https://youtu.be/LQOYKbhHG7Q

This episode of Geovoices explores the application of compositional data analysis in geochemistry, focusing on two key papers. The first paper, from the field of immunology, demonstrates the universal applicability of compositional data analysis techniques across scientific disciplines. It highlights the importance of treating immune cell data as compositional to avoid flawed statistical analyses and spurious correlations. The paper introduces log-ratio transformations and Dirichlet regression as solutions to handle compositional data effectively, methods that are directly transferable to geochemistry. The second paper introduces GeoCoDA, a structured workflow for analyzing lithogeochemical data. GeoCoDA emphasizes both unsupervised and supervised learning techniques to discover and validate geochemical processes. It leverages logratio transformations, particularly ALR and CLR, to identify key elemental ratios that can discriminate between different geological phases. The workflow successfully applied these techniques to a kimberlite dataset, demonstrating its ability to provide valuable insights for mineral exploration. The discussion also covers the handling of compositional zeros, the use of various multivariate techniques like PCA and correspondence analysis, and the incorporation of element amalgamations to enhance geochemical interpretation P. Geo. Ricardo A Valls, M. Sc. and Geo Gadfly Valls Geoconsultant ORCID ID- https://orcid.org/0000-0002-5421-0914 Scopus Author ID: 7003369619/35335510700 ResearcherID: S-6604-2018 If you like this content, please "buy me a coffee" https://www.buymeacoffee.com/goldendroplets