Co-flowering modularity and floral trait similarity help explain temporal changes in plant–pollinator network structure

Flowering phenology can be one of the most important factors mediating the temporal dynamics of plant–pollinator networks. However, most studies do not explicitly incorporate the effect of flowering phenology, which may bias conclusions about the structuring of plant–pollinator networks, obscuring our understanding of factors that explain the temporal variation of these networks. By analyzing co-flowering networks, floral traits similarity and pollinator sharing, in this study we aim to uncover the temporal dynamics of plant–pollinator network structure in two coastal communities. We recorded the flowering phenology of each plant species to construct co-flowering networks and the identity and frequency of floral visitors. We estimated and tested for differences in floral trait similarity and pollinator sharing across co-flowering modules. To disentangle the phenological effect of flowering on the structure of plant–pollinator networks, we constructed plant–pollinator subnetworks for each co-flowering modules and analyzed the role of the pollinators in each subnetwork. Floral trait similarity and pollinator sharing were related to changes in the structure of plant–pollinator networks, but these changes were community-dependent. The modular structure and network specialization index of plant–pollinator subnetworks were statistically persistent in both communities, suggesting the prevalence of specialized interactions throughout the flowering season. This result was consistent with the predominant peripheral role of most pollinator species across co-flowering modules in both communities. Our results highlight the importance of explicitly considering flowering phenology to advance our understanding of the mechanisms that explain temporal changes in the structure of plant–pollinator networks.