Contribution of traits and phylogenetic history to plant-pollinator network structure

Background/Question/Methods

Interaction webs define how the members of two or more trophic levels interact with one another. Different mutualistic interaction webs have strikingly consistent patterns, suggesting common mechanisms by which communities are built. Despite consistent patterns in structure, there is no consensus on what mediates network structure. Two general mechanisms, species traits and their phylogenetic history, can be important in mediating network structure. Traits mediate how species in communities interact with one another, and the resulting interaction structure of the community. There are few studies that have been able to use traits of both sides of a bipartite network to ask how traits contribute to network structure. A limitation of most studies exploring how traits relate to network structure is small dataset size: it is difficult to combine a large set of networks with trait data. Many species traits are phylogenetically conserved; closely related species tend to have similar traits. Understanding how phylogenetic signal of traits influences this relationship represents an important next step. We ask: How do phylogeny and traits interact to affect individual species network attributes, and community level network structures? 

Results/Conclusions

We found that species level network attributes, degree, interaction asymmetry, and a measure of specialization (Blüthgens d’), varied considerably across species. In addition, network level structures were quite variable across the 50 networks. Phylogenetic history was able to explain variation in species level and network level structures in only a small fraction of cases. Traits of plants were much better at explaining structures than pollinator traits. Overall, species traits were better at explaining species-level and network-level structures than phylogenetic history. Our findings suggest that understanding what drives network structure of plant-pollinator communities will require not only knowledge of phylogenetic history of the component species, but detailed measures of the species traits.