Deep storage of oceanic crust in a vigorously convecting mantle

Abstract

[1] Fractionated isotopic ratios in some oceanic basalts indicates the presence of recycled oceanic crust in the mantle. This crust must have escaped complete remixing for a significant period of time. Gravitational settling into a dense layer at the base of the mantle may facilitate this preservation. Christensen and Hofmann (1994) first demonstrated the dynamics of this process by developing scaling laws for extrapolating low convective vigor models to conditions estimated for the mantle. Here this sequestration is studied in models with more Earth-like convective vigor. Scaling laws for geodynamic parameters are updated and the isotopic evolution of the U-Pb and Sm-Nd systems examined. Significant accumulation is still found at high Rayleigh number, but only when the excess density of oceanic crust in the lower mantle is larger than currently suggested from laboratory experiments. These accumulations are found to maintain the fractionated isotopic signature of ancient crust for models with moderate to moderately high convective vigor relative to mantle estimates. At the highest convective vigor tested, the accumulations are not isotopically distinct.Citation: Brandenburg, J. P., and P. E. van Keken (2007), Deep storage of oceanic crust in a vigorously convecting mantle,

The effect of bulk composition and temperature on mantle seismic structure

“…subduction of differentiated oceanic lithosphere. Moreover, dynamical mechanisms have been proposed for segregating basalt from harzburgite in a convecting mantle, including slab delamination, and segregation due to the density contrast between basalt and harzburgite (Brandenburg and Van Keken, 2007;Christensen and Hofmann, 1994;Davies, 2006;Nakagawa and Buffett, 2005;Ringwood and Irifune, 1988;Xie and Tackley, 2004). Segregation of basalt from harzburgite provides a mechanism for producing radial and lateral variations in basalt fraction.…”

“…Allégre and Turcotte (1986) suggested a marble cake structure for the mantle in which subducted oceanic lithosphere is deformed into pervasive, narrow pyroxenite veins. Mantle convection simulations suggest a heterogeneous mantle made of a mechanical mixture of basalt and harzburgite (Brandenburg and Van Keken, 2007;Christensen and Hofmann, 1994;Davies, 2006;Nakagawa and Buffett, 2005;Xie and Tackley, 2004). A stirring time of the mantle between 250 and 750 Ma (Kellogg et al, 2002) limits the amount of stretching and folding of subducted heterogeneity that can occur in the mantle over the age of the Earth.…”

Continent-sized anomalous zones with low seismic velocity at the base of Earth's mantle

“…These are expected to have sharp compositional boundaries, which limit heat and chemical exchange with the surrounding mantle. On the other hand, piles may have been continuously forming from the early Earth to the present day, for example, by the accumulation of subducted oceanic crust 20,48,49,54 , or chemical reactions between the mantle and core 55,56 . One complication for the oceanic crust hypothesis is that it is difficult to accumulate intrinsically more-dense crust of present-day thickness (6 km) at the CMB because viscous stresses from mantle convection act to stir it into the background mantle 57 .…”

“…In the Pacific LLSVP, this has been interpreted as a lens of post-perovskite in compositionally distinct material, because the double-crossing of the post-perovskite phase transition is unlikely in higher-temperature regions of pyrolitic mantle 18 (though it is important to note that the deepest mantle temperature is not well constrained). Numerical convection calculations have shown that subducted oceanic crust is expected to become incorporated into thermochemical piles [19][20][21] . This basaltic material contains bridgmanite and free silica, which individually undergo phase transitions (to post-perovskite and to seifertite, respectively) at different lowermost depths 22,23 , and therefore, may contribute to a cause of the seismic wave reflections.…”

Subduction-transition zone interaction: A review

“…If there is any chemical difference, it would need to result in a density jump at 660 km depth that is less than 6%, or else it would lead to fully layered mantle convection (Christensen and Yuen, 1984), which seismic tomography, plate evolution modeling, and considerations about the convective and thermal evolution of the Earth have shown is implausible (Silver et al, 1988;Ricard et al, 1993;Van der Hilst et al, 1997;McNamara and Van Keken, 2000). Furthermore, if it does not lead to layered convection, a chemical interface cannot survive over many convective cycles, although chemical gradients may persist (Van Keken and Zhong, 1999;Tackley et al, 2005;Brandenburg and Van Keken, 2007). For example, Ballmer et al (2015) invoked compositional density gradients, due to a progressive enrichment in basaltic components with depth in the lower mantle, as a mechanism to stagnate slabs within the top of the lower mantle.…”