Rayleigh-Taylor Instability and Mantle Dynamics Beneath Mountain Belts

University Of Colorado At Boulder, Boulder CO

Investigators

Abstract

Molnar 0106909 In this ABR project the PI will continue work with Greg Houseman and colleagues that examines Rayleigh-Taylor instability relevant to the earth, with the ultimate goal of understanding how thickened mantle lithosphere can become gravitationally unstable. Most of the work until now has considered simple situations in which only one layer, that simulating mantle lithosphere overlies an inviscid fluid. They plan to examine more thoroughly the effect of an imposed horizontal shortening on a three-layered model: a light viscous layer (like crust) over a heavy viscous layer (like mantle lithosphere), which in turns overlies a layer (like asthenosphere) of low viscosity and slightly lower density than mantle lithosphere. The effect of buoyant crust can profoundly affect the nature of the instability, such that in some situations downwelling blobs of the dense layer form adjacent, not below, the thickest part of the overlying layer and therefore sink beneath the equivalent of the margins of mountain belts. Preliminary calculations show, not surprisingly, that the viscosity ratio that defines the transition between one or two downwellings depends also on the density of the upper layer, with lighter layers favoring two downwellings. The investigators plan to explore the parameter space more thoroughly, in particular, by quantifying the effects of different density ratios and thickness ratios of the two layers on the configuration of downwelling. They will construct "phase diagrams" that show the parameter ranges for which one or two downwellings occur and will examine non-Newtonian viscosity. They will consider the effects of somewhat different boundary conditions from those used so far. In particular, they plan to use conditions that do not mix varying strain rates at the top of the deforming layer, and instead consider how laterally varying viscosity might affect unstable growth. Finally, they will relate the calculations to geological and geophysical observations from belts where mantle lithosphere seems to have thinned or been removed entirely. Processes that motivate the study include the widespread normal faulting and post-orogenic volcanism in mountain belts, both of which must be associated with the evolving sub-crustal lithospheric structure.