ALBERT

Description: SUMMARY Self-consistent modelling of magmatic systems is challenging as the melt continuously changes its chemical composition upon crystallization, which may affect the mechanical behaviour of the system. Melt extraction and subsequent crystallization create new rocks while depleting the source region. As the chemistry of the source rocks changes locally due to melt extraction, new calculations of the stable phase assemblages are required to track the rock evolution and the accompanied change in density. As a consequence, a large number of isochemical sections of stable phase assemblages are required to study the evolution of magmatic systems in detail. As the state-of-the-art melting diagrams may depend on nine oxides as well as pressure and temperature, this is a 10-D computational problem. Since computing a single isochemical section (as a function of pressure and temperature) may take several hours, computing new sections of stable phase assemblages during an ongoing geodynamic simulation is currently computationally intractable. One strategy to avoid this problem is to pre-compute these stable phase assemblages and to create a comprehensive database as a hyperdimensional phase diagram, which contains all bulk compositions that may emerge during petro-thermomechanical simulations. Establishing such a database would require repeating geodynamic simulations many times while collecting all requested compositions that may occur during a typical simulation and continuously updating the database until no additional compositions are required. Here, we describe an alternative method that is better suited for implementation on large-scale parallel computers. Our method uses the entries of an existing preliminary database to estimate future required chemical compositions. Bulk compositions are determined within boundaries that are defined manually or through principal component analysis in a parameter space consisting of clustered database entries. We have implemented both methods within a massively parallel computational framework while utilizing the Gibbs free energy minimization program Perple_X. Results show that our autonomous approach increases the resolution of the thermodynamic database in compositional regions that are most likely required for geodynamic models of magmatic systems.

Description: The evolution of crustal magmatic systems is incompletely understood, as most studies are limited either by their temporal or spatial resolution. Exposed plutonic rocks represent the final stage of a long-term evolution punctuated by several magmatic events with different chemistry and generated under different mechanical conditions. Although the final state can be easily described, the nature of each magmatic pulse is more difficult to retrieve. This study presents a new method to investigate the compositional evolution of plutonic systems while considering thermal and mechanical processes. A thermomechanical code (MVEP2) extended by a semi-analytical dike/sill formation algorithm, is combined with a thermodynamic modelling approach (Perple_X) to investigate the feedback between petrology and mechanics. Melt is extracted to form dikes while depleting the source region. The evolving rock compositions are tracked on markers using a different phase diagram for each discrete bulk-rock composition. The rock compositional evolution is thus tracked with a high precision by means of a database with more than 58 000 phase diagrams. This database describes how density, melt fraction, chemical composition of melt and solid fractions and mineralogical assemblages change over crustal to uppermost mantle P–T conditions for a large range of rock compositions. Each bulk rock composition is composed of the 10 major oxides (SiO2–TiO2–Al2O3–Cr2O3–MgO–FeO–CaO–Na2O–K2O–H2O) including an oxygen buffer. The combined modelling approach is applied to study the chemical evolution of the crust during arc magmatism and related melt extraction and magma mixing processes. Basaltic sills are periodically injected into the crust to model heat/magma influx from the mantle. We find that accumulated sills turn into long-lived mush chambers when using a lower rock cohesion or assuming a higher intrusion depth. Associated partial melting of crustal host rocks occurs around densely distributed dikes and sills. High silica rocks (e.g. granites) are generated by partial melting of the host rocks, melt segregation within dikes, and from fractional crystallization of basalts. Although the volume of these rocks is relatively small in our models compared to rocks with a mafic to intermediate composition, they provide important information about the processes of magma differentiation within arc continental crust.