CLIMATE CHANGE, RANGE SHIFTS, AND DIFFERENTIAL GUILD RESPONSES OF
MICHIGAN BREEDING BIRDS
By
Jodi M. Kreuser

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Fisheries and Wildlife – Master of Science
2013

ABSTRACT
CLIMATE CHANGE, RANGE SHIFTS, AND DIFFERENTIAL GUILD RESPONSES OF
MICHIGAN BREEDING BIRDS
By
Jodi M. Kreuser

There are few large-scale macroecological studies to date that use multi-species data to evaluate
the influence of climate change on range shifts in wildlife. In part, this deficiency relates to the
lack of suitable data. Recent completion of consecutive Breeding Bird Atlases in 12 states offers
a valuable opportunity to explore the influence of climate change on avian communities. My
objectives were to test for poleward range shifts among diverse avian species across broad
temporal and geographic scales, and if I found systematic shifts, to further evaluate species
among foraging guilds and migratory strategies for differential shifts. I analyzed Michigan
Breeding Bird Atlas data, which provided 2 statewide surveys at a 20-year interval, (1983 to
1988 and 2001 to 2008), and represented over 1,000,000 occurrence records for more than 200
species of breeding birds. Analyses showed systematic shifts in the distribution of bird species
with wide-ranging life histories, suggesting that a fundamental ecological change is occurring. In
species with measurable range shifts, I further observed greater shifts in the northern boundary of
southerly species. This work is among the first empirical studies in the Midwestern United States
to quantify systematic range shifts for a diverse taxon at sufficient temporal and spatial scales for
compelling inference.

Copyright by
JODI M. KREUSER
2013

ACKNOWLEDGEMENTS

Funding for this research was provided by The Boone and Crockett Endowment at Michigan
State University. Professional enhancement awards were granted through the George J. Wallace
and Martha C. Wallace Foundation, Michigan Bird Conservation Initiative, Michigan State
University (MSU) Department of Fisheries and Wildlife, MSU College of Agriculture and
Natural Resources, MSU Graduate School, and The Wildlife Society. I am very grateful for the
support provided by each of these organizations and institutions.
I am especially thankful for the mentorship, patience, and encouragement of my advisor,
Dr. William Porter. Without his guidance and continuing assistance, this thesis would not have
been possible. Special thanks to my committee members, Dr. Amy Dechen Quinn, Dr. Ashton
Shortridge, and Dr. Benjamin Zuckerberg, for their involvement and support. This project also
benefited from the assistance and expertise of other faculty members and staff including Dr. Jen
Owen and Dr. Jeffrey Andresen. I would also like to thank the Departmental staff, especially
Marcia Baar and Jill Cruth, for their support and assistance.
I wish to extend my gratitude to the Michigan Department of Natural Resources Wildlife
Division for the opportunity to collaborate and strengthen the partnership between MSU and the
MDNR. In particular, I wish to thank Karen Cleveland and Marshall Strong for serving as
mentors during my project internship. They each shared their expertise and were instrumental in
accessing and working with the Michigan Breeding Bird Atlas Data. I am also indebted to the
thousands of Michigan Breeding Bird Atlas volunteers, whose many hours gathering data made
this research possible.

iv

This thesis benefited from the support and collaboration of the many talented people
around me. In 2010, I had the honor of joining Marta Jarzyna, Nathan Snow, and Andrea
Bowling as incoming graduate students in the newly formed Quantitative Wildlife Laboratory
(QWL) at Michigan State University. I am so grateful for their support and friendship. I would
like to thank QWL member Dr. David Williams, who has been an exceptional resource for
quantitative expertise and writing R code for analysis. I am also grateful for the assistance and
support from my other QWL colleagues, Heather Porter, Andrew Crosby, Chad Blass, and Bryan
Stevens.
Finally, I am forever indebted to my family for their love and support. They believe in me
even when I don’t believe in myself, and for that I am eternally thankful. I am especially grateful
for my Mom, who has supported my every endeavor, inspired me to take chances, and
encouraged me to follow my dreams. Many thanks go to my sister, Jessica, for never failing to
lift my spirits and reminding me to be fierce. And finally, to my nieces Olivia and Sophia, it has
been such a joy to explore our shared curiosity in the stars above and the birds around us.

v

TABLE OF CONTENTS

LIST OF TABLES

vii

LIST OF FIGURES

viii

CHAPTER 1
Introduction
Study Area
Methods
Breeding Bird Atlas Data
Range Shifts among Northerly and Southerly Species Groups
Differential Range Shifts among Foraging and Migratory Groups
Results
Center of Occurrence
Range Boundary Analysis
States of Occurrence
Discussion
Management Implications

1
6
10
10
11
20
21
21
23
29
35
44

APPENDICES
Appendix A. Supplementary tables for BBA data, species groups used
in analyses, and results for individual species
Appendix B. Vita

47
48

LITERATURE CITED

59

vi

57

LIST OF TABLES

Table 1. Statewide distribution and life history traits of 41 Michigan birds, 1983
to 2008.

14

Table 2. Center of statewide occurrence analysis quantifying changes in mean
latitude for 41 birds. Michigan Breeding Bird Atlas I and II. 1983 to 2008.

21

Table 3. Center of statewide occurrence analysis quantifying changes in mean
latitude by life history strategies. Michigan Breeding Bird Atlas I and II. 1983 to
2008.

23

Table 4. Range boundary analysis for change in mean latitude between Michigan
Breeding Bird Atlas I and II, 1983-2008.

24

Table 5. Estimating shifts from the ordinary least squares regression in the range
boundary analysis, Michigan Breeding Bird Atlas, 1983 to 2008.

28

Table A1. Michigan Breeding Bird Atlas I (1983-1988) and Atlas II (2001-2008)
observation records and survey block summaries.

49

Table B1. Breeding behavior classification codes, Michigan Breeding Bird Atlas,
1983-2008.

50

Table C1. Northerly and southerly species groups.

51

Table C2. Insectivorous and non-insectivorous species groups.

52

Table C3. Neotropical and non-neotropical species groups.

53

Table D1. Results for each species in the center of occurrence, range boundary,
and states of occurrence analyses.

54

vii

LIST OF FIGURES

Figure 1. Michigan Breeding Bird Atlas regions and township delineations, 1983
to 2008.

7

Figure 2. From 1980 to 2010 in Michigan, trend analyses for (A) the average rate
of change in the annual mean temperature (°C), (B) the average rate of change in
spring precipitation (mm) during March through May, and (C) the rate of change
in maximum spring temperature (°C) during March through May. Maps produced
by ClimateWizard©, University of Washington, and The Nature Conservancy,
2013. Base climate data from the PRISM Group, Oregon State University. For
interpretation of the references to color in this and all other figures, the reader is
referred to the electronic version of this thesis.

8

Figure 3. Theoretical northward shifts from Atlas I (1983-1998) to Atlas II
(2001-2008) in (a) the southern range boundary of northerly species, and (b) the
northern range boundary of southerly species. The dashed line represents the
shifted range boundary location in Atlas II.

12

Figure 4. Latitudinal change in range boundaries (km) plotted against the change
in occupancy for (A) 20 northerly, (B) 21 southerly, (C) all 41, D) 26 insectivore,
and E) 25 Neotropical species in Michigan between Atlas I (1983-1998) and
Atlas II (2001-2008). Northerly species demonstrated a northward shift in
southern range boundaries of 10.4 km (SE = 5.8 km) (P = 0.09). Southerly
species demonstrated a northward shift in northern range boundaries of 22.7 km
(SE = 10.3 km) (P = 0.04). All species demonstrated a northward shift in range
boundaries of 13.2 km (SE = 6.4 km) (P = 0.05). Insectivore species
demonstrated a northward range shift in range boundaries of 6.8 km (SE = 8.0
km) (P = 0.41). Neotropical species demonstrated a northward range shift in
range boundaries of 16.0 km (SE = 8.7 km) (P = 0.08).

25

Figure 5. In the states of occurrence analysis, values for the change in occupancy
distributed around zero suggest no overall changes in distribution. The change in
occupancy is represented for all species. Expanding distributions have positive
values toward 1.0, and contracting distributions have negative values moving to1.0.

30

Figure 6a-c. Atlas block gains and losses for northerly and southerly species in
the Michigan Breeding Bird Atlas I and II, 1983 to 2008. Each data point
represents the number of species in each block that demonstrated a gain or loss.
Gains represent blocks with no detection in Atlas I and detection in Atlas II.
Losses represent blocks with species occurrence in Atlas I but not Atlas II.

32

viii

CHAPTER 1

INTRODUCTION
The ecological consequences of climate change include altered habitats, mismatched
phenological relationships, and range shifts in terrestrial species (Parmesan 1996; Parmesan and
Yohe 2003; Root et al. 2003). As temperatures warm, species are shifting northwards towards
cooler climes (Thomas and Lennon 1999; Hitch and Leberg 2007; La Sorte and Thompson
2007). Range shifts are predicted in response to climate change for many taxonomic groups
including mammals, amphibians, and birds (e.g. Huntley et al. 2006; Pounds et al. 2006; Davies
et al. 2009). An implication of climate-induced range expansion and contraction are novel
species assemblages, which may alter competition and predator-prey community dynamics, and
may drive higher local extirpation rates among species less likely to adapt (Charmantier et al.
2008; Stralberg et al. 2009). The challenge is finding suitable data to evaluate these predictions.
Birds are an excellent group for large-scale research because of their diverse life
histories, ease of detection relative to other classes of organisms, and extensive data regarding
distribution, habitat selection, and abundance (Gibbons et al. 2007; Gill 2007; Niven et al. 2009).
Of particular importance are statewide breeding bird atlas (BBA) projects that yield large-scale
data sets with multi-species observations across many years. The large geographic scale of BBA
projects and the complex relationships birds maintain in trophic and phenological hierarchies
allow inference on large-scale phenomena such as climate change. A powerful advantage of
repeated BBA projects, now completed in 12 states, is the opportunity to test for measurable
changes in occupancy and range shifts for individual species and among guilds (Gaston 1996;
Gaston et al. 2000). This approach may elicit important findings about northward and elevational

1

shifts as potential responses to climate change (e.g. Araujo et al. 2005; Thomas et al. 2006; La
Sorte and Thompson 2007). Li et al. (2010) predicted climate change induced range shifts with
large-scale occurrence records for a family of widespread avian species in China. They further
described an association between species occurrence and six environmental variables of
temperature and precipitation. Jimenez-Valverde et al. (2011) demonstrated a positive
relationship between climatic factors and avian distribution structure. Additional predictive
models have identified potential impacts of climate on species distribution, and regional studies,
primarily focused on European species, have documented range shifts (Thomas and Lennon
1999; Lemoine et al. 2007; Hickling et al. 2006; Thomas et al. 2006).
Many studies to date have limited inference about distribution changes due to location,
small geographical extent, elevation, and narrow species breadth. Physical geography limits
poleward distribution changes in a number of studies; these limitations include physical
boundaries imposed by large bodies of water, elevational variation, absence of an adjacent land
mass, and highly fragmented landscapes. There is also a deficit in current literature about the
specific mechanism or fundamental ecological variables driving systematic population-level
changes across heterogeneous landscapes and disparate species (Walther et al. 2002; Parmesan
2006; Thomas et al. 2006). The need remains for studies using multi-species data to quantify
changes that allow us to describe the influence of climate change on observed range shifts.
Further, we need to reduce the ambiguity imposed by elevational variation, and replicate studies
that elucidate a correlative relationship between climate change and range shifts.
While the evaluation of a correlative relationship between climate change and range
shifts is complex, advancing research at large scales in different regions is necessary for making
robust inference about the influence of this relationship on wildlife (Parmesan and Yohe 2003;

2

Walther et al. 2005; Zuckerberg et al. 2009b). These research needs are specifically relevant in
central North America in the context of extreme weather events, shifting climate patterns
described by predictive models, and the need for case studies in the Great Lakes region
(Christensen et al. 2007; Hellman et al. 2010; Hayoe et al. 2010). To date, two studies in New
York and Ohio provide the only empirical research in the United States that quantified
systematic range shifts for a taxonomic group at sufficiently broad temporal and fine spatial
scales for compelling inference (Zuckerberg et al. 2009b; Batdorf 2012). With this in mind,
choosing suitable species and scale for quantifying range changes related to climatic variables is
essential for a better understanding of this dynamic, and the ecology influencing systematic
patterns (Thomas and Lennon 1999; Melles et al. 2011).
As climate-driven ranges shift across heterogeneous landscapes, we are likely to see
differential responses among avian species with varying resource selection, migratory behavior,
and relative sensitivity to environmental changes (e.g. Cotton 2003; Sparks et al. 2005; Donnelly
et al. 2009). Currently, there is no consensus regarding specific mechanisms driving diverse
responses to environmental changes; one theory in current research suggests a genetic
component to population-level responses (Balanyá et al. 2006; Bradshaw and Holapzfel 2006;
Gienapp et al. 2007). Alternatively, the other major theory suggests that adaptive phenotypic
plasticity among individual species allows for tracking environmental changes and resources at
the population level (Przybylo et al. 2000; Réale et al. 2003; Charmantier et al. 2008). Adaptive
plasticity is strongly correlated to reproductive fitness and timing, and may describe a specific
mechanism underlying range shifts influenced by climate change (Both et al. 2006; Gienapp et
al. 2007; Møller et al. 2010). As a result, we are then able to describe differential shifts among

3

migratory and foraging guilds, relative to the adaptive ability of species to track environmental
changes and resources, as a change that may be indicative of the mechanism driving range shifts.
As an extension of resource selection adaptation and sensitivity to changes in climate, we
are likely to observe greater range shifts among species with specialized resource niches, and
lesser range shifts among species better adapted to exploit a range of resources. Differential
responses may be intensified by the interacting effects of climate change and the altered timing
of seasonal activities including migration, leaf out, and insect emergence (Root et al. 2003;
Charmantier et al. 2008; Donnelly et al 2009). The timing of egg laying and emergence of
invertebrate food resources is a crucial and synchronous seasonal event between insects and
breeding birds (Visser et al. 1998; Visser et al. 2004; Charmantier et al. 2008). Any mismatched
phenological and trophic relationships pose ecological consequences, especially for species less
likely to adapt to changes in the environment (Both and Visser 2001; Both et al. 2006;
Charmantier et al. 2008). In addition to specialized resource selection, complex annual life cycles
and endogenous constraints on phenotypic plasticity further increase sensitivity to environmental
changes among Neotropical migrant and insectivorous species (Visser et al. 1998; Cotton 2003;
Jonzén et al. 2006). As a function of selecting for broader resources, resident and short-distance
migrants are better adapted to exploit a range of habitats for resource selection, and may exhibit
less of a response to changing ecological pressures (Turner et al. 1998; Sparks et al. 2005;
Lindell et al. 2007).
In the Great Lakes region of the United States, models predict earlier insect emergence,
increased annual fecundity, and invasion by southerly and pest species in the coming decades
(Hayoe et al. 2010; Hellman et al. 2010). The historical climate data that inform model
projections also illustrate the potential of corresponding ecological data. During the last century

4

in Michigan, there were broad trends of increasing average temperatures, greater precipitation
rates, and advancing growing seasons. However, during the last thirty years there has been an
abrupt escalation of temperature and precipitation changes (Hayoe et al. 2010). These abrupt
changes correspond with the time interval of the Michigan Breeding Bird Atlas, which provided
two statewide surveys at a 20-year interval, 1983 to 1988 (Atlas I) to 2001 to 2008 (Atlas II).
These data consist of more than 1,000,000 statewide occurrence records for 233 breeding birds in
approximately 7,000 sample units (Brewer et al. 1991; MDNR 2012a). The extensive study area,
breadth of species occurrence and distribution, and total number of records provided exceptional
data for robust evaluations of broad ecological changes across time.
I hypothesized that northward range shifts have occurred among distinct breeding bird
species in Michigan for 1983 to 1988 and 2001 to 2008. I also hypothesized that because
insectivorous and Neotropical species are more sensitive to climatic changes, they would
demonstrate greater northward range shifts. To look at how diverse species may be responding
similarly to external influences, my first objective was to evaluate changes in range boundaries,
center of statewide occurrence, and occurrence status between atlas periods. If I found evidence
of northward shifts overall, my second objective was to then determine if those shifts varied
among species grouped by foraging guilds and migration strategies. Using life history variables
within the same modeling framework for the first objective, I evaluated for differential range
shifts among insectivorous species and Neotropical migrant species.

5

STUDY AREA
The study area included the entire state of Michigan, USA, (147,121 km2) divided into 3 regions:
the Upper Peninsula (UP), southern Lower Peninsula (SLP), and northern Lower Peninsula
(NLP) (Figure 1). The state has a gradual north-south landcover gradient, limited elevation
variation (174 m to 603 m), and broad latitudinal range (41° N to 48° N). The contiguous nature
of forest landcover in northern Michigan is characterized by 3 major vegetation types:
1. Early-succession northern forest (aspen [Populus spp], and paper birch [Betula papyrifera]);
2. Mesic mixed forest (sugar maple [Acer saccharum], birch [Betula spp], eastern hemlock
[Tsuga canadensis], and American beech [Fagus grandifolia]); and
3. Mixed boreal/wet coniferous forest (balsam fir [Abies balsamea], spruce [Picea spp],
tamarack [Larix laricina], and white cedar [Thuja occidentalis]).
Southern Michigan is dominated by a fragmented agriculture-forest matrix, with 3 major
vegetation types that include:
1. Open land (farms, open wetlands, barrens, and other treeless areas);
2. Wet deciduous forest (maple [Acer spp], and ash [Fraxinus spp]); and
3. Dry deciduous forest (oak [Quercus spp]).

6

Upper
Peninsula (UP)

Northern Lower
Peninsula (NLP)

Southern Lower
Peninsula (SLP)

Figure 1. Michigan Breeding Bird Atlas regions and township delineations, 1983 to 2008.

Statewide climate is characterized by cold winters, temperate summers, and precipitation
throughout the year. Annual precipitation rates have increased 10% to 15% over the period 1930
to 2010. For the period 1850 to 2010, mean temperature increased approximately 0.8º C
(Christensen et al. 2007; Andresen et al. 2012). This increase is consistent with general global
trends. Although regional temperatures vary considerably between years that temperature shifts
have occurred, temperatures have changed much faster in recent decades. Mean average
temperatures increased approximately 0.059º C per decade during the past century, 0.12º C per
decade since 1950, and 0.26º C per decade since 1979, much of it concentrated during winter
7

months and at night (Brohan et al. 2006; Andresen et al. 2012) (Figure 2). In recent decades,
more mild winter temperatures have led to less ice cover on the Great Lakes, and seasonal spring
warm-up has occurred earlier than previously (Andresen 2009). Over the past 50 years, regional
growing seasons have advanced 1 to 1.5 days, and in coming decades, vegetation hardiness
zones are expected to shift northward, resulting in conditions similar to those currently found in
northern Alabama by 2100 (Schwartz et al. 2006; Hellman et al. 2010).

A.
Mean Temperature ( C) 1980 - 2010
Mean Temp Trend
0.04 C/yr
0.03
0.02
0.01
0.00
-0.01
-0.02
-0.03
-0.04

Figure 2. From 1980 to 2010 in Michigan, trend analyses for (A) the average rate of change in
the annual mean temperature (°C), (B) the average rate of change in spring precipitation (mm)
during March through May, and (C) the rate of change in maximum spring temperature (°C)
during March through May. Maps produced by ClimateWizard©, University of Washington, and
The Nature Conservancy, 2013. Base climate data from the PRISM Group, Oregon State
University. For interpretation of the references to color in this and all other figures, the reader is
referred to the electronic version of this thesis.

8

Figure 2 (cont’d)
B.
Precipitation 1980 - 2010
Precipitation Trend
0.50 mm/yr

0.25

0.00

-0.25

-0.50

C.
Maximum Temperature ( C) 1980 - 2010
Max Temp Trend

0.04 C/yr
0.03
0.02
0.01
0.00
-0.01
-0.02
-0.03
-0.04

9

METHODS
Breeding Bird Atlas Data
Long-term census data in the Michigan Breeding Bird Atlas documented the spatial occurrence
of breeding birds across the state (Brewer et al. 1991; MDNR 2012a). Atlas I resulted from 6
years of fieldwork (1983 to 1988) by approximately 1,300 field observers and an estimated
100,000 observer hours statewide. Atlas I produced a database with over 506,000 records with
breeding evidence for 233 species. Atlas II (2001 to 2008) involved approximately 1,600 field
observers, and data comparable to Atlas I, with over 501,000 records and breeding evidence for
238 species. Insufficient effort data exists for Atlas II, precluding reliable effort comparison
between the atlas surveys. In Atlas I and Atlas II, surveys utilized a grid system of townships
(9.66 km X 9.66 km) and one-quarter townships identified as atlas blocks (4.83 km X 4.83 km)
(Brewer et al. 1991; MDNR 2012a). The Michigan land survey grid system utilized for the BBA
represented 1,896 townships and 7,080 blocks; 6,115 blocks were surveyed in Atlas I and 5,795
blocks were surveyed in Atlas II. Each atlas block centroid was associated with spatial bounding
coordinates in standard geographic longitude and latitude decimal degree values in the Michigan
GeoRef (MDNR 2012b).
Sampling intensity was based on identifying between 1 to 4 priority blocks for surveying
in each township; a similar approach was applied in both atlas surveys. All 2,690 blocks in the
SLP were assigned priority status. Sampling intensity was lower in the NLP and UP, where
cover types were homogenous over large areas, and observer access was more limited. Two
blocks in each township unit were randomly chosen in the NLP for 1,097 total priority blocks;
one block in each township unit was randomly chosen in the UP for 514 total priority blocks.
After surveys were completed in the identified priority block for a given township, observers

10

could elect to survey in neighboring blocks. Survey goals for blocks were based on standards
established by the First Northeastern Breeding Bird Atlas Conference (Laughlin et al. 1982;
Brewer et al. 1991; MDNR 2012a). Minimum survey goals per block included 50 breeding
species and 10 to 20 hours of effort. Past research has shown that counts of 50 to 75 species
represent > 75% of the breeding birds present in most blocks, a normally accepted level of
survey effort in North America (Robbins and Geissler, 1990). The presence of unique species
was determined at the block level using visual and auditory surveys of breeding bird evidence
(MDNR 2012a). Field observers recorded species observations on a classification gradient of
increasing support for breeding behavior. The lowest class of breeding evidence was the
observation of a bird during the breeding season, and depending on evidence increased to
possible, probable, or confirmed, with each class including levels of support to identify breeding
behavior (Table B1). Regional and project coordinators reviewed, edited, and verified data
before including records in the project database (Brewer et al. 1991). The Michigan Department
of Natural Resources (MDNR), Wildlife Division, subsequently reviewed all data to ensure
accuracy in the final BBA project (MDNR 2012b).

Range Shifts among Northerly and Southerly Species Groups
To accomplish the first objective of testing for range shifts in Michigan breeding birds, I
analyzed BBA data for measurable changes in distribution. In my analyses, I evaluated 3
groupings representing all species, northerly distributed species, and southerly distributed species
(Table C1). For each species, I measured changes in the center of statewide occurrence,
statewide range boundaries, and localized colonization and extinction between atlas surveys

11

(Figure 3). The results from each of these steps informed further analysis of range dynamics
among foraging and migratory groupings.

Figure 3. Theoretical northward shifts from Atlas I (1983-1998) to Atlas II (2001-2008) in (a)
the southern range boundary of northerly species, and (b) the northern range boundary of
southerly species. The dashed line represents the shifted range boundary location in Atlas II.

Species selection and classification–The first step in evaluating range dynamics was selecting
species using criteria that helps account for potential bias in the data (Thomas and Lennon 1999,
Zuckerberg et al. 2009b). I selected species that had previously been shown to demonstrate
poleward range shifts in New York State (Zuckerberg et al. 2009b), and occurred in Michigan
during the breeding season. I eliminated uncommon species (occurrence in less than 37 atlas

12

blocks), because they represented endangered species or species with restricted detection
probability, to minimize bias from false absences in the data set. I excluded ubiquitous species
(occurrence in more than 3775 atlas blocks), because they did not exhibit a distribution with a
clear statewide range boundary. Hybrid, introduced, and game species were excluded because
their population dynamics are directly influenced by intentional human involvement. Eruptive
species with irregular seasonal or annual movements were also excluded.
Species with identifiable statewide range boundaries were classified according to
northerly or southerly breeding range distribution. To classify each species, I first delineated a
latitudinal boundary 100 km north and south of Michigan’s boundary to define the study area
extent (Zuckerberg et al. 2009b). A boundary extending beyond the state accommodated
discontinuous geographic ranges that confound precise range margin identification (Reif 2010). I
then used continental species accounts and avian range maps from Birds of North America
(BNA) (Poole 2005) to classify high-latitude temperate species as northerly if the southern range
boundary was within Michigan, and low-latitude temperate species as southerly if the northern
boundary range was within Michigan.
I further classified individual species by life history strategies (DeGraaf et al. 1985; Sauer
et al. 1999; DeGraaf and Yamasaki 2001; Poole 2005; BirdLife 2012) and reviewed these with
avian biologists (K. Cleveland, Michigan Department of Natural Resources, personal
communication; J. Owen, Michigan State University, personal communication). Each northerly
and southerly species was assigned a breeding season foraging guild (insectivore, carnivore,
piscivore, or omnivore), migratory strategy (Neotropical, short-distance, or resident), and
breeding habitat guild (grassland, scrub/shrub, wetland, and wooded). I assigned foraging guilds
by primary behavior during the breeding season and migratory strategies by behavior specific to

13

Michigan. Classification steps resulted in a final set of 41 species. These species displayed life
history traits among 3 migratory strategies, 4 foraging guilds, and 5 breeding habitats (Table 1).

Table 1. Statewide distribution and life history traits of 41 Michigan birds, 1983 to 2008.
a
d
Species
Scientific Name
Dist. Foraging Migratory Habitat
b
c
Guild
Strategy
Acadian Flycatcher
Insect
Neo
Wood
Empidonax virescens S
Blue-gray Gnatcatcher
S
Insect
Neo
Wood
Polioptila caerulea
Blue-headed Vireo
N
Insect
Neo
Wood
Vireo solitaries
Blackburnian Warbler
N
Insect
Neo
Wood
Dendroica fusca
Boreal Chickadee
N
Omni
Short
Wood
Poecile hudsonica
Brown Creeper
N
Insect
Short
Wood
Certhia americana
Black-throated Blue
N
Insect
Neo
Wood
Dendroica
Warbler
caerulescens
Black-throated Green
N
Insect
Neo
Wood
Dendroica virens
Warbler
Blue-winged Warbler
S
Insect
Neo
Scrub
Vermivora pinus
Carolina Wren
S
Insect
Res
Scrub
Thryothorus
ludovicianus
Canada Warbler
N
Insect
Neo
Wood
Wilsonia canadensis
Common Loon
N
Pisc
Short
Wet
Gavia immer
Chestnut-sided Warbler
N
Insect
Neo
Scrub
Dendroica
pensylvanica
Eastern Screech-Owl
S
Omni
Res
Wood
Megascops asio
Eastern Towhee
S
Omni
Short
Scrub
Pipilo
erythrophthalmus
Field Sparrow
S
Omni
Short
Scrub
Spizella pusilla
Gray Jay
N
Omni
Res
Wood
Perisoreus
canadensis
Great Egret
S
Carn
Neo
Wet
Ardea alba
Green Heron
S
Pisc
Neo
Wet
Butorides virescens
Grasshopper Sparrow
S
Insect
Neo
Grass
Ammodramus
savannarum
a
Statewide distribution: northerly (N) and southerly (S)
b
Breeding season foraging guilds: carnivore (Carn), insectivore (Insect), omnivore
(Omni), and piscivore (Pisc)
c
Migratory strategies: resident species (Res), short-distance migrants (Short), and longdistance Neotropical migrants (Neo)
d
Breeding habitats: grassland (Grass), scrub/shrub (Scrub), wetland (Wet), and wooded
(Wood)
14

Table 1 (cont’d)
Henslow’s Sparrow

Ammodramus
henslowii
Hermit Thrush
Catharus guttatus
Hooded Warbler
Wilsonia citrine
Least Bittern
Ixobrychus exilis
Lincoln’s Sparrow
Melospiza lincolnii
Louisiana Waterthrush
Seiurus motacilla
Magnolia Warbler
Dendroica magnolia
Nashville Warbler
Vermivora ruficapilla
Northern Cardinal
Cardinalis cardinalis
Northern Waterthrush
Seiurus
noveboracensis
Orchard Oriole
Icterus spurius
Olive-sided Flycatcher
Contopus cooperi
Red-bellied Woodpecker Melanerpes carolinus
Ruby-crowned Kinglet
Regulus calendula
Red-shouldered Hawk
Buteo lineatus
Swainson’s Thrush
Catharus ustulatus
Tufted Titmouse
Baeolophus bicolor
Willow Flycatcher
Empidonax traillii
Yellow-bellied
Empidonax
Flycatcher
flaviventris
Yellow-bellied Sapsucker Sphyrapicus varius
Yellow-throated Vireo
Vireo flavifrons

S

Insect

Short

Grass

N
S
S
N
S
N
N
S
N

Insect
Insect
Pisc
Omni
Insect
Insect
Insect
Omni
Insect

Short
Neo
Neo
Neo
Neo
Neo
Neo
Res
Neo

Wood
Wood
Wet
Scrub
Wood
Wood
Scrub
Scrub
Wood

S
N
S
N
S
N
S
S
N

Insect
Insect
Omni
Insect
Carn
Omni
Insect
Insect
Insect

Neo
Neo
Res
Short
Short
Neo
Res
Neo
Neo

Gen
Wood
Wood
Wood
Wood
Wood
Wood
Scrub
Wood

N
S

Omni
Insect

Short
Neo

Wood
Wood

To analyze dynamics of range centroids and boundaries, I generated observation histories
for each species for all blocks where they were observed. It was necessary to take this extra step
prior to analysis because the structure of the Atlas II dataset differed from the data structure in
Atlas I. Using Atlas II observation data (for each species, in all blocks, and each atlas year) I
selected records with the highest breeding evidence; this approach paralleled Atlas I records and
eliminated additional observations for lower breeding evidence that were included in the larger
Atlas II data set. I then amended the observation histories by identifying blocks with no records,
the highest breeding evidence per block by year, unique blocks with records in each atlas, and

15

unique blocks without records for each atlas (Table A1). I tested the final observation histories
for accurate representation of data class and format.
Center of statewide occurrence–To test for range changes I evaluated observation histories
for each northerly and southerly species, in all atlas blocks, for all years in both atlas surveys. I
used only the single greatest breeding observation records for a species in each block and each
year. From the unweighted mean of block centroids with observation records, I found the center
of statewide occurrence (by longitude and latitude) for each species in Atlas I and Atlas II. This
identified the center of statewide distributions based on the mean location of observed
occurrences. I then calculated the difference in center of occurrence between Atlas I and Atlas II.
I used an information-theoretic approach as an alternative to means testing to evaluate if
including a spatial variable was meaningful when calculating changes in mean latitude and
longitude. This approach quantified the strength of support for each of the hypotheses evaluated,
and allowed for comparison of results with earlier research (Anderson 2008; Zuckerberg et al.
2009b). The alternative hypothesis for northerly and southerly species assumed a change in the
mean latitude between Atlas I and II and a nonzero effect size. I calculated the second order
Akaike’s Information Criterion statistic (AICc) to compare support for models under the null and
alternative hypotheses, following Anderson (2008):

{

}

(where n = sample size, RSS = residual sum of squares, K = number of parameters). The null
hypothesis in the paired design does not have a term for differences between observations and
consequently has an effect size of zero (K = 1). RSS for the null hypothesis was calculated as:
16

∑

(where di = difference between the 2 sample distributions of mean observations). Assuming a
nonzero effect size, the value for the alternative hypothesis was calculated as:

̅)

∑(

(with a term for an assumed difference between the two sample distributions, K = 2). The
alternative hypothesis quantified model support for distinct latitudinal shifts between the two
sample observations; the null hypothesis calculated the probability of no changes in latitude
between observations (Zuckerberg et al. 2009b). I reported the probability of the null and
alternative hypotheses, effect size, RSS values, the number of parameters (K), AICc, ΔAICc, and
weight of evidence in support of the hypotheses (Anderson 2008) for northerly, southerly, and
ungrouped species between both atlas surveys.
Range boundary shifts–The range boundary analysis quantified shifts in northern range
boundaries of southerly species, and shifts in southern range boundaries of northerly species.
Zuckerberg et al. (2009b) explored whether or not their boundary analysis was sensitive to
increasing the number of blocks used in the calculation (ranging from 10 to 50); for northerly
species, recalculations had little effect and showed no support for the null model, and among
southerly species there was a small increase of support in model probabilities for a true intercept
17

value. They further suggested that using too many blocks for calculating statewide range
boundaries may not actually represent range boundaries, and diminished the signal of the yintercept. In my analysis, (having previously ruled out species with occurrence in less than 37
atlas blocks), I used data from 10 and 25 blocks for each species in the range boundary analysis.
I evaluated the n-most southern blocks for northerly species, and the n-most northern blocks for
southerly species. The mean statewide latitude and latitudinal range limits were calculated as the
mean, minimum, and maximum block centroid location for each species (Anderson et al. 2009).
The next step in evaluating for range shifts was comparison of range boundary shifts between
Atlas I and II against overall changes in occupancy between atlas periods. Following Zuckerberg
et al. (2009b) and Thomas and Lennon (1999), I calculated and plotted the change in occupancy
for each species by statewide distribution,

log10[occupied blocks Atlas II] – log10[occupied blocks Atlas I],

against the change in northern and southern range boundaries between Atlas I and Atlas II using
the ordinary least squares (OLS) approach. Using a regression model where parameters that can
take any real number, and a model with a forced zero y-intercept value, I evaluated the AICc for
proportional fit. I found the evidence ratio, which indicated the more probable model, using the
likelihood of each model, given the data, and the overall model likelihood. Support for northern
range shifts was demonstrated by greater evidence for the regression model and significantly
positive y-intercept values, (Zuckerberg et al. 2009b), while controlling for changes in overall
geographic distribution by incorporating the change in occupancy.

18

States of occurrence–To evaluate for changes in states of occurrence, I identified areas of
atlas block gains and losses for northerly and southerly species between both atlas surveys
(Thomas and Lennon 1999; Brommer 2004; Zuckerberg et al. 2009b, Bradbury 2011).
Controlling for overall changes demonstrated by expanding or contracting distribution helped
account for the variable nature of ranges, and allowed for stronger inference on the center of
statewide occurrence and range boundary analyses. Species with expanding distributions are
more likely to colonize towards the range margins and beyond, and those with contracting
distributions are likely to move towards distributional centers (Brommer 2004). In the states of
occurrence analysis, presence or absence was determined by observation records, (as opposed to
presence as it relates to occurrence probability specific to other types of analysis), for each
species in Atlas I and Atlas II. The absence or presence of species was assigned to each block in
Atlas I and Atlas II. Each atlas block was also identified by the spatial bounding coordinates of
the block centroid. Then, I quantified losses as the mean latitude and longitude of atlas blocks
with species occurrence in Atlas I but not Atlas II. Gains were the mean latitude and longitude in
atlas blocks with no detection in Atlas I and detection in Atlas II. Retention was identified by
atlas blocks with occurrence in both Atlas I and II, and absence represented blocks with no
occurrence (Zuckerberg et al. 2009b). I then incorporated distribution changes as the change in
occupancy between Atlas I and Atlas II (previously calculated in the range boundary analysis).
Values from the change in occupancy that were distributed around zero suggested no overall
changes in distribution, while directional changes in distribution were suggested as values moved
toward 1.0. For expanding distributions, I would expect to see positive values toward 1.0, and for
contracting distributions, negative values are expected toward -1.0 (Brommer 2004). By
conducting a regression of range margin changes on distribution changes, I also quantified the

19

expected changes in range margins without overall changes in distribution (Thomas and Lennon
1999; Brommer 2004).

Differential Range Shifts among Foraging and Migratory Groups
To accomplish the second objective to evaluate species groups for differential changes as a way
of exploring the underlying ecology, I further analyzed species that demonstrated range shifts.
First, I evaluated species organized into foraging groups as either insectivorous or noninsectivorous, and then evaluated species in migratory groups as either Neotropical or nonNeotropical), (Table C2 and Table C3). I followed the same analysis as above for the center of
occurrence, range boundary, and states of occurrence for foraging groups then migratory groups.
In the evaluation of changes in the center of statewide distributions, the alternative hypothesis for
the insectivorous guild (and subsequently Neotropical migrants) among the other groupings
supported a change in the mean latitude between the observations and a nonzero effect size. The
null hypothesis suggested no differences between observations and an effect size of zero. For
migratory groups and foraging groups, the range boundary analysis quantified shifts in northern
range boundaries of southerly species, and shifts in southern range boundaries of northerly
species. Finally, the states of occurrence analysis identified areas of atlas block gains and losses
for species in each group between Atlas I and II (Thomas and Lennon 1999; Brommer 2004;
Zuckerberg et al. 2009b; Bradbury 2011). I conducted all analyses in R 2.13.1 (R Development
Core Team, R; A language and environment for statistical computing, Vienna, Austria, R
Foundation for Statistical Computing, 2006).
Animal use and care exemption–The Michigan State University Institutional Animal Care and
Use Committee (IACUC) exempted this research from review, effective 28 December 2010.

20

RESULTS
Center of Occurrence
Shifts in the center of statewide occurrence ranged from a southward movement of 81.7 km for
the yellow-bellied flycatcher (Empidonax flaviventris) to a northward movement of 52.0 km for
the Henslow’s sparrow (Ammodramus henslowii). Results provided greater support for the
hypothesis for a change in the center of statewide occurrence between Atlas I and Atlas
|

II

(Table 2). The overall model showed changes in latitude, as

demonstrated by a large evidence ratio in favor of the alternative hypothesis (156.1), and ΔAICc
value of 10.1. A difference greater than 3.0 in the ΔAICc value further suggested a change in the
overall model for changes in mean latitude.

Table 2. Center of statewide occurrence analysis quantifying changes in mean latitude for 41
birds. Michigan Breeding Bird Atlas I and II. 1983 to 2008.

Null
a

L(h|data)

Prob(h|data)

Evidence Ratio

3.73E+10 2

0.0000

1.0000

0.9936

156.0915

10.1009

0.0064

0.0064

RSS

Alternative
b

Δ AICc

5.04E+10 1

Model
a

K

Alternative hypothesis quantifies support for latitudinal shifts between sample observations
Null hypothesis quantifies support for no significant changes in latitude between observations

b

Compared with all species, the southerly species group demonstrated similar shifts in
their statewide occurrence. However, a greater percentage of southerly species shifted polewards
than overall, and shifts were in the opposite direction than overall; while 39.0% of all species
demonstrated northward movement in their center of occurrence, I found that 38.1% of 21
southerly species showed southward movement, and 61.9% of southerly species showed
northward movement. Among northerly species, I found greater differences in the magnitude of
21

species that demonstrated shifts in the center of their statewide occurrence. For 20 northerly
species, only 15% showed a northerly shift in center of occurrence compared with 39.0% of all
species with northward movement. Eighty-five percent of northerly species demonstrated a
southerly shift, while 61.0% of all species had a southerly shift.
Among changes in center of occurrence for foraging guilds, I found greater support for a
shift in 26 insectivorous species

|

, relative to the lack of support for a

best model for the 15 non-insectivorous species

|

(Table 3). The

insectivore model indicated changes in latitude, as demonstrated by a ΔAICc value of 8.5 and
large evidence ratio (69.9) in favor of the alternative hypothesis for distinct shifts between
observations. Shifts in the center of statewide occurrence for the insectivore guild ranged from
southward movements of 81.7 km for the yellow-bellied flycatcher and 79.9 km for the brown
creeper (Certhia americana), to northward movements of 52.0 km for the Henslow’s sparrow
and 11.0 km for the willow flycatcher (Empidonax traillii). I found that 70.8% of insectivorous
species demonstrated northward movement in their center of occurrence, and 53.3% of noninsectivorous species shifted northward. Among changes in center of occurrence for migratory
groups, I found greater support for a shift in 25 Neotropical species

|

,

relative to the lack of support for a best model for the 16 non-Neotropical species
|

22

Table 3. Center of statewide occurrence analysis quantifying changes in mean latitude by life
history strategies. Michigan Breeding Bird Atlas I and II. 1983 to 2008.
Group
Insectivore

Model
Alt

a

Null
Non-insectivore Null
Alt
Alt
Neotropical
Null
Non-neotropical Null
Alt
a

RSS

K Δ AICc L(h|data)

Prob(h|data)

Evid. Ratio
69.882

2.69E+10 2 0.000
b

1

0.986

4.09E+10
9.52+10
8.39+10
1.91+10
3.19+10
1.9+10
1.7+10

0.014
1
0.670
1
0.006
1
0.597

0.014
0.599
0.401
0.99
0.006
0.626
0.374

1
1
2
2
1
1
2

8.494
0
0.802
0
10.368
0
1.032

1.494
178.361
1.676

Alternative hypothesis quantifies support for latitudinal shifts between sample observations

b

Null hypothesis quantifies support for no significant changes in latitude between observations

Range Boundary Analysis
In the analysis on range boundary changes and overall changes in occupancy, I found stronger
support for the hypothesis that birds are shifting their ranges northwards. Results from the range
boundary analysis suggest a change in the mean latitude between observations and a nonzero
effect size, and the importance of including a spatial component in the evaluation. I found
strongest model support for northward changes in range boundaries for the southerly species
group, while accounting for overall changes in occurrence (Table 4). Twenty-eight species
(68.3%) demonstrated a northward shift their range boundary (mean 41.7 km), and 13 species
(31.7%) had southward range boundary shifts (mean 25.2 km) (Table D1).

23

Table 4. Range boundary analysis for change in mean latitude between Michigan Breeding Bird
Atlas I and II, 1983-2008.
Group
Model
K
weight
Δ AICc
Alternative
3
0.0
0.567
Northerly species
Null
2
0.5
0.433
Alternative
3
0
0.738
Southerly species
Null
2
2.1
0.262
Alternative
3
0
0.721
All species
Null
2
1.9
0.279

For all 41 species I estimated a northward range boundary shift from a positive yintercept value of 13.2 km (SE = 6.41 km, P = 0.04) (Table 5). Species that demonstrated
positive changes in occupancy generally shifted their range boundaries northward (Figure 4c).
Results that suggested a northward shift has occurred using an OLS approach were corroborated
by an Information-Theoretic approach of comparing two competing models for all species while
controlling for overall changes in states of occurrence. Although I found that the null model
|

remains a possible model given the data
model for poleward range shifts is more than twice as likely
4).

24

, the alternative
|

(Table

60
40
20
0
-20
-40

Change in Southern Range Boundary (km)

A

-0.1

0.0

0.1

0.2

0.3

Change in Occupancy

Figure 4. Latitudinal change in range boundaries (km) plotted against the change in occupancy
for (A) 20 northerly, (B) 21 southerly, (C) all 41, D) 26 insectivore, and E) 25 Neotropical
species in Michigan between Atlas I (1983-1998) and Atlas II (2001-2008). Northerly species
demonstrated a northward shift in southern range boundaries of 10.4 km (SE = 5.8 km) (P =
0.09). Southerly species demonstrated a northward shift in northern range boundaries of 22.7 km
(SE = 10.3 km) (P = 0.04). All species demonstrated a northward shift in range boundaries of
13.2 km (SE = 6.4 km) (P = 0.05). Insectivore species demonstrated a northward range shift in
range boundaries of 6.8 km (SE = 8.0 km) (P = 0.41). Neotropical species demonstrated a
northward range shift in range boundaries of 16.0 km (SE = 8.7 km) (P = 0.08).

25

Figure 4 (cont’d)

150
100
50
0
-50
-100

Change in Northern Range Boundary (km)

B

-0.2

0.0

0.2
0.4
0.6
0.8
Change in Occupancy

1.0

-0.2

0.0

0.2
0.4
0.6
0.8
Change in Occupancy

1.0

100
50
0
-50
-100

Change in Range Boundaries (km)

150

C

26

0

50

100
-0.2

-50

Change in Neotropical Range Boundaries (km)
-100

-50

0

50

100

150

Change in Insectivore Range Boundaries (km)

Figure 4 (cont’d)
D

E
0.0
0.2
0.4
0.6
0.8
Change in Occupancy
1.0

-0.2

-0.1

0.0
0.1
0.2
Change in Occupancy

0.3

27

For 20 northerly species in the range boundary analysis, 75% expressed southern range
boundary shifts northward, while 25% showed southern range boundary shifts southward. When
northerly species had positive occupancy change values, 94% demonstrated southern range
boundary shifts northward (mean 19.4 km) (Figure 4a). When species had negative occupancy
change values, they tended to have southern range boundary shifts northward (mean 16.3 km). I
estimated a northward range boundary shift from a positive y-intercept value of 10.4 km (SE =
5.8 km, P = 0.08) (Table 5). However, I did not find strong evidence supporting probabilities for
either hypothesis when comparing the two competing models for northerly species
|

|

and

(Table 4).

Table 5. Estimating shift distances from the ordinary least squares regression in the range
boundary analysis, Michigan Breeding Bird Atlas, 1983 to 2008.
Group

Shift distance (km)

Std. Error (km)

Pr(>|t|)

Northerly Species

10.41

5.76

0.088

Southerly Species

22.67

10.25

0.039

All Species

13.19

6.41

0.046

For 21 southerly species in the range boundary analysis, 61.9% expressed northern range
boundary shifts northward (mean 68.0 km), while 38.1% showed northern range boundary shifts
southward (mean 30.4 km). When southerly species had positive occupancy change values, most
demonstrated northern range boundary shifts northward (mean 74.9 km) (Figure 4b). If species
demonstrated a negative occupancy change value, they tended to have northern range boundary
shifts southward (mean 33.9 km). I estimated a northward range boundary shift from a positive
y-intercept value of 22.7 km (SE = 10.3 km, P = 0.04) (Table 5). Model probabilities support the
28

alternative hypothesis for northward range shifts while controlling for overall changes in
occupancy

|

|

and

(Table 4).

Results from evaluating foraging guilds in the range boundary analysis indicated that for
26 insectivore species 73.1% showed northward shifts in their northern or southern range
boundaries (mean 36.8 km). While this was less than changes seen among all non-insectivore
species, of which 60% showed northward range boundary shifts (mean 51.8 km), more species in
the insectivore guild demonstrated changes and had greater shifts than other individual guilds.
When insectivore species demonstrated positive changes in occupancy, they generally shifted
their range boundaries northward (Figure 4d). Fourteen northerly insectivore species
demonstrated a mean northward shift in southerly range boundaries of 33.1 km. Among 12
southerly insectivore species, I found a mean northward shift of 35.8 km in northerly range
boundaries. Among Neotropical species, those with positive changes in occupancy tended to
shift their range boundaries northward (Figure 4e).

States of Occurrence
The states of occurrence analysis controlled for overall changes in distribution in the analyses for
center of statewide occurrence and range boundary results. In the states of occurrence results for
change in occupancy (Table D1), negative values indicated contracting distributions, and
positive values indicated expanding distributions. Values distributed around zero indicated little
to no changes in distribution and provided support toward the hypothesis for range shifts (Figure
5). After controlling for overall changes in distribution in the range boundary analysis, stronger
evidence for the regression model indicated ranges have shifted poleward. Results from the
evaluation of block status between Atlas I and II suggested consistency with changes in center of

29

occurrence and range boundaries, indicating range shifts have occurred; this consistency in

0.0
-1.0

-0.5

Occupancy Change

0.5

1.0

results lends further support for overall range shifts among diverse species.

All Species (n = 41)

Figure 5. In the states of occurrence analysis, values for the change in occupancy distributed
around zero suggest no overall changes in distribution. The change in occupancy is represented
for all species. Expanding distributions have positive values toward 1.0, and contracting
distributions have negative values moving to -1.0.

30

States of occurrence results for northerly species suggested northward trends in
movement; at the most southerly latitudes in Michigan, species were absent from blocks, and as
latitude increased, species exhibited block loss, block gain, and continued block presence, the
expected pattern as species shift northward (Figure 6a-c). I found a similar pattern in the states of
occurrence results among southerly species, which demonstrated a northward movement trend.
Here, species remained present at the most southerly latitudes, and as latitude increased,
demonstrated block loss, block gain, and remained absent at northerly latitudes. Similar to the
results for northerly species, block loss and block gain occurred in the same pattern at the center
of latitudinal occurrence. I found a third northward trend in the states of occurrence for both
Neotropical and insectivore species groups, which is consistent with the other range
characteristics indicating range shifts have occurred. Towards southerly latitudes, Neotropical
and insectivorous species demonstrated block losses; as latitude became more northerly, species
remained absent, experienced block gains, and then remained present.

31

A
Atlas Block Gains for Northerly Species

[1, 4.4]
(4.4, 7.8]
(7.8, 11.2]
(11.2, 14.6]
(14.6, 18]

Figure 6a-c. Atlas block gains and losses for northerly and southerly species in the Michigan
Breeding Bird Atlas I and II, 1983 to 2008. Each data point represents the number of species in
each block that demonstrated a gain or loss. Gains represent blocks with no detection in Atlas I
and detection in Atlas II. Losses represent blocks with species occurrence in Atlas I but not Atlas
II.

32

Figure 6 (cont’d)
B
Atlas Block Losses for Northerly Species

[1, 4.4]
(4.4, 7.8]
(7.8, 11.2]
(11.2, 14.6]
(14.6, 18]

33

Figure 6 (cont’d)
C
Atlas Block Losses for Southerly Species

[1, 4.4]
(4.4, 7.8]
(7.8, 11.2]
(11.2, 14.6]
(14.6, 18]

34

DISCUSSION
My research may be among the first empirical studies in the Midwestern United States, and one
of a small number in North America, to quantify systematic range shifts for a diverse taxonomic
group at large scales. I found northward range shifts for the majority of species, with various
breeding range distributions, foraging strategies, and migratory behaviors, (Figure 5, Table 2,
Table 4, and Table 5). While the majority of species demonstrated range shifts, there was
stronger support for northward shifts in the expanding ranges of southerly species, when
compared with northerly species. My results also suggest that a nuanced pattern may exist
among life history strategies within the larger context of range shifts.
I found consistent patterns within the results for the center of occurrence, range
boundary, and states of occurrence, which indicate the majority of species have shifted their
ranges polewards. After accounting for changes in occupancy, support for range shifts is
demonstrated in results showing a similar magnitude and direction between poleward shifts in
the center of occurrence, poleward shifts in range boundaries, and little change in states of
occurrence between Atlas I and Atlas II. I found that when the ranges of southerly species were
expanding, there were northward shifts in northern range boundaries, no change or a southward
shift in the center of occurrence, and increased occurrence between Atlas I and Atlas II. When
southerly species experienced range contractions, there were southward shifts in northern range
boundaries, no change or a southward shift in the center of occurrence, and decreased occurrence
between Atlas I and Atlas II. If the ranges of northerly species expanded, I found southward
shifts in southern range boundaries, no change or a southward shift in the center of occurrence,
and increased occurrence between Atlas I and Atlas II. When northerly species experienced
range contractions, there were northward shifts in southern range boundaries, no change or a

35

northward shift in the center of occurrence, and decreased occurrence between Atlas I and Atlas
II.
My findings elucidate trends in range shifts for Michigan, New York, and Ohio that
appear to be unique to North America. These findings contribute to a growing body of evidence
for climate-driven range shifts that represents heterogeneity in observable changes and regionspecific factors that influence long-term range trends (Chen et al. 2011; Tingley et al. 2012).
European studies using coarse-scaled data on a limited number of species found northward shifts
in northern range boundaries only (Thomas and Lennon 1999; Lemoine et al. 2007; Reif et
2010). In North America, Hitch and Leberg (2007) found evidence for northerly shifts in
northern boundaries only, at an estimated rate of 2.35 km per year. A 2007 study that evaluated
Christmas Bird Count data in the U.S. found similar northerly range shifts in northern boundaries
only, at an estimated rate of 1.5 km per year (La Sorte and Thompson 2007). A recent metaanalysis across taxonomic groups and continents compared range shifts, and estimated that
distributions were shifting to higher latitudes two to three times faster than previously reported,
at a median rate of 16.9 km per decade (Chen et al. 2011). The same study also suggests that
there is great variance in the rates of change for individual species, and that on average, about
one-quarter of species groups shift in the opposite direction than predicted; these findings are
consistent with those I identified for avian species in Michigan.
Of the few large-scale studies in North America that have quantified systematic range
shifts for a diverse taxonomic group, two have analyzed breeding bird data for a similar group of
species during a comparable period. Zuckerberg et al. (2009b) analyzed 129 species using the
New York State Breeding Bird Atlas (1980-1985 and 2000-2005). They found that birds
demonstrated northerly range shifts, with greater northerly shifts in southern range boundaries,

36

estimated at 11.4 km per year. Batdorf (2012) analyzed up to 67 species from Ohio Breeding
Bird Atlas projects conducted during 1982-1987, and 2006-2011. Results indicated that birds in
Ohio were demonstrating northerly shifts in northern boundaries, and stronger evidence for
southerly shifts in southern boundaries. When evaluating my results along with those identified
for a similar group of species in a different region of their North American breeding ranges, I
found similarities in overall poleward trends for northerly range shifts in southerly species. In
Michigan and Ohio, there was greater support for northerly range shifts in southerly species
compared to northerly species. When comparing Michigan species based on their statewide
distribution group, I also found greater support for northerly range shifts in southerly species.
In the center of occurrence analysis, I found a similar range of distances in southerly and
northerly shifts in the center of occurrence between Michigan and New York. A smaller
southerly shift for northerly species was the only trend in Ohio. While species in New York did
not show systematic trends in the center of occurrence across differing life history traits, I found
greater shifts among insectivorous species in Michigan. Although Zuckerberg et al. (2009b)
found more species overall showed a northerly shift in the center of occurrence than in Michigan,
I found a larger percentage of southerly species showed northerly shifts. Only a small percentage
of the northerly species shifted their center of occurrence northward compared to species in New
York State. In the range boundary analysis, I found that most northerly species expressed
southern range boundary shifts northward, whereas in New York, there was no trend either way
for the majority of northerly species moving northward or southward.
I found a similar distance in the estimated northward shift of southern range boundaries
in Michigan (n = 20, 10.4 km, SE = 5.8) and New York (n = 43, 11.4 km, SE = 3.1); in Ohio
there was a similar distance in the estimated shift for southern range boundaries (n = 68, 11.8

37

km, SE = 6.1), but the shift was southward. Range boundary analysis for southerly species
showed a comparable northward shift of northern boundaries in Michigan (n = 21, 22.7 km, SE =
10.3) and New York (n = 41, 15.9 km, SE = 8.5), and to a lesser magnitude in Ohio (n = 20, 6.8
km SE = 4.5). While Zuckerberg et al. (2009b) found support for a change in the mean latitude
among groups of northerly and southerly species, my results provided support for southerly
species, and to a lesser degree when grouped all together. This may be the result of the difference
in the larger sample size between New York (n = 129), the smaller group of species common to
Michigan (n = 41), and the influence this has in model probability results for the null or
alternative hypotheses. Collectively, the results from Michigan, New York, and Ohio suggest
that northward range shifts have occurred for many species across a large expanse of their
continental ranges. The results also suggest that the northern boundary, or leading edge, may be
more sensitive to climate change than the southern boundary, or trailing edge, of species.
While some degree of variation is expected along different regions in a continental range
boundary, the consistency across Michigan, New York, and Ohio suggests the following:
1. Across diverse species we can observe systemic trends in latitudinal range shifts, despite
various regional factors;
2. Despite a wide range in the magnitude and direction for individual species shifts, consistent
trends emerge in northward range shifts for southerly species, and are suggested for
insectivorous and Neotropical species; and
3. Changes are not just regional or the result of fluid range boundaries, and collectively, allow
for inference on large expanses of range boundaries.
There is a consistent direction and magnitude of shifts in center of occurrence for all species,
and species common only to the three studies across Michigan, New York, and Ohio. For

38

example, among northerly species there was a mean shift southward in changes of the center of
occurrence for all species (9.9 km) and species common in Michigan, New York, and Ohio (11.4
km). In general, all species evaluated versus only those common to Michigan, New York, and
Ohio showed slightly greater values in the magnitude of shifts. When looking at the center of
occurrence for northerly species by northward or southward shifts, the mean southward shifts
were two to three times greater than the northward shifts, and ranged from 21.5 km to 23.4 km.
While overall southerly species had a smaller magnitude in the change of center of occurrence in
each of the three states, they showed a much stronger change when looking at northward or
southward shifts only; southerly species also had greater northward shifts in center of occurrence
than for northward shifts in center of occurrence for northerly species. This suggests southerly
species (at their northern range boundaries), have been changing at a faster rate than northerly
species (at their southern range boundaries). We may then expect to observe southerly species
expand or shift distributions northward faster as populations colonize new areas. This may result
in greater competition for resources as southerly species shift into areas where northerly species
are slower to contract or shift their southern range boundaries.
Insectivorous and Neotropical species had bigger changes in the center of occurrence
than other life history groupings; among northward shifts only, the mean distance for changes in
the center of occurrence was between 8.3 km to 12.2 km, and for southward shifts only, the mean
distance for changes was from 11.6 km to 17.6 km. Most southerly Neotropical species in
Michigan, New York, and Ohio showed a much greater mean northward shift in northern range
boundaries (28.8 km) and at greater distances (70.07 km, blue-gray gnatcatcher [Polioptila
caerulea]) than mean southward shifts in northern boundaries (5.5 km) and distances (17.05 km,
Louisiana waterthrush [Seiurus motacilla]). The majority of southerly species in Michigan, New

39

York, and Ohio showed greater mean northward shifts (45.8 km) in northern boundaries than
southward shifts (4.3 km), and at greater distances northward (157.0 km, Carolina wren
[Thryothorus ludovicianus]) than southward (17.05 km, Louisiana waterthrush). Most northerly
Neotropical species demonstrated larger southern shifts in their southern boundary, and at greater
distances southward (68.5 km in Ohio and 26.6 km in Michigan, blue-headed vireo [Vireo
solitarius]), than northward (13.5 km in New York and 6.8 km in Michigan, chestnut-sided
warbler [Dendroica pensylvanica]).
Although I predicted finding range shifts, I did not expect to find the same level of shifts
relative to other studies. Not only do the Great Lakes play a significant role in shaping regional
climate, their proximity also imposes barriers to species movements. Compared to geographical
constraints in other studies, the broad north to south distance in Michigan was one factor that
may have contributed to the range shifts I observed, even with the constraint of a smaller number
of species analyzed. Michigan also lacks significant elevational variation, a factor that may have
constrained species movement in other studies, and further contributed to the measurable
changes I found in both northerly and southerly range boundaries in Michigan. For most species
with southern range boundaries in Michigan, the core of their continental breeding ranges lays to
the north in Canada. Despite this, I anticipated that the proximity of the Great Lakes would limit
species movement, decreasing measurable shifts. Lake Superior creates a barrier north of the
Upper Peninsula in Michigan, similar to the barriers imposed by Lake Erie to the north of Ohio,
and Lake Ontario north of New York. Despite the barrier imposed by Lake Superior, I was able
to identify contracting range shifts for northerly species, albeit to a lesser degree than shifts in
southerly species. While I found northward shifts in southern range boundaries and consistent
levels of occurrence, there were slightly dissimilar trends in changes of center of occurrence,

40

indicating less support for broad range shifts. In southern Michigan, the Lower Peninsula is
bordered to the east by Lake Michigan and west by Lake Huron. These geographical features do
not constrain poleward movement in a similar manner, and may be contributing factors in
explaining why I found greater shifts in southerly species. Not only have southerly species
demonstrated shifts in their northerly boundaries, they have also expanded their ranges
northward into the state. There were nine new species, each with a historical range south of the
state, that were observed in southern Michigan in Atlas II, further supporting my results for
southerly species shifting northward.
My findings are in line with predictions of greater sensitivity for different foraging and
migratory strategies, and support shifts in the center of occurrence and occupancy among
insectivorous and Neotropical species. These species tend to be specialized in their feeding
niches and resource selection, and may be limited in novel habitats and conditions (Lindell et al.
2007; Pineda-Diez de Bonilla et al. 2012). In addition to specialized resource selection, complex
annual life cycles and endogenous constraints on adaptability further increase sensitivity to
environmental changes among insectivorous and Neotropical migratory species (Visser et al.
1998; Pulido et al. 2001; Cotton 2003; Jonzén et al. 2006). My results in the center of occurrence
analysis are also consistent with predictions of lesser sensitivity among resident, short-distance
migratory, and omnivorous species, which demonstrated a lesser degree of shifts. As a function
of selecting for broader resources, resident, short-distance migrants, and omnivorous species are
better adapted to exploit a range of habitats for resource selection, and may exhibit less of a shift
under changing ecological pressures (Turner et al. 1998; Thomas and Lennon 1999). While there
was greater model support suggested for non-insectivorous and non-Neotropical species in range

41

boundary changes, I found greater support for northward shifts in the center of occurrence and
the magnitude of shifts for insectivorous and Neotropical species.
Pervasive range shifts across diverse species and regions correspond with dramatic
changes in climate, and provide compelling evidence for a correlative relationship between
climate change and range shifts. During the last thirty years, there has been an abrupt escalation
of temperature and precipitation changes in Michigan (Hayoe et al. 2010). Earlier timing of the
last spring freezing date has caused the growing season to increase by seven days in the Midwest
over the same time period. These abrupt changes correspond with the dramatic range shifts that
have occurred during the same time interval, and parallel changes in New York and Ohio.
Increasing average temperatures, greater precipitation rates, and advancing growing seasons
closely influence the timing of insect emergence. We would then expect that species that
maintain a close phenological relationship between the timing of insect emergence and breeding
might be more sensitive to changing environmental conditions. As invertebrate species have
shifted their phenology (Hodgson et al. 2011; Boggs and Inouye 2012; Ellwood et al. 2012), my
results suggest that insectivorous and Neotropical species may have responded to these changes.
The results lending support for northward shifts for insectivorous species suggest that this
foraging strategy may contribute to greater shifts under changing environmental conditions. As
an extension of these trends, results from analyzing differential guild shifts provide a stronger
link between insectivores, range shifts, and climate change, and help to describe specific
ecological variables influencing range shifts. Differential shifts may be intensified by the
interacting effects of climate change and the altered timing in seasonal activities including leaf
out, egg laying, and insect emergence (Walther et al. 2002; Root et al. 2003; Charmantier et al.
2008; Kovacs et al. 2011). As a result, any mismatched phenological and trophic relationships

42

pose further ecological consequences, especially for species less likely to adapt to changes in the
environment (Both and Visser 2001; Both et al. 2006; Stralberg et al. 2009).
While the BBA data analyzed here were not as complete or without error as was hoped
for at the outset, they still represent an exceptionally useful resource that is likely to prove
important to wildlife managers. Citizen science projects, including the BBA, pose challenges
related to obtaining data, methodology, and data quality. Because of its nature as a large citizenscience project, certain aspects of the BBA may lack comprehensive data, such as observer
effort. However, it is important to note that the extensive nature of the BBA provides a unique
opportunity to evaluate for large-scale population trends. Through close collaboration with the
professionals involved in the project and extensive exposure to data and background information,
it is clear that the BBA is a valuable project for resource managers, the public, and research.
Despite limitations or biases that may exist in the BBA data, extensive effort was directed at
ensuring quality and limiting errors wherever possible. By recognizing this and incorporating
important information and caveats within the project metadata, it is possible to utilize the BBA
project as a resource for breeding bird information and ecological trends. It is important that
future BBA projects implement consistent study design to allow for meaningful evaluation
between projects. With this in mind, it is also of the highest importance to implement procedures
to ensure accurate recording of observation data, as well as adequate collection of effort data.
Future atlas projects stand to benefit greatly from increasingly available technology that allows
observers to immediately and accurately record location, effort data, and observations. Future
research using BBA projects may be strengthened by incorporating additional environmental
variables, where available, that help describe the relationship between avian occurrence,
potential sampling bias, coarse climate data, block-level land cover, and land use change over

43

time. Finally, given the increasing number of states with repeated atlas projects, there are
significant opportunities to expand the scale of future research to evaluate for systematic shifts
across large regions of the United States.
In conclusion, I found evidence for systematic northward range shifts in the northern
boundary of southerly species and southern boundary of northerly species. These shifts have
occurred in the breeding ranges of a diverse group of species. The similarity of my findings with
those in New York and Ohio suggests that ecosystem changes in avian communities are
pervasive at least across the Great Lakes region. Among species demonstrating range shifts,
there are trends for greater changes in southerly species expanding north. Finally, consistency in
systematic distribution changes that coincide with dramatic changes in climate suggest a
correlative link between climate change and range shifts.

MANAGEMENT IMPLICATIONS
Faced with various ecological and economic consequences under climate change, my results
provide needed quantification of large-scale changes in the Great Lakes Region. My results
further provide an objective tool for monitoring large-scale ecological change. This is of great
importance because finding objective measures of ecosystem change has been a core issue in the
debate over climate change. Finding results indicating avian species have already shifted their
statewide ranges builds a foundation to inform conservation policy, including habitat
conservation plans, climate change adaption planning, and environmental reviews for managing
state threatened and endangered species. Given the various consequences projected under climate
models, it is important to understand how species have already responded under recent climate
change. Hellman et al. (2010) modeled suitable bird habitat, weather data, and associated tree

44

species importance with global climate models to project positive and negative changes in
habitat and distribution. They found that up to 76 of 147 avian species in the Great Lakes Region
may lose habitat, and among these, up to 47 species will lose half. By evaluating these model
predictions with my results for observable changes that have already occurred, we can identify
species and associated habitats that are particularly vulnerable to habitat loss and climate change.
Results for specific species, breeding habitats, and foraging resources are pertinent to the
Michigan State Forest Management Plan (SFMP). Under the SFMP sections for special resource
areas (Special Conservation Areas, High Conservation Value Areas, and Ecological Reference
Areas), my results may help to identify species and associated habitats that are particularly
vulnerable to ecological changes. As Michigan revisits its State Wildlife Action Plan, and
planning for management of public lands, there will need to be accommodation for shifting
species assemblages and attention paid to the longer-term affects of these changes. Incorporating
species changes demonstrated in my results may provide a baseline for future assessments and
identifying priority species for conservation. Between Atlas I and Atlas II, nine species with
historic ranges south of the state shifted their ranges into southern Michigan. The expansion of
southerly species may be an indicator that Michigan is likely to see expansion into the state by a
larger array of taxa. Current populations may be stable, but as populations shift and new species
create novel species assemblages, managers may need to revisit Vulnerability Assessments, plan
for new species, and update adaptation master plans. The retreat of northerly species suggests
that Michigan may be facing the loss of species that have long been native to the state, which
may further reshape priorities for wildlife management. Management decisions will need to
include accounting for invasive species shifting their ranges into Michigan, disease outbreak,
forest management, decreased native taxa, and vulnerable species in marginal habitats. In setting

45

conservation priorities for native plants, animals, aquatic animals, and natural ecosystems,
Michigan Natural Features Inventory (MNFI) now considers the influence of climate change and
population trends. Incorporating my results for species that are demonstrating northward range
shifts may be useful as MNFI updates their conservation priority rankings. The scope of
management plans extends to Michigan’s participation in the Upper Mississippi River and Great
Lakes Region Joint Venture and implementation plans such as the Landbird Habitat
Conservation Strategy and Waterbird Conservation Strategy. My results help identify species of
conservation opportunity, and areas encompassed by current range shifts or in the direction of
expected shifts. This research also addresses the near-term goal to increase knowledge of the
potential impacts of climate change on priority bird species.

46

APPENDICES

47

APPENDIX A

Supplementary tables for BBA data, species groups used in analyses, and results for individual
species

48

Table A1. Michigan Breeding Bird Atlas I (1983-1988) and Atlas II (2001-2008) observation
records and survey block summaries.
Species Atlas I & Atlas I
Atlas I
Atlas II Atlas I
Atlas II Atlas I
Atlas II
Atlas II
& Atlas blocks by blocks
blocks
blocks
unique
unique
total
II blocks max
by max
with no with no blocks
blocks
records
by max
breeding breeding records
records
with
with
breeding code
code
records
records
a
code
BOBO 4880
4626
2335
2291
4738
5584
2335
1490
BRCR
2027
1875
745
1130
6328
6166
745
907
CSWA 4278
3955
1594
2361
5479
5381
1594
1692
GCKI
1695
1570
562
1008
6511
6303
562
770
NAWA 4837
4229
1393
2836
5680
5359
1393
1714
NOWA 1309
1285
562
723
6511
6461
562
612
PISI
897
868
482
386
6591
6715
482
359
PUFI
3033
2957
1400
1557
5673
5853
1400
1220
WTSP
4948
4452
1667
2785
5406
5408
1667
1665
ACFL
1347
1174
420
754
6653
6583
420
490
BWWA 2109
1849
729
1120
6344
6320
729
753
CERW 567
480
217
263
6856
6894
217
179
HOFI
5999
4519
682
3837
6391
4901
682
2172
NOCA 11355
8353
2804
5549
4269
4387
2804
2686
NOMO 373
354
161
193
6912
6911
161
162
YTVI
3607
3184
1188
1996
5885
5760
1188
1313
Total
53261
45730
16941
28789
96227
94986
16941
18184
a
All references to “blocks by max breeding code” represent the observation records with the
greatest breeding evidence for each species, per block, in each survey year.

49

a

Table B1. Breeding behavior classification codes . Michigan Breeding Bird Atlas. 1983-2008.
b

Category

Code

Observed

O

Possible

PO
#
X

Probable

P
S
T

Confirmed

C
N
A
B
NB
PE
DD
UN
FL

ON
FY
FS
NE
NY
a

Behavior
Species (male or female) observed in a block during its breeding season,
but no evidence of breeding. Not in suitable nesting habitat. Includes a
wide range of species such as vultures or raptors, or a colonial nesting
species not at the nesting colony.
Species (male or female) observed in suitable nesting habitat during its
breeding season.
Species observed in suitable nesting habitat during its breeding season.
Singing male present in suitable nesting habitat during its breeding
season.
Pair observed in suitable habitat during its breeding season.
Singing male present at the same location on at least two dates at least 7
days apart (5 or more) singing males on the same date during the breeding
season.
Permanent territory presumed through defense of territory (chasing
individuals of the same species).
Courtship behavior or copulation.
Visiting probable nest-site.
Agitated behavior or anxiety calls from adult.
Nest building by wrens or excavation of holes by woodpeckers.
Nest building by all except woodpeckers and wrens.
Physiological evidence of breeding (e.g. highly vascularized, edematous
incubation [brood] patch or egg in oviduct based on bird in hand. To be
used by experienced bird banders on local birds during the nesting
season).
Distraction display or injury feigning.
Used nests or eggshells found. Caution: these must be carefully identified,
if they are to be accepted.
Recently fledged young (either precocial or altricial) incapable of
sustained flight, restricted to natal area by dependence on adults or limited
mobility.
Occupied nest: adults entering or leaving a nest site in circumstances
indicating occupied nest. To be used for nests which are too high (e.g. the
tops of trees) or enclosed (e.g. chimneys) for the contents to be seen.
Adults carrying food for young or feeding young.
Adult carrying fecal sac.
Nest with egg(s).
Nest with young seen or heard.

Courtesy of MDNR (2012b).
Similar breeding codes were used in Atlas I and Atlas II and comply with NORAC standards.

b

50

Table C1. Northerly and southerly species groups.
Northerly Species (n=20)
Southerly Species (n=21)
Common Name
Scientific Name
Common Name
Scientific Name
Blackburnian Warbler
Acadian Flycatcher
Dendroica fusca
Empidonax virescens
Black-throated Blue
Blue-gray
Dendroica
Polioptila caerulea
Warbler
Gnatcatcher
caerulescens
Black-throated Green
Blue-winged
Dendroica virens
Vermivora pinus
Warbler
Warbler
Blue-headed Vireo
Carolina Wren
Vireo solitarius
Thryothorus
ludovicianus
Boreal Chickadee
Eastern Screech-Owl Megascops asio
Poecile hudsonica
Brown Creeper
Eastern Towhee
Certhia americana
Pipilo
erythrophthalmus
Canada Warbler
Wilsonia canadensis Field Sparrow
Spizella pusilla
Chestnut-sided Warbler
Grasshopper
Dendroica
Ammodramus
Sparrow
pensylvanica
savannarum
Common Loon
Great Egret
Gavia immer
Ardea alba
Gray Jay
Green Heron
Perisoreus
Butorides virescens
canadensis
Hermit Thrush
Henslow’s Sparrow
Catharus guttatus
Ammodramus
henslowii
Lincoln’s Sparrow
Hooded Warbler
Melospiza lincolnii
Wilsonia citrine
Magnolia Warbler
Least Bittern
Dendroica
Ixobrychus exilis
magnolia
Nashville Warbler
Louisiana
Vermivora
Seiurus motacilla
Waterthrush
ruficapilla
Northern Waterthrush
Northern Cardinal
Seiurus
Cardinalis cardinalis
noveboracensis
Olive-sided Flycatcher
Orchard Oriole
Contopus cooperi
Icterus spurius
Ruby-crowned Kinglet
Red-bellied
Regulus calendula
Melanerpes carolinus
Woodpecker
Swainson’s Thrush
Red-shouldered
Catharus ustulatus
Buteo lineatus
Hawk
Yellow-bellied
Tufted Titmouse
Empidonax
Baeolophus bicolor
Flycatcher
flaviventris
Yellow-bellied
Willow Flycatcher
Sphyrapicus varius
Empidonax traillii
Sapsucker
Yellow-throated
Vireo flavifrons
Vireo

51

Table C2. Insectivorous and non-insectivorous species groups.
Insectivorous (n=26)
Non-insectivorous (n=15)
Common Name
Scientific Name
Common Name
Scientific Name
Acadian Flycatcher
Empidonax virescens Great Egret
Ardea alba
Blackburnian Warbler
Red-shouldered
Dendroica fusca
Buteo lineatus
Hawk
Black-throated Blue
Boreal Chickadee
Dendroica
Poecile hudsonica
Warbler
caerulescens
Black-throated Green
Eastern Screech-Owl Megascops asio
Dendroica virens
Warbler
Blue-gray Gnatcatcher
Eastern Towhee
Polioptila caerulea
Pipilo
erythrophthalmus
Blue-headed Vireo
Field Sparrow
Vireo solitarius
Spizella pusilla
Blue-winged Warbler
Gray Jay
Vermivora pinus
Perisoreus
canadensis
Brown Creeper
Lincoln’s Sparrow
Certhia americana
Melospiza lincolnii
Canada Warbler
Northern Cardinal
Wilsonia canadensis
Cardinalis
cardinalis
Carolina Wren
Red-bellied
Thryothorus
Melanerpes
Woodpecker
ludovicianus
carolinus
Chestnut-sided Warbler
Swainson’s Thrush
Dendroica
Catharus ustulatus
pensylvanica
Grasshopper Sparrow
Yellow-bellied
Ammodramus
Sphyrapicus varius
Sapsucker
savannarum
Henslow’s Sparrow
Common Loon
Ammodramus
Gavia immer
henslowii
Hermit Thrush
Green Heron
Catharus guttatus
Butorides virescens
Hooded Warbler
Least Bittern
Wilsonia citrine
Ixobrychus exilis
Louisiana Waterthrush
Seiurus motacilla
Magnolia Warbler
Dendroica magnolia
Nashville Warbler
Vermivora ruficapilla
Northern Waterthrush
Seiurus
noveboracensis
Olive-sided Flycatcher
Contopus cooperi
Orchard Oriole
Icterus spurius
Ruby-crowned Kinglet
Regulus calendula
Tufted Titmouse
Baeolophus bicolor
Willow Flycatcher
Empidonax traillii
Yellow-bellied
Empidonax
Flycatcher
flaviventris
Yellow-throated Vireo
Vireo flavifrons

52

Table C3. Neotropical and non-neotropical species groups.
Neotropical (n=25)
Non-neotropical (n=16)
Common Name
Scientific Name
Common Name
Scientific Name
Acadian Flycatcher
Empidonax virescens Carolina Wren
Thryothorus
ludovicianus
Blackburnian Warbler
Eastern Screech-Owl Megascops asio
Dendroica fusca
Black-throated Blue
Gray Jay
Dendroica
Perisoreus
Warbler
caerulescens
canadensis
Black-throated Green
Northern Cardinal
Dendroica virens
Cardinalis
Warbler
cardinalis
Blue-gray Gnatcatcher
Red-bellied
Polioptila caerulea
Melanerpes
Woodpecker
carolinus
Blue-headed Vireo
Tufted Titmouse
Vireo solitarius
Baeolophus bicolor
Blue-winged Warbler
Boreal Chickadee
Vermivora pinus
Poecile hudsonica
Canada Warbler
Brown Creeper
Wilsonia canadensis
Certhia americana
Chestnut-sided Warbler Dendroica
Common Loon
Gavia immer
pensylvanica
Grasshopper Sparrow
Eastern Towhee
Ammodramus
Pipilo
savannarum
erythrophthalmus
Great Egret
Field Sparrow
Ardea alba
Spizella pusilla
Green Heron
Henslow’s Sparrow
Butorides virescens
Ammodramus
henslowii
Hooded Warbler
Hermit Thrush
Wilsonia citrine
Catharus guttatus
Least Bittern
Red-shouldered
Ixobrychus exilis
Buteo lineatus
Hawk
Lincoln’s Sparrow
Ruby-crowned
Melospiza lincolnii
Regulus calendula
Kinglet
Louisiana Waterthrush
Yellow-bellied
Seiurus motacilla
Sphyrapicus varius
Sapsucker
Magnolia Warbler
Dendroica magnolia
Nashville Warbler
Vermivora ruficapilla
Northern Waterthrush
Seiurus
noveboracensis
Olive-sided Flycatcher
Contopus cooperi
Orchard Oriole
Icterus spurius
Swainson’s Thrush
Catharus ustulatus
Willow Flycatcher
Empidonax traillii
Yellow-bellied
Empidonax
Flycatcher
flaviventris
Yellow-throated Vireo
Vireo flavifrons

53

Table D1. Results for each species in the center of occurrence, range boundary, and states of
occurrence analyses.
c
Species
Scientific Name Da MSb
Center of
Range
States of
FG
Occurrence Boundary Occurrence
Shifts (km) Shifts (km) Change
Acadian
S Neo
Insect
-1.27
37.32
0.07
Empidonax
Flycatcher
virescens
Blue-gray
S Neo
Insect
3.13
70.07
0.15
Polioptila
Gnatcatcher
caerulea
Blue-headed
Insect
1.54
-26.64
0.33
Vireo solitaries N Neo
Vireo
Blackburnian
N Neo
Insect -54.13
12.45
0.11
Dendroica
Warbler
fusca
Boreal
N Short Omni
5.54
9.36
-0.14
Poecile
Chickadee
hudsonica
Brown Creeper Certhia
N Short Insect -79.87
8.26
0.09
americana
Black-throated Dendroica
N Neo
Insect -51.58
-1.89
0.21
Blue Warbler
caerulescens
Black-throated Dendroica
N Neo
Insect -13.28
-13.25
0.16
Green Warbler virens
Blue-winged
S Neo
Insect
-8.75
75.79
0.01
Vermivora
Warbler
pinus
Carolina Wren Thryothorus
S Res
Insect
9.03
148.48
1.06
ludovicianus
Canada
N Neo
Insect -55.74
25.06
-0.08
Wilsonia
Warbler
canadensis
Common Loon Gavia immer
N Short Pisc
-34.05
35.30
0.03
Chestnut-sided Dendroica
N Neo
Insect -19.37
6.82
0.03
Warbler
pensylvanica
Eastern
Omni
3.76
-43.69
-0.10
Megascops asio S Res
Screech-Owl
Eastern
S Short Omni
24.01
-0.07
-0.11
Pipilo erythroTowhee
phthalmus
Field Sparrow Spizella pusilla S Short Omni
17.98
13.11
-0.12
Gray Jay
N Res
Omni
10.82
-2.28
-0.02
Perisoreus
canadensis
a

Statewide distribution: northerly (N) and southerly (S)
Migratory strategies: resident species (Res), short-distance migrants (Short), and longdistance Neotropical migrants (Neo)
c
Breeding season foraging guilds: carnivore (Carn), insectivore (Insect), omnivore (Omni),
and piscivore (Pisc)
b

54

Table D1 (cont’d)
Great Egret
Ardea alba
Green Heron
Butorides
virescens
Grasshopper
Ammodramus
Sparrow
savannarum
Henslow’s
Ammodramus
Sparrow
henslowii
Hermit Thrush Catharus
guttatus
Hooded
Wilsonia citrine
Warbler
Least Bittern
Ixobrychus
exilis
Lincoln’s
Melospiza
Sparrow
lincolnii
Louisiana
Seiurus
Waterthrush
motacilla
Magnolia
Dendroica
Warbler
magnolia
Nashville
Vermivora
Warbler
ruficapilla
Northern
Cardinalis
Cardinal
cardinalis
Northern
Seiurus
Waterthrush
noveboracensis
Orchard Oriole Icterus spurius
Olive-sided
Contopus
Flycatcher
cooperi
Red-bellied
Melanerpes
Woodpecker
carolinus
Ruby-crowned Regulus
Kinglet
calendula
RedButeo lineatus
shouldered
Hawk
Swainson’s
Catharus
Thrush
ustulatus
Tufted
Baeolophus
Titmouse
bicolor
Willow
Empidonax
Flycatcher
traillii

S
S

Neo
Neo

Carn
Pisc

-22.89
8.88

131.59
-5.73

0.13
-0.13

S

Neo

Insect

-49.91

46.98

-0.15

S

Short

Insect

52.01

-107.54

-0.19

N

Short

Insect

-29.94

6.08

0.11

S

Neo

Insect

-6.00

65.44

0.25

S

Neo

Pisc

13.35

-44.90

-0.09

N

Neo

Omni

-29.57

58.25

0.04

S

Neo

Insect

4.55

-17.05

-0.20

N

Neo

Insect

-68.73

15.64

0.18

N

Neo

Insect

-45.04

15.69

0.09

S

Res

Omni

5.11

70.52

-0.02

N

Neo

Insect

-19.68

1.42

0.04

S
N

Neo
Neo

Insect
Insect

3.41
-26.47

22.64
-39.65

0.29
-0.12

S

Res

Omni

-13.91

96.80

0.18

N

Short

Insect

-62.82

14.45

-0.09

S

Short

Carn

-34.83

-6.24

0.02

N

Neo

Omni

-60.28

8.97

0.00

S

Res

Insect

-17.14

53.80

0.06

S

Neo

Insect

10.97

-18.17

-0.07

55

Table D1 (cont’d)
Yellow-bellied Empidonax
Flycatcher
flaviventris
Yellow-bellied Sphyrapicus
Sapsucker
varius
YellowVireo flavifrons
throated Vireo

N

Neo

Insect

-81.74

21.96

0.20

N

Short

Omni

-23.89

42.50

0.16

S

Neo

Insect

4.77

51.66

0.04

56

APPENDIX B

Vita

57

VITA

Name

Jodi Marie Kreuser

Education
Name and Location

Dates

Diploma

D.C. Everest High School
Schofield, WI

1992-1996

Diploma

University of Wisconsin-Madison
Madison, WI

2008-2010

B.S. Wildlife Ecology

Michigan State University
East Lansing, MI

2010-2013

M.S. Fisheries and
Wildlife Ecology

Employer

Dates

Job Title

Michigan State University

2010-2013

Research Assistant

Michigan Department of Natural
Resources

2011-2012

GIS Project Intern

University of Wisconsin-Madison

2009

Grassland Birds Field
Research Technician

Employment

58

LITERATURE CITED

59

LITERATURE CITED

Anderson, D. R. 2008. Model based inference in the life sciences: a primer on evidence.
Springer, New York, New York, USA.
Andresen, J., S. Hilberg, and K. Kunkel. 2012. Historical Climate and Climate Trends in the
Midwestern USA. In U.S. National Climate Assessment Midwest Technical Input Report. J.
Winkler, J. Andresen, J. Hatfield, D. Bidwell, and D. Brown, technical coordinators. Available
from the Great Lakes Integrated Sciences and Assessments Center, http://glisa.msu.edu/docs/
NCA/MTIT_Historical.pdf.
Araujo M. B., R. G. Pearson, W. Thuiller, and M. Erhard. 2005. Validation of species-climate
impact models under climate change. Global Change Biology 11:1504–1513.
Balanyá, J., J. M. Oller, R. B. Huey, G. W. Gilchrist, and L. Serra. 2006. Global genetic change
tracks global warming in Drosophila subobscura. Science 313: 1773-1775.
Batdorf, K. E. 2012. Distributional changes in Ohio’s breeding birds and the importance of
climate and land cover change. Thesis, The Ohio State University, Columbus, USA.
BirdLife International. 2012. IUCN Red List for birds. Downloaded from http://www.birdlife.org
on 18/09/2012.
Boggs, C. L. and D. W. Inouye. 2012. A single climate driver has direct and indirect effects on
insect population dynamics. Ecology Letters 15: 502-508.
Both, C., S. Bouwhuis, C. M. Lessells, and M. E. Visser. 2006. Climate change and population
declines in a long-distance migratory bird. Nature 441: 81-83.
Both, C. and M. E. Visser. 2001. Adjustment to climate change is constrained by arrival date in a
long-distance migrant bird. Nature 411: 296-298.
Bradbury, R. B., J. W. Pearce-Higgins, S. R. Wotton, G. J. Conway, and P. V. Grice. 2011. The
influence of climate and topography in patterns of territory establishment in a range-expanding
bird. Ibis 153: 336-344.
Bradshaw, W. E. and C. M. Holapzfel. 2006. Climate change – evolutionary response to rapid
climate change. Science 312: 1477-1478.
Brewer, R., G. A. McPeek, and R. J. Adams. 1991. The atlas of breeding birds of Michigan.
Michigan State University Press, East Lansing, USA.

60

Brohan, P., J. J. Kennedy, I. Harris, S. F. B. Tett, and P. D. Jones. 2006. Uncertainty estimates in
regional and global observed temperature changes: A new data set from 1850. Journal of
Geophysical Research, 111, D12106, doi:10.1029/2005JD006548, 2006.
Brommer, J. E. 2004. The range margins of northern birds shift polewards. Annales Zoologici
Fennici 41: 391-397.
Charmantier, A., R. H. McCleery, L. R. Cole, C. Perrins, L. E. B. Kruuk, and B. C. Sheldon.
2008. Adaptive phenotypic plasticity in response to climate change in a wild bird population.
Science 320: 800-803.
Chen, I. C., Jane K. Hill, R. Ohlemuller, D. B. Roy, and C. D. Thomas. 2011. Rapid range shifts
of species associated with high levels of climate warming. Science 333: 1024-1026.
Christensen, J.H., B. Hewitson, A. Busuioc, A. Chen, X. Gao, I. Held, R. Jones, R.K. Kolli, W.T. Kwon, R. Laprise, V. Magaña Rueda, L. Mearns, C.G. Menéndez, J. Räisänen, A. Rinke, A.
Sarr and P. Whetton. 2007: regional climate projections. In: Climate Change 2007: The Physical
Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M.
Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, USA.
Cotton, P. A. 2003. Avian migration phenology and global climate change. Proceedings of the
National Academy of Sciences of the United States of America 100: 12219-12222.
Davies T.J., A. Purvis, and J. L. Gittleman. 2009. Quaternary climate change and the geographic
ranges of mammals. American Naturalist 174: 297-307
DeGraaf, R.M., N. G. Tilghman, and S. H. Anderson. 1985. Foraging guilds of North American
birds. Environmental Management 9: 493–536.
DeGraaf, R.M. and M. Yamasaki. 2001. New England wildlife: habitat, natural history, and
distribution. University Press of New England, Hanover, New Hampshire, USA.
Donnelly, A., T. Cooney, E. Jennings, E. Buscardo, and M. Jones. 2009. Response of birds to
climatic variability-evidence from the western fringe of Europe. International Journal of
Biometeorology 53: 211-220.
Ellwood, E. R., J. M. Diez, I. Ibanez, R. B. Primack, H. Kobori, H. Higuchi, and J. A. Silander.
2012. Disentangling the paradox of insect phenology: are temporal trends reflecting the response
to warming? Oecologia 168: 1161-1171.
Gaston, K. J. 1996. The multiple forms of the interspecific abundance-distribution relationship.
Oikos 76: 211-220.

61

Gaston, K. J. and T. M. Blackburn. 2000. Pattern and Process in Macroecology. Blackwell
Science, Malden, Massachusetts, USA.
Gibbons, D. W., P. F. Donald, H.G. Bauer, L. Fornasari, and I. K. Dawson. 2007. Mapping avian
distributions: the evolution of bird atlases. Bird Study 54: 324-334.
Gienapp, P., R. Leimu, and J. Merila. 2007. Responses to climate change in avian migration time
– microevolution versus phenotypic plasticity. Climate Research 35: 25-35.
Gill, F. B. 2007. Ornithology. W. H. Freeman. New York. USA.
Hayhoe, K., J. VanDorn, T. Corley II, N. Schlegal, and D. Wuebbles. 2010. Regional climate
change projections for Chicago and the US Great Lakes. Journal of Great Lakes Research 36
(Supplement 2), 7–21.
Hellman, J. J., K. J. Nadelhoffer, L. R. Iverson, L. H. Ziska, S. N. Matthews, P. Meyers, A. M.
Prasad, and M. P. Peters. 2010. Climate change impacts on terrestrial ecosystems in metropolitan
Chicago and its surround, multi-state region. Journal of Great Lakes Research 36: 74-85.
Hickling R., D. B. Roy, J. K. Hill, R. Fox, and C. D. Thomas CD. 2006. The distributions of a
wide range of taxonomic groups are expanding polewards. Global Change Biology 12: 450–455.
Hitch, A. T. and P. L. Leberg. 2007. Breeding distributions of North American bird species
moving north as a result of climate change. Conservation Biology 21: 534-539.
Hodgson, J. A., C. D. Thomas, T. H. Oliver, B. J. Anderson, T. M. Brereton, and E. E. Crones.
2011. Predicting insect phenology across space and time. Global Change Biology 17: 1289-1300.
Huntley B., Y. C. Collingham, R. E. Green, G. M. Hilton, C. Rahbek, and S. G. Willis. 2006.
Potential impacts of climatic change upon geographical distribution of birds. Ibis 148: 8-28.
IPCC (Intergovernmental Panel on Climate Change). 2007. Climate change 2007: the physical
science basis. Contribution of Working Group I to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change. Cambridge University Press, England.
Jiménez-Valverde, A., N. Barve, A. Lira-Noriega, S. Maher, Y. Nakazawa, M. Papeş, J.
Soberón, J. Sukumaran, and A. Townsend Peterson. 2011. Dominant climate influences on North
American bird distributions. Global Ecology and Biogeography 20: 114-118.
Jonzén, N., A. Lindén, T. Ergon, E. Knudsen, J. O. Vik, D. Rubolini, D. Piacentini, C. Brinch, F.
Spina, L. Karlsson, M. Stervander, A. Andersson, J. Waldenström, A. Lehikoinen, E. Edvardsen,
R. Solvang, and N. C. Stenseth. 2006. Rapid advance of spring arrival dates in long-distance
migratory birds. Science 312: 1959-1961.

62

Kovacs, S. T. Csorgo, A. Harnos, P. Fehervari, and K. Nagy. 2011. Change in migration
phenology and biometrics of two conspecific Sylvia species in Hungary. Journal of Ornithology
152: 365-373.
La Sorte, F. A., and F. R. Thompson. 2007. Poleward shifts in winter ranges of North American
birds. Ecology 88: 1803-1812.
Laughlin, S. B., D. P. Kibbe, and P. F. J. Eagles. 1982. Atlasing the distribution of breeding birds
in North America. American Birds 36: 6-19.
Lemoine N., H. Bauer, M. Peintinger, and K. Boehning-Gaese. 2007. Effects of climate and
land-use change on species abundance in a central European bird community. Conservation
Biology 21: 495-503.
Li, R., H. Tian, and X. Li. 2010. Climate change induced range shifts of Galliformes in China.
Integrative Zoology 5: 154-163.
Lindell, C. A., S. K. Riffell, S. A. Kaiser, A. L. Battin, M. L. Smith, and T. D. Sisk. 2007. Edge
responses of tropical and temperate birds. The Wilson Journal of Ornithology 119: 205-220.
MDNR. 2012a. The Michigan Breeding Bird Atlas II Project Product Dataset. The Michigan
Department of Natural Resources (MDNR), Wildlife Division, Lansing, Michigan; contract
project coordinated by the Kalamazoo Nature Center, Kalamazoo, Michigan. Online connection
to BBA2 Information: http://michigan.gov/dnr/0,1607,7-153-10370_35523---,00.html
and
http://www.michigandnr.com/ftp/requested_geospatial/wildlife_breeding-bird-atlas2.
MDNR. 2012b. The Michigan Breeding Bird Atlas II Project Documentation and Metadata. The
Michigan Department of Natural Resources (MDNR), Wildlife Division, Lansing, Michigan;
contract project coordinated by the Kalamazoo Nature Center, Kalamazoo, Michigan. Online
connection to BBA2 Information: http://michigan.gov/dnr/0,1607,7-153-10370_35523--,00.html
and http://www.michigandnr.com/ftp/requested_geospatial/wildlife_breeding-birdatlas2.
Melles, S. J., M. J. Fortin, K. Lindsay, and D. Badzinski. 2011. Expanding northward: influence
of climate change, forest connectivity, and population processes on a threatened species’ range
shift. Global Change Biology 17: 17-31.
Møller, A. P., E. Flensted-Jensen, K. Klargor, W. Ardal, and J. T. Nielsen. 2010. Climate change
affects the duration of the reproductive season. Journal of Animal Ecology 79: 777-784.
Niven, D. K., G. S. Butcher, and G. T. Bancroft. 2009. Christmas Bird Counts and climate
change: northward shifts in early winter abundance. American Birds, 63: 10-15.
Parmesan, C. 1996. Climate and species’ range. Nature 382: 765-766.

63

Parmesan, C. 2006. Ecological and evolutionary responses to recent climate change. Annual
Review of Ecology Evolution and Systematics 37: 637-669.
Parmesan, C., N. Ryrolm, C. Stefanescu, J. K. Hill, C. D. Thomas, H. Descimon, B. Huntley, L.
Kaila, J. Kullber, T. Tammaru, W. J. Tennett, J. A. Thomas, and M. Warren. 1999. Poleward
shifts in geographic ranges of butterfly species associated with regional warming. Nature 399:
579-583.
Parmesan, C. and G. Yohe. 2003. A globally coherent fingerprint of climate change impacts
across natural systems. Nature 421: 37-42.
Pineda-Diez de Bonilla, E., J. L. Len-Cortes, and J. L. Rangel-Salazar. 2012. Diversity of bird
feeding guilds in relation to habitat heterogeneity and land-use cover in a human-modified
landscape in southern Mexico. Journal of Tropical Ecology 28: 369-376.
Poole,
A.
(Editor).
2005.
The
Birds
of
North
America
Online:
http://bna.birds.cornell.edu.proxy2.cl.msu.edu/BNA/. Cornell Laboratory of Ornithology, Ithaca,
NY.
Pounds, J. A., M. R. Bustamante, L. A. Coloma, J.A. Consuegra, M. P. L. Fodgen, P. N. Foster,
E. La Marca, K. L. Master, A. Merino-Viteri, R. Puschendorf, S. R. Ron, G. A. SanchezAzofeifa, C. J. Still, and B. E. Young. 2006. Widespread amphibian extinctions from epidemic
disease driven by global warming. Nature 439: 161-167.
Przybylo, R., B. C. Sheldon, and J. Merilä. 2000. Climatic effects on breeding and morphology:
evidence for phenotypic plasticity. Journal of Animal Ecology 69: 395-403.
Pulido, F. P. Berthold, G. Mohr, and U. Querner. 2001. Heritability of the timing of autumn
migration in a natural bird population. Proceedings of the Royal Society B-Biological Sciences
268: 953-959.
Réale, D., A. G. MacAdam, S. Boutin, and D. Bertaux. 2003. Genetic and plastic responses of a
northern mammal to climate change. Proceedings of the Royal Society of London B-Biological
Sciences 270: 591-596.
Reif, J., K. S. Astny, and V. Bejcek. 2010. Contrasting effects of climatic and habitat changes on
birds with northern range limits in central Europe as revealed by an analysis of breeding bird
distribution in the Czech Republic. Acta Ornithologica 45: 82-89.
Robbins, C. S. and P. H. Geissler. 1990. Survey methods and mapping grids. Pages 4.1-4.20
in: C. R. Smith, editor. Handbook for atlasing American breeding birds. Woodstock, Vermont
Institute of Natural Science, USA.
Root T. L., J. T. Price, and K. R. Hall. 2003. ‘Fingerprints’ of global warming on animals and
plants. Nature 421: 57-60.

64

Sauer J. R., J. E. Hines, I. Thomas, J. Fallon, and G. Gough. 1999. The North American
Breeding Bird Survey, results and analysis 1966-1998. Version 98.1. USGS Patuxent Wildlife
Research Center, Laurel, Maryland, USA.
Schwartz, M.D., R. Ahas, and A. Aasa. 2006. Onset of spring starting earlier across the
Northern Hemisphere. Global Change Biology 12: 343-351.
Sparks, T. H., F. Bairlein, J. G. Bojarinova, O. Huppop, E. A. Lehikoinen, K. Rainio, L. V.
Sokolov, and D. Walker. 2005. Examining the total arrival distribution of migratory birds.
Global Change Biology 11: 22-30.
Stralberg, D., D. Jongsomjit, C. A. Howell, M. A. Snyder, J. D. Alexander, J. A. Wiens, and T.
L. Root. 2009. Re-shuffling of species with climate disruption: a no-analog future for California
Birds? PLoS ONE 4(9): e6825. doi:10.1371/journal.pone.0006825.
Thomas, C.D., A. M. A. Franco, and J. K. Hill. 2006. Range retractions and extinction in the face
of climate warming. Trends in Ecology and Evolution 21: 415–416.
Thomas, C. D., and J. J. Lennon. 1999. Birds extend their ranges northwards. Nature 399: 213.
Tingley, M. W., M. S. Koo, C. Moritz, A. C. Rush, and S. R. Beissinger. 2012. The push and
pull of climate change causes heterogeneous shifts in avian elevational ranges. Global Change
Biology 18: 3279-3290.
Turner, J. R. G., J. J. Lennon, and J. A. Lawrenson. 1998. British bird species distributions and
the energy theory. Nature 335: 539-541.
Visser, M. E., A. J. van Noordwijk, J. M. Tinbergen, and C. M. Lessells. 1998. Warmer springs
lead to mistimed reproduction in great tits (Parsus major). Proceedings of the Royal Society BBiological Sciences 265: 1867-1870.
Visser, M. E., C. Both, and M. M. Lambrechts. 2004. Global climate change leads to mistimed
avian reproduction. Advances in Ecological Research 35: 89-110.
Walther, G. R., S. Berger, M. T. Sykes. 2005. An ecological ‘footprint’ of climate change.
Proceedings of the Royal Society B-Biological Sciences 272: 1427-1432.
Walther, G. R., E. Post, P. Convey, A. Menzel, C. Parmesan, T. J. C. Beebee, J. M. Fromentin,
O. Hoegh-Guldberg, and F. Bairlein. 2002. Ecological responses to recent climate change.
Nature 416: 389-395.
Zuckerberg, B., W. F. Porter, and K. Corwin. 2009a. The consistency and stability of abundanceoccupancy relationships in large-scale population dynamics. Journal of Animal Ecology 78: 172181.

65

Zuckerberg, B., A. M. Woods, and W. F. Porter. 2009b. Poleward shifts in breeding bird
distributions in New York State. Global Change Biology 15:1866-1883.

66