Predisposition to bark beetle attack by root herbivores and associated pathogens: Roles in forest decline, gap formation, and persistence of endemic bark beetle populations Brian H. Aukema a,b , Jun Zhu c , Jesper Møller d , Jakob G. Rasmussen d , Kenneth F. Raffa e, * a Canadian Forest Service, Natural Resources, Prince George, British Columbia, Canada b Ecosystem Science and Management, University of Northern British Columbia, Prince George, British Columbia, Canada c Department of Statistics, University of Wisconsin – Madison, Madison, USA d Department of Mathematical Sciences, Aalborg University, Aalborg, Denmark e Department of Entomology, University of Wisconsin – Madison, Madison 53706, USA 1. Introduction The ability of bark beetles to undergo largescale outbreaks and transform landscapes is widely recognized, and has been the attention of intense study (Coulson, 1979; Romme et al., 1986; Veblen et al., 1994). However, the dynamics by which populations of some beetle species transition from noneruptive to eruptive conditions, and the extent to which these dynamics reflect within- stand versus migratory processes, are not well understood, and pose some fundamental questions about population dynamics and forest disturbance (Okland et al., 2005; Friedenberg et al., 2007). One approach to better understanding these dynamics, and the overall role of bark beetles in forest ecosystems, is through improved knowledge of tree-killing species that do not undergo Forest Ecology and Management 259 (2010) 374–382 ARTICLE INFO Article history: Received 19 August 2009 Received in revised form 15 October 2009 Accepted 25 October 2009 Keywords: Leptographium Dendroctonus Ips Hylobius Disturbance Population dynamics ABSTRACT Bark beetles are largely known for their ability to undergo intermittent population eruptions that transform entire landscapes and pose significant economic hardships. However, most species do not undergo outbreaks, and eruptive species usually exert only minor disturbances. Understanding the dynamics of tree-killing noneruptive species can provide insights into how beetles persist at low densities, and how some spatiotemporal patterns of host predisposition may more likely favor breaching eruptive thresholds than others. Elucidating mechanisms behind low-density populations is challenging, however, due to the requirement of long-term monitoring and high degrees of spatial and temporal covariance. We censused more than 2700 trees annually over 7 years, and at the end of 17 years, in a mature red pine plantation. Trees were measured for the presence of bark beetles and wood borers that breed within the primary stem, root weevils that breed in root collars, and bark beetles that breed in basal stems. We quantify the sequence of events that drive this decline syndrome, with the primary emergent pattern being an interaction between below- and above-ground herbivores and their fungal symbionts. This interaction results in an expanding forest gap, with subsequent colonization by early- successional vegetation. Spatial position strongly affects the likelihood of tree mortality. A red pine is initially very likely to avoid attack by tree-killing Ips beetles, but attack becomes increasingly likely as the belowground complex spreads to neighboring trees and eventually make trees susceptible. This system is largely internally driven, as there are strong gap edge, but not stand-edge, effects. Additional stressors, such as drought, can provide an intermittent source of susceptible trees to Ips beetles, and elevated temperature slightly accentuates this effect. New gaps can arise from such trees as they subsequently become epicenters for the full complex of organisms associated with this decline, but this is not common. As Ips populations rise, there is some element of positive feedback, in that the proportion of killed trees that were not first colonized by root organisms increases. This positive feedback is very weak, however, and we propose the slope between beetle population density and reliance on host stress as a quantitative distinction along a gradient from noneruptive through eruptive species. Almost all trees colonized by Ips were subsequently colonized by wood borers, likely a source of negative feedback. We discuss implications to our overall understanding of cross-scale interactions, between-guild interac- tions, forest declines, and eruptive thresholds. ß 2009 Elsevier B.V. All rights reserved. * Corresponding author. Tel.: +1 608 262 1125. E-mail address: raffa@entomology.wisc.edu (K.F. Raffa). Contents lists available at ScienceDirect Forest Ecology and Management journal homepage: www.elsevier.com/locate/foreco 0378-1127/$ – see front matter ß 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.foreco.2009.10.032