COS 133-6
Urbanization homogenizes traits within and between bee communities

Friday, August 14, 2015: 9:50 AM
301, Baltimore Convention Center
Tina Harrison, Biology, University of Lousiana at Lafayette, Lafayette, LA
Rachael Winfree, Department of Ecology, Evolution, and Natural Resources, Rutgers University, New Brunswick, NJ
Background/Question/Methods

Trait homogenization occurs when anthropogenic change drives loss of trait diversity within communities, or increase of trait similarity between communities. Loss of trait diversity within communities is analogous to loss of species alpha diversity, while increase of trait similarity between communities is analogous to loss of species beta diversity. At either scale, trait homogenization is concerning because it may diminish ecosystem function and resilience to global change drivers such as land use change. To test whether land cover is associated with trait homogenization in native bees, we sampled bee communities at 36 sites divided among urban, agricultural, and forested landscapes. We matched observed species to 13 traits describing nest behavior, diet, body size and flight-season phenology. We first tested for within-community trait homogenization by using ANOVA to compare trait diversity across landscape types. We then tested for between-community trait homogenization by using a multivariate dispersion test to compare similarity of community weight mean (CWM) traits across landscape types. Both metrics of trait homogenization are weighted by species abundance, and are not directly influenced by species richness. Finally, to aid in interpretation, we compared species richness across landscape types using ANOVA.

Results/Conclusions

We captured 1717 bee specimens, which we identified to 108 species. Within communities, trait diversity was lower in agriculture and urban landscapes than in natural forest (ANOVA; F2,32=9.2, p=0.0007). However, only urban landscapes had a significantly more similar CWM trait composition between communities (multivariate dispersion test; F2,32 = 4.98, p = 0.011). Our results indicate that bee communities in agricultural landscapes are associated with reduced trait diversity, but retain natural levels of heterogeneity in trait identity across space. On the other hand, bee communities in urban landscapes are characterized by low trait diversity at both within- and between- community scales. Inspection of the CWM trait values revealed that urban bee communities are converging on smaller body size, higher percent eusociality, and longer and later flight season phenology. We did not find differences in species richness across landscape types. Therefore, we conclude that in the northeastern U.S., urban landscapes filter bee communities for a narrower set of traits that are nevertheless represented by a large number of species.