migratory stopover of songbirds in the western …...table 2.2- factor loadings for the first and...

Migratory Stopover of Songbirds in the Western Lake Erie Basin

THESIS

Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in

the Graduate School of The Ohio State University

By

Patrick Lyon Johnson

Graduate Program in Environment and Natural Resources

The Ohio State University

2013

Master's Examination Committee:

Dr. Paul Rodewald, Co-Advisor

Dr. Stephen Matthews, Co-Advisor

Dr. Robert Gates

Copyrighted by

Patrick Lyon Johnson

2013

ii

Abstract

Songbirds use multiple stopover locations to rest and refuel for subsequent flights

while migrating between breeding and non-breeding areas. Selection of high quality

stopover habitat may allow migrants to minimize time and energy spent in migration and

maximize fitness. Migrants should therefore attempt to select stopover habitat that affords

them suitable safety, shelter, and food. A better understanding of habitat attributes that

support high numbers of migrant landbirds during stopover is needed to develop

conservation strategies for these species, many of which are in decline. I examined how

migrant density during stopover in the Western Lake Erie Basin (WLEB) of Ohio was

influenced by local- and broad-scale habitat variables, specifically: patch vegetation

composition and structure, patch size and isolation, patch distance from the Lake Erie

shoreline, patch distance from a river or stream, and wetland cover surrounding a patch. I

used a generalized random tessellation stratified (GRTS) approach to select forested

study sites within 22 km of the lakeshore along a 70 km stretch of shoreline between

Toledo and Sandusky, Ohio, USA. Observers conducted over 800 point counts annually

from mid-April through late May in 2011 and 2012 at a total of 135 locations. Point

count data on Blackpoll Warbler (Setophaga striata), Black-throated Green Warbler (S.

iii

virens), and a guild of transient wood warblers (Parulidae) were analyzed using the Dail

and Madsen (2011) generalized hierarchical N-mixture model. This is an open-population

model that simultaneously estimates parameters that influence the abundance and

detection probability of study species. Detection probabilities for transient migrants

varied by survey technician, wind speed, and survey time, highlighting the importance of

accounting for detectability in bird migration studies. Broad-scale variables such as

distance to the lakeshore, patch isolation, and wetland cover were generally better

predictors of migrant abundance than proximity to a river or stream or patch area, and

local-scale variables (habitat structure and vegetation). Densities for both study species

were greatest in forest patches near the lakeshore. For the transient warbler guild,

densities declined about 3.4% per km from the lakeshore. Density of the transient warbler

guild was greater at sites that had more emergent aquatic midges (Chironomidae) in

2012, suggesting that midges could help explain the distribution and abundance of

migrants in the WLEB. While densities of transient migrants were greater near the coast,

inland forests often supported more transient migrants per patch than coastal forest

patches. Conservation efforts in the region should seek to protect, create, restore (1)

forested areas adjacent the lakeshore and (2) any forest within 0.5–10 km of the coast,

with priority given to forests that are larger (>20 ha) and closer to the lakeshore and

wetlands. Results from this study should be important in developing conservation plans

for migratory songbirds and for guiding decisions regarding local habitat management

and the placement of wind turbines within the landscape.

iv

For the birds

v

Acknowledgments

I would like to thank my advisors, Dr. Paul Rodewald & Dr. Stephen Matthews,

for their advice, support, and assistance throughout this project. I would also like to thank

my committee member, Dr. Robert Gates, for his role in shaping and encouraging my

research. The faculty and staff of the School of the Environment and Natural Resources

(SENR) provided logistical support for this research, specifically Olivia Ameredes, Mary

Capoccia, India Fuller, Dennis Hull, and Amy Schmidt. I owe thanks to a number of

current and former SENR graduate students and all of the members of the Terrestrial

Wildlife Ecology Lab, including: Kate Batdorf, Erin Cashion, Gabriel Colorado, Laura

Kearns, Mauri Liberati, Jenn Malpass, Molly McDermott, Desiree Narango, Keith Norris,

Ben Padilla, Dr. Amanda Rodewald, Linnea Rowse, Matt Shumar, Dave Slager, Jen

Thieme, Jason Tucker, and Karen Willard. This research would not have been possible

without the support of John Simpson and the Winous Point Marsh Conservancy, who

generously provided housing for field crews in 2011 and 2012. I appreciate all of the

landowners in the Western Lake Erie Basin who kindly gave me permission to conduct

bird surveys on their property. I am grateful for the hard work and dedication of my field

crews in 2011 (Annie Crary and Bob Riggs) and 2012 (Erin Cashion and Jackson Evans).

Thanks to Dave Ewert for his interest in my research and for providing feedback on

vi

earlier drafts of portions of this thesis. I owe thanks to the Black Swamp Bird

Observatory, and specifically Mark Shieldcastle, for feedback and guidance related to

this research. Thanks to Sarah Hadley and Richard Chandler for providing me with

example R code and answering many of my questions about hierarchical modeling.

Thanks to Chris Rimmer and Kent McFarland, for sparking and sustaining my

ornithological career. Thanks to all of my family and friends, without whom I would not

be where I am today. A special thanks to my parents, Claire Lyon and Peter Johnson, for

their wisdom and unconditional support. Thanks to my fiancée, Anne Mittnacht, for the

continuous love and support. This project was funded by the Ohio Department of Natural

Resources – Division of Wildlife and the School of Environment and Natural Resources

at The Ohio State University.

vii

Vita

Education

2007................................................................B.A. Conservation Biology, Middlebury

College, Middlebury, Vermont

Professional experience

2010-Present ..................................................Graduate Teaching and Research Associate,

The Ohio State University, Columbus, Ohio

2011................................................................Field technician, Ohio Breeding Bird Atlas

II, Ohio

2009-2010 ......................................................Conservation Biologist, Vermont Center for

Ecostudies, Norwich, Vermont & San

Francisco de Macoris, Dominican Republic

2006-2008 ......................................................Seasonal Field Biologist, Vermont Center

for Ecostudies, Norwich, Vermont

2007................................................................Field Biologist, The Nature Conservancy,

Killeen, Texas

viii

Fields of Study

Major Field: Environment and Natural Resources

ix

Table of Contents

Abstract ............................................................................................................................... ii

Acknowledgments............................................................................................................... v

Vita .................................................................................................................................... vii

Table of Contents ............................................................................................................... ix

List of Tables ..................................................................................................................... xi

List of Figures ................................................................................................................ xviii

Chapter 1 - Bird migration and migratory stopover............................................................ 1

Study overview ................................................................................................................ 1

Literature Cited ............................................................................................................... 8

Chapter 2 - The influence of local- and broad-scale habitat features on songbird

abundance during migratory stopover .............................................................................. 16

Introduction ................................................................................................................... 16

Study Area and Methods ............................................................................................... 20

Study area and site selection ..................................................................................... 20

Bird surveys ............................................................................................................... 21

x

Study species .............................................................................................................. 23

Local-scale variables ................................................................................................. 24

Broad-scale variables ................................................................................................ 26

Statistical methods ..................................................................................................... 28

Maps .......................................................................................................................... 32

Results ........................................................................................................................... 33

Blackpoll Warbler...................................................................................................... 33

Black-throated Green Warbler .................................................................................. 35

Transient warbler guild ............................................................................................. 36

Midges ....................................................................................................................... 38

Discussion ..................................................................................................................... 38

Broad-scale variables ................................................................................................ 38

Local-scale variables ................................................................................................. 43

Detection probability ................................................................................................. 45

Conclusions and management recommendations ...................................................... 46

Literature Cited ............................................................................................................. 48

Bibliography ................................................................................................................... 100

xi

List of Tables

Table 2.1 - Correlation matrix of habitat structure variables used in constructing habitat

structure PCA. Habitat structure variables were measured within 30 m of each of the 135

point count locations within the Western Lake Erie Basin of Ohio, USA. The habitat

variables used in the matrix were: stemhits = number of understory stems between 0.5

and 3.0 m, small trees = number of trees between 8–23 cm dbh, medium trees = number

of trees between 23–38 cm dbh, large trees = number of trees >38 cm dbh, medium and

large dead trees = number of medium (23–38 cm dbh), and large (>38 cm dbh) dead ash

trees, and average canopy height (m). .............................................................................. 60

Table 2.2- Factor loadings for the first and second principal component axes (PC1 and

PC2) in the habitat structure PCA. The first axis explained 29.2% of the total variation

(eigenvalue = 1.75). The second axis explained an additional 28.2% of the total variation

(eigenvalue = 1.69). The variables represent components of the habitat structure at each

of the 135 point count locations in the Western Lake Erie Basin, Ohio, USA. Variables

were measured within 30m of each point count location. The variables are: stemhits =

number of understory stems between 0.5 and 3.0 m, small trees = number of trees

between 8–23 cm dbh, medium trees = number of trees between 23–38 cm dbh, large

xii

trees = number of trees >38 cm dbh, medium and large dead trees = number of medium

(23–38 cm dbh), and large (>38 cm dbh) dead ash trees, and average canopy height (m).

........................................................................................................................................... 61

Table 2.3 - Correlation matrix of counts of trees >23 cm dbh in the eight most abundant

tree species groups used in constructing the species composition PCA. Counts of trees by

species were made within 30 m of each of the 135 point count locations in the Western

Lake Erie Basin of Ohio, USA. The species groups used in the matrix were: oaks

(Quercus spp.), maples (Acer spp.), boxelder (A. negundo), eastern cottonwood (Populus

deltoides), walnuts (Juglans spp.), hickories (Carya spp.), dead ashes (Fraxinus spp.),

and elm (Ulmus spp.). A Hellinger transformation was performed on the raw species

counts to make the data more appropriate for the PCA. ................................................... 62

Table 2.4- Factor loadings for the first and second principal component axes (PC1 and

PC2) of the species composition PCA. The first axis explained 27.7% of the total

variation (eigenvalue = 2.22). The second axis explained an additional 17.8% of the total

variation (eigenvalue = 1.42). The variables represent counts of trees >23 cm dbh within

the eight most abundant tree species groups. Counts were made within 30 m of each of

the 135 point count locations in the Western Lake Erie Basin, Ohio, USA. The species

groups are: oaks (Quercus spp.), maples (Acer spp.), boxelder (Acer negundo), eastern

cottonwood (Populus deltoides), walnuts (Juglans spp.), hickories (Carya spp.), dead

ashes (Fraxinus spp.), and elms (Ulmus spp.). ................................................................. 63

xiii

Table 2.5- Summary statistics for broad-scale variables for the 135 point count locations

used in 2011 and 2012. The variables are: area = area (ha) of the forest patch in which the

point count was conducted, distance to a river = distance (km) from point count location

to the nearest stream or river, distance to coast = distance (km) from point count location

to the nearest point of Lake Erie coastline, isolation = proportion of the landscape within

0.5 km of point count location classified as forest, wetland cover = proportion of the

landscape within 0.2 km of point count location classified as wetland. ........................... 64

Table 2.6 - Top models for gamma (arrival rate) for Blackpoll Warbler in 2011. nPars =

number of parameters in the model. AIC = Akaike Information Criterion. ΔAIC =

Difference in AIC between top-ranked model and specified model. Note that all models

include Julian date and a quadratic term for Julian date and that only models with ΔAIC <

10 are shown. Variables in the models are: area = area (ha) of the forest patch in which

the point count was conducted, distance to river = distance (km) from point count

location to nearest stream or river, distance to coast = distance (km) from point count

location to nearest point of Lake Erie coastline, isolation = proportion of the landscape

within 0.5 km of point count location classified as forest, wetland cover = proportion of

the landscape within 0.2 km of point count location classified as wetland, composition

PC1 = species composition PC1, composition PC2 = species composition PC2, structure

PC1 = habitat structure PC1, structure PC2 = habitat Structure PC2. FULL model (minus

area) = a model with all variables included except patch area (ha). ................................. 65

xiv




include Julian date and a quadratic term for Julian date and that only models with ΔAIC <

5 are shown. Variables in the models are: area = area (ha) of the forest patch in which the

point count was conducted, distance to river = distance (km) from point count location to

nearest stream or river, distance to coast = distance (km) from point count location to

nearest point of Lake Erie coastline, isolation = proportion of the landscape within 0.5 km

of point count location classified as forest, wetland cover = proportion of the landscape

within 0.2 km of point count location classified as wetland, composition PC1 = species

composition PC1, composition PC2 = species composition PC2, structure PC1 = habitat

structure PC1, structure PC2 = habitat Structure PC2. FULL model = a model with all

variables included except the variable inside the parentheses. ......................................... 66

Table 2.8- Coefficients for the top variables in the single-species models in each year.

95% confidence intervals for estimates are given inside the parentheses. Note that the

data are scaled and centered so interpretation of coefficients should be for strength and

direction of relationship. BLPW = Blackpoll Warbler. BTNW = Black-throated Green

Warbler. The variables are: area = area (ha) of the forest patch in which the point count

was conducted, distance to river = distance (km) from point count location to the nearest

stream or river, distance to coast = distance (km) from point count location to nearest

point of Lake Erie coastline, isolation = proportion of the landscape within 0.5 km of

xv

point count location classified as forest, wetland cover = proportion of the landscape

within 0.2 km of point count location classified as wetland, composition PC1 = species

composition PC1, composition PC2 = species composition PC2, structure PC1 = habitat

structure PC1, structure PC2 = habitat Structure PC2. ..................................................... 67

Table 2.9 - Top models for gamma (arrival rate) for Black-throated Green Warbler in

2011. nPars = number of parameters in the model. AIC = Akaike Information Criterion.

ΔAIC = Difference in AIC between top-ranked model and specified model. Note that all

models include Julian date and a quadratic term for Julian date and that only models with

ΔAIC < 4 are shown. Variables in the models are: area = area (ha) of the forest patch in

which the point count was conducted, distance to river = distance (km) from point count






PC1 = habitat structure PC1, structure PC2 = habitat Structure PC2. .............................. 68





xvi









Table 2.11 – Top gamma (finite rate of increase) models for the transient warbler guild in












xvii

Table 2.12 – Top gamma (finite rate of increase) models for the transient warbler guild in












xviii

List of Figures

Figure 2.1 – Location of the study area (defined in purple in the upper right and bottom)

in the Western Lake Erie Basin, Ohio, USA. In the upper left and right, darker grey areas

bordered in black represent the state of Ohio. Yellow areas in the upper right and left

represent the counties in which the study took place. ....................................................... 72

Figure 2.2 - Point count locations (dots) used in 2011 (grey), in 2012 (white), and in both

years (black) within the study area in the Western Lake Erie Basin, Ohio, USA. Distance

bands used in the stratified GRTS sampling are shown in yellow (0–0.573 km), orange

(0.573–2.648 km), pink (2.648–10.09 km), and blue (10.09–22 km). Forest patches

within the study area are shown in green. ......................................................................... 73

Figure 2.3- Detection probability for Blackpoll Warbler in 2011 as a function of observer

and time of day. Grey areas represent 95% confidence intervals for estimated detection

probabilities for each observer (colored lines).................................................................. 74

xix

Figure 2.4 - Number of Blackpoll Warblers gained between visits to a site (new arrivals)

in the Western Lake Erie Basin, Ohio, USA as a function of patch isolation in 2011. The

grey area represents the 95% confidence interval of the predicted estimate (black line). 75


in the Western Lake Erie Basin, Ohio, USA as a function of date in 2011. The grey area

represents the 95% confidence interval of the predicted estimate (black line). ................ 76

Figure 2.6 - Detection probability for Blackpoll Warblers in the Western Lake Erie Basin,

Ohio, USA in 2012 as a function of a categorical estimate for wind speed. Dots represent

estimates for detection probability and whiskers represent 95% confidence intervals of

the estimates. ..................................................................................................................... 77


in the Western Lake Erie Basin, Ohio, USA as a function of patch distance from the

lakeshore in 2012. The grey area represents the 95% confidence interval of the predicted

estimate (black line). ......................................................................................................... 78

Figure 2.8- Coefficient estimates and 95% confidence intervals for variables influencing

gamma (arrival rate) in the top models for Blackpoll Warbler and Black-throated Green

Warbler in 2011 and 2012. The red-dotted line is at X = 0. Variables are: Structure 1 =

habitat structure PC1, Structure 2 = habitat Structure PC2, Distance from coast = distance

xx

(km) from point count location to nearest point of Lake Erie coastline, Area = area (ha) of

the forest patch in which the point count was conducted, Composition 2 = species

composition PC2, Composition 1 = species composition PC1, Distance to a River =

distance (km) from point count location to nearest stream or river, Date - squared =

quadratic of Julian Date, Patch isolation = proportion of the landscape within 0.5 km of

point count location classified as forest, Date = Julian date, Wetland cover = proportion

of the landscape within 0.2 km of point count location classified as wetland. ................. 79

Figure 2.9 - Average density of Blackpoll Warblers (individuals per hectare) in forest

patches of the Western Lake Erie Basin, Ohio, USA, during each day of 2011 and 2012

spring migrations (approximately 5 May to 30 May annually). ....................................... 80

Figure 2.10 - Average number of Blackpoll Warblers per forest patch in the Western

Lake Erie Basin, Ohio, USA, during each day of 2011 and 2012 spring migrations

(approximately 5 May to 30 May annually). .................................................................... 81

Figure 2.11- Detection probability for Black-throated Green Warblers in the Western

Lake Erie Basin, Ohio, USA in 2011 as a function of observer and time of day. Grey

areas represent 95% confidence intervals for estimated detection probabilities for each

observer (colored lines)..................................................................................................... 82

xxi

Figure 2.12 - Number of Black-throated Green Warblers gained between visits to a site

(new arrivals) in the Western Lake Erie Basin, Ohio, USA as a function of date in 2011.

The grey area represents the 95% confidence interval of the predicted estimate (black

line). .................................................................................................................................. 83

Figure 2.13 - Estimated detection probabilities for Black-throated Green Warbler for

different observers in the Western Lake Erie Basin, Ohio, USA in 2012. Lines coming off

of estimates represent 95% Confidence intervals. ............................................................ 84


(new arrivals) in the Western Lake Erie Basin, Ohio, USA in 2012 as a function of

distance from the coast. The grey area represents the 95% confidence interval of the

predicted estimate (black line). ......................................................................................... 85


(new arrivals) in the Western Lake Erie Basin, Ohio, USA in 2012 as a function of date.

The grey area represents the 95% confidence interval of the predicted estimate (black

line). .................................................................................................................................. 86

Figure 2.16- Average density of Black-throated Green Warblers (individuals per hectare)

within forest patches of the Western Lake Erie Basin, Ohio, USA, during each day of

2011 and 2012 spring migrations (approximately 21 April – 29 May annually). ............ 87

xxii

Figure 2.17- Average number of Black-throated Green Warblers per forest patch in the

Western Lake Erie Basin, Ohio, USA, during each day of 2011 and 2012 spring

migrations (approximately 21 April to 29 May annually). ............................................... 88

Figure 2.18 – Raw counts of individuals in the transient warbler guild by species and

year. Counts were made during spring migration in the Western Lake Erie Basin, Ohio,

USA. The species are: Black-and-white Warbler (BAWW), Bay-breasted Warbler

(BBWA), Blackburnian Warbler (BLBW), Blackpoll Warbler (BLPW), Black-throated

Blue Warbler (BTBW), Black-throated Green Warbler (BTNW), Blue-winged Warbler

(BWWA), Canada Warbler (CAWA), Chestnut-sided Warbler (CSWA), Cerulean

Warbler (CERW), Cape May Warbler (CMWA), Golden-winged Warbler (GWWA),

Hooded Warbler (HOWA), Magnolia Warbler (MAWA), Yellow-rumped Warbler

(MYWA), Nashville Warbler (NAWA), Northern Parula (NOPA), Orange-crowned

Warbler (OCWA), Palm Warbler (PAWA), Pine Warbler (PIWA), Prairie Warbler

(PRAW), Tennessee Warbler (TEWA), Wilson’s Warbler (WIWA), and Yellow-throated

Warbler (YTWA). ............................................................................................................. 89

Figure 2.19 - Detection probability for the transient warbler guild in the Western Lake

Erie Basin, Ohio, USA as a function of wind speed and observer in 2011. (0) = no wind,

(5) = high wind. Whiskers indicate 95% confidence intervals of estimates. .................... 90

xxiii

Figure 2.20 - Detection probability for transient warbler guild in the Western Lake Erie

Basin, Ohio, USA as a function of time of day and observer in 2011. Grey areas represent

95% confidence intervals for estimated detection probabilities for each observer (colored

lines). ................................................................................................................................. 91



(4) = high wind. Whiskers indicate 95% confidence intervals of estimates. .................... 92


Erie Basin, Ohio, USA as a function of time of day and observer in 2012. Grey areas

represent 95% confidence intervals for estimated detection probabilities for each observer

(colored lines). .................................................................................................................. 93

Figure 2.23 – Average density of guild members (individuals per hectare) within forest

patches of the Western Lake Erie Basin, Ohio, USA, during each day of 2011 and 2012

spring migrations (approximately 19 April to 30 May annually). .................................... 94

Figure 2.24 - Average number of guild members per forest patch within the Western

Lake Erie Basin, Ohio, USA, during each day of 2011 and 2012 spring migrations

(approximately 19 April to 30 May annually). ................................................................. 95

xxiv

Figure 2.25 - Boxplot showing estimated midge abundance at survey locations in the

Western Lake Erie Basin, Ohio, USA, in relation to survey date in 2012. Categories used

to estimate midge abundance were: (0) = 0 midges, (1) = 1–10 midges, (2) = 11–100

midges, (3) = 101–500 midges, (4) = 501–1000 midges, (5) = 1001–5000 midges, (6) =

>5001 midges. ................................................................................................................... 96

Figure 2.26 – Boxplot showing estimated midge abundance at survey locations in the

Western Lake Erie Basin, Ohio, USA, in relation to wetland cover in the landscape in

2012. Categories used to estimate midge abundance were: (0) = 0 midges, (1) = 1–10


1001–5000 midges, (6) = >5001 midges. ......................................................................... 97

Figure 2.27- Boxplot showing estimated midge abundance at survey locations in the

Western Lake Erie Basin, Ohio, USA, in relation to distance to coast in 2012. Categories

used to estimate midge abundance were: (0) = 0 midges, (1) = 1-10 midges, (2) = 11-100

midges, (3) = 101-500 midges, (4) = 501-1000 midges, (5) = 1000-5000 midges, (6) =

>5001 midges. ................................................................................................................... 98

Figure 2.28 – Pairwise comparisons between transient warbler abundance in each midge

abundance category while accounting for Julian date and a quadratic of Julian date.

Estimates for the difference in transient warbler abundance between midge abundance

categories (Y-axis) are shown on the plot as dots. 95% Tukey family-wise confidence

xxv

intervals extend from estimated differences. Categories used to estimate midge

abundance were: (0) = 0 midges, (1) = 1-10 midges, (2) = 11-100 midges, (3) = 101-500

midges, (4) = 501-1000 midges, (5) = 1000-5000 midges, (6) = >5001 midges. A dotted

line is at X = 0. ‘*’ = P < 0.05, ‘**’ = P < 0.01, ‘***’ = P < 0.001. ................................ 99

1

Chapter 1 - Bird migration and migratory stopover

Study overview

Migration is a life-history strategy that allows mobile organisms to take advantage

of seasonal changes in environmental conditions (i.e., more food, water, shelter, space

and or fewer competitors and predators) in order to maximize their individual fitness

(Dingle and Drake 2007, Pulido 2007, Ramenofsky and Wingfield 2007). Migration is an

adaptation that has evolved in a large and diverse group of organisms (Bowlin et al. 2010,

Dingle 1980), and is prevalent among birds, with over 50% of the world’s 10,000 known

living species (roughly 50 billion individuals) undertaking some form of migration

annually (Berthold 1998). Migratory birds have evolved a variety of different migration

strategies (e.g., long-distance, partial, altitudinal, differential) to capitalize on seasonal

resources and opportunities (Alerstam and Hedenström 1998, Chapman et al. 2011,

Dingle and Drake 2007). Long-distance migration is perhaps the best-studied migration

strategy, and is used by approximately 36-38% of the world’s migratory avifauna (Kirby

et al. 2008, Sekercioglu 2007). Long-distance migration is believed to have been

common amongst birds for over 24 million years based on evidence in the fossil record

(Steadman 2005), but how this strategy evolved is not definitively known (Alerstam and

Enckell 1979, Bell 2000, Bruderer and Salewski 2008, Cox 1968, Cox 1985, Rappole and

Jones 2002, Salewski and Bruderer 2007).

2

Hundreds of species of landbirds migrate between non-breeding grounds in the

tropics of Central America, South America, and the Caribbean and temperate breeding

grounds in North America every spring (Greenberg 1980). These Nearctic-Neotropical

(hereafter referred to as simply Neotropical) migrants, may benefit by travelling to breed

in seasonally resource-rich environments in North America because with more food and

space available they can support larger clutch sizes (Alerstam et al. 2003, Greenberg

1980, Jetz et al. 2008, Lack 1968, Rappole and Jones 2002, Taverner 1904). Neotropical

migrants also benefit by travelling to the tropics during the winter, thereby avoiding

freezing temperatures and resource scarcity common at higher latitudes, and presumably

increasing their chance of survival (Alerstam et al. 2003, Greenberg 1980, Lack 1968,

Taverner 1904). Thus, migrants are able to successfully “glean the best of two worlds”

(Greenberg 1980). Yet, the benefits of migration, in terms of increased survival during

the non-breeding period and increased reproduction during the breeding period, are

balanced by the cost of high mortality associated with travelling between distant areas

(Paxton et al. 2007, Sillett and Holmes 2002). These costs are likely especially high for

hatch-year birds that have never experienced migration (Greenberg 1980, Moore et al.

1995).

Migration is thought to be the period of highest mortality in the annual cycle of

migratory landbirds, and likely plays a major role in limiting their populations (Holmes

2007, Johnson et al. 2006, Moore and Aborn 2000, Moore et al. 2005, Sillett and Holmes

2002). Analyses on long-term datasets from Europe (Berthold et al. 1998, Sanderson et

al. 2006) and North America (Robbins et al. 1989, Sauer et al. 2011) have revealed that

long-distance migratory bird populations are declining significantly faster than either

3

resident or short-distance migrant populations. Furthermore, Sillett and Holmes (2002)

estimated that 87-89 % of Black-throated Blue Warblers (Setophaga caerulescens) die

while away from the breeding and wintering grounds, presumably during migration.

Similarly, Paxton et al. (2007) found that 64% of annual mortality in Southwestern-

Willow Flycatchers (Empidonax traillii extimus) was concentrated during migratory

periods. Taken together, this evidence suggests that high morality during migration may

be typical for migratory landbirds.

Long-distance migratory birds have to stop to rest and refuel in multiple locations

while en route to their breeding or wintering grounds in order to acquire the energy

needed to continue their migration (Berthold 1975, Blem 1980, Hedenström and

Alerstam 1997). Migrants spend the majority of the time and energy utilized during

migration in these unfamiliar stopover habitats (Catry et al. 2004, Cochran and Wikelski

2005, Hedenström and Alerstam 1997, Lindström 2005, Wikelski et al. 2003), searching

for food, recuperating from long-flights, and taking shelter, before making subsequent

movements toward their final destinations. Migrants that quickly refuel at a stopover

location may be able to shorten the duration of their stopover (Biebach et al. 1986,

Cherry 1982, Loria and Moore 1990, Matthews and Rodewald 2010, Moore and

Kerlinger 1987), and consequently the time required to complete their migration (Moore

et al. 1995, Newton 2006, Tøttrup et al. 2012). By shortening time spent in migration,

individuals may be able to arrive earlier on the breeding grounds (Moore et al. 1995,

Tøttrup et al. 2012), and secure higher quality territories (Aebischer et al. 1996, Sergio

and Newton 2003, Smith and Moore 2005) and mates (Møller 1994), and experience

greater reproductive success (Hochachka 1990, Lozano et al. 1996, Moore et al. 2005,

4

Møller 1994, Sandberg and Moore 1996a, Smith and Moore 2003, Smith and Moore

2005). Therefore, migrants that can successfully navigate the landscape, locating high

quality stopover sites where they can quickly and safely rest and refuel, likely benefit

through increased survival and reproduction (Baker et al. 2004, Cohen et al. 2012, Drent

et al. 2006, Moore et al. 1995, Newton 2006, Weber et al. 1999).

Humans have radically altered terrestrial landscapes, reducing the amount of

habitat available to migratory birds during all phases of their annual cycle (Kirby et al.

2008). These changes have likely exacerbated the challenges migrants face during

migratory periods (Klaassen et al. 2012). For example, stopover habitats are smaller and

farther apart in fragmented landscapes, making it more difficult for migrants to locate

sufficient areas of quality stopover habitat (Klaassen et al. 2012). As a consequence of

limited habitat availability, migrants may become concentrated in remnant patches where

they could be more apt to experience increased competition and lower food availability

(Kelly et al. 2002, Moore et al. 1995), and more susceptible to disease or predation

(Klaassen et al. 2012). The challenges migrants face in fragmented landscapes may delay

their migration and have important consequences for their future survival and

reproduction (Baker et al. 2004, Moore et al. 1995, Weber et al. 1999).

Results from long-term monitoring efforts in North America have revealed that

populations of a number of Neotropical migratory bird species have declined over the last

several decades (Askins et al. 1990, Peterjohn et al. 1995, Sauer et al. 2011). These

findings have sparked research focused on identifying where, how, and why these species

are declining. The loss and fragmentation of habitats that birds utilize on the wintering

grounds (Keller and Yahner 2006, Rappole and McDonald 1994, Robbins et al. 1989,

5

Sherry and Holmes 1996) and on the breeding grounds (Böhning-Gaese et al. 1993,

Newton 2004) are often cited as the most likely causes for population declines. However,

it is now widely recognized that events occurring during migration, including the loss and

degradation of habitat available to migrants en route, may also play role in population

declines (Moore and Simons 1992, Moore et al. 1995, Newton 2006, Sherry and Holmes

1995). However, it is unlikely that declines in any population of migratory birds can be

attributed solely to events occurring in any one season (Faaborg et al. 2010b, Latta and

Baltz 1997, Sherry and Holmes 1995, Sherry and Holmes 1993). Consequently,

conservation plans must incorporate strategies for breeding, wintering, and migratory

periods to effectively manage migratory species (Faaborg et al. 2010a, Mehlman et al.

2005, Sheehy et al. 2011, Sillett and Holmes 2002), and recognize that events that occur

in one season can affect events in subsequent seasons (Bearhop et al. 2004, Gunnarsson

et al. 2005, Harrison et al. 2010, Holmes 2007, Marra et al. 1998, Mitchell et al. 2012,

Newton 2004, Newton 2006, Norris et al. 2004, Tøttrup et al. 2012). Our ability to

develop comprehensive conservation plans for migratory landbirds is currently hampered

by our incomplete knowledge of their ecology, especially on the wintering grounds and

during migration (Faaborg et al. 2010b, Mehlman et al. 2005, Petit 2000). Given the

ongoing population declines for a number of Neotropical migrants and ongoing habitat

loss throughout their ranges, there is an urgent need to identify and conserve important

habitats for these populations during all phases of their annual cycle (Faaborg et al.

2010a, Mehlman et al. 2005).

The conservation of a “network of stopover sites” that encompasses the full extent

of migratory routes is critical to the conservation of migratory species (Mehlman et al.

6

2005, Petit 2000). However, little is known about how migrants select stopover habitats

(Chernetsov 2006, Moore and Aborn 2000) or which habitat attributes consistently

support high numbers of migratory birds during stopover periods (Petit 2000).

Furthermore, the ways that migrants use stopover habitats may vary geographically,

temporally, or in relation to habitat availability in the landscape (Andrén 1994, Mehlman

et al. 2005, Petit 2000), suggesting that more research is needed during both spring and

fall migration and in a variety of regions and landscape configurations. Prior stopover

research has emphasized the importance of local habitat attributes (Rodewald and

Brittingham 2004, Rodewald and Brittingham 2007), but very few studies have

considered how patch- and broad-scale habitat features influence the abundance of

migrants (Buler et al. 2007, Buler and Moore 2011). Since migrants select stopover

habitat at multiple spatial-scales (Bowlin et al. 2005, Buler et al. 2007, Buler and Moore

2011, Hutto 1985b), it is important to evaluate how both broad and local features of the

landscape affect the distribution and abundance of migrants (Cushman and McGarigal

2002, Moore et al. 2005, Wiens 1989).

The goal of my research was to identify local- and broad-scale attributes of

stopover habitat associated with use by migrant songbirds in the Western Lake Erie Basin

(WLEB) of Ohio. Migrating songbirds concentrate in immense numbers during spring

and fall migration within shoreline forest patches of the WLEB (Ewert et al. 2006,

Rodewald 2007). Additional data on distribution and habitat-relationships of migrants

within the broader WLEB is needed as this region has been targeted for wind power

development (Black Swamp Bird Observatory 2011). I examined how abundance of

transient migrants varied with respect to local and broad-scale features, specifically: 1)

7

patch vegetation composition and structure, 2) patch size and isolation, 3) patch distance

from the lakeshore, 4) patch distance from a river or stream, and 5) wetland cover

surrounding a patch. Results from this study should be important in developing

conservation plans for migratory songbirds and for guiding decisions regarding local

habitat management and the placement of wind turbines within the landscape.

8

Literature Cited

Aebischer, A., N. Perrin, M. Krieg, J. Studer, and D.R. Meyer. 1996. The role of territory

choice, mate choice and arrival date on breeding success in the Savi's Warbler

(Locustella luscinioides). Journal of Avian Biology 27:143-152.

Alerstam, T., and A. Hedenström. 1998. The development of bird migration theory.

Journal of Avian Biology 29:343-369.

Alerstam, T., A. Hedenström, and S. Åkesson. 2003. Long-distance migration: evolution

and determinants. Oikos 103:247-260.

Alerstam, T., and P.H. Enckell. 1979. Unpredictable habitats and evolution of bird

migration. Oikos 33:228-232.

Andrén, H. 1994. Effects of habitat fragmentation on birds and mammals in landscapes

with different proportions of suitable habitat - a review. Oikos 71:355-366.

Askins, R.A., J.F. Lynch, and R. Greenberg. 1990. Population declines in migratory birds

in eastern North America. Pages 1-57 In D. M. Power, editor. Current Ornitholgy,

Volume 7. Plenum Press, New York.

Baker, A.J., P.M. González, T. Piersma, L.J. Niles, de Lima Serrano do Nascimento,

Inês, P.W. Atkinson, N.A. Clark, C.D.T. Minton, M.K. Peck, and G. Aarts. 2004.

Rapid population decline in red knots: fitness consequences of decreased refuelling

rates and late arrival in Delaware Bay. Proceedings of the Royal Society Series B

271:875-882.

Bearhop, S., G.M. Hilton, S.C. Votier, and S. Waldron. 2004. Stable isotope ratios

indicate that body condition in migrating passerines is influenced by winter habitat.

Proceedings of the Royal Society of London Series B 271:215-218.

Bell, C.P. 2000. Process in the evolution of bird migration and pattern in avian

ecogeography. Journal of Avian Biology 31:258-265.

Berthold, P. 1998. Bird migration: genetic programs with high adaptability. Zoology

101:235-245.

Berthold, P., W. Fiedler, R. Schlenker, and U. Querner. 1998. 25-year study of the

population development of central European songbirds: A general decline most

evident in long-distance migrants. Naturwissenschaften 85:350-353.

9

Berthold, P. 1975. Migration: control and metabolic physiology. Pages 77-128 In D. S.

Farner and J. R. King, editors. Avian Biology, Volume 5. Academic Press, New

York.

Biebach, H., W. Friedrich, and G. Heine. 1986. Interaction of bodymass, fat, foraging and

stopover period in trans-Sahara migrating passerine birds. Oecologia 69:370-379.

Black Swamp Bird Observatory. 2011. Black Swamp Bird Observatory speaks out

against wind turbines in migratory bird stopover habitat. URL:

http://www.bsbo.org/wind_energy.htm. Accessed: 11/28/2012.

Blem, C.R. 1980. The energetics of migration. Pages 175 - 224 In S. A. Gauthreaux Jr.,

editor. Animal migration, orientation, and navigation. Academic Press, New York.

Böhning-Gaese, K., M.L. Taper, and J.H. Brown. 1993. Are declines in North American

insectivorous songbirds due to causes on the breeding range. Conservation Biology

7:76-86.

Bowlin, M.S., W.W. Cochran, and M.C. Wikelski. 2005. Biotelemetry of new world

thrushes during migration: physiology, energetics and orientation in the wild.

Integrative and Comparative Biology 45:295-304.

Bowlin, M.S., I. Bisson, J. Shamoun-Baranes, J.D. Reichard, N. Sapir, P.P. Marra, T.H.

Kunz, D.S. Wilcove, A. Hedenström, C.G. Guglielmo, S. Åkesson, M. Ramenofsky,

and M. Wikelski. 2010. Grand challenges in migration biology. Integrative and

Comparative Biology 50:261-279.

Bruderer, B., and V. Salewski. 2008. Evolution of bird migration in a biogeographical

context. Journal of Biogeography 35:1951-1959.

Buler, J.J., and F.R. Moore. 2011. Migrant-habitat relationships during stopover along an

ecological barrier: extrinsic constraints and conservation implications. Journal of

Ornithology 152:101-112.

Buler, J.J., F.R. Moore, and S. Woltmann. 2007. A multi-scale examination of stopover

habitat use by birds. Ecology 88:1789-1802.

Catry, P., V. Encarnação, A. Araújo, P. Fearon, A. Fearon, M. Armelin, and P. Delaloye.

2004. Are long-distance migrant passerines faithful to their stopover sites? Journal of

Avian Biology 35:170-181.

Chapman, B.B., C. Brönmark, J.Å. Nilsson, and L.A. Hansson. 2011. The ecology and

evolution of partial migration. Oikos 120:1764-1775.

Chernetsov, N. 2006. Habitat selection by nocturnal passerine migrants en route:

mechanisms and results. Journal of Ornithology 147:185-191.

http://www.bsbo.org/wind_energy.htm

10

Cherry, J.D. 1982. Fat deposition and length of stopover of migrant White-crowned

Sparrows. Auk 99:725-732.

Cochran, W.W., M. Wikelski. 2005. Individual migratory tactics of New World Catharus

thrushes. Pages 274-289 In R. Greenberg and P. P. Marra, editors. Birds of two

worlds. The Johns Hopkins University Press, Baltimore.

Cohen, E.B., F.R. Moore, and R.A. Fischer. 2012. Experimental evidence for the

interplay of exogenous and endogenous factors on the movement ecology of a

migrating songbird. PloS ONE 7: e41818. doi: 10.1371/journal.pone.0041818.

Cox, G.W. 1985. The evolution of avian migration systems between temperate and

tropical regions of the new world. American Naturalist 126:451-474.

Cox, G.W. 1968. The role of competition in the evolution of migration. Evolution

22:180-192.

Cushman, S.A., and K. McGarigal. 2002. Hierarchical, multi-scale decomposition of

species-environment relationships. Landscape Ecology 17:637-646.

Dingle, H. 1980. Ecology and evolution of migration. Pages 1-101 In S. A. Gauthreaux

Jr., editor. Animal migration, orientation, and navigation. Academic Press, New

York.

Dingle, H., and V.A. Drake. 2007. What is migration? Bioscience 57:113-121.

Drent, R., A. Fox, and J. Stahl. 2006. Travelling to breed. Journal of Ornithology

147:122-134.

Ewert, D.N., J.G. Soulliere, R.D. Macleod, M.C. Shieldcastle, P.G. Rodewald, E.

Fujimura, J. Shieldcastle, and R.J. Gates. 2006. Migratory bird stopover site

attributes in the Western Lake Erie Basin. Final report to The George Gund

Foundation.

Faaborg, J., R.T. Holmes, A.D. Anders, K.L. Bildstein, K.M. Dugger, S.A. Gauthreaux

Jr., P. Heglund, K.A. Hobson, A.E. Jahn, D.H. Johnson, S.C. Latta, D.J. Levey, P.P.

Marra, C.L. Merkord, E. Nol, S.I. Rothstein, T.W. Sherry, S.T. Sillett, F.R.

Thompson III, and N. Warnock. 2010a. Conserving migratory land birds in the New

World: do we know enough? Ecological Applications 20:398-418.




Thompson III, and N. Warnock. 2010b. Recent advances in understanding migration

systems of New World land birds. Ecological Monographs 80:3-48.

11

Greenberg, R. 1980. Demographic aspects of long-distance migration. Pages 493-504 In

A. Keast and E. Morton, editors. Migrant birds in the Neotropics: ecology, behavior,

distribution, and conservation. Smithsonian Institution Press, Washington D.C.

Gunnarsson, T.G., J.A. Gill, J. Newton, P.M. Potts, and W.J. Sutherland. 2005. Seasonal

matching of habitat quality and fitness in a migratory bird. Proceedings of the Royal

Society Series B 272:2319-2323.

Harrison, X.A., J.D. Blount, R. Inger, R.D. Norris, and S. Bearhop. 2010. Carry-over

effects as drivers of fitness differences in animals. Journal of Animal Ecology 80:4-

18.

Hedenström, A., and T. Alerstam. 1997. Optimum fuel loads in migratory birds:

distinguishing between time and energy minimization. Journal of Theoretical

Biology 189:227-234.

Hochachka, W. 1990. Seasonal decline in reproductive performance of Song Sparrows.

Ecology 71:1279-1288.

Holmes, R.T. 2007. Understanding population change in migratory songbirds: long-term

and experimental studies of Neotropical migrants in breeding and wintering areas.

Ibis 149:2-13.

Hutto, R.L. 1985. Habitat selection by non-breeding, migratory land birds. Pages 455-476

In M. L. Cody, editor. Habitat selection in birds. Academic Press, Inc., Orlando.

Jetz, W., C.H. Sekercioglu, and K. Böhning-Gaese. 2008. The worldwide variation in

avian clutch size across species and space. Plos Biology 6: e303. doi:

10.1371/journal.pbio.0060303.

Johnson, M.D., T.W. Sherry, R.T. Holmes, and P.P. Marra. 2006. Assessing habitat

quality for a migratory songbird wintering in natural and agricultural habitats.

Conservation Biology 20:1433-1444.

Keller, G.S., and R.H. Yahner. 2006. Declines of migratory songbirds: evidence for

wintering-ground causes. Northeastern Naturalist 13:83-92.

Kelly, J.F., L.S. DeLay, and D.M. Finch. 2002. Density-dependent mass gain by Wilson's

Warblers during stopover. Auk 119:210-213.

Kirby, J.S., A.J. Stattersfield, S.H.M. Butchart, M.I. Evans, R.F.A. Grimmett, V.R. Jones,

J. O'Sullivan, G.M. Tucker, and I. Newton. 2008. Key conservation issues for

migratory land and waterbird species on the world's major flyways. Bird

Conservation International 18:49-73.

12

Klaassen, M., B.J. Hoye, B.A. Nolet, and W.A. Buttemer. 2012. Ecophysiology of avian

migration in the face of current global hazards. Philosophical Transactions of the

Royal Society Series B 367:1719-1732.

Lack, D. 1968. Bird migration and natural selection. Oikos 19:1-9.

Latta, S.C., and M.E. Baltz. 1997. Population limitation in Neotropical migratory birds:

comments. Auk 114:754-762.

Lindström, Å. 2005. Overview: migration itself. Pages 249 - 251 In R. Greenberg and P.

P. Marra, editors. Birds of Two Worlds. The Johns Hopkins University Press,

Baltimore.

Loria, D.E., and F.R. Moore. 1990. Energy demands of migration on Red-eyed Vireos

(Vireo olivaceus). Behavioral Ecology 1:24-35.

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

27:164-170.

Marra, P.P., K.A. Hobson, and R.T. Holmes. 1998. Linking winter and summer events in

a migratory bird by using stable-carbon isotopes. Science 282:1884-1886.

Matthews, S.N., and P.G. Rodewald. 2010. Urban forest patches and stopover duration of

migratory Swainson's Thrushes. Condor 112:96-104.

Mehlman, D.W., S.E. Mabey, D.N. Ewert, C. Duncan, B. Abel, D. Cimprich, R.D. Sutter,

and M. Woodrey. 2005. Conserving stopover sites for forest-dwelling migratory

landbirds. Auk 122:1281-1290.

Mitchell, G.W., A.E.M. Newman, M. Wikelski, and D.R. Norris. 2012. Timing of

breeding carries over to influence migratory departure in a songbird: an automated

radiotracking study. Journal of Animal Ecology 81:1024-1033.

Møller, A.P. 1994. Phenotype-dependent arrival time and its consequences in a migratory

bird. Behavioral Ecology and Sociobiology 35:115-122.

Moore, F.R., and D.A. Aborn. 2000. Mechanisms of en route habitat selection: How do

migrants make habitat decisions during stopover. Studies in Avian Biology 20:34-

42.

Moore, F.R., T.R. Simons. 1992. Habitat suitability and stopover ecology of Neotropical

landbird migrants. Pages 345-355 In J. M. Hagan and D. W. Johnston, editors.

Ecology and conservation of Neotropical migrant landbirds. Smithsonian Institution,

Washington.

13

Moore, F.R., S.A. Gauthreaux Jr., P. Kerlinger, and T.R. Simons. 1995. Habitat

requirements during migration: important link in conservation. In T. E. Martin and

D. M. Finch, editors. Ecology and management of Neotropical migratory birds. New

York.

Moore, F.R., M.S. Woodrey, J.J. Buler, S. Woltmann, and T.R. Simons. 2005.

Understanding the stopover of migratory birds: A scale dependent approach. Pages:

684-689 In C.J. Ralph and T.D. Rich, editors. Bird conservation implementation and

integration in the Americas: Proceedings of the Third International Partners in Flight

Conference, 20-24 March 2002. Albany, California.

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

74:47-54.

Moore, F.R., R.J. Smith, and R. Sandberg. 2005. Stopover ecology of intercontinental

migrants: en route problems and consequences for reproductive performance. Pages

251-261 In R. Greenberg and P. P. Marra, editors. Birds of two worlds. The Johns

Hopkins University Press, Baltimore.

Newton, I. 2006. Can conditions experienced during migration limit the population levels

of birds? Journal of Ornithology 147:146-166.

Newton, I. 2004. Population limitation in migrants. Ibis 146:197-226.

Norris, D.R., P.P. Marra, T.K. Kyser, T.W. Sherry, and L.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 271:59-64.

Paxton, E.H., M.K. Sogge, S.L. Durst, T.C. Theimer, and J.R. Hatten. 2007. The ecology

of the Southwestern Willow Flycatcher in central Arizona—a 10-year synthesis

report: U.S. Geological Survey Open-File Report 2007-1381.

Peterjohn, B.G., J.R. Sauer, and C.S. Robbins. 1995. Population trends from the North

American Breeding Bird Survey. Pages 3-39 In T. E. Martin and D. Finch M.,

editors. Ecology and management of Neotropical migratory birds. Oxford University

Press, New York.

Petit, D.R. 2000. Habitat use by landbirds along Nearctic-Neotropical migration routes:

implications for conservation of stopover habitats. Studies in Avian Biology 20:15 -

33.

Pulido, F. 2007. The genetics and evolution of avian migration. Bioscience 57:165-174.

Ramenofsky, M., and J.C. Wingfield. 2007. Regulation of migration. Bioscience 57:135-

143.

14

Rappole, J.H., and M.V. McDonald. 1994. Cause and effect in population declines of

migratory birds. Auk 111:652-660.

Rappole, J., and P. Jones. 2002. Evolution of old and new world migration systems.

Ardea 90:525-537.

Robbins, C.S., J.R. Sauer, R.S. Greenberg, and S. Droege. 1989. Population declines in

North American birds that migrate to the Neotropics. Proceedings of the National

Academy of Sciences of the United States of America 86:7658-7662.

Rodewald, P. 2007. Forest habitats in fragmented landscapes near Lake Erie: defining

quality stopover sites for migrating landbirds. A final research report submitted to

the Ohio Lake Erie Commission.

Rodewald, P.G., and M.C. Brittingham. 2004. Stopover habitats of landbirds during fall:

Use of edge-dominated and early-successional forests. Auk 121:1040-1055.

Rodewald, P.G., and M.C. Brittingham. 2007. Stopover habitat use by spring migrant

landbirds: The roles of habitat structure, leaf development, and food availability.

Auk 124:1063-1074.

Salewski, V., and B. Bruderer. 2007. The evolution of bird migration - a synthesis.

Naturwissenschaften 94:268-279.

Sandberg, R., and F.R. Moore. 1996. Fat stores and arrival on the breeding grounds:

reproductive consequences for passerine migrants. Oikos 77:577-581.

Sanderson, F.J., P.F. Donald, D.J. Pain, I.J. Burfield, and F.P.J. van Bommel. 2006.

Long-term population declines in Afro-Palearctic migrant birds. Biological

Conservation 131:93-105.

Sauer, J.R., J.E. Hines, J. Fallon, K.L. Pardieck, D.J.J. Ziolkowski and W.A. Link. 2011.

The North American breeding bird survey, results and analysis 1966 - 2010. Version

12.07.2011. URL: www.pwrc.usgs.gov. Accessed: 11/28/2012.

Sekercioglu, C.H. 2007. Conservation ecology: Area trumps mobility in fragment bird

extinctions. Current Biology 17:283-286.

Sergio, F., and I. Newton. 2003. Occupancy as a measure of territory quality. Journal of

Animal Ecology 72:857-865.

Sheehy, J., C.M. Taylor, and D.R. Norris. 2011. The importance of stopover habitat for

developing effective conservation strategies for migratory animals. Journal of


http://www.pwrc.usgs.gov/

15

Sherry, T.W., R.T. Holmes. 1995. Summer versus winter limitation of populations: what

are the issues and what is the evidence. Pages 85-120 In T. E. Martin and D. M.

Finch, editors. Ecology and management of Neotropical migratory birds. Oxford

University Press, New York.

Sherry, T.W., and R.T. Holmes. 1996. Winter habitat quality, population limitation, and

conservation of neotropical nearctic migrant birds. Ecology 77:36-48.

Sherry, T.W., R.T. Holmes. 1993. Are populations of neotropical migrant birds limited in

summer or winter - implications for management. Pages 47-57 In D. M. Finch and P.

W. Stangel, editors. Status and management of Neotropical migratory birds. General

Technical Report RM-229, U.S. Department of Agriculture Forest Service Rocky

Mountain Forest and Range Experiment Station, Fort Collins, Colorado.

Sillett, T.S., and R.T. Holmes. 2002. Variation in survivorship of a migratory songbird

throughout its annual cycle. Journal of Animal Ecology 71:296-308.

Smith, R.J., and F.R. Moore. 2005. Arrival timing and seasonal reproductive performance

in a long-distance migratory landbird. Behavioral Ecology and Sociobiology 57:231-

239.

Smith, R.J., and F.R. Moore. 2003. Arrival fat and reproductive performance in a long-

distance passerine migrant. Oecologia 134:325-331.

Steadman, D. 2005. Paleoecology and fossil history of migratory landbirds. Pages 5-17 In

R. Greenberg and P. P. Marra, editors. Birds of two worlds. The Johns Hopkins

University Press, Baltimore.

Taverner, P.A. 1904. A discussion of the origin of migration. Auk 21:322-333.

Tøttrup, A., R. Klaassen, M. Kristensen, R. Strandberg, Y. Vardanis, Å. Lindström, C.

Rahbek, T. Alerstam, and K. Thorup. 2012. Drought in Africa caused delayed arrival

of European songbirds. Science 338:1307-1307.

Weber, T.P., A.I. Houston, and B.J. Ens. 1999. Consequences of habitat loss at migratory

stopover sites: a theoretical investigation. Journal of Avian Biology 30:416-426.

Wiens, J.A. 1989. Spatial scaling in ecology. Functional Ecology 3:385-397.

Wikelski, M., E.M. Tarlow, A. Raim, R.H. Diehl, R.P. Larkin, and G.H. Visser. 2003.

Costs of migration in free-flying songbirds. Nature 423:704-704.

16

Chapter 2 - The influence of local- and broad-scale habitat

features on songbird abundance during migratory

stopover

Introduction

Songbirds use multiple stopover locations in order to rest and refuel for

subsequent flights while migrating between breeding and non-breeding areas (Alerstam

and Hedenström 1998, Moore et al. 1995). Migrants that locate quality stopover habitats

likely can regain energy more quickly and shorten stopover duration (Cherry 1982,

Matthews and Rodewald 2010, Moore et al. 1995). By shortening time spent in stopover,

which makes up the majority of the migratory period (Cochran and Wikelski 2005,

Hedenström and Alerstam 1997), migrants may be able to arrive at the breeding grounds

earlier (Moore et al. 1995, Newton 2006, Tøttrup et al. 2012), secure higher quality

territories (Aebischer et al. 1996, Lanyon and Thompson 1986, Smith and Moore 2005),

and experience higher reproductive success (Hochachka 1990, Lozano et al. 1996, Moore

et al. 2005, Møller 1994, Sandberg and Moore 1996a, Smith and Moore 2003, Smith and

Moore 2005). En route migrants should attempt to select stopover habitat that affords

them the greatest access to food, shelter, and safety in order to minimize time and energy

spent in migration and maximize fitness (Hedenström and Alerstam 1997, Moore et al.

1995, Newton 2006, Weber et al. 1999).

17

Habitat selection theory and field studies suggest that migrating birds initiate the

process of selecting a stopover site at a broad spatial scale, likely while descending from

nocturnal migration (Bowlin et al. 2005, Chernetsov 2006, Hutto 1985b, Mukhin et al.

2008, Schaub et al. 1999), and then refine their selection through a period of exploration

after making landfall (Aborn and Moore 1997, Chernetsov 2006, Hutto 1985b, Jenni and

Schaub 2003, Moore and Aborn 2000). The distribution and abundance of migrants in a

region can be influenced by broad-scale features of the landscape, such as proximity to

geographical obstacles to migration (e.g., an ocean, large lake, desert, or mountain range)

and habitat availability (Buler et al. 2007, Buler and Moore 2011, Ewert et al. 2011,

France et al. 2012, McCann et al. 1993, Moore et al. 1995). At a more local scale,

migrant abundance can be associated with habitat structure (Cashion 2011, Rodewald and

Matthews 2005, Rodewald and Brittingham 2007), tree species within a patch (Strode

2009, Wood et al. 2012), patch size (Cox 1988, Keller and Yahner 2007, Somershoe and

Chandler 2004), and the distribution of local food resources (Blake and Hoppes 1986,

Buler et al. 2007, Cohen et al. 2012, Ewert et al. 2011, Martin and Karr 1986). Since both

broad- and local-scale features of habitat shape patterns in the distribution and abundance

of migratory birds during stopover (Buler et al. 2007, Deppe and Rotenberry 2008,

Pennington et al. 2008) it is important to consider both scales in stopover research

(Moore et al. 2005). A better understanding of the habitat features that influence the

abundance and distribution of stopover migrant birds is needed to make informed

conservation decisions for these species (Ewert et al. 2006, Mehlman et al. 2005, Moore

et al. 2005, Petit 2000). This information is urgently needed as many migratory species in

North America have exhibited population declines over the last several decades (Sauer et

18

al. 2011), which may in part be attributable to the loss of stopover habitat along their

migratory routes (Newton 2006).

Stopover studies that involve the collection and analysis of survey data present

several challenges. Like all avian survey data, there is the issue of imperfect detection of

species and individuals that are available to be recorded (Burnham 1981, McCallum

2005), which masks the ecological patterns of scientific interest to researchers (Fiske

and Chandler 2011). Individuals or species may be missed during surveys due to

weather conditions, observer ability, species behavior, background noise, habitat

structure, and other factors (Alldredge et al. 2007b, Johnson 2008, Simons et al. 2009).

It is critical to account for imperfect detection when making inferences about the

abundance of a species to limit the risk of making misguided conclusions (Gu and

Swihart 2004, Kéry and Schaub 2012, Norvell et al. 2003, Royle and Dorazio 2008,

Thompson 2002). However, most models designed to account for imperfect detection

carry the assumption of population closure (e.g., the population does not change during

the study) (Alldredge et al. 2007a, Farnsworth et al. 2002, Nichols et al. 2000, Royle

2004), which makes them inappropriate for use on migration data when populations are

in a nearly continual state of flux. Dail and Madsen (2011) developed a generalized

hierarchical N-mixture model that allows for simultaneous estimates of parameters that

influence the abundance and detection of a species within an open-population (i.e., the

population changes during the study through immigration and emigration) making it

amenable to stopover data.

The Western Lake Erie Basin (WLEB) provides important stopover habitat to

many migratory birds (American Bird Conservancy 2010, Ewert et al. 2006). Previous

19

research in the region, has documented a decline in the relative abundance of en route

migrants with increasing distance from the Lake Erie shoreline (Rodewald 2007). No

study to date has examined migrant use of both lakeshore forests and forests located

farther inland (>5.5 kilometers from the lakeshore) while accounting for imperfect

detection of these species. I studied how the abundance and arrival rates of Blackpoll

Warbler (Setophaga striata), Black-throated Green Warbler (S. virens), and a guild of 24

transient wood warblers (Parulidae) varied in randomly selected forest patches in the

WLEB. I examined how both local- and broad-scale habitat features influenced migrant

abundance using the hierarchical N-mixture model developed by Dail and Madsen

(2011). Specifically, I looked at how migrant density was influenced by 1) patch

vegetation composition and structure, 2) patch size and isolation, 3) patch distance from

the Lake Erie shoreline, 4) patch distance from a river or stream, and 5) wetland cover

surrounding a patch. I predicted that the density of migrants would be greatest in patches

that 1) were more structurally complex (Rodewald and Brittingham 2007) and contained

more oak (Quercus spp.) and elm (Ulmus spp.) and fewer maple (Acer spp.) trees (Graber

and Graber 1983, Wood et al. 2012), 2) were smaller and more isolated (Moore et al.

1995), 3) were closer to the Lake Erie shoreline (Buler et al. 2007, Ewert et al. 2006,

Ewert et al. 2011), 4) were closer to a river or stream (Ewert et al. 2006, Skagen et al.

1998), and 5) had more wetland cover in the surrounding landscape (Ewert et al. 2011,

Smith et al. 2007).

20

Study Area and Methods

Study area and site selection

The study was conducted in the WLEB of Ohio from 19 April to 30 May in 2011

and 2012. The study area extended from Toledo to Sandusky, Ohio, and from 0-22 km

from the Lake Erie shoreline. This area includes all or portions of Erie, Huron, Lucas,

Ottawa, Seneca, Sandusky, and Wood counties (Figure 2.1). Land cover in the broader

WLEB is approximately 66% agriculture, 13% urban/residential, 11% grass/rural cover,

9% forest, and 1% water/wetlands (U.S. Army Corps of Engineers 2009). This region

was once covered by the Great Black Swamp, a large forested wetland that was largely

drained and cleared during the 19th

and early 20th

centuries (Kaatz 1955). Remaining

forests in the WLEB are largely composed of maple, elm, oak, ash (Fraxinus spp.),

hickory (Carya spp.), and eastern cottonwood (Populus deltoides). The majority of the

ash trees in the region are either dead or dying as result of the emerald ash borer (Agrilus

planipennis), an invasive non-native species that first arrived in this area of Ohio in 2004

(Herms et al. 2004, Herms et al. 2006).

I used the National Oceanic and Atmospheric Administration (NOAA) coastal

change analysis program (C-CAP) 2006 dataset to identify all forest patches larger than

0.36 ha within the study area (NOAA 2006). I used a quantile approach to stratify all

forest patches into four distance bands (0–0.57 km, 0.57–2.65 km, 2.65 km–10.09 km,

10.09–22 km) relative to the Lake Erie shoreline (shoreline defined by the U.S. Geologic

Survey (USGS) National Hydrology Dataset (NHD) (USGS 1999). I used these distance

bands to perform a generalized random tessellation stratified (GRTS) draw of forest

21

patches with replacement for each season (Stevens and Olsen 2004) using the ‘spsurvey’

package (Kincaid and Olsen 2012) in the statistical program R (R Development Core

Team 2012). I selected the nearest forested patch in the same distance band and within

0.5 km when a selected location was not forested due to a misclassification in the C-CAP

dataset. Different samples of survey locations were drawn for each year in order to

provide a more representative collection of forested sites within the broader study area. I

identified, contacted, and received permissions from landowners, agencies, and

organizations to conduct bird surveys at selected survey locations. In 2011, I evaluated

225 sites and received permission to conduct surveys at 86 locations (38.2% success

rate). In 2012, I evaluated 240 sites and received permission to conduct surveys at 91

sites (37.9% success rate).

Bird surveys

Migrant landbird abundance during stopover can vary from day to day, with many

days characterized by low numbers (DeSante 1983, Lowery Jr. 1945). I established fixed

point count locations 30 m from the forest edge, where spring migrants are more likely to

concentrate, to increase numbers of detections during point count surveys (Rodewald and

Brittingham 2007). Whenever possible, I located survey points near eastern edges where I

expected morning bird activity would be higher due to greater sunlight and insect activity

(Keller et al. 2009). I established point counts at 135 locations: 86 were used 2011, and

91 were used in 2012 (42 of which were the same locations used in 2011) (Figure 2.2).

For 22 of the 135 point count locations, I could not establish the point count along an

eastern edge due to inaccessibility, lack of land owner permissions, or nesting Bald

Eagles (Haliaeetus leucocephalus). For those sites, I located point counts along southern

22

edges, to maintain consistent amounts of sunlight near forest edges. Point count locations

were placed in the center of small woodlots to maintain a minimum distance of 20 m

from all forest edges.

Point count surveys were 10 minutes long, and split into four 2.5-minute sampling

periods. Observers recorded the number of individuals of each bird species detected, and

estimated the distance to each individual within 3 distance bands (0–15 m, 15–30 m, >30

m) during each sampling period. In addition, observers recorded the time when the

survey started and a categorical estimate for wind speed during the survey. Wind speed

categories were based on the Beaufort scale: (0) = < 1 km/hr, (1) = 1–5 km/hr, (2) = 6–11

km/hr, (3) = 12–19 km/hr, (4) = 20–30 km/hr, (5) = 31–39 km/hr (Cullen 2002). During

point counts in 2012, observers also recorded categorical estimates of midge abundance

within 5 m of point count locations following methods developed by Smith et al. (2007).

The categories used were: (0) = 0 midges, (1) = 1–10 midges, (2) = 11–100 midges, (3) =

101–500 midges, (4) = 501–1000 midges, (5) = 1001–5000 midges, (6) = >5001 midges

(Smith et al. 2007).

Point count locations were surveyed approximately twice weekly in both years

when weather permitted, between 0.75 and 5.75 hours after sunrise, from 19 April

through 30 May. Surveys were conducted at an average of 19.7 locations per day (range

= 7–27, SD = 4.6). Survey effort was divided among locations that were in close

proximity to the lakeshore, some farther inland (>6 km from the shore), and in both

eastern and western parts of the study area on each day of sampling. Observers rotated

among locations in both years so that only 22.6% of the time did the same observer

survey the same location on the subsequent visit. A total of 808 point counts were

23

conducted at 86 locations in 2011, an average of 9.4 surveys per location (range = 6–12,

SD = 1.4). A total of 810 point counts were conducted at 91 locations, an average of 8.9

surveys per location in 2012 (range = 6–12, SD = 0.93).

Study species

The Blackpoll Warbler and the Black-throated Green Warbler were selected as

focal species for these analyses. The Blackpoll Warbler (BLPW) is a long-distance

Neotropical-Nearctic migratory songbird, which breeds in boreal forests of the

northeastern United States and Canada and winters mainly in South America (eBird

2012, Peterjohn 1989). They do not breed in Ohio, but are common transients during

spring and fall migration (eBird 2012, Peterjohn 1989). The Black-throated Green

Warbler (BTNW) breeds in coniferous and mixed hardwood forests of the eastern United

States and Canada and winters predominantly in Central America and the Caribbean

(eBird 2012, Peterjohn and Rice 1991). The BTNW breeds within eastern Ohio, but there

are no breeding records within the study area, meaning all individuals surveyed were

highly likely to be transients (Ohio Breeding Bird Atlas II 2012, Peterjohn and Rice

1991).

To better understand how migrant landbirds respond to landscape features during

stopover, I created a guild comprised of 24 species of transient wood warblers. Migrants

in the guild are taxonomically similar and likely have similar foraging strategies during

stopover. Included in the guild were: Blue-winged Warbler (Vermivora cyanoptera),

Golden-winged Warbler (V. chrysoptera), Tennessee Warbler (Oreothlypis peregrina),

Orange-crowned Warbler (O. celata), Nashville Warbler (O. ruficapilla), Northern Parula

(Setophaga americana), Chestnut-sided Warbler (S. pensylvanica), Magnolia Warbler (S.

24

magnolia), Cape May Warbler (S. tigrina), Black-throated Blue Warbler (S.

caerulescens), Yellow-rumped Warbler (S. coronata), Black-throated Green Warbler,

Blackburnian Warbler (S. fusca), Yellow-throated Warbler (S. dominica), Pine Warbler

(S. pinus), Prairie Warbler (S. discolor), Palm Warbler (S. palmarum), Bay-breasted

Warbler (S. castenea), Blackpoll Warbler, Cerulean Warbler (S. cerulea), Hooded

Warbler (S. citrina), Black-and-white Warbler (Mniotilta varia), Wilson’s Warbler

(Cardellina pusilla), and Canada Warbler (C. canadensis).

Local-scale variables

I established a single 30-meter radius plot centered on each point count location to

quantify local habitat characteristics. Within each plot, I recorded the number of trees by

species in three diameter-at-breast-height (dbh) size classes: 8–23 cm dbh (small trees),

23–38 cm dbh (medium trees), >38 cm dbh (large trees). I recorded whether ash trees

were healthy, dead, or infected by emerald ash borer following a modified version of the

classification system developed by Smith (2006). I estimated understory stem density

(stemhits) between 0.5–3.0 m in height by counting the number of stems touching a

vertical pole every 2 m along 30 m transects that originated at the plot center and

extended in each cardinal direction. I measured canopy height in meters at a minimum of

two locations within the plot using a range finder in order to calculate average canopy

height.

I conducted a principal components analysis (PCA) on six habitat structure

variables collected at all 135 point count locations in the statistical computing program R

(R Development Core Team 2012). The variables used were stemhits, number of small

trees, number of medium trees, number of large trees, number of medium and large dead

25

ash trees, and average canopy height. I used a correlation matrix to construct the PCA

(Table 2.1) and followed the latent root criterion when selecting the number of axes to

interpret (McGarigal et al. 2000). The first three axes each had eigenvalues >1, but only

the first two axes were retained to minimize the number of parameters in the N-mixture

model (eigenvalue of third axis = 1.02). The first PCA axis accounted for 29.2% of the

total variation (eigenvalue = 1.75) and described a gradient of decreasing numbers of

large trees and canopy height, and increasing numbers of small and medium sized trees

and medium and large dead trees (Table 2.2). The second axis accounted for an additional

28.2% of the total variation (eigenvalue = 1.69) and described a gradient of decreasing

numbers of stemhits and increasing numbers of small, medium, and large trees (Table

2.2). I used the first two axes of the habitat structure PCA in subsequent N-mixture

models.

I performed another PCA using the tree species data. For each point count

location, I summed counts of medium and large trees for the eight most abundant species

groups found in the study area. Those species groups were oaks, hickories, eastern

cottonwood, elms, maples, boxelder (Acer negundo), walnuts (Juglans spp.) and dead

ashes. I performed a Hellinger transformation on the raw species counts for each species

group at each location using the package “vegan” (Oksanen et al. 2011) to make the data

more appropriate for a PCA analysis (Legendre and Gallagher 2001). I used a correlation

matrix to construct the PCA (Table 2.3) and followed the latent root criterion when

selecting the number of axes to interpret (McGarigal et al. 2000). The first three axes

each had eigenvalues >1, but only the first two axes were retained to minimize the

number of parameters in the N-mixture model (eigenvalue of third axis = 1.25). The first

26

axis accounted for 27.7% of the total variation (eigenvalue = 2.22) and described a

gradient of increasing numbers of oaks, hickories and walnuts and decreasing numbers of

maples and cottonwoods (Table 2.4). The second axis accounted for an additional 17.8%

of the total variation (eigenvalue = 1.42) and described a gradient of increasing numbers

of dead ashes, hickories, and cottonwoods, and decreasing numbers of elms (Table 2.4). I

used the first two axes of the species composition PCA in subsequent N-mixture models.

Broad-scale variables

I used data from the U.S. Fish and Wildlife Service’s (USFWS) 2012 National

Wetland Inventory (NWI) (USFWS 2012) and NOAA’s 2006 C-CAP (NOAA 2006) to

quantify forest cover within the study area for analyses. The 2012 NWI data were used to

identify forested areas near the coast, and the 2006 coastal change analysis program (C-

CAP) dataset was used to identify inland forests not covered by the NWI. The 2012 NWI

data more accurately classified coastal forest patches, but was not designed to classify

inland forests.

Broad-scale variables were calculated in ESRI’s ArcGIS Desktop 10 (ESRI

2011). I measured area, distance from the coast, distance from a river or stream, isolation,

and wetland cover in the landscape for each point count location in the study area (Table

2.5). I used the National Hydrology Dataset (USGS 1999) to define the Lake Erie

coastline and rivers and streams. I calculated the proportion of the landscape within 500

m of point count locations that was forested, as a proxy for patch isolation (Bender et al.

2003, Tischendorf et al. 2003). A 500 m scale was selected for the isolation variable

based on a post-hoc analysis using the computer program Focus (Holland et al. 2004). I

used Focus to assess the scale at which migrants responded to the amount of forest cover

27

within the landscape during stopover (Holland et al. 2004). I first calculated the

proportion of the landscape surrounding point count locations that was forested at 20

different spatial scales using radii of 0.5 to 10 km, in 0.5 km increments (Buler et al.

2007). I then used Focus to create 300 samples of 14 randomly selected and spatially

independent landscapes for each of the 20 spatial scales. Focus produced regressions

between the proportion of the landscape that was forested and the mean abundance of

migrants for the sites selected in each of the 300 samples (Buler et al. 2007, Holland et al.

2004). Focus provided mean r values (Pearson’s correlation coefficient) for the 300

correlations at each spatial scale, and for many migrants the scale with the greatest

absolute r value was 0.5 km (Holland et al. 2004).

Wetlands in the study area support high concentrations of emergent aquatic

midges (Diptera: Chironomidae) (MacDade et al. 2011), which are considered an

important food resource to migrating birds in the Great Lakes region, particularly in

spring (Ewert et al. 2011, Smith et al. 2007). Arthropod predators of midges, especially

spiders, are also consumed by migrating birds (Carlisle et al. 2012, Smith et al. 2007)

and their abundance can be positively correlated with midge abundance (Dreyer et al.

2012, Hoekman et al. 2011, Smith et al. 2007). Consequently, it is likely that areas near

wetlands offer greater food resources to migrating birds than areas farther from wetlands

(Smith et al. 2007). Wetlands were defined using the 2012 NWI data (USFWS 2012) for

analyses in this study. I removed all wetlands that had special modifiers (drained/ditched,

spoil, farmed, excavated, or artificial substrates), water regimes (temporarily flooded, or

saturated) or subclasses (unconsolidated bottom, or unconsolidated shore) (USFWS

2012) as these wetland types are unlikely to support a high abundance of emergent

28

aquatic midges (Robert Gates, Personal communication). I calculated the proportion of

the landscape within 200 m of each point count locations that was classified as wetland to

use as a covariate in N-mixture models. A 200 m buffer was selected for this variable as

midges tend to concentrate within this distance from the wetlands from where they

emerge (Dreyer et al. 2012).

Statistical methods

For these analyses, I excluded data from surveys conducted prior to a species’

first detection and after a species’ last detection during each year, as those zero counts

predominantly represented “false zeroes” as the species was very likely not present to be

detected (Martin et al. 2005, Wenger and Freeman 2008). Several extremely high counts

(>3 SD above mean) that likely were overestimates of true abundance were reduced to 2

SD above the mean. This included 5 extreme counts for Yellow-rumped Warbler and 3

for Palm Warbler, all occurred in 2011.

I used the generalized hierarchical N-mixture model created by Dail and Madsen

(2011) to analyze the count data for the Blackpoll Warbler, Black-throated Green

Warbler, and the transient warbler guild. This is a mechanistic open population model

that allows users to estimate changes in abundance over the duration of migration using

parameters for immigration (gamma - and emigration (omega - Ѡ), while accounting

for imperfect detection of study species (Dail and Madsen 2011). The model requires data

to be collected from i independent sites, during j secondary periods within t primary

periods (Dail and Madsen 2011, Royle and Chandler 2012). Primary periods were the

individual point count surveys conducted at a location and secondary periods were the

four 2.5-minute sub-sampling periods during each point count. Bird populations were

29

assumed to be closed during the 10-minute survey period at a site (a primary period), but

not between surveys (Dail and Madsen 2011). The mean abundance of migrants at a

location during primary period (t = 1) is governed in the model by the parameter lambda

(λ), and changes to that abundance between subsequent visits were modeled using the

parameters gamma (; i.e., number of new arrivals) and omega (Ѡ; i.e., apparent

survivorship between surveys) (Dail and Madsen 2011). The probability of detecting the

true number of individuals of a species or guild (Nit) present during a survey is governed

by the detection probability parameter p (Dail and Madsen 2011). Using these four

parameters (λ, , Ѡ, p), the model can generate abundance estimates for each survey

occasion, while accounting for imperfect detection of the study species (Dail and Madsen

2011). I used a Poisson distribution to model the counts of individuals detected on

surveys in both single-species models for both years. Since I was modeling changes in

abundance between migration surveys, it was necessary to analyze data from each year

separately.

I modeled count data for the transient warbler guild using an exponential model

described by Fiske and Chandler (2011), a modified version of the Dail and Madsen

model (2011). This model integrates immigration and emigration parameters into a single

parameter gamma (that represents population growth/loss (Fiske and Chandler 2011).

In this model, the number of individuals at site i on occasion t is equal to the number in

the previous period multiplied by or the finite rate of increase (Fiske and Chandler

2011). My decision to use the exponential model over the Dail and Madsen (2011) model

for the guild data was validated by improved Akaike Information Criterion (AIC) scores.

30

For the guild models, I used a Poisson distribution in 2011, and a negative binomial

distribution in 2012, as this distribution was favored by AIC.

Each of the parameters in the Dail and Madsen model (λ, , Ѡ, p) and the

exponential model (λ, , p) can be modeled as functions of covariates. Only variables for

which there was an a priori reason to believe influenced a given parameter were

considered. I used an information-theoretic approach to select appropriate variables for

each parameter (Burnham and Anderson 2002), and all continuous variables were scaled

and centered before being fit to any parameter by subtracting the mean of a variable from

all of its values and then dividing that number by the SD of the variable. I first modeled

candidate sets for detection probability, and identified the model with the greatest AIC

support. I then used the top detection model when testing the candidate set for gamma.

The most parsimonious model from the gamma set was considered the top overall model.

To plot relationships between variables and abundance, I created a model set of all

plausible models (< 2 AIC from the most parsimonious model) to use in generating

model-averaged predictions of relevant variables and parameters (Burnham and

Anderson 2002).

I modeled the initial counts of migrants at study sites (λ – lambda) as a constant

because there were very few detections of migrants early in migration, so modeling these

sparse data were not appropriate. I modeled immigration rates and population growth

(gamma - , using local- and broad-scale variables known to influence migrant landbird

abundance during stopover. A set of 60 candidate models were developed using the

following variables: patch area (Cox 1988, Martin 1980, Somershoe and Chandler 2004),

isolation (Buler et al. 2007, Martin 1980), distance from the coast (Buler et al. 2007,

31

Buler and Moore 2011, Ewert et al. 2011, Rodewald 2007), patch distance from rivers

and streams (Ewert et al. 2006, Skagen et al. 1998, Skagen et al. 2005), wetland cover

(Ewert et al. 2006, Ewert et al. 2011, Smith et al. 2004, Smith et al. 2007), tree species

composition (Graber and Graber 1983, Strode 2009, Wood et al. 2012), and habitat

structure (Rodewald and Brittingham 2007). Each of the 60 models included Julian date

and a quadratic term for Julian date because the abundance of the study species and the

guild was unimodal as counts generally peaked during mid-migration. Two variables

(area and isolation) were strongly correlated (|Pearson’s r| > 0.7), and so were not

included in the same models. Interactions were not considered in the analysis, as

understanding main effects in this novel framework was the primary focus of the study.

I modeled apparent survivorship of individuals between visits to a site (Ѡ –

omega) as a constant because I believed measured variables were not appropriate in

predicting the apparent survivorship of migrants. For the detection probability parameter

(p), I considered time of day, a quadratic effect for time of day, a categorical estimate of

wind speed, and the survey observer. Previous research has shown that the ability to

detect an individual bird that is present in an area during a survey can be influenced by:

observer skill and experience (Alldredge et al. 2007b, Diefenbach et al. 2003, Faanes and

Bystrak 1981, Farmer et al. 2012), observer age and hearing ability (Cyr 1981, Emlen and

Dejong 1992, Ramsey and Scott 1981, Saunders 1934), observer attentiveness (Farmer et

al. 2012, McCallum 2005, Robbins and Stallcup 1981), and observer expectations (Balph

and Balph 1983). Detection probabilities can also be influenced by background noise

(Alldredge et al. 2007b, Simons et al. 2007), time of day (Bart and Herrick 1984),

number of individuals present to be detected (Bart and Schoultz 1984, Kéry 2002,

32

Simons et al. 2007) and weather conditions (e.g., precipitation and wind) (Dawson 1981,

Emlen and Dejong 1981, Simons et al. 2009, Simons et al. 2007). Detectability also

depends on the behavior of the species itself, and how frequently species make

themselves available for detection (Alldredge et al. 2007b, Johnson 2008, McCallum

2005, Simons et al. 2009). I created a set of 12 candidate models using the variables of

interest for each of three distributions (Poisson, Negative Binomial, Zero-inflated

Poisson) and then tested each set for the detection parameter for each species and guild.

I assessed model fit using a parametric bootstrap procedure (Dixon 2002). The

procedure generated 250 replicate data sets for the single-species models with the greatest

support and 125 for the guild models, estimated parameters for those data, and computed

a fit statistic (Freeman-Tukey for the Dail and Madsen models and a Chi-square for the

exponential models) for each dataset (Fiske and Chandler 2011, Kéry et al. 2005). I

compared the fit statistic from the model with greatest support (observed) to the

distribution of the fit statistic created from the parametric bootstrap procedure (expected)

and then reported the proportion of the expected distribution greater than the observed

statistic (P-value) to assess the overall model fit (Fiske and Chandler 2011, Kéry and

Schaub 2012, Kéry et al. 2005). All model fitting, parameter estimations, and goodness of

fit testing were carried out in the program R version 2.15.1 (R Development Core Team

2012) using the package ‘unmarked’ version 0.9-9 (Fiske and Chandler 2011).

Maps

I created maps to show how the density of migrants varied in forest patches across

the landscape and to identify the forest patches that were most used by transient migrants

during spring migration. I made a raster of all forested locations using the NWI (USFWS

33

2012) and C-CAP (NOAA 2006) data. For each 1 ha cell of the raster, I assigned a value

for all variables within each top model that I was able to map (distance from the coast,

distance from a river or stream, patch isolation, wetland cover, patch area, and date). I

scaled and mapped all variables for each day of migration using the “raster” package

(Hijmans and van Etten 2012) in R (R Development Core Team 2012). I then used the

top model for each species or guild in each year to make area-weighted predictions of

bird abundance for each forest cell for each day of migration (Fiske and Chandler 2011).

I assumed that all individuals within 75 m of point count locations were exposed to

sampling in order to calculate densities. Given that surveyed areas were equal to 1.77 ha,

I divided predicted estimates by 1.77 before assigning estimated abundances to raster

cells that had areas of 1 ha. I also assumed that bird density was constant across the

patch. I extrapolated area-weighted predictions to the total area of the forest patch to

estimate numbers of individuals per patch.

Results

Blackpoll Warbler

In 2011, I detected a total of 95 BLPW during 514 surveys at 86 sites between 5

May and 29 May (0.18 BLPW/survey). There were four equally plausible models for the

detection parameter. The detection model with the greatest support included terms for

observer, time of day, and a quadratic of time of day. Detection probability was

positively correlated with time of day for all observers (Figure 2.3). There were three

equally plausible models for gamma (arrival rate of BLPW) (Table 2.6). The best-

34

supported model included terms for Julian date, a quadratic of Julian date, isolation, and

wetland cover, and showed moderate fit (P = 0.29). Arrival rates of BLPW in 2011 were

higher at sites with more wetland cover in the landscape, and at sites with less

surrounding forest cover (Figure 2.4). Arrival rates for BLPW peaked around 15 May in

2011 (Figure 2.5).

In 2012, I detected a total of 73 BLPW during 409 surveys at 91 sites between 9

May and 28 May (0.18 BLPW/survey). There were three equally plausible models for the

detection parameter, with the best-supported model including a term for wind. There was

a general pattern of decreasing detection probability with increasing wind speed (Figure

2.6). There were seven equally plausible models for gamma (arrival rate of BLPW)

(Table 2.7). The model with the greatest support included terms for Julian date, a

quadratic term for Julian date, distance from the coast, patch area, and wetland cover in

the surrounding landscape, and showed moderate fit (P = 0.38). Arrival rates in 2012

were higher at sites that had greater wetland cover, that were closer to the lakeshore

(Figure 2.7), and that were smaller in area, although this effect was marginal (Figure 2.8;

Table 2.8). Arrival rates for BLPW peaked around 15 May in 2012.

Arrival rates for BLPW in 2011 and 2012 were influenced largely by date,

wetland cover, patch isolation, and distance to the coast (Figure 2.8; Table 2.8). Forest

patches adjacent the Lake Erie shoreline and wetlands supported average densities that

were 9.9 to 144 times greater than forests that were 22 km inland and isolated from

wetlands (Figure 2.9). The forest patches that supported the highest average numbers of

BLPW during migration were within 9 km of the Lake Erie shoreline and were larger

than 15 ha (Figure 2.10). In many cases, smaller patches near the lakeshore supported as

35

many BLPW as large inland patches. For example, the average number of BLPW in a 8

ha patch adjacent to the lakeshore was approximately equal to the number supported in an

87 ha patch located 22 km inland.

Black-throated Green Warbler

In 2011, I detected a total of 202 BTNW during 727 surveys at 86 sites between

21 April and 29 May (0.28 BTNW/survey). There were four equally plausible models for

the detection parameter. The detection model with the greatest support included terms for

observer, time of day, and a quadratic term for time of day. Detection probability for

BTNW peaked mid-morning for all observers (Figure 2.11). There were seven equally

plausible models for gamma (arrival rate of BTNW) (Table 2.9). The model with the

greatest support included Julian date, a quadratic term for Julian date, distance from a

river or stream, species composition PC1, and species composition PC2 and showed

marginal fit (P = 0.12). Arrival rates of BTNWs in 2011 were most closely associated

with Julian date, with other variables showing little association (Figure 2.8; Table 2.8).

BTNW arrivals peaked around 6 May in 2011 (Figure 2.12).

In 2012, I detected a total of 202 BTNW during 753 surveys at 91 sites between

22 April and 28 May (0.27 BTNW/survey). There were two equally plausible models for

the detection parameter. The detection model with the greatest support included a term

for the survey observer, although parameter estimates overlapped zero (Figure 2.13).

There were six plausible models for gamma (Table 2.10). The model with the greatest

support included Julian date, a quadratic term for Julian date, distance from the lakeshore,

habitat structure PC1, habitat structure PC2, and patch isolation, and showed marginal fit

(P = 0.06). Arrival rates were primarily related to distance from the lakeshore (Figure

36

2.14) and date, as habitat structure and patch isolation were weakly associated with

arrival rates (Figure 2.8; Table 2.8). The peak in arrival rate was around 4 May in 2012

(Figure 2.15).

Arrival rates for BTNW in 2011 and 2012 were most strongly related to distance

from the lakeshore and date (Figure 2.8; Table 2.8). Predictive maps show that forest

patches within ~2 km of the Lake Erie shoreline supported densities of BTNW that were

between 1.5 and 2 times greater than forested areas 22 km inland (Figure 2.16). The

patches that supported the highest average numbers of BTNW per patch were larger than

58 ha and within 9 km of the lakeshore (Figure 2.17).

Transient warbler guild

In 2011, I detected a total of 3802 individuals in the transient warbler guild during

808 surveys at 86 sites between 19 April and 30 May (4.7 individuals/survey). The most

abundant transient migrant observed was the Yellow-rumped Warbler (50% of total),

followed by Palm Warbler (8%), Black-throated Green Warbler (7%), and Nashville

Warbler (6%) (Figure 2.18). The only probable model for the detection probability

parameter included terms for observer, time of day, a quadratic of time of day, and wind.

For all observers, detection probability generally declined with increasing wind speed

(Figure 2.19) and increased with time of day (Figure 2.20). There were five equally

plausible models for the finite rate of increase (Table 2.11). The model with the greatest

support included terms for Julian date, a quadratic of Julian date, and distance from the

coast and exhibited moderate fit (P = 0.16). The finite rate of increase was negatively

related to distance from the coast, and the 95% confidence interval for the estimate of the

coefficient overlapped zero (estimate for coefficient = -0.014, 95% confidence interval:

37

-0.030 to 0.002). The average density of the transient warbler guild during migration was

2.2 times greater in forest patches at the lakeshore compared to sites 22 km inland, and

declined by about 3.2% per km from the lakeshore.

In 2012, I detected a total of 1413 individuals in the transient warbler guild during

810 surveys at 91 sites between 19 April and 28 May (1.74 individuals/survey). The most

abundant transient migrant was the Yellow-rumped Warbler (36% of total), followed by

Black-throated Green Warbler (13%), Magnolia Warbler (6%), Nashville Warbler (6%),

and Tennessee Warbler (6%) (Figure 2.18). The one plausible model for the detection

probability parameter included terms for observer, time of day, a quadratic of time of

day, and wind. For all observers, detection probability generally declined with increasing

wind speeds (Figure 2.21), and increased with time of day (Figure 2.22). There were four

equally plausible models for the finite rate of increase (Table 2.12). The model with the

greatest support included terms for Julian date, a quadratic of Julian date, and distance

from the coast and exhibited moderate fit (P = 0.18). The finite rate of increase was

negatively related to distance from the coast, and the 95% confidence interval for the

estimate of the coefficient overlapped zero (estimate for coefficient = -0.019, 95%

confidence interval: -0.022 to 0.012). Mean density of the transient warbler guild was

about 2.4 times greater at sites within a km of the lakeshore than at sites 22 km inland,

declining about 3.9% with each km from the lakeshore.

Abundance of transient warblers in the WLEB in 2011 and 2012 was most

strongly related to the distance from the lakeshore. Mean density of transient warblers

during 2011 and 2012 was 2.2 times higher within a km of the lakeshore compared with

forests 22 km inland, and declined about 3.4% per km from the lakeshore (Figure 2.23).

38

The patches that supported the highest average number of guild members during

migration were larger than 58 ha and within 9 km of the lakeshore (Figure 2.24).

Midges

Categorical estimates of midge abundance in 2012 increased with Julian date

(Figure 2.25), wetland cover (Figure 2.26), and proximity to the coast (Figure 2.27). I

used empirical Bayes methods to estimate the number of transient warblers present at

each point count location during each survey period using the top model for the transient

warbler guild in 2012 (Fiske and Chandler 2011). I conducted a post-hoc ANOVA where

abundance of transient warblers was predicted using midge scores, Julian date, and a

quadratic of Julian date. There was a difference in the mean abundance of transient

warblers among midge classes (F6, 801 = 1.835, P = 0.0895). I found that surveys that had

midge scores of 3, 2, or 1 had significantly more transient warblers than surveys that had

midge scores of 0, and that surveys that had midge scores of 3 had significantly more

transient warblers than surveys that had midge scores of 1 (Figure 2.28).

Discussion

Broad-scale variables

Proximity to the coast was one of the strongest predictors of arrival rates for

Blackpoll Warblers in 2012 and Black-throated Green Warblers in 2012; it was also the

lone habitat variable in the top model for the finite rate of increase for the transient

warbler guild in both years, although confidence intervals for coefficient of the

realtionship overlapped zero in both years. The generally higher arrival rates and finite

39

rates of increase near the shoreline led to higher predicted densities of migrants in

lakeshore forest patches compared with more inland forests. These results are consistent

with other studies in the Great Lakes region (Bonter et al. 2009, Ewert et al. 2011, France

et al. 2012), along the Atlantic coast (McCann et al. 1993), and along the Gulf of Mexico

(Buler et al. 2007, Buler and Moore 2011), where migrant densities were consistently

found to be greatest in close proximity to large bodies of water. Although the decline in

migrant density with increasing distance from the Lake Erie coast was expected (Ewert et

al. 2006), no study had quantified this decline at distances more than 5.5 km from the

coast using on the ground surveys. The nature of the decline varied by species. For

example, BLPW abundance was more concentrated along the coast than BTNW or the

transient warbler guild. Migrants that are closer to their breeding grounds may respond

differently to features of the landscape than migrants that breed farther away (Keller and

Yahner 2007). Given that Yellow-rumped Warblers (the most abundant species in the

transient warbler guild) and BTNWs are closer to their breeding grounds in northwestern

Ohio than BLPW, they may be more apt to explore the local landscape in search of food

or potential breeding habitat than would BLPW. This would help explain differences in

the way the species responded to Lake Erie.

Why migrants concentrated near the shoreline is not entirely clear, but several

factors likely are involved. Migrants may encounter the lake and then decide to land in

lakeshore habitats in order to rest and refuel before attempting to cross (Ewert et al. 2006,

Ewert et al. 2011, Fortin et al. 1999). Concentrations of migrants may also build in

lakeshore habitats if migrants are circumventing the lake in order to minimize risk and

energy associated with directly crossing the water where there are few suitable locations

40

to land (Alerstam 2001, Sandberg and Moore 1996b). A local radar study of nocturnal

migration found that spring migrant landbirds largely fly northwest (parallel to the coast)

when they encounter the Lake Erie shoreline, suggesting that migrants are hesitant to

cross the Lake at night (Jeremy Ross, Personal communication). If spring migrants east

of the study area also head west when they reach Lake Erie, this would result in many

birds being directed into my study area, and would explain why this region hosts such

enormous numbers of migratory birds during spring migration (Ewert et al. 2006). More

research is needed to understand movements of migratory birds in relation to Lake Erie at

various locations along its shoreline and farther inland (Ewert et al. 2006).

Migrants may also be concentrating in lakeshore forest patches because these

areas offer greater food resources in the form of emergent aquatic midges and possibly

arthropods (Ewert et al. 2006, Ewert et al. 2011, Smith et al. 2007). Food is a critical

resource to migrating birds, and several studies have shown migrant abundance to be

positively correlated with food abundance during stopover across habitats (Cohen et al.

2012, Hutto 1985a, Smith and Hatch 2008) and within a habitat (Buler et al. 2007, Ewert

et al. 2011, Martin and Karr 1986). Migrant densities may increase in lakeshore habitats

over a period of several days if individual migrants stopover longer in these areas due to

high food availability and an increased opportunity to refuel rapidly (Martin and Karr

1986, Moore et al. 2005). Several pieces of evidence suggest that migrant abundance was

positively related to midge abundance in this study. Midge abundances along with

transient warbler abundance and 2012 BLPW and BTNW arrival rates were higher at

sites near the Lake Erie shoreline. In addition, midges were more abundant at sites with

more wetland cover in the landscape and so were BLPW arrival rates in both 2011 and

41

2012. Furthermore, a post-hoc analysis revealed that surveys during 2012 that had higher

estimates of midges also had significantly more transient warblers. However, because

distance to the lakeshore and wetland cover variables were correlated (|r| = 0.52), it was

not possible to tease apart whether migrants concentrated near the lake in response to the

ecological obstacle, the increased food abundance, or both. Yet, it is clear from this study

that lakeshore forest sites received high use by migrating birds in the WLEB and

therefore have high conservation value (Mehlman et al. 2005).

The amount of forest cover in the landscape was an important factor influencing

the abundance and arrival rates of BLPW in 2011. In that year, BLPWs were more

abundant in patches that were more isolated. This finding is in contrast to Buler et al.

(2007) who found that migrant abundance in the Gulf of Mexico was positively related to

increasing forest cover in the landscape at the 5 km scale. However, the landscape where

Buler et al. (2007) worked was 70% forested, whereas my study area was only 5.3%

forested. The lack of available habitat may have forced migrants to concentrate into

remnant patches at high densities (Moore et al. 1995). These densities may cause

migrants to experience greater competition and lower food availability (Kelly et al. 2002,

Moore et al. 1995), or be more exposed to disease or predation (Klaassen et al. 2012),

which in turn could influence their future survival and reproductive success (Baker et al.

2004, Moore et al. 1995). More research is needed to discern the influences of stopover

habitat quality and availability on the future reproductive success and survivorship of

migratory birds.

Patch proximity to a river or stream did not seem to strongly influence transient

migrant arrival rates or abundances. The variable was found in only one top model

42

(BTNW in 2012), but the 95% confidence interval for the coefficient overlapped zero.

However, inland point count locations (>10 km from the lakeshore) along rivers and

streams often had more migrants than other inland locations that were not situated along

rivers and streams. In contrast, migrants heavily used all lakeshore point count locations

(< 2 km from the lake), regardless of their proximity to a river or stream. This suggests

that distance to a river or stream may be a better predictor of migrant abundance and

arrival rates at inland sites (>10 km from the lakeshore) than at lakeshore sites. Future

research that examines abundance of migrants in relation to rivers and streams should

consider interaction-effects with larger-scale landscape features, especially obstacles to

migration such as large lakes, oceans, and mountains, as patterns in habitat-use by

migrants may shift with proximity to large-scale geographical features.

Surprisingly, patch area had little influence on the finite rate of increase in the

guild models, or arrival rate in either of the single-species models. In 2012, the only year

in which patch area was in a top model, BLPW arrival rates were generally higher in

smaller patches, but the effect was marginal as the confidence interval for the coefficient

of the relationship overlapped zero. The influence of patch area on both migrant arrival

rates and the finite rate of increase may be more complex than initially expected. For

example, a large inland forest patch likely has lower density of migrants than a smaller

patch located at the same distance from lakeshore (Martin 1980). However, near the

lakeshore (< 2 km) all patches, both large and small, may have equally high densities of

migrants because there is limited remaining habitat in this area. Thus, the effect of patch

area on migrant abundance may vary in relation to distance from the lakeshore, such that

patch area has a weaker influence on migrant density closer to the lakeshore.

43

Local-scale variables

Counter to my predictions, local-scale vegetation variables, patch composition,

and structure had little influence on transient migrant abundance or arrival rates. Previous

research has demonstrated that en route migrants selectively forage in oak and elm trees

and avoid maples during migration (Graber and Graber 1983, Strode 2009, Wood et al.

2012), likely because oaks (Summerville et al. 2003) and elms support greater

abundances of food resources (Graber and Graber 1983, Wood et al. 2012). Therefore, I

expected migrants would exhibit higher densities at sites that contained more oak and

elms and exhibit lower densities at sites that contained more maples. In 2012, the only

year in which patch composition was in a top model, BTNW arrival rates were negatively

related to PC1, suggesting greater arrival rates at sites with more maples and cottonwoods

and lower arrival rates at sites with more oak and hickory, but the relationship was not

strong. I also found that BTNW arrival rates were negatively related to PC2, suggesting

greater arrival rates at sites with more elms and lower arrival rates at sites with more dead

ash and hickory. Although the confidence interval on the coefficient did not overlap zero,

the influence of PC2 on arrival rates was marginal. The reason I found no relationship

between migrant abundance and species composition may have been related to the nature

of my data. I did not record which tree species migrants were foraging in during point

counts and so my data could only identify correlations between the abundance of tree

species and the abundance of migrants, but likely not preferences by migrants for

particular tree species.

Habitat structure also seemed to have little influence on migrant abundance and

arrival rates. Some studies have found that en route migrants seem to prefer sites with

44

complex and diverse vegetative structure (Cashion 2011, Rodewald and Brittingham

2007), but others have found structure to be weakly related to en route migrant

abundance during spring migration (Buler et al. 2007, Martin 1980, Packett and Dunning

2009, Somershoe and Chandler 2004). In the one model in which patch structure was a

top variable, BTNW arrival rates slightly declined with increasing values of PC1,

suggesting lower arrival rates at less mature sites (sites with more small and medium

trees and fewer large trees). However, the coefficient for the estimate of the relationship

overlapped zero. BTNW arrival rates showed a positive relationship with PC2,

suggesting higher arrival rates at sites with more trees (of all sizes) and denser

understories. While the 95% CI for the coefficient of that relationship did not overlap

zero, habitat structure (PC2) was not as strongly related to BTNW arrival rates as was

patch distance to the coast. Spring migrants in the WLEB have been shown to respond to

vegetation structure (Rodewald 2007), but that the pattern was not detected in this study.

There may have been too little variation in habitat structure across study sites to detect a

pattern between migrant abundance and structural variables. Furthermore, I focused

modeling efforts on arrival rates and abundance, which may be more strongly related to

broad-scale landscape features than local features (Buler et al. 2007). Fine-scale habitat

selection by migrants may be better addressed by focusing modeling efforts on omega or

apparent survivorship or alternatively by using radio-telemetry methods. Future efforts

could explore how apparent survivorship and stopover duration relates to various local-

and broad-scale habitat features.

45

Detection probability

Observer, time of day, and wind strongly influenced detection rates for transient

migrants. Other studies have shown that detection probability can vary among survey

observers (Diefenbach et al. 2003, Sauer et al. 1994) and decline with increasing wind

conditions (Simons et al. 2007). I found that detection probabilities during migration

generally increased with time of day, which is in contrast to numerous other studies that

have documented declines in detection probability over the course of morning surveys

(Farnsworth et al. 2002, Marques et al. 2007). A decline in detection probability as the

morning progresses is expected during the breeding season, as avian singing rates

generally decline from sunrise to noon (Bart and Herrick 1984). Since the majority of

detections during avian surveys are aural (Brewster and Simons 2009, Dawson and

Efford 2009), it is not surprising that this would lead to lower detection probabilities for

these species. The lower detection probability that was observed in my study during early

morning surveys may have been due to observers being exposed to high singing rates of

territorial breeding and transient birds at those times. Observers exposed to a high volume

of acoustic cues may have lower detection probabilities than observers exposed to fewer

signals (Bart and Schoultz 1984, Simons et al. 2007). For example, Simons et al. (2007)

found that observers detected 41 ± 5.2% fewer birds on simulated point counts when

there were 3 additional species of birds singing in the background. I also found

dramatically different detection probabilities among observers in 2011, which likely

resulted from a lower ability of one observer to hear high frequency vocalizations of

some species (Emlen and Dejong 1992).

46

Using guilds can be an effective way to make inferences about a suite of species

and to utilize data collected on rarer species for which counts were too low to use in more

detailed analyses (Mattsson and Marshall 2009). However, grouping similar species

together can also mask species-level variability to environmental gradients, especially for

less common species in the guild. I created and used a transient warbler guild in order to

understand more general patterns in the stopover habitat-use of transient migrants in the

WLEB. I recognize that species in this guild may have had different detection

probabilities (Diefenbach et al. 2003, Mattsson and Marshall 2009, Moore et al. 2004,

Reidy et al. 2011), which could lead to misleading results. However, I believe differences

in detection probability among species were marginal because species in the transient

warbler guild were taxonomically similar and shared similar foraging strategies during

migration. Moreover, others have had some success using guilds in an occupancy

framework while accounting for imperfect detection probability (Mattsson and Marshall

2009). While detection probability was lower for the guild than for single-species models,

the presence of larger numbers of guild members on surveys likely made it difficult for

observers to track all individuals, resulting in the observed lower detection rates (Bart and

Schoultz 1984, Simons et al. 2007).

Conclusions and management recommendations

The abundance and distribution of migrants in the WLEB during spring migration

was primarily shaped by broad-scale variables including wetland cover, patch isolation,

and especially proximity to the coast. There was some year-to-year variability in the raw

count totals for some species and in the variables that most strongly influenced the arrival

rates and abundance of migrants, which may have resulted, in part, from annual variation

47

in weather conditions. In general, lakeshore forests hosted the highest densities of

migrants, but this does not necessarily indicate that these areas are of higher quality

habitat than more inland forests (Johnson 2007, Van Horne 1983). However, lakeshore

forest sites consistently received high use for stopover in both 2011 and 2012 and thus

are of conservation value (Mehlman et al. 2005). Larger patches of forest (>20 ha)

located between 0.5 to 10 km from the lakeshore are also of high conservation value,

particularly those nearer the coast and wetlands, as these patches consistently supported

the greatest seasonal abundance of transient migrants in both years. Habitat conservation

away from the lakeshore could seek to provide a network of forest patches and potentially

travel corridors (e.g., along streams and rivers) that should benefit stopover migrants

(Ewert et al. 2006, Mehlman et al. 2005). With limited habitat remaining in the WLEB,

the conservation of any forested habitats will likely benefit migratory forest birds. Future

conservation efforts should prioritize the protection and restoration of (1) forested areas

adjacent the coastline and (2) forested areas within 0.5–10 km of the coast, with priority

given to forests that are larger than 20 ha and closer to wetlands and the lakeshore.

48

Literature Cited

Aborn, D.A., and F.R. Moore. 1997. Pattern of movement by Summer Tanagers (Piranga

rubra) during migratory stopover: A telemetry study. Behaviour 134:1077-1100.




Alerstam, T. 2001. Detours in bird migration. Journal of Theoretical Biology 209:319-

331.



Alldredge, M.W., K.H. Pollock, T.R. Simons, J.A. Collazo, S.A. Shriner, and D. Johnson.

2007a. Time-of-detection method for estimating abundance from point-count

surveys. Auk 124:653-664.

Alldredge, M.W., T.R. Simons, and K.H. Pollock. 2007b. Factors affecting aural

detections of songbirds. Ecological Applications 17:948-955.

American Bird Conservancy. 2010. Globally important bird areas in Ohio. URL:

http://www.abcbirds.org/abcprograms/domestic/iba/ohio.html. Accessed:

11/28/2012.





271:875-882.

Balph, D.F., and M.H. Balph. 1983. On the psychology of watching birds: the problem of

observer-expectancy bias. Auk 100:755-757.

Bart, J., and J.D. Schoultz. 1984. Reliability of singing bird surveys - changes in observer

efficiency with avian density. Auk 101:307-318.

Bart, J., and J. Herrick. 1984. Diurnal timing of bird surveys. Auk 101:384-387.

Bender, D., L. Tischendorf, and L. Fahrig. 2003. Using patch isolation metrics to predict

animal movement in binary landscapes. Landscape Ecology 18:17-39.

http://www.abcbirds.org/abcprograms/domestic/iba/ohio.html

49

Blake, J.G., and W.G. Hoppes. 1986. Influence of resource abundance on use of tree-fall

gaps by birds in an isolated woodlot. Auk 103:328-340.

Bonter, D.N., S.A. Gauthreaux Jr., and T.M. Donovan. 2009. Characteristics of important

stopover locations for migrating birds: remote sensing with radar in the Great Lakes

basin. Conservation Biology 23:440-448.




Brewster, J.P., and T.R. Simons. 2009. Testing the importance of auditory detections in

avian point counts. Journal of Field Ornithology 80:178-182.






Burnham, K.P. 1981. Summarizing remarks: environmental influences. Studies in Avian

Biology 6:324-325.

Burnham, K.P., D.R. Anderson. 2002. Model selection and multimodel inference: a

practical information-theoretic approach. Springer, New York.

Carlisle, J.D., K.L. Olmstead, C.H. Richart, and D.L. Swanson. 2012. Food availability,

foraging behavior, and diet of autumn migrant landbirds in the Boise foothills of

southwestern Idaho. Condor 114:449-461.

Cashion, E. 2011. Avian use of riparian habitats and the conservation reserve program:

migratory stopover in agroecosystems. Masters Thesis. The Ohio State University,

Columbus, Ohio.








50




Cox, J. 1988. The influence of forest size on transient and resident bird species occupying

maritime hammocks of northeastern Florida. Florida Field Naturalist 16:25-34.

Cullen, S. 2002. Trees and wind: wind scales and speeds. Journal of Arboriculture

28:237-242.

Cyr, A. 1981. Limitation and variability in hearing ability in censusing birds. Studies in


Dail, D., and L. Madsen. 2011. Models for estimating abundance from repeated counts of

an open metapopulation. Biometrics 67:577-587.

Dawson, D.G. 1981. Counting birds for a relative measure of density. Studies in Avian

Biology 6:12-16.

Dawson, D.K., and M.G. Efford. 2009. Bird population density estimated from acoustic

signals. Journal of Applied Ecology 46:1201-1209.

Deppe, J.L., and J.T. Rotenberry. 2008. Scale-dependent habitat use by fall migratory

birds: vegetation structure, floristics, and geography. Ecological Monographs

78:461-487.

DeSante, D.F. 1983. Annual variability in the abundance of migrant landbirds on

southeast Farallon Island, California. Auk 100:826-852.

Diefenbach, D.R., D.W. Brauning, and J.A. Mattice. 2003. Variability in grassland bird

counts related to observer differences and species detection rates. Auk 120:1168-

1179.

Dixon, P.M. 2002. Bootstrap resampling. Pages 212-220 In A. El-Shaarawi and W. W.

Piegorsch, editors. Encyclopedia of environmetrics. John Wiley & Sons, Chichester,

England.

Dreyer, J., D. Hoekman, and C. Gratton. 2012. Lake-derived midges increase abundance

of shoreline terrestrial arthropods via multiple trophic pathways. Oikos 121:252-258.

eBird. 2012. eBird: An online database of bird distribution and abundance. eBird, Ithaca,

New York. URL: http://www.ebird.org. Accessed: 11/28/2012.

Emlen, J.T., and M.J. Dejong. 1992. Counting birds - the problem of variable hearing

abilities. Journal of Field Ornithology 63:26-31.

http://www.ebird.org/

51

Emlen, J.T., and M.J. Dejong. 1981. The application of song detection threshold distance

to census operations. Studies in Avian Biology 6:346-352.

ESRI. 2011. ArcGis Desktop: Release 10. Redlands, CA: Environmental Systems

Research Institute. .




Foundation.

Ewert, D.N., M.J. Hamas, R.J. Smith, M.E. Dallman, and S.W. Jorgensen. 2011.

Distribution of migratory landbirds along the northern Lake Huron shoreline. Wilson

Journal of Ornithology 123:536-547.

Faanes, C., and D. Bystrak. 1981. The role of observer bias in the North American

Breeding Bird Survey. Studies in Avian Biology 6:353-359.

Farmer, R., M. Leonard, and A. Horn. 2012. Observer effects and avian-call-count survey

quality: rare-species biases and overconfidence. Auk 129:76-86.

Farnsworth, G.L., K.H. Pollock, J.D. Nichols, T.R. Simons, J.E. Hines, and J.R. Sauer.

2002. A removal model for estimating detection probabilities from point count


Fiske, I.J., and R.B. Chandler. 2011. Unmarked: An R Package for fitting hierarchical

models of wildlife occurrence and abundance. Journal of Statistical Software 43:1-

23. URL: http://www.jstatsoft.org/v43/i10/.

Fortin, D., F. Liechti, and B. Bruderer. 1999. Variation in the nocturnal flight behaviour

of migratory birds along the northwest coast of the Mediterranean Sea. Ibis 141:480-

488.

France, K.E., M. Burger, T.G. Howard, M.D. Schlesinger, K.A. Perkins, M. MacNeil, D.

Klein, and D.N. Ewert. 2012. Final report for Lake Ontario migratory bird stopover

project. Prepared by The Nature Conservancy for the New York State Department of

Environmental Conservation, in fulfillment of a grant from the New York Great

Lakes Protection Fund (C303907).

Graber, J.W., and R.R. Graber. 1983. Feeding rates of warblers in spring. Condor 85:139-

150.

Gu, W.D., and R.K. Swihart. 2004. Absent or undetected? Effects of non-detection of

species occurrence on wildlife-habitat models. Biological Conservation 116:195-

203.

http://www.jstatsoft.org/v43/i10/

52



Biology 189:227-234.

Herms, D.A., A.K. Stone, and M.K. Brewer. 2006. The battle of the borer continues in

Ohio - an emerald ash borer update. Pages 40 - 48 In J. A. Chatfield, E. A. Draper,

D. E. Dyke, P. J. Bennett, and J. F. Boggs, editors. Ornamental Plants: Annual

Reports and Research Reviews 2005. OARDC/OSU Extension Special Circular 197.

Herms, D.A., A.K. Stone, and J.A. Chatfield. 2004. Emerald Ash Borer: The beginning

of the end of Ash in North America? Pages 62-71 In J. A. Chatfield, E. A. Draper, H.

M. Mathers, D. E. Dyke, P. J. Bennett, and J. F. Boggs, editors. Ornamental Plants:

Annual Reports and Research Reviews 2003. OARDC/OSU Extension Special

Circular 193.

Hijmans, R.J., and J. van Etten. 2012. Raster. Geographic analysis and modeling with

raster data: R package version 2.0-08. URL: http://CRAN.R-

project.org/package=raster.


Ecology 71:1279-1288.

Hoekman, D., J. Dreyer, R.D. Jackson, P.A. Townsend, and C. Gratton. 2011. Lake to

land subsidies: experimental addition of aquatic insects increases terrestrial

arthropod densities. Ecology 92:2063-2072.

Holland, J.D., D.G. Bert, and L. Fahrig. 2004. Determining the spatial scale of species'

response to habitat. Bioscience 54:227-233.

Hutto, R.L. 1985a. Seasonal-changes in the habitat distribution of transient insectivorous

birds in southeastern Arizona - competition mediated. Auk 102:120-132.

Hutto, R.L. 1985b. Habitat selection by non-breeding, migratory land birds. Pages 455-

476 In M. L. Cody, editor. Habitat selection in birds. Academic Press, Inc., Orlando.

Jenni, L., M. Schaub. 2003. Behavioural and physiological reactions to environmental

variation in bird migration: a review. Pages 155-171 In P. Berthold, E. Gwinner, and

E. Sonnenschein, editors. Avian migration. Springer, New York.

Johnson, D.H. 2008. In defense of indices: the case of bird surveys. Journal of Wildlife

Management 72:857-868.

Johnson, M.D. 2007. Measuring habitat quality: A review. Condor 109:489-504.

Kaatz, M.R. 1955. The Black Swamp: a study in historical geography. Annals of the

Association of American Geographers 45:1-35.

http://cran.r-project.org/package=raster


53

Keller, G.S., and R.H. Yahner. 2007. Seasonal forest-patch use by birds in fragmented

landscapes of south-central Pennsylvania. Wilson Journal of Ornithology 119:410-

418.

Keller, G.S., B.D. Ross, D.S. Klute, and R.H. Yahner. 2009. Temporal changes in

migratory bird use of edges during spring and fall seasons in Pennsylvania.

Northeastern Naturalist 16:535-552.



Kéry, M., M. Schaub. 2012. Bayesian population analysis using WinBUGS: a

hierarchical perspective. Elsevier, Waltham, MA, USA.

Kéry, M. 2002. Inferring the absence of a species - A case study of snakes. Journal of

Wildlife Management 66:330-338.

Kéry, M., J.A. Royle, and H. Schmid. 2005. Modeling avian abundance from replicated

counts using binomial mixture models. Ecological Applications 15:1450-1461.

Kincaid, T., and A. Olsen. 2012. spsurvey: Spatial survey design and analysis. R package

version 2.5. URL: http://www.epa.gov.proxy.lib.ohio-state.edu/nheerl/arm/.




Lanyon, S.M., and C.F. Thompson. 1986. Site fidelity and habitat quality as determinants

of settlement pattern in male Painted Buntings. Condor 88:206-210.

Legendre, P., and E.D. Gallagher. 2001. Ecologically meaningful transformations for

ordination of species data. Oecologia 129:271-280.

Lowery Jr., G. 1945. Trans-Gulf spring migration of birds and the coastal hiatus. The

Wilson Bulletin 57:92-121.



27:164-170.

MacDade, L.S., P.G. Rodewald, and K.A. Hatch. 2011. Contribution of emergent aquatic

insects to refueling in spring migrant songbirds. Auk 128:127-137.

Marques, T.A., L. Thomas, S.G. Fancy, S.T. Buckland, and C. Handel. 2007. Improving

estimates of bird density using multiple-covariate distance sampling. Auk 124:1229-

1243.

http://www.epa.gov.proxy.lib.ohio-state.edu/nheerl/arm/

54

Martin, T.E. 1980. Diversity and abundance of spring migratory birds using habitat

islands on the great plains. Condor 82:430-439.

Martin, T.E., and J.R. Karr. 1986. Patch utilization by migrating birds - resource oriented.

Ornis Scandinavica 17:165-174.

Martin, T.G., B.A. Wintle, J.R. Rhodes, P.M. Kuhnert, S.A. Field, S.J. Low-Choy, A.J.

Tyre, and H.P. Possingham. 2005. Zero tolerance ecology: improving ecological

inference by modelling the source of zero observations. Ecology Letters 8:1235-

1246.



Mattsson, B.J. and M.R. Marshall. 2009. Occupancy modeling as a framework for

designing avian monitoring programs: A case study along Appalachian streams in

southern West Virginia. Pages: 617-632 In T.D. Rich, C. Arizmendi, D. Demarest

and C. Thompson, editors. Proceedings of the Fourth International Partners in Flight

Conference, 13-16 February 2008. McAllen, Texas.

McCallum, D.A. 2005. A conceptual guide to detection probability for point counts and

other count-based survey methods.

McCann, J.M., S.E. Mabey, L.J. Niles, C. Bartlett, and P. Kerlinger. 1993. A regional

study of coastal migratory stopover habitat for Neotropical migrant songbirds: land

management implications. Transactions of the North American Wildlife and Natural

Resources Conference 58:398-407.

McGarigal, K., S. Cushman, and S. Stafford. 2000. Multivariate statistics for wildlife and

ecology research. Springer-Verlag, New York.








42.




York.

55

Moore, F.R., M.S. Woodrey, J.J. Buler, S. Woltmann, and T.R. Simons. 2005a.





Moore, F.R., R.J. Smith, and R. Sandberg. 2005b. Stopover ecology of intercontinental




Moore, J.E., D.M. Scheiman, and R.K. Swihart. 2004. Field comparison of removal and

modified double-observer modeling for estimating detectability and abundance of

birds. Auk 121:865-876.

Mukhin, A., N. Chernetsov, and D. Kishkinev. 2008. Acoustic information as a distant

cue for habitat recognition by nocturnally migrating passerines during landfall.

Behavioral Ecology 19:716-723.



Nichols, J.D., J.E. Hines, J.R. Sauer, F.W. Fallon, J.E. Fallon, and P.J. Heglund. 2000. A

double-observer approach for estimating detection probability and abundance from

point counts. Auk 117:393-408.

NOAA. 2006. Coastal Change Analysis Program (C-CAP) Regional Land Cover.

Charleston, SC: National Oceanic & Atmospheric Administration (NOAA) Coastal

Services Center. URL: www.csc.noaa.gov/digitalcoast/data/ccapregional.

Norvell, R.E., F.P. Howe, J.R. Parrish, and F. Thompson III. 2003. A seven-year

comparison of relative-abundance and distance-sampling methods. Auk 120:1013-

1028.

Ohio Breeding Bird Atlas II. 2012. Unpublished Results (2006 - 2011). URL:

http://bird.atlasing.org/Atlas/OH/Main?viewResults=1. Accessed: 11/28/2012.

Oksanen, J., F.G. Blanchet, R. Kindt, et al. 2011. vegan: Community Ecology Package.

URL: http://CRAN.R-project.org/package=vegan.

Packett, D.L., and J.B. Dunning Jr. 2009. Stopover habitat selection by migrant landbirds

in a fragmented forest-agricultural landscape. Auk 126:579-589.

Pennington, D.N., J. Hansel, and R.B. Blair. 2008. The conservation value of urban

riparian areas for landbirds during spring migration: land cover, scale, and

vegetation effects. Biological Conservation 141:1235-1248.

http://www.csc.noaa.gov/digitalcoast/data/ccapregional

http://bird.atlasing.org/Atlas/OH/Main?viewResults=1

http://cran.r-project.org/package=vegan

56

Peterjohn, B.G. 1989. The birds of Ohio. Indiana University Press, Bloomington, Indiana.

Peterjohn, B.G., D.L. Rice. 1991. The Ohio Breeding Bird Atlas (1982 - 1987).

Department of Natural Resources, Columbus, OH.



33.

R Development Core Team. 2012. R: a language and environment for statistical

computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-

900051-07-0. URL: http://www.R-project.org/.

Ramsey, F., and M.J. Scott. 1981. Tests of hearing ability. Studies in Avian Biology

6:341-345.

Reidy, J.L., F.R. Thompson III, and J. Bailey. 2011. Comparison of methods for

estimating density of forest songbirds from point counts. The Journal of Wildlife


Robbins, C.S., and R.W. Stallcup. 1981. Problems with separating species with similar

habits and vocalizations. Studies in Avian Biology 6:360-365.




Rodewald, P.G., and S.N. Matthews. 2005. Landbird use of riparian and upland forest

stopover habitats in an urban landscape. Condor 107:259-268.



Auk 124:1063-1074.

Royle, A. and R. Chandler. 2012. An introduction to hierarchical modeling and inference

in ecology using unmarked. URL: http://www.pwrc.usgs.gov/Royalvideo.cfm.

Accessed: 11/28/2012.

Royle, J.A. 2004. N-mixture models for estimating population size from spatially

replicated counts. Biometrics 60:108-115.

Royle, J.A., R.M. Dorazio. 2008. Hierarchical modeling and inference in ecology.

Academic Press, Boston, Massachusetts.

Sandberg, R., and F.R. Moore. 1996a. Fat stores and arrival on the breeding grounds:


http://www.r-project.org/

http://www.pwrc.usgs.gov/Royalvideo.cfm

57

Sandberg, R., and F.R. Moore. 1996b. Migratory orientation of Red-eyed Vireos, Vireo

olivaceus, in relation to energetic condition and ecological context. Behavioral

Ecology and Sociobiology 39:1-10.




Sauer, J.R., B.G. Peterjohn, and W.A. Link. 1994. Observer differences in the North

American Breeding Bird Survey. Auk 111:50-62.

Saunders, W.E. 1934. Losing the bird songs. Auk 51:503-506.

Schaub, M., R. Schwilch, and L. Jenni. 1999. Does tape-luring of migrating Eurasian

Reed-Warblers increase number of recruits or capture probability? Auk 116:1047-

1053.

Simons, T.R., K.H. Pollock, J.M. Wettroth, M.W. Alldredge, K. Pacifici, and J.P.

Brewster. 2009. Sources of measurement error, misclassification error, and bias in

auditory avian point count data. Pages 237-254 In D. L. Thomson, E. G. Cooch, and

M. J. Conroy, editors. Modeling Demographic Processes in Marked Populations.

Environmental and Ecological Statistics, Series 3, Springer, New York.

Simons, T.R., M.W. Alldredge, K.H. Pollock, and J.M. Wettroth. 2007. Experimental

analysis of the auditory detection process on avian point counts. Auk 124:986-999.

Skagen, S.K., C.P. Melcher, W.H. Howe, and F.L. Knopf. 1998. Comparative use of

riparian corridors and oases by migrating birds in southeast Arizona. Conservation

Biology 12:896-909.

Skagen, S.K., J.F. Kelly, C. Van Riper III, R.L. Hutto, D.M. Finch, D.J. Krueper, and

C.P. Melcher. 2005. Geography of spring landbird migration through riparian

habitats in southwestern North America. Condor 107:212-227.



239.



Smith, R.J., F.R. Moore, and C.A. May. 2007. Stopover habitat along the shoreline of

northern Lake Huron, Michigan: emergent aquatic insects as a food resource for

spring migrating landbirds. Auk 124:107-121.


58

Smith, R.J., M.J. Hamas, D.N. Ewert, and M.E. Dallman. 2004. Spatial foraging

differences in American Redstarts along the shoreline of northern Lake Huron

during spring migration. Wilson Bulletin 116:48-55.

Smith, R.J., and M.I. Hatch. 2008. A comparison of shrub-dominated and forested habitat

use by spring migrating landbirds in northeastern Pennsylvania. Condor 110:682-

693.

Somershoe, S.G., and C.R. Chandler. 2004. Use of oak hammocks by Neotropical

migrant songbirds: The role of area and habitat. Wilson Bulletin 116:56-63.

Stevens, D., and A. Olsen. 2004. Spatially balanced sampling of natural resources.

Journal of the American Statistical Association 99:262-278.

Strode, P.K. 2009. Spring tree species use by migrating Yellow-rumped Warblers in

relation to phenology and food availability. Wilson Journal of Ornithology 121:457-

468.

Summerville, K.S., T.O. Crist, J.K. Kahn, and J.C. Gering. 2003. Community structure of

arboreal caterpillars within and among four tree species of the eastern deciduous

forest. Ecological Entomology 28:747-757.

Thompson, W.L. 2002. Towards reliable bird surveys: Accounting for individuals present

but not detected. Auk 119:18-25.

Tischendorf, L., D.J. Bender, and L. Fahrig. 2003. Evaluation of patch isolation metrics

in mosaic landscapes for specialist vs. generalist dispersers. Landscape Ecology

18:41-50.




U.S. Army Corps of Engineers. 2009. Western Lake Erie Basin - state of the basin.

USFWS. 2012. Classification of wetlands and deepwater habitats of the United States.

Washington D.C.: U.S. Department of the Interior Fish and Wildlife Service

(USFWS). URL: http://www.fws.gov.proxy.lib.ohio-

state.edu/wetlands/Data/index.html.

USGS. 1999. National Hydrology Dataset. Reston ,Virginia: United States Geological

Survey (USGS). URL: http://nhd.usgs.gov.proxy.lib.ohio-state.edu/.

Van Horne, B. 1983. Density as a misleading indicator of habitat quality. Journal of

Wildlife Management 47:893 -901.

http://www.fws.gov.proxy.lib.ohio-state.edu/wetlands/Data/index.html


http://nhd.usgs.gov.proxy.lib.ohio-state.edu/

59



Wenger, S.J., and M.C. Freeman. 2008. Estimating species occurrence, abundance, and

detection probability using zero-inflated distributions. Ecology 89:2953-2959.

Wood, E.M., A.M. Pidgeon, F. Liu, and D.J. Mladenoff. 2012. Birds see the trees inside

the forest: The potential impacts of changes in forest composition on songbirds

during spring migration. Forest Ecology and Management 280:176-186.

60

canopy

height

small

trees

medium

trees

large

trees stemhits

medium and

large dead trees

canopy height 1.00 -0.14 -0.18 0.53 -0.13 -0.18

small trees -0.14 1.00 0.40 -0.11 -0.32 -0.06

medium trees -0.18 0.40 1.00 0.05 -0.30 -0.04

large trees 0.53 -0.11 0.05 1.00 -0.21 -0.30

stemhits -0.13 -0.32 -0.30 -0.21 1.00 -0.11

medium and

large dead trees -0.18 -0.06 -0.04 -0.30 -0.11 1.00

Table 2.1 - Correlation matrix of habitat structure variables used in constructing habitat

structure PCA. Habitat structure variables were measured within 30 m of each of the 135

point count locations within the Western Lake Erie Basin of Ohio, USA. The habitat

variables used in the matrix were: stemhits = number of understory stems between 0.5 and

3.0 m, small trees = number of trees between 8–23 cm dbh, medium trees = number of trees

between 23–38 cm dbh, large trees = number of trees >38 cm dbh, medium and large dead

trees = number of medium (23–38 cm dbh), and large (>38 cm dbh) dead ash trees, and

average canopy height (m).

61

Variables PC1 PC2

canopy height -0.798 0.214

small trees 0.448 0.644

medium trees 0.376 0.68

large trees -0.758 0.427

stemhits -0.064 -0.732

medium and large dead trees 0.44 -0.22

Table 2.2- Factor loadings for the first and second principal component

axes (PC1 and PC2) in the habitat structure PCA. The first axis explained

29.2% of the total variation (eigenvalue = 1.75). The second axis

explained an additional 28.2% of the total variation (eigenvalue = 1.69).

The variables represent components of the habitat structure at each of the

135 point count locations in the Western Lake Erie Basin, Ohio, USA.

Variables were measured within 30m of each point count location. The

variables are: stemhits = number of understory stems between 0.5 and 3.0

m, small trees = number of trees between 8–23 cm dbh, medium trees =

number of trees between 23–38 cm dbh, large trees = number of trees >38

cm dbh, medium and large dead trees = number of medium (23–38 cm

dbh), and large (>38 cm dbh) dead ash trees, and average canopy height

(m).

62

oak maple boxelder cottonwood walnut hickory dead ash elm

oak 1.00 -0.02 -0.41 -0.48 -0.06 0.41 -0.08 0.13

maple -0.02 1.00 -0.17 -0.21 -0.29 0.00 0.07 -0.07

boxelder -0.41 -0.17 1.00 0.05 0.03 -0.29 -0.07 -0.13

cottonwood -0.48 -0.21 0.05 1.00 -0.21 -0.40 -0.30 -0.35

walnut -0.06 -0.29 0.03 -0.21 1.00 0.06 -0.10 0.19

hickory 0.41 0.00 -0.29 -0.40 0.06 1.00 -0.07 0.18

dead ash -0.08 0.07 -0.07 -0.30 -0.10 -0.07 1.00 0.12

elm 0.13 -0.07 -0.13 -0.35 0.19 0.18 0.12 1.00

Table 2.3 - Correlation matrix of counts of trees >23 cm dbh in the eight most abundant tree species

groups used in constructing the species composition PCA. Counts of trees by species were made within

30 m of each of the 135 point count locations in the Western Lake Erie Basin of Ohio, USA. The species

groups used in the matrix were: oaks (Quercus spp.), maples (Acer spp.), boxelder (A. negundo), eastern

cottonwood (Populus deltoides), walnuts (Juglans spp.), hickories (Carya spp.), dead ashes (Fraxinus

spp.), and elm (Ulmus spp.). A Hellinger transformation was performed on the raw species counts to

make the data more appropriate for the PCA.

62

63

Variables PC1 PC2

oaks 0.741 -0.114

maples -0.774 -0.087

boxelder 0.177 -0.179

cottonwood -0.522 0.329

walnuts 0.699 0.036

hickories 0.487 0.37

dead ashes 0.147 0.787

elms 0.142 -0.709

Table 2.4- Factor loadings for the first and second principal component

axes (PC1 and PC2) of the species composition PCA. The first axis

explained 27.7% of the total variation (eigenvalue = 2.22). The second

axis explained an additional 17.8% of the total variation (eigenvalue =

1.42). The variables represent counts of trees >23 cm dbh within the

eight most abundant tree species groups. Counts were made within 30

m of each of the 135 point count locations in the Western Lake Erie

Basin, Ohio, USA. The species groups are: oaks (Quercus spp.), maples

(Acer spp.), boxelder (Acer negundo), eastern cottonwood (Populus

deltoides), walnuts (Juglans spp.), hickories (Carya spp.), dead ashes

(Fraxinus spp.), and elms (Ulmus spp.).

64

variable mean median minimum maximum SD

area 30 ha 12 ha 1 ha 498 ha 64 ha

distance to river 4.5 km 1.6 km 0 km 19 km 4.8 km

distance to coast 5.6 km 2.9 km .03 km 21.8 km 6.4 km

isolation 0.29 0.23 0 0.89 0.21

wetland cover 0.36 0.32 0 1 0.35

Table 2.5- Summary statistics for broad-scale variables for the 135 point count

locations used in 2011 and 2012. The variables are: area = area (ha) of the forest

patch in which the point count was conducted, distance to a river = distance (km)

from point count location to the nearest stream or river, distance to coast =

distance (km) from point count location to the nearest point of Lake Erie coastline,

isolation = proportion of the landscape within 0.5 km of point count location

classified as forest, wetland cover = proportion of the landscape within 0.2 km of

point count location classified as wetland.

65

Model nPars AIC ΔAIC

patch isolation + wetland cover 12 873.06 0

distance to coast + isolation + wetland cover 13 873.72 0.66

distance to river + isolation + wetland cover 13 874.27 1.21

composition PC1 + composition PC2 + isolation

+ wetland cover 14 875.48 2.42

structure PC1 + structure PC2 + isolation +

wetland cover 14 877.05 3.99

FULL model (minus area) 18 881.67 8.61

Table 2.6 - Top models for gamma (arrival rate) for Blackpoll Warbler in 2011. nPars

= number of parameters in the model. AIC = Akaike Information Criterion. ΔAIC =


include Julian date and a quadratic term for Julian date and that only models with

ΔAIC < 10 are shown. Variables in the models are: area = area (ha) of the forest patch

in which the point count was conducted, distance to river = distance (km) from point

count location to nearest stream or river, distance to coast = distance (km) from point

count location to nearest point of Lake Erie coastline, isolation = proportion of the

landscape within 0.5 km of point count location classified as forest, wetland cover =

proportion of the landscape within 0.2 km of point count location classified as wetland,

composition PC1 = species composition PC1, composition PC2 = species composition

PC2, structure PC1 = habitat structure PC1, structure PC2 = habitat Structure PC2.

FULL model (minus area) = a model with all variables included except patch area (ha).

66


distance from coast + area + wetland cover 13 675.37 0.00

FULL model (minus area) 18 675.53 0.16


distance to coast + wetland cover 12 676.04 0.67

FULL model (minus isolation) 18 676.57 1.19

distance to river + distance to coast + wetland cover 13 676.61 1.23

distance to coast + composition PC1 + composition PC2 +


distance to coast + structure PC1 + structure PC2 + wetland

cover 14 678.37 3.00


structure PC1 + structure PC2 15 678.58 3.21

distance to river + distance to coast + structure PC1 +

structure PC2 14 678.68 3.31

distance to river + distance to coast 12 678.79 3.42

distance to river + distance to coast + composition PC1 +

composition PC2 14 679.51 4.14

distance to river + distance to coast + area 13 679.77 4.40

distance to river + distance to coast + isolation 13 680.25 4.88




include Julian date and a quadratic term for Julian date and that only models with ΔAIC

< 5 are shown. Variables in the models are: area = area (ha) of the forest patch in which

the point count was conducted, distance to river = distance (km) from point count






PC1 = habitat structure PC1, structure PC2 = habitat Structure PC2. FULL model = a

model with all variables included except the variable inside the parentheses.

67

Top variables 2011 – BLPW 2011 – BTNW 2012 – BLPW 2012 – BTNW

distance to coast

-3.61 (-5.29 to -1.93) -0.27 (-0.46 to -0.1)

distance to river

-0.13 (-0.28 to 0.02)

Julian date -0.25 (-0.69 to 0.19)

-0.84 ( -1.13 to -

0.54) -1.09 (-1.81 to -0.36) -1.43 (-1.82 to -1.05)

Julian date (quadratic) -0.87 (-1.32 to -0.43) -1.08 (-1.35 to -0.81) -0.95 (-1.54 to -0.36) -1.1 (-1.4 to -0.79)

area

-0.9 (-2.1 to 0.3)

composition PC1

-0.14 (-0.27 to 0)

composition PC2

-0.15 (-0.3 to -0.01)

isolation -0.71 (-1.01 to -0.41)

-0.14 (-0.29 to 0.02)

structure PC1

-0.03 (-0.18 to 0.1)

structure PC2

0.16 (0.02 to 0.3)

wetland cover 0.64 (0.4 to 0.89)

0.47 (0.17 to 0.76)

Table 2.8- Coefficients for the top variables in the single-species models in each year. 95% confidence intervals for estimates

are given inside the parentheses. Note that the data are scaled and centered so interpretation of coefficients should be for

strength and direction of relationship. BLPW = Blackpoll Warbler. BTNW = Black-throated Green Warbler. The variables

are: area = area (ha) of the forest patch in which the point count was conducted, distance to river = distance (km) from point

count location to the nearest stream or river, distance to coast = distance (km) from point count location to nearest point of

Lake Erie coastline, isolation = proportion of the landscape within 0.5 km of point count location classified as forest, wetland

cover = proportion of the landscape within 0.2 km of point count location classified as wetland, composition PC1 = species

composition PC1, composition PC2 = species composition PC2, structure PC1 = habitat structure PC1, structure PC2 = habitat

Structure PC2.

67

68


distance to river + composition PC1 + composition PC2 13 1855.72 0.00

composition PC1 + composition PC2 + wetland cover 13 1856.67 0.96

composition PC1 + composition PC2 12 1856.87 1.16

distance to river + composition PC1 + composition PC2 + wetland

cover 14 1857.14 1.43



distance to river + composition PC1 + composition PC2 + isolation 14 1857.34 1.62

distance to river + composition PC1 + composition PC2 + area 14 1857.44 1.73

distance to coast + composition PC1 + composition PC2 13 1857.99 2.27


distance to coast + composition PC1 + composition PC2 + wetland

cover 14 1858.63 2.91

composition PC1 + composition PC2 + area + wetland cover 14 1858.67 2.95

composition PC1 + composition PC2 + isolation + wetland cover 14 1858.67 2.96

composition PC1 + composition PC2 + area 13 1858.72 3.00


composition PC1 + composition PC2 + isolation 13 1858.86 3.14

composition PC1 + composition PC2 + structure PC1 + structure

PC2 14 1859.12 3.41

distance to river + composition PC1 + composition PC2 + structure

PC1 + structure PC2 15 1859.13 3.42

distance to river + wetland cover 12 1859.21 3.49


composition PC1 + composition PC2 + structure PC1 + structure

PC2 + wetland cover 15 1859.60 3.88



2011. nPars = number of parameters in the model. AIC = Akaike Information

Criterion. ΔAIC = Difference in AIC between top-ranked model and specified model.

Note that all models include Julian date and a quadratic term for Julian date and that

only models with ΔAIC < 4 are shown. Variables in the models are: area = area (ha)

of the forest patch in which the point count was conducted, distance to river =

distance (km) from point count location to nearest stream or river, distance to coast =

distance (km) from point count location to nearest point of Lake Erie coastline,

isolation = proportion of the landscape within 0.5 km of point count location

classified as forest, wetland cover = proportion of the landscape within 0.2 km of

point count location classified as wetland, composition PC1 = species composition

PC1, composition PC2 = species composition PC2, structure PC1 = habitat structure

PC1, structure PC2 = habitat Structure PC2.

69


distance to coast + structure PC1 + structure PC2 +

isolation 12 1856.18 0.00

distance to coast + structure PC1 + structure PC2 11 1857.31 1.13

distance to coast + isolation 10 1857.69 1.50

distance to coast + structure PC1 + structure PC2 + area 12 1857.99 1.81

distance to river + distance to coast + structure PC1 +

structure PC2 12 1858.09 1.91



distance to coast 9 1858.60 2.41

distance to coast + area 10 1858.89 2.71



distance to coast + structure PC1 + structure PC2 +








isolation 12 1860.38 4.20


distance to coast + area + wetland cover 11 1860.85 4.67

Table 2.10 - Top models for gamma (arrival rate) for Black-throated Green

Warbler in 2012. nPars = number of parameters in the model. AIC = Akaike

Information Criterion. ΔAIC = Difference in AIC between top-ranked model and

specified model. Note that all models include Julian date and a quadratic term for

Julian date and that only models with ΔAIC < 5 are shown. Variables in the

models are: area = area (ha) of the forest patch in which the point count was

conducted, distance to river = distance (km) from point count location to nearest

stream or river, distance to coast = distance (km) from point count location to

nearest point of Lake Erie coastline, isolation = proportion of the landscape

within 0.5 km of point count location classified as forest, wetland cover =

proportion of the landscape within 0.2 km of point count location classified as

wetland, composition PC1 = species composition PC1, composition PC2 =

species composition PC2, structure PC1 = habitat structure PC1, structure PC2 =

habitat Structure PC2.

70








isolation 15 12103.31 2.62

area 15 12103.60 2.91

distance to river 15 12103.83 3.15


isolation + wetland cover 16 12104.31 3.62








area + wetland cover 16 12104.71 4.03



distance to river + isolation 16 12105.29 4.61

distance to river + area 16 12105.59 4.90

Table 2.11 – Top gamma (finite rate of increase) models for the transient warbler

guild in 2011. nPars = number of parameters in the model. AIC = Akaike Information

Criterion. ΔAIC = Difference in AIC between top-ranked model and specified model.

Note that all models include Julian date and a quadratic term for Julian date and that

only models with ΔAIC < 5 are shown. Variables in the models are: area = area (ha)

of the forest patch in which the point count was conducted, distance to river = distance

(km) from point count location to nearest stream or river, distance to coast = distance

(km) from point count location to nearest point of Lake Erie coastline, isolation =

proportion of the landscape within 0.5 km of point count location classified as forest,

wetland cover = proportion of the landscape within 0.2 km of point count location

classified as wetland, composition PC1 = species composition PC1, composition PC2

= species composition PC2, structure PC1 = habitat structure PC1, structure PC2 =

habitat Structure PC2.

71







distance to river 15 8140.50 2.36

isolation 15 8140.55 2.40

area 15 8140.58 2.44










distance to river + isolation 16 8142.32 4.18

distance to river + area 16 8142.37 4.23

isolation + wetland cover 16 8142.45 4.31


area + wetland cover 16 8142.52 4.37


Table 2.12 – Top gamma (finite rate of increase) models for the transient warbler

guild in 2012. nPars = number of parameters in the model. AIC = Akaike

Information Criterion. ΔAIC = Difference in AIC between top-ranked model and

specified model. Note that all models include Julian date and a quadratic term for

Julian date and that only models with ΔAIC < 5 are shown. Variables in the models

are: area = area (ha) of the forest patch in which the point count was conducted,

distance to river = distance (km) from point count location to nearest stream or

river, distance to coast = distance (km) from point count location to nearest point of

Lake Erie coastline, isolation = proportion of the landscape within 0.5 km of point

count location classified as forest, wetland cover = proportion of the landscape

within 0.2 km of point count location classified as wetland, composition PC1 =

species composition PC1, composition PC2 = species composition PC2, structure

PC1 = habitat structure PC1, structure PC2 = habitat Structure PC2.

72

Figure 2.1 – Location of the study area (defined in purple in the upper right and

bottom) in the Western Lake Erie Basin, Ohio, USA. In the upper left and right,

darker grey areas bordered in black represent the state of Ohio. Yellow areas in

the upper right and left represent the counties in which the study took place.

73

Figure 2.2 - Point count locations (dots) used in 2011 (grey), in 2012 (white), and in both

years (black) within the study area in the Western Lake Erie Basin, Ohio, USA. Distance

bands used in the stratified GRTS sampling are shown in yellow (0–0.573 km), orange

(0.573–2.648 km), pink (2.648–10.09 km), and blue (10.09–22 km). Forest patches within

the study area are shown in green.

73

74

Figure 2.3- Detection probability for Blackpoll Warbler in 2011 as a function of observer and

time of day. Grey areas represent 95% confidence intervals for estimated detection

probabilities for each observer (colored lines).

74

75

Figure 2.4 - Number of Blackpoll Warblers gained between visits to a site (new arrivals) in the

Western Lake Erie Basin, Ohio, USA as a function of patch isolation in 2011. The grey area represents

the 95% confidence interval of the predicted estimate (black line).

75

76


Western Lake Erie Basin, Ohio, USA as a function of date in 2011. The grey area represents the

95% confidence interval of the predicted estimate (black line).

76

77

Figure 2.6 - Detection probability for Blackpoll Warblers in the Western Lake Erie Basin,

Ohio, USA in 2012 as a function of a categorical estimate for wind speed. Dots represent

estimates for detection probability and whiskers represent 95% confidence intervals of the

estimates.

77

78


Western Lake Erie Basin, Ohio, USA as a function of patch distance from the lakeshore in

2012. The grey area represents the 95% confidence interval of the predicted estimate (black

line).

78

79

Figure 2.8- Coefficient estimates and 95% confidence intervals for variables

influencing gamma (arrival rate) in the top models for Blackpoll Warbler and

Black-throated Green Warbler in 2011 and 2012. The red-dotted line is at X

= 0. Variables are: Structure 1 = habitat structure PC1, Structure 2 = habitat

Structure PC2, Distance from coast = distance (km) from point count location

to nearest point of Lake Erie coastline, Area = area (ha) of the forest patch in

which the point count was conducted, Composition 2 = species composition

PC2, Composition 1 = species composition PC1, Distance to a River =

distance (km) from point count location to nearest stream or river, Date -

squared = quadratic of Julian Date, Patch isolation = proportion of the

landscape within 0.5 km of point count location classified as forest, Date =

Julian date, Wetland cover = proportion of the landscape within 0.2 km of

point count location classified as wetland.

80

Figure 2.9 - Average density of Blackpoll Warblers (individuals per hectare) in forest patches of

the Western Lake Erie Basin, Ohio, USA, during each day of 2011 and 2012 spring migrations

(approximately 5 May to 30 May annually).

80

81

Figure 2.10 - Average number of Blackpoll Warblers per forest patch in the Western Lake Erie

Basin, Ohio, USA, during each day of 2011 and 2012 spring migrations (approximately 5 May to

30 May annually).

81

82

Figure 2.11- Detection probability for Black-throated Green Warblers in the Western

Lake Erie Basin, Ohio, USA in 2011 as a function of observer and time of day. Grey

areas represent 95% confidence intervals for estimated detection probabilities for each

observer (colored lines).

82

83

Figure 2.12 - Number of Black-throated Green Warblers gained between visits to a site (new arrivals)

in the Western Lake Erie Basin, Ohio, USA as a function of date in 2011. The grey area represents the

95% confidence interval of the predicted estimate (black line).

83

84

Figure 2.13 - Estimated detection probabilities for Black-

throated Green Warbler for different observers in the Western

Lake Erie Basin, Ohio, USA in 2012. Lines coming off of

estimates represent 95% Confidence intervals.

85

Figure 2.14 - Number of Black-throated Green Warblers gained between visits to a site (new arrivals) in

the Western Lake Erie Basin, Ohio, USA in 2012 as a function of distance from the coast. The grey area

represents the 95% confidence interval of the predicted estimate (black line).

85

86

Figure 2.15 - Number of Black-throated Green Warblers gained between visits to a site (new arrivals) in

the Western Lake Erie Basin, Ohio, USA in 2012 as a function of date. The grey area represents the 95%

confidence interval of the predicted estimate (black line).

86

87

Figure 2.16- Average density of Black-throated Green Warblers (individuals per hectare) within

forest patches of the Western Lake Erie Basin, Ohio, USA, during each day of 2011 and 2012

spring migrations (approximately 21 April – 29 May annually).

87

88

Figure 2.17- Average number of Black-throated Green Warblers per forest patch in the Western Lake

Erie Basin, Ohio, USA, during each day of 2011 and 2012 spring migrations (approximately 21 April

to 29 May annually).

88

89

Figure 2.18 – Raw counts of individuals in the transient warbler guild

by species and year. Counts were made during spring migration in the

Western Lake Erie Basin, Ohio, USA. The species are: Black-and-

white Warbler (BAWW), Bay-breasted Warbler (BBWA),

Blackburnian Warbler (BLBW), Blackpoll Warbler (BLPW), Black-

throated Blue Warbler (BTBW), Black-throated Green Warbler

(BTNW), Blue-winged Warbler (BWWA), Canada Warbler

(CAWA), Chestnut-sided Warbler (CSWA), Cerulean Warbler

(CERW), Cape May Warbler (CMWA), Golden-winged Warbler

(GWWA), Hooded Warbler (HOWA), Magnolia Warbler (MAWA),

Yellow-rumped Warbler (MYWA), Nashville Warbler (NAWA),

Northern Parula (NOPA), Orange-crowned Warbler (OCWA), Palm

Warbler (PAWA), Pine Warbler (PIWA), Prairie Warbler (PRAW),

Tennessee Warbler (TEWA), Wilson’s Warbler (WIWA), and

Yellow-throated Warbler (YTWA).

90



(5) = high wind. Whiskers indicate 95% confidence intervals of estimates.

90

91

Figure 2.20 - Detection probability for transient warbler guild in the Western Lake Erie

Basin, Ohio, USA as a function of time of day and observer in 2011. Grey areas

represent 95% confidence intervals for estimated detection probabilities for each


92

Figure 2.21 - Detection probability for the transient warbler guild in the Western Lake Erie

Basin, Ohio, USA as a function of wind speed and observer in 2012. (0) = no wind, (4) = high

wind. Whiskers indicate 95% confidence intervals of estimates.

92

93


Erie Basin, Ohio, USA as a function of time of day and observer in 2012. Grey areas

represent 95% confidence intervals for estimated detection probabilities for each


93

94

Figure 2.23 – Average density of guild members (individuals per hectare) within forest patches of

the Western Lake Erie Basin, Ohio, USA, during each day of 2011 and 2012 spring migrations

(approximately 19 April to 30 May annually).

94

95

Figure 2.24 - Average number of guild members per forest patch within the Western Lake Erie

Basin, Ohio, USA, during each day of 2011 and 2012 spring migrations (approximately 19 April to

30 May annually).

95

96

Figure 2.25 - Boxplot showing estimated midge abundance at survey locations

in the Western Lake Erie Basin, Ohio, USA, in relation to survey date in 2012.

Categories used to estimate midge abundance were: (0) = 0 midges, (1) = 1–10

midges, (2) = 11–100 midges, (3) = 101–500 midges, (4) = 501–1000 midges,

(5) = 1001–5000 midges, (6) = >5001 midges.

97

Figure 2.26 – Boxplot showing estimated midge abundance at survey locations

in the Western Lake Erie Basin, Ohio, USA, in relation to wetland cover in the

landscape in 2012. Categories used to estimate midge abundance were: (0) = 0


501–1000 midges, (5) = 1001–5000 midges, (6) = >5001 midges.

98

Figure 2.27- Boxplot showing estimated midge abundance at survey locations in the

Western Lake Erie Basin, Ohio, USA, in relation to distance to coast in 2012.

Categories used to estimate midge abundance were: (0) = 0 midges, (1) = 1-10 midges,

(2) = 11-100 midges, (3) = 101-500 midges, (4) = 501-1000 midges, (5) = 1000-5000

midges, (6) = >5001 midges.

99

Figure 2.28 – Pairwise comparisons between transient warbler

abundance in each midge abundance category while accounting for

Julian date and a quadratic of Julian date. Estimates for the difference in

transient warbler abundance between midge abundance categories (Y-

axis) are shown on the plot as dots. 95% Tukey family-wise confidence

intervals extend from estimated differences. Categories used to estimate

midge abundance were: (0) = 0 midges, (1) = 1-10 midges, (2) = 11-100

midges, (3) = 101-500 midges, (4) = 501-1000 midges, (5) = 1000-5000

midges, (6) = >5001 midges. A dotted line is at X = 0. ‘*’ = P < 0.05,

‘**’ = P < 0.01, ‘***’ = P < 0.001.

100

Bibliography

Aborn, D.A., and F.R. Moore. 1997. Pattern of movement by Summer Tanagers (Piranga

rubra) during migratory stopover: A telemetry study. Behaviour 134:1077-1100.




Alerstam, T. 2001. Detours in bird migration. Journal of Theoretical Biology 209:319-

331.



Alerstam, T., A. Hedenström, and S. Åkesson. 2003. Long-distance migration: evolution

and determinants. Oikos 103:247-260.

Alerstam, T., and P.H. Enckell. 1979. Unpredictable habitats and evolution of bird

migration. Oikos 33:228-232.

Alldredge, M.W., K.H. Pollock, T.R. Simons, J.A. Collazo, S.A. Shriner, and D. Johnson.

2007a. Time-of-detection method for estimating abundance from point-count


Alldredge, M.W., T.R. Simons, and K.H. Pollock. 2007b. Factors affecting aural

detections of songbirds. Ecological Applications 17:948-955.

American Bird Conservancy. 2010. Globally important bird areas in Ohio. URL:

http://www.abcbirds.org/abcprograms/domestic/iba/ohio.html. Accessed:

11/28/2012.

Andrén, H. 1994. Effects of habitat fragmentation on birds and mammals in landscapes

with different proportions of suitable habitat - a review. Oikos 71:355-366.

http://www.abcbirds.org/abcprograms/domestic/iba/ohio.html

101

Askins, R.A., J.F. Lynch, and R. Greenberg. 1990. Population declines in migratory birds

in eastern North America. Pages 1-57 In D. M. Power, editor. Current Ornitholgy,

Volume 7. Plenum Press, New York.





271:875-882.

Balph, D.F., and M.H. Balph. 1983. On the psychology of watching birds: the problem of

observer-expectancy bias. Auk 100:755-757.

Bart, J., and J.D. Schoultz. 1984. Reliability of singing bird surveys - changes in observer

efficiency with avian density. Auk 101:307-318.

Bart, J., and J. Herrick. 1984. Diurnal timing of bird surveys. Auk 101:384-387.

Bearhop, S., G.M. Hilton, S.C. Votier, and S. Waldron. 2004. Stable isotope ratios

indicate that body condition in migrating passerines is influenced by winter habitat.

Proceedings of the Royal Society of London Series B 271:215-218.

Bell, C.P. 2000. Process in the evolution of bird migration and pattern in avian

ecogeography. Journal of Avian Biology 31:258-265.

Bender, D., L. Tischendorf, and L. Fahrig. 2003. Using patch isolation metrics to predict

animal movement in binary landscapes. Landscape Ecology 18:17-39.

Berthold, P. 1998. Bird migration: genetic programs with high adaptability. Zoology

101:235-245.

Berthold, P., W. Fiedler, R. Schlenker, and U. Querner. 1998. 25-year study of the

population development of central European songbirds: A general decline most

evident in long-distance migrants. Naturwissenschaften 85:350-353.

Berthold, P. 1975. Migration: control and metabolic physiology. Pages 77-128 In D. S.

Farner and J. R. King, editors. Avian Biology, Volume 5. Academic Press, New

York.

Biebach, H., W. Friedrich, and G. Heine. 1986. Interaction of bodymass, fat, foraging and

stopover period in trans-Sahara migrating passerine birds. Oecologia 69:370-379.

102

Black Swamp Bird Observatory. 2011. Black Swamp Bird Observatory speaks out

against wind turbines in migratory bird stopover habitat. URL:

http://www.bsbo.org/wind_energy.htm. Accessed: 11/28/2012.

Blake, J.G., and W.G. Hoppes. 1986. Influence of resource abundance on use of tree-fall

gaps by birds in an isolated woodlot. Auk 103:328-340.

Blem, C.R. 1980. The energetics of migration. Pages 175 - 224 In S. A. Gauthreaux Jr.,

editor. Animal migration, orientation, and navigation. Academic Press, New York.

Böhning-Gaese, K., M.L. Taper, and J.H. Brown. 1993. Are declines in North American

insectivorous songbirds due to causes on the breeding range. Conservation Biology

7:76-86.

Bonter, D.N., S.A. Gauthreaux Jr., and T.M. Donovan. 2009. Characteristics of important

stopover locations for migrating birds: remote sensing with radar in the Great Lakes

basin. Conservation Biology 23:440-448.




Bowlin, M.S., I. Bisson, J. Shamoun-Baranes, J.D. Reichard, N. Sapir, P.P. Marra, T.H.

Kunz, D.S. Wilcove, A. Hedenström, C.G. Guglielmo, S. Åkesson, M. Ramenofsky,

and M. Wikelski. 2010. Grand challenges in migration biology. Integrative and

Comparative Biology 50:261-279.

Brewster, J.P., and T.R. Simons. 2009. Testing the importance of auditory detections in

avian point counts. Journal of Field Ornithology 80:178-182.

Bruderer, B., and V. Salewski. 2008. Evolution of bird migration in a biogeographical

context. Journal of Biogeography 35:1951-1959.






Burnham, K.P. 1981. Summarizing remarks: environmental influences. Studies in Avian

Biology 6:324-325.

http://www.bsbo.org/wind_energy.htm

103

Burnham, K.P., D.R. Anderson. 2002. Model selection and multimodel inference: a

practical information-theoretic approach. Springer, New York.

Carlisle, J.D., K.L. Olmstead, C.H. Richart, and D.L. Swanson. 2012. Food availability,

foraging behavior, and diet of autumn migrant landbirds in the Boise foothills of

southwestern Idaho. Condor 114:449-461.

Cashion, E. 2011. Avian use of riparian habitats and the conservation reserve program:

migratory stopover in agroecosystems. Masters Thesis. The Ohio State University,

Columbus, Ohio.

Catry, P., V. Encarnação, A. Araújo, P. Fearon, A. Fearon, M. Armelin, and P. Delaloye.

2004. Are long-distance migrant passerines faithful to their stopover sites? Journal of


Chapman, B.B., C. Brönmark, J.Å. Nilsson, and L.A. Hansson. 2011. The ecology and

evolution of partial migration. Oikos 120:1764-1775.











Cox, G.W. 1985. The evolution of avian migration systems between temperate and

tropical regions of the new world. American Naturalist 126:451-474.

Cox, G.W. 1968. The role of competition in the evolution of migration. Evolution

22:180-192.

Cox, J. 1988. The influence of forest size on transient and resident bird species occupying

maritime hammocks of northeastern Florida. Florida Field Naturalist 16:25-34.

Cullen, S. 2002. Trees and wind: wind scales and speeds. Journal of Arboriculture

28:237-242.

104

Cushman, S.A., and K. McGarigal. 2002. Hierarchical, multi-scale decomposition of

species-environment relationships. Landscape Ecology 17:637-646.

Cyr, A. 1981. Limitation and variability in hearing ability in censusing birds. Studies in


Dail, D., and L. Madsen. 2011. Models for estimating abundance from repeated counts of

an open metapopulation. Biometrics 67:577-587.

Dawson, D.G. 1981. Counting birds for a relative measure of density. Studies in Avian

Biology 6:12-16.

Dawson, D.K., and M.G. Efford. 2009. Bird population density estimated from acoustic

signals. Journal of Applied Ecology 46:1201-1209.

Deppe, J.L., and J.T. Rotenberry. 2008. Scale-dependent habitat use by fall migratory

birds: vegetation structure, floristics, and geography. Ecological Monographs

78:461-487.

DeSante, D.F. 1983. Annual variability in the abundance of migrant landbirds on

southeast Farallon Island, California. Auk 100:826-852.

Diefenbach, D.R., D.W. Brauning, and J.A. Mattice. 2003. Variability in grassland bird

counts related to observer differences and species detection rates. Auk 120:1168-

1179.

Dingle, H. 1980. Ecology and evolution of migration. Pages 1-101 In S. A. Gauthreaux

Jr., editor. Animal migration, orientation, and navigation. Academic Press, New

York.

Dingle, H., and V.A. Drake. 2007. What is migration? Bioscience 57:113-121.

Dixon, P.M. 2002. Bootstrap resampling. Pages 212-220 In A. El-Shaarawi and W. W.

Piegorsch, editors. Encyclopedia of environmetrics. John Wiley & Sons, Chichester,

England.

Drent, R., A. Fox, and J. Stahl. 2006. Travelling to breed. Journal of Ornithology

147:122-134.

Dreyer, J., D. Hoekman, and C. Gratton. 2012. Lake-derived midges increase abundance

of shoreline terrestrial arthropods via multiple trophic pathways. Oikos 121:252-258.

eBird. 2012. eBird: An online database of bird distribution and abundance. eBird, Ithaca,

New York. URL: http://www.ebird.org. Accessed: 11/28/2012.

http://www.ebird.org/

105

Emlen, J.T., and M.J. Dejong. 1992. Counting birds - the problem of variable hearing

abilities. Journal of Field Ornithology 63:26-31.

Emlen, J.T., and M.J. Dejong. 1981. The application of song detection threshold distance

to census operations. Studies in Avian Biology 6:346-352.

ESRI. 2011. ArcGis Desktop: Release 10. Redlands, CA: Environmental Systems

Research Institute. .




Foundation.

Ewert, D.N., M.J. Hamas, R.J. Smith, M.E. Dallman, and S.W. Jorgensen. 2011.

Distribution of migratory landbirds along the northern Lake Huron shoreline. Wilson

Journal of Ornithology 123:536-547.




Thompson III, and N. Warnock. 2010a. Conserving migratory land birds in the New

World: do we know enough? Ecological Applications 20:398-418.




Thompson III, and N. Warnock. 2010b. Recent advances in understanding migration

systems of New World land birds. Ecological Monographs 80:3-48.

Faanes, C., and D. Bystrak. 1981. The role of observer bias in the North American

Breeding Bird Survey. Studies in Avian Biology 6:353-359.

Farmer, R., M. Leonard, and A. Horn. 2012. Observer effects and avian-call-count survey

quality: rare-species biases and overconfidence. Auk 129:76-86.

Farnsworth, G.L., K.H. Pollock, J.D. Nichols, T.R. Simons, J.E. Hines, and J.R. Sauer.

2002. A removal model for estimating detection probabilities from point count


Fiske, I.J., and R.B. Chandler. 2011. Unmarked: An R Package for fitting hierarchical

models of wildlife occurrence and abundance. Journal of Statistical Software 43:1-

23. URL: http://www.jstatsoft.org/v43/i10/.

http://www.jstatsoft.org/v43/i10/

106

Fortin, D., F. Liechti, and B. Bruderer. 1999. Variation in the nocturnal flight behaviour

of migratory birds along the northwest coast of the Mediterranean Sea. Ibis 141:480-

488.

France, K.E., M. Burger, T.G. Howard, M.D. Schlesinger, K.A. Perkins, M. MacNeil, D.

Klein, and D.N. Ewert. 2012. Final report for Lake Ontario migratory bird stopover

project. Prepared by The Nature Conservancy for the New York State Department of

Environmental Conservation, in fulfillment of a grant from the New York Great

Lakes Protection Fund (C303907).

Graber, J.W., and R.R. Graber. 1983. Feeding rates of warblers in spring. Condor 85:139-

150.

Greenberg, R. 1980. Demographic aspects of long-distance migration. Pages 493-504 In

A. Keast and E. Morton, editors. Migrant birds in the Neotropics: ecology, behavior,

distribution, and conservation. Smithsonian Institution Press, Washington D.C.

Gu, W.D., and R.K. Swihart. 2004. Absent or undetected? Effects of non-detection of

species occurrence on wildlife-habitat models. Biological Conservation 116:195-

203.

Gunnarsson, T.G., J.A. Gill, J. Newton, P.M. Potts, and W.J. Sutherland. 2005. Seasonal

matching of habitat quality and fitness in a migratory bird. Proceedings of the Royal

Society Series B 272:2319-2323.

Harrison, X.A., J.D. Blount, R. Inger, R.D. Norris, and S. Bearhop. 2010. Carry-over

effects as drivers of fitness differences in animals. Journal of Animal Ecology 80:4-

18.



Biology 189:227-234.

Herms, D.A., A.K. Stone, and M.K. Brewer. 2006. The battle of the borer continues in

Ohio - an emerald ash borer update. Pages 40 - 48 In J. A. Chatfield, E. A. Draper,

D. E. Dyke, P. J. Bennett, and J. F. Boggs, editors. Ornamental Plants: Annual

Reports and Research Reviews 2005. OARDC/OSU Extension Special Circular 197.

Herms, D.A., A.K. Stone, and J.A. Chatfield. 2004. Emerald Ash Borer: The beginning

of the end of Ash in North America? Pages 62-71 In J. A. Chatfield, E. A. Draper, H.

M. Mathers, D. E. Dyke, P. J. Bennett, and J. F. Boggs, editors. Ornamental Plants:

Annual Reports and Research Reviews 2003. OARDC/OSU Extension Special

Circular 193.

107

Hijmans, R.J., and J. van Etten. 2012. Raster. Geographic analysis and modeling with

raster data: R package version 2.0-08. URL: http://CRAN.R-

project.org/package=raster.


Ecology 71:1279-1288.

Hoekman, D., J. Dreyer, R.D. Jackson, P.A. Townsend, and C. Gratton. 2011. Lake to

land subsidies: experimental addition of aquatic insects increases terrestrial

arthropod densities. Ecology 92:2063-2072.

Holland, J.D., D.G. Bert, and L. Fahrig. 2004. Determining the spatial scale of species'

response to habitat. Bioscience 54:227-233.

Holmes, R.T. 2007. Understanding population change in migratory songbirds: long-term

and experimental studies of Neotropical migrants in breeding and wintering areas.

Ibis 149:2-13.

Hutto, R.L. 1985a. Seasonal-changes in the habitat distribution of transient insectivorous

birds in southeastern Arizona - competition mediated. Auk 102:120-132.

Hutto, R.L. 1985b. Habitat selection by non-breeding, migratory land birds. Pages 455-

476 In M. L. Cody, editor. Habitat selection in birds. Academic Press, Inc., Orlando.

Jenni, L., M. Schaub. 2003. Behavioural and physiological reactions to environmental

variation in bird migration: a review. Pages 155-171 In P. Berthold, E. Gwinner, and

E. Sonnenschein, editors. Avian migration. Springer, New York.

Jetz, W., C.H. Sekercioglu, and K. Böhning-Gaese. 2008. The worldwide variation in

avian clutch size across species and space. Plos Biology 6: e303. doi:

10.1371/journal.pbio.0060303.

Johnson, D.H. 2008. In defense of indices: the case of bird surveys. Journal of Wildlife


Johnson, M.D. 2007. Measuring habitat quality: A review. Condor 109:489-504.

Johnson, M.D., T.W. Sherry, R.T. Holmes, and P.P. Marra. 2006. Assessing habitat

quality for a migratory songbird wintering in natural and agricultural habitats.

Conservation Biology 20:1433-1444.

Kaatz, M.R. 1955. The Black Swamp: a study in historical geography. Annals of the

Association of American Geographers 45:1-35.



108

Keller, G.S., and R.H. Yahner. 2006. Declines of migratory songbirds: evidence for

wintering-ground causes. Northeastern Naturalist 13:83-92.

Keller, G.S., and R.H. Yahner. 2007. Seasonal forest-patch use by birds in fragmented

landscapes of south-central Pennsylvania. Wilson Journal of Ornithology 119:410-

418.

Keller, G.S., B.D. Ross, D.S. Klute, and R.H. Yahner. 2009. Temporal changes in

migratory bird use of edges during spring and fall seasons in Pennsylvania.

Northeastern Naturalist 16:535-552.



Kéry, M., M. Schaub. 2012. Bayesian population analysis using WinBUGS: a

hierarchical perspective. Elsevier, Waltham, MA, USA.

Kéry, M. 2002. Inferring the absence of a species - A case study of snakes. Journal of

Wildlife Management 66:330-338.

Kéry, M., J.A. Royle, and H. Schmid. 2005. Modeling avian abundance from replicated

counts using binomial mixture models. Ecological Applications 15:1450-1461.

Kincaid, T., and A. Olsen. 2012. spsurvey: Spatial survey design and analysis. R package

version 2.5. URL: http://www.epa.gov.proxy.lib.ohio-state.edu/nheerl/arm/.

Kirby, J.S., A.J. Stattersfield, S.H.M. Butchart, M.I. Evans, R.F.A. Grimmett, V.R. Jones,

J. O'Sullivan, G.M. Tucker, and I. Newton. 2008. Key conservation issues for

migratory land and waterbird species on the world's major flyways. Bird

Conservation International 18:49-73.




Lack, D. 1968. Bird migration and natural selection. Oikos 19:1-9.

Lanyon, S.M., and C.F. Thompson. 1986. Site fidelity and habitat quality as determinants

of settlement pattern in male Painted Buntings. Condor 88:206-210.

Latta, S.C., and M.E. Baltz. 1997. Population limitation in Neotropical migratory birds:

comments. Auk 114:754-762.

http://www.epa.gov.proxy.lib.ohio-state.edu/nheerl/arm/

109

Legendre, P., and E.D. Gallagher. 2001. Ecologically meaningful transformations for

ordination of species data. Oecologia 129:271-280.

Lindström, Å. 2005. Overview: migration itself. Pages 249 - 251 In R. Greenberg and P.

P. Marra, editors. Birds of Two Worlds. The Johns Hopkins University Press,

Baltimore.

Loria, D.E., and F.R. Moore. 1990. Energy demands of migration on Red-eyed Vireos

(Vireo olivaceus). Behavioral Ecology 1:24-35.

Lowery Jr., G. 1945. Trans-Gulf spring migration of birds and the coastal hiatus. The

Wilson Bulletin 57:92-121.



27:164-170.

MacDade, L.S., P.G. Rodewald, and K.A. Hatch. 2011. Contribution of emergent aquatic

insects to refueling in spring migrant songbirds. Auk 128:127-137.

Marques, T.A., L. Thomas, S.G. Fancy, S.T. Buckland, and C. Handel. 2007. Improving

estimates of bird density using multiple-covariate distance sampling. Auk 124:1229-

1243.

Marra, P.P., K.A. Hobson, and R.T. Holmes. 1998. Linking winter and summer events in

a migratory bird by using stable-carbon isotopes. Science 282:1884-1886.

Martin, T.E. 1980. Diversity and abundance of spring migratory birds using habitat

islands on the great plains. Condor 82:430-439.

Martin, T.E., and J.R. Karr. 1986. Patch utilization by migrating birds - resource oriented.

Ornis Scandinavica 17:165-174.

Martin, T.G., B.A. Wintle, J.R. Rhodes, P.M. Kuhnert, S.A. Field, S.J. Low-Choy, A.J.

Tyre, and H.P. Possingham. 2005. Zero tolerance ecology: improving ecological

inference by modelling the source of zero observations. Ecology Letters 8:1235-

1246.



Mattsson, B.J. and M.R. Marshall. 2009. Occupancy modeling as a framework for

designing avian monitoring programs: A case study along Appalachian streams in

southern West Virginia. Pages: 617-632 In T.D. Rich, C. Arizmendi, D. Demarest

110

and C. Thompson, editors. Proceedings of the Fourth International Partners in Flight

Conference, 13-16 February 2008. McAllen, Texas.

McCallum, D.A. 2005. A conceptual guide to detection probability for point counts and

other count-based survey methods.

McCann, J.M., S.E. Mabey, L.J. Niles, C. Bartlett, and P. Kerlinger. 1993. A regional

study of coastal migratory stopover habitat for Neotropical migrant songbirds: land

management implications. Transactions of the North American Wildlife and Natural

Resources Conference 58:398-407.

McGarigal, K., S. Cushman, and S. Stafford. 2000. Multivariate statistics for wildlife and

ecology research. Springer-Verlag, New York.




Mitchell, G.W., A.E.M. Newman, M. Wikelski, and D.R. Norris. 2012. Timing of

breeding carries over to influence migratory departure in a songbird: an automated

radiotracking study. Journal of Animal Ecology 81:1024-1033.





42.

Moore, F.R., T.R. Simons. 1992. Habitat suitability and stopover ecology of Neotropical

landbird migrants. Pages 345-355 In J. M. Hagan and D. W. Johnston, editors.

Ecology and conservation of Neotropical migrant landbirds. Smithsonian Institution,

Washington.




York.

Moore, F.R., M.S. Woodrey, J.J. Buler, S. Woltmann, and T.R. Simons. 2005.





111

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

74:47-54.

Moore, F.R., R.J. Smith, and R. Sandberg. 2005. Stopover ecology of intercontinental




Moore, J.E., D.M. Scheiman, and R.K. Swihart. 2004. Field comparison of removal and

modified double-observer modeling for estimating detectability and abundance of

birds. Auk 121:865-876.

Mukhin, A., N. Chernetsov, and D. Kishkinev. 2008. Acoustic information as a distant

cue for habitat recognition by nocturnally migrating passerines during landfall.

Behavioral Ecology 19:716-723.



Newton, I. 2004. Population limitation in migrants. Ibis 146:197-226.

Nichols, J.D., J.E. Hines, J.R. Sauer, F.W. Fallon, J.E. Fallon, and P.J. Heglund. 2000. A

double-observer approach for estimating detection probability and abundance from

point counts. Auk 117:393-408.

NOAA. 2006. Coastal Change Analysis Program (C-CAP) Regional Land Cover.

Charleston, SC: National Oceanic & Atmospheric Administration (NOAA) Coastal

Services Center. URL: www.csc.noaa.gov/digitalcoast/data/ccapregional.

Norris, D.R., P.P. Marra, T.K. Kyser, T.W. Sherry, and L.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 271:59-64.

Norvell, R.E., F.P. Howe, J.R. Parrish, and F. Thompson III. 2003. A seven-year

comparison of relative-abundance and distance-sampling methods. Auk 120:1013-

1028.

Ohio Breeding Bird Atlas II. 2012. Unpublished Results (2006 - 2011). URL:

http://bird.atlasing.org/Atlas/OH/Main?viewResults=1. Accessed: 11/28/2012.

Oksanen, J., F.G. Blanchet, R. Kindt, et al. 2011. vegan: Community Ecology Package.

URL: http://CRAN.R-project.org/package=vegan.

http://www.csc.noaa.gov/digitalcoast/data/ccapregional

http://bird.atlasing.org/Atlas/OH/Main?viewResults=1

http://cran.r-project.org/package=vegan

112

Packett, D.L., and J.B. Dunning Jr. 2009. Stopover habitat selection by migrant landbirds

in a fragmented forest-agricultural landscape. Auk 126:579-589.

Paxton, E.H., M.K. Sogge, S.L. Durst, T.C. Theimer, and J.R. Hatten. 2007. The ecology

of the Southwestern Willow Flycatcher in central Arizona—a 10-year synthesis

report: U.S. Geological Survey Open-File Report 2007-1381.

Pennington, D.N., J. Hansel, and R.B. Blair. 2008. The conservation value of urban

riparian areas for landbirds during spring migration: land cover, scale, and

vegetation effects. Biological Conservation 141:1235-1248.

Peterjohn, B.G., J.R. Sauer, and C.S. Robbins. 1995. Population trends from the North

American Breeding Bird Survey. Pages 3-39 In T. E. Martin and D. Finch M.,

editors. Ecology and management of Neotropical migratory birds. Oxford University

Press, New York.

Peterjohn, B.G. 1989. The birds of Ohio. Indiana University Press, Bloomington, Indiana.

Peterjohn, B.G., D.L. Rice. 1991. The Ohio Breeding Bird Atlas (1982 - 1987).

Department of Natural Resources, Columbus, OH.



33.

Pulido, F. 2007. The genetics and evolution of avian migration. Bioscience 57:165-174.

R Development Core Team. 2012. R: a language and environment for statistical

computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-

900051-07-0. URL: http://www.R-project.org/.

Ramenofsky, M., and J.C. Wingfield. 2007. Regulation of migration. Bioscience 57:135-

143.

Ramsey, F., and M.J. Scott. 1981. Tests of hearing ability. Studies in Avian Biology

6:341-345.

Rappole, J.H., and M.V. McDonald. 1994. Cause and effect in population declines of

migratory birds. Auk 111:652-660.

Rappole, J., and P. Jones. 2002. Evolution of old and new world migration systems.

Ardea 90:525-537.

http://www.r-project.org/

113

Reidy, J.L., F.R. Thompson III, and J. Bailey. 2011. Comparison of methods for

estimating density of forest songbirds from point counts. The Journal of Wildlife


Robbins, C.S., and R.W. Stallcup. 1981. Problems with separating species with similar

habits and vocalizations. Studies in Avian Biology 6:360-365.

Robbins, C.S., J.R. Sauer, R.S. Greenberg, and S. Droege. 1989. Population declines in

North American birds that migrate to the Neotropics. Proceedings of the National

Academy of Sciences of the United States of America 86:7658-7662.




Rodewald, P.G., and S.N. Matthews. 2005. Landbird use of riparian and upland forest

stopover habitats in an urban landscape. Condor 107:259-268.

Rodewald, P.G., and M.C. Brittingham. 2004. Stopover habitats of landbirds during fall:

Use of edge-dominated and early-successional forests. Auk 121:1040-1055.



Auk 124:1063-1074.

Royle, A. and R. Chandler. 2012. An introduction to hierarchical modeling and inference

in ecology using unmarked. URL: http://www.pwrc.usgs.gov/Royalvideo.cfm.

Accessed: 11/28/2012.

Royle, J.A. 2004. N-mixture models for estimating population size from spatially

replicated counts. Biometrics 60:108-115.

Royle, J.A., R.M. Dorazio. 2008. Hierarchical modeling and inference in ecology.

Academic Press, Boston, Massachusetts.

Salewski, V., and B. Bruderer. 2007. The evolution of bird migration - a synthesis.

Naturwissenschaften 94:268-279.

Sandberg, R., and F.R. Moore. 1996a. Fat stores and arrival on the breeding grounds:


Sandberg, R., and F.R. Moore. 1996b. Migratory orientation of Red-eyed Vireos, Vireo

olivaceus, in relation to energetic condition and ecological context. Behavioral

Ecology and Sociobiology 39:1-10.

http://www.pwrc.usgs.gov/Royalvideo.cfm

114

Sanderson, F.J., P.F. Donald, D.J. Pain, I.J. Burfield, and F.P.J. van Bommel. 2006.

Long-term population declines in Afro-Palearctic migrant birds. Biological

Conservation 131:93-105.




Sauer, J.R., B.G. Peterjohn, and W.A. Link. 1994. Observer differences in the North

American Breeding Bird Survey. Auk 111:50-62.

Saunders, W.E. 1934. Losing the bird songs. Auk 51:503-506.

Schaub, M., R. Schwilch, and L. Jenni. 1999. Does tape-luring of migrating Eurasian

Reed-Warblers increase number of recruits or capture probability? Auk 116:1047-

1053.

Sekercioglu, C.H. 2007. Conservation ecology: Area trumps mobility in fragment bird

extinctions. Current Biology 17:283-286.

Sergio, F., and I. Newton. 2003. Occupancy as a measure of territory quality. Journal of

Animal Ecology 72:857-865.

Sheehy, J., C.M. Taylor, and D.R. Norris. 2011. The importance of stopover habitat for

developing effective conservation strategies for migratory animals. Journal of


Sherry, T.W., R.T. Holmes. 1995. Summer versus winter limitation of populations: what

are the issues and what is the evidence. Pages 85-120 In T. E. Martin and D. M.

Finch, editors. Ecology and management of Neotropical migratory birds. Oxford

University Press, New York.

Sherry, T.W., and R.T. Holmes. 1996. Winter habitat quality, population limitation, and

conservation of neotropical nearctic migrant birds. Ecology 77:36-48.

Sherry, T.W., R.T. Holmes. 1993. Are populations of neotropical migrant birds limited in

summer or winter - implications for management. Pages 47-57 In D. M. Finch and P.

W. Stangel, editors. Status and management of Neotropical migratory birds. General

Technical Report RM-229, U.S. Department of Agriculture Forest Service Rocky

Mountain Forest and Range Experiment Station, Fort Collins, Colorado.

Sillett, T.S., and R.T. Holmes. 2002. Variation in survivorship of a migratory songbird

throughout its annual cycle. Journal of Animal Ecology 71:296-308.


115

Simons, T.R., K.H. Pollock, J.M. Wettroth, M.W. Alldredge, K. Pacifici, and J.P.

Brewster. 2009. Sources of measurement error, misclassification error, and bias in

auditory avian point count data. Pages 237-254 In D. L. Thomson, E. G. Cooch, and

M. J. Conroy, editors. Modeling Demographic Processes in Marked Populations.

Environmental and Ecological Statistics, Series 3, Springer, New York.

Simons, T.R., M.W. Alldredge, K.H. Pollock, and J.M. Wettroth. 2007. Experimental

analysis of the auditory detection process on avian point counts. Auk 124:986-999.

Skagen, S.K., C.P. Melcher, W.H. Howe, and F.L. Knopf. 1998. Comparative use of

riparian corridors and oases by migrating birds in southeast Arizona. Conservation

Biology 12:896-909.

Skagen, S.K., J.F. Kelly, C. Van Riper III, R.L. Hutto, D.M. Finch, D.J. Krueper, and

C.P. Melcher. 2005. Geography of spring landbird migration through riparian

habitats in southwestern North America. Condor 107:212-227.



239.



Smith, R.J., F.R. Moore, and C.A. May. 2007. Stopover habitat along the shoreline of

northern Lake Huron, Michigan: emergent aquatic insects as a food resource for

spring migrating landbirds. Auk 124:107-121.

Smith, R.J., M.J. Hamas, D.N. Ewert, and M.E. Dallman. 2004. Spatial foraging

differences in American Redstarts along the shoreline of northern Lake Huron

during spring migration. Wilson Bulletin 116:48-55.

Smith, R.J., and M.I. Hatch. 2008. A comparison of shrub-dominated and forested habitat

use by spring migrating landbirds in northeastern Pennsylvania. Condor 110:682-

693.

Somershoe, S.G., and C.R. Chandler. 2004. Use of oak hammocks by Neotropical

migrant songbirds: The role of area and habitat. Wilson Bulletin 116:56-63.

Steadman, D. 2005. Paleoecology and fossil history of migratory landbirds. Pages 5-17 In

R. Greenberg and P. P. Marra, editors. Birds of two worlds. The Johns Hopkins

University Press, Baltimore.

116

Stevens, D., and A. Olsen. 2004. Spatially balanced sampling of natural resources.

Journal of the American Statistical Association 99:262-278.

Strode, P.K. 2009. Spring tree species use by migrating Yellow-rumped Warblers in

relation to phenology and food availability. Wilson Journal of Ornithology 121:457-

468.

Summerville, K.S., T.O. Crist, J.K. Kahn, and J.C. Gering. 2003. Community structure of

arboreal caterpillars within and among four tree species of the eastern deciduous

forest. Ecological Entomology 28:747-757.

Taverner, P.A. 1904. A discussion of the origin of migration. Auk 21:322-333.

Thompson, W.L. 2002. Towards reliable bird surveys: Accounting for individuals present

but not detected. Auk 119:18-25.

Tischendorf, L., D.J. Bender, and L. Fahrig. 2003. Evaluation of patch isolation metrics

in mosaic landscapes for specialist vs. generalist dispersers. Landscape Ecology

18:41-50.




U.S. Army Corps of Engineers. 2009. Western Lake Erie Basin - state of the basin.

USFWS. 2012. Classification of wetlands and deepwater habitats of the United States.

Washington D.C.: U.S. Department of the Interior Fish and Wildlife Service

(USFWS). URL: http://www.fws.gov.proxy.lib.ohio-

state.edu/wetlands/Data/index.html.

USGS. 1999. National Hydrology Dataset. Reston ,Virginia: United States Geological

Survey (USGS). URL: http://nhd.usgs.gov.proxy.lib.ohio-state.edu/.

Van Horne, B. 1983. Density as a misleading indicator of habitat quality. Journal of

Wildlife Management 47:893 -901.



Wenger, S.J., and M.C. Freeman. 2008. Estimating species occurrence, abundance, and

detection probability using zero-inflated distributions. Ecology 89:2953-2959.

Wiens, J.A. 1989. Spatial scaling in ecology. Functional Ecology 3:385-397.



http://nhd.usgs.gov.proxy.lib.ohio-state.edu/

117

Wikelski, M., E.M. Tarlow, A. Raim, R.H. Diehl, R.P. Larkin, and G.H. Visser. 2003.

Costs of migration in free-flying songbirds. Nature 423:704-704.

Wood, E.M., A.M. Pidgeon, F. Liu, and D.J. Mladenoff. 2012. Birds see the trees inside

the forest: The potential impacts of changes in forest composition on songbirds

during spring migration. Forest Ecology and Management 280:176-186.

migratory stopover of songbirds in the western …...table 2.2- factor loadings for the first and...

Documents