Testing and extending models of fire-induced tree mortality across the US
Managers can use the First Order Fire Effects Model (FOFEM) when planning prescribed burns to achieve mortality-related objectives and for creating post-fire salvage guidelines to predict which trees will die soon after fire. Of the preceding observations, 13,460 involved trees that burned twice. Researchers evaluated the post-fire tree mortality models in FOFEM for 45 species. Approximately 75% of models tested in the FOFEM had either excellent or good predictive ability. Models performed best for thick-barked conifer species. Models tend to overpredict mortality for conifers with moderate bark thickness and underpredict mortality in primarily angiosperms or thin-barked conifers. Managers who rely on these models can use the results to (1) be aware of the uncertainty and biases in model predictions and (2) choose a threshold for assigning dead and live trees that optimizes certainty in either identifying or predicting live or dead individuals.