Modeling wildfire potential in residential parcels: A case study of the north-central Colorado Front Range

Abstract

This study evaluated if present-day wildfire potential (i.e. potential fireline intensity and percentage crown fire) differs for residential parcels developed at different time periods in the north–central Colorado Front Range. To answer this question, a model of wildfire potential was built based on 2001 fuels and vegetation and compared the output to actual fire severity of the 2002 Hayman and 2004 Picnic Rock fires (measured by satellite imagery). Except for low-load fuel types such as grass, the modeled wildfire potential corresponded well to observed fire severity. Wildfire potential was then evaluated within 7 classes: developed (1880–1944, 1945–1959, 1960–1974, 1975–1989, 1990–2005) and undeveloped (either zoned or not zoned for development). The results suggest that there is one class characterized by relatively low wildfire potential (developed 1880–1944) and three classes characterized by relatively high wildfire potential (developed 1960–1974 and the two undeveloped parcel classes). These results hold both for 99th percentile (extreme) and 50th percentile (average) fuel conditions. The results suggest that under current zoning regulations, future structures are likely to be built on parcels that, on average, have somewhat higher potential fireline intensity and higher percentage of crown fire compared to currently developed parcels. However, the location of future development may be influenced by forest changes, such as the visual degradation and perceived fire hazard of trees killed by the continuing mountain pine beetle outbreak. Overall, this study introduces an improved method for quantifying wildfire potential in the rapidly developing wildland–urban interface that could be applied to other areas.