Lough and O'Brien: Life-stage recruitment models for Gadus morhua and Melanogrammus cieglefinus on Georges Bank 
129 
0.30 -| 
0.25 - 
0.00 l 1 1 1 1 1 1 1 1 
0 25 50 75 100 125 150 175 200 
Day of year 
Figure 4 
Weekly vector-averaged wind stress (in pascals, I Pa]) from January 
through June 1995-99 obtained from a combination of NOAA buoys 
44-005, 44-008, and 44-0011 versus larval cohort day of the year 
from 1 January. Wind stress was used to derive a relationship with 
larval cohort mortality in Figure 5. Larval cohort mortality rates 
were estimated from 10-day hatching bins ( January-June) from 
Georges Bank U.S. Global Ocean Ecosystems Dynamics (GLOBEC) 
broadscale surveys, 1995-99. 
forage, and avoid predators as they grow. A 
positive correlation was found between prey 
biomass concentration and growth rates for 
larval cod and haddock collected on Georges 
Bank during the GLOBEC surveys (Buckley 
and Durbin, 2006). Increased larval growth 
would reduce stage duration and presumably 
enhance survival. Wind stress was used as 
an indication of wind mixing or turbulence 
because it can affect larval encounter rates 
with zooplankton prey and subsequent feed- 
ing, growth, and survival for each cohort. 
Field studies indicated that the highest feed- 
ing ratio (mean number of prey per larval 
gut) for larval cod occurred when wind 
exceeded 10 m/s (Sundby and Fossum, 1990; 
Sundby et ah, 1994). Modeled simulations 
indicated that maximum ingestion occurred 
at wind speeds near 15.3 m/s (MacKenzie 
et al., 1994). In a Georges Bank field study, 
Lough and Mountain (1996) found that the 
feeding ratio of haddock larvae increased 
with wind speed up to 7.1 m/s in the upper 
20 m. There was an interaction between tur- 
bulence and prey density; the mean prey per 
gut increased with prey density at lower tur- 
bulence. Weekly vector-averaged wind stress 
(in pascals, [Pa]) from January through June 
1995-99 (Fig. 4) was used from a combination 
of NOAA buoys 44-005, 44-008, and 44-0011 
(Manning and Strout, 2001) to correlate with larval 
mortality. Larval cod and haddock cohort mortality rates 
were estimated from 10-day hatching bins (January- 
June) from Georges Bank GLOBEC broadscale surveys, 
1995-99 by Mountain et al. (2008). Cohort mortality 
represents the first 15 days of early life from hatching 
to about 6-mm standard length (SL). Trend lines were 
fitted to the mortality and wind stress data for all years 
and for the high larval growth years 1998-99 and the 
low growth years 1995-96. Insufficient data were avail- 
able for 1997, and therefore this year was omitted from 
the analysis. Both cod and haddock showed a decreasing 
mortality rate with increasing wind stress; however, the 
correlation was not significant for cod. For haddock, the 
best fit based on the coefficient of determination was for 
the high growth years 1998-99 ( r 2 =0.57, Fig. 5), where 
mortality declined with wind stress until about 0.15 Pa 
and then increased after about 0.20 Pa. Wind stress of 
0.15—0.20 Pa is equivalent to a wind speed of 9—12 m/s. 
The trend line for all years combined (not shown) was 
similar to the 1998-99 fit, but less significant (r 2 =0.24). 
For 1998 and 1999, the lowest cohort mortality was 
associated with the highest wind stress (0.15-0.20 Pa) 
that occurred from January to early March (Fig. 4). 
Wind stress declined uniformly throughout the spring. 
For 1995 and 1996, there was generally lower wind 
stress than optimal for survival, except for a brief pe- 
riod of high wind stress in February and a higher wind 
stress in April (Fig. 4). It is interesting that in both of 
these years there was an intense storm period in early 
April (day 95). Larval mortality rates for the haddock 
April 5 (day 95) cohort were 8.8 and 7.8 %/d for 1995 
and 1996, respectively, and 11.3 and 1.1 %/d for 1998 
and 1999. Cod mortality rates for the same cohort were 
6.1 and 7.9 %/d for 1995 and 1996, respectively, and 
1.1 and 0.0 %/d for 1998 and 1999. Mortality rates for 
the haddock April 15 (day 105) cohort were 10.4 and 
13.0 %/d for 1995 and 1996, and 14.2 and 7.4 %/d for 
1998 and 1999. Cod mortality rates for the April 15 
(day 105) cohort were 7.7 and 5.9 %-/d, for 1995 and 
1996, and 11.4 and 6.4 %/d for 1998 and 1999. Larval 
cod growth rates based on otolith daily ring analysis 
also were slightly better in April and May 1996 than 
1995 (Mountain et al., 2008), consistent with the lower 
mortality in 1996. 
Because the haddock mortality-wind relationship for 
1998-99 best fitted the reported studies of feeding as 
related to wind stress, this relationship also was used 
to estimate larval stage mortality for both cod and 
haddock: 
M = 741.70 W 2 -249.29 W + 23.75, 
r 2 =0.57, /z=27, (4) 
where W = wind stress (Pa). 
Pelagic juveniles 
There is little information on pelagic juvenile mortality, 
and because it is expected to be less that larval-stage 
