Modeling backing fires in California grassland fuels

This study had three objectives. The first was to develop a better measure of the depth of grassland fuels for use in wildland fire behavior modeling and prediction. The second was to develop a correction for BEHAVE's predictions of backing fire spread rate in this fuel type. The third was to correlate fuel bed characteristics (fuel load, depth, bulk density, and fuel moisture) with flame characteristics (spread rate and flame length).; The general approach that was adopted for this study was to conduct detailed sampling of the vertical distribution of biomass, transport natural fuel beds (1.0 x 1.0 m) to a laboratory, and document fire behavior in the absence of wind and slope effects following a line source ignition. Flame length was assessed using 35mm photography. Spread rate was derived from the temperature profiles generated using a series of Type-K thermocouples placed at 10 cm intervals along the length of the fuel bed.; The biomass sampling indicated that, on average, greater than 90% of the total biomass is located in the lowest 65% of a grass fuel bed. The depth of the fuel bed containing greater than 90% of the total fuel load was defined as the fuel bed's “effective depth.” Effective depth was found to be related to measured depth by the equation (ED = 0.75 D − 0.02), where ED is effective depth and D is the depth measured using standard protocols (R2 = 0.94).; The BEHAVE fire behavior prediction program was then tested for its ability to predict backing fires in this fuel type. Observed spread rates ranged from 0.7 to 2.2 m/min and flame lengths ranged from 0.7 to 1.2 m. The program consistently under predicted both spread rate and flame length. A correction factor was found to be a function of effective depth (C = −1.47 ln(ED) − 0.78, where C = observed spread rate/predicted spread rate and ED is the effective depth; R2 = 0.75).; Fuel characteristics were then analyzed for their effects on fire behavior. Fuel load class and fuel moisture class (low, medium, high) had an influence on spread rate, while effective depth class and bulk density class did not. Effective depth class and fuel moisture class had an influence on flame length, while fuel load class and bulk density class did not. In addition, the interaction effect of fuel moisture class and fuel load class was found to be significant.