The primary goal of the proposed work is to identify the scale or scales of the processes that control melt transfer in migmatitic rocks. This information will allow the production of realistic models of transport properties of migmatites with respect to melt, which will serve to perfect integrated numerical models of melt segregation and migration within, through, and out of migmatitic rocks. Specific steps will include: 1) the characterization, using high-resolution X-ray computed tomography, of the 3-D distribution of leucosome in a migmatite (as a proxy for mesoscopic melt advection network) from the Onawa contact metamorphic aureole; 2) the derivation of mesoscopic permeability estimates from deterministic models of flow through the network defined in 1; 3) the characterization of inter-leucosome compositional gradients (i.e., in mesosome) within that migmatite, using the electron microprobe analyzer and image analysis techniques; 4) the derivation of bulk diffusivities associated with these gradients by applying appropriate analytical solutions to the diffusion equation, as well as error minimization methods; 5) comparison of the resulting rates of inter-leucosome material transport with rates of melt advection in the leucosome network, in order to identify the scale or scales of the processes that control the overall movement of melt through these rocks; and 6) assessment of different types of spatial distribution models for transport properties of migmatites, in light of these results. Developing models of melt transfer in migmatites is a necessary step in understanding crustal-scale melt transport because any geometric model of focussed granite ascent requires extensive migration of material through the source region.
