Translator Disclaimer
8 August 2018 Where do winter crows go? Characterizing partial migration of American Crows with satellite telemetry, stable isotopes, and molecular markers
Author Affiliations +
Abstract

Partial migration—a strategy in which some individuals are resident and others are migratory within the same population—is widespread among avian species and could play an important transitional role in the evolution of migratory behavior. Nevertheless, detailed movement data are unavailable for most partial migrant species. We examined migration strategies of the American Crow (Corvus brachyrhynchos), a partially migratory species that overwinters in large communal roosts, from which some birds migrate north to breed. We used a combination of satellite telemetry, isotopic signatures (δ2H), and molecular markers (33 microsatellites) to describe and characterize the migratory movements of individuals from overwintering roosts on the east coast (Utica, New York) and west coast (Davis, California) of the United States. We collected 11,951 data points from 18 satellite-tagged individuals between 2014 and 2018, among which 14 (77.8%) were migratory (8 of 11 and 6 of 7 birds on the west and east coasts, respectively). Migration distances were 280–1,095 km and 177–793 km on the west and east coasts, respectively. Individual birds were consistent in their migratory behavior across years, and breeding-site fidelity was high: both migratory and resident birds returned to the same location in the breeding season of each year. Both isotopic signatures and molecular markers could generally differentiate residents from long-distance migrants (i.e. those breeding at latitudes >3.5°N of the resident populations), but they did not consistently differentiate residents from migrants with shorter migration distances. Overall, these data on the migratory movements of American Crows and the proportion of migrants in their roosts can serve as a baseline against which to test predictions about how partial migrants will respond to environmental alterations such as climate change and urbanization.

INTRODUCTION

Despite decades of research, the evolution of migration is still largely a mystery. Partial migration—a strategy in which some individuals are resident and others are migratory within the same population—is a central precursor in many of the hypotheses that have been set forth to explain its origin (Berthold 1999, Griswold et al. 2010, Zink 2011), potentially acting as a transitional stage in the evolution of complete migratoriness (Berthold 2001). Although partial migration is widespread across taxa, occurring in fish, mammals, insects, and birds (Shaw 2016), the phenomenon of partial migration is itself understudied and poorly understood (Chapman et al. 2011). Fundamental uncertainties remain, for example, about the extent to which it is a conditional strategy or a fixed trait with a genetic basis (Pulido et al. 1996, Gillis et al. 2008, Velez-Espino et al. 2013) and how environmental change might alter its expression (Nilsson et al. 2006, Chapman et al. 2011). Likewise, little is known about how partial migration might enable species to adapt to novel environments (Pulido 2011, Gilroy et al. 2016) or about the role of partial migration in disease transport (Daszak et al. 2001, Reed et al. 2003, Takekawa et al. 2010, Muzaffar et al. 2012, Bengtsson et al. 2016). Acquisition of movement data (e.g., routes, consistency, distance traveled) from partially migratory species, preferably from multiple populations (Chapman et al. 2011), is essential to understanding the evolution and implications of partial migration. We still lack detailed information about individual movements in many partial migrant species, however, despite rapid advances in tracking technology (Runge et al. 2014, Gilroy et al. 2016).

Here, we describe the migratory movements of the American Crow (Corvus brachyrhynchos; hereafter “crow”), a partial migrant that overwinters in populations comprising both year-round residents and migratory individuals that depart in spring to breed (Verbeek and Caffrey 2002, Ward and Raim 2011). In winter, crows aggregate in large nocturnal roosts (sometimes >10,000 birds; Hinton et al. 2015), often in areas of high human density and activity (Gorenzel and Salmon 1995, Hinton et al. 2015, Taff et al. 2016). Crows are of particular concern with respect to disease transmission because they are common urban and suburban birds in North America (McGowan 2001) that can harbor numerous pathogens important to humans, domestic animals, and wildlife, including West Nile virus (Wheeler et al. 2014, Hinton et al. 2015), Campylobacter jejuni (Weis et al. 2016, Taff and Townsend 2017), and other pathogens (Miller et al. 2010, Halová et al. 2014, Freund et al. 2016, Jamborova et al. 2017, Townsend et al. 2018).

Despite their conspicuous abundance in North America, their striking migration patterns and winter social behavior, and their potential relevance as disease vectors, published data on the migratory behavior of crows are limited. Spring and fall migratory movements have been reported in many parts of the range of both of the recognized northern subspecies (C. b. brachyrhynchos to the east and C. b. herperis to the west), and migration routes have been inferred in some locations through banding data (reviewed in Verbeek and Caffrey 2002). A single previous tracking study (using radio telemetry) described the spring migration of 6 migratory crows from Illinois (C. b. brachyrhynchos) to their breeding grounds in Michigan (Ward and Raim 2011). Nothing had been reported, however, about the extent to which migration is a fixed or flexible strategy or about the proportion of migratory birds within their overwinter roosts.

Our study had 2 objectives. First, we described the migratory movements of crows from 2 overwintering populations, one on the east coast (Utica, New York) and the other on the west coast (Davis, California) of the United States. We used satellite telemetry to track migratory routes of individuals, fidelity to breeding and wintering sites, and the extent to which individual birds were consistent in their migratory behavior across years. Second, we compared the efficacy of 2 intrinsic markers (molecular and stable isotope data) as tools to assess the breeding origin of overwintering crows in these large urban roosts. We then used these intrinsic and extrinsic markers to estimate the proportion of migrants at each roost. These data can serve as a baseline against which to test predictions about the responses of partial migrants to environmental change.

METHODS

Satellite Telemetry

We captured crows roosting in large (>7,500 birds) overwintering flocks on the west coast (Davis; 38.53°N, 121.75°W; C. b. herperis) and east coast (Utica; 43.09°N, 75.22°W; C. b. brachyrhynchos) of the United States. Crows were captured during the winter roosting period (January–March) using a net-launcher or baited drop-in trap. On the west coast, we captured 49 crows in 2014 and 2015; on the east coast, we captured 15 crows in 2016. Captured crows were marked with both a numbered USGS band and a unique color band. We determined age (i.e. second-year or after-second-year) by plumage characteristics (Pyle 1997). Blood samples (∼75 μL) were preserved in lysis buffer for genetic analysis, and 1 or 2 tail feathers were collected from each bird for isotopic analysis. We applied satellite tags to 11 of these birds in Davis (3 on March 8, 2014; 8 in January–February 2015) and to 7 of the birds in Utica (in February 2016). Satellite tags (12 g Solar PTT-100; Microwave Telemetry, Columbia, MD, USA) were secured by backpack harnesses made from 1.5 mm nylon cord following Taff et al. (2016). The weight of the tags was <3.5% of each bird's body weight (range: 353–470 g). After harnesses were attached, crows were released into an aviary for ≥1 hr to ensure that the harnesses did not impede movement. Crows were released after this holding period. Movements were monitored using the Argos satellite tracking system from time of release until April 23, 2018. Movement data were stored in the online database Movebank (Wikelski and Kays 2017) and analyzed in R 3.3.1 (R Core Team 2016).

Isotopic Analysis

In principle, stable hydrogen isotopes can be used to infer migratory origin because δ2H in feathers is derived from δ2H in food and water in the location where the feather was grown (Hobson 1999). In North America, δ2H in precipitation (δ2Hp) is generally lower at higher latitudes and elevations (Bowen et al. 2005). Crows undergo their annual postbreeding molt prior to autumn migration (Verbeek and Caffrey 2002, Townsend et al. 2009); therefore, deuterium in feathers (δ2Hf) collected on the wintering ground reflects δ2Hp at the breeding location. To assess the proportion of migratory birds in winter roosts, we collected and analyzed δ2Hf from feathers that we found beneath overwinter roosts from November through February (n = 93 and 45 individuals in California and New York, respectively). Feathers were cleaned twice (first with detergent and then with 2:1 chloroform:methanol; Paritte and Kelly 2009), weighed (California: 1.2 ± 0.01 mg; New York: 0.4 ± 0.01 mg), and submitted to the stable isotope facilities of the University of California at Davis (for the west coast feathers) or Cornell University (for the east coast feathers) for isotopic analyses using established keratin standards. Hydrogen isotope values (δ2H) were reported as the corrected delta value measured against a reference scale (Vienna Standard Mean Ocean Water).

We used δ2Hf in feathers collected from known-origin birds to validate our origin assignments. Known-origin samples included feathers from satellite-tagged birds with known breeding grounds (including both “known residents” and “known migrants”), passively molted feathers collected in the summer months (June–August) from adult resident crows on their breeding territories (“known residents”), as well as passively molted adult crow feathers collected in summer months at higher latitudes, presumably originating from local breeders in those locations. On the west coast, known-origin samples included 79 samples collected from “known resident” crows in Davis (38.53°N) in the summers of 2012–2014, and 20 samples from latitudes north of Davis, spanning a latitudinal range of 38.53–48.25°N. These northern samples included 8 satellite-tagged migratory birds with known breeding grounds and 12 passively molted feathers collected in the summer of 2014 along this latitudinal band. On the east coast, known-origin samples included 4 feathers collected from resident crows in Utica (43.09°N) and 5 feathers from our satellite-tagged migratory birds with breeding grounds ranging from 45.8°N to 48.3°N.

We developed a model to predict and analyze δ2Hp across the range of the crow on the east and west coasts in IsoMAP ( http://isomap.org). We selected precipitation isotope ratios from April, May, June, July, and August (the months encompassing the annual crow molt) from 1960 to 2010. We predicted δ2Hp as a function of elevation, latitude squared, and longitude squared (measured at each isotope monitoring station) with maps spanning known crow migration routes, based on satellite telemetry data. The δ2Hp isoscapes created for this study are available in  Supplemental Material Figure S1 and at  http://isomap.org (west coast: longitudinal range −123.3°W to −117.5°W, latitudinal range 38.5°N to 49.5°N, IsoMAP jobkey 66075; east coast: longitudinal range −82.3°W to −66.5°W, latitudinal range 41.4°N to 49.5°N, IsoMAP jobkey 66078).

Isoscapes generated by IsoMAP are based on precipitation isotope ratios, which can scale differently from δ2Hf. Therefore, we developed sample transfer functions on each coast to rescale δ2Hf to δ2Hp (Bowen et al. 2014). Transfer functions were made from a regression of known-origin δ2Hf values against δ2Hp values at those origins (estimated in IsoMAP), which we used to rescale the δ2Hf values for IsoMAP calculations. The west and east coast transfer functions were δ2Hp(west) = 31.95 + 0.73*δ2Hf(west) − 0.03*(δ2Hf(west) + 83.19)2 and δ2Hp(east) = −13.55 + 0.39*δ2Hf(east), respectively. We then generated a geographic likelihood assignment surface for each feather in IsoMAP using the “individual assignment” function. Standard deviations for individual assignments were derived from the residuals from the water/feather transfer functions (west coast: 6.3‰; east coast: 3.3‰).

Genetic Analysis

We genotyped birds at a panel of 33 microsatellite loci developed for the American Crow (Tarr and Fleischer 1998, Stenzler and Fitzpatrick 2002, Schoenle et al. 2007, Ernest et al. 2008, Verdugo et al. 2012). Locus characteristics (e.g., alleles per locus, null allele frequencies, and tests of Hardy-Weinberg equilibrium) and polymerase chain reaction conditions are given in  Supplemental Material Tables S1 and  S2. Mean allelic diversity was 9.8 ± 1.2 alleles locus−1 (range: 2–39 alleles locus−1), and frequencies of inferred null alleles were <0.1 at all loci (mean null allele frequency: 0.03 ± 0.01). Six of 33 loci deviated significantly from Hardy-Weinberg equilibrium (i.e. homozygosity excess); such departures were consistent with the regular occurrence of inbreeding in these crows (Townsend et al. 2010, A. K. Townsend personal observation). We scored all birds at a minimum of 30 loci; most were scored at the complete panel of 33 loci (mean proportion of loci scored >0.99). We confirmed that all feathers were derived from unique individuals by comparing genotypes and removing duplicates.

We visualized genetic structuring between the resident and migratory populations using STRUCTURE 2.3 (Pritchard et al. 2000). We limited our sample to birds for which we had other lines of evidence indicating resident or migratory status. We specified individuals as probable residents (west coast: n = 496 residents; east coast: n = 40 residents) when they were sampled during the breeding season or from late-summer roosts (May–August) or when satellite telemetry or resight data indicated year-round residency. We specified individuals as probable migrants (west coast: n = 38 migrants; east coast: n = 33 migrants) when satellite telemetry data indicated migratory behavior or if their isotopic signatures were ≥10‰ smaller than values observed among probable residents. On the west coast, δ2H of probable residents ranged from −68.9‰ to −95.5‰ (mean ± SE = −81.5 ± 0.9‰; n = 65 residents with corresponding isotopic and genetic data); on the east coast, it ranged from −84.5‰ to −95.2‰ (−89.8 ± 2.6 ‰; n = 4 residents with corresponding isotopic and genetic data). We therefore specified individuals as probable migrants when δ2H values were less than −105‰ (range of values among probable migrants: −107.4‰ to −147.2‰ on the west coast, −105.4‰ to −122.0‰ on the east coast). Additionally, the east coast “probable migrant” sample included 20 passively molted feathers collected from crows in summer months at higher latitudes along their migration route in 2017.

On each coast, we specified probable residents as belonging to “sampling population 1” and probable migrants as belonging to “sampling population 2.” We specified a LOCPRIOR model (burn-in = 10,000; MCMC repetitions =100,000; correlated allele frequencies; K = 2) using “sampling population” as an informative prior. This modeling framework assists in clustering when genetic structuring is weak: LOCPRIOR models place more prior weight on clustering outcomes that are correlated with these metadata yet do not tend to find structure when it is absent (Hubisz et al. 2009). Data were plotted in the program DISTRUCT (Rosenberg 2004). To facilitate comparisons with other studies, we estimated genetic differentiation (FST) between residents and probable migrants on each coast using GENEPOP 4.2. We did not attempt to determine the number of source populations using STRUCTURE because it estimates population number based on deviations from Hardy-Weinberg equilibrium (HWE) among populations. In this system, deviations from HWE occur within populations ( Supplemental Material Table S1), violating the assumptions of that analysis.

We characterized winter crows as migrants or residents using the program ONCOR (Anderson et al. 2008), which performs assignment tests and leave-one-out cross-validation analysis. To evaluate the performance of assignment tests, we first limited our baseline sample to the same putative migrants and residents from STRUCTURE analyses, specifying probable population of origin (resident or migratory) for each individual. We then ran the leave-one-out test, which removes each individual sequentially from the baseline population, estimates its origin, and records the fraction of correct classifications. To estimate the proportion of migratory birds at the winter roosts, we then performed assignment tests on all individuals sampled in winter. Assignment tests classify individuals to the baseline population (resident or migratory) that would have the highest probability of producing that genotype. The baseline file, again, was limited to birds of known origin.

RESULTS

Satellite Telemetry

We collected a total of 11,951 data points from 18 tagged crows (6,968 and 4,983 points from the west and east coasts, respectively). Tag transmission duration ranged from 54 to 1,305 days (mean = 388.8 ± 78.7 days [SE]). In total, 14 of the 18 satellite-tagged birds (77.8%) were migratory (8 of 11 and 6 of 7 birds on the west and east coasts, respectively). All resident birds remained within 25 km of the center of their breeding-season territory in all months of the year (Figure 1). Among the migratory birds, migration distance ranged from 280 to 1,095 km (mean = 589.3 ± 101.8 km) on the west coast and from 177 to 793 km (mean = 461.7 ± 81.3 km) on the east coast, corresponding to breeding latitudes ranging from 40.4°N to 48.3°N (mean = 43.6 ± 0.9°N) on the west coast and from 44.4°N to 48.4°N (mean = 46.3 ± 0.5°N) on the east coast. First recorded fall migratory movements of 13 of the 14 migrants are shown in Figure 2. After arrival at their breeding grounds (during March 26–April 8), all migrants remained within 25 km of the center of their territory for the entire breeding season (i.e. until at least September of that year).

FIGURE 1.

All recorded movements of (A) 3 satellite-tagged resident American Crows in Davis, California, and (B) 1 resident in Utica, New York, USA, over the duration of the study. Colors indicate different individuals. Data were recorded for each bird over a 12–42 mo period.

i0004-8038-135-4-964-f01.tif

FIGURE 2.

First recorded fall migratory movements of 8 satellite-tagged American Crows on the west coast and 5 on the east coast of North America. Colors indicate different individuals.

i0004-8038-135-4-964-f02.tif

Multiple breeding seasons (April–August) were tracked for 7 birds (2 breeding seasons for 4 birds; 3 breeding seasons for 2 birds; and 4 breeding seasons for a single bird). Two of these birds were residents and 5 were migratory. Breeding-site fidelity was high: migrants and residents returned to the same locations in the breeding season across years. In 2016, however, one migrant, a second-year pre-breeder, wandered for 78 days during its northward migration before settling down at one site for the remainder of the breeding season. In 2017, it returned to the same breeding site via a much more direct route ( Supplemental Material Figure S2).

Data from multiple winter seasons (November–February) were collected from 5 birds (2 winters from 4 migrants; 4 winters from 1 resident). The resident bird remained in the same geographic area (within 25 km of the center of its breeding-season territory) in all months of the year for 4 breeding and overwintering seasons. Data transmission for migratory birds was very low during the second winter (n = 3, 6, 21, and 36 transmissions). These transmissions showed that 2 of the 4 birds moved from site to site (some sites >300 km apart) during winter, and that 3 of the 4 birds spent at least part of the winter within 0–60 km and at the same latitudinal band (43.03–43.09°N) as their initial trapping site. For one bird, all winter transmissions (n = 36 transmissions) indicated that it did not return to its winter capture site in Utica but overwintered 260 km north in an entirely different location in the second winter (Ottawa, Ontario, Canada; 45.42°N, 75.69°W).

Isotopic Analysis

Feather isotope ratios (δ2Hf) were negatively correlated with latitude on both coasts (Figure 3). On the west coast, IsoMAP indicated a strong isotopic difference in δ2Hp between the Sacramento Valley (which encompassed the resident population of Davis) and areas north of the valley; however, no clear gradient in δ2Hp with latitude was otherwise apparent (Figure S1A). We therefore simply classified birds as “possible residents” when their likely origin included the Sacramento Valley (including Davis), or as “probable migrants” when their likely origin did not include Sacramento Valley. We quantified assignments for each individual by comparing their average likelihood surface within the Sacramento Valley at the latitude of Davis (38.53°N; longitude −121°W to −122°W) to the distribution of all likelihood values on the range map. If the assignment likelihood of an individual to Davis was greater than the mean likelihood value for all regions on the range map, that individual was considered a “possible resident.”

FIGURE 3.

Feather deuterium levels were negatively correlated with breeding latitude among known-origin, satellite-tagged American Crows on the (A) west coast (δ2Hf = −4.47 * latitude + 91.1; t10 = −5.4; P = 0.0003; r2 = 0.47) and (B) east coast (δ2Hf = −4.69 * latitude + 110.5; t7 = −3.8; P = 0.007; r2 = 0.67) of North America.

i0004-8038-135-4-964-f03.tif

This classification method discriminated residents from most migrants with relatively long migration distances. When we limited the sample to birds of known origin, all resident birds (n = 79; 38.53°N latitude) and all birds with breeding origins <2.5°N of the resident population (n = 13; 39°N to 40.94°N) were classified as “possible residents.” Among known migrants breeding at higher latitudes (42.39°N to 48.25°N), 5 of 6 (83.3%) were classified as “probable migrants.”

On the east coast, IsoMAP indicated that δ2Hp levels generally decreased with increasing latitude within the range of satellite-tagged crows; moreover, δ2Hp values in the Mohawk Valley (including the resident Utica population) were distinctly elevated in comparison with most northern latitudes (Figure S1B). We therefore classified birds as “possible residents” when their likely origin included Utica and as “probable migrants” when their likely origin did not include Utica. We quantified assignments for each individual by comparing their average likelihood surface within the Mohawk Valley at the latitude of Utica (latitude 43.09°N, longitude −75.1°W to −76.1°W) to the distribution of all likelihood values on the range map. As before, an individual was considered a “possible resident” if the assignment likelihood of an individual to Utica was greater than its mean likelihood value for all points on the range map. When we limited the sample to birds of known origin, all resident birds (n = 4; 43.09°N latitude) were classified correctly. Four of 5 known migrants originating from higher latitudes (45.80°N to 48.38°N) were classified as a “probable migrant” and the other one (breeding latitude 46.13°N) was misclassified as a “possible resident.”

Winter birds for which δ2Hf values were available included 93 and 45 individuals sampled near the winter roosts on the west and east coast, respectively. On the west coast, likelihood surfaces within IsoMAP indicated that ≥26 of 93 winter birds (28.0%) were probable migrants; on the east coast, ≥12 of 45 winter birds (26.7%) were probable migrants.

Genetic Analysis

Overall genetic differentiation between residents and putative migrants was low. Pairwise FST values between putative residents and migrants were 0.026 and 0.012 on the west and east coasts, respectively. Despite low differentiation, clear genetic distinctions between migrants and residents (Figure 4) were apparent in the program STRUCTURE using the LOCPRIOR modeling framework. On the west coast, leave-one-out cross-validation analysis in ONCOR indicated that the likelihood of assignment of known-origin birds to the correct population (resident or migrant population), based solely on molecular markers, was very high: 99% of putative residents were correctly assigned to the “resident” cluster, and 95% of putative migrants were correctly assigned to the migrant cluster. On the east coast, likelihood of correct assignment was lower, but still fairly high: 84% of putative residents were correctly assigned to the “resident” cluster, and 76% of putative migrants were correctly assigned to the “migrant” cluster. East coast resolution improved when we restricted our sample to residents vs. relatively long-distance migrants. After removing 4 birds that bred within 2°N of the resident population, 92% of putative residents were correctly assigned to the “resident” cluster, and 81% of putative migrants were correctly assigned to the migrant cluster.

FIGURE 4.

Membership coefficients (Q̂) of each individual American Crow to a given cluster (left cluster: known residents; right cluster: probable migrants). (A) On the west coast, 484 of 496 (97.6%) putative residents had a high probability (Q̂ > 0.9) of membership in the “resident” cluster, and none of the 38 probable migrants were assigned to the resident cluster with probability >0.53 (mean Q̂ of putative migrants to the resident cluster = 0.25 ± 0.01). (B) On the east coast, all putative residents had a high probability of membership in the “resident” cluster (Q̂ ≥ 0.94), and only 1 of 34 putative migrants was assigned to the resident cluster with probability ≥0.9 (Q̂ ≥ 0.925).

i0004-8038-135-4-964-f04.tif

We then ran assignment tests in ONCOR to estimate the proportion of winter birds that were migratory. These samples included 89 and 53 individuals sampled near the winter roosts of the west and east coast, respectively. On the west coast, ONCOR assigned 51.7% (46 of 89) of birds to the “resident” category with probability >0.96 (mean probability of membership in resident group = 0.99 ± 0.00; n = 46 birds); the other 48.3% of birds were classified to the migrant group with probabilities ranging from 0.2 to 1.0 (mean probability of membership in migrant group = 0.91 ± 0.04; n = 43 birds). On the east coast, ONCOR assigned 33.9% (18 of 53) of winter birds to the “resident” category with probability >0.91 (mean probability of membership in resident group = 0.98 ± 0.01; n = 18 birds); the other 66.0% of birds were classified in the migrant group with probabilities ranging from 0.2 to 1.0 (mean probability of membership in migrant group = 0.87 ± 0.05; n = 35 birds).

Comparison of Methods

The percentages of birds classified as migratory by each method are summarized in Table 1. More than one type of data (telemetry, molecular, or isotopic) was available for different subsets of winter-caught birds. On the west coast, both molecular and isotopic data were available from 89 winter birds; 11 of these were satellite-tagged. Overall congruence between classifications by molecular and isotopic data was 80.1% (72 of 89 samples). Almost all birds classified as residents in ONCOR (46 of 47 birds) were also classified as residents in IsoMAP, a result consistent with available telemetry data (n = 3 confirmed residents). Twenty-five birds were identified as migrants by both ONCOR and IsoMAP, again consistent with telemetry data (n = 5 confirmed migrants; all breeding at latitudes 3.84–9.7°N of the resident population). Seventeen birds were classified as migratory in ONCOR but resident in IsoMAP. The telemetry data in this subset (n = 3 confirmed migrants) indicated that this group included migrants with short or intermediate migration distances (breeding 1.9–3.9°N of the resident population). On the east coast, both molecular and isotopic data were available from 42 winter birds; 6 of these were satellite-tagged. Here, congruence between classifications by molecular and isotopic data was 64.3% (27 of 42 samples). Again, almost all birds classified as residents in ONCOR (16 of 17 putative residents) were also classified as residents in IsoMAP, a result consistent with available telemetry data (n = 1 confirmed resident). Eleven birds were identified as migrant by both ONCOR and IsoMAP, again consistent with telemetry data (n = 4 confirmed migrants; all breeding at latitudes 2.8–5.3°N of the resident population). Fourteen birds were classified as migratory in ONCOR but resident in IsoMAP. Telemetry data in this subset (n = 1 bird) indicated that this group included a migrant with an intermediate migration distance (breeding 3.1°N of the resident population).

TABLE 1.

Percentages of American Crows in overwintering roosts classified as migratory by satellite telemetry, microsatellite markers, and stable isotopes (deuterium). Proportions are shown in parentheses. West coast = roost in Davis, California, USA; east coast = roost in Utica, New York, USA.

i0004-8038-135-4-964-t01.tif

Considered in concert, these data suggest that both intrinsic markers usually were successful at discerning long-distance migrants (i.e. those breeding at latitudes >3.5°N of the resident populations) from residents but did not always distinguish residents from relatively short-distance migrants. Isotopic signatures were more likely than molecular markers to misclassify short-distance migrants as residents on both coasts.

DISCUSSION

We used satellite telemetry to track the partial migratory behavior of birds from 2 overwinter roosts of the northern subspecies of the American Crow. We then compared tracking data with characterizations based on molecular markers and isotopic signatures and estimated the proportion of migratory birds in the overwinter roosts. Tracking data indicated that 73% of tagged crows (8 of 11) in our western population (C. b. herperis) and 86% of tagged crows (6 of 7) in our eastern population (C. b. brachyrhynchos) migrated north for the breeding season (Figures 1 and 2), with mean migration distances of 589 km (range: 280–1,095 km) and 462 km (range: 177–793 km) on the west and east coasts, respectively. These data were consistent with published band recovery data, showing that some crows in these subspecies move between the southern provinces of Canada and the contiguous United States (longest recorded migration distance >2,800 km; Brewer et al. 2000) or make shorter migratory movements within the contiguous United States (reviewed in Verbeek and Caffrey 2002). The short-distance migratory distances that we observed were similar to those reported in a single previous tracking study of crows (based on radio telemetry data), documenting spring migratory movements of 6 migratory crows (C. b. brachyrhynchos), during which birds traveled an average of 571 km (range: 382–725 km) from Illinois to Michigan (Ward and Raim 2011). Our study was the first, however, to estimate the proportion of migratory birds in overwintering roosts, to track breeding- and winter-site fidelity of migrants, and to examine consistency in migratory behavior of individuals across years.

Partial migration may play an important role in the evolution of complete migration (Berthold 1999, Griswold et al. 2010, Zink 2011), the local persistence and long-distance spread of disease (Reed et al. 2003, Takekawa et al. 2010, Muzaffar et al. 2012), and population responses to environmental change (Gilroy et al. 2016), yet fundamental questions remain about the extent to which it is a plastic or fixed trait (Pulido et al. 1996, Gillis et al. 2008, Velez-Espino et al. 2013). We found no evidence that migration in crows was a facultative strategy. All satellite-tagged birds for which we had multiple years of data were consistent in their migratory behavior (migratory or resident) from year to year, and all were faithful to their breeding territories across years. We note, however, that the sample size of birds for which we had multiple years of data was small (5 migrants, 3 residents), limiting the scope of the inferences that we can draw.

Better understanding the degree to which birds can conditionally adjust their migratory behavior is particularly urgent as we seek to predict the responses of partial migrants to climate change. Recent work suggests that partial migration might buffer some species against a warming climate, although taxon-specific data on the effects of climate change on partial migration are scarce and the mechanisms underlying this pattern are unclear (Gilroy et al. 2016). It is possible, for example, that the proportion of sedentary birds in some partially migratory populations will increase as milder winters permit them to remain on their breeding grounds year-round (Berthold 1999), although empirical data are scarce and do not provide strong support for this prediction (Nilsson et al. 2006, Meller et al. 2016). Another possibility is that they may have the ability to shorten their migration distances by overwintering at higher latitudes in warmer years. Our tracking data—although sparse across winters—indicated that migratory crows were flexible with respect to their overwintering sites, moving among sites across and within winter seasons. Therefore, even if individual crows do not completely switch between migratory and sedentary behavior, they might be able to conditionally adjust the distance that they travel to their overwinter roosts in response to temperature.

Urbanization is another environmental alteration that could influence the behavior of partial migrants. Crows are a particularly valuable system for urban ecology studies because they often exploit urban areas for their overwinter roosts (Gorenzel and Salmon 1995, Hinton et al. 2015, Taff et al. 2016) and breeding territories (McGowan 2001). Some studies have found that urbanization is associated with increased sedentariness among partially migratory species (Adriaensen and Dhondt 1990, Partecke and Gwinner 2007, Evans et al. 2012), an effect that may be due, in part, to the milder winters of the urban “heat island” (Partecke and Gwinner 2007, Evans et al. 2012). If these patterns hold true for crows, we might expect them to exhibit increasing sedentariness in urban areas, or to shorten their southward migration distances and exploit urban overwinter sites at higher latitudes, in tempo with increasing temperatures.

The second major aim of this study was to evaluate the efficacy of intrinsic markers (isotopic and molecular signatures) as tools to characterize migratory propensity. In contrast with satellite tags, which are costly and are too heavy for many passerine species (Wikelski et al. 2007), intrinsic markers are relatively cost-effective, noninvasive, and easy to collect but can have coarse spatial resolution (Bowen et al. 2005, Ruegg et al. 2017). Using subsets of these markers in combination can improve the efficacy and resolution of origin assignment (Van Wilgenburg and Hobson 2011, Vander Zanden et al. 2015, Ruegg et al. 2017). In the present study, molecular markers indicated that ≥48% of birds on the west coast and ≥66% of birds on the east coast were migratory, whereas isotopic signatures identified only 28% of western birds and 27% of eastern birds as migrants. Comparison of these characterizations with actual origins of known-origin birds indicated that both isotopic and molecular signatures were generally successful in distinguishing residents from migrants with relatively long migration distances (i.e. those breeding at latitudes >3.5°N of the resident populations) but sometimes failed to distinguish residents from migrants with shorter migration distances. This limitation was more pronounced for characterization based solely on isotopic signatures. Considered in concert, our results indicate that—when telemetry data are not feasible to collect—genetic tools are a better option than isotopic signatures to identify long-distance migrants in crows. However, extreme isotopic values (as observed among long-distance migrants) would be useful to strengthen the certainty of assignment and could be used as priors in analyses of population structure. Similar conclusions were drawn in a study using genetic and isotopic markers to characterize migration in Wilson's Warblers (Cardellina pusilla; Ruegg et al. 2017).

Here, we compared migratory characterization results based on separate analyses of each data type (tracking, molecular, and isotopic data). Collection and comparison of more than one data type is become increasing common in migration studies (Cunjak et al. 2005, Boulet et al. 2006, Procházka et al. 2017), and some of these studies have begun to integrate 2 or more data types into a single analysis (e.g., band recovery and isotopic data [Van Wilgenburg and Hobson 2011]; band recovery, geolocator data, and other covariates [Procházka et al. 2017]; mark–resight and tracking data [Korner-Nievergelt et al. 2017]; isotopic, environmental, and molecular data [Ruegg et al. 2017]). Although, in some systems, specific data types (e.g., molecular markers) can perform nearly as well as integrated data types in population assignment (Ruegg et al. 2017), an integrated approach could reduce potential bias inherent in the collection of the individual data types and add detail and precision to migrant–resident characterizations (Robinson et al. 2010, Van Wilgenburg and Hobson 2011, Korner-Nievergelt et al. 2017, Ruegg et al. 2017). Ultimately, the development of a modeling framework that integrates the data types used in our study (satellite telemetry, microsatellites, and isotopes) would be a valuable tool for the migratory characterization of crows and other species for which this combination of data is available.

In summary, the data presented here on the migratory movements of American Crows in 2 of their northern subspecies, and the characterization of the proportion of migrants in 2 overwinter roosts, can serve as a baseline against which to test predictions about changes in migratory behavior with climate and urbanization. Observing the extent to which the expression of partial migration in crows matches theoretical expectations will allow us to better predict how environmental changes might affect other species of partially migratory passerines, the majority of which are, at present, too small to permit satellite tracking (Bridge et al. 2011) and for which migratory movements have not yet been elucidated (Runge et al. 2014, Gilroy et al. 2016).

ACKNOWLEDGMENTS

Thanks to M. Hinton, R. Logsdon, S. Wheeler, Y. H. Suh, M. Saberi, M. Jones, J. Hurd, W. Boyce, E. Paul, and J. Tremblay for assistance and support.

Funding statement: This project was supported by ANR Competitive Grant no. 13-2735 from the University of California, Davis, and by Agriculture and Food Research Initiative Competitive Grant no. 2014-67012-21622 from the USDA National Institute of Food and Agriculture.

Ethics statement: All fieldwork was performed under protocols approved by the Institutional Animal Care and Use Committees of the University of California, Davis, and Hamilton College.

Author contributions: A.T. conceived of the study. A.T. and C.T. designed the methods. A.T., C.T., B.F., and A.McG. conducted the research and collected the data. A.T. and A.McG. analyzed the data. A.T., B.F., and A.McG. wrote the paper.

LITERATURE CITED

  1. Adriaensen, F., and A. A. Dhondt (1990). Population dynamics and partial migration of the European Robin (Erithacus rubecula) in different habitats. Journal of Animal Ecology 59:1077–1090. Google Scholar
  2. Anderson, E. C., R. S. Waples, and S. T. Kalinowski (2008). An improved method for predicting the accuracy of genetic stock identification. Canadian Journal of Fisheries and Aquatic Sciences 65:1475–1486. Google Scholar
  3. Bengtsson, D., K. Safi, A. Avril, W. Fiedler, M. Wikelski, G. Gunnarsson, J. Elmberg, C. Tolf, B. Olsen, and J. Waldenström (2016). Does influenza A virus infection affect movement behaviour during stopover in its wild reservoir host?Royal Society Open Science 3. Google Scholar
  4. Berthold, P. (1999). A comprehensive theory for the evolution, control and adaptability of avian migration. Ostrich 70:1–11. Google Scholar
  5. Berthold, P. (2001). Bird Migration: A General Survey, second edition. Oxford University Press, Oxford, UK. Google Scholar
  6. Boulet, M., H. L. Gibbs, and K. A. Hobson (2006). 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 61:29–78. Google Scholar
  7. Bowen, G. J., Z. Liu, H. B. Vander Zanden, L. Zhao, and G. Takahashi (2014). Geographic assignment with stable isotopes in IsoMAP. Methods in Ecology and Evolution 5:201–206. Google Scholar
  8. Bowen, G. J., L. I. Wassenaar, and K. A. Hobson (2005). Global application of stable hydrogen and oxygen isotopes to wildlife forensics. Oecologia 143:337–348. Google Scholar
  9. Brewer, D., A. W. Diamond, E. J. Woodsworth, B. T. Collins, and E. H. Dunn (2000). Canadian Atlas of Bird Banding, vol. 1: Doves, Cuckoos, and Hummingbirds through Passerines, 1921–1995. Canadian Wildlife Service, Ottawa, ON, Canada. Google Scholar
  10. Bridge, E. S., K. Thorup, M. S. Bowlin, P. B. Chilson, R. H. Diehl, R. W. Fléron, P. Hartl, R. Kays, J. F. Kelly, W. D. Robinson, and M. Wikelski (2011). Technology on the move: Recent and forthcoming innovations for tracking migratory birds. BioScience 61:689–698. Google Scholar
  11. Chapman, B. B., C. Brönmark, J.-Å. Nilsson, and L.-A. Hansson (2011). The ecology and evolution of partial migration. Oikos 120:1764–1775. Google Scholar
  12. Cunjak, R. A., J. M. Roussel, M. A. Gray, J. P. Dietrich, D. F. Cartwright, K. R. Munkittrick, and T. D. Jardine (2005). Using stable isotope analysis with telemetry or mark–recapture data to identify fish movement and foraging. Oecologia 144:636–646. Google Scholar
  13. Daszak, P., A. A. Cunningham, and A. D. Hyatt (2001). Anthropogenic environmental change and the emergence of infectious diseases in wildlife. Acta Tropica 78:103–116. Google Scholar
  14. Ernest, H. B., J. A. Well, and J. D. Kurushima (2008). Development of 10 microsatellite loci for Yellow-billed Magpies (Pica nuttalli) and corvid ecology and West Nile virus studies. Molecular Ecology Resources 8:196–198. Google Scholar
  15. Evans, K. L., J. Newton, K. J. Gaston, S. P. Sharp, A. McGowan, and B. J. Hatchwell (2012). Colonisation of urban environments is associated with reduced migratory behaviour, facilitating divergence from ancestral populations. Oikos 121:634–640. Google Scholar
  16. Freund, D., S. S. Wheeler, A. K. Townsend, W. M. Boyce, H. B. Ernest, C. Cicero, and R. N. M. Sehgal (2016). Genetic sequence data reveals widespread sharing of Leucocytozoon lineages in corvids. Parasitology Research 115:3557–3565. Google Scholar
  17. Gillis, E. A., D. J. Green, H. A. Middleton, and C. A. Morrissey (2008). Life history correlates of alternative migratory strategies in American Dippers. Ecology 89:1687–1695. Google Scholar
  18. Gilroy, J. J., J. A. Gill, S. H. M. Butchart, V. R. Jones, and A. M. A. Franco (2016). Migratory diversity predicts population declines in birds. Ecology Letters 19:308–317. Google Scholar
  19. Gorenzel, W. P., and T. P. Salmon (1995). Characteristics of American Crow urban roosts in California. The Journal of Wildlife Management 59:638–645. Google Scholar
  20. Griswold, C. K., C. M. Taylor, and D. R. Norris (2010). The evolution of migration in a seasonal environment. Proceedings of the Royal Society B 277:2711–2720. Google Scholar
  21. Halová, D., I. Papoušek, I. Jamborova, M. Masarikova, A. Cizek, N. Janecko, V. Oravcova, L. Zurek, A. B. Clark, A. Townsend, J. C. Ellis, and I. Literak (2014). Plasmid-mediated quinolone resistance genes in Enterobacteriaceae from American Crows: High prevalence of bacteria with variable qnrB genes. Antimicrobial Agents and Chemotherapy 58:1257–1258. Google Scholar
  22. Hinton, M. G., W. K. Reisen, S. S. Wheeler, and A. K. Townsend (2015). West Nile Virus activity in a winter roost of American Crows (Corvus brachyrhynchos): Is bird-to-bird transmission important in persistence and amplification?Journal of Medical Entomology 52:683–692. Google Scholar
  23. Hobson, K. A. (1999). Tracing origins and migration of wildlife using stable isotopes: A review. Oecologia 120:314–326. Google Scholar
  24. Hubisz, M. J., D. Falush, M. Stephens, and J. K. Pritchard (2009). Inferring weak population structure with the assistance of sample group information. Molecular Ecology Resources 9:1322–1332. Google Scholar
  25. Jamborova, I., M. Dolejska, L. Zurek, A. K. Townsend, A. B. Clark, J. C. Ellis, I. Papousek, A. Cizek, and I. Literak (2017). Plasmid-mediated resistance to cephalosporins and quinolones in Escherichia coli from American Crows in the USA. Environmental Microbiology 19:2025–2036. Google Scholar
  26. Korner-Nievergelt, F., C. Prévot, S. Hahn, L. Jenni, and F. Liechti (2017). The integration of mark re-encounter and tracking data to quantify migratory connectivity. Ecological Modelling 344:87–94. Google Scholar
  27. McGowan, K. J. (2001). Demographic and behavioral comparisons of suburban and rural American Crows. InAvian Ecology and Conservation in an Urbanizing World ( J. M. Marzluff, R. Bowman, and R. Donelly, Editors). Kluwer Academic Press, Norwell, MA, USA. pp. 365–381. Google Scholar
  28. Meller, K., A. V. Vähätalo, T. Hokkanen, J. Rintala, M. Piha, and A. Lehikoinen (2016). Interannual variation and long-term trends in proportions of resident individuals in partially migratory birds. Journal of Animal Ecology 85:570–580. Google Scholar
  29. Miller, A. D., A. K. Townsend, K. J. McGowan, A. B. Clark, A. L. Glaser, L. A. Patrican, E. Dobson, and E. L. Buckles (2010). Non-West Nile virus-associated mortality in a population of American Crows (Corvus brachyrhynchos): A gross and histopathologic study. Journal of Veterinary Diagnostic Investigation 22:289–295. Google Scholar
  30. Muzaffar, S. B., N. J. Hill, J. Y. Takekawa, W. M. Perry, L. M. Smith, and W. M. Boyce (2012). Role of bird movements in the epidemiology of West Nile and avian influenza virus. Human–Wildlife Interactions 6:72–88. Google Scholar
  31. Nilsson, A. L. K., Å. Lindström, N. Jonzén, S. G. Nilsson, and L. Karlsson (2006). The effect of climate change on partial migration—The Blue Tit paradox. Global Change Biology 12:2014–2022. Google Scholar
  32. Paritte, J. M., and J. F. Kelly (2009). Effect of cleaning regime on stable-isotope ratios of feathers in Japanese Quail (Coturnix japonica). The Auk 126:165–174. Google Scholar
  33. Partecke, J., and E. Gwinner (2007). Increased sedentariness in European Blackbirds following urbanization: A consequence of local adaptation?Ecology 88:882–890. Google Scholar
  34. Pritchard, J. K., M. Stephens, N. A. Rosenberg, and P. Donnelly (2000). Association mapping in structured populations. American Journal of Human Genetics 67:170–181. Google Scholar
  35. Procházka, P., S. Hahn, S. Rolland, H. van der Jeugd, T. Csörgö, F. Jiguet, T. Mokwa, F. Liechti, D. Vangeluwe, and F. Korner-Nievergelt (2017). Delineating large-scale migratory connectivity of Reed Warblers using integrated multistate models. Diversity and Distributions 23:27–40. Google Scholar
  36. Pulido, F. (2011). Evolutionary genetics of partial migration—the threshold model of migration revis(it)ed. Oikos 120:1776–1783. Google Scholar
  37. Pulido, F., P. Berthold, and A. J. van Noordwijk (1996). Frequency of migrants and migratory activity are genetically correlated in a bird population: Evolutionary implications. Proceedings of the National Academy of Sciences USA 93:14642–14647. Google Scholar
  38. Pyle, P. (1997). Identification Guide to North American Birds, part 1. State Creek Press, Bolinas, CA, USA. Google Scholar
  39. R Core Team(2016). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Google Scholar
  40. Reed, K. D., J. K. Meece, J. S. Henkel, and S. K. Shukla (2003). Birds, migration and emerging zoonoses: West Nile virus, lyme disease, influenza A and enteropathogens. Clinical Medicine & Research 1:5–12. Google Scholar
  41. Robinson, W. D., M. S. Bowlin, I. Bisson, J. Shamoun-Baranes, K. Thorup, R. H. Diehl, T. H. Kunz, S. Mabey, and D. W. Winkler (2010). Integrating concepts and technologies to advance the study of bird migration. Frontiers in Ecology and the Environment 8:354–361. Google Scholar
  42. Rosenberg, N. A. (2004). DISTRUCT: A program for the graphical display of population structure. Molecular Ecology Notes 4:137–138. Google Scholar
  43. Ruegg, K. C., E. C. Anderson, R. J. Harrigan, K. L. Paxton, J. F. Kelly, F. Moore, and T. B. Smith (2017). Genetic assignment with isotopes and habitat suitability (GAIAH), a migratory bird case study. Methods in Ecology and Evolution 8:1241–1252. Google Scholar
  44. Runge, C. A., T. G. Martini, H. P. Possingham, S. G. Willis, and R. A. Fuller (2014). Conserving mobile species. Frontiers in Ecology and the Environment 12:395–402. Google Scholar
  45. Schoenle, L. A., A. K. Townsend, and I. J. Lovette (2007). Isolation and characterization of microsatellite loci in a cooperatively breeding corvid, the American Crow (Corvus brachyrhynchos). Molecular Ecology Notes 7:46–48. Google Scholar
  46. Shaw, A. K. (2016). Drivers of animal migration and implications in changing environments. Evolutionary Ecology 30:991–1007. Google Scholar
  47. Stenzler, L. M., and J. W. Fitzpatrick (2002). Isolation of microsatellite loci in the Florida Scrub-Jay Aphelocoma coerulescens. Molecular Ecology Notes 2:547–550. Google Scholar
  48. Taff, C. C., and A. K. Townsend (2017). Campylobacter jejuni infection associated with relatively poor condition and low survival in a wild bird. Journal of Avian Biology 48:1071–1076. Google Scholar
  49. Taff, C. C., A. M. Weis, S. Wheeler, M. G. Hinton, B. C. Weimer, C. M. Barker, M. Jones, R. Logsdon, W. A. Smith, W. M. Boyce, and A. K. Townsend (2016). Influence of host ecology and behavior on Campylobacter jejuni prevalence and environmental contamination risk in a synanthropic wild bird species. Applied and Environmental Microbiology 82:4811–4820. Google Scholar
  50. Takekawa, J. Y., D. J. Prosser, S. H. Newman, S. Bin Muzaffar, N. J. Hill, B. Yan, X. Xiao, F. Lei, T. Li, S. E. Schwarzbach, and J. A. Howell (2010). Victims and vectors: Highly pathogenic avian influenza H5N1 and the ecology of wild birds. Avian Biology Research 3:51–73. Google Scholar
  51. Tarr, C. L., and R. C. Fleischer (1998). Primers for polymorphic GT microsatellites isolated from the Mariana Crow, Corvus kubaryi. Molecular Ecology 7:253–255. Google Scholar
  52. Townsend, A. K., A. B. Clark, and K. J. McGowan (2010). Direct benefits and genetic costs of extrapair paternity for female American Crows (Corvus brachyrhynchos). The American Naturalist 175:E1–E9. Google Scholar
  53. Townsend, A. K., A. B. Clark, K. J. McGowan, and I. J. Lovette (2009). Reproductive partitioning and the assumptions of reproductive skew models in the cooperatively breeding American Crow. Animal Behaviour 77:503–512. Google Scholar
  54. Townsend, A. K., S. S. Wheeler, D. Freund, R. N. M. Sehgal, and W. M. Boyce (2018). Links between blood parasites, blood chemistry, and the survival of nestling American Crows. Ecology and Evolution 8. In press. Google Scholar
  55. Vander Zanden, H. B., A. D. Tucker, K. M. Hart, M. M. Lamont, I. Fujisaki, D. S. Addison, K. L. Mansfield, K. F. Phillips, M. B. Wunder, G. J. Bowen, M. Pajuelo, et al. (2015). Determining origin in a migratory marine vertebrate: A novel method to integrate stable isotopes and satellite tracking. Ecological Applications 25:320–335. Google Scholar
  56. Van Wilgenburg, S. L., and K. A. Hobson (2011). Combining stable-isotope (deltaD) and band recovery data to improve probabilistic assignment of migratory birds to origin. Ecological Applications 21:1340–1351. Google Scholar
  57. Velez-Espino, L. A., R. L. McLaughlin, and M. Robillard (2013). Ecological advantages of partial migration as a conditional strategy. Theoretical Population Biology 85:1–11. Google Scholar
  58. Verbeek, N. A., and C. Caffrey (2002). American Crow (Corvus brachyrhynchos), version 2.0. InBirds of North America Online ( A. F. Pooleand F. B. Gill, Editors). Cornell Lab of Ornithology, Ithaca, NY, USA. Google Scholar
  59. Verdugo, C., A. M. Clark, D. Prakoso, L. D. Kramer, and M. T. Long (2012). Multiplexed microsatellite loci in American Crow (Corvus brachyrhynchos): A severely affected natural host of West Nile virus. Infection, Genetics and Evolution 12:1968–1974. Google Scholar
  60. Ward, M. P., and A. Raim (2011). The fly-and-social foraging hypothesis for diurnal migration: Why American Crows migrate during the day. Behavioral Ecology and Sociobiology 65:1411–1418. Google Scholar
  61. Weis, A. M., D. B. Storey, C. C. Taff, A. K. Townsend, B. C. Huang, N. T. Kong, K. A. Clothier, A. Spinner, B. A. Byrne, and B. C. Weimer (2016). Genomic comparison of Campylobacter spp. and their potential for zoonotic transmission between birds, primates, and livestock. Applied and Environmental Microbiology 82:7165–7175. Google Scholar
  62. Wheeler, S. S., L. W. Woods, W. M. Boyce, C. D. Eckstrand, S. A. Langevin, W. K. Reisen, and A. K. Townsend (2014). West Nile virus and non-West Nile virus mortality and coinfection of American Crows (Corvus brachyrhynchos) in California. Avian Diseases 58:255–261. Google Scholar
  63. Wikelski, M., and R. Kays 2017. Movebank.  http://www.movebank.org Google Scholar
  64. Wikelski, M., R. W. Kays, N. J. Kasdin, K. Thorup, J. A. Smith, and G. W. Swenson, Jr. (2007). Going wild: What a global small-animal tracking system could do for experimental biologists. Journal of Experimental Biology 210:181–186. Google Scholar
  65. Zink, R. M. (2011). The evolution of avian migration. Biological Journal of the Linnean Society 104:237–250. Google Scholar
© 2018 American Ornithological Society.
Andrea K. Townsend, Blaire Frett, Anne McGarvey and Conor C. Taff "Where do winter crows go? Characterizing partial migration of American Crows with satellite telemetry, stable isotopes, and molecular markers," The Auk 135(4), (8 August 2018). https://doi.org/10.1642/AUK-18-23.1
Received: 31 January 2018; Accepted: 31 May 2018; Published: 8 August 2018
JOURNAL ARTICLE
11 PAGES


SHARE
ARTICLE IMPACT
Back to Top