-4-2 2 4 6 



INDEX OF TEMPERATURE ANOMALY 



Figure 3. — Relation between temperature anomaly at 

 pier of the Scripps Institution of Oceanography and 

 virtual availability of sardines in California, 1932-33 

 through 1957-58 seasons. Numerals denote the fishing 

 season, e.g., 32 indicates the season of 1932-33. 



in the efficiency of individual boats, they adjusted 

 the catch per luiit of effort using as base year the 

 1932-33 season at each port. The same authors 

 (1952) reported the data for the 1949-50 and 1950- 

 51 seasons in all California waters, as well as at 

 Monterey and San Pedro, and Clark (1956) 

 published the data for the four seasons 1951-52 

 through 1954-55 in southern California. In the 

 latter two works the authors chose the 1941-42 

 season as the new base year (Clark and Daugherty, 

 1950) to fit recent developments in the fisherj'. 



Because fishing effort should be expressed on 

 the same base year, I converted the catch per 

 unit of effort in the whole fishing season to the 

 value based on that of the 1941-42 season. Since 

 the fishing grounds off California are divided into 

 two regions of almost equal area, the north off 

 San Francisco and Monterey and the south off 

 San Pedro (Widrig, 1954), the total effective effort 

 is calculated by dividing the total California 

 catch by the sum of catches per imit effort in the 

 two regions. The deviations of catch showed a 

 significant multiple correlation coefficient of 0.594 

 with the temperature anomaly and the amount of 

 fishing effort. Partial correlation coefficients were 

 0.293 for fishing effort and 0.585 for temperature 

 anomalies. 



CATCH PER UNIT OF EFFORT 



Because fishing effort ajjiieared to show a 

 positive correlation with the deviation of catch 

 from the expected value, I repeated the above 

 calculations for total catch, using the catch per 

 unit of effort. Present evidence indicates that the 

 stock exploited in the waters off central California 

 is mainly composed of fish spawned in the waters 

 off southern California (northern subpopulation), 

 whereas the stock in southern California includes 

 fish spawned off central Baja California (southern 

 subpopulation) as well as fish from the northern 

 subpopulation (Felin, 1954; Marr, 1960). The 

 effort data are not given after the 1951-52 season 

 for central California where the fishery practically 

 disappeared by 1954. Catch per unit of effort at 

 San Francisco, Monterey, and San Pedro indicates 

 that the available stocks off the two northern 

 ports are correlated with a significant coefficient 

 of 0.810 but that the stock off San Pedro is not 

 correlated with either San Francisco (0.314) or 

 Monterey (0.398) during the 19 seasons from 

 1932-33 to 1950-51. This lack of correlation may 

 be attributed to differences in the subpopulation 

 supporting the fisheries. It is also possible that the 

 northern subpopidation, having migrated to cen- 

 tral California, left only a minor portion of the 

 stock in southern California or that major age 

 groups differed between these regions. 



Since it is desirable to forecast regional catches, 

 the following discussion is based on data by 

 region. In the northern waters the catch curves 

 for ages II to VI, based on catch per unit of 

 effort, were obtained for the 1930-44 year classes. 

 The curves were fairly regidar for most of the 

 year classes, except the last. The following table 

 shows that age and year class are significant 

 sources of variation of logarithmic catch, as was 

 true for total catch: 



•Significant at a probability of less than 0.1 percent. 



The following tabulation of the standard catch 

 curve based on logarithntic means of the catch per 

 unit of effort in central California for the year 

 classes 1930-44 shows that there is less change in 



ANALYSIS OF OATCH CURVE OF PACIFIC SARDINE 



591 



