Conclusions

Concerns about the ecological consequences of changes in bee biodiversity are leading to increased recognition of the importance of wild bee conservation and promoting wild bees in agricultural systems (Isaacs and Kirk, 2010, Biddinger et al., 2018, Reilly et al. 2020). But wild bee communities are diverse and dynamic, and little is known about what species or groups have the greatest conservation needs. Our intensive sampling across six years shows that bee communities vary greatly from month to month for all measures of biodiversity. For monitoring efforts to capture the full breadth of bee biodiversity, it is important to sample bees across all seasons, especially during spring when communities turnover rapidly. Biodiversity changes across the six years of our study were less dramatic than seasonal changes, but we found evidence of biodiversity loss over time with declines in abundance in 33% of the species. Notably, our results indicate that bee family is not a good predictor of changes over time, and in some cases, closely related species showed very different patterns. This suggests that species-specific traits may be more important predictors of long-term population dynamics than shared evolutionary history. We recommend that future monitoring efforts that seek to understand species-level dynamics for multiple co-occurring species, and whole-community patterns in biodiversity, utilize standardized collection methods repeated over multiple years. Longer-term and species-specific data on wild bee dynamics could provide greater insight into which species need targeted conservation efforts (Woodard et al., 2020).

Acknowledgments

We thank the orchard fruit grower Scott Slaybaugh, Jim Lerew, Bill Pulig, and Barry Rice for allowing us to conduct studies on their land and to Jim Gillis and Mace Vaughan for their support in establishing the USDA-NRCS wildflower strips. Thank you to Kathryn Wholaver, Lolita Miller, Sarah Heller, and many others at the Pennsylvania State Fruit Research and Extension Center. Jason Gibbs, Robert Jean, and Sam Droege identified bees and the members of the López-Uribe Lab provided helpful feedback on the previous versions of the manuscript. This research was supported by a Pennsylvania Department of Agriculture Grant number C940000555, and USDA-NIFA-AFRI Specialty Crop Research Initiative, Project 2012-51181-20105. DJB was funded through the USDA NIFA Appropriations under Project PEN04620. MML-U was funded through the USDA NIFA Appropriations under Projects PEN04716 and PEN04620.

Conflicts of interest

The authors declare no conflicts of interest.

Data availability statement

The data used in this study will be posted on Dryad after the paper is accepted for publication.

Author contributions

Nash E. Turley: Formal analysis (lead); Visualization (lead); Writing – original draft (lead); Writing – review & editing (equal)
David J. Biddinger: Conceptualization (equal); Funding acquisition (equal); Investigation (lead); Methodology (equal); Writing – review & editing (equal)
Neelendra K. Joshi: Conceptualization (equal); Investigation (equal); Methodology (equal); Writing – review & editing (equal)
Margarita M. López-Uribe: Conceptualization (equal), Funding acquisition (lead); Writing – review & editing (lead)