The chemical evolution of igneous rocks is fundamentally a process of crystal-liquid segregation. The objective of this research is to develop and extend a numerical approach to multiphase crystallization that explicitly allows for relative motion between crystals and melt simultaneously in the interior of the magma body and the boundary layers at the roof, walls and floor. This will include many processes that are central to geological applications such as thermal and compositional buoyancy, variable viscosity, kinetic effects of solidification and melting, crystal settling, crystal plumes and instabilities, compositional convection in regions of crystal mush, and entrainment of crystals from magma chamber floors. Some of the model is already implemented but important features relating to the details of nucleation and eutectic behavior and to variable rheological properties and functions require further refinement. Specific applications of this model to the cooling and dynamics of mid-ocean ridge basaltic systems and to systems involving mixing and disruption of cumulate piles by convective activity will be explored.
