THE ROLE OF PREALTERNATE MOULT IN NONBREEDING PERIOD CARRYOVER EFFECTS IN NEOTROPICAL MIGRATORY SONGBIRDS
by
SHAE TURNER
B.Sc., University of the Fraser Valley, 2017
A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF MASTER OF SCIENCE ENVIRONMENTAL SCIENCE
in the Department of Biological Sciences
Thesis examining committee:
Matt Reudink (PhD), Thesis Supervisor and Professor, Biological Sciences, Thompson
Rivers University
Mateen Shaikh (PhD), Committee Member and Assistant Professor, Statistics and Data
Science, Thompson Rivers University
Emily Studd (PhD), Committee Member and Assistant Professor, Biological Sciences,
Thompson Rivers University
Christopher Tonra (PhD), Committee Member and Associate Professor, School of
Environment and Natural Resources, The Ohio State University
Maggie MacPherson (PhD), External Examiner and Assistant Professor, Biological
Sciences, Western Illinois University

August 2024
Thompson Rivers University

Shae Turner, 2024

ii
Thesis Supervisor: Professor Matt Reudink
ABSTRACT
Feathers serve myriad functions, including flight, thermoregulation, and
communication. Because feathers are keratin structures, they cannot be repaired once
damaged so birds must shed and regrow new feathers each year through a process called
moult. Despite the importance of moult for feather function, it is one of the least
understood events in the annual cycle of migratory birds. For Nearctic-Neotropical
migratory songbirds that moult twice annually, little is known about the ecology of their
prealternate moult, which typically occurs on the stationary nonbreeding grounds prior to
pre-breeding (northward) migration. Improving our understanding of prealternate moult
is fundamental for identifying how it interacts with other life history stages, breeding and
migration, across the annual cycle, knowledge that is critical for determining periods of
the year that limit declining bird populations. Here, I provide a detailed quantification of
the timing, patterns, and intensity of prealternate moult for six warbler species (Family:
Parulidae) on their stationary nonbreeding grounds in Jamaica. I demonstrate that
prealternate moult is common for Northern Waterthrush (Parkesia noveboracensis),
Black-and-white Warbler (Mniotilta varia), American Redstart (Setophaga ruticilla),
Northern Parula (Setophaga americana), and Prairie Warbler (Setophaga discolor), and
for most species increases in frequency and intensity across the nonbreeding period. I
confirm the occurrence of prealternate moult in some Ovenbirds (Seiurus aurocapilla). I
then assess the role of prealternate moult in carry-over effects from the nonbreeding
period to the timing of migration departure for the Black-and-white Warbler, Northern
Waterthrush, and American Redstart. First, I assessed the influence of body condition on
the timing and intensity of prealternate moult in distinct habitats that differ in moisture
regime: dry second-growth scrub forest and wet mangrove. I analyzed stable-carbon
isotope ratios (δ13C) from birds’ red blood cells to support my understanding of relative
habitat quality for the study species. Using the Motus Wildlife Tracking System, I tracked
precise dates of migration departure so I could evaluate the influence of habitat quality,
body condition, and prealternate moult on departure timing. Moult intensity was

iii
generally higher in young birds but its relationship with body condition varied among
species, indicating that body condition may not be the most important driver of
prealternate moult phenology. For the American Redstart, the effect of age on moult
varied with habitat type, suggesting that birds with the poorest quality feathers, associated
with poorer quality, open habitat, require a more intense moult than older birds and those
in higher quality habitat. I found the first potential evidence of a carry-over effect from
prealternate moult to departure timing in a parulid warbler, the American Redstart. Male
American Redstarts with later, high-intensity moult departed on migration later,
suggesting that prealternate moult can represent an important seasonal interaction in
some birds. Ultimately, this study is an important step in beginning to understand the role
of prealternate moult in nonbreeding period carry-over effects. To that end, it will be
crucial for researchers in the global north to establish equitable partnerships with local
neotropical researchers to aid in furthering prealternate moult research on the
nonbreeding grounds of many migratory birds.
Keywords: migratory birds, nonbreeding ecology, prealternate moult, carry-over effects

iv
TABLE OF CONTENTS
ABSTRACT……………………………………………………………………………... ii
TABLE OF CONTENTS………………………………………………………………... iv
ACKNOWLEDGEMENTS……………………………………………………………... vi
LIST OF FIGURES……………………………………………………………………... ix
LIST OF TABLES……………………………………………………………………... xiii
CHAPTER 1: INTRODUCTION……………………………………………………...... 1
Understanding moult is important for conserving migratory birds……… 2
Study system……………………………………………………………... 5
Research goal………………………………………………………….... 10
LITERATURE CITED………………………………………………………….. 11
CHAPTER 2: QUANTIFYING PREALTERNATE MOULT TIMING, PATTERNS,
AND INTENSITY IN SIX NEARCTIC-NEOTROPICAL MIGRATORY
WARBLERS………………………………………….………………………………… 16
ABSTRACT…………………………………………………………………….. 16
INTRODUCTION……………………………………………………………… 16
METHODS……………………………………………………………………... 20
Study species……………………………………………………………. 20
Study site………………………………………………………………... 20
Field methods……………………………………………………. …….. 21
Quantifying moult intensity and prevalence……………………………. 23
Statistical analyses……………………………………………………… 24
RESULTS……………………………………………………………………….. 26
Species accounts………………………………………………………... 29
Ovenbird………………………………………………………... 29
Northern Waterthrush…………………………………………… 29
Black-and-white Warbler…………………………………….…. 33
American Redstart……………………………………………… 33
Northern Parula…………………………………………………. 35
Prairie Warbler………………………………………………….. 36

v
DISCUSSION…………………………………………………………………... 36
LITERATURE CITED………………………………………………………….. 40
CHAPTER 3: PREALTERNATE MOULT LINKS NONBREEDING HABITAT
CONDITIONS AND MIGRATION DEPARTURE TIMING IN NEARCTICNEOTROPICAL MIGRATORY SONGBIRDS………………………………………... 49
ABSTRACT…………………………………………………………………….. 49
INTRODUCTION……………………………………………………………… 50
METHODS……………………………………………………………………... 52
Study system……………………………………………………………. 52
Habitat quality and stable-carbon isotope sampling……………………. 53
Capture, body condition, and moult intensity…………………………... 55
Tracking migration departure dates…………………………………….. 56
Statistical analyses……………………………………………………… 57
Effect of δ13C on body condition……………………………….. 59
Predicting prealternate moult intensity…………………………. 60
Predicting departure dates………………………………………. 61
RESULTS……………………………………………………………………….. 61
Effect of δ13C on body condition……………………………………….. 61
Predicting prealternate moult intensity…………………………………. 62
Predicting departure dates………………………………………………. 66
DISCUSSION…………………………………………………………………... 71
LITERATURE CITED………………………………………………………….. 77
CHAPTER 4: CONCLUSION………………………………………………………… 87
Significance to the field of ornithology………………………………… 87
Limitations and directions for future study……………………………... 88
Management implications………………………………………………. 90
LITERATURE CITED………………………………………………………….. 92
APPENDIX A…………………………………………………………………………... 96
APPENDIX B…………………………………………………………………………... 98

vi
ACKNOWLEDGEMENTS
I want to start by first acknowledging the perspective through which I conducted
this research, as a descendant of white settlers, a resident of the Global North, and a
scientist trained in Western scientific methods, all of which come with inherent privileges
and biases. The field work component of this research took place on the stolen lands of
the Yamaye Taíno people, the first indigenous peoples of Yameyeka, also known as
Jamaica. As a student at Thompson Rivers University, I am based as an uninvited guest in
Secwepemcúlucw, on the stolen lands of the Tk’emlúps te Secwépemc people. Settlercolonialism and the Western scientific agenda have long-dominated the field of
environmental science in both Canada and Jamaica, often reinforcing systems of
oppression and exclusion. These systems continue to exclude Indigenous peoples and
peoples historically displaced because of settler-colonialism and slavery, from the study
and management of their lands. I would like to thank those of you who welcomed
conversation or provided guidance, as I sought to better understand the role of my
research in these dynamics and the inequities driven by Western science: Dr. Matt
Reudink, Dr. Lisa Cooke, Dr. Kara Lefevre, Sukh Heer Matonovich, Dr. Amie McLean,
Dr. Maggie MacPherson, and Ivy Ciaburri. I have much work to do but learned from all
of you.
I am immensely grateful to have worked with my supervisor, Dr. Matt Reudink.
Thank you for talking me into doing this thing, for sharing your knowledge, providing
constant support, and always holding my work to a high standard, yet reminding me to
take breaks. I appreciate and admire your desire to create an inclusive research
environment and support your student’s mental health. I have no doubts that my
experience working with you will shape the trajectory of my professional career going
forward.
My research benefited from the expertise of my supervising committee, Dr. Chris
Tonra, Dr. Mateen Shaikh, Dr. Emily Studd, and collaborators, Dr. Peter Marra, Dr.
Bryant Dossman, and Ivy Ciaburri. Thank you, Chris, for engaging me with exciting
research ideas in the beginning and for your insightful input through all stages of this
project, especially on my methodology. Thank you, Mateen, for helping me to work

vii
through challenging problems during my analyses. You are an excellent teacher and have
a superb ability to make statistics less daunting. Thank you, Emily, for your perspective
and your advice on my career beyond graduate studies. Pete and Bryant, thank you for
sharing your expert knowledge of the study system and for challenging me to think hard
about my research questions. Bryant, you are an incredibly patient mentor, and I am so
grateful that I got to learn from you in the field and while working through my analyses.
Having you as a mentor is a highlight of the past two years, as is your constant comic
relief. Ivy, you are truly the best research partner, field lead, and roommate I could have
asked for. I can’t imagine this experience without you. Thanks also to Dr. Jay Wright for
providing early advice on adapting my methods based on his experience.
Ivy and Christina, thank you for your teamwork and all of the hard work you put
into helping me succeed during the field season. I believe we all contributed literal blood,
sweat, and tears. I could not have pulled it off without you and value your friendship so
much. Thanks also to everyone who joined us in the field and boosted our morale,
especially Alicia, August, Ryan, and Henry.
Kelsey Freitag and Sam Gidora, I am so lucky to have gone through this program
with you. You are two of the most inspiring women I know. Sharing this experience with
you pushed me to be better every single day, and I think that’s rare. You have shared in
my successes and setbacks as if they were your own, and always remind me that I’m
capable.
Thank you to everyone in the BEAC Lab (past and present) and program cohort
who have made my time in Kamloops so special and made me feel part of a community,
especially Jacqueline, Sydney, and Natalie. Special thanks to Lindsay Veale for your
unwavering support and friendship since the day we met during my first month in
Kamloops. Thanks to Stephen Joly for creating the beautiful bird drawing used in
Chapter 2.
Thank you to Sukh Heer Matonovich and Dr. Marg Sonnenfeld. Working with and
learning from each of you was a meaningful part of my experience at TRU.
Thank you to my past mentors for helping me to develop professionally over the
years and encouraging me to pursue graduate studies: Dr. Debbie Wheeler, Kerry
Kenwood, Jay Rourke, Toby St. Clair, and Kyle Routledge. Deb, you have been my most

viii
influential mentor over the past 10 years. Thank you for sharing your love of birds with
me.
Finally, I want to thank my partner, Angus, my parents, Linda and Grey, and my
sister, Heidi. Thank you for being my biggest supporters in everything I do. I could not
have done this without your support. Special thanks to Penny for enthusiastically
enforcing my work-life balance.
I received funding for this research from a Ken Lepin Research Graduate Student
Award, British Columbia Graduate Scholarship, Graduate Student Research Mentor
Fellowship, Canadian Federation of University Women Memorial Fellowship, Sigma Xi
Grant in Aid of Research, Eastern Bird Banding Association Memorial Fund Research
Grant, and conference travel grants from the TRU Student Union, Association of Field
Ornithologists, and American Ornithological Society. This research was further supported
by Dr. Matt Reudink’s NSERC Discovery Grant, and in-kind support from Dr. Chris
Tonra and Dr. Peter Marra.
This research was conducted with permission from the National Environment and
Planning Agency of Jamaica, under a U.S.G.S. Federal Bird Banding Permit held by Dr.
Peter Marra (#24233), in accordance with Thompson Rivers University Animal Care
Committee (#103270) and the Georgetown Institutional Animal Care and Use Committee
(#2019-0081). The Petroleum Corporation of Jamaica granted us access to the field site.

ix
LIST OF FIGURES
Figure 1.1. An example of moult on the wing of a Black-headed Grosbeak (Pheucticus
melanocephalus), a small Nearctic-Neotropical migratory songbird. This bird grew the
black feathers more recently than the brown feathers; hence they are newer, fresher, and
in overall better quality. This example illustrates how moult can serve multiple functions:
to maintain the structural function of feathers, and to change the colour of a bird’s
plumage during the year to signal age or intrasexual dominance during the breeding
season. Photo taken by Shae Turner……………………………………………………... 3
Figure 1.2. Generalized diagram illustrating the annual life cycle of Nearctic-Neotropical
migratory birds that undergo prealternate moult on the stationary nonbreeding grounds.
Birds arrive to their temperate breeding grounds in Canada and the United States in April
– June, undergo prebasic moult after the breeding season, depart on post-breeding
migration Aug – Oct, spend the stationary nonbreeding period in the tropics Nov – Apr,
and undergo prealternate moult in the two to three months prior to departing on prebreeding migration. Photo of the Black-and-white Warbler (Mniotilta varia) taken by
Megan Buers……………………………………………………………………………... 5
Figure 1.3. Map showing the arrangement of study plots within Font Hill Nature
Preserve, (18.0391°N, 77.9411°W, <100 m above sea level), St. Elizabeth Parish, on the
southwest coast of Jamaica. Three mangrove plots are shown in green while two scrub
plots are shown in brown………………………………………………………………… 7
Figure 1.4. Range maps of A) Ovenbird, B) Northern Waterthrush, C) Black-and-white
Warbler, D) American Redstart, E) Northern Parula, and F) Prairie Warbler (Birds of the
World accounts; Porneluzi et al. 2020, Whitaker and Eaton 2020, Kricher 2020, Sherry et
al. 2020, Moldenhauer and Regelski 2020, Nolan Jr. et al. 2020). Photos taken by Shae
Turner………………………………………………………………..…………………… 8
Figure 1.5. Summary of the predicted relationships and their directions between
nonbreeding habitat quality, individual body condition, prealternate moult timing and
intensity, and pre-breeding migration departure date…………………………………... 10

x
Figure 2.1. Diagram of 17 contour feathers patches in which I scored prealternate moult
intensity. Illustration by Stephen Joly…………………………………………………... 22
Figure 2.2. The proportion of captured birds of six species moulting by week of year
from January 13 through April 18, 2023, at Font Hill Nature Preserve, Jamaica………. 28
Figure 2.3. The proportions of birds from six species moulting each of 17 contour feather
patches captured from January 13 through April 18, 2023, at Font Hill Nature Preserve,
Jamaica………………………………………………………………………………….. 30
Figure 2.4. The intensity of moult (scale of 0-4) for head and body feathers in six species
captured from January 13 through April 18, 2023, at Font Hill Nature Preserve,
Jamaica. The size of the circles represents the number (n) of birds sampled that had the
corresponding moult intensity score for each patch group.…………………………….. 31
Figure 2.5. The intensity of moult (scale of 0-4) per 17 feather patches in six species
captured from January 13 through April 18, 2023, at Font Hill Nature Preserve,
Jamaica. The size of the circles represents the number (n) of birds sampled that had the
corresponding moult intensity score for each patch.…………………………………… 32
Figure 2.6. Predicted linear relationships between contour feather moult intensity in
Black-and-white Warblers, measured as a Combined Contour Moult Index (CCMI) on a
scale of zero to four, and Julian date from the observed onset of moult, February 15,
2023, through April 18, 2023, interacting with age class. The solid lines represent the
predicted relationship between moult intensity and Julian date for each age class. The
shaded areas represent 95% confidence intervals………………………………………. 34
Figure 2.7. Predicted parabolic relationships between contour feather moult intensity in
American Redstarts, measured as a Combined Contour Moult Index (CCMI) on a scale of
zero to four, and Julian date from the observed onset of moult, February 15, 2023,
through April 18, 2023, grouped by age class. The solid lines represent the predicted
relationship between moult intensity and Julian date for each age class. The shaded areas
represent 95% confidence intervals…………………………………………………….. 35

xi
Figure 3.1. Analysis framework and summary of results from evaluating predictors of
residual prealternate moult intensity and pre-breeding migration departure dates in a)
Black-and-white Warbler, b) Northern Waterthrush, and c) American Redstart. I followed
a three-step process for the Black-and-white Warbler and Northern Waterthrush, adding
the first step of testing if δ13C, a proxy for habitat quality, predicted residual body
condition. I followed a two-step process for the American Redstart, using habitat type at
capture as an explanatory variable for habitat quality. Solid green arrows indicate a strong
effect (95% confidence intervals did not overlap zero), solid black arrows indicate a
moderate effect (90% confidence intervals did not overlap zero), and dashed arrows
indicate a variable with no effect, that was either not retained during model selection or
had 85% confidence intervals that overlapped zero. Stars indicate interaction terms with a
moderate or strong effect; only statistically meaningful interaction terms are depicted
here.……………………………………………………................................................... 58
Figure 3.2. Predicted relationship between residual moult intensity and residual body
condition by age class in the Black-and-white Warbler (n = 99). The residuals in this
model were derived from the linear regression equations body condition (mass/tarsus) ~
Julian date and Combined Moult Contour Index (CCMI) ~ Julian date. Shaded areas
represent 95% confidence intervals…………………………………………………….. 64
Figure 3.3. Predicted relationship between residual moult intensity and residual body
condition by age class in the American Redstart (n = 111). The residuals in this model
were derived from the linear regression equations body condition (mass/tarsus) ~ Julian
date and Combined Moult Contour Index (CCMI) ~ Julian date + Julian date2. Shaded
areas represent 95% confidence intervals………………………………………………. 66
Figure 3.4. Boxplots of residual moult intensity grouped by age class and habitat type in
the American Redstart (n = 111). Residual moult intensity was derived from the linear
regression equation Combined Moult Contour Index (CCMI) ~ Julian date + Julian
date2……………………………………………………………………………………... 66
Figure 3.5. δ13C ratios (‰) in the Northern Waterthrush (black circles). The line
represents the predicted relationship between pre-breeding departure date from Jamaica

xii
and δ13C ratio (‰) in the Northern Waterthrush (n = 14). Departure dates are in Julian
days; day 105 = April 15, 2023. The shaded area represents the 95% confidence
interval………………………………………………………………………………….. 69
Figure 3.6. Boxplots of pre-breeding migration departure dates from Jamaica grouped by
age and sex classes in the American Redstart (n = 48). Departure dates are Julian days;
day 110 = April 20, 2023……………………………………………………………….. 70
Figure 3.7. Predicted relationship between pre-breeding departure date from Jamaica and
residual moult intensity by sex in the American Redstart (n = 48). The residuals in this
model were derived from the linear regression equation Combined Moult Contour Index
(CCMI) ~ Julian date + Julian date2. Departure dates are in Julian days; day 100 = April
10, 2023. Shaded areas represent 95% confidence intervals…………………………… 71
Figure B.1. Boxplots of δ13C ratios (‰) sampled from A) Black-and-white Warbler, and
B) Northern Waterthrush, grouped by age class. δ13C ratios for each age class in the
Black-and-white Warbler are further grouped by sex because I could determine sex in the
field for this species……...……………………………………………………………... 98

xiii
LIST OF TABLES
Table 1.1. List of study species, pre-breeding migration timing for a bird departing
Jamaica and arriving in northern United States/southern Canada, and my pre-study
knowledge of the extent and timing of their prealternate moult. Information on migration
timing from eBird data (Fink et al. 2023) and Birds of the World accounts (Kricher 2020;
Moldenhauer and Regelski 2020; Nolan Jr. et al. 2020; Porneluzi et al. 2020; Sherry et al.
2020; Whitaker and Eaton 2020). Information on moult from Pyle (2022) and Birds of
the World accounts. Definitions of moult extent follow Pyle (2022): Absent = no moult;
limited = some, but not all, body feathers and no flight feathers; partial = most or all body
feathers and sometimes the tertial and/or central rectrices but no other flight feathers…. 9
Table 2.1. Summary for each study species of the unique individuals captured by age and
sex, the percentage of unique individuals scored for moult more than once, the total
number of moult scores, the percentage of birds scored for moult that were in active
moult, the mean ± SD Combined Contour Moult Index (CCMI) and the range of CCMI
values for moulting birds. n is the same for percentage in active moult (%) and mean ±
SD CCMI unless specified. n is the same for mean ± SD CCMI and CCMI
range.………………………………………………………………………………...….. 27
Table 3.1. Summary of the top ranked models (ΔAICc < 1) that predicted residual body
condition in the Black-and-white Warbler (n = 26) and Northern Waterthrush (n = 14).. 62
Table 3.2. Summary of the top ranked models (ΔAICc < 1) that predicted residual
prealternate moult intensity in the Black-and-white Warbler (n = 99), Northern
Waterthrush (n = 106), and American Redstart (n = 111)………………………………. 63
Table 3.3. Summary of the top ranked models (ΔAICc < 1) that predicted residual
prealternate moult intensity in the Black-and-white Warbler (n = 26) and Northern
Waterthrush (n = 14), using δ13C as an explanatory variable…………………………... 64
Table 3.4. Summary of the top ranked models (ΔAICc < 1) that predicted pre-breeding
migration departure dates from Jamaica in the Black-and-white Warbler (n = 21),
Northern Waterthrush (n = 16), and American Redstart (n = 48)………………………. 68

xiv
Table 3.5. Summary of the top ranked models (ΔAICc < 1) that predicted pre-breeding
migration departure dates from Jamaica in the Black-and-white Warbler (n = 17) and
Northern Waterthrush (n = 14), using δ13C as an explanatory variable………………… 68
Table A.1. Estimates for significant predictors (p-value < 0.05) of moult intensity as
measured by the Combined Contour Moult Index, in the American Redstart and Prairie
Warbler, modelled three ways using a linear model to handle individuals with repeated
measurements (recaptured birds): a) excluding the second capture, b) excluding the first
capture, and c) including both captures. Linear modelling followed a stepwise regression
(backward elimination), starting with the full initial model including the explanatory
variables: Julian date, Julian date2, age, and sex. For the American Redstart, I considered
the interaction term age*Julian date. It was not significantly related to moult intensity, so
I removed it from the final model………………………………………………………. 96
Table A.2. Sample sizes of the numbers of six species moulting and not moulting at
capture, after the observed onset of moult, used as input for a Pearson’s chi-squared
test………………………………………………………………………………………. 97
Table B.1. Stable-nitrogen (δ15N) ratios (‰) sampled from Black-and-white Warblers
and Northern Waterthrushes in March and April 2023, at Font Hill Nature Preserve,
Jamaica. I report corrected values measured in units of per mil (‰) relative to the
international standards Atmospheric Air for δ15N………………………………………. 99
Table B.2. Summary of estimates (β) and 85% confidence intervals for explanatory
variables from the top-ranked models (ΔAICc < 1) of all analyses that had no effect (85%
CIs overlapped zero) on the dependent variable…………………………………….… 100
Table B.3. Summary of the top ranked models (ΔAICc < 1) that predicted residual
prealternate moult intensity in the American Redstart, modelled three ways using a linear
model to handle individuals with repeated measurements (recaptured birds): a) excluding
the second capture, b) excluding the first capture, and c) including both captures. All
three modelling methods produced the same results………………………………….. 100

1
CHAPTER 1: INTRODUCTION
Animal migration is one of the most engaging phenomena in the natural world.
When colourful migratory songbirds arrive in temperate North America each spring, they
spark joy among people of many demographics and inspire them to connect with nature
and participate in environmental stewardship. Migratory birds play important roles in our
ecosystems, from plant pollination to agricultural pest control (Galloway et al. 2004,
Anderson et al. 2016, Díaz-Siefer et al. 2022) and can also act as indicators of
environmental health (Morrison 1986, Simons et al. 1999, Zöckler 2005). However,
migratory birds face threats such as habitat degradation and loss, changes in global
climate, and collisions with buildings (Machtans et al. 2013, Loss et al. 2014, Bairlein
2016, Zurrell et al. 2018, Rosenberg et al. 2019). As a result, migratory birds have
declined globally (Bairlein 2016, Zurrell et al. 2018, Rosenberg et al. 2019). In temperate
North America alone, we have lost an astounding 2.9 billion birds, or an estimated 29%,
of all birds between 1970 and 2017 (Rosenberg et al. 2019). Long-distance migrants,
such as Nearctic-Neotropical that migrate to the Caribbean and tropical regions of the
Americas during the season we call winter, have experienced more staggering declines
than non-migratory, native resident species (Rosenberg et al. 2019).
While Nearctic-Neotropical migrants might spend up to four months on their
breeding grounds in Canada and the United States, they spend a greater portion of the
year in-transit and on their tropical, stationary nonbreeding grounds (Albert et al. 2020)1.
Because migratory birds utilize diverse habitats throughout the year that are often
separated by thousands of kilometres, understanding the drivers of their population
decline (Marra et al. 2015) and identifying actions for conservation (Runge et al. 2015)
are uniquely challenging. Gaining this understanding requires identifying the temporal
and spatial periods of the annual life cycle in which bird populations are limited (Marra et
al. 2015). One such period for long-distance migrants can be the nonbreeding period. For
example, dry environmental conditions on the nonbreeding grounds can reduce the
1

This period is also known as “overwintering,” but that term is north-centric and inaccurate when
describing tropical seasons, so I use the terms “stationary nonbreeding,” or “nonbreeding” to refer to this
period, following Albert and Siegel (2023). Throughout the thesis I use “pre-breeding” and “post-breeding”
migration in place of the north-centric spring and fall migration (Albert and Siegel 2023)

2
annual survival of the American Redstart (Setophaga ruticilla) during northward, prebreeding migration (Dossman et al. 2023). Moreover, such effects can disproportionately
impact certain breeding populations. In this example, Dossman et al. (2023) found that
American Redstarts that breed further north, and therefore have a longer distance to
migrate, were most impacted by dry nonbreeding conditions. Undertaking research to
uncover such patterns, then applying this information to implement meaningful
conservation actions, requires massive cooperation across international and jurisdictional
boundaries (Runge et al. 2015, Albert et al. 2020).
Understanding moult is important for conserving migratory birds
Bird feathers serve myriad functions, including flight, thermoregulation,
protection from parasites, visual (and sometimes aural) communication, and others
(reviewed in Terrill and Shultz 2023). Because feathers are keratin structures, they cannot
repair themselves once damaged (Lennox and Rowlands 1969) and birds must shed and
regrow new feathers each year through a process called moult (Figure 1.1). Moult
requires protein to support feather synthesis and imposes a high energetic cost (Dietz et
al. 1992, Lindström et al. 1993, Cyr et al. 2008, Vézina et al. 2009, Lourenço 2014). The
energetic demands of moult can conflict with the energetic demands of the other
important annual life stages: migration and breeding (Barta et al. 2008). For this reason,
moult, migration, and breeding typically occur at different times during a bird’s annual
schedule without overlapping (Figure 1.2; Barta et al. 2008, Wingfield 2008, de la Hera et
al. 2012, Kiat et al. 2019), though there are exceptions. For example, moult-migrants
pause their migration at a “stopover” location to spend time moulting before proceeding
to their final destination (Tonra and Reudink 2018).

3

Figure 1.1. An example of moult on the wing of a Black-headed Grosbeak (Pheucticus
melanocephalus), a small Nearctic-Neotropical migratory songbird. This bird grew the
black feathers more recently than the brown feathers; hence they are newer, fresher, and
in overall better quality. This example illustrates how moult can serve multiple functions:
to maintain the structural function of feathers, and to change the colour of a bird’s
plumage during the year to signal age or intrasexual dominance during the breeding
season. Photo taken by Shae Turner.
The environmental conditions that a bird experiences during one life stage can
interact with the bird’s physiological state and “carry over” to shape their performance
(e.g., fitness) during the subsequent life stage, even creating a domino effect across
multiple stages, or seasons (Norris and Marra 2007, Harrison et al. 2011). For example,
songbirds that spend the stationary nonbreeding period in poor quality habitat tend to
have poorer body condition, which can delay pre-breeding migration (e.g., Marra et al.
1998, Studds and Marra 2005, Reudink et al. 2009), in turn delaying arrival to the
breeding grounds and culminating in reduced reproductive success when compared with
conspecifics from higher quality nonbreeding habitat (e.g., Norris et al. 2004, Reudink et
al. 2009). Identifying when and where these carry-over effects occur during the avian
annual cycle is critical for fully understanding when, where, and how declining
populations are limited (Marra et al. 2015). Some evidence suggests that the conditions a
bird experiences preceding moult can influence the timing of that moult (e.g., initiation;
Danner et al. 2015) or the production of plumage ornaments important for sexual
selection (Saino et al. 2004, Sparrow et al. 2017). The timing of moult, or the moult

4
strategy, can in turn carry over to influence the timing of birds’ migration (e.g.,
Stutchbury et al. 2011). Despite the year-round importance of moult, the potential role of
moult in carry-over effects has received far less study than breeding and migration,
leaving many gaps in the scientific community’s understanding of moult’s basic ecology
(Bridge 2011, Pyle et al. 2018, Kiat 2023).
Each year, adults of all songbird species moult most or all feathers after the
breeding season as part of the “prebasic” moult (Humphrey & Parkes 1959, Pyle 2022,
Jenni and Winkler 2020). Some species, including many Nearctic-Neotropical migratory
warblers (Family: Parulidae), moult some feathers a second time in the annual cycle,
prior to the breeding season (Figure 1.2; Pyle 2022). This second, partial moult is called
the “prealternate moult” (Howell et al. 2003). Parulid warblers that undergo prealternate
moult typically do so on the nonbreeding grounds, prior to pre-breeding migration (Pyle
2022). Because prealternate moult occurs during an understudied portion of the annual
cycle, the nonbreeding period, aspects of parulid prealternate moult such as its patterns,
intensity (i.e., percentage of feathers in moult at one time), and timing, are not well
understood when compared with the prebasic moult (see Chapter 2). Quantifying
prealternate moult phenology is an important first step toward being able to identify if
prealternate moult influences subsequent life history stages through carry-over effects. If
the prealternate moult period does limit bird populations, such information can be applied
to support the prioritization of high-quality nonbreeding habitats for conservation.

5

Figure 1.2. Generalized diagram illustrating the annual life cycle of Nearctic-Neotropical
migratory birds that undergo prealternate moult on the stationary nonbreeding grounds.
Birds arrive to their temperate breeding grounds in Canada and the United States in April
– June, undergo prebasic moult after the breeding season, depart on post-breeding
migration Aug – Oct, spend the stationary nonbreeding period in the tropics Nov – Apr,
and undergo prealternate moult in the two to three months prior to departing on prebreeding migration. Photo of the Black-and-white Warbler (Mniotilta varia) taken by
Megan Buers.
Study system
I completed field research from January to April 2023, at Font Hill Nature
Preserve (18.0391°N, 77.9411°W, <100 m above sea level), St. Elizabeth Parish, on the
southwest coast of Jamaica. I conducted this research as part of a long-term study (1987present) of a nonbreeding population of Nearctic-Neotropical migratory songbirds.
Research at Font Hill has primarily focused on the American Redstart as a model species,
but other work has investigated the nonbreeding ecology of the Ovenbird (Seiurus

6
aurocapilla), Black-and-white Warbler (Mniotilta varia), and Swainson’s Warbler
(Limnothlypis swainsonii). The decades-long body of work from this site has been
fundamental to the scientific community’s understanding of nonbreeding ecology and its
influence on the population dynamics of migratory birds through seasonal carry-over
effects. Numerous peer-reviewed articles and graduate theses have resulted from work at
Font Hill.
Font Hill Nature Preserve is owned by the Petroleum Company of Jamaica and is
adjacent to Font Hill Beach Park, a previously popular recreation site for locals and
tourists that is no longer in operation. Although hurricane damage during the 2010’s
altered the shoreline—and forest habitat structure within areas of the nature preserve—
this site remains under constant threat from development proposals.
I targeted bird capture in two habitat types: wet mangrove forest and a drier,
second-growth scrub forest. The study area included five, 5 to 7 ha plots (Figure 1.3). A
cluster of one scrub forest plot and two mangrove plots was separated by 200 m from one
mangrove plot adjoining a scrub forest plot.
I captured six warbler species (Family: Parulidae), Ovenbird, Northern
Waterthrush (Parkesia noveboracensis), Black-and-white Warbler, American Redstart,
Northern Parula (Setophaga americana), and Prairie Warbler (Setophaga discolor), and
used the data I collected on their prealternate moult in Chapter 2. In Chapter 3, I focused
only on the Northern Waterthrush, Black-and-white Warbler, and American Redstart, the
species for which capture rates were highest. For each species, their breeding and
nonbreeding ranges are shown in Figure 1.4. Pre-breeding migration timing from
Jamaica, and my pre-study knowledge of the extent and timing of the prealternate moult
for each species are summarized in Table 1.1.

7

Figure 1.3. Map showing the arrangement of study plots within Font Hill Nature
Preserve, (18.0391°N, 77.9411°W, <100 m above sea level), St. Elizabeth Parish, on the
southwest coast of Jamaica. Three mangrove plots are shown in green while two scrub
plots are shown in brown.

8

Figure 1.4. Range maps of A) Ovenbird, B) Northern Waterthrush, C) Black-and-white
Warbler, D) American Redstart, E) Northern Parula, and F) Prairie Warbler (Birds of the
World accounts; Porneluzi et al. 2020, Whitaker and Eaton 2020, Kricher 2020, Sherry et
al. 2020, Moldenhauer and Regelski 2020, Nolan Jr. et al. 2020). Photos taken by Shae
Turner.

9
Table 1.1. List of study species, pre-breeding migration timing for a bird departing
Jamaica and arriving in northern United States/southern Canada, and my pre-study
knowledge of the extent and timing of their prealternate moult. Information on migration
timing from eBird data (Fink et al. 2023) and Birds of the World accounts (Kricher 2020;
Moldenhauer and Regelski 2020; Nolan Jr. et al. 2020; Porneluzi et al. 2020; Sherry et al.
2020; Whitaker and Eaton 2020). Information on moult from Pyle (2022) and Birds of
the World accounts. Definitions of moult extent follow Pyle (2022): Absent = no moult;
limited = some, but not all, body feathers and no flight feathers; partial = most or all body
feathers and sometimes the tertial and/or central rectrices but no other flight feathers.
Species

Pre-breeding
migration timing

American
Redstart

Departs mid-Apr –
early May
Arrives mid-May

Black-andwhite Warbler

Departs mid-late Apr
– early May
Arrives late April –
mid-May
Departs late March –
late Apr
Earliest departures
from the Caribbean in
early Feb
Arrives mid-May
Departs late Apr –
early May
Arrives early – midMay
Departs mid-late Apr
Arrives late Apr –
mid-May

Northern
Parula

Northern
Waterthrush

Ovenbird

Prairie Warbler

Departs late Mar –
mid-Apr
Arrives late Apr –
mid-May

Pre-study knowledge of prealternate
moult
Extent
Timing
Absent – limited
Mar –
scattered head and body
Apr
feathers, occasionally 1-3
greater coverts in Mar – Aug
Limited – partial
Mar –
Body feathers
Apr

Absent – limited
Body feathers in some
individuals

Feb –
Apr

Limited – partial
Body feathers

Feb –
Apr

Absent

NA

Limited
Usually head and neck feathers

Feb –
Apr

10
Research goal
The purpose of my thesis is to evaluate the potential role of prealternate moult in
nonbreeding period carry-over effects in Nearctic-Neotropical migratory songbirds.
Chapter 2 is focused on quantifying the patterns, intensity, and timing of prealternate
moult in six parulid warbler species on their nonbreeding grounds in Jamaica to
understand variation among species, age, and sex classes across the moult period. For
three of those species, Chapter 3 is focused on testing the prediction that nonbreeding
habitat quality and body condition influence the timing and intensity of prealternate
moult, and in turn, carry over to influence the date birds depart Jamaica for pre-breeding
migration (Figure 1.5). Specifically, I predict that birds from high-quality habitat will
have better body condition, complete a high-intensity moult earlier, and depart from
Jamaica earlier when compared with birds from poor-quality habitats. This thesis
concludes with Chapter 4, in which I summarize the significance of my results to the
broader field and their implications for environmental management.

Figure 1.5. Summary of the predicted relationships and their directions between
nonbreeding habitat quality, individual body condition, prealternate moult timing and
intensity, and pre-breeding migration departure date.

11
LITERATURE CITED
Albert, S. K., & Siegel, R. B. (2023). Improving the language of migratory bird science
in North America. Ornithological Applications, 126, 1–4.
doi.org/10.1093/ornithapp/duad059
Albert, S., Wolfe, J. D., Kellerman, J., Sherry, T., Stutchbury, B. J. M., Bayly, N. J., &
Ruiz-Sánchez, A. (2020). Habitat ecology of Nearctic–Neotropical migratory
landbirds on the nonbreeding grounds. The Condor, 122(4), 1–18.
doi.org/10.1093/condor/duaa055
Anderson, S. H., Kelly, D., Robertson, A. W., & Ladley, J. J. (2016). Pollination by birds:
A functional evaluation. In Ç. H. Sekercioglu, D. G. Wenny, & C. J. Whelan
(Eds.), Why birds matter (pp. 73–106). University of Chicago Press.
Bairlein, F. (2016). Migratory birds under threat: Habitat degradation and loss, illegal
killings, and climate change threaten European migratory bird populations.
Science, 354(6312), 547–548. doi.org/10.1126/science.aah6647
Barta, Z., McNamara, J. M., Houston, A. I., Weber, T. P., Hedenström, A., & Feró, O.
(2008). Optimal moult strategies in migratory birds. Philosophical Transactions
of the Royal Society B: Biological Sciences, 363(1490), 211–229.
doi.org/10.1098/rstb.2007.2136
Bridge, E. S. (2011). Mind the gaps: What’s missing in our understanding of feather molt.
The Condor, 113(1), 1–4. doi.org/10.1525/cond.2011.100228
Cyr, N. E., Wikelski, M., & Romero, L. M. (2008). Increased energy expenditure but
decreased stress responsiveness during molt. Physiological and Biochemical
Zoology, 81(4), 452–462. doi.org/10.1086/589547
Danner, R. M., Greenberg, R. S., Danner, J. E., & Walters, J. R. (2015). Winter food
limits timing of pre‐alternate moult in a short‐distance migratory bird. Functional
Ecology, 29(2), 259–267. doi.org/10.1111/1365-2435.12322
de la Hera, I., DeSante, D. F., & Milá, B. (2012). Feather growth rate and mass in
Nearctic passerines with variable migratory behavior and molt pattern. The Auk,
129(2), 222–230. doi.org/10.1525/auk.2012.11212
Díaz-Siefer, P., Olmos-Moya, N., Fontúrbel, F. E., Lavandero, B., Pozo, R. A., & CelisDiez, J. L. (2022). Bird-mediated effects of pest control services on crop
productivity: A global synthesis. Journal of Pest Science, 95(2), 567–576.
doi.org/10.1007/s10340-021-01438-4

12
Dietz, M. W., Daan, S., & Masman, D. (1992). Energy requirements for molt in the
kestrel Falco tinnunculus. Physiological Zoology, 65(6), 1217–1235.
doi.org/10.1086/physzool.65.6.30158276
Dossman, B. C., Studds, C. E., LaDeau, S. L., Sillett, T. S., & Marra, P. P. (2023). The
role of tropical rainfall in driving range dynamics for a long-distance migratory
bird. Proceedings of the National Academy of Sciences, 120(52), e2301055120.
doi.org/10.1073/pnas.2301055120
Fink, D., Auer, T., Johnston, A., Strimas-Mackey, M., Ligocki, S., Robinson, O.,
Hochachka, W., Jaromczyk, L., Crowley, C., Dunham, K., Stillman, A., Davies, I.,
Rodewald, A., Ruiz-Gutierrez, V., & Wood, C. (2023). eBird Status and Trends
[Dataset]. doi.org/10.2173/ebirdst.2022
Galloway, T. S., Brown, R. J., Browne, M. A., Dissanayake, A., Lowe, D., Jones, M. B.,
& Depledge, M. H. (2004). A multibiomarker approach to environmental
ssessment. Environmental Science & Technology, 38(6), 1723–1731.
doi.org/10.1021/es030570+
Harrison, X. A., Blount, J. D., Inger, R., Norris, D. R., & Bearhop, S. (2011). Carry-over
effects as drivers of fitness differences in animals. Journal of Animal Ecology,
80(1), 4–18. doi.org/10.1111/j.1365-2656.2010.01740.x
Howell, S. N. G., Corben, C., Pyle, P., & Rogers, D. (2003). The first basic problem: A
review of molt and plumage homologies. The Condor, 105(4), 635–653.
Humphrey, P. S., & Parkes, K. C. (1959). An approach to the study of molts and
plumages. The Auk, 76(1), 1–31. doi.org/10.2307/4081839
Jenni, L., & Winkler, R. (2020). Moult and ageing of European passerines (Second
edition). Helm.
Kiat, Y. (2023). Moult terminology: Let’s make it simpler! Ibis, 165(2), 697–703.
doi.org/10.1111/ibi.13164
Kiat, Y., Vortman, Y., & Sapir, N. (2019). Feather moult and bird appearance are
correlated with global warming over the last 200 years. Nature Communications,
10(1), 2540. doi.org/10.1038/s41467-019-10452-1
Kricher, J. C. (2020). Black-and-white warbler (Mniotilta varia). In A. F. Poole (Ed.),
Birds of the world (version 1.0). Cornell Lab of Ornithology.
doi.org/10.2173/bow.bawwar.01
Lennox, F. G., & Rowlands, R. J. (1969). Photochemical degradation of keratins.
Photochemistry and Photobiology, 9(4), 359–367. doi.org/10.1111/j.17511097.1969.tb07300.x

13
Lindström, Å., Visser, G. H., & Daan, S. (1993). The energetic cost of feather synthesis is
proportional to basal metabolic rate. Physiological Zoology, 66(4), 490–510.
Loss, S. R., Will, T., Loss, S. S., & Marra, P. P. (2014). Bird–building collisions in the
United States: Estimates of annual mortality and species vulnerability. The
Condor, 116(1), 8–23. doi.org/10.1650/CONDOR-13-090.1
Lourenço, P. M. (2014). The energetic cost of pre-alternate body moult in black-tailed
godwits (Limosa l. Limosa) staging in Portugal during northward migration. Airo,
23, 3–9.
Machtans, C. S., Wedeles, C. H. R., & Bayne, E. M. (2013). A first estimate for Canada
of the number of birds killed by colliding with building windows. Avian
Conservation and Ecology, 8(2), 6. doi.org/10.5751/ACE-00568-080206
Marra, P. P., Cohen, E. B., Loss, S. R., Rutter, J. E., & Tonra, C. M. (2015). A call for full
annual cycle research in animal ecology. Biology Letters, 11(8), 20150552.
doi.org/10.1098/rsbl.2015.0552
Marra, P. P., Hobson, K. A., & Holmes, R. T. (1998). Linking winter and summer events
in a migratory bird by using stable-carbon isotopes. Science, 282(5395), 1884–
1886. doi.org/10.1126/science.282.5395.1884
Moldenhauer, R. R., & Regelski, D. J. (2020). Northern parula (Setophaga americana).
In A. F. Poole (Ed.), Birds of the world (version 1.0). Cornell Lab of Ornithology.
doi.org/10.2173/bow.norpar.01
Morrison, M. L. (1986). Bird populations as indicators of environmental change. In R.
F. Johnston (Ed.), Current ornithology (pp. 429–451). Springer US.
doi.org/10.1007/978-1-4615-6784-4_10
Nolan Jr, V., Ketterson, E. D., & Buerkle, C. A. (2020). Prairie warbler (Setophaga
discolor). In A. F. Poole (Ed.), Birds of the world (version 1.0). Cornell Lab of
Ornithology. doi.org/10.2173/bow.prawar.01
Norris, D. R., & Marra, P. P. (2007). Seasonal interactions, habitat quality, and population
dynamics in migratory birds. The Condor, 109(3), 535–547.
Norris, D. R., Marra, P. P., Kyser, T. K., Sherry, T. W., & Ratcliffe, L. M. (2004). Tropical
winter habitat limits reproductive success on the temperate breeding grounds in a
migratory bird. Proceedings of the Royal Society of London. Series B: Biological
Sciences, 271(1534), 59–64. doi.org/10.1098/rspb.2003.2569
Porneluzi, P., Van Horn, M. A., & Donovan, T. M. (2020). Ovenbird (Seiurus
aurocapilla). In A. F. Poole (Ed.), Birds of the World. Cornell Lab of Ornithology.
doi.org/10.2173/bow.ovenbi1.01

14
Pyle, P. (2022). Identification guide to North American birds: Part I (2nd ed). Slate Creek
Press.
Pyle, P., Saracco, J. F., & DeSante, D. F. (2018). Evidence of widespread movements
from breeding to molting grounds by North American landbirds. The Auk, 135(3),
506–520. doi.org/10.1642/AUK-17-201.1
Reudink, M. W., Marra, P. P., Kyser, T. K., Boag, P. T., Langin, K. M., & Ratcliffe, L. M.
(2009). Non-breeding season events influence sexual selection in a long-distance
migratory bird. Proceedings of the Royal Society B: Biological Sciences,
276(1662), 1619–1626. doi.org/10.1098/rspb.2008.1452
Rosenberg, K. V., Dokter, A. M., Blancher, P. J., Sauer, J. R., Smith, A. C., Smith, P. A.,
Stanton, J. C., Panjabi, A., Helft, L., Parr, M., & Marra, P. P. (2019). Decline of
the North American avifauna. Science, 366(6461), 120–124.
doi.org/10.1126/science.aaw1313
Runge, C. A., Watson, J. E. M., Butchart, S. H. M., Hanson, J. O., Possingham, H. P., &
Fuller, R. A. (2015). Protected areas and global conservation of migratory birds.
Science, 350(6265), 1255–1258. doi.org/10.1126/science.aac9180
Saino, N., Szép, T., Ambrosini, R., Romano, M., & Møller, A. P. (2004). Ecological
conditions during winter affect sexual selection and breeding in a migratory bird.
Proceedings of the Royal Society of London. Series B: Biological Sciences,
271(1540), 681–686. doi.org/10.1098/rspb.2003.2656
Sherry, T. W., Holmes, R. T., Pyle, P., & Patten, M. A. (2020). American Redstart
(Setophaga ruticilla). In A. F. Poole (Ed.), Birds of the World. Cornell Lab of
Ornithology. doi.org/10.2173/bow.amered.01
Simons, T. R., Rabenold, K. N., Buehler, D. A., Collazo, J. A., & Franzreb, K. E. (1999).
The role of indicator species: Neotropical migratory song birds. In Ecosystem
management for sustainability: Principles and practices illustrated by a regional
biosphere reserve cooperative (pp. 187–208). CRC Press, Lewis Publishers.
Sparrow, K. L., Donkor, K. K., Flood, N. J., Marra, P. P., Pillar, A. G., & Reudink, M. W.
(2017). Conditions on the Mexican moulting grounds influence feather colour and
carotenoids in Bullock’s orioles (Icterus bullockii). Ecology and Evolution, 7(8),
2643–2651. doi.org/10.1002/ece3.2836
Studds, C. E., & Marra, P. P. (2005). Nonbreeding habitat occupancy and population
processes: An upgrade experiment with a migratory bird. Ecology, 86(9), 2380–
2385. doi.org/10.1890/04-1145
Stutchbury, B. J. M., Gow, E. A., Done, T., MacPherson, M., Fox, J. W., & Afanasyev, V.
(2011). Effects of post-breeding moult and energetic condition on timing of

15
songbird migration into the tropics. Proceedings of the Royal Society B:
Biological Sciences, 278(1702), 131–137. doi.org/10.1098/rspb.2010.1220
Terrill, R. S., & Shultz, A. J. (2023). Feather function and the evolution of birds.
Biological Reviews, 98(2), 540–566. doi.org/10.1111/brv.12918
Tonra, C. M., & Reudink, M. W. (2018). Expanding the traditional definition of moltmigration. The Auk, 135(4), 1123–1132. https://doi.org/10.1642/AUK-17-187.1
Vézina, F., Gustowska, A., Jalvingh, K. M., Chastel, O., & Piersma, T. (2009). Hormonal
correlates and thermoregulatory consequences of molting on metabolic rate in a
northerly wintering shorebird. Physiological and Biochemical Zoology, 82(2),
129–142. doi.org/10.1086/596512
Whitaker, D. M., & Eaton, S. W. (2020). Northern Waterthrush (Parkesia
noveboracensis). In A. F. Poole (Ed.), Birds of the World (version 1.0). Cornell
Lab of Ornithology. doi.org/10.2173/bow.norwat.01
Wingfield, J. C. (2008). Organization of vertebrate annual cycles: Implications for control
mechanisms. Philosophical Transactions of the Royal Society B: Biological
Sciences, 363(1490), 425–441. doi.org/10.1098/rstb.2007.2149
Zöckler, C. (2005). Migratory bird species as indicators for the state of the environment.
Biodiversity, 6(3), 7–13. doi.org/10.1080/14888386.2005.9712769
Zurell, D., Graham, C. H., Gallien, L., Thuiller, W., & Zimmermann, N. E. (2018). Longdistance migratory birds threatened by multiple independent risks from global
change. Nature Climate Change, 8(11), 992–996. doi.org/10.1038/s41558- 018
0312-9

16
CHAPTER 2: QUANTIFYING PREALTERNATE MOULT TIMING, PATTERNS,
AND INTENSITY IN SIX NEARCTIC-NEOTROPICAL MIGRATORY WARBLERS
ABSTRACT
Moult is vital for maintaining year-round feather function yet is one of the least
understood events in the annual cycle of migratory birds. In Nearctic-Neotropical
migratory songbirds, prealternate moult has received far less study than prebasic moult
because it typically occurs during the least studied period of the annual cycle: the
stationary nonbreeding period. Improving our basic understanding of prealternate moult
is fundamental for identifying how it interacts with other life history stages across the
annual cycle. Here, I provide a detailed quantification of the timing, patterns, and
intensity of prealternate moult for six species of parulid warblers on their stationary
nonbreeding grounds in Jamaica. I demonstrate that head and body feather moult is
common for Northern Waterthrush (Parkesia noveboracensis), Black-and-white Warbler
(Mniotilta varia), American Redstart (Setophaga ruticilla), Northern Parula (Setophaga
americana), and Prairie Warbler (Setophaga discolor), and for most species increases in
frequency and intensity later in the nonbreeding period. Black-and-white Warbler and
American Redstart demonstrated age-specific patterns in moult intensity, with greater
moult intensity exhibited by first-cycle than definitive-cycle birds. I confirm the
occurrence of prealternate moult in at least some Ovenbirds (Seiurus aurocapilla) and
provide the first detailed description of their contour feather moult. These findings
represent the first quantification of contour feather moult intensity in these species and
will serve as a foundation for investigating the mechanisms that regulate prealternate
moult and potential carry-over effects from the nonbreeding grounds.
INTRODUCTION
An understanding of how life history stages interact across the full annual cycle is
needed to effectively conserve declining migratory bird populations (Marra 2015a,
Rosenberg et al. 2019). This task is particularly challenging for migratory species whose
life history stages are separated temporally and spatially across the annual cycle while
birds track seasonal resources across thousands of kilometers. Moult, the periodic

17
replacement of feathers, is one stage for which substantial knowledge gaps remain
(Bridge 2011, Pyle et al. 2018, Pyle 2022a, Kiat 2023). Moult is vital to maintaining a
myriad of feather functions from flight and movement overall (Hedenström 2002, Jenni
and Winkler 2020a), thermoregulation (Wolf and Walsberg 2000), visual communication
(Hill and McGraw 2006, Santos et al. 2011, Jenni and Winkler 2020a, Terrill et al. 2020),
protection from parasites (e.g., Gunderson et al. 2008) and solar radiation, among others
(for a full review of feather function, see Terrill and Shultz 2023). Since moult has a high
energetic cost (Dietz et al. 1992, Lindström et al. 1993, Cyr et al. 2008, Vézina et al.
2009, Lourenço 2014), moult in migratory birds is typically scheduled not to overlap with
breeding and migration (Barta et al. 2008, Wingfield 2008, de la Hera et al. 2012, Kiat et
al. 2019), two other energetically costly events, though moult and migration strategies
vary (Rohwer 2005). Some evidence suggests that moult strategy and/or timing can carry
over to influence birds’ migratory timing (e.g., Carlisle et al. 2005, Stutchbury et al. 2011,
Kiat and Izhaki 2016), and that conditions experienced prior to moult can influence its
timing (e.g., initiation date; Danner et al. 2015), and the production of phenotypic
plumage traits likely to have consequences on individual fitness (e.g., Saino et al. 2004,
Sparrow et al. 2017). Understanding the frequency, intensity, and timing of moult is
imperative given its year-round importance to feather function and its potential to
influence subsequent life history stages through seasonal interactions.
Each year, adults of all species of passerines moult most or all contour and flight
feathers as part the prebasic moult (Humphrey & Parkes 1959), typically occurring after
the breeding season in north-temperate resident and Nearctic-Neotropical migratory birds
(Pyle 2022b). Some species replace contour feathers a second time in the annual cycle
during a prealternate moult, typically occurring prior to the breeding season (Humphrey
& Parkes 1959, Pyle 2022b). The prealternate moult is thought to have evolved to
maintain feather function in long-distance migratory birds and as the mechanism through
which migrant passerines achieve seasonal dichromatism (Kiat et al. 2018, Terrill et al.
2020, Cuervo et al. 2022). Most Nearctic-Neotropical migrant songbirds that undergo
prealternate moult complete it on the stationary nonbreeding grounds (hereafter
“nonbreeding grounds”) prior to pre-breeding (northward) migration (Pyle 2022b),

18
though the number of species that employ a prealternate moult-migration strategy may be
underestimated (Leu and Thompson 2002, Tonra and Reudink 2018, Wright et al. 2018).
Given that the study of migratory birds has long been biased toward the breeding
period and breeding grounds (Marra et al. 2015a, Kiat 2023), it’s unsurprising that the
study of moult has focused heavily on the prebasic moult, leaving a significant
knowledge gap in our understanding of the prealternate moult. While detailed
information is available for the timing and extent of prebasic moult for almost all
Nearctic-Neotropical migrant songbirds, such information ranges from limited to absent
for those species’ prealternate moult (Pyle 2022b). Not only are many descriptions of the
timing and extent of prealternate moult lacking detail when compared with the prebasic
moult, but for some species the extent (e.g., Black-and-white Warbler (Mniotilta varia))
or occurrence of prealternate moult (e.g., Ovenbird (Seiurus aurocapilla)) remain
uncertain or unknown. Further, much study of the prealternate moult has focused on wing
feather moult and thereby the species that moult wing feathers as part of the prealternate
moult. This bias is likely because prealternate moult limits on passerines can be easily
detected on the wings of museum specimens (e.g., Pyle and Carnes 2022) and on birds
captured on the breeding grounds (e.g., Crary and Rodewald 2012, Jones et al. 2014,
Cimprich 2018, Carnes et al. 2021), prior to the prebasic moult. Once prealternate moult
is completed, moult limits between the tiny contour feathers of small birds are
challenging to detect in the field, becoming near impossible to detect when replaced
alternate feathers are the same colour as retained basic feathers. This challenge has likely
contributed to a biased focus on prealternate contour feather moult in particular regions
of the body that contribute to seasonal dichromatism (e.g., Boone 2006, Boone et al.
2010), leading to a reduced understanding of prealternate moult patterns on other regions
of the body. Studying birds in active prealternate moult, whether wild birds captured on
the moulting grounds (e.g., Mettke-Hofmann et al. 2010, Renfrew et al. 2011, Danner et
al. 2015, Wright et al. 2018) or museum specimens (e.g., Rohwer et al. 1983, Jackson et
al. 1992, Voelker and Rohwer 1998, Voelker and McFarland 2002, Sieburth and Pyle
2018), allows for greater capacity to accurately quantify the extent or intensity of
prealternate contour feather moult but these studies are relatively few in number.

19
Of the 46 warbler species (Family: Parulidae) that are Nearctic-Neotropical
migrants, 33 (72%) are thought to have a prealternate moult strategy, at least in some
individuals, and the occurrence of prealternate moult remains uncertain in additional 5
species (Pyle 2022b). For 8 of the 33 (24%) species thought to have a prealternate moult
strategy, there is some level of uncertainty in the apparent extent of that moult (Pyle
2022b). Aside from recent updates to species moult accounts in Identification Guide to
North American Birds, Part I (2nd Edition; Pyle 2022b), prealternate moult has largely
been overlooked in the study of wild Nearctic-Neotropical migratory warblers over the
past 24 years. A systematic search of scientific papers in The Web of Science yielded only
four papers since the year 2000 related to field study of the prealternate moult in any
Nearctic-Neotropical migratory parulid species (conducted June 2024, using the search
terms “prealternate moult” or “prealternate molt” alongside each species’ common name).
Only one of these studies aimed to quantify prealternate moult in all feather groups
during active moult; in the Lucy’s Warbler (Leiothlypis luciae) and Virginia’s Warbler
(Leiothlypis virginiae) (Voelker and McFarland 2002). Studies of the Magnolia Warbler
(Setophaga magnolia) (Boone et al. 2010) and Yellow Warbler (Setophaga petechia;
Quinlan and Green 2011, Crary and Rodewald 2012) aimed to understand links between
feathers moulted during the prealternate moult and another life history stage, using data
collected away from the moulting grounds. I know of one additional published study on
the Yellow Warbler (Jones et al. 2014) that was not captured by my search. While this
systematic search is likely to have accurately captured work published by scientists based
in the Global North, it excludes dissertations (e.g., Ames 2021) and may fail to account
for studies of prealternate moult by Neotropical researchers that are not published in
global-scope journals, particularly if those studies are descriptive (Ruelas Inzunza et al.
2023, Soares et al. 2023).
Collecting detailed information on the timing and patterns of prealternate moult in
all areas of the bird’s body is the necessary first step to facilitate future study of the
mechanisms that regulate prealternate moult, such as body condition, and how
prealternate moult interacts with other life history stages. As such, it will be critical to
gather fundamental information of how prealternate moult varies across species, age and
sex classes, and eventually across the nonbreeding period range for individual species.

20
Here, I provide a detailed quantification of the timing, intensity, and patterns of
prealternate moult for six species of Nearctic-Neotropical migratory warbler on their
nonbreeding grounds in Jamaica. This study represents the most detailed information
available for the moult of contour feathers across all areas of the head and body and is the
first quantification of moult intensity over time in the species studied.
METHODS
Study species
My study focused on six primarily insectivorous, long-distance, NearcticNeotropical migratory species that are commonly captured in Jamaica during their
nonbreeding period: Ovenbird, Northern Waterthrush (Parkesia noveboracensis), Blackand-white Warbler, American Redstart (Setophaga ruticilla), Northern Parula (Setophaga
americana), and Prairie Warbler (Setophaga discolor). While several of these species
have been the subject of extensive research on the nonbreeding grounds (e.g., Dugger et
al. 2004, Smith et al. 2011, Marra et al. 2015b, Kent and Sherry 2020, Cooper et al. 2021,
Kent et al. 2022), relatively little is known about the extent and timing of their
prealternate moult, which has been suggested to occur during this portion of the annual
cycle (Pyle 2022b, Kricher 2020, Moldenhauer and Regelski 2020, Nolan Jr. et al. 2020,
Porneluzi et al. 2020, Sherry et al. 2020, Whitaker and Eaton 2020). The occurrence of
prealternate moult in the Ovenbird requires confirmation (Porneluzi et al. 2020, Pyle
2022b); at least some individuals of the other five species complete a limited to partial
prealternate moult, as in some but not all contour feathers and typically no wing feathers
(Pyle 2022b).
Study site
I conducted this study on the southwest coast of Jamaica within the Font Hill
Nature Preserve (18.0391°N, 77.9411°W, <100 m above sea level), St. Elizabeth Parish,
from January 13 through April 18, 2023. Precipitation at this site is strongly seasonal,
with a wet season typically from August to November (average monthly precipitation
>100 mm) and dry season from December through May (average monthly precipitation
<25 mm; Powell et al. 2021).

21
I captured Nearctic-Neotropical migratory warblers in two second-growth dry
scrub forest plots and three mangrove plots, ranging in size from 5 to 7 ha each. A cluster
of one scrub forest plot and two mangrove plots was separated by 200 m from one
mangrove plot adjoining a scrub forest plot. See Cooper et al. (2021) for a description of
vegetation structure and composition in these two habitat types. Scrub plots remained dry
without standing water throughout the study period, typical conditions for this habitat
type during the dry season (Marra et al. 2015b, Cooper et al. 2021). Mangrove plots
typically hold 0.3 to 2 m of standing brackish water throughout the dry season (Cooper et
al. 2021). Due to relatively low levels of precipitation preceding and during the study
period, water levels in the mangrove plots remained low and at times, areas within plots
were completely dry. Similar conditions have occurred at the study site in previous dry
years (e.g., Cooper et al. 2015, Brunner et al. 2022, Dossman et al. 2023).
Field methods
I captured warblers by mist-netting on 47 days between January 13 and April 18,
2023, using a combination of passive netting and target netting with an audio lure. Each
week I captured birds in at least one scrub plot and one mangrove plot. Each capture day
usually involved opening at least ten 12 m long mist nets for five to six hours of passive
netting after sunrise. I used target netting with an audio lure to capture highly territorial
American Redstarts throughout the season, and from March through April to recapture
individual Black-and-white Warblers and Northern Waterthrushes. I typically used one or
two 12 m mist nets from the array for target netting or erected an additional 6 m mist net.
I aimed to recapture as many individuals as possible to observe the progression of moult
over time.
I banded each individual with a uniquely numbered United States Geological
Survey aluminum band. I determined sex for species with sex-based differences in
plumage characteristics (Pyle 2022b): American Redstart, Black-and-white Warbler,
Northern Parula, and Prairie Warbler. I was unable to determine sex using morphometric
measurements for Ovenbird or Northern Waterthrush (Pyle 2022b). I determined age
classes based on moult and plumage characteristics (Pyle 2022b), following the WRP age
classification system (Wolfe et al. 2010) most recently modified by Pyle et al. (2022).

22
Thereby, I assigned moulting individuals to one of two age classes, First Prealternate
Moult (FPA, hereafter “first-cycle”) or Definitive Prealternate Moult (DPA, hereafter
“definitive-cycle”).
I systematically searched for active contour feather moult (pinfeathers) by
blowing on the feathers (Mettke-Hoffman et al. 2010, Wright et al. 2018) in each of 17
patches across the head and body (Figure 2.1; Wright et al. 2018). For each patch, I
scored the percentage of feathers in active moult on a scale of 0 to 4, a method I adapted
from Greenwood et al. 1983 and Wright et al. 2018. The scale divides the percentage of
feathers in active moult into quarters, where 0 = 0%, 1 = 1–25%, 2 = 26–50%, 3 = 51–
75%, and 4 = 76–100% of active moult per patch. To minimize observer error, moult
scores for each bird were collected by one of two skilled observers (SAT and IAC). At the
beginning of the field season, we repeated the moult scores on the same birds to
standardize our measurements and reduce variability associated with the observer. We
periodically repeated this process for feather patches with active moult throughout the
season.

Figure 2.1. Diagram of 17 contour feathers patches in which I scored prealternate moult
intensity. Illustration by Stephen Joly.

23
I considered all instances of contour feather moult to be part of a prealternate
moult strategy, rather than part of a protracted or suspended (Jenni and Winkler 2020b,
Pyle 2022b) preformative/prebasic moult (Howell et al. 2003), or adventitious feather
replacement. Five of the six study species were not moulting during the first 1.5 weeks of
capture (at least 4 weeks without moult for most species), so I am confident that I
observed seasonal prealternate moult. I captured a Northern Waterthrush in moult on the
first day that I captured the species (January 25), so prealternate moult for some
individuals of the species may have commenced earlier in January than I could document.
I refer to the first observed occurrence of moult for each species as the “observed onset”
of moult for that species.
I could not easily distinguish retained contour feathers from those replaced on the
nonbreeding grounds prior to capture, except in first-cycle male American Redstarts
when recently replaced feathers were black. Prior to the observed onset of moult in the
species, plumage appearance varied considerably in males of the other focal species with
differing basic and alternate plumage, making it impossible to determine the extent of
contour feather moult completed before capture. For example, both first- and definitivecycle male Black-and-white Warblers displayed wide variation in the number and
position of black feathers on the cheeks and throat. For this reason, I used moult intensity,
rather than extent, as my moult metric.
I also examined the wings and rectrices for active moult, symmetrical or not, and
recorded the number and position of coverts, flight feathers, or rectrices in pin. The study
species are not known to moult wing or tail feathers as part of their prealternate moult
strategies (though some American Redstarts may moult one to three greater coverts; Pyle
2022b), so the few individuals I observed moulting these feathers likely did so
adventitiously, as previously documented in American Redstarts at the study site (Tonra
et al. 2014). I exclude this information from the statistical analyses.
Quantifying moult intensity and prevalence
To describe the intensity of active moult per individual, I calculated a combined
contour moult index (CCMI) modified from Wright et al.’s (2018) variation of the
Greenwood et al. (1983) method. Each bird received a CCMI score on a scale of 0 to 4,

24
the sum of all patch scores divided by the total number of patches (17). The CCMI score
divides the percentage of active moult across the entire head and body into quarters; for
example, a bird with a CCMI score of 1 was actively moulting between 1 and 25% of all
contour feathers.
Since I anticipated differences in moult intensity between the head and body in
most species (Chapter 1, Table 1.1), to facilitate comparison I also grouped the feather
patches by head (6 patches) and body (11 patches) and calculated a CCMI score for each
patch group. To compare moult intensity among moulting individuals across species, I
calculated the mean CCMI score for each species including only birds that were actively
moulting at capture and had a CCMI score recorded. I then calculated the mean CCMI
score for each patch, per species, only including individuals actively moulting that patch
at capture (Wright et al. 2018).
For each species’ population, I calculated the prevalence of moult by feather patch
as the proportion of all captures (including non-moulting individuals) actively moulting
each patch (Wright et al. 2018). In all calculations at the population level, I included data
from recaptures, regardless of the number of days passed since first (or previous) capture.
This decision was informed by multiple observations of moult starting in a new feather
patch only one day after the bird was captured and scored for active moult. In each of
these cases, I carefully compared patch scores in the field with those from the previous
capture to ensure that differing patch scores were not a result of observer error.
For each species, I also aimed to determine the population-level peak in moult
prevalence, as in the period where the greatest proportion of weekly captures was
moulting. When presenting the results I use the term population-level peak to describe the
clear peak only observed in the Northern Waterthrush and Ovenbird.
Statistical analyses
To compare the prevalence of prealternate moult across the six species, I used a
Pearson’s chi-squared test of the proportion of individuals from each species that were
moulting versus not moulting, excluding captures from before the observed onset of
moult in each species. I used ANOVA to compare mean moult intensity (CCMI) across
species. For all statistical analyses, I evaluated the significance of parameter estimates at

25
a α = 0.05. The results did not change qualitatively for Pearson’s chi-squared test or
ANOVA when I excluded recaptures, so I report test statistics from tests that included
recapture data. I performed all statistical analyses in R 4.4.0 (R Core Team 2024).
I used paired t-tests to determine if moult intensity, as measured by CCMI,
differed between the head and body patch groups within individuals of each species. To
compare the prevalence of head feather moult against body feather moult, I used chisquared tests of given probabilities of the proportion of moulting individuals of each
species moulting head or body feathers. I did not detect differences for any species in the
prevalence of active head versus body moult (p ≥ 0.14), so I do not present the individual
test statistics. The results of paired t-tests and chi-squared tests did not change
qualitatively when I excluded recaptures, so I report test statistics from paired t-tests that
included recapture data.
I examined the influence of Julian date, age, sex, and their interactions on moult
intensity (CCMI) for each species using a linear mixed-effect (function lmer, R package
lme4; Bates et al. 2015) or linear model (function lm) and a stepwise backward variable
removal procedure where I started with the initial full model and iteratively removed
non-significant terms (p > 0.05) to arrive at a single best fit model. I ran a separate model
for each species because exploratory analysis revealed a strong species effect on moult
intensity. I tested whether the explanatory variables predicted moult intensity during the
moult period by excluding capture data prior to the observed onset of moult. I ran a linear
mixed-effect model with the individual as a random effect to account for recaptures for
Ovenbird, Northern Waterthrush, and Black-and-white Warbler. I ran a linear model with
no random effect for the remaining species because I either had no recaptures (Northern
Parula), or the recaptures after the observed onset of moult did not contain all age-sex
combinations (American Redstart and Prairie Warbler), making model fitting unreliable.
For American Redstart and Prairie Warbler, I ran the model three ways to handle
individuals with repeated measurements. First, I included only the first capture record.
Second, I included only the second capture record. Finally, I included all capture records
without accounting for the effect of individual, for the sake of comparison. The estimates
for explanatory variables varied among approaches between 0 and 42.3%, but the results
did not differ qualitatively (p > 0.05), so here I report the models including only the first

26
capture records (see Appendix A, Table A.1 for the results of all approaches). For
American Redstart and Prairie Warbler only, I found a quadratic relationship between
moult intensity and Julian date, so I included Julian date2 as an explanatory variable in
the initial model to account for both linear and quadratic components of the relationship.
I present results below only for the species in which moult intensity was predicted by at
least one significant explanatory variable which included Northern Waterthrush, Blackand-white Warbler, American Redstart, and Prairie Warbler but excluded Ovenbird and
Northern Parula.
RESULTS
In total I recorded the presence or absence of contour feather moult 418 times for
six focal species combined. Those records came from 349 unique individuals, 54 (15%)
of which I recaptured at least once, with 15 individuals recaptured two, three, or four
times. The average time between successive captures was 24.8 ± 19.7 days (mean ± SD).
In total I determined that 64% of birds were actively moulting at capture (n = 268 of
418).
I assigned CCMI scores for 408 captures (first and recaptures combined); I was
unable to assign CCMI scores for 10 captures due to high capture volumes and time
constraints. For 51 individuals, I successfully scored moult two to five times throughout
the season. In Table 2.1, I summarize for each species the percentage of birds I scored for
moult more than once (i.e., recaptures), the total number of moult scores, and the
percentage of birds moulting at capture.
Each species showed active prealternate moult, involving some scattered head and
body feathers or the simultaneous moult of many feathers from multiple tracts. The
observed onset of moult in the study species ranged from late January to early March,
though for most species, moult commenced in early to mid-February (Figure 2.2). The
prevalence of moult generally increased in all species from the observed onset through
the end of the study period. After the observed onset of moult, the proportion of moulting
individuals varied markedly across species (Pearson’s chi-squared test: X2(5) = 22.3, p <
0.001, ntotal = 342; see Appendix A, Table A.2 for each species’ sample sizes).

27
Table 2.1. Summary for each study species of the unique individuals captured by age and sex, the percentage of unique
individuals scored for moult more than once, the total number of moult scores, the percentage of birds scored for moult that
were in active moult, the mean ± SD Combined Contour Moult Index (CCMI) and the range of CCMI values for moulting
birds. n is the same for percentage in active moult (%) and mean ± SD CCMI unless specified. n is the same for mean ± SD
CCMI and CCMI range.

a

Species

Number of
unique
individuals

Sex ratio
(M:F)

Percentage (%) of
unique individuals
with > 1 moult score
(i.e., recaptures)

Total number of
moult scores (initial
and recaptures
combined)

Percentage (%)
of scored birds
in active moult

Mean ± SD
CCMI

CCMI
range

Ovenbird

33; 9 firstcycle

NA

12 (n = 4)

37a

26 (n = 10)

0.29 ± 0.23
(n = 9a)

0.06 –
0.59

Northern
Waterthrush

95b; 39
first-cycle

NA

16 (n = 15)

110

67 (n = 75)

0.25 ± 0.26

0.06 –
1.41

Black-andwhite Warbler

73; 23 firstcycle

29:44

22 (n = 16)

99c

62 (n = 62)

0.74 ± 0.72
(n = 61c)

0.06 –
3.35

American
Redstart

100; 26
first-cycle

58:42

15 (n = 15)

112d

79 (n = 89)

0.50 ± 0.47
(n = 81d)

0.06 –
1.88

Northern
Parula

21; 8 firstcycle

13:8

0

21

67 (n = 14)

0.34 ± 0.23

0.06 –
0.77

Prairie
Warbler

27; 19 firstcycle

15:12

4 (n = 1)

28

64 (n = 18)

0.26 ± 0.21

0.06 –
0.59

One additional Ovenbird was moulting at first capture, but I did not collect the patch scores needed to calculate CCMI.
Four individuals were scored for moult but not aged.
c
One additional Black-and-white Warbler was moulting at second capture, but I did not collect the patch scores needed to calculate CCMI.
d
I recorded the presence of active contour moult for an additional 8 American Redstart captures (6 first capture, 1 second capture, 1 third
capture) but did not collect the patch scores needed to calculate CCMI.
b

28

Figure 2.2. The proportion of captured birds of six species moulting by week of year
from January 13 through April 18, 2023, at Font Hill Nature Preserve, Jamaica.
I observed both intra- and inter-specific variation in individual moult intensity.
However, while mean moult intensity varied across species (ANOVA: F(5) = 7.39, p <
0.001), all species had a mean CCMI score < 1 (Table 2.1), meaning that on average,
birds from all species were not actively moulting more than 25% of their contour feathers
at once. Only a few Black-and-white Warblers (n = 4) had CCMI scores ≥ 2, as in
simultaneously moulting more than 25% of contour feathers.

29
Species accounts
Ovenbird
I captured moulting Ovenbirds three to six weeks later than other species in this
study, between March 9 and April 9 (Figure 2.2). The prevalence of moult within the
population peaked from April 3 to April 6 (83% of captures that week were in moult). I
observed active moult in all feather patches except for the legs and wingpits, and most
frequently on the chin and cheeks (Figure 2.3). I did not detect a difference in moult
intensity between the head and the body when comparing CCMI scores for the two patch
groups (paired t-test: t(8) = 1.91, p = 0.09; Figure 2.4). While in other species I found clear
statistically significant trends in increasing moult intensity over time, Ovenbirds showed
a similar general pattern albeit not statistically significant (Julian date: df = 15, t = 1.68, p
= 0.11).
Northern Waterthrush
I captured the first moulting Northern Waterthrush earlier than all other species in
this study, on January 25 (Figure 2.2). The prevalence of moult peaked twice during the
study period, from February 19 to 25 and March 26 to April 1, with all captures moulting
during those weeks. After April 1, moult prevalence remained high (≥ 88% of weekly
captures in moult) through April 18. Waterthrush showed active moult in all feather
patches except for the belly (Figure 2.3). Active moult was most frequent on the chin,
throat, and cheek. Per feather patch, mean moult intensity was greatest on the rump, the
only feather patch with more than 50% of feathers in active moult (mean CCMI score >
2; Figure 2.5). Moult intensity also differed by patch group; mean CCMI was greater on
the head than the body (paired t-test: t(74) = 5.58, p < 0.0001; Figure 2.4). When I
examined changes in moult intensity over time from the observed onset of moult in the
study population (January 25), I found that CCMI increased linearly through the end of
the study period (Julian date: df = 104, t = 3.73, p = 0.003).

30

Figure 2.3. The proportions of birds from six species moulting each of 17 contour feather
patches captured from January 13 through April 18, 2023, at Font Hill Nature Preserve,
Jamaica.

31

Figure 2.4. The intensity of moult (scale of 0-4) for head feathers and body feathers in six species captured from January 13 through
April 18, 2023, at Font Hill Nature Preserve, Jamaica. The size of the circles represents the number (n) of birds sampled that had the
corresponding moult intensity score for each patch group.

32

Figure 2.5. The intensity of moult (scale of 0-4) per 17 feather patches in six species captured from January 13 through April 18,
2023, at Font Hill Nature Preserve, Jamaica. The size of the circles represents the number (n) of birds sampled that had the
corresponding moult intensity score for each patch.

33
Black-and-white Warbler
Black-and-white Warblers showed active moult between February 15 and April 18
(Figure 2.2). Moult was most prevalent within the population from March 19 onward
when most (≥ 75%) or all captures each week were moulting. I observed active moult in
all 17 feather patches, and most frequently on the chin, cheek, and throat (Figure 2.3).
Mean moult intensity per patch was greatest on the back and chin (mean CCMI scores >
2.8; Figure 2.5), and when compared between the head and body patch groups was
greater on the head (paired t-test: t(60) = 6.70, p < 0.0001; Figure 2.4).
When I examined changes in moult intensity over time from the observed onset of
moult (February 15), I found an Age*Julian Date interaction (df = 62.87, t = 3.27, p =
0.002) indicating that CCMI increased more strongly over time in first-cycle than
definitive-cycle birds (Figure 2.6). When I examined the age classes separately, both firstcycle (Julian date: df = 23.29, t = 5.12, p < 0.001) and definitive-cycle (Julian date: df =
69.79, t = 7.74, p < 0.001) birds saw an increase in CCMI over time.
American Redstart
I captured moulting American Redstarts between February 12 and April 18
(Figure 2.2). Moult was most prevalent within the population from February 26 onward
when most (≥ 86%) or all captures each week were moulting. American Redstarts showed
active moult in all 17 feather patches, and most frequently on the cheek, breast, and
crown (Figure 2.3). The back, rump, uppertail coverts, and undertail coverts had the
greatest mean moult intensities of all patches (mean CCMI scores >2; Figure 2.5).
However, moult intensity as measured by CCMI did not differ between the head and
body patch groups (paired t-test: t(80) = -0.70, p = 0.49; Figure 2.4). When I examined the
relationship between moult intensity and age from the observed onset of moult in the
study population (February 12), CCMI was greater for first-cycle than definitive-cycle
individuals (Age: df = 82, t = 3.55, p = 0.0007; Figure 2.7). Moult intensity followed a
parabola, with peak CCMI in the middle of the observed moult period based on the
quadratic term for time (Julian date: df = 82, t = 2.67, p = 0.009; Julian date2: df = 82, t =
-2.71, p = 0.008; Figure 2.7).

34

Intensity of contour feather moult
(CCMI, scale of 0−4)

2

1

0

−1

60

80

100

Julian date
Age

Definitive−cycle

First−cycle

Figure 2.6. Predicted linear relationships between contour feather moult intensity in
Black-and-white Warblers, measured as a Combined Contour Moult Index (CCMI) on a
scale of zero to four, and Julian date from the observed onset of moult, February 15,
2023, through April 18, 2023, interacting with age class. The solid lines represent the
predicted relationship between moult intensity and Julian date for each age class. The
shaded areas represent 95% confidence intervals.

35

Intensity of contour feather moult
(CCMI, scale of 0−4)

0.8

0.4

0.0

40

60

80

100

Julian date
Age

Definitive−cycle

First−cycle

Figure 2.7. Predicted parabolic relationships between contour feather moult intensity in
American Redstarts, measured as a Combined Contour Moult Index (CCMI) on a scale of
zero to four, and Julian date from the observed onset of moult, February 15, 2023,
through April 18, 2023, grouped by age class. The solid lines represent the predicted
relationship between moult intensity and Julian date for each age class. The shaded areas
represent 95% confidence intervals.
Northern Parula
I captured Northern Parulas in moult between February 6 and April 11 (Figure
2.2). Weekly capture sample sizes were small (n = 1, 3, or 7), making it difficult to
determine the population-level peak of moult prevalence. Northern Parulas showed active
moult in all feather patches except for the shoulders and wingpits (Figure 2.3). Active
moult most frequent on the chin, crown, and cheeks. Per feather patch, mean moult
intensity was greatest on the crown, chin, and throat (mean CCMI scores > 1.8; Figure

36
2.5). Northern Parulas moulted head feathers more intensely than body feathers, based on
a comparison of mean CCMI scores for the head and body patch groups (paired t-test:
t(13) = 3.24, p = 0.006; Figure 2.4).
Prairie Warbler
Prairie Warblers showed prealternate moult between February 12 and April 3
(Figure 2.2). All birds captured between February 12 and April 3 were moulting, aside
from 3 individuals (first-cycle male, first-cycle female, and definitive-cycle female)
captured on March 23, though sample sizes for weekly captures were small (n ≤ 6). I
observed active moult in all feather patches except for the belly, legs, and shoulders
(Figure 2.3). Active moult was most frequent on the cheeks, crown, and chin. Prairie
Warblers moulted the uppertail coverts most intensely of all 17 patches (Figure 2.5), but
moult intensity was greater on the head than the body when comparing the two patch
groups (paired t-test: t(17) = 4.31, p < 0.001; Figure 2.4). Moult intensity followed a
parabola, with peak CCMI in the middle of the observed moult period based on the
quadratic term for time (Julian date: df = 17, t = 2.96, p = 0.009; Julian date2: df = 17, t =
-3.23, p = 0.005).
DISCUSSION
Here, I provide detailed information on the timing, patterns, and intensity of
prealternate moult in six parulid warbler species–information that remains extremely
limited for most parulid warblers, despite the critical importance of feather quality for
these birds. I demonstrate that head and body feather moult is common in the Northern
Waterthrush, Black-and-white Warbler, American Redstart, Northern Parula, and Prairie
Warbler, and for most species increases in frequency and intensity later in the
nonbreeding period, though timing varied markedly across species. To the best of my
knowledge, I provide the first detailed account that some, though likely not all, Ovenbirds
complete a prealternate moult on the nonbreeding grounds. I also demonstrate agespecific patterns in the intensity of moult in the American Redstart and Black-and-white
Warbler.

37
The evolutionary drivers behind prealternate moult in birds remain unclear, but
the feather wear hypothesis proposed by Pyle and Kayhart (2010), and supported by
Wolfe and Pyle (2011), Terrill et al. (2020), and Cuervo et al. (2022), may explain the
age-specific differences in moult intensity that I observed in the American Redstart and
Black-and-white Warbler. The feather wear hypothesis proposes that prealternate moult
evolved primarily to renew feathers for maintaining feather function, rather than for
breeding plumage, but was then co-opted during the evolution of seasonal dichromatism.
In both the American Redstart (Sherry et al. 2020) and Black-and-white-Warbler (Kricher
2020), first-cycle birds of both sexes replace most or all juvenile contour feathers during
the preformative moult. Consistent with the feather wear hypothesis, first-cycle birds may
benefit more than definitive-cycle birds from an intense prealternate moult if they are
replacing relatively poor-quality contour feathers (i.e., either retained juvenile or
formative feathers). In the American Redstart, prealternate moult can contribute to the
slow acquisition of black (definitive) feathers in first-cycle males (Rohwer et al. 1983).
First-cycle males in this study did not exclusively replace grey contour feathers with
black feathers; some new feathers were black, some were grey, and some birds moulted
new feathers in both colours simultaneously. This is contrary to the results of Tonra et al.
(2014), where the experimentally regrown breast feathers of first-cycle males appeared to
be exclusively black. The production of grey alternate feathers in first-cycle males is
another example of a prealternate moult that does not contribute to plumage maturation
or seasonal dichromatism, suggesting that the replacement of these feathers may be
primarily driven by feather maintenance. Further supporting the feather wear hypothesis,
the patches most frequently moulted across all species tended to be areas of the body that
are exposed to the sun when birds are perching (Terrill et al. 2020): the head, in particular
the cheek, chin, and crown, followed by the breast.
In most other parulid warblers that complete a prealternate moult on the
nonbreeding grounds, their moult patterns are not known to differ by age or sex class
(Pyle 2022b), but more detailed studies of parulid warbler moult on the nonbreeding
grounds could elucidate age- or sex-based differences. I observed a pattern of higher
moult intensity in first-cycle Black-and-white Warblers and American Redstarts, a pattern
that was recently documented in the Prothonotary Warbler (Protonotaria critea; Ames

38
2021). For the American Redstart, Pyle (2022b) reports that prealternate moult is most
evident in first-cycle males. While I observed greater moult intensity in first-cycle
American Redstarts, I did not detect a significant difference in moult intensity between
the sexes. Contrary to Nolan 1978 (as cited in Nolan Jr et al. 2020), I did not find a sexbased difference in moult for Prairie Warblers, but my results may be limited in that I was
able to determine moult intensity at capture but not the full moult extent for the moult
period. Boone (2006) found age- and sex-based differences in the prealternate moult of
the Magnolia Warbler but captured migrating birds post-moult and quantified moult
extent based only on the wing coverts and amount of dark streaking on the breast, a part
of the plumage that changes in colour from the basic to alternate plumage. For species,
age, or sex classes that replace basic contour feathers in the same colour as alternate
feathers, as I have documented in all age and sex classes of American Redstarts, it is
possible that previous studies of moult and plumage may have missed the replacement of
same-colour alternate feathers or mischaracterized late-winter contour feather moult as
adventitious rather than obligate. I suggest that by collecting detailed moult data at the
feather patch level across the entire head and body during the moult period on the
nonbreeding grounds, along with more rigorous mark-recapture efforts through the moult
period, further studies can clarify our knowledge of prealternate moult and its potential
variation across species, age and sex classes, nonbreeding populations, and years.
Moult intensity often follows a parabolic curve from start to finish, peaking
during the middle of moult (e.g., Wright et al. 2018, Hutton et al. 2021, Guallar 2024). I
observed this pattern in the American Redstart and Prairie Warbler but only captured a
linear increase in moult intensity over time in the other species, indicating that the moult
period for the Ovenbird, Northern Waterthrush, Black-and-white Warbler, and Northern
Parula may have extended beyond April 18. Individuals from all six study species showed
active moult into early or mid-April, coinciding with timing windows for pre-breeding
migration departures from Jamaica (Fink et al. 2023; Chapter 1, Table 1.1). This result
suggests that for many birds, moult likely overlaps with migratory fattening, and that
some birds may continue moulting until departure. Mid-April aligns with early departures
for Ovenbird, Northern Waterthrush, Black-and-white Warbler, and American Redstart
(Studds and Marra 2007, 2011, Tonra et al. 2013, Cooper et al. 2015), but late departures

39
for Northern Parula and Prairie Warbler (Fink et al. 2023). As far as I know, most
Nearctic-Neotropical migrant parulids have prealternate moult schedules that occur
before pre-breeding migration (Billerman et al. 2022, Pyle 2022b), likely to avoid
energetic conflicts (Froehlich et al. 2005). Because migration constrains their prealternate
moult spatially and temporally, and the resources needed to facilitate energetically
demanding events vary seasonally in Jamaica (Parrish and Sherry 1994, Strong and
Sherry 2000, Johnson and Sherry 2001, Studds and Marra 2007, Studds and Marra 2011),
the study species may be faced with trade-offs between migratory fattening and
maximizing alternate feather quality by moulting closer to departure (Froehlich et al.
2005). Future studies should explore this potential trade-off and evaluate its
consequences by determining if the timing of individual moult is associated with the
timing of migratory fattening and pre-breeding migration departure. Collecting such data
for the same individuals over multiple years could begin to elucidate whether individuals
adjust their moult and migration strategies according to interannual variation in seasonal
resource availability during the nonbreeding period.
With an improved understanding of prealternate moult in six species of NearcticNeotropical migratory warblers, I can now use data on the timing and intensity of moult
to begin to evaluate how prealternate moult interacts across the annual cycle with other
crucial life events, including pre-breeding migration. Understanding the individual timing
of migration in relation to moult, together with the underlying influence of habitat quality
and individual body condition on moult and migration, are the next steps in disentangling
the role of prealternate moult in potential nonbreeding period carry-over effects in the
study species. This knowledge is critical for determining which life stages limit bird
populations in the annual cycle (Marra et al. 2015a). To that end, it will be crucial for
researchers in the global north to establish equitable collaborations with local neotropical
researchers (Soares et al. 2023) to aid in furthering prealternate moult research on the
nonbreeding grounds of many migratory birds.

40
LITERATURE CITED
Ames, E. M. (2021). Linking events across the annual cycle, in a Neotropical migratory
songbird of conservation concern, the Prothonotary Warbler (Protonotaria citrea)
[Ph.D. thesis]. The Ohio State University.
Barta, Z., McNamara, J. M., Houston, A. I., Weber, T. P., Hedenström, A., & Feró, O.
(2008). Optimal moult strategies in migratory birds. Philosophical Transactions
of the Royal Society B: Biological Sciences, 363(1490), 211–229.
doi.org/10.1098/rstb.2007.2136
Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015). Fitting linear mixed-effects
models using lme4. Journal of Statistical Software, 67(1).
doi.org/10.18637/jss.v067.i01
Billerman, S. M., Keeney, B. K., Rodewald, P. G., & Schulenberg, T. S. (Eds.). (2022).
Birds of the World. Cornell Lab of Ornithology.
https://birdsoftheworld.org/bow/home
Boone, A. T. (2006). Seasonal interactions between migration and winter in a migratory
songbird, the Magnolia Warbler (Dendroica magnolia) [Master’s thesis]. The
Ohio State University.
Boone, A. T., Rodewald, P. G., & DeGroote, L. W. (2010). Neotropical winter habitat of
the magnolia warbler: Effects on molt, energetic condition, migration timing, and
hematozoan infection during spring migration. The Condor, 112(1), 115–122.
doi.org/10.1525/cond.2010.090098
Bridge, E. S. (2011). Mind the gaps: What’s missing in our understanding of feather molt.
The Condor, 113(1), 1–4. doi.org/10.1525/cond.2011.100228
Brunner, A. R., Marra, P. P., & Tonra, C. M. (2022). Vulnerable Neotropical migratory
songbird demonstrates flexibility in space use in response to rainfall change.
Ornithology, 139(2), 1-14. doi.org/10.1093/ornithology/ukac005
Carlisle, J. D., Kaltenecker, G. S., & Swanson, D. L. (2005). Molt strategies and age
differences in migration timing among autumn landbird migrants in southwestern
Idaho. The Auk, 122(4), 1070–1085.
Carnes, B. H., Godwin, C. M., Foster, K. R., & Pyle, P. (2021). Clarification of molt
strategies in three Empidonax flycatchers. The Wilson Journal of Ornithology,
133(1), 34-42. doi.org/10.1676/20-00109
Cimprich, D. A. (2018). Molt limits reveal extent of molt and age in Black-capped
Vireos. North American Bird Bander, 43(4), 86–95.

41
Cooper, N. W., Sherry, T. W., & Marra, P. P. (2015). Experimental reduction of winter
food decreases body condition and delays migration in a long-distance migratory
bird. Ecology, 96(7), 1933–1942. doi.org/10.1890/14-1365.1
Cooper, N. W., Thomas, M. A., & Marra, P. P. (2021). Vertical sexual habitat segregation
in a wintering migratory songbird. The Auk, 138, 1–11.
doi.org/10.1093/ornithology/ukaa080
Crary, A. L., & Rodewald, P. G. (2012). Plumage coloration and ornamentation as
predictors of nest survival and number of young fledged in Yellow Warblers.
Journal of Field Ornithology, 83(2), 130–140. doi.org/10.1111/j.15579263.2012.00363.x
Cuervo, J. J., Morales, J., Soler, J. J., & Moreno, J. (2022). Sexual selection, feather wear,
and time constraints on the pre‐basic molt explain the acquisition of the pre‐
alternate molt in European passerines. Ecology and Evolution, 12(9), e9260.
doi.org/10.1002/ece3.9260
Cyr, N. E., Wikelski, M., & Romero, L. M. (2008). Increased energy expenditure but
decreased stress responsiveness during molt. Physiological and Biochemical
Zoology, 81(4), 452–462. doi.org/10.1086/589547
Danner, R. M., Greenberg, R. S., Danner, J. E., & Walters, J. R. (2015). Winter food
limits timing of pre‐alternate moult in a short‐distance migratory bird. Functional
Ecology, 29(2), 259–267. doi.org/10.1111/1365-2435.12322
de la Hera, I., DeSante, D. F., & Milá, B. (2012). Feather growth rate and mass in
Nearctic passerines with variable migratory behavior and molt pattern. The Auk,
129(2), 222–230. doi.org/10.1525/auk.2012.11212
Dietz, M. W., Daan, S., & Masman, D. (1992). Energy requirements for molt in the
kestrel Falco tinnunculus. Physiological Zoology, 65(6), 1217–1235.
doi.org/10.1086/physzool.65.6.30158276
Dossman, B. C., Studds, C. E., LaDeau, S. L., Sillett, T. S., & Marra, P. P. (2023). The
role of tropical rainfall in driving range dynamics for a long-distance migratory
bird. Proceedings of the National Academy of Sciences, 120(52), e2301055120.
doi.org/10.1073/pnas.2301055120
Dugger, K. M., Faaborg, J., & Arendt, W. J. (2004). Understanding survival and
abundance of overwintering warblers: Does rainfall matter? The Condor, 106(4),
744–760.
Fink, D., Auer, T., Johnston, A., Strimas-Mackey, M., Ligocki, S., Robinson, O.,
Hochachka, W., Jaromczyk, L., Crowley, C., Dunham, K., Stillman, A., Davies, I.,

42
Rodewald, A., Ruiz-Gutierrez, V., & Wood, C. (2023). eBird Status and Trends
[Dataset]. doi.org/10.2173/ebirdst.2022
Froehlich, D. R., Rowher, S., & Stutchbury, B. J. (2005). Spring molt constraints versus
winter territoriality: Is conspicuous winter coloration maladaptive? In R. S.
Greenberg & P. P. Marra (Eds.), Birds of two worlds: The ecology and evolution
of migration. The Johns Hopkins University Press.
Greenwood, H., Weatherhead, P. J., & Titman, R. D. (1983). A new age- and sex-specific
molt scheme for the Red-Winged Blackbird. The Condor, 85(1), 104–105.
doi.org/10.2307/1367901
Guallar, S. (2024). From partial to complete: Wing- and tail-feather moult sequence and
intensity depend on species, life-cycle stage, and moult completeness in
passerines. Avian Research, 15, 100163. doi.org/10.1016/j.avrs.2024.100163
Gunderson, A. R., Frame, A. M., Swaddle, J. P., & Forsyth, M. H. (2008). Resistance of
melanized feathers to bacterial degradation: Is it really so black and white?
Journal of Avian Biology, 39(5), 539–545. doi.org/10.1111/j.09088857.2008.04413.x
Hedenström, A. (2002). Aerodynamics, evolution and ecology of avian flight. Trends in
Ecology & Evolution, 17(9), 415–422. doi.org/10.1016/S0169-5347(02)02568-5
Hill, G. E., & McGraw, K. J. (Eds.). (2006). Bird coloration: Function and evolution:
Vol. II. Harvard University Press.
Howell, S. N. G., Corben, C., Pyle, P., & Rogers, D. (2003). The first basic problem: A
review of molt and plumage homologies. The Condor, 105(4), 635–653.
Humphrey, P. S., & Parkes, K. C. (1959). An approach to the study of molts and
plumages. The Auk, 76(1), 1–31. doi.org/10.2307/4081839
Hutton, P., McKenna, J., & McGraw, K. J. (2021). Urban links to molt schedule, body
condition and carotenoid‐based coloration in the house finch Haemorhous
mexicanus. Journal of Avian Biology, 52(5), jav.02761. doi.org/10.1111/jav.02761
Jackson, W. M., Wood, C. S., & Rohwer, S. (1992). Age-specific plumage characters and
annual molt schedules of Hermit Warblers and Townsend’s Warblers. The Condor,
94(2), 490–501. doi.org/10.2307/1369221
Jenni, L., & Winkler, R. (2020a). Moult and ageing of European passerines (Second
edition). Helm.
Jenni, L., & Winkler, R. (2020b). The biology of moult in birds. Helm.

43
Johnson, M. D., & Sherry, T. W. (2001). Effects of food availability on the distribution of
migratory warblers among habitats in Jamaica. Journal of Animal Ecology, 70(4),
546–560. doi.org/10.1046/j.1365-2656.2001.00522.x
Jones, T. B., Drake, A., & Green, D. J. (2014). Individual quality explains association
between plumage colouration, arrival dates and mate acquisition in yellow
warblers (Setophaga petechia). BMC Ecology, 14(1), 13. doi.org/10.1186/14726785-14-13
Kent, C. M., Huh, K. M., Hunter, S. C., Judson, K., Powell, L. L., & Sherry, T. W. (2022).
High resource overlap and small dietary differences are widespread in food‐
limited warbler (Parulidae) communities. Ibis, 164(1), 44–59.
doi.org/10.1111/ibi.13006
Kent, C. M., & Sherry, T. W. (2020). Behavioral niche partitioning reexamined: Do
behavioral differences predict dietary differences in warblers? Ecology, 101(8).
doi.org/10.1002/ecy.3077
Kiat, Y. (2023). Moult terminology: Let’s make it simpler! Ibis, 165(2), 697–703.
doi.org/10.1111/ibi.13164
Kiat, Y., & Izhaki, I. (2016). Moult strategies affect age differences in autumn migration
timing in east Mediterranean migratory passerines. PLOS ONE, 11(1), e0147471.
doi.org/10.1371/journal.pone.0147471
Kiat, Y., Izhaki, I., & Sapir, N. (2018). The effects of long‐distance migration on the
evolution of moult strategies in Western‐Palearctic passerines. Biological
Reviews, 94(2), 700–720. doi.org/10.1111/brv.12474
Kiat, Y., Vortman, Y., & Sapir, N. (2019). Feather moult and bird appearance are
correlated with global warming over the last 200 years. Nature Communications,
10(1), 2540. doi.org/10.1038/s41467-019-10452-1
Kricher, J. C. (2020). Black-and-white warbler (Mniotilta varia). In A. F. Poole (Ed.),
Birds of the world (version 1.0). Cornell Lab of Ornithology.
doi.org/10.2173/bow.bawwar.01
Leu, M., & Thompson, C. W. (2002). The potential importance of migratory stopover
sites as flight feather molt staging areas: A review for neotropical migrants.
Biological Conservation, 106(1), 45–56. doi.org/10.1016/S0006-3207(01)002282
Lindström, Å., Visser, G. H., & Daan, S. (1993). The energetic cost of feather synthesis is
proportional to basal metabolic rate. Physiological Zoology, 66(4), 490–510.

44
Lourenço, P. M. (2014). The energetic cost of pre-alternate body moult in black-tailed
godwits (Limosa l. Limosa) staging in Portugal during northward migration. Airo,
23, 3–9.
Marra, P. P., Cohen, E. B., Loss, S. R., Rutter, J. E., & Tonra, C. M. (2015a). A call for
full annual cycle research in animal ecology. Biology Letters, 11(8), 20150552.
doi.org/10.1098/rsbl.2015.0552
Marra, P. P., Studds, C. E., Wilson, S., Sillett, T. S., Sherry, T. W., & Holmes, R. T.
(2015b). Non-breeding season habitat quality mediates the strength of densitydependence for a migratory bird. Proceedings of the Royal Society B: Biological
Sciences, 282(1811), 20150624. doi.org/10.1098/rspb.2015.0624
Mettke-Hofmann, C., Sinclair, P. H., Hamel, P. B., & Greenberg, R. (2010). Implications
of prebasic and a previously undescribed prealternate molt for aging Rusty
Blackbirds. The Condor, 112(4), 854–861. doi.org/10.1525/cond.2010.100150
Moldenhauer, R. R., & Regelski, D. J. (2020). Northern parula (Setophaga americana).
In A. F. Poole (Ed.), Birds of the world (version 1.0). Cornell Lab of Ornithology.
doi.org/10.2173/bow.norpar.01
Nolan Jr, V., Ketterson, E. D., & Buerkle, C. A. (2020). Prairie warbler (Setophaga
discolor). In A. F. Poole (Ed.), Birds of the world (version 1.0). Cornell Lab of
Ornithology. doi.org/10.2173/bow.prawar.01
Parrish, J. D., & Sherry, T. W. (1994). Sexual habitat segregation by American Redstarts
wintering in Jamaica: Importance of resource seasonality. The Auk, 111(1), 38–49.
doi.org/10.2307/4088503
Porneluzi, P., Van Horn, M. A., & Donovan, T. M. (2020). Ovenbird (Seiurus
aurocapilla). In A. F. Poole (Ed.), Birds of the World. Cornell Lab of Ornithology.
doi.org/10.2173/bow.ovenbi1.01
Powell, L. L., Ames, E. M., Wright, J. R., Matthiopoulos, J., & Marra, P. P. (2021).
Interspecific competition between resident and wintering birds: Experimental
evidence and consequences of coexistence. Ecology, 102(2), e03208.
doi.org/10.1002/ecy.3208
Pyle, P. (2022a). Defining moults in migratory birds: A sequence‐based approach.
Journal of Avian Biology, 2022(5). doi.org/10.1111/jav.02958
Pyle, P. (2022b). Identification guide to North American birds: Part I (2nd ed). Slate
Creek Press.
Pyle, P., & Carnes, B. H. (2022). Molt strategies by age and subspecies in the Willow
Flycatcher. Western Birds, 53(3), 216–231. doi.org/10.21199/WB53.3.4

45
Pyle, P., Gahbauer, M., Johnson, E. I., Ryder, T. B., & Wolfe, J. D. (2022). Application of
a global age-coding system (“WRP”), based on molts and plumages, for use in
demographic and other studies of birds. Ornithology, 139(1), 1-12.
doi.org/10.1093/ornithology/ukab063
Pyle, P., & Kayhart, R. (2010). Replacement of primaries during the prealternate molt of
a Yellow Warbler. North American Bird Bander, 35(4), 178–181.
Pyle, P., Saracco, J. F., & DeSante, D. F. (2018). Evidence of widespread movements
from breeding to molting grounds by North American landbirds. The Auk, 135(3),
506–520. doi.org/10.1642/AUK-17-201.1
Quinlan, S. P., & Green, D. J. (2011). Variation in deuterium (δD) signatures of Yellow
Warbler Dendroica petechia feathers grown on breeding and wintering grounds.
Journal of Ornithology, 152(1), 93–101. doi.org/10.1007/s10336-010-0553-5
R Core Team. (2024). R: A language and environment for statistical computing
[Computer software]. R Foundation for Statistical Computing. https://www.rproject.org/
Renfrew, R. B., Frey, S. J. K., & Klavins, J. (2011). Phenology and sequence of the
complete prealternate molt of Bobolinks in South America. Journal of Field
Ornithology, 82(1), 101–113. doi.org/10.1111/j.1557-9263.2010.00312.x
Rohwer, S. (2005). Timing of molt in migratory birds. In R. Greenberg & P. P. Marra
(Eds.), Birds of two worlds: The ecology and evolution of migration (pp. 204–
216). The Johns Hopkins University Press.
Rohwer, S., Klein, Jr., W. P., & Heard, S. (1983). Delayed plumage maturation and the
presumed prealternate molt in American Redstarts. The Wilson Bulletin, 95(2),
199–208.
Rosenberg, K. V., Dokter, A. M., Blancher, P. J., Sauer, J. R., Smith, A. C., Smith, P. A.,
Stanton, J. C., Panjabi, A., Helft, L., Parr, M., & Marra, P. P. (2019). Decline of
the North American avifauna. Science, 366(6461), 120–124.
doi.org/10.1126/science.aaw1313
Ruelas Inzunza, E., Cockle, K. L., Núñez Montellano, M. G., Fontana, C. S., Cuatianquiz
Lima, C., Echeverry-Galvis, M. A., Fernández-Gómez, R. A., Montaño-Centellas,
F. A., Bonaccorso, E., Lambertucci, S. A., Cornelius, C., Bosque, C., Bugoni, L.,
Salinas-Melgoza, A., Renton, K., Freile, J. F., Angulo, F., Mugica Valdés, L.,
Velarde, E., … Miño, C. I. (2023). How to include and recognize the work of
ornithologists based in the Neotropics: Fourteen actions for Ornithological
Applications, Ornithology, and other global-scope journals. Ornithological
Applications, 125(1), duac047. doi.org/10.1093/ornithapp/duac047

46
Saino, N., Szép, T., Ambrosini, R., Romano, M., & Møller, A. P. (2004). Ecological
conditions during winter affect sexual selection and breeding in a migratory bird.
Proceedings of the Royal Society of London. Series B: Biological Sciences,
271(1540), 681–686. doi.org/10.1098/rspb.2003.2656
Santos, E. S. A., Scheck, D., & Nakagawa, S. (2011). Dominance and plumage traits:
Meta-analysis and metaregression analysis. Animal Behaviour, 82(1), 3–19.
doi.org/10.1016/j.anbehav.2011.03.022
Sherry, T. W., Holmes, R. T., Pyle, P., & Patten, M. A. (2020). American Redstart
(Setophaga ruticilla). In A. F. Poole (Ed.), Birds of the world. Cornell Lab of
Ornithology. doi.org/10.2173/bow.amered.01
Sieburth, D., & Pyle, P. (2018). Evidence for a prealternate molt-migration in the Rufous
Hummingbird and its implications for the evolution of molts in Apodiformes. The
Auk, 135(3), 495–505. doi.org/10.1642/AUK-17-231.1
Smith, J. A. M., Reitsma, L. R., & Marra, P. P. (2011). Multiple space-use strategies and
their divergent consequences in a nonbreeding migratory bird (Parkesia
noveboracensis). The Auk, 128(1), 53–60. doi.org/10.1525/auk.2011.10241
Soares, L., Cockle, K. L., Ruelas Inzunza, E., Ibarra, J. T., Miño, C. I., Zuluaga, S.,
Bonaccorso, E., Ríos-Orjuela, J. C., Montaño-Centellas, F. A., Freile, J. F.,
Echeverry-Galvis, M. A., Bonaparte, E. B., Diele-Viegas, L. M., Speziale, K.,
Cabrera-Cruz, S. A., Acevedo-Charry, O., Velarde, E., Cuatianquiz Lima, C.,
Ojeda, V. S., … Martins, P. V. R. (2023). Neotropical ornithology: Reckoning
with historical assumptions, removing systemic barriers, and reimagining the
future. Ornithological Applications, 125(1), 1–31.
doi.org/10.1093/ornithapp/duac046
Sparrow, K. L., Donkor, K. K., Flood, N. J., Marra, P. P., Pillar, A. G., & Reudink, M. W.
(2017). Conditions on the Mexican moulting grounds influence feather colour and
carotenoids in Bullock’s orioles (Icterus bullockii). Ecology and Evolution, 7(8),
2643–2651. doi.org/10.1002/ece3.2836
Strong, A. M., & Sherry, T. W. (2000). Habitat-specific effects of food abundance on the
condition of ovenbirds wintering in Jamaica: Ovenbird body condition. Journal of
Animal Ecology, 69(5), 883–895. doi.org/10.1046/j.1365-2656.2000.00447.x
Studds, C. E., & Marra, P. P. (2007). Linking fluctuations in rainfall to nonbreeding
season performance in a long-distance migratory bird, Setophaga ruticilla.
Climate Research, 35, 115–122. doi.org/10.3354/cr00718
Studds, C. E., & Marra, P. P. (2011). Rainfall-induced changes in food availability modify
the spring departure programme of a migratory bird. Proceedings of the Royal

47
Society B: Biological Sciences, 278(1723), 3437–3443.
doi.org/10.1098/rspb.2011.0332
Stutchbury, B. J. M., Gow, E. A., Done, T., MacPherson, M., Fox, J. W., & Afanasyev, V.
(2011). Effects of post-breeding moult and energetic condition on timing of
songbird migration into the tropics. Proceedings of the Royal Society B:
Biological Sciences, 278(1702), 131–137. doi.org/10.1098/rspb.2010.1220
Terrill, R. S., Seeholzer, G. F., & Wolfe, J. D. (2020). Evolution of breeding plumages in
birds: A multiple‐step pathway to seasonal dichromatism in New World warblers
(Aves: Parulidae). Ecology and Evolution, 10(17), 9223–9239.
doi.org/10.1002/ece3.6606
Terrill, R. S., & Shultz, A. J. (2023). Feather function and the evolution of birds.
Biological Reviews, 98(2), 540–566. doi.org/10.1111/brv.12918
Tonra, C. M., Marini, K. L. D., Marra, P. P., Germain, R. R., Holberton, R. L., &
Reudink, M. W. (2014). Color expression in experimentally regrown feathers of
an overwintering migratory bird: Implications for signaling and seasonal
interactions. Ecology and Evolution, 4(8), 1222–1232. doi.org/10.1002/ece3.994
Tonra, C. M., Marra, P. P., & Holberton, R. L. (2013). Experimental and observational
studies of seasonal interactions between overlapping life history stages in a
migratory bird. Hormones and Behavior, 64(5), 825–832.
doi.org/10.1016/j.yhbeh.2013.10.004
Tonra, C. M., & Reudink, M. W. (2018). Expanding the traditional definition of moltmigration. The Auk, 135(4), 1123–1132. doi.org/10.1642/AUK-17-187.1
Vézina, F., Gustowska, A., Jalvingh, K. M., Chastel, O., & Piersma, T. (2009). Hormonal
correlates and thermoregulatory consequences of molting on metabolic rate in a
northerly wintering shorebird. Physiological and Biochemical Zoology, 82(2),
129–142. doi.org/10.1086/596512
Voelker, G., & McFarland, S. L. (2002). Molt patterns and molting grounds of Lucy’s and
Virginia’s Warblers: Similar yet different. The Wilson Bulletin, 114(2), 255–263.
doi.org/10.1676/0043-5643(2002)114[0255:MPAMGO]2.0.CO;2
Voelker, G., & Rohwer, S. (1998). Contrasts in scheduling of molt and migration in
eastern and western warbling-vireos. The Auk, 115(1), 142–155.
doi.org/10.2307/4089119
Whitaker, D. M., & Eaton, S. W. (2020). Northern Waterthrush (Parkesia
noveboracensis). In A. F. Poole (Ed.), Birds of the world (version 1.0). Cornell
Lab of Ornithology. doi.org/10.2173/bow.norwat.01

48
Wingfield, J. C. (2008). Organization of vertebrate annual cycles: Implications for control
mechanisms. Philosophical Transactions of the Royal Society B: Biological
Sciences, 363(1490), 425–441. doi.org/10.1098/rstb.2007.2149
Wolf, B. O., & Walsberg, G. E. (2000). The role of the plumage in heat transfer processes
of birds. American Zoologist, 40, 575–584.
Wolfe, J. D., Ryder, T. B., & Pyle, P. (2010). Using molt cycles to categorize the age of
tropical birds: An integrative new system: Molt-Cycle Age-Categorization
System. Journal of Field Ornithology, 81(2), 186–194. doi.org/10.1111/j.15579263.2010.00276.x
Wolfe, J., & Pyle, P. (2011). Featured photo: First evidence for eccentric prealternate
molt in the Indigo Bunting: Possible implications for adaptive molt strategies.
Western Birds, 42, 257–262.
Wright, J. R., Tonra, C. M., & Powell, L. L. (2018). Prealternate molt-migration in Rusty
Blackbirds and its implications for stopover biology. The Condor, 120(3), 507–
516. doi.org/10.1650/CONDOR-17-177.1

49
CHAPTER 3: PREALTERNATE MOULT LINKS NONBREEDING HABITAT
CONDITIONS AND MIGRATION DEPARTURE TIMING IN NEARCTICNEOTROPICAL MIGRATORY SONGBIRDS
ABSTRACT
The quality of nonbreeding habitat can carry over to influence the performance of
migratory birds during subsequent life history stages. Yet, for birds that undergo
prealternate moult during the nonbreeding period, its potential role in seasonal
interactions has largely been overlooked. Here, I build on a detailed quantification of the
timing and intensity of prealternate moult in three species of Nearctic-Neotropical
migratory songbirds to assess the role of moult in carry-over effects from the nonbreeding
period to the timing of migration departure: Black-and-white Warbler (Mniotilta varia),
Northern Waterthrush (Parkesia noveboracensis), and American Redstart (Setophaga
ruticilla). I assessed the influence of body condition on the timing and intensity of
prealternate moult in distinct habitats in Jamaica that differ in moisture regime: xeric
second-growth scrub forest and mesic mangrove. I analyzed stable-carbon isotope ratios
(δ13C) from birds’ red blood cells to support my understanding of relative habitat quality
for the study species. Using the Motus Wildlife Tracking System, I tracked precise dates
of migration departure so I could evaluate the influence of habitat quality, body
condition, and prealternate moult on departure timing. Moult intensity was generally
higher in first-cycle birds but its relationship with body condition varied among species,
indicating that body condition may not be the most important driver of prealternate moult
phenology. For American Redstarts, the effect of age on moult varied with habitat type,
suggesting that birds with the poorest quality feathers, associated with poorer quality,
open habitat, require a more intense moult than older birds and those in higher quality
habitat. I found the first potential evidence of a carry-over effect from prealternate moult
to departure timing in a parulid warbler, the American Redstart. Male American Redstarts
with high-intensity moult prior to departure, indicating a later moult, departed later,
suggesting that prealternate moult can represent an important seasonal interaction in
some birds. Ultimately, this study is an important step in beginning to understand the role

50
of prealternate moult in nonbreeding period carry-over effects and highlights that moult
should not be overlooked when considering a birds’ full annual cycle.
INTRODUCTION
To identify the drivers of population decline and ultimately conserve declining
migratory bird populations (Rosenberg et al. 2019), it is imperative to understand how
birds’ life history stages interact across their annual cycle to influence individual
performance and population dynamics (Marra et al. 2015). Breeding, migration, and
moult, arguably the most important and energy-intensive stages in the annual cycle of
migratory birds, are influenced by a combination of interacting endogenous and
exogenous factors (Barta et al. 2008). Endogenous factors such as physiology, and
exogenous factors such as environmental conditions, vary seasonally as migratory birds
track resources from their breeding grounds to stationary nonbreeding (hereafter
“nonbreeding”) grounds. Both the timing of a life history stage and an individual’s
physiological state during that event can carry over to influence the timing of subsequent
stages, often with fitness consequences (Norris and Marra 2007, Harrison et al. 2011).
For example, a bird that occupies xeric nonbreeding habitat (an exogenous factor) may
have poorer body condition (an endogenous factor) than birds occupying mesic habitat,
resulting in a delayed pre-breeding migration (e.g., Marra et al. 1998, Studds and Marra
2005, Reudink et al. 2009a) or delayed arrival to the breeding grounds culminating in
reduced reproductive success (e.g., Norris et al. 2004, Reudink et al. 2009a). In other
cases, birds can mitigate carry-over effects from one event through plastic behaviour, for
example by adjusting their migration rate (e.g., González et al. 2020) or reducing time
spent at stop-over sites (e.g., Stutchbury et al. 2011) to arrive on time despite a late
departure. Carry-over effects from the nonbreeding grounds to pre-breeding migration or
the breeding period are well documented across bird families from passerines to
shorebirds (e.g., Gill et al. 2001, Clements et al. 2022) but the role of moult, particularly
the prealternate moult that occurs on the nonbreeding grounds, has largely been
overlooked.
Many species of Nearctic-Neotropical and Afro-Palearctic migrants evolved to
moult twice in their annual cycle: once post-breeding (prebasic moult) and again prior to

51
breeding (prealternate moult). The evolution of this moult strategy, called a Complex
Alternate Strategy (Howell et al. 2003), has been linked to long-distance migration and
an associated longer photoperiod (Svensson and Hedenström 1999, Terrill et al. 2020,
Pageau et al. 2021, Cuervo et al. 2022). Exposure to ultra-violet radiation (Lennox and
Rowlands 1969, Surmacki 2008) degrades the non-renewable keratin structures of wing
and body feathers, necessitating the replacement of at least some feathers more than once
per year. The feather wear hypothesis proposes that the prealternate moult evolved
primarily for birds to renew feathers for maintaining feather function (Terrill et al. 2020).
Prealternate moult was likely then co-opted under sexual selection pressures for the
evolution of seasonal dichromatism (Pyle and Kayhart 2010, Terrill et al. 2020), as the
mechanism through which some species moult from drab basic, or nonbreeding, plumage
into usually more colourful alternate, or breeding, plumage. Because the feathers
resulting from prealternate moult serve an important structural function, are critical for
social and sexual signaling, and come at a high energetic cost, the timing and intensity of
prealternate moult have the potential to play a role in birds’ migratory performance and
subsequent reproductive success.
Moult is in part cued by photoperiod and hormones, but aspects of its phenology
including onset, duration, extent, and intensity, can also be influenced by environmental
conditions (e.g., Danner et al. 2015, Hutton et al. 2021), individual condition (e.g.,
Piersma and Jukema 1993, Hutton et al. 2021), or timing constraints imposed by breeding
(e.g., Morton and Welton 1973, Borowske et al. 2017) and migration (e.g., Conklin and
Battley 2012). The environmental conditions that birds experience during moult can also
influence the expression of plumage ornaments (e.g., Saino et al. 2004, Sparrow et al.
2017), which can have a direct impact on fitness (Romano et al. 2017). Despite evidence
that nonbreeding conditions such as food availability can influence prealternate moult
phenology (e.g., Danner et al. 2015) and that the prebasic and preformative moults can
influence both inter- (e.g., Carlisle et al. 2005, Bozó et al. 2017) and intraspecific (e.g.,
Stutchbury et al. 2011) migration timing, potential carry-over effects from the
prealternate moult to pre-breeding migration have received extremely limited attention.
The influence of prealternate moult strategy on the timing of other life stages,
including pre-breeding migration, has been subject to some consideration in shorebirds

52
(e.g., Buehler and Piersma 2008, Lourenço and Piersma 2015). However, I am aware of
only two studies investigating the potential link between prealternate moult and prebreeding migration in Nearctic-Neotropical migratory warblers (Family: Parulidae).
Those studies found no link between the prealternate moult and timing of arrival to a
stopover site (Boone et al. 2010) or the breeding grounds (Jones et al. 2014). However,
they did not account for the birds’ entire prealternate moult, only focussing on a particular
region of contour feathers and/or wing coverts. To better understand how the prealternate
moult interacts with pre-breeding migration phenology, it will be important to consider
moult across the entire body. Further, carry-over effects from the nonbreeding period to
pre-breeding migration vary among species, sexes, age classes, and years (e.g., Rockwell
2013, Akresh et al. 2019), highlighting the need for further study of the prealternate
moult in relation to subsequent life stages.
Here, I investigated the relationships between nonbreeding habitat quality,
individual body condition, moult intensity, and departure date, in three migratory parulid
warblers to understand if prealternate moult carries over to affect pre-breeding migration
phenology. First, I assessed the influence of habitat quality and body condition on moult
intensity throughout the prealternate moult period. Using the Motus Wildlife Tracking
System (Taylor et al. 2017), I tracked precise departure dates from the nonbreeding
grounds to assess whether departure was influenced by prealternate moult intensity. I
predicted that a combination of exogenous and endogenous factors would influence
prealternate moult intensity and departure dates. Specifically, that individuals occupying
mesic mangrove habitat would be in better condition and, in turn, would undergo a more
intense moult, and individuals who completed moult earlier, indicated by low moult
intensity prior to departure, would depart earlier.
METHODS
Study system
I studied nonbreeding populations of the migratory Black-and-white Warbler
(Mniotilta varia), American Redstart (Setophaga ruticilla), and Northern Waterthrush
(Parkesia noveboracensis) in Font Hill Nature Preserve (18.0391°N, 77.9411°W, <100 m
above sea level), St. Elizabeth Parish, on the southwest coast of Jamaica. I conducted this

53
research from January 13 to April 18, 2023, during the birds’ nonbreeding period, in two
habitat types: xeric second-growth dry scrub forest and mesic mangrove. A description of
vegetation structure and composition in the habitat types is available in Cooper et al.
(2021).
Precipitation at this site follows the highly seasonal pattern typical of the
neotropics: a wet season from August to November (average monthly precipitation > 100
mm) followed by a dry season from December through May (average monthly
precipitation <25 mm) (Powell et al. 2021), though conditions vary annually (Studds and
Marra 2011, Brunner et al. 2022, Dossman et al. 2023a). Relatively low levels of
precipitation preceded my study period and as a result, mangrove habitat held little and at
times no standing water. While similar conditions have been observed in dry years
(Cooper et al. 2015), mangrove plots typically hold 0.3 to 2 m of brackish water through
the dry season (Cooper et al. 2021). Scrub habitat remained dry without standing water,
as expected for a typically water stressed habitat (Cooper et al. 2021). See Chapter 2 for a
complete description of the number and size of habitat plots sampled.
Habitat quality and stable-carbon isotope sampling
At this site, a large body of evidence indicates that mangrove is a higher quality
habitat than scrub for American Redstarts (Marra et al. 1998, Studds and Marra 2007,
Reudink et al. 2009a, Studds and Marra 2011) due to greater water and associated
arthropod availability (Studds and Marra 2007, 2011). American Redstarts are highly
territorial at this site (Studds and Marra 2005, 2011), so habitat type at capture likely
represents the predominant habitat they occupied during the study period and can be used
as an explanatory variable in statistical analyses.
The relative quality of mangrove and scrub habitats at the site are not well
understood for the Black-and-White Warbler or Northern Waterthrush, though I predict
that mangrove will be higher quality given the estimates of greater arthropod availability.
A recent study comparing mass change among habitats suggests that mangrove may be
higher quality for the Black-and-white Warbler at this site (Cooper et al. 2021), and other
research has linked better condition with mesic nonbreeding habitat in the species
(Paxton and Moore 2015). Links between habitat quality and individual body condition

54
have not been tested for Northern Waterthrush at this site, but habitat quality increases
with moisture (i.e., standing water) for the species in other Caribbean systems (Puerto
Rico; Smith et al. 2010). However, through recapture data and re-sighting banded birds, I
determined that Black-and-white Warblers and Northern Waterthrushes ranged further
and regularly utilize multiple habitat types, thus I deemed capture location an unreliable
indicator of the habitat type they occupied. For these two species only, I used stablecarbon (δ13C) isotope ratios from blood to infer habitat use during the stationary
nonbreeding period. At the study site, mangrove and scrub habitats tend to have distinct
δ13C ratios because of differences in water stress and photosynthetic system in the plants
that dominate each habitat type (Marshall et al. 2007), though there is some overlap in
isotopic ratios (e.g., Marra et al. 1998, Reudink et al. 2009a). The isotope ratios of the
corresponding habitat incorporate into bird tissues through the insectivorous prey they
consume (reviewed by Kelly 2000), resulting in depleted stable-carbon ratios in birds
foraging in wetter mangrove (e.g., -26 to -20) relative to enriched ratios in birds foraging
in drier scrub (e.g., -22 to -20) (Marra et al. 1998, Reudink et al. 2009a).
While the turnover rate for blood plasma can be as quick as a few days, the
cellular fraction of blood (i.e., red blood cells) has a turnover rate of a few weeks (Oppel
and Powell 2010), making it a suitable indicator of the birds’ foraging habitat and diet in
the weeks leading up to sampling. I collected blood samples from the brachial vein of 18
Northern Waterthrushes and 26 Black-and-white Warblers on the date that I applied their
nanotags (see below for a full description of the tagging procedure). I separated the red
blood cells from blood plasma by centrifuging each sample within four hours of
collection and immediately froze the separate components. All samples were processed
and analyzed by The Cornell Isotope Laboratory of Cornell University (Ithaca, NY) in
December 2023 using a Thermo Delta V isotope ratio mass spectrometer (IRMS)
interfaced to a NC2500 elemental analyzer. An in-house internal animal standard was run
after every ten samples resulting in an overall standard deviation of 0.06%, and two
additional in-house standards were used to perform two-point normalization (linear
regression) corrections. I report corrected values measured in units of per mil (‰)
relative to the international standard Vienna Pee Dee Belemnite.

55
Stable-nitrogen isotopes (δ15N) can be reliably used within a limited spatial area
to infer birds’ trophic level from diet items as enriched values indicate higher trophic
levels (reviewed by Kelly 2000). δ15N is commonly used to determine the ratio of
arthropod to fruits and seeds in a bird’s diet (e.g., Contina et al. 2022, Gonzalez et al.
2020). However, δ15N may not be a useful indicator of nonbreeding habitat quality (e.g.,
Gonzalez et al. 2020), and more work needs to be done to understand how it can be
applied in this study system. I sampled δ15N following the same procedure outlined above
and present the ratios in Appendix B, Table B.1, so that this information can be applied in
future work.
Capture, body condition, and moult intensity
I captured 268 individual warblers (73 Black-and-white Warbler, 100 American
Redstart, and 95 Northern Waterthrush) over 47 capture days using a combination of
passive mist-netting and target netting with conspecific playback of songs and calls from
each species. See Chapter 1 for complete details on capture effort. After capture, I banded
all birds with a uniquely numbered United States Geological Survey aluminum band.
Using moult and plumage characteristics from Pyle (2022), I assigned each bird to one of
two age classes: First Prealternate Moult (FPA, hereafter “first-cycle”) or Definitive
Prealternate Moult (DPA, hereafter “definitive-cycle”) following the WRP classification
system (Wolfe et al. 2010) most recently modified by Pyle et al. (2022). I determined sex
for Black-and-white Warblers and American Redstarts based on sex-based differences in
plumage characteristics (Pyle 2022).
I measured tarsus length (to the nearest mm) and body mass (to the nearest 0.1 g)
each time I captured or recaptured a bird and calculated the ratio of body mass divided by
tarsus length to estimate the bird’s body condition while accounting for structural body
size. One of three skilled observers (BCD, IAC, and SAT) took all measurements. At the
start of the study period, we repeated tarsus length measurements on the same birds to
standardize our measurements and reduce variability associated with the observer. If I
captured a bird more than once, I took the average of all tarsus measurements to use in its
body condition calculations. I was unable to use a scaled mass index following Peig and

56
Green (2009) as my body condition metric because tarsus length and mass were not
correlated in my species’ samples.
Each time I captured a bird I searched for the presence of active prealternate
moult in each of 17 feather patches across the head and body (Chapter 2, Figure 2.1) by
systematically blowing on the feather patches and scanning for pinfeathers (MettkeHoffman et al. 2010; Wright et al. 2018). I assigned a moult intensity score (adapted from
Greenwood et al. 1983 and Wright et al. 2018) to each feather patch between 0 and 4
which indicates the percentage by quarters of feathers in active moult: 0 = 0%, 1 = 1–
25%, 2 = 26–50%, 3 = 51–75%, and 4 = 76–100%. For each bird, I then summed all
patch scores and divided the sum by the total number of patches (17) to calculate a
Combined Contour Moult Index (CCMI; adapted from Greenwood et al. 1983 and Wright
et al. 2018). Moult scores were collected by one of two skilled observers (SAT and IAC)
and repeated across observers at the start of the season to reduce variability associated
with the observer. See Chapter 2 for complete details on quantifying moult intensity. I
recaptured 47 birds at least once (16 Black-and-white Warbler, 16 American Redstart, and
15 Northern Waterthrush) and repeated their body condition and moult intensity
measurements at each recapture, resulting in a total of 408 moult scores (Chapter 2).
Tracking migration departure dates
In March and April 2023, I fitted 20 Northern Waterthrushes, 28 Black-and-white
Warblers, and 59 American Redstarts with digitally coded nanotag transmitters (Lotek
Wireless, Newmarket, ON) using a leg-loop harness (Rappole and Tipton 1991) of 0.7mm nylon thread. Transmitters and harnesses combined weighed no more than 4% of a
bird’s body mass: 0.28 g NTQB2-1 tags on American Redstarts and 0.28 g NTQB2-1 or
0.32 g NTQB2-2 tags on Northern Waterthrushes and Black-and-white Warblers. An
array of five automated radio towers passively recorded the dates that birds departed Font
Hill Nature Preserve for pre-breeding migration. Each tower comprised a SensorGnome
receiver (https://www.sensorgnome.org) with four 3-element directional Yagi antennas
(Maple Leaf Communications, Everett, ON) mounted 9 m above ground.
After tag detection data was uploaded to The Motus Wildlife Tracking System
network (Taylor et al. 2017), I used R version 4.4.0 (R Core Team 2024), package Motus

57
(Birds Canada 2024), to visualize, clean, and explore the data according to Crewe et al.’s
(2018) guidelines for Motus data analysis. I visually inspected each bird’s signal strength
(dB) over time prior to its last detection to quantify explicit departure dates: a peak in
signal strength followed by a sharp decline at the bird’s last detection indicated a
migratory departure (Mills et al. 2011; Dossman et al. 2016; Taylor et al. 2017). Of the
total 107 tagged birds, 80% showed a departure signal (n = 16/20 Northern Waterthrush,
22/28 Black-and-white Warbler, and 48/59 American Redstart). I removed one outlier
from the statistical analyses that was three standard deviations (8.47 days) greater than
the mean for the species (May 1): a definitive-cycle male Black-and-white Warbler with a
departure date of May 27, late for the species and site when compared with general
departure timing windows (Fink et al. 2023) and previous data from the site (unpublished
Motus dataset 2022).
Statistical analyses
My goal was to evaluate, through statistical modelling, if nonbreeding habitat and
body condition predicted prealternate moult intensity and, if prealternate moult intensity
predicted pre-breeding migration departure dates. Because I could not use habitat type at
capture as an explanatory variable in my analyses for the Black-and-white Warbler and
Northern Waterthrush, I followed a three-step process for those two species. The first step
was to create a model set for each species to test whether δ13C predicted body condition
(Figure 3.1), a relationship which would indicate if the relative quality of mangrove and
scrub habitats differ for the species. Then, I evaluated predictors of prealternate moult
and departure dates. I followed a two-step process for the American Redstart, simply
using habitat type at capture as an explanatory variable in my models predicting
prealternate moult intensity and departure dates (Figure 3.1).
Exploratory analyses revealed that CCMI (see Chapter 2) and body condition (due
to migratory fattening) tended to increase over time in the study population for all three
species. Thus, to account for these trends while excluding the use of correlated
explanatory variables, I calculated, for each species, the residuals of CCMI and body
condition over time and used the residuals as the dependent or independent variables for
body condition and CCMI in my statistical analyses. Because the relationship between

58
CCMI and Julian date fit a quadratic relationship in the American Redstart (Chapter 2), I
calculated the residuals using the regression equation CCMI ~ Julian date + Julian date2
to account for both linear and quadratic components of the relationship. For the other
species’ CCMI and all species body condition, I used the linear regression equations
CCMI ~ Julian date and body condition ~ Julian date. I performed all analyses herein
using R 4.4.0 (R Core Team 2024).

Figure 3.1. Analysis framework and summary of results from evaluating predictors of
residual prealternate moult and pre-breeding migration departure dates in a) Black-andwhite Warbler, b) Northern Waterthrush, and c) American Redstart. I followed a three-

59
step process for the Black-and-white Warbler and Northern Waterthrush, adding the first
step of testing if δ13C, a proxy for habitat quality, predicted residual body condition. I
followed a two-step process for the American Redstart, using habitat type at capture as an
explanatory variable for habitat quality. Solid green arrows indicate a strong effect (95%
confidence intervals did not overlap zero), solid black arrows indicate a moderate effect
(90% confidence intervals did not overlap zero), and dashed arrows indicate a variable
with no effect, that was either not retained during model selection or had 85% confidence
intervals that overlapped zero. Stars indicate interaction terms with a moderate or strong
effect; only statistically meaningful interaction terms are depicted here.
Effect of δ13C on body condition
To examine the influence of δ13C on residual body condition in the Black-andwhite Warbler and Northern Waterthrush, I created linear models (function lm) and used
age as a covariate in both species. I included sex as a second covariate in the Black-andwhite Warbler because I could determine sex in the field. When I tested interaction terms
associated with δ13C ratios for either species, very high variance inflation factors (>
1,000) indicated multicollinearity and model unreliability. I found patterns of association
between δ13C and the covariates age or sex, as well as lower variation in δ13C ratios for
some age/sex classes relative to others (see Appendix B, Figure B.1 for boxplots).
Further, I had minimal representation for one of four age-sex classes for the Black-andwhite Warbler (n = 1 first-cycle female). For these reasons, I ran final, simplified models
including only the δ13C ratio with age (Northern Waterthrush) or age, sex, plus age*sex
(Black-and-white Warbler) as explanatory variables. To evaluate model reliability for
these models and all described herein, I confirmed that the variance inflation factor for
each explanatory variable was < 7.
Next, I used the dredge function (R package MuMIn; Bartón 2024) to create the
model set including a null model and ranked each model using Akaike’s information
criterion corrected for small sample sizes (AICc). I considered models with ΔAICc < 1
competitive. This decision was based on my models for the American Redstart, for which
a higher ΔAIC resulted in an excess of competitive models with highly variable, and
therefore less reliable, β estimates for my continuous explanatory variables. I used ΔAICc
< 1 in all analyses for consistency. After averaging the models with ΔAICc < 1, I
considered variables to have no effect if their model-averaged 85% confidence intervals
overlapped zero. I categorized the magnitude of effects as strong, moderate, and weak, if

60
their 95%, 90%, and 85% confidence intervals did not overlap zero, respectively
(following Buler et al. 2007 and Wright et al. 2020). Here, I report the estimates (β) and
corresponding confidence intervals only for the explanatory variables from top models
with at least a weak effect (see Appendix B, Table B.2 for the explanatory variables with
no effect). I followed the same process for all model selection described herein.
Predicting prealternate moult intensity
To examine the influence of residual body condition on residual moult intensity I
created global linear mixed-effect models (function lmer, R package lme4; Bates et al.
2015) with individual as a random effect to account for recaptures in the Black-and-white
Warbler and Northern Waterthrush. Again, I added age for both species and sex for the
Black-and-white Warbler. I included all possible interaction terms, including those with
residual body condition, in each species’ global model. Because δ13C was correlated with
body condition in the Black-and-white Warbler (Pearson’s product-moment correlation:
r(24) = 0.65, p < 0.001) and I found modelling δ13C with additional continuous
explanatory variables resulted in very high variance inflation factors (> 1,000) in both
species, I created two additional linear models to separately examine the effect of δ13C on
residual moult intensity. I included age as a covariate for the Northern Waterthrush and
age, sex, plus their interaction term for the Black-and-white Warbler.
For the American Redstart, I did not have sufficient recapture records for all
combinations of categorical covariates, making model fitting with a random effect
unreliable. To handle the individuals with repeated moult intensity measurements I ran a
linear model three ways and compared the results: once including only the first capture
record, once including only the second capture record, and finally including all capture
records without accounting for the effect of individual. The results of model selection
were identical, so here I report the models including only the first capture records (see
Appendix B, Table B.3 for the results of all approaches). I included habitat type at
capture, age, sex, residual body condition, and all interaction terms as predictors of
residual prealternate moult intensity.

61
Predicting departure date
I next created global linear models to examine the influence of residual body
condition, residual moult intensity, age, sex, and their interaction terms on departure dates
for each species. Again, I excluded sex from the Northern Waterthrush model and
included habitat type for the American Redstart model. Due to a small sample size, I
could not include all interaction terms in the Northern Waterthrush model. Because I did
not detect a difference in moult intensity between age classes for this species (Chapter 2),
I expected age*moult intensity would be less biologically important than residual body
condition’s interactions with age and moult intensity and excluded it from the model. For
birds that I captured more than once, I used their residual body condition and residual
moult intensity score from the last capture prior to departure. Again, I created additional
models for the Black-and-white Warbler and Northern Waterthrush to separately examine
the influence of δ13C on departure date. I included age as a covariate for the Northern
Waterthrush and age, sex, plus their interaction term for the Black-and-white Warbler,
excluding interactions with δ13C.
To test for differences in departure dates between age classes in the Black-andwhite Warbler and Northern Waterthrush, and between male and female American
Redstarts within each age class, I used Independent Samples t-tests and evaluated the
significance of p-values at α = 0.05.
RESULTS
Effect of δ13C on body condition
A model that contained only δ13C as a predictor was the only model within ΔAICc
<1 that explained body condition in the Black-and-white Warbler (n = 26, Table 3.1).
δ13C had a strong effect on residual body condition; enriched δ13C ratios, indicating drier
habitat, were associated with better condition (βδ13C = 0.03, model-averaged 95%CI =
0.003, 0.048). δ13C did not predict residual body condition in the Northern Waterthrush;
the null model was the top model (all other ΔAICc >1; Table 3.1, n = 14). The Northern
Waterthrushes I sampled had a wider range of δ13C ratios (-27.54 to -22.87‰) when
compared with Black-and-white Warblers (-24.60 to -22.63‰) with a lower minimum

62
value (more negative), indicating that some Northern Waterthrushes occupied wetter (i.e.,
more flooded) habitat than Black-and-white Warblers (Appendix B, Figure B.1).

Table 3.1. Summary of the top ranked models (ΔAICc < 1) that predicted residual body
condition in the Black-and-white Warbler (n = 26) and Northern Waterthrush (n = 14).
Species

AICc

Black-and-white Warbler

Top ranked models
(ΔAICc < 1)
δ13C

-106.9

ΔAICc AICc
weight
0.00
0.46

Northern Waterthrush

NULL MODEL

-30.0

0.00

0.61

Predicting prealternate moult intensity
After model selection, a single model (ΔAICc < 1) that included age, residual
body condition, and their interaction term interaction best explained variation in residual
moult intensity, in the Black-and-white Warbler (n = 99, Table 3.2). Residual moult
intensity was greater in first-cycle Black-and-white Warblers (βage = 0.55, 95% CI = 0.32,
0.77). The interaction between age and residual body condition indicated that when birds
were in better condition, first-cycle birds showed a sharper increase in residual moult
intensity than definitive-cycle birds (βage* residual body condition = 6.63, 95% CI = 1.53, 11.78;
Figure 3.2). When I separately examined the effect of δ13C, I found no effect on residual
moult intensity (n = 26, Table 3.3); the single top model included age alone. Age had a
strong effect on residual moult intensity, with greater residual moult intensity predicted
for first-cycle birds (βage = 0.85, 95% CI = 0.30, 1.41).
Neither age nor residual body condition influenced residual moult intensity in the
Northern Waterthrush; the top model was the null model (n = 106, Table 3.2). When I
examined the effect of δ13C residual prealternate moult intensity, I found no effect (n =
14, Table 3.3). Model selection resulted in only the null models with ΔAICc < 1.
Four models within ΔAICc < 1 explained residual moult intensity in the American
Redstart (n = 111, Table 3.2). Age and its interaction with body condition had strong
effects on residual moult intensity (Figure 3.3). The interaction between age and body
condition indicated that in first-cycle birds, moult intensity decreased with increasing
condition, but in definitive-cycle birds, moult intensity varied little with changes in body

63
condition (βage = 0.40, 95% model-averaged CI = 0.16, 0.64; βage*residual body condition = -8.73,
95% model-averaged CI = -17.22, -0.23). Sex had a moderate effect on residual moult
intensity, where moult intensity was higher in females than males (βsex = -0.15, 90%
model-averaged CI = -0.28, -0.02). The moderate effect of the interaction between age
and habitat type indicated that within the first-cycle age class, moult intensity was greater
in scrub than in mangrove habitat (βage*habitat type = -0.36, 90% model-averaged CI = -0.66,
-0.05; Figure 3.4).

Table 3.2. Summary of the top ranked models (ΔAICc < 1) that predicted residual
prealternate moult intensity in the Black-and-white Warbler (n = 99), Northern
Waterthrush (n = 106), and American Redstart (n = 111).
Species

Top ranked models (ΔAICc < 1)

AICc

ΔAICc AICc
weight
129.70 0.00
0.68

Black-andwhite Warbler

1. age + residual body condition +
age*residual body condition

Northern
Waterthrush

1. NULL MODEL

-1.00

0.00

0.69

American
Redstart

1. age + sex

111.9

0.00

0.08

2. age + residual body condition +
habitat type + sex + age*residual
body condition + age*habitat type

112.0

0.01

0.08

3. age + residual body condition +
habitat type + age*residual body
condition + age*habitat type

112.6

0.65

0.06

4. age + habitat type + sex +
age*habitat type

112.6

0.68

0.06

64

Residual moult intensity (CCMI ~ Julian date)

3

2

1

0

−1
−0.1

0.0

0.1

0.2

Residual body condition (mass/tarsus ~ Julian date)
Age

Definitive−cycle

First−cycle

Figure 3.2. Predicted relationship between residual moult intensity and residual body
condition by age class in the Black-and-white Warbler (n = 99). The residuals in this
model were derived from the linear regression equations body condition (mass/tarsus) ~
Julian date and Combined Moult Contour Index (CCMI) ~ Julian date. Shaded areas
represent 95% confidence intervals.
Table 3.3. Summary of the top ranked models (ΔAICc < 1) that predicted residual
prealternate moult intensity in the Black-and-white Warbler (n = 26) and Northern
Waterthrush (n = 14), using δ13C as an explanatory variable.
Species

AICc

ΔAICc

Black-and-white Warbler

Top ranked models
(ΔAICc < 1)
1. age

50.3

0.00

AICc
weight
0.57

Northern Waterthrush

1. NULL MODEL

20.6

0.00

0.68

Residual moult intensity (CCMI ~ Julian date + Julian date²)

65

2

1

0

−1

−2

−3
−0.1

0.0

0.1

0.2

Residual body condition (mass/tarsus ~ Julian date)
Age

Definitive−cycle

First−cycle

Figure 3.3. Predicted relationship between residual moult intensity and residual body
condition by age class in the American Redstart (n = 111). The residuals in this model
were derived from the linear regression equations body condition (mass/tarsus) ~ Julian
date and Combined Moult Contour Index (CCMI) ~ Julian date + Julian date2. Shaded
areas represent 95% confidence intervals.

Residual moult intensity (CCMI ~ Julian date + Julian date²)

66

1.0

0.5

0.0

−0.5
Definitive−cycle

First−cycle

Age
Habitat Type

Scrub

Mangrove

Figure 3.4. Boxplots of residual moult intensity grouped by age class and habitat type in
the American Redstart (n = 111). Residual moult intensity was derived from the linear
regression equation Combined Moult Contour Index (CCMI) ~ Julian date + Julian date2.
Predicting departure dates
Pre-breeding migration departure dates ranged from April 13 to May 27, with
Northern Waterthrushes generally departing earliest (mean ± SD = Apr 26 ± 4.23 days,
range = Apr 18 - May 2, n = 16), followed by Black-and-white Warblers (mean ± SD =
Apr 30 ± 6.43 days, range = April 21 – May 14, n = 21) and then American Redstarts
(mean ± SD = May 3 ± 8.52 days, range = April 13 – May 27), though American
Redstarts showed the widest range in departure dates.
After model selection, a single model (ΔAICc < 1) that included only sex best
explained variation in departure dates in the Black-and-white Warbler (n = 21, Table 3.4).
Males departed earlier than females (βsex = -5.51, 95%CI = -10.93, -0.09). δ13C did not
predict departure dates; model selection including δ13C resulted in a top model with sex

67
as the only predictor, followed by the null model (ΔAICc < 1) (n = 17, Table 3.5).
Consistent with the analysis excluding δ13C, males departed earlier than females, though
the relationship was weak (βsex = -2.60, 85% model-averaged CI = -9.62, -0.60). When I
examined age-based differences alone, mean departure dates (± SD) were similar for
first- (May 1 ± 5.74 days) and definitive-cycle birds (April 30 ± 6.94 days) (Independent
Samples t-test: t(14) = -0.23, p = 0.82).
I found no association between departure dates and age, residual body condition,
residual moult intensity, and the interaction terms age*residual body condition and
residual body condition*residual moult intensity for the Northern Waterthrush; the top
model (ΔAICc < 1) was the null model (n = 16, Table 3.4). However, I found that δ13C
was associated with departure date. Model selection resulted in a single model within
ΔAICc < 1 that included only δ13C (n = 14, Table 3.5). δ13C had a strong effect on
departure date, where birds with depleted δ13C ratios, consistent with occupying wetter
habitat, departed earlier than birds with enriched ratios (βδ13C = 2.38, 95%CI = 0.45, 4.30;
Figure 3.5). When I compared departure dates between age classes, mean departure dates
(± SD) were similar for first- (Apr 25 ± 5.80 days) and definitive-cycle birds (Apr 30 ±
9.65 days) (Independent Samples t-test: t(8) = 0.68, p = 0.52).
Three models within ΔAICc <1 best explained the variation in American Redstart
departure dates and included combinations of the variables age, residual moult intensity,
residual body condition, sex, and its interaction with residual moult intensity (n = 48,
Table 3.4). All variables included in the top models strongly affected departure dates,
except for residual body condition, which had no effect. Definitive-cycle birds departed
earlier than first-cycle birds (βage = 5.94, model-averaged 95%CI = 0.49, 11.39; Figure
3.6). Males tended to depart earlier than females (βsex = -5.33, model-averaged 95%CI = 10.20, -0.48), but the influence of sex was modified by residual moult intensity
(βsex*residual moult intensity = 8.88, model-averaged 95%CI = 1.75, 22.63; βresidual moult intensity = 6.98, model-averaged 95%CI = -18.52, -0.64; Figure 3.7). Residual moult intensity was
negatively associated with departure dates in females but positively associated in males.
When I examined sex-based differences within each age class, males departed
earlier than females on average in definitive-cycle (Independent Samples t-test: t(16) =
2.33, p = 0.03) but not first-cycle birds (Independent Samples t-test: t(9) = 0.56, p = 0.59).

68

Table 3.4. Summary of the top ranked models (ΔAICc < 1) that predicted pre-breeding
migration departure dates from Jamaica in the Black-and-white Warbler (n = 21),
Northern Waterthrush (n = 16), and American Redstart (n = 48).
Species

Top models (ΔAICc < 1)

Black-and-white
Warbler

1. sex

AICc ΔAICc AICc
weight
139.6 0.00
0.31

Northern
Waterthrush

1. NULL MODEL

95.4

0.00

0.45

American
Redstart

1. age + residual moult intensity +
sex
2. age + residual moult intensity +
residual body condition + sex
3. age + residual body condition +
sex

336.9 0.00

0.05

337.8 0.82

0.04

337.9 0.96

0.03

Table 3.5. Summary of the top ranked models (ΔAICc < 1) that predicted pre-breeding
migration departure dates from Jamaica in the Black-and-white Warbler (n = 17) and
Northern Waterthrush (n = 14), using δ13C as an explanatory variable.
Species
Black-and-white Warbler
Northern Waterthrush

Top ranked models
(ΔAICc < 1)
1. sex
2. NULL MODEL

AICc ΔAICc AICc
weight
115.5 0.00
0.31
115.6 0.07
0.30

1. δ13C

82.4

0.00

0.70

69

Figure 3.5. δ13C ratios (‰) in the Northern Waterthrush (black circles). The line
represents the predicted relationship between pre-breeding departure date from Jamaica
and δ13C ratio (‰) in the Northern Waterthrush (n = 14). Departure dates are in Julian
days; day 105 = April 15, 2023. The shaded area represents the 95% confidence interval.

70

Departue date (Julian days)

140

130

120

110

Definitive−cycle

First−cycle

Age
Sex

Female

Male

Figure 3.6. Boxplots of pre-breeding migration departure dates from Jamaica grouped by
age and sex classes in the American Redstart (n = 48). Departure dates are Julian days;
day 110 = April 20, 2023.

71

140

Departure date (Julian days)

130

120

110

100
−0.5

0.0

0.5

1.0

1.5

Residual moult intensity (CCMI ~ Julian date + Julian date²)
Sex

Male

Female

Figure 3.7. Predicted relationship between pre-breeding departure date from Jamaica and
residual moult intensity by sex in the American Redstart (n = 48). The residuals in this
model were derived from the linear regression equation Combined Moult Contour Index
(CCMI) ~ Julian date + Julian date2. Departure dates are in Julian days; day 100 = April
10, 2023. Shaded areas represent 95% confidence intervals.
DISCUSSION
The conditions that warblers experience on the stationary nonbreeding grounds
can directly influence the timing of their departure on pre-breeding migration (e.g.,
American Redstart; Marra et al. 1998, Studds and Marra 2005, Reudink et al. 2009a), yet
how the intensity of prealternate moult may modify this seasonal interaction remains
largely unknown. Here, I build on a detailed quantification of prealternate moult intensity
in three Nearctic-Neotropical migratory songbirds (Chapter 2) to demonstrate that, 1)
body condition, mediated by habitat quality, can be associated with moult intensity, and
2) moult intensity can be associated with departure date. These findings provide the first

72
potential evidence of prealternate moult carry-over effects from the nonbreeding period to
pre-breeding migration phenology in a parulid warbler and suggest that the timing and
intensity of prealternate moult may represent an important seasonal interaction in some
species. In addition, the intensity and timing of prealternate moult appears to be flexible
and vary in response to nonbreeding conditions. The influence of nonbreeding conditions
on moult intensity and their subsequent effects on departure date varied across age and
sex classes in addition to species, suggesting that moult and departure phenology are
driven by both endogenous and exogenous factors regardless of nonbreeding conditions.
Age was associated with moult intensity in the Black-and-white Warbler and
American Redstart, modified by an interaction with body condition. For the Black-andwhite Warbler, moult was generally more intense in first-cycle birds and increased with
condition. Moult was more intense in some first-cycle American Redstarts, but intensity
varied more than in definitive-cycle birds. When compared with Black-and-white
Warblers, first-cycle American Redstarts showed the opposite pattern—a negative
relationship between moult intensity and body condition. Moult intensity varied less with
changes in condition in definitive-cycle birds of both species. First-cycle Black-andwhite Warblers and American Redstarts in the study population may moult more
intensely than definitive-cycle birds because they can gain a greater benefit from
replacing relatively poorer quality contour feathers (as argued in Chapter 1). Birds tend to
have poorer quality feathers in their first cycle (Pyle 2022, e.g., Szép et al. 2019) and if
territorial, secure poorer quality territories on the nonbreeding grounds than their
definitive-cycle counterparts (e.g., Figuerola et al. 2001, Marra and Holmes 2001) which
might lead to increased feather wear throughout the nonbreeding period (Reudink et al.
2009b). In birds with a high incentive to moult, such as first-cycle Black-and-white
Warblers, being in better condition might facilitate a more intense moult. In contrast,
first-cycle American Redstarts in relatively poor condition but showing high intensity
moult might indicate the individuals with the poorest quality feathers, hence requiring the
most replacement. The differences in the relationships between moult and body condition
in these two species demonstrate the importance of future, detailed study on the
physiology of prealternate moult and its role within the annual cycle.

73
In this system, first-cycle American Redstarts tend to occupy nonbreeding
territories that are drier and have fewer food resources, resulting in relatively poor body
condition (Marra and Holmes 2001). When compared with definitive-cycle American
Redstarts, first-cycle territories also tend to be more open (unpublished long-term site
dataset 1987—2024), which may result in increased feather wear from solar radiation
(Bergman 1982, Froehlich et al. 2005, Reudink et al. 2009b, Terrill et al. 2020, Guallar et
al. 2021). In this study, not only did many first-cycle American Redstarts tend to moult
more intensely than definitive-cycle American Redstarts, but within first-cycle birds,
those occupying poorer quality scrub habitat moulted more intensely than those
occupying higher quality mangrove. Ames (2021) observed the same pattern in the
Prothonotary Warbler (Protonotaria critea), where prealternate moult intensity was
higher in first-cycle birds and for birds occupying dry nonbreeding habitat in Panama.
The difference in moult intensity among habitats was not as marked in definitive-cycle
American Redstarts, possibly because the quality of scrub habitat varies in the study plots
and anecdotally, older birds tend to secure higher quality, more closed-canopy scrub
territories (unpublished long-term site dataset 1987—2024). However, American
Redstarts in this system also have sex-biased habitat occupancy, with females more likely
to occupy scrub habitat (Parrish and Sherry 1994, Marra and Holberton 1998). Indeed, I
also found a moderate relationship between sex and moult intensity for American
Redstarts, where females moulted more intensely than males. The pattern of higher
intensity moult associated with more open habitat was found in comparative analyses of
the evolution of prealternate moult at the species level (Svensson and Hedenström 1999,
Cuervo et al. 2022). Migratory species that occupied open nonbreeding habitats were
more likely to evolve a prealternate moult while their congeneric relatives that occupied
more closed nonbreeding habitats did not. Further studies should directly measure habitat
openness and contour feather quality during the nonbreeding period preceding moult (i.e.,
October through January) to test this hypothesis for the American Redstart.
As expected, I found inter- and intraspecific variation in the relationship between
body condition and moult intensity. In a study of prebasic moult intensity in the House
Finch, the influence of body condition on moult intensity varied across the sexes, urbanrural habitat type, and at different stages of moult (i.e., beginning, peak, and end; Hutton

74
et al. 2021). Hutton et al. (2021) concluded that body condition limited feather synthesis
during the peak and end stages of moult but did not appear to be an important predictor
during the beginning of moult. Because moult intensity often peaks during the middle of
the moult period (e.g., Wright et al. 2018, Hutton et al. 2021, Guallar 2024), a
relationship I detected for American Redstarts (Chapter 2), I expect body condition to be
important mid-moult when energetic demands are high. Unlike the non-migratory House
Finch, moult in my study population appears to overlap with migratory fattening (Chapter
2). Therefore, body condition must also be important during the end of moult; though
moult intensity is low, energetic demands for migratory fattening are high and birds may
be faced with trade-offs for resource allocation (e.g., Bonier et al. 2007, Rubolini et al.
2022, Witkowska et al. 2024). Future study focused on parsing the effect of body
condition on moult intensity during different stages of the prealternate moult in this
system could improve our understanding of the differing relationships observed in the
Black-and-white Warbler and American Redstart, and potentially identify an effect in the
Northern Waterthrush. Further, examining habitat use and its relationship to condition
during different stages of moult could elucidate why δ13C predicted condition in the
Black-and-white Warbler but was not associated with moult intensity.
Moult intensity was strongly associated with departure date in American Redstarts
only and modified by an interaction with sex. However, the sexes showed relationships in
opposite directions. Because moult intensity followed a parabola in American Redstarts
over time (Chapter 2), lower moult intensity prior to departure likely indicates that moult
is nearer to completion. Males with higher moult intensity, likely indicating that moult
was completed later, departed later. This result suggests that the timing of moult may
impose a constraint on male American Redstarts but not females. Because males are
under strong selection pressure to arrive earlier to the breeding grounds to secure a highquality territory and mate(s), a delayed departure has the potential to incur greater
reproductive fitness costs (e.g., Lozano et al. 1996, Reudink et al. 2009a, Tonra et al.
2011, Rockwell et al. 2012). While some birds compensate for later departures by
increasing their migration speed (e.g., González et al. 2020), previous research shows that
when male American Redstarts depart later from Jamaica, they tend to arrive later to the
breeding grounds than birds with earlier departure dates (Reudink et al. 2009a).

75
In this study system, American Redstarts in better condition—typically a result of
territory quality—tend to depart the nonbreeding grounds earlier (Marra et al. 1998,
Studds and Marra 2005, Reudink et al. 2009a). If better condition, resulting from high
quality habitat, can also facilitate a more intense moult (e.g., Hutton et al. 2021), birds in
better condition should complete moult earlier via a higher moult intensity (de la Hera et
al. 2011, Rohwer and Rohwer 2013), allowing them to depart earlier. However, I did not
detect a direct relationship between condition and departure date in American Redstarts
or the other species. Contrary to my prediction, enriched δ13C ratios indicating drier
habitat were associated with better condition in the Black-and-white Warbler. At Font Hill
Nature Reserve, drier habitat was also better quality for the Swainson’s Warbler
(Lymnothlypis swainsonii) because it had more abundant arthropod biomass in the thicker
leaf litter found in scrub forest (Brunner et al. 2022). It’s possible that xeric scrub habitat
can be higher quality for the foliage-gleaning Black-and-white Warbler if their foraging
substrate (i.e., tree bark) hosts more abundant resources than trees in mangrove.
However, although condition was associated with moult intensity, neither condition nor
δ13C were linked to departure date.
Many migratory bird species exhibit age- and sex-based differences in migration
timing (reviewed in Newton 2011); males tend to migrate earlier than females (protandry)
and older birds tend to migrate earlier than young birds (e.g., passerines; Stewart et al.
2002). In my study population, I would expect earlier departures for males and definitivecycle birds. I detected two of those expected patterns: male Black-and-white Warblers
departed earlier than females, and American Redstarts in their definitive-cycle departed
earlier than those in their first-cycle. Given that such patterns of differential migration
among age or sex classes can vary interannually in relation to environmental conditions
on the nonbreeding grounds (e.g., Prairie Warbler; Akresh et al. 2019), it is not surprising
that I did not detect all three patterns among all three study species. Further, nonbreeding
populations may breed across a broad latitudinal range, and I was unable to account for
the expected differences in intraspecific migration distance (unpublished Motus dataset
2022, Dossman et al. 2023a). Within a nonbreeding population, differences in migration
distance can be associated with intraspecific migration timing (e.g., Bar-tailed Godwit
(Limosa lapponica): Conklin et al. 2010; American Redstart: Dossman et al. 2023b).

76
Evidence suggests that aspects of migratory behaviour, such as migration timing and
direction, are heritable, at least in the Eurasian Blackcap (Sylvia atricapilla; Berthold and
Helbig 1992, Bearhop et al. 2005; Delmore et al. 2020, Delmore et al. 2023). For prebreeding migration in particular, the timing of departure from the nonbreeding grounds
and arrival to the breeding grounds might be under strong endogenous control (Stanley et
al. 2012). Stanley et al. 2012 found that the timing of pre-breeding migration in the Wood
Thrush (Hylocichla mustelina) was highly repeatable over two or three years of tracking.
Males and older birds consistently departed the Neotropical nonbreeding grounds earlier,
and in addition, individual departure dates were highly consistent across years. Given that
my sample sizes for the departure analysis were relatively small and I could not account
for differences in migration distance, innate migration schedules based on the location of
breeding grounds likely explain some of the variation in departure dates and confound
my ability to fully understand the effect of moult intensity. In this system, a future study
that measures body condition, moult intensity, and departure date in the same individuals
across multiple years would greatly aid in improving our initial understanding of the
potential for prealternate moult carry-over effects in migratory parulid warblers.
In conclusion, my results provide the first potential evidence of a carry-over effect
from the prealternate moult to pre-breeding migration phenology in a parulid warbler, the
American Redstart. To my knowledge, this study represents only the third of its kind to
specifically consider the role of parulid prealternate moult in carry-over effects (see
Boone et al. 2010 and Jones et al. 2014), and the first to examine potential prealternate
moult carry-over effects in these species. Studies such as this are critical for building our
understanding of moult and its role in seasonal interactions among related species, age
and sex classes, and can be used in the future to make much needed comparisons across
years and nonbreeding populations. Ultimately, this study highlights the importance of
considering the role of moult when assessing carry-over effects across the annual cycle.

77
LITERATURE CITED
Akresh, M. E., King, D. I., & Marra, P. P. (2019). Examining carry‐over effects of winter
habitat on breeding phenology and reproductive success in prairie warblers
Setophaga discolor. Journal of Avian Biology, 50(4), e02025.
doi.org/10.1111/jav.02025
Ames, E. M. (2021). Linking events across the annual cycle, in a Neotropical migratory
songbird of conservation concern, the Prothonotary Warbler (Protonotaria
citrea) [Ph.D. thesis]. The Ohio State University.
Barta, Z., McNamara, J. M., Houston, A. I., Weber, T. P., Hedenström, A., & Feró, O.
(2008). Optimal moult strategies in migratory birds. Philosophical Transactions
of the Royal Society B: Biological Sciences, 363(1490), 211–229.
doi.org/10.1098/rstb.2007.2136
Bartón, K. (2024). Package “MuMIm.”http://cran.stat.unipd.it/web/packages/MuMIn
/MuMIn.pdfhttp://cran.stat.unipd.it/web/packages/MuMIn/MuMIn.pdf
Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015). Fitting linear mixed-effects
models using lme4. Journal of Statistical Software, 67(1).
doi.org/10.18637/jss.v067.i01
Bearhop, S., Fiedler, W., Furness, R. W., Votier, S. C., Waldron, S., Newton, J., Bowen,
G. J., Berthold, P., & Farnsworth, K. (2005). Assortative mating as a mechanism
for rapid evolution of a migratory divide. Science, 310(5747), 502–504.
doi.org/10.1126/science.1115661
Bergman, G. (1982). Why are the wings of Larus f. Fuscus so dark?. Ornis Fennica, 59,
77-83.
Berthold, P., & Helbig, A. J. (1992). The genetics of bird migration: Stimulus, timing, and
direction. Ibis, 134(s1), 35–40. doi.org/10.1111/j.1474-919X.1992.tb04731.x
Birds Canada. (2024). motus: Fetch and use data from the Motus Wildlife Tracking
System. https://motusWTS.github.io/motus/
Bonier, F., Martin, P. R., Jensen, J. P., Butler, L. K., Ramenofsky, M., & Wingfield, J. C.
(2007). Pre-migratory life history stages of juvenile Arctic birds: Costs,
constraints, and trade-offs. Ecology, 88(11), 2729–2735. doi.org/10.1890/070696.1
Boone, A. T., Rodewald, P. G., & DeGroote, L. W. (2010). Neotropical winter habitat of
the magnolia warbler: Effects on molt, energetic condition, migration timing, and
hematozoan infection during spring migration. The Condor, 112(1), 115–122.
doi.org/10.1525/cond.2010.090098

78
Borowske, A., Gjerdrum, C., & Elphick, C. (2017). Timing of migration and prebasic
molt in tidal marsh sparrows with different breeding strategies: Comparisons
among sexes and species. The Auk, 134(1), 51–64. doi.org/10.1642/AUK-16116.1
Bozó, L., Anisimov, Y., & Csörgő, T. (2022). Moult, sex and food are the most important
factors regulated the timing of migration of north Asian Passerines. Ornithology
Research, 30(4), 262–270. doi.org/10.1007/s43388-022-00108-y
Brunner, A. R., Marra, P. P., & Tonra, C. M. (2022). Vulnerable Neotropical migratory
songbird demonstrates flexibility in space use in response to rainfall change.
Ornithology, 139(2), ukac005. doi.org/10.1093/ornithology/ukac005
Buehler, D. M., & Piersma, T. (2008). Travelling on a budget: Predictions and ecological
evidence for bottlenecks in the annual cycle of long-distance migrants.
Philosophical Transactions of the Royal Society B: Biological Sciences,
363(1490), 247–266. doi.org/10.1098/rstb.2007.2138
Buler, J. J., Moore, F. R., & Woltmann, S. (2007). A multi-scale examination of stopover
habitat use by birds. Ecology, 88(7), 1789–1802. doi.org/10.1890/06-1871.1
Carlisle, J. D., Kaltenecker, G. S., & Swanson, D. L. (2005). Molt strategies and age
differences in migration timing among autumn landbird migrants in southwestern
Idaho. The Auk, 122(4), 1070–1085.
Clements, S. J., Loghry, J. P., Ballard, B. M., & Weegman, M. D. (2022). Carry‐over
effects of weather and decision‐making on nest success of a migratory shorebird.
Ecology and Evolution, 12(12), e9581. doi.org/10.1002/ece3.9581
Conklin, J. R., & Battley, P. F. (2012). Carry‐over effects and compensation: Late arrival
on non‐breeding grounds affects wing moult but not plumage or schedules of
departing bar‐tailed godwits Limosa lapponica baueri. Journal of Avian Biology,
43(3), 252–263. doi.org/10.1111/j.1600-048X.2012.05606.x
Conklin, J. R., Battley, P. F., Potter, M. A., & Fox, J. W. (2010). Breeding latitude drives
individual schedules in a trans-hemispheric migrant bird. Nature
Communications, 1(1), 67. doi.org/10.1038/ncomms1072
Contina, A., Pierce, A. K., Yanco, S. W., Bridge, E. S., Kelly, J. F., & Wunder, M. B.
(2022). Application of stable isotopes to study movement ecology and diet
variation in a migratory songbird. Journal of Field Ornithology, 93(4), 6.
doi.org/10.5751/JFO-00179-930406
Cooper, N. W., Sherry, T. W., & Marra, P. P. (2015). Experimental reduction of winter
food decreases body condition and delays migration in a long-distance migratory
bird. Ecology, 96(7), 1933–1942. doi.org/10.1890/14-1365.1

79
Cooper, N. W., Thomas, M. A., & Marra, P. P. (2021). Vertical sexual habitat segregation
in a wintering migratory songbird. The Auk, 138, 1–11.
doi.org/10.1093/ornithology/ukaa080
Crewe, T. L., Crysler, Z., & Taylor, P. (2018). Motus R Book. Online.
Cuervo, J. J., Morales, J., Soler, J. J., & Moreno, J. (2022). Sexual selection, feather wear,
and time constraints on the pre‐basic molt explain the acquisition of the pre‐
alternate molt in European passerines. Ecology and Evolution, 12(9), e9260.
doi.org/10.1002/ece3.9260
Danner, R. M., Greenberg, R. S., Danner, J. E., & Walters, J. R. (2015). Winter food
limits timing of pre‐alternate moult in a short‐distance migratory bird. Functional
Ecology, 29(2), 259–267. doi.org/10.1111/1365-2435.12322
de la Hera, I., Schaper, S. V., Díaz, J. A., Pérez-Tris, J., Bensch, S., & Tellería, J. L.
(2011). How much variation in the molt duration of passerines can be explained
by the growth rate of tail feathers? The Auk, 128(2), 321–329.
doi.org/10.1525/auk.2011.10181
Delmore, K. E., Van Doren, B. M., Ullrich, K., Curk, T., Van Der Jeugd, H. P., &
Liedvogel, M. (2023). Structural genomic variation and migratory behavior in a
wild songbird. Evolution Letters, 7(6), 401–412. doi.org/10.1093/evlett/qrad040
Delmore, K., Illera, J. C., Pérez-Tris, J., Segelbacher, G., Lugo Ramos, J. S., Durieux, G.,
Ishigohoka, J., & Liedvogel, M. (2020). The evolutionary history and genomics of
European blackcap migration. eLife, 9, e54462. doi.org/10.7554/eLife.54462
Dossman, B. C., Matthews, S. N., & Rodewald, P. G. (2018). An experimental
examination of the influence of energetic condition on the stopover behavior of a
Nearctic–Neotropical migratory songbird, the American Redstart (Setophaga
ruticilla). The Auk, 135(1), 91–100. doi.org/10.1642/AUK-17-67.1
Dossman, B. C., Mitchell, G. W., Norris, D. R., Taylor, P. D., Guglielmo, C. G.,
Matthews, S. N., & Rodewald, P. G. (2016). The effects of wind and fuel stores on
stopover departure behavior across a migratory barrier. Behavioral Ecology,
27(2), 567–574. doi.org/10.1093/beheco/arv189
Dossman, B. C., Studds, C. E., LaDeau, S. L., Sillett, T. S., & Marra, P. P. (2023a). The
role of tropical rainfall in driving range dynamics for a long-distance migratory
bird. Proceedings of the National Academy of Sciences, 120(52), e2301055120.
doi.org/10.1073/pnas.2301055120
Dossman, B. C., Rodewald, A. D., Studds, C. E., & Marra, P. P. (2023). Migratory birds
with delayed spring departure migrate faster but pay the costs. Ecology, 104(2),
e3938. doi.org/10.1002/ecy.3938

80
Figuerola, J., Jovani, R., & Sol, D. (2001). Age-related habitat segregation by Robins
Erithacus rubecula during the winter. Bird Study, 48(2), 252–255.
doi.org/10.1080/00063650109461225
Fink, D., Auer, T., Johnston, A., Strimas-Mackey, M., Ligocki, S., Robinson, O.,
Hochachka, W., Jaromczyk, L., Crowley, C., Dunham, K., Stillman, A., Davies, I.,
Rodewald, A., Ruiz-Gutierrez, V., & Wood, C. (2023). eBird Status and Trends
[Dataset]. doi.org/10.2173/ebirdst.2022
Froehlich, D. R., Rowher, S., & Stutchbury, B. J. (2005). Spring molt constraints versus
winter territoriality: Is conspicuous winter coloration maladaptive? In R. S.
Greenberg & P. P. Marra (Eds.), Birds of Two Worlds: The Ecology and Evolution
of Migration. The Johns Hopkins University Press.
Gill, J. A., Norris, K., Potts, P. M., Gunnarsson, T. G., Atkinson, P. W., & Sutherland, W.
J. (2001). The buffer effect and large-scale population regulation in migratory
birds. Nature, 412.
González, A. M., Bayly, N. J., & Hobson, K. A. (2020). Earlier and slower or later and
faster: Spring migration pace linked to departure time in a Neotropical migrant
songbird. Journal of Animal Ecology, 89(12), 2840–2851. doi.org/10.1111/13652656.13359
Greenwood, H., Weatherhead, P. J., & Titman, R. D. (1983). A New Age- and SexSpecific Molt Scheme for the Red-Winged Blackbird. The Condor, 85(1), 104–
105. doi.org/10.2307/1367901
Guallar, S. (2024). From partial to complete: Wing- and tail-feather moult sequence and
intensity depend on species, life-cycle stage, and moult completeness in
passerines. Avian Research, 15, 100163. doi.org/10.1016/j.avrs.2024.100163
Guallar, S., Rueda-Hernández, R., & Pyle, P. (2021). Evolution of the preformative molt
in Cardinalidae correlates with transitions from forest to open habitats.
Ornithology, 138(1), ukaa070. doi.org/10.1093/ornithology/ukaa070
Harrison, X. A., Blount, J. D., Inger, R., Norris, D. R., & Bearhop, S. (2011). Carry-over
effects as drivers of fitness differences in animals: Carry-over effects in animal
populations. Journal of Animal Ecology, 80(1), 4–18. doi.org/10.1111/j.13652656.2010.01740.x
Hedenström, A. (2002). Aerodynamics, evolution and ecology of avian flight. Trends in
Ecology & Evolution, 17(9), 415–422. https://doi.org/10.1016/S01695347(02)02568-5
Howell, S. N. G., Corben, C., Pyle, P., & Rogers, D. (2003). The First Basic Problem: A
Review of Molt and Plumage Homologies. The Condor, 105(4), 635–653.

81
Hutton, P., McKenna, J., & McGraw, K. J. (2021). Urban links to molt schedule, body
condition and carotenoid‐based coloration in the house finch Haemorhous
mexicanus. Journal of Avian Biology, 52(5), jav.02761. doi.org/10.1111/jav.02761
Jones, T. B., Drake, A., & Green, D. J. (2014). Individual quality explains association
between plumage colouration, arrival dates and mate acquisition in yellow
warblers (Setophaga petechia). BMC Ecology, 14(1), 13. doi.org/10.1186/14726785-14-13
Kelly, J. F. (2000). Stable isotopes of carbon and nitrogen in the study of avian and
mammalian trophic ecology. Canadian Journal of Zoology, 78(1), 1–27.
Lennox, F. G., & Rowlands, R. J. (1969). Photochemical degradation of keratins.
Photochemistry and Photobiology, 9(4), 359–367. doi.org/10.1111/j.17511097.1969.tb07300.x
Long-term dataset, Font Hill Nature Preserve, 1987-2024. [Unpublished data].
Lourenço, P. M., & Piersma, T. (2015). Migration distance and breeding latitude correlate
with the scheduling of pre-alternate body moult: A comparison among migratory
waders. Journal of Ornithology, 156(3), 657–665. doi.org/10.1007/s10336-0151175-8
Lozano, G. A., Perreault, S., & Lemon, R. E. (1996). Age, arrival date and reproductive
success of male American Redstarts Setophaga ruticilla. Journal of Avian Biology,
27(2), 164. doi.org/10.2307/3677146
Marra, P. P., Cohen, E. B., Loss, S. R., Rutter, J. E., & Tonra, C. M. (2015). A call for full
annual cycle research in animal ecology. Biology Letters, 11(8), 20150552.
doi.org/10.1098/rsbl.2015.0552
Marra, P. P., Hobson, K. A., & Holmes, R. T. (1998). Linking winter and summer events
in a migratory bird by using stable-carbon isotopes. Science, 282(5395), 1884–
1886. doi.org/10.1126/science.282.5395.1884
Marra, P. P., & Holberton, R. L. (1998). Corticosterone levels as indicators of habitat
quality: Effects of habitat segregation in a migratory bird during the non-breeding
season. Oecologia, 116(1–2), 284–292. doi.org/10.1007/s004420050590
Marra, P. P., & Holmes, R. T. (2001). Consequences of dominance-mediated habitat
segregation in American redstarts during the nonbreeding season. The Auk,
118(1), 92–104. doi.org/10.1093/auk/118.1.92
Marshall, J. D., Brooks, J. R., & Lajtha, K. (2007). Sources of Variation in the Stable
Isotopic Composition of Plants. In R. Michener & K. Lajtha (Eds.), Stable

82
Isotopes in Ecology and Environmental Science (1st ed., pp. 22–60). Wiley.
doi.org/10.1002/9780470691854.ch2
Mettke-Hofmann, C., Sinclair, P. H., Hamel, P. B., & Greenberg, R. (2010). Implications
of prebasic and a previously undescribed prealternate molt for aging Rusty
Blackbirds. The Condor, 112(4), 854–861. doi.org/10.1525/cond.2010.100150
Mills, A. M., Thurber, B. G., Mackenzie, S. A., & Taylor, P. D. (2011). Passerines use
nocturnal flights for landscape-scale movements during migration stopover. The
Condor, 113(3), 597–607. doi.org/10.1525/cond.2011.100186
Morton, M. L., & Welton, D. E. (1973). Postnuptial molt and its relation to reproductive
cycle and body weight in Mountain White-Crowned Sparrows (Zonotrichia
leucophrys oriantha). The Condor, 75(2), 184–189. doi.org/10.2307/1365866
Motus dataset, Jamaica Long-term Research—Font Hill Nature Preserve. (2022).
[Unpublished data].
Newton, I. (2011). Migration within the annual cycle: Species, sex and age differences.
Journal of Ornithology, 152(S1), 169–185. doi.org/10.1007/s10336-011-0689-y
Norris, D. R., & Marra, P. P. (2007). Seasonal interactions, habitat quality, and population
dynamics in migratory birds. The Condor, 109(3), 535–547.
Norris, D. R., Marra, P. P., Kyser, T. K., Sherry, T. W., & Ratcliffe, L. M. (2004). Tropical
winter habitat limits reproductive success on the temperate breeding grounds in a
migratory bird. Proceedings of the Royal Society of London. Series B: Biological
Sciences, 271(1534), 59–64. doi.org/10.1098/rspb.2003.2569
Oppel, S., & Powell, A. N. (2010). Carbon isotope turnover in blood as a measure of
arrival time in migratory birds using isotopically distinct environments. Journal of
Ornithology, 151(1), 123–131. doi.org/10.1007/s10336-009-0434-y
Pageau, C., Sonnleitner, J., Tonra, C. M., Shaikh, M., & Reudink, M. W. (2021).
Evolution of winter molting strategies in European and North American migratory
passerines. Ecology and Evolution, 11(19), 13247–13258.
doi.org/10.1002/ece3.8047
Parrish, J. D., & Sherry, T. W. (1994). Sexual habitat segregation by American Redstarts
wintering in Jamaica: Importance of resource seasonality. The Auk, 111(1), 38–49.
doi.org/10.2307/4088503
Paxton, K. L., & Moore, F. R. (2015). Carry-over effects of winter habitat quality on en
route timing and condition of a migratory passerine during spring migration.
Journal of Avian Biology, 46(5), 495–506. doi.org/10.1111/jav.00614

83
Peig, J., & Green, A. J. (2009). New perspectives for estimating body condition from
mass/length data: The scaled mass index as an alternative method. Oikos, 118(12),
1883–1891. doi.org/10.1111/j.1600-0706.2009.17643.x
Piersma, T., & Jukema, J. (1993). Red breasts as honest signals of migratory quality in a
long-distance migrant, the Bar-tailed Godwit. The Condor, 95(1), 163–177.
doi.org/10.2307/1369398
Powell, L. L., Ames, E. M., Wright, J. R., Matthiopoulos, J., & Marra, P. P. (2021).
Interspecific competition between resident and wintering birds: Experimental
evidence and consequences of coexistence. Ecology, 102(2), e03208.
doi.org/10.1002/ecy.3208
Pyle, P. (2022). Identification guide to North American birds: Part I (2nd ed). Slate Creek
Press.
Pyle, P., & Kayhart, R. (2010). Replacement of Primaries during the Prealternate Molt of
a Yellow Warbler. North American Bird Bander, 35(4), 178–181.
R Core Team. (2024). R: A language and environment for statistical computing
[Computer software]. R Foundation for Statistical Computing. https://www.rproject.org/
Rappole, J. H., & Tipton, A. R. (1991). New harness design for attachment of radio
transmitters to small passerines. Journal of Field Ornithology, 62(3), 335–337.
Reudink, M. W., Marra, P. P., Kyser, T. K., Boag, P. T., Langin, K. M., & Ratcliffe, L. M.
(2009a). Non-breeding season events influence sexual selection in a long-distance
migratory bird. Proceedings of the Royal Society B: Biological Sciences,
276(1662), 1619–1626. doi.org/10.1098/rspb.2008.1452
Reudink, M. W., Studds, C. E., Marra, P. P., Kyser, T. K., & Ratcliffe, L., M. (2009b).
Plumage brightness predicts non-breeding season territory quality in a longdistance migratory songbird, the American redstart Setophaga ruticilla. Journal of
Avian Biology, 40(1), 34–41. doi.org/10.1111/j.1600-048X.2008.04377.x
Rockwell, S. M. (2013). Carry-over effects from the non-breeding season influence
spring arrival dates, reproductive success and survival in an endangered
migratory bird, the Kirtland’s warbler (Setophaga kirtlandii) [Ph.D. thesis].
University of Maryland.
Rockwell, S. M., Bocetti, C. I., & Marra, P. P. (2012). Carry-over effects of winter
climate on spring arrival date and reproductive success in an endangered
migratory bird, Kirtland’s Warbler (Setophaga kirtlandii). The Auk, 129(4), 744–
752. doi.org/10.1525/auk.2012.12003

84
Rohwer, V. G., & Rohwer, S. (2013). How do birds adjust the time required to replace
their flight feathers? The Auk, 130(4), 699–707. doi.org/10.1525/auk.2013.13042
Romano, A., Saino, N., & Møller, A. P. (2017). Viability and expression of sexual
ornaments in the barn swallow Hirundo rustica: A meta‐analysis. Journal of
Evolutionary Biology, 30(10), 1929–1935. doi.org/10.1111/jeb.13151
Rosenberg, K. V., Dokter, A. M., Blancher, P. J., Sauer, J. R., Smith, A. C., Smith, P. A.,
Stanton, J. C., Panjabi, A., Helft, L., Parr, M., & Marra, P. P. (2019). Decline of
the North American avifauna. Science, 366(6461), 120–124.
doi.org/10.1126/science.aaw1313
Rubolini, D., Massi, A., & Spina, F. (2002). Replacement of body feathers is associated
with low pre‐migratory energy stores in a long‐distance migratory bird, the barn
swallow (Hirundo rustica). Journal of Zoology, 258(4), 441–447.
doi.org/10.1017/S0952836902001590
Saino, N., Szép, T., Ambrosini, R., Romano, M., & Møller, A. P. (2004). Ecological
conditions during winter affect sexual selection and breeding in a migratory bird.
Proceedings of the Royal Society of London. Series B: Biological Sciences,
271(1540), 681–686. doi.org/10.1098/rspb.2003.2656
Smith, J. A. M., Reitsma, L. R., & Marra, P. P. (2010). Moisture as a determinant of
habitat quality for a nonbreeding Neotropical migratory songbird. Ecology,
91(10), 2874–2882. doi.org/10.1890/09-2212.1
Sparrow, K. L., Donkor, K. K., Flood, N. J., Marra, P. P., Pillar, A. G., & Reudink, M. W.
(2017). Conditions on the Mexican moulting grounds influence feather colour and
carotenoids in Bullock’s orioles (Icterus bullockii). Ecology and Evolution, 7(8),
2643–2651. doi.org/10.1002/ece3.2836
Stanley, C. Q., MacPherson, M., Fraser, K. C., McKinnon, E. A., & Stutchbury, B. J. M.
(2012). Repeat tracking of individual songbirds reveals consistent migration
timing but flexibility in route. PLoS ONE, 7(7), e40688.
doi.org/10.1371/journal.pone.0040688
Stewart, R. L. M., Francis, C. M., & Massey, C. (2002). Age-related differential timing of
spring migration within sexes in passerines. The Wilson Bulletin, 114(2), 264–271.
doi.org/10.1676/0043-5643(2002)114[0264:ARDTOS]2.0.CO;2
Studds, C. E., & Marra, P. P. (2005). Nonbreeding habitat occupancy and population
processes: An upgrade experiment with a migratory bird. Ecology, 86(9), 2380–
2385. doi.org/10.1890/04-1145

85
Studds, C. E., & Marra, P. P. (2007). Linking fluctuations in rainfall to nonbreeding
season performance in a long-distance migratory bird, Setophaga ruticilla.
Climate Research, 35, 115–122. doi.org/10.3354/cr00718
Studds, C. E., & Marra, P. P. (2011). Rainfall-induced changes in food availability modify
the spring departure programme of a migratory bird. Proceedings of the Royal
Society B: Biological Sciences, 278(1723), 3437–3443.
doi.org/10.1098/rspb.2011.0332
Stutchbury, B. J. M., Gow, E. A., Done, T., MacPherson, M., Fox, J. W., & Afanasyev, V.
(2011). Effects of post-breeding moult and energetic condition on timing of
songbird migration into the tropics. Proceedings of the Royal Society B:
Biological Sciences, 278(1702), 131–137. doi.org/10.1098/rspb.2010.1220
Surmacki, A. (2008). Preen waxes do not protect carotenoid plumage from bleaching by
sunlight. Ibis, 150(2), 335–341. doi.org/10.1111/j.1474-919X.2007.00796.x
Svensson, E., & Hedenström, A. (1999). A phylogenetic analysis of the evolution of
moult strategies in Western Palearctic warblers (Aves: Sylviidae). Biological
Journal of the Linnean Society, 67(2), 263–276. doi.org/10.1111/j.10958312.1999.tb01864.x
Szép, T., Dobránszky, J., Møller, A. P., Dyke, G., & Lendvai, Á. Z. (2019). Older birds
have better feathers: A longitudinal study on the long-distance migratory Sand
Martin, Riparia riparia. PLOS ONE, 14(1), e0209737.
doi.org/10.1371/journal.pone.0209737
Taylor, P. D., Crewe, T. L., Mackenzie, S. A., Lepage, D., Aubry, Y., Crysler, Z., Finney,
G., Francis, C. M., Guglielmo, C. G., Hamilton, D. J., Holberton, R. L., Loring, P.
H., Mitchell, G. W., Norris, D. R., Paquet, J., Ronconi, R. A., Smetzer, J. R.,
Smith, P. A., Welch, L. J., & Woodworth, B. K. (2017). The Motus Wildlife
Tracking System: A collaborative research network to enhance the understanding
of wildlife movement. Avian Conservation and Ecology, 12(1), 8.
doi.org/10.5751/ACE-00953-120108
Terrill, R. S., Seeholzer, G. F., & Wolfe, J. D. (2020). Evolution of breeding plumages in
birds: A multiple‐step pathway to seasonal dichromatism in New World warblers
(Aves: Parulidae). Ecology and Evolution, 10(17), 9223–9239.
doi.org/10.1002/ece3.6606
Tonra, C. M., Marra, P. P., & Holberton, R. L. (2011). Migration phenology and winter
habitat quality are related to circulating androgen in a long-distance migratory
bird. Journal of Avian Biology, 42(5), 397–404. doi.org/10.1111/j.1600048X.2011.05333.x

86
Witkowska, M., Korniluk, M., Pinchuk, P., Tumiel, T., Karlionova, N., & Meissner, W.
(2024). The interplay between moult of flight feathers and fuelling conducted on
the breeding grounds of the Great Snipe Gallinago media from the eastern
European, lowland population [Preprint]. doi.org/10.1101/2024.01.23.576834
Wolfe, J. D., Ryder, T. B., & Pyle, P. (2010). Using molt cycles to categorize the age of
tropical birds: An integrative new system: Molt-Cycle Age-Categorization
System. Journal of Field Ornithology, 81(2), 186–194. doi.org/10.1111/j.15579263.2010.00276.x
Wright, J. R., Powell, L. L., Matthews, S. N., & Tonra, C. M. (2020). Rusty Blackbirds
select areas of greater habitat complexity during stopover. The Condor, 122(4),
duaa040. doi.org/10.1093/condor/duaa040
Wright, J. R., Tonra, C. M., & Powell, L. L. (2018). Prealternate molt-migration in Rusty
Blackbirds and its implications for stopover biology. The Condor, 120(3), 507–
516. doi.org/10.1650/CONDOR-17-177.1

87
CHAPTER 4: CONCLUSION
My goal was to evaluate the potential role of prealternate moult in carry-over
effects from the stationary nonbreeding period to pre-breeding migration in NearcticNeotropical migratory songbirds. Because there is a large gap in our basic knowledge of
prealternate moult (reviewed in Chapter 2), the first step in accomplishing my goal was to
quantify patterns, timing, and intensity of prealternate moult on the nonbreeding grounds
in Jamaica (Chapter 2). From January to April 2023, I quantified moult in six species of
parulid warbler. I demonstrated that head and body feather moult is common for the
Northern Waterthrush (Parkesia noveboracensis), Black-and-white Warbler (Mniotilta
varia), American Redstart (Setophaga ruticilla), Northern Parula (Setophaga americana),
and Prairie Warbler (Setophaga discolor), and increases in frequency and intensity later
in the nonbreeding period for most species. For the Ovenbird (Seiurus aurocapilla), I
confirmed the occurrence of prealternate moult in at least some individuals. Using this
new foundational knowledge of prealternate moult, I then assessed whether nonbreeding
habitat quality and body condition influenced the timing and intensity of moult, and, in
turn, if moult influenced the timing of departure on pre-breeding migration in three
species: Black-and-white Warbler, Northern Waterthrush, and American Redstart
(Chapter 3). I found that the intensity and timing of prealternate moult appears to be
flexible and vary in response to habitat quality and body condition. Ultimately, I
discovered links between nonbreeding conditions, prealternate moult intensity, and
departure dates in the American Redstart. I also found inter- and intraspecific variation in
the influence of nonbreeding conditions on prealternate moult and their subsequent
effects on departure timing, indicating that moult and departure phenology can be driven
by both endogenous and exogenous factors regardless of nonbreeding environmental
conditions.
Significance to the field of ornithology
The findings of my research fill gaps in our basic understanding of prealternate
moult phenology for migratory parulid warblers. To the best of my knowledge, this work
represents the first quantification of contour feather moult intensity in all six study

88
species and provide the first detailed description of contour feather moult patterns and
timing in the Ovenbird. This study appears to be only the third of its kind to investigate
the role of prealternate moult in nonbreeding period carry-over effects in parulid warblers
(see Boone et al. 2010 and Jones et al. 2014), and the first to examine it in the species I
studied. Thus, finding links between nonbreeding conditions, moult, and departure date in
the American Redstart is a meaningful advancement and highlights the importance of
considering moult in seasonal interactions across the avian annual cycle. While more
work must be done to understand the mechanisms regulating prealternate moult and the
variation across species, age and sex classes, my results serve as a jumping off point and
can be used for comparison with other nonbreeding populations and years. The data I
collected for this research can have long-term impacts through its inclusion in 1) the
publicly accessible Motus database of global migration tracking data (www.motus.org),
and 2) a long-term site-specific dataset for Font Hill Nature Preserve, Jamaica, which
over the past 30 plus years has been instrumental to the scientific community’s
understanding of the nonbreeding ecology of migratory birds. Further, my research will
be widely accessible through the publication of this thesis on an open access repository,
and peer-reviewed publication of the data chapters.
Limitations and directions for future study
From previous research the general migratory paths and breeding ranges of my
study species are known (i.e., eastern and/or east-central North America), but the
locations of breeding grounds vary widely within the study population. For example,
American Redstarts migrating from Jamaica likely breed from Alabama in the south to as
far north as southern Ontario, resulting in migration distances differing among
individuals by 2,000 kilometres or more (Dossman et al. 2023). Innate migration
schedules based on breeding latitude and associated migration distance likely explain
some of the intraspecific variation in departure dates (e.g., Neufeld et al. 2021) and
confound my ability to disentangle the effects of body condition and moult intensity on
departure timing. Accounting for migration distance, for example by determining
breeding latitude from stable-hydrogen isotopes in feathers grown on the breeding
grounds (e.g., Dossman et al. 2023), could clarify the relationships I detected. Further,

89
migration phenology can vary annually with variation in seasonal resources mediated by
tropical precipitation (e.g., Studds and Marra 2011, Rockwell et al. 2012, Akresh et al.
2019). Because the timing and intensity of moult appears to be flexible with variable
body condition—typically associated with habitat quality—a multi-year study to account
for interannual variation would be valuable.
Xeric habitat, which is typically poorer in quality than mesic habitat (Rodríguez
Vásquez and Taylor 2024), was associated with Black-and-white Warblers in better
condition during my study period, contrary to previous work (Paxton and Moore 2015,
Cooper et al. 2021). However, habitat moisture may not be the primary driver of habitat
quality for this species, as suggested for the Magnolia Warbler (Setophaga magnolia;
Boone et al. 2010), but further study is needed. I could not conclusively determine
relative habitat quality (i.e., habitat conditions associated with body condition) at my
study site for the Northern Waterthrush, potentially due to my small sample size (n = 14).
For these two species, delineating individual territories (e.g., Marra et al. 2015a, Cooper
et al. 2021) and increasing repeated body condition measurements across the season may
better indicate habitat use and quality and could improve our ability to determine their
combined effects on moult intensity.
To better understand the endogenous and exogenous mechanisms that regulate
prealternate moult, future studies should increase the number of repeated measurements
of moult intensity across the season in the same individuals, for example every two
weeks during the moult period. This approach can facilitate understanding the complete
extent of individual moult, which I could not determine from my moult intensity metric. I
detected a lower range in moult intensity scores than other studies of contour feather
moult in songbirds, during the prealternate (Wright et al. 2018) and prebasic moults
(Hutton et al. 2021). However, the biological significance, specifically the physiological
costs and benefits, of differing contour feather moult intensities is not known. Increased
repeated measurements of moult intensity and body condition across the nonbreeding
period, combined with measurements of physiological stress indicated by corticosterone
(e.g., Harms et al. 2015), could be used to evaluate if birds make trade-offs between
prealternate moult intensity and migratory fattening. Tracking departure dates in those

90
same individuals year-over-year (e.g., Studds and Marra 2011) would allow researchers
to evaluate the potential consequences of such trade-offs on migration timing.
Management implications
While the species I studied are currently listed as Least Concern by the
International Union for Conservation of Nature (IUCN 2024), Nearctic-Neotropical
migrants as a group have declined drastically over the past 50 years (Rosenberg et al.
2019) and continue to face threats such as habitat degradation and loss, exacerbated by
global climate change (Albert et al. 2020). My research serves as an important early step
in understanding the role of prealternate moult in carry-over effects within the annual
cycle. Considering how all life stages interact across the avian annual cycle is critical for
identifying the spatial and temporal periods that limit declining populations (Marra et al.
2015b), and then applying this information to prioritize high-quality habitats for
conservation.
Understanding if and how life stages that occur on the nonbreeding grounds are
limiting has become increasingly important as a long-term drying trend impacts the
Caribbean (Neelin et al. 2006). As a result of this trend, precipitation during the tropical
dry season in Jamaica has declined and become less predictable (Studds and Marra 2011),
and this unpredictable season coincides with the nonbreeding period for NearcticNeotropical migrants. Periods of drought, or reduced rainfall, have been linked to
reduced annual survival (e.g., Dossman et al. 2023), reduced breeding success (e.g.,
Rockwell et al. 2017), and shifts in breeding range (e.g., Dossman et al. 2023). Further
drying may amplify these effects, leading to further population declines. Projections also
suggest that the frequency of North Atlantic hurricanes will increase (Emanuel 2005,
reviewed in Knutson et al. 2010). Although their effect on migrant survival appears to be
minimal, these storms can alter the quality of habitat, causing migrants to shift toward
less damaged and potentially poorer-quality, habitat (Wunderle Jr. and Arendt 2017). In
fact, since I conducted my field research, Jamaica was hit by Hurricane Beryl, a category
4 storm (July 2024; NOAA 2024). Hurricane Beryl caused serious damage to the
southwest coast of the island (CBC 2024), where the study site is located. In the context
of global climate change, managing high quality nonbreeding habitats such as mesic

91
mangrove is now more important for migratory birds than ever (Albert et al. 2020),
especially because coastal habitats are under constant threat of degradation or loss
associated with unsustainable management and land development (Muñoz Sevilla and La
Bail 2017, Guimarais et al. 2021, Gómez et al. 2022).
People of diverse demographics across the Americas and the Caribbean share in
the delights, and ecological benefits, of Nearctic-Neotropical migratory birds.
International collaboration is essential to achieving the shared goal of reversing the
decline in migratory bird populations and safeguarding their habitats (Runge et al. 2015,
Albert et al. 2020). To this end, it is crucial that researchers and land managers from the
Global North (i.e., the United States and Canada) establish equitable collaborations with
Neotropical partners to support locally led work. Improving equity will require those
global northerners with current Neotropical projects to critically evaluate how their work
has in the past, or present, contributed to the systemic exclusion of Neotropical
researchers, much to the detriment of the field of ornithology (Soares et al. 2023). While
much work remains on this front, improving equity to emphasize local objectives and
leadership can strengthen both science and conservation (de Vos 2022, Soares et al.
2023).

92
LITERATURE CITED
Akresh, M. E., King, D. I., & Marra, P. P. (2019). Examining carry‐over effects of winter
habitat on breeding phenology and reproductive success in prairie warblers
Setophaga discolor. Journal of Avian Biology, 50(4), e02025.
doi.org/10.1111/jav.02025
Albert, S., Wolfe, J. D., Kellerman, J., Sherry, T., Stutchbury, B. J. M., Bayly, N. J., &
Ruiz-Sánchez, A. (2020). Habitat ecology of Nearctic–Neotropical migratory
landbirds on the nonbreeding grounds. The Condor, 122(4), 1–18.
doi.org/10.1093/condor/duaa055
Boone, A. T., Rodewald, P. G., & DeGroote, L. W. (2010). Neotropical winter habitat of
the magnolia warbler: Effects on molt, energetic condition, migration timing, and
hematozoan infection during spring migration. The Condor, 112(1), 115–122.
doi.org/10.1525/cond.2010.090098
CBC (2024, July 3). Hurricane Beryl pummels Jamaica after ripping through southeast
Caribbean. https://www.cbc.ca/news/world/hurricane-beryl-caribbean-1.7252868.
Accessed 2024-07-17.
Cooper, N. W., Thomas, M. A., & Marra, P. P. (2021). Vertical sexual habitat segregation
in a wintering migratory songbird. Ornithology, 138(1), ukaa080.
doi.org/10.1093/ornithology/ukaa080
De Vos, A. (2022). Stowing parachutes, strengthening science. Conservation Science and
Practice, 4(5), e12709. doi.org/10.1111/csp2.12709
Dossman, B. C., Studds, C. E., LaDeau, S. L., Sillett, T. S., & Marra, P. P. (2023). The
role of tropical rainfall in driving range dynamics for a long-distance migratory
bird. Proceedings of the National Academy of Sciences, 120(52), e2301055120.
doi.org/10.1073/pnas.2301055120
Emanuel, K. (2005). Increasing destructiveness of tropical cyclones over the past 30
years. Nature, 436(7051), 686–688. doi.org/10.1038/nature03906
Gómez, I., Silva, R., Lithgow, D., Rodríguez, J., Banaszak, A. T., & Van Tussenbroek, B.
(2022). A review of disturbances to the ecosystems of the Mexican Caribbean,
their causes and consequences. Journal of Marine Science and Engineering,
10(5), 644. doi.org/10.3390/jmse10050644
Guimarais, M., Zúñiga-Ríos, A., Cruz-Ramírez, C. J., Chávez, V., Odériz, I., Van
Tussenbroek, B. I., & Silva, R. (2021). The conservational state of coastal
ecosystems on the Mexican Caribbean coast: Environmental guidelines for their
management. Sustainability, 13(5), 2738. doi.org/10.3390/su13052738

93
Harms, N. J., Legagneux, P., Gilchrist, H. G., Bêty, J., Love, O. P., Forbes, M. R.,
Bortolotti, G. R., & Soos, C. (2015). Feather corticosterone reveals effect of
moulting conditions in the autumn on subsequent reproductive output and survival
in an Arctic migratory bird. Proceedings of the Royal Society B: Biological
Sciences, 282(1800), 20142085. doi.org/10.1098/rspb.2014.2085
Hutton, P., McKenna, J., & McGraw, K. J. (2021). Urban links to molt schedule, body
condition and carotenoid‐based coloration in the house finch Haemorhous
mexicanus. Journal of Avian Biology, 52(5), jav.02761. doi.org/10.1111/jav.02761
IUCN (2024). The IUCN Red List of Threatened Species (Version 2024-1).
https://www.iucnredlist.org. Accessed 17-July-2024.
Jones, T. B., Drake, A., & Green, D. J. (2014). Individual quality explains association
between plumage colouration, arrival dates and mate acquisition in yellow
warblers (Setophaga petechia). BMC Ecology, 14(1), 13. doi.org/10.1186/14726785-14-13
Knutson, T. R., McBride, J. L., Chan, J., Emanuel, K., Holland, G., Landsea, C., Held, I.,
Kossin, J. P., Srivastava, A. K., & Sugi, M. (2010). Tropical cyclones and climate
change. Nature Geoscience, 3(3), 157–163. doi.org/10.1038/ngeo779
Marra, P. P., Cohen, E. B., Loss, S. R., Rutter, J. E., & Tonra, C. M. (2015). A call for full
annual cycle research in animal ecology. Biology Letters, 11(8), 20150552.
doi.org/10.1098/rsbl.2015.0552
Marra, P. P., Studds, C. E., Wilson, S., Sillett, T. S., Sherry, T. W., & Holmes, R. T.
(2015). Non-breeding season habitat quality mediates the strength of densitydependence for a migratory bird. Proceedings of the Royal Society B: Biological
Sciences, 282(1811), 20150624. doi.org/10.1098/rspb.2015.0624
Muñoz Sevilla, N. P., & Le Bail, M. (2017). Latin American and Caribbean regional
perspective on Ecosystem Based Management (EBM) of Large Marine
Ecosystems goods and services. Environmental Development, 22, 9–17.
doi.org/10.1016/j.envdev.2017.01.006
Neelin, J. D., Münnich, M., Su, H., Meyerson, J. E., & Holloway, C. E. (2006). Tropical
drying trends in global warming models and observations. Proceedings of the
National Academy of Sciences, 103(16), 6110–6115.
doi.org/10.1073/pnas.0601798103
Neufeld, L. R., Muthukumarana, S., Fischer, J. D., Ray, J. D., Siegrist, J., & Fraser, K. C.
(2021). Breeding latitude is associated with the timing of nesting and migration
around the annual calendar among Purple Martin (Progne subis) populations.
Journal of Ornithology, 162(4), 1009–1024. doi.org/10.1007/s10336-021-01894w

94
National Oceanic and Atmospheric Administration (NOAA). (2024). Post-tropical
Cyclone Beryl at 43.1°N - 80.3°W.
https://www.star.nesdis.noaa.gov/GOES/floater.php?stormid=AL022024#tab2.
Accessed 2024-07-17.
Paxton, K. L., & Moore, F. R. (2015). Carry-over effects of winter habitat quality on en
route timing and condition of a migratory passerine during spring migration.
Journal of Avian Biology, 46(5), 495–506. https://doi.org/10.1111/jav.00614
Rockwell, S. M., Bocetti, C. I., & Marra, P. P. (2012). Carry-over effects of winter
climate on spring arrival date and reproductive success in an endangered
migratory bird, Kirtland’s Warbler (Setophaga kirtlandii). The Auk, 129(4), 744–
752. doi.org/10.1525/auk.2012.12003
Rodríguez Vásquez, F., & Taylor, C. (2024). An overview of the drivers of performance
in nonbreeding Nearctic–Neotropical migratory songbirds. Ornithology, ukae018.
doi.org/10.1093/ornithology/ukae018
Rosenberg, K. V., Dokter, A. M., Blancher, P. J., Sauer, J. R., Smith, A. C., Smith, P. A.,
Stanton, J. C., Panjabi, A., Helft, L., Parr, M., & Marra, P. P. (2019). Decline of
the North American avifauna. Science, 366(6461), 120–124.
doi.org/10.1126/science.aaw1313
Runge, C. A., Watson, J. E. M., Butchart, S. H. M., Hanson, J. O., Possingham, H. P., &
Fuller, R. A. (2015). Protected areas and global conservation of migratory birds.
Science, 350(6265), 1255–1258. doi.org/10.1126/science.aac9180
Soares, L., Cockle, K. L., Ruelas Inzunza, E., Ibarra, J. T., Miño, C. I., Zuluaga, S.,
Bonaccorso, E., Ríos-Orjuela, J. C., Montaño-Centellas, F. A., Freile, J. F.,
Echeverry-Galvis, M. A., Bonaparte, E. B., Diele-Viegas, L. M., Speziale, K.,
Cabrera-Cruz, S. A., Acevedo-Charry, O., Velarde, E., Cuatianquiz Lima, C.,
Ojeda, V. S., … Martins, P. V. R. (2023). Neotropical ornithology: Reckoning
with historical assumptions, removing systemic barriers, and reimagining the
future. Ornithological Applications, 125(1), duac046.
doi.org/10.1093/ornithapp/duac046
Studds, C. E., & Marra, P. P. (2011). Rainfall-induced changes in food availability modify
the spring departure programme of a migratory bird. Proceedings of the Royal
Society B: Biological Sciences, 278(1723), 3437–3443.
doi.org/10.1098/rspb.2011.0332
Wright, J. R., Tonra, C. M., & Powell, L. L. (2018). Prealternate molt-migration in Rusty
Blackbirds and its implications for stopover biology. The Condor, 120(3), 507–
516. doi.org/10.1650/CONDOR-17-177.1

95
Wunderle, Jr., J., & Arendt, W. (2017). The plight of migrant birds wintering in the
Caribbean: Rainfall effects in the annual cycle. Forests, 8(4), 115.
doi.org/10.3390/f8040115

96
APPENDIX A
Table A.1. Estimates for significant predictors (p-value < 0.05) of moult intensity as
measured by the Combined Contour Moult Index, in the American Redstart and Prairie
Warbler, modelled three ways using a linear model to handle individuals with repeated
measurements (recaptured birds): a) excluding the second capture, b) excluding the first
capture, and c) including both captures. Linear modelling followed a stepwise regression
(backward elimination), starting with the full initial model including the explanatory
variables: Julian date, Julian date2, age, and sex. For the American Redstart, I considered
the interaction term age*Julian date. It was not significantly related to moult intensity, so
I removed it from the final model.
Species

Model

Explanatory Estimate Residual
variable
standard
error
(degrees of
freedom)
Julian date
0.063
0.024 (82)
Julian date2
-0.0004
0.0002 (82)
Age
0.377
0.106 (82)

t-value

pvalue

2.667
-2.710
3.545

0.009
0.008
0.0007

b. excluding
first capture

Julian date
Julian date2
Age

0.061
-0.0004
0.307

0.023 (82)
0.0001 (82)
0.103 (82)

2.683
-2.817
2.976

0.009
0.006
0.004

c. including
both
captures

Julian date
Julian date2
Age

0.063
-0.0004
0.335

0.022 (90)
0.0001 (90)
0.100 (90)

2.930
-3.034
3.347

0.004
0.003
0.001

a. excluding
second
capture

Julian date

0.086

0.029 (17)

2.955

0.009

Julian date2

-0.0007

0.0002 (17)

-3.229

0.005

b. excluding
first capture

Julian date
Julian date2

0.056
-0.0005

0.026 (17)
0.0002 (17)

2.141
-2.461

0.047
0.025

c. including
both
captures

Julian date
Julian date2

0.060
-0.0005

0.027 (18)
0.0002 (18)

2.285
-2.585

0.035
0.019

American a. excluding
Redstart
second
capture

Prairie
Warbler

97
Table A.2. Sample sizes of the numbers of six species moulting and not moulting at
capture, after the observed onset of moult, used as input for a Pearson’s chi-squared test.
Species
Ovenbird
Northern Waterthrush
Black-and-white Warbler
American Redstart
Northern Parula
Prairie Warbler

# moulting
10
75
62
89
14
18

# not moulting
8
36
8
14
5
3

98
APPENDIX B

Figure B.1. Boxplots of δ13C ratios (‰) sampled from A) Black-and-white Warbler, and
B) Northern Waterthrush, grouped by age class. δ13C ratios for each age class in the
Black-and-white Warbler are further grouped by sex because I could determine sex in the
field for these species.

99
Table B.1. Stable-nitrogen (δ15N) ratios (‰) sampled from Black-and-white Warblers
and Northern Waterthrushes in March and April, 2023, at Font Hill Nature Preserve,
Jamaica. I report corrected values measured in units of per mil (‰) relative to the
international standards Atmospheric Air for δ15N.
Species

Age

Sex

δ15N ratio (‰)

Black-and-white Warbler

Definitive-cycle

Female

7.65
7.32
5.60
5.68
5.55
5.36
6.30
5.73
7.32
7.00
6.41
6.59
6.66
6.76
6.80
5.88
5.91
7.20
5.50
6.36
5.98
2.08
7.08
5.26
6.02
6.51
7.44
7.69
7.63
8.47
7.87
7.45
7.31
6.46
6.98
7.47
6.53
7.35
7.00

Male

Northern Waterthrush

First-cycle

Female
Male

Definitive-cycle

NA

First-cycle

100
Age

Species

δ15N ratio (‰)

Sex

6.12
Table B.2. Summary of estimates (β) and 85% confidence intervals for explanatory
variables from the top-ranked models (ΔAICc < 1) of all analyses that had no effect
(85%CIs overlapped zero) on the dependent variable.
Model type

Species

Predicting
prealternate
moult intensity

Black-and-white Warbler

Predicting
departure date

American Redstart

American Redstart

Explanatory
variable
residual body
condition

β (85%CI)

residual body
condition
habitat type

-0.15 (-2.66, 2.05)

residual body
condition

-37.07 (-132.39, 0.51)

2.05 (-0.06, 4.01)

-0.02 (-0.15, 0.11)

Table B.3. Summary of the top ranked models (ΔAICc < 1) that predicted residual
prealternate moult intensity in the American Redstart, modelled three ways using a linear
model to handle individuals with repeated measurements (recaptured birds): a) excluding
the second capture, b) excluding the first capture, and c) including both captures. All
three modelling methods produced the same results.
Top ranked models (ΔAICc < 1)
a 1. age + sex

AICc ΔAICc AICc
weight
111.9 0.00
0.08

2. age + residual body condition + habitat type + sex
+ age*residual body condition + age*habitat type

112.0 0.01

0.08

3. age + residual body condition + habitat type +
age*residual body condition + age*habitat type

112.6 0.65

0.06

4. age + habitat type + sex + age*habitat type

112.6 0.68

0.06

111.9

0.00

0.08

2. age + residual body condition + habitat type + sex
+ age*residual body condition + age*habitat type

112.0 0.01

0.08

3. age + residual body condition + habitat type +
age*residual body condition + age*habitat type

112.6 0.65

0.06

4. age + habitat type + sex + age*habitat type

112.6 0.68

0.06

b 1. age + sex

101
Top ranked models (ΔAICc < 1)
c

1. age + sex

AICc ΔAICc AICc
weight
111.9 0.00
0.08

2. age + residual body condition + habitat type + sex
+ age*residual body condition + age*habitat type

112.0 0.01

0.08

3. age + residual body condition + habitat type +
age*residual body condition + age*habitat type

112.6 0.65

0.06

4. age + habitat type + sex + age*habitat type

112.6 0.68

0.06