FISHERY BULLETIN: VOL. 78, NO. 4 



similar to those of the CalCOFI program (Kramer 

 et al. 1972) except we used 0.333 mm mesh net on 

 70 cm bongos vs. 0.55 mm mesh on a CalCOFI net. 

 The difference in mesh sizes can be corrected, 

 Equation (12), but because of the reduced avoid- 

 ance associated with bongo nets, larva catches 

 may be relatively greater resulting in high census 

 estimates and accordingly high biomass esti- 

 mates. 



Spawning frequency during the 2-mo spawning 

 period in the northern subpopulation is unknown. 

 Hunter and Goldberg (1980) indicated that north- 

 ern anchovy in the central subpopulation may 

 spawn a batch of eggs every 6 or 7 d during peak 

 spawning. If northern anchovy off Oregon respond 

 differently, additional error would be introduced 

 into the biomass estimate. The higher fecundity of 

 fish in the north may be balanced by less frequent 

 spawning. 



Spawning biomass estimates derived from the 

 Smith Larva Method are based on larva abun- 

 dance. Using the larva census estimates corrected 

 for mesh size differences, Equation (12), biomass 

 estimates were obtained by Equations (10) and 

 (11). These estimates were 737,736 and 764,692 t 

 in 1975 and 979,901 and 1,005,263 t in 1976 with 

 the Sette and Ahlstrom Census and slightly less 

 with the Smith Census, 696,479 and 723,705 in 

 1975 and 894,074 and 920,001 in 1976. In this case, 

 since the Smith Census is based on procedures and 

 data (Smith 1972) from which Equations (10) and 

 (11) are derived, biomass estimates based upon it 

 are probably better than those based on the Sette 

 and Ahlstrom Census. 



Because of diffusion and dispersion and length 

 of the larval period (about 30 d compared with 2-4 

 d for eggs), larvae are more evenly distributed 

 over a given geographic area than eggs and in turn 

 yield a less variable biomass estimate than eggs 

 (Pacific Fishery Management Council footnote 2). 

 This may, in part, account for the smaller year to 

 year difference in biomass estimates based on lar- 

 vae compared with those based on eggs. Interest- 

 ingly biomass estimates based on eggs decreased 

 greatly from 1975 to 1976 while those based on 

 larvae increased. 



Most Probable Biomass 



Spawning biomass estimates (Table 3) ranged 

 from 262,506 to 769,511 t in 1975 and 144,654 to 

 1,005,263 1 in 1976 (Table 3 ). The biomass estimate 

 based on acoustic survey of adults in July 1977 was 



870 



about 800,000 t (Smith footnote 3), using methods 

 given by Smith (1970). A reasonable conclusion is 

 that the actual spawning biomass of northern an- 

 chovy in the survey area laid or fluctuated be- 

 tween the extreme values in 1975 and 1976 (Houde 

 1977) and is probably <1 million t but >100,000 t. 

 This line of reasoning is supported by the fact that 

 the estimates from 1975 and 1976 (for a given 

 method) are within twofold of each other and less 

 than tenfold (for any method) from the acoustic 

 survey. 



These estimates only include mature spawoiing 

 adult fish. Laroche and Richardson ( 1981) reported 

 that northern anchovy in the northern subpopula- 

 tion attain first sexual maturity at the end of the 

 second year, i.e., in the third summer. Thus north- 

 ern anchovy <2 yr old are not included in the 

 estimates and may represent additional sizeable 

 biomass. These immature fish are segregated geo- 

 graphically from spawning adults during the 

 spawning season with the young fish occurring in 

 nearshore coastal areas (Laroche and Richardson 

 1981). 



Based on these estimates, spawning biomass of 

 northern anchovy off the Oregon-Washington 

 coast is less than that in the central subpopulation 

 which had a mean estimate of around 3,631,200 t 

 for 1965-72 (MacCall et al. 1976), although a re- 

 cent population decline has been recorded in 1978 

 (1,183,771 t) and 1979 (1,564,139 t) (Stauffer 

 1980). My spawning biomass estimates are more 

 comparable to that for the southern subpopulation 

 of 544,680 t (mean for 1965-69) (MacCall et al. 

 1976). 



YIELD ESTIMATES 



Using Gulland's (1971) formula for potential 

 yield to a fishery. Equation (14), with the range of 

 instantaneous natural mortality rates of 1.0-1.05, 

 estimates for a biomass of 144,654 t are 86,792- 

 91,132 t. For a biomass of 1,005,263 t they are 

 603,158-633,316 t. These potential yield estimates 

 are about 60% of the spawning biomass, which 

 may be dangerously high values for a species 

 known to undergo large year-to-year fluctuations 

 in abundance. 



Recommendations for harvest quotas by the 

 northern anchovy management plan for the cen- 

 tral subpopulation called for a more conservative 

 yield estimate of 70% of one-third, or about 23% , of 

 the spawning biomass in excess of 907,800 t 

 (Pacific Fishery Management Council footnote 2). 



