170 
Fishery Bulletin 120(2) 
1.2 
© Nominal CPUE 
e Standardized CPUE 
1.0 
] l l 
0.0 02 04 06 08 
ic 
int 
12) 
x 
fo) 
te) 
ds 
S 
Oo 
- 
& 
© 
=) 
2 
= 
ne) 
E 
uw 
=) 
oO 
Oo 
T T T T T l 
2006 2008 2010 2012 2014 2016 2018 
Year 
Figure 5 
Nominal and standardized catch per unit of effort (CPUE) of red drum 
(Sciaenops ocellatus) from bottom longline surveys conducted in the north- 
central Gulf of Mexico during 2006-2018. A mean of 100 stations (standard 
deviation 22) were sampled per year (range: 80-143 stations). Median values 
are shown in the index standardized by fitting a negative binomial general- 
ized linear model to the CPUE data. For 2009, there is no standardized CPUE 
estimate because of a lack of positive catch data with corresponding data for 
abiotic variables from that year. Error bars indicate 95% confidence intervals 
Despite the large sample size and 
broad size distribution of red drum cap- 
tured by using 2 fishery-independent 
gear types, individuals between 600 
and 800 mm TL (ages 3-6) were nota- 
bly rare in our study. Interestingly, it is 
in this size range that red drum in this 
region undergo maturation, according 
to mean estimates of size and age at 
maturity from Bennetts et al. (2019). 
Specifically, mean age at 50% maturity 
for males and females is approximately 
3 years, with fully mature individuals 
(spawning capable and elevated gonado- 
somatic index) undetected until ages 5 
and 6 (Bennetts et al., 2019). Therefore, 
although a multi-panel gill net can ade- 
quately sample fish of ages 0-2 and the 
bottom longline can adequately sample 
fish of age 7 and older, fish between the 
ages of 3 and 6 years are not selected by 
either gear type. Similar size selectivity 
has been reported for red drum off the 
West Florida Shelf. Using 3 fishery- 
of standardized CPUE. 
versus 31 years), a difference that illustrates the impor- 
tance of sampling enough large, presumably old individuals. 
Specifically, we collected more than 4 times more individu- 
als larger than 1000 mm TL than Bennetts et al. (2019); 2 of 
these fish, 1 male and 1 female, were assigned ages of 
36 years. Although fish older than 36 years are likely rare 
off Mississippi and Alabama, future efforts to model age and 
growth for red drum should consider collections that span 
the entirety of the range of the species. Future research also 
should account for the effects of gear selectivity, temporal or 
spatial changes in age structure, variable recruitment, and 
unexplained variance arising from individuals of undeter- 
mined sex, all of which are potential sources of bias in 
growth model parameters in this study. 
independent gear types (haul seine, 
trammel net, and purse seine), Winner 
et al. (2014) demonstrated that red drum 
that were 600-800 mm TL were not well 
represented in the catch in either haul seines or purse 
seines, yet they were dominant in trammel-net surveys. 
These examples illustrate that population dynamics are 
difficult to assess for red drum and that multiple gear 
types are needed to describe population dynamics across 
all life stages of this species. 
Surprisingly, a comprehensive review of life history stud- 
ies of red drum revealed that recent age-based estimates of 
M are not available for this species (SEDAR, 2016). During 
the most recent stock assessment, it was concluded that the 
updated Hoenig equation using longevity (Then et al., 2015) 
was the most robust approach for red drum. Our estimate of 
annual M based on the Then et al. (2015) approach was 0.14 
year ‘, a rate that is similar to the range of values used in 
Table 2 
Percentage of contribution of the 3 variables identified through analysis with boosted regression trees as the most influential on 
relative abundance of red drum (Sciaenops ocellatus) in the north-central Gulf of Mexico between 2006 and 2018. The area under 
receiver-operator-curve (AUC) and cross-validated (CV) AUC scores, with standard errors (SEs), were used to assess the model’s 
ability to discriminate between species presence and absence. 
Training 
data AUC CV AUC 
Season score score (SE) Variable 
Spring 0.90 
Summer 0.90 
Autumn 0.90 
Marginal Effect 1 
Marginal 
Marginal Effect 2 Effect 3 
Variable % 
% Variable % 
0.86 (0.01) Northward velocity of surface current 26.2 Surface temperature 20.7 Depth 14.7 
0.85 (0.01) Northward velocity of surface current 25.8 Surface temperature 20.4 Depth 14.6 
0.86 (0.01) Northward velocity of surface current 25.8 Surface temperature 20.4 Depth 14.4 
