3D density model of the western US lithosphere: Insights on chemistry, temperature, topography, and intraplate stress

Thursday, 18 December 2014
Will Levandowski, USGS, Golden, United States and Craig H Jones, University of Colorado at Boulder, Boulder, CO, United States
Although seismic velocity generally reflects material density, density models drawn solely from seismic interpretations suffer from the fact that temperature, melt-depletion, variations in quartz content, and in-situ melt—as well as myriad other factors—have different effects on the relationship between velocity and density. To wit, such models generally do not accurately recover gravity and/or topography variations. We have developed a probabilistic density modeling method that estimates density from seismic velocity and heat flow and refines these initial estimates in order to reproduce gravity and topography, accounting for lithospheric flexure. Both the input seismic velocity modeling and the refinement are Monte Carlo-type approaches, so the posterior distribution of models provides a direct measure of uncertainty. We leverage the aforementioned difference in sensitivity to separate density variations into thermal and compositional components, providing information on the chemistry and physical state of the crust and upper mantle. Using this approach and Transportable Array seismic data, we present a density model of the western US lithosphere (from central Kansas west) to a depth of 150 km that reveals: 1) remarkably uniform, near- to supra-solidus mantle temperatures beneath regions deformed in the Cenozoic--including the Colorado Plateau--that are ~400 °C higher than those beneath the nominally stable interior of North America; 2) crustal melt (~1%) beneath Miocene-Recent volcanic provinces; 3) depleted mantle lithosphere beneath the Wyoming craton and northern High Plains; 4) likely hydrated lower crust in the Colorado Plateau and Great Plains; and 5) that horizontal differences in lithostatic pressure create deviatoric extensional stress of ~10 MPa in the northern Basin and Range and along the margins of the Colorado Plateau. This density model is a rich source of information, shedding light on the causes and consequences of tectonism, crust-mantle interactions, and topography in the western US.