GRAVITY-DRIVEN STRAIN IN THE NORTH AMERICAN GREAT PLAINS

Gravitational body forces are the accepted explanation for modern deformation in both the Basin and Range and the Tibetan plateau. Yet topography is neither sufficient nor necessary to produce such stresses, which result from lateral contrasts in lithostatic pressure. To estimate body forces in the central United States, we derive a 3D lithospheric density model from seismic velocity, gravity, topography, and heat flow. Anomalous lower crust underlies each of the three highest concentrations of natural seismicity in the region. The Yavapai-Mazatzal boundary (Cheraw fault, SE Colorado) and Cheyenne Belt (Wheatland/Whalen faults, SE Wyoming) host buoyant, likely hydrated lower crust and extensional seismicity with regionally anomalous NW-SE moment tensor T-axes. Finite-element modeling predicts 10 MPa of localized tension in the seismogenic crust along these features and the observed 45-90° rotation of principal stress direction. Along the flanks of the Midcontinent Rift (Nebraska and Kansas), the contrast between dense rift-related intrusions and the flanking sedimentary basins creates up to 15 MPa of modeled rift-normal compression, the principal horizontal direction independently documented by earthquake moment tensors. In addition to the concentrated natural seismicity, a high proportion of injection wells in these three regions are associated with anthropogenic earthquakes (in the Raton basin, on the margins of the Powder River basin, and in southcentral Kansas, respectively). By contrast, in regions where we model low gravity-derived differential stress and there are abundant injection wells – the Denver basin and much of the western half of Kansas – induced seismicity is generally rare. Even in the low topographic relief central U.S., gravitational forces strongly influence the direction and magnitude of deviatoric stress, making specific areas more prone to both natural and anthropogenic earthquakes.