Brown-Peterson et al.: Meta-analysis of reproductive parameters of Lutjanus campechanus in the Gulf of Mexico 
45 
SlNNjM 
-i-1-r- 
-i-1-r- 
tl § 
c QQ 
3 LL. 
u O 
o U) 
o 01 
-O <D 
<1) *- 
> O 
■H 
_aj o 
CD 3 
QC 
T 394 T 338 
Figure 5 
Estimated red snapper (Lutjanus campechanus) relative batch fecundity (number eggs/g 
ovary-free body weight, OFBW) from 1991 through 2004 and 2007 through 2017 in the 
northern Gulf of Mexico (GOM). Individual relative batch fecundity values are represented 
by circles (northeastern GOM) and triangles (northwestern GOM). The solid line indicates 
the annual trend with the dashed lines showing the 50% and 95% credible intervals. The 
trend was estimated for regions combined (graph A) and separately for the northeast (graph 
B) and northwest (graph C). Extreme values are indicated by arrows in the region-specific 
panels but are omitted from the combined-region plot for visual clarity. 
that there is an overall reduction in surplus energy for 
growth and reproduction as the population increases 
(McBride et al., 2015, Porch et al., 2015). Decreases in 
fecundity with increasing fish density have been previ¬ 
ously reported (Rose et al., 2001, Forrester et al., 2011, 
Hixon et al., 2012), and this decrease has been hypoth¬ 
esized to be a function of competition, food availability, 
or habitat availability. Thus density-dependence pre¬ 
dicts a compensatory response wherein egg production 
may decline as population increases (Rose et al., 2001, 
Kulaw et al., 2017), possibly reflecting reproductive 
tradeoffs. Both conditions must be considered in order 
to track stock productivity over time (Trippel, 1995, 
Rose et al., 2001). 
Our temporal meta-analysis models provide a 
unique approach to understanding potential changes in 
female red snapper reproductive parameters over a 27- 
year period. Although some reproductive parameters 
of red snapper, such as female spawning fraction with 
fish size, region, depth, and month (Porch et al., 2015) 
and variations in fecundity by habitat (Karnauskas et 
al., 2017) have been previously modeled, our study is 
the first effort to use modeling to quantify decade-level 
changes in red snapper reproduction. Both Porch et 
al. (2015) and Karnauskas et al. (2017) explicitly mod¬ 
eled an age-structured population. Our purpose was 
to more fully evaluate whether there is evidence for a 
compensatory response by examining key elements of 
egg production as represented by spawning interval, 
fecundity, and duration of spawning season (i.e., based 
on the GSI). Compensation can be very difficult to de¬ 
tect from empirical field data which often come from 
short-term studies (Rose et al., 2001). This challenge 
prompted our focus on meta-analysis. We examined 
trends over 27 years and based on many separate stud¬ 
ies, and therefore age data were often not available. 
Therefore, we modeled spawning interval averaged 
across multiple age classes, and the size-age effect 
on fecundity was standardized to the degree possible 
by examination of RBF. Nor did we explicitly account 
