Landscape and local site variables differentially influence pollinators and pollination services in urban agricultural sites.

Urbanization has detrimental effects on biodiversity and ecosystem functioning, as agricultural and semi-natural habitats are converted into landscapes dominated by built features. Urban agricultural sites are a growing component of urban landscapes and have potential to serve as a source of biodiversity conservation and ecosystem service provisioning in urban areas. In 19 urban agricultural sites, we investigated how surrounding land cover and local site variables supported bees and pollination services. We found the abundance of bees differentially responded to landscape and local scale variables depending on body size and nesting habit. Large-bodied bees, Bombus and Apis species, were positively associated with increasing amounts of impervious cover, while the abundance of small-bodied soil nesting Halictus species increased as the proportion of flower area, a local variable, increased. Bee richness declined with increasing levels of impervious cover, while bee community composition changed along a gradient of increasing impervious cover. Pollination services, measured at each site using sentinel cucumber plants, declined as hardscape, a local variable, increased. To improve bee conservation and pollination services in urban agricultural sites, our results suggest urban planning strategies should minimize impervious cover at large spatial scales while land managers should focus locally on incorporating floral resources, which increases food and nesting resources especially for smaller bee species. Local site design coupled with regional urban planning can advance the success of urban agriculture, while benefiting biodiversity by creating opportunities for pollinator conservation in urban landscapes.

Modeling pollinator community response to contrasting bioenergy scenarios.

In the United States, policy initiatives aimed at increasing sources of renewable energy are advancing bioenergy production, especially in the Midwest region, where agricultural landscapes dominate. While policy directives are focused on renewable fuel production, biodiversity and ecosystem services will be impacted by the land-use changes required to meet production targets. Using data from field observations, we developed empirical models for predicting abundance, diversity, and community composition of flower-visiting bees based on land cover. We used these models to explore how bees might respond under two contrasting bioenergy scenarios: annual bioenergy crop production and perennial grassland bioenergy production. In the two scenarios, 600,000 ha of marginal annual crop land or marginal grassland were converted to perennial grassland or annual row crop bioenergy production, respectively. Model projections indicate that expansion of annual bioenergy crop production at this scale will reduce bee abundance by 0 to 71%, and bee diversity by 0 to 28%, depending on location. In contrast, converting annual crops on marginal soil to perennial grasslands could increase bee abundance from 0 to 600% and increase bee diversity between 0 and 53%. Our analysis of bee community composition suggested a similar pattern, with bee communities becoming less diverse under annual bioenergy crop production, whereas bee composition transitioned towards a more diverse community dominated by wild bees under perennial bioenergy crop production. Models, like those employed here, suggest that bioenergy policies have important consequences for pollinator conservation.

Floral diversity increases beneficial arthropod richness and decreases variability in arthropod community composition.

Declines in species diversity resulting from anthropogenic alterations of the environment heighten the need to develop management strategies that conserve species and ecosystem services. This study examined how native plant species and their diversity influence the abundance and richness of beneficial arthropods, a functionally important group that provides ecosystem services such as pollination and natural pest suppression. Beneficial arthropods were sampled in replicated study plots containing native perennials planted in one-, two-, and seven-species mixtures. We found plant diversity had a positive impact on arthropod richness but not on arthropod abundance. An analysis of arthropod community composition revealed that each flower species attracted a different assemblage of beneficial arthropods. In addition, the full seven-species mixture also attracted a distinct arthropod community compared to single-species monocultures. Using a multivariate approach, we determined whether arthropod assemblages in two- and seven-species plots were additive and could be predicted based on assemblages from their component single-species plots. On average, assemblages in diverse plots were nonadditive when compared to assemblages predicted using single-species plots. Arthropod assemblages in two-species plots most closely resembled those of only one of the flower species in the mixture. However, the arthropod assemblages in seven-species plots, although statistically deviating from the expectation of an additive model, more closely resembled predicted communities compared to the assemblages found in two-species plots, suggesting that variability in arthropod community composition decreased as planting diversity increased. Our study demonstrates that careful selection of plants in managed landscapes can augment beneficial arthropod richness and support a more predictable arthropod community, suggesting that planning and design efforts could shape arthropod assemblages in natural as well as managed landscapes to meet targeted conservation or management goals.