NYU wildfire forecasting prototype updates predictions with live observations

NYU researchers say a wildfire forecasting prototype produced near-real-time predictions in tests on three active fires, Phys.org reported. One Grasshopper Fire test achieved a 94.6% match between predicted and observed fire areas using a spatial overlap measure; agreement fell to 72.4% in another forecast after a longer gap between satellite observations. The system combines equations describing fire behavior with incoming observations, including weather, terrain, satellites, and field sensors. Users can select forecast horizons of up to 12 hours and ask plain-language questions about changing conditions. The prototype has not been deployed as an operational tool on active wildfires. The team wants further testing with fire departments and federal agencies before making it public.