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



Initial Conditions 



The initial conditions for a simulation are the 

 steady state solutions of the plankton dynamics 

 for a stratified water column (Fig. la). The in- 

 itial temperature gradient is 4°C over the upper 

 100 m. The daily average surface heating is 50 W 

 m"^ which is divided equally between short- 

 and long-wave radiation. A background eddy dif- 

 fusivity of 10""* m^ s"' throughout the 

 water column is assumed. 



RESULTS 



Multiple Wind Events 



Figure 2 shows the wind speed, level of tur- 

 bulent mixing at 3 m below the surface, and the 

 mixed layer depth for the simulation where 

 three wind events occur within a 15 d period. 

 Fifteen days is the critical period after yolk-sac 

 absorption when northern anchovy larvae are 

 susceptible to starvation (O'Connell 1980). Each 



Using Mellor • Yamada ( 1982 ) 

 Mixed Layer Model 



^ 20- 



w 10 

 E 5 



3 Wind Events in 1 5 Days 



4 Calm 

 Heating Days 



4 Calm 

 Heating Days 



4 Calm 

 Heating Days 



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



^ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 



Time (days) 



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



Figure 2. — The wind speed (a), vertical eddy diffusivity K^, 

 at 3 m depth (b), and mixed layer depth (c) during a sim- 

 ulated 15 d period. The Mellor-Yamada (1982) model was 

 used to calculate K„ and the mixed layer depth. 



wind event has a wind speed of 16 m s~^ and 

 a duration of 24 hours. Between the events, the 

 water column restratlfies due to a solar heat- 

 ing of 25 W m~^ at the surface and to a 

 short-wave radiation flux of an additional 25 W 

 m~^ penetrating the upper 10 m of the water 

 column. 



In the Mellor-Yamada (1982) model, the tur- 

 bulent diffusion coefficient K„ is predicted as a 

 function of depth; it has a maximum within the 

 mixed layer and decreases markedly below. 

 Since we are primarily concerned with dissipa- 

 tion of plankton near the surface where the con- 

 centration of prey is greatest (due to high 

 productivity of the phytoplankton), we plot in 

 Figure 2b the value of X,, at 3 m depth over the 

 15 d period. The deepening of the mixed 

 layer during each event and its shallowing after 

 each wind event is shown in Figure 2c. Note 

 that the mixed layer deepens slightly more with 

 each subsequent event, as the water column 

 does not completely restratify during the 4 d 

 period between each wind event. The back- 

 ground diffusivity maintains a mixed layer 5 m 

 deep. 



The model prediction of the increase in larval 

 anchovy mortahty for the three wind event case 

 is shown as curve b in Figure 3. For comparison 

 the mortality rate for larvae which experienced 

 only a single event of the same wind speed and 

 duration is shown as curve a in Figure 3. Curves 

 a and b are the same until day 5 when the second 

 wind event begins. The mortality rate for larvae 

 experiencing a second and then third wind event 

 continues to increase with time, while the mor- 

 tahty rate for larvae enduring only one event 

 declines as turbulence in the water column dis- 

 sipates and food concentrations are reestab- 

 lished (see Figure 1, panels c and d). 



The mortality rate after 15 days for larvae 

 having endured a single wind event is about 6% 

 d~^. If no wind event had occurred, their 

 mortality rate would be about 4% d^^ (curve 

 e in Figure 3). Thus a single event does not have 

 a great cumulative effect. However, for larvae 

 having endured 3 wind events over their 15 d 

 development period, the mortality rate increases 

 to 13% d"^ (curve b in Figure 3). 



The mortality rate increases dramatically with 

 the frequency of wind events. Curve c in Figure 

 3 shows the influence of 5 wind events each of 

 wind speed 16 m s~^ and 24 hours duration on 

 a cohort of larvae reaching the first feeding stage 

 at time zero. The mortality rate at the end of 15 

 days is 21% d"^ 



390 



