Starr et al.: Submersible-survey and acoustic-survey estimates of fish density 



1 19 



Percent of Targets 



ID 15 20 2? 30 



Station I 



Max Depth = 60 m 



(I 



10 



20 



£ 30 



S 40 



D. 



3 50 H 



60 

 70 



Figure 4 



Vertical distribution of fishes insonified at sta- 

 tions 1-3 during acoustic surveys conducted from 

 September through October 1991, off Newport, 

 Oregon. 



Discussion 



Comparison of submersible and acoustic 

 surveys 



On the basis of our previous submersible surveys of 

 rocky banks off Oregon and anecdotal information 

 from sport and commercial fishers, we expected to 

 observe demersal rockfishes from the submersible 

 and to insonify large schools of rockfishes in the wa- 

 ter column well above the rocky substrate of Stone- 

 wall Bank. Data collected from submersible transects 

 and bounce dives did indicate that rockfishes were 

 the predominate fish taxa inhabiting Stonewall Bank 

 at the time of the surveys. However, most schools 



Table 4 



Comparison of backscattering cross section (a) from dual- 

 beam estimates (DB a) and from Love's equation (Sub a). 

 Target strength (TS ) was converted to backscattering cross 

 section (cr ) by using the equation TS = 10 Log a. 



Station 



DB a 



Subcr 



DB a/Sub a 



iinoin -; 

 0.000214 

 0.000135 



0.000116 

 0.000045 

 0.000138 



1.06(0.65)' 

 4.76 (2.04)' 

 0.98(0.56)' 



' Ratio of DB a/Sub <j using mean lengths of nonschooling fishes 

 only I see text for details). 



appeared to be associated with bottom features and 

 were not isolated in the middle of the water column. 

 Most large schools of fishes extended from the bot- 

 tom upward for as much as 20 m into the water col- 

 umn, and submersible observers were able to view 

 the lower portions of these schools. 



The extension of rockfish schools from the bottom 

 upwards into the water column has been reported 

 from other submersible surveys. Krieger (1993) vi- 

 sually surveyed an area from to 10 m above the 

 bottom and noted that Pacific ocean perch, Sebastes 

 alutus, ranged from to 7 m above the bottom. He 

 observed that small groups of rockfishes were close 

 to the bottom, whereas larger groups were higher off 

 the bottom. Krieger observed no rockfish schools higher 

 in the water column than 7 m above the bottom. Re- 

 searchers in Alaska (O'Connell and Carlile, 1993), Brit- 

 ish Columbia (Murie et al., 1994), Oregon (Pearcy et 

 al., 1989; Stein et al., 1992), and California (Yoklavich 

 et al., 1995) also observed schools of semipelagic rock- 

 fishes above high-relief bottom features. 



"Plumes" of rockfish rising above pinnacles are 

 readily apparent on echograms from our acoustic 

 surveys (Fig. 5). Wilkins (1986) described this char- 

 acteristic shape for widow rockfish schools as tall, 

 narrow columns rising over an irregular bottom. 

 Richards et al. (1991) developed techniques to esti- 

 mate species composition of rockfishes in the plume 

 from the patterns of acoustic signals. They suggested 

 that computer processing of acoustic signals could 

 lead to a remotely operated technique for species 

 identification of schooling rockfishes. Acoustic iden- 

 tification of rockfish species would improve acoustic 

 surveys and seems possible, given the characteristic 

 acoustic shape of rockfish schools. However, single 

 species aggregations were atypical in our study. Spe- 

 cies discrimination based on acoustic pattern recog- 

 nition would have been difficult, although we were 

 able acoustically to identify aggregations of mixed 



