Maffei et al. (2025) Probabilistic approaches for the prediction of forest fire danger using optical and thermal satellite data

Identification

• Journal: Elsevier eBooks
• Year: 2025
• Date: 2025-12-06
• Authors: Carmine Maffei, Roderik Lindenbergh, Massimo Menenti
• DOI: 10.1016/b978-0-443-40296-8.00010-0

Research Groups

• Consorzio MedITech – Mediterranean Competence Centre 4 Innovation, Naples, Italy
• Department of Geoscience & Remote Sensing, Delft University of Technology, Delft, Netherlands
• State Key Laboratory of Remote Sensing Science, Chinese Academy of Sciences, Aerospace Information Research Institute, Lanzhou, China

Short Summary

This study develops and evaluates probabilistic approaches using optical and thermal satellite data to predict forest fire danger, demonstrating performance comparable to or better than the Fire Weather Index for extreme fire events.

Objective

• To develop and evaluate probabilistic approaches for predicting forest fire danger using optical and thermal satellite data, specifically assessing their performance against established indices like the Fire Weather Index.

Study Configuration

• Spatial Scale: Not explicitly defined, but implied regional to global coverage based on satellite data.
• Temporal Scale: Not explicitly defined, but implied continuous monitoring for fire danger prediction.

Methodology and Data

• Models used: Probabilistic approaches (specific model not named), Fire Weather Index (for comparison)
• Data sources: Optical and thermal satellite data; Derived products: Land surface temperature anomaly, Perpendicular Moisture Index

Main Results

• Observations of vegetation condition, specifically land surface temperature anomaly and the perpendicular moisture index, serve as independent covariates for fire burned area, duration, and rate of spread.
• The developed probabilistic approaches achieved comparable or superior performance to the Fire Weather Index in predicting the probability of extreme fire events.
• A physical interpretation of the results is proposed.

Contributions

• Introduction of novel probabilistic approaches for forest fire danger prediction utilizing optical and thermal satellite data.
• Demonstration of the effectiveness of satellite-derived land surface temperature anomaly and perpendicular moisture index as independent indicators of live fuel condition.
• Provision of a method that performs comparably to or better than the established Fire Weather Index for predicting extreme fire events.
• Offering a physical interpretation of the observed relationships, enhancing understanding of fire danger dynamics.

Funding

• Not specified in the provided text.

Citation

@article{Maffei2025Probabilistic,
  author = {Maffei, Carmine and Lindenbergh, Roderik and Menenti, Massimo},
  title = {Probabilistic approaches for the prediction of forest fire danger using optical and thermal satellite data},
  journal = {Elsevier eBooks},
  year = {2025},
  doi = {10.1016/b978-0-443-40296-8.00010-0},
  url = {https://doi.org/10.1016/b978-0-443-40296-8.00010-0}
}