JFSP_research_product_search

Project ID:

24-2-06-3

Year:

2024

Date Started:

09/01/2024

Date Completed:

Title:

Optimizing WUI Fire Risk Assessment: From Hazard Analysis to Damage Evaluation

Project Proposal Abstract:

Recent wildland fires that spread into the Wildland-Urban Interface (WUI) have devastated communities, with whole city blocks leveled and dozens of residents perished. Examples in the U.S. (California, Hawaii, Colorado, Tennessee), Canada, and Europe show that this is a monumental problem that will take enormous resources to tackle. The research community has developed mitigation approaches that can help prevent WUI fires from becoming disasters, such as fuels management, defensible space, home hardening, and evacuation and response planning; however, mitigation must be prioritized to the most vulnerable areas to maximize limited resources. Wildfire risk assessments have great potential to be used for this prioritization, as well as to help communities recognize risk and assess the impacts of these efforts. The wildland fire community has a great foundation for landscape-scale fire risk assessment, but these tools are not currently applicable within the WUI where spotting and home-to-home spread dominate losses, rather than Rothermel-type spread through vegetation. Current approaches therefore focus on probabilistic wildland fire modeling and applying an after-the-fact loss function to determine impacts on assets such as homes or communities. Recent advances in wildfire modeling including improved response (fragility) functions for structures based on exposure and the ability to directly model WUI fire spread between structures herald the potential to improve these methods.

We propose evaluation of complementary approaches of calculating fire risk in the WUI. First, the current state-of-the-art methodology using probabilistic modeling of wildland fire spread across a landscape will be used to generate conditional probabilities of fire behavior. This will include exposure from both fire intensity and ember deposition, which is critical for WUI fire spread. These outputs will be used with a set of new response (fragility) functions developed from past fire loss data in California to provide risk assessments. Second, a new WUI model coupled with an existing probabilistic landscape fire spread model (ELMFIRE) will be used to directly model fire spread from vegetation fires into and within WUI communities. By modeling structure-to-structure spread directly, features such as hardening of homes can be simulated and evaluated. These approaches will be systematically evaluated against a database of fire losses in California to determine the accuracy of these methods at characterizing the threat of wildfire to WUI and urban settings. It is hypothesized that different methods will perform better at different scales, potentially indicating the usefulness of different approaches depending on the application.

Losses in the WUI have skyrocketed in recent years. Whole communities have been lost in areas not traditionally considered at the highest risk for these events. Robust methods are needed to take the significant advances that have been produced by the wildland fire risk community and bring them to the WUI. There is incredible potential for these methods to assist communities in optimizing their future mitigation efforts, to better inform communities of their true current risks, and to set the stage for future development.

Principal Investigator:

Michael J. Gollner

Agency/Organization:

University of California-Berkeley

Branch or Dept:

Department of Mechanical Engineering

Other Project Collaborators

Type	Name	Agency/Organization	Branch or Dept
Agreements Contact	Joyce Y So	University of California-Berkeley	Sponsored Projects Office
Budget Contact	Damon A Hinson	University of California-Berkeley	Department of Mechanical Engineering