Predicting postfire Douglas-fir beetle attacks and tree mortality in the northern Rocky Mountains

Abstract

Douglas-fir ( Pseudotsuga menziesii (Mirb.) Franco) were monitored for 4 years following three wildfires. Logistic regression analyses were used to develop models predicting the probability of attack by Douglas-fir beetle ( Dendroctonus pseudotsugae Hopkins, 1905) and the probability of Douglas-fir mortality within 4 years following fire. Percent crown volume scorched (crown scorch), cambium injury, diameter at breast height (DBH), and stand density index for Douglas-fir were most important for predicting Douglas-fir beetle attacks. A nonlinear relationship between crown scorch and cambium injury was observed, suggesting that beetles did not preferentially attack trees with both maximum crown scorch and cambium injury, but rather at some intermediate level. Beetles were attracted to trees with high levels of crown scorch, but not cambium injury, 1 and 2 years following fire. Crown scorch, cambium injury, DBH, and presence/absence of beetle attack were the most important variables for predicting postfire Douglas-fir mortality. As DBH increased, the predicted probability of mortality decreased for unattacked trees but increased for attacked trees. Field sampling suggested that ocular estimates of bark char may not be a reliable predictor of cambium injury. Our results emphasize the important role of Douglas-fir beetle in tree mortality patterns following fire, and the models offer improved prediction of Douglas-fir mortality for use in areas with or without Douglas-fir beetle populations.