Testing How Depletion, Dehydration and Melt Affect Seismic Expressions of the Asthenosphere

Abstract:

Seismic images of the upper mantle beneath mid-ocean ridges and rift zones typically show discontinuities in wave speed that are often attributed to the presence of partial melt. To understand how lithosphere and asthenosphere differ in temperature, composition and melt content, it is necessary to test plausible dynamic scenarios against a wide range of observables. Here we take a first step at comparing seismic constraints with the seismic structure predicted for (1) a simple oceanic lithosphere-asthenosphere system, formed by melt extraction and dehydration at the ridge and half-space cooling as the lithosphere away from it and (2) continental break-up where there is significant partial melting. We have developed a relatively simple 2-D geodynamic model of decompression melting during extension that incorporates melt retention and dehydration. This model can explain the seismic structure below the East Pacific Rise as imaged by surface and body waves, including a double low velocity zone, with triangular anomaly above 50-60 km depth due to dry melting and low velocity layer between 60 and ~200km depth mainly resulting from solid-state anelasticity in hydrated mantle. We only require a minor contribution below the ridge of deep hydrous melt. Below Afar, in the East African Rift, S receiver functions have found a discontinuity in shear wave speeds at ~75km depth, consistent with the depth of the onset of dry melting. Yet, from the same forward model, no aspect of the melt zone leads to sufficiently sharp impedance contrast in isotropic velocity to explain the receiver-function signals. This suggests that something else is required to explain the seismic observations. We speculate that this may be a change in seismic anisotropy caused by a change in melt segregation characteristics.