Towards high-resolution forecasting of tropical deforestation with deep learning
Environmental Research Letters · 14.11.2026 · 21 (21) · 214045
DOI: 10.1088/1748-9326/aebda8 ↗Kurz gefasst
Fachartikel in der Zeitschrift Environmental Research Letters, online erschienen am 12.11.2026. Verfasst von Jan Pišl und 4 weiteren Autor:innen. Die Arbeit nennt 62 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: European Union.
Abstract (Originalsprache)
Reducing tropical deforestation effectively requires spatially explicit data on deforestation dynamics. While past deforestation is now detected and mapped with high accuracy, producing reliable future forecasts remains a challenge. We present a deep learning approach for high-resolution deforestation forecasting in the Brazilian Amazon (BA). We train a spatio-temporal Convolutional Long Short-Term Memory model on past deforestation maps and other covariates. Our model achieves an F 1 score of 0.47 when forecasting one year ahead, substantially improving over comparable prior work. For multi-year predictions, we obtain promising results within a 3 year horizon, while sensitivity to shifts in deforestation dynamics suggests limitations for longer-term forecasts. We generate forecasts for years 2016–2018 across the BA, and make them available through an interactive Google Earth Engine application. Our results demonstrate the potential for high-resolution, short- and medium-term forecasting that can support conservation efforts and help protect tropical forests.
Förderung
- European Union
Zitieren
APA
Pišl, J., Sumbul, G., Lenczner, G., Wegner, J. D., & Tuia, D. (2026). Towards high-resolution forecasting of tropical deforestation with deep learning. Environmental Research Letters, 21(21), 214045. https://doi.org/10.1088/1748-9326/aebda8
Vancouver
Pišl J, Sumbul G, Lenczner G, Wegner JD, Tuia D. Towards high-resolution forecasting of tropical deforestation with deep learning. Environmental Research Letters. 2026;21(21):214045. doi:10.1088/1748-9326/aebda8
BibTeX
@article{Pil2026,
author = {Pišl, Jan and Sumbul, Gencer and Lenczner, Gaston and Wegner, Jan Dirk and Tuia, Devis},
title = {Towards high-resolution forecasting of tropical deforestation with deep learning},
journal = {Environmental Research Letters},
year = {2026},
volume = {21},
number = {21},
pages = {214045},
doi = {10.1088/1748-9326/aebda8}
}
EUVIMED