https://orcid.org/0000-0003-3964-7015
Abstract
Rapid high-intensity molt of flight feathers occurs in many bird species and can have several detrimental consequences, including reductions in flight capabilities, foraging performance, parental care, and plumage quality. Many migratory New World warblers (family Parulidae) are known to have intense remigial molt, and recent work has suggested that simultaneous replacement of the rectrices may be widespread in the family as well. However, the phylogenetic distribution of simultaneous rectrix molt, and high-intensity flight feather molt more generally, has not been systematically investigated in warblers. We addressed this issue by examining flight feather molt in 13 species, representing 7 different warbler genera, at Powdermill Avian Research Center in southwestern Pennsylvania, USA. All 13 species replaced their 12 rectrices simultaneously, with the onset of rectrix molt occurring in the early-middle stages of high-intensity primary molt. As expected, single-brooded early migrants molted earlier than double-brooded species whose nesting activities extend into late summer. However, our finding that late-molting species replaced their primaries more slowly and less intensively than early molting species was unexpected, as late-molting species are widely hypothesized to be under stronger migration-related time constraints. This surprising result appears to be at least partially explained by a positive association between the pace of molt and daylength; shorter late-summer days may mandate reduced daily food intake, lower molt intensity, and a slower pace of molt. In comparison to other passerines, flight feather molt in warblers of eastern North America is extraordinarily intense; at its peak, individuals are simultaneously replacing 50–67% of their 48 flight feathers (all 12 rectrices and 6–10 remiges on each wing) for 2–3 weeks or more. Because molt of this intensity is likely to present numerous challenges for flight, avoiding predators, foraging, and parental care, the period of flight feather molt for warblers constitutes a highly demanding phase of their annual cycle.
Resumen
La muda rápida de alta intensidad de las plumas del vuelo ocurre en muchas especies de aves y puede tener consecuencias negativas, incluyendo reducción en la capacidad de vuelo, en el desempeño de forrajeo, en el cuidado parental y en la calidad del plumaje. Muchas currucas migratorias del Nuevo Mundo (familia Parulidae) son conocidas por presentar una muda intensa de las remeras, y trabajos recientes han sugerido que el reemplazo simultáneo de las timoneras también puede estar ampliamente presente en esta familia. Sin embargo, la distribución filogenética de la muda simultánea de las timoneras, y de modo más general de la muda de alta intensidad de las plumas del vuelo, no ha sido investigada sistemáticamente en las currucas. Analizamos este tema examinando la muda de las plumas del vuelo en 13 especies, representando 7 géneros diferentes de currucas, en el Centro de Investigación Aviar de Powdermill en el suroeste de Pensilvania, EEUU. Las 13 especies reemplazaron sus 12 timoneras simultáneamente, con el inicio de la muda de las timoneras ocurriendo en los estadios tempranos y medios de la muda primaria de alta intensidad. Como se esperaba, los migrantes tempranos de una sola nidada mudaron más temprano que las especies con dos nidadas cuyas actividades de anidación se extendieron hacia fines del verano. Sin embargo, nuestro hallazgo de que las especies de muda tardía reemplazaron sus primarias más despacio y con menor intensidad que las especies de muda temprana fue inesperado, ya que se supone que las especies de muda tardía tienen restricciones de tiempo más fuertes relacionadas con la migración. Este resultado sorprendente parece estar explicado al menos parcialmente por una asociacion positiva entre el ritmo de la muda y la duración del día; los días más cortos de fines de verano pueden imponer una reducción en el consumo diario de alimento, menor intensidad de muda y un ritmo más lento de muda. En comparación con otros paseriformes, la muda de las plumas del vuelo en las currucas del este de Norteamérica es extraordinariamente intensa; en su pico, los individuos están reemplazando simultáneamente 50–67% de sus 48 plumas del vuelo (las 12 timoneras y 6–10 remeras en cada ala) durante 2–3 semanas o más. Debido a que la intensidad de esta muda impone probablemente numerosos desafíos para el vuelo, para evitar depredadores, para el forrajeo y para el cuidado parental, el período de muda de las plumas del vuelo para las currucas constituye una fase de alta demanda de su ciclo anual.
Lay Summary
• Most birds undergo an energetically demanding annual molt in which they shed and replace all their flight feathers—the long feathers of the wing and tail that provide thrust, lift, and maneuverability.
• Rapid high-intensity molt, in which birds replace multiple flight feathers simultaneously, may be necessary in migratory species that are under strong migration-related time constraints but can seriously compromise flight capabilities.
• We examined the intensity of flight feather molt in 13 species of warblers that breed in eastern North America and undergo a complete late-summer molt before beginning their fall migration to the tropics and subtropics.
• We found that flight feather molt in warblers is extraordinarily intense; during its peak, individuals are replacing 50–67% of their 48 flight feathers simultaneously, an intensity that is among the highest reported for any songbird.
• Because molt of this intensity presents clear challenges for flight, foraging, avoiding predators, and caring for offspring, the late-summer molt constitutes a brief but highly demanding phase of the annual cycle for these colorful migratory songbirds, and a phase that warrants more attention from field ornithologists.
INTRODUCTION
Feather replacement during molt constitutes one of the most demanding stages of the life cycle of birds. Because molt requires considerable investments in both energy and protein synthesis (Murphy and King 1992, Lindström et al. 1993), molting birds may need to increase their foraging effort and food intake at a time when their ability to fly and forage may be compromised by the loss of multiple flight feathers (Swaddle and Witter 1997, Hedenström 2003, Bonier et al. 2007, Echeverry-Galvis and Hau 2012). These challenges can be especially acute in species where molt is rapid and intense; high-intensity molt is common in time-constrained species where rapid feather replacement may be unavoidable, such as long-distance migrants that breed at high latitudes (Jenni and Winkler 1994, Howell 2010). Rapid high-intensity molt, however, can produce numerous negative consequences, including substantially impaired flight or even temporary flightlessness (Haukioja 1971, Green and Summers 1975, Hedenström 2003, Tomotani et al. 2018), reduced ability to provide parental care (Svensson and Nilsson 1997, Mumme 2018), and increased energetic costs of flight resulting from the reduced wing area and increased body mass typical of wild birds during molt (Newton 1968, Heise and Rimmer 2000, Lind et al. 2004). Accelerated molt can also result in a poor quality of the newly grown feathers, which may increase future thermoregulatory costs (Dawson et al. 2000, Dawson 2004), differentially affect birds in poor body condition (Vágási et al. 2012), and increase the likelihood that future supplemental molts (Podlaszczuk et al. 2016), or even a second complete molt on the wintering grounds (Underhill et al. 1992), may be required to replace low-quality feathers.
The New World warblers (family Parulidae) include several species known to exhibit rapid, high-intensity molt of the flight feathers. Many members of the family, particularly long-distance migrants that breed at high latitudes in North America, replace an unusually large percentage of their remiges and rectrices simultaneously (Dwight 1900, Foster 1967, Rimmer 1988, Voelker and McFarland 2002, Ryder and Rimmer 2003, Mumme 2018, Rimmer and McFarland 2020). Several authors have noted that the intense flight feather molt of warblers is likely to seriously impair flight capability (Foster 1967, Hubbard 1980, Ryder and Rimmer 2003, Mumme 2018), and one has suggested that it “probably results in the virtual flightlessness of some birds” (Rimmer 1988, p. 154).
Recent work on the Hooded Warbler (Setophaga citrina) has highlighted an aspect of high-intensity flight feather molt that has previously received relatively little attention: simultaneous rectrix molt (Mumme 2018). Replacement of the tail feathers in Hooded Warblers begins with the loss of the central rectrices, followed within a few days by loss of the remaining rectrices, resulting in a completely tailless bird growing and replacing all 12 rectrices simultaneously. Simultaneous rectrix molt invariably begins during primary molt, at about the time the 5th primaries are lost, and has significant implications for late-season parental care; ~70% of breeding adults that initiate rectrix molt before the end of parental care desert their late-season nestlings or fledglings, leaving the mate responsible for all remaining care (Mumme 2018). Parental desertion is probably a consequence of molt-imposed reductions in flight capabilities (e.g., Tomotani et al. 2018), reductions in foraging performance (Hooded Warblers use their white tail spots and tail-flicking behavior to startle potential insect prey; Mumme 2014), the high nutritional and energetic demands of replacing all rectrices and multiple remiges simultaneously (Murphy and King 1992), or a combination of all three factors. Adults in simultaneous rectrix molt nearly always desert if their offspring are nestlings or young fledglings, but desertion is less certain if the offspring are older fledglings approaching independence, probably because the costs of providing parental care during simultaneous rectrix molt are lower when offspring are relatively mobile and partially self-sufficient (Mumme 2018).
Simultaneous rectrix molt within the Parulidae does not appear to be unique to Hooded Warblers; synchronous loss and replacement of all rectrices during primary molt has been reported from at least 10 other species representing 3 different genera of North American warblers (Foster 1967, Nolan 1978, Hubbard 1980, Rimmer 1988, Voelker and McFarland 2002, Rimmer and McFarland 2020, Bocetti et al. 2020, Ladd and Gass 2020, Mattsson et al. 2020). However, because some of the reported cases are based on limited or incidental data, a more systematic investigation of the phenomenon, across a wide range of species at the same location using consistent methods, is needed for a better understanding of its phylogenetic distribution and its importance in contributing to high-intensity molt in the family.
Here we provide such an analysis for adults of 13 species of parulid warblers, representing 7 different genera, undergoing their annual prebasic molt during the summer and early autumn in southwestern Pennsylvania, USA. The purpose of our study was to systematically examine simultaneous rectrix molt, and the ecology of high-intensity flight feather molt more broadly, for this 13-species assemblage. We specifically address five questions: (1) Is simultaneous molt of all rectrices unusual or phylogenetically widespread in parulid warblers? (2) When does rectrix molt occur in relation to molt of primaries and other flight feathers? (3) What ecological factors contribute to interspecific variation in timing of the onset of flight feather molt? (4) What ecological factors contribute to interspecific variation in the speed, intensity, and duration of flight feather molt? (5) How do parulid warblers compare to other passerines in the intensity of flight feather molt?
METHODS
Field Methods
Molt data were collected during summer and fall banding operations, 1986–2000, at Powdermill Avian Research Center (40.1637°N, 79.2674°W) in the Laurel Highlands of Westmoreland County, southwestern Pennsylvania, USA. Molt scores were obtained from 1,289 individual captures of 13 different species representing 7 different parulid genera (Table 1). Because all 13 focal species nest regularly at Powdermill and surrounding areas of eastern Westmoreland County (Brauning 1992, Wilson et al. 2012), individuals captured during molt were likely local breeders on or near their nesting grounds (Pyle et al. 2018). In all 13 focal species, birds were captured at all stages of prebasic molt, and we found no evidence that individuals arrested or suspended molt prior to migration. All 18 remiges on the right wing (9 primaries, 6 secondaries, and 3 tertials) and the 6 right rectrices were scored using the 0–5 molt scoring system of the British Trust for Ornithology: 0 = old feather, 1 = old feather missing or new pin feather, 2 = new feather emerging from sheath up to one-third grown, 3 = new feather between one and two-thirds grown, 4 = new feather more than two-thirds grown but still sheathed at base, and 5 = fully grown new feather with no trace of the sheath at its base (Ginn and Melville 1983).
Summary of 1289 molt records of 13 species and 7 genera of warblers (Parulidae) obtained from banding activities at Powdermill Avian Research Center in southwestern Pennsylvania, 1986–2000.
| Common name | Scientific name | Alpha code | Sample size |
|---|---|---|---|
| Ovenbird | Seiurus aurocapilla | OVEN | 76 |
| Louisiana Waterthrush | Parkesia motacilla | LOWA | 44 |
| Golden-winged Warbler | Vermivora chrysoptera | GWWA | 34 |
| Blue-winged Warbler | Vermivora cyanoptera | BWWA | 25 |
| Black-and-White Warbler | Mniotilta varia | BAWW | 76 |
| Kentucky Warbler | Geothlypis formosa | KEWA | 58 |
| Common Yellowthroat | Geothlypis trichas | COYE | 276 |
| Hooded Warbler | Setophaga citrina | HOWA | 149 |
| American Redstart | Setophaga ruticilla | AMRE | 428 |
| Magnolia Warbler | Setophaga magnolia | MAWA | 23 |
| Yellow Warbler | Setophaga petechia | YEWA | 60 |
| Chestnut-sided Warbler | Setophaga pensylvanica | CSWA | 25 |
| Canada Warbler | Cardellina canadensis | CAWA | 15 |
| Total | 1,289 |
| Common name | Scientific name | Alpha code | Sample size |
|---|---|---|---|
| Ovenbird | Seiurus aurocapilla | OVEN | 76 |
| Louisiana Waterthrush | Parkesia motacilla | LOWA | 44 |
| Golden-winged Warbler | Vermivora chrysoptera | GWWA | 34 |
| Blue-winged Warbler | Vermivora cyanoptera | BWWA | 25 |
| Black-and-White Warbler | Mniotilta varia | BAWW | 76 |
| Kentucky Warbler | Geothlypis formosa | KEWA | 58 |
| Common Yellowthroat | Geothlypis trichas | COYE | 276 |
| Hooded Warbler | Setophaga citrina | HOWA | 149 |
| American Redstart | Setophaga ruticilla | AMRE | 428 |
| Magnolia Warbler | Setophaga magnolia | MAWA | 23 |
| Yellow Warbler | Setophaga petechia | YEWA | 60 |
| Chestnut-sided Warbler | Setophaga pensylvanica | CSWA | 25 |
| Canada Warbler | Cardellina canadensis | CAWA | 15 |
| Total | 1,289 |
Summary of 1289 molt records of 13 species and 7 genera of warblers (Parulidae) obtained from banding activities at Powdermill Avian Research Center in southwestern Pennsylvania, 1986–2000.
| Common name | Scientific name | Alpha code | Sample size |
|---|---|---|---|
| Ovenbird | Seiurus aurocapilla | OVEN | 76 |
| Louisiana Waterthrush | Parkesia motacilla | LOWA | 44 |
| Golden-winged Warbler | Vermivora chrysoptera | GWWA | 34 |
| Blue-winged Warbler | Vermivora cyanoptera | BWWA | 25 |
| Black-and-White Warbler | Mniotilta varia | BAWW | 76 |
| Kentucky Warbler | Geothlypis formosa | KEWA | 58 |
| Common Yellowthroat | Geothlypis trichas | COYE | 276 |
| Hooded Warbler | Setophaga citrina | HOWA | 149 |
| American Redstart | Setophaga ruticilla | AMRE | 428 |
| Magnolia Warbler | Setophaga magnolia | MAWA | 23 |
| Yellow Warbler | Setophaga petechia | YEWA | 60 |
| Chestnut-sided Warbler | Setophaga pensylvanica | CSWA | 25 |
| Canada Warbler | Cardellina canadensis | CAWA | 15 |
| Total | 1,289 |
| Common name | Scientific name | Alpha code | Sample size |
|---|---|---|---|
| Ovenbird | Seiurus aurocapilla | OVEN | 76 |
| Louisiana Waterthrush | Parkesia motacilla | LOWA | 44 |
| Golden-winged Warbler | Vermivora chrysoptera | GWWA | 34 |
| Blue-winged Warbler | Vermivora cyanoptera | BWWA | 25 |
| Black-and-White Warbler | Mniotilta varia | BAWW | 76 |
| Kentucky Warbler | Geothlypis formosa | KEWA | 58 |
| Common Yellowthroat | Geothlypis trichas | COYE | 276 |
| Hooded Warbler | Setophaga citrina | HOWA | 149 |
| American Redstart | Setophaga ruticilla | AMRE | 428 |
| Magnolia Warbler | Setophaga magnolia | MAWA | 23 |
| Yellow Warbler | Setophaga petechia | YEWA | 60 |
| Chestnut-sided Warbler | Setophaga pensylvanica | CSWA | 25 |
| Canada Warbler | Cardellina canadensis | CAWA | 15 |
| Total | 1,289 |
Data Analysis and Statistical Methods
For each individual, we calculated a total primary molt score (0–45) and a total rectrix molt score (0–30) by summing the individual scores for each feather. To examine interspecific variation in the degree to which individuals molted all their rectrices simultaneously, we examined rectrix molt data from 582 individuals of all 13 species captured during active tail molt (total rectrix molt score > 0 and < 30). For each individual, we calculated two indices of synchrony in rectrix molt: (1) the maximum difference in score between any two rectrices of any molt score (MaxDiff; 0 = completely synchronous rectrix molt, 5 = completely staggered rectrix molt), and (2) the standard deviation of molt score for the 6 scored rectrices (SD; 0 = completely synchronous, 1.87 = completely staggered). Because neither MaxDiff nor SD was normally distributed, we made between-species comparisons using Kruskal–Wallis and Dunn’s tests, nonparametric equivalents of ANOVA and post-hoc multiple comparisons. JMP Pro 12.2 (SAS Institute, Cary, North Carolina, USA) was used for all statistical calculations.
To determine the status of primary molt at the onset of rectrix molt, we examined primary feather molt scores for 126 individuals of 10 species that were captured during the initial early stages of rectrix molt (total rectrix molt score > 0 and ≤ 6). We made between-species comparisons of total primary molt score for these 10 species using ANOVA and Tukey–Kramer multiple comparison tests, as the examination of normal quantile plots of residuals indicated that total primary molt scores were approximately normally distributed.
We explored interspecific variation in the timing and duration of molt by focusing on primary molt. Primary molt in parulid warblers follows the standard passerine pattern, with loss of the first (innermost) primary invariably indicating the start of flight feather molt, with sequential loss of the remaining primaries occurring in a regular proximal-to-distal pattern (Foster 1967, Nolan 1978, Rimmer 1988). Because of the regularity and predictability of primary molt, and because the period of primary replacement nearly always encompasses molt of the rectrices, secondaries, and tertials as well (e.g., Foster 1967), our focus on the timing and duration of primary molt is appropriate.
We estimated the day of the year when the midpoint of primary molt was achieved by fitting 3-parameter logistic models to the total primary molt score data for each species, using the Nonlinear Platform of JMP Pro 12.2. We focused on estimating the midpoint (or halfway date; see Jackson 2017, Erni 2018) rather than the onset of primary molt because the estimated midpoint was much less sensitive to curve-fitting assumptions; a variety of linear and non-linear curve-fitting approaches—including Pimm regression (Pimm 1976), 3-parameter logistic models, cubic splines, and kernel smoothing—all produced similar estimates of the midpoint of primary molt. With the fitted logistic models, we then used JMP’s Inverse Prediction tool to estimate the mean ± SE day of the year that the midpoint of primary molt (primary molt score = 22.5) was achieved for each species.
We estimated the pace of primary molt for each species from total primary molt scores of individual birds captured two or more times during the same season while in primary molt (primary molt score > 0 and < 45). We favored this approach over alternatives such as Pimm regression (Pimm 1976) and Underhill-Zucchini maximum likelihood models (Underhill and Zucchini 1988, Erni et al. 2013) because it allowed us to examine the pace of molt for individual birds and also avoid problems of meeting assumptions of regression and maximum likelihood models (Newton 2009, Rohwer 2013). For each species, we fit a general linear model to total primary molt score with two fixed effects: day of the year as a continuous variable, and band identification number as a categorical variable. The coefficient (mean ± SE) for the day of the year in these models represented an overall estimate of the slope of the relationship between primary molt score and day (change in molt score per day). All models were created using the Fit Model platform of JMP Pro 12.2. With estimates of both the midpoint and pace of primary molt, we were able to estimate the mean date of onset of primary molt and the mean duration of primary molt for each species. A total of 309 molt records from 137 individual bird-seasons and 12 species were included in these analyses; recapture data for Blue-winged Warblers (BWWA) were too limited to use this approach, and this species was not included in the analysis of timing and pace of primary molt. Recapture data for Canada Warbler (CAWA) were also limited, but because recaptures spanned almost the entire range of total primary molt scores (Appendix Figure 8), this species was included in the analysis. Our estimates of midpoint and pace of primary molt, however, should be viewed cautiously for this species.
For all 13 species we calculated two species-level estimates of primary molt intensity: (1) Peak intensity, which we defined as the mean number of primary feathers on the right wing being replaced by individuals with total primary molt scores > 15 and ≤ 35; our estimate of peak intensity, therefore, excludes individuals in the initial or terminal stages of primary molt when few primaries are being molted simultaneously. (2) Average intensity (as described by Rohwer and Rohwer 2013), the mean number of primaries on the right wing growing simultaneously for each feather between the 2nd and 8th primary, inclusive; we used this somewhat more conservative estimate, which includes some individuals in the early stages of molt replacing only 2–3 primaries, to allow us to compare primary molt intensities of the 13 species of warblers with the values calculated for other passerines by Rohwer and Rohwer (2013).
To examine the effects of daylength on the pace of primary molt, we used the online SolarTopo calculator (van der Staay 2020) to determine daylength on particular dates for the latitude (40.1637°N) of Powdermill Avian Research Center. For some species-level comparisons, we controlled for phylogenetic relationships via phylogenetic independent contrasts, using a recent phylogeny of the Parulidae (Lovette et al. 2010) and the package caper (1.01; Orme 2018) of R 3.6.2 (R Core Team 2019).
RESULTS
Synchrony of Rectrix Molt
The 13 focal species varied in the degree to which rectrices were replaced simultaneously. The two species of Geothlypis (KEWA and COYE; see species codes in Table 1) showed the greatest rectrix molt synchrony (lowest values of rectrix MaxDiff and SD), while BAWW and AMRE were the least synchronous (Figure 1). However, all 13 species were overall quite synchronous, with mean MaxDiff values (range: 0.24–1.06) generally < 1 and SD values (range: 0.10–0.46) much closer to completely synchronous rectrix molt (SD = 0) than to completely staggered rectrix molt (SD = 1.87; Figure 1). Between-species differences in synchrony, although modest, were statistically significant for both MaxDiff and SD (Kruskal–Wallis χ 2 = 113.3 and 97.8, respectively, df = 12, both P < 0.001). Post-hoc multiple comparisons (Dunn’s tests) showed that the two most synchronous species, KEWA and COYE, differed from the two least synchronous species, BAWW and AMRE (Figure 1), but that differences between other species pairs generally were not significant.
Degree to which individuals molt rectrices simultaneously in 13 species of warblers representing 7 genera from Powdermill Avian Research Center in southwestern Pennsylvania, 1986–2000. Mean ± SE for two measures of rectrix molt synchrony are shown: the maximum difference in score between any two rectrices (MaxDiff; 0 = completely synchronous, 5 = completely staggered), and the standard deviation of molt score for the 6 scored rectrices (SD; 0 = completely synchronous, 1.87 = completely staggered). Numbers to the right of the error bars for the standard deviation indicate sample size. Phylogenetic relationships among the 13 species are shown at left and reflect the phylogeny of Lovette et al. (2010).
Status of Primary Molt at the Onset of Rectrix Molt
For the 10 species with sufficient data, simultaneous rectrix molt began when individuals were in the early-middle stages of primary molt, usually around the time individuals were dropping their 6th primaries (Figure 2). The mean total primary molt score for birds at the onset of rectrix molt varied from 11.8 (HOWA) to 20.7 (LOWA) and averaged 16.0 across the 10 species (Figure 2). Between-species differences in primary molt score at the onset of rectrix score were statistically significant (F9,116 = 2.7, P = 0.007), but Tukey–Kramer post-hoc multiple comparisons revealed no species pairs that differed significantly from each other (all P > 0.1).
Molt scores of primary feathers for 10 species of warblers at the initial stages of rectrix molt (total rectrix molt score > 0 and ≤ 6) from Powdermill Avian Research Center in southwestern Pennsylvania, 1986–2000. Bars and error bars indicate the mean ± standard error of molt score for each primary feather. Mean ± SE of total primary molt score and sample size are shown numerically for each species. Phylogenetic relationships are indicated at left and reflect the phylogeny of Lovette et al. (2010).
Timing, Pace, and Duration of Primary Molt
Warblers showed considerable interspecific variation in both the timing and pace of primary molt (Table 2, Figure 3, Appendix Figure 8). LOWA, GWWA, and YEWA were the earliest molting species, with estimated mean onset dates of June 19, 20, and 21, respectively, more than five weeks in advance of the late-molting COYE (July 28; Table 2, Figure 3). The species with the most rapid pace of primary molt were YEWA, GWWA, and CAWA, with estimated molt duration of 39, 41, and 41 days, respectively (Table 2, Appendix Figure 8). In contrast, we estimated that the species with the slowest pace of primary molt, OVEN, and HOWA (Appendix Figure 8), would require 53 and 52 days to complete primary molt (Table 2).
Estimates of timing, intensity, and pace of primary molt for 12 species of warblers from Powdermill Avian Research Center in southwestern Pennsylvania, 1986–2000. Peak primary molt intensity is the number of primary feathers being replaced simultaneously on the right wing during the middle stages of primary molt (total primary molt score > 15 and ≤ 35). Pace of primary molt is the estimated daily rate of increase in primary molt score.
| Species alpha code | Midpoint of primary molt (mean ± SE) | Peak primary molt intensity (mean ± SE) | Pace of primary molt (mean ± SE) | Estimated mean onset of primary molt | Estimated mean duration of primary molt |
|---|---|---|---|---|---|
| OVEN | July 30 ± 1.4 day | 4.8 ± 0.9 | 0.851 ± 0.050 | July 3 | 53 day |
| LOWA | July 12 ± 2.4 day | 5.1 ± 1.2 | 1.000 ± 0.065 | June 19 | 45 day |
| GWWA | July 12 ± 2.3 day | 5.9 ± 1.3 | 1.107 ± 0.067 | June 21 | 41 day |
| BAWW | July 21 ± 1.8 day | 4.6 ± 1.1 | 0.988 ± 0.033 | June 27 | 46 day |
| KEWA | August 1 ± 1.8 day | 5.1 ± 0.8 | 0.952 ± 0.050 | July 8 | 48 day |
| COYE | August 21 ± 0.7 day | 4.6 ± 0.9 | 0.935 ± 0.018 | July 28 | 49 day |
| HOWA | August 16 ± 1.2 day | 4.3 ± 1.5 | 0.867 ± 0.030 | July 20 | 52 day |
| AMRE | July 19 ± 0.7 day | 4.9 ± 1.3 | 1.074 ± 0.030 | June 28 | 42 day |
| MAWA | August 9 ± 3.4 day | 4.6 ± 1.1 | 0.935 ± 0.029 | July 15 | 49 day |
| YEWA | July 10 ± 1.3 day | 5.3 ± 1.2 | 1.154 ± 0.055 | June 20 | 39 day |
| CSWA | August 5 ± 2.8 day | 5.0 ± 0.8 | 0.935 ± 0.063 | July 12 | 49 day |
| CAWA | July 29 ± 3.3 day | 5.0 ± 0.0 | 1.109 ± 0.144 | July 8 | 41 day |
| Species alpha code | Midpoint of primary molt (mean ± SE) | Peak primary molt intensity (mean ± SE) | Pace of primary molt (mean ± SE) | Estimated mean onset of primary molt | Estimated mean duration of primary molt |
|---|---|---|---|---|---|
| OVEN | July 30 ± 1.4 day | 4.8 ± 0.9 | 0.851 ± 0.050 | July 3 | 53 day |
| LOWA | July 12 ± 2.4 day | 5.1 ± 1.2 | 1.000 ± 0.065 | June 19 | 45 day |
| GWWA | July 12 ± 2.3 day | 5.9 ± 1.3 | 1.107 ± 0.067 | June 21 | 41 day |
| BAWW | July 21 ± 1.8 day | 4.6 ± 1.1 | 0.988 ± 0.033 | June 27 | 46 day |
| KEWA | August 1 ± 1.8 day | 5.1 ± 0.8 | 0.952 ± 0.050 | July 8 | 48 day |
| COYE | August 21 ± 0.7 day | 4.6 ± 0.9 | 0.935 ± 0.018 | July 28 | 49 day |
| HOWA | August 16 ± 1.2 day | 4.3 ± 1.5 | 0.867 ± 0.030 | July 20 | 52 day |
| AMRE | July 19 ± 0.7 day | 4.9 ± 1.3 | 1.074 ± 0.030 | June 28 | 42 day |
| MAWA | August 9 ± 3.4 day | 4.6 ± 1.1 | 0.935 ± 0.029 | July 15 | 49 day |
| YEWA | July 10 ± 1.3 day | 5.3 ± 1.2 | 1.154 ± 0.055 | June 20 | 39 day |
| CSWA | August 5 ± 2.8 day | 5.0 ± 0.8 | 0.935 ± 0.063 | July 12 | 49 day |
| CAWA | July 29 ± 3.3 day | 5.0 ± 0.0 | 1.109 ± 0.144 | July 8 | 41 day |
Estimates of timing, intensity, and pace of primary molt for 12 species of warblers from Powdermill Avian Research Center in southwestern Pennsylvania, 1986–2000. Peak primary molt intensity is the number of primary feathers being replaced simultaneously on the right wing during the middle stages of primary molt (total primary molt score > 15 and ≤ 35). Pace of primary molt is the estimated daily rate of increase in primary molt score.
| Species alpha code | Midpoint of primary molt (mean ± SE) | Peak primary molt intensity (mean ± SE) | Pace of primary molt (mean ± SE) | Estimated mean onset of primary molt | Estimated mean duration of primary molt |
|---|---|---|---|---|---|
| OVEN | July 30 ± 1.4 day | 4.8 ± 0.9 | 0.851 ± 0.050 | July 3 | 53 day |
| LOWA | July 12 ± 2.4 day | 5.1 ± 1.2 | 1.000 ± 0.065 | June 19 | 45 day |
| GWWA | July 12 ± 2.3 day | 5.9 ± 1.3 | 1.107 ± 0.067 | June 21 | 41 day |
| BAWW | July 21 ± 1.8 day | 4.6 ± 1.1 | 0.988 ± 0.033 | June 27 | 46 day |
| KEWA | August 1 ± 1.8 day | 5.1 ± 0.8 | 0.952 ± 0.050 | July 8 | 48 day |
| COYE | August 21 ± 0.7 day | 4.6 ± 0.9 | 0.935 ± 0.018 | July 28 | 49 day |
| HOWA | August 16 ± 1.2 day | 4.3 ± 1.5 | 0.867 ± 0.030 | July 20 | 52 day |
| AMRE | July 19 ± 0.7 day | 4.9 ± 1.3 | 1.074 ± 0.030 | June 28 | 42 day |
| MAWA | August 9 ± 3.4 day | 4.6 ± 1.1 | 0.935 ± 0.029 | July 15 | 49 day |
| YEWA | July 10 ± 1.3 day | 5.3 ± 1.2 | 1.154 ± 0.055 | June 20 | 39 day |
| CSWA | August 5 ± 2.8 day | 5.0 ± 0.8 | 0.935 ± 0.063 | July 12 | 49 day |
| CAWA | July 29 ± 3.3 day | 5.0 ± 0.0 | 1.109 ± 0.144 | July 8 | 41 day |
| Species alpha code | Midpoint of primary molt (mean ± SE) | Peak primary molt intensity (mean ± SE) | Pace of primary molt (mean ± SE) | Estimated mean onset of primary molt | Estimated mean duration of primary molt |
|---|---|---|---|---|---|
| OVEN | July 30 ± 1.4 day | 4.8 ± 0.9 | 0.851 ± 0.050 | July 3 | 53 day |
| LOWA | July 12 ± 2.4 day | 5.1 ± 1.2 | 1.000 ± 0.065 | June 19 | 45 day |
| GWWA | July 12 ± 2.3 day | 5.9 ± 1.3 | 1.107 ± 0.067 | June 21 | 41 day |
| BAWW | July 21 ± 1.8 day | 4.6 ± 1.1 | 0.988 ± 0.033 | June 27 | 46 day |
| KEWA | August 1 ± 1.8 day | 5.1 ± 0.8 | 0.952 ± 0.050 | July 8 | 48 day |
| COYE | August 21 ± 0.7 day | 4.6 ± 0.9 | 0.935 ± 0.018 | July 28 | 49 day |
| HOWA | August 16 ± 1.2 day | 4.3 ± 1.5 | 0.867 ± 0.030 | July 20 | 52 day |
| AMRE | July 19 ± 0.7 day | 4.9 ± 1.3 | 1.074 ± 0.030 | June 28 | 42 day |
| MAWA | August 9 ± 3.4 day | 4.6 ± 1.1 | 0.935 ± 0.029 | July 15 | 49 day |
| YEWA | July 10 ± 1.3 day | 5.3 ± 1.2 | 1.154 ± 0.055 | June 20 | 39 day |
| CSWA | August 5 ± 2.8 day | 5.0 ± 0.8 | 0.935 ± 0.063 | July 12 | 49 day |
| CAWA | July 29 ± 3.3 day | 5.0 ± 0.0 | 1.109 ± 0.144 | July 8 | 41 day |
Relationship between total primary molt score and ordinal day of the year for 12 species of warblers (A–L) from Powdermill Avian Research Center in southwestern Pennsylvania, 1986–2000. Curves fit to the data are 3-parameter logistic models used to estimate the midpoint of primary molt. Horizontal gray lines represent the midpoint of primary molt (molt score 22.5) and can be used to visualize the day of the year at which the midpoint of primary molt was achieved for each species.
Interspecific variation in the pace of primary molt was significantly dependent on both the timing of molt and daylength at the midpoint of primary molt (Figure 4). Early-molting species generally showed rapid primary molt, while late-molting species replaced their primaries more slowly (Figure 4A). This seasonal decline in the pace of molt appears to be at least partially related to seasonal declines in daylength, as pace of primary molt is positively related to daylength (Figure 4B); it also appears to be proximally mediated by seasonal declines in peak primary molt intensity, as we found a significant positive relationship between pace of molt and the number of primaries being replaced simultaneously during the middle stages of primary molt (Figure 4C). The significant relationships shown in Figure 4 are not confounded by phylogeny; phylogenetic independent contrasts confirmed that the mean pace of primary molt for the 12 species is significantly influenced by the timing of the midpoint of primary molt (t = –4.2, df = 10, P = 0.002), daylength at the midpoint of primary molt (t = 3.8, df = 10, P = 0.004), and the mean number of primaries being replaced simultaneously during the peak of primary molt (t = 3.2, df = 10, P = 0.01).
Relationship between the mean pace of primary molt and (A) the estimated midpoint (day of the year) of primary molt, (B) daylength at the estimated midpoint of primary molt, and (C) peak primary molt intensity—the mean number of primary feathers being replaced simultaneously during the middle stages of primary molt (total primary molt score > 15 and ≤ 35)—for 12 species of warblers from Powdermill Avian Research Center in southwestern Pennsylvania, 1986–2000.
The species-level positive association between pace of primary molt and daylength shown in Figure 4B is also evident at the level of individuals; across all 12 species, and within the 3 species best represented in the dataset (COYE, HOWA, and AMRE), individual birds that molted primaries under short-day conditions in late summer generally replaced them more slowly than did individuals that molted earlier in the summer when days were longer (Figure 5). Furthermore, general linear models suggest that interspecific variation in the pace of primary molt (Figure 4) is largely a consequence of among-species differences in daylength at the time individuals are molting. In simple models in which species is the only factor included, species identity is a significant predictor of the pace of an individual’s primary molt (F11,125 = 2.6, P = 0.003). However, in more complex models in which daylength during molt is included as an additional factor, daylength is a strong positive predictor of the pace of primary molt (F1,124 = 13.6, P < 0.001) but species identity has no significant predictive value (F11,124 = 0.9, P > 0.5; Figure 5A). Similar results are obtained when the analysis is restricted to just the 3 species best represented in the dataset (Figure 5B).
(A) Relationship between the pace of primary molt and daylength for 137 individual warblers of 12 species captured two or more times during primary molt from Powdermill Avian Research Center in southwestern Pennsylvania, 1986–2000. Pace of primary molt for each individual was estimated from the slope of the relationship between primary molt score and day. Daylength was plotted as the mean daylength for the period over which the pace of molt was calculated. In a general linear model, daylength was a significant predictor of pace of molt (F1,124 = 13.6, P < 0.001) but species was not (P > 0.5). (B) Same relationship as shown in (A) but restricted to the three species best represented in the dataset. In a general linear model, daylength was a significant predictor of pace of molt (F1,77 = 8.9, P = 0.004) but neither species nor the species × daylength interaction was a significant predictor of pace of molt (both P > 0.5).
Molt of Secondaries and Tertials
For all 13 species examined, secondary molt began with loss of the first (distal) secondary at approximately the same time as the loss of the 6th primary, and at the same time or shortly after the onset of rectrix molt. Secondary molt proceeded with the sequential loss of secondaries 2–4, followed by the more-or-less simultaneous loss of secondaries 5–6 at the latter stages of primary molt (total primary molt score 36–40). In all 13 species, the 3 tertials were molted earlier and more irregularly, with the middle (2nd) tertial shed in advance of tertials 1 and 3 and at approximately the same time as the loss of the 4th primary.
Intensity of Flight Feather Molt
Overall intensity of flight feather molt in relation to primary molt score is shown for all 1,289 individuals in the dataset in Figure 6. During the peak of flight feather replacement (primary molt score 20–40), a typical molting warbler is growing and replacing 24–32 of its 48 flight feathers simultaneously, including all 12 rectrices and 6–10 remiges on each wing (Figure 6). Depending on the species and the overall pace and duration of molt (Table 2), an individual warbler experiences this peak of molt intensity for 2–3 weeks or more (Figure 6).
Relationship between overall intensity of flight feather molt (mean number of right remiges and rectrices being replaced simultaneously) in relation to primary molt score for 1,289 individuals of 13 species of warblers from Powdermill Avian Research Center in southwestern Pennsylvania, 1986–2000. Molt intensities for primaries, rectrices, and secondaries are indicated separately. Gray reference lines frame the peak of flight feather molt intensity, when individuals were simultaneously replacing 24–32 flight feathers—all 12 rectrices plus 6–10 remiges on each wing.
In comparison to other passerines, primary molt in warblers of eastern North America is exceptionally intense (Figure 7). For all 13 species, we examined, our estimates of average primary molt intensity were greater than all, or all but a few, of the 44 passerine values derived from 35 species reported by Rohwer and Rohwer (2013), including 18 values from 17 migratory species with a post-breeding molt comparable to that of parulid warblers (Figure 7). The analysis of Rohwer and Rohwer (2013) included only one parulid, Lucy’s Warbler (Leiothlypis luciae) from western Mexico.
Box plots showing average primary molt intensity for all passerines (44 values for 35 species) reported by Rohwer and Rohwer (2013, their Supplementary Material Table S2) and for the 13 species of warblers from Powdermill Avian Research Center in southwestern Pennsylvania, 1986–2000. The 18 values (from 17 species) reported by Rohwer and Rohwer (2013) that are most directly comparable to Powdermill warblers are highlighted.
DISCUSSION
Our analysis indicates that simultaneous replacement of rectrices during primary molt is widespread and probably universal in parulid warblers that breed in the northeastern United States. All 13 species we examined showed high and comparable levels of synchrony in replacing their 12 rectrices simultaneously (Figure 1). Furthermore, simultaneous rectrix molt was always initiated in the early-middle stages of primary molt (Figure 2), when birds were already replacing 8–12 primaries (4–6 on each wing; Table 2, Figure 6), an intensity of primary molt that is greater than that of nearly every other passerine for which comparable data are available (Figure 7; Rohwer and Rohwer 2013).
When data on replacement of secondaries and tertials are added (Figure 6), our results indicate that molt of the flight feathers in warblers of eastern North America is extraordinarily intense; nearly all individuals are growing 24–32 (50–67%) of their 48 flight feathers simultaneously for 2–3 weeks or more (Table 2, Figure 6). Flight feather molt of this intensity is likely to entail significant energetic and nutritional burdens (Lindström et al. 1993, Bonier et al. 2007, Echeverry-Galvis and Hau 2012) and profoundly compromise flight capabilities (Swaddle and Witter 1997, Hedenström 2003, Tomotani et al. 2018), foraging performance (Winkler and Allen 1995, Nooker et al. 2005), and the ability of molting birds to provide parental care to dependent offspring (Winkler and Allen 1995, Svensson and Nilsson 1997, Mumme 2018). The period of flight feather molt in North American warblers therefore constitutes a brief but remarkably challenging and poorly understood phase of the full annual cycle (Marra et al. 2015) of these small migratory passerines.
Although we noted only modest interspecific variation in simultaneous rectrix molt and its timing in relation to primary molt (Figures 1, 2), we found substantial interspecific variation in the timing, pace, and duration of primary molt itself (Table 2, Figure 3, Appendix Figure 8). Our estimates of the mean onset of primary molt vary from June 19–21 for the three earliest-molting species (LOWA, YEWA, and GWWA) to July 28 for the latest-molting species (COYE), a range spanning more than 5 weeks (39 days; Table 2). Similar striking variation is evident in our estimates of the pace and duration of primary molt, with duration ranging from 39 days for YEWA to 53 days for OVEN. Interspecific variation in the timing of primary molt undoubtedly reflects differences in migratory and breeding strategies, with early migrants that are typically single-brooded (e.g., LOWA, GWWA, AMRE, YEWA; Mulvihill et al. 2009, Confer et al. 2020, Lowther et al. 2020, Mattsson et al. 2020, Sherry et al. 2020) being well represented among the early-molting species, and late-nesting double-brooders predominant among the late-molting species (e.g., COYE, HOWA, CSWA; Mumme 2018, Byers et al. 2020, Guzy and Ritchison 2020). The interspecific variation in molt timing that we observed is therefore not surprising given interspecific variation in parulid breeding and migration strategies.
On the other hand, our finding that species molting earlier in the summer have a more rapid pace of primary molt and shorter molt durations than species that molt later (Table 2, Figure 4A) was unexpected. This result is counterintuitive, as late-molting migratory species might be expected to be under strong selection to complete molt quickly (e.g., Hall and Fransson 2000, Dawson 2004) to avoid overlap between molt and fall migration. Many previous authors have noted the constraints that long-distance migration may place on the timing and duration of late-summer molt (e.g., Evans Ogden and Stutchbury 1996, Mulvihill et al. 2009, Dietz et al. 2013, Gow and Stutchbury 2013, Kiat and Sapir 2017) and, in that context, our finding that late-molting warblers have what appears to be a more leisurely pace of primary molt (Figure 4A) presents a paradox. However, our observation that the pace of primary molt is positively related to daylength (Figures 4B, 5) provides a novel potential solution to the paradox, one that to our knowledge has not been previously proposed. It suggests the hypothesis that for early-molting individuals at high latitudes, long days increase the amount of foraging time available, thereby allowing molting birds to secure more food, increase molt intensity (Figure 4C), and shorten molt duration. For late-molting individuals, on the other hand, shorter late-summer days may mandate reduced daily food intake, lower molt intensity, a slower pace of molt, and longer molt duration. Although this is an attractive hypothesis, our data also suggest that seasonal changes in foraging time probably cannot completely explain the positive association between the pace of molt and daylength. For example, we found that the pace of primary molt for the early-molting YEWA was 23% faster than that of the late-molting COYE (Table 2), but daylength at the midpoint of primary molt was only 11% longer (Figure 4B). Similarly, the linear regression equation shown in Figure 5A predicts that a 5% increase in daylength would on average produce about a 12% increase in an individual’s pace of primary molt. Thus, it is likely that factors besides seasonal changes in foraging time may also contribute to the positive relationship we observed between daylength and pace of primary molt. We can suggest three possibilities.
First, slower late-season molt could be explained by declining prey abundance. Late-season food declines on the breeding grounds have been implicated in the evolution of molt-migration in passerines breeding in arid habitats in western North America (Rohwer et al. 2005, Pageau et al. 2020), and some evidence suggests that late-season prey availability may be limiting to warblers in eastern North America as well (Nagy and Holmes 2005).
Second, warblers molting later in the season may molt slowly if they redirect energetic resources from molt to fat deposition and increased body mass in preparation for fall migration (Morton and Welton 1973, Lindström et al. 1994). However, a late-season increase in body mass of molting birds occurred in only one (COYE) of our 13 study species, suggesting that pre-migratory fueling in Powdermill warblers generally occurs after molt of the flight feathers has been completed.
Third, slow late-season molt may be explained by the demands of late-season parental care. For example, late-molting HOWAs in northwest Pennsylvania are nearly always engaged in post-fledging parental care, whereas the earliest molters typically have no parental responsibilities (Mumme 2018). If parental care delays the onset of molt similarly in other warbler species (e.g., Mulvihill et al. 2009), late molters may be parents unable to sustain a rapid pace of molt while also caring for dependent young. Experimental work on captive Zebra Finches (Taeniopygia guttata) has shown that molt-breeding overlap can substantially reduce the pace of molt (Echeverry-Galvis and Hau 2012), so this hypothesis is worthy of future investigation.
Birds can accelerate the pace of molt by increasing either: (1) the growth rate of individual feathers, or (2) molt intensity, the number of feathers being simultaneously replaced at a given moment in time (de la Hera et al. 2011, Rohwer and Rohwer 2013). Our study is silent on the first mechanism, as the molt scoring system we employed lacks the precision needed to accurately assess growth of individual flight feathers. However, we found strong support for the second mechanism; molt intensity was a significant predictor of species-level differences in the pace of primary molt in warblers (Figure 4C). In a recent literature review and comparative analysis, Rohwer and Rohwer (2013) estimated that 60% of the variation in molt duration reported in the literature is attributable to variation in molt intensity, with only 4.4% attributable to variation in feather growth rates. Peak and average primary molt intensity for the 13 species of warblers in our study (Table 2, Figure 7) were much greater than the range of average intensities (1.1–3.8) reported for passerines (figure 3 and table S2 of Rohwer and Rohwer 2013), but consistent with high primary molt intensities (5.2–6.3) reported for Yellow Warblers (YEWA) at high latitudes in North America (Ryder and Rimmer 2003).
Our study also highlights the particular set of challenges facing warblers that delay the onset of primary molt until mid-July or later (e.g., COYE, HOWA, MAWA; Table 2). For these birds, short late-summer days—and perhaps other factors associated with short late-summer days, such as reduced food availability, pre-migratory fueling, and late-season parental care—appear to mandate reduced molt intensity, a slower pace of molt, and a longer molt duration (Figures 4 and 5), potentially constraining late-season nesting, parental care, and the timing of fall migration. The option of increasing molt intensity and reducing the duration of molt simply may not be available to individual warblers molting in late summer, and the life-history consequences of that constraint merit further study.
The main question left unanswered by our work is whether the high-intensity flight feather molt that we have documented for warblers molting at 40°N latitude occurs throughout North America, or if molt intensity varies latitudinally. One possibility is that high-intensity molt, including simultaneous rectrix molt concurrent with intense molt of multiple remiges, is a constraint imposed on warblers by long-distance migration and breeding at high latitudes. To our knowledge only one previous study has investigated latitudinal variation in warbler molt; Ryder and Rimmer (2003) compared the molt of YEWA from 44°N (Vermont) and 51°N (Northern Ontario), finding that primary molt in Ontario birds was more intense and of shorter duration than it was for Vermont birds, but the differences were generally modest and not statistically significant. The average intensity of primary molt for Lucy’s Warbler at latitude ~24°N in western Mexico was 2.34 (Rohwer et al. 2009, Rohwer and Rohwer 2013), ~60% the intensity of Powdermill warblers at 40°N (Figure 7). Additional studies, particularly of species with broad latitudinal ranges (e.g., COYE, AMRE, YEWA) and including data from low-latitude (≤30°N) locations and the subtropics, would be particularly valuable in addressing this question. Studies following marked individuals through their breeding, molt, and pre-migratory periods would be especially valuable, and would help clarify how molt is integrated into the annual cycle across a latitudinal gradient. A detailed study of molt in nonmigratory tropical warblers (e.g., species in the genera Myiothlypis, Basileuterus, and Myioborus; Curson et al. 1994, Lovette et al. 2010) would also be valuable, as these tropical species would be expected to have a much slower pace of flight feather molt (e.g., Rohwer and Wang 2010, Johnson et al. 2012, Silveira and Marini 2012, Moreno-Palacios et al. 2018), with reduced intensity of primary molt and rectrix molt that is less synchronous and more staggered.
In summary, our study has shown that flight feather molt of warblers of the northeastern United States is extraordinarily intense, with simultaneous molt and replacement of all 12 rectrices broadly overlapping the concurrent replacement of 12–20 remiges (6–10 on each wing). Simultaneous replacement of 50–67% of the flight feathers is likely to compromise flight capabilities, foraging performance, and predator avoidance while also imposing significant energetic and nutritional burdens to support a large number of growing feathers. The period of flight feather molt, therefore, constitutes a brief but unusually demanding phase of the annual cycle of migratory warblers, one that merits increased attention from ornithologists seeking a more thorough understanding of that cycle.
Appendix
Trajectories of total primary molt score for 137 individual birds of 12 species of warblers (A–L) captured two or more times in the same season during primary molt from Powdermill Avian Research Center in southwestern Pennsylvania, 1986–2000. These data were used to estimate the mean pace of primary molt (change in primary molt score per day) for each species.
ACKNOWLEDGMENTS
The expertise and enthusiasm of the late K. C. Parkes in all matters of avian molt were instrumental in making this study possible. We thank R. C. Leberman, A. Leppold, M. Niedermeier, and many volunteer banding assistants at Powdermill Avian Research Center for help with collecting and coding the data used in this study. Three anonymous reviewers provided very helpful and constructive comments on an earlier draft of the manuscript.
Funding statement: Financial support for data analysis and manuscript preparation was provided by the Academic Support Committee of Allegheny College, which had no input on the content of the manuscript.
Ethics statement: The data reported here were collected as part of long-term bird-banding operations of Powdermill Avian Research Center, and authorized under a series of banding permits from the U.S. Fish and Wildlife Service and the Pennsylvania Game Commission.
Author contributions: R.S.M. and D.N. collected, curated, maintained, and provisionally analyzed the molt data. R.L.M conceived the project, fully analyzed the data, and prepared an initial draft of the manuscript. All three authors contributed to editing and revising the manuscript.
Data depository: Analyses reported in this article can be reproduced using the data provided by Mumme et al. (2021).
LITERATURE CITED
