Book Description

A wildland fire spreads when thermal energy heats up nearby fuel particles leading to their ignition. This heat transfer can only involve convection and radiation heat transfer. It is commonly assumed that radiation heat transfer is the dominant mechanism; that is, fire spread is primarily governed by radiation heat transfer. The purpose of this study was to quantity the contributions of convection and radiation prior to ignition and to test the assumption that radiation heat transfer is the dominant mechanism. The study used (a) mathematical modeling and (b) experimental methods. The mathematical model involved a two-dimensional, transient, finite-difference solution to the conduction heat equation using standard heat transfer equations. The mathematical model was not tuned to match the experimental data because the purpose of the model was to represent the physical processes. One set of experiments controlled fuel particle exposures to a radiant panel and another set of experiments had particles exposed to flame fronts during spreading fire. During the controlled experiments, irradiances were between 29.8 kW/m2 and 36.4 kW/m2. Fuel particles were cooled by free convection in some experiments and forced convection in others. All experimental fuel particles were fabricated from yellow poplar (Liriodendron tulipifera) and square in cross section. Particle sizes were 1, 3, 6, 9 and 12 mm for the controlled experiments and 1 and 12 mm for the fire spread experiments. The temperatures versus time plots predicted by the numerical model closely matched the shapes of the measured temperature profiles. Thus the mathematical model accurately captured the physics. Both experimental and numerical results from the controlled experiments showed that radiation heat transfer was not sufficient to ignite the 1 mm particle due to convective cooling. Experimental and numerical results from the fire spread experiments showed that convection (not radiation) was the dominant mechanism responsible for heating 1 mm particles to ignition for conditions relevant to wildland fires. These results indicate the need to consider both convective and radiative heat transfer at fuel particle scales in physical wildland fire spread models.