Matthews: Habitat use by rockfishes in central Puget Sound 



225 



Figure 2 



Nested experimental design showing habitat types in central Puget 

 Sound, sites within habitat types, and the three transects per site. 



The two low-relief rocky reefs were characterized by 

 flat, featureless cobble and rock bottom with a few 

 isolated areas of vertical relief (1-2 m). Transects ex- 

 tended from 8-10 m in depth, but not deeper than 10 

 m where the reef drops off into sand/cobble substrate. 

 The low-relief reefs were more uniform in depth when 

 compared with the high-relief reefs. Both low-relief 

 reefs have dense canopies of Nereocystis leutkeana May 

 through November and understories of perennial kelps 

 Agarum fimhriatum and Pterygophora californica. 

 Low-relief habitats underwent dramatic changes in the 

 fall and winter when they were virtually devoid of all 

 fish and algal structure. Bainbridge Island low-relief 

 rocky reef (LRl) is 500 m inshore from HRl, and the 

 two reefs are separated by cobble and sand. Blake 

 Island low-relief reef (LR2) is located 5 km south of 

 Bainbridge Island. Both low-relifef reefs are swept by 

 high currents up to 8.1 km/hour (4.5 knots). 



The two high-relief artificial reefs were character- 

 ized by vertical relief of up to 5 m and no surface 

 canopies of bull kelp, but with isolated patches of under- 

 story kelps. Transects extended from 15-20 m in depth. 

 The Blake Island artificial reef (ARl) was constructed 

 in 1980 (Lauile 1982) and is located 5 km south of Bain- 

 bridge Island on the southwestern side of Blake Island 

 approximately 500 m offshore of LRl. The reef con- 

 sists of concrete rubble, slabs, rectangular boxes, and 

 tires forming vertical relief up to 4 m. The Boeing- 

 Creek artificial reef (AR2) was constructed in late 1982 

 and is located 8 km north of Shilshole marina approx- 

 imately 600 m offshore. The reef consists of concrete 



rubble with vertical relief of up to 6 m. The two reefs 

 differ, as Blake Island is swept by strong currents up 

 to 8.1 km/hour (4.5 knots) while Boeing Creek ex- 

 periences little current. 



The sand/eelgrass areas were characterized by flat, 

 shallow, unconsolidated substrate with dense growth 

 of eelgrass Zostera marina May through November. 

 Eelgrass typically grows in low-current, sheltered 

 water and is restricted to depths less than 6 m (Phillips 

 1984), so the transects were placed at 5 m. The Redon- 

 do (SEl) sand/eelgrass area is located approximately 

 500 m north of Saltwater State Park, 24 km south of 

 Seattle. The Port Madison eelgrass area (SE2) is 

 located 500 m north of the Suquamish boat ramp, off 

 the northern side of Bainbridge Island. 



Survey methods 



From December 1986 through February 1988, I con- 

 ducted monthly SCUBA surveys and estimated 

 rockfish densities along three (30-m long, 3 m wide, and 

 1 m high) permanent transect lines at each of the eight 

 reefs (Fig. 2). After the February 1988 survey, surveys 

 were conducted in April, June, July, and August of 

 1988. In October 1988, surveys quantifying only YOY 

 rockfishes were conducted on four reefs where YOY 

 had been observed in the summer and fall of 1987: 

 HR2, LRl, AR2, and SE2. All eight sites were sam- 

 pled each month, and all three transects were sampled 

 on the same day. I swam along the transect line and 

 recorded individuals of each species as YOY (<80 mm 

 TL). subadult (80-200 mm TL), and adult (>200 mm 

 TL). During the summer and early fall when kelp cover 

 was the highest, I also swam through the canopy to 

 search for YOY rockfish. The ability to designate these 

 size categories was verified by periodically capturing 

 and measuring fish. Because YOY copper, quillback, 

 and brown rockfishes could not be identified to species 

 underwater, they were combined into one group. A 

 total of 456 transects was completed on all reefs: 24 

 transects conducted monthly for 15 consecutive months 

 December 1986-February 1988, and 24 transects con- 

 ducted monthly during April, June, July, and August 

 1988. Fifteen additional YOY surveys were completed 

 in October 1988. 



Other information collected along the transect lines 

 included estimates of vegetative cover and water tem- 

 perature. Kelp and eelgrass cover was qualitatively 

 assessed by estimating, at the end of each transect, the 

 percent cover of bullkelp, understory kelps, and the 

 height of eelgrass present. The percent cover ranged 

 from (no vegetation present) to 100% (transect line 

 completely covered by vegetation). Temperatures were 

 recorded along the transect lines on each survey with 

 a submersible thermometer. 



