AN ABSTRACT OF THE THESIS OF

Chiho Kimoto for the degree of Master of Science in Wildlife Science presented on December 3, 2010. Title: Effect of Livestock Grazing on Native Bees in a Pacific Northwest Bunchgrass Prairie.

Abstract approved:

Sandra J. DeBano

Native bees play an important role as pollinators of natural vegetation and agricultural crops. Yet many pollinators, including some native bees, are declining in numbers.

Some of the potential causes of these declines are habitat destruction and degradation by various human land uses, including urban development and sprawl, construction of roadways, and habitat conversion to agriculture such as crop production and livestock grazing. Livestock grazing is one of the most common land uses in western North

America and it can impact floral and nesting resources that are important to native bees. These effects are likely manifested through grazing’s effect on vegetation and soil characteristics. However, few studies have investigated how livestock grazing impacts native bees in North America. As a result, the overall goal of this research was to determine how a gradient of livestock grazing intensities impacts native bee communities in a threatened and poorly studied grassland of western North America, the Pacific Northwest Bunchgrass Prairie. Because no studies have examined the bee

fauna of this grassland habitat, our study had two objectives: 1) describe the native bee community in the Zumwalt Prairie in northeastern Oregon, which is one of the largest remaining remnants of this unique grassland type, and how it changes through time, and 2) investigate how livestock grazing affects that native bee fauna.

To address these objectives, we sampled pollinators during the summer of

2007 and 2008 in 16 40-ha pastures on a plateau in the Zumwalt Prairie using blue vane traps. Each pasture was assigned one of four cattle stocking rates (high, medium low, and no cattle), and grazing intensity was quantified by measuring utilization.

Grazing treatments were applied in the early summer for two years. We measured soil and vegetation characteristics that related to floral and nesting resources of bees as well as several metrics of the bee community, including diversity, richness, abundance, and community composition.

We found 92 species and 119 morphospecies of native bees in 27 genera. This diverse community of native bees showed strong interseasonal and interannual variation that appears to be related to weather and plant phenology. We also found that even after exposure to just two years of grazing, some effects on vegetation and soils were evident. For example, increased grazing intensity significantly reduced vegetation structure, the abundance of certain blooming plants, surface soil stability, and the amount of soil surface covered by herbaceous litter. In addition, increased grazing intensity significantly increased soil compaction and the amount of bare ground. Native bee communities responded grazing intensity through changes in abundance, richness, diversity and community composition. Different bee taxa

responded to grazing intensity differently and this response varied temporally. For example, bumble bees were sensitive to grazing intensity early in the season, showing reduced abundance, diversity, and/or richness with increased grazing intensity. In contrast, halictid bees did not respond to grazing intensity in any season. However, even within a genus or family, different species responded to grazing intensity in different manners, potentially because of variation in life histories. This research suggests that maintaining land with a mixture of livestock grazing intensities may be the best way to conserve this important and diverse pollinator group in the Pacific

Northwest Bunchgrass Prairie.

©Copyright by Chiho Kimoto December 3, 2010. All Rights Reserved

Effect of Livestock Grazing on Native Bees in a Pacific Northwest Bunchgrass Prairie

by Chiho Kimoto

A THESIS

Submitted to

Oregon State University

in partial fulfillment of the requirements for the degree of

Master of Science

Presented December 3, 2010 Commencement June 2011

Master of Science thesis of Chiho Kimoto presented on December 3, 2010.

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.

Chiho Kimoto, Author

ACKNOWLEDGEMENTS

I would like to thank my academic advisor and committee members. Dr. Sandra J.

DeBano was always there for me whenever I needed her help to understand the various aspects of the research and encouraged me through my master student’s life.

She introduced me to the beauty of terrestrial insects and always amazed me with not only her knowledge of science but also her knowledge of writing. I also would like to thank Dr. Andrew R. Moldenke who spent many hours identifying thousands of bees to species or morphospecies and was very generous in training me in the identification of some bee groups. I was always fascinated by his knowledge of bees. Whenever I had questions about scientific papers, he provided me with great insights; he also reviewed my manuscripts and gave me a great deal of detailed information. I also would like to thank Dr. Robert V. Taylor who provided me with great and useful input on my manuscripts and interesting scientific papers. In the field, I was always impressed by his knowledge of all aspect of plants and the Zumwalt Prairie.

I would like to thank Dr. Robbin W. Thorp who identified all Bombus bees to species at the University of California, Davis. Thank you very much for spending so many hours identifying bees and your great generosity in training me how to identify bumble bees to species. I was very fascinated by his knowledge of bees and willingness to share it with me.

I also would like to thank Drs. William P. Stephen and Sujaya Rao who provided bee training sessions and laboratory space during those bee sessions as I learned about bee morphology at Oregon State University. Thank you very much, Dr,

ACKNOWLEDGEMENTS (Continued)

Sujaya Rao, for reviewing my manuscripts and providing me great insights. I would like to say thank you very much to Dr. William P. Stephen for providing me with his fascinating knowledge about bees. I also appreciate very much his willingness to share his great knowledge about bees whenever I have questions about them.

I also would like to thank Drs. Terry L. Griswold and Molly G. Gee, who identified Megachile and Osmia to species at the USDA-ARS Bee Biology and

Systematics Laboratory, Utah State University, Logan. This detailed information improved my thesis very much.

I appreciate The Nature Conservancy for letting us conduct this study on their land and providing facilities.

I also would like to thank Dr. Christopher Marshall for providing me laboratory space at the Oregon State Arthropod Collection. It was a great experience for me to work in the museum so that I had opportunities to see so many beautiful terrestrial insects. I also offer my thanks to The Bohart Museum of Entomology at the

University of California, Davis, for providing me with workspace while I was being trained in bumble bee identification by Dr. Robbin W. Thorp.

I have many thanks for the hardest worker, Dr. Mahmut Doğramaci. Ms. Anne

Madsen and Ms. Kimberly Tanner helped with field work during very hot times of the year. I am very thankful to Ms. Heidi Schmalz and Ms. Tonya Vowles for helping me to identify all the blooming plants to species. I also would like to thank Mr. Austin

Gurwell and Ms. Merrie Richardson for putting identification tags on each bee

ACKNOWLEDGEMENTS (Continued) specimen. Your quality of work was greatly appreciated. I also would like to thank Ms.

Sam Colby and Ms. Ariadne DeMarco, who assisted me in the laboratory. I am grateful to Drs. David E. Wooster, Robert V. Taylor, Robbin W. Thorp, Andrew R.

Moldenke, and Sujaya Rao for reviewing my manuscripts and providing me great insights for improving my thesis.

Many thanks to Dr. Robert V. Taylor, Dr. Patricia Kennedy, Dr. Timothy

DelCurto, Ms. Heidi Schmaltz, Dr. Tracy Johnson, and Mr. Sam Wyfells for being so successful with their research. I appreciated all The Nature Conservancy staff members, Dr. Robert V. Taylor, Ms. Liza Jane Nichols, Mr. Phil Shepherd, and Ms.

Angie Freeman for providing me with housing facilities

Thank you very much to the funding agencies that made this research possible.

This project was supported by an Oren Pollak Research Grant from The Nature

Conservancy, a USDA National Research Initiative grant (#2006-35101-16572), and a

US EPA STAR grant (#RD83456601-0).

Finally, I would like to thank my family, especially my father Osami Kimoto, my mother Yoriko Kimoto, my older brother Takatomi Kimoto, relatives, friends, and my fellow graduate students who supported me during my graduate career. They always believed in me, supported me and encouraged me while I was at school.

Thank you very much all of you once again. All of your hard work and all of your support helped make this research successful.

CONTRIBUTION OF AUTHORS

Dr. Sandra J. DeBano helped design and conduct the research; collect, analyze, and interpret the data; and revise the manuscripts. Dr. Andrew R. Moldenke identified all bees, except bumble bees and Osmia, to species or morphospecies; trained me how to identify some of the groups to species or morphospecies; helped interpret the data; and reviewed my manuscripts. Dr. Robbin W. Thorp identified bumble bee specimens to species; trained me to identify bumble bee species; and reviewed my manuscripts. Drs.

Molly G. Gee and Terry L. Griswold identified Osmia to species. Dr. Sujaya Rao gave me the idea to use blue vane traps, and loaned me traps during my first field season.

Dr. William P. Stephen trained me initially in bee identification. Dr. Robert V. Taylor supervised soil and vegetation data collection; compiled that data; made that data available for my use; and reviewed my manuscripts. Dr. Tracey Johnson, Ms. Heidi

Schmaltz, and Mr. Samuel Wyffels shared their data on plant structure, soils, and utilization with me. Drs. Patricia Kennedy and Timothy DelCurto assisted in the experimental design and set up.

TABLE OF CONTENTS

Page

GENERAL INTRODUCTION ...... 1

Literature Cited ...... 7

INVESTING TEMPORAL PATTERNS OF A NATIVE BEE COMMUNITY IN NATIVE BEE COMMUNITY IN A REMNANT NORTH AMERICAN BUNCHGRASS PRAIRIE USING BLUE VANE TRAPS ...... 10 Abstract ...... 10

Introduction ...... 11

Methods ...... 15 Study Area ...... 15 Pollinator Sampling…………………………………..…………………..15 Floral Resources and Weather……………………………………………16 Data Summary and Analyses...... 17

Results ...... 18 Abundance, Richness, Eveness, Diversity, and Sex Ratio of the Bee Community...... 18 Patterns in Bee Genera between Years ...... 20 Seasonal Patterns in Bee Genera within Years ...... 21 Total Abundance, Richness, Eveness, Diversity, and Sex Ratio of Bombus...... 21 Patterns in Bombus Species between Years ...... 22 Seasonal Patterns in Bombus within Years ...... 23 Spatial and Temporal Variability in Abundance and Taxa Richness….....23 Floral Resources and Weather ...... 24

Discussion……………………………………………………………………....25 Pollinator Communities in the Pacific Northwest Bunchgrass Prairie ...... 25 Temporal Variability in Bee Communities ...... 27 Availability of Pollinators for a Threatened Grassland Plant Species ...... 32 Sampling Native Bee Fauna in Grasslands with Blue Vane Traps ...... 33 Conclusions and Broader Implications for Conservation ...... 37 Literature Cited ...... 39

TABLE OF CONTENTS (Continued) Page

EFFECT OF LIVESTOCK GAZING ON NATIVE BEE COMMUNITIES OF A PACIFIC NORTHWEST BUNCHGRASS PRAIRIE ...... 62 Abstract ...... 62

Introduction ...... 63

Methods ...... 69 Study Area ...... 69 Study Design ...... 70 Environmental Variable Measurement ...... 71 Pollinator Sampling ...... 73 Statistical Analyses ...... 74 Results ...... 77 Environmental Variables ...... 77 Patterns in Abundance ...... 78 Patterns in Richness and Diversity ...... 79 Patterns in Community Composition ...... 80 Discussion ...... 81 Effect of Grazing Intensity on Native Bee Diversity ...... 82 Effect of Grazing Intensity on Native Bee Diversity and Community Composition...... 84 Effet of Grazing Intensity on Common Taxa ...... 85 Temporal Variability ...... 93 Management Implications ...... 95 Literature Cited ...... 97

SUMMARY ...... 122

BIBLIOGRAPHY ...... 125

LIST OF FIGURES

Figure Page

2.1 a) Location of the Zumwalt Prairie in northeastern Oregon and location of pollinator traps in each pastures (all sites were sampled in each season of each year except that unshaded pastures and traps denoted with “*” were not sampled in June 2007) b) a blue vane trap…………………………………….…………..…....…….44 2.2 Estimated richness using rarefaction (Chao 1) in 2007 and 2008 for a) June, b) July, and c) August (2008 only). Note scale of both axes differ in each panel……………..…………..………45

2.3 Mean abundance of bees adjusted by trapping effort for common genera in 2007 and 2008 in a) June, b) July, and c) August (2008 only). Error bars are 95% confidence intervals……………..……….………………46

2.4 a) Mean abundance of bees adjusted by trapping effort of common Bombus species in 2007 and 2008 in a) June, b) July, and c) August (2008 only). Error bars are 95% confidence intervals…..…….47

2.5 Values of a) mean average daily temperature (°C) and b) total rainfall (cm) for spring and summer months in 2007 and 2008………48

3.1 a) Location of the Zumwalt Prairie in northeastern Oregon and location of pollinator traps in each pasture (all sites were sampled in each season of each year except that unshaded pastures and traps denoted with “*” were not sampled in June 2007), b) map showing stocking levels for each 40 ha pasture (H-high, M-medium, L-low, C-control) and a close-up view of the 36 subplots in each pasture, where vegetatopm amd soil characteristics were sampled, and c) a blue vane trap…………..……………………………….…..….……….105

3.2 Average utilization in a) 2007, and b) 2008……….…………………..…….106

3.3 Effect of grazing intensity on a) vegetation physical structure in 2007 and 2008, b) blooming forb abundance in July 2008, c) blooming balsamroot abundance in June 2008, d) blooming twin arnica abundance in July 2008, and e) dissimilarity of plant community composition from pre- to post-treatment…………………107

LIST OF FIGURES (Continued)

Figure Page

3.4 Effect of grazing intensity on a) soil compaction, b) surface soil stability, c) percent bare ground, and d) percent herbaceous litter…….……………………………………………………….108

3.5 Effect of grazing intensity on bee abundance for a) all bees in July 2007, b) Bombus in June 2007, and c) Bombus, Andrena, and Osmia in June 2008………………………………………………….….109

3.6 Effect of grazing intensity on bee richness of a) all bees and Bombus in June 2007, b) common genera in June 2008, and c) Megachile in August 2008……………………………………………..…110

3.7 Effect of grazing intensity on the Shannon diversity index of all bees and Bombus for a) June 2007, b) June 2008, c) July 2007, d) July 2008, and e) August 2008…..………………………...111

3.8 Bee community composition a) NMS scores for axis 2 of ordination for all bees in June 2008, b) total abundance of taxa showing significant correlations with axis 2 in June 2008, c) NMS scores axis 1 ordination for Bombus for June 2007 and d) abundance of taxa showing significant relationships with axis 1 in June 2007……………………………………………………...……...…..112

3.9 Effect of grazing intensity on the abundance of the most common species in June 2007: a) Lasioglossum morphospecies 8, b) Bombus flavifrons, c) Lasioglossum sisymbrii, d) Lasioglossum morphospecies 6………………………………………………………….….113

3.10 Effect of grazing intensity on the abundance of the most common species in July 2007: a) Bombus bifarius, b) Lasioglossum morphospecies 1, c) Bombus californicus, d) Lasioglossum morphospecies 3, and e) Bombus flavifrons…………….………….…...…...114

3.11 Effect of grazing intensity on the abundance of the most common species in June 2008: a) Lasioglossum morphospecies 1, b) Lasioglossum morphospecies 2, c) Lasioglossum morphospecies 3, d) Bombus californicus, and e) Bombus flavifrons…………………………………………………...……115

LIST OF FIGURES (Continued)

Figure Page

3.12 Effect of grazing intensity on the abundance of the most common species in July 2008: a) Melissodes rivalis, b) Melissodes agilis, c) Bombus flavifrons, and d) Bombus californicus……………………….…116

3.13 Effect of grazing intensity of the abundance of the most common species in August 2008: a) Lasioglossum morphospecies, b) Melissodes confusa, c) Megachile latimanus, d) Bombus californicus, and e) Lasioglossum morphospecies 5………...…117

LIST OF TABLES

Table Page

2.1 List of bee taxa found in the Zumwalt Prairie and their presence/absence for each season. Number following genus title is the number of morphospecies identified in that genus for that season………………………………………..…………………………...49

2.2 Sex ratios, as expressed by proportion of males, of common genera (>20 specimens) in each season. Sex ratios with more than 10% males are in bold type. “NA” indicates “non-applicable” because no sampling occurred in August 2008…………………..…….…………………55

2.3 Relative abundance of most common genera and Bombus species (>5%) in each season of each year……………………………..…….56

2.4 Percentage of males, queens, and workers in Bombus species in July 2007 and 2008. June and August are not shown because of the relatively low numbers collected (<30 for all species)………….…..……57

2.5 List of plant species blooming within 50 m of blue vane traps during each sampling period in 2008. Means are reported ± one standard error……………………………….….…59

2.6 Summary of studies of bee communities in North American grasslands. All studies were conducted in the Great Plains region of the USA. “NR” means “not reported”……………………………………………..……61

3.1 Results of linear regression models of abundance (number of bees per trap per hour) with utilization for all bees and common genera for each season. Relative abundance (“Rel Abund”) of common genera (>7% relative abundance) are listed for each season..….…118

3.2 Results of linear regression models of species richness and diversity with utilization for all bees and common genera. Total number of species for common genera (>7% relative abundance) are listed for each season………………………………………………………………119

3.3 Result of linear regression model of community composition with utilization for all bees and Bombus. The percent variation explained by each NMS axis listed……………………………………………………..120

LIST OF TABLES (Continued) Table Page

3.4 Results of lienear regression models of abudance (number of bees per trap per hour) with utilization for common species for each season. Relative abundance (“Rel Abund”) of each species (>7%) are listed for each season.…………………………………....121

1

GENERAL INTRODUCTION

Pollination is one of the most important ecosystem functions that animals fulfill.

Pollinators include vertebrates, such as bats, birds and some primates as well as many invertebrates, such as bees, wasps, beetles, butterflies, flies, and moths (Allen-Wardell et al. 1998) It is estimated that over 75% of the 250,000 species of flowering plants, including crops that make up 35% of the world’s food supply, are pollinated by animals (Klein et al. 2007; NRC 2007). One of the most important groups of pollinators is bees.

Bees pollinate many different crops and wild plants efficiently (Batra 1995;

Klein et al. 2007). Their pollination of agricultural crops is of great economic benefit to humans, with an estimated value of billions of dollars annually (Losey & Vaughn

2006; NRC 2007). Also of economic benefit are the byproducts of pollination by

European honey bees (Apis mellifera): honey and wax (LaSalle & Gauld 1993). In addition to their economic value to humans, the fitness of many non-self pollinated, non-cultivated plants depends on bee pollination. For example, many plants have co- evolved mutualistic relationships with particular pollinators (Kearns et al. 1998).

Some bees have evolved unique forms of body structure that allow them to specialize in obtaining nectar of certain types of flowers efficiently (Michener 2007). Without pollinators, some plants can self-pollinate but inbreeding depressions may result

(Michener 2007). In other words, bee pollinators are essential to maintaining the genetic diversity of many plant species. Bee pollination also indirectly impacts habitat 2

quality for other organisms. For example, some xeric shrubs that are pollinated by bees help prevent erosion and provide cover and food for wildlife (Michener 2007).

As a result, it is essential to maintain bee pollinators to protect many habitats. Finally, bees are a diverse group and thus contribute significantly to pollinator biodiversity.

World-wide, approximately 17,500 species of bees have been named, with many more undescribed, and all are obligate flower visitors (Michener 2007). The diversity of this group can be appreciated by comparing them to well-known vertebrate groups; there are more species of bees than birds and mammals combined (LaSalle & Gauld 1993).

Although European honey bees have received much attention for their role as pollinators in the USA, bees native to this country are also important (Allen-Wardell et al. 1998). First, they contribute up to $6.7 billion per year through crop pollination

(Buchmann & Nabhan 1996). Second, some native bees can forage in severe conditions such as windy, rainy, or cloudy days when honey bees cannot tolerate the conditions (Goulson 2003). Third, because they are native, they do not cause serious damage to the environment as do many introduced species (Thomson 2004). Fourth, native bees are not as expensive to use for crop pollination as honey bees. Farmers must raise or rent honey bees for their use as pollinators, while native bees are part of the natural ecosystem. Fifth, native bees, or at least non-Apis groups, appear to be less sensitive to parasites and pathogens that have caused decreases in populations of

European honey bees, such as Varroa destructor, a parasitic mite that may be one factor associated with Colony Collapse Disorder (NRC 2007; Winfree et al. 2007).

Thus, native bees may have the potential to play a more important role in pollinating 3

agricultural crops given the decline in numbers of European honey bees due to Colony

Collapse Disorder; in fact, recent research has supported this hypothesis (Winfree et al.

2007).

There is evidence suggesting that we are currently in the midst of a world-wide pollinator crisis, in which many invertebrate pollinator species, including the

European honey bee, as discussed above, are experiencing large declines (Buchmann

& Nabhan 1996; Kearns et al. 1998; São Paulo Declaration on Pollinators 1999; NRC

2007). Although debate continues over the extent of these declines, both taxonomically and geographically, and their effect on crop pollination (e.g., Ghazoul

2005, Ghazoul and Koh 2010), most research on the subject has focused on European honey bees because they are important pollinators of many agricultural products on which humans depend (Kearns et al. 1998; NRC 2007). However, European honey bees are not the only pollinators in decline; some native bees that are valued as important pollinators are also declining (Allen-Wardell et al. 1998; Biesmeijer et al.

2006; NRC 2007). For example, there is evidence that five native species of bumble bees (Bombus affinis, B. lucorum, B. franklini, B. occidentalis, and B. terricola) are either declining or are extinct in North America (Thorp 2003; Thorp & Shepherd

2005; Rao & Stephen 2007).

There are several reasons why native bees may be declining. First, although native bees may not be sensitive to many of the same parasites and pathogens that are currently having large impacts on European honey bees, there is evidence that some species of native bees can be negatively impacted by other parasites and pathogens. 4

For example, two protozoan species, Nosema bombi and Crithidia bombi, are believed to be partially responsible for the decline of some western bumble bee species, such as

B. occidentalis (Thorp 2003; Rao & Stephen 2007). These diseases were probably introduced to wild populations from individuals produced by captive mass rearing programs. Other factors that may be negatively impacting native bees are the introduction of non-native species, such as the European honey bee, and pesticide use

(Kearns et al. 1998; Thomson 2004; NRC 2007). Finally, native bees are believed to be impacted by habitat destruction and degradation caused by various human land uses.

Habitat destruction and degradation can impact a variety of resources important to pollinators, including host flower and nesting resources (NRC 2007). Urban development and sprawl, agricultural conversion, construction of roadways, and livestock grazing may all decrease the quality or amount of habitat for bees (NRC

2007). The impact of crop production and grazing is particularly pervasive, especially in grassland habitats. Historically, 42% of the world’s land was covered with grasslands; it is now less than 13% (Smith & Smith 2000).

Grasslands that have not been converted to agriculture are often used for livestock grazing, especially in the western United States, where it is economically important to many rural communities (Knight et al. 2002). Yet we know relatively little about how this land use affects native pollinators in this region’s grasslands.

Livestock grazing can impact bee communities via its influence on floral resources and nesting habitat, which are key habitat requirements of bees (Westrich 1996; Cane

2001). Of studies that have examined the effect of grazing on pollinators, most have 5

had significant limitations. Most work on the subject has been observational rather than manipulative, which limits the ability to establish cause and effect. Others have simply compared the presence or absence of grazing, rather than examining a variety of grazing intensities from which land managers may choose. Others have limited taxonomic resolution, making a detailed understanding of the effect of grazing on native bee communities difficult. Most importantly, there are few studies that have examined the effects of grazing intensity on native bee communities in any North

American grassland.

The need to understand the importance of these threatened grasslands as habitat for native bees and to understand how human land uses, such as livestock grazing, impact them was the motivation for this study. I chose to conduct my study at

The Nature Conservancy’s (TNC) 13,269 ha Zumwalt Prairie Preserve (lat 45º 3’N, long 116º 6’ W) in Wallowa County of northeastern Oregon, USA. This is the largest remnant of the once extensive Pacific Northwest Bunchgrass Prairie (Tisdale 1982). It is located at an elevation of 1,100-1,700 m and receives an average of ~43 cm/yr mean annual precipitation (30-year average, NOAA 2010). Although the Zumwalt Prairie has been used as summer pasture for horse, cattle, and sheep for over 100 years, the majority of the area remains dominated by native species with most ecological processes still intact, and is considered to be an important refuge for native biodiversity, including vertebrates, invertebrates, and plants (Kennedy et al. 2009).

The Zumwalt Prairie was also the location of a large-scale, manipulative livestock 6

grazing experiment, and I had the opportunity to examine the response of the native bees to grazing intensity in this unique habitat.

Thus, I had two major goals for this study. My first goal was to describe the native bee community in this portion of the Pacific Northwest Bunchgrass Prairie and quantify its seasonal and inter-annual variation. Although a few studies have been conducted in Great Plains grassland systems in the U.S. (Tepedino & Stanton 1981;

Hines & Hendrix 2005; Kearns & Oliveras 2009), no work describing the native bee communities in this threatened habitat has ever been conducted. Thus, meeting this goal would fulfill that need and provide baseline data that will be useful in monitoring the community effectively for future changes. I describe this study in Chapter 2. My second goal was to conduct the first manipulative study examining a gradient of livestock grazing intensities on a native bee community in a North American grassland.

This information will aid land managers and ranchers in understanding the potential effect of this land use on the providers of one ecosystem service, pollination, as they develop sustainable grazing practices. This work is described in Chapter 3. In my last chapter, I summarize the major conclusions and discuss the implications of this research.

7

Literature Cited

Allen-Wardell, G., et al. 1998. The potential consequences of pollinator declines on the conservation of biodiversity and stability of food crop yields. Conservation Biology 12:8-17.

Batra, S. W. T. 1995. Bees and pollination in our changing environment. Apidologie 26:361-370.

Biesmeijer, J. C., et al. 2006. Parallel declines in pollinators and insect-pollinated plants in Britain and the Netherlands. Science 313:351-354.

Buchmann, S. L., and G. P. Nabhan. 1996. The forgotten pollinators. Island Press, Washington, D.C.

Cane, J. H. 2001. Habitat fragmentation and native bees: a premature verdict? Conservation Ecology 5:3.

Ghazoul, J. 2005. Buzziness as usual? Questioning the global pollination crisis. Trends in Ecology and Evolution 20:367-373.

Ghazoul, J., and L. P. Koh. 2010. Food security not (yet) threatened by declining pollination. Frontiers in Ecology and the Environment 8:9-10.

Goulson, D. 2003. Bumblebees behaviour and ecology. 2003. Oxford University Press, United Kingdom.

Hines, H. M., and S. D. Hendrix. 2005. Bumble bee (Hymenoptera: Apidae) diversity and abundance in tallgrass prairie patches: effects of local and landscape floral resources. Environmental Entomology 34:1477-1484.

Kearns, C. A., D. W. Inouye, and N. M.Waser. 1998. Endangered mutualisms: the conservation of plant-pollinator interactions. Annual Review of Ecology and Systematics 29:83-112.

Kearns, C. A., and D. M. Oliveras. 2009. Environmental factors affecting bee diversity in urban and remote grassland plots in Boulder, Colorado. Journal of Insect Conservation 13:655-665.

Kennedy, P. L., S. J. DeBano, A. M. Bartuszevige, and A. S. Lueders. 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.

8

Klein, A. M., B. E. Vaissière, J. H. Cane, I. Steffan-Dewenter, S. A. Cunningham, C. Kremen, and T. Tscharntke. 2007. Importance of pollinators in changing landscapes for world crops. Proceedings of the Royal Society of London Series B-Biological Sciences 274:303-313.

Knight, R. L., W. C. Gilgert, and E. Marston. 2002. Ranching west of the 100th meridian: culture, ecology, and economics. Island Press, Washington, D.C.

LaSalle, J., and I. D. Gauld. 1993. Hymenoptera: their diversity, and their impact on the diversity of other organisms. Pages 1-26 in J. La Salle and I. D. Gauld, editors. Hymenoptera and biodiversity. CAB International, Wallingford, United Kingdom.

Losey, J. E., and M. Vaughan. 2006. The economic value of ecological services provided by insects. BioScience 56:311–323.

Michener, C. D. 2007. The bees of the world. 2nd edition. Johns Hopkins University Press, Baltimore, Maryland.

(NOAA) National Oceanic and Atmospheric Administration. 2010. National Weather Service Office. Available from http://www.weather.gov/climate/xmacis.php?wfo=pdt (accessed December 2010).

(NRC) National Research Council. 2007. Status of pollinators in North America. The National Academies Press, Washington, D.C.

Rao, S., and W. P. Stephen. 2007. Bombus (Bombus) occidentalis (Hymenoptera: Apiformes): In decline or recovery. The Pan-Pacific Entomologist 83:360-362.

São Paulo Declaration on Pollinators. 1999. Report on the recommendations of the workshop on the conservation and the sustainable use of pollinators in agriculture with emphasis on bees. Brazilian Ministry of the Environment, Brasilia, Brazil.

Smith, R. L., and T. M. Smith. 2000. Elements of ecology. 4th edition. Addison Wesley Longman, Benjamin/Cummings, Menlo, California.

Tepedino, V. J., and N. L. Stanton. 1981. Diversity and competition in bee-plant communities on short-grass prairie. Oikos 36:35-44.

Thomson, D. 2004. Competitive interactions between the invasive European honey bee and native bumble bees. Ecology 85:458-470.

9

Thorp, R. W. 2003. Bumble bees (Hymenoptera:Apidae): commercial use and environmental concerns. Pages 21-40 in K. Strickler and J. H. Cane, editors. For nonnative crops, whence pollinators of the future? Thomas Say Publications in Entomology: Proceedings. Entomological Society of America, Lanham, Maryland.

Thorp, R. W., and M. D. Shepherd. 2005. Profile. Subgenus Bombus Latreille 1802 (Apidae: Apinae: Bombini). Page 5 in M. D. Shepherd, D. M. Vaughan, S. H. Black, editors. Red list of pollinator insects of North America. Xerces Society for Invertebrate Conservation, Portland, Oregon.

Tisdale, E. W. 1982. Grasslands of western North America: the Pacific Northwest Bunchgrass. Pages 223-245 in A. C. Nicholson, A. McLean and T. E. Baker, editors. Grassland ecology and classification symposium proceedings. Information Services Branch B.C. Ministry of Forests, Victoria, British Columbia.

Westrich, P. 1996. Habitat requirements of central European bees and the problems of partial habitats. Pages 1-16 in A. Matheson, S. L. Buchmann, C. O’Toole, P. Westrich and I. H. Williams, editors. The conservation of bees. Academic Press, London.

Winfree, R., N. M. Williams, J. Dushoff, and C. Kremen. 2007. Native bees provide insurance against ongoing honey bee losses. Ecology Letters 10:1105-1113.

10

INVESTIGATING TEMPORAL PATTERNS OF A NATIVE BEE

COMMUNITY IN A REMNANT NORTH AMERICAN BUNCHGRASS

PRAIRIE USING BLUE VANE TRAPS

Abstract

Native bees are important ecologically and economically because their role as pollinators fulfills a vital ecosystem service. Pollinators are declining due to various factors, including habitat degradation and destruction. Grasslands, an important habitat for native bees, are particularly vulnerable. One highly imperiled and understudied grassland type in the United States is the Pacific Northwest Bunchgrass Prairie. No studies have examined native bee communities in this prairie type. To fill this gap, we examined the bee fauna of the Zumwalt Prairie, a large, relatively intact remnant of the Pacific Northwest Bunchgrass Prairie. We sampled pollinators during the summers of 2007 and 2008 in 16 40-ha study pastures on a plateau in northeastern Oregon using a sampling method not previously used in grassland studies – blue vane traps. We found that this grassland contained an abundant and diverse community of native bees that experienced marked seasonal and inter-annual variation, which appears to be related to weather and plant phenology. This temporal variability in bee abundance and taxa richness was relatively large in magnitude compared to the spatial variability detected over the sampling area. These results demonstrate that temporal variability in bee communities can have important implications for long-term monitoring protocols.

In addition, we found that the blue vane trap method appears to be well-suited for 11

studies of native pollinators in large expanses of grasslands or other open habitats and suggest that it may be a useful tool for monitoring native bee communities in these systems.

Keywords: bee monitoring, community composition, diversity, grasslands, native bees, Pacific Northwest Bunchgrass Prairie, pollinatiors, temporal variability

Introduction

Pollination is one of the most important ecosystem functions that animals fulfill. It is estimated that over 75% of the 250,000 species of flowering plants, including crops that make up 35% of the world’s food supply, are pollinated by animals (Klein et al.

2007; NRC 2007). One of the most important groups of pollinators is bees. Bees pollinate many different crops and wild plants efficiently (Batra 1995; Klein et al.

2007). Their pollination of agricultural crops is of great economic benefit to humans, with an estimated value of billions of dollars annually (Losey & Vaughan 2006; NRC

2007). In addition to their economic value to humans, the fitness of many cross- pollinated, non-cultivated plants depends on bee pollination. Even plants capable of self-pollination may benefit from pollinators through higher seed set and a reduction in inbreeding depression (Michener 2007). Thus, bee pollinators are essential to maintaining the genetic diversity of many plant species.

There is evidence suggesting that we are currently in the midst of a global pollinator crisis, in which many invertebrate pollinator species are experiencing large declines (Buchmann & Nabhan 1996; Kearns et al. 1998; São Paulo Declaration on 12

Pollinators 1999; NRC 2007). These declines not only include the domestic honey bee

(Apis mellifera) but also native bees that act as important pollinators in both natural and agricultural systems (Kearns et al. 1998; Allen-Wardell et al. 1998; Biesmeijer et al. 2006; NRC 2007). Although less well-studied than domestic honey bees, there are several reasons why some native bees are declining. First, while native bees may not be sensitive to many of the same parasites and pathogens that are currently impacting

A. mellifera, they are impacted by others, such as two protozoan species, Nosema bombi and Crithidia bombi that infect bumblebees (Thorp 2003; Rao & Stephen 2007).

Other factors that may be negatively impacting native bees are the overuse of insecticides, and the introduction of non-native species, such as A. mellifera (Kearns et al. 1998; Thomson 2004; NRC 2007). Finally, native bees can be negatively impacted by habitat destruction and degradation caused by human activities, such as urban development, construction of roadways, and agriculture, including crop production and livestock grazing (NRC 2007; Michener 2007). The impact of crop production and grazing is particularly pervasive, especially in grassland habitats. Historically, 42% of the land surface on earth was covered with grasslands; this coverage is now less than

13% (Smith & Smith 2000). Thus, documenting the diversity of pollinator communities in these threatened habitats and monitoring them effectively for future changes are pressing conservation priorities.

One of the most threatened and understudied grasslands in North America is the Pacific Northwest Bunchgrass Prairie, which historically covered over 8 million hectares in Oregon, Washington, Idaho, and Montana in the U.S. and British Columbia 13

and Alberta in Canada (Tisdale 1982). Over 90% of this unique grassland type has been converted to agriculture, yet little is known about the native bee communities that inhabit it. Although several studies have examined native bee communities and their temporal variability in tallgrass and shortgrass prairie in the Great Plains region of the Unites States (Tepedino & Stanton 1981; Hines & Hendrix 2005; Davis et al.

2008; Kwaiser & Hendrix 2008; Kearns & Oliveras 2009) and general descriptions of bee fauna of the Pacific Northwest and California exist (Stephen et al. 1969;

Moldenke & Neff 1974; Thorp et al. 1983), no published work has described native bee communities of the Pacific Northwest Bunchgrass Prairie. Here, we present results from a study examining the native bee community in this grassland type, at the

Zumwalt Prairie in northeastern Oregon.

Because of its moderately high elevation (>1,500 m), short growing season

(<150 days), and aridity (precipitation < 50 cm/yr) the Zumwalt Prairie has largely escaped conversion to cropland and thus is one of the largest (~ 65,000 ha) remaining remnants of the Pacific Northwest Bunchgrass Prairie (Kennedy et al. 2009). In addition, the Zumwalt Prairie also contains the largest known population of a threatened plant species, Silene spaldingii (Caryophyllaceae), which requires pollinators to maintain viable populations (Lesica 1993; Taylor et al. 2009).

Effectively managing this species, as well as Pacific Northwest Bunchgrass Prairie in general, depends on acquiring baseline data on the abundance and temporal variability of native pollinators. 14

This study also examined the potential usefulness of a trapping technique that has not been used in grassland habitat before to provide baseline data to compare with future changes in bee fauna. We used blue vane traps, which were originally designed for the collection of beetles, to sample native bees in grasslands. Stephen and Rao

(2005, 2007) discovered these beetle traps also attract a variety of native bees, and have used them in several studies of agroecosystems (Stephen & Rao 2005, 2007; Rao

& Stephen 2009, 2010; Stephen et al. 2009). This trapping method adds another option for sampling native bees to the traditional methods used that include sweep- and hand- netting, visual observations in plots or along transects, pan traps, and trap nests

(reviewed by Westphal et al. 2008). Ideal sampling methods for monitoring native bees would be those that are easily used with minimal training, would reduce the potential for observer bias or trap bias against particular taxa or sexes, and would be equally effective in all habitats. Evidence collected in agricultural systems suggests that blue vane traps may be more efficient at collecting greater numbers of individuals and species than some other trapping techniques, such as netting, and that the traps have some logistical advantages as well (Stephen & Rao 2007).

Thus, the objectives of this study were to use blue vane traps to: 1) describe the taxonomic composition, richness, diversity, and sex ratios of the bee community of the

Zumwalt Prairie of northeastern Oregon; 2) quantify seasonal and inter-annual variation in these characteristics; and 3) compare the relative magnitude of spatial and temporal variability in bee abundance and richness in this grassland. 15

Methods

Study Area

We conducted the study within The Nature Conservancy’s (TNC) 13,269 ha Zumwalt

Prairie Preserve (lat 45º 3’N, long 116º 6’ W) in Wallowa County of northeastern

Oregon, USA. Although the Zumwalt Prairie has been used as summer pasture for horse, cattle, and sheep for over 100 years, the majority of the area remains dominated by native species with most ecological processes still intact, and is considered to be an important refuge for an array of native biodiversity, including vertebrates, invertebrates, and plants (Kennedy et al. 2009). The Zumwalt Prairie is dominated by native grass species including Idaho fescue (Festuca idahoensis), Sandberg bluegrass

(Poa secunda), prairie Junegrass (Koeleria macrantha), and bluebunch wheatgrass

(Pseudoroegneria spicata) (Kennedy et al. 2009). In addition, over 112 forb species have been documented in the Zumwalt Prairie Preserve

(http://conserveonline.org/workspaces/ZumwaltPrairieWorkspace/documents/zumwalt

-prairie-plant-list-.pdf/view.html).

Pollinator Sampling

We sampled pollinators during the summers of 2007 and 2008 in 16 40-ha study pastures on a plateau located in the center of the TNC Zumwalt Prairie Preserve (Fig.

2.1a) using ultra-violet reflective blue vane traps. Blue vane traps consist of a plastic container (15 cm diameter x 15 cm high) with a blue polypropylene screw funnel with two 24 x 13 cm semitransparent blue polypropylene cross vanes of 3 mm thickness

(SpringStar™ LLC, Woodinville, WA, USA) (Stephen & Rao 2005). Traps were 16

suspended approximately 1.2 m from the ground with wire hangers inserted into aluminum pipes (Fig. 2. 1b). No liquids or other killing agents were used in traps.

Bees were sampled during two bouts in 2007 (June 18-20 and July 9-21) and three bouts in 2008 (June 7-16, July 10-18, and August 25-29). In June 2007 we sampled using 16 blue vane traps in 8 pastures; for all other sampling bouts we used

64 traps in 16 pastures (Fig. 2.1a). Elevation of traps ranged from 1372 to 1499 m. In

2007, traps were left open for two consecutive days and, because of high efficiency demonstrated in the first year, in 2008 they were left open for one day. Bees collected in the traps were frozen until they could be pinned, labeled, sexed, and identified to species, if possible, or morphospecies, if species identification was not possible.

Halictus and Lasioglossum were considered separate genera following Michener

(2007). We treat B. californicus and B. fervidus as two separate species, although uncertainty over their status is ongoing (Williams 2010). Representative specimens of all species and morphospecies are vouchered at the Oregon State Arthropod Collection at Oregon State University in Corvallis.

Floral Resources and Weather

Data on the presence of flowering plants and weather were also collected. We collected data on the presence of blooming forbs in 2008 along 50 m long, 0.3 m wide belt transects centered on each pollinator trap. The species and number of stems of each blooming forb that fell within the belt transect were recorded during each sampling period. Weather data were collected at a weather station located in the center 17

of the Zumwalt Prairie Preserve (lat 45º 34’39.88’N, long 116º 58’18.31’ W, elevation

1337 m) and less than 3 km from the nearest blue vane trap.

Data Summary and Analyses

To address our first two objectives (describing the overall native bee community of the grassland of the Zumwalt Prairie Preserve and how it changes through time), we characterized the native bee community with respect to abundance, richness, eveness, diversity, and species composition. Because sampling effort varied, both with regard to the number of traps used for each sampling bout and the time they operated, bee abundances were standardized by expressing them as the number of bees collected per trap per hour of daylight (henceforth referred to as “adjusted abundances”). Taxon richness, evenness, and Shannon diversity were calculated for each season in each year. In addition, in order to compare taxa richness among samples that varied in abundance, species richness estimates were generated by calculating the Chao1 richness estimator with log-linear 95% confidence intervals (Chao 1987) and rarefaction curves were generated for each season in each year using EstimateS

(Colwell 2005). Because our first two objectives focused on the entire native bee community of the Zumwalt Prairie Preserve, individual traps were not used as replicates but as subsamples. Thus, statistical analyses were not appropriately applied to these descriptions because the sample size was one.

To address our third objective of comparing the relative magnitude of spatial and temporal variability in bee abundance and richness in this grassland, we conducted one-way analyses of variance (ANOVA) with time as the factor and using each trap as 18

a replicate. The purpose of this analysis was to determine if spatial variability (at the scale of the trap) was large enough to swamp out temporal changes observed at the larger community level. If ANOVA tests were significant, we separated means using

Fisher’s LSD multiple comparison test. SYSTAT (1997) Version 7.0. was used all statistical analyses.

Results

A total of 7,124 bees were collected in 2007 and 2,034 bees in 2008. For both years combined, 92 species and 119 morphospecies in 27 genera were identified (Table 2.1).

We identified 60% of all specimens to species, 37% to morphospecies and 1% to genus only; 2% were too damaged to identify. The availability of regional generic taxonomic keys varied, resulting in some genera having more morphospecies than others. For example, 99% of specimens in the genus Bombus were identified to species and no morphospecies were used. In contrast, Lasioglossum, one of the most common genera, had a high proportion of morphospecies, with approximately 36% of all morphospecies belonging to this genus.

Abundance, Richness, Evenness, Diversity, and Sex Ratio of the Bee Community

There were large seasonal differences in the abundance of all bees; adjusted abundance was highest in July of both years, and lowest in August (Table 2.1). There were also differences between years; adjusted abundance for all bees in June and July decreased from 2007 to 2008 (Table 2.1). 19

Species richness showed similar trends as adjusted abundance. The number of species collected for each season ranged from 51 to 183, with the most species collected in July of each year (Table 2.1). There were also differences between years.

For June, more species were collected in 2008 than 2007, but for July, more species were collected in 2007 than 2008. Estimated species richness (Chao 1) displayed the same patterns (Table 2.1). Rarefaction curves showed that estimated species richness in June of both years was similar when comparing similar sample sizes; both years showed an estimated species richness of approximately 57 species with a sample size of 150 individuals (Fig. 2.2a). In addition, both curves appeared to begin to plateau at samples sizes of this size or smaller. In contrast, rarefaction curves for July suggest differences between years in total number of species active during the sampling period.

Estimated richness with samples of approximately 1,200 specimens was 142 species in

2007 and 114 in 2008 (Fig. 2.2b). No inter-annual comparisons could be made for

August, since only one year of data was available; however, the shape of the curve is similar to June 2007 and 2008, with a plateau beginning with samples of less than 150 individuals (Fig. 2.2c).

Community evenness remained fairly constant through the seasons and years, ranging from 0.73 to 0.86 (Table 2.1). Shannon diversity, although highest in July of each year, also did not show large differences through the seasons and years, with values ranging from 3.22 to 3.76 (Table 2.1).

Samples of nearly all genera in all sampling periods were heavily dominated by females (Table 2.2). The only exceptions were Sphecodes in July 2007, when 90% 20

of individuals collected were males, and Osmia in June 2008, when 73% were males.

Sex ratio also varied seasonally. For example, Bombus and Lasioglossum started the season with virtually no males, but the proportion of males increased in July of both years. In August 2008, the proportion of males of Lasioglossum reached its highest proportion (Table 2.2). Sex ratio for some genera also varied between years. For example, sex ratios in July for Bombus and Lasioglossum were more male biased in

2007 than 2008. In contrast, the proportion of male Melissodes in July was similar for both years.

Patterns in Bee Genera between Years

In June, patterns in adjusted abundance (Fig. 2.3a) and relative abundance of genera

(Table 2.3) of both years were similar, with the most common genera being

Lasioglossum and Bombus. However, although the adjusted abundance of

Lasioglossum in June decreased slightly from 2007 to 2008 (Fig. 2.3a), its relative abundance increased (Table 2.3). Also, while Hylaeus was the third most common genus in 2007, Andrena and Osmia were the third and fourth most common genera in

2008 (Table 2.3). In July, the adjusted abundance of the most common genera, except

Melissodes, decreased from 2007 to 2008 (Fig. 2.3b). However, general patterns in relative abundance in July were similar in both years, with the most common genera being Lasioglossum, Bombus, and Melissodes (Table 2.3). However, Melissodes became more dominant and Bombus less dominant in 2008, and Osmia replaced

Halictus as the fourth most dominant genus. No comparison of inter-annual patterns can be made for August, since bees were only sampled in 2008. 21

Seasonal Patterns in Bee Genera within Years

There were several similarities in seasonal patterns in 2007 and 2008. In both years, the adjusted abundance of the most common genera increased from June to July (Fig.

2.3a,b) and general patterns in relative abundance were similar, with the most common genus in all seasons of both years being Lasioglossum (Table 2.3). Bombus was also common in each month, reaching its peak in relative abundance in July of both years

(Table 2.3). In both years, Melissodes was uncommon in June and common in July.

Although only one year’s data are available for August, adjusted abundance of most genera decreased from July to August (Fig. 2.3b,c). Patterns in relative abundance in

August 2008 were similar to July of both years, except that Megachile occurred at a higher relative abundance in August compared to July (Table 2.3).

Total Abundance, Richness, Evenness, Diversity, and Sex Ratio of Bombus

Because of the availability of taxonomic keys for Bombus (Stephen 1957; Thorp et al.

1983) and because this genus comprises a major component of the Zumwalt bee fauna

(Table 2.3), we examined patterns in this genus in more depth. There were large seasonal differences in the abundance of bumble bees; adjusted abundance was highest in July of both years, and lowest in August (Table 2.1). There were also differences between years; adjusted abundance for bumble bees in June and July decreased from

2007 to 2008 (Table 2.1). Bumble bee species richness was the same in June of both years, but decreased in July from 2007 to 2008. Fourteen species of Bombus were identified (Table 2.1), with species richness for each season ranging from 9 to 14.

Bumble bee community evenness and Shannon diversity remained fairly constant 22

through the seasons and years, ranging from 0.68 to 0.78 and 1.57 to 1.71, respectively

(Table 2.1).

Sex ratios of bumble bees were examined for July only since fewer than 30 individuals per species were collected in June or August. In July of both years, almost all Bombus species were dominated by workers or queens except for B. insularis, which was composed of 79% males in 2007 (Table 2.4). For the genus overall, there was also a distinct difference between 2007 and 2008 in the proportion of females that were queens in June, with most females being workers in 2007, but most being queens in 2008 (Table 2.2).

Patterns in Bombus Species between Years

Although there were some similarities in patterns of Bombus species in 2007 and 2008, there were also strong differences. In June of both years, adjusted abundance was similar and comparatively low for all species (Fig. 2.4a). Patterns in relative abundance in June were also similar, with B. flavifrons being the most common species and B. bifarius also common in both years (Table 2.3). However, in 2007, B. nevadensis and B. appositus were also dominant, but in 2008, B. californicus was more common (Table 2.3). In July, the adjusted abundance of the most common species, except B. nevadensis, decreased strongly from 2007 to 2008 (Fig. 2.4b). In addition, the relative abundance of Bombus species in July differed between years

(Table 2.3). Although the same three species, B. bifarius, B. californicus and B. flavifrons, dominated in July of both years, their relative abundance differed (Table 23

2.3). No comparison of inter-annual patterns can be made for August, since bees were only sampled one year for that month.

Seasonal Patterns in Bombus within Years

The adjusted abundance of the most common Bombus species increased from June to

July in both years (Fig. 2.4a,b). However, inter-seasonal patterns in relative abundance differed between years. In 2007, the relative abundance of B. flavifrons decreased strongly from June to July while the relative abundance of B. bifarius increased strongly (Table 2.3). In contrast, in 2008 the relative abundance of the three most common species was relatively similar from June to July (Table 2.3). In 2008, adjusted abundance of all Bombus species except for B. huntii declined from July to

August (Fig. 2.4b,c) although two species common in July (B. californicus and B. bifarius) were also common in August (Table 2.3).

Spatial and Temporal Variability in Abundance and Taxa Richness

ANOVA was used to compare the relative magnitude of spatial and temporal variability in abundance and richness of all bees and bumble bees in this grassland.

Time had a significant effect on the total number of bees collected per trap per hour (F

= 167.4, df = 4, 226, p < 0.0001), the number of all species collected (F = 382.7, df = 4,

226, p < 0.0001), the number of bumble bees collected per trap per hour (F = 123.7, df

= 4, 226, p < 0.0001), and the number of bumble bees species collected (F = 197.3, df

= 4, 226, p < 0.0001). Multiple comparison tests showed that patterns observed at the community level described above were also observed at the trap level (Table 2.1). 24

Specifically, significantly more bees and bumble bees per trap were collected in July than June of each year (Table 2.1). Patterns in abundance between years at the trap level were also similar to those observed at the community level. Although adjusted abundance for all bees and bumble bees in June did not differ between 2007 and 2008, it significantly decreased in July from 2007 to 2008 (Table 2.1). Mean species richness per trap showed all the same significant seasonal and yearly patterns as abundance, except that decreases in richness for all bees and bumble bees in June from 2007 to

2008 were also statistically significant (Table 2.1). These results indicate that spatial variability in abundance and richness are not great enough to swamp out temporal patterns evident at the community level.

Floral Resources and Weather

Both floral resources and weather varied temporally. Blooming forb abundance and richness, which were only examined one year, showed strong seasonal changes. A total of 46 blooming forb species were found on transects adjacent to blue vane traps in 2008, with more stems and species in bloom in June than in July and no forb species blooming on transects in August (Table 2.5). Average daily temperature peaked in July of both years and was higher in 2007 than 2008 (Fig. 2.5a). In fact, mean temperature was higher in 2007 than 2008 in every month from March to July.

Spring 2007 was also drier than 2008 (Fig. 2.5b). 25

Discussion

Pollinator Communities in the Pacific Northwest Bunchgrass Prairie

Grasslands, in general, are of high conservation value for maintaining bee biodiversity; in temperate systems, grasslands are believed to support a richer bee fauna than other habitats, such as forests (Michener 2007). However, few studies have examined pollinator communities in North American grasslands, and those that have been conducted have focused on the prairies of the Great Plains (Table 2.6). Although sampling methods used in the Great Plains varied from study to study and some methods may have under-represented certain groups (e.g., pan traps likely underestimate cavity nesters, Westphal et al. 2008), all indicate that Great Plains grasslands support rich and abundant native bee communities. Likewise, in our study, the first to examine native bee communities in the Pacific Northwest Bunchgrass

Prairie, we found 211 species/morphospecies in 27 genera, with an estimated maximum species richness of 213. This level is comparable to the richness of Great

Plains grassland bee communities (Table 2.6), indicating that the unique and threatened Pacific Northwest Bunchgrass Prairie provides important habitat for native bee biodiversity and therefore is of key conservation interest.

In contrast to other studies of bee communities in North American grasslands, we examined the adjusted and relative abundance of all common taxa collected.

Describing community composition (i.e., the relative abundance of taxa) is important in understanding and conserving native pollinator communities because richness and diversity indices alone provide a limited picture of native pollinator communities, and 26

may mask or under-represent significant patterns of change, as discussed below and by

Williams et al. (2001).

We found a community heavily dominated by three genera: Bombus,

Lasioglossum, and Melissodes. The genus Bombus consists of bumble bees, which are mid- to large-sized bees (0.9-2.2 cm) that are mostly primitively eusocial and generally abundant in cooler, higher altitude habitats (Michener 2007). Bumble bees are typically generalist foragers and commonly nest in the ground, such as in rodent nests, or in cavities under bunchgrass or other vegetation. Bumble bee colonies are usually annual, and are started in early spring by the queen, who mates during the preceding fall. In contrast, Lasioglossum, commonly called sweat bees, is a large genus of primarily small bees (< 0.8 cm) that range from solitary to communal. Most are generalist foragers, with a long flight season, and members of this genus typically nest in burrows excavated in banks or flat soil (Potts & Willmer 1997). As with bumble bees, most females mate in fall and then overwinter until good weather in spring, when they emerge to establish their nests (Michener 2007). Melissodes, or long-horned bees, are small to medium-sized (0.8-1.8 cm) bees, many of which are specialists on plants in the Asteraceae family. Many are solitary, some are communal, and most nest in the ground.

The composition of this community is difficult to compare in much detail with other North American grasslands, since little information has been published on the relative abundance of individuals at any taxonomic level in other grassland systems in

North America. Kwaiser and Hendrix (2008) do list the relative abundance of the six 27

most common species in their study, which composed 52% of the fauna. Of these, three species of Lasioglossum comprised 24% of the fauna, and one species each of

Melissodes, Hylaeus, and Augochlorella comprised 10%, 7%, and 10%, respectively.

Thus, Lasioglossum, and Melissodes and Hylaeus to a lesser extent, are dominant taxa in this Iowan native bee community, a pattern also found in our study. Tepedino and

Stanton (1981) do not directly report the relative abundance of different genera or species, although they do report absolute abundance of Bombus for each year.

However, without knowing the total number of bees collected in their study, it is difficult to compare the dominance of Bombus between their and our study. Although

Kearns and Oliveras (2009) list the three most abundant species (generalists Apis mellifera, Augochlorella striata, and Halictus ligatus), no information on the relative abundance of genera or species is presented. In contrast to their study, we found no A. mellifera. Although this may potentially be a bias of blue vane traps (Stephen & Rao

2005, 2007), as we discuss below, extensive hand-netting in the Zumwalt Prairie has yet to produce a single specimen of this introduced species (DeBano & Kimoto, unpublished data), suggesting it is uncommon in the area.

Temporal Variability in Bee Communities

Bee fauna can vary greatly through time (Williams et al. 2001; Michener 2007). We found that the native bee community in the Zumwalt Prairie showed strong seasonal and inter-annual variation. Seasonally, the highest total abundance, richness, and diversity were found in July of both years. In addition, there were large seasonal differences in the dominance of particular genera and bumble bee species. For 28

example, although Lasioglossum was dominant throughout the range of collecting dates in both years, other genera, such as Bombus, were dominant in early to mid season, while others, such as Melissodes and Megachile, became more dominant in late season (Table 2.3). Even species within Bombus showed large seasonal fluctuations, with some species, such as B. flavifrons peaking in the early to mid season, and other species, such as B. bifarius, becoming more dominant late in the season. Whether these trends also occur in other North American grasslands is unknown, since no information on how relative abundance of community composition changes within the season has been reported (Tepedino & Stanton 1981; Reed 1995;

Hines & Hendrix 2005; Davis et al. 2008; Kwaiser & Hendrix 2008; Kearns &

Oliveras 2009).

Seasonal variation in bee communities is believed to be primarily the result of changes in floral resource availability and weather (Tepedino & Stanton 1981;

Michener 2007; Kearns & Oliveras 2009). However, with regard to floral resource availability, we found that blooming forb abundance and richness were highest in June, not in July when abundance, richness, and diversity of native bees were highest. One possible explanation for the asynchronicity between blooming forb abundance and richness and bee abundance and richness is that bees may not be limited by floral resources in July. Tepedino and Stanton (1981) provided evidence suggesting that bees are often not limited by floral resources in a Great Plains prairie. Weather may also have played a key role; as evidenced by Fig. 2.5, June is often cold and wet. Bees will delay emergence and/or decrease activity in inclement weather, regardless of how 29

many floral resources are available (Michener 2007). Seasonal variation in genera and species composition of native bee communities is also expected, not only because of variation among taxa in their ability to tolerate inclement weather (e.g., bumble bees can fly in colder and windier conditions than other bees (Goulson 2003)), but also because of variation in the degree of specialization and the phenology of plants upon which specialists depend (Michener 2007).

Inter-annual differences in native bee communities on the Zumwalt Prairie were also pronounced. Although overall abundance, richness, and diversity in June between years did not differ strongly, taxon composition and proportion of queens and sex ratio of some taxa did. Differences between years were even more pronounced in

July. Overall abundance, richness, and diversity decreased in 2008, and community composition varied, with Bombus becoming less dominant and Melissodes more so.

As with seasonal variation, weather also appears to play a major role in explaining inter-annual differences in bee fauna in both June and July. June 2008 was colder and wetter, which delayed phenology of flowering of forbs on the prairie by two weeks or more (Dingeldein et al. 2009). Bee phenology also appeared to be delayed in 2008, potentially as a direct response to cooler/wetter weather, or an indirect response to delayed plant phenology, or a combination of both. Evidence indicating that differences in bee communities were driven by delayed phenology as an indirect or direct response to weather include the collection of genera, such as

Andrena and Osmia, that are typically active early in the season (Wojcik et al. 2008) in June 2008, and their absence in June 2007 (when we presumably first sampled after 30

their peak), and the apparent lag in peaks of Bombus species in 2008 compared to

2007. The proportion of Bombus queens collected in 2007 and 2008 also suggests delayed phenology. Queens generally engage in most flight activity early in the season until workers are generated and take over foraging duties (Michener 2007). In June

2007, 91% of Bombus females were workers, while in June 2008, 98% of Bombus females were queens.

Although several studies of native pollinators in North American grasslands have included more than one year’s sampling (Tepedino & Stanton 1981; Reed 1995;

Hines & Hendrix 2005; Davis et al. 2008; Kearns & Oliveras 2009), most combine data across years. Only one study describes inter-annual variation in grassland bee communities. Tepedino and Stanton (1981) found large changes in bee communities in

Wyoming from one year to another that varied spatially. These changes included a steep decline in bumble bee abundance (but not in richness) during the second year, with only 13% of the Bombus collected compared to the previous year. They speculate that these differences were driven by changes in floral resources, which declined in the second year of the study. They hypothesize that bumble bees may have dispersed to live and forage in other areas in response to the decrease in floral abundance at that site, and suggest this response is particularly likely for bumble bees given the high energy needs of these relatively large bees. Decreased floral resources may also explain the decreased abundance of bumble bees on the Zumwalt Prairie in 2008.

Because of the delayed flowering of many plants, queens may have shifted their nesting to agricultural areas within 15 km of the preserve, where flowering crops may 31

have provided a more constant floral resource than native plants on the prairie that year. Although much attention has been directed at understanding the benefit of native areas next to croplands for enhancing pollinator activity for crops, in some cases, nearby croplands may provide an alternative resource in years when native plants are experiencing delayed and reduced flowering (Rao & Stephen 2010). More research on the potential interaction between agricultural and uncultivated habitats in supporting native bee communities is needed.

An alternative explanation to weather for the decrease in abundance of many genera in 2008 compared to 2007 is that sampling efforts of the previous year depressed populations the following year. We think this explanation is unlikely for several reasons. First, other taxa that were not destructively sampled showed similar trends. For example, the most common grassland bird species, Savannah sparrow

(Passerculus sandwichensis), also clearly decreased from 2007 to 2008 (and then increased the following year) (T. Johnson, personal communication). Second, although the total number of pollinators collected in 2007 was high, the rate of collection per trap was fairly low (maximum = 3.44 bees per hour in July), with each trap operating for 2 days or less through the sampling period and a trap density of only 1 trap per 10 ha.

Regardless of the reason for differences among seasons and between years, this study highlights the importance of considering phenology when monitoring bees. Our results support those of others (Williams et al. 2001) that native bee communities are diverse and greatly variable in time and space. In this study, temporal variability was 32

more important than spatial variability in driving patterns in abundance and richness, as evidenced by the fact that spatial variability was not high enough to swamp out temporal patterns when analyzed at the trap level. Thus, care must be taken when designing long-term monitoring protocols to ensure that sampling occurs during the same point of phenology of bee communities, not simply at the same date year after year. Our study demonstrates that even a few weeks between sampling periods can make a very large difference in abundance, species richness, and community composition, and that phenology can vary substantially from one year to another.

Thus, in an ideal situation, initial studies of seasonal variation for monitored sites should be conducted, and attempts to identify cues to initiate sampling identified.

These cues may be the presence of early (or late) season taxa, the presence (or absence) of queens outside the nest in eusocial species, or, by proxy, the blooming of plants whose phenologies are closely tied to bee community phenology. Without this information, long-term datasets will be difficult, if not impossible to interpret, as yearly increases or decreases may simply reflect seasonal variation in bee phenology.

Availability of Pollinators for a Threatened Grassland Plant Species

The threatened plant, Spalding’s catchfly, S. spaldingii, is found on the Zumwalt

Prairie and bumble bees are important pollinators for this species at the Zumwalt

Prairie (DeBano et al., unpublished data) as well as at other locations in the region

(Lesica & Heidel 1996). Our results show that bumble bees are an abundant portion of the native bee fauna on the prairie, forming up to 33% of the community in July, the month of peak flowering for S. spaldingii (Dingeldein et al. 2009). The Zumwalt 33

Prairie may support an abundant bumble bee fauna because of its large size and its relatively intact status. Thus, preserving large, relatively intact grasslands is not only important for preserving native bee biodiversity, but also the native plants that depend on them.

Sampling Native Bee Fauna in Grasslands with Blue Vane Traps

This study also examined the potential usefulness of the blue vane trapping method for sampling native bees in grassland habitats. An important step in conserving native bee fauna is understanding which species are experiencing declines in a given area. Thus, the importance of monitoring bee communities through time has long been recognized and the need for repeatable, standardized sampling methods that allow comparisons of trends across time and space is well appreciated (e.g., São Paulo Declaration on

Pollinators 1999; LeBuhn et al. 2003; Westphal et al. 2008).

Each technique used to sample native bees has advantages and disadvantages, including certain types of biases inherent in all sampling techniques (Westphal et al.

2008; Droege et al. 2010), and the magnitude of these will vary with many factors, including habitat type, personnel involved in the project, and the bee fauna being sampled. Active trapping methods, such as hand-netting, usually have the advantage of allowing the investigator to establish relationships between bee species and host flowers, although this advantage may be less important for monitoring studies that simply seek to describe changes in populations or communities through time. Active sampling has the additional advantage that, if specimens can be identified adequately in the field, bees can be sampled non-lethally. Drawbacks of active trapping methods 34

are that they are usually more time-consuming in the field since sampling occurs only when investigators are present, representative samples of bees require repeated sampling effort at a site throughout a day because diurnal patterns in bee activity vary by species (Michener 2007), and the effectiveness of the method is greatly affected by the experience and capabilities of the sampler (Westphal et al. 2008). In contrast, passive techniques, such as pan traps and blue vane traps, are generally more efficient than active ones because sampling continues in the absence of the investigator, and thus is generally more economical. Disadvantages to some passive techniques, such as pan traps, are that they involve lethal sampling. Passive techniques are also subject to trap biases because of their use of attractants of some type. For example, the color of pan traps has been shown to affect species and sexes collected (e.g., Leong & Thorp

1999; Campbell & Hanula 2007). Other passive techniques, such as trap nests, which attract cavity nesters, are specifically designed to sample subsets of the bee fauna

(Westphal et al. 2008). There is also a suggestion that some passive methods, such as pan traps, tend to be biased towards collecting smaller bees (Cane et al. 2000;

Roulston et al. 2007).

Sampling North American grasslands presents particular problems for native bee sampling. In contrast to many of the small remnant grasslands in Europe, most of which were the result of historic human intervention (Steffan-Dewenter & Tscharntke

2002), many North American grasslands still include very large, open remnants of native grasslands. Although these grasslands may appear superficially homogenous, many show a large degree of heterogeneity with regard to both soils and vegetation. 35

This presents a challenge in efficiently and adequately sampling a large, variable landscape with existing techniques that are logistically better suited for smaller spatial scales. Pan traps used to capture bees, for example, are recommended to be 96 ml plastic bowls placed on the ground in arrays of various sizes and shapes (Droege et al.

2010). Thus because of their small size and location on the ground, they are easily obscured by tall grasses. Other logistical disadvantages of using pan traps in these grassland habitats include their risk of blowing away, being stepped on by wildlife and domestic livestock, sliding downhill if placed on a slope, the need to transport liquid for use in traps often situated in remote locations, and the evaporation of liquids in traps in warm temperatures. Active sampling techniques, such as hand netting, also pose issues in these large, heterogeneous habitats. Given the labor intensive nature of active sampling, only a small area can be sampled, and this problem is compounded by the need to sample at different times of the day.

In this study, we show that the blue vane trapping technique overcame many of these logistical problems. First, in contrast to pan traps, blue vane traps do not wet or necessarily kill specimens. Non-lethal trapping allows for the potential of catch and release and may reduce the impact of trapping on pollinators, especially for long term studies (Stephen & Rao 2005). However, more research needs to be conducted on the viability of capture and release of pollinators collected with blue vane traps, including quantifying the amount of time required for handling and determining the best methods to prevent the recapture or double-counting of released individuals.

Logistically, blue vane traps suspended with aluminum poles are easy to work with in 36

field conditions, being light weight and not requiring liquids to be carried, or fluid levels to be monitored for evaporation. Fewer traps were needed to sample as compared to pan traps, specimen condition was excellent, and more than 80% of all specimens collected were bees, reducing the amount of time needed to sort through samples with non-focal individuals. In addition, field time is greatly reduced compared to active methods such as transect walks and observational blocks. These characteristics make the technique very applicable for use by volunteer organizations world-wide.

With regard to trap biases against certain taxa, previous work suggests that blue vane traps are more effective, both in terms of number of individuals and the number of species collected than net sampling or vacuuming. Stephen and Rao (2007) found that blue vane trapping collected 94% of the 33 species found in their study, sweeping collected 63%, and vacuuming and yellow-vane traps collected 54%; only A. mellifera was clearly underrepresented by blue vane trapping. Nevertheless, like all sampling methods, blue vane traps potentially over- or under-represent individuals of particular species or sex. Thus, we cannot say that blue vane traps give an unbiased assessment of the native bee community. However, as pointed out by Droege et al.

(2010), this is a problem common to all native bee sampling techniques, and no studies that we are aware of have yet demonstrated an independent and unbiased method of determining bee communities with which to compare different sampling methods. In addition, as long as biases of a particular method are consistent through 37

time, the method will be useful for monitoring the relative changes in bee communities it samples through time.

Conclusions and Broader Implications for Conservation

This study showed that the Pacific Northwest Bunchgrass Prairie, like other grasslands in North America, supports a rich and diverse native bee fauna, including species, such as the western bumble bee, B. occidentalis, that have virtually disappeared from other parts of its range (Rao et al. in press). Like grasslands world-wide, the remnants of Pacific Northwest Bunchgrass Prairie continue to be threatened by various types of human activities including conversion to cropland, improper livestock management, and invasions by non-native plants (Samson & Knopf 1994; Brennan & Kuvlesky

2005). Both conserving remaining intact habitats and restoring disturbed portions of this type of prairie will be important for conserving native bees and the plants, such as

Spalding’s catchfly, that depend on them.

This study also illustrates the importance of timing monitoring efforts aimed at determining trends in pollinator communities. Sampling for long-term monitoring needs to occur at the same point in bee community phenology. This can be done by tying sampling efforts to the phenology of common early, middle, or late season pollinator species, to queen/worker ratios in social species, or to the phenology of floral resources or some closely related surrogate. In addition, this study illustrates the importance of reporting species composition of communities (i.e., some measure of abundance of each taxa), and how they vary through seasons and years, to better understand trends in bee communities through time and space. Focusing only on 38

richness and total abundance of all bees or select groups can mask important changes in bee communities. Finally, this study illustrates the potential usefulness of a promising sampling technique for native pollinators that has never been used in North

American grasslands: the blue vane trap. This method appears to be well-suited for sampling native pollinators in these grasslands because it is economical, does not require lethal sampling, and can be easily implemented in large expanses. 39

Literature Cited

Allen-Wardell, G., et al. 1998. The potential consequences of pollinator declines on the conservation of biodiversity and stability of food crop yields. Conservation Biology 12:8-17.

Batra, S. W. T. 1995. Bees and pollination in our changing environment. Apidologie 26:361-370.

Biesmeijer, J. C., et al. 2006. Parallel declines in pollinators and insect-pollinated plants in Britain and the Netherlands. Science 313:351-354.

Brennan, L. A., and W. P. Kuvlesky, Jr. 2005. North American grassland birds: an unfolding conservation crisis? Journal of Wildlife Management 69:1-13.

Buchmann, S. L., and G. P. Nabhan. 1996. The forgotten pollinators. Island Press, Washington, D.C.

Campbell, J. W., and J. L. Hanula. 2007. Efficiency of Malaise traps and colored pan traps for collecting flower visiting insects from three forested ecosystems. Journal of Insect Conservation 11:399-408.

Cane, J. H., R. L. Minckley, and L. J. Kervin. 2000. Sampling bees (Hymenoptera: Apiformes) for pollinator community studies: pitfalls of pan-trapping. Journal of the Kansas Entomological Society 73:225-231.

Chao, A. 1987. Estimating the population size for capture-recapture data with unequal catchability. Biometrics 43:783-791.

Colwell, R. K. 2005. EstimateS: Statistical estimation of species richness and shared species from samples. Version 8.0. User’s guide and application. (http://purl.oclc.org/estimates).

Davis, J. D., S. D. Hendrix, D. M. Debinski, and C. J. Hemsley. 2008. Butterfly, bee and forb community composition and cross-taxon incongruence in tallgrass prairie fragments. Journal of Insect Conservation 12:69-79.

Dingeldein, J., H. Schmalz, R. V. Taylor, V. Jansen, D. Dostert, and L. Shirley. 2009. Plant phenology monitoring on the Zumwalt Prairie Preserve 2006-2009. The Nature Conservancy. Available from http://conserveonline.org/workspaces/ZumwaltPrairieWorkspace/documents/pl ant-phenology-monitoring-on-the-zumwalt-prairie/view.html (accessed January 2011). 40

Droege, S., V. J. Tepedino, G. LeBuhn, W. Link, R. L. Minckley, Q. Chen, and C. Conrad. 2010. Spatial patterns of bee captures in North American bowl trapping surveys. Insect Conservation and Diversity 3:15-23.

Goulson, D. 2003. Bumblebees behavior and ecology. Oxford University Press, United Kingdom.

Hines, H. M., and S. D. Hendrix. 2005. Bumble bee (Hymenoptera: Apidae) diversity and abundance in tallgrass prairie patches: effects of local and landscape floral resources. Environmental Entomology 34:1477-1484.

Kearns, C. A., D. W. Inouye, and N. M. Waser. 1998. Endangered mutualisms: the conservation of plant-pollinator interactions. Annual Review of Ecology and Systematics 29:83-112.

Kearns, C. A., and D. M. Oliveras. 2009. Environmental factors affecting bee diversity in urban and remote grassland plots in Boulder, Colorado. Journal of Insect Conservation 13:655-665.

Kennedy, P. L., S. J. DeBano, A. Bartuszevige, and A. S. Lueders. 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.

Klein, A. M., B. E. Vaissière, J. H. Cane, I. Steffan-Dewenter, S. A. Cunningham, C. Kremen, and T. Tscharntke. 2007. Importance of pollinators in changing landscapes for world crops. Proceedings of the Royal Society of London Series B-Biological Sciences 274:303-313.

Kwaiser, K. S., and S. D. Hendrix. 2008. Diversity and abundance of bees (Hymenoptera: Apiformes) in native and ruderal grasslands of agriculturally dominated landscapes. Agriculture, Ecosystems and Environment 124:200- 204.

LeBuhn, G., et al. 2003. A standardized method for monitoring bee populations – the bee inventory (BI) plot. Available from http://online.sfsu.edu/~beeplot/pdfs/Bee Plot 2003.pdf (accessed January 2011).

Leong, J. M., and R. W. Thorp. 1999. Colour-coded sampling: the pan trap colour preferences of oligolectic and nonoligolectic bees associated with a vernal pool plant. Ecological Entomology 24:329-335.

Lesica, P. 1993. Loss of fitness resulting from pollinator exclusion in Silene spaldingii (Caryophyllaceae). Madroňo 40:193-201. 41

Lesica, P. and B. Heidel. 1996. Pollination biology of Silene spaldingii. Unpublished report to Montana Field Office of The Nature Conservancy. Montana Natural Heritage Program, Helena, Montana.

Losey, J. E., and M. Vaughan. 2006. The economic value of ecological services provided by insects. BioScience 56:311–323.

Michener, C. D. 2007. The bees of the world. 2nd edition. Johns Hopkins University Press, Baltimore, Maryland.

Moldenke, A. R. and J. L. Neff. 1974. The bees of California: A catalogue with special relevance to pollination and ecological research. Board of Studies in Biology, University of California, Santa Cruz, California.

(NRC) National Research Council. 2007. Status of pollinators in North America. The National Academies Press, Washington, D.C.

Potts, S. G., and P. Willmer. 1997. Abiotic and biotic factors influencing nest-site selection by Halictus rubicundus, a ground-nesting halictine bee. Ecological Entomology 22:319-328.

Rao S, and W. P. Stephen. 2007. Bombus (Bombus) occidentalis (Hymenoptera: Apiformes): In decline or recovery? The Pan-Pacific Entomologist 83:360-362. Rao, S., and W. P. Stephen. 2009. Bumble bee pollinators in red clover seed production. Crop Science 49:2207-2214.

Rao. S., and W. P. Stephen. 2010. Abundance and diversity of native bumble bees associated with agricultural crops: The Willamette Valley experience. Psyche 2010:1-9.

Rao. S., W. P. Stephen, C. Kimoto, and S. J. DeBano. In Press. Bombus (Bombus) occidentalis (Hymenoptera: Apiformes) in the Zumwalt Prairie. Northwest Science.

Reed, C. C. 1995. Insects surveyed on flowers in native and reconstructed prairies (Minnesota). Restoration and Management Notes 13:210-213.

Roulston, T. H., S. A. Smith, and A. L. Brewster. 2007. A comparison of pan trap and intensive net sampling techniques for documenting a bee (Hymenoptera: Apiformes) fauna. Journal of the Kansas Entomological Society 80:179-181.

42

São Paulo Declaration on Pollinators. 1999. Report on the recommendations of the workshop on the conservation and the sustainable use of pollinators in agriculture with emphasis on bees. Brazilian Ministry of the Environment, Brasilia, Brazil.

Smith, R. L., and T. M. Smith. 2000. Elements of ecology. 4th edition. Addison Wesley Longman, Benjamin/Cummings, Menlo, California.

Steffan-Dewenter, I., and T. Tscharntke. 2002. Insect communities and biotic interactions on fragmented calcareous grasslands – a mini review. Biological Conservation 104:275-284.

Stephen, W. P., G. E. Bohart, and P. F. Torchio. 1969. The biology and external morphology of bees. Agricultural Experiment Station, Oregon State University, Corvallis, Oregon.

Stephen, W. P., and S. Rao. 2005. Unscented color traps for non-Apis bees (Hymenoptera: Apiformes). Journal of the Kansas Entomological Society 78:373-380.

Stephen, W. P., and S. Rao. 2007. Sampling native bees in proximity to a highly competitive food resource (Hymenoptera: Apiformes). Journal of the Kansas Entomological Society 80:369-376. Stephen, W. P., S. Rao, and L. White. 2009. Abundance, diversity and foraging contribution of bumble bees to blueberry production in Western Oregon. Acta Horticulture 810:557-562.

SYSTAT. 1997. SYSTAT 7.0 for Windows. SPSS Inc., Chicago, Illinois.

Taylor, R.V., V. Jansen, H. Schmalz, and J. Dingeldein. 2009. Mapping and monitoring Spalding’s catchfly (Silene spaldingii) on the Zumwalt Prairie Preserve 2006-2008. The Nature Conservancy. Available from http://conserveonline.org/workspaces/ZumwaltPrairieWorkspace/documents/ mapping-and-monitoring-spaldings-catchfly-silene/view.html (accessed January 2011).

Tepedino, V. J., and N. L. Stanton. 1981. Diversity and competition in bee-plant communities on short-grass prairie. Oikos 36:35-44.

Thomson, D. 2004. Competitive interactions between the invasive European honey bee and native bumble bees. Ecology 85:458-470.

43

Thorp, R. W., D. S. Horning, Jr. and L. L. Dunning. 1983. Bumble bees and cuckoo bumble bees of California (Hymenoptera: Apidae). Bulletin of the California Insect Survey 23:1-79.

Thorp, R. W. 2003. Bumble bees (Hymenoptera:Apidae): commercial use and environmental concerns. Pages 21-40 in K. Strickler and J. H. Cane, editors. For nonnative crops, whence pollinators of the future? Thomas Say Publications in Entomology: Proceedings. Entomological Society of America, Lanham, Maryland.

Tisdale, E. W. 1982. Grasslands of western North America: the Pacific Northwest Bunchgrass. Pages 223-245 in A. C. Nicholson, A. McLean and T. E. Baker, editors. Grassland ecology and classification symposium proceedings. Information Services Branch B.C. Ministry of Forests, Victoria, British Columbia.

Westphal, C., et al. 2008. Measuring bee diversity in different European habitats and biogeographical regions. Ecological Monographs 78:653-671.

Williams, N. M., R. L. Minckley, and F. A. Silveira. 2001. Variation in native bee faunas and its implications for detecting community changes. Conservation Ecology 5:7 (http://www.consecol.org/vol5/iss1/art7/).

Williams, P. 2010. Bombus (Th.) fervidus (Fabricius). Available from http://www.nhm.ac.uk/research- curation/research/projects/bombus/th.html#fervidus (accessed December 2010).

Wojcik, V. A., G. W. Frankie, R. W. Thorp, and J. L. Hernandez. 2008. Seasonality in bees and their floral resource plants at a constructed urban bee habitat in Berkeley, California. Journal of the Kansas Entomological Society 81:15-28. 44

a) b)

* ** * * * **** * * * ** Zumwalt Prairie * Area Enlarged

* * 0 500 1000 m * * * * OREGON * * * * N * * * * * * * *** ** * * * * **** * *

Figure 2.1. a) Location of the Zumwalt Prairie in northeastern Oregon and location of pollinator traps in each pastures (all sites were sampled in each season of each year except that unshaded pastures and traps denoted with “*” were not sampled in June 2007) b) a blue vane trap. 45

a) 80 70 60 50 40 2007 30 2008 20 Species Richness 10

0 0 100 200 300 400 Number of Individuals

b) 250

200

150

100 2007 2008 Species Richness 50

0 0 1000 2000 3000 4000 5000 6000 7000

Number of Individuals

c) 70

60

50

40

30 2008 20

Species Richness 10

0 0 50 100 150 200 250 300 Number of Individuals

Figure 2.2. Estimated richness using rarefaction (Chao 1) in 2007 and 2008 for a) June, b) July, and c) August (2008 only). Note scale of both axes differ in each panel. 46

1.40 a) 2007 1.20 2008

1.00 0.80 0.60

0.40

0.20

Abundance per trapper hour 0.00

Osmia Perdita Hylaeus Halictus Hoplitus Bombus Andrena Diadasia Megachile Anthophora

Emphoropsis Agapostemon Lasioglossum 1.40 b) 2007

1.20 2008

1.00 0.80

0.60

0.40 0.20 Abundance per trapper hour 0.00

Osmia

Perdita Hylaeus Colletes Hoplitus Bombus Andrena Diadasia Halictus Halictus Coelioxys Anthidium Megachile

Sphecodes Melissodes Anthophora Emphoropsis Ashmeadiella Agapostemon Lasioglossum 1.40 c) 2008 1.20

1.00

0.80

0.60 0.40

0.20

Abundance per trapper hour 0.00

Halictus Ceratina Bombus Megachile Melissodes Anthophora Agapostemon

Lasioglossum Figure 2.3. Mean abundance of bees adjusted by trapping effort for common genera in 2007 and 2008 in a) June, b) July, and c) August (2008 only). Error bars are 95% confidence intervals. 47

2007 a) 0.60 2008 0.50 0.40

0.30

0.20 0.10 Abundance per trapper hour 0.00

B. mixtus B. bifarius

B. fervidus B. insularis B. B. flavifrons B. appositus B. nevadensis rufocinctus B. B. californicus occidentalis B.

2007 b) 0.60 2008 0.50

0.40 0.30 0.20

0.10

Abundance per trapper hour 0.00

B. huntii B. mixtus B. vegans B. bifarius B. fervidus B. insularis B. fernaldae B. flavifrons B. appositus B. rufocinctus B. nevadensis B. griseocollis B. californicus B. occidentalis

c) 0.60 2008

0.50 0.40

0.30 0.20

0.10

Abundance per trapper hour 0.00

B. huntii B. bifarius B. fervidus B. rufocinctus B. californicus Figure 2.4. a) Mean abundance of bees adjusted by trapping effort of common Bombus species in 2007 and 2008 in a) June, b) July, and c) August (2008 only). Error bars are 95% confidence intervals. 48

a) 25.0

20.0

15.0

2007 10.0 2008

5.0

Air temperature (C°) temperature Air 0.0 March April May June July August September -5.0

b) 8.0

7.0 6.0

5.0

4.0 2007 2008 3.0

2.0

1.0 Total monthlyTotal rainfall (cm) 0.0 March April May June July August September

Figure 2.5. Values of a) mean average daily temperature (°C) and b) total rainfall (cm) for spring and summer months in 2007 and 2008. 49

Table 2.1. List of bee taxa found in the Zumwalt Prairie and their presence/absence for each season. Number following genus title is the number of morphospecies identified in that genus for that season.

2007 2008 June July June July August Family: Colletidae Subfamily: Colletinae Genus: Colletes (3) (1) C. simulans √ Subfamily: Hylaeinae Genus: Hylaeus (3) (4) (3) Family: Andrenidae Subfamily: Andreninae Genus: Andrena (4) (2) (10) (5) Subfamily: Panurginae Genus: Perdita P. oregonensis √ √ √ Family: Halictidae Subfamily: Halictinae Genus: Agapostemon A. texanus √ √ √ A. virescens √ √ √ √ Genus: Halitcus (1) H. confusus √ H. farinosus √ √ √ H. ligatus √ √ H. rubicundus √ √ √ √ H. tripartitus √ √ √ √ Genus: Lasioglossum (17) (38) (16) (29) (22) L. anhypops √ √ √ √ L. egregium √ √ √ √ L. olympiae √ √ L. pacificum √ √ L. rubicundus √ 50

Table 2.1. (Continued)

L. sisymbrii √ √ √ √ √ L. titusi √ √ √ Genus: Sphecodes (15) Subfamily: Rophitinae Genus Dufourea D. rufiventris √ Family: Megachilidae Subfamily: Megachilinae Genus: Anthidiellum A. notatum √ Genus: Anthidium (9) Genus: Ashmeadiella (3) A. sculleni √ Genus: Atoposmia (1) √ Genus: Coelioxys (4) Genus: Dianthidium D. singulare √ Genus: Stelis (2) Genus: Hoplitis (2) H. albifrons √ √ H. fulgida √ √ √ √ Genus: Megachile (2) M. bradleyi √ M. dentitarsus √ √ M. latimanus √ √ √ √ M. melanophaea √ M. mellitarsus √ M. nevadensis √ √ M. perihirta √ M. pugnata √ M. relativa √

51

Table 2.1. (Continued) M. rivalis √ M. wheeleri √ √ Genus: Osmia (4) (9) (4) (1) O. albolateralis √ O. atrocyanea √ √ √ O. bella √ O. brevis √ √ √ O. bruneri √ √ O. bucephala √ O. californica √ √ O. calla √ O. caulicola √ O. cobaltina √ O. coloradensis √ √ O. cyanella √ O. cyaneonitens √ √ O. ednae √ √ O. juxta √ O. kincaidii √ O. longula √ √ O. montana √ O. nanula √ O. nigrifrons or O. √ √ raritatis O. nifoata

O. pellax √ √ O. pentstemonis √ O. regulina √ O. sculleni √ O. subaustralis √ √ O. texana √ O. trevoris √ √ √

52

Table 2.1. (Continued) O. tristella √ O. vandykei √ √ Family: Apidae Subfamily: Apinae Genus: Anthophora (1) (1) (2) (1) A. bomboides √ √ A. pacifica √ √ A. terminalis √ A. urbana √ √ √ A. ursina √ Genus: Bombus B. appositus √ √ √ B. bifarius √ √ √ √ √ B. californicus √ √ √ √ √ B. fernaldae √ B. fervidus √ √ √ √ B. flavifrons √ √ √ √ √ B. griseocollis √ √ √ B. huntii √ √ B. insularis √ √ √ √ B. mixtus √ √ √ √ B. nevadensis √ √ √ √ √ B. occidentalis √ √ √ √ √ B. rufocinctus √ √ √ √ √ B. vagans √ Genus: Diadasia D. enavata √ √ D. nigrifrons √ √ √ Genus: Habropoda (1) (1) (1) (1)

53

Table 2.1. (Continued) Genus: Eucera (2) Genus: Melissodes (2) (2) (3) M. agilis √ √ √ √ M. bicolorata √ M. bimatris √ M. confusa √ √ √ M. metenua √ M. rivalis √ √ √ √ √ M. robustior √ √ √ Genus: Xeromelecta X. californica √ Subfamily: Xylocopinae Genus: Ceratina (1) (1) For all bees Total abundance 276 6,848 541 1,211 282

Adjusted abundance 0.83 3.44 0.67 1.28 0.33 (# bees/trap/hour) ± ± ± ± ± 0.10a,c,d 0.13b 0.10a 0.09c 0.04d

Species richness 54 183 58 92 51

Average bee species richness 11.75 36.26 5.94 11.49 3.16 ± ± ± ± ± 1.54 a 0.95 b 0.64 c 0.63 a 0.32 d

Chao 1 species richness est. 58 213 67 114 66 [95% CI] [50-82] [194- [57- [99- [55- 256] 101] 162] 101]

Evenness 0.86 0.73 0.80 0.81 0.84

Shannon diversity 3.38 3.76 3.22 3.64 3.27

54

Table 2.1. (Continued) For Bombus Total abundance 56 2,229 83 266 38

Adjusted abundance 0.16 1.11 0.10 0.28 0.04 (# bees/trap/hour) ± ± ± ± ± 0.04a,c 0.07b 0.02a 0.03c 0.01a

Total species richness 9 14 9 10 10

Average species richness 2.5 6.87 1.14 2.33 0.52 ± ± ± ± ± 0.33a 0.21b 0.21c 0.20a 0.09d

Evenness 0.78 0.68 0.71 0.72 0.74

Shannon diversity 1.71 1.69 1.57 1.65 1.71

55

Table 2.2. Sex ratios, as expressed by proportion of males, of common genera (>20 specimens) in each season. Sex ratios with more than 10% males are in bold type. “NA” indicates “non-applicable” because no sampling occurred in August 2008.

Genus Year June n July n August n Sex Ratio Sex Ratio Sex Ratio Bombus 07 0.00 56 0.13 2,229 NA NA (8.9% ♀♀=queens) (5.6% ♀♀=queens) 08 0.00 83 0.06 266 0.03 38 (98.0% ♀♀=queens) (21.1% ♀♀=queens) (21.1% ♀♀=queens) Lasioglossum 07 0.01 136 0.14 2,319 NA NA 08 0.00 312 0.04 424 0.42 121 Megachile 07 - - 0.19 77 NA NA 08 - - - - 0.39 28 Melissodes 07 - - 0.23 662 NA NA 08 - - 0.29 328 0.25 61 Osmia 07 - - 0.00 325 NA NA 08 0.73 37 0.00 80 - - Agapostemon 07 - - 0.02 99 NA NA Anthidium 07 - - 0.25 28 NA NA Anthophora 07 - - 0.11 79 NA NA Ashmeadiella 07 - - 0.11 27 NA NA Diadasia 07 - - 0.04 23 NA NA Halictus 07 - - 0.25 613 NA NA Perdita 07 - - 0.00 59 NA NA Sphecodes 07 - - 0.90 58 NA NA Andrena 08 0.39 41 - - - - 56

Table 2.3. Relative abundance of most common genera and Bombus species (>5%) in each season of each year.

Genera

June July August

Lasioglossum – 49% Lasioglossum – 34% Bombus – 20% Bombus – 33% N/A 2007 Hylaeus – 6% Melissodes – 10% Halictus – 9%

Lasioglossum – 58% Lasioglossum – 35% Lasioglossum – 43% Bombus – 15% Melissodes – 27% Melissodes – 22% 2008 Andrena – 8% Bombus – 22% Bombus – 13% Osmia – 7% Osmia – 7% Megachile – 10%

Bombus Species June July August

B. flavifrons – 45% B. bifarius – 47%

B. bifarius – 16% B. californicus – 17% 2007 N/A B. nevadensis – 11% B. flavifrons – 13%

B. appositus – 9% B. rufocinctus – 9%

B. flavifrons – 36% B. flavifrons – 38% B. californicus – 39% B. californicus – 34% B. californicus – 30% B. bifarius – 29% 2008 B. bifarius – 13% B. bifarius – 11% B. rufocinctus – 8% B. nevadensis – 8%

57

Table 2.4. Percentage of males, queens, and workers in Bombus species in July 2007 and 2008. June and August are not shown because of the relatively low numbers collected (n<30 for all species).

Bombus Species Year July n % ♂♂ % Queen % Workers B. appositus 07 61 15 13 72 08 11 0 9 91

B. bifarius 07 1,048 4 1 95 08 30 0 10 90

B. californicus 07 375 21 9 71 08 80 5 39 56

B. fernaldae1 07 2 0 100 NA 08 ------

B. fervidus 07 24 33 25 42 08 6 0 67 33

B. flavifrons 07 292 25 4 71 08 100 11 13 76

B. griseocollis 07 22 5 5 90 08 1 0 100 0

B. huntii 07 44 0 0 100 08 ------

B. insularis 1 07 52 79 21 NA 08 ------

B. mixtus 07 23 4 61 35 08 4 0 25 75

B. nevadensis 07 51 35 12 53 08 22 0 9 91

B. occidentalis 07 36 19 8 72 08 4 0 0 100

58

Table 2.4. (Continued) B. rufocinctus 07 195 8 8 84 08 8 0 0 100 B. vagans 07 1 100 0 0 08 ------1These species are cleptoparasites and therefore have no castes. 59

Table 2.5. List of plant species blooming within 50 m of blue vane traps during each sampling period in 2008. Means are reported ± one standared error. Family name Scientific name June July August Apiaceae Lomatium cous (cous biscuitroot) √ Lomatium triternatum (nineleaf biscuitroot) √ Asteraceae Achillea millefolium var. occidentalis (western √ yarrow) Agoseris glauca (pale agoseris) √ √ Antennaria anaphaloides (pearly pussytoes) √ Antennaria luzuloides (rush pussytoes) √ Arnica sororia (twin arnica) √ √ Balsamorhiza incana (hoary balsamroot) √ Erigeron chrysopsidis (dwarf yellow fleabane) √ √ Erigeron pumilus (shaggy fleabane) √ Senecio integerrimus (lambstongue ragwort) √ Solidago missouriensis (Missouri goldenrod) √ Taraxacum officinale (common dandelion) √ √ Brassicaceae Alyssum alyssoides (pale madwort) √ Draba verna (spring draba) √ Fabaceae Astragalus sheldonii (Sheldon's milkvetch) √ Lupinus caudatus (tailcup lupine) √ Lupinus leucophyllus (velvet lupine) √ Lupinus sericeus (silky lupine) √ √ Trifolium macrocephalum (largehead clover) √ Gentianaceae Frasera albicaulis (whitestem frasera) √ √ Frasera speciosa (elkweed) √ Geraniaceae Geranium viscosissimum (sticky purple geranium) √

60

Table 2.5. (Continued) Liliaceae Allium fibrillum (Cuddy Mountain onion) √ √ Allium tolmiei (Tolmie's onion) √ Calochortus eurycarpus (white mariposa lily) √ Camassia quamash (small camas) √ √ Onagraceae Clarkia pulchella (pinkfairies) √ Polemoniaceae Collomia linearis (tiny trumpet) √ Phlox hoodii (spiny phlox) √ Phlox pulvinata (cushion phlox) √ Polygonaceae Eriogonum douglasii (Douglas’ buckwheat) √ √ Polygonum douglasii (Douglas’ knotweed) √ Erigonum nudum (naked buckwheat) √ Portulacaceae Montia linearis (narrowleaf minerslettuce) √ √ Ranunculaceae Clematis hirsutissima (hairly clematis) √ Delphinium ×burkei (larkspur) √ Rosaceae Geum triflorum (old man's whiskers) √ √ Potentilla gracilis (slender cinquefoil) √ √ Rubiaceae Galium boreale (northern bedstraw) √ Saxifragaceae Lithophragma glabrum (bulbous woodland-star) √ Saxifraga integrifolia (wholeleaf saxifrage) √ Scrophulariaceae Collinsia parviflora (maiden blue eyed Mary) √ Castilleja sp. (Indian paintbrush) √ √ Orthocarpus tenuifolius (thinleaved owl’s-clover) √ Violaceae Viola adunca (hookedspur violet) √ Total species richness 36 23 0 Average species richness per transect 6.7 ± 0.4 2.8 ± 0.2 0 Average stem abundance per transect 61.7 ± 5.7 19.4± 3.0 0 61

Table 2.6. Summary of studies of bee communities in North American grasslands. All studies were conducted in the Great Plains region of the USA. “NR” means “not reported”.

Locations Grassland Type Duration # of Species Estimated Trapping Citation of Study Genera /Morphospecies Richness Method(s) Richness (estimator) Iowa tallgrass prairie 2 years NR 86 114 pan traps Davis et al. (ICE)1 2008

Colorado shortgrass 5 years 29 104 99-1012 pan traps Kearns and prairie (ACE3 and hand-netting Oliveras 2009 Chao2)

Iowa tallgrass prairie 1 year NR 56 81 pan traps Kwaiser and (ICE)1 sweep- Hendrix 2008 netting

Minnesota tallgrass prairie 3 years 33 127 N/A hand-netting Reed 1995

Wyoming high-altitude, 2 years 43 >200 N/A hand-netting Tepedino and shortgrass Stanton 1981 prairie 1 ICE = incidence-based coverage estimate 2Estimated species richness was only calculated for hand-netted samples, which had a species/morpho species richness of 88. 3ACE = abundance-base coverage estimator 62

EFFECT OF LIVESTOCK GRAZING ON NATIVE BEE COMMUNITIES OF

A PACIFIC NORTHWEST BUNCHGRASS PRAIRIE

Abstract

Livestock grazing is a widespread land use in western North America. This action may impact native bees by affecting floral and nesting resources. Native bees are considered to be one of the most important pollinators. However, few studies have investigated how livestock grazing impacts native bees in western North America. Our study is the first manipulative study to examine the effect of a gradient of livestock grazing on native bees. We conducted the study in 16 40-ha pastures on the Zumwalt

Prairie, one of the largest remnants of the once expansive Pacific Northwest

Bunchgrass Prairie. Each pasture was assigned one of four cattle stocking rates (high, medium, low, and no cattle), and grazing intensity was quantified by measuring utilization. Treatments were applied for two years. We measured soil and vegetation characteristics that related to floral and nesting resources as well as several metrics of the bee community, including diversity, richness, abundance, and community composition. Even after exposure to just two years of grazing, some effects on vegetation and soils were detected. For example, increased grazing intensity significantly reduced vegetation structure, the abundance of certain blooming plants, surface soil stability, and the percent of ground covered by herbaceous litter. In addition, increased grazing intensity significantly increased soil compaction and the amount of bare ground. Native bee communities showed corresponding responses to 63

grazing, with changes in abundance, richness, diversity, and community composition.

Different group of bees responded to grazing intensity differently and there was a seasonal difference in response of bee communities to grazing intensity. For example, bumble bees were sensitive to grazing intensity early in the season, showing reduced abundance, diversity, and/or richness with increased grazing intensity. In contrast, halictid bees appeared unaffected by grazing throughout the season. However, even within a genus or family, different species responded to grazing intensity in different manners, potentially because of variation in life histories. As a result, in order to conserve this important and diverse pollinator group in the Pacific Northwest

Bunchgrass Prairie, it is important to maintain land with a mixture of livestock grazing intensities.

Keywords: bumble bees, floral resources, grazing intensity, halictid bees, livestock grazing, native bees, nesting resources, pollinators, soil characteristics, vegetation characteristics

Introduction

Humans have grazed livestock for millennia; today, it remains a common land use throughout the world’s grasslands (Harris 2000). In many regions, such as Britain and

Europe, grazing is often seen as a positive management tool that promotes diversity and maintains habitats (Gibson et al. 1987; Pykälä 2000). However, in other areas, such as western North America, where livestock grazing is one of the most widespread land uses (Crumpacker 1984; Fleischner 1994), it is often viewed as a threat to 64

biodiversity and ecological function (Fleischner 1994; Noss 1994; Belsky &

Blumenthal 1997; Donahue 1999). Regardless of the region, most studies that have examined the effect of livestock grazing on biodiversity have focused on plants and vertebrates, rather than invertebrates (see Fleischner 1994 and Jones 2000 for reviews).

Even fewer studies have examined native bees, which not only form a major component of biodiversity but also play pivotal roles in providing a valuable ecosystem service, pollination.

It is estimated that over 75% of the 250,000 species of flowering plants, including crops that make up 35% of the world’s food supply, are pollinated primarily by bees (Klein et al. 2007; NRC 2007). The fitness of many cross-pollinated, non- cultivated plants depends on bee pollination and even plants capable of self-pollination may benefit from pollinators through higher seed set and a reduction in inbreeding depression (Michener 2007). Thus, bee pollinators are essential for maintaining the genetic diversity of many plant species. However, there is evidence suggesting that we are currently in the midst of a global pollinator crisis, in which many invertebrate pollinators, including native bees, are experiencing large declines (Buchmann &

Nabhan 1996; Kearns et al. 1998; São Paulo Declaration on Pollinators 1999; NRC

2007). Factors implicated in the decline of native bees include exposure to parasites and pathogens, the overuse of insecticides, the introduction of non-native species, such as A. mellifera, and habitat destruction and degradation associated with human disturbance (Kearns et al. 1998; Thomson 2004; NRC 2007; Michener 2007). Of these human disturbances, the impact of livestock grazing on pollinators has received 65

relatively little attention even though grasslands have been shown to support a diverse and abundant bee fauna (Tepedino & Stanton 1981; Reed 1995; Davis et al. 2008;

Kwaiser & Hendrix 2008; Kearns & Oliveras 2009; Chapter 2).

Native bees should be sensitive to livestock grazing because of its impact on plant growth, architecture, diversity, and quality, as well as soil characteristics and microhabitat temperature and relative humidity (Fleischner 1994; Loftin et al. 2000;

Kruess & Tscharntke 2002; DeBano 2006b). Some of these effects such as simplified plant architecture, reduced plant biomass, and increased regrowth by some plants after grazing, are apparent even after short-term exposure to grazing (Morris 1967; Morris

1981; Purvis & Curry 1981). Other effects, including changes in plant community composition and soil compaction are more likely to occur after long-term exposure

(Day & Detling 1990; Huntly 1991). General ecological studies not aimed at examining grazing effects have found that native bees respond to changes in environmental variables including soil structure, bare ground, surface stability, soil compaction, and vegetation, and that these responses may vary seasonally

(Brockmann 1979; Cane 1991; Potts & Willmer 1997; Wuellner 1999; Kells &

Goulson 2003; Potts et al. 2005; Michener 2007).

Of studies that have examined livestock grazing effects on native bees, most have found that livestock grazing is associated with decreased abundance, richness, and/or diversity of bees (Morris 1967; Kruess & Tscharntke 2002; Steffan-Dewenter and Leschke 2003; Hatfield and LeBuhn 2007; Sjödin 2007; Xie et al. 2008; Kearns and Oliveras 2009), while other studies have found a positive relationship of grazing 66

with bee abundance and/or richness (Carvell 2002; Vulliamy et al. 2006; Yoshihara et al. 2008). Whether livestock grazing has a positive or negative effect on bee communities may depend upon various factors including the intensity of grazing, the types of grazers (e.g. cattle, goat, sheep), how long the grazing has occurred, the land use history, and the type of habitat.

However, previous work on native bees suffers from some common limitations associated with many livestock grazing studies. First, most studies have been observational (i.e., they have not experimentally manipulated grazing level), and thus lack the ability to infer a causal relationship (Sugden 1985; Carvell 2002; Kruess &

Tscharntke 2002; Steffan-Dewenter & Leschke 2003; Vulliamy et al. 2006; Hatfield &

LeBuhn 2007; Sjödin 2007; Xie et al. 2008; Kearns & Oliveras 2009).

Second, many studies have compared only the presence or absence of grazing, rather than a gradient of grazing intensities (e.g., Morris 1967; Sugden 1985; Xie et al.

2008). Studies that have looked at grazing intensities have had other shortcomings.

For example, Hatfield and LeBuhn (2007) examined three intensities of grazing but they used cows for their moderate intensity grazing treatment, and sheep for high, so that it is unclear whether the differences that they detected were due to grazing intensity or to different browsers. Although Yoshihara et al. (2008) examined a gradient of grazing intensities, they did not have a control that was ungrazed by livestock, so it is unknown how their results compare to a system without livestock.

Because many rural economies depend upon livestock grazing (Knight et al. 2002), grazing will likely continue to be a significant use of private land in the U.S. Thus, 67

understanding the relative costs and benefits of different intensities of grazing will allow ranchers and other land managers to make more informed decisions about sustainable grazing levels.

Third, many previous studies lack taxonomic resolution, reporting little information about livestock grazing at the species or even genus level (e.g., Gess &

Gess 1993; Steffan-Dewenter & Leschke 2003; DeBano 2006a, Hatfield and LeBuhn

2007; Yoshihara et al. 2008; Kearns & Oliveras 2009). Even within the same genus, different species exhibit variation in life histories. For example, some bumble bee species nest above ground and others nest underground (Goulson 2003). In fact,

Sjödin (2007) found that different species of bumble bees responded to grazing intensities differently. Thus, species level identification gives us more detailed information of how livestock grazing intensity affects bee communities, and allows conservation biologists and land managers to develop proper land management strategies for focal taxa.

Finally, few studies of livestock grazing effects on native bees have been conducted in North America. Hatfield and LeBuhn (2007) examined the effect of livestock grazing intensity on one group of bees, bumble bees, in one habitat – meadows in forested areas of the California Sierras, Kearns and Oliveras (2009) examined the effect of grazing regime and other environmental factors in urban and remote grassland plots in Colorado, and Sugden (1985) examined the potential impact of sheep grazing in California’s portion of the Great Basin. Thus, we have little information on how grazing impacts native pollinators of large scale grasslands that 68

form the majority of U.S. rangelands. Here, we describe the first manipulative study examining a gradient of livestock grazing intensities on native bee communities in a native grassland of North America; this large scale experiment overcomes many limitations of previous studies.

We have three mutually exclusive hypotheses with regard to how livestock grazing intensity affects native bee communities. One hypothesis is that increased grazing intensity will not impact native bee communities. Livestock might primarily eat plants that native bees do not use as nectar or pollen sources. For example, Wyffels

(2009) found that 92% of the biomass consumed by cattle was composed of grasses rather than forbs. Alternatively, increased grazing intensity may positively or negatively impact native bee communities. For example, grazing intensity may lead to greater soil compaction and more bare ground, which might enhance nesting habitats

(Potts et al. 2005; Vulliamy et al. 2006). Conversely, increased intensity may negatively affect bee abundance, richness or diversity because as vegetation is eaten, fewer flowers may be available, and plants may be killed directly by trampling

(Morris 2000; Sjödin 2007). Finally, the intermediate disturbance hypothesis predicts that moderate grazing will reduce the dominance of competitive plant species, such as some grasses, and result in a higher plant species richness as less competitive plants, like some forbs, increase (Curry 1994). In fact, several studies have suggested that plant diversity is highest at intermediate grazing (Bowers 1993; Noy-Meir 1995).

Some hypothesize that increased plant richness will result in higher bee richness or diversity (Potts et al. 2003). Thus, the intermediate disturbance hypothesis predicts 69

that the greatest bee diversity should occur at moderate levels of grazing. The specific objectives of this study were 1) to document changes in environmental variables associated with grazing intensity and 2) to describe associated responses in native bee abundance, richness, diversity, and species composition to the grazing gradient.

Methods

Study Area

We conducted the study within The Nature Conservancy’s (TNC) 13,269 ha Zumwalt

Prairie Preserve (lat 45º 3’N, long 116º 6’ W) in Wallowa County of northeastern

Oregon, USA (Fig. 3.1 a). Located at an elevation of 1,100-1,700 m, the Preserve receives a mean annual precipitation of 43.3 cm/yr, has a mean annual temperature of

6.4ºC, an average maximum temperature of 26.9ºC occurring in August, and an average minimum temperature of -7.9ºC occurring in December (30-year average,

1971-2000, at the Joseph, Oregon weather station, < 30 km from study site, NOAA

2010). Although the Zumwalt Prairie has been used as summer pasture for horse, sheep, and cattle for over 100 years, the majority of the area remains dominated by native species with most ecological processes still intact, and is considered to be an important refuge for an array of native biodiversity, including vertebrates, invertebrates, and plants (Kennedy et al. 2009). The Zumwalt Prairie is dominated by native grass species including Idaho fescue (Festuca idahoensis), Sandberg bluegrass

(Poa secunda), prairie Junegrass (Koeleria macrantha), and bluebunch wheatgrass

(Pseudoroegneria spicata) (Kennedy et al. 2009). In addition, over 112 forb species 70

have been documented in the Zumwalt Prairie Preserve

(http://conserveonline.org/workspaces/ZumwaltPrairieWorkspace/documents/zumwalt

-prairie-plant-list-.pdf/view.html). This rich forb community is associated with a diverse bee community with an estimated richness of more than 200 species in 27 genera; the most common genera in this prairie are Lasioglossum (Halictidae) (relative abundance up to 58%) and Bombus (Apidae) (relative abundance up to 33%), although the composition of bee community shows strong seasonal and yearly changes (Chapter

2).

Study Design

The study was designed as a large scale manipulation of four levels of stocking rates: high (24 cow-calf pairs or 0.93 Ha/AUM), medium (16 cow-calf pairs or 1.39

Ha/AUM), low (8 cow-calf pairs or 2.78 Ha/AUM), and no grazing. The moderate grazing intensity was designed to reflect prevailing stocking rates in the region.

Treatments were randomly assigned to 16 40-ha pastures on a plateau in the center of the Preserve in a randomized complete block design (Fig. 3.1b). Treatments were applied for two summers; in 2007, cattle grazed from 20 May to 2 July and in 2008, cattle grazed from 28 May to 8 July.

Utilization, or the percent of aboveground biomass removed by grazers, was estimated each year within a week after cattle were removed from all 16 pastures to quantify the intensity of grazing resulting from the stocking rates. The same stocking rate may result in different grazing pressure because of differences in grassland condition, cow behavior and physiology, and micro-distribution of resources in a 71

pasture (Allison 1985). Utilization was quantified using ocular estimation in 0.5 m2 (1 m x 0.5 m) areas at 36 uniformly spaced plots in each of the 16 pastures (for a total of

576 plots) (Fig. 3.1b) (Parsons et al. 2003; Wyffels 2009). Observers were trained to recognize five utilization categories: 1) 0% use, 2) 1-25% use, 3) 26-50% use, 4) 51-

75% use and 5) 76-100% use. Before collecting data in the field, each observer estimated the utilization in ten 0.25 m2 test plots that had been clipped at various levels. Then, the remaining vegetation of the test plots was clipped to within 2 cm of ground level and weighed to determine actual utilization values. Regression equations of estimated versus actual utilization values were developed for each observer to correct for observer bias (Wyffels 2009).

Environmental Variable Measurement

To evaluate changes in vegetation physical structure occurring during the experiment, pasture-level estimates of visual obstruction were obtained in 2007 and 2008. Visual obstruction was measured using a Robel pole held perpendicular to the ground, where an observer recorded the height of the lowest visible decimeter from 4 m away and 1 m off the ground (Robel et al. 1970). Vegetation height was measured at 10 m intervals along eight randomly placed 100 m transects within each pasture within 24 hours after cattle were removed from each pasture in early July (Johnson 2010). Data on the presence of blooming forbs were also collected in June, July, and August 2008.

We recorded the number of stems of each species of blooming forb that fell within a

50 m long, 0.3 m wide belt transect centered on each pollinator trap location

(described below). Each pasture had four transects. 72

To determine if short term exposure to grazing led to changes in plant composition, we measured pre- and post-treatment vegetation in 2006 and 2009 in 0.5 m2 areas within 36 uniformly placed 3 m radius sampling plots in each of the 16 pastures. Each sampling plot was separated by approximately 90 m (Fig. 3.1b).

Biomass of each plant species in each 0.5 m2 area was measured by clipping during the peak blooming season from mid-June to late July in 2006 and in July 2009, drying plant material with a forced air oven at 55 °C, and then weighing (Daubenmire 1959).

Plants were identified to species level, if possible, using Hitchcock and Cronquist

(1973). Biomass was estimated for each of 150 plant species at each of 576 plots. The data set was then reduced to 24 plant species that occurred at least 15% or more sites, as suggested by McCune and Grace (2002), in order to reduce the bulk and noise of the data set. We calculated a Sorenson dissimilarity index for each of the 576 plots (36 in each of 16 pastures) comparing plant community composition pre- and post- treatment.

Post-treatment soil characteristics, including surface coverage, compaction, and surface soil stability, were measured after cows were removed from pastures.

Compaction and surface soil stability was measured in August and September 2008 and percent soil coverage was measured in July 2009. Percent soil surface coverage was divided into five categories: bare ground, mosses and liverworts, lichen or other biological soil crust, herbaceous litter, and rock fragments (>5 mm) (see Damiran et al.

2007 for more detail). Soil compaction was measured up to a depth of 2 cm with a dynamic cone penetrometer using a 2-kg hammer dropped from a height of 60 cm 73

(Herrick & Jones 2002). Surface soil stability was measured at 0-3 mm depth using a modified slake test (Herrick et al. 2001), which is a qualitative visual estimation of how easily a small sample of soil falls apart when submersed in water. Soil stability data are presented on a scale of 1-6, with 6 being the most stable and 1 being the least stable (Schmalz, in prep.).

Pollinator Sampling

We sampled pollinators during the summers of 2007 and 2008 in each study pasture using ultra-violet reflective blue vane traps. Blue vane traps consist of a plastic container (15 cm diameter x 15 cm high) with a blue polypropylene screw funnel with two 24 x 13 cm semitransparent blue polypropylene cross vanes of 3 mm thickness

(SpringStar™ LLC, Woodinville, WA, USA; Stephen & Rao 2005). Traps were suspended approximately 1.2 m from the ground with wire hangers inserted into aluminum pipes (Fig. 3.1c). No liquids or other killing agents were used in traps.

Traps were placed in each of the cardinal directions halfway between the center of the pasture and the perimeter, resulting in four traps per pasture, except for June 2007, when only two traps were used per pasture. Traps within each pasture were separated by approximately 200 m from their nearest neighbor and were 360 m or more from traps in neighboring pastures (Fig. 3.1a).

Pollinators were sampled during two bouts in 2007 (18-20 June and 9-21 July) and three bouts in 2008 (7-16 June, 10-18 July, and 25-29 August). In June 2007 we sampled using 16 blue vane traps in 8 pastures; for all other sampling bouts we used

64 traps in 16 pastures (Fig. 3.1a). Elevation of traps ranged from 1372 to 1499 m. In 74

2007, traps were left open for two consecutive days and, because of high efficiency demonstrated in the first year, in 2008 they were left open for one day. Bees collected in the traps were frozen until they could be pinned, labeled, sexed, and identified to species, if possible, or morphospecies, if species identification was not possible. The availability of regional generic taxonomic keys varied, resulting in some genera having more morphospecies than others. For example, 99% of the specimens in the genus Bombus were identified to species and no morphospecies were used. In contrast,

Lasioglossum, one of the most common genera, had a high proportion of morphospecies, with approximately 36% of all morphospecies belonging to this genus.

Because of the accuracy with which Bombus could be identified, and the fact that it comprises a major component of the bee fauna, we focused on this genus in more depth for diversity and community composition analyses (see below). Halictus and

Lasioglossum were considered separate genera following Michener (2007). We treat B. californicus and B. fervidus as two separate species, although the uncertainty over their status is ongoing (Williams 2010). Representative specimens of all species and morphospecies are vouchered at the Oregon State Arthropod Collection at Oregon

State University in Corvallis. Abundances are expressed as the number of bees collected per trap per hour of daylight.

Statistical Analyses

Although the manipulation was designed as a randomized complete block design, quantification of grazing intensity using utilization showed that treatment effects displayed a continuous distribution rather than a categorical one (Fig. 3.2a,b). Since 75

utilization provides a more accurate measure of grazing intensity than stocking rate, linear regression analysis was used to determine how grazing intensity, as measured by utilization, affected environmental variables and native bee communities.

Environmental variables regressed against utilization included both vegetation and soil characteristics. Vegetation characteristics examined were vegetation physical structure in 2007 and 2008, abundance and richness of blooming forbs in 2008, and the

Sorenson dissimilarity index of the plant community from 2006 to 2009. Post- treatment soil characteristics examined included surface coverage of bare ground and herbaceous litter, compaction, and surface soil stability. For bee communities, we regressed utilization against abundance (the number of bees collected per trap per hour of daylight), species richness, Shannon diversity, and community composition. We examined abundance and species richness of all bees, most common genera (>7% relative abundance) and the most common species (>4% relative abundance) in each season. Shannon diversity indices were calculated for all bees and Bombus.

In addition, bee community and Bombus community composition were characterized by non-metric multidimensional scaling (NMS) ordinations. NMS ordination is a robust technique that is based on ranked distances and performs well with data that are not normally distributed and contain numerous zero values (McCune

& Grace 2002). NMS was run on the abundance of taxa using Sorenson’s distance measure. The best solution was determined through 250 runs of randomized data and dimensionality was determined by evaluating the relationship between final stress and the number of dimensions. We used Pearson’s correlation coefficients to quantify 76

relationships between bee species abundance and ordination axes results (McCune &

Mefford 2006). Ordinations were run using PC-ORD, version 5.19, set on “autopilot” mode (McCune & Mefford 2006). NMS scores for each pasture were used as dependent variables in regression analyses.

Because pasture was our experimental unit, all variables used in regressions were averaged for each pasture. We tested variables for normality using Lillefors’ test and log transformed, if non-normal. We averaged utilization over both years (2007 and 2008) and used this averaged utilization in all regressions for environmental and bee community variables except flowering richness and abundance, which were measured in 2008 only; for these variables we used utilization in 2008 only.

Each sampling month of each year was analyzed separately because bees in this area show substantial temporal variability in abundance, richness and community composition both among months and between years (Chapter 2). SYSTAT (1997)

Version 7.0. was used for linear regression analyses and PC-ORD, version 5.19

(McCune & Mefford 2006) was used for calculating dissimilarity and diversity indices and for ordinations. We used an α of 0.05 for all significant tests. Although multiple comparisons were made (i.e., for all bees and for common genera and species), we chose not to adjust α levels using Bonferroni corrections in these cases because we viewed these corrections as overly conservative, increasing the chances of dismissing ecologically meaningful results (Moran 2003). While Bonferroni corrections control the probability of Type I error (rejecting a false null hypothesis when it is true), this 77

control comes at the cost of committing at Type II error (failing to reject a null hypothesis when it is false).

Results

Environmental Variables

We found grazing intensity affected several environmental variables thought to be important in structuring bee communities including vegetation and soil characteristics.

Several immediate responses to grazing intensity were quantified during the treatment years. Vegetation structure decreased significantly as grazing intensity increased in

2007 and 2008 (Fig. 3.3a; r2=0.71, p<0.01; r2=0.64, p<0.01, respectively). There was no significant effect of grazing intensity on the richness of blooming forbs in June or

July of 2008 (r2=0.003, p=0.83; r2=0.000, p=0.99, respectively). No blooming forbs were recorded on any transect in August 2008. In addition, grazing intensity had no significant effect on the abundance of blooming forbs observed on belt transects in

June 2008 (r2=0.00, p=0.99), although bloom abundance showed a negative trend with grazing intensity in July 2008 (Fig. 3.3b; r2=0.19, p=0.09). The abundance of blooms of two common forb species showed a negative response to grazing intensity.

Increased utilization resulted in significantly fewer blooms of hoary balsamroot

(Balsamorhiza incana) in June 2008 (Fig. 3.3c; r2=0.35, p=0.02 ) and significantly fewer blooms of twin arnica in (Arnica sororia) in July 2008 (Fig. 3.3d; r2=0.30, p=0.03). Dissimilarity indices comparing vegetation composition from 2006 to 2009 tended to increase with increased grazing intensity (Fig. 3.3e; r2=0.18, p=0.10). 78

After two years of exposure to the grazing treatments, changes in soil characteristics were also detected. Grazing intensity significantly increased soil compaction (Fig. 3.4a; r2=0.80, p<0.001) and significantly decreased surface soil stability (Fig. 3.4b; r2=0.34, p=0.01). Grazing intensity also significantly increased the percent of bare ground (Fig. 3.4c; r2=0.40, p<0.01) and significantly decreased the percent of ground covered by herbaceous litter (Fig. 3.4d; r2=0.41, p<0.01).

Patterns in Abundance

A total of 7,124 bees were collected in 2007 and 2,034 bees in 2008. There were no statistically significant relationships of total bee abundance with grazing intensity for any season (Table 3.1), although in July 2007, there was a trend for a negative relationship of total bee abundance with grazing intensity (Table 3.1; Fig. 3.5a). At the genus level, in June of both years, Bombus and Lasioglossum, the most common genera, showed a consistent pattern. Specifically grazing intensity had a significant negative effect on Bombus abundance (Fig. 3.5b,c) but had no effect on Lasioglossum abundance (Table 3.1). In addition, in June 2008, Osmia abundance tended to decrease with grazing intensity while Andrena abundance tended to increase (Table 3.1; Fig.

3.5c) although these patterns were not statistically significant at α = 0.05. In July of both years and August 2008, there were no statistically significant relationships of abundance of any common genera with grazing intensity (Table 3.1). 79

Patterns in Richness and Diversity

For both years combined, 92 species and 119 morphospecies in 27 genera were identified (see Chapter 2, for species list). There were no statistically significant relationships of total species richness with grazing intensity in any season, although there was a negative trend in June 2007 (Table 3.2; Fig. 3.6a). The patterns in richness of common genera were similar to those shown in abundance of both years. In particular, in June 2008, Bombus richness decreased significantly and Osmia richness tended to decrease with grazing intensity (Table 3.2; Fig. 3.6b). In July of both years, there was no statistically significant effect of grazing intensity on the species richness of common genera (Table 3.2). In August 2008, Megachile richness significantly increased with grazing intensity (Table 3.2; Fig. 3.6c).

Shannon diversity of total bees was significantly affected by grazing intensity in multiple seasons (Table 3.2). In June, grazing intensity had a negative effect on bee diversity in both years, although this effect was only statistically significant in 2007

(Table 3.2; Fig. 3.7a,b). In July, total bee diversity was not affected by grazing intensity in 2007, while grazing intensity tended to have a positive effect on bee diversity in 2008 (Table 3.2; Fig. 3.7c,d). In August 2008, there were no statistically significant relationships of total bee diversity and grazing intensity (Table 3.2; Fig. 3.

7e). For Bombus, grazing intensity significantly decreased diversity in June 2008

(Table 3.2; Fig. 3.7b). In June 2007, July of both years, and August 2008, there were no statistically significant relationships of Bombus diversity and grazing intensity

(Table 3.3; Fig. 3.7a,c,d,e). 80

Patterns in Community Composition

Ordination results revealed that bee and Bombus community composition during most seasons varied among pastures, with 1-3 dimensional solutions that explained 59-94% of the variation in bee community composition (Table 3.3). There was no statistically significant relationship with total bee community composition with utilization in any season except in June 2008, when one axis was significantly affected by grazing intensity (Table 3.3; Fig. 3.8a). This axis, which explained 37% of the variation in community composition, was significantly correlated with: Bombus californicus

(r=0.78), Andrena morphospecies 1 (r=-0.72), Osmia morphospecies 1 (r=0.58), and

Andrena morphospecies 12 (r=-0.56) (Fig. 3.8b). Although grazing intensity did not have a statistically significant effect on Bombus community composition (Table 3.3), for one axis explaining 78% of the variation, there was a trend with grazing intensity in June 2007 (Table 3.3; Fig. 3.8c). The following species were correlated with the axis: Bombus appositus (r=-0.75), B. californicus (r=-0.75), B. flavifrons (r=0.89; Fig.

3.8d), B. mixtus (r=0.69), and B. nevadensis (r=0.87; Fig. 3.8d).

Patterns in the abundance of the most common species in each season were explored. Five species showed increased abundance as grazing intensity increased, while 17 species showed decreased abundance, and the abundance of one

Lasioglossum morphospecies did not show an increase or decrease (Table 3.4; Figs.

3.9 to 3.13). However, only one of these relationships was statistically significant; in

June 08, the abundance of B. californicus was significantly reduced by livestock grazng intensity (Table 3.4). 81

Discussion

If livestock grazing impacts native bees, it should do so by its effect on food and/or nesting resources – two of the most important needs of bees (Westrich 1996; Cane

2001). Flowering plants provide nectar and pollen resources to specialist and generalist bees alike (Michener 2007), plant material is used in nest construction, and the physical structure of plants plays a role for some bees such as those bumble bees that build above-ground nests in plants (Michener 2007). Likewise, soil characteristics are also key, primarily in their effect on ground-nesting bees, which, depending on the species, may prefer different levels of soil compaction, bare ground, and stability

(Potts & Willmer 1997, 1998; Wuellner 1999; Potts et al. 2005). We predicted that livestock grazing intensity could affect both plant and soil characteristics and, if so, that these effects would, in turn, impact native bee communities in this Pacific

Northwest Bunchgrass Prairie. Our results indicate that grazing intensity strongly reduced the physical structure of vegetation, the number of blooms of two common flower species (B. incana and A. sororia), and tended to decrease total bloom abundance and alter plant community composition. In addition, increased grazing intensity also affected soil characteristics by increasing soil compaction and the amount of bare ground and decreasing soil surface stability and the coverage of the soil surface by herbaceous litter. We found that bee communities also responded to grazing intensity, presumably in response to these alterations in environmental variables, as evident by changes in a variety of metrics including diversity, richness, and community composition. 82

Effect of Grazing Intensity on Native Bee Diversity

Our study demonstrated that patterns in total bee diversity were not consistent with the intermediate disturbance hypothesis. Patterns in bee diversity did not show “hump- shaped” curves consistent with this hypothesis. Instead, the relationship of grazing intensity with diversity was generally linear, with the slope of the relationship varying by season. In June of both years, increased grazing intensity was associated with decreasing diversity (although only statistically significantly so in June 2007), while in

July 2008, it was weakly associated with increasing diversity. Increases and decreases in the Shannon diversity index reflect changes in species richness and/or evenness.

The only time period that showed a trend for grazing intensity to affect total bee species richness was June 2007. Thus, the decrease in bee diversity associated with grazing intensity during that time period was probably primarily due to decreased species richness. In other months, when richness did not vary with grazing intensity, changes in diversity are likely due to changes in evenness. For example, in July 2008, the tendency for diversity to increase with grazing intensity is potentially due to decreased dominance of species such as Melissodes rivalis.

As with the entire bee community, patterns in diversity of bumble bees also did not support the intermediate disturbance hypothesis; moderate grazing intensities were not associated with the highest bumble bee diversity or richness. In June 2008, grazing intensity negatively affected bumble bee diversity, a pattern driven at least partially by reduced bumble bee species richness with increased grazing intensity that occurred during that month. 83

Our results relative to the intermediate disturbance hypothesis are difficult to compare with other studies because most do not report diversity indices. Diversity indices that incorporate both richness and evenness, such as the Shannon diversity index used here, provide the best test of the intermediate disturbance hypothesis because richness and evenness are hypothesized to change under the intermediate disturbance hypothesis paradigm. Examining just species richness is less than ideal because as systems move from low to moderate disturbance, strongly competitive species (e.g., k-selected species) that dominate the community are predicted to lose their competitive edge and decline in abundance. This may result in the competitive release of other species, which may increase in abundance. Under this scenario, changes in evenness may be more pronounced than changes in species richness.

Surprisingly, only one study on the effect of livestock grazing on native bee communities examined diversity indices. In contrast to our results, Kearns and

Oliveras (2009) found some support for the intermediate disturbance hypothesis in grasslands of Colorado, where bee diversity was positively correlated with flowering richness, and flowering richness was highest in plots with intermediate grazing disturbance. Other authors have used species richness in an attempt to test the intermediate disturbance hypothesis, and like us, did not find support for the hypothesis. For example, Vulliamy et al. (2006), who measured the effect of grazing intensity on species richness in forest habitat in Israel, suggested that the intermediate disturbance hypothesis was supported with regard to floral resources because, like

Kearns and Oliveras (2009), their moderate level grazing treatment had the highest 84

species richness of flowers. However, Vulliamy et al. (2006) did not see a corresponding relationship with bee richness. Instead, their most intensely grazed sites had the highest bee richness. Xie et al. (2008) also found no support for the intermediate disturbance hypothesis in their work with yak grazing in Tibet, finding that increased grazing intensity showed a significantly negative relationship with bumble bee richness.

Effect of Grazing Intensity on Native Bee Abundance and Community

Composition

Unlike diversity, total bee abundance did not show strong responses to grazing intensity in most seasons. However, there was a trend for decreased abundance with increasing grazing intensity in July 2007, a result that is consistent with some previous studies (Kruess & Tscharntke 2002; Xie et al. 2008; Kearns & Oliveras 2009) but contrasts with others that found that total abundance of bees increased with grazing intensity (e.g., Vulliamy et al. 2006). The lack of effect in most seasons on total bee abundance in our study is not surprising given the fact that some native bee taxa appeared to be negatively affected by livestock grazing, others appeared relatively insensitive, and yet others appeared to be positively affected by grazing. These differences in community composition were reflected both by changes in species composition and through differences in abundance and richness at the genus level in one or both years. 85

Effect of Grazing Intensity on Common Taxa

Bumble bees as a genus appear to be particularly sensitive to grazing intensity in the

Pacific Northwest Bunchgrass Prairie. Grazing intensity significantly reduced bumble bee abundance, richness, and diversity in June 2008, and significantly reduced abundance in June 2007. These effects were not apparent later in the season (i.e., in

July of either year or in August 2008). One mechanism that may be responsible for this decrease in abundance, richness, and diversity of bumble bees found in our study is that grazing may have resulted in a decrease of floral resources important to this group. Vegetation structure was significantly reduced by grazing, and resulted in significantly fewer flowers of two common flowering plants, as well as an overall tendency for fewer flowers of all types. Several studies have suggested that grazing effects on floral resources should be particularly important to bumble bees (Carvell

2002; Hatfield & LeBuhn 2007; Xie et al. 2008) and various general ecological studies have found that bumble bee abundance and/or species richness can be greatly influenced by floral resources (Steffan-Dewenter & Tscharntke 2001; Potts et al.

2003; Hines & Hendrix 2005; Holzschuh et al. 2007). Other studies on the effect of livestock grazing on bumble bees support this idea. Several studies have found that increased grazing resulted in decreased floral resources with a corresponding decrease in bee abundance and/or richness (Carvell 2002; Hatfield & LeBuhn 2007; Sjödin

2007; Xie et al. 2008). For example Hatfield and LeBuhn (2007) found grazing intensity negatively affected bumble bee richness and abundance in some years in montane meadows in the Sierra Nevadas of California. Sjödin (2007), who measured 86

the effect of grazing intensity in a semi-natural grassland in Sweden, indicated that livestock grazing decreased floral abundance and richness and this resulted in a lower abundance of bumble bees. Xie et al. (2008) also found a decrease in both bumble bee abundance and richness with increased yak grazing in Tibet. In contrast, in other systems, increased grazing has resulted in increases in the abundance and richness of some flowers, at least with a moderate level of grazing. Carvell (2002) found that some species of flowers in chalk grasslands of northwestern Europe were most abundant in land recently grazed by cattle, and this, in turn, was associated with significantly greater abundance and species richness of bumble bees, including both long-tongued species and short-tongued species. However, she found that abundance and richness of bumble bees decreased with more intensive grazing by sheep. These studies highlight the fact that the effect of grazing on bumble bees may ultimately depend on how floral resources are impacted, and this may vary depending on various factors, including the evolutionary history of the area, the grazer involved, and/or the type of habitat. Interestingly, none of these studies examined other environmental variables associated with nesting habitat, such as vegetation structure for bumble bees that build nests above ground in grass tussocks, and soil compaction or stability for those species that nest in the ground (Michener 2007). Our results suggest that these variables vary concomitantly with grazing intensity in our system, and thus may also be negatively impacting bumble bees species. Further research is needed to tease apart the relative importance of these factors. 87

It is important to note that even though bumble bees, in general, appear to be sensitive to grazing intensity, we found that different species of bumble bees appear to be more sensitive to grazing than others. In our study, grazing intensity tended to alter bumble bee community composition in June 2007, and in June 2008, one bumble bee species, B. californicus, which made up 5% of the entire native bee fauna, was among the species making a significant contribution to total bee community composition differences. Grazing intensity significantly reduced the abundance of this common bumble bee species in that month. Other common species of Bombus showed similar patterns. For example, B. flavifrons, which was the most common species in June 2007, comprising 11% of all bees, showed a tendency to be negatively affected by grazing intensity. In contrast, other common species, such as B. bifarius which comprised 3% of the community, did not appear to respond to grazing intensity. These apparent differences in sensitivity may be driven by differences in life history, including morphology associated with foraging strategy (e.g., tongue length) and different requirement for nesting (Goulson 2003). Surprisingly, few studies have examined the effect of grazing intensity on the species composition of bumble bees. Carvell (2002) found that the highest abundance of four species of bumble bees was found in the recently cattle grazed areas (compared to ungrazed and sheep grazed areas), but another bumble bee species had higher abundances in disturbed edge habitat. Sjödin

(2007) found that four species of bumble bees decreased with increased grazing intensity, while one species increased. Each of these studies demonstrate that different species of bumble bees can respond to grazing intensity differently, and that a 88

particular species response potentially depends on a number of factors, including how grazing intensity affects the floral resources a particular species uses, whether the bee is a specialist or generalist, the type of grazing regime, the habitat and its evolutionary history, the season of the year, and the type of grazer.

Unlike Bombus, we found genera in the family Halictidae (e.g., Lasioglossum and Halictus) to be relatively insensitive to grazing intensity. This may be because 1) the family as a whole is not negatively impacted by grazing, or 2) different halictid species with varying life histories respond to grazing intensity differently, making family or genus level patterns less obvious. Surprisingly, none of the studies on the effect of grazing intensity on bee communities examined the effect of grazing intensity on the abundance and richness of this common family except for Vulliamy et al.

(2006). Their work suggests that the family as whole was actually positively affected by grazing, with halictid abundance showing a strong positive response to grazing intensity. However, of note is the fact that in their study, one halictid species,

Lasioglossum marginatum, comprised ~93% of all Halictidae, so the family level pattern in their study may be driven by a single species. In fact, they speculate that the response of halictids (and thus primarily this one species) may account for the positive effect of grazing on total bee abundance because this species made up 19.6% of the bee community. In our study, the most common halictid genus was Lasioglossum, which was comprised of 47 species and made up 34-58% of the community, depending on the season. Unlike Vulliamy et al. (2006), we found several common halictid species showed tendencies to respond to grazing intensity in different ways. 89

For example, in June 2007, the abundance of two Lasioglossum species, which together comprised 19% of the community, tended to increase with grazing intensity.

In contrast, in July 2007, two different species of Lasioglossum, which comprised

10% of the community, tended to decrease with increased grazing intensity. As such, different halictid species may be responding to grazing intensity differently, resulting in a lack of pattern at the genus level. However, none of the responses were statistically significant, suggesting that grazing intensity is not exerting a strong effect, regardless of the species.

Halictids may not be strongly negatively affected by grazing in our system because some of the changes in environmental variables associated with grazing may benefit the group, or at least offset negative effects that grazing has on other environmental variables. In general, members of this group are thought to prefer bare ground and compacted soils for their nesting sites (Potts & Willmer 1997, 1998; Potts et al. 2005; Vulliamy et al. 2006). We found, like others, that soil compaction and the amount of bare ground increased with grazing intensity (Sugden 1985; Gess & Gess

1993; Vulliamy et al. 2006), suggesting that grazing intensity may improve nesting habitat for halictids. In contrast, floral resources for Halictidae may decrease with grazing in our system. Other work has indicated that many halictid bees prefer to forage in small composites (Proctor et al. 1996). In fact, Vulliamy et al. (2006) found that halictid abundance was positively correlated with composite cover, and in their system, the number of composite flowers increased with higher grazing intensity.

This may explain why Vulliamy et al. (2006) found a positive effect of grazing 90

intensity on halictids and we found weak to no effects. In our system, grazing resulted in increased soil compaction and bare ground, but also resulted in decreased floral resources of some composites, such as B. incana and A. sororia. Thus, grazing in our system may have impacted one set of environmental variables in a way that benefited halictids, but impacted a different set in a way that negatively affected this group. In contrast, in the system that Vulliamy et al. (2006) worked in, grazing affected both vegetation and soil characteristics in a way that should benefit halictids. One reason

Vulliamy et al. (2006) found a higher abundance of composites with increased grazing intensity may be related to land use history. Israel has a long history of grazing, and plant species have potentially evolved adaptations to relatively high grazing pressure

(Naveh & Whittaker 1980).

In contrast to bumble bees and halictid bees, one common genus showed a tendency to be positively influenced by increased grazing intensity. In June 2008,

Andrena, which comprised 8% of relative abundance of the bee community, tended to be more abundant with increased grazing intensity. In addition, we found two Andrena morphospecies in June 2008 (that made up more than 3% of the community) were among the species making a significant contribution to bee community composition differences in that month, with each showing increased abundance with grazing intensity. Few studies have examined the effect of grazing on this genus. Vulliamy et al. (2006) found no correlation between abundance of Andrenidae and grazing intensity. In contrast, Sjödin et al. (2008) found four species of Andrena were more dominant in semi-natural grasslands exposed to low intensive grazing in Sweden while 91

two species were more dominant in grasslands exposed to intensive grazing. One reason why Andrena may have been positively influenced by increased grazing intensity in our study is because of grazing’s influence on soil characteristics. All

Andrena species nest in the ground (Michener 2007), and Kearns and Oliveras (2009) showed that the abundance of ground nesting bees in Colorado was highest in areas that had been grazed for one year compared to areas with lower and higher intensity grazing. As a result, nesting sites for Andrena may have been enhanced by the two- year grazing exposure in our study. Another possibility is that cattle grazing may have positively influenced flowering resources for Andrena in our study. Although most species of Andrena are polylectic, some are oligolectic (Michener 2007). If livestock grazing decreased the abundance of some dominant plants, it may have increased the abundance of others (a la the intermediate disturbance hypothesis), potentially including some that are preferred by certain species of Andrena. This may explain the results of Sjödin et al. (2008), who found different species of Andrena responded differently to grazing intensity. Although we found no evidence of increases in the number of blooms of any plant, our results indicated that vegetation community composition showed a tendency to change pre- and post-treatment. This may indicate that certain preferred floral resources for Andrena may have increased. However, more research is needed to understand 1) which flowers these Andrena use, and 2) how those flower species are affected by livestock grazing intensity in the Pacific

Northwest Bunchgrass Prairie. 92

In contrast to Andrena, another early season genus tended to show a negative response to increased grazing intensity. In June 2008, Osmia comprised 7% of the bee community, and tended to be less species rich and abundant with increased grazing intensity. In addition, one Osmia morphospecies, which comprised 1% of the bee community in June 2008, was among the species making a significant contribution to differences in bee community composition in that month, showing decreased abundance with increased grazing intensity. In general, Osmia species are generalist feeders and most nest in cavities in pithy stems and wood, including abandoned beetle burrows, in crevices under or between stones, in the soil, including in abandoned

Anthophora bee nests, in abandoned Sceliphron wasp nests, and even in empty snail shells (Cane et al. 2007; Michener 2007). In our study, seven species that comprise

62% of the Osmia collected in this study are believed to nest in the soil, on the surface of the soil, or on sides of rocks (Cane et al. 2007). These nests may be disturbed by direct trampling, or decreasing soil stability and compacting soils may lead to the filling in of cavities used for nesting. Unfortunately, no other studies of livestock grazing intensity have examined the effect on this genus. Vulliamy et al. (2006) examined Megachilidae (the family that includes Osmia) as a whole, but found that it was not affected by grazing intensity. Similarly, Kearns and Oliveras (2009) found that abundance and richness of cavity nesting bees in general were not affected by grazing intensity. One possible reason the findings of these studies differ from ours is that the groups they examined included many genera beyond Osmia. Other genera in

Megachilidae or in the general groups of cavity nesters may respond differently. 93

Indeed, in our study we found that species richness of one genus, Megachile, a member of Megachilidae, was slightly higher with increased grazing intensity in

August 2008. Thus, these results suggest that even within a family that includes primarily cavity nesters, different genera may respond to grazing intensity differently.

Temporal Variability

We found substantial variation in the effects of grazing intensity on bee communities, and even the direction of that effect, based on season and year. The strongest effects of grazing intensity on bee communities occurred in June rather than July or August, and those responses varied from one year to another. Other studies have also found that the effect of grazing intensity on both plant and bee communities varies temporally, both according to season and from year to year (Hatfield & LeBuhn 2007; Sjödin 2007).

Several factors potentially play a role in causing temporal variation in bee communities including the timing of the grazing regime. In our system, cattle were grazed in early season, and were removed from pastures by early July, which may explain why we found the largest effects on bees in June. In contrast, Hatfield and

LeBuhn (2007) found that grazing intensity negatively affected bumble bees primarily late in the season, and hypothesized that this pattern was due to the fact that their most intensive grazing regime (sheep grazing) occurred late in the season. As Hatfield and

LeBuhn (2007) point out, late season grazing regimes may be more harmful than early grazing regimes to some native bees, like bumble bees, that produce reproductive members at the end of the season, since the reproductive success of these individuals will determine the future fitness of the colony. Coincidentally, in Hatfield and 94

LeBuhn’s study peak bumble bee abundance occurred late in the season, when the impact of grazing was largest; in our system bumble bee abundance is highest in July

(Chapter 2).

Changes in temperature and moisture can also affect the timing of livestock grazing effects on bees by impacting the phenology of the bee communities directly, or indirectly through their effects on plant phenologies. In the Zumwalt Prairie, temperatures not only vary strongly from June to August, but also from year to year.

In the period of our study, mean daily temperature in 2007 was 12.7 C° in June and

20.6 C° in July. Early summer in 2008 was much colder and wetter, with a mean daily temperature of 11.4 C° in June, 16.6 C° in July, and 16.2 C° in August (Hansen et al.

2010). In 2008, both plant and bee community phenologies were delayed (Dingeldein et al. 2009; Chapter 2). For example in June 2008 Andrena and Osmia, which are early season bees (Wojcik et al. 2008), were much more common than in June 2007. Thus, the reason why these two genera showed patterns with regard to grazing intensity in

June 2008 and not in June 2007 is simply because they were common in 2008 compared to 2007. In addition, changes in plant phenology affect the availability of nectar and pollen sources, so even for bee taxa that are active the entire summer, grazing intensity may affect a particular food source in one part of the season and not another. Finally, the value of particular environmental characteristics may vary depending on the season and year. A south-facing slope in a cool season may be more valuable (both because it may have more flowers blooming and warmer temperatures) than in the middle of August in a particularly warm year. More work is needed to 95

understand how temporal variation affects the interaction between bee communities, their resources, and their physical environment.

Management Implications

This study is the first large-scale manipulative study of the effect of grazing intensity on native bee communities in a North American grassland. Our results indicate that grazing intensity on the Pacific Northwest Bunchgrass Prairie affects native bee communities, whose members are vital pollinators for many native flowering plants.

These native bees also have the potential of providing ecosystem services in the form of enhanced pollination for adjacent agricultural lands, if properly managed.

Understanding how livestock grazing intensity affects these pollinators in grasslands that form a major component of U.S. rangelands is important for guiding sustainable agriculture production. Grazing is an important component of rural economies, and as such, will continue on both private and public land in the western United States

(Knight et al. 2002). As a result, understanding the relative costs and benefits associated with different grazing intensities is important for land managers. Because different species of bees respond to grazing intensity differently, to ensure a diverse native bee community, areas with different levels of grazing should be encouraged. In the western United States, since many grasslands are already used as rangeland, this research suggests the importance of maintaining some ungrazed or minimally grazed lands, to serve as reservoirs for those bees that are negatively impacted by more intense grazing. Our research does not support the idea that moderate grazing intensity leads to higher bee diversity in these grasslands, at least under short term grazing 96

regimes, as suggested by the intermediate disturbance hypothesis. This may be due to the fact the Pacific Northwest Bunchgrass Prairie is not believed to have evolved in the presence of large native ungulates, like bison, which were common in midwestern grasslands (Vulliamy et al. 2006). Finally, our research indicates that certain groups, as a whole, may be more sensitive than others. In our study, bumble bees appeared to be more sensitive than other groups. Bumble bees are often of special conservation interest because of 1) the existence of several threatened or extinct species in the western United States (Thorp 2003; Thorp & Shepherd 2005; Rao & Stephen 2007),

2) their role as the major pollinator of some rare or threatened plants, like Spalding’s catchfly (Silene spaldingii) in the Pacific Northwest Bunchgrass Prairie (Lesica 1993), and 3) their potential economic value as crop pollinators (Goulson et al. 2008). Thus, if sensitive groups are of special conservation interest, grazing intensity may need to be adjusted accordingly, or larger areas of ungrazed or lightly grazed reservoirs may need to be maintained. Finally, similar research is needed to understand how other components of biodiversity and ecological function are affected by grazing intensity in western North American grasslands, so a more holistic view of the relative costs and benefits of different management regimes can be assessed.

97

Literature Cited

Allison, C. D. 1985. Factors affecting forage intake by range ruminants: a review. Journal of Range Management 38:305-311.

Belsky, A. J., and D. M. Blumenthal. 1997. Effects of livestock grazing on stand dynamics and soils in upland forests of the interior West. Conservation Biology 11:315-327.

Bowers, M. A. 1993. Influence of herbivorous mammals on an old-field plant community: years 1-4 after disturbance. Oikos 67:129-141.

Brockmann, H. J. 1979. Nest-site selection in the great golden digger wasp, Sphex ichneumoneus L. (Sphecidae). Ecological Entomology 4:211-224.

Buchmann, S. L., and G. P. Nabhan. 1996. The forgotten pollinators. Island Press, Washington, D.C.

Cane, J. H. 1991. Soils of ground-nesting bees (Hymenoptera: Apoidea): texture, moisture, cell depth and climate. Journal of the Kansas Entomological Society 64:406-413.

Cane, J. H. 2001. Habitat fragmentation and native bees: a premature verdict? Conservation Ecology 5:3: http://www.ecologyandsociety.org/vol5/iss1/art3. Cane, J. H., T. Griswold, and F. D. Parker. 2007. Substrates and materials used for nesting by North American Osmia bees (Hymenoptera: Apiformes: Megachilidae). Annals of the Entomological Society of America 100:350- 358.

Carvell, C. 2002. Habitat use and conservation of bumblebees (Bombus spp.) under different grassland management regimes. Biological Conservation 103:33-49.

Crumpacker, D. W. 1984. Regional riparian research and a multi-university approach to the special problem of livestock grazing in the Rocky Mountains and Great Plains. Pages 413-423 in R. E. Warner and K. M. Hendrix, editors. California riparian systems: ecology, conservation, and productive management. University of California Press, Berkeley, California.

Curry, J. P. 1994. Grassland Invertebrates: ecology, influence on soil fertility and effects of plant growth. Chapman and Hall, London. Damiran D., T. DelCurto, E. Darambazar, P. L. Kennedy, R. V. Taylor, and A. A. Clark. 2007. Northwestern bunchgrass prairie monitoring sites databases. 98

Circular of information No. 5, Union, Oregon, USA: Eastern Oregon Agricultural Research Center, Oregon State University.

Daubenmire, R. 1959. A canopy-coverage method of vegetational analysis. Northwest Science 33:43-64.

Davis, J. D., S. D. Hendrix, D. M. Debinski, and C. J. Hemsley. 2008. Butterfly, bee and forb community composition and cross-taxon incongruence in tallgrass prairie fragments. Journal of Insect Conservation 12:69-79.

Day, T. A., and J. K. Detling. 1990. Grassland patch dynamics and herbivore grazing preference following urine deposition. Ecology 71:180-188.

DeBano, S. J. 2006a. Effects of livestock grazing on aboveground insect communities in semi-arid grasslands of southeastern Arizona. Biodiversity and Conservation 15:2547-2564.

DeBano, S. J. 2006b. The effect of livestock grazing on the rainbow grasshopper: population differences and ecological correlates. Western North American Naturalist 66:222-229.

Donahue, D. L. 1999. The western range revisited: removing livestock from public lands to conserve native biodiversity. University of Oklahoma Press, Norman, Oklahoma.

Fleischner, T. L. 1994. Ecological costs of livestock grazing in western North America. Conservation Biology 8:629-644.

Gess, F. W., and S. K. Gess. 1993. Effects of increasing land utilization on species representation and diversity of aculeate wasps and bees in the semi-arid areas of southern Africa. Pages 83-113 in J. La Salle and I. D. Gauld, editors. Hymenoptera and biodiversity. CAB International, Wallingford, United Kingdom.

Gibson C. W. D., T. A. Watt, and V. K. Brown. 1987. The use of sheep grazing to recreate species-rich grassland from abandoned arable land. Biological Conservation 42:165-183.

Goulson, D. 2003. Bumblebees behaviour and ecology. Oxford University Press. United Kingdom.

Goulson, D., G. C. Lye, and B. Darvill. 2008. Decline and conservation of bumble bees. Annual Review of Entomology 53:191-208.

99

Hansen, J.D., R.V. Taylor, and V.S. Jansen, 2010. Zumwalt weather and climate annual report (2006-2009). The Nature Conservancy. Available from http://conserveonline.org/workspaces/ZumwaltPrairieWorkspace/documents/z umwalt-weather-and-climate-annual-report-2006/@@view.html (accessed Janurary 2011).

Harris, P. S. 2000. Grassland resource assessment for pastoral systems. Food and Agriculture Organization of the United Nations, Rome, Italy.

Hatfield, R. G., and G. LeBuhn. 2007. Patch and landscape factors shape community assemblage of bumble bees, Bombus spp. (Hymenoptera: Apidae), in montane meadows. Biological Conservation 139:150-158.

Herrick, J. H., W. G. Whitford, A. G. de Soyza, J. W. Van Zee, K. M. Havstad, C. A. Seybold, and M. Walton. 2001. Field soil aggregate stability kit for soil quality and rangeland health evaluations. Catena 44:27-35.

Herrick, J. E., and T. L. Jones. 2002. A dynamic cone penetrometer for measuring soil penetration resistance. Soil Science Society of America Journal 66:1320-1324.

Hines, H. M. and S. D. Hendrix. 2005. Bumble bee (Hymenoptera: Apidae) diversity and abundance in tallgrass prairie patches: effects of local and landscape floral resources. Environmental Entomology 34:1477-1484.

Hitchcock, C. L., and A. Cronquist. 1973. Flora of the Pacific Northwest: an illustrated manual. University of Washington Press, Seattle, Washington.

Holzschuh, A., I. Steffan-Dewenter, D. Kleijn, and T. Tscharntke. 2007. Diversity of flower-visiting bees in cereal fields: effects of farming system, landscape composition and regional context. Journal of Applied Ecology 44:41-49.

Huntly, N. 1991. Herbivores and the dynamics of communities and ecosystems. Annual Review of Ecology and Systematics 22:477-503.

Johnson, T. M. 2010. The effects of livestock grazing intensity on grassland songbird nestling diet composition, prey selectivity, and growth rates. Dissertation, Oregon State University.

Jones, A. 2000. Effects of cattle grazing on North American arid ecosystems: a quantitative review. Western North American Naturalist 60:155-164.

Kearns, C. A., D. W. Inouye, and N. M.Waser. 1998. Endangered mutualisms: the conservation of plant-pollinator interactions. Annual Review of Ecology and Systematics 29:83-112. 100

Kearns, C. A., and D. M. Oliveras. 2009. Environmental factors affecting bee diversity in urban and remote grassland plots in Boulder, Colorado. Journal of Insect Conservation 13:655-665.

Kells, A. R., and D. Goulson. 2003. Preferred nesting sites of bumblebee queens (Hymenoptera: Apidae) in agroecosystems in the UK. Biological Conservation 109:165-174.

Kennedy, P. L., S. J. DeBano, A. M. Bartuszevige, and A. S. Lueders. 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.

Klein, A. M., B. E. Vaissière, J. H. Cane, I. Steffan-Dewenter, S. A. Cunningham, C. Kremen, and T. Tscharntke. 2007. Importance of pollinators in changing landscapes for world crops. Proceedings of the Royal Society of London Series B-Biological Sciences 274:303-313.

Knight, R. L., W. C. Gilgert, and E. Marston. 2002. Ranching west of the 100th meridian: culture, ecology, and economics. Island Press, Washington, D.C.

Kruess, A., and T. Tscharntke. 2002. Grazing intensity and the diversity of grasshoppers, butterflies, and trap-nesting bees and wasps. Conservation Biology 16:1570-1580.

Kwaiser, K. S., and S. D. Hendrix. 2008. Diversity and abundance of bees (Hymenoptera: Apiformes) in native and ruderal grasslands of agriculturally dominated landscapes. Agriculture, Ecosystems and Environment 124:200- 204.

Lesica, P. 1993. Loss of fitness resulting from pollinator exclusion in Silene spaldingii (Caryophyllaceae). Madroňo 40:193-201.

Loftin, S. R., C. E. Bock, J. H. Bock, and S. L. Brantley. 2000. Desert grasslands. Pages 53-96 in R. Jemison and C. Raish, editors. Livestock management in the American Southwest: ecology, society, and economics. Elsevier, New York.

McCune, B., and J. B. Grace. 2002. Analysis of Ecological Communities. MjM Software, Gleneden Beach, Oregon.

McCune, B., and M. J. Mefford. 2006. PC-Ord, Multivariate Analysis of Ecological Data, Version 5.19. MjM Software, Gleneden Beach, Oregon. 101

Michener, C. D. 2007. The bees of the world. 2nd edition. Johns Hopkins University Press, Baltimore, Maryland.

Moran, M. D. 2003. Arguments for rejecting the sequential Bonferroni in ecological studies. Oikos 100:403-405.

Morris, M. G. 1967. Differences between the invertebrate faunas of grazed and ungrazed chalk grassland. I. Responses of some phytophagous insects to cessation of grazing. Journal of Applied Ecology 4:459-474.

Morris, M. G. 1981. Reponses of grassland invertebrate to management by cutting. III. Adverse effects on Auchenorhyncha. Journal of Applied Ecology 18:107-123.

Morris, M. G. 2000. The effects of structure and its dynamics on the ecology and conservation of arthropods in British grasslands. Biological Conservation 95:129-142.

Naveh, Z., and R. H. Whittaker. 1980. Structural and floristic diversity of shrublands and woodlands in northern Israel and other Mediterranean areas. Vegetatio 41:171-190.

(NOAA) National Oceanic and Atmospheric Administration. 2010. National Weather Service Office. Available from http://www.weather.gov/climate/xmacis.php?wfo=pdt (accessed December 2010).

Noss, R. F. 1994. Cows and conservation biology. Conservation Biology 8:613-616.

Noy-Meir, I. 1995. Interactive effects of fire and grazing on structure and diversity of Mediterranean grasslands. Journal of Vegetation Science 6:701-710.

(NRC) National Research Council. 2007. Status of pollinators in North America. The National Academies Press, Washington, D.C.

Parsons, C. T., P. A. Momont, T. DelCurto, M. Mcinnis, and M. L. Porath. 2003. Cattle distribution patterns and vegetation use in mountain riparian areas. Journal of Range Management 56:334-341.

Potts, S. G., and P. Willmer. 1997. Abiotic and biotic factors influencing nest-site selection by Halictus rubicundus, a ground-nesting halictine bee. Ecological Entomology 22:319-328.

102

Potts, S. G., and Willmer, P. 1998. Compact housing in built-up areas: spatial pattering of nests in aggregations of a ground-nesting bee. Ecological Entomology 23:427-432.

Potts, S. G., B. Vulliamy, A. Dafni, G. Ne’eman, and P. Willmer. 2003. Linking bees and flowers: how do floral communities structure pollinator communities? Ecology 84:2628-2642.

Potts, S. G., B. Vulliamy, S. Roberts, C. O’Toole, A. Dafni, G. Ne’eman, and P. Willmer. 2005. Role of nesting resources in organizing diverse bee communities in a Mediterranean landscape. Ecological Entomology 30:78-85.

Proctor, M. C. F., P. Yeo, and A. Lack. 1996. The natural history of pollination. Timber Press, Portland Oregon.

Purvis, G., and J. P. Curry. 1981. The influence of sward management on foliage arthropod communities in a ley grassland. Journal of Applied Ecology 18:711- 725.

Pykälä, J. 2000. Mitigating human effects on European biodiversity through traditional animal husbandry. Conservation Biology 14:705-712.

Rao S, and W. P. Stephen. 2007. Bombus (Bombus) occidentalis (Hymenoptera: Apiformes): In decline or recovery? The Pan-Pacific Entomologist 83:360-362.

Reed, C. C. 1995. Insect surveyed on flowers in native and reconstructed prairies (Minnesota). Restoration and Management Notes 13:210-213.

Robel, R. J., J. N. Briggs, A. D. Dayton, and L. C. Hulbert. 1970. Relationships between visual obstruction measurements and weight of grassland vegetation. Journal of Range Management 23:295-297.

São Paulo Declaration on Pollinators. 1999. Report on the recommendations of the workshop on the conservation and the sustainable use of pollinators in agriculture with emphasis on bees. Brazilian Ministry of the Environment, Brasilia, Brazil.

Schmalz, H. In prep.Characterizing fine-scale soil heterogeneity to improve ecological understanding in grazing research.

Sjödin, N. E. 2007. Pollinator behavioural responses to grazing intensity. Biodiversity and Conservation 16:2103-2121.

103

Sjödin, N. E., J. Bengtsson and B. Ekbom. 2008. The influence of grazing intensity and landscape composition on the diversity and abundance of flower-visiting insects. Journal of Applied Ecology 45:763-772.

Steffan-Dewenter, I., and T. Tscharntke. 2001. Succession of bee communities on fallows. Ecography 24:83-93

Steffan-Dewenter, I., and K. Leschke. 2003. Effects of habitat management on vegetation and above-ground nesting bees and wasps of orchard meadows in Central Europe. Biodiversity and Conservation 12:1953-1968.

Stephen, W. P., and S. Rao. 2005. Unscented color traps for non-Apis bees (Hymenoptera: Apiformes). Journal of the Kansas Entomological Society 78:373-380.

Sugden, E. A. 1985. Pollinators of Astragalus monoensis Barneby (Fabaceae): new host records; potential impact of sheep grazing. Great Basin Naturalist 45:299- 312.

SYSTAT. 1997. SYSTAT 7.0 for Windows. SPSS Inc., Chicago, Illinois.

Tepedino, V. J., and N. L. Stanton. 1981. Diversity and competition in bee-plant communities on short-grass prairie. Oikos 36:35-44.

Thomson, D. 2004. Competitive interactions between the invasive European honey bee and native bumble bees. Ecology 85:458-470.

Thorp, R. W. 2003. Bumble bees (Hymenoptera:Apidae): commercial use and environmental concerns. Pages 21-40 in K. Strickler and J. H. Cane, editors. For nonnative crops, whence pollinators of the future? Thomas Say Publications in Entomology: Proceedings. Entomological Society of America, Lanham, Maryland.

Thorp, R. W., and M. D. Shepherd. 2005. Profile. Subgenus Bombus Latreille 1802 (Apidae: Apinae: Bombini). Page 5 in M. D. Shepherd, D. M. Vaughan, S. H. Black, editors. Red list of pollinator insects of North America. Xerces Society for Invertebrate Conservation, Portland, Oregon.

Vulliamy, B., S. G. Potts, and P. G. Willmer. 2006. The effects of cattle grazing on plant-pollinator communities in a fragmented Mediterranean landscape. Oikos 114:529-543.

Westrich, P. 1996. Habitat requirements of central European bees and the problems of partial habitats. Pages 1-16 in A. Matheson, S. L. Buchmann, C. O’Toole, P. 104

Westrich and I. H. Williams, editors. The conservation of bees. Academic Press, London, United Kingdom.

Williams, P. 2010. Bombus (Th.) fervidus (Fabricius). Available from http://www.nhm.ac.uk/research- curation/research/projects/bombus/th.html#fervidus (accessed December 2010).

Wuellner, C. T. 1999. Nest site preference and success in a gregarious, ground-nesting bee Dieunomia triangulifera. Ecological Entomology 24:471-479.

Wyffels, S. A. 2009. Influence of stocking density on grazing beef cattle performance, diet composition foraging efficiency, and diet quality on a late-spring early summer native bunchgrass prairie. Master’s Thesis, Oregon State University.

Xie, Z., P. H. Williams, and Y. Tang. 2008. The effect of grazing on bumblebees in the high rangelands of the eastern Tibetan Plateau of Sichuan. Journal of Insect Coservation 12:695-703.

Yoshihara, Y., B. Chimeddorj, B. Buuveibaatar, B. Lhagvasuren, and S. Takatsuki. 2008. Effects of livestock grazing on pollination on a steppe in eastern Mongolia. Biological Conservation 141:2376-2386. 105

a) * ** * * * **** * * * ** Zumwalt Prairie * Area Enlarged * * 0 500 1000 m * * * * OREGON * * * * N * * * * * * * * **** * * * * **** * *

b)

c)

Figure 3.1. a) Location of the Zumwalt Prairie in northeastern Oregon and location of pollinator traps in each pasture (all sites were sampled in each season of each year except that unshaded pastures and traps denoted with “*” were not sampled in June 2007), b) map showing stocking levels for each 40 ha pasture (H-high, M-medium, L- low, C-control) and a close-up view of the 36 subplots in each pasture, where vegetation and soil characteristics were sampled, and c) blue vane traps. 106

a) 60

50

40

30

20. (%) Utilization 10

0 Control Low Medium High Stocking rate b) 80

70

60 50

40

30

(%) Utilization 20

10

0 Control Low Medium High Stocking rate

Figure 3.2. Average utilization in a) 2007, and b) 2008. 107

a) 1.6 1.4

1.2

1.0 2007 0.8 2008 0.6 0.4

Robel of measurement 0.2 vegetation structure (dm) 0.0 0 10203040506070 Utilization (%) b) 80.0 70.0

60.0

50.0

40.0

30.0

20.0

10.0

transect / forbs # of blooming 0.0 0 10203040506070 Utilization (%) c) 5.0

4.0

3.0

2.0

1.0

0.0

/ transect # of forbs blooming -1.0 0 10203040506070 Utilization (%) d) 30.0 25.0

20.0

15.0

10.0

5.0

0.0 # of blooming forbs / transect -5.0 0 10203040506070 Utilization (%) e) 0.6 0.5 0.4 0.3 0.2 0.1 Disassimilarity Index Disassimilarity

0 0 10203040506070 Utilization (%) Figure 3.3. Effect of grazing intensity on a) vegetation physical structure in 2007 and 2008, b) blooming forb abundance in July 2008, c) blooming balsamroot abundance in June 2008, d) blooming twin arnica abundance in July 2008, and e) dissimilarity of plant community composition from pre- to post-treatment. 108

a) 2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 Compaction (strikes/cm) Compaction 0.0 0 10203040506070 Utilization (%) b) 6.0 5.0

4.0

3.0

2.0

1.0

Surface soil stability (Slake rnaks) 0.0 0 10203040506070 Utilization (%)

c) 30.0 25.0

20.0

15.0

10.0

(%) ground Bare 5.0 0.0 0 10203040506070 Utilization (%) d) 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 Herbaceous litter (%) 0.0 0 10203040506070 Utilization (%) Figure 3.4. Effect of grazing intensity on a) soil compaction, b) surface soil stability, c) percent bare ground, and d) percent herbaceous litter. 109

a) 6.00

5.00

4.00

3.00

2.00

1.00 # of bees / / hour of trap # bees

0.00 0 102030405060

Utilization (%) b) 0.30

0.25 0.20

0.15

0.10

0.05 # of bees / trap/ hour

0.00 0 5 10 15 20 25 30 35 Utilization (%) c) 0.25

0.20

0.15 Bombus 0.10 Andrena Osmia

# of bees / hour bees / trap of # 0.05

0.00 0 102030405060 Utilization (%) Figure 3.5. Effect of grazing intensity on bee abundance for a) all bees in July 2007, b) Bombus in June 2007, and c) Bombus, Andrena, and Osmia in June 2008. 110

a) 18.0 16.0 14.0 12.0 10.0

8.0 Total Bee 6.0 Bombus

Species richness 4.0 2.0 0.0 0 5 10 15 20 25 30 35 Utilization (%)

b) 3.0

2.5

2.0

1.5 Bombus Andrena 1.0 Osmia Species richness 0.5

0.0 0 102030405060 Utilization (%) c) 0.8 0.7

0.6

0.5 0.4 0.3

0.2

Species richness 0.1 0 0 10203040506070 Utilization (%) Figure 3.6. Effect of grazing intensity on bee richness of a) all bees and Bombus in June 2007, b) common genera in June 2008, and c) Megachile in August 2008. 111

a) 3.5 3.0 2.5

2.0

1.5 All bees Bombus 1.0 Shannon diversity 0.5

0.0 0 5 10 15 20 25 30 35 Utilization (%) b) 3.0 2.5 2.0

1.5

1.0 All bees Bombus 0.5 Shannon diversity Shannon 0.0

-0.5 0 102030405060 Utilization (%) c) 4.0 3.5 3.0

2.5 2.0 All bees 1.5 Bombus 1.0 Shannon diversity Shannon 0.5

0.0 0 102030405060 Utilization (%) d) 4.0 3.5 3.0

2.5

2.0 All bees 1.5 Bombus 1.0

Shannon diversity 0.5

0.0 0 10203040506070 Utilization (%) e) 3.0 2.5

2.0

1.5 All bees 1.0 Bombus

diversity Shannon 0.5

0.0 0 10203040506070 Utilization (%) Figure 3.7. Effect of grazing intensity on the Shannon diversity index of all bees and Bombus for a) June 2007, b) June 2008, c) July 2007, d) July 2008, and e) August 2008. 112

a) 1.50 1.00

0.50

0.00

NMS score NMS -0.50 -1.00 -1.50 0 102030405060 Utilization (%) b) 0.12 0.10

0.08

0.06 Andrena species 1 0.04 Andrena species 12 Osmia species 1 0.02 Bombus californicus

hour / / trap bees of # 0.00

-0.02 0 102030405060 Utilization (%) c) 2.00 1.50 1.00 0.50

0.00

-0.50 NMS score -1.00 -1.50 -2.00 0 5 10 15 20 25 30 35 Utilization (%) d) 0.20

0.15

0.10

Bombus bifarius 0.05 Bombus flavifrons Bombus nevadensis 0.00

of# bees / trap /hour

-0.05 0 10203040 Utilization (%) Figure 3.8. Bee community composition a) NMS scores for axis 2 of ordination for all bees in June 2008, b) total abundance of taxa showing significant correlations with axis 2 in June 2008, c) NMS scores axis 1 ordination for Bombus for June 2007 and d) abundance of taxa showing significant relationships with axis 1in June 2007. 113

a) 0.40 0.35 0.30 0.25

0.20

0.15

0.10 0.05 # of bees / trap / hour 0.00

-0.05 0 5 10 15 20 25 30 35 Utilization (%) b) 0.20 0.18 0.16 0.14 0.12 0.10 0.08 0.06 0.04 # of bees / trap / hour 0.02 0.00 0 5 10 15 20 25 30 35 Utilization (%) c) 0.14 0.12

0.10

0.08

0.06

0.04

# of bees / trap / hour 0.02

0.00 0 5 10 15 20 25 30 35 Utilization (%) d) 0.10 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 / / hour of trap # bees 0.01 0.00 0 5 10 15 20 25 30 35 Utilization (%) Figure 3.9. Effect of grazing intensity on the abundance of the most common species in June 2007: a) Lasioglossum morphospecies 8, b) Bombus flavifrons, c) Lasioglossum sisymbrii, and d) Lasioglossum morphospecies 6.

114

2.00 a) 1.80 1.60 1.40 1.20 1.00 0.80 0.60 0.40 # of bees / trap hour / / trap of bees # 0.20 0.00 0 102030405060 Utilization (%) b) 0.70 0.60 0.50 0.40

0.30

0.20

# of/ bees# traphour / 0.10

0.00 0 102030405060 Utilization (%) c) 0.40 0.35

0.30 0.25 0.20

0.15

0.10

# of bees / traps /of/ hour bees traps# 0.05

0.00 0 102030405060 Utilization (%) d) 0.35 0.30

0.25

0.20

0.15 0.10

# of bees / hour / of trap # bees 0.05

0.00 0 102030405060 Utilization (%)

0.50 e) 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 # of bees / trap / hour 0.05 0.00 0 102030405060 Utilization (%) Figure 3.10. Effect of grazing intensity on the abundance of the most common species in July 2007: a) Bombus bifarius, b) Lasioglossum morphospecies 1, c) Bombus californicus, d) Lasioglossum morphospecies 3, and e) Bombus flavifrons.

115

0.90 a) 0.80

0.70 0.60 0.50 0.40 0.30 0.20 # of bees / trap / hour / trap of bees # 0.10

0.00 0 102030405060 Utilization (%) b) 0.40 0.35 0.30 0.25 0.20 0.15

0.10 # of bees / trap / hour 0.05 0.00 0 102030405060 Utilization (%) c) 0.35 0.30

0.25

0.20 0.15 0.10 # of# bees/ trap / hour 0.05 0.00 0 102030405060 Uilization (%) d) 0.12 0.10

0.08

0.06 0.04 0.02 # of bees / trap / hour / / trap of # bees 0.00 -0.02 0 102030405060 Utilization (%) 0.12 e) 0.10 0.08 0.06

0.04

# of bees / trap / hour 0.02 0.00 0 102030405060 Utilization (%) Figure 3.11. Effect of grazing intensity on the abundance of the most common species in June 2008: a) Lasioglossum morphospecies 1, b) Lasioglossum morphospecies 2, c) Lasioglossum morphospecies 3, d) Bombus californicus, and e) Bombus flavifrons. 116

a) 0.80 0.70 0.60 0.50 0.40 0.30 0.20

# of bees / trap / hour 0.10 0.00 0 10203040506070 Utilization (%)

b) 0.35 0.30

0.25 0.20 0.15

0.10

# of bees / / hour of trap # bees 0.05

0.00 0 10203040506070 Utilization (%) c) 0.40 0.35 0.30 0.25 0.20 0.15 0.10

# of bees / trap / hour 0.05

0.00 0 10203040506070 Utilization (%) d) 0.30 0.25

0.20

0.15

0.10

# of bees/ trap / hour 0.05

0.00 0 10203040506070 Utilization (%) Figure 3.12. Effect of grazing intensity on the abundance of the most common species in July 2008: a) Melissodes rivalis, b) Melissodes agilis, c) Bombus flavifrons, and d) Bombus californicus. 117

a) 0.30 0.25

0.20

0.15

0.10

# of# bees/ trap / hour 0.05

0.00 0 10203040506070 Utilization (%) b) 0.12 0.10

0.08

0.06

0.04

# of# bees / trap / hour 0.02

0.00 0 10203040506070 Utilization (%) c) 0.09 0.08

0.07 hour 0.06 0.05

0.04 0.03 0.02 # of bees / trap / 0.01

0.00 0 10203040506070 Utilization (%)

d) 0.06 0.05

0.04

0.03

0.02

# of bees / / hour trap of # bees 0.01

0.00 0 10203040506070 Utilization (%)

e) 0.10 0.09 0.08

hour 0.07 0.06 0.05 0.04 0.03 0.02 # of bees / trap / 0.01 0.00 0 10203040506070 Utilization (%) Figure 3.13. Effect of grazing intensity of the abundance of the most common species in August 2008: a) Lasioglossum morphospecies, b) Melissodes confusa, c) Megachile latimanus, d) Bombus californicus, and e) Lasioglossum morphospecies 5. 118

Table 3.1. Results of linear regression models of abundance (number of bees per trap per hour) with utilization for all bees and common genera for each season. Relative abundance (“Rel Abund”) of common genera (>7% relative abundance) are listed for each season.

June 2007; n=5 July 2007; n=16 June 2008; n=14 July 2008; n=16 August 2008; n=16 All Bees r2=0.27, p=0.37 r2=0.22, p=0.07 r2=0.02, p=0.65 r2=0.03, p=0.51 r2=0.01, p=0.76 Common Genera Lasioglossum Lasioglossum Lasioglossum Lasioglossum Lasioglossum Rel Abund: 49% Rel Abund: 34% Rel Abund: 58% Rel Abund: 35% Rel Abund: 43% r2=0.05, p=0.72 r2=0.04, p=0.44 r2=0.003, p=0.86 r2=0.02, p=0.61 r2=0.02, p=0.65

Bombus Bombus Bombus Melissodes Melissodes Rel Abund: 20% Rel Abund: 33% Rel Abund: 15% Rel Abund: 27% Rel Abund: 22% r2=0.85, p=0.03 r2=0.09, p=0.26 r2=0.40, p=0.02 r2=0.08, p=0.30 r2=0.004, p=0.81

Melissodes Andrena Bombus Bombus Rel Abund: 10% Rel Abund: 8% Rel Abund: 22% Rel Abund: 13% r2=0.07, p=0.34 r2=0.23, p=0.08 r2=0.06, p=0.37 r2=0.003, p=0.83

Halictus Osmia Osmia Megachile Rel Abund: 9% Rel Abund: 7% Rel Abund: 7% Rel Abund: 10% r2=0.03, p=0.56 r2=0.22, p=0.09 r2=0.01, p=0.71 r2=0.02, p=0.63

119

Table 3.2. Results of linear regression models of species richness and diversity with utilization for all bees and common genera. Total number of species for common genera (>7% relative abundance) are listed for each season.

June 2007; n=5 July 2007; n=16 June 2008; n=14 July 2008; n=16 August 2008; n=16 Total Bee Richness Species Richness: Species Richness: Species Richness: Species Richness: Species Richness: 54 183 58 92 51 r2=0.64, p=0.10 r2=0.14, p=0.15 r2=0.02, p=0.63 r2=0.001, p=0.93 r2=0.00, p=0.95 Common Genera Lasioglossum Lasioglossum Lasioglossum Lasioglossum Lasioglossum Species Richness: 19 Species Richness: 45 Species Richness: 18 Species Richness: 35 Species Richness: 26 r2=0.21, p=0.44 r2=0.05, p=0.43 r2=0.01, p=0.72 r2=0.03, p =0.53 r2=0.01, p=0.68 Bombus Bombus Osmia Osmia Bombus Species Richness: 9 Species Richness: 14 Species Richness: 11 Species Richness: 13 Species Richness: 10 r2=0.37, p=0.27 r2=0.13, p=0.18 r2=0.25, p=0.07 r2=0.01, p=0.70 r2=0.00, p=0.98 Melissodes Andrena Bombus Melissodes Species Richness: 9 Species Richness: 10 Species Richness: 10 Species Richness: 6 r2=0.002, p=0.88 r2=0.11, p=0.25 r2=0.001, p=0.91 r2=0.02, p=0.66 Halictus Bombus Melissodes Megachile Species Richness: 5 Species Richness: 9 Species Richness: 6 Species Richness: 2 r2=0.04, p=0.45 r2=0.32, p=0.04 r2=0.05, p=0.41 r2=0.27, p=0.04 Total Bee Shannon Diversity r2=0.81, p=0.04, r2=0.01, p=0.76, r2=0.26, p=0.07, r2=0.23, p=0.06, r2=0.00, p=0.97, n=5 n=16 n=13 n =16 n =16 Bombus Shannon Diversity r2=0.34, p=0.30, r2=0.01, p=0.72, r2=0.33, p=0.05, r2=0.12, p=0.19, r2=0.002, p =0.87, n=5 n =16 n =12 n =16 n=13

120

Table 3.3. Result of linear regression model of community composition with utilization for all bees and Bombus. The percent variation explained by each NMS axis is listed.

June 2007 July 2007 June 2008 July 2008 August 2008 All Bee N/A Axis 1 (7%) Axis 1 (49%) Axis 1 (59%) Axis 1 (23%) r2=0.02, p=0.58, n =16 r2=0.15, p=0.19, n =13 r2=0.05, p=0.40, n=16 r2=0.02, p=0.60, n=16 Axis 2 (87%) Axis 2 (37%) r2=0.02, p=0.66, n =16 r2=0.43, p=0.02, n =13 Axis 2 (13%) r2=0.05, p=0.39, n=16

Axis 3 (51%) r2=0.002, p=0.88, n=16

Bombus Axis 1 (78%) Axis 1 (85%) N/A Axis 1 (60%) N/A r2=0.68, p=0.09, r2=0.01, p=0.77, n=16 r2=0.02, p=0.61, n=16 n=5 Axis 2 (26%) r2=0.02, p=0.59, n=16 “N/A” means no solution was found.

121

Table 3.4. Results of linear regression models of abundance (number of bees per trap per hour) with utilization for common species for each season. Relative abundance (“Rel Abund”) of each species (>7%) are listed for each season.

June 2007; n=5 July 2007; n=16 June 2008; n=14 July 2008; n=16 August 2008; n=16 Most Common Species Lasioglossum sp8 Bombus bifarius Lasioglossum sp1 Melissodes rivalis Lasioglossum sp Rel Abund: 14% Rel Abund: 16% Rel Abund: 23% Rel Abund: 14% Rel Abund: 12% r2=0.61, p=0.12 r2=0.07, p=0.33 r2=0.002, p=0.88 r2=0.20, p=0.08 r2=0.00, p =0.95

Bombus flavifrons Lasioglossum sp1 Lasioglossum sp2 Melissodes agilis Melissodes confusa Rel Abund: 11% Rel Abund: 5% Rel Abund.: 13% Rel Abund: 9% Rel Abund: 12% r2=0.58, p=0.13 r2=0.04, p=0.46 r2=0.03, p=0.53 r2=0.02, p=0.58 r2=0.007, p=0.75

Lasioglossum Bombus californicus Lasioglossum sp3 Bombus flavifrons Megachile latimanus sisymbrii Rel Abund: 5% Rel Abund: 7% Rel Abund: 8% Rel Abund: 10% Rel Abund: 8% r2=0.06, p=0.38 r2=0.03, p=0.57 r2=0.13, p=0.18 r2=0.007, p=0.76 r2=0.58, p=0.13 Lasioglossum sp3 Bombus californicus Bombus Bombus californicus Lasioglossum sp6 Rel Abund: 5% Rel Abund: 5% californicus Rel Abund: 5% Rel Abund: 5% r2=0.18, p=0.10 r2=0.52, p=0.004 Rel Abund: 7% r2=0.05, p=0.43 r2=0.17, p=0.49 r2=0.008, p=0.75 Bombus flavifrons Bombus flavifrons Lasioglossum sp5 Rel Abund: 4% Rel Abund: 4% Rel Abund: 5% r2=0.02, p=0.64 r2=0.10, p=0.27 r2=0.06, p=0.37

122

SUMMARY

Bees provide one of the most important ecosystem services, pollination. They pollinate both natural vegetation and agricultural crops. However, there is evidence suggesting that pollinators including not only European honey bees (Apis mellifera) but also some native bees are declining in numbers. One of the potential causes for this decline is the effect of anthropogenic activities including livestock grazing, one of the most widespread land uses in western North American grasslands. Because livestock production is an important driver of rural economies, there is a great need to conduct research on sustainable grazing practices. However, previous studies on the effect of livestock grazing on native bees had experimental or geographical limitations.

In particular, there were very few studies directed at understanding the relationship between grazing intensity and native bee communities in grasslands of North America.

One important grassland type on this continent is the Pacific Northwest Bunchgrass

Prairie, a threatened and under-studied grassland in North America. Thus, the goal of this work was two-fold: 1) describe the native bee community of this unique grassland and its temporal variability, and 2) examine how livestock grazing intensity affects variables important for food and nesting for bees, and how it impacts native bee communities.

We found that the Zumwalt Prairie Preserve, a large remnant of the once extensive Pacific Northwest Bunchgrass Prairie, supported an abundant and diverse bee community. Using blue vane traps, we conducted pollinator sampling over the course of two years, and described a community with 92 species and 119 123

morphospecies in 27 genera. We also found substantial seasonal and interannual variation in the bee community. This temporal variability appeared to be driven by changes in weather and corresponding changes in floral resource phenology. In June

2008, it was colder and wetter, which caused a delay in plant phenology on the prairie by two or more weeks compared to the previous year. Bee phenology also appeared to be delayed in June and July of 2008. This may be because of a direct response of bees to the weather, a response to delayed plant phenology, or a combination of both. In conclusion, the Zumwalt Prairie was found to support a diverse bee community that varied strongly through time. This temporal variation has important implications for the timing of long-term monitoring efforts; sampling should be timed to coincide with the same point in bee community phenology each year. In addition, reporting species composition and how those patterns change through time within season and between years is vital for understanding the temporal patterns of bee communities.

We also examined whether environmental variables that were associated with native bee communities were affected by grazing intensity and how native bee communities responded to grazing intensity. Vegetation and soil characteristics were significantly affected by grazing intensity, with vegetation structure strongly reduced during both years of the study, and the abundance of blooms of two common composites decreasing with increased grazing intensity. Both soil compaction and soil stability were impacted by grazing intensity as well, with more compact soils and less surface soil stability with increased grazing intensity. Bees responded to grazing intensity, presumably in response to these alterations in environmental variables, as 124

evident by changes in a variety of metrics including diversity, richness, and community composition. Changes in total bee community diversity were not consistent with the intermediate disturbance hypothesis. Instead, diversity varied linearly with grazing intensity, with the direction of the slope dependent upon season.

We also found that different groups of bees responded to grazing intensity differently.

For example, bumble bees seemed to be sensitive to grazing intensity, showing significant decreases in abundance, richness, and diversity in one or both years of the study. In contrast, other common bees, such as those in the family Halictidae, did not show strong patterns relative to grazing intensity. Abundance of other groups, such as

Andrena, showed a positive trend with grazing intensity early in the season of one year, while other groups, such as Megachile, showed increased richness with grazing intensity late in the season.

The results of this work suggest that maintaining diverse bee communities in the Zumwalt Prairie requires a mixture of habitat with different grazing intensities.

This is most likely to provide a diversity of food and nesting resources needed by various species of bees. At a larger scale, this research may indicate that more ungrazed or lightly grazed areas need to be maintained in large expanses of rangelands in western North America to provide refuges for native bee taxa that are sensitive to grazing pressure. Further work is needed in a variety of areas, including understanding the life histories of native bees and how grazing intensity specifically affects them, and how longer-term grazing regimes and different rotation types may affect native bee communities. 125

BIBLIOGRAPHY

Allen-Wardell, G., et al. 1998. The potential consequences of pollinator declines on the conservation of biodiversity and stability of food crop yields. Conservation Biology 12:8-17.

Allison, C. D. 1985. Factors affecting forage intake by range ruminants: a review. Journal of Range Management 38:305-311.

Batra, S. W. T. 1995. Bees and pollination in our changing environment. Apidologie 26:361-370.

Belsky, A. J., and D. M. Blumenthal. 1997. Effects of livestock grazing on stand dynamics and soils in upland forests of the interior West. Conservation Biology 11:315-327.

Biesmeijer, J. C., et al. 2006. Parallel declines in pollinators and insect-pollinated plants in Britain and the Netherlands. Science 313:351-354.

Bowers, M. A. 1993. Influence of herbivorous mammals on an old-field plant community: years 1-4 after disturbance. Oikos 67:129-141.

Brennan, L. A., and W. P. Kuvlesky, Jr. 2005. North American grassland birds: an unfolding conservation crisis? Journal of Wildlife Management 69:1-13.

Brockmann, H. J. 1979. Nest-site selection in the great golden digger wasp, Sphex ichneumoneus L. (Sphecidae). Ecological Entomology 4:211-224.

Buchmann, S. L., and G. P. Nabhan. 1996. The forgotten pollinators. Island Press, Washington, D.C.

Campbell, J. W., and J. L. Hanula. 2007. Efficiency of Malaise traps and colored pan traps for collecting flower visiting insects from three forested ecosystems. Journal of Insect Conservation 11:399-408.

Cane, J. H. 1991. Soils of ground-nesting bees (Hymenoptera: Apoidea): texture, moisture, cell depth and climate. Journal of the Kansas Entomological Society 64:406-413.

Cane, J. H., R. L. Minckley, and L. J. Kervin. 2000. Sampling bees (Hymenoptera: Apiformes) for pollinator community studies: pitfalls of pan-trapping. Journal of the Kansas Entomological Society 73:225-231.

126

Cane, J. H. 2001. Habitat fragmentation and native bees: a premature verdict? Conservation Ecology 5:3. Cane, J. H., T. Griswold, and F. D. Parker. 2007. Substrates and materials used for nesting by North American Osmia bees (Hymenoptera: Apiformes: Megachilidae). Annals of the Entomological Society of America 100:350- 358.

Carvell, C. 2002. Habitat use and conservation of bumblebees (Bombus spp.) under different grassland management regimes. Biological Conservation 103:33-49.

Chao, A. 1987. Estimating the population size for capture-recapture data with unequal catchability. Biometrics 43:783-791.

Colwell, R. K. 2005. EstimateS: Statistical estimation of species richness and shared species from samples. Version 8.0. User’s guide and application. (http://purl.oclc.org/estimates).

Crumpacker, D. W. 1984. Regional riparian research and a multi-university approach to the special problem of livestock grazing in the Rocky Mountains and Great Plains. Pages 413-423 in R. E. Warner and K. M. Hendrix, editors. California riparian systems: ecology, conservation, and productive management. University of California Press, Berkeley, California.

Curry, J. P. 1994. Grassland Invertebrates: ecology, influence on soil fertility and effects of plant growth. Chapman and Hall, London.

Damiran D., T. DelCurto, E. Darambazar, P. L. Kennedy, R. V. Taylor, and A. A. Clark. 2007. Northwestern bunchgrass prairie monitoring sites databases. Circular of information No. 5, Union, Oregon, USA: Eastern Oregon Agricultural Research Center, Oregon State University.

Daubenmire, R. 1959. A Canopy-coverage method of vegetational analysis. Northwest Science 33:43-64.

Davis, J. D., S. D. Hendrix, D. M. Debinski, and C. J. Hemsley. 2008. Butterfly, bee and forb community composition and cross-taxon incongruence in tallgrass prairie fragments. Journal of Insect Conservation 12:69-79.

Day, T. A., and J. K. Detling. 1990. Grassland patch dynamics and herbivore grazing preference following urine deposition. Ecology 71:180-188.

127

DeBano, S. J. 2006a. Effects of livestock grazing on aboveground insect communities in semi-arid grasslands of southeastern Arizona. Biodiversity and Conservation 15:2547-2564.

DeBano, S. J. 2006b. The effect of livestock grazing on the rainbow grasshopper: population differences and ecological correlates. Western North American Naturalist 66:222-229.

Dingeldein, J., H. Schmalz, R. V. Taylor, V. Jansen, D. Dostert, and L. Shirley. 2009. Plant phenology monitoring on the Zumwalt Prairie Preserve 2006-2009. The Nature Conservancy. Available from http://conserveonline.org/workspaces/ZumwaltPrairieWorkspace/documents/p lant-phenology-monitoring-on-the-zumwalt-prairie/view.html (accessed Janurary 2011).

Donahue, D. L. 1999. The western range revisited: removing livestock from public lands to conserve native biodiversity. University of Oklahoma Press, Norman, Oklahoma.

Droege, S., V. J. Tepedino, G. LeBuhn, W. Link, R. L. Minckley, Q. Chen, and C. Conrad. 2010. Spatial patterns of bee captures in North American bowl trapping surveys. Insect Conservation and Diversity 3:15-23.

Fleischner, T. L. 1994. Ecological costs of livestock grazing in western North America. Conservation Biology 8:629-644.

Gess, F. W., and S. K. Gess. 1993. Effects of increasing land utilization on species representation and diversity of aculeate wasps and bees in the semi-arid areas of southern Africa. Pages 83-113 in J. La Salle and I. D. Gauld, editors. Hymenoptera and biodiversity. CAB International, Wallingford, United Kingdom.

Ghazoul, J. 2005. Buzziness as usual? Questioning the global pollination crisis. Trends in Ecology and Evolution 20:367-373.

Ghazoul, J., and L. P. Koh. 2010. Food security not (yet) threatened by declining pollination. Frontiers in Ecology and the Environment 8:9-10. Gibson C. W. D., T. A. Watt, and V. K. Brown. 1987. The use of sheep grazing to recreate species-rich grassland from abandoned arable land. Biological Conservation 42:165-183.

Goulson, D. 2003. Bumblebees behavior and ecology. Oxford University Press, United Kingdom.

128

Goulson, D., G. C. Lye, and B. Darvill. 2008. Decline and conservation of bumble bees. Annual Review of Entomology 53:191-208.

Harris, P. S. 2000. Grassland resource assessment for pastoral systems. Food and Agriculture Organization of the United Nations, Rome, Italy.

Hatfield, R. G., and G. LeBuhn. 2007. Patch and landscape factors shape community assemblage of bumble bees, Bombus spp. (Hymenoptera: Apidae), in montane meadows. Biological Conservation 139:150-158.

Herrick, J. H., W. G. Whitford, A. G. de Soyza, J. W. Van Zee, K. M. Havstad, C. A. Seybold, and M. Walton. 2001. Field soil aggregate stability kit for soil quality and rangeland health evaluations. Catena 44:27-35.

Herrick, J. E., and T. L. Jones. 2002. A dynamic cone penetrometer for measuring soil penetration resistance. Soil Science Society of America Journal 66:1320-1324. Hines, H. M., and S. D. Hendrix. 2005. Bumble bee (Hymenoptera: Apidae) diversity and abundance in tallgrass prairie patches: effects of local and landscape floral resources. Environmental Entomology 34:1477-1484.

Hitchcock, C. L., and A. Cronquist. 1973. Flora of the Pacific Northwest: an illustrated manual. University of Washington Press, Seattle, Washington.

Holzschuh, A., I. Steffan-Dewenter, D. Kleijn, and T. Tscharntke. 2007. Diversity of flower-visiting bees in cereal fields: effects of farming system, landscape composition and regional context. Journal of Applied Ecology 44:41-49.

Huntly, N. 1991. Herbivores and the dynamics of communities and ecosystems. Annual Review of Ecology and Systematics 22:477-503.

Johnson, T. M. 2010. The effects of livestock grazing intensity on grassland songbird nestling diet composition, prey selectivity, and growth rates. Dissertation, Oregon State University.

Jones, A. 2000. Effects of cattle grazing on North American arid ecosystems: a quantitative review. Western North American Naturalist 60:155-164.

Kearns, C. A., D. W. Inouye, and N. M.Waser. 1998. Endangered mutualisms: the conservation of plant-pollinator interactions. Annual Review of Ecology and Systematics 29:83-112.

129

Kearns, C. A., and D. M. Oliveras. 2009. Environmental factors affecting bee diversity in urban and remote grassland plots in Boulder, Colorado. Journal of Insect Conservation 13:655-665.

Kells, A. R., and D. Goulson. 2003. Preferred nesting sites of bumblebee queens (Hymenoptera: Apidae) in agroecosystems in the UK. Biological Conservation 109:165-174. Kennedy, P. L., S. J. DeBano, A. Bartuszevige, and A. S. Lueders. 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.

Klein, A. M., B. E. Vaissière, J. H. Cane, I. Steffan-Dewenter, S. A. Cunningham, C. Kremen, and T. Tscharntke. 2007. Importance of pollinators in changing landscapes for world crops. Proceedings of the Royal Society of London Series B-Biological Sciences 274:303-313.

Knight, R. L., W. C. Gilgert, and E. Marston. 2002. Ranching west of the 100th meridian: culture, ecology, and economics. Island Press, Washington, D.C.

Kruess, A., and T. Tscharntke. 2002. Grazing intensity and the diversity of grasshoppers, butterflies, and trap-nesting bees and wasps. Conservation Biology 16:1570-1580.

Kwaiser, K. S., and S. D. Hendrix. 2008. Diversity and abundance of bees (Hymenoptera: Apiformes) in native and ruderal grasslands of agriculturally dominated landscapes. Agriculture, Ecosystems and Environment 124:200- 204.

LaSalle, J., and I. D. Gauld. 1993. Hymenoptera: their diversity, and their impact on the diversity of other organisms. Pages 1-26 in J. La Salle and I. D. Gauld, editors. Hymenoptera and biodiversity. CAB International, Wallingford, United Kingdom.

LeBuhn, G., et al. 2003. A standardized method for monitoring bee populations – the bee inventory (BI) plot. Available from http://online.sfsu.edu/~beeplot/pdfs/Bee Plot 2003.pdf (accessed January 2011).

Leong, J. M., and R. W. Thorp. 1999. Colour-coded sampling: the pan trap colour preferences of oligolectic and nonoligolectic bees associated with a vernal pool plant. Ecological Entomology 24:329-335.

130

Lesica, P. 1993. Loss of fitness resulting from pollinator exclusion in Silene spaldingii (Caryophyllaceae). Madroňo 40:193-201.

Lesica, P. and B. Heidel. 1996. Pollination biology of Silene spaldingii. Unpublished report to Montana Field Office of The Nature Conservancy. Montana Natural Heritage Program, Helena.

Loftin, S. R., C. E. Bock, J. H. Bock, and S. L. Brantley. 2000. Desert grasslands. Pages 53-96 in R. Jemison and C. Raish, editors. Livestock management in the American Southwest: ecology, society, and economics. Elsevier, New York.

Losey, J. E., and M. Vaughan. 2006. The economic value of ecological services provided by insects. BioScience 56:311–323.

McCune, B., and J. B. Grace. 2002. Analysis of Ecological Communities. MjM Software, Gleneden Beach, Oregon.

McCune, B., and M. J. Mefford. 2006. PC-Ord, Multivariate Analysis of Ecological Data, Version 5.19. MjM Software, Gleneden Beach, Oregon.

Michener, C. D. 2007. The bees of the world. 2nd edition. Johns Hopkins University Press, Baltimore, Maryland.

Moldenke, A. R. and J. L. Neff. 1974. The bees of California: A catalogue with special relevance to pollination and ecological research. Board of Studies in Biology, University of California, Santa Cruz, California.

Moran, M. D. 2003. Arguments for rejecting the sequential Bonferroni in ecological studies. Oikos 100:403-405.

Morris, M. G. 1967. Differences between the invertebrate faunas of grazed and ungrazed chalk grassland. I. Responses of some phytophagous insects to cessation of grazing. Journal of Applied Ecology 4:459-474.

Morris, M. G. 1981. Reponses of grassland invertebrate to management by cutting. III. Adverse effects on Auchenorhyncha. Journal of Applied Ecology 18:107-123.

Morris, M. G. 2000. The effects of structure and its dynamics on the ecology and conservation of arthropods in British grasslands. Biological Conservation 95:129-142.

131

Naveh, Z., and R. H. Whittaker. 1980. Structural and floristic diversity of shrublands and woodlands in northern Israel and other Mediterranean areas. Vegetatio 41:171-190. (NOAA) National Oceanic and Atmospheric Administration. 2010. National Weather Service Office. Available from http://www.weather.gov/climate/xmacis.php?wfo=pdt (accessed December 2010).

Noss, R. F. 1994. Cows and conservation biology. Conservation Biology 8:613-616.

Noy-Meir, I. 1995. Interactive effects of fire and grazing on structure and diversity of Mediterranean grasslands. Journal of Vegetation Science 6:701-710.

(NRC) National Research Council. 2007. Status of pollinators in North America. The National Academies Press, Washington, D.C.

Parsons, C. T., P. A. Momont, T. DelCurto, M. Mcinnis, and M. L. Porath. 2003. Cattle distribution patterns and vegetation use in mountain riparian areas. Journal of Range Management 56:334-341.

Potts, S. G., and P. Willmer. 1997. Abiotic and biotic factors influencing nest-site selection by Halictus rubicundus, a ground-nesting halictine bee. Ecological Entomology 22:319-328.

Potts, S. G., and Willmer, P. 1998. Compact housing in built-up areas: spatial pattering of nests in aggregations of a ground-nesting bee. Ecological Entomology 23:427-432.

Potts, S. G., B. Vulliamy, A. Dafni, G. Ne’eman, and P. Willmer. 2003. Linking bees and flowers: how do floral communities structure pollinator communities? Ecology 84:2628-2642.

Potts, S. G., B. Vulliamy, S. Roberts, C. O’Toole, A. Dafni, G. Ne’eman, and P. Willmer. 2005. Role of nesting resources in organizing diverse bee communities in a Mediterranean landscape. Ecological Entomology 30:78-85.

Proctor, M. C. F., P. Yeo, and A. Lack. 1996. The natural history of pollination. Timber Press, Portland Oregon.

Purvis, G., and J. P. Curry. 1981. The influence of sward management on foliage arthropod communities in a ley grassland. Journal of Applied Ecology 18:711- 725.

132

Pykälä, J. 2000. Mitigating human effects on European biodiversity through traditional animal husbandry. Conservation Biology 14:705-712.

Rao S, and W. P. Stephen. 2007. Bombus (Bombus) occidentalis (Hymenoptera: Apiformes): In decline or recovery? The Pan-Pacific Entomologist 83:360-362. Rao, S., and W. P. Stephen. 2009. Bumble bee pollinators in red clover seed production. Crop Science 49:2207-2214.

Rao, S., and W. P. Stephen. 2010. Abundance and diversity of native bumble bees associated with agricultural crops: the Willamette Valley experience. Psyche 2010:1-9.

Rao, S., W. P. Stephen, C. Kimoto, and S. J. DeBano. In Press. Bombus (Bombus) occidentalis (Hymenoptera: Apiformes) in the Zumwalt Prairie. Northwest Science.

Reed, C. C. 1995. Insects surveyed on flowers in native and reconstructed prairies (Minnesota). Restoration and Management Notes 13:210-213.

Robel, R. J., J. N. Briggs, A. D. Dayton, and L. C. Hulbert. 1970. Relationships between visual obstruction measurements and weight of grassland vegetation. Journal of Range Management 23:295-297.

Roulston, T. H., S. A. Smith, and A. L. Brewster. 2007. A comparison of pan trap and intensive net sampling techniques for documenting a bee (Hymenoptera: Apiformes) fauna. Journal of the Kansas Entomological Society 80:179-181.

São Paulo Declaration on Pollinators. 1999. Report on the recommendations of the workshop on the conservation and the sustainable use of pollinators in agriculture with emphasis on bees. Brazilian Ministry of the Environment, Brasilia, Brazil.

Schmalz, H. In prep. Characterizing fine-scale soil heterogeneity to improve ecological understanding in grazing research.

Sjödin, N. E. 2007. Pollinator behavioural responses to grazing intensity. Biodiversity and Conservation 16:2103-2121.

Sjödin, N. E., J. Bengtsson and B. Ekbom. 2008. The influence of grazing intensity and landscape composition on the diversity and abundance of flower-visiting insects. Journal of Applied Ecology 45:763-772.

133

Smith, R. L., and T. M. Smith. 2000. Elements of ecology. 4th edition. Addison Wesley Longman, Benjamin/Cummings, Menlo, California.

Steffan-Dewenter, I., and T. Tscharntke. 2001. Succession of bee communities on fallows. Ecography 24:83-93.

Steffan-Dewenter, I., and T. Tscharntke. 2002. Insect communities and biotic interactions on fragmented calcareous grasslands – a mini review. Biological Conservation 104:275-284.

Steffan-Dewenter, I., and K. Leschke. 2003. Effects of habitat management on vegetation and above-ground nesting bees and wasps of orchard meadows in Central Europe. Biodiversity and Conservation 12:1953-1968.

Stephen, W. P., G. E. Bohart, and P. F. Torchio. 1969. The biology and external morphology of bees. Agricultural Experiment Station. Oregon State University, Corvallis, Oregon.

Stephen, W. P., and S. Rao. 2005. Unscented color traps for non-Apis bees (Hymenoptera: Apiformes). Journal of the Kansas Entomological Society 78:373-380.

Stephen, W. P., and S. Rao. 2007. Sampling native bees in proximity to a highly competitive food resource (Hymenoptera: Apiformes). Journal of the Kansas Entomological Society 80:369-376. Stephen, W. P., S. Rao, and L. White. 2009. Abundance, diversity and foraging contribution of bumble bees to blueberry production in Western Oregon. Acta Horticulture 810:557-562.

Sugden, E. A. 1985. Pollinators of Astragalus monoensis Barneby (Fabaceae): new host records; potential impact of sheep grazing. Great Basin Naturalist 45:299- 312. Taylor, R.V., V. Jansen, H. Schmalz, and J. Dingeldein. 2009. Mapping and monitoring Spalding’s catchfly (Silene spaldingii) on the Zumwalt Prairie Preserve 2006-2008. The Nature Conservancy. Available from http://conserveonline.org/workspaces/ZumwaltPrairieWorkspace/documents/ mapping-and-monitoring-spaldings-catchfly-silene/view.html (accessed January 2011).

Tepedino, V. J., and N. L. Stanton. 1981. Diversity and competition in bee-plant communities on short-grass prairie. Oikos 36:35-44.

134

Thomson, D. 2004. Competitive interactions between the invasive European honey bee and native bumble bees. Ecology 85:458-470.

Thorp, R. W., D. S. Horning, Jr. and L. L. Dunning. 1983. Bumble bees and cuckoo bumble bees of California (Hymenoptera: Apidae). Bulletin of the California Insect Survey 23:1-79.

Thorp, R. W. 2003. Bumble bees (Hymenoptera:Apidae): commercial use and environmental concerns. Pages 21-40 in K. Strickler and J. H. Cane, editors. For nonnative crops, whence pollinators of the future? Thomas Say Publications in Entomology: Proceedings. Entomological Society of America, Lanham, Maryland.

Thorp, R. W., and M. D. Shepherd. 2005. Profile. Subgenus Bombus Latreille 1802 (Apidae: Apinae: Bombini). Page 5 in M. D. Shepherd, D. M. Vaughan, S. H. Black, editors. Red list of pollinator insects of North America. Xerces Society for Invertebrate Conservation, Portland, Oregon.

Tisdale, E. W. 1982. Grasslands of western North America: the Pacific Northwest Bunchgrass. Pages 223-245 in A. C. Nicholson, A. McLean and T. E. Baker, editors. Grassland ecology and classification symposium proceedings. Information Services Branch B.C. Ministry of Forests, Victoria, British Columbia.

Vulliamy, B., S. G. Potts, and P. G. Willmer. 2006. The effects of cattle grazing on plant-pollinator communities in a fragmented Mediterranean landscape. Oikos 114:529-543.

Westphal, C., et al. 2008. Measuring bee diversity in different European habitats and biogeographical regions. Ecological Monographs 78:653-671.

Westrich, P. 1996. Habitat requirements of central European bees and the problems of partial habitats. Pages 1-16 in A. Matheson, S. L. Buchmann, C. O’Toole, P. Westrich and I. H. Williams, editors. The conservation of bees. Academic Press, London, United Kingdom.

Williams, N. M., R. L. Minckley, and F. A. Silveira. 2001. Variation in native bee faunas and its implications for detecting community changes. Conservation Ecology 5:7 (http://www.consecol.org/vol5/iss1/art7/).

Williams, P. 2010. Bombus (Th.) fervidus (Fabricius). Available from http://www.nhm.ac.uk/research- curation/research/projects/bombus/th.html#fervidus (accessed December 2010). 135

Winfree, R., N. M. Williams, J. Dushoff, and C. Kremen. 2007. Native bees provide insurance against ongoing honey bee losses. Ecology Letters 10:1105-1113.

Wojcik, V. A., G. W. Frankie, R. W. Thorp, and J. L. Hernandez. 2008. Seasonality in bees and their floral resource plants at a constructed urban bee habitat in Berkeley, California. Journal of the Kansas Entomological Society 81:15-28.

Wuellner, C. T. 1999. Nest site preference and success in a gregarious, ground-nesting bee Dieunomia triangulifera. Ecological Entomology 24:471-479.

Wyffels, S. A. 2009. Influence of stocking density on grazing beef cattle performance, diet composition foraging efficiency, and diet quality on a late-spring early summer native bunchgrass prairie. Master’s Thesis, Oregon State University.

Xie, Z., P. H. Williams, and Y. Tang. 2008. The effect of grazing on bumblebees in the high rangelands of the eastern Tibetan Plateau of Sichuan. Journal of Insect Coservation 12:695-703.

Yoshihara, Y., B. Chimeddorj, B. Buuveibaatar, B. Lhagvasuren, and S. Takatsuki. 2008. Effects of livestock grazing on pollination on a steppe in eastern Mongolia. Biological Conservation 141:2376-2386.