Abstract

SUMMARY Seismic anisotropy in the Earth, particularly in the mantle, is commonly interpreted as the result of solid-state deformation by dislocation creep that induces a lattice preferred orientation (LPO). Diffusion creep operates where stress levels are lower and/or grain sizes smaller. It is often assumed that diffusion creep induces grain rotations that eventually destroy any existing LPO. A new numerical test of this assumption shows that it is not necessarily the case: diffusion creep will create some relative grain rotations, but rotation rates decrease through time. Hence, when microstructural change due to diffusion creep dominates that due to grain growth, defined here as ‘type P’ behaviour (the converse being ‘type O’ behaviour), the model indicates that LPO will be weakened but preserved (for a variety of strain paths including both pure and simple shear). One measure of anisotropy is the proportional difference in shear wave velocities for different polarization vectors (AVs). A model olivine microstructure with equant grains and initial maximum AVs of 10.0 percent has this value reduced to 6.7 per cent when ‘rotational steady state’ is attained. Other models with different initial maximum AVs values exhibit final maximum AVs values more than half the initial values. If the grains are initially elongate by a factor of 2, maximum AVs is reduced just slightly, to 8.5 per cent. Thus, when grain growth plays a subordinate role to the deformation, diffusion creep weakens seismic anisotropy by a factor of less than 2 (using maximum AVs as a measure and olivine as an example). Consequently, the link between seismic anisotropy and deformation mechanism in the mantle requires reappraisal: regions with LPO may comprise material which once deformed by dislocation creep, but is now deforming by diffusion creep in a rotational steady state.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call