Phenology and density-dependent dispersal predict patterns of mountain pine beetle (Dendroctonus ponderosae) impact

For species with irruptive population behavior, dispersal is an important component of outbreak dynamics. We developed and parameterized a mechanistic model describing mountain pine beetle (Dendroctonus ponderosae Hopkins) population demographics and dispersal across a landscape. Model components include temperature-dependent phenology, host tree colonization determined by an Allee effect, and random-walk dispersal with motility conditioned by host tree density. The model was parameterized at a study site in central Idaho, United States (US), and evaluated at an independent site in northern Washington, US. Phloem and air temperatures, MPB spatial impact data from USDA Forest Service aerial detection surveys, and remotely sensed host tree density data were used to parameterize the model using a maximum likelihood approach. At both study sites the model was highly accurate (>84%) in predicting annual pattern formation when the model was re-initiated each year with the location of new patches of infested trees. Prediction of annual population growth at both sites was also good (>90%), although the model under-predicted area impacted at the Washington site, and at both sites was unable to predict initiation of new small patches. Our model extends previous research by providing a mechanistic description of the link between motility, dispersal and temperature-dependent MPB phenology.