Page Intentionally Left Blank

Page intentionally left blank

AN ABSTRACT OF THE THESIS OF

Scott Richard Mitchell for the degree of Master of Science in Wildlife Science presented on March 5, 2020

Title: Impacts of Range Management Decisions on Native Pollinators: Innovative Grazing Practices and Riparian Restoration

Abstract approved: ______Sandra J. DeBano

Abstract Animal pollination is critical to plant reproduction in agricultural and wildland ecosystems. Much of the production of seeds and fruits in natural areas, which underlie many food webs, depends on pollination services by insects. The taxon responsible for delivering the bulk of these services in most temperate systems is bees. While colony collapse disorder in the nonnative European honey bee (Apis mellifera) is a significant concern and one that has generated much media and scientific interest, recent studies have indicated that native, unmanaged bees may be declining as well. To increase the likelihood of continued delivery of pollination services, land managers need to have a comprehensive understanding of how management actions may affect native bees. One dominant land type, especially in the Pacific Northwest of the United States (US), is rangelands. Rangelands are known to support diverse pollinator communities and face land management challenges such as multiple-use and historic degradation. My thesis examines how one potentially important stressor, livestock grazing, affects native bee and plant communities and how shrubs used in the restoration of riparian areas common in rangelands can provide resources to a diverse community of native bees.

My first chapter provides a general introduction to the topic of native pollinators and a broad overview of the importance of pollinators, their basic habitat

and biological needs, and some of the stressors that may be affecting populations. The purpose of this chapter is to provide context and background knowledge for the chapters that follow.

Chapter two focuses on one potentially important stressor of native bee communities in the American West – livestock grazing. In this chapter, I examine the effects of, late-season, moderate intensity, rotational cattle grazing on native bees and the blooming plants they depend on. To understand the effects of grazing on bee and plant communities, we conducted bee and blooming plant surveys at 28 sites located at two eastern Oregon study locations in the summer of 2018. One of the locations was a riparian wet-meadow system and the other was a bunchgrass prairie system. At each location, half of the sites were grazed by cattle at some point in the summer and half were not. Bee and plant communities at both locations varied through the growing season, with peaks in species richness and diversity occurring early to mid- season. We found location specific effects of cattle grazing on bee and blooming plant communities. At the riparian meadow location, we found that cattle grazing had short-term effects in reducing bloom abundance, species richness, and Shannon diversity in July and August, but these effects did not translate to any significant effects on bee communities. At the bunchgrass prairie location, we found no significant short-term effects of grazing on blooming plant communities and no negative effects on native bees. In fact, we detected higher bee abundance and richness in grazed sites at this location. An analysis of longer term-grazing at the second location revealed no detectable effect of grazing on blooming plant or native bee communities.

Chapter three focuses on the interactions of native bees and flowering plants in a restored riparian area. To understand how bees interact with flowering plants, we conducted extensive hand-net surveys of bees in 2018 and 2019 from April to September in the restored area. To understand the changes in blooming plant community richness throughout the season, we conducted plant richness surveys throughout the 2018 growing season. We found a diverse community of bees that interacted with a diverse community of blooming plants. Early in the season (April), we found that blooming forbs were significantly more abundant and species rich than

blooming shrubs. While forbs were more abundant in the early-season, we found no evidence that bees foraged on forbs at a higher rate than shrubs. Willow seemed to be important in supporting several apparent specialist bee foragers in the early-season. Later in the season, both shrubs and forbs remained important resources for bees, and we found bees that were apparent specialists on both shrub and forb blooms.

Chapter four summarizes the key findings of the previous two chapters and provides some basic recommendations for land management with objectives related to pollinator health. I also make some suggestions for future research directions that could build on the findings described in this thesis. Collectively, the results of this thesis provide hope for pollinator conservation in inland Pacific Northwest grasslands. The second chapter indicates that late-season, rotational, cattle grazing may help mitigate some of the negative effects of cattle grazing on native bees that have been observed in some systems. The third chapter indicates that riparian restoration activities, especially planting blooming shrubs, may provide important forage resources for native bees in Pacific Northwest rangelands.

Impacts of Range Management Decisions on Native Pollinators: Innovative Grazing Practices and Riparian Restoration

by Scott Richard Mitchell

A THESIS

submitted to

Oregon State University

in partial fulfillment of the requirements for the degree of

Master of Science

Presented March 5, 2020 Commencement June 2020

Master of Science thesis of Scott Richard Mitchell presented on March 5, 2020

APPROVED:

Major Professor, representing Wildlife Science

Head of the Department of Fisheries and Wildlife

Dean of the Graduate School

I understand that my thesis will become part of the permanent collection of Oregon State University libraries. My signature below authorizes release of my thesis to any reader upon request.

Scott Richard Mitchell, Author

ACKNOWLEDGEMENTS

I would like to thank all the incredible people and organizations that contributed to this project and its successful completion, both directly and through their support. First, I would like to thank my major professor, Dr. Sandy DeBano for all of her support, advice, encouragement, help in the field, and commitment to her students. Without Sandy this project would not have been possible, and I feel incredibly grateful to have had the opportunity to work with her for the past several years. My other committee members, Dr. Dana Sanchez and Dr. Gail Langellotto, both provided helpful suggestions and edits throughout the planning of this project and writing of this thesis, and always had useful advice. Critical to the completion of this work was funding from multiple sources, including a grant from the Foundation for Food and Agriculture (Research Grant #549031) that supported my NIFA Fellowship, the OSU Branch Experiment Station Internship program, the Oregon Department of Forestry and the United States Forest Service. Second, I would not have been able to complete this thesis and my degree without the support of the faculty members and staff at Oregon State University. Faculty and staff at the Hermiston Agricultural Research and Extension Center in Hermiston, and those at the main Corvallis campus contributed immensely to my ability to complete this project. I would especially like to thank Dr. Raymond Malewitz, at the Corvallis campus for his edits and tips on writing and the writing process as I began writing this thesis. The staff and researchers at The Nature Conservancy and the United States Forest Service Pacific Northwest Research Station were all critical to the completion of this project. In particular Mary M. Rowland and Heidi Schmalz helped immensely with the logistics of the project, some field work, and identification of some challenging plant species. Josh Averett was another person who contributed substantially to this project and taught me how to identify willow species in Starkey. Josh also contributed important feedback to the design and analysis of this work. I would also like to thank Skyler Burrows for all the work he

did identifying all the bee specimens that were collected during this study and his responsiveness to all my questions. I would like to especially thank the field and laboratory assistants who helped us accomplish the monumental task of collecting and preparing tens of thousands of bees—washing, sorting, and pinning them. Special thanks to James McKnight, Marisa McCaskey, and Coltyn Kidd for all their hard work in both the field and the laboratory. I would also like to thank all of the other graduate students, both in the Department of Fisheries and Wildlife and those in the Pollinator Reading Group for providing friendship, support, and feedback on my work throughout the project. In particular, my fellow bee researcher and lab-mate Katie Arstingstall who helped with field work related to this project and helped supervise the interns who worked for us. Last, I would like to thank my family for always encouraging me and providing me with a rich upbringing that inspired me to pursue a career in science. My brothers and parents always encouraged my interest in nature and are probably the reason I decided I wanted to study bugs, flowers, plants, and ecology. I would also like to thank my girlfriend, Kenzie Fleischman, for her unwavering support in my pursuit of my degree. She has been a critical part of my life for most of my adult life and endured multiple years of long-distance, many late nights, and a move across state-lines to support my goals and aspirations.

CONTRIBUTION OF AUTHORS

Sandra J. DeBano contributed to all parts of this thesis including study design, implementation, data analysis, and writing in all chapters. Skyler Burrows identified all bee specimens collected in both studies. Mary M. Rowland contributed to the design and implementation of studies described in both chapters. Lesley R. Morris and Scott B. Lukas contributed to the design and implementation of the study described in the second chapter. Heidi Schmalz contributed to the study design and implementation and helped with plant identification at the Zumwalt Prairie Preserve.

TABLE OF CONTENTS Page

CHAPTER 1: GENERAL INTRODUCTION ...... 1

Literature Cited ...... 4

CHAPTER 2: LATE-SEASON LIVESTOCK GRAZING: A SUSTAINABLE STRATEGY FOR NATIVE BEES IN GRASSLANDS OF THE INTERIOR PACIFIC NORTHWEST, USA? ...... 8

Abstract ...... 8 Introduction ...... 9 Methods ...... 12 Study Areas: The Nature Conservancy’s Zumwalt Prairie Preserve ...... 12 Study Areas: The United States Forest Service Starkey Experimental Forest and Range ...... 13 Plant Community Sampling ...... 14 Bee Community Sampling ...... 15 Statistical Analyses ...... 15 Abundance, Richness, and Diversity Responses ...... 15 Blooming Plant and Bee Community Analyses ...... 16 Results ...... 17 General Results ...... 17 Seasonal Variation in Plant and Bee Communities ...... 17 Abundance, Richness and Diversity Responses ...... 18 Plant and Bee Community Composition Relative to Season, Location, and Treatment ...... 19 Discussion ...... 20 Literature Cited ...... 22 Figures and Tables ...... 29

CHAPTER 3: FEED THE BEES AND SHADE THE STREAMS: FLOWERING SHRUBS PLANTED FOR RIPARIAN RESTORATION PROVIDE FORAGE FOR BEES ...... 39

Abstract ...... 39

Introduction ...... 40 Methods ...... 43 Study Area ...... 43 Plant and Bee Sampling ...... 43 Statistical Analyses ...... 45 Early-season Communities ...... 45 Seasonal Phenology ...... 45 Bee and Plant Interactions ...... 46 Results ...... 47 Early-season Communities ...... 47 Seasonal Phenology ...... 48 Bee and Plant Interactions ...... 48 Discussion ...... 49 Literature Cited ...... 52 Figures and Tables: ...... 58

CHAPTER 4: GENERAL CONCLUSIONS...... 67

General Bibliography ...... 70

APPENDIX 1: Additional Figures and Tables for Chapter 2 ...... 82

APPENDIX 2: Additional Figures and Tables for Chapter 3 ...... 95

LIST OF FIGURES Figure Page

2.1: Study locations and sampling set-up ...... 29

2.2: Timeline of sampling events and cattle grazing events ...... 30

2.3: Seasonal variability in bees and plants ...... 31

2.4: Short-term effects of cattle grazing ...... 32

2.5: Rarefied bee richness of Zumwalt sites in July ...... 33

2.6: NMS ordinations of bee and plant communities ...... 34

3.1: Study sites and native bees ...... 58

3.2: Seasonal variation in forb and shrub richness ...... 59

3.3: Average bloom abundance, species richness, and # bees/flower ...... 60

3.4: Bloom proportions and bee visitations for April 2018 ...... 61

3.5: Blooming plant species in bee community space ...... 62

3.6: Season-long network of bee and plant interactions ...... 63

3.7: Shannon diversity of bee visitors to shrubs at Meadow Creek...... 64

LIST OF TABLES Table Page

2.1: Locations by the numbers ...... 35

2.2: Short-term effects of grazing on bee and plant communities ...... 36

2.3: Long-term grazing effects on bee and plant communities ...... 37

2.4: MRPP Analyses Results ...... 38

3.1: Phenology of shrub blooms. Numbers of bees collected on each of 16 shrub species blooming in April, May, June, and July of 2018 and 2019...... 65

3.2 Pearson correlations for plant ordinations ...... 66

LIST OF APPENDIX FIGURES Figure Page

A2.1: Pie chart of bee visits to flowers ...... 103

LIST OF APPENDIX TABLES Table Page

A1.1: Complete bee species list and occurrence at each location and reference for identification...... 83

A1.2: Pearson correlations for bee ordination ...... 88

A1.3: Pearson correlations for plant ordination ...... 89

A1.4: Relative proportion of taxa making up 90% of all bees collected each month at the Zumwalt Prairie in 2018...... 90

A1.5: Relative proportion of taxa making up 90% of all bees collected each month at Starkey in 2018 ...... 91

A1.6: Relative proportion of species making up 90% of blooms counted each month at the Zumwalt Prairie in 2018...... 92

A1.7: Relative proportion of species making up 90% of counted blooms collected each month at Starkey in 2018...... 93

A2.1: Complete list of bee species collected over the two sampling years...... 96

CHAPTER 1: GENERAL INTRODUCTION

Globally, more than three-quarters of flowering plants, including crops and wild plants, rely on some form of insect pollination (Ollerton et al. 2011). While the bulk of economically relevant pollination services in the US are delivered by the non- native, managed European honey bee (Apis mellifera), the economic value provided by wild pollinators alone is estimated to be $3.07 billion dollars (Crane & Walker 1984; Losey & Vaughan 2006; Klein et al. 2007). Declines in pollinator populations and the risk of potential economic impacts from the loss of pollination services has been recognized for over 20 years (Kearns et al. 1998). The recognition of a need for basic research and implementation of pollinator habitat restoration came soon after (Kremen & Ricketts 2000). More recently, observed declines in wild bee and invertebrate populations (Potts et al. 2010; Colla et al. 2012; Hallmann et al. 2017) have resulted in continued interest in managing for wild pollinators and their habitat. Additionally, colony collapse disorder in honey bees, which has resulted in annual losses of 30-40% of private and commercial hives (LeBuhn et al. 2013; Lee et al. 2015; Kulhanek et al. 2017), has sparked interest in managing for healthy wild pollinators to increase the stability of pollination services. With over 20,000 bee species worldwide and over 4,000 in the US alone (Wilson & Carril 2015), remarkably little is known about many bee populations and there is a need to characterize species diversity, abundance, and population trends (LeBuhn et al. 2013). To effectively manage for healthy populations of wild pollinators, it is increasingly important to understand environmental processes and habitat features that can help maintain wild bee populations. Many factors impact pollinator populations around the world including habitat loss from agricultural intensification and urbanization, pesticide application, and issues with diseases and pests (Le Féon et al. 2010; Potts et al. 2010; Vanbergen & Insect Pollinator Initiative 2013; Weiner et al. 2014). While agricultural intensification may be responsible for declines in some pollinator populations, many of these systems depend on pollination services of honey bees or wild bees. A pressing research need is investigating ways in which agricultural practices could be

changed to facilitate sustainable pollinator populations (Kovács-Hostyánszki et al. 2017). In addition to farming, livestock production can also affect native bee habitat, including in Oregon rangelands (Kimoto et al. 2012b). Oregon has a significant ranching industry and each year several million cattle are produced, making cattle one of Oregon’s top commodities (Walker 2017). Cattle grazing occurs on the majority of public land, including in National Forests, and effective management of rangelands should include managing for pollinator health (Black et al. 2011). While grazing represents a potentially important stressor on blooming plant and bee communities, restoration projects represent potential opportunities for improving pollinator habitat (Kremen & Ricketts 2000). In areas that have been heavily grazed or otherwise degraded, range managers may pursue habitat restoration projects, especially in sensitive habitats like riparian zones. Many shrub species (e.g. willows (Salix spp.), snowberry (Symphoricarpos albus), hawthorn (Crataegus douglasii), currants (Ribes spp.) are already commonly planted by land managers in riparian restoration projects for the benefits they provide to stream health and wildlife. Some shrub species, including those that are primarily wind-pollinated, have been noted as food sources for foraging insect pollinators (Saunders 2018), and several native species (including several species of willow) are documented as dependent on bees for pollination (NRCS Plant Resources Database, https://plants.sc.egov.usda.gov/java/). Since some shrubs bloom earlier in the season than many forbs, they may be critical early-season resources for pollinators, and thus restorations that use shrubs may be more valuable than previously thought. Additionally, planting shrubs could help provide complimentary resources to wild bees in agricultural settings (Saunders 2018; Bentrup et al. 2019). Several studies have examined the effects of livestock grazing on pollinators (Kruess & Tscharntke 2002; Vulliamy et al. 2006; Le Féon et al. 2010; Kimoto, et al. 2012b). Work by Hatfield and LeBuhn (2007) indicated that the impacts of livestock grazing on bees varies with grazing species and with grazing intensity. Despite the number of studies on the effects of grazing on bees, the impacts of innovative grazing strategies, designed for sustainability, on pollinators are not well understood.

Understanding how such grazing practices impact pollinator populations is a high priority, as pollinators are likely critical for rangeland health (Black et al. 2011) and grasslands and riparian meadows often support a high diversity of flowering plants and pollinators (Kimoto et al. 2012a, b; DeBano et al. 2016). In Oregon, native bees have been studied in agricultural systems (Bosch & Kemp 2005; Rao & Stephen 2009; Broussard et al. 2011), in eastern Oregon grassland systems (Kimoto et al. 2012a, b; DiCarlo et al. 2019) and in riparian meadows of the Blue Mountains (DeBano et al. 2016; Roof et al. 2018). Ongoing work in restored riparian meadows of the Blue Mountains at the USFS Starkey Experimental Forest and Range (Starkey) has characterized pollinator communities and impacts of wild ungulate herbivory on native bees and plants (DeBano et al. 2016; Averett et al. 2017; Roof et al. 2018). Some of this work has focused on how cattle grazing affects bees (Kimoto et al. 2012b) and other work has focused on bee interactions with plants in restored riparian areas (DeBano et al. 2016, Roof et al. 2018). The studies relating to grazing has focused on single systems over short durations and simultaneous comparisons of pollinator responses across multiple systems has not been conducted in Oregon systems or elsewhere. Studies conducted in restored riparian areas have focused broadly on bee and plant interactions and not on interactions with those plants that are often planted in restoration projects. The research described here will contribute to our current state of knowledge of livestock grazing effects on bees and will evaluate how bees interact with plants that are commonly used in riparian restoration. In the second chapter of my thesis, I examined whether late-season rotational cattle grazing affects bee and blooming plant communities in two grassland locations in eastern Oregon. This chapter focuses on analyzing differences in bee and blooming plant abundance, richness, diversity and community composition in grazed and ungrazed sites and determining if effects differ in the short and long-term. I also describe spatiotemporal patterns evident in blooming plant and bee communities and relate those patterns to how they may influence the effect of grazing on plant and native bee communities.

In the third chapter of my thesis, I examined bee and blooming plant interactions in a restored riparian area focusing on interactions between bees and plants commonly used in restorations and co-occurring native forbs. In this chapter, I describe the abundance and species richness of blooming shrubs and forbs in late- spring/early-summer and examine how bee visitation rates differ for forbs and shrubs during this period. I also describe phenological changes in blooming plant abundance, richness, and diversity throughout the season and characterize bee-plant interactions for the entire season. Last, I describe which shrub species hosted the greatest diversity of bee species to provide some management relevant recommendations to restoration managers. In my final chapter, I describe the key findings of the previous two chapters and provide some basic recommendations for land management aimed at improving pollinator habitat. I also make suggestions for future research directions that could build on the findings described in this thesis.

Literature Cited

Averett JP, Endress BA, Rowland MM, Naylor BJ, Wisdom MJ (2017) Wild ungulate herbivory suppresses deciduous woody plant establishment following salmonid stream restoration. Forest Ecology and Management 391:135-144 Bentrup G, Hopwood J, Adamson NL, Vaughan, M (2019) Temperate agroforestry systems and insect pollinators: A review. Forests 10: 981-1001 Black SH, Shepherd M, Vaughan M (2011) Rangeland management for pollinators. Rangelands 33: 9-13 Bosch J, Kemp WP (2005) Alfalfa leafcutting bee population dynamics, flower availability, and pollination rates in two Oregon alfalfa fields. Journal of Economic Entomology 98:1077-1086 Broussard M, Rao S, Stephen WP, White L (2011) Native bees, honeybees, and pollination in Oregon cranberries. HortScience 46:885-888 Colla SR, Gadallah F, Richardson L, Wagner D, Gall L (2012) Assessing declines of North American bumble bees (Bombus spp.) using museum specimens. Biodiversity and Conservation 21: 3585-3595

Crane E & Walker P (1984) Pollination directory for world crops. International Bee Research Association, London, UK DeBano SJ, Roof SM, Rowland MM, Smith LA (2016) Diet overlap of mammalian herbivores and native bees: Implications for managing co-occurring grazers and pollinators. Natural Areas Journal 36:458-477 Hallman CA, Sorg M, Jongejans E, Siepel H, Hofland N, Schwan H, Stenmans W, Müller A, Sumser H, Hörren T, Goulson D, de Kroon H (2017) More than 75 percent decline over 27 years in total flying insect biomass in protected areas. PLOS One 10:1-21 Hatfield RG & LeBuhn G (2007) Patch and landscape factors shape community assemblage of bumble bees, Bombus spp. (Hymenoptera: Apidae), in montane meadows. Biological Conservation 139:150-158 DiCarlo Smith LA, DeBano SJ, Burrows S (2019) Short-term response of two beneficial invertebrate groups to wildfire in an arid grassland system, United States. Rangeland Ecology & Management 72:551-560 Kearns CA, Inouye DW, Waser NM (1998) Endangered mutualisms: The conservation of plant-pollinator interactions. Annual Review of Ecology and Systematics 2:83-112 Kimoto C, DeBano SJ, Thorp RW, Robbin W, Rao S, Stephen WP (2012a) Investigating temporal patterns of a native bee community in a remnant North American bunchgrass prairie using blue vane traps. Journal of Insect Science 12:1-23 Kimoto C, DeBano SJ, Thorp RW, Taylor RV, Schmalz H, DelCurto T, Johnson T, Kennedy PL, Rao S (2012b) Short-term responses of native bees to livestock and implications for managing ecosystem services in grasslands. Ecosphere 3:88-107 Klein A-M, Vaissière BE, Cane JH, Steffan-Dewenter I, Cunningham SA, Kremen C, Tscharntke T (2007) Importance of pollinators in changing landscapes for world crops. Proceedings: Biological Sciences 274:303-313

Kovács-Hostyánszki A, Espındola A, Vanbergen AJ, Settele J, Kremen C, Dicks LV (2017) Ecological intensification to mitigate impacts of conventional intensive land use on pollinators and pollination. Ecology Letters 20:673-689 Kremen C & Ricketts T (2000) Global perspectives on pollination disruptions. Conservation Biology 14:1226-1228 Kruess A & Tscharntke T (2002) Grazing intensity and the diversity of grasshoppers, butterflies, and trap-nesting bees and wasps. Conservation Biology 16:1570- 1580 Kulhanek K, Steinhauer N, Rennich K, Caron DM, Sagili RR, Pettis JS, Ellis JD, Wilson ME, Wilkes JT, Tarpy DR, Rose R, Lee K, Rangel J, vanEngelsdorp D (2017) A national survey of managed honey bee 2015–2016 annual colony losses in the USA. Journal of Apicultural Research 56:328-340 Le Féon V, Schermann-Legionnet A, Delettre Y, Aviron S, Billeter R, Bugter B, Hendrickx F, Burel F (2010) Intensification of agriculture, landscape composition and wild bee communities: A large scale study in four European countries. Agriculture, Ecosystems & Environment 137:143-150 LeBuhn G, Droege S, Connor EF, Gemmill-Herren B, Potts SG, Minckley RL, Griswold T, Jean R, Kula E, Roubik DW, Cane J, Wright KW, Frankie G, Parker F (2013) Detecting insect pollinator declines on regional and global scales: Detecting pollinator declines. Conservation Biology 27:113-120 Lee KV, Steinhauer N, Rennich K, Wilson ME, Tarpy DR, Caron DM, Rose R, Delaplane KS, Baylis K, Lengerich EJ, Pettis J, Skinner JA, Wilkes JT, Sagili R, vanEnglersdorp D (2015) A national survey of managed honey bee 2013- 2014 annual colony losses in the USA. Apidologie 46:292-305 Losey JE & Vaughan M (2006) The economic value of ecological services provided by insects. BioScience 56:311-323 Ollerton J, Winfree R, Tarrant S (2011) How many flowering plants are pollinated by animals? Oikos 120:321-326 Potts SG, Biesmeijer JC, Kremen C, Neumann P, Schweiger O, Kunin WE (2010) Global pollinator declines: trends, impacts and drivers. Trends in Ecology & Evolution 25:345-353

Rao S & Stephen WP (2009) Bumble bee pollinators in red clover seed production. Crop Science 49:2207-2214 Roof SM, DeBano SJ, Rowland MM, Burrows S (2018) Associations between blooming plants and their bee visitors in a riparian ecosystem in eastern Oregon. Northwest Science 92:119-135 Saunders ME (2018) Insect pollinators collect pollen from wind-pollinated plants: Implications for pollination ecology and sustainable agriculture. Insect Conservation and Diversity 11:13-31 Vanbergen AJ & Insect Pollinators Initiative (2013) Threats to an ecosystem service: pressures on pollinators. Frontiers in Ecology and the Environment 11:251- 259 Vulliamy B, Potts SG, Willmer PG (2006) The effects of cattle grazing on plant- pollinator communities in a fragmented Mediterranean landscape. Oikos 114:529-543 Walker, K (2017) State of Oregon Agriculture. Oregon Department of Agriculture. [Industry report] Weiner CN, Werner M, Linsenmair KE, Blüthgen N (2014) Land-use impacts on plant–pollinator networks: interaction strength and specialization predict pollinator declines. Ecology 95:466-474 Wilson JS & Carril OJM (2015) The bees in your backyard: A guide to North America’s bees. Princeton University Press, Princeton, New Jersey

CHAPTER 2: LATE-SEASON LIVESTOCK GRAZING: A SUSTAINABLE STRATEGY FOR NATIVE BEES IN GRASSLANDS OF THE INTERIOR PACIFIC NORTHWEST, USA?

Scott R. Mitchell, Sandra J. DeBano, Mary M. Rowland, Lesley R. Morris, Heidi Schmalz, Skyler Burrows and Scott B. Lukas

Abstract

Bees are the most important pollinators in most ecosystems and are critical to plant reproduction and ecosystem health in wild and managed landscapes. Livestock grazing occurs on private and public lands at broad scales in the western United States and represents a potentially large stressor for pollinator populations. Effects of livestock on bees and other pollinators are expected to be greatest when the periods of activity of both herbivore groups overlap spatially and/or temporally. Late-season grazing may be one way to reduce impacts of cattle on native bees by reducing temporal overlap in foraging of bees and cattle. In 2018 we simultaneously sampled bee and plant communities throughout the growing season at two locations (a riparian meadow system and a bunchgrass prairie remnant) to examine short-term responses to late-season cattle grazing in the Pacific Northwest. We also examined effects of long-term exposure to livestock grazing at one site, where grazing treatments have been in place since 2004. Overall, within-season variability was high in bee and plant communities. At the riparian meadow site, bloom richness peaked in June and bloom diversity, bee richness and bee diversity peaked in July. At the bunchgrass prairie location, bloom and bee richness and diversity peaked in June. While short-term grazing resulted in significantly lower abundance, richness, and species diversity of blooms at the riparian meadow location later in the summer, we detected no effect on bee communities. At the bunchgrass prairie location, we found no significant effect on blooming plant communities and no negative effect on bee communities. In fact, bee abundance and species richness were higher at grazed sites relative to ungrazed sites. No significant long-term effects were detected in plant and bee communities at the location where grazing had occurred since 2004. In sum, we found the short-term effects of late-season cattle grazing on blooming plant communities varied by

location; however, our results indicate that late-season grazing may not have negative short-term effects on native bees in Pacific Northwest grasslands, and may represent a viable option when designing regionally sustainable grazing practices that taken into account pollinator conservation.

Introduction

Invertebrates have long been recognized as critical members of ecological communities, due in part to the disproportionately large role they play in ecosystem processes (Wilson 1987). Pollination is no exception to this rule. Pollination is a crucial service that occurs in nearly every ecosystem on the planet and is mediated by a wide range of animals including bats, birds, rodents, and invertebrates (Roulston & Cane 2000). Invertebrates are more speciose than other pollinating taxa and are also responsible for delivering the bulk of pollination services in most systems (Kearns & Inouye 1997; NRCS 2007; Potts et al. 2010). In the United States (US), the value of ecosystem services provided by invertebrate pollinators is estimated to be approximately $3.07 billion a year (Crane & Walker 1984; Losey & Vaughan 2006; Klein et al. 2007). Most of this value is generated by managed bees (e.g., honey bees, blue orchard mason bees) and native bees (e.g., bumble bees, mining bees, leafcutter bees); however, in US range and wildland systems, honey bees are not common flower visitors (Hung et al. 2018) and native bees are responsible for the bulk of animal-mediated pollination (Ballantyne et al. 2015). Native bees are a diverse group of over 4,000 species in North America (Michener 2007) that vary tremendously in size, flower preference, nesting habits, and social organization (Michener 2007). While most people are familiar with the European honey bee (Apis mellifera), the life history of this single species is not representative of most other bee species. Most native bee species are small, solitary insects that emerge in the spring and work alone to forage pollen and nectar from flowers to provision brood cells and nectar to feed themselves. Of native bees, around 70-80% of species nest below ground in preexisting holes, holes they construct, or in abandoned rodent nests (Michener 2007; Harmon-Threatt 2020; Liczner & Colla

2019). Other species use above-ground resources, including downed wood, standing trees, and pithy stems. Thus, native bee communities need landscapes with abundant floral resources and varied nesting habitat. Lands, such as publicly-owned federal and state lands in the US, that have not been converted to agriculture or urban and suburban environments, are likely to have a higher density of floral and nesting resources and may provide prime pollinator habitat (Black et al. 2011, Cane 2011, Kimoto et al. 2012a, b). In fact, these types of areas may provide spillover pollination services to neighboring agricultural land (Öckinger & Smith 2007; Garibaldi et al. 2011). However, even public lands and open green spaces can be vulnerable to habitat degradation. For example, public lands are often designated for “multiple-use,” often including livestock grazing, which can influence ecosystems in various ways (Fleischner 1994; Jones 2000; Freilich et al. 2003; Schönbach et al. 2011) and may be a significant stressor for pollinators (Black et al. 2011; Cane 2011). Livestock grazing is widespread in the US. For example, in 2017, cattle grazing occurred on 30 million ha of US Forest Service land and 62 million ha of land managed by the Bureau of Land Management (Vincent 2019). This represents over half of the combined land managed by both agencies (Vincent 2019). As ranching plays a significant role in US economies (Vincent 2019) and culture (Kirner 2015), it is likely that livestock grazing will continue to be a prevalent land use on western US landscapes. When widespread grazing first began in the US, it was often practiced without much thought to the long-term sustainability of pasture and rangelands, resulting in many degraded rangelands (Borman 2005). Grazing currently occurs at substantially lower rates than those seen in the early 1900s (Borman 2005). While modern grazing intensity is much lower than historic levels (Galbraith & Anderson 1991), range management is often more intensive than it was historically (Baker et al. 2010). Modern grazing often involves calculating plant growth of rangelands and using plant growth to determine stocking rate on a landscape (Butler et al. 2003). Factors that may be manipulated to make grazing plans more sustainable include timing in the year, stocking rate, type of grazing animal, and rotation of animals.

Livestock grazing may impact bee communities through multiple pathways, including through dietary overlap (livestock consuming the same floral resources used by bees) (DeBano et al. 2016), by affecting nesting habitat (Kurz et al. 2006; Kimoto et al. 2012b; Schmalz et al. 2013), or by altering foraging behavior (Sjödin 2007). In fact, multiple studies have shown that grazing significantly affects bee communities. Some have found that grazing reduced bee diversity, richness and/or abundance (Kruess & Tscharntke 2002; Sjödin 2007; Hatfield & LeBuhn 2007; Xie et al. 2008; Le Féon et al. 2010; Kimoto et al. 2012b). Others have found positive effects of grazing on bee communities, with increased bee species richness and abundance in grazed locations (Vulliamy et al. 2006; Lázaro et al. 2016; Shapira et al. 2020), although all of these studies were conducted in Mediterranean systems, and their applicability to North American grasslands is unclear. Variability in response of native bee communities to livestock grazing may be due to differences in the type of grazer, intensity of grazing, evolutionary history of the system with regard to large ungulate grazers, and timing of grazing (Kimoto et al. 2012b). Timing may be a particularly important factor influencing community level responses; for example, although Lázaro et al. (2016) found generally higher bee abundance and richness at intermediate grazing levels, they found early-season grazing resulted in lowered species richness. Other studies that detected significant negative effects of cattle grazing on bee communities were conducted in systems where grazing occurred early in the season (Kruess & Tscharntke 2002; Xie et al. 2008; Kimoto et al. 2012b). To our knowledge, only one study has examined pollinator responses to late-season grazing; Sjödin et al. (2007) found that most observed pollinator species in a Swedish grassland system occurred at higher rates at sites where grazing was delayed until mid-summer. Given differences between European and North American grasslands, it is not clear that the same results could be expected in the US. Here, we examine whether late-season grazing may be one approach to allow livestock and bees to share rangelands by temporally separating resources. We conducted cattle grazing experiments in two grasslands systems in the Pacific Northwest, US to examine both short-term and long-term responses to late-season

grazing. Our specific objectives were to 1) quantify seasonal variation in abundance, richness, diversity, and community composition of blooming plant and bee communities, 2) measure the short-term responses of those communities to late- season grazing, and 3) examine the effects of long-term exposure to sustainable grazing practices in one system.

Methods

Study Areas: The Nature Conservancy’s Zumwalt Prairie Preserve

The Zumwalt Prairie Preserve (the Zumwalt) is in Wallowa County, Oregon (45.55o N, 116.95o W, Figure 2.1). The Zumwalt, covering 13,269 ha, is owned and managed by The Nature Conservancy (TNC) and includes one of the largest remaining intact remnants of Pacific Northwest Bunchgrass Prairie (Tisdale 1982; Kennedy et al. 2009). Most of the prairie is located between 1,060 - 1,680 m and receives an average of 48 cm of precipitation each year (NOAA Climatic Data Center 2010). Precipitation occurs primarily from November to June. The coldest month of the year is December with average temperatures ranging from -8.3 to 1.7o C, while the warmest month of the year is July with average temperatures ranging from 6.8 to 29.5o C (NOAA Climatic Data Center 2010). Human activity on the prairie occurred for thousands of years prior to Euro- American settlement in the late 1800s. Indigenous peoples of the Nez Perce and other tribes collected plants and animals on the prairie and began livestock (mostly horses) grazing in the early 1700s (Bartuszevige et al. 2012). Following the forced relocation of indigenous peoples onto reservations in 1877, Euro-Americans continued to graze livestock on the prairie until the present day. From the 1880s until the 1940s, livestock grazing on the prairie was dominated by sheep but has since been dominated by cattle (Bartuszevige et al. 2012). Portions of the prairie have been cultivated; however, poor soil condition and the relatively cool climate historically limited continual cultivation of widespread areas on the prairie (Bartuszevige et al. 2012). While the prairie was historically used to produce crops and livestock (sheep, cattle, and horses), large portions of the prairie still have relatively intact native plant

communities. Communities are dominated by Idaho fescue (Festuca idahoensis), prairie junegrass (Koeleria macrantha), and bluebunch wheatgrass (Psuedoroegneria spicata) and a wide variety of forbs (Bartuszevige et al. 2012). Since 2000, TNC has owned and managed the Zumwalt with the goal of understanding how to sustain and enhance conservation values of rangelands in the context of sustainable livestock grazing. While experimental studies on the effects of grazing on prairie ecosystems have been conducted at the Zumwalt since 2007, our study is nested within the context of a long-term project examining the effects of fire and grazing on prairie ecosystems that began in 2018. Our study used 16 sites (8 grazed, 8 ungrazed) from an established grazing experiment distributed over four blocks. In each block, we sampled two sites where cattle have been excluded since 2004/2005 and two sites that have been grazed at a moderate rate (0.2 – 0.4 AUM/acre) since 2006. During 2018, cattle were rotated through pastures beginning on June 19 (June sampling occurred prior to grazing) and ending on August 9 (Table 2.1; Figure 2.2). Stocking occurred at moderate rates (0.28-0.30 AUM/acre), which are considered representative of typical stocking in the region (Kimoto et al. 2012b).

Study Areas: The United States Forest Service Starkey Experimental Forest and Range

The US Forest Service Starkey Experimental Forest and Range (Starkey) is in Union County of Oregon (45o 12’ N and 118o 3’ W, Figure 2.1). The climate at Starkey is typical of forested systems in the Blue Mountains of eastern Oregon and is characterized by warm dry summers and cool wet winters. Starkey receives an annual average precipitation of 42 cm (NOAA Climatic Data Center 2010), most of which falls between November and June. The coldest month is December with average temperatures ranging from -4.6 to 3.1o C, while the warmest month is July with average temperatures ranging from 12.1 to 29.7o C (NOAA Climatic Data Center 2010). Starkey has been used since 1955 for various studies focusing on ungulate herbivory and cattle grazing, and pastures along Meadow Creek used in this study

were installed during the 1970s (Rowland et al. 1997). Our study was conducted in three of the five pastures that lie along an 11 km section of Meadow Creek. Meadow Creek is a tributary to the Grand Ronde River and is characterized by a small band of riparian meadows surrounded by drier uplands, which are dominated by mixed coniferous forests. Along the 11 km section surveyed, the elevation of Meadow Creek is 1,100 - 1,200 m and the surrounding uplands have elevations as high as 1,500 m. The meadow system is dominated by various grass and sedge species; annual forb species: ( e.g., Myosotis stricta, Collomia linearus, Draba verna); perennial forb species: (e.g., Potentilla gracilis, Viola nuttallii, Achillea millefolium); and some shrubs and trees (DeBano et al. 2016). Between 2012 and 2013, the riparian area of Meadow Creek was planted with over 50,000 woody shrubs as part of a large-scale riparian restoration project (Averett et al. 2017). While this system had been historically grazed prior to the 2012-2013 restoration, grazing did not occur following 2012 restoration until 2017. Our study was conducted within fenced exclosures and open sites that were installed in three of the five pastures as a part of a large manipulative grazing experiment that began in 2017. Within each of the three pastures along Meadow Creek, we examined two sites that were grazed by cattle and two sites that were ungrazed by cattle for a total of six ungrazed sites and six grazed sites. As with the Zumwalt, rotational grazing practices were used (Table 2.1; Figure 2.2). During 2018, cattle were rotated through pastures beginning on June 28 (June sampling occurred prior to grazing) and ending on September 12 (Table 2.1; Figure 2.2). Stocking occurred at moderate rates (0.1-0.9 AUM/acre), which are representative of stocking rates in the region (Kimoto et al. 2012b).

Plant Community Sampling

At every site, the blooming plant community was surveyed concurrent with bee sampling (Figure 2.2). Plants were sampled along the same five parallel transects where bees were sampled (Figure 2.1). All blooming stems encountered in each belt transect were identified and counted by the same individual during each sampling period to reduce observer bias. Apart from two taxa at the Zumwalt, all blooming

plants were identified to species; the two taxa at the Zumwalt identified to genus were Crepis and Lupinus.

Bee Community Sampling

Bees were sampled concurrently with plant sampling once per month at the Zumwalt in June, July, and August of 2018 and at Starkey in May, June, July and September of 2018 (Figure 2.2). At every site, an array of 15 240 ml pan traps was laid out in groups of three (one fluorescent blue, one fluorescent yellow, and one white), with each group placed at the center of five parallel, 20-m long and 0.5-m wide belt transects separated by 15 m (Figure 2.1). Traps were filled with approximately 180-200 ml of slightly soapy water. At Starkey, the trapping array was laid out perpendicular to Meadow Creek while at the Zumwalt, transects were centered on each plot and oriented north to south. All traps were set out for approximately 48 hours, and upon collection the number of disturbed traps was noted. In the laboratory, collected specimens were washed and pinned. Most bees were identified to species or morphospecies. Taxa that were frequently not identified to species include the following: Epeolus, Lasioglossum (subgenus Dialictus), Nomada, Sphecodes, and Stelis. For more information on bee species identification see Kuhlman & Burrows (2017).

Statistical Analyses

Abundance, Richness, and Diversity Responses

Univariate community responses were evaluated for each sampling month using one-way analysis of variance (ANOVA). Separate analyses were conducted evaluating short-term and long-term effects of cattle grazing on bee and plant communities. The community responses that were evaluated included bee abundance (bee/trap/hour), bee species richness, bee diversity (Shannon-Weiner Diversity Index), bloom abundance (total blooms counted on transects), blooming plant species richness, and blooming plant diversity (Shannon-Weiner Diversity Index). For short- term grazing analyses, “grazed sites” were defined as sites that had been grazed in the

2018 season prior to the time of sampling. For long-term grazing analyses, Zumwalt sites were used that had been grazed since 2006 and may or may not have been grazed in the 2018 year. Longer term grazing was not considered for Starkey as sites had only been grazed in one year prior to sampling. Assumptions of normality were evaluated visually, and data were found to meet assumptions for ANOVA analysis. All ANOVAs were conducted in the software R (R Core Team 2019), boxplots and bar charts were constructed using the package ggplot2 (Wickham 2016), and boxplots were arranged using the gridExtra package (Auguie 2017). Species rarefaction curves were constructed using the iNEXT package in R (Hsieh et al. 2016) with methods outlined by Hsieh et al. (2016).

Blooming Plant and Bee Community Analyses

Communities were visualized using non-metric multidimensional scaling (NMS) separating sites in plant community and bee community space. These analyses were conducted using Sorenson distance measure, a maximum of 500 iterations, and a random starting point with a step length of 0.20; 250 runs were conducted with real data and 250 runs were conducted with randomized data. Pearson correlations were used to quantify relationships between plant and bee species abundance, and environmental variables (sampling month, bloom abundance, bloom diversity, and stocking rate) with ordination axes. To determine if blooming plant and bee community composition differed between grazed and ungrazed sites, multi-response permutation procedures (MRPP) were used. MRPP analysis returns an “A” value which provides an estimate of in group agreement and a p-value which provides an estimate of how likely it would be to see the given community by chance (McCune et al. 2002). All multivariate analyses were conducted in PC-ORD Version 7 (McCune & Mefford 2006).

Results

General Results

For blooming plants, we counted 68,568 blooming stems from 156 plant species across both locations. On Zumwalt plant transects, we counted 45,642 blooms from 85 plant species. On Starkey plant transects, we counted a total of 22,926 blooms from 102 species. Over the course of 2018 sampling, we collected 7,128 bees from 172 different taxa. See Table A1.1 for bee species list and occurrence at each location. At the Zumwalt, a total of 3,608 bees were collected, 3,317 of which were identified to one of 122 species or morphospecies. Of the specimens not identifiable to species, 284 specimens were identified to subgenus (283 of which were Lasioglossum (Dialictus)) and 7 were only identifiable to genus. At Starkey, we collected a total of 3,520 bees, 3,362 of which were identified to one of 119 species or morphospecies. Of the remaining specimens, 119 were only identifiable to subgenus (117 of which were Lasioglossum (Dialictus)) and 39 of which were only identifiable to genus.

Seasonal Variation in Plant and Bee Communities

Blooming plant abundance, richness, and diversity varied through the season. At the Zumwalt, plant abundance was highest in June, species richness was highest in June and July, and species diversity was highest in June (Figure 2.3). The number of plants making up the top 90% of blooms counted was 18 in June, 11 in July, and 4 in August (Table A1.6). At Starkey, bloom abundance peaked in May, blooming plant richness peaked in June, and bloom diversity peaked in July (Figure 2.3). The number of plant species making up the top 90% of blooms was 9 in May, 11 in June, 15 in July, 6 in August, and 3 in September (Table A1.7) Bee abundance and richness also varied through the season. At the Zumwalt, bee species richness and diversity peaked in June, and bee abundance peaked in August (Figure 2.3). The bee community in August was highly dominated by two species of sweat bees, Lasioglossum incompletum and Halictus tripartitus, which made up more than 55% of all bees collected that month. In fact, the number of

species making up the top 90% of all bees collected at the Zumwalt decreased throughout the season (from 31 species in June, 27 in July, and 14 in August, Table A1.4). At Starkey, both bee richness and abundance peaked in July and bee diversity was relatively similar from May through July (Figure 2.3). As with the Zumwalt in August, Starkey in September was also dominated by two sweat bee species: H. tripartitus and H. ligatus, which made up more than 58% of all bees collected that month. The number of species making up the top 90% of all bees collected was 23 in May, 24 in June, 25 in July, and 16 in September (Table A1.5).

Abundance, Richness and Diversity Responses

While we found no significant differences in bee abundance, richness, or Shannon diversity between recently grazed and ungrazed sites at Starkey (Table 2.2), bee abundance and species richness at the Zumwalt in July were significantly higher at sites that had been grazed prior to sampling in the 2018 season (Table 2.2, Figure 2.4). Rarified richness curves were constructed for the two sites that had been grazed in July (Graze 1 and Graze 2) and the two control sites (Control 1 and Control 2) from the same pastures for July to examine the if rarified richness differed between paired grazed and ungrazed sites for July. The 95% confidence intervals around rarefaction curves show significant overlap and indicate that rarefied species richness did not differ between grazed and ungrazed sites (Figure 2.5). While we observed no statistically significant short-term responses of blooming plant abundance, species richness, or Shannon diversity to grazing at the Zumwalt (Table 2.2), we did find significant responses in all three variables at Starkey in July and August; average bloom abundance, species richness, and Shannon diversity were all lower in sites that had been grazed prior to sampling in those months (Table 2.2, Figure 2.4). The results of the ANOVA analyses indicated no significant long-term effects of cattle grazing on blooming plant or bee abundance, species richness and Shannon diversity at the Zumwalt (Table 2.3).

Plant and Bee Community Composition Relative to Season, Location, and Treatment

When analyzing community data for all sites and all sampling periods simultaneously, we found that plant communities were strongly structured by sampling month and by study location. Later sample months were correlated with positive values on axis 1 and earlier months were correlated with negative values on axis 1. In the plant ordination, total R2n (nonmetric fit) was 0.99 with axis 1 explained 35% of the variance in the data and axis 2 explained 5%. Bee and plant communities did not differ significantly relative to grazing treatment (Table 2.4). In both bee and plant ordinations, sites separated by location on axis two with Starkey communities above, and Zumwalt communities below on axis 2 (Figure 2.6, see Table A1.2 and A1.3 for significant axis correlations). Axis 1 of the NMS ordination of plant communities was most strongly correlated with month and to a lesser extent with stocking rate. Axis 2 was more strongly correlated with stocking rate and less correlated with month. Plant species most strongly, positively associated with axis 1 were Perideridia gairdneri, Solidago missouriensis, and Hypericum scouleri. These were also plants that were strongly associated with late-season bloom communities. The plants most negatively associated with axis 1 include Myosotis stricta, Collinsia parviflorum, and Draba verna. These three plants are all abundant early-season annual plants that occurred at both Starkey and the Zumwalt. The plants most strongly positively associated with axis 2 include Taraxacum officinale, Veratrum californicum, and Symphyotrichum spathulatum. All of these were plants that occurred at Starkey but were never observed at the Zumwalt. The plants most strongly negatively associated with axis 2 were Arnica sororia, Erigeron pumilus, and Lupinus sp. These species commonly occurred in the Zumwalt but were not common or were not observed at Starkey. See Supporting Information: Table A1.3 for details on axes correlations. Bee communities were strongly structured by location sampling month and within each location were strongly structured by month, especially at Starkey (Figure 2.6). In the bee ordination, total R2n (nonmetric fit) for the ordination was 0.98, with a final stress of 12.7 and an association value, A = 0.27 (a measure of similarity between columns, based on randomization tests). Axis 1 explained 48% of variance

in the data and axis 2 explained 16% of variance in the data. Axis 1 of the NMS ordination of bee communities was most strongly related with month followed by bloom abundance and blooming plant diversity. Stocking rate was correlated with axis 1 but to a lesser degree. Axis 2 was most strongly correlated with month and weakly correlated with bloom abundance, bloom diversity and stocking rate. When bee and plant communities were analyzed within a single location and for a single month (e.g. Starkey, bee community, June), no significant differences were found between grazed and ungrazed sites (Table 2.4).

Discussion

The results of this study suggest that late-season grazing may be an effective strategy for reducing negative effects of livestock on native bee communities in grasslands of the Pacific Northwest. Although we found some evidence of decreases in floral abundance, richness, and diversity at one location (riparian meadows) in response to short-term grazing, we detected no negative effects of short-term grazing on blooming plants at the other location (bunchgrass prairie). Further, we found no negative effects of late-season, short-term grazing on any aspect of the native bee community. Our results are generally consistent with a study conducted in European grasslands (Sjödin 2007) and add to evidence suggesting that late-season grazing may be effective in mitigating the negative impacts of grazing on bee communities observed in other study systems that were grazed early-mid season (Kruess & Tscharntke 2002; Xie et al. 2008; Kimoto et al. 2012b; Lázaro et al. 2016). A major reason why late-season grazing may reduce impacts on native bees is because it generally occurs after peaks in species richness and diversity of bees and the floral resources they use. We found strong temporal variability in both the plant and bee communities, with floral richness peaking in early or mid-season at both locations, a result consistent with other work conducted in the region (Kimoto et al. 2012a,b) and in other western US locations (Golet et al. 2011; Rhoades et al. 2018). Peak bee richness occurred just before or just after peak bloom richness at both sites, a pattern seen in other studies (Potts et al. 2003; Ebeling et al. 2008; Holzschuh et al. 2012). While bee abundance peaked at the bunchgrass location in August, a high

proportion of bees collected (60%) were two generalist sweat bee species. By grazing cattle later in the year when bloom and bee species richness are both lower, it is possible that the dietary overlap between most bee species and cows is reduced (DeBano et al. 2016). In addition to finding few short-term effects of late-season livestock grazing at these two locations, we also did not detect any long-term effects of grazing on bees or plants at the Zumwalt. Starkey was not evaluated for long-term effects due to the relatively recent reintroduction of cattle into the system. The lack of long-term effects on the communities examined in this study, coupled with the presence of some short- term impacts may indicate that these ecosystems are resilient to short “pulse” disturbances and may be able to “reset” each year (Lake 2000). However, studies have indicated that perennial forbs can be long-lived (Treshow & Harper 1974; Ehrlén & Lehtilä 2002), and longer term surveys should be made to determine if blooming plant abundance changes with longer term grazing than was examined in this study (12 years of grazing at the Zumwalt sites). The different responses of the blooming plant communities to short-term effects at Starkey and Zumwalt indicate that the sensitivity of these habitats to grazing disturbances may differ and that riparian meadows of Starkey may be more vulnerable to grazing pressure. Since Starkey was only grazed for one season prior to 2018 sampling, it will be important to continue monitoring bee and blooming plant communities along Meadow Creek to determine if long-term effects on communities at Starkey emerge over time. It is also possible that nesting habitat on the Zumwalt is not a limiting factor and alterations to soil substrate associated with livestock grazing simply does not alter soil nesting substrates (positively or negatively) enough to observe any significant trends. In some contexts, estimating bee abundance using passive traps may be influenced by local floral abundance or alteration of the physical structure of vegetation (O’Connor et al. 2019). For example, higher abundance at the bunchgrass prairie in August may simply be an artifact of the trapping method used, given that low floral abundance in that month may have made traps relatively more attractive. Likewise, the significantly higher abundance and species richness in short-term

grazed sites in the same location in July may also be due to fewer floral resources in grazed areas, or increased visibility of traps because of reduced vegetation structure. Regardless of the reason for the higher abundance, the significantly higher species richness at those sites in July on was likely simply due to the higher abundance, as suggested by the fact that rarefied richness of grazed sites for July at the Zumwalt did not differ significantly from ungrazed sites. Despite questions regarding the accuracy of pan traps in estimating bee abundance and richness, they remain a standard method and are useful for comparing rarified richness, diversity, and community composition. Our results suggest that efforts towards developing more sustainable grazing practices for plants and pollinators may be working in inland Northwest grassland systems. Based on the results of our study, late-season grazing may be a good approach for range managers attempting to reduce the negative effects of grazing on bee and blooming plant communities. Additionally, surveying sites for pollinator diversity, richness and abundance may help range managers identify sensitive habitats that support complex bee and plant communities and less-sensitive sites that contain fewer bees and blooming plants. This would allow range managers to map these less- sensitive areas on their range and they could graze livestock on these sites in the early-season when bees are most vulnerable, rotating cattle onto more sensitive habitats later in the season. Many rangelands have at least some areas that do not have abundant pollinators or blooming plants (e.g., previously cultivated fields). To better understand community responses to sustainable grazing practices, future research efforts should investigate the following areas: the effect of landscape variables (e.g., floral abundance) on pan trap efficiency, the effects of grazing on individual taxa of conservation concern, trait- and phylogeny-based analyses of grazing effects on bee and plant communities (e.g., annual vs. perennial plants), longer term studies, and interactions between grazing and other disturbances (e.g., fire).

Literature Cited

Auguie B (2017) GridExtra: Miscellaneous Functions for “Grid” Graphics (version 2.3). R. https://CRAN.R-project.org/package=gridExtra

Averett JP, Endress BA, Rowland MM, Naylor BJ, Wisdom MJ (2017) Wild ungulate herbivory suppresses deciduous woody plant establishment following salmonid stream restoration. Forest Ecology and Management 391:135-144 Ballantyne G, Baldock KCR, Willmer PG (2015) Constructing more informative plant–pollinator networks: Visitation and pollen deposition networks in a heathland plant community. Proceedings of the Royal Society B: Biological Sciences 282: 11-30 Baker S, Bohle M, Bouck MJ, Cardon G, Cash D, Cheyney C, Church J, Crane K, Dreves A, Engel C, Falen C, Fife T, Fisher G, Fransen S, Gerrish J, Gillaspy B, Glaze JB, Graw W, Griggs T, Hall J, Hannaway D, Hooper L, Horneck D, Hulting A, Jensen K, Jensen S, Koenig RT, McLain B, Neher O, Neibling H, Ogle D, Shewmaker G, St. John L, Whitesides R, Williams S, Wilson R (2010) Pasture and grazing management in the Northwest. University of Idaho Extension PNW 614 Bartuszevige AM, Kennedy PL, Taylor RV (2012) Sixty-seven years of landscape change in the last, large remnant of the Pacific Northwest bunchgrass prairie. Natural Areas Journal 32:166-170 Black SH, Shepherd M, Vaughan M (2011) Rangeland management for pollinators. Rangelands 33: 9-13 Borman MM (2005) Forest stand dynamics and livestock grazing in historical context. Conservation Biology 19:1658-1662 Butler LD, Cropper JB, Johnson RH, Norman AJ, Peacock GL, Shaver PL, Spaeth KE, Pierson Jr. FB, Weltz MA (2003) The national range and pasture handbook (NRPH). NRCS 190-VI-NRPH, rev. 1 Cane JH (2011) Meeting wild bees’ needs on Western US rangelands. Rangelands 33:27-32 Crane E & Walker P (1984) Pollination directory for world crops. International Bee Research Association, London, UK DeBano SJ, Roof SM, Rowland MM, Smith LA (2016) Diet overlap of mammalian herbivores and native bees: Implications for managing co-occurring grazers and pollinators. Natural Areas Journal 36:458-477

Ebeling A, Klein A-M, Schmacher J, Weisser WW, Tscharntke T. (2008) How does plant richness affect pollinator richness and temporal stability of flower visits? Oikos 117:1808-1815 Ehrlén J & Lehtilä K (2002) How perennial are perennial plants? Oikos 98:308-322 Fleischner TL (1994) Ecological costs of livestock grazing in Western North America. Conservation Biology 8:629-644 Freilich JE, Emlen JM, Duda JJ, Freeman DC, Carfaro PJ (2003) Ecological effects of ranching: A six-point critique. BioScience 53:759-765 Galbraith WA & Anderson EW (1991) Grazing history of the Northwest. Rangelands 13:213-218 Garibaldi LA, Steffan-Dewenter I, Kremen C, Morales JM, Bommarco R, Cunningham SA, Carvalheiro LG, Chacoff NP, Dudenhöffer JH, Greenleaf SS, Holzschuh A, Isaacs R, Krewenka K, Mandelik Y, Mayfield MM, Morandin LA, Potts SG, Ricketts TH, Szentgyörgyi H, Viana BF, Westphal C, Winfree R, Klein AM (2011) Stability of pollination services decreases with isolation from natural areas despite honey bee visits: Habitat isolation and pollination stability. Ecology Letters 14:1062-1072 Golet GH, Gardali T, Hunt JW, Koenig DA, Williams NM (2011) Temporal and taxonomic variability in response of fauna to riparian restoration. Restoration Ecology 19:126-135 Harmon-Threatt A (2020) Influence of nesting characteristics on health of wild bee communities. Annual Review of Entomology 65:39-56 Hatfield RG & LeBuhn G (2007) Patch and landscape factors shape community assemblage of bumble bees, Bombus spp. (Hymenoptera: Apidae), in montane meadows. Biological Conservation 139:150-158 Holzschuh A, Dudenhöffer JH, Tscharntke T (2012) Landscapes with wild bee habitats enhance pollination, fruit set and yield of sweet cherry. Biological Conservation 153:101-107 Hsieh TC, Ma KH, Chao A (2016) iNEXT: an R package for rarefaction and extrapolation of species diversity (Hill numbers). Methods in Ecology and Evolution 7:1451-1456

Hung KLJ, Kingston JM, Albrecht M, Holway DA, Kohn JR (2018) The worldwide importance of honey bees as pollinators in natural habitats. Proceedings of the Royal Society B: Biological Sciences 285:1-8 Jones A (2000) Effects of cattle grazing on North American arid ecosystems: A quantitative review. Western North American Naturalist 60: 155-164 Joshi NK, Leslie T, Rajotte EG, Kammerer MA, Otieno M, Biddinger DJ (2015) Comparative trapping efficiency to characterize bee abundance, diversity, and community composition in apple orchards. Annals of the Entomological Society of America 108:785-799 Kearns CA & Inouye DW (1997) Pollinators, flowering plants, and conservation biology. BioScience 47:297-307 Kennedy PL, DeBano SJ, Bartuszevige AM, Lueders AS (2009) Effects of native and non-native grassland plant communities on breeding passerine birds: Implications for restoration of northwest bunchgrass prairie. Restoration Ecology 17:515-525 Kimoto C, DeBano SJ, Thorp RW, Robbin W, Rao S, Stephen WP (2012a) Investigating temporal patterns of a native bee community in a remnant North American bunchgrass prairie using blue vane traps. Journal of Insect Science 12:1-23 Kimoto C, DeBano SJ, Thorp RW, Taylor RV, Schmalz H, DelCurto T, Johnson T, Kennedy PL, Rao S (2012b) Short-term responses of native bees to livestock and implications for managing ecosystem services in grasslands. Ecosphere 3:88-107 Kirner KD (2015) The cultural heritage of family ranches. Rangelands 37:85–89 Klein A-M, Vaissière BE, Cane JH, Steffan-Dewenter I, Cunningham SA, Kremen C, Tscharntke T (2007) Importance of pollinators in changing landscapes for world crops. Proceedings: Biological Sciences 274:303-313 Kruess A & Tscharntke T (2002) Grazing intensity and the diversity of grasshoppers, butterflies, and trap-nesting bees and wasps. Conservation Biology 16:1570- 1580

Kuhlman M & Burrows S (2017) Checklist of bees (Apoidea) from a private conservation property in west-central Montana. Biodiversity Data Journal. Kurz I, O’Reilly CD, Tunney H (2006) Impact of cattle on soil physical properties and nutrient concentrations in overland flow from pasture in Ireland. Agriculture, Ecosystems & Environment 113:378-390 Lake PS (2000) Disturbance, patchiness, and diversity in streams. Journal of the North American Benthological Society 19:573-592 Lázaro A, Tscheulin T, Devalez J, Nakas G, Petanidou T (2016) Effects of grazing intensity on pollinator abundance and diversity, and on pollination services. Ecological Entomology 41:400-412 Le Féon V, Schermann-Legionnet A, Delettre Y, Aviron S, Billeter R, Bugter B, Hendrickx F, Burel F (2010) Intensification of agriculture, landscape composition and wild bee communities: A large scale study in four European countries. Agriculture, Ecosystems & Environment 137:143-150 Liczner AR & Colla SR (2019) A systematic review of the nesting and overwintering habitat of bumble bees globally. Journal of Insect Conservation 23:787-801 Losey JE & Vaughan M (2006) The economic value of ecological services provided by insects. BioScience 56:311-323 McCune B, Grace JB, Urban DL (2002) Analysis of ecological communities. 1st ed. MjM Software Design McCune B & Mefford MJ (2006) PC-ORD: Multivariate Analysis of Ecological Data. MjM Software, Gleneden Beach, Oregon, USA Michener CD (2007) The bees of the world. 2nd ed. Johns Hopkins University Press, Baltimore, Maryland National Oceanic and Atmospheric Administration (1981-2010) National Climatic Data Center https://catalog.data.gov/dataset/u-s-hourly-climate-normals-1981- 2010 (accessed 18 Dec 2019) National Research Council U. S. Committee on the Status of Pollinators in North America (2007) Status of pollinators in North America. National Academies Press, Washington, D.C.

Öckinger E & Smith HG (2007) Semi-natural grasslands as population sources for pollinating insects in agricultural landscapes. Journal of Applied Ecology 44:50-59 O’Connor RS, Kunin WE, Garratt MPD, Potts SG, Roy HE, Andrews C, Jones CM, Peyton JM, Savage J, Harvey MC, Morris RKA, Roberts SPM, Wright I, Vanbergen AJ, Carvell C (2019) Monitoring insect pollinators and flower visitation: The effectiveness and feasibility of different survey methods. Methods in Ecology and Evolution 10:2129-2140 Potts SG, Biesmeijer JC, Kremen C, Neumann P, Schweiger O, Kunin WE (2010) Global pollinator declines: trends, impacts and drivers. Trends in Ecology & Evolution 25:345-353 Potts SG, Vulliamy B, Dafni A, Ne’Eman G, Willmer P (2003) Linking bees and flowers: How do floral communities structure pollinator communities? Ecology 84:2628-2642 R Core Team. 2019. R: A Language and Environment for Statistical Computing. R. Vienna, Austria: R Foundation for Statistical Computing. https://www.R- project.org/ Rhoades PR, Davis TS, Tinkham WT, Hoffman CM (2018) Effects of seasonality, forest structure, and understory plant richness on bee community assemblage in a southern Rocky Mountain mixed conifer forest. Annals of the Entomological Society of America 111:278-284 Roulston TH & Cane JH (2000) Pollen nutritional content and digestibility for animals. Plant Systematics and Evolution; Heidelberg 222:187-209 Rowland MM, Bryant LD, Johnson BK, Noyes JH, Wisdom MJ, Thomas JW (1997) Starkey project: history facilities, and data collection methods for ungulate research. USFS PNW-GTR-396. United States Forest Service, Washington D.C. Schmalz HJ, Taylor RV, Johnson TN, Kennedy PL, DeBano SJ, Newingham BA, McDaniel PA (2013) Soil morphologic properties and cattle stocking rate affect dynamic soil properties. Rangeland Ecology & Management 66:445- 453

Schönbach P, Wan H, Gierus M, Bai Y, Müller, Lin L, Susenbeth A, Taube F (2011) Grassland responses to grazing: effects of grazing intensity and management system in an Inner Mongolian steppe ecosystem. Plant Soil 340:103-115 Shapira T, Henkin Z, Dag A, Mandelik Y (2020) Rangeland sharing by cattle and bees: moderate grazing does not impair bee communities and resource availability. Ecological Applications (in press) Sjödin NE (2007) Pollinator behavioural responses to grazing intensity. Biodiversity and Conservation 16:2103-2121 Tisdale EW (1982) Grasslands of western North America: The Pacific Northwest bunchgrass. Pages 223–245 In: Grassland ecology and classification symposium, Kamloops, British Columbia Treshow M & Harper K (1974) Longevity of perennial forbs and grasses. Oikos 25:93-96 Vincent CH (2019) Grazing fees: Overview and issues. CRS21232 Congressional Research Service, Washington, D.C. Vulliamy B, Potts SG, Willmer PG (2006) The effects of cattle grazing on plant- pollinator communities in a fragmented Mediterranean landscape. Oikos 114:529-543 Wickham H. 2016. Ggplot2: Elegant Graphics for Data Analysis. New York: Springer-Verlag. https://ggplot2.tidyverse.org Wilson EO (1987) The little things that run the world (the importance and conservation of invertebrates). Conservation Biology 1:344-346 Xie Z, Williams PH, Tang Y (2008) The effect of grazing on bumblebees in the high rangelands of the eastern Tibetan Plateau of Sichuan. Journal of Insect Conservation 12:695-703

Figures and Tables

2.1: Study locations and sampling set-up a) Both study locations were in northeastern Oregon, located 130 km from each other; b) The Nature Conservancy’s Zumwalt Prairie Preserve is a large remnant of native bunchgrass prairie; c) the United States Forest Service Starkey Experimental Forest and Range is home to a riparian meadow network along Meadow Creek (pictured); d) schematic of the sampling set up used at Starkey and the Zumwalt; and e) native bees are diverse in both locations. (See Table 2.1 for more in-depth location comparisons.)

2.2: Timeline of sampling events and cattle grazing events Dark gray shows when cattle grazing occurred on grazed sites (grazed site names are given next to or inside boxes). Light gray boxes show timing of plant sampling bouts, and white boxes show bee sampling bouts. During most sampling bouts, plant sampling occurred concurrently with bee sampling.

2.3: Seasonal variability in bees and plants Total bloom abundance, richness and diversity of blooms and bees at Starkey and the Zumwalt. Total bee abundance is broken up by 3 most common taxa for each month. Data is summarized for each location (12 sites at Starkey and 16 sites at the Zumwalt, analyzed together).

2.4: Short-term effects of cattle grazing Red triangles are group means, upper, middle, and lower extents of boxes show 25th, 50th (median), 75th quartiles respectively, lines show 1.5 interquartile range above and below the outer quartiles. a) Blooming plants at Starkey, based on 2 grazed sites and 10 ungrazed sites. b) Significant effects of grazing were observed in August for blooming plants at Starkey, based on 4 grazed sites and 8 ungrazed sites. c) Significant effects of grazing were detected in July for bees at the Zumwalt, based on 2 grazed and 14 ungrazed sites. See Table 2.2 for ANOVA results.

2.5: Rarefied bee richness of Zumwalt sites in July Figure 2.5: Rarefied bee richness for Zumwalt in July. Rarefied bee richness for short-term grazing sites in the Zumwalt sampled in July. Two sites were used for grazed data (Graze 1 and Graze 2) and two sites from the same pastures were used for ungrazed data (Control 1 and Control 2). Shaded areas around the lines shows 95% confidence interval for richness estimates.

2.6: NMS ordinations of bee and plant communities Figure 2.6: NMS Ordinations of bee and plant communities. Species with r2 correlations > 0.1 are marked with asterisks. Species correlations with axes are listed in Table A1.1.

Table 2.1: Locations by the numbers Comparison of the two locations sampled for this study. Locations by the numbers

Starkey Zumwalt Annual Precipitation 42 cm 48 cm Elevation 1,100 - 1,200 m 1,300 - 1,500 m Stocking Rate (AUM/acre) 0.1 - 0.9 (site dependent) 0.3 Total Area (acres) 28,000 33,000 Start Year of Grazing 2017 2004 Months Sampled May, Jun, Jul, Aug (plants), Sept Jun, Jul, Aug Bee Taxa Richness 128 122 Bloom Species Richness 102 85 Ownership US Forest Service Nature Conservancy

Table 2.2: Short-term effects of grazing on bee and plant communities Significant effects are highlighted and shown in Figure 2.4. For short-term grazing effects sites were defined as grazed if they had been grazed prior to sampling in the 2018 season. *In the Zumwalt all treatment sites had been grazed by August in the 2018 season, so the values for these months are the same as in Table 2.2 (below). Starkey Short Term Effects Bees Blooming Plants Month Parameter dF F- p-value dF F-stat p-value stat July Abundance 1, 10 0.4 0.6 1, 10 11.9 0.006 Species Richness 1, 10 2.3 0.2 1, 10 16.9 0.002 Shannon Diversity 1, 10 2.4 0.2 1, 10 42.4 < 0.001 August Abundance - - - 1, 10 7.9 0.02 Species Richness - - - 1, 10 8.1 0.02 Shannon Diversity - - - 1, 10 8.6 0.01 Sept Abundance 1, 10 0.002 0.97 1, 10 0.1 0.7 Species Richness 1, 10 0.3 0.6 1, 10 0.8 0.4 Shannon Diversity 1, 10 0.1 0.7 1, 10 0.9 0.4 Zumwalt Short Term Effects Bees Blooming Plants Month Parameter dF F- p-value dF F-stat p-value stat July Abundance 1, 14 5.2 0.04 1, 14 0.02 0.9 Species Richness 1, 14 5.4 0.04 1, 14 0.2 0.7 Shannon Diversity 1, 14 1.9 0.2 1, 14 0.4 0.6 August* Abundance 1, 14 2.0 0.2 1, 14 1.8 0.2 Species Richness 1, 14 0.02 0.9 1, 14 0.07 0.8 Shannon Diversity 1, 14 1.2 0.3 1, 14 0.05 0.8

Table 2.3: Long-term grazing effects on bee and plant communities No significant effects were observed from long-term cattle grazing on bee or plant communities the Zumwalt. Grazed sites were defined as sites that had been grazed seasonally since 2006 and may or may not have been grazed yet in 2018. For short- term effects, see Table 2.2 (above). Zumwalt Bees Blooming Plants Month Parameter dF F-stat p-value dF F-stat p-value June Abundance 1, 14 0.2 0.7 1, 14 0.5 0.5 Species Richness 1, 14 0.01 0.9 1, 14 0.9 0.4 Shannon Diversity 1, 14 0.01 0.9 1, 14 0.07 0.8 July Abundance 1, 14 0.7 0.4 1, 14 1.3 0.3 Species Richness 1, 14 0.2 0.6 1, 14 0.1 0.7 Shannon Diversity 1, 14 0.04 0.9 1, 14 0.6 0.5 August Abundance 1, 14 2.0 0.2 1, 14 1.8 0.2 Species Richness 1, 14 0.02 0.9 1, 14 0.07 0.8 Shannon Diversity 1, 14 1.2 0.3 1, 14 0.05 0.8

Table 2.4: MRPP Analyses Results Multi response permutation procedures (MRPP) for each month of analyses, comparing grazed and ungrazed sites.

Plants Bees Month A – Value p- value A – Value p- value

May -0.021 0.78 -0.0081 0.56

June -0.0078 0.57 -0.035 0.95 July -0.012 0.70 -0.015 0.71

Starkey August -0.0053 0.50 -- -- Sept -0.091 0.96 -0.015 0.71

June -0.010 0.76 -0.0065 0.67 July -0.0082 0.60 0.0028 0.36 August -0.011 0.62 0.030 0.10

Zumwalt

CHAPTER 3: FEED THE BEES AND SHADE THE STREAMS: FLOWERING SHRUBS PLANTED FOR RIPARIAN RESTORATION PROVIDE FORAGE FOR BEES

Scott R. Mitchell, Sandra J. DeBano, Mary M. Rowland, and Skyler Burrows

Abstract

With evidence of widespread declines in pollinator populations and an increased focus on habitat restoration, there is growing interest in investigating floral resources available to native bees. While woody shrubs can provide forage for native bees, relatively few studies have looked at the pollinator communities of native shrubs and none have examined plant-pollinator interactions in riparian restoration areas of the northwestern United States (US). We conducted extensive hand-net surveys over two years in a large restoration project on Meadow Creek in the US Forest Service Starkey Experimental Forest and Range to better understand bee and flowering plant interactions to help guide future restoration work. In early-season surveys conducted in April of 2018, we counted over 20,000 blooms on 27 plant species and observed 37 bee-flower interactions. Throughout the summers of 2018 and 2019, we collected nearly 2,000 bees from 150 species foraging on 92 plant species. We found that forb bloom abundance and richness is higher in the early-season, but 57% of bees observed on flowers in these surveys were foraging on willow (Salix spp.). We found that bee community composition on shrub species differed from communities found on forbs. On shrubs, we found that bee species diversity early in the season (April and May) was highest on wax currant (Ribes cereum) and willow (Salix sp.) and later in the season (June and July) was highest on black hawthorn (Crataegus douglasii) and mallow ninebark (Physocarpus malvaceus). Apparent specialist bees occurred on wax currant and black hawthorn while several rare bees were only found on mallow ninebark. Our results suggest that riparian restoration practitioners should consider targeting shrubs such as those listed above, that provide forage resources to a diverse community of pollinators, when planning restoration projects. Planting bee-friendly shrubs can help accomplish multiple restoration goals (e.g. stream shading and bee habitat improvement).

Introduction

Bees are of significant conservation interest due to their importance as pollinators in wildland and cropland pollination. While honey bees provide substantial pollination services in cultivated systems, native bees still contribute an estimated $3.07 billion dollars to pollination in these systems (Crane & Walker 1984; Losey & Vaughan 2006; Klein et al. 2007), and this number is likely an underestimate, given that valuation of pollination services can be challenging (Melathopoulos et al. 2015; Hanley et al. 2015). While honey bees are responsible for a greater share of pollination in agricultural systems, wild, native bees provide the bulk of pollination services in wildland systems (Aslan et al. 2016). With over 20,000 bee species worldwide and an estimated 4,000 or more in North America (Michener 2007), remarkably little is known about many bee populations and there is a need to characterize species diversity, abundance, population trends, and life histories (LeBuhn et al. 2013). To effectively manage for healthy populations of wild pollinators, it is increasingly important to understand plants and landscape features that can help maintain wild bee populations. More recently, observed declines in wild bee populations (Potts et al. 2010; Colla et al. 2012) have resulted in increased interest in managing and restoring habitat for wild bees. One habitat type that supports highly diverse and abundant native bee communities is riparian habitat (DeBano et al., 2016). Riparian zones are ecosystems that exist at the interface of water and land and can provide long linear corridors that connect habitat types across large geographic areas (Naiman et al. 1993; Galindo et al. 2017) and can support diverse invertebrate and pollinator communities (DeBano & Wooster 2004; Roof et al. 2018). Historically, these systems have been degraded through various land management practices including logging (Sweeney et al. 2004), livestock grazing (Schulz & Leininger 1990; Belsky et al. 1999), and stream channel alteration (Sweeney et al. 2004). With an estimated 71% of US streams and rivers in “poor” or “fair” condition, and lack of adequate vegetation as a primary stressor (US Environmental Protection Agency 2017), there is growing interest in restoring ecosystem structure and function of these systems (Goodwin et al. 1997; Bernhardt et al. 2005; Wohl et al. 2015), including those associated with pollinator health

(Williams 2011). In areas that have been heavily grazed or otherwise degraded, range managers may pursue habitat restoration projects, especially in sensitive habitats like riparian areas. Many approaches to riparian restoration emphasize the use of woody plants, including shrubs and trees, in riparian restoration (Guillozet et al. 2014; Averett et al. 2017b). Shrubs and trees are often planted because they benefit stream ecosystem health by shading streams (Wondzell et al. 2019), stabilizing banks (Hughes 2016), and providing and connecting habitat (Rockwell & Stephens 2018; Stanford et al. 2020). While riparian restoration objectives do not often include pollinator health, pollinators may benefit from restoration activities that involve planting flower- producing plants, including flowering shrubs. Although some studies have shown that bees can benefit from and use flowering shrubs (Morandin & Kremen 2013; Bareke et al. 2017; Bentrup et al. 2019), fewer studies have focused on flowering shrubs specifically as forage resources (Reddersen 2001; Dumroese & Luna 2016; Saunders 2018), and most have been conducted in agricultural contexts (Hannon & Sisk 2009; Kovács-Hostyánszki et al. 2013; Morandin & Kremen 2013, Bentrup et al. 2019). To our knowledge, no studies have yet examined bee communities on woody plants commonly used in riparian restoration in the Pacific Northwest of the US. By understanding the interactions between bees and shrubs in riparian habitats, restoration practices can be refined to not only restore streams and meet other riparian restoration goals, but to provide optimal forage resources for bees. Since financial and logistical support for conservation and restoration is limited (Holl & Howarth 2000; Iftekhar et al. 2017), combining multiple restoration goals is important for riparian restoration managers (González et al. 2017). Although land managers have not traditionally considered invertebrates in restoration planning, that is changing, and pollinators are of increasing conservation concern (Winfree 2010; Hanula et al. 2016). In addition to their season-long usefulness to riparian restoration, woody shrubs may be particularly valuable to bees as resources in spring, when many native bee species emerge from overwintering. This time of year is particularly critical for many bee species. For example, bumble bee queens are just emerging, and colony

size is dependent on the availability of floral resources (Williams et al. 2012). While the colony cycle is unique to bumble bees among native bee species in the US, there is some evidence that solitary bee abundance increases following years with higher floral abundance during emergence (Crone 2013); therefore, availability of floral resources in spring may be a key contributor to native bee fitness. Some shrub species, such as willow (Salix spp.), are already commonly planted in riparian restoration projects and grow readily near streams and rivers. While willows are often thought of as wind-pollinated, some species are visited by bees and actually depend of insects for pollination, and from the bee’s perspective, the value of wind-pollinated willows can be high, as long as their pollen and nectar contain sufficient nutrition (Saunders 2018). This study seeks to determine which shrubs may be good candidates for pollinator restoration by examining the interactions between bees and shrubs in an extensively restored riparian area on Meadow Creek in the United States Forest Service (USFS) Starkey Experimental Forest and Range in northeastern Oregon. The large-scale riparian restoration of Meadow Creek involved planting over 50,000 shrubs and trees, several of which produce blooms, along an 11 km section of the creek. The primary goal of the restoration was to benefit Endangered Species Act (ESA) listed salmonid fish (Averett et al. 2017a). Because of its scale and emphasis on shrubs, it presents an ideal opportunity to study shrub use by bees. To understand the relationships between bees and flowering shrubs in restored areas of Meadow Creek, we sought to answer the following questions: 1) How do the abundance and species richness of blooming shrubs compare to the abundance and species richness of blooming forbs early in the season? 2) Does bee visitation rate vary between shrub blooms and forb blooms early in the season? 3) When does shrub and forb bloom richness peak during the growing season and when are bees foraging on common shrub species in riparian areas? 4) Are there differences between shrubs and forbs in terms of the bee communities that interact with them and which bee species interact with which blooming species? and 5) Which shrubs support the highest diversity of bee species?

Methods

Study Area

This study was conducted in riparian meadow sites along 11 km of Meadow Creek in the USFS Starkey Experimental Forest and Range (Starkey) (Figure 3.1). The riparian area is approximately 157 ha that range in elevation from 1,100 – 1,200 m (Averett et al. 2017a, b). The climate and habitats at Starkey are typical of forested riparian systems in the Blue Mountains of eastern Oregon. Average annual precipitation at Starkey is 42 cm, most of which is received between November and June (NOAA Climatic Data Center 2010). This meadow system was part of a large- scale restoration project conducted in 2012 and 2013 focused on restoring stream function for the benefit of salmonid fishes. As part of the restoration efforts, over 50,000 woody plants were planted along the riparian area including flowering shrubs such as willows (Salix spp.), currants (Ribes spp.), and black hawthorn (Crataegus douglasii).

Plant and Bee Sampling

We addressed our first two questions in April 2018. To answer our first question concerning the relative abundance and species richness of blooming shrubs compared to blooming forbs early in the season, we conducted exhaustive searches of all shrub and forb blooms in 11 of 12 pastures along Meadow Creek in Starkey. During exhaustive searches, we systematically walked through pastures counting every bloom present. Each pasture survey took approximately one hour to complete, and all surveys were conducted between April 26-28, 2018. To answer our second question about whether bee visitation rates vary between shrub and forb blooms early in the season, any bee observed foraging during our exhaustive searches was counted (but not necessarily caught) and the species of plant on which the bee was foraging was noted. Bees were not always caught during this portion of the study as the priority was to estimate abundance of bees on each plant species and interrupting plant counts to capture bees could result in inaccurate bloom counts. This, in combination with the plant data collected during the exhaustive searches, allowed us

to calculate a visitation rate for each site (#bees/blooms/site). The 12th pasture was not sampled due to inclement weather (hailstorm) which obscured forb blooms growing low to the ground and eliminated the possibility of observing any bees that happened to be foraging during a hailstorm. To address our third question about the timing of peak shrub and forb bloom richness and the timing of bee foraging on common shrub species, we conducted 20- min presence/absence surveys of blooming plants throughout each of the 12 pastures once per month through the growing season (May, June, July and September of 2018) to document changes in plant species richness of each growth form (shrub or forb). Richness in April was determined based on the exhaustive search surveys. To address our fourth question about differences between shrubs and forbs in terms of the bee communities that interact with them, we sampled bees with hand- nets from flowering shrubs and forbs from April to September of 2018 and May to September of 2019 in the restored riparian area of Meadow Creek. We also used these data on bees collected from shrubs to address the second part of our third question regarding timing of bee foraging on shrub species. Bees were only collected when they were observed to be actively foraging on flowers and coming into contact with the reproductive parts of the flowers (anthers and stigma). To address our fifth question concerning which shrubs support the highest diversity of bee species, bees were collected during bouts (2018 and 2019) that targeted the most common blooming shrubs in the study area including: willows (Salix spp.), wax currants (Ribes cereum), snowberry (Symphoricarpos albus), elderberry (Sambucus nigra), mallow ninebark (Physocarpus malvaceus), red osier dogwood (Cornus sericea), wild rose (Rosa sp.) and black hawthorn (C. douglasii). We also collected bees on less common shrub species including spiraea (Spiraea beautifolia), thimbleberry (Rubus parviflorus), sticky currant (Ribes viscossissimum) and English hawthorn (Craetagus laevigata) whenever we encountered them blooming. All bees collected with hand-nets were euthanized with ethyl acetate, preserved for later species identification, and vouchered at OSU’s Hermiston Agricultural Research and Extension Center’s Invertebrate Ecology Laboratory collection.

Statistical Analyses

Early-season Communities

To address our first and second questions, we compared the early-season abundance, richness, and visitation rate of blooming forbs and shrubs at each site in April 2018 using Wilcoxon rank sum tests. Wilcoxon rank sum tests were used because data violated the assumption of heterogeneity of errors for parametric tests. Visitation rate of bees to shrub flowers was calculated by dividing the total number of bees observed visiting shrub flowers in a site by the total number of shrub flowers # 표푓 푏푒푒푠 푣𝑖푠𝑖푡𝑖푛푔 푠ℎ푟푢푏푠 that were counted in a site ( ). Visitation rate of bees to forb # 표푓 푠ℎ푟푢푏 푏푙표표푚푠 flowers was calculated in the same way. Wilcoxon rank sum tests were conducted in R, using R-Studio (R Core Team 2019).

Seasonal Phenology

To address our third question about the timing of peak shrub and forb bloom richness and the timing of bee foraging on common shrub species, we plotted boxplots of shrub and forb species richness by month. We used one-way analysis of variance tests (ANOVA) to determine if the average richness of blooms in a site differed significantly among months for each plant growth type. Pairwise comparisons of means for each month were made using Tukey’s honest significant differences tests (Tukey’s HSD). Data were checked and found to fit the assumptions of ANOVA prior to analyses. ANOVAs and Tukey’s HSD tests were conducted in R using R-Studio (R Core Team 2019), graphs were produced using the ggplot2 package (Wickham 2016) and arranged using the gridExtra package (Auguie 2017). To answer the second part of question 3, we noted the month in which each shrub species of interest was blooming and noted when bees were collected from each shrub species.

Bee and Plant Interactions

To answer the first part of our fourth question concerning whether there are differences between bee communities interacting with shrubs and forbs, we visualized bee communities foraging on shrub and forb flowers using non-metric multidimensional scaling procedures (NMS). Multi-response permutation procedures (MRPP) were used to determine if bee communities differed between forbs and shrubs. For both NMS and MRPP analyses, rare species were removed; for bees, any species with fewer than five specimens collected were removed, and for plants, any species with fewer than five bee observations were removed prior to analyses. This resulted in a matrix with 48 rows (plant species) and 58 columns (bee species). NMS analysis was conducted to show bloom species in bee species space and with the following options selected: Sorenson distance measure, a maximum of 500 iterations, and using random starting coordinates with a step length of 0.20. The analyses was conducted with 100 runs with real data and 500 runs with randomized data to generate a final 3-d configurations. All multivariate analyses were conducted using PC-ORD Version 7 (McCune & Mefford 2006). To answer the second part of our fourth question, we described bee and flower interactions by constructing a pollinator network using all data from both years. Bipartite networks (with bees on one side and plants on the other) allow for visualization of two node networks where box size of species of each trophic level (plants or bees) is proportional to the number of specimens collected and the lines connecting plants to bees is proportional to the number of observations of a link. To further explore the interactions between bees and blooms, we calculated a paired difference index (PDI) value for each bee species to look at degree of specialization. PDI describes the degree of specificity for bee species based on their interactions with plants. Since PDI incorporates interaction strength (# of observations) into calculations, it is preferable to other specialization indices (such as resource range) when quantitative data are available (Poisot et al. 2012). Resource range rankings are primarily influenced by species richness of plants that individual bees were visiting and were not used for these analyses. All network analyses were conducted using the bipartite package in R (Dormann et al. 2008, 2009).

To address our fifth question, we calculated the Shannon diversity of bee visitors to different shrub species. Shannon diversity was used instead of species richness of bee visitors because Shannon diversity accounts for relative abundance of different species. This is relevant to conservation because a shrub that evenly provides resources for a number species is more likely to support diverse bee fauna than one that is visited by an equal number of species, but only occasionally by most species. Shannon diversity was calculated in the bipartite package of R (Dormann et al. 2009).

Results

Over the course of 2018 and 2019, we collected 1,886 bees of 150 species (see Table A2.1) or morphospecies foraging on one of 16 species of shrubs (641 specimens) or on one of 76 species of forbs (1,245 specimens). Of the 150 species of bees collected, 30 were only observed on shrubs, 70 were only observed on forbs, and 50 were observed on both forbs and shrubs. Around 20% of bee species were only observed once during the sampling period (24 species on forbs, 13 species on shrubs).

Early-season Communities

In early spring sampling (April 2018), we counted 20,831 blooms from 27 plant species (25 forb species, 2 shrub species) and observed 37 bees foraging on those flowers during exhaustive searches of Meadow Creek sites. In April 2018, forb blooms were more abundant at each site than shrub blooms (Wilcoxon W = 120, p < 0.001) and forbs had a higher blooming species richness than shrubs (Wilcoxon W = 121, p < 0.001) (Figure 3.3). However, there was not a significant difference in the number of bees per forb flower and the number of bees per shrub flowers in the 11 surveyed sites (Wilcoxon W = 76, p = 0.66) (Figure 3.3). While there was not a significant difference in the number of bees per forb vs. shrub bloom, it is notable that of the 37 bees observed during early-season surveys, 57% were observed visiting willow flowers (Salix spp.) even though willow made up less than 7% of total observed blooms (Figure 3.4).

Seasonal Phenology

Blooming shrub species richness peaked in May. While the average richness of shrub blooms was statistically higher in May than in September, it was not significantly higher than June or July (Figure 3.2). Blooming forb richness peaked in June. While average richness was significantly higher in June than in April, May, and September, it was not significantly higher than blooming forb species richness in July (Figure 3.2). Bees were collected on blooming shrubs in April, May, June, and July (Table 3.1). No shrubs were observed blooming after July. In April and May, most bees were collected on willow (Salix spp.) and wax currant (Ribes cereum). In June, most bees were collected on black hawthorn (Craetagus douglasii), wild rose (Rosa sp.), red- osier dogwood (Cornus sericea), and snowberry (Symphoricarpos albus), but were also caught on five other species in fewer numbers. In July, most bees were collected on blue elderberry (Sambucus nigra) and snowberry (Symphoricarpos albus) but were also caught on five other shrub species in fewer numbers.

Bee and Plant Interactions

Community analyses of blooming shrub species in bee community space revealed that bee communities foraging on forbs and shrubs were distinct from each other (MRPP: A = 0.02, p < 0.001). Final stress of the 3-dimensional ordination was 12.6 and the R2n (nonmetric fit) for the data by the ordination was 0.98. The I-value for final ordination was 0.6 (0 = random, 1 = perfect fit), which is considered quite good for community data (McCune et al. 2002). Axis 1 explained 23% of the variation present in the data, axis 2 explained 12%, and axis 3 explained 10%. NMS ordinations showed some separation between shrub and forb species on axis 1 and on axis 3 (Figure 3.5). Shrub species were associated with positive values on axis 1 and thus were associated with several Andrena and Osmia species and negative values on axes 2 and 3 and thus were associated with several Bombus, Lasioglossum, and Osmia species (Figure 3.3, Table 3.3). Forb species were associated with negative values on axis 1 and positive values on axes 2 and 3 and were thus associated with several species from the following genera: Melissodes, Lasioglossum, and Halictus.

Network analyses showed that bees exhibit a mix of foraging patterns with some species exhibiting strong preferences or specialist interactions with certain plants and other species exhibiting more generalist interactions. For the 35 bee species that had greater than 10 specimens collected (named in the network diagram, Figure 3.6), the five most specialized bee species were Andrena porterage (15 specimens, 100% on wax currant), Andrena flocculosa (13 specimens, 100% on red osier dogwood), Anthophora pacifica (104 specimens, 97% on wax currant), Diadasia nigrifrons (25 specimens, 92% on Oregon checker-mallow), and Bombus vosnesenskii (10 specimens, 80% on bull thistle). The most generalist species observed were Bombus mixtus (101 specimens, collected on 25 plant species), Andrena salicifloris (13 specimens, collected on 7 plant species), Bombus flavifrons (141 specimens, collected on 38 plant species), Hoplitis fulgida (12 specimens, collected on 8 plant species), and Bombus centralis (55 specimens, collected on 21 plant species). Some species that were ranked as more generalist species, while visiting multiple plant species, primarily visited plants of particular genera and thus could be considered oligolectic (specializing on a few closely related plants). One example of this is Andrena angustitarsata which was caught on 11 different plant species; however, 78% of specimens were caught on one of four species of willow, suggesting some level of oligolecty or preference for willows (see Table A2.1 complete list of PDI scores and Figure A2.1 for diet breadth of the top 35 bees). Andrena salicifloris was also a species that was ranked as a generalist, but 46% of all specimens were collected on one or three willow species and the remaining 54% were collected on one of four other plant species.

Discussion

While other studies have explored the relationships between bees and shrubs in other parts of the world (Reddersen 2001; Saunders 2018) and in agricultural contexts (Hannon & Sisk 2009; Kovács-Hostyánszki et al. 2013), this study is the first of its kind to explore in-depth relationships between bees and shrubs in wildlands in the northwestern US. Our results indicate that a species-rich community of native bees forage on a diverse community of blooming forbs and shrubs in restored riparian

areas. The diversity of floral resources varies through the season and, as other authors have indicated, shrubs are an important forage resource for bees and should be considered in plantings (Saunders 2018; Bentrup et al. 2019). We found that a diverse and, in many cases, distinct community of bees visit shrubs in a riparian restoration area, with several species of Andrena seeming to exhibit specialist interactions with willows early in the season. Later in the season, shrubs such as currants, hawthorns, and snowberry can support diverse bees for most of the growing season. Perhaps most importantly we found that while shrubs produce significantly fewer total blooms of significantly fewer species than forbs in the early-season, certain species such as willow are visited at high rates, especially by certain species of bees. In fact, willow was the plant upon which most bees were observed during exhaustive searches in April 2018. Part of the reason for this could be that while willow did not bloom in many of our pastures, when it was blooming, there was a concentration of blooms in one area that had abundant nectar and pollen. Willow pollen has been found to have comparable availability of protein relative to other forb and shrub blooms (Weiner et al. 2010) and may be why bees frequently forage on it when these other shrubs and forbs are not yet blooming. Of the many forb species observed in April, a large number were small annuals such as spring draba (Draba verna) and blue-eyed Mary (Collinsia parviflora) that likely do not provide much nectar or pollen for bees (Pywell et al. 2005). Future studies could compare the pollen protein content, nectar volume, and sugar content of early spring blooming forbs to further explore this relationship. When considering restoration in riparian areas, land managers should also consider the palatability of plants to ungulate herbivores. Ungulate herbivores are capable of depressing shrub establishment and growth in riparian areas, including at Starkey (Averett et al. 2017a, 2019). Additionally, since bees may have a significant dietary overlap with cattle and native ungulates, in particular with forbs that bees forage on, unpalatable shrubs could be a useful addition to restoration projects (DeBano et al. 2016). Some studies have indicated that shrubs such as hawthorn and currants are not palatable to ungulates, and tend to be avoided in favor of other plants (Holechek et al. 1982), a hypothesis consistent with studies showing these species

survive better than other shrub species in riparian restoration projects in the presence of ungulate herbivory (Averett et al. 2017b). In the early-season, 30 species of bees were collected on wax currant (Ribes cereum), and later in the season 24 species were collected on black hawthorn (Crataegus douglasii) and between the two plants, 50 bee species (or one-third of total observed species richness) were collected. Since both shrubs are long-lived, hearty to ungulate herbivory, and provide forage to a diverse community of bees, they would be worth considering in riparian restoration projects. Our results represent an important first step in assessing the value of riparian restorations to pollinator communities. While our work makes important observations of bees in riparian communities, future work should assess pollinator communities pre- and post-restoration to see if shrub plantings are increasing richness, diversity and abundance of bees in riparian areas. Other studies should be conducted to more closely examine specialist interactions. One such way to do this would be to make use of combined behavioral and metabarcoding data (as in Arstingstall et al., in review).

Literature Cited

Aslan CE, Liang CT, Galindo B, Hill K, Topete W (2016) The role of honey bees as pollinators in natural areas. Natural Areas Journal 36: 478-88 Auguie B (2017) GridExtra: Miscellaneous Functions for “Grid” Graphics (version 2.3). R. https://CRAN.R-project.org/package=gridExtra Averett JP, Endress BA, Rowland MM, Naylor BJ, Wisdom MJ (2017a) Wild ungulate herbivory suppresses deciduous woody plant establishment following salmonid stream restoration. Forest Ecology and Management 391:135-144 Averett JP, Wisdom MJ, Endress BA (2019) Livestock riparian guidelines may not promote woody species recovery where wild ungulate populations are high. Rangeland Ecology & Management 72: 145-49 Averett JP, Wisdom MJ, Naylor BJ, Rowland MM, Endress BA (2017b) Data and analyses of woody restoration planting survival and growth as a function of wild ungulate herbivory. Data in Brief 14: 168-74 Bareke T, Addi A, Wakjira K (2017) Screening of potential shrubs for bee forage development. Journal of Plant Science 5:160-164 Belsky AJ, Matzke A, Uselman S (1999) Survey of livestock influences on stream and riparian ecosystems in the western United States. Journal of Soil and Water Conservation 54: 419-31 Bentrup G, Hopwood J, Adamson NL, Vaughan, M (2019) Temperate agroforestry systems and insect pollinators: A review. Forests 10: 981-1001 Bernhardt ES, Palmer MA, Allan JD, Alexander G, Barnas K, Brooks S, Carr J, Clayton S, Dahm C, Follstad-Shah J, Galat D, Gloss S, Goodwin P, Hart D, Hassett B, Jenkinson R, Katz S, Kondolf GM, Lake PS, Lave R, Meyer JL, O’Donnell TK, Pagano L, Powell B, Sudduth E (2005) Synthesizing U.S. river restoration efforts. Science 308:308-309 Colla SR, Gadallah F, Richardson L, Wagner D, Gall L (2012) Assessing declines of North American bumble bees (Bombus spp.) using museum specimens. Biodiversity and Conservation 21: 3585-3595

Crone, EE (2013) Responses of social and solitary bees to pulsed floral resources. The American Naturalist 182:465-73 DeBano SJ, Roof SM, Rowland MM, Smith LA (2016) Diet overlap of mammalian herbivores and native bees: Implications for managing co-occurring grazers and pollinators. Natural Areas Journal 36:458-477 DeBano SJ & Wooster DE (2004) Chapter 10: Insects and other invertebrates: Ecological roles and indicators of riparian and stream health. Pages 215-237 In: Baker MB, Ffolliott PF, DeBano LF, Neary DG (1st Ed) Riparian Areas of the Southwestern United States: Hydrology, Ecology, and Management. CRC Press, Washington, D.C. Dormann CF, Fründ J, Blüthgen N, Gruber B (2009) Indices, graphs and null models: Analyzing bipartite ecological networks. The Open Ecology Journal 2:7-24 Dormann CF, Gruber B, Fründ J (2008) Introducing the bipartite package: Analysing ecological networks. R News 8:8-11 Dumroese RK, & Luna T (2016) Growing and marketing woody species to support pollinators: An emerging opportunity for forest, conservation, and native plant nurseries in the northeastern United State. 59:49-60 Crane E & Walker P (1984) Pollination directory for world crops. International Bee Research Association, London, UK Galindo V, Calle Z, Chará J, Armbrecht I (2017) Facilitation by pioneer shrubs for the ecological restoration of riparian forests in the central Andes of Colombia. Restoration Ecology 25:731-37 González E, Felipe-Lucia MR, Bourgeois B, Boz B, Nilsson C, Palmer G, Sher AA (2017) Integrative conservation of riparian zones. Biological Conservation 211:20-29 Goodwin CN, Hawkins CP, Kershner JL (1997) Riparian restoration in the western United States: Overview and perspective. Restoration Ecology 5:4-14 Guillozet P, Smith K, Guillozet K (2014) The rapid riparian revegetation approach. Ecological Restoration 32:113-124

Hanley H, Breeze TD, Ellis C, Goulson D (2015) Measuring the economic value of pollination services: Principles, evidence and knowledge gaps. Ecosystem Services 14:124-132. Hannon LE & Sisk TD (2009) Hedgerows in an agri-natural landscape: Potential habitat value for native bees. Biological Conservation 142: 2140-2154 Hanula JL, Ulyshen MD, Horn S (2016) Conserving pollinators in North American forests: A review. Natural Areas Journal 36: 427-39 Holechek JL, Vavra M, Skovlin J, Krueger WC (1982) Cattle diets in the Blue Mountains of Oregon II forests. Journal of Range Management 35:239-242 Holl KD & Howarth RB (2000) Paying for restoration. Restoration Ecology 8: 260- 267 Hughes AO (2016) Riparian management and stream bank erosion in New Zealand. New Zealand Journal of Marine and Freshwater Research 50:277-290 Iftekhar MS, Polyakov M, Ansell D, Gibson F, Kay GM (2017) How economics can further the success of ecological restoration. Conservation Biology 31:261- 268 Klein AM, Vaissière BE, Cane JH, Steffan-Dewenter I, Cunningham SA, Kremen C, Tscharntke T (2007) Importance of pollinators in changing landscapes for world crops. Proceedings: Biological Sciences 274:303-313 Kovács-Hostyánszki A, Haenke S, Batáry P, Jauker B, Báldi A, Tscharntke T, Holzschuh A (2013) Contrasting effects of mass-flowering crops on bee pollination of hedge plants at different spatial and temporal scales. Ecological Applications 23:1938-1946 LeBuhn G, Droege S, Connor EF, Gemmill-Herren B, Potts SG, Minckley RL, Griswold T, Jean R, Kula E, Roubik DW, Cane J, Wright KW, Frankie G, Parker F (2013) Detecting insect pollinator declines on regional and global scales: Detecting pollinator declines. Conservation Biology 27:113-120 Losey JE & Vaughan M (2006) The economic value of ecological services provided by insects. BioScience 56:311-323 McCune B, Grace JB, Urban DL (2002) Analysis of ecological communities. 1st ed. MjM Software Design

McCune B & Mefford MJ (2006) PC-ORD: Multivariate Analysis of Ecological Data. MjM Software, Gleneden Beach, Oregon, USA Melathopoulos AP, Cutler GC, Tyedmers P (2015) Where is the value in valuing pollination ecosystem services to agriculture? Ecological Economics 109: 59– 70. Michener CD (2007) The bees of the world. 2nd ed. Johns Hopkins University Press, Baltimore, Maryland Morandin LA & Kremen C (2013) Bee preference for native versus exotic plants in restored agricultural hedgerows. Restoration Ecology 21: 26-32 Naiman RJ, Decamps H, Pollock M (1993) The role of riparian corridors in maintaining regional biodiversity. Ecological Applications 3: 209-212 National Oceanic and Atmospheric Administration (1981-2010) National Climatic Data Center https://catalog.data.gov/dataset/u-s-hourly-climate-normals-1981- 2010 (accessed 18 Dec 2019) National water quality inventory: Report to Congress (2017) EPA 841-R-16-011. National Water Quality Inventory. Environmental Protection Agency, Washington, D.C. Poisot T, Canard E, Mouquet N, Hochberg ME (2012) A comparative study of ecological specialization estimators: Species-level specialization. Methods in Ecology and Evolution 3:537-544 Potts SG, Biesmeijer JC, Kremen C, Neumann P, Schweiger O, Kunin WE (2010) Global pollinator declines: trends, impacts and drivers. Trends in Ecology & Evolution 25:345-353 Pywell RF, Warman EA, Carvell C, Sparks TH, Dicks LV, Bennett D, Wright A, Critchley CNR, Sherwood A (2005) Providing foraging resources for bumblebees in intensively farmed landscapes. Biological Conservation 121: 479-494 R Core Team. 2019. R: A Language and Environment for Statistical Computing. R. Vienna, Austria: R Foundation for Statistical Computing. https://www.R- project.org/

Reddersen J (2001) SRC-willow (Salix viminalis) as a resource for flower-visiting insects. Biomass and Bioenergy 20:171-179 Rockwell SM & Stephens JL (2018) Habitat selection of riparian birds at restoration sites along the Trinity River, California. Restoration Ecology 26:767-777 Roof SM, DeBano SJ, Rowland MM, Burrows S (2018) Associations between blooming plants and their bee visitors in a riparian ecosystem in eastern Oregon. Northwest Science 92:119-135 Saunders ME (2018) Insect pollinators collect pollen from wind-pollinated plants: Implications for pollination ecology and sustainable agriculture. Insect Conservation and Diversity 11:13–31 Schulz TT & Leininger WC (1990) Differences in riparian vegetation structure between grazed areas and exclosures. Journal of Range Management 43:295- 299 Stanford B, Holl KD, Herbst DB, Zavaleta E (2020) In-stream habitat and macroinvertebrate responses to riparian corridor length in rangeland streams. Restoration Ecology 28:173-184 Sweeney BW, Bott TL, Jackson JK, Kaplan LA, Newbold JD, Standley LJ, Hession WC, Horwitz LJ (2004) Riparian deforestation, stream narrowing, and loss of stream ecosystem services. Proceedings of the National Academy of Sciences 101:14132-14137. Weiner CN, Hilpert A, Werner M, Linsenmair KE, Blüthgen N (2010) Pollen amino acids and flower specialisation in solitary bees. Apidologie 41: 476–487 Wickham H. 2016. Ggplot2: Elegant Graphics for Data Analysis. New York: Springer-Verlag. https://ggplot2.tidyverse.org Williams NM (2011) Restoration of nontarget species: Bee communities and pollination function in riparian forests. Restoration Ecology 19: 450-459 Williams NM, Regetz J, Kremen C (2012) Landscape-scale resources promote colony growth but not reproductive performance of bumble bees. Ecology 93:1049- 1058 Winfree R (2010) The conservation and restoration of wild bees: Wild bee conservation. Annals of the New York Academy of Sciences 1195:169-197

Wohl E, Lane SN, Wilcox AC (2015) The science and practice of river restoration. Water Resources Research 51:5974-5997 Wondzell SM, Diabat M, Haggerty, R (2019) What matters most: Are future stream temperatures more sensitive to changing air temperatures, discharge, or riparian vegetation? Journal of the American Water Resources Association 55:116-132

Figures and Tables:

Figure 3.1: Study sites and native bees. The United States Forest Service Starkey Experimental Forest and Range located in northeastern Oregon: a) map of Meadow Creek with approximate sampling locations indicated. Black arrows indicate location of sites used in plant sampling efforts, blue arrows indicate additional sites where significant bee sampling occurred; b) planted one-color willow (Salix monochroma) in bloom; c) bee foraging on one-color willow (Salix monochroma) in April; d) Bee foraging on snowberry (Symphoricarpos albus). *plant sampling site not sampled during early season foraging bouts due to weather.

Figure 3.2: Seasonal variation in forb and shrub richness Seasonal variation in forb and shrub species richness at 12 sites along Meadow Creek in 2018. Significant group differences (ANOVA and Tukey’s HSD, p < 0.05) are indicated by different letters above boxplots.

Figure 3.3: Average bloom abundance, species richness, and number of bees/flower Average bloom abundance, species richness, and number of bees/flower of forbs and shrubs per site along Meadow Creek in April 2018.

Figure 3.4: Bloom proportions and bee visitations for April 2018 Percent of total bees observed (left) and percent of total blooms count (right) observed in 11 sites during April 2018. Plants with no observed bee visits and making up less than 5% of the total count of blooms are not shown.

Figure 3.5: Blooming plant species in bee community space

Axis 1 versus axis 2 and axis 1 versus axis 3 are shown in the ordination. Percent of variation explained by each axes is shown on the graph. Points are plant species and are labelled with USDA plant codes. The bee communities observed foraging on shrubs and forbs were significantly different from each other based on MRPP results (A = 0.02, p < 0.001). Bee species with r2 > 0.1 correlation to each axis are shown.

Figure 3.6: Season-long network of bee and plant interactions

Network of bee and plant interactions for 150 bee species and 92 plant species (16 shrubs, 76 forbs). Names of bee and plant species with 10 or more observations are displayed. Box height and lines connecting bee and plant species are proportional to the number of observations.

Figure 3.7: Shannon diversity of bee visitors to shrubs occurring at Meadow Creek.

Tables: Table 3.1: Phenology of shrub blooms. Numbers of bees collected on each of 16 shrub species blooming in April, May, June, and July of 2018 and 2019.

Common Name Species name April May June July Willow sp. Salix sp. 23 20 - - Wax currant Ribes cereum 6 167 3 - Geyer’s willow Salix geyeriana - 23 - - Onecolor willow Salix monochroma - 19 - - Lemmon’s willow Salix lemmonii - 12 - - Shining willow Salix lasiandra - 3 - - Black hawthorn Crataegus douglasii - - 101 - Wild rose Rosa sp. - - 46 5 Red-osier dogwood Cornus sericea - - 40 3 Mallow ninebark Physocarpus malvaceus - - 15 2 English hawthorn Crataegus laevigata - - 2 - Sticky currant Ribes viscosissimum - - 1 - Blue elderberry Sambucus nigra - - 8 53 Snowberry Symphoricarpos albus - - 39 48 Thimbleberry Rubus parviflorus - - - 1 White spiraea Spiraea betulifolia - - - 1 Grand Total 29 244 255 113

Table 3.2 Pearson correlations for plant ordinations Pearson correlations for plant ordination axis 1, axis 2, and axis 3. Species correlated with axis with r2 higher than 0.1. Cells are highlighted in grey to show correlations with the axis of interest in the NMS ordinations shown in Figure 3.5.

Axis 1 Axis 2 Axis 3 Bee Species r r2 r r2 r r2 Melissodes rivalis -0.35 0.12 -0.13 0.02 0.01 < 0.01 Andrena microchlora 0.34 0.12 0.18 0.03 0.04 < 0.01 Osmia lignaria 0.36 0.13 -0.11 0.01 -0.17 0.03 Andrena salicifloris 0.52 0.27 -0.18 0.03 -0.16 0.03 Andrena angustitarsata 0.61 0.37 -0.1 0.01 -0.12 0.02

Halictus tripartitus -0.21 0.05 0.33 0.11 0.08 0.01 Halictus ligatus -0.26 0.07 0.33 0.11 -0.05 < 0.01 Lasioglossum sp. -0.19 0.04 0.42 0.18 -0.17 0.03 Lasioglossum egregium -0.27 0.07 0.44 0.19 0.01 < 0.01 Lasioglossum incompletum -0.07 0.01 0.48 0.23 -0.17 0.03

Bombus flavifrons -0.22 0.05 -0.19 0.04 -0.41 0.17 Bombus mixtus -0.19 0.04 -0.03 < 0.01 -0.38 0.14 Bombus centralis -0.29 0.08 -0.17 0.03 -0.34 0.11 Lasioglossum egregium < 0.01 < 0.01 0.09 0.01 -0.34 0.11 Osmia bucephala -0.13 0.02 -0.08 0.01 -0.33 0.11

CHAPTER 4: GENERAL CONCLUSIONS Bees are critical members of agricultural and natural ecosystems because of the role they play in animal-mediated pollination. In North America, native bee species make up the vast majority of bee species richness. While colony collapse disorder has affected non-native honey bees and attracted abundant media and scientific attention, recent observed declines in native bee species has increased the importance of understanding how land management actions affect these species. Understanding the stressors that affect bee communities and land management actions that may benefit bees are both areas of ongoing research that will contribute to the long-term sustainability of land management for bees.

One potentially important stressor to native bees is cattle grazing. We examined the effects of cattle grazing at 28 sites in two eastern Oregon locations throughout the growing season of 2018. One location was in Pacific Northwest bunchgrass prairie (The Nature Conservancy’s Zumwalt Prairie Preserve) and the other was a riparian meadow in a forested system (the US Forest Service Starkey Experimental Forest and Range). While we found strong temporal and spatial patterns structuring bee and plant communities, we did not detect significant responses of the blooming plant community to short-term grazing at the Zumwalt. Short-term grazing resulted in higher bee abundance and richness in July, but not higher Shannon diversity. It is possible that the increased richness and abundance of bees was due to higher trap efficiency in grazed sites, and so may not reflect increases in bee abundance in response to grazing. We also did not detect long-term significant effects of low-moderate intensity, rotational, late-season grazing on blooming plant or bee communities at the Zumwalt. At Starkey we found reduced bloom abundance, species richness, and species diversity in sites that had been recently grazed but did not detect any significant effect on bee communities. These results contrast with other studies in the region that have found cattle grazing to negatively impact bee populations, and we suggest late-season rotational grazing may be an effective way to mitigate negative impacts of grazing on bee communities in inland Northwest grasslands, potentially by minimizing dietary overlap between bees and cattle.

Riparian restoration represents a potential opportunity to restore forage resources for bees while simultaneously accomplishing stream restoration objectives. We conducted extensive hand-net surveys of bees and blooming plants in a riparian restoration area to understand how riparian restoration may benefit bee communities. We focused on the interactions between bees and flowering shrubs, which are commonly planted to shade streams and revegetate habitat. We found that early in the bloom season (April), blooming forbs outnumbered shrubs in both abundance of blooms and richness of species. While forbs dominated early-season blooming communities, most of the foraging bees we observed were on shrub flowers. Willow seemed to be important to supporting several apparent specialist bee foragers in the early-season. One important finding was that wax currant and black hawthorn have complementary bloom times and included over 1/3 of total species richness observed over the two years of study. Other researchers have indicated that these plants are less palatable to ungulate herbivores and may succeed in areas with intense browsing pressure. Since these two plant species are more resistant to browsing and provide forage for such a significant part of the bee community, both should be considered by land managers looking to enhance stream habitat and provide resources for bees.

The results of these studies indicate that late-season cattle grazing and riparian restoration activities may benefit bees and should be considered by land managers. With regards to riparian restoration, shrubs and forbs make up important parts of the floral landscape and both are used by a diversity of bees. Restoration practitioners should consider planting shrubs and forbs to create high quality, season-long forage for native bees. When restoration goals include shading streams or revegetation with shrubs, flowering shrubs such as willow, hawthorn, wax currant, and dogwood may be valuable to accomplish multiple restoration goals in the Pacific Northwest by also providing abundant, season-long forage for bees.

Some considerations for future research directions include sampling questions, such as how trapping method may influence abundance estimates of pollinators, and biological questions such as how the effects of grazing in riparian areas will differ from drier grasslands when grazing occurs over many years. Additionally, experimental studies should be conducted on riparian restoration in

streams to determine how limited funding can be maximized to meet both stream and pollinator health goals. Such experiments could include examining bee communities in sites before and after the application of restoration treatments which utilize different planting mixes.

General Bibliography

Auguie B (2017) GridExtra: Miscellaneous Functions for “Grid” Graphics (version 2.3). R. https://CRAN.R-project.org/package=gridExtra Aslan CE, Liang CT, Galindo B, Hill K, Topete W (2016) The role of honey bees as pollinators in natural areas. Natural Areas Journal 36: 478-88 Averett JP, Endress BA, Rowland MM, Naylor BJ, Wisdom MJ (2017a) Wild ungulate herbivory suppresses deciduous woody plant establishment following salmonid stream restoration. Forest Ecology and Management 391:135-144 Averett JP, Wisdom MJ, Endress BA (2019) Livestock riparian guidelines may not promote woody species recovery where wild ungulate populations are high. Rangeland Ecology & Management 72: 145-49 Averett JP, Wisdom MJ, Naylor BJ, Rowland MM, Endress BA (2017b) Data and analyses of woody restoration planting survival and growth as a function of wild ungulate herbivory. Data in Brief 14: 168-74 Ballantyne G, Baldock KCR, Willmer PG (2015) Constructing more informative plant–pollinator networks: Visitation and pollen deposition networks in a heathland plant community. Proceedings of the Royal Society B: Biological Sciences 282: 11-30 Bareke T, Addi A, Wakjira K (2017) Screening of potential shrubs for bee forage development. Journal of Plant Science 5:160-164 Baker S, Bohle M, Bouck MJ, Cardon G, Cash D, Cheyney C, Church J, Crane K, Dreves A, Engel C, Falen C, Fife T, Fisher G, Fransen S, Gerrish J, Gillaspy B, Glaze JB, Graw W, Griggs T, Hall J, Hannaway D, Hooper L, Horneck D, Hulting A, Jensen K, Jensen S, Koenig RT, McLain B, Neher O, Neibling H, Ogle D, Shewmaker G, St. John L, Whitesides R, Williams S, Wilson R (2010) Pasture and grazing management in the Northwest. University of Idaho Extension PNW 614 Bartuszevige AM, Kennedy PL, Taylor RV (2012) Sixty-seven years of landscape change in the last, large remnant of the Pacific Northwest bunchgrass prairie. Natural Areas Journal 32:166-170

Belsky AJ, Matzke A, Uselman S (1999) Survey of livestock influences on stream and riparian ecosystems in the western United States. Journal of Soil and Water Conservation 54: 419-31 Bentrup G, Hopwood J, Adamson NL, Vaughan, M (2019) Temperate agroforestry systems and insect pollinators: A review. Forests 10: 981-1001 Bernhardt ES, Palmer MA, Allan JD, Alexander G, Barnas K, Brooks S, Carr J, Clayton S, Dahm C, Follstad-Shah J, Galat D, Gloss S, Goodwin P, Hart D, Hassett B, Jenkinson R, Katz S, Kondolf GM, Lake PS, Lave R, Meyer JL, O’Donnell TK, Pagano L, Powell B, Sudduth E (2005) Synthesizing U.S. river restoration efforts. Science 308:308-309 Black SH, Shepherd M, Vaughan M (2011) Rangeland management for pollinators. Rangelands 33: 9-13 Borman MM (2005) Forest stand dynamics and livestock grazing in historical context. Conservation Biology 19:1658-1662 Bosch J & Kemp WP (2005) Alfalfa leafcutting bee population dynamics, flower availability, and pollination rates in two Oregon alfalfa fields. Journal of Economic Entomology 98:1077–1086 Broussard M, Rao S, Stephen WP, White L (2011) Native bees, honeybees, and pollination in Oregon cranberries. HortScience 46: 885–888 Butler LD, Cropper JB, Johnson RH, Norman AJ, Peacock GL, Shaver PL, Spaeth KE, Pierson Jr. FB, Weltz MA (2003) The national range and pasture handbook (NRPH). NRCS 190-VI-NRPH, rev. 1 Cane JH (2011) Meeting wild bees’ needs on Western US rangelands. Rangelands 33:27-32 Colla SR, Gadallah F, Richardson L, Wagner D, Gall L (2012) Assessing declines of North American bumble bees (Bombus spp.) using museum specimens. Biodiversity and Conservation 21: 3585-3595 Crane E & Walker P (1984) Pollination directory for world crops. International Bee Research Association, London, UK Crone EE (2013) Responses of social and solitary bees to pulsed floral resources. The American Naturalist 182:465-73

DeBano SJ, Roof SM, Rowland MM, Smith LA (2016) Diet overlap of mammalian herbivores and native bees: Implications for managing co-occurring grazers and pollinators. Natural Areas Journal 36:458-477 DeBano SJ & Wooster DE (2004) Chapter 10: Insects and other invertebrates: Ecological roles and indicators of riparian and stream health. Pages 215-237 In: Baker MB, Ffolliott PF, DeBano LF, Neary DG (1st Ed) Riparian Areas of the Southwestern United States: Hydrology, Ecology, and Management. CRC Press, Washington, D.C. DiCarlo Smith LA, DeBano SJ, Burrows S (2019) Short-term response of two beneficial invertebrate groups to wildfire in an arid grassland system, United States. Rangeland Ecology & Management 72:551–560 Dormann CF, Fründ J, Blüthgen N, Gruber B (2009) Indices, graphs and null models: Analyzing bipartite ecological networks. The Open Ecology Journal 2:7-24 Dormann CF, Gruber B, Fründ J (2008) Introducing the bipartite package: Analysing ecological networks. R News 8:8-11 Dumroese RK, & Luna T (2016) Growing and marketing woody species to support pollinators: An emerging opportunity for forest, conservation, and native plant nurseries in the northeastern United State. 59:49-60 Ebeling A, Klein A-M, Schmacher J, Weisser WW, Tscharntke T. (2008) How does plant richness affect pollinator richness and temporal stability of flower visits? Oikos 117:1808-1815 Ehrlén J & Lehtilä K (2002) How perennial are perennial plants? Oikos 98:308-322 Crane E & Walker P (1984) Pollination directory for world crops. International Bee Research Association, London, UK Fleischner TL (1994) Ecological costs of livestock grazing in Western North America. Conservation Biology 8:629-644 Freilich JE, Emlen JM, Duda JJ, Freeman DC, Carfaro PJ (2003) Ecological effects of ranching: A six-point critique. BioScience 53:759-765 Galbraith WA & Anderson EW (1991) Grazing history of the Northwest. Rangelands 13:213-218

Galindo V, Calle Z, Chará J, Armbrecht I (2017) Facilitation by pioneer shrubs for the ecological restoration of riparian forests in the central Andes of Colombia. Restoration Ecology 25:731-37 Garibaldi LA, Steffan-Dewenter I, Kremen C, Morales JM, Bommarco R, Cunningham SA, Carvalheiro LG, Chacoff NP, Dudenhöffer JH, Greenleaf SS, Holzschuh A, Isaacs R, Krewenka K, Mandelik Y, Mayfield MM, Morandin LA, Potts SG, Ricketts TH, Szentgyörgyi H, Viana BF, Westphal C, Winfree R, Klein AM (2011) Stability of pollination services decreases with isolation from natural areas despite honey bee visits: Habitat isolation and pollination stability. Ecology Letters 14:1062-1072 Golet GH, Gardali T, Hunt JW, Koenig DA, Williams NM (2011) Temporal and taxonomic variability in response of fauna to riparian restoration. Restoration Ecology 19:126-135 González E, Felipe-Lucia MR, Bourgeois B, Boz B, Nilsson C, Palmer G, Sher AA (2017) Integrative conservation of riparian zones. Biological Conservation 211: 20-29 Goodwin CN, Hawkins CP, Kershner JL (1997) Riparian restoration in the western United States: Overview and perspective. Restoration Ecology 5: 4-14 Guillozet P, Smith K, Guillozet K (2014) The rapid riparian revegetation approach. Ecological Restoration 32:113-124 Hallman CA, Sorg M, Jongejans E, Siepel H, Hofland N, Schwan H, Stenmans W, Müller A, Sumser H, Hörren T, Goulson D, de Kroon H (2017) More than 75 percent decline over 27 years in total flying insect biomass in protected areas. PLOS One 10:1-21 Hanley H, Breeze TD, Ellis C, Goulson D (2015) Measuring the economic value of pollination services: Principles, evidence and knowledge gaps. Ecosystem Services 14:124-132. Hannon LE & Sisk TD (2009) Hedgerows in an agri-natural landscape: Potential habitat value for native bees. Biological Conservation 142: 2140-2154 Hanula JL, Ulyshen MD, Horn S (2016) Conserving pollinators in North American forests: A review. Natural Areas Journal 36: 427-39

Harmon-Threatt A (2020) Influence of nesting characteristics on health of wild bee communities. Annual Review of Entomology 65:39-56 Hatfield RG & LeBuhn G (2007) Patch and landscape factors shape community assemblage of bumble bees, Bombus spp. (Hymenoptera: Apidae), in montane meadows. Biological Conservation 139:150-158 Holechek JL, Vavra M, Skovlin J, Krueger WC (1982) Cattle diets in the Blue Mountains of Oregon II forests. Journal of Range Management 35:239-242 Holl KD & Howarth RB (2000) Paying for restoration. Restoration Ecology 8: 260- 267 Holzschuh A, Dudenhöffer J-H, Tscharntke T (2012) Landscapes with wild bee habitats enhance pollination, fruit set and yield of sweet cherry. Biological Conservation 153:101-107 Hsieh TC, Ma KH, Chao A (2016) iNEXT: an R package for rarefaction and extrapolation of species diversity (Hill numbers). Methods in Ecology and Evolution 7:1451-1456 Hughes AO (2016) Riparian management and stream bank erosion in New Zealand. New Zealand Journal of Marine and Freshwater Research 50:277-290 Hung KLJ, Kingston JM, Albrecht M, Holway DA, Kohn JR (2018) The worldwide importance of honey bees as pollinators in natural habitats. Proceedings of the Royal Society B: Biological Sciences 285:1-8 Iftekhar MS, Polyakov M, Ansell D, Gibson F, Kay GM (2017) How economics can further the success of ecological restoration. Conservation Biology 31:261- 268 Jones A (2000) Effects of cattle grazing on North American arid ecosystems: A quantitative review. Western North American Naturalist 60: 155-164 Joshi NK, Leslie T, Rajotte EG, Kammerer MA, Otieno M, Biddinger DJ (2015) Comparative trapping efficiency to characterize bee abundance, diversity, and community composition in apple orchards. Annals of the Entomological Society of America 108:785-799 Kearns CA & Inouye DW (1997) Pollinators, flowering plants, and conservation biology. BioScience 47:297-307

Kearns CA, Inouye DW, Waser NM (1998) Endangered mutualisms: The conservation of plant-pollinator interactions. Annual Review of Ecology and Systematics 2:83-112 Kennedy PL, DeBano SJ, Bartuszevige AM, Lueders AS (2009) Effects of native and non-native grassland plant communities on breeding passerine birds: Implications for restoration of northwest bunchgrass prairie. Restoration Ecology 17:515-525 Kimoto C, DeBano SJ, Thorp RW, Robbin W, Rao S, Stephen WP (2012a) Investigating temporal patterns of a native bee community in a remnant North American bunchgrass prairie using blue vane traps. Journal of Insect Science 12:1-23 Kimoto C, DeBano SJ, Thorp RW, Taylor RV, Schmalz H, DelCurto T, Johnson T, Kennedy PL, Rao S (2012b) Short-term responses of native bees to livestock and implications for managing ecosystem services in grasslands. Ecosphere 3:88-107 Kirner KD (2015) The cultural heritage of family ranches. Rangelands 37:85–89 Klein AM, Vaissière BE, Cane JH, Steffan-Dewenter I, Cunningham SA, Kremen C, Tscharntke T (2007) Importance of pollinators in changing landscapes for world crops. Proceedings: Biological Sciences 274:303-313 Kovács-Hostyánszki A, Espındola A, Vanbergen AJ, Settele J, Kremen C, Dicks LV (2017) Ecological intensification to mitigate impacts of conventional intensive land use on pollinators and pollination Irwin, R, editor. Ecology Letters 20:673–689 Kovács-Hostyánszki A, Haenke S, Batáry P, Jauker B, Báldi A, Tscharntke T, Holzschuh A (2013) Contrasting effects of mass-flowering crops on bee pollination of hedge plants at different spatial and temporal scales. Ecological Applications 23:1938-1946 Kremen C & Ricketts T (2000) Global perspectives on pollination disruptions. Conservation Biology 14:1226–1228

Kruess A & Tscharntke T (2002) Grazing intensity and the diversity of grasshoppers, butterflies, and trap-nesting bees and wasps. Conservation Biology 16:1570- 1580 Kuhlman M & Burrows S (2017) Checklist of bees (Apoidea) from a private conservation property in west-central Montana. Biodiversity Data Journal. Kulhanek K, Steinhauer N, Rennich K, Caron DM, Sagili RR, Pettis JS, Ellis JD, Wilson ME, Wilkes JT, Tarpy DR, Rose R, Lee K, Rangel J, vanEngelsdorp D (2017) A national survey of managed honey bee 2015–2016 annual colony losses in the USA. Journal of Apicultural Research 56:328–340 Kurz I, O’Reilly CD, Tunney H (2006) Impact of cattle on soil physical properties and nutrient concentrations in overland flow from pasture in Ireland. Agriculture, Ecosystems & Environment 113:378-390 Lake PS (2000) Disturbance, patchiness, and diversity in streams. Journal of the North American Benthological Society 19:573-592 Lázaro A, Tscheulin T, Devalez J, Nakas G, Petanidou T (2016) Effects of grazing intensity on pollinator abundance and diversity, and on pollination services. Ecological Entomology 41:400-412 Le Féon V, Schermann-Legionnet A, Delettre Y, Aviron S, Billeter R, Bugter B, Hendrickx F, Burel F (2010) Intensification of agriculture, landscape composition and wild bee communities: A large scale study in four European countries. Agriculture, Ecosystems & Environment 137:143-150 Lebuhn G, Droege S, Connor EF, Gemmill-Herren B, Potts SG, Minckley RL, Griswold T, Jean R, Kula E, Roubik DW, Cane J, Wright KW, Frankie G, Parker F (2013) Detecting insect pollinator declines on regional and global scales: Detecting pollinator declines. Conservation Biology 27:113–120 Lee KV, Steinhauer N, Rennich K, Wilson ME, Tarpy DR, Caron DM, Rose R, Delaplane KS, Baylis K, Lengerich EJ, Pettis J, Skinner JA, Wilkes JT, Sagili R, vanEnglersdorp D (2015) A national survey of managed honey bee 2013– 2014 annual colony losses in the USA. Apidologie 46:292–305 Liczner AR & Colla SR (2019) A systematic review of the nesting and overwintering habitat of bumble bees globally. Journal of Insect Conservation 23:787-801

Losey JE & Vaughan M (2006) The economic value of ecological services provided by insects. BioScience 56:311-323 McCune B, Grace JB, Urban DL (2002) Analysis of ecological communities. 1st ed. MjM Software Design. McCune B & Mefford MJ (2006) PC-ORD: Multivariate Analysis of Ecological Data. MjM Software, Gleneden Beach, Oregon, USA Melathopoulos AP, Cutler GC, Tyedmers P (2015) Where is the value in valuing pollination ecosystem services to agriculture? Ecological Economics 109: 59– 70. Michener CD (2007) The bees of the world. 2nd ed. Johns Hopkins University Press, Baltimore, Maryland Morandin LA & Kremen C (2013) Bee preference for native versus exotic plants in restored agricultural hedgerows. Restoration Ecology 21: 26-32 Naiman RJ, Decamps H, Pollock M (1993) The role of riparian corridors in maintaining regional biodiversity. Ecological Applications 3: 209-212 National Oceanic and Atmospheric Administration (1981-2010) National Climatic Data Center https://catalog.data.gov/dataset/u-s-hourly-climate-normals-1981- 2010 (accessed 18 Dec 2019) National Research Council U. S. Committee on the Status of Pollinators in North America (2007) Status of pollinators in North America. National Academies Press, Washington, D.C. National water quality inventory: Report to Congress (2017) EPA 841-R-16-011. National Water Quality Inventory. Environmental Protection Agency, Washington, D.C. O’Connor RS, Kunin WE, Garratt MPD, Potts SG, Roy HE, Andrews C, Jones CM, Peyton JM, Savage J, Harvey MC, Morris RKA, Roberts SPM, Wright I, Vanbergen AJ, Carvell C (2019) Monitoring insect pollinators and flower visitation: The effectiveness and feasibility of different survey methods. Methods in Ecology and Evolution 10:2129-2140

Öckinger E & Smith HG (2007) Semi-natural grasslands as population sources for pollinating insects in agricultural landscapes. Journal of Applied Ecology 44:50-59 Ollerton J, Winfree R, Tarrant S (2011) How many flowering plants are pollinated by animals? Oikos 120:321–326. Poisot T, Canard E, Mouquet N, Hochberg ME (2012) A comparative study of ecological specialization estimators: Species-level specialization. Methods in Ecology and Evolution 3:537-544 Potts SG, Vulliamy B, Dafni A, Ne’Eman G, Willmer P (2003) Linking bees and flowers: How do floral communities structure pollinator communities? Ecology 84:2628-2642 Pywell RF, Warman EA, Carvell C, Sparks TH, Dicks LV, Bennett D, Wright A, Critchley CNR, Sherwood A (2005) Providing foraging resources for bumblebees in intensively farmed landscapes. Biological Conservation 121: 479-494 R Core Team. 2019. R: A Language and Environment for Statistical Computing. R. Vienna, Austria: R Foundation for Statistical Computing. https://www.R- project.org/ Rao S & Stephen WP (2009) Bumble bee pollinators in red clover seed production. Crop Science 49:2207-2214 Reddersen J (2001) SRC-willow (Salix viminalis) as a resource for flower-visiting insects. Biomass and Bioenergy 20:171-179 Rhoades PR, Davis TS, Tinkham WT, Hoffman CM (2018) Effects of seasonality, forest structure, and understory plant richness on bee community assemblage in a southern Rocky Mountain mixed conifer forest. Annals of the Entomological Society of America 111:278-284 Rockwell SM & Stephens JL (2018) Habitat selection of riparian birds at restoration sites along the Trinity River, California. Restoration Ecology 26:767-777 Roof SM, DeBano SJ, Rowland MM, Burrows S (2018) Associations between blooming plants and their bee visitors in a riparian ecosystem in eastern Oregon. Northwest Science 92:119-135

Roulston TH & Cane JH (2000) Pollen nutritional content and digestibility for animals. Plant Systematics and Evolution; Heidelberg 222:187-209 Rowland MM, Bryant LD, Johnson BK, Noyes JH, Wisdom MJ, Thomas JW (1997) Starkey project: history facilities, and data collection methods for ungulate research. USFS PNW-GTR-396. United States Forest Service, Washington D.C. Saunders ME (2018) Insect pollinators collect pollen from wind-pollinated plants: Implications for pollination ecology and sustainable agriculture. Insect Conservation and Diversity 11:13-31 Schmalz HJ, Taylor RV, Johnson TN, Kennedy PL, DeBano SJ, Newingham BA, McDaniel PA (2013) Soil morphologic properties and cattle stocking rate affect dynamic soil properties. Rangeland Ecology & Management 66:445- 453 Schönbach P, Wan H, Gierus M, Bai Y, Müller, Lin L, Susenbeth A, Taube F (2011) Grassland responses to grazing: effects of grazing intensity and management system in an Inner Mongolian steppe ecosystem. Plant Soil 340:103-115 Schulz TT & Leininger WC (1990) Differences in riparian vegetation structure between grazed areas and exclosures. Journal of Range Management 43:295- 299 Shapira T, Henkin Z, Dag A, Mandelik Y (2020) Rangeland sharing by cattle and bees: moderate grazing does not impair bee communities and resource availability. Ecological Applications (in press) Sjödin NE (2007) Pollinator behavioural responses to grazing intensity. Biodiversity and Conservation 16:2103-2121 Stanford B, Holl KD, Herbst DB, Zavaleta E (2020) In-stream habitat and macroinvertebrate responses to riparian corridor length in rangeland streams. Restoration Ecology 28:173-184 Sweeney BW, Bott TL, Jackson JK, Kaplan LA, Newbold JD, Standley LJ, Hession WC, Horwitz LJ (2004) Riparian deforestation, stream narrowing, and loss of stream ecosystem services. Proceedings of the National Academy of Sciences 101:14132-14137.

Tisdale EW (1982) Grasslands of western North America: The Pacific Northwest Bunchgrass. In: Grassland Ecology and Classification Symposium, Kamloops, BC. pp. 223–245. Treshow M & Harper K (1974) Longevity of perennial forbs and grasses. Oikos 25:93-96 Vanbergen AJ & Insect Pollinators Initiative (2013) Threats to an ecosystem service: pressures on pollinators. Frontiers in Ecology and the Environment 11:251- 259 Vincent CH (2019) Grazing fees: Overview and issues. CRS21232 Congressional Research Service, Washington, D.C. Vulliamy B, Potts SG, Willmer PG (2006) The effects of cattle grazing on plant- pollinator communities in a fragmented Mediterranean landscape. Oikos 114:529-543 Walker, K (2017) State of Oregon Agriculture. Oregon Department of Agriculture. [Industry report] Weiner CN, Werner M, Linsenmair KE, Blüthgen N (2014) Land-use impacts on plant–pollinator networks: interaction strength and specialization predict pollinator declines. Ecology 95:466-474 Weiner CN, Hilpert A, Werner M, Linsenmair KE, Blüthgen N (2010) Pollen amino acids and flower specialisation in solitary bees. Apidologie 41: 476–487 Wickham H. 2016. Ggplot2: Elegant Graphics for Data Analysis. New York: Springer-Verlag. https://ggplot2.tidyverse.org Williams NM (2011) Restoration of nontarget species: Bee communities and pollination function in riparian forests. Restoration Ecology 19: 450-459 Williams NM, Regetz J, Kremen C (2012) Landscape-scale resources promote colony growth but not reproductive performance of bumble bees. Ecology 93:1049- 1058 Wilson EO (1987) The little things that run the world (the importance and conservation of invertebrates). Conservation Biology 1:344–346 Wilson JS, Carril OJM (2015) The bees in your backyard: A guide to North America’s bees. Princeton University Press, Princeton, New Jersey

Winfree R (2010) The conservation and restoration of wild bees: Wild bee conservation. Annals of the New York Academy of Sciences 1195:169-197 Wohl E, Lane SN, Wilcox AC (2015) The science and practice of river restoration. Water Resources Research 51:5974-5997 Wondzell SM, Diabat M, Haggerty, R (2019) What matters most: Are future stream temperatures more sensitive to changing air temperatures, discharge, or riparian vegetation? Journal of the American Water Resources Association 55:116-132 Xie Z, Williams PH, Tang Y (2008) The effect of grazing on bumblebees in the high rangelands of the eastern Tibetan Plateau of Sichuan. Journal of Insect Conservation 12:695-703

APPENDIX 1: Additional Figures and Tables for Chapter 2 This appendix includes seven tables that are supplementary materials to Chapter 2.

Supporting Information Table A1.1: Complete bee species list and occurrence at each location and reference for identification for bees identified in Chapter 2.

Genus Species Name Starkey Zumwalt Author and Date Agapostemon Agapostemon femoratus X X Crawford 1901 Agapostemon texanus X X Cresson 1872 Agapostemon virescens X X (Fabricius 1775) Andrena Andrena amphibola X (Viereck 1904) Andrena angustitarsata X X Viereck 1904 Andrena astragali X X Viereck & Cockerell 1914 Andrena caerulea X Smith 1879 Andrena cressonii X X Robertson 1891 Andrena cyanophila X Cockerell 1906 Andrena evoluta X X Linsley & MacSwain 1961 Andrena gordoni X Ribble 1974 Andrena lawrencei X Viereck & Cockerell 1914 Andrena medionitens X Cockerell 1902 Andrena melanochroa X X Cockerell 1898 Andrena merriami X Cockerell 1901 Andrena microchlora X X Cockerell 1922 Andrena miranda X X Smith 1879 Andrena nigrocaerulea X X Cockerell 1897 Andrena nivalis X X Smith 1853 Andrena pallidifovea X X (Viereck 1904) Andrena prunorum X X Cockerell 1896 Andrena raveni X Linsley & MacSwain 1961 Andrena salicifloris X X Cockerell 1897 Andrena schuhi X LaBerge 1980 Andrena scutellinitens X Viereck 1916 Andrena sp. X No author date Andrena thaspii X X Graenicher 1903 Andrena transnigra X X Viereck 1904 Andrena venata X X LaBerge & Ribble 1975 Andrena vierecki X X Cockerell 1904 Andrena washingtoni X Cockerell 1901 Andrena w-scripta X Viereck 1904 Anthidium Anthidium atrifrons X Cresson 1868 Anthidium utahense X X Swenk 1914 Anthophora Anthophora bomboides X Kirby 1838 Anthophora curta X Provancher 1895 Anthophora pacifica X Cresson 1878 Anthophora urbana X X Cresson 1878 Anthophora ursina X Cresson 1869 Apis Apis mellifera X Linnaeus 1758

Genus Species Name Starkey Zumwalt Author and Date Ashmeadiella Ashmeadiella bucconis X X (Say 1837) Ashmeadiella difugita X Michener 1939 Atoposmia Atoposmia abjecta X (Cresson 1878) Bombus Bombus appositus X X Cresson 1878 Bombus bifarius X X Cresson 1878 Bombus californicus X X Smith 1854 Bombus centralis X X Cresson 1864 Bombus fernaldae X (Franklin 1911) Bombus fervidus X (Fabricius 1798) Bombus flavifrons X X Cresson 1863 Bombus huntii X Greene 1860 Bombus insularis X (Smith 1861) Bombus mixtus X X Cresson 1878 Bombus nevadensis X X Cresson 1874 Bombus occidentalis X Greene 1858 Bombus rufocinctus X X Cresson 1863 Bombus sp. X No author date Bombus vosnesenskii X Radoszkowski 1862 Ceratina Ceratina acantha X Provancher 1895 Ceratina nanula X X Cockerell 1897 Ceratina pacifica X H.S. Smith 1907 Chelostoma Chelostoma phaceliae X Michener 1938 Coelioxys Coelioxys moesta X Cresson 1864 Coelioxys rufitarsis X Smith 1854 Colletes Colletes consors X Cresson 1868 Colletes fulgidus X X Swenk 1904 Colletes phaceliae X Cockerell 1906 Colletes sp. X No author date Diadasia Diadasia enavata X (Cresson 1872) Diadasia nigrifrons X (Cresson 1878) Dianthidium Dianthidium parvum X (Cresson 1878) Dianthidium subparvum X Swenk 1914 Dianthidium ulkei X (Cresson 1878) Dufourea Dufourea dilatipes X Bohart 1948 Dufourea trochantera X Bohart 1948 Epeolus Epeolus sp. X No author date Eucera Eucera edwardsii X (Cresson 1878) Eucera frater X X (Cresson 1878) Eucera hurdi X (Provancher 1888) Halictus Halictus confusus X X Smith 1853 Halictus farinosus X X Smith 1853 Halictus ligatus X X Say 1837 Halictus rubicundus X X (Christ 1791)

Genus Species Name Starkey Zumwalt Author and Date Halictus tripartitus X X Cockerell 1895 Hoplitis Hoplitis albifrons X X (Cresson 1864) Hoplitis fulgida X X (Cresson 1864) Hoplitis grinnelli X Cockerell 1910 Hoplitis plagiostoma X Michener 1947 Hoplitis producta X X (Cresson 1864) Hylaeus Hylaeus coloradensis X (Cockerell 1896) Hylaeus conspicuus X X (Metz 1911) Hylaeus episcopalis X (Cockerell 1896) Hylaeus modestus X (Cockerell 1896) Hylaeus verticalis X (Cresson 1869) Lasioglossum Lasioglossum aberrans X (Crawford 1903) Lasioglossum albipenne X X (Robertson 1890) Lasioglossum cooleyi X X (Crawford 1906) Lasioglossum egregium X X (Vachal 1904) Lasioglossum glabriventre X (Crawford 1907) Lasioglossum incompletum X X (Crawford 1907) Lasioglossum laevissimum X (Smith 1853) Lasioglossum nevadense X X (Crawford 1907) Lasioglossum olympiae X X (Cockerell 1898) Lasioglossum ovaliceps X (Cockerell 1898) Lasioglossum pacificum X (Cockerell 1898) Lasioglossum sedi X X (Sandhouse 1924) Lasioglossum sisymbrii X X (Cockerell 1895) Lasioglossum sp. X X Gibbs ms Lasioglossum sp. 2 X No author date Lasioglossum sp. 3 X X No author date Lasioglossum sp. 4 X No author date Lasioglossum sp. 5 X No author date Lasioglossum titusi X X (Crawford 1902) Lasioglossum trizonatum X X (Cresson 1874) Lasioglossum versans X X (Lovell 1905) Megachile Megachile apicalis X Spinola 1808 Megachile gemula X Cresson 1878 Megachile melanophaea X Smith 1853 Megachile montivaga X Cresson 1878 Megachile onobrychidis X X Cockerell 1905 Megachile parallela X X Smith 1853 Megachile perihirta X X Cockerell 1898 Megachile pugnata X Say 1837 Melecta Melecta pacifica X X Cresson 1878 Melecta separata X Cresson 1879 Melissodes Melissodes ablusus X Cockerell 1926

Genus Species Name Starkey Zumwalt Author and Date Melissodes agilis X Cresson 1878 Melissodes Melissodes lupinus X X Cresson 1878 (cont.) Melissodes microstictus X X Cockerell 1905 Melissodes rivalis X X Cresson 1872 Melissodes sp. X X No author date Nomada Nomada sp. X X No author date Osmia Osmia albolateralis X Cockerell 1906 Osmia atrocyanea X Cockerell 1897 Osmia austromaritima X Michener 1936 Osmia bella X Cresson 1878 Osmia brevis X X Cresson 1864 Osmia bruneri X X Cockerell 1897 Osmia bucephala X Cresson 1864 Osmia californica X X Cresson 1864 Osmia calla X Cockerell 1897 Osmia cyaneonitens X Cockerell 1906 Osmia densa X Cresson 1864 Osmia ednae X X Cockerell 1907 Osmia indeprensa X X Sandhouse 1939 Osmia integra X Cresson 1878 Osmia juxta X Cresson 1864 Osmia kincaidii X X Cockerell 1897 Osmia longula X X Cresson 1864 Osmia melanopleura X X Cockerell 1916 Osmia montana X Cresson 1864 Osmia nemoris X X Sandhouse 1924 Osmia nigrifrons X Cresson 1878 Osmia pusilla X X Cresson 1864 Osmia raritatis X Michener 1957 Osmia simillima X Smith 1853 Osmia sp. X X No author date Osmia sp. 10 X No author date Osmia sp. 2 X No author date Osmia sp. 3 X No author date Osmia sp. 9 X No author date Osmia thysanisca X Michener 1957 Osmia trevoris X X Cockerell 1897 Panurginus Panurginus gracilis X Michener 1935 Panurginus torchio X X No author date Perdita Perdita lingualis X Cockerell 1896 Perdita wyomingensis X Cockerell 1922 Pseudopanurgus Pseudopanurgus didirupa X (Cockerell 1908) Sphecodes Sphecodes sp. X X No author date

Genus Species Name Starkey Zumwalt Author and Date Stelis Stelis sp. X No author date Stelis sp. B X No author date Triepeolus Triepeolus heterurus X (Cockerell & Sandhouse 1924)

Table A1.2: Pearson correlations for plant ordination Pearson correlations for whole season bee ordination axis 1 and axis 2. Strongest correlations for axis 1 and axis 2 are shown. Cells are highlighted in gray to show axis of interest. Environmental variables were only significantly correlated to axis 1, so axis 2 correlation values are not shown (Figure 2.3). Only the top three species positively and negatively correlated with each axis are shown.

Axis 1 Axis 2 R R2 R R2 Month -0.6 0.3 - - Environmental Bloom Abundance 0.5 0.3 - - Variables Bloom Diversity 0.5 0.3 - - Stocking Rate -0.4 0.1 - -

Lasioglossum sp. 2 0.5 0.3 0.4 0.2 Andrena melanchroa 0.4 0.2 -0.2 0.03 Eucera frater 0.4 0.2 0.2 0.02 Halictus tripartitus -0.6 0.4 0.2 0.03 Lasioglossum -0.5 0.3 -0.2 0.02 Bee species incompletum associations Lasioglossum versans -0.5 0.3 -0.03 0.001 Eucera hurdi -0.05 0.003 0.4 0.1 Andrena transnigra -0.04 0.001 0.3 0.1 Osmia trevoris 0.03 0.001 -0.4 0.2 Andrena venata 0.2 0.05 -0.3 0.1 Osmia thyamisca 0.1 0.01 -0.3 0.1

Table A1.3: Pearson correlations for plant ordination Pearson correlations for whole season plant ordination axis 1 and axis 2 (Figure 2.3) Cells are highlighted in gray to show axis of interest. Environmental variables were only significantly correlated to axis 1, so axis 2 correlation values are not shown. Only correlations with r2 > 0.1 are shown.

Axis 1 Axis 2 R R2 R R2 Environmental Month 0.9 0.9 - - variables Stocking rate 0.3 0.1 - -

Myosotis stricta -0.5 0.3 -0.006 ≈ 0 Collinsia parviflorum -0.5 0.2 -0.1 0.02 Draba verna -0.4 0.2 -0.08 0.007 Taraxacum officinale -0.4 0.2 -0.3 0.08 Plant Species Viola adunca -0.4 0.2 -0.2 0.04 Fragaria virginiana -0.4 0.1 -0.2 0.03 Montia linearis -0.4 0.1 -0.2 0.03 BAOR -0.3 0.1 -0.2 0.02 Arnica sororia -0.3 0.09 -0.4 0.2

Table A1.4: Relative proportion of taxa making up 90% of all bees collected each month at the Zumwalt Prairie in 2018.

June % of total July % of total August % of total Andrena venata 13 % Lasioglossum sp. 23 % Lasioglossum incompletum 46 % Osmia trevoris 11 % Osmia trevoris 14 % Halictus tripartitus 14 % Andrena microchlora 10 % Lasioglossum titusi 7 % Lasioglossum sp. 6 % Lasioglossum incompletum 8 % Panurginus torchio 5 % Lasioglossum titusi 4 % Halictus tripartitus 5 % Halictus tripartitus 4 % Melissodes lupinus 3 % Andrena nigrocaerulea 5 % Bombus californicus 4 % Bombus californicus 3 % Lasioglossum sedi 5 % Lasioglossum incompletum 4 % Halictus ligatus 3 % Lasioglossum sp. 5 % Lasioglossum sedi 4 % Lasioglossum cooleyi 2 % Lasioglossum sp. 3 3 % Lasioglossum cooleyi 4 % Lasioglossum vesans 2 % Halictus rubicundus 3 % Pseudopanurginus didirupa 3 % Lasioglossum albipenne 2 % Panurginus torchio 3 % Andrena melanochroa 2 % Lasioglossum aberrans 1 % Panurginus gracilis 2 % Halictus rubicundus 2 % Melissodes ablusus 1 % Halictus confusus 2 % Perdita wyomingensis 2 % Halictus rubicundus 1 % Andrena evoluta 1 % Lasioglossum albipenne 2 % Halictus farinosus 1 % Lasioglossum aberrans 1 % Nomada sp. 2 % Andrena melanochroa 1 % Andrena venata 1 % Eucera edwardsii 1 % Andrena nigrocaerulea 1 % Eucera frater 1 % Eucera frater 1 % Andrena merriami 1 % Halictus confusus 1 % Andrena thaspii 1 % Lasioglossum versans 1 % Lasioglossum cooleyi 1 % Osmia melanopleura 1 % Nomada sp. 1 % Osmia nemoris 1 % Osmia kincaidii 1 % Osmia thysanisca 1 % Andrena cressonii 1 % Bombus centralis 1 % Andrena gordoni 1 % Osmia bruneri 1 % Bombus californicus 1 % Osmia ednae 1 % Lasioglossum trizonattum 1 % Andrena evoluta < 1 % Osmia thysanisca 1 % Andrena washingtoni 1 % Lasioglossum titusi 1 %

Table A1.5: Relative proportion of taxa making up 90% of all bees collected each month at Starkey in 2018

May % of June % of July % of September % of total total total total Halictus tripartitus 19 % Lasioglossum olympiae 16 % Lasioglossum nevadense 17 % Halictus tripartitus 43 % Lasioglossum incompletum 11 % Lasioglossum sp. 2 15 % Hoplitis fulgida 15 % Halictus ligatus 15 % Lasioglossum nevadense 10 % Lasioglossum sedi 10 % Lasioglossum incompletum 7 % Lasioglossum incompletum 6 % Osmia densa 6 % Andrena nigrocaerulea 10 % Halictus tripartitus 6 % Lasioglossum sp. 5 % Andrena microchlora 5 % Andrena melanochroa 6 % Bombus californicus 5 % Lasioglossum cooleyi 4 % Lasioglossum cooleyi 5 % Eucera frater 5 % Halictus farinosus 5 % Lasioglossum nevadense 3 % Osmia juxta 5 % Lasioglossum sp. 5 % Osmia trevoris 4 % Lasioglossum versans 3 % Andrean transnigra 4 % Panurginus torchio 4 % Osmia pusilla 3 % Halictus farinosus 3 % Osmia atrocyanea 3 % Dufourea dilatipes 3 % Lasioglossum sp. 3 % Agapostemon virescens 2 % Eucera hurdi 2 % Lasioglossum nevadense 3 % Osmia albolateralis 3 % Lasioglossum sp. 3 2 % Andrena nigrihirta 2 % Melecta pacifica 2 % Osmia juxta 3 % Halictus confusus 1 % Lasioglossum sedi 2 % Bombus flavifrons 1 % Bombus bifarius 3 % Bombus bifarius 1 % Lasioglossum trizonatum 2 % Halictus rubicundus 1 % Hoplitis albifrons 2 % Ceratina acantha 1 % Osmia sp. 2 % Lasioglossum cooleyi 1 % Lasioglossum cooleyi 2 % Halictus rubicundus 1 % Lasioglossum sp. 2 % Lasioglossum laevissimum 1 % Diadasia nigrifrons 2 % Lasioglossum laevissimum 1 % Osmia californica 1 % Lasioglossum sisymbrii 1 % Bombus flavifrons 2 % Dianthidium subparvum 1 % Halictus farinosus 1 % Osmia densa 1 % Halictus rubicundus 2 % Lasioglossum glabriventre 1 % Osmia juxta 1 % Osmia atrocyanea 2 % Lasioglossum sp. 3 1 % Andrena evoluta 1 % Osmia densa 1 % Lasioglossum sp. 2 1 % Andrena pallidifovea 1 % Bombus appositus 1 % Andrena nigrocaerulea 1 % Andrena prunorum 1 % Halictus ligatus 1 % Andrena evoluta 1 % Anthophora pacifica 1 % Melissodes microstictus 1 % Andrena nivalis 1 % Bombus appositus 1 % Lasioglossum versans 1 % Bombus californicus 1 % Bombus centralis 1 % Lasioglossum glabriventre 1 %

Table A1.6: Relative proportion of species making up 90% of blooms counted each month at the Zumwalt Prairie in 2018.

June % of total July % of total August % of total Arnia sororia 16 % Orthocarpus tenuifolius 23 % Erigeron pumilus 36 % Myosotis stricta 12 % Arnica sororia 21 % Epilobium brachycarpum 21 % Geum triflorum 10 % Clarkia pulchella 20 % Achillea millefolium 20 % Frasera albicaulis 9 % Achillea millefolium 6 % Perideridia gairdneri 17 % Lupinus spp. 7 % Polygonum polygaloides 5 % Astragalus sheldonii 6 % Lupinus spp. 4 % Microsteris gracilis 6 % Epilobium brachycarpum 3 % Castilleja tenuis 5 % Collinsia parviflorum 2 % Collomia linearis 4 % Hieracium cynoglossoides 2 % Antennaria luzuloides 3 % Castilleja tenuis 2 % Arennaria aculeata 3 % Potentilla gracilis 2 % Potentilla gracilis 3 % Alyssum alyssoides 2 % Zigodenus venenosus 2 % Draba verna 2 % Collinsia parviflorum 1 % Delphinium nuttallianum 1 % Lithophragma parviflorum 1 %

Table A1.7: Relative proportion of species making up 90% of counted blooms collected each month at Starkey in 2018.

% of % of % of % of % of May June July August September total total total total total Collomia Perideridia Symphyotrichum Myosotis stricta 27 % 28 % Achillea millefolium 15 % 32 % 38 % linearis gairdneri spathulatum Prunella Symphyotrichum Symphyotrichum Epilobium Draba verna 19 % 16 % 14 % 23 % 36 % vulgaris spathulatum spathulatum brachycarpum Collinsia Microsteris Epilobium Polygonum 16 % 11 % Monardella odoratissima 11 % 11 % 18 % parviflorum gracilis brachycarpum douglasii Sedum Viola nuttalli 13 % 8 % Eriogonum heracleoides 9 % Achillea millefolium 10 % stenopetalum Microsteris Thermopsis Solidago 5 % 7 % Lotus unifoliolatus 9 % 10 % gracilis montana missouriensis Achillea Erigeron Viola adunca 4 % 4 % Galium boreale 7 % 5 % millefolium corymbosus Polemonium Montia linearis 3 % 4 % Solidago missouriensis 6 % occidentale Fragaria Myosotis 3 % 4 % Senecio serra 5 % virginiana stricta Taraxacum Potentilla 2 % 3 % Perideridia gairdneri 3 % officinale gracilis Eriogonum 3 % Erigeron corymbosus 3 % heracleoides Hypericum 1 % Potentilla gracilis 3 % anagalloides Dianthus armeria 3 %

Sidalcea oregana 2 %

Epilobium brachycarpum 2 %

Polygonum douglasii 2 %

APPENDIX 2: Additional Figures and Tables for Chapter 3 This appendix includes one table and one figure that are supplementary materials to Chapter 3.

Table A2.1: Complete list of bee species collected on shrubs and forbs over the two sampling years. The number of bees captured on shrubs, forbs, and trees are noted for each species. The paired differences index (PDI) score for each species is listed and the reference for the taxonomic identification are included. This list refers only to bees captured in Chapter 3.

Bees species observed on forbs and shrubs Bee Species Forb Shrub Tree Total Bees PDI Score Reference

Andrena angustitarsata 14 51 65 0.977 Viereck 1904 Andrena cressonii 1 3 4 0.967 Robertson 1891 Andrena miranda 3 6 9 0.978 Smith 1879 Andrena nivalis 1 9 10 0.995 Smith 1853 Andrena salicifloris 4 9 13 0.963 Cockerell 1897 Andrena sp. 2 1 3 0.978 No author date Andrena transnigra 1 6 7 0.998 Viereck 1904 Andrena vierecki 1 46 47 0.993 Cockerell 1904 Andrena walleyi 2 1 3 0.995 Cockerell 1932 Anthophora pacifica 3 101 104 0.999 Cresson 1878 Anthophora ursina 6 12 18 0.995 Cresson 1869 Bombus appositus 7 3 10 0.983 Cresson 1878 Bombus bifarius 193 6 3 204 0.984 Cresson 1878 Bombus californicus 36 8 44 0.971 Smith 1854 Bombus centralis 33 21 1 55 0.969 Cresson 1864 Bombus flavifrons 97 44 141 0.967 Cresson 1863 Bombus insularis 19 1 20 0.987 (Smith 1861) Bombus mixtus 52 36 13 101 0.962 Cresson 1878 Eucera frater 10 2 12 0.995 (Cresson 1878), (Provancher 1888) Halictus farinosus 11 11 22 0.989 Smith 1853

Halictus rubicundus 4 2 7 0.945 (Christ 1791) Halictus tripartitus 8 1 9 0.961 Cockerell 1895 Hoplitis fulgida 11 1 12 0.967 (Cresson 1864) Hylaeus modestus 3 1 4 0.989 (Cockerell 1896) Hylaeus verticalis 1 3 4 0.989 (Cresson 1869) Lasioglossum cooleyi 26 5 32 0.985 (Crawford 1906) Lasioglossum laevissimum 17 1 18 0.995 (Smith 1853) Lasioglossum nevadense 14 11 25 0.986 (Crawford 1907) Lasioglossum olympiae 55 66 121 0.978 (Cockerell 1898) Lasioglossum ruidosense 1 2 3 0.995 (Cockerell 1897) Lasioglossum sp. 30 3 33 0.975 Gibbs ms Lasioglossum sp. 2 89 24 113 0.973 No author date Lasioglossum sp. 3 4 4 8 0.923 No author date Lasioglossum sp. 4 5 3 8 0.982 No author date Lasioglossum trizonatum 2 1 3 0.978 (Cresson 1874) Melecta pacifica 2 2 5 0.989 Cresson 1878 Nomada sp. I 1 1 2 0.989 No author date Osmia albolateralis 6 3 9 0.978 Cockerell 1906 Osmia atrocyanea 8 1 9 0.986 Cockerell 1897 Osmia bruneri 1 1 2 0.989 Cockerell 1897 Osmia bucephala 8 2 10 0.993 Cresson 1864 Osmia densa 3 1 4 0.967 Cresson 1864 Osmia pusilla 3 1 4 0.989 Cresson 1864 Osmia sp. 5 1 6 0.989 No author date Osmia trevoris 1 1 2 0.989 Cockerell 1897 Panurginus gracilis 1 1 2 0.989 Michener 1935 Panurginus ineptus 10 4 14 0.995 Cockerell 1922 Panurginus torchio 13 15 28 0.985 No author date

Perdita wyomingensis 2 1 3 0.995 Cockerell 1922 Sphecodes sp. 2 2 4 0.989 No author date Bees species observed on forbs only Bee Species Forb Shrub Tree Total Bees PDI Score Reference Agapostemon texanus 1 1 1 Cresson 1872 Agapostemon virescens 2 2 1 (Fabricius 1775) Andrena barbilabris 1 1 1 (Kirby 1802) Andrena cyanophila 7 7 0.998 Cockerell 1906 Andrena evoluta 6 6 0.998 Linsley & MacSwain 1961 Andrena frigida 1 1 1 Smith 1853 Andrena laminibucca 2 2 0.989 Viereck & Cockerell 1914 Andrena melanochroa 44 44 0.990 Cockerell 1898 Andrena microchlora 42 42 0.985 Cockerell 1922 Andrena nigrocaerulea 3 3 0.978 Cockerell 1897 Andrena scutellinitens 3 3 0.995 Viereck 1916 Andrena sladeni 4 4 1 Viereck 1924 Andrena sp. A 1 1 1 No author date Andrena washingtoni 2 2 1 Cockerell 1901 Anthidium atrifrons 1 1 1 Cresson 1868 Anthidium utahense 1 1 1 Swenk 1914 Anthophora bomboides 2 2 1 Kirby 1838 Anthophora terminalis 1 1 1 Cresson 1869 Anthophora urbana 2 2 0.989 Cresson 1878 Ashmeadiella bucconis 1 1 1 (Say 1837) Bombus fernaldae 16 16 0.976 (Franklin 1911) Bombus fervidus 2 2 1 (Fabricius 1798)

Bombus huntii 1 1 1 Greene 1860 Bombus rufocinctus 2 2 0.989 Cresson 1863 Bombus vosnesenskii 10 10 0.997 Radoszkowski 1862 Ceratina acantha 1 1 1 Provancher 1895 Coelioxys rufitarsis 2 2 1 Smith 1854 Diadasia nigrifrons 25 25 0.999 (Cresson 1878) Dianthidium subparvum 1 1 1 Swenk 1914 Epeolus americanus 1 1 1 (Cresson 1878) Eucera hurdi 4 4 0.989 (Timberlake 1969) Halictus confusus 9 9 0.986 Smith 1853 Halictus ligatus 97 97 0.981 Say 1837 Hylaeus annulatus 1 1 1 (Linnaeus 1758) Hylaeus basalis 3 3 0.995 (Smith 1853) Lasioglossum egregium 8 8 0.982 (Vachal 1904) Lasioglossum glabriventre 3 3 1 (Crawford 1907) Lasioglossum incompletum 5 5 0.956 (Crawford 1907) Lasioglossum ovaliceps 4 4 0.989 (Cockerell 1898) Lasioglossum sandhousiellum 1 1 1 Gibbs 2010 Lasioglossum sedi 10 10 0.984 (Sandhouse 1924) Lasioglossum tenax 1 1 1 (Sandhouse 1924) Lasioglossum titusi 2 2 0.989 (Crawford 1902) Lasioglossum versans 3 3 1 (Lovell 1905) Megachile montivaga 1 1 1 Cresson 1878 Megachile perihirta 5 5 0.997 Cockerell 1898 Megachile relativa 1 1 1 Cresson 1878 Melissodes microstictus 6 6 0.989 Cockerell 1905 Melissodes rivalis 7 7 0.998 Cresson 1872 Nomada sp. 1 2 2 0.989 No author date

100

Nomada sp. C 2 2 0.989 No author date Nomada sp. D 2 2 1 No author date Osmia brevis 4 4 1 Cresson 1864 Osmia californica 5 5 0.997 Cresson 1864 Osmia cobaltina 1 1 1 Cresson 1878 Osmia coloradensis 2 2 0.989 Cresson 1878 Osmia ednae 1 1 1 Cockerell 1907 Osmia indeprensa 3 3 0.995 Sandhouse 1939 Osmia integra 4 4 0.989 Cresson 1878 Osmia juxta 7 7 0.972 Cresson 1864 Osmia kincaidii 1 1 1 Cockerell 1897 Osmia nigrifrons 3 3 0.978 Cresson 1878 Osmia obliqua 1 1 1 White 1952 Osmia raritatis 3 3 0.978 Michener 1957 Osmia simillima 2 2 0.989 Smith 1853 Osmia sp. 2 3 3 0.995 No author date Osmia subaustralis 1 1 1 Cockerell 1900 Osmia tanneri 1 1 1 Sandhouse 1939 Osmia tristella 5 5 0.984 Cockerell 1897 Stelis montana 1 1 1 Cresson 1864 Bees species observed on shrubs only Bee Species Forb Shrub Tree Total Bees PDI Score Reference

Andrena amphibola 3 3 0.995 (Viereck 1904) Andrena candida 1 1 1 Smith 1879 Andrena crataegi 3 3 0.995 Robertson 1893 Andrena flocculosa 13 13 1 LaBerge & Ribble 1972 Andrena hippotes 2 2 1 Robertson 1895

101

Andrena lupinorum 2 2 1 Cockerell 1906 Andrena medionitens 1 1 1 Cockerell 1902 Andrena milwaukeensis 1 1 1 Graenicher 1903 Andrena pallidifovea 1 1 1 (Viereck 1904) Andrena porterae 15 15 1 Cockerell 1900 Andrena prunorum 2 2 1 Cockerell 1896 Andrena schuhi 1 1 1 LaBerge 1980 Andrena thaspii 7 7 0.992 Graenicher 1903 Andrena topazana 1 1 1 Cockerell 1906 Andrena vicinoides 3 3 1 Viereck 1904 Andrena w-scripta 3 3 0.995 Viereck 1904 Bombus nevadensis 2 2 1 Cresson 1874 Bombus occidentalis 1 1 1 Greene 1858 Colletes kincaidii 1 1 1 Cockerell 1898 Habropoda cineraria 5 5 1 (Smith 1879) Hylaeus episcopalis 2 2 1 (Cockerell 1896) Lasioglossum sisymbrii 1 1 1 (Cockerell 1895) Megachile gemula 5 5 0.997 Cresson 1878 Melecta separata 1 1 0.989 Cresson 1879 Nomada sp. 3 3 1 No author date Nomada sp. 8 1 1 1 No author date Nomada sp. G 2 2 1 No author date Osmia lignaria 14 14 0.993 Cresson 1864, Say 1837 Osmia sp. Y5 1 1 1 No author date Stelis pavonina 1 1 1 (Cockerell 1908)

102

103

Figure A2.1: Pie charts of bee visits to flowers Bee foraging habits of 38 bee species represented by 10 or more individuals. Pie chart size is scaled to number of bees caught. Color corresponds to the color indicated in Figure 3.6 (proportion of individuals caught on shrubs). Plants were grouped to genus when multiple species within a genus were present in a sample. Bee species are ranked (left to right, top to bottom) by PDI value.