Cooling of the upper mantle as energy source of convection

Abstract

For heat-induced effects, the cooling of a fluid -layer at its top is formally indistinguishable from heating at the bottom. This cooling can be taken as the chief energy source of convection. The density difference of material rising into the crust is a weak second. The extreme shallowness of the pattern- can largely be explained by the known layering: a hard and brittle lithosphere on top of a deformable but relatively thin asthenosphere. No model of upper-mantle convectionis fully satisfactory unless it gives a hint of why the observed motions seem strongly confined to the last 5% of the earth's life. Only two geophysical effects are known that yield proeesses of the duration of gigayears: termal conduction, and radioactive decay. The simplest model is one in which, to start with, the mantle was uniformly hot to the top. The "cold" penetrates gradually into depth, the processes being slowed down in its earlier stages by radioactive heat, more uniformly distríbuted tan now. This, together with a plausible hypothesis about the distribution of some viscosity lowering compounds (H 20, C02) seems to be enough to account for a great deal of what is known.