First Street Foundation Wildfire Model Technical Methodology

Executive Summary

The need for an understanding of the wildfire risk across the United States has been described by a number of stakeholders and researchers (Burke et al., 2021; Westerling et al., 2006; Vose et al., 2018). To address the need to accurately describe this risk at the community level and therefore allow for the formulation of appropriate disaster responses, the U.S. Federal Government supported the creation of the publicly-available Wildfire Risk to Communities (see WildfireRisk.org; Scott et al., 2020) which was published in April, 2020. Wildfire Risk to Communities conveys relative fire risk for communities following analysis at a 270m horizontal resolution and based on the US Forest Service’s 2014 Landscape Fire and Resource Management Planning Tools (LANDFIRE, 2020) database which provides open data on the composition and state of fuels across the United States. Given the rapid increase in wildfire occurrence (Smith, 2022) and the resultant consequences ($8.5 billion between 2012 and 2016, with over $79.8 billion in costs associated with wildfires between 2018 and 2021; NOAA, 2022) over the past 30 years that is often ascribed to the impacts of climatic changes (Abatzgalou and Williams, 2016), there is an apparent need to evaluate the probable changes in wildfire exposure which may be expected in the near future in order to provide US residents a more informed understanding of their risk.

Following the open science approach taken by the nonprofit First Street Foundation for climate-adjusted flood risk (Armal et al. 2020; Bates et al. 2021; Kearns et al. 2020), a new analysis was undertaken that would estimate the wildfire exposure at a property-by-property scale across the United States. The technical documentation for the methodology used by the First Street Foundation Wildfire Model (hereafter, FSF-WFM) to compute estimates of the 30-year, climate-adjusted aggregate wildfire hazard for the contiguous United States at 30 meter horizontal resolution is presented.

The FSF-WFM integrates several existing methods from the wildfire science community and implements computationally efficient and scalable modeling techniques to allow for new high-resolution, hazard generation across the contiguous United States (CONUS). Using U.S. Federal Government open data as a baseline, additional data were added from a variety of sources which facilitate both a high resolution (30m horizontal) and future-facing (for today and for 30 years ahead) product allowing individuals, communities, businesses, and governments to better understand and prepare for their wildfire risk. Burn probability, flame length, and ember spread for the years 2022 and 2052 are computed from two 10-year representative Monte Carlo simulations of wildfire behavior, utilizing augmented LANDFIRE fuels estimates updated with best-available disturbance information. FSF-WFM utilizes ELMFIRE, an open-source, Rothermel-based wildfire behavior model, and multiple US Federal Government open data sources to drive the simulations. LANDFIRE non-burnable fuels classes within the Wildland Urban Interface (WUI) are replaced with fuel estimates from machine-learning models trained on data from selected historical fires to allow the propagation of wildfire through the WUI in the model. Historical wildfire ignition locations and NOAA’s hourly time series of surface weather at 2.5km resolution are used to drive ELMFIRE to produce wildfire hazards representative of the 2022 and 2052 conditions at 30m resolution, with the future weather conditions scaled to the IPCC CMIP5 RCP4.5 model ensemble predictions. As simplifying assumptions, both winds and vegetation were kept unchanged between the 2022 and 2052 simulations, and thus climate change’s impacts on the future fuel conditions are the main contributors to the changes observed in the 2052 results. In addition, as a number of communities in the built environment have suffered extensive losses (e.g. Gatlinburg, TN 2016; Paradise, CA 2018; Boulder County, CO 2021) there is also a need to capture the spread and risk of wildfire within the Wildlands Urban Interface (WUI; Glickman & Babbitt, 2001).

The compound wildfire risk product resulting from this model has been named “Fire Factor” as part of a wider “Risk Factor” information platform provided by the First Street Foundation for free, public use under noncommercial license terms (see riskfactor.com). Nonzero wildfire exposure is estimated for 71.8 million out of 140 million properties across CONUS. Climate change impacts add another 11% properties to this non zero exposure class over the next 30 years. Much of this change in wildfire exposure is observed in forested areas east of the Mississippi River. “Major” aggregate wildfire exposure of greater than 6% over the 30 year analysis period from 2022-2052 is estimated for 10.2 million properties. The FSF-WFM represents a notable contribution to the ability to produce property-specific, climate-adjusted wildfire risk assessments in the US.