Clark and Tracey: Population changes of Hoplostethus atlanticus on the Challenger Plateau 247 



a reduction in the number of spawning schools, 

 and a marked decline in biomass. Stock reduction 

 analyses have estimated virgin biomass to be 

 about 110,000 t. The stock had declined to about 

 20% of this by 1991, well below the optimal long- 

 term biomass of 30% of virgin levels predicted by 

 computer modelling under an F 1 fishing strat- 

 egy (Clark, 1992). 



It is clear that the Challenger Plateau spawn- 

 ing population declined rapidly and substantially 

 with commercial fishing in the 1980s. However, 

 it is uncertain exactly how much of the change 

 was directly attributable to the fishery. There are 

 no data on size of the stock prior to its exploita- 

 tion, so the level of any natural fluctuations in 

 population size and distribution is unknown. It 

 is possible that stock size might have decreased 

 in the absence of any fishing, but it seems very 

 unlikely, given the longevity of orange roughy, 

 that such changes would occur as rapidly as we 

 observed. There could also have been a progres- 

 sive change in the availability of fish in the area 

 covered by the trawl surveys or bulk of the com- 

 mercial fleet. Adult orange roughy do not neces- 

 sarily spawn each year (e.g. Bell et al., 1992; 

 authors' unpubl. data) and hence may not migrate 

 to the general spawning area. In the early years 

 of the fishery, large catches were taken over much 

 of the year, but the period of large catches became 

 progressively reduced to the winter months (Clark, 

 1991a). This suggests that initially there were resi- 

 dent fish on the grounds, with a migratory compo- 

 nent which used the area for spawning only. If this 

 latter group had a variable number of spawners 

 each year, this could have affected how well trawl 

 survey or CPUE data reflected true biomass. How- 

 ever, if this occurred, greater variation in abundance 

 indices between years than observed would be ex- 

 pected. There could also have been other spawning 

 grounds that fish from the 'Challenger stock' used. 

 However, there were no indications of this from ei- 

 ther commercial or research survey trawling beyond 

 the main grounds. A further alternative explanation 

 for the observed decline in abundance could be a 

 change in vulnerability to the trawl gear, either by 

 fish residing above the bottom in midwater, or by 

 heavy fishing pressure disrupting existing aggrega- 

 tions or preventing their formation. There is no in- 

 formation on the former (although no fish marks 

 were noted on the echosounder above the bottom 

 during the last two research surveys), but the lat- 

 ter is likely to have occurred (Clark and Tracey, 

 1991). The effects of fishing on school stability could 

 have resulted in a greater decline in catch rates 

 than true abundance, but reduced stability of schools 



