We propose a detailed mass budget analysis of the solid Earth, that is mantle and lithosphere, from geodynamic a-priori considerations. For the lithosphere and upper mantle, we will build on a published 3-D density model (3SMAC, Nataf and Ricard, 1996). For the deeper mantle, we will infer thermally induced density anomalies from a history of plate motion assimilated in global mantle circulation models. Key advances to the mantle convection model include (1) a thermodynamically self-consistent formulation of the mantle mineralogy and (2) a very high numerical resolution sufficient to resolve (for the first time) global mantle flow at Earth-like convective vigor. The benefit of the latter is to resolve thermal boundary layers at the time-and length-scales appropriate for the Earth’s mantle. We will use the forward models to gain better insight on the impact of hot thermal upwellings on the mantle density budget. This is important because recent geodynamic studies have favoured a larger heat flux across the core mantle boundary than prior estimates. We will test the sensitivity of mantle density models to a range of core heat flux values.