A PILOT-STUDY OF THE DISTRIBUTION AND ABUNDANCES OF ROCKFISHES IN RELATION TO NATURAL ENVIRONMENTAL-FACTORS AND AN OFFSHORE OIL AND GAS-PRODUCTION PLATFORM OFF THE COAST OF SOUTHERN CALIFORNIA

Platform Hidalgo is located in 123 m of water, approximately 10 km SW of Pt. Arguello, California. A pilot study was conducted to examine th e distribution and abundance of rockfishes around Hidalgo (by scuba an d by ROV photosurveys) and at eight adjacent natural reefs (by ROV pho tosurveys). To test for differences in fish assemblages among reefs us ing factorial ANOVA, the natural reefs were classified according to tw o levels each of three factors: ''low'' (0.2-0.5 m) versus ''high'' (> 1 m) relief height; ''nearfield'' (<3 km from Hidalgo) versus ''farfie ld'' (>3 km from Hidalgo) proximity to the platform; and ''shallow'' ( 113-160 m) versus ''deep'' (195-213 m) reef depth. Fishes were also ta gged over natural reefs with breakaway hook tags. Surveys and tagging studies were conducted from July to October 1990. During surveys in Ju ly and August, large numbers of young-of-the year (YOY) rockfishes (pr imarily Sebastes entomelas, S. flavidus and S. hopkinsi) were found in the surface and midwaters around Hidalgo. At the base of the platform , 1+ yr rockfishes (mainly S. entomelas and S. flavidus) were very abu ndant. When the platform was resurveyed in October, following the firs t gale of the year, nearly all the YOY rockfishes had disappeared, tho ugh the If yr fishes were still abundant at the platform's base. While the ultimate fate of the YOY rockfishes is not known, our preliminary data suggest the hypothesis that Platform Hidalgo acts as a producer of fish biomass by providing recruitment habitat for pelagic larvae to settle and grow before dispersing as small juveniles. Both multivaria te cluster analysis and univariate ANOVA show that the rockfish assemb lage around Hidalgo is different in species composition and abundance from those over the natural reefs, although several species co-occur a t the platform and the reefs. Three-way factorial ANOVA (without repli cation) suggests that among the natural reefs there are no statistical ly significant differences (at P less than or equal to 0.05) in abunda nces of rockfish populations or in diversity of rockfish assemblages d ue to the effect of relief height, depth, or proximity to the platform . However, pattern-recognition techniques (cluster analysis) reveal di stinct groupings of these fauna that are related foremost to reef dept h, followed by proximity to the platform and finally relief height. Fi sh density differed significantly (P less than or equal to 0.05) betwe en sampling times at selected sites. We propose that the factor relief height be dropped in subsequent definitive studies of these offshore reef assemblages. By doing so, the effects of remaining factors, inclu ding possible platform effects, on the distribution and abundances of these fishes can be tested with a more powerful replicated factorial d esign. Sampling also should be conducted on a number of occasions to e xamine patterns and sources of natural temporal variability.