Intensity and direction of lattice-preferred orientation of olivine: are electrical and seismic anisotropies of the Australian mantle reconcilable?

Abstract

It has been hypothesised that seismic and electrical anisotropy at the base of the lithosphere are caused by strain-induced lattice-preferred orientation (LPO) of olivine [100] axes parallel to present-day plate motion. This would imply that seismic and electrical anisotropy observations can provide geodynamicists with fundamental information for characterising mantle flow. The qualitative agreement between the fast direction of SV-waves and direction of maximum electrical conductance modelled deeper than 150 km below the North Central craton of Australia appear to support a common alignment mechanism, and the observed, anisotropic electrical conductances can be generated by hydrogen diffusivity in a water-poor (<1000 ppm H/Si) olivine mantle. A quantitative test is proposed for the hypothesis that electrical anisotropy is generated by anisotropic hydrogen diffusion rates (D) in olivine. Electrical anisotropy factors are computed using random resistor network models assuming that D[100]≈20×D[010]≈40×D[001]. Electrical and seismic anisotropies calculated from olivine LPO angular distribution functions modelled for a range of shear strains under a simple shear deformation demonstrate that the intensity of olivine [100] alignments (and associated shear strains) that would be required to explain the electrical anisotropy in the mantle below central Australia are significantly greater than predicted by Rayleigh wave anisotropies. The poor agreement between the observed electrical anisotropies and the electrical anisotropies that would be predicted from the Rayleigh wave anisotropies indicates that either (i) electrical anisotropy in the upper mantle below central Australia is not generated by hydrogen diffusivity alone or (ii) the seismic anisotropy is underestimated. The orientation of the olivine [100] axes maxima is inferred to be ∼30° rotated relative to the direction of present-day absolute plate motion (APM) that is determined relative to the hotspot reference frame (HS2-NUVEL1). Both the APM direction that is determined relative to a reference frame defined by requiring no-net rotation of the lithosphere (NNR-NUVEL1) and GPS-derived plate motion vectors fit the geophysical observations of upper mantle anisotropy better. This may support the contention that hotspots are not stationary relative to the deep mantle.