Rodgveller et al: Evidence of hook competition in longline surveys 
365 
Stations sampled (▲) during the annual National Oceanic and Atmospheric Admin- 
istration Alaska Fisheries Science Center (AFSC) sablefish ( Anoplopoma fimbria) 
longline surveys, 1979-2003 within the six sablefish management areas (1-Bering 
Sea, 2-Aleutian Islands, 3-Western Gulf of Alaska, 4-Central Gulf of Alaska, 5-West 
Yakutat, and 6-East Yakutat and Southeast [Alaska]). The AFSC Gulf of Alaska 
slope trawl survey stations (not shown) are also located along the continental slope 
above the 1000-m isobath. 
remain (Sigler, 2000). Therefore, it is more likely that 
sablefish are outcompeting grenadier and shortraker 
and rougheye rockfish for hooks. 
Negative correlations between catch rates of two spe- 
cies could be evidence of gear saturation, or could be 
caused by other factors such as differing habitat pref- 
erences or direct competition between species. If the 
cause of negative correlations is competition for hooks, 
catch rate trends or depth distribution shifts could 
be confounded. Additionally, it may not be possible to 
evaluate the effects of factors such as habitat and other 
environmental variables until competition effects are 
taken into account. For this study, we examined the cor- 
relations between sablefish and grenadier and between 
sablefish and shortraker and rougheye rockfish to iden- 
tify any negative relationships on the longline surveys. 
We also compared correlations of catch rates from the 
longline surveys to correlations of catch rates from the 
AFSC bottom trawl surveys in the Gulf of Alaska to 
investigate the cause of the relationships on the longline 
surveys (Britt and Martin, 2001). Because trawl survey 
catches are not generally susceptible to gear saturation, 
catches are a reflection of natural fish densities and 
therefore a good control to compare to longline catch 
rates (Gunderson, 1993). Although feeding history, wa- 
ter temperature, and other variables could affect the 
strength of competition between sablefish and grenadier 
and between sablefish and shortraker and rougheye 
rockfish, we were unable to control for these variables 
in this analysis. 
Materials and methods 
Sampling 
AFSC longline surveys The continental slope throughout 
the eastern Bering Sea, the eastern Aleutian Islands, 
and the Gulf of Alaska have been systematically sampled 
each summer since 1979 during the annual AFSC long- 
line surveys. The survey is divided among six sablefish 
management areas: 1) Bering Sea; 2) Aleutian Islands; 
3) Western Gulf of Alaska; 4) Central Gulf of Alaska; 
5) West Yakutat; and 6) East Yakutat and Southeast 
(Alaska). Stations are placed 30-50 km apart and at 
each station depths from 150 to 1000 m are sampled 
(Fig. 1). Catches are pooled by management area and an 
abundance index is computed (Sigler, 2000). The gear on 
the survey closely resembles gear used by the sablefish 
fishery in Alaska. The basic unit of gear is a skate; each 
skate consists of 45 hooks, baited with squid, spaced 
2 m apart. At the end of each skate a 3-kg lead ball is 
