Wildland Fire Air Quality Response Program

Smoke Modeling

Smoke impact prediction modeling is an important component to creating daily forecasts about how smoke will impact air quality. US Forest Service scientists with the AirFire team develop and customize such models to support the Program. Specifically, the SmartFire2 system uses satellite fire detections and other datasets such as fire perimeters to provide fire activity information (fire location and size) to the BlueSky smoke modeling framework which links together mapped vegetation datasets, fuel consumption and emission models, and plume rise algorithms to provide information on smoke transport and concentration. Daily smoke (e.g. 1-hr PM2.5 concentrations) predictions are provided on a variety of scales, from local highly resolved 1-km scales, to continental 12-km scales. These smoke predictions are used by Air Resource Advisor’s and others to produce products and information useful to the public so they may assess their risk and take necessary actions to protect themselves.

Smoke Model Runs

image of BlueSky Daily Run file download site
image of animated websky map

Access to smoke modeling tools:


Caution on interpreting smoke model output:

  • Disclaimer: Due to the uncertain nature of the smoke model runs, output from these runs must be interpreted with care and placed into a larger context. Air Resource Advisors are specially trained at this type of interpretation, including looking across the BlueSky runs, weather model runs, and other air quality model runs and their past performance to build up a smoke outlook. The model runs are not intended for use by the public or untrained individuals; any use is at the user's own risk. Model runs are for research purposes and may change without notice. Smoke Outlooks are issued for public use and may be found here.