Laidig et al.; Relationship between abundance of juvenile Sebastes spp. and environmental variables 



41 



We related three oceanographic data sets (sea level 

 anomaly, nearshore temperature, and offshore Ekman 

 transport) to abundance data. Sea level anomaly data 

 were collected from a shore station (with an acoustic 

 gauge measuring in mml at Arena Cove, CA (38°54'N), 

 and monthly means were obtained from the University 

 of Hawaii Sea Level Center. These data represent a 

 measure of change in sea level height over time and 

 were adjusted for local atmospheric conditions. Sea level 

 anomalies reflect water movement — a positive anomaly 

 being associated with poleward flow and a negative 

 anomaly associated with equatorward flow. Nearshore 

 temperature data were continuously collected at the 

 Dark Gulch site throughout the study. Temperature 

 monitors were located in two areas of the kelp bed, one 

 at 8 m and the other at 15 m water depth. Tempera- 

 ture was recorded hourly and these data were averaged 

 daily, monthly, and seasonally. The offshore Ekman 

 transport data were determined from data gathered by 

 a buoy 20 km offshore of 39°N (Albion, CA). Monthly 

 mean offshore Ekman transport was derived from geo- 

 strophic wind stress data determined from atmospheric 

 pressure values at mean sea level (at 6-hour intervals) 

 and obtained from Environmental Research Division, 

 Southwest Fisheries Science Center (NOAA''). Offshore 

 Ekman transport was used as a measure of cross-shelf 

 water transport. The offshore flow of surface water 

 causes deep, cold, nutrient-rich water to be brought to 

 the surface to replace the displaced surface waters (the 

 process of upwelling). Monthly and seasonal values were 

 determined for all oceanographic data sets and years, 

 and the seasonal values represented the mean from 

 January to June. The months of July to December were 

 not considered in our analysis because this time period 

 was after the time of juvenile rockfish recruitment to 

 the kelp bed. 



The comparison of the abundance index for each spe- 

 cies and the oceanographic variables involved numer- 

 ous statistical tests. Cross-correlation coefficients were 

 computed in pairwise comparisons of the index and the 

 oceanographic data. For each year in the time series, we 

 used both seasonal and monthly means of the oceano- 

 graphic variables and the annual log-transformed index. 

 We also applied principal components analysis (PCA) 

 to evaluate the relationships among monthly oceano- 

 graphic variables, and also the relationships among 

 the log-transformed index of the three species. We used 

 canonical correlation analysis (CCA) to explicitly ex- 

 amine the correlations between these oceanographic 

 and rockfish PCAs. PCA develops an ordination space 

 constrained to be a linear combination of the variables 

 (i.e., the eigenvectors). CCA does the same, except the 

 ordination space is constrained to be a linear combina- 

 tion of the other set of variables. The fish data, there- 



fore, are displayed in a space defined by a combination 

 of the oceanographic data (the canonical scores). 



Year-specific catch-at-age data for adult yellowtail 

 rockfish were compared with the juvenile yellowtail 

 rockfish abundance index to evaluate the relationship 

 between juvenile rockfish abundance and subsequent 

 landings of adults in the fishery. No age data were 

 available for the adult blue or black rockfishes. Age 

 and catch data of adult yellowtail rockfish commercially 

 caught in trawls from 1997 to 1999 were retrieved from 

 the CALCOM database^ for the three closest ports to 

 our study sites (Bodega Bay, Fort Bragg, and Eureka, 

 CA). We used expanded age compositions by year class, 

 which were determined from a subset of aged fish for 

 that region. For example, in 1997 there were 10 year- 

 classes in a subset of fish that were aged from Bodega 

 Bay, whereas only six year classes existed in the subset 

 of aged fish for 1998 from Bodega Bay. Year-specific 

 catch-at-age data were determined for each year and 

 port and regressed against the log-transformed index. 

 Significance levels were determined for each port-year 

 combination and for all ports combined. 



Results 



Over 21 years, 172 dives and 3333 one-minute surveys 

 were completed, averaging 8.2 dives/year and 158.7 

 one-minute surveys/year (Table 1). The average number 

 of one-minute surveys per dive was 19.4 surveys/dive 

 (calculated by dividing the total number of one-minute 

 surveys by the total number of dives). Total number of 

 dives per year varied from 4 in 1993 to 14 in 1983, and 

 the number of one-minute surveys varied from 80 in 

 1993 to 330 in 1983. The total number of dives at Dark 

 Gulch was 113, whereas only 59 dives were conducted 

 at Salmon Point. Similarly, nearly twice as many one- 

 minute surveys were completed at Dark Gulch (2165) 

 compared to Salmon Point (1168). This difference was 

 mainly due to sea conditions; Salmon Point sometimes 

 experiences larger ocean swells (because it is less pro- 

 tected), making sampling unsafe. 



The abundance index for each species varied among 

 years but was not significantly different (t-test; n=21; 

 P>0.05) between the two study sites. For this reason, 

 the data from both sites were pooled. The index was 

 highly variable for all species (Fig. 2). For blue rock- 

 fish, the index varied from 181 fish/min in 1987 to 0.26 

 fish/min in 1992. For yellowtail rockfish, the index 

 ranged from 162 fish/min in 1985 to 0.03 fish/min in 

 1994. Abundance of black rockfish peaked in 1999 at 22 

 fish/min and was lowest in 1998 at 0.01 fish/min. 



■• NOAA (National Oceanic and Atmospheric Administration). 

 Environmental Research Division. Southwest Fisheries 

 Science Center. NOAA/NMFS/SWFSC. 1352 Lighthouse 

 Ave., Pacific Grove, CA 93950-2097. Website: http://www. 

 pfeg.noaa.gov (accessed 15 March 2006). 



^ CALCOM (California Cooperative Survey). Commercial 

 landings sampling program maintained by California Depart- 

 ment of Fish and Game, 350 Harbor Blvd., Belmont, CA 

 94002; Pacific States Marine Fisheries Commission, 350 

 Harbor Blvd., Belmont, CA 94002; and Fisheries Ecology 

 Division, SWFSC, NMFS, NOAA, 110 Shaffer Rd., Santa 

 Cruz. CA 95060. Website: 128.114.3.187 (accessed on 15 

 March 2006). 



