Biotic homogenization in bee communities

Background/Question/Methods

Biotic homogenization is the process of historically distinct ecological communities becoming more similar due to human driven increases in widespread, disturbance-adapted species and/or decreases in sensitive endemic species. As weedy plants and animals dominate regions around the world, we will face important consequences for biodiversity, ecosystem function and evolution. Although land use change is known to be the primary driver of species loss, how land use causes biotic homogenization is not well studied. I will explore the role of LUC in biotic homogenization, using native bees as a model system. Bees are primary pollinators for most of the world’s plant species, yet there are few studies investigating how LUC affects bee community composition. Based on studies of other taxa and my preliminary analyses, we hypothesize that LUC is driving biotic homogenization of bee communities. We tested this hypothesis using a space-for-time substitution design with a large dataset of 130,000 spatially referenced bee specimens obtained through a USGS bee monitoring scheme.  We used GIS tools to quantify surrounding land at 1500m and used a robust k-means clustering algorithm to classify sites by land cover type. 

Results/Conclusions

Sites were strongly clustered into four groups characterized as forest/pasture, forest/wetland, developed and coastal.  Using ANOVA we found significantly different community richness and abundance among the four classified land use types (F=101.18, df=3, 21305, p<0.05). In particular, coastal sites had high abundance and diversity, suggesting these areas are potentially important for bee conservation. We also ordinated communities using non-metric multidimensional scaling, and found complex patterns of community similarity across different land use types.