Stem et al Fish-habitat associations at edge of Oregon continental shelf 



541 



44° 10' N 



44 N 



At each station we made three 

 dayhght dives, each by a differ- 

 ent observer (DS, MH, WB). 

 Dives began and ended at least 

 an hour after dawn and an hour 

 before sunset, respectively, min- 

 imizing the possible effects of 

 diurnal migration by fishes. Al- 

 most all dives at each station 

 were made on the same day. 



Our methods basically follow 

 those developed for use by scuba 

 divers working on shallow reefs 

 (Brock 1954, Ebeling 1982). Each 

 observer made two 30-minute 

 visual belt transects during each 

 dive, yielding 6 transects per sta- 

 tion (i.e., a total of 36 transects, 

 12 by each observer). To deter- 

 mine if there were any discern- 

 ible effects from lights or motor 

 noise of the submersible on the 

 fishes, a 10-minute rest was 

 taken with all lights and machin- 

 ery off between each pair of tran- 

 sects. To minimize variability 

 caused by within-transect sub- 

 strate changes, all transects 

 within a station started as close- 

 ly as possible at the same posi- 

 tion, as determined by Loran C. 

 However, due to limits in the ac- 

 curacy of Loran C and variabil- 

 ity in current speed and direc- 

 tion, transects within stations 

 were usually 100-300 m apart. 



The observer in the submer- 

 sible viewed the bottom through 

 a single bow port which limited 

 observation to about a 90° view. 

 Submersible altitude above bot- 

 tom (at height of observers' eyes 

 from the bottom) was held as 

 closely as possible to 2 m, as mea- 

 sured by an altimeter on the vehicle and by a chain 

 suspended from the submersible (see below). Widths 

 of the viewing path at altitudes of 0.5-2.0 m were deter- 

 mined empirically by "flying" the vehicle at right 

 angles across a decimeter-striped 3 m pole placed on 

 the bottom and noting the length of the pole visible to 

 the observer between two fixed points on the submer- 

 sible. At 2m altitude, the transect width was 2.3m. 

 Thus, the density of fishes (no./m^) was calculated as 

 the number of fishes seen along a transect divided by 

 2.3 times the transect length in meters. To aid in 



124 50' W 



124° 40' W 



Bathymetric chart 

 sampled. 



Figure 1 



of Heceta Bank, Oregon, indicating locations of the six stations 



estimating fish length and maintaining vehicle altitude, 

 an ~0.4m long fiberglass rod, striped in alternating 

 black and white decimeters, was hung by chain from 

 the vehicle within the observer's view. Chain length 

 was adjusted so that when the rod was just above the 

 bottom, the observer's altitude was 2m. 



The goal of the observer during a dive was to iden- 

 tify, count, and estimate the lengths (to the nearest 

 decimeter) of all fishes seen along the transect. Fishes 

 were categorized into "schooling" when five or more 

 individuals formed a polarized group (i.e., all fish 



