The Potential of Copernicus Satellites for Disaster Response: Retrieving Building Damage from Sentinel-1 and Sentinel-2

TL;DR

This study investigates using Copernicus Sentinel-1/2 data for rapid building-damage assessment by extending the xBD dataset with xBD-S12, a 10,315 image-pair collection aligned to VHR references. It demonstrates that damage can be detected at a 10 m GSD, with large-area disturbances like wildfires and floods yielding robust performance, while localized damage remains challenging. The authors compare multiple architectures and feature extractors, finding that simpler two-step localization+damage strategies generalize better to unseen disasters than highly engineered models, and that geospatial foundation models offer limited practical benefits for this task. The work provides a publicly available dataset, code, and trained models, highlighting Copernicus data as a viable, rapid, wide-area damage assessment source that complements VHR imagery in disaster response workflows.

Abstract

Natural disasters demand rapid damage assessment to guide humanitarian response. Here, we investigate whether medium-resolution Earth observation images from the Copernicus program can support building damage assessment, complementing very-high resolution imagery with often limited availability. We introduce xBD-S12, a dataset of 10,315 pre- and post-disaster image pairs from both Sentinel-1 and Sentinel-2, spatially and temporally aligned with the established xBD benchmark. In a series of experiments, we demonstrate that building damage can be detected and mapped rather well in many disaster scenarios, despite the moderate 10$\,$m ground sampling distance. We also find that, for damage mapping at that resolution, architectural sophistication does not seem to bring much advantage: more complex model architectures tend to struggle with generalization to unseen disasters, and geospatial foundation models bring little practical benefit. Our results suggest that Copernicus images are a viable data source for rapid, wide-area damage assessment and could play an important role alongside VHR imagery. We release the xBD-S12 dataset, code, and trained models to support further research.

Figures (13)

• Figure 1: Copernicus satellites are surprisingly effective at mapping building damage, despite the moderate GSD of 10$\,$m. Left: Sentinel-2 TCI product. Center: zero-shot prediction from Sentinel-1 and Sentinel-2. Right: reference map, adapted from palisadeFireMaps.
• Figure 2: xBD-S12 comprises 10,315 image pairs across 16 disaster events. Colours denote the event-based split.
• Figure 3: Geographical distribution of xBD-S12, grouped by disaster type as introduced by DisasterAdaptiveNet25.
• Figure 4: Example patches from xBD-S12. For visualisation purposes, we display the True Color Image product for Sentinel-2 and the VV-polarised (log-)amplitude for Sentinel-1. All tiles are to 128$\times$128$\,$px ($\approx$4$\,$m GSD). On the right, VHR images (1024$\times$1024$\,$px, $\approx$0.5$\,$m GSD) and labels from the original high-resolution xBD dataset are shown for reference. See \ref{['app:additional_visu']} for more.
• Figure 5: Effect of buffer size on building localization. Left: Predicted footprints with different training buffer sizes. Right: F1loc score as a function of buffer size.