Dynamic topography and long-term sea-level variations: There is no such thing as a stable continental platform

Globally uniform (i.e., eustatic) sea-level trends with time scales approaching 100 Myr have been inferred from both seismic and backstripping stratigraphic analysis at a small set of geographic sites that are presumed to lie on stable continental platforms and passive continental margins characterized by simple thermal subsidence histories (e.g., the New Jersey margin, the western African margin). We demonstrate, using mantle flow simulations based on high resolution seismic tomography, that both the New Jersey margin and the conjugate western African margin have been subject to orders of 100 m of dynamic (i.e., flow induced) topography change over the last 30 Myr. We also show that the changing pattern of downwelling mantle flow associated with plate subduction is a significant contributor to the background eustatic sea-level trend, which is also of order 100 m during the 30 Myr time window. Therefore, Late-Cenozoic variations of dynamic topography on these passive margin sites are comparable to the eustatic sea-level changes and can partially mask the latter. Furthermore, even if the observed trend could be accurately corrected for local dynamic topography variation, the residual eustatic signal does not merely reflect changes in mean spreading rates at mid-ocean ridges. We conclude that the observed long term sea-level variations at so-called \textquotedblleftstable\textquotedblright sites cannot be interpreted as eustatic. Moreover, previous analyses that have used long-term sea-level trends as a proxy for spreading rates and geochemical fluxes must be revisited.