Answers are grounded in this paper and cite their sources.
### OverviewThis dataset provides multidecadal (1985–2025), wall-to-wall annual burned-area maps (binary and probabilistic) for the entire Iberian Peninsula (mainland Spain and Portugal) at 20 m (Sentinel-2) and 30 m (Landsat) resolution. Generated using the open-source OtsuFire framework, it offers physically auditable supervision without relying on external training labels or prior fire locations. The dataset is based on the paper:Quintero, N., Viedma, O., & Moreno, J. M. (2026). Learning what is not burned: OtsuFire, a self-labelling framework for wall-to-wall burned-area mapping that transfers across sensors and decades. ### Dataset Contents- Annual binary burned-area maps (1985–2025)- Annual probabilistic burned-area maps (1985–2025)- Spatial resolution: 20 m (Sentinel-2 era, 2018–2025) and 30 m (Landsat era, 1985–2017)- Geographic coverage: Iberian Peninsula (mainland Spain and Portugal) ### Code & Companion Resources- R Package: https://github.com/olgaviedma/OtsuFire- Workflow Book & GEE Scripts: https://github.com/nataliaquintero/OtsuFire-book
No takes yet. Share an insight, caveat, or question.
Quintero et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: