Wildfire’s destruction of homes is an increasingly serious global problem. Research indicates that characterizing home hardening and defensible space at the individual structure level may reduce loss through enriched understanding of structure susceptibility in the built environment. However, improved data and methods are required to accurately characterize these features at scale. This paper does three things: (1) Identifies features correlated with structure loss. (2) Compares methods of characterizing structure susceptibility, including home assessments and emerging fire spread models. (3) Evaluates methods and open data sources used to measure these features. We find that relative feature importance varies widely among studies due to data limitations and scale issues. Built-environment fire spread models show limited inclusion of structure-level features. Additional research, model validation, improved data, and improved data collection methods are needed to bridge the gaps between primary research, susceptibility indices, and built-environment fire spread models. Advancing scalable methods for characterizing built-environment fuels and susceptibility will refine risk mitigation efforts globally.
Extreme spread events (ESEs), often characterized by high intensity and rapid rates of spread, can overwhelm fire suppression and emergency response capacity, threaten responder and public safety, damage landscapes and communities, and result in high socioeconomic costs and losses. Advances in remote sensing and geospatial analysis provide an improved understanding of observed ESEs and their contributing factors; however, there is a need to improve anticipatory and predictive capabilities to better prepare, mitigate, and respond. Here, leveraging individual-fire day-of-arrival raster outputs from the FSim fire modeling system, we prototype and evaluate methods for the simulation and categorization of ESEs. We describe the analysis of simulation outputs on a case study landscape in Colorado, USA, summarize daily spread event characteristics, threshold and probabilistically benchmark ESEs, spatially depict ESE potential, and describe limitations, extensions, and potential applications of this work. Simulation results generally showed strong alignment with historical patterns of daily growth and the proportion of cumulative area burned in the western US and identified hotspots of high ESE potential. Continued analysis and simulation of ESEs will likely expand the horizon of uses and grow in salience as ESEs become more common.