Early stage invasion of a flowering shrub fails to disrupt native plant-pollinator interactions

Background/Question/Methods

Biological invasions can be highly disruptive to native communities through direct competition and predation, or indirectly by altering existing abiotic environments or biotic interactions. Plant-pollinator interactions are an example of a critical ecosystem service that has suffered detrimental effects from exotic plant invasions. The majority of past research, however, has focused on single pairs of exotic and native plants out of their community context. In addition, the influence of early-stage invasions on native plant-pollinator interactions is little studied, but could provide valuable insight into the transition of species from establishment to becoming widespread and dominant. We experimentally manipulated the presence of an exotic woody shrub, Rosa multiflora, to investigate its effects on the composition, richness, and visitation rate of pollinators to co-flowering native species at early invasion levels (low relative floral density). To further explore our results, we compared the pollination biology of R. multiflora at early invasion levels to its known potential to disrupt plant-pollinator interactions at high invasion levels from the previous year. We also considered diet preferences of pollinators in the native community to better understand which pollinators were more likely to visit an exotic plant.

Results/Conclusions

We found that at low to moderate levels of invasion, the presence of Rosa multiflora had little effect on the pollinator composition, richness, and visitation rate to co-flowering native plants. R. multiflora shared a large proportion of pollinators with co-flowering natives, and was visited by pollinators with a wider diet breadth than expected by chance. Compared to co-flowering natives, R. multiflora was visited at the lowest rate by the lowest richness of pollinators. At early invasion levels, the pollinator richness and visitation rate of R. multiflora was less than 50% of its observed rates at high densities from the previous year. Together, these results suggest that at low to moderate levels of invasion, the exotic shrub R. multiflora was less attractive to existing pollinators than co-flowering natives. In addition, generalist pollinators readily included R. multiflora in their diet without detriment to the pollination of native plants in the community.