Wildfire risk models use historical fires, weather models, estimates of fuel load, and terrain to estimate the risk of fire for each plot of land.
This video is a very detailed description of how CalFire creates the fire hazard severity zones
CalFire and the Department of Forestry are responsible for producing maps to show the severity zones for fire hazard, covering the entire state. The maps have been recently remade, including the current weather and climate data, land use and vegetation coverage.
Insurers have similar models, which are kept confidential and are unique to each insurer. IBHS, which is sponsored by the insurers, defines standards for the Wildfire Protected Home, which is a comprehensive set of measures to keep your home safe from wildfire.
This paper, published in Nature Communications in 2025, describes the results of a study of the factors which led to destruction of structures in five major historical WUI fires in California.
The study used machine learning (ML) analysis of on-the-ground post-fire data collection, remotely sensed data, and fire reconstruction modeling to assess patterns of structure loss and mitigation effectiveness wildfire modeling. The fires which were included in the study are the 2017 Tubbs, 2017 Thomas, 2018 Camp, 2019 Kincaid and the 2020 Glass fires.
Structure survivability could be predicted to 82% accuracy using these tools.
The factor that was the most critical for survivability was structure to structure spacing. While structure spacing cannot be remediated, the study did outline the relative merits of Zone 0 and home hardening. With home hardening and defensiblespace remediation in both Zone 0 and Zone 1, survivability of your home increases from a mere 20% to 52%.
This chart, taken from the Nature Communications publication, shows that home hardening increases survivability by 5% and defensible space in Zone 0 increases an additional 15% survivability, effectively doubling the rate of survival.
The paper can be downloaded here: Fire risk to structures in California's Wildland-Urban Interface