Automatic Estimation of Daily Volcanic Sulfur Dioxide Gas Flux From TROPOMI Satellite Observations: Application to Etna and Piton de la Fournaise

Abstract

AbstractUnderstanding the dynamics of sulfur dioxide (SO2) degassing is of primary importance for tracking temporal variations in volcanic activity. Here we introduce the novel “disk method,” which aims at estimating the daily volcanic SO2 mass flux from satellite images (such as those provided by Sentinel‐5P/TROPOspheric Monitoring Instrument [TROPOMI]). The method calculates a “proto‐flux” using a regression, as a function of distance, of SO2 mass integrated in a series of nested circular domains centered on a volcano. After regression, a single multiplication by plume speed suffices to deduce the SO2 mass flux, without requiring a subsequent regression. This way, a range of plume speed and plume altitude scenarios can be easily explored. Noise level in the image is simultaneously evaluated by the regression, which allows for estimating posterior uncertainties on SO2 flux and improving the level of detection for weak sources in noisy environments. A statistical test is also introduced to automatically detect occurrences of volcanic degassing, lowering the risk of false positives. Application to multi‐year time‐series at Etna (2021) and Piton de la Fournaise (2021–2023) demonstrates (a) a reliable quantification of SO2 emissions across a broad range of degassing styles (from passive degassing to effusive or paroxysmal events), and (b) a reasonable day‐to‐day correlation between SO2 flux and seismic energy. The method is distributed as an open‐source software, and is implemented in an interactive web application within the “Volcano Space Observatory Portal,” facilitating near‐real‐time exploitation of the TROPOMI archive for both volcano monitoring and assessment of volcanic atmospheric hazards.