Long-distance movements and associated diving behaviour of ringed seals (Pusa hispida) in the eastern Canadian Arctic

Quantifying animal movement is essential to understand the dynamic structure of ecosystems (Nathan et al. 2008; Wilmers et al. 2015). Marine environments consist of a heterogeneous distribution of resources shaped by spatiotemporal variation in oceanographic conditions that ultimately influence the abundance, movement, and behaviour of species. As animals move, they facilitate the transport of nutrients within and across ecosystems (Doughty et al. 2016). For example, anadromous fishes transport marine nutrients to riverine and terrestrial systems (Darimont et al. 2003; Hanson et al. 2010), whales transport nutrients from deep ocean basins to surface waters (Roman and McCarthy 2010), and reef sharks transfer resources from pelagic to reef systems (McCauley et al. 2012). These connections, however, are particularly difficult to observe in the marine environment due to the great depth, remoteness, expanse, and opacity of ocean ecosystems. Satellite telemetry provides a key tool to overcome these difficulties in understanding basic natural history, allowing monitoring of individuals and indirect observation of animal movement and behavior in relation to environmental conditions across marine environments (Cooke et al. 2004; Hussey et al. 2015).

Given their often large spatial ranges, marine mammals integrate broad biophysical processes across systems into interpretable metrics; therefore, some species can act as sentinels of environmental change in marine ecosystems (Moore 2008; Hazen et al. 2019; Moore and Kuletz 2019). Arctic marine mammals have adapted to extreme intra-annual fluctuations in environmental conditions, such as seasonal ice cover and daylight — key features that affect their movement ecology (Moore 2008). For example, narwhal (Monodon monoceros Linnaeus, 1758) exhibit annual migrations between the winter pack ice in northern Baffin Bay and the open-water fjords and bays between Lancaster Sound and Peel Sound (Heide-Jørgensen et al. 2003). As climate change has modified oceanographic processes in the Arctic at an accelerating pace, including dramatically reduced summer sea ice, increased freshwater inputs, surface-layer warming, and altered sedimentation (Wassmann et al. 2011; IPCC 2014), significant trophodynamic shifts have occurred, including alterations to zooplankton communities and changing interactions due to northward range expansions of subarctic species (Blanchard 2015; Kortsch et al. 2015; Pecl et al. 2017; Yurkowski et al. 2018). It is, therefore, important to establish reference information on the movements and dive behaviour of Arctic marine mammals relative to the environment to determine key areas of habitat use, nutrient transfer, and connectivity between populations and systems, as well as the potential for overlap with predators, competitors, conspecifics, and human influences.

The pagophilic ringed seal (Pusa hispida (Schreber, 1775)) is widely distributed, highly abundant, and occupies a diversity of habitats, ranging from shallow coastal zones to deeper ocean basins during the open water period and land-fast ice to moving pack-ice during the ice-covered period (Smith 1987). As a result of their dependence on sea ice and their sensitivity to changes in snow and ice conditions (McLaren 1958; Smith 1987; Chambellant et al. 2012; Iacozza and Ferguson 2014; Reimer et al. 2019), ringed seals act as an indicator species of environmental change in Arctic ecosystems (Laidre et al. 2008). After the energetically expensive spring breeding and moulting periods, when energy reserves are expended and blubber is thinnest (Young and Ferguson 2013), ringed seals move among prey patches during the open-water season in summer and fall (Yurkowski et al. 2016), foraging intensively to rebuild energy stores (Smith 1987; Freitas et al. 2008). Throughout the open-water period, both adult and subadult seals make extensive movements into open water, presumably to forage (e.g., Teilmann et al. 1999, Gjertz et al. 2000, Kelly et al. 2010, Crawford et al. 2012, Harwood et al. 2012, Hamilton et al. 2015). During the winter, their movements are more variable, with seals establishing territories and maintaining breathing holes in the shore-fast ice, moving throughout the pack ice, or moving into open water (McLaren 1958; Kelly et al. 2010; Crawford et al. 2012; Harwood et al. 2015).

Here, we describe the movements and dive behaviour of ringed seals in the less-studied eastern Canadian Arctic. With the aid of Inuit partners, ringed seals were equipped with satellite telemetry tags during summer in the Tallurutiup Imanga National Marine Conservation Area (Lancaster Sound), Nunavut, Canada and were tracked to southeast Baffin Island. We fit a two-state switching Bayesian state-space model (BSSM) to satellite-telemetry-derived location data for 12 ringed seals to examine their movements and identify periods of area-restricted movement in addition to investigating diving parameters. Specifically, we compared individual movements, movement paths, and behavioral states, as well as mean daily dive depth and duration and mean daily surface duration. Results will improve our understanding of eastern Canadian Arctic ringed seal movement ecology and connectivity within and between High and Low Arctic areas and act as a foundation for future comparative research.

Twelve ringed seals were captured and tagged, of which 10 were subadults (eight male and two female) and two were adults (one male and one female). Deployment durations for tagged ringed seals ranged from 19 to 248 days and averaged 96.3 ± 21.9 days (Table 1). Excluding days where no transmissions were received, total tracked days ranged from 19 to 205 days and averaged 81.6 ± 17.1 days (Table 1). All ringed seals were tagged in August or September, with the two longest-transmitting tags providing data until May of the following year. Total distances travelled by all ringed seals throughout the tagging period averaged 2171 ± 481 km (range: 166 – 4872 km; Table 2).

Transmissions for the short-duration individuals (one adult and five subadults) ceased in October (16 October ± 3.5 d; range: 4 October – 26 October) while in Parry Channel and nearby areas, having moved relatively short distances prior to transmission termination (Table 1, 2, Fig. 1). These individuals travelled an average distance of 702 ± 465 km (range: 166 – 1574 km; Table 2). The dates on which long-ranging seals initiated their travel southward ranged from 19 September to 25 November (14 October ± 9.6 d; Table 2). Local ice concentration at the start of these movements was highly variable, ranging from 0% to 75+% (29.2% ± 14.5%; Fig. 2). Long-ranging ringed seals travelled, on average, 1209 ± 252 km (range: 649 – 2021 km; Table 2) before moving southward. During their southward movement, distance travelled averaged 2023 ± 175 km (range: 1637 – 2577 km; Table 2), whereas, after reaching southeastern Baffin Island, seals moved much less (614 ± 339 km; range: 143 – 1587 km; Table 2).

Ringed seals varied markedly in their movement paths, with three of the six long-ranging individuals moving south along the coast of Baffin Island (107840, 159284, 159289), whereas the other three moved into the open water of Baffin Bay (107832, 107836, 107839), one of which reached the coast of Disko Island, Greenland (107839; 69.8319° N, 53.9962° W; Fig. 1). Long-ranging ringed seals moved along a similar eastward path into central Baffin Bay but diverged as they moved south (Fig. 1), with most having different final destinations. One subadult (est. age 0; 107832) reached Angijak Island (65.6729° N, 62.3339° W) on 25 November 2012 (36 d of long-distance travel) and remained there for ∼4 months (mean local sea-ice concentration: 90.4% ± 4.5%), with tag transmissions terminating on 27 March 2013. For two individuals, 107836 (subadult, est. age 0) and 159284 (subadult, est. age 3), tags stopped transmitting near Brevoort Island (63.4626° N, 64.2982° W) on 1 December 2013 (58 d of long-distance travel) and 31 December 2017 (36 d of long-distance travel), respectively, with 107836 having inhabited this area for ∼2 weeks (mean local sea-ice concentration: 7.3% ± 6.7%). Seal 107839’s (subadult, est. age 1) tag stopped transmitting east of Resolution Island (61.5087° N, 64.9475° W) on 9 November 2013 (44 d of long-distance travel) while still in transit. Two long-ranging ringed seals reached and spent a significant amount of time overwintering in Cumberland Sound: one (107840; adult, est. age 7) entered Cumberland Sound on 2 January 2014 2014 (78 d of long-distance travel) and inhabited the moving ice areas of Cumberland Sound for ∼3 months (mean local sea-ice concentration: 88.1% ± 2.5%) before moving northward on 8 April 2014, with transmissions ceasing on 6 May 2014 near Qikiqtarjuaq (67.5556° N, 64.0257° W); the second individual (159289; subadult, est. age 4–5) entered Cumberland Sound on 13 December 2018 (62 d of long-distance travel) and stayed near its mouth, around Abraham Bay (65.1394° N, 64.3613° W), for ∼5.5 months (mean local sea-ice concentration: 83.9% ± 2.2%) before the tag ceased transmitting on 25 May 2019.

Distance travelled was not biased by tag duration, as there was no significant difference in distance travelled between short-duration ringed seals (702 ± 214 km) and long-ranging ringed seals prior to their travels southward (1209 ± 252 km; t = 1.5, df = 9.7, p = 0.16). Daily movement rate, however, differed between short-duration ringed seals (16.6 ± 1.3 km/d) and long-ranging ringed seals before their travels southward (23.7 ± 1.1 km/d; t = 7.2, df = 430.9, p < 0.0001). Whereas the three long-ranging ringed seals that moved into central Baffin Bay and then southward (107832, 107836, 107839) made relatively fast, directed movements, the other three that travelled southward along Baffin Island (107840, 159284, 159289) made more gradual and meandering movements, with multiple periods of area-restricted movement both within and outside of fjords.

Movement rates for long-ranging seals varied significantly by travel period (F_2,847.0 = 159.3, p < 0.001), with individuals moving fastest during their travel southward (39.8 ± 1.3 km/d) compared with before and after (23.7 ± 1.1 km/d and 9.7 ± 0.7 km/d, respectively). Post-hoc analysis identified significant differences in movement rates between all pairwise time-period comparisons (p < 0.0001). Time spent exhibiting area-restricted movements also differed between travel periods for long-ranging ringed seals (F_2,8.6 = 10.3, p < 0.01). Post-hoc analysis revealed that this difference was significant between the during (31.1% ± 10.4%) and after (88.5% ± 6.2%) periods (p < 0.01). In addition, although coastal-travelling long-ranging seals appeared to spend more time exhibiting area-restricted movement during long-distance movements (49.1% ± 7.7%) than those that travelled through central Baffin Bay (13.2% ± 12.5%), this difference was not statistically significant (t = –2.5, df = 3.3, p = 0.08), likely as a result of the low statistical power resulting from a low sample size.

In total 78,751 dives were recorded by ringed seals during the study period. Individual seals made, on average, a minimum of 6560 ± 1380 dives throughout their respective tagging period (range: 2257 – 17,115; Table 2). Dive depth was plotted against dive duration, and 28 unrealistic dive outliers (e.g., 782 m in 8 s, 750 m in 240 s) were identified and removed from the data set. From the remaining data (n = 78,723), the maximum dive depth was 558 m in 1020 s (17 min) by an adult male ringed seal (Table 2). Maximum depth reached by each seal averaged 294 m ± 38 m. Overall, 70.5% of all recorded dives were shallower than 50 m depth, 82.9% were shallower than 100 m, 93.6% were shallower than 200 m, and 98.7% were shallower than 300 m.

Similar to distance travelled, mean daily surface duration did not appear to be biased by tag duration, with short-duration seals spending similar amounts of time at the surface between dives (77.0 ± 1.8 s) compared with long-ranging seals before their travel southward (83.4 ± 4.1 s; t = 1.1, df = 331.2, p = 0.28, Fig. 3). Differences in correlated variables of mean daily dive depth and mean daily dive duration (Kendall rank correlation, z = 35.8, τ = 0.70, p < 0.0001) were observed between short-duration seals (23.2 ± 1.1 m) and long-ranging seals prior to their travel southward (68.3 ± 3.2 m; t = 13.4, df = 483.9, p < 0.0001; Fig. 3). This difference was primarily driven by the three coastal-travelling long-ranging seals (107840, 159284, 159289; Fig. 3) that dived deeper on average prior to their travel southward (100.0 ± 4.1 m) than did long-ranging seals that travelled through central Baffin Bay (25.4 ± 1.1 m; t = 19.8, df = 286.5, p < 0.0001). Similarly, mean daily dive duration was greater for long-ranging seals prior to their travel southward (228.8 ± 7.2 s) than for seals tagged for a shorter duration (137.7 ± 4.4 s; t = 10.1, df = 472.5, p < 0.0001; Fig. 3). This, again, was driven by the coastal-travelling long-ranging seals, which made longer dives during this movement period (307.4 ± 8.1 s) than those that travelled through central Baffin Bay (122.5 ± 2.9 s; t = 20.3, df = 291.1, p < 0.0001).

Long-ranging ringed seals differed in mean daily dive depth throughout their travel southward (F_2,957.8 = 140.1, p < 0.0001), with mean daily dive depths of 68.3 ± 3.2 m, 78.5 ± 3.2 m, and 46.9 ± 2.4 m before, during, and after, respectively (Fig. 3). Post-hoc analysis showed significant differences between the before and after periods (p < 0.0001) and between the during and after (p < 0.0001) periods. There were also significant differences in mean daily dive duration for long-ranging seals throughout their travel southward (F_{2, 957.6} = 43.3, p < 0.0001), with mean daily dive durations of 228.8 ± 7.2 s, 278.8 ± 8.9 s, and 221.6 ± 7.1 s before, during, and after, respectively (Fig. 3). Post-hoc analysis revealed that mean daily dive duration after southward travel was significantly lower than mean daily dive duration before (p < 0.0001) and during (p < 0.0001). Mean daily surface duration varied by travel period as well (F_2,959.6 = 109.4, p < 0.0001), with seals spending 77.0 ± 1.8 s, 77.8 ± 1.9 s, and 139.2 ± 4.1 s at the surface between dives before, during, and after their travel southward, respectively (Fig. 3). This was driven by mean daily surface duration in the after period, as values in this period were significantly greater than both the before (p < 0.0001) and during (p < 0.0001) periods.

Understanding how animals move through time and space is an important requirement of their successful management and conservation. In the present study, we examined the movements and dive behaviour of 12 ringed seals tagged in the Canadian High Arctic near Resolute Bay and Tremblay Sound, Nunavut, to improve understanding of travel pathways, scales of movements, and the level of inter- and intra-individual variation in movement dynamics. Half of the tags (6/12) transmitted for only 1–2 months, at which point the seals were still in High Arctic areas of Parry Channel and Barrow Strait. Tags on the remaining individuals (6/12) recorded data for much longer and captured long-distance movements to southern Baffin Island, with different final locations recorded. Notably, three of these long-ranging ringed seals travelled into central Baffin Bay and southward, whereas the other three travelled southward along the coast of Baffin Island. The three seals that travelled through central Baffin Bay made shallower and shorter dives than those that travelled southward along the Baffin Island coast. Distances travelled and the dates of their travel southward relative to sea ice concentration were also variable. Possible reasons for these differences in movement are numerous and likely include a combination of both abiotic and biotic drivers. Results indicate that, although many ringed seals may exhibit long-distance movements southward from the High Arctic, individuals vary in their movement strategies and do not appear to follow a singular route, as was observed in the western Canadian Arctic (Harwood et al. 2012), or reach a similar southern destination.

Long-distance movements by ringed seals have been documented in other Arctic regions, including the Beaufort and Chukchi Seas (subadults: 2793 km, adults: ∼1500 km, Harwood et al. 2012; adults: ∼1700 km, Kelly et al. 2010), Northern Baffin Bay (subadults: ∼2500 km, Teilmann et al. 1999), the Bering and Chukchi Seas (subadults: ∼1000 km, Crawford et al. 2012), and Svalbard (subadult: 5393 km, Hamilton et al. 2015; adult: ∼1500 km, Gjertz et al. 2000), although, in most instances, recorded movements are typically <500 km (Heide-Jørgensen et al. 1992; Gjertz et al. 2000; Freitas et al. 2008). Ringed seals tagged at higher Arctic latitudes travelled longer distances and have higher movement rates than those tagged at lower Arctic latitudes (Yurkowski et al. 2016; Ferguson et al. 2019). In the present study, ringed seals made one-way long-distance movements of approximately 1769–3518 km, some of the longest recorded distances observed for both subadults and adults. Most notably, an adult male ringed seal travelled from Resolute Bay to Cumberland Sound, which, excluding foraging stops in bays and fjords along Baffin Island, represents a distance of approximately 2200 km. To our knowledge, this is one of the longest recorded one-way movements by an adult male ringed seal. In contrast to adult ringed seals, subadults commonly travel long distances (Teilmann et al. 1999; Crawford et al. 2012; Harwood et al. 2012; Hamilton et al. 2015), as they do not need to establish territories for mating and are competitively excluded from land-fast ice areas by adult seals during winter (Smith et al. 1991). Although the greatest distance travelled in the present study was by an adult seal, low sample sizes did not allow for statistical comparison between adult and subadult age classes.

The adult male that travelled the greatest distance also undertook the deepest dive, reaching a depth of 558 m. Dives of this depth are relatively uncommon for this species and are likely near the limit of ringed seals’ diving capabilities (Hammill 2018). Although a few studies have documented maximal depths similar to this (e.g., 500+ m, Born et al. 2004; 542 m, Harwood et al. 2015), the vast majority of recorded dives are below 100 m, with occasional dives deeper than 300 m (Gjertz et al. 2000; Born et al. 2004; Benoit et al. 2010; Harwood et al. 2012; Hamilton et al. 2016; Crawford et al. 2019). This trend was confirmed by the data in this study. As an adult male, its large body size likely allowed it to reach these depths, as larger seals are typically able to dive deeper due to increased physiological capabilities compared with smaller seals (Lydersen et al. 1992; Kelly and Wartzok 1996).

Ringed seals that travelled southward took different routes, displaying either a coastal movement behaviour along Baffin Island or an offshore movement behaviour through central Baffin Bay. The reasons for these two distinct movement behaviours are unclear, but possible explanations include surface currents, competition, predator avoidance, ice conditions, and foraging strategies. Based on the present data, however, there is insufficient information to further test these hypotheses and whether individual or interacting factors drive the observed movements. Sea surface current directions in Baffin Bay do not overlap with the directionality of ringed seal movements to a large extent, particularly for those that travelled through open water (Fig. 1; Aksenov et al. 2010; Hamilton and Wu 2013), and few studies have documented current-following in pinnipeds (e.g., northern fur seal, Callorhinus ursinus (Linnaeus, 1758); Ream et al. 2005). There is also insufficient evidence to support competitive exclusion of subadults by territorial adult seals (Krafft et al. 2007) driving open-water movements; ringed seals that travelled offshore did not appear to make initial attempts to travel along the coast. Crawford et al. (2012) similarly found that territoriality by adult ringed seals did not adequately explain long-distance movements by subadults in the Bering and Chukchi Seas. Based on the time of year and associated ice conditions when individuals initiated their travel southward, it seems unlikely that predator avoidance could explain these open-water paths. Polar bears (Ursus maritimus Phipps, 1774), the main predators of ringed seals, hunt on the sea ice most intensely from May to July (Hammill and Smith 1991; Smith and Lydersen 1991; Messier et al. 1992; Galicia et al. 2015). Ringed seals began their movements before ice had formed in the area, suggesting that predation risk was low during their travel south. Furthermore, ice conditions at the initiation of their travel southward varied considerably between individuals, suggesting that ringed seals do not conform to a single strategy of ice association and instead exhibit individual variation in their movement and behaviour. We accept, however, that the low number of tags that transmitted long enough to capture long-distance movements may be too low to identify a pattern. To our knowledge, no other study has assessed sea-ice concentration as a trigger for long-distance movements in ringed seals, although it has been examined in belugas (Delphinapterus leucas (Pallas, 1776)) (Bailleul et al. 2012; Hauser et al. 2017), and further study is needed. Increased foraging opportunities could explain this movement, and, although we do not have the data to determine this with certainty, patterns of dive behaviour and behavioural state results support this as a potential explanatory factor, both for routes taken and final destinations.

In summer and early fall, ringed seals spent more than half of their time exhibiting area-restricted movement behavior, with intermediate daily movement rates, behavioural states, and diving parameters compared with other time periods. At this time, seals were likely foraging intensively to rebuild fat stores following the spring fasting season, which occurs primarily from April to June (McLaren 1958; Welch et al. 1992; Quakenbush et al. 2011; Young and Ferguson 2013).

During fall and early winter, daily movement rates of long-ranging seals were at their highest, and, correspondingly, time in area-restricted movement behaviour was at its lowest. For open-water seals, dives were shorter and shallower than those of coastal-travelling seals. Ringed seals are known to switch from deeper, benthic feeding to pelagic feeding depending on their local environment and prey availability (McLaren 1958). The observed patterns in diving suggest that open-water seals might be foraging on pelagic prey, such as amphipods, requiring shallower dives (McLaren 1958; Auel and Werner 2003; Ogloff et al. 2019). In support of this hypothesis, one of the main amphipod prey species (Siegstad et al. 1998; Wathne et al. 2000; Holst et al. 2001; Labansen et al. 2011; Ogloff et al. 2019) of ringed seals, Themisto libellula (Lichtenstein, 1822), is common in the upper 25 m of the pelagic environment (Wathne et al. 2000; Auel and Werner 2003; Havermans et al. 2019), which overlaps with the mean daily dive depth (25.4 m) of ringed seals that travelled through central Baffin Bay. In contrast, the coastal-travelling seals, which more often occupied shallower water over the shelf and in bays and fjords along Baffin Island, were able to dive to the sea floor and access benthic prey fish, such as Boreogadus saida (Lepechin, 1774) and sculpin (McLaren 1958; Wathne et al. 2000; Labansen et al. 2011; Ogloff et al. 2019). Similar findings were documented for ringed seals in the North Water Polynya (Born et al. 2004). Importantly, however, vertical overlap with prey does not confirm with certainty that ringed seals forage extensively during this time, as ringed seals are known to make travel dives, defined by their directionality and relatively shallow depth (Kelly and Wartzok 1996; Simpkins et al. 2001). The short surface duration during long-distance movements, in addition to relatively shallow diving and seals being predominantly in a travelling behavioral state (∼69% of the time), may provide evidence for this type of movement.

During the winter and early spring, the four tags that were still transmitting recorded extended periods of residency of 2 weeks up to ∼5.5 months in an area. Surface duration and time exhibiting area-restricted movement behaviour were at their highest, whereas daily movement rate, distance travelled, mean daily dive depth, and mean daily dive duration were at their lowest. Given that these periods of residency occurred primarily in winter, with high sea-ice concentrations, it is possible that individuals were spending more time hauled out, maintaining breathing holes, or moving throughout the pack ice (McLaren 1958). Specifically, seal 107832, an immature individual, spent ∼4 months at Angijak Island before tag transmissions stopped in late March. Based on timing, this individual was likely foraging and building fat reserves in preparation for the spring moult, which takes place from mid-May to mid-July (McLaren 1958). Seal 107836 remained near Brevoort Island for ∼2 weeks in late November, perhaps foraging in the mostly-open water in preparation for winter (McLaren 1958). Two ringed seals (159289 and 107840) resided in Cumberland Sound for a large portion of the winter (∼5.5 and ∼3 months, respectively). A mature male, 107840, moved clockwise around Cumberland Sound over the winter before exiting and travelling northward and residing near Qikiqtarjuaq for a short time before tag transmissions stopped. As this ringed seal was mature, it is likely that this extensive residency in Cumberland Sound served to build energy reserves in preparation for spring breeding, which may have occurred while still in Cumberland Sound, and the following moult, which may have occurred near Qikiqtarjuaq. In contrast, the subadult, 159289, stayed in a small area near the mouth of Cumberland Sound for almost 6 months in the land-fast ice, most likely maintaining a breathing hole and foraging, before transmissions stopped in late May, perhaps during moulting (McLaren 1958). As noted previously, subadult ringed seals are often excluded from prime mating and feeding territory by larger, aggressive conspecifics as early as ice-formation in the fall, resulting in subadults being relegated to peripheral habitats with less-stable sea ice (McLaren 1958; Smith 1987; Smith and Lydersen 1991; Krafft et al. 2007; Kelly et al. 2010; Crawford et al. 2012). This could explain why subadults tended to occupy peripheral areas throughout the study area, whereas the adult male moved deep into Cumberland Sound. Both adult and subadult ringed seals can also occupy the pack-ice during the breeding season (Finley et al. 1983), so it is possible that seal 159289 was a young adult employing this strategy.

Long-distance southward movements by ringed seals in the High Arctic facilitate the transfer of nutrients and genes from the High Arctic, where the greatest fat deposition occurs during intensive foraging in the open-water season, to lower latitudes, where seals may breed or are more likely to be consumed by predators (Smith and Hammill 1981; Furgal et al. 1996; Young and Ferguson 2013; Pilfold et al. 2014; Galicia et al. 2015). Tags deployed on seals 107839 and 159284 stopped transmitting at Resolution Island and Brevoort Island, respectively, while seals were still in transit, so it is unclear whether they displayed any residency behaviour to these areas or elsewhere. Seal 107839, having reached the southernmost tip of Baffin Island while still in a transiting state when its tag stopped transmitting, may have been travelling into Hudson Strait, Ungava Bay, or the Labrador Sea. This suggests that mixing between High Arctic and Low Arctic ringed seals may occur; however, this individual was immature and more data for adult individuals would be required to confirm this.

It is not known whether these long-distance travels southward represent seasonal migrations, with seals returning northward and transferring nutrients from the Low Arctic to the High Arctic. Inter-annual site fidelity to breeding or foraging areas has been documented for ringed seals in other areas of the Arctic (Smith and Hammill 1981; Krafft et al. 2007; Freitas et al. 2008; Kelly et al. 2010; Harwood et al. 2015), suggesting that some seals do return to the same area across multiple years. It is also unknown how common this long-distance movement strategy is for High Arctic ringed seals and whether tagging site or year might influence observed patterns. Given that the behaviour of short-duration ringed seals differed little from the behaviour of long-ranging seals prior to their travel southward, and that all tags that transmitted throughout the fall and winter captured a long-distance movement, it is plausible that many of the ringed seals that summer around northern Baffin Island make similar long-distance travels. Increased sample sizes in future studies would be needed to quantify how common this movement strategy is for High Arctic ringed seals, as some individuals also overwinter in our High Arctic study area (Kelly et al. 2010; Young et al. 2019; Yurkowski et al. 2019), and would allow for comparison among age classes, High Arctic locations, and years. Given the scale of ringed seal movements, the proximity of Resolute Bay and Tremblay Sound (<500 km), and the overlap in ringed seals’ movement paths around northern Baffin Island during summer and early fall (prior to southward movements), we suspect that tagging location would not be a significant predictor of long-distance southward movements by ringed seals.

These preliminary data suggest flexibility and high variability in ringed seal movement behaviour among individuals, which may allow ringed seals to adjust to changing sea ice conditions during long-distance travels; however, the strong association between ringed seals and sea ice (McLaren 1958; Smith and Stirling 1975) for breeding may concurrently limit their resilience to these changes. Future studies in the High Arctic should aim to increase sample sizes and age and sex representation while incorporating environmental data, prey distribution, and physiology/energetics into analyses. This study provides an improved understanding of the natural history and movements of eastern Canadian Arctic ringed seals, the main prey of polar bears and an important subsistence species for harvesters in Nunavut. An important aspect of this study is the inclusion of local expertise and Indigenous knowledge by working directly with Indigenous partners to capture and tag seals. Understanding the general movement ecology of ringed seals will allow for more well-informed and effective management and conservation of ringed seals given impending ecological challenges, such as increased shipping traffic, oil exploration, development, and climate change.

We thank the communities of Resolute and Pond Inlet, as well as the Resolute Hunters and Trappers Association and the Mittimatalik Hunters and Trappers Organization, and especially their hunters, Peter, Jeff, Star, and Uluriak Amarualik, for assistance in the field. We also thank the Polar Continental Shelf Program for logistical support. All members of the Ecosystem Approach to Tremblay Sound (EAT) team, especially Robert Hodgson (Fisheries and Oceans Canada; DFO), Troy Neale (Ocean Wise), Melissa Galicia (York University), and Stephen Petersen (Assiniboine Park Zoo), are thanked for their assistance with field logistics and seal capture/tagging in Tremblay Sound. Thanks also to Kevin Scharffenberg for assistance analysing sea-ice data. This study was supported by funding from Natural Sciences and Engineering Research Council (NSERC)-Ocean Tracking Network, NSERC-Discovery, Fisheries and Oceans Canada, Government of Nunavut, Nunavut Wildlife Research Fund, World Wildlife Fund, Ocean Wise, and ArcticNet to ATF, SHF, NEH, and MM, as well as the University of Windsor, Ontario Graduate Scholarship, The W. Garfield Weston Foundation, NSERC, and Earth Rangers to DJY.