Abstract book

Pre-Conference Workshop

Web design

Practical Thermodynamic Modelling for Petrology

The Pre-Conference Workshop “Practical Thermodynamic Modelling for Petrology” concluded successfully at the Institute of Geology, Mongolian Academy of Sciences, delivering an intensive day of cutting-edge computational training ahead of the main GICA International Geological Conference. Led by Associate Professor Ondrej Lexa from the Institute of Petrology and Structural Geology at Charles University (Prague, Czechia), the masterclass gathered advanced graduate students, postdoctoral researchers, and industry professionals specializing in metamorphic, igneous, and economic geology.
Assoc. Prof. Ondrej Lexa (Institute of Petrology and Structural Geology, Charles University
15 September 2026
Coding the Crust:
 
Reconstructing Central Asian Tectonics
 
The workshop focused directly on the complex geological architecture of the Central Asian Orogenic Belt (CAOB). Rather than relying on traditional, manual estimation methods, participants spent the 8-hour session using open-source programming stacks to quantitatively model magmatic and metamorphic systems.
The curriculum moved seamlessly from raw lab measurements to high-performance parallel thermodynamic calculations across three core modules:
  • Data Ingestion & Cleansing: Attendees used the Python library petropandas to process raw EPMA spot analyses and whole-rock XRF matrices. The session covered automated structural formula calculations (APFU) and isolated local thermodynamic equilibrium domains within altered mineral systems.
  • Automated Phase Diagrams: Using pypsbuilder, participants bypassed tedious manual command-line interfaces to drive a THERMOCALC backend autonomously. The group successfully mapped out univariant lines and generated automated P-T pseudosections to decode ancient subduction and exhumation pathways.
  • High-Performance Minimization: The final module leveraged MAGEMin and MAGEMinApp, exposing researchers to multi-core parallel Gibbs free energy minimization written in C and Julia. This allowed participants to model liquid lines of descent, extract seismic velocity profiles (\(V_p/V_s\)), and calculate density variations during lower-crustal melting events.

A Modern Workflow for Today’s Geologists

By utilizing a modern stack composed of Python, Julia, and interactive Jupyter Notebooks, the course provided attendees with a reproducible workflow ready to be deployed directly into their active research projects. The Institute of Geology and the GICA organizing committee extend their sincere thanks to Associate Professor Lexa for sharing his expertise and empowering the regional community with these state-of-the-art computational tools.