The Pennsylvania State University

The Graduate School

Department of Entomology

PESTICIDE USE, HABITAT MANIPULATION AND MANAGEMENT

CHANGES AS FACTORS IN POLLINATOR SUSTAINABILITY IN

PENNSYLVANIA APPLE ORCHARDS

A Thesis in

Entomology

Sarah Elizabeth Shugrue

Submitted in Partial Fulfillment

of the Requirements

for the Degree of

Master of Science

August 2016

ii The thesis of Sarah Shugrue was reviewed and approved* by the following:

David J. Biddinger Associate Professor of Entomology Thesis Co-Advisor

Edwin G. Rajotte Professor of Entomology Thesis Co-Advisor

Richard T. Roush Dean, College of Agricultural Sciences

Gary W. Felton Professor of Entomology Head of the Department of Entomology

*Signatures are on file in the Graduate School

iii ABSTRACT

Pollinator declines, and their impact on crop production, have become a major concern in recent years. A variety of factors have been implicated in the wild and managed bee declines including pathogens, parasites, pesticides, and habitat loss. This thesis focuses specifically on the assessment of pesticide exposure to pollinators, differences in pollinator community composition associated with orchards and floral enhancement plantings, and the interaction between pest management practices and pollinator health. Pesticide residue levels in apple pollen and nectar were measured to determine the oral exposure of bees to pesticides, and dose mortality curves depicting the oral toxicity of commonly used pesticides to honey bees were created. Furthermore, the bee communities in commercial apple orchards were compared to the bee communities in government-funded pollinator habitats.

Pesticide residue levels in the pollen and nectar were significantly lower than the application rate, and laboratory assays indicate that these levels did not pose a threat to honey bees. Additionally, the bee communities in the orchards were different from the bee communities in the managed pollinator habitat, but the pollinator habitats did not support more tree fruit pollinating or rare bee species than the orchards did. The results from these experiments should be influential in instructing new policy regarding pollinator health, specifically pesticide toxicity testing on pollinators, and the use of managed bee habitats as a conservation tool.

iv TABLE OF CONTENTS

LIST OF FIGURES ...... vii

LIST OF TABLES ...... viii

ACKNOWLEDGEMENTS ...... viiii

Chapter 1 INTRODUCTION ...... 1

Pollination Services and Their Decline ...... 1 The Role of Pesticides in Bee Decline ...... 3 The Role of Habitat in Bee Decline ...... 4 Integrated Pest and Pollinator Managment ...... 5 Literature Cited ...... 7

Chapter 2 POLLINATOR EXPOSURE TO SYSTEMIC INSECTICIDES FROM PRE- BLOOM AND LATE FALL APPLICATIONS IN PENNSYLVANIA APPLE ORCHARDS ...... 11

INTRODUCTION ...... 11 MATERIALS AND METHODS ...... 14 Experimental Design ...... 14 Prebloom Applications ...... 15 Fall Applications ...... 16 Pollen and Nectar Sampling ...... 17 Pesticide Residue Extraction ...... 18 Pesticide Residue Analysis ...... 19 Statistical Analysis ...... 19 RESULTS ...... 20 DISCUSSION ...... 22 Integrated Pest and Pollinator Managment ...... 23 Pollinator Toxicity ...... 24 Future Directions ...... 26 Literature Cited ...... 27

Chapter 3 ACUTE ORAL TOXICITY OF SEVERAL FORMULATED NEONICOTINOIDS AND RELATED INSECTICIDES USED IN APPLE ORCHARD PEST MANAGEMENT PROGRAMS TO THE HONEY BEE (APIS MELLIFERA (L.)) ... 31

INTRODUCTION ...... 31 MATERIALS AND METHODS ...... 34 Insects and Treatment ...... 34 Data Collection and Analysis ...... 36 RESULTS ...... 36 DISCUSSION ...... 38 Literature Cited ...... 41

Chapter 4 THE USE OF POLLINATOR HABITAT AUGMENTATION IN PENNSYLVANIA APPLE ORCHARDS ...... 44

INTRODUCTION ...... 44 MATERIALS AND METHODS ...... 47 Study Sites ...... 47 Sampling ...... 51 Statistical Analysis ...... 52 RESULTS ...... 54 DISCUSSION ...... 63 Bee Community Differences ...... 64 Tree Fruit Pollination Benefits ...... 65 Rare Species Conservation ...... 66 Policy Implications and Suggestions ...... 67 Literature Cited ...... 69

Chapter 2 CONCLUSIONS ...... 74

Literature Cited ...... 77

vi LIST OF FIGURES

Figure 2-1. The names and chemical structures of the six neonicotinoid insecticides registered for use on apples...... 12

Figure 3-1. Dose mortality data for (a) Assail and (b) Sivanto using mortality 48 hours post-feeding ...... 37

Figure 4-1. A proposed hierarchy of the effects of pollinator habitat in crop production (modified from Wratten et al. 2012)...... 46

Figure 4-2. Pollinator planting species mix with bloom times *Grasses, no designated bloom time...... 50

Figure 4-3. Mean Shannon Diversity Index for orchard sites (blue) versus pollinator strip sites (orange) (a) for the full set of data (b) for the known tree fruit pollinator species and (c) for specimens collected during tree fruit bloom...... 59

Figure 4-4. Individual-based rarefaction curves depicting (A) full season (B) known tree fruit pollinating and (C) bloom time species accumulation in orchard (blue) and pollinator strips (orange). 95% confidence intervals are shown for the orchards (light blue) and the pollinator strips (light orange). *Increased sampling in the orchards led to longer rarefaction curves...... 59

Figure 4-5. Proportional abundance of the ten most abundant species in the orchards and pollinator strips...... 60

Figure 4-6. Non-metric multidimensional scaling ordination of study sites and types according to bee species composition. The ordination is based on the relative Sørensen index, which separates sites based on proportional abundance. Stress= 0.1136, P=0.816 ...... 61

Figure 4-7. Non-metric multidimensional scaling ordination of study sites and types according to known tree fruit pollinating bee species composition. The ordination is based on the relative Sørensen index, which separates sites based on proportional abundance. Stress= 0.1395, P=0.001 ...... 62

Figure 4-8. RDA ordination biplot depicting associations between site types and known tree fruit pollinating species. Species names have been abbreviated, full names are in Table 4-2...... 63

Figure 4-9. A Venn diagram depicting the rare bee species found in the orchards, pollinator strips, and both...... 67

vii LIST OF TABLES

Table 2-1. Pesticide application timings, rates, and concentrations for pre-bloom apple sprays to control Rosy Apple Aphid ...... 16

Table 2-2. Fall 2013 Pesticides application dates, rates, and concentrations for BMSB control ...... 17

Table 2-3. Systemic pesticide residue levels in apple pollen and nectar sampled after spring pre-bloom sprays ...... 21

Table 2-4. Systemic pesticide residue levels in apple pollen and nectar sampled after fall sprays ...... 22

Table 3-1. Dose mortality curve slopes, LC50, and LC90 for Actara, Assail, Closer, and Sivanto using mortality 48 hours post-feeding ...... 37

Table 4-1. Site Information ...... 49

Table 4-2. Species list and abundances of bees in pollinator strips and orchards in Adams Co., PA; 2012-2014. = Known tree fruit pollinator, ◊ = Rare species ...... 54

viii

ACKNOWLEDGEMENTS

I would like to start by thanking my advisors, Dave Biddinger and Ed Rajotte, as well as

Richard Roush, for their encouragement and guidance on my unconventional path through graduate school. Similarly, gratitude is due to Chris Mullin and Julia Fine for all the technical support, and for providing me with nearly every piece of equipment used in Chapters 2 and 3.

Thank you to Tim Leslie and the crew at the Penn State Data Learning Center for the statistical support, I’d be powerless without it. I would also like to extend my gratitude to the State

Horticultural Association of Pennsylvania for getting me out of the trailer, and to the fruit growers of Adams County, for the use of their orchards. Thank you to Jim Gillis and Kelly Gill, for the technical support and crash courses in government work.

I’m forever grateful to Lindsay Carubia for jump-starting my entomology career, and

Bronwyn Bleakley for continuing to advise me long after my years at Stonehill. To every single member of the FREC family, thank you for the life lessons, laughs, and endless supply of fruit. I am so blessed to have spent the last three summers in Biglerville, and I know I’ll leave a piece of my heart here when I go. I’d especially like to thank Lolita Miller, and the Fissel family, for taking me in and turning me into an Adams County local. To Anna, Fern, and all the members of the Entomology Department- I don’t know that there is another department as supportive and compassionate as ours. I couldn’t be more proud to call you all my friends.

To Zachary Heller, for being my rock, and providing the ultimate motivation to finish my degree. And finally a huge thank you to my family, for answering every panicked phone call, for the love and laughter, and for always reminding me what I’m capable of.

CHAPTER 1

INTRODUCTION

Pollination Services and Their Decline

Pollination is the transfer of pollen from the male anther to the female stigma of the same or a different flower, enabling fertilization of the ova and formation of a seed and fruit. Although pollen can be transferred by wind, water, and animals, the majority of plants are specialized for pollination by insects (O’Toole and Raw 2004). Insect pollinators, and bees (Hymenoptera: Apoidea) in particular, are essential, as they support the reproduction of nearly 85% of the world’s flowering plants and 35% of global crop production (Hopwood et al. 2012, Ollerton et al. 2011, Klein et al. 2007). Of all the crops grown in Pennsylvania, apples are among the most valuable, at nearly $75 million in

2010 (USDA 2011). Pennsylvania produces 400-500 million pounds of apples per year, and ranks fourth in the nation for apple production. The majority of that production occurs in Adams County, where the Pennsylvania State University Fruit Research and

Extension Center is located, and the research for this thesis was centered (Pennsylvania

State University 2016). Most apple cultivars require cross pollination to ensure commercial quality fruit and yields, (Pennsylvania State University 2016), making insect pollination an important part of apple production.

Honey bees are the most widely used bee species for apple pollination because of their ability to find flowers, recruit their nest mates to the floral resources, and they are

2 relatively easy to maintain and transport. Despite successful use of the honey bee for many years, the reliance on this single species for commercial pollination has become threatened by high rates of colony loss over the last decade (Lee et al. 2015, Potts et al.

2010b). Alternatively, other bee species can be reliable pollinators. Wild and other social species can be effective, however, declines of wild bee populations, although more difficult to measure, also raise concerns regarding pollination deficits in the future

(Bartomeus et al. 2013, Burkle et al. 2013). In 2009, the economic benefits from native bee pollination in the United States were valued at more than 9 billion dollars (Pollinator

Health Task Force 2014). A recent survey of Pennsylvania and New York apple growers indicated that most growers have been relying on wild pollinators rather than honey bees, and in PA some large farm operations have not been using honey bees for over 30 years, with no discernable decline in fruit quality or yield (Joshi et al. 2011, Park et al. 2010).

Despite often going unnoticed, pollination by wild bees contributes greatly in the production of many crops and declines in wild bee populations would have a negative impact.

Research on bee declines suggests that they are most likely caused by the exposure to multiple interacting stressors such as pesticide use, pathogens, parasites, and a reduction in appropriate floral and nesting resources (Goulson et al. 2015, Potts et al. 2010a). This thesis focuses specifically on the assessment of pesticides exposure to pollinators, differences in pollinator community composition associated with orchards and floral enhancement plantings, and the interaction between pest management practices and pollinator health, which underpins the emerging concept of Integrated Pest and Pollinator

3 Management as a new paradigm within Integrated Pest Managment (Biddinger and

Rajotte 2015).

The Role of Pesticides in Bee Decline

The consequences of the exposure of pollinators to pesticides has received a disproportionate amount of media attention due to the presence of agricultural chemicals in honey bee wax and pollen (Mullin et al. 2010), despite the lack of evidence that they are the main cause of CCD (Cresswell 2011). One class of pesticides, known as neonicotinoids, have been implicated in increased mortality of honey bee colonies

(Maxim and van der Sluis 2010) and have become the subject of much debate, leading to public-driven reactions such as the moratorium on neonicotinoid use in Europe (Gross

2013).

Neonicotinoids are synthetic nicotine-related insecticides that act as agonists at nicotinic acetylcholine receptors, blocking the transmission of nerve impulses in insects’ central nervous system (Matsuda et al. 2001, Tomizawa and Casida 2003). They were developed as a more human- and environmentally-safe replacement for the vertebrate- toxic organophosphate and carbamate insecticides (Namba et al. 1971, Bardin et al. 1994) that were being phased out due to the Food Quality Protection Act of 1996.

Although much research has been conducted on the effects of neonicotinoids on bee health (Creswell 2011), few studies have moved beyond the laboratory to address the effects in a field realistic manner (Carreck and Ratnieks 2014). For example, a single

4 laboratory study examining the synergism of neonicotinoids with fungicides concluded that synergism of over 1,000-fold could occur, but in the field trials the findings could not be reproduced (Iwasa et al. 2004). A more recent paper examining the synergism of similar neonicotionids and fungicides used in apple orchards on the honey bee and a

Mason bee found synergism to be non-significant in most treatments and barely significant at less than 10-fold for one product when formulated product was used

(Biddinger et al. 2013).

Federal safety testing for pesticide registration requires risk to pollinators be calculated through short-term laboratory studies on only honey bees, using the active ingredient of the pesticide dissolved in acetone rather than the formulated product in water (Office of Pesticide Programs 2013). To mitigate the impacts of pesticides on pollinator health, we have to understand the risks involved in the most field realistic manner, and consider the effects on all bee species through time and at a community level.

The Role of Habitat in Bee Decline

Loss of the habitat that provides food and nesting resources for pollinators is suspected as a long-term contributor to bee declines (Potts et al. 2010a). Although many bee species may respond favorably to moderate landscape disturbances (Winfree et al.

2007), extreme urbanization and monoculture cropping are generally unfavorable to managed and wild bees (Bates et al. 2011). Although not well understood, climate change could lead to range shifts or constrictions (Kerr et al. 2015), and temporal separation of

5 pollinators and the plants they pollinate. Herbicides are used regularly to control weeds in most cropping systems, and this may reduce the availability of season-long flowers for pollinators (Goulson et al. 2008). Floral and nesting resources are required for bees to reproduce and survive, so the reduction of these resources can have a negative impact on bee populations.

The importance of a rich and robust pollinator community to crop production, and the threat of reduced floral resource and habitat to this community, is recognized in the

National Strategy to Promote the Health of Honey Bees and Other Pollinators. In that report, improving habitat acreage is one of the three main goals, with the aim being to restore or enhance 7 million acres of land for pollinators by 2020 (Pollinator Health Task

Force 2015). Much of this effort is coming in the form of managed pollinator hedgerows, despite little research on the effectiveness of these habitats.

Integrated Pest and Pollinator Management

The economic impacts of pollinator shortages on US specialty crops such as fruit, vegetables and nuts could be considerable. Inadequate pollination can reduce crop quality and yield. In apple or pear, pollination efficiency affects seed set which in turn affects size and quality and hence the profitability to growers (Kevan and Viana 2003). The US almond crop alone requires almost $300 million in pollination services (Bond et al.

2014). Because demand for honey bees is so high, there has been a dramatic increase in bee colony rental costs – from $35/hive for Pennsylvania apple growers in 2006 to over

$100/hive currently (Biddinger, unpublished). Rising pollination costs combined with

6 declining yields from pollinator shortages would lead to an increase in the prices of nuts, fruits and vegetables, and increased imports of these commodities from other countries.

Luckily, there is already a paradigm in place, which can be used to protect both the honey bees and wild pollinators in agricultural habitats. Biddinger and Rajotte (2015) theorized that IPM, which is a decision making process that relies on the use of a variety of tactics to protect crops while minimizing the economic, health, and environmental impacts, can be adjusted to include pollinators as a protected natural resource. To do this, we have to have a better understanding of the threats to pollinator health, and we must consider them in the greater context of crop production.

This thesis focuses on two threats to pollinator health, pesticide exposure and habitat loss, and how these are impacting pollination in Pennsylvania apple orchards.

Chapter 2 is an investigation into the major route of exposure of bees to systemic pesticides, contaminated pollen and nectar. Chapter 3 explores how consumption of different levels of these pesticides impact pollinator health. In Chapter 4, the focus shifts to managed pollinator habitats and their value as a conservation and pollination supplementation tool. Future directions and policy suggestions to combat these threats to pollinator health are included in Chapter 5.

7 Literature Cited

Bardin, P. G., S. F. van Eeden, J. A. Moolmam, A. P. Fodon, and J. R. Joubert. 1994. Organophosphate and carbamate poisoning. Archives of Internal Medicine 154(13): 1433-1441.

Bartomeus, I., J. S. Ascher, J. Gibbs, B. N. Danforth, D. L. Wagner, S. M. Hedtke, and R. Winfree. 2013. Historical changes in northeastern US bee pollinators related to shared ecological traits. Proceedings of the National Academy of Sciences, 110(12): 4656–4660.

Bates, A.J., J. P. Sadler, A. J. Fairbrass, S. J. Falk, J. D. Hale, and T. J. Matthews. 2011. Changing bee and hoverfly pollinator assemblages along an urban-rural gradient. PLoS ONE 6: e23459.

Biddinger, D.J., and E. G. Rajotte. 2015. Integrated pest and pollinator management- adding a new dimension to an accepted paradigm. Current Opinion in Insect Science 10: 204-209.

Biddinger, D. J., J. L. Robertson, C. Mullin, J. Frazier, S. A. Ashcraft, E. G. Rajotte, N. K. Joshi, and M. Vaughn. 2013. Comparative toxicities and synergism of apple orchard pesticides to Apis mellifera (L.) and Osmia cornifrons (Radoszkowski). PLoS ONE 8(9): e72587.

Bond, J., K. Plattner, and K. Hunt. 2014. Fruit and Tree Nuts Outlook: Economic Insight. US Pollination-Services Market. USDA Economic Research Service Situation and Outlook FTS-357SA.

Burkle, L.A., J. C. Marlin, and T. M. Knight. 2013. Plant–pollinator interactions over 120 years: loss of species, co-occurrence, and function. Science, 339:1611-1615.

Carreck, N. L., and F. L. W. Ratnieks. 2014. The dose makes the poison: have “field realistic” rates of exposure of bees to neonicotinoid insecticides been overestimated in laboratory studies? Journal of Apicultural Research 52(5): 607- 614.

Creswell, J. E. 2011. A meta-analysis of experiments testing the effects of a neonicotinoid insecticide (imidacloprid) on honey bees. Ecotoxicology 20(1): 149-157.

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

8 Goulson D., E. Nicholls, C. Botias, and E. L. Rotheray. 2015. Bee declines driven by combined stress from parasites, pesticides, and lack of flowers. Science 347(6229): 1255957-1 - 1255957-9.

Gross, M. 2013. EU ban puts spotlight on complex effects of neonicotinoids. Current Biology, 23(11), R462-R464.

Hopwood, J., M. Vaughn, M. Shepherd, D. J. Biddinger, E. Mader, S. H. Black, and C. Mazzacano. 2012. Are Neonicotinoids Killing Bees? A review of research into the effects of neonicotinoid insecticides on bees, with recommendations for action. Xerces Society for Invertebrate Conservation, USA.

Iwasa, T., N. Motoyama, J. T. Ambrose, and R. M. Roe. 2004. Mechanism for the differential toxicity of neonicotinoid insecticides in the honey bee, Apis mellifera. Crop Protection 23: 371–378.

Joshi, N. K., D. J. Biddinger, and E. G. Rajotte. 2011. “A survey of apple pollination practices, knowledge and attitudes of fruit growers in Pennsylvania,” in 10th International Pollination Symposium (Puebla, Mexico).

Kerr, J. T., A. Pindar, P. Galpern, L. Packer, S. G. Potts, S. M. Roberts, P. Rasmont, O. Schweiger et al. 2015. Climate change impacts bumblebees coverage across continents. Science 349 (6244): 177-180.

Kevan, P. G., and B. F. Viana. 2003. The global decline of pollination services. Biodiversity 4(4), 3-8.

Klein, A. M., B. E. Vaissiere, 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 B: Biological Sciences, 274(1608), 303-313.

Lee, K. V., N. Steinhauer, K. Rennich, M. E. Wilson, D. R. Tarpy, D. M. Caron, et al. 2015. A national survey of managed honey bee 2013-2014 annual colony losses in the USA. Apidologie 46(3):292–305.

Matsuda, K., S. D. Buckingham, D. Kleier, J. J. Rauh, M. Grauso, and D. B. Sattelle. 2001. Neonicotinoids: insecticides acting on insect nicotinic acetylcholine receptors. Trends in Pharmacological Sciences 22(11): 573-580.

Maxim, L., and J. P. van der Sluis. 2010. Expert explanations of honeybee losses in areas of extensive agriculture in France: Gaucho® compared with other supposed causal factors. Environmental Research Letters 5(1): 014006.

9 Mullin, C. A., M. Frazier, J. L. Frazier, S. Ashcraft, R. Simonds, and J. S. Pettis. 2010. High levels of miticides and agrochemicals in North American apiaries: implications for honey bee health. PLoS ONE, 5(3), e9754.

Namba T., C. T. Nolte, J. Jackrel, and D. Grob. 1971. Poisoning due to organophosphate insecticides: Acute and chronic manifestations. The American Journal of Medicine 50(4): 475-492.

Office of Pesticide Programs (United States Environmental Protection Agency). 2013. Guiding Principles for Data Requirements. Available at: https://www.epa.gov/sites/production/files/2016-01/documents/data-require- guide- principle.pdf

Ollerton, J., R. Winfree, and S. Tarrant. 2011. How many flowering plants are pollinated by animals? Oikos 120:321–326.

O’Toole, C., and A. Raw. 2004. Bees of the World. Facts on File, New York, NY.

Park, M., M. Orr, and B. Danforth. New York Fruit Quarterly 18: 21-25 (New York State Horticultural Society, Geneva, NY, 2010).

Pennsylvania State University. 2016. Penn State Tree Fruit Production Guide 2015-2016. Pennsylvania State University.

Pollinator Health Task Force. 2014. National strategy to promote the health of honey bees and other pollinators (The White House, Washington, DC).

Potts, S. G., J. C. Biesmeijer, C. Kremen, P. Neumann, O. Schweiger, and W. E. Kunin. 2010a. Global pollinator declines: Trends, impacts and drivers. Trends in Ecology and Evolution 25: 345–353.

Potts, S. G., S. P. M. Roberts, R. Dean, G. Marris, M. A. Brown, R. Jones, P. Neumann, and J. Settele. 2010b. Declines of managed honey bees and beekeepers in Europe. Journal of Apiculture Research 49(1):15–22.

Tomizawa, M., and J. E. Casida. 2003. Selective toxicity of neonicotinoids attributable to specificity of insect and mammalian nicotinic receptors. Annual Review of Entomology 48: 339-364.

USDA (United States Department of Agriculture). 2011. Pennsylvania Agricultural Statistics 2010-2011. Available at: https://www.nass.usda.gov/Statistics_by_State/Pennsylvania/Publications/Annual _Statistical_Bulletin/2010_2011/Fruit_coes-apple.pdf

Winfree, R., N. M. Williams, J. Dushoff, and C. Kremen. 2007. Native bees provide insurance against ongoing honey bee losses. Ecology Letters 10(11), 1105-1113.

CHAPTER 2

POLLINATOR EXPOSURE TO SYSTEMIC INSECTICIDES FROM PRE-BLOOM AND LATE FALL APPLICATIONS IN PENNSYLVANIA APPLE ORCHARDS

INTRODUCTION

The effect of pesticides on pollinator decline has been a widely debated topic in recent years, but little has been done to understand the true field impacts and how to mitigate them. In particular, systemic pesticides that are water soluble and transported through the plant tissues are of concern because insects can be exposed through ingestion, even though direct contact was avoided. One group of systemic pesticides known as neonicotinoids are thought to harm pollinators due to their widespread use in many agricultural crops and increased mortality of honey bee colonies in areas treated with neonicotinoids (Maxim and van der Sluis, 2010, van Engelsdorp and Meixner 2010,

Hopwood et al. 2012).

They were developed as a more consumer- and environmental-safe replacement for the mammalian-toxic organophosphate and carbamate insecticides (Namba et al. 1971,

Bardin et al. 1994) that were being phased out due to the Food Quality Protection Act of

12 1996. Neonicotinoids were developed in the 1990’s and integrated in to tree fruit IPM over the following decade (Agnello et al. 2009).

Systemic pesticides such as neonicotinoids (Figure 2-1) allow protection against many sap feeding arthropod pests. Neonicotinoids can be applied to the root as a seed coating, through pre-plant soil drenches prior to planting, through chemigation, tree trunk injections, or in most fruits and vegetables, as foliar sprays. Their variety of uses and their importance in controlling insecticide resistant pests have made them the most widely used group of insecticides in the United States and worldwide (Simon-Delso et al.

2014).

The use of neonicotinoids (IRAC Code 4A) and neonicotinoid-related (IRAC

Code 4C, 4D, etc.) systemic pesticides has increased dramatically since their introduction

(USGS Pesticide Synthesis Project 2013). In Eastern US apple orchards, this includes multiple applications in the fall for the invasive Brown Marmorated Stink Bug

13 (Halyomorpha halys) (Stål) (Hemiptera: Pentatomidae), a single application in the spring prior to bloom for the organophosphate-, carbamate-, and pyrethroid-resistant Rosy

Apple Aphid (Dysaphis plantaginea) (Passerini) (Hemiptera: Aphididae), and post-bloom summer applications for various leafhopper and aphid pests which are also now resistant to organophosphates (Pennsylvania State University 2016). Flonicamid, a novel non- neonicotinoid insecticide (IRAC Code 29), has also been tested for control of pesticide resistant Rosy Apple Aphid. It is also plant systemic, but thought to be safe to bees

(European Food Safety Authority 2010).

To varying degrees, surface residues from foliar applications of plant systemic pesticides are quickly absorbed into the plant, leaving little contact activity for both pests and beneficial insects such as pollinators (Hopwood et al. 2012). However, plant systemic pesticides, including neonicotinoids, can be translocated and absorbed into pollen and nectar (Yañez et al. 2014, Stoner and Eitzer 2012, Dively and Kamel 2012).

Much research has been done on the toxicity of these chemicals (Hopwood et al. 2012,

Biddinger et al. 2013, Blacquiere et al. 2012, Creswell 2011), but little is known about what non-target insects, and bees in particular, are actually exposed to in the field

(Carreck and Ratnieks 2014). The levels of neonicotinoid and other systemic pesticide residues in the pollen and nectar that are consumed by bees have been inadequately studied. Given the importance of pollinators in agriculture, it is essential to understand which pesticides pollinators are being exposed to, as well as the routes, levels, and duration of exposure.

In this study, we aim to identify the realistic exposure of bees to systemic pesticides in Pennsylvania apple orchards by addressing the question: What levels of the

14 systemic pesticides commonly used in apples are found in the pollen and nectar of apple trees (Malus domestica) after spring and fall pesticide applications? We also investigate the possibility of changing the timing of applications to reduce pollinator pesticide exposure while still maintaining commercial levels of pest control. Apple pollen and nectar samples were collected from trees treated with known amounts of systemic insecticides and analyzed for pesticide residues. Insecticide applications simulated commonly used pest control programs for the fall pre-harvest control of the Brown

Marmorated Stink Bug and spring pre-bloom applications for Rosy Apple Aphid control

(Pennsylvania State University 2016).

We hypothesized that pesticide residues could be present in the nectar and/or pollen in varying amounts. Higher residues should be present in the pollen and nectar when treatments are applied closer to flowering, because the levels of pesticide residues will decrease over time. Therefore, residues from fall sprays should be lower. This information would us to determine the possible routes and levels of exposure of systemic pesticides to pollinators, and to make recommendations to fruit growers that would minimize pollinator exposure and still maintain acceptable pest control by simply adjusting the timing and choice of the spring insecticide applications.

MATERIALS AND METHODS

Experimental Design

Three separate field studies were conducted at the Penn State Fruit Research and

Extension Center in Adams County, PA. In a complete randomized design, trees were

15 treated with a specific, formulated, commercially available pesticide, with a Durand-

Wayland airblast sprayer, calibrated to deliver 937 l/ha at 3.9 km/h. Trees in each study were 20 years old with average canopy and were planted to a spacing of 6 x 9 meters. No other insecticides were applied to the trees during this study other than their designated treatments. Pollen and nectar were collected and analyzed for pesticide residues.

Pre-bloom Applications

Pre-bloom application experiments took place in 0.9-hectare plot of Golden

Delicious apple trees (Malus domestica 'Gibson Golden'). In the spring of 2013 pesticides were applied at the commercially recommended timing of the pink stage (approximately

5-7 days before bloom), to control rosy apple aphid (Pennsylvania State University

2016). Trees were treated with either acetamiprid, sulfoxaflor, thiacloprid, thiamethoxan, imidicloprid, or unsprayed as a control, five trees per treatment (Table 2-1). Pollen and nectar samples were collected five days later, during peak bloom.

In 2014, the effect of application timing was tested. Pesticide treatments were paired with a single application at half-inch green stage on April 17th, and compared with an application at the pink stage on April 28th. Trees were treated with either acetamiprid, sulfoxaflor, flupyradifurone, or flonicamid, four trees per phenology stage, per treatment.

Again, four untreated trees were sampled as a control (Table 2-1). Pollen and nectar samples of all treatments were collected at peak bloom on May 8th.

Table 2-1. Pesticide application timings, rates, and concentrations for pre-bloom apple sprays to control Rosy Apple Aphid Product Formulation Application Rate Application (g AI/ha) Concentration (PPM) Spring 2013 Assail 30SG 30% Acetamiprid 126.0 134 Closer 240SC 21.8% Sulfoxaflor 52.9 94 Calypso 480SC 480 g AI/L 210.8 150 Thiacloprid Actara 25WP 25% Thiamethoxam 96.4 103

Provado 1.6 EC 192 g AI/L 84.3 60 Imidacloprid Spring 2014 Assail 30SG 30% Acetamiprid 311.4 134 Beleaf 50SG 50% Flonicamid 242.4 105 Closer 240SC 240 g AI/L 130.7 104 Sulfoxaflor Sivanto 200SL 200 g AI/L 153.7 164 Flupyradifurone

Fall Applications

Fall pesticide application experiments took place in an adjacent 0.9-hectar plot of

Fuji apple trees (Malus domestica ‘Fuji’). Pesticide applications took place in August or

September, simulating the Brown Marmorated Stink Bug pesticide application. Trees were treated with thiamethoxam, dinotefuran, clothianidin, sulfoxaflor, imidacloprid, or

17 an unsprayed control, five trees per treatment (Table 2-2). Pollen and nectar samples

were collected during the following spring, at peak bloom.

Table 2-2. Fall 2013 Pesticides application dates, rates, and concentrations for BMSB control Product Timing Formulation Application Application Rate (g Concentration AI/ha) (PPM) Endigo ZC 8/26 141.1 g AI/L Thiamethoxam 62.3 67

Leverage 9/5 239.7 g AI/L Imidacloprid 49.2 53 Scorpion 8/29 388.3 g AI/L Dinotefuran 340.8 365 35SL 9/5 340.8 365 9/12 340.8 365 9/19 340.8 365 Belay 2.13SC 9/12 255.3 g AI/L Clothianidin 112.2 120 9/19 112.2 120 Closer 240SC 8/29 240 g AI/L Sulfoxaflor 96.6 104 9/5 96.6 104

Pollen and Nectar Sampling

Pollen and nectar were collected from 50 blossoms per replicate tree, and

combined as one sample. Nectar was extracted from the nectary via capillary action using

a one-microliter micro capillary tubes and anthers were combed from the stamens using

eyelash combs.

18 Pesticide Residue Extraction

Samples were analyzed using a QuEChERS procedure. Pollen (50 flowers equivalent, range of mg) was weighed into a 50 mL plastic centrifuge tube. Two ml of water and 2 ml of acetonitrile with internal standard (IS; 2.5 ppm capramide) were added and vortexed for 30 seconds. Next, 0.8 g of anhydrous magnesium sulfate and 0.2 g of anhydrous sodium acetate were added and vortexed for 30 seconds. The tube was placed on ice for 5 minutes, then centrifuged at 5000 rpm for 10 minutes at 25 ◦C. One ml of the supernatant was transferred into a 2 ml dispersed solid phase extraction (SPE) tube (UCT,

Bristol, PA, USA) containing 0.15 grams of CUPSA, 0.05 grams of CEC18 and 0.15 grams of magnesium sulfate, and then vortexed for 30 seconds. The tube was centrifuged again at 5000 rpm for 10 minutes at 25 ◦C. The supernatant was transferred to a 1.5 ml vile and dried under air, then reconstituted with 1/10 volume acetonitrile/water (50/50).

The sample was centrifuged at 10000 rpm for ten minutes and the supernatant was transferred to an auto-sampler vial for LC-MS analysis.

Capillary tubes (50 per sample) containing nectar were crushed in a tissue grinder, and 1 ml of acetonitrile/water (20/80) was added containing 2.5 ppm IS. The samples were sonicated for 30 minutes, then the supernatant was transferred and centrifuged at

10000 rpm for 10 minutes. Five hundred µl of supernatant was transferred and dried under air, then reconstituted with 50 µl acetonitrile/water (20/80). This was centrifuged at

10000 rpm for ten minutes and the supernatant was transferred to an auto-sampler vial for analysis.

19 Pesticide Residue Analysis

The liquid chromatography-mass spectrometry (LC-MS) separation was performed on a 2.0×100 mm XR-ODS column (Shimadzu, Kyoto, Japan), maintained at

50 ◦C, A 10 µl injection of sample was made with a solvent flow rate of 0.35 ml/min.

Mobile phases A and B were water and acetonitrile (containing 2.5% water) respectively, both containing 2 mM of ammonium formate and 0.01% formic acid. The solvent gradient started at 20% B, changed to 65% B after 7 minutes, and to 100 % B at 9 minutes and kept for 2 minutes. Then, the column was equilibrated back to 20% B from

11 to 15 minutes in the run.

The LC-MS 2020 system was equipped with an electrospray ionization (EIS) source, and operated in positive ion mode. Interface voltage was 4.5 kV. The flow rate of the nebulizing gas (nitrogen) and drying gas (nitrogen) were 0.5 l/minute and 15 l/minute.

Desorption line (DL) and heat block temperature were 250 ◦C and 200 ◦C respectively.

Detector voltage was -1.1 kV. The event time was set to 2 seconds, with micro scan at

0.5u. Mass ions at m/z 223 (acetamiprid), m/z 250 (clothianidin), m/z 203 (dinotefuran), m/z 230 (flonicamide), m/z 289 (flupyradifurone), m/z 278 (sulfoxaflor), m/z 292

(thiamethoxam), and m/z 213.1 (IS) were monitored for quantitation

Statistical Analysis

Pollen and nectar residue levels were averaged across treatments. Mean residue levels in the pollen and nectar and standard error of the means were calculated for each pesticide. Zero PPM was used during calculations for any residues that were below the detection threshold.

20 RESULTS

Control treatments from all three sampling periods were found to have no detectable pesticide residues. Limit of detection and pollen and nectar residue levels are in Table 2-3 for pre-bloom applications, and Table 2-4 for fall applications.

21 Table 2-3. Systemic pesticide residue levels in apple pollen and nectar sampled after spring pre-bloom sprays Treatment Phenology at time Limit of Level in Level in of application Detection Pollen (PPM) Nectar (PPM) (PPM) ± SE ± SE Spring 2013 Acetamiprid Pink 0.025 0.0681 ± 0.0683 ± 0.0049 0.0133 Sulfoxaflor Pink 0.011 0.0797 ± Not Detected 0.0181 Thiacloprid Pink 0.015 0.0097 ± 0.0023 ± 0.0022 0.0003 Thiamethoxam Pink 0.015 0.0485 ± 0.0177 ± 0.0029 0.0012 Imidacloprid Pink 0.015 0.0070 ± 0.0033 ± 0.0020 0.0003 Spring 2014 Acetamiprid Pink 0.010 0.0177 ± 0.0058 ± 0.0005 0.0007 Acetamiprid Half-Inch Green 0.010 0.0109 ± Not Detected 0.0025 Flonicamid Pink 0.050 0.5124 ± Not Detected 0.0745 Flonicamid Half-Inch Green 0.050 0.2141 ± Not Detected 0.0265 Sulfoxaflor Pink 0.011 0.0444 ± Not Detected 0.0043 Sulfoxaflor Half-Inch Green 0.011 Not Detected Not Detected

Flupyradifurone Pink 0.009 0.0430 ± 0.0115 ± 0.0078 0.0020 Flupyradifurone Half-Inch Green 0.009 0.0219 ± Not Detected 0.0028

22 Table 2-4. Systemic pesticide residue levels in apple pollen and nectar sampled after fall sprays Treatment Phenology at time Limit of Level in Level in of application Detection Pollen (PPM) Nectar (PPM) (PPM) Thiamethoxam N/A 0.015 Not Detected Not Detected Dinotefuran N/A 0.010 Not Detected Not Detected Clothianidin N/A 0.010 Not Detected Not Detected Sulfoxaflor N/A 0.011 Not Detected Not Detected Imidacloprid N/A 0.015 Not Detected Not Detected

DISCUSSION

All pesticide residue levels were well below the application concentration. This was expected, as the pesticides are applied over large areas and degrade over time.

Residue levels from trees treated at half-inch green stage were consistently lower than those treated at pink stage. No pesticide residues were detected from fall pesticide applications.

We found pesticide residues from all spring applications to be much lower than the application concentration. For example, acetamiprid was applied in the spring of 2013 at a rate of 134 PPM, which would be the estimated contact exposure of a pollinator to this pesticide if the application were made during the day while the pollinator was foraging. However, it was detected in the pollen and nectar at a mere 0.0681 and 0.0683

PPM respectively, which would be the estimated oral exposure of a pollinator to this pesticide. The lethal concentration of acetamiprid needed to kill 50% of a population

(LC50) of honey bees is 1450 PPM (Hopwood et al. 2012). This value is more than

10,000 fold higher than the levels detected in the pollen and nectar, leading us to

23 conclude that consumption of pollen and nectar tainted with systemic insecticides poses little threat to pollinators.

No pesticide residues were detected in the pollen or nectar from flowers in the spring following stink bug applications the previous fall. This was a concern because neonicotinoids have been found to carry over from season to season in seed and soil applications (Hopwood et al. 2012, Goulson 2013). The Brown Marmorated Stink Bug

(BMSB) is a relatively new pest in the United States, caused $37 Million in losses to

Mid-Atlantic apple growers in 2010 (US Apple Association 2011). It is difficult to control with most insecticides because it continues to immigrate from the habitat adjacent to fruit orchards, requiring multiple applications of insecticide. One particular neonicotinoid, dinotefuron, has been given an emergency label in tree fruit specifically for BMSB control in several mid-Atlantic states. It is vital to continue using all the management techniques we can, including neonicotinoids, to control this pest. We can confirm that concern over the impacts of fall BMSB pesticide applications on bees appears unnecessary, as no detectable pesticide residues were present in the pollen and nectar the following spring.

Integrated Pest and Pollinator Management

The data support our hypothesis that it is possible to reduce pollinator pesticide exposures by adjusting pest management practices. For rosy apple aphid management in the spring, shifting the pesticide application from pink stage to half-inch green stage, 11 days earlier, reduced the pesticide residue levels in the pollen and nectar for every insecticide tested, and in some cases lowered residues to non-detectable levels. While

24 pollinator exposure was decreased by shifting the application timing, rosy apple aphid control was not compromised (DJB unpublished data). This simple management adjustment can reduce the exposure of pollinators to systemic pesticides while still controlling the pest population, and it is now recommended to all fruit growers in the mid-Atlantic states (Pennsylvania State University 2016). Rather than advocating for the complete ban of all neonicotinoids, we propose retaining the use of the least bee toxic compounds for certain key pests until alternative pesticides, such as flonicamid, are further developed.

This study supports the idea that pollinator protection can be incorporated into integrated pest management (IPM) practices, forming the new practice of integrated pest and pollinator management (IPPM) (Biddinger and Rajotte 2015). They suggest that pollinator health and conservation can be included into the already successful IPM paradigm in the same way that other beneficial insects such as predators and parasitoids are conserved. Presently, fruit growers reduce the pesticide exposure to predatory insects by making small shifts in pesticide product selection and application timing.

Incorporating pollinators into the IPM system will allow fruit growers to protect the valuable pollination resources while still managing pest pressure.

Pollinator Toxicity

Establishing to which pesticides the pollinators are actually exposed to and the concentration and route of exposure, as we did here, is essential in determining the field realistic toxicities of pesticides to pollinators and methods to mitigate them. Although systemic pesticide residues have been found to be variable (Blacquiere et al. 2012), many

25 laboratory studies use doses that are beyond the range of pesticide residues found in the field, and therefore represent a “worst case” scenario (Carreck and Ratnieks 2014). In order to determine the actual exposure of pollinators to pesticides, researchers must consider this realistic dose, as well as a variety of other factors. Multiple types of bioassays must be conducted to consider all routes of exposure, as bees can encounter systemic pesticides in a variety of ways (Krupke et al. 2012, Hopwood et al. 2012). For apple pollinators in the Eastern United States consumption of pollen and nectar contaminated by systemic pesticides will be the most pressing, as insecticides are not applied during the short 7 to 10-day apple bloom period, and bees do not normally contact foliage, which virtually eliminates contact toxicity. Differential toxicity depending on life stage has been shown in many insects (Biddinger et al. 1998, Banken and Stark 1997, Yates and Sherman 1970) and therefore must be considered. Although the residue levels we detected are relatively low compared to what many researchers are using to understand neonicotinoid toxicity to pollinators, they only represent an acute dose, which is not what pollinators experience in the field. Long term studies of chronic exposure provide a more accurate understanding of exposure, because the length of exposure has been shown to produce differential toxicities in bees (Suchail et al. 2001), and sublethal effects of systemic pesticides on reproductive, nesting, and foraging behavior (Dively et al. 2015, Artz and Pitts-Singer 2015) are not likely observed in acute studies. Lastly, using formulated product rather than active ingredient is essential, as it best mimics pesticide pharmacology in the field.

26 Future Directions

Interestingly, some pesticides (flonicamid, sulfoxaflor) were detected in the pollen but not in the nectar. Because systemic pesticides are a relatively new field of research, little is known about their movement within the plant tissues. To better understand the movement of systemic pesticides to pollen and nectar, we need to understand the source of pollen and nectar production. Nectar is believed to be fed by the phloem (De la Barrera and Nobel 2004), but pollen is formed from an undifferentiated mass of cells (Pratt 1988) that may be fed by the vascular bundle which contains both xylem and phloem. It is possible that some pesticides associate with the xylem and not the phloem, due to their chemical composition. This would explain why some systemic pesticide residues are found in the pollen and not the nectar, but further research is needed.

Time played a significant role in the pesticide residue levels. Developing a decay curve for these pesticides in the pollen and nectar would allow us to better understand the impact they have on pollinators and further influence management strategies. It is also important to note that as these pesticides degrade, they produce other chemicals that may be equally as toxic to pollinators.

While we do have multiple years of data for a few pesticides, we believe it is important to repeat pesticide residue analyses due to the impact of environmental factors.

The flow of water, and thus the water-soluble systemic pesticides, is affected by environmental conditions. Heat and moisture have been shown to have a significant effect on the overall residue levels of neonicotinoids in pollen and nectar (Kamel 2012).

27 Literature Cited

Agnello, A., A. Atanassov, J. C. Bergh, D. J. Biddinger, L. J. Gut, M. J. Haas, J. K. Harper, H. W. Hogmire, L. A. Hull, L. F. Kime et al. 2009. Reduced-risk pest management programs for eastern U.S. apple and peach orchards: a 4-year regional project. American Entomologist 55:184-197.

Artz, D. R., and T. L. Pitts-Singer. 2015. Effects of fungicide and adjuvant sprays on nesting behavior in two managed solitary bees, Osmia lignaria and Megachile rotundata. PLoS ONE 10(8): e0135688.

Banken, J. A. O., and J. D. Stark. 1997. Stage and age influence on the susceptibility of Coccinella septempunctata (Coleoptera: Coccinellidae) after direct exposure to Neemix, a Neem insecticide. Journal of Economic Entomology 90 (5): 1102- 1105.

Bardin, P. G., S. F. van Eeden, J. A. Moolmam, A. P. Fodon, and J. R. Joubert. 1994. Organiphosphate and carbamate poisoning. Archives of Internal Medicine 154(13): 1433-1441.

Biddinger, D. J., L. A. Hull, and E. G. Rajotte. 1998. Stage specificity of various insecticides to tufted apple bud moth larvae (Lepidoptera:Tortricidae). Journal of Economic Entomology 91(1): 200-208.

Biddinger, D.J., and E. G. Rajotte. 2015. Integrated pest and pollinator management- adding a new dimension to an accepted paradigm. Current Opinion in Insect Science 10: 204-209.

Blacquiere, T., G. Smagghe, C. A. M. van Gestel, and V. Mommaerts. 2012. Neonicotinoids in bees: a review on concentration, side effects and risk assessment. Ecotoxicology 21: 973-992.

Creswell, J. E. 2011. A meta-analysis of experiments testing the effects of a neonicotinoid insecticide (imidacloprid) on honey bees. Ecotoxicology 20(1): 149-157.

De la Barrera, E., and P. S. Nobel. 2004. Nectar: properties, floral aspects, and speculations on origin. Trends in Plant Science 9(2): 65-69.

Dively, G. P., M. S. Embrey, A. Kamel, D. J. Hawthorne, and J. S. Pettis. 2015. Assessment of chronic sublethal effects of imidacloprid on honey bee colony health. PLoS ONE 10(3): e0118748.

Dively, G. P., and A. Kamel. 2012. Insecticide residues in pollen and nectar of a cucurbit crop and their potential exposure to pollinators. Journal for Agricultural Food Chemistry, 60(18): 4449-4456.

European Food Safety Authority. 2010. Conclusion on the peer review of the pesticide risk assessment of the active substance flonicamid. EFSA Journal 2010, 8(5):1445. Available at: http://www.efsa.europa.eu/sites/default/files/scientific_output/files/main_docume nts/1445.pdf

Goulson, D. 2013. Review: An overview of the environmental risks posed by neonicotinoid insecticides. Journal of Applied Ecology 50(4): 977-987.

Kamel, A. 2012. Insecticide residues in pollen and nectar of cucurbit crop and their potential exposure to pollinators. Journal of Agricultural and Food Chemistry, 60: 4449-4456.

Krupke, C. H., G. J. Hunt, B. D. Eitzer, G. Andino, and K. Given. 2012. Multiple Routes of Pesticide Exposure for Honey Bees Living Near Agricultural Fields. PLoS ONE 7(1): e29268.

Namba T., C. T. Nolte, J. Jackrel, and D. Grob. 1971. Poisoning due to organophosphate insecticides: Acute and chronic manifestations. The American Journal of Medicine 50(4): 475-492.

Pennsylvania State University. 2016. Penn State Tree Fruit Production Guide 2015-2016. Pennsylvania State University.

Pratt, C. 1988. Apple flower and fruit: morphology and anatomy. Horticultural Reviews 10: 273-308.

Simon-Delso, N., V. Amaral-Rogers, L. P. Belzunces, J. M. Bonmatin, M. Chagnon, C. Downs, L. Furlan, D. W. Gibbons, C. Giorio, V. Girolami, D. Goulson, et al. 2014. Systemic insecticides (neonicotinoids and fipronil): trends, uses, mode of action and metabolites. Environmental Science and Pollution Research 22(1): 5- 34.

Stoner, K. A., and B. D. Eitzer. 2012. Movement of soil-applied imidacloprid and thiamethoxam into nectar and pollen of squash (Cucurbita pepo). PLoS ONE, 7(6): e39114.

Suchail, S., D. Guez, and L. P. Belzunces. 2001. Discrepancy between acute and chronic toxicity induced by imidacloprid and its metabolites in Apis mellifera. Environmental Toxicology and Chemistry 20(11): 2482-2486.

Tomizawa, M., and J. E. Casida. 2003. Selective toxicity of neonicotinoids attributable to specificity of insect and mammalian nicotinic receptors. Annual Review of Entomology 48: 339-364.

US Apple Growers Association. 2011. Brown Marmorated Stink Bug Causes $37 Million in Losses to Mid-Atlantic Apple Growers. April 14, 2011. Available at: http://www.growingproduce.com/ article/21057/brown-marmorated-stink-bug- causes-37-million- in-losses-to-mid-atlantic-apple-growers.

USGS (United States Geological Survey). 2013. Pesticide National Synthesis Project. Available at: http://water.usgs.gov/nawqa/pnsp/usage/maps/compound_listing.php van Engelsdorp, D., and M. D. Meixner. 2010. A historical review of managed honey bee populations in Europe and the United States and factors that may affect them. Journal of Invertebrate Pathology 103: S80–S95.

Yañez, K. P., M. T. Martin, J. L. Bernal, and M. J. Nozal. 2014. Trace analysis of seven neonicotinoid insecticides in bee pollen be soild-liquid extraction and liquid chromatography coupled to electrospray ionization mass spectrometry. Food Analytical Methods 7(2): 490-499.

30 Yates, J. R., and M. Sherman. 1970. Latent and differential toxicity of insecticides to larvae and adults of six fly species. Journal of Economic Entomology 63(1): 18- 23.

CHAPTER 3

ACUTE ORAL TOXICITY OF SEVERAL FORMULATED NEONICOTINOIDS AND RELATED INSECTICIDES USED IN APPLE ORCHARD PEST MANAGEMENT PROGRAMS TO THE HONEY BEE (APIS MELLIFERA (L.))

INTRODUCTION

Pollinators provide essential ecosystem services, supporting approximately 85% of the world’s flowering plants (Ollerton et al. 2011) and 35% of the global crop production (Klein et al. 2007). Pollinator health has become a pressing issue as both managed and native pollinators decline worldwide (Potts et al. 2010). Honey bees (Apis mellifera L.) are easily the most studied pollinator and are considered the most valuable.

Honey bee populations, however, have declined for the past several years due to a combination of factors referred to as Colony Collapse Disorder (CCD Steering

Committee 2007). Studies suggest that bee decline, in general, are caused by the interaction of many stress factors, including several pathogens, parasitic mites, pesticides, lack of forage and nesting habitat due to intensive monoculture, stress from poor nutrition and the transport of hives for commercial pollination (van Engelsdorp et al. 2009,

Goulson et al. 2015). Pesticide exposure in particular has received a disproportionate amount of media attention due to the presence of agricultural chemicals detected in honey bee wax and pollen (Mullin et al. 2010), despite the lack of evidence that they are the main cause of CCD (Creswell 2011).

32 One class of pesticides, known as neonicotinoids, have been implicated in increased mortality of honey bee colonies (Maxim and van der Sluis, 2010), and have become the subject of much debate. Neonicotinoids are systemic pesticides that are transported throughout the plant’s vascular tissue and protect against sap-feeding insects.

Neonicotinoids act as nicotinic acetylcholine receptors, blocking the transmission of nerve impulses in insects’ central nervous system (Matsuda et al. 2001, Tomizawa and

Casida 2003).

Although much research has been done on the effects of some neonicotinoids on bee health (Creswell 2011), few studies have been done on the newer neonicotinoids and related compounds that are applied as foliar applications in tree fruit crops. Even fewer studies have moved beyond the laboratory to address the effects in a field realistic manner (Carreck and Ratnieks 2014). One of the key factors contributing to the realistic exposure of pollinators to pesticides is determining the routes of exposure. Bees can encounter systemic insecticides, such as neonicotinoids, in many ways such as: direct contact with foliar sprays or particles released during treated seed planting, contacting residues on treated plants or nesting substrate, or consuming contaminated pollen, nectar, and water (Krupke et al. 2012, Hopwood et al. 2012). In Eastern US apple orchards, insecticides are generally not applied during the 7 to 10-day bloom period. Therefore, oral exposure through contaminated pollen and nectar is the primary route of exposure in orchards.

Federal safety testing for pesticide registration requires pesticide risk to pollinators be calculated through short-term acute laboratory studies on honey bees, using

33 the active ingredient of a substance rather than the formulated product (Office of

Pesticide Programs 2013). Independent studies are also conducted to gather more information, but often use doses that are above the range of realistic agricultural pesticide residues, and therefore represent a “worst case” scenario (Carreck and Ratnieks 2014).

Because of the variety of toxicity testing methods and doses used, the results of these studies are inconsistent (Blacquiere et al. 2012).

To determine the field realistic risk to honey bees associated with the use of neonicotinoids and neonicotinoid-related products, we assessed the acute oral toxicities of two formulated neonicotinoids (Assail® and Actara®, IRAC group 4A) and two formulated neonicotinoid-like products (Closer®, IRAC group 4C and Sivanto®, IRAC group 4D) commonly used in Eastern United States apple production. Bees were fed a range of doses of each pesticide in order to create a dose mortality curve that could be used to predict responses in future studies. We then compared the toxicity of each pesticide to known pesticide residues determined from field applications in research orchards (Chapter 2). Despite previous research showing that in contact assays, formulated product was less toxic to honey bees than active ingredient (Biddinger et al.

2013), we anticipated that the LC50 of the formulated pesticides would be lower than the reported values, indicating they are more toxic, due to the additional toxicity of the unknown inactive ingredients in formulated pesticides.

Pesticide residue levels found in the nectar from field applications (Chapter 2) would then be compared to the dose mortality curves we generated to determine expected levels of mortality in the field. Unlike in IPM and pest control journals, published LC50

34 values almost never report the slope or additional LC values to make such calculations, nor the 95% confidence limits around those values. Without this information, it is impossible to determine what level of mortality can be expected from the low pesticide levels found in our field residue study (Robertson et al. 2007).

MATERIALS AND METHODS

Insects and Treatment

Adult worker bees were sourced from three different hives from the Penn State apiary. These hives are treated traditionally and the bees were in good health. After bees were collected from the hives, they were starved for four hours and placed into group feeding cages. Five bees from each hive were placed into each cage to eliminate hive effects, 15 bees per cage total.

Cages were constructed of a Petri dish (100×20 mm) encasing the open ends of a

100 mm-long wire mesh cylinder. Each petri dish had a hole made with a heated cork- borer, one side with a larger hole to put bees in and remove dead bees, and the other side with a smaller hole where the feeder hung. The feeder was a 1.7 milliliter centrifuge tube with two pin-sized holes in the bottom to allow ad libidum feeding.

The commercial formulations (AI% ; manufacturer) of the treatments were the neonicotinoids (IRAC Code 4A) Assail 30SG (acetamiprid 30%; United Phosphorous

Inc., King of Prussia, PA) and Actara 25WP (25.0% thiamethoxam, Syngenta,

Wilmington, DE), and the neonicotinoid related compounds (IRAC Code 4D and 4C,

35 respectively) Sivanto 200SL (flupyradifurone 17.09%, Bayer CropScience, Research

Triangle Park, NC) and Closer 240SC (sulfoxaflor 21.8%, Dow AgroSciences LLD,

Indianapolis, IN). One dose of Cygon 400 (dimethoate 43.5%; Drexel Chemical

Company, Memphis, TN) was included as a positive control (Gough et al 1994), and negative control bees had access to 50% sucrose solution.

A range of 5 treatment concentrations were chosen based off the published LD50 for the active ingredient of each pesticide (acetamiprid: EC 2004, thiamethoxam:

Syngenta Group 2005, flupyradifurone: Bayer CropScience 2013, sulfoxaflor: average derived from multiple reports, EFSA 2014, Dow AgroSciences 2013). These values were converted into a concentration per bee, with 50% sucrose solution as the solvent. After the starvation period bees were given a feeder with 400 microliters of the assigned treatment. Bees were watched until about 200 microliters of the treatment was consumed, at which point the feeder was replaced with a feeder containing 50% sucrose solution only.

Three replicate cages of bees were tested with each treatment, for a total of 45 bees per dose and 225 bees per chemical. Cages were placed inside large plastic containers, which also contained a jar of saturated NaCl solution to maintain ~75% relative humidity. Plastic containers were kept in a growth chamber at 34.5°C in darkness.

36 Data Collection and Analysis

The amount of treatment solution consumed by weight was recorded and converted into an average dose per bee assuming equal consumption by individuals in a treatment cage. Mortality and behavioral observations were recorded at zero, 12, 24, and

48 hours post-exposure. Bees were considered dead when they remained absolutely still during the 30 second observation period (similar to Laurino et al. 2011).

Dose-mortality regressions were estimated assuming the normal distribution (i.e., probit model) with the computer program POLOPlus 2.0 (LeOra Software 2005) as described by Robertson et al. (Robertson et al. 2007). LC50 and LC90 values with 95% confidence intervals were also calculated using POLOPlus 2.0.

RESULTS

Negative control mortality (n=45) was 0%, and positive control mortality (n=45) was 4.4%. Dose mortality curves (Figure 3-1) and LC50 and LC90 values (Table 3-1) were generated for Assail® and Sivanto®. A valid curve could not be generated by POLOPlus

2.0 for Actara® nor Closer®, because the data points were concentrated at very low mortality, giving too much variance for proper analysis.

Table 3-1. Dose mortality curve slopes, LC50, and LC90 for Actara, Assail, Closer, and Sivanto using mortality 48 hours post-feeding

Product n Dose Mortality LC50 (PPM) LC90 (PPM) (Active Ingredient) Curve Slope ± (95% CL) (95% CL) SEM Actara 60 - - - (thiamethoxam) Assail 165 4.36 ± 0.58 405.94 799.13 (acetamiprid) (328.16-479.18) (623.33-1307.13) Closer 120 - - - (sulfoxaflor) Sivanto 150 3.03 ± 0.76 118.79 314.19 (flupyradifurone) (81.82-483.19) (155.18-7037.10)

38 DISCUSSION

We expected to be able to create dose mortality curves for all chemicals tested by using a range of concentrations, as those doses should produce a full range of mortalities necessary for POLOPlus to generate valid curves (Robertson and Preisler 1992,

Robertson et al. 2007). However, for all four pesticides, the data points were concentrated at the lower end of the dose mortality curve, suggesting that the published LD values were lower than the toxicities we observed. This could be due to the method of testing, as some groups, including the EPA, use active ingredient rather than formulated product.

The LC values determined by this experiment were lower than the published toxicity values obtained using active ingredient. This indicates that these formulated products are more toxic than active ingredient alone, although without confidence intervals from the previously published data we cannot draw definitive conclusions.

There is much debate over the oral toxicity of the inactive ingredients of formulated pesticides, and consumption of these “inert” ingredients have been shown to cause learning impairments in honey bees. (Mullin et al. 2015, Ciarlo et al. 2012).

In this present study, the range of mortalities observed was great enough to generate the LC50 and LC90 for Assail® and Sivanto®. Number of individuals included in the analysis (n), the slope of the dose mortality curve, and 95% confidence intervals for the LC values are included. These values are not usually reported in conventional toxicity testing, but they are essential for making comparisons between different chemicals, different bee species, different populations of the same species, and for extrapolating the toxicity of a known dose (Robertson and Preisler 1992, Robertson et al. 2007).

39 Nearly all of the mortality occurred within 12 hours of feeding. However, all bees were sustained on sucrose solution long after the experiment was over, and only the untreated control bees survived beyond 7 days. The poisoning symptoms observed mirrored those already reported in similar studies (Decourtye and Devillers 2010).

Sublethal effects such as staggering, partial paralysis, abdomen tucking, and twitching were observed to some degree in all treated bees, but most often in those treated with

Actara and Closer. These observations could be an indication of subletal and long term effects of pesticide consumption.

The current practice of using A. mellifera as a surrogate species for all other non- target pollinators is flawed. Previous research has found that there are differences in the contact toxicity of systemic pesticides to A. mellifera and another apple pollinator, Osmia cornifrons (Biddinger et al. 2013). We attempted to repeat this study on O. cornifrons with little success. Osmia as a species are much more active than honey bees and do not like to be contained. We had difficulty keeping them in cages, and could not coax them into consuming enough sucrose solution to be considered a dose of pesticide. Research has been done on a feeding assay method for Osmia species (Ladurner et al. 2003), but we experienced little success because O. cornifrons is more active than the species tested.

When we compared the dose mortality curves to the pesticide residue values presented in Chapter 2, we found that the doses bees are exposed to in a commercial orchard setting is much lower than the LD50. For Assail® and Sivanto®, the pollen and nectar residues equate to less than 1% of the published LD50. Although we could not extrapolate an LD50 for Actara® or Closer®, based on the trends of the mortality data we

40 can assume the pollen and nectar residues are much less than an LD50. This suggests that the exposure to formulated neonicotinoids and neonicotinoid-related products under field applications may not be as much of a threat to bees as previously hypothesized. This would explain why the toxic effects of many laboratory studies cannot be replicated in the field (Blacquiere et al. 2012, Carreck and Ratnieks 2014)

The data reported on neonicotinoid toxicity to pollinators varies greatly due to inconsistent testing methods (Blacquiere et al. 2012). More research is needed on the field realistic toxicity of these pesticides to pollinators, including: toxicity testing on all life stages, long term studies to observe sublethal effects, choice tests, and field scale assays (Carreck and Ratnieks 2014). This range of exposure testing is not currently considered during the registration or regulation of insecticides by the Environmental

Protection Agency (EPA 1996). Moving forward, we believe field scale studies where pesticide application is highly controlled and bee behavior and reproduction can be observed would be best. Also, diversifying the bees studied is essential if we want to understand the impact on wild and alternative pollinators. A uniform toxicity testing technique that emphasizes field realistic conditions with formulated pesticides could reduce the variability in reported data and produce results that better represent the effects seen in the field. Also, mandating the reporting of dose mortality curve slopes and confidence intervals around LD and LC values would allow allow more detailed conclusions to be drawn, such as which chemical is more toxic and what the toxicity of these chemicals are at a variety of doses.

41 Literature Cited Bayer Crop Science. 2013. Flupyradifurone Technical Information. Available at http://www.sivanto.com/doc/Technical-Information-SIVANTO.pdf

Blacquiere, T., G. Smagghe, C. A. M. van Gestel, and V. Mommaerts. 2012. Neonicotinoids in bees: a review on concentration, side effects and risk assessment. Ecotoxicology 21: 973-992.

CCD Steering Committee. 2007. “Colony Collapse Disorder Action Plan.” USDA-ARS

Ciarlo, T. J., C. A. Mullin, J. L. Fraizer, D. R. Schmehl. 2012. Learning impairment in honey bees caused by agricultural spray adjuvant. PLoS ONE 7(7): e40848.

Creswell, J. E. 2011. A meta-analysis of experiments testing the effects of a neonicotinoid insecticide (imidacloprid) on honey bees. Ecotoxicology 20(1): 149-157.

Decourtye, A., and J. Devillers. 2010. “Ecotoxicity of neonicotinoid insecticides to bees.” Insect nicotinic acetylcholine receptors. Springer New York, 2010. 85-95.

Dow AgroSciences. 2013. Isoclast Active Technical Bulletin. Available at http://msdssearch.dow.com/PublishedLiteratureDAS/dh_08fd/0901b803808fd73f. pdf?fil epath=nz/pdfs/noreg/012-10813.pdf&fromPage=GetDoc

EC (European Comission). 2004. “Acetamiprid. Review Report for the Active Substance Acetamiprid. SAN-CO/1392/2001- Final.” European Comission Health and Consumer Protection Directorate- General. Available at http://ec.europa.eu/food/plant/protection/evaluation/newactive/acetamiprid.pdf

EFSA (Eurpoean Food Saftey Authority). 2014. Conclusion on the peer review of the pesticide risk assessment of the active substance sulfoxaflor. EFSA Journal. 2014;12: 3692.

42 EPA (United States Environmental Protection Agency). 1996. Ecological Effects Test Guidelines. OPPTS 850.3020. Honey Bee Acute Contact Toxicity. Available at: http://www.eppa.gov/ocspp/pubs/frs/publications/OPPTS_Harmonized/850_Ecol ogical_ Effects_Test_Guidelines/Drafts/850-3020.pdf.

Goulson, D., E. Nicholls, C. Botias, and E. L. Rotheray. 2015. Bee declines driven by combined stress from parasites, pesticides, and lack of flowers. Science 347(6229): 1255957-1 - 1255957-9.

Krupke, C. H., G. J. Hunt, B. D. Eitzer, G. Andino, and K. Given. 2012. Multiple Routes of Pesticide Exposure for Honey Bees Living Near Agricultural Fields. PLoS ONE 7(1): e29268.

Ladurner, E., J. Bosch, S. Maini, and W. Kemp. 2003. A method to feed individual bees (Hymenoptera: Apiformes) known amounts of pesticides. Apidologie 34(6): 597-602.

Laurino, D., M. Porporato, A. Patetta, and A. Manino. 2011. Toxicity of neonicotinoid insecticides to honey bees: laboratory tests. Bulletin of Insectology 64(1): 107- 113.

LeOra Software, POLOPlus 2.0. 2005. Available: http://www.LeOraSoftware.com. Accessed 2015 November 1.

Mullin, C. A., M. Frazier, J. L. Frazier, S. A. Ashcraft, R. Simonds, D. van Engelsorp, and J. S. Pettis. 2010. High levels of miticides and agrochemicals in North American apiaries: Implications for honey bee health. PLoS ONE 5: e9754.

Mullin, C. A., J. Chen, J. D. Find, M. T. Frazier, and J. L. Frazier. 2015. The formulation makes the honey bee poison. Pesticide Biochemistry and Physiology 120: 27-35.

Ollerton, J. R., R. Winfree. and S. Tarrant. 2011. How many flowering plants are pollinated by animals? Oikos 120: 321-326.

Pollinator Health Task Force. 2014. “National strategy to promote the health of honey bees and other pollinators” (The White House, Washington, DC).

Potts, S. G., J. C. Biesmeijer, C. Kremen, P. Neumann, O. Schweiger, and W. E. Kunin. 2010. Global pollinator declines: Trends, impacts and drivers. Trends in Ecology and Evolution. 25: 345–353.

Robertson, J. L. and H. K. Preisler. 1992. Bioassays with Arthropods. Boca Raton: CRC Press. 127 p.

Robertson, J. L., R. M. Russell, H. K. Preisler, and N. E. Savin. 2007. Bioassays with Arthropods. Boca Raton: CRC Press. 199 p.

Syngenta Group. 2005. “Envirofacts. Syngenta Crop Protection Fact Sheet: Thiamethoxam.” Available at http://www.syngentacropprotection.com/env_stewardship/futuretopics/Thiometho xamEn virofacts_7-19-05.pdf

Tomizawa, M., and J. E. Casida. 2003. Selective toxicity of neonicotinoids attributable to specificity of insect and mammalian nicotinic receptors. Annual Review of Entomology, 48: 339-364. van Engelsdorp, D., J. D. Evans, C. Saegerman, C. Mullin, E. Haubruge, B. K. Nguyen, M. Frazier, J. Frazier, D. Cox-Foster, Y. Chen, et al. 2009. Colony collapse disorder: a descriptive study, PLoS ONE 4: e6481.

CHAPTER 4

THE USE OF POLLINATOR HABITAT AUGMENTATION IN PENNSYLVANIA APPLE ORCHARDS

INTRODUCTION

Insect pollinators are essential in nearly all terrestrial ecosystems, and the ecosystem services they provide are vital to both wild plant communities and agricultural production. Bees are the primary insect pollinators in agricultural ecosystems, and they provide an estimated global service worth $215 billion to food production (Gallai et al.

2009). Most crop producers rely on managed bees, such as the honey bee (Apis mellifera

L.), for their pollination needs. Despite successful use of the honey bee for many years, the reliance on this single species for commercial pollination has become threatened by high rates of colony loss over the last decade (Lee et al. 2015, Potts et al. 2010b).

Declines of wild bee populations, although more difficult to measure, also raise concerns regarding pollination deficits (Koh et al. 2016, Bartomeus et al. 2013, Burkle et al. 2013).

Bee declines are most likely caused by the exposure to multiple interacting stressors such as pesticide use, pathogens, parasites, and a reduction in appropriate floral and nesting resources (Goulson et al. 2015).

Many factors may limit access to the feeding and nesting resources bees need

(Goulson et al. 2015). Loss of the habitat providing these resources has been a long-term contributor to bee declines (Potts et al. 2010a). Although many bees may respond favorably to moderate landscape disturbances that create nesting sites and stimulate

45 alternative nectar and pollen sources (Winfree et al. 2007), extreme urbanization and monoculture cropping create a fragmented habitat and limited diet, which can be generally unfavorable to managed and wild bees (Bates et al. 2011, Donaldson-Matasci and Dornhaus 2012, Cane and Tepedino 2001). Although not well understood, climate change could lead to range shifts or constrictions (Kerr et al. 2015), and temporal separation of pollinators and the plants they pollinate. Herbicides are used regularly to control weeds in most cropping systems, and this may reduce the availability of season- long flowers for pollinators (Goulson et al. 2008).

Recognizing the importance of all pollinators and the threat of reduced floral resource and habitat, the National Strategy to Promote the Health of Honey Bees and

Other Pollinators has addressed pollinator habitat acreage as one of its three main goals, aiming to restore or enhance 7 million acres of land for pollinators by 2020 (Pollinator

Health Task Force 2015). Pollination (and subsequently crop yields) is generally higher in areas located closer to natural or semi-natural habitats (Klein et al. 2012, Nayak et al.

2015). Because of this, many researchers and policy makers are encouraging the creation of flower-rich habitats such as hedgerows, field borders, or cover crops to conserve bee populations and increase crop pollination (Figure 4-1) (Williams et al. 2015). However, it is still unclear whether these habitats actually increase the population of the specific pollinators required for targeted crop pollination (Sidhu and Joshi 2016), and flowers in these habitats may compete with the crop and interfere with crop pollination (Somerville

1999, Free 1993).

Crop yield increases Pollination rate in crops increases Rare pollinator species become less rare Number of pollinators overwintering successfully increases Colony size increases Biological fitness of pollinator larvae and/or adults increases Nest pollen stores increase in size and quality The proportion of bees with pollen in the corbiculae/ full corbiculae increases The quality of pollen/nectar taken by individual pollinators increases Pollinators aggregate on the added floral resource Figure 4-1. A proposed hierarchy of the effects of pollinator habitat in crop production

(modified from Wratten et al. 2012)

In the present study, we examined the bee communities found in managed floral resource pollinator strips and how they compared to the bee communities found in commercial Pennsylvania apple orchards and their surrounding forests. We focused on three questions: (1) Are the bee communities in the orchard different than those in the pollinator strips? (2) Are the communities of known tree fruit pollinators in the orchard different than those in the pollinator strips? (3) Do we find more rare pollinator species in the pollinator strips or in the orchards? We chose these two subsets of the bee communities to act as indicators of the utility of the pollinator strips; known tree fruit pollinators representing their value in improving crop pollination, and the presence of rare species indicating overall conservation.

We expected that the bee communities in the orchard would be less rich and diverse than those in the pollinator strips because the pollinator strips were designed to have diverse floral resources for a longer blooming period. However, we hypothesized

47 that the orchards would have a more diverse and rich assemblage of bees that are known to be tree fruit pollinators because they would be more likely to move into the orchard during fruit bloom periods. We also hypothesized that the pollinator strips would support more species identified as rare because the increase in floral and nesting resources would increase the presence of rare species.

MATERIALS AND METHODS

Study Sites

Multi-year field studies were conducted at six separate apple orchards and eight pollinator strips in fruit grower orchards in Adams County, PA from 2012-2014 (Table 4-

1). In this area of Pennsylvania, which is the leading tree fruit producing county in the state, orchards have steep slopes, well drained soils and a landscape matrix of approximately 8% orchards, 24% arable and pasture land, 9% developed area and 56% forests. Orchards are small and bordered by undeveloped scrub, forest, or fence rows.

This orchard-forest interface is home to a diverse plant community, supporting 146 plant species (Kammerer et al. 2015). The orchards tested ranged in age from 15-25 years, and contained primarily ‘York’, ‘Golden Delicious’, and ‘Honey Crisp’ apple varieties.

Pollinator strips were established under a project funded through contracts with the United States Department of Agriculture’s Natural Resource Conservation Service

(USDA-NRCS) under the Environmental Quality Incentive Program (EQIP) and

Conservation Stewardship Program (CSP). These pollinator strips were planted according

48 to guidelines developed in conjunction with the Xerces Conservation Society, which specified seeding rates, plant species, and establishment and maintenance programs.

Pollinator strip seed mixtures were developed by USDA-NRCS and the Xerces

Conservation Society and supplied by Ernst Conservation Seed (884 Mercer Pike,

Meadville, PA 16335) to best support a diversity of local bees (see Figure 4-2). We established pollinator strips next to apple orchards, with a focus on (a) providing blooming wild plants throughout the growing season, avoiding apple bloom time; (b) including plant species that would not become weeds in the primary crop, and (c) including plant species that would not serve as alternate hosts of crop pests and diseases.

Orchards were maintained similarly by conventional IPM practices using both conventional and reduced risk pesticides and biological control by the land owners.

Pollinator strips were maintained with yearly mowing and spot spraying for noxious weeds, accordance with USDA-NRCS/ Xerces Society recommendations.

Oct

Sept

Aug

July

June

May

Apr

Mar

- - 12% 2% 9% 2% 6% 7-8% 7-9% 7-10% 2% 1% 0-5% 0-2% 5-6% 8-11% 0-1% 2% 0-2% 3% 4% 5% 8% Percent of Percent Mix

Common Name Common Apple Maple Red Willow Black coreopsis Lancelead Ohio Spiderwort Beardtounge Hairy Susan Black-Eyed Coneflower Purple Bergamot Wild Mountain Virginia Mountain Bigleaf Cupplant Master Rattlesnake Goldenrod Early Goldenrod Stiff Pea Partridge Mint Dotted Rough Blazingstar Wingstem Aster New England rye wild Virginia Bluestem Big Lovegrass Purple Bluestem Little

. Pollinator planting species mix with bloom times *Grasses, no bloom mixtimes .*Grasses, designated plantingbloom time with species Pollinator 2 - 4

Figure

Scientific Name Scientific Malus domestica rubrum Acer nigra Salix Coreopsis lanceolate ohiensis Tradescantia hirsutus Penstemon hirta Rudbeckia purpurea Echinacea Monarda fistulosa virginiaum Pycnanthemum muticum Pycnanthemum perfoliatum Silphium yuccifolium Eryngium juncea Solidago rigidum Oligoneuron fasciculate Chamaecrista Monarda punctate Liatris spicata alternifolia Verbesina Symphyotrichum novae-angliae virginicus* Elymus Andropogon gerardii* Eragrostis spectabilis* scoparium* Schizachyrium

Sampling

We sampled pollinator populations in the orchards and the pollinator strips using blue vane traps (model Z-BVT, SpringStar Inc., Woodinville, WA), which are highly efficient in sampling pollinators (Kimoto et al. 2012, Joshi et al. 2015). The blue vane trap consists of a yellow plastic container (0.71 l volume capacity) with a blue plastic funnel and two blue plastic cross vanes. Each pollinator planting had two traps, while orchards had five or ten traps, depending on the size of the orchard. Traps were suspended approximately 1 m above the ground with nylon rope and a metal pole.

Supertech® antifreeze (ethylene glycol (107-21-1), diethylene glycol (111-46-6), water

(7732- 18-5), Wal-Mart Stores, Inc., Bentonville, AR) diluted with tap water (60:40) was used to preservative for trapped specimens. Trapped bees were collected weekly, from early April to first frost each year.

Bees were collected from traps using a stainless steel strainer and a pair of flathead forceps, and immediately placed in a clear plastic vial containing 70% ethanol for further processing. In the laboratory, bee samples were dried, pinned, labeled, and identified to species level. Identifications were done by Jason Gibbs at Michigan State

University (Lasioglossum), Robert Jean (Andrenidae), Karen Wright at the University of

New Mexico (Melissodes). The remaining bees were identified by Rick Donvall (USDA-

APHIS) and David Biddinger (Penn State University) using keys from Mitchell (1960,

1962), Michener (2000), and Discover Life (Discover Life n.d.).

Statistical Analysis

For all analyses, data were combined across dates and traps at each site within a year. For some analyses, data was limited to include only known tree fruit pollinators

(determined by net collections in Pennsylvania orchards at tree fruit bloom), rare species

(determined by the expert opinion, Sam Droege and David Biddinger), and only specimens collected by the blue vane traps during tree fruit bloom time (mid-April to mid-May). The Shannon Diversity Index, which accounts for both abundance and evenness of species present, was calculated for each site-year combination. Means were compared in RStudio using generalized linear mixed models using the lme4 package

(RStudio Team 2015, Bates et al. 2015) in which site and year were assigned as random factors, and site type (orchard versus pollinator strip) as a fixed explanatory factor.

Species richness was estimated using individual-based rarefaction curves developed in EstimateS 8.2 (Colwell 2005). The rarefaction curves allow for a comparison of species richness by indicating the statistical expectation of species accumulation based on 100 permutations of the species-by-sample matrix. Significance was determined by non-overlapping confidence intervals (Payton et al. 2003). Species abundance was represented by plotting rank abundance curves for the ten most abundant species at each site. Community evenness was calculated using Pielou's index of evenness (Pielou, 1966).

Differences in species composition of the bee communities in the orchards versus those in the pollinator strips were visualized with non-metric multidimensional scaling

(NMDS) using the metaMDS function in the vegan package in RStudio (Oksanen et al.

53 2016, RStudio Team 2015). NMDS is an ordination technique used to visualize the similarity between data points (in this case, each site/year combination) based on the dissimilarity of any number of variables (in this case, the abundance of each species)

(Hoand 2008). We conducted two NMDS analyses; the first analysis included all bee species collected in this study, and the second analysis only included the subset of bees that were known to be tree fruit pollinators. Statistical significance was determined by a

PERMANOVA using the adonis function in the vegan package in RStudio (Oksanen et al. 2016, RStudio Team 2015).

To investigate associations between individual tree fruit pollinating species and site types (orchard and pollinator strip), we examined community assemblage patterns using redundancy analysis (RDA), a constrained ordination technique, in CANOCO 4.5

(ter Braak and Šmilauer 2002). Species data were centered and standardized, and year was included as a covariable in the analysis. The significance of site type was determined using Monte Carlo simulations with 499 permutations of the data. Species associations with orchard or pollinator strips were then visualized using biplots developed in

CanoDraw 4.5 (ter Braak and Šmilauer 2002).

Percent abundance of species identified as known fruit pollinators and species identified as rare was calculated for each site each year, and analyzed in RStudio using generalized linear mixed models using the lme4 package (RStudio Team 2015, Bates et al. 2015), in which site and year were assigned as random factors, and site type (orchard versus pollinator strip) as a fixed explanatory factor.

RESULTS

Over the three years of the study 24,155 bees were collected, which were identified to 138 species: 7,507 specimens and 100 species in the pollinator strips versus

16,648 specimens and 116 species in the orchards. Percent abundance was calculated for each species as a proportion of the total number of individuals collected at that site type

(Table 4-2).

Table 4-2. Species list and abundances of bees in pollinator strips and orchards in Adams Co., PA; 2012-2014.  = Known tree fruit pollinator, ◊ = Rare species Percent Abundance Species name Author Orchard Pollinator Strip Andrenidae Andrena barbara  Bouseman & LeBerge 0.01 0.01 Andrena bisalicis  Viereck 0.04 0.03 Andrena carlini  Cockerell 0.23 0.07 Andrena commode  Smith 0.26 0.13 Andrena cornelli Viereck 0.01 Andrena cressonii  Robertson 0.01 Andrena distans ◊ Provancher 0.01 Andrena dunning  Cockerell 0.50 0.03 Andrena erythrogaster (Ashmead) 0.01 Andrena forbesii  Robertson 0.02 0.04 Andrena heraclei  Robertson 0.02 Andrena hippotes  Robertson 0.01 Andrena imitatrix  Cresson 0.05 0.11 Andrena mandibularis  Robertson 0.01 Andrena miserabilis  Cresson 0.04 0.24 Andrena nasonii  Robertson 0.01 0.01 Andrena nivalis Smith 0.01 Andrena nuda  Robertson 0.01 Andrena perplexa  Smith 0.41 0.12

Andrena pruni  Robertson 0.05 Andrena rugosa  Robertson 0.05 0.01 Andrena tridens  Robertson 0.10 Andrena vicina  Smith 0.02 Andrena violae  Robertson 0.19 0.08 Andrena wilkella  (Kirby) 0.11 0.13 Calliopsis andreniformis Smith 0.02 0.16 Pseudopanurgus (Robertson) 0.01 compositarum

Apidae Anthophora abrupta  Say 0.16 0.01 Anthophora bomboides Kirby 0.25 0.31 Anthophora plumipes ◊ (Pallas) 0.04 Anthophora terminalis Cresson 0.47 0.13 Apis mellifera  Linnaeus 4.41 6.61 Bombus auricomus Robertson 0.01 0.07 Bombus bimaculatus  Cresson 1.71 2.42 Bombus fervidus (Fabricius) 1.14 3.84 Bombus griseocollis  (DeGeer) 0.28 1.27 Bombus impatiens  Cresson 7.80 4.92 Bombus insularis (Smith) 0.01 Bombus perplexus  Cresson 1.81 2.24 Bombus vagans  Smith 9.77 8.58 Cemolobus ipomoeae ◊ (Robertson) 0.02 0.08 Ceratina calcarata  Robertson 17.09 5.90 Ceratina dupla  Say 3.12 0.35 Ceratina mikmaqi Rehan and Sheffield 0.01 Ceratina strenua  Smith 0.35 0.16 Eucera atriventris (Smith) 0.01 0.01 Eucera dubitata (Cresson) 0.03 0.01 Eucera hamata (Bradley) 1.21 16.21 Eucera rosae ◊ (Robertson) 0.01 Melissodes bimaculata (Lepeletier) 3.56 8.70 Melissodes denticulata Smith 0.10 0.35 Melissodes dentiventris Smith 0.01 0.08 Melissodes desponsa Smith 0.86 0.97 Melissodes druriella (Kirby) 0.02 0.04 Melissodes illata Lovell and Cockerell 0.01

Melissodes subillata LaBerge 0.03 Melissodes trinodis Robertson 0.08 5.33 Melitoma taurea (Say) 0.45 0.60 Nomada maculata Cresson 0.01 Nomada ovata  (Robertson) 0.01 Peponapis pruinosa (Say) 2.09 5.66 Ptilothrix bombiformis (Cresson) 0.17 1.11 Svastra obliqua (Say) 0.04 Triepeolus concavus ◊ (Cresson) 0.01 Triepeolus lunatus (Say) 0.01 Xylocopa virginica  (L.) 0.07 0.44 Colletidae Colletes compactus Cresson 0.01 Hylaeus affinis (Smith) 0.02 0.07 Hylaeus hyalinatus Smith 0.03 Hylaeus mesillae (Cockerell) 0.03 Hylaeus modestus Say 0.07 0.01 Halictidae Agapostemon sericeus (Forster) 0.05 Agapostemon splendens (Lepeletier) 0.01 Agapostemon texanus Cresson 0.05 0.29 Agapostemon virescens (F.) 0.43 4.41 Augochlora pura  (Say) 35.74 3.44 Augochlorella aurata (Smith) 0.91 1.29 Augochlorella persimilis (Viereck) 0.01 Augochloropsis metallica  (Fabricius) 0.01 Halictus confusus  Smith 0.05 0.11 Halictus ligatus Say 0.37 3.61 Halictus rubicundus  (Christ) 0.15 0.03 Lasioglossum admirandum  (Sandhouse) 0.10 0.08 Lasioglossum bruneri (Crawford) 0.02 0.01 Lasioglossum callidum (Sandhouse) 0.01 0.07 Lasioglossum coeruleum (Robertson) 0.01 Lasioglossum coriaceum (Smith) 0.07 0.03 Lasioglossum cressonii  (Robertson) 0.03 0.03 Lasioglossum ephialtum Gibbs 0.01 0.01 Lasioglossum forbesii  (Robertson) 0.02 Lasioglossum foxii  (Robertson) 0.04 Lasioglossum gotham  Gibbs 0.04 0.01

Lasioglossum hitchensi Gibbs 0.05 0.21 Lasioglossum imitatum (Smith) 0.01 0.01 Lasioglossum katherineae Gibbs 0.01 Lasioglossum macoupinense (Robertson) 0.01 Lasioglossum oblongum (Lovell) 0.01 0.01 Lasioglossum obscurum (Robertson) 0.01 Lasioglossum paradmirandum (Knerer & Atwood) 0.01 Lasioglossum pectorale (Smith) 0.01 0.03 Lasioglossum pilosum  (Smith) 0.43 7.23 Lasioglossum quebecense  (Crawford) 0.04 Lasioglossum subviridatum (Cockerell) 0.02 Lasioglossum tegulare (Robertson) 0.17 0.11 Lasioglossum truncatum  (Robertson) 0.02 Lasioglossum versans  (Lovell) 0.05 0.04 Lasioglossum versatum  (Robertson) 0.14 0.04 Lasioglossum weemsi (Mitchell) 0.01 Lasioglossum zephyrum  (Smith) 0.01 Lasioglossum zonulum ◊ (Smith) 0.01 Megachilidae Anthidium manicatum (Linnaeus) 0.12 Anthidium oblongatum (Illiger) 0.01 0.09 Coelioxys rufitarsis Smith 0.01 Coelioxys sayi Robertson 0.01 Heriades carinata Cresson 0.04 Hoplitis pilosifrons (Cresson) 0.01 0.04 Hoplitis truncate (Cresson) 0.01 Megachile addenda Cresson 0.01 Megachile brevis Say 0.02 0.07 Megachile campanulae (Robertson) 0.02 0.09 Megachile centuncularis (Linnaeus) 0.01 Megachile gemula Cresson 0.05 Megachile integra Cresson 0.01 Megachile melanophaea Smith 0.01 Megachile mendica Cresson 0.08 0.08 Megachile montivaga Cresson 0.05 Megachile pugnata Say 0.02 Megachile relativa Cresson 0.01 Megachile rotundata (Fabricius) 0.08 Megachile sculpturalis Smith 0.03

Osmia atriventris  Cresson 0.31 0.05 Osmia bucephala Cresson 0.16 0.08 Osmia cornifrons  (Radosz.) 0.20 0.01 Osmia lignaria  Say 0.03 0.03 Osmia pumila  Cresson 0.50 0.03 Osmia taurus  Smith 0.05 0.01 Osmia texana ◊ Cresson 0.01 0.01 Osmia virga Sandhouse 0.01

Throughout the duration of the study, pollinator strips were significantly more diverse than the orchards (Z = -3.219, P = 0.0013) (Figure 4-3a). However, there was no significant difference in species richness (Figure 4-4a). Across all samples collected, the known tree fruit pollinators were significantly more abundant in the orchards compared to the pollinator strips (Z = 3.479, P = 0.0005). However, after fruit bloom the abundance of known tree fruit pollinating bees in the pollinator strips increased significantly (Z = -

6.436, P = 1.23e-10.). The pollinator strips had a significantly more diverse community of known tree fruit pollinators than the orchards (Z = -2.056, P = 0.0397) (Figure 4-3b).

There was no significant difference in richness of known tree fruit pollinators between the orchards and the pollinator strips (Figure 4-4b).

(H’)

Diversity

60 Rank abundance graphs were created for the ten highest ranking species in orchards and pollinator strips (Figure 4-5). Based on Pielou’s evenness index, community evenness was higher in the pollinator plantings (J’ = 0.559) and lower in the orchard (J’ = 0.477).

61 The bee communities visiting orchards and pollinator strips were not

differentiable in the NMDS ordination (Stress=0.1136, P=0.816) (Figure 4-6). However,

when the data were limited to known tree fruit pollinators, the communities visiting the

orchards and the pollinator strips are significantly different (Stress=0.1395, P=0.001)

(Figure 4-7).

Figure 4-7: Non-metric multidimensional scaling ordination of study sites and types according to known tree fruit pollinating bee species composition. The ordination is based on the relative Sørensen index, which separates sites based on proportional abundance. Stress= 0.1395, P=0.001

Among the known tree fruit pollinators, there was a significant gradient (F = 8.51,

P = 0.002) in the community between orchards and pollinator strips, with most species

being more closely associated with the orchards (Figure 4-8). The first axis, which

delineates the community gradient between the two site types, explained 22.7% of the

variance in species data.

Figure 4-8. RDA ordination biplot depicting associations between site types and known tree fruit pollinating species. Species names have been abbreviated, full names are in Table 4-2

There was no significant difference in rare species abundance between the

orchards and the pollinator strips (Z = - 0.137, P = 0.891).

DISCUSSION

The overall bee communities in Pennsylvania apple orchards and their

surrounding forests were not significantly different from the bee communities in the

pollinator strips. However, the pollinator strips hosted more diverse bee community, and

64 perhaps acted as a reservoir for tree fruit pollinating species after fruit bloom. There were no differences in the abundance of rare species in the orchard versus the pollinator strips.

Bee Community Differences

Different statistical techniques produced contradictory results. The traditional community ecological diversity measurement suggests that the pollinator strip bee communities are significantly different than the orchard bee communities. However, this method focuses on a characteristic of the communities as an indicator whereas the dissimilarity-based non-metric multidimensional scaling is a measurement of total community composition. NMDS indicated that the bee communities in the orchards and the pollinator strips do not differ significantly.

The orchards have a few species (A. pura, C. calcarata, and B. vagans) comprising more than half of the abundance. These species are all known tree fruit pollinators, that are readily found in the orchards in this area (Biddinger et al. 2016, Joshi et al. 2016). A. pura and C. calcarata nest in rotting wood, stems, or pith, which makes the orchard- forest interface an ideal habitat for them. All of these species are also multivoltine, which would contribute to their high abundance. These few species being so abundant is apparent in Pielou’s index of evenness, and it is the cause of the orchard bee populations low diversity.

The increased bee diversity in the pollinator strips may be explained by the increased plant diversity. More natural, diverse habitats are associated with increased pollinator diversity (Kremen et al. 2002, Ricketts et al. 2008, Kennedy et al. 2013).

Although the prevalent orchard/forest interface habitat has been found to support a large

65 population of plant species (Kammerer et al. 2015), the pollinator plantings were designed to bloom for a much longer period of time than the orchard/forest interface (see

Table 4-2). Many of the plants in the orchard/forest interface are early season perennials

(Kammerer et al. 2015). It’s also possible that the orchard/forest interface is impacted by orchard herbicide applications that would reduce plant diversity and thus bee diversity.

Tree Fruit Pollination Benefits

Both statistical techniques suggest that there are significant differences in the tree fruit pollinating communities of the orchards and the pollinator strips. The RDA ordination shows this is driven by most of the known tree fruit pollinating species associating strongly with the orchard sites. X. virginica, B. griseocollis, L. pilosum, and

A. miserabilis were the only species of known tree fruit pollinators that associated with the pollinator strips. All four of these species are generalists which may be utilizing the late season blooming resources in the pollinator plantings.

Although there is a higher diversity of tree fruit pollinating species in the pollinator strips, these species are much more abundant in the orchards. Once again, the diverse plant community in the pollinator plantings is attracting a diverse set of tree fruit pollinating species. This diverse floral provisioning may be acting as a nest site resource, rather than a food source, which would explain the high diversity but low abundance. An increase in pollinator diversity has been shown to increase fruit set in coffee (Klein et al.

2003) and pumpkins (Hoehn et al. 2008), though it is unclear whether the diversity of tree fruit pollinators we see in our pollinator strips are actively pollinating the orchard trees.

66 To better understand these interactions, a paired experiment with orchards with and without pollinator plantings where yield was observed would be useful.

The pollinator plantings may be acting as a reservoir for tree fruit pollinators, as we saw an increase in the abundance of tree fruit pollinating species in the pollinator plantings after fruit bloom. However, the high abundance of tree fruit pollinators found in the orchards suggests that pollination augmentation may not be necessary in

Pennsylvania apple orchards. The orchard-forest interface may be acting as a pollinator reservoir, since we see very little nesting in the orchards (Biddinger, pers. obs.). Also, apples only require 2-8% of the blossoms to set for a good commercial crop (Chaplin and

Westwood 1980). Despite most of the apple growers in Adams County PA not renting honey bees anymore (Joshi et al. 2011) and many not having pollinator strips on their property, there isn’t a shortage of apple production. Perhaps focusing on a crop that is more dependent on pollination services, such as cherries (Holzschuh et al. 2012), would be more useful.

Rare Species Conservation

There was no significant difference in the abundance of rare species in the orchards and the pollinator strips. In fact, there were six rare species found in the orchards and only three rare species found in the pollinator plantings (Figure 4-9). T. concavus, the only rare species found in the pollinator plantings and not the orchards, is a known parasite of the genus Colletes. Research suggests that the diversity and abundance of parasitic bees is indicative of the status of the bee community (Sheffield et al. 2013).

This supports our hypothesis that pollinator plantings are a more diverse habitat.

67 However, because the rare species were found in such low abundances, even with our extensive sampling, we can conclude that the pollinator strips do not act to conserve rare bee species.

Figure 4-9. A Venn diagram depicting the rare bee species found in the orchards, pollinator strips, and both

Policy Implications and Suggestions

These pollinator plantings met the design goals in Pennsylvania apple orchards.

Anecdotally, we did not observe issues with the plantings becoming weeds in the crop, or an increase in pests that used the pollinator planting species as hosts. Furthermore, when the data was limited to only bees collected during fruit bloom, the orchards were significantly more diverse than the pollinator strips, indicating that the pollinator strips were not attracting pollinators away from the orchards during fruit bloom. However, we did witness problems with weed management and sustainability of the pollinator

68 plantings. Canadian thistle, poison ivy, Virginia creeper and field bindweed were abundant in many of the plantings, and many of the species intentionally established in the pollinator plantings did not survive past two years.

Overall, the pollinator strips attracted a more diverse community of pollinators, but they have little value as a method of augmenting pollination or as a form of rare species conservation. In order for them to act as a pollination service supplementation, they must be tailored more specifically to the crop and the species that pollinate it. Only a small minority of common bee species provide most of the crop pollination services

(Kleijn et al. 2015), so those bees must be targeted when creating habitat. In this case, it would have been beneficial to have more early season plants, blooming before the crop in the pollinator habitats, as many of the apple pollinating species are early season bees.

These pollinator plantings increase bee populations as a whole, but we believe the best strategy for pollination supplementation and conservation in Pennsylvania orchards would be to develop management practices for hedgerows and woods-edge (see Vaughan et al. 2014). Because of the diverse floral and nesting resources already present in the orchard-forest interface (Kammerer et al. 2015), pollinator plantings are redundant. With the goal of 7 million acres of pollinator habitat established in the United States by the

Pollinator Health Task Force (2015), we must consider specified plant mixes tailored to the ecosystem they’ll be established in and the services that ecosystem already provides in order to maximize bee conservation.

69 Literature Cited

Bates, A.J., J. P. Sadler, A. J. Fairbrass, S. J. Falk, J. D. Hale, and T. J. Matthews. 2011. Changing bee and hoverfly pollinator assemblages along an urban-rural gradient. PLoS ONE 6: e23459.

Bates, D., M. Maechler, B. Bolker, and S. Walker. 2015. Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software, 67(1), 1-48.

Biddinger, D.J., E. G. Rajotte, and N. K. Joshi. 2016. Integrating pollinator health into tree fruit IPM- A case study of Pennsylvania apple production. In: The pollination of cultivated plants: a compendium for practitioners (2nd edition) (Editor- D. Roubik) FAO, Rome. In Press.

Burkle, L.A., J. C. Marlin, and T. M. Knight. 2013. Plant–pollinator interactions over 120 years: loss of species, co-occurrence, and function. Science, 339:1611-1615.

Cane, J. H. and V. J. Tepedino. 2001. Causes and extent of declines among native North American invertebrate pollinators: detection, evidence, and consequences. Conservation Ecology 5(1): 1.

Chaplin, M.H., and M. N. Westwood. 1980. Relationship of nutritional factors to fruit set. Journal of Plant Nutrition 2(4): 477-505.

Colwell, R. K. 2005. Estimate: Statistical estimation of species richness and shared species from samples. Version 8.2. (purl.oclc.org/estimates)

Discover Life. (n.d.). www.discoverlife.org.

Donaldson-Matasci, M. C., and A. Dornhaus. 2012. How habitat affects the benefits of communication in collectively foraging honey bees. Behavioral Ecology & Sociobiology 66 (4): 583-592.

Free JB. 1993. Insect Pollination of Crops, 2nd ed. Academic Press, London, 684 pp.

Gallai, N., J. M. Salles, J. Settele, and B. E. Vaissière. 2009. Economic evaluation of the vulnerability of world agriculture confronted with pollinator decline. Ecological Economics. 68: 810–821.

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

Goulson D., E. Nicholls, C. Botias, and E. L. Rotheray. 2015. Bee declines driven by combined stress from parasites, pesticides, and lack of flowers. Science 347(6229): 1255957-1 - 1255957-9.

Hoand, S. M. 2008. Non-metric multidimensional scaling (MDS). University of Georgia, Athens. Available at: http://strata.uga.edu/software/pdf/mdsTutorial.pdf.

Hoehn, P., T. Tscharntke, J. M. Tylianakis, and I. Steffan-Dewenter. 2008. Functional group diversity of bee pollinators increases crop yield. Proceedings of the National Academy of Sciences B, 275, 2283-2291.

Holzschuh, A., J. H. Dudenhöffer, and T. Tscharntke. 2012. Landscapes with wild bee habitats enhance pollination, fruit set and yield of sweet cherry. Biological Conservation, 153, 101-107.

Joshi, N. K., T. Leslie, E. G. Rajotte, M. A. Kammerer, M. Otieno, and D. J. Biddinger. 2015. Comparative trapping efficiency to characterize bee abundance, diversity and community composition in apple orchards. Annals of the Entomological Society of America 108: 1-15.

Joshi, N. K., M. Otieno, E. G. Rajotte, S. J. Fleischer, and D. J. Biddinger. 2016. Proximity to natural woodland and landscape structure drives pollinator visitation in uniformly managed apple orchard systems. Frontiers in Ecology and Evolution 4:38.

Kammerer, M. A., D. J. Biddinger, E. G. Rajotte, and D. A. Mortensen. 2015. Local plant diversity across multiple habitats supports a diverse wild bee community in Pennsylvania apple orchards. Environmental Entomology 45(1): 1–8.

Kennedy, C.M., E. Lonsdorf, M. C. Neel, N. M. Williams, T. H. Ricketts, R. Winfree, R. Bommarco, C. Brittain, A. L. Burley, D. Cariveau, et al. 2013. A global quantitative synthesis of local and landscape effects on wild bee pollinators in agroecosystems. Ecological Letters 16: 584–599.

71 Kimoto, C., S. J. DeBano, R. W. Thorp, S. Rao, and W. P. Stephen. 2012. Investigating temporal patterns of a native bee community in a remnant North American bunchgrass prairie using blue vane traps. Journal of Insect Science 12:108.

Kleijn, D., R. Winfree, I. Bartomeus, L. G. Carvalheiro, M. Henry, R. Isaacs, A. M. Klein, C. Kremen, L. K. M’Gonigle, R. Rader et al. 2015. Delivery of crop pollination services is an insufficient argument for wild pollinator conservation. Nature communications 6: 7414.

Klein, A. M., I. Steffan-Dewenter, and T. Tscharntke. 2003. Fruit set of highland coffee increases with the diversity of pollinating bees. Proceedings of the National Academy of Sciences B, 270, 955-961.

Klein, A. M., C. Brittain, S. D. Hendrix, R. Thorp, N. M. Williams, and C. Kremen. 2012. Wild pollination services to California almond rely on semi-natural habitat. Journal of Applied Ecology, 49: 723–732.

Koh, I., E. V. Lonsdorf, N. M. Williams, C. Brittain, R. Isaacs, J. Gibbs, T. H. Ricketts. 2016. Modeling the status, trends, and impacts of wild bee abundance in the United States. Proceedings of the National Academy of Sciences 113(1): 140-145.

Kremen, C., N. M. Williams, and R. W. Thorp. 2002. Crop pollination from native bees at risk from agricultural intensification. Proceedings of the National Academy of Sciences, 99(26): 16812-16816.

Michener, C. D. 2000. The bees of the world. The Johns Hopkins University Press, Baltimore, MD.

Mitchell, T. B. 1960. Bees of the eastern United States.Vol.1. Technical Bulletin No. 141, North Carolina Agric. Exp. Stn.

Mitchell, T. B. 1962. Bees of the eastern United States, Vol.2. Technical Bulletin No. 152, North Carolina Agric. Exp. Stn.

Nayak, G. K., S. P. M. Roberts, M. Garratt, T. D. Breeze, T. Tscheulin, J. Harrison- Cripps, I. N. Vogiatzakis, M. T. Stirpe, and S. G. Potts. 2015. Interactive effect of floral abundance and semi-natural habitats on pollinators in field beans (Vicia faba). Agriculture, Ecosystems and Environment. 199: 58–66.

72 Oksanen, J., F. G. Blanchet, R. Kindt, P. Legendre, P. R. Minchin, R. B. O'Hara, G. L. Simpson, P. Solymos , M. Henry, H. Stevens, and H. Wagner. 2016. vegan: Community Ecology Package. R package version 2.3-4. https://CRAN.R- project.org/package=vegan

Palmer, M. W. (n.d.). Ordination methods- an overview. Available at: http://ordination.okstate.edu/overview.htm

Payton, M. E., M. H. Greenstone, and N. Schenker. 2003. Overlapping confidence intervals or standard error intervals: what do they mean in terms of statistical significance? Journal of Insect Science. 3:34.

Pielou, E. C. 1966. The measurement of diversity in different types of biological collections. Journal of Theoretical Biolology 13: 131-144.

Pollinator Health Task Force. 2014. National strategy to promote the health of honey bees and other pollinators (The White House, Washington, DC).

Potts, S. G., J. C. Biesmeijer, C. Kremen, P. Neumann, O. Schweiger, and W. E. Kunin. 2010a. Global pollinator declines: Trends, impacts and drivers. Trends in Ecology and Evolution 25: 345–353.

Ricketts, T. H., J. Regetz, I. Steffan-Dewenter, S. A. Cunningham, C. Kremen, A. Bogdanski, B. Gemmill-Herren, S. S. Greenleaf, A. M. Klein, M. M. Mayfield, et al. 2008. Landscape effects on crop pollination services: are there general patterns? Ecology Letters 11: 499–515.

RStudio Team. 2015. RStudio: Integrated Development for R. RStudio, Inc., Boston, MA URL http://www.rstudio.com/.

Sheffield, C. S., A. Pindar, L. Packer, and P. G. Kevan. 2013. The potential of cleptoparasitic bees as indicator taxa for assessing bee communities. Apidologie 44: 501-510.

Sidhu, C. S., and N. K. Joshi. 2016. Establishing wildflower pollinator habitats in agricultural farmland to provide multiple ecosystem services. Frontiers in Plant Science, 7 (363).

Somerville, D., 1999. Honey bees in cherry and plum pollination, NSW Agriculture Agnote DAI 126.

73 ter Braak, C. J. F., and P. Šmilauer. 2002. CANOCO reference manual and CanoDraw for Windows user’s guide: software for canonical community ordination, 4.5 edition. Microcomputer Power, Ithaca, NY.

Vaughan, M., E. Mäder, J. K. Cruz, J. Goldenetz-Dollar, K. Gill, and B. Borders. 2014. Hedgerow Planting (422) for Pollinators, Pennsylvania Installation Guide and Job Sheet. The Xerces Society for Invertebrate Conservation. Available at: http://www.xerces.org/wp- content/uploads/2013/01/InstallGuideJobSheet_Pennsylvania_Hedge.pdf

Williams, N. M., K. L. Ward, N. Pope, R. Isaacs, J. Wilson, E. A. May, J. Ellis, J. Daniels, A. Pence, K. Ullmann, and J. Peters. 2015. Native wildflower plantings support wild bee abundance and diversity in agricultural landscapes across the Unites States. Ecological Applications 25(8): 2119-2131.

Winfree, R., N. M. Williams, J. Dushoff, and C. Kremen. 2007. Native bees provide insurance against ongoing honey bee losses. Ecology Letters 10(11), 1105-1113.

Wratten, S. D., M. Gillespie, A. Decourtye, E. Mader, and N. Desneux. 2012. Pollinator habitat enhancement: benefits to other ecosystem services. Agriculture, Ecosystems, and Environment 159:112–122.

CHAPTER 5

CONCLUSIONS

The information provided in this thesis poses more new questions. There is a need for policy change in order to protect pollinators and conserve their ecosystem services.

Standardized toxicity testing protocols that reflect field realistic exposure by using formulated product, examining all life stages, and using long term studies to observe sublethal effects are necessary. To truly understand the affect a pesticide has on an organism, field scale studies must be performed. When registering a pesticide, field scale studies are required to prove its efficacy, and differential efficacies have been observed in the laboratory versus the field (Cox et al. 1995). We must adapt this mindset when considering the toxicity of pesticides, and stop relying on laboratory data to determine the effects of pesticides on pollinators.

Also, consistent reporting protocols by the US-EPA that would require reporting the slope of dose mortality curves and confidence intervals will help researchers to accurately assess the toxicity of these products using results from their own field research and bioassays (Robertson et al. 2007). Lastly, diversifying the bee species researched is essential if we want to conserve the declining wild bee populations in addition to honey bees, as research suggests bee species may respond different to the same dose of pesticide

(Biddinger et al. 2013).

75 Neonicotinoids are currently a necessary tool for IPM in Eastern United States apple orchards. They replaced the unsafe organophosphates and carbamates that were phased out, and neonicotinoids are the only product available to combat pests such as

Rosy Apple Aphid and Brown Marmorated Stink Bug. The ban on neonicotinoids in the

UK has cost farmers nearly $33 million in pest damage (Scott and Bilsborrow 2015). We cannot ban our only effective product, but we must rather look towards management strategies to use neonicotinoids safely. We have proven that it is possible to reduce the risk of pollinators to neonicotinoids with proper spray timing (Chapter 2). We have also shown that all neonicotinoids do not produce the same toxic response in bees (Chapter 3).

Using the least toxic neonicotinoids and spraying responsibly can significantly reduce the risk to pollinators.

The value of managed pollinator habitat needs to be examined and research prioritized in the context of landscape, as not all areas of the nation may need this type of enhancement. Habitat enhancement is expensive and its objective needs to be clarified.

Research suggests that habitat enhancement will not conserve all bee species, but only those that pollinate the specific plants used in the habitat (Kleijn et al. 2015). Our study confirms that in PA pollinator plantings, few rare species were found (Chapter 4). These pollinator plantings may not be a viable source of rare bee species conservation, but with some consideration it is possible to use the pollinator habitats to enhance pollination in a specific cropping system. The research in this thesis indicates that planting such pollinator habitat is redundant around most PA apple orchards and that priority should be given to areas of intensive agricultural areas consisting of mostly monocultures. Federal money would be better spent in our area to conserve existing fencerows and forested

76 habitat adjacent to apple orchards. While land enhancement may work for some pollinator species (pasturing honey bees between crop pollination seasons), not all pollinator species may benefit from a mix of only a dozen plant species that may be a poor substitute for unmanaged forest edges which may have over a 100 plant species

(Kammerer et al. 2015).

Because of our cultural demands, it is impossible to completely eliminate pesticide use or habitat destruction to protect managed and wild pollinators. There are, however, simple changes that can be made to protect pollinators in agriculture using

Integrated Pest and Pollinator Management (Biddinger and Rajotte 2015). We must focus on maintaining our agricultural practices in a way that minimizes impacts to pollinators, in the same way that IPM attempts to minimize negative impacts on other beneficial insects. Including pollinator protection in IPM education, encouraging cost share programs such as the USDA/NRCS, and providing pollinator safety education to consumers are some of the simple changes we can make to conserve wild and managed pollinators.

77 Literature Cited

Biddinger, D.J., and E. G. Rajotte. 2015. Integrated pest and pollinator management- adding a new dimension to an accepted paradigm. Current Opinion in Insect Science 10: 204-209.

Cox, D. L., A. L. Knight, D. J. Biddinger, J. A. Lasota, B. Pikounis, L. A. Hull, and R. A. Dybas. 1995. Toxicity and field efficacy of avermectins against codling moth (Lepidoptera: Tortricidae) on apples. Journal of economic entomology 88(3), 708-715.

Robertson, J. L., R. M. Russell, H. K. Preisler, and N. E. Savin. 2007. Bioassays with Arthropods. Boca Raton: CRC Press. 199 p.

Scott, C., and P. Bilsborrow. 2015. An interim impact assessment of the neonicotinoid seed treatment ban on oilseed rape production in England. Available at: http://www.fbspartnership.co.uk/documents/Interim%20Assessment%20of%20N eonicotinoid%20Ban%20on%20Oilseed%20Rape.pdf