Operational use of mesoscale models in fire weather forecasting

As in most areas of the western United States, fire weather forecasting in the Rocky Mountains is challenging due to the complex terrain of the area. Mesoscale numerical models can provide a more detailed representation of this terrain and therefore guidance that takes into account the mesoscale forcing mechanisms influencing the weather conditions. This guidance is especially useful in preparing forecasts for a specific fire location.

During the fire season of 1999, the National Weather Service (NWS) Office in Grand Junction, CO began using adapted output from the Colorado State University (CSU) Regional Atmospheric Modeling System (RAMS) that provided specific guidance for fire weather forecasting. Researchers at CSU provided parameters specific to the fire weather forecasting problem (Haines index, mixing height, transport winds), and supplied time series plots of weather parameters for fire weather observing stations. This output gave detailed information to help forecasters predict the various fire weather parameters and their trends.

To evaluate the usefulness of this guidance, a verification study was initiated to compare the guidance from RAMS with the other numerical guidance available to the NWS offices. Also, field personnel provided feedback on the usefulness of this more detailed guidance. The results of this study are presented.