Pollinator-mediated interactions between plants can play an important role for the dynamics of plant communities. Pollination services depend on the abundance and the foraging behaviour of pollinators, which in turn respond to the availability and distribution of floral resources (notably nectar sugar). However, it is still insufficiently understood how the ‘sugar landscapes’ provided by flowering plant communities shape pollinator-mediated interactions between multiple plant species and across different spatial scales. A better understanding of pollinator-mediated interactions requires an integrative approach that quantifies different aspects of sugar landscapes and investigates their relative importance for pollinator behaviour and plant reproductive success. In this study, we quantified such sugar landscapes from individual-based maps of Protea shrub communities in the Cape Floristic Region, South Africa. The 27 study sites of 4 ha each jointly comprise 127,993 individuals of 19 species. We analysed how rates of visitation by key bird pollinators and the seed set of plants respond to different aspects of sugar landscapes: the distribution of nectar sugar amounts, as well as their quality, taxonomic purity and phenology.
Results/Conclusions
We found that interactions between Protea species and pollinators depend on different spatial and temporal scales, can either be positive or negative. Pollinator visitation rates strongly depended on phenological variation of site-scale sugar amounts. The seed set of focal plants increased with nectar sugar amounts of conspecific neighbours and with site-scale sugar amounts. Seed set increased particularly strongly if site-scale sugar amounts were provided by plants that offer less sugar per inflorescence. These combined effects of the amount, quality, purity and phenological variation of nectar sugar show that nectar sugar is a common interaction currency that determines how multiple plant species interact via shared pollinators. The responses of pollinator-mediated interactions to different aspects of this interaction currency alter conditions for species coexistence in Protea communities and may cause community-level Allee effects that promote extinction cascades. The identification of such interaction currencies is crucial both for developing a more general understanding of community dynamics and for predicting community dynamics in changing environments. It is timely to test whether resource landscapes play similar roles in other ecosystems such as agricultural systems where mass flowering field crops (MFCs) rely on pollination services and pollinators on floral resource availability of MFCs.