Automated Mapping of Lava: Classifcation of the Lava Flow Field From the 2021 Fagradalsfjall Eruption

Efficient mapping of lava flow fields during volcanic eruptions is crucial for hazard mitigation. Here we present a machine learning approach to classify the lava flow field of the 2021 Fagradalsfjall eruption on the Reykjanes Peninsula, Iceland using orthomosaics and layers derived from digital elevation models. A supervised classification method achieves a 96.7% accuracy against the manually digitized outline of the lava flow field after post-processing. The addition of an unsupervised clustering step yields a 99.8% accuracy. These results are a stepping stone for automated monitoring of lava flows on the Reykjanes Peninsula, which can improve the efficiency of hazard assessment.