330 
Fishery Bulletin 118(4) 
intermediate longevities (Frazier et al., 2014). For these 
reasons, modeling of growth with tag-recapture data and 
newer methods of modeling that combine tag-recapture 
data with age estimates have been recommended (Eveson 
et al., 2007; Aires-da-Silva et al., 2015; Francis et al., 2016; 
Natanson and Deacy, 2019). 
Growth models that use tag-recapture data also have 
limitations and biases. To accurately represent growth 
in a population, recapture data should include individ- 
uals throughout the range of lengths found in the pop- 
ulation. However, this goal is rarely achieved with most 
data sets. Further, growth of fish may be affected by tag- 
ging (Kelly and Barker, 1963; Saunders and Allen, 1967), 
including elasmobranch growth (Gruber, 1982; Parsons, 
1987; Davenport and Stevens, 1988; Kalish and John- 
ston, 2001). Measurement error in body length of tagged 
and recaptured fish can also bias models; however, some 
methods of modeling data can incorporate measurement 
error as a parameter (Francis, 1988b). When sufficient 
tag-recapture data are available, models can be used to 
verify or compare growth rates and estimates of longev- 
ity (maximum age) and age at maturity with those from 
traditional age-based models (e.g., Natanson et al., 2002; 
Welsford and Lyle, 2005; Natanson and Deacy, 2019). One 
advantage of models based on tag-recapture data is that, 
if recapture data spans multiple seasons, information 
about growth variability and seasonal changes in growth 
can be determined (Francis, 1988b). Such information 
can be especially useful when comparing region-specific 
growth rates between or within populations. 
The bonnethead (Sphyrna tiburo) is a relatively small 
shark species, with individuals reaching a maximum 
size of 150 cm total length, that is commonly found in 
the coastal and estuarine waters of the western North 
Atlantic Ocean from North Carolina to southern Brazil, 
including the Gulf of Mexico (GOM) and the Caribbean 
Sea (Compagno, 1984). Significant differences in life 
history characteristics exist between bonnetheads cap- 
tured off the Atlantic coast of the southeastern United 
States (hereafter referred to as the Atlantic region) 
and those caught in the eastern GOM (Frazier et al., 
2014). Regional variation in life history between popu- 
lations in the Atlantic region and in the GOM has been 
found for other coastal shark species, including the 
Atlantic sharpnose shark (Rhizoprionodon terraenovae) 
(Carlson and Loefer!), blacknose shark (Carcharhinus 
acronotus) (Driggers et al., 2004), and finetooth shark 
(C. isodon) (Drymon et al., 2006; Vinyard et al., 2019). 
The regional differences observed in bonnetheads are 
greater than those that have been described for any 
adjoining populations of other elasmobranch species. Sig- 
nificant differences in growth characteristics of bonnet- 
heads have been found between populations in the GOM 
' Carlson, J. K., and J. Loefer. 2007. Life history parameters for 
Atlantic sharpnose sharks, Rhizoprionodon terraenovae, from the 
United States South Atlantic Ocean and northern Gulf of Mexico. 
Southeast Data, Assessment, and Review SEDAR13-DW-08, 7 p. 
[Available from website.] 
and the Atlantic region (Frazier et al., 2014). For the pop- 
ulation in the Atlantic region, estimated maximum age 
(males: 12.0 years; females: 17.9 years) and age at 50% 
maturity (males: 3.9 years; females: 6.7 years) are more 
than twice the estimates for the population in the GOM 
(males: 5.5+ years for maximum age, 1.7 years for age at 
50% maturity; females: 7.5+ years for maximum age, 2.9 
years for age at 50% maturity) (Lombardi-Carlson et al., 
2003; Frazier et al., 2014). Previous age and growth stud- 
ies of bonnetheads in the eastern GOM (Parsons, 1993; 
Carlson and Parsons, 1997; Lombardi-Carlson et al., 
2003) found significant latitudinal variation in life his- 
tory traits; however, latitudinal variation has not been 
detected in the population off the Atlantic coast (Frazier 
et al., 2014). High degrees of site fidelity have been doc- 
umented for this species (Heupel et al., 2006; Driggers 
et al., 2014), and tagging data indicate that there is no 
mixing between populations in the GOM and Atlantic 
region (Kohler and Turner, 2019). 
The population status of bonnetheads in U.S. waters 
was most recently assessed as a single stock in 2013; how- 
ever, because of observed differences in life history, tagging 
data, and genetic population structure, the results of that 
stock assessment were rejected and it was recommended 
that regional populations (i.e., those in the GOM and 
Atlantic region) be assessed as separate stocks (SEDAR, 
2013). As such, the population status for both stocks is cur- 
rently considered to be unknown (SEDAR, 2013). 
Recent studies of bonnethead population structure 
(Escatel-Luna et al., 2015; Portnoy et al., 2015) found 
that the populations in the GOM and Atlantic region are 
genetically distinct, with evidence of fine-scale genetic 
structure within populations. Using a combination of 
mtDNA and nuclear single nucleotide polymorphisms, 
Portnoy et al. (2015) found evidence of female philopatry 
with male-mediated gene flow. Results of further analysis 
indicate that over half of a small sample of outlier single 
nucleotide polymorphism loci has signatures of latitudinal 
selection. Portnoy et al. (2015) proposed that philopatry 
can lead to adaptive variation on a local scale and that, 
when combined with sex-biased dispersal, adaptive vari- 
ation can move among locations and environments. The 
high degree of site fidelity and latitudinal variation in life 
history observed for bonnetheads, in addition to localized 
adaption to environmental conditions, could explain the 
dissimilarities in life histories between populations in 
the different regions. However, differences in aging tech- 
niques, age estimation, or spatiotemporal biases could also 
explain observed differences (Campana, 2001; Cailliet and 
Goldman, 2004). 
The objectives of this study were 1) to use 2 long-term 
mark-and-recapture data sets and an age-independent 
model, GROTAG, to estimate region-specific growth rates; 
2) to generate estimates for age-independent life history 
parameters and compare results to region-specific esti- 
mates based on length-at-age data for verification of current 
life history information; and 3) to estimate region-specific 
seasonal growth and growth variability of bonnetheads in 
the GOM and the Atlantic region. 
