LAUSR

Abstract Two large, seismically slow regions in the lower mantle beneath Africa and the Pacific Ocean are sometimes referred to as “superplumes”. This name evokes images of large-scale active upwellings. However, it remains unclear whether these features are real or represent collections of multiple regular mantle plumes. Here, we investigate the implications of these upwellings for dynamic topography. We combine detailed measurements of oceanic residual topography from Hoggard et al. (2016) with continental constraints derived from CRUST1.0 to produce a global model expanded in spherical harmonics. Observed dynamic topography is subsequently compared to predictions derived from mantle flow following Steinberger (2016) using tomographic density models. Results yield relatively good overall agreement and amplitude spectra with similar slopes, except for degree two (i.e. > 10,000 km wavelengths) where predicted amplitude is more than two times as large and is dominated by contributions from the lower mantle. Predictive models suggest two large-scale uplifted regions above the “superplumes” that are barely seen in the observed topography. We suggest that this mismatch can only partly be reconciled by altering the seismic velocity to density conversion factor or by including the effects of lower mantle chemical heterogeneity. In addition, it may be important to consider more significant revisions to the lower mantle flow patterns, such as those possibly induced by different radial viscosity profiles and laterally-varying or anisotropic lower mantle viscosity.