TREE SWALLOW BREEDING BIOLOGY AND THE PHENOLOGY OF
AQUATIC EMERGENT PREY IN ARTIFICIAL AND NATURAL WETLANDS
A
THESIS
SUBMITTED TO THE SCHOOL OF GRADUATE STUDIES
OF
BLOOMSBURG UNIVERSITY OF PENNSYLVANIA

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

PROGRAM IN BIOLOGY
DEPARTMENT OF BIOLOGICAL AND ALLIED HEALTH SCIENCES

BY
VICTORIA G. ROPER

BLOOMSBURG, PENNSYLVANIA
2020

III
ABSTRACT
Over the past 80-100 years, aerial insectivore populations have declined across
North America. Climate change and agricultural intensification are hypothesized to be
responsible for the declines by negatively affecting their main food source, aerial insects.
Some management strategies in agroecosystems have focused on creating aquatic habitats
(e.g. prairie ponds and artificial wetlands) to mitigate nutrient runoff, but their use by
avian species has not been well studied. Tree swallows (Tachycineta bicolor) are aquatic
aerial specialists and forage over aquatic habitats on emergent adult aquatic insects such
as Dipteran and Ephemoptera. Subsequently, their breeding success could be negatively
affected by having reduced insect abundances, having implications for future population
dynamics. This study investigated whether artificial wetlands in agroecosystems from
corn cultures in northeast Pennsylvania and current climate conditions are creating
ecological traps for reproducing populations of Tree swallows. I investigated whether
trophic mismatch between emerging aquatic Diptera prey and reproductive biology
existed. I evaluated aquatic insect emergence from artificial and natural wetlands
surrounded by intensive crop cultures as potential prey availability to Tree swallows in
northeast Pennsylvania. I predicted the Diptera prey abundance will be similar in the
natural versus artificial wetland because of the agriculture land surrounded by it. To do
this, I used emergence traps to assess abundance of aquatic Diptera emergence in
different wetland types (artificial and natural) utilized by breeding Tree swallow breeding
pairs. The wetland sites were selected based on surrounding agroecosystems of intensive
cultures.

IV
Reproductive timing of egg laying, chick hatching, and subsequent chick growth
was quantified for each site. I predicted that reproductive timing will correspond with
peak insect emergence of aquatic Diptera and if peak aquatic insect prey abundance is
synchronized with reproduction it will have a significant effect on clutch size and chick
growth. I found while peak aquatic insect Diptera prey abundance was synchronized with
Tree swallow reproductive timing at two artificial wetland sites this did not have a
significant effect on clutch size or chick growth. An implication from my results are that
populations are able to benefit from lower Diptera aquatic insect emergence than
hypothesized and peak emergence is not driving Tree swallow breeding phenology. Also,
terrestrial prey subsidies may be more important than accounted for in this research. This
suggests that aquatic sampling alone may not capture Tree swallow’s dietary flexibility in
artificial and natural wetlands. My results highlight the importance of aquatic ecosystems
for reproducing populations of Tree swallows.

V
ACKNOWLEDGEMENTS
First, I want to thank Dr. Clay E. Corbin for providing mentorship support
throughout my graduate career. His work studying aerial insectivores for the past 20
years has inspired this research. He taught me how to band birds, become an avian
phlebotomist, analyze data, conduct statistical analyses that were beyond me to answer
my questions. Moreover, he always encouraged me to focus on “the question” and to not
get lost in the details. He has been endlessly supportive of me delving into ornithological
systems and methods that were brand-new to me all in order to answer exciting
ecological questions. I am thankful for Dr. Corbin always believing in me and pushing
me to become a better scientist and future ornithologist.

I next want to thank my committee members Dr. Steve Rier, Dr. Thomas Klinger,
and Dr. Lauri Green. Dr. Rier has always had an open-door policy encouraging students
to ask questions and seek advice which I am thankful for as I always seemed to be at his
door seeking advice. Additionally, it was his guidance that steered the direction for the
development of my thesis methods, and ultimately as Dr. Corbin would say “the
question”. Dr. Klinger has always offered his guidance and support with enthusiasm.

I would like to thank my parents, Tim Roper and Marian Comiskey Roper; my
sisters, Bridget Kane Kelly, Erin Kelly Furby, and Alex Roper; my friend, Christopher
Spaid for always supporting my education. This work would not have been possible
without their input. Finally, I would like to extend a special thank you to the students I
mentored at Bloomsburg University, as each helped shape my future.

VI
TABLE OF CONTENTS

Title Page …………………………………………………………………………………1
Approval Page…………………………………………………………………………......2
Acknowledgements……………………………………………………………………......3
List of Tables……………………………………………………………………………...5
List of Figures……………………………………………………………………………..5
List of Appendices………………………………………………………………………...4
Manuscript………………………………………………………………………………...7
Abstract……………………………………………………………………………………7
Introduction………………………………………………………………………………10
Materials and Methods…………………………………………………………………...21
Results……………………………………………………………………………………25
Discussion………………………………………………………………………………..35
Literature Cited…………………………………………………………………………..36
Appendices………………………………………………………………………………49

VII
LIST OF TABLES:
Table 1. Raw Data on Numbers of Emergent Insects Captured by emergent traps..........35
Table 2. Raw Data on Chick Growth…………………………………………………….36
Table 3. Interpolated data for emergent insects captured by emergent traps for two sites
(4&2). The microhabitat types are open water and algae while the sample days are both 3.
The abundances of insects that was derived using the polynomial equations are listed in
the table above. ………………………………………………………………………….36
LIST OF FIGURES:
Figure 1. Map of study sites in Pennsylvania……………………………………………37
Figure 2. United States Department of Agriculture; National Agriculture Statistics
Service 1983-2017 in Columbia County, Pennsylvania. Intensive crops harvested: corn
and soybeans and extensive cultures harvest: hay……………………………………….38
Figure 3. Daily numerical density averaged over three microhabitats, for all emergent
insects and for the Dipteran order, from Fry (top left), Murray (top right), Lourwist
(bottom left), and Tanner (bottom right) wetlands, 2019. Drop lines indicate the Julian
date of each sample. Total numerical emergence is equal to the area under the curve….39
Figure 4. Relationship between egg hatching date and insect abundance for Site 1 wetland
site. Depicted are the total number of Tree swallow eggs and newly hatched chicks in all
nest boxes, and the number of insect (individuals per m2 per day) captured in emergence
traps on each
date………………………………………………………………………..……………..40
Figure 5. Relationship between egg hatching date and insect abundance for Site 2 wetland
site. Depicted are the total number of Tree swallow eggs and newly hatched chicks in all
nest boxes, and the number of insect (individuals per m2 per day) captured in emergence
traps on each date..............................................................................................................41
Figure 6. Relationship between egg hatching date and insect abundance for Site 3
wetland site. Depicted are the total number of Tree swallow eggs and newly hatched
chicks in all nest boxes, and the number of insect (individuals per m2 per day) captured in
emergence traps on each date …………………………………………………………42
Figure 7. Relationship between egg hatching date and insect abundance for Site 4 wetland
site. Depicted are the total number of Tree swallow eggs and newly hatched chicks in all
nest boxes, and the number of insect (individuals per m2 per day) captured in emergence
traps on each date …......................................................................................................43

VIII

LIST OF APPENDICES:
Appendix 1. Institutional Animal Care and Use Committee Approval…………………47
Appendix 2. Raw Data…………………………………………………………………...48
Appendix 3. Tables of Finalized Results………………………………………………...49
Appendix 4. Figures Depicting Finalized Results………………………………………...5

1

INTRODUCTION
Since 1970, North America has seen a significant decline in avifauna (Rosenberg
et. al 2019). Over 528 different avian species spanning across nine different biomes (e.g.
Coasts, Arid Lands, Eastern Forest, Arctic Tundra, Western Forest, Boreal Forest, and
Grassland) had an approximate 2.8 billion net population decline (Rosenberg et. al 2019).
One North American foraging guild of avian species, aerial insectivores (hereafter “aerial
insectivores”) are experiencing the most significant population declines (Rosenberg et. al
2019, Nebel et al. 2010). The geographic patterns of decline are well-documented in the
North American Breeding Bird Survey (BBS). Since 1970, aerial insectivorous
populations have declined by ~32% in the United States and by ~45% in Canada. The
declines are strongest in populations that breed in northeast North America, however the
explanation for this trend is still unclear (Nebel et al. 2010, Smith et al. 2015, Michel et
al. 2016).
Aerial insectivores specialize on feeding on aerial invertebrates. They encompass
over thirty species in North America including swifts, nightjars, swallows, and
flycatchers. The main hypotheses to explain aerial insectivore declines are changes in
aerial insect phenology, abundance, and availability (Nebel et al. 2010). Key drivers of
changes in prey availability are agricultural intensification and climate change (Nebel et
al. 2010).

2
Agricultural intensification impact on aquatic insect emergence and avian
reproductive timing
Agricultural intensification has negatively affected insect abundance and
biodiversity through disruptive farming practices, wetland drainage, pesticide usage, and
habitat fragmentation (Nebel et al. 2010). Agricultural intensification has been shown to
reduce insect prey for terrestrial consumers including farmland birds in Europe. Recently,
in Canada, agroecosystem research has narrowed its focus on intensive cultures versus
intensive cultures on aerial insectivore declines by reducing Diptera prey (Ghilain and
Belisle 2008). Intensive cultures include soybeans, maize, and other cereals while
extensive cultures include hayfields and livestock pastures.
Intensive culture agriculture may have negative effects on breeding success of
Tree swallows. Regions with higher extensive agriculture than intensive agriculture
fledge twice more Tree swallow nestlings that those breeding in areas of highly intensive
agriculture and low extensive agriculture (Ghilain and Belisle 2008). Landscapes with
intensive crops support less invertebrate prey and can affect the survival of chicks before
fledging. Aquatic habitats such as artificial wetlands surrounded by intensive agriculture
may be more susceptible to reduced insect availability from pesticide accumulation and
habitat simplification creating an ecological trap for Tree swallow breeding on wetlands
surrounded by intensive cultures. Tree swallows forage heavily on emergent insect prey
with aquatic larval stages that are sensitive to agricultural activities particularly Diptera
(McCarty & Winkler 1999, Lenat & Crawford 1994). Focusing on wetlands surrounded
by intensive cultures that produce aquatic Diptera insect prey abundance for Tree
swallows show be a priority. Though agricultural practices such as livestock pasture and

3
hedgerows can increase insect availability the creation of a potential ecological trap from
specifically intensive cultures may reduce aquatic Diptera emergent prey availability but
has not been extensively studied (Moller 2001, Evans et al. 2007, Gruuebler et al. 2007,
Paquette et al. 2013).
Climate change impact on aquatic insect emergence and avian reproductive timing
In the 20th century, human‐caused emissions of greenhouse gases have global
mean in air temperatures to increase at a rate of 0.047 ± 0.050 °C per year having
implications on the phenology of aerial insects emerging from aquatic ecosystems. Aerial
insect emergences from aquatic ecosystems are mediated by temperature and photoperiod
(Ratte et al. 1984, Quayle 2002). Climate change affects the rate and timing of the
development of aerial insects that emerge from aquatic ecosystems, hereafter referred to
as emergent aquatic insects. Changes in ambient temperatures have direct consequences
on metabolic rates, activity patterns, and developmental rates of emergent insects
(Rempel et al. 1987). Additionally, emergent aquatic insects have decreased growth and
fecundity when spring temperatures are high because of increased metabolic costs
(Sweeney & Schnack, 1977, Rempel et al. 1987, Greig et al., 2012). Emergent aquatic
insects respond to changes in temperatures by expressing thermal plasticity in their
lifecycle. For example, Jonsson (2015) showed that aquatic insects are emerging earlier
with increasing spring temperatures. This plasticity could result in emergence occurring
earlier in the season (Jonsson 2015).
The development of multivoltine emergent aquatic insects in the insect orders
Ephemeroptera, Odonata, and Diptera is temperature-dependent and they may be more
susceptible to having “lost generations”. Ephemeroptera include numerous multivoltine

4
species and the number of generations within their life cycles varies with other insect
orders (Van Dyck et al. 2015). The development of both Ephemeroptera eggs and larvae
are temperature dependent (Van Dyck et al. 2015). Increasing temperatures from climate
change may delay the emergence of the last generation of Ephemeroptera causing a
phenological change and this generation would fall victim to the lost generation
hypothesis (Van Dyck et al. 2015). A lost generation would negatively impact their
population and may have implications for terrestrial predator populations. There is
evidence that Odonates emerge later in the season (Van Dyck et al. 2015). The end result
is a decrease in the population over time leading to a lost generation of individuals (Doi
2008). Dipteran emergence rate can increase at higher temperatures (~28°C) it may
decline and reduce insect abundances (Bayoh 2013). This leads to a phenological shift in
emergence and a decrease emergence rate over time because the increase water
temperature causes developmental problems (Bayoh 2013). Overall, majority of emergent
aquatic insect orders from aquatic ecosystems are emerging earlier in the season and are
susceptible to developmental traps leading to a lost generation of emerging aquatic
insects (Rempel & Carter 1987, Greig et al. 2012). The phenology of emergent aquatic
insects particularly aquatic Diptera prey under current climate conditions has not been
extensively explored in agroecosystems and may affect terrestrial consumer populations
that rely on aquatic insects as prey (Baxter et al. 2004, Stracevisius et al. 2013).
Another consequence of warming temperatures is that birds are breeding earlier
(Dunn &Winkler 1999, Sala et al. 2000, Visser 2004). Single brood species are showing
significant advancement in laying date, likely in an attempt to time their reproduction
with prey abundances (Visser 2004). Tree swallows are typically a single brood species

5
and the egg‐laying date has advanced up to nine days between 1959 and 1991 (Dunn and
Winkler 1991, Robertson et al. 1992). This advance in phenology was associated with
increasing surface air temperatures at the time of breeding. This phenological shift has
implications for breeding success (Dunn and Winkler 1999, Dunn and Moller 2014).
Birds advancing their breeding are experiencing negative responses such as reproductive
failure or poor recruitment (Visser et al. 2006, Moller 2008, Both et al. 2009, Vatka
et al. 2011). Temperature and photoperiod are regulators in most non-tropical avian
species and influencing the timing of breeding (Bentley 2003). Increasing photoperiods
stimulate secretion of gonadotropin-releasing hormone and consequent gonadal
maturation in avian species (Dawson et al. 2001). It is not clear which cue (photoperiod
or temperature) has greater influence in reproductive success. Avian species that rely on
photoperiod rather than temperature for the timing of their reproduction, may miss
increases in prey availability as emergent insects are more sensitive to temperature cues
than photoperiod.
Importance of Dipteran aquatic insects as prey for aerial insectivores: aquatic
nutritional subsidies
Aquatic insect emergence is an important energy transfer from primary producers
to higher trophic levels in wetlands (Mitch and Gosselink 1993, Batzer and Wissinger
1996) and aquatic to terrestrial ecosystems that support avian species (Stagliano et al.
1998, Nakano and Murakami 2001, Bartrons 2018). Nakano and Murakami (2001) found
that aquatic emergent insects provided 50-90% of the monthly energy budget for avian
species. Emerging aquatic insects are captured by Tree swallows and used for egg
development and as a food source to feed to their young (Blancher 1991). Diptera

6
emerging from water are utilized by Tree swallows, particularly to feed to growing
nestlings and is the main prey to finance their reproduction.
Changes in biomass of aquatic emergent insects are a strong, positive predictor of
aerial insectivore fledging success, whereas terrestrial insects had no impact on fledging
success (Twinging et al. 2016). Furthermore, aquatic emergent insects, compared to
terrestrial insects, contain a higher quality nutritional composition of omega-3 highly
unsaturated fatty acids (HUFAs) (Twinging et al 2016). Twinging et al. (2016) also found
that chicks fed high-omega 3’s or long chain polyunsaturated fatty acids grew faster,
were in better condition, and had greater immune function. Few studies have focused on
possible effects of a trophic mismatch between emergent aquatic insects particularly
Diptera in wetlands and evaluated the influence on avian species that rely on energy from
the aquatic–terrestrial interface (Ballinger & Lake, 2006, Donaldson & Vander Zanden,
2008, Greig et al., 2012).
Mitigating aerial insectivore declines by constructing artificial wetlands
Creating terrestrial non-crop habitats such as livestock pastures instead of
intensive cultures can mitigate some of the declines in insect populations that aerial
insectivores eat during the breeding season. However, this conservation strategy may not
be sufficient to support breeding populations of aerial insectivores. Insects that emerge
from terrestrial landscapes, compared to aquatic insects, lack nutrients important for
chick growth and development (Johnson and Lombardo 2000, Poulin et al. 2010, Elgin
2019). In Canada, terrestrial non-crop habitats were not a substitute for aquatic habitats in
agroecosystems in studies of for Tree swallows (Morrissey et al. 2017). This suggests
that aquatic insect availability could be more important than overall insect abundance for

7
breeding aerial insectivore populations. Importantly, the phenology of aquatic insects
may have implications for breeding success (Twingings et al. 2018).
The creation of aquatic habitats such as ponds, lakes, wetlands, and restoration of
wetlands that produce high-quality aquatic Diptera emergent prey, may be a method to
reduce the widespread declines of aerial insectivores. Construction of prairie ponds in
Canada and artificial wetlands have been used to support a broad range of other avian
species. Most studies comparing avian and insect communities in natural and artificial
wetlands focused on metrics of richness, abundance, density (Batzer and Wissinger 1996,
Arena, Battisti et al. 2011, Giosa, Mammides et al. 2018). Few studies exist on the
phenological mismatches of high-quality Diptera prey abundance, emergent aquatic
insects and aerial insectivore reproductive biology (Both et al. 2006, Lyon et al. 2008,
Dunn et al. 2011, Kunkel et al. 2013, Paquette et al. 2013, Stanto et al 2017).
Though wetlands may play an important role in supporting biodiversity, it is
critical to understand whether they can be utilized to reduce the rapid declines of key
species such as aerial insectivores. Management strategies in agricultural landscapes may
need to consider phenological mismatches and abundance of Diptera prey to mitigate the
potentially negative effects of agricultural intensification in the breeding season.
Over the last 100 years, at least 50% of wetlands in the United States have been
lost due to urbanization or agriculture (Whigham 1999). The Clean Water Act specifies
that artificial wetlands can be created to offset destruction or deterioration of natural sites
(EPA 2005). One report estimated that approximately 74% of wetlands in the US are
constructed (Brown 1994), that percentage is likely higher now. Over the last 50 years,
created wetlands have played an important role in supporting biodiversity and ecosystem

8
services in the United States (Alphin and Posey 2000, Moseman et al. 2004) including
avian populations (Darnell and Smith 2004). These constructed wetlands may serve as a
lifeline for threatened avian species when natural wetlands are unavailable or extensively
deteriorated. Hence, they provide an opportunity to test whether artificial wetlands can be
utilized to reduce the rapid declines of avian species. Though the construction of artificial
wetlands or restoration of wetland habitats attempts to replace biodiversity losses,
questions about the effectiveness of artificial wetlands to support avian populations
remain. This is especially true given the gaps in our understanding of temporal variation
of emergent Diptera insects from artificial wetlands.
Mismatch hypothesis
Increases in spring temperatures are shifting emergence patterns across different
aquatic insect orders and avian species have also altered their reproductive biological
timing. Unfortunately, breeding earlier is not advantageous if prey species have not
shifted to the same degree, causing mismatches between the metabolic needs of growing
chicks and food availability. Biological consequences from climate change are evident in
mismatches seen in the temporal variation of phenological responses between different
species on the trophic pyramid (Walther et al. 2002). Food availability mismatches
caused by climate change have been demonstrated in a number of species and are linked
to population declines (Gaston et al. 2009, McKinnon et al. 2012, Visser et al. 2012). The
temperature sensitivity of the phenology of emergent aquatic insects and avian species
can differ, leading to different response rates to climate change (Winder 2004).
Asynchronies have been studied in migratory bird species in Europe that have
experienced population declines from the mistiming of their arrival on the breeding

9
grounds (Post 2007, Moller et al. 2008). In aquatic ecosystems, trophic interactions are
normally regarded as strong because of the increase in consumption efficiency such as
primary producers supporting higher trophic levels compared to terrestrial ecosystems but
may be more susceptible to rising spring temperatures from climate change (Likens
2004).
Since emergent aquatic insect phenology can be highly variable within and among
species, seasons, and localities (Primack et al. 2009, Hodgson et al. 2010), it is important
to study phenological trends across sites that vary with respect to natural and
anthropogenic factors to better inform management strategies for avian species that rely
on these energy subsidies. This is especially true during energetically demanding periods
such as breeding.
The mismatch hypothesis predicts that birds will adjust their breeding biology to
exploit seasonal pulses of prey availability within their local environment (Visser et al.
1998, Stenseth & Mysterud 2002). Lack (2009) hypothesized that birds will time their
breeding so the nestling period corresponds with the period of peak food availability.
Additionally, alterations to this synchronization may have negative implications for bird
populations (Visser et al. 2011). A reduction in prey availability during a critical growth
period of their young could result in population declines (Lack 1954, Norris 1993, Both
et al. 2006, Moller et al. 2008, Low et al. 2015). A phenological decoupling between the
emergence and reproductive timing may reduce fitness and cause population declines.
Trophic mismatches in the phenology of reproductive timing and prey abundance has
been studied in pied flycatchers (Ficedula hypoleuca), great tits (Parus major) and Tree
swallows (Nussey et al. 2005, Both et al. 2006, Paquette & Dunn 2011). One study tested

10
the mismatch hypothesis in Tree swallows and found 79% of Tree swallows bred earlier
when temperatures were higher (Dunn et al. 2011).
To my knowledge, no study has evaluated the possible trophic mismatch of
emergent aquatic insect prey phenology focused on the insect order Diptera and the
reproductive timing and chick growth in artificial and natural wetlands in agroecosystems
with intensive cultures for Tree swallows in northeast North America. Aerial insects from
aquatic and terrestrial ecosystems are captured by Tree swallows and used as a food
source for egg production and to feed to their young (Winkler & Allen 1995 Nooker et al.
2005, Ardia et al. 2006). Prey abundance affects Tree swallow breeding success and
nestling growth (Hussell and Quinney 1987, McCarty 2002). However, Tree swallows
may exhibit dietary flexibility and opportunistically feed on terrestrial versus aquatic
origin prey if it is more readily available. A consequence of the switch to terrestrial prey
is a reduction in prey quality. Recent research has shown that Tree swallow females will
travel from their central place “nest box” to forage and prefer to forage in aquatic habitats
with increased in temperature and nestling age (Morrisey 2017). This is a significant
finding as temperature is the major driver of aquatic insect emergence, and increases
from climate change may be shifting emergence patterns. Air temperature drives Tree
swallow foraging activity and subsequent breeding performance which makes this aerial
insectivore highly susceptible to climate change (Winkler et al. 2013). Tree swallows
may be more sensitive than other aerial insectivores to the accessibility of artificial
wetlands, since females select for aquatic habitats. As central place foragers, Tree
swallows will travel farther from their nest to aquatic habitats that may have, in the past,

11
provided higher emergent insect densities; the farther Tree swallows travel from their
nest, the more likely they are to select aquatic habitats (Morrissey 2017).
Tree swallows are typically a single-brood species in North America and this may
make their population more sensitive to climate change, as they have one chance to time
their reproduction with high-quality food availability compared to birds that have
multiple broods. This is compounded by the fact that Tree swallows are income breeders
that rely on financing their reproductive costs from resources acquired from the
environment during the breeding period. Therefore, reproductive performance may be
particularly susceptible to trophic mismatches (Jonsson 1997, Stephens 2009) compared
to multi-brood species or capital foragers that rely on their internal reserves for
reproduction.
Temporal mismatches between high-quality prey and egg laying and chick
hatching may pose a potential threat to Tree swallow populations because it may reduce
the number of young, they fledge. Population growth rates in short-lived birds with larger
clutch sizes are typically driven by fecundity such as the number of broods per year and
clutch size. This suggests that in Tree swallow fecundity rates surrounding breeding
success might be important drivers of population dynamics. However, a recent study
found poor fledge success was an important driver of population declines in Tree
swallows (Saether & Bakke 2000).
The phenological link between emergent aquatic insect focusing on Diptera insect
order from artificial and natural wetlands in agroecosystems from intensive cultures and
Tree swallow reproductive biological timing was explored to evaluate whether artificial
wetlands can mitigate declining aerial insectivores. I evaluated whether Tree swallow

12
reproduction matched prey availability of aquatic insects focusing on Diptera insect order
emerging from artificial and natural wetlands in northeast Pennsylvania.
Each of the wetlands were bordered on one side by intensive cultures (soybeans
and maize). My study had four main objectives: 1) determine the dates of Tree swallow
egg laying and chick hatching, 2) determine peak emergence rates for aquatic insects
particularly Diptera prey emerging from the wetlands 3) evaluate whether Tree swallow
egg laying and chick hatching coincided with peak Diptera emergence 4) Evaluate
emergent insect productivity and specifically aquatic Diptera productivity across each site
(artificial versus natural wetland) 4) determine a baseline nest box occupancy rate of Tree
swallows breeding in northeast Pennsylvania at each site (artificial versus natural
wetland) to contribute to future conversation studies. I predicted that Tree swallows
would time egg-laying with peak Diptera emergence, and emergent productivity will be
similar in artificial and natural wetland sites. If true, then artificial wetlands, even those
bordered by intensive cultures may mitigate declines in Tree swallow populations.
METHODS
Study species and model organism to test trophic mismatch
Tree swallows are migratory aerial insectivores that breed throughout central and
northern regions of North America (Shutler et al. 2012). Adults arrive in Pennsylvania to
breed by mid-April. The typical Tree swallow clutch is five to six eggs, and one egg is
laid per day with ~14 days of incubation (Winkler et al. 2013). The parents of both sexes
feed the nestlings until they fledge after 18-25 days. Tree swallows are specialized aerial
insectivores that predominantly prey on flying insects, particularly Diptera that emerge
from wetlands (Quinney and Ankney 1985, Bellarance et al. 2018). Lower aerial insect

13
abundance is associated with reduced breeding success (Hussell and Quinney 1987) and
slower nestling growth (McCarty 2002). Tree swallow occupancy rates can be used to
study population growth rates in Tree swallows because individuals with larger clutch
sizes might be strong predictors of population growth in Tree swallows
(Saether & Bakke 2000).
Study Sites
This study was conducted in three artificial and one natural wetland located in
Pennsylvania (Figure 1) surrounded by intensive cultures (Figure 2). Since 1983, the
region has seen an increase in intensive cultures (e.g. soybean) and a decrease in
extensive cultures (e.g hay) (Figure 2). Pennsylvania lies along the border between
increasing and decreasing occupancy trends, and research on Tree swallow occupancy
rates might provide insight to best manage other aerial insectivores species in the feeding
guild that are at risk (Nebel et al. 2010).
Nest box deployment
Prior to the breeding season in February 2019, a total of 99 nest boxes spaced in
grids were deployed (Robertson & Rendell 1990). Nest boxes were placed at three
artificial wetlands Site 1 (41.047021, -76.379825) (n=30), Site 2 (41.025144, 76.722078) (n=32), Site 3 (41.090783, -76.707109) (n=21) (Figure 1). Additionally, nest
boxes were set up at Site 4, a natural wetland (41.016516, -76.260169) (n=16) (Figure 1).
The number of boxes at each site was determined by the available area surrounding each
wetland. Nest boxes were deployed on freestanding galvanized steel conduit pipes (1.5
cm i.d.) that were placed over rebar (1.5 cm diameter) driven approximately 0.5 m into
the ground. Nest boxes were secured to the conduit pipes at a hight of ~1.5 m above the

14
ground, with the entrance facing southeast. Boxes were placed 1-3 m from the water’s
edge and 28 m apart. Vegetation around the nest box did not exceed 0.6 m and was
routinely mowed throughout the breeding season. To reduce predation by squirrels,
raccoons, and black rat snakes, cylindrical metal guards were fitted beneath the nest box.
Nest box monitoring
Nest boxes were visited every three days during the nest building stage in late
April. In May, when majority of nests were close to completion, they were visited daily
to determine occupancy, clutch initiation (laying date of the first egg), clutch size, and
number of chicks hatched. Since weight and body condition are the two most important
predictors of Tree swallow fledging success and survival in natural systems, chicks were
weighed and body condition recorded. Chicks were individually identified by a unique
combination of nail clips. Nails were periodically clipped again if a nail grew out.
Beginning on hatch day (day 0), chicks were measured every three days until day 12.
Weights were measured (±0.1 g) using Ohaus digital scale. On day 12, all individuals
were banded with a federal U.S. Geological Survey metal band. On day 25, all nest boxes
were checked, and number of the dead chicks recorded.
Emergence traps
We used emergence traps to quantify the timing and abundance of emerging
aquatic insects (Miller-Rushing et al. 2008). Similar emergence traps have been used to
sample wetland ecosystems (Whiles and Goldowitz 2001, MacKenzie and Kaster 2004).
Three emergence traps were placed in each wetland from 1 May through 14 June 2019.
Emergence traps similar to Cadmus (2016), were constructed from polyvinyl chloride

15
tubes (PVC) (1.5 cm i.d.) to make a pyramidal trap with a basal area of 0.35 m2. Traps
were fitted with white no-see-um fabric (mesh size of 1.1 x 1.3 mm).
Binder clips were used to create a closed tent around the frame of the trap. The
bottom perimeter was fitted with foam to allow traps to float on the surface of the water.
To prevent traps from drifting out of place and enable them to rise and fall with
fluctuating water levels, they were loosely tied to rebar that had been pushed into the
sediment. Emergence traps were checked a minimum of once per week for tears and
gaps. Traps were fitted with one-liter bottles suspended at the apex and filled with 100 ml
of 70% ethanol as a preserving agent (Davies 1984, Merrit and Cummins 1996). Aquatic
insects emerging from the water would attempt to escape at the highest point of the trap
and get euthanized in the ethanol (Davies 1984, Merrit and Cummins 1996).
Emergence traps were placed within 100 m of Tree swallow nest boxes as during
egg-laying and chick rearing, swallows will reduce the distance they forage from the nest
from 10km to <500m (Nooker et al. 2005, McCarty et al. 1999, Stapleton et al. 2006).
Since emergence rates are known to vary based on wetland microhabitat (Wellnitz et al.
2014, Alexander and Stewart 1996), Jamiez-Cuellar and Tierno de Figueroa 2005, and
Fenoglio et al. 2008), I decided to capture this variation by placing a single trap in the
following microhabitats: emergent vegetation, suspended algae, and open water.
Insects captured in traps were collected over 24-hour periods approximately each
week. The 24-hour collection period was to reduce insect mortality from the high summer
temperatures and predation from other invertebrates (Sandrock 1978, Palmen 1962,
LeSage and Harrison 1979). When the bottles were emptied, living insects present in the
traps were collected using a Sonic Technology Bugbuster®. The insect vacuum was

16
swept over the interior of the trap continuously for two to four minutes. Living insects
that did not fall into the ethanol were added to the existing collection and stored in 125
ml Nalgene HDPE bottles containing 70% ethanol until they were identified to order and
enumerated (Morrissey et al. 2016). Emergent vegetation was allowed to grow inside the
traps but trimmed back if the vegetation started to push on the mesh material (Whiles
2001).
Emergent aquatic insect prey abundance
Emergent aquatic insects were identified to order using Triplehorn and Johnson
(2005) under a dissection microscope. To determine emergent numbers for each sampling
day (Whiles 2001), abundances from all traps were standardized to 1 m2 and averaged
across the three microhabitats per site. Poor weather and equipment malfunction
prevented insect collection for Sites 2 and 4 on . Therefore, we interpolated the missing
data by regressing insect abundance and sampling date for each site then using fitting a
polynomial equation to the trend.
Statistical analysis
To determine the relationship between the timing of emergent insect abundance
during the egg laying period, the 25th percentile of clutch initiation dates (laying dates)
was used as an index of the beginning of laying for each study site population (Dunn
2011). Using the 25th percentile of clutch initiation avoids extremely early nests and
represents the beginning of laying better than the mean laying date (Dunn et al 2011).
Laying date was defined as when the first egg of the clutch was laid.
Occupancy rates were too low at two sites (Sites 3 and 4) to statistically analyze.
For Sites 1 and 2, day 12 nestling body mass was analyzed by conducting a one-way

17
analysis of covariance controlling for initial mass on day 0. To estimate the entire
emergent insect productivity and specifically aquatic Diptera productivity across each site
over the whole sampling period, I integrated daily emergence over the four-day sample
days (5 May 2019-14 June 2019). The area under the curve (AUC) per wetland site was
derived for the total number of emergent aquatic insects m-2 (densities averaged over the
three microhabitats). The AUC for the insects in the order Diptera were also separately
calculated and graphed because of their importance in Tree swallow diets, influence on
nestling body mass, and fledging success (Davies 1984).
RESULTS
Emergent aquatic insect prey abundance
Across all sampling dates and sites, I collected 11 insect orders (Table 1).
Dipterans dominated the emergent insect assemblage with 95.9% of total abundance
(Table 1), and most were Chironomids. Odondata was the second most dominant but
comprised just 1.3% of total samples (Table 1). The remaining insect orders
(Ephemoptera, Trichoptera, Hemiptera, Lepidoptera, Plecopteran, Coleoptera, and
Phasmatodea) constituted <1% of the total (Table 1). Diptera peaked on sample day three
(31 May 2019) at the Site 1 and Site 3 and peaked on sample day one (5 May 2019) at the
Site 2 and Site 4. I collected a total of 807 insects m-2 from Site 1, 420 insects m-2 from
Site 2, 632 insects m-2 from Site 3, and 447 insects m-2 from Site 4.

Synchronization between peak emergence and Tree swallow egg laying reproductive
timing

18
The average start of egg-laying (25th percentile of laying dates) varied from 8
May 2019- to 14 May 2019 in all populations (mean+/- 10 May 2019 +/- 2.6 d, n=4 site
for one year (2019)). The Tree swallow egg laying at Site 2 synchronized with the second
highest peak insect emergence abundance (Figure 5). The first peak insect emergence
was on 5 May 2019 with a peak of 187 insects (no. m-2. d-1) and the other was on 31 May
2019 with peak of 177 (no. m-2. d-1). The two peaks differed in insect emergence by 10
insects (no. m-2. d-1). During the second highest peak insect emergence on 31 May 2019,
68% of the eggs had hatched. Sixty eight percent of egg hatching coincides with this peak
in insect emergence for Site 2. There was no effect of laying date on clutch size in in Site
2 (r=0.180, N=16, P=0.102). Clutch size was constant during the breeding season.
Synchronization between peak emergence and Tree swallow chick rearing
The Tree swallow population at Site 1 did not have a synchronization between the
peak prey abundance and Tree swallow egg laying but rather with chick hatching (Figure
4). At Site 1, peak egg laying occurred on 12 May 2019 when emergent insect abundance
was increasing and reached peak insect emergence insects (no. m-2. d-1) on 31 May 2019
(Figure). The peak insect emergence recorded on 31 May 2019 coincided with the peak
in chick hatching. There was no effect of laying date on clutch size in Tree swallow
population in site two (r=0.018, N=20, P=0.621) (Figure 2) as clutch size was constant
during the breeding season.

Selection of Tree swallow reproductive timing with peak emergent aquatic insect prey
abundance

19
While peak aquatic insect prey abundance was synchronized with Tree swallow
reproductive timing at two sites there was no evidence for synchronized reproduction.
Peak aquatic insect prey abundance synchronization with reproduction did not have a
significant effect on clutch size and chick growth. Results of the nested ANOVA revealed
there was no significant difference in mean chick mass at day 12, (F24=1.575, P=0.225)
between either peak aquatic insect prey abundance at egg laying or chick rearing
controlling for the effect of initial mass on day 0. Initial mass on day 0 was not a
significant covariate (F24=0.2734, P=0.606.)
No synchronization between peak aquatic insect emergence and Tree swallow
reproductive timing
At the Site 4, the 25th percentile of egg laying occurred on 22 May 2019 when
emergent insect abundance decreased from 547 insects (no. m-2. d-1) on 5 May 2019 to 5
insects (no. m-2. d-1) on 24 May 2019 (Figure 7). This peak coincided with peak chick
hatching at Site 4. At Site 3, the 25th percentile of egg laying occurred on 18 May 2019
when emergent insect abundance increased from 17 insects (no. m-2. d-1) on 5 May 2019
to 27 insects on 24 May 2019 (no. m-2. d-1) (Figure 7). At Site 3, a peak aquatic insect
emergence on 6 June 2019 of 415 insects (no. m-2. d-1) (Figure 6). The peak aquatic insect
emergence recorded on 6 June 2019 at Site 3 did not coincide with peak egg laying or
chick hatching (Figure 6).

DISCUSSION
Synchronization between peak emergence and Tree swallow reproductive timing
Emergence patterns of aquatic insects are shifting earlier to keep pace with the
increasing spring temperatures. Avian species that use photoperiod rather than

20
temperature as a cue for reproductive timing may miss opportunities to exploit seasonal
prey availability. Pulses in insect abundance are used to finance avian reproduction but
emergent insects are sensitive to temperature rather than photoperiod. If phenological
shifts between emergent aquatic insects and Tree swallow reproduction, a trophic
mismatch may occur. The hypothesis that Tree swallow egg laying would coincide with
peak aquatic insect emergence was not supported.

Prey availability independently affects both laying date and clutch size,
suggesting a plasticity in egg laying date. For example, Davies et al. (1985)
experimentally manipulated Tree swallow food supply and advanced the egg laying date
of that population for three days (Davies et al. 1985). Supplementing food to breeding
Song sparrows (Melospiza melodia) resulted in larger clutch sizes and subsequently more
young fledging from the nest (Arcese & Smith 1988). I predicted that the emergent
aquatic prey abundance would influence the laying date of Tree swallows. However,
there was no effect of laying date on clutch size in my Tree swallow population at Site 2.
Moreover, clutch size was constant during the breeding season. Site 2 was investigated
because while results indicate that while there was a synchronization with the phenology
of the emergent aquatic insects, I cannot determine if it is constraining Tree swallow
reproductive biology. Though eggs are energetically expensive to produce, I did not find
that aquatic insect abundance constrained clutch size.

Previous studies found that timing egg laying during peak food abundance may
give some females an advantage as more energetic resources are available invest into egg
production (Wiebe 1994). However, my results suggest that females timing egg laying

21
with a peak emergent aquatic insect prey abundance did not have significantly larger
clutches. Though peak aquatic insect prey abundance was synchronized with egg laying
at Site 2 there was no evidence for synchronized reproduction. Peak aquatic insect prey
abundance synchronization with reproduction did not have a significant effect on clutch
size, and there was no effect on laying date on clutch size.

The hypothesis that Tree swallow chick hatching, and subsequent chick rearing
would coincide with peak insect emergence was partially supported. However, at Site 1,
peak egg laying occurred on 12 May 2019 when emergent aquatic insect abundance was
increasing and reached peak emergence on 31 May 2019. The peak emergent aquatic
insect emergence recorded on 31 May 2019 coincided with the peak in the chick hatching
period. Tree swallows at two of the four sites did not benefit from the brief pulse of
Dipteran abundance, by optimizing reproductive success with maximal food availability.
My results do not show evidence for synchronization between reproduction and
phenology. Peak aquatic insect prey abundance synchronization with reproduction did
not have a significant effect on chick growth. Females that timed chick hatching with
peak emergent aquatic insect prey abundance did not have significantly heavier chicks on
Day 12. My results indicate there may not be an advantage to Tree swallows timing their
reproduction to coincide with peak emergent insects. Additionally, there did not seem to
be a constraint clutch size or chick mass on day 12 based on prey availability.

At two of my four wetland study sites there was no relationship between emergent
aquatic insect prey abundance and Tree swallow reproductive timing. To date, studies
seem evenly divided in their support or refutation that Tree swallows time their

22
reproduction with peak food abundance. Dunn et. al (1999 and 2011) found no
synchronization between reproductive timing and peak food abundance, whereas
Blancher et al. (1991) and Twining et al. (2016) found a synchronization between
reproductive timing and peak food abundance in Tree swallows. At all my study sites,
food abundance generally decreased during the nestling period, as well as across the
entire breeding season. This differed from that of Winkler et. al (2001) who showed that
food abundances increased over the course of the reproductive season.
This study highlights that Tree swallow populations that attempt to optimize
reproduction based on pulses of aquatic emergent insects may be more susceptible to
trophic mismatches compared to Tree swallows who do not attempt to optimize
reproduction with peak prey availability. My results can serve as a baseline for future
trophic mismatch studies on avian species in wetlands. An important finding from my
study was that I found different reproductive biological events synchronized with the
peak aquatic insect food abundance depending on the study site.
Relationship between prey abundance and Tree swallow reproductive timing with
occupancy trends
Tree swallows are income breeders and finance their reproduction during the
breeding season. This differs from capital breeders such as capital breeding birds such as
Zebra finches (Taeniopygia guttata) and Common Eiders (Somateria mollissima) that
rely on reserves accumulated prior to the breeding season to finance reproduction.
Tree swallows mainly feed on Diptera, which was a majority of the insects emerging
from these wetlands. Differences in prey availability affects occupancy rates of Tree
swallows. If emergent insect availability were the main driver of Tree swallow

23
occupancy, then sites with higher AUC values should have greater occupancy than sites
with lower AUC. However, we found that Site 2 had the second highest occupancy rates
but the greatest AUC values for emergent insects during the pre-egg laying period. This
may be an important finding because certain avian species use current environmental
cues to make reproductive timing decisions. It is possible that Tree swallows occupying
Site 1 used environmental cues that were not measured in this study.
Limitations and future directions
This study showed that the emergence of aquatic insects from artificial and
natural wetlands in agricultural landscapes may represent an energy transfer from aquatic
to terrestrial ecosystems. The emergence patterns were generally a pulse a highly
synchronized fashion from the water. Ephemoptera, Plecoptera, and Trichoptera
emergence peaks commonly follow a staircase pattern throughout the year with valleys of
highs and lows (Morrissey 2017). Unfortunately, these bimodal patterns were not
captured because of limited sampling days. However, this study did catch the bimodal
peak in Diptera throughout the four wetland study sites. Progar et al (2009) showed that
Diptera commonly emerge during the summer in a manner similar to my results. This
contrasts with the quadratic curve patterns found in other studies (Douglas et al. 2010,
Paquette 2013). However, this pattern might be explained by the limited sampling periods
that I performed compared to other studies. The four wetland study sites were similar in
that Dipteran emergence is proportionally higher compared to Ephenmoptera, Plecoptera,
Trichoptera as with other published studies (Batzer et al. 1996). I found that Diptera
peaked in the end of May and early June. This is different to other systems that found
Diptera abundance peaked in early July (Morrissey 2017). Since 1901, the average

24
surface temperature across the contiguous 48 states has risen at an average rate of 0.14°C
per decade. My study may be revealing that Diptera emergence is occurring earlier in the
season compared to 2006 and 2008. With Dipteran’s importance to Tree swallow
reproduction this can have possible future population dynamics implications.
A future area for research would be to investigate the nest-box projects that are
relying on limited aquatic Diptera abundance food abundance as opposed to an increasing
food abundance throughout the breeding season such as in more terrestrial landscapes.
Another future direction of this study would be to capture the emergence insect
abundance from all insect orders during the post-fledging stage. Poor post fledging
survival will substantially influence recruitment success (Naef-Daenzer et al. 2001;
Gruebler et al. 2014).
My study revealed that relying solely on samples of aquatic insect prey may be
insufficient in understanding prey availability for Tree swallows. For one thing, sampling
aquatic insects will become more challenging with climate change. Climate change is
altering storm frequencies, which shifts patterns of insect emergence in wetlands though
no studies report this. I may have reported lower insect emergence due to the large
number of storm events occurred during the sampling period. Whiles et al. (2001) found
that fluctuating hydroperiods resulted in lower emergent insect densities. Furthermore,
though Tree swallows are generally aquatic insect specialists it may not be true in the
artificial and natural wetlands sites. Sweep nets, which capture both terrestrial and
aquatic aerial insects, should be used in conjunction with emergence sampling to capture
a wider range of prey availability. Capturing both parameters may strengthen evidence
for the importance of synchronization of Tree swallow reproduction with insect

25
phenology. This information is crucial in identifying whether artificial wetlands bordered
by intensive cultures are suitable habitats for avian aerial insectivores such as Tree
swallows. Moreover, it will enable researchers to test for the presence of ecological traps
at the very wetlands designed to support biodiversity in this ever-changing anthropogenic
landscape.

29

LITERATURE CITED
Baines, D., Wilson, I. A., & Beeley, G. (1996). Timing of breeding in black grouse
(Tetrao tetrix) and capercaillie (Tetrao urogallus) and distribution of insect food for the
chicks. Ibis, 138(2), 181-187.
Batzer, D. P., & Wissinger, S. A. (1996). Ecology of insect communities in nontidal
wetlands. Annual review of entomology, 41, 75.
Bartrons, M., Sardans, J., Hoekman, D., & Peñuelas, J. (2018). Trophic transfer from
aquatic to terrestrial ecosystems: a test of the biogeochemical niche
hypothesis. Ecosphere, 9(7), e02338.
Bayoh, M. N., & Lindsay, S. W. (2003). Effect of temperature on the development of the
aquatic stages of Anopheles gambiae sensu stricto (Diptera: Culicidae). Bulletin of
entomological research, 93(5), 375-381.
Bentley, G. E., Audage, N. C., Hanspal, E. K., Ball, G. F., & Hahn, T. P. (2003).
Photoperiodic response of the hypothalamo‐pituitary‐gonad axis in male and female
canaries, Serinus canaria. Journal of Experimental Zoology Part A: Comparative
Experimental Biology, 296(2), 143-151.
Blancher, P. J., & McNicol, D. K. (1991). Tree swallow diet in relation to wetland
acidity. Canadian Journal of Zoology, 69(10), 2629-2637.
Both, C., Bouwhuis, S., Lessells, C. M., & Visser, M. E. (2006). Climate change and
population declines in a long-distance migratory bird. Nature, 441(7089), 81-83.
Edmondson, W. T., & Vinberg, G. G. (1971). Manual on methods for the assessment of
secondary productivity in fresh waters.
Dunn, P. O., & Winkler, D. W. (1999). Climate change has affected the breeding date of
tree swallows throughout North America. Proceedings of the Royal Society of London.
Series B: Biological Sciences, 266(1437), 2487-2490.
Dunn, P. O., Winkler, D. W., Whittingham, L. A., Hannon, S. J., & 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(2), 450-461.
Durant, J. M., Hjermann, D. O., Ottersen, G., & Stenseth, N. C. (2007). Climate and the
match or mismatch between predator requirements and resource availability. Climate
research, 33(3), 271-283.

30

Finch, D. M. (1990). Effects of predation and competitor interference on nesting success
of House Wrens and Tree Swallows. The Condor, 92(3), 674-687.
Ghilain, A., & Bélisle, M. (2008). Breeding success of tree swallows along a gradient of
agricultural intensification. Ecological Applications, 18(5), 1140-1154.
Hussell, D. J., & Quinney, T. E. (1987). Food abundance and clutch size of Tree
Swallows Tachycineta bicolor. Ibis, 129, 243-258.
IUCN, S. (2018). Amphibian Specialist Group (2015) Lithobates catesbeianus. The
IUCN Red List of Threatened Species 2015: e. T58565A53969770.
Jones, J. (2003). Tree swallows (Tachycineta bicolor): a new model organism?. The
Auk, 120(3), 591-599.
Jonsson, K. I. (1997). Capital and income breeding as alternative tactics of resource use
in reproduction. Oikos, 57-66.
Jonsson, M., Hedstrom, P., Stenroth, K., Hotchkiss, E. R., Vasconcelos, F. R., Karlsson,
J., & Byström, P. (2015). Climate change modifies the size structure of assemblages of
emerging aquatic insects. Freshwater Biology, 60(1), 78-88.
Kneib, R. T. (1984). Patterns of invertebrate distribution and abundance in the intertidal
salt marsh: causes and questions. Estuaries, 7(4), 392-412.
Lack, D. (1954). The natural regulation of animal numbers. The Natural Regulation of
Animal Numbers.
Møller, A. P., Fiedler, W., & Berthold, P. (2006). Birds and climate change. Academic
Press.
LeSage, L., & AD, H. (1979). Improved traps and techniques for the study of emerging
aquatic insects.
Likens, G. E., & Bormann, F. H. (1974). Linkages between terrestrial and aquatic
ecosystems. BioScience, 24(8), 447-456.
Low, M., Arlt, D., Pärt, T., & Oberg, M. (2015). Delayed timing of breeding as a cost of
reproduction. Journal of Avian Biology, 46(4), 325-331.
McCarty, J. P., & Winkler, D. W. (1999). Foraging ecology and diet selectivity of tree
swallows feeding nestlings. The Condor, 101(2), 246-254.

31

McCarty, J. P. (2002). The Number of Visits to the Nest by Parents Is an Accurate
Measure of Food Delivered to Nestlings in Tree Swallows. Journal of Field Ornithology,
9-14.
McCarty, J. P., & Secord, A. L. (1999). Nest-building behavior in PCB-contaminated tree
swallows. The Auk, 116(1), 55-63.
Nebel, S., Mills, A., McCracken, J., & Taylor, P. (2010). Declines of aerial insectivores
in North America follow a geographic gradient. Avian Conservation and Ecology, 5(2).
Neckles, H.A., Murkin, H.R., & Cooper, J.A. (1990). Influences of seasonal flooding on
macroinvertebrate abundance in wetland habitats. Freshwater biology, 23(2), 311-322.
Nooker, J. K., Dunn, P. O., & Whittingham, L. A. (2005). Effects of food abundance,
weather, and female condition on reproduction in tree swallows (Tachycineta
bicolor). The Auk, 122(4), 1225-1238.
Norris, K., & Evans, M. R. (2000). Ecological immunology: life history trade-offs and
immune defense in birds. Behavioral Ecology, 11(1), 19-26.
Nussey, D. H., Postma, E., Gienapp, P., & Visser, M. E. (2005). Selection on heritable
phenotypic plasticity in a wild bird population. Science, 310(5746), 304-306.
Paquette, S. R., Garant, D., Pelletier, F., & Bélisle, M. (2013). Seasonal patterns in tree
swallow prey (Diptera) abundance are affected by agricultural intensification. Ecological
Applications, 23(1), 122-133.
Palmen, E. (1962). Studies on the ecology and phenology of the Chironomids (Dipt.) of
the Northern Baltic. 1. Allochironomus crassiforceps, 137-168.
Post, E., & Forchhammer, M. C. (2008). Climate change reduces reproductive success of
an Arctic herbivore through trophic mismatch. Philosophical Transactions of the Royal
Society B: Biological Sciences, 363(1501), 2367-2373.
Quayle, W. C., Peck, L. S., Peat, H., Ellis-Evans, J. C., & Harrigan, P. R. (2002).
Extreme responses to climate change in Antarctic lakes.(Climate
Change). Science, 295(5555), 645-646.
Quinney, T. E., & Ankney, C. D. (1985). Prey size selection by tree swallows. The
Auk, 102(2), 245-250.
Ratte, H. T. (1984). Temperature and insect development in environmental physiology
and biochemistry of insects (pp. 33-66). Springer, Berlin, Heidelberg.

32

Rosenberg, K. V., Dokter, A. M., Blancher, P. J., Sauer, J. R., Smith, A. C., Smith, P. A.,
& Marra, P. P. (2019). Decline of the North American avifauna. Science, 366(6461), 120124.
Rioux Paquette, S., Pelletier, F., Garant, D., & Belisle, M. (2014). Severe recent decrease
of adult body mass in a declining insectivorous bird population. Proceedings of the Royal
Society B: Biological Sciences, 281(1786), 20140649.
Sandrock, F. (1978). A Comparison of the Insects Emerging from Two Brooks in the
Region of Schlitz (Breitenbach and Rohrwiesenbach 1970-1971). Arch. Hydrobiol
(Ger.)/Suppl, 54(3), 328.
Stapleton, M. K., & Robertson, R. J. (2006). Female tree swallow home-range
movements during their fertile period as revealed by radio-tracking. The Wilson Journal
of Ornithology, 118(4), 502-507.
Tammaru, T., & Haukioja, E. (1996). Capital breeders and income breeders among
Lepidoptera: consequences to population dynamics. Oikos, 561-564.
Shutler, D., Hussell, D. J., Norris, D. R., Winkler, D. W., Bonier, F., Belisle, M., ... &
Thorpe, P. A. (2012). Spatiotemporal patterns in nest box occupancy by Tree Swallows
across North America. Avian Conservation & Ecology, 7(1), 3.
Twining, C. W., Brenna, J. T., Lawrence, P., Shipley, J. R., Tollefson, T. N., & Winkler,
D. W. (2016). Omega-3 long-chain polyunsaturated fatty acids support aerial insectivore
performance more than food quantity. Proceedings of the National Academy of
Sciences, 113(39), 10920-10925.
Wegge, P., & Kastdalen, L. (2008). Habitat and diet of young grouse broods: resource
partitioning between Capercaillie (Tetrao urogallus) and Black Grouse (Tetrao tetrix) in
boreal forests. Journal of Ornithology, 149(2), 237.
Whiles, M. R., & Goldowitz, B. S. (2001). Hydrologic influences on insect emergence
production from central Platte River wetlands. Ecological Applications, 11(6), 18291842.
Winder, M., & Schindler, D. E. (2004). Climate change uncouples trophic interactions in
an aquatic ecosystem. Ecology, 85(8), 2100-2106.
Winkler, D. W., Luo, M. K., & Rakhimberdiev, E. (2013). Temperature effects on food
supply and chick mortality in tree swallows (Tachycineta bicolor). Oecologia, 173(1),
129-138.

33

Appendix 1. Institutional Animal Care and Use Approval

Table 1. Raw data of emergent insects captured by emergent traps

Appendix 1. Raw data

34

35

Table 2. Chick Growth data for wetland sites

Location

Date

Box

Growth Day

Foot

Nail #

Chk (g)

Lurowist

5/31/19

1

0

R

4

3.2

Lurowist

5/31/19

1

0

R

1

2.34

Lurowist

5/31/19

1

0

R

2

3.24

Lurowist

5/31/19

1

0

L

4

3.03

Lurowist

5/31/19

1

0

L

2

3.03

Lurowist

5/31/19

1

0

L

3

2.29

Lurowist

6/3/19

1

3

L

3

7.51

Lurowist

6/3/19

1

3

R

4

9.55

Lurowist

6/3/19

1

3

L

2

10

Lurowist

6/3/19

1

3

R

2

9.75

Lurowist

6/3/19

1

3

R

1

7.38

Lurowist

6/3/19

1

3

L

4

9.12

Lurowist

6/6/19

1

6

L

2

15.82

Lurowist

6/6/19

1

6

R

3

13.68

Lurowist

6/6/19

1

6

R

4

16.2

Lurowist

6/6/19

1

6

L

4

18.9

Lurowist

6/6/19

1

6

R

2

17.33

Lurowist

6/6/19

1

6

R

3

14.82

Lurowist

6/8/19

1

9

R

4

21.5

Lurowist

6/8/19

1

9

L

2

23.85

Lurowist

6/8/19

1

9

R

1

20.34

Lurowist

6/8/19

1

9

R

2

20.6

Lurowist

6/8/19

1

9

L

4

22.36

Lurowist

6/8/19

1

9

R

3

21.14

Lurowist

6/12/19

1

12

R

4

20.32

Lurowist

6/12/19

1

12

R

1

21.26

Lurowist
Lurowist

6/12/19
6/12/19

1
1

12
12

R
L

2
4

22.71
22.24

Lurowist

6/12/19

1

12

R

3

24.12

Lurowist

6/12/19

1

12

L

2

22.88

Lurowist

5/28/19

2

0

R

4

2.55

Lurowist

5/28/19

2

0

R

2&4

2.57

Lurowist

5/28/19

2

0

L

2

2.3

36

Lurowist

5/28/19

2

0

L

3

1.96

Lurowist

5/31/19

2

3

L

3

Lurowist

5/31/19

2

3

R

2&4

Lurowist

5/31/19

2

3

L

3

Lurowist

6/3/19

2

6

L

3

13.95

Lurowist
Lurowist

6/3/19
6/3/19

2
2

6
6

R
L

2&4
2

16.74
15.32

Lurowist

6/3/19

2

6

R

4

15.63

Lurowist

6/6/19

2

9

L

3

15.92

Lurowist

6/6/19

2

9

R

2&4

21.31

Lurowist

6/6/19

2

9

L

2

17.6

Lurowist

6/6/19

2

9

R

4

19.3

Lurowist

6/9/19

2

12

R

2&4

23.92

Lurowist

6/9/19

2

12

L

2

24.29

Lurowist

6/9/19

2

12

R

4

25.1

Lurowist

6/9/19

2

12

L

3

21.5

Lurowist

6/5/19

6

1

R

4

3.15

Lurowist

6/5/19

6

1

R

2

3.94

Lurowist

6/5/19

6

1

L

4

2.71

Lurowist

6/5/19

6

1

L

3

4.05

Lurowist

6/10/19

6

6

R

4

19.73

Lurowist

6/10/19

6

6

R

2

19.44

Lurowist

6/10/19

6

6

L

4

18.15

Lurowist

6/10/19

6

6

L

3

19.28

Lurowist

6/13/19

6

10

R

4

23.87

Lurowist

6/13/19

6

10

L

4

22.3

Lurowist

6/13/19

6

10

L

3

21.88

Lurowist

6/13/19

6

10

R

2

22.52

Lurowist

5/29/19

26

0

R

4

1.59

Lurowist

5/29/19

26

0

L

4

1.85

Lurowist

5/29/19

26

0

L

1

1.74

Lurowist

5/29/19

26

0

R

2

1.75

Lurowist

6/2/19

26

3

R

2

7.3

Lurowist

6/2/19

26

3

R

4

7.77

Lurowist

6/2/19

26

3

L

4

8.54

Lurowist

6/2/19

26

3

L

1

7.33

Lurowist

6/4/19

26

6

R

2

13.67

Lurowist

6/4/19

26

6

R

4

13.77

37

Lurowist
Lurowist

6/4/19
6/4/19

26
26

6
6

L
L

1
4

13.3
15.22

Lurowist

6/7/19

26

6

L

1

20.21

Lurowist

6/7/19

26

6

R

4

21.2

Lurowist

6/7/19

26

6

L

4

21.33

Lurowist

6/9/19

26

12

R

2

21.98

Lurowist

6/9/19

26

12

R

4

23.47

Lurowist

6/9/19

26

12

L

1

23.3

Lurowist

6/9/10

26

12

L

4

23.8

Lurowist

5/30/19

8

0

R

2

1.2

Lurowist

5/30/19

8

0

R

1

1.31

Lurowist

5/30/19

8

0

R

4

1.27

Lurowist

6/11/19

29

13

L

4

21.88

Lurowist

6/11/19

29

13

L

2

20.32

Lurowist

6/11/19

29

13

R

2

21.86

Lurowist

6/11/19

29

13

R

4

22.03

Lurowist

6/11/19

29

13

R

3

22.2

Lurowist

5/22/19

24

0

R

3

1.55

Lurowist

5/22/19

24

0

R

4

1.73

Lurowist

5/22/19

24

0

L

4

1.43

Lurowist

6/1/19

24

9

R

4

21.9

Lurowist

6/1/19

24

9

R

2

21.66

Lurowist

6/1/19

24

9

Lurowist

5/29/19

7

0

L

2

1.7

Lurowist

5/29/19

7

0

L

4

1.75

Lurowist

5/29/19

7

0

R

2

1.57

Lurowist

5/30/19

7

0

R

2,1

2.04

Lurowist

5/30/19

7

0

L

3

1.8

Tanner

6/1/19

5

6

R

2

20.75

Tanner

6/1/19

5

6

R

4

21.78

Tanner

6/1/19

5

6

R

3

21.93

Tanner

6/2/19

9

3

R

4

7.32

Tanner

6/2/19

9

3

R

3

7.54

Tanner

6/2/19

9

3

L

4

6.36

Tanner

6/2/19

9

3

R

2

6.83

Tanner

6/2/19

9

3

L

2

7.52

Tanner

6/2/19

9

2

L

1

5.63

Tanner

6/6/19

9

7

R

4

13.71

22.46

38

Tanner

6/6/19

9

7

L

2

13.47

Tanner

6/6/19

9

7

R

2

13.04

Tanner

6/6/19

9

7

R

3

13.72

Tanner

6/6/19

9

6

L

1

10.71

Tanner

6/8/19

9

9

R

2

21.78

Tanner

6/8/19

9

9

R

3

22.03

Tanner

6/8/19

9

9

L

2

29.69

Tanner

6/8/19

9

8

L

1

18.57

Tanner

6/8/19

9

9

L

4

18.98

Tanner

6/8/19

9

9

R

4

20.89

Tanner

6/11/19

9

12

L

2

20.81

Tanner

6/11/19

9

12

R

3

20.59

Tanner

6/11/19

9

12

R

4

20.1

Tanner

6/11/19

9

12

R

2

20.45

Tanner

6/11/19

9

12

L

4

20.38

Tanner

6/11/19

9

11

L

1

18.48

Tanner

6/2/19

11

6

L

1

8.74

Tanner

6/2/19

11

6

R

3

16.33

Tanner

6/2/19

11

6

L

2

16.63

Tanner

6/2/19

11

6

L

3

14.35

Tanner

6/2/19

11

6

R

4

14.18

Tanner

6/2/19

11

6

R

2

13.38

Tanner

6/2/19

11

6

R

1

15.72

Tanner

6/5/19

11

9

L

2

20.9

Tanner

6/5/19

11

9

R

3

20.22

Tanner

6/5/19

11

9

R

2

17.33

Tanner

6/5/19

11

9

R

4

19.13

Tanner

6/5/19

11

9

R

1

19.95

Tanner

6/5/19

11

9

L

3

17.82

Tanner

6/5/19

11

9

L

1

11.81

Tanner

6/9/19

11

12

R

3

20.83

Tanner

6/9/19

11

12

R

2

20.81

Tanner

6/9/19

11

12

R

1

22.03

Tanner

6/9/19

11

12

R

4

21.81

Tanner

6/9/19

11

12

L

2

23.2

Tanner

6/9/19

11

12

L

1

17.73

Tanner

6/9/19

11

12

L

3

20.62

Tanner

6/1/19

6

1

R

2

2.5

39

Tanner

6/1/19

6

1

R

3

2.94

Tanner

6/3/19

6

3

R

2

Tanner

6/3/19

6

3

R

3,1

Tanner

6/3/19

6

3

R

2,4

Tanner

6/3/19

6

2

L

Tanner

6/6/19

6

6

UNMARKED

Tanner

6/6/19

6

6

L

Tanner

6/6/19

6

6

R

2,4

13.61

Tanner

6/6/19

6

6

R

3,1

15.44

Tanner

6/6/19

6

6

R

2

12.94

Tanner

6/13/19

6

12

R

3

21.85

Tanner

6/13/19

6

12

R

2

21.95

Tanner

6/13/19

6

12

L

2

18.98

Tanner

6/13/19

6

12

L

4

19.93

Tanner

6/13/19

6

12

R

Tanner

6/5/19

12

0

Tanner

6/5/19

Tanner

4
5.39
4

2,4

7.7

22.08

NOT CLIPPED

0.9

12

NOT CLIPPED

1.6

6/5/19

12

L

1

1.66

Tanner

6/14/19

12

9

R

4

19.59

Tanner

6/14/19

12

9

R

2

22.03

Tanner

6/14/19

12

9

L

4

19.64

Tanner

5/23/19

5

0

R

2

1.93

Tanner

5/23/19

5

0

R

4

1.81

Tanner

5/24/19

5

1

R

2

2.74

Tanner

5/24/19

5

1

R

4

2.42

Tanner

5/24/19

5

0

R

3

1.72

Tanner

5/25/19

5

2

R

3

4.17

Tanner

5/25/19

5

2

R

4

4.5

Tanner

5/25/19

5

2

R

2

4.19

Tanner

5/26/19

5

3

R

4

8.51

Tanner

5/26/19

5

3

R

2

8.17

Tanner

5/27/19

5

3

R

3

8.85

Tanner

5/30/19

5

6

R

2

18.3

Tanner

5/30/19

5

6

R

1

17.27

Tanner

5/30/19

5

6

R

4

18.62

Tanner

5/27/19

10

L

4

5.95

Tanner

5/27/19

10

L

2

6.1

Tanner

5/27/19

10

R

3

6.08

40

Tanner

5/27/19

10

R

4

6.2

Tanner

5/25/19

4

0

R

4

1.51

Tanner

5/25/19

4

0

L

4

1.51

Tanner

5/25/19

4

0

L

1

1.46

Tanner

5/26/19

4

0

R

2

2.23

Tanner

5/26/19

4

0

R

1

2.13

Tanner

5/26/19

4

0

L

2

1.82

Tanner

5/31/19

4

6

UNMARKED

Tanner

5/31/19

4

6

R

1

13.33

Tanner

5/31/19

4

6

L

1

13.69

Tanner

5/31/19

4

6

R

4

15.45

Tanner

5/31/19

4

6

L

2

14.2

Tanner

5/27/19

11

0

R

4

2.07

Tanner

5/27/19

11

0

L

4

1.93

Tanner

5/27/19

11

0

R

1

2.57

Tanner

5/27/19

11

0

R/L

2

2.47

Tanner

5/27/19

11

0

R

3

2.47

Tanner

5/27/19

11

0

L

3

2.1

Tanner

5/30/19

11

3

R

4

5.73

Tanner

5/30/19

11

3

L

4

5.43

Tanner

5/30/19

11

3

R

1

6.78

Tanner

5/30/19

11

3

L

2

7.04

Tanner

5/30/19

11

3

R

3

7.41

Tanner

5/30/19

11

3

L

3

5.39

Tanner

5/30/19

11

3

R/L

2

5.38

Tanner

5/30/19

9

0

R

4

1.64

Tanner

5/30/19

9

0

L

4

1.39

Tanner

5/30/19

9

0

L

2

1.7

Tanner

5/30/19

9

0

R

2

1.71

Tanner

5/30/19

9

0

R

3

1.61

Tanner

6/1/19

18

1

R

4

2.63

Tanner

6/1/19

18

1

R

2

2.07

Tanner

6/1/19

18

1

R

3

2.46

Tanner

6/1/19

18

0

L

2

1.97

Tanner

6/2/19

18

0

NOT CLIPPED

Tanner

6/2/19

5

R

1

24.6

Tanner

6/2/19

5

R

4

23.55

Tanner

6/2/19

5

R

1

22.69

11.78

2.35

41

Fry

5/30/19

1

R

2

2.24

Fry

5/30/19

1

R

4

2.66

Fry

5/30/19

1

L

4

2.86

Table 3. Chick Growth data for chick growth days 0,3,6,8, and 12 days for wetland sites
Tanner, Lurowist, and Fry

42

Appendix 2: Interpolated data

Table 2. Interpolated data Open water and Algae microhabitat type for sample day 3

Site
4

Micro
Habitat Type
Open water

2

Algae

Interpolated data
Sample
Abundance
Day
3
238
3

2

Polynomial Equation
y = 0.14x2 - 12204x +
3E+08
y = 0.0257x2 - 2241.4x +
5E+07

Table. 2 Interpolated data for emergent insects captured by emergent traps for two
sites (4&2). The microhabitat types are open water and algae while the sample days
are both 3. The abundances of insects that was derived using the polynomial
equations are listed in the table above.

43

Appendix 3 . Figures Depicting Finalized Results

Figure 1. Map of study sites in Columbia and Montour
County, Pennsylvania

38

30,000
25,000

Acres

20,000
15,000

Soybean
Corn

10,000

Hay

5,000
0
1983

1993

2003

2013

2017

YEAR

Figure 2. United States Department of Agriculture; National Agriculture Statistics
Service 1983-2017 in Columbia County, Pennsylvania. Intensive crops harvested: corn
and soybeans and extensive cultures harvest: hay.

39

EMERGENT INSECTS
DENSITY (no.m-2d-1)

600

450
400
350
300
250
200
150
100
50
0

DENSITY (no.m-2d-1)

DIPTERA

Emergent insects
Diptera

500
400
300
200
100
0

125

144
157
JULIAN DATE

125

165

DENSITY (no.m-2d-1)

DENSITY (no.m-2d-1)

700
600
500
400
300
200
100
0
125

151
153
JULIAN DATE

165

144
157
JULIAN DATE

165

200
180
160
140
120
100
80
60
40
20
0
125

151

153

165

JULIAN DATE

Figure 3. Daily numerical density averaged over three microhabitats, for all emergent insects
and for the Dipteran order, from 1 (top left), 2 (top right), 3 (bottom left), and 4 (bottom right)
wetlands, 2019. Drop lines indicate the Julian date of each sample. Total numerical emergence
is equal to the area under the curve.

40

Figure. 4 Relationship between egg hatching date and insect abundance for Site
1 wetland site. Depicted are the total number of Tree swallow eggs and newly
hatched chicks in all nest boxes, and the number of insect (individuals per m2 per
day) captured in emergence traps on each date.

41

DATE (MONTH/DAY/YEAR)

Figure. 5 Relationship between egg hatching date and insect abundance for Site
2 wetland site. Depicted are the total number of Tree swallow eggs and newly
hatched chicks in all nest boxes, and the number of insect (individuals per m2
per day) captured in emergence traps on each date.

42

Figure. 6 Relationship between egg hatching date and insect abundance for Site
3 wetland site. Depicted are the total number of Tree swallow eggs and newly
hatched chicks in all nest boxes, and the number of insect (individuals per m2
per day) captured in emergence traps on each date.

5/11/19
5/13/19

5/13/19

per day) captured in emergence traps on each date.

4 wetland site. Depicted are the total number of Tree swallow eggs and newly

hatched chicks in all nest boxes, and the number of insect (individuals per m2
6/26/19

6/24/19

6/22/19

6/26/19

6/24/19

6/22/19

6/20/19

6/18/19

7

6/20/19

6/16/19

6/14/19

6/12/19

6/10/19

6/8/19

6/6/19

6/4/19

6/2/19

5/31/19

5/29/19

5/27/19

5/25/19

5/23/19

5/21/19

5/19/19

5/17/19

8

6/18/19

6/16/19

6/14/19

6/12/19

6/10/19

6/8/19

6/6/19

6/4/19

6/2/19

5/31/19

5/29/19

5/27/19

5/25/19

5/23/19

5/21/19

5/19/19

5/17/19

5/15/19

5/9/19

5/11/19
5/15/19

5/7/19

5/9/19

0
5/5/19

5

5/7/19

10
5/3/19

15

5/5/19

20

5/3/19

25
5/1/19

0

5/1/19

43

9

# OF INSECTS
# OF EGGS
# OF CHICKS

6

5

4

3

2

1

Figure. 7 Relationship between egg hatching date and insect abundance for Site