Skip to main content
Dryad

Explaining building damage from wildfires in California

Bar, Somnath1; Li, Shu1; Banerjee, Tirtha1

Published Jun 23, 2026 on Dryad. https://doi.org/10.5061/dryad.r2280gbrw

Data files

Jun 23, 2026 version files 387.34 MB

Click names to download individual files

Abstract

Wildfires continue to devastate California's wildland-urban interface (WUI), where buildings intersect flammable landscapes. We present a physically interpretable, data-driven framework to elucidate causal interactions, model, and predict wildfire-induced building damage across California. Using over 100,000 damage inspection records (2013-2024), we identify the key environmental, structural, and exposure factors driving wildfire impacts on the built environment. Using Random Forest classifiers, building damage is modeled from static environmental variables (topography, vegetation, human footprint), dynamic weather inputs, and a Composite Building Flammability Rating (CBFR). Three model configurations were tested: (1) a comprehensive model integrating all variables, (2) an enviro-weather hybrid excluding CBFR, and (3) an environmental exposure model excluding both weather and CBFR. Applying a strict 200m spatial dead-zone constraint to eliminate local autocorrelation, the comprehensive model achieved 88% (±0.4%) accuracy, dropping to 82.9% (±0.6%) without CBFR and to 74.5% (±0.5%) without both weather and CBFR. Spatial grid-based cross-validation demonstrated a diverse accuracy of 68.0% (±17%), 66.0% (±16%), and 62.0% (±13%), respectively. Building flammability, dew point temperature, and near-surface wind vector were the most important predictors of damage. Causal analysis revealed spatially heterogeneous Conditional Average Treatment Effects, suggesting mitigation strategies should target specific environmental and anthropogenic gradients. We developed a 100-meter-resolution Wildfire Building Damage Risk Index (WBDRI) to highlight high-risk damage zones. Findings emphasize that wildfire impacts in the WUI result from a confluence of structural vulnerability, atmospheric dryness, and fuel exposure, offering scalable tools for risk forecasting, defensible space planning, and climate-resilient infrastructure development.