2018 ESA Annual Meeting (August 5 -- 10)

OOS 38-8 - Genetic evidence for complex dispersal patterns across urban landscapes in the Eastern carpenter bee (Xylocopa virginica)

Friday, August 10, 2018: 10:30 AM
348-349, New Orleans Ernest N. Morial Convention Center
Kimberly M. Ballare, Integrative Biology, University of Texas, Austin, TX and Shalene Jha, Integrative Biology, University of Texas at Austin, Austin, TX
Background/Question/Methods

Urbanization can fragment natural landscapes and impede dispersal of reproductive individuals. Limited dispersal can create isolated populations vulnerable to inbreeding and genetic drift. However, human-mediated dispersal of many organisms is likely common across urban areas. Xylocopa virginica are large bees found in both urban and rural areas across eastern North America, constructing their nests in dead wood, utilizing both natural and human-associated materials (ex. firewood). Xylocopa females are often nest-loyal, and so may not naturally disperse far from natal sites. However, both natural and human mediated long-distance dispersal could likely occur, related to their large body size and nesting habits. However, little is known about their nesting densities and dispersal patterns across urban landscapes. To better understand this, we amplified DNA from ~600 female X. virginica at 10 polymorphic microsatellites. Specimens were collected at 14 sites varying in land-use across the rapidly urbanizing I-35 corridor between Austin and Dallas in Central Texas. To estimate diversity and between-site gene-flow, we calculated Ne, G’st, and conducted analyses in STRUCTURE. We also calculated pairwise genetic resistance distances between sites, estimating how landscape affects movement by correlating genetic information with both map and local habitat resource data.

Results/Conclusions

Results suggest a complex pattern of X. virginica dispersal in our study area. First, STRUCTURE analyses reveal that the likely number of genetic populations is K=2, with individuals from more rural regions in one cluster, and individuals within Austin and Dallas clustering together. This pattern suggests that there exist two possible reproductive and dispersal processes, one facilitated by natural habitat and the other aided by human development. We also show that X. virginica populations located further apart were more genetically distinct, as we found significant isolation by distance between sampling locations (Mantel test, p<0.001). Additionally, we found that the most rural regions are the most genetically distinct, and differentiation values of ~0.20 (Gst) suggests relatively high levels of genetic differentiation among sites. This set of results may indicate that X. virginica exhibit shorter-range dispersal patterns or higher levels of nest fidelity in the more rural areas. Ongoing work explores the finer scale patterns of sibship and local relatedness to determine drivers of local site fidelity. As important pollinators of native and cultivated plant species, better understanding the dispersal patterns of X. virginica help us to understand how this and other related species can persist in rapidly urbanizing global environments.