The danger of destructive wildfires has become a major issue in the U.S. due to decades of fuel accumulation from fire suppression efforts. With more people moving closer to or living in the wildland-urban interface, the risk of wildfires has increased dramatically. Wildfire risk analysis is aiming at predicting when and where wildfires will likely occur, and measures can be taken in advance to reduce the number and the intensity of disasters. In general, wildfire risk analysis can be divided into ignition risk analysis and fire behavior risk analysis. This study focuses on an ignition fire risk analysis in Rabun County, Georgia. Four fire risk related factors — human activity, illumination, elevation and vegetation type — were derived from spatial datasets. Three of them were used as linguistic variables, and then fuzzy set theory and fuzzy inference were applied to these variables to model wildfire risk levels (low, possible, substantial and high). The final result is a thematic risk level map that suggests that fuzzy logic can be used as a powerful tool in the field of fire risk analysis to predict more interpretable risk levels for a certain area. Some advantages and disadvantages of using fuzzy logic over traditional approaches were also discussed in the paper.
Keywords: wildfire risk, fuzzy logic, fuzzy inference, linguistic variables, membership function