Landscape and local correlates of bee abundance and species richness in urban community gardens

Background/Question/Methods

As cities become more populated and open spaces shrink, biodiversity and the ecosystem services biodiversity provides are threatened. Green spaces, like parks and community gardens, have the potential to preserve biodiversity as urban population densities increase, but this will strongly depend on the characteristics of those green spaces and the landscapes in which they are embedded. One taxonomic group critical for providing an essential ecosystem service (pollination) is bees. This study examines the role of local and landscape scale variables and how they influence bee abundance, species richness, and community composition in urban gardens. We investigated whether local and landscape characteristics are important correlates of bee abundance and species richness in urban community gardens. We sampled bees using aerial nets and pan traps, measured vegetation at 1, 20, and 100 meter scales, and used the USGS National Land Cover Database to classify the landscape surrounding our garden study sites at 500m, 1km, and 2km scales. We examined the correlations between >40 local and landscape variables and abundance and species richness of bees with conditional inference trees.

Results/Conclusions

We collected a total of 2566 individual bees, including 2215 females and 351 males. The most common species were Halictus tripartitus (39% of individuals), Apis mellifera (30%) and Bombus vosnesenskii (6%). Overall, we collected 55 morphospecies of bees. The genera with the highest number of morphospecies were Lasioglossum (15 morphospecies), Hylaeus (6), and Halictus (6). Bee abundance increased with increases in open space, agricultural land, natural habitat (all at 1 km scale) and open space at the 5km scale. Significant predictors of increasing bee abundance among local variables included number of ornamental, crop, and weed species, and decreases in concrete cover. Bee morphospeices richness increased with decreases in percent of ground covered by bare soil, and increases in non-woody vegetation cover, and the number of white flower - all local variables. Thus, both local and landscape factors are important for bee communities, but landscape factors appear to be more important drivers of bee abundance and local factors for bee richness in urban community gardens. We hope that this research will highlight the importance of urban green spaces in maintaining biodiversity in urban ecosystems.