Crustal structure of the incoming Pacific Plate off Hokkaido, the southern Kuril Trench subduction zone, by an active-source seismic survey using dense linear OBS array

Abstract:

An incoming oceanic plate transports water into the subduction zone by hydration before subduction. Previous seismic studies in the Japan and Kuril Trenches concluded that plate-bending normal faults developed at the trench outer slope play important roles for incoming plate hydration. However, a state of the hydrated plate after subducting is still unknown due to poor resolution of seismic velocity structure at the deep trench axis due to lack of instruments. Ultra-depth ocean bottom seismometers (OBSs) are expected to improve our understanding of the hydration process of the oceanic plate after it initiates subduction. We conducted an active-source seismic survey using a dense linear OBS array with ultra-depth OBSs at the Kuril Trench axial region.

The survey was conducted by R/Vs Kairei and Yokosuka of JAMSTEC in 2019. A 208 km-length survey line runs from the land slope to the trench outer slope where developed normal faults and a seamount exist. 80 of OBSs were deployed every ~2 km along the line. An airgun array with a total volume of 7,800 cu.in. was used as a controlled source and shoot every 200 m. Multi-channel seismic (MCS) reflection data was also collected by towing a 5.5 km-length multichannel hydrophone streamer cable.

As results of a first arrival traveltime inversion by a Monte Carlo approach and a reflection traveltime mapping method (TMM), a P-wave velocity (Vp) model, TMM and MCS images were obtained. The model and images present mutually consistent characteristic in both overriding and incoming plates. The subducting oceanic mantle, immediately beneath the Moho defined by the MCS image, has lower Vp (~7.5 km/s) than that at far oceanward part of the trench axis, indicating the presence of hydrated mantle. Depths to the Moho in TMM and MCS images, derived from wide-angle and near-vertical reflection arrivals, show a systematic difference by ~0.5 s in two-way time. One possible explanation for the difference would be the presence of a crust-mantle transition zone, showing gradual increase of Vp around the Moho depth, associated with serpentinizing the uppermost mantle material. Our result basically suggests that the oceanic mantle remains hydrated after initiating subduction.