Development and application of a probabilistic method for wildfire suppression cost modeling

Abstract

Wildfire activity and escalating suppression costs continue to threaten the financial health of federal land management agencies. In order to minimize and effectively manage the cost of financial risk, agencies need the ability to quantify that risk. A fundamental aim of this research effort, therefore, is to develop a process for generating risk-based metrics for annual suppression costs. Our modeling process borrows from actuarial science and the process of assigning insurance premiums based on distributions for the frequency and magnitude of claims, generating parameterized probability distributions for fire occurrence and fire cost. A compound model of annual suppression costs is built from the coupling of a wildfire simulation model and a suppression cost model. We present cost modeling results for a set of high cost National Forests, with results indicating variation in expected costs due to variation in factors driving financial risk. We describe how our probabilistic cost models can be used for a variety of applications, in the process furthering the Forest Service's movement towards increased adoption of risk management principles for wildfire management. We review potential strengths and limitations of the cost modeling process, and conclude by discussing policy implications and research needs.