Fire behavior icon

Fire Behavior

Advancing research on fuel mapping and fire spread to inform next generation wildfire models

Operational wildfire behavior models are empirical and cannot be adapted to predict extreme fire behaviors resulting from prolonged heat release by large woody fuels and deep duff layers typical of modern California forests. Current fire and fuel models were developed in the 1970s to predict the steady advance of a thin linear flame zone assuming heat release only from flaming combustion of continuous fine fuels for periods of 10–30 seconds. Fuel classification and mapping ignores the 10–100 times greater fuel biomass of large-diameter woody material and deep duff layers now commonplace. These fuels burn for hours, leading to large-area fires (not thin flame zones) and heat release rates that vary widely depending on fire-induced ventilation of solid-phase combustion and smoldering-to-flaming transition. Without a fundamental understanding of solid-fuel combustion under large wildfire conditions we will never approximate the actual fire potential associated with the loads and extent of woody fuels as millions of drought-killed trees fall to the ground.

To integrate results from experimental work with detailed and extensive fuel characterizations , WG2 will use an ensemble approach to draw from measured fuel distributions and determine the variability in fire behavior predictions for the current and projected forest fuel conditions. Nonlinear dependence of fire behavior on fuel characteristics means that fuel ensembles are essential to modeling extreme fire behaviors and their relative frequencies in mixed-conifer forests with high tree mortality. The current spatially averaged fuel descriptions and fire behavior models ignore these critical dependencies. Results from WG2 will enhance the efforts to forecast near-term risk (WG3). The two groups work in parallel. WG2 focused on addressing known limitations in the contemporary fire-spread models while WG3 pushes the envelope of our current capacity to capture fire risk to electricity assets. The strategy of linking application with basic research ensures that immediate needs are met while also addressing likely future challenges. Such an anticipatory design is essential to keep pace with rapidly changing environmental conditions.

Contemporary wildfire events

Current fire models fail to address the fundamental physics behind contemporary wildfire events in the State of California. This team aims to deliver fundamental scientific advances to give rise to the next generation of wildfire models, including a new fuel mapping technique and fire spread model based on fire physics.

Devise experimental apparatus and test in a laboratory setting the predicted heat release rates across the range of fuel structures and environmental conditions found in wildland areas

Develop and employ a new fuel measurement and mapping system to resolve the essential fuel components and spatial heterogeneity in fuels occurring at multiple scales

Map current and projected future fuel conditions in areas of elevated tree mortality

Evaluate how to integrate the products into near-term risk forecasts

High Impact Products

Web map icon

A new fuel measurement and mapping system

Data icon

A new fire behavior model

Analysis icon

Analysis of fuels over a 20 year life-cycle in areas of elevated tree mortality across the State

Fire Behavior Team

Scott Stephens

Scott Stephens, PhD

UC Berkeley

Lead for Workgroup #2 - Fire Behavior.

John Battles

John Battles, PhD

UC Berkeley

Co-lead for Workgroup #2 - Fire Behavior. Field Campaign Lead

Brandon Collins

Brandon Collins, PhD

UC Berkeley

Field data collection and analysis

Michael Gollner

Michael Gollner, PhD

UC Berkeley

Laboratory test and analysis

David Marvin

David Marvin, PhD

Salo Sciences

Fuels mapping

Chris Anderson

Chris Anderson, PhD

Salo Sciences

Fuels mapping

Mark Finney

Mark Finney, PhD

US Forest Service - Missoula Fire Lab

Lead for laboratory testing and analysis

David Saah

David Saah, PhD

Spatial Informatics Group and University of San Francisco

Principle Investigator

Shane Romsos

Shane Romsos, M.Sc.

Spatial Informatics Group

Project Manager