V026-08
Heterogeneity in crystal zonation records variability in the crystal settling dynamics.

Thursday, 10 December 2020: 19:28
Virtual
Cansu Culha1, Tobias Keller2 and Jenny Suckale1, (1)Stanford University, Department of Geophysics, Stanford, CA, United States, (2)Stanford University, Stanford, CA, United States
Abstract:
Chemical zonations of crystals are indicative of the dynamic changes within a magmatic system. A change in magma temperature can alter the thermodynamic equilibrium of the crystal and get recorded as a crystal zonation. Observed crystal populations show a large degree of thermal zonation heterogeneity, suggesting that each crystal has a unique perspective of the magmatic system. System-scale processes like thermal convection and magma mixing have been proposed to explain the crystalline data. Here, we explore whether thermal zonation heterogeneity in the absence of a system-scale flow field can result from a crystalline-scale process. We use crystal settling as the process driving the local-flow field, but our results can be applicable to other local-flow fields like magma injection or bubble driven dynamics.

We model the cooling and crystallization of both hot basaltic and dacitic magmas after being injected into a cold magma reservoir. We zoom into the cooling interface, where crystallization can initiate the flow dynamics. We couple the magma dynamics at the crystalline scale to the thermodynamic processes governing crystal formation and melt property variation. We resolve the physics of individual crystal interactions as they settle. By resolving the individual crystals, we can capture crystal zonations as crystals are subjected to different thermal conditions along their flow path.

Our results show that the dense crystals that form in magma settle as a cluster, which creates meso-scale thermal heterogeneity. The crystals along this warming path drag the cooler melt from the cooling interface and mix into hotter magma. Other crystals flow along a cooling path. Although some crystals experience either heating or cooling, most crystals experience both. Since we force thermodynamic equilibrium by forming or dissolving crystals, we can track the crystals that dissolve due to thermodynamic disequilibrium. The majority of the simulation crystals experience disequilibrium.

By comparing our results to natural samples from Parinacota Volcano and Holyoke Flood Basalt, we conclude that the thermal zonations and their heterogeneity may be a result of local scale heterogeneity overprinting broader system-scale trends.