Effect of meadow regeneration on bee (Hymenoptera: Apoidea) abundance and diversity in southern Ontario, Canada

To investigate how bee (Hymenoptera: Apoidea) communities repopulate habitat following severe disturbances, we compared assemblages in new, regenerating landfill sites planted in 2003, recent landfill sites planted in 2000, and control meadows relatively undisturbed for >40 years. All sites were identically sampled using pan traps and sweep netting, from early May to late September 2003, equalising collection effort among sites. In addition, we carried out five-minute aerial net samplings wherever sites contained large patches of wildflowers. We predicted that abundance and diversity of bees would be highest in recent sites and lowest in new sites. This prediction was partially supported: bees were most abundant in recent sites followed by control, then new sites, but species richness was highest (82 species) in recent sites, followed by new sites (67 species), then control (66 species). A randomisation analysis showed that there were more species than expected in new sites and fewer than expected in control sites. Differences in blossom availability likely explain differences in bee abundance and diversity among habitat regeneration levels. Overall, our results suggest that the bee community recolonised newly available sites immediately in the first year and that bee diversity and abundance increased for at least three years, subsequently declining between three and 40 years.