Partial Melt Systems in Plate-Driven Corner Flow: Evaluating the Formation of Porosity Bands as a Mechanism for Magma Focusing at Mid-Ocean Ridges

Abstract:

The imposition of an external shear on a system of partial melt will result in compaction of the solid matrix and concentration of the interstitial liquid melt leading to the formation of regions of contrasting high and low porosity. In experiments, direct and torsional shear geometries have demonstrated that these regions of varying porosity form in bands orientated at low angles relative to the shear plane. A variety of numerical models have been employed to recreate these experimental results. Simple shear, pure shear and torsional shear geometries have been used in both linear and nonlinear numerical settings to model the formation of the porosity bands. In this contribution the numerical models utilize a shear geometry derived from the velocity field of the plate-driven corner flow of a mid-ocean ridge. Motivation for using the velocity field of a mid-ocean ridge comes from evidence that suggests the existence of lateral melt channeling from either side of the ridge axis. Imposing the shear from a mid-ocean ridge corner flow allows for the evaluation of the resulting porosity bands in terms of suitability for channeling melt laterally toward the ridge axis. This is done using both slow and fast spreading ridge geometries. The degree of similarity between previous numerical and experimental results has been found to be greatly influenced by the imposed viscosity law of the solid matrix phase. In order to keep this in mind, the numerical models in this contribution use three different matrix viscosity laws: strain-rate independent, strain-rate dependent and anisotropic. Of these rheologies, strain-rate independence results in the poorest orientation for channeling melt directly to the ridge axis. The strain-rate dependent and anisotropic viscosities present more favorable direct-channeling orientations for the fastest growing porosity bands, but in both cases the background flow will rotate bands to less ideal orientations over time. However, these less favorable orientations for direct melt channeling may result in a more confined region of melt channeling at the base of the plate where a high porosity layer is proposed to exist due to the eventual freezing of upwelling mantle melt. In this high porosity region, melt would be transported toward the ridge axis along the base of the plate under the influence of gravity.