Experimental evidence supporting a global melt layer at the base of the Earth\u2019s upper mantle

D Freitas,D Andrault,N Bolfan-Casanova,J Chantel,F Schiavi,G Manthilake,M A Bouhifd

doi:10.1038/s41467-017-02275-9

Abstract

The low-velocity layer (LVL) atop the 410-km discontinuity has been widely attributed to dehydration melting. In this study, we experimentally reproduced the wadsleyite-to-olivine phase transformation in the upwelling mantle across the 410-km discontinuity and investigated in situ the sound wave velocity during partial melting of hydrous peridotite. Our seismic velocity model indicates that the globally observed negative Vs anomaly (−4%) can be explained by a 0.7% melt fraction in peridotite at the base of the upper mantle. The produced melt is richer in FeO (~33 wt.%) and H2O (~16.5 wt.%) and its density is determined to be 3.56–3.74 g cm−3. The water content of this gravitationally stable melt in the LVL corresponds to a total water content in the mantle transition zone of 0.22 ± 0.02 wt.%. Such values agree with estimations based on magneto-telluric observations.

Highlights

The low-velocity layer (LVL) atop the 410-km discontinuity has been widely attributed to dehydration melting
The current velocity models are based on either low pressure (
At a fixed pressure of 12 GPa, the sound velocity of peridotite before melting is characterized as expected by a gradual decrease of compressional and shear wave velocities with increasing temperature

Summary

Introduction

The low-velocity layer (LVL) atop the 410-km discontinuity has been widely attributed to dehydration melting. Our seismic velocity model indicates that the globally observed negative Vs anomaly (−4%) can be explained by a 0.7% melt fraction in peridotite at the base of the upper mantle. The highly wetting character of hydrous silicate melt at high pressure[11] implies that even a small degree of melting can dramatically affect the propagation of seismic waves at high depths Both experimental[12,13,14] and theoretical models[15,16,17,18,19,20,21,22] have investigated the magnitude of the velocity drop that would be associated with the melt fraction. We discuss the geophysical and geochemical consequences of a globally distributed melt layer at the base of the Earth’s upper mantle

Methods

Results

Conclusion