PASSIVE ORIGIN OF THE MIDCONTINENT RIFT AND UPLIFT OF THE COLORADO PLATEAU AND HIGH PLAINS BY LOWER CRUSTAL HYDRATION: EVIDENCE FROM HIGH PRECISION DENSITY MODELS

The midcontinent rift (MCR) is the most prominent gravity high in the central U.S., but its density anomaly does not generate surface relief, and it is collocated with low shear-velocity upper crust. Precise knowledge of the enigmatic 3D density structure of the region may illuminate not only the mechanics of rifting but also the origin of topographic relief between the broader midcontinent and other nominally stable Proterozoic regions that were at sea level during the Cretaceous--namely the Colorado Plateau (2 km elevation) and the High Plains (>1 km). We improve upon a density modeling technique that was previously successful in understanding topography and gravity in the western U.S. but which fails utterly near the MCR. Our new automated, probabilistic algorithm jointly satisfies gravity, topography, receiver functions, surface wave dispersion to 80s (~150 km depth), and heat flow. It has a precision of +/- 5 mGal and +/- 20 meters of elevation. The final density model reveals 1) an increase in mafic crustal rock along strike of the rift, consistent with initiation of the rift as a wedge-shaped, passive feature; 2) systematic south (Proterozoic) to north (Archean) increase in mantle depletion that is not enhanced beneath the rift, suggesting that this depletion is ancient and that rift magmas were generated below 150 km; and 3) that ~0.9 km of Colorado Plateau and ~0.5 km of High Plains relief relative the non-rifted midcontinent is due to crustal density differences. These differences increase systematically with depth, consistent with density loss by lower crustal hydration, which is documented by xenoliths in both provinces. Hydration is Laramide in age in the Colorado Plateau and undated in the Plains.