336 
Fishery Bulletin 118(4) 
O Nylon dart 
4 Rototag 
A 
4a 
A 
T 
? _.*506--~7000 1500 2000 2500 
Oo 
Oo 
A 
Change in residuals 
Time at liberty (d) 
Figure 2 
Changes in residuals for ages of recaptured bonnetheads (Sphyrna tiburo) in 
the estuarine waters of the Atlantic coast of the southeastern United States 
plotted against their times at liberty, by type of tag: nylon dart (squares) and 
3000 3500 
y=0.0054x— 14.886, R?=0.0053 
overlap (Table 3). For sharks in the GOM, 
published estimated maximum ages are 
significantly below calculated maximum 
ages from models based on age and on 
tag-recapture data, with published val- 
ues below 95% Cls for all models with the 
exception of the GROTAG model used for 
estimating growth of males; a lower limit 
of the 95% CI could not be estimated 
because of the size of the tagged shark 
used in relation to the lower limit of the 
95% CI for the estimate of L,,. For bonnet- 
heads in the Atlantic region, calculated 
maximum ages are also significantly 
larger than published estimates, with the 
lower limit of the 95% CI falling above 
maximum estimated age for all models. 
Published estimates of age at 50% 
maturity are lower than those calculated 
by using the GROTAG models (Table 3). 
We were unable to calculate the lower 
limit of the 95% CI for age at 50% matu- 
rity of males with either GROTAG 
model, and 95% CIs based on the GOM 
data were not published; therefore, we 
cannot determine if differences are sig- 
nificant for these estimates. Upper and 
lower confidence intervals for estimates 
from both length-based and age-based 
models were available only for females 
in the Atlantic region. Although the 
GROTAG model for females in the Atlan- 
tic region produced an estimate of age at 
50% maturity that is slightly larger than 
that from the age-based model, 95% CIs 
for the estimates overlap. 
rototag (triangles). If tagging has a negative effect on growth, the majority 
of data points should be negative and the slope of the dashed line should be 
significantly different from zero. The equation for the trend line and the coef- 
Discussion 
ficient of multiple determination (R) are provided. Residuals were calculated 
by using lengths at initial tagging and at recapture and by using ages and the 
von Bertalanffy growth curve from Frazier et al. (2014). 
deviate between models as individuals grow beyond the 
upper reference length. 
Longevity and age at 50% maturity 
For the population in the GOM, data for recaptured bonnet- 
heads from long-term tag-recapture studies, a male and a 
female with times at liberty of 1639 and 1833 d and initial 
FLs of 600 and 800 mm, respectively, were used to calcu- 
late maximum age in each model. For the population in the 
Atlantic region, data for a recaptured male and a recap- 
tured female, with times at liberty of 2659 and 3263 d and 
FLs of 763 and 980 mm, respectively, were used. Calculated 
ages from the GROTAG model are larger than those from 
the age-based models for both regions, although 95% CIs 
This study confirms previously pub- 
lished significant differences in region- 
specific growth and life history 
characteristics of bonnetheads between 
the northeastern GOM and the Atlantic region by using 
age-independent methods of modeling growth. Growth 
rates, age at 50% maturity, and longevity all differ for both 
sexes between these 2 regions. Despite low sample sizes and 
poor coverage of ranges of lengths for males, we were able 
to produce reasonable growth estimates for the populations 
in the northeastern GOM and the Atlantic region by using 
the GROTAG model; however, we were unable to determine 
significance of some life history characteristics because of 
the high uncertainty in estimates of growth and growth 
parameters. The low sample sizes of males from the Atlan- 
tic region available for growth modeling is reflective of the 
nearshore distribution of male bonnetheads, with tagging 
effort heavily skewed toward females in estuarine waters 
where most fishery-independent sampling and tagging 
