CCAMLR

This paper assesses the status and abundance of M. carinatus on BANZARE Bank (Statistical Division 58.4.3/1). The long-term precautionary yield was estimated using the Generalised Yield Model (GYM) used previously by the CCAMLR Working Group on Fish Stock Assessment. Estimates of population and biological parameters were not available for this species at BANZARE Bank. Length and weight data were taken from a trawl survey conducted at Macquarie Island in 1999. Where parameters were not directly available for M. carinatus, estimates were obtained from the literature for similar species elsewhere in the world. The trawl survey data was used to calculate estimates of biomass and density for the area and showed that M. carinatus did not appear to be aggregated over any part of BANZARE Bank. The long-term annual yield calculated for M. carinatus was 550 tonnes, based on a critical value of y of 0.033 (the proportion of the estimate of pre-exploitation biomass) found using the CCAMLR decision rules. The binding rule for this assessment was the decision rule on the probability of depletion below 0.2 of the median pre-exploitation spawning biomass. This assessment could be used as a basis for setting catch rates of M. carinatus in longline fisheries targeting Dissostichus eleginoides. Applying the critical y to the mean density observed in the survey results gives a catch rate of 5.81 kg/km2 which translates into a precautionary yield of 17.9 tonnes per fine-scale rectangle. Such a yield represents 18 % of the total catch allowed for D. eleginoides in fine scale areas in new and exploratory fisheries. This catch rate may be useful in setting general by-catch rules for M. carinatus to protect the species from localised depletions.

Based on a survey of Dissostichus eleginoides in the Heard Island region in April 1999 and information from an observer program on the commercial fishery begun in 1997, this paper aims (i) to provide an assessment of biomass and stock structure in the vicinity of Heard, (ii) to examine the relationship between commercial fishing grounds and the wider plateau and banks where surveys have been undertaken previously, and (iii) to revise the population parameters, where possible, including fishing selectivity, growth, maturity and recruitment. The results demonstrate clear differences between populations of Dissostichus eleginoides at Heard Island in the southern Indian Ocean and South Georgia Island in the South Atlantic. For example, Patagonian toothfish at Heard Island are considerably older and slower growing than those at South Georgia. This raises questions as to the appropriateness of the current range of natural mortality used in the assessment of yield at Heard Island. The revised estimates of recruitment indicate that the variability of recruitment at Heard Island is much greater than originally estimated and that, in recent years, there has been an influx of a large number of young fish compared to the period during the first surveys in the early 1990s. A notable outcome of the 1999 survey is the identification of Ground B as the most important area for 7 to 8 year old fish in the Heard Island region. Careful thought will need to be given to how best to manage a stock that aggregates in this manner while still juvenile and with some years before becoming reproductively mature.

The results of icefish ecology were summarized for the period of above 20 years of South Georgia Subarea study. Ecological aspects were considered taking into account temporal-spatial variability. The functional structure of Ch.gunnari distribution area was described on the basis of juvenile fish distribution at different stages of physiological state. Juvenile fish mainly distribute in the south, immature fish - in the cast and southwest, while adult at the north of the shelf. It was revealed that the life mode changed from pelagic one occurring at early life stages, to near-bottom - pelagic one after 20 cm length reaching and the bottom one at length above 40 cm. Ch.gunnari undertakes vertical migrations: young fish above 4 cm in length distribute within 75 cm from the bottom, and migrate into the less depths in the day-time and to the bottom at night. Immature fish distribute in the water column for the most time and is fished with midwater trawl. Its diurnal migrations pattern is similar to that of the larger fish. Diurnal migrations of adult fish are opposite to those of juveniles: it migrated to the bottom in the day-time and into the water column at night. Mature fish form dense aggregations near the bottom where it is fished with bottom gears. The annual cycle of fish is subdivided into 3 periods: feeding (October-March), spawning (April-June) and wintering (July-September). Feeding migrations of fish above 25 cm in length occur in the northern shelf and associate to migrations of krill being the main, food item of icefish. Two major directions of prespawning migrations were revealed: from the north-east to coastal areas in the northern and north-eastern part and to the open shelf areas in the southeast, while in the north-west part it migrated westwards and south-westwards. Post-spawning migrations had the opposite direction. Overwintering of icefish occurred at the depths of 200-250 m far from the coast mainly northwards of the Island.

The results of experiments on the South Georgia Shelf carried out in the period of 1986 to 1990 have been presented. Six (6) daily stations were occupied where vertical distribution of fry and young individuals (including adult specimens) of Champsocephalus gunnari, Champsocephalus aceratus, Pseudochaenichthys georgianus (Channichthiidae family) and Nototheniops larseni (Nototheniidae family) was studied. The most representative data were obtained for Ch.gunnari and N.larseni. The vertical migration trend was the most distinctive in Ch.gunnari fry. The individuals used to ascend towards the water column before the sunrise, being finished their return migration before the sunset. Juvenile and adult individuals were found in the pre-bottom layer by day, being presented their small quantities in the water column, including the superficial layer, at night. N larseni fry vertical migration pattern was poorly expressed during 24 hours. It was distinctly observed in juvenile and adult individuals: they were kept close to the bottom by daylight and were not caught within the pre-bottom layer. At night (after the sunset) they ascended into the water column where they were found in the superficial layer. No substantial variations in middle size fry of both species were found at different depth levels during 24-hour periods. Ch. aceratus fry were found within the whole water column during 24-hour periods, being mainly observed within two lower depth levels. Juvenile and adult individuals were kept by day close to the bottom and within the whole water column at night. However major catches were taken within the limits of the pre-bottom depth level, being higher as compared to those taken by day. Ps.georgianus fry were caught during 24-hours nearly in equal quantities in the pre-bottom layer by day and at higher depth levels at night. Juvenile and adult individuals were almost not found in the day-time, being observed within the whole water column at night and mainly caught within the pre-bottom layer.

1. Bottom trawl surveys on the South Georgia shelf indicate episodic declines in the abundance of the mackerel icefish Champsocephalus gunnari that, since 1990, are not directly attributable to commercial fishing.
2. The greatest effect has been observed in years when krill are known to have been scarce on the South Georgia shelf.
3. It is thought that C. gunnari survivorship is closely related to, but indirectly influenced by, krill availability.
4. Examination of food chain relationships indicates that predation by fur seals is likely to play a major role in the observed changes in C. gunnari abundance.
5. Periodic shifts in food chain relationships, involving krill, C. gunnari and fur seals, are postulated that could explain the episodic declines in the abundance of C. gunnari in the absence of fishing.
6. The study demonstrates how a key predator can have an over-riding impact on a commercial fish species, highlighting the importance of the ecosystem approach to fisheries management.