Dear Editor, In a paper published in the January 2016 issue of Fire Technology, Hoffman et al. provide an assessment of crown fire rate of spread predictions of two physics-based models, FIRETEC and the Wildland-urban interface Fire Dynamics Simulator (WFDS), through an indirect comparison with a large data set of wildfire observations (n = 57) published by us, which they refer to as AC06. The AC06 data set was compiled from various published sources for the purpose of evaluating the performance of an empirical model we developed for predicting active crown fire rate of spread.

We commend the authors for their effort to evaluate the outputs from such complex model systems against real-world data. However, we disagree with certain statements made by Hoffman et al. regarding: (1) their view of the presumed limitations of our data set derived from case study information of Canadian and U.S. wildfires and (2) a number of conclusions they have reached in their evaluation of FIRETEC and WFDS for the prediction of crown fire rate of spread.