Gradual entrainment of a chemical layer at the base of the mantle by overlying convection

Abstract

SUMMARY A dense chemical layer may exist at the base of the mantle. The overlying convection tends to entrain some of this layer reducing its thickness through time. The lower more dense fluid is dragged into cusps beneath upwelling limits of the overlying convection and some material is entrained. It is obvious that a high density of intrinsically more dense material retards this entrainment. An upper limit, and probably a reasonable estimate for the amount of entrainment, is obtained by considering the carrying capacity of the vertical flow above cusps and ignoring the complications within the cusps themselves. Flow through conduits formed by the reduced viscosity of hot material is less effective at entrainment than broad-scale flow controlled by the background viscosity of the normal mantle. Cylindrical plumes are less effective at entrainment than 2-D flow. Intrinsic density contrasts of around 6 per cent are needed for the layer to persist through geological time (assuming the thermal expansion coefficients is 2 × 10-5). In addition, the chemical layer may convect internally. This causes the layer to entrain less dense material from above and thus aids entrainment from above by reducing the excess density of the layer. Convection in the chemical layer and stripping of the layer into cusps by the overlying flow also cause much of the temperature contrast to be in the overlying mantle rather than the chemical layer. The difficulty of having more than a few 100 °K temperature contrast across the layer, unless it is internally stratified, negates a reason for supposing the layer exists.