Uncertainty-aware deep learning for wildfire danger forecasting
- 1National Observatory of Athens and National Technical University of Athens
- 2Universitat de València
- 3Athena Research Center
Abstract
Wildfires are among the most severe natural hazards, posing a significant threat to both humans and natural ecosystems. The growing risk of wildfires increases the demand for forecasting models that are not only accurate but also reliable. Deep Learning (DL) has shown promise in predicting wildfire danger; however, its adoption is hindered by concerns over the reliability of its predictions, some of which stem from the lack of uncertainty quantification. To address this challenge, we present an uncertainty-aware DL framework that jointly captures epistemic (model) and aleatoric (data) uncertainty to enhance short-term wildfire danger forecasting. In the next-day forecasting, our best-performing model improves the F1 Score by 2.3% and reduces the Expected Calibration Error by 2.1% compared to a deterministic baseline, enhancing both predictive skill and calibration. Our experiments confirm the reliability of the uncertainty estimates and illustrate their practical utility for decision support, including the identification of uncertainty thresholds for rejecting low-confidence predictions and the generation of well-calibrated wildfire danger maps with accompanying uncertainty layers. Extending the forecast horizon up to ten days, we observe that aleatoric uncertainty increases with time, showing greater variability in environmental conditions, while epistemic uncertainty remains stable. Finally, we show that although the two uncertainty types may be redundant in low-uncertainty cases, they provide complementary insights under more challenging conditions, underscoring the value of their joint modeling for robust wildfire danger prediction. In summary, our approach significantly improves the accuracy and reliability of wildfire danger forecasting, advancing the development of trustworthy wildfire DL systems.
Related
- An intelligent weighted adaptive regression process for reliable and uncertainty aware drought forecasting
- Uncertainty-Aware End-to-End AI Weather Forecasting: Disentangling Observation and Model Contributions
- Advancing Near-Term Water Quality Forecasting with Explainable Machine Learning and Probabilistic Uncertainty Quantification
- KNOU-Flood: Advancing global streamflow forecasting with lead-aware probabilistic machine learning
- A machine learning model for early season wildfire risk in Ghana’s Guinea savannah assessing discrimination, operational prevalence, spatial transferability, and CMIP6 climate scenarios
BibTeX
@article{kondylatos2026uncertaintyaware,
title = {{Uncertainty-aware deep learning for wildfire danger forecasting}},
author = {Spyros Kondylatos and Nikolas Papadopoulos and Gustau Camps‐Valls and Ioannis Papoutsis},
journal = {Machine Learning Earth},
year = {2026},
month = jul,
doi = {10.1088/3049-4753/ae89b9},
url = {https://doi.org/10.1088/3049-4753/ae89b9},
note = {CC BY},
}
RIS
TY - JOUR
TI - Uncertainty-aware deep learning for wildfire danger forecasting
AU - Spyros Kondylatos
AU - Nikolas Papadopoulos
AU - Gustau Camps‐Valls
AU - Ioannis Papoutsis
PY - 2026
DA - 2026/07/14
JO - Machine Learning Earth
DO - 10.1088/3049-4753/ae89b9
UR - https://doi.org/10.1088/3049-4753/ae89b9
AB - Wildfires are among the most severe natural hazards, posing a significant threat to both humans and natural ecosystems. The growing risk of wildfires increases the demand for forecasting models that are not only accurate but also reliable. Deep Learning (DL) has shown promise in predicting wildfire danger; however, its adoption is hindered by concerns over the reliability of its predictions, some of which stem from the lack of uncertainty quantification. To address this challenge, we present an uncertainty-aware DL framework that jointly captures epistemic (model) and aleatoric (data) uncertainty to enhance short-term wildfire danger forecasting. In the next-day forecasting, our best-performing model improves the F1 Score by 2.3% and reduces the Expected Calibration Error by 2.1% compared to a deterministic baseline, enhancing both predictive skill and calibration. Our experiments confirm the reliability of the uncertainty estimates and illustrate their practical utility for decision support, including the identification of uncertainty thresholds for rejecting low-confidence predictions and the generation of well-calibrated wildfire danger maps with accompanying uncertainty layers. Extending the forecast horizon up to ten days, we observe that aleatoric uncertainty increases with time, showing greater variability in environmental conditions, while epistemic uncertainty remains stable. Finally, we show that although the two uncertainty types may be redundant in low-uncertainty cases, they provide complementary insights under more challenging conditions, underscoring the value of their joint modeling for robust wildfire danger prediction. In summary, our approach significantly improves the accuracy and reliability of wildfire danger forecasting, advancing the development of trustworthy wildfire DL systems.
SN - 3049-4753
ER -