Refining Fire Weather Models: Improvements in Data Assimilation and Downscaling Approaches

Fire weather warnings and short-range forecasts are critical for saving lives and minimizing damage during wildfire events. Traditional data assimilation and dynamic downscaling techniques, particularly those using terrain-following coordinates, have been employed with some success in providing these vital warnings and forecasts. However, in complex terrains such as mountainous regions and densely built urban areas, the accuracy of these analyses and forecasts often falls short, necessitating further advancements.

Building on recent successes in meso-scale forecasting, we have upgraded an object-oriented meteorological data assimilation system to incorporate new coordinate options tailored specifically for fire weather forecasting. Among these options is a cut-cell coordinate system, designed to better handle the intricacies of complex terrains. In this presentation, we will discuss the enhanced analysis and downscaling capabilities of this system, demonstrating its effectiveness at resolutions of 100 meters and finer. We will also compare the performance of the cut-cell coordinate system with that of the traditional terrain-following approach, highlighting the improvements and potential implications for fire weather prediction.