A demographic approach to evaluating the impact of stressors on bumble bee colonies

1. Natural and anthropogenic stressors threaten the sustainability of bumble bees and evaluating their impact is essential to the stewardship of these valuable pollinators. Demographic modelling provides a framework for testing hypotheses about the impacts of stressors, but it has not previously been applied to bumble bees. 2. I therefore formulated a demographic model for a bumble bee colony and then quantified the impact of two stressors, pesticide exposure and spider predation, by perturbing it with their known effects. 3. By simulating a laboratory exposure of Bombus terrestris L. to dietary imidacloprid (a neonicotinoid insecticide), I tested whether the observed colony declinewas explained solely by a toxic effect on the fecundity of the foundress queen. By simulating field observations of B. terricola Kirby, I tested whether predation by crab spiders reduced colony fitness sufficiently to provide an adaptive explanation for avoidance behaviours seen when bumble bees encounter spiders. 4. In B. terrestris, a dose-appropriate decrease in fecundity predicted the observed colony decline, which implicates this as a principal mechanism of toxicity. In B. terricola, doubling the rate of spider predation reduced a colony's production of new queens by 11%, which implies that spider avoidance is highly adaptive. 5. These analyses illustrate the utility of demographic modelling for quantifying the impacts of stressors on bumble bees. In the future, models of this type could be used to investigate a wider range of stressors and to produce thereby knowledge and tools useful for safeguarding bumble bees and the pollination services that they provide.