Seasonal interactions in the Yellow Warbler

(Setophaga petechia): winter habitat use,

migration and demography

by

Anna Evelyn Gray Drake

M.Sc. (Animal Science), University of British Columbia, 2007 B.Sc. (Ecology and Evolution), Simon Fraser University, 2003

Thesis Submitted In Partial Fulfillment of the

Requirements for the Degree of

Doctor of Philosophy

in the

Department of Biological Sciences

Faculty of Science

Anna Evelyn Gray Drake 2013

SIMON FRASER UNIVERSITY

Spring 2013

All rights reserved. However, in accordance with the Copyright Act of Canada, this work may

be reproduced, without authorization, under the conditions for "Fair Dealing". Therefore, limited reproduction of this work for the purposes of

private study, research, criticism, review and news reporting is likely to be in accordance with the law, particularly if cited appropriately.

ii

Approval

Name: Anna Evelyn Gray Drake

Degree: Doctor of Philosophy (Biology)

Title of Thesis: Seasonal interactions in the Yellow Warbler (Setophaga petechia): winter habitat use, migration and demography

Examining Committee: Chair: Arne Mooers

Professor

David J. Green

Senior Supervisor Associate Professor

David B. Lank Supervisor Adjunct Professor

Ronald C. Ydenberg

Supervisor Professor

Wendy J. Palen

Internal Examiner Assistant Professor Department of Biological Sciences

Peter P. Marra

External Examiner Adjunct Professor Department of Biology Queen’s University

Date Defended: April 16, 2013

iii

Partial Copyright Licence

Ethics Statement

The author, whose name appears on the title page of this work, has obtained, for the research described in this work, either:

a. human research ethics approval from the Simon Fraser University Office of Research Ethics,

or

b. advance approval of the animal care protocol from the University Animal Care Committee of Simon Fraser University;

or has conducted the research

c. as a co-investigator, collaborator or research assistant in a research project approved in advance,

or

d. as a member of a course approved in advance for minimal risk human research, by the Office of Research Ethics.

A copy of the approval letter has been filed at the Theses Office of the University Library at the time of submission of this thesis or project.

The original application for approval and letter of approval are filed with the relevant offices. Inquiries may be directed to those authorities.

Simon Fraser University Library Burnaby, British Columbia, Canada

update Spring 2010

iv

Abstract

Numerous bird species in the western hemisphere – Nearctic-Neotropical migrants –

breed and rear young in North America but spend two-thirds of the year south of the

Tropic of Cancer. The annual cycle of these species can be divided into a distinct

breeding, migratory and non-breeding period. Events in one period may alter

demographic rates in a subsequent period; however, the spatial and temporal separation

of these events and the difficulty of tracking individuals over the annual cycle make

detecting seasonal interactions/carry-over effects difficult. In this thesis I confirm that

13C and 15N isotopic signatures in Yellow Warbler (Setophaga petechia) tissues vary

with winter habitat use and can therefore be used as a means of inferring the wintering

conditions experienced by individuals captured on the breeding grounds. I show that

isotopic signatures suggestive of drier and possibly more southerly winter habitat use

are associated with delayed clutch initiation and lower productivity in young female

Yellow Warblers breeding in Revelstoke, British Columbia. In contrast, wintering isotopic

signatures do not influence the breeding phenology of other age/sex-classes in

Revelstoke or any age/sex-class in Inuvik, Northwest Territories, 2000 km further north.

Differences between Revelstoke and Inuvik in how productivity declines seasonally

suggest that one pathway through which carry-over effects can act is reduced or absent

in the north. I also examine the role of climate variables operating at discrete periods of

the annual cycle on the demography of Yellow Warblers in Revelstoke. I demonstrate

that wind conditions during spring migration are the best predictor of adult survival, male

arrival date and female clutch initiation date. These timing effects, in turn, impact

reproductive success. This finding indicates that the spring migration is a key period for

western Yellow Warblers, having a larger impact than either winter habitat use or

wintering conditions and influencing both demographic rates.

Keywords: migration; seasonal interactions; carry-over effects; breeding phenology; population demography; Setophaga petechia

v

Dedication

To Bus for walking with me when I was small

and sharing with me your love of nature, and

to my Mom and Dad for always sharing with

us your engagement and interest in the world

and in life

vi

Acknowledgements

Thank you to my advisory committee, Dr. Ron Ydenberg and Dr. David Lank, for your

help throughout the process: Ron for playing devil’s advocate and therefore making the

arguments and the resulting science much stronger and Dov for thoughtful feedback and

difficult questions which helped me better frame my thoughts and arguments. Thank you

to my supervisor Dr. David Green, for all of your help and for always being invested and

supportive. Thank you for asking me after my first field season when I was going to take

a break rather than when I was going to start analyzing the data. Connie Smith, Monica

Court and Marlene Nguyen: without your help I couldn’t have completed this degree.

Thank you to my examining committee, Dr. Peter Marra and Dr. Wendy Palen and to my

defense chair, Dr. Arne Mooers for investing your time and energy in reviewing and

critiquing my work, and in the thesis exam itself.

I owe many thanks to my field assistants. Stephanie Topp for your hard work over two

seasons in Inuvik and for your strong ethical standards regarding both the research sites

and the birds. The quality and quantity of the data from Inuvik owes a lot to your

excellent field and observational skills. Matthew Pennell for your dedication and

enthusiasm in 2010: thank you for forging ahead through long hours and millions of

mosquito and blackfly bites. Thank you to Simon Valdez for your indispensible help in

2011 (89 birds in 41 days!). Thank you also for your company and support: you made

the degree easier and everything else a lot more fun.

The Nihtat Gwich’in Renewable Resources Council, the Inuvik Hunters and Trappers

Committee and the Inuvik Town Council allowed me to collect data in the Inuvik area for

which I am grateful. The staff and visiting researchers at the Aurora Research Institute

were a great help in 2009 and 2010 and their company made my time in Inuvik

enjoyable. Thank you especially: Don Ross, Annika Trimble, Lisa Greenland, Debbie

Gully, William Hurst, and Pippa Seccombe-Hett. Environment Canada provided in-kind

support that made the Inuvik work feasible. A special thank-you goes to Joel Ingram and

Blake Bartzen at Environment Canada who managed the administrative details for this,

provided local bird knowledge, and offered assistance. Thank you to Dr. Guillermo

vii

Fernández at the Universidad Nacional Autónoma de México (UNAM), who enabled me

to conduct work in Mexico and received, held, and exported samples for the project. I am

greatly appreciative of your help and of the time and effort you put into this. The staff and

researchers at the UNAM field stations made my stay in two beautiful reserves

productive and enjoyable. Specific thanks goes to Dr. Jorge Vega and Rosamond

Coates for sharing their knowledge of wintering migrant habitat use within the Chamela-

Cuixmala and Los Tuxtlas regions. Thank you also to Rosamond and to the local

farmers and ranchers at Punta Pérula and Emiliano Zapata, JA and La Barra and

Sontecomapan, VZ for generously providing access to your properties for surveys and

banding.

I am also grateful to the undergraduate work-study students and assistants who helped

in Inuvik and did data processing over the years: Kate Snow, Daniel and Andrew Fehr,

Gabrielle Pang, Lindsay Davidson, Amirhossein Manavi, Pipreesh Gaind and especially

Kristina Hunter, who did the work of two assistants in 2011.

To my parents, thank you for your constant encouragement, your love, and the many,

many ways you have offered me support over my graduate career. To the rest of my

family, particularly Emily and Jol, and to each of my friends: thank you for your

encouragement and support. My fellow grad students (past and present) offered

feedback, assistance, coffee breaks, and entertainment and made my time at the CWE

memorable. Special thanks to Danica Hogan, Michelle Martin and Heidi Currier: when

the options were to laugh or cry, you made laughing much easier.

My graduate work would not have been possible without scholarship and research

support from the Natural Sciences and Engineering Research Council of Canada

(NSERC), the Northern Studies Training Program (NSTP), the Centre for Wildlife

Ecology, and from Simon Fraser University.

Finally, thank you to the birds: for putting up with the stress of capture, handling, feather

pulling, harassment at the nest, and the oddity of plastic rings on your legs. Three

seasons of working with Yellow Warblers have made me conclude that they are a little

like tiny, feathered bulldogs: occasionally bad tempered and always determined and

tenacious. My respect goes to the pairs that attempted to lure us away from the nest with

viii

injury displays, the females who did not flush when approached, and particularly

‘BwDgP’ who valiantly stayed at the nest edge to battle me, only 7000× her size, over

nest measures by grabbing and attempting to toss the ruler.

Aldo Leopold said that despite intense study we can never know all of the salient facts

that relate to the hundreds of little dramas that occur in the woods and fields. I think that

this is true, but that there is something of spiritual value in the attempt. I was witness to

hundreds of dramas in my time in Inuvik: nestlings and adults lost to predators, late

clutches abandoned, cuckoldry, mates that disappeared for unknown reasons, a female

who brooded an empty nest, and, more happily, a breeding pair who - after having

previously lost two clutches and built nests in three different locations - hatched a single

egg at the tail end of the season: an egg which both I and my field assistants had been

sure was infertile. Anyone casually walking in the immediate vicinity of these events

would not know about them; in fact, without some effort of observation, this person might

not even know there were little yellow birds going about their lives a few feet away. My

work allowed me to make this effort, to be an observer within this parallel world, and to

attempt to see things from a new, non-human perspective. I hope many people get this

kind of opportunity, because it is a door out of a limited, human-centered worldview into

something much more diverse and fascinating.

ix

Table of Contents

Approval .......................................................................................................................... ii Partial Copyright Licence................................................................................................ iii Abstract .......................................................................................................................... iv Dedication ....................................................................................................................... v Acknowledgements ........................................................................................................ vi Table of Contents ........................................................................................................... ix List of Tables .................................................................................................................. xi List of Figures................................................................................................................ xiii Introductory Image ........................................................................................................xvi

Chapter 1. General Introduction .............................................................................. 1 The role of the non-breeding period in population dynamics............................................ 2 Measuring the influence of the non-breeding period ........................................................ 3 Objectives ....................................................................................................................... 4 References ...................................................................................................................... 6 Figures .......................................................................................................................... 11

Chapter 2. Habitat use differs by sex in wintering Yellow Warblers ................... 13 Abstract ......................................................................................................................... 13 Introduction ................................................................................................................... 14 Methods ........................................................................................................................ 16

Study species ....................................................................................................... 16 Study areas .......................................................................................................... 17 Habitat searches, point counts and targeted mist-netting ..................................... 18

13C and 15N Isotope Signatures ......................................................................... 20 Statistical Analysis ................................................................................................ 20

Results .......................................................................................................................... 23 Habitat Use........................................................................................................... 23 Variation in morphology ........................................................................................ 23 Abundance and habitat use by sex, age and body size ........................................ 23 Within-habitat variation in 13C and 15N signatures .............................................. 25

Discussion ..................................................................................................................... 26 References .................................................................................................................... 31 Tables and Figures........................................................................................................ 37

Chapter 3. Carry-over effects of winter habitat vary with age and sex in Yellow Warblers (Setophaga petechia) ............................................... 40

Abstract ......................................................................................................................... 40 Introduction ................................................................................................................... 41 Methods ........................................................................................................................ 44

Study Species and location .................................................................................. 44 Monitoring and breeding ....................................................................................... 45 Sexing and aging .................................................................................................. 45 Feather samples and stable isotope analysis ....................................................... 46 Winter habitat use and 13C and 15N ................................................................... 46

x

Data Analysis ....................................................................................................... 48 Results .......................................................................................................................... 49

Age differences in phenology and breeding performance ..................................... 49 Winter habitat use and 13C and 15N ................................................................... 49 Carry-over effects ................................................................................................. 50

Discussion ..................................................................................................................... 51 References .................................................................................................................... 55 Tables and Figures........................................................................................................ 62

Chapter 4. Winter habitat use does not influence spring arrival dates or the reproductive success of Yellow Warblers breeding in the arctic ..................................................................................................... 67

Abstract ......................................................................................................................... 67 Introduction ................................................................................................................... 68 Methods ........................................................................................................................ 70

Study Species ...................................................................................................... 70 Study site and population monitoring .................................................................... 72 Sexing and aging .................................................................................................. 73 Tissue samples and stable isotope analysis ......................................................... 73 Carry-over effects ................................................................................................. 74

Results .......................................................................................................................... 75 Population traits .................................................................................................... 75 Carry-over effects ................................................................................................. 76 Comparison with mid-latitude sites ....................................................................... 77

Discussion ..................................................................................................................... 77 References .................................................................................................................... 81 Tables and Figures........................................................................................................ 87

Chapter 5. Wind speed during migration influences the survival, timing of breeding, and productivity of a Neotropical migrant..................... 92

Introduction ................................................................................................................... 92 Results and Discussion ................................................................................................. 95 Materials and Methods .................................................................................................. 96 References .................................................................................................................. 100 Tables and Figures...................................................................................................... 104

Chapter 6. Significance and Future Directions .................................................. 114 General Conclusions ................................................................................................... 114 Future directions ......................................................................................................... 116

Non-breeding effects – are there longitudinal patterns? ...................................... 116 References .................................................................................................................. 119 Tables and Figures...................................................................................................... 123

Appendices ............................................................................................................. 129 Appendix A. Derivation of wind data presented in Figure 6.1 .................................. 130 References: ................................................................................................................. 130

xi

List of Tables

Table 2.1 Morphometric traits (± SD) of Yellow Warblers wintering on the Pacific and Gulf coast of Mexico. Sample sizes are reported in brackets. The length of the 6th primary feather was measured in 2011 only and therefore the sample sizes for wing shape (P6/P9) are smaller than other metrics. ............................................................................................... 37

Table 3.1 Summary of stable isotope signatures (n, mean ±SD and range) from winter-grown feathers collected from Yellow Warblers breeding in Revelstoke, British Columbia. Yearling individuals are in their first breeding season; older birds are at least 2 years old. .................................. 62

Table 3.2 Results of general linear model analyses for each stage of the proposed pathway from winter habitat use to fledgling productivity. Relationships for yearling and older Yellow Warbler are reported separately for (A) males and (B) females. Standardized beta coefficients for significant continuous variables are reported in Figure 3.2. .............................................................................................................. 63

Table 4.1 Multiple regression analyses testing the influence of winter habitat

(inferred from 13C and 15N) on Inuvik Yellow Warbler breeding metrics. For predicted relationships see Figure 4.1. Age and Year plus second-order interaction terms were included in initial models and dropped sequentially if they did not contribute to fit. Below each model

we report the contribution of 13C and 15N to model fit (italics) if these variables were to be independently added back into the final model........................................................................................................... 88

Table 4.2 Location effects on the relationship between first-clutch initiation date and productivity in Yellow Warblers from the perspective of (A) breeding males and (B) breeding females. For both sexes the best fitting model predicts a negative relationship at our mid-latitude (Revelstoke) site and no relationship at our high-latitude (Inuvik) site. See also Figure 4.2. ..................................................................................... 89

Table 5.1 Models describing apparent annual survival of Yellow Warblers breeding in Revelstoke, British Columbia (n=279 individuals, 437 encounters). Models are ranked using QAICc. Model number refers to regionally-specific climate variables described in the text. Age was included in all models as a covariate (see methods). The number of parameters in the model (K), Akaike’s information criterion (QAICc), QAICc difference from the top model (ΔQAICc), and Akaike weight (ωi) are reported. ....................................................................................... 104

xii

Table 5.2. Ranked summary of AICc support for the candidate climate models and the null model (YEAR + AGE) describing male Yellow Warbler arrival date (n=183) in Revelstoke, British Columbia. Age (SY=1 yr; ASY>2 yrs) was included in all models as a covariate (see methods). Model adjusted r2, the number of parameters in the model (K), Akaike’s information criterion adjusted for small sample size (AICc), AICc difference from the top model (ΔAICc), and Akaike weight (ωi) are reported. .............................................................................................. 106

Table 5.3 Ranked summary of AICc support for the candidate climate models and the null model (YEAR + AGE) describing female Yellow Warbler clutch initiation date (n=155) in Revelstoke, British Columbia. Age (SY=1 yr; ASY>2 yrs) was included in all models as a covariate (see methods). Model adjusted r2, the number of parameters in the model (K), Akaike’s information criterion adjusted for small sample size (AICc), AICc difference from the top model (ΔAICc), and Akaike weight (ωi) are reported. ............................................................................ 107

Table 5.4. Correlation (r) matrix of explanatory climate variables and time (year) (n=7). Significant relationships (Spearman’s ρ) are starred (*=0.01 ≤ P ≤ 0.05). Variables where the predicted relationship with survival and breeding phenology would suggest a possible confound between models are marked with (‡) ....................................................................... 108

Table 6.1. Overview of cross-seasonal studies involving Neotropical migrant passerines. Study number indicates study location in Figure 6.2. .............. 124

Table 7.1. Region limits and night wind vectors (March-May averages, 2003-2012) presented in Figure 6.1(b) ................................................................ 130

xiii

List of Figures

Figure 1.1 Data collection sites and their associated chapters within this thesis. Base map courtesy of Wikimedia Commons. ............................................... 11

Figure 1.2 Data collection sites shown within the Yellow Warbler breeding and wintering range (shaded regions). Dashed lines indicate the geographical limit of the eastern genetic line during the breeding (top) and wintering (bottom) period (Boulet et al. 2006). The probability of eastern haplotypes being present to the west of these lines is less than 0.1 (Boulet et al. 2006). ........................................................................ 12

Figure 2.1 Patterns of habitat segregation in Yellow Warblers on the (a) Pacific coast (“Chamela”) and (b) Gulf coast (“Los Tuxtlas”) of Mexico. Left columns represent point count results with individuals identified as older (ASY) males (solid dark grey) or ‘other’ (solid light grey). Right columns represent banding data where the ‘other’ category is separated into older females (solid white), yearling (SY) females (stippled), and yearling males (grey, diagonal lines). Sample sizes are indicated above each column. In Chamela, riparian habitat contained more males than agriculture and both of these habitats contained more males than scrub. In Los Tuxtlas, coastal and pasture habitats did not significantly differ in the proportion of males present. ....................... 38

Figure 2.2 Variation in Yellow Warbler blood-15N signatures within habitats in relation to body size. Habitat-types were classified as (a) agricultural habitat, Pacific coast, (b) riparian habitat, Pacific coast, (c) scrub/seasonal saline marsh, Pacific coast and (d) “Los Tuxtlas”, Gulf of Mexico. This last category merges data from birds using coastal and pasture habitats on the Gulf of Mexico as habitats were not isotopically distinct. Unfilled circles indicate females, filled circles indicate males. ............................................................................................. 39

Figure 3.1 Biplot of the mean (±SD) 15N (‰) and 13C (‰) values of red blood cells from Yellow Warblers over-wintering in different habitat-types on the southern Gulf coast (Veracruz) and the Pacific slope (Jalisco) of Mexico. Bracketed letters indicate signature overlap: birds from habitats with shared letters were not significantly different from each

other in terms of tissue 15N (A-C) and/or 13C (D-F). .................................. 64

Figure 3.2 Model pathways for Yellow Warbler males and females by age class. Black solid lines indicate significant relationships between variables (thickest line P<0.0001, mid-thickness P<0.001, thin line P<0.05). Dashed lines indicate no statistically-detectable relationship between variables. Standardized beta coefficients for significant continuous variables are shown above each arrow. Whole-model r2 values are reported in parentheses above the dependent variables for significant relationships only. ........................................................................................ 65

xiv

Figure 3.3 Relationship between winter isotope signatures andclutch initiation date in yearling female Yellow Warblers. Initiation dates are presented

as residuals from a model controlling for year and 15N (a) or for year

and 13C (b). ............................................................................................... 66

Figure 4.1 Suggested pathways through which winter-habitat-driven carry-over effects could act. For females in our study population, clutch initiation date was used as a proxy for arrival date (see text) and therefore condition- and timing-pathways to clutch initiation are not separate.

Depleted 13C and enriched 15N tissue signatures are expected to be associated with high quality winter habitat use. ............................................ 90

Figure 4.2 The relationship between first-clutch initiation date and annual productivity (total number of young fledged) among Yellow Warbler females within our high-latitude breeding population (Inuvik, top) and our mid-latitude breeding population (Revelstoke, bottom). See also Table 4.2. .................................................................................................... 91

Figure 5.1 The location of our study (A), the Yellow Warbler wintering range (blue), and the area used to calculate wind speed values during migration (yellow). Image courtesy of NASA. ............................................. 109

Figure 5.2 Relationship between westerly wind speed during migration and annual survival of Yellow Warblers. Points are apparent survival (± SE) of young (1 yr, open points) and older (≥2 yrs, filled points) birds (7 years, n=279 individuals, 437 encounters). Solid lines and shading represent predicted survival (ϕ) ± 95% CI from the top model assuming an average southerly wind vector. ............................................. 110

Figure 5.3 Predicted relationship between southerly wind speed during migration and annual survival of Yellow Warblers in Revelstoke, British Columbia (see also Fig. 1). Lines and shading represent predicted survival (ϕ) ± 95% CI for young (1 yr, dashed) and older (≥2 yrs, solid) birds from the top model assuming an average westerly wind vector (7 years, n=279 individuals, 437 encounters). ............................................ 111

Figure 5.4 Male arrival date (circles) and female clutch initiation date (diamonds) for Yellow Warblers as a function of westerly wind speed during migration. Points represent mean dates ± SE for young (1yr, open points) and older (≥2yrs, filled points) birds (6 years, n=338). .................... 112

Figure 5.5 Within-individual changes in arrival date (males; n=49) and clutch initiation date (females; n=21) associated with inter-annual variation in average westerly wind vectors. Box-plots show the median change, 10th and 90th percentile, and outliers. Positive values represent delayed breeding phenology in stronger wind years; negative values represent advanced breeding phenology in stronger wind years. Data was restricted to individuals that were ≥ 2 years of age in both years. Males and females exhibited a median shift of +2 and +4 days respectively................................................................................................ 113

xv

Figure 6.1 (A) The suggested migratory routes used by Neotropical migrants between temperate and tropical regions (Lincoln 1998) and (B) nighttime wind vectors encountered by migrants during the March-May period (10-year average (2003-2012)). Average vectors were calculated within regions bounded by a longitudinal and latitudinal grid (see Appendix A). Boxed regions indicate the approximate area for which vectors were calculated, but are unadjusted for projection. Arrow length corresponds to wind strength with the exception of (*) where length was halved relative to other vectors in the interest of space. ........................................................................................................ 127

Figure 6.2 The geographical distribution and main findings of cross-seasonal studies involving Neotropical migrant passerines. Evidence gathered from wintering isotopic signatures are summarized by age- and sex-class in (A), evidence from wintering climatological data is summarized in (B). Study details are presented in Table 6.1. .................... 128

xvi

Introductory Image

1

Chapter 1. General Introduction

Each year North America experiences a dramatic increase in bird abundance, driven by

the seasonal movement of Nearctic-Neotropical migrants from wintering habitat in

Mexico, Central, and South America to breeding locations in the United States and

Canada. This movement involves 252 landbird species and amounts to 3.5 billion

individuals redistributing themselves across an 18.5 million km2 region (Rich et al. 2004).

Within weeks, the number of birds in Canada and the United States quadruples (Rich et

al. 2004; e.g. Holmes 2007). In the subsequent months, these individuals transform the

abundant food of the northern summer into young and their numbers briefly jump to

about 5 billion (Michel et al. 2011). This tide of birdlife then recedes as migrants move

southward to the wintering grounds where they will spend the remaining two-thirds of the

year.

A feature of seasonal movement is that single populations are dependent on

conditions within several regions. For both migrant and resident species, demographic

rates in one period may be influenced by events in an earlier period (“seasonal

interactions”) (Webster and Marra 2005). For migrants, however, this earlier period is

geographically separate. Similarly, factors that limit resident numbers occur within their

breeding range; migrant numbers may be limited by conditions encountered at any

geographic location used during the annual cycle (Sutherland 1996; Newton 2004, 2006)

The role of non-breeding period in regulating and limiting migrant numbers is of

particular interest to ecologists and managers. Migrants tend to be well-studied on the

breeding grounds but they receive little attention during the wintering period and are

difficult to track during the migratory period (Faaborg et al. 2010). This creates a gap in

knowledge that makes migratory birds more vulnerable than resident species: we don’t

often know where or when populations are most limited, or what actions will better

facilitate the recovery of those species in decline (Kelly and Finch 1998).

2

The role of the non-breeding period in population dynamics

How might the non-breeding period influence survival and productivity? Migrants

can die at any point in the annual cycle; however, some periods may have a stronger

influence on demography than others. While mortality during migration is high (Sillett and

Holmes 2002), the overwintering period may play a larger role in population limitation.

Winter habitat availability and density dependence may determine population carrying

capacities (“winter-limited” populations; Newton 2004). Sexual habitat segregation and

competition on the wintering grounds may result in more young and female birds dying

than older males (Sherry and Holmes 1995, 1996; Marra and Holmes 2001). These

processes can limit population growth by lowering over-winter survival and juvenile

recruitment rates, and by producing male-biased sex ratios on the breeding grounds that

reduce population productivity (Marra and Holmes 2001; Steifetten and Dale 2006). The

non-breeding period may also alter productivity via a more complex route. The quality of

habitat occupied by individuals on the wintering grounds has been linked to the

maintenance of body weight in the period prior to spring migration (Marra and Holmes

2001; Studds and Marra 2005). Birds in high quality habitat are able to gain mass and

depart the wintering grounds earlier in spring (Marra et al. 1998; Studds and Marra

2005). These birds then arrive earlier on the breeding grounds and initiate reproduction

before their conspecifics (Norris et al. 2004). Because reproducing earlier in the season

tends to result in higher productivity, the repercussions of winter habitat use “carries-

over” to impact individual breeding success (Norris et al. 2004; Reudink et al. 2009;

Tonra et al. 2011). In this thesis I refer to this type of carry-over effect as “time-mediated”

because it is produced by wintering events that act entirely through the timing of

individual arrival on the breeding grounds. Winter habitat use can also influence the

condition of individuals when they arrive on the breeding grounds (González-Prieto

2012). Good wintering habitat may confer physical advantages such as greater fat

reserves, improved muscle condition or higher quality winter-moulted plumage

(González-Prieto 2012; Lindsay 2008; Jones, Drake, and Green, unpublished data). In

this thesis I refer to carry-over effects that do not act via arrival date as “condition-

mediated”. Birds arriving on the breeding grounds in better condition than conspecifics

may pair earlier, pair with a higher quality mate, initiate clutches earlier, or be better

providers. Productivity advantages for these birds follow from the advancement of

3

breeding phenology after arrival and/or from direct benefits. Individual carry-over effects

can scale-up to influence population productivity when a large proportion of the

population experiences poor non-breeding conditions.

Broad-scale climatic variation plays a role in the environment encountered by

migratory birds during their annual cycle (e.g. Saino et al. 2004; Mazerolle et al. 2011).

The effects of climate can be additive and/or interact with local-scale habitat quality to

influence the conditions faced by migrant populations (Brown and Sherry 2006; Studds

and Marra 2007). Climate conditions during the non-breeding period can affect migrant

survival and productivity. For example, links between climate cycles such as the El Niño

Southern Oscillation (ENSO) and the abundance, apparent annual survival, and

breeding phenology of migrant birds have been shown in multiple studies (Sillett et al.

2000; Mazerolle et al. 2005; Macmynowski et al. 2007). This association is often

attributed to variation in winter habitat quality produced by variation in rainfall. Similar

patterns where direct measures of wintering-ground rainfall or primary productivity are

used offer support for this interpretation (e.g. Nott et al. 2002; Saino et al. 2004;

Rockwell et al. 2012).

Measuring the influence of the non-breeding period

Migratory birds must be followed in space as well as over time in order to model

population dynamics accurately. For small-bodied species, spatial tracking is extremely

difficult: mark-resight/recapture is ineffective over large geographical scales (Wassenaar

and Hobson 2001) and remote monitoring devices are currently too heavy and generate

too much drag for species under 40g (Bowlin et al. 2010; Fraser et al. 2012). Ideally

technological advancements in the near future will allow us to follow individuals through

the whole annual cycle (e.g. Wikelski et al. 2007); in the meantime, alternative

approaches must be used. One approach is to infer habitat use or geographic origin

from intrinsic (naturally occurring, tissue-based) markers such as stable isotopes

(Hobson et al. 2010). Stable isotope ratios (particularly those of carbon (13C), nitrogen

(15N), and hydrogen (D)) provide a means of collecting cross-seasonal data because

they vary spatially in relation to physical, climatological and biological processes (Kelly

2000; Hobson et al. 2010). These ratios are incorporated into tissues grown while an

4

animal resides and feeds in a given area; individuals retain these signatures until this

tissue is replaced (Hobson 1999). For example, birds that undergo pre-alternate moult

on the wintering grounds incorporate wintering signatures into plumage that they will

retain for a full a year (Hobson 1999). Winter feathers collected on the breeding grounds

therefore offer a means of inferring certain aspects of non-breeding habitat use. A

second option for collecting cross-seasonal data is to use our current (albeit inexact)

knowledge of migrant population movements and distributions (e.g. Boulet et al. 2006;

Norris et al. 2006). By matching this data with climate data we can assess the role of

conditions at different points of the annual cycle in altering migrant phenology and in

limiting their numbers. In this approach, it is necessary to test regionally-specific models

as climate cycles influence large areas and can impact breeding, wintering and migration

conditions simultaneously (Nott et al. 2002; Forchhammer et al. 2002).

The influence of the non-breeding period on populations can be measured in

terms of between-individual differences (after controlling for annual effects), or it can be

measured in terms of within-individual or within-population differences between years. A

between-individual approach requires there to be variation in the types of non-breeding

habitat occupied by individuals within a single population (e.g. Marra et al. 1998). This

variation is detected using isotope signatures in tissue samples and the strength of the

habitat effect is then measured by contrasting the breeding performance of individuals

within the population. A between-year approach requires there to be year-to-year

variation in the climatic conditions faced by populations in non-breeding regions (e.g.

Saino et al. 2004; Rockwell et al. 2012). This variation is detected using regional climate

data and the strength of the effect is then measured by contrasting the performance of

the same individual or population between years. Differences in survival probabilities

with changing non-breeding conditions may also be detected using the second

approach, and be indicative of the role of these periods in population limitation (Baillie

and Peach 1992; Wilson et al. 2011).

Objectives

In this thesis, I seek to expand our current knowledge of how migratory bird

populations are influenced by events occurring outside of the breeding season. To do

5

this, I make use of both stable isotope and regional climate data to quantify the

conditions experienced by my study populations in the non-breeding period. I measure

the impact these conditions might have on individuals or breeding populations as a

whole by using between-individual (Chapter 3 and 4) and between-year approaches

(Chapter 5). My study species is the northern Yellow Warbler (Setophaga petechia).

Yellow Warblers are a good model species with which to investigate seasonal

interactions: they breed throughout North America and their migratory patterns and

wintering distributions have been delineated to a greater extent than those of other small

migrants (Boulet et al. 2006).

In Chapter 2, I collect data on Yellow Warblers wintering in on the Pacific and

Atlantic coasts of the isthmus region of Mexico (Figure 1.1). I document habitat use, and

age- and sex-distributions within habitat types. I then analyze blood collected from

wintering birds for its carbon (13C) and nitrogen (15N) values in order to determine the

extent to which habitat use can be inferred from tissue in this species.

In my third chapter I measure the strength of time- and condition-mediated carry-

over effects among Yellow Warblers breeding in Revelstoke, British Columbia (50°58’N,

118°12’W; Figure 1.1). To do this, I relate between-individual differences in breeding

phenology and performance to variation in winter habitat use inferred from 13C and 15N

signatures obtained from winter-grown plumage. In Chapter 4, I move 2000 km further

north - to the limit of the Yellow Warbler summer range - and collect breeding data from

birds within the Mackenzie watershed near Inuvik, Northwest Territories (68º 21’N, 133º

45’W; Figure 1.1). The different costs and benefits incurred by breeding at different

latitudes would be expected to influence the strength of carry-over effects. I therefore

compare my breeding populations in terms of phenology, productivity and the

relationship between these parameters and inferred winter habitat use.

Finally, in Chapter 5, I use a long-term dataset from the Revelstoke population to

assess the role of climate, acting at different points in time and in different regions, on

breeding phenology and apparent annual survival. To reduce the confounding effects

produced by broad climate cycles that impact multiple regions at the same time, I

compete regionally-specific hypotheses.

6

The data in this thesis focuses on the annual cycle of migrants in western North

America. Both breeding sites in this study are used by Yellow Warblers of the western

genetic lineage and sampling sites in Mexico were selected due to the prevalence of

western migrants in these areas (Boulet et al. 2006; Figure 1.2). Chapter 3 and 4 are the

first studies to measure carry-over effects in a western landbird, while Chapter 5

assesses the role conditions in western Mexico play in breeding success and in survival.

Such information is needed, as most of the current North American literature is based on

research conducted with eastern birds. There is increasing evidence that eastern

migratory systems differ from western ones in terms of how strongly they are influenced

by wintering conditions (e.g. Wilson et al. 2011; LaManna et al. 2012). This suggests

that the migrant story is complex, and I offer my interpretation of the existing patterns

and the results of my research in the General Conclusion.

References

Baillie, Stephen R., and Will J. Peach. 1992. Population Limitation in Palaearctic-African Migrant Passerines. Ibis 134: 120–132. doi:10.1111/j.1474-919X.1992.tb04742.x.

Boulet, Marylène, H. Lisle Gibbs, and Keith A. Hobson. 2006. Integrated Analysis of Genetic, Stable Isotope, and Banding Data Reveal Migratory Connectivity and Flyways in the Northern Yellow Warbler (Dendroica petechia; Aestiva Group). Ornithological Monographs (61): 29–78.

Bowlin, Melissa S., Per Henningsson, Florian T. Muijres, Roel H. E. Vleugels, Felix Liechti, and Anders Hedenström. 2010. The Effects of Geolocator Drag and Weight on the Flight Ranges of Small Migrants. Methods in Ecology and Evolution 1 (4): 398–402. doi:10.1111/j.2041-210X.2010.00043.x.

Brown, David, and Thomas Sherry. 2006. Food Supply Controls the Body Condition of a Migrant Bird Wintering in the Tropics. Oecologia 149 (1): 22–32. doi:10.1007/s00442-006-0418-z.

Faaborg, John, Richard T. Holmes, Angela D. Anders, Keith L. Bildstein, Katie M. Dugger, Sidney A. Gauthreaux, Patricia Heglund, et al. 2010. Recent advances in understanding migration systems of New World land birds. Ecological Monographs 80 (1): 3–48.

Forchhammer, M. C., E. Post, and N. C. Stenseth. 2002. North Atlantic Oscillation Timing of Long- and Short-distance Migration. Journal of Animal Ecology 71: 1002–1014.

7

Fraser, Kevin C., Bridget J. M. Stutchbury, Cassandra Silverio, Patrick M. Kramer, John Barrow, David Newstead, Nanette Mickle, et al. 2012. Continent-wide Tracking to Determine Migratory Connectivity and Tropical Habitat Associations of a Declining Aerial Insectivore. Proceedings of the Royal Society B: Biological Sciences 279 (1749): 4901–4906. doi:10.1098/rspb.2012.2207.

González-Prieto, Ana María. 2012. Factors Influencing Body Condition and Arrival Phenology of Neotropical Migrants at a Northern Spring Stopover Site. M.Sc., University of Saskatchewan.

Hobson, Keith A. 1999. Tracing Origins and Migration of Wildlife Using Stable Isotopes: a Review. Oecologia 120: 314–326. doi:10.1007/s004420050865.

Hobson, Keith A., Rachel Barnett-Johnson, and Cerling Thure. 2010. Using Isoscapes to Track Animal Migration. In Isoscapes: Understanding Movement, Pattern, and Process on Earth Through Isotope Mapping, pp. 273–298.

Holmes, Richard T. 2007. Understanding Population Change in Migratory Songbirds: Long-term and Experimental Studies of Neotropical Migrants in Breeding and Wintering Area. Ibis 149 (Suppl. 2): 2–13.

Kelly, Jeffrey F. 2000. Stable Isotopes of Carbon and Nitrogen in the Study of Avian and Mammalian Trophic Ecology. Canadian Journal of Zoology 78: 1–27. doi:10.1139/z99-165.

Kelly, Jeffrey F., and Deborah M. Finch. 1998. Tracking Migrant Songbirds with Stable Isotopes. Trends in Ecology and Evolution 13 (2): 48–49.

LaManna, J. A., T. L. George, J. F. Saracco, M. P. Nott, and D. F. Desante. 2012. El Nino-Southern Oscillation Influences Annual Survival of a Migratory Songbird at a Regional Scale. The Auk 129 (4): 1–10.

Lindsay, Andrea M. 2008. Seasonal Events and Associated Carry-over Effects in a Neotropical Migratory Songbird, the Yellow Warbler (Dendroica petechia). M.Sc.,

Ohio State University.

Macmynowski, Dena P., Terry L. Root, Grant Ballard, and Geoffrey R. Geupel. 2007. Changes in Spring Arrival of Nearctic-Neotropical Migrants Attributed to Multiscalar Climate. Global Change Biology 13 (11): 2239–2251. doi:10.1111/j.1365-2486.2007.01448.x.

Marra, Peter P., Keith A. Hobson, and Richard T. Holmes. 1998. Linking Winter and Summer Events in a Migratory Bird by Using Stable-Carbon Isotopes. Science 282 (5395): 1884 –1886. doi:10.1126/science.282.5395.1884.

Marra, Peter P., and Richard T. Holmes. 2001. Consequences of Dominance-Mediated Habitat Segregation in American Redstarts During the Nonbreeding Season. The Auk 118 (1): 92–104.

8

Mazerolle, Daniel F., Kevin W. Dufour, Keith A. Hobson, and Heidi E. den Haan. 2005. Effects of Large-scale Climatic Fluctuations on Survival and Production of Young in a Neotropical Migrant Songbird, the Yellow Warbler Dendroica petechia.

Journal of Avian Biology 36 (2): 155–163. doi:10.1111/j.0908-8857.2005.03289.x.

Mazerolle, Daniel F., Spencer G. Sealy, and Keith A. Hobson. 2011. Interannual Flexibility in Breeding Phenology of a Neotropical Migrant Songbird in Response to Weather Conditions at Breeding and Wintering Areas. Ecoscience 18 (1): 18–25. doi:10.2980/18-1-3345.

Michel, N., D. F. Desante, D. R. Kaschube, and M. Philip Nott. 2011. The Monitoring Avian Productivity and Survivorship (MAPS) Program Annual Reports, 1989-2006. NBII/MAPS Avian Demographics Query Interface. http://www.birdpop.org/nbii2006/NBIIHome.asp (January 2011).

Newton, Ian. 2004. Population Limitation in Migrants. Ibis 146 (2): 197–226. doi:10.1111/j.1474-919X.2004.00293.x.

———. 2006. Can Conditions Experienced During Migration Limit the Population Levels of Birds? Journal of Ornithology 147 (2): 146–166. doi:10.1007/s10336-006-0058-4.

Norris, D. Ryan, Peter P. Marra, Gabriel J. Bowen, Laurene M. Ratcliffe, Andrew J Royale, and T. Kurt Kyser. 2006. Migratory Connectivity of a Widely Distributed Songbird, the American Redstart (Setophaga ruticilla). Ornithological Monographs 61: 14–28.

Norris, D. Ryan, Peter P. Marra, T. Kurt Kyser, Thomas W. Sherry, and Laurene M. Ratcliffe. 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:10.1098/rspb.2003.2569.

Nott, M. Philip, David F Desante, Rodney B Siegel, and Peter Pyle. 2002. Influences of the El Niño/Southern Oscillation and the North Atlantic Oscillation on Avian Productivity in Forests of the Pacific Northwest of North America. Global Ecology and Biogeography 11 (4): 333–342. doi:10.1046/j.1466-822X.2002.00296.x.

Reudink, Matthew W, Peter P Marra, T. Kurt Kyser, Peter T Boag, Kathryn M Langin, and Laurene M Ratcliffe. 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:10.1098/rspb.2008.1452.

Rich, T.D., C. J. Beardmore, H. Berlanga, P. J. Blancher, M. S. W. Bradstreet, G. S. Butcher, D. W. Demarest, et al. 2004. Partners in Flight North American Landbird Conservation Plan. Cornell Lab of Ornithology. Ithaca, NY. http://www.partnersinflight.org/cont_plan/.

9

Rockwell, Sarah M., Carol I. Bocetti, and Peter P. Marra. 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:10.1525/auk.2012.12003.

Saino, Nicola, Tibor Szép, Roberto Ambrosini, Maria Romano, and Anders Pape Møller. 2004. Ecological Conditions During Winter Affect Sexual Selection and Breeding in a Migratory Bird. Proceedings of the Royal Society B: Biological Sciences 271 (1540): 681–686.

Sherry, Thomas W., and Richard T. Holmes 1995. Summer Versus Winter Limitation of Populations: Conceptual Issues and Evidence. In Ecology and Management of Neotropical Migratory Birds: A Synthesis and Review of Critical Issues, edited by T. E. Martin and D. M. Finch, pp. 85–120. Oxford University Press, New York.

Sherry, Thomas W., and Richard T. Holmes. 1996. Winter Habitat Quality, Population Limitation, and Conservation of Neotropical-Nearctic Migrant Birds. Ecology 77 (1): 36–48. doi:10.2307/2265652.

Sillett, T. Scott, and Richard T Holmes. 2002. Variation in Survivorship of a Migratory Songbird Throughout Its Annual Cycle. Journal of Animal Ecology 71 (2): 296–308. doi:10.1046/j.1365-2656.2002.00599.x.

Sillett, T. Scott, Richard T. Holmes, and Thomas W. Sherry. 2000. Impacts of a Global Climate Cycle on Population Dynamics of a Migratory Songbird. Science 288 (5473): 2040–2042. doi:10.1126/science.288.5473.2040.

Steifetten, Oyuind, and Svein Dale. 2006. Viability of an endangered population of Ortolan Buntings : The effect of a skewed operational sex ratio. Biological Conservation 132 (1): 88–97.

Studds, Colin E, and Peter P Marra. 2007. Linking Fluctuations in Rainfall to Nonbreeding Season Performance in a Long-distance Migratory Bird, Setophaga ruticilla. Climate Research 35: 115–122. doi:10.3354/cr00718.

Studds, Colin E., and Peter P. Marra. 2005. Nonbreeding Habitat Occupancy and Population Processes: An Upgrade Experiment with a Migratory Bird. Ecology 86 (9): 2380–2385. doi:10.1890/04-1145.

Sutherland, William J. 1996. Predicting the Consequences of Habitat Loss for Migratory Populations. Proceedings of the Royal Society of London. Series B: Biological Sciences 263 (1375): 1325–1327. doi:10.1098/rspb.1996.0194.

Tonra, Christopher M., Peter P. Marra, and Rebecca L. Holberton. 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:10.1111/j.1600-048X.2011.05333.x.

10

Wassenaar, Leonard I., and Keith A. Hobson. 2001. A Stable-Isotope Approach to Delineate Geographical Catchment Areas of Avian Migration Monitoring Stations in North America. Environmental Science & Technology 35 (9): 1845–1850. doi:10.1021/es0010204.

Webster, Michael S., and Peter P. Marra. 2005. The Importance of Understanding Migratory Connectivity and Cross-seasonal Interactions. In Birds of Two Worlds: The Ecology and Evolution of Migration, edited by Russell Greenberg and Peter P. Marra, pp. 199–209. Baltimore, MD: Johns Hopkins University Press.

Wikelski, Martin, Roland W. Kays, N. Jeremy Kasdin, Kasper Thorup, James A. Smith, and George W. Swenson. 2007. Going Wild: What a Global Small-animal Tracking System Could Do for Experimental Biologists. Journal of Experimental Biology 210 (2): 181–186. doi:10.1242/jeb.02629.

Wilson, Scott, Shannon L. Ladeau, Anders P. Tøttrup, and Peter P. Marra. 2011. Range-wide effects of breeding- and nonbreeding-season climate on the abundance of a Neotropical migrant songbird. Ecology 92 (9): 1789–1798.

11

Figures

Figure 1.1 Data collection sites and their associated chapters within this thesis. Base map courtesy of Wikimedia Commons.

12

Figure 1.2 Data collection sites shown within the Yellow Warbler breeding and wintering range (shaded regions). Dashed lines indicate the geographical limit of the eastern genetic line during the breeding (top) and wintering (bottom) period (Boulet et al. 2006). The probability of eastern haplotypes being present to the west of these lines is less than 0.1 (Boulet et al. 2006).

13

Chapter 2. Habitat use differs by sex in wintering Yellow Warblers

Anna Drake, Simon O. Valdez Juarez, David J. Green1

Abstract

Sex-biases in habitat use are common among wood warblers (Parulidae) that are

territorial over winter. We show that in tropical deciduous forest on the Pacific coast of

Mexico, male Yellow Warblers predominate in riparian habitat and to a lesser extent in

agricultural field margins while females predominate in scrub/seasonal saline marsh

habitat near the coast. In contrast, we find no evidence of sexual habitat segregation

between the pasture and coastal habitats used by Yellow Warblers on the Gulf of

Mexico. Body size does not play a significant role in predicting habitat occupancy on the

Pacific coast, suggesting that segregation is produced either by differences in

preference or by competitive asymmetry produced by sex-specific traits other than

physical size. Additional, more subtle, differences in habitat use are suggested by blood-

isotope samples. While variation in individual 13C and 15N signatures are most strongly

related to occupancy of a particular habitat-type and region, body size (but not age or

sex) explains additional variation in 15N signatures within scrub and riparian habitats on

the Pacific coast. This pattern may be the product different foraging behaviours or

differences in microhabitat use within broader habitat-types. We discuss sexual habitat

segregation in relation to regional climate and suggest that the seasonally dry climate of

1 A version of this chapter has been submitted to The Condor

14

the northern and western portion of the Yellow Warbler wintering range might create

steeper habitat quality gradients that result in segregation.

Introduction

Wood warblers (Parulidae) exhibit a diversity of over-wintering strategies (Stutchbury et

al. 2005; Faaborg et al. 2010). Species that exploit predictable, defendable food sources

such as the American Redstart (Setophaga ruticilla) and Black-throated Blue Warbler

(Setophaga caerulescens) maintain permanent territories throughout the wintering

period and show site fidelity between years (Brown and Long 2007; Holmes, et al. 1989;

Wunderle and Latta 2000). Territoriality in other species is more flexible. Northern

Waterthrush (Parkesia noveboracensis) will aggressively defend winter territories in high

quality areas but allow extensive home-range overlap to occur in poor habitat (Smith et

al. 2011) and Tennessee Warblers (Oreothlypis peregrina) will form ephemeral territories

around productive nectar sources (Greenberg 1986). Winter flock formation is observed

in many species. Coniferous forest breeders such as Bay-breasted (Setophaga

castanea), Yellow-rumped (Setophaga coronata) and Blackpoll (Setophaga striata)

Warblers that feed on insects, fruits and nectar in the tropics will form intraspecific flocks

(Greenberg 1979) while species feeding primarily on insects, such as Golden-cheeked

(Setophaga chrysoparia), Black-throated Grey (Setophaga nigrescens), and Black-and-

white (Mniotilta varia) Warblers, will join mixed-species flocks (Rappole et al. 1999; Hutto

1994).

Winter habitat use varies with sex and age in many territorial warbler species.

Use patterns tend to be consistent, with males associated with forest, females found in

more open and early successional habitat (Ornat and Greenberg 1990; Parrish and

Sherry 1994; Stutchbury et al. 2005) and older individuals tending to occupy high quality

habitats (Latta and Faaborg 2002; Marra and Holmes 2001). These sex- and age-biases

among habitats can be driven by differing preference or can be the product of the

competitive exclusion of certain classes (females and young birds in all existing

examples) from high quality or less variable habitat. For example, studies suggest that

differing habitat preference produces sexual habitat segregation in Hooded Warblers

(Setophaga citrina). Male Hooded Warblers appear to exhibit a preference for vertical

15

landscape elements (Morton 1990) and natural disturbances that produce changes in

structure within forested territories can result in sites being abandoned by males and

occupied by females (Morton et al. 1993). Removal experiments within male-dominated

habitats do not result in females occupying vacated territories (Morton et al. 1987). In

contrast, studies involving American Redstarts suggest that sexual habitat segregation in

this species is a product of competitive exclusion (Marra and Holmes 2001). Male-

biased, mangrove habitat has greater and less variable insect biomass over the

wintering period (Studds and Marra 2005, 2007). Males removed from territories in

mangrove are readily replaced by females (Marra et al. 1993; Studds and Marra 2005),

who subsequently maintain better condition over the wintering period than females in

second-growth scrub habitat (Studds and Marra 2005).

Habitat segregation by sex or age can have a diversity of consequences for

populations. Where preference produces segregation, differences in the availability of

preferred habitat may mean that winter carrying capacities differ between age- and sex-

classes. Where competition for high quality habitat produces segregation, outcomes are

more complex. In competitive exclusion scenarios, classes without access to good

quality habitat may occupy marginal habitat and consequently be in poorer condition and

experience reduced over-winter survival (Sherry and Holmes 1995; Marra 2000). If

juveniles are forced to winter in more marginal habitat, juvenile survival and recruitment

into breeding populations may be reduced. Similarly, higher female mortality may skew

operational sex ratios on the breeding grounds (Sherry and Holmes 1995; Marra and

Holmes 2001; Townsend et al. 2009) and reduce population productivity (Steifetten and

Dale 2006). Meanwhile, population responses to landscape changes will be complex:

exclusion rates (and therefore the condition and survival) of subordinate age- and sex-

classes will vary with both the availability of high quality habitat and with overall

population size (Sherry and Holmes 1995; Marra and Holmes 2001). Despite the

potential to influence population demography, we know very little about the habitat use

patterns exhibited by most warbler species (Morton and Stutchbury 2005; Faaborg et al.

2010).

Yellow Warblers are one of the most abundant North American warblers, with an

estimated population size of 30 million (Rich et al. 2004). They have been the subject of

many breeding-ground studies (from Bigglestone 1913 to Quinlan and Green 2012) but

16

there is still relatively little is known about their wintering ecology. In this study, we

compare the morphology of Yellow Warbler wintering within the Pacific and Gulf coast

regions of the isthmus of Mexico, assess habitat use within these regions, and examine

the extent to which habitat use varies with sex, age and body size. Body size is not the

only factor that could determine the outcome of intraspecific competitions (Holmgren and

Lundberg 1993; Arizaga and Bairlein 2011), however, body size differences among

habitats could indicate competitive exclusion. Previous work has shown that the type of

winter habitat occupied by Yellow Warblers can be discerned from 13C and 15N

signatures in tissue samples (Drake et al. 2013; Chapter 3). We use 13C and 15N to

further evaluate whether there is evidence of different within-habitat, microhabitat use

associated with sex, age, or body size.

Methods

Study species

Yellow Warblers are small, insectivorous birds that breed in riparian margins and

wet deciduous forest throughout North America. They winter primarily in lowland regions

from the southern Baja peninsula (25°N) to northern Brazil (1°N) where they occupy both

natural and human-modified habitats (Hutto 1980, 1989; Binford 1989; Lynch 1989;

Estrada et al. 1997). Molecular and stable isotope data suggests that eastern and

western populations of Yellow Warblers differ in their overwintering distributions, with

eastern populations occupying South and Central America and western populations

wintering in Mexico and Central America (Boulet et al. 2006). In the regions where we

conducted our study, wintering populations are composed largely of individuals who

breed in northwestern North America and the probability of eastern haplotypes being

present is <0.1 (Boulet et al. 2006).

Several studies indicate that Yellow Warblers hold territories over winter.

Territorial behaviour coupled with a high degree of interspecies aggression has been

documented among individuals using riparian trees bordering pasture in Chiapas,

Mexico (Greenberg et al. 1994; Greenberg and Ortiz 1994), sun-coffee plantations in

Tucurú, Guatemala (Greenberg et al. 1996), and second-growth scrub forest in Panama

17

(Neudorf and Tarof 1998). However, Morton (1976) reports that young Yellow Warblers

in Panama will join wandering mixed-species flocks. There is also evidence from band

returns (n=7; Lowther et al. 1999) and marked populations (Valdez Juarez, unpublished

data) that Yellow Warblers show winter site fidelity.

Study areas

We examined the habitat use of Yellow Warblers within the Pacific and the Gulf

coast regions of the isthmus of Mexico. On the Pacific coast, we worked within and near

the Chamela-Cuixmala Biosphere Reserve, in the state of Jalisco (19° 28' 44" N 105° 2'

38" W; Figure 1.1; hereafter “Chamela”). Chamela is predominantly lowland tropical

deciduous forest (Leguminosae, Euphorbiaceaea, and Sapindacea (Hutto 1994)) with

narrow bands of semi-deciduous tropical forest in riparian areas along the Cuixmala

River and Chamela Arroyo (including Salix gooddingii, Acacia hindsii, Bursera arborea,

and Astianthus viminalis). Coastal regions are predominantly dry scrub (e.g. Prosopis

juliflora) and seasonal saline marshland (Conocarpus erecta and Avicennia germinans)

(Lott 1993). Local agricultural areas (in the communities of Punta Pérula and Emiliano

Zapata) consist mostly of papaya, chili pepper, corn, and tomatoes bordered by shrubby

fence margins or tree rows planted as wind breaks. During the Yellow Warbler wintering

period (September-May) the region has mean temperatures of 24.4 ± 2.4°C and a total

of 380.4 ± 216.7 mm of rainfall. Rainfall is highly seasonal and only 3% of precipitation in

the wintering period falls between February and May (3.3 mm month-1) (UNAM 2007).

On the Gulf of Mexico, we worked within the Los Tuxtlas Biosphere Reserve, on

the Catemaco coast, Veracruz (18° 33' 41"N, 95° 1' 56”W; Figure 1.1; hereafter “Los

Tuxtlas”). Coastal regions in Los Tuxtlas are dominated by short (<4m) vegetation

including Randia laetevirens and Hibiscus tiliaceus. Shoreline vegetation within the

Laguna de Sontecomapan is predominantly red mangrove (Rhizophora mangle). Inland

areas would naturally be lowland tropical moist- and wet-forest (Gutiérrez-García and

Ricker 2011), however, the majority of the region is deforested (80-90% of the landscape

(Estrada et al. 1997)) and the landscape is primarily cattle pasture bounded by live-

fences (lines of Bursera simaruba and Gliricidia sepium trees strung with barbed wire).

Primary forest remains within the boundaries of the protected zone of the reserve and

some patches of second-growth exist near the coastline. Agricultural crops account for

18

approximately 3% of the landscape in Los Tuxtlas (Estrada et al. 1997). Within our study

area, plantings were citrus, corn, beans and chili. During the Yellow Warbler wintering

period, Los Tuxtlas has mean temperatures of 24.3 ± 2.3°C and a total of 2371 ± 137.8

mm of rainfall. As in Chamela, conditions become drier just prior to spring migration,

however, rainfall does not decrease as dramatically. The period between February and

May accounts for 16% of rainfall within the Yellow Warbler wintering period (93.8mm

month-1) (Gutiérrez-García and Ricker 2011).

Habitat searches, point counts and targeted mist-netting

We surveyed for Yellow Warblers in Chamela within a 220-km2 region between

February 2 and February 14, 2011. In Los Tuxtlas, we sought birds within a 120-km2

region between February 25 and March 12, 2011. We conducted surveys within lowland

deciduous forest, riparian semi-deciduous forest, agricultural land, and scrub/seasonal

saline marsh in Chamela, and, in coastal vegetation, red mangrove, primary lowland

tropical forest, second-growth, and cattle pasture in Los Tuxtlas. Agricultural plantings

within our study area in Los Tuxtlas were small (<4 ha) and situated within pasture

habitat. We therefore did not distinguish these areas from pasture during surveys. For

each habitat-type, we sought Yellow Warblers at two to four locations separated by at

least 4 km. Surveys were conducted over one to two hours and consisted of walking a

minimum of 0.5 km through the given habitat, seeking birds by sight. Visual surveys

were interspersed with playback of male song and chip calls (recorded during the

breeding season) followed by silent, stationary observation. Playback attracted wintering

Yellow Warblers in both regions; male and females responded aggressively to the audio

in areas where they were seen foraging.

Where Yellow Warblers were found within a given habitat-type, we subsequently

conducted point counts at two replicate sites in order to determine the density of

individuals and the proportion of adult males present (identifiable on the basis of their

bright plumage). Distances between replicate sites averaged 13.1 km (range: 4.6-28.5

km). Ten counts were done per replicate between 7:00 and 10:00 CST. Points were

placed 200m apart along the length of two parallel, 1km-long transects in broadly

distributed habitat or along one, 2km-long transect in areas where available habitat was

narrowly distributed (riparian sites in Chamela, and one coastal site in Los Tuxtlas).

19

Limited access to private property meant that agricultural point counts in Chamela were

randomly placed (≥ 200m apart) at the edges of fields. Counts consisted of 10 minutes

of silent observation followed by five minutes of male song and chip playback (Sliwa and

Sherry 1992). For each location we noted the distance of observed Yellow Warblers

from the center of the point during both silent and playback periods. Playback increased

our detection of individuals and drew birds closer to the observer. Yellow Warblers

cannot be aged except in the hand, and it is not possible to distinguish between females

and young males by sight. We therefore noted if birds were bright (older males), dull

(young males or females of both age-classes), or “unknown” (when birds were heard but

not seen).

Targeted mist-netting was also conducted within each habitat-replicate in order to

provide an alternative dataset with which to examine sex, age and body size differences

between habitats. Yellow Warblers in Chamela were captured between February 3 and

February 20, 2011 and between January 8 and April 7, 2012. Yellow Warblers in Los

Tuxtlas were captured between February 26 and March 14, 2011. In 2011, 7-10 birds

were captured per habitat-replicate: 52 birds in Chamela and 32 birds in Los Tuxtlas. In

2012, 8-13 additional individuals were captured per habitat-replicate in Chamela (n=52

birds).

Individuals were drawn to the net using playback and decoy. Once extracted,

they were fitted with a Canadian Wildlife Service-issued aluminum band and a unique

combination of three color bands (AC Hughes, UK). We aged and sexed individuals

based on physical and plumage characteristics. Birds hatched in the previous summer

(“second year” or SY) were identified by their pale lower mandibles, tapered primary

coverts and retrices and the limited yellow coloration on the inner webs of their outer tail

relative to older (“after-second year” or ASY) birds (Pyle 1997). Once aged, birds could

be sexed using plumage coloration (brightness and degree of rufous streaking) with

males being the brighter bird within their age-class. Prior to release, we measured the

individual’s tarsus and head-bill length (nearest 0.1mm) and tail and wing chord (9 th

primary) (nearest 1mm). We subsequently defined individual “size” as the first principal

component score (PC1) derived from these metrics (n=133 birds). This score was

normally distributed and described the independent metrics fairly well, capturing 36, 46,

57, and 65% of their variation respectively.

20

13C and 15N Isotope Signatures

Small birds have a whole blood half-life of approximately 4-6 days for 13C

(Hobson and Bairlein 2003; Pearson et al. 2003) and approximately 11 days for 15N

(Hobson and Bairlein 2003; Evans Ogden et al. 2004). Isotopic signatures in blood

should therefore reflect the diet/foraging area of an individual within 12 days prior to

sampling for 13C and within 22 days prior to sampling for 15N (or two half-lives; Hobson

and Clark 1993). To infer habitat use, we collected an average of 30 µl of blood from the

brachial vein of each bird banded in 2011. Blood was cooled in the field then separated

into its components and frozen within 5.5 hours. In the lab, RBCs were dried at 60°C for

3 days. 1 mg (±0.2) of material was then analyzed at the University of California Davis

Stable Isotope Facility in California, USA to determine its 13C and 15N isotope ratios.

Delta values are expressed relative to international standards: V- PDB (Vienna PeeDee

Belemnite) for carbon and Air for nitrogen; the estimated measurement precision was

0.2‰ for 13C and 0.3‰ for 15N. Within-sample replicates indicated that repeatability

was high for both isotopes (13C: r=0.96, F19,1=75.9, P<0.001; 15N: r=1.0, F19,1=129.6,

P<0.001) (Lessells and Boag 1987).

Statistical Analysis

We first used ANOVA to assess whether Yellow Warbler morphology differed

between our study regions. Work by Boulet et al. (2006) indicates that wintering Yellow

Warbler populations on the Pacific and Gulf coast of Mexico differ in their average

breeding latitude. Morphological differences between Chamela and Los Tuxtlas might

therefore be expected as a product of subspecies differences and/or differences in

subspecies mixing. Sex was included in all analyses, as female Yellow Warblers tend to

be smaller than males. Age was included in wing and tail measures as the primaries and

tail feathers of young (SY) birds tend to be shorter than those found in adults. In 2011

we measured 6th primary length in addition to wing chord (9th primary length). The 6th to

9th primary ratio is reported to vary with breeding latitude in Yellow Warblers

(Wiedenfield 1991) and we used this ratio as an additional estimate of regional

differences.

21

We subsequently examined how warbler densities varied with habitat type in

Chamela and Los Tuxtlas. Yellow Warblers were recorded at less than half of point

counts during silent observation (39/100). Playback enabled us to detect individuals at

significantly more points (63/100; one-sided binomial test P<0.001). We therefore

calculated habitat densities using the maximum number of individuals seen or heard

within a 50m radius during both silent and playback periods. We used a generalized

linear mixed model (GLMM; Poisson distribution) to examine habitat-within-region

differences in individuals counted per point, with replicate site as a random factor and

habitat as a fixed effect.

Within each region, we used point count data to calculate the ratio of older males

(bright individuals) to young males and females (dull individuals) within habitat types. We

supplemented these results with capture data where we could assign individuals to a

definitive sex- and age-class (Figure 2.1). Sex- and age-ratio differences between

habitats were assessed using GLMMs: for point count data, the sample replicate was the

point while for capture data the sample replicate was the individual. All models were fit

using a binomial distribution (logit-link) with sampling site as a random effect and habitat

type as a fixed effect. Within each sex we assessed age distributions across habitats.

Finally, a between-region comparison of habitat sex ratios was done using GLMM and

capture data.

We examined how body size (PC1) influenced habitat occupancy within regions

using nominal logistic regression. Our initial analysis for Chamela included age- and sex-

class as covariates plus ‘body size×class’ interaction terms. No significant interactions

were detected (both P>0.70) and the model was rerun using only main effects. For Los

Tuxtlas we did not test interactions as our sample size was small. In our second

analysis, we examined if body size alone influenced habitat occupancy. We predicted

that, if competitive exclusion based on body-size was driving sex distributions, physically

larger birds would be more likely to hold territories in male-dominated habitats. If

physical size alone determines occupancy, body size in our second model would

account for most of the variance explained by sex and age in our first model. In contrast,

if sex- and age-specific traits other than body size (such as aggression, experience, or

innate preference) play a role in habitat use, our second model would explain

substantially less variance than our first.

22

Finally, we examined whether 13C and 15N signatures in Yellow Warbler RBC

samples varied with an individual’s sex, age, or body size, after controlling for habitat

type. Previous work shows that Yellow Warbler 13C and 15N signatures at our study

sites are related to broad-scale habitat use (Drake et al. 2013; Chapter 3). Briefly, 13C is

most enriched among individuals captured in Los Tuxtlas and in agricultural areas in

Chamela. Birds using scrub habitat in Chamela are more 13C depleted, and individuals

in riparian habitat in Chamela have the most 13C-depleted signatures in our dataset.

15N signatures are more enriched in Chamela. Significant differences in 15N exist

between Los Tuxtlas birds, birds using scrub habitat in Chamela, and birds using

agricultural or riparian areas in Chamela. Together 13C and 15N signatures enable

differentiation between habitats used in Chamela and between Chamela and Los

Tuxtlas, but do not allow for the separation of the coastal and pasture habitats used by

birds in Los Tuxtlas (Drake et al. 2013; Chapter 3). We hypothesized that, as habitat-

types are relatively heterogeneous, individuals might show non-random differences in

small scale, microhabitat use within-habitat. Such differences might be reflected in

blood-isotope signatures. To test this, we divided habitats into four, isotopically-distinct

categories: scrub/seasonal saline marsh, agriculture, and riparian in Chamela plus a

merged pasture and coastal habitat category for the Gulf of Mexico (“Los Tuxtlas”). We

then examined small-scale variance in blood-13C and -15N signatures using linear

mixed models (LMM (REML)) with sampling site included as a random effect and habitat

category, sex, age, and body size as fixed effects. We anticipated that heterogeneity

might differ between habitat-types and therefore included habitat interaction terms with

our predictive variables. Non-significant interaction terms (P>0.10) were dropped from

our final model.

All statistical analyses were run using JMP 9.0.2 (JMP 2010). Values reported in

the Results are means ± SD.

23

Results

Habitat Use

In Chamela, Yellow Warblers were detected in riparian semi-deciduous forest,

agricultural areas, and scrub/seasonal saline marsh. They were not detected within the

lowland deciduous forest. In Los Tuxtlas, Yellow Warblers were found in coastal shrub

habitat, cattle pasture, and agricultural plantings. They were not detected within forested

areas (including forested riparian or lake margins) or within mangrove that did not border

(within 50m) on cattle pasture. Playback in forested regions yielded responses from

Wilson’s Warblers (Cardellina pusilla) and American Redstarts (Setophaga ruticilla)

indicating that the dense growth in these areas did not obstruct sound transmission.

Variation in morphology

Yellow Warbler morphology varied with sex and age. Males were physically

larger than females (head-bill: P=0.01, tarsus: P=0.01, wing: P<0.001 tail: P<0.001), and

young birds had truncated plumage (wing: P<0.001, tail: P<0.001). After controlling for

sex and age, population differences existed between our Pacific and Gulf coast

populations (Table 2.1). Birds in Chamela had shorter wing chords (P=0.03) and

possibly shorter tarsi than birds in Los Tuxtlas (P=0.07). Head-bill and tail length did not

differ between regions (P=0.44 and P=0.56 respectively). Wing morphology did not differ

between sex or age classes but did differ between regions. On average, Los Tuxtlas

birds had lower 6th to 9th primary ratios (P6/P9) (F1,81=23.9, P<0.001) and there was a

trend toward less variance in wing shape within this population relative to individuals in

Chamela (Brown-Forsythe test: F1,81=3.10, P=0.08; P6/P9 range, Chamela: 0.93-1.05,

n=51, Los Tuxtlas: 0.98-1.02, n=32).

Abundance and habitat use by sex, age and body size

The number of Yellow Warblers sighted at point counts differed significantly

between habitats in Chamela (χ22=6.8, P=0.03) and in Los Tuxtlas (χ2

1=7.9, P=0.005). In

Chamela, the number of individuals found in riparian and agricultural habitats were

approximately double that found in scrub (2.4 ± 1.0 and 2.3 ± 1.1 versus 1.2 ± 0.8

24

individuals per point; Z=2.8, P=0.005, Z=2.5, P=0.01 respectively). In Los Tuxtlas,

densities were higher in pasture than along the coast (1.6 ± 1.2 versus 0.4 ± 0.8

individuals per count; Z=3.6, P<0.001).

Bright individuals (i.e. older males) were not disproportionally drawn to playback

relative to dull individuals (paired-t99=0.6, P=0.54) and we therefore used total birds

sighted per point to calculate the ratio of bright to dull individuals by habitat. Habitat type

was a significant predictor of the proportion of ASY males present at different point

counts in Chamela (χ22=6.2, P=0.04). Counts conducted in riparian and agricultural

habitats did not differ significantly in their proportion of adult males (Z=0.8, P=0.43) but

scrub habitat had fewer adult males than riparian (Z=2.9, P=0.004) or agricultural habitat

(Z=2.3, P=0.02) (Figure 2.1). Habitat type did not predict the proportion of ASY males at

different point counts in Los Tuxtlas (χ21=0.5, P=0.48).

Capture data mirrored our point count data. Habitat in Chamela was a significant

predictor of the proportion of ASY males captured (χ2 2=6.5, P=0.04). Riparian and

agricultural habitats did not differ in their proportion of adult males (Z=0.25, P=0.80) but

both had more adult males than scrub habitat (agriculture: Z=2.8, P=0.005; riparian:

Z=2.6, P=0.009). As dull individuals could be reliably sexed, we could also compare

overall sex ratios between habitats. These differed significantly by habitat in Chamela

(χ22=10.6, P=0.005): more males were present in riparian habitat than in either

agricultural (Z=2.1, P=0.03) or scrub habitat (Z=4.2, P<0.001) and males were also more

common in agricultural sites than in scrub (Z=2.7, P=0.006). Young (SY) individuals in

Chamela accounted for 34% of birds banded (25% in 2011 and 42% in 2012); age ratios

within sex did not differ significantly between habitats (female age: χ22=2.1, P=0.34, male

age: χ22=4.2 P=0.12) (Figure 2.1).

In Los Tuxtlas capture data did not indicate differences in the proportion of ASY

individuals within coastal and pasture habitats (χ21=2.1, P=0.15). Sex and age-within-sex

ratios also did not differ between habitats (sex: χ21=1.0, P=0.31; male age: χ2

1=0.52,

P=0.47; female age: χ21=0.0, P=0.92). Young individuals in Los Tuxtlas accounted for

30% of the total number of individuals banded (Figure 2.1).

Comparisons of habitat sex ratios between regions using capture data indicated

that males were more common in pasture on the Gulf of Mexico than in scrub habitat on

25

the Pacific coast (Z=-2.1, P=0.03) but less common than in riparian habitat on the Pacific

(Z=1.9, P=0.05). Males were less abundant in coastal habitat on the Gulf of Mexico than

in riparian habitat on the Pacific (Z=2.9, Z=0.004) but did not differ significantly from

other Pacific habitat types.

Body size was not a determinant of habitat use in Chamela after controlling for

sex biases in habitat use (whole model: χ26=29.0, P<0.001, n=101; sex: χ2

2=24.2,

P<0.001, age: χ22=2.0, P=0.36, body size: χ2

2=1.2, P=0.55). Body size alone was also

not a significant predictor of habitat use (χ22=2.4, P=0.31, n=101). Together these

models indicate that (1) sex distributions across habitats in Chamela were not explained

by size differences between males and females and (2) that physical size was poor

predictor of habitat use.

In Los Tuxtlas body size did not predict habitat use (whole model: χ23=2.2,

P=0.53, n=32; sex: χ21=0.0, P=0.85, age: χ2

1=0.6, P=0.44, body size: χ21=0.4, P=0.50;

reduced model (body size): χ21=1.6, P=0.21, n=32).

Within-habitat variation in 13C and 15N signatures

Much of the variance in 13C and 15N was explained by the habitat-type used by

an individual (13C whole model: r2=0.65, n=81; habitat: F=9.1, P=0.01; 15N whole

model: r2=0.84, n=81; habitat: F=101.2, P<0.001; see also Drake et al. 2013; Chapter 3).

After accounting for this variation, neither sex, age, nor body size explained additional

variance in blood-13C values (sex: F=2.1, P=0.15; age: F=0.6, P=0.44; body-size:

F=0.2, P=0.64). 15N signatures also did not vary with sex (F=0.0, P=0.92) or age

(F=1.5, P=0.23). However, body size was correlated with 15N signatures in two habitats

(body size: F=0.0, P=0.84; body size×habitat: F=4.6, P=0.005). Physically larger birds in

riparian habitat tended to have more depleted 15N signatures than smaller birds while,

in scrub habitat, the pattern was reversed: larger birds were 15N enriched (Figure

2.2b,c).

26

Discussion

Yellow Warblers are reported to use a diversity of natural and anthropogenically-

modified habitats over winter (Dunn and Garrett 1997). Previous work on the Pacific

slope of Mexico indicates that this species makes use of six of eight broad vegetation

zones found within the lowland areas of the region (Hutto 1980). They have not been

observed within tropical deciduous forest or as members of mixed-species insectivorous

flocks that use the forest canopy during the non-breeding season (Hutto 1980, 1989,

1994). On the Gulf of Mexico, diversity surveys conducted across habitats in Los Tuxtlas

captured the majority of Yellow Warblers (n=102) within citrus plantations and live-

fences (30% and 35% of captures) and the minority within mixed crops and forested

areas (4% and 6% of captures) (Estrada et al. 1997). Our habitat surveys in Chamela

and Los Tuxtlas were not exhaustive and multi-year data is needed to assess changes

in occupancy patterns associated with variation in climate and bird abundance between

years. Despite this fact, our results are consistent with the above studies. We found

Yellow Warblers in natural and man-made landscapes but they were largely absent from

the dominant (or historically dominant) forest cover in both regions. Birds in our study

foraged in defined areas and were attracted by and aggressive toward call playback,

suggesting territoriality. Contrary to more southerly observations by Morton (1976), this

behaviour extended to the 45 young birds we captured.

Sex-biases in habitat use are a common phenomenon among territorial

overwintering wood-warblers (Stutchbury et al. 2005). Two previous studies that

assessed sexual habitat segregation in Yellow Warblers, however, differed in their

findings. Greenberg et al. (1997) conducted surveys in the Ocosingo valley, Chiapas,

Mexico that indicated that Yellow Warblers in gallery forest along arroyos (equivalent to

what we define as riparian habitat) were predominately male (83%; n=12), while birds in

drier acacia plantations were predominantly female (76%; n=23). As Greenberg’s were

survey results, it is likely that the observed males in gallery forest were older individuals

and that birds identified as females in acacia plantations were actually a mixture of

females and young males that could only be distinguished in the hand. In contrast to the

sexual habitat segregation found in Chiapas, Lopez-Ornat and Greenberg (1990) found

an even distribution of sexes among Yellow Warblers captured in scrub and open

habitats in the Sian Ka’an Biosphere Reserve, Quinta Roo, Mexico (n=62). Yellow

27

Warblers in our study showed regional differences in how sexes were distributed in the

landscape. Birds in Chamela exhibited sexual habitat segregation. Riparian areas were

predominately occupied by males (92% of individuals), while almost two thirds of

occupants of coastal scrub/seasonal saline marsh areas were female (63%). Agricultural

areas were also male dominated (71%) but were not as severely skewed as riparian

habitat. Of the Yellow Warblers we banded in scrub, 79% would be visually identified as

dull-coloured, suggesting that the distributions we observed are very similar to those

found acacia plantations in Chiapas (Greenberg et al. 1997). In Los Tuxtlas, we found no

significant differences in sex distributions between pasture and coastal habitats. Males

made up 71% and 53% of the individuals captured in in these habitats respectively. Our

sample size in Los Tuxtlas was relatively small (n=32). This may have limited our ability

to detect sex-differences between habitat types on the Gulf coast. Between-region

comparisons indicate that a sex-skew in pasture habitat, if present, is more moderate

than observed in riparian habitat on the Pacific coast. This could support within-region

results (i.e. reduced habitat biases in Los Tuxtlas) but it may also reflect differences in

regional sex ratios and/or population density.

Climate differences between study locations could offer a possible explanation

for differences in the degree of sexual habitat segregation between regions. The Pacific

slope of Mexico, where Chamela is located, has a highly seasonal climate (García-Oliva

et al. 1991). Low rainfall would be expected to produce sharp quality differences

between habitat-types at the end of the wintering period (Studds and Marra 2007). As

male-dominated riparian corridors are the only areas in Chamela with continuous access

to fresh water and as these corridors make up a very small percentage of the landscape,

we might expect greater competition for access. In contrast, the more moderate climate

of the southern Gulf coast might be expected to reduce quality differences between

habitats in the region. Our inability to identify habitat use by 13C or 15N in Los Tuxtlas

might further suggest that these habitats are qualitatively similar (Drake et al. 2013;

Chapter 3). Differences in how female and male Yellow Warblers distribute themselves

within the Ocosingo valley (Greenberg et al. 1997) and Sian Ka’an Biosphere Reserve

(Lopez Ornat and Greenberg 1990) are consistent with a climate explanation. The

Ocosingo valley, like Chamela, receives very little rainfall in the spring (21mm month-1

(February-May)) while Sian Ka’an falls within the same climate zone as Los Tuxtlas

(Kottek et al. 2006) and receives more rain than either Ocosingo or Chamela (although

28

less than Los Tuxtlas) in the same period (45mm month-1 (February-May)). Increased

competition in dry wintering regions would also be consistent with the reported latitudinal

segregation of Yellow Warblers on the wintering grounds (Komar et al. 2005). The

northern regions of the Yellow Warbler winter range have seasonal climates with low

mean annual precipitation (Legates and Willmott 1990; Kottek et al. 2006) and males

may exclude females to a greater degree from good quality habitat in these areas.

Females may, as a result, experience lower mortality if they overwinter in the wetter

southern and Caribbean-coast regions.

We predicted that, if sexual habitat segregation within habitats in Chamela was a

product of competition, body size might be a determinant of habitat occupancy. One

source of competitive asymmetry between sexes is size, as males tend to be physically

larger than females. In American Redstarts, habitat occupancy was a significant

predictor of female size: individuals in high quality (mangrove) habitat tended to be

larger than those in lower quality (scrub) habitat (Marra 2000). However, we found no

evidence that body size predicted habitat occupancy or explained the differences in

habitat use between sexes in Chamela. Age effects were also absent in this model

suggesting that experience or prior occupancy was not a determining factor in habitat

use. Our data therefore does not allow us to rule out the possibility that Yellow Warbler

distributions in Chamela are the product of sex-specific preference. It is also possible

that sex biases within habitats are due to competitive asymmetries based on factors

other than size or experience (Arizaga and Bairlein 2011). For example, males may be

physiologically more capable than females of expressing aggressive behaviours during

territory formation. Male American Redstarts are more aggressive than females in

response to simulated territorial intrusions (Marra 2000). Similarly, wintering

observations indicate that male Cape May Warblers (Setophaga tigrina) are more

aggressive than females when defending resources (Latta et al. 2001). In (breeding)

male Song Sparrows (Melospiza melodia), the presence of challengers triggers

increased testosterone secretion, a process that Wingfield et al. (1987) suggests could

facilitate the expression of aggression over an extended timeframe. Females, with lower

testosterone production (Møller et al. 2005), may not be able to mount a similar

response and may therefore be more likely to lose contests through attrition.

29

Broad-scale habitat use in wintering Yellow Warblers can be partially

differentiated using 13C and 15N signatures in tissue samples (Drake et al. 2013;

Chapter 3). After accounting for this source of variance, we investigated whether fine-

scale differences in 13C or 15N signatures might be non-random in relation to sex, age

or body size. 13C signatures did not suggest additional differences in habitat use

between classes or between large and small individuals once broad-scale habitat use

was accounted for. There was, however, an apparent relationship between 15N and

body size in riparian and scrub habitats in Chamela. These trends opposed each other

and explained 31% of the variation in 15N within riparian corridors and 21% of the

variation in 15N within scrub habitat. Increasing 15N with increasing body size was

evident in both females and males in scrub habitat, indicating that the relationship was

not the product of more subtle sex-differences in habitat use (Figure 2.2c). In

investigating this pattern, we found that birds with similar 15N signatures in scrub (but

not riparian) habitat were clustered geographically, suggesting that scrub habitat was

patchy. Larger, 15N enriched birds tended to border on higher vegetation mixed with

trees while smaller, 15N depleted birds tended to be in low-lying vegetation. Differences

in 15N could indicate that the mean trophic level of prey items consumed by Yellow

Warblers differed between these vegetation-types or that differences in plant-level 15N

exist between patches (Post 2002). Plant-level 15N differences could occur if high-

vegetation patches contain more deeply rooted species than shrubby areas (Hyodo et al.

2010). Foraging behaviour might also produce a relationship between 15N and body

size if differences in foraging strategy meant individuals consumed prey from different

trophic levels. Foraging behaviours such as hover-picking are more common in smaller,

more agile species (Greenberg 1979; Forstmeier and Keßler 2001) and might also be

more common among smaller individuals within-species (Nyström 1991). Identical shifts

foraging strategies could produce opposing trends where prey communities differ. While

our results remain ambiguous without supporting behavioural and insect sampling data,

they do indicate that blood-15N data can be used as a tool for identifying potential

differences in microhabitat use in passerines, once broad habitat variables are taken into

account.

Point count data indicate that Yellow Warblers density varied between habitats

and between regions with the greatest number of individuals per count sighted in riparian

30

and agricultural habitat in Chamela. Scrub in Chamela, and pasture and coastal areas in

Los Tuxtlas all had low densities. Our point counts were conducted primarily to

determine age and sex distributions. As density estimates, these results are crude. The

use of playback means that the number of individuals per count could reflect real

differences in bird number, but also behavioural differences between habitats. For

example, riparian and agricultural birds in Chamela may be more responsive to

perceived territorial intrusions than birds in scrub or in Los Tuxtlas (Silwa and Sherry

1994). Estimates of habitat-specific territory size are needed to achieve better habitat-

density estimates for this species.

Genetic and isotopic work indicate that Yellow Warblers who breed in the

Northwest Territories and Yukon are more common on the Gulf coast of Mexico than the

Pacific coast during the winter (Boulet et al. 2006). In addition to northern breeders,

Pacific wintering populations contain individuals from southern Canada and from the

USA (Boulet et al. 2006). Where we found morphology differences between our study

regions, they were in agreement with these patterns. Birds in Los Tuxtlas tended to have

longer wing chords and more pointed wing morphology (Table 2.1), which is suggestive

of more northern subspecies (Wiedenfeld 1991). Similarly, higher average values (i.e.

more rounded wing morphology) and possibly greater variance in P6/P9 measures in

Chamela are consistent with the presence of southern breeders and a wider range of

breeding origins among Pacific birds.

Our data suggest that Yellow Warblers exhibit regional differences in sexual

habitat segregation on their wintering grounds in Mexico. Differences in habitat use by

sex were evident on the Pacific but not detected on the Gulf coast. Where sexual habitat

segregation did occur, males were most common in mesic, forested habitats while

females were most common in drier, open habitats. This type of distribution is consistent

with patterns reported in other wintering warbler species (Stutchbury et al. 2005). Further

work is needed to determine the mechanism by which sex-distributions in Yellow

Warblers are produced. If it is a response to regional climate conditions as we propose,

and if differences in habitat quality and competitive asymmetry are responsible for

female- and male-skewed habitats on the drier Pacific coast (this study) and in the state

of Chiapas (Greenberg et al. 1997), then we would predict from climate maps (Kottek et

al. 2006) that segregation occurs over approximately 40% of the Yellow Warbler

31

wintering range. However, dry habitats account for approximately 80% of the region

occupied by Yellow Warblers within Mexico. As western breeding populations

predominate in Mexico, sexual habitat segregation on the wintering grounds could be

much more common phenomenon in this lineage. This warrants further investigation, as

such differences could contribute to demographic and population-response differences

between eastern and western Yellow Warbler populations.

References

Arizaga, J., and F. Bairlein. 2011. The role of age, sex, subspecies, body size and fuel load in determining the social status of a migratory passerine during the non-breeding period. Ornis Fennica 88:154–160.

Bennett, S. E. 1980. Interspecific competition and the niche of the American Redstart (Setophaga ruticilla) in wintering and breeding communities. Pages 319–335 in

A. Keast and E. S. Morton, editors. Migrant birds in the neotropics: ecology, behaviour, distribution, and conservation. Smithsonian Institution Press, Washington, D. C.

Bigglestone, H. C. 1913. A Study of the Nesting Behavior of the Yellow Warbler (Dendroica æstiva æstiva). The Wilson Bulletin 25:49–67. doi: 10.2307/4154476.

Binford, L. C. 1989. A distributional survey of the birds of the Mexican state of Oaxaca. Ornithological monographs 43:392–418. doi: 10.2307/40167673

Boulet, M., H. L. Gibbs, and K. A. Hobson. 2006. Integrated Analysis of Genetic, Stable Isotope, and Banding Data Reveal Migratory Connectivity and Flyways in the Northern Yellow Warbler (Dendroica petechia; Aestiva group). Ornithological

Monographs:29–78. doi:10.2307/40166837

Brown, D. R., and J. A. Long. 2007. What isa Winter Floater? Causes, Consequences, and Implications for Habitat Selection. The Condor 109:548–565. doi: 10.2307/4500991.

Drake, A., C. Rock, S. P. Quinlan, and D. J. Green. 2013. Carry-over effects of winter habitat vary with age and sex in Yellow Warblers (Setophaga petechia). Journal of Avian Biology, in press. doi: 10.1111/j.1600-048X.2013.05828.x.

Dunn, J., and K. Garrett. 1997. A field guide to warblers of North America. Houghton Mifflin Harcourt.

Estrada, A., R. Coates-Estrada, and D. A. J. Meritt. 1997. Anthropogenic landscape changes and avian diversity at Los Tuxtlas, Mexico. Biodiversity and Conservation 6(1):19-43.

32

Evans Ogden, L. J., K. A. Hobson, and D. B. Lank. 2004. Blood isotopic (δ13C and δ15N) turnover and diet-tissue fractionation factors in captive Dunlin (Calidris alpina pacifica). The Auk 121:170–177. doi: 10.1642/0004-

8038(2004)121[0170:BICANT]2.0.CO;2.

Faaborg, J., R. T. Holmes, A. D. Anders, K. L. Bildstein, K. M. Dugger, S. A. Gauthreaux, P. Heglund, K. A. Hobson, A. E. Jahn, D. H. Johnson, S. C. Latta, D. J. Levfy, P. P. Maria, C. L. Merkord, E. Nol, S. I. Rothstein, T. W. Sherry, T. S. Sillett, F. R. Thompson, and N. Warnock. 2010. Recent advances in understanding migration systems of New World land birds. Ecological monographs 80:3–48. doi:10.1890/09-0395.1

Forstmeier, W., and A. Keßler. 2001. Morphology and foraging behaviour of Siberian Phylloscopus warblers. Journal of Avian Biology 32:127–138. doi:

10.1034/j.1600-048X.2001.320205.x.

Fujio Hyodo, Ayato Kohzu, and Ichiro Tayasu. 2010. Linking aboveground and belowground food webs through carbon and nitrogen stable isotope analyses. Ecological Research 25:745–756. doi: 10.1007/s11284-010-0719-x.

García-Oliva, F., E. Ezcurra, and L. Galicia. 1991. Pattern of rainfall distribution in the central pacific coast of Mexico. Geografiska Annaler. Series A, Physical Geography 73:179–186. doi: 10.2307/521023.

Greenberg, R. 1979. Body Size, breeding habitat, and winter exploitation systems in Dendroica. The Auk 96:756–766. doi: 10.2307/4085662.

Greenberg, R. 1986. Competition in migrant birds in the nonbreeding season. Current Ornithology 3:281–307. doi:10.1007/978-1-4615-6784-4_6

Greenberg, R., P. Bichier, and J. Sterling. 1997. Acacia, cattle and migratory birds in southeastern Mexico. Biological Conservation 80:235–247. doi: 10.1016/S0006-3207(96)00137-1

Greenberg, R., and J. S. Ortiz. 1994. Interspecific defense of pasture trees by wintering Yellow Warblers. The Auk 111:672–682.

Greenberg, R., J. S. Ortiz, and C. M. Caballero. 1994. Aggressive competition for critical resources among migratory birds in the Neotropics. Bird Conservation International 4:115–127. doi: 10.1017/S0959270900002719.

Greenberg, R., R. Reitsma, and A. C. Angon. 1996. Interspecific aggression by Yellow Warblers in a sun coffee plantation. The Condor 98:640–642. doi: 10.2307/1369579.

Gutiérrez-García, G., and M. Ricker. 2011. Climate and climate change in the region of Los Tuxtlas (Veracruz, Mexico): A statistical analysis. Atmosfera 24(4): 347-373.

33

Hobson, K. A., and R. G. Clark. 1993. Turnover of 13 C in Cellular and Plasma Fractions of Blood: Implications for Nondestructive Sampling in Avian Dietary Studies. The Auk 110:638–641. doi: 10.2307/4088430.

Hobson K.A., and Bairlein F. 2003. Isotopic fractionation and turnover in captive Garden Warblers (Sylvia borin): implications for delineating dietary and migratory associations in wild passerines. Canadian Journal of Zoology 81:1630–1635. doi: 10.1139/z03-140

Holmes, R. T., T. W. Sherry, and L. Reitsma. 1989. Population structure, territoriality and overwinter survival of two migrant Warbler species in Jamaica. The Condor 91:545–561. doi: 10.2307/1368105

Holmgren, N., and S. Lundberg. 1993. Despotic Behaviour and the Evolution of Migration Patterns in Birds. Ornis Scandinavica 24:103–109. doi: 10.2307/3676359.

Hutto, R. L. 1980. Winter habitat distribution of migratory land birds in western Mexico, with special reference to small foliage-gleaning insectivores. Page 576 in A. Keast and E. S. Morton, editors. Migrant Birds in the Neotropics: Ecology, Behaviour, Distribution and Conservation. Smithsonian Institution Press, Washington, D.C.

Hutto, R. L. 1989. The Effect of Habitat Alteration on Migratory Land Birds in a West Mexican Tropical Deciduous Forest: A Conservation Perspective. Conservation Biology 3:138–148. doi: 10.1111/j.1523-1739.1989.tb00066.x.

Hutto, R. L. 1994. The Composition and Social Organization of Mixed-Species Flocks in a Tropical Deciduous Forest in Western Mexico. The Condor 96:105–118. doi: 10.2307/1369068.

JMP. 2010. Version 9.0.2. SAS Institute Inc.

Komar, O., B. J. O’Shea, A. T. Peterson, and A. G. Navarro-Sigüenza. 2005. Evidence of latitudinal sexual segregation among migratory birds wintering in Mexico. The Auk 122:938–948. doi: 10.1642/0004-8038(2005)122[0938:EOLSSA]2.0.CO;2

Kottek, M., J. Grieser, C. Beck, B. Rudolf, and F. Rubel. 2006. World Map of the K ppen-Geiger climate classification updated. Meteorologische Zeitschrift 15:259–263. doi: 10.1127/0941-2948/2006/0130.

Latta, S. C., and J. Faaborg. 2002. Demographic and population responses of Cape May Warblers wintering in multiple habitats Ecology 83:2502–2515. doi:10.2307/3071810

Latta, S. C., H. A. Gamper, and J. R. Tietz. 2001. Revising the Convergence Hypothesis of Avian Use of Honeydew: Evidence from Dominican Subtropical Dry Forest. Oikos 93:250–259. doi: 10.2307/3547303.

34

Legates, D. R., and C. J. Willmott. 1990. Mean seasonal and spatial variability in gauge-corrected, global precipitation. International Journal of Climatology 10:111–127. doi: 10.1002/joc.3370100202.

Lessells, C. M., and P. T. Boag. 1987. Unrepeatable Repeatabilities: A Common Mistake. The Auk 104:116–121. doi: 10.2307/4087240.

Lott, E. J. 1993. Annotated checklist of the vascular flora of the Chamela Bay region, Jalisco, Mexico. Occasional Papers of the California Academy of Sciences 148:60 pp.

Lynch, J. F. 1989. Distribution of Overwintering Nearctic Migrants in the Yucatan Peninsula, I: General Patterns of Occurrence. The Condor 91:515–544. doi: 10.2307/1368104.

Marra, P. P. 2000. The role of behavioural dominance in structuring patterns of habitat occupancy in a migrant bird during the nonbreeding season. Behavioral Ecology 11:299–308. doi: 10.1093/beheco/11.3.299.

Marra, P. P., and R. T. Holmes. 2001. Consequences of Dominance-Mediated Habitat Segregation in American Redstarts during the Nonbreeding Season. The Auk 118:92–104. doi: 10.2307/4089760

Marra, P. P., T. W. Sherry, and R. T. Holmes. 1993. Territorial Exclusion by a Long-Distance Migrant Warbler in Jamaica: A Removal Experiment with American Redstarts (Setophaga ruticilla). The Auk 110:565–572. doi:10.2307/4088420

Møller, A. P., L. Z. Garamszegi, D. Gil, S. Hurtrez-Boussès, and M. Eens. 2005. Correlated evolution of male and female testosterone profiles in birds and its consequences. Behavioral Ecology and Sociobiology 58:534–544. doi: 10.1007/s00265-005-0962-2.

Morton, E. S. 1976. The Adaptive Significance of Dull Coloration in Yellow Warblers. The Condor 78:423. doi: 10.2307/1367713.

Morton, E. S. 1990. Habitat Segregation by Sex in the Hooded Warbler: Experiments on Proximate Causation and Discussion of Its Evolution. The American Naturalist 135:319–333. doi:10.2307/2462249.

Morton, E. S., M. Van Der Voort, and R. Greenberg. 1993. How a warbler chooses its habitat : field support for laboratory experiments. Animal Behaviour 46:47–53. doi:10.1006/anbe.1993.1160.

Morton, E. S., J. F. Lynch, K. Young, and P. Mehlhop. 1987. Do male Hooded Warblers exclude females from nonbreeding territories in tropical forest? The Auk 104:133–135. doi: 10.2307/4087245.

35

Morton, E. S., and B. J. M. Stutchbury. 2005. The Significance of Mating System and Nonbreeding Behavior to Population and Forest Patch Use by Migrant Birds. Pages 285–289. in C.J. Ralph and T. D. Rich, editors. Proceedings of the 3rd

International Partners in Flight Conference, 2002, Asilomar, CA.

Neudorf, D. L., and S. A. Tarof. 1998. The Role of Chip Calls in Winter Territoriality of Yellow Warblers (El rol de Las llamadas “Chip” en la Territorialidad Invernal de Dendroica petechia). Journal of Field Ornithology 69:30–36.

Nyström, K. G. Kenneth. 1991. On sex-specific foraging behaviour in the Willow Warbler, Phylloscopus trochilus. Canadian Journal of Zoology 69:462–470.

doi:10.1139/z91-072

Ornat, A. L., and R. Greenberg. 1990. Sexual Segregation by Habitat in Migratory Warblers in Quintana Roo, Mexico. The Auk 107:539–543. doi: 10.2307/4087916.

Parrish, J. D., and T. W. Sherry. 1994. Sexual habitat segregation by american redstarts wintering in Jamaica: importance of resource seasonality. The Auk 111:38–49. doi:10.2307/4088503.

Pearson, S. F., D. J. Levey, C. H. Greenberg, and C. M. del Rio. 2003. Effects of elemental composition on the incorporation of dietary nitrogen and carbon isotopic signatures in an omnivorous songbird. Oecologia 135:516–523. doi: 10.1007/s00442-003-1221-8.

Pyle, P. 1997. Identification Guide to North American Birds, Part I: Columbidae to Ploceidae. Slate Creek Press.

Quinlan, S. P., and D. J. Green. 2012. Riparian habitat disturbed by reservoir management does not function as an ecological trap for the Yellow Warbler (Setophaga petechia). Canadian Journal of Zoology 90:320–328. doi: 10.1139/z11-138.

Rappole, J. H., D. I. King, and W. C. Barrow Jr. 1999. Winter Ecology of the Endangered Golden-Cheeked Warbler. The Condor 101:762–770. doi: 10.2307/1370063.

Rich, T.D., C. J. Beardmore, H. Berlanga, P. J. Blancher, M. S. W. Bradstreet, G. S. Butcher, D. W. Demarest, E. H. Dunn, W. C. Hunter, E. E. Iñigo-Elias, J. A. Kennedy, A. M. Martell, O. Panjabi, D. N. Pashley, K. V. Rosenberg, C. M. Rustay, J. S. Wendt, and T.C. Will. 2004. Partners in Flight North American Landbird Conservation Plan. Cornell Lab of Ornithology. Ithaca, NY. Retrieved from http://www.partnersinflight.org/cont_plan/.

Sherry, T. W., and R. T. Holmes. 1995. Summer versus winter limitation of populations: Conceptual issues and evidence. Pages 85–120 in T. E. Martin and D. M. Finch,

editors. Ecology and Management of Neotropical Migratory Birds: A Synthesis and Review of Critical Issues. Oxford University Press, New York.

36

Sliwa, A., and T. W. Sherry. 1992. Surveying Wintering Warbler Populations in Jamaica: Point Counts with and without Broadcast Vocalizations. The Condor 94:924–936. doi: 10.2307/1369289.

Smith, J. A. M., L. R. Reitsma, and P. P. Marra. 2011. Multiple Space-Use Strategies and Their Divergent Consequences in a Nonbreeding Migratory Bird (Parkesia noveboracensis). The Auk 128:53–60. doi: 10.1525/auk.2011.10241.

Steifetten, O., and S. Dale. 2006. Viability of an endangered population of ortolan buntings : The effect of a skewed operational sex ratio. Biological conservation 132:88–97. doi:10.1016/j.biocon.2006.03.016

Studds, C. E., and P. P. Marra. 2005. Nonbreeding Habitat Occupancy and Population Processes: An Upgrade Experiment with a Migratory Bird. Ecology 86:2380–2385. doi:10.1890/04-1145

Studds, C. E., and P. P. Marra. 2007. Linking fluctuations in rainfall to nonbreeding season performance in a long-distance migratory bird, Setophaga ruticilla.

Climate Research 35:115–122. doi:10.3354/cr00718

Stutchbury, B. J. M., E. S. Morton, and T. E. Pitcher. 2005. Sex roles in migrants: extra-pair mating systems and winter social systems. Pages 307–320 in R. Greenberg

and P. P. Marra, editors. Birds of two worlds: the ecology and evolution of migration. Johns Hopkins University Press, Baltimore, Maryland, USA.

Townsend, J. M., C. C. Rimmer, and K. P. McFarland. 2009. Elucidating the limiting factors of a rare, vulnerable species: Bicknell’s Thrush. Pages 91–95 in T. D. Rich, C. Arizmendi, D. Demarest, and C. Thompson, editors. Proceedings of the 4th International Partners in Flight Conference. McAllen, Texas, USA.

UNAM. 2007. Median Temperature Data: Estación de Biología Chamela. Retrieved from www.ibiologia.unam.mx/ebchamela/www/tmedia.html.

Wiedenfeld, D. A. 1991. Geographical Morphology of Male Yellow Warblers. The Condor 93:712–723. doi: 10.2307/1368203.

Wingfield, J. C., G. F. Ball, A. M. Dufty, R. E. Hegner, and M. Ramenofsky. 1987. Testosterone and Aggression in Birds. American Scientist 75:602–608.

Wunderle, J. M. 1995. Population Characteristics of Black-Throated Blue Warblers Wintering in Three Sites on Puerto Rico. The Auk 112:931–946. doi: 10.2307/4089024.

37

Tables and Figures

Table 2.1 Morphometric traits (± SD) of Yellow Warblers wintering on the Pacific and Gulf coast of Mexico. Sample sizes are reported in brackets. The length of the 6th primary feather was measured in 2011 only and therefore the sample sizes for wing shape (P6/P9) are smaller than other metrics.

Morphometrics Pacific coast (Chamela-Cuixmala) Gulf coast (Los Tuxtlas)

Sex ASY SY ASY SY

Wing Chord (mm) Female 59.2 ± 2.2 (20) 57.6 ± 1.4 (13) 59.9 ± 1.6 (7) 58.8 ± 1.3 (5)

Male 62.0 ± 2.1 (49) 60.2 ± 1.7 (20) 62.5 ± 1.6 (15) 62.2 ± 1.1 (5)

Tail (mm) Female 42.4 ± 1.8 (20) 41.8 ± 1.0 (13) 42.7 ± 1.2 (7) 40.6 ± 2.3 (5)

Male 44.8 ± 2.0 (48) 42.8 ± 2.5 (21) 44.0 ± 1.6 (15) 44.2 ± 1.9 (5)

Wing shape (P6/P9) Female 1.02 ± 0.02 (12) 1.00 ± 0.03 (5) 1.00 ± 0.01 (7) 1.00 ± 0.01 (5)

Male 1.01 ± 0.02 (27) 1.02 ± 0.02 (7) 0.99 ± 0.01 (15) 0.99 ± 0.01 (5)

Head- bill (mm) Female 28.6 ± 0.6 (20) 28.4 ± 0.6 (13) 28.7 ± 0.4 (7) 28.6 ± 0.4 (5)

Male 28.8 ± 0.6 (49) 28.8 ± 0.6 (22) 28.9 ± 0.6 (15) 28.9 ± 0.5 (5)

Tarsus (mm) Female 18.8 ± 0.5 (20) 18.8 ± 0.5 (13) 18.7 ± 0.6 (7) 19.1 ± 0.5 (5)

Male 19.0 ± 0.6 (49) 19.0 ± 0.6 (22) 19.3 ± 0.6 (15) 19.2 ± 0.7 (5)

Mass (g) Female 8.2 ± 0.4 (17) 8.1 ± 0.4 (13) 8.7 ± 0.3 (7) 8.8 ± 0.4 (5)

Male 8.5 ± 0.5 (45) 8.3 ± 0.4 (21) 9.1 ± 0.5 (15) 9.6 ± 0.3 (5)

38

Figure 2.1 Patterns of habitat segregation in Yellow Warblers on the (a) Pacific coast (“Chamela”) and (b) Gulf coast (“Los Tuxtlas”) of Mexico. Left columns represent point count results with individuals identified as older (ASY) males (solid dark grey) or ‘other’ (solid light grey). Right columns represent banding data where the ‘other’ category is separated into older females (solid white), yearling (SY) females (stippled), and yearling males (grey, diagonal lines). Sample sizes are indicated above each column. In Chamela, riparian habitat contained more males than agriculture and both of these habitats contained more males than scrub. In Los Tuxtlas, coastal and pasture habitats did not significantly differ in the proportion of males present.

39

Figure 2.2 Variation in Yellow Warbler blood-15N signatures within habitats in relation to body size. Habitat-types were classified as (a) agricultural habitat, Pacific coast, (b) riparian habitat, Pacific coast, (c) scrub/seasonal saline marsh, Pacific coast and (d) “Los Tuxtlas”, Gulf of Mexico. This last category merges data from birds using coastal and pasture habitats on the Gulf of Mexico as habitats were not isotopically distinct. Unfilled circles indicate females, filled circles indicate males.

40

Chapter 3. Carry-over effects of winter habitat vary with age and sex in Yellow Warblers (Setophaga petechia)

Anna Drake, Christine Rock, Sam P. Quinlan, David J. Green2

Abstract

We use stable isotope data to investigate the role of winter habitat use in altering

the breeding phenology of Yellow Warblers (Setophaga petechia). We first confirm that

13C and 15N isotopic signatures vary with winter habitat use in this species. We then

examine the relationship between winter habitat use, breeding phenology and

productivity within four age-sex-classes, since life history theory would predict that carry-

over effects should vary with age and gender. The 13C signatures of Yellow Warblers

using riparian habitats over winter were more depleted than the signatures of those

using agricultural or scrub habitat. Individuals on the Pacific coast of Mexico were also

more 15N enriched than those on the southern Gulf of Mexico. 13C and 15N signatures

were only correlated with earlier clutch initiation and subsequent higher productivity in

first-breeding-season females. We estimate that shifts in 13C equivalent to a shift from

scrub to riparian winter habitat would be associated with the production of 0.8 more

fledglings by yearling females. Pre-breeding events that influence the timing of breeding

could also influence the reproductive performance of older males and females, but we

found little evidence that winter habitat use influenced breeding season phenology in

these birds. 2 A version of this chapter is published in the Journal of Avian Biology doi: 10.1111/j.1600-

048X.2013.05828.x.

41

Introduction

Carry-over effects are defined as events and/or processes that occur in one

period of the annual cycle and affect an individual’s fitness in a subsequent period due to

the individual making the transition between periods in a different state (Harrison et al.

2011). Carry-over effects in migratory birds, particularly those driven by winter habitat

use and affecting subsequent breeding productivity, have received considerable

attention over the past decade. To date, such effects have been documented in four

migratory bird species and indicated in a fifth: Black-tailed Godwits (Limosa limosa

islandica; Gill et al. 2001; Gunnarsson et al. 2005, Gunnarsson et al. 2006) and Barn

Swallows (Hirundo rustica; Saino et al. 2004) from populations breeding in Europe, and

American Redstarts (Setophaga ruticilla; Marra et al. 1998, Norris et al. 2004, Reudink et

al. 2009, Tonra et al. 2011), Kirkland’s Warblers (Setophaga kirtlandii; Rockwell et al.

2012), and potentially Black-throated Blue Warblers (Setophaga caerulescens; Bearhop

et al. 2004) from populations breeding in eastern North America. These studies indicate

that loss, degradation, or climate-driven variation in the quality of winter habitat could

influence migratory populations beyond simply impacting over-winter survival. However,

few of these studies included age as a factor in their models: three limited their data to

older individuals (Norris et al. 2004; Reudink et al. 2009, Tonra et al. 2011) and five

combined age-classes in their analysis (Marra et al. 1998; Gill et al. 2001; Norris et al.

2004; Gunnarsson et al. 2005; Gunnarsson et al. 2006).

Life history theory would suggest that the strength of carry-over effects should

vary with both age- and sex-class. This is because the ultimate effect of state on an

individual’s seasonal fitness will differ depending on the costs and constraints faced by

that individual (as seen in staging Brent Geese (Branta bernicla bernicla); Ebbinge and

Spaans 1995). As different age- and sex-classes are likely to vary predictably in the

costs and constraints they face (Brown and Roth 2002), it follows that the presence and

magnitude of carry-over effects may also vary predictably between classes within a

population.

Habitat-driven carry-over effects act through non-fatal differences in the physical

condition of individuals. Such differences may have a direct impact on fitness or they

may result in variance in biologically relevant events such as onset of migration, the date

42

at which individuals arrive on the breeding grounds and establish a territory, the timing of

breeding, and/or the timing of moult. In the latter case, it is the variance in these events

that ultimately impact fitness and can be considered the mechanism through which

carry-over effects act. In this paper we use the term “time-mediated” carry-over effect, to

distinguish this mechanism from where physical condition directly impacts fitness.

Theoretically, time-mediated carry-over effects resulting from winter habitat use

should be more evident in males than in females. The timing of arrival on the breeding

grounds typically has a stronger impact on male breeding success: arrival-date-

dependent changes in within-pair and extra-pair paternity, polygyny, and greater odds of

failing to acquire a territory or mate (and any reproductive success) mean that males

may experience a greater range in productivity than females (Lozano et al. 1996; Marra

and Holmes 1997; Hasselquist 1998; Langefors et al. 1998; Reudink et al. 2009). In

contrast, females are likely to experience less dramatic fitness declines with later arrival

dates, driven by decreasing territory and mate quality, mate-sharing, and later clutch

initiation dates (Alatalo et al. 1986; Bensch and Hasselquist 1992; Rowe et al. 1994;

Brown and Roth 2002; Smith and Moore 2005; Huk and Winkel 2006). Additionally, if

female migration is influenced by other factors under stronger selection (such as

improving body condition at stop-over sites prior to arrival (e.g. Lavee et al. 1991; Yong

et al. 1998)), winter-habitat quality may show no relationship to arrival date. Empirical

data offers support for this concept: winter habitat quality affected male but not female

arrival dates in both American Redstarts (Marra et al. 1998; Norris et al. 2004), and

Black-tailed Godwits (Gunnarsson et al. 2006).

Time-mediated carry-over effects may also be more evident in older birds for

several reasons. First, familiarity with breeding sites may allow early-arriving,

experienced males to select high-quality territories before leaf emergence; lack of prior

knowledge may reduce the advantages of early arrival for yearling males (Lozano et al.

1996). Secondly, older individuals may reduce or forgo reproductive investment late in

the breeding season in favour of improving the odds of over-winter survival and future

reproduction (Brown and Roth 2002). This behaviour should produce a strong negative

relationship between clutch initiation date and fledge number. Yearling birds, with poorer

over-winter survival, may continue to invest in current reproduction and thus show a

weaker negative relationship between clutch initiation date and fledge number (Brown

43

and Roth 2002). Finally, younger, inexperienced birds generally have lower reproductive

success than older individuals (reviewed in Forslund and Pärt 1995; Fowler 1995) and

poor overall performance may simply overwhelm carry-over effects.

It is also possible that greater experience enables older birds to compensate for

some of the negative effects of late arrival. For example, late-arriving older females may

be able to initiate and complete nest-building more rapidly than late-arriving yearling

birds. Such compensation would result in weaker carry-over effects for older birds and

could be differentiated from other age differences by weaker effect-slopes for older birds

for events occurring after arrival.

Understanding whether the strength of carry-over effects vary among age- and

sex-classes is necessary for accurate population modeling. For example, breeding

populations of many migratory songbirds contain a large proportion of yearling

individuals (e.g. Setophaga caerulescens: 38% of males (Graves 1997); Seiurus

aurocapilla: 39% of males (Porneluzi and Faaborg, 1999); Hylocichla mustelina: 48% of

males and 54% of females (Brown and Roth 2002)). These individuals tend not to be

evenly distributed throughout the species range (Graves 1997; Rohwer 2004).

Differences in carry-over effect strength between age classes could therefore create

regional variation in how a species responds to shifts in winter habitat quality.

In this study we examine whether the quality of winter habitat used by Yellow

Warblers in Mexico/Central America, influences their breeding performance in western

Canada and whether the strength of time-mediated carry-over effects varies between

sex- and age-classes. Specifically we ask whether winter habitat type, measured

indirectly through 13C and 15N isotope ratios in tissue samples, influences when males

establish territories, when females initiate clutches, and whether the timing of these

events impact breeding performance. We expect the relationship between breeding

season phenology and performance to be negative: late territory establishment and

clutch initiation will result in fewer fledged young. We predict that the strength of time-

mediated carry-over will vary with age and sex; specifically, that (1) both winter habitat to

territory establishment/territory arrival date and habitat to total productivity effects will be

stronger in males, and that (2) carry-over effects will be weaker in yearling individuals.

44

Methods

Study Species and location

Yellow Warblers are small, insectivorous, Nearctic – Neotropical migrants that

commonly breed in wet deciduous and riparian habitat. Their breeding range covers the

majority of the North American continent, and extends as far north as the arctic tree-line

(Lowther et al. 1999). Recent genetic and isotopic work suggest that eastern and

western populations undergo parallel migration with western populations wintering in

Mexico and Central America and eastern populations wintering in Central and South

America (Boulet et al. 2006).

Yellow Warblers are found in a diversity of natural winter habitats ranging from

mangrove, swamp and riparian forests to dry coastal scrub and are also found in a

variety of human modified habitats such as pasture, sun/semi-shade coffee plantations,

and other cropland (Binford 1989; Greenberg et al. 1996; Garrett and Dunn 1997). The

species undergoes a pre-alternate moult prior to spring migration, replacing a variable

number of greater covert, body, and crown feathers (Pyle 1997; Quinlan and Green

2010). As Yellow Warblers are territorial and exhibit winter site fidelity (Greenberg and

Ortiz, 1994; Greenberg and Salewski, 2005; Morton, 1976) these feathers are expected

to incorporate local isotopic signatures that reflect habitat use (Hobson 1999).

We have studied a banded population of Yellow Warblers near Revelstoke,

British Columbia (50.97 °N, 118.20°W; Figure 1.1) since 2004. Birds are monitored at

three, 30-39 ha study plots in riparian habitat bordering the northern section of the Upper

Arrow Lakes Reservoir (elevation 435-441 m). These sites consist of seasonally flooded

grassland, grading into isolated willow thickets (Salix sp.) and ultimately mature black

cottonwood (Populus trichocarpa) forest at elevations above the high-water level

(>438m). The average breeding density of Yellow Warblers within the study sites is 0.7

pairs/ha. In the years encompassed by this study, yearling (first-breeding season or

“SY”) females accounted for 43.8% of the breeding females at our study sites (annual

range 25-57%, n=5 years) and yearling (SY) males accounted for 30.0% of the breeding

males (annual range 13-48%, n=5 years).

45

Monitoring and breeding

Study sites were monitored from early-May until late-July. Territory establishment

dates were determined during surveys of the three study plots conducted every 1-2

days. Males were easily detected and captured, as they would begin to sing once on

territory and were highly aggressive toward song-playback. Males of both age-classes

were typically caught within three days of their appearance using targeted mist-netting.

Females were more difficult to detect and capture as they were less conspicuous and

less responsive to playback. Females were either caught with their mates or later in the

season by placing mist-nets across flight paths used during nest-building, incubation, or

while provisioning young. Upon capture, all previously unmarked individuals were fitted

with a Canadian Wildlife Service-issued aluminum band and a unique combination of

three colour bands (AC Hughes, UK).

Pairs were monitored from early May to late July in 2005-2006, and 2008-2010 in

order to determine the onset of breeding and the fate of all nesting attempts. Only pairs

that initiated clutches were included in the data that were analysed. Nests were located

through close observation of females during nest building and checked every 3 days to

determine clutch initiation dates, clutch size, brood size and fledging success. Clutch

initiation date was defined as the laying date of the first egg in a female’s first nesting

attempt of the season. Nestlings were banded seven days after hatching. We assumed

that all nestlings present on day 7 successfully fledged unless signs of predation were

found in or around the nest after fledging. Fledging was confirmed, in most cases, by

observing parents feeding or defending fledglings. Annual productivity for an individual

bird was defined as the number of fledglings produced over all nesting attempts.

Although rare, some males (3.9%, 7/178 males over 6 years) obtained two social mates,

and it was possible for their seasonal productivity to be higher than that of females.

Sexing and aging

Birds were sexed using plumage (brightness/degree of rufous streaking),

behaviour and breeding characteristics (brood patch/cloacal protuberance). Birds were

classified as yearling or older birds based on plumage characteristics. Yearling (SY)

birds have narrow, tapered primary coverts, tapered retrices with less yellow colouration

on the inner webs of the outer tail, and worn prebasic plumage; older (ASY) birds have

46

broad, truncate primary coverts with narrow yellow-olive edging, and truncate retrices

with more extensive yellow colouration on the inner webs (Pyle 1997).

Feather samples and stable isotope analysis

We collected three new inner greater covert feathers from all individuals captured

on the breeding grounds in 2008-2010. These feathers are easily identifiable because of

their brighter pigmentation and lack of wear and carry isotopic signatures from habitat

used during pre-alternate moult prior to spring migration (Pyle 1997; Quinlan and Green

2010).

Feathers were washed in 2:1 chloroform:methanol solution for 24hrs, drained,

and then air-dried in a fume hood for an additional 24h to remove excess solvent. One

mg (±0.2) of tissue was placed in 9×5mm smooth-walled tin capsules (Elemental

Microanalysis, UK). Tissue samples were analyzed using a PDZ Europa ANCA-GSL

elemental analyzer interfaced to a PDZ Europa 20-20 isotope ratio mass spectrometer

(Sercon Ltd., Cheshire, UK) at the University of California Davis Stable Isotope Facility in

California, USA. Delta values are expressed relative to international standards: V-PDB

(Vienna PeeDee Belemnite) for carbon and air for nitrogen; estimated measurement

precision was 0.2‰ for 13C and 0.3‰ for 15N.

Winter habitat use and 13C and 15N

In February and March 2011, we took tissue samples (n=89) from territorial

Yellow Warblers wintering on the Pacific slope (Chamela, Jalisco; 19.52 °N, 105.08°W),

and the southern Gulf coast (Los Tuxtlas, Veracruz; 18.57°N, 95.05°W) of Mexico. As

birds had not undergone pre-alternate moult we collected red blood cell (RBC) samples

to obtain local habitat signatures. Samples were collected from replicate habitat-types

across the range of habitat used to determine associated 13C and 15N signatures.

Within each habitat-replicate we collected 30 µl of blood from the brachial vein of 7-10

birds captured using targeted mist-netting. Blood was held in a heparinized

microcapillary tube and cooled on ice for a maximum of 5 1/2 hours (average of 3 hours)

before being spun down for 5 minutes at 11,000 rpm in a ZIPocrit centrifuge (LW

Scientific Inc., USA). Red blood cells were separated and frozen. Once in the lab

47

(Burnaby, Canada) samples were thawed, transferred to eppendorf tubes and placed in

a drying oven at 60°C for 3 days. Dried RBCs then weighed and analyzed in the same

fashion as feather tissue described in the preceding section.

13C patterns are largely driven by plant photosynthetic pathways that differ with

water availability (Hobson 2005). Use of cooler, mesic winter habitats is expected to

result in depleted 13C signatures. 15N in terrestrial animals can show complex variation

related the trophic level of dominant prey items and to local environmental factors (e.g.

marine inputs, waste, cultivation, and artificial fertilizer) (Kelly 2000; Amundson et al.

2003). However, 15N also varies more predictably with factors such as climate,

temperature, and water and/or nutritional stress (van der Merwe et al. 1990; Hobson et

al. 1993; Cormie and Schwarcz 1996; Kelly 2000; Amundson et al. 2003) and enriched

15N may suggest xeric and/or poor habitats.

Mesic (13C depleted) habitats have been shown to be better over-wintering sites

for American Redstarts (Studds and Marra 2005), Cape May Warblers (Setophaga

tigrina; Latta and Faaborg 2002), and Northern Waterthrushes (Parkesia

noveboracensis; Smith et al. 2010; Smith et al. 2011). Such habitat has greater insect

biomass than drier habitat (Studds and Marra 2005; Smith et al. 2010, Smith et al.

2011), resulting in less mass-loss and higher muscle mass for warblers during the dry

season (Marra and Holberton 1998; Latta and Faaborg, 2002; Smith et al. 2010), earlier

fat deposition prior to spring migration (Smith et al. 2010) and earlier spring migration

dates (Marra et al. 1998). Mesic riparian habitats in Chiapas, and Jalisco, Mexico

contain a greater proportion of older male Yellow Warblers than drier habitat-types,

suggesting that riparian habitat is of higher quality (Greenberg et al. 1997, Drake, in

preparation). We therefore expected that depleted 13C would be associated with earlier

territory establishment and clutch initiation in our breeding population (Marra et al. 1998;

Norris et al. 2004). We predicted that Yellow Warblers from poor wintering habitat might

have enriched 15N signatures if the use of such habitat results in nutritional deficiencies

by the end of the wintering period (Hobson et al. 1993). If poor habitats are also dry

habitats, then feather samples should be further 15N enriched and we would predict

high 15N values to be associated with high 13C values and later territory establishment

and clutch initiation dates.

48

Data Analysis

We first assessed whether 15N and 13C signatures in blood samples collected

from birds in Mexico differed across winter habitat-types using Analysis of Variance and

Tukey-Kramer HSD post-hoc tests for each element (Figure 3.1). We then evaluated

whether 13C or 15N of winter-grown feathers collected from birds in Revelstoke varied

by sex or age-within-sex-classes (Table 3.1). Revelstoke signatures were non-normally

distributed and we therefore used Wilcoxon tests for comparisons between classes. We

subsequently used general linear models to evaluate whether winter habitat use

influences the territory establishment dates, onset of breeding, and productivity of male

and female Yellow Warblers within different age classes.

The proposed pathways for carry-over effects in male and female warblers are

illustrated in Figure 3.2. For males we hypothesized that winter habitat could influence

territory establishment dates, because occupation of high quality habitat would allow

individuals to initiate spring migration earlier and/or in better condition, allowing them to

arrive on the breeding grounds earlier. We also tested whether winter habitat use could

directly affect clutch initiation, based on the hypothesis that males in high quality habitat

might have greater access to resources during winter moult and birds with superior

plumage might acquire a mate more rapidly (Studd and Robertson 1985a, 1985b). We

used a simpler pathway for females because female arrivals on territories were more

difficult to detect than males and later capture and banding made female identity prior to

clutch initiation less certain (Figure 3.2). Since the female model connects winter habitat

use directly to clutch initiation, it may incorporate winter-habitat effects on bird condition

in addition to those acting through arrival-on-territory date (see Discussion).

Data for yearling and older birds were analyzed separately because age

influences all aspects of breeding (see Results). Year was included as a potential

explanatory variable in all models. Sample sizes for the different steps in the proposed

pathways differ as tissue samples were not obtained from all individuals in all years.

Where winter habitat quality carried-over and impacted breeding productivity, we

used standardized partial regression coefficients (s) for each step in the pathway in

order to calculate the total effect of 13C and 15N on fledgling number. We then used the

shift in mean 13C and 15N values between wintering habitat-types with the greatest

49

isotopic difference to estimate the expected total shift in offspring number associated

with variation in winter habitat use (see Bart and Earnst 1999; Norris et al. 2004).

Results

Age differences in phenology and breeding performance

We collected data on male territory establishment dates, female nest initiation

dates and annual productivity for 132 males and 121 females over f ive years (2005-

2006; 2008-2010). Older males established territories an average of 8.1 days before

yearling males (partial F1,131=31.4, P<0.0001) and their social mates initiated nesting an

average 8.4 days before those of yearling males (partial F1,123=39.4, P<0.001). Older

males fledged twice as many young as yearling males (2.4 ± 1.9 vs. 1.1 ± 1.6 (partial

F1,129=11.7, P<0.001)). Older females initiated nests an average of 5.3 days before

yearling females (partial F1,115=18.0 P<0.001) and produced significantly more fledglings

(2.4 ± 0.22 vs. 1.4 ± 0.27 (partial F1,120=7.7, P=0.006)).

Winter habitat use and 13C and 15N

Yellow Warblers on the Pacific slope of Mexico wintered in riparian forest

corridors, agricultural field margins and coastal scrub habitat. Those on the southern

Gulf coast wintered within cattle pasture and along coastal habitat. Red blood cell 13C

and 15N signatures differed significantly among these habitats (13C: F4,78=22.16,

P<0.0001; 15N: F4,78=81.49, P<0.0001; Figure 3.1). Tukey HSD tests showed that 15N

signatures were significantly more depleted on the southern Gulf coast than on the

Pacific slope (P<0.0001). 13C signatures on the Pacific slope varied with habitat use

(P=0.05), being most depleted in riparian habitats. 13C signatures in pasture and

coastal habitat on the Gulf of Mexico did not differ (P=0.52).

Within our Revelstoke breeding population, 13C and 15N signatures were only

weakly correlated (r=0.13, P=0.09, n=171) indicating that these signatures were

reflecting different aspects of wintering habitat use. 13C and 15N values did not differ

significantly between sex- and age-within-sex-classes (Table 3.1; 13C: Gender, Z=0.09,

50

P=0.93; Age-class females, Z=-0.36, P=0.72; Age-class males, Z=-0.84, P=0.40; 15N:

Gender, Z=-1.51, P=0.13, Age-class females, Z=0.30, P=0.76, Age-class males, Z=-

0.05, P=0.96).

Carry-over effects

Winter habitat use only had a carry-over effect on the productivity of yearling

females (Table 3.2, Figure 3.2). More depleted 13C signatures and more enriched 15N

signatures were correlated with earlier clutch initiation dates in this group (Table 3.2,

Figure 3.3). Earlier clutch initiation dates were, in turn, associated with the production of

more fledglings by yearling females (Figure 3.2). Using the isotopic values obtained from

wintering birds of known habitat origin in Mexico, this result suggests that birds

originating in mesic habitat within a dry/seasonal climate region performed better than

those with 13C and 15N signatures consistent with wetter climate regions.

The total effect of 13C on the productivity of yearling females was -0.17. Thus, a

shift in one standard deviation of 13C would be expected to result in a shift of (-0.17)

standard deviations in fledglings. Using our wintering isotope data, our model would

predict that yearling females occupying habitat at the mesic end of the 13C gradient (i.e.

13C depleted Pacific riparian forest) would be able to produce 0.8 more fledglings the

following summer than counterparts in 13C enriched coastal or pasture habitat in the

southern Gulf of Mexico. Similarly, the total effect of 15N on the productivity of yearling

females was 0.15. Our model would thus predict that females with 15N signatures

equivalent to the average of our Pacific slope site would produce 0.6 more fledglings

than birds with signatures equivalent to the southern Gulf coast.

Although we found no evidence of winter habitat effects in our other age-sex

classes, early territory establishment was associated with early onset of reproduction

and greater annual productivity in older males (Table 3.2). Similarly, early onset of

reproduction in older females was associated with greater annual productivity (Table

3.2). Although territory establishment dates were related to the onset of reproduction in

yearling males, clutch initiation date had no impact on reproductive success (Table 3.2,

Figure 3.2).

51

Annual differences (“year”) had significant direct effects on older male territory

establishment dates and on yearling female clutch initiation dates (Figure 3.2).

Discussion

Carry-over effects have generated considerable attention because of their ability

to explain some of the observed variation in individual fitness within populations and in

population processes between years (Webster and Marra 2005). Our study offers

additional support for the concept that events occurring in one season can affect

important biological processes in a subsequent season. We expand upon previous work

(e.g. Marra et al. 1998; Gill et al. 2001; Norris et al. 2004; Gunnarsson et al. 2005;

Gunnarsson et al. 2006; Reudink et al. 2009; Tonra et al. 2011) by investigating age-

and sex-class effects. We predicted that males would experience stronger time-

mediated carry-over effects due to the importance of arrival date with regards to territory

acquisition and the greater range in productivity experienced by this sex on breeding

grounds. Similarly, we predicted that age-dependent differences in experience and

performance would make time-mediated carry-over effects weaker in yearling birds.

Contrary to our predictions, there was no evidence that winter habitat, inferred using

13C and 15N, influenced the territory establishment date of Yellow Warbler males or the

clutch initiation dates of their social mate. This differs from research on American

Redstarts (Marra et al. 1998, Norris et al. 2004) but is consistent with other Yellow

Warbler work in Ohio that found no relationship between male winter 13C signatures

and the clutch initiation dates of their social mates (Lindsay 2008).

Time-mediated carry-over effects did act on yearling female Yellow Warblers. For

this age-sex class, depleted 13C and enriched 15N wintering signatures were

associated with earlier clutch initiation dates and higher subsequent productivity. Our

13C results indicate the yearling females originating from mesic winter habitats have

higher fitness than those with signatures suggesting drier habitat use. This finding is

consistent with previous migration phenology and carry-over work (Marra et al. 1998,

Norris et al. 2004, Smith et al. 2010 (departure delay inferred from body condition)).

52

The relationship between breeding season phenology and wintering 15N

signatures has not been previously investigated in landbirds. Our blood 15N signatures

of Yellow Warblers sampled on the wintering grounds align with predicted soil 15N

enrichment patterns produced by regional temperature and precipitation gradients

(Amundson et al. 2003; Handley et al. 1999). Coastal Jalisco falls within the As/Aw

climate zones (Köppen-Geiger climate designation; Kottek et al. 2006) where high

temperatures and low rainfall would produce 15N enriched soil. Southern Veracruz (Am

climate zone) receives 3 times more precipitation and its soil 15N should be

correspondingly depleted. As/Aw climate zones cover the northern portion of the Yellow

Warbler wintering range and follow the Pacific slope of Central America as far south as

Panama. Wetter climate regions first arise at the southern-most portion of the Gulf of

Mexico and then continue southward on the Caribbean coast of Central America. If 15N

signatures in our Yellow Warblers reflect regional soil 15N signatures, we would expect

birds using the northern portion of the wintering range to be uniformly 15N enriched,

while birds wintering further south would have mixed signatures. Delayed breeding

associated with depleted 15N may therefore be related to wintering latitude, with

yearling females wintering further south arriving later than counterparts from the northern

portion of the wintering range.

Tissue 13C values in our study population showed a similar range to those

reported for American Redstarts (Marra et al. 1998; Norris et al. 2005; Reudink et al.

2009) (Table 3.1). This, when coupled with our wintering data, would suggest that there

is sufficient among-individual variation in winter habitat use to produce detectable carry-

over effects in other age-sex classes within our population. It is possible that similar

habitat variation does not influence spring migration phenology in western songbirds as

strongly as it does in eastern populations of American Redstarts. Such differences could

be driven by regional differences in habitat tolerance or by flyway geography.

Western migrants make greater use of seasonal tropical environments that are

mostly avoided by their eastern counterparts (Terborgh 1989). It has been suggested

that this pattern is driven by eastern populations becoming specialized with regards to

the mesic forests that dominate eastern North America (Terborgh 1989; Rotenberry et

al. 1995). Yellow Warblers occupy a much broader breeding range than American

53

Redstarts, which might suggest that they are less specialized in the tropics as well. An

additional feature of western systems is the absence of large geographical barriers.

Birds wintering on Caribbean islands, or those crossing the Gulf of Mexico must obtain a

threshold mass before initiating migration across open water (Schaub et al. 2008). This

could produce a close relationship between the rate of mass gain on wintering territories

and departure dates in eastern populations, a pattern which may not be present in the

west.

Lack of support for winter habitat-driven carry-over effects in males was not due

to the absence of timing effects on the breeding grounds. Territory establishment dates

had a large effect on the date by which male Yellow Warblers had attracted a mate and

that mate had initiated reproduction. For older males, the timing of clutch initiation

ultimately influenced productivity; for yearling males, it did not, principally because of the

low overall reproductive success in this age class. Because breeding season phenology

did influence the reproductive success of older males, it is possible that seasonal

interactions mediated through migration phenology could be operating in this older age-

sex class. Factors such as inter-annual variation in climate (Nott et al. 2002; Vähätalo et

al. 2004) deserve exploration, as “year” had a significant effect on the territory

establishment dates of older males in this study.

We predicted that time-mediated carry-over effects would be weaker in female

Yellow Warblers. In American Redstarts 13C had no effect on female arrival dates but

instead had a direct effect on the fledging date of those females that successfully reared

a clutch (Norris et al. 2004). Fledge date subsequently influenced the productivity of

those females. Although female condition would be the most likely mechanism through

which winter habitat could influence fledge date, no such correlation was found (Norris et

al. 2004) and this relationship may instead be linked to intrinsic quality. As in the merged

female age classes assessed in Norris et al. and Lindsay (2008), we found no evidence

of a winter habitat effect on the clutch initiation dates of older Yellow Warbler females.

Contrary to our expectations, we did find evidence of carry-over effects operating via

clutch initiation date in yearling females. Genders did not differ significantly in terms of

their 13C and 15N values and yearling females did not differ from older females, making

it unlikely that the presence of carry-over effects in this class is due to these individuals

occupying a wider range of habitats or relatively worse habitat than other classes (Table

54

3.1). It possible that carry-over effects are not seen in older, experienced, females

because they are better able to compensate for late arrival on territories by initiating

clutches more rapidly. However, our female arrival data does not support this, with lag

times to clutch initiation not differing significantly by age. Older females may also differ

from yearling females in their migration behaviour (with differing departure schedules or

stop-over timing) in a way that reduces their spread in arrival dates and subsequent

clutch initiation dates. Again, such a hypothesis was not supported by the available data:

older females showed a wider absolute range of dates as a product of earlier clutch

initiations than yearling females.

That time-mediated carry-over effects vary by age and by sex in Yellow Warblers

has population-level implications. In the 5 years encompassed by this study, 43.8% of

breeding females were yearling birds. As a result, changes in mean winter habitat quality

(Norris and Taylor 2006) could be expected to have productivity repercussions for

roughly half of the pairs at our study site. The total effect score of 13C and 15N on the

productivity of yearling females was -0.17 and 0.15 respectively, independently

corresponding to an estimated shift of 0.8 and 0.6 fledged young across the range of

mean isotope values presented in Figure 3.1. This response is less than that reported for

female American Redstarts, where a comparable shift in 13C resulted in an estimated

shift of 2 fledged young. However, like American Redstarts, Yellow Warblers generally

rear a single brood per season and have an average first clutch size of four offspring.

Thus, even small shifts in fledgling number represent a large shift in total productivity for

an individual. An additional implication of age-specific carry-over effects is geographic. If

a greater proportion of yearling birds are found at the periphery of a species breeding

range (Graves 1997; Rohwer 2004) and carry-over effects are present in yearling but not

older individuals, winter habitat degradation may lead to reduced productivity at the

edges of the breeding range but have little effect on productivity within higher-quality

breeding areas.

As with all among-individual assessments of carry-over effects, our results have

a caveat. In this type of correlational study it is not possible to isolate winter habitat

effects from the potential effect of the intrinsic quality of individuals (Harrison et al.

2011). It is possible that intrinsic quality may be driving both winter habitat occupancy

and reproductive success, leaving these factors confounded. Within-individual

55

differences in breeding season phenology and winter signatures across years, or studies

that incorporate winter habitat alteration (such as food supplementation) and record

migration date would be needed to confirm a causal link.

Our data are consistent with other work that has shown sex differences in carry-

over effect presence or strength (Marra et al. 1998, Norris et al. 2004, Gunnarsson et al.

2005, Gunnarsson et al. 2006). However, these previous studies found arrival-mediated

carry-over in males and not in females. Our study also indicates that carry-over effects

can vary between age classes. Saino et al. (2004) found differences in between-year

carry-over effects by age; to our knowledge this is the first paper to show within-year

differences.

The wintering ecology of many Neotropical warblers is poorly understood. The

majority of our current knowledge is based on research conducted on eastern

populations wintering in the Caribbean. Carry-over effects and age-sex class variation in

carry-over effects may be more evident (and more important from a conservation

standpoint (Norris et al. 2004)) among migratory bird species that prefer mesic winter

habitat. Such habitats buffer the impact of the late-winter dry season experienced in

many regions of the Neotropics (Rotenberry et al. 1995) and marked fitness differences

may exist between individuals with access to it and those occupying drier habitat.

Additional work is needed within the wintering range of western populations and with

western warbler species. These populations face different winter climate regimes, and

possibly possess a greater tolerance for drier winter habitat types than their eastern

counterparts. We advocate incorporating biologically relevant differences such as age-

class and sex into carry-over analyses and population models that include carry-over

effects in order to better understand how these effects act on individuals within a

population and to better predict population effects associated with habitat loss.

References

Alatalo, R. V., Lundberg, A., and Glynn, C. 1986. Female pied flycatchers choose territory quality and not male characteristics. – Nature 323: 152-153. doi:10.1038/323152a0

56

Amundson, R., Austin, A. T., Schuur, E. A. G., Yoo, K., Matzek, V., Kendall, C., Uebersax, A., et al. 2003. Global patterns of the isotopic composition of soil and plant nitrogen. – Glob. Biogeochem. Cycles 17: 1031-1041. doi:200310.1029/2002GB001903

Bart, J., and Earnst, S. L. 1999. Relative importance of male and territory quality in pairing success of male rock ptarmigan (Lagopus mutus). – Behav. Ecol.

Sociobiol. 45: 355–359. doi:10.1007/s002650050571

Bayne, E. M., and Hobson, K. A. 2001. Effects of habitat fragmentation on pairing success of ovenbirds: importance of male age and floater behaviour. – Auk 118: 380–388. doi: 10.2307/4089799

Bearhop, S., Hilton, G. M., Votier, S. C., and Waldron, S. 2004. Stable isotope ratios indicate that body condition in migrating passerines is influenced by winter habitat. – P. Roy. Soc. B-Biol. Sci. 271 (Suppl 4): S215–S218. doi:10.1098/rsbl.2003.0129

Bensch, S., and Hasselquist, D. 1992. Evidence for active female choice in a polygynous warbler. – Anim. Behav. 44: 301–311. doi:10.1016/0003-3472(92)90036-9

Binford, L. C. 1989. A distributional survey of the birds of the Mexican state of Oaxaca. – Ornithol. Monogr. 43: 392–418. doi:10.2307/40167673

Boulet, M., Gibbs, H. L., and Hobson, K. A. 2006. Integrated analysis of genetic, stable isotope, and banding data reveal migratory connectivity and flyways in the Northern Yellow Warbler (Dendroica petechia; Aestiva group). – Ornithol. Monogr. 61: 29–78. doi: 10.2307/40166837

Brown, W. P., and Roth, R. R. 2002. Temporal patterns of fitness and survival in the Wood Thrush. – Ecology 83: 958–969. doi:10.2307/3071905

Cormie, A., Schwarcz, H.P. 1996. Effects of climate on deer bone δ15N and δ13C: Lack of precipitation effects on δ15N for animals consuming low amounts of C4 plants. – Geochim. Cosmochim. Ac. 60: 4161–4166. doi:10.1016/S0016-7037(96)00251-7

Ebbinge, B. S., and Spaans, B. 1995. The importance of body reserves accumulated in spring staging areas in the temperate zone for breeding in Dark-Bellied Brent Geese Branta b. bernicla in the High Arctic. – J. Avian Biol. 26: 105–113. doi:10.2307/3677058

Forslund, P., and Pärt, T. 1995. Age and reproduction in birds — hypotheses and tests. – Trends Ecol. Evol. 10: 374–378. doi:10.1016/S0169-5347(00)89141-7

Fowler, G. S. 1995. Stages of age-related reproductive success in birds: simultaneous effects of age, pair-bond duration and reproductive experience. – Am. Zool. 35: 318–328. doi:10.1093/icb/35.4.318

Garrett, K. and Dunn, J. 1997. A field guide to warblers of North America. – Houghton Mifflin Harcourt.

57

Gill, J. A., Norris, K., Potts, P. M., Gunnarsson, T. G., Atkinson, P. W., and Sutherland, W. J. 2001. The buffer effect and large-scale population regulation in migratory birds. – Nature 412: 436–438. doi:10.1038/35086568

Graves, G. R. 1997. Geographic clines of age ratios of black-throated blue warblers (Dendroica caerulescens). – Ecology 78: 2524–2531. doi:10.2307/2265911

Greenberg, Russel, Bichier, P., and Sterling, J. 1997. Acacia, cattle and migratory birds in southeastern Mexico. – Biol. Conserv. 80: 235–247. doi:10.1016/S0006-3207(96)00137-1

Greenberg, Russell, and Ortiz, J. S. 1994. Interspecific defense of pasture trees by wintering Yellow Warblers. – Auk 111: 672–682.

Greenberg, R., Reitsma, R., and Angon, A. C. 1996. Interspecific aggression by Yellow Warblers in a sun coffee plantation. – Condor 98: 640–642. doi:10.2307/1369579

Greenberg, R., and Salewski, V. 2005. Ecological correlates of wintering social systems in new world and old world migratory passerines. – In: Greenberg, R. and Marra, P.P. (eds.), Birds of two worlds: the ecology and evolution of migration. Johns Hopkins University Press, pp. 336–358.

Gunnarsson, T. G., Gill, J. A., Atkinson, P. W., Gélinaud, G., Potts, P. M., Croger, R. E., Gudmundsson, G. A., et al. 2006. Population-scale drivers of individual arrival times in migratory birds. – J. Anim. Ecol. 75: 1119–1127. doi:10.1111/j.1365-2656.2006.01131.x

Gunnarsson, T. G., Gill, J. A., Newton, J., Potts, P. M., and Sutherland, W. J. 2005. Seasonal matching of habitat quality and fitness in a migratory bird. – P. Roy. Soc. B-Biol. Sci. 272: 2319 –2323. doi:10.1098/rspb.2005.3214

Handley, L. L., Austin, A. T., Robinson, D., Scrimgeour, C. M., Raven, J. A., Heaton, T. H. E., Schmidt, S., et al. 1999. The 15N natural abundance (δ15N) of ecosystem samples reflects measures of water availability. – Aust. J. Plant Physiol. 26: 185–199. doi:10.1071/PP98146

Harrison, X. A., Blount, J. D., Inger, R., Norris, D. R., and Bearhop, S. 2011. Carry-over effects as drivers of fitness differences in animals. – J. Anim. Ecol. 80: 4–18. doi:10.1111/j.1365-2656.2010.01740.x

Hasselquist, D. 1998. Polygyny in Great Reed Warblers: a long-term study of factors contributing to male fitness. – Ecology 79: 2376–2390. doi:10.2307/176829

Hobson, K. A. 2005. Stable isotopes and the determination of avian migratory connectivity and seasonal interactions. – Auk 122: 1037-1048. doi:10.1642/0004-8038(2005)122[1037:SIATDO]2.0.CO;2

Hobson, K. A. 1999. Tracing origins and migration of wildlife using stable isotopes: a review. – Oecologia 120: 314–326. doi:10.1007/s004420050865

58

Hobson, K. A., Alisauskas, R. T., and Clark, R. G. 1993. Stable-nitrogen isotope enrichment in avian tissues due to fasting and nutritional stress: implications for isotopic analyses of diet. – Condor 95: 388–394. doi:10.2307/1369361

Huk, T., and Winkel, W. 2006. Polygyny and its fitness consequences for primary and secondary female Pied Flycatchers. – P. Roy. Soc. B-Biol. Sci. 273: 1681–1688. doi:10.1098/rspb.2006.3485

Kelly, J. F. 2000. Stable isotopes of carbon and nitrogen in the study of avian and mammalian trophic ecology. – Can. J. Zool. 78: 1–27. doi:10.1139/z99-165

Kottek, M., Grieser, J., Beck, C., Rudolf, B., and Rubel, F. 2006. World map of the Köppen-Geiger climate classification updated. – Meteorol. Z. 15: 259–263. doi:10.1127/0941-2948/2006/0130

Langefors, Å., Hasselquist, D., and von Schantz, T. 1998. Extra-pair fertilizations in the Sedge Warbler. – J. Avian Biol. 29: 134–144. doi:10.2307/3677191

Latta, S. C., and Faaborg, J. 2002. Demographic and population responses of Cape May Warblers wintering in multiple habitats. – Ecology 83: 2502–2515. doi:10.2307/3071810

Lavee, D., Safriel, U. N., and Meilijson, I. 1991. For how long do trans-Saharan migrants stop over at an oasis? – Ornis Scand. 22: 33–44. doi:10.2307/3676619

Lindsay, A. M. 2008. Seasonal Events and Associated Carry-over Effects in a Neotropical Migratory Songbird, the Yellow Warbler (Dendroica petechia). –

M.Sc. Ohio State University.

Lowther, P. E., Celada, C., Klein, N. K., Rimmer, C. C., and Spector, D. A. 1999. Yellow Warbler (Dendroica petechia). – In: Poole, A. and Gill, F. (eds.), The birds of

North America, No. 454. The Academy of Natural Science, Philadelphia, and The American Ornithologists’ Union, Washington, D.C.

Lozano, G. A., Perreault, S., and Lemon, R. E. 1996. Age, arrival date and reproductive success of male American Redstarts Setophaga ruticilla. – J. Avian Biol. 27:

164–170. doi:10.2307/3677146

Marra, P. P., Hobson, K. A., and Holmes, R. T. 1998. Linking winter and summer events in a migratory bird by using stable-carbon isotopes. – Science 282: 1884 –1886. doi:10.1126/science.282.5395.1884

Marra, P. P., and 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:10.1007/s004420050590

Marra, P. P., and Holmes, R. T. 1997. Avian removal experiments: do they test for habitat saturation or female availability? – Ecology 78: 947–952. doi:10.2307/2266072

59

Morton, E. S. 1976. The adaptive significance of dull coloration in Yellow Warblers. – Condor, 78: 423. doi:10.2307/1367713

Norris, D. R., and Marra, P. P. 2007. Seasonal interactions, habitat quality, and population dynamics in migratory birds. – Condor 109: 535-547. doi:10.1650/8350.1

Norris, D. R., Marra, P. P., Kyser, T. K., and Ratcliffe, L. M. 2005. Tracking habitat use of a long-distance migratory bird, the American Redstart Setophaga ruticilla, using

stable-carbon isotopes in cellular blood. – J. Avian Biol. 36: 164–170. doi:10.1111/j.0908-8857.2005.03398.x

Norris, D. R., Marra, P. P., Kyser, T. K., Sherry, T. W., and Ratcliffe, L. M. 2004. Tropical winter habitat limits reproductive success on the temperate breeding grounds in a migratory bird. - P. Roy. Soc. B-Biol. Sci. 271: 59–64. doi:10.1098/rspb.2003.2569

Norris, D. R., and Taylor, C. M. 2006. Predicting the consequences of carry-over effects for migratory populations. – Biol. Letters 2: 148–151. doi:10.1098/rsbl.2005.0397

Nott, M. P., Desante, D. F., Siegel, R. B., and Pyle, P. 2002. Influences of the El Niño/Southern Oscillation and the North Atlantic Oscillation on avian productivity in forests of the Pacific Northwest of North America. – Global Ecol. Biogeogr. 11: 333–342. doi:10.1046/j.1466-822X.2002.00296.x

Porneluzi, P. A., and Faaborg, J. 1999. Season-long fecundity, survival, and viability of ovenbirds in fragmented and unfragmented landscapes. – Conserv. Biol. 13: 1151–1161. doi:10.1046/j.1523-1739.1999.98455.x

Pyle, P. 1997. Identification Guide to North American Birds, Part I: Columbidae to Ploceidae. Slate Creek Press.

Quinlan, S. P., and Green, D. J. 2010. Variation in deuterium (δD) signatures of Yellow Warbler Dendroica petechia feathers grown on breeding and wintering grounds.

– J. Ornithol. 152: 93–101. doi:10.1007/s10336-010-0553-5

Reudink, M. W., Marra, P. P., Kyser, T. K., Boag, P. T., Langin, K. M., and Ratcliffe, L. M. 2009. Non-breeding season events influence sexual selection in a long-distance migratory bird. – P. Roy. Soc. B-Biol. Sci. 276: 1619–1626. doi:10.1098/rspb.2008.1452

Rockwell, S. M., Bocetti, C. I., and 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). – Auk 129: 744–752. doi:

10.1525/auk.2012.12003

Rohwer, S. (2004). Using age ratios to infer survival and despotic breeding dispersal in hybridizing warblers. – Ecology 85: 423–431. doi:10.1890/02-0380

60

Rotenberry, J. T., Cooper R. J., Wunderle, J. M., and Smith K. S. 1995. When and how are populations limited: the roles of insect outbreaks, fires, and other natural disturbances. – In: T.E. Martin and D.M. Finch (eds.) Ecology and management of neotropical migratory birds: a synthesis and review of critical issues. Oxford University Press, pp. 55-84.

Rowe, L., Ludwig, D., and Schluter, D. 1994. Time, condition, and the seasonal decline of avian clutch size. – Am. Nat. 143: 698–722. doi:10.1086/285627

Runge, M. C., and Marra, P. P. 2005. Modelling seasonal interactions in the population dynamics of migratory birds. – In: Greenberg, R. and Marra, P.P. (eds.), Birds of two worlds: the ecology and evolution of migration. Johns Hopkins University Press, pp. 336–358.

Saino, N., Szép, T., Ambrosini, R., Romano, M., and Møller, A. P. 2004. Ecological conditions during winter affect sexual selection and breeding in a migratory bird. – P. Roy. Soc. B-Biol. Sci. 271: 681–686. doi:10.1098/rspb.2003.2656

Schaub, M., Jenni, L., and Bairlein, F. 2008. Fuel stores, fuel accumulation, and the decision to depart from a migration stopover site. – Behav. Ecol. 19: 657–666. doi:10.1093/beheco/arn023

Smith, J. A. M., Reitsma, L. R., and Marra, P. P. 2010. Moisture as a determinant of habitat quality for a nonbreeding Neotropical migratory songbird. – Ecology 91: 2874–2882. doi:10.1890/09-2212.1

Smith, J. A. M., Reitsma, L. R., and Marra, P. P. 2011. Multiple space-use strategies and their divergent consequences in a nonbreeding migratory bird (Parkesia noveboracensis). – Auk 128: 53–60. doi:10.1525/auk.2011.10241

Smith, R. J., and Moore, F. R. 2005. Arrival timing and seasonal reproductive performance in a long-distance migratory landbird. – Behav. Ecol. Sociobiol. 57: 231–239. doi:10.1007/s00265-004-0855-9

Studd M. V., Robertson, R. J. 1985a. Evidence for reliable badges of status in territorial yellow warblers (Dendroica petechia). – Anim. Behav. 33: 1102–1113. doi:10.1016/S0003- 3472(85)80169-X

Studd M.V., Robertson R.J. 1985b. Sexual selection and variation in reproductive strategy in male yellow warblers (Dendroica petechia). – Behav. Ecol. Sociobiol.

17: 101–109. doi:10.1007/BF00299242

Studds, C. E., and Marra, P. P. 2005. Nonbreeding habitat occupancy and population processes: an upgrade experiment with a migratory bird. – Ecology 86: 2380–2385. doi:10.1890/04-1145

Terborgh, J. 1989. Where have all the birds gone?: essays on the biology and conservation of birds that migrate to the American tropics. Princeton University Press.

61

Tonra, C. M., Marra, P. P., and Holberton, R. L. 2011. Migration phenology and winter habitat quality are related to circulating androgen in a long-distance migratory bird. – J. Avian Biol. 42: 397–404. 10.1111/j.1600-048X.2011.05333.x

Vähätalo, A. V., Rainio, K., Lehikoinen, A., and Lehikoinen, E. 2004. Spring arrival of birds depends on the North Atlantic Oscillation. – J. Avian Biol. 35: 210–216. doi:10.1111/j.0908-8857.2004.03199.x

van der Merwe, N. J., Lee-Thorp, J. A., Thackeray, J. F., Hall-Martin, A., Kruger, F. J., Coetzee, H., Bell, R. H. V., et al. 1990. Source-area determination of elephant ivory by isotopic analysis. – Nature 346: 744–746. doi:10.1038/346744a0

Webster, M. S., and Marra, P. P. 2005. The importance of understanding migratory connectivity and cross-seasonal interactions. – In: Greenberg, R. and Marra, P.P. (eds.), Birds of two worlds: the ecology and evolution of migration. Johns Hopkins University Press, pp. 336–358.

Yong, W., Finch, D. M., Moore, F. R., and Kelly, J. F. 1998. Stopover ecology and habitat use of migratory Wilson’s Warblers. – Auk 115: 829–842. doi:10.2307/4089502

62

Tables and Figures

Table 3.1 Summary of stable isotope signatures (n, mean ±SD and range) from winter-grown feathers collected from Yellow Warblers breeding in Revelstoke, British Columbia. Yearling individuals are in their first breeding season; older birds are at least 2 years old.

Age-Sex Class N 13C (‰) 15N (‰)

min Q1 med Q3 max min Q1 med Q3 max

Yearling Females 42 -24.6 -22.3 -21.5 -20.9 -18.1 5.1 7.6 8.2 9.6 13.3

Older Females 41 -23.6 -22.3 -21.7 -21.1 -18.8 5.3 7.6 8.4 9.6 11.4

Yearling Males 21 -23.7 -22.5 -21.8 -21.1 -17.4 6.4 7.2 8.5 10.9 12.5

Older Males 64 -24.8 -22.3 -21.5 -21.0 -16.8 5.2 7.6 8.9 10.2 15.5

Pooled 168 -24.8 -22.3 -21.6 -21.0 -16.9 5.1 7.6 8.5 9.9 15.5

63

Table 3.2 Results of general linear model analyses for each stage of the proposed pathway from winter habitat use to fledgling productivity. Relationships for yearling and older Yellow Warbler are reported separately for (A) males and (B) females. Standardized beta coefficients for significant continuous variables are reported in Figure 3.2.

A) Males Yearling (SY) Older (ASY)

r2 n p F ratio Prob>F r2 n p F ratio Prob>F

Arrival 0.18 19 0.55

0.15 63 0.05

Year 0.12 0.89 4.70 0.01

13C 0.00 0.97 1.82 0.18

15N 2.28 0.15 0.42 0.52

Clutch Initiation 0.61 15 0.09

0.42 58 <0.001

Year 0.18 0.83 1.86 0.17

Arrival 13.23 0.005 29.21 <0.001

13C 0.24 0.64 2.48 0.12

15N 1.02 0.34 0.07 0.79

Fledge number 0.14 39 0.55

0.13 114 0.02

Year 0.84 0.53 1.76 0.13

Clutch initiation 1.09 0.3 8.89 0.004

B) Female Yearling (SY) Older (ASY)

r2 n p F ratio Prob>F r2 n p F ratio Prob>F

Clutch Initiation 0.33 38 0.009 0.10 39 0.47

Year 3.37 0.05 1.54 0.23

13C 5.45 0.03 0.52 0.48

15N 4.35 0.04 0.37 0.55

Fledge number 0.27 61 0.007

0.13 84 0.08

Year 1.48 0.21 1.71 0.14

Clutch initiation 14.93 <0.001 5.65 0.02

64

Figure 3.1 Biplot of the mean (±SD) 15N (‰) and 13C (‰) values of red blood cells from Yellow Warblers over-wintering in different habitat-types on the southern Gulf coast (Veracruz) and the Pacific slope (Jalisco) of Mexico. Bracketed letters indicate signature overlap: birds from habitats with shared letters were not significantly different from

each other in terms of tissue 15N (A-C) and/or 13C (D-F).

65

Figure 3.2 Model pathways for Yellow Warbler males and females by age class. Black solid lines indicate significant relationships between variables (thickest line P<0.0001, mid-thickness P<0.001, thin line P<0.05). Dashed lines indicate no statistically-detectable relationship between variables. Standardized beta coefficients for significant continuous variables are shown above each arrow. Whole-model r2 values are reported in parentheses above the dependent variables for significant relationships only.

66

Figure 3.3 Relationship between winter isotope signatures andclutch initiation date in yearling female Yellow Warblers. Initiation dates are

presented as residuals from a model controlling for year and 15N

(a) or for year and 13C (b).

67

Chapter 4. Winter habitat use does not influence spring arrival dates or the reproductive success of Yellow Warblers breeding in the arctic

Anna Drake, Michaela Martin, David J. Green1

Abstract

Winter-habitat use can influence the breeding success of migratory songbirds in

temperate regions due to its impact on bird condition and breeding phenology. How such

carry-over effects vary with latitude and therefore how they are incorporated into

migration trade-offs is unknown. To address this question, we examined how winter

habitat use, inferred from 13C and 15N signatures in winter-grown feathers, influenced

the breeding phenology and productivity of Yellow Warblers (Setophaga petechia) at the

extreme north of their range in the Canadian arctic (68ºN) and compared this population

to mid-latitude Yellow Warbler (50ºN) and American Redstart (Setophaga ruticilla; 44ºN)

populations reported in previous studies. In the arctic, we examined male arrival dates,

female clutch initiation dates and the relationship between these timing variables and the

number and quality of offspring produced within the season. In contrast to warblers

breeding at mid-latitudes, we find no support for an impact of winter habitat use on

timing or productivity measures. Male arrival dates and female clutch initiation dates in

both young and older individuals were not correlated with isotopic signatures acquired

on the wintering grounds. Males with enriched 15N signatures paired more rapidly after

arrival, indicating a possible relationship between winter habitat use and condition. This

1 A version of this chapter has been submitted to the Journal of Polar Biology

68

relationship did not translate into productivity advantages as the negative relationship

between breeding phenology and reproductive success was significantly weaker in our

northern population when compared to Yellow Warblers breeding further south. This

reduction or absence of timing effects on productivity effectively removes one pathway

through which carry-over effects can act.

Introduction

Among migratory birds, individuals breeding at high latitudes are expected to

have higher annual productivity, but lower inter-annual survival odds than their more

southern counterparts (Alerstam et al. 2003). Increased productivity is expected because

of reduced nest predation at higher latitudes (McKinnon et al. 2010; Salgado-Ortiz et al.

2008) and greater resource availability (Cox 1985; Alerstam et al. 2003). Conversely,

energetic cost and presumably mortality risk will increase with traveling distance (Bell

1996; Wikelski et al. 2003). Breeding latitude may therefore be seen as a trade-off

between movement costs and breeding benefits. In the absence of range shifts,

productivity and mortality should balance each other and populations at different

latitudes should perform equally (Alerstam and Enckell 1979; Sutherland 1996).

Studies have shown that the quality of winter habitat occupied by migrant birds

can influence bird condition (Sherry and Holmes 1996; Studds and Marra 2005; Smith,

et al. 2011). Poor winter conditions and delayed pre-migratory fattening can delay spring

migration and therefore the timing of events on the breeding grounds (e.g. Marra et al.

1998; Saino et al. 2004). Birds arriving late may produce fewer offspring and this chain

of events therefore represents a cross-seasonal or “carry-over” effect (Norris et al.

2004). The influence of wintering-ground events on individual breeding performance

might be expected to vary with latitude for several reasons. First, the role of breeding

phenology in individual productivity should vary with latitude. Suitable conditions for

rearing young exist for a progressively smaller window of time as latitude increases

(Newton 2008). A shortened breeding season should increase the reproductive costs

associated with delayed arrival at northern breeding sites. Secondly, conditions at the

start of the breeding period are harsher at more northern latitudes and body condition

and energy reserves may become more important with increasing latitude (Sandberg

69

1996). Thirdly, for birds wintering in the same region, the energetic costs of migration will

be greater with increasing breeding latitude (Wikelski et al. 2003). We might therefore

predict that the importance of bird condition at the onset of migration would increase with

breeding latitude (Marra 2012) and that time- and condition-mediated carry-over effects

will be more pronounced in northern populations. However, it is also possible that, with

increasing migration distance, events occurring during migration increase in their relative

contribution to individual arrival date and condition. This would be expected to dampen

the impact of winter events on breeding processes and make winter to breeding-ground

carry-over effects weaker in northern populations.

While carry-over effects have been investigated at mid-latitudes using songbird

populations (e.g. Saino et al. 2004; Norris et al. 2004; Rockwell et al. 2012; Drake et al.

2013; Chapter 3). All high-latitude studies have been conducted with geese and waders

(e.g. Ebbinge and Spaans 1995; Gunnarsson et al. 2005; Legagneux et al. 2011).

Differences in breeding biology complicate comparisons between these groups and

single-species work is needed. Yellow Warblers have one of the largest breeding ranges

of all wood-warblers (Parulidae), making them an ideal species to examine latitudinal

variation in the strength of carry-over effects. We have previously demonstrated that

stable 13C and 15N isotope signatures in winter-grown tissue is related to habitat use

among Yellow Warblers wintering in Mexico and that winter habitat quality influences the

breeding performance of young females within a mid-latitude population (50°N,

Revelstoke, BC; Drake et al. 2013; Chapter 3). Here, young females with isotopic

signatures consistent with the use of dry winter habitat (enriched 13C) experienced

breeding delays and lower productivity than those with signatures suggestive of more

mesic winter habitat use (Drake et al. 2013; Chapter 3). Birds breeding in northwestern

regions of North America (such as Revelstoke) share their wintering range with birds

which breed at higher latitudes (Boulet et al. 2006). We therefore asked whether carry-

over effects in Yellow Warblers breeding at the extreme north of the species range in

Inuvik, Northwest Territories, Canada (68°N) were stronger or weaker than those

indicated in these conspecifics further south.

We hypothesized that winter habitat quality could carry-over to impact breeding

success in our northern breeding birds through two, non-exclusive, mechanisms:

individual timing (time-mediated) and individual condition (condition-mediated) (Figure

70

4.1). In the first scenario, individuals occupying high-quality winter habitat are able to

fatten more rapidly and depart the wintering grounds earlier in the spring (Marra et al.

1998; Smith et al. 2010 (departure inferred)). This allows them to arrive at the breeding

grounds earlier (Marra et al. 1998; Norris et al. 2004) and obtain the productivity

advantages that are associated with advanced breeding phenology (Lozano et al. 1996;

Smith and Moore 2005; Reudink et al. 2009). In the second scenario, individuals

occupying high-quality winter habitat are in better physical condition than their

counterparts when they reach the breeding grounds. For males, larger energy reserves,

more muscle mass, or higher quality plumage could offer an advantage in territory

acquisition/defense and influence female choice, both of which should result in more

rapid pairing (Studd and Robertson 1985; Gottlander 1987; Yezerinac and Weatherhead

1997). For females, larger energy reserves might result in more rapid egg-laying

(Sandberg and Moore 1996 but see Smith and Moore 2003). Both of these factors could

advance breeding phenology in both sexes from the point of clutch initiation rather than

from arrival. In addition, habitat-based variation in parental condition could have a direct

influence on fledging state and number because females in good condition may produce

larger, higher quality clutches (Smith and Moore 2003) and parents in good condition

may provision at a higher rate. We evaluated evidence for time- and condition-mediated

carry-over effects in Yellow Warblers breeding in Inuvik using information on winter

habitat use inferred from 13C and 15N signatures in winter-grown feathers, the dates of

male arrival on territories, the timing of breeding, and number and condition of offspring

produced by breeding pairs. We subsequently evaluated whether our results differed

from those reported in the carry-over effect literature for Yellow Warblers and American

Redstarts breeding in British Columbia and Ontario (Drake et al. 2013; Norris et al.

2004).

Methods

Study Species

Yellow Warblers are insectivorous Nearctic-Neotropical migrants in the family

Parulidae. During the summer they are found in riparian and wet deciduous habitat

throughout North America as far north as the Arctic tree-line (Lowther et al. 1999).

71

Genetic and isotopic work suggest that eastern and western populations undergo

parallel migration with western populations wintering in Mexico and Central America and

eastern populations wintering in Central and South America (Boulet et al. 2006).

Substantial wintering overlap is seen in mid- and high-latitude populations in the west

(Boulet et al. 2006) indicating that northern populations migrate further to reach their

breeding grounds than mid-latitude populations.

Overwinter, Yellow Warblers occupy native habitat ranging from mangrove,

swamp and riparian forests to dry coastal scrub as well as a diversity of human-modified

habitat such as pasture and cropland (Dunn et al. 1997; Greenberg et al. 1996; Binford

1989). The species undergoes pre-alternate moult prior to spring migration, replacing a

variable number of greater covert, body, and crown feathers (Pyle 1997; Quinlan and

Green 2010). Yellow Warblers are territorial on the wintering grounds (Greenberg and

Ortiz 1994, Valdez Juarez, unpublished data) and the habitat used by an individual

influences the 13C and 15N signatures in their blood (Drake et al. 2013; Chapter 3) and

ultimately the feathers they grow during this period (Hobson 1999).

Depleted 13C is indicative of more mesic habitat use (Hobson et al. 2010).

Previous work suggests this type of habitat is of higher quality for many wintering

warblers (Studds and Marra 2005; Latta and Faaborg 2002; Smith et al. 2011) including

Yellow Warblers (Drake et al. 2013; Chapter 3). 15N ratios are influenced by regional

temperature and precipitation (Amundson et al. 2003), and, at the local-level, by factors

such as soil cultivation, and marine and artificial fertilizer inputs (Hobson 1999;

Amundson et al. 2003; Croll et al. 2005). 15N in animal tissue may become further

enriched with water and/or nutritional stress (Hobson et al. 1993; Kelly 2000) and with

trophic level (Post 2002). Among Yellow Warblers overwintering in Mexico, the greatest

variation in tissue 15N was regional and we suggest that depleted 15N is indicative of

wintering origins in higher rainfall regions such as eastern Mexico and southern Central

America, while enriched 15N suggests wintering origins in drier northern and Pacific

coastal regions of the Yellow Warbler wintering range (Drake et al. 2013; Chapter 3).

We predicted that the relationship between 13C, 15N and breeding phenology

and performance in Inuvik would be the same as in our southern study (Drake et al.

2013; Chapter 3): winter habitats conferring more depleted 13C and more enriched 15N

72

signatures would be associated with earlier arrival and clutch initiation dates and better

breeding performance by individual birds.

Study site and population monitoring

Between 2009 and 2011, we monitored a banded population of Yellow Warblers

in the taiga plains ecozone near Inuvik, Northwest Territories (68º 21’ N, 133º 45’ W;

Figure 1.1; hereafter Inuvik). Our study site covered 20.0 ha of riparian habitat bordering

the Mackenzie River (3-6 m elevation). Low-lying portions of this habitat are dominated

by several species of willow shrubs (Salix spp.) and by Green Alder (Alnus crispa) all of

which experience seasonal flooding during spring ice-break. These species are

intermixed with White Spruce (Picea glauca) at higher elevations (Gill 1973). Yellow

Warblers experience an average of 23 hrs of daylight per day during the breeding period

and, in the years of our study, mean daily temperatures of 11.5 ± 0.4°C, range: -2.2-

25.5°C (Environment Canada 2012).

Dates of male arrival on breeding territories were determined during surveys

conducted every 1-2 days beginning in mid-May. Male Yellow Warblers sing when

establishing territorial boundaries making them easy to detect. Males are also highly

aggressive toward song-playback and were typically caught within 1-2 days of arrival

using targeted mist-netting. Females were more difficult to detect, but were either caught

with their mates or later in the season by placing mist-nets across flight paths used

during nest-building, incubation, or while provisioning young. Upon capture, all

previously unmarked individuals were fitted with a Canadian Wildlife Service-issued

aluminum band and a unique combination of three colour bands (AC Hughes, UK).

Breeding pairs were monitored through the entire season (mid-May to late-July).

Nests were located by following females during the building period and then checked

every 3 days to determine clutch initiation dates, clutch size, brood size and fledgling

number. Clutch initiation date was defined as the laying date of the first egg in a female’s

first nesting attempt of the season. Nestlings were banded, weighed and had tarsus

measures taken seven days after hatching. We calculated the average condition of

nestlings within a clutch by calculating the mean tarsus length (mm) and mass (g) for all

individuals in the nest. We then regressed mean mass against mean tarsus for the

73

population and used the residuals (i.e. body-size controlled mass) as a unit-less metric

of average clutch ‘condition’. We assumed that all nestlings present on day 7

successfully fledged unless signs of predation were found in or around the nest after

fledging. Successful fledging was usually confirmed by observing parents feeding or

defending young out of the nest. Annual productivity for an individual bird was defined as

the number fledglings produced over all nesting attempts. Although rare, some males

(4%, n=5/128 males over 3 years) obtained two social mates, and it was possible for

their seasonal productivity to be higher than that of females.

Sexing and aging

Birds were sexed using breeding characteristics (brood patch/cloacal

protuberance), plumage (brightness/degree of rufous streaking) and behaviour. They

were then classified as yearling or older birds based on plumage. Yearlings (“second

year” or SY birds) have narrow, tapered primary coverts, tapered retrices with less

yellow colouration on the inner webs of the outer tail, and worn prebasic plumage; older

(“after-second year” or ASY) birds have broad, truncate primary covets with narrow

yellow-olive edging, and truncate retrices with more extensive yellow colouration on the

inner webs (Pyle 1997).

Tissue samples and stable isotope analysis

To obtain winter habitat 13C and 15N signatures we collected three newly

moulted greater covert feathers (identifiable by their brighter pigment and lack of wear)

from all individuals captured in 2009-2011. Feathers were washed in 2:1

chloroform:methanol solution for 24hrs, drained, and then air-dried in a fume hood for an

additional 24h to remove excess solvent. 1mg (±0.2) of tissue was placed in 9x5mm

smooth-walled tin capsules (Elemental Microanalysis, UK) and samples were then

analyzed using a PDZ Europa ANCA-GSL elemental analyzer interfaced to a PDZ

Europa 20-20 isotope ratio mass spectrometer (Sercon Ltd., Cheshire, UK) at the

University of California Davis Stable Isotope Facility in California, USA. Delta values are

expressed relative to international standards: V-PDB (Vienna PeeDee Belemnite) for

carbon and air for nitrogen; estimated measurement precision was 0.2‰ for 13C and

0.3‰ for 15N.

74

Carry-over effects

We used general linear models (JMP 9.0.2, SAS Institute Inc.) to evaluate the

effect of winter habitat use on: male arrival date, the onset of breeding in males and

females, and productivity and offspring condition for both genders. Predicted habitat

effects follow the pathway illustrated in Figure 4.1. We did not examine the relationship

between winter habitat and female arrival as lower detection probabilities and later

banding dates made female arrival dates less certain and female identity unreliable until

nest construction had begun. Clutch initiation is therefore used as proxy for arrival date

in females. Connecting winter habitat use directly to clutch initiation merges condition-

vs. timing-mediated effects in the female pathway.

All models included ‘Age’ and ‘Year’ effects as covariates as we predicted that

timing effects could vary between age-classes and among years. We also predicted that

the relationships between timing variables could vary with age and year and included

‘Arrival×Age’ and ‘Arrival×Year’ in models predicting male clutch initiation dates, and

‘Clutch Initiation×Age’ and ‘Clutch Initiation×Year’ in models predicting fledgling number

and condition. Carry-over effect strength has been shown to vary by age classes (Drake

et al. 2013; Chapter 3) and by year (Norris et al. 2004). We therefore included ‘13C /15N

×Age’ and ‘13C/ 15N ×Year’ interaction terms in all models.

Contributing variables for each step in the path analysis were selected using a

stepwise procedure where the probability to enter and to leave the model was ≤ 0.05

and ≥ 0.1 respectively. Interaction terms were restricted such that they could only enter

the model with their precedent terms. We determined the contribution of 13C and 15N to

model fit by independently reintroducing them to the final model if they were dropped

during the stepwise procedure.

Power to detect carry-over effects observed at mid-latitude sites

We used the linear multiple regression test in G*Power 3.1 (Faul et al. 2007) to

calculate the power of our Inuvik data to detect isotope effect sizes equivalent to those

found in warbler populations breeding at mid-latitude sites. First, we calculated our

power to detect a 13C and 15N effect equivalent to that found in young female Yellow

Warblers breeding in Revelstoke, British Columbia (50ºN) (Drake et al. 2013; Chapter 3).

75

In Revelstoke, variation in 13C and 15N values independently explained 11.1% and

8.9% of the variation in young female clutch initiation dates. Jointly these isotopes

explained 17.1% of the variation in this data (n=38; for detailed study description see

Drake et al. 2013; Chapter 3). We used these values and the residual variance in

reduced Inuvik models that included only young females (‘Year + 13C or 15N’ and

‘Year’) to calculate the predicted effect size (f2) of 13C and 15N in Inuvik if these

isotopes explained an equivalent amount of variance as documented in Revelstoke. We

then used f2 and the sample size of young females in Inuvik to calculate Power (1- β).

We additionally calculated our power to detect a 13C effect equivalent to that shown in

older American Redstart males at Chaffey’s Lock, Ontario (44ºN; Norris et al. 2004).

Here variation in 13C explained 32% of the variation in arrival dates (n=57). We used

this value and the residual variance in arrival date within a reduced Inuvik model (‘Year’)

for older males to calculate f2. We then used the sample size of older males in Inuvik to

calculate Power (1- β).

We also compared the strength of the relationship between clutch initiation date

and productivity for males and females in Inuvik to that shown within our mid-latitude site

by combining Inuvik breeding data with breeding data collected in Revelstoke (for

detailed study description see Drake et al. 2013) in the same manner and over the same

time period (2009-2011; n=69 females and 76 males). We used a multiple linear

regression analysis to test for a ‘Location×Clutch Initiation’ interaction in models

predicting fledgling number after accounting for ‘Year’ and ‘Year×Location’ effects.

Results

Population traits

Breeding density in Inuvik was 3.8 pairs/ha. This was higher than in our

comparison population in Revelstoke, B.C. (0.7 pairs/ha; Drake et al. 2013; Chapter 3)

but lower than that reported for some other mid-latitude populations (e.g. 10 pairs/ha;

Mazerolle et al. 2005). First breeding season (SY) individuals accounted for 41% of the

female breeding population (range 38-47%, n=3 years) and 10% of the male breeding

76

population (range 9-12%, n=3 years). Breeding pairs fledged an average of 3.3±2.1

offspring per year (n=130 pairs over 3 years).

Male and female Yellow Warblers in Inuvik did not differ significantly in their

winter 13C signatures (13C (±SD): -23.3 ± 2.0 vs. -23.1 ± 2.3, t(229)=-0.81, P=0.42.

Males, however, had more enriched winter 15N signatures than females (15N (±SD):

8.9 ± 1.5 vs. 8.4 ± 1.4, t(230)=2.64, P(two-tailed)=0.009). Wintering signatures of young

and older birds did not differ (13C: ‘Age’, P=0.52, ‘Age×Sex’, P=0.60; 15N: ‘Age’,

P=0.40, ‘Age×Sex’, P=0.92). Winter 13C and 15N signatures were not correlated

(Spearman’s ρ=0.10, P=0.12, n=231).

Carry-over effects

13C and 15N signatures in winter-grown tissue showed no relationship to either

male arrival or female clutch initiation date. Although local weather conditions in May

were not notably different among years, male arrival dates were later in 2009 than in

2010 or 2011 (‘Year’, Table 4.1). Female clutch initiation dates were delayed in 2011 in

response to a period of freezing temperatures in early June (‘Year’, Table 4.1). After

controlling for arrival date, males with enriched 15N signatures obtained mates who

initiated clutches earlier than the mates obtained by males with depleted 15N signatures

(P=0.01; Table 4.1). Male 13C signatures showed no direct relationship with clutch

initiation date.

13C and 15N signatures were not directly related to the number of fledglings

produced by either sex. Productivity of males and females varied among years and was

significantly lower in 2011 when sub-zero temperatures in June killed eggs and led to the

abandonment of clutches initiated before the freeze (Table 4.1). Unexpectedly, we found

little evidence that the productivity of male or female warblers declined significantly with

later clutch initiation dates (Table 4.1). There was some suggestion that, in males,

seasonal effects on productivity varied among years (‘Year×Clutch Initiation’, P=0.06;

Table 4.1). However, this interaction was not caused by seasonal declines in productivity

in any of the three years, but rather a positive relationship between clutch initiation date

and male productivity in 2011 when males whose social mates initiated their first clutch

after the freeze experienced greater success.

77

The 13C signature of either social mate was not directly related to the average

condition of their nestlings at day 7 post-hatch. For males, none of the variables we

considered explained any of the variation in nestling condition (Table 4.1). Among

females, there was an indication that 15N influenced nestling condition in some years

but not others (‘Year×15N’, P=0.05; Table 4.1). This interaction was the product of a

negative relationship between 15N and nestling condition in 2010. Overall, the model did

not explain a significant amount of the variation in the dataset.

Comparison with mid-latitude sites

Our power to detect winter habitat effects on the breeding phenology of young

female Yellow Warblers in Inuvik of the same magnitude as that observed in young

females in Revelstoke, British Columbia was high (13C alone: 1-β=0.97; 15N alone: 1-

β=0.92; joint effect: 1-β=1.0; n=50). Similarly a winter habitat effect on arrival date

among older males in Inuvik of comparable strength to that observed in older male

American Redstarts in Chaffy’s Lock, Ontario would have been detectable given our

sample sizes (1-β=1.0; n=106).

The relationship between clutch initiation date and productivity differed

significantly between Inuvik and Revelstoke, confirming that the timing of breeding in

Inuvik had little impact on productivity. The number of offspring fledged decreased with

clutch initiation date in Revelstoke but did not vary with clutch initiation date in Inuvik

(‘Location×Clutch Initiation’: males, P=0.02; females, P=0.03; Table 4.2, Figure 4.2).

Discussion

We predicted that carry-over effects of winter habitat use would be more

pronounced among migrants breeding the arctic. We anticipated this because northern

breeders travel further during migration and encounter harsher environmental conditions

upon arrival. Both factors should result in higher energetic costs. Migrants in the north

additionally experience a shorter breeding season, which should strongly penalize

delayed reproduction. Instead we found little support for an effect of winter-habitat use

on the breeding phenology or productivity of Yellow Warblers in the far north. Our study

78

instead offers convincing support for the concept that increasing migration distance

dampens carry-over effects.

We found no evidence that 13C and 15N signatures in winter grown tissue

influenced male arrival, female clutch initiation dates, or pair productivity in Inuvik. Power

analysis indicated that 13C effects in older males equivalent to those found in American

Redstarts breeding at mid-latitudes in eastern Canada (Norris et al. 2004) would have

been detectable if present. The absence of a winter habitat effect on older males in our

northern population was not unexpected as we also found no support for such an effect

among older male Yellow Warblers breeding further south in Revelstoke, British

Columbia (Drake et al. 2013; Chapter 3). For young females, our ability to detect a

relationship between 13C, 15N and clutch initiation date equivalent to that found in

Revelstoke (Drake et al. 2013; Chapter 3) was also high. Differences in the relationship

between 13C and arrival date in adult male warblers between populations in eastern and

western North America may relate to differences in migration routes rather than

migration distance (see Chapter 5). However, differences in the relationship between

13C and 15N and young female clutch initiation dates between our arctic site and

Revelstoke would suggest that migration distance reduces the importance of wintering

habitat use.

Evidence of condition-mediated carry-over effects in Inuvik was limited. 13C had

no effect on the speed with which males obtained a mate and initiated reproduction. 13C

and 15N also had no impact on male or female productivity or nestling condition after

controlling for breeding phenology. Enriched 15N signatures were associated with

earlier clutch initiation in males once arrival date was accounted for. Enriched feather

15N is associated with higher feather chroma in Inuvik (Jones, Drake and Green,

unpublished data). This plumage trait may have enabled 15N-enriched males to pair

faster, but it remains unclear why 15N and chroma itself are related. Despite advances

in breeding phenology through clutch initiation date for 15N enriched males, this shift did

not represent a condition mediated carry-over effect as males did not experience higher

productivity with earlier clutch initiation. Similarly, clutch initiation date in Inuvik was not

directly associated with nestling condition. Such an effect might have produced a carry-

over effects if nestlings that leave the nest in better condition are more likely to survive

79

and recruit into the breeding population (Magrath 1991). We could not test two additional

avenues through which winter habitat use might impact fitness. It is possible that 15N-

enriched males experience higher rates of extra-pair paternity if advanced pairing is the

product of increased attractiveness. It is also possible that rapid pairing is the product of

rapid territory acquisition. This may mean that 15N-enriched males experience less

male-male conflict, which could in turn influence survival.

It was expected that individuals who arrived on the breeding grounds and

initiated clutches early would have higher productivity than individuals arriving later in the

breeding season (e.g. Smith and Moore 2005; Rockwell et al. 2012). We did not observe

this pattern in Inuvik Yellow Warblers. Clutch initiation date in Inuvik was not significantly

correlated with productivity even after excluding data from 2011, where freezing

temperatures penalized individuals who began nesting earlier in the season (males:

Spearman’s ρ=-0.16, P=0.15, n=83; females: Spearman’s ρ=-0.14, P=0.25, n=75). In

contrast, declines in seasonal productivity are observed among Yellow Warblers in

Revelstoke (Drake et al. 2013; Chapter 3). That seasonal productivity declines in Inuvik

were indeed weaker than those in more southerly populations was confirmed by

significant slope differences in the relationship between clutch initiation and fledgling

number between Revelstoke and Inuvik (Table 4.2). This result is similar to that reported

for ptarmigan (Lagopus spp.) where later first clutch dates had less of an influence on

clutch size for birds breeding in the arctic than for those in the alpine at lower latitudes

(Martin and Wiebe 2004) but counter to consistent seasonal declines in first clutch size

reported for American Pipits (Anthus rubescens) across latitudes (Hendricks 1997).

There are several reasons why productivity declines might not be observed in our Inuvik

population. First, inhospitable conditions in the early spring may mean that northern

birds arrive and initiate reproduction more synchronously than more southerly

populations. In the three years of our study, Inuvik birds initiated nests over a 22-day

period while Revelstoke birds began to breed a week earlier in the season and initiated

nests over a 27-day period (Figure 4.2). This reduced spread in clutch initiation dates in

Inuvik would reduce productivity differences among individuals if timing effects were

present. Secondly, Inuvik Yellow Warblers, like other birds breeding in the north,

experience lower predation rates than more southern populations (Martin 2013;

McKinnon et al. 2010). This means that fewer first-nests fail (39% (Inuvik, n=128 nests

over 3 years) vs. 52% (Revelstoke, n=152 nests over 6 years)) and, consequently,

80

additional time for re-nesting is less of an advantage for the majority of individuals.

Among females in Inuvik who failed in their first nesting attempt, individuals who initiated

their first clutch later in the season were less likely to successfully fledge a subsequent

clutch (χ21,45=12.9, P<0.001), however, these individuals represented too small a

proportion of the breeding population to influence overall trends. Yellow Warblers in

Inuvik are almost always single-brooded; in three years only one female in our study

reared a second clutch after successfully fledging a first. Fledging first nests early

therefore does not result in increased productivity through second broods. Finally, clutch

size in Inuvik is a quadratic function of clutch initiation date, with early clutches

containing fewer eggs than mid-season clutches (Martin, 2013). This pattern suggests

that early clutch initiation dates are costly, possibly due to the shorter window of suitable

climatic conditions and food availability in the north. Small, early clutch-sizes followed by

a mid-season peak and decline may flatten any negative linear relationship between

clutch initiation date and fledge number.

Yellow warblers breeding in Inuvik (particularly males) had more depleted 13C

signatures (indicative of more mesic habitat use (Hobson 2007)) and higher productivity

than reported for Yellow Warblers breeding in Revelstoke (Martin 2013). Mesic habitat is

considered to be higher quality habitat for wintering warblers relative to drier habitats

(e.g. Bearhop et al. 2004; Studds and Marra 2005; Smith et al. 2010) and signature

differences could indicate that Yellow Warblers in Inuvik are more socially dominant on

the wintering grounds than Revelstoke birds. This is similar to seasonal matching

documented among Black-tailed Godwits (Limosa limosa islandica) in the U.K., where

individuals from high quality winter habitat also occupy high quality breeding locations.

However, seasonal matching has not been documented in songbirds that breed across a

large latitudinal gradient and wintering isotope signatures and productivity data from

multiple breeding populations across the Yellow Warbler range would be needed to test

this concept.

In conclusion, we found no evidence of an effect of wintering habitat use on

individual productivity among Yellow Warblers at northerly extreme of their range. This

was due to (1) any potential effect of winter habitat use on breeding phenology being

weaker than that observed among warblers at mid-latitudes and (2) productivity patterns

in the north that undermine the time-mediated pathway through which carry-over effects

81

have previously been shown to act in passerines (Norris et al. 2004, Saino et al. 2004,

Drake et al. 2013; Chapter 3). We found some support for a non-breeding season effect

on male condition, which advanced male breeding phenology from the point of clutch

initiation. This effect did not translate into increased productivity or improved nesting

condition. In total, the absence of a detectable winter-habitat effect on fledgling number

in Inuvik indicates that timing effects are weaker in this population than in populations

where declines in fledgling number have been shown. Increased migratory distance

therefore appears to dampen the impact of winter habitat use on breeding phenology

while productivity advantages associated with breeding in the far north appear to reduce

the role of phenology in breeding success.

References

Alerstam, Thomas, and Pehr H. Enckell. 1979. Unpredictable Habitats and Evolution of Bird Migration. Oikos 33 (2): 228–232. doi:10.2307/3543999.

Alerstam, Thomas, Anders Hedenström, and Susanne Åkesson. 2003. Long-distance Migration: Evolution and Determinants. Oikos 103 (2): 247–260. doi:10.1034/j.1600-0706.2003.12559.x.

Amundson, Ronald, A. T. Austin, E. A. G. Schuur, K. Yoo, V. Matzek, C. Kendall, A. Uebersax, D. Brenner, and W. T. Baisden. 2003. Global Patterns of the Isotopic Composition of Soil and Plant Nitrogen. Global Biogeochemical Cycles 17: 1031-1041. doi:200310.1029/2002GB001903.

Bearhop, Stuart, Geoff M Hilton, Stephen C Votier, and Susan Waldron. 2004. Stable Isotope Ratios Indicate that Body Condition in Migrating Passerines isInfluenced by Winter Habitat. Proceedings of the Royal Society B: Biological Sciences 271 (Suppl 4): S215–S218.

Bell, Christopher P. 1996. Seasonality and Time Allocation as Causes of Leap-Frog Migration in the Yellow Wagtail Motacilla flava. Journal of Avian Biology 27 (4): 334–342. doi:10.2307/3677265.

Binford, Laurence C. 1989. A Distributional Survey of the Birds of the Mexican State of Oaxaca. Ornithological Monographs 43: 392–418.

Boulet, Marylène, H. Lisle Gibbs, and Keith A. Hobson. 2006. Integrated Analysis of Genetic, Stable Isotope, and Banding Data Reveal Migratory Connectivity and Flyways in the Northern Yellow Warbler (Dendroica petechia; Aestiva Group).

Ornithological Monographs (61): 29–78.

82

Cox, George W. 1985. The Evolution of Avian Migration Systems Between Temperate and Tropical Regions of the New World. The American Naturalist 126 (4): 451–474. doi:10.2307/2461532.

Croll, D. A., J. L. Maron, J. A. Estes, E. M. Danner, and G. V. Byrd. 2005. Introduced Predators Transform Subarctic Islands from Grassland to Tundra. Science 307 (5717): 1959–1961. doi:10.1126/science.1108485.

Drake, A., C. Rock, S. P. Quinlan, and D. J. Green. 2013. Carry-over effects of winter habitat vary with age and sex in Yellow Warblers (Setophaga petechia). Journal of Avian Biology, in press. doi: 10.1111/j.1600-048X.2013.05828.x.

Dunn, Jon, and Kimball Garrett. 1997. A Field Guide to Warblers of North America. Houghton Mifflin Harcourt.

Ebbinge, Barwolt S., and Bernard Spaans. 1995. The Importance of Body Reserves Accumulated in Spring Staging Areas in the Temperate Zone for Breeding in Dark-Bellied Brent Geese Branta b. bernicla in the High Arctic. Journal of Avian

Biology 26 (2): 105–113. doi:10.2307/3677058.

Environment Canada. 2012. National Climate Data and Information Archive, Environment Canada. www.climate.weatheroffice.gc.ca/climateData.

Faul, Franz, Edgar Erdfelder, Albert-Georg Lang, and Axel Buchner. 2007. G*Power 3: a Flexible Statistical Power Analysis Program for the Social, Behavioral, and Biomedical Sciences. Behavior Research Methods 39 (2): 175–191.

Gill, Don. 1973. Ecological Modifications Caused by the Removal of Tree and Shrub Canopies in the Mackenzie Delta. Arctic 26 (2): 95–111.

Gottlander, Karin. 1987. Variation in the Song Rate of the Male Pied Flycatcher Ficedula hypoleuca: Causes and Consequences. Animal Behaviour 35 (4): 1037–1043.

doi:10.1016/S0003-3472(87)80160-4.

Greenberg, Russell, and Javier Salgado Ortiz. 1994. Interspecific Defense of Pasture Trees by Wintering Yellow Warblers. The Auk 111 (3): 672–682.

Greenberg, Russell, Robert Reitsma, and Andrea Cruz Angon. 1996. Interspecific Aggression by Yellow Warblers in a Sun Coffee Plantation. The Condor 98 (3): 640–642. doi:10.2307/1369579.

Gunnarsson, Tómas Grétar, Jennifer A Gill, Jason Newton, Peter M Potts, and William J Sutherland. 2005. Seasonal Matching of Habitat Quality and Fitness in a Migratory Bird. Proceedings of the Royal Society B: Biological Sciences 272 (1578): 2319 –2323. doi:10.1098/rspb.2005.3214.

Hendricks, Paul. 1997. Geographical Trends in Clutch Size: A Range-Wide Relationship with Laying Date in American Pipits. The Auk 114 (4): 773–778. doi:10.2307/4089300.

83

Hobson, K. A. 1999. Stable-carbon and nitrogen isotope ratios of songbird feathers grown in two terrestrial biomes : Implications for evaluating trophic relationships and breeding origins. The Condor 101 (4): 799–805.

Hobson, Keith A. 1999. Tracing Origins and Migration of Wildlife Using Stable Isotopes: a Review. Oecologia 120: 314–326. doi:10.1007/s004420050865.

———. 2007. Isotopic Tracking of Migrant Wildlife. In Stable Isotopes in Ecology and Environmental Science, edited by Robert H. Michener and Kate Lajtha, 2nd ed., 155–175. Wiley-Blackwell.

Hobson, Keith A., Ray T. Alisauskas, and Robert G. Clark. 1993. Stable-Nitrogen Isotope Enrichment in Avian Tissues Due to Fasting and Nutritional Stress: Implications for Isotopic Analyses of Diet. The Condor 95 (2): 388–394. doi:10.2307/1369361.

Hobson, Keith A., Rachel Barnett-Johnson, and Cerling Thure. 2010. Using Isoscapes to Track Animal Migration. In Isoscapes: Understanding Movement, Pattern, and Process on Earth Through Isotope Mapping, 273–298.

Kelly, Jeffrey F. 2000. Stable Isotopes of Carbon and Nitrogen in the Study of Avian and Mammalian Trophic Ecology. Canadian Journal of Zoology 78: 1–27. doi:10.1139/z99-165.

Latta, S C, and J. Faaborg. 2002. Demographic and Population Responses of Cape May Warblers Wintering in Multiple Habitats. Ecology 83 (9): 2502–2515.

Legagneux, Pierre, Peter L. F. Fast, Gilles Gauthier, and Joël Bêty. 2011. Manipulating Individual State During Migration Provides Evidence for Carry-over Effects Modulated by Environmental Conditions. Proceedings of the Royal Society B: Biological Sciences: 279: 876-883. doi:10.1098/rspb.2011.1351.

Lowther, P. E., C. Celada, N. K. Klein, C. C. Rimmer, and D. A. Spector. 1999. Yellow Warbler (Dendroica petechia). Edited by A. Poole and F. Gill. The Birds of North

America Online. doi:10.2173/bna.454.

Lozano, G. A., S. Perreault, and R. E. Lemon. 1996. Age, Arrival Date and Reproductive Success of Male American Redstarts Setophaga ruticilla. Journal of Avian

Biology 27 (2): 164–170. doi:10.2307/3677146.

Magrath, Robert D. 1991. Nestling Weight and Juvenile Survival in the Blackbird, Turdus Merula. Journal of Animal Ecology 60 (1): 335–351. doi:10.2307/5464.

Marra, Peter P., Keith A. Hobson, and Richard T. Holmes. 1998. Linking Winter and Summer Events in a Migratory Bird by Using Stable-Carbon Isotopes. Science 282 (5395): 1884 –1886. doi:10.1126/science.282.5395.1884.

Martin, Kathy, and Karen L. Wiebe. 2004. Coping Mechanisms of Alpine and Arctic Breeding Birds: Extreme Weather and Limitations to Reproductive Resilience. Integrative and Comparative Biology 44 (2): 177–185. doi:10.1093/icb/44.2.177.

84

Martin, Michaela. 2013. Latitudinal variation in life history traits and incubation patterns in Yellow Warblers. M.Sc., Simon Fraser University.

Mazerolle, Daniel F., Kevin W. Dufour, Keith A. Hobson, and Heidi E. den Haan. 2005. Effects of Large-scale Climatic Fluctuations on Survival and Production of Young in a Neotropical Migrant Songbird, the Yellow Warbler Dendroica petechia. Journal of Avian Biology 36 (2): 155–163. doi:10.1111/j.0908-8857.2005.03289.x.

McKinnon, L., P. A. Smith, E. Nol, J. L. Martin, F. I. Doyle, K. F. Abraham, H. G. Gilchrist, R. I. G. Morrison, and J. Bêty. 2010. Lower Predation Risk for Migratory Birds at High Latitudes. Science 327 (5963): 326–327. doi:10.1126/science.1183010.

Newton, Ian. 2008. The Migration Ecology of Birds. Academic Press.

Norris, D. Ryan, Peter P. Marra, T. Kurt Kyser, Thomas W. Sherry, and Laurene M. Ratcliffe. 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:10.1098/rspb.2003.2569.

Post, David M. 2002. Using stable isotopes to estimate trophic position: Models, methods, and assumptions. Ecology 83 (3): 703–718.

Pyle, Peter. 1997. Identification Guide to North American Birds, Part I: Columbidae to Ploceidae. Slate Creek Press.

Quinlan, Sam P., and David J. Green. 2010. 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:10.1007/s10336-010-0553-5.

Reudink, Matthew W, Peter P Marra, T. Kurt Kyser, Peter T Boag, Kathryn M Langin, and Laurene M Ratcliffe. 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:10.1098/rspb.2008.1452.

Rockwell, Sarah M., Carol I. Bocetti, and Peter P. Marra. 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:10.1525/auk.2012.12003.

Saino, Nicola, Tibor Szép, Roberto Ambrosini, Maria Romano, and Anders Pape Møller. 2004. Ecological Conditions During Winter Affect Sexual Selection and Breeding in a Migratory Bird. Proceedings of the Royal Society B: Biological Sciences 271 (1540): 681–686.

85

Salgado-Ortiz, Javier, Peter P. Marra, T. Scott Sillett, and Raleigh J. Robertson. 2008. Breeding Ecology of the Mangrove Warbler (Dendroica petechia bryanti) and

Comparative Life History of the Yellow Warbler Subspecies Complex. The Auk 125 (2): 402–410. doi:10.1525/auk.2008.07012.

Sandberg, Roland. 1996. Fat Reserves of Migrating Passerines at Arrival on the Breeding Grounds in Swedish Lapland. Ibis 138 (3): 514–524. doi:10.1111/j.1474-919X.1996.tb08072.x.

Sandberg, Roland, and Frank R. Moore. 1996. Fat Stores and Arrival on the Breeding Grounds: Reproductive Consequences for Passerine Migrants. Oikos 77 (3): 577–581. doi:10.2307/3545949.

Sherry, Thomas W., and Richard T. Holmes. 1996. Winter Habitat Quality, Population Limitation, and Conservation of Neotropical-Nearctic Migrant Birds. Ecology 77 (1): 36–48. doi:10.2307/2265652.

Smith, Joseph A. M., Leonard R. Reitsma, and Peter P. Marra. 2010. Moisture as a Determinant of Habitat Quality for a Nonbreeding Neotropical Migratory Songbird. Ecology 91: 2874–2882. doi:10.1890/09-2212.1.

———. 2011. Multiple Space-Use Strategies and Their Divergent Consequences in a Nonbreeding Migratory Bird (Parkesia noveboracensis). The Auk 128: 53–60. doi:10.1525/auk.2011.10241.

Smith, Robert J, and Frank R Moore. 2003. Arrival Fat and Reproductive Performance in a Long-distance Passerine Migrant. Oecologia 134 (3): 325–331. doi:10.1007/s00442-002-1152-9.

Smith, Robert J., and Frank R. Moore. 2005. Arrival Timing and Seasonal Reproductive Performance in a Long-Distance Migratory Landbird. Behavioral Ecology and Sociobiology 57 (3): 231–239.

Studd, Micheal V., and Raleigh J. Robertson. 1985. Evidence for Reliable Badges of Status in Territorial Yellow Warblers (Dendroica petechia). Animal Behaviour 33

(4): 1102–1113. doi:10.1016/S0003-3472(85)80169-X.

Studds, Colin E., and Peter P. Marra. 2005. Nonbreeding Habitat Occupancy and Population Processes: An Upgrade Experiment with a Migratory Bird. Ecology 86 (9): 2380–2385. doi:10.1890/04-1145.

Sutherland, William J. 1996. From Individual Behaviour to Population Ecology. Oxford University Press.

Wikelski, Martin, Elisa M. Tarlow, Arlo Raim, Robert H. Diehl, Ronald P. Larkin, and G. Henk Visser. 2003. Costs of migration in free-flying songbirds. Nature 423 (6941): 704–704.

86

Yezerinac, Stephen M., and Patrick J. Weatherhead. 1997. Extra–pair Mating, Male Plumage Coloration and Sexual Selection in Yellow Warblers (Dendroica petechia). Proceedings of the Royal Society of London. Series B: Biological

Sciences 264 (1381): 527 –532. doi:10.1098/rspb.1997.0075.

87

Tables and Figures

88

Table 4.1 Multiple regression analyses testing the influence of winter habitat

(inferred from 13C and 15N) on Inuvik Yellow Warbler breeding metrics. For predicted relationships see Figure 4.1. Age and Year plus second-order interaction terms were included in initial models and dropped sequentially if they did not contribute to fit. Below

each model we report the contribution of 13C and 15N to model fit (italics) if these variables were to be independently added back into the final model.

89

Table 4.2 Location effects on the relationship between first-clutch initiation date and productivity in Yellow Warblers from the perspective of (A) breeding males and (B) breeding females. For both sexes the best fitting model predicts a negative relationship at our mid-latitude (Revelstoke) site and no relationship at our high-latitude (Inuvik) site. See also Figure 4.2.

A) Males r2 df, n p Std B F ratio Prob>F

Fledge number: 0.15 7, 199 <0.001

Location (Revelstoke-Inuvik) 0.36 20.50 <0.001

Clutch Initiation -0.16 4.17 0.05

Location×Clutch Initiation 0.18 5.41 0.02

Year - 1.80 0.17

Year×Clutch Initiation - 0.77 0.46

B) Females r2 df, n p Std B F ratio Prob>F

Fledge number: 0.13 7, 190 <0.001

Location (Revelstoke-Inuvik) 0.32 13.76 <0.001

Clutch Initiation -0.16 3.78 0.05

Location×Clutch Initiation 0.19 4.99 0.03

Year - 2.99 0.05

Year×Clutch Initiation - 0.53 0.59

90

Figure 4.1 Suggested pathways through which winter-habitat-driven carry-over effects could act. For females in our study population, clutch initiation date was used as a proxy for arrival date (see text) and therefore condition- and timing-pathways to clutch initiation are not

separate. Depleted 13C and enriched 15N tissue signatures are expected to be associated with high quality winter habitat use.

91

Figure 4.2 The relationship between first-clutch initiation date and annual productivity (total number of young fledged) among Yellow Warbler females within our high-latitude breeding population (Inuvik, top) and our mid-latitude breeding population (Revelstoke, bottom). See also Table 4.2.

92

Chapter 5. Wind speed during migration influences the survival, timing of breeding, and productivity of a Neotropical migrant

Anna Drake, Christine Rock, Sam P. Quinlan, Michaela Martin, David J. Green1

Abstract

Migratory bird populations can be limited by climatic conditions on breeding

grounds, wintering grounds or during migration. Identifying the timing and location of

events that alter demographic rates would improve the accuracy of population models

and aid conservation efforts. We examine how climate variables operating at discrete

periods of the annual cycle influence the demography of Yellow Warblers (Setophaga

petechia) that breed in western Canada and overwinter in Mexico. We demonstrate that

wind conditions during spring migration are the best predictor of adult survival, male

arrival date, female clutch initiation date and, via these timing effects, reproductive

success. Our study emphasizes that events that occur during the migratory period are

important for the population dynamics of Nearctic-Neotropical migrants.

Introduction

Climate events have been related to variation in the phenology, survival, and

reproductive success of migratory birds (Sillett et al. 2000; Nott et al. 2002; Dugger et al.

1 This chapter is modified from a manuscript formatted for PNAS.

93

2004; Studds and Marra 2011) and have consequently been incorporated into

conceptual models examining the population dynamics of Nearctic-Neotropical migrants

(Sherry and Holmes 1995). However, our ability to predict the relative importance of

climate events operating at different stages of the annual cycle has been limited by

several factors: 1) climate variables are often correlated between periods and across

regions (Nott et al. 2002; Forchhammer et al. 2002); 2) events in one period may impact

processes occurring in subsequent ones (Norris et al. 2004; Saino et al. 2004); and 3)

most studies have focused on species that winter in the Caribbean and breed in eastern

North America. In this study, we evaluate the relative importance of conditions during

the winter, migratory, and breeding season on the demographic rates of Yellow

Warblers breeding in western Canada.

Yellow Warblers have a broad breeding distribution in North America. Genetic-

isotopic work suggests eastern and western lineages have parallel migration systems

and differing wintering distributions (Boulet et al. 2006). Northwestern populations, such

as those at our study site near Revelstoke, British Columbia (50.97°N, -118.20°W;

Figure 1.1), winter primarily in lowland areas in the occidental and isthmus region of

Mexico. In spring, they migrate along the western and central flyway (>97°W) (Boulet et

al. 2006) arriving at northern breeding grounds by mid-May. Males establish territories in

riparian and wet deciduous habitat and the majority of pairs rear a single clutch of four

nestlings before migrating back to Mexico in late August.

Climatic conditions operating during the wintering period, spring migration and

breeding period could all explain annual variation in the survival, breeding phenology

and subsequent reproductive success of Yellow Warblers in Revelstoke. We therefore

developed seven regionally-specific climate models that could potentially describe this

variation. We assessed the relative support for these models using an Information

Theoretic approach (Burnham and Anderson 2004). Wintering period models. Primary

productivity and insect abundance in overwintering areas, which is linked to rainfall

(Brown and Sherry 2006; Studds and Marra 2007), has been shown to increase survival

and induce earlier spring migration in some species (Dugger et al. 2004; Studds and

Marra 2007). The El Niño Southern Oscillation (ENSO) impacts rainfall patterns in

Mexico and we used the Southern Oscillation Index (SOI) to infer wintering conditions

for our population. The majority of rainfall (>60%) in the occidental and isthmus regions

94

of Mexico falls during the summer monsoon (May-August) (Caso et al. 2007) and,

consequently, this period impacts the conditions encountered by birds on the wintering

grounds. Weak monsoon years are associated with El Niño (negative SOI) phases of

ENSO (Caso et al. 2007) and likely result in reduced food availability for wintering birds.

We therefore predicted that negative mean SOI values in the May-August period would

be associated with low survivorship and delayed phenology in our population the

following spring (Model 1). It is possible that late winter rainfall is disproportionately

important to Neotropical migrants. Such rainfall contributes significantly less moisture to

over-winter habitat than the monsoon, but occurs at a critical time, when regions are

experiencing drought conditions. In winter El Niño conditions (negative SOI) promote

greater precipitation (Caso et al. 2007). We therefore predicted that negative SOI values

in the December-February period would be associated with favourable conditions that

would improve survivorship and advance breeding phenology (Model 2). Spring

migration models. Strong winds and poor weather can impact energetic costs and speed

of movement during migration (Liechti 2006). Birds facing hostile conditions should have

lower survival or later arrival dates due to slowed movement and increases in stop-over

number or duration (Liechti 2006). We used mean nighttime wind vectors (westerly (U-)

and southerly (V-) components) and precipitation over the western flyway region (Figure

5.1) between March and May as measures of migration costs. High wind speeds

(westerly (Model 3), southerly (Model 4), and combined (U+V; Model 5) and greater

precipitation (Model 6) were expected to be associated with reduced survivorship and

delayed breeding phenology. Breeding period models. Local temperatures at the

beginning of the breeding season impact the timing of insect development and therefore

food availability (Visser et al. 2006). Migratory birds have been shown to adjust arrival

and/or clutch initiation partly in response to local/regional conditions (Mazerolle et al.

2011). Increased food availability and earlier breeding phenology may also enhance bird

condition at the end of the breeding season, reducing overwinter mortality. Higher mean

May temperatures at our study site were therefore expected to be associated with

earlier breeding phenology in the same year, and improved adult survivorship in the

following year (Model 7).

95

Results and Discussion

Variation in survival in Revelstoke was best described by the three migration

wind speed models (3, 4, 5), which received 78% of the total model support (Table 5.1).

Model 5 best described survival estimates, which declined in association with increasing

westerly and southerly wind speeds. Adult apparent survival was predicted to decline

from 75 to 37% across the range of westerly (Figure 5.2) and 68 to 46% across the

range of southerly (Figure 5.3) wind speeds in our dataset.

Wind speed also best described male arrival and female clutch initiation dates.

Wind speed models describing male arrival date received 68% of total model support

(Table 5.2). The westerly wind vector during spring migration was the best predictor of

male arrival (Figure 5.4, Table 5.2). Wind speed models describing female clutch

initiation date received 80% of total model support with the westerly wind vector also

being the best predictor (Figure 5.3, Table 5.3). Individuals that returned in more than

one year showed similar delays in years where westerly winds were stronger (male

arrival: P=0.001, t test, n=49; female clutch initiation: P=0.002, t-test, n=21; Figure 5.5).

Wind speed effects on breeding phenology and annual productivity varied with

age. Older males anticipated young male arrival by 12 days under the most favourable

wind conditions and arrived at the same time as young males in the year with the

strongest wind (Figure 5.4). Westerly wind vectors on the flyway are weaker in May,

suggesting that delaying migration could reduce costs for individuals. It is possible that

older male Yellow Warblers trade-off migration costs against the productivity

advantages associated with early arrival (Reudink et al. 2009) while young males, who

arrive late regardless of wind conditions, may simply minimize migration costs. Over the

range of westerly wind speeds, clutch initiation varied by 8 days in young females and 3

days in older females (Figure 5.4). This difference may arise because older females are

better able to compensate for delayed arrival by initiating reproduction more rapidly. We

calculate that, across the range of observed wind speeds, delays in clutch initiation

would result in declines in productivity of 0.5 fledglings per year for young females and

0.2 fledglings per year for older females.

Mean SOI values between March and May are correlated with westerly wind

vectors in the same time period (r2=0.36, P<0.0001, n=50). Our data therefore

96

corroborates and offers an additional explanation for delayed movement of migrants

through California in La Niña years (Macmynowski et al. 2007) as well as increased

migrant productivity in the Pacific Northwest during El Niño years (Nott et al. 2002).

Research conducted on Neotropical migrants that over-winter in the Caribbean

and breed in the eastern US and Canada indicate that populations are limited by events

during the wintering and/or breeding period (e.g. Wilson et al. 2011; Sillett et. al. 2000).

In contrast, we find only weak evidence that climate factors acting during the wintering

and breeding periods influence Yellow Warblers using western migratory routes and far

more compelling support for migration effects (see also LaManna et al. 2012). We argue

that this difference is due to 1) wind patterns that oppose the north/north-westerly

course of western migrants while facilitating spring movement in the east (Boulet et al.

2006; Gauthreaux 1980) and 2) a weaker dependency on wintering habitat quality for

birds using over-land (western) versus over-water (eastern) migration routes. Birds

moving through western North America can compensate for poor departure conditions

through shorter flights and increased stopovers: a behavioural option not available to

eastern birds who must cross large ecological barriers (Alerstam 1979; Schaub et al.

2008). We suggest that whereas pre-migration fattening in wintering habitat may be

essential for the movement and survival of eastern populations, en route conditions and

habitats are more important for western birds.

Materials and Methods

Population Monitoring

Survival

Breeding birds were captured and fitted with a Canadian Wildlife Service-issued

aluminum band and a unique combination of three color-bands (AC Hughes, UK). Birds

were aged as either SY (first breeding season) or ASY (at least 2 years old) individuals

based on plumage characteristics (Pyle 1997). Apparent annual adult survival for the

period 2004-2011 was calculated as the probability of an adult returning to the study site

(φ) after adjusting for the re-sighting probability of banded individuals (p) using methods

described by Lebreton et al. (1992) implemented using program MARK version 5.1

97

(White and Burnham 1999). The dataset contained 279 individuals (141 females, 138

males) and 437 encounter histories. The global model that allowed adult survival to vary

as a function of gender, age (SY vs. ASY), and year and re-sighting probability to vary

as a function of gender and between years with extensive monitoring (2005-2006 and

2008-2011) and one year with less extensive monitoring (2007) fit the data well (median

procedure, ĉ=1.11). We first determined the best model structure for the resighting rate,

then modeled survival rates with candidate models containing gender, age, year, and all

possible interactions, and finally investigated the influence of climate variables. We used

Akaike’s Information Criterion corrected for small sample sizes and over-dispersion

(QAICc), to rank competing models (Table 5.1). Recapture rates were best modeled

with only a gender term; males were estimated to have higher resighting probabilities

than females (male=0.89 ± 0.04, female=0.63 ± 0.09). Apparent annual survival rates

were best described by a model that included year and age (AIC weight=0.53) and a

model that included year, age and gender (AIC weight=0.41). The final candidate model

set therefore included both these models, models that included age, gender and climate

variables as main effects, only age and climate variables as main effects, and models

that included main effects, age×climate interactions and/or gender×climate interactions

(Table 5.1 lists the 47 models in the final candidate set).

Breeding Phenology

Breeding data was collected between 2005-2006, and 2008-2011. Male arrival dates

(n=183) were determined during surveys conducted every 1-2 days from early-May to

late-June. Pairs were monitored every 3 days from early-May to late-July in order to

determine the clutch initiation date (n=155), clutch size, fate and fledging success of all

nesting attempts. Clutch initiation date was defined as the laying date of the first egg in

a female’s first nesting attempt of the season. Nestlings were banded seven days post-

hatch and the number of nestlings present at this date was assumed (in the absence of

evidence of predation) to be the number of young fledged from successful nests.

Successful fledging was usually confirmed by observing parents feeding or defending

young out of the nest.

Age has a strong influence on individual behaviour, timing and breeding success

(e.g. 28, 29) (Fowler 1995; Stewart et al. 2002). Models examining climate effects on

98

male arrival and female clutch initiation dates included age as a main effect and as an

interaction term. The candidate model set therefore included 15 models (Table 5.2 and

5.3).

Carry-over Effect on Female Productivity

Productivity was defined as the total number of young fledged across all nesting

attempts made by a given individual. Later initiation dates in our population resulted in

lower female productivity (SY females: r2=0.18, n=61, P<0.001; ASY female: r2=0.08,

n=92, P=0.004). We multiplied the standardized partial regression coefficients (βs) for

the two steps in the pathway between wind speed and fledgling number to calculate the

total effect (TE) of wind speed on productivity for each age class. The TE score for ASY

females was less than for SY birds (-0.03 vs. -0.11). TE and the standard deviation of

both wind speed and age-specific fledge numbers in our data set were then used to

calculate predicted change in fledge number across the range of wind speeds observed

(Norris et al. 2004; Bart and Earnst 1999).

Climate variable sources and derivation.

All climate variables were derived from publically available datasets. Wintering

period. SOI values were averaged standardized monthly SOI data for May to August

and December to February (NOAA SOI 2012). Spring migration. Wind and rainfall on

migration was derived from modeled climate data extracted from the National Center of

Environmental Prediction (NCEP) Reanalysis 1 data archives at the NOAA-CIRES

Climate Diagnostics Center at Boulder, Colorado, USA (NOAA NCEP 2012) using the

RNCEP program (Kemp et al. 2012). This data has a spatial resolution of 2.5° latitude

and longitude and temporal resolution of six hours. We defined the western flyway for

our population as the overland region west of the easternmost portion of the continental

divide (107°W), beginning at the northern extent of the Yellow Warbler wintering range

(25°N) and ending at the latitude of our study site (50°N). We used surface precipitation

data (Kg/m2/s) and wind vector data (U- and V-wind speed components (m/s)) averaged

from the 850mb (1500m) and 925mb (700m) level (which equates with the mean altitude

of passerine migration (1094±319 m (Alerstam et al. 2011) for the March-May period,

dropping noon values so that the series represented conditions encountered during

nighttime migration (between 18:00h and 6:00h). Points that corresponded to locations

99

over the Pacific were dropped while those over the Baja peninsula and Gulf of California

were retained. Breeding period. Averaged daily mean temperatures (°C) on the

breeding grounds during the month of May (Mazerolle et al. 2011) were drawn from

“Revelstoke A” weather station (ID: 1176749; 50°57’40N, -118°11’00W, elevation

444.7m (Environment Canada 2012).

There was some collinearity between the five climate variables (Table 5.4), but in

all but one case the predicted effects were in opposite directions allowing our

predictions to be evaluated. SOIDEC-FEB and U-wind were positively correlated (r(6)=0.74,

P=0.05) and both were predicted to be negatively correlated with breeding phenology

and survival. However, model support for SOIDEC-FEB was consistently low (Table 5.1-5.3)

and we did not consider it to be competitive explanation for survival or breeding

phenology. SOIMAY-AUG and U-wind were also positively correlated with sampling year in

our data set (Table 5.4).

We evaluated whether temporal trends in breeding phenology and survival could

have influenced model selection by creating an alternative candidate model set that

replaced climate variables with the residuals of a climate metric-year regression and a

continuous time variable.

Conclusions regarding climate effects on breeding phenology did not change

when de-trended climate metrics were used. De-trended wind models received more

support (48%) than the next best models (SOIMAY-AUG; 30%) in candidate sets examining

variation in male arrival. De-trended wind models also received more support (58%)

than the next best model (the null model, “continuous time + YEAR (nominal) + AGE”;

19%) in candidate sets examining variation in female clutch initiation date. Annual

variation in apparent survival could be explained by linear declines in survival over the 7

years of the study since the null model (the continuous time variable alone) received

more support than any model that also included de-trended climate metrics. However,

the magnitude of the annual variation in apparent survival observed is unlikely to reflect

population trends in survival and we interpret support for the null model as reflecting the

close correlation between westerly wind vectors and year over our sampling period.

We assessed the relationship between mean SOI values between March and

May and westerly wind speed in the same time period (r2=0.36, n=50, P<0.0001) by

100

deriving these values, as above, for the years 1962-2011. Westerly wind speed for the

850mb (1500m) and 925mb (700m) levels alone did not differ from each other in their

relationship with SOI (850mb: r2=0.37, n=50, P<0.0001; 925mb: r2=0.33, n=50,

P<0.0001).

Wind speed declined from March to May (U-wind: F2,147=3.13, P=0.05; V-wind:

F2,147=13.22, P<0.0001). Post-hoc comparisons showed both vectors were significantly

weaker in May relative to earlier in the migratory period. U-wind speed dropped only in

May (students-t: March vs. April, P=0.97; April vs. May, P=0.03; March vs. May, P=0.03)

while V-wind showed a steady decline over the three-month period (students-t: March

vs. April, P=0.001; April vs. May, P=0.09; March vs. May, P<0.0001).

Statistical programs

Survival QAICc were run in MARK version 5.1 (White and Burnham 1999).

NCEP data extraction and breeding phenology cAIC analyses were run in R version

2.14.1 (2011, The R Foundation for Statistical Computing). All other statistical analyses

were run in JMP (8.0, SAS Institute Inc.).

References

Alerstam, Thomas. 1979. Wind as Selective Agent in Bird Migration. Ornis Scandinavica 10 (1): 76–93.

Alerstam, Thomas, Jason W. Chapman, Johan Bäckman, Alan D. Smith, Håkan Karlsson, Cecilia Nilsson, Don R. Reynolds, Raymond H. G. Klaassen, and Jane K. Hill. 2011. Convergent Patterns of Long-distance Nocturnal Migration in Noctuid Moths and Passerine Birds. Proceedings of the Royal Society B: Biological Sciences 278 (1721): 3074–3080. doi:10.1098/rspb.2011.0058.

Bart, Jonathan, and Susan L. Earnst. 1999. Relative Importance of Male and Territory Quality in Pairing Success of Male Rock Ptarmigan (Lagopus mutus). Behavioral

Ecology and Sociobiology 45: 355–359. doi:10.1007/s002650050571.

Boulet, Marylène, H. Lisle Gibbs, and Keith A. Hobson. 2006. Integrated Analysis of Genetic, Stable Isotope, and Banding Data Reveal Migratory Connectivity and Flyways in the Northern Yellow Warbler (Dendroica petechia; Aestiva Group).

Ornithological Monographs (61): 29–78.

101

Brown, David, and Thomas Sherry. 2006. Food Supply Controls the Body Condition of a Migrant Bird Wintering in the Tropics. Oecologia 149 (1): 22–32. doi:10.1007/s00442-006-0418-z.

Burnham, Kenneth P., and David R. Anderson. 2004. Multimodel Inference Understanding AIC and BIC in Model Selection. Sociological Methods & Research 33 (2): 261–304. doi:10.1177/0049124104268644.

Caso, Margarita, Charlotte González-Abraham, and Exequiel Ezcurra. 2007. Divergent Ecological Effects of Oceanographic Anomalies on Terrestrial Ecosystems of the Mexican Pacific Coast. Proceedings of the National Academy of Sciences 104 (25): 10530–10535. doi:10.1073/pnas.0701862104.

Dugger, Katie M., John Faaborg, Wayne J. Arendt, and Keith A. Hobson. 2004. Understanding Survival and Abundance of Overwintering Warblers: Does Rainfall Matter? The Condor 106 (4): 744–760. doi:10.1650/7632.

Environment Canada. 2012. National Climate Data and Information Archive, Environment Canada. www.climate.weatheroffice.gc.ca/climateData.

Forchhammer, M. C., E. Post, and N. C. Stenseth. 2002. North Atlantic Oscillation Timing of Long- and Short-distance Migration. Journal of Animal Ecology 71: 1002–1014.

Fowler, Gene S. 1995. Stages of Age-Related Reproductive Success in Birds: Simultaneous Effects of Age, Pair-Bond Duration and Reproductive Experience. American Zoologist 35 (4): 318–328. doi:10.1093/icb/35.4.318.

Gauthreaux, S.A. 1980. The Influence of Global Climatological Factors on the Evolution of Bird Migratory Pathways. Proceedings XVII International Ornithological Congress 17: 517–525.

Kemp, Michael U., E. Emiel van Loon, Judy Shamoun-Baranes, and Willem Bouten. 2012. RNCEP: Global Weather and Climate Data at Your Fingertips. Methods in Ecology and Evolution 3 (1): 65–70. doi:10.1111/j.2041-210X.2011.00138.x.

LaManna, J. A., T. L. George, J. F. Saracco, M. P. Nott, and D. F Desante. 2012. El Nino-Southern Oscillation Influences Annual Survival of a Migratory Songbird at a Regional Scale. The Auk 129 (4): 1–10.

Lebreton, Jean-Dominique, Kenneth P. Burnham, Jean Clobert, and David R. Anderson. 1992. Modeling Survival and Testing Biological Hypotheses Using Marked Animals: A Unified Approach with Case Studies. Ecological Monographs 62 (1): 67–118. doi:10.2307/2937171.

Liechti, Felix. 2006. Birds: Blowin’ by the Wind? Journal of Ornithology 147 (2): 202–211. doi:10.1007/s10336-006-0061-9.

102

Macmynowski, Dena P., Terry L. Root, Grant Ballard, and Geoffrey R. Geupel. 2007. Changes in Spring Arrival of Nearctic-Neotropical Migrants Attributed to Multiscalar Climate. Global Change Biology 13 (11): 2239–2251. doi:10.1111/j.1365-2486.2007.01448.x.

Mazerolle, Daniel F., Spencer G. Sealy, and Keith A. Hobson. 2011. Interannual Flexibility in Breeding Phenology of a Neotropical Migrant Songbird in Response to Weather Conditions at Breeding and Wintering Areas. Ecoscience 18 (1): 18–25. doi:10.2980/18-1-3345.

NOAA NCEP 2012. NCEP/NCAR Reanalysis 1. Earth System Research Laboratory, National Oceanic and Atmospheric Administration. www.esrl.noaa.gov/psd/data/gridded/data.ncep.reanalysis.html.

NOAA SOI. 2012. Southern Oscillation Index (SOI) Sea Level Pressure Standardized Data. Climate Prediction Center, National Oceanic and Atmospheric Administration. www.cpc.ncep.noaa.gov/data/indices/soi.

Norris, D. Ryan, Peter P. Marra, T. Kurt Kyser, Thomas W. Sherry, and Laurene M. Ratcliffe. 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:10.1098/rspb.2003.2569.

Nott, M. Philip, David F Desante, Rodney B Siegel, and Peter Pyle. 2002. Influences of the El Niño/Southern Oscillation and the North Atlantic Oscillation on Avian Productivity in Forests of the Pacific Northwest of North America. Global Ecology and Biogeography 11 (4): 333–342. doi:10.1046/j.1466-822X.2002.00296.x.

Pyle, Peter. 1997. Identification Guide to North American Birds, Part I: Columbidae to Ploceidae. Slate Creek Press.

Reudink, Matthew W, Peter P Marra, T. Kurt Kyser, Peter T Boag, Kathryn M Langin, and Laurene M Ratcliffe. 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:10.1098/rspb.2008.1452.

Saino, Nicola, Tibor Szép, Roberto Ambrosini, Maria Romano, and Anders Pape Møller. 2004. Ecological Conditions During Winter Affect Sexual Selection and Breeding in a Migratory Bird. Proceedings of the Royal Society B: Biological Sciences 271 (1540): 681–686.

Schaub, Michael, Lukas Jenni, and yranz Bairlein. 2008. Fuel Stores, Fuel Accumulation, and the Decision to Depart from a Migration Stopover Site. Behavioral Ecology 19 (3): 657–666. doi:10.1093/beheco/arn023.

Sherry, T. W., and R. T. Holmes. 1995. Summer Versus Winter Limitation of Populations: Conceptual Issues and Evidence. In Ecology and Management of Neotropical Migratory Birds: A Synthesis and Review of Critical Issues, edited by T. E. Martin and D. M. Finch, pp. 85–120. Oxford University Press, New York.

103

Sillett, T. Scott, Richard T. Holmes, and Thomas W. Sherry. 2000. Impacts of a Global Climate Cycle on Population Dynamics of a Migratory Songbird. Science 288 (5473): 2040–2042. doi:10.1126/science.288.5473.2040.

Stewart, Rebecca L. M., Charles M. Francis, And Cheryl Massey. 2002. Age-Related Differential Timing of Spring Migration Within Sexes in Passerines. The Wilson Bulletin 114 (2): 264–271. doi:10.1676/0043-5643(2002)114[0264:Ardtos]2.0.Co;2.

Studds, Colin E, and Peter P Marra. 2007. Linking Fluctuations in Rainfall to Nonbreeding Season Performance in a Long-distance Migratory Bird, Setophaga ruticilla. Climate Research 35: 115–122. doi:10.3354/cr00718.

Studds, Colin E., and Peter P. Marra. 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:10.1098/rspb.2011.0332.

Visser, Marcel, Leonard Holleman, and Phillip Gienapp. 2006. Shifts in Caterpillar Biomass Phenology Due to Climate Change and Its Impact on the Breeding Biology of an Insectivorous Bird. Oecologia 147 (1): 164–172. doi:10.1007/s00442-005-0299-6.

White, Gary C., and Kenneth P. Burnham. 1999. Program MARK: Survival Estimation from Populations of Marked Animals. Bird Study 46 (sup001): S120–S139. doi:10.1080/00063659909477239.

Wilson, Scott, Shannon L. Ladeau, Anders P. Tøttrup, and Peter P. Marra. 2011. Range-wide effects of breeding- and nonbreeding-season climate on the abundance of a Neotropical migrant songbird. Ecology 92 (9): 1789–1798.

104

Tables and Figures

Table 5.1 Models describing apparent annual survival of Yellow Warblers breeding in Revelstoke, British Columbia (n=279 individuals, 437 encounters). Models are ranked using QAICc. Model number refers to regionally-specific climate variables described in the text. Age was included in all models as a covariate (see methods). The number of parameters in the model (K), Akaike’s information criterion (QAICc), QAICc difference from the top model (ΔQAICc), and Akaike weight (ωi) are reported.

Period Model Variables K QAICc ∆QAICc ωi

Migration 5 U-WIND + V-WIND + AGE 6 526.46 0.00 0.168

Migration 5 U-WIND + V-WIND + AGE + SEX 7 526.72 0.26 0.148

Migration 3 U-WIND + AGE 5 527.41 0.96 0.104

Migration 3 U-WIND + AGE + SEX 6 527.65 1.20 0.092

Migration 3 U-WIND + AGE + SEX + U-WIND*SEX 7 528.43 1.97 0.063

Migration 5 U-WIND + V-WIND + AGE + SEX + U-WIND*SEX + V-WIND*SEX

9 528.55 2.09 0.059

- - YEAR(nominal) + AGE 10 528.73 2.27 0.054

- - YEAR(nominal) + AGE + SEX 11 529.07 2.61 0.046

Migration 3 U-WIND + AGE + U-WIND*AGE 6 529.32 2.86 0.040

Migration 3 U-WIND + AGE + SEX + U-WIND*AGE 7 529.66 3.2 0.034

Migration 3 U-WIND + AGE + SEX + U-WIND*AGE + U-WIND*SEX 8 530.39 3.93 0.024

Migration 5 U-WIND + V-WIND + AGE + U-WIND*AGE + V-WIND*AGE 8 530.46 4.00 0.023

Winter 1 SOIMAY-AUG + AGE 5 530.46 4.01 0.023

Migration 5 U-WIND + V-WIND + AGE + SEX + U-WIND*AGE + V-WIND*AGE

9 530.83 4.37 0.019

Winter 1 SOIMAY-AUG + AGE + SEX 6 530.87 4.41 0.019

Winter 1 SOIMAY-AUG + AGE + SEX + SOIMAY-AUG *SEX 7 532.22 5.76 0.009

Winter 1 SOIMAY-AUG + AGE + SOIMAY-AUG *AGE 6 532.36 5.90 0.009

Migration 5 U-WIND + V-WIND + AGE + SEX + U*AGE + V*AGE + U*SEX + V*SEX

11 532.71 6.25 0.007

Winter 1 SOIMAY-AUG + AGE + SEX + SOIMAY-AUG *AGE 7 532.77 6.31 0.007

Breeding 7 MAY MEAN°C + AGE 5 532.81 6.35 0.007

Breeding 7 MAY MEAN°C + AGE + SEX 6 533.30 6.84 0.005

105

Table 5.1 continued

Period Model Variables K QAICc ∆QAICc ωi

Winter 2 SOIDEC-FEB + AGE 5 533.61 7.15 0.005

Winter 2 SOIDEC-FEB + AGE + SEX 6 534.06 7.60 0.004

Breeding 7 MAY MEAN°C + AGE + MAY MEAN°C *AGE 6 534.07 7.61 0.004

Winter 1 SOIMAY-AUG + AGE + SEX + SOIMAY-AUG *AGE + SOIMAY-AUG *SEX

8 534.11 7.65 0.004

Breeding 7 MAY MEAN°C + AGE + SEX + MAY MEAN°C*SEX 7 534.30 7.84 0.003

Breeding 7 MAY MEAN°C + AGE + SEX + MAY MEAN°C*AGE 7 534.46 8.00 0.003

Winter 2 SOIDEC-FEB + AGE + SEX + SOIDEC-FEB*SEX 7 535.08 8.62 0.002

Breeding 7 MAY MEAN°C + AGE + SEX + MEAN°C*AGE + MEAN°C*SEX

8 535.22 8.76 0.002

Winter 2 SOIDEC-FEB + AGE + SOIDEC-FEB*AGE 6 535.57 9.11 0.002

Winter 2 SOIDEC-FEB + AGE + SEX + SOIDEC-FEB*AGE 7 535.927 9.47 0.001

Migration 6 MIGRATION RAIN + AGE + MIG RAIN*AGE 6 535.99 9.53 0.001

Migration 6 MIGRATION RAIN + AGE + SEX + MIG RAIN*AGE 7 536.03 9.57 0.001

Migration 6 MIGRATION RAIN + AGE 5 536.38 9.92 0.001

Migration 4 V-WIND + AGE 5 536.93 10.47 0.001

Migration 6 MIG RAIN + AGE + SEX 6 537.03 10.57 0.001

Winter 2 SOIDEC-FEB + AGE + SEX + SOIDEC-FEB*AGE + SOIDEC-FEB*SEX 8 537.10 10.64 0.001

Migration 6 MIGRATION RAIN + AGE + SEX + MIG RAIN*AGE + MIG RAIN*SEX

8 537.50 11.04 0.001

Migration 4 V-WIND + AGE + SEX 6 537.59 11.13 0.001

Migration 4 V-WIND + AGE + SEX + V-WIND*SEX 7 537.87 11.41 0.001

Migration 6 MIGRATION RAIN + AGE + SEX + MIG RAIN*SEX 7 538.20 11.75 0

Migration 4 V-WIND + AGE + V-WIND*AGE 6 538.88 12.42 0

Migration 4 V-WIND + AGE + SEX + V-WIND*AGE 7 539.61 13.15 0

Migration 4 V-WIND + AGE + SEX + V-WIND*AGE + V-WIND*SEX 8 539.92 13.46 0

- - YEAR(nominal)*AGE*SEX 30 551.93 25.47 0

106

Table 5.2. Ranked summary of AICc support for the candidate climate models and the null model (YEAR + AGE) describing male Yellow Warbler arrival date (n=183) in Revelstoke, British Columbia. Age (SY=1 yr; ASY>2 yrs) was included in all models as a covariate (see methods). Model adjusted r2, the number of parameters in the model (K), Akaike’s information criterion adjusted for small sample size (AICc), AICc difference from the top model (ΔAICc), and Akaike weight (ωi) are reported.

Period Model Variables r2 K AICc ∆AICc ωi

Migration 3 U-WIND + AGE + U*AGE 0.23 5 1273.80 0.00 0.546

Winter 1 SOIMAY-AUG + AGE + SOIMAY-AUG*AGE 0.22 5 1275.15 1.35 0.278

Migration 5 V-WIND + U-WIND + AGE + V-WIND*AGE + U-WIND*AGE

0.22 7 1277.35 3.55 0.092

Migration 3 U-WIND + AGE 0.20 4 1279.73 5.93 0.028

Winter 1 SOIMAY-AUG + AGE 0.20 4 1280.35 6.55 0.021

Migration 5 V-WIND + U-WIND + AGE 0.20 5 1280.81 7.01 0.016

Winter 2 SOIDEC-FEB + AGE + SOIDEC-FEB*AGE 0.20 5 1281.56 7.77 0.011

- - YEAR + AGE 0.20 8 1283.82 10.03 0.004

Winter 2 SOIDEC-FEB + AGE 0.18 4 1283.98 10.18 0.003

Breeding 7 MAY MEAN°C + AGE 0.12 4 1296.18 22.38 0

Breeding 7 MAY MEAN°C + AGE + MAY MEAN°C *AGE 0.12 5 1297.19 23.39 0

Migration 4 V-WIND + AGE 0.11 4 1299.00 25.19 0

Migration 4 V-WIND + AGE + V-WIND *AGE 0.11 5 1300.32 26.52 0

Migration 6 MIGRATION RAIN + AGE 0.09 4 1303.22 29.42 0

Migration 6 MIGRATION RAIN + AGE + MIGRATION RAIN*AGE

0.08 5 1305.33 31.53 0

107

Table 5.3 Ranked summary of AICc support for the candidate climate models and the null model (YEAR + AGE) describing female Yellow Warbler clutch initiation date (n=155) in Revelstoke, British Columbia. Age (SY=1 yr; ASY>2 yrs) was included in all models as a covariate (see methods). Model adjusted r2, the number of parameters in the model (K), Akaike’s information criterion adjusted for small sample size (AICc), AICc difference from the top model (ΔAICc), and Akaike

weight (ωi) are reported.

Period Model Candidate Model r2 K AICc ∆AICc ωi

Migration 3 U-WIND + AGE + U*AGE 0.16 5 1045.30 0.00 0.317

Migration 5 V-WIND + U-WIND + AGE 0.15 5 1046.30 1.00 0.192

Migration 3 U-WIND + AGE 0.14 4 1046.34 1.04 0.188

- - YEAR + AGE 0.16 8 1047.01 1.70 0.135

Migration 5 V-WIND + U-WIND + AGE + V-WIND*AGE + U-WIND*AGE

0.15 7 1047.79 2.48 0.091

Winter 1 SOIMAY-AUG + AGE 0.12 4 1050.32 5.02 0.026

Winter 1 SOIMAY-AUG + AGE + SOIMAY-AUG*AGE 0.12 5 1051.81 6.51 0.012

Migration 6 MIGRATION RAIN + AGE 0.11 4 1052.82 7.52 0.007

Winter 2 SOIDEC-FEB + AGE + SOIDEC-FEB*AGE 0.11 5 1052.93 7.62 0.007

Winter 2 SOIDEC-FEB + AGE 0.11 4 1053.02 7.72 0.007

Breeding 7 MAY MEAN°C + AGE 0.10 4 1053.44 8.14 0.005

Migration 4 V-WIND + AGE 0.10 4 1053.81 8.51 0.005

Migration 6 MIGRATION RAIN + AGE + MIGRATION RAIN*AGE

0.10 5 1054.75 9.45 0.003

Migration 4 V-WIND + AGE + V-WIND *AGE 0.10 5 1055.14 9.84 0.002

Breeding 7 MAY MEAN°C + AGE + MAY MEAN°C *AGE 0.10 5 1055.29 9.99 0.002

108

Table 5.4. Correlation (r) matrix of explanatory climate variables and time (year) (n=7). Significant relationships (Spearman’s ρ) are starred (*=0.01 ≤ P ≤ 0.05). Variables where the predicted relationship with survival and breeding phenology would suggest a possible confound between models are marked with (‡)

SOIMAY-AUG SOIDEC-FEB Mig. Rain U-Wind V-Wind Year

Mean May °C -0.50 -0.05 -0.52 -0.63 0.34 -0.64

SOIMAY-AUG 0.75* 0.28 0.74* 0.11 0.73*

SOIDEC-FEB -0.38 0.74*‡ -0.16 0.52

Mig. Rain -0.07 0.50 0.14

U-Wind -0.48 0.77*

V-Wind -0.07

109

Figure 5.1 The location of our study (A), the Yellow Warbler wintering range (blue), and the area used to calculate wind speed values during migration (yellow). Image courtesy of NASA.

110

Figure 5.2 Relationship between westerly wind speed during migration and annual survival of Yellow Warblers. Points are apparent survival (± SE) of young (1 yr, open points) and older (≥2 yrs, filled points) birds (7 years, n=279 individuals, 437 encounters). Solid lines and shading represent predicted survival (ϕ) ± 95% CI from the top model assuming an average southerly wind vector.

111

Figure 5.3 Predicted relationship between southerly wind speed during migration and annual survival of Yellow Warblers in Revelstoke, British Columbia (see also Fig. 1). Lines and shading represent predicted survival (ϕ) ± 95% CI for young (1 yr, dashed) and older (≥2 yrs, solid) birds from the top model assuming an average westerly wind vector (7 years, n=279 individuals, 437 encounters).

Fig. S1. Predicted relationship between southerly wind speed during migration and annual

survival of Yellow Warblers in Revelstoke, British Columbia (see also Fig. 1). Lines and

shading represent predicted survival (ϕ) ± 95% CI for young (1 yr, dashed) and older (≥2 yrs,

solid) birds from the top model assuming an average westerly wind vector (7 years, n=279

individuals, 437 encounters).

112

Figure 5.4 Male arrival date (circles) and female clutch initiation date (diamonds) for Yellow Warblers as a function of westerly wind speed during migration. Points represent mean dates ± SE for young (1yr, open points) and older (≥2yrs, filled points) birds (6 years, n=338).

Fig. 3. Male arrival date (circles) and female clutch initiation date (diamonds) for Yellow

Warblers as a function of westerly wind speed during migration. Points represent mean dates ±

SE for young (1yr, open points) and older (≥2yrs, filled points) birds (6 years, n=338).

113

Figure 5.5 Within-individual changes in arrival date (males; n=49) and clutch initiation date (females; n=21) associated with inter-annual variation in average westerly wind vectors. Box-plots show the median change, 10th and 90th percentile, and outliers. Positive values represent delayed breeding phenology in stronger wind years; negative values represent advanced breeding phenology in stronger wind years. Data was restricted to individuals that were ≥ 2 years of age in both years. Males and females exhibited a median shift of +2 and +4 days respectively.

Fig. S2. Within-individual changes in arrival date (males; n=49) and clutch initiation date

(females; n=21) associated with inter-annual variation in average westerly wind vectors. Box-

plots show the median change, 10th

and 90th

percentile, and outliers. Positive values represent

delayed breeding phenology in stronger wind years, negative values represent advanced breeding

phenology in stronger wind years. Data was restricted to individuals that were ≥ 2 years of age

in both years. Males and females exhibited a median shift of +2 and +4 days respectively.

114

Chapter 6. Significance and Future Directions

General Conclusions

My research shows that Yellow Warblers wintering in the isthmus of Mexico

exhibit sex- but not age-segregation (Chapter 2). As in American Redstarts (Setophaga

ruticilla), male Yellow Warblers in the Pacific slope region predominate in limited, wetter

habitat (riparian corridors) while females represent a larger proportion of individuals

using drier, more abundant, shrub habitat (coastal scrub/seasonal saline marsh).

However, sex-segregation appeared to vary by region and was not evident in habitats

used by Yellow Warblers on the Gulf of Mexico. Regional differences, if real, may arise

because of regional rainfall patterns. Water limitation is a more common within the

tropical deciduous forests of the Pacific slope than in the wetter Gulf region. This may

produce steeper habitat gradients and possibly greater competition in western Mexico.

In both regions wintering Yellow Warblers incorporated local isotopic signatures into

their tissues. This was evidenced by significant differences in 13C and 15N between

blood samples collected within different habitat types and within different regions. As

isotope ratios will be incorporated into feather tissue during pre-alternate moult (which

occurs in Yellow Warblers on the wintering grounds (Pyle 1997, pers. observation),

winter-grown feathers collected on the breeding grounds can be used as a source of

cross-seasonal data.

It is important to note that the inference of winter habitat use from feather

isotopes involves several levels of uncertainty. First, in this thesis I assume that isotopic

differences in Yellow Warbler tissue reflect winter habitat quality differences. This

assumption is based on isotope-habitat relationships reported for other species of

wintering Parulidae, but has not been explicitly demonstrated in the Yellow Warbler.

115

Secondly, “ground-truthed” wintering isotope values in Chapter 3 were obtained from

blood samples as most birds captured in Mexico had not completed pre-alternate moult

at the time of capture. Blood tends to be 13C and 15N depleted relative to feathers due

to tissue-specific differences in isotope discrimination (Caut et al. 2009). For 15N, this

means that only relative values can be compared between tissue types. For 13C, blood

discrimination values also increase slightly as the diet becomes more 13C depleted

(Caut et al. 2009). Feathers do not show this shift in discrimination making comparisons

of relative 13C values between tissue types more difficult. Feather signatures should

more accurately reflect habitat-13C while 13C differences between habitats presented

in Figure 3.1 are likely to be slightly underestimated. Finally, carry-over models assume

a linear relationship between wintering isotope signatures and breeding variables. If

wintering habitat use is, in fact, a discrete variable (e.g. “high” versus “low quality”

habitat) the relationship between habitat, isotope values, and breeding variables may be

non-linear. Further work on the wintering grounds is needed to (a) quantify the

relationship between 13C, 15N and winter habitat quality for Yellow Warblers and (b)

determine species-specific discrimination values for feather and blood samples.

In Chapter 3 I showed evidence for a relationship between winter habitat use

(inferred using 13C and 15N in winter-grown feathers), breeding phenology, and

subsequent productivity among young (first-breeding season) female Yellow Warblers in

Revelstoke, British Columbia. Within this age-sex class, depleted 13C and enriched

15N was associated with earlier clutch initiation dates. This suggested that young

females wintering in wetter, and possibly more northern locations were able to breed

earlier the following spring. As the timing of clutch initiation was associated with higher

productivity, winter habitat use carried-over to impact young female fitness. No similar

effects were seen in other age-sex classes.

An equivalent analysis involving Yellow Warblers breeding in Inuvik, Northwest

Territories in Chapter 4 yielded less support for winter habitat carry-over effects. Here I

found no evidence that 13C or 15N was associated with timing effects in this

population. Support existed for a 15N effect on male condition: earlier clutch initiation

dates occurred among pairs where males had enriched 15N signatures. However, this

effect did not carry-over to influence productivity. Comparisons with the existing

116

American Redstart data (Norris et al. 2004) and with the Revelstoke data presented in

Chapter 3, indicated that Yellow Warblers breeding in Inuvik differ from southern

populations in both how they are influenced by winter habitat use and in how the timing

of breeding influences their productivity. In total, carry-over effects were weaker in this

extreme northern population.

Finally, in Chapter 5 I show strong evidence that wind conditions during the

spring migratory period play an important role in the timing of breeding events in

Revelstoke and in annual differences in apparent survival. This same analysis provided

only weak support for an influence of the wintering and breeding periods on this

population. This chapter, and Chapters 3-4 align with recent work (Wilson et al. 2011;

LaManna et al. 2012) that suggests a longitudinal difference exists in the contribution of

non-breeding conditions to events that occur on the breeding grounds. This is an

important avenue for future study. Below, I review the existing research, discuss

possible reasons why a longitudinal effect might be expected, and make predictions

about how the non-breeding period might impact migrants across North America.

Future directions

Non-breeding effects – are there longitudinal patterns?

The current paradigm for carry-over effects in Neotropical migrants is largely

built upon studies of eastern American Redstarts wintering in the Greater Antilles. Here

birds in higher quality wintering territories maintain their weight over the wintering period

and, in spring, depart earlier than their conspecifics (Marra et al. 1998; Marra and

Holmes 2001; Studds and Marra 2005). Similarly, increased food availability in high

rainfall years is associated with earlier departure schedules (Studds and Marra 2007;

Studds and Marra 2011). Early departing individuals subsequently arrive on the

breeding grounds earlier, where they obtain reproductive advantages relative to later-

arriving individuals (Norris et al. 2004; Reudink et al. 2009; Tonra et al. 2011; McKellar

2013). However, this pattern does not appear to be universal among passerines

breeding in North America (Figure 6.2, Table 6.2). A strong influence of winter

conditions on eastern breeding populations is supported by the majority of studies (8:1).

Results from the central region of the continent are more ambivalent (3:2) and effects

117

are more strongly supported for species that winter predominantly in the Caribbean

(González-Prieto 2012). West of the Rocky Mountains, the minority of studies found

support for wintering effects (1:3): two instead found strong evidence of a migratory

effect. As briefly discussed in Chapter 5, there are two reasons why this may be the

case: (1) eastern migrants likely face different movement costs than western migrants

and (2) migrants wintering on islands may face unique constraints relative to mainland

individuals.

It is likely that the majority of eastern birds cross ocean during migration, either

by island-hopping or by making the uninterrupted 18-40 hr. flight across the Gulf of

Mexico (Figure 1 A; Moore and Kerlinger 1987). Flights over water allow no opportunity

to rest or refuel: to make the crossing, individuals must be in good physical condition

and they must be able to obtain a threshold mass prior to departure (Alerstam 1979;

Schaub et al. 2008). Island birds cross open water immediately after departing their

winter territories. For these birds, on-territory fattening to a set threshold is likely

essential in order to initiate spring movement. Among eastern populations wintering in

Central and South America, the ocean barrier is encountered later in migration. For

these individuals, on-territory fattening may be less crucial for the initiation of migration,

but food availability at pre-barrier stopover sites (such as the Yucatan) may be

particularly important. Mainland individuals may also choose to follow an overland

(circum-Gulf) route to the breeding grounds (e.g. Stutchbury et al. 2009). This

behavioural strategy may reduce dependency on any one fuelling location by allowing

birds to make shorter flights. Evidence of lower fat loads carried by migrants at

southwestern (fall) stopover sites relative to stopover sites along the Gulf Coast of

Mexico would support this (Kelly and Hutto 2005).

Western birds are more likely to use exclusively overland routes (Figure 1A) and,

as for mainland eastern populations, on-territory fattening may not be as essential for

the initiation of spring movement. While geographical barriers exist in the west in the

form of the southwestern deserts, these barriers are not as extreme (birds may rest

even if they cannot refuel), they occur later in migration (northern Mexico), and are they

interspersed with ‘habitat islands’ such as riparian corridors that are heavily used

(Skagen et al. 2005a; Kelly and Hutto 2005). For a subset of western migrants,

however, our data indicates that there is a climatological barrier to spring movement.

118

For individuals moving west of the Sierra Nevada, wind patterns at the altitude of

migrant movement oppose north and westward movement (Figure 1B). This would

indicate that migration becomes more costly for western birds once they enter the

western United States. In contrast, conditions for birds in the central and eastern regions

of the US are less hostile with wind vectors facilitating a north-easterly movement

(Figure 1B; Appendix A; Gauthreaux 1980).

Taking these geographical factors into consideration, I would make several

predictions with respect to longitudinal patterns: (1) for eastern migrant populations

wintering in the Caribbean, winter habitat will play a strong role in the timing of

movement; (2) for eastern populations that winter in Central and South America,

stopover sites before water-crossings will be of greater importance than wintering

conditions for timing; (3) for migrants that breed to the west of the continental divide,

wind conditions in California will have a stronger influence than wintering conditions on

timing. Finally, (4) for both eastern and western populations wintering on the mainland,

survival may also be strongly influenced by conditions at stopover sites, which may

mask evidence of a wintering-ground effect. No such masking will occur in Caribbean

populations because wintering habitat and pre-barrier fuelling sites heavily overlap. By

the same token, individual-level carry-over effects in populations that winter on the

mainland may stem from access to foraging opportunities at stopover sites, rather than

on winter territories.

Further work is needed to test this interpretation. Our data suggests that

conditions on migration are correlated with the El Niño Southern Oscillation (ENSO) and

are more hostile during La Niña years. Two western studies where timing and

productivity shifts could be attributed to wintering conditions might also be interpreted

from a migration perspective, as both were correlated with ENSO in the direction we

would predict (Nott et al. 2002; Macmynowski et al. 2007). Evidence of a positive

relationship between previous-year spring temperatures and American Redstart

abundance in the US Northern Rockies Bird Conservation Region (BCR) (Wilson et al.

2011) may also be related to a migration effect. Higher temperatures in this region occur

during El Niño years. In this case, a connection with migration conditions is tentative

given the time lag in the effect and as similar patterns were not seen further north.

119

LaManna et al.’s (2012) work on Swainson’s Thrush (Catharus ustulatus)

breeding in the northwestern United States indicated that higher rainfall in the

southwestern portion of the spring flyway is linked to higher apparent annual survival.

This suggests that wind-speed may not be the only variable influencing populations

during migration. I found no support for a rainfall model with Yellow Warblers (Chapter

5). A post-hoc analysis using the regions used by LaManna et al. (2012) also was not

competitive with wind models (unpublished data). However, the impact of rain may be

species-specific or additive. Higher rainfall is associated with El Niño and therefore

reduced wind-speed years; this confound may mask rain effects in the Yellow Warbler

system.

An understanding of how non-breeding events alter breeding bird abundance,

survival and productivity is needed in order to make effective conservation decisions.

Without this knowledge the outcome of management practices on the breeding and

wintering grounds will be unpredictable and effort and money may be wasted. If

migration is the most costly period for western birds, stopover habitats in the

southwestern United States may play a significant role in maintaining western

populations. These habitats, limited and threatened by development (Skagen et al. 2005

a,b; van Riper et al. 2008), may need to be prioritized.

References

Alerstam, Thomas. 1979. Wind as Selective Agent in Bird Migration. Ornis Scandinavica 10 (1): 76–93.

Bearhop, Stuart, Geoff M Hilton, Stephen C Votier, and Susan Waldron. 2004. Stable Isotope Ratios Indicate that Body Condition in Migrating Passerines is Influenced by Winter Habitat. Proceedings of the Royal Society B: Biological Sciences 271 (Suppl 4): S215–S218.

Caut, Stéphane, Elena Angulo, and Franck Courchamp. 2009 Variation in discrimination factors (Δ15N and Δ13C): the effect of diet isotopic values and applications for diet reconstruction. Journal of Applied Ecology 46: 443-453.

Gauthreaux, S.A. 1980. The Influence of Global Climatological Factors on the Evolution of Bird Migratory Pathways. Proceedings XVII International Ornithological Congress 17: 517–525.

120

González-Prieto, Ana María. 2012. Factors Influencing Body Condition and Arrival Phenology of Neotropical Migrants at a Northern Spring Stopover Site. M.Sc., University of Saskatchewan.

Kelly, Jeffrey F., and Richard L. Hutto. 2005. An East-West Comparison of Migration in North American Wood Warblers. The Condor 107 (2): 197–211. doi:10.2307/4096504.

LaManna, J. A., T. L. George, J. F. Saracco, M. P. Nott, and D. F Desante. 2012. El Nino-Southern Oscillation Influences Annual Survival of a Migratory Songbird at a Regional Scale. The Auk 129 (4): 1–10.

Lincoln, Frederick C., Steven R. Peterson, and John L. Zimmerman. 1998. Migration of Birds. U.S. Department of the Interior, U.S. Fish and Wildlife Service, Washington, D.C. Circular 16. Jamestown, ND: Northern Prairie Wildlife Research Center Online. http://www.npwrc.usgs.gov/resource/birds/migratio/index.htm.

Lindsay, Andrea M. 2008. Seasonal Events and Associated Carry-over Effects in a Neotropical Migratory Songbird, the Yellow Warbler (Dendroica petechia). M.Sc.,

Ohio State University.

Macmynowski, Dena P., Terry L. Root, Grant Ballard, and Geoffrey R. Geupel. 2007. Changes in Spring Arrival of Nearctic-Neotropical Migrants Attributed to Multiscalar Climate. Global Change Biology 13 (11): 2239–2251. doi:10.1111/j.1365-2486.2007.01448.x.

Marra, Peter P. 2012. Studying birds in the context of the annual cycle: Carry-over effects and seasonal interactions. presented at the North American Ornithological Conference, August, Vancouver, British Columbia.

Marra, Peter P., Keith A. Hobson, and Richard T. Holmes. 1998. Linking Winter and Summer Events in a Migratory Bird by Using Stable-Carbon Isotopes. Science 282 (5395): 1884 –1886. doi:10.1126/science.282.5395.1884.

Marra, Peter P., and Richard T. Holmes. 2001. Consequences of Dominance-Mediated Habitat Segregation in American Redstarts During the Nonbreeding Season. The Auk 118 (1): 92–104.

Mazerolle, Daniel F., Kevin W. Dufour, Keith A. Hobson, and Heidi E. den Haan. 2005. Effects of Large-scale Climatic Fluctuations on Survival and Production of Young in a Neotropical Migrant Songbird, the Yellow Warbler Dendroica petechia.

Journal of Avian Biology 36 (2): 155–163. doi:10.1111/j.0908-8857.2005.03289.x.

Mazerolle, Daniel F., Spencer G. Sealy, and Keith A. Hobson. 2011. Interannual Flexibility in Breeding Phenology of a Neotropical Migrant Songbird in Response to Weather Conditions at Breeding and Wintering Areas. Ecoscience 18 (1): 18–25. doi:10.2980/18-1-3345.

121

McKellar, Ann E. 2013. Winter rainfall predicts phenology in widely separated populations of a migrant songbird. Oecologia, in press. doi: 10.1007/s00442-

012-2520-8

Moore, F., and P. Kerlinger. 1987. Stopover and Fat Deposition by North American Wood-warblers (Parulinae) Following Spring Migration over the Gulf of Mexico. Oecologia 74 (1): 47–54. doi:10.1007/BF00377344.

Norris, D. Ryan, Peter P Marra, T. Kurt Kyser, Thomas W Sherry, and Laurene M Ratcliffe. 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:10.1098/rspb.2003.2569.

Nott, M. Philip, David F Desante, Rodney B Siegel, and Peter Pyle. 2002. Influences of the El Niño/Southern Oscillation and the North Atlantic Oscillation on Avian Productivity in Forests of the Pacific Northwest of North America. Global Ecology and Biogeography 11 (4): 333–342. doi:10.1046/j.1466-822X.2002.00296.x.

Pyle, Peter. 1997. Identification Guide to North American Birds, Part I: Columbidae to Ploceidae. Slate Creek Press.

Reudink, Matthew W, Peter P Marra, T. Kurt Kyser, Peter T Boag, Kathryn M Langin, and Laurene M Ratcliffe. 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:10.1098/rspb.2008.1452.

van Riper, Charles, Kristina Paxton, Carena van Riper, Kimberly van Riper, Laura McGrath, and J. Fontaine. 2008. The Role of Protected Areas as Bird Stop-over Habitat: Ecology and Habitat Utilization by Migrating Land Birds Within Colorado River Riparian Forests of Southwestern North America. Nebraska Cooperative Fish & Wildlife Research Unit -- Staff Publications - Paper 78. http://digitalcommons.unl.edu/ncfwrustaff/78.

Rockwell, Sarah M., Carol I. Bocetti, and Peter P. Marra. 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:10.1525/auk.2012.12003.

Schaub, Michael, Lukas Jenni, and Franz Bairlein. 2008. Fuel Stores, Fuel Accumulation, and the Decision to Depart from a Migration Stopover Site. Behavioral Ecology 19 (3): 657–666. doi:10.1093/beheco/arn023.

Sillett, T. Scott, Richard T. Holmes, and Thomas W. Sherry. 2000. Impacts of a Global Climate Cycle on Population Dynamics of a Migratory Songbird. Science 288 (5473): 2040–2042. doi:10.1126/science.288.5473.2040.

122

Skagen, Susan K., Jeffrey F. Kelly, Charles van Riper, Richard L. Hutto, Deborah M. Finch, David J. Krueper, and Cynthia P. Melcher. 2005a. Geography of Spring Landbird Migration Through Riparian Habitats in Southwestern North America. The Condor 107 (2): 212–227. Doi:10.1650/7807.

Skagen, Susan K., Rob Hazlewood, and Michael L. Scott. 2005b. The Importance and Future Condition of Western Riparian Ecosystems as Migratory Bird Habitat. USDA Forest Service Gen. Tech. Rep. PSW-GTR-191.

Studds, Colin E, and Peter P Marra. 2007. Linking Fluctuations in Rainfall to Nonbreeding Season Performance in a Long-distance Migratory Bird, Setophaga ruticilla. Climate Research 35: 115–122. doi:10.3354/cr00718.

Studds, Colin E., and Peter P. Marra. 2005. Nonbreeding Habitat Occupancy and Population Processes: An Upgrade Experiment with a Migratory Bird. Ecology 86 (9): 2380–2385. doi:10.1890/04-1145.

———. 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:10.1098/rspb.2011.0332.

Stutchbury, Bridget J. M., Scott A. Tarof, Tyler Done, Elizabeth Gow, Patrick M. Kramer, John Tautin, James W. Fox, and Vsevolod Afanasyev. 2009. Tracking Long-Distance Songbird Migration by Using Geolocators. Science 323 (5916): 896–896. doi:10.1126/science.1166664.

Tonra, Christopher M., Peter P. Marra, and Rebecca L. Holberton. 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:10.1111/j.1600-048X.2011.05333.x.

Wilson, Scott, Shannon L. Ladeau, Anders P. Tøttrup, and Peter P. Marra. 2011. Range-wide effects of breeding- and nonbreeding-season climate on the abundance of a Neotropical migrant songbird. Ecology 92 (9): 1789–1798.

123

Tables and Figures

124

Table 6.1. Overview of cross-seasonal studies involving Neotropical migrant passerines. Study number indicates study location in Figure 6.2.

125

Table 6.1. Continued

126

Table 6.1. Continued

127

Figure 6.1 (A) The suggested migratory routes used by Neotropical migrants between temperate and tropical regions (Lincoln 1998) and (B) nighttime wind vectors encountered by migrants during the March-May period (10-year average (2003-2012)). Average vectors were calculated within regions bounded by a longitudinal and latitudinal grid (see Appendix A). Boxed regions indicate the approximate area for which vectors were calculated, but are unadjusted for projection. Arrow length corresponds to wind strength with the exception of (*) where length was halved relative to other vectors in the interest of space.

128

Figure 6.2 The geographical distribution and main findings of cross-seasonal studies involving Neotropical migrant passerines. Evidence gathered from wintering isotopic signatures are summarized by age- and sex-class in (A), evidence from wintering climatological data is summarized in (B). Study details are presented in Table 6.1.

129

Appendices

130

Appendix A. Derivation of wind data presented in Figure 6.1

As also described in Chapter 5, wind on migration was derived from modeled climate data extracted from the National Center of Environmental Prediction (NCEP) Reanalysis 1 data archives at the NOAA-CIRES Climate Diagnostics Center at Boulder, Colorado, USA (NOAA NCEP 2012) using the RNCEP program (Kemp et al. 2012). This data has a spatial resolution of 2.5° latitude and longitude and temporal resolution of six hours. Vector data (U- and V-wind speed components (m/s)) were averaged from the 850mb (1500m) and 925mb (700m) level for the March-May period between 2003 and 2012. Noon values were dropped so that the series represented conditions encountered during nighttime migration (between 18:00h and 6:00h). Data was then averaged spatially for regions bounded by the longitudinal and latitudinal limits described below (Table 7.1).

Table 7.1. Region limits and night wind vectors (March-May averages, 2003-2012) presented in Figure 6.1(b)

Region Latitude °N Longitude °W U-wind (m/s) V-wind (m/s)

South America 5-15 72.5-57.5 -7.2 0.1

Southern Central America 5-15 92.5-75 -3.8 -1.1

Caribbean 17.5-27.5 92.5-72.5 -3.0 1.4

Mexico 17.5-27.5 115-95 0.4 0.6

Southeastern USA 30-40 90-67.5 4.2 1.0

South-central USA 30-40 112.5-93 2.0 2.4

Southwestern USA 30-40 125-115 2.6 -2.8

Northeastern USA 42.5-52.5 90-67.5 2.6 -0.4

North-central USA 42.5-52.5 112.5-93 1.5 0.2

Northwestern USA 42.5-52.5 125-115 2.0 1.1

Northwest Territories 55-65 135-112.5 0.6 1.1

Alaska/Yukon 55-65 160-137.5 -0.9 1.0

References:

Kemp, Michael U., E. Emiel van Loon, Judy Shamoun-Baranes, and Willem Bouten. 2012. RNCEP: Global Weather and Climate Data at Your Fingertips. Methods in Ecology and Evolution 3 (1): 65–70. doi:10.1111/j.2041-210X.2011.00138.x.

NOAA NCEP 2012. NCEP/NCAR Reanalysis 1. Earth System Research Laboratory, National Oceanic and Atmospheric Administration. www.esrl.noaa.gov/psd/data/gridded/data.ncep.reanalysis.html.