Explaining building damage from wildfires in California

Area of Science:

• Environmental Science
• Climate Science
• Risk Assessment

Background:

• Wildfires pose a significant threat to buildings, particularly in the wildland-urban interface.
• Accurate modeling of wildfire-induced building damage is crucial for effective risk management and mitigation strategies.

Purpose of the Study:

• To develop a physically interpretable, data-driven framework for modeling and predicting wildfire-induced building damage.
• To identify key environmental, weather, and structural factors contributing to building damage.
• To create a high-resolution Wildfire Building Damage Risk Index for California.

Main Methods:

• Utilized over 100,000 damage inspection records (2013-2024) and static environmental variables (topography, vegetation, human footprint).
• Incorporated dynamic weather inputs and a Composite Building Flammability Rating (CBFR).
• Tested three model configurations with a 200-meter spatial dead-zone constraint and employed spatial grid-based cross-validation.

Main Results:

• The comprehensive model achieved 88% accuracy, significantly outperforming models excluding CBFR (82.9%) or both weather and CBFR (74.5%).
• Building flammability, dew point temperature, and near-surface wind speed were identified as key damage predictors.
• Vapor pressure deficit showed the strongest causal effect on damage probability, with spatial variations in climate and geographic variable effects.

Conclusions:

• Wildfire impacts result from a combination of structural vulnerability, atmospheric dryness, and fuel exposure.
• The developed framework and risk index offer scalable tools for forecasting wildfire risk.
• Findings support improved defensible space planning and the development of climate-resilient infrastructure.