CCAMLR

A discriminant function using simple carapace measurements corretly determined the sex of 87% of male and 89% of female Antarctic krill Euphausia superba. This facilitated the use of sex-specific regression models which increased the accuracy of total length estimates.

The distribution of whales and krill in two survey boxes north of South Georgia was examined by comparing sightings and underway acoustic data collected as part of a multidisciplinary research cruise during January/February 1998. A total of 222 cetaceans of 10 species was recorded with southern right whale (Eubalaena glacialis) and humpback whale (Megaptera novaeangliae) the two most frequent. The largest aggregation of cetaceans (21 southern right whales, 18 fin whales (Balaenoptera physalus), 4 sei whales (B. borealis), 1 humpback whale and 8 hourglass dolphins (Lagenorhynchus cruciger)) occurred close to the, largest single aggregation of krill. The level of association between baleen whales and krill was examined at a number of spatial scales. There was a positive relationship between whale abundance and mean krill density at the largest spatial scale examined (80 x 100km ). At progressively smaller scales the relationship weakened, due mainly to the increased frequency of areas of high krill density where whales were not recorded. In particular whales were absent from inshore areas (up to 300 m depth) that had higher mean krill densities compared with areas where whales were recorded. To compare krill and whale distribution, particularly at smaller scales, may require information on krill swarm structure and density. Such information may also be crucial to understanding the role of scale-dependence in potential interspecies competition among krill-feeding marine predators.

The influence of two sampling protocols on diet determination of two marine predators, the Gentoo penguin Pygoscelis papua and Antarctic fur seal Arctocephalus gazella, was investigated. The collection of diet samples on three occasions over a two week period was compared with collecting all samples during a single session, as current CEMP monitoring protocols recommend. Some differences in the mass of food recovered from penguins were found but this was attributed to the mass of penguin sampled. There were no differences in diet composition between protocols and although body mass was a significant determinant of the mean length of krill Euphausia superba recovered from penguins, there were no differences between sampling protocols. This study has shown that differences between sampling frequencies are small and a variety of sampling protocols can produce results acceptable for inter-annual monitoring. Mass of sampled individuals can account for significant variation and should be recorded, especially if sampling frequencies and sizes are low.

This study examined the feasibility of using time depth recorders (TDRs) to measure light levels experienced by a diving marine mammal. TDRs were deployed on 10 female Antarctic fur seals (Arctocephalus gazella) at Bird Island, South Georgia (54° 00' S 38° 02' W) in the 1994/5 austral summer. Depth and light measurements were made during 11 foraging trips which lasted on average (± SE) 7.1 ± 0.7 d. A total of 25657 dives were recorded with a mean dive depth and duration of 18.0 ± 3.6 m and 49.5 ± 6.9 s, respectively. Depending on time of day, fur seals experienced between 6 and 57 % of the surface illumination when diving. Illuminance ranged from full sunlight (102 lux) at the surface to minimum starlight (10-6 lux) at night and during deep daytime dives. The change in il1uminance with depth did not follow a simple exponential decrease as recorded in clear oceanic waters. TDRs recorded strong light attenuation in the top 50 m of the water column. The mean attenuation coefficient was 0.140 ± 0.014 m-1 which was in the upper range of values measured by ship surveys at South Georgia. When fur seals made shorter and shallower dives TDRs also recorded greater light attenuation. These findings suggest that Antarctic fur seals forage in relatively turbid waters containing large concentrations of phytoplankton and 1 or dense aggregations of krill. Night-time foraging and deep daytime diving indicates that fur seals are well adapted to locating prey in low illumination.

After almost a ten-year gap in fishing for Champsocephalus gunnari in the area of South Georgia and Shag Rocks, the Russian large-sized trawler ZAKHAR SOROKIN (the MOONZUND-type) attempted a short-time fishing operation in February-March, 1999. Overall, 85 pelagic trawlings yielded 264 tons of C. gunnari. Search for the fish stocks covered all the traditional sites; this allowed observers to determine a clear correlation between concentrations of adult fish and krill on the north-western shelf of South Georgia. At the same time, lasting of the feeding stocks of C. gunnari depended on weather conditions. Comparison of the obtained results with data of trawl-and-sonar surveys revealed regular occurrence of dense concentrations of C. gunnari on the north-western shelf of South Georgia.

This paper describes the diet and foraging behaviour of Ad6lie penguins at B6chervaise Island during 1998-99, a season of high chick mortality. One thousand nests perished in the three weeks following first hatch, when the chicks were very small. A total of 802 chicks reached crèche age from 1880 nests compared to previous "good" seasons in which between 1200 and 1800 chicks crèched on the island. Evidence from analyses of foraging location, foraging trip duration and diet led to the conclusion that the death of chicks during the guard stage resulted from decreased feeding frequency due to adult birds spending longer at sea foraging than in normal years. In most previous seasons birds have foraged both at the continental shelf edge (particularly females) as well as locally (particularly males). This season male penguins carried out fewer local trips, and both sexes spent longer at sea than in "good" years. Meal masses brought back to the chicks appeared normal, but feeding frequencies were reduced. These findings contrast with observations in 1994-95 Ca season in which all chicks starved to death) when smaller meals were delivered and birds foraged further offshore than in this or any other season studied. The significance to CEMP of these variations in foraging behaviour is discussed.

We irradiated captive juvenile Euphausia superba in the laboratory with lower than spring surface levels of ultraviolet-B, ultraviolet-A and photosynthetically active radiation, in order to examine their response in terms of mortality and generalised activity. Levels of photosynthetically active radiation 3–5 times below surface irradiance caused krill to die within a week, while animals in the dark survived. Addition of ultraviolet-B typical of depths up to 15 m were found to significantly accelerate mortality and lead to a drop in activity in all experiments. A drop in activity in krill exposed to ultraviolet-A wavelengths was evident without an increase in mortality. The protein content of animals from various treatments was found not to vary.

This background paper consists of the agenda and information for the Second International Symposium on Krill which has been partially sponsored by CCAMLR. The symposium will be held at the University of California, Santa Cruz, from 23–27 August 1999. Further information can be found on the website: http://www2.ucsc.edu/people/msmangel/Krill Symposium.html

The global abundance of krill is estimated using recent estimates of krill density from acoustic surveys and historical information on the overall range of krill. The biomass estimates fall between 64 and 137 million tonnes - at the low end of values that have been suggested in the past. The differences between our estimates and others can be explained by a number of factors such as: an underestimation of the range, or of the acoustic biomass estimates, the possibility of a large, undetected krill population, and the overestimation of the demand for krill by predators. Even if these low global krill biomass estimates are correct, regional and global precautionary limits are still likely to rise as a result of new surveys because the method used to calculate precautionary limits uses a value of 11 % of the biomass. Additionally, the current precautionary catch limits in the South Atlantic are set using the old target strength which has effectively underestimated the krill biomass by a factor of three so the new survey of the South Atlantic is likely to result in an effective biomass which is greater than that used in the past. The seasonal and local consequences of elevated catch limits will have to be taken into account when managing an expanded krill fishery and the appropriateness of using the existing statistical divisions as management areas will have to be considered.