Northeastern Naturalist 236 M. Blyth, C.W.E. Paine, and C.E.T. Paine 22001199 NORTHEASTERN NATURALIST 2V6(o2l). :2263,6 N–2o5. 02 The Decline in Nesting Success of Tachycineta bicolor (Tree Swallow) Over 38 Years on Hardwood Island, Maine Marcia Blyth1, C.W. Eliot Paine2, and C.E. Timothy Paine3,* Abstract - The abundance of Tachycineta bicolor (Tree Swallow) has fallen by half across North America over the past 50 years. This study presents 38 years of observations on their nesting success from coastal Maine. We document long-term declines in nest-box occupancy and fledging success. We show that nest-box occupancy was affected by proximity to other nest boxes and to buildings, and that it increased with time since mowing. The number of young fledged per nest, on the other hand, decreased in wet years and years with many cold days, and it increased with time since mowing. These local factors do not, however, explain the long-term decline in nesting success, which we tentatively attribute to anthropogenic effects on the wintering grounds or along the migration route. Introduction In the spring, Tachycineta bicolor Vieillot (Tree Swallow) migrate north throughout North America as they seek sites to breed and raise their young (Knight et al. 2018). Tree Swallows are specialized aerial insectivores, primarily consuming Diptera, though other insect orders are also taken (Bellavance et al. 2018). They are cavity nesters and readily use artificial nest boxes, which has facilitated studies on their behavior and population dynamics (Robertson and Rendell 1990). Female Tree Swallows construct grass-cup nests within nest boxes and typically lay 4–6 eggs, 1 per day. Incubation begins when the penultimate egg is laid, and eggs usually hatch 14–16 d after they were laid, with nestlings fledging 18–22 d later (Hussell and Quinney 1985, Nooker et al. 2005). Tree Swallows are generally regarded as monogamous, although extra-pair paternity is frequent (Lifjeld et al. 1993). Tree Swallows typically raise a single brood per year. After the breeding season, they gather in communal roosts near their breeding grounds prior to their mid-summer migration to the southern US, Mexico, and Central America (Knight et al. 2018). Across North America, the abundance of Tree Swallows has fallen by 49% since 1966 (Sauer et al. 2017). Regionally, their population trends vary, with increases in the Pacific northwest, and particularly sharp declines in New England and maritime Canada (Michel et al. 2016). At a continental scale, this decline has occurred concomitant with habitat destruction on their wintering grounds in Central America (Robbins et al. 1989) and agricultural intensification on their breeding grounds (Ghilain and Bélisle 2008, Stanton et al. 2017). At local scales, fluctuating weather conditions, habitat features, and anthropogenic activity affect nesting success 1Biological and Environmental Sciences, University of Stirling, Stirling FK9 4LA, UK. 254 Birch Bay Drive, Bar Harbor, ME 04609. 3Environmental and Rural Science, University of New England, Armidale, New South Wales 2350, Australia. *Corresponding author - timothy.paine@une.edu.au. Manuscript Editor: Heather York Northeastern Naturalist Vol. 26, No. 2 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 237 (Rioux Paquette et al. 2014, Weegman et al. 2017), which may also affect their overall abundance. Many previous studies of Tree Swallow nesting success have focused on single factors (Ghilain and Bélisle 2008, McArthur et al. 2017, Winkler et al. 2013) or examined them for just 1 or a few breeding seasons (Coe et al. 2015, Male et al. 2006, McCarty and Winkler 1999a). Here, we present 38 years of observations on Tree Swallow breeding, over which time nesting success declined steeply, even though the site experienced only minor anthropogenic perturbation. We monitored nest-box occupancy, whether nestlings fledged from each nest, and the number of fledglings per nest. We also explored local weather, landscape features, and mowing as explanations for the observed decline in nesting success of Tree Swallows. Weather on the breeding ground may affect nesting success of Tree Swallows, both directly and through indirect effects on food supply (Weegman et al. 2017). Brief bouts of cold temperature cause egg temperatures to fall and the metabolic demands of nestlings to increase (Weegman et al. 2017), with long-lasting negative implications for their growth and post-fledging survival (Ardia et al. 2010, McCarty 2001). Cold weather can also reduce the foraging efficiency of adults and their propensity to incubate (Coe et al. 2015, Dunn et al. 2011). These effects may lead females to reduce clutch sizes or adjust the timing of egg laying (Bourret et al. 2015, Dunn et al. 2000). Cold weather also indirectly affects nesting success by reducing the abundance of flying insects (Winkler et al. 2013). Decreased insect abundance has been associated with reduced body condition of female Tree Swallows, causing them to produce smaller clutch-sizes and fledge fewer young (McCarty and Winkler 1999a, Pérez et al. 2008), which in turn have poor body condition (Nooker et al. 2005). We therefore expected that an increased frequency of cold days would be associated with decreased nest-box occupancy, the frequency of fledging, and the number of fledglings per nest. Hereafter, we refer to these 3 response variables collectively as nesting success. In dry areas, Tree Swallows can be positively influenced by moderate precipitation, as it promotes the dispersal and flight of insects (McArthur et al. 2017). However, more substantial precipitation can reduce nesting survival and fledging success of Tree Swallows (Weegman et al. 2017). Temperature and precipitation may covary; thus, we further investigated the degree to which they interact to affect nesting success. The choice of nest site, including the proximity of nests to landscape features, can impact reproductive success (Forsman et al. 2008). Increasing the distance between nest boxes may be positively associated with nest-box occupancy because Tree Swallows can compete with each other for food, be highly defensive of their nests, and chase nearby conspecifics (Hussell 2012, Robillard et al. 2013). However, Tree Swallows probably also benefit from the presence of conspecifics, in part, due to increased opportunities for extra-pair fertilization and a better capacity to detect and deter nest predators. Thus, the effects of nest-box proximity on nesting success are not known (Male et al. 2006). Owing to anthropogenic noise and disturbance, we hypothesized that nesting success would be reduced by nearby Northeastern Naturalist 238 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 Vol. 26, No. 2 buildings and active paths (MacGregor-Fors et al. 2010). Human activity is predicted to negatively impact nesting success by interrupting feeding and causing birds to expend greater energy in flight and vigilance (Miller et al. 2003). Finally, habitat characteristics, including prey, parasite, and predator abundance, may affect nesting success. The occasional removal of vegetation by mowing at our study site may negatively affect nesting success of Tree Swallows (Ghilain and Bélisle 2008) via a reduction in insect abundance (Rioux Paquette et al. 2014, Robillard et al. 2013). Mowing also changes vegetative composition of the landscape from a more structurally complex mixture of shrubs, forbs, and grasses to a relatively uniform grassy composition, which may affect the selection of prey available to foraging Tree Swallows. We therefore expected that the number of nesting seasons since the last mowing of the meadow would affect nesting success. Field-site Description Hardwood Island (44º18'N, 68.26'W) lies in Blue Hill Bay, ME, ~1.4 km from Mount Desert Island and Acadia National Park, and ~6.0 km from the mainland. The 78-ha, peanut-shaped island is mostly covered with Picea (spruce) forest and has a 7.4-ha meadow on the eastern shore. To prevent shrub encroachment, the meadow was mowed with a brush-hog mower after the Tree Swallow breeding seasons in 1981, 1985, 1989, 2007, and 2016. Tree Swallows arrive on the coast of Maine in mid-April and finish building nests by the end of May (Dunn and Winkler 1999, EBird 2017). Nestlings typically fledge by mid-July, and they begin their southward migration a few weeks later . In 1981, prior to the breeding season, we established 7 nest boxes (floor: 10 cm x 10 cm, back: 20 cm tall) with angled, overhanging roofs, which we placed on 1.5-m metal poles. We established more boxes in subsequent years, reaching a maximum of 22 in 1994. They were haphazardly arranged, with boxes separated by between 9 m and 46 m. There are 2 houses on the island, located at the western edge of the meadow. Nest boxes were located between 17 m and 124 m from the houses and between 11 m and 140 m from a gravel path connecting the houses with a dock on the eastern shore. We utilized a Garmin GPSMap 62s GPS (Garmin Ltd., Olathe, KS) to map all nest-box locations. We used Google Earth to calculate distances between the nest boxes, and from the nest boxes to the nearest building and to the main path. Hardwood Island is ideal for the study of Tree Swallows, as their typical foraging range of 400 m means that they rarely leave the island during the breeding season (McCarty and Winkler 1999b). Hardwood Island is generally free of predators, but Procyon lotor (L.) (Raccoon) accessed nest boxes in 1985, 1989, 1990, and 1991, disrupting nesting and killing nestlings. Thereafter, predator guards were installed on the nest boxes, and Raccoons were removed from the island. Methods We assessed 3 response variables: the percentage of nest boxes occupied, the probability that at least 1 nestling fledged from a nest, and the number of nestlings Northeastern Naturalist Vol. 26, No. 2 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 239 that fledged per nest. We calculated nest occupancy as the number of boxes in which nests were built. The number of nest boxes available varied from 7 to 22 among years, as some boxes were damaged over winter. We inferred the probability that at least 1 nestling fledged from a nest by the presence of nestling fecal matter in an intact nest at a post-breeding census. At these censuses, we also noted boxes that showed signs of predation or nest abandonment, such as broken eggshells or dead nestlings. In 13 of the 38 years, we made additional observations during the breeding season, during which we recorded how many birds fledged from each nest. In 10 additional years, although we did not preserve detailed records of the number of fledglings per box, we have records of the total number of fledglings across all nests. To assess the effects of weather on nest-box occupancy and fledgling success, we obtained data from the weather stations at Acadia National Park and at Ellsworth, ME, using NOAA’s online climate data (https://www.ncdc.noaa.gov/cdo-web/). These stations are within 20 km of the study site and, together, provided a complete weather record spanning the duration of the study. We analyzed the effect of temperature on nesting success by counting the number of cold days, which we defined as days with a maximum temperature ≤18.5 °C. This threshold is a critical temperature for insect-flight activity, as temperatures below this can result in a flying-insect abundance of 0 (Winkler et al. 2013). We experimented with the threshold used to define a cold day, varying it between 15.0 ºC and 25.0 ºC, and found that 18.5 ºC provided the best discrimination among years, in agreement with Winkler et al. (2013). To assess the effects of weather on nest occupancy, we used data from the 45-d period from 16 April to 31 May. For analyses of fledging, we used the 45-d period from 1 June to 15 July (Hussell and Quinney 1985). Statistical analyses We used generalized mixed-effect models to assess temporal changes in nesting success, as well as the effects of weather, proximity of landscape features, and time since mowing. We predicted the probabilities of nest occupancy and fledging at least 1 nestling per box as binomial (yes–no) processes with a logit link, whereas we considered the number of fledlings to be a Poisson (counting) process with a log link. We used mixed-effect models to account for variation in nesting success among nest boxes that derived from their idiosyncratic differences, such as orientation, age, or odor. Thus, the temporal models, the weather models, and the time-since-mowing model all included nest box as a random effect. In contrast, in the proximity model, we included year as a random effect to account for the variation in nesting success among years. Prior to analysis, we detrended the temporally varying predictor variables (weather and time since mowing) by taking the residuals after regressing them against year. This procedure focused the analysis on the predictors themselves and reduced the possibly confounding element of time. In the weather models, we investigated the statistical support for an interaction between precipitation and the number of cold days. We evaluated the support for these interactions by comparing models on an Akaike’s information criterion (AIC) basis (Burnham and Anderson 2002). For the landscape-features models, we Northeastern Naturalist 240 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 Vol. 26, No. 2 did not include interactions between the distances to the nearest nest box, nearest building, or main path, as there were not enough degrees of freedom to do so. We log-transformed these distances to improve model fits, and because the effects of competition and disturbance decline rapidly with distance (Canham et al. 2004). All predictors were centered and standardized to a standard deviation of 1 to facilitate comparisons of effect size (Schielzeth 2010). We assessed multi-collinearity using the variance-inflation factor and consistently found it to be absent, indicating that predictor variables were independent of one another. We assessed the models with binomially distributed errors for over-dispersion, but it was never significant. We performed analyses in R 3.5.0 (R Core Team 2018) using the lme4 library (Bates et al. 2012). Results Long-term trends Nest-box occupancy significantly decreased over the 38-y study. Between 1981 and 2018, the predicted occupancy fell from 98.5% to 28.7%, with the steepest decline occurring after 2000 (Fig. 1A). Nests were built in none of the 11 boxes available in 2017. Averaging across all years, 65% of nests fledged young. The frequency of fledging varied widely among years but showed no trend through time (Fig. 1B). The number of fledglings per nest decreased from an average of 3.04 in 1981 to 0.90 in 2018 (Fig. 1C). Very low nesting-success in 1985, 1989, 1990, and 1991 can be attributed to predation by Raccoons, which managed to reach into nest boxes and perturb nesting pairs. In 1989, for example, we found a total of 35 dead nestlings at the post-breeding census. Nevertheless, sporadic predation does not explain long-term declines in nesting success. Weather Nest-box occupancy was positively but non-significantly associated with the precipitation that fell between the arrival of the Tree Swallows in mid-April and the completion of nest building at the end of May (P = 0.121; Fig. 2A), and it was independent of the number of cold days during that period (P = 0.47). In contrast, the probability of fledging young was negatively affected by the precipitation during June and the first half of July (P = 0.0002; Fig. 2B), but it was independent of the number of cold days during that period (P = 0.69). When rainfall was 50 mm less than normal, 75% of nests were predicted to fledge young; only 56% of nests were predicted to fledge young with 50 mm more rain than normal. Both increasing number of cold days and increasing precipitation in the nesting season significantly reduced the number of fledglings per nest (cold days: P = 0.0059, precipitation: P = 0.0175; Fig. 2C); with 50 mm of rain less than normal, 2.31 nestlings per nest fledged and 1.36 young per nest fledged with 50 mm more rain than normal. In years with average rainfall, an additional cold day during the fledging phase reduced the number of fledglings per nest from 1.80 to 1.54. Temperature and precipitation had additive, rather than interactive, effects on all 3 aspects of nesting success (temperature x precipitation interactions: P ≥ 0.42). Northeastern Naturalist Vol. 26, No. 2 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 241 Nest-box proximity Across the 38-y study, occupancy varied widely among nest boxes, with 3 occupied in 20 or more years, whereas 1 was occupied in only 4 y. The probability of Figure 1. Long-term trends in nesting success for Tree Swallows on Hardwood Island, Maine, between 1981 and 2018. (A) Nest-box occupancy has declined dramatically, especially since 2000, whereas (B) the fraction of nests fledging young has remained unchanged. (C) The number of fledglings per nest also declined over the study period. Predictions, derived from generalized linear mixed-effect models, are shown with raw data. Years in which Northern Raccoon (Procyon lotor) preyed upon Tree Swallow nests (1985, 1989, 1990, and 1991) are shown with filled points. Response variables were not available in all of these years; thus, only 3 “raccoon” years appear in panels (A) and (C) and only 1 appears in panel (B). Solid and dashed lines indicate significant (P ≤ 0.05) and non-significant relationships, respectively. The shaded area represents 95% confidence region, derived from parametric bootstrapping. Northeastern Naturalist 242 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 Vol. 26, No. 2 nest-box occupancy significantly declined with increasing distance to the nearest nest box (P = 0.0001; Fig. 3A). Increased spacing between nest boxes, from 20 m to 50 m, reduced the probability of occupancy from 74.7% to 51.2%. Occupancy Figure 2. Aspects of nesting success, as predicted by weather. (A) Nest-box occupancy is positively but non-significantly associated with the number of cold days in the second half of April and May. (B) The fraction of nests that fledged young significantly declined with increased total rainfall in June and the first half of July. (C) The number of fledglings per nest was reduced by both the number of cold days (≤18.5 °C) and the total precipitation in June and July. Precipitation is shown as the residuals of a regression against year. Thus, negative numbers indicate relatively dry years, and positive numbers indicate relatively wet years. Solid and dashed lines indicate significant (P ≤ 0.05) and nonsignificant relationships, respectively. The shaded area represents 95% confidence region, derived from parametric bootstrapping. Northeastern Naturalist Vol. 26, No. 2 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 243 increased with increasing distance to the nearest building (P < 0.0001; Fig. 3B). Increasing the distance from a nest box to a building from 20 m to 50 m increased the probability of occupancy from 31.5% to 65.7%. Occupancy was independent of distance to the main path through the meadow (P = 0.18; Fig 3C). Both the probability of fledging and the number of young fledged were independent of the distances to the nearest nest box, the nearest building, and the main path (P ≥ 0.24; results not shown). Number of nesting seasons since last mowing Nest-box occupancy increased significantly with time since mowing (P = 0.0001; Fig. 4A). Five years after mowing, predicted nest occupancy increased from 66.6% to 76.4%. The probability of fledging young was independent of time since mowing (P = 0.94; Fig. 4B), whereas the number of fledglings per nest decreased markedly with time since mowing (P = 0.0238; Fig. 4C). The predicted number of fledglings per nest decreased from 1.75 in the year of mowing to 1.23 5 years after mowing. Discussion Our study provides support for broad-scale observations of a decline in abundance of Tree Swallows across northeastern North America (Michel et al. 2016, Nebel et al. 2010, Sauer et al. 2017). We highlight 2 demographic factors that have contributed to this decline: striking decreases in nest-box occupancy and the number of young fledging from each nest (Fig. 1). The decline in the number Figure 3. Distances to landscape features. Nest-box occupancy was (A) negatively associated with distance to the nearest neighboring box, (B) positively associated with distance to the nearest building, and (C) independent of distance to the main path through the meadow. The fraction of nests that fledged young and the number of fledglings per nest were independent of the distances to other objects.Solid and dashed lines indicate significant (P ≤ 0.05) and non-significant relationships, respectively. The shaded area represents 95% confidence region, derived from parametric bootstrapping. Northeastern Naturalist 244 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 Vol. 26, No. 2 of fledglings began in the 1980s, whereas nest occupancy began to decline in the 2000s. This difference in timing suggests that multiple factors may have reduced nesting success at this site. Tree Swallows are income breeders, as food availability Figure 4. Time since mowing. (A) Nest-box occupancy was positively associated with time since mowing, whereas (B) the fraction of nests that fledged young was independent of time since mowing, and (C) the number of fledglings per nest decreased with time since mowing. Time since mowing is shown as the residuals of a regression against year; negative numbers indicate periods in which mowing had recently occurred, and positive numbers indicate periods many years after mowing. Solid and dashed lines indicate significant (P ≤ 0.05) and non-significant relationships, respectively. The shaded area represents 95% confidence region, derived from parametric bootstrapping. Northeastern Naturalist Vol. 26, No. 2 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 245 mediates their nesting behavior, including nest construction, egg laying, incubation, and nestling provisioning (Dunn et al. 2011, Nooker et al. 2005). Precipitation affected nesting success more strongly than did the frequency of cold days (Fig. 2). Nest-box occupancy increased slightly, though non-significantly, in rainy years (Fig. 2A). This finding suggests that Tree Swallows are not deterred by cold or wet weather during nest construction and incubation. However, consistent with previous studies (Weegman et al. 2017), both the percentage of nests with fledglings and the number of fledglings were reduced in rainy years (Fig. 2). The number of fledglings was further reduced in breeding seasons with many cold days (Fig. 2C). These results are to be expected, as nest temperatures fall rapidly and nesting survival declines during cold, wet weather (Ardia et al. 2010, Winkler et al. 2013), in part because adults make more short-duration foraging bouts (Coe et al. 2015). This pattern in nesting success is not only seen in Tree Swallows but also in other closely related aerial insectivores, such as Delichon urbica (L.) (House Martin) and Hirundo rustica L. (Barn Swallow) (Bryant 2008, Dunn et al. 2011). It is not clear why nesting success at this site was less sensitive to the frequency of cold days than was observed in previous studies (Winkler et al. 2013). However, Winkler and colleagues (2013) found the frequency of cold snaps of 3 days’ duration to be more predictive of nestling survival than were 1- or 2-d periods of cold. In our dataset, however, multi-day cold snaps were scarce, precluding their use in statistical analyses. In contrast with previous studies (Hussell 2012), distance between nest boxes was associated with a decrease in nest-box occupancy (Fig. 3A). Increased proximity between nest boxes can be a trigger for aggressive behavior between adult Tree Swallows, which can have negative consequences for nest construction (Male et al. 2006). These interactions may, however, only occur when food availability is low (Hussell 2012). Unfortunately, direct measurements of insect availability were beyond the scope of this study. The positive effect of nest-box proximity may result from Tree Swallows seeking the presence of other individuals, looking for opportunities for extra-pair fertilizations, and a better capacity to detect nest predators. It is also possible that the distances between nest boxes on Hardwood Island were great enough, nest-box density in the meadow low enough, or insect availability abundant enough to preclude competitive behavior between adults. A stronger predictor of nest-box occupancy was the distance from a nest box to the nearest building. Nest boxes closest to buildings consistently had reduced occupancy (Fig. 3B), suggesting that anthropogenic disturbance perturbed nest construction. This finding is surprising, however, as no people are resident on Hardwood Island during the nest-construction period of mid-April to the end of May, and the amount of activity in the houses then is minimal. The result suggests that the presence of the buildings themselves, aside from the daily activity of humans, reduces the inclination to build a nest. To improve occupancy rates, nest boxes should be moved farther from the buildings. The effects of proximity to buildings fade rapidly; even a distance of 50 m appears to be adequate to avoid disturbance (Fig. 3). Northeastern Naturalist 246 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 Vol. 26, No. 2 Nest-box occupancy increased slightly, whereas the number of fledglings per nest declined substantially, with time since mowing (Fig. 4). The decline in the Figure 5. Temporal variation in predictor variables. There has been great variation in (A) the frequency of cold days in the second half of April and May, (B) cold days in June and the first half of July, (C) precipitation in the second half of April and May, (D) precipitation in June and the first half of July, and (E) the time since mowing. Only the frequency of cold days in June and July has increased significantly over time. Solid and dashed lines indicate significant (P ≤ 0.05) and nonsignificant relationships, respectively. The shaded area represents 95% confidence region, derived from parametric bootstrapping. Northeastern Naturalist Vol. 26, No. 2 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 247 number of fledglings suggests that food availability declined as ecological succession advanced with grassy vegetation being replaced by woody shrubs. This pattern stands in contrast to the results of Ghilain and Bélisle (2008), but they investigated Tree Swallow nesting in a more intensively managed agricultural landscape. To promote fledgling success, we suggest that mowing occur after the nesting season at least once every 5 years, although this may adversely affect nest-box occupancy. Which factor has caused the observed decline in Tree Swallow nesting success? This study identified factors related to weather, geography, and vegetation that affect the nesting success of Tree Swallows on the coast of Maine. Which of these underlies the long-term declines in nest-box occupancy and in the number of fledglings per nest (Fig 1.)? Briefly, none of them do. The reason is simple: most of these factors have not changed directionally over the 38-y duration of this study (Fig. 5). Only the frequency of cold days during the fledging phase increased significantly over time (P = 0.0008; Fig. 5B), and it was only associated with a minor reduction on the number of fledglings per nest (Fig. 2C). Aside from the distances to landscape features, which have not moved, the other factors varied substantially through time, and their effects on nesting success likewise have been variable (Figs. 2, 3). They are unlikely, however, to have caused the major long-term decline in nesting success. This finding suggests that the decline in nesting success has other origins. The decline in nest-box occupancy may be driven by fewer adult Tree Swallows returning each summer to the coast of Maine or by returning adults opting not to build nests. The most likely explanation is that the substantial decline in Tree Swallow abundance observed across northeastern North America over the past 50 years has led to fewer adults returning to the study site (Sauer et al. 2017, Shutler et al. 2012). It is less likely that the local Tree Swallow abundance has remained steady with an unidentified factor causing nest-box occupancy to decline. The decrease in the number of fledglings per nest is more problematic. It is possible that the increased frequency of cold days during the fledging phase may have contributed to this decrease. In general, however, Hardwood Island appears to remain as ideal a site for Tree Swallow nesting and breeding as it has ever been, at least in terms of the variables we considered. Thus, we speculate that the body condition of returning Tree Swallows has deteriorated over time, causing them to fledge fewer young per nest (Rioux Paquette et al. 2014). Although insecticide use can reduce return rates and body condition (Stanton et al. 2017), insecticides have never been applied at the study site. Therefore, we infer that anthropogenic effects on the wintering grounds or along the migration route are the primary causes of reduced nesting success (Ghilain and Bélisle 2008, Robbins et al. 1989, Stanton et al. 2017), which suggests that halting or reversing the declines in abundance for Tree Swallows will require concerted and large-scale conservation action. Northeastern Naturalist 248 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 Vol. 26, No. 2 Acknowledgments We thank the many investigators, including Pamela Manice and other instructors at the Hardwood Island Biological Station, who contributed to data collection. Two anonymous reviewers and the editor, Heather York, provided helpful comments on an earlier version of this paper. Literature Cited Ardia, D.R., J.H. Perez, and E.D. Clotfelter. 2010. Experimental cooling during incubation leads to reduced innate immunity and body condition in nestling Tree Swallows. Proceedings of the Royal Society of London. Series B: Biological Sciences 277:1881–1888. Bates, D., M. Mächler, B. Bolker, and S. Walker. 2012. Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67:51. Bellavance, V., M. Bélisle, J. Savage, F. Pelletier, and D. Garant. 2018. Influence of agricultural intensification on prey availability and nestling diet in Tree Swallows (Tachycineta bicolor). Canadian Journal of Zoology 96:1053–1065. Bourret, A., M. Bélisle, F. Pelletier, and D. Garant. 2015. Multidimensional environmental influences on timing of breeding in a Tree Swallow population facing climate change. Evolutionary Applications 8:933–944. Bryant, D.M. 2008. Breeding biology of House Martins, Delichon urbica, in relation to aerial insect abundance. Ibis 117:180–216. Burnham, K.P., and D.R. Anderson. 2002. Model Selection and Multi-Model Inference: A Practical Information-Theoretical Approach. Springer Verlag, New York, NY. 496 pp. Canham, C.D., P.T. LePage, and K.D. Coates. 2004. A neighborhood analysis of canopytree competition: Effects of shading versus crowding. Canadian Journal of Forest Research 34:778–787. Coe, B.H., M.L. Beck, S.Y. Chin, C.M.B. Jachowski, and W.A. Hopkins. 2015. Local variation in weather conditions influences incubation behavior and temperature in a passerine bird. Journal of Avian Biology 46:385–394. Dunn, P.O., and D.W. Winkler. 1999. Climate change has affected the breeding data of Tree Swallows throughout North America. Proceedings of the Royal Society of London. Series B: Biological Sciences 266:2487–2490. Dunn, P.O., K.J. Thusius, K. Kimber, and D.W. Winkler. 2000. Geographic and ecological variation in clutch size of Tree Swallows. The Auk 117:215–221. Dunn, P.O., D.W. Winkler, L.A. Whittingham, S.J. Hannon, and R.J. Robertson. 2011. A test of the mismatch hypothesis: How is timing of reproduction related to food abundance in an aerial insectivore? Ecology 92:450–461. EBird. 2017. eBird: An online database of bird distribution and abundance. eBird, Cornell Lab of Ornithology, Ithaca, NY. Available online at http://www.ebird.org. Accessed 2 July 2018. Forsman, J.J.T., M.B. Hjernquist, J. Taipale, and L. Gustafsson. 2008. Competitor-density cues for habitat quality facilitating habitat selection and investment decisions. Behavioral Ecology 19:539–545. Ghilain, A., and M. Bélisle. 2008. Breeding success of Tree Swallows along a gradient of agricultural intensification. Ecological Applications 18:1140–1154. Hussell, D.J.T. 2012. The influence of food abundance on nest-box occupancy and territory size in the Tree Swallow, a species that does not defend a feeding territory. The Condor 114:595–605. Northeastern Naturalist Vol. 26, No. 2 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 249 Hussell, D.J.T., and T.E. Quinney. 1985. Food abundance and clutch size of Tree Swallows, Tachycineta bicolor. Ibis 129:243–258. Knight, S.M., D.W. Bradley, R.G. Clark, E.A. Gow, M. Bélisle, L.L. Berzins, T. Blake, E.S. Bridge, L. Burke, R.D. Dawson, P.O. Dunn, D. Garant, G.L. Holroyd, D.J.T. Hussell, O. Lansdorp, A.J. Laughlin, M.L. Leonard, F. Pelletier, D. Shutler, L. Siefferman, C.M. Taylor, H.E. Trefry, C.M. Vleck, D. Vleck, D.W. Winkler, L.A. Whittingham, and D.R. Norris. 2018. Constructing and evaluating a continent-wide migratory-songbird network across the annual cycle. Ecological Monographs 88:445–460. Lifjeld, J.T., P.O. Dunn, R.J. Robertson, and P.T. Boag. 1993. Extra-pair paternity in monogamous Tree Swallows. Animal Behaviour 45:213–229. MacGregor-Fors, I., L. Morales-Pérez, and J.E. Schondube. 2010. Migrating to the city: Responses of Neotropical migrant-bird communities to urbanization. The Condor 112:711–717. Male, S.K., J. Jones, and R.J. Robertson. 2006. Effects of nest-box density on the behavior of Tree Swallows during nest building. Journal of Field Ornithology 77:61–66. McArthur, S.L., A.E. McKellar, N.J. Flood, and M.W. Reudink. 2017. Local weather and regional climate influence breeding dynamics of Mountain Bluebirds (Sialia currucoides) and Tree Swallows (Tachycineta bicolor): A 35-year study. Canadian Journal of Zoology 95:271–277. McCarty, J.P. 2001. Variation in growth of nestling Tree Swallows across multiple temporal and spatial scales. The Auk 118:176–190. McCarty, J.P., and D.W. Winkler. 1999a. Relative importance of environmental variables in determining the growth of nestling Tree Swallows, Tachycineta bicolor. Ibis 141:286–296. McCarty, J.P., and D.W. Winkler. 1999b. Foraging ecology and diet selectivity of Tree Swallows feeding nestlings. The Condor 101:246–254. Michel, N.L., A.C. Smith, R.G. Clark, C.A. Morrissey, and K.A. Hobson. 2016. Differences in spatial synchrony and interspecific concordance inform guild-level population trends for aerial insectivorous birds. Ecography 39:774–786. Miller, J.R., J.A. Wiens, N.T. Hobbs, and D.M. Theobald. 2003. Effects of human settlement on bird communities in lowland riparian areas of Colorado (USA). Ecological Applications 13:1041–1059. Nebel, S., A. Mills, J.D. McCracken, and P.D. Taylor. 2010. Declines of aerial insectivores in North America follow a geographic gradient. Avian Conservation and Ecology 5(2):1. Nooker, J., P.O. Dunn, and L.A. Whittingham. 2005. Effects of food abundance, weather, and female condition on reproduction in Tree Swallows (Tachycineta bicolor). The Auk 122:1225–1238. Pérez, J.H., D.R. Ardia, E.K. Chad, and E.D. Clotfelter. 2008. Experimental heating reveals nest temperature affects nestling condition in Tree Swallows (Tachycineta bicolor). Biology Letters 4:468–471. R Core Team. 2018. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Rioux Paquette, S., F. Pelletier, D. Garant, and M. Bélisle. 2014. Severe recent decrease of adult body mass in a declining insectivorous bird population. Proceedings of the Royal Society of London. Series B: Biological Sciences 281:20140649. Robbins, C.S., J.R. Sauer, R.S. Greenberg, and S. Droege. 1989. Population declines in North American birds that migrate to the Neotropics. Proceedings of the National Academy of Sciences 86:7658–7662. Northeastern Naturalist 250 M. Blyth, C.W.E. Paine, and C.E.T. Paine 2019 Vol. 26, No. 2 Robertson, R.J., and W.B. Rendell. 1990. A comparison of the breeding ecology of a secondary cavity-nesting bird, the Tree Swallow (Tachycineta bicolor), in nest boxes and natural cavities. Canadian Journal of Zoology 68:1046–1052. Robillard, A., D. Garant, and M. Bélisle. 2013. The swallow and the sparrow: How agricultural intensification affects abundance, nest-site selection, and competitive interactions. Landscape Ecology 28:201–215. Sauer, J.R., D.K. Niven, J.E. Hines, J.D J. Ziolkowski, K.L. Pardieck, J.E. Fallon, and W.A. Link. 2017. The North American Breeding Bird Survey, Results and Analysis 1966–2015. USGS Patuxent Wildlife Research Center, Laurel, MD. Available online at https://www.mbr-pwrc.usgs.gov/bbs/. Accessed 15 July 2018. Schielzeth, H. 2010. Simple means to improve the interpretability of regression coefficients. Methods in Ecology and Evolution 1:103–113. Shutler, D., D.J.T. Hussell, D.R. Norris, D.W. Winkler, R.J. Robertson, F. Bonier, W.B. Rendell, M. Bélisle, R.G. Clark, R.D. Dawson, N.T. Wheelwright, M.P. Lombardo, P.A. Thorpe, M.A. Truan, R. Walsh, M.L. Leonard, A.G. Horn, C.M. Vleck, D. Vleck, A.P. Rose, L.A. Whittingham, P.O. Dunn, K.A. Hobson, and M.T. Stanback. 2012. Spatiotemporal patterns in nest-box occupancy by Tree Swallows across North America. Avian Conservation and Ecology 7:3–12. Stanton, R., R.G. Clark, and C.A. Morrissey. 2017. Intensive agriculture and insect-prey availability influence oxidative status and return rates of an aerial insectivore. Ecosphere 8. Weegman, M.D., T.W. Arnold, R.D. Dawson, D.W. Winkler, and R.G. Clark. 2017. Integrated population models reveal local weather conditions are the key drivers of population dynamics in an aerial insectivore. Oecologia 185:119–130. Winkler, D.W., M.K. Luo, and E. Rakhimberdiev. 2013. Temperature effects on food supply and chick mortality in Tree Swallows (Tachycineta bicolor). Oecologia 173:129–138.