Volume 135, 2018, pp. 733–747
DOI: 10.1642/AUK-17-195.1
RESEARCH ARTICLE
Variation in nest characteristics and brooding patterns of female Black-
throated Blue Warblers is associated with thermal cues
Maria G. Smith,1,2a* Sara A. Kaiser,1,2b T. Scott Sillett,3 and Michael S. Webster1,2
1 Department of Neurobiology and Behavior, Cornell University, Ithaca, New York, USA
2 Macaulay Library, Cornell Lab of Ornithology, Ithaca, New York, USA
3 Migratory Bird Center, Smithsonian Conservation Biology Institute, National Zoological Park, Washington, D.C., USA
a Current address: Department of Ecology and Evolutionary Biology, Princeton University, Princeton, New Jersey, USA
b Current address: Center for Conservation Genomics, Smithsonian Conservation Biology Institute, National Zoological Park,
Washington, D.C., USA
* Corresponding author: mgsmith@princeton.edu
Submitted September 30, 2017; Accepted March 26, 2018; Published June 6, 2018
ABSTRACT
Thermal variation poses a problem for nesting birds and can result in reduced offspring growth rates and survival. To
increase the thermal stability of the nest, females can adjust nest characteristics and nest attendance in response to
changes in environmental conditions. However, it is unclear how and to what extent females modify parental
behaviors during various stages of offspring development. We tested the hypothesis that females adjust nest
characteristics and brooding patterns in response to thermal variation during the nest-building and nestling stages,
respectively. We examined elevational variation in nest location, nest construction, and brooding patterns in the
migratory Black-throated Blue Warbler (Setophaga caerulescens) across a 28C gradient at the Hubbard Brook
Experimental Forest, New Hampshire, USA. Density of woody stems at nest sites and nest wall thickness increased from
low to high elevation, corresponding to decreasing temperatures, but we found no relationship between weather
during nest building and nest characteristics. However, weather during the nestling stage was associated with female
brooding patterns: at lower temperatures and with higher rainfall, females spent more time off the nest, which was
associated with lower nestling mass near fledging. These results suggest that thermal cues during nest building may
be unreliable as predictors of future conditions for developing nestlings and also that females might favor their own
self-maintenance and compromise nestling growth under adverse thermal conditions.
Keywords: behavioral plasticity, brooding, environmental cues, nest construction, nest microclimate, nestling
condition
La variacio´n en las caracterı´sticas del nido y en los patrones de crianza de las hembras de Setophaga
caerulescens esta´ asociada con sen˜ales te´rmicas
RESUMEN
La variacio´n te´rmica representa un problema para las aves que anidan y puede resultar en una reduccio´n en las tasas de
crecimiento y en la supervivencia de las cr´ıas. Para aumentar la estabilidad te´rmica del nido, las hembras pueden ajustar las
caracter´ısticasdel nidoy lapresencia enel nidoen respuestaa los cambios en las condicionesambientales. Sinembargo, esta´
poco claro co´mo y en que´ medida las hembrasmodifican los comportamientos de los progenitores durante varios estadios
del desarrollode las cr´ıas. Evaluamos la hipo´tesisque lashembras ajustan las caracterı´sticas del nidoy los patronesdecrianza
en respuestaa la variacio´n te´rmicadurante las etapasdeconstruccio´ndel nidoydepolluelos, respectivamente. Examinamos
la variacio´n altitudinal en la localizacio´n del nido, la construccio´n del nido y los patrones de crianza en la especie migratoria
Setophagacaerulescensa trave´s deungradientede28CenelBosqueExperimentalHubbardBrook,NH, EEUU. Ladensidadde
tallos len˜osos en los sitios de anidacio´n y el espesor de la pared de los nidos aumento´ desde la elevacio´n baja hacia la alta,
correspondiendo a una disminucio´n de la temperatura, pero no encontramos una relacio´n entre el clima durante la
construccio´n del nido y las caracter´ısticas del nido. Sin embargo, el clima durante la etapa de polluelo estuvo asociado con
los patrones de crianza de la hembra: a bajas temperaturas y con mayor precipitacio´n, las hembras pasaron ma´s tiempo
afuera del nido, lo que estuvo asociado con una menor masa de los polluelos cerca del emplumamiento. Estos resultados
sugieren que las sen˜ales te´rmicas durante la construccio´n del nido pueden ser poco confiables como predictores de las
futuras condiciones para desarrollar volantones y tambie´n que las hembras podrı´an favorecer su propio auto
mantenimiento y comprometer el crecimiento de los volantones bajo condiciones te´rmicas adversas.
Palabras clave: condicio´n del volanto´n, construccio´n del nido, crianza, microclima del nido, plasticidad
comportamental, sen˜ales ambientales
Q 2018 American Ornithological Society. ISSN 0004-8038, electronic ISSN 1938-4254
Direct all requests to reproduce journal content to the AOS Publications Office at pubs@americanornithology.org
INTRODUCTION
Temperature fluctuations pose a challenge for nesting
birds (DuRant et al. 2013). Offspring developing under
extreme or highly variable temperatures can suffer reduced
growth rates (Olson et al. 2006), leading to lower survival
(Ardia 2013). Such temperature conditions might be
particularly problematic for altricial species because their
young are incapable of thermoregulating during early
development (Whittow and Tazawa 1991, Dawson et al.
2005). Altricial species should therefore use environmental
cues to modify their parental behaviors to maintain
thermal stability for their developing offspring and to
maximize offspring survival and their own fitness (Britt
and Deeming 2011, Deeming et al. 2012). However, there is
still uncertainty regarding how and to what extent females
respond adaptively to environmental cues by adjusting
their parental behaviors during different stages of offspring
development.
Females can increase thermal stability at their nests by
adjusting nest location and construction (Walsberg 1985,
Windsor et al. 2013). For example, selection of nest sites in
densely vegetated or sheltered areas can influence nest
microclimate by decreasing cooling from wind and
radiative heat loss (Walsberg 1985, D’Alba et al. 2009).
Constructing nests in patches of dense, woody vegetation
may minimize cooling of nest contents and shield nest
contents from precipitation (Walsberg 1985). Microcli-
mate can, in turn, influence nest construction and
insulation properties. More protected nest sites may
require less investment by females during nest building
to insulate the nest and less effort during incubation and
brooding to maintain thermal stability. Building more
insulated nests in cooler areas has been shown to mitigate
the effects of lower ambient temperatures. For instance,
birds at high elevations and latitudes, where breeding
conditions are cooler and more variable, may construct
nests with denser or thicker walls than birds at low
elevations (Kern and van Riper 1984) and latitudes
(Rohwer and Law 2010, Crossman et al. 2011, Mainwaring
et al. 2014). However, nest wall thickness generates a trade-
off, in that thicker nests lose heat less rapidly (Whittow
and Berger 1977) but absorb more water and dry more
slowly, increasing evaporative cooling (Rohwer and Law
2010). Females may adjust nest characteristics differently
across populations, given the relative importance of
cooling and water absorption (Botero-Delgadillo et al.
2017). Moreover, few studies have been designed to test for
a direct link between functional differences in nest
characteristics and temperature and rainfall during nest
building (Sua´rez et al. 2005, Deeming et al. 2012,
Mainwaring et al. 2014).
Nest temperature also can be mediated through changes
in female attentiveness at the nest (Sanz and Tinbergen
1999, DuRant et al. 2013). For example, adjustments in
incubation patterns in response to changes in environ-
mental conditions, such as temperature and rainfall, are
well documented (Haftorn 1978, 1988, Morton and
Pereyra 1985, Joyce et al. 2001, Kovarˇ´ık et al. 2009).
Generally, females take shorter incubation off-bouts under
cooler and wetter conditions to maintain a relatively
constant nest temperature (DuRant et al. 2013). However,
other adverse conditions for the nest may include times
when females experience increased energetic demands,
potentially creating conflicts over self-maintenance and
incubating or brooding (Morton and Pereyra 1985,
MacDonald et al. 2013). Overall, female adjustments in
brooding patterns in response to environmental variation
and how these patterns are influenced by the nest
microclimate and nest construction are not well under-
stood.
We tested the hypothesis that females adjust nest
characteristics and brooding patterns in response to
thermal variation during different nest stages. We exam-
ined nest-building and brooding behaviors in a population
of migratory Black-throated Blue Warblers (Setophaga
caerulescens) breeding on 3 study plots at different
elevations that spanned a 28C gradient (Rodenhouse et
al. 2003). We first assessed plot-level variation in nest
location, nest construction, and brooding patterns. Next,
we investigated whether nest characteristics were associ-
ated with temperature and rainfall experienced during nest
building. We then examined whether nest characteristics
and/or variation in temperature and rainfall during the
nestling stage were correlated with female brooding
patterns. Lastly, we determined whether brooding patterns
were associated with the mass of nestlings near fledging,
which can be a determinant of fledgling survival (Monro´s
et al. 2002, Gren˜o et al. 2008).
METHODS
Study Population
We studied a breeding population of the migratory Black-
throated BlueWarbler at the Hubbard Brook Experimental
Forest (HBEF), Woodstock, New Hampshire, USA
(43.938N, 71.758W). The study area is a northern
hardwood forest dominated by sugar maple (Acer saccha-
rum), American beech (Fagus grandifolia), and yellow
birch (Betula alleghaniensis), with red spruce (Picea
rubens), balsam fir (Abies balsamea), and white birch (B.
papyrifera) increasing in abundance on the ridges
(Schwarz et al. 2003). Basic breeding biology for our study
population is detailed in Holmes et al. (2017). In brief,
females choose nest sites in the understory, primarily in
dense hobblebush (Viburnum lantanoides), and build
nests over 3–5 days. At the HBEF, nests are constructed
with strips of bark from yellow or white birch and formed
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
734 Female behavioral responses to thermal variation M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster
by spider webs. Nests are lined with rootlets, pine needles,
moss, and mammal hair. Females lay 1 egg day!1 (mean
clutch size ¼ 3.6, range: 2–5 eggs) and incubate without
assistance from males for ~12 days. Both sexes feed
nestlings (mostly lepidopteran larvae) for ~9 days until
fledging. Females brood nestlings frequently, particularly
early in the nestling stage.
Field Methods
We collected data on 3 study plots classified as low-
elevation (250–350 m; 85 ha), mid-elevation (450–600 m;
65 ha), and high-elevation (750–850 m; 35 ha) during
May–August, 2010–2013. All plots were on south-facing
slopes. We marked adults, mapped male territories, and
monitored all nesting attempts. Adults were captured in
mist nets and marked with a unique combination of 3
colored leg bands and a federal leg band. Nests were found
through intensive searching and were monitored every
other day throughout all nest stages, with daily checks near
building completion, clutch initiation and completion, and
anticipated hatch and fledge dates. We weighed and placed
a federal band on young on day 6 (hatching¼ day 0) of the
nestling stage. Because brood size can affect the thermal
environment of the nest (Chaplin et al. 2002), we
accounted for cases of brood reduction by calculating
the average brood size over nestling days 3–5.
We measured the density of woody stems at the nest site
and nest wall thickness after each nest became inactive.
Stem density was measured as the number of stems per
square meter within a 1 m radius of the nest. Mean nest
height was 53.2 cm (range: 11–700 cm). A horizontal plane
was created on the ground by placing the end of a pole 1 m
in length underneath the nest and rotating the pole in a
circle around the nest. We counted all bases of woody
stems .0.1 m in height that emerged from the ground
within the plane. Nest wall thickness was measured as the
difference in the diameter of the outer nest cup and inner
nest cup (both measured to the nearest 0.1 cm). We
measured stem density at all nest sites (n¼ 490) and nest
wall thickness of intact nests that were not deformed by
weather or nest depredation (n ¼ 388).
Temperatures inside the nest cup and from the ambient
microclimate of each nest site were measured with iButton
thermochrons (DS1921G-F5; Embedded Data Systems,
Lawrenceburg, Kentucky, USA). Similar thermochron
models have been shown to be accurate within 60.58C
of reference temperatures when tested in cold and warm
conditions and are highly consistent (Davidson et al. 2003,
Hubbart et al. 2005, Smith et al. 2010). The afternoon
before data recording, we placed one thermochron in the
nest cup to collect ‘‘nest temperature.’’ We affixed a second
thermochron to a small piece of cardboard 5 m from the
nest, and tied to vegetation of a similar type and height as
the vegetation supporting the nest, to collect mean
maximum daily ambient ‘‘microclimate temperature’’ at
the nest site. Thermochrons were automated to begin
recording near dawn (0400 hours) on nestling day 3 and
recorded temperature to the nearest 0.58C at 4 min
intervals until they were retrieved after the nest fledged or
failed.
For each nest, we calculated weather conditions during
nest building (5 days prior to clutch initiation) and the
early nestling stage (nestling days 3–5) when brooding is
most likely to occur. When nests were found after laying,
the date of clutch initiation was estimated to within 1 day
by back-dating 1 egg day!1 from the date of the first egg of
known lay date and/or the hatch date. We assumed a 12-
day incubation period (Holmes et al. 2017). Plot-level daily
ambient temperature (8C) and rainfall (millimeters) were
measured at 3 permanent U.S. Forest Service weather
stations adjacent to each study plot. On average, birds
nesting at high elevations experienced cooler (28C) and
wetter (30 mm more rainfall) conditions over the course of
the study than birds nesting at low elevations. We
measured total daily rainfall accumulated at weather
stations during both the nest-building and nestling stages.
‘‘Nest-building temperature’’ is the mean maximum daily
temperature measured at weather stations. For ‘‘nestling-
stage temperature,’’ we used microclimate temperature
collected by thermochrons (defined above), which better
represents the temperature experienced by females nesting
in the understory than temperature measured at weather
stations.
We estimated female brooding patterns using data
recorded by thermochrons. We paired and aligned
temperature data from nest and ambient thermochrons
at a nest site by time of day and calculated the difference
between nest and ambient temperatures for each 4 min
interval. We defined female off-bouts as intervals with a
monotonic decrease in the difference between nest and
ambient temperatures of #0.58C, indicating that the nest
was cooling in relation to the ambient temperature. We
identified off-bouts by visualizing output files produced by
Rhythm (Cooper and Mills 2005) in Raven Pro 1.4
(Bioacoustics Research Program 2011). We provide a
validation of thermochrons for detecting female off-bouts
(Appendix). We excluded data collected after sunset (i.e.
between 2030 and 0500 hours) because the difference
between nest and ambient temperature showed a contin-
uous decrease during this time, making off-bout assign-
ment difficult without a validation method (e.g., video
recording of nests at night). From these data, we calculated
3 measures of brooding patterns: mean off-bout duration,
mean off-bout frequency, and total proportion of time off
the nest over the early nestling stage (i.e. nestling days 3–
5).We also calculated these brooding measures during 2 hr
time windows to analyze fine-scale patterns because
averaging over several days could potentially obscure
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster Female behavioral responses to thermal variation 735
short-term effects of temperature on brooding. We
excluded from analyses thermochron data collected after
day 5, when nestlings began to generate more heat, which
reduced the accuracy of off-bout detection by thermo-
chrons (for details, see Appendix).
Using thermochron data, we identified a total of 25,801
daytime off-bouts over nestling days 3–5 from 193 nests.
In models of brooding patterns, we included only nests
with complete temperature data from 0500–2030 hours on
nestling days 3–5 (n ¼ 146). Nests with incomplete
temperature data resulted from late deployment of
thermochrons due to logistical difficulties or failure to
find the nest prior to nestling day 3 (n¼ 27), nest failures
prior to nestling day 5 (n ¼ 19), and thermochron
malfunctioning (n ¼ 1).
Statistical Analyses
We used a 2-stage process to investigate how nest
characteristics and brooding patterns varied with (1) study
plot and (2) nest-stage-specific temperature and rainfall.
We first used analysis of variance (ANOVA) and Tukey’s
test with a¼ 0.05 to test for significant differences in nest
location, nest construction, and brooding patterns among
study plots at low-, mid-, and high-elevation zones. We
calculated g2 (Lakens 2013) and the estimated change in
mean values between elevations (b) with associated 95%
confidence intervals (CI) as measures of effect size for
ANOVAs and Tukey’s tests, respectively, using the R
package ‘‘lsr’’ (Navarro 2015). We then built linear mixed
models (LMMs) and used an information-theoretic
approach to identify which weather and nest-site predic-
tor(s) best explained differences in nest characteristics,
female brooding behavior, and nestling mass. We con-
structed LMMs using ‘‘lme4’’ (Bates et al. 2015) and fitted
models with maximum likelihood (ML) to account for
different fixed-effects model structures (Zuur et al. 2009).
Study plot and elevation were not included in LMMs
because elevation was correlated with temperature (linear
model; P , 0.001, adjusted R2 ¼ 0.04). We conducted all
analyses in R 3.2.4 (R Core Team 2016).
For LMMs, we used our knowledge of the study system
to select 2–4 predictor variables per model to test a priori
hypotheses that nest microclimate was correlated with nest
characteristics, brooding behavior, and nestling mass.
Models testing for associations between microclimate
and brooding and between microclimate and nestling
mass also included nest-site variables (e.g., nest wall
thickness) as predictors. Each candidate model set (range:
3–17 models) included a global model with all fixed
effects, an additive model with each combination of fixed
effects, and a model with the intercept only. We examined
models for homogeneity of variance by plotting model
residuals against fitted values (Pinheiro and Bates 2000)
and for normality of model residuals by visually inspecting
quantile-quantile plots (Crawley 2013). We conducted a
square-root transformation of stem density and a natural-
log transformation of mean bihourly off-bout duration to
satisfy model assumptions (McDonald 2014).
We examined associations of (1) temperature and
rainfall during nest building with square-root-transformed
stem density at the nest site (n ¼ 297 nests); (2)
temperature and rainfall during nest building and stem
density with nest wall thickness (n ¼ 297 nests); and (3)
microclimate temperature and rainfall during the nestling
stage, stem density, and nest wall thickness with off-bout
duration, off-bout frequency, and proportion of time off
the nest (n ¼ 144 nests). To examine potential short-term
adjustments in brooding patterns, we built 3 LMMs (n ¼
104 nests) for which the response variables were log-
transformed bihourly off-bout duration, off-bout frequen-
cy, and proportion of time off the nest during 2 hr periods.
We divided daylight (0500–2030 hours) into eight 2 hr
periods (1.5 hr for the last period, 1900–2030 hours), for a
total of twenty-four 2 hr periods over the 3 days, and
included mean microclimate temperature over each 2 hr
period (‘‘bihourly microclimate temperature’’) as a fixed
effect. In summary, average brooding patterns were
examined both over the early nestling stage and bihourly;
rainfall and nest characteristics were included as fixed
effects only in models for brooding patterns over the early
nestling stage. Brood size has been found to be associated
with variation in brooding: females spend less time
brooding larger broods because they retain heat better
than smaller broods during periods of parental inattention
(Clark 1985, Sanz and Tinbergen 1999, Chastel and
Kersten 2002, Archard et al. 2006). Therefore, in all 6
model sets for brooding patterns, we included mean brood
size as a predictor in each model to account for known
variation. We included year and female identity (or, for
bihourly models, nest identity nested within female
identity) as random effects in each model. We did not
include year as a random effect in the final models for
mean off-bout duration (nestling stage), proportion of time
off the nest (nestling stage), and mean off-bout duration
(bihourly) because its estimated variance was equal to zero
(Zuur et al. 2009).
The final LMMs examined the factors influencing mean
nestling mass near fledging, which has been correlated
with survival probability in birds (Monro´s et al. 2002,
Gren˜o et al. 2008). Each analysis used a subset of nests that
had the same set of measurements without missing data.
The 2 candidate model sets were (1) weather and brooding
(n¼ 149 nests), which included microclimate temperature,
rainfall during the nestling stage, and proportion of time
off the nest (nestling stage) as fixed effects; and (2) nest
location and construction (n¼ 169 nests), which included
stem density and nest wall thickness as fixed effects. In all
models for mean nestling mass, we included (1) mean
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
736 Female behavioral responses to thermal variation M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster
brood size to account for known variation and (2) hatch
date to account for unmeasured conditions (e.g., food
resources) that vary over the breeding season (Holmes et
al. 1986) and could influence nestling mass. To further
examine potential fitness consequences on nest success,
we constructed 2 generalized linear mixed models with
fledge (1) or fail (0) as a response variable with a binomial
error distribution with a logit link function. We included
the same fixed effects as for the nestling mass models but
removed hatch date and mean brood size to facilitate
model convergence. Model-selection and model-averaging
results were qualitatively and quantitatively similar to the
nestling mass results, except that proportion of time off
the nest was not associated with fledging probability
(Supplemental Material Tables S3 and S4).
Model selection for all LMM candidate model sets was
based on second-order Akaike’s Information Criterion
adjusted for small sample sizes (AICc; Burnham and
Anderson 2002) using ‘‘MuMIn’’ (Barton 2014). We report
maximized log likelihood (logL) and number of estimated
parameters (K) for each model. We ranked models within a
candidate set by the difference in AICc between each
model and the best model (Di). We averaged all models in
each candidate model set to obtain coefficient estimates
for each fixed effect. We summed Akaike weights (wi)
across all models containing a given fixed effect to evaluate
the importance of each model parameter. If a parameter
had a summed wi #0.75 and a confidence interval that did
not include zero, we considered it important in explaining
a given response variable. If a parameter had a 95% CI
containing zero, we concluded that the parameter and
response variable were not associated. The full list of
models included in each candidate model set is given in
Supplemental Material Table S1. We detected no multi-
collinearity among variables included as fixed effects in the
global model for each candidate model set; all variance
inflation factors were ,1.5 (Supplemental Material Table
S2), well below the threshold of 10 suggested by Hair et al.
(2010). Results are reported as means 6 SE.
RESULTS
Nest characteristics and brooding patterns differed among
study plots (Figure 1). Mean stem density and nest wall
thickness increased with plot elevation (ANOVA; stem
density: F¼ 5.5, df¼ 2 and 487, P¼ 0.004, g2¼ 0.02; nest
wall thickness: F ¼ 5.3, df ¼ 2 and 385, P ¼ 0.005, g2 ¼
0.03). Mean stem density at nest sites was significantly
higher at high-elevation sites than at low-elevation sites
(Tukey’s test; P¼ 0.01, b¼ 0.49 stems m!2, 95% CI: 0.09–
0.90) and mid-elevation sites (P ¼ 0.005, b ¼ 0.48 stems
m!2, 95% CI: 0.12–0.85), but we detected no statistical
difference in stem density between high- and mid-
elevation sites (P ¼ 0.99, b ¼ 0.01 stems m!2, 95% CI:
!0.31 to 0.33). Mean nest wall thickness was significantly
greater at high-elevation sites than at low-elevation sites (P
¼ 0.004, b¼ 0.46 cm, 95% CI: 0.13–0.79), but we detected
no statistical difference between mid- and low-elevation
sites (P ¼ 0.06, b ¼ 0.30 cm, 95% CI: !0.01 to 0.60) or
between high- and mid-elevation sites (P ¼ 0.32, b ¼ 0.17
cm, 95% CI: !0.10 to 0.43). Variation in mean off-bout
duration and proportion of time off the nest differed
significantly among plots (ANOVA; off-bout duration: F¼
5.5, df¼2 and 190, P¼0.005, g2¼0.05; proportion of time
off nest: F ¼ 3.6, df ¼ 2 and 190, P ¼ 0.03, g2 ¼ 0.04).
Females had longer off-bouts and spent more time off their
nests at high-elevation sites than at mid-elevation sites
(Tukey’s test; off-bout duration: P¼ 0.003, b¼ 23.50 s, 95%
CI: 6.59–40.41; proportion of time off nest: P ¼ 0.03, b ¼
0.02, 95% CI: 0.002–0.04). However, we detected no
statistical difference in brooding patterns between mid-
and low-elevation sites (off-bout duration: P ¼ 0.31, b ¼
!12.21 s, 95% CI: !32.00 to 7.59; proportion of time off
nest: P¼ 0.32, b¼!0.01, 95% CI:!0.04 to 0.01) or between
high- and low-elevation sites (off-bout duration: P¼ 0.43,
b¼ 11.29 s, 95% CI:!10.19 to 32.77; proportion of time off
nest: P¼0.76, b¼0.01, 95% CI:!0.02 to 0.03).We detected
no statistical difference in mean off-bout frequency among
plot elevations (ANOVA; F¼ 0.1, df¼ 2 and 190, P¼ 0.90,
g2 ¼ 0.001).
Daily brooding patterns varied over the early nestling
stage (Figure 2). Mean off-bout duration and proportion of
time off the nest decreased from dawn to dusk and from
nestling day 3 to day 5. Mean off-bout frequency remained
relatively constant on day 3 and peaked between 0900 and
1100 hours on days 4 and 5. Ambient temperature at nest
sites reached a maximum at midday (1300–1500 hours).
Variation in stem density and nest wall thickness was
not associated with temperature or rainfall during nest
building (Tables 1 and 2). The top-ranked model for stem
density included the intercept only (Table 1). Confidence
intervals for nest-building temperature and rainfall in
models for both stem density and nest wall thickness
encompassed zero (Table 2). Although weather did not
explain variation in nest construction, stem density was
associated with nest wall thickness; nest walls were thicker
at nest sites with higher stem density.
Variation in some female brooding patterns was
associated with weather, but none was influenced by nest
characteristics (Tables 1 and 2). Mean off-bout duration
and proportion of time off the nest, measured over the
nestling stage or at bihourly intervals, increased with
decreasing microclimate temperatures. Mean off-bout
duration and proportion of time off the nest increased
with increasing rainfall during the nestling stage and with
decreasing brood size (Table 2). Mean off-bout duration
and proportion of time off the nest were not associated
with stem density or nest wall thickness; confidence
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster Female behavioral responses to thermal variation 737
intervals encompassed zero. Mean off-bout frequency over
the nestling stage, which showed little variation across plot
elevations (Figure 1), was not associated with any of the
measured environmental factors; the top-ranked model
included the intercept only (Table 1). Mean bihourly off-
bout frequency increased with decreasing microclimate
temperature (Table 2).
Mean nestling mass near fledging was associated with
the proportion of time females were off nests and with
mean brood size (Tables 1 and 2). Females that spent more
time off nests fledged offspring with lower mass. By
contrast, nest microclimate temperature, rainfall, stem
density, nest wall thickness, and hatch date (Table 2) were
not correlated with nestling mass.
DISCUSSION
Maintaining a stable thermal environment is necessary for
the development of altricial young, yet the extent to which
females respond behaviorally to changes in nest microcli-
FIGURE 1. Median values, quartiles, and range in parameters across low-elevation, mid-elevation, and high-elevation study plots at
Hubbard Brook Experimental Forest, New Hampshire, USA, May–August, 2010–2013. Parameters include (A) stem density at the nest
site, (B) nest wall thickness, (C) mean off-bout duration, (D) mean off-bout frequency, and (E) proportion of time off the nest for
female Black-throated Blue Warblers. Letters denote statistically significant differences (P, 0.05). Sample sizes are given along the x-
axes.
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
738 Female behavioral responses to thermal variation M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster
mate is not well understood. At the scale of our 3 study
plots, stem density at nest sites and nest wall thickness
increased from low to high elevation, corresponding to a
broad pattern of decreasing temperatures (Rodenhouse et
al. 2003). However, we found no direct link between
temperature and rainfall during nest building and either
stem density at nest sites or nest wall thickness.
Adjustments in brooding patterns, by contrast, were
associated with weather. Moreover, females that brooded
less under cooler and wetter conditions produced nestlings
with lower mass near fledging. This indicates that females
might favor their own self-maintenance and compromise
nestling growth under adverse thermal conditions.
Stem density at nest sites and nest wall thickness were
generally greater at higher elevations, although the propor-
tion of the variance in these factors explained by study plot
was low. Breeding conditions are cooler and wetter and
understory vegetation is more dense at high elevations
(Rodenhouse et al. 2003, 2008). The finding that stem
density at nest sites increased with elevation could reflect
the overall pattern of higher stem density at higher
elevations. However, previous work at the HBEF has shown
that female Black-throated Blue Warblers select nest sites
with denser vegetation than at random points, which may
be due, in part, to microclimate benefits (Holway 1991).
Moreover, females constructed nests with thicker walls at
higher elevation, which suggests a mechanism to insulate
FIGURE 2. Mean (6 SE) brooding patterns of female Black-throated Blue Warblers, measured every 2 hr during the daytime (0500–
2030 hours) over nestling days 3–5 (n¼146 nests), at Hubbard Brook Experimental Forest, New Hampshire, USA, May–August, 2010–
2013. Vertical black lines separate days.
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster Female behavioral responses to thermal variation 739
their developing offspring from cooler temperatures. These
results are also consistent with other studies that have found
that nest wall thickness and/or stem density was greater at
higher elevations or latitudes (Kern and van Riper 1984,
Rohwer and Law 2010, Crossman et al. 2011).
Differences in nest characteristics were, however, not
associated with thermal cues or rainfall during nest
building. Nest wall thickness increases with decreasing
temperatures during nest building in some open-cup-
nesting species (e.g., Eurasian Skylark [Alauda arvensis]
TABLE 1. Model-selection results examining the effects of microclimate temperature, rainfall, stem density, nest wall thickness, and
mean brood size on nest location, nest construction, brooding patterns (nestling stage and bihourly), and nestling mass near
fledging in Black-throated Blue Warblers at Hubbard Brook Experimental Forest, New Hampshire, USA, May–August, 2010–2013.
Model type Response variable Model logL K Di wi
Nest characteristics Stem density Intercept only !486.74 4 0.00 0.52
Nest-building temperature !486.61 5 1.81 0.21
Nest-building rainfall !486.69 5 1.96 0.20
Nest wall thickness Nest-building temperature þ stem density !390.63 6 0.00 0.34
Nest-building rainfall þ stem density !390.94 6 0.63 0.25
Stem density !392.09 5 0.85 0.22
Nest-building temperature þ nest-building rainfall
þ stem density
!390.19 7 1.22 0.19
Brooding patterns Mean off-bout
duration
(nestling stage)
Microclimate temperature þ nestling-stage rainfall
þ brood size
!730.79 6 0.00 0.32
Nestling-stage rainfall þ brood size !732.83 5 1.89 0.12
Microclimate temperature þ nestling-stage rainfall
þ nest wall thickness þ brood size
!730.65 7 1.92 0.12
Microclimate temperature þ nestling-stage rainfall
þ stem density þ brood size
!730.66 7 1.93 0.12
Mean off-bout
frequency
(nestling stage)
Intercept only 35.15 4 0.00 0.34
Proportion of time
off nest
(nestling stage)
Microclimate temperature (nestling stage) þ
nestling-stage rainfall þ stem density þ brood
size
243.52 7 0.00 0.27
Microclimate temperature (nestling stage) þ
nestling-stage rainfall þ brood size
242.04 6 0.74 0.19
Microclimate temperature (nestling stage) þ
nestling-stage rainfall þ stem density þ nest wall
thickness þ brood size
243.69 8 1.89 0.11
Mean off-bout
duration
(bihourly)
Microclimate temperature (bihourly) þ brood size !253.62 6 0.00 1.00
Mean off-bout
frequency
(bihourly)
Microclimate temperature (bihourly) þ brood size !244.90 7 0.00 0.57
Intercept only !247.56 5 1.31 0.30
Proportion of time
off nest (bihourly)
Microclimate temperature (bihourly) þ brood size !591.14 7 0.00 1.00
Nestling mass Mean nestling mass
(weather and
brooding)
Nestling-stage rainfall þ proportion of time off nest
(nestling stage) þ brood size þ hatch date
!144.74 8 0.00 0.46
Proportion of time off nest (nestling stage)
þ brood size þ hatch date
!146.38 7 1.04 0.28
Mean nestling mass
(nest location and
construction)
Brood size þ hatch date !174.15 6 0.00 0.30
Intercept only !176.33 4 0.08 0.28
Stem density þ brood size þ hatch date !173.41 7 0.69 0.21
Nest wall thickness þ brood size þ hatch date !174.01 7 1.90 0.11
Notes: Maximized log likelihood (logL), number of estimated parameters (K), difference in AICc between current and best model (Di),
and Akaike weight (wi) are reported for each candidate model set. Models with DAICc ,2 are shown. Additive effects are indicated
by a plus sign. The lowest values of AICc for each model set are as follows: stem density, 981.62; nest wall thickness, 793.54; mean
off-bout duration (nestling stage), 1,474.20; mean off-bout frequency (nestling stage),!62.02; proportion of time off nest (nestling
stage),!472.21; mean off-bout duration (bihourly), 519.27; mean off-bout frequency (bihourly), 503.84; proportion of time off nest
(bihourly), 1,196.32; mean nestling mass (weather and brooding), 306.50; mean nestling mass (nest location and construction),
360.82.
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
740 Female behavioral responses to thermal variation M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster
and Greater Short-toed Lark [Calandrella brachydactyla];
Herranz et al. 2004) but not in others (e.g., Tawny Pipit
[Anthus campestris]; Sua´rez et al. 2005). Some species may
show less plasticity, or high within-individual repeatability,
in nest construction behavior (Schleicher et al. 1996,
Møller 2006, Walsh et al. 2010) or may modify nest
characteristics other than nest wall thickness to increase
nest insulation. For instance, in some cavity-nesting
species, such as Blue Tits (Cyanistes caeruleus) and Great
Tits (Parus major), females increased nest insulation
properties or nest mass in response to lower temperatures
during nest building (Britt and Deeming 2011, Deeming et
al. 2012). Rainfall during nest building has not been
associated with nest characteristics (Mainwaring et al.
2014). Our results suggest that thermal cues during nest
building may not be reliable indicators of nestling
developmental conditions in the open-cup-nesting Black-
throated Blue Warbler. Alternatively, females might adjust
characteristics other than those we measured. Further
work should evaluate whether nest characteristics are
associated with unmeasured environmental variation and/
or with individual variation in nest building that may be
correlated with elevation (i.e. differences in habitat quality
and associated individual settlement patterns).
Mean off-bout duration and proportion of time off the
nest decreased as nestlings aged and with larger brood
sizes, but off-bout frequency varied little over the nestling
stage. Increased nest attendance over the early nestling
stage is likely unrelated to thermoregulation, given that
older nestlings should require less warming by the female
(Archard et al. 2006). As nestlings grow, they can consume
larger prey items or larger amounts of food (Haggerty
1992, Goodbred and Holmes 1996). Females may therefore
be able to increase nest attentiveness, and decrease their
time spent provisioning, if they or their social mates bring
larger prey loads during each visit to the nest (Stodola et al.
2010). Because higher nest temperature is associated with
greater nestling mass and greater nestling survival
(Dawson et al. 2005), females should devote as much time
to brooding early in the nestling stage as energetic
constraints allow, in order to maximize their reproductive
success. We included mean brood size in each model of
brooding behavior to account for the known relationship
between brooding patterns and brood size. Females spent
less time attending nests with smaller broods, contrary to
results of previous studies (Sanz and Tinbergen 1999,
Chastel and Kersten 2002, Archard et al. 2006). This result
may reflect differential investment by females in small vs.
large broods. For instance, although a small brood may be
less insulated and therefore may require more brooding
(Clark 1985, Chastel and Kersten 2002), females may
spend more time brooding larger broods because they are
more valuable and are worth the additional parental
investment (Coleman and Gross 1991, Hanssen et al.
2003). Alternatively, nests may sufficiently insulate a small
brood such that heat loss does not pose a large problem. In
this case, warming a small brood may simply require less
brooding time than warming a large brood.
Both mean off-bout duration and proportion of time off
the nest increased with decreasing temperatures and with
increasing rainfall, although effect sizes were relatively
small. Female adjustments in brooding in response to
weather conditions, though slight, were opposite to
patterns reported in other studies (Beintema and Visser
1989, Rosa and Murphy 1994, Sanz and Tinbergen 1999,
Archard et al. 2006; but see Poiani 1993). For example,
female Eastern Kingbirds (Tyrannus tyrannus) spent more
time brooding at lower temperatures (Rosa and Murphy
1994), whereas female Black-throated Blue Warblers spent
less time brooding at lower temperatures. Several possi-
bilities could explain the brooding patterns we observed.
First, females might increase off-bout duration at low
temperatures and during heavy rainfall to shift the
allocation of limited time to self-maintenance activities,
such as foraging. Such a pattern has been documented in
some incubating songbirds (Haftorn 1978, 1988, Morton
and Pereyra 1985, MacDonald et al. 2013). Second, females
may require more time off the nest to forage to provision
their offspring, given that decreased prey activity at low
temperatures effectively reduces food availability (Bryant
1975, Davies 1977, Jones 1987). Lastly, the off-bout pattern
we observed could be due to increased nest attentiveness
at higher temperatures to shade nestlings (Murphy 1985,
Archard et al. 2006). For example, in 2010, a record-
breaking hot and dry year in the Northeast (Blunden et al.
2011), females on video recordings were observed shading
nestlings more than in other years (e.g., panting and
holding their wings outstretched over nestlings; S.A. Kaiser
personal observation). The effect of shading on nest
temperature is poorly understood, and further validation
of thermochrons under such conditions would be neces-
sary to determine how to identify shading events using
nest temperature data.
Females that brooded less had nestlings with lower mass
near fledging. The effect size was considerable, given that a
decrease in 0.4 g with a 10% increase in the female’s time
off the nest is likely meaningful for a nestling with a mass
of 6–9 g. However, fledging probability was not correlated
with proportion of time off the nest (Supplemental
Material Tables S3 and S4). Offspring that are left exposed
for longer periods, especially at low temperatures, may
need to reallocate energy from growth to thermoregulation
(Dawson et al. 2005). Experimental reductions of nest
temperature have not been conducted during the brooding
stage. However, nestling Tree Swallows (Tachycineta
bicolor) in experimentally heated nest boxes had higher
mass and survival to fledging (Dawson et al. 2005), which
suggests that higher nest temperatures increase nestling
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster Female behavioral responses to thermal variation 741
T
A
B
LE
2
.F
u
ll
m
o
d
el
-a
ve
ra
g
ed
p
ar
am
et
er
es
ti
m
at
es
(6
u
n
co
n
d
it
io
n
al
SE
),
u
n
co
n
d
it
io
n
al
95
%
co
n
fid
en
ce
in
te
rv
al
s
(C
I),
an
d
su
m
m
ed
A
ka
ik
e
w
ei
g
h
ts
(w
i)
fo
r
ea
ch
p
ar
am
et
er
in
cl
u
d
ed
in
lin
ea
r
m
ix
ed
m
o
d
el
s
ex
am
in
in
g
th
e
ef
fe
ct
s
o
f
m
ic
ro
cl
im
at
e
te
m
p
er
at
u
re
,r
ai
n
fa
ll,
st
em
d
en
si
ty
,n
es
t
w
al
lt
h
ic
kn
es
s,
an
d
m
ea
n
b
ro
o
d
si
ze
o
n
n
es
t
lo
ca
ti
o
n
,n
es
t
co
n
st
ru
ct
io
n
,
b
ro
o
d
in
g
p
at
te
rn
s,
an
d
n
es
tl
in
g
m
as
s
n
ea
r
fle
d
g
in
g
in
B
la
ck
-t
h
ro
at
ed
B
lu
e
W
ar
b
le
rs
at
H
u
b
b
ar
d
B
ro
o
k
Ex
p
er
im
en
ta
l
Fo
re
st
,
N
ew
H
am
p
sh
ir
e,
U
SA
,
M
ay
–
A
u
g
u
st
,2
01
0–
20
13
.S
u
m
m
ed
w
i
va
lu
es
ar
e
n
o
t
p
ro
vi
d
ed
fo
r
b
ro
o
d
si
ze
o
r
h
at
ch
d
at
e
b
ec
au
se
th
es
e
fa
ct
o
rs
w
er
e
in
cl
u
d
ed
in
ea
ch
m
o
d
el
in
a
g
iv
en
ca
n
d
id
at
e
m
o
d
el
se
t
to
ac
co
u
n
t
fo
r
kn
o
w
n
va
ri
at
io
n
.
P
ar
am
et
er
s
in
b
o
ld
h
av
e
co
n
fid
en
ce
in
te
rv
al
s
th
at
d
o
n
o
t
in
cl
u
d
e
ze
ro
.
M
o
d
el
ty
p
e
R
es
p
o
n
se
va
ri
ab
le
P
ar
am
et
er
Es
ti
m
at
e
C
I
w
i
Ef
fe
ct
a
N
es
t
ch
ar
ac
te
ri
st
ic
s
St
em
d
en
si
ty
(s
q
u
ar
e-
ro
o
t
tr
an
sf
o
rm
ed
)
N
es
t-
b
u
ild
in
g
te
m
p
er
at
u
re
!0
.0
03
6
0.
01
1
!0
.0
45
,
0.
02
7
0.
29
þ9
3
10
6
st
em
s/
1
m
2
/1
8C
N
es
t-
b
u
ild
in
g
ra
in
fa
ll
0.
00
02
6
0.
00
2
!0
.0
05
,
0.
00
7
0.
27
þ4
3
10
8
st
em
s/
1
m
2
/1
m
m
ra
in
N
es
t
w
al
l
th
ic
kn
es
s
N
es
t-
b
u
ild
in
g
te
m
p
er
at
u
re
0.
01
1
6
0.
01
4
!0
.0
06
,
0.
04
8
0.
53
þ0
.0
11
cm
n
es
t
w
al
l/
18
C
N
es
t-
b
u
ild
in
g
ra
in
fa
ll
!0
.0
01
6
0.
00
2
!0
.0
08
,
0.
00
2
0.
43
!0
.0
01
cm
n
es
t
w
al
l/
1
m
m
ra
in
S
te
m
d
e
n
si
ty
0.
02
3
6
0.
00
6
0.
01
2,
0.
03
5
1.
00
þ0
.0
23
cm
n
es
t
w
al
l/
st
em
/
1
m
2
B
ro
o
d
in
g
p
at
te
rn
s
M
ea
n
o
ff
-b
o
u
t
d
u
ra
ti
o
n
(n
es
tl
in
g
st
ag
e)
M
ic
ro
cl
im
a
te
te
m
p
e
ra
tu
re
(n
e
st
li
n
g
st
a
g
e
)
!1
.3
88
6
1.
10
5
!3
.5
88
,
!0
.1
03
0.
75
!1
.3
88
s
d
u
ra
ti
o
n
o
ff
/1
8C
N
e
st
li
n
g
-s
ta
g
e
ra
in
fa
ll
0.
51
4
6
0.
31
9
0.
10
4,
1.
10
8
0.
85
þ0
.5
14
s
d
u
ra
ti
o
n
o
ff
/1
m
m
ra
in
N
es
t
w
al
l
th
ic
kn
es
s
!0
.6
92
6
2.
38
9
!1
0.
26
7,
5.
43
8
0.
29
!0
.6
92
s
d
u
ra
ti
o
n
o
ff
/1
cm
n
es
t
w
al
l
St
em
d
en
si
ty
0.
06
4
6
0.
26
4
!0
.6
80
,
1.
14
6
0.
27
0.
06
4
s
d
u
ra
ti
o
n
o
ff
/s
te
m
/
1
m
2
B
ro
o
d
si
ze
!2
0.
57
6
6
4.
34
5
!2
9.
16
6,
!1
1.
98
6
–
!2
0.
57
6
s
d
u
ra
ti
o
n
/n
es
tl
in
g
M
ea
n
o
ff
-b
o
u
t
fr
eq
u
en
cy
(n
es
tl
in
g
st
ag
e)
M
ic
ro
cl
im
at
e
te
m
p
er
at
u
re
(n
es
tl
in
g
st
ag
e)
!0
.0
01
6
0.
00
2
!0
.0
11
,
0.
00
5
0.
21
!0
.0
01
o
ff
-b
o
u
ts
/1
h
r/
18
C
N
es
tl
in
g
-s
ta
g
e
ra
in
fa
ll
0.
00
03
6
0.
00
1
!0
.0
01
,
0.
00
3
0.
23
þ0
.0
00
3
o
ff
-b
o
u
ts
/1
h
r/
1
m
m
ra
in
N
es
t
w
al
l
th
ic
kn
es
s
0.
00
6
6
0.
01
4
!0
.0
14
,
0.
06
1
0.
27
þ0
.0
06
o
ff
-b
o
u
ts
/1
h
r/
1
cm
n
es
t
w
al
l
St
em
d
en
si
ty
0.
00
02
6
0.
00
1
!0
.0
03
,
0.
00
6
0.
18
þ0
.0
00
2
o
ff
-b
o
u
ts
/1
h
r/
st
em
/1
m
2
B
ro
o
d
si
ze
!0
.0
02
6
0.
01
7
!0
.0
45
,
0.
03
8
–
!0
.0
02
o
ff
-b
o
u
ts
/1
h
r/
n
es
tl
in
g
P
ro
p
o
rt
io
n
o
f
ti
m
e
o
ff
n
es
t
(n
es
tl
in
g
st
ag
e)
M
ic
ro
cl
im
a
te
te
m
p
e
ra
tu
re
(n
e
st
li
n
g
st
a
g
e
)
!0
.0
02
6
0.
00
1
!0
.0
04
,
!0
.0
00
2
0.
79
!0
.2
%
ch
an
g
e
in
ti
m
e
o
ff
/
18
C
N
e
st
li
n
g
-s
ta
g
e
ra
in
fa
ll
0.
00
1
6
0.
00
03
0.
00
01
,
0.
00
1
0.
87
þ0
.1
%
ch
an
g
e
in
ti
m
e
o
ff
/
1
m
m
ra
in
N
es
t
w
al
l
th
ic
kn
es
s
0.
00
1
6
0.
00
3
!0
.0
06
,
0.
01
3
0.
31
þ0
.1
%
ch
an
g
e
in
ti
m
e
o
ff
/
1
cm
n
es
t
w
al
l
St
em
d
en
si
ty
0.
00
05
6
0.
00
1
!0
.0
00
2,
0.
00
2
0.
55
þ0
.0
5%
ch
an
g
e
in
ti
m
e
o
ff
/
st
em
/1
m
2
B
ro
o
d
si
ze
!0
.0
18
6
0.
00
5
!0
.0
28
,
!0
.0
08
–
!1
.8
%
ch
an
g
e
in
ti
m
e
o
ff
/
n
es
tl
in
g
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
742 Female behavioral responses to thermal variation M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster
T
A
B
LE
2
.
C
o
n
ti
n
u
ed
.
M
o
d
el
ty
p
e
R
es
p
o
n
se
va
ri
ab
le
P
ar
am
et
er
Es
ti
m
at
e
C
I
w
i
Ef
fe
ct
a
M
ea
n
o
ff
-b
o
u
t
d
u
ra
ti
o
n
(b
ih
o
u
rl
y;
n
at
u
ra
l-
lo
g
tr
an
sf
o
rm
ed
)
M
ic
ro
cl
im
a
te
te
m
p
e
ra
tu
re
(b
ih
o
u
rl
y)
!0
.0
20
6
0.
00
2
!0
.0
22
,
!0
.0
17
1.
00
!1
.9
8%
ch
an
g
e
in
d
u
ra
ti
o
n
o
ff
(s
)/
18
C
B
ro
o
d
si
ze
!0
.0
56
6
0.
01
2
!0
.0
80
,
!0
.0
32
–
!5
.4
4%
ch
an
g
e
in
d
u
ra
ti
o
n
o
ff
(s
)/
n
es
tl
in
g
M
ea
n
o
ff
-b
o
u
t
fr
eq
u
en
cy
(b
ih
o
u
rl
y)
M
ic
ro
cl
im
a
te
te
m
p
e
ra
tu
re
(b
ih
o
u
rl
y)
!0
.0
02
6
0.
00
2
!0
.0
06
,
!0
.0
00
4
0.
57
!0
.0
02
o
ff
-b
o
u
ts
/1
h
r/
18
C
B
ro
o
d
si
ze
0.
00
3
6
0.
00
8
!0
.0
13
,
0.
02
1
–
þ0
.0
03
o
ff
-b
o
u
ts
/1
h
r/
n
es
tl
in
g
P
ro
p
o
rt
io
n
o
f
ti
m
e
o
ff
n
es
t
(b
ih
o
u
rl
y)
M
ic
ro
cl
im
a
te
te
m
p
e
ra
tu
re
(b
ih
o
u
rl
y)
!0
.0
23
6
0.
00
2
!0
.0
26
,
!0
.0
19
1.
00
!2
.3
%
ch
an
g
e
in
ti
m
e
o
ff
/1
8C
B
ro
o
d
si
ze
!0
.0
51
6
0.
01
4
!0
.0
79
,
!0
.0
23
–
!5
.1
%
ch
an
g
e
in
ti
m
e
o
ff
/n
es
tl
in
g
N
es
tl
in
g
m
as
s
M
ea
n
n
es
tl
in
g
m
as
s
(w
ea
th
er
an
d
b
ro
o
d
in
g
)
M
ic
ro
cl
im
at
e
te
m
p
er
at
u
re
(n
es
tl
in
g
st
ag
e)
0.
00
04
6
0.
00
8
!0
.0
30
,
0.
03
3
0.
25
þ0
.0
00
4
g
/1
8C
N
es
tl
in
g
-s
ta
g
e
ra
in
fa
ll
!0
.0
04
6
0.
00
4
! 0
.0
14
,
0.
00
1
0.
62
!0
.0
04
g
/1
m
m
ra
in
P
ro
p
o
rt
io
n
o
f
ti
m
e
o
ff
n
e
st
(n
e
st
li
n
g
st
a
g
e
)
!4
.2
34
6
1.
18
9
!6
.5
33
,
!1
.9
80
0.
99
!4
.2
34
g
/1
00
%
ch
an
g
e
in
ti
m
e
o
ff
B
ro
o
d
si
ze
!0
.2
07
6
0.
07
8
!0
.3
60
,
!0
.0
54
–
!0
.2
07
g
/n
es
tl
in
g
H
at
ch
d
at
e
!0
.0
03
6
0.
00
4
!0
.0
10
,
0.
00
4
–
!0
.0
03
g
/d
ay
M
ea
n
n
es
tl
in
g
m
as
s
(n
es
t
lo
ca
ti
o
n
an
d
co
n
st
ru
ct
io
n
)
St
em
d
en
si
ty
0.
00
3
6
0.
00
6
!0
.0
04
,
0.
02
2
0.
31
þ0
.0
03
g
/s
te
m
/1
m
2
N
es
t
w
al
l
th
ic
kn
es
s
!0
.0
08
6
0.
03
0
!0
.1
49
,
0.
07
5
0.
21
!0
.0
08
g
/1
cm
n
es
t
w
al
l
B
ro
o
d
si
ze
!0
.1
02
6
0.
08
8
!0
.2
82
,
!0
.0
01
–
!0
.1
02
g
/n
es
tl
in
g
H
at
ch
d
at
e
0.
00
1
6
0.
00
3
!0
.0
05
,
0.
00
8
–
þ0
.0
01
g
/d
ay
a
Ef
fe
ct
o
f
1-
u
n
it
in
cr
ea
se
in
p
ar
am
et
er
in
te
rm
s
o
f
re
sp
o
n
se
va
ri
ab
le
.
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster Female behavioral responses to thermal variation 743
growth rate. When nests were cooled experimentally
during incubation, females spent less time incubating,
resulting in lower nestling mass and immunity (Ardia et al.
2010). Our results suggest that females may modify their
parental investment during cooler and wetter periods, but
possibly at a cost to their offspring’s growth and survival.
The ability of females to adjust nest characteristics may
be an important determinant of reproductive success. We
found no association between either stem density around
nests or nest wall thickness and (1) nestling mass, an index
of reproductive success (Monro´s et al. 2002, Gren˜o et al.
2008); or (2) nesting success (Supplemental Material
Tables S3 and S4). We know of no similar studies of
open-cup-nesting passerines. Studies of cavity-nesting
species have found a positive relationship between fledging
success and nest mass (Britt and Deeming 2011), nest size
(A´lvarez and Barba 2008, 2011), and nest insulation
(Lombardo et al. 1995), but other studies have not found
these relationships (Alabrudzin´ska et al. 2003, Toma´s et al.
2006). Black-throated BlueWarbler nests in areas of higher
stem density and with thicker nest walls were typically at
higher elevations at the HBEF. Thus, cooler ambient
temperatures experienced by young in those nests could
have counteracted any effects of nest characteristics on
nestling mass. Alternatively, brooding may be more
effective than nest structure at buffering nestlings from
thermal fluctuations in this and similar species.
In conclusion, our findings demonstrate that nest
characteristics and brooding behavior of female Black-
throated Blue Warblers may be related to nest-site
temperatures and rainfall. At higher elevations, where
conditions are cooler, females construct nests in areas of
higher stem density and with thicker nest walls. However,
females do not adjust nest characteristics in response to
weather conditions experienced during nest building.
Females do, however, modify brooding patterns in
response to nest-stage-specific thermal variation in ways
that could influence fledgling survival. During cooler and
wetter periods, brooding females spent more time off the
nest, which suggests that they may have been investing in
self-maintenance activities at the expense of offspring
growth and development. Plasticity in brooding behavior
provides a mechanism by which open-cup-nesting song-
birds can cope with thermal stress, but this might be
constrained by female energetic thresholds. Further studies
designed to quantify the effects of local temperature on
parental investment and offspring growth and develop-
ment will be useful for understanding how wild bird
populations will respond to environmental change (Du-
Rant et al. 2013). Longitudinal studies examining within-
individual plasticity in female response to various weather
conditions, including extreme weather events, across years
and locations would help in predicting the effects of rapid
environmental change on reproductive success.
ACKNOWLEDGMENTS
We are grateful to the undergraduates and field technicians
who contributed their efforts to this study, including A.
Locatelli for helping pilot field methods for deploying
thermochrons. We thank C. Cooper and H. Mills for technical
support and W. Hochachka for statistical guidance. We
especially thank R. T. Holmes and N. L. Rodenhouse for
logistical support. D. J. Weber and two anonymous reviewers
provided helpful comments that improved the manuscript.
This research was supported by U.S. National Science
Foundation grants, and M.G.S. was supported on a Research
Experience for Undergraduates supplement. This research is a
contribution of the Hubbard Brook Ecosystem Study, part of
the Long-Term Ecological Research network supported by the
National Science Foundation.
Funding statement: This research was supported by National
Science Foundation grants awarded to Cornell University
(0640470), the Smithsonian Institution (0640732), and Well-
esley College (064082300). None of our funders had any
influence on the content of the submitted or published
manuscript. None of our funders requires approval of the final
manuscript to be published.
Ethics statement: All research activities were performed
under protocols approved by the animal care and use
committees of the authors’ institutions. This research was
conducted in compliance with the Guidelines to the Use of
Wild Birds in Research (Fair et al. 2010).
Author contributions: S.A.K. designed the study and
supervised the research and data collection. M.G.S. and
S.A.K. drafted the manuscript. M.G.S. analyzed the data. T.S.S.
and M.S.W. secured funding and revised the manuscript. All
authors gave final approval for publication and agreed to be
accountable for all aspects of the work.
Data deposits: Data will be made available via the Dryad
Digital Repository.
LITERATURE CITED
Alabrudzin´ska, J., A. Kalin´ski, R. Słomczyn´ski, J. Wawrzyniak, P.
Zielin´ski, and J. Ban´bura (2003). Effects of nest characteristics
on breeding success of Great Tits Parus major. Acta
Ornithologica 38:151–154.
A´lvarez, E., and E. Barba (2008). Nest quality in relation to adult
bird condition and its impact on reproduction in Great Tits
Parus major. Acta Ornithologica 43:3–9.
A´lvarez, E., and E. Barba (2011). Nest characteristics and
reproductive performance in Great Tits Parus major. Ardeola
58:125–136.
Archard, G. A., R. J. Robertson, D. Jones, J. Painter, R. Crozier, and
M. F. Clarke (2006). Brooding behaviour in the cooperatively
breeding Bell Miner (Manorina melanophrys). Emu 106:105–
112.
Ardia, D. R. (2013). The effects of nestbox thermal environment
on fledging success and haematocrit in Tree Swallows. Avian
Biology Research 6:99–103.
Ardia, D. R., J. H. Pe´rez, and E. D. Clotfelter (2010). Experimental
cooling during incubation leads to reduced innate immunity
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
744 Female behavioral responses to thermal variation M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster
and body condition in nestling Tree Swallows. Proceedings of
the Royal Society B 277:20092138.
Barton, K. (2014). MuMIn: Multi-Model Inference. R package
1.12.1. https://cran.r-project.org/web/packages/MuMIn/
index.html
Bates, D., M. Ma¨chler, B. M. Bolker, and S. C. Walker (2015). Fitting
linear mixed-effects models using lme4. Journal of Statistical
Software 67:1–48.
Beintema, A. J., and G. H. Visser (1989). The effect of weather on
time budgets and development of chicks of meadow birds.
Ardea 77:181–192.
Bioacoustics Research Program (2014). Raven Pro: Interactive
Sound Analysis Software (version 1.4). Cornell Lab of
Ornithology, Ithaca, New York, USA.
Blunden, J., D. S. Arndt, and M. O. Baringer (Editors) (2011). State
of the climate in 2010. Bulletin of the American Meteorolog-
ical Society 92:S1–S236.
Botero-Delgadillo, E., N. Orellana, D. Serrano, Y. Poblete, and R. A.
Va´squez (2017). Interpopulation variation in nest architecture
in a secondary cavity-nesting bird suggests site-specific
strategies to cope with heat loss and humidity. The Auk
134:281–294.
Britt, J., and D. C. Deeming (2011). First-egg date and air
temperature affect nest construction in Blue Tits Cyanistes
caeruleus, but not in Great Tits Parus major. Bird Study 58:78–
89.
Bryant, D. M. (1975). Breeding biology of House Martins Delichon
urbica in relation to aerial insect abundance. Ibis 117:180–
216.
Burnham, K. P., and D. R. Anderson (2002). Model Selection and
Multimodel Inference: A Practical Information-Theoretic
Approach, second edition. Springer, New York, NY, USA.
Chaplin, S. B., M. L. Cervenka, and A. C. Mickelson (2002).
Thermal environment of the nest during development of
Tree Swallow (Tachycineta bicolor) chicks. The Auk 119:845–
851.
Chastel, O., and M. Kersten (2002). Brood size and body
condition in the House Sparrow Passer domesticus: The
influence of brooding behaviour. Ibis 144:284–292.
Clark, L. (1985). Consequences of homeothermic capacity of
nestlings on parental care in the European Starling.
Oecologia 65:387–393.
Coleman, R. M., and M. R. Gross (1991). Parental investment
theory: The role of past investment. Trends in Ecology &
Evolution 6:404–406.
Cooper, C. B., and H. Mills (2005). New software for quantifying
incubation behavior from time-series recordings. Journal of
Field Ornithology 76:352–356.
Crawley, M. J. (2013). The R Book. Wiley, Chichester, UK.
Crossman, C. A., V. G. Rohwer, and P. R. Martin (2011). Variation
in the structure of bird nests between northern Manitoba
and southeastern Ontario. PLOS One 6:e19086.
D’Alba, L., P. Monaghan, and R. G. Nager (2009). Thermal benefits
of nest shelter for incubating female eiders. Journal of
Thermal Biology 34:93–99.
Davidson, A. J., F. Aujard, B. London, M. Menaker, and G. D. Block
(2003). Thermochron iButtons: An inexpensive method for
long-term recording of core body temperature in untethered
animals. Journal of Biological Rhythms 18:430–432.
Davies, N. B. (1977). Prey selection and the search strategy of the
Spotted Flycatcher (Muscicapa striata): A field study on
optimal foraging. Animal Behaviour 25:1016–1022.
Dawson, R. D., C. C. Lawrie, and E. L. O’Brien (2005). The importance
of microclimate variation in determining size, growth and
survival of avian offspring: Experimental evidence from a cavity
nesting passerine. Oecologia 144:499–507.
Dawson, W. R., and F. C. Evans (1957). Relation of growth and
development to temperature regulation in nestling Field and
Chipping sparrows. Physiological Zoology 30:315–327.
Deeming, D. C., M. C. Mainwaring, I. R. Hartley, and S. J. Reynolds
(2012). Local temperature and not latitude determines the
design of Blue Tit and Great Tit nests. Avian Biology Research
5:203–208.
DuRant, S. E., W. A. Hopkins, G. R. Hepp, and J. R. Walters (2013).
Ecological, evolutionary, and conservation implications of
incubation temperature-dependent phenotypes in birds.
Biological Reviews 88:499–509.
Fair, J. M., E. Paul, and J. Jones (Editors) (2010). Guidelines to the
Use of Wild Birds in Research, third edition. Ornithological
Council, Washington, DC, USA.
Goodbred, C. O’N., and R. T. Holmes (1996). Factors affecting
food provisioning of nestling Black-throated Blue Warblers.
The Wilson Bulletin 108:467–479.
Gren˜o, J. L., E. J. Belda, and E. Barba (2008). Influence of
temperatures during the nestling period on post-fledging
survival of Great Tit Parus major in a Mediterranean habitat.
Journal of Avian Biology 39:41–49.
Haftorn, S. (1978). Egg-laying and regulation of egg temperature
during incubation in the Goldcrest Regulus regulus. Ornis
Scandinavica 9:2–21.
Haftorn, S. (1988). Incubating female passerines do not let the
egg temperature fall below the ‘physiological zero temper-
ature’ during their absences from the nest. Ornis Scandinav-
ica 19:97–110.
Haggerty, T. M. (1992). Effects of nestling age and brood size on
nestling care in the Bachman’s Sparrow (Aimophila aestivalis).
The American Midland Naturalist 128:115–125.
Hair, J. F., Jr., W. C. Black, B. J. Babin, and R. E. Anderson (2010).
Multivariate Data Analysis, seventh edition. Prentice Hall,
Upper Saddle River, NJ, USA.
Hanssen, S. A., K. E. Erikstad, V. Johnsen, and J. O. Bustnes (2003).
Differential investment and costs during avian incubation
determined by individual quality: An experimental study of
the Common Eider (Somateria mollissima). Proceedings of the
Royal Society B 270:531–537.
Herranz, J., J. Traba, M. B. Morales, and F. Sua´rez (2004). Nest size
and structure variation in two ground nesting passerines, the
Skylark Alauda arvensis and the Short-toed Lark Calandrella
brachydactyla. Ardea 92:209–218.
Holmes, R. T., S. A. Kaiser, N. L. Rodenhouse, T. S. Sillett, M. S.
Webster, P. Pyle, and M. A. Patten (2017). Black-throated Blue
Warbler (Setophaga caerulescens), version 3.0. In Birds of
North America Online (P. G. Rodewald, Editor). Cornell Lab of
Ornithology, Ithaca, NY, USA.
Holmes, R. T., T. W. Sherry, P. P. Marra, and K. E. Petit (1992).
Multiple brooding and productivity of a Neotropical migrant,
the Black-throated Blue Warbler (Dendroica caerulescens), in
an unfragmented temperate forest. The Auk 109:321–333.
Holmes, R. T., T. W. Sherry, and F. W. Sturges (1986). Bird
community dynamics in a temperate deciduous forest: Long-
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster Female behavioral responses to thermal variation 745
term trends at Hubbard Brook. Ecological Monographs 56:
201–220.
Holway, D. A. (1991). Nest-site selection and the importance of
nest concealment in the Black-throated Blue Warbler. The
Condor 93:575–581.
Hubbart, J., T. Link, C. Campbell, and D. Cobos (2005). Evaluation
of a low-cost temperature measurement system for environ-
mental applications. Hydrological Processes 19:1517–1523.
Jones, G. (1987). Time and energy constraints during incubation
in free-living swallows (Hirundo rustica): An experimental
study using precision electronic balances. Journal of Animal
Ecology 56:229–245.
Joyce, E. M., T. S. Sillett, and R. T. Holmes (2001). An inexpensive
method for quantifying incubation patterns of open-cup
nesting birds, with data for Black-throated Blue Warblers.
Journal of Field Ornithology 72:369–379.
Kaiser, S. A., T. S. Sillett, and M. S. Webster (2014). Phenotypic
plasticity in hormonal and behavioural responses to changes
in resource conditions in a migratory songbird. Animal
Behaviour 96:19–29.
Kern, M. D., and C. van Riper III (1984). Altitudinal variations in
nests of the Hawaiian honeycreeper Hemignathus virens
virens. The Condor 86:443–454.
Kovarˇı´k, P., V. Pavel, and B. Chutny´ (2009). Incubation behaviour
of the Meadow Pipit (Anthus pratensis) in an alpine
ecosystem of Central Europe. Journal of Ornithology 150:
549–556.
Lakens, D. (2013). Calculating and reporting effect sizes to
facilitate cumulative science: A practical primer for t-tests and
ANOVAs. Frontiers in Psychology 4:863.
Lombardo, M. P., R. M. Bosman, C. A. Faro, S. G. Houtteman, and
T. S. Kluisza (1995). Effect of feathers as nest insulation on
incubation behavior and reproductive performance of Tree
Swallows (Tachycineta bicolor). The Auk 112:973–981.
MacDonald, E. C., A. F. Camfield, J. E. Jankowski, and K. Martin
(2013). Extended incubation recesses by alpine-breeding
Horned Larks: A strategy for dealing with inclement weather?
Journal of Field Ornithology 84:58–68.
Mainwaring, M. C., D. C. Deeming, C. I. Jones, and I. R. Hartley
(2014). Adaptive latitudinal variation in Common Blackbird
Turdus merula nest characteristics. Ecology and Evolution 4:
841–851.
McDonald, J. H. (2014). Handbook of Biological Statistics, third
edition. Sparky House, Baltimore, MD, USA.
Møller, A. P. (2006). Rapid change in nest size of a bird related to
change in a secondary sexual character. Behavioral Ecology
17:108–116.
Monro´s, J. S., E. J. Belda, and E. Barba (2002). Post-fledging
survival of individual Great Tits: The effect of hatching date
and fledging mass. Oikos 99:481–488.
Morton, M. L., and M. E. Pereyra (1985). The regulation of egg
temperatures and attentiveness patterns in the Dusky
Flycatcher (Empidonax oberholseri). The Auk 102:25–37.
Murphy, M. T. (1985). Nestling Eastern Kingbird growth: Effects
of initial size and ambient temperature. Ecology 66:162–170.
Navarro, D. J. (2015). Learning statistics with R: A tutorial for
psychology students and other beginners (version 0.5).
University of Adelaide, Adelaide, Australia.
Olson, C. R., C. M. Vleck, and D. Vleck (2006). Periodic cooling of
bird eggs reduces embryonic growth efficiency. Physiological
and Biochemical Zoology 79:927–936.
Pinheiro, J. C., and D. M. Bates (2000). Mixed-Effects Models in S
and S-PLUS. Springer, New York, NY, USA.
Poiani, A. (1993). Effects of clutch size manipulations on
reproductive behaviour and nesting success in the cooper-
atively breeding Bell Miner (Manorina melanophrys). Evolu-
tionary Ecology 7:329–356.
R Core Team (2016). R: A Language and Environment for
Statistical Computing. R Foundation for Statistical Comput-
ing, Vienna, Austria.
Rodenhouse, N. L., S. N. Matthews, K. P. McFarland, J. D. Lambert,
L. R. Iverson, A. Prasad, T. S. Sillett, and R. T. Holmes (2008).
Potential effects of climate change on birds of the Northeast.
Mitigation and Adaptation Strategies for Global Change 13:
517–540.
Rodenhouse, N. L., T. S. Sillett, P. J. Doran, and R. T. Holmes
(2003). Multiple density-dependence mechanisms regulate a
migratory bird population during the breeding season.
Proceedings of the Royal Society B 270:2105–2110.
Rohwer, V. G., and J. S. Y. Law (2010). Geographic variation in
nests of Yellow Warblers breeding in Churchill, Manitoba, and
Elgin, Ontario. The Condor 112:596–604.
Rosa, S. M., and M. T. Murphy (1994). Trade-offs and constraints
on Eastern Kingbird parental care. The Wilson Bulletin 106:
668–678.
Sanz, J. J., and J. M. Tinbergen (1999). Energy expenditure,
nestling age, and brood size: An experimental study of
parental behavior in the Great Tit Parus major. Behavioral
Ecology 10:598–606.
Schleicher, B., H. Hoi, and F. Valera (1996). Seasonal change in
female mate choice criteria in Penduline Tits (Remiz
pendulinus). Ardeola 43:19–29.
Schwarz, P. A., T. J. Fahey, and C. E. McCulloch (2003). Factors
controlling spatial variation of tree species abundance in a
forested landscape. Ecology 84:1862–1878.
Smith, A. D. H., D. R. Crabtree, J. L. J. Bilzon, and N. P. Walsh
(2010). The validity of wireless iButtons and thermistors for
human skin temperature measurement. Physiological Mea-
surement 31:95–114.
Stodola, K. W., E. T. Linder, D. A. Buehler, K. E. Franzreb, D. H. Kim,
and R. J. Cooper (2010). Relative influence of male and female
care in determining nestling mass in a migratory songbird.
Journal of Avian Biology 41:515–522.
Sua´rez, F., M. B. Morales, I. Mı´nguez, and J. Herranz (2005).
Seasonal variation in nest mass and dimensions in an open-
cup ground-nesting shrub-steppe passerine: The Tawny Pipit
Anthus campestris. Ardeola 52:43–51.
Toma´s, G., S. Merino, J. Moreno, J. J. Sanz, J. Morales, and S.
Garcı´a-Fraile (2006). Nest weight and female health in the
Blue Tit (Cyanistes caeruleus). The Auk 123:1013–1021.
Walsberg, G. E. (1985). Physiological consequences of microhab-
itat selection. In Habitat Selection in Birds (M. L. Cody, Editor).
Academic Press, Orlando, FL, USA. pp. 389–413.
Walsh, P. T., M. Hansell, W. D. Borello, and S. D. Healy (2010).
Repeatability of nest morphology in African weaver birds.
Biology Letters 6:149–151.
Whittow, G. C., and A. J. Berger (1977). Heat loss from the nest of
the Hawaiian honeycreeper, ‘‘Amakihi.’’ The Wilson Bulletin
89:480–483.
Whittow, G. C., and H. Tazawa (1991). The early development of
thermoregulation in birds. Physiological Zoology 64:1371–
1390.
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
746 Female behavioral responses to thermal variation M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster
Windsor, R. L., J. L. Fegely, and D. R. Ardia (2013). The effects of
nest size and insulation on thermal properties of Tree
Swallow nests. Journal of Avian Biology 44:305–310.
Zuur, A., E. N. Ieno, N. Walker, A. A. Saveliev, and G. M. Smith
(2009). Mixed Effects Models and Extensions in Ecology with
R. Springer, New York, NY, USA.
APPENDIX
Validation of Thermochrons for Detecting Female Off-
Bouts
We verified the accuracy of off-bout identification using
thermochron data by comparing off-bouts detected using
thermochrons to off-bouts documented in 2 hr video
recordings of nestling provisioning on nestling days 3 and
7 from a complementary study conducted during 2010–
2012 (Kaiser et al. 2014). We randomly selected 10 video
recordings across all plots and years taken at different
times during the breeding season for each nestling age to
compare with thermochron data. In 2013, we collected 5
video recordings on nestling day 5 to test the accuracy of
the thermochrons midway through the nestling stage. We
recorded start and end times of all off-bouts, defined here
as any time the female was not sitting inside the nest cup
to brood. We included time spent perched on the nest rim
as an off-bout. When analyzing thermochron data, we
defined female off-bouts as intervals with a monotonic
decrease in the difference between nest and ambient
temperatures of #0.58C, indicating that the nest was
cooling in relation to the ambient temperature.
For days 3, 5, and 7 of the nestling stage, we calculated
the percentage of off-bouts that were detected by
thermochrons and whose durations as recorded by
thermochrons were within 4 min of the actual durations
recorded by video. When videos showed 2 off-bouts
separated by an on-bout of ,2 min, we counted this as a
single off-bout because thermochron resolution was not
fine enough to record 2 off-bouts in these situations.
Thermochrons accurately detected 74.24% (n ¼ 66 off-
bouts from 10 nests), 70.37% (n ¼ 27 off-bouts from 5
nests), and 30.61% (n¼ 49 off-bouts from 10 nests) of off-
bouts on days 3, 5, and 7, respectively. Of all off-bouts
recorded by thermochrons, the percentages that did not
correspond to actual off-bouts (false positives) were 7.55%
(n¼ 53 off-bouts recorded by thermochrons), 20.83% (n¼
24 off-bouts recorded by thermochrons), and 55.88% (n¼
34 off-bouts recorded by thermochrons) on days 3, 5, and
7, respectively. When a thermochron recorded an actual
off-bout as multiple off-bouts, we evaluated whether the
first off-bout recorded was within 4 min of the actual
duration and did not count subsequent off-bouts as false
positives because they occurred during a known off-bout.
We examined whether ambient temperature during off-
bouts was associated with whether thermochrons detected
actual off-bouts, missed off-bouts, or recorded false
positives but found no differences (ANOVA; F ¼ 0.2, df
¼ 2 and 167, P ¼ 0.80).
On the basis of these results, we excluded thermochron
data after day 5 from all analyses. Although thermochron
accuracy was poorer for day 5 than for day 3, accuracy was
substantially lower on day 7 of the nestling stage than on
day 3 or 5. Nestling homeothermy begins on day 5 or 6 in
the Field Sparrow (Spizella pusilla) and Chipping Sparrow
(S. passerina), which develop over 8–10 days (Dawson and
Evans 1957). The Black-throated Blue Warbler has a
similar nestling development period (Holmes et al. 1992),
so this seems to be an appropriate cut-off to avoid the
effects of heat produced by thermoregulating nestlings.
Given the fairly high accuracy of thermochrons prior to
nestling day 5, we retained our initial off-bout criterion of
any monotonic decrease in the difference between nest
and ambient temperatures. We imposed no minimum
duration or rate of cooling when determining off-bouts.
Although off-bout data from thermochrons were not
completely accurate, using thermochrons in place of video
recording allowed us to have a much larger sample size.
We excluded nocturnal data from all analyses because
we often observed a decrease in the difference between
nest and ambient temperatures throughout the night,
likely due to consistent cooling of the nest contents as
ambient temperature decreased. We could not verify
female activity after sundown, so we could not identify
female off-bouts during 2030–0500 hours with any
certainty.
The Auk: Ornithological Advances 135:733–747, Q 2018 American Ornithological Society
M. G. Smith, S. A. Kaiser, T. S. Sillett, and M. S. Webster Female behavioral responses to thermal variation 747