36 
Fishery Bulletin 1 15(1) 
Table 3 
The effect of sea surface temperature (SST), regulatory period, and geographic zone on the immediate mor- 
tality of blue and tiger sharks in the U.S. Atlantic pelagic longline fishery during the period 1992-2008. NS 
denotes a nonsignificant effect of the factor on mortality. An asterisk (*) denotes significant results with P- 
values less than 0.05. An interaction effect was not examined in the model for immediate mortality of tiger 
shark (represented by N/A below). 
Species 
SST 
Regulatory 
period 
Geographic 
zone 
Regulatory period 
X 
geographic zone 
Deviance 
explained 
Blue shark 
* 
* 
* 
* 
1.6% 
X2=15 
X2=2761 
X^=4648 
P<0.05 
P<0.05 
P<0.05 
P<0.05 
1.6% 
Tiger shark 
NS 
* 
* 
N/A 
4.5% 
5 ( 2=49 
X^=137 
P<0.05 
P<0.05 
data set and the mandatory collection of such data, the 
USPL data set will be a powerful and economical tool 
in future studies for analyzing shark mortality trends, 
but the current problem of reliable species identifica- 
tion needs to be addressed. 
The rates of immediate mortality for 2 occasionally 
retained species deviated from the rates reported in the 
POP data set: 6% higher for porbeagles and 10% lower 
for oceanic whitetip sharks. Despite differences with 
the POP data set, the rates of immediate mortality for 
these species reported in the USPL data set were de- 
termined to be within the range of immediate mortal- 
ity rates reported in other sources (Table 2). Although 
this comparability indicates that the reporting of im- 
mediate mortality in the USPL for these species may 
be accurate, we recommend interpreting these values 
with caution. During the screening of data, all longline 
sets that retained individuals of the species being ex- 
amined were removed. This approach left only longline 
sets that had discarded sharks in the data set but it 
was assumed that the likelihood of retention was the 
same for dead and live sharks. This supposition is false 
because fishermen may be more likely to retain sharks 
that are dead or moribund than they are to keep live 
sharks (Gilman et ah, 2008). Because of this limita- 
tion, for occasionally retained species, the factors that 
infiuence immediate mortality during capture cannot 
be accurately assessed by using the designated catego- 
ries in the USPL data set. Therefore, it would be nec- 
essary for the USPL to include the number of live and 
dead or moribund animals retained in order for data 
to be reliable when some animals are kept as part of 
a catch. 
Factors contributing to mortality in blue and tiger sharks 
The establishment of fin-to-carcass weight ratios has 
afforded greater protection for tiger sharks than the 
implementation of the mandatory use of circle hooks. 
The large number of sharks discarded dead before 
March 1993 may be attributable to the high value of 
the fins of tiger shark compared with that of the fins 
of other shark species because “finned sharks” (sharks 
whose fins were removed in the catch) were likely to be 
recorded as discarded dead rather than retained dur- 
ing this period (Clarke et ah, 2006). The effect of circle 
hooks on immediate mortality is likely to be species- 
specific (Serafy et ah, 2012), and immediate mortality 
of tiger sharks did not decrease after the start of regu- 
lations for the mandatory use of circle hooks. 
In contrast with the effect on survival of tiger sharks, 
the prohibition of the use of J-style hooks had a more 
positive effect on the survival of blue sharks than the 
establishment of fin-to-carcass ratios (for zone-specific 
effects, see Fig. 4). We are unsure why differences in 
mortality rates varied according to geographic zones 
between the period before March 1993 and the period 
March 1993-August 2004. However, statistically signif- 
icant decreases in rates of immediate mortality for blue 
sharks in every geographic zone analyzed after August 
2004 have indicated that the mandated change in hook 
type has had a positive effect on the survival of blue 
shark caught as bycatch. 
Several studies have reported that the use of circle 
hooks instead of J-style hooks can reduce immediate 
mortality of blue sharks (Campana et ah, 2009; Car- 
ruthers et ah, 2009; Afonso et al., 2011; Afonso et ah, 
2012; Epperly et ah, 2012; Godin et al., 2012; Gilman 
et al., 2016; but see Kerstetter and Graves, 2006; Yo- 
kota et ah, 2006; Curran and Bigelow, 2011; Serafy et 
al., 2012) because such hooks decrease the likelihood of 
hooking in the gut (gut-hooking) in this species (Wat- 
son et al., 2005; Afonso et al., 2011). We cannot deter- 
mine the specific postrelease survival rate for released 
animals, but delayed death due to gut-hooking may 
have been a further significant cause of mortality that 
