magmasource.caltech.edu

This is the home page for the two day workshop at the Goldschmidt 2020 conference; for the NSF sponsored two to three day workshops held at Caltech, and elsewhere, please see the alphaMELTS NSF workshops page.

We are pleased to announce a two day workshop on the use of MELTS software for calculation of phase equilibria and tracking of trace elements during igneous processes. The MELTS family of algorithms are widely used by petrologists and geochemists to predict the outcomes of melting and crystallization processes, to compare with experiments or to design experimental campaigns, to assess the energy budgets of igneous processes, to teach the principles of igneous petrology and geochemistry, and more. Instructors Paul Asimow and Paula Antoshechkina from Caltech are the designers and maintainers of alphaMELTS 2 and related software, and will cover background to the MELTS models, software installation and use, real world examples, as well as some undocumented tricks that are important to getting reliable results from MELTS calculations.

Although there are a variety of ways to access MELTS models, this workshop will focus on the newly released alphaMELTS 2 interface and access to alphaMELTS from MATLAB, Python, and (if there is sufficient interest) R programming environments. alphaMELTS 2 is a text-based front end to the rhyolite-MELTS, pMELTS, and pHMELTS models, with built-in trace element calculations and a variety of unique features and workflows. This workshop will differ from previous ones in that it will be structured so that users can attend either or both days, depending on their needs and previous experience. Day 1 will focus on introductory material and basic functionality for users new to alphaMELTS. Day 2 will focus on more advanced features, larger-scale calculations, and special problems suggested by attendees. We will also have time for a "MELTS clinic" with one-on-one support or break out groups for users' own MELTS calculations (including any using the graphical user interface, alphaMELTS 1.X etc.). Day 1 will constitute sufficient preparation for a novice to be able to appreciate Day 2. Users who have previously attended an alphaMELTS workshop, especially one of the recent NSF-sponsored GeoInformatics short courses, should also find Day 2 helpful. We will start the second day with a refresher/update of alphaMELTS 2's newest features (e.g. the expanded trace element engine).

Title: Thermodynamic Modeling of Magmatic Processes with alphaMELTS 2
Organizers: Paula Antoshechkina, Paul Asimow (Caltech)
Date: June 20th - 21st, 2020, 9.00AM - 5.00PM each day
Venue: Building and room TBD, University of Hawaii
Registration fee: $125 (two days) / $65 (one day), per particpant; payable by credit/debit card
Workshop registration deadline: May 21st, 2020
Early registration deadline for Goldschmidt conference: April 21st, 2020

The workshop is being hosted at the University of Hawaii, approximately 1.5 miles from the conference location and is open to all participants at the Goldschmidt 2020 meeting. The $125/$65 fee includes refreshments and lunch for each day, together with a $5 materials fee. The deadline for workshop registration coincides with the close of standard online registration for the conference. Details on the logistics of staying in Hawaii, such as travel and accommodation for the workshop, are also provided on the Goldschmidt 2020 conference website.

Registration and payment for the workshop are available together as three options ('Thermodynamic Modeling of Magmatic Systems with alphaMELTS2', 'Thermodynamic Modeling of Magmatic Systems with alphaMELTS2 - Saturday only', or 'Thermodynamic Modeling of Magmatic Systems with alphaMELTS2 - Sunday only') during the conference registration process. Or you can select 'Buy' from the alphaMELTS 2 workshop details on the Goldschmidt workshops page; after entering your contact information you will be able to choose the day(s) of the workshop that you wish to attend.

We intend to cover a range of topics during the workshop, including:

• Background to, and strengths and limitations of the computational phase equilibria approach taken by the MELTS family of algorithms
• Comparison of MELTS models (rhyolite-MELTS 1.0.2, 1.1.0, 1.2.0; pMELTS; pHMELTS) and suitability for particular applications
• Common errors in running the models and important ways to apply them efficiently and accurately
• Introduction to the alphaMELTS 2, and alphaMELTS for MATLAB/Python interfaces
• The trace element engine in alphaMELTS 2, including the expanded D(P,T,X) functionality of alphaMELTS 2.
• traceDs, a database of experimental studies involving trace element distribution between liquid, and solid phases, and how to use traceDs search results in alphaMELTS 2
• Brief discussion of the next generation of models (e.g. high pressure phase equilibria in xMELTS) and planned extensions to the existing interfaces

Talks describing the models and software will be kept brief to allow more time for first hand experience with the software. Particular features will be higlighted on the platforms for which they are most convenient (e.g. alphaMELTS for trace elements; alphaMELTS for MATLAB for open system behavior). Application to magma chamber processes and mantle melting will be discussed, and short demonstrations will be included as illustrations. Examples from the introductory sessions will be expanded on, and some special calculations will highlight features that are only available (or only convenient) on individual platforms.

The schedule will be something like (subject to change):

• 20th June 2020: Introduction to alphaMELTS 2, and alphaMELTS for MATLAB/Python
  • AM session 1: Welcome to the workshop; brief overview of MELTS software programs that we will be using and how to install them
  • 9.30 AM: Talk on model construction for the MELTS family of algorithms; modeling modeling magma chamber processes, including liquid lines of descent (LLD), assimilation and fractional crystallization (AFC)
  • 10.30 AM: Coffee break (exact time TBD; software installation clinic)
  • 11.00 AM: Practical session, getting started with alphaMELTS 2; how to generate input files, and plot results, modeling fractional crystallization, LLD etc.
  • 12.30 PM Lunch break
  • PM sesion 1: Continued from previous session, modeling magma chamber processes with alphaMELTS 2
  • 3.00 PM: Refreshments served (exact time TBD)
  • PM session 2: Introduction to alphaMELTS for MATLAB/Python*: modeling mantle melting, phase diagram mode etc.
  • 5.00 PM: End of afternoon session
• 21st June 2020: Trace elements and advanced features in alphaMELTS, MELTS clinic
  • AM session 1: Refresher/update on new alphaMELTS 2 features, and trace elements in alphaMELTS 2. Very brief introduction to ENKI and the tools hosted there, and discussion of how alphaMELTS and other MAGMA@Caltech tools fit in with the ENKI project (Paula Antoshechkina)
  • 10.30 AM Coffee break (exact time TBD)
  • AM session 2: Advanced calculations: e.g. assimilation and other (user-defined) open-system behavior, polybaric mantle melting of two sources that are thermal but not chemcial equilibrium
  • 12.30 PM Lunch break
  • PM session 1: Work through problems suggested by participants. Possibility to break out into smaller groups with common interests or to work individually with instructors
  • 3.00 PM: Refreshments served (exact time TBD)
  • PM session 2: Continued from previous session, followed by group discussion - concluding remarks, feedback, and suggestions for features that workshop participants would like to see in alphaMELTS 2 soon
  • 5.00 PM: End of afternoon session

*Note that having access to MATLAB is not a requirement for participation. The data visualization part of the exercises will be possible in MATLAB Online, which we hope to have a workshop license for. Python (or possibly R) alternatives for each exercise will be available, if needed; the MATLAB and Python interfaces to alphaMELTS are near-identical.

We will have time for Q&A with individual workshop organizers to address participants' particular interests during the break-out sessions. Please feel free to bring examples that you would like us to look at. This could be a calculation that you have attempted that does not seem to work (not necessarily in alphaMELTS 2; we can offer support for the GUI, alphaMELTS 1.X, MELTS for Excel etc.). Alternatively, simply bring a composition (major elements in wt% or grams of oxides; trace elements, optional for alphaMELTS, in ppm) and conditions that you would see calculated.

We will provide a thumb drive (that you can keep) with all the software and sample input files. Links to example files, from a 3-day workshop, are available by clicking the Links tab above (since previous 1- and 2-day workshops at Goldschmidt conference using alphaMELTS 1.9, we are updating to the newer alphaMELTS 2 format). The files will be updated nearer the time of this workshop. Some examples (such as MORB crystallization) will be provided for more than one front-end, so that you can see similarities and differences.

Links to software front-ends that will be demonstrated at the workshop are available by clicking the Links tab above. If you decide to install the software in advance, you can test the installation using the appropriate 'Quick Start using MELTS' tutorial example; otherwise there will be plenty of opportunity during the workshop to install the software of your choice, with help from the organizers. Either way, it is a good idea to have VirtualBox installed so that you can get going with the first day's practical sessions using VM alphaMELTS if there are any software installation issues.

Paula Antoshechkina (née Smith) is an Associate Research Scientist at Caltech. She is responsible for most of the alphaMELTS front-end, forum and MAGMA website. At the moment she is focused on pMELTS+CO₂ and other calibrations, libalphaMELTS and libpHMELTS interfaces, and various projects that fit between the tools available on the ENKI portal and traditional MELTS software.

Paul Asimow is Professor of Geology and Geochemistry at Caltech. He developed pHMELTS and many of the tools since incorporated into alphaMELTS. He has supervised a number of students and postdocs with MELTS-related projects, both to extend and improve the underlying thermodynamic models and to apply the software to real world magmatic processes. He is also involved in shock-wave experiments and theoretical calculations on silicate liquid and minerals.