Antagonisms, mutualisms and commensalisms affect outbreak dynamics of the southern pine beetle.

Feedback from community interactions involving mutualisms are a rarely explored mechanism for generating complex population dynamics. We examined the effects of two linked mutualisms on the population dynamics of a beetle that exhibits outbreak dynamics. One mutualism involves an obligate association between the bark beetle, Dendroctonus frontalis and two mycangial fungi. The second mutualism involves Tarsonemus mites that are phoretic on D. frontalis ("commensal"), and a blue-staining fungus, Ophiostoma minus. The presence of O. minus reduces beetle larval survival ("antagonistic") by outcompeting beetle-mutualistic fungi within trees yet supports mite populations by acting as a nutritional mutualist. These linked interactions potentially create an interaction system with the form of an endogenous negative feedback loop. We address four hypotheses: (1) Direct negative feedback: Beetles directly increase the abundance of O. minus, which reduces per capita reproduction of beetles. (2) Indirect negative feedback: Beetles indirectly increase mite abundance, which increases O. minus, which decreases beetle reproduction. (3) The effect of O. minus on beetles depends on mites, but mite abundance is independent of beetle abundance. (4) The effect of O. minus on beetles is independent of beetle and mite abundance. High Tarsonemus and O. minus abundances were strongly correlated with the decline and eventual local extinction of beetle populations. Manipulation experiments revealed strong negative effects of O. minus on beetles, but falsified the hypothesis that horizontal transmission of O. minus generates negative feedback. Surveys of beetle populations revealed that reproductive rates of Tarsonemus, O. minus, and beetles covaried in a manner consistent with strong indirect interactions between organisms. Co-occurrence of mutualisms embedded within a community may have stabilizing effects if both mutualisms limit each other. However, delays and/or non-linearities in the interaction systems may result in large population fluctuations.