Anisotropic Shear-Velocity Structure of the Lithosphere-Asthenosphere System in the Central Pacific from the NoMelt Experiment

Abstract:

Recent theoretical models of the seismic properties of mantle rocks predict seismic velocity profiles for mature oceanic lithosphere that are fundamentally inconsistent with the best observations of seismic velocities in two ways. Observations of strong positive velocity gradients with depth, and a very sharp and very shallow low-velocity asthenosphere boundary (LAB), both suggest that non-thermal factors such as bulk composition, mineral fabric, grain size, and dehydration play important roles in controlling the formation of the lithosphere, and thus the underlying LAB. There is little consensus on which of these factors are dominant, in part because observations of detailed lithosphere structure are limited. To address this discrepancy, we conducted the NoMelt experiment on ~70 Ma Pacific lithosphere between the Clarion & Clipperton fracture zones. The experiment consists of a 600x400 km array of broad-band ocean bottom seismometers (OBS) and magnetotelluric instruments, and an active-source reflection/refraction experiment. Here we characterize the shear-velocity structure and its seismic anisotropy across the lithosphere-asthenosphere system beneath the array using surface-wave dispersion. Of the 27 deployed instruments, 21 were recovered, all of which produced useful data on the seismometer and the differential pressure gauge in the 20-200 s period band. Energetic, high S/N Rayleigh waves and useful love waves are observed from over 21 and 8 events with Mw > 6.5 respectively. Models of shear velocity as a function of depth derived from intra-array Rayleigh-wave phase velocities are characterized by a relatively constant, high-velocity lithosphere to ~70 km depth, with a rapid drop in velocity below that depth. The combination of a high-velocity lid with an abrupt transition to the low-velocity zone cannot be explained by plate-cooling models. The Rayleigh waves display strong azimuthal anisotropy with a fast direction parallel to fossil spreading at all periods; the anisotropy is strongest at shortest periods, suggesting that the underlying fabric is dominantly in the lithosphere. We are developing a mode-interference analysis to evaluate the Love-wave propagation across the system. These data will provide constraints on the radial anisotropy across the lithosphere-asthenosphere system.