ern anchovy larvae for 15 days from first feed- 

 ing. From histological indications of tissue condi- 

 tion, O'Connell (1980) showed that anchovy 

 larvae less than 10 mm SL are vulnerable to 

 starvation. For northern anchovy, yolk-sac ab- 

 sorption and first feeding occurs 3 to 4 days 

 (depending on temperature) after egg hatching. 

 Larvae obtain a standard length of 10 mm about 

 19 days after hatching (Blaxter and Hunter 

 1982). Thus the larvae are subject to starvation 

 for 15 days after first feeding. The larvae meta- 

 morphose at 34 to 40 mm SL, about 56 to 60 days 

 after hatching. 



O'Connell (1980) suggested that larvae above 

 10 mm SL may be less vulnerable to starvation 

 because of increasing nutriment (protein, car- 

 bohydrate, and lipid) reserves with growth. 

 Early post-yolk-sac larvae have negligible re- 

 serves and will survive only two or three days 

 vdthout any food (Lasker et al. 1970; O'Connell 

 and Raymond 1970). Larger (35 mm SL) larvae 

 can survive two weeks of starvation as their lipid 

 content declines (Hunter 1977). 



Because of their small mouth size, first-feed- 

 ing northern anchovy larvae are restricted to 

 feeding on small prey (Hunter 1977). First-feed- 

 ing northern anchovy are initially able to subsist 

 on a diet of the unarmored dinoflagellate Gy7n- 

 nodinium splendens (Lasker et al. 1970). How- 

 ever, after a few days, their growth rate will be 

 greatly depressed unless their diet includes 

 more typical foods of young clupeoid larvae such 

 as tintinnids, cihates, copepod eggs, naupliar, 

 and copepodite stages (Arthur 1976; Blaxter and 

 Hunter 1982; Theilacker 1987). Therefore, in our 

 model, we consider the prey of young anchovy 

 larvae to be microzooplankton. We do discuss 

 later the availability of G. splendens to first- 

 feeding larvae under the turbulent mixing condi- 

 tions predicted by the model. 



Larval Fish Dynamics 



Our formulation of growth and mortality of 

 larval northern anchovy (see Wroblewski 1984) 

 is based on the laboratory experiments of 

 O'Connell and Raymond (1970), who determined 

 the effect of various zooplankton prey concentra- 

 tions on the survival of larvae over the fii'st 12 

 days of life. The prey in the O'Connell and Ray- 

 mond (1970) experiments were wild crustacean 

 nauplii with some tintinnids and phytoplankton 

 also present. 



The equation for growth rate of larval anchovy 

 expresses the difference between metabolic 



FISHERY BULLETIN: VOL. 87, NO. 3, 1989 



gains and losses, 



growth = ingestion - egestion and excretion. 



Ingestion of prey by larval anchovy is calculated, 

 using the Ivlev (1955) formulation for the feeding 

 of fishes. Egestion of fecal matter by anchovy 

 larvae is taken to be a constant fraction of the 

 ingested ration. MetaboUc excretion is assumed 

 to occur at a basal rate plus an additional excre- 

 tion associated with feeding activity. 



Larval anchovy mortahty is expressed in the 

 model as a function of weight at age, or in other 

 words, the feeding history of the larvae, 



mortality = baseline growth rate/actual growth 

 rate. 



Mathematical formulations for larval anchovy 

 growth and mortality used here are the same as 

 equations (5) and (6) in Wroblewski and Richman 

 (1987). 



Prey Dynamics 



The equation for the concentration and ver- 

 tical distribution of the prey of larval anchovy 

 (microzooplankton) is one of a set of coupled 

 partial differential equations describing the 

 plankton ecosystem. These equations for phyto- 

 plankton, zooplankton, and dissolved nutrient 

 (nitrate) in a one-dimensional water column are 

 given as equations (1) to (4) in Wroblewski and 

 Richman (1987). Analytical solutions to these 

 plankton equations have been derived and sensi- 

 tivity analyses have been performed so that one 

 can fully understand how the concentration of 

 prey responds to changes in the biological pa- 

 rameter values (see Franks et al. 1986; Wrob- 

 lewski and Richman 1987). 



The biological parameter values used here are 

 the same as in table I of Wroblewski and Rich- 

 man (1987), with the exception that the gi'owth 

 rate of the phytoplankton is reduced. In the pre- 

 vious study, it was assumed that the average 

 growth rate in the euphotic zone (waters with 

 greater than 1% surface light intensity) was 2 

 doublings day"\ if nutrients were not limit- 

 ing. Here we assume that the maximal growth 

 rate of the phytoplankton at the surface is 

 2 d ^ Below the surface, the growth rate of 

 the plants is given by V = V„, exp(-/c z), where 

 V,„ is 2 d~^ and the light extinction coeffi- 

 cient fc is 0.1 m~\ The effect of this change in 

 plant growth rate is to reduce the concentration 



388 



