A caveat applies to the use of densities of evolving magmatic melts in the modelling of fluid dynamic processes in magma chambers. Models that are dependent on the concepts of stratified liquids within a magma chamber, or the rising of liquid plumes through such columns until a level of neutral buoyancy is reached, often assume that the melts are entirely crystal-free. Both field and physical evidence suggest, however, that this assumption may not be true and that melts at or below the liquidus commonly retain some crystals in suspension. The effective density of the melt will then include the contribution of the suspended crystals. Because of the large-scale density difference between the liquid and some early-crystallized solid phases, as little as ca. 1.5% of suspended load will overwhelm any density changes that may be attributed to the evolution of a cooling, fractionating melt that is assumed to have shed its crystal load. Although there are many factors involved in the dynamics of magma chambers (e.g., evolving viscosity), this density effect alone raises profound doubts about the validity of some current models of magma chambers.
Read full abstract