Krieger and Ito: Distribution and abundance of Sebastes borealis and 5. aleutianus 



265 



the net is towed. The samphng efficiency 

 could be less than 100% if fish are distrib- 

 uted above the headrope, protected by 

 structures such as boulders, swim out of 

 the path of the net, or escape under the 

 net. Conversely, if fish are herded into the 

 trawl by the bridles and doors and do not 

 escape above or under the net, the sam- 

 pling efficiency may be greater than 100%. 

 Typically, smooth (trawlable) substrates 

 are sampled during trawl surveys. Catch 

 rates fi-om trawlable substrates are then 

 applied to all substrates for estimating bio- 

 mass. Bottorn-trawl surveys may not pro- 

 vide reliable biomass estimates of short- 

 raker and rougheye rockfish because 1 ) the 

 sampling efficiency may not be 100% for 

 the distance between the wingtips of the 

 net, 2) these species may use untrawlable 

 substrates at a different rate than they do 

 trawlable substrates, and 3) the sampling 

 fi-equency may not be sufficient, depend- 

 ing on the distribution patterns of the tar- 

 get species. 



Minimal information is available on 

 the distribution of shortraker and 

 rougheye rockfish. Fishermen report 

 that rockfish school above bottom in 

 steep-slope areas. From a manned sub- 

 mersible Krieger (1992) observed 20 

 shortraker rockfish on the continental 

 shelf; these fish were in contact with the 

 seafloor and were distributed as solitary individuals 

 on shallow-sloped, smooth habitat. Shortraker rock- 

 fish were observed only at sites where boulders were 

 common, and six of the fish were found next to boul- 

 ders 0.5-1.5 m in diameter (Krieger, 1992). Catches 

 of shortraker and rougheye rockfish during longline 

 surveys indicate they are most abundant on the up- 

 per continental slope at 300-400 m depths (Sigler 

 and Zenger, 1994), but most of this substrate is con- 

 sidered untrawlable and is seldom sampled during 

 bottom-trawl surveys. For example, only eight trawl 

 hauls were completed along the 500-km continental 

 slope in southeastern Alaska during the last four 

 bottom-trawl surveys ( 1987, 1990. 1993, 1996). 



We need to understand the distribution and habi- 

 tats of shortraker and rougheye rockfish to assess 

 them effectively with bottom trawls or other sam- 

 pling gear. In this study, a manned submersible was 

 used to observe their spatial distributions and habi- 

 tat associations. These species were also quantified 

 from the submersible for comparison with abundance 

 estimates from bottom-trawl surveys. The two spe- 

 cies were combined and are referred to as rockfish in 



137 W 



Gulf of Alaska 



Figure 1 



Submersible survey sites for shortraker and rougheye rockfish in the eastern 

 Gulf of Alaska in May 1992. 



this paper because distinguishing between them was 

 not always possible from the submersible. 



Materials and methods 



Study area 



This study was conducted in May 1992 on the upper 

 continental slope in the eastern Gulf of Alaska between 

 lat. 56=10' and 58°10'N (Fig. 1). This region has consis- 

 tently produced high catch rates of shortraker and 

 rougheye rockfish during the annual longline surveys 

 in the Gulf of Alaska (Rutecki et al., 1997). The study 

 area spanned more than 200 km to include a variety of 

 habitats. Distances separating adjacent sites ranged 

 from 0.2 to 84.2 km. Dives were conducted during day- 

 light, between 0600 and 1900 hours. 



Submersible 



The two-man submersible Delta was chartered for 

 all dives. This battery-powered submersible is 4.7 m 



