Kneger Distribution and abundance of Sebastes spp 



89 



tower with a panoramic view and assisted in counting 

 fish above the observer's view and in monitoring fish 

 behavior completely around the submersible. All 

 observations were audio- and video-recorded for sub- 

 sequent analysis and verification. All dives were dur- 

 ing daylight between 0600 and 1900 h. 



Rockfish densities were derived from the number of 

 fish counted and the seafloor area searched. The sea- 

 floor area searched was the distance traveled (1.7- 

 2.2 km/dive) times estimates of the lateral distance from 

 the submersible at which rockfish were visible. Lat- 

 eral distance estimates varied between 5 and 6 m be- 

 cause of changes in water clarity; illumination was 

 provided entirely by the submersible lights and re- 

 mained constant. Estimated distances were compared 

 with true distances using three methods: (DA length 

 of pipe marked at meter intervals was laid on the 

 seafloor on two dives, (2) a hand-held sonar gun pro- 

 vided distance readouts to rock formations on six dives, 

 and (3) the submersible's sonar provided distance read- 

 outs to the seafloor during descent and ascent. Esti- 

 mates of distance were consistently within lm of true 

 distances. About 10 m was monitored above the sub- 

 mersible: 5-6 m by the observer and an additional 

 4-5 m by the pilot. 



Dive sites were classified as trawlable, marginally 

 trawlable, or untrawlable based on bottom type and 

 extent of relief. Trawlable sites contained pebble sub- 

 strate interspersed with cobble <0.5 m in diameter on 

 flat bottom; marginally trawlable sites contained pebble 

 substrate interspersed with cobble and boulders of 

 0.5-5.0 m in diameter on low-relief bottom; untrawlable 

 sites contained mainly bedrock substrate with a vari- 

 ety of rugged habitats including boulders, coral, ledges, 

 rocky outcroppings, and pinnacles on high-relief 

 bottom. 



Bottom trawling 



Bottom trawling was used in 1989 to identify fish spe- 

 cies at submersible dive sites and to derive population 

 densities from catch rates. Trawling was conducted 

 from the NOAA ship RV Townsend Cromwell, using a 

 400-mesh Eastern otter trawl equipped with 1.5x2.1 m 

 doors, each weighing 386 kg. Trawling was during day- 

 light, usually within 4h after completion of a submers- 

 ible dive. The time between dives and trawls depended 

 on ship's operations, including how long it took to pro- 

 cess the catch. The sampling strategy was to trawl at 

 6.0km/h, intersecting the four parallel submersible 

 transects at each dive site. 



Trawl catches were processed for total number and 

 weight by species. The fork lengths of rockfish were 

 measured to the nearest centimeter. Fish density esti- 



mates were derived from the number of fish captured 

 and the seafloor area swept by the net. The seafloor 

 area swept was the distance trawled (0.93-1.35 km/ 

 haul) times the horizontal opening of the net (14m, 

 based on measurements between the wing tips, 12.2- 

 14.3 m for the 400-mesh Eastern otter trawl; NMFS 

 1990). Measured vertical openings were 1.4—1.8 m. 



Data analysis 



The off-bottom distance monitored from the submers- 

 ible was about 10 m, whereas the trawl sampled to 

 about 2 m off bottom. Correlation between the percent 

 composition offish species observed from the submers- 

 ible and captured with bottom trawls was determined 

 using correlation analysis (Sokal & Rohlf 1981). Cor- 

 relations were determined for rockfish, flatfish, short- 

 spine thornyhead Sebastolobus alascanus, and wall- 

 eye pollock Theragra chalcogramma. 



Correlation analysis also was used to examine the 

 correlation between densities of rockfish observed from 

 the submersible and densities derived from trawl catch 

 rates. Ratio estimates (Cochran 1977) of observed and 

 trawl densities were then used to determine the catch 

 efficiency of bottom trawls for rockfish. For these analy- 

 ses, Pacific ocean perch, sharpchin rockfish S. 

 zacentrus, redstripe rockfish S. proriger, and harlequin 

 rockfish S. variegatus >25cm long were categorized as 

 "large"; whereas those <25cm were categorized as 

 "small." Most "large" rockfish observed from the sub- 

 mersible were identified as adult Pacific ocean perch 

 (based on symphyseal knob and body shape), whereas 

 "small" rockfish could not be consistently identified. 

 Rockfish were visually categorized as either large or 

 small from the submersible, whereas trawl-caught fish 

 were measured. Other rockfish species observed 

 from the submersible included redbanded S. babcocki, 

 rosethorn S. helvomaculatus, dusky S. ciliatus, 

 silvergray S. brevispinis, yelloweye S. ruberrimus, and 

 greenstriped S. elongatus. These solitary, demersal 

 rockfish were identified from the submersible by their 

 distinct color patterns, and categorized as "other" rock- 

 fish. 



Results and discussion 



Submersible dives 



Six submersible dives were completed in 1988 and four- 

 teen in 1989 at 188-290 m depths. Thirteen of the dive 

 sites were classified as trawlable, three as marginally 

 trawlable, and four as untrawlable. Of the 9278 fish 

 observed from the submersible, 5317 were rockfish 

 (Table 1). Rockfish were the most abundant fish on 11 



