Shark interactions in pelagic longline fisheries

Abstract

Substantial ecological, economic and social problems result from shark interactions in pelagic longline fisheries. Improved understanding of industry attitudes and practices towards shark interactions assists with managing these problems. Information on fisher knowledge and new strategies for shark avoidance may benefit sharks and fishers. A study of 12 pelagic longline fisheries from eight countries shows that incentives to avoid sharks vary along a continuum, based on whether sharks represent an economic disadvantage or advantage. Shark avoidance practices are limited, including avoiding certain areas, moving when shark interaction rates are high, using fish instead of squid for bait and deeper setting. Some conventionally employed fishing gear and methods used to target non-shark species contribute to shark avoidance. Shark repellents hold promise; more research and development is needed. Development of specifically designed equipment to discard sharks could improve shark post release survival prospects, reduce gear loss and improve crew safety. With expanding exploitation of sharks for fins and meat, improved data collection, monitoring and precautionary shark management measures are needed to ensure that shark fishing mortality levels are sustainable.

Full text

Page 1

Author's personal copy
Marine Policy 32 (2008) 1–18
Shark interactions in pelagic longline fisheries
Eric Gilmana,b,c,?, Shelley Clarked, Nigel Brotherse, Joanna Alfaro-Shiguetof,
John Mandelmang, Jeff Mangelf, Samantha Petersenh, Susanna Piovanoi, Nicola Thomsonj,
Paul Dalzellk, Miguel Donosol, Meidad Gorenh, Tim Wernerg
aBlue Ocean Institute, USA
bSchool of Geography and Environmental Studies, University of Tasmania, Australia
c2718 Napuaa Place, Honolulu, Hawaii 96822, USA
dImperial College London, Silwood Park Campus, Manor House, Buckhurst Road, Ascot SL5 7PY, UK
e178 South Arm Drive, Wonga Beach, Queensland 4873, Australia
fPro Delphinus, Octavio Bernal 572-5, Lima 11, Peru
gNew England Aquarium, Central Wharf, Boston, MA 02110, USA
hBirdLife South Africa, P.O. Box 52026, Waterfront, Cape Town 8002, South Africa
iDipartimento di Biologia Animale e dell’Uomo, Universita ` di Torino, Via Accademia Albertina 13, I-10123 Torino, Italy
jEnvironment Consultants Fiji, Box 2041, Government Buildings, Suva, Fiji
kWestern Pacific Regional Fishery Management Council, 1164 Bishop Suite, Suite 1400, Honolulu, HI 96813, USA
lInstituto de Fomento Pesquero, Blanco 839, Valparaiso, V Region, Chile
Received 23 March 2007; received in revised form 22 May 2007; accepted 22 May 2007
Abstract
Substantial ecological, economic and social problems result from shark interactions in pelagic longline fisheries. Improved
understanding of industry attitudes and practices towards shark interactions assists with managing these problems. Information on fisher
knowledge and new strategies for shark avoidance may benefit sharks and fishers. A study of 12 pelagic longline fisheries from eight
countries shows that incentives to avoid sharks vary along a continuum, based on whether sharks represent an economic disadvantage or
advantage. Shark avoidance practices are limited, including avoiding certain areas, moving when shark interaction rates are high, using
fish instead of squid for bait and deeper setting. Some conventionally employed fishing gear and methods used to target non-shark
species contribute to shark avoidance. Shark repellents hold promise; more research and development is needed. Development of
specifically designed equipment to discard sharks could improve shark post release survival prospects, reduce gear loss and improve crew
safety. With expanding exploitation of sharks for fins and meat, improved data collection, monitoring and precautionary shark
management measures are needed to ensure that shark fishing mortality levels are sustainable.
r 2007 Elsevier Ltd. All rights reserved.
Keywords: Bycatch; Depredation; Finning; Fishery; Pelagic longline; Shark
1. Introduction
Bycatch1in marine fisheries is an increasingly prominent
international issue [1–16]. Bycatch raises ecological con-
cerns, as some bycatch species of cetaceans, seabirds, sea
ARTICLE IN PRESS
www.elsevier.com/locate/marpol
0308-597X/$-see front matter r 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.marpol.2007.05.001
?Corresponding author. Tel.: +18087225424; fax: +18089881440.
E-mail addresses: egilman@blueocean.org (E. Gilman), scclarke@
biznetvigator.com (S. Clarke), brothersbone@yahoo.com.au
(N. Brothers), prodelphinus@prodelphinus.org (J. Alfaro-Shigueto),
jmandelman@neaq.org (J. Mandelman), Seabirds@birdlife.org.za
(S. Petersen), susanna.piovano@unito.it (S. Piovano), watling@connect.
com.fj (N. Thomson), paul.dalzell@noaa.gov (P. Dalzell), mdonoso@
ifop.cl (M. Donoso), twerner@neaq.org (T. Werner).
1‘‘Bycatch’’ is the retained catch of non-targeted species or ‘‘incidental
catch’’, plus all discards [1, 17]. ‘‘Target’’ catch is the catch of a species or
species assemblage primarily sought in a fishery, while ‘‘non-target’’ catch
is the catch of a species or species assemblage not primarily sought.
‘‘Incidental’’ catch is the portion of non-target catch that is retained, while
‘‘discards’’ is the portion of non-target catch that is not retained [1, 17].

Page 2

Author's personal copy
turtles, sharks2and other fish species are particularly
sensitive to increased mortality above natural levels due to
their life history traits, including being long-lived, having
delayed maturity and having low reproductive rates
[9,12,18,19]. Bycatch can alter biodiversity by removing
top predators and prey species at unsustainable levels. It
also alters foraging habits of species that learn to take
advantage of discards [11,13]. Economic effects on fisheries
from bycatch include the imposition of a range of
restrictions, closed areas, embargos, and possible closures;
fishery interactions, where bycatch in one fishery reduces
target catch in another, and bycatch of juvenile and
undersized individuals of a commercial species, can
adversely affect future catch levels [13]. Discarded bycatch
is a social issue over waste [1, 4].
Depredation, the partial or complete removal of hooked
fish and bait from fishing gear, is conducted primarily by
cetaceans and sharks in pelagic longline fisheries (Fig. 1).
Economic losses from depredation can be substantial [20,
21]. Depredation also raises ecological concerns as these
interactions may change cetacean and shark foraging
behavior and distribution, increase fishing effort, and
confound fish stock assessments as well as result in
deliberate injury and mortality of cetaceans and sharks
by fishers to discourage depredation and avoid future
interactions [11].
Much progress has been made to identify effective,
commercially viable, and even operationally beneficial
methods to significantly reduce seabird and sea turtle
bycatch in longline fisheries [8, 12, 22–24]. Relatively little
progress has been made to reduce cetacean [11] and shark
interactions in longline fisheries.
In some pelagic longline fisheries, shark interactions pose
substantial ecological, economic and social problems. As
demonstrated in some fisheries to address seabird and sea
turtle bycatch [8, 9, 12, 24, 25], collaborative approaches,
which tap fishers’ large repository of knowledge, may
likewise successfully reduce unwanted shark interactions.
We collect information from longline industries ranging
from small-scale artisanal fisheries to large-scale industrial
distant water fleets to obtain a more complete under-
standing of shark-pelagic longline interactions, current
fisher attitudes and practices employed in response to shark
interactions, identify methods to avoid shark interactions,
identify research priorities and assess the effects of
legislation that affect longline practices in catching
and processing sharks. Information on existing fisher
knowledge and new strategies for shark avoidance may
benefit sharks and fishers wanting to reduce shark
interactions. Improving the understanding of longline
industry attitudes and practices towards shark interac-
tions provides industry and management authorities with
better information to address these problems. A detailed
project report from this study has been produced by
Gilman et al. [26].
2. Methods and overview of fisheries
Information was collected from the following 12 pelagic
longline fisheries from eight countries: (i) Australia longline
tuna (Thunnus spp) and billfish (Istiorphoridae spp) fishery,
(ii) Chile artisanal mahi mahi (dolphinfish) (Coryphaena
spp) and shark fishery, (iii) Chile swordfish (Xiphias
gladius) fishery, (iv) Fiji longline tuna fishery, (v) Italy
Mediterranean industrial longline swordfish fishery, (vi)
Japan distant water longline fishery, (vii) Japan offshore
longline fishery, (viii) Japan nearshore longline fishery, (ix)
Peru artisanal mahi mahi and shark fishery, (x) South
Africa longline tuna and swordfish fishery, (xi) US Hawaii
longline tuna fishery, and (xii) US Hawaii longline
swordfish fishery. From January to December 2006, 149
vessel captains, fishing masters, crew, vessel and company
owners, fishing cooperative staff and port officials from
these 12 fisheries were interviewed at 24 fishing seaports
(nine seaports in Australia, including the main port of
Mooloolabah; Arica, Iquique and Valparaiso, Chile; Suva,
Fiji; Sicily, Italy; Kesennuma, Kii-Katsuura, Yaizu and
Misaki, Japan; Ilo, Paita and Salaverry, Peru; Cape Town
Harbour, Hout Bay Harbour and Richards Bay Harbour,
South Africa; and Honolulu, USA).
Information from the interviews, analyses of available
logbook and observer data, and a review of the literature
were collected and analyzed to:
? Determine shark catch rates, disposition of caught
sharks and costs and benefits from shark interactions
to better understand longline industry interest in
reducing shark interactions;
? Describe the range of longline industry attitudes
towards shark capture and depredation to understand
the degree of interest in shark avoidance;
ARTICLE IN PRESS
Fig. 1. Shark-damaged yellowfin tuna caught in the Hawaii pelagic
longline fisheries (photo courtesy of US National Marine Fisheries Service
Hawaii Pelagic Longline Observer Program).
2The term ‘sharks’ refers to the Chondrichthyan fishes, which comprise
elasmobranchs (sharks, skates and rays) and holocephalans (chimaeroids).
E. Gilman et al. / Marine Policy 32 (2008) 1–18
2

Page 3

Author's personal copy
? Identify practices employed by longline fishers in
response to shark interactions;
? Identify promising concepts not currently practiced to
reduce shark capture, reduce depredation and gear
damage, improve discard methods, and determine what
obstacles must be overcome to implement these concepts;
? Identify priority research and development, monitoring
and management measures; and
? Identify economic, social and ecological effects of
legislation affecting shark practices, assess if the
legislation has resulted in reduced interest in capturing
and retaining sharks, and discuss how these laws may
have affected shark fishing mortality levels.
The 12 fisheries range from small-scale domestic
artisanal fisheries to modern mechanized industrial fleets
of distant water fishing nations. Distant water vessel fishing
grounds range throughout the world’s oceans on trips
lasting two to three months, while smaller vessels fish in
nearshore waters on trips lasting a few days. The number
of vessels in each fleet also varies, from the South Africa
longline fleet with about 17 vessels, to the Japanese and
Peruvian fleets with about 1500 vessels each. Some of these
fisheries never target and rarely retain sharks, while in
other fisheries sharks are an important target species.
There are several fishing gear designs and operational
characteristics that are likely to affect shark interactions,
including the location of fishing grounds, depth of baited
hooks, timing of gear deployment and retrieval, use of wire
leaders and lightsticks on branch lines and type and size of
bait. For example, Peru artisanal longline vessels, which
are about 15m in length, target mahi mahi during the
austral summer and target sharks from autumn to spring.
Baited hooks are set at depths between 10 and 16m. Wire
leaders are not typically used during the mahi mahi season
but are always used during the shark season to maximize
shark retention and reduce gear loss. Giant squid, mackerel
and flying fish are used for bait. Lightsticks are not used.
Gear soaks during the daytime. Hawaii longline tuna
vessels are a bit larger, between 15 and 31m in length, set
baited hooks deeper at depths between 35 and 224m, use a
wire trace, use fish for bait, do not use lightsticks, and gear
soaks during the day. However, there may be large
variability in fishing gear and methods between vessels in
a fleet and even for an individual vessel. For instance, some
vessels in the Fiji longline tuna fleet fish at grounds within
the Fiji Exclusive Economic Zone (EEZ), while larger
vessels fish at grounds much more distant from their home
port, on the high seas and in other nation’s EEZs, and
these two categories of vessels have substantially different
gear characteristics. In some fisheries, vessels will substan-
tially alter their gear seasonally when they change their
primary target species (e.g., Chile and Peru artisanal
longline mahi mahi and shark fisheries, Japan offshore
and nearshore pelagic longline tuna fisheries). Gear
characteristics may also vary substantially between sea-
ports within a fishery.
3. Shark catch rates and disposition
Table 1 summarizes the catch rates and retention of
caught sharks, where available. Several of these entries
are based on limited data from small sample sizes. For
the Japan distant water, offshore and nearshore longline
fisheries, the shark catch rates reported in Table 1 are based
on logbook data. These figures’ relationship to the actual
number of sharks caught and retained is expected to vary
with logbook recording behavior [27]. The Italy Mediter-
ranean large-scale longline swordfish fishery is the only
fishery included in this study where there is a lack of a local
market for shark fins, and as a result, fishers do not fin
sharks. The shark catch rates for the two Chile fisheries are
estimated from fisher interviews. Shark catch rates for the
Chile longline swordfish fishery and artisanal longline mahi
mahi and shark fishery are available only in units of weight
per unit of effort (0.36 and 0.28kg/hk, respectively [28]),
and not as number of sharks per unit of effort. Information
on the Fiji longline tuna fishery is based on observer data
ARTICLE IN PRESS
Table 1
Shark catch rate and disposition in 12 pelagic longline fisheries for the
most current year for which data are available
Pelagic longline fisheryShark catch
rate (number
per 1000
hooks)
Shark retention
(fins and/or
carcass) (% of
total number
caught sharks)
Australia tuna and billfish
longline fishery
5.5a
Not available
Chile artisanal mahi mahi and
shark longline fishery
24b
4 99b
Chile longline swordfish fishery8b
4 99b
Fiji longline tuna fishery 1.178–90
Italy mediterranean industrial
longline swordfish fishery
0.74Not available
Japan distant water longline
tuna fishery
0.021c
Not available
Japan offshore longline fishery0.175c
Not available
Japan nearshore longline
fishery
0.020c
Not available
Peru artisanal longline mahi
mahi and shark fishery during
mahi season
0.9984
South Africa longline tuna and
swordfish fishery
4.080
USA-Hawaii tuna2.22.1
USA-Hawaii swordfish16.70.2
aRough estimate based on Australian Commonwealth Scientific and
Research Organization, unpublished data from a subset of the fleet and
time period, possibly not representative.
bRough estimate based on interview responses.
cBased on number of sharks recorded in vessel logbooks [27].
E. Gilman et al. / Marine Policy 32 (2008) 1–18
3

Page 4

Author's personal copy
from 1999 and 2002 to 2005 (Secretariat of the Pacific Com-
munity, unpublished data). Statistics for the Peru artisanal
longline mahi mahi and shark fishery are based on 2004–2006
onboard observer data taken only during the mahi mahi
season from four ports for a total of 27 trips and 197 sets.
The large number of ‘‘not available’’ entries and entries
based on rough estimates in Table 1 suggests either that there
is insufficient data collection and management measures for
shark species or that relevant data are collected but have not
been analyzed. For fisheries where there is high confidence in
available shark catch rates, these range from 0.7 to 17 sharks
per 1000 hooks. The location of fishing grounds and
characteristics of fishing gear and methods are likely primary
factors determining a fleet’s shark catch rate. Certain gear
designs (e.g., use of a wire leader, use of squid for bait, use of
lightsticks, and setting baited hooks at shallow depths)
contribute to high shark catch rates.
The proportion of total catch comprised of sharks by
number varies widely for the fisheries. In the Australia
fishery, sharks comprise about 27% of the total catch.
Sharks comprised 4 25% of the total number of fish
caught by Fiji longline tuna vessels based on observer data
from 1999 [29], while Secretariat of the Pacific Community
observer program data for 1999 and 2002–2005 found that
sharks comprised only 5.5% of the total number of caught
fish. From 1998 to 1999, sharks comprised about 18% of
the total catch in the Italian longline swordfish fishery [30].
In the Peru artisanal longline fishery, during the mahi mahi
season in the port of Ilo for 2005–2006, sharks comprise
less than 1% of the total catch by number. In the South
Africa longline fishery, from 1998 to 2005, sharks
comprised 16.2% of the total number of caught fish. In
2001, pelagic sharks comprised about 50% of the catch
composition of swordfish sets and 16% for tuna sets in the
Hawaii longline fishery [31]. However, since 2004 the shark
catch rate in the swordfish fishery dropped 36% when the
fishery was required to switch from using J hooks with
squid bait to wider circle hooks with fish bait [8].
Results are generally consistent with the literature, which
shows that a large quantity of pelagic sharks is taken as
bycatch in pelagic longline fisheries with tuna and
swordfish as their primary target species [29, 32–36]. For
example, in the western Pacific, shark species account for
the highest category of bycatch in tropical fisheries, where
sharks comprise 27% of total bycatch, and in subtropical
fisheries, where sharks are 18% of total bycatch [32, 37]. In
the US Atlantic longline swordfish and tuna fisheries,
sharks and rays constituted 25% of total catch between
1992 and 2003 [38]. Beerkircher et al. [33] found that sharks
comprised 15% of the total catch in the southeastern US
pelagic longine swordfish and tuna fisheries. Bonfil [18]
found that the same numbers of sharks are caught in
directed fisheries as are caught as bycatch mostly in
longline tuna fisheries. However, the recent development
of longline directed shark fisheries, especially in the Pacific,
may mean that directed shark fisheries are now catching
more sharks [39–42].
For fisheries where information on shark catch composi-
tion is available, blue sharks comprise the largest propor-
tion of shark catch. Blue sharks comprise 47% of total
shark catch by number of fish in the Australia longline
tuna and billfish fishery; 49% in the Fiji longline tuna
fishery; X70% in Japan longline fisheries; 57% for the
Peru artisanal mahi mahi and shark longline fishery (Pro
Delphinus, unpublished data); 69% for the South Africa
longline tuna and swordfish fishery; and 82% and 92% for
the US Hawaii longline tuna and swordfish longline
fisheries, respectively.
Identifying effective and commercially viable methods to
reduce unwanted shark bycatch in longline fisheries would
contribute to reducing shark fishing mortality. Increasing
the proportion of caught blue sharks that are discarded in
pelagic longline fisheries would likely reduce fishing
mortality of this species, as blue sharks are usually alive
when hauled to the vessel. Beerkircher et al. [33] found that
the condition of sharks caught in pelagic longline gear
(dead versus alive when hauled to the vessel) varied widely
by species, where, for example, blue sharks had a relatively
low 12.2% mortality, while silky sharks (the most
dominant species of shark by number caught in the
observed southeastern US pelagic longline swordfish and
tuna fisheries, 31.4% of elasmobranch catch) had a 66.3%
mortality. Over 89% of sharks caught in the Hawaii-based
longline swordfish fishery and over 93% of sharks caught
in the Hawaii-based longline tuna fishery are alive when the
gear is retrieved. Eighty-seven percent of sharks caught in
the Peru artisanal mahi mahi and shark longline fishery
were alive when gear was retrieved. An analysis of
Secretariat of the Pacific Community observer program
data from 2002 to 2005 for the Fiji longline tuna fishery
indicates that over 94% of blue sharks and over 84%
of combined species of sharks were alive when hauled to
the vessel.
4. Summary and effects of national/EC legislation on shark
interactions
Table 2 summarizes legally binding measures that
influence longline industry practices and attitudes towards
shark bycatch and depredation. The two Chile longline
fisheries, Fiji longline fishery, and three Japan longline tuna
fisheries are not subject to legally binding measures that
manage shark interactions, and are not included in Table 2.
Japan and Fiji distant water vessels may comply with
voluntary measures adopted by Regional Fishery Manage-
ment Organizations, and vessels operating in EEZs of other
nations through foreign license access agreements may be
required to comply with restrictions on shark catch,
retention and use under these agreements.
Legislation prohibiting the removal of shark fins and tail
and discarding the remainder of the shark at sea in pelagic
longline fisheries exists in four of the eight countries
included in this study (Australia, Italy, South Africa, and
USA) [43–46]. In the Australia longline tuna and billfish
ARTICLE IN PRESS
E. Gilman et al. / Marine Policy 32 (2008) 1–18
4

Page 5

Author's personal copy
fishery, a rule that disallows possession, carrying and
landing of shark fins unless attached to the trunk of the
shark has likely substantially reduced shark fishing
mortality, as finning was a widespread practice before this
measure was instituted, while about 75% of caught sharks
are now released alive [47, 48]. In the Hawaii longline tuna
and swordfish fisheries, observer data indicate that the
restriction on shark finning, which requires the retention of
shark carcasses for corresponding retained fins, has like-
wise substantially reduced shark fishing mortality. As
many as 76% and 64% of caught sharks were finned in
the Hawaii tuna and swordfish fisheries, respectively, prior
to this rule, while in 2006, 91% and 93% of caught sharks
were released alive in the tuna and swordfish fisheries,
respectively. Revenue from shark fins had comprised
10–11% of Hawaii longline crew salaries [49]. In the South
Africa longline tuna and swordfish fishery, all interviewed
fishers stated that prior to the finning restriction, they
would fin and discard the carcass of all caught sharks
excluding makos, which were retained for the sale of
both fins and meat. Thus, the restriction on finning in
South Africa has also substantially reduced shark retention
and increased discards. In these fisheries, shark finning
restrictions have caused substantial reductions in revenue
to industry.
All 17 interviewed owner–operators from the Italy
Mediterranean industrial longline swordfish fishery were
unaware of a restriction on shark finning practices: the
legislation does not affect their practices. However, no
shark finning is reported to occur in the fishery due to the
lack of a local market for fins.
Japan does not have legislation restricting shark finning
practices, however, the distant-water fleet fishes in EEZs of
nations that do have finning restrictions (e.g., South Africa,
Brazil, Costa Rica). Vessels in the Japan distant-water
longline tuna fishery will likely fin caught sharks and
discard the carcass unless they are fishing in the EEZ of a
nation that prohibits this practice, in which case the vessel
may choose to retain the whole shark carcass and land the
carcass in ports where there are markets for shark meat.
Thus, Japanese longline fishermen have adapted to finning
regulations applicable in some areas by landing sharks in
recently developed local markets. In waters without finning
regulations, including Japanese waters and the North
Pacific, sharks are either finned or landed whole, and in
either case the ability to sell shark products has contributed
to a lack of interest in reducing shark bycatch.
A 20 shark carcass per trip limit for the retention of
sharks in the Australia East coast longline tuna and billfish
fishery has not altered the number of sharks retained by
fishers, as fewer than 20 sharks are typically caught during
an average length trip, and only a small proportion of the
sharks caught on a trip are of species (makos and
threshers) for which there is sufficient value for their meat.
Furthermore, many operators only retain a shark of a
marketable species if it is dead or dying when hauled to the
vessel, which can be safely and easily landed.
The South Africa longline tuna and swordfish fishery is
subject to a shark landing limit of 10% of the total
swordfish and tuna catch. In theory, this has been
economically detrimental to the industry. From 1998 to
2005, the total number of caught sharks was 18% of the
total number of caught swordfish and tunas. Thus, vessels
would need to discard about 44% of caught sharks to
comply with the shark limit. Because only about 18% of
caught blue sharks and 10% of makos have been observed
to be released alive in this fishery, and the literature
demonstrates that a much larger proportion of these shark
ARTICLE IN PRESS
Table 2
Legal framework that influence practices and attitudes towards shark bycatch and depredation in six pelagic longline fisheries [43–46]
Legal constraintsa
Pelagic longline fisheries by flag stateRetention of fins
requires retention of
corresponding carcassb
Shark
retention
limitc
Prohibit
wire trace
Prohibit retention
of specified shark
species
Size
Limit
Australia tuna and billfishXXXX
Italy Mediterranean industrial swordfishX
Peru artisanal mahi mahi and sharkX
South Africa tuna and swordfishXXX
USA-Hawaii tunaX
USA-Hawaii swordfishX
aJapan and Fiji distant water longline tuna vessels may comply with voluntary measures adopted by Regional Fishery Management Organizations, and
vessels operating in EEZs of other nations through foreign license access agreements may be required to comply with restrictions on shark catch, retention
and use under these access agreements.
bUSA, Italy (European Union), and South Africa require the total weight of retained shark fins to be p5% of the total dressed ‘‘live’’ weight of shark
carcasses. Australia requires fins to be attached to the shark carcass when landed.
cAustralia has a 20 shark carcass per trip retention limit for longline tuna and billfish fisheries. South Africa has a shark landing limit of 10% of the total
swordfish and tuna catch.
E. Gilman et al. / Marine Policy 32 (2008) 1–18
5

Page 6

Author's personal copy
species (4 85%) are likely alive when hauled to the vessel
(e.g., [33]), it is likely that fishers are not complying with
this measure.
A prohibition on the use of a wire trace in the Australian
longline tuna and billfish fishery and South Africa longline
tuna and swordfish fishery has likely resulted in an
increased economic cost from shark interactions as this
has likely caused an increase in the loss of terminal tackle
to sharks. A substantially larger number of hooks, bait and
line are likely now bitten off of branch lines compared to
when wire trace was used. However, fishers generally do
not consider this to be a large concern. It is not known how
the injury to sharks from retaining hook and trailing
monofilament line affects their survival prospects. This is a
research priority. This may be an improvement to their
previous fate when caught on lines with wire trace when
they would soak on the gear for hours, be gaffed and
hauled onboard the vessel and then have hooks removed by
cutting with a knife or pulled out by force. Available but
limited information indicates that a large proportion of
sharks caught in longline gear that are released after
removal of the hook will survive (Section 7.4; [50]). In
fisheries where a large proportion of caught sharks are
killed either for retention or discarding, prohibiting the use
of wire leaders will reduce shark fishing mortality.
Prohibiting wire leaders may exacerbate seabird bycatch
problems. Fishers will be less likely to attach weights close
to hooks on branch lines lacking a wire leader due to safety
concerns, thus, reducing the baited hook sink rate, and
increasing seabird catch rates.
Shark fisheries in Peru are regulated by the Ministry of
Fishery through size limits for certain elasmobranch
species. However, there is little enforcement of these
regulations and few fishers are aware of the regulations.
Of the few interviewed fishers who reported that they were
aware of the regulations (5%), all report that they still
retain sharks that are under the minimum size limit.
5. Economic, practical, ecological and social problems from
shark–longline interactions
5.1. Economic and practical concerns
Shark interactions in pelagic longline fisheries result in
substantial inconveniences and adverse economic effects,
including:
(i) Depredation: Lost revenue from shark damage to
target species can amount to several thousand US
dollars in a single set in some fisheries.
(ii) Damage and loss of gear: Sharks bite off terminal
tackle (e.g., baited hook, leader, weighted swivel, and
line) from branch lines, stretch and chafe branch lines,
break the main line, and some shark species will pull
the gear down causing branch lines to entangle.
(iii) Reduced catch of marketable species: When baited
hooks are occupied or removed by sharks, fewer
hooks are available to catch marketable non-shark
species.
(iv) Risk of injury: It is dangerous to handle caught sharks
and there is a risk of being hit by weights when branch
lines containing sharks break during gear retrieval.
(v) Expenditure of time: Substantial time is spent removing
sharks from gear, retrieving terminal tackle and
repairing and replacing gear due to shark interactions.
In fisheries with a demand for shark products, where
vessels continuously or periodically target sharks, fisher-
men generally perceive these costs to be a minor incon-
venience and are not problematic enough to create an
incentive to avoid sharks. However, in fisheries with
restrictions on finning, a lack of market for shark meat
or a per-trip limit on shark retention, where shark catch
rates are relatively high, shark interactions are perceived to
be a major inconvenience and can represent a substantial
economic cost.
In the Australia longline tuna and billfish fishery,
fishermen estimate that they lose 20% of their catch of
target species due to shark damage, while damage and loss
of gear from shark interactions amount to a loss of about
AUD 100 per set. Considerable time is also expended to
discard caught sharks. The average catch rate of sharks is
about 5.5 sharks per 1000 hooks compared to the catch
rate of target and incidental fish of about 20.5 fish per 1000
hooks. However, on a given set, the shark catch can be
extremely high (hundreds of sharks).
Fishers in the Chile mahi mahi and shark fishery and
swordfish fishery report that sharks are an important target
or incidental catch species. Fishers perceive that revenue
from catching sharks exceeds costs from shark interactions.
In a typical mahi mahi set, costs from the loss and damage
to gear is about USD 18.5 and in the swordfish fishery
50–100 branch lines are damaged from shark interactions
on a typical set. Fishermen reported having an average of
six mahi mahi and three swordfish damaged from shark
depredation on a typical set in the artisanal mahi mahi
fishery and swordfish fishery, respectively. This represents a
loss of about USD 146 per mahi mahi set and USD 1063
per swordfish set.
In the Fiji longline tuna fishery, almost all caught sharks
are finned (Table 1) and carcasses are usually discarded due
to the low value of shark meat. Fishers generally perceive
that costs from shark interactions, including economic
costs and time spent to deal with the interactions, exceed
the benefit from revenue to crew from fins.
In the Italian Mediterranean industrial longline sword-
fish fishery, where the shark catch rate is low and sharks
are occasionally retained for the sale of meat, fishermen
find the costs of shark interactions to be a minor
inconvenience. Few (0–10) branch lines are damaged or
lost to sharks on a typical set, and very rarely is a target
species damaged by sharks. At most, two target species are
damaged by sharks per set. However, despite the perceived
low frequency of shark interactions and nominal economic
ARTICLE IN PRESS
E. Gilman et al. / Marine Policy 32 (2008) 1–18
6

Page 7

Author's personal copy
cost from shark interactions, fishermen perceive that the
revenue from catching sharks is exceeded by costs from
shark interactions, and there is concern over the safety risk
of handling caught sharks. As a result, fishers are interested
in reducing shark interactions as long as this does not
adversely affect their catch rate of target species.
In Japanese longline fisheries, where fins are retained
from the majority of caught sharks and in some cases
carcasses are retained for their meat, costs of shark
interactions are perceived to be minor. Gear damage and
loss from shark interactions are considered a much less
important problem than shark damage to hooked tunas
and billfishes, which can result in the damage of as many as
three fish per set, where shark depredation of one fish every
3–5 sets is more typical.
Fishers in the Peru artisanal mahi mahi and shark
longline fishery report that the revenue from catching
sharks exceeds costs from shark interactions. Sharks are an
important incidental catch species during the mahi mahi
season and the main target species during the remainder of
the year. Fishers estimate that they incur a cost of USD 11
per set due to damage and loss of gear, and incur a loss of
about USD 30 from 7 to 8 mahi mahi being damaged from
sharks on a typical set during the mahi mahi season.
In the South Africa longline tuna and swordfish fishery,
fishers are concerned over shark damage to their gear and
the loss of bait from shark interactions. On average, they
lose the terminal tackle of 10–30 branch lines, although this
is highly variable from set to set. On typical sets, 2–5
marketable fish are damaged or lost to sharks.
In the US Hawaii longline swordfish fishery, 4 99% of
caught sharks were discarded in 2006 when the shark catch
rate was 16.7 sharks per 1000 hooks and the catch rate of
retained fish was 23 fish per 1000 hooks. In the Hawaii
longline tuna fishery, 4 97% of caught sharks were
discarded in 2006 when the shark catch rate was 2.2 sharks
per 1000 hooks and the catch rate of retained fish was 13
fish per 1000 hooks [9]. In these fisheries, fishers perceive
the time required to remove sharks from gear and to
rebuild damaged and lost gear to be a substantial
inconvenience. Risk of injury from caught sharks is also
a substantial concern. Economic costs from the damage
and loss of gear are nominal, costing an estimated USD 19
and USD 50 per typical tuna and swordfish set, respec-
tively. Fishers report having an average of three market-
able fish species damaged from sharks on a typical longline
tuna set and five damaged on a typical swordfish set. This
can represent a loss of several thousand dollars. This is a
much higher rate of depredation of caught fish than
reported by Lawson [20] for Western and Central Pacific
Ocean pelagic longline tuna fisheries, where 1.8% (46 of
2555) of caught tunas were discarded due to shark damage.
On an especially bad set, as many as 50% of target species
may be damaged to a degree that they cannot be sold.
Many pelagic longline fisheries targeting species other
than sharks, when not prevented by regulation, will retain
the fins of captured sharks, which fetch a high value in the
Asian dried seafood trade, and occasionally will retain
meat and other parts (cartilage, liver oil, skin) from
marketable species of sharks when markets for these
products are available (e.g., [34, 36, 49, 51]. High demand
for shark fins in Asia means that few sharks caught in
pelagic longline fisheries, where finning is not prohibited or
resources for enforcement are scarce, are released alive [34,
36]. For instance, from 1995 to 1999, before restrictions on
shark finning were instituted, the Hawaii-based longline
swordfish fishery finned 65% of caught sharks, when about
50% of the catch by number was elasmobranch bycatch. In
the Fiji longline tuna fishery, 78–90% of caught sharks are
finned (Secretariat of the Pacific Community, unpublished
data). Francis et al. [34] found that about half of the catch
by number on New Zealand tuna longlines was elasmo-
branch bycatch, from which usually only the fins were
retained. Williams [36] found that in western and central
Pacific longline tuna fisheries, the fate of shark bycatch was
species-specific: Certain species, such as pelagic stingray,
were always discarded, trunks of silky and blue shark were
occasionally retained (45.8% and 5.4% of the time,
respectively), fins of blue sharks were retained 84.1% of
the time, and fins of silky sharks were retained 47.5% of
the time. Crew in many pelagic longline fisheries have a
strong economic incentive to catch sharks and fin as many
of the sharks that are caught as possible as they receive the
proceeds from shark fins [36, 49]. For instance, Williams
[36] reported that crew of some longline tuna vessels
operating in the western and central Pacific obtain half of
their wage from shark fin revenue. McCoy and Ishihara
[49] estimated that Hawaii longline crew had obtained over
10% of their annual wage from shark fin sales, prior to the
promulgation of rules restricting finning practices. In some
fisheries, shark discarding and retention practices are also a
result of the value of the species of caught shark, whether
the shark is caught at the beginning or end of a fishing trip,
how much hold space remains, whether or not the shark
is alive or dead when hauled to the vessel and the size
of the shark.
However, to address the social concern that shark
finning is wasteful when a large portion of the shark is
discarded, and ecological concerns over the sustainability
of shark exploitation in fisheries, there have been several
recent international initiatives3and adoption of national
legislation addressing shark finning (Section 4 Table 2).
Fisheries that are required to retain and land entire shark
carcasses if they wish to retain the fins have a high
economic incentive to avoid shark bycatch in areas where
there is a lack of markets for shark meat. Some fisheries
may lack access to markets for shark products, as
documented in Italy, creating an incentive to avoid
catching sharks.
ARTICLE IN PRESS
3For instance, the International Commission for the Conservation of
Atlantic Tunas, Inter-American Tropical Tuna Commission and North
Atlantic Fisheries Organization have adopted legally binding measures to
prohibit shark finning practices.
E. Gilman et al. / Marine Policy 32 (2008) 1–18
7

Page 8

Author's personal copy
There are pelagic longline fisheries where revenue from
sharks exceeds costs from shark interactions, a large
proportion of caught sharks are retained, and sharks are
either always an important target species, are targeted
seasonally or at certain fishing grounds proximate to ports
where there is demand for shark fins and meat, or are an
important incidental catch species. For instance, sharks
comprised 70% of landings by the Spanish North Atlantic
and Mediterranean longline swordfish fishery in 1991–1992
based on sampling at the Algeciras fish market in southern
Spain [52]. While the majority of pelagic longline fisheries
target tunas and billfishes [53], as documented in this study
(Chile, Peru, Japan) there are a growing number of pelagic
longline fisheries where the main target species are pelagic
or coastal sharks [19, 39, 40–42]. While some directed shark
fisheries are large industrial practices, the majority of shark
catches comes from small-scale primarily gillnet fisheries
[19, 54]. Chondrichthyan fisheries have substantially grown
in developing countries over the past several decades.
Developing countries’ shark catches increased from 76,000
to 575,031 metric tons from 1950 to 2000 for a value in the
year 2000 of USD 515 million [39, 41]. From 1985 to 2000,
elasmobranch catches reported to the Food and Agricul-
ture Organization of the United Nations increased
annually by an average of 2% [55]. However, actual
elasmobranch catches are likely much higher than reported
due to a lack of accurate data collection programs and to
purposeful underreporting [51, 56].
Results from this study reveal that there has been a large
increase in the demand for shark fins and meat and catch of
sharks over the past several decades [36, 41, 49]. For
instance, shark catch by weight in Chile fisheries has
increased an order of magnitude from about 1000ton in
1950 to over 10,000ton in 2005 [41]. Also, the shark catch
in Peruvian fisheries and export market for frozen shark
meat have grown, where the revenue from shark meat
exceeds revenue from fins on a per-trip basis for a vessel in
the Peru artisanal mahi mahi and shark longline fishery.
From 2000 to 2005, exports of shark meat from Peru
tripled, with main export markets including Uruguay,
Spain, Brazil and Colombia [57]. The Japan component of
this study has identified a trend in expanding demand for
shark meat in a few regions in Japan where offshore and
nearshore vessels land their catch as well as at several
foreign seaports where distant water longline vessels land
their catch, including Cape Town (South Africa), Callao
(Peru), Las Palmas (Spain), Balboa (Panama), Cartagena
(Venezuela) and Port Louis (Mauritius), and a concomi-
tant increase in retention and landing of shark carcasses by
Japan longline fisheries. The shark meat landed in Callao,
Cape Town and Las Palmas may be exported to European
markets in Italy and Spain.
5.2. Ecological concerns
There is an ecological basis for concern over shark
interactions in pelagic longline fisheries. In the last decade,
as elasmobranch catches have increased in both directed
and incidental fisheries, there has been increasing concern
about the status of some shark stocks, the sustainability of
their exploitation in world fisheries, and ecosystem-level
effects from shark population declines [4, 19, 58–60].4
Most shark species are predators at the top of the food
chain and characterized by relatively late maturity, long
life, slow growth, low fecundity and productivity (small
and infrequent litters), long gestation periods, high natural
survivorship for all age classes, and low abundance
(K-selected life history strategies) relative to bony fish
such as tunas and billfishes and to organisms at lower
trophic levels [66]. Some shark species may also aggregate
by sex, age and reproductive stage [37, 67]. These life
history characteristics make sharks particularly vulner-
able to overexploitation and slow to recover from
large population declines [19, 67]. Directed shark fish-
eries in North America provide examples of overfishing
and population declines, such as occurred in directed
fisheries for the porbeagle (Lamna nasus) [68], soupfin
shark (Galeorhinus zyopterus) [69], and spiny dogfish
(Squalus acanthias) [70]. Also, for example, the lack of
monitoring of primarily discarded bycatch of the barn-
door skate(Dipturus laevis) in the western
Atlantic bottom trawl fisheries resulted in a large popula-
tion decline [71].
The main threats faced by chondrichthyans are various
fishing activities and habitat degradation and loss [72].
Reviews of assessments of the threatened status of sharks
and related taxa undertaken to date indicate that the taxa
at highest risk include commercially exploited species of
deepwater sharks, species restricted to freshwater and
brackish water habitats and coastal endemics whose entire
range overlaps with fishing effort [73]. However, a lack of
both fundamental biological information and fishery-
dependent data for most shark species [67, 71] means
that there is a high degree of uncertainty in the status of
these species. The biology of the chondrichthyan fishes is
the least understood of all the major marine vertebrate
groups, wheredetailedinformation
and reproductive dynamics is not available for all but a
few of species important for directed fisheries [67]. There
is a general lack of reliable and sufficiently detailed
fishery-dependent data on shark species to enable sustain-
able management [71, 74]. Pelagic longline fisheries
operating on the high seas are not likely interacting with
these shark species identified as highest-risk, while some
North
onlife history
ARTICLE IN PRESS
4Atlantic blue sharks are among those species reported to have
undergone considerable population declines [58]. Consistent with previous
arguments against treating CPUE as an index of abundance [61], the
reported blue shark decline (?60% since 1986) postulated by Baum et al.
[58] has been questioned by several authors (e.g. [62–64] on the basis that
potential reasons for drops in CPUE aside from abundance declines were
not accounted for, such as underreporting, changes in fishing grounds and
changes in fishing gear such as not using wire leaders [65]. It is
acknowledged, however, that the species has likely endured some level
of decline in recent years [64, 65].
E. Gilman et al. / Marine Policy 32 (2008) 1–18
8

Page 9

Author's personal copy
coastal pelagic longline fleets could be catching at-risk
coastal endemics. In particular, blue sharks (Prionace
glauca), the dominant species of shark caught in most
pelagic longline fisheries operating on the high seas (e.g.,
[34, 36]), are less vulnerable to overfishing relative to other
shark species due to their being relatively prolific and
resilient [75, 76]. Blue sharks comprise the largest propor-
tion of shark species caught in all 12 of the fisheries
included in this study, ranging from 47% to 92% of shark
catch in fisheries where this information is available.
Kleiber et al. [77] conducted a stock assessment of blue
sharks in the North Pacific and concluded that blue sharks
are not being overfished in the North Pacific. However,
more recent research by Clarke et al. [51] suggests that blue
sharks globally are being captured at levels close to
or possibly exceeding maximum sustainable yield. Clarke
et al. [51] estimated global shark catches using shark
fin trade records, and found that shark biomass in
the fin trade is three to four times higher than shark catch
figures reported by the Food and Agriculture Organization
of the United Nations, which is the sole existing global
database. Additional stock assessments for other pelagic
sharks have been conducted only by the International
Commission for the Conservation of Atlantic Tunas
for blue and shortfin mako sharks in the North and
South Atlantic.
5.3. Social concerns
Shark finning, where fins from caught sharks are
retained and the remainder of the carcass is discarded,
raises the social issue of waste. This has received recent
international and national attention (Section 4). Concern
has also been raised that finning practices are cruel to
sharks based on the presumption that fishers remove fins
and discard sharks alive. However, results from this
study document that, in all fisheries where shark finning
occurs, to avoid injury and increase efficiency, crew first kill
the fish before removing fins, and do not remove fins
from live sharks. Also, discarded bycatch in general
raises the social issue of waste [1, 4], however, in the
case of shark discards, available information suggests
that in pelagic longline fisheries, shark post release
survival prospects are high [50] and most sharks caught
in pelagic longline fisheries are alive when hauled to
the vessel [36].
6. Industry attitudes
Table 3 summarizes predominant attitudes related to
shark interactions held by fishers of 12 pelagic longline
fisheries. The existence of restrictions on shark finning and
shark retention limits (Table 2) has a large influence on
industry attitudes towards shark interactions in the
Australia, South Africa and Hawaii longline fisheries,
where legal constraints have caused shark interactions to
be an economic disadvantage. In these fisheries, fishers
have a large incentive to avoid shark interactions. In the
Italy longline fishery, despite a lack of market for shark
products, low shark interaction rates result in low incentive
to reduce shark interactions. The predominant attitude in
the Fiji longline fishery towards shark interactions is
unexpected. In this fishery, where almost all caught sharks
are finned and carcasses discarded, fishers perceive that
costs from shark interactions exceed the economic benefit.
In the Chile, Japan and Peru longline fisheries, where
restrictions on shark finning and retention are lacking,
there is no incentive to reduce shark interactions, as
revenue from sharks exceeds costs.
7. Industry shark avoidance and discarding practices
Table 4 identifies practices in use by longline fishers in
response to shark interactions with longline gear. A
practice is checked for a fishery only when it is employed
predominantly for the purpose of addressing shark
interactions, and not if the practice is employed primarily
as a normal part of fishing operations to maximize catch
rates of non-shark target species.
Fishers identified numerous fishing methods and gear
characteristics that they employ to maximize catch rates of
non-shark target species, which may contribute to reducing
shark catch rates. For instance, the depth of baited hooks,
timing of gear setting, soak and hauling, location of fishing
grounds in relation to topographic and oceanographic
features as well as sea surface temperature, type and size of
bait and hook, selection of material for the leader on
branch lines, use of lightsticks, and other fishing methods
and gear designs selected to maximize non-shark target
species catch rates may be effective shark avoidance
strategies. More research is needed to improve the under-
standing of the shark avoidance efficacy of many of these
practices.
7.1. Avoid peak areas and periods of shark abundance
In fisheries where there is an incentive to avoid shark
interactions, common practices are to avoid areas known
to have high shark abundance and move position when
shark interaction rates are high and the non-shark target
species catch rates are not particularly high. Because
there is high spatial and temporal variability in shark
catch rates (e.g., [78, 79]), in addition to fishing gear
characteristics, the location of fishing grounds to target
non-shark species and perhaps the capability of a vessel
captain to avoid areas where sharks are abundant appear
to be important factors determining a vessel’s shark
catch rate.
Fishing position in relation to (i) certain sea surface
temperatures, (ii) topographic features such as shelf breaks
and sea mounts, and (iii) oceanographic features such as
currents, fronts, and gyres, may affect shark interaction
rates. Australian fishermen identified setting on the colder
side of fronts in order to reduce shark catch levels.
ARTICLE IN PRESS
E. Gilman et al. / Marine Policy 32 (2008) 1–18
9

Page 10

Author's personal copy
Catch rates of blue sharks have been found to decline
by 9.7–11.4% in response to only a 0.61C increase in
sea surface temperature [24]. Not surprisingly, it has also
been shown that blue sharks tend to prefer sub-surface
depths that possess cooler temperatures (e.g. [80]). How-
ever, more comprehensive studies on blue shark distribu-
tion according to full water column temperature profiles
and thermocline dynamics are necessary before amending
fishing practices in accordance with patterns in sea-surface
temperatures.
7.2. Reduce shark detection of baited hooks
Very few interviewed fishers believe that refraining from
chumming during the set and refraining from discarding
offal and spent bait during the haul will affect shark
interactions. Chumming during setting is not a common
practice. Offal and spent bait are typically discarded during
hauling operations. Many respondents explained that it
would be impractical to retain spent bait and offal to
discard at the end of hauling due to limited space. Some
ARTICLE IN PRESS
Table 3
Industry attitudes towards shark bycatch and depredation prevalent in each of 12 pelagic longline fisheries
Longline industry attitude
Pelagic
longline
fishery
Want to
minimize
shark
interactions
due to time
required to
repair gear
and discard
sharks
Want to minimize
shark interactions
because revenue
from catching
sharks is exceeded
by costs from
shark interactions
Little incentive
to reduce shark
interactions
because they
are infrequent
and result in
nominal costs
Want to
maximize
shark catch
because
revenue
exceeds
economic costs
from shark
interactions
Want to minimize
shark catch to
avoid injuring
crew when
landing sharks
from projectile
swivels and shark
bites
Want to
minimize
shark fishing
mortality
because are
concerned
with
overfishing
Want to
minimize shark
catch to make
more baited
hooks
available to
more valuable
fish species
Shark
interactions
are an
expected
and
unavoidable
part of
longline
fishing
No incentive to
reduce shark
interactions
because revenue
from sharks
exceeds costs
from shark
interactions
Australia tuna
and billfish
fishery
XXXXX
Chile
artisanal mahi
mahi and
shark fishery
XX
Chile
swordfish
fishery
X
Fiji tuna
fishery
XX
Italy
mediterranean
industrial
swordfish
fishery
X
Japan distant
water,
offshore and
nearshore
tuna fisheries
XXX
Peru artisanal
mahi mahi
and shark
fishery
XX
South africa
tuna and
swordfish
fishery
XXXX
US hawaii
tuna fishery
XXXXX
US hawaii
swordfish
fishery
XXXXX
E. Gilman et al. / Marine Policy 32 (2008) 1–18
10

Page 11

Author's personal copy
Australian fishers avoid using lightsticks because they
believe lightsticks increase shark catch rates.
7.3. Reduce shark catch rate through deeper setting, leader
material and type of bait or hook
Some fishers avoid using certain types of bait to reduce
shark interactions, or perceive that avoiding certain bait
types will reduce shark capture rates (e.g., Italy and Japan,
avoid squid; Australia, avoid oily pilchard and squid).
Few fishers believe that hook shape has a large effect on
shark catch rates. Furthermore, some fishers indicated
that they set their gear at a certain depth or perceive
that setting deeper would contribute to reducing shark
interactions.
Most fishers believe that the depth of baited hooks and
timing of the gear soak influence shark catch rates. The
deployment depth of hooks and timing of the soak and
haul (day versus night) can have an influence on fish species
CPUE, including sharks, perhaps due to different water
temperature preferences by each species [36, 81, 82]. For
example, Rey and Munoz-Chapuli [82] found higher mako
ARTICLE IN PRESS
Table 4
Prevalent industry practices employed to address shark interactions in 12 pelagic longline fisheries
Pelagic longline fishery
PracticeAustralia
tuna and
billfish
fishery
Chile
artisanal
mahi
mahi and
shark
fishery
Chile
swordfish
fishery
Fiji tuna
fishery
Italy
Mediterr-
anean
industrial
swordfish
fishery
Japan
distant
water,
offshore
and
nearshore
fisheries
Peru
artisanal
mahi
mahi and
shark
fishery
South
Africa
tuna and
swordfish
fishery
US
Hawaii
tuna
fishery
US
Hawaii
swordfish
fishery
Move position if shark
interactions are high and
target species CPUE is
low
XXXXXXX
Avoid fishing grounds
with high shark
abundance from past
experience or
communication from
other vessels
XXXXX
Reduce shark detection
of baited hooks
Set gear deeperX
Use or avoid type of bait
or hook
X
To discard sharks, cut
branch line or remove
hook by making cut in
shark mouth
XXXXX
No wire trace to reduce
retention of sharks
X
Do not use lightsticksX
Avoid setting in specific
sea surface temperature
X
Set during daytimeX
Minimize gear soak timeX
Kill sharks before
discarding to avoid re-
catching
X
Do nothing to reduce
shark catch because
shark catch is desirable
or shark interactions are
rare
XXX
E. Gilman et al. / Marine Policy 32 (2008) 1–18
11

Page 12

Author's personal copy
shark (Isurus oxyrinchus) CPUE on shallower set hooks,
and no mako capture on the three deepest hooks in a
basket, which were estimated to be set to between 370 and
460m deep, in a Spanish tropical eastern Atlantic surface
longline swordfish and mako shark fishery. Williams [36]
found that main pelagic shark species, with the main
exception being the mako sharks, tend to be taken at a
higher rate in more shallow-set gear than vessels setting
gear deeper in central and western Pacific pelagic longine
tuna fisheries. Blue shark, silky shark, pelagic stingray, and
oceanic whitetip CPUE were 2.7, 6.4, 1.1, and 2.8 times
higher, respectively, in shallow versus deep set gear. Setting
baited hooks below a threshold depth may reduce bycatch
and depredation by certain species of sharks in certain
areas, but shark interaction rates may also depend on when
it is that the hooks are at these depths.
One fisher in the Hawaii-based longline tuna fishery tried
various types of artificial baits to determine their ability to
catch target species and to reduce shark capture. He found
that the artificial baits did not catch tuna well and that they
were not durable enough as he lost about 90% of the
artificial baits after one trip. One fisher in the Peru artisanal
mahi mahi and shark longline fishery tried artificial bait,
which did not reduce shark interactions. An artificial bait
was tested in the Alaska demersal longline sablefish
(Anoplopoma fimbria) and Pacific halibut (Hippoglossus
stenolepis) fishery. Results indicated that the artificial bait
caught as many or more target species and reduced bycatch
of spiny dogfish shark (Squalus acanthias), skate (Raja
spp.), arrowtooth flounder (Atheresthes stomias), and
Pacific cod (Gadus macrocephalus) by more than ten times
compared to a control of fishing with herring bait, the
conventional bait used in this fishery [83].
The type of hook and natural bait used might affect
shark CPUE and depredation rates [8, 36]. Results from
controlled experiments in the Azores and US North
Atlantic longline fisheries indicate that fishing with fish
instead of squid for bait causes a significant decrease in
shark CPUE, while using a wider circle hook instead of a
narrower J hook may cause a significant but small increase
in shark CPUE. Research in the Azores longline swordfish
and blue shark fishery, during a year when blue sharks
were not being targeted due to low market demand, found
that non-offset 16/0 circle hooks had a significantly higher
blue shark CPUE than a non-offset 9/0 J hook [84]. In
another study in the Azores fishery, during a year when
blue sharks were being targeted, non-offset 16/0 and non-
offset 18/0 circle hooks caught significantly more blue
sharks than a non-offset 9/0 J hook [85]. A study
conducted in the US North Atlantic longline swordfish
fishery found that a non-offset and 101 offset 18/0 circle
hook with squid bait resulted in a small but significant
increase in blue shark CPUE (8% and 9% increases,
respectively) compared to a 9/0 J hook with squid [24].
Watson et al. [24] also found that a 101 offset 18/0 circle
hook with mackerel bait and a 9/0 J hook with mackerel
bait resulted in a significant and large reduction in blue
shark CPUE by 31% and 40%, respectively, compared to a
9/0 J hook with squid. Research in an experimental
Japanese North Pacific longline fishery found no difference
in the capture rate of blue sharks between a circle and
Japan tuna hook [86, 87].
An assessment of observer data from the Hawaii-based
longline swordfish fishery is consistent with results from
these controlled experiments [8]. Shark combined species
CPUE was significantly lower by 36% after regulations
came into effect, which required the fishery to switch from
using a 9/0 J hook with squid bait to using a 101 offset 18/0
circle hook with fish bait [8].
Retention of sharks on branch lines with wire leaders or
other durable material is substantially higher than in gear
where nylon monofilament is connected directly to the
hook (Section 4 (e.g., [36])). Several fishermen who target
sharks seasonally use a wire leader when they wish to
maximize their shark catch rates and do not use a wire
leader when they are targeting non-shark species. Avoiding
certain material for branch lines could also reduce shark
depredation and catch rates. For instance, the use of
rope/steel (‘‘Yankee’’) gangions resulted in lower juvenile
sandbar shark catch rates than when using monofilament
gangions [88]. In another study, percent-capture of blue
shark with the use of monofilament gangions (66%)
exceeded that when employing multifilament gangions
(34%) [89]. Shortfin mako shark catches adhered to the
same pattern (60% and 40%) for ‘‘mono’’ and ‘‘multi’’
line, respectively. Stone and Dixon [89] hypothesized that
the aversion to the multifilament gangion was a function of
strong visual acuity.
7.4. Shark discard practices
A large proportion of pelagic shark species are alive
when gear is retrieved (Section 3), suggesting that improved
shark handling and release methods may reduce fishing
mortality of discarded sharks. While some fishers report
routinely killing caught sharks to retrieve terminal tackle
and to avoid the inconvenience of recapture, empirical
information from Australia, Hawaii and Peru shows that a
very small proportion of caught sharks that are alive when
hauled to the vessel are killed before discarding. When a
caught shark is discarded, the majority of fishers indicate
that most of the time they cut branch lines, cut the hook
out of the shark’s mouth with a knife or pull the hook out
by force in order to retrieve the terminal tackle before
discarding the shark. It is uncommon for fishers to kill
a shark to retrieve terminal tackle or to prevent future
shark interactions.
The survival of sharks that are not finned, that are
deeply hooked (where the shark has swallowed the hook)
and have hooks removed by fishers pulling the hook out
likely depends on where they were hooked and how much
damage is done by pulling out the hook. For deeply
hooked sharks, as is believed to be the case for sea turtles
[8, 9, 24], prospects for post release survival might be
ARTICLE IN PRESS
E. Gilman et al. / Marine Policy 32 (2008) 1–18
12

Page 13

Author's personal copy
improved by having fishers cut the line as close to the shark
as safely possible. A large proportion of sharks caught in
longline gear that is released after removal of the hook
from the mouth are expected to survive. Research using
pop-up satellite archival tags found that 97.5% of pelagic
sharks released after capture in longline gear survived (1 of
40 captured sharks died), while another study found that
94% of 17 tagged shorfin mako sharks survived beyond
two months after release from being caught in longline gear
[50]. While shark post-release survival prospects are high,
and while many fishers do not see a need for new hook
removal methods, development of especially designed
equipment to discard sharks could improve shark post
release survival prospects, reduce the loss of terminal tackle
and improve crew safety.
None of the interviewed pelagic longline fishermen use
dehookers to discard live sharks. Only two fishers, one
from the Chile artisanal mahi mahi and shark longline
fishery and one from the Peru artisanal mahi mahi and
shark longline fishery, report that they use a dehooker to
recover hooks from sharks when these are onboard and
already dead. Most fishers perceive commercially avail-
able dehooker devices to be impractical and potentially
dangerous for use with sharks. Some caught sharks will
twist and spin when hauled to the vessel, which could cause
the dehooker to be lost overboard and be a hazard for crew
to handle before being lost overboard. Because sharks may
be on the sea surface when being hauled, some fishers were
concerned that the incidence of having branch lines break if
a shark pulls the line would increase with use of a dehooker
because use of a dehooker requires bringing the shark close
to the vessel.
8. Promising new strategies for shark avoidance
8.1. Shark repellents
For fisheries with an incentive to reduce shark interac-
tions, chemical, magnetic, electropositive rare earth metal
and electrical repellents are promising shark deterrents.
Some of these strategies are concepts requiring substantial
investment to develop the technology for application in
longline gear. Research and commercial demonstrations
are needed to assess their efficacy at repelling sharks and
effects on target and incidental species. Research and
development is also needed to reduce the per-unit cost of
these repellents to make them economically viable for use
in pelagic longline fisheries.
Chemical deterrents, including a protein extract called
‘‘pardaxin’’ from an excretion from the moses sole
(Pardachirus marmoratus), sodium and lithium lauryl
sulfate (components of common soap and shampoo) and
sodium dodecyl sulfate, a related compound, have been
found to repel some species of sharks under certain
conditions [90]. Some fishers from Peru artisanal mahi
mahi and shark fishery report retain all shark offal until the
end of a haul because they believe that discharging offal
during the haul would repel sharks and reduce their shark
catch rate. The US-based company Shark Defense LLC
has identified a semiochemical-based shark repellent, which
in preliminary trials caused six shark species to leave an
area after the chemical was disbursed without repelling
teleost fish [91]. Presumably, this can be ascribed to an
apparent aversion in sharks to certain chemicals, inclu-
ding ammonium acetate (a major component in decaying
shark flesh) and other semiochemicals emitted from
predators [92]. When mixtures of semiochemicals were
introduced into feeding aggregations of sharks and
teleost fish in reef habitat, sharks quickly left the feeding
area while bony fish stayed in the area and continued
feeding [50].
A preliminary study was conducted by the US National
Marine Fisheries Service Pacific Islands Fisheries Science
Center in Hawaii in early 2005 to compare the catch of
target species and sharks in sets using bait injected with
synthetic shark semiochemicals produced by Shark De-
fense LLC to catch rates in sets using untreated bait.
Results were inconclusive, in part, because the research
design prevents conclusions to be drawn on the single
factor effect of the presence or absence of semiochemicals
in the bait, and because it was not possible to confirm that
bait injected with semiochemicals retained the chemical
throughout the gear soak [50].
Since conducting this 2005 trial in Hawaii, the chemical
is now available in a hydroxypropyl cellulose and glycol
ether ester gel matrix [91]. In current ongoing trials the gel
is placed in biodegradable, porous muslin bags filled with
100ml of the gel at 30.5cm above the bait. The gel has been
observed to dissolve evenly over an 8-h period while the
fishing gear soaks. The gel could also be syringed directly
into a bait or a bag of the gel could be stuffed into a bait
[91]. One bag filled with 100ml of the gel matrix would
cost about USD 1.05. Pre-treated baits may be a less
expensive option.
Shark Defense is also conducting preliminary trials of
neodymium-iron-boride (Nd2Fe14B) magnets as a possible
shark deterrent in longline gear. It is hypothesized that a
10cm?4cm NdFeB magnet’s field would be effectively
detected by sharks up to a 0.3-m range. Preliminary
research conducted in 2005 on the effect of Nd2Fe14B
magnets by the Inter-American Tropical Tuna Commission
on captive yellowfin tuna and by the University of Miami
on cobia indicates that the presence of the magnet versus a
control produced no significant difference in feeding
behavior [91]. Preliminary research in a demersal longline
fishery in the Bahamas is underway. A 2.5cm?2.5cm
Nd2Fe14B nickel-coated cylinder with a center bore costs
about USD 300 for 100 magnets.
Shark Defense is also exploring the shark deterrent
efficacy of electropositive metals (e.g. neodymium, praseo-
dymium, early lanthanide metals, mischmetal, and magne-
sium). These metals, which are also present in rare earth
magnets, may be responsible for some of the repellency
effect seen with permanent magnets and present a more
ARTICLE IN PRESS
E. Gilman et al. / Marine Policy 32 (2008) 1–18
13

Page 14

Author's personal copy
practical alternative to the magnets. A correlation has been
found between standard oxidation potential of these
metals and their behavioral response using immobilized
sharks [91].
An electrical shark avoidance device was tested in a
coastal midwater trawl fishery in the Sea of Japan
(Ishikawa Prefecture) in 2004. The purpose of the device
was to deter predation by sharks on the cod end of the
trawl during hauling. The device, mounted on the fishing
vessel, emitted an electrical pulse into the waters in the
immediate vicinity. It was believed by fishermen to be
effective based on qualitative observations of sharks
suddenly moving away from the cod end and the vessel
once the electrical pulse was emitted.
The Shark Protective Ocean Device is a device designed
to be worn by scuba-divers that emits an electrical field
with a radius of 4–6m to repel sharks from divers. The
device costs about USD 700. This technology theoretically
could be modified to deter sharks from foraging on bait
and catch on longline hooks.
Acoustic deterrents may reduce shark–longline interac-
tions, but have not been tested in longline fisheries for any
shark species.
8.2. Hotspot avoidance through fleet communication and
protected areas
The distribution of sharks, as well as other species
groups such as seabirds, sea turtles and cetaceans, is often
unpredictable, and may be spatially contagious or aggre-
gated. Consequently, fleet communication programs may
be employed by a fishing industry to report near real-time
observations of hotspots to substantially reduce fleet-wide
depredation and bycatch of sharks [11]. In addition,
fleet coordination of daily fishing positions and times, a
current practice in many fleets, may minimize per vessel
shark interaction levels relative to vessels that fish in
isolation [10].
Area and seasonal closures can also contribute to
reducing shark–longline interactions. Establishing pro-
tected areas within a nation’s EEZ is potentially an
expedient method to reduce shark–longline interactions.
However, establishing and managing high seas marine
protected areas to protect sharks, which would require
extensive and dynamic boundaries and extensive buffers,
may not be a viable short-term solution. This is due in part
to the extensive time anticipated to (i) resolve legal
complications with international treaties, including creat-
ing legally binding mechanisms for multilateral designation
and management of high seas protected areas;5(ii) achieve
international consensus and political will; (iii) provide
requisite extensive resources for surveillance and enforce-
ment, in part, to control illegal, unreported and unregu-
lated fishing activities; and (iv) improve the scientific basis
for designing high seas marine protected areas, which can
be effective at reducing shark interactions only where the
location and times of occurrence of shark hotspots are
known and predictable. However, establishing and mana-
ging a representative system of protected area networks on
the high seas to contribute to the management of
interactions between marine capture fisheries and highly
migratory sensitive species groups, including sharks, may
eventually be realized.
Consequences of establishing a protected area need to be
carefully considered, as resource use restrictions of a
marine protected area may displace effort to adjacent and
potentially more sensitive and valuable areas, where
weaker management frameworks may be in place [58, 93].
Also, measures adopted by regional fishery management
organizations and other international bodies are only
binding to parties to the Convention that established the
organization, and will not control activities by non-party
States. Thus, another consideration for employing high
seas marine protected areas to manage problematic fish-
eries bycatch is that closing areas to fisheries only of party
States could result in increased effort in this area by fleets
from non-party States with fewer or no control to manage
bycatch, exacerbating the problem for which the MPA was
established to address.
9. Conclusions
Incentives for pelagic longline fishers to reduce shark
interactions vary along a continuum, based on whether
sharks represent an economic disadvantage or advantage.
On one extreme, there are pelagic longline fisheries with a
regulatory framework limiting shark catches or placing
restrictions on shark handling, or lack of markets for shark
products, resulting in negligible retention of sharks. In
these fisheries, the costs from shark interactions exceed
benefits from revenue from sharks. On the other extreme,
there are pelagic longline fisheries where revenue from
sharks exceeds costs from shark interactions, a large pro-
portion of caught sharks are retained, and sharks are either
always an important target species, are targeted seasonally
or at certain fishing grounds proximate to ports where
there is demand for shark products, or are an important
incidental catch species.
In fisheries where there is an incentive to avoid shark
interactions, predominant shark avoidance practices are to:
(i) Avoid fishing in areas known to have high shark
interactions; and
ARTICLE IN PRESS
5It is already possible to establish high seas MPAs for discrete areas by
agreement by individual countries. However, there remains a need for an
international framework with specific language to identify the criteria to
establish a representative system of high seas MPA networks, and
management and enforcement measures for the individual MPAs. Several
RFMOs are updating their scope and legal mandate to include ecosystem-
based management and biodiversity conservation under the auspices of
(footnote continued)
the Fish Stocks Agreement. The Commission for the Conservation of
Antarctic Marine Living Resources has made some preliminary progress
toward establishing a system of MPAs in the Southern Ocean.
E. Gilman et al. / Marine Policy 32 (2008) 1–18
14

Page 15

Author's personal copy
(ii) Change fishing grounds when shark interactions are
high but the target species catch rate is low.
Longline fishers identified numerous fishing methods and
gear characteristics that they conventionally employ to
maximize catch rates of non-shark target species, which
may contribute to reducing shark catch rates. For instance,
experimental trials have shown that using fish instead of
squid as bait results in a significant and large decrease in
shark catch rates [8, 24, 84, 85]. Also, deeper setting helps
reduce catch rates of most pelagic shark species [36].
Research is needed to improve the understanding of the
shark avoidance efficacy of some of these other practices.
Beyond these strategies, the state of knowledge for shark
avoidance in pelagic longline fisheries is poor. Chemical,
magnetic, electropositive rare earth metals and electrical
shark deterrents hold promise. Research is needed to assess
their efficacy at repelling sharks, effect on target species
catch rates and reduce the cost for commercially viable
employment in longline fisheries. Fleet communication
programs and marine protected areas also hold promise to
reduce unwanted shark–longline interactions.
A large proportion of pelagic shark species are alive
when gear is retrieved. Most sharks that are alive when
hauled to the vessel and will be discarded are released alive.
When a caught shark is discarded, most of the time fishers
either cut branch lines, cut the hook out of the shark’s
mouth or pull the hook out by force in order to retrieve the
terminal tackle before discarding the shark. It is uncom-
mon for fishers to kill a shark to retrieve terminal tackle or
to prevent future shark interactions. Most fishers perceive
commercially available dehooker devices to be impractical
and potentially dangerous for use with sharks. Develop-
ment of especially designed equipment to discard sharks
could improve shark post release survival prospects, reduce
the loss of terminal tackle and improve crew safety. In
fisheries where shark finning occurs, to avoid injury and
increase efficiency, crew first kill the fish before removing
fins, and do not remove fins from live sharks.
The Japan, Chile and Peru components of this study
documented growing markets for shark meat at several
ports worldwide. This trend toward more utilization of
shark meat may be beneficial in the short term in that fully
utilized sharks are more likely to be reported in logbooks
and landings statistics than are the retention and landing of
just shark fins. However, if the shark meat market
continues to grow, this could increase shark catch rates
and fishing mortality. This study shows that fishers possess
the knowledge to modify their fishing gear and methods to
maximize shark catch. There are few fisheries with
measures to manage shark catch levels. Thus, to prepare
for a possible increase in demand for shark meat fishery
management authorities are encouraged to institute data
collection, monitoring and precautionary management
measures to ensure that shark catches are sustainable.
Most national fishery management authorities of the 12
fisheries included in this study demonstrate a low priority
for monitoring and managing chondrichthyan fishes,
consistent with the results of a global review by Shotton
[74]. Few regional fishery management organizations are
using fishery-dependent data to conduct shark stock
assessments (only the International Commission for the
Conservation of Atlantic Tunas, for blue and shortfin
mako sharks in the North and South Atlantic). Sustainable
management of chondrichthyan populations is hampered
by this general lack of fishery-dependent data and manage-
ment measures for sharks [71]. The expanding exploitation
of sharks, for their fins as well as meat, largely in the
absence of management frameworks and the lack of
reliable fishery-dependent data and fundamental under-
standing of the biology of most shark species warrant
concern for the health of shark populations as well as
ecosystem-level effects from population declines. Ap-
proaches to sustainably managed cartilaginous fishes will
necessarily differ from traditional fishery management
methods for teleosts due to cartilaginous fishes’ relatively
low reproductive potential [72].
Acknowledgments
We are extremely grateful for the contribution of fishers
and other longline industry representatives. Staff from the
following organizations shared their considerable knowl-
edge of the fisheries and their management: Australian
Fisheries Management Authority; Australian Common-
wealth Scientific and Research Organization, Marine and
Atmospheric Research Laboratory; Oceanic Fisheries
Programme, Secretariat of the Pacific Community; Fiji
Department of Fisheries; StoneFish (Fiji) Limited; Uni-
versity of the South Pacific; Greenpeace South Pacific
Programme; General Association of Italian Cooperatives-
Fishery sector; National Research Institute for Far Seas
Fisheries, Japan Fisheries Research Agency; Organization
for Promotion of Responsible Tuna Fisheries; and Global
Guardian Trust. This research was made possible through
financial support from the Western Pacific Regional
Fishery Management Council, Regional Seas Programme
of the United Nations Environment Programme and the
Gordon and Betty Moore Foundation.
References
[1] Alverson DL, Freeberg MH, Murawski SA, Pope JG. A global
assessment of fisheries bycatch and discards. Fisheries Technical
Paper No. 339. Food and Agriculture Organization of the United
Nations, Rome, Italy; 1994.
[2] FAO. International plan of action for reducing incidental catch of
seabirds in longline fisheries. Rome, Italy: Food and Agriculture
Organization of the United Nations; 1999.
[3] FAO. International plan of action for the conservation and
management of sharks. Rome, Italy: Food and Agriculture Organi-
zation of the United Nations; 1999.
[4] FAO. The state of world fisheries and aquaculture. Rome, Italy:
Food and Agriculture Organization of the United Nations; 1999.
[5] FAO. Technical consultation on sea turtles conservation and
fisheries, Bangkok, Thailand, 29 November–2 December 2004. Sea
ARTICLE IN PRESS
E. Gilman et al. / Marine Policy 32 (2008) 1–18
15