Hydrogen isotope analysis of feather keratin (δ2HF) has become an essential tool for tracking movements between breeding and wintering populations of migratory birds. In particular, δ2HF has been used to create δ2HF isoscapes that can be used to assign the geographic origins of molt. The majority of past studies have sampled a portion of a single feather as an isotopic proxy for the entire plumage although surprisingly little is known about variation of stable isotopes within and between feather tracts of individuals in local populations. Here we examine δ2HF variation in 24 pterylographic variables (9 primaries, 6 secondaries, 6 rectrices, and 3 patches of ventral contour feathers) in individual specimens of black-throated blue warbler (Setophaga caerulescens) breeding in the Big Santeetlah Creek watershed (5350 ha), southern Appalachian Mountains. By restricting our study to territorial ASY males (after second year) inhabiting a small watershed, we could focus on δ2HF variation generated during the complete prebasic annual molt in a circumscribed population while factoring out age and sexual differences in foraging behavior, isotopic incorporation, and post-breeding dispersal. Summed within-individual variation (δ2HF) across 24 pterylographic variables ranged from 12 to 60‰ (= 21.8 ± 9.4‰), with 81% of the individuals exhibiting variation ≥ 16‰ (reproducibility of analyses was ≤ 4 ‰). Adjacent feathers in feather tracts tend to have more similar δ2HF values than feathers grown weeks apart, consistent with the stepwise replacement of flight feathers. The pooled population sample exhibited significant δ2HF variation in primaries (-78 to -21‰), secondaries (-80 to -17‰), rectrices (-78 to -23‰), and ventral contour feathers (-92 to -32‰). Strong year effects in δ2HF variation were observed in each of the 24 pterylographic variables. Altitudinal effects were observed only in ventral contour feathers. The current findings demonstrate that within-individual variation (δ2HF) may be much greater than previously thought in migratory species that molt on or near breeding territories. Our study also highlights the need for greater pterylographic precision in research design of isotope-based studies of avian movement. Within-individual and within-population δ2HF variation should be incorporated in geographic assignment models. In a broader context, the staggered Staffelmauser pattern of molt in wood warblers provides an exceptional view of the seasonal variation of hydrogen isotopes circulating in blood plasma during the six-week period of annual molt.
Citation: Graves GR, Newsome SD, Fogel ML (2018) Stable hydrogen isotope variability within and among plumage tracts (δ2HF) of a migratory wood warbler. PLoS ONE 13(4): e0193486. https://doi.org/10.1371/journal.pone.0193486
Editor: Claudia Mettke-Hofmann, Liverpool John Moores University, UNITED KINGDOM
Received: July 11, 2017; Accepted: February 12, 2018; Published: April 3, 2018
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Data Availability: Raw data are appended as S9 Table in Supporting Information.
Funding: This study was supported by Alexander Wetmore Fund, Smithsonian Institution, to GRG; Smoketree Trust (private financial support), to GRG; National Science Foundation (NSF) grant DEB-1343015, to SDN; National Science Foundation (NSF) grant DEB-1437845, to MLF.
Competing interests: The authors have declared that no competing interests exist.
Naturally occurring stable isotopes of several light and heavy elements (i.e., H, C, O, S, and Sr) of feather keratins have been used singly or in combination to infer the foraging ecology, dispersal, and migratory connectivity of birds [1–9]. Hydrogen isotope analysis (δ2H) has figured prominently because continental isoscapes for precipitation (δ2HP) exhibit strong latitudinal gradients in North America and Europe [10–12] that are propagated through local surficial waters and food sources, which in turn are assimilated by birds and used to synthesize tissues [13, 14]. In theory, spatial variation in the isotope signatures of avian tissues should closely map the underlying isoscapes of assimilated drinking water and diet, although there are still several important unknowns such as the fluctuating variation of δ2H in diet and water available to local populations [14–16].
For species whose molt schedules are well understood, feather keratin is an ideal tissue for δ2H analysis because feathers are metabolically inert after a brief growth period (a few weeks). The isotopic composition of feathers (δ2HF) is fixed, with the exception of potentially exchangeable hydrogen that represents ~8–15% of the hydrogen budget in tissues [8, 17–19]. Several recent studies have shown that δ2HF values for species that molt on the breeding grounds in North America and Europe typically exhibit strong correlations with amount-weighted, growing season δ2HP [9, 20]. As a consequence, some investigators have made biological and geographical interpretations of relatively small differences in δ2HF values (<10‰) observed among individuals and populations [21, 22]. Others have questioned whether δ2HF values differing by <9‰ can be confidently assigned to different geographical locations [23, 24].
Pterylography is the study of feather arrangement and feather tracts of birds. Songbirds are typically plumaged with 1,000–3,000 feathers . The majority of δ2HF studies sample a single feather, typically a sub-section of a feather, as an isotopic proxy for the entire plumage of a bird [2, 22, 26–30]. Does the choice of feather have an effect on isotope-based models of movement? A few studies have made limited investigations of δ2HF variation within individual feathers, variation among feathers within feather tracts, or variation among different feather tracts sampled from the same individual [1, 23, 26, 31–33]. Analyses that strive for an “apples to apples” comparison of individuals and populations sample the same feather (e.g., third rectrix) from every individual. This is an implicit acknowledgment that δ2HF values may vary within and among plumage tracts and that the initiation date of molt may have an effect on δ2HF [23, 34, 35]. In contrast, studies that pool data from different feathers within a feather tract or from feathers sampled from different feather tracts make the explicit assumption that within-individual δ2HF variation is small compared to δ2HF variation observed among individuals and among populations [1, 8, 30, 36].
Many studies of δ2HF were conducted before it was recognized that temporal variability in in the isotopic composition of drinking water and diet might affect δ2HF values of different feathers sampled from an individual. Studies of wood warblers, for example, have pooled δ2HF data from contour (body) feathers and rectrices [1, 8], from contour feathers and primaries , from retrices and crown feathers , from the innermost and outmost rectrices , and from primaries and rectrices [20, 36]. In some cases, investigators have failed to specify the pterylographic origin of sampled feathers [6, 34, 38] or have vaguely referred to flight feathers , or have analyzed an unspecified primary  or rectrix [22, 26, 29, 39–41]. The extent to which pterylographic imprecision and the pooling of data from different feather tracts weakens or obfuscates research findings remains to be determined.
The δ2HF profile of the black-throated blue warbler (Setophaga caerulescens) is among the best studied of North American songbirds [1, 5, 8, 42–44]. This small insectivorous warbler (body mass, 9–10 g) breeds in mixed deciduous-coniferous forests in eastern North America and winters primarily in the Greater Antilles [45–47]. Older males (ASY, after second year) show a high degree of year-to-year fidelity to breeding sites . ASY males sampled from breeding populations exhibit significant latitudinal and longitudinal δ2HF variation in ventral contour feathers [1, 8]. δ2HF was highest in samples from the southern end of the breeding range in Georgia and lowest near the northern margins of its range in Ontario. δ2HF also increased systematically from west (Michigan) to east (New Brunswick). The influence of altitude on δ2HF has not been systematically studied [1, 5, 8]. In the southern Appalachian Mountains, ASY males are believed to forage within a restricted altitudinal zone centered on breeding territories not only during the early part of the breeding season  but also during molting period  as inferred from the altitudinal lapse rates (the rate at which isotope values vary with altitude) of δ13C values observed in liver, pectoral muscle, and feather keratin.
Fidelity to a relatively small breeding territory during molt should result in constrained variation of δ2HF among feather tracts within individuals based on the premise that δ2H of food and water resources will be less variable in a narrow geographic or altitudinal zone than in a wider one. If flight feather molt begins on breeding territories and ends while individuals are engaged in post-breeding wandering to higher or lower altitudes or during migration to their wintering grounds, then δ2HF values of ASY males could exhibit chaotic or staggered variation within feather tracts. Staggered variation in δ2HF values within feather tracts could also result from seasonal changes in δ2HP within territories during the course of an extended molt period. In any case, knowledge of within- and between-individual variation of δ2HF values within circumscribed breeding populations is critical for evaluating the application of hydrogen isotopes in population “connectivity” studies .
In this paper, we investigate δ2HF variation in 24 pterylographic variables (9 primaries, 6 secondaries, 6 rectrices, and 3 patches of ventral contour feathers) in the plumage of individual ASY black-throated blue warblers from a small watershed in the southern Appalachian Mountains. We characterize (i) δ2HF variation within and between feather tracts in individuals; (ii) examine between-individual variation within the Big Santeetlah Creek breeding population; and (iii) document strong year effects in δ2HF. Finally, we address the implications of δ2HF variation for connectivity studies of breeding and wintering populations of songbirds.
The study was conducted in western North Carolina in the Big Santeetlah Creek watershed (35° 21' N; 84° 00' W), nestled in the highest subsidiary mountain range in the Appalachian Mountains south of the Little Tennessee River [42, 43, 46, 49–52]. The watershed (680–1689 m above sea level; 5350 ha) is embedded in the largest contiguous tract of montane forest in eastern North America and supports one of the most diverse woody floras north of Mexico [53, 54]. Forestry practices since the 1920's have resulted in a mosaic of relict old-growth trees and even-aged stands of hardwood forest.
The black-throated blue warbler is a common breeding bird between 850 m and 1450 m in forests of the southern Appalachians [46, 55, 56]. Males return to breeding territories (0.75–3.0 ha) at the lower altitudinal limit of their breeding range as early as 1 May but settlement at higher altitudes (> 1400 m) may be delayed by 2–3 weeks.
Molt patterns of the flight feathers (primaries, secondaries, rectrices) and contour plumage in the black-throated blue warbler have not been studied in detail, but like most Nearctic-Neotropical members of the genus Setophaga that breed in eastern North America, breeding birds are thought to undergo a complete prebasic molt (including all flight feathers) on or near their breeding territories before fall migration [57–59]. A supplementary pre-alternate molt restricted to the replacement of feathers on the head and throat occurs on the wintering grounds before spring migration . In principle, δ2HF values observed in the flight feathers of territorial birds sampled during the current breeding season (year T) reflect the δ2H of the food and water ingested during June, July, and August of the previous breeding season (year T-1). This property and the continental δ2HP isoscape make it possible to infer the connectivity of breeding and wintering populations [1, 8].
First time breeders (SY, second year, first alternate plumage) can be distinguished in the field from older male breeders (ASY, after second year, definitive alternate plumage) by their dull primary coverts and olive margins on the alula coverts . We restricted our study to territorial ASY males because older birds exhibit higher site fidelity from year to year . In other words, we sampled individuals in their second or later breeding season which were likely to have bred in the Big Santeetlah Creek watershed during the previous year. Annual survivorship of males returning to breeding territories is roughly 50% . By omitting females and SY males (which retain flight feathers from the juvenal plumage), we were able to focus on δ2HF variation generated during the molting period while factoring out age and sexual differences in foraging behavior, isotopic incorporation, and post-breeding dispersal.
All feathers of adult black-throated blue warblers are replaced during the annual molt cycle. The principal wing feathers (Fig 1), primaries (P1–P9) and secondaries (S1–S9), are part of the alar plumage tract. Both primaries and secondaries are numbered respectively from the middle of the wing . Primary molt is thought to progress in the common Staffelmauser pattern , commencing with the replacement of the first primary (P1) in the middle of the wing and followed in staggered fashion by the remaining primaries to the wingtip [57, 58]. The duration of primary molt (from loss of P1 to the full renewal of P9) is unknown in this species but it ranges from 38 to 42 days in the prairie warbler (Setophaga discolor) . Molt of the secondaries commences with S1 in the middle of the wing and progresses toward the body , except for the three innermost secondaries (S4-S6) adjacent to the body, which may molt out of sequence [57, 58]. The rectrices (R1–R6), which constitute the most important component of the caudal plumage tract, are numbered outward from the center of the tail. Molt commences with the innermost rectrix (R1) and progresses rapidly outward. Molt of ventral contour feathers (upper breast, breast, lower flank) coincides with the replacement of the flight feathers.
The Smithsonian IACUC approved the research protocol. Specimen collection permits were granted by the U.S. Fish & Wildlife Service, U.S. Forest Service (Department of Agriculture), and North Carolina Wildlife Resources.
Territorial ASY males were collected 10–12 June 2013 (n = 15) and 16–17 June 2014 (n = 17) with a small bore shotgun under state and federal permits for multiple research purposes along an altitudinal transect that extended from 720 m-1540 m (above sea level). GPS coordinates were recorded at each collecting site. Altitude of collection sites was subsequently determined with Google Earth Pro. The mean altitudinal distribution of specimens in 2013 (range, 855–1491 m) and 2014 (range, 856–1394 m) was similar (ANOVA, F1,30 = 0.78; P = 0.38). Males collected in this study responded to playback of recorded songs and exhibited territorial behavior. Females were observed on many of the territories, but we did not systematically search for females or nests. The sampling period (10–17 June) coincided with the overlap of the fledging period for the first brood and nest building and egg laying for the second brood. Annual standardized censuses along a 14.5 km altitudinal transect indicated that the removal of males in the watershed had no demonstrable year-to-year effect on the census population.
Specimens were packaged in multiple layers of insulating tissue paper and aluminum foil and frozen within 30 minutes of death in liquid nitrogen. Voucher specimens (flat skins, partial skeletons, tissues, stomach contents) were deposited in the research collection of the National Museum of Natural History, Smithsonian Institution (USNM 647334–647338, 647340, 647342–647350, 648884–648885, 648887–648889, 648891–648901, and 648903).
We removed a section of vane (10–12 mm) from the middle of each primary (P1–P9), secondary (S1–S6), and rectrix (R1–R6) from the left or right side (but not both) of each specimen. The feather rachis was not included owing to the potential for δ2H enrichment of the rachis compared to the feather vane . Given average growth rates of rectrices (3 mm/day) observed in the prairie warbler , our vane samples from flight feathers represent 3–4 days of growth during molt. Small contour feathers were sampled from three 5 × 5 mm patches (V1 = upper breast, V2 = breast, V3 = lower flank) on the ventral plumage tract from one side of each specimen; both vane and rachis were included in contour feather samples. All samples were washed in 2:1 chloroform:methanol to remove surface contaminants, rinsed with distilled water, and dried at 60°C. Prior to weighing, cleaned feather samples were held at the University of California Merced Stable Isotope Laboratory for at least three weeks to equilibrate possible exchangeable hydrogen. Then, approximately 0.1–0.2 mg of each sample was homogenized with surgical scissors and loaded into a 3x5 mm silver capsule for δ2H analysis. Encapsulated samples and keratin reference materials with known non-exchangeable δ2H values (see below) were kept in a drying oven at 45°C under a steady stream of N2 until analysis (see below).
Stable isotope analysis
Stable isotope data are expressed as δ values, δ2H = 1000* [(Rsample− Rstandard/Rstandard)], where Rsample and Rstandard are the 2H/1H of the sample and standard, respectively. The internationally accepted standard for δ2H is Vienna Standard Mean Ocean Water (V-SMOW) and the units are expressed as parts per thousand or per mil (‰). δ2H values were measured with a Thermo Scientifc thermal conversion/elemental analyzer (TC/EA) operated at 1430°C (Thermo Fisher Scientific Inc., Waltham, MA, USA) and coupled to a Thermo Scientific Delta V Plus isotope ratio mass spectrometer at the University of California, Merced (Merced, CA, USA).
Approximately 8–15% of the organically bound hydrogen in proteins can potentially exchange with hydrogen in atmospheric water vapor at ambient temperatures [18, 19]. To correct for exchangeable hydrogen in unknown samples, we used two feather keratin internal reference materials for which the δ2H composition of the non-exchangeable hydrogen fraction was determined via comparative equilibration experiments and statistical analyses identical to those described in . δ2H values of the non-exchangeable fraction of hydrogen in these keratin reference materials were -55‰ (turkey feathers) and -95‰ (chicken feathers), which brackets most of the δ2H values of the warbler feathers we analyzed. Analyses of these two keratin reference materials, two oil reference materials, and pure stearic acid were used to determine the within-run precision (standard deviation) and accuracy of analyses, which was determined to be ≤ 4 ‰ relative to V-SMOW.
Variation of δ2HF among individuals
We subjected each of the 24 pterylographic variables to a separate analysis of covariance (ANCOVA) test for year (2013 and 2014) and altitude effects on δ2HF using the GLM (general linear model) module in SYSTAT Version 12. The alpha level (σ < 0.05) was adjusted for the number of simultaneous tests (0.05/24 = 0.002). We found differences in mean δ2HF values between years, which are reported as Δ (equal to the mean of 2013 values minus the mean of 2014 values rounded to the nearest whole number). Mean δ2HF values were higher in 2013 than in 2014 for all feather types (P < 0.0001; Fig 2): primaries (Δ = 14–28‰), secondaries (Δ = 17–21‰), rectrices (Δ = 15–19‰), and ventral contour feathers (Δ = 17–23‰). We separated the data by year in subsequent analyses. Only V1 (upper breast, ventral contour feathers) of the 24 pterylographic variables showed noticeable altitude effects in δ2HF values (P = 0.013).
Box plots depict minimum values, first quartile, median, third quartile, and maximum values observed among population samples for primaries (P1-P9), secondaries (S1-S6), rectrices (R1-R6), and ventral contour feathers (V1-V3). Data are presented by year (2013 and 2014).
The cumulative range of δ2HF values observed across all primaries (P1– P9) varied from -67 to -16‰ in 2013 and from -78 to -54‰ in 2014 (Fig 2, S1 Table). In 2013, mean δ2HF gradually decreased from P1 (-40‰), the first primary to molt, to P9 (-49‰) at the wingtip. The opposite pattern was observed in 2014 when mean δ2HF increased slightly from P1 (-67‰) to P9 (-63‰). In secondaries (S1–S6), δ2HF ranged from -68 to -17‰ in 2013 and from -80 to -53‰ in 2014 (S2 Table). δ2HF values for rectrices (R1–R6) varied from -66 to -23‰ in 2013 and from -78 to -44‰ in 2014 (S3 Table). δ2HF values for ventral contour feathers (V1–V3) varied from -92 to -32‰ in 2013 and from -92 to -47‰ in 2014 (S4 Table).
Pearson correlation coefficients (r) for δ2HF values of pairwise combinations of primaries showed significant variation in 2013 (r = 0.26–0.92) and 2014 (r = 0.19–0.86)(S5 Table). In 2013, correlations involving P1 and other primary feathers were relatively high (r > 0.5). In 2014, P1 correlations decreased precipitously from the middle of the wing toward the wingtip. Correlations for pairwise combinations of secondaries (S6 Table) also exhibited considerable variation in 2013 (r = 0.64–0.97) and in 2014 (r = 0.06–0.92). Pairwise correlations for rectrices were similarly varied in 2013 (r = 0.59–0.92) and 2014 (r = 0.20–0.83)(S7 Table). Pairwise correlations for ventral contour feathers exhibited a comparatively narrow range of coefficients in 2013 (r = 0.66–0.79) and 2014 (r = 0.39–0.59)(S8 Table). Overall, δ2HF values for adjacent feathers within feather tracts exhibited higher correlation coefficients than did pairs of feathers whose growth periods overlapped less or not at all (e.g., P1 and P9)(S9 Table).
Variation of δ2HF within individuals
Within-individual variation of δ2HF exceeded analytical precision (± 4‰) in one or more of the three feather tracts (alar, caudal, ventral) in all specimens sampled in 2013 and 2014 (Fig 3). The range of δ2HF values (δ2HF maximum - δ2HF minimum) observed in primaries sampled from individual birds was substantial in 2013 (8–31‰) and in 2014 (4–21‰). The range of δ2HF observed in secondaries was also substantial in 2013 (2–26‰) and in 2014 (4–22‰). Fewer individuals exhibited δ2HF variation in rectrices that exceeded analytical precision, but δ2HF still varied extensively in 2013 (4–25‰) and in 2014 (3–26‰). The range of δ2HF observed in ventral contour feathers was also considerable in 2013 (1–30‰) and in 2014 (2–27‰). Summed across all 24 pterylographic variables, the range of within-individual variation in δ2HF was large in 2013 (14–60‰; = 23.7 ± 11.7‰) and in 2014 (12–33‰; = 20.2 ± 6.7‰)(Fig 4). The range of δ2HF values observed among the pooled flight feathers (primaries, secondaries, and rectrices) within individuals was uncorrelated (P > 0.05) with altitude of the collecting site.
Individuals that exhibited a range of δ2HF values > 8‰ (twice the maximum range of analytical variability observed within runs for laboratory reference materials) within the selected feather tracts are shown in red. Upper values for each bin are shown on the x-axis.
Our study provides the first fine-grained view of within-individual δ2HF variation in the plumage of wild songbirds. Perhaps the most significant finding was the substantial δ2HF heterogeneity that occurs within and between feather tracts of individual black-throated blue warblers. For example, the range of δ2HF values observed across primaries (P1–P9) varied from 4 to 31‰ (= 15 ± 6‰). The range of within-individual variation observed across all 24 pterylographic variables (Fig 4) varied from 12 to 60‰ (= 22 ± 9‰), with 81% of the individuals exhibiting ranges ≥ 16‰. Given the paucity of published data on within-individual heterogeneity (δ2HF), it is impossible to say whether the δ2HF variation observed in black-throated blue warblers is unusual for a songbird species that molts on the breeding grounds. Comparable δ2HF data from a small collection of Swainson’s thrush (Catharus ustulatus; n = 9) collected in Quebec  exhibited within-feather heterogeneity of 2 to 17‰ (= 5 ± 5‰) based on three subsamples from a single flight feather per specimen. If the sampling had been expanded to include feathers from other feather tracks, the cumulative within-individual variation would likely have been larger.
The general pattern that emerged in the correlation matrices of pairwise combinations of feathers in the current study was that δ2HF values of adjacent feathers within a feather tract tended to be more similar than those of randomly chosen pairs of feathers. This observation is consistent with the stepwise replacement of flight feathers in wood warblers [58, 59]. Feathers grown at or near the same time tend to have more similar δ2HF values than feathers grown weeks apart. This pattern and the more general observation of within-individual heterogeneity (δ2HF) could be caused by several covarying factors: (i) seasonal variation in δ2HP  propagated through drinking water and the insect diet of warblers during the 6-week molting period; (ii) physiological variation in isotopic discrimination linked to variation in dietary macromolecular content, the contribution of metabolic water to tissue δ2H, and the potential for isotopic routing during molt and pre-migratory conditioning (iii) seasonal variation in heat stress and evaporative cooling in breeding and molting individuals resulting in isotopic enrichment of body water [33, 63]; (iv) local post-breeding wandering  in isotopically heterogeneous habitats when molt is still active; or (v) replacement of feathers lost accidentally during migration or on the wintering grounds rather than during the summer breeding season [58, 65].
Lacking data on the location, movement, diet, and drinking water of individuals during the molting period, we are unable to identify the exact causes of within-individual heterogeneity of δ2HF in our breeding population sample. Suffice to say, within-individual heterogeneity of δ2HF was uncorrelated with the altitude of the individual’s territory at the time of collection. Additionally, we observed only a single individual in which δ2HF values of adjacent primaries or rectrices (the feather tracts most frequently analyzed in previous studies) differed to an extent (> 20‰) that suggested the loss and regrowth of a feather during migration or on the Antillean wintering grounds (see Fig 2 in reference 34).
A second prominent finding was that the range of population-level heterogeneity of δ2HF (76‰) observed in the Santeetlah Creek watershed across the 24 pterylographic variables (n = 32 specimens) was similar in extent to δ2HF variation in ventral contour feathers of black-throated blue warblers collected at multiple breeding locations in the Appalachians from 35° to 44° N [1, 8]. Basiolli (2008) observed a slightly smaller range of within-population heterogeneity (63‰) in rectrices of black-throated blue warblers that returned to breeding territories in New Hampshire. Basiolli’s data (n = 43 individuals) included three outliers that were thought to have molted south of the breeding grounds . Collectively, these data indicate that within-population δ2HF variation contributes a substantial fraction of the total variance in δ2HF observed in black-throated blue warblers across its geographic range. Causes of population-level variation would include the roster of factors previously mentioned for within-individual heterogeneity as well as tangible differences in the water and diet resources present in non-overlapping breeding territories in both local populations and in geographically separated populations.
A third significant but perhaps less surprising finding was a strong year effect in δ2HF in each of the 24 pterylographic variables. δ2HF values averaged 20 ‰ higher in 2013 than in 2014 (S1 Table), suggesting a systemic environmental cause for the difference. Strong year effects in carbon isotopes (δ13C) of feather keratin was reported by , as well as pectoral muscle and liver δ13C values  collected from black-throated blue warblers in the same watershed . Whatever the cause, significant year-to-year fluctuations in δ2HF and δ13CF within a local watershed indicate that temporal effects clearly influence analyses of continent-wide isotope gradients in feather keratin if geographically dispersed populations are sampled in different years. Previous geographic studies of δ2HF variation in black-throated blue warblers did, in fact, pool population samples collected during several breeding seasons [1, 8]. The collection of specimens at a number of widely separated locations within the warbler’s extensive breeding range during a single (brief) breeding season would be logistically difficult. Both of these studies were completed years before it was widely recognized that annual variation in δ2HP could override the long-term latitudinal variation in δ2HP signal that is used to discriminate among local bird populations via δ2HF analysis of growing feathers [35, 66–68]. Our study adds additional support to recommendations that year effects should be factored into geographic assignment models .
We draw several conclusions from the exceptional δ2HF variability observed in the Santeetlah Creek population of black-throated blue warblers. First, accuracy of geographic assignment tests in studies of migratory connectivity hinges on the accuracy of transfer functions to convert δ2Hp into δ2HF . Transfer functions, which implicitly incorporate a variety of abiotic, physiological, and ecological processes are ideally estimated by documenting the hydrogen isotope discrimination factor between feather keratins and local precipitation at each of many known molting areas for each study species. In practice, species-specific transfer functions were traditionally based on the intercept of linear regression of δ2HF on δ2Hp of the pooled sample collected at many locations [6, 9]. More recent models have incorporated data on foraging guild, foraging substrate, nesting substrate, diet, and age class among other variables [20, 35]. Conspicuously absent from the latest iterations of both transfer functions and geographic assignment models is the explicit consideration of population-specific data on within-individual heterogeneity of δ2HF, the observed variation within and between feather tracts of individual birds.
Second, the finding of high levels of within-individual variation in δ2HF in black-throated blue warblers emphasizes the necessity of homologous sampling when studies aim to compare δ2HF variation among populations but are limited to a single feather sample per individual. Prebasic molt is a prolonged process that requires up to six weeks to complete in wood warblers. Based on our data, the first and last feathers grown during a single molt cycle on the breeding grounds would in most individuals be classified as originating from molting sites that are separated by several degrees of latitude using the stringency benchmarks commonly employed in δ2HF studies of avian movement [21, 22, 26, 38, 69]. In any event, we recommend the incorporation of within-individual variation of δ2HF in geographic assignment models. In a broader context, the Staffelmauser pattern of molt in wood warblers provides a unique window on the seasonal variation of hydrogen isotopes circulating in blood plasma during the six week period of annual molt on the breeding grounds.
S1 Table. Summary statistics for δ2HF values for primaries (P1-P9).
S2 Table. Summary statistics for δ2HF values for secondaries (S1-S6).
S3 Table. Summary statistics for δ2HF values for rectrices (R1-R6).
S4 Table. Summary statistics for δ2HF values for ventral contour feathers (V1-V3).
S5 Table. Pearson correlation coefficients for δ2HF values for pairwise combinations of primaries (P1-P9).
S6 Table. Pearson correlation coefficients for δ2HF values for pairwise combinations of secondaries (S1-S6).
S7 Table. Pearson correlation coefficients for δ2HF values for pairwise combinations of rectrices (R1-R6).
S8 Table. Pearson correlation coefficients for δ2HF values for pairwise combinations of ventral contour feathers (V1-V3).
We thank Brian Schmidt and Chris Milensky for preparing specimens. The authors thank David Araiza and Christina Bradley for laboratory support, sample preparation, and analysis at the University of California Merced, and Chris Romanek and an anonymous reviewer for comments on the manuscript. Jacqueline Mask prepared the schematic illustration of flight feather topography.
- 1. Rubenstein DR, Chamberlain CP, Holmes RT, Ayres MP, Waldbauer JR, Graves GR, et al. Linking breeding and wintering ranges of a migratory songbird using stable isotopes. Science. 2002;295:1062–5. pmid:11834833
- 2. Sellick MJ, Kyser TK, Wunder MB, Chipley D, Norris DR. Geographic variation of strontium and hydrogen isotopes in avian tissue: implications for tracking migration and dispersal. PLoS One. 2009;4:e4735. pmid:19266102
- 3. Wassenaar LI, Hobson KA. Stable-carbon and hydrogen isotope ratios reveal breeding origins of red-winged blackbirds. Ecological Applications. 2000;10:911–6.
- 4. Kelly JF, Johnson MJ, Langridge S, Whitfield M. Efficacy of stable isotope ratios in assigning endangered migrants to breeding and wintering sites. Ecological Applications. 2008;18:568–76. pmid:18488617
- 5. Royle JA, Rubenstein DR. The role of species abundance in determining breeding origins of migratory birds with stable isotopes. Ecological Applications. 2004;14:1780–8.
- 6. Hobson KA, Wassenaar LI. Linking breeding and wintering grounds of neotropical migrant songbirds using stable hydrogen isotopic analysis of feathers. Oecologia. 1997;109:142–8.
- 7. Wunder MB, Norris DR. Improved estimates of certainty in stable-isotope-based methods for tracking migratory animals. Ecological Applications. 2008;18:549–59. pmid:18488615
- 8. Chamberlain CP, Blum JD, Holmes RT, Feng X, Sherry TW, Graves GR. The use of isotope tracers for identifying populations of migratory birds. Oecologia. 1997;109:132–41.
- 9. Hobson KA, Wassenaar LI. Tracking animal migration with stable isotopes. London: Academic; 2008.
- 10. Poage MA, Chamberlain CP. Empirical relationships between elevation and the stable isotope composition of precipitation and surface waters: considerations for studies of paleoelevational change. American Journal of Science. 2001;301:1–15.
- 11. Bowen GJ, Wassenaar LI, Hobson KA. Global application of stable hydrogen and oxygen isotopes to wildlife forensics. Oecologia. 2005;143:337–48. pmid:15726429
- 12. West JB, Bowen GJ, Dawson TE, Tu KP, editors. Isoscapes: understanding movement, pattern, and process on Earth through isotope mapping. 1st ed. New York: Springer; 2009.
- 13. Estep MF, Dabrowski H. Tracing food webs with stable hydrogen isotopes. Science. 1980;209:1537–8. pmid:17745967
- 14. Hobson KA, Atwell L, Wassenaar LI. Influence of drinking water and diet on the stable-hydrogen isotope ratios of animal tissues. Proc Natl Acad Sci USA 1999;96:8003–6. pmid:10393937
- 15. Wolf N, Newsome SD, Fogel ML, Martínez del Rio C. An experimental exploration of the incorporation of hydrogen isotopes from dietary sources into avian tissues. Journal of Experimental Biology. 2012;215:1915–22. pmid:22573770
- 16. Wolf N, Bowen GJ, Martínez del Rio C. The influence of drinking water on the δ2H and δ18O values of house sparrow plasma, blood and feathers. Journal of Experimental Biology. 2011;214:98–103. pmid:21147973
- 17. Wassenaar LI, Hobson KA. Comparative equilibration and online technique for determination of non-exchangeable hydrogen of keratins for use in animal migration studies. Isotopes in Environmental and Health Studies. 2003;39:211–7. pmid:14521282
- 18. Wassenaar LI, Hobson KA. Improved method for determining the stable-hydrogen isotopic composition (delta D) of complex organic materials of environmental interest. Environmental Science and Technology. 2000;34:2354–60.
- 19. Sauer PE, Schimmelmann A, Sessions AL, Topalov K. Simplified batch equilibrium for D/H determination of non-exchangeable hydrogen in solid organic material. Rapid Communication in Mass Spectrometry 2009;23:949–56.
- 20. Hobson KA, Van Wilgenburg SL, Wassenaar LI, Larson K. Linking hydrogen (δ2H) isotopes in feathers and precipitation: sources of variance and consequences for assignment to isoscapes. PLoS One. 2012;7:e35137. pmid:22509393
- 21. Langin KM, Reudink MW, Marra PP, Norris DR, Kyser TK, Ratcliffe LM. Hydrogen isotopic variation in migratory bird tissues of known origin: implications for geographic assignment. Oecologia. 2007;152:449–57. pmid:17370093
- 22. Studds CE, Kyser TK, Marra PP. Natal dispersal driven by environmental conditions interacting across the annual cycle of a migratory bird. Proceedings of the National Academy of Sciences USA. 2008;105:2929–33.
- 23. Wassenaar LI, Hobson KA. Stable-hydrogen isotope heterogeneity in keratinous materials: mass spectrometry and migratory wildlife tissue subsampling strategies. Rapid Communications in Mass Spectrometry. 2006;20:2505–10. pmid:16862621
- 24. Farmer A, Cade BS, Torres-Dowdall J. Fundamental limits to the accuracy of deuterium isotopes for identifying the spatial origin of migratory animals. Oecologia. 2008;158:183–92. pmid:18810500
- 25. Wetmore A. The number of contour feathers in passeriform and related birds. Auk. 1936;53:159–69.
- 26. Norris DR, Marra PP, Montgomerie R, Kyser TK, Ratcliffe LM. Reproductive effort, molting latitude, and feather color in a migratory songbird. Science. 2004;306:2249–50. pmid:15618516
- 27. Clegg SM, Kelly JF, Kimura M, Smith TB. Combining genetic markers and stable isotopes to reveal population connectivity and migration patterns in a Neotropical migrant, Wilson's warbler (Wilsonia pusilla). Molecular Ecology. 2003;12:819–30. pmid:12753204
- 28. Kelly JF. Stable isotope evidence links breeding geography and migration timing in wood warblers (Parulidae). Auk. 2006;123:431–7.
- 29. Paxton KL, Van Riper C III, Theimer TC, Paxton EH. Spatial and temporal migration patterns of Wilson's Warbler (Wilsonia pusilla) in the southwest as revealed by stable isotopes. Auk. 2007;124:162–75.
- 30. Hobson KA, Van Wilgenburg SL, Faaborg J, Toms JD, Rengifo C, Sosa AL, et al. Connecting breeding and wintering grounds of Neotropical migrant songbirds using stable hydrogen isotopes: a call for an isotopic atlas of migratory connectivity. Journal of Field Ornithology. 2014;85:237–57.
- 31. Wiley AE, Ostrom PH, Stricker CA, James HF, Gandhi H. Isotopic characterization of flight feathers in two pelagic seabirds: sampling strategies for ecological studies. Condor. 2010;112:337–46.
- 32. Perez GE, Hobson KA. Isotopic evaluation of interrupted molt in northern breeding populations of the Loggerhead Shrike. Condor. 2006;108:877–86.
- 33. Meehan TD, Rosenfield RN, Atudorei VN, Bielefeldt J, Rosenfield LJ, Stewart AC, et al. Variation in hydrogen stable-isotope ratios between adult and nestling Cooper's Hawks. Condor. 2003;105:567–72.
- 34. Hobson KA, Van Wilgenburg SL, Dunn EH, Hussell DJT, Taylor PD, Collister DM. Predicting the origins of passerines migrating through Canadian migration monitoring stations using stable-hydrogen isotope analyses of feathers: a new tool for bird conservation. Avian Conservation and Ecology. 2015;10:3. http://dx.doi.org/10.5751/ACE-00719-100103.
- 35. Nordell CJ, Haché S, Bayne EM, Sólymos P, Foster KR, Godwin CM, et al. Within-site variation in feather stable hydrogen isotope (δ2Hf) values of boreal songbirds: implications for assignment to molt origin. PLoS ONE. 2016;11(11):e0163957. pmid:27806037
- 36. Boulet MB, Gibbs HL, Hobson KA. Integrated analysis of genetic, stable isotope and banding data reveal migratory connectivity and flyways in the northern Yellow Warbler (Dendroica petechia; Aestiva group). Ornithological Monographs. 2006;61:29–78.
- 37. Holberton RL, Van Wilgenburg SL, Leppold AJ, Hobson KA. Isotopic (δ2Hf) evidence of "loop migration" and use of the Gulf of Maine flyway by both western and eastern breeding populations of Blackpoll Warblers. Journal of Field Ornithology. 2015;86:213–28.
- 38. Hobson KA, Wassenaar LI, Bayne E. Using isotopic variance to detect long-distance dispersal and philopatry in birds: an example with ovenbirds and American redstarts. Condor. 2004;106:732–43.
- 39. Dugger KM, Faaborg J, Arendt WJ, Hobson KA. Understanding survival and abundance of overwintering warblers: does rainfall matter? Condor. 2004;106:744–60.
- 40. Podlesak DW, McWilliams SR, Hatch KA. Stable isotopes in breadth, blood, feces and feathers can indicate intra-individual changes in the diet of migratory songbirds. Oecologia. 2005;142:501–10. pmid:15586297
- 41. Biasiolli TG. Using stable isotopes to understand seasonal interactions in a long-distance migratory songbird: M.S. thesis, University of Vermont; 2008.
- 42. Graves GR, Romanek CS. Mesoscale patterns of altitudinal tenancy in migratory wood warblers inferred from stable carbon isotopes. Ecological Applications. 2009;19:1264–73. pmid:19688933
- 43. Graves GR, Romanek CS, Rodriguez Navarro A. Stable isotope signature of philopatry and dispersal in a migratory songbird. Proceedings of the National Academy of Sciences USA. 2002;99:8096–100.
- 44. Blum JD, Taliaferro EH, Holmes RT. Determining the sources of calcium for migratory songbirds using stable strontium isotopes. Oecologia. 2001;126:569–74. pmid:28547242
- 45. Holmes RT, Rodenhouse NL, Sillett TS. Black-throated Blue Warbler (Setophaga caerulescens). The birds of North America Ithaca, New York, Cornell Laboratory of Ornithology, http://bnabirdscornelledu/bna/species/087. 2005. https://doi.org/10.2173/bna.87
- 46. Graves GR. Geographic clines of age ratios of black-throated blue warblers (Dendroica caerulescens). Ecology. 1997;78:2524–31.
- 47. Bent AC. Life histories of North American wood warblers. Bulletin of United States National Museum. 1953;203:1–734.
- 48. Webster MS, Marra PP, Haig SM, Bensch S, Holmes RT. Links between worlds: unraveling migratory connectivity. Trends in Ecology and Evolution. 2002;17:76–83.
- 49. Graves GR. Age determination of free-living male Black-throated Blue Warblers during the breeding season. Journal of Field Ornithology. 1997;68:443–9.
- 50. Fallon SM, Fleischer RC, Graves GR. Malarial parasites as geographic markers in migratory birds? Biology Letters. 2006;2:213–6. pmid:17148365
- 51. Graves GR. Testicular volume and asymmetry are age-dependent in Black-throated Blue Warblers (Dendroica caerulescens). Auk. 2004;121:473–85.
- 52. Grus WE, Graves GR, Glenn TC. Geographic variation in the mitochondrial control region of Black-throated Blue Warblers Auk. 2009;126:198–210.
- 53. Lorimer CG. Age structure and disturbance history of a southern Appalachian virgin forest. Ecology. 1980;61:1169–84.
- 54. Braun EL. Deciduous forests of eastern North America. New York: Hafner; 1950.
- 55. Haney JC, Lee DS, Wilbert M. A half-century comparison of breeding birds in the southern Appalachians. Condor. 2001;103:268–77.
- 56. Wilcove DS. Changes in the avifauna of the Great Smoky Mountains: 1947–1983. Wilson Bulletin. 1988;100:256–71.
- 57. Dwight JJ. The sequences of plumages and moults of the passerine birds of New York. Annals of the New York Academy of Science. 1900;13:73–360.
- 58. Pyle P. Identification guide to North American birds. Part 1. Columbidae to Ploceidae. Bolinas, California: Slate Creek Press; 1997.
- 59. Nolan V Jr. The ecology and behavior of the Prairie Warbler Dendroica discolor. Ornithological Monographs. 1978;26:1–595.
- 60. Sillett TS, Holmes RT. Variation in survivorship of a migratory songbird throughout its annual cycle. Journal of Animal Ecology. 2002;71:296–308.
- 61. Stresemann E, Stresemann V. Die Mauser der Vogel. Journal für Ornithologie. 1966;107:1–53.
- 62. Bowen GJ, Chesson L, Nielson K, Cerling TE, Ehleringer JR. Treatment methods for the determination of d2H and d18O of hair keratin by continuous-flow isotope-ratio mass spectrometry. Rapid Communication in Mass Spectrometry. 2005;19:2371–8.
- 63. Wolf BO, Martinez del Rio C. Use of saguaro fruit by white-winged doves: isotopic evidence of tight ecological association. Oecologia. 2000;124:536–43. pmid:28308392
- 64. Betts MG, Hadley AS, Rodenhouse N, Nocera JJ. Social information trumps vegetation structure in breeding-site selection by a migrant songbird. Proceedings of the Royal Society B. 2008;275:2257–63. pmid:18559326
- 65. Reudink MW, Marra PP, Langin KM, Studds CE, Kyser TK, Ratcliffe LM. Molt-migration in the American Redstart (Setophaga ruticilla) revisited: Explaining variation in feather δD signatures. Auk. 2008;125:744–8.
- 66. Hache S, Hobson KA, Villard M-A, Bayne EM. Assigning birds to geographic origin using feather hydrogen isotope ratios (δ2H): importance of year, age, and habitat. Canadian Journal of Zoology. 2012;90:722–8.
- 67. van Dijk JGB, Meissner W, Klaassen M. Improving provenance studies in migratory birds when using feather hydrogen stable isotopes. Journal of Avian Biology. 2013;45:103–8.
- 68. Van Wilgenburg SL, Hobson KA, Brewster KR, Welker JM. Assessing dispersal in threatened migratory birds using stable hydrogen isotope (δD) analysis of feathers. Endangered Species Research. 2012;16(1):17–29.
- 69. Meehan TD, Lott CA, Sharp ZD, Smith RB, Rosenfield RN, Stewart AC, et al. Using hydrogen isotope geochemistry to estimate the natal latitudes of immature Cooper's hawks migrating through the Florida Keys. Condor. 2001;103:11–20.