THE RHEOLOGICAL PATH OF SOLIDIFYING MAGMA: CONTROLS, FEEDBACK RELATIONS, AND END-MEMBER SCENARIOS

Conduit models and lava flow models have advanced our understanding of eruptive processes and associated hazards. However, there is still a lot of room for improvement. Many models do not include temperature-dependence of physical and thermal properties, and those that do typically include parameterizations of crystallization and rheological behavior that are overly deterministic, and do not incorporate kinetic factors or disequilibrium. This contribution explores the effects of thermo-rheological feedbacks and disequilibrium (thermal and thermodynamic) in eruptive processes and seeks to identify the most important variables.

The starting point is the concept of “viscosity paths”, introduced by Whittington et al. (2009 Bulletin of Volcanology), which simply considers the range of possible physico-chemical states that an aliquot of magma can achieve on changing pressure-temperature conditions. End-member scenarios range from equilibrium crystallization (resulting in complete crystallization) to zero crystallization (quenching to glass), and from equilibrium degassing and zero degassing. Lavas typically undergo disequilibrium crystallization and degassing, resulting in partially crystallized and partially degassed quenched products. These materials are two- or three-phase suspensions (liquid ± crystals ± bubbles), and exhibit strain-rate-dependent (non-Newtonian) rheological behavior.

Experimental investigations of the rheology of crystallizing magma can determine end-member behaviors at constant temperature, in which case strain-rate can be varied, or at constant strain-rate, in which case temperature can be varied. There are of course an infinite number of permutations arising from strain-rate and cooling-rate. Additional complications arise from the effects of different crystal- and bubble-size distributions (polydispersity), the effects of strain localization, and the possibility of separation of the three phases during ascent, especially by outgassing of volatiles through connected permeable networks. Feedbacks between rheology, ascent rate / flow velocity, thermal history and phase assemblages are complex, but a systematic experimental approach to studying these relations can provide useful constraints on the uncertainties implicit in numerical modeling.