Integrated machine learning approach for volcanic cloud tracking - A Case Study of Etna’s Lava Fountains (2020-2022)

Aug 4, 2026

Upload by Aurelia

Between December 2020 and February 2022, Mount Etna generated extraordinary lava fountains that formed eruptive columns and volcanic clouds rising several kilometers above the vent. Monitoring these clouds is essential to evaluate environmental, health, and aviation impacts, yet rapid response is challenging because geostationary thermal infrared satellites deliver high-frequency data every 5 minutes. This study develops an integrated machine learning workflow using EUMETSAT MSG SEVIRI to detect and track clouds, segmenting them via deep learning and classifying key components (ash-rich, SO2-rich, mixed) with supervised pixel-level models, then retrieving plume height for characterization.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages14

Reading Time35 min

AccessFree to download

Introduction

Volcanic cloud hazards and impacts

Remote sensing with geostationary infrared satellites

Machine learning for automatic cloud detection

ANNALS OF GEOPHYSICS, 68, 2, V224, 2025; doi:10.4401/ag-9189 OPEN ACCESS Integrated machine learning approach for volcanic cloud tracking: A Case Study of Etna’s Lava Fountains (2020‑2022) Federica Torrisi *,1,2 (1) Istituto Nazionale di Geofisica e Vulcanologia, Osservatorio Etneo, Catania, Italy (2) University of Catania, Department of Electrical, Electronic and Computer Science Engineering, Catania, Italy Article history: received October 20, 2024; accepted March 14, 2025 Abstract Between December 2020 and February 2022, Mt. Etna produced extraordinary lava fountains which developed into eruptive columns rising several kilometers above thevent. It is crucial to monitor the volcanic clouds produced during these eruptions to assess their impact on the environment, human health, and aviation. Geostationary satellite missions provide high‑frequency thermal infrared data, which are crucial for monitoring volcanic clouds during intense explosive eruptions. However, the large volume of satellite data necessitates automatic and accurate processing algorithms, especially when dealing with global‑scale observations every 5 minutes. In this work, a robust machine learning approach is developed to identify and track volcanic clouds using images from the EUMETSAT MSG SEVIRI (Meteosat Second Generation – Spinning Enhanced Visible and InfraRed Imager) . This approach combines two distinct machine learning models: a deep learning (DL) model for volcanic cloud detection and a supervised machine learning (ML) model for identifying its primary components. The DL model segments volcanic cloudsin SEVIRI images by analyzing both the spatial and spectral intensity data. The supervised ML model is able to distinguish the main components of a volcanic cloud by classifying the pixels as ash‑rich, SO2‑rich, or characterized by mixed components. Once an accurate mask of the volcanic cloud is obtained, the volcanic plume height is retrieved from satellite observations for further characterization. This integrated ML approach was applied to characterize the volcanic clouds produced during some of the lava fountains occurred at Etna volcano (Italy) between 2020 and 2022. Keywords: Volcanic clouds; Geostationary satellite sensors; Machine learning; Convolutional Neural Networks; Etna 2020‑2022 lava fountains 1. Introduction Etna is one of the most active volcanoes in the world, characterized by a frequent explosive activity (Branca and Carlo, 2005; Freret‑Lorgeril et al., 2018; Calvari and Nunnari, 2024; Corsaro et al., 2024). At Etna volcano, between December 2020 and February 2022, a total of 66 lava fountains occurred (Calvari and Nunnari, 2022), producing Federica Torrisi lava flows and large volcanic clouds that spread over the surrounding areas. Most of these clouds caused disruptions, frequently leading to the closure of Catania airport. Volcanic clouds are produced during explosive volcanic eruptions, which release significant amount of silicate particles and gases, primarily composed ofwater vapor (H 2O), carbon dioxide (CO 2) and sulphur dioxide (SO 2) (Scollo et al., 2012) . These emissions can affect human health and ecosystems (Mather et al., 2003; Durant et al., 2010), and are one of the most important natural sources of pollutants in the atmosphere (Filonchyk et al., 2022). Silicate particles, in particular, can damage aircrafts (Guffanti et al., 2009), crops and infrastructure (Wilson et al., 2014) and also cause respiratory diseases (Gudmundsson, 2011). SO2, on the other hand, can lead to acid rain and, if it reaches the stratosphere, can alter the aerosol optical depth, potentially causing climatic disturbances that may last for several years following major volcanic eruptions (Pitari and Mancini, 2002) . Large volcanic eruptions are not only known as significant hazards to aviation and health but also playa crucial role in the climate’s natural variability (Ge et al., 2016; Marshall et al., 2025) . Volcanic aeros

FAQs

Why is volcanic cloud monitoring important for Mt. Etna eruptions?

Volcanic clouds affect the environment, human health, and aviation, and often disrupt operations such as airport activities. Accurate monitoring is therefore essential for impact assessment and risk management.

What satellite data and sensing approach does the study use?

The workflow uses EUMETSAT MSG SEVIRI geostationary observations, leveraging high-frequency thermal infrared measurements. The method focuses on infrared information where SO2 absorption and ash behavior differ.

How does the integrated machine learning approach identify and characterize volcanic clouds?

A deep learning model segments volcanic clouds in SEVIRI images using spatial and spectral intensity information. A supervised machine learning model then classifies cloud pixels into ash-rich, SO2-rich, or mixed components, enabling plume height retrieval for further characterization.

Aug 4, 2026

Upload by Aurelia