Investigating how the cross-sectional shape of ULVZs provides information on the intrinsic viscosity of LLSVPs and ULVZs

Abstract:

Seismic tomography models reveal two Large Low Shear Velocity Provinces (LLSVPs) that are hypothesized to be caused by large-scale, long-lived compositional reservoirs in the Earth’s lowermost mantle beneath central Pacific and Africa. Because the composition of the LLSVPs can’t be determined directly, their origin remains elusive. High resolution seismic studies have detected Ultra Low Velocity Zones (ULVZs) oftentimes occurring at the margins of or within the LLSVPs as tiny disparate patches around two-order-of-magnitude smaller than the LLSVPs. Core-reflection precursor seismic studies can map out the size and shape of ULVZs more precisely. If the morphology of ULVZs could provide insights into the rheological nature of LLSVPs, this may help constrain the origin of LLSVPs. Earlier geodynamical studies have found that ultra-dense materials could formulate ULVZs with asymmetrical cross-sectional shapes along the margins of large compositional reservoirs (which are hypothesized to be the cause of LLSVPs). When background mantle, compositional reservoirs, and the ultra-dense ULVZs have the same temperature-dependent rheology, the ULVZs are thicker on the outside edges of LLSVPs (and thinner on the inside edges) due to differential viscous coupling. If the compositional reservoirs have a larger grain size than the background mantle, their intrinsic diffusion creep viscosity will increase. If the ULVZs are enriched in iron and partial melt or have a larger grain size, the intrinsic viscosity of ULVZs could decrease or increase correspondingly. This work is a continuation of Marin and McNamara (2016). We performed spherical geodynamical calculations to explore how the change of intrinsic viscosity of LLSVPs and ULVZs affect the shape of ULVZs. We find that higher intrinsic viscosity of reservoirs and ULVZs leads to thicker and narrower cross-sectional shapes of ULVZs that are more symmetrical in shape, while lower intrinsic viscosity leads to more flattened and less symmetrical-shaped ULVZs. In addition, ULVZs’ inboard height-width ratio is higher if LLSVPs are intrinsically more viscous than ULVZs. Combined with seismic observations on the cross-sectional shapes of ULVZs, these results can provide insight into the rheological nature and origin of LLSVPs and ULVZs.