ABSTRACT Title of Thesis: THE MARYLAND NATIVE POLLINATOR SURVEY: COMBINING CITIZEN SCIENCE WITH A SPECIMEN COLLECTION TO DETERMINE THE FLORAL PREFERENCES OF MARYLAND’S NATIVE POLLINATORS Olivia Marie Bernauer, Master of Entomology, 2017 Thesis Directed By: Professor Dennis vanEngelsdorp Department of Entomology This citizen science project collected data on pollinator visits to flowering plants. Participants were trained, received educational materials, and identification skills were measured with pictorial quizzes. Participants learned and retained identification skills. Bee specimens were collected to compare community composition between observational and collected specimens; community composition differed between data collection methods. The data generated in the citizen science project were used to calculate a pollinator attractiveness score (PAS) for 23 plant species using five metrics: bloom length, visitation rate, pollinator richness and evenness, and specialist value. The field season was divided into three portions: ‘early’, ‘middle’, and ‘late’. In the ‘early’ season Asclepias syriaca scored highest and Thymus vulgaris lowest, in the ‘middle’ season, Asclepias incarnata scored highest, and Thymus citriodorus lowest. Symphyotrichum novae-angliae scored highest in the late season and Rudbeckia triloba the lowest. PAS provides a framework for assessing the relative plant attractiveness to pollinators. THE MARYLAND NATIVE POLLINATOR SURVEY: COMBINING CITIZEN SCIENCE WITH A SPECIMEN COLLECTION TO DETERMINE THE FLORAL PREFERENCES OF MARYLAND’S NATIVE POLLINATORS by Olivia Marie Bernauer Thesis submitted to the Faculty of the Graduate School of the University of Maryland, College Park, in partial fulfillment of the requirements for the degree of Master of Entomology 2017 Advisory Committee: Assistant Professor Dennis vanEngelsdorp, Chair Associate Professor Paula Shrewsbury Professor Mike Raupp © Copyright by Olivia Marie Bernauer 2017 ii Dedication I dedicate this work to those closest to me, who have supported me tirelessly over the past few years. Mom and dad, thank you for your endless love and encouragement throughout my academic career. Thank you for believing in me and for allowing me to pursue my dreams. I would not be where I am today without you two as my role models. iii Acknowledgements First, I would like to thank my advisor, Dennis vanEngelsdorp. I have enjoyed being a member of the bee lab and am thankful for the endless opportunities afforded to me to learn about both native bees and honey bee keeping. I am thankful for my committee members, Dr. Mike Raupp and Dr. Paula Shrewsbury, for their willingness to serve on my committee and for their support and advice throughout each stage of this project. I am appreciative of your collective mentorship, specifically in the extension and education aspect of my research. I also want to sincerely thank each of my citizen science participants. A big thank you to the Maryland Master Gardener program and to Jon Traunfeld, the Maryland Master Gardener coordinator for his help organizing volunteers. This research would not have been possible without the countless hours participants spent in the hot sun, and for that I am endlessly grateful. This project was more successful than I could have ever imagined and would not have been possible without your dedication and enthusiasm for pollinators. I also want to thank my funding source, an eIPM USDA NIFA grant, for making this research possible. Karen Rennich and Heather Eversole, thank you for running an organized lab and for coordinating the undergraduate work force. I want to thank the undergraduates who helped to collect data for this project, in no particular order: Judy Joklik, Laura Tiffany, Emily Starobin, Max O’Grady, and Meg Wickless. Thank you to Nathalie Steinhauer for being eager and willing to help with statistical analyses. I am thankful for your expertise and continuous desire to learn and teach. iv I want to thank Sam Droege, Gene Scarpulla, and the technicians at the USGS BIML lab for their help with bee species identification. I want to say thank you to the friends I have grown close to the past few years. I will forever treasure the memories we have made. Thank you to Morgan Thompson, Kelly Kulhanek, Andrew Garavito, Nathalie Steinhauer, and Jackie Hoban for your support, smiles, and laughs. To Meghan McConnell, thank you for taking me under your wing and sharing with me your love for Maryland. I am grateful for your friendship, and am glad to have shared many of the graduate school “firsts” with you. I look forward to seeing where life takes each of us and for more adventures to come. v Table of Contents Dedication .................................................................................................................... ii Acknowledgements .................................................................................................... iii List of Tables ............................................................................................................. vii List of Figures ........................................................................................................... viii Chapter 1: The Maryland Native Pollinator Survey: a useful method for citizen scientists to survey pollinators visiting floral resources .......................................... 1 Abstract: ..................................................................................................................................... 1 Introduction: ............................................................................................................................... 1 Methods: .................................................................................................................................... 5 Citizen Science - Volunteer selection ............................................................................... 5 Citizen Science – Training ................................................................................................ 6 Citizen Science -Exit survey ............................................................................................. 9 Specimen collection ........................................................................................................ 10 Data preparation and analysis ......................................................................................... 12 Results: ..................................................................................................................................... 15 Data preparation .............................................................................................................. 15 Citizen science - Training ............................................................................................... 16 Citizen science – Exit survey .......................................................................................... 20 Citizen science – pollinator observations ........................................................................ 22 Specimen collection ......................................................................................................... 24 Discussion ................................................................................................................................ 29 Future directions .............................................................................................................. 34 Chapter 2: An Assessment of the Attractiveness of Pollinator Plants Using a Combination of Citizen Science Observational and Hand-Collected Specimen Data ............................................................................................................................ 36 Abstract: ................................................................................................................................... 36 Introduction: ............................................................................................................................. 36 Methods: .................................................................................................................................. 40 Data preparation: ............................................................................................................. 40 Plant attractiveness metrics ............................................................................................. 41 Data analysis: .................................................................................................................. 44 Results: ..................................................................................................................................... 44 Data preparation .............................................................................................................. 44 Plant attractiveness .......................................................................................................... 46 Discussion: ............................................................................................................................... 48 Appendix A: Demographic surveys of citizen science participants ...................... 51 2015.......................................................................................................................................... 51 2016.......................................................................................................................................... 55 Appendix B: Monitoring and Pocket Guides ......................................................... 59 2015.......................................................................................................................................... 59 Monitoring guide ............................................................................................................. 59 Pocket guide .................................................................................................................... 94 2016.......................................................................................................................................... 98 Monitoring guide ............................................................................................................. 98 Pocket guide .................................................................................................................. 158 vi Appendix C: List of standard and supplemental plants for participants to observe upon ............................................................................................................ 165 Appendix D: Citizen science pollinator monitoring data sheets ......................... 166 2015........................................................................................................................................ 166 2016........................................................................................................................................ 167 Appendix E: Exit survey questionnaires .............................................................. 169 2015........................................................................................................................................ 169 2016........................................................................................................................................ 170 Appendix F: Lists of pollinator species by functional group .............................. 172 2015........................................................................................................................................ 172 2016........................................................................................................................................ 174 Appendix G: List of species collected in the 2016 specimen collection .............. 176 References ................................................................................................................ 181 vii List of Tables Table 1.1: Periods used to categorize observations and hand-collected specimens for comparison. ………………………………………………………………………….13 Table 1.2: Citizen science observational data summary from 2015 and 2016. ……..17 Table 1.3: Participants were quizzed at the beginning and end of their training sessions with pictures of pollinators with an improvement in score demonstrating participants learning to identify pollinator functional groups. All participants in both years improved their quiz score after the training session indicating that the training sessions were successful at teaching participants to identify pollinator functional groups. Even returning participants (2nd year participants) in 2016 improved quiz scores after the initial training session, showing that participants of all previous skill levels were benefitting from the training session. In 2016 there was a follow-up training session conducted 6-8 weeks after the initial training session where participants were given another picture quiz and a quiz of pinned pollinator specimens to identify. Both first- and second-year participants scored similarly to their post-quiz score from the initial training session demonstrating that participants were able to retain the pollinator knowledge that they learned at the initial training session. First year participants averaged 59% on their specimen quiz while second year participants averaged 6. ………………………………………………………...18 Table 1.4: New county records of bee species collected in the 2016 specimen survey. These records come from both hand-collected and bowl trap specimens. ………….26 Table 1.5: SIMPER analysis results indicated the average abundance of each functional group of bees observed or hand-collected per two-week period (see Table 1.1) and the percentage that each functional group contributes to the community composition differences between data types (observational vs. hand-collected specimen data). Honey bees and bumble bees contributed the most to the differences between the community compositions of the observed and hand-collected data...26-27 Table 2.1: A ranking of the 23 plant species assessed as follows. Each plant was assessed with five different metrics that were summed and turned into a percentage. The percentage was used to rank each plant within each of the three portions of the field season. Plants that are not native to Maryland are bolded………………… 46-47 viii List of Figures Figure 1.1: Three different colored bowls were used to collect bee species in pollinator gardens; bowls were left in the garden for 24 hours. Together, fluorescent yellow (a), fluorescent blue (b) and white (c) bowls effectively sample the entire bee community (Droege et al. 2010). Using this method to collect bees in addition to hand-collection will effectively sample the entire bee community present in an individual garden. ………..………………………………………………………… 11 Figure 1.2: The locations where native bees and butterflies were monitored in the Maryland Native Pollinator Survey in both years. Sites that were only monitored in 2015 (n=6), are marked in red, 2016 in light blue (n=19) and sites monitored in both years are marked purple (n=6). The locations where bee specimens were collected are denoted by a black box with an “X” through it (n=4). Prince George’s county did not submit observational data for their demonstration garden even though bee specimens were collected there, indicated by the specimen site lacking a corresponding field site circle of any color. …………………………………………………………………. 19 Figure 1.3: Participant effort, in minutes spent observing, was high at the beginning of the field season, but declined over time in both years (a. & b.). Raw bee abundances in both years decline similarly to the reduction in participant effort (c. & d.). Standardizing observations by effort (e. & f.) allows for a comparison of data over the season and between years. Even after adjusting for effort we document a steep decline in bee abundance per minute effort in period 9 (e.) and period 7 (f.)... 24 Figure 1.4: 2016 hand-collected bee specimen data standardized by effort in minutes. There is no late season decline in bees per minute as experienced in both seasons of observationally collected data. Hand-collected specimen data from period 10 is not available due to a lack of flowering resources in the monitored gardens at this time 28 Figure 1.5: Relationships between the observed and hand-collected data collected in 2016. All data points are standardized by plant, period (see Table 1.1), and time. Data presented are from the three locations where both observational and hand-collection specimen data are available: Baltimore, Allegany and Frederick counties. Based on Spearman rank calculations, strong relationships were found for the honey bees (a; rs=0.617) and bumble bees (b; rs=0.781)………………………………………….. 29 Figure 1.6: In Period 9, participants observed (a) 0.23 “bumble bees” per minute and 0.02 “large carpenter bees” per minute as opposed to the 0.07 “bumble bees” per minute and 0.18 “large carpenter bees” per minute hand-collected (b). These differences are likely due to participants misidentifying the similar-looking “bumble bees” (c) and “large carpenter bees” (d) late in the field season. In the early fall, bumble bee colonies have finished producing reproductive gynes and workers are no longer foraging to support the colony, thus bumble bee observations should decrease at this time. In contrast, juvenile large carpenter bees forage in the fall to store up energy to survive the winter months. Hand-collected specimen data from period 10 is not available due to a lack of blooming floral resources at the monitored sites. Photos c. & d. courtesy of John Baker………………………………………………….. 34-35 Figure 2.1: The pollinator networks constructed from the plant records used to create recommendations for the early (a), middle (b), and late (c) portions of the field season. The plant species are on the left and the pollinator functional groups are on ix the right. The width of the grey bars corresponds to the number of records between the plant and pollinator; when this value is small, the grey bars become thin and may be black. Nonnative plant species are bolded………………………………….…… 48 1 Chapter 1: The Maryland Native Pollinator Survey: a useful method for citizen scientists to survey pollinators visiting floral resources Abstract: Citizen science research projects have the potential to gather large amounts of data in a quick and economical manner. Data generated by citizen science research projects can provide answers to large-scale ecological questions, and engage and enlighten citizens about the scientific method and current environmental issues. This citizen science project aims to create a method to collect accurate data on the identity of pollinators visiting specific flowers. To improve the accuracy of the citizen scientist’s observations, they were trained to identify floral visitors to functional groups. Participants who attended a training session received educational materials for field identification purposes. The impact of these training sessions was evaluated using a pictorial pre- and post-quiz of native pollinators. In the first year, 2015, volunteers observed functional groups of native bees and in 2016 functional groups of butterflies were added. In year two, a follow-up training occurred where questions about data collection and pollinator identification were answered and a specimen and additional pictorial quizzes were administered. To further assess the efficacy of this data collection method, bee specimens were collected using two methods: hand-collection and bowl trapping, was conducted in the second field season. For both years, volunteers scored higher on the post-quiz than they did on the pre-quiz. The volunteers retained the identification knowledge they learned at the initial training into the follow-up training session. When the community composition of the observational data and hand-collected specimen data were compared using a PERMANOVA test, it was discovered that the community compositions were different between the two data collection methods. This difference was evaluated using a SIMPER analysis which determined that the “honey bee” and “bumble bee” functional groups were contributing the most to the community differences (17.48 and 15.43%, respectively). Despite these community differences, the observational data generated by this study can still prove useful for drawing conclusions about pollinator floral preference. Introduction: Citizen scientists, members of the public involved in organized research efforts (Dickinson and Bonney 2012), can generate data that can provide valuable answers to socially significant, landscape-level ecological questions by amassing large datasets that were previously labor or cost-prohibitive (Dickinson et al. 2010). While the generation of data for research purposes benefits scientists, citizens participating in informal science also benefit by improving their scientific literacy and knowledge, and as a result tend to become better environmental stewards (National Research Council et al. 2009; Bonney 2 2009; Crall et al. 2013; Domroese and Johnson 2017). This valuable method of data collection has become increasingly popular since Irwin et al. (1994) coined the phrase “citizen science” and its versatility has been further revitalized by technological advancements (Triezenberg et al. 2012; Dickinson et al. 2012). The concept of citizen science far outdates the recent advent of this terminology; members of the public have long played an important role in collecting ecological data (Triezenberg et al. 2012; Dickinson et al. 2012). The first citizen science project in the United States began in 1880 when lighthouse keepers began recording which bird species struck their windows (Dickinson and Bonney 2012). Since then, many other successful citizen science projects on birds have evolved to include: the long-running National Audubon Society’s Christmas bird counts, initiated in 1900 and continues today in the United States, Canada, and Central and South America (LeBaron 2016); the Cornell Lab of Ornithology’s Project FeederWatch (Bonter 2012); and Neighborhood Nestwatch (Evans et al. 2005). Birds make high quality subjects for citizen science research as they are diurnal, charismatic, frequently encountered and already observed by many ‘bird watchers’ (Sullivan et al. 2009). Aside from monitoring bird populations, citizen science has proven useful in understanding landscape-level environmental trends and in documenting population change and size, in several types of organisms including plants and insects. Project Budburst seeks to gather information on the impact climate change has on plants by using citizen scientists to monitor specific events in both native and non-native plant life cycles throughout the year (Meymaris et al. 2008; Henderson et al. 2012). In the United Kingdom, citizen scientists across the landscape monitored the pollination services 3 received by the fava bean, Vicia faba L., by assessing changes in plant yield after manipulating the pollination environment to either receive hand pollination, insect pollination, or prevent pollination by excluding insects (Birkin and Goulson 2015). The Monarch Larva Monitoring Project is a good example of the use of citizen scientists to monitor population size and trends. This project recruits citizens to monitor monarch eggs and larvae during the breeding season (Kountoupes and Oberhauser 2008). Similarly, citizen scientists in France photographed floral visitors in a variety of land-use types to determine how land-use change is affecting pollinator populations (Deguines et al. 2012). While citizen science can collect valuable ecological data in a variety of contexts, there are some limitations to this method. First, citizen scientists are amateur naturalists and therefore are prone to making mistakes (i.e. incorrectly identifying organisms). To counteract this, volunteers are often given specific training on how to collect data for individual projects. Training sessions are intended to help improve the data quality from the start of the project. To further validate the data quality throughout the project, many citizen science projects have a mechanism in place to double check submitted data. For instance, in the Buckeye Lady Beetle Blitz (Gardiner et al. 2012), participants attended an in-person training session where they received a lady beetle identification guide, protocol guidelines and equipment for data collection. After completing a training session, participants collected and identified lady beetles. To check the accuracy of the lady beetle identifications they were verified by entomologists. Similarly, when pollinator populations were surveyed photographically in France, these photographs were checked 4 by trained entomologists, however, in contrast, participants in this study received no formal training ahead of time (Deguines et al. 2012). A further challenge of working with insects in general, and pollinators specifically, is that they are often difficult to identify without extensive training. The identification of individual pollinators to the species level often requires specialized taxonomic skills (Hopkins and Freckleton 2002), to make pollinator identifications easier, functional groups, scientifically-meaningful classification of organisms that share similar characteristics within a community (Stanley 1979), can be used. Functional groups can describe organisms either based on physical characteristics (i.e. metallic coloration) or their ecological role in the environment (i.e. predator). Using functional groups simplifies the pollinator identification process for the citizen scientists. The Xerces society, an international, nonprofit organization striving to protect and conserve invertebrates, focuses some of their current conservation effort on pollinators. In 2007, the Xerces society initiated the ‘California Pollinator Project’, a citizen science project monitoring bee diversity and abundance over time in Yolo County, CA (Ullmann et al. 2007). Participants in the ‘California Pollinator Project’ were instructed to walk transects in meadows, hedgerows, or crops, at least three times per season and record all bees visiting the reproductive parts of flowers to one of ten functional groups. The work done by the Xerces society ultimately set the groundwork for the Maryland Native Pollinator survey, a citizen science project to monitor pollinator visitations to flowering resources. This project strives to determine if citizen scientists can collect reliable ecological data on pollinator visitation rates to floral resources. To reduce the potential for participant error, pollinators were organized into functional 5 groups based on visible physical characteristics. More obvious pollinators like the Monarch butterfly (Danaus plexippus L.) or the honey bee (Apis mellifera L.) are readily identified, and were not placed into a functional group. The focus of this chapter is twofold; first, to document the impact of training on citizen scientist ability to identify pollinators, and second, to quantify the accuracy of the in-field observations with a comparison to a hand-collected bee specimen collection. Methods: Citizen Science - Volunteer selection In 2015 and 2016, Master Gardener coordinators were contacted to recruit participants. Master Gardeners were the target participants for this survey because they already understand the basics of the scientific process and have experience in plant identification. Therefore, Master Gardeners only needed to learn pollinator identification skills to participate in this survey. Master Gardener groups that were contacted first had a strong member base or had independently reached out, expressing an interest in participating. Master Gardener coordinators were emailed a summary of the survey and were then asked to contact their Master Gardeners to gauge interest in the survey. If the Master Gardeners in that county expressed interest in participating, they were then included in the study for the upcoming field season. The goal was to have a total of three participating counties in the first year and six in the second year. In 2015 Prince George’s, Montgomery, and Baltimore counties participated in the survey and in 2016 Baltimore, Prince George’s, Baltimore city, Frederick, Allegany, and Howard counties participated. 6 Citizen Science – Training Training sessions were held in-person, once per county. At the beginning of the training session information on trainee demographics was collected using a paper survey attached to the pre-quiz questionnaire (Appendix A: Demographic surveys of citizen science participants). This survey collected information on their: gender, age, race, and occupation. Prior to the start of the training session, participants were asked to provide a self-reported score of pollinator knowledge from 1-5, where 1 indicated little to no knowledge of pollinators, and 5 indicated a thorough knowledge of pollinators. To assess if training sessions were successful in teaching potential participants to identify pollinator functional groups, pictorial quizzes were administered prior to and after the completion of each training session. The pictorial quizzes included photographs that the participants would not see during the training session. In 2015, the pictorial quiz included ten photos of bees and in 2016 the pictorial quiz included fifteen photos, seven of bees, seven of butterflies and one “other” floral visitor that participants were asked to identify to functional group. Each photo in the pictorial quiz was shown to the participants for approximately two seconds to simulate in-the-field observations. To enhance the training sessions and aid in pollinator identification, volunteers were provided with both a monitoring guide and pocket guide (Error! Reference source not found.). The monitoring guide included information on the importance of pollinators, insect anatomy and defining characteristics, the individual pollinator functional groups, and how to collect and submit data. A companion Pocket Identification Guide was developed to serve as an in-field reference guide for identifying observed pollinators. The 7 monitoring guide, pocket guide, and data sheets were based on the ‘Pennsylvania Native Bee Survey’ (Ullmann et al. 2007; Donovall and vanEngelsdorp 2009). The 2015 field season focused solely on native bee functional groups, similar to what was done in Pennsylvania (Donovall and vanEngelsdorp 2009). In 2016, improvements were made to the monitoring and pocket guides based upon feedback from the 2015 exit survey as outlined in the results. Butterfly functional groups were monitored in addition to bee functional groups in the second field season. The 2015 guides were modified to reflect this addition and the survey was appropriately renamed the ‘Maryland Native Pollinator Survey’. To properly identify different pollinators to functional groups, participants were first taught basic insect anatomy. Important insect characteristics and body part modifications were discussed (i.e. location of scopa or pollen basket in bees, the presence of wing tails in butterflies). Then, each pollinator functional group was examined in detail, highlighting their distinguishing characteristics. Because there are some insects that closely mimic certain pollinators (i.e. some wasps in the family Chrysididae mimic green sweat bees in the family Halictidae), a special emphasis was placed on teaching volunteers to differentiate similar-looking insects. Volunteers were next taught the standardized methods for collecting data. To track volunteer data submissions throughout the field season, each participant was given a unique identification number. Participants were asked to collect observational data four times per month: twice in the morning and twice in the afternoon. These observations could take place on the same day or on different days, but were requested to occur at roughly the same time each month (e.g. the 5th and 15th of every month). Volunteers 8 began collecting data as soon as they received training and were instructed to collect data until the first frost in the fall, typically in late October or early November. Participants were instructed to choose six to eight in-bloom plant species to observe pollinators on during a single observation period. When possible, these plants should include as many of the standard plant genera or species as possible. In 2015, the standard plants included: Agastache, Asclepias, Asters, Echinacea, Helianthus, Monarda, Oregano, and Thyme. These plants were chosen because collectively their bloom times span the field season. In 2016, Rudbeckia and Solidago were added to the standard plant list to bolster the number of late-season blooming plants. Suggestions for supplementary plant species to observe upon were also discussed during training sessions (Appendix C: List of standard and supplemental plants for participants to observe upon). Volunteers were asked to observe each different type of plant for ten minutes and to record all the insects that visited the reproductive parts of the flowers during this time. Insects of the plant leaves, stems, or petals were not recorded. If a pollinator that was not a bee or a butterfly arrived on the flower, or if a bee or butterfly was unidentifiable, volunteers recorded this floral visitor in the ‘Other’ category on the data sheet and a description of the insect was included in the notes section of the data sheet. During each individual observation period, participants completed a data sheet to record environmental conditions and pollinator observations (Appendix D: Citizen science pollinator monitoring data sheets). They recorded environmental conditions that included: the temperature (50s, 60s, 70s, 80s, 90s, 100+ °F), wind (still, light breeze, windy, gusty), and cloud cover (clear, partly cloudy, mostly cloudy, overcast). Additional blooming floral resources and significant environmental changes (i.e. new plants planted, 9 a large rainstorm caused erosion in the garden, etc.) were also noted. The plot size (i.e. 1/3 of an acre, 50 sq. ft., etc.) and number of individuals collecting data during the observational period were recorded. After completing an observational period, volunteers were instructed to submit their datasheets. In 2015, all data sheets were submitted via paper mail or electronically scanned and emailed. In 2016, participants also had the option to submit data via a Google form. During the 2016 field season an additional follow-up training was held 6 to 8 weeks after the first training. All those who attended the first training were welcome to attend the follow-up training, however, generally only those who had already begun collecting data or were planning on collecting data attended. Volunteers were asked to reflect upon: how the observations were going, what was working well for them, what they were having trouble with, and any other issues surrounding the collection of data. After this guided discussion, the volunteers were walked through a very quick pictorial review of the pollinator functional groups covered in the initial training session. After this review, another pictorial post-quiz, the same one conducted in the initial training session, and a specimen quiz were proctored. The specimen quiz consisted of fourteen pinned pollinators that participants identified to functional group. Citizen Science -Exit survey After the end of each field seasons, participants were sent an exit survey to gather additional information about their experiences, suggestions for survey improvement, and whether participants were likely to return for another season of pollinator observations (Appendix E: Exit survey questionnaires). In 2015, only participants that collected data were given the opportunity to respond while in 2016 the survey was offered to all 10 volunteers who were trained. This provided feedback on the training sessions from both those who did and did not participate in the survey. The exit survey questions in 2016 were adapted from the 2015 exit survey. The open-ended questions asked in 2015 provided more insight and therefore, the 2016 exit survey included more open-ended questions. In 2016, volunteers were also asked to self-report their score of pollinator knowledge again after the field season was over. Specimen collection To determine how closely the observational data reflected true values, a native bee specimen collection was implemented in the 2016 field season. Bees were collected in the Master Gardener demonstration gardens in four of the counties (Allegany, Frederick, Baltimore, and Prince George’s counties) where participants performed their observations every two weeks throughout the field season. The bee specimen collection took place on the same schedule. Bees were collected using two methods, hand-collection and bowl trapping. Both methods for collecting bees were conducted within the same 24- hour period for each sampling event. Hand-collection took place on the same list of standard plant species that the Master Gardeners were instructed to observe. Bees were hand-collected on six to eight plant species per collection period both in the morning and again in the afternoon. Bees were hand-collected for ten minutes per blooming plant species. The collected bees were placed into a 50-mL FalconTM plastic tube (Fischer Scientific, Pittsburgh, PA) filled with soapy water to kill the specimens. A total of twenty-four bowl traps, eight of each fluorescent blue, fluorescent yellow, and white bowls, were placed throughout the Master Gardener demonstration 11 gardens approximately 3 meters apart (Error! Reference source not found..1). The bowls were filled with a mixture of dish soap and water and left to sit for 24-hours (Droege et al. 2010). After 24-hours, the bowl traps were collected, the contents were strained through a 190-micron fine-mesh paint filter (Pro Marine Supplies, Venice, FL) and placed into a Whirl-Pak® (Nasco, Fort Atkinson, WI) for storage until processing. Figure 2.1: Three different colored bowls were used to collect bee species in pollinator gardens; bowls were left in the garden for 24 hours. Together, fluorescent yellow (a), fluorescent blue (b) and white (c) bowls effectively sample the entire bee community (Droege et al. 2010). Using this method to collect bees in addition to hand-collection will effectively sample the entire bee community present in an individual garden. Hand-collecting bees directly from flower species creates records of foraging preference at the species level while bowl-trapping helps to characterize the whole bee community. It is important to use both methods together to fully sample a bee community as some larger bees, like the eastern carpenter bees (Xylocopa virginica L.), are big enough that they can pull themselves out of the bowl traps and avoid being collected (Droege et al. 2010). Because the bowl traps are placed in the garden for a 24-hour period, they can collect other bee species that may forage earlier or later in the day than when the hand-collection took place. 12 We did not collect bees that were easily identified in the field (i.e. honey bees – A. mellifera, or eastern carpenter bees – X. virginica). If there was any uncertainty about a bee’s identity, it was always collected. Sometimes bees were missed in the field (i.e. the bee evaded captured) and these bees were recorded to their corresponding functional groups (i.e. “small dark bee”). Once in the lab, the soapy water was rinsed from the both the bowl trap and hand- collected specimen samples. Once rinsed, each sample was placed in a plastic Whirl- Pak® (Nasco, Fort Atkinson, WI) bag filled with enough 70% ethanol to completely cover all bees in the sample. Samples were stored at 5°C until further processing could occur. To identify many bee species, taxonomists rely heavily on the bee’s hair color or patterns. To facilitate identification, bees were removed from alcohol, washed, dried, and pinned (Sam Droege 2015). Once pinned, location labels were generated on discoverlife.org, printed on appropriate paper (acid-free, archival paper; BioQuip, Rancho Dominguez, CA), and affixed to pinned specimens. Once all specimens were pinned and labeled, they were subsequently identified. Bee identifications were confirmed or corrected by Sam Droege at the USGS Bee Inventory and Monitoring Lab in Beltsville, Maryland Data preparation and analysis Data submitted by participants was entered into a spreadsheet and the data was prepared prior to conducting statistical analyses. The contents of the ‘other’ floral visitor column, an open-ended column, with space for notes, was reviewed before data analysis and either disregarded (i.e. observation was not a pollinator: “grasshopper”) or moved to 13 its respective functional group (i.e. “fritillaries” were moved to “brushfoot butterflies”). If a volunteer submitted fewer than four data sheets, or one complete month of observations, during a field season, their observations were not used for data analysis. This data was removed to ensure that the analyzed data came from sites with more than a few data points, providing a bigger-picture view of the pollinator populations. Because the observational data and hand-collected specimen data were not always collected on the same date, the observational and hand-collected specimen data were portioned into ten two-week periods (Table 1.1). Table 1.3: Periods used to categorize observations and hand-collected specimens for comparison. Period Start date End date 1 6/1 6/15 2 6/16 6/30 3 7/1 7/15 4 7/16 7/31 5 8/1 8/15 6 8/16 8/31 7 9/1 9/15 8 9/16 9/30 9 10/1 10/15 10 10/16 10/31 In 2015, participants were asked to record the time they started and completed observing for a given observational period, but were not asked to record when they started and completed observing on a specific plant. Therefore, when standardizing bee observations by time, it was assumed that observers spent 10 minutes per plant, as instructed in the training sessions. In 2016, participants were asked to record both the observational period and specific plant observational period start and end times. If 14 participants forgot to record the start or end time of their observation on a plant, it was assumed that ten minutes was spent observing this plant. All statistical analyses were conducted in R version 3.3.2 (R Core Team 2016). When comparing the bee functional groups from the hand-collected specimen data to the observed data, only data collected from the same garden were compared. To visually compare the changes in community composition over time between the observational and specimen collections, stacked area graphs were created (Wickham 2009). Next, the observed and hand-collected specimen data were standardized by time spent observing (bees/minute), paired by plant and period (i.e. Asclepias sp. in Period 3), and then were plotted by bee functional group (i.e. honey bees). The linear relationship between the data types and the associated Spearman rank correlation coefficients were calculated. To assess the differences in community composition between data collection methods, the adonis function in the package vegan was used (Oksanen et al. 2017). The adonis function runs a PERMANOVA test to analyze the similarity in community composition as determined by the Bray-Curtis similarity measure. This function was utilized to compare the difference in bee community composition at the functional group level between data types (observed vs. collected), sites (counties), and to see if there was an interaction between location and data type on community composition. To determine which bee functional groups contribute the most to differences between the observed and collected community compositions, a SIMPER analysis was conducted (Oksanen et al. 2017). SIMPER analyses looks at the similarity of the percentages of each member of a community, or in this case, each bee functional group, to determine where and by what 15 percentage do each functional group contribute to the overall community-level differences. In Prince George’s county, no observational data was submitted and therefore this data was not included in the PERMANOVA and SIMPER analyses. These analyses were only conducted the remaining three counties where both observational and specimen data were collected. The observational data was assessed for biological accuracy and corrected when appropriate. For example, the “other metallic bees” represent bees in the genus Osmia which complete their life cycle in early spring (Bosch and Kemp 2000) and should not be reported after June. After corrections were made, the PERMANOVA and SIMPER analyses were recalculated. Results: Data preparation In 2015, 13 locations were monitored by participants but 2 sites lacked sufficient data and so were removed from analyses. Similarly, in 2016, of the 25 sites where data was collected, 8 sites were removed. In total, participants conducted 156 observational periods in 2015 and 316 in 2016; however, observations sometimes began earlier or continued later in the season than the hand-collected specimens were. Observational data collected outside of the 10 periods (Table 1.1) were not included in the data analysis (n=37 observations; 1.81%). There were 23 (1.12%) instances in which participants in 2016 failed to record either the start or end time of their observations on a specific plant. In the hand-collected specimen data set, bees that were reported but not hand- collected, represent 31% (n=628) of the total number of bees. Almost half of the bees 16 reported but not hand-collected were the “large carpenter bees” with 4.4% (n=14) of their records coming from hand-collection events and an additional 305 records based on visual identification in the field. Honey bees and bumble bees were recorded but not hand-collected 42.6% (n=136) and 26.6% (n=134) of the time, respectively. Citizen science - Training In 2015, five 75-minute initial training session were conducted and 46 individuals attended (Table 1.2). Of the 46 trained participants, 21 (18 females, 3 males) submitted data at least once. The volunteers monitored native bees in 13 different locations (Error! Reference source not found.1.2). The average pre-training score was 37 ± 15.0%, and the post-training score averaged 55 ± 20.3%. The post-training quiz scores were higher than the pre-training quiz scores (mean improvement: 18 ± 20.0%, t44=-5.966, p<0.0001; Error! Reference source not found.). On average, 42 volunteers self-reported their pre- training pollinator identification knowledge as 2.7± 0.98 out of 5. 17 2015 2016 Data sheets submitted 156 316 Hours of observation 161 349.4 Number of pollinators observed 10,024 18,479 Bees 9,585 14,088 Butterflies n/a 1,974 Other 439 2,098 Most commonly observed bees Bumble bees (n=3,643; 36.3%) Bumble bees (n=4,037; 21.8% of total, 28.7% of bees) Most commonly observed butterflies Not observed in 2015. White & Sulphurs (n=694; 3.76% of total, 35.2% of butterflies) Table 1.4: Citizen science observational data summary from 2015 and 2016. 18 2015 2016 1st year participants 1st year participants 2nd year participants Initial training Pre-quiz (%) 37 ± 15.0 (n=45) 40 ± 15.7 (n=85) 64 ± 16.1 (n=11) Post-quiz (%) 55±20.3 (n=45) 59 ± 18.5 (n=83) 79 ± 18.0 (n=11) Paired t-test t44 = -5.966; p <0.0001 t78 = -10.821; p <0.0001 t10 = -5.036; p <0.0001 Follow-up training Post-quiz (%) n/a 73 ± 16.4 (n=12) 83 ± 13.3 (n=6) Paired t-test n/a t10 = -1.1016; p > 0.1 t5 = -0.25482; p > 0.1 Specimen quiz (%) n/a 59 ± 11.0 (n=12) 67 ± 8.7 (n=6) Table 1.3: Participants were quizzed at the beginning and end of their training sessions with pictures of pollinators with an improvement in score demonstrating participants learning to identify pollinator functional groups. All participants in both years improved their quiz score after the training session indicating that the training sessions were successful at teaching participants to identify pollinator functional groups. Even returning participants (2nd year participants) in 2016 improved quiz scores after the initial training session, showing that participants of all previous skill levels were benefitting from the training session. In 2016 there was a follow-up training session conducted 6-8 weeks after the initial training session where participants were given another picture quiz and a quiz of pinned pollinator specimens to identify. Both first- and second-year participants scored similarly to their post-quiz score from the initial training session demonstrating that participants were able to retain the pollinator knowledge that they learned at the initial training session. First year participants averaged 59% on their specimen quiz while second year participants averaged 79%. 19 Figure 1.2: The locations where native bees and butterflies were monitored in the Maryland Native Pollinator Survey in both years. Sites that were only monitored in 2015 (n=6), are marked in red, 2016 in light blue (n=19) and sites monitored in both years are marked purple (n=6). The locations where bee specimens were collected are denoted by a black box with an “X” through it (n=4). Prince George’s county did not submit observational data for their demonstration garden even though bee specimens were collected there, indicated by the specimen site lacking a corresponding field site circle of any color. During the 2016 field season, 105 volunteers participated in one of six two-hour training sessions. Of the 105 trained participants, 28 (23 females, 5 males) submitted data (Table 1.2). The volunteers monitored either bee functional groups, butterfly functional groups or both. Most (87.8%) volunteers chose to monitor both bee and butterfly functional groups. In 2016 volunteers monitored 25 sites in seven counties (Figure 1.2). Of the six counties that hosted initial training sessions, only Howard County did not submit any observational data. Two additional counties, Garret and Montgomery, were monitored in 2016 by volunteers. Both first year and second year participants had higher post-quiz scores (first-year: 59 ± 18.5% (n=83); second-year: 79 ± 18.0% (n=11)) when compared to pre-quiz scores (first-year: 40 ± 15.7% (n=85); second year: 64 ± 16.1% 20 (n=11) Error! Reference source not found..3). However, in 2016, the second-year participants scored higher than first-year participants in both the pre- (t12=-4.6875, p<0.001) and post-quizzes (t13=-3.4576, p<0.01). Collectively, 2016 participants self- reported their pre-training pollinator identification knowledge as 2.3 ± 0.87 out of 5 on average. The secondary pictorial-quiz was administered to 28 people at the follow-up training sessions. The scores of this quiz were slightly higher, although not significantly higher, or lower than the post-quiz scores at the end of the initial training session for both first- and second-year participants (first-year: t10 = -1.1016, p=0.2964; second-year: t5 = - 0.25482, p=0.809; Error! Reference source not found..3). The average score of specimen quiz presented to the volunteers at the follow-up training was 59 ± 11.0% (n=12) for first-year participants and 67 ± 8.7% (n=6) for second-year participants. Citizen science – Exit survey Six participants returned the 2015 exit survey. Most (4) volunteers found the initial training session, monitoring guide, and pocket guide helpful. When asked if they would participate again in 2016, two thirds of respondents indicated that they were likely to. Survey respondents made several suggestions to improve the survey, including a desire to have more clear and intuitive bee functional groups, more variety in the monitoring guide pictures, greater transparency about time commitment needed to participate, and better methods to record and submit data (i.e. a fillable PDF or online form). Feedback from the 2015 participants indicated that some bee functional groups needed renaming or restructuring (Appendix F: Lists of pollinator species by functional 21 group). For clarity, some of the bee functional groups were renamed: the “hairy-leg bees” became the “long-horned bees”, the “green sweat bees” became the “green metallic bees”, and the “metallic hairy belly bees” became the “other metallic bees”. Participants also indicated difficulty in distinguishing the “dark bees” from the “dark hairy belly bees”. These groups were distinguished by determining where the bee carries pollen, either on the underside of their abdomen as in the “dark hairy belly bees” or in another in location. To assist with dark bee identification, in 2016, the dark bees were no longer broken up into functional groups based on where they carried pollen but by size: “large dark bees” (>1 cm) and “small dark bees” (<1 cm). If volunteers were well-versed in bee identification they had the option to still distinguish between large and small dark bees that did and did not carry pollen on the underside of their abdomen. Finally, the functional group containing kleptoparasitic bees known as the “cuckoo bees” were removed from the survey in 2016. This group of bees was rarely observed in 2015 (n=11, 0.11% of total observations) and participants reported having difficulty identifying this bee functional group as they were unsure if they could distinguish “cuckoo bees” from wasps. Twenty-nine volunteers responded to the 2016 exit survey, 21 of which submitted data sheets during the field season. Some respondents (8), who attended one or both training sessions but did not participate in data collection indicated that they did not feel confident with their level of pollinator identification following the training sessions. When asked about the initial training session, 23 respondents found the initial training session helpful while 4 respondents either did not or did not find it sufficient for confident functional group identifications. The follow up training was evaluated by 16 22 respondents, 15 of which felt that the training session was helpful. Participants (27) found the monitoring guide helpful for classifying floral visitors to functional group. Most (22) indicating that they referred to it often while conducting observations and in-between observations. Similarly, 23 respondents indicated that the pocket guide was useful and was frequently used during observations. When asked about future participation, 19 of 29 respondents indicated that they were likely to participate again. The data generated in both years of this survey can provide insights into local pollinator communities that may be useful to Master Gardeners, as they frequently give educational talks or present at outreach events. When asked if they would use the results of this survey for educational purposes, 3 of 6 respondents in 2015 indicated they would use the collected data and in 2016, 13 of 29 respondents indicated that they had an interest in using the results from this survey. Citizen science – pollinator observations In the 2015 field season, a total of 156 data sheets were submitted documenting 10,024 pollinator observations. This represented 161 hours of observations. Generally, as the season progressed participants spent less and less time observing floral visitors (Error! Reference source not found.a.). The most commonly observed group of bees were “bumble bees” (n=3,643 or 36.3% of all observations). The “other” category accounted for 439 (4.38%) of the observations and included unidentifiable bees, butterflies, moths, skippers, wasps, beetles, flies, and hummingbirds. 23 0.0 0.5 1.0 1.5 1 2 3 4 5 6 7 8 9 10 Period Vb t 0 250 500 750 1000 1250 1 2 3 4 5 6 7 8 9 10 Period su m va lue 0 250 500 750 1000 1250 1 2 3 4 5 6 7 8 9 10 Period su m va lue 0.0 0.3 0.6 0.9 1 2 3 4 5 6 7 8 9 10 Period Vb t 200 400 600 800 1 2 3 4 5 6 7 8 9 10 Period su m tim e 0 500 1000 1500 1 2 3 4 5 6 7 8 9 10 Period su m tim e Ef fo rt ( mi n. ) Ef fo rt ( mi n. ) Be e a bu nd an ce (in div idu als ) Be e a bu nd an ce (in div idu als ) Be e a bu nd an ce st an da rd ize d b y e ffo rt ( ind ivi du als /m in. ) (a) (b) (c) (d) (e) (f) 2015 2016 Be e a bu nd an ce st an da rd ize d b y e ffo rt ( ind ivi du als /m in. ) 24 Figure 1.3: Participant effort, in minutes spent observing, was high at the beginning of the field season, but declined over time in both years (a. & b.). Raw bee abundances in both years decline similarly to the reduction in participant effort (c. & d.). Standardizing observations by effort (e. & f.) allows for a comparison of data over the season and between years. Even after adjusting for effort we document a steep decline in bee abundance per minute effort in period 9 (e.) and period 7 (f.). Volunteers submitted 316 data sheets during the 2016 field season, representing 349.4 hours of observation time. As in the previous year participants spent less and less time observing flowers as the season progressed (Figure 1.3 a & b). In total, volunteers observed 18,479 pollinators including 1,974 butterflies, 14,088 bees, and 2,098 “other” pollinators. The “other” pollinator group included wasps, beetles, flies, hummingbirds, hummingbird hawk moths, and other moths. As in 2015, the “bumble bees” were the most commonly observed bee functional group with 4,037 (21.8% of total observations, 28.7% of bees observed) observations recorded. The most frequently observed butterfly functional group were the “whites and sulphur butterflies” with 694 observations reported (3.76% of total observations, 35.2% of butterfly observations). Specimen collection In 2016, 661 bees were collected in bowl traps and 1,276 bees were hand- collected from flowers. The collected bees represented 83 species from 27 genera (Appendix G: List of species collected in the 2016 specimen collection). Based on the records in discoverlife.org for Maryland, eight new county records, seven in Baltimore county and one in Allegany county, were created from this bee collection (Error! Reference source not found. 25 Table 1.4: New county records of bee species collected in the 2016 specimen survey. These records come from both hand-collected and bowl trap specimens. The community composition of bee functional groups from observational data differed from the composition of the hand-collected specimen data (PERMANOVA test df= 1, p<0.01). Honey bees (17.5%), bumble bees (15.4%), and small dark bees (11.8%) were responsible for most of the differences between the community composition of the two data types (hand-collected vs. observed, Error! Reference source not found.). Community compositions, as expected, also differed between the three sites (df=1, p<0.001). However, there was no interaction between data type and site (df=1, p>0.05). This indicates that the community composition differences between data types were the same regardless of site. Bee species: Bee gender: County: Lasioglossum versatum Robertson female Baltimore Lasioglossum trigeminum Gibbs female Baltimore Triepeolus remigatus Fabricius female Baltimore Epeolus bifasciatus Cresson male Baltimore Hylaeus mesillae Cockerell female Baltimore Heriades carinata Cresson female Baltimore Megachile exilis Cresson female Alleghany Megachile exilis Cresson male Baltimore Bee functional group Average abundance (bees/period) Contribution to difference % Observations Specimen collection Honey bees 28.48 14.28 17.48 Bumble bees 21.13 19.88 15.43 Small dark bees 19.70 16.64 11.84 Large dark bees 8.91 6.60 8.61 Large Carpenter bees 7.74 6.00 6.20 Green metallic bees 3.96 4.48 3.40 Other metallic bees 1.83 0.12 1.32 Long-horned bees 0.13 1.44 0.95 26 Table 1.5: SIMPER analysis results indicated the average abundance of each functional group of bees observed or hand-collected per two-week period (see Table 1.1) and the percentage that each functional group contributes to the community composition differences between data types (observational vs. hand-collected specimen data). Honey bees and bumble bees contributed the most to the differences between the community compositions of the observed and hand-collected data. “Other metallic bees” were not collected at any site after period 2, however they were observed by participants in multiple sites after this time. When the “other metallic bees” after period 2 were removed, the community composition between data collection methods no longer differed (PERMANOVA test, df=1, p>0.1). Similarly, when observed values of large carpenter bees and bumble bees were corrected in periods 9 and 10, community composition no longer differed between data collection methods (df=1, p>0.1 Figure 1.4). 27 Figure 1.4: 2016 hand-collected bee specimen data standardized by effort in minutes. There is no late season decline in bees per minute as experienced in both seasons of observationally collected data. Hand-collected specimen data from period 10 is not available due to a lack of flowering resources in the monitored gardens at this time. Strong relationships were found between the number of bees hand-collected and observed for the “honey bee” (Spearman rank coefficient: rs= 0.617, p<0.05) and “bumble bee” (rs=0.781, p<0.05) functional groups. The “green metallic bees”, and “small dark bees” demonstrated a moderate relationship (rs=0.550, p<0.05; rs=0.474 0.0 0.3 0.6 0.9 1 2 3 4 5 6 7 8 9 Period Vb t variable Honey.bees Bumble.bees Large.carpenter.bees Long.horned.bees Large.dark.bees..total Small.dark.bees..total Green.metallic.bees Metallic.bees..Other.metallic.bees Be e ab un da nc e s ta nd ar diz ed b y e ffo rt ( in di vid ua ls/ m in ute ) 0.0 0.3 0.6 0.9 1 2 3 4 5 6 7 8 9 Period Vb t variable Honey.bees Bumble.bees Large.carpenter.bees Long.horned.bees Large.dark.bees..total Small.dark.bees..total Green.metallic.bees Metallic.bees..Other.metallic.bees Hon y bees Large Ca p nter bees Bum l bees Long-horned bees Large dark bees bees Small dark bees Green metallic bees Other metallic bees Legend: 28 p<0.05 , respectively) while the remaining bee functional groups demonstrated weak relationships between the observational and hand-collected specimen data (“large carpenter bees” rs= 0.371, p<0.05; “large dark bees” rs= NA, p>0.05; “long-horned bees” rs= -0.039, p>0.05; “other metallic bees” rs= -0.027, p>0.05; Error! Reference source not found.). Figure 1.5: Relationships between the observed and hand-collected data collected in 2016. All data points are standardized by plant, period (see Table 1.1), and time. Data presented are from the three locations where both observational and hand-collection specimen data are available: Baltimore, Allegany and Frederick counties. Based on p = 2.407e−05 p = 0.8623 p = 1.283e−07 p = 1.664e−12 p = 0.1804 p = 0.872 p = 0.03409 p = 1.325e−07 Green.metallic.bees Metallic.bees..Other.metallic.bees Long.horned.bees Large.dark.bees..total Small.dark.bees..total Honey.bees Bumble.bees Large.carpenter.bees 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 Hand−collected Ob se rv ed (a) (b) (c) (d) (e) (f) (g) (h) be umble b es Large C bees - rne b es Large dark bees Small dark bees reen ic b es Other metallic bees p <0.0001, rs=0.617 p <0.0001, rs=0.781 p <0.05, rs=0.371 p >0.1, rs=-0.039 p >0.1, rs=NA p >0.1, rs=-0.027 p <0.0001, rs=0.474 p <0.0001, rs=0.550 Ob se rva tio ns (b ee ab un da nc e/m inu te ) Hand-collected specimens (be abundance/minute) 29 Spearman rank calculations, strong relationships were found for the honey bees (a; rs=0.617) and bumble bees (b; rs=0.781). Discussion Certain bee functional groups, honey bees and bumble bees, demonstrated linear relationships between the citizen scientist observational data and the hand-collected specimen data. This is an encouraging finding as the use of citizen science to monitor pollinator communities. Pollinator communities are diverse and species-rich, and for many of these insects, little is known about their floral preferences. Citizen science has the potential to collect the information needed about pollinator community composition, floral preference, and population dynamics to fill existing knowledge gaps about pollinator biology. However, our results show that the relationships that we observe are not always strong and can improve. Ways in which we can improve our results include: additional participant training leading to better observer accuracy and corrections based on annual life cycles. The ability of citizen scientists to properly identify floral visitors improved with training, and continued to improve with further training and experience, as indicated by an average quiz score improvement of approximately 17% over both years (Table 1.3). Others citizen science projects have documented similar findings, known as the “first- year” effect (Dickinson et al. 2010). For example, members of iSpot, an interactive online community where individuals submit and identify photos of plants or wildlife, increased their ability to correctly identify photographed organisms over time. iSpot participants also have access to online quizzes of varying difficulty to self-test and improve their identification skills ((Scanlon et al. 2014). When analyzing data from the long-term 30 Breeding Bird Survey data, removing data collected by first-year participants between 1966 and 1991 helped to reduce bias in population trend estimates (Kendall et al. 1996). Future use of the monitoring guide and the methods used in this project to evaluate pollinator communities may benefit by implementing a more regular self- or peer-to-peer testing and review process. For instance, online quizzes and review sessions that participants need to complete prior to bi-weekly observations. More participants generate larger datasets, and so retaining participants for a complete field season is important (Dickinson et al. 2010). In our study, most participants who started collecting data, did so for the entire season. Even so, retaining more of the participants would have created a more spatially complete and valuable dataset. Other studies have increased retention rates by increasing communication between participants and researchers (Dickinson et al. 2012; Birkin and Goulson 2015) or by providing incentives. Successful incentives for participation incorporate rewards of either monetary value (Rogstadius et al. 2011) or intrinsic value (i.e. contributions toward scientific research (Silvertown 2009)). For instance, to encourage participation in the 2012 Great Backyard Bird Count, researchers offered “The Great Horned Owl Award”, a certificate presented for participation (Dickinson et al. 2012). eBird, a citizen-based bird observation network, saw an increase in participation after a shift in the perspective of their research goals from aiming to create a scientifically valuable bird database to creating a useful resource for bird watchers (Sullivan et al. 2009). In this survey, the participant effort declined over time, with a sudden decrease in minutes spent observing in the late summer and early fall (Error! Reference source not found.). Both field seasons show this trend, which may correspond to participants going 31 on vacation or getting busier as the academic school year begins. We did not attempt to elucidate the cause of this decline in our study, however, research to better understand the driver(s) of this trend would be beneficial. Future citizen science pollinator monitoring studies may benefit from implementing incentive strategies to better retain participation throughout the field season. Honey bees and bumble bees correlated strongly between the observational and hand-collected specimen data, however these two groups also contributed the most to differences in community composition between data collection methods. Indicating that while participant observations did not match the hand-collected specimen data, we can still define a relationship between these data sets. It is interesting to note that some of the larger, common bees (“honey bees” and “bumble bees”) were the bees contributing the most to the differences between community compositions with the observers reporting more “honey bees” and “bumble bees” than were collected. It may be that observers counted the same bee more than once, or that the removal of foraging honey bees prevented the bee from recruiting more foragers upon returning to the colony (Visscher and Seeley 1982). Another reason for the differences detected between the observational and hand- collected specimen data, is confusion of bee functional groups. This is likely the case when it comes to “large carpenter bees” and “bumble bees” late in the field season (late- September and October). While “bumble bees” were over-reported by participants, in contrast, “large carpenter bees” were under-reported by observers (Error! Reference source not found.a). Large carpenter bees are abundant in the spring with fewer adults foraging during the summer months. In the fall, the large carpenter bees become 32 abundant again as the first-year juvenile bees first forage in search of nectar (Gerling and Hermann 1978). At this point in the year, the bumble bee colonies have produced queens and the number of foraging workers has declined, sometimes dropping to zero (Williams et al. 2014). Since “bumble bees” and “large carpenter bees” are very similar looking (Error! Reference source not found. c &d), it is likely that in the fall, when “bumble bees” are still frequently observed, some observers were confusing these groups of bees. If the observational data from periods 9 and 10 are corrected with the generated linear relationships, the community composition between the data collection methods no longer differ. While this result implies that late season misidentification of “large carpenter bees” and “bumble bees” contributes to the differences in community composition differences, this result should be interpreted cautiously as neither voucher specimens nor photographs exist to confirm the identity of these bees. 33 Figure 1.6: In Period 9, participants observed (a) 0.23 “bumble bees” per minute and 0.02 “large carpenter bees” per minute as opposed to the 0.07 “bumble bees” per minute and 0.18 “large carpenter bees” per minute hand-collected (b). These differences are likely due to participants misidentifying the similar-looking “bumble bees” (c) and “large carpenter bees” (d) late in the field season. In the early fall, bumble bee colonies have finished producing reproductive gynes and workers are no longer foraging to support the colony, thus bumble bee observations should decrease at this time. In contrast, juvenile large carpenter bees forage in the fall to store up energy to survive the winter months. 0.0 0.1 0.2 0.3 0.4 1 2 3 4 5 6 7 8 9 10 Period Be e ab un da nc e sta nd ar diz ed b y e ffo rt 0.0 0.1 0.2 0.3 0.4 1 2 3 4 5 6 7 8 9 10 Period Be e ab un da nc e sta nd ar diz ed b y e ffo rt (a) (b) Observational Hand-collected specimen (c) (d) Be e ab un da nc e s ta nd ar diz ed b y e ffo rt ( in di vid ua ls/ m in ute ) 34 Hand-collected specimen data from period 10 is not available due to a lack of blooming floral resources at the monitored sites. Photos c. & d. courtesy of John Baker. While the “other metallic bees”, or bees in the genus Osmia, did not contribute significantly to the differences in community composition between the two groups, observers reported seeing this functional group throughout the field season, likely long after their natural annual life cycle had finished. The identification accuracy of this group of bees may increase if the study began earlier in the spring to encompass more of the Osmia sp. phenology, giving participants the opportunity to witness these bees foraging. While we know that the bees that participants reported seeing in this functional group late in the season were not Osmia sp., they cannot be confidently reclassified into another bee functional group without specimens or photos. “Other metallic bee” specimens were not hand-collected after Period 2 in June at any site. Removing the “other metallic bee” observations after June (Periods 3-10) from the PERMANOVA analysis, alters the end results: community compositions are no longer different. Future directions The use of the internet in citizen science projects aided in its resurgence, providing volunteers with the opportunity to instantaneously upload data while in the field (Dickinson and Bonney 2012; Scanlon et al. 2014). A native pollinator survey, like this one, could benefit from the incorporation of more technological components (e.g. an app with a compatible website for submitting and viewing data). A user-friendly app or website could more easily cater to a younger, and potentially untapped, demographic such as school children (Evans et al. 2001; Weckel et al. 2010; Tweddle et al. 2012; Hidalgo-Ruz and Thiel 2013) of participants and could expedite the data entry process. 35 For example, nearly 1000 school children from 39 schools documented plastic debris on Chilean beaches in the ‘National Sampling of Small Plastic Debris’ citizen science project (Hidalgo-Ruz and Thiel 2013). Furthermore, younger participants may have sharper eyesight, helping them to better distinguish pollinator functional groups. While bees and butterflies are the primary pollinators, there are many other important pollinators such as beetles, moths, flies, and hummingbirds visiting flowers. Many studies that examine pollinator floral preferences focus on a small set of pollinators: bumble bees: (Patten et al. 1993), bees: (Frankie et al. 2005; Tuell et al. 2008; Williams et al. 2015), bees and syrphid flies: (Bahlai and Landis 2016), or butterflies: (Jennersten 1984; Corbet 2000; Shackleton and Ratnieks 2016), but few studies examine the entire community of pollinators at once (but see: (Corbet et al. 2001; Bartomeus et al. 2008). Pollinator communities are complex and involve many intricate interspecies relationships. These communities are altered by factors such as the introduction of invasive species (Bartomeus et al. 2008), climate change (Breed et al. 2013) and habitat loss (Fortuna and Bascompte 2006; Ricketts et al. 2008; Winfree et al. 2009). Understanding how the entire pollinator community responds to these changes can help guide conservation and restoration efforts. Expanding this study to encompass all pollinators would increase understanding about relationships between plants and pollinator communities in a variety of settings. 36 Chapter 2: An Assessment of the Attractiveness of Pollinator Plants Using a Combination of Citizen Science Observational and Hand-Collected Specimen Data Abstract: Native and managed pollinators experience stress due to habitat loss and fragmentation in response to urbanization. Despite frequent habitat fragmentation, suburban landscapes retain the potential to host a diverse native pollinator community in green spaces, conservancies, and gardens. Plantings intended to benefit pollinators have become a popular installment in many yards and managed gardens. Most gardens are planned using pre-existing pollinator garden planting lists or books. However, many of these lists are based upon the author’s personal experiences and cover a broad spatial area. To create more reliable planting recommendations, a small-scale, data-driven assessment of pollinator preference is needed. This study examines the relative attractiveness of 23 plant species in Maryland (5 non-native, 18 native species) to pollinators. The data used to assess these pollinator plants was obtained from the Maryland Native Pollinator Survey in 2015 and 2016. These plants were assessed on five metrics: bloom length, visitation rates, pollinator richness and evenness, and value to specialist pollinators. Plants received a score based on the sum of these five metrics to determine their attractiveness to pollinators. Plants were divided up into three portions of the field season: ‘early’, ‘middle’, and ‘late’. In the early season Asclepias incarnata scored highest and Thymus vulgaris scored the lowest. Similarly, in the ‘middle’ season, Asclepias incarnata scored highest, and Thymus citriodorus scored the lowest. Symphyotrichum novae-angliae scored highest in the late season and Rudbeckia triloba scored the lowest. These scores can provide a framework for assessing the relative value of pollinator plants on a smaller scale. This method can be applied to other locations to determine how the value of a pollinator plant changes throughout the landscape and over time. Introduction: Globally, native and managed pollinators have experienced increased stressors over the last five decades (Potts et al. 2016). A myriad of factors are responsible for these stressors including: exposure to pesticides (Scott-Dupree et al. 2009; Potts et al. 2010; Smith et al. 2014; Bernauer et al. 2015), the introduction and spread of new and possibly more virulent parasites and pathogens (Genersch et al. 2006; Colla et al. 2006; Vanbergen and The Insect Pollinators Initiative 2013; Goulson et al. 2015), a decline in habitat quality (Jennersten 1984; Curtis et al. 2015; Potts et al. 2016) and habitat loss and fragmentation due to anthropogenic factors such as urbanization or climate change (Fortuna and Bascompte 2006; Forister et al. 2010; Potts et al. 2010; Goulson et al. 37 2015). Pollinators provide the ecosystem service of pollination that is vital to agricultural crop, nut, and seed production (Klein et al. 2007; Potts et al. 2010; Ollerton et al. 2011). While the major agricultural pollinator is the honey bee (Calderone 2012), many native bee species also contribute to crop pollination (Vicens and Bosch 2000; Kremen et al. 2002; Winfree et al. 2007; Garibaldi et al. 2013; Gaines-Day and Gratton 2016) and are often responsible for facilitating plant reproduction in many natural systems (Wcislo and Cane 1996; Fortuna and Bascompte 2006). Though suburban landscapes are often composed of a mosaic of land use types, they retain the potential to host a diverse native pollinator community (Smith et al. 2005, 2006; Fetridge et al. 2008; Owen 2010); often in the form of green spaces, conservancies and gardens (Salisbury et al. 2015). Similarly, hedgerows of flowers planted along agricultural fields can support pollinator and natural enemy communities by providing forage and nesting resources (Patten et al. 1993; Morandin and Kremen 2013; Williams et al. 2015). While agricultural hedgerows are valuable to many pollinators, the focus of this paper is on plantings in urban and suburban settings. When establishing or improving existing green spaces, one may consult a book or website for guidelines and suggestions to ensure their garden will best support local pollinators. Many educational resources exist that provide lists of pollinator-friendly plants (Tallamy 2007; The Xerces Society 2011, 2016b, a; Holm 2014; Hayes 2016). However, there are some issues with many of these lists including: the spatial scale that lists are created for and that the lists are often based upon the authors personal experiences rather than on experimentally collected data. 38 Pollinator garden planting recommendations are often made at large, regional scales, and cover many distinct physiographic regions. This ignores microclimatic and macro-environmental differences within the region. For instance, Maryland is typically considered a part of the mid-Atlantic region along with: North Carolina, Virginia, West Virginia, Delaware, New Jersey, Pennsylvania, and Washington D.C. (Adamson et al. 2015). Yet, Maryland alone has five distinct physiographic regions including: three mountainous, a piedmont plateau, and a coastal plain region, each with their own unique biota (Reger and Cleaves 2008). Throughout these physiographic regions, the environmental conditions change and so do the accompanying flora and fauna, including pollinators. To make accurate planting recommendations for pollinators in a state like Maryland, studies that focus on smaller areas are needed. More recently, studies have begun to use experimental data to determine which pollinator plants are the most valuable in their respective region, but these recommendations are mainly from just a few locations (i.e. Michigan (Fiedler and Landis 2007; Tuell et al. 2008; Williams et al. 2015), California, and Florida (Williams et al. 2015), Europe (Garbuzov and Ratnieks 2014, 2015; Shackleton and Ratnieks 2016)) and each paper uses a different method for assessing which plants are the most attractive to pollinators. For example, (Tuell et al. 2008) collected bee specimens and conducted observations to determine which bees were visiting the species flowers in their study. They examined floral characteristics like floral area per meter and corolla width to explain bee preference. In contrast, (Williams et al. 2015) looked at the attractiveness of pollinator-friendly plant mixes instead of specific plant species. 39 There are many floral traits that play a role in a flower’s respective attractiveness to pollinators (i.e. quality and quantity of nectar and pollen, corolla length, color, etc. (Fenster et al. 2004). Furthermore, different pollinators are looking for particular things in a flower. For example, when considering the nutritional reward that pollinators receive from visiting a flower, most butterflies only need nectar (O’Brien et al. 2004) while bees need both nectar and the pollen (Michener 2007). And as such, pollinators will utilize different floral resources. Often plants and pollinators have close-knit relationships, that have evolved together (Hu et al. 2008). For example, the tropical plant species Zaluzianskya microsiphon (Kuntze) is primarily pollinated by a long-tongued fly species. The corolla depth of this plant species varies throughout its range and similarly, the tongue length of the corresponding fly pollinator varies (Anderson and Johnson 2008). Pollinators that rely on a single or a few plant species for their food sources are considered specialists on those plant species (Waser and Ollerton 2006). However, some pollinators, like honey bees, are generalist foragers in that they will feed on many different flowers (Menzel et al. 1993). Therefore, the value of plants utilized by specialist pollinators is greater than plants only used by generalists, as their foraging range is wider (Fowler and Droege 2016). Specialists are pollinating insects that forage on a narrow set of plants. This is often at the family level, but some bees are known to further specialize on a particular plant genus (Fowler and Droege 2016). In this study, we assessed different commonly recommended pollinator plant species for their respective attractiveness to different pollinators in Maryland. This study takes a novel approach to quantifying floral attractiveness to pollinators by combining 40 multiple metrics for attractiveness. We used several different metrics to grade flowering plants including bloom length, pollinator visitation rates, the richness and evenness of the pollinator communities visiting the flowers, and the value to specialist pollinators. By assessing these metrics together, we can determine which pollinator plants among those studied are the most valuable to Maryland’s pollinators. Furthermore, this study improves upon exiting planting lists by making Maryland-specific, data-driven recommendations. Methods: Data preparation: To assess the attractiveness of each flower species, data from the Maryland Native Pollinator Survey was used (Bernauer 2017). This project generated observational and hand-collected specimen data on which pollinators were visiting specific plant species in Maryland. For the methods of data collection, see: Chapter 1: The Maryland Native Pollinator Survey: a useful method for citizen scientists to survey pollinators visiting floral resources (Bernauer 2017). For the purposes of this study a plant record is a single 10-minute hand-collected specimen or observational data collection period; for the remainder of this paper observational and specimen data will no longer be separated. Before data analysis, data was gathered at the level of individual plant species. Plant species not included in the standard plant list specified in methods section of Chapter 1: The Maryland Native Pollinator Survey: a useful method for citizen scientists to survey pollinators visiting floral resources were not analyzed. These plants were removed to keep the focus of this study narrow and to complement the results from Chapter 1. While the data collection for the Maryland Native Pollinator Survey was supposed to take place on days with optimal foraging conditions, this was not always the 41 case. When data was collected, the weather was also measured in three ways: the temperature (50s, 60s, 70s, 80s, 90s, 100+ °F), wind (still, light breeze, windy, gusty), and cloud cover (clear, partly cloudy, mostly cloudy, overcast). Optimal foraging conditions in this study are days in which the wind was “still” or there was a “light breeze” and the cloud cover was “clear” or “partly cloudy”. Plant records that were collected on days with sub-optimal weather conditions were excluded from analyses so that the results of this study can be useful when comparing to other similar studies where data was only collected during optimal foraging conditions (Tuell et al. 2008; Isaacs et al. 2009; Williams et al. 2015). Plants bloom at different intervals throughout the field season. To make meaningful comparisons of pollinator plant attractiveness, only plants that bloom at the same time were contrasted. The field season was portioned into three parts: the early season (May to June), mid or middle season (July to August), and late season (September through November), following the example of (Tuell et al. 2008). Plant species with fewer than five plant records per portion of the field season were removed. Plant attractiveness metrics We quantified five different metrics to assess plant attractiveness to pollinators: bloom length, visitation rates, pollinator richness and evenness, and specialist value. The five criteria for assessing plant attractiveness were calculated and each metric was scaled down to a maximum value of 1. Metrics were scaled so that each of the five metrics were equally important to the total score. The first metric we assessed was bloom length (B), or a sum of the number of weeks that the plant species was in bloom. Bloom length was determined by the span of 42 weeks that plant records existed for a particular species and was averaged between years when the data was available. To scale down the bloom length metric, the plant’s bloom length in weeks was divided by the total number of weeks in the growing season. Based on the scope of the data set from the Maryland Native Pollinator Survey, this was determined to span May 1 thru the second week in November, for a total of 29 weeks. No single species bloomed throughout the entire 29 week period, thus every plant scores less than 1 for this metric. The visitation rate metric (V) is a measure of the number of pollinators visiting a given plant per minute. This was calculated by summing the total number of pollinators recorded on a plant divided by the time spent collecting data on said plant. The visitation rate metric was scaled down by dividing by the highest value of pollinators per minute recorded on one of the plant species analyzed in this study: 6.5. For the richness metric (R), the number of pollinator functional groups recorded on a plant was counted. Floral visitors were categorized into 14 different functional groups of bees and butterflies and the maximum value a plant could score was 14. To scale down the richness metric, the number of recorded pollination functional groups was divided by 14. A plant’s evenness metric (E) was determined by Simpson's diversity index. Simpson’s index (Simpson 1949) provides a measure of the evenness of the pollinator community recorded on each assessed plant. Plants that support a more even pollinator community (i.e. no single species dominates) have a higher Simpson’s index making them more valuable to the pollinator community overall. Because the Simpson’s diversity index has a maximum score of one, it was not further scaled down. 43 Lastly, the value to specialist pollinators (S) was determined for each plant at the genus or family level. A bee or butterfly is a specialist on a specific plant species if the plant either provides a larval host for butterflies and/or adult pollen and/or nectar source to at specialist bee or butterfly species. To determine the number of specialist pollinators that use a plant, two lists were used: the Central Maryland (Piedmont) list produced by the North American Butterfly Association (Gibbs et al. 1997) for butterflies and the Specialist bees of the Mid-Atlantic and Northeastern United States list (Fowler and Droege 2016) for bees. Each plant received a score between 0-3 for their value to specialist pollinators, one point was awarded for each of the following: supporting bee, adult butterfly, or larval butterfly specialists. For example, a plant that supported bee specialists but did not support adult or larval butterfly specialists would receive a score of 1. On the other hand, a plant that served as a larval host plant for a specialist butterfly and adult butterflies would score a 2. Similarly, a plant that supported larval butterflies and bee specialists would also score a 2 and so on. For a plant to score a 3, it supported specialist bees, adult butterflies, and provides a larval host plant for a specialist butterfly. The value to specialist pollinators metric was divided by three, giving a plant species a maximum score of one. Once the metrics were calculated and scaled, they were averaged by plant species for a single portion of the field season. The averaged metrics for a plant species were summed and then divided by 5. The resulting value is a percentage, which can be converted to a final score out of 100. This final score will be known as the “pollinator attractiveness rating” or PAS for short and is illustrated in Equation (1). (1) 44 PAS = B + V + R + E + S5 ∗ 100 Data analysis: All statistical analyses were conducted in R version 3.3.2 (R Core Team 2016). Pollination networks were constructed with the bipartite package (Dormann et al. 2008, 2009) to visualize the interconnectedness of plant-pollinator interactions and how they change throughout the field season. Results: Data preparation The complete data set contained 2,962 plant records encompassing 210 plant species from 151 genera. After removing the plant records without species-level information available, there were 2,093 plant records remaining from 119 plant genera and 210 species. Removing the plants species not on the standard list of plants leaves 1000 plant records for 45 plant species representing 12 genera. Removing plant records with sub-optimal foraging weather yields 620 total observations, and 41 plant species from 12 genera. Plant species without five records in at least one portion of the field season were removed, leaving 542 plant records encompassing 23 species of plants in 11 genera. Eighteen of the remaining plants assessed are native to Maryland while the other five are non-natives (Table 2.1). 45 Season Plant B lo om le ng th V is ita tio n ra te R ic hn es s Ev en ne ss V al ue to Sp ec ia lis ts Su m Pe rc en ta ge R an ki ng Early Asclepias incarnata L. 0.45 0.52 0.20 0.53 0.67 2.36 47.30 1 Early Asclepias syriaca L. 0.38 0.36 0.21 0.59 0.67 2.20 44.06 2 Early Monarda fistulosa L. 0.28 0.56 0.09 0.52 0.67 2.12 42.31 3 Early Asclepias tuberosa L. 0.52 0.32 0.16 0.43 0.67 2.09 41.82 4 Early Echinacea purpurea (L.) Moench 0.62 0.08 0.11 0.51 0.33 1.65 33.02 5 Early Monarda didyma L. 0.28 0.16 0.12 0.47 0.33 1.36 27.21 6 Early Rudbeckia hirta L. 0.59 0.16 0.09 0.04 0.33 1.21 24.11 7 Early Thymus citriodorus (Pers.) Schreb. 0.24 0.32 0.13 0.43 0.00 1.12 22.40 8 Early Thymus vulgaris L. 0.52 0.12 0.09 0.25 0.00 0.98 19.53 9 Middle Asclepias incarnata L. 0.45 0.52 0.24 0.60 0.67 2.47 49.41 1 Middle Origanum laevigatum Boiss. 0.38 0.88 0.49 0.70 0.00 2.45 48.90 2 Middle Asclepias syriaca L. 0.38 0.64 0.29 0.44 0.67 2.41 48.23 3 Middle Asclepias tuberosa L. 0.52 0.44 0.14 0.48 0.67 2.25 44.94 4 Middle Origanum vulgare L. 0.48 1.00 0.27 0.45 0.00 2.20 44.08 5 Middle Echinacea purpurea (L.) Moench 0.62 0.48 0.19 0.55 0.33 2.18 43.54 6 Middle Monarda fistulosa L. 0.28 0.56 0.20 0.31 0.67 2.01 40.25 7 Middle Rudbeckia laciniata L. 0.48 0.56 0.19 0.43 0.33 1.99 39.84 8 Middle Eurybia divaricata (L.) G.L.Nesom 0.55 0.40 0.19 0.47 0.33 1.95 38.96 9 Middle Helianthus annuus L. 0.48 0.60 0.14 0.28 0.33 1.84 36.78 10 Middle Rudbeckia hirta L. 0.59 0.40 0.15 0.28 0.33 1.75 34.99 11 Middle Monarda didyma L. 0.28 0.28 0.15 0.57 0.33 1.61 32.18 12 Middle Monarda punctata L. 0.34 0.24 0.11 0.50 0.33 1.53 30.65 13 Middle Agastache foeniculum (Pursh) Kuntze 0.48 0.40 0.19 0.46 0.00 1.53 30.57 14 Middle Rudbeckia fulgida Aiton 0.28 0.12 0.11 0.41 0.33 1.25 24.93 15 Middle Thymus citriodorus (Pers.) Schreb. 0.24 0.16 0.11 0.65 0.00 1.17 23.31 16 Late Symphyotrichum novae- angliae (L.) G.L.Nesom 0.17 0.84 0.31 0.59 0.67 2.58 51.67 1 Late Solidago canadensis L. 0.31 0.80 0.19 0.60 0.67 2.57 51.40 2 46 Season Plant B lo om le ng th V is ita tio n ra te R ic hn es s Ev en ne ss V al ue to Sp ec ia lis ts Su m Pe rc en ta ge R an ki ng Late Solidago caesia L. 0.31 0.60 0.19 0.34 0.67 2.11 42.20 3 Late Rudbeckia laciniata L. 0.48 0.44 0.24 0.57 0.33 2.06 41.24 4 Late Eurybia divaricata (L.) G.L.Nesom 0.55 0.56 0.19 0.41 0.33 2.05 40.96 5 Late Symphyotrichum laeve (L.) Á & D. Löve 0.34 0.52 0.24 0.50 0.33 1.94 38.82 6 Late Origanum laevigatum Boiss. 0.38 0.52 0.31 0.46 0.00 1.67 33.33 7 Late Solidago rugosa Mill. 0.28 0.20 0.17 0.30 0.67 1.61 32.28 8 Late Agastache foeniculum (Pursh) Kuntze 0.48 0.16 0.06 0.71 0.00 1.41 28.20 9 Late Rudbeckia triloba L. 0.21 0.04 0.04 0.50 0.33 1.12 22.32 10 Table 2.1: A ranking of the 23 plant species assessed as follows. Each plant was assessed with five different metrics that were summed and turned into a percentage. The percentage was used to rank each plant within each of the three portions of the field season. Plants that are not native to Maryland are bolded. Plant attractiveness Pollinator attractiveness scores (PAS) for the 23 plant species assessed in this study ranged from a low of 19.53 to a high of 51.67. The bloom length metric varied from 0.172 to 0.621, visitation rate from 0.04 to 1, richness from 0.036 to 0.486, evenness 0.04 to 0.71, and specialist value 0 to 0.667. Thymus vulgaris scored the lowest of all plants assessed with a PAS score of 19.53 and the overall highest-scoring pollinator plant was Symphyotrichum novae-angliae with a score of 51.67 (Table 2.1). Of the plants assessed in the early season, the most attractive pollinator plant was Asclepias incarnata (PAS = 47.30) and the least attractive of the plants assessed was Thymus vulgaris (PAS = 19.53). The most attractive pollinator plant in the middle season was again Asclepias incarnata (PAS = 49.41) and the least attractive plant was Thymus citriodorus (PAS = 47 23.31). In the late season, Symphyotrichum novae-angliae (PAS = 51.67) was the most attractive on the plant and Rudbeckia triloba (PAS = 22.32) was the least attractive. The pollination networks for the early, middle, and late portions of the field season were generated (Figure 2.1). The early and late season pollination networks are simpler than the middle season pollination network. Figure 2.1: The pollinator networks constructed from the plant records used to create recommendations for the early (a), middle (b), and late (c) portions of the field season. The plant species are on the left and the pollinator functional groups are on the right. The width of the grey bars corresponds to the number of records between the plant and Brushfoots Whites.and.Sulphurs Gossamer.wings Swallowtails Green.metallic.bees Dark.bees Bumble.bees Long.horned.bees Other.metallic.bees Honey.bees Skippers Large.carpenter.bees Thymus.vulgaris Rudbeckia.hirta Asclepias.tuberosa Echinacea.purpurea Asclepias.incarnata Thymus.citriodorus Monarda.fistulosa Asclepias.syriaca Monarda.didyma Brushfoots Whites & Sulphurs Gossamer wings Swallowtails Gr en metallic bees Dark bees Bumble bees Long-horned bees Other metallic bees Honey bees Skippers Large Carpenter bees Thymus vulgaris Rudbeckia hirta Asclepias tuberosa Echin ea purpurea Thymus citriodorus Monarda fistulosa Asclepias syriaca Monarda didyma Asclepias incarnata Satyrs.and.Wood.Nymphs Dark.bees Large.carpenter.bees Bumble.bees Skippers Green.metallic.bees Other.metallic.bees Gossamer.wings Whites.and.Sulphurs Long.horned.bees Honey.bees Brushfoots Monarchs Agastache.foeniculum Rudbeckia.triloba Eurybia.divaricata Solidago.caesia Solidago.canadensis Rudbeckia.laciniata Symphyotrichum.laeve Solidago.rugosa Symphyotrichum.novae−angliae Origanum.laevigatum Satyrs & Wood Nymphs Dark bees Large Carpenter bees Bumble bees Skippers Green metallic bees Other metallic bees Gossamer wings Whites & Sulphurs Long-horned bees Honey bees Bru hfoots Monarchs Origa um laevigatum Symphyotrichum novae-angliae Solidago rugosa Symphyotrichum laeve Rudbeckia laciniata Solidago canadensis Solidago caesia Eury a divaricata Rudbeckia triloba Agastache foeniculum Swallowtails Bumble.bees Skippers Large.carpenter.bees Monarchs Honey.bees Brushfoots Long.horned.bees Dark.bees Gossamer.wings Satyrs.and.Wood.Nymphs Whites.and.Sulphurs Green.metallic.bees Other.metallic.bees Helianthus.annuus Monarda.punctata Rudbeckia.laciniata Echinacea.purpurea Monarda.fistulosa Agastache.foeniculum Asclepias.incarnata Asclepias.syriaca Monarda.didyma Thymus.citriodorus Origanum.laevigatum Rudbeckia.fulgida Rudbeckia.hirta Asclepias.tuberosa Eurybia.divaricata Origanum.vulgare Swallowtails Bumble bees Skippers Large carpenter bees Monarchs Honey bees Brushfoots Long-horned bees Dark bees Gossamer wings Satyrs & Wood Nymphs Whites & Sulphurs Green metallic bees Other metallic bees Origanum vulgare Eurybia divaricata Ascle ias tuberosa Rudbeckia hirta Rudbeckia fulgida Origanum laevigatum Thymus citriodorus Monarda didyma Asclepias syriaca Asclepias incarnata Agasta e foeniculum Monarda fistulosa Echinacea purpurea Rudb kia laciniata Monarda punctata Helianthus an us(a) Early (b) Middle (c) Late 48 pollinator; when this value is small, the grey bars become thin and may be black. Nonnative plant species are bolded. Discussion: This study demonstrates a useful metric to assess the value of plant species to pollinators. Because the PAS score is calculated using experimentally collected data, the results of this study are objective. While the data analyzed in this study are from Maryland, this method is not system- or location-dependent and can be applied to any number of study systems; it is not restrained to use in urban or suburban garden plantings and therefore is applicable in natural and agricultural systems. Furthermore, the PAS score can be useful for nurseries and growers to advertise high quality pollinator plants, similar to the NuVal nutritional index system often displayed in grocery stores (Katz et al. 2010). We see that the native plants frequently outscore the non-native plants in terms of their value to pollinators. The non-native plants analyzed with this metric consistently scored in the bottom half of the plants analyzed. This could be due to specialists, evolutionary relationships, or introduced pollinator species from the plant’s native range. In contrast, in the middle portion of the field season, O. laevigatum has a PAS score of 48.90 and ranks second of 14 in attractiveness to pollinators. In the late portion of the field season, however, O. laevigatum drops to a PAS score of 33.33 and ranks seventh of 10. Origanum vulgare, while only assessed in the middle portion of the field season, also scored relatively high, ranking fifth of 14. Origanum sp. accounted for two of the five non-native plant species assessed in this study. Similarly, researchers in Britain, where Origanum sp. are native, found that they also are beneficial to pollinator communities there (Garbuzov and Ratnieks 2014). (Razanajatovo et al. 2015) demonstrated that non- 49 naturalized alien plants were visited less frequently by pollinators than native or naturalized alien plants, which is generally consistent with the findings of this study. Although Origanum sp. do not occur in nature in Maryland, it is possible that pollinators have evolved a taste for this culinary herb in the last 200 years. Plant-pollinator communities are dynamic and change throughout the field season. Many solitary bee species are short lived as adults, with some only foraging for a few days to weeks (Michener 2007). Bumble bee colonies, which may persist for many weeks, have dynamic needs as they shift from rearing workers to gynes (Heinrich 2004). Similarly, some butterfly species, like the Baltimore checkerspot in Maryland, only forage for approximately six weeks as adults (Durkin et al. 2013). This implies that the value of a plant species to the pollinator community may change as the pollinator community does so as well. Pollinator networks can help to visualize the changing interactions between plant and pollinator species during a field season. The pollinator networks for the early and late portion of the field season are less complex than the middle portion of the field season (Figure 2.1). As the pollinator species change throughout the season, it is likely that the value of plant species also changes. For example, Echinacea purpurea increased in PAS score almost ten points in the middle season as opposed to the early season. While some of the studied plants rank quite high over two portions of the field season (i.e. Asclepias sp.), it is important to keep in mind that planting the highest- ranking plant and only this plant is not the best choice. Increased plant diversity has been shown to better support a more diverse pollinator community (Salisbury et al. 2015). For gardeners and garden planners to create the most valuable pollinator garden, they should 50 choose a variety of early, middle, and late season plants that rank highly with this metric. To ensure a high-quality habitat of continuously blooming flowers for specialist and generalist pollinators alike. A limitation of this study is that only bee and butterfly pollinator plant preferences were assessed, however to make this metric more robust, a full assessment of the pollinating insects visiting each plant (i.e. beetles, flies, and day flying moths) should be included. While the PAS scores generated in this study are based on two years of data, the scores are likely to change with increased years of data collection. Alarcón et al. (2008) found that the pollinator networks varied year-to-year in a montane meadow in California, likely due to severe drought conditions during the second of three field seasons. It is reasonably to suggest that environmental conditions impact pollinator networks in other locations, and so analyzing data from multiple years can help to separate pollinator-preference from annual variation in environmental conditions. 51 Appendix A: Demographic surveys of citizen science participants 2015 Site ID number Pre-training questions Date: _______________ Age: _______________ Gender: ______________ Race:_____________________ Occupation: ____________________________ Are you a Master Gardener? Yes No If yes, how many years have you been a Master Gardener for? _____________ Which Master Gardener Group are you a part of? _______________________ On a scale from 1 to 5, 1 being little or no experience, 5 being lots of experience, rate how familiar you are with pollinators?: 1 2 3 4 5 Where do you plan to observe pollinators? __________________________________ Do you have any experience working with pollinators? Yes No If yes, please describe your experiences below: At the end of the study you will have full access to the data collected. Do you intend to use the data or results of this study? Yes No If yes, please explain how you intend to use this data. Bee identification. Use the photos provided in the PowerPoint. Please categorize these bees into one of the following categories: honey bee, bumble bee, large carpenter bee, hairy leg bee, large dark bee, small dark bee, green sweat bee, metallic hairy belly bee, dark hairy belly bee, or cuckoo bee. Each choice may be used once, more than once, or not at all. Use the white ovals on the screen to determine the approximate size and shape of the bee. 1. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 2. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 3. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 4. 52 Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 5. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 6. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 7. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 8. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 9. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 10. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 53 Post-training questions Site ID number Date: ________________ How interesting did you find today’s training? Very boring Boring Indifferent Interesting Very interesting How helpful did you find today’s training? Very unhelpful Unhelpful Indifferent Helpful Very helpful How much did you previously know about pollinators and/or native bees? Nothing A little Some A lot How much did you learn at today’s training? Nothing A little Some A lot Do you have any additional questions that were not answered in today’s training? If so, list them here. If you would like to be contacted with a response, please leave your email or phone number as well. Do you have any suggestions for future training sessions? If so, write them here Use the photos provided in the PowerPoint. Please categorize these bees into one of the following categories: honey bee, bumble bee, large carpenter bee, hairy leg bee, large dark bee, small dark bee, green sweat bee, metallic hairy belly bee, dark hairy belly bee, or cuckoo bee. Each choice may be used once, more than once, or not at all. Circle your answers below. 1. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 2. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 3. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 4. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 54 5. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 6. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 7. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 8. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 9. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 10. Honey bee Bumble bee Large carpenter bee Hairy leg bee Large dark bee Small dark bee Green sweat bee Metallic hairy belly bee Dark hairy belly bee Cuckoo bee 55 2016 The Pre-quizzes asked about demographic data first, then had space to record the pictorial quiz answers. The Post-quiz had the same space to record the pictorial quiz answer but did not include the questions about demographic questions. Site ID number Pre-Quiz Maryland Native Pollinator Survey 2016 Date: _______________ Age: _______________ Gender: ______________ Race:_____________________ Occupation: ____________________________ - Are you: a Master Gardener? Yes No a Master Naturalist? Yes No - If yes, how many years have you been a Master Gardener/Naturalist for? _____________ - Which Master Gardener/Naturalist Group are you a part of? _______________________ - What garden(s) are you planning on monitoring? - Did you participate in the 2015 Maryland Native Bee Survey? Yes No - Rate your knowledge of pollinators on a scale from 1 to 5 (1 being little or no knowledge, 5 being very knowledgeable) 1 2 3 4 5 - Do you keep honey bees? Yes No If yes, for how many years have you been keeping honey bees? _______________ - Do you have any experience working with pollinators? Yes No If yes, please describe your experiences below: Pollinator Identification 1. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 2. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 3. 56 Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 4. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 5. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 6. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 7. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 57 8. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 9. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 10. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 11. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 12. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 58 13. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 14. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 15. Honey bee Bumble bee Large carpenter bee Long-horned bee Dark bee Metallic bee Swallowtail butterfly Monarch butterfly White or Sulphur butterfly Gossamer wing butterfly Satyr or Wood Nymph butterfly Brushfoot butterfy Skipper Other 59 Appendix B: Monitoring and Pocket Guides 2015 Monitoring guide 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 Pocket guide 95 96 97 98 2016 Monitoring guide 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 Pocket guide 159 160 161 162 163 164 165 Appendix C: List of standard and supplemental plants for participants to observe upon List of plants to monitor Maryland Native Pollinator Survey ** Please note that this list is NOT exhaustive, its purpose is to serve as suggestions of possible plants to monitor. ** Standard plants: • Asters: o Eurybia divaricata o Symphyotrichum novae- angliae • Asclepias spp. o Asclepias tuberosa o Asclepias syriaca o Asclepias incarnata • Agastache spp. o Agastache foeniculum • Echinacea spp. o Echinacea purpurea • Helianthus o Helianthus annuus o Helianthus divaricatus • Monarda spp. o Monarda fitsulosa o Monarda didyma o Monarda punctata • Oregano • Thyme o Thymus serpyllum o Thymus citriodorus • Rudbeckia spp. (2016 addition) o Rudbeckia hirta o Rudbeckia laciniata • Solidago spp. (2016 addition) o Solidago caesia o Solidago canadensis Additional plants: • Baptisia australis • Penstemon spp. o Penstemon digitalis o Penstemon smalii • Veronicastrum virginicum • Pycnanthemum spp. o Pycnanthemum muticum o Pycnanthemum tenuifolium • Vernonia noveboracensis • Helenium autumnale • Eupatorium/Eutrochium dubium • Eupatorium/Eutrochium dubium ‘Baby Joe’/’Little Joe’ • Coreopsis spp. o Coreopsis verticillata • Zizia spp. o Zizia aptera o Zizia aurea • Sedum spp. o 166 Appendix D: Citizen science pollinator monitoring data sheets 2015 The following is an example of the data sheet provided to observers in 2015. The only difference in the morning and afternoon datasheets was the word “MORNING” in the upper right corner. Each set of data sheets provided to observers also included their unique site ID number. PO L L IN A T O R M O N IT O R IN G D A T A SH E E T D ate: / /2015 M O R N IN G O B SE R V A T IO N PE R IO D IN ST R U C T IO N S: Fill in the date and requested tim es, and circle the relevant w eather-related inform ation. For the “H abitat changes… ” section, fill in all of the habitat changes and w eather anom alies (i.e. heavy precipitation, high w inds, tem perature extrem es, etc.) since the last tim e you observed bees at this site. In the table, fill in the “Floral R esource” colum n w ith the flow er species you are observing (i.e. type, cultivar, etc.); for each species, count the total num ber of plants (“# Plants”) and count the num ber of bees visiting each floral resource for ten (10) m inutes per species; record the num ber of plants you actually m ade your bee counts from (“# O bs.”). If you cannot determ ine w hich group a bee belongs to, record the bee in the “O ther” colum n and note w hether you could tell it w as not a honey bee or bum ble bee in the “O bservational N otes” section at the bottom of the page, along w ith any additional com m ents. Please record additional bloom ing species at the bottom of the other side of this datasheet, under “A dditional B loom ing Plants”. N am e: _____________________________________ Plot size: _______________________________________________ Site N am e: O bservation Start Tim e: : End Tim e: : W eather: Tem p: 50s / 60s / 70s / 80s / 90s / 100+ W ind: Still / Light B reeze / W indy / G usty Sky: C lear / Partly C loudy / M ostly C loudy / O vercast Floral R esource (G enus species) # Plants/ # O bs. H oney B ees B um ble B ees Large C arpenter B ees H airy Leg B ees D ark B ees G reen Sw eat B ees M etallic H airy B elly B ees D ark H airy B elly B ees C uckoo B ees O ther - D escribe in N otes H abitat changes since the last observation/collection date: Observational N otes: 167 2016 Below is an example of the front page of a data sheet for the 2016 field season. The data sheets are a little different looking than the 2015 data sheets as they needed to also include space for the butterfly functional groups. The back side of the data sheet is on the next page. Site ID num ber: D ate: _ _ _ _ /_ _ _ _ _ /_ _ _ _ _ M aryland N ative Pollinator Survey 2016 Instructions: Fill in the date, garden size (approxim ate square footage or acreage of area devoted to flow ering plants), num ber of participants, nam es can be included if desired, and start and end tim es of the observational period. D ocum ent the w eather, record either an approxim ate O R actual tem perature if this is possible (you do not need to fill out both) the level of w indiness and the cloudiness. For the H abitat changes since last observation period include here any m ajor w eather events, plants that are now flow ering, plants that are no longer flow ering, and any changes to the landscape. Choose six plant species to observe pollinators, preferably Aster, Asclepias, Agastache, Echinacea, H elianthus, M onarda, oregano, thym e, R udbeckia, and/or Solidago. For each plant, record its scientific nam e, being as specific as possible, the num ber of plants or the area the plants take up in the garden, the bloom status (entering bloom , full bloom , w aning) and w hat tim e you begin and end observing at each plant species patch, aim for at least 10 m inutes of observations per plant. R ecord floral visitors in the table provided, describing ‘other’ floral visitors in m ore detail in the larger space provided. G arden size: N um ber of participants: O bservation period start tim e: ____:____ am / pm O bservation period end tim e: ____:____ am / pm W eather inform ation: W ind: Still / Light breeze / W indy / O ccasional strong gusts Sky: Clear / Partly Cloudy / M ostly Cloudy / O vercast A pproxim ate tem perature: 50-59°F / 60-69°F / 70-79°F / 80-89°F / 90-99°F / 100°F + O R A ctual tem perature: _________ °F H abitat changes since last observation period: 1. Plant nam e: _______________________ N um ber of plants or area: _________ B loom status: Entering Bloom / Full Bloom / W aning Start tim e for observations on plant # 1. : ___:___ am / pm E nd tim e for observations on plant # 1. : ___:___ am / pm 2. Plant nam e: _______________________ N um ber of plants or area: _________ B loom status: Entering Bloom / Full Bloom / W aning Start tim e for observations on plant # 1. : ___:___ am / pm E nd tim e for observations on plant # 1. : ___:___ am / pm Swallowtails M onarchs W hites and Sulphurs Gossam ers Satyrs and W ood Nym phs Brushfoots Skippers Other: Honey bees Bum ble bees Large carpenter bees Long-horned bees Large dark bees Sm all dark bees M etallic bees Abdom inal scopa No abd. scopa Abdom inal scopa No abd. scopa Green m etallic Other m etallic Swallowtails M onarchs W hites and Sulphurs Gossam ers Satyrs and W ood Nym phs Brushfoots Skippers Other: Honey bees Bum ble bees Large carpenter bees Long-horned bees Large dark bees Sm all dark bees M etallic bees Abdom inal scopa No abd. scopa Abdom inal scopa No abd. scopa Green m etallic Other m etallic 02021 168 169 Appendix E: Exit survey questionnaires 2015 Maryland Native Plant Survey: Exit survey Starred (*) questions were mandatory for participants to answer. 1. * How helpful was the native bee training you received in Spring or Early summer? a. Very helpful b. Helpful c. Neutral d. Not helpful e. Very unhelpful f. NA, did not attend training 2. * How helpful was the Monitoring Guide when you were observing native bees? (The larger of the two booklets given out) a. Very helpful b. Helpful c. Neutral d. Not helpful e. Very unhelpful 3. * How helpful was the Pocket Guide when you were observing native bees? (The smaller of the two booklets given out) a. Very helpful b. Helpful c. Neutral d. Not helpful e. Very unhelpful 4. * Do you have any suggestions for improvement of the Monitoring Guide? (The larger of the two booklets given out) 5. * Do you have any suggestions for improvement of the Pocket Guide? (The smaller of the two booklets given out) 6. * Do you have any overall suggestions or comments to improve this study in future years? 7. * How likely are you to participate in this study again next year? (Next year’s study will include native bees, and butterflies) a. Very likely b. Likely c. Not likely d. Unlikely 8. * Do you plan to use the data or results from the survey for anything? a. Yes b. No 9. * If you plan on using the data or results from this survey, what do you plant to use them for? 170 2016 2016 Maryland Native Pollinator Survey: Exit survey This questionnaire has some final questions about the 2016 Maryland Native Pollinator survey. If you attended a training but did not actually collect and submit data, please still fill this out! Responses are anonymous, so please be complete and honest. Thank you so much for taking the time to complete this survey and for participating in my pollinator survey! Starred (*) questions were mandatory for participants to answer. 1. * Did you participate in the 2015 Maryland Native Bee survey? a. Yes b. No 2. * Did you attend an initial training session (in May or June of 2016)? a. Yes b. No 3. If you attended an initial training session, did you find it to be useful? Why or Why not? What could have been improved? 4. * Did you attend a follow-up training session (in July or August of 2016)? a. Yes b. No 5. If you attended a follow-up training session, did you find it to be useful? Why or why not? What could have been improved? 6. * Did you submit data either electronically or via paper mail for the 2016 Native Pollinator Survey? a. Yes b. No 7. If you submitted data electronically, what were your thoughts on that system? Did you like that option? Did it work well? Was it hard to use? What could be improved? 8. * Now that the 2016 Native Pollinator Survey is over with, please rate your knowledge of pollinators on a scale from 1 to 5 (1 being little to no knowledge of pollinators, 5 being very knowledgeable). a. 1 b. 2 c. 3 d. 4 e. 5 9. * Please tell me something you learned about pollinators after participating in the 2016 Native Pollinator Survey. 10. * Now think about the Monitoring guide you received. How helpful was the Monitoring guide when you were observing pollinators? Did you refer to it after the initial training session? 11. Do you have suggestions for improvement of the Monitoring Guide? If so, what are they? 171 12. * Now think about the Pocket guide you received. How helpful was the Pocket guide when you were observing pollinators? Did you refer to it after the initial training session? 13. Do you have suggestions for improvement of the Pocket Guide? If so, what are they? 14. Do you have any suggestions or comments to improve the study in future years? If so, what are they? 15. * How likely are you to participate in this study again next year? a. Very likely b. Likely c. Neutral/unsure d. Not likely e. Unlikely 16. If you had access to the results of this survey, would you use them for anything? (i.e. a pollinator talk) a. Yes b. No c. Maybe 17. * If you collected data for the 2016 Native Pollinator Survey, have you already submitted all of your data sheets? a. Yes b. No c. No, did not participate/collect data 172 Appendix F: Lists of pollinator species by functional group 2015 Functional group Bee scientific names Common names "Honey bees" Apis mellifera European honey bee "Bumble bees" Bombus species (14 species) Bumble bees "Large Carpenter bees" Xylocopa virginica Eastern carpenter bee "Large Dark bees" Andrena species (85 species) Mining bees Cemolobus ipomoeae Colletes species (14 species) Plasterer and Cellophane bees Dieunomia species (2 species) Florilegus condignus Macropis species (2 species) Melitoma taurea Melitta species (2 species) Nomia species (2 species) Xenoglossa strenua “Small Dark bees” "Yellow- faced bees" Calliopsis andreniformis Hylaeus species (13 species) Yellow-masked bees Perdita species (7 species) Protandrena abdominalis Pseudopanurgus species (6 species) "Dark Sweat bees" Halictus species (6 species) Lasioglossum species (77 species) "Small Carpenter bees" Ceratina species (5 species) Small or Dwarf Carpenter bees "Small Mining bees" Panurginus species (3 species) 173 "Green Sweat bees" Agapostemon species (4 species) Augochlora pura Augochlorella species (2 species) Augochloropsis species (2 species) "Dark Hairy Belly bees" Anthidiellum notatum Anthidium species (2 species) Wool Carder Bees Chelostoma philadelphi Heraides species (4 species) Hoplitis species (6 species) Lithurgus chrysurus Medierranean Wood-boring bee Megachile species (23 species) Resin and Leaf-cutting bees Paranthidium jugatorum Pseduanthidium nanum "Metallic Hairy Belly Bees" Osmia species (19 species) Orchard and Mason bees "Cuckoo Bees" Coelioxys species (10 species) Epeoloides pilosula Epeolus species (8 species) Holcopasites species (3 species) Nomada species (32 species) Sphecodes species (16 species) Stelis species (3 species) Triepeolus species (10 species) 174 2016 Functional group Bee scientific names Common names "Honey bees" Apis mellifera European honey bee "Bumble bees" Bombus species (14 species) Bumble bees "Large Carpenter bees" Xylocopa virginica Eastern carpenter bee "Large Dark bees" with abdominal scopa Anthidium manicatum Wool carder bee Lithurgus chrysurus Medierranean Wood-boring bee Megachile species (26 species) Resin and Leaf- cutting bees without abdominal scopa Andrena species (85 species) Mining bees Cemolobus ipomoeae Colletes species (14 species) Plasterer and Cellophane bees Dieunomia species (2 species) Florilegus condignus Macropis species (2 species) Melitoma taurea Melitta species (2 species) Nomia species (2 species) Xenoglossa strenua “Small Dark bees” with abdonminal scopa Anthidiellum notatum Anthidium oblongatum Wool Carder bees Chelostoma philadelphi Heraides species (4 species) Hoplitis species (6 species) Megachile species (4 species) Paranthidium jugatorum Pseduanthidium nanum 175 “Small Dark bees” without abdominal scopa "Yellow- faced bees" Calliopsis andreniformis Hylaeus species (13 species) Yellow-masked bees Perdita species (7 species) Protandrena abdominalis Pseudopanurgus species (6 species) "Dark Sweat bees" Halictus species (6 species) Lasioglossum species (77 species) "Small Carpenter bees" Ceratina species (5 species) Small or Dwarf Carpenter bees "Small Mining bees" Panurginus species (3 species) Metallic bees Metallic green bees Agapostemon species (4 species) Augochlora pura Augochlorella species (2 species) Augochloropsis species (2 species) Other metallic bees Osmia species (19 species) Orchard and Mason bees 176 Appendix G: List of species collected in the 2016 specimen collection Bee species Baltimore Frederick Prince George's Allegany Bowl trap Hand collection Bowl trap Hand collection Bowl trap Hand collection Bowl trap Hand collection Agapostemon texanus Cresson 1 0 0 0 0 0 0 0 Agapostemon virescens Fabricius 23 5 6 0 20 0 3 0 Andrena alleghaniensis Viereck 0 1 0 0 0 0 0 0 Andrena crataegi Robertson 0 2 0 0 0 0 0 0 Andrena cressonii Robertson 0 57 0 0 0 0 0 0 Andrena forbsii Robertson 0 2 0 0 0 0 0 0 Andrena hippotes Robertson 0 1 0 0 0 0 0 0 Andrena illini Bouseman and LaBerge 0 4 0 0 0 0 0 0 Andrena imitatrix Cresson /morrisonella Viereck 0 3 0 0 0 0 0 0 Andrena nasonii Robertson 0 19 0 1 3 0 0 0 Andrena nuda Robertson 0 1 0 0 0 0 0 0 Andrena perplexa Smith 1 23 0 0 0 0 0 0 Andrena personata Robertson 0 1 0 0 0 0 0 0 Andrena robertsonii Dalla Torre 0 11 0 0 0 0 0 0 Andrena spp. 0 1 0 0 0 0 0 0 Andrena vicina Smith 0 29 0 0 0 0 0 1 Anthidium manicatum L. 0 0 0 6 0 0 0 5 Anthidium oblongatum Illiger 2 0 0 1 1 0 2 0 177 Bee species Baltimore Frederick Prince George's Allegany Bowl trap Hand collection Bowl trap Hand collection Bowl trap Hand collection Bowl trap Hand collection Apis mellifera L. 13 132 4 6 0 1 3 24 Augochlora pura Say 2 6 0 0 11 1 1 1 Augochlorella aurata Smith 27 37 7 20 13 9 1 3 Augochloropsis metallica metallica Fabricius 0 3 0 1 0 0 0 0 Bombus bimaculatus Cresson 0 24 1 0 0 6 1 3 Bombus fervidus Fabricius 0 1 0 0 0 0 0 0 Bombus griseocollis De Geer 0 53 0 3 0 0 0 18 Bombus impatiens Cresson 87 182 0 37 0 4 0 9 Bombus perplexus Cresson 0 16 0 0 0 0 0 0 Bombus spp. 0 3 0 0 0 0 0 0 Calliopsis andreniformis Smith 9 0 2 0 3 0 2 1 Ceratina calcarata Robertson 2 3 3 2 0 5 1 7 Ceratina dupla Say 0 2 0 4 0 4 0 0 Ceratina mikmaqi Rehan and Sheffield 5 1 4 4 11 9 1 0 Ceratina spp. 2 3 0 1 1 5 0 0 Ceratina strenua Smith 15 4 7 12 69 23 3 1 Coelioxys sayi Robertson 0 3 0 0 0 2 0 0 Colletes latitarsis Robertson 0 0 0 0 0 0 0 2 Colletes thoracicus Smith 0 1 0 0 0 0 0 0 Epeolus bifasciatus Cresson 0 3 0 0 0 0 0 0 178 Bee species Baltimore Frederick Prince George's Allegany Bowl trap Hand collection Bowl trap Hand collection Bowl trap Hand collection Bowl trap Hand collection Halictus confusus Smith 4 3 0 1 10 4 0 1 Halictus ligatus Say/poeyi Lepeletier 10 71 6 38 13 34 5 10 Halictus parallelus Say 0 0 0 0 1 0 0 0 Heriades carinata Cresson 0 4 1 0 0 0 0 0 Hoplitis pilosifrons Cresson 1 0 0 0 0 0 0 0 Hoplitis spp. 0 0 0 0 0 0 1 0 Hylaeus affinis Smith/modestus Say 3 2 2 12 0 2 0 0 Hylaeus mesillae Cockerell 3 3 0 1 0 0 0 0 Hylaeus modestus Say 0 0 0 1 0 0 0 0 Lasioglossum admirandum Sandhouse 0 0 0 0 1 1 0 0 Lasioglossum apocyni Mitchell 0 0 0 0 0 0 4 7 Lasioglossum bruneri Crawford 0 1 1 0 10 0 0 0 Lasioglossum callidum Sandhouse 5 3 1 0 0 0 0 0 Lasioglossum coreopsis Robertson 0 0 0 0 1 0 0 0 Lasioglossum cressonii Robertson 0 0 2 0 0 0 0 0 Lasioglossum ephialtum Gibbs 0 1 0 0 0 0 0 0 Lasioglossum fuscipenne Smith 0 0 0 0 1 0 0 0 Lasioglossum hitchensi Gibbs 36 22 10 6 27 3 10 4 Lasioglossum illinoense Robertson 2 0 1 1 0 0 1 0 Lasioglossum imitatum Smith 0 6 2 9 17 12 0 1 Lasioglossum leucocomum Lovell 0 0 0 0 2 0 0 0 179 Bee species Baltimore Frederick Prince George's Allegany Bowl trap Hand collection Bowl trap Hand collection Bowl trap Hand collection Bowl trap Hand collection Lasioglossum pilosum Smith 8 4 1 0 21 0 0 0 Lasioglossum platyparium Robertson 3 0 1 1 1 0 0 0 Lasioglossum spp. 9 6 2 4 5 9 2 0 Lasioglossum tegulare Robertson 7 1 5 0 6 2 1 1 Lasioglossum trigeminum Gibbs 42 9 8 3 7 1 1 0 Lasioglossum versatum Robertson 12 3 0 0 0 0 1 0 Lasioglossum weemsi Mitchell 3 2 1 0 2 0 0 0 Lasioglossum zephyrum Smith 0 0 1 0 0 0 0 0 Megachile brevis Say 0 0 1 0 1 0 0 0 Megachile campanulae Robertson 0 0 0 0 0 1 0 0 Megachile exilis Cresson 0 3 0 0 0 0 0 1 Megachile frugalis Cresson 1 0 0 0 0 0 0 1 Megachile inimica Cresson 0 0 0 2 0 0 0 0 Megachile mendica Cresson 0 5 0 0 1 2 0 2 Megachile pugnata Say 0 0 0 0 0 0 0 1 Megachile rotundata Fabricius 0 1 0 0 0 0 0 1 Megachile sculpturalis Smith 0 1 0 2 1 0 0 1 Melissodes bimaculata Lepeletier 0 0 0 2 0 0 0 0 Melissodes denticulata Smith 0 15 0 0 0 0 2 0 Melissodes spp. 0 0 0 14 0 0 0 0 Melissodes tepaneca Cresson 0 1 0 1 0 0 2 0 Melissodes trinodis Robertson 0 5 0 0 0 0 0 0 180 Bee species Baltimore Frederick Prince George's Allegany Bowl trap Hand collection Bowl trap Hand collection Bowl trap Hand collection Bowl trap Hand collection Nomada pygmaea Cresson 0 0 0 0 0 0 0 0 Osmia caerulescens L. 0 0 0 0 1 0 0 0 Osmia pumila Cresson 0 0 0 0 0 0 0 1 Peponapis pruinosa Say 0 0 0 0 0 0 1 2 Ptilothrix bombiformis Cresson 0 0 0 0 3 0 0 0 Stelis lateralis Cresson 1 0 0 0 2 0 1 0 Triepeolus remigatus Fabricius 0 1 0 0 0 0 0 0 Xylocopa virginica L. 0 2 2 0 0 0 0 0 181 References Adamson NL, Borders B, Cruz JK, et al (2015) Pollinator Plants: Mid-Atlantic Region. Alarcón R, Waser NM, Ollerton J (2008) Year-to-year variation in the topology of a plant–pollinator interaction network. Oikos 117:1796–1807. doi: 10.1111/j.0030- 1299.2008.16987.x Anderson B, Johnson SD (2008) The Geographical Mosaic of Coevolution in a Plant– Pollinator Mutualism. Evolution 62:220–225. doi: 10.1111/j.1558- 5646.2007.00275.x Bahlai CA, Landis DA (2016) Predicting plant attractiveness to pollinators with passive crowdsourcing. R Soc Open Sci 3:150677. doi: 10.1098/rsos.150677 Bartomeus I, Vilà M, Santamaría L (2008) Contrasting Effects of Invasive Plants in Plant-Pollinator Networks. Oecologia 155:761–770. Bernauer OM (2017) The Maryland Native Pollinator Survey: a useful method for citizen scientists to survey pollinators visiting floral resources. Master, University of Maryland Bernauer OM, Gaines-Day HR, Steffan SA (2015) Colonies of Bumble Bees (Bombus impatiens) Produce Fewer Workers, Less Bee Biomass, and Have Smaller Mother Queens Following Fungicide Exposure. Insects 6:478–488. doi: 10.3390/insects6020478 Birkin L, Goulson D (2015) Using citizen science to monitor pollination services. Ecol Entomol 40:3–11. doi: 10.1111/een.12227 Bonney R (2009) Public Participation in Scientific Research: Defining the Field and Assessing Its Potential for Informal Science Education. A CAISE Inquiry Group Report. Bonter D (2012) From Backyard Observations to Continent-Wide Trends: Lessons from the First Twenty-two Years of Project FeederWatch. In: Citizen Science: Public participation in environmental research. Cornell University Press, Ithaca, NY, Bosch J, Kemp WP (2000) Development biology and rearing methods for Osmia bees used as crop pollinators. Breed GA, Stichter S, Crone EE (2013) Climate-driven changes in northeastern US butterfly communities. Nat Clim Change 3:142–145. doi: 10.1038/nclimate1663 Calderone NW (2012) Insect Pollinated Crops, Insect Pollinators and US Agriculture: Trend Analysis of Aggregate Data for the Period 1992–2009. PLoS ONE 7:e37235. doi: 10.1371/journal.pone.0037235 182 Colla SR, Otterstatter MC, Gegear RJ, Thomson JD (2006) Plight of the bumble bee: Pathogen spillover from commercial to wild populations. Biol Conserv 129:461– 467. doi: 10.1016/j.biocon.2005.11.013 Corbet SA (2000) Butterfly nectaring flowers: butterfly morphology and flower form. Entomol Exp Appl 96:289–298. doi: 10.1046/j.1570-7458.2000.00708.x Corbet SA, Bee J, Dasmahapatra K, et al (2001) Native or Exotic? Double or Single? Evaluating Plants for Pollinator-friendly Gardens. Ann Bot 87:219–232. doi: 10.1006/anbo.2000.1322 Crall AW, Jordan R, Holfelder K, et al (2013) The impacts of an invasive species citizen science training program on participant attitudes, behavior, and science literacy. Public Underst Sci 22:745–764. doi: 10.1177/0963662511434894 Curtis RJ, Brereton TM, Dennis RLH, et al (2015) Butterfly abundance is determined by food availability and is mediated by species traits. J Appl Ecol 52:1676–1684. doi: 10.1111/1365-2664.12523 Deguines N, Julliard R, Flores M de, Fontaine C (2012) The Whereabouts of Flower Visitors: Contrasting Land-Use Preferences Revealed by a Country-Wide Survey Based on Citizen Science. PLOS ONE 7:e45822. doi: 10.1371/journal.pone.0045822 Dickinson J, Zuckerberg B, Bonter D (2010) Citizen Science as an Ecological Research Tool: Challenges and Benefits. Annu Rev Ecol Evol Syst 41:149–172. doi: 10.1146/annurev-ecolsys-102209-144636 Dickinson JL, Bonney R (2012) Citizen Science: Public Participation in Environmental Research. Cornell University Press Dickinson JL, Shirk J, Bonter D, et al (2012) The current state of citizen science as a tool for ecological research and public engagement. Front Ecol Environ 10:291–297. doi: 10.1890/110236 Domroese MC, Johnson EA (2017) Why watch bees? Motivations of citizen science volunteers in the Great Pollinator Project. Biol Conserv 208:40–47. doi: 10.1016/j.biocon.2016.08.020 Donovall L, vanEngelsdorp D (2009) Pennsylvania Native Bee Survey: Citizen Scientist Pollinator Monitoring Guide. Dormann CF, Fründ J, Blüthgen N, Gruber B (2009) Indices, Graphs and Null Models: Analyzing Bipartite Ecological Networks. Dormann CF, Gruber B, Fründ J (2008) Introducing the bipartite Package: Analysing Ecological Networks. R news 8:8–11. 183 Droege S, Tepedino VJ, Lebuhn G, et al (2010) Spatial patterns of bee captures in North American bowl trapping surveys. Insect Conserv Divers 3:15–23. doi: 10.1111/j.1752-4598.2009.00074.x Durkin P, Frye J, Gibbs D, et al (2013) Conservation and Management of the Baltimore Checkerspot (Euphydryas phaeton Drury) in Maryland: Strategies for Statewide Monitoring and for Wetland Restoration, Captive Breeding and Release in the Piedmont Region. Evans C, Abrams E, Reitsma R, et al (2005) The Neighborhood Nestwatch Program: Participant Outcomes of a Citizen-Science Ecological Research Project. Conserv Biol 19:589–594. doi: 10.1111/j.1523-1739.2005.00s01.x Evans CA, Abrams ED, Rock BN, Spencer SL (2001) Student/Scientist Partnerships. Am Biol Teach 63:318–323. doi: 10.1662/0002-7685(2001)063[0318:SSP]2.0.CO;2 Fenster CB, Armbruster WS, Wilson P, et al (2004) Pollination Syndromes and Floral Specialization. Annu Rev Ecol Evol Syst 35:375–403. doi: 10.1146/annurev.ecolsys.34.011802.132347 Fetridge ED, Ascher JS, Langellotto GA (2008) The Bee Fauna of Residential Gardens in a Suburb of New York City (Hymenoptera: Apoidea). Ann Entomol Soc Am 101:1067–1077. doi: 10.1603/0013-8746-101.6.1067 Fiedler AK, Landis DA (2007) Attractiveness of Michigan Native Plants to Arthropod Natural Enemies and Herbivores. Environ Entomol 36:751–765. doi: 10.1603/0046-225X(2007)36[751:AOMNPT]2.0.CO;2 Forister ML, McCall AC, Sanders NJ, et al (2010) Compounded effects of climate change and habitat alteration shift patterns of butterfly diversity. Proc Natl Acad Sci 107:2088–2092. doi: 10.1073/pnas.0909686107 Fortuna MA, Bascompte J (2006) Habitat loss and the structure of plant–animal mutualistic networks. Ecol Lett 9:281–286. doi: 10.1111/j.1461- 0248.2005.00868.x Fowler J, Droege S (2016) Specialist bees of the Mid-Atlantic and Northeastern United States. Frankie GW, Thorp RW, Schindler M, et al (2005) Ecological Patterns of Bees and Their Host Ornamental Flowers in Two Northern California Cities. J Kans Entomol Soc 78:227–246. Gaines-Day HR, Gratton C (2016) Crop yield is correlated with honey bee hive density but not in high-woodland landscapes. Agric Ecosyst Environ 218:53–57. doi: 10.1016/j.agee.2015.11.001 184 Garbuzov M, Ratnieks FLW (2014) Quantifying variation among garden plants in attractiveness to bees and other flower-visiting insects. Funct Ecol 28:364–374. doi: 10.1111/1365-2435.12178 Garbuzov M, Ratnieks FLW (2015) Using the British National Collection of Asters to Compare the Attractiveness of 228 Varieties to Flower-Visiting Insects. Environ Entomol 44:638–646. doi: 10.1093/ee/nvv037 Gardiner MM, Allee LL, Brown PM, et al (2012) Lessons from lady beetles: accuracy of monitoring data from US and UK citizen-science programs. Front Ecol Environ 10:471–476. doi: 10.1890/110185 Garibaldi LA, Steffan-Dewenter I, Winfree R, et al (2013) Wild Pollinators Enhance Fruit Set of Crops Regardless of Honey Bee Abundance. Science 339:1608–1611. doi: 10.1126/science.1230200 Genersch E, Yue C, Fries I, de Miranda JR (2006) Detection of Deformed wing virus, a honey bee viral pathogen, in bumble bees (Bombus terrestris and Bombus pascuorum) with wing deformities. J Invertebr Pathol 91:61–63. doi: 10.1016/j.jip.2005.10.002 Gerling D, Hermann HR (1978) Biology and mating behavior of Xylocopa virginica L. (Hymenoptera, Anthophoridae). Behav Ecol Sociobiol 3:99–111. doi: 10.1007/BF00294984 Gibbs D, Smith D, Mason S (1997) Piedmont (Central Maryland). Goulson D, Nicholls E, Botías C, Rotheray EL (2015) Bee declines driven by combined stress from parasites, pesticides, and lack of flowers. Science 347:1255957. doi: 10.1126/science.1255957 Hayes RF (2016) Pollinator Friendly Gardening: Gardening for Bees, Butterflies, and Other Pollinators. Voyageur Press Heinrich B (2004) Bumblebee Economics. Harvard University Press Henderson S, Ward D L, Meymaris KK, et al (2012) Project Budburst: Citizen Science for All Seasons. In: Citizen Science: Public Participation in Enivronmental Research. Cornell University Press, Ithaca, NY, pp 50–57 Hidalgo-Ruz V, Thiel M (2013) Distribution and abundance of small plastic debris on beaches in the SE Pacific (Chile): A study supported by a citizen science project. Mar Environ Res 87–88:12–18. doi: 10.1016/j.marenvres.2013.02.015 Holm H (2014) Pollinators of Native Plants: Attract, Observe and Identify Pollinators and Beneficial Insects with Native Plants. Pollination Press LLC 185 Hopkins GW, Freckleton RP (2002) Declines in the numbers of amateur and professional taxonomists: implications for conservation. Anim Conserv 5:245–249. doi: 10.1017/S1367943002002299 Hu S, Dilcher DL, Jarzen DM, Taylor DW (2008) Early steps of angiosperm–pollinator coevolution. Proc Natl Acad Sci 105:240–245. doi: 10.1073/pnas.0707989105 Irwin A, Georg S, Vergragt P (1994) The social management of environmental change. Futures 26:323–334. doi: 10.1016/0016-3287(94)90018-3 Isaacs R, Tuell J, Fiedler A, et al (2009) Maximizing arthropod-mediated ecosystem services in agricultural landscapes: the role of native plants. Front Ecol Environ 7:196–203. doi: 10.1890/080035 Jennersten O (1984) Flower visitation and pollination efficiency of some North European butterflies. Oecologia 63:80–89. doi: 10.1007/BF00379789 Katz DL, Njike VY, Rhee LQ, et al (2010) Performance characteristics of NuVal and the Overall Nutritional Quality Index (ONQI). Am J Clin Nutr 91:1102S–1108S. doi: 10.3945/ajcn.2010.28450E Kendall WL, Peterjohn BG, Sauer JR (1996) First-Time Observer Effects in the North American Breeding Bird Survey. The Auk 113:823–829. doi: 10.2307/4088860 Klein A-M, Vaissière BE, Cane JH, et al (2007) Importance of pollinators in changing landscapes for world crops. Proc R Soc B Biol Sci 274:303–313. doi: 10.1098/rspb.2006.3721 Kountoupes DL, Oberhauser KS (2008) Citizen Science and Youth Audiences: Educational Outcomes of the Monarch Larva Monitoring Project - ProQuest. 1:10–20. Kremen C, Williams NM, Thorp RW (2002) Crop pollination from native bees at risk from agricultural intensification. Proc Natl Acad Sci 99:16812–16816. doi: 10.1073/pnas.262413599 LeBaron GS (2016) The 116th Christmas Bird Count Summary. In: Audubon. http://www.audubon.org/news/the-116th-christmas-bird-count-summary. Accessed 4 Feb 2017 Menzel R, Greggers U, Hammer M (1993) Functional Organization of Appetitive Learning and Memory in a Generalist Pollinator, the Honey Bee. In: Papaj DR, Lewis AC (eds) Insect Learning. Springer US, pp 79–125 Meymaris K, Henderson S, Alaback P, Havens K (2008) Project BudBurst: Citizen Science for All Seasons. 186 Michener CD (2007) The bees of the world, 2nd edn. The Johns Hopkins University Press, Baltimore, Maryland Morandin LA, Kremen C (2013) Hedgerow restoration promotes pollinator populations and exports native bees to adjacent fields. Ecol Appl 23:829–839. doi: 10.1890/12-1051.1 National Research Council, Division of Behavioral and Social Sciences and Education, Center for Education, et al (2009) Learning Science in Informal Environments: People, Places, and Pursuits. National Academies Press O’Brien DM, Boggs CL, Fogel ML (2004) Making eggs from nectar: the role of life history and dietary carbon turnover in butterfly reproductive resource allocation. Oikos 105:279–291. doi: 10.1111/j.0030-1299.2004.13012.x Oksanen J, Blanchet FG, Friendly M, et al (2017) vegan: Community Ecology Package. Ollerton J, Winfree R, Tarrant S (2011) How many flowering plants are pollinated by animals? Oikos 120:321–326. doi: 10.1111/j.1600-0706.2010.18644.x Owen J (2010) Wildlife of a Garden: A Thirty-year Study. Royal Horticultural Society Patten KD, Shanks CH, Mayer DF (1993) Evaluation of herbaceous plants for attractiveness to bumble bees for use near cranberry farms. J Apic Res 32:73–79. doi: 10.1080/00218839.1993.11101290 Potts SG, Biesmeijer JC, Kremen C, et al (2010) Global pollinator declines: trends, impacts and drivers. Trends Ecol Evol 25:345–353. doi: 10.1016/j.tree.2010.01.007 Potts SG, Imperatriz-Fonseca V, Ngo HT, et al (2016) Safeguarding pollinators and their values to human well-being. Nature 540:220–229. doi: 10.1038/nature20588 R Core Team (2016) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria Razanajatovo M, Föhr C, Fischer M, et al (2015) Non-naturalized alien plants receive fewer flower visits than naturalized and native plants in a Swiss botanical garden. Biol Conserv 182:109–116. doi: 10.1016/j.biocon.2014.11.043 Reger JP, Cleaves ET (2008) Explanatory Text for the Physiographic Map of Maryland. Ricketts TH, Regetz J, Steffan-Dewenter I, et al (2008) Landscape effects on crop pollination services: are there general patterns? Ecol Lett 11:499–515. doi: 10.1111/j.1461-0248.2008.01157.x Rogstadius J, Kostakos V, Kittur A, et al (2011) An Assessment of Intrinsic and Extrinsic Motivation on Task Performance in Crowdsourcing Markets. 187 Salisbury A, Armitage J, Bostock H, et al (2015) EDITOR’S CHOICE: Enhancing gardens as habitats for flower-visiting aerial insects (pollinators): should we plant native or exotic species? J Appl Ecol 52:1156–1164. doi: 10.1111/1365- 2664.12499 Sam Droege (2015) The Very Handy Manual: How to Catch and Identify Bees and Manage a Collection. Scanlon E, Woods W, Clow D (2014) Informal participation in science in the UK: identification, location and mobility with iSpot. J Educ Technol Soc 17:58–71. Scott-Dupree CD, Conroy L, Harris CR (2009) Impact of Currently Used or Potentially Useful Insecticides for Canola Agroecosystems on Bombus impatiens (Hymenoptera: Apidae), Megachile rotundata (Hymentoptera: Megachilidae), and Osmia lignaria (Hymenoptera: Megachilidae). J Econ Entomol 102:177–182. doi: 10.1603/029.102.0125 Shackleton K, Ratnieks FLW (2016) Garden varieties: How attractive are recommended garden plants to butterflies? J Insect Conserv 20:141–148. doi: 10.1007/s10841- 015-9827-9 Silvertown J (2009) A new dawn for citizen science. Trends Ecol Evol 24:467–471. doi: 10.1016/j.tree.2009.03.017 Simpson EH (1949) Measurement of diversity. Nature 163:688. Smith KM, Loh EH, Rostal MK, et al (2014) Pathogens, Pests, and Economics: Drivers of Honey Bee Colony Declines and Losses. EcoHealth 10:434–445. doi: 10.1007/s10393-013-0870-2 Smith RM, Thompson K, Hodgson JG, et al (2006) Urban domestic gardens (IX): Composition and richness of the vascular plant flora, and implications for native biodiversity. Biol Conserv 129:312–322. doi: 10.1016/j.biocon.2005.10.045 Smith RM, Warren PH, Thompson K, Gaston KJ (2005) Urban domestic gardens (VI): environmental correlates of invertebrate species richness. In: Hawksworth DL, Bull AT (eds) Human Exploitation and Biodiversity Conservation. Springer Netherlands, pp 75–98 Stanley SM (1979) Macroevolution, Pattern and Process. W. H. Freeman Sullivan BL, Wood CL, Iliff MJ, et al (2009) eBird: A citizen-based bird observation network in the biological sciences. Biol Conserv 142:2282–2292. doi: 10.1016/j.biocon.2009.05.006 Tallamy DW (2007) Bringing Nature Home: How Native Plants Sustain Wildlife in Our Gardens. Timber Press 188 The Xerces Society (2011) Attracting Native Pollinators: The Xerces Society Guide to Conserving North American Bees and Butterflies and Their Habitat. Storey Publishing The Xerces Society (2016b) Gardening for Butterflies: How You Can Attract and Protect Beautiful, Beneficial Insects. Timber Press The Xerces Society (2016a) 100 Plants to Feed the Bees: Provide a Healthy Habitat to Help Pollinators Thrive. Storey Publishing Triezenberg HA, Knuth BA, Yuan YC, Dickinson JL (2012) Internet-Based Social Networking and Collective Action Models of Citizen Science. In: Dickinson JL, Bonney R, Louv R (eds) Citizen Science: Public Participation in Environmental Research. Cornell University Press, Ithaca, NY, pp 214–225 Tuell JK, Fiedler AK, Landis D, Isaacs R (2008) Visitation by Wild and Managed Bees (Hymenoptera: Apoidea) to Eastern U.S. Native Plants for Use in Conservation Programs. Environ Entomol 37:707–718. doi: 10.1603/0046- 225X(2008)37[707:VBWAMB]2.0.CO;2 Tweddle JC, Robinson LD, Pocock MJO, Roy HE (2012) Guide to citizen science: developing, implementing and evaluating citizen science to study biodiversity and the environment in the UK. NERC/Centre for Ecology & Hydrology, Wallingford Ullmann K, Shih T, Vaughan M, Kremen C (2007) California Pollinator Project: Citizen Scientist Pollinator Monitoring Guide. Vanbergen AJ, The Insect Pollinators Initiative (2013) Threats to an ecosystem service: pressures on pollinators. Front Ecol Environ 11:251–259. doi: 10.1890/120126 Vicens N, Bosch J (2000) Pollinating Efficacy of Osmia cornuta and Apis mellifera (Hymenoptera: Megachilidae, Apidae) on “Red Delicious” Apple. Environ Entomol 29:235–240. doi: 10.1093/ee/29.2.235 Visscher PK, Seeley TD (1982) Foraging Strategy of Honeybee Colonies in a Temperate Deciduous Forest. Ecology 63:1790–1801. doi: 10.2307/1940121 Waser NM, Ollerton J (2006) Plant-Pollinator Interactions: From Specialization to Generalization. University of Chicago Press Wcislo WT, Cane JH (1996) Floral Resource Utilization by Solitary Bees (Hymenoptera: Apoidea) and Exploitation of Their Stored Foods by Natural Enemies. Annu Rev Entomol 41:257–286. doi: 10.1146/annurev.en.41.010196.001353 Weckel ME, Mack D, Nagy C, et al (2010) Using Citizen Science to Map Human– Coyote Interaction in Suburban New York, USA. J Wildl Manag 74:1163–1171. doi: 10.2193/2008-512 189 Wickham H (2009) ggplot2 - Elegant Graphics for Data Analysis. Springer Williams NM, Ward KL, Pope N, et al (2015) Native wildflower plantings support wild bee abundance and diversity in agricultural landscapes across the United States. Ecol Appl 25:2119–2131. doi: 10.1890/14-1748.1 Williams PH, Thorp RW, Richardson LL, Colla SR (2014) Bumble Bees of North America: An Identification Guide. Princeton University Press, Princeton Winfree R, Aguilar R, Vázquez DP, et al (2009) A meta-analysis of bees’ responses to anthropogenic disturbance. Ecology 90:2068–2076. doi: 10.1890/08-1245.1 Winfree R, Williams NM, Dushoff J, Kremen C (2007) Native bees provide insurance against ongoing honey bee losses. Ecol Lett 10:1105–1113. doi: 10.1111/j.1461- 0248.2007.01110.x