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



S 20 ms"^ 





_L 



[ Z ] required tor optimal growtji 

 s-storm ( Z ] 



I Z 1 as a result of storm 



0.3 r 



mortality rale for post yolk-sac larvae 

 existing before storm event 



X3 



0.2 





0.1 



mortality rate if no storm had occurred 



mortality rate for lan/ae first feeding 2 days after storm 



initial rate 



mortality rate for larvae first feeding 4 days after storm 



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 

 Time ( days ) 



Figure 6. — Mortality rate predicted for larval northern anchovy positioned 

 at 3 m depth in the simulation shown in Figure 5. Curve a shows the 

 predicted mortality rate where winds are calm during the simulated 15 d 

 period. Mortahty rate increases because food is limiting larva grovrth. Curve 

 b shows the increase in zooplankton biomass at 3 m depth owing to the wind 

 event which entrains nutrients into the euphotic zone. Curve c shows the 

 predicted mortahty rate for larvae feeding in the water column during and 

 after the storm. Day zero is the day of first feeding for the cohort of larvae 

 described in curves a and c. Curves d and e show the enhanced survival of 

 larvae first feeding 2 days and 4 days respectively, after passage of the 

 storm. 



sage of a storm (Checkley et al. 1988). The strat- 

 egy of northern anchovy is to spawn contin- 

 uously (Methot 1981) so that some larvae will be 

 emerging from the yolk-sac stage at just the 

 right time to feed upon high concentrations of 

 microzooplankton induced by a storm. 



Since any wind event which falls on the heals 

 of the first beneficial storm will be detrimental, 

 optimal conditions consist of a storm followed by 

 a period of calm during which the larvae can 

 grow into a less vulnerable stage. Any beneficial 

 effect of the first storm will depend on the time 

 required to concentrate new prey biomass. If the 

 first storm is quickly followed by subsequent 

 wind events, the prey will remain disbursed and 

 the first-feeding larvae may not withstand the 

 starvation period. But if the storm is followed by 

 a period of calm winds, turbulence wall dissipate, 



the water column will restratify, and the en- 

 hanced secondary production will become con- 

 centrated. As the time to complete development 

 from post-yolk-sac stage to this less vulnerable 

 stage is about 15 days (O'Connell 1980), a 2 wk 

 period of calm following a storm strong enough 

 to deepen the mixed layer into the nutricline is 

 theoretically ideal for the survival of the north- 

 ern anchovy larvae. 



These theoretical results may apply as well to 

 other clupeoid species, such as the Atlantic 

 menhaden. Checkley et al. (1988) inferred that 

 the adaptive significance of spawning by B. 

 tyrannus during storms derives, in part, from 

 the enhancement of microzooplankton prey for 

 young larvae owing to the storm-induced upwell- 

 ing. One criterion for maximal survival of the 

 larvae of any clupeoid species would be a close 



394 



