16 
Fishery Bulletin 11 7(1-2) 
100°W 95° W 90° W 85° W 
Figure 1 
Spatial distribution of release locations for bluefin tuna (Thunnus thyn- 
nus) tagged with pop-up satellite archival tags in the pelagic longline 
fishery of the Gulf of Mexico from 2010 through 2015. Data for the num¬ 
ber of fish released are presented in 1° grids of density. The 2 open rect¬ 
angles indicate the areas closed to the use of pelagic longline gear during 
April and May. Sources for satellite image: Esri, DigitalGlobe, GeoEye, 
Earthstar Geographies, CNES/Airbus DS, USDA, USGS, AeroGRID, IGN, 
and the GIS user community. 
implementation of Amendment 7 to the Consolidated 
Atlantic Highly Migratory Species Fishery Manage¬ 
ment Plan (Federal Register, 2014), which created a 
system to assign individual quotas for bluefin tuna to 
vessels and closed 2 areas that cover the majority of 
the spawning habitat of bluefin tuna in the northern 
GOM to the use of PLL gear during the peak spawn¬ 
ing months of April and May (Fig. 1). The closure of 
these areas, in conjunction with the requirement to use 
weak hooks, has greatly reduced total discards of dead 
bluefin tuna. 
The effectiveness of management measures that 
require or promote release of fish hinges on 2 compo¬ 
nents of mortality associated with interactions with 
fishing operations. The first component of mortality is 
the fraction of fish dead at-vessel upon retrieval of the 
gear, and the second component is the fraction of fish 
that die after being released. At-vessel mortality and 
survival have been documented from commercial long- 
line fishing operations for several species of billfishes, 
tunas, and sharks (Serafy et al., 2012a; Walter et al., 
2012; Musyl et al., 2015) but have yet to be quantified 
for bluefin tuna in the GOM PLL fishery. Similarly, the 
second component, postrelease mortality from fishing 
operations, has been quantified for bluefin tuna from 
recreational fisheries (Marcek and Graves, 2014; Gold¬ 
smith et al., 2017) and for other spe¬ 
cies on PLL operations (Kerstetter et 
al., 2003; Musyl et al., 2011a) but has 
not been evaluated for bluefin tuna 
from the U.S. GOM PLL fishery oper¬ 
ating under normal fishing conditions. 
Both components of mortality are 
necessary to determine the total mor¬ 
tality associated with fishing interac¬ 
tions and to evaluate the efficacy of 
management regulations (Coggins et 
al., 2007) designed to promote release 
of live fish. Mortality from fishing op¬ 
erations can have a substantial effect 
on populations; therefore, it is critical 
to consider such mortality in popula¬ 
tion assessments (Musyl et al., 2015). 
For this study, we quantified both 
components of mortality associated 
with interactions of bluefin tuna with 
the U.S. GOM PLL fishery. We first 
examined the database of the NOAA 
Southeast Fisheries Science Center’s 
Pelagic Observer Program (POP) to 
determine an at-vessel mortality rate 
as a function of several covariates. 
Next we electronically tagged bluefin 
tuna caught incidentally by the U.S. 
PLL fishery in the GOM to obtain es¬ 
timates of postrelease mortality that 
apply to the fishery operating under 
normal commercial fishing practices. 
Fish were tagged from commercial 
fishing vessels, and all live fish cap¬ 
tured, regardless of apparent condition, were tagged. 
Finally, we combined the results of our tagging study 
with the proportion of bluefin tuna reported dead at- 
vessel from the POP database to determine an overall 
mortality estimate for interactions of bluefin tuna with 
PLL gear in the GOM. 
Materials and methods 
Examination of Pelagic Observer Program database 
The POP deploys NMFS-trained observers on a portion 
of PLL vessel trips to collect details on gear configura¬ 
tion, catch composition, and environmental conditions 
(for further details about observer protocols, see the 
training manual available from the Southeast Fisher¬ 
ies Science Center at website). To perform analyses 
similar to those used by Serafy et al. (2012a), we used 
a logistic regression to examine data for the influence 
of several key variables on the probability of mortality 
of 1498 bluefin tuna captured in the GOM PLL fishery 
during 1993-2017. For the logistic regression, we used 
the PROC GENMOD procedure in SAS/STAT 2 software 
2 Mention of trade names or commercial companies is for iden¬ 
tification purposes only and does not imply endorsement by 
the National Marine Fisheries Service, NOAA. 
