C-26 



lyfeLCklntosh, 197O; Everson, in press). Public expectations 

 are even higher (e.g., 200 x 10° tons, Gwynne, 1977). 



Changing the perspective from only krlll and whales to 

 the whole ecosystem, estimated harvesting levels appear to 

 be too large. A harvest of 150 x 10° tons would displace 

 half of the current predators, producing dramatic changes In 

 the ecosystem. 



The estimate of 330 x 10° tons for krlll production Is 

 a sum over the whole southern ocean area. A local harvest, 

 say 10 x 10° tons In the Scotia Sea, would probably cause 

 observable changes In predator populations there. My 

 present educated guess for a sustainable yield, based on 

 possible replacement rates and ecosystem Interactions, Is 

 30 to 60 X 10° tons per year for the southern ocean, with the 

 larger value less likely to be sustainable over a long time. 



To try to maintain balance within the food web, preda- 

 tors could be harvested In proportion to prey (krlll) 

 harvested. Sustainable harvests of seals, birds, fish and 

 cephalopods cannot be estimated well at present because of 

 lack of Information on replacement rates within each of 

 those stocks. As a rule of thumb for general consideration 

 only, a total predator blomass of about 10^ of the total 

 krlll harvest could be taken. Thus, 30 x 10° tons of krlll 

 harvested would imply a total of 3 x 10° tons for seals, 

 birds, fish and cephalopods. About half of that total 

 would be crabeater seals. A concomitant harvest would 

 necessitate management of all the resources, including 

 krlll, by one agency. To meet the objective of maintaining 

 a balanced ecosystem, the geographic locations of all the 

 harvests would have to be coordinated as closely as the 

 total removed. Predators should be removed from essentially 

 the same location as prey. Population dynamics within each 

 species must also be taken into account. 



