CCAMLR

Total scattering cross-sections (st) of Antarctic krill, Euphausia superba, were acoustically measured over a broadbandwidth (36 to 202 kHz) using a new technique [J. De Rosny and P. Roux, 2001, J. Acoust. Soc. Am. 109(6): 2587:2597]. From 18 February to 9 March 2002, mean total target strengths ( ??10log(??4?) t TTS ), were measured from groups of 57 to 1169 krill (average standard length = 31.6 mm; standard deviation = 6.6 mm), at the Cape Shirreff field station, Livingston Island, Antarctica, and aboard R/V Yuzhmorgeologiya. Chirp pulses were sequentially transmitted with an omnidirectional emitter into one of three glass carboys containing groups of krill swimming in 9.3, 19.3 or 45.9 l of seawater ( 0.6 ??temperature??3.6 °C). Between each pulse, the krill moved within the fixed-boundary-tank and the modulated reverberations were sensed bi-statically with three omnidirectional receivers. At each center frequency (fc), the coherent energy in 200-pulse ensembles identified sound scattered from the tank. The incoherent energy described total sound scattering from the krill. Thus, the TTS at each fc was extracted from a correlation analysis of energy reverberated in the tank. Measurement accuracy was determined using standard metal spheres for references [Demer et al., in-press, J. Acoust. Soc. Am.], and the precision was estimated from the variability in krill TTS measurements. Moreover, empirical estimates of mean s t were statistically compared to a recently proposed krill scattering model based on the distorted wave Born approximation (DWBA) [McGehee et. al, 1998, Deep Sea Res. Pt2, 45(7): 1273-1294], which has been enhanced to account for the stochastic nature of sound scattering (SDWBA) [Demer and Conti, submitted, ICES J. Mar. Sci.], and integrated over all scattering angles (SDWBATTS). This study improves upon methods for acoustical identification and target strength estimation for Antarctic krill, thus reducing the uncertainty in biomass estimation using multi-frequency echo sounder data and echo integration methods.

Demographic parameters (age-specific mortality rates, fecundity levels and population numbers) were analysed for the Béchervaise Island Adélie penguin colony in eastern Antarctica after 12 years of CEMP monitoring. A life table was constructed, and predicted rates of population growth and breeding success calculated. As for most long-lived seabird species, growth/decline rates of the Adélie penguin population at Béchervaise Island were found to be more sensitive to changes in annual survival rates, especially of young breeding adults, than to changes in fecundity parameters. These findings are discussed briefly in relation to other CEMP parameters, environmental factors and fishery regulation.

Seven fledging Adélie penguin (Pygoscelis adeliae) chicks and four post-moult adults were satellite-tracked using the Argos system during the winters of 1995-97 and 1998 respectively. Six fledglings departed from Béchervaise Island near Mawson station (67°35'S, 62°49'E) during late February 1996 and 1997 and were tracked for up to five months before transmissions stopped. The seventh chick left Magnetic Island near Davis station (68°33'S, 77°54'E) in February 1995 and was tracked for 32 days. All fledglings travelled northward initially, then westward along the edge of the fast-ice or in the pack-ice. Fledglings had travelled between 536 and 1931 km to the west of their natal colonies by the time transmissions ceased. Adult Adélie penguins were tracked between March and October 1998, following completion of their annual moult at Bechérvaise Island. Instruments were factory-set to transmit intermittently for the first five months and one day in four thereafter. Adult birds travelled westward until July after which time they moved north within the expanding pack-ice into known areas of high krill concentration before returning eastward toward their breeding sites. Penguin movements over the winter months were closely related to those of sea-ice in the region. Ice motion patterns were in turn influenced by gyral oceanic current systems and wind. We propose that large gyral oceanic systems provide a means for Adélie penguins to reduce costs of transport as they travel into regions of high productivity during winter and return to their breeding colonies in spring. Locations of boundaries of oceanic gyres may thus be useful to CCAMLR in the regulation of the Antarctic krill fishery as a means of delimiting management regions.

We examined temporal variability in a series of CEMP parameters collected over the period 1991/92 to 2001/02 on Adélie penguins at Béchervaise Island. Parameters relating to chick survival (brooding counts, 2/3 crèche counts and fully crèched chick counts) show large temporal variability while the other parameters were more stable. We also explored the correlation between CEMP parameters, and the correlation of each parameter with breeding success. Breeding success was measured in terms of 1) the absolute number of chicks on the island which fully crèche and 2) the number of chicks crèched per nest with eggs. We found a low degree of correlation between incubating and brooding nest counts which supports the notion that events occurring during the hatching period are crucial for chick survival. The sex of the foraging birds and the timing of foraging trips were important in determining whether foraging trip duration was correlated negatively to breeding success. Additionally, we examined within season penguin weights, the simplest output from the Automated Penguin Monitoring System (APMS), in terms of their correlation with breeding success. This analysis showed that lower weights of females at the time they depart after egg laying would appear to be the first indication that a season may have low breeding success. The results obtained throughout these analyses indicate that knowledge of the sex of birds is important for understanding interrelationships between CEMP parameters and breeding success.

In 2000 WG-EMM discussed the need for region-wide and circum-Antarctic estimates of abundance of land-based predators. At the 2001 meeting of WG-EMM it was proposed that a workshop be held in 2002 to assess the feasibility of such broad-scale surveys, and a small intersessional correspondence group was formed to scope out the terms of reference and method of operation for the workshop. The intersessional group developed a draft decision-making framework for planning regional scale surveys and agreed that it be applied at the workshop to a selected region as a means of both assessing and/or developing the framework and undertaking preliminary plans for the region. The framework is designed to be general enough to apply to any species or scale, and addresses the issues of survey specification, survey design, review and use existing information, and estimation of availability (proportion of total population not able to be surveyed) as an integral part of the survey design. The framework also recognises that new technologies and methodologies need to be considered for use wherever possible.

Naganobu et al. (1999) had assessed variability in krill recruitment and density with hypothesized environmental factors; strength of westerly winds (westerlies) determined from sea-level pressure differences across the Drake Passage, sea ice cover and chlorophyll-a in the Antarctic Peninsula area during 1982-1998. They found significant correlations between krill recruitment and those factors. The westerlies were especially regarded as a key environmental index. Fluctuations in the westerlies across the Drake Passage were referred to as the Drake Passage Oscillation Index (DPOI).
We planned to extend time series of DPOI using historical data. We searched the historical data and found the time series since the 1950s until 1988 at the Web sites of the Carbon Dioxide Information Analysis Centre for Rio Gallegos and since the 1940s until 1998 at British Antarctic Survey for Esperanza. Here we calculated a time series of DPOI from 1952 to 1988 at this stage. Time series since 1988 will be soon calculated after obtaining appropriate data.
The total number of monthly data used from 1952 to 1988 was 420. The mean, median, and mode were 13.6, 13.8, and 14.0 hPa, respectively. The maximum, minimum, and range were 27.5, -6.4, and 33.9 hPa, respectively. The standard deviation was 6.2. Linear regression declined as a whole from 1952 to 1988. A time series of 3-month running mean suggested considerable seasonal variability of climate. A time series of 12-month running mean indicated various yearly changes without seasonal variability. High DPOI periods, not less than 16 hPa, were mostly observed in the period before 1964 and only in 1973 and 1986-88 after 1964. Low DPOI periods, less than 14 hPa, were continued longer after 1964. A low DPOI less than 10 hPa appeared in 1967 and 1980. Intervals between the years of low DPOI were generally observed for around 3 years except 6 years between 1958 and 1964.

We assessed the environmental variability of Antarctic krill (Euphausia superba) distribution with comparison between the CCAMLR 2000 Survey and similar scale datasets partially by the Japanese R/V Kaiyo Maru Survey in the 1987/88 austral summer season in the Scotia Sea. There were distinct differences between the 2000 and 1987/88 Surveys with regard to sea ice extent, oceanographic structure and krill distribution. The sea ice cover in 1987/88 extended northward widely during the last winter season such that sea ice remained around the South Orkney Islands until December 1987. In contrast, the sea ice cover in 1999/2000 reduced southward such that no sea ice remained around the South Orkney Islands in December 1999. The Antarctic Surface Water mass, consisting of Winter Water and Summer Surface Water, in 1987/88 extended northward and covered a large area in the Scotia Sea. In contrast, the Antarctic Surface Water in 2000 reduced southward. Geographical distribution of krill, which approximates the area of the Antarctic Surface Water, in 1987/88 extended northward with high density. In contrast, the distribution of krill in 2000 reduced southward with low density. To generally understand the above relationships between oceanographic structure and krill distribution, we introduced integrated water temperature from the surface to 200m (Q200) as an environmental index indicating the structure of the upper ocean, that is referred to as the Environmental Index (EI Q200 ). The isoline of EI Q200 =0.0 ? was located near 60S northward off the South Shetland Islands in 1987/88. In contrast, the isoline of 0.0 ? in 2000 was located in the Bransfield Strait and Weddell Sea southward off the South Shetland Islands. The Antarctic Surface Water in 1987/88 clearly developed northward compared with 2000 reduced southward. The geographical distribution of krill ranged over the area under the isolines of EI Q200 =1.0 ? in the western waters and 2.0 ? in the eastern waters of the Scotia Sea. Krill density became higher with the colder isolines of EI Q200 =0.0 ? , especially south of its steep gradient, namely, the Southern Boundary of the Antarctic Circumpolar Current. It suggested that the geographical distribution of three krill size clusters in the 2000 Survey (Siegel et al., 2002) corresponded with the distribution pattern of EI Q 200 on the whole.

Over 120 adult Adélie penguins were found dead in unusual circumstances on Welch Island near Mawson between 23 November and 4 December, 2001. It is concluded the most likely cause of death was severe injury from being crushed by ice at the ice land interface. Circumstance surrounding the death of the penguins suggested initially that disease may have been implicated and so investigations were treated accordingly. The sampling protocols developed by CEMP were used in these investigations.

One step toward defining small-scale management units is to determine the areas most likely to be foraged by predators from one year to the next, i.e. what is a predator’s feeding range taking into account interannual variation in foraging locations? This paper considers the issues to be addressed in answering that question. The proposed method for defining foraging ranges is based on an approach used to define fishing grounds. The data which is considered as an example here consists of location/time recordings from a satellite tracking system. This data is used to generate a map of feeding effort and this is used to delineate a feeding ground for Adélie penguins on the Mawson coast in eastern Antarctica. This method builds on existing methods but incorporates tools for pooling information across colonies, years, and species to define individual species and pooled foraging ranges. The establishment of these foraging ranges for the purposes of small-scale management units may need to be examined in three parts. The first part is to determine whether the results would be different for different seasons e.g. summer vs winter. The second part is to establish the combined foraging ranges across a number of species for which tracking information is available – pooled foraging ranges. The third consideration is whether some species with low colony biomass have large proportions of their foraging ranges falling outside of the specified pooled foraging ranges. If this makes those species vulnerable in the management process then consideration will need to be given to including those ranges as special extensions to the foraging grounds. Thus, a comparison of foraging ranges for individual species with the pooled foraging range would be a useful step in this process.

This paper provides a method for delineating krill fishing grounds in Area 48 based on commercial catch data for the region held in the CCAMLR database. It also summarises available information on krill distribution and abundance and movement for the region, which can be used to help understand the relationship between the fishing grounds and the krill population. We define a fishing ground as being a predictable location where the fishery obtains relatively reliable catches from one year to the next over a number of years. The quantity of interest is not only the total catch obtained from a location, say a 10 x 10 nmile area, over the years but how important that location is to the fishery each year, which is judged by that location providing a reasonable catch in a given year and that the catch remains sufficiently high on average over a number of years. We call this value the normalised longterm average catch (shortened to the term ‘normalised catch’). An important consideration is the threshold for the normalised catch, such that locations would generally only be considered for inclusion in a fishing ground if their values were greater than the threshold. A method for choosing a threshold is given. The boundary for a fishing ground should predominantly include only locations for which the normalised catch is greater than the threshold. Some simple criteria for designating fishing grounds are presented. The type of analytical tool needed to convert the data to a longitudelatitude grid of normalised catches and for determining boundaries on the grid according to the criteria is also discussed. The components of this process are developed using the commercial krill catch data available in the CCAMLR database.