A simplified model of continental extension including lower crustal flow is developed. The model employs the thin sheet approximation in estimating lithospheric yield strength and gravitational buoyancy forces arising from lateral variations in crustal thickness and temperature. The effect of advection and diffusion of heat on the temperature structure and yield strength of the extending region is calculated. The viscosity of the lower crust, which controls its rate of flow, is estimated in a manner consistent with the yield strength calculation. The change in the force required to extend the lithosphere is calculated after a finite amount of extension. When this force increases, the zone of extension is assumed to widen; when it decreases, the extension remains localized. Material parameters affecting the model results are the density difference between the crust and mantle, the gradient of brittle yield stress with depth, the abundance of heat-producing elements within the crust, and the ductile strengths of the crust and mantle as a functions of temperature and strain rate. The model predicts three distinct modes of extension depending on the model crustal thickness, heat flow, and strain rate.

The modes are (1) core complex mode (concentrated upper crustal extension with lower crustal thinning over a broad area and any mantle lithosphere extending in a local area), (2) wide rift mode (uniform crustal and mantle lithospheric thinning over a width greater than the lithospheric thickness), and (3) narrow rift mode (concentrated crustal and mantle lithospheric extension). Models with a dry olivine mantle rheology and a dry quartz or dry anorthosite crustal rheology are consistent with geologic observations of crustal thickness and heat flow for all three extensional modes. An analysis of the force changes due to extension, ignoring conduction of heating is carried out to give greater insight into how variations in crust and mantle rheologic parameters affect model results. Crustal rheologies several orders of magnitude stronger or weaker than dry quartz give results inconsistent with observations. The model implies a temporal progression from core complex mode to wide rift mode to narrow rift mode. Data from the southern Basin and Range Province are consistent with this progression. ¿American Geophysical Union 1991