Subducted slabs and the geoid: Constraints on mantle rheology and flow

Abstract

Geoid anomalies are primarily the result of the density contrasts driving mantle convection and plate motion. The total geoid anomaly resulting from a given density contrast in a convecting earth is affected by the mass anomalies associated with the flow‐induced deformation of the upper surface and internal compositional boundaries as well as by the density contrast itself. These boundary deformations, and hence the total gravity field, depend on the radial distribution of effective viscosity. If the internal density contrasts can be estimated, as is the case for subducted slabs, useful constraints can be placed on the depth and on the variation of viscosity with depth of the convecting system. The degree 4–9 components of the observed long‐wavelength geoid are highly correlated with those predicted by a density model for seismically active subducted slabs. The (positive) sign of the correlation requires that the effective viscosity increases with depth by a factor of 30 or more. The amplitude of the correlation cannot be explained by the density contrasts associated with just the seismically active parts of subducted slabs, however. The amplitude can be explained if the density contrasts associated with subduction extend into the lower mantle or if old lithosphere is piled up at the base of the upper mantle beneath subduction zones to a thickness in excess of 350 km over horizontal distances of thousands of kilometers. Mantlewide convection in a mantle that has a viscosity increasing with depth provides a simple explanation of the long‐wavelength geoid anomalies over subduction zones.