Skip to content

Introducing Orange-Volcanoes a Visual Tool for Petro-Volcanological Data Analysis: A Case Study in Petrological Volcano Monitoring

Alessandro Musu ^1,2^, Valerio Parodi3, Marko Toplak4, Alessandro Carfì3, Mónica Ágreda-López2, Fulvio Mastrogiovanni3, Diego Perugini2, Zupan Blaž4, Maurizio Petrelli2

  • Affiliations: 1Department of Lithospheric Research, University of Vienna, UZA2, Josef-Holaubek-Platz 2, 1090 Vienna, Austria; 2Department of Physics and Geology, University of Perugia, Piazza dell\'Università, 1, Perugia, 06123, Italy; 3Department of Informatics, Bioengineering, Robotics and Systems Engineering, University of Genoa, Via Opera Pia 13, Genova, 16145, Italy; 4Faculty of Computer and Information Science, University of Ljubljana, Večna pot 113, SI-1000 Ljubljana, Slovenia.

  • Presentation type: Poster

  • Presentation time: Monday 16:30 - 18:30, Room Poster Hall

  • Poster Board Number: 155

  • Programme No: 3.1.47

  • Theme 3 > Session 1

Abstract

We present Orange-Volcanoes, an extension (add-on) of the open-source platform Orange Data Mining, specifically designed to advance data-driven analysis in petrology, geochemistry, and volcanology. Orange-Volcanoes expands the core capabilities of Orange by integrating tools for: (i) Compositional Data Analysis (CoDA), (ii) cleaning and preprocessing of geochemical and petrological data, and (iii) mineral and liquid thermobarometry. The combination of these tools enables the rapid application of machine learning, statistical analysis, and predictive modeling to large petro-volcanological datasets, supporting visual and interactive data exploration. The platform's visual programming environment fosters collaborative research, ensuring reproducibility and accessibility for scientists, educators, and students without requiring programming skills. The ability to apply diverse statistical and machine learning tools to geochemical data, while interactively visualizing step-by-step results, makes Orange and Orange-Volcanoes valuable assets for managing large multivariate datasets and supporting petrological volcano monitoring. The ability to apply explainable artificial intelligence techniques, such as feature importance and Shapley additive explanations, allows users to better interpret the underlying drivers of geochemical variability, enhancing understanding of magmatic processes. Through case studies, we demonstrate the application of Orange-Volcanoes in clustering geochemical data and conducting petrological assessments. As the volume of geochemical, petrological, and volcanological data grows, this tool facilitates the future integration of data mining and machine learning techniques into scientists' workflows. Orange-Volcanoes represents a significant step toward a transparent, reproducible, and collaborative scientific approach.