Subseasonal Predictability of Fire Weather Metrics for Decision Support

Wildfires have become more prevalent and devastating in recent decades. Pre-planning efforts can improve resource allocation and communications when weather and fuel conditions create high risks for wildfires. Forecasters rely on fire weather metrics, such as the Hot Dry Windy Index (HDWI), that incorporate metrics with distinct predictability timescales to estimate where and when conditions have the potential to exacerbate wildfire management efforts. These metrics are limited by the forecast length (typically 7-14 days), accuracy, and reliability – all of which are highly variable and model dependent. Extended forecasts (32-45 days) are now available from the Unified Forecast System (UFS) and Subseasonal Experiment (SubX), but are untested for fire-weather decision support. Subseasonal-to-seasonal (S2S) forecasts will allow wildland fire managers to (1) better plan for fire conditions, (2) better allocate resources, and (3) improve communication of upcoming risks to local communities. Our work will generate a database of fire weather variables for UFS and SubX retrospective forecasts and multiple reanalysis datasets. We will test calibration schemes and multiple downscaling protocols (e.g., dynamic: EPIC, statistical: bilinear interpolation, machine learning) to evaluate the value of different post-processing steps for fire weather forecast skill and usability. In this session, we will cover the preliminary status of this project, including (a) stakeholder feedback on key fire weather variables needed for boots-on-the-ground decision making, (b) our scientific approach, and (c) timelines for data availability. We invite open discussion on the next steps and collaborations.