230 



Fishery Bulletin 88(2), 1990 



1987) were the only rockfishes seen and only during 

 surveys conducted in late July and August (Fig. 6). 

 These observations of rockfishes coincided with the 

 peak growth of eelgrass up to 1.5 m high; in the fall 

 and winter the height of eelgrass beds was reduced to 

 <0.5 m. No quillback rockfish were ever observed on 

 sand/eelgrass. 



Nested analysis of variance 



Habitat type comparison When density differences 

 were detected, the highest densities were observed on 

 the high-relief rocky and artificial reefs (i-'<0.05 nested 

 ANOVA, Table 1). Highest densities of large copper 

 rockfish were observed on high-relief rocky reefs dur- 

 ing both the summer of 1987 and 1988, whereas ar- 

 tificial reefs had the highest densities in the fall and 

 winter. Similar low densities of 80-200 mm copper 

 rockfish were observed on all four habitat types all 

 seasons except fall 1987. High-relief rocky reefs had 

 the highest densities of large quillback rockfish all 

 seasons; sand/eelgrass and low-relief rocky reefs were 

 similar as large quillback rockfish were rarely observed. 

 Artificial reefs had the highest densities of 80-200 mm 

 quillback rockfish all seasons except winter and sum- 

 mer 1988. No density differences were observed for 

 both size groups of brown rockfish on any of the habitat 

 types. Similarly, YOY rockfishes were observed on all 

 four habitat types, and no differences were detected 

 in densities of YOY rockfish among the four habitats. 



Replicate site comparison For many (23/42 groups) 

 size groups of copper, quillback, brown, and YOY rock- 

 fishes, there were significant differences in densities 

 between the two replicate sites for most seasons 

 (P<0.05 nested ANOVA, Table 1). The major excep- 

 tion was that similar densities of large copper rockfish 

 were observed on replicate sites all seasons except 

 spring 1987. 



Year-to-year comparison 



For most size categories of copper, quillback, and 

 brown rockfishes, there were no significant difference 

 between their densities for the two periods tested: 

 Summer 1987-summer 1988, and winter 1987-winter 

 1988 (Table 2). For the summer comparisons, the prin- 

 cipal differences in densities were for YOY rockfish on 

 all sites where they were observed: HR2, LRl, LR2, 

 AR2, and SE2. YOY were observed at all five reefs 

 in the summer of 1987; however, no YOY were ob- 

 served in the summer of 1988, except at AR2. Other 

 differences between summer 1987-1988 densities were 

 80-200 mm quillback rockfish on HR2; densities of 



Sand/eelgrass 

 Copper rockfish 



D86 F87 A87 J87 Ae7 087 D87 F8B A88 J8B 



Survey month CZ) SE 1 ^B 

 Brown rockfish 



)200 mm 



ABB 

 SE 2 



_ 2 



D86 Fa7 AS? J87 AB7 087 DB? FB8 ABB J8B ASS 

 Survey month CU SE 1  SE 2 



YOY rockfish (80 mm 



086 ra? AB7 J87 A87 087 087 F88 A86 J88 A8B 088 



Survey month ^B SE 2 



Figure 6 



Mean monthly densities (-1-95% Cl)of y(iung-of-the-year(<80 mm), 

 suhadult (80-20(1 mm), and adult (>200 mm) rockfishes on the two 

 sand/eelgrass areas in central Puget Sound, December 1986-AugT.ist 

 1988; YOY surveys December 1986-()clober 1988. No quillback 

 rockfish (80-200 mm or >200 mm), subadult copper rockfish, or 

 subadult brown rockfish were observed. 



small quillback rockfish were higher in the summer of 

 1988 after the YOY recruitment of 1987. 



Similar to the summer 1987-summer 1988 compari- 

 sons, there were few significant density differences for 

 winter 1987-winter 1988 (Table 2). The principal dif- 

 ferences were that large copper rockfish and small 



