Subduction at plate boundaries can have thermal, chemical, and physical impacts on broad regions of the continental interior, but these interactions are not as readily obvious as deformation near the continental margin. Such cryptic alteration has produced surface uplift in the Colorado Plateau and western Great Plains of North America, which have risen—largely undeformed—1.6 and 1.3 km, respectively, relative to the eastern Great Plains during the Cenozoic. Accumulation of Cretaceous–Cenozoic sediments accounts for only 300 m of uplift of the Colorado Plateau and 400 m of the western Great Plains, leaving 1.3 km and 0.9 km, respectively, unexplained. To determine the physical causes of this enigmatic epeirogeny, we derived three-dimensional (3-D) lithospheric density models from seismic velocity, gravity, topography, and heat-flow data. Lower-crustal density decreases systematically westward across the Great Plains, accounting nearly perfectly for the remaining 900 m of uplift of the western Great Plains and the modern east-west topographic gradient. Lower-crustal dedensification beneath the Colorado Plateau accounts for a similar 900 m of uplift. Lower-crustal xenoliths in both regions show progressive hydration-induced retrogression of garnet-bearing assemblages with increasing modern elevation, and Th-Pb dating of the Colorado Plateau retrogression gives end-Cretaceous dates (xenoliths from the Great Plains have not yet been dated). We hypothesize that lower-crustal density variations—and much of the surface relief—in North America’s Proterozoic interior terranes reflect varying degrees of metasomatic retrogression, such as by fluids exsolved from the Farallon slab. The remaining 400 m of Colorado Plateau uplift is most plausibly due to elevated mantle temperature. We present thermal models that suggest that 25–70 km of Cenozoic lithospheric thinning can explain the modern elevation and density structure.

Gold Open Access: This paper is published under the terms of the CC-BY-NC license.

Supplementary data