Fernandez-Carvalho et al.: Age and growth of Alopias superciliosus in the Atlantic Ocean 
477 
Table 3 
Estimated means of growth parameters for bigeye thresher ( Alopias superciliosus ) col- 
lected from the North and South Atlantic Ocean between 2007 and 2009, obtained with 
the VBGF model with a fixed size at birth (Lq = 84 cm in fork length [FL] ). For the 
models for each sex and region, the parameter means are given with their respective 
standard errors (SE) and 95% confidence intervals (CIs). L; n (=asymptotic maximum 
length, given in FL; &=growth coefficient per year. 
95% Cl 
Sex 
Atlantic 
Parameter 
Estimate 
SE 
Lower 
Upper 
Males 
North 
L inf 
211.77 
4.64 
202.62 
220.93 
k 
0.18 
0.01 
0.15 
0.21 
South 
Linf 
229.00 
5.10 
218.85 
239.15 
k 
0.13 
0.01 
0.11 
0.15 
Females 
North 
Linf 
219.20 
4.34 
210.64 
227.77 
k 
0.16 
0.01 
0.14 
0.18 
South 
Linf 
265.70 
11.34 
243.01 
288.43 
k 
0.09 
0.01 
0.07 
0.11 
siderably older than the age at which band counting 
is no longer possible for aging (Francis et al., 2007; 
Andrews et ah, 2011; Hamady et ah, 2014; Passerotti et 
al., 2014; Andrews and Kerr, 2015). Underestimation of 
longevity may cause little change in the estimation of 
growth parameters, but the life-time reproductive pro- 
ductivity of long-lived sharks would be greater (Pas- 
serotti et ah, 2014). Despite the lack of validation for 
bigeye thresher in this study, the growth data present- 
ed are the most comprehensive age estimates for this 
species for the North and South Atlantic Ocean and, as 
such, are an important contribution to our understand- 
ing of the biology of this species. Nonetheless, it should 
be noted that these estimates should be used with cau- 
tion for stock assessment and management decisions 
until a definitive age validation is accomplished. 
In this study, all growth parameters estimated with 
the 4 growth models that converged had biologically 
reasonable values. The differences between the AIC 
values of the VBGF model with the lowest AIC (3-pa- 
rameter VBGF model) and the 3-parameter GGF for 
both sexes were small (A aic males= 4 - 4 ; a AIC females=5.2), 
but the differences between the VBGF model with the 
lowest AIC and both the VBGF and GGF models with 
a fixed Lq were high ( > 10); therefore, these models with 
a fixed Lq should be discarded, as suggested by Katsa- 
nevakis (2006). Nevertheless, it should be noted that 
when considering VBGF and GGF models with a fixed 
Lq, different authors give different values for size at 
birth (e.g., Bass et ah, 1975; Moreno and Moron, 1992; 
Gilmore, 1993) and that changing this value will affect 
the estimation of the other parameters (Pardo et al., 
2013). On the other hand, although the GGF models 
also produced realistic growth parameters, this growth 
function has been described as better suited for batoids 
(or elasmobranchs that hatch from eggs), for which vol- 
ume increases more with age than with length (e.g., 
Myliobatiformes) (Goldman et al., 2012). Because the 
growth parameters obtained by the 3-parameter VBGF 
model had the best statistical fit and seemed to be bio- 
logically realistic, we recommend its use for describing 
the growth of bigeye thresher. 
To our knowledge, this study is the first comprehen- 
sive one for the age and growth of bigeye thresher, cov- 
ering both the North and South Atlantic Ocean. Initial 
growth curves have been produced by Fernandez-Carv- 
alho et al. (2011) for the region of the Cape Verde Ar- 
chipelago in the tropical northeastern Atlantic Ocean. 
Because the sample size and coverage area in that 
study were relatively small, these samples collected 
in Cape Verde Archipelago have been included in the 
study described in this article. In comparisons of the 
parameters from our most recent study with our previ- 
ous estimates for the Cape Verde region, the main dif- 
ferences were observed among Males; among females, 
the growth parameters were similar (Table 4). The k 
value obtained in this study for males (k~0. 09/year) 
was lower than our first estimate (^=0. 18/year), and 
the L; n f of 245.6 cm FL for this study was higher than 
the Li n f of 206.0 cm FL for just the Cape Verde region. 
The growth parameters obtained in this study are 
comparable with those parameters generated by Liu et 
al. (1998) for the population in the northwestern Pa- 
cific Ocean and by Mancini (2005) for the southwestern 
Atlantic Ocean — with some differences. Females in our 
study grew to a larger size (Li n f=284.2 cm FL) but did 
so at a slower rate (^=0. 06/year) than did the females in 
the study in the northwestern Pacific Ocean (Li n f=241.7 
cm FL, k=0. 09/year) (Table 4). On the other hand, 
males sampled in our study grew to a slightly larger 
size (Li n f=245.6 cm FL) but did so at a rate similar 
to that of males described for the northwestern Pacific 
Ocean (Li n f=235.5 cm FL, ^=0. 09/year) (Table 4). Man- 
cini (2005) presented L [ n f values higher than those of 
