The effects of impervious surface area, tree canopy cover, and floral richness on bee abundance, richness, and diversity across an urban landscape Eden Gerner Thesis submitted to the University of Ottawa in partial fulfillment of requirements for the Master of Science Department of Biology Faculty of Science University of Ottawa © Eden Gerner, Ottawa, Canada, 2020 Table of Contents Title Page........................................................................................................................................ i Table of Contents........................................................................................................................... ii List of Tables…………………………………………………………………………………………………………………………… iv List of Figures…………………………………………………………………………………………………………………….…….v Acknowledgements……………………………………………………………………………………………………………..…. vi Abstract………………………………………………………………………………………………………………………………..… vii Chapter 1: Overview Literature Review........................................................................................................................... 1 References……………………………………............................................................................................... 6 Chapter 2: Landscape and Local Impacts on Wild Bees Introduction………........................................................................................................................... 12 Materials & Methods.....................................................................................................................15 Site Description & Selection……………………………….............................................................15 Bee Collection…….. ………………………………………………………………………………………………..……17 Bee Identification………………..…………………………………………………………………………………..…. 18 Local Factors……………………….…………………………………………………………………………………..…. 20 Statistical Analysis………………………………………………………………………………………………..…….. 21 Results………………………………………………………………………………………………………………………………..….. 22 Bee Community Composition…………………………………………………………………………………..….22 Local Factors……………………….………………………………………………………………………………..……. 23 Multiple Regression Analyses …………………………………………………………………………………..…….23 Discussion……………………………........................................................................................................24 Bee Community Response to Landscape Factors…………………………………………………..…….25 Bee Community Response to Local Factors………………………………….………………………..…… 27 Scale of Effect………………………………………………………………………………………………………………27 Limitations & Future Research………..…………………………………………………………………………..29 Conclusion.…………………………………………………………..……………………………………………………. 30 References……………………………………………………………………………………………………………………………… 31 Chapter 3: Conclusion Conclusion…………………………………………………………………………………………………………………………….…44 References……………………………………………………………………………………………………………………………… 45 ii Tables & Figures……………………………………………………………………………………………………………………… 46 Table 2.1……………………………………………………………………………………………………………………… 46 Table 2.2……………………………………………………………………………………………………………………… 47 Table 2.3……………………………………………………………………………………………………………………… 49 Table 2.4……………………………………………………………………………………………………………………… 51 Table 2.5……………………………………………………………………………………………………………………….53 Table 2.6……………………………………………………………………………………………………………………….54 Table 2.7……………………………………………………………………………………………………………………….55 Table 2.8……………………………………………………………………………………………………………………… 56 Table 2.9…………………………………………………………………………………………………………………..…. 57 Figure 2.1…………………………………………………………………………………………………………………...…58 Figure 2.2…………………………………………………………………………………………………………………..… 59 Figure 2.3…………………………………………………………………………………………………………………..… 60 Figure 2.4………………………………………………………………………………………………………………….… 61 Figure 2.5……………………………………………………………………………………………………………………. 62 Figure 2.6……………………………………………………………………………………………………………………. 63 Figure 2.7……………………………………………………………………………………………………………………. 64 Appendix A: Site Volunteer Survey 95 Tables……………………………………………………………………………………………………………………………………… 65 Appendix B: Site Percentages of Landscape Gradients 98 Tables…………………………………………………………………………………………………………………………………..... 66 101 Appendix C: Floral Resources Tables…………………………………………………………………………………………………………………………………..… 69 References…………………………………………………………………………………………………………………………..…. 83 Appendix D: Tree Species Tables………………………………………………………………………………………………………………………………..…… 89. Appendix E: Bee Ecological Characteristics Introduction……………………………………………………………………………………………………………………………. 92 Materials & Methods……………………………………………………………………………………………………………… 93 Results………………………………………………………………………………………………………………………………….… 94 Discussion………………………………………………………………………………………………………………………….…… 95 References……………………………………………………………………………………………………………………………… 97 Tables…………………………………………………………………………………………………………………………………..…. 101 Appendix F: Site Nesting Characteristics Tables………………………………………………………………………………………………………………………………….…. 105 References….………………………………………………………………………………………………………………….………. 107 iii List of Tables Table 2.1. Sizes of the surveyed residential yards Table 2.2. All models comparing bee abundance against predictor variables. Models are in order of the lowest AIC value. Hyphens indicate that the variable was absent in the model. Values in brackets indicate standardized coefficients (Z scores) Table 2.3. All models comparing bee richness against predictor variables. Models are in order of the lowest AIC value. Hyphens indicate that the variable was absent in the model. Values in brackets indicate standardized coefficients (Z scores) Table 2.4. All models comparing bee diversity against predictor variables. Models are in order of the lowest AIC value. Hyphens indicate that the variable was absent in the model. Values in brackets indicate standardized coefficients (Z scores) Table 2.5. Pairwise correlations for predictor variables used in bee abundance and richness models Table 2.6. Pairwise correlations for predictor variables used in bee diversity models Table 2.7. VIF for models with R2 values within 10% of the top model for bee abundance Table 2.8. VIF for models with R2 values within 10% of the top model for bee richness Table 2.9. VIF for models with R2 values within 10% of the top model for bee diversity iv List of Figures Figure 2.1. The categories used for site selection, based on the two landscape gradients, percent impervious surface area and tree cover Figure 2.2. The model used in ArcGIS to obtain the levels of urbanization and tree cover at each site Figure 2.3. Correlation matrix for predictor variables used in bee abundance and richness models Figure 2.4. Correlation matrix for predictor variables used in bee diversity models Figure 2.5. Illustration of the steps used in the site selection process. A) Volunteer sites that met the criteria were evaluated on their standing within the two gradients. B) Selected sites and their standing within the two gradients. Figure 2.6. Map of the study region in Ottawa, Ontario, Canada. Sites shown are placed into four categories based on the surrounding percent urbanization and tree cover at 500m. Green colouring represents tree cover, while black colouring represents impervious surface area. Figure 2.7. R2 values plotted against landscape scale for bee abundance, richness, and diversity for the response gradients of A) impervious surface and B) tree cover v Acknowledgements I would like to thank my advisor, Dr. Risa Sargent, for her guidance, support, and numerous edits that helped me through this process. I would also like to thank my committee members Dr. Jeremy Kerr and Dr. Lenore Fahrig for their time and feedback. Thank you to Nick Stowe from the City of Ottawa, for providing me with the GIS data that this project is based on. A huge thank you to all my site volunteers (chosen or not), without whom, this research could not have been collected. Thank you to my field assistant Magean Ng who spent an inordinate amount of time counting and identifying flowers in backyards. I am also grateful to all of the Kharouba, Forrest, and Sargent lab members who provided feedback on my presentations and analysis. In particular, thank you to Emma Gaudreault, Lindsay Bennett, and Anneke Golemiec for their comradery and numerous edits to the various iterations of my thesis. Thank you to the Ottawa Field Naturalist Club and the Government of Canada (through Risa Sargent’s NSERC Discovery Grant) for funding my research. I thank my parents for their continued love and support as well as to my brother for sharing the coffee maker during our pandemic isolation. Finally, thank you to my partner, Gurjap Singh, for all of his editing, fieldwork assistance, and support throughout this project. vi Abstract As urbanization increases globally, habitat loss is increasing at an unprecedented rate, eroding the suitability of many landscapes for most forms of wildlife, including bees. At least some of this habitat loss is through the ongoing expansion of urban areas, a process termed ‘urbanization’. Studies of the effects of urbanization and urban land use on bees have reported a mixture of results, including some instances where at least some species appear to do better in urban areas than they do in lands surrounding urban areas. While the impacts of urbanization on bee communities has been investigated, tree canopy cover has been largely overlooked as a contributor to urban bee distributions, despite their potential importance as a predictor of bee activity. I investigated the impacts of urban land use and tree canopy cover on bee communities across a variety of neighbourhoods in a medium-sized Canadian city (Ottawa, Ontario). In total, I surveyed bee communities in 27 residential