COS 35-9
Within-season temporal dynamics of plant-pollinator community interactions
Ecological communities are interconnected via complex networks of species interactions. Analyses of ecological networks have provided insight into the structural properties of communities and species interactions – especially for plants and pollinators. Despite evidence of strong temporal components, few studies have explicitly examined the temporal dynamics in plant-pollinator interaction networks within a season. In this study, we investigate the turnover of interactions using well-resolved, season-wide plant-pollinator interaction network from the Colorado Rocky Mountains. Specifically we explore three questions: (1) how do plant-pollinator interactions turnover across a season (from week to week), (2) what are the relative effects of species turnover and interaction rewiring on the turnover of interactions, and (3) what factors are associated with plant-pollinator interaction turnover? Plant-pollinator interactions and floral resource abundance were sampled weekly across the majority of the growing season in 2013 and 2014 at two subalpine meadows sites at the Rocky Mountain Biological Laboratory in Gothic, CO, USA (2,900m); in each season, this sampling occurred over a 3-month period from late May until early September (ca. 15 weeks, or 105 days).
Results/Conclusions
Turnover of plant-pollinator interactions was high and changed rapidly across the season in both years. Most interactions were ephemeral: 56% of interactions were present in only one week, and 75% lasted for two weeks or less. In contrast, the turnover of plant and pollinator species was much less rapid: most plant species were present for about a month, and most pollinator species were present for about three weeks. Across both years, interaction turnover decreased across the season, which was partly due to changes in plant and pollinator species composition. However, interaction rewiring did not vary directional across the season. In general, the relative effect of interaction rewiring on the turnover of interactions was greater than the relative effect of species turnover, and on average accounted for more than 50% of the changes in interactions from week to week. The richness of flowering plant species was strongly related to the amount of interaction rewiring. These results provide evidence that plant-pollinator interactions can exhibit strong temporal dynamics at fine temporal scales. This suggests that, at least in some communities, fine-scale temporal sampling is necessary to accurately capture the temporal dynamics of species interactions; without this, many interactions may be missed. These results also provide insight into the assembly and disassembly of plant-pollinator community interactions.