The internal structure and evolution of planetary bodies are influenced by endogenous processes that operate deep within the body and become apparent at the surface through various geologic and tectonic features. Although the evolution, composition, and interiors of terrestrial, or earth-like, planets have some features in common, such as iron-rich cores, slowly convecting silicate mantles, and rigid crustal layers – they are entirely different in many other respects. The main focus of the group’s research includes numerical modeling of the internal structure and thermo-chemical evolution of planetary bodies as well as the development and scientific support of in-situ space experiments – in particular measurement of physical properties.
What is the data science project’s research question? Can machine learning-based surrogates replace computationally intensive thermodynamic models to predict the magma production of terrestrial planets?
What data will be worked on? An extensive database of magma compositions, temperature and physical properties (e.g. density) created with thermodynamic models such as pMELTS and Perple_X. A starting database will be assembled prior to the start of the project and could be expanded based on needs.
What tasks will this project involve? Through data analysis and discussions with researchers of various backgrounds (physicists, geoscientists and computer scientists), identify the best predictive model of magma composition, melt fraction and density based on temperature-pressure conditions and the chemical/mineralogical compositions of planetary interiors.
Design a practical model that can be implemented and tested with a 3D finite-volume mantle convection mode.
What makes this project interesting to work on? The candidate will contribute to expand the knowledge of how the terrestrial planets evolved through time, and why they are covered by different rock types. The formation of planetary crusts is also tied to the development of atmospheres, which in some cases led to surface conditions suitable for life. Understanding the chronology of those events and identifying the key controlling parameters is at the center of the field of “planetary thermo-chemical evolution”. The candidate will be an active member of a team consisting of physicists, chemists, geologists and computer scientists and will be involved in unprecedentedly interdisciplinary research.
What is the expected outcome? Contribution to research paper, Contribution to software development
What infrastructure, programs and tools will be used? Can they be used remotely? The project can be carried out solely using open source tools such as Python and Scikit-learn/Keras/Pytorch
What skills are necessary for this project? Data analytics / statistics, Computational models, Deep learning, Visualization
Is the data open source? Yes
Interested candidates should be at Master level. Doris Breuer is looking for 1 visiting scientist, working on the project with Sabrina Schwinger (email@example.com) as supervisor with the team.