Abstract

Chair: Martin Rogers Abstract: There are ~1500 subaerial volcanoes with the potential to erupt, but less than 10% are instrumentally monitored. Routine acquisition by the European Sentinel-1 radar mission now offers the potential to monitor most of them with at least two acquisitions every twelve days. We have developed a system to routinely apply radar interferometry (InSAR) whenever a new image is acquired by Sentinel-1 over a volcano. Displacement of the ground between images shows up in the resulting “interferograms”, but there are too many of these to inspect individually. We have therefore developed a machine learning approach to identify signs of new deformation, and changes in the rate of existing deformation patterns, using independent component analysis. We first use a set of training interferograms to identify components associated with background deformation and common atmospheric patterns. We then analyse new interferograms in the context of those components, and flag changes in rate above the background variation, and significant unexplained new signals. We demonstrate this approach on a range of volcanoes, including Sierra Negra in the Galapagos, where the algorithm detects the acceleration in uplift some two years prior to eruption in 2018, as well as the eruption itself. More recently we have developed a more general alternative approach, using a convolutional neural network, which can locate deformation in interferograms and distinguish it from atmospheric signals.