Lithosphere and Asthenosphere Properties beneath Oceans and Continents and their Relationship with Domains of Partial Melt Stability in the Mantle

The depth of the lithosphere-asthenosphere boundary (LAB) and the change in properties across the lithosphere, asthenosphere, and LAB in various tectonic settings are captured in a variety of geophysical data, including seismic velocities and electrical conductivity. A sharp drop in shear wave velocity and increase in electrical conductivity can potentially be caused by the appearance of partial melt at or below the LAB but the chemical and dynamic stability of partial melt across lithosphere and at LAB remain debated.

Here I apply the recent models of mantle melting in the presence of water and carbon [1, 2] to evaluate the domains of stability of partial melt both beneath continents and oceans. The model allows prediction of the possible presence, the fraction, and composition of partial melt as a function of depth, bulk C and H₂O content, and fO₂ [3] in various geologic/tectonic settings.

The results show that while a hydrous, carbonated melt is stable only beneath LAB and in the asthenospheric mantle beneath oceans, continental mantle can contain a carbonate-rich melt within the lithosphere. For geotherms corresponding to surface heat flux (SHF) of 40-50 mW m⁻², which also match P-T estimates beneath cratons based on thermo-barometry of peridotite xenoliths [4], the solidus of fertile peridotite with trace amount of CO₂ and H₂O is crossed at depths as shallow as 80-120 km [5]. If elevated geotherms of the Proterozoic and Phanerozoic terrains are applied, carbonatitic melt becomes stable somewhat shallower. These depths are similar to those argued for a mid-lithospheric discontinuity (MLD) where a negative velocity gradient has been detected much shallower than the proposed depth of LAB in many places. With a drop in oxygen fugacity with depth, a freezing of carbonatitic melt may be expected at intermediate depths (~150-200 km). At 200-250 km a hydrous, carbonated silicate melt may reappear owing to the interplay of fO₂ and freezing point depression effect of CO₂ + H₂O, but such melting may be variable depending on redox heterogeneity of the mantle at depths.

[1] O’Leary et al. (2010). EPSL 297, 111-120. [2] Dasgupta et al. (2013). Nature 493, 211-215. [3] Stagno et al. (2013). Nature 493, 84-88. [4] Lee et al. (2011). Ann. Rev. EPS 39, 59-90. [5] Dasgupta (2013). RiMG 75, 183-229.