Seismic Probing of the Base of a Tectonic Plate from Subduction Zone to Trench Outer Rise: Hikurangi Margin, New Zealand

Abstract:

The nature of the base of a tectonic plate (LAB) is the 3^rd dimension of plate tectonics. Recent seismic studies of the LAB have revealed planar features that show very different characteristics. In the oceans, the top of the S-wave low velocity zone shows a systematic deepening with plate age that fits plate-cooling models. However, the change in radial anisotropy has a markedly constant depth of ~70 km, regardless of plate age. A recent land-based study (SAHKE 2) of the subducted Pacific Plate in the southern Hikurangi margin has imaged a pair of distinct reflectors defining a ~10 km thick channel parallel to and ~70 km below the top of the ~15° dipping plate. Low velocities indicate that the channel is a zone of partial melt or high volatile content, acting as a weak base to the plate. Receiver function studies along the Japan margin have also imaged layers at these depths, parallel to the top of the plate and dipping up to 45°. We propose probing the base of the tectonic plate by tracing potential LAB seismic reflectors from their dipping portions in the subduction zone to where they bend in the trench outer rise. If the seismically identified boundary represents a ‘frozen-in’ feature, created at the mid ocean ridge, then it will remain parallel to the top of the plate, and its nature will remain unchanged as it is tracked over the outer rise. Alternatively, if the base of the plate is a thin channel of partial melt, then one would expect thickening of the channel beneath the outer rise due to melt ponding in the core of the flexure; this melt ponding may be the source of volcanic activity. A 500-km survey will trace the Pacific plate LAB from the subduction zone into the trench outer rise. The deeper part of the line coincides with the part of the plate where the ~10 km thick ‘melt’ channel was clearly imaged with the SAHKE 2 experiment. We show with synthetic experiments that given seismic energy scatter and attenuation comparable to that observed in the SAHKE 2 experiment, and the typical noise of OBS records, the previously detected LAB should be capable of being imaged with a high resolution OBS/low-frequency air-gun deployment, with ~10 km OBS spacing and ~60 sec recording time, using the RV Langseth. In addition, we give examples of air-gun experiments elsewhere in the world that provide further support that the LAB can be imaged this way.