Regular issues
Special Issues

Northeastern Naturalist
    NENA Home
    Range and Scope
    Board of Editors
    Editorial Workflow
    Publication Charges

Co-published Journals
    Southeastern Naturalist
    Caribbean Naturalist
    Urban Naturalist

EH Natural History Home

Habitat Use and Foraging Behavior of Eastern Bluebirds (Sialia sialis) in Relation to Winter Weather
Todd J. Weinkam, Gregg A. Janos, and David R. Brown

Northeastern Naturalist,Volume 24, Special Issue 7 (2017): B1–B18

Full-text pdf (Accessible only to subscribers.To subscribe click here.)


Site by Bennett Web & Design Co.
Northeastern Naturalist 1 T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 Vol. 24, Special Issue 7 Habitat Use and Foraging Behavior of Eastern Bluebirds (Sialia sialis) in Relation to Winter Weather Todd J. Weinkam1, Gregg A. Janos1, and David R. Brown1,* Abstract - Population declines of songbirds following severe winters draw attention to a need to better understand behavioral responses to inclement weather. We used observations of radio-tracked Sialia sialis (Eastern Bluebird) wintering in Kentucky to examine the effects of weather on habitat use, group size, foraging behavior, and diet. Home ranges were smaller than published estimates, and consisted of more open than wooded habitat, in proportion to availability. Although habitat use appeared unchanged during inclement weather, Bluebirds increased group sizes, and shifted from insectivory to frugivory during periods of sub-freezing temperatures and snow cover. Fecal analysis confirmed the weather-driven shift of diet. Inclement winter weather likely lowers the efficiency of insectivory, leading to changes in social and foraging behaviors. Introduction Environmental factors and winter-specific resource constraints can affect the behavior (Duriez et al. 2005), abundance (Meehan et al. 2004), and range limits (Zuckerberg et al. 2011) of birds. Severe winter weather can reduce winter survival, population size, and impact reproductive success in subsequent seasons (Porter et al. 1983, Sauer and Droege 1990). Our ability to predict and interpret these processes may be enhanced by a better understanding of winter-specific habitat and resource requirements. Specifically, information about the interactions between weather, habitat, and behavior may help explain how winter events influence population patterns. Sialia sialis L. (Eastern Bluebird, hereafter Bluebird) have been widely studied during the breeding season (Gowaty and Plissner 2015), but much less is known about their winter ecology. Reports of Bluebird mortality during harsh winter weather (Pitts 1978, Wilson and Stamm 1960) followed by observable population declines (Monroe 1978; Palmer-Ball 2015; Sauer and Droege 1990; Stamm 1979a, b; Wilson 1962) demonstrate the potential for winter weather to affect their populations. Recently, Wetzel and Krupa (2013) reported a positive correlation between mean winter temperature and Bluebird abundance during subsequent breeding seasons in central Kentucky, and suggested that the Bluebird population there may be particularly susceptible to cold winters, in part because most birds are residents. Although several investigators have examined Bluebird breeding habitat use (e.g., Plissner and Gowaty 1995, Sloan and Carlson 1980, Stanback and Rockwell 2003), less is known about use of habitat during the winter. Allen and Sweeney 1Department of Biological Sciences, Eastern Kentucky University, Richmond, KY 40475.*Corresponding author - Manuscript Editor: Noah Perlut Winter Ecology: Insights from Biology and Histor2y4(Special Issue X):XX–XX 2017 Northeastern Naturalist 24(Special Issue 7):B1–B18 Northeastern Naturalist T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 2 Vol. 24, Special Issue 7 (1991) reported that winter home ranges of Bluebirds in South Carolina averaged 113.1 ha in size, an almost 10-fold increase from that of the breeding season. They attributed this increase to greater winter energy requirements, and the need to search outside their home range for food. Allen and Sweeney (1991) also found that wintering Bluebirds used edge habitat more than expected based on availability. Region-specific information on habitat requirements and animal behaviors is important because population responses to environmental stimuli can vary across a species’ range (Mehlman 1997, Whittingham et al. 2007). For instance, despite a trend for increasing Bluebird abundance nationally over the past 4 decades, Bluebird populations in central Kentucky have recently declined (Sauer et al. 2014 ). Mehlman (1997) demonstrated a trend for a decline in Bluebird abundance following a series of severe winters in the 1970s, but this effect varied across the species’ range. Many wintering birds join flocks; larger groups allow individuals to spend more time foraging as a result of decreased anti-predator vigilance by each individual, and may enhance the efficiency of finding food (Morse 1970, Sridhar et al. 2009). Group size may increase during periods of higher energetic demands (Caraco 1979), such as those imposed by inclement winter weather. Factors that might contribute to adjustments in group size by Bluebirds during unfavorable weather have not been described. Winters in temperate regions can create metabolically demanding conditions for birds. Arthropods, an important food source for Bluebirds, are not always available during winter, and Bluebirds include a wide variety of fruits in their winter diet (Pinkowski 1977, Pitts 1979). Fruit is a critical component of the diet of many other wintering neo-temperate bird species (Baird 1980), but its relative importance for bluebirds is not clear. Snow is likely to affect availability of ground-active arthropods, whereas ice, which can accumulate on tree branches during freezing-rain events, may also impact fruit availability, and has been shown to disproportionately affect the abundance of tree foraging bird species in subsequent years as compared to open-habitat, ground-foraging species (Blais et al. 2001). If fruit is an important food resource for Bluebirds in winter, identifying and describing conditions that cause a shift to frugivory is important for better understanding the ability of Bluebirds to survive such conditions. Because unpredictable winter events are correlated with fluctuations of Bluebird populations (Gowaty and Plissner 2015, Wetzel and Krupa 2013), identifying the habitat requirements and factors potentially limiting populations is important, especially within a regional context since Bluebird population changes are spatially variable (Mehlman 1997, Sauer et al. 2014). The goals of this study were to describe the habitat composition and size of the winter home ranges of Bluebirds in the Bluegrass ecoregion of Kentucky, and to examine how weather influences Bluebird habitat and space use, group size, group composition, foraging decisions, and diet. We used radio-telemetry to locate individual Bluebirds throughout the winter, and then related their behaviors to habitat availability and weather conditions. Northeastern Naturalist 3 T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 Vol. 24, Special Issue 7 Field-site Description All field research was conducted at the US Department of Defense Blue Grass Army Depot (BGAD), located in Madison County, KY (37°40'55"N, 84°13'16"W; Fig. 1). The installation consists of ~5907 ha, within which we focused on a study area of ~850 ha containing open fields and pastures dissected by woodlots, wooded stream corridors, roads, and buildings. The local population of Bluebirds presumably included both residents and migrant Bluebirds that breed at higher latitudes. Figure 1. (A) Location of study area within the state of Kentucky. (B) Home ranges (95% kernel isopleths) of Bluebirds within the Bluegrass Army Depot (white boundary line), and (C) at a larger geographic scale within the study area. Northeastern Naturalist T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 4 Vol. 24, Special Issue 7 Methods During January–February of 2010 and 2011, we located Bluebirds across the study area and used 12 m x 2.5 m x 36 mm mist-nets to capture 1 or more birds in each observed group. We used playback of Bluebird vocalizations to attract birds into the mist-nets. We used elastic nylon-string harnesses (Rappole and Tipton 1991) to attach radio-transmitters (Holohil, BD-2; 0.9 g; average = 2.5 ± 0.2 [SD] % of body mass) to 19 Bluebirds (2010: n = 11 birds, 2011: n = 8 birds), and uniquely marked each bird using 1 USGS numbered aluminum band and 3 color bands. We determined the age of captured birds (second-year [SY] or after-secondyear [ASY]) by examining the 10th primary coverts (Pitts 1985), and the sex of the birds by using plumage characteristics (Pyle 1997). Each radio-tagged Bluebird was tracked during daylight (08:00–18:00 hrs), typically 4–6 days per week, using a Yagi 3-element antenna and Telonics TR-4 receiver (Telonics, Inc., Mesa, AZ) for the duration of the transmitter battery life (60 days), or until transmitters fell off, or birds were depredated. We used a homing method to locate birds and visually confirmed their identity by sighting color bands or the transmitter’s whip antenna. Upon locating a radio-tagged bird, we used a portable GPS/data management device (Trimble Juno SB; Trimble Navigation Ltd., Sunnyvale, CA) to record the individual’s geo-referenced location, date, and time. During each observation, we recorded all data from a distance (typically >20 m) that appeared sufficient to avoid influencing the behavior of individuals or flocks. The average (± SD) interval between same-day observations was 81 ± 73 min. Foraging observations were recorded whenever possible as part of radio-tracking and began as soon as birds were visually located. Radio-telemetry allowed us to locate birds from a distance (typically 25−100 m). Observations lasted 2−5 minutes and were conducted on all birds visible in the group. The average (± SD) number of observations per group was 21 ± 5. Observations occurred throughout the day, with an average (± SD) of 30 ± 11 per daylight hour. All Bluebirds in a group typically foraged for the same type of prey. Foraging directed at arthropods was treated as a single category because of difficulty identifying prey. We assumed Bluebirds were foraging for arthropods when they exhibited their characteristic drop-foraging behavior (ground sallying; Goldman 1975) and this assumption was often confirmed by observation of arthropod ingestion. The absence of leaves from most woody plants made instances of frugivory and type of fruit consumed quickly discernible. Frugivory was recorded based on plant species, and we later combined all such instances into a single category for analysis. Occasionally 2 different food items were being consumed by Bluebirds in a group (e.g., frugivory interrupted by sporadic drop-foraging by one or more group members), in which case we used the dominant foraging mode to assign food type. Individual Bluebirds were typically observed in conspecific groups in a relatively small area. We estimated the size of the group and the sex of each group member during observations of radio-tagged birds. Determination of group size was often facilitated by extended flights of the entire group into open habitat, but estimates made in forest habitat or when movements were minimal may be biased Northeastern Naturalist 5 T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 Vol. 24, Special Issue 7 towards lower numbers. Thus, the estimates reported here likely represent minimum group sizes. We opportunistically collected fecal samples while banding birds and stored them on filter paper. Dried fecal samples were scraped into a petri dish for sorting and identification. Working under a dissecting microscope, we dripped small amounts of ethyl alcohol (70%) into the dishes and separated seeds, arthropod parts, and other materials using forceps and probes (Burger et. al. 1999). Arthropods were sorted and identified to Order using references by Ralph et al. (1985), Borror et al. (1989), and Burger et al. (1999), and counted based on the number of heads, or pairs of mandibles, wings, elytra, chelicerae, or other distinguishable body segments. For example, a head capsule was counted as one individual and every two mandibles of particular taxa were counted as one individual. We sorted by species and counted all seeds found in the samples. To identify seeds, we used reference material including books (Jones 2005), the seed collection of the EKU Herbarium, and dissected field collections from the study area. We also consulted with an entomologist (A. Braccia, Eastern Kentucky University, Richmond, KY, pers. comm.) and a botanist (R. Jones, Eastern Kentucky University, Richmond, KY, pers. comm.) for verification. All GPS locations were downloaded into ArcMap version 10.0 (ESRI, Redlands, CA) and projected using the NAD 1983 state plane (feet) Kentucky coordinate system. We generated home-range estimates using kernel density estimation (KDE) techniques that produce utilization distributions (UD) based on the relative density of telemetry locations over an area. We used the Geospatial Modelling Environment (Beyer 2012), a Program-R based supplement to ArcGIS, to conduct KDE analyses for all birds with >20 locations (n = 9). Results of KDE analysis are sensitive to both the resolution of the evaluation area (grid size) and the bandwidth (Seaman and Powell 1996). We used a fixed-bandwidth value of 20,000 for all analysis because it provided a suitable balance between over- and under-estimation of home ranges. We depicted a home-range boundary as the 95% isopleth of the kernel probability density function (Seaman et al. 1999). The core home-range area, representing areas of intensive use, is defined here by a 50% isopleth of the kernel probability function. The nonselective capture of Bluebirds used in our sampling design excluded many individuals present in the study area, so comprehensive insight of Bluebird home range overlap across the study area was not possible. However, 5 birds in 2010 appeared to occupy distinct, yet adjacent, home ranges. We calculated the percentage of overlapping area between these adjacent home ranges using ArcGIS. We used satellite imagery and a maximum-likelihood supervised image-classification process to categorize habitats (Palmeirim 1988), employing the image classification tool in ArcMap 10.0 to designate each pixel of the 1-m–resolution RGB satellite image of the study area (US Department of Agriculture, 2006) as either (1) wooded or (2) open habitat types based on the color profiles of the pixels. Representative training samples of the satellite imagery were selected using unambiguous areas of land cover within the study area and then further Northeastern Naturalist T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 6 Vol. 24, Special Issue 7 adjusted to minimize overlap within resultant scatterplots of the red, green, and blue visual spectra for the 2 habitat category profiles. Based on the color profiles of these training samples, we assigned all pixels in the study area to either open or wooded categories. Further correction of the resulting raster image map (e.g., removal of artifacts such as improperly categorized shadows) was completed manually using Adobe Photoshop CS (version 5.1, Adobe™) and referencing the original satellite imagery. Weather data were collected from an on-site weather monitoring station maintained by the US Army. Air temperature and wind speed were recorded every 15 min; thus, no weather observations were more than 7 min removed from an instantaneous reading. Snow depth was recorded daily 53 km north at the Lexington Bluegrass Airport (38°2' 23.99"N, 84°36'35.99"W) and collected from the National Oceanic and Atmospheric Administration’s National Climatic Data Center (www. Statistical analysis We quantified descriptive Bluebird 95% and 50% home-range metrics including area, habitat composition, and overlap between adjacent home ranges with ArcGIS. We compared the habitat composition of home ranges to that of all unused habitat in the study area using a one-sample t-test. In this case, the study area was defined as a polygon bound by the most-outward vertices of all bird home ranges over 2 years. We quantified the habitat surrounding each bird location within a 5-m radius to determine habitat use at each location. These data were used to compare habitat use both when snow cover was and was not present (i.e., ≥2.54 and 0 cm, respectively, as recorded from the Lexington Bluegrass airport) using a Mann-Whitney U-test. We used a nested-ANOVA to investigate the potential effect of snow cover on the distance of birds to edge habitat (the boundary between open and wooded spaces). For this analysis, we were interested in whether Bluebirds preferentially used edge habitats during periods of snow cover. To account for the repeated observations of individuals, each Bluebird location was nested by the identity of its group as determined by the individual radio-tagged bird that was used to find the group. Thus, group identity is based on the radio-tagged individual. We also used a nested-ANOVA to examine the potential effect of freezing temperatures (i.e., ≤0 °C) on Bluebird group size, again with each observation nested within group identity, as based on individual radio-tagged birds. Similarly, group sizes were compared during periods of snow cover (≥2.54 cm) and when snow was absent (0 cm) using a nested-ANOVA, with each observation of group size nested within a flock represented by the radio-tagged bird. We used a paired samples t-test to compare the number of males and females in groups. Sex composition of groups was also compared between periods of snow cover and when snow was absent using a nested-ANOVA, with the count of each sex nested within the flock as represented by the radio-tagged bird. Observations were not included in these analyses when the sex of one or more group members was not determined. To test for influence of snow cover, air temperature, and wind speed on the foraging behavior of Bluebirds, we used a logistic regression analysis. For this Northeastern Naturalist 7 T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 Vol. 24, Special Issue 7 analysis, foraging behavior, the response variable, was classified dichotomously as directed towards fruit or arthropod prey. Group identity, as determined by the presence of a radio-tagged individual, was included as a random effect to account for repeated sampling. For diet, we report the percent occurrence of taxa (i.e., the percent of samples in which a taxa occurs; Rosenberg et. al. 1990). To test for differences in diet based on weather, we used a 2 x 2 contingency table analysis to compare the relative occurrence of fecal samples with and without seeds when the temperatures were greater or less than 0 °C. All analyses were performed using SPSS v. 18.0 (SPSS Statistics 2009). All means are reported ± SE, unless otherwise noted. Statistical significance was accepted at α = 0.05. Results There were 21 and 19 days of snow cover of ≥2.54 cm during January–February of 2010 and 2011, respectively (Fig. 2). Daytime temperatures averaged 31.2 °C and 31.5 °C during January–February of 2010 and 2011, respectively. We radio-tagged and tracked 19 individuals, including 15 males (7 SY and 8 ASY) and 4 females (2 SY and 2 ASY), for an average span of 23 ± 15 (SD) days (range = 1–51). We obtained an average of 20 ± 13 (SD) locations per individual (range = 2–44). Each day of tracking yielded an average of 1.7 ± 0.6 (SD) locations per individual. Twelve of the 19 radio-transmitters were subsequently recovered either on the ground or among foliage. Whereas most of the recovered transmitters appeared to have fallen off (the birds were later re-sighted), at least 3 Bluebirds with transmitters appeared to have been killed by predators as indicated by a transmitter located among or near numerous feathers or other body parts. In 2 cases, location data of 2 Bluebirds were combined for analyses. One pair of males from the same group was captured and radio-tagged simultaneously; 87% of 39 locations were shared by both birds (i.e., the two birds were typically found moving or foraging together). The other 2 birds (a male and female) were also captured simultaneously, but were tracked over separate time periods; the female was tracked for 14 days, after which the radio-tag fell off. We then recaptured and radio-tagged the male at the same location, and tracked it for 11 days. An exploratory home-range analysis for the 2 birds revealed nearly identical overall and core home ranges, so we combined their locations into a single home-range analysis. Including the 2 shared home ranges described above, 9 Bluebirds (or pairs) had >20 location observations (average = 30.3 ± 7.5 [SD], range = 22–44). For these 9 birds, the average home-range area (95% utilization distribution) was 29.2 ± 2.4 ha (range = 16.3–42.3 ha), and the average core home-range area (50% utilization distribution) was 7.1 ± 0.6 ha (range = 3.8–9.9 ha) (Fig. 1). Of 5 birds that had an adjacent home range, an average of 9.4 ± 2.5% of the 95% home-range–estimate area was shared by neighboring birds. Core home ranges did not overlap. Northeastern Naturalist T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 8 Vol. 24, Special Issue 7 Bluebird home ranges consisted of 39.6 ± 2.6% wooded habitat and 60.4 ± 2.6% open habitat. The percentage of wooded habitat within home ranges did not differ significantly from the overall percentage of wooded habitat in the entire study area (35.7%) (t8 = 1.5, P = 0.18). Similarly, core home-range areas consisted of 41.1 ± 3.5% wooded habitat and 58.9 ± 3.5% open habitat, which was not significantly different from the composition of available habitat in the study area (t8 = 1.6, P = 0.16). The mean percentage of wooded area surrounding each observation (5-m radius) was 60.0 ± 2.0% (n = 344). Habitat within the 5-m radius around each point was frequently either entirely (100%) wooded or entirely open (0% wooded). The mean percentage of wooded habitat per observation (5-m radius) during periods of snow cover (54.4 ± 4.3%, n = 94) did not differ from that during periods when snow was Figure 2. Snow depth at Lexington Bluegrass Airport during January–February of 2010 and 2011. Northeastern Naturalist 9 T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 Vol. 24, Special Issue 7 absent (63.1 ± 2.5%, n = 250) (Mann-Whitney U-test: P = 0.14). The distance of Bluebirds to the edge of wooded habitat did not differ during periods of snow cover (F1,91 = 0.01, P = 0.92). Radio-tagged Bluebirds were almost always found in conspecific groups (97% of observations). Average group size was 5 ± 0.1 individuals (n = 300), with a maximum of 16 individuals observed in a group that included a radio-tagged bird (Fig. 3). Groups included more males (2 ± 0.1) than females (1 ± 0.1) (t272 = 20.0, P < 0.001), and the presence of snow cover did not influence the sex composition of groups (F1,83 = 2.6, P = 0.11). Mean Bluebird group size during below-freezing temperatures (5 ± 0.2 individuals) was larger than when temperatures were above 0 °C (4 ± 0.2 individuals) (F1,62 = 4.8, P = 0.032). Similarly, group size during periods of snow cover (6 ± 0.3 individuals) was larger than that when snow was absent (5 ± 0.1 individuals) (F1,266 = 5.0, P = 0.02). Flocks were typically found with all members foraging on either arthropods or fruit (91% of n = 235 observations), but not both. Most foraging attempts were directed at arthropods (>65% of observations for either year), and relative frequencies of frugivory for different fruits varied between years, with Phoradendron leucarpum Raf. (Oak Mistletoe) most common in 2010, and Celtis occidentalis L. (Hackberry) the most common in 2011 (Table 1). Logistic regression analysis indicated that frugivory was more likely than foraging on arthropods during observations with low temperatures, low wind speed, and the presence of snow (Table 2). Analysis of fecal samples from 33 birds included 105 items identified from 7 arthropod orders (Mesostigmata, Araneae, Orthoptera, Hemiptera, Hymenoptera, Figure 3. Eastern Bluebird group sizes (n = 300) during January–February of 2010 and 2011. Northeastern Naturalist T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 10 Vol. 24, Special Issue 7 Coleoptera, and Lepidoptera) and 5 species of seeds (Table 1). Seeds represented 60% of all items and occurred in 22 samples. There was an equal frequency of fecal samples that contained only arthropods (33%), only fruit seeds (33%), and both arthropods and fruit seeds (33%). Fecal samples collected when the temperature was below 0 °C (n = 20) were 4.4 times more likely to have seeds present than samples collected at higher temperatures (n = 13) (χ2 = 4.06, P = 0.04, Fig. 4). Table 2. Results of logistic regression analysis on forage type (fruit or arthropods) by weather conditions. Low air temperature, low wind speed, and snow presence increased the likelihood that eastern bluebirds foraged on fruit. Group identity was included as random variable to account for repeated measurements so coefficients and odds ratio are not reported. 95% CI Wald’s P-value B Lower Upper Odds ratio Low air temperature 14.5 less than 0.001 0.19 0.09 0.29 1.22 Low wind speed 8.2 0.004 0.57 0.18 0.96 1.77 Snow presence 5.3 0.020 1.23 0.18 2.28 3.44 Group identity 13.0 0.791 - - - - Table 1. Percent foraging observations of Eastern Bluebirds in 2010 (n = 139), 2011 (n = 96), and in both years combined among different food categories. Other fruit included Prunus serotine Ehrhart (Black Cherry), Symphoricarpos orbiculatus Moench( Coralberry), Lonicera japonica Thunb. (Japanese Honeysuckle)*, and Toxicodendron radicans L. (Poison Ivy)*. Plant species with an asterisk were also detected in fecal samples. Food category Both years (%) 2010 (%) 2011 (%) Arthropods 71.9 76.3 65.6 Hackberry* 10.2 0.1 24.0 Oak Mistletoe* 8.5 13.7 1.0 Red Cedar* 6.0 7.9 3.1 Other fruit 3.4 1.4 6.3 Figure 4. Percent of fecal samples with seeds present (i.e., only seeds, or seeds and arthropods) or with only arthropods collected from Eastern Bluebirds during January– February of 2010 and 2011 when temperatures were above or below freezing. Northeastern Naturalist 11 T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 Vol. 24, Special Issue 7 Discussion Home range and habitat use The mean estimated 95% winter home range size of Bluebirds in this study (29.2 ha) was smaller than that reported by Allen and Sweeney (1991) in South Carolina (113.1 ha). However, this difference is likely at least partially due to the use of different home-range estimation techniques; the minimum convex polygon method used by Allen and Sweeney (1991) is known to overestimate home-range sizes (Anderson 1982), whereas the kernel-based utilization distribution that we used emphasizes focal areas of intensive use and minimizes the influence of distant, isolated locations (i.e., outliers). The mean size of 95% winter home ranges of Bluebirds in this study was similar to that of Bluebirds during seasonal transition periods (i.e., August–November and February–April) described by Savereno (1991) in South Carolina (28.0 ha). In his study, home-range area was calculated using cumulative area curves (Odum and Kuenzler 1955), which is more comparable to the kernelbased utilization distribution used in this study than the minimum convex polygon method. Our sample size was too small to examine the possible effects of sex and age on the size of Bluebird winter home ranges. With the exception of 2 pairs whose space use data were combined for home-range analysis, we found little (95% home-range estimates) and no (core home-range estimates) overlap in the home ranges of Bluebirds. Territory maintenance and defense during the non-breeding season has been documented in resident and migratory bird species, and is typically driven by variation in the abundance of limited food resources (Brown et al. 2000, Safina and Utter 1989, Townsend et al. 2010). The distinct, minimally overlapping home ranges of Bluebirds in this study may also represent resource-influenced territoriality, but, in this case, the territories appear to be specific to groups. The 2 pairs that occupied almost identical home ranges over 45 days of radio-tracking suggest that Bluebird groups (or certain pairs) persist for extended periods of the winter , but the specific roles of these relationships and other potential effects of group behavior on space use of wintering Bluebirds have yet to be determined. It is also possible that winter territoriality is related to defense of breeding territories. Thomas (1946) noted that wintering Bluebirds in Arkansas sometimes engaged in intraspecific competition (i.e., fighting and singing) around nest boxes, suggesting that defense of nest sites may occur throughout the year, though to a lesser degree in winter. Thus, in some populations, the co-occurrence of a male and female within home ranges may be related to the winter maintenance of a breeding territory. Home ranges of Bluebirds in this study were composed of more open habitat than wooded habitat, although wooded areas made up almost 40% of home ranges. Habitat structure was not measured in this study, but, in all cases, home ranges included a heterogeneous mixture of both open and wooded areas, including lone trees and wooded corridors. We found no difference in habitat composition between Bluebird home ranges and the study area outside of those home ranges, and also no difference in habitat composition between core (50%) and 95% home ranges. Because we used radio-telemetry to locate birds, our observations should not be Northeastern Naturalist T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 12 Vol. 24, Special Issue 7 biased by differences in detectability between habitats. Allen and Sweeney (1990) and Savereno (1991) reported that wintering Bluebirds in South Carolina used edge habitat more than expected, and Levey et al. (2008), also in South Carolina, found that movement along edges was preferred by wintering Bluebirds over using open space within corridors. Because we only categorized open and wooded habitat, our data cannot test if Bluebirds prefer edge habitat, as has been found in other studies (Allen and Sweeney 1990, Levey et al. 2008, Savereno 1991). Mean distance to edge did not change during periods of snow cover, nor did the mean percentage of wooded habitat within a 5-m radius. These results suggest that Bluebirds do not dramatically shift habitat occupancy during periods of snow cover, and that their presumed preference for edge habitat is unaffected. Petit (1989) demonstrated that wintering woodland birds move into habitat patches with greater cover (i.e., mature pine stands) during harsh weather, but habitat occupancy by Bluebirds in this study appeared to be unchanged by snow cover of at least 2.54 cm in depth. Brotons (1997) described how Parus ater L. (Coal Tit) responded to snow presence by changing foraging methods within a patch rather than seeking a new habitat, and our results suggest that Bluebirds may respond to the presence of snow in a similar way. For example, Pinkowski (1977) described a positive relationship between Bluebird foraging height and temperature as well as sunshine percentage (i.e., the proportion of time that shadows were cast), possibly as a response to changing insect detectability related to weather. In that scenario, a change of habitat as temperatures drop may be unnecessary to meet immediate foraging needs, requiring only a behavioral response within the habitat. Even a shift toward frugivory may not require movement to a new habitat (e.g., into a wooded area) if fruits are available nearby. Despite no apparent change in habitat occupancy during snow cover, Bluebirds may cope with unfavorable weather and its associated energetic demands in other ways, such as by shifting diet or group size. Several studies have documented communal night roosts in trees and bird houses during cold weather and snowstorms (Forbush 1929, Frazier and Nolan 1959, Pitts 1978, Thomas 1946), suggesting weather influences bluebird habitat use and group size during nighttime. Because of site access limitations during the night, we were unable to include such behavior as part of this study, although we did observe a daytime roost in a tree of 19 individuals during a severe winter storm. Foraging behavior Bluebirds in this study were more likely to engage in frugivory when temperatures and wind speeds were low, and when snow cover was present, with snow cover having the strongest effect. Low temperatures decrease arthropod activity (Mellanby 1939), and snow reduces insect availability to Bluebirds (Frazier and Nolan 1959, Pitts 1978), particularly by covering ground-active arthropods. Since drop foraging is the most common foraging method used by Bluebirds in all seasons (Gowaty and Plissner 2015), it’s not surprising that snow cover causes a shift to alternative methods. Together, low temperatures and snow cover may make insectivory less reliable and possibly increase the cost of this strategy to a suboptimal level. Increased frugivory when wind speeds were low may indicate an ability of Northeastern Naturalist 13 T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 Vol. 24, Special Issue 7 Bluebirds to manage thermal stress by altering foraging behavior. Grubb (1975) found that several species of woodland birds foraged at lower heights during periods of high wind (i.e., 2–3 m/s), and suggested that birds seek areas of low thermal stress (i.e., closer to the ground) under windy conditions. Similarly, birds have been shown to avoid windward edges of habitat patches and favor leeward edges (Dolby and Grubb 1999). High winds may encourage insectivory and discourage frugivory as a way to minimize thermal loss to wind, which would likely increase when Bluebirds forage for fruits high in trees. Both Oak Mistletoe and Hackberry fruit, the 2 most common targets of frugivory, were typically located well above ground level. The presence of high wind may limit the ability of Bluebirds to forage efficiently during low temperatures and snow cover; so, when combined, these factors may create the most energetically demanding (and behaviorally restrictive) conditions that Bluebirds experience during winter. The fruits most frequently consumed by Bluebirds in this study differed between years. For example, based on observational data we found that Oak Mistletoe frugivory accounted for 13.7% of all foraging observations in 2010, and only 1% in 2011. Instances of Hackberry frugivory displayed an opposite pattern with 0.07% and 24.0% of all observations in 2010 and 2011, respectively. Although both are native species, Oak Mistletoe tends to have relatively high lipid content (Stiles 1993), whereas Hackberry has low levels of lipids (Johnson et al. 1985, Stiles 1980). The availability of food was not determined in this study, so it is not possible to determine whether the relative abundances of these food items accounted for this pattern. Nonetheless, the composition of fruit in the diet of wintering Bluebirds occupying the same general area can differ considerably between years. Therefore, the relative value of any one fruit resource over another may vary annually, just as the relative value of fruit over arthropods may depend on immediate weather circumstances. Although Bluebirds are known to consume at least 60 types of fruit in winter (Hoyo et al. 1992). Most winter studies of Bluebirds in temperate North America document the consumption of fewer than 10 fruit species, most of which represent only a minor dietary component (Morland 1978, Pinkowski 1977, Pitts 1978, Savereno, 1991). We observed Bluebirds foraging on 7 types of fruit, but most foraging was directed toward fruit of Hackberry, Mistletoe, and Juniperus virginiana L. (Red Cedar). Although fruits have been reported as important for wintering Bluebirds elsewhere, these 3 species were absent or only a minor dietary component for wintering Bluebirds in Ohio (Morland 1978), South Carolina (Savereno 1991), and Tennessee (Pitts 1979), suggesting regional or temporal differences, likely based on availability. Arthropod segments, comprising at least 7 different orders, were the most common items in fecal samples, followed in decreasing order of occurrence by seeds of Red Cedar, Oak Mistletoe, Poison Ivy, and Japanese Honeysuckle. Seeds were more commonly found in fecal samples collected during periods with sub-freezing temperatures reinforcing our conclusions from foraging observations that Bluebird diets shift toward fruit during periods with low temperatures. Northeastern Naturalist T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 14 Vol. 24, Special Issue 7 Group size If the metabolic costs of birds increase during unfavorable weather conditions (i.e., low temperature), individuals may have to spend more time foraging to meet metabolic demands. One strategy for increasing an individual’s foraging time is termed the “group-size effect”, whereby individual foraging time can increase with flock size as a result of improved collective anti-predator vigilance (Beauchamp 1998, Caraco 1979). Hogstad (1988) observed this effect among flocks of wintering Poecile montanus von Baldenstein (Willow Tit) during cold days. Large Bluebird flocks during snowstorms have been sporadically reported elsewhere during winter (e.g., Thomas 1946), including accounts of Sialia currucoides Bechstein (Mountain Bluebird) and Sialia Mexicana Swainson (Western Bluebird) (Allen and Brewster 1883, Henderson 1903). By assembling in larger numbers during inclement weather, Bluebirds may be able to locate new food sources while simultaneously benefiting from a group-size effect. Because Bluebird group size was larger during periods with snow cover, when frugivory was also higher, increased flocking behavior as a strategy to locate fruit resources seems possible (e.g., Elgar and Catterall 1982, Ficken 1981). Thomas (1946) described “roaming and shifting” among Bluebird flocks during winter, suggesting a lack of cohesiveness within Bluebird groups. Although accurately estimating group membership was not always possible, it was not uncommon to repeatedly observe the same uniquely banded individuals together over a period of weeks. The 2 pairs of Bluebirds whose home-range estimates showed considerable overlap is an additional indication that group cohesiveness, as well as stable territoriality, can be maintained during the winter months. So it appears that Bluebirds maintain winter territories in small groups through most of the winter, but group membership may change, particularly during periods of inclement weather when group size increases. More work is needed to understand how group size and membership is related to foraging. Our results suggest Bluebirds respond to inclement winter weather by changing foraging behaviors and increasing group size; however, it remains to be determined how these responses are driven by the distribution of food, their ability to find resources, predator avoidance, and even thermoregulation (e.g., communal roosting). Although we detected no mortality as a result of winter weather, our sample size to do so was small. Winter weather may lead to mortality by reducing food availability and contributing to physiological stress through heat loss. Concurrent changes in foraging and space-use behavior may also make birds more vulnerable to predation. We found evidence for changes in behavior in response to weather, but more work is needed to understand the population-level impacts. Acknowledgments We thank A. Colwell and N. White of the Bluegrass Army Depot for providing logistical support and access to the site. We’re grateful for taxonomic consultations with A. Braccia and R. Jones. G. Ritchison, and C. Elliott provided comments on an earlier version of the manuscript. S. Smedley, T. Wickman, and 2 anonymous reviewers provided valuable Northeastern Naturalist 15 T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 Vol. 24, Special Issue 7 comments on the manuscript. This work was supported by a grant from Eastern Kentucky University. Procedures related to the capture and handling of Bluebirds were reviewed by Eastern Kentucky University’s Institutional Animal Care and Use Committee and approved as Protocol #01-2010. Literature Cited Allen, D.H., and J.R. Sweeney. 1990. Home range and habitat use of Eastern Bluebirds in South Carolina. Proceedings Annual Conference Southeastern Association Fish and Wildlife Agencies 44:295–303. Allen, J.A., and Brewster, W. 1883. List of birds observed in the vicinity of Colorado Springs, Colorado, during March, April, and May, 1882. Bull. Nuttall Ornithol. Club 8:189–198. Anderson, D.J. 1982. The home range: A new nonparametric estimation technique. Ecology 63:103–112. Baird, J.W. 1980. The selection and use of fruit by birds in an eastern forest. Wilson Bulletin 92:63–73. Beauchamp, G. 1998. The effect of group size on mean food intake rate in birds. Biological Reviews of the Cambridge Philosophical Society 73:449–472. Beyer, H.L. 2011. Geospatial Modelling Environment (GME) version Available online at Accessed 22 February 2012. Blais J., J.P. Savard, and J. Gauthier. 2001. Impact of an ice storm on resident bird populations in eastern North America. The Forestry Chronicle 77:661–666. Borror, D.J., C.A. Triplehorn, and N.F. Johnson. 1989. An Introduction to the Study of Insects, 6th Edition. Saunders College Publishing, Toronto, ON, Canada. 800 pp. Brotons, L. 1997. Changes in foraging behavior of the Coal Tit (Parus ater) due to snow cover. Ardea 85:249–257. Brown, D.R., P.C. Stouffer, and C.M. Strong. 2000. Movement and territoriality of wintering Hermit Thrushes in southeastern Louisiana. Wilson Bulletin 112:347–353. Burger, J.C., M.A. Patten, J.T. Rotenberry, and R.A. Redak. 1999. Foraging ecology of the California Gnatcatcher deduced from fecal samples. Oecologia 120:304–310. Caraco, T. 1979. Time budgeting and group size: A test of theory. Ecology 60:618–627. Dolby, A.S., and T.C. Grubb Jr. 1999. Effects of winter weather on horizontal and vertical use of isolated forest fragments by bark-foraging birds. Condor 101:408–412. Duriez, O., H. Fritz, F. Binet, Y. Tremblay, and Y. Ferrand. 2005. Individual activity rates in wintering Eurasian Woodcocks: Starvation versus predation risk trade-off? Animal Behaviour 69:39–49. Easterling, D.R., G.A. Meehl, C. Parmesan, S.A. Changnon, T.R. Karl, and L.O Mearns. 2000. Climate extremes: Observations, modeling, and impacts. Science 289:2068–2074. Elgar, M.A., and C.P. Catterall. 1982. Flock size and feeding efficiency in House Sparrows. Emu 82:109–111. Environmental Systems Research Institute, Inc. (ESRI). 2010. ArcView GIS version 10.0. Redlands, CA. Ficken, M.S. 1981. Food finding in Black-capped Chickadees: Altruistic communication? Wilson Bulletin 93:393–394. Forbush, E.H. 1929. Birds of Massachusetts and other New England States, Part III. Berwick and Smith Company, Norwood, MA. 466 pp. Frazier, A., and V. Nolan. 1959. Communal roosting by the Eastern Bluebird in winter. Bird-Banding 30:219–226. Northeastern Naturalist T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 16 Vol. 24, Special Issue 7 Gitzen, R.A., J.J. Miillspaugh, and B.J. Kernohan. 2006. Bandwidth selection for fixedkernel analysis of animal utilization distributions. Journal of Wildlife Management 70:1334–1344. Goldman, P. 1975. Hunting behavior of Eastern Bluebirds. Auk 92:798–801. Gowaty, P.A, and J.H. Plissner. 2015. Eastern Bluebird (Sialia sialis).No. 381, In A. Poole (Ed.). The Birds of North America Online. Cornell Lab of Ornithology, Ithaca, NY. Available online at Accessed 19 June 2015. Grubbs, T.C. 1975. Weather-dependent foraging behavior of some birds wintering in a deciduous woodland. Condor 77:175–182. Hall, P., J.S. Marron, and B.U. Park. 1992. Smoothed cross validation. Probability Theory and Related Fields 90:149–163. Henderson, J. 1903. Mountain Bluebird increasing in Boulder, Colorado. Wilson Bulletin 15:74–75. Hogstad, O. 1988. Advantages of social foraging of Willow Tits (Parus montanus). Ibis 130:275–283. Hoyo, J. del, A. Elliott, J. Sargatal, and J. Cabot. 1992. Handbook of the Birds of the World. Lynx Edicions, Barcelona, Spain. 696 pp. Johnson, R.A., M.F. Willson, J.N. Thompson, and R.I. Bertin. 1985. Nutritional values of wild fruits and consumption by migrant frugivorous birds. Ecology 66:819–827. Jones, R.L., Plant Life of Kentucky. University Press of Kentucky, Lexington, KY. 833 pp. Levey, D.J., J.J. Tewksbury, and B.M. Bolker. 2008. Modelling long-distance seed dispersal in heterogeneous landscapes. Journal of Ecology 96:599–608. Meehan, T.D., W. Jetz, and J.H. Brown. 2004. Energetic determinants of abundance in winter landbird communities. Ecology Letters 7:532–537. Mehlman, D. W. 1997. Change in avian abundance across the geographic range in response to environmental change. Ecological Applications 7:614–624. Mellanby, K. 1939. Low temperature and insect activity. Proceedings of the Royal Society of London 127:473–487. Monroe, B.L., Jr. 1978. Analysis of Kentucky’s breeding birds: Declining species. Kentucky Warbler 54:19–26. Morland, K. 1978. The provision of roosting boxes as a management technique for wintering Eastern Bluebirds and notes on associated behavior. M.Sc. Thesis. The Ohio State University, Columbus, OH. Morse, D.H. 1970. Ecological aspects of some mixed-species foraging flocks of birds. Ecological Monographs 40:119–168. Odum, E.P., and E.J. Kuenzler. 1955. Measurement of territory and home range size in birds. Auk 72:128–137. Palmeirim, J.M. 1988. Automatic mapping of avian species habitat using satellite imagery. Oikos 52:59–68. Palmer-Ball, B. 2015. Breeding bird survey evidence of Eastern Bluebird mortality during winter. The Kentucky Warbler 91:93. Petit, D.R. 1989. Weather-dependent use of habitat patches by wintering woodland birds. Journal of Field Ornithology 60:241–247. Pinkowski, B.C. 1977. Foraging behavior of the Eastern Bluebird. Wilson Bulletin 89:404– 414. Pitts, T.D. 1978. Eastern Bluebird mortality at winter roosts in Tennessee. Bird-banding 49:77–78. Northeastern Naturalist 17 T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 Vol. 24, Special Issue 7 Pitts, T.D. 1979. Foods of Eastern Bluebirds during exceptionally cold weather in Tennessee. Journal of Wildlife Management 43:752–754. Pitts, T.D. 1985. Identification of second-year and after-second-year Eastern Bluebirds. Journal of Field Ornithology 56:422–424. Plissner, J.H., and P.A. Gowaty. 1995. Eastern Bluebirds are attracted to two-box sites. Wilson Bulletin 107:289–295. Porter, W.F., G.C. Nelson, K. Mattson. 1983. Effects of winter conditions on reproduction in a northern Wild Turkey population. Journal of Wildlife Management 47:281–290. Pyle, P. 1997. Identification Guide to North American Birds. Slate Creek Press, Bolinas, CA. 732 pp. Ralph, C.P., S.E. Nagata, and C.J. Ralph. 1985. Analysis of droppings to describe diets of small birds. Journal of Field Ornithology 56:165–174. Rappole, J.H., and A.R. Tipton. 1991. New harness design for attachment of radio transmitters to small passerines. Journal of Field Ornithology 62:335–337. Rappole, J.H., and M.V. McDonald. 1994. Cause and effect in population declines of migratory birds. Auk 111:652–660. Rosenberg, K.V., and R.J. Cooper. 1990. Approaches to avian diet analysis. Studies in Avian Biology 13:80–90. Safina, C., and J.M. Utter. 1989. Food and winter territories of Northern Mockingbirds. Wilson Bulletin 101:97–101. Sauer, J.R., and S. Droege. 1990. Recent population trends of the Eastern Bluebird. Wilson Bulletin 102:239–252. Sauer, J.R., J.E. Hines, J.E. Fallon, K.L. Pardieck, D.J. Ziolkowski, Jr., and W.A. Link. 2014. The North American Breeding Bird Survey: Results and analysis 1966–2013, Version 1.30.2015. USGS Patuxent Wildlife Research Center, Laurel, MD. Available online at, Accessed 9 August 2016. Savereno, A.J. 1991. Seasonal home ranges of and habitat use by Eastern Bluebirds. M.Sc. Thesis. Clemson University, Clemson, SC. 97 pp. Seaman, D.E., and R.A. Powell. 1996. An evaluation of the accuracy of kernel density estimators for home-range analysis. Ecology 77:2075–2085. Seaman, D.E., J.J. Millspaugh, B.J. Kernohan, G.C. Brundige, K.J. Raedeke, and R.A. Gitzen. 1999. Effects of sample size on kernel home-range estimates. Journal of Wildlife Management 63:739–747. Sloan, N.F., and D.J. Carlson. 1980. Eastern Bluebird home-range determination using radio telemetry. Inland Bird Banding 52:20–22. SPSS Statistics. 2009. Version 18.0.0. SPSS, Inc., Chicago, IL. Sridhar, H., G. Beauchamp, K. Shanker. 2009. Why do birds participate in mixed-species foraging flocks? A large-scale synthesis. Animal Behaviour 78:337–347. Stamm, A.L. 1979a.The spring season of 1979. Kentucky Warbler 55:46–50. Stamm, A.L. 1979b.The nesting season, summer 1979. Kentucky Warbler 55:55–58. Stanback, M.T., and E.K. Rockwell. 2003. Nest-site fidelity in Eastern Bluebirds (Sialia sialis) depends on the quality of alternative cavities. Auk 120:1029–1032. Stiles, E.W., 1980. Patterns of fruit presentation and seed dispersal in bird-disseminated woody plants in the eastern deciduous forest. American Naturalist 116:670–688. Stiles, E. 1993. The influence of pulp lipids on fruit preference by birds. Vegetatio 107/108: 227–235. Thomas, R.H. 1946. A study of Eastern Bluebirds in Arkansas. Wilson Bulletin 58:143–183. Townsend, J.M., C.C. Rimmer, and K.P. McFarland. 2010. Winter territoriality and spatial behavior of Bicknell’s Thrush (Catharus bicknelli) at two ecologically distinct sites in the Dominican Republic. Auk 127:514–522. Northeastern Naturalist T.J. Weinkam, G.A. Janos, and D.R. Brown 2017 18 Vol. 24, Special Issue 7 US Department of Agriculture. 2006. NAIP digital georectified image. Farm Service Agency. Accessed 20 November 2010. Webster, P.J., G.J. Holland, J.A. Curry, and H.R. Change. 2005. Changes in tropical cyclone number, duration, and intensity in a warming environment. Science 309:1844–1846. Wetzel, D.P., and J.J. Krupa. 2013. Where are the Bluebirds of the Bluegrass? Eastern Bluebird decline in central Kentucky. American Midland Naturalist 169:398–408. Whittingham, M.J., J.R. Krebs, R.D. Swetnam, J.A. Vickery, J.D. Wilson, and R.P. Freckleton, 2007. Should conservation strategies consider spatial generality? Farmland birds show regional not national patterns of habitat association. Ecology Letters 10:25–35. Wilson, G. 1962. Field Notes: The tragedy of the Bluebirds. Kentucky Warbler 38:49–50. Wilson, G., and A.L. Stamm. 1960. The 1959–60 winter season and its effect on Kentucky birdlife. Kentucky Warbler 36:39–43. Zuckerberg, B., D.N. Bonter, W.M. Hochachka, W.D. Koenig, A.T. DeGaetano, and J.L. Dickinson. 2011. Climatic constraints on wintering bird distributions are modified by urbanization and weather. Journal of Animal Ecology 80:403–413.