"This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5)."
Volcanoes are places where molten rock erupts on the surface of the Earth; they are both major natural hazards and fascinating targets for scientific study because they are messengers from the interior of our planet, where melting of rocks takes place. Their message, however, is encoded in the chemistry of lavas and in order to read that code we need tools that relate the chemical and physical make-up of the source region to the quantity and composition of magma generated. Computational thermodynamics (CT) has proven to be an essential tool for predicting these connections and has become a necessary component of modern petrologic modeling of melting and magma transport. Previous work by the lead investigators in CT (the MELTS, pMELTS, and phMELTS software packages and a substantial body of work on mineral and melt thermodynamics) is utilized by thousands of researchers and students whose interests range from volcanic eruptions, to the transport of volatiles from the Earth's interior to its atmosphere and oceans, to the formation of ore deposits.
This project funds collaboration between researchers at Caltech and OFM Research to improve and extend thermodynamic models of the minerals garnet and pyroxene, both of which occur in the mantles of terrestrial planets. The motivation for this effort is the desire to complete the xMELTS software package, a CT tool that will permit modeling of partial melting of terrestrial planets to pressures of ~40 GPa. Once completed, this tool will allow scientists to quantitatively assess the chemistry of magma production in the upper 1200 km of the Earth and throughout other terrestrial bodies like the Moon, Mars and Europa. This assessment will help us understand how these planets have evolved chemically since their formation to the present day. The thermodynamic models and software tools developed under this grant will be rapidly disseminated to the research community; they will be incorporated into xMELTS and phMELTS, they will be programmed into web services for distributed computing access from the OFM website, and they will be made available for use in Excel-based applications. There is every expectation that the proposed work will result in thermodynamic models that contribute to a computational infrastructure that is well utilized in the classroom at both the undergraduate and graduate level, and will support the research efforts of a broad community of scientists. In addition, the proposed work will support graduate education and training.
Incorporation of both the garnet and pyroxene models into xMELTS is essential for realistic forward modeling of partial melting in peridotite and pyroxenite bulk compositions. The proposed garnet solid solutions model will apply to compositions in the system (Ca,Mg,Fe2+)3Al2Si3O12-(Mg,Fe2+)4Si4O12 containing minor amounts of Cr3+, Na, Fe3+, and Ti4+, which is inclusive of majorite. The pyroxene model will extend the Sack and Ghiorso model [with general formula (Na,Ca,Mg,Fe2+)(Mg,Fe2+,Ti4+,Fe3+)(Si,Al,Fe3+)2O6] to include the components CaCrAlSiO6 and KAlSi2O6. The garnet model will be based on previously pub-lished low-pressure models of pyralspite garnet thermodynamics and will be ex-tended to higher-pressures principally by evaluation of exchange equilibria be-tween garnet and spinel, garnet-olivine, and garnet-pyroxene. The pyroxene model extension will be calibrated mainly from exchange equilibria involving pyroxene-spinel. In both cases, literature data will be supplemented with new experiments as required.
