Killer whales

On 23 January 2004, we observed C ecotype killer whales (Pitman & Ensor Reference Pitman and Ensor2003), seemingly feeding, off Cape Washington and travelling within 2 m of the pack ice edge toward Campbell Glacier tongue. We followed them by helicopter as they tracked along the regular pack ice edge; ice thickness was about 2 m. The animals stopped at a small right-angle embayment in the ice formed by a large multi-year floe. Two adult (female-sized) animals emerged from under the ice floe and then spy hopped. In the meantime, an adult male arrived, remaining motionless for about 20 sec before diving under the ice. Of the two previously mentioned animals, one moved away to then return to the ice, while the other dove under the ice floe belly up. At this stage other animals gathered. A mother-calf pair emerged close to the ice emitting bubbles of air, joined soon by another female-calf pair. These eight animals were joined by four more, spy hopping around the ice floe, which then cracked in two. No penguins or seals were observed in the immediate area of the episode.

The event lasted about 5 min, in the course of which the 12 animals displayed a behaviour similar to the herding phase of the ‘carousel method’ (Similä & Ugarte Reference Similä and Ugarte1996). We noted that individuals swam with the white underside of their body towards the ice floe and emitted bubbles close to the surface, in a way comparable to the cooperative feeding reported by these authors for killer whales preying on herring (Clupea harengus) in northern Norway.

Therefore we interpreted this behaviour as a feeding-related activity intended to encircle the (fish?) prey to prevent them from escaping to deeper waters. To date there is little direct information on the diet of the C form of the killer whale (Ainley et al. Reference Ainley, Ballard and Dugger2006a). It is recognized as a fish-eating species targeting toothfishes (Dissostichus spp.; Berzin & Vladimirov Reference Berzin and Vladimirov1983), but it is thought to also eat Antarctic silverfish (Pleuragramma antarcticum Boulenger) (Ainley et al. Reference Ainley, Ballard and Dugger2006a, Reference Ainley, Toniolo, Ballard, Barton, Eastman, Karl, Focardi, Kooyman, Lyver, Olmastroni, Stewart, Testa and Wilson2006b). A similar behaviour has been observed in a Weddell seal (Leptonychotes weddellii Lesson) in McMurdo Sound (Ross Island vicinity), where the seal attempted to catch fish in a crevice under the ice by expelling blasts of air (Davis et al. Reference Davis, Fuiman, Williams, Collier, Hagey, Kanatous, Kohin and Horning1999). In that area, the Weddell seal feeds extensively on the Antarctic silverfish (Burns et al. Reference Burns, Trumble, Castellini and Testa1998) but also the bald rock cod/notothen (Pagothenia borchgrevinki Boulenger; Kim et al. Reference Kim, Conlan, Malone and Lewis2005). Between 1996 and 2006 at Cape Crozier, GB and DA frequently observed groups of 10–30 type C killer whales or of 1–12 minke whales apparently foraging along the Ross Ice Shelf edge and nearby fast ice edges. In several instances, killer whales were observed near the surface with belly up. The whales were typically seen either travelling to and from the ice edges, or diving at or under the ice edge repeatedly over several hours, indicating that most foraging associated with either glacial or fast ice edges.

Other predators

On 9 January 2007, sighting from the 60 m high bluff of Cape Royds (Ross Island) at a distance of about 500 m, DA observed one minke whale, in loose pack ice, swimming just beneath the surface in a tight circle having a radius of about 30 m, its back to the centre. It then would disappear from sight for a couple of minutes, emerging again in the immediate vicinity. It did this several times throughout the day in an area called “Backdoor Bay”, in McMurdo Sound.

In the same general area, from the same vantage point, DA also observed several instances of small flocks of Adélie penguins, 10–15 per flock, generating vortices by swimming round and round just beneath the surface, the radius of each vortex being about 20 m. The flock would completely disappear but then appear in a rush, popping vigorously to the surface within the vortex centre. At times, they would then porpoise as fast as they could in one direction for about 30–40 m, whereupon they would make a right angle turn as they dove. They would appear 10 m from where they dove. They would then porpoise in the opposite direction, repeating the same manoeuvre. At the time, individuals in the population at Cape Royds were feeding only on fish, likely silverfish, as judged by the food regurgitated to chicks (Ainley et al. Reference Ainley, Ballard and Dugger2006a).

In most years at Cape Crozier, we have observed penguins generating vortices in waters along the face of the Ross Ice Shelf, but usually at distances at 1–2 km or more, too far to be certain about what they were doing. On one occasion (December 2002), closer observations of such vortices made from the ice edge (by GB and others) revealed groups of several hundred Adélie penguins and dozens of emperor penguins repeatedly making extended dives under the fast ice locked between the Ross Ice Shelf and the Cape Crozier Adélie penguin colony. At any given time, one to several dozen penguins would dive simultaneously, having spent 20–30 sec at the surface, and we could see that the penguins were diving towards the ice edge. We do not know how deeply they were diving but at least 10 m (potentially much deeper). The emperor penguins were observed feeding fish, probably silverfish (judging from the black guano), to their nearly-fledged chicks, which had gathered nearby.

Although, a priori, we cannot exclude other fish species (such as Trematomus newnesi Boulenger (Williams Reference Williams1988) or Pagothenia borchgrevinki, which is found in association with ice shelves (Gutt Reference Gutt2002)), the predatory activity performed by the orcas and the other predators that we observed in Terra Nova Bay and off capes Royds and Crozier was also likely directed at silverfish. We have rarely found other fish species in the diet of penguins nesting at Royds or Crozier (Ainley et al. Reference Ainley, Ballard, Barton, Karl, Rau, Ribic and Wilson2003), although we have a photograph of a penguin trying to swallow a large Trematomus spp. Pleuragramma antarcticum is a small, shoaling, herring-like species dominating, in both number and biomass, the water column fish fauna of most shelf areas of the Southern Ocean (Eastman Reference Eastman1993) and is recognised as a principal prey species for many predators in the Ross Sea (La Mesa et al. Reference La Mesa, Eastman and Vacchi2004, Ainley et al. Reference Ainley, Toniolo, Ballard, Barton, Eastman, Karl, Focardi, Kooyman, Lyver, Olmastroni, Stewart, Testa and Wilson2006b). In particular, the coastal region of our observations is within a geographic area very important for the life cycle of silverfish, as deduced by a high abundance of larvae (Granata et al. Reference Granata, Cubeta, Guglielmo, Sidoti, Greco, Vacchi and La Mesa2002), and of a recently discovered large spawning ground near the Italian Mario Zucchelli Station (Vacchi et al. Reference Vacchi, La Mesa, Dalù and Macdonald2004). In the same area, P. antarticum represents the most important prey for the Cape Washington emperor penguins at the colony (Cherel & Kooyman Reference Cherel and Kooyman1998); they are important as well in the diet of Adélie penguins at Edmonson Point (Clarke et al. Reference Clarke, Manly, Kerry, Gardner, Franchi, Corsolini and Focardi1998).