Skip to content

Decadal patterns of pre-eruptive pressure and temperature at Mt Etna from machine learning thermobarometry

Martín Miranda-Muruzábal1, Rosa Anna Corsaro2, Luca Caricchi1

  • Affiliations:  1 Department of Earth sciences, University of Geneva, Geneva, Switzerland; 2 Osservatorio Etneo, Istituto Nazionale di Geofisica e Vulcanologia, Catania, Italy 

  • Presentation type: Talk

  • Presentation time: Tuesday 10:30 - 10:45, Room S150

  • Programme No: 3.2.1

  • Theme 3 > Session 2

Abstract

The identification and study of fluctuations and patterns in the eruptive behavior of volcanoes can provide valuable information to predict future activity as well as understanding the dynamics of an ongoing eruption. Mt Etna, one of the world's most active volcanoes, is a natural laboratory exhibiting frequent volcanic activity with different eruptive styles. This constitutes an ideal framework to investigate eruptive patterns and gather insights on the link between the architecture and magmatic processes of the plumbing system and eruptive dynamics. Here, we apply machine learning thermobarometry and chemometry to minerals and glasses (from the literature and newly acquired analyses) from historical and recent Mt Etna eruptions, spanning from 1651 to 2024. The resulting pressure (P), temperature (T) and melt chemistry (Xmelt) are compared across different eruptions and correlated with key volcanic parameters such as total erupted volume and duration of the eruptive event. Preliminary results reveal significant differences in pre-eruptive P-T-Xmelt estimates between eruptive events associated with varying styles and mechanisms. For instance, some eruptions, such as the 1974 event, display a pressure range spanning the entire continental crust, whereas others, like the 1669 eruption, exhibit a more restricted range confined within the crust.