Comparison of a coupled weather-wildland fire model simulation and FireMapper thermal imaging data of a Santa Ana-driven Wildfire

Beyond estimating the direction and rate of spread of a fire's leading edge, modeling aims to reproduce and explain fire phenomena and why large fire events unfolded as they did. This work examines the Esperanza fire, a Santa Ana-driven wildfire that occurred in complex mountainous terrain in spatially heterogeneous chaparral fuels, using airborne remote sensing imagery from the FireMapper thermal-imaging radiometer and the CAWFE coupled numerical weather prediction-wildland fire model. Radiometer data mapped fire intensity and was used to reveal dynamically active regions of the fire fronts, their intensity, and their depth, and to evaluate simulations of these properties. The model simulated the meteorological flow in and near the wildfire, the fire's growth as driven by weather, and the interactions between the fire and weather through fire-induced winds during the fire's first day. The airflow was characterized by thermally stratified, two-layer flow channeled between the San Bernardino and San Jacinto mountain ranges with transient flow accelerations impacting the fire in the lee of Cabazon Peak. The simulation had rough agreement in area, shape, and spread direction compared to known positions and remote sensing data and reproduced several distinguishing features of the fire – the rapid spread to the west-southwest, runs up canyons, splitting of the fire head, and feathering at the leading edge.