Introduction

In the frame of our long-term study of the quantitative distribution at sea of seabirds and marine mammals in polar marine ecosystems, the main aims are, on the one hand, to deepen the understanding of the mechanisms influencing their distribution (water masses, fronts, pack ice, ice edge and eddies) and on the other hand to try and detect spatial and temporal evolutions with special attention to possible consequences of global climatic changes, and of (krill) fisheries. As upper trophic levels, their distribution reflects prey abundance: zooplankton, krill, nekton, and small fish and so integrates the ecology and the biological production of the whole water column. They constitute thus the best approach to identify and localise areas of high biological production, and to detect temporal changes (Hunt 1990; Bost and Le Maho 1993; Furness and Camphuysen 1997; Joiris 2007; Joiris and Falck 2010).

This paper reports on data collected about the upper trophic levels in the south-western Atlantic sector, from South America to the South Shetland Islands, i.e. a poorly studied area in autumn.

Materials and methods

During the ANT XXVIII/4 expedition of icebreaking RV Polarstern from Punta Arenas to the South Shetland Islands and back, from March 14 to April 8, 2012, 333 transect counts were devoted to seabird and marine mammal observation. The transect counts were lasting half an hour, without width limitation from the bridge at 18 m above sea level on a continuous basis, light and visibility allowing (see description and discussion in Joiris 2007; Joiris and Falck 2010). The animals were detected with the naked eye, and observations confirmed and complemented with binoculars (10 × 42) when useful; photographic documents were also used, especially for species difficult to identify. Results are presented as basic unmodified data, i.e. numbers encountered per half-an-hour transect count. Density was also calculated on the basis of specific detection limits established by this team in function of size, colour and behaviour (jizz) (Joiris 2007; Joiris and Falck 2010) and mean ship’s speed: 10 knots in open water and 5 knots in ice covered areas.

Water temperature and salinity were continuously recorded on board Polarstern with a thermo-salinometer at subsurface sampling (keel −10 m). Ice cover was evaluated by us from the bridge and expressed as  % coverage within an approximate range of 500 m around the ship. A navigation echo-sounder was continuously running, mainly in order to follow water depth, but also showing echo’s in the water column, such as zooplankton, nekton and fish (Simrad Kongsberg EK60 with multifrequency at 70 or 120 kHz and an angle of 7°).

Results

The route followed by Polarstern is shown in Fig. 1, as well as the position and numbering of the transect counts, the positions of the slope and ice edge (Online Resource 1). During the 333 counts devoted to the upper trophic levels, 113,000 birds were encountered belonging to 40 species, without taking into account the strictly coastal ones. The most numerous species was southern fulmar Fulmarus glacialoides with 77,000 individuals, i.e. 68 % of the total; next came grey-headed albatross Thalassarche chrysostoma with 7,000 individuals; and Antarctic prion Pachyptila desolata with 6,400 individuals (Table 1). Similarly, fin whale Balaenoptera physalus formed the vast majority of the identified large cetaceans: 308 individuals out of 330, i.e. 94 % (Table 1). Concentrations of these species were highest within the South Shetland Islands area, especially off Elephant Island (Fig. 2). Mean numbers for this area are out of 99,700 seabirds in 205 counts: 76,800 southern fulmars (77 %)—of which 57,600 in one count west of Elephant Island, 6,940 grey-headed albatrosses (Table 1) and 308 fin whales, corresponding to 105 sightings. These numbers of fin whales are probably an underestimate, considering that the majority of unidentified large whales (n = 87, Table 1) were actually fin. This is why data from this area were analysed and discussed in some detail (Table 2). Density was calculated in this limited area for fin whale and the most abundant seabird species (N ≥ 100): mean density was 0.03/km2 for fin whale; a mean density of more than 9/km2 was found for southern fulmar, of 6.2/km2 for grey-headed albatross, and 17/km2 for Antarctic prion, one of the most abundant species in the southern seas.

Fig. 1
figure 1

Route of RV Polarstern during expedition ANT XXVIII/4, March/April 2012; position and numbering of the transect counts for seabirds and marine mammals; position of the shelf slope indicated by a dotted line (top); pack ice northern limit in Bransfield Strait indicated by a dotted line (bottom): situation on April 4, 2012; based on a “Polar View” map (University of Bremen, Germany: http://iup.physik.uni-bremen.de) (online Resource 1)

Full size image
Table 1 Seabirds and marine mammals encountered between Punta Arenas and the South Shetland Islands
Full size table
Fig. 2
figure 2figure 2

Counting results for the most abundant species, totals per transect count in classes of abundance, birds: southern fulmar F. glacialoides (a), grey-headed albatross T. chrysostoma (b), Antarctic prion P. desolata (c); cetaceans: fin whale B. physalus (circles) and unidentified large whale (triangles) (d)

Full size image
Table 2 Seabirds and fin whales encountered around the South Shetland Islands area
Full size table

Moreover, these very high values were concentrated in few counts in a limited area north-west of Elephant Island, reflecting a huge patchiness in distribution: 100 fin whales in count 127 (March 21, 61°S, 55.50°W), 67,570 fulmars in three neighbouring counts, i.e. 1 % of the total (142, 143, 145, of which 57,600 in count 143, March 23, 61°S, 56°W), 457 Antarctic prions in count 143 and 1,480 grey-headed albatrosses in count 144. A colloquial summary becomes “it all happened there”. Translated into density (N/km2), these maximal numbers encountered correspond to 2.2 fin whales, 1,400 southern fulmars and 270 grey-headed albatrosses.

High prey availability was reflected by visual observation of large subsurface swarms of krill Euphausia superba, extending down to 50–100 m deep above the slope of the “continental” shelf of Elephant Island as seen on the echo-sounder screen (Online Resource 2).

Some observations deserve comments: 3 bird species were recorded more than 1,000 km south or southwest of their known range, according to Shirihai (2007); 2 sooty albatrosses Phoebetria fusca on March 16 at 58°S, 60°W, a few soft-plumaged petrels Pterodroma mollis in the Drake Passage, on the way to Elephant Island on March 17. A common diving-petrel Pelecanoides urinatrix was found alive as a shipwrecked bird on March 25 at about 50 km west of Elephant Island and released. Three hourglass dolphins Lagenorhynchus cruciger correspond to the group present far from the coast in open water (Santora 2011; Dellabianca et al. 2012). Temporal evolution was noted during the expedition. Southern fulmar, for instance, was close to absent from the south route and present in significant numbers during the return north transect (Fig. 2a).

This study area is covering the south-western Atlantic at the boundary of the Antarctic Circumpolar Current ACC and the Polar Front PF, i.e. at the limit between low and high Antarctic, as reflected by water temperature and salinity data (Fig. 3) and corresponding to the shelf slope. In the north and east of the South Shetland Islands, the bathymetry (slope of the continental shelf) might cause food-rich upwelling. On the other hand, the vicinity of a deep canyon strongly influences krill and top predator abundance (Santora and Reiss 2011). The biological production is high, as reflected for instance by high oxygen concentration (Orsi et al. 1995).

Fig. 3
figure 3

Hydrological data collected on board Polarstern: salinity (a) and water temperature (°C) (b)

Full size image

Discussion

Water masses and fronts, pack ice and ice edge are the main hydrological factors influencing the distribution of the upper trophic levels—seabirds and marine mammals—in the ocean. This is known for decades (Wynne-Edwards 1935; Joiris 1979; Pocklington 1979; Kinder et al. 1983) and was confirmed more recently by many studies (Elphick and Hunt 1993; Joiris and Falck 2010). Boundaries such as ice edge and slope are known to play a very important role, especially in Antarctic ecosystems (Ainley et al. 1998, 2007, 2012; Ribic et al. 1991, 2011; Jacobs 1991; Joiris 1991; van Franeker 1992; Bost et al. 2009; Branch 2011). Seabird and baleen whale distribution is depending on prey abundance, krill, nekton and fish (Heinemann et al. 1989; Ryan and Cooper 1989; Reid et al. 2000), mainly krill in this case. Few publications, however, simultaneously include birds and cetaceans (this team; Friedlaender 2006; Friedlaender et al. 2008; Santora et al. 2009a, 2010; Ballard et al. 2011). The South Shetland Islands, especially Elephant Island, are known for their high krill concentrations attracting seabird predators (Hunt et al. 1990; Hewitt and Demer 1993; Agnew 1997). They also support an important krill fishery, the main one together with South Georgia (Croxall and Nicol 2004) even if it strongly decreased during the last decades (Everson and Goss 1991; Jones and Ramm 2004). Important variations in krill density were recorded, directly influencing seabird foraging (Santora et al. 2009b). An exceptionally large “super krill swarm” was detected in 1981 (Tarling et al. 2009), then not any longer for decades (Lawson et al. 2008) and again detected in 2009 in the western Antarctic Peninsula in association with very high concentration of humpback whales (up to 5/km2) (Nowacek et al. 2011). The data presented here, even if semi-quantitative for krill (Online Resource 1), seem to reflect the presence of such a huge swarm again.

The high numbers of fin whales, southern fulmars and grey-headed albatrosses represent an exceptionally large aggregation and were concentrated in a very small subarea at the front, even if not strictly in the same counts, reflecting a huge patchiness. Most individuals of the three species were actively subsurface feeding and resting, both bird species mainly sitting on the water in “rafts”, an unusual observation for albatrosses. The quantitative importance of the area for the main species is to be considered in relation to their global population as well. Southern fulmar has one million breeding pairs, mainly on the islands of the Scotia Sea Arc (Creuwels et al. 2007). Our observed total of 7,000 grey-headed albatrosses, mainly adults, represents more than 1 % of the global population of about 600,000 individuals or 3 % of the estimated 250,000 adults, of which 90,000–95,000 breeding pairs (Shirihai 2007). Obviously, their absolute number in the area must be a lot higher than the observed ones. Many individuals are known to regularly visit this sector of the southern seas (ACAP 2009). This makes the South Shetland Islands area a major feeding ground for this species classified as “vulnerable and decreasing” (ACAP 2009; IUCN 2012). For fin whale, an endangered species as well (IUCN 2012), 100,000–120,000 individuals are present in the southern hemisphere, south of 50°S in summer (Gambell 1985; Shirihai 2007). They are usually encountered in groups of 3–10 or 20 with some concentrations of 100 (Gambell 1985). During previous studies in the area, they were noted in numbers varying between 9 and 30 sightings (Pankow and Kock 2000; Branch and Butterworth 2011; Sirovic et al. 2006; Scheidat et al. 2007). The large number we encountered was not mentioned before. Other species were noted in high numbers, but their relative abundance was low: e.g. Antarctic prion with 6,400 individuals out of a population estimated as up to 20 million breeding pairs (Shirihai 2007), probably the most numerous marine bird species in the southern hemisphere. Other species were concentrated in the same area on the slope of the “continental” shelf of Elephant Island: e.g. chinstrap penguin Pygoscolis antarctica and black-browed albatross Thalassarche [melanophrys] melanophrys, confirming the special importance of the shelf slope (Fig. 4).

Fig. 4
figure 4

Position of the peak abundance for selected seabird and cetacean species

Full size image

Seasonal factors might be of importance as well, namely the probable autumn migration and accumulation of adult krill migrating before overwintering under pack ice (Lascara et al. 1999; Hewitt et al. 2004; Nicol 2006) on the one hand. On the other hand the seasonal migration of large cetaceans towards the tropics in autumn, fin whales were possibly aggregating in a pre-migration feeding situation. Concerning seabirds, the species composition was very different during the breeding season (Santora et al. 2009a, b); this difference probably reflects an important post-breeding movement for most bird species leaving their breeding grounds.

Conclusions

Low biodiversity ecosystems such as the polar marine ones are characterised by low numbers of species, of which a few only form the majority of individuals. Moreover, their distribution is very patchy with huge aggregations in very limited zones, usually at the polar front in mixed Polar/Antarctic Water or Polar/Arctic Water, respectively. These areas thus show high biological production in the water column. If such concentrations were not covered by geographical studies, the numbers and densities are dramatically underestimated. Considering that upper trophic levels distribution basically reflects prey abundance, one might guess that, similarly, studies not covering these areas also dramatically underestimate prey numbers and density.

Present observations stress the international importance of the South Shetland Islands for several species, especially in autumn. This area therefore deserves a strong protection management, including interdiction of “scientific” whaling and krill fishery. More research outside summer or winter is needed in order to confirm and deepen the knowledge of such important autumn feeding aggregations.