342 
Fishery Bulletin 118(4) 
GOM age based 
Atlantic age based 
& GOM GROTAG 
} Atlantic GROTAG 
> 
oO 
® 
> 
= 
LW 
E 
E 
= 
D 
c 
a) 
i 
w 
8 
4 
10 20 30 
Agao In length (mm FL/year) 
Figure 6 
more reliable, estimates of longevity 
and age-independent estimates of matu- 
rity. Although methods of modeling 
growth with tag-recapture data have 
historically been considered inferior to 
age-based methods (Harry, 2018), given 
recent doubts about the suitability of 
vertebrae for aging, methods based on 
tag-recapture data may be more reliable 
than age-based methods (Harry, 2018; 
Natanson and Deacy, 2019). When pos- 
sible, more computationally advanced 
methods, such as those of Aires-da-Silva 
et al. (2015) and Francis et al. (2016), 
which incorporate all available growth 
data (length based, tag and recapture, 
and age based) that can be integrated 
directly into stock assessment models, 
should be considered because they can 
combine all available data sources to 
estimate growth characteristics of a 
population. Additionally, methods that 
involve tag-recapture data may be useful 
in providing growth information for spe- 
cies that are difficult to age and for pop- 
ulations for which sacrifice of specimens 
is not preferred or fishery-dependent 
catch are not available for sampling (e.g., 
for protected species). 
Given the large discrepancies between 
the age-based model and the model based 
on tag-recapture data for bonnetheads in 
the GOM, the length-based tag-recapture 
estimates generated in our study (espe- 
cially age at 50% maturity and longevity) 
likely more accurately describe growth of 
Plot of bootstrap estimates of growth rates (g) at the 2 reference lengths of 
555 and 830 mm fork length (FL), parameters g;,, and 83), from GROTAG 
models based on tag-recapture data for female bonnetheads (Sphyrna tiburo) 
tagged in the northeastern Gulf of Mexico (GOM) during 1993-2006 and off 
the Atlantic coast of the southeastern United States (Atlantic region) during 
1998-2019 and from region-specific, age-based von Bertalanffy growth func- 
tions generated by using original length-at-age data from Lombardi et al. 
(2007) for sharks in the GOM and from Frazier et al. (2014) for sharks in the 
Atlantic region. Mean growth rates for age-based models were calculated from 
bootstrapped estimates of lengths at ages ®, X, and Y (parameters lg, ly, ly). 
Estimates are shown with 50%, 80%, and 95% polygon ellipsoid confidence 
intervals. Bootstrapping was done with 5000 iterations. 
bonnetheads in the GOM than age-based 
estimates. Parameters of the VBGF are 
commonly used in modeling estimates 
of mortality, maturity, and longevity 
and in modeling for stock assessments. 
Use of current age-based parameters 
for the population in the GOM should 
be carefully considered until growth in 
the population can be reexamined. Using 
parameters that underestimate age and 
maturity can lead to overly optimistic 
estimates of growth and mortality rates, 
increasing the potential for overexploita- 
tion of stocks. Results of this study indi- 
Results of recent studies indicate that in some species of 
sharks, band-pair formation may be a function of growth 
and body girth, not age (Natanson et al., 2018b). The 
results of our study indicate that the use of tag-recapture 
methods can produce estimates of growth that are similar 
to those produced with age-based methods. 
When sufficient tag-recapture data are available, mod- 
els should be used to verify growth estimates from age- 
based models and provide independent, and perhaps 
cate that bonnetheads in the GOM may be more susceptible 
to overexploitation than previously believed. The status of 
the stocks of bonnetheads in the northeastern GOM and 
off the Atlantic coast of the southeastern United States are 
currently considered to be unknown, and no stock assess- 
ments are pending. Given these results, we recommend 
that region-specific assessments be conducted and that 
they incorporate these new age-independent estimates of 
life history characteristics. 
