380 
Fishery Bulletin 95(2), 1 997 
CZZD AGE 2 ■ SEPSW --A--TOTSE --K--NOVSES O MAR SUN 
Year class 
Figure 3 
Percent anomalies of the number of days of September southwesterly wind (SEPSW), num- 
ber of days of southeasterly wind from August through December (TOTSE), and number of 
storm days in November with southeasterly winds (NOVSES) and March sunshine (MAR 
SPIN ) compared with anomalies of the 1965-89 age-2 Atlantic herring abundance estimates. 
Table 3 
Mean and standard deviation (in parentheses) and f-test results assuming unequal variances for the environmental factors 
associated with the high and low age-2 Atlantic herring estimates, ns = not significant. 
Environmental factor 
Low age-2 estimate (n= 4) 
High age-2 estimate (n= 4) 
t-test probability 
November storms 
7.20 (1.64) 
7.00(5.23) 
ns 
March sunshine 
40.75 (6.94) 
57.00 (3.16) 
0.012 
September SW wind 
11.40(2.79) 
17.00 (1.63 
0.008 
Aug-Dec SE wind 
14.40(2.30) 
19.75 (3.30) 
0.039 
November SE storms 
0.20 (0.451) 
1.75 (0.50) 
0.003 
normal in those years. Average year classes were 
much more common during the 25-yr period and, in 
14 of these 17 years, August-December southeast- 
erly winds were also below average. September 
southwesterly winds and November southeasterly 
storms, on the other hand, were below average dur- 
ing about half of these years. 
Discussion 
In our study, we observed associations of sunshine, 
wind direction, and velocity with the number of age- 
2 herring estimated to recruit to the coastal Atlantic 
stock. Strong year classes were produced in years 
with more days of southerly fall (September) winds 
and storms (November), and weak year classes were 
produced in years with less sunshine in March and 
fewer days of southeasterly fall (August-December) 
winds. However, because southwest wind velocities 
are lower than velocities from other directions in the 
Gulf of Maine in fall, it is not possible to differenti- 
ate between the effects of wind direction and strength 
in this study. We believe that our results, without 
specifically addressing how wind events influence 
herring larval survival, show that recruitment suc- 
cess and, therefore, larval survival are related to wind 
events. In general, wind-driven transport and tur- 
