Development of finite strain in the convecting lower mantle and its implications for seismic anisotropy

Abstract

Seismological observations have revealed patches of seismic anisotropy in regions related to mantle upwelling and paleosubduction within an otherwise isotropic lower mantle. A combination of numerical modeling and mineral physics is used to constrain the source of anisotropy in these regions in an effort to better understand lower mantle dynamics and mineral physics. Specifically, it is investigated whether lattice-preferred orientation (LPO) can explain the anisotropy observed in regions of paleosubduction. Since LPO is caused by dislocation creep and is destroyed by diffusion creep, we can develop deformation mechanism maps to determine which regions allow for the development of a mineral fabric. Strain is then calculated in these regions and is related to mineral physics experiments combined with high-pressure elastic constants of lower mantle minerals in order to assess the predicted seismic anisotropy. Uncertainties in rheological parameters such as the transition stress between dislocation creep and diffusion creep necessitate a full evaluation of the parameter range. The effect of variations in transition stress, activations parameters, and strength of slabs on fabric development is investigated. It is shown that LPO is a likely candidate for the cause of lowermost mantle anisotropy in regions of paleosubduction.