Bark beetle populations phase between epidemic, outbreak levels, and low population density, endemic levels. The majority of scientific research is focused on outbreak populations because of the associated economic, ecological, and social impacts. Endemic populations are rarely studied but could provide information about the triggers that cause outbreaks. The goal of this thesis was to gain a better understanding of how endemic populations persist in a landscape through time by looking at the spatial distribution and susceptibility of host trees in southwestern US forested landscapes. To do this, I (1) analyzed 21 years of field data to examine the population dynamics of bark beetles and the factors that affect them, (2) created a statistical model for predicting the absolute risk of individual trees to bark beetle-cause mortality using tree, stand, and beetle pressure variables, and (3) simulated a forest landscape to develop a framework for applying tree-level risk assessments. In 1995, forty-five sites were established throughout the southwestern US to measure bark beetle activity and associated tree and stand characteristics. The plots were periodically revisited through 2012 resulting in over twenty years of bark beetle data with highly variable population densities over time and space. Site maximum dbh and the number of ponderosa pines per acre were significant (P <.029) for predicting the probability a rise in the population density of bark beetles. Tree, stand, and beetle pressure were significant (P < .001) in predicting the probability of beetle caused tree mortality per year. Using GIS, remote sensing, and ground truth data, a ponderosa pine forest was simulated with information about the size and configuration of trees in the landscape. This simulated landscape was used to develop a framework for tree-level risk assessments. The results are discussed further in the context of bark beetle management and further research opportunities. In 1995, forty-five sites were established throughout the southwestern US to measure bark beetle activity and associated tree and stand characteristics. The plots were periodically revisited through 2012 resulting in over twenty years of bark beetle data with highly variable population densities over time and space. Site maximum dbh and the number of ponderosa pines per acre were significant (P <.029) for predicting the probability a rise in the population density of bark beetles. Tree, stand, and beetle pressure were significant (P < .001) in predicting the probability of beetle caused tree mortality per year. Using GIS, remote sensing, and ground truth data, a ponderosa pine forest was simulated with information about the size and configuration of trees in the landscape. This simulated landscape was used to develop a framework for tree-level risk assessments. The results are discussed further in the context of bark beetle management and further research opportunities.
