Lava Flow Detection Using Dual Polarimetric Sentinel-1 Sar Data

In this study, a dual-polarization (DP) processing chain is presented to identify lava flow after a volcanic eruption. The proposed methodology is applied to two very challenging case studies, related to the volcanic eruptions that occurred on 24 December 2018 on Mount Etna, Italy, and on 22 May 2021 on Mount Nyiragongo, Democratic Republic of the Congo. The different geomorphology of the two areas together with the different types of lava (i.e., different rheology and rock composition) and surrounding environment is an unprecedented opportunity to analyze the radar polarimetric contribution. First, a DP change detection scheme, which uses a pair of SAR images collected before and after the eruption, is used. Then, the lava flow is detected using a constant false alarm rate detector. Finally, to quantify the performance of the detector, the SAR-based lava flow maps are contrasted with the lava flow footprint provided by Copernicus Emergency Management Service (EMS). Experimental results show that the use of polarimetric information improves lava flow detection in different kinds of environments.