Convective Mixing in the Mantle: The Role of Viscosity Differences

The fate of the subducted oceanic crust has been a subject of controversy since the hypothesis of plate tectonics was first proposed. There is now considerable evidence to support the mantle-mixing hypothesis according to which the subducted oceanic lithosphere is heated and softened and then becomes entrained in the convective flows in the mantle. Normal strains stretch and thin the oceanic crust until diffusion processes complete the homogenization. Since solid-state diffusion processes are effective only on the scale of centimetres, then 6 km thick oceanic crust must be stretched by a factor of 105 or more before losing its identity. Direct observational evidence for this hypothesis comes from studies of high-temperature peridotite massifs. In these massifs, pyroxenite layers with a basaltic composition are embedded in a lherzolitic matrix; the pyroxenite layers with thicknesses varying from centimetres to a few metres may be identified with the stretched, subducted oceanic crust. With the aim of improving our understanding of the role of convective mixing, we have carried out a study of the influence of viscosity differences on the stretching process. We have obtained equations that describe the stretching of a symmetrical two-dimensional body of highly viscous fluid embedded in a normal strain flow. These can be solved in closed form when the body is elliptical, and asymptotically for a general shape. We find that for mantle conditions, viscosity differences are unlikely to inhibit the stretching of the subducted oceanic crust significantly.