Utilization of Satellite-derived Information for Improved Wildland Fire Behavior Forecasting

The United States has entered a new era of increasing wildfire frequency, size, and intensity, which has culminated in many devastating wildfire seasons over the past decade. Fire behavior is becoming more extreme, with many fires becoming large and hot enough that they create their own weather. This creates a situation of enhanced danger for firefighters and first responders. Thus, there is an urgent need for nowcasts of fire behavior for operational decision-making, especially in the early, critical hours of a fire when intervention is most likely to change the outcome. In those early hours, satellite-derived fire detections are usually the only source of information for initializing fire behavior nowcasts. The primary goal of this project is development and dissemination of fire behavior nowcasts to be used specifically within NOAA’s Next Generation Fire System (NGFS) that aims to provide high-quality, low-latency products salient to fire forecasters and other operational users within the fire weather community.

This presentation will report on the work we’ve done to leverage the development of the open-source Weather Research and Forecasting Fire Spread (WRF-SFIRE) modeling system supported by NASA Applied Sciences to integrate with NOAA’s NGFS. The version-1 methodology and workflow developed to use satellite-derived information from NGFS in fire spread forecasting will be described. For the first time, we have a fully automated system capable of running forecasts for all new named incidents over the continental United States, triggered by new satellite fire detections. We successfully implemented this system in 2023, and the hundreds of forecasts made during the 2023 fire season will be evaluated in terms of fire spread and fire radiative power. The impact of satellite-derived information on fire spread forecasting will be evaluated. Computational costs of the system are quantified. The use of WRF-SFIRE to identify the fires with the greatest explosive potential will be discussed. Automated processes used to deliver fire prediction products to users will be described. The presentation will discuss planned enhancements to the version-1 methodology to be implemented before the start of the next fire season.