WindNinja Simulations in Canyons and Complex Terrain

Historically, wildland fires that occur over complex terrain exhibit extreme fire behavior and pose the biggest safety threat to wildland firefighters. Fire spread over complex terrain is not well understood and is still very challenging to predict. A fire’s behavior is extremely impacted by changes in wind speed and direction, something that is highly variable over complex terrain. Thus, it is vital that we understand more about atmospheric flow over complex terrain and the effects that it has on fire spread in these scenarios.

To better understand wind flow over complex terrain, simulations on a local scale were performed with WindNinja, one of the many tools used by emergency responders during wildland fires. The microscale processes happening in complex terrain are hard to capture with conventional weather models, such as the Weather Research and Forecasting model (WRF), which operates on a larger scale. Therefore, WindNinja will be used due to its ability to capture terrain-induced wind flow at smaller scales with fine resolution. This means that it is possible to simulate flow through valleys and over individual ridges where wind speeds tend to be the highest. The ability to simulate winds over complex terrain at finer resolutions is of immense value to agencies such as utility companies, who have powerlines in remote places over this type of terrain, where strong winds pose a significant risk of fire ignition. This type of forecast would directly aid in the decision-making of risk management plans such as public safety power shutoffs.

In this research, WindNinja’s ability to simulate flow over complex terrain will be evaluated. The two different solvers within WindNinja will be evaluated and compared. The conducted simulations were executed with different input data. First simulations were executed using observational data from surface weather stations as input. Further simulations were conducted with gridded weather model data from WRF. These simulations with differing input data were then compared to evaluate what data yielded better results within WindNinja. This study also aims to understand in which scenarios WindNinja performs best by simulating cases where, for example, a downslope windstorm or on-shore flow took place. The performance of WindNinja will additionally be evaluated during these scenarios to understand its ability to resolve flow over complex terrain in comparison to other conventional weather models during downslope windstorms and on-shore flow.