Abstract

Vail et al. (1977) recognized that there was no plausible mechanism in non-glacial times for their proposed higher frequency (“third-order”) cycles of sea-level. An explanation for these one to ten million year events has been elusive: at least this part of the Vail eustatic model remains unproven. A review of research on mantle processes over the past three decades shows that mantle-induced vertical motions of Earth’s surface can occur over time intervals from several tens of millions of years down to less than one million years, and can have amplitudes of hundreds of metres, even at the shorter intervals. These vertical motions can occur worldwide and are not merely associated with major hotspots. Mantle convection therefore provides a mechanism for explaining some second- and third-order sea-level cycles. The planform pattern of the convection exerts a first-order control on stratigraphy by creating swells or depressions with typical diameters of one to two thousand kilometres, and durations of uplift or subsidence from a few million years to tens of millions of years. Vertical motions that result from episodic pulsing of mantle plumes are characterized by rapid rates over intervals of several million years down to less than one million years. These effects are evident in the sedimentary record of North Atlantic basins, with peak uplift seen at Middle Jurassic, Early Cretaceous, Eocene and Neogene. Caution is therefore required before interpreting rapid (less than one million years) and high amplitude (tens of metres) sea-level fluctuations as eustatic in origin.