Examining the ability of numerical models to forecast pyrocumulus

Over the past decade the interest in including wildfires in real time forecasts has increased. In addition to the impact that smoke from wildfires can have upon near-surface regional air quality, several recent journal articles have provided evidence that very often wildfire smoke, through pyrocumulus, reaches the lower stratosphere. At high altitude wildfire smoke impacts not only the regional weather, but stratospheric chemistry and the global climate as well. These pyrocumulus clouds are also important to the fire incident command as these storms can produce strong gusty wind, significantly decreased visibility and intense fire whirls. Therefore, the inclusion of pyrocumulus forecasts is of interest to the general wildfire and weather forecasting communities. But despite the interest in forecasting of pyrocumuls, little is known about the numerical model's ability to accurately predict their occurrence.

To begin addressing this question, a retrospective simulation using the WRF model is being made for the 2009 Station wildfire. Using output from the numerical model, the simulated environmental conditions are being examined to determine the indicators for when and where pyrocumulus will develop in the model. In addition, satellite and ground-based observational data from the Station fire are being used to initialize the model and provide evidence of the numerical simulation's validity. An overview of this preliminary work will be presented.