LA-ICP-MS trace elements analyses of groundmass glasses for petrological monitoring: a data-driven study of magmatic processes in the Dec 2020–Feb 2022 lava fountains series at Mt. Etna

Near real-time petrological monitoring represents a major advancement for volcanology, offering potential insights into magmatic plumbing systems, unravelling mechanisms driving volcanic eruptions, with strong implications for eruption forecasting and volcanic hazard assessment, aiding decision-making during volcanic crises.

Thanks to recent technological and procedural advances, petrological monitoring using trace elements geochemistry of groundmass glasses by LA-ICP-MS is highly promising, as it can be completed in a relatively short time without requiring additional sample preparation compared to SEM analysis. Here we present LA-ICP-MS trace elements analyses of groundmass glasses of lapilli samples from the 2020–2022 paroxysmal sequence produced by the South-East Crater (SEC) at Mt. Etna volcano (Sicily). Mt. Etna is an open-vent basaltic volcano characterized by periods of explosive activity of variable intensity, which, in some cases, lasts months. Between December 2020 and February 2022, about sixty paroxysms (lava fountains) occurred at SEC, with frequency varying throughout the period. Paroxysms were divided into two eruptive sequences^1,2: 13 December 2020 to 1 April 2021 (SEQ1) and 19 May to 23 October 2021 (SEQ2), separated by 48 days of stasis; two paroxysms also occurred on 10 and 21 February 2022. The strong impact of erupted pyroclastic material on aviation, traffic, agriculture, and human life led us to investigate whether it is possible to quickly identify waxing and waning phases of activity.

Previous studies based on major elements, mineral compositions, and diffusion timescales have linked magma dynamics and eruptive activity, and, thanks to sampling of most lava fountains, a near real-time petrological monitoring was addressed mainly through major elements analyses of groundmass glasses and bulk-rock^1,2. Trace elements analyses of groundmass glass are more sensitive than major elements to small and subtle changes in melt composition due to processes such as recharge, mixing, and crystal fractionation, changes that may be partly hidden in major oxides data of high-porphyritic bulk-rock. Thus, trace elements analyses of groundmass glasses can potentially fasten and improve petrological insights detecting magma evolution trends during eruptions.

We analyzed changes in the chemical composition of erupted magma throughout the entire sequence using a data-driven approach on glass analyses, employing hierarchical clustering (HC) to identify compositional groups based on trace element chemistry. This method allowed the fast and accurate recognition of different types of magmas involved in each examined lava fountain, the detection of mafic recharges, as well as the involvement of magma stored in different portions of the plumbing system, as stated in previous works^1,2.

This information is also post-validated combining HC with textural analysis of investigated samples, providing a more complete knowledge of eruptive dynamics. From a petrological monitoring perspective, the acquisition of a large dataset on volcanic glass, combined with unsupervised learning, allows tracking eruptive episodes by recognizing different magmas using objective compositional criteria. At well-monitored volcanoes, this approach can aid in tracking the progression of eruptive activity, its climax, and/or its decline, offering valuable insights to complement data and results from geophysical and geochemical monitoring networks.

¹ Corsaro, R.A., Miraglia, L. (2022); https://doi.org/10.3389/feart.2022.828026;

² Corsaro R.A., Miraglia L., Arienzo I., Di Renzo V. (2024); https://doi.org/10.1007/s00445-024-01770-4;