Skip to main content

Synchronous timing of return to breeding sites in a long-distance migratory seabird with ocean-scale variation in migration schedules



Migratory birds generally have tightly scheduled annual cycles, in which delays can have carry-over effects on the timing of later events, ultimately impacting reproductive output. Whether temporal carry-over effects are more pronounced among migrations over larger distances, with tighter schedules, is a largely unexplored question.


We tracked individual Arctic Skuas Stercorarius parasiticus, a long-distance migratory seabird, from eight breeding populations between Greenland and Siberia using light-level geolocators. We tested whether migration schedules among breeding populations differ as a function of their use of seven widely divergent wintering areas across the Atlantic Ocean, Mediterranean Sea and Indian Ocean.


Breeding at higher latitudes led not only to later reproduction and migration, but also faster spring migration and shorter time between return to the breeding area and clutch initiation. Wintering area was consistent within individuals among years; and more distant areas were associated with more time spent on migration and less time in the wintering areas. Skuas adjusted the period spent in the wintering area, regardless of migration distance, which buffered the variation in timing of autumn migration. Choice of wintering area had only minor effects on timing of return at the breeding area and timing of breeding and these effects were not consistent between breeding populations.


The lack of a consistent effect of wintering area on timing of return between breeding areas indicates that individuals synchronize their arrival with others in their population despite extensive individual differences in migration strategies.


Stages in the annual cycle of animals are intricately linked and the timing of each event can have cascading effects on the timing of subsequent events, usually referred to as carry-over effects [1, 2]. Temporal carry-over effects may be particularly pronounced in animals that undertake long and challenging migrations between their breeding and wintering areas, where delays in one phase of the annual cycle can lead to a temporal mismatch with resource availability in subsequent phases, ultimately affecting survival and reproductive success [3, 4]. The degree to which delays can be buffered, preventing amplification of carry-over effects across time, may vary as a function of the tightness of scheduling within the annual cycle [5]. Quantifying the geographical variation in timing of annual cycles and carry-over effects is therefore important for our understanding of how phenology drives variation in life-history traits [6]. In addition, considering that daily survival rates may differ between stages of the annual cycle [7, 8], variation in annual schedules can drive meta-population dynamics [2, 9, 10]. Insight in the drivers of meta-population dynamics is important to inform location-specific conservation efforts.

Population-level differences in the scheduling of the annual cycle may arise as a function of the location of the breeding and wintering areas [11, 12] and the conditions experienced there [13]. Annual schedules are generally shifted to later in the season at higher breeding latitudes, where the optimal time for reproduction is associated with a later time of return to the breeding site [14, 15, 16, 17, but see 18]. In turn, this affects the duration and timing of subsequent phases of the annual cycle, such as migration, given that the degree of seasonality in resources and photoperiod leads to differences in optimal migration speed [19, 20]; however, this is little studied. In addition to an effect of the breeding area, the timing of annual cycles may also depend on the choice of wintering area and migration routes, as more distant non-breeding areas require longer travel time. As such, unless departure is earlier or travel speeds higher, arrival will be later [21, 22]. When fueling conditions in the wintering area itself or along the migration route are sub-optimal, delays can arise as a result of choice of wintering area. These delays can subsequently carry-over to timing of return to the breeding area and egg laying [23, 24] and can reduce survival [25]. Thus, temporal carry-over effects are expected to be more pronounced when distances between breeding and non-breeding areas are greater, and there is less leeway in the annual cycle to compensate for delays.

We define the net carry-over effect as the timing of an event for an individual or group relative to the population mean for a sequence of events (Fig. 1a). The strength of temporal carry-over effects can be conceptualized as the degree to which timing of an event affects the scheduling of a following event (Fig. 1a–c). If there is full compensation for a delay, each day lost in one phase translates to one less day spent in the next phase, resulting in a similar timing of the following event. In the case of complete carry-over of delays, the timing of the next event is delayed by a day, i.e. there is no compensation. Delays may also be amplified at later stages, i.e. each day of delay translates to more than a day delay in following events.

Fig. 1
figure 1

Conceptual framework of the two interrelated measures of carry-over effects between events in the annual cycle. The framework assumes the population mean represents the overall optimal timing, while throughout the range of timings different advantages and disadvantages can be attributed. A Here an example delay (\({d}_{1}\)) relative to the population mean timing (\({\overline{x}}_{{t}_{1}}\)) of event 1 affects the timing of a subsequent event \({t}_{2}\) relative to the population mean \({\overline{x}}_{{t}_{2}}\) to different degrees. The resulting degree of advancement or delay in event (\({t}_{2}\)) is the net of carry-over effects and is described by its strength as partial, complete, or amplified carry-over effect, or a complete compensation. The strength of the carry-over effect can be quantified as the slope between \({t}_{1}\) and \({t}_{2}\) (B), or, between \({t}_{1}\) and the duration of the period between \({t}_{1}\) and \({t}_{2}\) (C)

Although long-distance migration has evolved in many taxonomic groups [26], geographical variation in the timing of annual cycles and carry-over effects have mainly been studied in single breeding populations of migratory birds [13, 27, 28], that use a single wintering area or multiple wintering areas at similar latitudes [11, 12, 15] or in multiple breeding populations that breed at similar latitudes [29]. Ideally, quantifying and testing the relative effects of breeding and wintering location on the timing of annual schedules and on carry-over effects requires substantial geographic variation in both breeding and wintering areas and mixing of individuals from different breeding populations within wintering areas and vice versa (weak migratory connectivity [30]). Such studies are rare, because of the challenges of performing large-scale tracking studies and because few species show sufficient variation in both breeding and wintering latitudes. Hence, the extent to which the timing of events and carry-over effects in the annual cycle of migratory birds is driven by both breeding and wintering locations is largely unresolved [31].

We studied carry-over effects on the timing of different phases in the annual cycle of a long-distance migratory seabird, the Arctic Skua Stercorarius parasiticus, from multiple breeding populations. Its circumpolar breeding distribution ranges from northern temperate to high arctic zones ([32]; Fig. 2). Based on ring recoveries and field sightings, the main wintering areas in the Atlantic are thought to span productive continental shelf and pelagic areas across a large (latitudinal) range, from Iberia to Patagonia and South Africa [32, 33]. By tracking the full annual cycle of many individuals over several years from multiple breeding sites ranging from East Greenland to West Siberia (59°N to 79°N in latitude and 24°W to 69°E in longitude), we present the full extent of their wintering distribution. We quantify the migratory connectivity [30] and the consistency of individuals from year to year in their migration strategy. We quantify the relative effects of breeding and wintering areas on the timing of key events and duration of intervening periods in the annual cycle. Based on the northern hemisphere (breeding area) viewpoint on annual schedules, the non-breeding area is referred to as wintering area throughout. Using the variation in annual schedules, we study compensation versus carry-over of variation in timing between subsequent events, and how these depend on breeding and wintering areas. Later timing of events and tighter schedules are predicted to be associated with more northerly breeding areas [15, 16]; later timing is also expected to lead to different allocation of time to phases of the annual cycle. Time spent in more distant wintering areas is expected to be shorter due to longer time spent on migration. Among different annual schedules, we expect carry-over effects to arise particularly among schedules with less leeway to compensate for delays and that the wintering period generally buffers the variation in timing of autumn migration, but less so when the wintering period is shorter. As such, migrations to more distant wintering areas are expected to be associated with higher risks of delayed spring return in the breeding area—a net carry-over effect [24, 34].

Fig. 2
figure 2

Wintering areas of Arctic Skuas tracked from breeding areas between East Greenland and West Siberia. Dots represent centroids of wintering positions for each track, coloured per wintering area (boxes). Green shaded areas reflect the breeding range, with red-filled dots showing the studied breeding sites, which are enlarged in the upper panel to show pie charts representing the proportion of individuals wintering in each of the seven wintering areas, with the size corresponding to the total number of individuals tracked. Sample sizes are shown as the number of tracks/number of individuals. The main stopover in the North Atlantic is shown as a dark green polygon. Abbreviations for breeding sites are SVA = Svalbard, KVP = Karupelv, HOC = Hochstetter Forland, ICE = Iceland, FAR = Faroe Islands, FAI = Fair Isle, ROU = Rousay, SVA = Svalbard, SLE = Slettnes, BRH = Brensholmen, TOB = Tobseda, ERK = Erkuta and FIN = Finland, and for breeding areas are GRE = Greenland, SCO = Scotland, NOR = mainland Norway and RUS = Russia. Abbreviations for wintering areas are PAT = Patagonian Shelf, BEN = Benguela region, GUL = Gulf of Guinea, CAR = Caribbean region, CAN = Canary Current, MED = Mediterranean Sea and IND = Indian Ocean. Sites within countries were merged for statistical analyses


Device deployments and sample size

Adult Arctic Skuas were fitted with geolocators (Global Location Sensor or GLS) loggers at twelve breeding sites in eight countries. Data from sites within countries were combined to avoid very small sample sizes in models and are henceforth referred to as breeding areas, as follows: East Greenland (Karupelv, 72° 30′ N 24° 00′ W, and Hochstetter Forland, 75° 09′ N, 19° 40′ W), Iceland (Hroarstunga, 65° 35′ N, 14° 22′ W), Faroe Islands (Fugloy, 62° 20′ N, 6° 19′ W), Scotland (Fair Isle, 59° 54′ N, 1° 63′ W and Rousay, 59° 09′ N 3° 02′ W), mainland Norway (Brensholmen, 69° 35′ N, 18° 01′ E and Slettnes, 71° 05′ N, 28° 12′ E), Spitsbergen (Kongsfjorden, Svalbard, 79° 57′ N, 12° 06′ E), Finland (Satakunta, 61° 33′ N, 21° 27′ E), and Russia (Tobseda, 68° 36′ N, 52° 19′ E, and Erkuta, 68° 14′ N, 69° 09′ E) (Fig. 2).

Birds were captured either on the nest using bow nets, tent spring traps or walk-in traps, remote release nooses, or away from the nest using net guns. Geolocators were attached to a Darvic ring fitted around the tarsus. Various models of geolocators were used: C65, C250 (Migrate Technology, Cambridge, UK), Mk9, Mk13, Mk15 and Mk18H (British Antarctic Survey, Cambridge, UK) and Mk3006 (Biotrack, Wareham, UK). Geolocators measured ambient light in lux (Migrate Technology) or arbitrary units (BAS/Biotrack) every 1 min and saved the maximum value every 5 min. In subsequent breeding seasons, geolocators were removed and data downloaded.

In total, data for 276 interbreeding periods (i.e. one migration cycle, referred to as a ‘track’ below) from 155 individuals were obtained in 2009–2021. Sample sizes per breeding area ranged from 103 tracks of 57 individuals in mainland Norway to 8 tracks of 3 individuals in East Greenland (Fig. 2). In total, 34, 23 and 12 individuals were tracked for 2, 3 and 4–6 years, respectively. Some of the tracks (n = 48) were incomplete due to battery or electronics failure.

Geolocator data processing

Positions were estimated for noon and midnight from the light measurements by the geolocators (see electronic Additional file 1 for details).

We describe the phenology from the perspective of a breeding bird, i.e. ‘autumn migration’ is post-breeding, and ‘spring migration’ is pre-breeding. We define the following key events in the annual cycle: (1) departure from the breeding area; (2) arrival in the wintering area; (3) departure from the wintering area; (4) arrival in the North Atlantic north of 35°N (mostly around the sub-polar frontal zone where many Arctic Skuas made a stopover before returning to the breeding grounds [35, 36]); (5) return to the breeding area; (6) clutch initiation. The duration of the intervening periods (in days) were calculated for the: (1) autumn migration; (2) wintering period; (3) spring migration, including the time in the North Atlantic; (4) time in the North Atlantic; (5) pre-laying period; (6) breeding period. As loggers were deployed in the colony and retrieved before departure, a composite index of the length of the breeding period for each track was calculated as the sum of the periods from 1 July (all individuals present in the breeding area) to departure in the first year, and from clutch initiation in the second year to 30 June. Timing of movements were identified by inspecting raw position estimates. For birds breeding in areas with continuous daylight, first position estimates after leaving the zone with continuous daylight were close to the breeding site, but last position estimates, just before returning to the zone with continuous daylight, were around 60°N and distant from the breeding area. To estimate time of departure from or arrival to the breeding area under continuous daylight, we assumed birds traveled at a speed of ca. 850 km d−1, which has been recorded for the closely related Long-tailed Skua Stercorarius longicaudus [37], between the breeding site and the first or last position estimate. Rapid movement to or away from the wintering area, roughly entering or leaving the boxes shown in Fig. 2, was taken as the timing of arrival at and departure from the wintering area. The date of arrival in the North Atlantic during spring migration was taken as the day birds crossed 35°N. For birds wintering in the Mediterranean Sea, the day after wintering area departure was taken. Clutch initiation dates were inferred from the light data, as incubation stints are apparent as alternating periods of light and darkness each lasting more than an hour [38].

Based on positions in the wintering areas, tracks were assigned to one of seven discrete areas: (1) Caribbean region, Guiyana, Gulf of Mexico and north of Brazil; (2) Patagonian Shelf; (3) from the Benguela region west to Tristan da Cuñha; (4) Gulf of Guinea and west towards eastern Brazil; (5) Canary Current and off north-west Africa/Iberia; (6) Mediterranean Sea; (7) Indian Ocean and adjacent seas (Fig. 2). Although skuas moved during the wintering period, sometimes over substantial distances (mean = 1009 km; range = 47–2735 km), none had arrival and departure positions in different wintering areas.

Statistical analysis

Wintering centroids were calculated as the geographical mean of positions between arrival and departure. Minimum migration distance was calculated as the great-circle distance between the wintering centroid and the breeding site. As a measure of migratory connectivity, the degree to which individuals from the same breeding site stay together at wintering areas [30], we calculated the Mantel correlation \({R}_{mantel}\) using the ‘MigConnectivity’ package in R [39], using the first track of each individual. Significance of \({R}_{mantel}\) was assessed by 1000 random permutations.

To quantify individual consistency in choice of wintering area, inter-centroid distances were measured between successive years in (a) the same individual and (b) different individuals from the same breeding site [40]. To test for greater consistency within than between individuals, a random distribution was generated 10,000 times by re-distributing within- and between-individual labels and calculating the within- and between-individual distances between centroids. In addition, the frequency with which individuals switched between wintering areas was assessed.

Timing of key events and duration of periods were modeled as a function of breeding latitude, with random intercepts for each combination of breeding and wintering area as well as for individuals, in Bayesian Generalized Linear Mixed-effects Models (GLMMs).

We then assessed whether wintering in certain areas led to earlier or later arrival compared to birds from the same breeding area, by constructing GLMMs with breeding and wintering area as fixed effects and random intercepts for individuals. We choose Bayesian GLMMs for these and other models as these often perform better when sample sizes are small and do not require adjustments for multiple comparisons [41]. Data with at least three birds for each combination of breeding and wintering areas were selected. We performed two analyses to show the effect of wintering area. First, all pairwise comparisons of wintering area effects within breeding areas were made based on the posterior distributions of the contrast between each set of two wintering areas. Posterior contrast probabilities where 0 was outside the 95% high density interval (HDI) were considered important. Second, to test whether wintering areas led to advances or delays relative to birds from the same breeding area but wintering elsewhere, the 95% HDI of the posterior distributions of each breeding-wintering area combination were compared to the region of practical equivalence (ROPE), defined as the mean timing of birds from the same breeding area but wintering elsewhere, ± 0.1 * sd [42]. If the 95% HDI of the posterior distribution fell completely outside the ROPE, the wintering area was taken to have an important effect. Variance components of the fixed and random effects were used to show partitioning of variation between and within individuals (see Additional file 1).

To show partitioning of variation among breeding and wintering area as well as between and within individuals, variance components were extracted for the fixed effects \({\sigma }_{\beta }^{2}\) as well as between \({\sigma }_{\alpha }^{2}\) and within individuals \({\sigma }_{\epsilon }^{2}\) of the GLMM with timing of events or duration of periods as the response variable, and breeding and wintering area as fixed effects. Variance components were also used to calculate the marginal repeatability as \({R}_{m}={\sigma }_{\alpha }^{2}/\left({\sigma }_{\alpha }^{2}+{\sigma }_{\epsilon }^{2}\right)\) [43].

Using only data from breeding areas with the largest data sets (FAR: n = 27 tracks, NOR: n = 103 tracks and SVA: n = 100 tracks), we tested whether the duration of periods between events changed according to timing of the onset of each period, by fitting GLMMs with duration as the response, the onset of the period as a fixed effect, random intercepts and slopes for breeding-wintering area combinations and random intercepts for individuals. To test whether the duration of periods also changed as a function of timing within each breeding-wintering area combination, GLMMs were constructed with duration as the response, slopes with timing per breeding-wintering area as fixed effects and random intercepts for individuals.

GLMMs were fitted using the brms package in R [44], which provides an interface to stan [45], with 6000 iterations, a burn-in of 2000, a thinning interval of five and default uninformative priors. Model results were checked for mixing of and autocorrelation within the chains, and for convergence (\(\widehat{R}\) ~ 1).


Migratory connectivity and spatial consistency of individuals

For any given colony, there was a range of wintering areas used by individual Arctic Skuas (Fig. 2). The largest variation was among individuals from Slettnes, mainland Norway (largest sample size, n = 46 individuals), which wintered in all seven areas identified in this study. Variation among individuals was smallest in the small samples from Russian sites that all wintered in the Indian Ocean (n = 6), and from Greenland, that all wintered in the Caribbean region (n = 3). Arctic Skuas breeding in Svalbard (n = 42) wintered primarily in the Caribbean region (51%), and none migrated to the southern wintering areas, the Patagonian Shelf and Benguela region. In contrast to the Svalbard population, few individuals from mainland Norway wintered in the Caribbean region (9%, n = 57). A considerable proportion of individuals breeding in mainland Norway and Finland wintered on the Patagonian Shelf (43%, n = 63). Birds breeding in Iceland, the Faroe Islands and Scotland wintered mainly in the Benguela region (31%), Canary Current (36%) and the Patagonian shelf (23%, n = 39), but none in the Caribbean or Mediterranean Sea.

Individual wintering-site fidelity was very high: out of 69 individuals tracked for 2–6 years (n = 190 tracks), only two birds (2.9%) switched area between years. The first individual, a bird breeding in Brensholmen (mainland Norway), went to the Gulf of Guinea in the first two years of tracking, but wintered in the Canary Current in the third year. Another individual, breeding in Svalbard, wintered in the Caribbean region in the first and in the Gulf of Guinea in the second year. The large overall consistency in wintering area is reflected in the smaller distance between wintering centroid of the same individuals (485 ± 558 km) compared with those of different individuals from the same breeding area (5634 ± 3152 km, p < 0.001).

The considerable degree of mixing of individuals from different breeding sites in wintering areas indicated weak overall migratory connectivity, as is expressed in the \({R}_{mantel}\) of 0.19 (95% HDI: 0.11–0.26); excluding the birds from the Russian sites resulted in a slightly lower \({R}_{mantel}\) of 0.14 (95% HDI: 0.07–0.22).

During both autumn and spring migration, most birds wintering in the Atlantic staged in a broad area across the central North Atlantic, roughly between 35°N and 55°N (Fig. 2). Individuals breeding at Svalbard and wintering in the Canary Current, however, skipped this staging area in spring and followed a more eastern route along Western Europe [36].

Annual schedules in relation to migration strategies

Birds at more northerly colonies started breeding (i.e. laid the first egg) later (Fig. 3, Additional file 1: Fig. S1) and also started and finished migration later. The duration of spring migration, the time spent in the North Atlantic and the time between return to the breeding area and clutch initiation all decreased with breeding latitude (Fig. 3). However, the duration of autumn migration, wintering and breeding periods did not change with breeding latitude (Fig. 3).

Fig. 3
figure 3

Relation between breeding latitude and timing (upper panels) or duration (lower panels) in Arctic Skuas tracked from breeding areas between East Greenland and West Siberia to wintering areas in the Atlantic. Shaded areas are 95% HDIs

Birds that used different wintering areas also showed differences in timing and duration of migration (Fig. 4). There was a clear trend of longer migration periods and shorter wintering period when wintering further south for all breeding areas. Birds wintering on the Patagonian Shelf and the Benguela region (the two furthest wintering areas, at 12,647–14,664 km and 9803–11800 km, respectively, from colonies) generally stayed for shorter periods in the wintering areas and had longer spring migrations than individuals wintering in the Canary Current (the closest wintering area, at 4021–7140 km from colonies), with the Guinean and Caribbean regions showing intermediate duration of migration and wintering periods. Birds from the same breeding population returned to the North Atlantic at the same time regardless of wintering area, with the exception of later timing among birds wintering on the Patagonian Shelf, and earlier timing among Norwegian birds wintering in the Canary Current. In all breeding areas except mainland Norway, the timing of return to the breeding area was similar between birds from different wintering areas. Among mainland Norwegian birds, we found weak evidence that return to the breeding grounds was slightly delayed among birds that wintered on the Patagonian Shelf and was advanced among those that wintered in the Canary Current. Although timing of clutch initiation varied among Norwegian birds—birds that wintered in the Caribbean region bred substantially earlier—there was no relation between wintering area and clutch initiation date across breeding areas.

Fig. 4
figure 4

Parameter estimates of the timing of events (upper panels) or duration of phases (lower panels) per breeding and wintering area for Arctic Skuas tracked from breeding areas in the North Atlantic to wintering areas in the Atlantic. Breeding areas are ordered from South (SCO) to North (SVA) and wintering areas from North (CAN) to South (PAT). Error bars show 95% HDIs. Within breeding areas, estimates accompanied by a common letter are equivalent based on whether 0 was included in the 95% HDI of their contrast posterior distribution (values are in Additional file 1: Table S1). Filled squares on the right y-axes indicate that the 95% HDI was outside the ROPE of birds from the same colony wintering elsewhere. Colours correspond to the wintering areas in Fig. 2

From wintering area arrival to clutch initiation in the next breeding season, individuals were increasingly consistent in their timing (Fig. 5a). The variance between individuals was substantially higher than the variance within individuals for timing of arrival to and departure from the wintering area and arrival in the North Atlantic, leading to highest marginal repeatabilities (\({R}_{m}\) ~ 0.7). Variance between and within individuals was similar for the timing of return to the breeding area and clutch initiation, with accordingly lower marginal repeatabilities (\({R}_{m}\) ~ 0.55 and 0.25, respectively, Fig. 5c). With respect to the duration of periods, individuals were most consistent in the time spent in the North Atlantic and least in the duration of the wintering period (Fig. 5b). Similarly, the repeatability of the duration of periods was ca. 0.7 for autumn migration to the time spent in the North Atlantic, after which it dropped to ca. 0.4 due to similar variances between and within individuals (Fig. 5d).

Fig. 5
figure 5

Marginal repeatability (upper panels) and squared posterior means of model variance components of GLMMs of timing and duration (lower panels) of timing of events (left) and duration of phases (right) in the annual cycle of Arctic Skuas tracked from breeding areas in the North Atlantic to wintering areas in the Atlantic

Strength of temporal carry-over effects

Both between and within breeding populations, the duration of each migration phase was shorter when started later (Fig. 6). There was no consistent pattern among breeding areas in the steepness of this relationship, and slope estimates (a measure of the strength of the carry-over effect) were not consistently associated with breeding or wintering areas (Additional file 1: Fig. S2). Overall, the decline of duration with time was steepest for the wintering period (Additional file 1: Fig. S2b), when slope estimates around − 1 indicate complete compensation of the timing of arrival in the wintering area (Fig. 1). There were two exceptions to this pattern: slope estimates around − 0.5 for the two farthest wintering areas suggested only partial compensation for late arrival in a) Norwegian birds wintering at the Patagonian Shelf (\(\beta\) = − 0.5, 95% HDI =  − 0.83 to − 0.15) and b) Faroese birds wintering in the Benguela region (\(\beta\) = − 0.48, 95% HDI =  − 1.2 to 0.24). Note, however, that the estimate for the Faroese birds wintering in the Benguela region was associated with a very wide HDI overlapping with both − 1 and 0, and that the slope estimate for birds from mainland Norway wintering in the Benguela region indicates complete compensation (\(\beta\) = − 0.97, 95% HDI =  − 1.53 to − 0.43). The duration of spring migration and the time spent in the North Atlantic were tightly related to their timing, with overlapping distributions across breeding and wintering areas (Fig. 6c, d). The slope estimates between − 0.5 and 0 indicated a partial to complete carry-over effect of the arrival in the North Atlantic on return to the breeding area (Fig. 6d, e). The timing of return partially carried-over to the timing of clutch initiation, with most slope estimates between − 1 and − 0.5 (Additional file 1: Fig. S2e).

Fig. 6
figure 6

The relationship between duration of each period and the timing of its start in Arctic Skuas tracked from breeding areas in the North Atlantic to wintering areas in the Atlantic. Thick lines show overall slope estimates, thin lines show regression lines per combination of breeding and wintering area. Note that the aspect ratio of all plots is 1, but the axis scales differ


Our study demonstrates substantial variation within and between breeding populations in both wintering areas and annual schedules of the Arctic Skua, a long-distance migratory seabird with a huge wintering range. Arctic Skuas showed weak overall migratory connectivity in the Atlantic, with very high variation among individuals, including individuals within breeding sites, in their use of wintering areas separated by thousands of kilometers. Individuals from the colony with the most tracked birds (Slettnes, mainland Norway) wintered in all productive areas of the Atlantic between 50°S and 30°N, and also in the Mediterranean Sea and the Persian Gulf (Fig. 2), together representing roughly a tenth of the world’s ocean surface area. The lower variability in wintering destinations of birds from other breeding sites may partly reflect smaller sample sizes. The wintering range of Arctic Skuas is as large or larger than most other well-studied long-distance migratory seabirds [46,47,48,49] and also far exceeds the variation shown by a congeneric species, the Long-tailed Skua Stercorarius longicaudus [50]. Such ocean-scale variation in wintering area selection makes the Arctic Skua exceptionally well-suited for directly testing the relationships between the timing of annual events and wintering site selection within and between breeding colonies.

Migration timing of the tracked Arctic Skuas varied principally with breeding latitude, with later-timed schedules of more northern breeders associated with faster spring migrations and shorter periods between spring arrival at the breeding grounds and clutch initiation. Variation in timing of autumn migration was completely compensated during the wintering period, even in the most distant wintering areas, where skuas spent less time. Despite the large variation in migration distances and schedules and the strong carry-over effect of timing during spring migration, the timing of return to the breeding area was highly synchronized among individuals from the same site, leaving no substantial net carry-over effect.

In Arctic Skuas, breeding latitude not only shifted the timing of annual schedules, as expected [12, 15, 16, 51], but later timing also affected the duration of subsequent key periods within the annual cycle. Most notably, Arctic Skuas breeding at high latitudes had faster spring migrations and shorter pre-laying periods compared to birds breeding at lower latitudes. The faster migration of later-timed spring migration of Arctic Skuas contrasts with little or no geographical variation in the duration of spring migration of two insectivorous songbirds [16, 51] and a long-distance migratory seabird, the Arctic Tern Sterna paradisaea [52]. Arctic Skuas may face faster seasonal changes in conditions encountered en route or may be more flexible in where they fuel for migration/reproduction, by regulating the amount of pre-migratory fueling in the wintering area, foraging en route and the time spent in staging areas, in particular the North Atlantic staging area where Arctic Terns stage for no or few days in spring [52]. Arctic Skuas from Svalbard spent fewer days during spring migration at stopovers compared to birds from more southern breeding areas (Fig. 6d; [36]), which may be explained by deteriorating foraging conditions in the North Atlantic for Arctic Skuas during April to May, when the potential host species to kleptoparasitize [53], such as Arctic Terns and Sabine’s Gull Xema sabini, return to their breeding areas [52, 54, 55]. Shorter time spent at stopovers suggests that birds from more northern breeding areas need to fuel up before migration rather than depend on foraging en route to fuel the migration flight, which is common among migrating seabirds [56], and to return with large body reserves in the breeding area [57]. These body stores are also used for egg production: in a population breeding at 61°N, a latitude similar to our most southern study sites and thus presumably with a relatively long pre-laying period, 16% of the protein in eggs originated from distantly-acquired resources [58]. Carrying large nutrient stores on migration is however costly and hence less beneficial as migratory distance increases. This trade-off potentially explains why Arctic Skuas breeding in the high Arctic do not migrate to the southernmost wintering areas. However, carrying large nutrient stores on migration can also be advantageous as it increases flight speed when adopting the ‘dynamic soaring’ flight technique [59], which skuas utilize at high wind speeds. Furthermore, carrying nutrient stores is relatively less costly in larger-bodied birds [19]. Interestingly, Arctic Skuas from Svalbard were heavier and larger than those from mainland Norway, the Faroe Islands and Scotland (B. Moe, R.S.A. van Bemmelen, K.R.S. Snell, R.A. Phillips unpublished data). Selection for larger body size to reduce the relative costs of carrying nutrient stores during migration provides an alternative hypothesis to explain latitudinal clines in body size, for which Bergman’s rule is usually invoked [60].

Arctic Skuas were able to compensate for the timing of the autumn migration during the wintering period. Thereby, our study adds to a growing number of studies of ducks, waders and passerines reporting the wintering period to be the main period in the annual cycle where delays are compensated [17, 27, 61, 62], although the migration periods may additionally act as time buffer in some species [63, 64]. The ability to shorten the wintering period to compensate for later timing of autumn migration has also been shown by experimentally increasing the duration of parental care [65]. We investigated whether birds are less able to compensate for timing of autumn migration if the wintering period is shorter, as observed in more distant wintering areas, and evidence was inconsistent. Indeed, reduced compensation in the wintering period was found among skuas breeding in northern Norway and wintering at the furthest wintering area, the Patagonian Shelf (Fig. 4), where the wintering period is up to 50 d shorter than in the northernmost wintering area (but still about 50 d longer than the 122 d needed for the moult of the primary feathers; [66]). However, no conclusive evidence for reduced compensation was found among individuals wintering in the Benguela region, at a similar latitude and migration distance. Therefore, site-specific effects, rather than migration distance alone, may explain the difference in the degree to which birds compensate earlier timing variation during the wintering period, which corresponds to findings for the Sanderling Calidris alba [13].

Our study shows largely synchronized timing of return to the breeding grounds as well as clutch initiation within breeding sites, despite differences between wintering areas in migration distance of thousands of kilometers and in average spring migration duration of up to 19 d (in Scotland). Despite the longer duration of spring migration from more distant wintering areas and the increasing strength of the carry-over effect of timing during spring migration, the similar departure dates across wintering areas did not result in substantial net carry-over effects on the breeding grounds (Fig. 1). This indicates that, rather than departing earlier from the wintering grounds, Arctic Skuas with longer migrations increased travel speeds to ensure a timely return to the breeding grounds, which agrees with empirical data from songbirds [21]. The similarity in timing of departure across wintering areas observed in Arctic Skuas suggests control by endogenous biological ‘clocks’, which are thought to be fine-tuned by selection [31]. While such endogenous rhythms may dictate the general timing of departure, individuals select the timing of their departure relative to their nutrient stores and environmental conditions [67, 68] and may optimise their timing with age [69, 70]. Indeed, we found that individuals varied their timing of departure from the wintering area from year to year, with a standard deviation of about a week (Fig. 5). Whether this variation represents sufficient flexibility for Arctic Skuas to adjust to fast directional environmental change, such as the effects of global warming on oceanic primary productivity in the North Atlantic [71], is yet unclear.

Birds from widely separated wintering areas arrived back in the breeding area, on average, within less than a week. This synchrony underpins the strong selection on timing of spring arrival and breeding when competition for territories is high [72]. Later arrival at the breeding grounds of individuals from more distant wintering areas has been reported for several short- to long-distance migrants [28, 73,74,75,76,77], but only in European Shags Gulosus aristotelis and Eurasian Spoonbills Platalea leucorodia did this also lead to later reproduction (6 d in shags, 12 d in male spoonbills) and lower reproductive success. In Arctic Skuas, however, we found partial compensation for later return at the breeding grounds: on average, a 1 d later spring arrival led to 0.6 d later clutch initiation, but with substantial individual variation. We did not have data on reproductive success, but as wintering area had a minimal influence on timing of return to the breeding area (on average up to 6 d later) and laying dates (up to 3 d), a substantial effect of wintering area on reproductive success via timing of breeding seems unlikely. Furthermore, unless an effect on productivity is substantial, demonstrating a carry-over effect on reproduction would be difficult considering the multitude of local effects at the breeding grounds—effects that have been shown to impact reproduction and thus drive population declines in Arctic Skuas, at least in Scotland and mainland Norway [78, 79].

By describing the migratory connectivity and ocean-scale geographic variation in annual schedules and carry-over effects on timing of a long-distance migratory seabird, our study contributed to our understanding of potential drivers of meta-population dynamics in migratory species. Our results indicate that choice of wintering area is unlikely to affect population trends through a carry-over effect on timing of breeding, irrespective of breeding latitude. Of course, this does not rule out effects on survival or reproduction via other mechanisms. For example, as daily survival rate may differ between stages of the annual cycle [7, 8], wintering area selection may affect annual survival rates and therefore population trends, given their concomitant annual schedules. While the weak migratory connectivity means that conditions in each wintering area can affect multiple breeding populations, each breeding population may be buffered from the effect of conditions at specific wintering areas due to the large spread of individuals across wintering areas [30]. Interestingly, the population decline in Svalbard (from where no individuals wintered in the two most distant wintering areas) is less than in the Faroe Islands, Scotland and mainland Norway [78,79,80] and the longevity record holder of the species, a bird from Finland, wintered in one of the closer wintering areas, the Canary Current [81]. If and how breeding area, wintering area, and the corresponding annual schedules affect seasonal survival rates in Arctic Skuas [82] across populations, and whether these can explain differences in population trends, remains to be investigated.

Availability of data and materials

Geolocator data is stored on under study IDs: 2746281685 (Svalbard, Norway), 968984187 (Slettnes, Norway), 2746341928 (Brensholmen, Norway), 976692287 (Tobseda, Russia), 1260558178 (Greenland), 2746320914 (Iceland), 2746186408 (Faroe Islands), 2743674038 (Scotland) and 2747912295 (Finland). The studies can also be found by searching for the study IDs using the search functionality at this website.


  1. Piersma T. Hink, stap of sprong? Reisbeperkingen van arctische steltlopers door voedselzoeken, vetopbouw en vliegsnelheid. Limosa. 1987;60:185–94.

    Google Scholar 

  2. Harrison XA, Blount JD, Inger R, Norris DR, Bearhop S. Carry-over effects as drivers of fitness differences in animals. J Anim Ecol. 2011;80:4–18.

    Article  PubMed  Google Scholar 

  3. Alerstam T. Bird migration. Cambridge: Cambridge University Press; 1990.

    Google Scholar 

  4. Norris DR, Taylor CM. Predicting the consequences of carry-over effects for migratory populations. Biol Lett. 2006;2:148–51.

    Article  PubMed  Google Scholar 

  5. Wingfield JC. Organization of vertebrate annual cycles: implications for control mechanisms. Philos Trans R Soc B: Biol Sci. 2008;363:425–41.

    Article  Google Scholar 

  6. Marra PP, Cohen EB, Loss SR, Rutter JE, Tonra CM. A call for full annual cycle research in animal ecology. Biol Lett. 2015;2015(11):0552.

    Google Scholar 

  7. Klaassen RHG, Hake M, Strandberg R, Koks BJ, Trierweiler C, Exo K-M, et al. When and where does mortality occur in migratory birds? Direct evidence from long-term satellite tracking of raptors. J Anim Ecol. 2014;83:176–84.

    Article  PubMed  Google Scholar 

  8. Buechley ER, Oppel S, Efrat R, Phipps WL, Carbonell Alanís I, Álvarez E, et al. Differential survival throughout the full annual cycle of a migratory bird presents a life-history trade-off. J Anim Ecol. 2021;90:1228–38.

    Article  PubMed  Google Scholar 

  9. Norris DR, Marra PP. Seasonal interactions, habitat quality, and population dynamics in migratory birds. The Condor. 2007;109:535–47.

    Article  Google Scholar 

  10. Bauer S, Lisovski S, Hahn S. Timing is crucial for consequences of migratory connectivity. Oikos. 2016;125:605–12.

    Article  Google Scholar 

  11. Bogdanova MI, Butler A, Wanless S, Moe B, Anker-nilssen T, Frederiksen M, et al. Multi-colony tracking reveals spatio-temporal variation in carry-over effects between breeding success and winter movements in a pelagic seabird. Mar Ecol Prog Ser. 2017;578:167–81.

    Article  Google Scholar 

  12. Gow EA, Burke L, Winkler DW, Knight SM, Bradley DW, Clark RG, et al. A range-wide domino effect and resetting of the annual cycle in a migratory songbird. Proc R Soc B: Biol Sci. 2018;286:20181916.

    Article  Google Scholar 

  13. Reneerkens J, Versluijs TSL, Piersma T, Alves JA, Boorman M, Corse C, et al. Low fitness at low latitudes: wintering in the tropics increases migratory delays and mortality rates in an Arctic breeding shorebird. J Anim Ecol. 2019;89:671–932.

    Google Scholar 

  14. van der Jeugd HP, Eichhorn G, Litvin KE, Stahl J, Larsson K, van der Graaf AJ, et al. Keeping up with early springs: rapid range expansion in an avian herbivore incurs a mismatch between reproductive timing and food supply. Glob Change Biol. 2009;15:1057–71.

    Article  Google Scholar 

  15. Conklin JR, Battley PF, Potter MA, Fox JW. Breeding latitude drives individual schedules in a trans-hemispheric migrant bird. Nat Commun. 2010;1:67.

    Article  PubMed  Google Scholar 

  16. Briedis M, Hahn S, Gustafsson L, Henshaw I, Träff J, Král M, et al. Breeding latitude leads to different temporal but not spatial organization of the annual cycle in a long-distance migrant. J Avian Biol. 2016;47:743–8.

    Article  Google Scholar 

  17. Gow EA, Knight SM, Bradley DW, Clark RG, Winkler DW, Bélisle M, et al. Effects of spring migration distance on tree swallow reproductive success within and among flyways. Front Ecol Evol. 2019;7:380.

    Article  Google Scholar 

  18. Pedersen L, Jackson K, Thorup K, Tøttrup AP. Full-year tracking suggests endogenous control of migration timing in a long-distance migratory songbird. Behav Ecol Sociobiol. 2018;72:139.

    Article  Google Scholar 

  19. Alerstam T, Lindström Å. Optimal bird migration: the relative importance of time, energy and safety. In: Gwinner E, editor. Bird migration: the physiology and ecophysiology. Berlin: Springer; 1990. p. 331–51.

    Chapter  Google Scholar 

  20. Lisovski S, Ramenofsky M, Wingfield JC. Defining the degree of seasonality and its significance for future research. Integr Comp Biol. 2017;57:934–42.

    Article  PubMed  Google Scholar 

  21. Morbey YE, Hedenström A. Leave earlier or travel faster—optimal mechanisms for managing arrival time in migratory songbirds. Front Ecol Evol. 2020;7:492.

    Article  Google Scholar 

  22. Sorte FAL, Fink D, Hochachka WM, DeLong JP, Kelling S. Population-level scaling of avian migration speed with body size and migration distance for powered fliers. Ecology. 2013;94:1839–47.

    Article  PubMed  Google Scholar 

  23. Marra PP, Hobson KA, Holmes RT. Linking winter and summer events in a migratory bird by using stable-carbon isotopes. Science. 1998;282:1884–6.

    Article  CAS  PubMed  Google Scholar 

  24. Tøttrup AP, Klaassen RHG, Kristensen MW, Strandberg R, Vardanis Y, Lindström Å, et al. Drought in Africa caused delayed arrival of European songbirds. Science (New York, NY). 2012;338:1307.

    Article  Google Scholar 

  25. Dossman BC, Rodewald AD, Studds CE, Marra PP. Migratory birds with delayed spring departure migrate faster but pay the costs. Ecology. 2023;104:e3938.

    Article  PubMed  Google Scholar 

  26. Dingle H. Migration. The biology of life on the move. 2nd ed. Oxford: Oxford University Press; 2014.

    Book  Google Scholar 

  27. Senner NR, Hochachka WM, Fox JW, Afanasyev V. An exception to the rule: carry-over effects do not accumulate in a long-distance migratory bird. PLoS ONE. 2014;9:e86588.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Dufour P, Wojczulanis-Jakubas K, Lavergne S, Renaud J, Jakubas D, Descamps S. A two-fold increase in migration distance does not have breeding consequences in a long-distance migratory seabird with high flight costs. Mar Ecol Prog Ser. 2021;676:117–26.

    Article  Google Scholar 

  29. Pedersen L, Onrubia A, Vardanis Y, Barboutis C, Waasdorp S, van Helvert M, et al. Remarkably similar migration patterns between different red-backed shrike populations suggest that migration rather than breeding area phenology determines the annual cycle. J Avian Biol. 2020;51:jav.02475.

    Article  Google Scholar 

  30. Webster MS, Marra PP, Haig SM, Bensch S, Holmes RT. Links between worlds: unraveling migratory connectivity. Trends Ecol Evol. 2002;17:76–83.

    Article  Google Scholar 

  31. Åkesson S, Ilieva M, Karagicheva J, Rakhimberdiev E, Tomotani B, Helm B. Timing avian long-distance migration: from internal clock mechanisms to global flights. Philos Trans R Soc B: Biol Sci. 2017;372:20160252.

    Article  Google Scholar 

  32. Furness RW. The Skuas. London: T. & A.D. Poyser; 1987.

    Google Scholar 

  33. Olsen KM, Larsson H. Skuas and jaegers. A guide to the skuas and jaegers of the world. East Sussex: Pica Press; 1997.

    Google Scholar 

  34. Briedis M, Hahn S, Adamík P. Cold spell en route delays spring arrival and decreases apparent survival in a long-distance migratory songbird. BMC Ecol. 2017;17:11.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Davies TE, Carneiro APB, Tarzia M, Wakefield E, Hennicke JC, Frederiksen M, et al. Multispecies tracking reveals a major seabird hotspot in the North Atlantic. Conserv Lett. 2021;14:e12824.

    Article  Google Scholar 

  36. O’Hanlon NJ, van Bemmelen RSA, Snell KRS, Conway GJ, Thaxter CB, Aiton H, et al. Atlantic populations of a declining oceanic seabird have complex migrations and weak migratory connectivity to staging areas. Mar Ecol Prog Ser. 2024;730:113–29.

    Article  Google Scholar 

  37. Sittler B, Aebischer A, Gilg O. Post-breeding migration of four Long-tailed Skuas (Stercorarius longicaudus) from North and East Greenland to West Africa. J Ornithol. 2011;152:375–81.

    Article  Google Scholar 

  38. Verhoeven MA, Loonstra AHJ, McBride AD, Macias P, Kaspersma W, Hooijmeijer JCEW, et al. Geolocators lead to better measures of timing and renesting in black-tailed godwits and reveal the bias of traditional observational methods. J Avian Biol. 2020;51:jav.02259.

    Article  Google Scholar 

  39. Cohen EB, Hostetler JA, Hallworth MT, Rushing CS, Sillett TS, Marra PP. Quantifying the strength of migratory connectivity. Methods Ecol Evol. 2018;9:513–24.

    Article  Google Scholar 

  40. McFarlane Tranquilla LA, Montevecchi WA, Fifield DA, Hedd A, Gaston AJ, Robertson GJ, et al. Individual winter movement strategies in two species of murre (Uria spp) in the Northwest Atlantic. PLoS ONE. 2014;9:e90583.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Gelman A, Hill J, Yajima M. Why we (usually) don’t have to worry about multiple comparisons. J Res Educ Effect. 2012;5:189–211.

    Google Scholar 

  42. Kruschke JK. Doing Bayesian data analysis: A tutorial with R, JAGS, and Stan. 2014.

  43. Nakagawa S, Schielzeth H. Repeatability for Gaussian and non-Gaussian data: a practical guide for biologists. Biol Rev. 2010;85:935–56.

    Article  PubMed  Google Scholar 

  44. Bürkner P-C. Brms: an R package for Bayesian multilevel models using Stan. J Stat Softw. 2017;80:1–28.

    Article  Google Scholar 

  45. Team SD. Stan Modeling Language Users Guide and Reference Manual. 2021.

  46. Dias MP, Granadeiro JP, Phillips RA, Alonso H, Catry P. Breaking the routine: Individual Cory’s shearwaters shift winter destinations between hemispheres and across ocean basins. Proc Biol Sci/R Soc. 2011;278:1786–93.

    Google Scholar 

  47. Phillips RA, Catry P, Silk JRD, Bearhop S, McGill R, Afanasyev V, et al. Movements, winter distribution and activity patterns of Falkland and brown skuas: insights from loggers and isotopes. Mar Ecol Prog Ser. 2007;345:281–91.

    Article  Google Scholar 

  48. Delord K, Cherel Y, Barbraud C, Chastel O, Weimerskirch H. High variability in migration and wintering strategies of brown skuas (Catharacta antarctica lonnbergi) in the Indian Ocean. Polar Biol. 2017;41:59–70.

    Article  Google Scholar 

  49. Weimerskirch H, Tarroux A, Chastel O, Delord K, Cherel Y, Descamps S. Population-specific wintering distributions of adult south polar skuas over three oceans. Mar Ecol Prog Ser. 2015;538:229–37.

    Article  CAS  Google Scholar 

  50. van Bemmelen R, Moe B, Hanssen S, Schmidt N, Hansen J, Lang J, et al. Flexibility in otherwise consistent non-breeding movements of a long-distance migratory seabird, the long-tailed skua. Mar Ecol Prog Ser. 2017;578:197–211.

    Article  Google Scholar 

  51. Neufeld LR, Muthukumarana S, Fischer JD, Ray JD, Siegrist J, Fraser KC. Breeding latitude is associated with the timing of nesting and migration around the annual calendar among Purple Martin (Progne subis) populations. J Ornithol. 2021;162:1009–24.

    Article  Google Scholar 

  52. Wong JB, Lisovski S, Alisauskas RT, English WB, Harrison A-L, Kallett DK, et al. Variation in migration behaviors used by Arctic Terns (Sterna paradisaea) breeding across a wide latitudinal gradient. Polar Biol. 2022;45:909–22.

    Article  Google Scholar 

  53. Furness BL. The feeding behaviour of Arctic Skuas Stercorarius parasiticus wintering off South Africa. Ibis. 1983;125:245–51.

    Article  Google Scholar 

  54. Stenhouse IJ, Egevang C, Phillips RA. Trans-equatorial migration, staging sites and wintering area of Sabine’s Gulls Larus sabini in the Atlantic Ocean. Ibis. 2012;154:42–51.

    Article  Google Scholar 

  55. Hromádková T, Pavel V, Flousek J, Briedis M. Seasonally specific responses to wind patterns and ocean productivity facilitate the longest animal migration on Earth. Mar Ecol Prog Ser. 2020;638:1–12.

    Article  Google Scholar 

  56. Bonnet-Lebrun A-S, Dias MP, Phillips RA, Granadeiro JP, de Brooke ML, Chastel O, et al. Seabird migration strategies: flight budgets, diel activity patterns, and lunar influence. Front Mar Sci. 2021;8:683071.

    Article  Google Scholar 

  57. Belopol’skii LO. Ecology of sea colony birds of the Barents Sea. Jerusalem: Israel Programme for Scientific Translations; 1961.

  58. Hobson KA, Sirois J, Gloutney ML. Tracing nutrient allocation to reproduction with stable isotopes: a preliminary investigation using colonial waterbirds of Great Slave Lake. Auk. 2000;117:760–74.

    Article  Google Scholar 

  59. Pennycuick CJ. Modelling the flying bird. London: Academic Press; 2008.

    Google Scholar 

  60. Bergmann C. Ober die Verhaltnisse der Warmeokonomie der Thiere zu ihrer Grosse. Gott Stud. 1847;3:595–708.

    Google Scholar 

  61. Briedis M, Krist M, Král M, Voigt CC, Adamík P. Linking events throughout the annual cycle in a migratory bird non-breeding period buffers accumulation of carry-over effects. Behav Ecol Sociobiol. 2018;72:93.

    Article  Google Scholar 

  62. Carneiro C, Gunnarsson TG, Alves JA. Annual schedule adjustment by a long-distance migratory bird. Am Nat. 2023;201:353–62.

    Article  PubMed  Google Scholar 

  63. Stutchbury BJM, Gow EA, Done T, MacPherson M, Fox JW, Afanasyev V. Effects of post-breeding moult and energetic condition on timing of songbird migration into the tropics. Proc R Soc B: Biol Sci. 2011;278:131–7.

    Article  Google Scholar 

  64. van Wijk RE, Schaub M, Bauer S. Dependencies in the timing of activities weaken over the annual cycle in a long-distance migratory bird. Behav Ecol Sociobiol. 2017;71:1–8.

    Google Scholar 

  65. Fayet AL, Freeman R, Shoji A, Kirk HL, Padget O, Perrins CM, et al. Carry-over effects on the annual cycle of a migratory seabird: an experimental study. J Anim Ecol. 2016;85:1516–27.

    Article  PubMed  PubMed Central  Google Scholar 

  66. van Bemmelen RSA, Clarke RH, Pyle P, Camphuysen CJ. Timing and duration of primary molt in Northern Hemisphere skuas and jaegers. Auk. 2018;135:1043–54.

    Article  Google Scholar 

  67. Åkesson S, Helm B. Endogenous programs and flexibility in bird migration. Front Ecol Evol. 2020;8:78.

    Article  Google Scholar 

  68. Burnside RJ, Salliss D, Collar NJ, Dolman PM. Birds use individually consistent temperature cues to time their migration departure. Proc Natl Acad Sci. 2021;118:e2026378118.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  69. Sergio F, Tanferna A, De Stephanis R, Jiménez LL, Blas J, Tavecchia G, et al. Individual improvements and selective mortality shape lifelong migratory performance. Nature. 2014;515:410–3.

    Article  CAS  PubMed  Google Scholar 

  70. Battley PF, Conklin JR, Parody-Merino ÁM, Langlands PA, Southey I, Burns T, et al. Interacting roles of breeding geography and early-life settlement in godwit migration timing. Front Ecol Evol. 2020;8:52.

    Article  Google Scholar 

  71. Edwards M, Hélaouët P, Goberville E, Lindley A, Tarling GA, Burrows MT, et al. North Atlantic warming over six decades drives decreases in krill abundance with no associated range shift. Commun Biol. 2021;4:644.

    Article  PubMed  PubMed Central  Google Scholar 

  72. Kokko H. Competition for early arrival in migratory birds. J Anim Ecol. 1999;68:940–50.

    Article  Google Scholar 

  73. Bearhop S, Fiedler W, Furness RW, Votier SC, Waldron S, Newton J, et al. Assortative mating as a mechanism for rapid evolution of a migratory divide. Science. 2005;310:502–4.

    Article  CAS  PubMed  Google Scholar 

  74. Bregnballe T, Frederiksen M, Gregersen J. Effects of distance to wintering area on arrival date and breeding performance in Great Cormorants. Ardea. 2006;94:619–30.

    Google Scholar 

  75. Grist H, Daunt F, Wanless S, Burthe SJ, Newell MA, Harris MP, et al. Reproductive performance of resident and migrant males, females and pairs in a partially migratory bird. J Anim Ecol. 2017;86:1010–21.

    Article  PubMed  PubMed Central  Google Scholar 

  76. Lok T, Veldhoen L, Overdijk O, Tinbergen JM, Piersma T. An age-dependent fitness cost of migration? Old trans-Saharan migrating spoonbills breed later than those staying in Europe, and late breeders have lower recruitment. J Anim Ecol. 2017;86:998–1009.

    Article  PubMed  Google Scholar 

  77. Nightingale J, Gill JA, Gunnarsson TG, Rocha AD, Howison RA, Hooijmeijer JCEW, et al. Does early spring arrival lead to early nesting in a migratory shorebird? Insights from remote tracking. Ibis. 2023.

  78. Perkins A, Ratcliffe N, Suddaby D, Ribbands B, Smith C, Ellis P, et al. Combined bottom-up and top-down pressures drive catastrophic population declines of Arctic skuas in Scotland. J Anim Ecol. 2018;87:1497–748.

    Article  Google Scholar 

  79. van Bemmelen RSA, Schekkerman H, Hin V, Pot M, Janssen K, Ganter B, et al. Heavy decline of the largest European Arctic Skua Stercorarius parasiticus colony: interacting effects of food shortage and predation. Bird Study. 2021;68:13.

    Google Scholar 

  80. Santos I. Survival and breeding success of the declining Arctic Skua population of the Faroe Islands. University of Copenhagen, Denmark. MSc Thesis. 2018.

  81. Mäntylä E, Mäntylä K, Nuotio J, Nuotio K, Sillanpää M. Longevity record of arctic skua (Stercorarius Parasiticus). Ecol Evol. 2020;ece3.6875.

  82. Santos IAM dos, Snell KRS, Bemmelen RS van, Moe B, Thorup K. Wintering, rather than breeding, oceanic conditions contribute to declining survival in a long-distance migratory seabird. bioRxiv; 2023. p. 2023.11.10.566398.

  83. Fair JM, Jones J. Guidelines to the use of wild birds in research. Ornithol Counc; 2010.

Download references


This study would not have been possible without the help of the many fieldworkers and others that invested substantial time and resources. In Slettnes, we were assisted in the field by Geert Aarts, Daniël van Denderen, Jan van Dijk, Maria van Leeuwe, Daan Liefhebber, Morrison Pot, Marc van Roomen, Janne Schekkerman and Rinse van der Vliet. Other help was provided by Cees and Mimi Tesselaar, Barbara Ganther, Hans-Ulrich Rösner, Karl-Birger Strann, Jeanette Hickman, Torstein Johnsrud, Niels Westpahl and Benedicte Færevåg. In Svalbard and Brensholmen, help was provided by Erlend Lorentzen, Elise Biersma, Fokje Schaafsma, Anouk Goedknegt, Oebele Dijk, Maarten Loonen, Anette Fenstad, Elise Skottene, Liv Monica Trondrud, Nora Bjørnlid, Heidi Kilen, Eline Rypdal, Emilly Hill, Melissa Fontenille and Thomas Oudman. Expeditions to Tobseda in 2014-2015 were organized by Thomas Lameris and in 2018 by Götz Eichorn (NIOO-KNAW), and joined by Kees Schreven, Stefan Sand, Jasper Koster and Chiel Boom. In the Faroe Islands, help was provided by Jón Aldará, Leivur Janus Hansen and Anthony Weatherhill; we kindly acknowledge the landowners on Fugloy for permission to work on their property. In Greenland, Adrian Aebischer, Vadim Heuacker, Eric Buchel, Brigitte Sabard and Vladimir Gilg helped at Hochstetter, and Anita Lang, Mark Nitze and Felix Norman at Karupelv Valley. In Erkuta, help was provided by Natalia Sokolova, Ivan Fufachev and Stijn Hofhuis. In Finland, Kari Mäntylä and Jukka Nuotio assisted during fieldwork. Support in the field in Scotland was provided by Helen and David Aiton on Rousay and the Fair Isle Bird Observatory. James Fox (Migrate Technology Ltd) and Glen Fowler (Biotrack Ltd) kindly retrieved data from loggers that failed to download.


RvB (Slettnes) by the Netherlands Organisation for Scientific Research (project 866.13.005), BM and SAH and (Svalbard & Brensholmen) by the Fram Center flagship “Climate Change in Fjord and Coast” (Grant 2019147470 1152018), the County Governor of Troms and the County Governor of Finnmark, KS and SH (Faroe Islands) by the Faroese Research Council and the National Museum of the Faroe Islands, OG and Loïc Bollache (Hochstetter) by the French Polar Institute-IPEV (Grant Interactions-1036) and the Agence Nationale de la Recherche (programs ILETOP ANR-16-CE34-0005 and PACS ANR-21-CE02-0024), DE (Erkuta) by the terrestrial flagship of the Fram Center (project Yamal EcoSystem 362259), LH, SH and JC (Scotland) by a donor programme managed by David Agombar, EM, KN and MS (Finland) by the Kone Foundation (application 28-1235), the Finnish Cultural Foundation (Grant to EM) and the ERC Grant nr 669609 (EM), GTH, SRV and YK (Iceland) by the University of Iceland Research Fund (Grant to GTH) and by the Southwest Iceland Nature Research Centre.

Author information

Authors and Affiliations



Conceptualization: RSAvB, BM, HS, KRSS and IT; methodology, formal analysis and visualization: RSAvB; fieldwork: RSAvB; BM, HS, SAH, KRSS, SH, EM, KN, MS, DE, SRV, GTH, YK, SH, JC, OG, JL, BS, AS, IT writing original draft: RSAvB with input from BM, HS, IT, KRSS, EMH and RP; writing and editing: RSAvB, BM, HS, IT, KRSS, EMH, JC and RP; supervision: BM and IT; funding acquisition: RSAvB, BM, EMH, JC, EM and IT.

Corresponding authors

Correspondence to Rob S. A. van Bemmelen or Ingrid Tulp.

Ethics declarations

Ethics approval and consent to participate

This study was carried out in accordance with guidelines in the use of wild birds for research [83]. Catching and tagging Arctic Skuas was approved for the following countries by the appropriate licensing authority: East Greenland (Government of Greenland, #C11-31, C12-28, C13-4(29), C13-11, C14-4(23), C14-16, C15-4(10), C15-14, C16-4(15), C16-20, C17-3(28), C17-23), Svalbard/Norway (Governor of Svalbard and Norwegian Food Safety Authority, FOTS ID 2086, 3817, 6329, 8538, 15726); Iceland (Icelandic Bird Ringing Scheme); Faroe Islands (Copenhagen Bird Ringing Centre and The National Museum of the Faroe Islands); Scotland (Special Methods Technical Panel of the British Trust for Ornithology Ringing Committee); Finland (Centre for Economic Development, Transport and the Environment, VARELY/418/07.01/2012) and Russia (Federal Law No. 52-FZ of April 24, 1995, Article 44).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1.

Geolocation data processing, figure S1 (summary of timing data), table S1 (posterior probailities) and figure S2 (slope estimates of the duration of period).

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

van Bemmelen, R.S.A., Moe, B., Schekkerman, H. et al. Synchronous timing of return to breeding sites in a long-distance migratory seabird with ocean-scale variation in migration schedules. Mov Ecol 12, 22 (2024).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: