hilst , ; Nakamura and Yamamoto , ]. Although this feature improved the fit, we were unable to identify a combination of anisotropy parameters for the wedge mantle portion of our model that allowed us to reproduce all the transverse component features.…”

“…[5] Previous shear wave splitting studies in Ryukyu have reported strong mantle wedge anisotropy, with delay times of~1 s [e.g., Long and van der Hilst, 2006;Nakamura and Yamamoto, 2009]. The subducting slab beneath the stations used in these studies is relatively shallow (~40 km) [Syracuse and Abers, 2006], and accruing a shear wave splitting delay time of~1 s would require the presence of strong anisotropy.…”