EBELING ET AL.: ANNUAL VARIABILITY OF REEF FISH 



km west of Santa Barbara, Calif, (lat. 34°25' N, 

 long. 119°57' W). Measuring 275 x 80 m (2.2 ha), 

 the reef surface averaged 8-12 m in depth, though 

 some crests projected to within 5 m of the surface. 

 The reef was surrounded by flat sand or cobble 

 bottom, 16-20 m deep, with smaller rock outcrops. 

 Above the reef, the kelp canopy usually prolifer- 

 ated during spring and summer, but thinned dur- 

 ing late fall and winter. 



Island observations were made at a site centered 

 about Fry's Harbor on the north side of Santa Cruz 

 Island (Figure 1). The subtidal substrates here 

 were mostly rocky, with boulder areas, ledges, and 

 caves interspersed occasionally with sand or flat- 

 faced rock. The bottom sloped rather steeply to 

 sand at depths of 15-25 m about 20-50 m from 

 shore. Most sampling was conducted at depths of 

 3-15 m. Here, the kelp canopy extended a short 

 distance seaward over greater depths and shore- 

 ward to meet steep rock cliffs. In contrast with the 

 mainland observations, therefore, most island ob- 

 servations were made within about 10-50 m of the 

 shore, over an area of rapidly increasing bottom 

 depth. 



We saw anglers and divers at both sites. Yet, we 

 rarely observed concentrated fishing effort, prob- 

 ably because of the erratic state of the Santa Bar- 

 bara partyboat industry during the early 1970's 

 (Love and Ebeling 1978). Fishermen in small 

 boats were more frequently seen casting bait and 

 lures near the surface at Naples Reef than at the 

 island site. Hence, catches of kelp bass and other 

 surface predators were probably substantial at 

 Naples Reef only. Sport divers exploited both sites, 

 albeit more sporadically than boat fishermen and 

 seldom during the sampling periods. We suspect 

 that catches of bottom fishes were not large and 

 about the same at both sites. 



We noted no kelp cutting and harvesting in the 

 area of either site. About the island site, kelp beds 

 are limited to a narrow band along the steep shore, 

 and so are inaccessible and too small for harvest. 

 Kelp in the mainland area is harvested only in- 

 shore of Naples Reef, which is left undisturbed. 



Cinetransects 



We sampled fish and observed habitat charac- 



*Ron H. McPeak, Senior Marine Biologist, Kelco Corp., 2145 

 East Belt St., San Diego, CA 921 13, pers. commun. October 1979. 



'Bruce W. W. Harger, General Manager, Neushul Mariculture 

 Corp., 275 Orange St., Goleta, CA 93017, pers. commun. October 

 1979. 



teristics by means of "cinetransects." These were 

 2.5-min, Super-8 mm movie films taken at 24 

 frames/s by scuba divers. The use of 50 ft ( 15.24 m) 

 film cartridges standardized sampling time, and 

 allowed rapid changing of film. High-speed color 

 film yielded good photographs when water visibil- 

 ity exceeded 3 m. Starting from opportune points 

 within the kelp-rock habitat, divers swam at rela- 

 tively constant speeds, and, with the camera held 

 level or pointed slightly downward (for bottom 

 transects), steadily panned in about 10° arcs. 

 Large aggregations of fish were photographed in 

 one sweep of the camera; thereafter the camera 

 was not pointed at the aggregation. This proce- 

 dure allowed rapid and accurate enumeration of 

 aggregations, and avoided redundant sampling of 

 fish. During a given transect, a diver would keep to 

 the same general depth and terrain, so that each 

 transect could be classified by its habitat charac- 

 teristics. For each transect, he measured depth of 

 filming, depth of bottom, underwater visibility, 

 temperature, and depth of thermocline. Films 

 were taken in two general habitats: bottom and 

 kelp canopy. Canopy transects were made at 

 depths of 2-3 m, just below the mat of floating 

 fronds. Bottom transects were taken from about a 

 meter above the bottom, at depths ranging from 

 about 3 to 15 m. All cinetransects were photo- 

 graphed during September of the years 1971-74. 

 This was in the midst of the season of maximum 

 thermal stratification, when water was predicta- 

 bly calm and clear (Brown 1974; Love and Ebeling 

 1978). 



Two observers counted and tallied individuals 

 per species from cinetransects projected at low 

 speed and stop action. For each film, observers also 

 scored bottom relief and algal density from 1 (low) 

 to 5 (high). They often stopped, reversed, and 

 reran the film to accurately count fish in dense 

 clusters. When observers disagreed, they re- 

 counted, and recorded the average of the two 

 closest values. All species but two were tallied 

 separately. The sibling rockfishes Sebastes car- 

 natus andS. chrysomelas, which were identifiable 

 by color only, were tallied as one, because subtle 

 color differences were not always discernible in 

 cinetransects filmed at greater depths or in more 

 turbid water. Observers did not count small 

 (young-of-year) juveniles, such as the reddish 

 growth stage of blue rockfish,S. mystinus. Nor did 



'"Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



363 



