Koob et al.: Age validation of the northern stock of Centropristis striata in the Atlantic Ocean 267 
selected for further analysis. No interac- 
tion between Age Bin and Region was 
observed (F=1.717, df=2, P=0.1800); how- 
ever, there was an interaction between 
Season and Region (F=3.593, df=3, 
P=0.0132). Marginal increment ratios 
were higher in the winter, spring, and fall 
for otoliths from fish sampled south of the 
Hudson Canyon than those for otoliths 
from fish collected north (P<0.010, 
P<0.05, and P<0.0001, for each season 
respectively), but MIRs for regions were 
similar in summer (P=0.4486). Figure 7, 
A and B, shows that there was one mini- 
mum MIR per year for each region and 
that it occurred in summer and was dif- 
ferent from the MIRs for all other seasons 
(P<0.0001). This minimum occurred once 
per year for each region and was verified 
by plotting monthly means (Fig. 6, D 
Figure 4 and E). An interaction was also detected 
Bland—Altman bias plot of differences in age determinations between read- between Season and Age Bin (F=16.602, 
ers (reader 2—reader 1) of sagittal otoliths from black sea bass (Centropristis df=6, P<0.0001), corroborating results 
striata) caught in the Atlantic Ocean off the northeastern United States from the previous model, meaning there 
during 2013-2017. The dotted black line (located at 0) represents no bias were differences in MIR between age bins 
in age estimates, the solid black line indicates the degree and direction of in each season (P<0.0100). The only vari- 
bias in age estimates for samples used in this study, and the dashed black ante GES Hiner IMIS Pap NEM engl NBD Sa 
lines indicate 95% confidence limits. Size of circles indicates the number of 
samples. 
@Q 
— 
© 
a) 
> 
= 
® 
1S) 
Cc 
o 
—_ 
© 
= 
re) 
Co) 
D 
<x 
3 4 5 6 7 8 
Average age (years) 
summer were similar (P=0.4830), an out- 
come that was also observed in the model 
with the predictor Month Bin. 
Marginal increment ratio 
Jan— Mar— May-— Jul- Sep— Nov— Jan— Mar— May-— Jul—- Sep— Nov— Jan— Mar— May— Jul- Sep— Nov— 
Feb Apr Jun Aug Oct Dec Feb Apr Jun Aug Oct Dec Feb Apr Jun Aug Oct Dec 
Figure 5 
Box plots of marginal increment ratios for sagittal otoliths from black sea bass (Centropristis striata) captured in the Atlantic 
Ocean off the northeastern United States during 2013-2017, by month bin for (A) age bin 1 (ages 1-2), (B) age bin 2 (ages 3-4), 
and (C) age bin 3 (ages 5+). Letters above the boxes denote significant differences (significance level=0.05; tested with Tukey’s 
honestly significant difference by using estimated marginal means). In each box plot, the thick horizontal line indicates the 
median, the areas above and below the median represent the 25th and 75th percentiles, the thin vertical lines indicate the 95% 
confidence limits, and the points indicate outliers. 
