CCAMLR

The ecosystem approach to fisheries attempts to define objectives for the target species, the wider ecosystem and, critically, the fishery itself. Proposals for implementing this approach often include spatial restrictions on harvesting and it is therefore important to understand how these will affect fishery performance. One metric of potential performance is the probability of encountering exploitable densities of the target species at the scale of fishing operations. The probability of encountering exploitable densities of Antarctic krill, Euphausia superba, at the 1 nm scale during an acoustic survey was predicted by both bathymetry and the mean krill density at the larger scale at which the fishery is managed. This suggests that the risk to fishery performance will increase if management actions relocate the fishery into deeper water. The results also suggest that ecosystem models resolved to the spatial scale of management units could usefully predict effects at the scale of fishing operations. However, correct parameterisation of these models will require better characterisation of threshold densities for efficient exploitation. Finally, the distribution of both catch and fishing effort over an entire fishing season reflected the distribution of krill density observed during the survey.

Penguins are adapted to live in extreme environments, but they can be highly sensitive to climate change, which disrupts penguin life history strategies when it alters the weather, oceanography and critical habitats. For example, in the southwest Atlantic, the distributional range of the ice-obligate emperor and Ade´lie penguins has shifted poleward and contracted, while the ice-intolerant gentoo and chinstrap penguins have expanded their range southward. In the Southern Ocean, the El Niño-Southern Oscillation and the Southern Annular Mode are the main modes of climate variability that drive changes in the marine ecosystem, ultimately affecting penguins. The interaction between these modes is complex and changes over time, so that penguin responses to climate change are expected to vary accordingly, complicating our understanding of their future population processes. Penguins have long life spans, which slow microevolution, and which is unlikely to increase their tolerance to rapid warming. Therefore, in order that penguins may continue to exploit their transformed ecological niche and maintain their current distributional ranges, they must possess adequate phenotypic plasticity. However, past species-specific adaptations also constrain potential changes in phenology, and are unlikely to be adaptive for altered climatic conditions. Thus, the paleoecological record suggests that penguins are more likely to respond by dispersal rather than adaptation. Ecosystem changes are potentially most important at the borders of current geographic distributions, where penguins operate at the limits of their tolerance; species with low adaptability, particularly the ice-obligates, may therefore be more affected by their need to disperse in response to climate and may struggle to colonize new habitats. While future sea-ice contraction around Antarctica is likely to continue affecting the iceobligate penguins, understanding the responses of the ice-intolerant penguins also depends on changes in climate mode periodicities and interactions, which to date remain difficult to reproduce in general circulation models.

Correctly quantifying the impacts of rare apex marine predators is essential to ecosystem-based approaches to fisheries management, where harvesting must be sustainable for targeted species and their dependent predators. This requires modelling the uncertainty in such processes as predator life history, seasonal abundance and movement, size-based predation, energetic requirements, and prey vulnerability. We combined these uncertainties to evaluate the predatory impact of transient leopard seals on a community of mesopredators (seals and penguins) and their prey at South Georgia, and assess the implications for an ecosystem-based management. The mesopredators are highly dependent on Antarctic krill and icefish, which are targeted by regional fisheries. We used a state-space formulation to combine (1) a mark-recapture open-population model and individual identification data to assess seasonally variable leopard seal arrival and departure dates, numbers, and residency times; (2) a size-based bioenergetic model; and (3) a size-based prey choice model from a diet analysis. Our models indicated that prey choice and consumption reflected seasonal changes in leopard seal population size and structure, size-selective predation and prey vulnerability. A population of 104 (90–125) leopard seals, of which 64%were juveniles, consumed less than 2% of the Antarctic fur seal pup production of the area (50% of total ingested energy, IE), but ca. 12–16% of the local gentoo penguin population (20% IE). Antarctic krill (28% IE) were the only observed food of leopard seal pups and supplemented the diet of older individuals. Direct impacts on krill and fish were negligible, but the “escapement” due to leopard seal predation on fur seal pups and penguins could be significant for the mackerel icefish fishery at South Georgia. These results suggest that: (1) rare apex predators like leopard seals may control, and may depend on, populations of mesopredators dependent on prey species targeted by fisheries; and (2) predatory impacts and community control may vary throughout the predator’s geographic range, and differ across ecosystems and management areas, depending on the seasonal abundance of the prey and the predator’s dispersal movements. This understanding is important to integrate the predator needs as natural mortality of its prey in models to set prey catch limits for fisheries. Reliable estimates of the variability of these needs are essential for a precautionary interpretation in the context of an ecosystem-based management.

Goudier Island is located in the Palmer Archipelago, to the west of the Antarctic Peninsula; it is one of the most frequently visited tourist sites in Antarctica. A number of gentoo penguin (Pygoscelis papua) breeding colonies are located on the island and these have been the focus of one of the longest running experiments to examine the impacts of tourist numbers upon penguin breeding performance anywhere in the Antarctic. In this paper we describe the population trends and breeding productivity (chicks per nest) of the 10 colonies on Goudier Island, all of which have now been monitored for 12 consecutive years beginning in the 1996/1997 breeding season. Our results demonstrate that all colonies show considerable inter-annual variability for both the number of breeding pairs and breeding productivity. Of the six visited colonies, two showed an important and significant statistical decline in the number of breeding pairs. One of these declining colonies is used to determine the breeding chronology dates for all other colonies, an important part of the monitoring procedure used to assess breeding success. Our results suggest that in the future, it would be useful to control for this additional disturbance. Our results further suggest that understanding all of the many subtle influences that impact upon gentoo penguin breeding numbers is complex and that some factors may never be completely identified.