AI-Powered Volcanic SO₂ Retrieval using MSG-SEVIRI and Sentinel-5P TROPOMI

Sulfur dioxide (SO₂) represents one of the most important volcanic gases released by magma degassing in the shallow crust. Its monitoring provides information on magma ascent rates, conduit dynamics, and eruption style and intensity, thereby supporting volcano monitoring and hazard assessment.The TROPOMI instrument onboard Sentinel-5 Precursor, launched in 2017, is the most recent sensor which delivers daily measurements of atmospheric SO₂ column densities at an unprecedented spatial resolution of 5.5 km × 3.5 km at nadir.The aim of the present study is to improve the current SO₂ detection capabilities by combining TROPOMI products with data from the MSG‑SEVIRI radiometer, which offers higher spatial resolution (~3 km × 3 km at nadir) and revisit times of 15 minutes, or 5 minutes in Rapid Scan mode.To enhance SO₂ retrieval capabilities, a data-driven AI model was implemented to estimate SO₂ vertical column densities at SEVIRI spatial and temporal resolution, using TROPOMI observations as reference. In particular, a multilayer perceptron was designed and trained, consisting of two hidden layers with 128 and 64 neurons, respectively, followed by a single linear output neuron. The model was trained for up to 200 epochs and optimized by minimizing the Mean Squared Error, with an early-stopping strategy applied to prevent overfitting. Model performance was then evaluated on the test set using the Mean Absolute Error, which measures the average absolute difference between predicted and observed SO₂ Vertical Column Density (VCD) values and provides a reliable indication of the prediction accuracy.This approach allows SEVIRI data to inherit the sensitivity of TROPOMI while preserving their native high-frequency coverage. The method substantially increases measurement density and improves spatial detail, enabling more refined and continuous monitoring of volcanic degassing.The methodology is applied to Mount Etna (Sicily, Italy), an open‑conduit volcano characterized by persistent degassing sustained by shallow convecting magma, with typical SO₂ fluxes ranging from 500 to 5000 t/day. The satellite‑based results are quantitatively validated against measurements from ground‑based monitoring networks.Results show that the AI-enhanced SEVIRI-based SO₂ VCDs differ from the original TROPOMI values by 5–10%, confirming the robustness and reliability of the approach.This integrated technique offers a promising tool for rapid and robust volcanic hazard assessment, introducing improvements to current retrieval methods, and enhancing early warning capabilities for aviation safety, as well as studies of climate impacts from volcanic emissions.