CCAMLR

Method of estimation of krill drift velocity in water current,using repeated acoustic transects on rectangular polygon is discribed.Dimensions of polygon and frequency of acoustic observations have to be determined with taking into account distribution patterns of krill,its patchiness, between-swarm distance (BSD) and between-clusters distance (BCD).For ocean current velocity- 20 sm/sec,ship speed-8 knots,mean BSD-5km,mean BCD-6.3 km dimensions of polygon must be 10 12 km and discretness of acoustic transects 4 hours.

Physical characteristics of 437 krill aggregations detected acoustically in the area o£ S.0rkney are discribed.All swarm parameters were registered and processed automatically.The aggregations thickness and length (2.5 m and 10.4 m)were considerably smaller, then those published by. other authers. The aggregations were on average 2.2 km apart. It is shown that aggregation parameters depend in great extent from spatial resolution capability of swarm measurements.

Observations are reported of icefish being taken as bycatch during krill fishing operations from a research vessel. The results indicate that the bycatch of fish in the commercial krill fishery may be significant in some areas of the South Georgia shelf. The problem is thought to be least in open ocean krill fishing.

Estimates are calculated for the parameter λ in the yield equation Y = λ M B0 for a kri11 fishery in which both the fishery and kri11 growth are seasonal. The criterion used for the calculations is that the probability that the kri11 spawning biomass drops below 20% of its average pre-exploitation level over a 20 year period of harvesting should not exceed l0%. The value of λ depends strongly on the value of M (natural mortality) and σR (recruitment variability), and also on the relative values of the ages at maturity and first-capture. Seasonal effects seem to be of lesser importance.