Song Discrimination Between Two Subspecies of Vesper Sparrow: Pooecetes gramineus affinis and Pooecetes gramineus confinis

Item

Title

Eng Song Discrimination Between Two Subspecies of Vesper Sparrow: Pooecetes gramineus affinis and Pooecetes gramineus confinis

Date

Eng 2021

Creator

Eng Leque, Timothy

Identifier

Eng Thesis_MES_2021_LequeT

extracted text

Song Discrimination Between Two Subspecies of Vesper Sparrow:
Pooecetes gramineus affinis and Pooecetes gramineus confinis

by
Timothy Colin Leque

A Thesis Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
June, 2021

iii

© 2021 by Timothy Leque. All rights reserved.

iv

This Thesis for the Master of Environmental Studies Degree
by Timothy Leque

has been approved for
The Evergreen State College
by

John Withey, Ph.D.
Member of the Faculty

______________________
Date

v

Abstract

Song Discrimination Between Two Subspecies of Vesper Sparrow:
Pooecetes gramineus affinis and Pooecetes gramineus confinis
Timothy Leque

Vesper sparrows (Pooecetes gramineus) are grayish-brown songbirds of the family
Passerellidae, found in open spaces such as prairies, meadows, and sagebrush steppe. Like other
songbirds, male vesper sparrows sing throughout the breeding season to attract mates, as well as
to delineate and defend territories. The Oregon vesper sparrow (Pooecetes gramineus affinis) is a
subspecies endemic to the Pacific Northwest that has been identified as a species of conservation
concern throughout its range. There is little research on the Oregon vesper sparrow, with some
uncertainty regarding taxonomic status due to a lack of genetic analysis. Western vesper
sparrows (Pooecetes gramineus confinis) occur east of the Cascade Mountains and are common
and widespread throughout the western United States. The degree to which the boundary of the
Cascades affects speciation among vesper sparrows is unknown, as wintering ranges for the two
subspecies overlap in California. Differences in territorial response to conspecific song
playbacks are often associated with evolutionary divergence between subspecies. This study
involved exposing individual male vesper sparrows of two subspecies to playback of
consubspecific and heterosubspecifc songs. The vesper sparrows in this study demonstrated
some discrimination between subspecific song with differences in flight behavior between
playback treatments. There were also differences in behavioral responses to playback, with
Oregon vesper sparrows responding to playback with more singing, and western vesper sparrows
responding to playback with more flights.

vi

Table of Contents
Abstract .......................................................................................................................................... vi
Table of Contents ........................................................................................................................... iv
List of Figures ................................................................................................................................. v
List of Tables ................................................................................................................................. vi
Acknowledgements ....................................................................................................................... vii
Chapter 1: Introduction ................................................................................................................... 1
Chapter 2: Literature Review .......................................................................................................... 4
Bird Song: Function and Evolution ................................................................................................ 4
Sexual Selection .......................................................................................................................... 5
Male Territoriality ....................................................................................................................... 7
Vocal Tutoring, Cultural Transmission and Evolution ............................................................... 9
Avian Biogeography of the Pacific Northwest ............................................................................. 11
Playback Experiments ................................................................................................................... 14
Vesper Sparrow (Pooecetes gramineus) ....................................................................................... 17
Species Description ................................................................................................................... 17
Oregon Vesper Sparrow (Pooecetes gramineus affinis) ........................................................... 20
Western Vesper Sparrow (Pooecetes gramineus confinis) ....................................................... 25
Song of the Vesper Sparrow ..................................................................................................... 28
Listing Considerations for the Oregon Vesper Sparrow ............................................................... 30
Conclusion .................................................................................................................................... 31
Chapter 3: Methods ....................................................................................................................... 33
Oregon vesper sparrow ............................................................................................................. 33
Western vesper sparrow ............................................................................................................ 35
Chapter 4: Results & Discussion .................................................................................................. 45
Results ........................................................................................................................................... 45
Discussion ..................................................................................................................................... 52
References ..................................................................................................................................... 57

iv

List of Figures
Figure 1. Oregon vesper sparrow (Pooecetes gramineus affinis) on Joint Base Lewis-McChord,
Washington. .................................................................................................................................. 21
Figure 2. Western vesper sparrow (Pooecetes gramineus confinis) in Wenas Wildlife Area near
Ellensburg, Washington. ............................................................................................................... 26
Figure 3. P. g. affinis study sites within the South Puget Sound region of western Washington
State. Two training areas on Joint Base Lewis-McChord support the majority of breeding pairs of
P. g. affinis left in Washington…………………………………………………………………..34
Figure 4. P. g. confinis study sites within the Columbia Basin region of eastern Washington
State. Playback trials were conducted at two WDFW-managed wildlife areas in 2020, with some
songs recorded at a Nature Conservancy-owned site near Ephrata…………………………...…36
Figure 5. Spectrograms of six song exemplars, each from a unique individual P. g. affinis
occurring at different sites in the Puget Lowlands………………………………………………39
Figure 6. Spectrograms of six song exemplars, each from a unique individual P. g. confinis
occurring at different sites in the Columbia Basin…………………………………………...….40
Figure 7. Plots of parameter effects on the second principal component explaining flight
behavior (PC2fly). Plot (A) shows mean PC2fly values (with 95% CI) for playback trials separated
by subspecific stimuli type, with ‘consubspecific’ stimuli shown on the left, and
‘heterosubspecific’ stimuli on the right. Plot (B) shows mean PC2fly values (with 95% CI) for
playback trials separated by subject subspecies, with ‘P. g. affinis’ shown on left, and ‘P. g.
confinis’ on the right…………………………………………………………………………….49
Figure 8. Plot of parameter effects on the third principal component explaining song behavior
(PC3song). Plot (A) shows mean PC3song values (with 95% CI) for playback trials separated by
subject subspecies, with ‘P. g. affinis’ shown on left, and ‘P. g. confinis’ on the right……...…50

v

List of Tables
Table 1. Loadings of the principal component’s axis 1, 2, and 3 scores. Based on the specific
response variables with the highest loadings for each PC axis (shown in bold), PC1 was
designated the ‘Approach Behavior’ axis, PC2 the ‘Flight Behavior’ axis, and PC3 the ‘Song
Behavior’ axis. The cumulative proportion of variance explained by PC1 through PC3 was
65.4%.............................................................................................................................................45
Table 2. Model selection results using Akaike’s Information Criterion corrected for small
sample size (AICc). Principal components related to vesper sparrow approach, flight, and song
behavior written as PC1approach, PC2fly, and PC3song. Models with a delta value less than 2.0
(ΔAICc ≤ 2.00) are shown, along with the null (intercept-only). K equals the number of model
parameters, log(ℒ) equals the maximized log-likelihood value, Δ equals delta (the change in AICc
from the top model), and wi equals the Akaike weight for each well-supported model. Bold
indicates parameters with 95% CIs that do not overlap zero. The direction of influence for each
parameter is indicated with positive/negative signs (+/-). Null (intercept-only) subscripts indicate
the random effects included in all models for each principal component, where ID = Bird ID and
Obs = observer. n = 63 trials…………………………….……………………………………….47
Table 3. The influence of fixed effects from the top-ranked models for vesper sparrow
approach, flight, and song behavior. Principal components explaining approach, flight, and song
behavior denoted PC1approach, PC2fly, and PC3song. Parameter levels for treatment stimuli
(consubspecific, heterosubspecific) and subspecies (P. g. affinis, P. g. confinis) shown in
parentheses. Intercept and categorical level of reference condition (parentheses) included for
each component. Strong effects with 95% CIs that do not overlap zero are indicated in bold….48
Table 4. Means and standard deviations for each raw response variable collected during
playback trials, separated by treatment stimuli type……………………………………………..51

vi

Acknowledgements
This project would not have been possible without the help from numerous individuals, I
apologize if I have left anyone out of my acknowledgements.
First, I would like to thank my friends and colleagues with the Avian Conservation
Program, who all contributed to this project with their support and ornithological knowledge.
Specifically, I would like to thank Veronica Reed because her expertise in the execution and
analysis of playback experiments improved the quality of this thesis tremendously. I also would
like to thank Gary Slater, who proposed the idea of a subspecific playback experiment involving
the Oregon vesper sparrow, advised on study design, and assisted with the logistics of field work.
Without Gary’s consistent support in and out of the field this project simply would not have
happened. Also, thanks to Karla Kelly who helped conduct playback experiments and whose
expertise on the vesper sparrows of the Rainier Training Areas was a major contribution.
Several people from Joint Base Lewis-McChord’s Fish and Wildlife Program also made
essential contributions to this project. Jim Lynch facilitated access to the Artillery Impact Area
and Rainier Training Areas and advised on locations of vesper sparrows. Jim was especially
supportive of my thesis and engaged me in some great discussions on the Oregon vesper
sparrow. Dan Grosboll provided his explosive ordinance expertise to ensure my safety while
collecting data in the Artillery Impact Area. Dan also recommended several state wildlife areas
for recording vesper sparrows that I would end up using as backup study areas.
Thanks to John Withey, my thesis reader who helped me with the statistical analysis of
this project and provided substantive feedback on my thesis document.
I would also like to thank several other people who helped me along the way: Emily Lind
from Klamath Bird Observatory provided song recordings of vesper sparrows in Oregon and
advised me on equipment and protocol for recording birds in the field. Alison Styring of The
Evergreen State College advised me on song recording and sampling design very early on when I
was first planning my thesis and got me started in the right direction. Sanders Freed from the
Center for Natural Land Management provided access to Tenalquot Preserve.
And a special thank you to my partner, Gabriela Santiago, who assisted with field work
in eastern Washington and was patient and supportive throughout this time-consuming project.

vii

Chapter 1: Introduction

Washington State is home to two subspecies of vesper sparrow (Pooecetes gramineus):
the western vesper sparrow (Pooecetes gramineus confinis), which inhabits shrub-steppe and
pine savannah habitats in eastern Washington, and the Oregon vesper sparrow (Pooecetes
gramineus affinis), which is limited to remnant coastal prairies in western Washington. The
breeding ranges for these subspecies do not overlap, and they are somewhat different
morphologically. The Oregon vesper sparrows are slightly smaller on average, and with darker
upperparts and a buff-tinged belly (King, 1968a, 1968b; Pyle, 1997; Rising, 1996). This study
aimed to determine whether the two subspecies of vesper sparrow are able to discriminate
between each other’s song, and whether they exhibit any other differences in territorial response
behavior.
The two subspecies of vesper sparrow that occur in Washington also differ considerably
in population size and distribution. P. g. confinis is widespread in the Columbia Basin, Great
Basin, and Great Plains while P. g. affinis is limited to remnant grassland habitat in the Puget
Lowlands in Washington; and in the Willamette Valley, Umpqua Valley, Klamath Mountains,
and Rogue Basin in Oregon (Altman, Stinson, & Hayes, 2020). While P. g. confinis is among the
most abundant breeding birds found in sagebrush steppe habitat east of the Cascades, P. g. affinis
is a rare breeder within grassland habitats of coastal Washington and Oregon. In recent decades
P. g. affinis has experienced population decline, enough that they have been listed as endangered
in Washington and U.S. Fish and Wildlife service has been petitioned to list them under the
Endangered Species Act (Altman et al., 2020; American Bird Conservancy, 2016).

1

The subspecies designations for Pooecetes gramineus are accepted by most authors, but
there has never a genetic analysis to confirm their distinctiveness. Rising (1996) asserts that the
subspecies of Pooecetes gramineus are indistinguishable in the field, and that P. g. affinis and P.
g. confinis cannot be reliably distinguished in the hand. The variable and individualistic nature of
Pooecetes gramineus songs make it impossible for human observers to distinguish the subspecies
by ear. While humans may not be able to distinguish the song syllables and sequences unique to
a particular subspecies, it is possible that the birds themselves have this ability, and that these
differences contribute to reproductive barriers.
Most Passerines use song as a territorial signal, and this plays a key role in both
reproductive selection as well as the defense of resources. Among the oscines, these signals are
learned by juvenile birds (Baptista & Petrinovich, 1986). Cultural transmission of songs within
isolated bird populations over time can result in song divergence, which likely contributes to
genetic divergence and speciation (Mason et al., 2017; Podos & Warren, 2007). When divergent
populations meet again in secondary contact zones, these diverged signals can act as
reproductive barriers between subspecies or even smaller populations (Toews, 2017). Male
songbirds may not perceive a foreign song or dialect as an immediate threat, whereas the song of
a local bird indicates a confirmed competitor.
One common method for determining territorial response to birdsong used by
ornithologists is the playback experiment. Exposing subjects (birds) to different stimuli (song
recordings), researchers can tally territorial responses given by subjects to each treatment.
Statistical analysis can then be used to determine whether the sample of individual birds exposed
to stimuli differed in their responses to one stimulus type over another. Review of multiple
playback studies show that birds typically respond more aggressively to songs of their own
2

subspecies over songs from a foreign subspecies (T. H. Parker, Greig, Nakagawa, Parra, &
Dalisio, 2018). However, birds also react more aggressively to local songs versus those of a
disjunct population, and therefore results from playback experiments must be carefully examined
before drawing any conclusions regarding apparent discrimination.
The essence of this thesis is the use of song playback to help determine whether P. g.
affinis and P. g. confinis are able to discriminate between each other’s songs. The results of
these playback trials can provide evidence supporting or contradicting the current subspecies
designations of Pooecetes gramineus in western North America. Due to the limited research on
Pooecetes gramineus taxonomy, and the imperiled status of P. g. affinis, any additional
information on distinctions between P. g. confinis will assist in listing determinations of P. g.
affinis. The hope for this document is to provide additional evidence informing the taxonomic
status of P. g. affinis. This thesis has been written in four chapters. Chapter one (this
introduction) outlines the research question and methods, as well as the context for why this
research is important. Chapter two includes the literature review, which provides background on
the biological function of birdsong, the history and applications of playback experiments, and
summarizes subspecies descriptions of Pooecetes gramineus. Chapter three describes the field
and analysis methods that were used to address the research questions. Chapter four concludes
this thesis with a discussion of the biological meaning that could be interpreted from the results
of the experiment.

3

Chapter 2: Literature Review

The purpose of this thesis is to determine whether two subspecies of vesper sparrow (P.
g. affinis and P. g. confinis) can discriminate between each other’s songs. Answering this
question requires comprehension of the role that signaling plays in the life histories of songbirds.
One must also understand how avian signaling has evolved into the behavior we observe today,
and how it continues to change. The spatial extent of this study is the Pacific Northwest region of
North America, as well as geographic features within the region such as the Cascade Mountains.
The primary method of this study is the use of audio playbacks, a popular experimental method
that has contributed considerable knowledge on the communication of animals (Falls, 1992).
This method has been used to infer the degree of speciation among populations of numerous
avian taxa (T. H. Parker et al., 2018). As for the subjects of this research, the taxonomy of
Pooecetes gramineus is based on a variety of sources, including several dating back to the 19th
century. There has never been a genetic analysis of P. g. affinis, and Jones and Cornely (2002)
describe the subspecies designation of the vesper sparrow as “weakly defined to moderately
distinct.” The song of Pooecetes gramineus has been characterized as variable and individualistic
(Kroodsma, 1972; Ritchison, 1981), having implications on the designs, results, and discussion
of this study.
Bird Song: Function and Evolution
Birds produce sound as a means of communication, broadcasting signals throughout their
environment to be received by conspecifics sharing the same habitat. Simple calls are often used
to communicate information of immediate importance such as the location of individuals
(Marler, 2004) or the presence of predators (Smith, 1965), but more complex ‘songs’ are
4

primarily produced by males and are usually associated with breeding behavior. It is widely
accepted that the main functions of birdsong are for sexual selection, and for the defense and
sorting of territorial boundaries (Collins, 2004). While other bird groups extensively produce
sound as a means for communication and often ‘sing’, complex singing behavior is most
developed in the Passerines, or songbirds, the largest and most diverse order of birds. Singing is
essential to the breeding ecology of nearly all passerine birds and can determine reproductive
success (Potvin, Crawford, MacDougall-Shackleton, & MacDougall-Shackleton, 2015).
Nearly all birds possess syrinxes, noise producing organs believed to have developed in
an extinct common ancestor. These organs serve no apparent purpose besides the production of
noise signals and are therefore believed to provide an essential biological purpose. Birds have
evolved to utilize a wide variety of social strategies, ranging from solitary to highly communal,
and sound communication often plays a vital role in these interactions. Aural communication in
birds is highly variable, not only between orders and species, but also phenologically. This
variability is demonstrated by the contrast between winter flocking behavior, when contact and
alarm calls are used to locate conspecifics as well as evade predators, and spring breeding
territoriality, when these same species will partition themselves separately within the
environment and compete for mates and resources.
Sexual Selection
Behaviors associated with sexual selection and breeding are incredibly diverse in birds.
Examples of variable mating strategies include the coordinated ‘dances’ of Clark’s grebes
(Aechmophorus clarkii), the lekking of the greater sage-grouse (Centrocercus urophasianus),
building and presentation of decorative structures by bowerbirds (Ptilonorhynchidae), and food
gifting, or ‘tidbitting’ among gallinaceous birds (Stokes & Williams, 1971). In the Passerines,
5

song appears to serve a primary role in the selection of reproductive partners. The time of the
year when passerines sing is concurrent with the breeding season, when birds are selecting mates
and partitioning resources within their habitat.
Important traits determining mate selection in birds are thought to include some measures
of evolutionary fitness, and these can be physical, such as diet-influenced plumage ornaments
(G. E. Hill, 1990) or acoustic, such as fast trill rates or wide frequency bandwidths (Ballentine,
Hyman, & Nowicki, 2004; Collins, 2004). These characteristics often reflect higher-quality
males who are more successful in the acquisition and defense of resources. Certain structural
characteristics of a bird’s song can be associated with differences in body mass, virility, and
other measures of sexual fitness (Moseley & Podos, 2014). Projecting vocalizations requires
sophisticated motor function and stamina, and a strong singer indicates good physical condition
and strong motor skills to nearby conspecifics, advertising the survivability of that individual
(Moseley & Podos, 2014). The ability to perform well is vital to a male bird’s breeding success,
as higher quality performances are more likely to solicit a response from a female (Ballentine et
al., 2004). Stress or lack of food during the developmental stages of a bird’s life can be reflected
in weaker vocal performance as a breeding adult, negatively influencing reproductive success
(Moseley & Podos, 2014).
Another metric of birdsong that communicates an individual’s fitness is repertoire size.
Song repertoires are defined as the variety of syllables a bird is capable of performing, as well as
the number of combinations in which those syllables are vocalized. Repertoire size is often
associated with male survivorship, and females of several bird species have shown preferences
for large song repertoires in both laboratory settings and in the field (Collins, 2004; Potvin et al.,
2015). In a study of song sparrows (Melospiza melodia), large repertoires were associated with
6

territory possession and reproductive success, while birds with small repertoires were more
likely to lose their territories (Hiebert, Stoddard, & Arcese, 1989). Another study of song
sparrows found that repertoire size was more indicative of reproductive success than territory
location (Potvin et al., 2015). Despite this research, the ways in which repertoire size is related
to reproductive success is overall poorly understood. Some have hypothesized that the ability to
learn more songs is related to the size of certain areas in the avian brain, which can be affected
by developmental stress early in life (Collins, 2004). A reduced song repertoire may act as an
indicator of poor overall fitness, rendering that male an undesirable partner (Nowicki, Searcy, &
Peters, 2002). Song repertoire is undoubtedly an important function in sexual selection of many
bird species, however the ways in which song repertoire conveys biological fitness to potential
mates in poorly understood.
Male Territoriality
Territoriality is a common trait throughout the animal kingdom, and in migratory birds,
territoriality is exhibited during the spring and summer when birds have migrated to their
Northern breeding grounds. Passerine species often flock together for safety while on their
wintering grounds but compete over space and resources during the breeding season. Animal
territories can generally be defined as a large area in which breeding, nesting, and raising
fledglings occurs. Territory boundaries, along with the resources within them are defended from
neighboring conspecifics (Hinde, 1956). During the breeding season many passerines will
restrict themselves to their territories, habitually signaling their ownership to others by singing
and confronting any trespassers along territory boundaries (Hinde, 1956). Birdsong is widely
considered to serve the dual function of attracting sexual partners while simultaneously repelling
rival conspecifics competing for the same limited space and resources. The purpose of acoustic
7

signals among birds is therefore largely based on the identity of the receiver: a female seeking to
mate or a competing male of the same species. In the same ways that a song might communicate
physical condition to a potential mate, this information is also received by male conspecifics that
must determine whether to engage in a territorial dispute. A song that communicates physical
prowess through increased vocal performance may deter weaker males from intruding into the
territory (J. N. Phillips & Derryberry, 2017). For example, Moseley et al. (2013) found that
swamp sparrows (Melospiza georgiana) responded less aggressively to songs with artificially
weakened trill rates than to control songs. Those same birds also responded less aggressively to
artificially strengthened trills, unless the subject was a strong vocal performer, in which case they
responded more aggressively. Repertoire size is also associated with higher male performance,
and males with smaller repertoires are often ejected from their territories (Hiebert et al., 1989).
Male vocal performance is not the only factor influencing territorial response to songs
from conspecifics. Proximity of breeding territories also influences the strength of a male’s
response. The ‘dear enemy phenomenon’ is common throughout the animal kingdom, in which
territorial (usually) males respond less aggressively to individuals from neighboring territories
(Ydenberg, Giraldeau, & Falls, 1988). Among birds, these types of interactions are most
common in breeding territory situations, with males responding less aggressively the closer a
neighboring territory is to their own (Temeles, 1992). In other words, an individual might
exhibit a weakened response to a song from a bird they are more familiar with, such as a male
from an adjacent territory.
As male birds are the primary signalers during breeding, most studies of birdsong have
focused on the responses of territorial males. Across the numerous studies of avian
communication, these responses have been measured in a variety of ways. Songbirds will often
8

approach speakers broadcasting songs from conspecifics, and will even attack mounts coupled
with song playback (Akçay, Tom, Holmes, Campbell, & Beecher, 2011). Distance to speaker has
been a primary territorial response measure in many studies. ‘Soft’ or low frequency singing is a
lesser known territorial behavior, but has been observed in many passerines and has been
measured as a response in several studies (Searcy & Beecher, 2009). Another infrequent measure
of territoriality is ‘wing waves’, in which a bird usually puffs itself and flutters its wings (Akçay
et al., 2011; J. N. Phillips & Derryberry, 2017). Increased rate of birdsong is also commonly used
as a response measure, but song type matching/switching and overlapping are less reliable
responses, as research on their importance have produced variable results (Kolesnikova, Liu,
Kang, & Opaev, 2019; Searcy & Beecher, 2009). Recent literature analyzing song as a response
measure have focused on adjustments to the receiver’s signal in response to the original signaler
(Illes, Hall, & Vehrencamp, 2006). ‘Latency,’ or the time elapsed between signal broadcast and
the aforementioned responses, is another way territorial response can be quantified (McGregor,
1992). While distance to speaker or ‘signaler’ is considered the most reliable territorial response
measure, many researchers choose to use a combination of behaviors to quantify how breeding
birds react to competitive signaling, allowing for a more robust analysis.
Vocal Tutoring, Cultural Transmission and Evolution
Among the passerines, song is transmitted to offspring either innately or culturally, and
these different learning mechanisms are exemplified by two distinct groups within the order. The
suboscines, or Tyranni, have innate songs that young birds are able to produce even with the
absence of a vocal tutor (Kroodsma & Konishi, 1991). In contrast, oscine species including the
sparrows, thrushes, larks and finches have learned songs that are mimicked by young birds
exposed to conspecific vocal tutors (Slater, 1986). Members of this group may even have an
9

innate preference for learning songs of their own subspecies (D. A. Nelson, 2000), but can also
mimic other species songs when deprived of songs from their own (Kroodsma, 1972). Song
learning often occurs on natal grounds, when young and developing birds are exposed to songs
of their own species (Baptista & Gaunt, 1994), and later begin practicing their ‘plastic’ song
during dispersal (Marler & Tamura, 1964). Studies of some species have concluded that vocal
learning ceases once a bird is past the early developmental stage (Hiebert et al., 1989), while
others suggest that birds continue to learn songs from neighbors into their first breeding season
(D. A. Nelson, 2000).
For many birds, geographic isolation of breeding grounds can result in not only allopatric
speciation, but also differences in culturally transmitted songs. These culturally transmitted
songs can change over time due to imprecise copying by juveniles (Podos & Warren, 2007;
Slater, 1986), but also through selection influenced by structural and temporal differences in the
birds environments (Derryberry et al., 2018; Karin, Cicero, Koo, & Bowie, 2018; Slabbekoorn &
Smith, 2002; Wilkins, Seddon, & Safran, 2013). Consequently, populations of a species
geographically isolated in different ecosystems over an extended period would develop distinct
differences in song traits. These effects are apparent in comparisons of historical and current
song types from urban versus rural populations (Derryberry, 2011; Moseley, Phillips,
Derryberry, & Luther, 2019), and rapid song divergence among small isolated groups of
translocated individuals (K. A. Parker, Anderson, Jenkins, & Brunton, 2012). Species with high
site fidelity often develop dialects between isolated populations (Baker & Cunningham, 1985),
and if these dialects become different enough may act as breeding barriers, eventually leading to
speciation (Toews, 2017). There are several examples contradicting this theory particularly in

10

hybrid zones (Kenyon, Alcaide, Toews, & Irwin, 2017), indicating that the link between cultural
evolution and genetic isolation could be weak in some species.
While differences in culturally transmitted behavior cannot necessarily be equated to
genetic differences, the former does seem to be a reliable indicator of the latter (Mason et al.,
2017). It would be expected however, that the birds themselves, cuing into aspects of the signals
not immediately apparent to humans (frequencies, song length), would be able to discriminate
between the song of a closely related competitor and a more benign foreign individual (Mason et
al., 2017). Recent studies continue to provide evidence of the correlation between genetic and
acoustic differences between oscines at the species and subspecies levels (Demko, Sosa-López,
& Mennill, 2019; Pegan et al., 2015; Sosa-López, Martínez Gómez, & Mennill, 2016).
Summary of published literature on song recognition appears to support the assertion that
genetically isolated bird populations can discriminate between each other’s songs (Freeman &
Montgomery, 2017; T. H. Parker et al., 2018).
Avian Biogeography of the Pacific Northwest
While birds are able to disperse and colonize new areas more easily than other terrestrial
animals, geographic barriers do contribute to speciation among many populations by restricting
movement and by producing stark climatic differences. In North America, large mountain ranges
mostly running north to south, are massive physical barriers to animal dispersal and produce
distinct climatic regions. It is no coincidence that mountain ranges often coincide with the
boundaries of speciation for many organisms (Swenson & Howard, 2005). These geographic
variations have resulted in elaborate species diversification among many families of birds. The
dynamics of speciation across geographic barriers within the intermountain west are incredibly

11

complex (Behle, 1978; Stein, Kutner, Hammerson, Master, & Morse, 2000), and the ability of
birds to disperse to desirable habitat adds further difficulty in generalizing these processes.
Reconstructing the evolutionary histories of migratory birds is further complicated by differences
in geographic area of breeding versus wintering ranges (Barker, Burns, Klicka, Lanyon, &
Lovette, 2015). This section of the literature review focuses on the geographic extent of the study
area, contrasting between coastal and interior habitats on either side of the Cascade mountain
range.
The mountains along the Pacific Crest of North America act as a boundary between
numerous endemic coastal bird species or subspecies and those of the interior west (Behle,
1978). While the avian endemism of coastal California west of the Sierra-Nevada mountains is
well known, similar dynamics in the Pacific Northwest are present, with coastal Washington
harboring more breeding and wintering bird species than in the interior of the state (Stein et al.,
2000). Several endemic or near-endemic avian species of the Pacific Northwest include the redbreasted sapsucker (Sphyrapicus ruber), chestnut-backed chickadee (Poecile rufescens), and
sooty grouse (Dendragapus fuliginosus); as well as numerous endemic subspecies such as the
streaked horned lark (Eremophila alpestris strigata), Puget Sound white-crowned sparrow
(Zonotrichia leucophrys pugetensis), and black merlin (Falco columbarius suckleyi). The
Cascade Mountains act as a high-elevation barrier between coastal and interior habitats, despite a
relatively short distance of just over 100 miles. The climatic and vegetative differences across
this distance is remarkable with temperate forests of the Puget Lowlands on one end of the
spectrum and the sagebrush steppe of the Columbia Plateau on the other, divided by the Pacific
cordillera.

12

Despite the geographic barriers and climatic contrasts, there is still considerable gradation
in ecotones within the Pacific Northwest, allowing for clinal zones among a variety of closely
related bird species and subspecies. Well-studied examples of clines include the hybridization
zone in Western Washington and Oregon between Townsend’s warblers (Setophaga townsendi)
and hermit warblers (Setophaga occidentalis) (Krosby & Rohwer, 2010), and hybrid zones
among Sphyrapicus woodpeckers (Seneviratne, Davidson, Martin, & Irwin, 2016). While
hybridization between different species might be the most visible example of this phenomenon, it
also occurs between subspecies within secondary contact zones, where previously isolated
population segments begin to overlap (Short, 1969). Hybridization between formerly isolated
populations has been documented in a number of bird species, with a considerable amount of
research on subspecies of white-crowned sparrows (Brooks & Wimberger, 2018; Lipshutz,
Overcast, Hickerson, Brumfield, & Derryberry, 2017).
These clinal zones can be particularly dynamic in mountain environments, where habitat
conditions are variable from year to year. Coupling this fact with the ability of birds to shift their
breeding ranges based on annual habitat conditions produces a serious taxonomic challenge for
scientists attempting to designate subspecies boundaries across mountain gradients. North
American mountain ranges, including the Cascades, are known to be hotspots of hybridization
due to their cooccurrence with species and subspecies boundaries (Swenson & Howard, 2005).
The presence of intergradation between population segments occurring in these areas means that
morphological traits can be an unreliable indicator of species status. Researchers wishing to
determine the locations of boundaries may need to take into account genetic and behavioral
differences as well.

13

Playback Experiments
Song playback is a tool most commonly known for its practical applications but has also
been a popular experimental method in the fields of animal behavior and bioacoustics,
particularly for birds. Playback is used by biologists, wildlife enthusiasts and hunters alike for
drawing wildlife close enough to survey, view, photograph and kill; as well as by land managers
for repelling certain animals from airfields, livestock and crops (Falls, 1992). Experimental
playback conducted by researchers usually involves exposing animal subjects to recordings of
audio stimuli broadcast through a speaker, either in the field or in laboratories. Often animal
subjects are exposed to several different “treatments” in the forms of artificially manipulated
animal signals, or more often, signals from conspecifics varying by geographic location. Avian
playback experiments have contributed to knowledge on the pertinent structural characteristics of
signals used in communication (Illes et al., 2006; Moseley et al., 2013), effects of anthropogenic
noise on signaling (Luther & Magnotti, 2014; Moseley et al., 2019), parallels between species
and song divergence (T. H. Parker et al., 2018), and determining taxonomic statuses (Alström &
Olsson, 1999; Alström, Rasmussen, Olsson, & Sundberg, 2008; Randler et al., 2012). The
number of playback studies testing differences in geographically separated populations of birds
has allowed for meta-analyses attempting to generalize the cumulative results of these
experiments (Freeman & Montgomery, 2017; T. H. Parker et al., 2018).
Playbacks have been used in numerous experiments involving a variety of taxa including
mammals, reptiles, amphibians, insects, and fish but not unexpectedly, these methods have been
utilized most frequently with birds. Although the use of playback on birds can be found in
literature dating back to the 1930s, the first study employing experimental playback treatments
was Dilger’s (1956) comparison of thrush (Hylocichla and Catharus spp.) responses to songs of
14

other species versus their own (Falls, 1992). Some of these early playback studies informed
ornithologists on neighbor/stranger discrimination (Weeden & Falls, 1959), vocal tutoring
(Thorpe, 1958), repertoires (Hinde, 1958), and important structural components of songs (Abs,
1963; Falls, 1992). While many playback studies test the responses of males actively defending
territories, experiments measuring female response to playback are far more scarce (Falls, 1992).
A large portion of these studies compare responses of territorial males to intrusions from
different types of signals, what the subject believes to be other birds. These can be digitally
manipulated versions of the original signal to determine structural characteristics subjects are
responding to like trill rate (Illes et al., 2006) or frequency (Luther & Magnotti, 2014).
In 1984, Hulbert published Pseudoreplication and the Design of Ecological Field
Experiments, a critical piece that spurred debate among practitioners of song playback
experiments. Hulbert identified pseudoreplication as a common design flaw in many of the
published studies involving ecological field experiments. Donald Kroodsma published several
papers discussing the design of animal playback experiments, with a particular focus on
pseudoreplication in the wake of Hulbert’s work (Kroodsma, 1986, 1989). Many of the early
playback studies had committed pseudoreplication (Kroodsma, 1989), and recent analysis of
these studies found that they produce more variable results than experiments that adequately
sampled stimuli (T. H. Parker et al., 2018). In many of these experiments, only one or a few
songs would be selected as stimuli and explanatory variables. If the song selected happened to be
from an individual with a below or above average song performance, this could in turn weaken
or strengthen the response to one of the playback treatments. In 1992, The Thornbridge Hall
NATO ARW Consensus was published summarizing best study design and execution practices
recommended by practitioners throughout the field. Essentially, the authors argue for the

15

sampling of stimuli along with the sampling of subjects to accurately measure the response to the
song types being tested. Several of the authors of the consensus revisited this work about a
decade later (Kroodsma, Byers, Goodale, Johnson, & Liu, 2001; Mcgregor, 2000). Research
published since the pseudoreplication debate have generally adopted these better study design
practices, although there are still examples of recent studies committing this kind of sampling
error (T. H. Parker et al., 2018).
Frequently the different types of signals used as test stimuli are chosen to represent
individuals sourced from a finite geographic area. A large body of research on the role of song in
the evolutionary divergence of passerines includes numerous studies employing playback
experiments as the primary methodology. Male territorial responses have been tested for
differences in subspecies (Liu, Lohr, Olsen, & Greenburg, 2008; Matessi, Dabelsteen, & Pilastro,
2001; Petrinovich & Patterson, 1981), regional dialects (D. A. Nelson, 1998; Petrinovich &
Patterson, 1981), and simply local versus non-local birds (Searcy, Nowicki, & Hughes, 1997). In
most of these studies, results suggest that birds can discriminate between songs of their own and
those of ‘foreign’ individuals (Freeman & Montgomery, 2017). A meta-analysis of local versus
foreign song discrimination found that treatments between subspecies produced the most
convincing examples of vocal discrimination by subjects (T. H. Parker et al., 2018). Observing
differences in male territorial response between songs of separate subspecies may indeed serve
as an appropriate indicator of speciation.

16

Vesper Sparrow (Pooecetes gramineus)
Species Description
Pooecetes gramineus or the Vesper Sparrow, previously known as the “bay-wing
bunting” and “grass finch”, is a member of the family Passerellidae, which includes towhees,
sparrows, buntings and longspurs. The only member of their genus, Pooecetes gramineus are
relatively large sparrows, colored grayish brown to light tan, with dark brown streaking (Jones &
Cornely, 2002; Pyle, 1997). They can be differentiated from other similarly drab sparrows by a
white eye-ring, white outer retrices, and chestnut-brown lesser coverts (Jones & Cornely, 2002;
Pyle, 1997). There is little sexual dimorphism within the species, with female measurements
averaging slightly smaller than males (Pyle, 1997).
Pooecetes gramineus is a Nearctic migratory passerine confined to North and Central
America, with four recognized subspecies (Jones & Cornely, 2002). Pooecetes gramineus
gramineus, or “Eastern vesper sparrow” occurs from along the east coast of North America to the
western edge of their range, Minnesota down through Texas (Berger, 1968; Pyle, 1997).
Pooecetes gramineus confinis, also known as “Western vesper sparrow” or “Great Basin vesper
sparrow” can be found from western Nebraska all the way to the eastern slopes of the Cascade
mountains, and south into Mexico during winter (King, 1968b; Pyle, 1997). Pooecetes
gramineus affinis, the Oregon Vesper Sparrow can be found along the Pacific coast, west of the
Cascade and Sierra Nevada mountain ranges (Altman, 2011, 2017; King, 1968a; Pyle, 1997).
Pooecetes gramineus altus, the “mountain vesper sparrow” has the most limited range, confined
to parts of Arizona, New Mexico, Utah, and Colorado (Johnson & Dickerman, 2006; A. R.
Phillips, 1964; Pyle, 1997).

17

Pooecetes gramineus received their common name “vesper sparrow” due to the fact they
are often the last bird heard singing into the evening, with “vesper” translating to “evening” in
Latin (Jones & Cornely, 2002). The males often sing from elevated perches, while females select
nest sites on the ground, building small grass cups at the base of clumps of vegetation, sticks or
sod (Jones & Cornely, 2002). Pooecetes forage on the ground, eating a mix of grass and forb
seeds, but mostly insects during the breeding season. Mothers feed the young of the year
primarily insects (Jones & Cornely, 2002).
Habitat used by Pooecetes gramineus can be generally characterized as dry open spaces
dominated by short and sparse vegetation, with some shrub or tree cover and bare ground (Camp
& Best, 1993; Dechant et al., 2002; Jones & Cornely, 2002). Wiens (1969) and Harrison (1974)
both described Pooecetes gramineus microhabitat preferences as being “xeric, sparsely
vegetated” (p. 40, p. 37). Pooecetes gramineus breed in a variety of ecosystems, including
sagebrush steppe, montane meadows, cropland, sandplain grasslands, reclaimed surface mines,
coastal prairie and desert shrub and grasslands (Dechant et al., 2002; Jones & Cornely, 2002;
Vickery, Hunter, & Wells, 1999; Wray, Strait, Whitmore, & Sparrow, 1982). Historically,
Pooecetes gramineus likely used early successional habitats created by natural disturbances such
as wildfires, erosion, grazing and trampling by bison (Best & Rodenhouse, 1984; Jones &
Cornely, 2002). Many of the current habitats used by Pooecetes gramineus can be described as
edges of anthropogenic openings, with artificial disturbance regimes tied to human land
management practices such as grazing, crop production, mowing, and prescribed burning. This
preference for disturbed areas can lead Pooecetes gramineus to breed in less than ideal habitats,
such as in active agricultural land (Rodenhouse & Best, 1983), mowed airports, and reclaimed
surface mines (Wray et al., 1982). Pooecetes gramineus habitat preferences have been

18

characterized as ecotonal (Owens & Myres, 1973), breeding along fencerows adjacent to
cropland (Roadhouse & Best, 1983) and within grassland-woodland transitions (Dechant et al.,
2002; Finzel, 1964).
During the breeding season, male Pooecetes gramineus spend a large portion of their
time singing from elevated perches to delineate and defend their territories (Jones & Cornely,
2002). Multiple studies have identified perches as an important factor during breeding territory
selection (Berger, 1968; Best & Rodenhouse, 1984; Castrale, 1983; Dechant et al., 2002;
Rodenhouse & Best, 1983; Schaid, Uresk, Tucker, & Linder, 1983; Wiens, 1969). Berger (1968)
observed that P. g. gramineus preferred perches greater than seven and a half meters tall in
territories along edges of woodland. Best and Rodenhouse (1984) found associations with P. g.
gramineus territory selection and pairing success to proximity of fencerows and shrub cover in
Iowa cropland. In this study the authors attributed this association to the availability of singing
perches along fencerows in a landscape otherwise devoid of perches. Castrale (1983) found P. g.
confinis perch selection was related to intershrub distance and shrub density across several
breeding sites in Utah.
Structure seems to be the most important characteristic of adequate perches, as Pooecetes
gramineus apparently have no preference for live versus dead shrubs (Best, 1972; Castrale,
1983). A variety of reported perch heights suggests that Pooecetes gramineus probably do not
select from a fixed range of heights but utilize perches that contrast structurally from the
dominant vegetation height across a site, as well as the heights and volumes of other available
perches (Castrale, 1983). Similarly, Harrison (1977) found no height preference by P. g.
gramineus using artificial perches in Michigan. Although perches are an important habitat

19

requirement of Pooecetes gramineus, they prefer low densities of elevated perches within a
sparse grassland landscape.
Oregon Vesper Sparrow (Pooecetes gramineus affinis)
Miller provided the first formal description of P. g. affinis in 1888, although J. G. Cooper
noted the presence of vesper sparrows in the Puget lowlands nearly 30 years prior (Suckley &
Cooper, 1860). Miller (1888, pp. 404-405) differentiated P. g. affinis from neighboring
subspecies by its smaller size, and “… having the ground color above buffy-brown rather than
grayish-brown. All the lighter areas of the plumage (including crissum, under wing-coverts and
lining of wings) suffused with pinkish buff.” (Figure 1). Pyle’s (1997) measurements of P. g.
affinis were consistent with Miller’s as being smaller than other subspecies of vesper sparrow
and describes plumage characteristics as well: “upperparts medium-dark grayish brown;
underparts white with a buff tinge” (p. 558). P. g. affinis has since been accepted as a distinct
subspecies due to these physical differences as well as their geographic isolation from other
vesper sparrows, west of the Pacific crest (Altman, 2017; King, 1968a; Pyle, 1997).
P. g. affinis spend April through July on their breeding grounds in Washington and
Oregon, migrating in August and September to spend the remainder of the year on their
wintering grounds in California (Altman, 2017). The current known P. g. affinis breeding range
extends north into Washington, with the majority of birds breeding on the South Sound Prairies
within Joint Base Lewis-McChord (JBLM), and a few occasionally spotted on the eastern side of
the Olympic Peninsula, San Juan Islands, and along the lower Columbia River (Altman, 2017).
In 1860, Suckley and Cooper wrote that P. g. affinis was “[r]ather abundant on the Nisqually
plains, Puget Sound” (p. 200). This passage suggests that the current breeding sites of P. g.
affinis on JBLM are likely within the core of their historical range in Washington. In Oregon, P.
20

g. affinis breed in the Willamette Valley, Umpqua Valley and Rogue Basin (Altman, 2017). P. g.
affinis migrate to California during winter, where they are found at low elevations from the
Central Valley to Northern Baja California (Erickson, 2008). During winter, P. g. affinis range
overlaps with P. g. confinis, which makes monitoring the wintering habits of these subspecies
difficult (Altman, 2017; Erickson, 2008). P. g. affinis’ breeding range historically extended into
Northern California as well as Southwest British Columbia and Northwest Washington, but the
species has been extirpated from most of their former breeding locations in these areas (Altman,
2011, 2017; Beauchesne, 2002; Roger, 2000). As Altman (2017) points out, the northward and

Figure 1. Oregon vesper sparrow (Pooecetes gramineus affinis) on Joint Base Lewis-McChord,
Washington.

21

southward range extractions experienced by P. g. affinis populations are consistent with common
extirpation patterns among bird communities. Curnutt, Pimm & Maurer (1996) found that
peripheral sites with variable abundance of sparrows had populations that were likely less
resilient to random environmental factors than those sparrow populations in the cores of their
ranges. This pattern parallels that of the P. g. affinis range contractions. Massive habitat loss and
degradation, especially within the northern extent of P. g. affinis’ former range (Chappell, Gee,
Stephens, Crawford, & Farone, 2001), would explain the pattern of local extirpations on small,
isolated remnant prairies. In this respect, the decline of P. g. affinis resembles that of other
coastal prairie obligate species listed under the Endangered Species Act, such as the streaked
horned lark (Eremophila alpestris strigata), Mazama pocket gopher (Thomomys Mazama), and
Taylor’s checkerspot butterfly (Euphydryas editha taylori).
Pooecetes gramineus affinis range retraction also parallels those of other prairie-oak
associated birds once more widespread across the Pacific Northwest, such as western
meadowlark (Sturnella neglecta), western bluebird (Sialia mexicana), and slender-billed whitebreasted nuthatch (Sitta carolinensis aculeate) (Altman, 2011). Similar trends of grassland bird
habitat loss, habitat fragmentation and habitat degradation due to residential and agricultural
development have been documented across the continent (Brennan & Kuvlesky Jr., 2005;
Vickery & Herkert, 2001). These effects have been more pronounced in the Pacific Northwest,
where available habitat was historically limited and of which only a small fraction remains
(Chappell et al., 2001).
Current breeding sites of P. g. affinis include restored prairie, airports, river dredge
deposits, Christmas tree farms, and grazed pastureland (Altman, 2017). Vickery, Hunter &
Melvin (1994) found that P. g. gramineus reached a 50% rate of occurrence on grassland patches

22

20 hectares (50 acres) or more in area, and that in general, presence of P. g. gramineus was
positively correlated with patch size; current breeding sites in Washington range in the hundreds
of acres. In Washington, known P. g. affinis breeding populations occur on sites with artificial
disturbance regimes. JBLM’s prairies frequently burn due to artillery sparked wildfires as well as
an active prescribed burn program. Airports throughout the region adhere to FAA mowing
standards, and Columbia River dredge deposits are frequently covered with new material (U.S.
Army, 2014). The largest populations of P. g. affinis in Washington (150-200 birds; Altman,
2017) occur on JBLM, where available habitat is regularly burned, approximately every three to
five years (Hill, Kronland, & Martin, 2017; Tveten & Fonda, 1999). These burned areas are then
seeded with native forbs and grasses to meet the installation’s habitat restoration goals. JBLM’s
Fish and Wildlife program also manages perch structure in occupied and potentially occupied
training areas to promote favorable conditions for P. g. affinis (Jim Lynch, personal
communication). On JBLM, most P. g. affinis nests have been found at the base of native
bunchgrasses, especially Roemer’s Fescue (Festruca roemeri), the dominant plant species seeded
onto restoration prairies (Kronland, Hill, & Martin, 2018). P. g. affinis regularly perch and sing
from man-made structures such as tanks, signs, and airfield lights.
Sparsely vegetated habitat associated with regular exposure to fire was once widespread
across the Southern Puget Sound’s fescue-dominated prairies (Chappell & Crawford, 1997).
Prairie-oak habitat throughout the Puget Sound region had been maintained through burning by
Indigenous peoples, who burned tracts of land to aid in the production of root and berry crops as
well as providing openings for hunting (Boyd, 1999; Deur & Turner, 2005; Norton, 1979).
Indigenous people did not limit fires to the prairies; they also allowed them to carry into the
forest (Deur & Turner, 2005), resulting in a much broader transitional edge between grassland

23

and prairie than what can now be found on the remaining prairies of JBLM. P. g. affinis breeding
locations in Washington, are generally found along forest-prairie transitions on JBLM (Jim
Lynch, personal communication). The colonization of the region by European and American
settlers brought Indigenous burning practices to an end, which led to loss of prairie habitat to
widespread conifer and shrub encroachment, particularly by Douglas fir (Pseudotsuga menziesii)
and Scotch broom (Cytisus scoparius) (Foster & Shaff, 2003; Tveten & Fonda, 1999). The lack
of fire and the colonization of open land by exotic pasture grasses introduced by homesteaders
(White, 1980) has increased the height and density of vegetation on what were historically
prairies (Dennehy et al., 2011). These changes have been especially dramatic in the northern
reaches of P. g. affinis’ former range in British Columbia and northern Puget Sound, resulting in
extirpation of those populations (Altman, 2011; Beauchesne, 2002; White, 1980). Removal of
regular disturbance, introduction of exotic plant species and the resulting change of vegetation
structure, has most likely degraded habitat in P. g. affinis’ former range to a degree that most of
those prairie remnants have been rendered unusable.
In the Willamette Valley of Oregon, home to a population of P. g. affinis, modern
agricultural burning has replaced historical indigenous burning (Johannessen, Davenport, Millet,
& Mcwilliams, 1971). Many of the other P. g. affinis breeding sites in Oregon are ranches and
ranges subject to cattle grazing, which may also mimic natural disturbances (Altman, 2017).
These anthropogenic disturbance regimes are likely creating and sustaining the open, sparse
areas of patchy vegetation that P. g. affinis favor for breeding, somewhat resembling the natural
disturbances of the past.
In recent years, P. g. affinis have become increasingly rare in the Pacific Northwest, with
current estimates at 300 individuals left in Washington State (Altman et al., 2020). This
24

subspecies currently holds state protected status throughout its entire breeding and wintering
ranges, and the U.S. Fish and Wildlife Service has been petitioned to list P. g. affinis under the
federal Endangered Species Act (ABC, 2016). Limited published literature is available on this
subspecies, so there is a need for additional research to inform conservation efforts. Altman
(2017) states “[t]he highest priority research need is to understand the role of demographic
parameters on population status. Two other important research needs include genetic evaluation
of the boundaries of subspeciation, and determination of factors influencing populations on the
wintering grounds.”
Western Vesper Sparrow (Pooecetes gramineus confinis)
The Western vesper sparrow (Pooecetes gramineus confinis), also known as the Great
Basin vesper sparrow, was first described by Baird in 1858, and is the largest of the Pooecetes
gramineus subspecies (Jones & Cornely, 2002; King, 1968b; Pyle, 1997; Rising, 1996).
Compared with P. g. gramineus, P. g. confinis also has a thinner bill, slender streaking and paler
gray coloration, but can be difficult to distinguish from P. g. affinis, even in hand (Rising, 1996).
Plumage differences between the subspecies are also apparent in juvenile birds, with young P. g.
confinis exhibiting lighter edging and buff along the back and crown (Figure 2; King, 1968b).
Pyle (1997) offers a similar description of bill and coloration, adding that the tail of P. g. confinis
average relatively longer than other subspecies. He also notes that the P. g. confinis of the
Columbian Plateau average darker in plumage, leading to the designation of these birds as a
separate subspecies by Jewett et al. (1953). However, this is the only source that separates the
Columbian P. g. confinis as a distinct subspecies and Pyle attributes the darker coloration to
intergradation with nearby populations of P. g. affinis (Pyle, 1997; Roger, 2000). The vesper
sparrows of the Columbia Basin are widely considered synonymous with P. g. confinis across
25

Figure 2. Western vesper sparrow (Pooecetes gramineus confinis) in Wenas Wildlife Area near
Ellensburg, Washington.

authors (AOU, 1957; Jones & Cornely, 2002; King, 1968b; Pyle, 1997; Rising, 1996; Roger,
2000).
The Northern extent of P. g. confinis’ range reaches into eastern British Columbia (K. J.
Nelson & Martin, 1999), with the southern extent of their wintering range in Mexico (Pulliam &
Mills, 1977). P. g. confinis breeds as far west as the eastern slopes of the Sierra-Nevada
mountains and as far east as western Nebraska (Pyle, 1997). Throughout western North America,
P. g. confinis is a common species associated with sagebrush (Artemisia spp.) steppe habitats
across a wide altitudinal range, although they can also be found in open juniper and ponderosa
26

pine woodlands as well as montane meadows (Dechant et al., 2002; Finzel, 1964; King, 1968b;
Schaid et al., 1983). Due to the large availability of these habitats in the intermountain west, P. g.
confinis is far more widespread and numerous than P. g. affinis, whose suitable habitat is scarce
within their already limited range.
The general habitat characteristics of Pooecetes gramineus are consistent with the
preferences of P. g. confinis: sparse and open grasslands with scattered perches. In Wyoming, for
example, P. g. confinis were found to be common in grasslands that were transitioning to forest,
characterized as grassland with a few scattered conifers, early successional trees and moderate
shrub cover (Finzel, 1964). In the Northern Great Plains, presence of sagebrush was a limiting
factor for P. g. confinis on reclaimed surface mines, suggesting that available perches are a
habitat requirement of P. g. confinis (Schaid et al., 1983). Castrale (1983) found that in Utah, P.
g. confinis showed a preference for the most prominent perches available, and showed an
association with intermediate shrub densities relative to the other species present. While strongly
associated with sagebrush, P. g. confinis showed no preference for living versus dead shrubs in
Montana, suggesting the structural characteristics the plants provided was more important than
any species-dependent characteristic (Best, 1972). Cumulatively these studies show that P. g.
confinis can be found in sparse, open habitats with scattered perches across Western North
America, and are particularly abundant in sagebrush steppe as this ecosystem is often
characterized by their preferred habitat structure.
Although P. g. confinis is not of immediate conservation concern, as a species Pooecetes
gramineus have been declining since the 1960s across North America (Altman, 2017; Sauer et
al., 2014). P. g. confinis is a common native of the grassland bird communities found across the
intermountain west, with an affection for sagebrush steppe habitat. Across North America
27

scientists are alarmed by losses in biodiversity, with avifauna experiencing population declines
over the last several decades. North America has seen the loss of nearly 3 billion birds, with
sparrows and grassland species experiencing some of the greatest loss in numbers (Rosenberg et
al., 2019). Conservation of grassland birds has been described as an “unfolding conservation
crisis” (Brennan & Kuvlesky Jr., 2005), and P. g. confinis is one of many species that make up
these communities. In Washington, the sagebrush steppe of the Columbia plateau is host to
several state endangered and threatened species such as the Columbian sharp-tailed grouse
(Tympanuchus phasianellus), pygmy rabbit (Brachylagus idahoensis), and ferruginous hawk
(Buteo Regalis). This region historically was dominated by sagebrush, and although sizable tracts
of native sagebrush steppe is still intact, about a quarter of the plateau’s land is used for
agriculture (Groves et al., 2000). While much of the remaining sagebrush communities occur on
publicly owned lands, very little (4%) has been set aside for biodiversity protection, and resource
extraction is allowed across most of it (Groves et al., 2000). So, while P. g. confinis is not
currently threatened, the species is likely declining along with grassland bird communities in
general, and especially in Washington, the sagebrush habitat on which it depends is increasingly
fragmented and degraded.
Song of the Vesper Sparrow

The song of the vesper sparrow was originally described in detail by Borror (1961), and
is generally characterized as a series of two to four introductory whistles followed by one to two
seconds of variable trills (Kroodsma, 1972; Rising, 1996; Sibley, 2003). Several authors have
noted that the repertoires of the vesper sparrow are highly variable and individualistic (Hing,
2014; Kroodsma, 1972; Ritchison, 1981). Kroodsma (1972) found that P. g. affinis had
remarkably large repertoires, noting that one individual sang 218 different song variations among
28

a sample of 400 songs. These large repertoire sizes may be important for sexual selection, similar
to other sparrow species (Hiebert et al., 1989). Kroodsma (1972) suggested that juvenile vesper
sparrows in the Willamette Valley learned songs from adults while on their natal grounds and
noted weak dialects in the introductory notes of P. g. affinis songs. In Minnesota, Ritchison
(1981) found no evidence of dialects among P. g. gramineus and attributed the differences from
Kroodsma’s (1972) findings to high dispersal rates related to occupancy of regularly disturbed
breeding habitats. Ritchison (1981) also noted very little sharing of song syllables between
individual vesper sparrows and no instances of shared song sequences. Hing (2014) found that in
Montana, male P. g. confinis did not share song characteristics among neighboring territories,
but with birds an intermediate distance away. Hing (2014) suggested these results could be
attributed to vesper sparrows using song complexity as a mechanism for mating selection, with
individual males attempting to differentiate themselves from their neighbors. These similarities
in songs of intermediate distances could be explained by fidelity to the general location of natal
sites. In published literature there have been no studies comparing songs between subspecies of
vesper sparrows, and the species has not been the subject of playback experiments.
Several descriptions of an extended flight song can also be found in the limited literature
on Pooecetes gramineus songs. Wells & Vickery (1994) described the extended flight song in
detail, and discussed some possible functions of this signal. During the 1993 breeding season, the
extended songs were mostly heard on one day in late July. Wells & Vickery suggested that the
extended flight song might serve the purpose of rounding up juvenile birds together, possibly as
a warning of nearby predators. This behavior was also mentioned by Burroughs (1905) who gave
a colorful description of the song:

29

“One summer, up in the Catskills, I added another name to my list of ecstatic
singers—that of the vesper sparrow. Several times I heard a new song in the air, and caught
a glimpse of the bird as it dropped back to the earth. My attention would be attracted by a
succession of hurried, chirping notes, followed by a brief burst of song, then by the
vanishing form of the bird. One day I was lucky enough to see the bird as it was rising to
its climax in the air, and to identify it as the vesper sparrow. The burst of song that crowned
the upward flight of seventy-five or one hundred feet was brief; but it was brilliant and
striking, and entirely unlike the leisurely chant of the bird while upon the ground. It
suggested a lark, but was less buzzing or humming. The preliminary chirping notes, uttered
faster and faster as the bird mounted in the air, were like the trail of sparks which a rocket
emits before its grand burst of color at the top of its flight.”
J. Burroughs, The Ways of Nature (1905).

Listing Considerations for the Oregon Vesper Sparrow
The subjects of this study were chosen because of the need for better understanding of the
subspecies boundaries for the rare and declining P. g. affinis. Controversy inevitably follows the
listing of a new species under the Endangered Species Act (Wilde, 2014), and with the ability to
list “species, subspecies, and distinct population segments” the distinction between significant
population segments and genetically distinct taxonomic units can be contentious (Haig & Elia,
2010). What constitutes a distinct population segment is left unclear by the Endangered Species
Act, and so it is left to taxonomists and conservationists to debate which species and populations
deserve legal protection and finite resources. While North American birds are some of the most
well studied in the world, there is still taxonomic uncertainty among many subspecies (Zink,
1996), with some listed species having relatively low genetic distinctiveness (Zink &
Barrowclough, 2008). Some argue for a holistic approach to species designation, taking into
account genetic, morphological, distributional and cultural factors (Alström et al., 2008).
The results of this study could have potential impacts on current land use taking place on
sites hosting remnant populations of P. g. affinis. Evidence supporting the designation P. g.
30

affinis as a distinct population segment of Pooecetes gramineus would strengthen the argument
that protection of this subspecies under the Endangered Species Act is warranted. In Washington,
nearly all remaining P. g. affinis occur on JBLM’s artillery impact area and Rainier training
areas. Listing of P. g. affinis could have the consequence of restricting training exercises on
JBLM, an indispensable military installation and part of the United States defense apparatus.
Several other listed species occur on JBLM, some of which share habitat with P. g. affinis, such
as the Mazama pocket gopher (Thomomys mazama), Taylor’s checkerspot butterfly (Euphydryas
editha taylori) and streaked horned lark (Eremophila alpestris strigata). The fact that P. g. affinis
breed within priority habitat of endangered species with training restrictions already in place
would mean that listing of the subspecies might not have a profound impact on military
exercises.
Conclusion
The songs of vesper sparrows are highly variable and individualistic (Kroodsma, 1972;
Ritchison, 1981), and therefore are difficult to discriminate by ear. Two western subspecies, P. g.
affinis and P. g. confinis are morphologically similar, impossible to differentiate at a distance and
even difficult up close (Rising, 1996). The degree to which Pacific cordillera boundary affects
speciation among vesper sparrows is unknown as wintering ranges for P. g. affinis and P. g.
confinis overlap in southern and central California (Altman, 2017; Erickson, 2008). Altman
(2017) also notes uncertainty regarding subspecies boundaries within the Klamath Mountains of
Oregon, and Pyle (1997) suggests intergradation with P. g. affinis among the P. g. confinis of the
Columbian Plateau. Furthermore, it is unclear to what degree differences in song culture act as a
boundary to gene flow between the two subspecies. While differences in habitat associations and
geographic distributions may not be adequate in themselves to constitute subspecies
31

designations, considering morphological and behavioral differences can provide a more robust
set of evidence in the absence of genetic analysis. To answer the question of whether vesper
sparrows distinguish individuals of another subspecies, we can ‘ask’ this question to the birds
themselves through playback experiments.

32

Chapter 3: Methods

Study Area
Oregon vesper sparrow
The last place in Washington where P. g. affinis breeds in numbers is on Joint Base
Lewis McChord (JBLM), an 86,000-acre military installation in the South Puget Lowlands of
Washington. The base encompasses the largest remaining patches of glacial outwash prairie and
oak-savannah habitat left in western Washington. The song recording and playback trials for P.
g. affinis took place within the two areas of the installation where P. g. affinis is known to breed:
the Artillery Impact Area (AIA) and the Rainier Training Areas (RTAs; Figure 3). Access to the
Artillery Impact Area is highly restricted due to the presence of unexploded ordnance, and work
in this area required support by an explosive ordnance disposal specialist. The Rainier Training
Areas have far less restrictions, so there was consistent access to that site throughout the
breeding season. Several pairs also breed on Tenalquot Preserve, a private nature reserve owned
and managed by the Center for Natural Lands Management. Tenalquot Preserve shares a border
with the southernmost portion of the RTAs, and the birds that breed there contribute to the RTA
population of P. g. affinis. Breeding territories were typically located in the ecotonal habitat
between prairie and Douglas fir forest.

33

Figure 3. P. g. affinis study sites within the South Puget Sound region of western Washington
State. Two training areas on Joint Base Lewis-McChord support the majority of breeding pairs of
P. g. affinis left in Washington.

34

Western vesper sparrow
East of the Cascade crest, P. g. confinis is a common and widespread breeding species
found in a variety of open upland habitats throughout the Columbia Basin. Study areas were
located using the vesper sparrow species map from eBird.org (eBird, 2019), and were selected
based on high frequencies of vesper sparrow observations, proximity to P. g. affinis study areas,
and public access. Recordings and playback trials with P. g. confinis as the focal subspecies
occurred on state and private land in Yakima, Kittitas, and Douglas counties. State public lands
included Wenas, Quilomene, and Whiskey Dick state wildlife areas, all located in the foothills
outside of Ellensburg, Washington (Figure 4). These wildlife areas are all co-owned and
managed by Washington Department of Fish and Wildlife and Washington Department of
Natural Resources. In one instance, a private landowner offered access to his property for song
recordings. Some song recordings also took place on publicly accessible portions of The Nature
Conservancy’s Beezley Hills Preserve outside of Ephrata, Washington. Habitat within the P. g.
confinis study areas can generally be described as sagebrush steppe, with a few sites in the
transitional zone to ponderosa pine (Pinus ponderosa) forest.

35

Figure 4. P. g. confinis study sites within the Columbia Basin region of eastern Washington
State. Playback trials were conducted at two WDFW-managed wildlife areas in 2020, with some
songs recorded at a Nature Conservancy-owned site near Ephrata.

36

Song Recordings
The songs that sourced the playback stimuli were recorded from unprovoked wild vesper
sparrows in the spring of 2019. P. g. affinis songs were recorded in the AIA and RTAs of JBLM
from April 29th until June 10th. Songs of P. g. confinis were recorded at Beezley Hills Preserve
and at Wenas Wildlife Area from May 11th until June 9th. Recording efforts were staggered
between subspecies, site, and spread out across the breeding season. Singing male vesper
sparrows were located by ear, approached within a distance of approximately 30 meters and
recorded singing for up to 15 minutes. Songs were recorded with 44.1 kHz sampling rate in 16bit wav format with an Olympus LS-100 Multi-track PCM recorder and a Sennheiser ME62
microphone with a parabolic reflector. Two stimuli used in the trials were sourced from a
colleague who recorded P. g. confinis songs at Wright’s Meadows in Klamath County, Oregon
during spring of 2018 using a Marantz PMD-660 solid-state recorder and Sennheiser ME62
microphone with a parabolic reflector. Those songs were also recorded in 16-bit wav format but
with a 48.0 kHz sampling rate.
Songs of both subspecies consistently matched descriptions by Sibley (2003) and Rising
(1996) of two to four introductory whistles followed by a series of 4-7 musical trills. While the
types of syllables are highly variable between individual vesper sparrows, they only have one
primary song-type. During bouts of singing, vesper sparrows will broadcast this primary songtype repeatedly, each time using a different variation of syllables. Kroodsma (1972) for example,
found 218 variations out of 400 songs analyzed from a single vesper sparrow.
Raven Pro (ver. 1.6.1) software was used to normalize amplitude of the song recordings
as well as remove background noise with band filters (Center for Conservation Bioacoustics,
2019). Songs were spaced out at 10-second intervals throughout each playback recording using
37

Audacity ® (2019) software. This spacing was based on the mean inter-song interval within the
raw song recordings of both subspecies. Playback stimuli were constructed using the maximum
number of possible song variations available from each source recording. Songs were edited into
six-minute wav files to be used in the field as playback stimuli. Thirty-six playback stimuli were
produced in total, 18 from each subspecies, with each recording sourced from a unique
individual to avoid pseudoreplication (Kroodsma, 1989).

38

P. g. affinis

Lower Weir – 01

Range 53 – 01

Upper Weir – 04

Range 74 – 01

Tenalquot – 02

Range 76 – 06

Figure 5. Spectrograms of six song exemplars, each from a unique individual P. g. affinis
occurring at different sites in the Puget Lowlands.

39

P. g. confinis

BBQ Horse Camp - 01

Beezley Hills – 01

Wenas – 01

Beezley Hills – 02

Wenas – 04

Beezley Hills – 05

Figure 6. Spectrograms of six song exemplars, each from a unique individual P. g. confinis
occurring at different sites in the Columbia Basin.

Playback Experiment
Playback trials for P. g. affinis were conducted on JBLM from May 14th through July
10th, 2020. Seventeen P. g. affinis subjects were exposed to song recordings from both
subspecies, with 6 occurring in the Artillery Impact Area (AIA) and 11 occurring within the
40

Rainier Training Areas. Most P. g. affinis subjects selected were color-banded, which ensured
that the same individual male was observed responding to both playback treatments. Effort was
made to stagger trials between sites throughout the season, but due to access restrictions in the
AIA, trials at this site occurred opportunistically. The minimum length of time between
treatments was one day, and the maximum length between treatments was 13 days.
Trials for P. g. confinis were conducted on Wenas, Quilomene, and Whiskey Dick
wildlife areas from May 24th through July 3rd, 2020. Sixteen P. g. confinis subjects were exposed
to playback treatments but two individuals were not relocated for the second treatment. Eight P.
g. confinis subjects were selected at Wenas wildlife area, seven at Whiskey Dick wildlife area,
and one at Quilomene wildlife area. None of the P. g. confinis subjects were color-banded and it
was assumed that subjects who responded to playback with territorial aggression were males. In
all of the P. g. confinis playback trials, the length of time between playback treatments was one
day. P. g. confinis trials were staggered between sites and spread out between May and July.
Each subject was exposed to two song recordings: one from their own subspecies and one
from a foreign subspecies, with the order both randomized and equally weighted among subjects.
Subjects were never exposed to stimuli sourced from the same site as the subject, which limited
the chance that a subject would be exposed to song of a bird it was familiar with (Temeles,
1992). Trials were aborted if a conspecific male approached the speaker, or if a predator was
observed within the trial area. Trials began at sunrise and continued until around noon. Subjects
were most often located by ear, especially in the P. g. confinis trials, but effort was made to
choose visually detected individuals as well. Subjects could not always be easily located upon
return for second trial, in which case playbacks would begin after a thorough search of the area.

41

In the majority of these cases, the subject was hiding nearby and would appear shortly after the
playback started.
Playback treatments were broadcast from a single UBL JFLIP 4 Bluetooth speakers
which played wav files through an auxiliary cord sourced from an iPod (Apple, Inc.). Stimuli
were broadcast at approximately 90 dB(A), calibrated using SPLnFTT Noise Meter version 7.0
(2020) cellphone application placed 1 meter from the speaker. The speaker was mounted on a
scope stand and placed at a height of two meters, based on the median and mode perch height
noted from vesper sparrows recorded in 2019 (n = 45). Each mounted speaker was placed next to
a preferred perch within a subject’s territory to maximize the likelihood of a territorial response.
Non-bright flagging was placed at 10 meters from the speaker in each cardinal direction. The
same speaker location was used during the second playback trial for each subject.
Each treatment began with brief, categorial observations during speaker set-up, followed
by the six-minute playback, and then another six minutes of silent observation resulting in 12
minutes of total observation for each treatment (Liu et al., 2008). Pre-playback observations
included noting whether the bird was singing, stationary, the bird’s initial proximity to the
speaker, and whether there were any conspecifics present. Response variables recorded were: 1)
number of songs (total count), 2) number of flights (total count), 3) minimum distance to speaker
(m), 4) time spent within 10 meters of the speaker (seconds), 5) whether the bird wing-waved
(binary), and 6) whether the bird performed a “soft” song (binary). Subjects were generally
observed from approximately 20-30 meters from the speaker. Minimum distance to the speaker
was defined as the closest distance the bird came to speaker in flight, perched, or on the ground.
Observers were blind to which subspecies song was being played, which cannot be reliably
distinguished by ear. Observers dictated the descriptions of each subject’s territorial response
42

before, during and after each playback trial onto an Olympus WS-852 digital voice recorder and
were transcribed at a later date.
Analysis
Eight territorial response variables to playback measured for each subject across all study
sites were reduced using principal component analysis (PCA; McGregor, 1992). A correlation
matrix was produced to confirm that the data met the assumptions of the analysis. For the pooled
dataset, the four variables, each collected in both treatment and post-treatment periods, were
reduced to three principal components. Linear mixed effects models were then created for the
three principal components using the parameters that were most likely to have influenced
territorial responses. Model ranking using Akaike’s Information Criterion corrected for small
sample size was then conducted for each principal component (AICc; Phillips & Derryberry,
2017).
The global model included playback stimuli type (coded as ‘consubspecific’ or
‘heterosubspecific’, based on the subspecies of the subject and the playback stimuli), subject
subspecies (P. g. confinis or P. g. affinis), subject’s initial distance to speaker (m), and Julian
date as fixed effects. The interaction between stimuli and subspecies was also included as a fixed
effect. Study site, observer, subject-ID, and stimuli-ID (the specific playback recording) were
designated as random effects (Greig, Baldassarre, & Webster, 2015). Random effects were
removed from the model if they did not account for any variance (Bates, Machler, Bolker, &
Walker, 2015). All models included subject-ID as a random effect, with the only other random
effect included being observer in the model with the third principal component axis as the
dependent variable. Variance inflation factors and quantile-quantile residual plots were examined
for each model to check for collinearity and residual normality.
43

The top models with ΔAICc < 2.0 for each principal component were examined for
parameters with strong influence. Model parameters with 95% confidence intervals that did not
overlap zero were considered to have a strong influence on vesper sparrow response behavior
(Reed et al., in press). All statistical analyses were conducted with R version 3.5.3 (R Core
Team, 2020) using the “psych” (Revelle, 2020), “lme4” (Bates et al., 2015), “car” (Fox &
Weisberg, 2019), and “MuMIn” (Barton, 2020) packages.

44

Chapter 4: Results & Discussion

Results
Principal component analysis of the data identified three components with eigenvalues
higher than one (Table 1). PC1approach was most heavily loaded with variables related to approach
distance, with the treatment and post-treatment variables “minimum distance to speaker” and
“number of seconds spent within 10 meters of the speaker” represented within this component.
PC2fly was most heavily loaded with treatment and post-treatment flight behavior, or “number of
flights”. PC3song was most heavily loaded with treatment and post-treatment vocal behavior or
“number of songs”. The cumulative percent of variation in the data accounted for by these three
principal components was 65.4%.
Table 1.
Loadings of the principal component’s axis 1, 2, and 3 scores. Based on the specific response
variables with the highest loadings for each PC axis (shown in bold), PC1 was designated the
‘Approach Behavior’ axis, PC2 the ‘Flight Behavior’ axis, and PC3 the ‘Song Behavior’ axis.
The cumulative proportion of variance explained by PC1 through PC3 was 65.4%.
Loadings
Response Variables

PC1approach

PC2fly

No. songs during treatment
No. flights during treatment

0.845
0.244

0.757

-0.747

-0.206

Time within 10m during treatment

0.788

0.172

No. songs post-treatment

0.215

Distance during treatment (m)

No. flights post-treatment
Distance post-treatment

PC3song

0.128
0.751

0.828
-0.717

-0.231

Time within 10m post-treatment

0.799

-0.254

0.193

Eigenvalue

2.441

1.457

1.334

Proportion of variance

0.305

0.182

0.167

45

Vesper sparrows approached within ten meters of the speaker in 90% of playback trials
and approached within one meter of the speaker in 48% of trials. While the subjects in this study
showed strong overall approach behavior (PC1approach), none of the parameters included in the
global model had a strong influence on PC1approach scores. When ranked by AICc (Table 2), the
best-supported model explaining PC1approach included treatment as the only fixed effect, with
‘consubspecific’ stimuli having a positive effect on PC1approach scores (i.e., shorter distances to
the playback during and post-treatment, and more time spent within 10m). The null model was
ranked second, with a similar weight as the top-model, and other top ranked models included
combinations of treatment and subject’s initial distance as fixed effects. All of the fixed effects
among the top-ranked (Δ AICc < 2.00) PC1approach models had 95% confidence intervals that
overlapped zero, including the treatment (Table 3). Among the raw response variables related to
approach behavior (Table 4), both subspecies averaged closer distances and more time spent
within 10 meters to consubspecific stimuli, although this was less pronounced among P. g.
confinis.
Vesper sparrows responded with increased flight behavior (PC2fly) when exposed with
consubspecific playback stimuli. Among the two subspecies, P. g. confinis subjects responded
with more flights overall than P. g. affinis subjects. The best-supported PC2fly model included
treatment, subject subspecies, and Julian date, with all three parameters having a 95% CI that did
not overlap zero (Table 3). The second-ranked model included treatment and subject subspecies
only. Principal component scores for PC2fly were strongly influenced by both the treatment and
which subspecies was the subject of the playback trial (Figure 7). Julian date also had a strong
effect on flight behavior, with vesper sparrows responding with reduced flight behavior later in
the breeding season.

46

Table 2.
Model selection results using Akaike’s Information Criterion corrected for small sample size
(AICc). Principal components related to vesper sparrow approach, flight, and song behavior
written as PC1approach, PC2fly, and PC3song. Models with a delta value less than 2.0 (ΔAICc ≤ 2.00)
are shown, along with the null (intercept-only). K equals the number of model parameters, log(ℒ)
equals the maximized log-likelihood value, Δ equals delta (the change in AICc from the top
model), and wi equals the Akaike weight for each well-supported model. Bold indicates
parameters with 95% CIs that do not overlap zero. The direction of influence for each parameter
is indicated with positive/negative signs (+/-). Null (intercept-only) subscripts indicate the
random effects included in all models for each principal component, where ID = Bird ID and
Obs = observer. n = 63 trials.
K

log( )

AICc

Δ

wi

Treatment (+)

4

-81.30

171.29

0.00

0.35

NullID

3

-82.63

171.67

0.38

0.29

Treatment (+), Distance (-)

5

-80.56

172.18

0.89

0.22

Distance (-)

4

-82.21

173.10

1.81

0.14

Subspecies (+), Treatment (+), Julian date (-)

6

-77.94

169.38

0.00

0.71

Subspecies (+), Treatment (+)

5

-80.09

171.23

1.85

0.29

NullID

3

-86.18

178.78

9.40

-

Subspecies (-), Julian date (-)

6

-77.28

168.06

0.00

0.72

Subspecies (-), Julian date (-), Distance (-)

7

-76.97

169.97

1.88

0.28

NullID+Obs

4

-83.95

176.58

8.52

-

Model
PC1approach

PC2fly

PC3song

Although the subspecific treatment did not have a strong influence on vesper sparrow
song behavior (PC3song), the subject’s subspecies did have a strong effect. P. g. affinis subjects
responded with more songs overall than P. g. confinis subjects. The best-supported PC3song
model included subject subspecies and Julian date as fixed effects (Table 2). The second-ranked
model included these parameters with the addition of subject’s initial distance to speaker.

47

Table 3.
The influence of fixed effects from the top-ranked models for vesper sparrow approach, flight,
and song behavior. Principal components explaining approach, flight, and song behavior denoted
PC1approach, PC2fly, and PC3song. Parameter levels for treatment stimuli (consubspecific,
heterosubspecific) and subspecies (P. g. affinis, P. g. confinis) shown in parentheses. Intercept
and categorical level of reference condition (parentheses) included for each component. Strong
effects with 95% CIs that do not overlap zero are indicated in bold.
β  S

Lower CI

Upper CI

Intercept (Heterosubspecific)

-0.13 ± 0.17

-0.47

0.21

Treatment (Consubspecific)

0.26 ± 0.16

-0.04

0.57

Intercept (Heterosubspecific, P. g. affinis)

2.35 ± 1.40

-0.38

5.08

Treatment (Consubspecific)

0.44 ± 0.17

0.1

0.78

Subspecies (P. g. confinis)

0.68 ± 0.25

0.18

1.18

Julian date

-0.02 ± 0.01

-0.03

-0.001

Intercept (P. g. affinis)

3.80 ± 1.25

1.34

6.27

Subspecies (P. g. confinis)

-0.73 ± 0.23

-1.20

-0.28

Julian date

-0.02 ± 0.01

-0.03

-0.005

Parameter (level)
PC1approach

PC2fly

PC3song

Playback stimulus type did not have a strong effect on song behavior among either P. g. affinis
or P. g. confinis, with the treatment fixed effect not included in any of the models with ΔAICc <
2.00 (Table 2). Subject subspecies and Julian date were the only parameters in any of the top
ranked models with 95% confidence intervals that did not overlap zero (Table 3). Similar to
vesper sparrow flight behavior, Julian date had a strong influence on the response in song
behavior, with reduced singing later in the season.

48

Figure 7. Plots of parameter effects on the second principal component explaining flight
behavior (PC2fly). Plot (A) shows mean PC2fly values (with 95% CI) for playback trials separated
by subspecific stimuli type, with ‘consubspecific’ stimuli shown on the left, and
‘heterosubspecific’ stimuli on the right. Plot (B) shows mean PC2fly values (with 95% CI) for
playback trials separated by subject subspecies, with ‘P. g. affinis’ shown on left, and ‘P. g.
confinis’ on the right.

49

P. g. affinis wing-waved during 29.4 % of trials when exposed to con-subspecific stimuli,
versus 12.5% of trials when exposed to hetero-subspecific stimuli. P. g. confinis wing-waved
during 26.7% of trials when exposed to con-subspecific stimuli, and during 26.7 % of trials when
exposed to hetero-subspecific stimuli. P. g. affinis responded with soft-singing to 86.9 % of trials
with con-subspecific stimuli versus 56.2 % of trials with hetero-subspecific stimuli. P. g. confinis
responded with soft-singing to 86.7 % of trials with con-subspecific stimuli versus 66.7 % of
trials with hetero-subspecific stimuli.

Figure 8. Plot of parameter effects on the third principal component explaining song behavior
(PC3song). Plot (A) shows mean PC3song values (with 95% CI) for playback trials separated by
subject subspecies, with ‘P. g. affinis’ shown on left, and ‘P. g. confinis’ on the right.

50

Table 4.
Means and standard deviations for each raw response variable collected during playback trials,
separated by treatment stimuli type.
Raw response (mean ± SD)
Response Variables

Consubspecific

Heterosubspecific

No. songs during treatment

25.1 ± 12.1

24.6 ± 9.2

No. flights during treatment

7.8 ± 5.7

5.8 ± 5.3

Distance during treatment (m)

3.4 ± 5.9

7.3 ± 11.8

177.1 ± 106.2

154.2 ± 128.6

No. songs post-treatment

29.4 ± 12.0

25.2 ± 10.0

No. flights post-treatment

2.8 ± 2.9

2.0 ± 2.8

Distance post-treatment

5.6 ± 5.8

15.2 ± 25.6

196.7 ± 152.3

144.9 ± 161.8

No. songs during treatment

15.7 ± 9.9

14.9 ± 10.9

No. flights during treatment

11.8 ± 6.8

9.1 ± 5.5

Distance during treatment (m)

3.1 ± 5.2

3.8 ± 6.7

Time within 10m during treatment

168.1 ± 86.8

163.3 ± 85.4

No. songs post-treatment

20.3 ± 11.0

20.1 ± 12.8

No. flights post-treatment

4.2 ± 3.1

3.1 ± 1.7

Distance post-treatment

7.9 ± 6.5

12.6 ± 12.2

85.8 ± 126.9

112.2 ± 121.3

P. g. affinis

Time within 10m during treatment

Time within 10m post-treatment
P. g. confinis

Time within 10m post-treatment

51

Discussion

Results from this experiment suggest the existence of song discrimination between the
two subspecies of vesper sparrow, P. g. affinis and P. g. confinis. The strength of this response
appeared to be moderate when compared to similar studies investigating passerine song
discrimination at the subspecies level (Lipshutz et al., 2017; Liu et al., 2008). Based on these
results, discriminatory response to subspecific stimuli among both subspecies was limited to
flight behavior, with a lack of strong treatment effect among approach and song behavior. This
study also identified several differences in territorial response behavior between the two
subspecies, with P. g. confinis responding to playback with more flights, and P. g. affinis
responding to playback with more songs.
Nearly all playback trials that were attempted garnered responses from subjects, with
birds singing, approaching, and making short flights around the speaker. The vesper sparrows in
this study would approach the speaker by flying or walking along areas of unvegetated bare
ground. Vesper sparrows would most often sing from perches, but would also frequently sing
from the ground, which contradicted some previous descriptions (Castrale, 1983). In a portion of
the trials, subjects wing-waved or performed soft-singing. Although measures of latency were
not analyzed, in some trials subjects would approach the speaker immediately after the first song
was broadcast. The average distance of subjects prior to playback was 44.0 ± 21.6 meters. The
broadcast of a song by the speaker would be immediately followed by a flight from some
subjects, or by a song in others. Subjects would often fly from perch to perch within the playback
area, usually making a close pass at the speaker in response to broadcast of a song. Many

52

subjects would remain near the speaker and continue singing after the six-minute playback
treatment had ended.
Searcy, Anderson and Nowicki (2006) identified distance to playback speaker as the most
reliable measure of territorial response to bird song playback when compared to other response
behaviors. In this study, principal component analysis identified PC1approach as accounting for the
highest proportion of variance (30.5%) in the data and was heavily loaded with variables related
to approach distance to playback speaker (Table 1). Of the response variables recorded during
the playback trials, four were related to the subject’s approach distance to the speaker as opposed
to two for flight behavior and two for song behavior. When examining parameters of the bestsupported PC1approach models, none were shown to have a strong influence on approach behavior.
While the influence of the treatment on approach behavior was not significant, review of raw
response averages suggests a weak treatment effect (Table 4). Among response variables related
to approach behavior, subjects averaged closer approach distances and more time spent within
ten meters of the speaker when exposed to consubspecific stimuli, although this pattern is mainly
present in the P. g. affinis response variables.
Loadings for PC2fly were greatest for “number of flights during treatment” and “number
of flights post-treatment” (Table 1), which suggested that this component was related to flight
behavior among subjects. Flight behavior was the only response type in which vesper sparrows
showed strong discrimination between subspecific stimuli. Many of the flights during the
playback trials were associated with close passes at the speaker, which could be interpreted as an
attempt at physical confrontation with a perceived intruder. Therefore, subjects expended more
energy towards physical confrontation with a perceived intruder when the subject detected an
intruder of the same subspecies. Some researchers have used taxidermic mounts during playback
53

experiments and studied the territorial response of subjects when a visual target is presented
(Greig et al., 2015; Searcy et al., 2006). Utilizing taxidermic mounts in future playback
experiments among vesper sparrows might help determine whether flight behavior is in fact
related to aggressive intent.
Vesper sparrow song behavior was not influenced by the treatment, but by the subject’s
subspecies and Julian date. P. g. affinis subjects responded to playback with significantly more
songs than P. g. confinis subjects (Figure 8), reverse of the subject’s subspecies effect on flight
behavior. PC3song was primarily loaded with the “number of songs during treatment” and
“number of songs post-treatment” variables (Table 1), which indicated that this component was
representative of vesper sparrow song behavior. Julian date had a strong influence on vesper
sparrow song behavior, with subjects responding to playback with less singing as the season
progressed. Near the end of the sampling period, vesper sparrows in Eastern Washington had
become noticeably less vocal, which was likely due to breeding phenology, and possibly changes
in weather (it was becoming hotter and drier).
Local versus non-local effects were controlled for in this experiment by only exposing
subjects to stimuli sourced from a different study area. Each subject would hear a song of their
own subspecies, but never from the specific population it was a part of. However, the P. g. affinis
study areas were in closer proximity than the P. g. confinis study areas, and therefore there is a
possibility of a mild local effect among the P. g. affinis trials. The P. g. confinis trials had a
limited sampling period compared to the P. g. affinis trials. Visits to the Eastern Washington
study sites only occurred on weekends, while on JBLM trials occurred throughout the week. For
this reason, all P. g. confinis subjects received both playback treatments separated by only 24
hours. Some researchers have recommended a minimum of 48 hours between playback

54

treatments in their studies, but due to time and access constraints subjects were visited on
subsequent days.
Differences in flight and song behavior between P. g. affinis and P. g. confinis could be
attributed to either biological or ecological factors. While both vesper sparrow subspecies are
morphologically similar, their habitats and some aspects of their life histories are distinct. In the
Quilomene and Wenas study areas, vesper sparrows were among the most common passerine
species. On JBLM vesper sparrows are not abundant, although territories tend to be clustered
within suitable habitat. In Washington, P. g. confinis generally migrate to their breeding sites a
few weeks earlier than P. g. affinis, and as a result, finish their breeding season earlier as well
(Jones and Cornely, 2002). P. g. confinis breeding sites also differ in climate, with the study
areas in the Columbia Basin being much drier than the P. g. affinis study areas on JBLM.
Qualitatively, the Columbia Basin study areas appeared more abundant in perches and could be
described as shrublands whereas the sites in the Puget Lowlands are more accurately
characterized as ‘prairies’, as they are dominated by grasses and forbs with low numbers
scattered conifers and small shrubs providing perches. The abundance of perches among P. g.
confinis study areas could have possibly contributed to the increased flight behavior among that
subspecies. More perches might allow subjects to change location more frequently as they
searched for the perceived intruder, while less perches could force the subjects to remain on the
limited perches within their territory.
Approach behavior, often considered a reliable measure territorial response, was not
strongly influenced by the subspecific treatment. However, flight behavior was strongly
influenced by the treatment, which could be interpreted as energy invested towards attempts at
aggression. The treatment also appeared to influence the performance of soft songs and wing

55

waves among subjects, though this was less pronounced among P. g. confinis. Searcy, Anderson
and Nowicki (2006) found that along with distance to playback, soft singing provided another
reliable measure of aggressive intent. Regrettably, the binary variables that noted instances of
soft-singing and wing-waving were not included in the main analysis of this study.
Overall, the results of this experiment support Jones and Cornely’s (2020) description of
vesper sparrow subspecies taxonomy as “moderately distinct”. Both subspecies showed some
degree of discrimination between subspecific song stimuli, although this was mainly limited to
flight behavior. P. g. confinis responded to playback with more flights than P. g. affinis overall.
Differences in singing behavior were also significant between the two subspecies, with P. g.
affinis responding to playback with more singing than P. g. confinis. It is unclear whether the
differences between the subspecies are biological, or due to temporal and environmental
conditions during the limited sampling period. Limitations in time and site access undoubtedly
had influence on the results of this study, but overall, there appears to be a moderate trend of
subspecific discrimination.

56

References

Abs, M. (1963). Essential Components of European Nightjar Song. Proc. Int. Ornithol. Congr.,
13(1), 202–205. https://doi.org/10.1093/auk/94.3.615
Akçay, Ç., Tom, M. E., Holmes, D., Campbell, S. E., & Beecher, M. D. (2011). Sing Softly and
Carry a Big Stick: Signals of Aggressive Intent in the Song Sparrow. Animal Behaviour,
82(2), 377–382. https://doi.org/10.1016/j.anbehav.2011.05.016
Alström, P., & Olsson, U. (1999). The Golden-Spectacled Warbler: a Complex of Sibling
Species, Including a Previously Undescribed Species. Ibis, 141, 545–568.
https://doi.org/10.1111/j.1474-919x.1999.tb07363.x
Alström, P., Rasmussen, P. C., Olsson, U., & Sundberg, P. (2008). Species Delimitation Based
on Multiple Criteria: The Spotted Bush Warbler Bradypterus thoracicus Complex (Aves:
Megaluridae). Zoological Journal of the Linnean Society, 154, 291–307.
https://doi.org/10.1111/j.1096-3642.2008.00418.x
Altman, B. (2011). Historical and Current Distribution and Populations of Bird Species in
Prairie-Oak Habitats in the Pacific Northwest. Northwest Science, 85(2), 194–222.
https://doi.org/10.3955/046.085.0210
Altman, B. (2017). Conservation Assessment for Oregon Vesper Sparrow (Pooecetes gramineus
affinis). Retrieved from https://cascadiaprairieoak.org/documents/conservationassessment-for-oregon-vesper-sparrow-pooecetes-gramineus-affinis
Altman, B., Stinson, D., & Hayes, G. (2020). Status Report for the Oregon Vesper Sparrow.
(February), 31 + iii pp.
57

American Bird Conservancy. (2016). Petition to List Oregon Vesper Sparrow (Pooecetes
gramineus affinis) as Endangered or Threatened. Retrieved from https://abcbirds.org/wpcontent/uploads/2016/.../Oregon-Vesper-Sparrow-Petition.pdf
American Ornithologists’ Union. (1957). Checklist of North American Birds (5th Editio).
Baltimore: American Ornithologists’ Union.
Baker, M. C., & Cunningham, M. A. (1985). The Biology of Bird-Song Dialects. Behavioral and
Brain Sciences, 8(1), 85–133. https://doi.org/10.1017/S0140525X00019932
Ballentine, B., Hyman, J., & Nowicki, S. (2004). Vocal Performance Influences Female
Response to Male Bird Song: An Experimental Test. Behavioral Ecology, 15(1), 163–
168. https://doi.org/10.1093/beheco/arg090
Baptista, L. F., & Gaunt, S. L. L. (1994). Advances in Studies of Avian Sound Communication.
The Condor, 96(3), 817–830.
Baptista, L. F., & Petrinovich, L. (1986). Song Development in the White-crowned Sparrow:
Social Factors and Sex Differences. Animal Behaviour, 34(5), 1359-1371.
Barker, F. K., Burns, K. J., Klicka, J., Lanyon, S. M., & Lovette, I. J. (2015). New Insights into
New World Biogeography: An Integrated View from the Phylogeny of Blackbirds,
Cardinals, Sparrows, Tanagers, Warblers, and Allies. The Auk, 132(2), 333–348.
https://doi.org/10.1642/auk-14-110.1
Barton, K. (2020). MuMIn: Multi-Model Inference. Retrieved from https://cran.rproject.org/package=MuMIn

58

Bates, D., Machler, M., Bolker, B., & Walker, S. (2015). Fitting Linear Mixed-Effects Models
Using lme4. Journal of Statistical Software, 67(1), 1–48.
https://doi.org/10.18637/jss.v067.i01
Beauchesne, S. M. (2002). Coastal Vesper Sparrow Stewardship Account For the Garry Oak
Ecosystems of Southwestern British Columbia. Retrieved from
http://goert.ca/documents/SA_pooegram.pdf
Behle, W. H. (1978). Avian Biogeography of the Great Basin and Intermountain Region. Great
Basin Naturalist Memoirs, 2, 55–80.
Berger, A. J. (1968). Eastern Vesper Sparrow. In O. L. Austin Jr. (Ed.), Life Histories of North
American Cardinals, Grosbeaks, Buntings, Towhees, Finches, Sparrows, and Allies (pp.
868–882).
Best, L. B. (1972). First-Year Effects of Sagebrush Control on Two Sparrows. The Journal of
Wildlife Management, 36(2), 534–544.
Best, L. B., & Rodenhouse, N. L. (1984). Territory Preference of Vesper Sparrows in Cropland.
The Wilson Bulletin, 96(1), 72–82.
Borror, D. J. (1961). Songs of Finches (Fringillidae) of Eastern North America. The Ohio
Journal of Science, 61, 161–174.
Boyd, R. (1999). Introduction. In R. Boyd (Ed.), Indians, Fire, and the Land in the Pacific
Northwest (pp. 1–30). Corvallis, Oregon: Oregon State University Press.
Brennan, L. A., & Kuvlesky Jr., W. P. . (2005). North American Grassland Birds : An Unfolding
Conservation Crisis? Journal of Wildlife Management, 69(1), 1–13.
59

Brooks, W., & Wimberger, A. P. (2018). The Role of Song in Reproductive Isolation in a New
Secondary Contact Zone of White-crowned Sparrows (Zonotrichia leucophrys).
University of Puget Sound. Retrieved from:
https://soundideas.pugetsound.edu/summer_research/304
Burroughs, J. (1905). The Ways of Nature. Retrieved from
https://cdm16021.contentdm.oclc.org/digital/collection/p16021coll3/id/115
Camp, M., & Best, L. B. (1993). Bird Abundance and Species Richness in Roadsides Adjacent to
Iowa Rowcrop Fields. Wildlife Society Bulletin, 21(3), 315–325.
Castrale, J. S. (1983). Selection of Song Perches by Sagebrush-Grassland Birds. The Wilson
Bulletin, 95(4), 647–655.
Center for Conservation Bioacoustics. (2019). Raven Pro: Interactive Sound Analysis Software
(Version 1.6.1). Retrieved from http://ravensoundsoftware.com/
Chappell, C B, & Crawford, R. C. (1997). Native Vegetation of the South Puget Sound Prairie
Landscape. Ecology and Conservation of the South Puget Sound Prairie Vegetation, 107–
124. Retrieved from http://w.southsoundprairies.org/tech/Native Veg of S Puget Sound
Prairies.pdf
Chappell, C. B., Gee, M. S., Stephens, B., Crawford, R., & Farone, S. (2001). Distribution and
Decline of Native Grasslands and Oak Woodlands in the Puget Lowlands and Willamette
Valley Ecoregions, Washington. In S. H. Reichard, P. W. Dunwiddie, J. G. Gamon, A. R.
Kruckeberg, & D. L. Salstrom (Eds.), Conservation of Washington’s Native Plants and
Ecosystems (pp. 124–139).

60

Collins, S. (2004). Vocal Fighting and Flirting: The Functions of Birdsong. In P. Marler & H.
Slabbekoorn (Eds.), Nature’s Music: The Science of Birdsong (pp. 39–78).
Curnutt, J. L., Pimm, S. L., Maurer, B. A., Curnutt, J. L., Pimm, S. L., & Maurer, B. A. (1996).
Population Variability of Sparrows in Space and Time. Oikos, 76(1), 131–144.
Dechant, J. A., Dinkins, M. F., Johnson, D. H., Igl, L. D., Goldade, C. M., & Euliss, B. R.
(2002). Effects of Management Practices on Grassland Birds: Vesper Sparrow. Northern
Prairie Wildlife Research Center, Jamestown ND, pp 1–25.
Demko, A. D., Sosa-López, J. R., & Mennill, D. J. (2019). Subspecies Discrimination on the
Basis of Acoustic Signals: A Playback Experiment in a Neotropical Songbird. Animal
Behaviour, 157, 77–85. https://doi.org/10.1016/j.anbehav.2019.08.021
Dennehy, C., Alverson, E. R., Anderson, H. E., Clements, D. R., Gilbert, R., & Kaye, T. N.
(2011). Management Strategies for Invasive Plants in Pacific Northwest Prairies,
Savannas, and Oak Woodlands. Northwest Science, 85(2), 329–351.
https://doi.org/10.3955/046.085.0219
Derryberry, E. P. (2011). Male Response to Historical and Geographical Variation in Bird Song.
Biology Letters, 7(1), 57–59. https://doi.org/10.1098/rsbl.2010.0519
Derryberry, E. P., Seddon, N., Derryberry, G. E., Claramunt, S., Seeholzer, G. F., Brumfield, R.
T., & Tobias, J. A. (2018). Ecological drivers of song evolution in birds: Disentangling
the effects of habitat and morphology. Ecology and Evolution, 8(3), 1890–1905.
https://doi.org/10.1002/ece3.3760

61

Deur, D., & Turner, N. J. (Eds.). (2005). Keeping it Living: Traditions of Plant Use and
Cultivation on the Northwest Coast of North America. Seattle, WA: University of
Washington Press.
Dilger, W. C. (1956). Hostile Behavior and Reproductive Isolating Mechanisms in the Avian
Genera Catharus and Hylocichla. The Auk, 73(3), 313–353.
https://doi.org/10.1525/auk.2008.61008
eBird. (2019). Vesper Sparrow (Pooecetes gramineus) - Species Map - eBird. Retrieved October
23, 2019, from https://www.ebird.org
Erickson, R. A. (2008). Oregon Vesper Sparrow (Pooecetes gramineus affinis). In W. D. Shuford
& T. Gardali (Eds.), Studies of Western Birds (Vol. 1, pp. 377–381).
Falls, J. B. (1992). Playback: A Historical Perspective. Playback and Studies of Animal
Communication, 11–33. https://doi.org/10.1007/978-1-4757-6203-7_2
Finzel, J. E. (1964). Avian Populations of Four Herbaceous Communities in Southeastern
Wyoming. The Condor, 66(6), 496–510. https://doi.org/10.2307/1365226
Foster, J. R., & Shaff, S. E. (2003). Forest Colonization of Puget Lowland Grasslands. Northwest
Science, 77, 283–296.
Fox, J. & Weisberg, S. (2019). An {R} Companion to Applied Regression, Third Edition.
Thousand Oaks, CA.
Freeman, B. G., & Montgomery, G. A. (2017). Using Song Playback Experiments to Measure
Species Recognition Between Geographically Isolated Populations: A Comparison with
Acoustic Trait Analyses. The Auk, 134(4), 857–870. https://doi.org/10.1642/auk-17-63.1
62

Greig, E. I., Baldassarre, D. T., & Webster, M. S. (2015). Differential Rates of Phenotypic
Introgression are Associated with Male Behavioral Responses to Multiple Signals.
Evolution, 69(10), 2602–2612. https://doi.org/10.1111/evo.12756
Groves, C. R., Kutner, L. S., Storms, D. M., Murray, M. P., Scott, J. M., Schafale, M., …
Pressey, R. L. (2000). Owning Up to Our Responsibilities: Who Owns Lands Important
for Biodiversity? In B. A. Stein, L. S. Kutner, & J. S. Adams (Eds.), Precious Heritage:
The Status of Biodiversity in the United States (pp. 275–300). New York, New York:
Oxford University Press.
Haig, S. M., & Elia, J. D. (2010). Avian Subspecies and the U.S. Endangered Species Act.
Ornithological Monographs, 67(67), 24–34. https://doi.org/10.1525/om.2010.67.1.24.24
Harrison, K. G. (1974). Aspects of Habitat Selection in Grassland Birds. Master’s Thesis,
Western Michigan University. Retrieved from
https://scholarworks.wmich.edu/cgi/viewcontent.cgi?article=3555&context=masters_thes
es
Harrison, K. G. (1977). Perch Height Selection of Grassland Birds. The Wilson Bulletin, 89(3),
486–487.
Hiebert, S. M., Stoddard, P. K., & Arcese, P. (1989). Repertoire Size, Territory Acquisition and
Reproductive Success in the Song Sparrow. Animal Behaviour, 37, 266–273.
https://doi.org/10.1016/0003-3472(89)90115-2
Hill, G. E. (1990). Female House Finches Prefer Colourful Males: Sexual Selection for a
Condition-Dependent Trait. Animal Behaviour, Vol. 40, pp. 563–572.
https://doi.org/10.1136/vr.143.6.169
63

Hill, K., Kronland, B., & Martin, A. (2017). Fire Effects Monitoring Joint Base Lewis-McChord
Annual Report. Retrieved from https://cascadiaprairieoak.org/documents/fire-effectsmonitoring-joint-base-lewis-mcchord-annual-report
Hinde, R. A. (1956). The Biological Significance of the Territories of Birds. Ibis, 98(3), 340–
369. https://doi.org/10.1111/j.1474-919X.1956.tb01419.x
Hinde, R. A. (1958). Alternative Motor Patterns in Chaffinch Song. Animal Behaviour, 6(3–4),
211–218. https://doi.org/10.1016/0003-3472(58)90053-8
Hing, J. S. (2014). Examining Vesper Sparrow (Pooecetes gramineus) Songs on the National
Bison Range in Relation to Inter-individual Distances. Retrieved from
https://underc.nd.edu/assets/174532/fullsize/hing_final_paper.pdf
Illes, A. E., Hall, M. L., & Vehrencamp, S. L. (2006). Vocal Performance Influences Male
Receiver Response in the Banded Wren. Proceedings of the Royal Society B: Biological
Sciences, 273(1596), 1907–1912. https://doi.org/10.1098/rspb.2006.3535
Jewett, S. G., Taylor, W. P., Shaw, W. T., & Aldrich, J. W. (1953). The Birds of Washington
State. Seattle, WA: University of Washington Press.
Johannessen, C. L., Davenport, W. A., Millet, A., & Mcwilliams, S. (1971). The Vegetation of
the Willamette Valley. Annals of the Association of American Geographers, 61(2), 286–
302. https://doi.org/10.1111/j.1467-8306.1971.tb00783.x
Johnson, A. B., & Dickerman, R. W. (2006). Distribution of the Mountain Vesper Sparrow
(Pooecetes gramineus altus) Phillips 1964. NMOS Bulletin, 34(1–2), 11.

64

Jones, S. L., & Cornely, J. E. (2002). Vesper Sparrow (Pooecetes gramineus) (A. Poole & F.
Gill, Eds.). https://doi.org/10.2173/bna.624
Karin, B. R., Cicero, C., Koo, M. S., & Bowie, R. C. K. (2018). The Role of History and
Ecology as Drivers of Song Divergence in Bell’s and Sagebrush Sparrows
(Artemisiospiza, Aves: Passerellidae). Biological Journal of the Linnean Society, 125(2),
421–440. https://doi.org/10.1093/BIOLINNEAN/BLY090
Kenyon, H. L., Alcaide, M., Toews, D. P. L., & Irwin, D. E. (2017). Cultural Isolation is Greater
than Genetic Isolation Across an Avian Hybrid Zone. Journal of Evolutionary Biology,
30(1), 81–95. https://doi.org/10.1111/jeb.12989
King, J. R. (1968a). Oregon Vesper Sparrow. In O. L. Austin Jr. (Ed.), Histories of North
American Cardinals, Grosbeaks, Buntings, Towhees, Finches, Sparrows, and Allies. (pp.
884–886).
King, J. R. (1968b). Western Vesper Sparrow. In O. L. Austin Jr. (Ed.), Histories of North
American Cardinals, Grosbeaks, Buntings, Towhees, Finches, Sparrows, and Allies. (pp.
882–884).
Kolesnikova, Y., Liu, M., Kang, Z., & Opaev, A. (2019). Song Does Not Function as a Signal of
Direct Aggression in Two Leaf-Warbler Species. Ornithological Science, 18(1), 17.
https://doi.org/10.2326/osj.18.17
Kronland, B., Hill, K., & Martin, A. (2018). Prairie Habitat Enhancement Joint Base LewisMcChord 2017 Annual Report.

65

Kroodsma, D. E. (1972). Variations in Songs of Vesper Sparrows in Oregon. The Wilson
Bulletin, 84(2), 173–178.
Kroodsma, D. E. (1986). Design of Song Playback Experiments. The Auk, 103(3), 640–642.
Kroodsma, D. E. (1989). Suggested Experimental Designs for Song Playbacks. Animal
Behaviour, 37(PART 4), 600–609. https://doi.org/10.1016/0003-3472(89)90039-0
Kroodsma, D. E., Byers, B. E., Goodale, E., Johnson, S., & Liu, W. C. (2001). Pseudoreplication
in Playback Experiments, Revisited a Decade Later. Animal Behaviour, 61(5), 1029–
1033. https://doi.org/10.1006/anbe.2000.1676
Kroodsma, D. E., & Konishi, M. (1991). A Suboscine Brd (Eastern Phoebe, Sayornis phoebe)
Develops Normal Song without Auditory Feedback. Animal Behaviour, 42(3), 477–487.
https://doi.org/10.1016/S0003-3472(05)80047-8
Krosby, M., & Rohwer, S. (2010). Ongoing Movement of the Hermit Warbler X Townsend’s
Warbler Hybrid Zone. PLoS ONE, 5(11). https://doi.org/10.1371/journal.pone.0014164
Lipshutz, S. E., Overcast, I. A., Hickerson, M. J., Brumfield, R. T., & Derryberry, E. P. (2017).
Behavioural Response to Song and Genetic Divergence in Two Subspecies of Whitecrowned Sparrows (Zonotrichia leucophrys). Molecular Ecology, 26(11), 3011–3027.
https://doi.org/10.1111/mec.14002
Liu, I. A., Lohr, B., Olsen, B., & Greenburg, R. (2008). Macrogeographic Vocal Variation in
Subspecies of Swamp Sparrow. The Condor, 110(1), 102–109.
https://doi.org/10.1525/cond.2008.110.1.102

66

Luther, D., & Magnotti, J. (2014). Can Animals Detect Differences in Vocalizations Adjusted for
Anthropogenic Noise? Animal Behaviour, 92, 111–116.
https://doi.org/10.1016/j.anbehav.2014.03.033
Marler, P. (2004). Bird Calls: Their Potential for Behavioral Neurobiology. Annals of the New
York Academy of Sciences, 1016, 31–44. https://doi.org/10.1196/annals.1298.034
Marler, P., & Tamura, M. (1964). Culturally Transmitted Patterns of Vocal Behavior in
Sparrows. Science, 146(3650), 1483–1486.
https://doi.org/10.1126/science.146.3650.1483
Mason, N. A., Burns, K. J., Tobias, J. A., Claramunt, S., Seddon, N., & Derryberry, E. P. (2017).
Song Evolution, Speciation, and Vocal Learning in Passerine Birds. Evolution, 71(3),
786–796. https://doi.org/10.1111/evo.13159
Matessi, G., Dabelsteen, T., & Pilastro, A. (2001). Subspecies Song Discrimination in a
Mediterranean Population of the Reed Bunting Emberiza schoeniclus intermedia. Italian
Journal of Zoology, 68(4), 311–314. https://doi.org/10.1080/11250000109356424
Mcgregor, P. K. (2000). Playback Experiments : Design and Analysis. Acta Ethologica, 3, 3–8.
https://doi.org/10.1007/s102110000023
McGregor, P. K. (1992). Quantifying Responses to Playback: One, Many, or Composite
Multivariate Measures? Playback and Studies of Animal Communication, 79–96.
https://doi.org/10.1007/978-1-4757-6203-7_6

67

Moseley, D. L., Lahti, D. C., & Podos, J. (2013). Responses to Song Playback Vary with the
Vocal Performance of both Signal Senders and Receivers. Proceedings of the Royal
Society B: Biological Sciences, 280(1768). https://doi.org/10.1098/rspb.2013.1401
Moseley, D. L., Phillips, J. N., Derryberry, E. P., & Luther, D. A. (2019). Evidence for Differing
Trajectories of Songs in Urban and Rural Populations. Behavioral Ecology, 1–9.
https://doi.org/10.1093/beheco/arz142
Moseley, D. L., & Podos, J. (2014). Vocal Performance in Songbirds: Territorial Defense and the
Development of Male Song and Female Mating Preferences. Organismic and
Evolutionary Biology, Doctor of, 95.
Nelson, D. A. (1998). Geographic Variation in Song of Gambel’s White-Crowned Sparrow.
Behaviour, 135(3), 321–342.
Nelson, D. A. (2000). A Preference for Own-Subspecies’ Song Guides Vocal Learning in a Song
Bird. Proceedings of the National Academy of Sciences of the United States of America,
97(24), 13348–13353. https://doi.org/10.1073/pnas.240457797
Nelson, K. J., & Martin, K. (1999). Thermal Aspects of Nest-Site Location for Vesper Sparrows
and Horned Larks in British Columbia. Studies in Avian Biology, (19), 137–143.
Norton, H. (1979). The Association Between Anthropogenic Prairies and Important Food Plants
in Western Washington. Northwest Anthropological Research Notes, 13(2), 175–200.
Nowicki, S., Searcy, W. A., & Peters, S. (2002). Quality of Song Learning Affects Female
Response to Male Bird Song. Proceedings of the Royal Society B: Biological Sciences,
269(1503), 1949–1954. https://doi.org/10.1098/rspb.2002.2124

68

Owens, R. A., & Myres, M. T. (1973). Effects of Agriculture Upon Populations of Native
Passerine Birds of an Alberta Fescue Grassland. Canadian Journal of Zoology, 51(7),
697–713. https://doi.org/10.1139/z73-104
Parker, K. A., Anderson, M. J., Jenkins, P. F., & Brunton, D. H. (2012). The Effects of
Translocation-Induced Isolation and Fragmentation on the Cultural Evolution of Bird
Song. Ecology Letters, 15(8), 778–785. https://doi.org/10.1111/j.14610248.2012.01797.x
Parker, T. H., Greig, E. I., Nakagawa, S., Parra, M., & Dalisio, A. C. (2018). Subspecies Status
and Methods Explain Strength of Response to Local Versus Foreign Song by Oscine
Birds in Meta-Analysis. Animal Behaviour, 142, 1–17.
https://doi.org/10.1016/j.anbehav.2018.05.023
Pegan, T. M., Rumelt, R. B., Dzielski, S. A., Ferraro, M. M., Flesher, L. E., Young, N., …
Freeman, B. G. (2015). Asymmetric Response of Costa Rican White- breasted Woodwrens (Henicorhina leucosticta) to Vocalizations from Allopatric Populations. PLoS
ONE, 10(12), 1–16. https://doi.org/10.1371/journal.pone.0144949
Petrinovich, L., & Patterson, T. L. (1981). The Responses of White‐crowned Sparrows to Songs
of Different Dialects and Subspecies. Zeitschrift Für Tierpsychologie, 57(1), 1–14.
https://doi.org/10.1111/j.1439-0310.1981.tb01309.x
Phillips, A. R. (1964). Notas Sistematicas Sobre Aves Mexicanas, III. Revista de La Sociedad
Mexicana de Historia Natural, 25, 217–242.

69

Phillips, J. N., & Derryberry, E. P. (2017). Vocal Performance is a Salient Signal for Male–Male
Competition in White-Crowned Sparrows. The Auk, 134(3), 564–574.
https://doi.org/10.1642/auk-17-2.1
Podos, J., & Warren, P. S. (2007). The Evolution of Geographic Variation in Birdsong.
Advances in the Study of Behavior, 37(07), 403–458. https://doi.org/10.1016/S00653454(07)37009-5
Potvin, D. A., Crawford, P. W., MacDougall-Shackleton, S. A., & MacDougall-Shackleton, E.
A. (2015). Song Repertoire Size, not Territory Location, Predicts Reproductive Success
and Territory Tenure in a Migratory Songbird. Canadian Journal of Zoology, 93(8), 627–
633. https://doi.org/10.1139/cjz-2015-0039
Pulliam, H. R., & Mills, G. S. (1977). The Use of Space by Wintering Sparrows. 58(6), 1393–
1399.
Pyle, P. (1997). Identification Guide to North American Birds. Bolinas, California: Slate Creek
Press.
R Core Team. (2020). R: A Language and Environment for Statistical Computing. Retrieved
from https://www.r-project.org/
Randler, C., Förschler, M. I., Gonzalez, J., Aliabadian, M., Bairlein, F., & Wink, M. (2012).
Phylogeography, Pre-Zygotic Isolation and Taxonomic Status in the Endemic Cyprus
Wheatear Oenanthe cypriaca. Journal of Ornithology, 153(2), 303–312.
https://doi.org/10.1007/s10336-011-0744-8

70

Reed, V. A., Toth, C. A., Wardle, R. N., Gomes, D. G. E., Barber, J. R., & Francis, C. D. (in
press). Natural Noise Affects Conspecific Signal Detection and Territorial Defense
Behaviors in Songbirds. Behavioral Ecology.
Revelle, W. (2020). psych: Procedures for Psychological, Psychometric, and Personality
Research. Evanston, Illinois: Northwestern University.
Rising, J. D. (1996). A Guide to the Identification and Natural History of The Sparrows of the
United States and Canada. San Diego, California: Academic Press.
Ritchison, G. (1981). Variation in the Songs of Vesper Sparrows Pooecetes gramineus. The
American Midland Naturalist, 106(2), 392–398.
Rodenhouse, N. L., & Best, L. B. (1983). Breeding Ecology of Vesper Sparrows in Corn and
Soybean Fields. The American Midland Naturalist, 110(2), 265–275.
Roger, R. E. (2000). The Status and Microhabitat Selection of Four Grassland Bird Species in
Western Washington. The Evergreen State College.
Rosenberg, K. V, Rosenberg, K. V, Dokter, A. M., Blancher, P. J., Sauer, J. R., Smith, A. C., …
Marra, P. P. (2019). Decline of the North American Avifauna. Science, 1313(September),
1–10. https://doi.org/10.1126/science.aaw1313
Sauer, J. R., Hines, J. E., Fallon, J. E., Pardieck, K. L., Ziolkowski jr., D. J., & Link, W. A.
(2014). The North American Breeding Bird Survey, Results and Analysis 1966 - 2012.
Version 02.19.2014.

71

Schaid, T. A., Uresk, D. W., Tucker, W. L., & Linder, R. L. (1983). Effects of Surface Mining
on the Vesper Sparrow in the Northern Great Plains. Journal of Range Management,
36(4), 500. https://doi.org/10.2307/3897953
Searcy, W. A., Anderson, R. C., & Nowicki, S. (2006). Bird Song as a Signal of Aggressive
Intent. Behavioral Ecology and Sociobiology, 60(2), 234–241.
https://doi.org/10.1007/s00265-006-0161-9
Searcy, W. A., & Beecher, M. D. (2009). Song as an Agressive Signal in Songbirds. Animal
Behaviour, 78(6), 1281–1292. https://doi.org/10.1016/j.anbehav.2009.08.011
Searcy, W. A., Nowicki, S., & Hughes, M. (1997). The Response of Male and Female Song
Sparrows to Geographic Variation in Song. The Condor, 99(3), 651–657.
https://doi.org/10.2307/1370477
Seneviratne, S. S., Davidson, P., Martin, K., & Irwin, D. E. (2016). Low Levels of Hybridization
Across Two Contact Zones Among Three Species of Woodpeckers (Sphyrapicus
Sapsuckers). Journal of Avian Biology, 47(6), 887–898.
https://doi.org/10.1111/jav.00946
Short, L. L. (1969). Taxonomic Aspects of Avian Hybridization. The Auk, 8(1), 84–105.
Sibley, D. A. (2003). The Sibley Field Guide to Birds of Western North America. New York:
Alfred A. Knopf, Inc.
Slabbekoorn, H., & Smith, T. B. (2002). Bird Song, Ecology and Speciation. Philosophical
Transactions of the Royal Society B: Biological Sciences, 357(1420), 493–503.
https://doi.org/10.1098/rstb.2001.1056

72

Slater, P. J. B. (1986). The Cultural Transmission of Bird Song. Trends in Ecology and
Evolution, 1(4), 94–97. https://doi.org/10.1016/0169-5347(86)90032-7
Smith, J. M. (1965). The Evolution of Alarm Calls. The American Naturalist, 99(904), 59–63.
Sosa-López, J. R., Martínez Gómez, J. E., & Mennill, D. J. (2016). Divergence in Mating Signals
Correlates with Genetic Distance and Behavioral Responses to Playback. Journal of
Evolutionary Biology, 29(2), 306–318. https://doi.org/10.1111/jeb.12782
Stein, B. A., Kutner, L. S., Hammerson, G. A., Master, L. L., & Morse, L. E. (2000). State of the
States: Geographic Patterns of Diversity, Rarity and Endemism. In B. A. Stein, L. S.
Kutner, & J. S. Adams (Eds.), Precious Heritage: The Status of Biodiversity in the United
States (pp. 119–157). New York, New York: Oxford University Press.
Stokes, A. W., & Williams, H. W. (1971). Courtship Feeding in Gallinaceous Birds. The Auk,
88(3), 543–559.
Suckley, G., & Cooper, J. G. (1860). The Natural History of Washington Territory and Oregon.
Final Reports on the Survey of the Northern Pacific Railroad Route. Retrieved from
https://www.biodiversitylibrary.org/item/184210#page/9/mode/1up
Swenson, N. G., & Howard, D. J. (2005). Clustering of Contact Zones, Hybrid Zones, and
Phylogeographic Breaks in North America. American Naturalist, 166(5), 581–591.
https://doi.org/10.1086/491688
Temeles, E. J. (1992). The Role of Neighbors in Territorial Systems: When are they “Dear
Enemeies”? Animal Behaviour, 47, 339–350.

73

Thorpe, W. H. (1958). The Learning of Song Patterns by Birds, With Especial Reference to the
Song of the Chaffinch Fringilla coelebs. Ibis, 100, 535–570.
Toews, D. P. L. (2017). From Song Dialects to Speciation in White-crowned Sparrows.
Molecular Ecology, 26(11), 2842–2844. https://doi.org/10.1111/mec.14104
Tveten, R. K., & Fonda, R. W. (1999). Fire Effects on Prairies and Oak Woodlands on Fort
Lewis, Washington. Northwest Science, 73(3), 145–158.
United States Army Corps of Engineers Portland District. (2014). Biological Assessment for the
Continued Operations and Maintenance Dredging Program for the Columbia River
Federal Navigation Channel. Retrieved from
https://cdm16021.contentdm.oclc.org/digital/collection/p16021coll3/id/115
Vickery, P. D., & Herkert, J. R. (2001). Recent Advances in Grassland Bird Research: Where Do
We Go from Here? The Auk, 118(1), 11–15. https://doi.org/10.2307/4089754
Vickery, P. D., Hunter, M. L., & Melvin, S. M. (1994). Effects of Habitat Area on the
Distribution of Grassland Birds in Maine. Conservation Biology, 8(4), 1087–1097.
https://doi.org/10.1046/j.1523-1739.1994.08041087.x
Vickery, P. D., Hunter, M. L., & Wells, J. V. (1999). Effects of Fire and Herbicide Treatment on
Habitat Selection in Grassland Birds in Southern Maine. Studies in Avian Biology, (19),
149–159.
Weeden, J. S., & Falls, J. B. (1959). Differential Responses of Male Ovenbirds to Recorded
Songs of Neighboring and More Distant Individuals. The Auk, 76(3), 343–351.

74

Wells, J. V., & Vickery, P. D. (1994). Extended Flight-Songs of Vesper Sparrows. The Wilson
Bulletin, 106(4), 696–702.
White, R. (1980). Land Use, Environment, and Social Change: The Shaping of Island County,
WA. Seattle and London: University of Washington Press.
Wiens, J. A. (1969). An Approach to the Study of Ecological Relationships among Grassland
Birds. Ornithological Monographs, (8), 1–93.
Wilde, C. (2014). Evaluating The Endangered Species Act : Trends in Mega-Petitions , Judicial
Review , and Budget Constraints Reveal a Costly Dilemma. Villanova Environmental
Law Journal, 25(1), 307–350.
Wilkins, M. R., Seddon, N., & Safran, R. J. (2013). Evolutionary Divergence in Acoustic
Signals: Causes and Consequences. Trends in Ecology and Evolution, 28(3), 156–166.
https://doi.org/10.1016/j.tree.2012.10.002
Wray, T., Strait, K. A., Whitmore, R. C., & Sparrow, F. (1982). Reproductive Success of
Grassland Sparrows on a Reclaimed Surface Mine in West Virginia. The Auk, 99, 157–
164.
Ydenberg, R. C., Giraldeau, L. A., & Falls, J. B. (1988). Neighbours, Strangers, and the
Asymmetric War of Attrition. Animal Behaviour, 36(2), 343–347.
https://doi.org/10.1016/S0003-3472(88)80004-6
Zink, R. M. (1996). Comparative Phylogeography in North American Birds. Evolution, 50(1),
308. https://doi.org/10.2307/2410802

75

Zink, R. M., & Barrowclough, G. F. (2008). Mitochondrial DNA Under Siege in Avian
Phylogeography. Molecular Ecology, 17(9), 2107–2121. https://doi.org/10.1111/j.1365294X.2008.03737.x

76

Media

Thesis_MES_2021_LequeT.pdf