Pollination services are responsible for the reproduction of the vast majority of plant species worldwide, and species conservation likely requires maintenance of these essential ecosystem services. Although ecologists have a solid understanding of the species-area relationship and its importance for conservation and land management, little is known about how and why changes in land area influence the richness, structure, and function of species interactions. To address this critical gap, we investigated plant-pollinator interaction networks in Ozark glades of different areas. We explored whether the scaling of species interactions with area was predictable by a null model based solely on changes in the abundances of individuals, and thus changes in species richness, as well as the contribution of biological mechanisms to this pattern.
Results/Conclusions
We found that interaction richness increased more steeply with area than did species richness, explained partially by the null model. Two biological mechanisms that could further explain this pattern are the presence of a greater richness of generalist species and/or particularly generalist pollinator behavior in larger glades. We eliminated the first biological mechanism because larger glades contained more species of diet specialists, while smaller glades contained nested subsets of species present in larger glades, including primarily the common, generalist species. Instead, we found particularly generalist pollinator behavior across species in larger glades. The proportion of possible interactions realized (the proportion of known food plants visited by pollinators) increased with area. Smaller glades failed to support their interaction potential because pollinators exhibited disproportionately restricted diets in small glades. Importantly, the increased generalist foraging behavior of pollinators in large glades represents the summed population-level resource use of the species, while individual pollinators did not display the full range of diet breadth exhibited at the species level. It is this individual behavior which is important for pollination service and plant reproduction. We concluded that land area can affect the structure and function of plant-pollinator interaction networks, with implications for long-term biodiversity maintenance.