Description
In large parts of the US the chaparral zone typifies the natural landscape. Chaparral habitat is one of grass lands containing short vegetation where larger shrubs dot the landscape. This creates a wildfire scenario where flames spread across the grasslands (fuel bed) and, if conditions are right, will ignite the overhead vegetation (crown fuel) in the shrubs. An effort to study this type of fire spread has taken place over the years. Dr. Watcharapong Tachajapong and researchers at UC Riverside and the Pacific Southwest Research Station of the USDA Forest Service conducted a series of experiments that examined to effects of altering various parameters related to fire behavior. In addition to the experimental work there have been previous efforts of Drew Castle at SDSU and Dr. William Mell to model this these experiments using Fire Dynamic Simulator. This work is being continued in this body of research with more simulations performed in an updated specialized version of FDS6 to model certain scenarios that were created in UC Riverside experiments. Other bodies of research also exist around the areas of Froude number scaling, flame tilt angle, and numerical modeling of wildfires and were used as a basis of comparison to the simulation results. In the case of the comparisons between the UC Riverside experiments and this research, there were strong correlations between the numerical model results and the experimental results for cold flow behavior, flame rate of spread (ROS) and flame tilt angle. Flame spread from surface fuel to crown had mixed correlation to the experiments. Results were also compared against literature on flame tilt angle and were found to be substantially different than predicted by the literature (around 75% difference). Froude number scaling predictions compared very closely between the experimental scale and the naturally occurring wildfires. The correlation factor associated with wind velocity was also compared and found to closely match between the predicted difference on flame ROS and that in the numerical models.
