-
Title
-
Eng
Comparing Endangered Streaked Horned Lark Fecundity to Other Grassland Birds
-
Date
-
2010
-
Creator
-
Eng
Anderson, Jeffrey K
-
Subject
-
Eng
Environmental Studies
-
extracted text
-
Comparing endangered Streaked Horned Lark
(Eremophila alpestris strigata)
fecundity to other grassland birds
by
Jeffrey K. Anderson
A Thesis submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Study
The Evergreen State College
September 2010
© 2010 by Jeffrey K. Anderson. All rights reserved.
This Thesis for the Master of Environmental Study Degree
by
Jeffrey K. Anderson
has been approved for
The Evergreen State College
by
________________________
Dr. Alison Styring
Member of the Faculty
________________________
Dr. Scott Pearson
Westside Research Team Leader and Senior Scientist
Washington Department of Fish and Wildlife
________________________
Dr. Timothy Quinn
Member of the Faculty
________________________
Date
ABSTRACT
Comparing endangered Streaked Horned Lark
(Eremophila alpestris strigata) fecundity to other grassland birds
Jeffrey K. Anderson
The Streaked Horned Lark (Eremophila alpestris strigata) is a critically
endangered subspecies which breeds on prairie remnants in Washington and
Oregon. Dramatic losses in grassland habitat have pushed the lowland Puget
populations to the brink of extinction, with projected population losses at 40% a
year. In order to investigate potential mechanisms driving this decline, I
conducted a case study of Streaked Horned Larks at 13th Division Prairie, Fort
Lewis, Washington over a two year period, 2007 and 2009. I analyzed nesting
data of all species comprising the grassland ground nesting guild, and compared
Streaked Horned Lark fecundity with those of the larger guild to determine if the
breeding site itself is a sink, or if low fecundity is specific to Larks. I compared
fecundity in two separate groups: (1) Larks vs. the ground nesting guild and (2)
Larks vs. Savannah Sparrows (Passerculus sandwichensis). In these comparisons,
Streaked Horned Larks had significantly lower values in all measures of
reproductive success when compared to both the guild and Savannah Sparrows.
Furthermore, the Streaked Horned Lark’s low egg hatching rate of 44% suggests
that inbreeding depression may be playing a role in the decline of Larks at 13th
Division Prairie. Although analyses of nest site habitat variables confirmed that
Streaked Horned Larks have unique nesting preferences, cross-year, interspecific
comparisons of vital rates and nest site characteristics did not indicate site-wide
environmental causes driving Streaked Horned Lark declines. Since these
findings are based on a case study of a single breeding site, I recommend further
monitoring of this site and other remaining breeding sites, with emphasis on
potential inbreeding depression.
TABLE OF CONTENTS
LIST OF FIGURES
v
LIST OF TABLES
vi
ACKNOWLEDGEMENTS
vii
INTRODUCTION
1
METHODS
4
Study Site
4
Data Collection
5
Nest Discovery and Monitoring
6
Nest Site Habitat Sampling
7
Vital Rates and Annual Fecundity
8
Data Analysis
12
RESULTS
15
Streaked Horned Lark and Guild Comparison
15
Streaked Horned Lark and Savannah Sparrow Comparison
17
Impact of Nest Exclosures
19
Nest Site Habitat Comparison
20
DISCUSSION
25
Clutch size
26
Proportion hatched
26
Fledglings per nest
28
Proportion nests depredated
29
Proportion nests abandoned
29
Annual Fecundity
30
Nest Site Habitat Analysis
30
Impacts of Environmental Factors on Vital Rates
31
RECOMMENDATIONS FOR FURTHER RESEARCH
33
LITERATURE CITED
35
iv
LIST OF FIGURES
Figure 1. Current and historic Streaked Horned Lark
2
breeding sites and possible historic nesting or uncertain
breeding season locations.
Figure 2. Western Washington State with star designating
4
the study site location (13th Division Prairie) for both years
of the field study.
Figure 3. Vital rates by year for Streaked Horned Lark and
17
guild nests.
Figure 4. Annual fecundity (mean annual female fledglings
19
per pair) for Streaked Horned Larks and Savannah
Sparrows.
Figure 5. Mean proportion substrate cover of nest area for
22
successful and failed STHL and guild nests.
Figure 6. Mean percent of non-vegetated nest area for
25
STHL and guild nests based on total nests and nest success.
v
LIST OF TABLES
Table 1. Vital rates of Streaked Horned Lark and guild nests
16
(2007 & 2009) from 13th Division Prairie, Ft. Lewis, WA.
Table 2. Vital rates of Streaked Horned Lark and Savannah
18
Sparrow nests (2007 & 2009) from 13th Division Prairie, Ft.
Lewis, WA.
Table 3. Comparison of Streaked Horned Lark nest site
21
substrate variables between successful nests (produced at
least 1 fledgling) and failed nests (abandoned or
depredated) from 2007 and 2009.
Table 4. Comparison of guild nest site substrate variables
21
between successful nests (produced at least 1 fledgling)
and failed nests (abandoned or depredated) from 2007 and
2009.
Table 5. Comparison of Streaked Horned Lark and guild
23
nest site substrate variables from 2007 and 2009.
Table 6. Comparison of Streaked Horned Lark and guild
24
nest site vegetative functional group variables from 2009.
vi
ACKNOWLEDGEMENTS
I would like to thank the head of my thesis committee, Dr. Alison Styring, for her
valuable insights, experience, and encouragement through this process and for
her outstanding classroom teaching throughout my time as a graduate student. I
owe a huge thanks to Dr. Scott Pearson who has provided the ideas and
inspiration that made this project possible. I would also like to thank my
committee member Dr. Timothy Quinn for his exceptional editing skills and
suggestions regarding this document. I am grateful to Hannah Anderson, the
Nature Conservancy, Washington Department of Fish and Wildlife, and Fort
Lewis for giving me the opportunity to study Streaked Horned Larks, and I thank
Mark Hopey for collecting roughly half of the data used in this study. I must
thank my wife Laura, not only for her hours of editing and feedback, but for her
enduring love and support. Lastly, I thank my son Kai for his constant reminders
of what’s truly important in life.
vii
INTRODUCTION
Grassland ecosystems are among the most imperiled in the United States and
have been altered to a greater degree than any other biome in North America (Samson
and Knopf 1994). Accompanying this loss of habitat is a widespread decline in North
American grassland bird populations (Robbins et al. 1989, Knopf 1994, Vickery et al.
1994). This precipitous decrease is not only widespread, but is progressing at a faster
and less variable rate than in any other guild of North American birds (Peterjohn and
Sauer 1993, Knopf 1994).
Although primarily known for its forested ecosystems, Washington State is also
home to native prairie habitats that are disappearing at a rapid rate (Kruckeberg 1995,
Stinson 2005). As the remaining prairies of the Pacific Northwest face persisting threats
from human development, we continue to lose flora and fauna that have evolved along
with these rare, treeless, flat open-spaces (Crawford and Hall 1997, Pearson and Altman
2005). One rapidly disappearing subspecies associated with prairies is the Streaked
Horned Lark (Eremophila alpestris strigata).
The Streaked Horned Lark (referred to as “Lark” throughout this thesis) is a rare
subspecies of ground-nesting bird that inhabits open grassland habitats of Washington,
Oregon, and (previously) British Columbia. In Canada, where they are believed to be
extirpated, Larks are listed as endangered by the Species at Risk Act, and in the United
States, Streaked Horned Larks are a federal candidate for listing under the Endangered
Species Act (Beauchesne and Cooper 2003). At the state level, they are listed in
Washington as endangered and in Oregon as a sensitive species, critical category (ODFW
1
2006, Pearson et al. 2008). Genetic
data confirm that this subspecies is
unique, isolated and possesses very
little genetic diversity (Drovetski et al.
2005). In addition, recent research
estimates that Streaked Horned Lark
populations are declining at a rate of
40% per year (Pearson et al. 2008,
Schapaugh 2009, Camfield et al.
2010).
There are historical records of
Streaked Horned Larks breeding at
the northern end of their range in
southern British Columbia, the San
Juan Islands and additional coastal
areas north of Tacoma, but these
sites appear to currently be devoid of
Figure 3. Current and historic Streaked
Horned Lark breeding sites and possible
historic nesting or uncertain breeding
season locations. Figure reproduced from
(Pearson and Altman 2005).
any Lark populations (Fig 1.) Mirroring the
loss of breeding sites to the north, the
southern end of the Streaked Horned Lark’s
range has shrunk towards the north and Larks are no longer found in the Rogue River
Valley of southern Oregon (Rogers 2000, Beauchesne and Cooper 2003, Stinson 2005). It
appears that the Lark range is retracting towards its core- the wintering habitat of the
Willamette Valley and lower Columbia River islands of Oregon and Washington.
2
Each remaining population of Streaked Horned Larks is estimated to be below
500 individuals: Puget lowlands (222 birds), Washington coast (86 birds), lower
Columbia River (68 birds) and the Willamette Valley (398 birds) (Pearson and Altman
2005, Stinson 2005).
As the amount of suitable Streaked Horned Lark nesting habitat continues to
shrink and historic breeding locations cease to be used, it has become critical to assess
Lark reproductive output on a site-by-site basis in order to allocate recovery efforts and
funds to the places where Larks are successfully reproducing. The intent of this paper is
to compare the annual fecundity of Streaked Horned Larks to other ground nesting
grassland species at a single breeding site, to determine if low fecundity is unique to
streaked horned larks and to identify the factors contributing to low fecundity. I have
employed three separate approaches to help answer this question:
1. Compare fecundity of Streaked Horned Larks with all other ground nesting
species that nest at the site; with an additional emphasis on Savannah Sparrows.
Through these comparisons, it may be determined if all species at the site are
experiencing low fecundity, or if it is only Streaked Horned Lark productivity that
is low.
2. Compare nest site habitat characteristics of Streaked Horned Larks with those of
all other ground nesting species at 13th Division Prairie. Through these
comparisons it can be established which nest site characteristics are most
closely related to each species, and to what extent these characteristics are
found in nest site plots of both successful and failed nests.
3
3. Discuss the possible impacts of environmental factors vs. endogenous factors on
fecundity. A look at annual trends among nesting species can provide insights
into the mechanisms driving nesting success or failure.
METHODS
Study Site
Research was conducted on a 202.6 hectare section of 13th Division Prairie (47°
01’N 122° 26’W) (Anderson 2005)) located on the U.S. Army’s Fort Lewis in Washington
State (Fig. 2). Fort
Lewis was established
in 1917, with active
military training
taking place on the
surrounding prairies
starting in the late
1930s (Dunwiddie et
al. 2006). A glacial
outwash prairie, 13th
Division Prairie is
among a handful of
south Puget Sound
Figure 4. Western Washington State with star designating the study site
location (13th Division Prairie) for both years of the field study.
4
prairies that have remained undeveloped and retain at least a portion of native Puget
prairie flora and fauna (Kruckeberg 1995, Dunwiddie et al. 2006). Although it is merely
a remnant of the once extensive Northwest prairie expanse, 13th Division Prairie is
among the largest patches of undeveloped, native prairie left in Washington’s south
Sound region (Kruckeberg 1995, Dunwiddie et al. 2006).
Data Collection
Vital rates and habitat data were collected from April to August for the breeding
seasons of 2007 and 2009. Although identical methods of data collection were used for
both seasons, I only collected data for the 2009 season. All data from the 2007 season
was collected by Mark Hopey, working under the guidance of Dr. Scott Pearson from
Washington Department of Fish and Wildlife. The data from 2007 was collected as part
of a study that attempted to identify Streaked Horned Lark predators by setting up
video cameras at nest locations. In Pearson and Hopey’s (2008) study, nests of other
species that share a similar nesting ecology to the Streaked Horned Lark were also
located and outfitted with cameras in order to ascertain the suite of predators.
The data collected for the 2009 breeding season was part of a larger project
under the auspices of Washington Department of Fish and Wildlife, Oregon Department
of Fish and Wildlife, and The Nature Conservancy. This project, designed to test
predator exclusion nest-cages, enclosed nests in chicken wire cages that were
approximately 1m x 1m x1m. These exclosures have openings large enough for Larks to
come and go, but restrict nest access by larger predators (mammals, corvids, raptors,
etc.). To randomize the experiment, roughly every other Lark nest discovered was
5
outfitted with an exclosure. The potential impacts of the nest exclosure experiment on
the results of my study will be discussed further in the ‘Results’ section of the paper.
Nest Discovery and Monitoring
The standardized methodology for grassland nest searches (Martin and Geupel
1993) was used to locate and monitor nests throughout the 2007 and 2009 breeding
seasons. Nests were located by one of four methods: observation of adults carrying
food or nesting material back to the site, flush of an incubating or brooding adult,
systematic search of areas where adults were routinely observed, and the rope-dragging
technique (Martin and Geupel 1993). Nearly all nests were located using search effort
and observation centered on Streaked Horned Lark nests. The nests of all other species
were found opportunistically through their close proximity to Streaked Horned Lark
nests or by through systematic walking transects. Every week (from the end of April to
the beginning of August) we walked a site-wide grid transect of 150m intervals in
alternating directions (north to south, east to west; northeast to southwest; and
northwest to southeast). In addition, we divided the study site into a grid of 7
approximately equal sized polygons which were systematically searched by walking
transects spaced at 50m intervals at a typical rate of 1 (of the 7) polygon searched per
day.
Once nests were discovered, we monitored them once every 3-5 days, or more
frequently if the nest was close to hatching or fledging. Nest observations were made
as quickly as possible with efforts being made to visit the nests while the adults were
away foraging. The date of the nest check, number of eggs, nestlings and approximate
ages were recorded for each nest visit. During nest checks, we also recorded
6
observations of adults that included their sex, locations and behaviors. The information
from adult observations was used in determining if nests were abandoned, or still being
tended after a disturbance.
A nest was considered successful if evidence indicated successful fledging of at
least one chick. Such evidence included observation of the parents making food
deliveries near the nest area, nestlings observed outside the nest, or if the nest rim was
flattened with droppings located on top of it or outside the nest area, along with no
signs of predation.
Nest Site Habitat Sampling
Habitat variables were measured using the methods of Pearson and Hopey
(2005) that were derived from Barbour et al. (1980). Vegetation and substrate were
measured using a 1m long wooden pin-drop frame that was broken into gradations of
10 cm. At each interval a metal pin was dropped through a hole in the wooden frame
and the number of vegetative hits on that pin were counted and keyed out to species.
Additionally, the underlying substrate of the pin was recorded, along with the maximum
vegetation height at each hit. The pin-drop frame was placed with the midpoint of the
meter span directly over the nest and in a north –south orientation. Once vegetation
variables were measured, the frame was re-positioned in an east-west orientation and a
second round of data was collected.
Substrate results are reported in mean percent cover of the nest area plot and
were averaged from pin drops for each nest, resulting in 100% of the nest area falling
into one of four categories: thatch, bare ground, rock, or moss/lichen.
7
For the functional groups, pin drop hits were categorized into individual plant
species and then placed into the categories of native annual forbs (NAF), non-native
annual forbs (NNAF), native perennial forbs (NPF), non-native perennial forbs (NNPF)
native annual grasses (NAG), non-native annual grasses (NNAG), native perennial
grasses (NPG), and non-native perennial grasses (NNPG) for analysis. Results should be
interpreted as percent coverage of the nest area, but unlike substrate variables,
functional group coverage can exceed 100% due to pins contacting multiple species
within more than one group on some of the drops. Another variable accompanies the
analysis for functional groups: Vegetation height (cm) was calculated by taking the
highest point of plant/pin intersection for each pin, adding all the heights and dividing
by the number of pins for each nest. This gives an average maximum vegetation height
for the overall nest site area
In addition to the above habitat variables, total non-vegetated hits were
calculated by adding the number of pins that did not touch any plant and then dividing
by the number of pins. This measure gives an approximation of the percent nonvegetated cover.
Vital Rates and Annual Fecundity
We measured 6 vital rates for all bird species:
1) Clutch size (
was determined from nests that were observed with eggs prior to
hatching, or, if nests were discovered during the nestling phase, nestlings were counted
and added to any unhatched eggs that were also inside the nest. Although counting
nestlings may have resulted in artificially low clutch size estimates (i.e., eggs could have
8
been removed from the nest during the nestling phase) it was necessary to include nests
discovered during the nestling phase (n=5, 17%) in order to achieve a sufficient sample
size to calculate proportion of eggs hatched. I removed one clutch size record from
analysis as an extreme outlier: A Streaked Horned Lark nest was found in 2007 with 11
eggs, which were probably multiple broods laid by the same female, none of which
hatched. This nest (with over 3.5 times the mean egg counts of Lark nests) affected the
data considerably due to the small sample size of Lark nests. Because the 11 egg nest
was more than three standard deviations from the mean, (Osborne and Overbay 2004)
it was eliminated from all calculations specific to eggs and their hatching. When this
nest was removed from the data set, the standard deviation changed from 1.61 to 0.60
and the mean clutch size decreased from 3.30 to 3.03, which is closer to numbers
reported in the literature [3.05±0.07 n=135 nests (Pearson et al. 2008, Camfield et al.
2010)}.
2) Proportion of eggs hatched is the number of eggs in a nest that hatched relative to
the number present at hatching (Briskie and Mackintosh 2004). Calculation of this rate
precludes eggs from nests that were depredated or abandoned before a full incubation
period. Again, this may have resulted in an overestimation of the actual proportion of
hatched eggs thus an overestimation of hatch rates.
3) Fledglings per nest is the total number of fledglings produced by each nest and
includes all nests that had a known outcome.
4) Nest survival was determined with Mayfield (1975) estimators and measures the
probability of a nest to fledge at least one nestling.
9
5) Proportion depredated represents the proportion of all discovered nests that
appeared to be destroyed by a predator. This includes nests where young or eggs went
missing from the nest before a probable fledge time could be attributed for their
absence. If only a portion of the eggs or chicks in a nest were depredated, the nest was
monitored for adult presence. In all of these instances, the nests were abandoned
resulting in no fledglings.
6) Proportion abandoned is the proportion of all nests that were abandoned throughout
any phase of the nesting process.
7) Annual fecundity was estimated using an equation from Ricklefs and Bloom (1977)
designed to calculate annual production of total fledglings per pair (P). As annual
fecundity is the number of female fledglings (Pearson et al. 2008, Camfield et al. 2010),
P was divided by two assuming an equal distribution of the two sexes between fledging.
The formula for annual production of fledglings (P) is:
(
)
is the number of days in the breeding season and is corrected for the variance in
breeding effort across the months of the breeding season with the formula:
)
Where
is the proportion of clutches that were laid in each month , and e is the base
of natural logarithms.
is the number of young fledged/pair/day and is calculated as:
10
Where
is clutch size, is breeding success (measured in fledglings per egg laid) and
is the rate of nest initiation (clutches/pair/day) and is calculated as:
Nest mortality rate ( ) is the proportion of nests failing per day and was calculated
using the midpoint method for the Mayfield (1975) estimator. After the fledging of a
successful clutch,
represents the time before the next clutch is initiated, and
is the
time interval between a failed clutch and a new one. Probability of a nest failing before
fledging is designated as
and calculated:
= 1Where
is the probability that a nest will successfully fledge at least one young and is
calculated:
Where
= the length of the nest cycle from clutch initiation to fledging in days.
was
calculated for Streaked Horned Larks as 12 days of incubation + 9 days until fledging + a
laying day for each egg in the clutch ( ) (Beason 1995).
was calculated similarly for
Savannah Sparrows, with the exception of an 11-day fledging period (Wheelwright
2008).
11
Data Analysis
In order to compare the nesting data of Streaked Horned Larks with that of the
other species at 13th Division Prairie, two different comparisons of this data set were
performed. In the first comparison, Streaked Horned Lark vital rates from 2007 and
2009 were compared to those of a nesting guild comprised of all other ground nesting
grassland species at 13th Davison Prairie from the same breeding seasons. This was
done by treating nests of all species as a single species (the guild) as described below.
The second comparison matches Streaked Horned Larks with Savannah Sparrows.
The purpose of this case study is to more finely assess potential mechanisms
driving the decline of a Streaked Horned Lark population, by comparing Lark breeding
success with that of the guild. Root (Root 1967) defines a guild as “a group of similar
species that exploit a resource in a similar fashion”. Guilds can group animal species on
the basis of habitat use or behavioral characteristics (Severinghaus 1981, Brooks and
Croonquist 1990). For the purpose of this study, the guild was based on nesting habitat.
Although there is an inference of differences in microhabitat use between members of
the guild, overall, it has been found that the guild concept can be particularly effective
in increasing samples sizes in studies like mine and decreasing statistical variability by
virtue of larger sample sizes (Verner 1983, 1984, Block et al. 1986). In addition, use of
the guild comparison allows us to explore the idea that Streaked Horned Lark declines
are a function of environmental change that would affect all species in the guild (Block
et al. 1986). Guild-based studies can also reflect the biological integrity of an area in a
more complete way than a look at a single species (Angermeier and Karr 1994, Bishop
12
and Myers 2005). Biological integrity can be defined as, "the ability of an environment
to support and maintain a biota (both structural and functional performance)
comparable to the natural habitats of the region." (Angermeier and Karr 1994)
The species that comprised the ground nesting grassland guild were Savannah
Sparrow, Western Meadowlark (Sturnella neglecta), Vesper Sparrow (Pooecetes
gramineus), Common Nighthawk (Chordeiles minor), and Killdeer (Charadrius
vociferous). This guild represents the entirety of species that nest on the ground at this
particular prairie. It is possible that Northern Harrier (Circus cyaneus) and Short-eared
Owl (Asio flammeus) could also be included in this group, but no nests of these species
were discovered. The pooled vital rates of these species were then compared to those
of Streaked Horned Larks.
In addition to comparisons between Larks and the guild, statistical comparisons
were also calculated between Larks and Savannah Sparrows. Savannah Sparrows made
for strong pair-wise comparisons because they made up 29 of the 46 guild nests and
share a very similar ecology to Streaked Horned Larks. Both Larks and Savannah
Sparrows inhabit open country and share a similar diet and foraging behaviors (Beason
1995, Wheelwright 2008). In addition, both species have similar incubation and
fledgling times (Martin 1951, Maher 1979, Meunier and Bedard 1984, Beason 1995).
Savannah Sparrows and Streaked Horned Larks differ in some aspects of their breeding
ecology: Savannah Sparrows select more densely vegetated sites for their nests (Beason
1995, Wheelwright 2008), have slightly longer nestling periods (Wheelwright 2008), and
have larger clutch sizes. Clutch size varies geographically, but Horned Larks typically lay
2 to 5 eggs with a mean of 2.5 in Washington and British Columbia (Beason 1995)
13
whereas Savannah Sparrows lay between 2 and 6 eggs with a mean of 4 eggs across
North America. Although the two species have some ecological differences, comparing
them with one another eliminates some of the confounding variables inherent with the
guild approach.
All 6 vital rates (clutch size, proportion hatched, fledglings per nest, nest
survival, proportion nests depredated, and proportion nests abandoned) were
compared between 2007 and 2009, between Larks and the guild, and between Larks
and Savannah Sparrows.
Annual fecundity calculations were calculated using replacement nest interval
( ) and multiple brood interval ( ) data from existing literature. Calculations for
Savannah Sparrows were done using the interval means of =19 and
2008)and =22 and
=5 (Wheelwright
=22.25 interval numbers for Streaked Horned Larks (Pearson et
al. 2008) No statistical comparisons were done on annual fecundity calculations due to
small sample sizes (n=2).
Habitat characteristics around each nest site were compared in two different
groupings: substrate (bare ground, rock, moss/lichen, or thatch) and vegetative
functional groups (native and non-native, annual and perennial grasses and forbs). Nest
site substrate comparisons were performed between successful and failed Lark nests,
between successful and failed guild nests, and between all Lark and guild nests. Nest
site functional group comparisons were performed between Lark and guild nest sites.
Data from 2007 and 2009 were pooled for all analyses except those regarding
annual fecundity and functional group habitat variables. Due to an incomplete data set,
14
the functional group analyses were only performed with the data from 2009, and
consequently are based on smaller sample sizes than the data for the substrate and
non-vegetated hit analyses.
All comparisons for vital rates and nest site habitat variables were performed
with two-sample Wilcoxon Rank Sum tests in the program R (Team 2006). Overall
significance for these tests was designated at =0.05 and all totals are reported as
means ±SE, unless otherwise noted. In order to decrease the chance of Type 1 errors
from multiple comparisons, Bonferroni corrections (α=0.05/n) were made for vital rate,
substrate and functional group calculations (Rice 1989). After these corrections, the
overall significance (α=0.05) was adjusted to α=0.008 for vital rates, α=0.0125 for
substrate variables, and α=0.008 for functional group variables. Due to the conservative
nature of the Bonferroni corrections, the calculated p-values are also included in the
results in order to assess which comparisons might be biologically meaningful, albeit not
statistically significant (Cabin and Mitchell 2000).
RESULTS
Streaked Horned Lark and Guild Comparison
Vital rates of Lark nests were significantly lower than guild vital rates for 4 of 6
measures of reproductive success (Table 1). The only category where Streaked Horned
Larks had significantly higher averages was Proportion of Nests Abandoned, which is
equated with nesting failure (Table 1).
15
In two instances, Lark numbers were more than 50% lower than those of the
guild: proportion hatched (Streaked Horned Lark 52% lower than guild), and fledglings
per nest (Streaked Horned Lark 64% lower than guild) (Table 1).
Table 7. Vital rates of Streaked Horned Lark and guild nests (2007 & 2009) from 13th Division Prairie, Ft.
Lewis, WA. Comparisons that demonstrated significance after Bonferroni corrections (p<0.008) are in
bold. Values are means ±SE with number of nests in parentheses. W and p statistics from Wilcoxon Rank
Sum tests.
Streaked
Horned Lark
Guild
Statistic
P value
(E. a. strigata)
Clutch size
3.03±0.12
(29)
3.38±0.15
(39)
W = 699
0.08
Proportion
hatched
0.44±0.09
(17)
0.91±0.03
(29)
W = 410
<0.0001
Fledglings per
nest
0.66± 0.20
(27)
1.82±0.26
(40)
W = 738
0.003
Nest survival
0.27±.03
(30)
0.46±0.04
(44)
W=1029
<0.0001
Proportion nests
depredated
0.33±0.09
(30)
0.32±0.07
(46)
W = 685
0.9531
Proportion nests
abandoned
0.27±0.08
(30)
0.00±0
(46)
W = 506
0.0002
For four indicators of fecundity (hatch rate, fledglings per nest, clutch size, and
fledglings per egg) the annual differences in vital rates between Larks and the guild
showed no clear pattern (Fig. 3). In 2007, Streaked Horned Lark nests had lower
productivity in hatch rate, fledglings per nest, clutch size and fledglings per egg than
they did in 2009. In contrast, guild nests actually had higher productivity in 2007 for
hatch rate, fledglings per nest, and fledglings per egg than they did for 2009.
16
4
3.5
3
2.5
STHL 2007
2
STHL 2009
Guild 2007
1.5
Guild 2009
1
0.5
0
Hatch Rate
Fledglings per
nest
Clutch Size
Fledglings per
egg
Figure 3. Vital rates by year for Streaked Horned Lark and guild nests.
Streaked Horned Lark and Savannah Sparrow Comparison
In the Lark vs. Savannah Sparrow comparison, these species differed significantly
on all but one of the vital rates: proportion of nests that suffered predation (Table 2). As
with the guild comparisons, Streaked Horned Lark vital rates were only significantly higher
in one category: proportion of nests abandoned (Table 2). As with the guild comparisons,
there were two vital rates that differed by a margin of more than 50%: proportion
hatched (Streaked Horned Lark 54% lower than Savannah Sparrow), fledglings per nest
(Streaked Horned Lark 66% lower than guild). In these two comparisons that differed by
more than 50%, Streaked Horned Lark results were lower when compared with Savannah
Sparrows than they were against the guild as a whole.
17
Table 8. Vital rates of Streaked Horned Lark and Savannah Sparrow nests (2007 & 2009) from 13 th Division
Prairie, Ft. Lewis, WA. Comparisons that demonstrated significance after Bonferroni corrections
(p<0.008) are in bold. Values are means ±SE with number of nests in parentheses. W and p statistics from
Wilcoxon Rank Sum tests.
Streaked Horned
Lark
Savannah
Sparrow
(E. a. strigata)
(P. sandwichensis)
Statistic
P value
Clutch size
3.03±0.12
(29)
3.61±.18
(23)
W = 460.5
0.01
Proportion hatched
0.44±0.09
(17)
0.96±0.02
(17)
W = 248.5
0.0001
0.96±0.31
(28)
W= 519.5
0.004
Fledglings per nest
0.66± 0.20
(27)
Nest survival
0.27±0.03
(30)
0.39±0.03
(29)
W = 705
<0.0001
Proportion nests
depredated
0.33±0.09
(30)
0.34±0.09
(29)
W = 440
0.9337
Proportion nests
abandoned
0.27±0.08
(30)
0.00±0.00
(29)
W = 319
0.003
Annual fecundity
0.99
(2)
3.25
(2)
Average Streaked Horned Lark annual fecundity for the two breeding seasons
was 70% lower than that of Savannah Sparrows (Table 2).
18
4
3.5
Female fledglings
3
2.5
2
2007
2009
1.5
1
0.5
0
Streaked Horned Lark
Savannah Sparrow
Species
Figure 4. Annual fecundity (mean annual female fledglings per pair) for Streaked Horned Larks and
Savannah Sparrows.
Both Larks and Savannah Sparrows had higher annual fecundity in 2009 than in
2007 (Fig. 4). Estimated annual fecundity for Larks was 0.34 in 2007 and 1.63 in 2009.
Savannah Sparrow fecundity was 2.93 for 2007 and 3.72 for 2009. Compared to 2007,
the 2009 breeding season represented a 79% increase in annual fecundity for Larks and
a 21% increase for Savannah Sparrows.
Impact of Nest Exclosures
The predation rates of Streaked Horned Lark nests in this study may be
artificially low due to the nest-exclosure experiment that was carried out during the
19
2009 breeding season. As the exclosure experiment results have not been published
(the study is ongoing) the data here can only reflect their effectiveness at 13th Division
Prairie during the 2009 breeding season. Out of the six exclosed nests, three failed and
three produced fledglings. Therefore, predation rates could conceivably have been as
high as 55% for the 2009 season (that is, if all three successful exclosed nests were
never exclosed and ended up being depredated). It should also be noted here that
three of the nests that were exclosed still failed: two from predation and one from
starvation, possibly due to a severely malformed beak on the nest’s single nestling.
Nest Site Habitat Comparison
A comparison between successful and failed (depredated or abandoned)
Streaked Horned Lark nests revealed that nests that successfully fledged at least one
young were built in substrates that contained much higher percentages of moss/lichen
than thatch (Table 3). The percentage of ground covered in moss or lichens for
successful Lark nests was 34.2% higher than moss and lichen coverage surrounding
failed nests, whereas failed nests were situated among a 34.8% higher percentage of
thatch covered substrate than successful nests.
20
Table 9. Comparison of Streaked Horned Lark nest site substrate variables between successful nests
(produced at least 1 fledgling) and failed nests (abandoned or depredated) from 2007 and 2009.
Comparisons that demonstrated significance after Bonferroni corrections (p<0.0125) are in bold. Values
are means ±SE percent cover with number of nests in parentheses. W and p statistics are from Wilcoxon
Rank Sum tests.
Failed
Nests
(17)
16.7±4.9
Statistic
P value
Bare Ground
Successful
Nests
(9)
9.3±2.6
W = 80.5
0.84
Moss/Lichen
50.9±9.0
16.7±5.3
W = 28.5
0.009
Rock
25.0±7.7
16.2±5.9
W = 53
0.20
Thatch
15.7±3.2
50.5±3.2
W = 129.5
0.004
At the guild level, all nests, both failed and successful had a higher percentage
of thatched substrate than other substrates, with the other three substrate variables
combined filling less than 15% of nest site areas (Table 4). At the guild level, there were
no significant differences between substrate variables for failed versus successful nests
(Table 4).
Table 10. Comparison of guild nest site substrate variables between successful nests (produced at least 1
fledgling) and failed nests (abandoned or depredated) from 2007 and 2009. Values are means ±SE
percent cover with number of nests in parentheses. W and p statistics are from Wilcoxon Rank Sum tests.
Successful Nests
(26)
Failed Nests
(12)
Statistic
P value
Bare Ground
1.9±1.1
1.4±0.9
W = 162
0.77
Moss/Lichen
7.4±2.8
8.3±5.3
W = 149.5
0.81
Rock
4.8±2.9
4.2±3.5
W = 157.5
0.96
85.9 ±5.3
86.8±8.7
W = 178
0.46
Thatch
21
The graph below (Fig.5) combines Tables 3 and 4 for a visual representation of the
interplay between nest fate and substrate variables.
Proportion of nest site area covered
1.2
1
0.8
Thatch
0.6
Rock
Moss
0.4
Bare Ground
0.2
0
STHL Successful STHL Failed
(n=9)
(n=17)
Guild
Successful
(n=26)
Guild Failed
(n=12)
Figure 5. Mean proportion substrate cover of nest area for successful and failed STHL and guild nests.
Number of nests is given in parenthesis.
Nest site substrate variables were significantly different between Streaked
Horned Lark and guild nests in all four categories. Although a thatched substrate is the
highest of the four associated with Streaked Horned Lark nests, it is still 47% lower than
thatch coverage associated with guild nests (Table 5).
22
Table 11. Comparison of Streaked Horned Lark (STHL) and guild nest site substrate variables from 2007
and 2009. Comparisons that demonstrated significance after Bonferroni corrections (p<0.0125) are in
bold. Values are means ±SE percent cover with W and p statistics from Wilcoxon Rank Sum tests.
STHL (n=26)
Guild (n=38)
Statistic
P value
Bare ground
14.1±3.4
1.8±0.8
W=226.5
<0.00001
Moss/Lichen
28.5±5.6
7.7±2.5
W=249.5
0.0003
Rock
19.2±4.7
4.6±2.3
W=280
0.0005
Thatch
38.5±5.7
86.2±4.5
W=877
<0.00001
Nest site coverage by functional group showed no significant differences
between Streaked Horned Lark nests and those of guild species. While there was only
one significant difference between functional group variables (non-vegetated nest area),
some of the other differences are also worth noting; in particular, differences in percent
cover of native perennial grasses (Larks= 19.4±6% vs. Guild=51.9±8.3%; p=0.03) and
differences in vegetation height (Larks=15.9 ±3.4 cm vs. Guild=21.9±2.0 cm; p=0.11)
(Table 6).
23
Table 12. Comparison of Streaked Horned Lark (STHL) and guild nest site vegetative functional group
variables from 2009. Comparisons that demonstrated significance after Bonferroni corrections (p<0.008)
are in bold. Values are means ±SE percent cover, except for Vegetation Height, which is the mean
maximum height of vegetation at the nest site. The data for mean percent cover of non-vegetated nest
area is from 2007 and 2009. W and p statistics are from Wilcoxon Rank Sum tests.
STHL (n=9)
Guild (n=19)
Statistic
P value
Non-native
annual grass
11.1±5.0
5.7±2.0
W=74.5
0.57
Non-native
perennial forb
12.0±3.4
9.6±3.1
W=68
0.38
Non-native
perennial grass
38.9±10.6
43.4±7.3
W=93
0.73
Native
perennial forb
7.4±4.0
5.3±1.8
W=83
0.91
Native
perennial grass
19.4±6.9
51.9±8.3
W=129
0.03
Vegetation
height (cm)
15.9 ±3.4
21.9±2.0
W=118.5
0.11
Non-vegetated
nest area
29.8±4.9
14.5±3.6
W=289
0.004
Although differences in non-vegetated area between Larks and the guild are
significant, there is very little difference in non-vegetated area within each group,
regardless of nest outcome (Fig. 6).
24
35.00%
Percent of non-vegetated nest area
30.00%
25.00%
20.00%
15.00%
10.00%
5.00%
0.00%
Figure 6. Mean percent of non-vegetated nest area for STHL and guild nests based on total nests and nest
success. Number of nests in parenthesis.
DISCUSSION
Streaked Horned Larks at our study site had reduced fecundity when compared
with either the ground nesting guild as a whole, or Savannah Sparrows specifically. For
both guild and Savannah Sparrow comparisons, Streaked Horned Lark rates were
considerably lower in all categories except proportion of depredated nests and clutch
size.
25
Clutch size
In this study, clutch size is not a very informative measure of reproductive
success. Clutch size is species specific and not necessarily an indication of relative
fitness. For example, Common Nighthawks almost always have a clutch size of two
(Poulin et al. 2006). The Streaked Horned Lark clutch size reported in this study (3.03
eggs per nest) is very similar to that found by Pearson et al (2008) in a study of 135 Lark
nests (3.05 eggs per nest). Although clutch size may not be the most effective measure
of fecundity in cross-species studies, Streaked Horned Lark clutch sizes are lower than
other subspecies of Horned Larks (Camfield et al. 2010), and this may indicate a
disadvantage.
Proportion hatched
Every egg laid reflects a considerable expenditure in energy for that particular
bird (Koenig 1982). For this reason, most passerine species across the world average a
high hatch rate of about 90% (Koenig 1982). The results of this study were consistent
with this finding for the guild and Savannah Sparrows, whose mean (across two seasons)
hatch rates were 91% and 96% respectively. Although Streaked Horned Lark hatch rates
were much lower (44%), it should be noted that all subspecies of Horned Larks may
have relatively low hatch rates. A sample of three studies of three different Horned Lark
subspecies returns three different hatch rates: Pickwell’s (1931) study of 82 eggs had a
hatch rate of 79%, Beason and Franks (1974) study of 26 eggs had a hatch rate of 50%,
and in a study of 65 eggs Camfield et al (2010) had a hatch rate of 92%. In addition,
Camfield et al (2010) reported a Streaked Horned Lark hatch rate of 83% for a sample of
61 eggs in Washington State. Although there is a lot of variation in the hatch rates
26
found in these studies, they are all still higher than the 44% hatch rate reported in this
study. Low hatch rates do occur in wild populations of other endangered species, and
hatch rates of less than 50% have been routinely observed in a suite of endangered bird
species in New Zealand (Briskie and Mackintosh 2004, Congdon and Briskie 2010).
Although the mechanisms that explain variation in egg hatching proportions are
not fully understood (Knape et al. 2008), low egg hatchability can be a result of
environmental effects (such as calcium deficiency), contamination from pollutants
(DDT), and environmental changes which force large percentages of the population to
alter typical behaviors (Congdon and Briskie 2010). However, if these factors could
explain the low hatch rates among Streaked Horned Larks, then I would have expected
to see similar low hatch rates in other species at my study site that have similar diets
(Beason 1995, Poulin et al. 2006, Wheelwright 2008). Although there is a possibility that
hatch rates are affected by influences at the wintering grounds in the Willamette and
Columbia River Valleys, the more likely source of low hatch rates seems to be inbreeding
depression.
While the source of low hatch rates for Streaked Horned Larks at 13th Division
Prairie is unknown, Drovetski et al. (2005) hypothesize that Streaked Horned Lark
declines are due to genetic factors resulting from a population bottleneck. As Drovetski
et al. (2005) point out, several pieces of evidence point to a bottleneck leading to
inbreeding depression including, low genetic diversity, (Drovetski et al. 2005) combined
with a well documented contraction in range (Rogers 2000, Beauchesne and Cooper
2003, Pearson and Altman 2005, Stinson 2005), and genetic patterns consistent with a
bottleneck (Drovetski et al. 2005). Congdon and Briskie (2010) define population
27
bottlenecks as abrupt and temporary reductions in population size, in which populations
can suffer the loss of genetic variation and a subsequent increase in inbreeding.
Inbreeding depression can significantly affect the viability of a population through lower
birth weights, survival, reproductive success, and resistance to environmental stress,
predation and disease (Keller and Waller 2002). Briskie and Mackintosh (2004) found
that in 11 species of New Zealand birds that passed through a bottleneck of <150
individuals, there were significantly higher rates of hatching failure. Of the 4 Streaked
Horned Lark populations, (see Introduction) all but the Willamette Valley have Lark
numbers close to or below Briskie and Mackintosh’s (2004) threshold of <150
individuals. The Larks at my study site are part of the Puget lowlands population of
approximately 222 birds (Pearson and Altman 2005, Stinson 2005), and of those perhaps
25 breed at 13th Division Prairie. Given the estimated declines of 40% per year for the
three Washington populations (Pearson et al. 2008, Camfield et al. 2010) and the
infrequent dispersal of individuals between Puget lowland breeding sites (Pearson et al.
2008) it would seem that genetic exchange at 13th Division prairie will only become
more limited with each breeding season.
Fledglings per nest
Fledglings per nest is perhaps one of the most important vital rates, and one in
which Streaked Horned Larks fall far behind the other breeding species at 13th Division
Prairie. In essence, the number of fledglings produced each breeding season might be a
more telling metric of reproductive success than Mayfield nest survival, as was the case
with Streaked Horned Larks in 2009. In 2009, eight of the 11 nests survived (Mayfield
nest survival of 0.465) but they only produced a mean of 1.27±0.33 fledglings per nest.
28
Compare this to the numbers for Savannah Sparrows in the same year, in which many
nests were depredated and they had a Mayfield nest survival of 0.227 (less than half of
Streaked Horned Larks’) but still had a higher mean number of fledglings per nest with
1.67±0.43.
Proportion nests depredated
Predation is the leading cause of nest failure among grassland birds (Best 1978,
Johnson and Temple 1990) and the leading cause of Streaked Horned Lark nest failures
(Pearson and Altman 2005). The results of this study show that although predation was
indeed the leading cause of nest failure for all species studied, the rates of predation for
Streaked Horned Lark nests were not much different than those of the guild nests
(Streaked Horned Larks= 33%; guild nests=32%).
Proportion nests abandoned
Nest abandonment had a large impact on the breeding failures of Streaked
Horned Larks, but was confined to the 2007 season when almost a third of Lark nests
were abandoned. In fact, no other nests of any species were abandoned throughout
the study. This, of course, brings up the question of why so many Lark nests were
abandoned that particular year. In 2007 cameras were placed at nests in order to
identify the suite of predators at 13th Division Prairie. Although this could have
something to do with nest abandonment, it seems unlikely; after cameras were placed
at nests, all incubating females were back on the nests within 15-20 minutes (Pearson
and Hopey 2008). Although their study did not involve camera surveillance, Beason and
Franks (1974) report in their study of Horned Larks in Illinois that although some
29
abandonment occurred, it was never a result of researchers briefly checking a nest or
measuring eggs or young. Conversely, in a study of videotaped grassland bird nests in
North Dakota, 23% of 69 nests were abandoned within one day of camera installation
(Pietz and Granfors 2000), however it should be noted that this was not a study of
Horned Lark nests. Although the reason behind such high abandonment rates remains
unknown, it is clear that within the scope of this study, it seriously impacted Streaked
Horned Lark vital rates.
Annual Fecundity
This study found the two-year mean annual fecundity of Streaked Horned Larks
to be far below that of Savannah Sparrows (STHL=0.99 vs. Sav. Sparrow=3.25). The
annual fecundity of 0.99 female fledglings per female per year that I estimated is similar
to the 0.91 annual fecundity found by Pearson et al. (2008) for Streaked Horned Larks in
Washington from 2003-2006. Annual fecundity for Streaked Horned Larks in
Washington is much lower compared to the annual fecundity of 3.40 estimated by
Ricklefs and Bloom (1977) for a population of Horned Larks in Kansas, and an annual
fecundity of 1.75 estimated by Camfield et al (2010) for a population of Horned Larks in
British Columbia.
Nest Site Habitat Analysis
Habitat variables at the nest site scale were significantly different between Larks
and the guild in regards to substrate, but not significantly different in regards to
functional group variables. Additionally, Lark nests were located in less densely
vegetated areas than guild nests. Given that Savannah Sparrows (who nest in dense
30
vegetation) made up two-thirds of the guild, these findings seem to agree with the
literature (Beason 1995, Wheelwright 2008). In other studies, Streaked Horned Lark
nests were typically associated with sparsely vegetated areas (Rogers 2000, Pearson and
Hopey 2005), as was the case in my study.
Although functional group variables did not vary significantly between Lark and
guild nests, the difference in vegetation height did vary considerably (Lark 15.9cm vs.
Guild 21.9cm; p=0.11), as did the difference between Lark nests and guild nests in
percentage of non-vegetated hits (Lark 29.8% vs. Guild 14.5%; P=0.004). These findings
reinforce the conclusion that Larks tend to prefer nest site habitat that is short and
sparsely vegetated (Beason 1995, Rogers 2000, Pearson and Hopey 2005).
Impacts of Environmental Factors on Vital Rates
For each species at a shared breeding site, there are optimal sites, structures,
and locations for nests that are the result of the evolutionary importance of nest success
in regards to fitness (Cody 1981, Bekoff et al. 1989, Wiebe and Martin 1998). Although
this study found significant differences in nest site habitat variables between Lark and
guild nests, this difference did not appear to affect rates of predation on Lark nests.
Predation rates between the two groups were nearly identical, which suggested that
predation had a similar effect on nest success across species, regardless of
environmental factors.
Clutch size, though not one of the more informative metrics for an interspecific
comparative study, can also be influenced by environmental factors, but only to a very
small degree (Haywood and Perrins 1992). One way of assessing if environmental
31
factors are driving clutch size and other vital rates is to look at them in a year by year
comparison. If a vital rate shows declines for two or more species in the same year,
then environmental factors might be having an influence on that particular measure of
nesting success. For example, if a breeding area is hit by a late freeze during incubation,
hatch rates might be low for all species breeding in that area during the freeze. Figure 4
(see Results) presents yearly breakdowns of four vital rates that are critical to
productivity and shows that, although there were differences in fecundity between
2007 and 2009 for both Larks and the guild, there are no annual trends that matched
between Larks and the guild; guild nests were more productive in 2007, while Lark nests
were more productive in 2009. This lack of annual trends lends credence to, but does
not necessarily prove, the hypothesis that low productivity among Streaked Horned
Larks at 13th Division Prairie is primarily due to endogenous and not environmental
factors.
Still, exogenous environmental factors such as vegetative cover and thatch
should not be underestimated. Past Lark population declines were primarily driven by
loss of suitable breeding habitat (Pearson and Altman 2005), combined with changes in
vegetation at remaining breeding grounds (Crawford and Hall 1997). However, current
declines in key measures of productivity, such as fledglings per nest and hatch rate
would suggest that endogenous factors have also come to play an increasingly
detrimental role in Streaked Horned Lark declines.
32
RECOMMENDATIONS FOR FURTHER RESEARCH
If pending analyses of nest exclosure data support a decrease in rates of
predation with no increase in rates of abandonment, I recommend that the nest
exclosure program be continued. Although rates of predation were not significantly
different between Larks and the guild in this study, Lark numbers are so low that any
measure that increases nest success can only help. In addition, although predation rates
were nearly identical between Streaked Horned Larks and the nesting guild at 13th
Division Prairie, a depredated Lark nest has a much greater impact on that species’ total
population than, say, a depredated Savannah Sparrow nest.
Secondly, given the extremely low numbers of Streaked Horned Larks left at
each remaining breeding site, in addition to their strong nest site fidelity, I highly advise
further investigation into the possible effects of inbreeding depression in Streaked
Horned Larks, as well as research of potential strategies to help alleviate these effects.
For example, management might consider pursuing an egg exchange experiment in
order to increase genetic diversity among breeding sites. When nests are discovered,
egg age could be determined and a portion of those eggs could be swapped with eggs of
roughly the same age in other breeding areas. Although this is a relatively drastic
measure and the intricacies of such a program lie beyond the scope of this paper, I
nevertheless believe it should be investigated and pursued, particularly if the nest
exclosure program continues to employ personnel intensively searching for nests each
breeding season.
33
Lastly, given the dramatic loss of grassland habitat in the Pacific Northwest, it is
now more important than ever to maintain the quality of what little Streaked Horned
Lark habitat remains. In addition to protections being placed on the lands where
Streaked Horned Larks nest, management actions might focus on maintaining sparsely
vegetated areas in remaining grassland breeding grounds, and perhaps even
establishing new breeding sites. There are still lowland Puget grasslands such as 13th
Division Prairie that can continue to function as healthy breeding grounds for grassland
birds if protected and maintained.
34
LITERATURE CITED
Anderson, H. E. 2005. Nest predation of the Streaked Horned Lark (Eremophila alpestris
strigata) on lowland Puget prairie remnants, Washington State. The Evergreen
State College, Olympia, Washington.
Angermeier, P. L. and J. R. Karr. 1994. Biological integrity versus biological diversity as
policy directives- protecting biotic resources. Bioscience 44:690-697.
Barbour, M. G., J. H. Burk, and W. D. Pitts. 1980. Terrestrial plant ecology.
Benjamin/Cummings Publishing Co., Menlo Park, California, USA.
Beason, R. C. 1995. Horned Lark (Eremophila alpestris). The Birds of North America
Online (A. Poole, Ed.), Ithaca: Cornell Lab of Ornithology.
Beason, R. C. and E. C. Franks. 1974. Breeding behavior of the Horned Lark. Auk 91:6574.
Beauchesne, S. and J. Cooper. 2003. COSEWIC status report on the Horned Lark Strigata
subspecies Eremophila alpestris strigata. Status report prepared for the
Committee on the Status of Endangered Wildlife in Canada. COSEWIC
Secretariat c/o Canadian Wildlife Service, Environment Canada, Ottawa,
Ontario.
Bekoff, M., A. C. Scott, and D. A. Conner. 1989. Ecological analyses of nesting success in
Evening Grosbeaks. Oecologia 81:67-74.
Best, L. B. 1978. Field Sparrow reproductive success and nesting ecology. Auk 95:9-22.
Bishop, J. A. and W. L. Myers. 2005. Associations between avian functional guild
response and regional landscape properties for conservation planning.
Ecological Indicators 5:33-48.
Block, W. M., L. A. Brennan, and R. J. Gutierrez. 1986. The use of guilds and guildindicator species for assessing habitat suitability. Pages 109-113 in J. Verner, M.
L. Morrison, and C. J. Ralph, editors. Wildlife 2000: modeling habitat
relationships of terrestrial vertebrates. University of Wisconsin Press, Madison,
USA.
Briskie, J. V. and M. Mackintosh. 2004. Hatching failure increases with severity of
population bottlenecks in birds. Proceedings of the National Academy of
Sciences of the United States of America 101:558-561.
35
Brooks, R. P. and M. J. Croonquist. 1990. Wetland habitat and trophic response guilds
for wildlife species in Pennsylvania USA. Journal of the Pennsylvania Academy of
Science 64:93-102.
Cabin, R. J. and R. J. Mitchell. 2000. To Bonferroni or not to Bonferroni: when and how
are the questions. Bull. Ecol. Soc. Am. 81:246-248.
Camfield, A. F., S. F. Pearson, and K. Martin. 2010. Life history variation between high
and low elevation subspecies of horned larks Eremophila spp. Journal of Avian
Biology 41:273-281.
Cody, M. L. 1981. Habitat selection in birds - the roles of vegetation structure,
competitors, and productivity. Bioscience 31:107-113.
Congdon, N. M. and J. V. Briskie. 2010. Effect of population bottlenecks on the egg
morphology of introduced birds in New Zealand. Ibis 152:136-144.
Crawford, R. and H. Hall. 1997. Changes in the south Puget prairie landscape. Pages 1115 in Ecology and Conservation of the South Puget Sound Prairie Landscape.,
The Nature Conservancy, Seattle, WA.
Drovetski, S. V., S. F. Pearson, and S. Rohwer. 2005. Streaked horned lark Eremophila
alpestris strigata has distinct mitochondrial DNA. Conservation Genetics 6:875883.
Dunwiddie, P., E. Alverson, A. Stanley, R. Gilbert, S. Pearson, D. Hays, J. Arnett, E. Delvin,
D. Grosboll, and C. Marschner. 2006. The Vascular Plant Flora of the South Puget
Sound Prairies. Davidsonia 14(2): 51:69.
Haywood, S. and C. M. Perrins. 1992. Is clutch size in birds affected by environmentalconditions during growth. Proceedings of the Royal Society of London Series BBiological Sciences 249:195-197.
Johnson, R. G. and S. A. Temple. 1990. Nest predation and brood parasitism of tallgrass
prairie birds. Journal of Wildlife Management 54:106-111.
Keller, L. F. and D. M. Waller. 2002. Inbreeding effects in wild populations. Trends in
Ecology & Evolution 17:230-241.
Knape, J., M. Skoeld, N. Jonzen, M. Akesson, S. Bensch, B. Hansson, and D. Hasselquist.
2008. An analysis of hatching success in the great reed warbler Acrocephalus
arundinaceus. Oikos 117:430-438.
Knopf, F. L. 1994. Avian assemblages on altered grasslands. Pages 247-257 in J. R. J. Jehl
and N. K. Johnson, editors. A century of avifaunal change in western North
America: proceedings of an Internations Symposium at the centennial meeting
of the Cooper Ornithological Society. Series: Studies in Avian Biology,
Sacramento, CA.
36
Koenig, W. D. 1982. Ecological and social-factors affecting hatchability of eggs. Auk
99:526-536.
Kruckeberg, A. R. 1995. The natural history of Puget Sound country. The University of
Washington Press, Seattle, Washington.
Maher, W. J. 1979. Nestling diets of prairie passerine birds at Matador, Saskatchewan,
Canada. Ibis 121:437-452.
Martin, A. C. 1951. American wildlife & plants, a guide to wildlife food habits; the use of
trees, shrubs, weeds, and herbs by birds and mammals of the United States.
McGraw-Hill, New York,.
Martin, T. E. and G. R. Geupel. 1993. Nest-monitoring plots - methods for locating nests
and monitoring success. Journal of Field Ornithology 64:507-519.
Mayfield, H. F. 1975. Suggestions for calculating nest success. Wilson Bulletin 87:456466.
Meunier, M. and J. Bedard. 1984. Nestling foods of the savannah sparrow. Canadian
Journal of Zoology-Revue Canadienne De Zoologie 62:23-27.
Oregon Department of Fish and Wildlife. 2006. Oregon Conservation Strategy. Oregon
Department of Fish and Wildlife, Salem, OR.
Osborne, J. W. and A. Overbay. 2004. The power of outliers (and why researchers should
ALWAYS check for them). Practical Assessment, Research & Evaluation 9.
Pearson, S. F. and B. Altman. 2005. Range-wide Streaked Horned Lark (Eremophila
alpestris strigata) Assessment and Preliminary Conservation Strategy.
Washington Department of Fish and Wildlife, Wildlife Science Division, Olympia,
WA.
Pearson, S. F., A. F. Camfield, and K. Martin. 2008. Streaked Horned Lark fecundity,
survival, population growth and site fidelity: Research progress report.
Washington Department of Fish and Wildlife, Wildlife Science Division, Olympia,
WA.
Pearson, S. F. and M. Hopey. 2005. Streaked Horned Lark Nest Success, Habitat
Selection, and Habitat Enhancement Experiments for the Puget Lowlands,
Coastal Washington and Columbia River Islands. Washington Dept. of Natural
Resources, Olympia, WA.
Pearson, S. F. and M. Hopey. 2008. Identifying streaked horned lark nest predators.,
Washington Department of Fish and Wildlife, Wildlife Science Division, Olympia,
WA.
Peterjohn, B. G. and J. R. Sauer. 1993. North American Breeding Bird Survey annual
summary, 1990-1991. Bird Popul. 1:52-67.
37
Pickwell, P. B. 1931. The Prairie Horned Lark. St. Louis Acad. Sci. Trans. 27:1-153.
Pietz, P. J. and D. A. Granfors. 2000. Identifying predators and fates of grassland
passerine nests using miniature video cameras (vol 64, pg 71, 2000). Journal of
Wildlife Management 64:1099-1099.
Poulin, R. G., G. S.D., and B. R.M. 2006. Common Nighthawk (Chordeiles minor). The
Birds of North America Online (A. Poole, Ed.). Cornell Lab of Ornithology, Ithaca,
NY.
Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43:223-225.
Ricklefs, R. E. and G. Bloom. 1977. Components of avian breeding productivity. Auk
94:86-96.
Robbins, C. S., S. Droege, and J. R. Sauer. 1989. Monitoring bird populations with
breeding bird survey and atlas data. Annales Zoologici Fennici 26:297-304.
Rogers, R. E. 2000. The status and microhabitat selection of Streaked Horned Lark,
Western Bluebird, Oregon Vesper Sparrow, and Western Meadowlark in
Western Washington. The Evergreen State College, Olympia, Washington.
Root, R. B. 1967. Niche exploitation pattern of Blue-Gray Gnatcatcher. Ecological
Monographs 37:317-&.
Samson, F. and F. Knopf. 1994. Prairie conservation in North America. Bioscience
44:418-421.
Schapaugh, A. W. 2009. The Dynamics and Viability of the Endangered Streaked Horned
Lark (Eremophila alpestris strigata). The Evergreen State College, Olympia, WA.
Severinghaus, W. D. 1981. Guild theory development as a mechanism for assessing
environmental-impact. Environmental Management 5:187-190.
Stinson, D. W. 2005. Draft Washington State Status Report for the Mazama Pocket
Gopher, Streaked Horned Lark, and Taylor's Checkerspot. Washington
Department of Fish and Wildlife, Wildlife Science Division, Olympia, WA.
Team, R. D. C. 2006. A language and environment for statistical computing. R
Foundation for statistical computing, Vienna, Austria.
Verner, J. 1983. An integrated system for monitoring wildlife on the Sierra National
Forest. Transactions of the North American Wildlife and Natural Resources
Conference 48:355-366.
Verner, J. 1984. The guild concept applied to management of bird populations.
Environmental Management 8:1-13.
38
Vickery, P. D., M. L. Hunter, and S. M. Melvin. 1994. Effects of habitat area on the
distribution of grassland birds in Maine. Conservation Biology 8:1087-1097.
Wheelwright, N. T. 2008. Savannah Sparrow (Passerculus sandwichensis). J. D. Rising,
editor. The Birds of North America Online (A. Poole, Ed.), Ithaca: Cornell Lab of
Ornithology.
Wiebe, K. L. and K. Martin. 1998. Costs and benefits of nest cover for ptarmigan:
changes within and between years. Animal Behaviour 56:1137-1144.
39
