62 
Fishery Bulletin 1 12(1) 
Table 2 ( continued ) 
Model 
Component 1 
D 
D 
Pd 
DIC 
ADIC 
BF 
LBF 
M2B 
NEFSC 
330.08 
320.71 
9.37 
339.46 
bt 
0.28 
0.07 
0.22 
0.50 
SEAMAP 
117.95 
114.63 
3.68 
121.63 
a 2 
4.17 
4.17 
0.00 
4.17 
°P 
5.04 
5.04 
0.00 
5.04 
Total 
457.52 
444.25 
13.27 
470.80 
MIBvs. M2B: 4.52 
0.10 
-4.51 
M2N 
NEFSC 
317.22 
307.33 
9.89 
327.11 
bt 
0.82 
0.51 
0.31 
1.13 
SEAMAP 
115.64 
111.40 
4.23 
119.87 
a 2 
0.55 
0.55 
0.00 
0.55 
5.04 
5.04 
0.00 
5.04 
Total 
439.26 
424.83 
14.43 
453.69 
Ml vs. M2N: 0.32 
2.32 
1.69 
M2rUN 
NEFSC 
313.96 
303.66 
10.26 
324.22 
bt 
1.27 
1.16 
0.11 
1.38 
SEAMAP 
113.97 
109.69 
4.28 
118.24 
a 2 
0.55 
0.55 
0.00 
0.55 
5.037 
5.037 
0.00 
5.037 
Total 
434.77 
420.12 
14.65 
449.42 
MlrU vs. M2rUN: 1.53 
1.76 
1.13 
M2BN 
NEFSC 
328.73 
319.19 
9.54 
338.28 
bt 
0.36 
0.09 
0.27 
0.63 
SEAMAP 
118.23 
114.49 
3.74 
121.98 
a 2 
0.55 
0.55 
0.00 
0.55 
0 P 
5.04 
5.04 
0.00 
5.04 
Total 
452.91 
439.36 
13.56 
466.47 
MIBvs. M2BN: 0.20 
2.06 
1.45 
1 The components accounted for in DIC calculations for models without MWET were the Northeast Fisheries Science Center 
(NEFSC) fall index, the Southeast Area Monitoring and Assessment Program (SEAMAP) fall index, the depletion time series 6 t , 
and the process error variance Op ; the components for models incorporating MWET also included the coefficient a. Note that 
the DIC calculations should only include the components for indices and the depletion, but WinBUGS unexpectedly included the 
process error variance and a as well. 
especially when the commercial removals, the SESTF 
bycatch, or both suddenly increased. (Following low 
biomasses in the early 1970s and 1980s, the models 
predicted large but imprecise depletion levels that 
were needed to support the upsurge of the fisheries 
removals). The biomass ratios indicate an overfished 
stock of Atlantic Croaker in most years except in the 
mid-1970s, mid-1980s, and perhaps in 1991, 2004, and 
2007 when the NEFSC index had peaked after years of 
low total fishery removals. 
The harvest ratios, TTj/^MSY (Fig.6, C and D), showed 
trends opposite of the biomass ratios, and their preci- 
sion was generally consistent over time. They indicated 
that the Atlantic Croaker stock likely experienced over- 
fishing during 1993-2001. 
The risks for the Atlantic Croaker stock being over- 
fished (Fig. 6, E and F) coincided with the lowest es- 
timates for the NEFSC index and culminated in years 
when both this index was lowest and the estimates 
of total fishery removals in the preceding years were 
highest (e.g., 1972-74, 1978-83, and 1995-2001). This 
result reflected the model structures and behaviors in 
that, in a given year, the estimated stock biomass was 
largely driven by that year’s NEFSC index and the to- 
tal fishery removals of the preceding year. On the other 
hand, the models interpreted the magnitude of total 
fishery removals as a signal of overfishing risk: this 
one was highest in years of larger total fishery remov- 
als (1993-2001), lowest otherwise (Fig. 6, E and F). For 
the period of 2002-08, the risk of overfishing averaged 
0.2 and the risk of the overfished status averaged 0.7. 
Biomass depletions, ratios, and overfished risks 
were insensitive to model variants, including the re- 
jected ones, because all models fitted the biomass indi- 
ces equally. The harvest ratios and risks of overfishing 
trended similarly across models. In some years, howev- 
er, the latter statistic was largest for models including 
the SESTF bycatch. 
Discussion 
We used BDMs to improve understanding of the cli- 
mate effects on Atlantic Croaker production dynamics 
along the U.S. Atlantic coast. An age-structured pro- 
duction model has been applied to address the same 
