An Analysis of Fire-Atmosphere Interactions on Fire Propagation in Steep Canyons Using Idealized WRF-SFIRE Simulations

Wildfires are known to create their own fire-induced circulation which can contribute to the spread of the fire. One of the most important places to monitor these condition changes are in steep canyons, which tend to funnel winds and accelerate fire propagation. These are also among the most dangerous places where fire blow-ups can threaten firefighters. This study analyzed how fire-induced conditions produced by fires burning in steep canyons affect fire acceleration. The fires were represented as two idealized WRF-SFIRE simulations, one looking at conditions with, and one without, the fire. The fire-induced perturbation was estimated by subtracting a given variable simulated in the “No Fire Run” from the output of the “Fire Run” (Fire - No Fire). The runs analyzed three different canyon heights: 10 m, 210 m, and 480 m, with different slope angles of 0º, 20º, and 40º, respectively, to investigate fire acceleration and fire-induced conditions. For the purpose of this study, the wind speed, temperature, and pressure fields simulated with and without fire were analyzed in order to assess the spatial and temporal variability in the fire-induced circulation. Fire rate of spread (ROS) was also be analyzed to determine how canyon terrain impacts eruptive fire behavior. Two wind regimes, one with a constant southerly wind at 5 m/s through the whole vertical atmosphere, and another with a constant 5 m/s southerly wind until 250 m and with varying wind shear afterwards, were analyzed in order to see how vertical changes in wind can affect fire spread on the surface. In this study, it was found that the 10 m canyon configurations generally had the weakest fire-induced circulations and ROS, while the 480 m canyon had the greatest. In addition, ROS values increased when wind shear was introduced in the atmosphere and the fire was able to reach the walls of the canyon. Lastly, these idealized cases were compared to actual laboratory cases from Viegas & Pita, 2004 in order to determine how well fire-atmosphere coupled models resolved eruptive fire behavior. Overall, the fire spread maps showed similar patterns to that of the laboratory studies, as the fires always started off with elliptical fire growth, and once it reaches the canyon walls, a heart shaped lobe can develop depending on the wind conditions. Understanding and validating coupled fire-atmosphere model runs is very important to establish the current understanding of fire modeling capabilities and finding any improvements for the future, as these models still are not perfect. This study into canyons was designed as the first step into understanding to what degree simple coupled fire-atmosphere models can capture eruptive fire behavior and how important fire-atmosphere coupling is in accelerating fire progression. Further research into the processes behind the fire-induced circulation will lead to a better understanding of fire behavior and the extent to which fire can alter the environment surrounding it. This will help to assess limitations of currently used uncoupled models and develop strategies that will improve firefighter safety during fires in complex terrain.