An Evaluation of the FDS and Clark Coupled Wildfire Model

Computational fluid dynamical models are important tools for the study of wildland fires, especially severe wildland fires. However before using these models to study wildfire, validation is necessary, and model results need to be compared to real fires. In this study, the Fire Dynamics Simulator, or FDS, and the Clark coupled atmosphere--fire model, representing two types of fluid dynamical wildfire models, are used to simulate stationary, non-spreading fires in the Meteotron fire experiment. Comparisons between the Meteotron experiments and simuations are made to evaluate the models' performance. Since FDS can produce 'smoke' data, the plume properties in the FDS simulation are compared with the Meteotron observations using a soot--based criterion. There is no 'smoke' data in the Clark coupled model simulations completed in this study, and therefore plume properties in the Clark coupled atmosphere--fire model simulations are compared with the FDS simulation using a plume vertical-velocity-based, or W, criterion. Two different methods of putting energy from the fire combustion process are evaluated in the Clark coupled model simulations on the basis of plume properties and buoyancy fluxes, and the validity of each numerical approach and method is discribed.