Research Paper

INTRODUCTION

Aerial insectivores have recently received considerable attention because numerous studies indicated significant population declines (Sauer et al. 2017, Rosenberg et al. 2019, Spiller and Dettmers 2019). Aerial insectivores have the steepest decline of any bird group, i.e., 59% since 1970 (North American Bird Conservation Initiative Canada 2019), with the most precipitous decline beginning in the 1980s. These declines follow a geographic gradient and are greatest in northeastern North America (Nebel et al. 2010, Smith et al. 2015). Causes of declines for aerial insectivores are likely linked to habitat loss (i.e., agricultural intensification and forest maturation), declining insect populations, and changes in phenology spurred by climate change (North American Bird Conservation Initiative Canada 2019, Spiller and Dettmers 2019).

The Eastern Whip-poor-will (Antrostomus vociferus), one such aerial insectivore, has experienced a 69% population decline across its breeding range since 1970 (Rosenberg et al. 2016) and was recently designated as Near Threatened on the IUCN Red List (BirdLife International 2018). Eastern Whip-poor-wills follow a similar pattern as other species of avian aerial insectivores, with some of the highest declines in the northeastern portion of its range. Indeed, in Massachusetts, Breeding Bird Survey data indicate that Eastern Whip-poor-wills have experienced an annual decline of 4.0% (1966–2019; Sauer et al. 2020). Key threats linked to population declines include habitat loss and declines in high-quality food resources, i.e., Saturnidae moths (English et al. 2018, Spiller and Dettmers 2019, Cink et al. 2020), but little information is known about these factors on migration or the wintering grounds (Cink et al. 2020).

Several recent studies highlight the lack of information regarding wintering grounds for aerial insectivores. In Canada, English et al. (2017) used light-level geolocators attached to birds on the breeding grounds to identify Eastern Whip-poor-will wintering areas, migratory routes, migratory stopovers, and variability in timing of migratory movements. The group identified wintering areas along the Gulf Coast of North and Central America and called for further research into the precise location of wintering territories and how wintering birds use them. For Eastern Whip-poor-wills breeding throughout the Midwest United States, studies used archival GPS tags to identify and measure winter locations, winter home range habitat composition, and migratory connectivity (Tonra et al. 2019, Skinner et al. 2022). Tonra et al. (2019) calculated discrete home ranges (0.50–10.85 ha) primarily composed of closed-canopy forest and indicated that a few birds had two wintering territories, choosing to relocate in the middle of the winter. Studies documenting large-scale movement on the wintering grounds (Fraser et al. 2017, Ng et al. 2018) also called for more research on the frequency of mid-season relocations and driving factors for these movements because they may result in lower fitness (Smith et al. 2011).

Avian ecology during the wintering period remains understudied relative to that of the breeding season, even though winter events can limit populations and be drivers of population change for migrant birds (Sherry and Holmes 1996, Taylor and Stutchbury 2016, Rushing et al. 2017). Habitat quality on the wintering grounds can influence avian survival and body condition and have carry-over effects into other portions of the annual cycle (Marra and Holmes 2001, Norris et al. 2004, Harrison et al. 2011). Resources, competition, safety, knowledge of a location, and niche conservatism can drive habitat selection (see Albert et al. 2020). Unfortunately, loss of tropical forests, up to 120 million hectares by 2030, is expected to continue via intensification of agriculture, whereas, in contrast, temperate forests continue to see increases in forest cover (Jenkins 2003, Gibbs et al. 2010, Song et al. 2018). Research on migratory bird use of disturbed and agricultural habitats during the winter has included an array of habitats, such as secondary forest and coffee, tree, and oil palm plantations (Komar 2006, Bennett et al. 2018, Şekercioğlu et al. 2019). Still, studies target diurnal migratory bird species. Little to no information is available about Nightjar (Caprimulgidae sp.) use of these habitats and the role of loss or conversion of forest to agriculture in affecting Nightjar populations.

This study examined Eastern Whip-poor-will winter ecology of birds nesting in the northeastern United States by documenting arrival and departure timing, wintering ground locations, and winter site fidelity. We also sought to describe winter home ranges, e.g., size, and examine habitat selection at multiple spatial scales, covering fine-, local-, and landscape-scale selection, i.e., multi-order selection (Johnson 1980). In particular, we examined land cover types (e.g., forest, agriculture, developed) at the home range, local (2-km), and landscape (5-km) scales. We employed a multi-scale approach to examine if selection patterns differed across spatial scales (Jackson and Fahrig 2015). We evaluated three research questions related to land cover: (1) Does land cover differ between used and available plots at the home range, 2-km, and 5-km scales? (2) Does forest cover at first home ranges differ for sedentary (single home range) and mobile (multiple home ranges) birds? (3) Does forest cover at first and second home ranges differ for mobile birds? Based on information from habitat use on the breeding grounds, we hypothesized that habitat selected by Eastern Whip-poor-wills would differ from random sites available to birds, and birds would choose home ranges within large forest blocks adjacent to open habitat for foraging (Cink et al. 2020).

METHODS

GPS tag deployment and collection

In May–June of 2018 and 2019, we used a Foxpro broadcasting unit with conspecific playback recordings to lure territorial adult male Eastern Whip-poor-wills into mist-nets, where we opened mist-nets and initiated playback shortly after sunset. We focused capture efforts at three sites in Massachusetts with high-density breeding populations of Eastern Whip-poor-will, including Bolton Flats Wildlife Management Area (WMA; 42.475°N, 71.647°W), Montague Plains WMA (42.569°N, 72.534°W), and Joint Base Cape Cod (41.689°N, 70.546°W). When captured, we banded birds with U.S. Geological Survey bands and recorded the bird’s age (i.e., second-year, after-second-year) and sex. In addition, we placed 1.2-g archival GPS loggers (PinPoint 10 GPS tag, Lotek Wireless Inc.), that were approximately 2% of the bird’s mass, on each bird’s back using a leg-loop harness (Rappole and Tipton 1991) constructed of 0.7 mm stretch cord (Stretch Magic®) and crimp tubes (Beadalon). To deploy the leg-loop harnesses efficiently and minimize the handling time of birds, we constructed them before going into the field. In summers 2019 and 2020, we conducted recapture efforts to retrieve prior-year GPS tags, where we placed mist-nets in similar locations (i.e., within 50 m of original capture) to where we previously captured birds. If we tagged a bird the previous year, the GPS logger was removed and stored for future processing and, in most cases, was replaced with another tag. We did not place new GPS tags on birds recaptured in 2020.

Before deployment, we programmed GPS loggers to collect location information at 62 points over a year on the following schedule: one point collected on the night of deployment, one point every seven days from 2 September to 19 September, one point every three days from 22 September to 27 November, one point every seven days from 4 December to 28 February, and one point every three days from 4 March to 6 May. We based data collection times on reasonable periods for fall migration, wintering, and spring migration (English et al. 2017). Like Tonra et al. (2019), we programmed loggers to collect data at 2200 EST when birds were most likely active and foraging and not located under the tree canopy to increase the chance that loggers located satellites.

Home range estimation

We defined the onset of migration as movement > 100 km from a breeding or wintering home range and the onset of the winter stage as > three consecutive points within 1 km of each other. To reduce outliers and inaccurate locations, we discarded data points during the overwinter period with GPS fixes with dilution of precision values > five (Forrest et al. 2022), a measure of positional precision. We followed methods in Calabrese et al. (2016) and Tonra et al. (2019) to determine range residency and generate home range estimates that account for autocorrelation between consecutive locations (R Package ctmm; Calabrese et al. 2016, R Core Team 2021). First, we determined range residency for an individual if a clear asymptote was reached in the variogram, indicating a bounded home range during the study period (Calabrese et al. 2016, Fleming et al. 2019). Next, for all range residents, we used continuous time movement model selection, with the “ctmm.select” function, to generate and rank models (via AICc) where there is no autocorrelation (i.e., independent identically distributed, IID) and when there is autocorrelation among points (i.e., Ornstein-Uhlenbeck [OU], and Ornstein-Uhlenbeck F [OUF]; see Calabrese et al. 2016). Using the best model, we generated kernel density 50% utilization (KD50) and kernel density 95% utilization (KD95) using autocorrelated kernel density estimation (AKDE; Calabrese et al. 2016) and exported shapefiles into ArcGIS. For birds with a bounded home range, we documented winter residence time, defined as the number of days between the arrival at a winter home range and the initiation of spring migration.

Land cover classification

We quantified land cover types from 2019 data using the 100-m Copernicus Global Land Service layer (Buchhorn et al. 2020). According to Buchhorn et al. (2020), we classified land cover types as (a) closed forest, including any land dominated with trees with > 70% canopy cover (CForest), (b) open forest consisting of a tree canopy of 15–70% with shrubs or grasses underneath (OForest), (c) shrubland consisting of land with low (< 5-m tall) woody vegetation (Shrubs), (d) herbaceous vegetation where plants did not have persistent stems and tree canopy was < 10% (HerbVeg), (e) wetland including areas with a permanent mixture of vegetation and water (HerbWet), (f) cultivated and managed lands covered with vegetation and crops (Crops), (g) barren, as land with exposed soil, sand, or rocks (Barren), (h) freshwater or saltwater bodies of water such as rivers, lakes, and oceans (Water), and (i) developed, including land covered by buildings, impervious surfaces, and other manufactured structures (Develop). We quantified land cover within a 5-km and 2-km radius circle and home range around a point centered on the winter home range (i.e., “used”) and a point centered on a randomly available polygon (i.e., “available”) within 35 km of the actual home range location. We generated randomly available locations using the “create random points” tool in ArcMap toolbox (Data Management Tools, Sampling). We used the 35-km measurement because this was the shortest distance between primary and secondary wintering home ranges for Whip-poor-wills that relocated during the wintering season.

Using the geoprocessing features of ArcMap, we quantified land cover at a 5-km scale to examine landscape-level habitat selection. We based this value on the Buler et al. (2007) study that found forest cover at the 5-km scale was a cue for birds when landing at the end of a migratory flight. Next, we repeated land cover quantification at a 2-km scale, representing two times the largest winter home range. Finally, we quantified land cover within each home range (KD95). Because home ranges were small (i.e., < 100-m across), and we noticed some misclassifications of the Copernicus global land cover types at the home range scale based on aerial imagery from 2019 Google Earth Pro 7.3.2 (Google, Mountain View, California; i.e., closed forest was classified as crop cover), we corrected these obvious misclassifications within two home ranges. Because percent shrub cover was significantly negatively correlated with percent closed forest cover at 5-km and 2-km scales (all r_s < -0.72 and p < 0.01), we excluded shrub cover from further analyses. We retained closed forest cover, given that forest cover is positively correlated with Whip-poor-will occupancy on the breeding grounds (Vala et al. 2020), and excluded variables in ≤ two home ranges (i.e., herbaceous wetland, bare, water). At the home range scale, we retained the variables closed forest, open forest, and crops, and dropped herbaceous vegetation and developed land covers because these were present in one and zero home ranges, respectively.

STATISTICAL ANALYSES

Habitat selection

We used a resource selection function (RSF; Johnson et al. 2006, Manly et al. 2002) approach with a conditional logistic regression analysis (R Package survival; Therneau 2021) to examine which land covers were most important in Whip-poor-will habitat use when differentiating between used and available locations. Conditional logistic regression allowed for the pairing of each used plot with the available plot by individuals to account for their non-independence (Duchesne et al. 2010). Because there is little information about habitat selection on the wintering grounds by Whip-poor-wills, we performed an exploratory data analysis that considered all possible combinations of candidate variables. Using Spearman correlation, we assessed land cover variables for collinearity and dropped one of any highly correlated covariates in further analyses (r > 0.70; Dormann et al. 2013). We used automated model selection (R Package MuMIn; Bartón 2020) to identify and rank the best models (ΔAIC_c < 2) for each spatial scale. After selecting the best models, we performed a likelihood ratio test to examine model fit. We assessed predictive ability of best models with concordance (c) where values range between 0.5 (completely random) to 1.0 (perfect discrimination), and values > 0.70 and > 0.80 indicate acceptable and excellent discrimination, respectively (Hosmer and Lemeshow 2000, Austin and Steyerberg 2012, Harrell 2015). Next, we calculated variable importance as the sum of relative weights across all models that contained the variable. Finally, we predicted habitat selection probabilities at each spatial scale using the coefficients of the variable of the highest importance from the best model (Wagner et al. 2017). Furthermore, we calculated means and standard errors for candidate variables to provide a quantitative description of used and available habitats.

We used descriptive statistics to describe differences in land cover between birds with sedentary and mobile movement strategies at the primary home range. Birds that had one home range throughout the winter were considered sedentary (n = 15), and mobile birds were individuals that had two home ranges during the winter period (n = 4). Finally, we provide descriptive statistics of the land cover variables for the four mobile birds that relocated from their first winter home range to a second home range. We did not run statistical analyses because of the low sample size of mobile birds.

RESULTS

GPS tag deployment and collection

In the 2018 and 2019 summers, we deployed 21 and 30 GPS loggers, respectively, on male Eastern Whip-poor-wills. In summers 2019 and 2020, we recaptured 12 and 18 of these birds, respectively, and recovered their GPS tags. All of the twelve males recaptured in 2019 received another logger, and we recaptured ten in 2020 (hereafter, multi-year birds). We did not recapture a bird where the logger had fallen off. Second-year (SY) males had relatively low recapture rates of 27% (3/11) in 2019 and 30% (3/10) in 2020. Conversely, after-second-year (ASY) males had recapture rates of 82% (9/11) and 75% (15/20) in 2019 and 2020, respectively. On two occasions, birds originally captured as SY birds were not recaptured until two years post-banding. We did not document this pattern for any ASY birds, despite banding a larger sample size of that age cohort. We could not retrieve data from two GPS tags placed on males in 2019 even after returning the units to the manufacturer to attempt data retrieval. Across the two years of winter data and for 19 individual birds, the PinPoint GPS tags collected 496 locations, and we retained 451 locations for analyses based on dilution of precision values < five, which provided a 3D location accurate to ~10 m (Tonra et al. 2019, Forrest et al. 2022).

Wintering locations, timing, and site fidelity

Winter home range locations and arrival and departure dates varied substantially for Eastern Whip-poor-wills tagged for this study. Nine birds spent the winter in Mexico, three in Guatemala, two in Honduras, two in the United States, one in Belize, and one in El Salvador (Table 1, Fig. 1). Birds arrived and settled on their winter home range from approximately 1 October to 4 December. Birds with winter home ranges at higher latitudes averaged an earlier arrival and longer winter residence times than birds that settled at lower latitudes that migrated greater distances. Four of 19 (21%) Whip-poor-wills demonstrated mobile wintering behavior and occupied two home ranges on the wintering grounds, moving between 32 km and 73 km (Table 2). Typically, these birds moved to a second winter home range, in a northern direction, between mid- and late February and stayed there for four to six weeks before initiating spring migration. The first (or primary) wintering home range was always where the bird stayed the longest.

For multi-year birds, one GPS tag failed during fall migration (i.e., no winter data). Still, the remaining nine birds demonstrated 100% site fidelity and returned to the same home range occupied the previous year (Fig. 2). Interestingly, one of the multi-year birds with the mobile strategy retained the strategy of moving to the identical second home range for approximately the same amount of time (i.e., ~43 and 41 days). For the other multi-year bird with a mobile strategy, during the second year of data collection, the battery expired prior to the expected date of movement to its second home range.

Home range

We examined variograms to determine range residency for the following 28 instances: nine individuals with two years of data and 10 individuals with one year of data. In all cases except one (i.e., 27 of 28; see Table 1), birds demonstrated a bounded home range during the wintering period, and we used this range in the movement model selection to generate home range sizes. For 17 occasions, there was no evidence for autocorrelation among consecutive locations, and we generated home range estimates using the IID model (Table 1; Calabrese et al. 2016). OUF and OU models were selected as the top movement models on six and four occasions, respectively, indicating positive autocorrelation and were used to generate home range estimates in those instances (Table 1; Calabrese et al. 2016).

For all birds, the median KD95 home range area was 1.95 ha (range: 0.34–54.74 ha), which was nearly four times larger than the core home range size (KD50) that ranged from 0.09 to 8.91 ha, with a median value of 0.53 ha (Table 1). We present median values because the two males that overwintered in the United States had substantially larger home ranges than other birds (i.e., KD95 = 54.74 ha and 18.51 ha). Interestingly, the much smaller KD95 (3.84 ha) home range of the male in South Carolina during the second year of data collection for this bird (2019–2020) was almost entirely contained within the core home range from the prior year (Fig. 2).

Habitat selection

At the 5-km scale, model selection identified the CLR model containing crops, open forest, and closed forest as the best model (Table 3), with crops having a sum of weights of 0.97. The odds-ratio estimate for crops is 0.72 (Table 4), indicating avoidance of this land cover at the 5-km scale, and available locations had greater amounts of crops than used locations (Table 5). In particular, the probability of use drops dramatically once the percent of crops is > 25% of the 5-km landscape (Fig. 3a). Two additional models containing the variables crops and closed forest were also ranked within ΔAIC_c <2 (Table 3). The top model fit the data well (LRT χ² = 9.04, df = 1, p = 0.002) and had high predictive ability (c = 0.86). Variables ranked by importance were crops, closed forest, open forest, developed, and herbaceous vegetation, and all top models fit the data well (LRT χ² > 9.04, p < 0.03; Table 3).

Model selection at the 2-km scale identified three CLR models with ΔAIC_c <2 and included crops, herbaceous vegetation, and open forest (Table 3). All models fit the data well (all LRT χ² > 6.30, p < 0.002), and the third-best model containing crops and open forest together had the highest predictive ability (c = 0.78). Again, crops had the highest importance (sum of weights = 0.71), followed by open forest, closed forest, herbaceous vegetation, and developed land covers. The odds ratio for percent crops was 0.93 (Table 4), indicating a lack of strong avoidance or preference. Predicted probability of use showed a negative relationship between use and percent crops in the 2-km landscape (Fig. 3b). More crop cover was in available compared to used locations (Table 5).

At the home range scale, four models were included in the best set with ΔAIC_c <2 (Table 3). All models fit the data well (all LRT χ² > 4.44, p < 0.04); variables ranked by importance in decreasing order were open forest (0.73), closed forest, and crops. The model containing all three variables had the greatest predictive ability (c = 0.76). Open forest and closed forest land covers were greater and crops were lower in used than in available locations (Table 5). The 95% CI of the odds ratio for open forest is above one (Table 4), indicating a preference for this land cover type within the home range.

Primary home ranges of sedentary birds had two times greater mean closed forest and 3.7 times less open forest than birds that moved to a second home range (Fig. 4; Table 5). Birds that were sedentary also had crop cover in their home range, but mobile birds had little to no crop cover in their first or second home range. The largest difference between the first and second home range for mobile birds was a 29% reduction in open forest cover and a 30% increase in closed forest cover (Table 5).

DISCUSSION

Several variables emerged as consistent predictors of habitat selection by Eastern Whip-poor-wills on the wintering grounds. We found that Eastern Whip-poor-wills selected for open forest and closed forest habitats at the home range scale and avoided crop cover at the landscape scale. These results are similar to the findings of Tonra et al. (2019) that Eastern Whip-poor-wills wintering home ranges are strongly associated with forest habitat. However, they did not differentiate between different types of forest. Additionally, birds in this study used a variety of landscapes ranging from peri-urban (e.g., Houston outskirts) to agricultural (e.g., mixed coffee, corn, etc.) to highly forested (e.g., Parque Nacional Sierra del Lacandón in Honduras). Similarly, Ng et al. (2018) found that Common Nighthawks (Chordeiles minor) varied in their wintering habitat selection where some birds used landscapes with a mix of land uses (e.g., forest, cropland, and shrubland) and others selected predominantly forest habitat. In our study, open forest cover at the home range and 2-km scale was 129% and 41% greater, respectively, than at available locations. Although some landscapes used by Whip-poor-wills had high crop cover, crop cover averaged 2.0, 2.7, and 3.7 times greater in available plots than used plots across the largest to smallest spatial scale.

Similar to other studies, we found that a small percentage (4/19) of Eastern Whip-poor-wills moved to a second home range late in the winter period and prior to migration. Tonra et al. (2019) also documented 3/11 Eastern Whip-poor-wills moving to a second wintering area in early February with each bird spending a minimum of 24 days at the site. For Common Nighthawks, Ng et al. (2018) found 1/6 birds wintering in Brazil established a second roosting home range on the winter grounds, and Knight et al. (2021) documented 7/52 birds relocated to a second winter area. This intra-seasonal movement on the wintering grounds is not unique to Nightjars. Purple Martins (Progne subis) have been documented making large-scale movements to different roost locations within the Amazon Forest in Brazil (Fraser et al. 2017). To evaluate a possible cause of this pattern, we examined if forest cover differed between first and second home ranges for mobile birds. Interestingly, their second home range had less open forest cover and higher closed forest cover than their first home range. In fact, mobile birds moved to a second home range that had a similarly high value of closed forest cover as their sedentary counterparts (65%). Movements on the wintering grounds did not resemble transience, or the continual movement among habitats, but directed movement to another location, where they remained for an extended period. Indeed, a bird with two years of winter data demonstrated the same mobile strategy, with site fidelity and similar timing of movements in both years. One caveat of our and similar studies is the low sample size, especially for birds that had two home ranges (English et al. 2017, Tonra et al. 2019), and thus a more extensive study of the role of forest cover and the resources they provide during the winter period is warranted.

Given that Eastern Whip-poor-will winter home ranges are mainly composed of open and closed forests, reducing forest cover on the wintering grounds could contribute to population declines. This is particularly important given that the documented wintering areas for Whip-poor-wills lie in a region with some of the highest forest conversion rates to agriculture (Jenkins 2003, Song et al. 2018). Although not a replacement for primary forest, agricultural landscapes that provide tree and canopy cover can provide winter habitat, as an open forest matrix, for migratory birds and should be considered in conservation and management strategies (Bennett et al. 2018, Şekercioğlu et al. 2019, Ritterson et al. 2021). However, prior research on these landscapes has focused on diurnal migratory birds. There is a lack of data on the role and use of agricultural landscapes for Nightjars, given that these species are nocturnal and often quiet, making them hard to detect during the winter period. Although agriculture may provide open spaces for aerial insectivores to forage, agricultural intensification and pesticide use have been identified as drivers of insect population declines (Spiller and Dettmers 2019) and may contribute to aerial insectivore population declines (Nebel et al. 2010, English et al. 2018).

We found a lack of migratory connectivity for Eastern Whip-poor-wills that breed across Massachusetts because birds exemplified a wintering distribution that overlapped most of the species’ wintering range (Cink et al. 2020). Although some winter home ranges were located in mountainous terrain (e.g., > 2000 m in the Sierra Madre de Chiapas in Guatemala), others were in coastal plains (e.g., 5 m elevation in South Carolina). Consistent with our results, breeding Whip-poor-wills from the Midwest United States (Tonra et al. 2019, Skinner et al. 2022) and central Canada (English et al. 2017) were documented with a broad range of winter locations from Texas to as far south as Costa Rica. However, Eastern Whip-poor-wills have rarely been documented wintering as far north as coastal South Carolina (Post and Gauthreaux 1989, as cited in Cink et al. 2020). However, La Sorte and Thompson (2007) found a northern boundary expansion for Eastern Whip-poor-will wintering in North America when analyzing CBC data from 1975 to 2004. Cink et al. (2020) noted that more research is needed to examine whether some individuals winter farther north in mild winters. Although our results of 100% site fidelity on the wintering grounds suggest that adults do not modify their wintering areas annually, juveniles may be more flexible in their wintering site selection.

All Eastern Whip-poor-wills demonstrated bounded home ranges on the wintering grounds. The two birds that overwintered ≥ 30° N (i.e., Texas and South Carolina) had 95KD home ranges that were ten times larger than home ranges of birds ≤ 20° N (i.e., southern Mexico through Central America). Home range size on the wintering grounds may vary by resource availability, predation pressure, and intra- and interspecific competition (Brown and Sherry 2008), and these factors also may vary by latitude. Low latitudes with low seasonality should support higher and more consistent food abundance through the wintering period, especially for an aerial insectivore, compared to highly seasonal temperate environments at higher latitudes (Smith et al. 2010). In addition, winter home range size can vary across years because birds may modify behaviors and space use to accommodate for environmental (e.g., climate) and biological (e.g., competitors) factors (Miller et al. 2017, Ruiz-Sánchez et al. 2017). For example, the Whip-poor-will that overwintered in South Carolina had a 12x smaller home range in the second year of the study, one that coincided with the core (50KD) from the previous year. More inter- and intraspecific studies are needed to examine the role and interactions of mechanisms driving patterns of variation in winter home range sizes with contrasting contexts by latitude.

Recapture rates of ASY males were high (i.e., 75–82%) and indicate a high apparent annual survival of adult birds. Our value corresponded to the 77% return rates for adults in Kansas (Cink et al. 2020). The Eastern Whip-poor-will appears to be a relatively long-lived species, with two records of birds recaptured > 13 years after initial capture (Cink et al. 2020), but more research is needed on longevity of the species because this may influence movement and settlement patterns of younger birds. For example, males tagged as SY birds had 2.5–3 times lower recapture rates. We suspect the lower recapture rates of SY males result from a lack of site fidelity on the breeding grounds because young males may not have established a territory. However, they also may have a somewhat lower survival rate, especially on the wintering ground (Ritterson et al. 2021). Understanding how events on the breeding and wintering grounds interact during these young life stages may be critical in understanding population recruitment of the species. Although this study only examines males, we initially intended to collect data on both male and female birds. However, female Whip-poor-wills were more challenging to capture and recapture with the playback techniques. Of the six females we placed GPS tags on in 2018, we only recaptured one of these birds (in 2020), and we could not retrieve data from the GPS unit. Another female captured in two different years never received a GPS tag. Compared to males, females may exhibit different strategies and behaviors on the wintering grounds, as found in other species (Komar et al. 2005). Research on female Eastern Whip-poor-wills is needed on their overwintering range, movements, habitat selection, and survival to determine how that contributes to population dynamics (Cink et al. 2020).

Our study documented serial residency (Cresswell 2014) with 100% winter site fidelity for Eastern Whip-poor-wills with two years of overwintering locations (n = 9). Skinner et al. (2022) document one multi-year Whip-poor-will, banded in southern Ohio, that returned to the same winter home range in eastern Guatemala. Although English et al. (2017) reported three Eastern Whip-poor-will males that appeared to return to the same winter location, that study used light-level geolocators that lacked the precision to pinpoint locations. All home range locations of birds overlapped in the two years of data, most often with the 50KD also overlapping. One caveat of our study is that all birds were adults (i.e., SY or ASY) when they were first tagged on the breeding grounds. We do not have multiple years of overwintering data starting with their first winter period and cannot say if first-year birds are faithful to a first overwinter site or even occupy a bounded home range. However, Evens et al. (2017) reported a European Nightjar (Camprimulgus europaeus) juvenile returned to the same winter location as a SY bird.

Overwinter site fidelity is common among migratory birds and is a more accurate measure of annual survival given the lack of need for mate acquisition on the breeding grounds (see Blackburn and Cressswell 2016). Benefits of site fidelity for overwintering birds include familiarity of resources and changes in seasonal conditions, reduced food search costs, and increased foraging efficiency and predator avoidance (Warkentin and Hernández 1996, Latta and Faaborg 2001). However, site fidelity of forest habitats combined with the continued loss of forest in the Neotropics can place birds at high risk in rapidly changing landscapes (Koronkiewicz et al. 2006, Monroy-Ojeda et al. 2013, Cresswell 2014). For example, the decline of the Appalachian population of Golden-winged Warblers has been linked to forest loss on the wintering grounds of Columbia and Venezuela (Kramer et al. 2018). Although this study provides critical information to guide conservation planning efforts, continued research is needed to examine impacts of land cover types on survival and body condition of overwintering Eastern Whip-poor-wills, especially in increasingly agricultural (e.g., Veracruz, Mexico) and urban (e.g., Houston outskirts) landscapes.

LITERATURE CITED

Albert, S., J. D. Wolfe, J. Kellerman, T. Sherry, B. M. Stuctchbury, N. J. Bayly, and A. Ruiz-Sánchez. 2020. Habitat ecology of Nearctic-Neotropical migratory landbirds on the nonbreeding grounds. Ornithological Applications 122(4). https://doi.org/10.1093/condor/duaa055

Austin, P. C., and E. W. Steyerberg. 2012. Interpreting the concordance statistic of a logistic regression model: relation to the variance and odds ratio of a continuous explanatory variable. BMC Medical Research Methodology 12:82. https://doi.org/10.1186/1471-2288-12-82

Bartón, K. 2020. MuMIn: Multi-model inference. R package version. 1.43.17. https://CRAN.R-project.org/package=MuMIn

Bennett, R. E., W. Leuenberger, B. B. Bosarreyes Leja, A. Sagone Cáceres, K. Johnson, and J. Larkin. 2018. Conservation of Neotropical migratory birds in tropical hardwood and oil palm plantations. PLoS One 13(12):e0210293. https://doi.org/10.1371/journal.pone.0210293

BirdLife International. 2018. Eastern Whip-poor-will Antrostomus vociferus. The IUCN Red List of Threatened Species. International Union for Conservation of Nature and Natural Resources, Cambridge, UK. https://doi.org/10.2305/IUCN.UK.2018-2.RLTS.T22736393A131617918.en

Blackburn, E., and W. Cresswell. 2016. High winter site fidelity in a long-distance migrant: implications for wintering ecology and survival estimates. Journal of Ornithology 157:93-108. https://doi.org/10.1007/s10336-015-1252-z

Brown, D. R., and T. W. Sherry. 2008. Alternative strategies of space use and response to resource change in a wintering migrant songbird. Behavioral Ecology 19(6):1314-1325. https://doi.org/10.1093/beheco/arn073

Buchhorn, M., M. Lesiv, N. E. Tsendbazar, M. Herold, L. Bertels, and B. Smets. 2020. Copernicus global land cover layers—collection 2. Remote Sensing 12:1044. https://doi.org/10.3390/rs12061044

Buler, J. J., F. R. Moore, and S. Woltmann. 2007. A multi-scale examination of stopover habitat use by birds. Ecology 88(7):1789-1802. https://doi.org/10.1890/06-1871.1

Calabrese, J. M., C. H. Fleming, and E. Gurarie. 2016. ctmm: An R package for analyzing animal relocation data as a continuous-time stochastic process. Methods in Ecology and Evolution 7(9):1124-1132. https://doi.org/10.1111/2041-210X.12559

Cink, C., P. Pyle, and M. Patten. 2020. Eastern Whip-poor-will (Antrostomus vociferus). In P. G. Rodewald, editor. Birds of the World. Cornell Lab of Ornithology, Ithaca, New York, USA. https://doi.org/10.2173/bow.whip-p1.01

Cresswell, W. 2014. Migratory connectivity of Palaearctic-African migratory birds and their responses to environmental change: the serial residency hypothesis. Ibis 156(3):493-510. https://doi.org/10.1111/ibi.12168

Dormann, C. F., J. Elith, S. Bacher, C. Buchmann, G. Carl, G. Carré, J. R. G. Marquéz, B. Gruber, B. Lafourcade, P. J. Leitão, T. Münkemüller, C. McClean, P. E. Osborne, B. Reineking, B. Schröder, A. K. Skidmore, D. Zurell, and S. Lautenbach. 2013. Collinearity: a review of methods to deal with it and a simulation study evaluating their performance. Ecography 36(1):27-46. https://doi.org/10.1111/j.1600-0587.2012.07348.x

Duchesne, T., D. Fortin, and N. Courbin. 2010. Mixed conditional logistic regression for habitat selection studies. Journal of Animal Ecology 79(3):548-555. https://doi.org/10.1111/j.1365-2656.2010.01670.x

English, P. A., D. J. Green, and J. J. Nocera. 2018. Stable isotopes from museum specimens may provide evidence of long-term change in the trophic ecology of a migratory aerial insectivore. Frontiers in Ecology and Evolution 6:14. https://doi.org/10.3389/fevo.2018.00014

English, P. A., A. M. Mills, M. D. Cadman, A. E. Heagy, G. J. Rand, D. J. Green, and J. J. Nocera. 2017. Tracking the migration of a nocturnal aerial insectivore in the Americas. BMC Zoology 2:5. https://doi.org/10.1186/s40850-017-0014-1

Evens, R., N. Beenaerts, N. Witters, and T. Artois. 2017. Repeated migration of a juvenile European Nightjar Camprimulgus europaeus. Journal of Ornithology 158:881-886. https://doi.org/10.1007/s10336-017-1459-2

Fleming, C. H., M. J. Noonan, E. P. Medici, and J. M. Calabrese. 2019. Overcoming the challenge of small effective sample sizes in home-range estimation. Methods in Ecology and Evolution 10(10):1679-1689. https://doi.org/10.1111/2041-210X.13270

Forrest, S. W., M. R. Recio, and P. J. Seddon. 2022. Moving wildlife tracking forward under forested conditions with the SWIFT GPS algorithm. Animal Biotelemetry 10:19. https://doi.org/10.1186/s40317-022-00289-9

Fraser, K. C., A. Shave, A. Savage, A. Ritchie, K. Bell, J. Siegrist, J. D. Ray, K. Applegate, and M. Pearman. 2017. Determining fine-scale migratory connectivity and habitat selection for a migratory songbird by using new GPS technology. Journal of Avian Biology 48(3):339-345. https://doi.org/10.1111/jav.01091

Gibbs, H. K., A. S. Ruesch, F. Achard, M. K. Clayton, P. Holmgren, N. Ramankutty, and J. A. Foley. 2010. Tropical forests were the primary sources of new agricultural land in the 1980s and 1990s. PNAS 107(38):16732-16737. https://doi.org/10.1073/pnas.0910275107

Harrell Jr, F. E. 2015. Regression modeling strategies: with applications to linear models, logistic and ordinal regression, and survival analysis. Springer, New York, New York, USA.

Harrison, X. A., J. D. Blount, R. Inger, D. R. Norris, and S. Bearhop. 2011. Carry-over effects as drivers of fitness differences in animals. Journal of Animal Ecology 80(1):4-18. https://doi.org/10.1111/j.1365-2656.2010.01740.x

Hosmer, D. W., and S. Lemeshow. 2000. Applied logistic regression. John Wiley and Sons, New York, New York, USA.

Jackson, H. B., and L. Fahrig. 2015. Are ecologists conducting research at the optimal scale? Global Ecology and Biogeography 24(1):52-63. https://doi.org/10.1111/geb.12233

Jenkins, M. 2003. Prospects for biodiversity. Science 302(5648):1175-1177. https://doi.org/10.1126/science.1088666

Johnson, C. J., S. E. Nielsen, E. H. Merrill, T. L. McDonald, and M. S. Boyce. 2006. Resource selection functions based on use-availability data: theoretical motivation and evaluation methods. Journal of Wildlife Management 70(2):347-357. https://doi.org/10.2193/0022-541X(2006)70[347:RSFBOU]2.0.CO;2

Johnson, D. H. 1980. The comparison of usage and availability measurements for evaluating resource preference. Ecology 61(1):65-71. https://doi.org/10.2307/1937156

Knight, E. C., A.-L. Harrison, A. L. Scarpignato, S. L. Van Wilgenburg, E. M. Bayne, J. W. Ng, E. Angell, R. Bowman, R. M. Brigham, B. Drolet, W. E. Easton, T. R. Forrester, J. T. Foster, S. Haché, K. C. Hannah, K. G. Hick, J. Ibarzabal, T. L. Imlay, S. A. Mackenzie, A. Marsh, L. P. McGuire, G. N. Newberry, D. Newstead, A. Sidler, P. H. Sinclair, J. L. Stephens, D. L. Swanson, J. A. Trembley, and P. P. Marra. 2021. Comprehensive estimation of spatial and temporal migratory connectivity across the annual cycle to direct conservation efforts. Ecography 44(5):655-679. https://doi.org/10.1111/ecog.05111

Kramer, G. R., D. E. Andersen, D. A. Buehler, P. B. Wood, S. M. Peterson, J. A. Lehman, K. R. Aldinger, L. P. Bulluck, S. Harding, J. A. Jones, J. P. Loegering, C. Smalling, R. Vallender, and H. M. Streby. 2018. Population trends in Vermivora warblers are linked to strong migratory connectivity. PNAS 115(14):E3192-E3200. https://doi.org/10.1073/pnas.1718985115

Komar, O. 2006. Priority contribution. Ecology and conservation of birds in coffee plantations: a critical review. Bird Conservation International 16(1):1-23. https://doi.org/10.1017/S0959270906000074

Komar, O., B. J. O’Shea, A. Townsend Peterson, and A. G. Navarro-Sigüenza. 2005. Evidence of latitudinal sexual segregation among migratory birds wintering in Mexico. Auk 122(3):938-948. https://doi.org/10.1642/0004-8038(2005)122[0938:EOLSSA]2.0.CO;2

Koronkiewicz, T. J., M. K. Sogge, C. Van Riper, III, and E. H. Paxton. 2006. Territoriality, site fidelity, and survivorship of willow flycatchers wintering in Costa Rica. Condor 108(3):558-570. https://www.jstor.org/stable/4151077

La Sorte, F. A., and F. R. Thompson, III. 2007. Poleward shifts in winter ranges of North American birds. Ecology 88(7):1803-1812. https://doi.org/10.1890/06-1072.1

Latta, S. C., and J. Faaborg. 2001. Winter site fidelity of prairie warblers in the Dominican Republic. Condor 103(3):455-468. https://doi.org/10.1650/0010-5422(2001)103[0455:WSFOPW]2.0.CO;2

Manly, B. F. L., L. McDonald, D. L. Thomas, T. L. McDonald, and W. P. Erickson. 2002. Resource selection by animals: statistical design and analysis for field studies. Second edition. Kluwer, Boston, Massachusetts, USA.

Marra, P. P., and R. T. Holmes. 2001. Consequences of dominance-mediated habitat segregation in American redstarts during the nonbreeding season. Auk 118(1):92-104. https://doi.org/10.1093/auk/118.1.92

Miller, T. A., R. P. Brooks, M. J. Lanzone, J. Cooper, K. O’Malley, D. Brandes, A. Duerr, and T. E. Katzner. 2017. Summer and winter space use and home range characteristics of Golden Eagles (Aquila chrysaetos) in eastern North America. Condor 119(4):697-719. https://doi.org/10.1650/CONDOR-16-154.1

Monroy-Ojeda, A., M. Grosselet, G. Ruiz, and E. Del Valle. 2013. Winter site fidelity and winter residency of six migratory Neotropical species in Mexico. Wilson Journal of Ornithology 125(1):192-196. https://doi.org/10.1676/12-072.1

Nebel, S., A. Mills, J. McCracken, and P. Taylor. 2010. Declines of aerial insectivores in North America follow a geographic gradient. Avian Conservation and Ecology 5(2):1. https://doi.org/10.5751/ACE-00391-050201

Ng, J., E. Knight, A. Scarpignato, A. Harrison, E. Bayne, and P. Marra. 2018. First full annual cycle tracking of a declining aerial insectivorous bird, the Common Nighthawk (Chordeiles minor), identifies migration routes, nonbreeding habitat, and breeding site fidelity. Canadian Journal of Zoology 96(8):869-875. https://doi.org/10.1139/cjz-2017-0098

Norris, D. R., P. P. Marra, T. K. Kyser, T. W, Sherry, and L. M. Ratcliffe. 2004. Tropical winter habitat limits reproductive success on the temperate breeding grounds in a migratory bird. Proceedings of the Royal Society B, Biological Sciences 271(1534):59-64. https://doi.org/10.1098/rspb.2003.2569

North American Bird Conservation Initiative Canada. 2019. The state of Canada’s birds. Environment and Climate Change Canada, Ottawa, Canada. http://nabci.net/resources/state-of-canadas-birds-2019/

Rappole, J., and A. Tipton. 1991. New harness design for attachment of radio transmitters to small passerines. Journal of Field Ornithology 62(3):335-337. https://www.jstor.org/stable/20065798

R Core Team. 2021. R: A language and environment for statistical computing. Version 4.1.1. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/

Ritterson, J. D., D. I. King, and R. B. Chandler. 2021. Habitat-specific survival of golden-winged warblers Vermivora chrysoptera during the non-breeding season in an agricultural landscape. Journal of Avian Biology 52(3):e02442. https://doi.org/10.1111/jav.02442

Rosenberg, K. V., A. M. Dokter, P. J. Blancher, J. R. Sauer, A. C. Smith, P. A. Smith, J. C. Stanton, A. Panjabi, L. Helft, M. Parr, and P. Marra. 2019. Decline of the North American avifauna. Science 366(6461):120-124. https://doi.org/10.1126/science.aaw1313

Rosenberg, K. V., J. A. Kennedy, R. Dettmers, R. P. Ford, D. Reynolds, J. D. Alexander, C. J. Beardmore, P. J. Blancher, R. E. Bogart, G. S. Butcher, A. F. Camfield, A. Couturier, D. W. Demarest, W. E. Easton, J. J. Giocomo, R. H. Keller, A. E. Mini, A. O. Panjabi, D. N. Pashley, T. D. Rich, J. M. Ruth, H. Stabins, J. Stanton, and T. Will. 2016. Partners in Flight landbird conservation plan: 2016 revision for Canada and continental United States. Partners in Flight Science Committee, McAllen, Texas, USA. https://www.partnersinflight.org/resources/the-plan/

Ruiz-Sánchez, A., K. Renton, and R. Rueda-Hernández. 2017. Winter habitat disturbance influences density and territory size of a Neotropical migratory warbler. Journal of Ornithology 158:63-73. https://doi.org/10.1007/s10336-016-1368-9

Rushing, C. S., J. A. Hostetler, T. S. Sillett, P. P. Marra, J. A. Rotenberg, and T. B. Ryder. 2017. Spatial and temporal drivers of avian population dynamics across the annual cycle. Ecology 98(11):2837-2850. https://doi.org/10.1002/ecy.1967

Sauer, J. R., W. A. Link, and J. E. Hines. 2020. The North American breeding bird survey, analysis results 1966–2019. U.S. Geological Survey data release. https://doi.org/10.5066/P96A7675

Sauer, J. R., K. L. Pardieck, D. J. Ziolkowski, A. C. Smith, M. R. Hudson, V. Rodriguez, H. Berlanga, D. K. Niven, and W. A. Link. 2017. The first 50 years of the North American breeding bird survey. Condor 119(3):576-593. https://doi.org/10.1650/CONDOR-17-83.1

Şekercioğlu, Ç. H., C. D. Mendenhall, F. Oviedo-Brenes, J. J. Horns, P. R. Ehrlich, and G. C. Daily. 2019. Long-term declines in bird populations in tropical agricultural countryside. PNAS 116(20):9903-9912. https://doi.org/10.1073/pnas.1802732116

Sherry, T. W., and R. T. Holmes. 1996. Winter habitat quality, population limitation, and conservation of Neotropical-Nearctic migrant birds. Ecology 77(1):36-48. https://doi.org/10.2307/2265652

Skinner, A., M. P. Ward, I. Souza-Cole, J. R. Wright, F. R. Thompson III, T. J. Benson, S. N. Matthews, and C. M. Tonra. 2022. High spatiotemporal overlap in the non-breeding season despite geographically dispersed breeding locations in the Eastern Whip-poor-will (Antrostomus vociferus). Diversity and Distributions 28(4):712-726. https://doi.org/10.1111/ddi.13477

Smith, A., M. Hudson, C. Downes, and C. Francis. 2015. Change points in the population trends of aerial-insectivorous birds in North America: synchronized in time across species and regions. PLoS ONE 10(7):e0130768. https://doi.org/10.1371/journal.pone.0130768

Smith, J. A. M., L. R. Reitsma, and P. P. Marra. 2010. Moisture as a determinant of habitat quality for a nonbreeding Neotropical migratory songbird. Ecology 91(10):2874-2882. https://doi.org/10.1890/09-2212.1

Smith, J. A. M., L. R. Reitsma, and P. P. Marra. 2011. Multiple space-use strategies and their divergent consequences in a nonbreeding migratory bird (Parkesia noveboracensis). Auk 128:53-60. https://doi.org/10.1525/auk.2011.10241

Song, X., M. C. Hansen, S. V. Stehman, P. V. Potapov, A. Tyukavina, E. F. Vermote, and J. R. Townshend. 2018. Global land change from 1982 to 2016. Nature 560:639-643. https://doi.org/10.1038/s41586-018-0411-9

Spiller, K. J., and R. Dettmers. 2019. Evidence for multiple drivers of aerial insectivore declines in North America. Condor 121:duz010. https://doi.org/10.1093/condor/duz010

Taylor, C. M., and B. J. M. Stutchbury. 2016. Effects of breeding versus winter habitat loss and fragmentation on the population dynamics of a migratory songbird. Ecological Applications 26(2):424-437. https://doi.org/10.1890/14-1410

Therneau, T. M. 2021. survival: Survival analysis. R package version 3.2-13. https://CRAN.R-project.org/package=survival

Tonra, C. M., J. R. Wright, and S. N. Matthews. 2019. Remote estimation of overwintering home ranges in an elusive, migratory nocturnal bird. Ecology and Evolution 9(22):12586-12599. https://doi.org/10.1002/ece3.5723

Vala, M. A., G. W. Mitchell, K. C. Hannah, J. Put, and S. Wilson. 2020. The effects of landscape composition and configuration on Eastern Whip-poor-will (Caprimulgus vociferous) and Common Nighthawk (Chordeiles minor) occupancy in an agroecosystem. Avian Conservation and Ecology 15(1):24. https://doi.org/10.5751/ACE-01613-150124

Wagner, R. O., M. B. Connior, C. A. Melder, B. S. Cooper, D. Hightower, and S. Pearce. 2017. Movement and resource selection of Baird’s pocket gopher within a longleaf pine ecosystem. Southeastern Naturalist 16(3):397-410. https://doi.org/10.1656/058.016.0307

Warkentin, I. G. and D. Hernández. 1996. The conservation implications of site fidelity: a case study involving nearctic-neotropical migrant songbirds wintering in a Costa Rican mangrove. Biological Conservation 77(2-3):143-150. https://doi.org/10.1016/0006-3207(95)00146-8