Temporal- and density-dependent impacts of an invasive plant on pollinators and pollination services to a native plant

Abstract

Pollinators and pollination services are under threat globally, and invasive plants have been implicated in their decline. Results of previous studies suggest that consequences of invasion for pollinators and plant-pollinator interactions are context specific. Investigating factors such as the density of an invasive plant and its phenology may provide a nuanced understanding of invasive species impacts. We conducted a 2-yr study in Montana to investigate how local pollinator abundance, richness, community composition, and visitation patterns varied with invasive Centaura stoebe density and phenology, and whether C. stoebe altered the reproduction of a co-flowering native plant, Heterotheca villosa, through changes in pollinator visitation. In an observational study, we found that during its peak bloom in August, Centaurea stoebe provided abundant floral resources to late-season pollinators. However, prior to C. stoebe bloom, native floral density and pollinator abundance and richness of these plots were lower compared to plots where C. stoebe was low or absent. Pollinator community composition in plots without C. stoebe was different compared to plots with C. stoebe (both high and low C. stoebe density), and these differences in pollinator composition strongly depended on the time of season. In an experimental study, we found that there was little evidence of competition between C. stoebe and H. villosa for pollinators at low relative densities of C. stoebe. Using experimental pollen supplementation, we observed no evidence of pollen limitation of seed set in H. villosa with increasing density of experimentally added C. stoebe. Our results suggest that the impact of an invasive plant on pollinators and plant-pollinator interactions depends on the relative density of the invasive plant and the timing of its bloom. Differences in pollinator visitation patterns over the growing season suggest that although C. stoebe provides abundant resources to lateseason pollinators, displacement of native plants at high C. stoebe density may indirectly harm pollinators that are active before C. stoebe blooms or that prefer native plants. Based on our results, restricting C. stoebe to low densities may help mitigate negative repercussions to native plant reproduction and may even be beneficial to some pollinators.