Understanding of the conditions that contribute to wildfire ignitions and impacts increases capacity to mitigate wildfire risks. The Fire Program Analysis Fire-Occurrence Database (FPA FOD) contains information on the location, jurisdiction, discovery time, cause, and final size of more than 2 million wildfires from 1992 through 2020. To each of those wildfire records, we added information on 267 physical, biological, social, and administrative attributes. As we will demonstrate, these publicly available data can be used to answer numerous questions about the circumstances associated with human- and lightning-caused wildfires. We will share examples of how the enhanced FPA FOD data can support descriptive, diagnostic, predictive, and prescriptive wildfire analytics, including the development of machine learning models.
Presenters:
Mojtaba Sadegh, Boise State University
Karen Short, USDA Forest Service