USING DNA BARCODING to INVESTIGATE the DIET and FOOD SUPPLY of a DECLINING AERIAL INSECTIVORE, the BARN SWALLOW (Hirundo Rustica)

USING DNA BARCODING TO INVESTIGATE THE DIET AND FOOD SUPPLY OF A DECLINING AERIAL INSECTIVORE, THE BARN SWALLOW (Hirundo rustica)

A Thesis Submitted to the Committee on Graduate Studies in Partial Fulfillment of the Requirements for the Degree of Master of Science in the Faculty of Arts and Science

TRENT UNIVERSITY Peterborough, Ontario, Canada © Copyright by Beverly McClenaghan 2017 Environmental and Life Sciences M.Sc. Graduate Program January 2018 Abstract

Using DNA barcoding to investigate the diet and food supply of a declining aerial insectivore, the Barn Swallow (Hirundo rustica) Beverly McClenaghan

Barn Swallow (Hirundo rustica) populations have declined in North America over the past 40 years and they are listed as Threatened in Ontario, Canada. Changes in the food supply have been hypothesized as a potential cause of this population decline. I used DNA barcoding to investigate the diet and food supply of Barn Swallows and to determine if the food supply affects their reproductive performance. In two breeding seasons, I monitored nests, collected fecal samples, and monitored prey availability by collecting insects from the habitat surrounding breeding sites using Malaise traps. I used

DNA barcoding to identify insect specimens collected from the habitat and to identify prey items from Barn Swallow nestling fecal samples. I found that Barn Swallow nestlings were fed a very broad range of prey items but were fed larger prey items more frequently. Prey availability was not related to the timing of reproduction, the number of nests at a breeding site, or the reproductive output of individual nests. This study provides information on the diet composition of Barn Swallows in North America and suggests that food limitation during the breeding season may not be a major factor in their population decline.

Keywords: aerial insectivore, diet, DNA barcoding, Hirundo rustica, metabarcoding, reproductive success

Acknowledgements

I would like to thank my supervisors, Dr. Kevin Kerr and Dr. Erica Nol, for their support, guidance and encouragement throughout my project. Thank you to my committee members, Dr. Dave Beresford and Dr. Joe Nocera, for their useful comments and insight which was always appreciated. I am very grateful to those who helped and accompanied me in the field: Ariel Lenske, who was great company and a great resource during my first field season, and Melissa Brochu and Chelsea May, who were generous enough to help with my fieldwork while I was away. I would like to thank the landowners I worked with - Wayne Bolton, Bill Ogilvie, John Cooper, Troy, Barb Chatten, Bob Lenard, Steve Plunkett, the Saunders, Pete Hogan, and Jim Seymour - for their interest in Barn Swallows, for allowing me access to their property at all hours of the day and night, and for their kindness. My fellow Nol lab members and my officemates, both new and old, were always there to talk about statistical problems, to give support and encouragement, to take some much needed breaks, and to always keeping me smiling. I would like to acknowledge financial support from the Toronto Zoo through my NSERC IPS, the Purple Martin Conservation Association and the Toronto Entomologists Association. This project was also funded through a Species at Risk Research Fund of Ontario grant awarded to Dr. Kevin Kerr. Finally, I’d like to thank my family: my parents for instilling a scientific curiosity in me and all my siblings for their interest and encouragement. A big thank you goes to my aunt and uncle, Andy McClenaghan and Fiona Tokic, for taking me in as their honorary daughter while I worked at the Toronto Zoo.

iii

Author Contributions & Formatting

This thesis was a collaborative effort and I would like to differentiate my work from the work of my collaborators. In addition, Chapters 2 & 3 were written for publication and formatted accordingly. Chapter 2 is formatted for submission to Proceedings of the Royal Society B: Biological Sciences. Chapter 3 is formatted for submission to the Journal of Avian Biology. The methods in Chapters 2 and 3 were designed by my advisors, Erica Nol and Kevin Kerr, and myself. I collected and analyzed the data and wrote these chapters. To reflect these contributions, I use "we" and "our" in these chapters when referring to the study. I wrote the General Introduction and Discussion and therefore, use “I” and “my” with reference to the overall study in these sections.

Table of Contents

Abstract ...... ii Acknowledgements ...... iii Author Contributions & Formatting ...... iv Table of Contents ...... v List of Figures ...... vii List of Tables ...... viii Chapter 1: General Introduction ...... 1 DIET Background ...... 2 Methods of Diet Analysis ...... 4 REPRODUCTIVE BEHAVIOUR Background ...... 7 Food Availability and Reproduction ...... 8 SCOPE & OBJECTIVES ...... 11 Chapter 2: DNA metabarcoding reveals the broad and flexible diet of a declining aerial insectivore ABSTRACT ...... 18 INTRODUCTION ...... 18 METHODS Study Site and Sample Collection ...... 22 DNA Extraction, Amplification, and Sequencing ...... 23 Custom Reference Library ...... 25 Taxonomic Identification ...... 26 Data Analysis ...... 27 RESULTS Prey Identity ...... 28 Prey Selection ...... 29 Dietary Flexibility...... 29 DISCUSSION ...... 30 REFERENCES ...... 36

Chapter 3: Does prey availability affect the reproductive performance of Barn Swallows, Hirundo rustica, breeding in Ontario, Canada? ABSTRACT ...... 46 INTRODUCTION ...... 47 METHODS Study Site ...... 52 Nest Monitoring ...... 52 Insect Collection ...... 53 Insect Identification ...... 53 Data Analysis ...... 55 RESULTS Colony Size and Rates of Second Brooding ...... 58 Timing of Reproduction ...... 59 Reproductive Output...... 59 DISCUSSION ...... 60 REFERENCES ...... 66 Chapter 4: General Discussion ...... 79 REFERENCES ...... 85 Appendix A ...... 87 Appendix B ...... 117

List of Figures

Figure 1.1 Location of the breeding sites shown by red dots with numbers indicating the number of pairs at each site in each year of the study (2015/2016). The red star on the inset map shows the location of the study area within Canada...... 12

Figure 2.1 Accumulation curve of arthropod families identified in the diet of Barn Swallow nestlings. 95% confidence intervals are represented in grey...... 42

Figure 2.2 The frequency of occurrence of different arthropod families in the diet of Barn Swallows compared to the availability of these arthropod families in the habitat presented as a proportion of the total prey available in the habitat. Each bar on the x-axis represents a different family and bars are coloured by order. The five most frequently consumed families were: Tachinidae, Tipulidae, Limoniidae, Calliphoridae, and Anthomyiidae. The five most abundant insect families in the habitat were: Chironomidae, Sciaridae, Cecidomyiidae, Ceratopogonidae, and Staphylinidae...... 43

Figure 2.3 The frequency of occurrence of different sizes of arthropod in the diet of Barn Swallows compared to the availability of these sizes of arthropod prey in the habitat presented as a proportion of the total prey available in the habitat. Bars are coloured to show the proportion of operational taxonomic units (OTUs) in each size class represented by each order...... 44

Figure 3.1 Linear regressions of the number of old nests at each site with the number of pairs nesting at each site in (A) 2015 and (B) 2016. 95% confidence intervals are represented in grey. There was a significant relationship in both years (2015: r2 = 0.56, p = 0.01 and 2016: r2 = 0.71, p =0.002)...... 71

Figure 3.2 Prey availability (A) and prey demand (B) curves for Barn Swallows at 7 breeding sites plotted as proportional occurrence of the season total by day of the year. Each breeding site is represented by a separate curve...... 72

Figure 3.3 Prey availability and demand curves for Barn Swallows plotted together for each breeding site by day of the year. The red line represents the demand curve and the black line represents the prey availability curve. Curves are plotted as proportional occurrence of the season totals. Overlap is shaded in red...... 73

Figure 3.4 Predicted values from generalized linear models of (A) the numbers of eggs and clutch initiation date and (B) the number of fledglings and the hatch date. First broods and second broods are plotted separately. 95% confidence intervals are represented in grey...... 74

vii

List of Tables

Table 2.1 Summary of multivariate GLMs of the prey community in the diet of Barn Swallows and of the available prey community in the habitat...... 45

Table 3.1 Summary of the reproductive output and success of Barn Swallows at ten breeding sites in 2015 and 2016...... 75

Table 3.2 Linear models of the effects of five (2015) and nine (2016) single predictors on the number of pairs nesting at a site. Models with significant predictors are shown in bold (α = 0.05)...... 76

Table 3.3 Linear models of the proportion of pairs at each breeding site that had a second brood with twelve measures of prey availability and peak clutch initiation date as single predictors. Year and site were included as random effects in all models...... 77

Table 3.4 Reproductive output models using the number of eggs and the number of fledglings as measures of reproductive output. Single predictors included eight measures of prey availability, clutch initiation date and hatch date. All measures of prey availability for the fledgling model were from the 3-week period after hatching. Year, site, and brood were included in all models as random effects. Models with significant predictors are shown in bold (α = 0.05)...... 78

viii

Chapter 1: General Introduction

Barn Swallows (Hirundo rustica) are aerial foraging insectivorous birds with a worldwide distribution (Brown and Brown 1999). They are long-distance migrants that breed across the Northern Hemisphere and winter throughout the Southern Hemisphere

(Brown and Brown 1999, Turner 2006). Barn Swallows have a close association with humans since they are commonly found nesting on manmade structures, including barns, bridges, and culverts, and their distribution and abundance is often linked to human activities (Zink et al. 2006). While they remain abundant on a worldwide scale, local population declines have been observed across the range (Turner 2006, Lee et al. 2011,

Ambrosini et al. 2012,) and widespread declines have been observed in North America

(Nebel et al. 2010). Since 1980, Barn Swallow populations have declined by 77% in

Canada (Environment Canada 2014) and 66% in Ontario, where they are listed provincially as a species at risk (Heagy et al. 2014).

The population decline of Barn Swallows is part of a guild-wide population decline in aerial insectivores in North America over the past 30 years (Nebel et al. 2010).

This decline is steepest in eastern and northern North America (Nebel et al. 2010) where birds are at the northern edge of their range (Environment Canada 2014). Many possible causes for the decline have been suggested including loss of nesting or foraging habitat, climate change, and increased environmental contamination through pesticide applications (Heagy and McCraken 2004, COSEWIC 2011, NABCI Canada 2012).

Aerial insectivores represent a wide taxonomic breadth, occupy various habitats and exhibit different life histories (Cadman et al. 2007, Nocera et al. 2012). This suggests that

1 population declines in this guild may be related to the common food source that unites them: flying insects. A change in the supply of aerial insect prey could result from a shift in the timing of insect prey availability (Dunn et al. 2011) or a decrease in the overall abundance of insect prey (Benton et al. 2002). While there are few studies on the long- term monitoring of insect populations, local broad-scale declines in insect populations and global decreases in the abundance of certain insect groups have been documented

(Hallmann et al. 2017, Dirzo et al 2014). Worldwide declines in insect abundance could be caused by habitat loss, invasive species, anthropogenic climate change, and/or pesticide use (Potts et al 2010, Dirzo et al 2014).

The breeding season is an energetically demanding time for birds and the rapid growth of chicks requires abundant food resources (Martin 1987). Food limitation during the breeding season can have negative impacts on birds’ reproductive success and annual fecundity, thereby limiting their population size (Martin 1987, Sæther & Bakke 2000,

Ruffino et al. 2014). Understanding the food resources required by Barn Swallows and how food availability affects their reproductive behaviour will help determine if the food supply on the breeding grounds is contributing to their population decline and provide information for the management and conservation of Barn Swallows.

DIET

Background

Much of the research on the biology of Barn Swallows, including their diet, has focused on the nominate subspecies, Hirundo rustica rustica, that breeds across Europe

(Turner 2004). Barn Swallows breeding in North America are a different subspecies,

Hirundo rustica erythrogaster, which is distinguished by morphology and distribution.

The North American subspecies also displays different life history traits and behaviours

(e.g., male incubation; Turner 2006). Based on the analysis of mitochondrial DNA it has been suggested that H. r. erythrograster, is in the early stages of speciation from the Old

World subspecies (Zink et al. 1995, Sheldon et al. 2005, Dor et al 2010). The population dynamics of these two subspecies are different. There was a widespread population decline in H. r. rustica across Western Europe in the late 20th century that was associated with farmland bird population declines but this decline has mostly stabilized (Gregory et al. 2004, Sanderson et al. 2006). Local population declines are still being observed in

Denmark and Italy (Turner 2006, Ambrosini et al. 2012). The ongoing population decline of Barn Swallows in North America started in the 1980s and is associated with a guild- wide decline in aerial insectivores (Nebel et al. 2010). A change in the food supply has been suggested as a potential driver of population declines in North America and our understanding of the diet and prey selection of Barn Swallows will be important in determining how the changing food supply is affecting Barn Swallows.

Our knowledge of Barn Swallow diet in North America is very limited. A single, very old study, predating the population decline, examined the stomach contents of adult

Barn Swallows to show that they consumed prey items from many insect families in six different orders (Beal 1918). Diptera was the most abundant order identified but this order contained fewer prey items identified to genus and species compared to Coleoptera, which was less abundant but had a high level of genus and species identification (Beal

1918). This is likely due to differences in digestibility between prey items of different orders, which could bias the results (Major 1990). These results are consistent with the diet of adult Barn Swallows observed elsewhere in their range (e.g., Europe), where a

3 range of insect orders are consumed (Turner 1982a, b, Møller 2001). The taxonomic specificity of these studies has been very coarse, where prey items were generally identified to the order level. Prey selection patterns could be occurring at lower taxonomic levels since there is considerable variation in morphology and behaviour within insect orders. Studies have shown that adult Barn Swallows prefer large prey items, regardless of their taxonomic affiliation (Waugh 1978, Turner 1982a). Adult Barn

Swallows are generalist foragers and select prey items by size and they show a similar selection pattern when choosing prey to feed nestlings. Prey from a range of taxa are fed to nestlings in Europe and in North America (Hebblethwaite 1989, Orłowski and Karg

2011). Additionally, in Europe, Orlowski and Karg (2011) showed that Barn Swallow nestlings were fed large prey items more often than small prey items, a relationship that has not been tested in North America.

According to optimal foraging theory, larger prey is more profitable for Barn

Swallows, despite moving faster than smaller prey and requiring more energy to capture

(Turner 1982a). What we know of Barn Swallow diet suggests that they mostly follow optimal foraging theory since they do consume large prey more often than small prey

(Turner 1982a, Møller 2001, Orłowski and Karg 2011). However, they also consume more small prey items than optimal foraging theory predicts (Turner 1982a). Barn

Swallows might take advantage of local abundances of small prey items while still generally selecting larger prey items.

Methods of Diet Analysis

Many methods have been used to investigate the diet of insectivorous birds. Older methods include examining the stomach contents of dead birds and using neck collars as

4 ligatures allowing food boluses to be recovered (Beal 1918, Johnson et al. 1980). Prey items are only moderately digested and they can be readily identified and quantified.

However, these invasive methods are problematic when studying threatened taxa. An alternative method is to identify prey items morphologically from undigested remnants in feces (Ralph et al. 1985). This method can be used to quantify prey items (Kunz and

Whitaker Jr. 1982), but it can underestimate items consumed in large quantities and small or soft bodied prey which are more easily digested (Dickman and Huang 1988).

Additionally, prey items cannot easily be identified beyond the order level. Prey selection can occur at lower taxonomic levels with prey taxa in the same family varying greatly in size or life history (e.g., Oliver 1971, Chown and Gaston 2010). The diet of generalist consumers is harder to elucidate from fecal samples than that of specialists due to the greater variety of prey items that need to be identified and the range in digestibility and ease of identification (Major 1990). Most method validation involves feeding trials which use a limited number of prey items compared to the number of prey items generalists can consume in the wild (Kunz and Whitaker Jr. 1982, Dickman and Huang 1988).

More recently, molecular techniques have been used for diet analysis. DNA barcoding involves the sequencing of a standardized region of the mitochondrial genome, cytochrome c oxidase I (COI), from an unknown specimen and querying this sequence against a database of known sequences to identify the species (Hebert et al. 2003). This method allows for rapid identification of specimens with less taxonomic expertise

(Hebert et al. 2003). DNA metabarcoding refers to the use of DNA barcoding with high- throughput sequencing so that the target region of the genome can be sequenced from multiple specimens simultaneously in a single sample. In the context of diet analysis,

5 researchers can sequence the COI region from the remains of all the prey items in a single fecal sample. The sequences can then be compared to a database of known sequences to identify the unknown prey items. The DNA of prey items is degraded in fecal samples, but it can still be recovered and sequenced by targeting a short mini-barcode sequence

(Hajibabaei et al. 2006, Pompanon et al. 2012). DNA metabarcoding has been used to identify prey items from the feces of a variety of predator taxa, including insectivorous birds (Trevelline et al. 2016, Jedlicka et al. 2017). These studies make use of arthropod specific primers that target a mini-barcode within the COI region developed for use in the analysis of the diet of bats (Zeale et al. 2010).

DNA metabarcoding provides a non-invasive method of diet analysis which can reveal greater prey diversity than older methods (Bowser et al. 2013, Gerwing et al.

2016). Furthermore, DNA metabarcoding can facilitate the identification of prey items at higher taxonomic levels compared to older methods. In studying the diet of a generalist predator, this method could reveal previously unreported prey items in the diet. This method does not quantify prey items accurately within a sample due to the differential digestion of prey items and taxon-specific biases in PCR (Pompanon et al. 2012, Deagle et al. 2013), but the frequency of occurrence of a prey taxa across many samples can be used as a semi-quantitative measure of the importance of a prey taxa in the diet of a population of birds. DNA metabarcoding is a versatile tool that can be used to identify degraded prey items for non-invasive diet analysis.

REPRODUCTIVE BEHAVIOUR

Background

Barn Swallows arrive on the breeding grounds in Ontario between mid-April and mid-May of each year. Adults are philopatric and will usually return to the same nesting site as previous years while first year birds typically select a breeding site different than their natal site but within a few kilometres (Turner 2006). The number of birds nesting at a site can vary from a single pair to upwards of 30 pairs (Turner 2006). For this study, each breeding site consisted of a barn and associated buildings, surrounded by open foraging habitat. When selecting a nesting site, Barn Swallows require a horizontal or vertical nesting substrate underneath a horizontal overhang, which is typically a manmade structure, such as a barn, shed, bridge, or culvert, and they require foraging habitat within approximately 1 km of the nest site (Brown and Brown 1999). Barn Swallows are aerial predators and need open foraging habitat, such as hayfields, pasture, and cropland.

Pair formation and nest-building begin within two weeks of arrival on the breeding grounds (Samuel 1971). Most pairs will reuse old nests and renovate them by adding mud to the rim and lining the inside, which can take 2-5 days (Anthony and Ely 1976). A small proportion of pairs build new nests, taking between 3 and 10 days to construct

(Samuel 1971).

Clutch initiation of the first brood of the season usually begins in early May and continues into mid-June (Smith and Montgomerie 1991). Clutch initiation can take place over an extended period because of variation in the timing of arrival of females on the breeding grounds (Turner 2006). The degree of synchrony in egg-laying between females at a breeding site is variable (Turner 2006). Barn Swallows typically lay between 3 and 6

7 eggs per clutch, but clutch size can range from 2-7 eggs (Brown and Brown 1999). In

North America, both males and females incubate the eggs (Smith and Montgomerie

1991). Once the young hatch, both parents feed the nestlings and it takes on average 20-

21 days for the young to fledge (Brown and Brown 1999). Juveniles continue to be fed by their parents for up to a week post-fledging (Brown and Brown 1999). Many Barn

Swallow pairs will have two broods in a breeding season. Second broods tend to be smaller than first broods, with fewer eggs laid (Turner 2006). Second clutches are laid in late June and July and young typically fledge by the end of August (Brown and Brown

1999). In North America, Barn Swallows have never been recorded rearing more than two broods (Brown and Brown 1999). Studies report the percentage of pairs having a second brood between 6% and 92% across the global range of Barn Swallows (Turner

2006). In North America, between 34% and 49% of pairs were reported to double brood

(Turner 2006).

Barn Swallows show considerable variation in many of their reproductive behaviours, including the number of pairs nesting at a site, the number of broods in a year, the timing of reproduction, and the number of eggs laid. These reproductive behaviours may be influenced by physiological and environmental factors, including food availability (Turner 2006). Variation in these reproductive behaviours can lead to changes in seasonal fecundity, which affects population dynamics (Nagy and Holmes 2005).

Food Availability and Reproduction

The breeding season is an energetically demanding time where adults need to invest energy into self-maintenance as well as nest-building, egg-laying, and chick provisioning (Martin 1987). The food resources on the breeding grounds during this time

8 are thus very important. Barn Swallows are income breeders and they undertake a long migration to take advantage of seasonal food availability on the breeding grounds (Turner

2006). Limited food availability could impact various components of the Barn Swallow nesting cycle, each of which contribute to annual fecundity and each of which have been shown to be limited by the food supply (Martin 1987). Persistent food limitation and low fecundity over many breeding seasons could lead to declines in population size (Newton

1980).

As colonial breeders, the number of pairs nesting at a breeding site can affect the availability of food to each pair through competition (Turner 2006). Sites that support larger colonies would need to have more food available for the greater number of pairs to successfully raise young or parents would have to travel farther to obtain the resources they require, costing them more energy. Ideally, Barn Swallows should select sites that maximize their energy intake by balancing interspecific competition with food resource availability such that the distribution of individuals between sites matches the distribution of resources between sites (ideal free distribution theory; Fretwell and Lucas 1969).

When populating breeding sites, it could be more advantageous for a swallow to occupy a lower quality site with fewer competitors. If Barn Swallows follow the ideal free distribution theory, the prey availability per nesting pair at each site should be approximately equal. Limited prey availability at a site would limit the number of pairs that could successfully nest at a site. Møller (1987) found that Barn Swallow colony size in Denmark is positively correlated with insect abundance and the number of sites offering feeding opportunities in bad weather.

Barn Swallows use the concurrent intake of resources to provide for their reproductive attempts, therefore, the number of eggs laid and the number of young fledged could both be affected by food availability on the breeding grounds. In Europe, higher reproductive success and larger clutch sizes were associated with livestock farms, which provided more large prey items (Møller 2001, Ambrosini et al. 2002, Grüebler et al. 2010). In urban areas, higher reproductive success was observed at nest sites at the periphery, where there was more prey available, compared to the centre (Teglhøj 2017).

Møller (2001) also found higher rates of second brooding at sites with higher prey availability. Therefore, at sites with higher prey availability, pairs fledge more young per nest and pairs are more likely to have multiple broods, increasing the annual fecundity of these birds. If the same is true in North America, population declines could be driven by the effect of limited food availability on the fecundity of Barn Swallows (i.e., smaller clutch size, brood size, and/or number of broods). The success of second broods is especially dependent on the food supply, which decreases in late summer and can limit the survival of young from second broods (Grüebler and Naef-Daenzer 2008). Second broods tend to have lower survival during the post-fledging period and are less likely to recruit into the population (Raja-Aho et al. 2017). Prey availability is an important factor for the success of second broods as well as for the rate of second brooding.

For many insectivorous bird species, the timing of breeding has evolved such that the peak energy demands of reproduction occur at the same time as the peak in prey abundance (Both and Visser 2005). Climate change, leading to warmer spring temperatures, can cause a shift in the timing of these ecological events (Both 2010).

Insects and birds do not necessarily respond to the same environmental cues; therefore,

10 birds and insects may shift their phenologies at different rates (Both 2010). This shift can affect the food supply of Barn Swallows by causing a mismatch between the availability of prey and the demand for prey. Clutch initiation in Barn Swallows is limited by the arrival of females on the breeding grounds, but Barn Swallows can adjust the timing of clutch initiation based on several factors, including temperature, weather conditions, livestock, and the abundance of insects (Ambrosini et al. 2002, Turner 2006). Due to their ability to respond to these cues, Barn Swallows may be able to shift the timing of breeding to match the phenology of their insect prey and maximize the prey available during breeding.

SCOPE & OBJECTIVES

My thesis examines the diet and food supply of Barns Swallows and the role of the food supply on their reproductive performance at ten breeding sites in Ontario,

Canada (Figure 1.1). In Chapter 2, my objectives were to (1) determine what prey taxa

Barn Swallow nestlings are consuming using DNA metabarcoding, (2) determine whether Barn Swallows exhibit prey-size selectivity, and (3) determine how nestling

Barn Swallow diet changes in response to fluctuating prey availability. In Chapter 3, my objective was to investigate the effects of prey availability on colony size, reproductive success, double brooding and the timing of reproduction. The results from this work will increase our knowledge of the biology of Barn Swallows breeding in Ontario and it will provide insight into the cause of their population decline.

FIGURES

REFERENCES Ambrosini, R., Bolzern, A. M., Canova, L., Arieni, S., Møller, A. P. and Saino, N. 2002. The distribution and colony size of barn swallows in relation to agricultural land use. - J. Appl. Ecol. 39: 524–534. Ambrosini, R., Rubolini, D., Trovò, P., Liberini, G., Bandini, M., Romano, A., Sicurella, B., Scandolara, C., Romano, M. and Saino, N. 2012. Maintenance of livestock farming may buffer population decline of the Barn Swallow Hirundo rustica. - Bird Conserv. Int. 22: 411–428. Anthony, L. W. and Ely, C. A. 1976. Breeding biology of Barn Swallows in west-central Kansas. - Kansas Ornithol. Soc. Bull. 27: 37–43. Beal, F. E. L. 1918. Food habits of the swallows: a family of valuable native birds. - United States Dep. Agric. Bull. Benton, T. G., Bryant, D. M., Cole, L. and Crick, H. Q. P. 2002. Linking agricultural practice to insect and bird populations: a historical study over three decades. - J. Appl. Ecol. 39: 673–687. Both, C. 2010. Food availability, mistiming, and climactic change. - In: Møller, A. P. et al. (eds), Effects of Climate Change on Birds. Oxford University Press, pp. 129– 147. Both, C. and Visser, M. E. 2005. The effect of climate change on the correlation between avian life-history traits. - Glob. Chang. Biol. 11: 1606–1613. Bowser, A. K., Diamond, A. W. and Addison, J. A. 2013. From puffins to plankton: A DNA-based analysis of a seabird food chain in the Northern Gulf of Maine. - PLoS One 8: 1–16. Brown, C. R. and Brown, M. B. 1999. Barn Swallow Hirundo rustica, the Birds of North America Online. - Birds North Am. Online Cadman, M. D., Sutherland, D. A., Beck, G. G., Lepage, D., Couturier, A. R. 2007. Atlas of the Breeding Birds of Ontario. - Bird Studies Canada, Environment Canada, Ontario Field Ornithologists, Ontario Ministry of Natural Resources, and Ontario Nature. Chown, S. L. and Gaston, K. J. 2010. Body size variation in insects: a macroecological perspective. - Biol. Rev. 85: 139–169. COSEWIC 2011. Assessment and Status Report on the Barn Swallow Hirundo rustica in Canada. - Committee on the Status of Endangered Wildlife in Canada. Deagle, B. E., Thomas, A. C., Shaffer, A. K., Trites, A. W. and Jarman, S. N. 2013. Quantifying sequence proportions in a DNA-based diet study using Ion Torrent amplicon sequencing: which counts count? - Mol. Ecol. Resour. 13: 620–33.

Dickman, C. R. and Huang, C. 1988. The reliability of fecal analysis as a method for determining the diet of insectivorous mammals. - J. Mammal. 69: 108–113. Dirzo, R., Young, H. S., Mauro, G., Ceballos, G., Isaac, N. J. B. and Collen, B. 2014. Defaunation in the Anthropocene. - Science 345: 401-406. Dor, R., Safran, R. J., Sheldon, F. H., Winkler, D. W. and Lovette, I. J. 2010. Phylogeny of the genus Hirundo and the Barn Swallow subspecies complex. - Mol. Phylogenet. Evol. 56: 409-418. Dunn, P. O., Winkler, D. W., Whittingham, L. A., Hannon, S. J. and Robertson, R. J. 2011. A test of the mismatch hypothesis: How is timing of reproduction related to food abundance in an aerial insectivore? - Ecology 92: 450–461. Environment Canada 2014. North American Breeding Bird Survey - Canadian Trends Web Site, Data-Version 2012. Fretwell, D. S. and Lucas, H. L. 1969. On territorial behaviour and other factors influencing habitat distribution in birds. - Acta Biotheor. 19: 16–32. Gerwing, T. G., Kim, J.-H., Hamilton, D. J., Barbeau, M. A. and Addison, J. A. 2016. Diet reconstruction using next-generation sequencing increases the known ecosystem usage by a shorebird. - Auk 133: 168–177. Gregory, R. D., Noble, D. G. and Custance, J. 2004. The state of play of farmland birds: population trends and conservation status of lowland farmland birds in the United Kingdom. - Ibis. 146: 1-13. Grüebler, M. U. and Naef-Daenzer, B. 2008. Fitness consequences of pre- and post- fledging timing decisions in a double-brooded passerine. - Ecology 89: 2736– 2745. Grüebler, M. U., Korner-Nievergelt, F. and von Hirschheydt, J. 2010. The reproductive benefits of livestock farming in barn swallows Hirundo rustica: quality of nest site or foraging habitat? - J. Appl. Ecol. 47: 1340–1347. Hallmann, C., Sorg, M., Jongejans, E., Siepel, H., Hofland, N., Schwan, H., Stenmans, W., Müller, A., Sumser, H., Hörren, T., Goulsons, D. and de Kroon, H. 2017. More than 75 percent decline over 27 years in total flying insect biomass in protected areas. – PLoS One 12: e0185809. Hajibabaei, M., Smith, M. A., Janzen, D. H., Rodriguez, J. J., Whitfield, J. B. and Hebert, P. D. N. 2006. A minimalist barcode can identify a specimen whose DNA is degraded. - Mol. Ecol. Notes 6: 959–964. Heagy, A. and McCraken, J. 2004. Monitoring the State of Ontario’s Migratory Landbirds. - Bird Studies Canada.

Heagy, A., Badzinski, D., Bradley, D., Falconer, M., McCraken, J., Reid, R. A. and Richardson, K. 2014. Recovery strategy for the Barn Swallow (Hirundo rustica) in Ontario. Ontario Recovery Strategy Series. - Ontario Ministry of Natural Resources and Forestry. Hebblethwaite, M. L. 1989. Foraging behavior of the Barn Swallow (Hirundo rustica) in the Adirondacks. PhD Thesis. Hebert, P. D. N., Cywinska, A., Ball, S. L. and DeWaard, J. R. 2003. Biological identifications through DNA barcodes. - Proc. R. Soc. Biol. Sci. 270: 313–321. Jedlicka, J. A., Vo, A.-T. E. and Almeida, R. P. P. 2017. Molecular scatology and high- throughput sequencing reveal predominately herbivorous insects in the diets of adult and nestling Western Bluebirds (Sialia mexicana) in California vineyards. - Auk 134: 116–127. Johnson, E. J., Best, L. B. and Heagy, P. A. 1980. Food sampling biases associated with the “ligature method” - Condor 82: 186–192. Kunz, T. H. and Whitaker Jr., J. O. 1983. An evaluation of fecal analysis for determining food habits of insectivorous bats. - Can. J. Zool. 61: 1317–1321. Lee, S.-D., Ellwood, E. R., Park, S. and Primack, R. B. 2011. Late-arriving barn swallows linked to population declines. - Biol. Conserv. 144: 2182–2187. Major, R. E. 1990. Stomach flushing of an insectivorous bird: an assessment of differential digestibility of prey and the risk to birds. - Aust. Wildl. Res. 17: 647– 657. Martin, T. E. 1987. Food as a limit on breeding birds: a life-history perspective. - Annu. Rev. Ecol. Syst. 18: 453–487. Møller, A. P. 1987. Advantages and disadvantages of coloniality in the swallow, Hirundo rustica. - Anim. Behav. 35: 819–832. Møller, A. P. 2001. The effect of dairy farming on barn swallow Hirundo rustica abundance, distribution and reproduction. - J. Appl. Ecol. 38: 378–389. NABCI Canada. 2012. The State of Canada’s Birds. - Environment Canada. Nagy, L. R. and Holmes, R. T. 2005. Food limits annual fecundity of a migratory songbird: an experimental study - Ecology 86: 675–681. Nebel, S., Mills, A., McCracken, J. D. and Taylor, P. D. 2010. Declines of aerial insectivores in North America follow a geographic gradient. - Avian Conserv. Ecol. 5: 1. Newton, I. 1980. The role of food in limiting bird numbers. - Ardea 68: 11–30.

Nocera, J. J., Blais, J. M., Beresford, D. V., Finity, L. K., Grooms, C., Kimpe, L. E., Kyser, K., Michelutti, N., Reudink, M. W. and Smol, J. P. 2012. Historical pesticide applications coincided with an altered diet of aerially foraging insectivorous chimney siwfts. - Proc. R. Soc. Biol. Sci. 279: 3114-3120. Oliver, D. R. 1971. Life story of the Chironomidae. - Annu. Rev. Entomol. 16: 211–230. Orłowski, G. and Karg, J. 2011. Diet of nestling Barn Swallows Hirundo rustica in rural areas of Poland. - Cent. Eur. J. Biol. 6: 1023–1035. Pompanon, F., Deagle, B. E., Symondson, W. O. C., Brown, D. S., Jarman, S. N. and Taberlet, P. 2012. Who is eating what: diet assessment using next generation sequencing. - Mol. Ecol. 21: 1931–50. Potts, S. G., Biesmeijer, J. C., Kremen, C., Neumann, P., Schweiger, O. and Kunin, W. E. 2010. Global pollinator declines: trends, impacts and drivers. – Trends Ecol. Evolut. 25: 345-353. Raja-Aho, S., Eeva, T., Suorsa, P., Valkama, J. and Lehikoinen, E. 2017. Juvenile Barn Swallows Hirundo rustica from late broods start autumn migration younger, fuel less effectively and show lower return rates than juveniles from early broods. - Ibis in press. Ralph, C. P., Nagata, S. E. and Ralph, C. J. 1985. Analysis of droppings to describe diets of small birds. - J. F. Ornithol. 56: 165–174. Ruffino, L., Salo, P., Koivisto, E., Banks, P. B. and Korpimäki, E. 2014. Reproductive responses of birds to experimental food supplementation: a meta-analysis. - Front. Zool. 11: 1-13. Sæther, B.-E. and Bakke, Ø. 2000. Avian life history variation and contribution of demographic traits to the population growth rate. - Ecology 81: 642–653. Samuel, D. E. 1971. The breeding biology of Barn and Cliff Swallows in West Virginia. - Wilson Bull. 83: 284–301. Sanderson, F. J., Donald, P. F., Pain, D. J., Burfield, I. J. and van Bommel F. P. J. 2006. Long-term population declines in Afro-Palearctic migrant birds. - Biol. Cons. 131: 93-105. Sheldon, F. H., Whittingham, L. A., Moyle, R. G., Slikas, B. and Winkler, D. W. 2005. Phylogeny of swallows (Aves: Hirundinidae) estimated from nuclear and mitochondrial DNA sequences. - Mol. Phylogenet. Evol. 35: 254-270. Smith, H. G. and Montgomerie, R. 1991. Sexual selection and the tail ornaments of North American Barn Swallows. - Behav. Ecol. Sociobiol. 28: 195–201.

Teglhøj, P. G. 2017. A comparative study of insect abundance and reproductive success of barn swallows (Hirundo rustica) in two urban habitats. - J. Avian Biol. 48: 846–853. Trevelline, B. K., Latta, S. C., Marshall, L. C., Nuttle, T. and Porter, B. A. 2016. Molecular analysis of nestling diet in a long-distance Neotropical migrant, the Louisiana Waterthrush (Parkesia motacilla). - Auk 133: 415–428. Turner, A. K. 1982a. Optimal foraging by the Swallow (Hirundo rustica, L): Prey size selection. - Anim. Behav. 30: 862–872. Turner, A. K. 1982b. Timing of laying by Swallows (Hirundo rustica) and Sand Martins (Riparia riparia). - J. Anim. Ecol. 51: 29-46. Turner, A. K. 2004. Swallows and Martins (Hirundinidae). - In: del Hoyo, J. et al. (eds), Handbook of the Birds of the World. Lynx Edicions. Turner, A. 2006. The Barn Swallow. - T & A D Poyser. Waugh, D. R. 1978. Predation strategies in aerial feeding birds. Zeale, M. R. K., Butlin, R. K., Barker, G., Lees, D. C. and Jones, G. 2010. Taxon-specific PCR for DNA barcoding arthropod prey in bat faeces. - Mol. Ecol. Resour. 11: 236-244. Zink, R. M., Rohwer, S., Andreev, A. V., Dittmann, D. L. 1995. Trans-Beringia comparisons of mitochondrial DNA differentiation in birds. - Condor. 97: 639- 649. Zink, R. M., Pavlova, A., Rohwer, S. and Drovetski, S. V 2006. Barn Swallows before barns: population histories and intercontinental colonization. - Proc. R. Soc. B Biol. Sci. 273: 1245–1251.

Chapter 2: DNA metabarcoding reveals the broad and flexible diet of a declining aerial insectivore

ABSTRACT

Investigating predator-prey interactions in the wild can be difficult when mobile predators consume diverse and abundant prey. In this study, we used an integrated DNA barcoding method to monitor prey availability in Barn Swallow (Hirundo rustica) habitat and identify the prey items nestlings consumed. As with most aerial insectivore species,

Barn Swallows have experienced dramatic population declines over the past 30 years in

North America. It has been suggested that this decline is related to declines in insect prey populations. We collected insects using Malaise traps at ten breeding sites and identified specimens using standard DNA barcoding to create a custom reference database of prey availability for our study area. We identified dietary items from nestling fecal samples collected below nests using high-throughput DNA sequencing and compared the sequences against our custom reference database. We found that Barn Swallows fed nestlings a range of prey items but showed selection for larger prey species. Barn

Swallow nestling diet was variable between breeding sites, between breeding seasons, and within the breeding season. This flexible diet should make Barn Swallows resilient to changing prey availability on the breeding grounds. Our study demonstrates the utility of custom reference databases when employing molecular methods for dietary analysis.

INTRODUCTION

Predator-prey interactions can be challenging to observe in nature, especially when direct observation of predation is not possible due to the small size of prey or the

18 location of feeding [1]. The challenges of studying these interactions are even greater for predators that feed on a broad range of prey items and when prey items vary substantially over time and space [2]. Arthropods, particularly insects, dominate terrestrial ecosystems and they represent an abundant food source for predators [3,4]. Observing insectivorous predator-prey interactions can be challenging because direct observations of feeding can be difficult due to the small size of insects. In addition, most insects fly during some life stage so predation can occur by aerial insectivores on the wing [5]. Insectivores are typically generalist consumers feeding on numerous prey species, although they can show some diet specialization (e.g., hard vs. soft bodied prey [6]). Insects are a diverse group that occupy various habitats and functional niches and they are highly variable in their distribution across time and space [4]. In the wild, it can be difficult to investigate the diet of insectivores while also continually monitoring and identifying this diverse, variable prey source. Understanding how predators respond to changing prey availability in changing environments is important to our understanding of predator-prey interactions

[7].

Molecular methods are becoming more commonly used for diet analysis, most prominently including DNA barcoding. DNA barcoding of animals uses the sequence from a standardized region of the mitochondrial genome, cytochrome c oxidase I (COI), to taxonomically identify unknown specimens using a reference database of sequences from known specimens [8]. DNA barcoding can be used to identify large numbers of diverse specimens at high taxonomic levels and requires less time and expertise than morphological identification [8]. Thus, DNA barcoding can provide an efficient way to monitor prey availability. DNA metabarcoding uses the same principles as DNA

19 barcoding but makes use of high-throughput sequencing to sequence DNA from multiple specimens in a single sample [9]. This method can be used to sequence DNA from different prey items in a fecal sample. Sequences are identified by matching unknown prey sequences from the fecal samples to a reference database of known sequences. By applying the same technique to identify prey items in the diet and in the habitat, the DNA sequences generated from prey in the habitat can be used as a custom reference database to link prey in the diet to prey availability. DNA metabarcoding has been successfully used to study the diet of a variety of taxa, including insects, ungulates, bats, and birds

[10–13]. For generalists that consume a variety of different prey species this method can reveal a greater diversity of prey items than alternative methods of diet analysis [14,15].

Furthermore, DNA metabarcoding has higher sensitivity and taxonomic resolution compared to other methods of diet analysis [16]. This method cannot be used to reliably quantify prey items within a sample [17] but frequency of occurrence across samples can be used as a semi-quantitative measure to compare how often predators are consuming different prey items [15].

Aerial-foraging insectivorous birds have suffered severe population declines in

North America over the past 30 years [18]. This population decline has been estimated at

77% in Canada since 1970 [19] and it is the steepest decline seen in any ecological guild of birds in North America [20]. Many possible causes have been suggested for the decline of breeding populations of aerial insectivores including habitat loss and degradation, decrease in food availability, climate change, and pesticide use [20–22]. The aerial insectivores comprise a taxonomically diverse guild of birds with varied life histories, diverse habitat requirements and different wintering grounds, suggesting that

20 their common food source may be a key factor causing their decline [18,23]. Recent studies support the food supply as a factor limiting the population of some aerial insectivores, including Chimney Swifts (Chaetura pelagica) and Eastern Whip-Poor-

Wills (Antrostomus vociferous) [24,25]. A decrease in food availability could be caused by phenological shifts leading to a mismatch between insect prey abundance and food demands and/or large-scale declines in insect prey abundance [18]. To understand how prey availability could be affecting aerial insectivores, we need to know what prey these birds are consuming and we need a reliable method of comparing insect prey in the diet to insect availability in the habitat. Here, we use DNA barcoding to investigate the diet and food supply of Barn Swallows (Hirundo rustica), a widespread species of aerial insectivore. While Barn Swallows have a worldwide distribution, population declines have been observed in many parts of their range, including North America [18,26–28].

Due to the proximity of Barn Swallow nesting and foraging habitat to humans (i.e., barns for nesting and agricultural fields for foraging), Barn Swallows are exposed to many anthropogenic effects on the environment, such as pesticide applications [29].

Little is known of Barn Swallow diet in North America, despite the importance of prey in understanding the cause of recent population declines in this species. In Europe,

Barn Swallows have been reported feeding on a variety of insect orders, with Coleoptera and Diptera identified as the most abundant prey items [30–32]. Most of these studies used morphological identification of prey remains in feces and did not identify prey at taxonomic levels higher than order. Despite this broad level of classification, the studies to date support that Barn Swallows are generalist foragers. However, insect prey varies in size and nutritional quality and insectivores select their prey accordingly [33,34].

Previous dietary analysis shows Barn Swallows selecting larger prey items from various arthropod orders, which is predicted by optimal foraging theory where larger prey items are more profitable for Barn Swallows despite taking more effort to catch [30–32].

Our objective was to use an integrated DNA barcoding method to determine what insect prey are available to Barn Swallows during the breeding season, and to determine what prey items Barn Swallow nestlings are consuming. The prey fed to nestlings is a critical resource since the nestling period is the most energetically demanding time during the breeding season [35]. Based on previous work on the diet of this species in Europe

[30-34], we hypothesized that Barn Swallows are broad-scale generalists and select prey based on size. We expected to identify a wide range of insect taxa in nestling fecal samples but to identify large prey items more frequently than their availability in the habitat. We expected to identify different prey items in the diet of nestlings as prey availability in the habitat changed seasonally and when prey availability differed between nesting sites.

METHODS

Study Site and Sample Collection

We studied Barn Swallow diet at 10 breeding sites in Peterborough County and the City of Kawartha Lakes in Ontario, Canada (Figure 1.1). Each breeding site had between one and 25 nesting pairs (mean: 8.1 ± 1.48 SE) and consisted of a barn and associated buildings surrounded by a mixture of cropland, pasture, and/or hayfields. At each site, nests were monitored every 2 to 3 days and active nests were identified. Barn

Swallow nestlings begin defecating off the side of the nest as early as day 5 after hatching

[36] and all nestlings defecate off the side of the nest by day 12 after hatching [29]. We

22 placed clean, plastic squares (approximately 30 cm x 30 cm) below all active nests. We collected and pooled 3 fresh nestling fecal samples from each nest every 2 days from the plastic squares using flame sterilized forceps, starting at day 8 after hatching and ending at fledging. After sampling, we placed a new plastic square below each nest. Fecal samples were stored in 95% ethanol at -20°C until processing.

We used land and air Malaise traps with bottom collectors (BioQuip #2869 &

#2892) to monitor insect abundance and diversity at each breeding site. At each site, we set up a trap 1 m above the ground and 15-30 m from the structure used for nesting. Barn

Swallows usually forage <10 m in height and often <1 m from the ground [29] and spend most of their time foraging within a few hundred metres of the nest site [37]. Malaise traps paired with bottom collectors are effective at collecting flying insects from a range of taxa [38,39] and should represent the diversity of prey available to Barn Swallows.

Collection bottles were placed on traps at dawn (sunrise +/- 30 minutes) and removed at dusk (sunset +/- 30 minutes). In 2015, traps were open for 2 days/week at 9 sites and 6 days/week at one site. We began collecting insects on 20 May 2015 at 2 sites and began collecting insects on 1 June 2015 at the remaining 8 sites. Collection continued at all sites until 29 August 2015. In 2016, traps were open for 2 days a week at all sites and collection began on 25 April and continued through to 29 August. Specimens were stored in 95% ethanol at -20°C until processing.

DNA Extraction, Amplification, and Sequencing

Fecal samples were processed at the Canadian Centre for DNA Barcoding

(CCDB) at the University of Guelph, Guelph, Ontario, Canada. The samples were homogenized by vortexing in a 15 mL tube containing 2 ceramic beads, followed by

23 subsampling and extraction using one of two protocols. Both protocols used a manual plate-based method of DNA extraction modified from [40]. In the first protocol, the homogenate was dried to remove excess liquid and a subsample of ~100 µL was transferred to 1.5 mL tubes. To lyse samples, 350 µL of an insect lysis buffer, ProK, and

2% PVP mixture was added and samples were incubated at 56°C overnight. The whole lysate was mixed with 700 µL of binding mix (3 M guanidine thiocyanate, 10 mM EDTA pH 8.0, 5 nM Tris-HCl pH 6.4, 2% Triton X-100, 50% ethanol) and 850 µL of the mixture was transferred to a glass fibre plate. The plate was washed with 700 µL of protein wash buffer followed by two 700 µL washes with wash buffer and then eluted in

40 µL of elution buffer. The second protocol was an updated extraction protocol by the

CCDB to maximize DNA recovery in which, the entire homogenate was lysed using 5 mL of insect lysis buffer, ProK and 2% PVP mixture and incubated at 56°C overnight. A

50 µL subsample of lysate was mixed with 100 µL of binding mix and transferred to a glass fibre plate. The plate was washed with 180 µL of protein wash buffer followed by

750 µL of wash buffer and then eluted in 40 µL of elution buffer.

A 157 bp target region of the COI gene was amplified using primers described in

[16]. For each 96-well plate, the samples were uniquely tagged using a combination of 12 forward multiplex identifier (MID) tags and 8 reverse MID tags [41]. The PCR mixture consisted of 6.25 µL of 10% trehalose, 2 µL of dH2O, 1.25 µL of 10X buffer, 0.625 µL

50 nM MgCl2, 0.125 µL of 10µM of each the forward and reverse primer, 0.0625 µL of

10 µM dNTP, 0.06 µL of Platinum Taq (5U/µL) and 2 µL of DNA. PCR conditions were as follows: 2 min at 94°C, followed by 60 cycles of 30s at 94°C, 30s at 53°C and 30s at

72°C, followed by a final extension of 5 min at 72°C and then held at 10°C. Amplicons

24 were visualized using 4 µL on an E-Gel (Invitrogen G700802; Thermo Fisher Scientific,

Waltham, Massachusetts, USA). DNA was successfully extracted and amplified from

271 fecal samples out of a total 281 samples.

PCR products for each plate were pooled and amplicons were purified using a magnetic bead protocol outlined in [42] using double-size selection to purify for the target amplicon length (~284 bp). The cleaned product was quantified using a Qubit 2.0 fluorometer (Invitrogen Q32866; Thermo Fisher Scientific) and adjusted to 1 ng µL-1.

The sequencing library was prepared by templating and enriching with the Ion OneTouch

2 System (Ion 4474779; Thermo Fisher Scientific). The library was sequenced using a

316 v.2 chip on an Ion Torrent PGM (Ion 4462921; Thermo Fisher Scientific) following the manufacturer’s instructions.

Custom Reference Library

To create a custom reference database of insect prey available to Barn Swallows, we sorted insect specimens collected in Malaise traps to morphospecies [43]. We measured the body length to the nearest 0.1 mm (excluding appendages such as antennae or cerci) of up to five specimens of each morphospecies if available. A representative of each morphospecies was identified using DNA barcoding. Specimens were processed at the CCDB according to their standardized extraction, amplification, and uni-directional sequencing protocols (see [40,44]). The resulting sequences were used to identify the specimens via the BOLD Identification System (http://www.boldsystems.org/; [45]) and were subsequently used as reference sequences to identify unknown specimens from the fecal samples. Matching individual specimens in the diet to specimens in the habitat would typically require species-level resolution. Developing a custom reference database

25 allowed us to match prey items consumed by Barn Swallows to specimens collected in traps and their associated abundance and size using sequence similarities, even when sequence identification at high taxonomic resolution was not possible. For any insect specimens that did not sequence successfully or did not match a reference sequence on

BOLD, we used photos to identify the specimens to order and, where possible, to family using [46].

Taxonomic Identification

For sequencing each run, the reads were separated into twelve subsets by forward

MID tag using the Ion Torrent Server and these subsets were then separated into 96 samples by reverse MID tag using Galaxy (https://usegalaxy.org/). The primer and adapter sequences were trimmed using Cutadapt [47]. Reads were filtered based on quality (minimum QV of 20) and length (minimum length of 100 bp) using Sickle

(github.com/ucdavis-bioinformatics/sickle). The filtered reads were dereplicated to remove duplicated reads using FASTX Collapser

(http://hannonlab.cshl.edu/fastx_toolkit/index.html) and clustered into operational taxonomic units (OTUs) at 97% using UPARSE [48]. OTUs with infrequent haplotypes

(<10 reads) were excluded from downstream analysis. Taxonomic identification was assigned to each OTU using a BLAST search against our custom database. Species level identifications were assigned at a minimum of 99.3% sequence similarity and genus level identifications were assigned at a minimum of 95% based on [16]. Any sequences that did not match a reference sequence in the custom database were identified by performing a BLAST search against all Canadian animal sequences in the BOLD database

(downloaded December 2016). Spurious identifications, including sequences identified as non-arthropod (e.g., algae, nematodes), were omitted from subsequent analysis.

Data Analysis

Data analysis was conducted using R Studio v 1.0.136. Amplicon read counts do not reliably reflect abundances of prey items in fecal samples [17], therefore, arthropod sequences were considered as present or absent within each fecal sample. We calculated frequency of occurrence of families and of size classes in the diet based on presence/absence across our samples (e.g., if a taxon was present in 10 of 100 samples, the frequency would be 0.10). To calculate the frequency of occurrence of different prey sizes, we used only the prey items identified using our custom database because they were associated with voucher specimens whose length was measured. An accumulation curve of arthropod families in the diet was created using the vegan package [49]. We used the package mvabund [50] to model the change in the family-level prey community in the diet over time and by site using multivariate GLMs. Site, year, and brood were included as categorical predictor variables and significant effects were determined using the anova.manyglm function with α = 0.05. We ran the same models for the family-level community of prey available in the habitat. Families whose frequency of occurrence in the diet changed significantly with site, year, or brood were identified using univariate tests in the anova.manyglm function. For these families, we compared the change in frequency of occurrence in the diet to the change in availability in the habitat between sites, years, and broods.

RESULTS

Custom Reference Library We collected over 60,000 insect specimens over two breeding seasons. Of the specimens submitted for sequencing, we recovered sequences from 87.4% of specimens.

These sequences were used as our custom reference library, where 99.1% of sequences were identified to the level of order, 97.5% to the level of family, 71.7% to the level of genus, and 42.0% to species level. We identified over a thousand species from 262 different families. All sequences are publicly available on BOLD under the dataset DS-

AABARS.

Prey Identity

We recovered 12,479,453 reads representing 472 unique taxa from 271 fecal samples (see Figure A.1 and Table A.1 in Appendix A). Of the OTUs that successfully matched a reference sequence, 100% identified to the order level, 98.5% identified to the family level, 82.8% identified to the genus level and 14.5% identified to the species level.

Family level identifications represented the highest level of taxonomic resolution where a high proportion of OTUs were identified so we performed subsequent analysis at the family level. We identified 130 arthropod families in the diet from 13 different orders.

The accumulation curve of taxonomic richness at the family-level in the diet suggests that the analysis of more fecal samples would result in the identification of more prey taxa

(Figure 2.1), indicating that we are not reporting an exhaustive list of families or arthropods consumed by Barn Swallows. The most frequently consumed order was

Diptera, the most frequently consumed family was Tachinidae (Diptera), and the most

28 frequently consumed genus and species was Pollenia pediculata (Calliphoridae: Diptera), a pollinator species of cluster fly that is parasitic on earthworms [51]. Many human and livestock pests were consumed frequently as prey such as mosquitoes, horseflies, and stable flies. Agricultural pests were also consumed but at a lower frequency. Barn

Swallow nestlings consumed specialized nest parasites of the nestlings themselves,

Protocalliphora sp. Out of the 472 unique prey items we identified, 35 were identified as pests (see Table A.2 in Appendix A).

Prey Selection

The most frequently consumed families did not represent a large proportion of the prey available, while the insect families that were most available in the habitat were consumed but not at high frequency (Figure 2.2). Small prey items (<3 mm in length) represented most of the prey available in the habitat but were consumed at low frequency while large prey items (>3 mm in length) were consumed very frequently but represented a very small proportion of the prey items available in the habitat (Figure 2.3). Prey items detected in the diet ranged from 1.1 - 20.3 mm in length while prey items available in the habitat ranged from 0.1 – 38.9 mm in length.

Dietary Flexibility

Site, brood, and year and their interactions all affected both Barn Swallow diet composition and prey availability in the habitat (Table 2.1). For both the diet and habitat models, site was the predictor that explained most of the deviance. Thus, Barn Swallows consumed different prey items over time and at different sites as prey availability also changed temporally and between sites (see Table A.3 in Appendix A). The following eight families in the order Diptera were consistently consumed at high frequency across

29 sites, broods and years: Tachinidae, Tabanidae, Anthomyiidae, Tipulidae, Limoniidae,

Calliphoridae, Culicidae, and Muscidae. The differences in diet between sites and times were mostly driven by changes in less frequently consumed families. The change in diet we observed between sites and years reflected changes in prey availability between sites and years. Most prey families were consumed more frequently when they were more abundant in the habitat. The change in diet between broods did not mirror changes in prey availability in the habitat. Most insect families that differed in frequency of occurrence in the diet between broods were more frequently consumed in the second brood, despite greater abundance in the habitat during the first brood. The preferred prey families, listed above, were more abundant in the habitat in the first brood as well. These families decreased in abundance during the second brood when Barn Swallows consumed alternative prey at a higher frequency.

DISCUSSION

Barn Swallow nestlings consumed a broad range of prey items whose availability in the environment around their nesting locations changed over time. DNA metabarcoding allowed us to identify prey items from a wide range of taxonomic groups and at higher resolution than previous studies of Barn Swallow diet. Additionally, we successfully monitored the variation in diet concurrently with the variation in prey availability of a diverse and abundant food source using DNA barcoding. We identified most prey items to the genus level and a notable number to the species level. These levels of identification often cannot be reached using other methods of diet analysis. Due to this high taxonomic resolution and the sensitivity of this method, we detected many prey items that were previously unreported in Barn Swallow diet. This broad range of prey

30 items includes species from a range of foraging guilds and trophic levels, from terrestrial and aquatic environments, and includes multiple pest species. Barn Swallows nesting in barns can provide an ecosystem service by controlling human, livestock, and agricultural pests. Interestingly, Barn Swallow nestlings consumed blowfly nest parasites although we could not determine whether the nestlings were consuming the parasitic larvae or the adult flies. Blowfly parasites have been detected in the diet of other species of insectivorous birds using molecular methods and similarly life stage could not be determined [52].

Within the context of a generalist forager, we also observed evidence for prey selection. Swallows were consuming larger prey items from certain families more frequently relative to their abundance in the habitat. Larger prey items are more profitable prey for Barn Swallows to capture [30], yet we also observed Barn Swallows eating small prey items. Turner [30] showed that Barn Swallows in Scotland eat more small prey items than would be predicted by optimal foraging theory (OFT). OFT may not apply well to Barn Swallows since they feed on a diverse array of mobile prey; OFT may apply better to systems where there is a low number of available prey types with uniform nutrient contents [53] and OFT may not apply well to foragers that feed on mobile prey

[54]. We detected prey selection due to the taxonomic resolution we achieved in identifying prey items and our ability to match prey items in the diet with specimens from the habitat. Many diet studies have been limited to the taxonomic level of order. Those studies would not have been able to provide evidence for selection at this level since, in our study, Diptera was the most abundant order in the habitat and in the diet. It was at the family level that we saw differences in which families were most abundant in the diet and

31 in the habitat. Using an integrated DNA barcoding approach allowed us to match prey items in the diet to specimens in the habitat and their associated size, even without high taxonomic resolution. This allowed us to detect prey size selection, which would be difficult to detect using DNA metabarcoding without a custom reference database unless genus- or species-level taxonomic resolution was achieved for most prey items.

In addition to showing a diverse diet and prey selection, Barn Swallow diet varied significantly over time and between sites. This variation in diet gives Barn Swallows the ability to respond to changing prey availability and to take advantage of fluctuating and unpredictable prey resources. The variation in diet between sites and years appears to be opportunistic, where Barn Swallows are consuming more of certain prey items when they are more available in the habitat. For the most part, the variation in diet matched the yearly and site-specific variation in prey availability. By contrast, the difference in diet between broods was not driven by prey availability. Instead, it could be due to changes in the availability of preferred prey items. In the second brood, there were fewer of the preferred families available, so Barn Swallows may have needed to include more alternative prey as their main prey source declined. While the variation in diet may be caused by different factors, Barn Swallows appear to be able to adjust their diet in response to changing prey availability and take advantage of whatever resources are available. Seasonal changes in diet are observed in other insectivore species which also show a temporal change in their selectivity [55]. While swallows are flexible in their diet, prey items vary in quality in terms of the energy and other nutrients (e.g., fatty acids [56]) they provide. Barn Swallows can still feed on lower quality prey items but this could lead

32 to negative consequences for their health and survival which could carry over into the post-breeding period.

Dietary generalists that can take advantage of changing prey availability should be more resilient to a fluctuating food supply compared to dietary specialists, which are more susceptible [57]. This is because generalists can take advantage of alternate prey sources when their preferred prey is not available [58]. We have demonstrated that Barn

Swallows are generalists with a flexible diet which suggests that they should be resilient to a changing food supply. For generalist insectivores, overall prey abundance is thought to be a more important driver of productivity than taxa-specific abundance [2]. If the decline in the Barn Swallow population is related to the food supply, it would, most likely, be due to a broad-scale decline in insect prey abundance across taxa, rather than changes in the availability of specific prey groups, due to phenological shifts for example. It is likely that the declines in this guild are caused by multiple interacting factors [59].

While we were successful in using DNA barcoding for this study, there are limitations in the use of this method to identify unknown specimens and prey items from fecal samples. DNA barcoding is very dependent on the quality of the reference database

[60]. A comprehensive reference database is required for unknown sequences to match a sequence in the database and even if there is a match, the level of identification of unknown specimens can be limited by the level of taxonomic identification of the reference sequence. While we used our own custom database generated by DNA barcoding insect specimens, our reference library was limited by the success of DNA sequencing and identification of these specimens. Not all insect specimens (~13%) that

33 we collected in the habitat were identified via DNA barcoding and not all those that were identified via barcoding were identified to a high taxonomic level. By integrating traditional and molecular taxonomic knowledge to expand reference databases, identification success and the utility of reference databases could be improved. Another limitation is that some taxa can be difficult to distinguish from one another due to low variation in the COI gene between these taxa [61]. This problem can be further complicated when targeting a shorter segment rather than the standard DNA barcode, which is necessary to recover DNA in degraded samples such as feces [62]. The short target sequence we used was shown to be reliable in distinguishing species in the prey taxa we were targeting [16].

DNA metabarcoding cannot be used as a quantitative method due to a variety of biological and technical factors [9]. We used frequency of occurrence as a semi- quantitative measure which applies well to the large sample sizes that we processed.

Molecular diet analysis could be improved if it were possible to reliably quantify prey items within samples. Current work is attempting to minimize the quantitative biases of this method [63]. The detection of secondary consumption, where the predator of interest consumes another predator that contained prey in its guts, is a possibility when using

DNA metabarcoding for diet analysis [9,14]. Some of the taxa we detected may result from secondary consumption but the most frequently consumed prey items we detected are not predatory as adults.

Most, but not all sequences from the fecal samples were identified using our custom database, indicating that some prey items were not collected in our Malaise traps, or were collected but were not successfully sequenced. We could identify these

34 sequences using a broader database, but it suggests that not all potential prey items were represented in our Malaise trap catch. The landscape surrounding the breeding sites was heterogeneous and prey resources may have displayed a patchy distribution with local abundances in microhabitats (e.g., hedgerows [64]) that were not reflected in our Malaise trap catch. Additionally, swallows typically forage within a few hundred metres of the nest site and a single Malaise trap could not capture the diversity of prey in this whole area. Operating multiple Malaise traps around the breeding site could improve measures of prey availability, but logistically this was not possible for this study. Malaise traps can underrepresent certain taxa, such as microhymenopterans, but there is no evidence of bias against the preferred prey items Barn Swallows consumed frequently [65].

We made use of an integrated DNA barcoding method to determine the diet of

Barn Swallows and simultaneously monitor prey abundance. We found that Barn

Swallows were generalist consumers but showed selection for larger prey items that were mostly Dipterans. Barn Swallows changed their diet in response to changing prey conditions, suggesting that they can obtain food when overall insect abundance is sufficiently high, even if their preferred prey is not available in high quantities.

However, the alternative food consumed by Barn Swallows may not be of equivalent quality which could have negative effects during or after the breeding season. This information expands our knowledge of Barn Swallow biology in North America and can be used to determine how changes in prey availability are affecting Barn Swallow populations. This integrated DNA barcoding method using a custom reference database of the prey available in the local area can be applied to other systems with hard to observe predator-prey interactions and hard to monitor prey populations.

REFERENCES

1. Fisher DO, Dickman CR. 1993 Diets of insectivorous marsupials in arid Australia: selection for prey type, size or hardness? J. Arid Environ. 25, 397–410. (doi:10.1006/jare.1993.1072) 2. Durst SL, Theimer TC, Paxton EH, Sogge MK. 2008 Age, habitat, and yearly variation in the diet of a generalist insectivore, the Southwestern Willow Flycatcher. Condor 110, 514–525. (doi:10.1525/cond.2008.8493) 3. Bell GP. 1980 Habitat use and response to patches of prey by desert insectivorous bats. Can. J. Zool. 58, 1876–1883. (doi:10.1139/z80-256) 4. Kremen C, Colwell RK, Erwin TL, Murphy DD, Noss RF, Sanjayan MA. 1993 Arthropod assemblages: their use in conservation planning. Conserv. Biol. 7, 796–808. 5. Sane SP. 2003 The aerodynamics of insect flight. J. Exp. Biol. 206, 4191– 4208. (doi:10.1242/jeb.00663) 6. Freeman PW. 1979 Specialized insectivory: beetle-eating and moth-eating molossid bats. J. Mammal. 60, 467–479. (doi:10.2307/1380088) 7. Wilmers CC, Post E, Hastings A. 2007 The anatomy of predator-prey dynamics in a changing climate. J. Anim. Ecol. 76, 1037–1044. (doi:10.1111/j.1365-2656.2007.01289.x) 8. Hebert PDN, Cywinska A, Ball SL, DeWaard JR. 2003 Biological identifications through DNA barcodes. Proc. R. Soc. Biol. Sci. 270, 313– 321. 9. Pompanon F, Deagle BE, Symondson WOC, Brown DS, Jarman SN, Taberlet P. 2012 Who is eating what: diet assessment using next generation sequencing. Mol. Ecol. 21, 1931–50. (doi:10.1111/j.1365- 294X.2011.05403.x) 10. Ibanez S et al. 2013 Plant functional traits reveal the relative contribution of habitat and food preferences to the diet of grasshoppers. Oecologia 173, 1459–70. (doi:10.1007/s00442-013-2738-0) 11. Erickson DL, Reed E, Ramachandran P, Bourg NA, McShea WJ, Ottesen A. 2017 Reconstructing a herbivore’s diet using a novel rbcL DNA mini- barcode for plants. AoB Plants 9, plx015. (doi:10.1093/aobpla/plx015) 12. Clare EL, Symondson WOC, Fenton MB. 2014 An inordinate fondness for beetles? Variation in seasonal dietary preferences of night-roosting big brown bats (Eptesicus fuscus). Mol. Ecol. 23, 3633–3647. (doi:10.1111/mec.12519) 13. Trevelline BK, Latta SC, Marshall LC, Nuttle T, Porter BA. 2016 Molecular analysis of nestling diet in a long-distance Neotropical migrant,

the Louisiana Waterthrush (Parkesia motacilla). Auk 133, 415–428. (doi:10.1642/AUK-15-222.1) 14. Gerwing TG, Kim J-H, Hamilton DJ, Barbeau MA, Addison JA. 2016 Diet reconstruction using next-generation sequencing increases the known ecosystem usage by a shorebird. Auk 133, 168–177. (doi:10.1642/AUK- 15-176.1) 15. Bowser AK, Diamond AW, Addison JA. 2013 From puffins to plankton: A DNA-based analysis of a seabird food chain in the Northern Gulf of Maine. PLoS One 8, 1–16. (doi:10.1371/journal.pone.0083152) 16. Zeale MRK, Butlin RK, Barker G, Lees DC, Jones G. 2010 Taxon- specific PCR for DNA barcoding arthropod prey in bat faeces. Mol. Ecol. Resour. 11, 236-244. (doi:10.1111/j.1755-0998.2010.02920.x) 17. Deagle BE, Thomas AC, Shaffer AK, Trites AW, Jarman SN. 2013 Quantifying sequence proportions in a DNA-based diet study using Ion Torrent amplicon sequencing: which counts count? Mol. Ecol. Resour. 13, 620–33. (doi:10.1111/1755-0998.12103) 18. Nebel S, Mills A, McCracken JD, Taylor PD. 2010 Declines of aerial insectivores in North America follow a geographic gradient. Avian Conserv. Ecol. 5, 1. 19. Environment Canada. 2014. North American Breeding Bird Survey - Canadian Trends Web Site, Data-Version 2012. 20. NABCI Canada. 2012 The State of Canada’s Birds. Ottawa: Environment Canada. 21. COSEWIC. 2011 Assessment and Status Report on the Barn Swallow Hirundo rustica in Canada. Ottawa: Committee on the Status of Endangered Wildlife in Canada. See www.sararegistry.gc.ca/status/status_e.cfm. 22. Heagy A, McCraken J. 2004 Monitoring the State of Ontario’s Migratory Landbirds. Bird Studies Canada. 23. Cadman MD, Sutherland DA, Beck GG, Lepage D, Couturier AR. 2007 Atlas of the Breeding Birds of Ontario. Toronto: Bird Studies Canada, Environment Canada, Ontario Field Ornithologists, Ontario Ministry of Natural Resources, and Ontario Nature. 24. Nocera JJ et al. 2012 Historical pesticide applications coincided with an altered diet of aerially foraging insectivorous chimney swifts. Proc. R. Soc. Biol. Sci. 279, 3114–3120. (doi:10.1098/rspb.2012.0445) 25. English PA, Nocera JJ, Pond BA, Green DJ. 2016 Habitat and food supply across multiple spatial scales influence the distribution and abundance of a

nocturnal aerial insectivore. Landsc. Ecol. 32:343-359. (doi:10.1007/s10980-016-0454-y) 26. Ambrosini R et al. 2012 Maintenance of livestock farming may buffer population decline of the Barn Swallow Hirundo rustica. Bird Conserv. Int. 22, 411–428. (doi:10.1017/S0959270912000056) 27. Vickery JA, Ewing S, Smith KW, Pain DJ, Bairlein F, Škorpilová J, Gregory RD. 2014 The decline of Afro-Paleartic migrants and an assessment of potential causes. Ibis 156, 1–22. (doi:10.1111/ibi.12118) 28. Lee SD, Ellwood ER, Park S, Primack RB. 2011 Late-arriving barn swallows linked to population declines. Biol. Conserv. 144, 2182–2187. (doi:10.1016/j.biocon.2011.05.009) 29. Brown CR, Brown MB. 1999 Barn Swallow Hirundo rustica, the Birds of North America Online. Birds North Am. Online. See http://bna.birds.cornell.edu.cat1.lib.trentu.ca:2048/bna/species/452 (accessed on 7 December 2015). 30. Turner AK. 1982 Optimal foraging by the swallow (Hirundo rustica, L): Prey size selection. Anim. Behav. 30, 862–872. (doi:10.1016/S0003- 3472(82)80160-7) 31. Orłowski G, Karg J. 2013 Partitioning the effects of livestock farming on the diet of an aerial insectivorous passerine, the Barn Swallow Hirundo rustica. Bird Study 60, 111–123. 32. Orłowski G, Karg J. 2013 Diet breadth and overlap in three sympatric aerial insectivorous birds at the same location. Bird Study 60, 475–483. 33. Razeng E, Watson DM. 2015 Nutritional composition of the preferred prey of insectivorous birds: Popularity reflects quality. J. Avian Biol. 46, 89–96. (doi:10.1111/jav.00475) 34. Borg C, Toft S. 2000 Importance of insect prey quality for grey partridge chicks Perdix perdix: A self-selection experiment. J. Appl. Ecol. 37, 557– 563. (doi:10.1046/j.1365-2664.2000.00510.x) 35. Martin TE. 1987 Food as a Limit on Breeding Birds: A Life-History Perspective. Annu. Rev. Ecol. Syst. 18, 453–487. (doi:10.1146/annurev.es.18.110187.002321) 36. Turner A. 2006 The Barn Swallow. London: T & A D Poyser. 37. Møller AP. 1987 Advantages and disadvantages of coloniality in the swallow, Hirundo rustica. Anim. Behav. 35, 819–832. 38. van Achterberg K. 2009 Can Townes type Malaise traps be improved? Some recent developments. Entomol. Ber. 69, 129–135.

39. Campos WG, Pereira DBS, Schoereder JH. 2000 Comparison of the efficiency of flight-interception trap models for sampling Hymenoptera and other insects. An. Soc. Entomol. do Bras. 29, 381–389. 40. Ivanova N V., deWaard JR, Hebert PDN. 2006 An inexpensive, automation-friendly protocol for recovering high-quality DNA. Mol. Ecol. Notes 6, 998-1002. (doi:10.1111/j.1471-8286.2006.01428.x) 41. Clarke LJ, Czechowski P, Soubrier J, Stevens MI, Cooper A. 2014 Modular tagging of amplicons using a single PCR for high-throughput sequencing. Mol. Ecol. Resour. 14, 117–121. (doi:10.1111/1755- 0998.12162) 42. Prosser SWJ, Hebert PDN. 2017 Rapid identification of the botanical and entomological sources of honey using DNA metabarcoding. Food Chem. 214, 183–191. (doi:10.1016/j.foodchem.2016.07.077) 43. Oliver I, Beattie AJ. 1996 Invertebrate morphospecies as surrogates for species: a case study. Conserv. Biol. 10, 99–109. 44. deWaard JR, Ivanova N V, Hajibabaei M, Hebert PDN. 2008 Assembling DNA barcodes: Analytical protocols. In Environmental Genomics (ed C Martin), pp. 275–293. Totawa: Humana Press. (doi:10.1007/978-1-59745- 548-0_15) 45. Ratnasingham S, Hebert PDN. 2007 BOLD : The Barcode of Life Data System (www.barcodinglife.org). Mol. Ecol. Notes 7, 355–364. (doi:10.1111/j.1471-8286.2006.01678.x) 46. Marshall SA. 2006 Insects: their natural history and diversity, with a photographic guide to insects of eastern North America. Buffalo, NY: Firefly Books Ltd. 47. Martin M. 2011 Cutadapt removes adapter sequences from high- throughput sequencing reads. EMBnet J. 17, 10–12. 48. Edgar RC. 2013 UPARSE: Highly accurate OTU sequences from microbial amplicon reads. Nat. Methods 10, 996–998. 49. Oksanen J et al. 2017 vegan: community ecology package. 50. Wang Y, Naumann U, Wright S, Eddelbuettel D, Warton D. 2017 mvabund: statistical methods for analysing multivariate abundance data. Methods Ecol. Evol. 3, 471-474. 51. Jewiss-Gaines A, Marshall SA, Whitworth TL. 2012 Cluster flies (Calliphoridae: Polleniinae: Pollenia) of North America. Can. J. Arthropod Indentif. 19, 1-19. 52. Jedlicka JA, Vo A-TE, Almeida RPP. 2017 Molecular scatology and high- throughput sequencing reveal predominately herbivorous insects in the

diets of adult and nestling Western Bluebirds (Sialia mexicana) in California vineyards. Auk 134, 116–127. (doi:10.1642/AUK-16-103.1) 53. Korpimaki E. 1986 Diet variation, hunting habitat and reproductive output of the kestrel Falco tinnunculus in the light of the optimal diet theory. Ornis Fenn. 63, 84–90. 54. Sih A, Christensen B. 2001 Optimal diet theory: when does it work, and when and why does it fail? Anim. Behav. 61, 379–390. 55. Agosta SJ, Morton D, Kuhn KM. 2003 Feeding ecology of the bat Eptesicus fuscus: ‘preferred’ prey abundance as one factor influencing prey selection and diet breadth. J. Zool. 260, 169–177. (doi:10.1017/S0952836903003601) 56. Twining CW, Brenna JT, Lawrence P, Shipley JR, Tollefson TN, Winkler DW. 2016 Omega-3 long-chain polyunsaturated fatty acids support aerial insectivore performance more than food quantity. Proc. Natl. Acad. Sci. 113, 10920-10925. 57. McKinney ML. 1997 Extinction vulnerability and selectivity : combining ecological and paleontological views. Annu. Rev. Ecol. Syst. 28, 495–516. 58. Dennis RLH, Dapporto L, Fattorini S, Cook LM. 2011 The generalism- specialism debate: The role of generalists in the life and death of species. Biol. J. Linn. Soc. 104, 725–737. (doi:10.1111/j.1095-8312.2011.01789.x) 59. Michel NL, Smith AC, Clark RG, Morrissey CA, Hobson KA. 2016 Differences in spatial synchrony and interspecific concordance inform guild-level population trends for aerial insectivorous birds. Ecography. 39, 774–786. (doi:10.1111/ecog.01798) 60. Joly S, Davies TJ, Archambault A, Bruneau A, Derry A, Kembel SW, Peres-Neto P, Vamosi J, Wheeler TA. 2014 Ecology in the age of DNA barcoding: the resource, the promise and the challenges ahead. Mol. Ecol. Resour. 14, 221–232. (doi:10.1111/1755-0998.12173) 61. Meier R, Shiyang K, Vaidya G, Ng P. 2006 DNA barcoding and taxonomy in Diptera: a tale of high intraspecific variability and low identification success. Syst. Biol. 55, 715–728. (doi:10.1080/10635150600969864) 62. Valentini A, Pompanon F, Taberlet P. 2008 DNA barcoding for ecologists. Trends Ecol. Evol. 24, 110–117. (doi:10.1016/j.tree.2008.09.011) 63. Thomas AC, Deagle BE, Eveson JP, Harsch CH, Trites AW. 2015 Quantitative DNA metabarcoding: improved estimates of species proportional biomass using correction factors derived from control material. Mol. Ecol. Resour. 16, 714–726. (doi:10.1111/1755-0998.12490) 64. Lewis T. 1969 The distribution of flying insects near a low hedgerow. J. Appl. Ecol. 6, 443–452. 40

65. Darling CD, Packer L. 1988 Effectiveness of Malaise traps in collecting Hymenoptera: the influence of trap design, mesh size, and location. Can. Ent. 120, 787-796.

FIGURES

Figure 2.1 Accumulation curve of arthropod families identified in the diet of Barn Swallow nestlings. 95% confidence intervals are represented in grey.

Figure 2.2 The frequency of occurrence of different arthropod families in the diet of Barn Swallow nestlings compared to the availability of these arthropod families in the habitat presented as a proportion of the total prey available in the habitat. Each bar on the x-axis represents a different family and bars are coloured by order. The five most frequently consumed families were: Tachinidae, Tipulidae, Limoniidae, Calliphoridae, and Anthomyiidae. The five most abundant insect families in the habitat were: Chironomidae, Sciaridae, Cecidomyiidae, Ceratopogonidae, and Staphylinidae.

Figure 2.3 The frequency of occurrence of different sizes of arthropod in the diet of Barn Swallow nestlings compared to the availability of these sizes of arthropod prey in the habitat presented as a proportion of the total prey available in the habitat. Bars are coloured to show the proportion of operational taxonomic units (OTUs) in each size class represented by each order.

TABLES

Table 2.1 ANOVA summary of multivariate GLMs of the prey community in the diet of Barn Swallow nestlings and of the available prey community in the habitat with site, year and brood as predictor variables.

Predictor Res.Df Dev Pr(>Dev) Diet Site 261 1054.7 0.001 Brood 260 163 0.013 Year 259 162.4 0.05 Site:Brood 250 447.2 0.001 Site:Year 242 561.8 0.001 Brood:Year 241 95.2 0.001 Site:Brood:Year 235 178.7 0.001

Habitat Site 231 5225 0.001 Brood 230 1447 0.001 Year 229 431 0.001 Site:Brood 220 1744 0.001 Site:Year 211 3660 0.001 Brood:Year 210 290 0.001 Site:Brood:Year 201 1244 0.001

Chapter 3: Does prey availability affect the reproductive performance of Barn Swallows, Hirundo rustica, breeding in Ontario, Canada?

ABSTRACT

Food availability during the breeding season can be an important regulator of population size in birds by influencing annual fecundity. Barn Swallows (Hirundo rustica), a species of aerial insectivore, have experienced population declines in the past

30 years in North America that are thought to be related to changes in prey availability.

Decreases in insect populations or shifts in insect phenology could lead to limited food availability during periods of high food demand, which could limit the annual fecundity of Barn Swallows and contribute to population declines. We monitored Barn Swallow reproductive behaviour at ten breeding sites in Ontario, Canada and collected aerial insects using Malaise traps to monitor food availability throughout two breeding seasons.

We found no relationship between the number of eggs laid or the number of young fledged and food availability. Additionally, we did not observe higher rates of second brooding or more pairs nesting at breeding sites with higher food availability. We did see a relationship between the number of pairs and the number of old nests at a breeding site, although there was no apparent reproductive benefit to nesting at sites with more old nests. Barn Swallows did not time their reproductive effort to maximize prey availability during the nesting period, but any mismatch in phenology of prey and bird reproduction at a site was not associated with lower reproductive success. These results do not support

46 our hypotheses that food availability limits the reproductive output and affects the reproductive behaviour of Barn Swallows. While there may be effects that we could not detect, such as low post-fledging survival, the results of this study suggest that Barn

Swallow populations in eastern North America are not currently limited during the breeding season by low reproductive success due to food limitation on the breeding grounds.

INTRODUCTION

The breeding season is an energetically demanding time for birds and food availability is important for successful reproduction (Martin 1987). Limited food availability during the breeding season can limit annual fecundity (Zanette et al. 2003,

Nagy and Holmes 2005a, Preston and Rotenberry 2006) and high fecundity is important for maintaining populations in short-lived species (Sæther and Bakke 2000). Therefore, food availability during the breeding season can, in turn, affect population sizes and limited food availability could be associated with population declines.

Food availability during the breeding season can impact annual fecundity (defined as the number of fledglings produced per pair per year) through various components of the breeding cycle including clutch size, fledging success, and the number of broods in a year (Martin 1987, Nagy and Holmes 2005b). Lack (1947) suggested that, in altricial birds, clutch size is limited by the number of nestlings that parents can adequately feed.

Support for Lack’s (1947) hypothesis comes from a number of species where larger clutches are laid when there is more food available (Song Sparrow, Melospiza melodia:

Zanette et al. 2006; review: Ruffino et al. 2014) and smaller clutches are laid when food

47 is limited (Korpimäki and Hakkarainen 1991, Korpimäki and Wiehn 1998). Also, for many species greater food availability has been positively correlated with higher fledging success rates (Bryant 1975, Granbom and Smith 2006, Schoech et al. 2008). Several species are more likely to double brood in areas or during years with higher food availability (Common Tern, Sterna hirundo: Moore and Morris 2005; Black-Throated

Blue Warbler, Setophaga caerulescens: Nagy and Holmes 2005a, b). However, changes in food availability do not cause changes in annual fecundity in all species (e.g., Hooded

Warbler, Setophaga citrina: Nagy and Smith 1997; Red-Eyed Vireo, Vireo olivaceus:

Marshall et al. 2002). The lack of response in reproductive success to limited food availability in these species may be because the effects of food limitation are strongest during the post-fledgling period or because predation was a stronger driver of reproductive success than food availability (Nagy and Smith 1997, Marshall et al. 2002).

Most studies investigating the effects of food availability on reproductive performance experimentally reduce or supplement the food supply during the breeding season

(Newton 1980, Ruffino et al. 2014). By contrast, there are fewer studies examining how natural variation in food availability affects the annual fecundity of birds through various components of the breeding cycle.

Barn Swallows, Hirundo rustica, as well as most North American aerial insectivores, have experienced dramatic population declines over the past 30 years (Nebel et al. 2010). While the cause of these population declines has not been determined, many possible factors have been suggested. Due to the foraging strategy shared by the birds in this guild, a decline in their food source – flying insects – is one of the commonly suggested causes for the decline in aerial insectivores (Nebel et al. 2010, Nocera et al.

2012). A decrease in the quality or quantity of insect prey available during the breeding season could be limiting the reproductive output of Barn Swallows and contributing to the population decline. A decrease in food availability could be caused by a broad-scale decrease in insect abundance which may be the result of widespread pesticide use reducing insect populations (Dirzo et al. 2014). Pesticide application could also cause a change in the prey community available, where high quality prey items are more affected by pesticides than low quality items (Nocera et al 2012). A decrease in food availability could also be caused by phenological shifts in insect emergence that are not matched by shifts in Barn Swallow reproduction (Both et al. 2010). The timing of insect emergences is advancing due to warmer spring temperatures as a result of climate change, while migrant birds are limited in their ability to shift the timing of reproduction (Dunn et al.

2011). Phenological shifts in food availability have led to asynchrony between the peak in food availability and the peak food demand for several species (e.g., Visser et al. 2006,

Dunn et al. 2011, McKinnon et al. 2012). A mismatch between the timing of reproduction and food availability, where reproduction is not occurring during peak food availability, can lead to lower reproductive success (Both 2010).

Reproductive success in European Barn Swallows has been linked to prey availability (Turner 2006). Higher reproductive success and larger clutch sizes were associated with the presence of livestock on farms and livestock farming was associated with a greater availability of large prey items, which suggests that higher reproductive success may be due to increased prey availability (Møller 2001, Ambrosini et al. 2002,

Grüebler et al. 2010). Furthermore, Møller (2001) observed higher rates of second brooding at nesting sites with livestock farming, which had higher prey availability than

49 sites without livestock. Barn Swallows also experienced higher reproductive success at nests at the periphery of urban areas in Denmark, where food availability was higher compared to nests in the centre of the urban area (Teglhøj 2017). In North America, there is no published relationship between the presence of livestock at a nesting site and clutch size or reproductive success (Bossuyt, unpublished). Barn Swallow reproductive success in North America has been linked with weather variables, where cold, wet weather and very hot, dry weather both lead to lower prey availability and poor breeding success

(Turner 2006).

The timing of clutch initiation by Barn Swallows has also been linked to the presence of livestock at nesting sites in Europe, where clutch initiation was later at sites without livestock (Ambrosini et al. 2002). Later clutch initiation could be due to either higher prey availability at sites with livestock or due to higher temperatures in barns with livestock. Temperature can predict clutch initiation date in Barn Swallows (Saino et al.

2004). Ambient temperature is an important factor governing insect development (Ratte

1984) therefore by using temperature as a cue for the timing of laying, swallows may also be timing their reproduction to maximize prey availability. In some parts of Europe, Barn

Swallows appear to be laying earlier in response to warming spring temperatures (Crick and Sparks 1999, Møller 2007).

Barn Swallows are colonial breeders and the number of pairs nesting at a breeding site can influence the prey available to each pair (Turner 2006). Therefore, prey availability can affect reproductive decisions at the scale of breeding site. As colony size increases, competition for prey resources may also increase although, sites with more

50 prey available could support more pairs. A variety of factors, including food availability, have been shown to affect the colony size of Barn Swallows in Europe. The presence of livestock and traditional stables, lee from prevailing winds, and high food availability were all associated with larger colony sizes (Møller 1987, Ambrosini et al. 2002).

Multiple studies from across the Barn Swallows’ range have shown a positive correlation between the number of old nests and the number of pairs nesting at a breeding site

(Safran 2004, Ringhofer and Hasegawa 2014). Old nests provide a resource to Barn

Swallows by removing the need to build a new nest and they provide an indirect social cue of the presence and reproductive success of conspecifics in previous years (Ringhofer and Hasegawa 2014). Ultimately, Barn Swallows should select their breeding site to maximize reproductive success, which could be impacted by food availability and competition at the breeding site.

We studied the reproductive performance of Barn Swallows breeding in Ontario,

Canada, and addressed three hypotheses. First, we hypothesized that the number of pairs nesting at a breeding site and the number of those pairs that double brood are dependent on the prey availability at the breeding site. We predicted that sites with higher prey availability would have more nesting pairs and more of these pairs would have second broods. Secondly, we hypothesized that the timing of clutch initiation of the first brood will be related to prey availability in the habitat. We predicted that nests would be laid later at sites with a later peak in prey availability. Finally, we hypothesized that Barn

Swallow reproductive output is limited by prey availability during the nesting season. We predict that Barn Swallows will lay more eggs and fledge more young when there is more prey available.

METHODS

Study Site

We conducted our study at 10 Barn Swallow breeding sites in Peterborough

County and the City of Kawartha Lakes in Ontario, Canada (Figure 1.1). Each breeding site consisted of a barn and associated buildings surrounded by a mixture of cropland, pasture, and/or hayfields. We conducted our study over the 2015 and 2016 breeding seasons. The breeding season was separated into the arrival period and the nesting period for analyses. The arrival period was defined as a 4-week period beginning in late April, when Barn Swallows begin arriving at nest sites in Ontario, and ending in late May, before the first nests hatched. This period included clutch initiation and incubation for early nesting pairs at some sites. The nesting period was a 12-week period that began when the first nests hatched.

Nest Monitoring

At each breeding site, we checked all nests every 2 to 3 days. Any newly constructed nests were noted and subsequently checked. We recorded clutch initiation, hatching, and fledging dates and counted the number of eggs, nestlings, and fledglings for each nest. If clutch initiation occurred on a day we did not check nests, the initiation date was calculated based on the number of eggs in the nest, assuming that one egg was laid per day (Brown and Brown 1999, Møller et al. 2005). If eggs hatched on a day when we did not check the nests, hatch date could be determined by the age of nestlings, as described in Turner (2006). If the young were not observed fledging, we considered the most recent nestling count made on day 16 or later as the number of fledglings (Barclay

1988).

Insect Collection

We used land and air Malaise traps with bottom collectors (BioQuip #2869 &

#2892) to monitor insect abundance and diversity at each breeding site. At each site, we set up a trap 1 m above the ground and 15-30 m from the structure used for nesting. Barn

Swallows usually forage <10 m in height and often <1 m from the ground (Brown and

Brown 1999) and spend most of their time foraging within a few hundred metres of the nest site (Møller 1987). Malaise traps paired with bottom collectors are effective at collecting flying insects from a range of taxa (Campos et al. 2000, van Achterberg 2009).

Collection bottles were placed on traps at dawn (sunrise +/- 30 minutes) and removed at dusk (sunset +/- 30 minutes). In 2015, we began collecting insects on 20 May 2015 at 2 sites and began collecting insects on 1 June 2015 at the remaining 8 sites. Collection continued at all sites until 29 August 2015. Traps were open for 2 days/week at 9 sites and 6 days/week at one site. Logistically, we could not operate all traps daily because of the distance between sites and we found that the range of prey items Barn Swallows consumed was captured using 2 days of collection. Therefore, in 2016, all traps were open for 2 days/week at all sites. Collection began on 25 April and continued through to

29 August. Specimens were stored in 95% ethanol at -20°C until processing.

Insect Identification

We sorted insect specimens collected in Malaise traps to morphospecies (Oliver and Beattie 1996). We measured the body length to the nearest 0.1 mm (excluding appendages such as antennae or cerci) of up to five specimens of each morphospecies, depending on abundance. Average lengths were calculated for each morphospecies.

Representative photos were taken of one specimen from each morphospecies. A

53 representative from each morphospecies was identified using DNA barcoding. Specimens were processed at the Canadian Centre for DNA Barcoding (CCDB) following their standardized extraction, amplification and uni-directional sequencing protocols (see

Ivanova et al. 2006, deWaard et al. 2008). The resultant sequences were identified using the BOLD Identification System (www.boldsystems.org; Ratnasingham and Hebert

2007). For any specimens that did not sequence successfully or did not match a reference sequence on BOLD, we used photos to identify the specimens to order and, where possible, to family (Marshall 2006).

The number of insect specimens caught per day of collection was used as an index of insect abundance. Insect biomass was estimated using length-mass equations from

Sabo et al. (2002) for insect orders and families, where possible. Prey abundance was calculated by removing insect families that were not detected in Barn Swallow diet (see

Chapter 2) from the total insect abundance. Family level identifications were used because that was the highest level of taxonomic resolution where most insects were identified in both the diet and from Malaise traps. Our measure of prey abundance was further narrowed to preferred prey abundance based on prey size. Barn Swallows prefer larger prey items, as seen in previous studies (Turner 1982, Orłowski and Karg 2013,

Chapter 2). We included prey larger than 3 mm in length as preferred prey since there was a large increase in the frequency of occurrence of prey items larger than 3 mm in length in the diet compared to smaller prey items (Chapter 2).

Data Analysis

All statistical analyses were conducted using R Studio v 1.0.136. We report the total number of nest attempts across all sites, hatch success (= the percentage of nests that successfully hatched at least one chick), nest success (= the percentage of nests that fledged at least one young), predation (= the percentage of nests known to be depredated), the mean number of eggs, and the mean number of young fledged. Means are presented with standard errors. The number of eggs laid and the number of young fledged per nest was compared between years and broods using generalized linear models with a Poisson family distribution for count data. Model fit was assessed with a chi- square goodness-of-fit test.

Colony Size. We modelled colony size from both 2015 and 2016 against the predictor variables prey abundance, large prey abundance, prey biomass, and large prey biomass, using specimens collected over the 12-week period nesting period. Additionally, we modelled colony size from 2016 with prey abundance, large prey abundance, prey biomass and large prey biomass during the arrival period. We also modelled colony size with the number of old nests present at each breeding site at the beginning of the breeding season. To determine if the presence of old nests was associated with higher breeding success, we modelled the number of old nests with the average number of successful nests and fledglings produced per pair at each breeding site.

Second Brooding. Using linear mixed models implemented using the ‘lme4’ and

‘lmerTest’ packages (Bates et al. 2015, Kuznetsova et al. 2016), we modelled the proportion of pairs at each site that had a second brood by the abundance and biomass of

55 prey and the abundance and biomass of large prey over the whole nesting season, during the early brood and during the late brood. We also modelled the proportion of pairs with a second brood with the average clutch initiation date of the first brood at each breeding site. Year and site were included as random effects in all models.

Timing of Reproduction. To compare the timing of peak food demand and peak prey availability, we used data from 2016 only, when insect collection began early enough in the season to include the beginning of the laying period. Prey phenology was quantified as the available prey distribution using prey biomass over the breeding season and the demand distribution for Barn Swallows was quantified as the predicted temporal distribution of 13-day-old broods, which is when the energy demands of the nest are the highest (Jones 1987). Prey biomass and the number of 13-day-old broods were plotted as percentiles of the season total value across all breeding sites, to standardize scales for direct comparison between the two distributions and to facilitate comparisons between breeding sites. Following Kwon (2016), a smoothing curve was applied to the available food distribution and the demand distribution at each breeding site against the day of the year using P-splines and a smoothing parameter of 0.6 in the ‘mgcv’ package in R (Wood

2006). The smoothed curve for demand was overlaid with the food availability curve and the area of overlap between the two curves was calculated using the integrate.xy function in the ‘sfsmisc’ package (Maechler 2015). The coefficient of overlap between prey availability and demand at each breeding site was calculated, where a lower coefficient of overlap indicates more mismatch (Kwon 2016). To determine if phenological overlap had any effect on reproductive measures, the coefficient of overlap was compared to the average number of fledglings per pair and the average number of successful nests per pair

56 at each breeding site using linear models. To determine if swallows breed later at sites where prey availability peaks later, the date of peak prey availability was identified from the smoothed curves for each site and compared to the date of peak demand at each breeding site using a linear model.

Reproductive Output. We modelled the impacts of predictor variables on the number of eggs laid and the number of fledglings using generalized linear mixed models with a

Poisson distribution for count data in the ‘glmm’ package (Knudson 2016). To model the number of eggs per nest, we included prey abundance and prey biomass during the week prior to clutch initiation (2 days of insect collection) as predictors. These models excluded nests that were initiated prior to the start of our insect collection in 2015. The fledgling models included all 2015 and 2016 nests and include the following predictors: prey abundance, prey biomass, prey abundance index, prey biomass index, large prey abundance, large prey biomass, large prey abundance index, and large prey biomass index. These variables were included for the three weeks after hatching or 6 days of insect collection, since the Barn Swallow nestling period usually lasts 20 to 21 days

(Brown and Brown 1999). Index values were calculated by dividing the abundance or biomass on each collection day by the number of active nests at that site on that day. This standardization controls for variable colony sizes and provides a measure of the amount of prey available per nest at a site. Year, brood, and site were included in all models as random effects.

Correlation matrices of prey availability predictor variables for each response variable (colony size, rate of double brooding, and reproductive success) can be found in figures B.1 - B.3 in Appendix B.

RESULTS

Breeding sites had between one and 26 nesting pairs each year with a mean of

12.6 ± 2.28 nest attempts at each site per breeding season and sites were very consistent in the numbers of nesting pairs between years (Figure 1.1). In both years, we observed a high level of hatching success and nest success (Table 3.1). Hatching and nest success were both higher in 2016 than in 2015. The percent of nests lost to predators varied considerably and there were no losses to predators in the second brood in 2016 (Table

3.1). There was no significant difference between years or between broods in the number of eggs laid or the number of young fledged. In 2015, the rate of second brooding was

2 60.2 % and in 2016, it was 37.9 % (two-sample proportion test, X = 7.162, df = 1, p =

0.007).

Colony Size and Rates of Second Brooding

In both years, the number of old nests was a significant predictor of the number of pairs present (Figure 3.1, Table 3.2). In 2016, the abundance of large prey during the nesting season was also a significant predictor of the number of pairs but did not explain as much of the variation as the number of old nests (Table 2). Combining the number of old nests and the abundance of large prey items into one model did not improve the fit of the model and only the number of old nests was significant (r2 = 0.71, old nests: p = 0.03, large prey abundance: p = 0.88). In both years, the best predictor of the number of nesting

58 pairs was the number of old nests. The number of old nests showed no relationship with the number of successful nests per pair (r2 = 0.09, p = 0.2) or the number of fledglings produced per pair (r2 = 0.0005, p = 0.9). No measures of prey availability were significant predictors of the proportion of pairs that had a second brood nor was the average clutch initiation date a significant predictor of this variable (Table 3.3).

Timing of Reproduction

The peak in prey availability and the amount of overlap between the prey availability and the demand curve varied considerably between breeding sites, while the peak in food demand was much less variable between sites (Figure 3.2, 3.3). There was no significant relationship between the date of peak prey availability and the date of peak demand (r2 = 0.04, p = 0.57). Nor was there a significant relationship between the coefficient of overlap and the average number of fledglings per pair (r2 = 0.01, p = 0.8) or the average number of successful nests per pair at each breeding site (r2 = 0.09, p = 0.5).

Reproductive Output

No measures of prey availability were significant predictors of the number of eggs laid or the number of young fledged (Table 3.4). Clutch initiation date was a significant predictor of the number of eggs laid and hatch date was a significant predictor of the number of fledglings. The number of eggs laid and the number of young fledged decreased with later clutch initiation and hatch dates, while controlling for brood as a random effect (Figure 3.4).

DISCUSSION

Overall, we observed a high level of hatching success and reproductive success of

Barn Swallows in central Ontario, Canada. Our results are comparable to other studies of

Barn Swallow reproduction across their range, which show high reproductive success with approximately 90% of eggs hatching and 80-90% of chicks fledging (Shields and

Crook 1987, Safran 2004, 2006). Nest predation was relatively low overall, although we did observe a seasonal decline in nest predation indicated by lower predation on second broods than first broods. The main nest predators of Barn Swallows in our study area were raccoons, Procyon lotor, and squirrels, Tamiasciurus hudsonicus and Sciurus carolinensis (pers. obs., Brown and Brown 1999), although ravens (Corvus corax) were also observed at one site entering the barn, knocking down nests, and eating nestlings.

Studies show mixed results for the change in predation risk over the breeding season for different bird species (Brown and Roth 2002, Davis 2005, Grant et al. 2005, Kroll and

Haufler 2009). A decline in predation risk over the breeding season can be due to changes in predator food availability with more alternate food sources available later in the season

(Borgmann et al. 2013).

Contrary to our prediction, we did not observe any relationship between prey availability and reproductive performance in Barn Swallows. Despite wide variation in prey availability over time and between sites, our models showed that factors other than prey availability explained more variation in colony size, second brooding, clutch size, and the number of young fledged. The lack of relationship between prey availability and reproductive success suggests that food is not currently limiting the number of offspring that Barn Swallows can produce. Additionally, the dates of peak food demand were not

60 related to the dates of peak prey availability at each breeding site and the coefficient of overlap between the prey availability and demand curves was not related to reproductive success. This suggests that Barn Swallows are not adjusting the timing of their reproductive behaviour in response to local prey availability on the breeding grounds and that currently, there is no selection pressure on the timing of reproduction to maximize the prey available to offspring within the nesting period.

More Barn Swallows nested at breeding sites with more old nests leading to larger colonies at sites with more old nests. This relationship has been observed previously in

Barn Swallows (Safran 2004, Ringhofer and Hasegawa 2014). These old nests may be a reliable cue that can reduce the time needed for Barn Swallows to assess nest sites.

However, old nests can be used as cues even when they are no longer a reliable cue of current habitat quality. Habitat degradation or loss could occur in the area decreasing the quality of nest sites, but nests can be used for up to 12-15 years (Turner 2006). In this study, old nests did not appear to provide any cues about breeding site quality as there was no relationship between the average nest success or the number of young produced at each site and the number of old nests. Alternatively, old nests may be attractive to Barn

Swallows because nest reuse can have reproductive benefits. Safran (2006) found that

Barn Swallows reusing nests raised more young than those that built new nests. Reusing nests may not always be advantageous since old nests are more likely to harbour parasites that are detrimental to chick health (Møller 1990). Very few nests in our study were newly constructed so we could not compare the reproductive output between old and new nests.

We observed a range in the proportion of double brooding between sites but this variation was not explained by variation in prey availability among sites. By contrast, studies on other species have shown a higher rate of second brooding and higher survival of second broods when food availability is high (Moore and Morris 2005, Nagy and

Holmes 2005b, Grüebler and Naef-Daenzer 2008). Barn Swallows are more likely to attempt a second brood if their first brood was unsuccessful and/or if the first clutch was laid early (Verboven and Verhulst 1996). In turn, the timing and success of the first brood is affected by other factors, including weather conditions and predation (Turner 2006).

Older, more experienced females have more second broods (Bańbura and Zieliński 1998) which may be explained by the fact that older females also tend to lay earlier (Turner

2006). We did not observe a higher rate of second brooding at sites with earlier peak clutch initiation dates. The variation we observed in second brooding between sites could be the result of differential reproductive success between nests in the first brood or due to age differences between females. From 2015 to 2016, we saw a decrease in the overall rate of second brooding across all sites. In 2015, the weather was warm and wet while in

2016, the weather was hot and very dry. This change in weather conditions was related to drastically reduced prey abundance during the first brood but not the second brood (see

Figure B.4 in Appendix B). This decrease in prey abundance was not associated with fewer fledglings from nests in the first brood, but it may have affected individual decisions to second brood. If parents had to expend more energy providing food for the first brood, they may not have had enough energy to have a second brood (Nagy and

Holmes 2005b). We had only two years of data for this study and without more years of

62 data, we cannot make any broad conclusions about how weather conditions affect prey abundance and reproductive success.

Individual variation between females may have played a role in the variation in reproductive success and output. Younger females tend to lay smaller clutches (Bańbura and Zieliński 1998, Safran 2006). The condition of the female upon arrival on the breeding grounds may also affect the reproductive success of the nest (Turner 2006). We did not monitor female age or condition and were unable to account for this in our models. We did control for clutch initiation date since clutches laid earlier in the season are larger and fledge more young (Turner 2006). Studies show mixed results for whether ambient temperature affects clutch size. Differences in ambient temperature at breeding sites in Europe were mostly driven by the presence/absence of livestock at the breeding sites (Saino et al. 2004, Ward and Bryant 2006). While we did not measure ambient temperature in barns, previous work at these sites showed no effect of temperature on clutch size or clutch initiation date (Bossuyt, unpublished).

We observed considerable variation in first clutch initiation dates between sites

(range of peak clutch initiation dates: 11 May – 16 Jun) but no effect of prey availability on these dates. While many studies report differences in clutch initiation due to local weather conditions such as temperature and precipitation (Turner 2006), these sites were relatively close (~20 km) and experienced very similar weather conditions. Thus, variation in clutch initiation between sites does not appear to be explained by weather variables. The difference in clutch initiation date could be due to individual differences in birds, including female condition, age, and arrival date at the nest site (Turner 2006). If

63 prey availability is not currently limiting reproductive success as our results suggest, then there would not be strong selective pressure for clutch initiation to occur at a certain time within the nesting period. In addition, as generalist opportunistic foragers Barn Swallows should not be affected by mismatch as much as specialist predators. This was further supported by a lack of relationship between the degree of overlap between prey availability and demand curves and reproductive success at each breeding site. We observed considerable variability in the timing of peak prey availability between sites, which means that selection cannot act on a population level to synchronize the timing of reproduction with peak prey availability.

While our results suggest that prey availability is not currently affecting reproductive performance, it is possible that prey availability is having an effect that we were unable to detect. Low prey availability may have decreased chick quality leading to low post-fledgling survival. We did not monitor chick quality during the nestling period.

Alternatively, there must be a minimum threshold of food requirements and, as long as prey availability is above that threshold, nests can be raised successfully and variation in clutch/brood size is explained by other factors, such as those discussed above. It is possible that prey availability was always high enough but there was not always enough high quality prey available. Barn Swallows consume a wide range of prey items and their diet is flexible so they can take advantage of whatever prey is available even if high quality prey is not available. To control for this, we quantified the total prey availability and the availability of large prey to represent high quality prey and we used both biomass and abundance as measures of availability. This suggests that Barn Swallows are not currently limited by low availability of high quality prey. However, prey may vary in

64 quality based on other measures such fatty-acid or micronutrient levels (Twining et al.

2016). The Malaise traps we used cannot represent the patchiness of prey resources which may display local abundances in microhabitats, such as along hedgerows (Lewis

1969). We demonstrated that the traps caught most of the prey diversity consumed by

Barn Swallows (see Chapter 2) however small-scale changes in prey availability across the landscape are not reflected in our trap catch. It is possible that these local abundances could offset low prey abundance in our trap catch.

Limited food availability during the breeding season is a proposed driver of the population decline in Barn Swallows in North America. Our results do not support our hypotheses that prey availability is affecting colony size, reproductive success, second brooding, or the timing of reproduction of Barn Swallows. We observed a high overall reproductive success and no relationship between prey availability and reproductive performance. This suggests that prey availability on the breeding grounds is not currently limiting the population of Barn Swallows in our study area since reproductive performance was not explained by food availability as measured here. Food limitation during the breeding season may still affect the population by causing negative carry-over effects on adults and/or juveniles during migration and wintering. Furthermore, this study only included two years of data and food limitation during the breeding season may have affected Barn Swallow populations prior to this study. Nevertheless, the results presented here and the high overall reproductive success we observed suggest that prey availability on the breeding grounds is not currently affecting Barn Swallow populations during the breeding season.

REFERENCES

Ambrosini, R., Bolzern, A. M., Canova, L., Arieni, S., Møller, A. P. and Saino, N. 2002. The distribution and colony size of barn swallows in relation to agricultural land use. - J. Appl. Ecol. 39: 524–534. Bańbura, J. and Zieliński, P. 1998. Timing of breeding, clutch size and double- broodedness in Barn Swallows Hirundo rustica. - Ornis Fenn. 75: 177–183. Barclay, R. M. R. 1988. Variation in the costs, benefits, and frequency of nest reuse by Barn Swallows (Hirundo rustica). - Auk 105: 53–60. Bates, D., Maechler, M., Bolker, B. and Walker, S. 2015. Fitting linear mixed-effects models using lme4. - J. Stat. Softw. 67: 1–48. Borgmann, K. L., Conway, C. J. and Morrison, M. L. 2013. Breeding phenology of birds: mechanisms underlying seasonal declines in the risk of nest predation. - PLoS One 8: e65909 Bossuyt, S. Effects of livestock and other environmental variables on the reproductive success of North American Barn Swallows. Honours Thesis. Both, C. 2010. Food availability, mistiming, and climactic change. - In: Møller, A. P. et al. (eds), Effects of Climate Change on Birds. Oxford University Press, pp. 129– 147. Both, C., Van Turnhout, C. a M., Bijlsma, R. G., Siepel, H., Van Strien, A. J. and Foppen, R. P. B. 2010. Avian population consequences of climate change are most severe for long-distance migrants in seasonal habitats. - Proc. R. Soc. Biol. Sci. 277: 1259–1266. Brown, C. R. and Brown, M. B. 1999. Barn Swallow Hirundo rustica, the Birds of North America Online. - Birds North Am. Online Brown, W. P. and Roth, R. R. 2002. Temporal patterns of fitness and survival in the Wood Thrush. - Ecology 83: 958–969. Bryant, D. M. 1975. Breeding biology of house martins Delichon urbica in relation to aerial insect abundance. - Ibis (Lond. 1859). 117: 180–216. Campos, W. G., Pereira, D. B. S. and Schoereder, J. H. 2000. Comparison of the efficiency of flight-interception trap models for sampling Hymenoptera and other insects. - An. Soc. Entomol. do Bras. 29: 381–389. Crick, H. Q. P. and Sparks, T. H. 1999. Climate change related to egg-laying trends. - Nature 399: 423–424. Davis, S. K. 2005. Nest-site selection patterns and the influence of vegetation on nest survival of mixed- grass prairie passerines. - Condor 107: 605–616.

66 deWaard, J. R., Ivanova, N. V, Hajibabaei, M. and Hebert, P. D. N. 2008. Assembling DNA barcodes: Analytical protocols. - In: Martin, C. (ed), Environmental Genomics. Humana Press, pp. 275–293. Dirzo, R., Young, H. S., Galetti, M., Ceballos, G., Isaac, N. J. B. and Collen, B. 2014. Defaunation in the Anthropocene. - Science 345: 401–406. Dunn, P. O., Winkler, D. W., Whittingham, L. A., Hannon, S. J. and Robertson, R. J. 2011. A test of the mismatch hypothesis: How is timing of reproduction related to food abundance in an aerial insectivore? - Ecology 92: 450–461. Granbom, M. and Smith, H. G. 2006. Food limitation during breeding in a heterogeneous landscape. - Auk 123: 97–107. Grant, T. A., Shaffer, T. L., Madden, E. M. and Pietz, P. J. 2005. Time-specific variation in passerine nest survival: new insights into old questions. - Auk 122: 661–672. Grüebler, M. U. and Naef-Daenzer, B. 2008. Fitness consequences of pre- and post- fledging timing decisions in a double-brooded passerine. - Ecology 89: 2736– 2745. Grüebler, M. U., Korner-Nievergelt, F. and von Hirschheydt, J. 2010. The reproductive benefits of livestock farming in barn swallows Hirundo rustica: quality of nest site or foraging habitat? - J. Appl. Ecol. 47: 1340–1347. Ivanova, N. V., deWaard, J. R. and Hebert, P. D. N. 2006. An inexpensive, automation- friendly protocol for recovering high-quality DNA. - Mol. Ecol. Notes 6: 998- 1002. Jones, G. 1987. Parent-offspring resource allocation in swallows during nestling rearing: an experimental study. - Ardea 75: 145–168. Knudson, C. 2016. glmm: Generalized linear mixed models via Monte Carlo likelihood approximation. R package version 1.1.1. https://CRAN.R- project.org/package=glmm Korpimäki, E. and Hakkarainen, H. 1991. Fluctuating food supply affects the clutch size of Tengmalm’s owl independent of laying date. - Oecologia 85: 543–552. Korpimäki, E. and Wiehn, J. 1998. Clutch size of kestrels: seasonal decline and experimental evidence for food limitation under fluctuating food conditions. - Oikos 83: 259–272. Kroll, A. J. and Haufler, J. B. 2009. Age and clutch size variation in dusky flycatcher nest survival. - J. Ornithol. 150: 409–417. Kuznetsova, A., Brockhoff, P. B. and Christensen, R. H. B. 2016. lmerTest: Tests in linear mixed effects models. R package version 2.0-33. https://CRAN.R- project.org/package=lmerTest

Kwon, E. 2016. Effects of climate change on the breeding ecology and trophic interactions of Arctic-breeding shorebirds. PhD Thesis. Lack, D. 1947. The significance of clutch-size. - Ibis 90: 302–352, 25–45. Lewis, T. 1969. The distribution of flying insects near a low hedgerow. - J. Appl. Ecol. 6: 443–452. Maechler, M. 2015. sfsmisc: Utilities from Seminar fuer Statistik ETH Zurich. Marshall, S. A. 2006. Insects: their natural history and diversity, with a photographic guide to insects of eastern North America. - Firefly Books Ltd. Marshall, M. R., Cooper, R. J., Dececco, J. A., Strazanac, J. and Butler, L. 2002. Effects of experimentally reduced prey abundance on the breeding ecology of the Red- Eyed Vireo. - Ecol. Appl. 12: 261–280. Martin, T. E. 1987. Food as a limit on breeding birds: a life-history perspective. - Annu. Rev. Ecol. Syst. 18: 453–487. McKinnon, L., Picotin, M., Bolduc, E., Juillet, C. and Bêty, J. 2012. Timing of breeding, peak food availability, and effects of mismatch on chick growth in birds nesting in the High Arctic. - Can. J. Zool. 90: 961–971. Møller, A. P. 1987. Advantages and disadvantages of coloniality in the swallow, Hirundo rustica. - Anim. Behav. 35: 819–832. Møller, A. P. 1990. Effects of parasitism by a haematophagous mite on reproduction in the Barn Swallow. - Ecology 71: 2345–2357. Møller, A. P. 2001. The effect of dairy farming on barn swallow Hirundo rustica abundance, distribution and reproduction. - J. Appl. Ecol. 38: 378–389. Møller, A. P. 2007. Interval between clutches, fitness, and climate change. - Behav. Ecol. 18 :62–70. Møller, A. P., Mousseau, T. A., Milinevsky, G., Peklo, A., Pysanets, E. and Szép, T. 2005. Condition, reproduction and survival of barn swallows from Chernobyl. - J. Anim. Ecol. 74: 1102–1111. Moore, D. J. and Morris, R. D. 2005. The production of second slutches in the Common Tern: proximate effects of timing and food supply. - Waterbirds 28: 458–467. Nagy, L. R. and Smith, K. G. 1997. Effects of insecticide-induced reduction in lepidopteran larvae on reproductive success of hooded warblers. - Auk 114: 619– 627. Nagy, L. R. and Holmes, R. T. 2005a. Food limits annual fecundity of a migratory songbird: an experimental study - Ecology 86: 675–681.

Nagy, L. R. and Holmes, R. T. 2005b. To double-brood or not? Individual variation in the reproductive effort in black-throated blue warblers (Dendroica caerulescens). - Auk 122: 902–914. Nebel, S., Mills, A., Mccracken, J. D. and Taylor, P. D. 2010. Declines of aerial insectivores in North America follow a geographic gradient. - Avian Conserv. Ecol. 5: 1. Newton, I. 1980. The role of food in limiting bird numbers. - Ardea 68: 11–30. Nocera, J. J., Blais, J. M., Beresford, D. V, Finity, L. K., Grooms, C., Kimpe, L. E., Kyser, K., Michelutti, N., Reudink, M. W. and Smol, J. P. 2012. Historical pesticide applications coincided with an altered diet of aerially foraging insectivorous chimney swifts. - Proc. Biol. Sci. 279: 3114–20. Oliver, I. and Beattie, A. J. 1996. Invertebrate morphospecies as surrogates for species: a case study. - Conserv. Biol. 10: 99–109. Orłowski, G. and Karg, J. 2013. Partitioning the effects of livestock farming on the diet of an aerial insectivorous passerine, the barn swallow Hirundo rustica. - Bird Study 60: 111–123. Preston, K. L. and Rotenberry, J. T. 2006. Independent effects of food and predator- mediated processes on annual fecundity in a songbird. - Ecology 87: 160–168. Ratnasingham, S. and Hebert, P. D. N. 2007. BOLD: The Barcode of Life Data System (www.barcodinglife.org). - Mol. Ecol. Notes 7: 355–364. Ratte, H. T. 1984. Temperature and Insect Development. - In: Hoffmann, K. (ed), Environmental Physiology and Biochemistry of Insects. Springer, pp. 33–66. Ringhofer, M. and Hasegawa, T. 2014. Social cues are preferred over resource cues for breeding-site selection in Barn Swallows. - J. Ornithol. 155: 531–538. Ruffino, L., Salo, P., Koivisto, E., Banks, P. B. and Korpimäki, E. 2014. Reproductive responses of birds to experimental food supplementation: a meta-analysis. - Front. Zool. 11: 1-13. Sabo, J. L., Bastow, J. L. and Power, M. E. 2002. Length–mass relationships for adult aquatic and terrestrial invertebrates in a California watershed. - J. North Am. Benthol. Soc. 21: 336–343. Sæther, B.-E. and Bakke, Ø. 2000. Avian life history variation and contribution of demographic traits to the population growth rate. - Ecology 81: 642–653. Safran, R. J. 2004. Adaptive site selection rules and variation in group size of Barn Swallows: individual decisions predict population patterns. - Source Am. Nat. 164: 121–131. Safran, R. J. 2006. Nest-site selection in the Barn Swallow, Hirundo rustica: What predicts seasonal reproductive success? - Can. J. Zool. 84: 1533–1539.

Saino, N., Romano, M., Ambrosini, R., Ferrari, R. P. and Møller, A. P. 2004. Timing of reproduction and egg quality covary with temperature in the insectivorous Barn Swallow, Hirundo rustica. - Funct. Ecol. 18: 50–57. Schoech, S. J., Bridge, E. S., Boughton, R. K., Reynolds, S. J., Atwell, J. W. and Bowman, R. 2008. Food supplementation: A tool to increase reproductive output? A case study in the threatened Florida Scrub-Jay. - Biol. Conserv. 141: 162–173. Shields, W. M. and Crook, J. R. 1987. Barn Swallow coloniality: a net cost for group breeding in the Adirondacks? - Ecology 68: 1373–1386. Teglhøj, P. G. 2017. A comparative study of insect abundance and reproductive success of Barn Swallows (Hirundo rustica) in two urban habitats. - J. Avian Biol. 48: 846–853. Turner, A. K. 1982. Optimal foraging by the swallow (Hirundo rustica, L): Prey size selection. - Anim. Behav. 30: 862–872. Turner, A. 2006. The Barn Swallow. - T & A D Poyser. Twining, C. W., Brenna, J. T., Lawrence, P., Shipley, J. R., Tollefson, T. N. and Winkler, D. W. 2016. Omega-3 long-chain polyunsaturated fatty acids support aerial insectivore performance more than food quantity. – Proc. Natl. Acad. Sci. U.S.A. 113: 10920-10925. van Achterberg, K. 2009. Can Townes type Malaise traps be improved? Some recent developments. - Entomol. Ber. 69: 129–135. Verboven, N. and Verhulst, S. 1996. Seasonal variation in the incidence of double broods: the date hypothesis fits better than the quality hypothesis. - J. Anim. Ecol. 65: 264–273. Visser, M. E., Holleman, L. J. M. and Gienapp, P. 2006. Shifts in caterpillar biomass phenology due to climate change and its impact on the breeding biology of an insectivorous bird. - Oecologia 147: 164–172. Ward, S. and Bryant, D. M. 2006. Barn Swallows Hirundo rustica form eggs mainly from current food intake. - J. Avian Biol. 37: 179–189. Wood, S. N. 2006. Generalized Additive Models: An Introduction with R. - Chapman and Hall/CRC. Zanette, L., Smith, J. N. M., Van Oort, H. and Clinchy, M. 2003. Synergistic effects of food and predators on annual reproductive success in Song Sparrows. - Proc. R. Soc. Biol. Sci. 270: 799–803. Zanette, L., Clinchy, M. and Smith, J. N. M. 2006. Food and predators affect egg production in Song Sparrows. - Ecology 87: 2459–2467.

FIGURES

A B

Figure 3.2 Prey availability (A) and prey demand (B) curves for Barn Swallows at 7 breeding sites plotted as proportional occurrence of the season total by day of the year (day 120 represents April 29th). Each breeding site is represented by a separate curve.

Figure 3.3 Prey availability and demand curves for Barn Swallows plotted together for each breeding site by day of the year (day 120 represents April 29th). The red line represents the demand curve and the black line represents the prey availability curve. Curves are plotted as proportional occurrence of the season totals. Overlap is shaded in red.

Figure 3.4 Predicted values from generalized linear models of (A) the numbers of eggs and clutch initiation date and (B) the number of fledglings and the hatch date. Dates are plotted as day of the year (day 150 represents May 30th). First broods and second broods are plotted separately. 95% confidence intervals are represented in grey.

TABLES

Table 3.1 Summary of the reproductive output and success of Barn Swallows at ten breeding sites in 2015 and 2016.

% Hatch % Nest % # Nests # Eggs # Fledged Success Success Depredated

2015 126 88.1 76.9 19 4.5 ± 0.09 3.3 ± 0.16 Brood 1 80 85 72.5 22.5 4.7 ± 0.13 3.4 ± 0.22 Brood 2 46 93.4 84.7 13 4.4 ± 0.11 3.2 ± 0.23

2016 116 91.3 83.6 12.1 4.5 ± 0.11 3.6 ± 0.17 Brood 1 86 89.5 80.2 16.3 4.6 ± 0.12 3.6 ± 0.21 Brood 2 30 96.7 93.3 0 3.9 ± 0.19 3.4 ± 0.25

Total 242 89.7 80.1 15.7 4.51 ± 0.07 3.4 ± 0.12

Table 3.2 Linear models of the effects of five (2015) and nine (2016) single predictors on the number of pairs nesting at a site. Models with significant predictors are shown in bold (α = 0.05).

Model Estimate SE t-value p-value r2

2015 Nesting Prey Abundance -0.001 0.003 -0.19 0.85 0.01 Nesting Prey Biomass -0.001 0.003 -0.42 0.68 0.02 Nesting Large Prey Abundance -0.004 0.010 -0.41 0.69 0.02 Nesting Large Prey Biomass -0.001 0.003 -0.37 0.72 0.01 Number of Old Nests 0.190 0.060 3.22 0.01 0.56

2016 Arrival Prey Abundance 0.001 0.010 0.09 0.92 0.001 Arrival Prey Biomass 0.002 0.020 0.09 0.93 0.001 Arrival Large Prey Abundance 0.020 0.040 0.46 0.66 0.02 Arrival Large Prey Biomass 0.006 0.025 0.26 0.80 0.01 Nesting Prey Abundance 0.008 0.004 1.86 0.10 0.30 Nesting Prey Biomass 0.006 0.005 1.18 0.27 0.15 Nesting Large Prey Abundance 0.031 0.012 2.47 0.04 0.43 Nesting Large Prey Biomass 0.007 0.006 1.13 0.29 0.13 Number of Old Nests 0.207 0.046 4.47 0.002 0.71

Model Estimate SE t-value p-value

Total Prey Abundance 0.00001 0.0002 0.071 0.94 Total Prey Biomass 0.00006 0.0001 0.42 0.68 Brood 1 Prey Abundance 0.00009 0.0001 0.775 0.45 Brood 1 Prey Biomass 0.00009 0.0001 0.552 0.59 Brood 2 Prey Abundance 0.00001 0.0002 0.074 0.94 Brood 2 Prey Biomass 0.00005 0.0001 0.45 0.66

Total Large Prey Abundance 0.0003 0.0003 0.99 0.34 Total Large Prey Biomass 0.00004 0.00007 0.874 0.4 Brood 1 Large Prey Abundance 0.0001 0.0004 0.266 0.79 Brood 1 Large Prey Biomass 0.0001 0.0002 0.539 0.54 Brood 2 Large Prey Abundance 0.0002 0.0005 0.351 0.73 Brood 2 Large Prey Biomass 0.00006 0.0001 0.514 0.62

Peak Clutch Initiation Date -0.004 0.008 -0.549 0.59

Model Estimate SE z-value p-value

Eggs ~ Prey Abundance 0.0002 0.0005 0.368 0.71 Eggs ~ Prey Biomass -0.00007 0.0006 -0.108 0.91 Eggs ~ Clutch Initiation Date -0.003 0.002 -2.096 0.03

Fledglings ~ Total Prey Abundance 0.00009 0.0002 0.457 0.65 Fledglings ~ Total Prey Biomass 0.00004 0.0002 0.226 0.82 Fledglings ~ Total Prey Abundance Index -0.0001 0.0003 -0.363 0.71 Fledglings ~ Total Prey Biomass Index -0.0004 0.0004 -0.114 0.91 Fledglings ~ Large Prey Abundance 0.0002 0.0007 0.348 0.73 Fledglings ~ Large Prey Biomass 0.00009 0.0002 0.423 0.67 Fledglings ~ Large Prey Abundance Index -0.0003 0.001 -0.241 0.81 Fledglings ~ Large Prey Biomass Index -0.000006 0.0004 -0.015 0.98 Fledglings ~ Hatch Date -0.011 0.001 -5.512 < 0.001

Chapter 4: General Discussion

The Barn Swallow population in North America has experienced a population decline in the past 30 years, the cause of which is unknown (Nebel et al. 2010). In this thesis, I used molecular tools to investigate one of the factors commonly thought to contribute to the population decline: the food supply. I identified the prey items Barn

Swallow nestlings consumed during the breeding season and determined how provisioning parents select their prey. Next, I compared the reproductive performance to the prey availability at different nesting sites throughout the breeding season to determine if the food supply was limiting the annual fecundity of Barn Swallows in Ontario.

Using a DNA barcoding approach to dietary analysis, a broad range of prey items were identified in the diet of Barn Swallows, including 472 different taxa. A number of taxa previously unreported in Barn Swallow diet were also identified. Barn Swallows showed selection for larger prey items and consumed large flies most frequently. This is consistent with what is known of Barn Swallow diet elsewhere in their range (Turner

2006), although prey items in this study were identified to higher taxonomic levels. This taxonomic resolution allowed us to determine the preferred prey items of Barn Swallows in Ontario more precisely. Future work could determine which foraging habitats provide the most of these preferred prey items and, therefore, represent higher quality foraging habitat for Barn Swallows. It would also be useful to examine the quality of these preferred prey items in terms of the energy they provide, the nutritional quality (e.g., lipid content), and the contaminant load of these prey items. Additionally, Barn Swallow nestling diet composition changed between breeding sites, between years, and between

79 broods. Barn Swallows can take advantage of fluctuating prey resources at different times and at different sites depending on what prey is available. Similar flexibility in nestling diet between years, between breeding sites, and over the breeding season has been observed in other insectivorous birds (Bańbura et al. 1994). Examining the quality of different prey items would also be useful in terms of this variation in diet. Since nestling diet varies between sites, there may be differences in the quality of the different prey items being fed to nestlings which could impact the growth and survival of nestlings at different breeding sites. A flexible diet allows Barn Swallows to take advantage of whatever prey resources are available at a site but these resources may not always be high quality prey resources.

Prey availability did not affect the reproductive behaviour of Barn Swallows breeding in Ontario. The timing of reproduction, the number of pairs nesting at a site, the reproductive output, and the proportion of pairs second brooding were not affected by prey availability, as measured here. These results suggest that the reproductive performance of Barn Swallows in Ontario is not limited by food availability during the nesting season and that there are other factors causing the variation in reproductive performance we observed. A similar trend, where food limitation did not affect reproductive performance, has been observed in other bird species and other factors – such as female arrival time, female condition, and predation – were suggested as causing the variation in reproductive success (Nagy and Smith 1997, Marshall et al. 2002). Future work should be done to confirm that food limitation during the breeding season is not affecting Barn Swallows after fledging or causing carry-over effects during migration and wintering. Additionally, since we saw relatively high productivity, it would be beneficial

80 to examine the full annual cycle to determine how the migration and wintering stages contribute to the population dynamics.

The reproductive performance of Barn Swallows was not limited by food availability. Barn Swallows are generalist consumers and they can respond to changing prey availability, which should make them more resilient to changing insect populations.

These results taken together suggest that Barn Swallow populations are not currently being affected by low prey availability on the breeding grounds. Food may be limiting the Barn Swallow population during other parts of their annual cycle or there could be other factors affecting the population on the breeding and/or wintering grounds (Nebel et al. 2010, NABCI Canada 2012). Work on other aerial insectivore species suggests that diet and food supply are major factors in their decline (Finity 2011). The decline of the aerial insectivore guild in North America is likely due to a combination of multiple, interacting factors (Michel et al. 2016) and individual species may have a different combination of factors driving their population declines. Low juvenile recruitment and/or low adult survival may be contributing factors in the decline in Barn Swallows. Low survival or recruitment could be due to low post-fledging survival of juveniles or high mortality during migration and/or wintering. Mortality on the wintering grounds could be driven by the food supply, where prey quality or quantity could be negatively affected by habitat loss or pesticide use. Alternatively, habitat loss and contamination could directly impact bird mortality.

In my dietary analysis, many of the prey items detected in the diet did not match specimens collected in Malaise traps, indicated by 47% of the OTUs from fecal samples

81 not matching a reference sequence in our custom database. Our custom reference database was limited because it did not include all insect specimens we collected in

Malaise traps. DNA extraction and sequencing was unsuccessful for 12.6% of the insect specimens submitted for sequencing. Due to cost limitations we could not re-sequence these specimens and therefore, they were not represented in our custom database. In addition, Malaise trap sampling, as with all passive insect sampling methods, is very dependent on trap placement (Matthews and Matthews 1971). Each trap can only be placed in one spot at a given time and therefore may poorly reflect heterogeneity in the habitat surrounding breeding sites. Even traps placed in the same habitat a few metres apart can collect different insect communities (Matthews and Matthews 1971) and simply having a single trap in a single location is unable to capture all prey diversity at a breeding site. There is some evidence for trapping bias with Malaise traps. Small

Hymenoptera are underrepresented (Darling and Packer 1988) as well as odonates and butterflies (Owen 1976, De Almieda et al. 2013). None of these insect groups made up a large portion of Barn Swallow diet in my study so this trapping bias has minimal impact on my conclusions. No bias against large flies, such as those frequently consumed by

Barn Swallows, has been reported in the literature (e.g., Roberts 1972). To account for trap placement bias and any potential taxonomic bias, future studies should use multiple traps and potentially use multiple trapping methods at each breeding site. In this study, it was not logistically feasible to operate multiple traps and types at each breeding site.

The integrated DNA barcoding approach used in this study, where a custom reference database was built from insect sampling in the habitat, is not commonly used to concurrently monitor a prey resource and the diet of predators. This approach has many

82 benefits including that it allowed us to make direct links between items in the diet and in the habitat. By making these links we knew the size of the items consumed and their abundance in the habitat. Furthermore, this method required less time and taxonomic expertise to identify the abundant and diverse insect specimens from the habitat. A downside to using this method is that DNA sequencing is expensive (approximately $23

CAD/specimen), particularly for such an abundant prey resource with many individuals to identify. Despite the higher cost, it gave us a better understanding of Barn Swallow diet and improved on older methods by identifying more unique prey items at higher taxonomic resolution. If it is financially viable, I would recommend this method to others wanting to monitor the diet and prey availability of generalist insectivore predators.

This study included a large sample size for the number of nests and the number of fecal samples included in this study, however, I was limited in the number of breeding sites I monitored. The results comparing colony size and rates of second brooding with prey availability could be improved by including more sites and possibly sites from a wider geographic area. Additionally, the study period lasted two breeding seasons. This is a limited time frame when investigating a potential cause of long-term population declines. I did see considerable variation in prey availability between years and a longer study would have allowed some of this yearly variation to be accounted for. This change in prey availability was associated with a change in diet. Long term data on insect populations would greatly improve our understanding of how the food supply affects the diet of Barn Swallows and how it might relate to population declines.

The results presented here provide a knowledge base for future work on the protection and management of Barn Swallow populations. Having identified the preferred prey items Barn Swallows consume, researchers can now identify the foraging habitats that provide the most preferred prey and can investigate the contaminant loads of these preferred prey items to ensure a stable food supply for Barn Swallows. Future work can build on the results shown here by incorporating prey quality and post-fledging survival of chicks. Without historical insect data and diet data, we cannot confirm how the diet and food supply may have contributed to the population decline in the past but the results presented here suggest that there is currently no effect of food availability on reproduction. Other potential drivers of population decline on the breeding grounds and on the wintering grounds should also be investigated.

This thesis provides new information on the biology of Barn Swallows breeding in Ontario and fills gaps in our knowledge as identified in Heagy et al. (2014). An integrated DNA barcoding method for dietary analysis was used that can be applied to other aerial insectivore species to better understand their diet composition and prey selection. No evidence was found to show that the food supply was limiting Barn

Swallow reproduction in my study area of central Ontario, Canada during the study period. The cause of the Barn Swallow population decline in North America is still unknown but, this thesis provides information than can be used for the protection and management of Barn Swallows in Ontario.

REFERENCES

Bańbura, J., Blondel, J., de Wilde-Lambrechts, H., Galan, M.-J. and Maistre, M. 1994. Nestling diet variation in an insular Mediterranean population of Blue Tits Parus caeruleus: effects of years, territories and individuals. - Oecologia 100: 413–420. Darling, C. D. and Packer, L. 1988. Effectiveness of Malaise traps in collecting Hymenoptera: the influence of trap design, mesh size and location. - Can. Ent. 120: 787-796. De Almieda, M. V. O., Pinto, A. P., Carvalho, A. L. and Takiya, D. M. 2013. When rare is just a matter of sampling: unexpected dominance of clubtail dragonflies (Odonata, Gomphidae) through different collecting methods at Parque Nacional da Serra do Cipo, Minas Girais State, Brazil. - Rev. Bras. Entomol. 57: 417-423. Finity, L. K. 2011. The role of habitat and dietary factors in Chimney Swift (Chaetura pelagica) population declines. MSc Thesis. Heagy, A., Badzinski, D., Bradley, D., Falconer, M., McCraken, J., Reid, R. A. and Richardson, K. 2014. Recovery strategy for the Barn Swallow (Hirundo rustica) in Ontario. Ontario Recovery Strategy Series. – Ontario Ministry of Natural Resources and Forestry. Marshall, M. R., Cooper, R. J., Dececco, J. A., Strazanac, J. and Butler, L. 2002. Effects of experimentally reduced prey abundance on the breeding ecology of the Red- Eyed Vireo. - Ecol. Appl. 12: 261–280. Matthews, R W. and Matthews, J. R. 1971. The Malaise trap: its utility and potential for sampling insect populations. - Great Lakes Entomol. 4: 117-122. Michel, N. L., Smith, A. C., Clark, R. G., Morrissey, C. A. and Hobson, K. A. 2016. Differences in spatial synchrony and interspecific concordance inform guild-level population trends for aerial insectivorous birds. - Ecography 39: 774–786. NABCI Canada. 2012. The State of Canada’s Birds. - Environment Canada. Nagy, L. R. and Smith, K. G. 1997. Effects of insecticide-induced reduction in lepidopteran larvae on reproductive success of Hooded Warblers. - Auk 114: 619– 627. Nebel, S., Mills, A., McCracken, J. D. and Taylor, P. D. 2010. Declines of aerial insectivores in North America follow a geographic gradient. - Avian Conserv. Ecol. 5: 1. Owen, D. F. 1976. Conservation of butterflies in garden habitats. - Env. Cons. 3: 285- 290.

Roberts, R. H. 1972. The effectiveness of several types of Malaise traps for the collection of Tabanidae and Culicidae. - Mosq. News 32: 542-547. Turner, A. 2006. The Barn Swallow. - T & A D Poyser.

Appendix A

r2 = 0.38 p < 0.0001

2 r = 0.05 p = 0.0002

r2 = 0.03 p = 0.004

Figure A.1 Correlations between the number of raw reads and the number of OTUs, the number of reference sequences matched and the number of families identified in each fecal sample.

Table A.1 Summary of the sequencing success for each sample showing the number of raw reads produced, the number of reads after filtering for quality and length, the number of OTUs in each samples and the number of OTUs that matched a reference sequence.

Sample # Raw Reads # Filtered Reads # OTUs # OTUs Matched 2015_2_1_C2 18709 678 3 0 2015_2_1_D1 46733 41387 86 84 2015_2_1_E1 140543 132843 161 151 2015_2_2_F1 41601 12515 20 20 2015_2_2_F2 9014 8892 9 9 2015_2_2_G1 49960 29885 27 26 2015_2_3_E2 146015 136006 137 127 2015_2_3_F1 52632 50310 113 111 2015_2_3_F2 19230 571 2 2 2015_2_5_J2 36002 35661 57 55 2015_2_5_K1 108397 85805 61 55 2015_2_5_K2 45523 5634 32 32 2015_20_16_M1 33554 31218 155 140 2015_20_2_E1 39228 37734 86 86 2015_20_3_E2 101251 67638 119 104 2015_20_3_F1 24784 22751 71 66 2015_20_4_F1 30303 25926 53 50 2015_20_4_F2 7630 699 8 7 2015_20_4_M1 9022 7622 60 52 2015_20_4_M2 86119 48041 77 73 2015_20_5_F1 57665 53229 177 168 2015_20_5_H2 16698 15384 96 90 2015_24_1_D1 88293 76647 184 167 2015_24_10_K2 37259 34162 113 110 2015_24_10_L1 17924 7723 5 5 2015_24_11_M1 15180 14766 55 48 2015_24_11_M2 44268 3799 17 17 2015_24_12_M2 26700 26074 84 70 2015_24_3_F1 21951 21014 65 65 2015_24_7_F2 31808 29688 82 76 2015_24_7_G1 137228 128759 220 210 2015_24_8_G1 65738 58478 154 142 2015_24_9_K1 28549 27635 77 76 2015_24_9_K2 55810 32704 53 49 2015_26_1_D1 40995 35858 115 113 2015_26_1_J1 40823 40118 61 57

2015_26_10_E1 39724 35738 115 112 2015_26_10_L2 25349 24420 53 51 2015_26_11_E2 37375 35448 92 90 2015_26_13_F1 21445 19805 34 34 2015_26_14_E2 58624 57264 116 110 2015_26_15_J1 24370 23637 58 55 2015_26_16_F2 35274 31965 138 128 2015_26_17_H2 40788 38376 89 88 2015_26_18_I7 45324 43018 78 77 2015_26_18_L2 34633 33391 111 110 2015_26_19_F2 30596 30018 49 48 2015_26_2_D1 16573 13364 89 85 2015_26_20_F2 70279 58730 224 210 2015_26_22_K2 16285 15545 84 81 2015_26_23_G1 37684 34404 87 75 2015_26_24_G1 40870 39496 93 90 2015_26_26_H1 48282 46882 55 48 2015_26_27_J2 70046 67899 142 134 2015_26_28_K2 29288 28081 57 56 2015_26_29_L1 69291 67094 123 122 2015_26_3_D1 33063 29386 92 92 2015_26_30_L2 44196 41047 153 142 2015_26_31_M1 17814 15925 80 64 2015_26_32_M2 27738 26401 77 77 2015_26_4_D1 32219 29596 120 113 2015_26_5_D3 39543 38447 74 71 2015_26_6_D2 51175 45179 163 158 2015_26_7_E1 29042 27417 91 91 2015_26_9_E1 21336 18732 56 54 2015_26_9_K2 23131 21327 103 98 2015_34_1_D3 11535 7814 50 48 2015_34_1_E1 68546 55023 152 145 2015_34_1_L2 60783 59458 115 112 2015_34_1_M1 35795 25305 80 72 2015_34_2_D2 62347 57134 205 196 2015_34_2_F1 44372 22888 45 43 2015_34_3_F2 84489 78465 183 174 2015_34_4_G1 34715 33252 55 51 2015_34_5_M1 41600 39716 131 126 2015_34_6_M2 35793 34155 83 76 2015_34_7_M3 25805 23982 139 122 2015_41_1_E1 8951 970 5 5

2015_41_1_E2 124569 98219 215 202 2015_41_1_F1 60650 58143 83 80 2015_41_1_F2 122846 104161 130 126 2015_41_2_F1 16189 445 3 3 2015_41_2_F2 54747 51784 118 115 2015_41_2_G1 34793 22586 13 13 2015_41_2_G2 160073 156463 169 166 2015_42_1_D1 22657 19187 69 62 2015_42_1_J2 33761 32543 33 30 2015_42_11_L2 59384 56981 161 150 2015_42_2_D2 36856 35531 109 105 2015_42_3_D1 42287 40150 112 109 2015_42_3_K1 26188 24406 94 90 2015_42_4_D2 45545 42482 127 115 2015_42_4_L1 29388 28454 51 47 2015_42_5_E2 30767 29342 88 86 2015_42_6_F1 44124 39824 127 125 2015_42_7_F1 36014 33522 104 102 2015_42_7_L1 38517 37252 109 105 2015_42_8_L1 24209 23431 51 49 2015_42_9_G1 31528 30164 66 64 2015_43_1_D2 65721 54673 140 129 2015_43_13_K2 11937 11403 44 40 2015_43_14_K1 27238 26160 69 64 2015_43_14_K2 4419 1886 11 8 2015_43_16_K2 50234 37517 86 82 2015_43_17_K2 26500 25070 60 59 2015_43_2_D1 23 4 0 0 2015_43_20_L2 74891 64315 291 270 2015_43_4_E2 41695 40375 83 78 2015_43_5_F2 36807 36146 80 79 2015_43_6_G2 60703 57428 105 96 2015_43_7_G2 43319 40065 107 98 2015_77_1_F1 32549 25889 23 22 2015_77_1_F2 43596 42475 81 80 2015_77_1_G1 93520 86042 99 88 2015_77_2_E1 85739 62136 88 87 2015_77_2_E2 20187 7855 11 11 2015_77_2_F1 31980 30207 63 60 2015_77_2_F2 83847 52258 92 90 2015_77_2_G1 37003 12529 5 5 2015_9_1_E2 34410 33484 46 46

2015_9_2_E2 48542 41774 153 145 2015_9_2_F1 76217 30451 67 62 2015_9_4_F1 47075 31506 48 46 2015_9_4_G1 27128 26138 46 43 2015_9_6_G2 21882 19391 86 83 2015_9_6_H1 18883 4 0 0 2015_9_7_J2 28706 28140 65 64 2015_9_7_L2 32934 31780 58 57 2015_9_8_L2 41757 38974 107 102 2015_9_8_M1 85430 47105 48 44 2015_9_9_M1 118551 100730 58 39 2016_2_1_P1 53182 49244 119 102 2016_2_1_P2 13679 2 0 0 2016_2_3_P1 13118 409 2 2 2016_2_3_P2 56085 41319 41 38 2016_2_3_P3 20071 6204 13 12 2016_2_4_I1 71636 69933 99 87 2016_2_4_J1 163776 162611 120 102 2016_2_4_J2 114275 111213 103 88 2016_2_6_M1 27894 21500 46 42 2016_2_6_M2 47873 38112 63 49 2016_2_6_M3 93858 82271 233 181 2016_2_6_N1 13564 94 1 1 2016_20_11_M1 52398 47205 70 69 2016_20_11_M2 48438 34405 60 56 2016_20_13_N1 157363 142997 153 146 2016_20_13_O1 39090 19870 75 73 2016_20_13_O2 47324 33915 59 57 2016_20_14_P1 30039 28900 90 78 2016_20_16_M2 42218 3457 9 9 2016_20_2_I2 66094 60423 158 147 2016_20_2_I3 14593 252 2 2 2016_20_5_J1 44479 43243 67 64 2016_20_5_J2 47378 43650 54 54 2016_20_8_L2 35605 22875 59 52 2016_20_8_L3 40739 21160 42 42 2016_24_1_H1 30381 27247 53 52 2016_24_11_N1 64877 62229 87 42 2016_24_12_O2 5342 2 2 2 2016_24_12_O3 10236 1564 0 0 2016_24_2_I1 58622 55014 71 68 2016_24_3_I2 55568 52097 56 52

2016_24_4_I2 68726 59130 122 111 2016_24_4_I3 9760 2811 6 6 2016_24_5_I2 18338 17043 44 41 2016_24_6_I2 56653 55104 105 96 2016_24_7_J1 69784 68734 84 76 2016_24_8_K1 64168 60051 180 165 2016_24_9_K1 24480 16719 61 58 2016_26_1_G1 41996 41718 56 56 2016_26_1_G2 55527 47406 62 60 2016_26_1_H1 19492 139 2 2 2016_26_1_H2 36231 31485 63 58 2016_26_10_I2 53930 49170 121 104 2016_26_11_I1 40839 27396 42 39 2016_26_13_I1 263 2 0 0 2016_26_13_I2 52598 33639 36 36 2016_26_13_J1 46756 42331 98 90 2016_26_13_J2 83487 79647 60 55 2016_26_15_I1 17881 17751 39 39 2016_26_16_J1 59558 52689 135 111 2016_26_17_I1 42574 35108 88 82 2016_26_17_J1 11697 1401 1 1 2016_26_17_J2 25084 24451 19 19 2016_26_17_J3 22716 5850 22 20 2016_26_18_J1 43505 40594 54 50 2016_26_19_J2 68413 60643 125 112 2016_26_2_G2 67725 59819 91 82 2016_26_2_I1 55350 54970 22 22 2016_26_20_J1 50989 33164 111 103 2016_26_20_J2 63234 61954 39 35 2016_26_20_K1 22333 9036 27 27 2016_26_21_J1 44252 41333 118 107 2016_26_22_K1 5004 323 5 5 2016_26_23_J1 62690 56054 74 68 2016_26_23_J2 47118 41911 54 52 2016_26_23_K1 5 0 0 0 2016_26_23_K2 29727 19283 38 32 2016_26_24_N2 70534 64067 162 136 2016_26_25_N1 84326 76639 189 165 2016_26_26_N1 28331 6615 9 4 2016_26_26_N2 61951 42380 127 114 2016_26_26_O1 64778 62073 94 90 2016_26_26_O2 63316 58898 73 69

2016_26_27_O1 70055 64147 136 125 2016_26_28_P1 59709 53900 156 126 2016_26_29_P1 51878 45710 97 80 2016_26_3_H1 84172 80599 116 112 2016_26_30_J3 55919 46016 55 45 2016_26_30_P1 48213 26624 37 18 2016_26_31_P1 61588 59234 67 60 2016_26_4_G1 8997 55 1 1 2016_26_4_H1 23762 22493 36 34 2016_26_4_H2 16846 13431 27 26 2016_26_4_H3 27255 3711 15 15 2016_26_5_I2 33928 26238 62 58 2016_26_6_I2 3759 6 0 0 2016_26_7_I2 56073 49308 109 91 2016_26_8_I1 72083 66402 141 129 2016_26_8_I2 63999 59126 114 98 2016_26_8_J1 67032 61644 95 78 2016_26_8_J2 2339 940 11 11 2016_26_9_I2 69327 64890 113 103 2016_34_1_J1 25787 19858 31 26 2016_34_1_J2 9153 309 3 3 2016_34_1_K1 37581 4546 9 9 2016_34_2_J1 33736 15594 12 11 2016_34_2_J2 43968 40202 65 60 2016_34_2_J3 20987 340 1 1 2016_34_2_K1 11141 1234 6 6 2016_34_5_M2 24142 14736 66 62 2016_41_1_J1 74302 69137 123 120 2016_41_1_J2 21610 10069 13 13 2016_41_1_J3 66903 42939 62 59 2016_42_1_H1 29393 2007 4 4 2016_42_1_N1 6788 785 5 5 2016_42_1_N2 27993 12351 19 18 2016_42_2_H1 71173 67553 72 60 2016_42_3_I2 73444 71356 102 94 2016_42_4_I1 69020 61116 59 58 2016_42_4_I2 80201 67924 34 30 2016_42_4_P2 36516 24420 46 41 2016_42_5_I2 78135 73612 93 85 2016_42_5_J1 21352 10833 13 12 2016_42_6_I1 78272 75269 136 127 2016_42_7_J2 73048 67186 166 148

2016_42_8_J2 18565 445 6 6 2016_43_10_J2 43665 35143 48 45 2016_43_11_J1 34381 8044 4 4 2016_43_12_L2 72027 63724 55 50 2016_43_13_L2 11185 46 1 1 2016_43_14_O1 79289 77182 78 74 2016_43_14_O3 41471 31105 84 75 2016_43_15_O1 35508 20963 33 28 2016_43_16_O2 26870 1900 4 4 2016_43_18_R1 61801 32906 44 41 2016_43_2_M1 12597 4729 5 5 2016_43_3_I3 0 0 0 0 2016_43_4_I2 33019 24075 24 24 2016_43_5_I1 19401 983 2 2 2016_43_6_I2 5841 512 3 3 2016_43_7_I2 13086 1256 5 4 2016_43_8_I2 6150 1991 4 4 2016_43_9_J2 12681 5515 2 2 2016_77_2_L1 32488 0 0 0 2016_77_2_L2 13463 0 0 0 2016_77_2_M1 22660 2371 4 4 2016_77_2_M2 44924 4252 5 5 2016_9_1_G1 52625 37142 95 94 2016_9_10_P1 53339 50293 162 144 2016_9_11_P1 92221 88720 53 49 2016_9_2_I1 68018 63508 34 33 2016_9_3_I1 34665 23484 46 44 2016_9_5_K1 59181 52415 97 96 2016_9_5_K2 25916 13294 21 20 2016_9_6_K1 14584 3900 7 7 2016_9_7_L2 37352 33987 32 31 2016_9_8_M1 49837 45678 144 133 2016_9_8_M3 28026 9074 11 10 2016_9_9_M2 35016 33409 86 79 2016_9_9_N1 13657 333 2 2

Table A.2 Species list of arthropod prey items from Barn Swallow nestling fecal samples, identified using a custom reference database and Canadian COI records from the BOLD database, with a minimum of 99.3% similarity for species identifications and a minimum of 95% similarity for genus level identifications. The number of samples the prey item was detected in is presented as well as the frequency of occurrence of that prey item. Prey items with a frequency of occurrence of 5% or higher are indicated in bold. Pest species are indicated by a letter corresponding to what they are pests of: stored product – SP, human – H, livestock – L, crop – C, bird – B, and forest – F.

Number of Frequency of Class Order Family Genus Species Pest Samples Occurrence Arachnida Trombidiformes Arrenuridae Arrenurus sp. 1 0.004 - Unionicolidae sp. 1 0.004 - Neumania sp. 1 0.004 - Diplopoda Julida Julidae Julus scandinavius 1 0.004 - Insecta Coleoptera Anthicidae Omonadus floralis 1 0.004 SP Anthribidae Anthribus nebulosus 1 0.004 - Cantharidae Rhagonycha recta 1 0.004 - Carabidae Acupalpus sp. 2 0.007 - Agonum sp. 2 0.007 - Amara sp. 11 0.041 - Anisodactylus sp. 1 0.004 - Bembidion mimus 2 0.007 - versicolour 1 0.004 - Bradycellus sp. 1 0.004 - Chlaenius lithophilus 1 0.004 - Lebia sp. 1 0.004 - Loricera pilicornis 1 0.004 - Trechus quadristriatus 1 0.004 - Cerambycidae Tetrops praeusta 1 0.004 -

Insecta Coleoptera Chrysomelidae sp. 3 0.011 - Oulema melanopus 1 0.004 C Systena blanda 1 0.004 - Cleridae sp. 1 0.004 - Isohydnocera tabida 1 0.004 - Coccinellidae Harmonia axyridis 1 0.004 - Cryptophagidae sp. 1 0.004 - Atomaria fuscata 1 0.004 - Curculionidae sp. 2 0.007 - Dryocoetes autographus 1 0.004 - Hypera nigrirostris 1 0.004 - postica 9 0.033 C Phyllobius oblongus 5 0.018 - Polydrusus formosus 4 0.015 - Rhinoncus bruchoides 1 0.004 - castor 1 0.004 - Sitona sp. 12 0.044 C Dermestidae sp. 1 0.004 - Dermestes undulatus 1 0.004 - sp. 2 0.007 - Elateridae Ampedus miniipennis 1 0.004 - Hydrophilidae sp. 1 0.004 - Cercyon haemorrhoidalis 2 0.007 - lateralis 3 0.011 - Cryptopleurum minutum 1 0.004 - Cymbiodyta sp. 1 0.004 - Enochrus ochraceus 1 0.004 - Enochrus sp. 1 0.004 - Helophorus nitiduloides 1 0.004 -

Insecta Coleoptera Latridiidae Corticarina sp. 1 0.004 - Nitidulidae sp. 1 0.004 - Epuraea sp. 1 0.004 - Phalacridae Acylomus sp. 1 0.004 - Stilbus apicialis 1 0.004 - Ptinidae sp. 1 0.004 - Hemicoelus carinatus 5 0.018 - Ripiphoridae Ripiphorus fasciatus 3 0.011 - Scarabaeidae sp. 1 0.004 - Onthophagus nuchicornis 1 0.004 - Scirtidae sp. 5 0.018 - Cyphon obscurus 1 0.004 - Scirtes tibialis 1 0.004 - Staphylinidae Aleochara bimaculata 1 0.004 - Oxypoda opaca 1 0.004 - Diptera sp. 16 0.059 - Agromyzidae Ophiomyia sp. 2 0.007 - Pseudonapomyza sp. 1 0.004 - Anisopodidae Sylvicola alternatus 1 0.004 - sp. 1 0.004 - Anthomyiidae sp. 10 0.037 - Anthomyia sp. 2 0.007 - Botanophila sp. 5 0.018 - Delia florilega 21 0.077 C sp. 7 0.026 C Egle sp. 1 0.004 - Eustalomyia sp. 5 0.018 - Fucellia sp. 1 0.004 - Hydrophoria lancifer 5 0.018 -

Insecta Diptera Anthomyiidae Hylemya sp. 2 0.007 - Hylemyza sp. 1 0.004 - Lasiomma sp. 3 0.011 - Paradelia sp. 1 0.004 - Pegomya flavifrons 1 0.004 - sp. 1 0.004 - Zaphne implicata 1 0.004 - sp. 1 0.004 - Asilidae sp. 2 0.007 - Dioctria sp. 12 0.044 - Holcocephala abdominalis 1 0.004 - calva 1 0.004 - Leptogaster flavipes 1 0.004 - Bibionidae Dilophus sp. 2 0.007 - Bolitophilidae Bolitophila sp. 1 0.004 - Calliphoridae Pollenia pediculata 20 0.074 - griseotomentosa 6 0.022 - sp. 5 0.018 - Calliphora sp. 2 0.007 - Cynomya sp. 1 0.004 - Lucilia sp. 3 0.011 - Melanomya sp. 1 0.004 - Phormia regina 4 0.015 - Protocalliphora sp. 1 0.004 B Cecidomyiidae sp. 2 0.007 - Ceratopogonidae Clinohelea sp. 1 0.004 - Dasyhelea sp. 1 0.004 - Chaoboridae Chaoborus punctipennis 1 0.004 - sp. 1 0.004 -

Insecta Diptera Chironomidae sp. 20 0.074 - Chironomus acidophilus 2 0.007 - sp. 4 0.015 - Conchapelopia sp. 1 0.004 - Cryptochironomus sp. 2 0.007 - Microtendipes sp. 2 0.007 - Parachironomus sp. 1 0.004 - Paracladopelma winnelli 1 0.004 - Phaenopsectra punctipes 1 0.004 - Tanytarsus sp. 1 0.004 - Chloropidae Chaetochlorops sp. 1 0.004 - Conioscinella sp. 1 0.004 - Eribolus sp. 1 0.004 - Incertella sp. 1 0.004 - Rhopalopterum sp. 1 0.004 - Siphonella sp. 1 0.004 - Thaumatomyia sp. 2 0.007 - Culicidae sp. 3 0.011 H/L/B Aedes excrucians 1 0.004 H/L/B japonicus 1 0.004 H/L/B provocans 7 0.026 H/L/B stimulans 1 0.004 H/L/B vexans 2 0.007 H/L/B sp. 7 0.026 H/L/B Coquillettidia perturbans 6 0.022 H/L/B sp. 3 0.011 H/L/B Culex sp. 1 0.004 H/L/B Culiseta minnesotae 2 0.007 H/L/B morsitans 3 0.011 H/L/B

Insecta Diptera Diastatidae Diastata sp. 1 0.004 - Dolichopodidae sp. 2 0.007 - Dolichopus comatus 1 0.004 - discifer 1 0.004 - sp. 3 0.011 - Drosophilidae sp. 2 0.007 - Chymomyza sp. 3 0.011 - Drosophila affinis 5 0.018 - algonquin 5 0.018 - falleni 1 0.004 - Scaptomyza adusta 1 0.004 - Stegana sp. 1 0.004 - Empididae sp. 1 0.004 - Rhamphomyia longicauda 5 0.018 - sp. 3 0.011 - Ephydridae sp. 1 0.004 - Calocoenia sp. 1 0.004 - Discocerina sp. 2 0.007 - Lytogaster sp. 1 0.004 - Notiphila pallidipalpis 1 0.004 - pulchra 1 0.004 - Ochthera sp. 4 0.015 - Paracoenia sp. 1 0.004 - Parydra sp. 1 0.004 - Fanniidae sp. 1 0.004 - Fannia armata 1 0.004 - sp. 8 0.030 - Hybotidae sp. 2 0.007 - Crossopalpus sp. 1 0.004 -

100

Insecta Diptera Hybotidae Euhybus triplex 2 0.007 - Platypalpus holosericus 1 0.004 - stabilis 1 0.004 - sp. 4 0.015 - Syndyas dorsalis 1 0.004 - Syneches simplex 3 0.011 - Keroplatidae sp. 1 0.004 - Lauxaniidae Homoneura sp. 1 0.004 - sp. 1 0.004 - Limoniidae sp. 51 0.188 - Antocha saxicola 1 0.004 - Dicranomyia frontalis 1 0.004 - longipennis 1 0.004 - sp. 25 0.092 - Discobola sp. 1 0.004 - Erioptera caliptera 1 0.004 - chlorophylla 3 0.011 - septemtrionis 2 0.007 - sp. 1 0.004 - Euphylidorea luteola 1 0.004 - platyphallus 2 0.007 - Geranomyia sp. 4 0.015 - Helius flavipes 3 0.011 - sp. 1 0.004 - Limnophila sp. 1 0.004 - Limonia sp. 1 0.004 - Metalimnobia immatura 1 0.004 - solitaria 1 0.004 - sp. 1 0.004 -

101

Insecta Diptera Limoniidae Neolimnophila sp. 1 0.004 - Ormosia affinis 9 0.033 - Pilaria tenuipes 2 0.007 - Pseudolimnophila inornata 14 0.052 - luteipennis 5 0.018 - Rhipidia sp. 2 0.007 - Symplecta sp. 2 0.007 - Lonchaeidae Lonchaea sp. 1 0.004 - Lonchopteridae Lonchoptera bifurcata 1 0.004 - Milichiidae sp. 1 0.004 - Muscidae Coenosia tigrina 4 0.015 - sp. 1 0.004 - Dasyphora sp. 2 0.007 - Graphomya sp. 1 0.004 - Gymnodia delecta 1 0.004 - Haematobia sp. 1 0.004 L Hebecnema nigra 1 0.004 - Helina evecta 5 0.018 - reversio 1 0.004 - setiventris 1 0.004 - sp. 11 0.041 - Hydrotaea sp. 3 0.011 - Macrorchis ausoba 1 0.004 - Morellia sp. 1 0.004 - Musca sp. 4 0.015 - Muscina levida 2 0.007 - pascuorum 1 0.004 - stabulans 2 0.007 - sp. 2 0.007 -

102

Insecta Diptera Muscidae Mydaea detrita 1 0.004 - flavicornis 1 0.004 - Myospila meditabunda 3 0.011 - Neodexiopsis calopyga 1 0.004 - Neomyia sp. 1 0.004 - Pentacricia aldrichii 1 0.004 - Phaonia tuguriorum 1 0.004 - sp. 2 0.007 - Schoenomyza sp. 1 0.004 - Spilogona sp. 1 0.004 - Stomoxys sp. 4 0.015 L Mycetophilidae sp. 3 0.011 - Exechia sp. 1 0.004 - Mycetophila alea 1 0.004 - fungorum 1 0.004 - sp. 1 0.004 - Opomyzidae Geomyza tripunctata 3 0.011 - Phoridae Conicera sp. 3 0.011 - Diplonevra sp. 1 0.004 - Megaselia sp. 3 0.011 - Pipunculidae Pipunculus hertzogi 1 0.004 - torus 1 0.004 - sp. 3 0.011 - Platypezidae sp. 1 0.004 - Platystomatidae Rivellia steyskali 1 0.004 - Psychodidae sp. 1 0.004 - Psychoda sp. 2 0.007 - Ptychopteridae Bittacomorpha clavipes 1 0.004 - Rhagionidae Chrysopilus proximus 1 0.004 -

103

Insecta Diptera Rhagionidae Chrysopilus sp. 2 0.007 - Rhagio lineola 1 0.004 - sp. 2 0.007 - Rhinophoridae sp. 1 0.004 - Sarcophagidae sp. 5 0.018 - Blaesoxipha sp. 1 0.004 - Boettcheria sp. 2 0.007 - Ravinia querula 5 0.018 - stimulans 3 0.011 - sp. 5 0.018 - Sarcophaga subvicina 11 0.041 - sp. 2 0.007 - Sarcotachinella sp. 6 0.022 - Scathophagidae Scathophaga stercoraria 3 0.011 - Sciomyzidae Renocera brevis 1 0.004 - Sepedon fuscipennis 1 0.004 - sp. 1 0.004 - Sepsidae Sepsis punctum 5 0.018 - sp. 1 0.004 - Simuliidae Simulium sp. 1 0.004 H/L Sphaeroceridae Coproica acutangula 1 0.004 - ferruginata 1 0.004 - hirtula 1 0.004 - Copromyza sp. 2 0.007 - Lotophila sp. 2 0.007 - Spelobia bifrons 1 0.004 - sp. 4 0.015 - Sphaerocera curvipes 1 0.004 - Stratiomyidae Actina sp. 1 0.004 -

104

Insecta Diptera Stratiomyidae Allognosta fuscitarsis 3 0.011 - Odontomyia cincta 1 0.004 - sp. 1 0.004 - Syrphidae Cheilosia sp. 1 0.004 - Epistrophe sp. 1 0.004 - Eristalis sp. 1 0.004 - Eupeodes sp. 2 0.007 - Lejops lineatus 1 0.004 - Melanostoma sp. 1 0.004 - Orthonevra anniae 1 0.004 - Parhelophilus sp. 1 0.004 - Platycheirus hyperboreus 1 0.004 - sp. 1 0.004 - Rhingia nasica 1 0.004 - Sphaerophoria sp. 2 0.007 - Syritta pipiens 1 0.004 - Syrphus sp. 1 0.004 - Tropidia quadrata 1 0.004 - Xylota sp. 1 0.004 - Tabanidae sp. 4 0.015 H/L Haematopota sp. 1 0.004 H/L Hybomitra illota 2 0.007 H/L sp. 11 0.041 H/L Tabanus novaescotiae 11 0.041 H/L similis 3 0.011 H/L sp. 16 0.059 H/L Tachinidae sp. 27 0.100 - Actia interrupta 1 0.004 - Allophorocera arator 1 0.004 -

105

Insecta Diptera Tachinidae Allophorocera sp. 4 0.015 - Archytas sp. 1 0.004 - Belvosia sp. 23 0.085 - Billaea sibleyi 1 0.004 - Carcelia amplexa 2 0.007 - sp. 1 0.004 - Chaetogaedia sp. 2 0.007 - Cryptomeigenia sp. 4 0.015 - Dinera grisescens 1 0.004 - Drino galii 7 0.026 - Erynnia sp. 1 0.004 - Euexorista rebaptizata 7 0.026 - Eulasiona sp. 4 0.015 - Eumea sp. 5 0.018 - Euthelyconychia sp. 1 0.004 - Exorista larvarum 2 0.007 - sp. 1 0.004 - Gymnoclytia sp. 2 0.007 - Gymnosoma nudifrons 1 0.004 - sp. 1 0.004 - Houghia coccidella 1 0.004 - Hypovoria cauta 1 0.004 - Leschenaultia exul 4 0.015 - fulvipes 11 0.041 - sp. 3 0.011 - Lespesia archippivora 1 0.004 - datanarum 1 0.004 - sp. 1 0.004 - Linnaemya sp. 2 0.007 -

106

Insecta Diptera Tachinidae Myxexoristops bonsdorffi 3 0.011 - Panzeria sp. 3 0.011 - Patelloa sp. 4 0.015 - Peleteria sp. 1 0.004 - Phorocera sp. 1 0.004 - Siphosturmia sp. 1 0.004 - Spallanzania hesperidarum 2 0.007 - Uramya sp. 1 0.004 - Winthemia rufopicta 6 0.022 - sp. 6 0.022 - Zaira sp. 1 0.004 - Tephritidae Procecidochares atra 1 0.004 - sp. 1 0.004 - Rhagoletis sp. 1 0.004 C Strauzia sp. 1 0.004 C Urophora sp. 2 0.007 - Tipulidae sp. 5 0.018 - Angarotipula illustris 36 0.133 - Dolichopeza obscura 1 0.004 - Nephrotoma alterna 2 0.007 - eucera 3 0.011 - ferruginea 32 0.118 - occipitalis 1 0.004 - sp. 14 0.052 - Tipula hermannia 1 0.004 - oropezoides 2 0.007 - penobscot 1 0.004 - sulphurea 3 0.011 - tricolour 1 0.004 -

107

Insecta Diptera Tipulidae Tipula sp. 20 0.074 - Xylophagidae Xylophagus sp. 1 0.004 - Ephemeroptera Baetidae Cloeon sp. 1 0.004 - Caenidae Caenis punctata 5 0.018 - youngi 6 0.022 - Ephemerellidae Eurylophella temporalis 1 0.004 - Ephemeridae Hexagenia atrocaudata 3 0.011 - limbata 1 0.004 - Heptageniidae Maccaffertium vicarium 1 0.004 - Hemiptera Acanthosomatidae Elasmostethus cruciatus 1 0.004 - Aphididae Acyrthosiphon pisum 1 0.004 C Aphrophoridae Philaenus spumarius 2 0.007 - sp. 2 0.007 - Cercopidae Aphrophora sp. 1 0.004 - Neophilaenus lineatus 1 0.004 - Cicadellidae Aphrodes diminuta 4 0.015 - Athysanus argentarius 1 0.004 - Gyponana sp. 1 0.004 - Clastopteridae Clastoptera obtusa 1 0.004 - Corixidae Sigara decoratella 1 0.004 - Lygaeidae Kleidocerys resedae 1 0.004 - Nysius sp. 1 0.004 C Membracidae Tortistilus sp. 1 0.004 - Miridae Lygus lineolaris 2 0.007 C Neolygus sp. 2 0.007 - Pappus sp. 1 0.004 - Stenodema pilosipes 1 0.004 - Pentatomidae Neottiglossa undata 1 0.004 - Rhopalidae Arhyssus sp. 1 0.004 -

108

Insecta Hemiptera Rhyparochromidae Megalonotus sp. 1 0.004 - Sphragisticus sp. 2 0.007 - Hymenoptera Braconidae Meteorus sp. 1 0.004 - Microplitis sp. 2 0.007 - Diprionidae Diprion Diprion similis 1 0.004 - Monoctenus sp. 1 0.004 - Eurytomidae Eurytoma sp. 1 0.004 - Figitidae sp. 3 0.011 - Ichneumonidae sp. 9 0.033 - Cylloceria sp. 1 0.004 - Exenterus Exenterus confusus 1 0.004 - Ichneumon sp. 1 0.004 - Mesochorus sp. 1 0.004 - Mesoleptus sp. 1 0.004 - Promethes Promethes sulcator 1 0.004 - sp. 1 0.004 - Perilampidae Euperilampus sp. 1 0.004 - Tenthredinidae sp. 1 0.004 - Lepidoptera sp. 1 0.004 - Autostichidae Oegoconia deauratella 1 0.004 - Batrachedridae Batrachedra pinicolella 1 0.004 - Coleophoridae Coleophora deauratella 1 0.004 - Crambidae Fissicrambus mutabilis 1 0.004 - Neodactria sp. 1 0.004 - Petrophila canadensis 1 0.004 - Erebidae Caenurgina crassiuscula 1 0.004 - Ctenucha virginica 1 0.004 - Hyphantria sp. 2 0.007 - Idia sp. 1 0.004 -

109

Insecta Lepidoptera Erebidae Rivula sp. 1 0.004 - Gelechiidae Agnippe sp. 1 0.004 - Scrobipalpa sp. 1 0.004 - Geometridae Pasiphila rectangulata 2 0.007 - Hesperiidae Polites sp. 6 0.022 - Thymelicus lineola 1 0.004 - Lasiocampidae Malacosoma sp. 2 0.007 - Lycaenidae Satyrium iparops 1 0.004 - Momphidae Mompha epilobiella 1 0.004 - Nepticulidae sp. 1 0.004 - Noctuidae Apamea devastator 1 0.004 - sordens 1 0.004 - Cucullia intermedia 1 0.004 - Hyppa sp. 1 0.004 - Lacanobia sp. 1 0.004 - Loscopia velata 2 0.007 - Mythimna sp. 3 0.011 - Protodeltote albidula 2 0.007 - Xestia sp. 2 0.007 - Nymphalidae Coenonympha tullia 1 0.004 - Pyralidae Pococera expandens 1 0.004 - Tineidae Niditinea fuscella 1 0.004 - Tortricidae sp. 1 0.004 - Acleris sp. 1 0.004 - Archips sp. 1 0.004 F Catastega aceriella 2 0.007 - Cnephasia sp. 1 0.004 - Dichrorampha sp. 2 0.007 - Epiblema obfuscana 1 0.004 -

110

Insecta Lepidoptera Tortricidae Eucosma sp. 1 0.004 - Olethreutes baccatana 1 0.004 - Phtheochroa sp. 1 0.004 - Sparganothis sp. 1 0.004 - Spilonota sp. 1 0.004 - Zeiraphera sp. 1 0.004 - Ypsolophidae Ochsenheimeria sp. 2 0.007 - Neuroptera Chrysopidae Chrysoperla sp. 1 0.004 - Coniopterygidae sp. 1 0.004 - Hemerobiidae Hemerobius humulinus 3 0.011 - Micromus posticus 1 0.004 - Psectra sp. 1 0.004 - Odonata Coenagrionidae Enallagma sp. 1 0.004 - Orthoptera Gryllidae Allonemobius sp. 1 0.004 - Trichoptera Leptoceridae Limnephilus sp. 1 0.004 - Oecetis inconspicua 1 0.004 - Triaenodes aba 1 0.004 - sp. 2 0.007 -

111

Table A.3 Summary table of the frequency of occurrence of arthropod families in the diet of Barn Swallow nestlings by year, brood and breeding site.

Year Brood Site Family 2015 2016 1 2 77 43 42 41 34 26 24 20 9 2 Tachinidae 0.57 0.57 0.49 0.71 0.10 0.52 0.74 0.27 0.58 0.46 0.75 0.87 0.67 0.61 Tipulidae 0.70 0.42 0.67 0.34 0.40 0.37 0.63 0.64 0.79 0.60 0.63 0.52 0.50 0.30 Limoniidae 0.60 0.48 0.64 0.36 0.40 0.37 0.44 0.64 0.63 0.76 0.50 0.52 0.33 0.26 Calliphoridae 0.72 0.36 0.55 0.50 0.50 0.44 0.63 0.55 0.47 0.60 0.42 0.52 0.63 0.35 Anthomyiidae 0.51 0.28 0.44 0.31 0.40 0.30 0.33 0.55 0.32 0.46 0.38 0.39 0.46 0.26 Muscidae 0.48 0.23 0.31 0.42 0.50 0.37 0.33 0.45 0.32 0.37 0.13 0.48 0.46 0.17 Tabanidae 0.45 0.25 0.34 0.35 0.10 0.37 0.44 0.18 0.11 0.48 0.21 0.22 0.42 0.26 Culicidae 0.48 0.11 0.38 0.11 0.30 0.22 0.30 0.18 0.32 0.33 0.25 0.30 0.33 0.17 Curculionidae 0.22 0.31 0.21 0.38 0 0.33 0.33 0.09 0.21 0.25 0.54 0.26 0.25 0.17 Sarcophagidae 0.20 0.22 0.22 0.19 0 0.19 0.19 0.36 0.16 0.28 0.21 0.30 0.13 0.09 Miridae 0.31 0.08 0.21 0.15 0.20 0.15 0.22 0.18 0.47 0.20 0.25 0.13 0.08 0.04 Syrphidae 0.19 0.06 0.12 0.13 0.10 0.11 0.11 0.18 0.05 0.17 0.04 0.04 0.21 0.13 Asilidae 0.14 0.11 0.16 0.06 0 0.11 0.19 0.18 0.11 0.16 0.13 0.17 0 0.04 Scirtidae 0.19 0.03 0.09 0.13 0 0.11 0.07 0.18 0.16 0.14 0.04 0.09 0.13 0 Agromyzidae 0.14 0.07 0.11 0.09 0 0.11 0.15 0.09 0.11 0.12 0.17 0.09 0.08 0 Fanniidae 0.19 0.02 0.09 0.12 0 0.11 0.22 0.27 0 0.05 0.17 0.22 0.04 0.04 Chironomidae 0.17 0.04 0.11 0.08 0 0.04 0.22 0.09 0.11 0.11 0.13 0.09 0.08 0.04 Hybotidae 0.12 0.07 0.12 0.05 0.10 0.04 0.15 0.09 0.16 0.16 0.08 0.04 0 0 Sphaeroceridae 0.16 0.02 0.09 0.08 0.20 0.15 0.11 0.09 0.16 0.11 0 0 0 0.09 Cicadellidae 0.09 0.09 0.08 0.10 0 0.19 0.04 0 0.05 0.12 0.17 0.13 0 0 Ptinidae 0.13 0.04 0.13 0 0.10 0 0.19 0 0.05 0.06 0.21 0.13 0.13 0

112

Carabidae 0.08 0.09 0.07 0.11 0 0.11 0.04 0.09 0.05 0.08 0.25 0.17 0 0 Noctuidae 0.14 0.03 0.12 0.02 0 0.07 0.15 0 0.11 0.14 0 0.04 0.04 0 Hydrophilidae 0.09 0.08 0.09 0.06 0.30 0.11 0 0.18 0 0.12 0 0.04 0.13 0 Drosophilidae 0.13 0.01 0.09 0.03 0.30 0 0.04 0.09 0 0.06 0.04 0.04 0.04 0.26 Empididae 0.11 0.04 0.10 0.02 0.10 0 0.07 0 0.21 0.07 0.13 0.09 0 0.04 Ichneumonidae 0.09 0.06 0.09 0.04 0 0 0.07 0 0.11 0.11 0.04 0.09 0.13 0 Ephydridae 0.09 0.06 0.09 0.03 0 0 0.04 0.09 0 0.11 0.17 0.09 0.08 0 Tortricidae 0.13 0.01 0.08 0.04 0 0 0.07 0.09 0.11 0.05 0.04 0.17 0.13 0.04 Dolichopodidae 0.12 0.01 0.08 0.04 0.10 0.04 0.07 0 0.05 0.10 0.04 0.04 0.04 0.04 Chrysomelidae 0.09 0.03 0.09 0.01 0.10 0 0 0.36 0.11 0.05 0.04 0.04 0 0.13 Sepsidae 0.09 0.02 0.08 0.02 0.10 0.07 0 0 0.11 0.07 0.08 0.04 0 0.04 Stratiomyidae 0.09 0.03 0.07 0.03 0.10 0 0 0 0.11 0.06 0.08 0.22 0 0 Rhagionidae 0.08 0.01 0.03 0.07 0.30 0 0 0 0 0.07 0 0.04 0.08 0 Phoridae 0.08 0.01 0.03 0.07 0 0 0 0 0.05 0.10 0 0 0 0.13 Mycetophilidae 0.06 0.03 0.05 0.04 0 0.04 0.11 0.09 0.11 0.04 0.04 0 0.04 0 Hesperiidae 0.05 0.04 0.07 0 0 0.04 0.11 0 0.05 0.08 0 0 0 0 Caenidae 0.05 0.04 0.06 0 0 0 0.04 0.27 0 0.02 0.17 0 0.04 0 Chloropidae 0.05 0.03 0.03 0.04 0 0 0.04 0 0.05 0.06 0.04 0.04 0 0.04 Ceratopogonidae 0.05 0.01 0.04 0.02 0 0 0 0 0.05 0.07 0 0.04 0.04 0 Tephritidae 0.04 0.03 0.02 0.06 0 0 0.04 0 0.05 0.02 0.08 0.09 0.04 0 Crambidae 0.05 0.01 0.05 0 0.10 0.04 0 0 0.05 0.05 0 0.04 0 0 Erebidae 0.05 0.01 0.03 0.03 0 0.04 0 0 0.11 0.06 0 0 0 0 Figitidae 0.04 0.02 0.01 0.06 0 0 0.07 0 0 0.02 0.04 0.04 0.08 0 Aphrophoridae 0.04 0.02 0.03 0.03 0 0.07 0 0 0 0.02 0 0.13 0 0.04 Membracidae 0.03 0.03 0.04 0.01 0 0.04 0 0 0.11 0.04 0.04 0.04 0 0 Leptoceridae 0.04 0.01 0.03 0.01 0 0 0.04 0 0.05 0.02 0.08 0 0.04 0 Pipunculidae 0.03 0.02 0.04 0 0 0.04 0 0 0 0.02 0.04 0 0.04 0.09 Staphylinidae 0.03 0.01 0.02 0.02 0.10 0 0.04 0 0.05 0.02 0 0 0.04 0

113

Dermestidae 0.02 0.02 0.02 0.02 0 0.04 0.04 0 0 0.02 0 0 0.04 0.04 Nitidulidae 0.04 0 0.01 0.03 0 0.04 0.04 0 0.05 0.01 0.04 0 0 0 Gryllidae 0.03 0.01 0 0.05 0 0.04 0 0 0.05 0.02 0 0 0.04 0 Geometridae 0.02 0.01 0.03 0 0.10 0 0 0 0.05 0.01 0 0.04 0 0.04 Ephemeridae 0.02 0.01 0.01 0.04 0 0 0 0 0 0.06 0 0 0 0 Sciomyzidae 0.02 0.02 0.02 0.01 0 0 0 0 0 0.02 0.04 0.04 0.04 0 Isonychiidae 0.02 0.01 0.02 0.01 0 0 0 0 0 0.02 0 0 0 0.09 Hemerobiidae 0.02 0.01 0.02 0.01 0 0 0 0 0.05 0.02 0 0 0.04 0 Scathophagidae 0.02 0.01 0.01 0.03 0 0.04 0.04 0 0 0.02 0 0 0 0 Opomyzidae 0.02 0.01 0.01 0.02 0 0 0 0 0 0.01 0.08 0 0.04 0 Braconidae 0.02 0.01 0.02 0 0 0.04 0.04 0 0.05 0.01 0 0 0 0 Simuliidae 0.02 0 0.01 0.01 0 0 0.04 0 0 0.01 0 0.04 0 0 Milichiidae 0.02 0 0.01 0.01 0 0 0 0 0.05 0 0.04 0.04 0 0 Cercopidae 0.02 0 0.01 0.02 0 0 0 0 0 0.02 0 0 0.04 0 Ypsolophidae 0.02 0.01 0.02 0 0 0 0.04 0 0 0.01 0 0 0.04 0 Psychodidae 0.02 0.01 0.01 0.01 0 0 0 0 0 0.02 0 0 0 0.04 Histeridae 0.02 0.01 0.01 0.02 0 0 0.04 0 0 0.02 0 0 0 0 Ditomyiidae 0.02 0.01 0.01 0.01 0 0.04 0.04 0 0 0.01 0 0 0 0 Cecidomyiidae 0.01 0.01 0.02 0 0 0 0 0 0 0.01 0.08 0 0 0 Ripiphoridae 0 0.02 0 0.03 0 0 0.04 0 0 0 0 0.04 0 0.04 Rhyparochromidae 0 0.02 0.01 0.02 0 0 0 0 0 0.04 0 0 0 0 Lasiocampidae 0 0.02 0.02 0 0 0 0 0 0.05 0.01 0 0 0.04 0 Coleophoridae 0 0.02 0.01 0.01 0 0 0.04 0 0 0.01 0 0 0 0.04 Therevidae 0.02 0 0.01 0.01 0 0.04 0 0 0 0 0 0.04 0 0 Scarabaeidae 0.02 0 0.01 0.01 0 0 0 0 0 0.01 0 0.04 0 0 Phalacridae 0.02 0 0 0.02 0 0 0 0 0.05 0.01 0 0 0 0 Gelechiidae 0.02 0 0.01 0 0 0.04 0 0 0 0.01 0 0 0 0 Depressariidae 0.02 0 0.01 0 0 0 0.04 0 0.05 0 0 0 0 0

114

Bibionidae 0.02 0 0.01 0.01 0 0 0.07 0 0 0 0 0 0 0 Unionicolidae 0.01 0.01 0.01 0 0 0.04 0 0 0 0.01 0 0 0 0 Rhopalidae 0.01 0.01 0 0.02 0 0.04 0 0 0 0 0 0.04 0 0 Lygaeidae 0.01 0.01 0.01 0.01 0 0 0 0.09 0 0.01 0 0 0 0 Lauxaniidae 0.01 0.01 0 0.02 0 0.04 0 0 0 0 0 0.04 0 0 Diprionidae 0.01 0.01 0.01 0 0 0 0.04 0 0 0.01 0 0 0 0 Anisopodidae 0.01 0.01 0.01 0.01 0 0 0 0 0.05 0.01 0 0 0 0 Sciaridae 0 0.01 0 0.02 0.20 0 0 0 0 0 0 0 0 0 Xylophagidae 0.01 0 0.01 0 0 0 0.04 0 0 0 0 0 0 0 Tineidae 0.01 0 0 0.01 0 0 0 0 0 0 0 0 0.04 0 Rhinophoridae 0.01 0 0 0.01 0 0 0 0 0.05 0 0 0 0 0 Pyralidae 0.01 0 0.01 0 0 0 0 0 0.05 0 0 0 0 0 Ptychopteridae 0.01 0 0 0.01 0 0 0 0 0 0 0.04 0 0 0 Perilampidae 0.01 0 0 0.01 0 0 0 0 0 0.01 0 0 0 0 Pentatomidae 0.01 0 0.01 0 0 0 0 0.09 0 0 0 0 0 0 Nymphalidae 0.01 0 0 0.01 0 0 0 0 0 0.01 0 0 0 0 Lycaenidae 0.01 0 0 0.01 0 0 0.04 0 0 0 0 0 0 0 Lonchopteridae 0.01 0 0 0.01 0 0 0 0 0 0.01 0 0 0 0 Latridiidae 0.01 0 0.01 0 0 0.04 0 0 0 0 0 0 0 0 Keroplatidae 0.01 0 0.01 0 0 0 0 0 0 0 0 0 0.04 0 Heptageniidae 0.01 0 0.01 0 0 0 0 0 0 0 0 0 0 0.04 Ephemerellidae 0.01 0 0.01 0 0 0 0 0 0 0 0.04 0 0 0 Cryptophagidae 0.01 0 0 0.01 0 0 0 0 0 0.01 0 0 0 0 Coccinellidae 0.01 0 0 0.01 0 0 0 0 0 0 0 0 0.04 0 Clastopteridae 0.01 0 0 0.01 0 0 0 0 0 0.01 0 0 0 0 Chrysopidae 0.01 0 0.01 0 0 0 0 0 0 0.01 0 0 0 0 Chaoboridae 0.01 0 0.01 0 0 0 0 0 0 0 0.04 0 0 0 Cerambycidae 0.01 0 0.01 0 0 0 0 0 0 0.01 0 0 0 0

115

Cantharidae 0.01 0 0.01 0 0 0 0 0 0 0.01 0 0 0 0 Batrachedridae 0.01 0 0.01 0 0 0 0 0 0 0 0 0 0.04 0 Autostichidae 0.01 0 0.01 0 0 0 0.04 0 0 0 0 0 0 0 Anthribidae 0.01 0 0.01 0 0 0 0.04 0 0 0 0 0 0 0 Anthicidae 0.01 0 0 0.01 0 0 0 0 0.05 0 0 0 0 0 Acanthosomatidae 0.01 0 0.01 0 0 0.04 0 0 0 0 0 0 0 0 Tenthredinidae 0 0.01 0.01 0 0 0 0 0 0 0.01 0 0 0 0 Platystomatidae 0 0.01 0 0.01 0 0 0 0 0 0 0.04 0 0 0 Platypezidae 0 0.01 0 0.01 0 0 0 0 0 0.01 0 0 0 0 Nepticulidae 0 0.01 0.01 0 0 0 0 0 0 0.01 0 0 0 0 Momphidae 0 0.01 0.01 0 0 0 0 0 0 0.01 0 0 0 0 Lonchaeidae 0 0.01 0.01 0 0 0 0 0 0 0 0 0 0.04 0 Julidae 0 0.01 0.01 0 0 0 0 0 0 0 0.04 0 0 0 Eurytomidae 0 0.01 0.01 0 0 0 0 0 0 0 0 0.04 0 0 Elateridae 0 0.01 0.01 0 0 0 0.04 0 0 0 0 0 0 0 Diastatidae 0 0.01 0.01 0 0 0 0 0 0 0.01 0 0 0 0 Corixidae 0 0.01 0 0.01 0 0 0 0 0 0.01 0 0 0 0 Coniopterygidae 0 0.01 0 0.01 0 0 0 0 0 0 0 0 0.04 0 Coenagrionidae 0 0.01 0.01 0 0 0 0 0 0 0.01 0 0 0 0 Cleridae 0 0.01 0.01 0 0 0 0.04 0 0 0 0 0 0 0 Bolitophilidae 0 0.01 0.01 0 0 0 0 0 0 0.01 0 0 0 0 Baetidae 0 0.01 0.01 0 0 0 0 0 0 0 0 0.04 0 0 Arrenuridae 0 0.01 0.01 0 0 0.04 0 0 0 0 0 0 0 0 Aphididae 0 0.01 0.01 0 0 0 0.04 0 0 0 0 0 0 0 Erythraeidae 0 0 0 0 0 0 0 0 0 0 0 0 0 0

116

Appendix B

Figure B.1 Correlation matrix of the prey availability predictor variables used in colony size models.

117

Figure B.2 Correlation matrix of the prey availability predictor variables used in rate of second brooding models.

118

Figure B.3 Correlation matrix of the prey availability predictor variables used in reproductive success models.

119

Figure B.4 Total prey abundance from all Barn Swallow breeding sites summarized by year and brood.

120