Contrasting changes in space use induced by climate change in two Arctic marine mammal species

1. Background

Climate change is having serious consequences for global biodiversity [1]. Long-lived, high trophic level species are experiencing direct and indirect impacts of climate change, with the rapid pace of change rendering genetic adaptation unfeasible [2]. Distributional changes and various expressions of behavioural and dietary plasticity will likely be the first observable responses within ecosystems [2,3]. However, time series of sufficient length to assess these changes are rare in the Arctic [4,5].

The Arctic is the bellwether of climate change. Air temperatures are increasing three times faster than the global average, sea-ice extents are declining and glaciers are retreating [6]. Climate change impacts on Arctic species will likely have far-reaching impacts across ecosystems [4,7]. As long-lived species that are dependent on sea ice, and in some regions glacier fronts, Arctic marine mammals are expected to be negatively affected by climate change [7,8]. Owing to different ecological relationships with sea ice (or glacier fronts) and varying degrees of behavioural plasticity, species responses to climate change will likely vary [7].

Ringed seals (RS; Pusa hispida) and white whales (WW; Delphinapterus leucas) are Arctic marine mammals with circumpolar distributions [7]. Most populations of both species are found in areas containing sea ice throughout the year and both species forage predominantly on ice-associated prey [7]. Tidal glacier fronts are important areas for both species in some regions for foraging [8]. Both species will likely be impacted directly and indirectly (i.e. through changes in their prey base) by sea-ice reductions and glacier retraction.

RS and WW live year-round in waters surrounding Svalbard, Norway (74–81° N, 10–35° E). More than half of Svalbard's landmass is covered by glaciers and 60% of the glaciated area terminates in the sea ([8], figure 1). This archipelago has variable oceanographic regimes with the West Spitsbergen Current (WSC) transporting warm, Atlantic water northwards along the continental shelf-break in the west while eastern Svalbard is primarily influenced by Arctic water, which is transported around the southern tip of Svalbard and then northward along the west coast by the East Spitsbergen Current (ESC). Water mass exchange occurs across the polar front that forms between the WSC and ESC, resulting in intrusions of Atlantic water into west coast fjords and Storfjorden (east; figure 1) [9,10]. The magnitude of Atlantic water intrusions varies intra- and inter-annually [10].

• Download figure
• Open in new tab
• Download PowerPoint

In 2006, the sea-ice regime in Svalbard unexpectedly collapsed with the altered sea-ice conditions persisting to the present day. The land-fast sea-ice extent declined sharply, especially along the west coast [11]. This is partly due to the increased temperature of the WSC and more frequent penetration of the WSC across the polar front [10,11]. Svalbard and the northern Barents Sea region have had the greatest decrease in the seasonal duration of sea-ice cover in the Arctic [5]. The number of tidal glacier fronts in Svalbard is also decreasing [8].

Biotelemetry data from RS and WW were collected between 1995 and 2003 to study their basic ecology. The unexpected change in environmental conditions in 2006 presented the opportunity for a natural experiment. Repeat sampling after 2006 created unique biotelemetry datasets spanning two decades that were used herein to investigate how the large environmental changes in Svalbard have impacted the space-use patterns of these two ice-affiliated species during summer and autumn. These seasons are important foraging periods for both species and are times when the fjords are equally accessible to both species.

2. Material and methods

RS (28 in 1996–2003, 28 in 2010–2016) and WW (18 in 1995–2001, 16 in 2013–2016) were equipped with biotelemetry devices in Svalbard waters, providing animal movement data (table 1; electronic supplementary material, figure S1 and tables S1 and S2) [12]. Generalized additive mixed-effect models (GAMMs—binomial family and logistic link) were used to investigate how the proportion of time spent within 5 km of tidal glacier fronts (distance ≤ 5 km = 1, distance > 5 km = 0) changed between these two periods. Linear models were used to assess if glacier front use was associated with calving length or water depth. See electronic supplementary material for further details.

3. Results and discussion

Two decades ago, RS and WW spent approximately half of their time affiliated with glacier fronts (figure 2) and had diets dominated by polar cod (Boreogadus saida) during the summer and autumn [13–16]. However, RS and WW display contrasting responses to the environmental changes that have occurred in Svalbard waters, with RS now spending significantly higher proportions of time near tidal glacier fronts, while WW spend significantly less time near tidal glacier fronts (figure 2; electronic supplementary material, figure S2).

• Download figure
• Open in new tab
• Download PowerPoint

Negribreen was the glacier most visited by WW (both periods) and RS (first period) (figure 1; electronic supplementary material, tables S3 and S4). This glacier is large and has a long calving front that occurs in deep water. RS also heavily used Sonklarbreen (first period) and Kongsbreen (second period), which have similar characteristics to Negribreen (figure 1; electronic supplementary material, table S3). Time spent in front of other glaciers was relatively low and relative use of them was not explained by their characteristics for RS, though for WW frontal length remained important (electronic supplementary material, tables S5 and S6). Differences in tagging locations in the two study years (electronic supplementary material, figure S1 and tables S1 and S2) are unlikely to have impacted our results because WW move across much of the archipelago constantly [16] and RS results were not dependent on tagging location in the analyses herein.

Concomitant with the physical changes (increased Atlantic water intrusion and decreased sea ice) that have occurred over the last decade in Svalbard waters, large ecosystem changes have taken place, including a general ‘borealization’ of the fish community. Atlantic species are increasingly common and the ranges of Arctic and subarctic species are shifting northward [17,18]. Diets of some seabirds and marine mammals in the Svalbard area have changed to include more Atlantic and less Arctic prey [19,20]. However, Arctic and subarctic zooplankton, which are the main prey of polar cod, still dominate the innermost parts of glacial fjords [21] and polar cod are still abundant in these areas [22]. Calved glacier ice pieces also provide haul-out platforms for RS. Tidal glacier fronts appear to be serving as Arctic ‘refugia’ for RS, explaining why this species has increased the amount of time spent near glaciers, resulting in smaller home ranges following the sea-ice collapse (figure 2; electronic supplementary material, figure S2). Foraging effort by RS has also increased following the sea-ice collapse [15].

In contrast with RS, WW are not retracting into Arctic glacial refugia. They had larger home ranges and spent less time near glacier fronts and more time in the centre of fjords (figure 2; electronic supplementary material, figure S2) in 2013–2016 compared with 1995–2001 [16]. It is likely that they have shifted to foraging on Atlantic prey such as capelin (Mallotus villosus) and herring (Clupea harengus), similar to the situation in the Canadian Arctic [23]. WW have been observed milling in the centre of fjords in recent years, which was never seen previously in Svalbard waters (K.M.K. & C.L. 2013–2016, unpublished data). WW tend to be dietary generalists, in contrast with RS, which are more commonly individual specialists [24]. Although competition between these two species cannot be ruled out, a difference in dietary plasticity between them is likely the primary factor influencing their contrasting responses to a shared environmental change.

4. Conclusion

The different changes observed in the space-use patterns of RS and WW in Svalbard waters, using unique long-term biotelemetry datasets, highlight that ecosystem changes are affecting top trophic level predators differently. The flexible response shown by WW improves their chances of adapting to warming conditions, while RS's retraction into Arctic refugia, which are declining in number, with an ongoing dependence on prey that are also in decline, reflects limited adaptability and resilience. Plasticity in foraging and other responses to habitat change will be important in successfully adjusting to the ongoing environmental changes driven by global warming. Species and subpopulations that are not able to make such changes are almost certain to decline, perhaps to extinction where refugial areas become too limiting for species survival.

Ethics

Animal handling protocols were approved by the Norwegian Animal Research Authority and the Governor of Svalbard (RIS numbers: 2014/00067-9, 2014/00067-14, 16/01341-4, 16/01621-3).

Data accessibility

Data are available at the Norwegian Polar Data Centre (doi:10.21334/npolar.2019.e1cd54e1).

Authors' contributions

C.D.H., J.V.-G., R.A.I., C.L. and K.M.K. conducted fieldwork. J.K. provided glacier data. C.D.H. and J.V.-G. analysed the data. C.D.H., J.V.-G., C.L. and K.M.K. interpreted the results. All authors wrote the manuscript, approved the final version and agree to be accountable for the manuscript contents.

Competing interests

We have no competing interests.

Funding

Funding was provided by the Norwegian Research Council (ICE-WHALES, TIGRIF, GLAERE), NPIs ICE Centre and Fram Centre's Fjord and Coast Flagship.