EGG PRODUCTION OF THE CENTRAL STOCK OF 

 NORTHERN ANCHOVY, ENGRAULIS MORDAX, 1951-82 



Nancy C. H. Lo' 



ABSTRACT 



A model was developed for estimating daily production of eggs of northern anchovy from counts of the 

 total numbers of eggs and size-frequency distribution of larvae. Estimates of egg production using this 

 model were compared with three estimates based on the mortality rates of staged (aged) eggs. The 

 model was used to calculate daily egg production of anchovy for a 24-year time series (1951-82) (data 

 were collected each year from 1951 to 1966 and 1979 to present and every 3 years from 1966 to 1979). 

 Comparisons of this index of stock abundance with ones based on the standing stock of larvae indicate 

 that the present model is a better index of spawning biomass. It was found from the 1979-81 data that 

 the eggs and larvae (< 20 days) have different forms of instantaneous mortality rate (IMR): The larval 

 IMR was age dependent, i.e., zit) = pit for tc < t whereas the egg IMR was constant 2(n = a for t < tc 

 where tc is incubation time or yolk-sac absorption. Based upon this model, the daily egg production, 

 and egg-larval mortality rates for larvae <20 days (<8 mm preserved length), were estimated for 

 1951-82 from data collected with 1 m ring nets and bongo nets. Egg production varies with stock size 

 proportionally if the reproduction effort remains constant. The egg production is a better index of stock 

 size than the larval abundance because the latter is subject to the inherent egg and larval mortality in 

 addition to reproductive output. 



Ichthyoplankton data have been used extensively 

 for estimating biomass (or spawning biomass) of 

 marine fish stocks (Murphy 1966; Ahlstrom 1968; 

 Smith 1972). One of the tacit assumptions under- 

 lying most of the methods used for estimating 

 biomass from ichthyoplankton data is that egg or 

 larval mortality is constant among years. In 

 recent years, however, it has become increasingly 

 evident that egg and larval mortality is quite 

 variable among years and among life stages (Ahl- 

 strom 1954; Marr 1956; Colton 1959; Burd and 

 Parnell 1972; Gushing 1973; Fager 1973; Harding 

 and Talbot 1973). As a result, biomass indices 

 based on standing stock of eggs or larvae are 

 subject to a considerable bias if the interannual 

 variability in mortality is not taken into account. 

 In order to eliminate the bias, attempts were made 

 to estimate the spawning biomass by using the egg 

 production and reproduction parameters (Saville 

 1964; Beverton and Holt 1965; Ciechomski and 

 Capezzani 1973). The basic model is 



Po = BaRiE/W) 



where Po = egg production at age zero, 

 Ba = spawning biomass, 



(1) 



'Southwest Fisheries Center La Jolla Laboratory, National 

 Marine Fisheries Service, NOAA, PO. Box 271, La Jolla, CA 

 92038. 



R = 



E = 

 W = 



proportion of spawning biomass be- 

 ing female, 



average batch fecundity, 

 average mature female weight. 



Equation (1) is adequate for species that spawn 

 only once during a season. But for the multiple 

 spawners, like northern anchovy, Engraulis mor- 

 dax, one needs to include another adult parameter, 

 the proportion of mature spawning female (F), in 

 the equation (Parker 1980). Moreover, Parker 

 chose to use egg production per day, as this could 

 be easily estimated from a single cruise. Thus, the 

 egg production model (EPM) for northern anchovy 

 (or any multiple spawning stock) becomes 



Po = BaRF{E/W). 



(2) 



Manuscript accepted April 1984. 



FISHERY BULLETIN: VOL. 83, NO. 2, 1985. 



Staged eggs are used to estimate the daily egg 

 production (number of eggs per day) of the popula- 

 tion (Po) while adult fish are sampled to estimate 

 the number of eggs produced per fish weight 

 (E/W), sex ratio (i?), and proportion of mature 

 spawning female iF). This method is, without 

 doubt, the best of all ichthyoplankton biomass 

 estimation techniques. It is, however, a data rich 

 method requiring both ichthyoplankton and adult 

 sampling plus staging of eggs and various labora- 

 tory measurements which may not be available. 

 In this report, I present an alternative method for 



137 



