A glance at any geological map of North America reveals a large area of relatively young and essentially undeformed sedimentary rocks extending from the McKenzie River delta to the Gulf of Mexico. In Canada these rocks are Devonian (around 400 m.y.) to Tertiary in age, although the rocks exposed at surface are mostly Cretaceous.  They include thousands of metres of strata with significant oil reserves and coal deposits and with thick evaporite beds. Most of the sediments were deposited in marine environments, although some terrestrial sedimentation did take place, especially in late Cretaceous and Tertiary times (since 70 m.y. ago).

The predominance of marine rocks within these sequences indicates that for several hundred million years much of central North America was below sea level. Elevations throughout most of this area are now consistently above 500 m, in some parts more than 1500 m above sea level.

The significant vertical displacement of huge areas of continental crust in North America and elsewhere, is now attributed to mantle processes, as has been recently summarized by CalTech geologist Michael Gurnis. Two quite different processes appear to be responsible for the ups and downs of the crust. One of these is the existence of hot mantle "super plumes", and the other the slow descent of cold subducted oceanic slabs.

Seismic tomography is the use of earthquake-generated seismic waves to provide a 3-dimensional picture of the mantle velocity.  For the past 20 years seismologists have been using this information to map the distribution of relatively warm (low velocity) and relatively cool (high velocity) mantle rock, and we now have a pretty clear picture of where the cool and warm areas are, and how the  convection systems are distributed. The most conspicuous anomaly in the mantle is the African Super-Plume, an upwelling of hot mantle material several thousand kilometres across which contributes to the significant elevation of most of Africa*. 

The explanation of crustal down-welling – of the type that might have explained central North America’s submerged past – is more complex. It involves looking at the nature and timing of past subduction, and the fate of the relatively cold subducted oceanic slabs. Up until about 40 m.y. ago the Farallon Plate was being pushed down under western North America. There is evidence that this subduction took place at an uncharacteristically shallow angle, which accounts for some of the current elevated topography of the western U.S. (see the article on The history of subduction beneath western North America). For much of its subduction history, therefore, the Farallon Plate did not sink deep into the mantle until it reached the central part of the continent. This sinking slab created a downward flow of mantle in its wake, pulling down the continental rocks in the process. Once the slab had descended to a sufficient depth it no longer exerted a significant pull, and North America bobbed back up into its natural position above sea level.

There is similar evidence that part of Australia was submerged during the Cretaceous, and again this can be traced to an oceanic slab which originally subducted along the east coast of that continent.

Reference

Gurnis, M., Sculpting the earth from inside out, Scientific American, March 2001

Steven Earle, 2000. Return to Earth Science News