CCAMLR

The survey comprised 48 bottom trawl hauls taken in subarea 48.3. About 80% of the hauls were intended as replicates of the position of hauls in the 1994 survey, allowing site by site comparison between both cruises. Of these, 20 hauls were located at less than 0.5 miles and 11 between 0.5 and 1.0 miles from those in the previous survey. Mean densities were obtained considering the spatial aggregation of stations in a nested ANOVA model. An increase over last year's values was observed in the density of several species, particularly Champsocephalus gunnari at the shelf of the South Georgia Islands. The age and size structure of this species in the whole subarea 48.3 is characterized the predominance of the younger fraction (below 4 years: 99.3% at the South Georgia Islands and 93.4 % at Shag Rocks). Along the cruise, 128 oceanographical stations were occupied. Oceanographical conditions were largely within the historical limits except for a the possible presence of a large eddy in the northern part of the Scotia Sea. Plankton samples contained krill more often than last year and an hydroacoustical survey for krill which covered a total of 800 nautical miles in the South Georgia shelf has shown the presence of krill aggregations in larger quantities than in last years survey.

Mean densities and length structure of Champsocephalus gunnari obttained in the 1994 and 1995 surveys conducted in subarea 48.3 by BIP Dr: E L HOLMBERG are compared. Haul locations were determined by a search procedure yielding randomly selected clusters of stations located within the 50–150; 151–250 and 251–500 m depth strata. The cluster structure is considered as a nested factor in an ANOVA design. Forty five stations in the first cruise and 48 in the second were analyzed. Of these, 51 pairs of 1995 and 1994 hauls located at less than one nautical mile were formed and their differences analyzed in a nested ANOVA and a paired t-test. Significant increases in density have been detected at all depth strata in the South Georgia Islands shelf, while density remained stable at Shag Rocks. The fit of a probability function to mean densities should be improved, but their differences at replicated sites fulfilled the requirements of the statistics used, stressing the value of site comparison in monitoring the stock. A relative stability was found in the components of the variance of the mean density, thus allowing their application to design optimization. The analysis of length distributions confirmed the predominance of young specimens and has shown to the North and East of the South Georgia shelf the presence of relatively large quantities of recruits. The stock is considered to be experimenting a recovery after an overfishing event.

Haul by haul Chilean and Argentinian data show that fishing forDissostichus eleginoides in 1994/95 took place in two major fishing grounds, West of Shag Rocks (Wshag) and North-West of South Georgia (NWSG). In March there was a high proportion of large females at Wshag (72%) but a nearly equal proportion of males and females at NWSG. It is suggested that a strong influx of small fish at Wshag in April reduced the mean length in the catch and increased CPUE (numbers of fish per hook), although sex ratio remained biassed in favour of females. Similar changes in mean length and CPUE were not seen at NWSG, although the sex ratio did change in favour of males (60% male by early May). The changes at Wshag are hypothesised to be the result of a prespawning migration of sub-adults, especially males, from shallow waters to deeper spawning grounds just prior to spawning in June-July