Dimarco et al. (2025) FireRisk-Zone-LR: A Logistic Regression-Based Wildfire Hazard Zoning Framework for Mediterranean Forests in Tangier, Morocco

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Springer Link (Chiba Institute of Technology)
• Year: 2025
• Date: 2025-12-12
• Authors: Nicola Aimane Dimarco, Ibtissam Faraji, Miriam Wahbi, Mustapha Maatouk, Hakim Boulaassal, Otman Yazidi Aalaoui, Omar El Kharki
• DOI: 10.1051/e3sconf/202567602006/pdf

Research Groups

[Information not provided in the paper text.]

Short Summary

This study developed an open-access, reproducible geospatial workflow integrating satellite-derived indicators with logistic regression to assess wildfire susceptibility in northern Morocco, finding that Normalized Difference Vegetation Index (NDVI) is a significant predictor with the model achieving an AUC of 0.72.

Objective

• To develop an open-access, reproducible geospatial workflow that integrates satellite-derived indicators with logistic regression to assess wildfire susceptibility in Mediterranean forests, specifically in northern Morocco.

Study Configuration

• Spatial Scale: Local to regional scale, focusing on Mediterranean forests in northern Morocco, particularly around Tangier.
• Temporal Scale: Contemporary assessment of wildfire susceptibility based on recent satellite data, without a specific study period mentioned for the susceptibility mapping itself.

Methodology and Data

• Models used: Logistic regression.
• Data sources:
  • Satellite: Sentinel-2 (vegetation indices), SRTM (topography), MODIS (land-surface temperature anomalies).
  • Observation: Burned-area masks (used to generate a dataset of 734 labeled pixels).

Main Results

• The logistic regression model achieved an Area Under the Curve (AUC) of 0.72, indicating reasonable discriminatory power for wildfire susceptibility.
• Normalized Difference Vegetation Index (NDVI) was identified as a significant predictor of fire probability.
• Slope was found not to be statistically significant as a predictor of fire probability.
• The developed workflow demonstrates a scalable and transparent approach to fire-risk assessment using freely available satellite data and cloud-based tools.

Contributions

• Provides an open-access and reproducible geospatial workflow for wildfire susceptibility assessment, addressing limitations of existing expert-weighted or coarse meteorological index-based risk maps.
• Integrates multiple freely available satellite-derived indicators (vegetation indices, topography, land-surface temperature anomalies) with logistic regression for a robust assessment.
• Offers a scalable and transparent methodology applicable to Mediterranean forest systems beyond the study area in Morocco.

Funding

[Information not provided in the paper text.]

Citation

@article{Dimarco2025FireRiskZoneLR,
  author = {Dimarco, Nicola Aimane and Faraji, Ibtissam and Wahbi, Miriam and Maatouk, Mustapha and Boulaassal, Hakim and Aalaoui, Otman Yazidi and Kharki, Omar El},
  title = {FireRisk-Zone-LR: A Logistic Regression-Based Wildfire Hazard Zoning Framework for Mediterranean Forests in Tangier, Morocco},
  journal = {Springer Link (Chiba Institute of Technology)},
  year = {2025},
  doi = {10.1051/e3sconf/202567602006/pdf},
  url = {https://doi.org/10.1051/e3sconf/202567602006/pdf}
}