Dapp et al.: Immediate mortality of Prionace glauca and Galeocerdo cuvier caught by pelagic longiines 
35 
Table 2 
The observed immediate mortality, calculated as the percentage of total discards that were dead discards, in the U.S. pelagic 
fishery logbook (USPL) data set for the period 1992-2008, after data screening was completed. Expected immediate mortality 
rate is given as a mean and range from existing sources that reported the immediate mortality rates of a species examined 
during longline capture. The expected immediate mortality rate column also provides the mean rate of immediate mortality 
recorded by the Pelagic Observer Program (POP), National Marine Fisheries Service, in Gallagher et al. (2014). Superscript 
denotes that the observed immediate mortality rate is likely to be inaccurate because of the discrepancy between observed 
(USPL) and expected (POP) percentages. 
Species 
Number 
examined 
Observed 
imm.8diate 
mortality 
Expected 
immediate 
mortality 
Sources 
Blue shark 
806,598 
17.7% 
Range: 5-34% 
Mean: 19% 
POP: 15% 
(Francis et al., 2001; Moyes 
et al., 2006; Hight et al., 2007; 
Petersen, 2008; Campana et al., 
2009; Coelho et al.'; Musyl et al., 
2011; Bromhead et al., 2012; Coelho 
et al., 2012; Griggs and Baird, 2013; 
Gallagher et al., 2014) 
Tiger shark 
15,474 
3.1% 
Range: 0-7% 
Mean: 3% 
POP: 3% 
(Morgan et al., 2009; Scott-Den- 
ton et al., 2011; Coelho et al., 
2012; Afonso and Hazin, 2014; Gal- 
lagher et al., 2014) 
Oceanic whitetip shark^ 
6348 
15.8%^ 
Range: 5-34% 
Mean: 24% 
POP: 26% 
(Musyl et al., 2011; Bromhead et 
al., 2012; Coelho et al., 2012; 
Gallagher et al., 2014) 
Porbeagle^ 
2619 
27.2%^ 
Range: 21-39% 
Mean: 32% 
POP: 21% 
(Francis et al., 2001; Griggs and 
Baird, 2013; Gallagher et al., 
2014) 
^Coelho, R., P. G. Lino, and M. N. Santos. 2011. At-haulback mortality of elasmobranchs caught on the Portuguese longline 
swordfish fishery in the Indian Ocean. Indian Ocean Tuna Comm. IOTC-2011-WPEB07-31, 9 p. lOTC, Victoria Mahe, Sey- 
chelles. [Available from website, accessed March 2015.] 
gust 2004, with a mean difference in percentages of 
3.0% (95% Cl: 2.6% to 3.5%; (Fig. 3). 
The immediate mortality model for blue sharks had 
a statistically significant interaction effect between 
regulatory period and geographic zone. Rates of imme- 
diate mortality for blue sharks caught before March 
1993 were lower in 2 zones (NCA and NED), higher in 
3 zones (MAB, NEC, and SAB), and not significantly 
different in 4 zones {CAR, FEC, SAR, and TUN) than 
rates of immediate mortality for blue sharks caught 
between March 1993 and August 2004 (Figs. 1 and 4). 
However, after the implementation in August 2004 of 
regulations for the mandatory use of circle hooks, the 
immediate mortality of blue sharks had a statistically 
significant decrease in every geographic zone analyzed 
when compared with the regulatory period between 
March 1993 and August 2004 (Figs. 1 and 4). Between 
the period March 1993-August 2004 and the period 
August 2004-December 2008, the rate of immediate 
mortality in geographic zones decreased by a mean of 
8.0% (standard error [SE] 0.5) and by a range of 4.4- 
10.1% for 9 geographic zones examined. 
Discussion 
Accuracy of logbook data 
Despite being limited to blue and tiger sharks, our 
analysis with the USPL data set provided useful in- 
formation about immediate mortality over a temporal 
and spatial scale not possible with other data sets. In 
previous studies, where USPL and POP data sets were 
compared, catch rates were determined to have been 
reported accurately for easily identifiable, commonly 
captured, and marketable species of sharks (blue, tiger, 
porbeagle, and oceanic whitetip sharks) (Mandelman 
et al., 2008; Baum and Blanchard, 2010). Our results 
show that the numbers of dead and live discards of 
bycatch species (i.e., blue and tiger sharks) recorded 
by commercial fishermen in the USPL data set are also 
similar to those recorded in the POP data set, indicat- 
ing accuracy, and that the USPL is a useful, massive 
source of long-term data with value for management of 
sharks and fisheries, particularly in relation to shark 
bycatch over time. Given the large sample size of the 
