Effects of melt migration on the dynamics and melt generation of diapirs ascending through asthenosphere

Abstract

[1] Melting of a plume head can affect its dynamics by creating melt retention buoyancy. Melt migration controls the distribution of the melt retention buoyancy and affects the dynamics of the plume. We investigate in detail the effects of melt migration on the dynamics and partial melting of a 20-km radius mantle diapir, using axisymmetric two-phase flow models. We first study melt migration in a diapir with a 10% initial melt where no further melting is allowed. The diapir dynamics are modeled for permeable and impermeable cases. In the permeable model, melt migrates within the diapir, whereas no relative motion is allowed between melt and solid matrix in the impermeable model. The permeable model shows a progressive increase of melt fraction in the top portion and decrease in the bottom part of the diapir. This results in a melt buoyancy polarization that elongates the diapir and increases its upwelling velocity. We then model the dynamics of the permeable and impermeable diapirs allowing melting to occur. The velocity of the permeable diapir and its melt generation are significantly larger than that of the impermeable diapir. In general, the permeable diapirs lack the mushroom shape observed for the impermeable ones. Because of intensive computational demand of the two-phase flow modeling, our detailed studies are limited to small diapirs. However, we also investigate a 200-km-diameter mantle plume and show that melt migration produces a nibble of higher melt fraction at the top of the plume that ascends much faster than the bulk of the plume.

Melting phase relations of the Udachnaya-East Group-I kimberlite at 3.0–6.5 GPa: Experimental evidence for alkali-carbonatite composition of primary kimberlite melts and implications for mantle plumes

“…Dobretsov and Shatskiy, 2012). A number of scientists have suggested that this is possible because of the high ascent rate of plumes via the liquid-phase enhanced creep (an order of magnitude faster than the solid-state creep) promoted by the presence of partial melt (Ghods, 1999;Ghods and Arkani-Hamed, 2002;Dobretsov and Shatskiy, 2012;Kirdyashkin and Kirdyashkin, 2013;Shatskiy et al, 2013a). The migration properties of a partial melt (e.g.…”

“…Experimental and numerical modeling demonstrate that the migration rate of a melt within plume is significantly faster than the ascent rate of the solid-state plume (0.2-20 m/year vs a few centimeters per year) and this implies that melt would be accumulated in the top of the mantle plume (Hammouda and Laporte, 2000;Ghods and Arkani-Hamed, 2002;Shatskiy et al, 2013a).…”

“…This suggests disequilibrium conditions as the kimberlite travels through the lithosphere by hydraulic crack propagation. This mechanism of melt migration is not viable beneath the lithosphere, and melt can reach the base of the lithosphere only by the ascent of the melting front (Ghods and Arkani-Hamed, 2002). In the sublithospheric mantle dissolutionprecipitation should be the dominant mechanism of melt migration, which implies continuous reequilibration between melt and surrounding mantle (Hammouda and Laporte, 2000;Shatskiy et al, 2013a).…”

“…The plume models, which considered only thermal buoyancy, constrain a number of prerequisites for plumes: very high temperature, thick conduit (thick zone of low shear-wave velocity) and mushroom shape (see review in Ghods (1999) and Ghods and Arkani-Hamed (2002)). However, seismic observations under the present-day active hotspots indicate only narrow zones of low shearwave velocity and no mushroom shape (see review in Ghods (1999) and Ghods and Arkani-Hamed (2002)).…”

Porosity‐driven convection and asymmetry beneath mid‐ocean ridges

“…That study incorporated Darcian magmatic flow, but neglect compaction and associated stresses. Other authors have considered the two-phase dynamics of diapirs and plumes away from ridges [Tackley and Stevenson, 1993;Ghods and Arkani-Hamed, 2002;Schmeling, 2010].…”