Constraints on the Thermal and Compositional Nature of the Oceanic Lithosphere-Asthenosphere Boundary from Seismic Anisotropy

Abstract:

In this study we modeled S-wave velocities, radial and azimuthal anisotropy beneath the Pacific ocean, and compared our model with detections of the Gutenberg (G) discontinuity at 40–100 km depth to evaluate its context and relation to the lithosphere-asthenosphere boundary (LAB). The G is often associated with the LAB, but its sharpness and the low correlation between its depth and oceanic plate age suggest a compositional origin, in contradiction with tomographic models of isotropic wave velocities.

Here, we inverted fundamental and higher mode anisotropic Rayleigh wave phase velocity maps to which we applied non-linear crustal corrections. Our model defines three layers within the upper 250km of the mantle. The bottom layer is characterized by relatively low velocities, strong (3%) azimuthal anisotropy, fast seismic directions that follow the absolute plate motion (APM), and strong (5%) radial anisotropy with V_SH>V_SV. This suggests alignment of olivine fast axes with mantle flow direction in the asthenosphere. The middle layer has fast axes aligned with the paleospreading directions, and the boundary between the bottom and middle layers follows a half-space cooling model. This suggests a thermal origin of the LAB if we use the change in alignment of the fast axes with the APM as a proxy for the LAB.

Remarkably, a change in azimuthal anisotropy is found between the two top layers at a roughly constant depth that coincides with the location of the G. The G is therefore located within the thermal lithosphere and is primarily associated with a vertical gradient in azimuthal anisotropy, which may result from compositional changes. Dehydration of the mantle underlying mid-ocean ridges offers a possible explanation for our results. It could generate a chemically depleted, viscous layer that becomes overprinted by lowered temperatures as the plate cools and migrates away from the ridge. The olivine fast axes would align with the spreading direction at the ridge in the underlying hydrated, more deformable bottom layer, yielding an anisotropic discontinuity at the base of the dehydrated layer. This alignment would be frozen in the part of the lithosphere that lies below the chemical depletion boundary as the lid cools and thickens, whereas flow in the warmer asthenosphere would align with present-day APM.