WildfireGenome: Interpretable Machine Learning Reveals Local Drivers of Wildfire Risk and Their Cross-County Variation
This work addresses wildfire risk assessment for local decision-makers by providing interpretable, decision-scale analytics, though it is incremental as it builds on existing methods like Random Forest and SHAP.
The paper tackled the problem of opaque wildfire risk assessments by developing WildfireGenome, an interpretable machine learning framework that fuses federal indicators and uses SHAP analyses to reveal local drivers like needleleaf forest cover, achieving accuracies of 0.755-0.878 and Kappa up to 0.951 across seven U.S. counties.
Current wildfire risk assessments rely on coarse hazard maps and opaque machine learning models that optimize regional accuracy while sacrificing interpretability at the decision scale. WildfireGenome addresses these gaps through three components: (1) fusion of seven federal wildfire indicators into a sign-aligned, PCA-based composite risk label at H3 Level-8 resolution; (2) Random Forest classification of local wildfire risk; and (3) SHAP and ICE/PDP analyses to expose county-specific nonlinear driver relationships. Across seven ecologically diverse U.S. counties, models achieve accuracies of 0.755-0.878 and Quadratic Weighted Kappa up to 0.951, with principal components explaining 87-94% of indicator variance. Transfer tests show reliable performance between ecologically similar regions but collapse across dissimilar contexts. Explanations consistently highlight needleleaf forest cover and elevation as dominant drivers, with risk rising sharply at 30-40% needleleaf coverage. WildfireGenome advances wildfire risk assessment from regional prediction to interpretable, decision-scale analytics that guide vegetation management, zoning, and infrastructure planning.