Lava flow monitoring from EO microwave imageries

In this study, microwave satellite images, and in particular Synthetic Aperture Radar (SAR), are used to detect volcanic lava flow. We propose methods that exploit both single and dual polarization channels to map the deposits that occurred in three case studies: the Etna Volcano (Italy) 2007-2008 eruption and the Sangay Volcano (Ecuador) eruption that took place in December 2021.

The objective of the work is to analyse the different information carried out by single and dual polarization data, in different volcanic environments, aiming to identify the most suitable solution for each setting.

For the Etna case study, we applied two change detection methods. The first one exploits the normalised difference and is used on a pair single polarization (SP) image captured by ENVISAT mission. The second method relies on the dual polarization (DP) data acquired by ALOS-PALSAR satellite. The latter is based on the covariance matrix derived from DP images, and the normalized difference of pre and post-event matrices.

As far as the Sangay case study, the results are undertaken over a pair of images collected by the C-band Sentinel-1 DP SAR sensor. The outputs are then analyzed to detect the different phenomena (i.e., lava flow, pyroclastic currents, landslides), that occurred over the scene.

The work demonstrates the complementary evidence provided by the co- and cross-polarized channels, suggesting the combination of them to obtain additional and more accurate information.

This activity is part of a INGV funded project, SAFARI - an AI-based StrAtegy For volcano hAzaRd monItoring from space, a research project that aims at developing a comprehensive space-based strategy for the near-real-time characterization of volcanic state of activity, based on the extraction of satellite-derived input parameters to physical models for rapid scenario forecasting during the eruptive phases.