Reliable forecasting of regional weather and wind flow patterns is critical for effective fighting of wildland fires and operational management of prescribed burns. Accurate forecast of future wind fields, relative humidity, and stability are essential for predicting fire behavior; smoke dispersion, and mitigation of public health risks. Regional weather forecasts are currently produced by 3D mesoscale simulation models, which are quite sensitive to lower boundary conditions defined by the surface exchange of heat and water vapor between terrestrial ecosystems and the atmosphere. These exchange processes are poorly described in current meso-models. This causes great uncertainties in the weather predictions beyond 10 hours.

The paper discusses a new initiative to dramatically improve the representation of land-surface biophysical processes in the MM5 mesoscale model. The goal is to significantly increase the accuracy of regional forecasts of fire weather and airflow. We plan to achieve this by coupling MM5 with a state-of-the-art biophysical model of soil-vegetation-atmosphere interactions called FORFLUX. The paper presents details about the FORFLUX model, its input data requirements, and verification history, as well as methods for linking it to MM5 as a new surface scheme.