Testing the hypothesis of nest site limitation of wild bees using experimental and observational approaches

Reclaimed mined lands commonly persist as grasslands and can be rich in floral resources for bees. Nevertheless, the paucity of trees and shrubs may limit the suitability of reclaimed mine habitat for bee species that use woody nesting substrate.  Over four years, I surveyed the diversity and abundance of wild bees, floral resource abundance, and the availability various nesting substrates in 24 sites located on reclaimed mine land. I used these data to predict bee abundance and diversity. In these same sites, I experimentally augmented woody nesting substrate by providing artificial domiciles. I recorded the number of nests and the number of offspring produced by bees in the family Megachilidae. Although floral richness significantly predicted bee richness, nesting substrate variables were more important in predicting bee richness and abundance. The addition of artificial domiciles slightly, but not significantly, increased abundance of wood-nesting bees netted. This effect was greatest in sites located farthest from remnant forest patches. The number of nests and offspring produced by bees nesting within artificial domiciles increased over the four year period, but one exotic bee species dominated, suggesting that diversity of nesting substrates, not just availability, is important for bee diversity. The greater importance of nesting resources than floral resources suggests that wild bee populations are limited by nest site availability.  If nest sites are limiting, restoration of reclaimed mine lands with native flowers may provide only partial habitat for wild bees.  Attention to nesting substrate could enhance habitat quality for native bees. In the long run, however, artificial domiciles may do more harm than good for bee communities by promoting a prolific exotic species over native species.