The Department of Defense uses prescribed fire to manage millions of acres across a variety of ecosystems in the United States to accomplish multiple objectives. To properly implement prescribed fire and limit potential escapes, an improved understanding is needed of fundamental science questions related to combustion and fire propagation in natural fuel beds that are a mix of both live and dead fuels. This study examined aspects of pyrolysis, the thermal breakdown of solid wildland fuels to produce the gases that combust resulting in fire. To improve the understanding and modeling capability of pyrolysis in physics-based fire spread models, three goals were identified: 1) provide more detailed description of pyrolysis and the evolution of its products for a greater variety of southern fuels than is currently known, 2) determine how convective and radiative heat transfer from flames to live fuel particles influenced pyrolysis and ignition at laboratory and field scales and 3) gain more detailed insight into pyrolysis, combustion and heat transfer processes in wildland fire spread through the use of high-fidelity physics-based models.

These goals were achieved by accomplishing three technical objectives (tasks) which were supported by a 0th foundational task: 0) characterize the physical, chemical, compositional and spatial structure of wildland fuels used in this project, 1) characterize pyrolysis products by measurement of a variety of live and dead foliar fuels in laboratory and small-scale field experiments, 2) determine the effects of convective and radiant heat transfer on pyrolysis and 3) perform high-fidelity physics-based modeling of pyrolysis for bench-scale and wind tunnel experiments. Physical and chemical properties were determined using a variety of analytical methods. Wildland fuels were described using a variety of traditional two-dimensional (2D) and innovative three-dimensional (3D) sampling methods using Light Detection and Ranging and terrestrial laser scanning. Pyrolysis gases were generated from single leaves at bench-scale by slow heating in pyrolyzer and fast heating using combinations of convective and radiant heating in a flat-flame burner. Pyrolysis gases from fuel beds of live and dead fuels were measured in a wind tunnel and in the field in small, prescribed burns at Ft. Jackson, SC by capturing gases in canisters or in real-time using nonintrusive Fourier Transform InfraRed (FTIR) spectroscopy. Bench scale, wind tunnel and field gases were identified using gas chromatography/mass spectrometry and gas chromatography/flame ionization detection. Wind tunnel and field gases were also identified using static and dynamic (time-dependent) gas collection and identification using FTIR spectroscopy. Data were analyzed using a variety of statistical methods. Modeling of bench-scale pyrolysis experiments was accomplished using the General Pyrolysis (GPYRO) model coupled with the Fire Dynamics Simulator (FDS) model. Wind tunnel experiments were modeled using the FDS model.

Composition of foliar fuels was found to be appreciably different from wood. The 3D description of fuel beds provided more information for physics-based fire models than traditional 2D sampling. The statistical field of compositional data analysis was applied to pyrolysis and combustion gas mixtures for the first time. In the bench-scale measurements, the relative amounts of pyrolysis gases were affected by moisture content and heating mode (convective versus radiant). First successful measurement and description of pyrolysis gases under realistic fire conditions was accomplished, both in the wind tunnel and in the low intensity prescribed burns at Ft. Jackson. Fuel heating rates, maximum fuel temperatures and fuel conditions were found to be similar for the wind tunnel fires and the prescribed burns. The relative amounts of pyrolysis gases differed, however, between the wind tunnel and the field experiments. Dynamic changes in gas composition measured by FTIR were correlated with fire phase determined by infrared camera in wind tunnel experiments. The GPyro model was modified substantially to improve modelling of evaporation from foliar fuels. Replacement of an Arrhenius-based model with an equilibrium model for evaporation had greater impact on high fuel moisture fuels. Drying dynamics from the equilibrium model is more consistent with the physics of evaporation. Modeling revealed that fluid dynamics play a distinctive role in evaporation, pyrolysis, ignition, combustion, and burnout behavior of leaves. Fluid flow was influenced by leaf orientation (horizontal or vertical). The addition of radiative heating to a combined heat flux reduced the time that a fuel particle lost 50 percent of initial mass by only a third suggesting that convective heating had a greater impact on pyrolysis and burning of an individual leaf.

The primary benefits of the project are the information and modeling related to pyrolysis of intact wildland fuels. Prior pyrolytic work strove to minimize the effects of heat transfer and fuel moisture on pyrolysis. The present study showed that heat transfer mode and fuel moisture are both important factors that should be considered when modeling pyrolysis in modern physics-based fire models. Similarly, results from laboratory fires under standard conditions differed from field measurements. Comparison of results under standard conditions with bench-scale results in oxidizing and non-oxidizing environments will be limited to a small subset of the measured gases and is outside the scope of this project. Improvements to the modeling of evaporation and a more complex pyrolysis and combustion framework in GPYRO-FDS better model these processes; however, the impact of these improvements on fire spread predictions is unknown and needs to be explored in future work to determine if the improvements warrant the additional complexity. The introduction of compositional data analysis, a branch of statistics thus far overlooked by the wildland fire community and a significant outcome of this project, has the potential to provide scientific results based on statistical analyses suited to the nature of many types of wildland fire data describing the composition of things.