Yoklavich et al : Larval rockfishes and their physical environment off central California 



771 



During winter 1993, a lens of warm surface water 

 deepened from late January to mid-February, with 

 13.5°C water at 60 m along the transect line (Fig. 

 4B). Winds were variable, with no distinct periods of 

 reversal. Unlike 1992, however, persistent wind from 

 the northwest (favorable to upwelling) commenced 

 in early March 1993 (Fig. 5A) and resulted in cooler 

 water temperatures in the upper 80 m off Daven- 

 port by early April (Fig. 5B). This was the first evi- 

 dence of upwelling during either year of the 

 ichthyoplankton surveys, as indicated by the 10°C 

 isotherm sloping up toward the coast from 19 km off- 

 shore in the 6 April 1993 CTD profile series (Fig. 4B). 



Ichthyoplankton surveys 



Patterns of distribution and abundance The 130 



bongo-net tows made during ten ichthyoplankton 

 surveys in 1991-92 (Table 1) contained 9,389 rock- 

 fish larvae; 39 tows made during four surveys in 1993 

 included 3,492 rockfish larvae. Rockfish larvae were 

 collected during all surveys, occurring in 78.7% of 

 all tows and in 100% of the tows made at the three 

 stations 7-19 km from shore. Forty-seven percent of 

 the rockfish larvae collected in 1992 and 66% in 1993 

 were identified as eight species or species-groups 

 (Table 2). Shortbelly rockfish were numerically domi- 

 nant both years, representing 46% of the total catch 

 in 1993 and 29% in 1992. Shortbelly rockfish larvae 

 occurred from late December to early May, largely in 

 the offshore samples (13 and 19 km). Although much 

 less numerous than shortbelly rockfish, blue, 

 stripetail, and squarespot rockfish larvae also were 

 relatively abundant each year (3—6% of the total). 

 Although members of the "copper complex +" repre- 

 sented only 1-2% of the total catch and generally 

 occurred in greater numbers offshore, they were the 

 most abundant rockfish taxon identified at inshore 

 stations (1 and 3 km). 



Abundance of rockfish larvae varied among sam- 

 pling dates and locations in 1991-92, with signifi- 

 cant interaction between the two factors (ANOVA, 

 P<0.001; Fig. 6A). Larvae were relatively abundant 

 from January to April. Pairwise comparisons indi- 

 cated that the large catches in early February 1992 

 at all stations were unique among all other time pe- 

 riods and that abundances across all stations early 

 in the season (December and early January) were 

 not significantly different from those that followed 

 later (late-February, March, and April). Pairwise 

 comparison of interstation abundance indicated two 

 groups, one of stations closest to shore (1, 3, and 7 

 km) and another of the two offshore stations (13 and 

 19 km). The offshore stations had significantly 

 greater abundance through time than did those in- 



shore and, while abundance at most stations gener- 

 ally declined from a peak in early February, num- 

 bers of larvae remained relatively high through April 

 at stations farthest from shore (13 and 19 km). 



As with the 1991-92 collections, larval rockfish 

 abundance varied significantly among sampling 

 dates in 1993, with greatest abundances in Febru- 

 ary (Fig. 6B). Average abundances of larval rockfishes 

 estimated at comparable stations and times from 

 January through April were not significantly differ- 

 ent between years (mean=602 larvae/10 m 2 , SE=161 

 in 1992; mean=376 larvae/10 m 2 , SE=179 in 1993; 

 /-test, P>0.383), but distribution patterns differed. 

 In contrast to the previous year, large catches of rock- 

 fish larvae occurred only at the two offshore stations 

 in 1993. Significantly greater numbers of larvae oc- 

 curred at the three stations closest to shore (1, 3, 

 and 7 km) during January and February 1992, com- 

 pared with the same months in 1993 (Kolmogorov- 

 Smirnov tests, P<0.05). 



Size, age, and birthdate distributions Size of lar- 

 val shortbelly rockfishes collected on the same days 

 in 1992 varied significantly among the 7, 13, and 19 

 km stations (ANOVA, P<0.001). Virtually all short- 

 belly rockfish larvae collected at the 7 km station 

 were greater than 5 mm SL and significantly larger 

 than those at the two offshore stations (Table 3). 

 Shortbelly rockfish larvae were collected at the 

 7 km station only from early January to mid-March, 

 and size composition remained similar throughout 

 this period (overall mean=6.5 mm SL, SE=0.1). Size 

 composition of shortbelly rockfishes did not differ 

 significantly between the two offshore stations (over- 

 all mean=5.3 mm SL, SE=0.04), and the smallest 

 larvae (<5 mm SL) were relatively abundant in all 

 samples through early April. 



Fewer samples and larvae collected at onshore sta- 

 tions in 1993 precluded among-station comparisons, 

 although size composition of shortbelly rockfishes at 

 the two offshore stations differed significantly (e.g. 

 smaller larvae at the 13 km station compared with 

 the 19 km station). The smallest larvae (<5 mm SL) 

 were most abundant at the 13 km station in Febru- 

 ary and at the 19 km station in early April (Table 3). 



Ages of 302 larval shortbelly rockfish collected in 

 January-May 1992 and February-April 1993 were 

 1-28 d (Table 4). A least-squares linear model, Length 

 = a + b (Age), was fitted to separate sets of age-length 

 data for each month in both years. Because no sig- 

 nificant differences in growth rate were found among 

 months (ANCOVA, P>0.64 [1992]; P>0.73 [1993]), 

 age-length data were combined for each year. More- 

 over, there was no interannual difference in growth 

 rate (ANCOVA, P>0. 90). A least-squares linear model 



