Collaborative Research: Sustained Resources: Cyberinfrastructure for petrological modeling with alphaMELTS and Python

About This Grant

Models of the physical and chemical behavior of partially molten rocks form a key component of scientists' ability to study and understand volcanic systems. Together with field-based studies, experimental analyses, and investigations of prior eruptions, these models are important for advancing our knowledge of potentially hazardous volcanoes in the United States and globally. This project continues development of a flexible, powerful, and easy-to-use suite of modeling software tools used by thousands of Earth scientists: alphaMELTS. This project will expand the scenarios that alphaMELTS can model, increase integration with igneous rock and experimental databases, link to other geochemical modeling tools written in Python, and support users with workflows for increased reproducibility. Online resources and outreach workshops will extend applications of the software in teaching and training. These workshops, both virtual and in-person, will equip scientists from various career stages and experience levels with quantitative tools for modern Earth science research. This project will implement a framework to align alphaMELTS petrologic modeling software and workflows with FAIR principles (findable, accessible, interoperable, and reusable). Planned developments will make alphaMELTS fully open source, easy to install with standard tools, interoperable with associated modeling and visualization tools and databases, callable from many programming and data visualization environments, and fully versioned, logged, and documented. The work will focus on Python-based tools – in particular, via continued development of the PetThermoTools package for beginner-to-intermediate MELTS users, and machine-learning assisted acceleration for high-end users with high-volume throughputs – but also support those who prefer a simpler graphical user interface. A systematic assessment of model performance will guide users towards applications where the software is verifiably accurate. Expanded functionality will include modeling of reverse crystallization and post-entrapment crystallization of melt inclusions, integration with packages like Thermobar and PySulfSat, and a new trace element engine that gives users total control over partition coefficients. This project will increase integration with IEDA2 databases and the VICTOR portal, and engage with users though workshops, online resources, and creation of instructional materials. The core software development team will grow to include two early-career researchers, as well as two graduate student researchers, which will foster the continuity of alphaMELTS services moving forward. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.