Abstract
Modeling flow in vegetative fuel beds is a key component in any detailed physics-based tool for simulating wildland fire dynamics. Current approaches for drag modeling, particularly those employed in multiphase computational fluid dynamics (CFD) models, tend to take a relatively simple form and have been applied to a wide range of fuel structures. The suitability of these approaches has not been rigorously tested for conditions which may be encountered in a wildland fire context. Here, we focus on beds of Pinus rigida needle litter and undertake a two-part study to quantify the drag and evaluate the capabilities of a multiphase large eddy simulation CFD model, the NIST Fire Dynamics Simulator. In the first part, bed drag was measured in a wind tunnel under a range of conditions. The results were fit to a Forchheimer model, and the bed permeability was quantified. A traditional approach employed in the multiphase formulation was compared to the parameterized Forchheimer equation and was found to over-predict the drag by a factor of 1.2–2.5. In the second part, the development of a velocity profile above and within a discrete fuel layer was measured. Using the Forchheimer equation obtained in the first part of the study, the CFD model was able to replicate a qualitatively consistent velocity profile development. Within the fuel bed, the model appeared to under-predict the velocity magnitudes, which may be the result of unresolved pore-scale flow dynamics.