Comparison of Three AVHRR-Based Fire Detection Algorithms for Interior Alaska

Three satellite fire detection models (threshold, contextual, and fuel mask) were compared and evaluated using National Oceanographic and Atmospheric Administration (NOAA)-11, NOAA-12, and NOAA-14 Advanced Very High Resolution Radiometer sensor data from interior Alaska. The fixed threshold model compared the radiant temperature of each pixel to predetermined threshold values. The contextual model compared the radiant temperature of each pixel to its surrounding (background) pixels. The fuel mask model is similar to the contextual model, but pixels were tested for fuel availability according to pre-fire vegetation index values. Fire location data from the Alaska Fire Service was used to assess the accuracy of the fire detection models. Fire detection accuracy: (a) was highest using the fuel mask model; (b) was lowest using the fixed threshold model; (c) increased as fire size increased; (d) was considerably greater in afternoon images than morning or night images. Fire detection methods may be less accurate in taiga/tundra regions such as interior Alaska due to landscape heterogeneity and relatively low aboveground fuel.