Modelling the effects of the repellent scent marks of pollinators on their foraging efficiency and the plant-pollinator community.

Pollinator insects forage in complex and unpredictable resource landscapes, often using social information from congeneric individuals to acquire knowledge about their environment. It has long been recognized that this process allows them to exploit floral resources more efficiently and thus increase individual fitness. However, by creating correlations between the behaviors of pollinators within a population, this could also indirectly influence the entire plant-pollinator community. One type of social information used by pollinators is the scent mark left on the corolla of flowers by previous visitors, which can be used as a cue to avoid recently depleted resources. We developed a spatially explicit agent-based model to examine the effects, at both individual and community levels, of pollinators using these scent marks. The model simulates a population of pollinators foraging on flowers in a continuous 2D space in which we can vary the density of pollinators. We showed that the use of scent marks as a source of information significantly increased the foraging efficiency of pollinators except when competition between pollinators was very low. At the community level, this also resulted in a marked homogenization between floral resources within the landscape: in the absence of scent marks, the coefficient of variation of the remaining nectar quantity per flower strongly increased with greater pollinator competition, but it remained low at all levels of competition when scent marks were used by the pollinators. Finally, the use of scent marks markedly decreased the number of pollinator flower visits, especially at high levels of pollinator competition, which can potentially reduce the pollination service.