! 16 



Fishery Bulletin 94(1), 1996 



As a second stratum, we summarized the acoustic 

 information collected from 4 m above the bottom 

 down to the bottom. These data represent acoustic 

 information collected for the same area that was sur- 

 veyed by the submersible ("below sub" stratum). 



Navigation 



The support ship crew navigated using GPS instru- 

 mentation with an expected positional accuracy 

 within 100 m (Hum, 1989). At two stations, submers- 

 ible observers and pilots saw the same object on sub- 

 sequent dives. In each case, the object was within a 

 few meters of the intended path of the submersible. 

 Although these observations provided evidence that 

 the navigational precision of the support vessel was 

 better than 100 m, we did not expect the submers- 

 ible to duplicate the path of a previous dive. Simi- 

 larly, GPS variation, and wind and sea conditions 

 prevented the support ship from duplicating an 

 acoustic transect. Thus, the acoustic and submers- 

 ible transects did not cover identical segments of the 

 bottom. Instead, each survey provided three esti- 

 mates offish density in a rectangle that was approxi- 

 mately 300 m wide by 2,500 m long (Fig. 2). 



Results 



Fish species observed 



On submersible transects, observers counted a total 

 of 1,928 fish from 28 taxa. Rockfish species accounted 

 for 15 of the taxa observed and 89% of all fishes 

 counted (Table 1). Mean length of fishes observed on 

 submersible transects ranged from 19.3 to 21.0 cm 

 (Table 2). On bounce dives into large schools in the 

 water column, submersible observers saw mixtures 

 of three species: blue iSebastes mystinus), yellowtail 

 (S. flavidus), and widow (S. entomelas) rockfish. 

 Schools in the middle of the water column all con- 

 tained large (> 30 cm) adult rockfishes. 



Rockfishes observed from the submersible fell into 

 three strata with respect to location in the water col- 

 umn: there were fishes touching the bottom, fishes 

 swimming a short distance dm) off the bottom, and 

 fishes schooling well off the bottom. Species such as 

 rosethorn rockfish, S. helvomaculatus , were more 

 frequently observed on the bottom or in crevices; 

 species such as redstripe rockfish, S. proriger, were 

 more frequently observed swimming singly or in 

 small groups about 1 m off the bottom; and species 

 such as blue and yellowtail rockfish were more com- 

 monly observed in small or large schools that ex- 

 tended upward from 1 m off the bottom to past the 



44 30.5' N 



Figure 2 



Example of multiple submers- 

 ible and acoustic transects that 

 occurred at station 3 from Sep- 

 tember through October 1991, 

 off Newport, Oregon. At this 

 station, submersible and acous- 

 tic transects sampled an area 

 100 m wide by 2,500 m long. 

 Dashed lines indicate acoustic 

 transects. Solid lines represent 

 submersible transects. 



limit of visibility. Juvenile rockfish of unknown spe- 

 cies occasionally occurred in very large schools 1 m 

 or more above the bottom. 



Fish density 



Estimates of areal density generated by submersible 

 surveys ranged from 76.2 fishha" 1 to 1,101.7 fishha' 1 . 

 Estimates obtained from acoustic surveys of the en- 

 tire water column ranged from 3.4 fishha" 1 to 5,716.6 

 fish-ha 1 (Table 3). We observed no significant dif- 

 ference in mean fish density between acoustic sur- 

 veys conducted before and after submersible 

 transects (Wilcoxon paired sample test, n=8, z= 

 -0.14, P=0.89). Mean densities estimated from sub- 

 mersible and acoustic surveys exhibited a significant 

 positive correlation (Kendall Rank Correlation, n =9, 

 t=0.72, 2=2.71, P=0.007). 



