Chapter Four Interactions among conifer terpenoids and bark beetles across multiple levels of scale: An attempt to understand links between population patterns and physiological processes

Transferring information from specific components of a plant-herbivore interaction to population and landscape-level impacts poses a major challenge to ecologists. Bark beetle-conifer-microbial interactions comprise a valuable model for addressing this issue, because host plant compounds are known to affect multiple components of these relationships. In particular, terpenoids play important roles in host acceptance, beetle aggregation, host defense, establishment of microbial symbionts, exposure to and avoidance of predators, and other functions. Some bark beetle species undergo dramatic population eruptions in which they convert from relatively stable to outbreak dynamics. These eruptions both play major roles in ecosystem processes, and pose significant economic and natural resource management challenges. A wealth of information has been developed for each individual component of bark beetle-fungal-conifer interactions. However, we have limited ability to scale across multiple layers of biological organization, which is essential for an integrated understanding of the system and for judicious management decisions. We propose that focusing on one group of compounds that plays an important role at each stage of colonization, and whose effects are density-dependent, can provide a useful approach to achieving integration. We also identify biological thresholds, whose outcomes are qualitative but whose determinant inputs are quantitative, as a major challenge to both mechanistic and landscape approaches, and which need to be addressed in an integrated fashion. Based on these analyses, it appears that linkages among plant defense physiology, individual host acceptance decisions, cooperative behavior, and beetle density can constrain or generate eruptions in a fashion consistent with bimodal equilibria theory, including Allee effects. Moreover, chemically mediated interactions with predators and competiors can constrain these eruptions, but their ability to do so may be linked to the spatial and temporal distribution of agents compromising tree defenses, which in turn both reflects and contributes to host selection behavior. A narrow set of host, climatic, and natural enemy conditions, and distribution patterns of each, is needed to release populations to eruptive levels. Our specific conclusions are: 1) Individual compounds can affect interactions across multiple levels of scale, from molecular throught landscape; 2) At each level of scale, the same compounds can be sources of both positive and negative feedback. Their interactions across scales can be amplified or buffered, depending on these feedback processes; 3) Host selection behavior can be an important link between physiological and population processes, particularly where responses to phytochemicals are plastic; 4) Tritrophic interactions mediated by chemical cues can be either important or ineffective constraints on eruptive behavior, depending on how prey are spatially and temporally distributed, which in turn reflects their host selection behavior; 5) Each level of scale is characterized by thresholds, whose qualitative outcome is determined by quantitative factors. Based on these conclusions, we identify two areas in particular need of future research: 1) Chemically informed, spatially explicit studies on interactions among natural enemies, population dynamics, and predisposing agents that affect host tree chemistry and physiology, can improve both our understanding of linkages across multiple trophic levels, and how single chemical groups function at multiple levels of scale; 2) Integrated studies incorporating landscape and mechanistic approaches are needed to bridge our understanding of pattern and process.