C 55.302: M 31
SEP 1 4 1998
jction
S. Depository Copy
Vto»a,.
50%
7% (12)
8% (13)
75% (119)
II
1-50%
<1%
III
ll/lll 2
0-50%
10% (15)
' See the glossary at the end of this document.
2 10% (15) of the fisheries included in this assessment were classified as Category ll/lll due to the
inclusion of several fisheries with different MMPA categories in a single classification.
purpose of compiling these statistics was to quan-
tify the biological, economic, and social signifi-
cance of each fishery to the extent possible. The
most recent estimates of discards of each species
or species group were used for each fishery. Dis-
cards for a species or species group were not
estimated if no statistically reliable information
was available.
A total of 148 unique species or species
groups were identified as discards associated with
the 152 fisheries defined nationwide. Of these
species or species groups, 92 (62%) were finfish,
crustaceans, or molluscs and 56 (38%) were "pro-
tected" species (i.e., marine mammals, turtles, or
birds). Protected species were not included in
the review unless positive identification — fre-
quently to the species level — and exact enumera-
tion were possible.Thus, information on discards
of protected species is available in much greater
detail than for fish, and caution must be exer-
cised when comparing species or species group
counts between finfish and protected resources.
Some protected species are represented by a
single occurrence, whereas the resolution for fish
was in terms of metric tons or thousands offish.
A species or species group was frequently iden-
tified as discard in more than one fishery. For
example, snow crab was listed as a discard spe-
cies in 25 of the 54 fisheries in Alaska, and pe-
lagic species were listed as discards in 9 of the 35
fisheries in the Southeast.
The quality and quantity of discard and
other bycatch information on species or species
groups varied considerably among the regions.
Regions with large data collection programs
were able to provide information at a much finer
level of resolution, frequently at the species level,
than were regions that had either minimal or no
quantitative information on discards in the
region's fisheries. When no quantitative infor-
mation on discards for a fishery was available,
general descriptive categories, such as "ground-
fish," were created; when quantitative informa-
tion was available, individual species were listed
separately. Similarly, simple classification of fish-
eries based on targeted species and gear results
in all fisheries being equivalent and can mask
the importance of a fishery and potential im-
pact of discards on it. Thus, analyses were con-
ducted at the regional level and considered the
volume of the discards in the fishery if possible.
Data were compiled to provide a general pic-
ture of how much is known about discards in
the nation's fisheries and to identify major trends
National
Persjtecture*
within fisheries and regions. Due to the varying
level and quantity of information available, data
in the matrix cannot be used to calculate total
discards for a particular region or fishery or to
make comparisons about discard rates and
amounts among regions.
Quantitative estimates of finfish discards
were available for 52% (48 of 92) of unique dis-
card species or species groups in the nation's fish-
eries.The fractions of discarded species for which
quantitative estimates were available were dis-
proportionate among regions (Table 3). These
numbers do not imply that precise or accurate
measures for 52% of the species discards are avail-
able. Only in Alaska groundfish and some shell-
fish fisheries is there sufficient information to
estimate total fish discards for some fisheries. For
protected species some quantitative data on
by catch are available for 61% (43 of 57) of pro-
tected species or protected species groups.
Reasons for Discards in the
Nations Fisheries
Four categories were identified as poten-
tial reasons for discards: (1) discards of protected
species; (2) regulatory-induced discards — e.g.,
quotas, trip limits, prohibited species, size or sex
limits; (3) discretionary discards, which may oc-
cur, for example, when no market exists for a
particular species; and (4) catch-and-release dis-
cards, as in recreational fisheries. Analyses of the
reasons for discards can be affected by the de-
gree of classification of the species discarded. This
assessment was conducted using both nominal
counts of the reasons for discarding species or '
species groups and quantitative measures (weight
or numbers) where available.
Clearly, when only the occurrence of a spe-
cies/group is considered, regulatory-induced
discards are dominant in most regions (Figure
1). Regulatory and discretionary discards occur
together in a significant proportion of fisheries
in some regions, and account for the most sub-
stantial portion, by volume and occurrence, of
discards in all regions. Protected species discards
occurred in all regions. Catch and release was
not the dominant factor influencing discards in
any fishery.
Significance of Discards in the
Nations Fisheries
Information on the current status of target
and discard species was obtained from Our Liv-
ing Oceans (NMFS 1996a). Two measures of stock
status were specified: (1) the rate of utilization
(over-, fully, or underutilized) and (2) the cur-
rent stock size relative to the size necessary to
produce the maximum long-term potential yield
(below, near, above). These criteria are impor-
tant when considering the effects discards may
Table, 3.
Percentage and number of discarded species or species groups
were available, exclusive of protected species.
for which quantitative estimates
Region
Percent
Number (Total)
Alaska
Pacific pelagic and insular area
Atlantic & Gulf pelagic
89
24 (27)
57
8(14)
50
33
30
5(10)
Southeast
3(9)
3(10)
5(23)
West Coast
Northeast
22
National By catch
figure, 1.
Reasons for discarding species or species groups. Classification reflects occurrence, not amount, of
each type of discard.
Northeast Fisheries
Atlantic tutd gulf Pelagic Fisheries
Southeast Fisheries
Pacific Pelagic and Insular Fisheries
West Coast Fisheries
Alaska, Fisheries
Discretionary Discards
Discretionary & Regulatory
Regulatory Discards
Protected Species
have in contributing to the exploitation status
of stocks.
For fishery resources,Table 4 describes each
discarded fish species/group according to its sta-
tus of utilization (over-, fully, or underfished) in
relation to its long-term potential yield. Taken
together, these two criteria indicate that the
magnitude of fishery discards of some species or
species groups may be important in determin-
ing the health of these stocks. For instance, for
the species for which information is available,
50% of the fish species that are discarded in the
fisheries for Atlantic and Gulf highly migratory
pelagic species are below their long-term po-
tential yield and are over- or fully utilized. This
means that the stocks of these species have sus-
tained heavy fishing pressure and are depleted
to levels below the maximum long-term aver-
age catch that can be sustained. For these stocks,
discard mortality can be an important additional
source of fishing pressure that should be ac-
counted for in fishery analyses. Regionally, us-
ing both criteria, the status of bycatch species or
species groups varies, with 82% of the discard
species or species groups in the Northeast, 80%
of Atlantic and Gulf highly migratory pelagic-
species, 75% in the Southeast, 60% on the West
Coast, and 52% in Alaska classified as fully or
overutilized and at or below their long-term
potential yield. The status of 45% of discard spe-
cies or species groups in the Pacific pelagic and
insular fisheries is unknown with respect to ei-
ther of these criteria.
Discard mortality, in combination with di-
rected fishing mortality and unobserved mor-
tality, contributes to the current status of stocks.
In the case of overfished fisheries, reducing some
component of fishing mortality — either directed,
incidental, or unobserved mortality — is critical
to rebuilding these stocks to sustainabilitv.
The significance of discards was further
evaluated through the use of two related quali-
National
Perst
TabU 4.
Current level of utilization and long-term potential yield of discard species or species groups.
Long-Term Potential Yield
Level of Utilization % Below % Near
% Above
% Unknown
% Total
Northeast Fisheries 64 23
Over 55
Full 9 18
Under 5
Unknown
13
5
4
4
100
60
31
9
Southeast Fisheries 76 13
Over 63
Full 12
Under 12
Unknown
West Coast Fisheries 70 20
Over 20
Full 30 10
Under 20 10
Unknown
Atlantic and Gulf 50 30
Pelagic Fisheries
Over 40
Full 10 30
Under
Unknown
13
13
100
63
12
12
13
10
10
100
20
40
30
10
20
20
100
40
40
20
Pacific Pelagic and 9 45
Insular Fisheries
Over 9
Full 9
Under 9
Unknown 27
Alaska Fisheries 36 28
Over
Full 32 20
Under 4 8
Unknown
36
12
24
46
45
100
9
9
9
72
100
64
36
National
tative measures — nature and level. The nature of
discards identifies the following categories of
concerns: population status (of the discarded spe-
cies), social and economic concerns, ecosystem
concerns, or public concerns. In the review,
population effects of discards was listed as the
primary concern if discards contributed signifi-
cantly to the current status of a species or spe-
cies group. Public concern was frequently listed
as the primary determinant when discard of a
species or species group is low relative to other
sources of mortality.
As shown in Figure 2, population concerns
dominated in the fisheries for Atlantic highly
migratory pelagic species and in the Northeast,
while social and economic concerns dominated
in the Western Pacific area, the Southeast, and
Alaska. Social and economic and population
concerns were about equal in the Pacific Coast.
Population issues were the overwhelming con-
cern for protected species in all regions, except
for Alaska, where public concern regarding the
impacts of discards on populations of marine
mammals and birds was the primary factor.
Evaluation of discards may be problematic.
For example, uncertainty regarding the effects
of discards on population status may generate
public concerns and may have economic con-
sequences for the industry. In these cases, mul-
tiple causes for concern are ranked by priority
in the review, and the most important factor in
determining the nature of discarding is used for
this analysis (Figure 2).
The level of concern about discards describes
in subjective, relative terms the importance dis-
cards have for one or more of the following at-
tributes: population status of the discarded spe-
cies, the economic and social status of fisheries
that may target the discarded species, or the ef-
fects on the ecosystem from which the discarded
figures 2.
Primary nature of concern affecting the determination of the significance of discards for species or
species groups.
Northeast fisheries
Atlantic and, Quit Pelagic fisheries
Southeast fisheries
Pacific Pelagic omA Insular fisheries
West Coast fisheries
Alaska, fisheries
Population Concerns
Public Concerns
Social/Economic Concerns
Ecosystem Concerns
Vevspecturt-
species is taken. This is not a measure of the ab-
solute magnitude of the discards for a -species or
species group. Four categories of discard level
used were high, moderate, low, and unknown.
Regional data on discard levels for all fisheries
are compiled in Figure 3. Information for pro-
tected species was not used in this analysis be-
cause it is available at a much greater level of
resolution than for fish. Some protected species
are represented by a single occurrence, whereas
the resolution for fish was in terms of metric
tons or thousands of fish. Note that the same
discard stock may be counted more than once if
it occurs in more than one fishery (hence there
was a total of 447 cases) . Overall, there is a ten-
dency for the level of concern to be high or
moderate for over- and fully utilized stocks. For
protected resources (marine mammals, turtles,
and birds), the level of concern for the vast
majority of discards is considered high or
moderate.
■
Adequacy of Information for
Managing Bycatch
NMFS developed a systematic hierarchical
approach to identifying and evaluating the in-
formation available for managing bycatch. The
hierarchy consists of seven steps that can be used
to identify problems, evaluate potential solutions,
and implement effective management programs.
It provides a measurable framework that is adapt-
able to region- and fishery-specific character-
izations that can be used widely across NMFS
regions and fisheries.
The seven steps, described in detail in Ap-
pendix A, are: (1) determine the quality of in-
formation on the magnitude of bycatch; (2)
evaluate the impacts of current bycatch prac-
tices on populations, fisheries, and ecosystems;
(3) evaluate the effectiveness of current bycatch
management measures; (4) identify potential
management alternatives; (5) evaluate the popu-
lation, ecosystem, and socio-economic effects of
Figure* 3.
Level of concern for population, social and economic impacts on species or species groups.
Northeast fisheries
Atlantic AtteL Qulf Pelagic Fisheries
Southeast Fisheries
Pacific Pelagic and, Insular Fisheries
West Coast Fisheries
Alaska, Fisheries
High
Low
Moderate
Unknown
National B\
* An
each alternative; (6) choose and implement an
alternative; and (7) evaluate the effectiveness of
the implemented measures.
A hierarchical description of data quality
and progress was used to assess the agency's
current capabilities for addressing bycatch is-
sues. Generally, little or no information is avail-
able on the unobserved mortality portion of
bycatch; the results summarized here address
discards only. Information relating to regional
progress in completing these seven steps also
follows.
Information/ on, the Magnitude of Bycatch
The quality of information available on
discards is greatest in Alaska and in Atlantic pe-
lagic species, and poorest for the Southeast,
Northeast and Pacific Coast regions. Nationwide
the quality of information is only slightly better
than isolated snapshots of information. Informa-
tion on the unobserved mortality compo-
nent of bycatch is lacking in nearly every
fishery.
Impact Analyses of Bycatch
There is little information on the popula-
tion, social, economic, and ecosystem impacts of
discards. Some quantitative information, mixed
with qualitative information, is available on the
population impacts of discards. Limited qualita-
tive information is available for evaluating the
social and economic impacts of discards. No re-
gion has yet completed quantitative or qualita-
tive evaluations of the impacts of discarding on
ecosystems.
Effectiveness of Current
Management Measures
The adequacy of current bycatch manage-
ment measures was evaluated in terms of their
population, ecosystem, social, and economic ef-
fects. The evaluation indicated that most fisher-
ies require identification of additional manage-
ment alternatives.
identification^ of Potential
Management Alternatives
Progress in identifying management alter-
natives was evaluated to determine if the practi-
cality of proposed alternatives has been assessed
in terms of industry acceptability and fishery
management council policy. Nationally, major
factors influencing discards have been identified,
and input in terms of management alternatives
is being sought in many cases. Within the re-
gions, progress is quite variable, as those with
the highest-priority discard problems have re-
ceived greater attention than others.
Evaluation^ of Impacts of Bycatch
Management Alternatives
The population, social, and economic im-
pacts of alternatives have been evaluated to a lim-
ited extent in all regions. In general, however,
these evaluations are based on qualitative infor-
mation and, either no evaluations have been
made or, in some cases, qualitative judgments
on the ecosystem impacts of management alter-
natives have been made.
Implementation; of Alternative
Management Measures
Nationwide there has been little progress
in developing the regulatory, enforcement, or
monitoring infrastructure necessary to imple-
ment effective discard-reduction programs.
Adequacy of Monitoring Programs
Monitoring programs are best developed
in Alaska. In other regions, they are generally
not capable of routinely monitoring the effec-
tiveness of bycatch reduction measures, although
programs may be in place for selected high-pro-
file fisheries.
ftal goal caul Objectives
This plan reflects the aggregate knowledge
and experience of the National Marine Fisher-
ies Service and its many partners, including con-
tributions from many regional and national
bycatch workshops held from 1992 through
1995. The national bycatch goal and objectives
described here were developed after consider-
ation of these perspectives, as well as the regional
perspectives provided in the second section of
this plan. Bycatch planning must be a dynamic
process that continually incorporates informa-
tion and views from all these sources. Finally,
the plan does not propose to direct activities of
nonfederal sectors, but rather to focus national
and regional bycatch research and management
needs for NMFS.
National Goal
The fundamental national goal of NMFS'
bycatch-related activities is to implement con-
servation and management measures for living
marine resources that will minimize, to the ex-
tent practicable, bycatch and the mortality of
bycatch that cannot be avoided. Inherent in this
goal is the need to avoid bycatch, rather than
create new ways to use bycatch.
The national bycatch goal reflects the es-
sential bycatch management purpose of the
major marine resource statutes (the Magnuson—
Stevens Fishery Conservation and Management
Act, the Marine Mammal Protection Act
(MM PA) and the Endangered Species Act (ESA))
to reduce bycatch and bycatch mortality for spe-
cies managed under the acts. It also reflects the
commitment to cooperate with the U.S. Fish
and Wildlife Service in monitoring and report-
ing the bycatch of seabirds listed under the ESA
and those protected under the Migratory Bird
Treaty Act.
Despite this similarity of purpose, the acts —
and thus bycatch management of the appropri-
ate species — have several important differences.
The goal of the MMPA is to reduce bycatch "to
insignificant levels approaching zero mortality
and serious injury rate [by April 30, 2001]," rather
than the Magnuson— Stevens Act's "to the extent
practicable" [Sec. 118 (b) (1) 16 U.S.C. 1387].
The ESA proscribes the taking of listed species
based upon the biological status of the species
(16 U.S.C. 1531 et seq.).The incidental catch of
protected species, such as marine mammals and
ESA-listed salmon, turtles and seabirds is man-
aged by take-reduction teams and recovery plans,
respectively.The Migratory Bird Treaty Act gov-
erns any taking of seabirds in addition to the
ESA-listed species (16 U.S.C. 703 et seq.). Na-
tional Standard 9 in the Magnuson— Stevens Act
highlighted the need for the statement of a similar
management goal for living marine resources
managed under fishery management plans.
While the bycatch management measures
employed to manage protected species differ
from those for other species, it is the intention
of this plan to lay the groundwork for an inte-
grated, comprehensive approach to all aspects of
the bycatch problem. This will allow NMFS to
build on successful existing bycatch management
National Goal
and, Objectives
programs, such as the take-reduction teams, while
identifying areas where further research and
management are needed to address bycatch. Spe-
cific concerns generated by the workshops, Con-
gressional directives, and NMFS support the
achievement of the fundamental national goal
and have been cast as objectives for this plan.
National Objectives
The following objectives are based upon
findings of the National Assessment that was
conducted during development of this plan.
These objectives support achievement of NMFS'
national bycatch goal.
/. Determine the Magnitude of Bycatch
Determining the magnitude and character of the
bycatch in a given fishery is critical to the effec-
tive conservation and management of the stocks
in question. As pointed out in many of the re-
cent bycatch workshops and symposia, the cur-
rent debate on bycatch is often driven by the
lack of information on how much, where, when,
and what type of bycatch is occurring.
Strategy 1
Review and, where necessary, improve collec-
tion methods, data sources, and applications of
data to determining the magnitude of
bycatch.
a. Identify required data elements for estima-
tion of bycatch mortality.
b. Conduct a review of government and non-
government sources of bycatch data, includ-
ing observer programs, fishery-dependent
and independent surveys, and other data col-
lection programs.
c. Develop a methodology to estimate unob-
served mortality.
d. Conduct a periodic review of the available
data on the character and magnitude of
bycatch.
e. Solicit the input of fishery scientists, manag-
ers, industry representatives, and conserva-
tion groups on methods to assess the quan-
tity and type of bycatch.
Strategy 2
Standardize the collection of bycatch data.
a. Coordinate pilot programs to ensure that
estimates of bycatch are comparable across
programs.
b. Design and test sampling protocols to pro-
vide precision and accuracy of data at the
lowest cost.
c. Evaluate the accuracy and precision of the
data and their usefulness in estimating the
magnitude of the bycatch.
d. Make the collection of bycatch data part of
the NMFS core statistics program.
e. Assess bycatch mortality in commercial and
recreational fisheries.
f. Solicit the input of fishery scientists, manag-
ers, industry representatives and conserva-
tion groups on methods to establish stan-
dards for bycatch data collection.
g. Integrate the collection of economic and
social information (e.g., operating costs, fleet
size, and vessel characteristics) with the col-
lection of biological information.
//; Determine the Population, Ecosystem,
Social, and Economic Impacts
The current state of knowledge on the impacts
of bycatch and bycatch mortality on populations
and ecosystems, and on the social and economic
NcubLotial
Perspective*
implications of bycatch, is highly variable. Some
fisheries have a substantial amount of informa-
tion on the population effects of bycatch, while
others have very little data. Generally, very little
or no information is available on the ecosystem
or economic impacts of bycatch, or the social
and economic impacts of bycatch reduction strat-
egies. NMFS must determine the impacts of
bycatch in order to establish research and man-
agement priorities.
Strategy 1
Identify the type and quality of the information
that currently exists. Consider the availability of
expertise and information from the commercial
and recreational fishing sectors, the councils,
conservation groups, and the interstate marine
fisheries commissions.
Strategy 2
Establish research and management priorities on
a fishery-by-fishery basis.
Strategy 3
Develop a fully integrated data collection sys-
tem that includes biological, economic and so-
cial information.
Strategy 4
Identify ecosystem-wide issues that can be ad-
dressed through a well-coordinated research
program.
Strategy 5
Assess the impacts of bycatch.
a. Use bycatch statistics programs to help de-
termine the population impacts of bycatch.
b. Consider the lost benefits due to bycatch.
c. Assess the impact of bycatch mortality on
fishing communities.
d. Develop models for assessing the indirect
impacts of bycatch mortality.
e.
Include analyses of single-species and
multispecies impacts.
f. Identify gear impacts on species.
g. Build partnerships and increase information
sharing with government and nongovern-
ment scientists, particularly of ecosystem im-
pacts of bycatch and other sources of fishing
mortality.
///; Determine Whether Current
Conservation and Management
Measures Minimize Bycatch. If
Necessary, Choose New
Alternatives
Conservation and management measures to
minimize bycatch to the extent practicable will
be executed, primarily at the regional level. It is
generally the responsibility of NMFS and the
respective fishery management councils to evalu-
ate current and proposed management measures.
Strategy 1
Evaluate current management measures.
a. Assess the precision and accuracy of quanti-
tative and qualitative information used in the
evaluation process. Include evaluation of user
conflicts and competition, harvester response,
and unintended effects.
b. Identify similarities between bycatch and
other management problems.
c. Assess the contribution of current manage-
ment schemes and regulations to bycatch
problems.
d. Ensure that decision makers and stakehold-
ers are informed of the relative precision and
accuracy of information used in the evaluation.
e. Consider fisherman response to bycatch
regulations and the economic and social im-
pacts of the regulations.
Strategy 2
If existing measures do not adequately address
denned management goals, develop, evaluate, and
prioritize potential alternatives.
a. For each alternative, identify factors that af-
fect bycatch, bycatch mortality, species popu-
lation levels, and social, economic and eco-
system effects.
b. Identify information requirements and avail-
ability to successfully implement alternative
management measures.
c. If necessary, (1) develop alternatives that in-
volve incentives/disincentives, compensation
programs, or other market-based or indi-
vidual responsibility approaches; (2) seek in-
formation on pertinent solutions from other
regions; and (3) identify opportunities to
increase compliance with mitigation measures.
d. Identify legal or jurisdictional constraints to
proposed management alternatives.
e. Ensure that all interested groups are provided
opportunities to become involved in devel-
oping and evaluating alternatives, and not
merely comment on proposed plans.
f. Ensure that alternatives consider industry
views and agency/council policy.
Strategy 3
Develop an implementation plan based upon a
preferred alternative that includes monitoring
and enforcement measures.
Strategy 4
Expand the capacity of individual fishing op-
erations to reduce bycatch.
a. Examine incentives to develop technologies,
fishing practices and monitoring methods to
reduce bycatch and bycatch mortality.
b. Encourage mechanisms to fund, share, and
transfer new and improved technologies and
fishing practices, and to involve all interested
groups in their design, testing, and monitor-
ing.
IV: Implement and Monitor the
Preferred Alternative
Effective monitoring programs require assess-
ment of bycatch and the population, ecosystem,
social, and economic effects of the mitigation
measure. Implementation of the preferred alter-
native requires the support of concerned inter-
ests, and cooperation and coordination among
the fishing sectors, managers, enforcement agen-
cies, and scientists.
Strategy 1
Ensure coordination with domestic and inter-
national organizations.
a. Identify opportunities for cooperative plan-
ning to eliminate inconsistencies among state,
federal, tribal, and international fishery man-
agement organizations.
b. Promote international agreements for effec-
tive bycatch management of transboundary
or straddling stocks and highly migratory
stocks.
Strategy 2
Implement monitoring systems.
a. Identify opportunities for cooperative data
collection, especially with fishermen and
processors.
b. Evaluate monitoring and enforcement alter-
natives for practicality, cost, and effectiveness.
National
Perst
c. Identify opportunities for coordinating data
management for cost-efficiency and to avoid
duplication of effort.
d. Provide for timely communication of fish-
eries data among fishermen and managers.
e. Routinely evaluate monitoring effectiveness,
including social and economic factors; in-
corporate results into research and manage-
ment planning.
Strategy 3
Implement an enforcement and compliance system.
a. Identify opportunities for cooperative en-
forcement with other involved agencies (e.g.,
the U.S. Coast Guard and state, territorial,
and tribal agencies).
b. Identify opportunities for cooperative com-
pliance efforts with the commercial and rec-
reational fishing communities (e.g., self-re-
porting, dealer reporting).
c. Evaluate new enforcement technologies that
can be used to improve or reduce the costs
of compliance.
d. Routinely evaluate factors contributing to
noncompliance; incorporate results into re-
search and management planning.
V: Improve Communications on
Bycatch Issues
Priority must be given to improving communi-
cation among concerned interests on bycatch
issues and achievements, and to providing op-
portunities for interactions.
Strategy 1
Identify outreach contacts for the exchange of
by catch-related information.
a. Develop, update, and distribute lists of gov-
ernment, industry, conservation, professional,
and other organizations interested in bycatch,
including contacts at each.
b. Coordinate with the NOAA Office of Pub-
he Affairs to develop, update and distribute a
list of media contacts (trade publications, gen-
eral news media, and conservation newsletters).
Strategy Z
Provide accurate and timely information on
bycatch-related information issues, regulations,
and activities.
a. Distribute timely reports on the status of
bycatch and on progress in reducing bycatch.
b. Distribute timely and accurate information
on regional bycatch regulations.
Strategy 3
Establish partnerships to prepare and distribute
bycatch information.
a. Work with partners to develop regional and
national information bycatch "media kits,"
including a glossary of terms, pertinent laws
and regulations, visuals, NMFS contacts, and
World Wide Web sites.
b. Work with partners to compile and update
a computerized bibliography of bycatch lit-
erature.
c. Prepare articles for lay audiences.
d. Sponsor — in cooperation with Sea Grant,
industry associations, and interstate marine
fisheries commissions — technology-transfer
workshops to introduce gear innovations and
new fishing practices.
e. Prepare national and regional bycatch ex-
hibits for trade and boat shows, professional
society meetings, and other general public
and industry displays.
National Goal
and Objectives
f. Archive bycatch-related informational ma-
terials produced by external organizations.
VI: Improve the Effectiveness of
External Partnerships
Fishermen, managers, scientists, conservationists,
and other interested groups must work together
to craft a balanced approach to addressing bycatch
issues. NMFS and its partners must develop ways
to strengthen and expand cooperative relationships
to meet common bycatch management goals.
in
Strategy 1
Create opportunities for partner involvement
planning and monitoring bycatch reduction.
a. Promote a cooperative network of partners
in the coordination of bycatch planning and
research.
b. Develop infrastructure for long-term, con-
tinuous working relationships with partners
to address bycatch issues.
c. Sponsor symposia and conferences for part-
ners to exchange information and identify
needs on bycatch technology and management.
d. Solicit partners' views on bycatch research
needs.
e. Seek opportunities to provide incentives for
industry-sponsored gear studies, experimen-
tal fisheries, and/or development of innova-
tive management measures.
f. Inform partners of Saltonstall-Kennedy, 1
MARFIN 2 and other solicitations for
bycatch grants and contracts, through Web
sites, public and trade media, and special
bulletins.
Strategy Z
Provide easy access to NMFS bycatch databases.
VII: Coordinate NMFS Activities to Effec-
tively Implement the Bycatch Plan
Effective communication, planning, and coor-
dination among NMFS program offices and
other NOAA units is required to make the best
use of available fiscal and human resources, avoid
duplication of effort and programmatic activi-
ties, and enhance overall efficiency of the agency
to implement bycatch research and management
initiatives.
Strategy 1
Integrate bycatch management needs and pro-
grams within NOAA and NMFS.
a. Provide for NMFS Offices of Protected
Resources and Enforcement, Sustainable
Fisheries, and Science and Technology,
NOAA General Counsel for Fisheries, and
NOAA Sea Grant representation in the
bycatch planning system.
b. Integrate protected resources objectives into
the bycatch plan.
Strategy 2
Develop regional implementation plans consis-
tent with the national goals and objectives.
Strategy S
Develop or identify funding sources for meet-
ing the objectives of the bycatch plan.
The Saltonstall-Kennedy (S-K) Grant Program is a competitive program that provides grants for research and development
projects to benefit the U.S. fishing industry.The S-K Act, as amended [15 U.S.C. 713(c)(3)] is the program's statutory authority.
The Marine Fisheries Initiative, or MARFIN program, brings together scientific, technical, industry, resource conservation, and
management talents to conduct cooperative programs to facilitate and enhance the management of the marine fishery resources of the
Gulf of Mexico and South Atlantic.
kaI Recommendations
Some general issues of bycatch are com-
mon to all regions — concern about waste, im-
pacts on populations taken as bycatch (whether
finfish, invertebrates, mammals, turtles, or birds),
and impacts on other fisheries. Bycatch issues in
the separate regions and in the diverse fisheries
within regions can be very different in nature,
information needs, and potential solutions to
problems.
The following recommendations focus on
determining the magnitude of bycatch, assess-
ing the impact of bycatch, evaluating the effec-
tiveness of current bycatch management mea-
sures, identifying potential management alter-
natives, evaluating the impacts of bycatch miti-
gation alternatives, implementing alternative
management measures, and assessing the ad-
equacy of monitoring programs. They identify
bycatch research and management needs com-
mon to all regions, and are based on findings of
the national bycatch assessment that was con-
ducted during development of this plan. Spe-
cific regional recommendations are included in
the second section of the plan at the conclusion
of each regional discussion. Full implementation
of these recommendations will require coop-
eration among all concerned interests, an orga-
nizational commitment to bycatch reduction, and
stable long-term funding dedicated to bycatch
management and biological, social, economic,
and ecosystem research on bycatch.
These recommendations are not listed in
order of their priority. Actual priorities must be
determined on a fishery-by-fishery basis through
a process that includes all stakeholders in the fishery.
Monitoring and Data
Collection Programs
• Develop a fully integrated scientific approach
to the collection of biological, economic, and
social data on bycatch.
• Develop strategies for the long-term collec-
tion of fully integrated reliable, scientifically
valid data that provide fishery-specific and
species-specific estimates of total catch, as
well as spatial and temporal variabilities in
bycatch and bycatch mortality. Strategies
could include the use of at-sea observer pro-
grams, satellite or other at-sea monitoring
technologies, logbooks, fish tickets, or indus-
try surveys.
• Where appropriate, increase the level and
broaden the scope of observer programs suf-
ficiently to allow quantitative estimates of
total catch, discards, and incidental takes of
living marine resources, with acceptable lev-
els of precision and accuracy, for inclusion
in stock assessments. A review of observer
coverage levels as well as observer data col-
lection methods and associated catch esti-
mation procedures should be initiated to
ensure that these programs meet the expec-
tations of scientists, managers, and the in-
dustry cost-effectively
National
RecoHiHi£tu(a±u»v
• Develop strategies to distribute observer ca-
pability among the various fisheries requir-
ing coverage, with the goal of completing
basic quantification of bycatch. In coopera-
tion with appropriate fishery management
councils and industry representatives, develop
and implement at-sea observer programs in
fisheries where coverage is required.
• Resolve legal and legislative constraints on
long-term funding of data collection pro-
grams.
• Develop adequate funding and staff resources
for a long-term fishery observer capability.
• Pursue options for the procurement of ob-
server services that would reduce the po-
tential for conflicts of interest, and provide
incentives for quality observers to remain
with the program.
• Integrate collection of total catch and
bycatch statistics into the core statistics pro-
gram of NMFS.
• Collaborate with the fishing industry to bet-
ter utilize industry resources to collect
bycatch information.
• Develop methods to assess unobserved mor-
tality.
• Evaluate the effectiveness of bycatch moni-
toring and data collection methods, and in-
corporate the results into research and man-
agement planning.
Gear Technology and
Selectivity Research
• Increase regional conservation engineering
programs to develop, test, and certify spe-
cies- and size-selective fishing gears to ad-
dress critical conservation programs in the
region (e.g., groundfish, scallops, protected
species). This effort should make maximum
use of existing expertise in states, universi-
ties, and the industry.
• Allocate additional observer sea-days to
evaluate new or existing technologies or to
certify modifications to existing gear to al-
low fisheries to proceed under the bycatch
constraints or potential biological removal
limits.
• Provide adequate funding for research and
development capabilities in gear technology.
• Develop and implement methods for assess-
ing the response offish to fishing gear to aid
in the design of more selective fishing gear
and to promote high survival of bycatch.
Effects of Bycatch
• Improve methods to assess the population,
ecosystem, social, and economic effects of
bycatch, and the effects of management al-
ternatives for reducing bycatch.
• Develop a research program to estimate un-
observed fishing mortality and its effects on
populations of living marine resources.
Incentive Programs
• Evaluate existing incentive programs and
their effectiveness to minimize bycatch and
bycatch mortality.
• Identify new solutions that increase incen-
tives to minimize bycatch and bycatch mor-
tality.
• Identify legal impediments that prevent
implementation of incentive programs.
• Encourage research on market-based incentive
programs, including compensation programs, that
could be effectively monitored and enforced with-
out undue costs to the agency or industry.
Conservation and
Management Measures
• Assess the effectiveness of current manage-
ment measures to minimize bycatch.
• Develop performance measures to assess the
bycatch effects of proposed conservation and
management actions.
• Identify and implement more effective man-
agement measures to reduce bycatch.
• Establish monitoring and enforcement com-
pliance programs to implement and evalu-
ate management measures in terms of ex-
pected bycatch population, ecosystem, so-
cial, and economic effects.
Information Exchange and
Cooperation
• Improve public access to bycatch informa-
tion.
• Develop information exchange and distri-
bution programs for the recreational and
commercial fishing sectors, other manage-
ment agencies and the general public con-
cerning the magnitude of bycatch and ef-
forts to reduce it.
• Promote partnerships to increase informa-
tion sharing with government and nongov-
ernment scientists.
• Develop infrastructure for long-term coop-
erative working relationships on bycatch
management with industry, conservation
groups, fishery management councils, inter-
state commissions, tribal organizations, and
other agencies and organizations.
regional
JfRSPECTIVES
>vth
east Fisheries
Regional Characteristics
Northeast fisheries (Maine south to north-
ern North Carolina) generate about three-quar-
ters of a billion dollars in ex- vessel revenue per
year, and employ about 35,000 fishermen (both
full and part time; NEFSC 1995). The greatest
volume of landed fish is derived from small
pelagics (menhaden and Atlantic herring); the
greatest value of wild-caught species is from
American lobster, sea scallop, menhaden,
monkfish (goosefish), and Atlantic surfclam.
Groundfish fisheries, targeting gadoids (cod-like
fish) and flounders in New England, and sum-
mer flounder, scup, and black sea bass in the mid-
Atlantic region, collectively generate substantial
landings and income, although many of these
species have been severely overfished, and popu-
lations and landings have declined greatly.
Groundfish fishing is primarily by otter
trawling, which accounts for about 70% of land-
ings. The target species or species assemblage of
trawlers can be quite diverse, and is dictated pri-
marily by where and when fishing occurs
(Gabriel 1993). In the Gulf of Maine, otter trawl
target species include cod and mixed flatfishes
(witch flounder and American plaice; Murawski
et al. 1991). On Georges Bank, cod, yellowtail
flounder and mixed species are generally targeted
(Overholtz and Tyler 1985). In Southern New
England, groundfish fisheries primarily target
whiting (silver hake), yellowtail flounder, winter
flounder, and monkfish (NEFSC 1995). In the
Middle Atlantic, groundfish trawling targets sum-
mer flounder, scup, black sea bass, monkfish, win-
ter flounder, tautog, and a variety of other spe-
cies (Shepherd and Terceiro 1994; Gabriel 1996).
In the Gulf of Maine, fixed-gear fisheries
using gill nets and set lines (locally termed "tub-
trawls" or "longlines") target primarily cod, pol-
lock, and white hake. Groundfish gill nets are
increasingly being used to target monkfish, par-
ticularly as effort-control programs attempt to
limit fishing on traditional groundfish species.
Fishing for spiny dogfish has intensified in re-
cent years as other species have declined. Gill
netting for dogfish occurs in summer and early
autumn in the Gulf of Maine, and during the
winter off North Carolina, as the species mi-
grates southward seeking warmer waters (Rago
et al. 1994) .Trawl fisheries for dogfish occur prin-
cipally around Cape Cod. Most recently, a di-
rected monkfish gill-net fishery has begun to
target the species, particularly in deep waters of
the Mid-Atlantic.
Lobster landings are mostly taken with
baited traps, with about 70% of landings from
the Gulf of Maine (Maine, Massachusetts, and
New Hampshire; NEFSC 1996a). Some land-
ings of lobster occur by otter trawling, where it
is legal to do so (e.g., outside of Maine). Sea
scallop landings are derived principally from
dredge fisheries (particularly on Georges Bank
and in the Middle Atlantic; NEFSC 1996b).
Trawling and diving account for the rest of scal-
lop landings. Other important invertebrate fish-
eries are for northern shrimp (trawls and pots),
surfclam/ocean quahog (hydraulic dredges), and
two species of squids (trawls).
Recreational fishing is a significant com-
ponent of the region's fisheries, accounting for a
substantial proportion of the catch of a number
of species, including bluefish (~80% of catch),
summer flounder, striped bass, scup, black sea bass,
winter flounder, cod, and large pelagics. In 1996,
about 3 million recreational anglers took 23
million fishing trips in the Northeast.
Regional Bycatch Issues
Fishery Resources
Regulatory discards (i.e., discard of under-
sized or trip-quota limited stocks) are an issue
in the Northeast region's groundfish fisheries.
Historically, managers often selected minimum
legal sizes for groundfish that resulted in the se-
lection of undersized fish, given the characteris-
tics of nets used in the fishery, often resulting in
substantial discards (Alverson et al. 1994). Regu-
latory discards also occur when catches of cer-
tain stocks are limited by trip quotas. Managers
are attempting to reduce regulatory discards, but
this must be accomplished against a background
of increasingly severe regulations intended to
dramatically reduce fishing mortality on nearly
two-thirds of the region's resources, which are
considered overfished and at a low level of his-
torical abundance (NEFSC 1995).
Although the total magnitude of discards
in the region's fisheries is not great relative to
some other areas of the nation, discards of fin-
fish and shellfish can represent a significant pro-
portion of the catch, and thus an important
source of fishing-related mortality. One of the
factors that contributed to high discard rates was
the open-access nature of most fisheries, which
contributed to very high fishing mortality rates
and recruitment and growth overfishing. Because
abundance of large fish declined due to over-
fishing, many of the region's fisheries became
"recruitment fisheries" (i.e., targeting incoming,
but infrequent recruitment events). Differential
targeting of these small fish, combined with in-
appropriate mesh size and inadequate enforce-
ment sometimes resulted in extremely high dis-
card rates and economic and biological waste of
the resource.
Management programs that control fish-
ing mortality rates have been adopted for most
of the region's fisheries. For example, since 1994
the groundfish and sea scallop fisheries through-
out the Northeast are regulated primarily by
maximum allowed days at sea per vessel. The
program substantially reduced the allocations of
allowed fishing days in both fisheries, over the
base periods before effort-based management.
The effects of effort management on discards
are not precisely known. Eventually, however, it
is anticipated that with sufficient effort reduc-
tion, combined with other management regu-
lations, the fisheries will become less dependent
on incoming recruitment, thus reducing the
potential catch of undersized animals and, thus,
regulatory discards.
Northeast
Fisheries
One consequence of reduced target spe-
cies abundance is that mobile gears are towed
for longer intervals between haulback. Towing
times of three hours or more are not uncom-
mon for the New England and Middle Atlantic
groundfish trawl fishery (Murawski 1996). Be-
cause the species composition of individual
catches diversifies as various depth and bottom-
type habitats are crossed the fisheries have be-
come less directed to a single target species or
group. The time of towing has been found to
significantly influence the overall discard rate of
trawl fisheries (Murawski 1996).
Trip limits contribute to the discarding of
three species — summer flounder, haddock, and
Atlantic cod. Trip limits for summer flounder
are invoked when individual states approach their
allocated share of regionwide total allowable
catch (TACs). Depending on both the length of
time trip limits are in effect, and on targeting by
the fleet, discarding of fish may be significant.
Sea sampling of this fishery is conducted to esti-
mate trip-limit-induced discarding, and these
projected discards are included inTAC calcula-
tions (NEFSC 1996a). The potential for sum-
mer flounder discards in both the commercial
and recreational fisheries represents a controver-
sial issue in both the assessment and the man-
agement of this recovering stock.
Currently, trip limits for haddock are set at
1,000 pounds per day fished on a trip, up to a
maximum of 10,000 pounds, until such time as
75% of the target TAC has been caught. The
haddock trip limit then reverts to 1,000 pounds.
This trip limit scheme was set to remove eco-
nomic incentives to target aggregations of this
critically overfished species. Obviously, if man-
agement efforts are successful in stock rebuild-
ing, then the trip limit will become constrain-
ing to an increasing fraction of trips. Major un-
certainty exists in establishing trip limits that
would minimize discards of haddock taken as
truly accidental catches, while not encouraging
vessels to target them or to fish in areas where
the incidental catch of haddock is more prob-
able. Cod trip limits have been invoked for the
Gulf of Maine region to limit exploitation of
the cod resource in that region. It is too early to
evaluate the effects on discard rates of this change
in the management system.
Minimum size regulations, as well as eco-
nomic factors contribute to relatively high dis-
card rates in a number of mid-Atlantic fisheries,
especially for scup and, to some extent, black sea
bass. Discard estimates for these species are so
tentative, and potentially of such magnitude, that
the lack of better discard information precludes
the assessment of these stocks by traditional
catch-at-age methods.
Small-mesh fisheries in the Northeast Re-
gion have undergone a great deal of scrutiny, as
managers have sought to minimize the catch of
undersized groundfish, particularly in trawl fish-
eries. The trawl fishery for northern (pandalid)
shrimp now requires the use of finfish excluder
devices, which, when fished properly, reduces the
overall proportional weight of nonshrimp catch,
particularly of flatfish and gadoids (NEFSC
1995). Sea sampling of this fishery has shown
that shrimp catch rates are slightly improved
when excluders are used, possibly due to changes
in hydrodynamics of the net. Bycatch rates of
some smaller groundfish may have increased (e.g.,
very small flounders and pollock), but overall,
the program has reduced finfish bycatch from
about half of the total quantity of catch (in
weight) to about 10% (Richards and
Hendrickson, unpub.).
Other small-mesh trawl fisheries of the re-
gion targeting silver and red hakes, herring,
mackerel, squids, butterfish, ocean pout, and dog-
fish are subject to a performance criterion ot
less than or equal to 5% of the total catch com-
prised of regulated groundfish species (e.g.. cod.
Regional
?&spectWG*
haddock, redfish, pollock, white hake, and five
flounder species). On Georges Bank, a small-
mesh fishery is allowed for whiting, but only in
prescribed locations (e.g., Cultivator Shoals) and
only in summer months. Some fisheries have
been curtailed altogether or geographically re-
stricted to meet this performance criterion. Squid
fisheries in the mid-Atlantic and southern New
England potentially generate discards of a num-
ber of commercial species, but sea sampling has
not been of sufficient magnitude or distribution
among various components of the squid fishery
(e.g. refrigerated sea water "wet" boats, freezer
trawlers, offshore vs. inshore fisheries) to ad-
equately characterize discards.
Bycatch is also an important source of
allocative conflict among the region's fishermen.
For example, Atlantic cod are targeted primarily
by three gear types — otter trawls, gill nets, and
demersal longlines. Mobile gears tend to have
the highest overall discard rates. Gill nets using
appropriate mesh are generally more selective
than both trawls and hooks. Gear sectors are in
competition for small overall targetTACs for cod,
and regulations are likely to change the relative
proportions of the catch derived by the various
gear types. Debate continues on the merits of
explicit policy decisions to allocate shares of the
catch to gears that exhibit low discard rates. The
issue is particularly problematic, given the need
to reduce overall harvest rates by about 80% from
1994 levels (NEFSC 1994a).
Kept bycatch can also be an important
source of overall income to specific fisheries and
a source of conflict when the bycaught species
is targeted by other fleets. For example, monkfish
have become the single most valuable finfish
taken in the offshore fishery, generating $33
million ex- vessel in 1995 — nearly equal to the
value of cod, haddock, and yellowtail flounder
combined. A large portion of the monkfish catch
is bycatch in the sea scallop dredge fishery; this
bycatch provides significant income to this fish-
ery. Monkfish are being increasingly targeted by
trawlers as an alternative to declining ground-
fish resources, and additional gears, including gill
nets, are being used to target monkfish. Thus,
there are conflicts regarding the appropriate use
of the resource, particularly as restrictive regula-
tions are enacted.
The greatest magnitude of discarded catch
occurs when low-valued species are taken coin-
cident with target species (Murawski 1994,
NEFSC 1995). These discretionary discards can
account for 40% or more of the volume of the
catch. Recent diversification of the fisheries has
resulted in greater utilization of these low-val-
ued species (e.g., dogfish), but others still have
little market value (e.g., small skates, sculpins)
and continue to be discarded in quantity.
Recreational fisheries of the region are re-
sponsible for a substantial quantity and propor-
tion of catch discarded (VanVoorhees et al. 1992).
These discards are due to regulatory (fish below
minimum sizes or bag limits), discretionary (un-
wanted species or sizes), or catch-and-release
considerations. Overall, the rate of recreational
fishery discard has increased steadily, from about
30% of the catch in 1980, to about 60% of the
catch in 1996 (NMFS, unpublished data). De-
pending on the species, the proportion of the
recreational catch that is released alive varies
considerably with high and low release rates of
25—70% typical for unregulated species, and 33—
70% typical for regulated species. Most of the
increase has been due to the imposition of size
and bag limits in specific fisheries (VanVoorhees
et al. 1992). Not all discarded recreational fish
die, and the proportion surviving release can be
a major factor in stock assessments of species,
including striped bass, bluefish, and summer
flounder.
t Northeast
Fisheries
Protected Species
Takes of marine mammals and sea turtles
are problematic in several of the region's fisher-
ies (Blaylock et al. 1995). Bottom-tending gill-
net fisheries targeting groundfish in the Gulf of
Maine and Southern New England entangle
harbor porpoise in numbers sufficient to be of
concern to the long-term stability of the harbor
porpoise resource (NEFSC 1995). Reasons for
these entanglements are not clear, and may vary
in location from year to year. Takes of harbor
porpoise in these fisheries are substantially above
the "potential biological removal" of the stock,
and bycatch mitigation is required. Gill-net fish-
eries in the Gulf of Maine also entangle large
whales, including the endangered right whale;
take-reduction team activities have been focused
on these fisheries to reduce interactions. Gill-
net fisheries also result in mortalities of some
seabirds, including shearwaters, gulls, and gan-
nets. Middle Atlantic coastal gill-net fisheries also
take harbor porpoises and bottlenose dolphins.
Pelagic drift-net and longline fisheries for
tunas and swordfish result in takes of a variety of
marine mammals and turtles (Blaylock et al.
1995). Pelagic longlines, primarily set for sword-
fish and tuna, take leatherback and green sea
turtles, as well as pilot whales and dolphins. Pe-
lagic drift-nets take marine mammal species, such
as saddleback dolphin, bottleneck dolphin, and
Risso's grampus dolphin, and occasionally other
species, such as pilot whales, beaked whales, and
other dolphins.
Although infrequent, entanglements of
whales in lobster gear are of particular concern.
Given the status of right whales (Blaylock et al.
1995), any fishing activities that generate mor-
talities of this species are subject to mitigation
measures. Thus, the lobster pot fishery has been
reclassified as Category I (likely to exceed po-
tential biological removal for protected species)
under the Marine Mammal Protection Act on
the basis of right whale interactions.
Nearshore trawl fisheries in the Middle
Atlantic have generated some takes of sea turtles,
particularly in summer months. The use of turtle
excluder devices in coastal trawl fisheries in the
Middle Atlantic, when turtles are present, has
been proposed. Coastal gill-net fisheries in the
Middle Atlantic set for monkfish, dogfish, blue-
fish, and other species are currently being moni-
tored to assess their potential impacts on marine
mammal species.
Regional Bycatch Programs
Bycatch monitoring and assessment pro-
grams are an integral part of bycatch manage-
ment programs in the Northeast.
Bycatch Monitoring and Assessment
Bycatch in Northeast commercial fisheries
is monitored primarily through the Fishery
Observer Program of the Northeast Fisheries
Science Center (NEFSC 1995). Several states also
undertake some monitoring activities in their
waters.The Fishery Observer Program is funded
through several NMFS offices, and primarily
focuses on estimates of takes of protected spe-
cies. A private contractor currently coordinates
the deployment of observers. Training of at-sea
observers is conducted by NEFSC staff, who are
also responsible for archiving observer data files.
This program has operated since 1989.
The observer program conducts about
1 ,500 vessel deployments per year, comprising
about 3,000 days at sea. The vast majority of at-
sea observer coverage for the region's fisheries is
expended to monitor protected species takes. The
sink gill-net fishery in the Gulf of Maine ac-
counts for about one-third of the sea sampling
coverage due to the need to monitor harbor
porpoise takes. About 6% oi~ the sink gill-net
trips are sampled annually. Proportionally, the
most heavily sampled fisheries are the drift-net
fishery for swordfish and the purse-seine fishery
for tuna. Coastal trawl and gill-net fisheries in
the Middle Atlantic Region are monitored for
takes of turtles and marine mammals.
Days-at-sea allocated for nonprotected spe-
cies surveillance have been prioritized to moni-
tor fisheries for northern shrimp, summer floun-
der, sea scallop, and to a limited extent, large-
mesh groundfish trawlers. Overall, however, the
level of coverage of observed trips is very low
(much less than 1% of the fleet-days at sea) and
insufficient to generate reliable estimates of dis-
card mortalities for inclusion in stock assessment
for all but a few species due to the lack of preci-
sion and concerns that such few trips may be
biased. The level of coverage is not sufficient for
evaluating the effectiveness of bycatch mitiga-
tion measures in most fisheries.
Preliminary analyses of statistical proper-
ties of sea sample data indicate that the sensitiv-
ity of discard estimates to the design features of
sampling programs, the level of sampling, the
choice of estimator, and the assumption that se-
lected trips are unbiased (Brodziak 1991, Hayes
1991, NEFSC 1991).
For some fish stock assessments, bycatch
mortalities are such a large fraction of the catch
that they cannot be ignored without seriously
compromising the assessment. These cases in-
clude yellowtail, summer, witch, and winter
flounders, American plaice, and scup. In these
cases, analysts have used available discard sam-
pling information, and sometimes have com-
bined historical information from captains' in-
terviews and estimates derived from use of fish-
ery-independent resource surveys (e.g., the yel-
lowtail flounder assessment in NEFSC (1994b)).
Historical size-selection patterns of the fishery
have been applied to population-length com-
positions from survey data to estimate the pro-
portion of the catch likely discarded by the fish-
ery. Such methods have produced surprisingly
consistent estimates, but are useless when the
selection patterns of the fishery change (due to
increased mesh, population size, and other regu-
lations) .
Discard data are also sought from fisher-
men in their mandatory logbook submissions.
Preliminary information from this self-report-
ing program was correlated with observer esti-
mates from identical trips (NEFSC 1996a). Al-
though analyses suggest no obvious discrepan-
cies, this may be due to the effect of the pres-
ence of the observer. Much more analysis of in-
formation and communication with fishermen
is necessary before self-reported estimates of dis-
cards can routinely be incorporated into stock
assessments.
Recreational discards are based almost ex-
clusively on interview information provided as
part of the marine recreational fishery statistics
survey (VanVoorhees et al. 1992). Private boats
have not been subject to sea sampling coverage,
and only a few party boats have been so sampled
to date under the Northeast fishery observer
program.
Bycatch Management
Bycatch management in northeast fisher-
ies uses minimum mesh size regulations, trip lim-
its, finfish excluder devices, and closed areas,
among other measures, to reduce bycatch of fin-
fish and protected resources.
fishery Resources
Bycatch management has been fundamen-
tal to the development of overall proposals to
eliminate overfishing and rebuild depleted stocks.
Managers are particularly concerned that valu-
able fish are not wasted due to regulatory-in-
duced discards, particularly given the depleted
nature of many of the Northeast Region's re-
sources. Nevertheless, the overriding concern at
this point is to eliminate the overfished condi-
tion of most of the region's stocks, and to re-
build them.
Amendment 5 of the Northeast Multispecies
Fishery Management Plan increased trawl and gill-
net mesh sizes in most fisheries to a minimum
of 6 inches (stretched). At the same time, how-
ever, minimum fish sizes were not increased, so
as to reduce the capture of undersized ground-
fish. Because of the performance criterion for
small-mesh fisheries of <5% regulated ground-
fish, there has been increased interest in the de-
velopment of species-selective trawling gear. Vari-
ous designs are being proposed and tested for
potential application to groundfish and sea scal-
lop fisheries.
In December 1994 three large areas on
Georges Bank and in southern New England
were closed to all fishing gears, except lobster
pots, to protect groundfish resources. Southern
New England was an area of historical concen-
tration of age-two yellowtail flounder, tradition-
ally the age class most subject to discarding. The
closed areas on Georges Bank are historical con-
centration areas for haddock and cod.
Minimum net mesh sizes apply to a variety
of other fisheries in an attempt to minimize catch
of juveniles and improve yield per recruit. Be-
cause of the highly mixed nature of catch, and
the fact that different target species have differ-
ent optimum mesh sizes, no one mesh is best for
all cases.
Trip limits apply for the summer flounder
fishery, when individual state allocations of the
total allowable catch have been met. Likewise, a
trip limit for haddock is applied year-round.
Managers have sought alternatives to the trip
limits that would give equivalent conservation
benefits while reducing the need for regulatory
discards. Alternatives considered include ex-
panded closed areas, larger mesh sizes, and closed
seasons.
Other regulations designed specifically to
address bycatch have included mandatory use of
finfish excluder devices in the northern shrimp
fishery and increased minimum net mesh and
ring size requirements for sea scallop dredges (the
top of the dredge is usually a net, while the bot-
tom and sides are steel rings). Discretionary dis-
cards have not been the subject of specific regu-
lations.
Protected Resources
Managers are attempting to reduce harbor
porpoise takes through a series of phased time
and area closures. These closed areas potentially
benefit the overfished groundfish species as well.
Specific boundaries of closure areas are prima-
rily based upon the historical "hot spots" of por-
poise bycatch. Although the timing of the peak
bycatch may change from year to year, the "hot"
locations remain relatively constant.
Acoustic deterrence of harbor porpoise
from gill nets is also under experimentation.
Some preliminary experiments with these
"pingers" have been promising, but it is unclear
if the use of these devices as a general bycatch
reduction measure would be sufficient by them-
selves, or in combination with reduced area clo-
sures, in decreasing harbor porpoise mortalities
below the potential biological removal. A take-
reduction team is examining information from
field experiments and related modeling and fish-
ery observer data to determine their effective-
ness.
The swordfish drift-net fishery in the At-
lantic has been responsible for hundreds of ma-
rine mammal mortalities. A long-term average
is approximately one marine mammal taken per
overnight set. The offshore species taken include
the critically endangered North Atlantic right
whale, as well as sperm whale, common dolphin,
and five species of beaked whales. The fishery is
currently under an emergency closure and may
tonal
Vevspecturts
only be reopened with very stringent require-
ments placed on it by the Atlantic Offshore Take-
Reduction Team (TRT). These requirements
include time/area closures, open access to the
swordfish quota (to eliminate the derby fishery),
use of a net set allocation, limited entry, and 100%
observer coverage.
The Atlantic longline fishery also has also
come under scrutiny from the Offshore TRT as
that fishery takes a large number of marine mam-
mals and sea turtles. However, in the longline
fishery the vast majority of these takes are re-
leased alive. Questions about the long-term sur-
vival of these released animals are being asked
by the team, and studies are being initiated to
determine their fate. Several effort-reduction
measures on the longline fishery have been in-
troduced, including a limit on the number of
hooks and total length of line deployed, limited
entry to the fishery, increased observer cover-
age, reverse retrieval of gear, and a requirement
to move to a new area after a marine mammal
interaction.
The Gulf of Maine lobster pot fisheries are
currently designated as Category I fisheries due
to serious injuries and mortalities of right and
humpback whales. Gear modifications and gear
marking requirements have been developed to
reduce the likelihood of such interactions. Be-
cause of the relatively rare occurrence of these
interactions, the precision and accuracy of these
estimates remains low.
Regional Recommendations
The most important bycatch monitoring
need is for data collection programs sufficient to
estimate the magnitude of bycatch mortalities
and incidental catch of protected species for in-
clusion in stock assessments. While observer cov-
erage does not need to be universal, current cov-
erage for most fisheries is not high enough to
estimate fish discards or protected-species
bycatch with acceptable precision for inclusion
in stock assessments or for impact evaluation.
Given the diversity of regional fisheries, the
amount and breadth of observer coverage need '
to be expanded greatly if the goal of adequate
discard estimates for all important resources is
to be achieved.
There is also a need to provide ongoing
advice to managers on whether the use of spe-
cific gears or fishing in particular areas will com-
promise their bycatch reduction goals. This can
be best accomplished by using some observer
coverage in an experimental, rather than moni-
toring, mode. This approach needs to be ex-
panded, particularly if greater emphasis is placed
on gear-based solutions to bycatch problems.
Assessing the population consequences of
bycatch involves evaluating all sources of mor-
tality on harvested populations, including land-
ings, natural deaths, and injuries and mortalities
of animals that encounter the gear, but are not
retained (e.g., fish that squeeze through the
meshes, are injured by rollers, or that drop off
prior to the gear being hauled aboard). Collect-
ing discard data must be included in a core sta-
tistics program that provides mortality estimates
with acceptable precision. Unobserved mortali-
ties of nonretained animals are potentially the
most difficult to measure, and will require a com-
bination of field and laboratory experiments to
obtain usable estimates.
Evaluating the economic and social impacts
of bycatch requires information on factors, such
as costs of mitigation alternatives, prices, and
participation by various fleet sectors. Without
such information, evaluation of appropriate miti-
gation measures will be subjective.
Regionally, emphasis on the continued re-
ductions in fishing effort prescribed in the
Northeast Multispecies and Sea Scallop Fishery
Management Plans may be the single most ef-
Northeast
ttikerits
fective bycatch mitigation measure currently in
place. These reductions, if effective in reducing
fishing mortality rates, should decrease effort
directed to recruits and thus increase retention
rates. However, until stocks are rebuilt and age
compositions of the populations are expanded,
there will be a great emphasis by management
on gear-based solutions, trip-based quotas for
some species, and closed areas.
Effort reductions should also reduce some
takes of protected species in fixed-gear fisheries.
In the short term, however, efforts to reduce
bycatch of protected species will most likely fo-
cus on seasonal area closures combined with gear
technology adaptations.
Managers, fishermen, environmental
groups, and the general news media have all ex-
pressed the need for timely, accurate, and widely
available information on discard rates of various
fisheries and fleet sectors. Given the increased
profile of bycatch issues, additional resources al-
located to effective communication of bycatch
goals, programs, and information are required.
Following are specific recommendations for
Northeast fisheries:
• Increase the level and broaden the scope of
the fishery observer program sufficiently to
allow quantitative estimates of discards of
fishery resources and incidental catch of pro-
tected species, with acceptable levels of pre-
cision and accuracy for inclusion in stock
assessments.
At the discretion of the Regional Adminis-
trator, allocate additional observer sea-days to
evaluate new or existing technologies or to
certify modifications to existing gear to allow
fisheries to proceed under the bycatch con-
straints or potential biological removal limits.
Increase the ability to assess the population,
ecosystem, social, and economic effects of dis-
cards, and the impacts of management alter-
natives developed to reduce them through in-
tegrated data collection and analysis systems.
Increase research on acute and long-term
mortalities of animals encountering fishing
gears, but not retained. Specifically, evaluate the
fate of animals that escape through net meshes,
the hook-and-release mortality of recreational
fishes, and the effects of bottom-tending mo-
bile fishing gears on benthic communities.
Increase regional conservation engineering
programs to develop, test, and certify spe-
cies- and size-selective fishing gears to ad-
dress critical conservation programs in the
region (e.g., groundfish, scallops, protected
species). This program should make maxi-
mum use of existing expertise in states, univer-
sities, and the indusuy.
Develop effective information exchange and
distribution programs to communicate with
the industry, regulators, and general public
concerning the magnitude of bycatch and
efforts to reduce it.
3t Qtttf Pelagic Fisheries
Regional Characteristics
U.S. fishing vessels, both commercial and
recreational, fish for Atlantic highly migratory
species (HMS) in the Atlantic Ocean, Gulf of
Mexico, and Caribbean Sea. Commercial U.S.
fisheries for Atlantic HMS target tunas (includ-
ing bluefin, bigeye, albacore, yellowfin, and skip-
jack), tuna-like species (bonito, mahi-mahi, and
wahoo), swordfish, and sharks. Recreational fish-
eries target tunas, tuna-like species, shark, and
billfish.There is no directed U.S. commercial fish-
ery for Atlantic billfish, and the sale of Atlantic-
caught billfish in the United States is prohib-
ited. A once-popular recreational fishery for
swordfish has declined due the decrease in the
availability of swordfish in nearshore waters.
NMFS manages Atlantic tunas, swordfish,
and billfish under the dual management author-
ity of the Magnuson— Stevens Fishery Conser-
vation and Management Act and the Atlantic
Tunas Convention Act (ATCA).ATCA autho-
rizes the Secretary of Commerce, acting through
NMFS, to issue regulations to implement the
recommendations of the International Commis-
sion for the Conservation of Atlantic Tunas
(ICCAT).This international cooperative body
manages the fisheries for and conducts research
on the stocks of Atlantic tunas, swordfish, and
billfish. It does not have management authority
for Atlantic sharks, though its scientific body is
collecting data on shark bycatch in fisheries tar-
geting ICCAT species.
Because a fishery management plan for
Atlantic tunas has not yet been implemented
under the Magnuson— Stevens Act, they are man-
aged under ATCA in the United States. NMFS
is developing a comprehensive fishery manage-
ment plan for Atlantic tunas, swordfish, and sharks
that will amend the existing shark and swordfish
plans and create a new plan for tunas. Atlantic
billfish are managed under ATCA as well as un-
der the fishery management plan for Atlantic
billfish; NMFS is currently amending the bill-
fish plan to meet new requirements of the
Magnuson— Stevens Act.
i
Several stocks of Atlantic HMS have been
subjected to prolonged decline due to a combi-
nation of domestic and international overfish-
ing. In a recent report to Congress on the status
of U.S. fishery stocks relative to overfishing, At-
lantic bluefin tuna, Atlantic swordfish, the 22
species that comprise the large coastal shark
management unit, and Atlantic blue and white
marlin were identified as overfished (NMFS
1997b). Under the Magnuson-Stevens Act,
NMFS must develop rebuilding programs for
those species identified as overfished. Rebuild-
ing of Atlantic HMS stocks is complicated by
the fact that these species are fished by many
nations. For example, in 1996, 7% of Atlantic-
wide billfish mortality was attributable to US.
fishing activities; the remaining 93% can be at-
tributed to other countries. Despite the imple-
mentation of and compliance with conservation-
oriented billfish management measures (e.g.,
minimum size requirements, ban on sale) by U.S.
recreational and commercial fishermen, the rela-
tively small U.S. influence on total mortality frus-
trates domestic efforts to rebuild the stocks.With-
out the cooperation of other countries in imple-
menting and enforcing conservation-oriented
management measures, stock rebuilding is greatly
impeded. For overfished HMS stocks where the
U.S. share of total mortality is low, development
of a cooperative international strategy to slow
Atlantic-wide overfishing is essential to an ef-
fective domestic rebuilding strategy.
Regional Bycatch Issues
Bycatch issues in the fisheries for Atlantic
highly migratory pelagic species are driven by
population concerns about depleted stocks of
HMS and protected species and also by alloca-
tion concerns among user groups.
Fishery Resources
The directed swordfish fishery is limited by
regulation to longline, harpoon, and drift gill-
net gear. Catches by other gear are restricted to
bycatch trip limits of two to 1 5 swordfish per
trip, depending on gear type. Longline vessels
account for the vast majority of swordfish land-
ings, followed by drift gill-net vessels and har-
pooners. Drift gill-net vessels primarily target
swordfish, but also take tunas and sharks. Finfish
bycatch in the drift gill-net fishery includes blue-
fin tuna, little tunny, skipjack tuna, rays, and ocean
sunfish, most of which is discarded (Cramer
1996a) .The drift gill-net fishery has been closed
under an emergency rule since December 1,
1996, due to concern about interactions with
right whales.
The pelagic longline fishery for Atlantic
HMS targets primarily swordfish, sharks, bigeye
tuna, and yellowfin tuna. The longline fishery
may also retain bluefin tuna under an incidental
catch limit that is subject to target catch require -
ments.The discard of undersized swordfish, blue-
fin tuna, and billfish is an important issue in the
pelagic longline fishery for swordfish, tuna, and
sharks. In 1996 the longline fishery discarded
approximately 579 metric tons of swordfish,
equivalent to about 40,000 fish (NMFS 1997a).
Time/area closures are frequently proposed as
management measures to reduce mortality on
undersized swordfish, although further analysis
is warranted.
Bycatch of Atlantic billfish in the pelagic
longline fishery for tunas, swordfish, and sharks
is a contentious population- and allocation-re-
lated issue. Atlantic billfish (blue and white mar-
lin, spearfish, and sailfish) are prized by recre-
ational anglers and are encountered as bycatch
in the longline fishery. Due to concern about
the declining populations for these species.
NMFS prohibited the landing and sale of Atlan-
tic-caught billfish in the United States. When a
longline vessel hooks a billfish, the leader must
be cut as close to the fish as possible without
removing the fish from the water.
Estimates of the billfish bycatch discarded
dead in the U.S. commercial longline fishery in
1996 were 196.6 metric tons for blue marlin,
67.6 metric tons for white marlin, and 71.6
metric tons for sailfish (NMFS 1997a). Both blue
and white marlin are classified as overfished
(NMFS 1997b), and the stocks are estimated to
be at 61% and 32%, respectively, of the levels
needed to support maximum sustainable
yield.
Recreational and conservation groups are
very concerned that billfish mortality as bycatch
in the longline fishery is impeding recovery of
these overfished stocks. The longline industry,
on the other hand, is concerned that insufficient
data on the magnitude of landings and of post-
release mortality in the recreational catch-and-
release billfish fishery may obscure a significant
source of fishing mortality to billfish stocks. Both
user groups express concern that, because the
U.S. share of Atlantic billfish mortality is low
(generally less than 10% of Atlantic-wide mor-
tality), bycatch management for these species
must include stock-wide conservation and man-
agement measures that are adopted by all na-
tions that fish the stock.
ICC AT has recommended that the United
States implement measures designed to reduce
dead discards of Atlantic bluefin tuna captured
incidentally in the fisheries for other tunas,
swordfish, and sharks in 1996—97. Discards of
Atlantic bluefin tuna are generated by regula-
tory minimum size requirements and, of par-
ticular concern, by incidental target catch re-
quirements for the longline fishery. Longline and
drift gill-net vessels may obtain an Incidental
Catch permit that allows them to retain "large
medium" and "giant" Atlantic bluefin tuna (de-
fined by regulation as 73—81" and >81" curved
fork length, respectively) as incidental catch. The
amount of bluefin tuna that can be retained is
based on several factors, such as vessel type, lo-
cation of fishing, and season. Vessels that hold
Incidental Catch permits must meet a variety of
target catch requirements in order to retain in-
cidentally captured bluefin tuna. In 1996 U.S.
longline vessels discarded an estimated 570 dead
bluefin tuna (about 73 metric tons), and US.
drift gill-net vessels discarded an estimated 32
dead bluefin tuna (about 4 metric tons). Dead
discards of bluefin tuna for 1 996 decreased by
almost half compared with 1995 levels.
The purse seine fishery for Atlantic tunas
is a limited-access fishery that targets bluefin tuna,
particularly giant bluefin, yellowfin tuna, and
skipjack tuna. Bycatch can occur in this fishery
when vessels set on mixed schools of tunas that
include undersized fish and fish that cannot be
marketed. Discard data are generally unavailable
for several other fisheries for Atlantic highly
migratory pelagic species, including the harpoon
and handline fisheries. In these fisheries, Atlan-
tic bluefin tunas less than the minimum size are
discarded.
Bycatch of sharks, in both directed shark
fisheries and other fisheries, is of increasing con-
cern. Sharks are particularly vulnerable to over-
fishing due to most species' low fecundity, slow
maturation, and long reproductive cycles. Fur-
thermore, shark species are difficult to distin-
guish from each other, and discard data often do
not accurately reflect the species composition
of the discarded sharks. Small coastal shark
bycatch can comprise a large portion of the to-
tal catch in southeast shrimp trawl fisheries. Stock
status and basic life history are poorly under-
stood for many species of small coastal sharks,
and there is concern that high volumes of bycatch
may be depleting these populations.
AtLuttic&.gulf
Pelaqic Fisheries
Bycatch issues in the recreational fisheries
for Atlantic HMS are driven primarily by allo-
cation concerns and by the difficulty of estimat-
ing total fishing mortality in the recreational sec-
tor. Data on recreational angling are collected
through the NMFS Large Pelagic Survey, a com-
bination of dockside intercepts and phone in-
terviews conducted between Maine and North
Carolina, and by the Marine Recreational Fish-
ery Statistical Survey. NMFS also conducts tour-
nament sampling. These survey techniques esti-
mate the type and amount of fishing mortality
and fishing effort for marine and large pelagic
species from the recreational sector. Due to the
highly disparate nature of recreational fisheries,
it is very difficult to standardize techniques for
estimating fishing mortality. Also of particular
concern in these fisheries is the lack of infor-
mation on post-release mortality in catch-and-
release fisheries.
Protected Species
Concern about bycatch of protected spe-
cies is particularly high in the drift gill-net fish-
ery for tunas, swordfish, and sharks. This fishery
is classified as a Category I fishery under the
Marine Mammal Protection Act. Concern is also
high in the Category III pelagic longline fish-
ery for tunas, sharks, and swordfish.
Based on 1996 observer reports, bycatch
of protected species for drift gill-net vessels in-
cluded True's beaked whales, Sowerby's beaked
whale, spotted dolphin, striped dolphin, long-
finned pilot whales, short-finned pilot whales,
loggerhead turtles, and leatherback turtles. The
swordfish-directed drift gill-net fishery is cur-
rently under an emergency closure due to con-
cerns about bycatch of the protected right whale.
Bycatch of protected species in the 1 996 longline
fishery included leatherback, loggerhead, and
Kemp's ridley turtles, most of which were re-
leased unharmed (Cramer 1996a). Representa-
tives of the drift gill-net and longline fisheries
participated in the work of the Offshore Ceta-
cean Take-Reduction Team, which was charged,
in part, with determining how to reduce bycatch
of marine mammals in these fisheries to levels
approaching zero. The team recommended a
number of options, including time/area closures,
acoustic devices to warn cetaceans of fishing gear,
and effort controls to reduce the derby nature
of the drift gill-net fishery.
Bycatch of protected species also occurs in
the purse seine fishery for Adantic HMS. In 1996,
95% of purse seine trips were covered by NMFS-
contracted observers. Observers recorded the
capture and release unharmed of one humpback
whale, one minke whale, and six pilot whales.
No purse seine trips were observed in 1997.
Regional Bycatch Programs
Currently, participants in the HMS com-
mercial fisheries submit daily logbook reports,
weigh out and/or tally sheets, and dealer reports.
Recreational fishermen are subject to the Large
Pelagic Survey and the Marine Recreational
Fishery Statistics Survey. Atlantic bluefin tuna
fishermen are required to report their catch on
a toll-free phone line. NMFS is planning two
pilot surveys to supplement data collection in
the recreational fisheries for HMS in 1998.
In addition, scientific observer coverage of
the U.S. pelagic longline fleet was initiated by
the Southeast Fisheries Science Center (SEFSC)
in early 1992. In conjunction with the North-
east Fisheries Science Center's Woods Hole
Laboratory, the SEFSC uses contracted and
NMFS observers to collect catch-and-discard
data aboard longline vessels fishing m the waters
of the northwest Atlantic Ocean, Gulf of Mexico,
and Caribbean Sea. Selection of vessels is based
on a random sampling of the number of sets
reported by the longline fleet (approximately 5%
jiotuit
Perspectives
of sets are observed). A total of 2,857 sets was
observed by personnel from the SEFSC and
NEFSC programs from May 1992 to Decem-
ber 1996. Observers have recorded over 50,000
fish (primarily swordfish, tunas, and sharks), as
well as marine mammals, turtles, and seabirds
caught and discarded during this time period.
A higher proportion of drift gill-net trips
is sampled due to concern over potential bycatch
of protected species (marine mammals and sea
turtles). In 1996, the NEFSC placed observers
aboard six different domestic drift gill-net ves-
sels targeting tuna, swordfish, and sharks. Ob-
servers made 1 3 trips (totaling 1 40 days) on these
vessels in 1996, representing 81% of the total 16
trips made in the fishery in 1996. Bycatch man-
agement measures for the drift gill-net and
longline fisheries are being considered by NMFS
upon recommendation by the Atlantic Offshore
Cetacean Take-Reduction Team.
In response to the 1996 ICCAT recom-
mendation that calls for the United States to
adopt measures designed to reduce dead discards
ofbluefin tuna during 1997-98, NMFS has per-
formed preliminary analyses to examine the vi-
ability of different options for reducing discards.
The options being considered include changing
the current target weight catch requirement, lim-
iting the number of days per trip, and imple-
menting time/area closures. Logbook and dealer
weigh-out slips from 1991 through 1995 were
collected, and initial results indicate significant
differences between the number ofbluefin tuna
caught and discarded per trip by season and re-
gion. NMFS plans to expand these analyses to
develop more conclusive results as a basis for
management action. In the meantime, restric-
tive management measures on the target fisher-
ies in which bluefin are taken as a bycatch ap-
pear to be having an effect on bluefin discards.
Swordfish and shark quotas have been reduced
(50% for large coastal sharks), and limited entry
is scheduled to be implemented in both fisher-
ies. The recently formed HMS Advisory Panel
will assist NMFS in considering options, such as
time/area closures, to reduce discards of billfish
and undersized tunas, swordfish, and sharks.
Data on shark catch and bycatch are being
collected by ICCAT's Scientific Committee on
Research and Statistics. Increasingly concerned
about shark bycatch in Atlantic-wide directed
tuna fisheries, the committee initiated a shark
bycatch data collection program in 1995. Data
for 1996 indicate that, for the entire Atlantic, 47
shark species were taken as bycatch in longline
fisheries, 16 in drift gill-net fisheries, and 11 in
purse-seine fisheries (ICCAT 1997). Data in the
U.S. commercial shark fisheries are collected
through logbooks and dealer reporting and
through an observer program run by the Gulf
and South Atlantic Fisheries Development Foun-
dation. Data in the recreational fisheries for At-
lantic sharks are collected through NMFS' Large
Pelagic Survey and Marine Recreational Fisheries
Statistical Survey, as well as by tournament sampling.
Regional Recommendations
With several economically and recreationally
important stocks of Atlantic HMS overfished,
bycatch issues are particularly contentious in
these fisheries. In many cases, these fisheries op-
erate as multispecies fisheries with overlap in gear
use, participants and target species. Bycatch rec-
ommendations focus on reducing discard mor-
tality for overfished species, such as Atlantic blue-
fin tuna, blue and white marlin, swordfish, and
large coastal sharks. Stock rebuilding and ongo-
ing allocation disputes also demand improve-
ments in bycatch mortality estimates and mini-
mization of bycatch mortality. Following are spe-
cific recommendations for Atlantic HMS:
Improve data on the character and magni-
tude of bycatch to allow quantitative esti-
mates of discards in the fisheries for use in
stock assessments and making management
decisions.
Improve gear-handling techniques to reduce
discard mortality.
Conduct research on gear-deployment
methods that will reduce interactions be-
tween and mortality of protected species that
encounter fishing gear.
Work cooperatively with the fishing indus-
try to transfer new knowledge and tech-
niques between fishermen and researchers.
Reduce bycatch and bycatch mortality of
undersized swordfish and tunas.
Improve knowledge of (1) basic biology and
stock status of shark species in the North-
west Atlantic and (2) of the effects of bycatch
mortality on shark populations.
Increase research on the role of apex preda-
tors in structuring marine ecosystems, and
assess the effects of bycatch of these stocks.
Reduce mortality and bycatch mortality of
billfish captured in the directed fisheries for
Atlantic HMS.
Determine the status of sailfish populations.
Conduct research on post-release mortality
of recreationally caught billfish, tunas, and sharks.
Improve data collection and monitoring of
the recreational tuna, shark, and billfish fish-
eries.
t Fisheries
Regional Characteristics
Southeast fisheries (North Carolina to
Texas) generate about $900 million in ex-vessel
revenue per year (NMFS 1997). Fisheries of the
Southeast reflect the very diverse fauna of the
region, with many small fisheries working over
200 stocks.
Two fisheries dominate economically. The
menhaden purse seine fishery is the volume
leader in the Southeast, with annual landings
approaching 2 billion pounds. About 60% come
from the Gulf of Mexico and 40% from the At-
lantic. The shrimp trawl fishery generates the
largest revenue regionally, and sometimes nation-
ally.The Gulf of Mexico shrimp fishery accounts
for about 70% of the entire U.S. wild shrimp
production. About half the commercial value of
fisheries other than shrimp and menhaden con-
sists of shellfish fisheries (blue crabs, oysters, and
other invertebrates), generally harvested from
state waters, and managed by the states. The re-
mainder of the commercial harvest consists of
finfish from many stocks, including reef fish (red
snapper, red grouper, etc.); coastal pelagic (e.g.,
king and Spanish mackerel); and oceanic pelagics
(sharks, swordfish, and tunas).
Marine recreational fishing is a very im-
portant part of the Southeast harvest. Typically,
4-6 million participants make 30—40 million trips
annually. The bulk of recreational harvest con-
sists of small fish of the drum family (croakers
and seatrouts) and catfish, but many of the prized
commercial species are also prized recreationally
(e.g., red snapper and other reef species, and king
and Spanish mackerel). This shared usage makes
every conservation issue an allocation issue as
well.
In many cases, management targets have
been set toward retaining the historical shares of
catch between commercial and recreational com-
ponents. For example, the allocation ratio for
the recreational and commercial fisheries for red
snapper are set at about 50:50, and at about 70:30
for king mackerel. The recreational sector as a
whole appears to respond very quickly to
changes in abundance of individual species — if
abundance of a species increases from year to
year, catch patterns suggest that recreational fish-
ing effort may be quickly shifted to it, while
total effort may remain roughly constant. This
has led to some management paradoxes, in that
to maintain yield targets, reductions in bag lim-
its have sometimes been needed to respond to
improvements in abundance.
Southeast
Fisheries
Regional Bycatch Issues
The commercial shrimp trawl fishery con-
sistently generates the highest ex-vessel value of
any fishery in the United States, totaling $468
million in 1996 (NMFS 1997). In the Southeast
United States, the shrimp trawl fishery, made up
of thousands of small, independent firms, catches
and discards all manner of living marine organ-
isms, the vast bulk of which are of little interest
commercially or recreationally. Inconspicuous
within this bycatch are juveniles of much less
abundant, but highly prized, species that are killed
at a rate that has a substantial impact on their
populations. More conspicuous, but less frequent,
are captures of endangered marine turtles. The
shrimp industry is large and diverse (about
20,000 vessels). The major challenge may be to
make "stakeholders" out of the thousands of
shrimpers who individually have a very minor
impact, but collectively have a very major im-
pact.
Capture and drowning in shrimp nets was
identified as the single largest source of mortal-
ity for sea turtles, especially the highly endan-
gered Kemp's ridley turtle (NRC 1990). Mor-
tality can be reduced considerably with the use
of turtle excluder devices (TEDs), which have
been available for many years. However, the road
to full implementation of these devices by the
fishery has been long and contentious. Shrimp-
ers claim the devices cause loss of shrimp from
the nets, but data collected by observers aboard
commercial shrimp vessels do not support that
claim. In the Gulf of Mexico, where turtle catch
rates are low, the average shrimper withoutTEDs
might encounter a turtle every three or four
months. The quantity of shrimp effort is so high,
however, and the turtle populations so depressed,
that the fleet's impact on the turtle population
was considerable. Interestingly, along the Atlan-
tic Coast, turtle catch rates were much higher,
and perhaps as a consequence, resistance to the
use ofTEDs was much less hostile.
Finfish bycatch by the shrimp industry has
been cited as a potential problem in the scien-
tific literature since the 1930s. The weight of
finfish caught and discarded by the shrimp fish-
ery exceeds the weight of the shrimp harvest, in
some areas by severalfold. Much of the bycatch
consists of juveniles and small adults of several
hundred species. The bulk of the bycatch con-
sists of species, such as croaker, spot, and longspine
porgy, that are of limited commercial or recre-
ational interest in most areas. Within the mass of
fish taken, however, are juveniles of prized spe-
cies, such as red snapper, king and Spanish mack-
erel, and weakfish. Although not conspicuous in
the bycatch because of their much lower abun-
dance, the shrimping effort is high enough that
the impact of the bycatch removals on the popu-
lations of these highly valued species can be con-
siderable. In the Gulf of Mexico, most attention
has been focused on red snapper, which, due to
a temporal and spatial distribution similar to that
of the target shrimp, may be one of the most
highly affected species. Along the Atlantic Coast,
bycatch of weakfish and mackerels has also been
a major issue.
Steps toward managing and reducing fin-
fish bycatch have centered on development of
bycatch-reduction devices, although area or sea-
sonal closures may also be useful for bycatch re-
duction for some species. Several candidate de-
vices show strong promise in reducing finfish
bycatch without compromising shrimping effi-
ciency. Finfish species were found to differ con-
siderably in their behavior in trawls, affecting the
efficacy of bycatch-reduction efforts much more
than expected. Red snapper proved to be one of
the most difficult bycatch species to exclude; this
species is structure-oriented, and a shrimp trawl
makes a very attractive structure to the juvenile
red snapper.
tonal
xpecturts
As with the TED issue, many in the indus-
try remain skeptical of the need for finfish
bycatch reduction and distrust devices offered
as solutions. As with turtles, the low catch rates
of the prized species hidden within the bulk of
the bycatch means an individual shrimper may
feel little stake in contributing to bycatch re-
duction. For example, the average catch rate of
red snapper in the Gulf of Mexico shrimp fish-
ery is about six fish per hour. However, it is the
4—5 million hours of effort per year by the fleet
that significantly affects the snapper population.
Ultimate management authority for imple-
mentation of bycatch-reduction devices is spread
among the Gulf of Mexico Fishery Management
Council, the South Atlantic Fishery Management
Council, the Atlantic States Marine Fisheries
Commission, and the individual states. Most of
these entities are currently considering or in the
process of implementing bycatch-reduction
regulations.
Other southeastern bycatch issues center
on the general lack of knowledge needed for
quantifying bycatch in particular fisheries.While
the bycatch of the offshore shrimp fishery has
been extensively studied, the quantities taken by
the inshore shrimp fishery are essentially un-
known. There have been a few attempts to char-
acterize the bycatch of the menhaden purse seine
fishery, but the high variability of bycatch among
sets has made analysis problematic. Menhaden
catch is fairly clean (a few percent is bycatch),
but even a few percent of a billion pounds a
year might have a considerable impact on some
populations within the bycatch. Bycatch in
longline, bandit reel, and pot fisheries has been
characterized in several studies, but there are no
long-term programs for estimating bycatch, and
recent observer effort has been reduced. There
have been quite a few bycatch studies on men-
haden over the last 100 years, although each study
has tended to be limited in coverage (temporal
spatial, at dock versus at sea). Regulatory bycatch
is an issue in some fisheries, such as capture of
red snapper out of season in general reef fish
fisheries in the region. Regulatory discard of
undersized fish is a contentious issue in almost
every fishery with minimum-size regulations in
the region.
Large numbers of finfish are released alive
by recreational anglers in the Southeast. In 1996,
recreational anglers released over one half of the
total estimated recreational catch of 170 million
fish. The proportion of the catch released alive
varies considerably by species, ranging from over
90% being released for some species, such as sea
robins and dogfish, to less than 20% for highly
prized species, such as king mackerel and dol-
phins. Releases of many species (e.g., red snap-
per, groupers, and red drum) are governed by
size limits and bag limits in existing manage-
ment plans. Typically, over 50% of the recreational
catch of these species are released alive. Even
though anglers report that fish are being released
alive, there is still a question of how many of the
released fish actually survive. Short-term studies
indicate that upwards of 70% of some species
may survive, however, survival may be affected
by environmental conditions prevailing at the
time of release and care in handling.
Regional Bycatch Programs
Partnerships with other fishery manage-
ment agencies (e.g., state fishery management
agencies, interstate marine fisheries commissions,
state Sea Grant College programs, and the Gulf
and South Atlantic Fishery Development Foun-
dation) have been crucial to addressing bycatch
issues in the Southeast Region. Efforts in this
region pre-date many of the regional and na-
tional workshops held in other areas of the coun-
try. The Southeast formally began to address
bycatch in the shrimp trawl fishery in 1 990 and
Southeast
Fisheries
developed a strategic research document focus-
ing on this important issue (Hoar et al. 1992).
This strategic document led to implementation
of a formal Regional Research Program, coor-
dinated by the Gulf and South Atlantic Fishery
Development Foundation. The major compo-
nents of the program were observer programs
to quantify bycatch mortality, and gear technol-
ogy research and development to reduce finfish
bycatch.
The Regional Research Program actually
established several separate observer programs to
counter industry mistrust of data collected solely
by the government. NMFS, the Gulf and South
Atlantic Fishery Development Foundation, and
the Texas Shrimp Association all deployed ob-
servers. The separate programs were highly co-
ordinated: a common protocol was developed,
all observers received the same training, a col-
lected database from all programs was developed
and is managed by the NMFS Galveston Labo-
ratory, and estimates of bycatch of the various
species are supplied to area stock assessment sci-
entists for inclusion in total removals.
A four-phase development program for
bycatch-reduction devices for shrimp trawls is
currently under way under the Regional Re-
search Program. Throughout the development
process, each of the following phases is coordi-
nated by a gear review panel composed of gear
technical specialists from both NMFS and the
shrimp industry.
Phase 1: Initial design and prototype develop-
ment — In this phase, the full technical range
of trawl design and modification approaches
is identified. Emphasis initially was placed
on existing gear. Industry techniques, ideas
solicited from fishermen, net shop designs,
and research studies conducted by various
groups are evaluated. Fish behavior, gear
interaction, and gear performance studies
are conducted on each design using scuba,
acoustic instrumentation, remote video
cameras, and other techniques made nec-
essary due to local water conditions. This
work evaluates fish behavior and feasibility
of concept.
Phase 2: Proof of concept — The objectives of
this phase are to evaluate prototype devices
on key species, determine finfish reduction
rates, and establish shrimp catch rates. Proof
of concept testing also evaluates the ad-
equacy of the design for safety and for prob-
lems with operational use.
Phase 3: Operational evaluation — The main
objective in this phase is to test the new
gear against a standard gear under condi-
tions encountered during commercial op-
erations. Observers are placed aboard co-
operating commercial vessels to collect the
data.
Phase 4: Industry evaluation — The commer-
cial shrimp industry is responsible for fleet
testing of candidate designs for bycatch-re-
duction devices. Vessels are used to test de-
vices on commercial shrimp grounds and
to maintain logbooks on results. Observers
are placed on a subset of vessels whose cap-
tains agree to keep logbooks to collect
bycatch data by species.
Establishing and maintaining the distinc-
tion among these four phases have proven sur-
prisingly useful, both to the orderly progression
of candidate gear through the development pro-
gram, and to communicating the nature of dif-
ferent types of data and research. Within this
framework, actual research and development of
candidate devices have been carried out inde-
pendently by NMFS, Sea Grant, state agencies.
universities, and industry, drawing on a variety
of funding sources, primarily the Saltonstall—
tonal
ij&ctitres
Kennedy (S-K) and MARFIN (Marine Fisher-
ies Initiative) grants programs.
Research on the economics and sociology
of management of shrimp fishery bycatch was
initiated by the NMFS Southeast Regional Of-
fice in the late 1980s and continues to the present.
Universities that successfully competed for fund-
ing under the MARFIN and S-K grants pro-
grams have conducted additional research. Eco-
nomic analysis of bycatch issues in other fisher-
ies has been sparse, and the only fishery explic-
itly considered to date is the red snapper fishery.
Bycatch characterization and reduction re-
search has been conducted for other fisheries in
the Southeast, but not through a formal pro-
gram structure as for shrimp. Longline fisheries
for tuna, swordfish, and sharks have a history of
observer programs for general characterization
of the fisheries, including bycatch. However,
none of these programs has been sustained over
consecutive periods or conducted throughout
the range of the fishery during a single year, even
within U.S. waters.
MARFIN and S— K grants have also funded
characterization research on bycatch in the men-
haden purse-seine fisheries of the Gulf and At-
lantic Coasts. The menhaden industry has already
developed some gear innovations to release
bycatch alive during harvest. Estimates of fish
caught, but not retained, in recreational fisheries
are made through the national Marine Recre-
ational Fisheries Statistics Survey (MRFSS) pro-
gram for much of the Southeast Region. There
have been S— K awards for short-duration projects
assessing recreational bycatch in some geographic
areas not covered by MRFSS. A number of
MARFIN and S— K grants have been awarded
to examine mortality of hooked-and-released
fish; species addressed include red snapper, red
grouper, king and Spanish mackerel, and sharks.
Short-duration observer programs have been
conducted in some areas in the Gulf of Mexico
to examine bycatch of the commercial hook-
and-line fishery for reef fish. There have been
S— K research grants directed at bycatch of stur-
geon in coastal shad fisheries. Short-term research
has been conducted on bycatch in trap fisheries
for fmfish and crustaceans, with most projects
focused on developing escape structures for un-
wanted or prohibited catch, and for reduction
of ghost fishing by lost traps.
Evaluations of impacts of bycatch on the
fish stocks, and thus on directed fisheries, are
made through traditional stock assessments
whenever estimates of bycatch are available.
Evaluations of the effects of bycatch in the shrimp
fisheries are most advanced. Incorporation of
bycatch information from other fisheries in stock
assessments is often less adequate due to lack of
time-series estimates for bycatch.
Bycatch management in the Southeast
shrimp fisheries is progressing rapidly.TEDs have
been required in all but hand-operated shrimp
trawls for several years. The state of North Caro-
lina took the lead in establishing bycatch reduc-
tion device (BRD) requirements in state waters
in 1992. Both the South Atlantic and Gulf of
Mexico Fishery Management Councils are ac-
tively implementing BRD-based management
of bycatch in shrimp fisheries. The South Atlan-
tic Council began requiring BRDs in shrimp
trawls in April 1997. Amendment 9 to the Gulf
Council's shrimp plan, requiring BRDs in
shrimp trawls, was approved by NMFS in July
1997. Bycatch reduction for weakfish caught in
shrimp trawls is required under the management
plan coordinated by the Atlantic States Marine Fish-
eries Commission, and is implemented by the states.
Regional Recommendations
Several adjustments must be made to the
Regional Research Program upon implemen-
tation of BRD-based management of bycatch
Southeast
Fisheries
in Southeastern shrimp fisheries. Continued
monitoring of bycatch in shrimp trawls will be
necessary to establish mortality rates with re-
duction gear in place. BRD monitoring should
explicitly address shrimp-loss rates because this
factor determines whether a particular design is
"practicable." New BRDs are certain to be pro-
posed and developed. Provisions must be made
for all four phases of device testing under what-
ever regulations are finally adopted; procedures
must be finalized to certify new BRDs as meet-
ing the requirements for reduction when ap-
propriate. (The South Atlantic Council has al-
ready adopted a certification protocol.)
Over the longer term, impacts on the stocks
are not known for many species most promi-
nent in the bycatch. Full stock assessments will
have to be developed for several of these species
due to their primary importance in the coastal
ecosystem, and their secondary — but not
trivial — importance in the fisheries of the re-
gion. Possible multispecies impacts of bycatch
and its manipulation is already a contentious is-
sue in the region. Research and modeling have
begun, and probably must be expanded.
Priorities for other fisheries are dominated
by the need to estimate bycatch either initially
and/or on a continuing basis. For those fisheries
where bycatch exerts a significant impact on
other stocks, estimating bycatch must be con-
sidered in the same light as estimating commer-
cial and recreational harvest — a responsibility
continuing into perpetuity. Priority fisheries in-
clude the inshore shrimp fishery, longline fish-
eries, menhaden fisheries, and reef fish fisheries.
Partly because of the success in developing
TEDs and BRDs, reduction through gear tech-
nology will probably be viewed as the primary
candidate for a management tool in nontrawl
fisheries in the Southeast, although research into
modifying fishing strategies and into season/area
management options should prove productive.
Ecosystem models to determine the im-
pacts of bycatch reduction in the Gulf of Mexico
are currently under development. This is impor-
tant to establish the cost/benefit analyses for the
various bycatch-reduction options. Following are
specific recommendations for Southeast fisheries:
• Establish estimation of bycatch removals as an
integral part of collecting basic fishery statistics.
• Develop stable, long-term funding for a long-
term fishery observer capability.
• Develop strategies to distribute observer ca-
pability among the various fisheries requir-
ing coverage in such a manner as to com-
plete basic quantification of bycatch for all
critical fisheries, and to provide continuing
coverage in those fisheries deemed to exert
significant impact on populations of species
taken in the bycatch.
• Provide stable funding for research and de-
velopment capabilities in gear technology.
This will enhance NMFS' ability to work co-
operatively with experts in gear technology
spread among other agencies, universities, and
industry, providing rapid innovation and de-
velopment of bycatch management tools.
• Improve and develop multispecies model-
ing capabilities that focus on bycatch man-
agement issues and impacts. Bycatch estima-
tion is of little value without a context to
evaluate impacts. Stock assessment provides
that context at the population level, and
multispecies modeling provides it at higher
levels of organization.
• Improve and develop economic and social re-
search and monitoring programs that provide
the context for bycatch impact evaluation.
• Initiate a program to collect detailed shrimp
fishing effort data to aid in estimating mor-
tality of bycatch species.
Pelagic Si Insular fisheries
Regional Characteristics
Pacific pelagic and insular fisheries (Hawaii,
American Samoa, Guam, Northern Mariana Is-
lands, and other U.S. islands in the Pacific) are
biologically healthy, economically valuable, and
important for cultural and subsistence users. Fish-
ery management regimes in the area appear to
be effective. Three Western Pacific ports (Pago
Pago, American Samoa; Agana, Guam; and Ho-
nolulu, Hawaii) rank among the 10 U.S. fishing
ports with the highest value of landings. The ex-
vessel value of marine fish landings in these ports
in 1994 totaled $341 million (34% of the total ex-
vessel value of the landings in the top 10 U.S. ports) .
While research and proactive management ap-
pear to be effective, care must be taken to maintain
the productivity of these fisheries. Many basic popu-
lation and ecosystem aspects of the fisheries are poorly
understood. This is particularly true of the na-
ture and impact of bycatch, which are increas-
ingly controversial issues that could eventually
limit the continuation of fisheries, such as the
large and economically important longline fish-
ery for highly migratory species. Bycatch issues
involving threatened and endangered sea turtles,
monk seals, dolphins and other marine mam-
mals, seabirds, and other living marine resources
must be accorded a high level of attention.
Regional Bycatch Issues
Bycatch issues for Pacific pelagic and insular fish-
eries focus on population concerns, particularly for
seabirds and protected turdes and marine mammals.
Western Pacific Longline Fishery for
Highly Migratory Species
The Western Pacific pelagic fisheries landed
approximately 14,100 metric tons of pelagic spe-
cies in 1995, valued at approximately $53 mil-
lion. The number of vessels in the Hawaii
longline fishery increased from around 40 in
1983-87 to 141 m 1991. The Western Pacific
Fishery Management Council established a
moratorium on new entry to the fishery be-
tween 1991 and 1994, and established a limited
entry system (166 Hawaii longline permits) for
this fishery in 1994. In 1995, 110 vessels were
active in the fishery (the fewest since 1988), al-
though the number of trips and the number of
hooks set increased by 1 1%.
Early concerns about protected resources
focused on the endangered Hawaiian monk seal.
Primary bycatch species of current concern are
turtles, seabirds, and sharks. Monk seals, turtles,
and seabirds are each the subject of legislative
prohibitions (Endangered Species Act, Migra-
tory Bird Treaty Act) and have generated con-
siderable controversy and management attention.
Pacific Pelaaicfk.
Insular Fisheries
In 1991, the Western Pacific Fishery Man-
agement Council established an exclusion area of
50 nautical miles (nm) around the northwestern
Hawaiian islands to protect endangered monk seals.
It also closed an area within 50-75 nm of the
main Hawaiian islands and within 50 nm of Guam
to prevent gear conflicts between longliners and
smaller fishing boats targeting pelagic stocks. In
1994, the council implemented a mandatory ves-
sel monitoring system for the Hawaii longline fish-
ery to track the position of longliners within the
U.S. exclusive economic zone (EEZ) to ensure
that the vessels complied with the exclusion areas.
An important consideration in the central
and western Pacific Ocean is that swordfish and
tuna stocks migrate through the U.S. EEZ and
international waters.The US. component of the
longline fishery for these stocks is less than sev-
eral percent of the total effort in the area. There
are disjunct comprehensive international mecha-
nisms for gathering and reporting statistics, and
separate international management authorities
to manage these species in the Pacific.
The NMFS Honolulu Laboratory staff
compiles data for the domestic longline fishery
from the mandatory federal logbook program,
which began in 1990, and from a market moni-
toring program, which began in 1 984.
Sea Turtle' By catch
There is substantial and growing concern
about the status of populations of all species of
sea turtles. Sea turtles are designated worldwide
as threatened and endangered species. Popula-
tion declines are especially prominent in the
Pacific Islands because of nesting habitat loss and
excessive, intensive harvesting for commercial,
cultural and subsistence purposes. The principal
species of concern in the Pacific are green,
hawksbill, olive ridley, leatherback, and logger-
head turtles. The last two are the species of prin-
cipal concern regarding incidental take in pe-
lagic longline fisheries in the Pacific, conducted
mainly by Japan, Taiwan, Korea, and the United
States. Monitoring and assessment of sea turtle
bycatch is carried out primarily through the
Hawaii longline observer program.
In a 1994 biological opinion, NMFS con-
cluded that the Hawaii-based pelagic longline
fishery adversely affects, but does not jeopardize,
sea turtle populations. Nevertheless, limits were
set on estimated incidental take and mortalities.
The estimated 1994 and 1995 take and mortali-
ties of turtles in total and by species were within
allowable limits stipulated in the most recent
biological opinion, except for loggerheads in
1995 (Table 5). Consequently, NMFS has now
Tables.
Take and mortality of sea
turtles in the Hawaii longline fishery
r.
Species
1994
1995
Allowable Take in
Any Single Year
Take
-90% CL, .
take
Mortality
Take
-90% CL .
take
Mortality
Take
Mortality
Loggerhead
Leatherback
207
122
70-403
31
413
153-764
62
12
305
271
46
41
41-233
18
81
0-187
Olive Ridley
78
0-180
12
81
0-191
NR
12
215
23
Green
34
0-95
5
NR
NR
119
18
Hawksbill
NR
NR
575
2
1
All Species
441
238-688
67
272-970
87
849
129
Note: CL - confidence level; NR = none recorded.
jional
Perspectives
reinitiated an Endangered Species Act Section 7
consultation on the longline fishery interactions
with sea turtles, with a focus on loggerheads.
In general, lack of information about sea
turtle survival, age at maturity, and other bio-
logical parameters, coupled with a dearth of data
on human harvests and incidental takes and the
great expense of an adequate observer program,
makes assessment of the impact of this interac-
tion particularly difficult. There is a need for in-
ternational cooperation in collecting and ana-
lyzing data on the effects of the Pacific longline
fishery on sea turtle populations.
Skavks
The Hawaii longline fishery targets prima-
rily swordfish and tunas, but has a substantial
bycatch of pelagic sharks. As noted in Table 6,
shark bycatch doubled between 1991 and 1993
and then declined by one-third in 1995. This
probably reflects changes in the operations of
the fisheries during that time, rather than varia-
tions in shark stocks. Blue sharks make up more
than 90% of the shark bycatch; all blue sharks
that are kept are believed to be finned.
There is no U.S. -directed shark fishery in
the central and western Pacific. Most sharks that
are taken incidentally in the Hawaii domestic
longline fishery are not marketable because of
their species or size, and are discarded. Of the
68.8% of sharks that are released, observer re-
ports indicate that 80% of them are alive, al-
though the long-term mortality is not known.
The fins of some sharks, especially blue sharks,
are taken and dried on board for future sale, and
the carcasses discarded.
The estimated total round weight of sharks
that are kept for processing is approximately
1 ,590 metric tons (mt).The estimated weight of
dried shark fins is 22.2 mt. The value of pro-
cessed shark products to the Hawaii longline fish-
ery in 1995 was $830,000.
Some data on shark catch and disposition
are obtained by the domestic longline observers
and from logbook data. Relatively little infor-
mation is available on the biological status of
pelagic shark species, and the volume and im-
pact of shark bycatch and discards. There is a
need for better collection of shark bycatch data
in both domestic longline fisheries and those that
occur throughout the western Pacific.
Seaburds
Controversy over the bycatch of several
species of albatross in the Western Pacific longline
fishery is growing both locally and internation-
ally. The impact of seabird mortality in the
longline fisheries is unknown, but is probably
quite large. Albatross ingest bait and hooks, or
become entangled in longline gear during gear
TabU 6.
Sharks catch in the Hawaii domestic longline fishery.
Year
No. of Sharks Caught CPUE
(No./1, 000 Hooks)
% Blue Sharks % Kept
(All Sharks)
1991
71,183 5.77
92.0 3.2
1992 94,897 8.11
94.1 3.8
1993
154,608 11.87
97.2 10.8
1994
114,656 9.56
96.1 14.4
1995
101,773 7.52
93.7 31.2
set and retrieval.Very few seabirds survive hook-
ing or entanglement.
An estimated 54,000 breeding pairs of
black-footed albatross and 616,000 pairs of
Laysan albatross exist in the world. More than
99% of both species nest in the northwestern
Hawaiian Islands. Although inadequate scientific
knowledge of seabirds makes it very difficult or
impossible to assess the true impact of fishing or
other causes of mortality, there is little question
that the populations of these two species of al-
batross are in serious decline.
Preliminary analysis of longline observer
data indicates that in 1,286 Laysan albatross and
2,135 black-footed albatross were taken in 1994
in the Pacific longline fishery. In 1995, the
longline fishery took 1,942 Laysan albatross and
1,796 black-footed albatross. In the opinion of
the U.S. Fish and Wildlife Service, the black-
footed albatross population cannot sustain this
level of take.
The National Marine Fisheries Service will
continue to work with the U.S. Fish and Wild-
life Service to estimate mortality and develop
mitigation measures. Recently, the Western Pa-
cific Fishery Management Council reprinted (in
English and Vietnamese) the booklet Catchitig
Fish Not Birds — A Gtride to Improving Your Long
Line Fishing Efficiency (Nigel Brothers, Parks and
Wildlife Service,Tasmania, Australia), which it is
distributing widely And, in cooperation with the
U.S. Fish and Wildlife Service, the Council re-
cently held the first in a series of workshops with
leading participants in the longline fishery whose
basic theme was that "every hook that catches a
bird will not catch a fish."
Potential methods of albatross bycatch re-
duction include (1) putting weights near the
hooks to sink the bait faster; (2) using faster-
sinking thawed, rather than frozen, bait; (3) set-
ting longlines at night when birds are not as ac-
tive;and (4) flying streamers and other devices to scare
the birds away while the longlines are being set.
Monk, Seals
The Hawaiian monk seal is the only en-
dangered marine mammal found entirely within
U.S. waters. Its abundance has declined by 60%
since the late 1950s, and the current population
is about 1,300—1,400 animals. Beach counts over-
all have declined by 5% a year. To some degree,
growth of smaller populations in some areas is
offsetting losses in other areas.
The Western Pacific Fishery Management
Council addressed initial concerns about inter-
actions with monk seals by imposing a strict pro-
hibition on longlining within a 50-mile area
surrounding the northwest Hawaiian Islands. No
direct bycatch of monk seals in the fishery is
currently known, but extreme care must be taken
because of the endangered status of the species.
Fishermen are concerned that the closure zones
are unnecessarily large and exclude them from
valuable fishing opportunities. Because this con-
cern is expected to grow, research into monk
seal ecology must continue.
A specific fisheries-related peril of particular
concern is the entanglement of monk seals in
marine fisheries debris. Each year, monk seals
are found entangled in fishing nets on the beach
or in nets snagged on coral reefs. The source of
these nets is not known. The observed minimum
rate of entanglement in beach debris alone is
about 1 % of the entire monk seal population
per year. Lethal entanglement in unobserved
shallow reef areas is probably greater.
Martins
There is growing pressure to ban the sale
of blue marlin that is landed incidentally in the
swordfish and tuna longline fishery. Blue marlin
is a principal target species for the recreational
and charter fisheries, especially in Hawaii. The
Western Pacific Council has been asked to con-
sider expanding a longline area closure to re-
duce the catch of blue marlin near a major sport
fishing center. For commercial vessels, the landed
value of this species is very important economi-
cally, with some fishermen claiming that rev-
enues from incidentally caught marlin often
make or break a fishing trip. Also, there is con-
cern that blue marlin may be overfished, and
that bycatch of blue marlin may have an increas-
ing impact on the conservation of this species.
The economically important and influen-
tial recreational fishery asserts that the commer-
cial incidental harvest of marlins diminishes the
economic and social returns to the recreational
sector, compared with a purely recreational, in-
creasingly catch-and-release fishery.
Western Pacific Crustacean Fishery
(Northwestern Hawaiian Islands)
Bycatch of protected species, such as monk
seals, has been addressed effectively in this fish-
ery through (1) designing the size of trap open-
ings to prevent entrapping monk seals; (2) clos-
ing the waters around Laysan Island less than 10
fathoms deep and within atolls; and (3) imple-
menting framework regulatory measures to al-
low rapid action (including closure of the fish-
ery if necessary) to respond to actual or suspected
mortality of seals in the fishery.
Until the 1996 fishing season, the harvest
quota system in the commercial lobster fishery
included minimum size limits and a prohibition
on retention of egg-bearing lobsters. A signifi-
cant concern under this system was the belief
that the mortality of the bycatch of small and
egg-bearing lobsters that resulted from on-deck
injury and exposure, and predation upon return
to the ocean, are extremely high — perhaps
greater than 75%.
The immediate bycatch problem has now
been addressed to some degree by a recently
implemented "retain-all" system in the fishery,
under which fishermen can retain subadult and
berried lobsters that are then counted as part of
the quota, rather than being lost uncounted be-
cause of on-deck or post-release mortality. How-
ever, the retain-all approach is relatively untested
in lobster fisheries and must be monitored closely
to confirm its efficacy and impact.
Subadult and egg-bearing lobsters may
make up 50% or more of the total catch in some
areas. If fishermen still "high-grade" to an ap-
preciable extent (retain only the most valuable
portions of the catch, which are probably the
larger lobsters), many subadult and berried lob-
sters could suffer discard mortality and be lost
to the population.
Current regulations require that all traps
deployed in the northwestern Hawaiian Islands
include two escape panels, each with four escape vents
to help subadult lobsters escape. However, a large pro-
portion of the catch is composed of subadults.
Eastern Tropical Pacific Tuna
Purse Seine Fishery
The Eastern Tropical Pacific tuna purse-
seine fishery (primarily for yellowfin and skip-
jack tuna) has been controversial because of the
bycatch and subsequent mortality of large num-
bers of dolphins that were caught when the purse
seiners targeted and encircled mixed schools of
tuna and dolphin. The bycatch of small tunas,
turtles, sharks, billfish, and other species is now
receiving attention because of the relatively re-
cent change to fishing around schools of tuna
and or fishing near logs and other floating de-
bris, rather than setting on dolphin.
Observers are required on all vessels to
record and report bycatch. The bycatch of dol-
phins is managed largely as an international is-
sue through the Inter-American Tropical Tuna
Commission. Because dolphin bycatch was sub-
stantial in the 1970s, purse seine gear was im-
proved and procedures were developed to safely
release dolphins. Subsequently, U.S. legislation
provided that a tuna product could not be des-
ignated "dolphin safe" if dolphins were involved
in the catch. This resulted in a dramatic 97%
reduction of mortality in the overall fishery to
low levels in recent years — from nearly 130,000
dolphins in the late 1980s to about 4,000 dol-
phins in 1994. The U.S. fishery (about six ves-
sels) reduced its take to fewer than 500 dolphins
in 1992. Despite these reductions, dolphin
bycatch has remained a volatile issue.
An emerging bycatch issue in this
"nondolphin" tuna purse seine fishery is that sets
made on tuna under logs or other aggregating
debris catch the community associated with such
debris— such as small, immature tunas, mahi
mahi, wahoo, billfish, sharks, rays, and other im-
portant living marine resources. Turtles are not
supposed to be retained, and markets are not
available for small tunas, sharks, and most other
of the bycatch species. Thus, fishermen usually
discard this bycatch dead. Adequate information
is not available on the size and species composi-
tion of bycatch in this fishery, or on the biologi-
cal impacts of bycatch in this fishery, but there is
growing concern. Proposals that could moder-
ate the impact of log fishing were hotly debated
in the U.S. Congress in 1996, but failed to pass
into U.S. law. As part of the 1988 Marine Mam-
mal Protection Act amendments, the National
Research Council performed a review of alter-
native methods of harvesting tuna without en-
circling dolphins, and made a number of rec-
ommendations that are being explored.
Western and Central Pacific
Tuna Fisheries
The Central- Western Pacific tuna purse
seine fishery has a bycatch that is largely un-
known, but is probably similar to that in the
Eastern Pacific, except that setting on dolphin is
not a practice in the Western Pacific. Lack of a
market or low price inhibits the retention and
utilization of bycatch in this fishery.
The Oceanic Fisheries Program of the
South Pacific Commission concluded that". . .
not enough information was available to accu-
rately determine the levels of by-catch in the
western and central Pacific tuna fisheries." How-
ever, the review noted that "... although defi-
nite estimates were not possible, observer data
suggest that by-catch may constitute between
0.35% and 0.77% of the total catch (by weight)
for unassociated sets, and between 3.0% and 7.3%
for log sets. Purse-seine sets on floating objects
(compared to unassociated sets) produce the larg-
est amounts, highest incidence, and greatest va-
riety offish and other species" (Bailey 1996).
Data collected under the South Pacific Com-
mission program are retained by the commis-
sion and are available to the United States only
on an aggregated basis.
The review reported further there is no
evidence that dolphins are deliberately set on or
incidentally caught in the Western Pacific; ma-
rine turtles are occasionally caught, but the ma-
jority are released alive; significant numbers of
sharks are taken, but overall estimates of exploi-
tation are impossible due to non- and under-
reporting on logsheets (a problem common for
most nontarget species); accidental marine mam-
mal capture is rare; and seabird capture is well
documented, with management measures pro-
posed. Also, mortality of small bigeye tuna as
bycatch in the purse-seine fishery may adversely
affect bigeye tuna stocks.
The report also noted that observer data
collection is the most reliable means for collect-
ing information in the fisheries. However, only
two compliance-related and scientific-data-re-
lated observer programs were operational in the
Western Tropical Pacific in the 1980s — one was
operated by the Micronesian Maritime Author-
ity; the other, the U.S. Multilateral Treaty Ob-
Regional
Perspectives
server Program, was established and supervised
by the Forum Fisheries Agency. There has re-
cently been an increase in several other observer
activities in the region.
The low level of coverage of fishing activi-
ties makes it difficult to estimate levels of bycatch.
In 1995, 4.3% of purse-seine trips and 0.3% of
longline trips had observer coverage. The report
concludes further that ". . .it remains evident
that the current levels of monitoring fall well
short of providing the information required for
effective conservation and monitoring of the
species in question" (Bailey 1996).
California/Oregon Drift Gill-Net Fishery for
Swordfish and Sharks
The West Coast fishery for offshore pelagic
species (primarily for swordfish and sharks off
California) is conducted with drift gill nets,
longlines, and harpoons. The drift gill-net fish-
ery has a bycatch of marine mammals, sharks,
sea turtles, and billfish. Because of the marine
mammal interactions, this fishery is a Category
I fishery under the Marine Mammal Protection
Act.A number of other species with market value,
such as tunas and sharks, are taken incidentally
to the directed swordfish catch. Small numbers
of blue sharks, pelagic rays, and inedible fish are
also taken.
Marine mammal stocks of particular con-
cern (strategic stocks) include the short-finned
pilot whale, Baird's beaked whale, mesoplodont
beaked whales, Cuvier's beaked whales, pygmy
sperm whale, sperm whale, and humpback whale.
The California Department of Fish and
Game regulates the fishery with laws passed by
the California legislature. California state law lim-
its the number of vessels in the drift gill-net fish-
ery to 185 permits statewide; about 90 of these
are estimated to be active on a full-time basis. In
1 993, approximately 990 metric tons, or about
82% of the total landings, were swordfish.
Fishermen are required to maintain and
submit a logbook detailing their fishing activi-
ties. Management consists primarily of area clo-
sures, seasons, limited entry, and minimum mesh
sizes. Fishermen, as a practicality, set nets several
meters below the surface to avoid higher billfish
and mammal catches.
Since 1990, the NMFS Southwest Region
has placed observers in the drift gill-net fishery
to monitor incidental taking of marine mam-
mals, collect specimens, and record other bycatch
data, such as net-related variables and location
of mammals in the net. Each year, overall ma-
rine mammal mortality is estimated from ob-
server data and estimates of total effort in the
fishery. Relatively few strategic stocks were ob-
served taken over the five-year observed period.
Of all observed sets (759 in 1994, or approxi-
mately 15% of total sets made), 1.4% contained
one or more cetaceans from a strategic stock,
and 10.5% contained one or more cetaceans from
other stocks. Continued observer data collec-
tion is very important for this effort.
The existing Mexican drift gill-net fishery
also interacts with some of the species of con-
cern, and may have high marine mammal takes
without any regulations regarding marine mam-
mal bycatch. Therefore, the take-reduction team
"... suggests that NMFS consider ways to re-
solve this issue and strongly encourages interna-
tional cooperation aimed at conserving marine
mammal populations."
Bycatch in Other Fisheries
The U.S. troll fishery for albacore, which
operates in both the North and South Pacific,
produces a very small bycatch of turtles (un-
known species) and finfish, mostly dolphinfish,
yellowtail, and skipjack tuna. This fishery also
discards an unknown amount of small (less than
59 cm) juvenile albacore for economic reasons.
The quantity of discarded albacore is not likely
IftSi
Velaaio8t
Flmerles
to exceed 10% of the total number offish caught
(2.8 million fish in 1996). The mortality of the
discarded fish is unknown, but is presumed to
be high. This information is based on a limited
amount of observer data from the North Pacific.
Few or no known bycatch problems exist
in the northwest Hawaiian Islands bottomfish
fishery or in the Hawaii precious corals fishery.
Regional Bycatch Programs
Because of various biological opinions re-
garding sea turtle bycatch, NMFS has imple-
mented several "reasonable and prudent mea-
sures" and "conservation recommendations."
When turtles are taken on longline gear, fisher-
men are required to return them to the sea
whether the turtles are alive or dead. NMFS
places observers on vessels according to a statis-
tical design to document turtle takes. Because
of uncertainties regarding the level of turtle in-
teractions in the fishery as reported through the
logbook program, monitoring and assessment of
sea turtle bycatch are carried out primarily
through the NMFS Southwest Region's Hawaii
mandatory longline observer program. The ob-
server program, which began in 1994, employs
a pilot stratified random survey design to esti-
mate the take rate (turtles per hook) in the ag-
gregate and by species. The trip-coverage rate of
the program to date is about 4%. Workshops have
been held to develop and inform fishermen of
methods to reduce the bycatch of turtles. Also,
six recovery plans for U.S. Pacific species of sea
turtles are nearing final agency approval.
In February 1 996, NMFS convened a take-
reduction team in accordance with provisions
of the Marine Mammal Protection Act to de-
velop a take reduction plan to reduce the inci-
dental taking of marine mammals in the Cali-
fornia/Oregon drift gill-net fishery for sword-
fish and sharks. The immediate goal of the plan
is to reduce, within six months of its implemen-
tation, the incidental mortality and serious in-
jury of strategic stocks to less than the "poten-
tial biological removal" levels established for those
stocks. The plan's long-term goal is to reduce
the rates to zero within five years of implemen-
tation.
The draft take-reduction plan, reached by
consensus, contains four primary strategies to
reduce take rates: (1) a multiyear test of the ef-
fectiveness of acoustical devices (pingers) to de-
termine whether to require their use; (2)
fleetwide deployment of a six-fathom minimum
buoy line extender length (nets set several meters
below the surface reduce the bycatch of marine
mammals and billfish); (3) skipper workshops to
generate and consider potential additional take-
reduction strategies; and (4) continuation of
California's current policy of not issuing new
shark and swordfish drift gill-net permits to re-
place those that lapse. NMFS and the take-re-
duction team will continue to meet every six
months to monitor the plan's implementation
until NMFS determines that the plan's objec-
tives have been met.
The take-reduction plan also recommended
that additional research be conducted, includ-
ing determining the optimal minimum extender
length, increasing the level of observer coverage,
increasing the understanding of cetacean hear-
ing ranges and why pingers work in some cases,
and considering "buying out" permit holders to
reduce potential effort.
Regional Recommendations
The issue of sea turtle bycatch in the Ha-
waii longline fishery has the focused and effec-
tive attention of conservation groups. It is very
important that NMFS devote greater resources
to (1) identify factors associated with the take o\
turtles in longline gear; (2) develop further CO-
Regional
Perspectives
operation in the international arena to assess the
status of Pacific sea turtle populations and the
impact of the Hawaii longline fishery on them;
and (3) investigate other dimensions of this
bycatch problem. Immediate approaches will
include developing and implementing an alter-
native survey design for conducting the longline
observer program, expanding the trip-coverage
rate of the observer program to 10%, determin-
ing factors associated with turtle takes, and de-
veloping and implementing mitigation measures
in the longline fishery.
Shark bycatch in the Hawaii tuna/sword-
fish longline fishery has also generated a great
deal of public concern. It is necessary to (1) in-
crease research to estimate bycatch for all shark
species, (2) evaluate logbook performance in
documenting disposition of sharks, (3) under-
take biological research in support of stock as-
sessment, and (4) pursue cooperative research of
bycatch with major foreign fishing nations.
Interactions between fishing operations and
endangered monk seals are of great concern in
Western Pacific fisheries. Because of the largely
unknown nature of this problem, further work
is required to (1) monitor and assess the six main
reproductive populations of monk seals; (2) study
monk seal ecology (particularly in the pelagic
habitat), biology, and natural history; and (3) in-
vestigate and mitigate problems impeding recov-
ery of this endangered species. Additional re-
sources should also be devoted to removing net-
ting and other marine debris that pose entrap-
ment dangers in the monk seal environment.
Following are other recommendations for man-
aging bycatch in Western Pacific fisheries:
• Increase the level, broaden the scope, and en-
sure the continuity of fishery observer pro-
grams sufficiently to allow quantitative esti-
mates of catch and other fishery data, in-
cluding discards of fishery resources and pro-
tected species, with acceptable levels of pre-
cision and accuracy.
Increase the ability to assess the effects of dis-
cards (population, ecosystem, social, and eco- .
nomic effects) and of management alternatives.
Increase research on immediate and post-
release mortalities of animals encountering
fishing gear — but not retained — in particu-
lar, sea turtles that have been hooked or en-
tangled and released.
Work closely with the Western Pacific Fish-
ery Management Council, industry, environ-
mental interests, and others to develop al-
ternative solutions to real and perceived
bycatch problems, including transfer of
knowledge and techniques to reduce the
bycatch of seabirds in the longline fishery.
Improve knowledge of basic biology and
stock status of shark species in the Pacific,
and of the effects of fishery-related mortal-
ity on shark populations.
Implement recommendations of take-reduc-
tion plans to reduce incidental taking of ma-
rine mammals in the California/Oregon drift
gill-net fishery for swordfish and sharks.
Enhance research on the impacts of fishing
on blue and striped marlin populations,
through domestic research programs and in-
ternational scientific cooperation.
Develop mitigation techniques to reduce
mortality of lobster bycatch, including re-
search on (1) gear design and operations
and (2) handling and release techniques.
Develop and evaluate modifications to exist-
ing fishing gear to allow a reduction in the
retention of the "legal bycatch" of small size
classes of lobster, and increase the subsequent
recruitment of lobsters to the fishery.
Coast Fisheries
Regional Characteristics
Fisheries of the West Coast (coastal Cali-
fornia, Washington and Oregon) primarily tar-
get several species of groundfish and salmon,
while anchovy, sardines, mackerel, shrimp, crab,
squid, and other shellfish and molluscs provide
important alternative markets. These fisheries are
harvested using a variety of gear types (trawls,
seines, pots, hook and line, etc.) that produce
about 462,000 metric tons (mt) annually, and
have an ex-vessel value of approximately $300
million. About one-third of the harvest is taken
within coastal state waters (0—3 mile zone).
In the groundfish fishery, nearly 200,000
mt of Pacific whiting are taken annually by large
mid-water trawl and catcher/processor vessels
that have replaced foreign and joint- venture fleets
of the 1970s and 1980s. Bottom trawls harvest
about 75,000 mts annually of other groundfish
species, including several species of rockfish, flat-
fish, lingcod, and Pacific cod, as well as a deep-
water complex of thornyhead rockfish, Dover
sole, and sablefish.
The five species of Pacific salmon support
important commercial, recreational, and tribal
fisheries in the states of Washington, Oregon,
California, and Idaho. Salmon are part of the
culture and heritage of the Pacific Northwest,
having been harvested for ceremonial and sub-
sistence purposes by Native Americans for mil-
lennia. Commercial, recreational, and tribal fish-
ermen harvest salmon from the Pacific Ocean,
Puget Sound, estuaries, and rivers along spawn-
M
1
1
' ** '**J A 1
ft
\
ing migration routes using trolling gear, seines,
gill nets, and hook and line. Salmon fisheries yield
about 23,000 mt annually Harvests have been
declining, however, as habitat degradation and
overfishing have threatened specific populations
of salmon. Several species of salmon have been
or are proposed for listing under the Endangered
Species Act.
Recreational angling is important to the
West Coast fisheries, with about 80% of the es-
timated 25 million fish landed annually taken
from California waters, about 15% taken from
Washington waters, and about 5% taken from
Oregon waters. Anglers reportedly spend about
$850 million each year in the West Coast fisher-
ies. A large portion of the recreational catch is
released, including releases of protected species.
Regional
Perspectives
Additional information on post-release surviv-
ability of these fish would be useful.
Management and enforcement of West
Coast fisheries rely heavily on actions of the
Pacific Fishery Management Council and on the
cooperation among the federal, state, and tribal
fishery management agencies.
Regional Bycatch Issues
Pacific Groundfish
Bycatch in the Pacific groundfish fishery
comes in many different shapes and forms and is
a significant issue. Bycatch discards occur in ev-
ery sector of the groundfish fishery. Discarded
bycatch includes (1) nongroundfish species (pro-
hibited species, such as salmon, Pacific halibut,
and crab) that are the target species in other fish-
eries, (2) targeted groundfish species that are
caught in a species complex and discarded to
stay within species or species-complex trip-land-
ing limits, (3) discards of target species resulting
from harvest guidelines or quotas being achieved
for some species and not others, and (4) unmar-
ketable groundfish and nongroundfish species.
With the exception of the mid-water trawl
fishery for Pacific whiting, bycatch is not com-
prehensively monitored or precisely estimated.
Lack of a comprehensive at-sea observer pro-
gram to collect bycatch and other biological data
is the main reason information is lacking or es-
timates are considered to be very "soft." Some
useable data have been collected through lim-
ited observer programs conducted under research
activities or under experimental fishing permits.
Bycatch information on many groundfish spe-
cies is needed to better assess and account for
total mortalities in the different fishing strate-
gies. Bycatch of salmon includes species listed
under the Endangered Species Act.
Pacific Whiting
Pacific whiting are taken by large mid- wa-
ter trawls in the spring and summer each year.
The annual whiting harvest guideline is allo-
cated among those vessels that deliver at sea to'
floating processors (moth erships), those that catch
and process at sea (catcher/processors), and those
that land whiting at shoreside processing plants
(shoreside). Pacific whiting landings generally
range between 150,000 mt and 300,000 mt an-
nually, making up the single largest component
of the Pacific groundfish fishery. The majority
of the catch is either headed and gutted or made
into surimi and is, primarily, exported.
Salmon bycatch is a sensitive issue because
the extremely depressed status of many wild
salmon stocks, some of which have been listed
as either "threatened" or "endangered" under the
Endangered Species Act, has resulted in signifi-
cant restrictions to directed commercial and rec-
reational salmon fisheries, with serious economic
impacts to coastal communities. Salmon bycatch
can be a problem at the beginning of the season
when vessels are exploring for abundances of
whiting of the right size for processing. Salmon
bycatch occurs intermittently with little consis-
tency in season or location where it occurs.
Estimating the total salmon bycatch is pos-
sible because all at-sea processing vessels have
at-sea observers on board, and vessels landing
shoreside currently are allowed to land unsorted
catches under the authority of experimental fish-
ing permits, so that the bycatch of salmon can
be counted when the catch is unloaded at
shoreside plants. Salmon bycatch by the at-sea
processing sector is discarded, whereas salmon
recovered at shoreside plants are confiscated by
the state and given to charity. All of the at-sea
processing vessels that participate in the Pacific
whiting fishery also participate in the North
Pacific groundfish fisheries off Alaska, where they
West Coast
Fisheries
are required to carry at-sea observers. No simi-
lar regulatory requirement currently exists for
at-sea processors in the Pacific whiting fishery
because all have voluntarily agreed to carry the
same NMFS-certified observers necessary to fish
off Alaska ever since the processors began par-
ticipating in the whiting fishery in 1990. The
Pacific Fishery Management Council (PFMC)
has recommended that all at-sea processor ves-
sels longer than 125 feet that participate in the
Pacific whiting fishery also be required to carry
an at-sea observer. The implementation of this
regulation has been delayed because all vessels
already carry observers voluntarily; nevertheless,
it is likely to be promulgated soon to legally
ensure that each at-sea processor continues to
carry an observer.
The current biological opinion (NMFS
1995) resulting from consultation under Section
7 of the Endangered Species Act requires that a
monitoring program be continued at a level that
maintains the current capability under the ex-
perimental fishing permit program to estimate
the salmon bycatch by vessels that deliver whit-
ing to shoreside processing plants. Each experi-
mental fishing permit requires the vessel opera-
tor to take an observer, if asked, and requires
that his entire catch be delivered, unsorted, to a
shoreside processing plant where technicians
sample the entire catch. The intermittent nature
of salmon bycatch and the fact that not all hauls
are sampled and some sampled hauls are only
partly sampled, introduce significant uncertain-
ties when extrapolating salmon bycatch for a
single vessel or area, but are more reliable for
estimating the total bycatch for the entire sea-
son and area. Salmon bycatch has averaged about
11,000 fish (98% chinook salmon) during the
last decade.
Marine mammal bycatch in the Pacific
whiting mid- water trawl fishery is also of con-
cern. Since 1990, limited mortality takes have
included individuals from six marine mammal
species — specifically, California sea lion, Steller
sea lion, harbor seal, northern elephant seal, Pa-
cific white-sided dolphin, and Dall's porpoise.
During the 1996—97 fishing season, observers
reported an annual marine mammal mortality
take of six to eight marine mammals, a level that
is not considered significant.
This observer program is providing infor-
mation not only on the actual bycatch of salmon,
but on the bycatch of other groundfish species
as well. Some species of rockfish, such as yel-
lowtail rockfish and Pacific Ocean perch, are
occasionally taken as bycatch in large numbers,
but are accounted for by the monitoring pro-
grams. The bycatch of yellowtail rockfish is an
immediate concern because the most recent
stock assessment indicates yellowtail have been
overharvested and future catch must be reduced.
Yellowtail rockfish bycatch in the Pacific whit-
ing fishery, which is either discarded or made
into fish meal, is deducted from the annual har-
vest guideline, thus reducing the amount of yel-
lowtail available for the directed fishery. In 1996,
for example, the yellowtail rockfish bycatch was
estimated at 631 metric tons, of which only 12
metric tons was retained. Pacific Ocean perch
are overfished, are subject to a rebuilding pro-
gram, and have had an annual allowable biological
catch of zero for many years. The stock shows
no signs of rebuilding.The total rockfish bycatch
during 1994—96 has averaged around 1,000 mt
annually.
The whiting industry and the PFMC have
developed a variety of bycatch avoidance mea-
sures, some voluntary and some regulatory. The
industry has adopted a PFMC-endorsed volun-
tary guideline of 0.05 salmon per metric ton of
whiting as a bycatch rate ceiling for the entire
whiting season. Time and area closures have also
been implemented to avoid areas of high chinook
salmon abundance. For example, at-sea process-
Lotud
ing and all trawling for Pacific whiting in depths
shallower than 100 fathoms are prohibited off
California to reduce the salmon bycatch. At-sea
processing vessels are also testing a pilot pro-
gram based on real-time feedback to vessels iden-
tifying bycatch "hotspots" encountered by indi-
vidual vessels within the fleet so that other ves-
sels may avoid those areas.
Bottom, Trawl Fishery
The bottom trawl fishery targets individual
rockfish, flatfish, roundfish, and different species
aggregations of rockfish, as well as the deep-wa-
ter complex consisting of thornyheads rockfish,
Dover sole, and sablefish. All types of regulatory
discards plus discretionary (economic) discards
occur in the bottom trawl fishery. Reasons for
discard include prohibited species designation
(Pacific halibut, salmon, crab) ; unmarketable size
or species; and overages of trip-landing limits,
harvest guidelines, and quotas.
Information on bycatch has been derived
from a variety of sources, primarily research stud-
ies or other short-term programs that sample
only a small portion of the bottom trawl fleet.
Fishermen are required to record bycatch in log-
books, but these have not been used to generate
bycatch estimates because of inaccuracies in
bycatch records. The Pacific Fishery Manage-
ment Council has developed the guidelines for
a comprehensive data collection program, includ-
ing at-sea observers, but it has never been imple-
mented due to lack of funding.
Monitoring the total removals by the fish-
ery is an important component of any fishery
analysis program. In the bottom trawl fishery,
total landed catch is well monitored by the state-
run fish sales ticket system, but catch discarded
at sea is still unknown for most segments of the
fishery.
Based on various "snapshots" from spe-
cific research studies, it is thought that the an-
nual salmon bycatch in the bottom trawl fish-
ery may range from 6,000 to 9,000 fish, nearly
equal to the magnitude of the bycatch in the
whiting fishery, although this cannot be cor-
roborated because of lack of sufficient at-sea
monitoring. Chinook salmon, several popula-
tions of which are listed under the Endangered
Species Act, are particularly vulnerable to bot-
tom trawls.
Pacific halibut also frequent waters where
the groundfish and the shrimp bottom trawl fish-
eries occur. The distribution of halibut is very
spotty throughout waters off Washington, Or-
egon, and northern California, which constitute
the extreme southern range of halibut. Halibut
are found primarily in localized concentrations
called halibut "hotspots." The International Pa-
cific Halibut Commission is currently develop-
ing a new stock assessment method. It is also
devising a method for determining the total al-
lowable catch by management area that includes
bycatch compensation features that deduct the
adult bycatch from the current equilibrium yield,
thereby directly reducing fixed-gear (longlines
and fish pots) harvest guidelines.
Because of the lack of an at-sea observer
program, the estimates of Pacific halibut bycatch
in the bottom trawl fishery are based on the in-
cidence of halibut observed during an experi-
mental observer program designed to investigate
the extent of discards induced by trip-landing
limits during 1985—87. Pacific halibut bycatch
rates were estimated by multiplying the observed
bycatch rate in the experimental program by the
estimated total hours of fishing obtained from
logbooks. The most recent estimate of Pacific
halibut bycatch in the bottom trawl fishery is
448 mt. In the absence of more accurate esti-
mates of halibut bycatch, overestimates of the
bycatch could directly reduce the fishing op-
portunity for the directed longline and recre-
ational fisheries for halibut, with the extreme
possibility of virtually eliminating both directed
fisheries.
The primary economic management ob-
jective for groundfish management on the West
Coast is to have seafood processors provide a
continuous, year-round flow offish to fresh fish
markets to produce a variety of benefits, includ-
ing promoting continuous employment in
coastal communities. However, overcapitalization,
increased effort, and either declining or stable
total allowable catch have resulted in the need
to significantly slow catch rates to spread the
catch of each species or species complex for
which there is a harvest guideline over the en-
tire year.
The PFMC has chosen trip-landing limits
as the vehicle to slow the catch. Because almost
all species managed by trip limits are harvested
in a multispecies mixture with other trip-limit
species, vessels are forced to discard valuable
market species once the trip limit for that spe-
cies is reached, while the vessel continues to fish
on the trip limit for other species. As trip limits
become more restrictive and as more species
come under trip-limit management, discards
increase.
Although data are not precise, estimates of
trip-limit-induced bycatch are made for some
trip-limit species and are either factored into the
in-season catch estimates so the harvest guide-
line includes total mortality or deducted from
the allowable biological catch before setting the
harvest guideline pre-season. The level of dis-
card managers currently assume as trip-limit-
induced ranges from 5% for Dover sole to 20%
for sablefish. However, if discard estimates are
too high, then the industry is foregoing some
short-term yield; if discard estimates are too low,
then the long-term health of the fish stock may
be jeopardized. The PFMC has attempted to
reduce trip-limit-induced discards by extend-
ing the trip-limit period from weekly, daily, or
single trip-landing limits to monthly. Now most
species are managed under two-month cumula-
tive trip-landing limits.
Quota-induced discards also can occur
when fishermen continue to harvest other spe-
cies when the harvest guideline of a single spe-
cies is reached and further landings of that spe-
cies are prohibited. Discretionary discards of
unmarketable species or sizes are known to oc-
cur widely, although they are largely unmeasured.
In the absence of a comprehensive at-sea
observer program, other more limited programs
have been conducted to obtain bycatch data.
Limited research studies have been funded by
NMFS and under the Saltonstall— Kennedy grant
program to investigate bycatch in different fish-
ing strategies and recommend changes in strate-
gies or gear modifications. These included the
experimental observer program conducted be-
tween 1985 and 1987 by the University ofWash-
ington, which provided some useful insights into
bycatch by fishing strategy. However, a program
to comprehensively estimate discard rates and
mortality is still necessary to provide accurate
data on total catch and mortality. Recently, the
Oregon Trawl Commission, an Oregon indus-
try group, has initiated a limited voluntary ob-
server program linked to an enhanced fishery
logbook program to estimate bycatch discards
in the deep-water complex fishery under an
experimental fishing permit. Observers began
riding vessels in November 1995. Since then the
program has expanded to include vessels land-
ing in both Washington and California. The
project has observed nine vessels on 52 trips and
has put enhanced logbooks on board four addi-
tional vessels.
The other major West Coast bottom trawl
fishery is the shrimp trawl fishery. Bycatch dis-
cards in this fishery are known to include
groundfish species, Pacific halibut, chinook
salmon, and squid. Although the amount oi~
..
LohaI
-jtectives
groundfish bycatch in the shrimp trawl fishery
is unknown because of the lack of an at-sea sam-
pling program, its existence is recognized. Dis-
card wastage is intended to be minimized by fed-
eral regulations that provide a landing allowance
(other than the bycatch allowance, the fishery is
state-managed). Thro ugh 1996, shrimp trawlers
were permitted to land up to 1,500 pounds of
groundfish per trip to prevent discard wastage.
Because of the recent reduction in the yellow-
tail rockfish allowable catch, however, the 1997
fishing regulations will reduce the bycatch land-
ing allowance from 1,500 to 500 pounds per
trip. Some work has also been done under a
Saltonstall— Kennedy grant to develop and test
finfish excluder devices; some shrimp fishermen
are now using them routinely.
The International Pacific Halibut Commis-
sion estimates the Pacific halibut bycatch in the
shrimp bottom trawl fishery to be 56 mt.
Other groMtdfcfa Fisheries
Other groundfish fisheries include bottom
longline and pot (fish trap) fisheries for sable-
fish; other line (vertical longline, etc.) fisheries
for rockfish; bottom gill nets for rockfish; and
the recreational groundfish fishery, which is sig-
nificant for some species such as lingcod and
bocaccio rockfish.Very little is known regarding
the amount of bycatch discards; mortalities; and
the social, economic, or biological impacts of
the bycatch in these fisheries.
For the West Coast sablefish longline fish-
ery, the International Pacific Halibut Commis-
sion estimates the Pacific halibut bycatch based
on a relationship between halibut and sablefish
exploitation rates by the sablefish fisheries of the
West Coast and Alaska, since there are no direct
data derived from the West Coast sablefish fish-
ery.The current Pacific halibut bycatch estimate
is 41 mt.
Pacific Salmon
The federally managed ocean salmon fisher-
ies are divided into commercial troll and recre-
ational fisheries. Both groups use hook-and-line
gear. Inside-water commercial fisheries, which
are managed by the states and treaty tribes, use
gill nets and purse seines. Bycatch in the ocean
commercial troll and recreational salmon fisher-
ies has two major components. The first is the
catch and discard of depressed or endangered
salmon species, for which there is no total al-
lowable catch in a mixed-stock fishery with other
salmon species. The second is the catch and dis-
card of salmon species either coastwide or by
management area, where the quota for one spe-
cies of salmon is taken before the quota for the
other species.
The primary salmon species taken in the
ocean fisheries are chinook and coho salmon.
Since 1994, because of very depressed coho
salmon stocks (both hatchery and wild), reten-
tion of coho has been prohibited off the coasts
of both Oregon and California. Even though
retention was prohibited, it was estimated that
hook-and-release mortality of coho salmon taken
incidentally in chinook salmon fisheries was 8%
for the recreational fishery and 26% for the com-
mercial troll gear fishery, with an additional 5%
for each fishery for drop-offs (fish hooked, but
not landed). Coho salmon bycatch rates would
be higher if chinook harvests were not con-
strained to limit the bycatch of coho salmon.
The reverse has occurred off the coast ofWash-
ington where coho harvests have been allowed
some years, but chinook salmon retention is pro-
hibited to protect weak and endangered Colum-
bia River chinook salmon stocks. Estimates of
salmon encounter rates and hook-and-release
mortalities in nonretention fisheries are limited
to a few specific areas, but are essential to assess
the impacts of harvest on weak and endangered
West Coast
Fisheries
stocks. Recent studies have updated past esti-
mates, but more work is necessary. Some of this
work has been conducted using fishermen funded
by the Northwest Emergency Assistance Program.
The states of Oregon andWashington have
begun to mass-mark hatchery coho salmon, be-
ginning with the 1995 brood year, with the in-
tent of prosecuting selective fisheries on return-
ing adults in 1998. In a selective fishery, fisher-
men would keep only fin-marked hatchery fish,
while releasing unmarked native or wild fish. To
evaluate the potential impacts on wild fish, bet-
ter information is needed on both encounter
and hooking-mortality rates of unmarked fish.
These studies are a high priority for funding
under the Salmon Disaster Relief Program,
which began in 1994 in the wake of major
salmon stock collapses.
The bycatch of seabirds (common murres
and endangered marbled murrelets) occurs in
gill-net and purse-seine fisheries in the Colum-
bia River and Puget Sound. In recent years, bio-
logical opinions prepared by the U.S. Fish and
Wildlife Service have required observer programs
to assess the incidence of seabird bycatch. The
Saltonstall— Kennedy program is currently fund-
ing experimental gear research in the purse-seine
fishery. Bycatch of marine mammals, mainly
harbor porpoise, occurs in the net fisheries. In a
recent Stock Assessment Report (Barlow et al.
1 995), NMFS found that the minimum total fish-
ery mortality and serious injury of harbor por-
poise cannot be considered insignificant and that
the status of the harbor porpoise stock be re-
viewed during 1997.
Pacific Coastal Pelagic Fisheries
The major target species in the Pacific
coastal pelagic fishery are the northern anchovy,
jack mackerel, Pacific sardine, and Pacific mack-
erel. These species are naturally dynamic, highly
responsive to environmental conditions, and sub-
ject to wide fluctuations in abundance and dis-
tribution, even in the absence of a fishery. They
are very important as live bait in the recreational
fisheries for gamefish, groundfish, and salmon.
The species also support a low-volume, but high-
value fishery for dead bait, pet food, and dried
fish as well as lower-value fisheries for canning
or reduction.
The fisheries are distributed internation-
ally, with components in the exclusive economic
zones of Mexico and Canada. There is no bilat-
eral agreement with Mexico regarding anchovy
management. The fishery management plan al-
locates 70% of the annual optimum yield to the
U.S. reduction fishery, and 70% of the quota for
nonreduction purposes to the U.S. exclusive eco-
nomic zone.
Commercial landings are monitored from
information provided from processors'"fish tick-
ets" and a California Department of Fish and
Game port sampling program.
The Pacific Fishery Management Council
manages the anchovy fishery under the North-
ern Anchovy Fishery Management Plan. Pacific
sardine and Pacific mackerel are managed by the
state of California. Jack mackerel north of lat.
39° N are managed under the Pacific Coast
Groundfish Fishery Management Plan.
Several of the species are particularly im-
portant in the ecosystem. Anchovies, for example,
are an important food source for the endangered
brown pelican. Coastal pelagics are important
for the endangered least tern. While these spe-
cies are a key component of marine food webs
and the primary prey of many seabirds, it is not
currently possible to estimate the total amount
of the species that is necessary to sustain the
predator populations. However, the fishery man-
agement plan for anchovy specifies a threshold
for its optimum-yield determination to prevent
anchovy depletion and provide adequate forage
for marine fish, mammals, and birds.
ional
Coastal pelagic species support a multispecies
fishery in which by catch is common. By catch
usually consists of other coastal pelagic species,
but may include other species as well. The di-
rected fishery for anchovy has little bycatch.
Bycatch of sardines and Pacific mackerel may be
important from an economic or allocative point
of view when harvest quotas or guidelines for
one species are reached and another is not. Un-
der a multispecies quota management system,
discards of species for which the quota has been
met may increase while fishing activity contin-
ues for other species in the complex. However,
with this possibility in mind, California regu-
lates bycatch with a system of bycatch allow-
ances and overall incidental reserves by retain-
ing a portion of any harvest guideline to ap-
portion at a later time if the fishery for one
species could be closed because the harvest
guideline has been reached for another species.
This management approach appears to work
well.
Because of the nature of the stocks, it is
unlikely that bycatch poses a biological risk to
the species. Little or no information on bycatch
of marine mammals or protected species is avail-
able, but the impact is thought to be nonexist-
ent or very small.
Regional Recommendations
The need for monitoring bycatch and de-
termining ways to minimize bycatch mortality
remains a high priority in West Coast ground-
fish fisheries and Pacific salmon fisheries. Sev-
eral populations of Pacific salmon that have ei-
ther been listed or are being considered for list-
ing under the Endangered Species Act are taken
in directed fisheries or as bycatch in groundfish
fisheries. Observer data on salmon bycatch have
been available since 1990 from all at-sea pro-
cessing vessels in the Pacific whiting fishery on
a voluntary basis. There are no ongoing com-
prehensive at-sea observer programs in the
bottomfish fishery to determine bycatch of
salmon, Pacific halibut, or other regulated spe-
cies. The Pacific Fisheries Management Coun-
cil has recommended mandatory at-sea observer
programs for at-sea processors with vessels 125
feet long or longer in the Pacific whiting fishery
and is currently formulating recommendations
for bycatch data collection that will most likely
require an at-sea observer program for
bottomfish fisheries.
In recent years, the bycatch of seabirds (e.g.,
common murres and endangered marbled
murrelets) and marine mammals has been moni-
tored in gill-net fisheries. Better documentation
of the bycatch in these fisheries is needed, as
well as additional gear research to reduce or avoid
the taking of sea birds and marine mammals.
Based on research done to date, the state ofWash-
ington has imposed time, area, and gear restric-
tions to reduce the bycatch of seabirds in the
Puget Sound sockeye salmon net fishery.
Pacific coastal pelagic species are impor-
tant in West Coast recreational fisheries, in lim-
ited commercial fisheries, and are essential com-
ponents of the eastern North Pacific Ocean eco-
system. While current state and federal manage-
ment of individual pelagic species appears ad-
equate to minimize bycatch, little is known about
the interaction of these species as a complex,
the impacts of commercial and recreational fish-
ing (e.g., magnitude of discarding), and environ-
mental effects, such as El Nino, on population
fluctuations.
The high cost of obtaining information and
data on the magnitude of bycatch requires a
greater effort to utilize all existing data and to
be more selective about collecting new data.
Improved cooperation on the collection and
sharing of bycatch information and data with
the states, tribes, commercial and recreational
fishing industry, academia, conservation groups,
and other interested parties provides an oppor-
tunity to enhance "core" statistics and informa-
tion bases that will be essential to evaluate the
population, ecosystem, social, and economic
impacts of proposed bycatch management mea-
sures. Following are specific recommendations
for West Coast fisheries:
• Assess the magnitude of bycatch in West
Coast groundfish fisheries. With the excep-
tion of the mid-water whiting fishery, little
is known about the magnitude and compo-
sition of the bycatch in the bottomfish trawl
fishery.
• Develop and implement an at-sea observer
program in the Pacific groundfish bottom
trawl fishery cooperatively with the Pacific
Fishery Management Council and the Pa-
cific Coast fishing industry.
• Resolve legal and other issues that are im-
pediments to commercial and recreational
industry involvement in groundfish bycatch
data collection and research. NMFS and the
industry need to develop a more collabora-
tive relationship that can better utilize the
fishing industry to assist in the collection of
bycatch data.
• Explore management and enforcement poli-
cies that discourage regulatory discards in
groundfish fisheries.
Collaborate with the groundfish, salmon,
and coastal pelagic fishing industries to bet-
ter utilize industry resources to collect com-
mercial and recreational bycatch informa-
tion.
Develop better estimates of hook encoun-
ter and hooking mortality rates of salmon
taken in commercial and recreational salmon
fisheries. Encounter and mortality rates are
key components of assessing the impacts of
single-salmon-species fisheries on
nonretention species, and of the impacts of
selective fisheries for mass-marked hatchery
salmon on wild salmon stocks.
Develop selective harvest techniques in
ocean and freshwater fisheries that can be
used to target healthy, harvestable stocks of
salmon while protecting weak or recover-
ing salmon stocks.
Minimize the salmon bycatch in nonsalmon
fisheries through the collection of better in-
formation on the magnitude, distribution,
and stock composition of salmon bycatch,
and using the information to develop and
implement either voluntary or mandatory
bycatch-minimization measures.
Develop better documentation of the sea-
bird bycatch in the purse-seine and gill-net
salmon fisheries, and conduct additional gear
research to reduce or avoid the taking of sea
birds and marine mammals.
Fisheries
Regional Characteristics
Groundfish Fisheries
Alaska groundfish were harvested prima-
rily by foreign nations until the mid-1980s. The
foreign catches were replaced in the late 1980s
by joint- venture harvests by domestic fishermen
delivering to foreign processors. Fully domestic
operations developed rapidly in the late 1980s
and by 1991 were the only form of operation.
Currently, about 90% of the groundfish harvest
is taken with trawl gear, although harvest
amounts with hook-and-line, pot, and jig gear
are increasing. The selectivity of these gear types
in the multispecies groundfish fisheries varies by
target species, area, and time of year.
Groundfish stocks generally are in a healthy
and stable condition. The optimum yield of the
groundfish resource is established as a range in
the Bering Sea and Aleutian Islands (BASI) man-
agement area (1.4-2.0 million mt) and the Gulf
of Alaska (116,000-800,000 mt). All Alaska
groundfish stocks have fluctuated in abundance
over the years, but no widespread trend toward
decline is evident (NPFMC 1996a, 1996b). The
annual harvest of Alaska groundfish approaches
2.3 million mt.
Management of the Alaska groundfish fish-
eries is directed to maintain total harvest amounts
within annually specified total-allowable-catch
amounts. An extensive program that includes
monitoring by NMFS-certified observers and
an industry catch-reporting requirement is used
to estimate total fishing mortality. Management
tries to account for all sources of fishing mortal-
ity; estimated discard amounts of groundfish are
charged against the annual total-allowable-catch
amounts. When NMFS determines that the al-
lowable harvest level for a species has been taken,
the fishery is closed for the year. In 1995, the
total harvest of Alaska groundfish species (2.14
million mt) accounted for only about 64% of
the total acceptable biological catch (3.33 mil-
lion mt; NPFMC 1996a, 1996b).
Commercial Crab Fisheries
The management of the king and Tanner
crab pot gear fisheries in the BSAI area largely is
deferred to the state of Alaska under the federal
Fishery Management Plan for the Commercial King
and Tanner Crab Fisheries of the Bering Sea /Aleu-
tian Islands Area (NPFMC 1996c). Other crab
fisheries are managed by the state of Alaska with-
out federal overview.
The history of the eastern Bering Sea crab
fisheries extends back to the 1 930s, but large-
Alaska,
Fisheries
scale commercial efforts were not undertaken
until development of the foreign king crab fish-
eries in the 1950s. Foreign Tanner crab fisheries
were developed in the 1 960s. Foreign fishing for
king crab ceased in 1974, and foreign fishing for
Tanner crab in U.S. waters was prohibited un-
der the Magnuson Act in 1980 (Otto 1989).
Offshore areas of Bristol Bay have supported
large domestic fisheries for red king crab, snow
crab (Chionoecetes opilio), and Tanner crab (C.
bairdi).
In recent years, however, these stocks have
declined to low levels. In the 1995 eastern Bering
Sea fishery, 60.6 and 1.8 million snow and Tan-
ner crab were harvested, respectively. The 1994—
95 Bristol Bay red king crab fishery was closed
due to low abundance, but was reopened in 1996.
Commercial fisheries for other species of crab
exist, but at volumes less than those for the three
species that historically have supported large
commercial operations.
Salmon Fisheries
The management of the Alaska salmon fish-
eries is deferred to the state of Alaska under the
federal Fishery Management Plan for the Salmon
Fisheries of the EEZ off Alaska (NPFMC 1990).
State management of the salmon fishery is based
on sustainable optimal yield. It has resulted in
healthy salmon stocks for all species and record
harvest levels for all species, except chinook
salmon, which remains under conservative man-
agement. Management of the Alaska salmon fish-
ery strives to protect, to the extent possible, any
depressed stock, including those originating
south of the Alaska border.
Commercial fishing is conducted in both
state and federal waters using troll, drift gill-net,
set gill-net, and purse-seine gear. All five Pacific
salmon species are harvested by commercial, rec-
reational, and subsistence fishermen. Chinook
salmon are the most highly prized species be-
cause of their large size and excellent food qual-
ity. In Alaska, approximately 1 million chinook
salmon are harvested annually. While this is less
than 1% of the annual salmon catch off Alaska,
chinook salmon typically are the focus of a dis-
proportionately larger amount of management
and regulatory effort because of the conserva-
tion concerns and intense allocation issues for
this species. Increased focus on the Southeast
Alaska commercial salmon troll fisheries oc-
curred with the listing of Snake River sockeye,
Snake River spring/summer chinook, and Snake
River fall chinook in 1991 and 1992 under the
Endangered Species Act (NMFS 1997c).
Pacific Halibut Fishery
Commercial and recreational fisheries ex-
ist for Pacific halibut off Alaska. The Interna-
tional Pacific Halibut Commission (IPHC) has
the primary responsibility for managing the Pa-
cific halibut resource off Alaska. Under author-
ity of the North Pacific Halibut Act, the North
Pacific Fishery Management Council is autho-
rized to develop regulations that are in additional
to, but not in conflict with, regulations adopted
by the IPHC.The Council adopted an individual
fishing quota (IFQ) for the commercial Alaska
halibut fishery in 1 992. NMFS implemented the
program in 1995. Under the IFQ program, in-
dividual fishermen were assigned a quota share
based on past participation in the fishery and
other criteria developed by the Council. The
annual halibut quota established by the IPHC is
allocated among fishermen based on their indi-
vidual quota share. These quota shares are trans-
ferable harvest privileges within specified limi-
tations. Under the IFQ program, fishermen arc
able to harvest their halibut IFQ whenever and
however such harvest is most economical to their
fishing operation, subject to program limitations
and seasons. In 1996, over 18,000 mt of halibut
were harvested in the IFQ fishery. Recent im-
p
provements to the halibut stock assessment mod-
els used by the IPHC resulted in an estimate of
halibut abundance that is above the long-term
potential yield.
Other Fisheries
The commercial scallop fishery off Alaska
is managed under the Fishery Management Plan
for the Scallop Fishery off Alaska (NPFMC 1996d).
Federal regulations governing this fishery gen-
erally mirror Alaska state regulations. Participa-
tion in the federal water scallop fishery is con-
strained by a vessel moratorium implemented
in 1997. A total of 18 vessels currently are eli-
gible to participate. Alaska recently has imple-
mented a separate limited-entry program for state
waters. The North Pacific Fishery Management
Council currently is pursuing an amendment to
the federal fishery management plan that would
defer most scallop fishery management measures
to the state. The commercial landings of shucked
scallop meats have varied widely since the late
1960s, with peak annual landings in excess of
1.8 million pounds. In 1996, about 583,000
pounds of shucked meat were landed.
Other fisheries off Alaska that are managed
by the state include commercial and subsistence
herring fisheries, as well as numerous small-scale
coastal fisheries for finfish, shellfish, and other
invertebrates.
Regional Bycatch Issues
Groundfish Fisheries
Since the late 1980s, a dramatic increase in
harvesting and processing capacity in the do-
mestic open-access groundfish fisheries has re-
sulted in an extremely competitive race for fish,
with every vessel pressured to catch its share of
the quotas before the fleet harvests the ground-
fish quotas or before prohibited species bycatch
restrictions close the fishery. This situation frus-
trates any inclination vessel operators may have
to alter fishing practices to reduce bycatch if such
action puts them at a competitive disadvantage
relative to other participants in the fishery. For
this reason, the controversial option of individual
fishing quotas has been promoted by some fish-
ery participants as a means to allow a market-
driven incentive to reduce bycatch.
The overall bycatch and discard rate in the
Alaska groundfish fishery is not exceptional com-
pared to other major fisheries in the world
(Alverson et al. 1994), although individual fish-
ery or vessels rates can be high.kHowever, the
2.3-million-mt fishery is so immense that the
absolute volume of discards and the foregone
opportumT7"tlTeyTepresent have rai sed nationa l
and industry consciousness, and pose_a_signifi-
canTcohcern tcTother fisheries dependent on
s ome of the byca tch species. For example, the
1995 Bering Sea mid- water pollock fishery har-
vested over 1 . 1 million mt of fish, of which al-
most 46,000 mt were discarded (a discard rate
of only 4%) (NMFS 1996c) .The Bering Sea rock
sole, flathead sole, and other flatfish trawl fisher-
ies typically experience high discard rates rela-
tive to other Alaska groundfish fisheries (about
55% of the total catch in 1995), although other
small-scale trawl and hook-and-line fisheries
have exceeded this rate. Overall, the 1995 dis-
card rates in the Alaska trawl and hook-and-line
fisheries were 14% and 18%, respectively. By
volume, however, discard amounts in the trawl
fisheries accounted for 91% of the total 1995
discards in the Alaska groundfish fisheries (NMFS
1996c).
The NMFS Alaska Region catch reports
for 1995 estimate that total discards in the
groundfish fisheries include about 285,000 mt
of groundfish, 7, 190 mt of dead halibut, 123,300
(individual) salmon, 1 ,020 mt of herring, and
almost 8 million (individual) crab (mostly Tan-
ner crab). Pacific halibut, salmon, herring, and
Alaska,
Fisheries
crab are prohibited species in the groundfish fish-
eries and must be discarded under existing regu-
lations. Regulations also limit the amount of a
groundfish species that may be retained on board
a vessel if the species is closed to directed fish-
ing. Catch amounts of these species that exceed
the maximum retainable bycatch amount must
be returned to the sea. Most groundfish discard
reflects discretionary decisions on the part of
industry (e.g., undersize fish, no market, male
fish in roe fisheries), although regulatory discards
also account for a significant portion of the
groundfish bycatch that is returned to the seas.
The absolute percentage of discretionary versus
regulatory discards is not known.
The bycatch of prohibited species (Pacific
halibut, crab, salmon, and herring) in the ground-
fish fisheries has been a major focus of attention
since the days of foreign fishing. Stocks of some
of these species have declined, particularly some
crab stocks, and management agencies are con-
cerned about all sources of mortality, including
bycatch mortality. Furthermore, the pressure to
address the allocative implications of bycatch
mortality of these fully utilized species, as well
as concerns about the potential impact of fish-
ing operations on crab habitat, have propelled
the NPFMC to recommend numerous manage-
ment measures to address these concerns and
mitigate potentially adverse impacts on declin-
ing stocks of prohibited species. Foremost among
these measures is the establishment of area clo-
sures and prohibited species bycatch limits that,
when reached, result in groundfish fishery clo-
sures.
Bycatch limits, area closures, and other pro-
hibited-species bycatch mitigation measures limit
the overall bycatch mortality of these species in
the groundfish fisheries, and have protected sen-
sitive habitat areas. However, they also have cre-
ated barriers to harvesting groundfish total-al-
lowable-catch amounts, and have generated tre-
mendous allocative controversy among various
users of species taken as bycatch in the ground-
fish fisheries. Furthermore, the multispecies na-
ture of the bycatch problem in the groundfish
fisheries creates a situation where a solution for
one species' bycatch problem often exacerbates
the bycatch problem for a different species.
Recently, several high-valued groundfish
species that have relatively low acceptable-bio-
logical-catch levels have posed significant bycatch
issues that are similar to those experienced for
prohibited species. These species, such as
Greenland turbot or several species of rockfish,
often are not open to directed fishing because
the full total allowable catch (TAC) is needed to
support bycatch needs in other fisheries. How-
ever, bycatch amounts of these species can be
retained up to a specified percentage of other
retained catch. Once a species' TAC is reached,
further retention is prohibited. In some cases,
continued bycatch amounts approach or reach
the overfishing level, and fisheries that cannot
avoid the bycatch of the affected species are
closed. These closures can prevent fishermen
from harvesting other groundfish quotas. In many
cases, allocative issues can develop to the extent
that the bycatch in one fishery can be sufficiently
large to exceed the TAC and can approach over-
fishing levels early in the year, thus preempting
other fisheries that start later in the year from
opening or harvesting available groundfish quo-
tas.
Concerns about marine mammal bycatch
in the Alaska groundfish fisheries exist, particu-
larly for killer whale interactions in the hook-
and-line gear fisheries and Steller sea lions in
the groundfish trawl fishery. Incidental takes of
Steller sea lions averaged 12 per year during
1991-95 in the Bering Sea/ Aleutian Islands
groundfish trawl fishery, a level of take that may
be a cause for concern for a stock that continues
to decline for unknown reasons (Hill et al. 1997).
Regional
Persj*ectw&
Recently, the mortality of marine birds (includ-
ing the short-tailed albatross, an endangered spe-
cies) in the Alaska hook-and-line gear fisheries
has received a great deal of attention. In 1997,
regulations were implemented for the ground-
fish hook-and-line gear fisheries that require
mandatory use of bird-avoidance gear and fish-
ing methods.The NPFMC also has adopted these
measures for the Pacific halibut fishery.The bio-
logical opinion developed as part of a recent
Endangered Species Act section 7 consultation
for short-tailed albatross (USFWS 1997) requires
NMFS to develop a research plan to assess the
effectiveness of seabird by catch mitigation mea-
sures, with the understanding that measures
implemented to date would be adjusted as nec-
essary to reduce seabird bycatch mortality.
Commercial Crab Fisheries
The selective nature of commercial crab
pots results in very limited bycatch amounts of
noncrab species. Although bycatch of ground-
fish and Pacific halibut does occur, the small
amounts caught have not instigated resource
management or allocative concerns. Crab
bycatch mortality in the directed crab fisheries
is receiving increased scrutiny given the overall
decline in crab stocks. Crab bycatch includes
females of target species, sublegal males of target
species, and nontarget crab. Due to the differ-
ence in legal size versus market size for snow
crab, a portion of the legal crab are not retained.
The number of crab taken as bycatch in the 1995
commercial crab fisheries is estimated at over 75
million (Appendix A).
Some discarded crab die because of han-
dling mortality or predation. Estimates of han-
dling mortality rates range widely based on gear
type, species, molting stage, number of times
handled, temperature, and exposure time
(Murphy and Kruse 1995). Crab mortality also
is caused by ghost fishing, which is the term used
to describe continued fishing by lost or derelict
pot gear. Crab captured in lost pots may die of
starvation or by predation. The impact of ghost
fishing on crab stocks remains unknown, al-
though management agencies hope that pot lim-
its and mandatory pot gear escape mechanisms
have reduced ghost fishing due to pot loss in
recent years.
Salmon Fisheries
The bycatch of nonsalmon species in the
directed salmon fisheries is not monitored or
quantified under the assumption that bycatch
amounts are small and do not affect population
levels. The bycatch problem in the state-man-
aged Alaska salmon fisheries centers around the
interception of other salmon species or runs. This
interception creates allocation issues and, in some
cases, gives rise to conservation concerns.
Of particular interest are the salmon fish-
eries in Southeast Alaska, which intercept salmon,
including ESA-listed Pacific Northwest stocks,
passing through the marine waters off the coast
of Alaska on their way to more southerly spawn-
ing grounds. This interception is the focus of
ongoing negotiations and debate among Alas-
kan, Canadian, and Pacific Coast fishermen,
management agencies, and governments.
Another important bycatch issue in the
commercial and recreational hook-and-line fish-
eries is the capture of undersized chinook salmon
which must released.While the majority of these
fish survive the hooking encounter, large num-
bers can be hooked and substantial mortality
incurred. Larger mature chinook salmon must
also be released during commercial troll and net
fisheries for coho salmon and other species once
season quotas for retaining chinook salmon are
reached. These quotas are part of the United
States/Canada salmon treaty process. Under these
conditions in the commercial troll fishery for
coho salmon, the nonretention catch of chinook
Alaska,
Fisheries
salmon often exceeds 100,000 maturing fish that
are subject to significant hook-and-release mor-
talities.
Due to bycatch of marine mammals, the
salmon gill-net fisheries are classified as Category
II fisheries under the Marine Mammal Protec-
tion Act. A limited observer program in the
Prince William set and drift gill-net fisheries
documented significant seabird bycatch, and the
persistence of intentional lethal taking of ma-
rine mammals. NMFS is currently developing a
comprehensive observer program for other
Alaska gill-net fisheries with the primary focus
of determining the nature and extent of marine
mammal interactions in these fisheries; seabird
bycatch information will also be collected.
Other Fisheries
Bycatch of crab in the Alaska scallop dredge
fishery is limited by area-specific crab bycatch
limits. Observer data are used to estimate crab
bycatch. When the bycatch limits are reached in
an area, it is closed to fishing for scallops. Ob-
server data do not suggest that significant bycatch
of other species occur in the scallop fishery.
The Pacific halibut fishery does not have
an observer program to monitor the discard
mortality of undersized halibut or other species
in the fishery, although logbook data are used
by the International Pacific Halibut Commis-
sion to estimate halibut dead loss in the fishery.
Logbook data also are collected on bycatch of
some groundtish species. Seabird bycatch mor-
tality in the halibut fishery is of concern, and
regulations have been proposed to implement
mandatory gear and fishery operation restric-
tions to reduce seabird bycatch.
The impact of nongroundfish fisheries on
marine mammals or other protected species is
largely unknown because of the lack of data on
interactions. To date, an Alaska take-reduction
team has not been formed because reliable in-
formation regarding take levels of marine mam-
mals in unobserved Alaska commercial fisheries
is not available to make a determination as to
whether a take-reduction team is warranted.
Regional Bycatch Programs
Research and monitoring programs to ad-
dress the bycatch problem off Alaska primarily
are based on data collected from an industry-
funded mandatory observer program. Research
is focused on (1) how bycatch operates within
various fisheries and gear types, (2) gear modifi-
cation to reduce bycatch rates, (3) mortality as-
sociated with discards by gear and fishery, and
(4) the relationship of bycatch in terms of abun-
dance to the stock status of bycatch species and
the effect of bycatch on other fisheries.
Numerous regulatory approaches have
been implemented to address or reduce bycatch
in the Alaska groundfish and shellfish fisheries.
These include bycatch limits for prohibited spe-
cies, gear restrictions, season delays or time/area
closures, a vessel incentive program, an individual
fishing quota program for hook-and-line sable-
fish and halibut, mandatory retention and in-
creased utilization of pollock and Pacific cod
(proposed program that would be expanded to
include rock sole and yellowfin sole within five
years), and voluntary industry initiatives. The af-
fected industry has been instrumental in the
development and successful implementation of
most of these programs. Despite a high level of
compliance with these programs, difficulties ex-
ist in assessing the effectiveness of these regula-
tions in promoting either a long-term reduc-
tion in bycatch to the extent practicable or posi-
tive responses in abundance of species of con-
cern. Although absolute bycatch mortality of
prohibited species is reduced by bycatch limits
and time/area closures, few of these measures
Regional
Perspectives
actively promote independent efforts to under-
stand the cause and effect of bycatch.
Bye ate h Monitoring and
Assessment Strategies
Qrotmdfish Observer Program
An important element in determining the
magnitude and character of the bycatch prob-
lem in the Alaska groundfish fisheries is the
monitoring program that has been implemented
for the domestic fishery since 1990. Observer
catch data are submitted to NMFS on a weekly
(or daily, if necessary) basis. Observer data on
groundfish catch and bycatch rates of halibut,
salmon, crab, and herring are blended with in-
dustry-reported groundfish catch to derive a
"blend" estimate (based on an established "blend
algorithm") of groundfish catch and associated
prohibited-species bycatch amounts. This infor-
mation is used for in-season monitoring of
groundfish catch and prohibited-species bycatch
amounts, and for analysis of present and future
management measures. The observer program
also collects data on the viability of Pacific hali-
but bycatch for use in estimating discard mor-
tality rates in specified groundfish fisheries.
The observer data on species catch com-
position and amount in the groundfish fisher-
ies provide substantial, but not complete, infor-
mation on the characteristics of bycatch. In re-
cent years, other observer priorities have pre-
vented the collection of adequate size and sex
composition data for crab bycatch. Stock iden-
tification information is relatively limited, and
considerable uncertainty exists concerning the
handling mortality rates for discards. Increasing
concern about seabird bycatch has prompted
interest to collect additional information on
seabird interactions that observers are currently
unable to collect because of other data collec-
tion priorities.
The observer program also does not pro-
vide estimates of the bycatch mortality that oc-
curs when fish and shellfish come in contact with
fishing gear, but are not brought up with the
gear. This includes fishing mortality caused by
lost gear and fish that escape the gear, but not
without incurring fatal injuries.
The mandatory groundfish observer pro-
gram has an annual cost of more than $8 mil-
lion, of which more than $6 million is paid by
the vessels and processing plants that are required
to have observers. To fish, vessels 125 feet long
or longer must have an observer on board at all
times.Vessels 60-124 feet long must have an ob-
server on board 30% of the days that fishing gear
is retrieved and groundfish are retained.
Mothership and shoreside processors receiving
less than 1 ,000 mt of groundfish during a month
must have an observer present 30% of the days
groundfish are received or processed; those pro-
cessors that receive greater amounts of ground-
fish must have an observer present each day of
operation.
Alaska, State- Shellfish Observer Program,
At-sea observers are required by Alaska state
regulation on all vessels processing king or Tan-
ner crab at sea throughout Alaska and on all ves-
sels participating in the brown king crab fishery
in the Aleutian Islands area. At-sea observers are
required as a special permit condition for all ves-
sels participating in other crab fisheries. Alaska
state regulations also require 1 00% observer cov-
erage on vessels fishing for scallops, although
certain exemptions exist for the small-boat fleet
fishing in Cook Inlet. Federal regulations imple-
menting the Fishery Management Plan for the
Alaska Scallop Fishery (NPFMC 1996d) mirror
the state's observer coverage requirements.
Data collection by shellfish observers is es-
sential to the Alaska Department of Fish and
Game (ADFG) as a primary means for gather-
ing the data that are used for research, in-season
management, and development of management
measures, as well as for enforcement of regula-
tions. Shellfish observers currently collect data
to assess the magnitude of bycatch and by catch
discard in the crab and scallop fisheries. ADFG
believes the mortality of crab discarded in the
shellfish fisheries is significantly less than 100%,
although the actual mortality rate can vary
among fisheries and vessel types.
Currently, crab and scallop vessel owners/
operators must pay for observers. The state is
exploring alternative cost-recovery programs to
nullify the issue of costs to vessel operators. Al-
ternative programs could provide more manage-
ment flexibility to deploy observers in a manner
appropriate to meet the changing needs for shell-
fish resource management and research.
Other Observer Programs
At present, no other observer programs exist
other than for the groundfish and shellfish fish-
eries. NMFS is developing a proposal for the
implementation of a marine mammal interac-
tion monitoring program for commercial fish-
eries off Alaska. The proposed program is based
on a feasibility study conducted in 1995 through
a contract with Marine Mammal Protection Act
funds (Wynne and Merklein 1996). The intent
of the proposed program would be to achieve a
basic understanding of the rate of mortality and
serious injuries occurring to marine mammals
in Alaska Category II fisheries.
The initial proposal is intended to be the
start-up phase for a long-term monitoring pro-
gram to assess the impact of commercial fisher-
ies on marine mammal stocks, and to collect
information on the level and types of interac-
tions. To date, logbooks have been the primary
source ot information on marine mammal/com-
mercial fisheries interactions in Alaska because
only two of the current 13 Category II fisheries
in Alaska have been observed. Under the pro-
posed observer program (subject to funding),
eight previously unobserved fisheries would be
monitored for one fishing season each over the
next three years (1998—2000) to obtain an ini-
tial, reliable estimation of mortality and serious
injury levels. All eight fisheries target salmon and
are Alaska state-managed fisheries.
CstixJisReftorting (mil Monitoring
A comprehensive record-keeping and re-
porting program has been established for the
Alaska groundfish fisheries, which supplements
the data collected by observers. Processor ves-
sels are required to maintain daily cumulative
production logbooks that record the amount of
discards, and the amount and type of product
produced from retained catch. This information
is submitted to NMFS weekly, although moni-
toring requirements for a fast-paced fishery may
require that this information be submitted daily.
Shoreside processors record landed weight of
each species and associated discard amounts. This
information also is reported to NMFS on a
weekly or daily basis. NMFS estimates total
groundfish catch based on a combination of
observer data and weekly catch reports from
processors. Discard rates from these observer data
are applied to the shoreside groundfish landings
to estimate total at-sea discards from both ob-
served and unobserved vessels.
The principal objective of the groundfish
observer program is to provide adequate esti-
mates of total catch by species and not to differ-
entiate between retained and discarded catch. For
at-sea processors, the observers generally esti-
mate total groundfish catch directly, as opposed
to estimating retained catch and discarded catch
separately and adding the two estimates. How-
ever, the total catch estimate for shoreside pro-
cessors is the sum of observer estimates of at-sea
discards by catcher vessels and the catch that is
Regional
Perspectives
delivered to shoreside processors and reported
in the processors' weekly reports.
Voluntary Industry tiiforuwdLon, Systems
Participants in the Bering Sea bottom trawl
fisheries for flatfish voluntarily have developed
an information system to distribute to the fish-
ing fleet timely data on prohibited-species
bycatch rates and bycatch hot spots (Gauvin et
al. 1996). In the program, observer data on catch
and bycatch are electronically transmitted from
each participating vessel to Sea State, a private
contractor located in Seattle. Sea State conducts
statistical expansions from observer data to cal-
culate an average bycatch rate per vessel for each
24-hour period. Daily bycatch rates are then
placed in a format where the relationship be-
tween bycatch rates and locations is accessible
to vessel operators and vessel companies. Sea State
relays this information to participants every 24
hours via fax or by a computer file loaded into a
plotting program provided to the vessel. The goal
of the program is to allow the fleet to rapidly
respond (both individually and collectively) to
high bycatch rates and to reduce bycatch rates
of prohibited species. Assessments of observed
vessel bycatch rates in the Pacific cod and flat-
fish fisheries indicate that vessels participating
under the Sea State program experience signifi-
cantly reduced bycatch rates compared to non-
participating vessels. The Sea State program is
recognized by industry and management agen-
cies as an effective tool for monitoring bycatch
in the groundfish fisheries.
A separate information system has been
voluntarily developed for vessels participating in
the Bering Sea hook-and-line fishery for Pa-
cific cod (Smith 1996). Participants in this fish-
ery developed a careful release procedure to de-
crease the mortality of Pacific halibut inciden-
tally taken while targeting Pacific cod. This pro-
cedure ultimately was incorporated into regula-
tions implemented by NMFS. Working with
Fisheries Information Services (FIS), a private
consultant, the freezer-longline fleet organized
an industry monitoring program for halibut
bycatch mortality. Each week, vessels fax pre-
liminary observer data on the physical condi-
tion of released halibut to FIS. FIS calculates the
halibut mortality for each vessel and faxes it back
to the vessel operators who immediately learn if
they are fishing in a high bycatch area or if their
crew are mishandling halibut. Two-thirds of the
fleet participated in the program during 1995,
which is credited by the fleet to have reduced
halibut discard mortality substantially.
Bycatch Management Program
The management of the crab and scallop
fisheries generally is geared to minimize crab
discard mortality through gear restrictions, sea-
son closures, bycatch limits, and area closures. In
response to concerns about bycatch in the Bering
Sea/ Aleutian Islands area and Gulf of Alaska
groundfish fisheries, the North Pacific Fishery
Management Council (NPFMC) has recom-
mended, and the Secretary of Commerce has
approved and implemented, a variety of man-
agement measures that, in part, were intended
to help control the bycatch of prohibited spe-
cies as well as reduce discard of groundfish. Of
the more than 40 amendments to the Fishery
Management Plan for the Groundfish Fishery of the
Bering Sea and Aleutian Islands Area (NPFMC
1997) that have been considered by the NPFMC
since 1982, about a third addressed primarily
bycatch issues, and about another fourth ad-
dressed some aspect of bycatch management.
Tutte/Area, Closures
Time/area trawl closures are implemented
around Kodiak Island, around the Pribilof Is-
lands, and in the Bristol Bay area of the south-
eastern Bering Sea to protect sensitive king and
Tanner crab habitat areas and to avoid bycatch
Alaska;
Fisheries
of crab during the molting season. Some of these
closures are year-round, others are seasonal.
Time/area closures are also implemented to re-
duce the bycatch rates of chum salmon in the
Bering Sea pollock fishery.
Bycatch Lvnuts
Bycatch limits for C. iWrr// Tanner crab, red
king crab, herring, chinook salmon, and chum
salmon are established for Bering Sea/Aleutian
Islands (BSAI ) management area trawl fisheries.
Bycatch limits for C. opilio Tanner crab have been
approved by the North Pacific Fishery Manage-
ment Council. Halibut bycatch mortality limits
also are established for all the BSAI and Gulf of
Alaska groundfish fisheries. Bycatch limits for
king crab and Tanner crab also are established
for the commercial king and Tanner crab fisher-
ies as well as the scallop fishery. These bycatch
limits may be apportioned among fisheries as
bycatch allowances.
Bycatch limits are effective measures to
control bycatch amounts of specified species, but
the potential costs to the affected industry
through foregone harvest opportunity can be
large. With the exception of the halibut bycatch
mortality limits, the attainment of a fishery
bycatch allowance triggers a time/area closure.
The attainment of a fishery bycatch mortality
allowance for halibut in the BSAI or Gulf of
Alaska closes the entire BSAI or Gulf to that fishery.
Vessel Incentive, Program
A vessel incentive program (VIP) was
implemented in 1991 to reduce halibut and red
king crab bycatch rates in the Bering Sea and
Gulf of Alaska groundfish trawl fisheries. Under
the VIP, halibut and red king crab bycatch stan-
dards are established semiannually.
Vessel operators and owners who exceed
these standards based on observer data are sub-
ject to prosecution. Large enforcement and le-
gal staff resources are required to develop and
prosecute a VIP case. Appeal procedures may
delay the final resolution of a potential violation
for years. To date, three cases have been brought
before an administrative law judge and were ruled
in favor of the National Marine Fisheries Ser-
vice. A fourth case was settled out of court.
The VIP has raised the consciousness of the
industry relative to individual vessel bycatch rates,
but whether the program has resulted in fleet-
wide reductions in halibut or crab bycatch rates
is difficult to assess. Frustration exists within the
industry because Constitutional due process and
other legal and enforcement constraints do not
enable NMFS to take more timely action against
individual vessels that exhibit chronically high
bycatch rates, take a disproportionate amount of
established bycatch limits, and increase the rate
at which fisheries are closed upon attainment of
those limits.
Qeav Restrictions
Pelagic Trawl Gear. Regulations specify a con-
figuration for pelagic trawl gear to more effec-
tively minimize the incidental take of halibut
and crab, limit the number of crab that may be
on board a vessel at any time, encourage vessel
operators to fish pelagic trawl gear off the bot-
tom when NMFS has closed fishing with
nonpelagic trawl gear.
Mandatory Procedures for Careful Release of Hali-
but in the Hook-and-Line Gear Fisheries. Pacific
halibut discard mortality rates in the Alaskan
groundfish fisheries are routinely estimated from
viability data collected by at-sea observers. These
data are analyzed by staff of the International
Pacific Halibut Commission and NMFS, which
result in recommendations to the North Pacific
Fishery Management Council for managing
halibut bycatch limits in the upcoming fishing
year.
MMMfc-^^H
Current regulations require vessels using
hook-and-line gear to release halibut in a man-
ner that minimizes handling mortality. The in-
tent of this measure is to reduce not only mor-
tality rates but also the amount of halibut re-
quired by these fisheries to harvest available
amounts of groundfish under halibut bycatch
restrictions.
Pot Gear. Regulations require that groundfish pot
gear be fitted with halibut excluder devices and
biodegradable escape panels.
Season, Pelays or Seasonal Affiwrtionments
of Total Allowables Catch
Fishing seasons for specified groundfish
species are delayed to avoid high bycatch rates
of halibut. Similarly, annual total allowable catch
amounts and/or prohibited species bycatch al-
lowances may be seasonally apportioned to mini-
mize fishing operations when bycatch rates for
prohibited species are high.
Allocation, of Bering Sea, Pacific-
Cod Among gear Types
Regulations establish the allocation of
Bering Sea/ Aleutian Islands Pacific cod among
vessels using trawl and fixed gear. Although this
management provision is not solely based on
bycatch considerations, it is thought to reduce
halibut bycatch mortality in the cod fishery by
allocating more of the total allowable catch to
the fixed-gear fishery, which has a lower halibut
bycatch mortality rate, and by allowing the fixed-
gear fishery an increased opportunity to fish in
ways that further reduce halibut bycatch mor-
tality rates.
Regulations also authorize the seasonal
apportionment of the amount of Pacific cod al-
located to vessels using fixed gear.Their intent is
to avoid significant harvests of Pacific cod dur-
ing summer months, when halibut bycatch rates
are highest.
Indundual Pishing Quota, Program
An individual fishing quota (IFQ) program
for the Alaska sablefish and halibut fisheries was
implemented in 1995. The program is expected
to reduce halibut bycatch mortality in part by
slowing the pace of the sablefish hook-and-line
gear fisheries. Until a fisherman has used all of
his halibut IFQ, legal-sized halibut taken in the
sablefish fishery must be retained, rather than
discarded. The total catch of halibut is assumed
to be more effectively monitored as a result.
NMFS estimates that the total halibut discard
mortality in the 1995 Alaska hook-and-line
sablefish fishery was 148 mt, as compared to 615
mtin 1994 (NMFS,Alaska Region, unpublished
data) .
Voluntary Industry Initiatives to Reduce,
Prokibvt&L'Species Bycatck,
Several voluntary programs have been de-
veloped by trawl industry members to reduce
halibut bycatch in the yellowfin sole and Pacific
cod fisheries. Industry initiatives also resulted in
the publication of analyses of historical observer
data on fishery-specific bycatch rates of halibut
and other prohibited species, and in rulemaking
that authorizes the release of observer data on
vessel bycatch or bycatch rates of prohibited spe-
cies. This information is used by the industry to
identify sensitive times and areas of prohibited-
species bycatch and to provide an initial assess-
ment of proposed management measures to ad-
dress the halibut bycatch problem. More recently,
participants in the Bering Sea flatfish fisheries
have developed an in-season information sys-
tem to reduce prohibited-species bycatch rates.
In 1993, the industry formed the Salmon
Research Foundation to address the chinook
salmon bycatch problem in the Bering Sea trawl
fisheries. Vessels volunteering to participate in
the foundation's program agreed to pay a $20
fee for each chinook salmon taken during trawl
operations. Monies collected from the volun-
tary fee programs were intended to fund selected
research projects designed to address the salmon
bycatch problem. Subsequent action by the
North Pacific Fishery Management Council and
NMFS to establish salmon bycatch restrictions
and associated time/area closures greatly dimin-
ished the industry's initiative to continue the
voluntary fee-collection program and fulfill the
intent of the foundation. Nonetheless, fees col-
lected in 1993—94 were used by the foundation
to fund extra observer coverage in 1995—96 to
collect tissue samples necessary to enhance chum
salmon stock identification research under way
by NMFS. Reports have been submitted on the
genetic stock identification for samples taken in
1994 and 1995.
Sainton Donation Program
At the urging of the industry and the North
Pacific Fishery Management Council, together
with the experience gained under an experi-
mental donation program, NMFS has imple-
mented a program authorizing the voluntary
retention, processing, and donation of salmon
incidentally taken in the groundfish fisheries to
economically disadvantaged individuals through
a NMFS-authorized distributor. Currently, a
single authorized distributor, Northwest Food
Strategies, successfully administers donations
from almost 25 processors and numerous asso-
ciated catcher vessels under the salmon dona-
tion program. The Council has adopted a simi-
lar donation program for Pacific halibut taken
by trawl catcher vessels that deliver unsorted
catch to shoreside processors.
Imjurovut Retention and Utilization
Program,
The North Pacific Fishery Management
Council has approved an improved retention and
utilization program for the Alaska groundfish
fisheries that would require 100% retention of
pollock and Pacific cod caught in any ground-
fish fishery. Specified flatfish species-retention
requirements would follow, but would be de-
layed for a period of five years to allow for de-
velopment of markets and gear technology nec-
essary for vessels to effectively comply with the
requirements. The Council adopted a minimum
utilization rate of 15% for pollock and Pacific
cod. NMFS is proceeding with rulemaking that,
if approved by the Secretary of Commerce,
would implement the Council's action by 1998.
Seabird Avoidance, Program,
Federal regulations require operators of
vessels fishing for Alaska groundfish with hook-
and-line gear to conduct fishing operations in a
specified manner and to employ specified sea-
bird avoidance. Similar measures have been
adopted by the North Pacific Fishery Manage-
ment Council for the Pacific halibut hook-and-
line gear fishery.
Regional Research Initiatives
Gear research to reduce bycatch in the
Alaska fisheries has focused on changes in gear
technology and fishing methods to improve gear
selectivity. Some individual vessel operators and
fishing companies experiment with equipment
designed to avoid or reduce bycatch of nontar-
get species. The competitive nature of the open-
access groundfish fisheries, however, generally is
not conducive to voluntary adjustments in fish-
ing gear to reduce bycatch, especially if changes
necessary to achieve lower bycatch also result in
lower catch rates of target species.
The Alaska Fisheries Development Foun-
dation initiated a project supported by
Saltonstall-Kennedy program funding to assess
the effectiveness of experimental separator pan-
els in trawl nets to reduce the bycatch of Pacific
Regional
Perspectives
halibut in the Pacific cod fishery (Stone and
Bublitz 1996). Although preliminary results were
promising, the competitive nature of this open-
access fishery reduces the incentive of individual
fishermen to improve the selectivity of their fish-
ing gear. Fishermen who experiment with new
devices to reduce bycatch risk incurring opera-
tional costs and losing valuable fishing time, while
other competing vessels continue to use nonse-
lective nets.
Other research has been conducted on the
behavior offish encountering commercial trawl
gear in the North Pacific (Rose 1996). Species-
specific differences in fish behavior have been
observed using underwater video cameras, some
of which have applications for improving trawl
selectivity. The information provided by video
observations allows iterative development and
testing of gear modifications and fishing tech-
niques to find effective ways to reduce bycatch.
An independent trawl vessel association recently
was issued an experimental fishing permit by
NMFS to expand upon this research to reduce
the bycatch of groundfish (primarily pollock and
Pacific cod) in flatfish fisheries off Alaska.
Industry members, as well as the NPFMC,
have considered limiting the harvest of Alaska
pollock to mid-water trawl operations to reduce
halibut and crab bycatch. However, the open-
access nature of the groundfish fishery again frus-
trates this approach by aggravating the trade-off
between the gains associated with a reduction
in bycatch and the increased allocation and op-
eration costs that would ensue from restrictions
on the use of nonpelagic trawl gear in the pol-
lock fishery (Pereyra 1996).
A great deal of attention has been focused
on the use of trawl mesh restrictions to reduce
the catch of undersized fish in the Alaska ground-
fish trawl fisheries. Bublitz (1996) conducted
research to provide a predictive capability to as-
sess mesh-selectivity needs in the Alaska pollock
fishery. Other researchers (Pikitch et al. 1996)
pose a cautionary note on the effectiveness of
trawl mesh restrictions, particularly in high- vol-
ume fisheries, where escape rates decrease as
catch volume increases, regardless of mesh size
or configuration. The deleterious result of
"blocking" of cod-end meshes may be reduced
or eliminated by using sorting devices that per-
mit the escape of undersized fish before they
reach the cod end.
Other researchers have proposed an alter-
native type of cod end with very small mesh
size to reduce relative water velocity and en-
hance the ability offish to escape through vari-
ous bycatch-reduction devices (Loverich 1996).
The concept of cod ends made of very small
mesh size or even impermeable material runs
contrary to traditional thinking on cod-end se-
lectivity and escapement associated with cod ends
made of large-sized mesh.
Efforts also have been expanded to research
ways to reduce bycatch in the crab fisheries or
to reduce the unobserved mortality of crab as-
sociated with ghost fishing of" derelict" pots. The
Alaska Department of Fish and Game has imple-
mented minimum mesh size restrictions to en-
courage the escape of female and undersized
male crabs, as well as mandatory use of cotton
thread sewn into the bottom of all crab pots to
minimize ghost fishing in lost pots. King crab
excluder devices also are required to reduce tun-
nel height openings in the Tanner crab fisheries.
Research is ongoing to address factors affecting
crab entry into pots, improving the ability of
small crabs to escape, reducing discard mortality
due to damage while sorting unwanted catch,
and reducing mortality associated with ghost
fishing of lost pots (Stevens 1996,Wyman 1996).
With the exception of regulatory gear re-
strictions to reduce bycatch that may be applied
fleetwide, little incentive exists for individual
fishermen to voluntarily take action to change
*£
likerl&s
fishing gear or practices to reduce bycatch. As
stated by Stone and Bublitz (1996), "Unless there
becomes an economic advantage to fish cleanly,
such as in the case of individual bycatch ac-
countability, there is not likely to be any large-
scale trend toward the use of improved fishing
methods."
Regional Recommendations
Monitoring total catch, including discards,
and decreasing bycatch mortality in the Alaska
groundfish fisheries have been priorities of the
North Pacific Fishery Management Council
since it was established in 1976. An extensive at-
sea observer program and a comprehensive catch
reporting program generally are thought to pro-
vide adequate estimates of total catch by species
for the groundfish fishery as a whole. However,
the array of management measures that have been
used in whole or in part to decrease or limit
bycatch has not yet minimized bycatch to the
extent practicable.
A more difficult task will be to assess the
effectiveness of various bycatch-reduction mea-
sures that have been implemented. The domes-
tic Alaska groundfish fisheries are relatively new
and dynamic. The evolving nature of these
multispecies fisheries, together with the matrix
of different management programs governing
them that may affect the spatial or temporal dis-
tribution of fishing effort and associated bycatch
rites, creates a situation where impacts of spe-
cific management measures on bycatch rates may
be cumulative and difficult to assess individually.
An assessment of overall progress toward reduc-
ing bycatch can be attained through observer
data on catch composition and discard.
The difficulty in adequately addressing the
bycatch problem has resulted in an increasing
awareness of the necessity for better informa-
tion on winch to base bycatch decisions by fish-
ermen, fishery managers, and the public. This
input requires better understanding of (1) the
levels of bycatch; (2) the fishing practices and
techniques that can decrease bycatch mortality;
and (3) the population, ecosystem, social, and
economic effects of bycatch and of bycatch man-
agement measures. In addition, improved deci-
sions require increased efforts to ensure that
fishermen, fishery managers, and the public more
fully consider the impacts of their bycatch deci-
sions. Following are specific recommendations
for Alaska fisheries:
• Develop stable long-term funding for North
Pacific groundfish observer programs. Fund-
ing is currently totally contingent on "pass-
through" funding from several sources.
• Develop appropriate contractual arrange-
ments for observer services that would re-
duce the potential for conflicts of interest,
encourage the best observers to remain with
the program, and improve operational con-
trol of the program by NMFS.
• Improve data collection and catch estimation
procedures. A review of observer coverage lev-
els as well as observer data collection meth-
ods and associated catch estimation procedures
should be initiated to ensure that observer pro-
grams meet expectations of scientists, man-
agers, and the industry cost-effectively
• Improve the information concerning the
population, ecosystem, social, and economic
effects of bycatch and of bycatch manage-
ment measures.
• Require that proposed bycatch management
actions include clear statements o\ objec-
tives and performance criteria.
• Develop models to assess the probable fleet
response to alternative bycatch management
measures.
jiohaI
Support the establishment of international
guidelines for managing bycatch.
Increase involvement of the industry and
academic communities in analyzing factors
that affect bycatch rates by improving ac-
cess to observer and oceanographic data.
Conduct research on the survivability (acute
and chronic mortality) and recovery of
bycatch species from stresses imposed by cap-
ture.
Increase industry's involvement in the de-
velopment and testing of methods to reduce
bycatch mortality.
Improve technology transfer of bycatch re-
duction methods through reports, videos, and
workshops.
Establish a process for NMFS and the fish-
ing industry to examine the bycatch incen-
tives faced by fishermen and the degree to
which bycatch is a consequence of current
incentives and regulations. Identify regula-
tory changes that could be pursued to re-
duce regulatory discards.
Improve individual accountability by devel-
oping programs that improve incentives to
fishermen to consider the full costs and ben-
efits of their bycatch decisions and that al-
low fishermen to use the most cost-effec-
tive methods for reducing bycatch.
Conduct legal research to explore ways to
overcome potential impediments to indi-
vidual incentive programs and decrease the
monitoring and enforcement costs of bycatch
management programs that can provide catch
and bycatch accountability for individual
fishing operations.
Develop improved estimates of catch-and-re-
lease mortalities, especially for chinook salmon,
in both commercial and recreational fisheries.
Increase knowledge of the type and magnitude
of marine mammal bycatch, including ghost
fishing by lost nets, in the salmon drift-net and
Southeast Alaska purse-seine salmon fisheries.
r&mc&'
Keferi
wees
Alaska Sea Grant College Program. 1996. Solving Bycatch: Considerations for Today and Tomorrow. Alaska Sea Grant
College Program Report No. 96-03, University of Alaska, Fairbanks.
Alverson, D.L., M.H. Freeberg, S.A. Murawski, and J.G. Pope. 1994. A global assessment of fisheries bycatch and dis-
cards. Food and Agriculture Organization Fisheries Technical Paper No. 339. Rome, Italy, FAO, 233 pp.
Bailey, CM. 1996. The integrated shark research and management program: 1996 semi-annual report (DRAFT).
National Marine Fisheries Service, Silver Spring, Maryland.
Barlow, J., R.L. Brownell,Jr., D.P DeMaster, K.A. Forney, M.S. Lowry, S. Osmek,TJ. Ragen, R.R. Reeves, and R.J.
Small. 1995. U.S. Pacific marine mammal stock assessments. NOAA Technical Memorandum NMFS-SWFSC-
219.162 pp.
Blaylock, R.A., J.W. Hain, L.J. Hansen, D.L. Palka, and G.T. Waring. 1995. U.S. Atlantic and Gulf of Mexico
marine mammal stock assessments. NOAA Technical Memorandum NMFS-SEFSC-363, Washington,
D.C. 211 pp.
Brodziak.J. 1991. Bootstrap estimators of discard rates using domestic sea sampling data. Northeast Fisheries Science
Center SAW/ 12/P1/3, National Marine Fisheries Service,Woods Hole, Massachusetts.
Bublitz, C.G. 1996. Mesh size and shape: Reducing the capture of undersized fish. In Solving Bycatch: Consid-
erations for Today and Tomorrow. Alaska Sea Grant College Program Report No. 96—03, University of
Alaska, Fairbanks.
Center for Marine Conservation. 1996. Outreach strategy to promote a constructive public discourse on bycatch. A
report to the National Marine Fisheries Service, Washington, D.C.
Cheung, S.N.S. 1970. The structure of a contract and the theory of a non-exclusive resource. Journal of Law and
Economics 13:49-70.
Copes, P. 1972. Factor rents, sole ownership and the optimum levels of fisheries exploitation. Manchester School of
Economics 40:145-163.
CramerJ. 1996a. Species reported caught in the U.S. commercial pelagic longline, gill net, and pair trawl fisheries from
1987 to 1995. NMFS Southeast Fisheries Science Center, Miami Laboratory Contribution MIA-95/96-38,
Florida.
Edwards, S.F. 1991. A critique of three "economics" arguments commonly used to influence fishery allocations. North
American Journal of Fisheries Management 11:212-130.
Endangered Species Act. 16 U.S.C. 1531-1544 (1985).
FAO (Food and Agriculture Organization). 1995. Code of conduct for responsible fisheries. Food and Agriculture
Organization of the United Nations, Rome, Italy. 41 pp.
FAO (Food and Agriculture Organization). 1997. Report of the technical consultation on reduction of wastage in
fisheries. Food and Agriculture Organization of the United Nations, Rome, Italy. 27 pp.
Gabriel, W. 1993. Persistence of demersal fish assemblages between Cape Hatteras and Nova Scotia, Northwest Atlan-
tic. Journal of Northwest Atlantic Fishery Science 14:29-46.
Gabriel, W 1996. The role of targeted species in identification of technological interactions in Mid- Atlantic Bight
groundfish fisheries. Journal of Northwest Atlantic Fisheries Science 19:1 1-20.
Gates, J. M. 1976. Demand price, fish size, and the price offish. Canadian Journal of Agricultural Economics 22:1-12.
Resource
Material
Gates, J.M., and V.J. Norton. 1974. The benefits of fisheries regulation: A case study of the New England yellowtail
flounder fishery. Marine Technical Report No. 21, University of Rhode Island, Kingston.
Gauvin,J.R., K. Haflinger, and M. Nerini. 1996. Implementation of a voluntary bycatch avoidance program in the
flatfish fisheries of the eastern Bering Sea. In Solving Bycatch: Considerations for Today and Tomorrow. Alaska Sea
Grant College Program Report No. 96—03, University of Alaska, Fairbanks.
GMFMC (Gulf of Mexico Fishery Management Council). 1980. Fishery management plan for the shrimp fishery of
the Gulf of Mexico, United States waters. Gulf of Mexico Fishery Management Council, Tampa, Florida.
GMFMC (Gulf of Mexico Fishery Management Council). 1981. Environmental impact statement and fishery man-
agement plan for the reef fish resources of the Gulf of Mexico. Gulf of Mexico Fishery Management Council,
Tampa, Florida. 154 pp.
GMFMC (Gulf of Mexico Fishery Management Council). 1997. Draft amendment Number 9 to fishery management
plan for the shrimp fishery of the Gulf of Mexico, United States waters with supplemental environmental impact
statement, regulatory impact review and initial regulatory flexibility analysis. Gulf of Mexico Fishery Manage-
ment Council, Tampa, Florida. 154 pp.
Gordon, H.S. 1954. The economic theory of a common-property resource. Journal of Political Economy 62:124—142.
Gulf and South Atlantic Fisheries Development Foundation. 1995. An industry workshop addressing bycatch issues in
southeastern U.S. Fisheries. A report to the National Marine Fisheries Service, Tampa, Florida.
Hayes, D. 1991 . Exploratory analysis of four methods for estimating discards from sea sampling data. Northeast Fisher-
ies Science Center, SAW/12/P1/2, National Marine Fisheries Service, Woods Hole, Massachusetts.
Hill, P.S., DP. DeMaster, and R.J. Small. 1997. Alaska marine mammal stock assessments 1996. NOAA Technical
Memorandum NMFS-AFSC-78. 150 pp.
Hoagland, P., D.Jin, P. Lee, C. Croft, L. Davidson, and S. Wallis. 1996. Market-based incentives to reduce fisheries
bycatch. A report to the National Marine Fisheries Service, Silver Spring, Maryland.
Hoar, P., J. HoeyJ. Nance, and C. Nelson. 1992. A research plan addressing finfish bycatch in the Gulf of Mexico and
South Atlantic shrimp fisheries. Gulf and South Atlantic Fisheries Development Foundation. 114 pp.
ICCAT (International Commission for the Conservation of Atlantic Tunas). 1996. Report of the shark
working group: Sub-committee on by-catch meeting (Miami, Florida, USA, February 26-28, 1996).
COM-SCRS/96/.
ICCAT (International Commission for the Conservation of Atlantic Tunas). 1997. Report of the working group on
sharks. ICCAT Sub-Committee on Bycatch.
ICES (International Council for the Exploration of the Seas). 1986. Report of the working group on methods offish
stock assessment. November 20-26, 1985. ICES CM 1986/ Assess: 10, Copenhagen, Denmark.
ICES (International Council for the Exploration of the Seas). 1995. Report of the study group on unaccounted
mortality in fishes. ICES CM B:l, Copenhagen, Denmark.
Loverich, G.F. 1996. Thinking beyond the traditional codends. /// Solving Bycatch: Considerations for Today and
Tomorrow. Alaska Sea Grant College Program Report No. 96-03, University of Alaska, Fairbanks.
Magnuson Fishery Conservation and Management Act. 16 U.S.C. 1821 (1996).
Marine Mammal Protection Act. 16 U.S.C. 1361 (1994).
McCaughran, DE. 1992. Standardized nomenclature and methods of defining bycatch levels and implications. In
Proceedings of the National Industry Bycatch Workshop, pp. 200-201. Natural Resources Consultants, Inc.,
Seattle, Washington.
Migratory Bird Treaty Act. 16 U.S.C. 703 (1976).
Murawski, S.A. 1994. Opportunities in bycatch mitigation. In R.H. Stroud, editor. Conserving America's Fisheries.
Proceedings of a national symposium on the Magnuson Act, pp. 299-31 1. New Orleans, Louisiana, March 8-10,
1993. National Coalition for Marine Conservation, Savannah, Georgia.
Murawski, S.A. 1996. Factors influencing by-catch and discard rates: analyses from multispecies/multifishery sea sam-
pling.Journal of Northwest Atlantic Fishery Science 19:31-39.
Murawski, S.A.,A.M. Lange, and J.S. Idoine. 1991. An analysis of technological interactions among Gulf of Maine
mixed-species fisheries. ICES Marine Science Symposia 193:237-252.
Murphy, M.C., and G.H. Kruse. 1995. An annotated bibliography of capture and handling effects on crabs and lobsters.
Alaska Fishery Research Bulletin 2(l):23-75.
NEFSC (Northeast Fisheries Science Center). 1991. Report of the 12th Northeast regional stock assessment work-
shop (12th SAW), the plenary. Northeast Fisheries Science Center, Center Reference Document 91— 03, Woods
Hole, Massachusetts.
NEFSC (Northeast Fisheries Science Center). 1994a. Report of the 18th Northeast regional stock assessment work-
shop (18th SAW), the plenary. Northeast Fisheries Science Center, Center Reference Document 94-23, Woods
Hole, Massachusetts.
NEFSC (Northeast Fisheries Science Center). 1994b. Report of the 17th Northeast regional stock assessment work-
shop (17th SAW). Northeast Fisheries Science Center, Center Reference Document 94-06,Woods Hole, Massachusetts.
NEFSC (Northeast Fisheries Science Center). 1995. Status of fishery resources off the northeastern United States for
1994. NOAA Technical Memorandum NMFS-NE- 108. Woods Hole, Massachusetts. 140 pp.
NEFSC (Northeast Fisheries Science Center). 1996a. Report of the 22nd Northeast regional stock assessment work-
shop (SAW 22). Northeast Fisheries Science Center, Center Reference Document 96— 13, Woods Hole, Massa-
chusetts.
NEFSC (Northeast Fisheries Science Center). 1996b. Report of the 23rd Northeast regional stock assessment work-
shop (SAW 23). Northeast Fisheries Science Center, Center Reference Document, Woods Hole, Massachusetts.
NMFS (National Marine Fisheries Service). 1995. Reinitiation of consultation on 1994-1998 operation of the federal
Columbia River power system and juvenile transportation program in 1995 and future years. National Marine
Fisheries Service, Portland, Oregon.
NMFS (National Marine Fisheries Service). 1996a. Our Living Oceans. Report on the Status of U.S. Living Marine
Resources, 1995. NOAA Technical Memorandum NMFS-F/SPO-19. Silver Spring, Maryland. 160 pp.
NMFS (National Marine Fisheries Service). 1996b. Commercial catch data — fisheries of the U.S., 1995. National
Marine Fisheries Service, Silver Spring, Maryland.
NMFS (National Marine Fisheries Service). 1996c. Draft economic status of the groundfish fisheries off Alaska, 1995.
Alaska Fisheries Science Center, National Marine Fisheries Service, Seattle, Washington.
NMFS 1997a. National report of the United States to the International Commission for the Conservation of Atlantic
Tunas. October 1997. NMFS, Office of Sustainable Fisheries, Silver Spring, Maryland 20910.
NMFS 1997b. Report to Congress: Status of fisheries of the United States. September 1997. NMFS, Office of Sustain-
able Fisheries, Silver Spring, Maryland.
NMFS. 1997c. Public review draft environmental assessment for salmon fisheries in the EEZ and state waters oft" the
coast of Alaska. May 29, 1997. NMFS.Alaska Region, P.O. Box 21668, Juneau, Alaska.
4
Resource-
Material
J5 1
m
NOAA (National Oceanic and Atmospheric Administration). 1996. NOAA strategic plan: A vision for 2005. May
1996. U.S. Department of Commerce, Washington, D.C.
NPFMC (North Pacific Fishery Management Council). 1990. Fisheries management plan for the salmon fisheries in
the EEZ off the coast of Alaska. North Pacific Fishery Management Council, 605 W. 4th Avenue, Suite 306,
Anchorage, AK 99501-2252. 51 pp.
NPFMC (North Pacific Fishery Management Council). 1996a. Stock assessment and fishery evaluation report for the
groundfish resources of the Gulf of Alaska as projected for 1997. North Pacific Fishery Management Council,
Anchorage, Alaska.
NPFMC (North Pacific Fishery Management Council). 1996b. Stock assessment and fishery evaluation report for the
groundfish resources of the Bering Sea and Aleutian Islands management area as projected for 1997. North Pacific
Fishery Management Council, Anchorage, Alaska.
NPFMC (North Pacific Fishery Management Council). 1996c. Fishery management plan for the commercial king
and Tanner crab fisheries of the Bering Sea/Aleutians. North Pacific Fishery Management Council, 605 W. 4th
Avenue, Suite 306, Anchorage, AK 99501-2252. 39 pp.
NPFMC (North Pacific Fishery Management Council). 1996d. Fishery management plan for the scallop fishery off
Alaska. North Pacific Fishery Management Council, Anchorage, Alaska.
NPFMC (North Pacific Fishery Management Council). 1997. Fishery management plan for the groundfish fishery of
the Bering Sea and Aleutian Islands area. North Pacific Fishery Management Council, Anchorage, Alaska.
NRC (National Research Council). 1990. Decline of the sea turtles; causes and prevention. National Academy Press,
Washington, D.C. 259 pp.
Otto, Robert S. 1989. An overview of eastern Bering Sea king and Tanner crab fisheries. Proceedings of the Interna-
tional Symposium on King and Tanner Crabs. November 1989, Anchorage, Alaska. Alaska Sea Grant College
Program AK-SG— 90— 04. University of Alaska, Fairbanks.
Overholtz,WJ., and A.V.Tyler. 1985. Long-term responses of the demersal fish assemblages of Georges Bank. Fishery
Bulletin 83(4):507-520.
Pearse, PH., and C.J.Walters. 1992. Harvesting regulation under quota management systems for ocean fisheries: Deci-
sion-making in the face of natural variability, weak information, risks and conflicting incentives. Marine Policy
16:167-182.
Pereyra, W.T. 1996. Midwater trawls and the Alaskan pollock fishery: a management perspective. In Solving Bycatch:
Considerations for Today and Tomorrow. Alaska Sea Grant College Program Report No. 96-03, University of
Alaska, Fairbanks.
Pikitch, E., D. Erickson,J. A. Perez-Comas., and P. Suuronen. 1996. Codend size-selection: good concept, but does it
really work? In Solving Bycatch: Considerations for Today and Tomorrow. Alaska Sea Grant College Program
Report No. 96—03, University of Alaska, Fairbanks.
Rago, P.,W Gabriel, and M. Lambert. 1993 .Assessment of yellowtail flounder, Pleuronectes femtgineus. Northeast
Fisheries Science Center Research Document 94— 02. Woods Hole, Massachusetts.
Rago, P., K. Sosebee,J. Brodziak, and E.Anderson. 1994. Distribution and dynamics of North Atlantic spiny dogfish
(Squalus acanthias). Northeast Fisheries Science Center, Center Research Document 94-19, Woods Hole, Massachusetts.
Richards, R.A., and L. Hendrickson. Manuscript. Effectiveness of the Nordmore grate in reducing bycatch in the
northern shrimp fishery. Northeast Fisheries Science Center, Woods Hole, Massachusetts.
Rose, OS. 1996. Behavior of North Pacific groundfish encountering trawls: Applications to reduce bycatch. In Solving
Bycatch: Considerations for Today and Tomorrow. Alaska Sea Grant College Program Report No. 96-03. Univer-
sity of Alaska, Fairbanks.
Saila, S. 1983. Importance and assessment of discards in commercial fisheries. United Nations Food and Agriculture
Organization, Rome, Italy. FAO Circ. 765. 62 pp.
Shepherd, G.R., and M.Terceiro. 1994. The summer flounder, scup and black sea bass fishery of the Middle Atlantic
Bight and southern New England. NOAA Technical Report NMFS-122, National Marine Fisheries Service,
Woods Hole, Massachusetts.
Smith, W.T. 1996. Reduction of halibut bycatch and associated mortality in the Bering Sea cod fishery. In Solving
Bycatch: Considerations for Today and Tomorrow. Alaska Sea Grant College Program Report No. 96—03, Univer-
sity of Alaska, Fairbanks.
Squires, D., and J. Kirkley 1991. Production quota in multiproduct Pacific fisheries. Journal of Environmental Economics
and Management 21:109-126.
Stevens, B. 1996. Crab bycatch in pot fisheries: Causes and solutions. In Solving Bycatch: Considerations for Today and
Tomorrow. Alaska Sea Grant College Program Report No. 96—03, University of Alaska, Fairbanks.
Stone, M., and C.G. Bublitz. 1996. Cod trawl separator panel: potential for reducing halibut bycatch. In Solving
Bycatch: Considerations for Today and Tomorrow. Alaska Sea Grant College Program Report No. 96-03, Univer-
sity of Alaska, Fairbanks.
UN (United Nations). 1995. Agreement for the implementation of the provisions of the United Nations Convention
on the Law of the Sea of 10 December 1982 relating to the conservation and management of straddling fish stocks
and highly migratory fish stocks. United Nations General Assembly, Sixth Session, July— August 1995.
United States Fish and Wildlife Service (USFWS). 1997. February 19, 1997, amendment to the biological opinion on
reasonable and prudent measures to mitigate potentially adverse impacts of the Alaska groundfish fisheries on
shortailed albatross.
VanVoorhees, D.A.,J.F Witzig, M.F. Osborn, M.C. Holliday and RJ. Essig. 1992. Marine recreational fishery statistics
survey, Atlantic and Gulf Coasts, 1990-1991. National Marine Fisheries Service, Current Fishery Statistics 9204,
Silver Spring, Maryland. 275 pp.
Warren, B., ed. 1994. Win-win bycatch solutions: A handbook for collaboration. National Fisheries Conservation
Center, Seattle, Washington. 112 pp.
Warren, B., and D. L. Alverson. 1995. Interim draft report on recommendations for a national bycatch strategy. Seattle,
Washington. 22 pp.
Wyman, E. 1996. Selective groundfish pots offer solutions to bycatch problems. In Solving Bycatch: Considerations for
Today and Tomorrow. Alaska Sea Grant College Program Report No. 96-03, University of Alaska, Fairbanks.
Wynne, K.M., and M.M. Merklein. 1996. Marine mammal observer program design considerations: A survey of eight
Alaskan small-boat fisheries. Alaska Sea Grant Marine Advisory Program. University of Alaska, Fairbanks.
Resource,
Material
glossary
Allowable biological catch (ABC) — The maximum allowable catch for a species or species
group for a particular fishing year. It is set each year by a scientific group created by the management
agency. It is developed by reducing the maximum optimum yield as necessary, based on stock
assessments. The agency then takes the ABC estimate and sets the annual total allowable catch.
Bycatch reduction device (BRD) — Any of a number of implements that have been certified to reduce
the likelihood of capturing nontarget species.
Catch per unit effort (CPUE) — The amount offish that is caught by a given amount of effort. Typi-
cally, effort is a combination of gear type, gear size, and length of time gear is used.
Category I, II, and HI fisheries — Categories of commercial fisheries under the Marine Mammal Pro-
tection Act:
Category I — A commercial fishery with frequent incidental mortality and serious injuries of marine
mammals. A Category I fishery is by itself responsible for the annual removal of 50% or more of any
stock's potential biological removals (PBRs).
Category II — A commercial fishery with occasional incidental mortality and serious injury of ma-
rine mammals. Collectively with other fisheries, a Category II fishery is responsible for the annual
removal of more than 10% of any marine mammal stock's PBR. By itself is responsible for the
annual removal of between 1% and 50%, exclusive of any stock's PBR.
Category III — A commercial fishery that has a remote likelihood of, or no known, incidental mor-
tality and serious injury of marine mammals. Collectively with other fisheries, a Category III fishery
is responsible for the annual removal of 10% or less of any marine mammal stock's PBR, or more
than 10% of any marine mammal stock's PBR. By itself it is responsible for the annual removal of
1% or less of that stock's PBR.
Conservation engineering — The practice of determining the modification in gear design that will
meet conservation objectives, such as decreasing bycatch and bycatch mortality by increasing the
selectivity of gear and increasing the survival of fish and other living marine resources that fishing
gear encounters inadvertently.
Demersal — Fish and animals that live near the bottom of an ocean.
Derby fishery — Generally, a fishery operated under conditions where each vessel has an incentive to
catch the greatest number offish in the least amount of time.
Endangered species— A species is considered "endangered" if it is in danger of extinction throughout
a significant portion of its range; it is considered "threatened" if it is likely to become an endangered species.
Ex-vessel value — The amount paid to a vessel's owner or operator for its catch, excluding any value
added by at-sea processing.
Exclusive economic zone (EEZ) — The zone contiguous to the territorial sea of the United States, the
inner boundary of which is a line coterminous with the seaward boundary of each of the coastal
states and the outer boundary of which is a line drawn in such a manner that each point on it is 200
nautical miles from the baseline from which the territorial sea is measured.
Fishery management plan — A plan developed by a regional fishery management council, or the Sec-
retary of Commerce under certain circumstances, to manage a fishery resource in the U.S. EEZ
pursuant to the Magnuson— Stevens Fishery Conservation and Management Act. It includes data,
analyses, and management measures for a fishery.
Ghost fishing — The capture offish or other living marine resources by lost or discarded fishing gear.
Groundfish — A species offish, usually finfish, that lives on or near the sea bottom part of the time.
Haulback — The period in fishing operations during which the gear is hauled from the water back
onto the fishing vessel.
Individual fishing quota (IFQ) — A federal permit under a limited-access system to harvest a quantity
offish, expressed by a unit or units representing a percentage of the total allowable catch of a fishery
that may be received or held for exclusive use by a person.
Level of utilization — A comparison of existing fishing effort with that required to achieve long-term
potential yield (LTPY).
Overutilized — Fishing effort is in excess of that needed to achieve long-term potential yield.
Fully utilized — Fishing effort is at a level that will support the achievement of long-term potential
yield.
Underutilized — Fishing effort is below the level at which long-term potential yield will be achieved.
Limited entry — A program that restricts the persons or vessels that may participate in a fishery. Li-
cense limitation and individual fishing quota programs are two examples of limited entry.
Long-term potential yield — The maximum long-term average catch that can be achieved from the
resource.
MARFIN (Marine Fisheries Initiative) — A program that brings together scientific, technical, industry,
resource conservation, and management talents to conduct cooperative programs to facilitate and
enhance the management of the marine fishery resources of the Gulf of Mexico and South Atlantic.
Maximum sustainable yield — The largest average catch that can be taken from a stock under existing
environmental conditions.
Metric ton — 2204.6 pounds.
National standards — A set of 10 conservation and management standards included in the Magnuson-
Stevens Fishery Conservation and Management Act. Each fishery management plan must be consis-
tent with all 10 national standards. National Standard 9 requires that conservation and management
Resource,
Material
Si
measures shall, to the extent practicable (1) minimize by catch and (2) to the extent that bycatch
cannot be avoided, minimize the mortality of such bycatch.
Open-access fishery — A fishery in which anyone may participate at any time.
Optimum yield (OY) — The amount offish (1) that will provide the greatest overall benefit to the
United States, with particular reference to food production and recreational opportunities; and (2)
that is prescribed as such on the basis of maximum sustainable yield from such fishery as modified by
any relevant ecosystem or social and economic factors.
Pelagic species — Fish and animals that live in the open sea.
Potential biological removal (PBR) — The maximum number of animals, not including natural mortali-
ties, that may be removed from a marine mammal stock while allowing that stock to reach or
maintain its optimum sustainable population. The PBR level is the product of the following factors:
(1) the minimum population estimate of the stock, (2) one-half the maximum theoretical or esti-
mated net productivity rate of the stock at a small population size, and (3) a recovery factor of
between 0.1 and 1.0.
Protected species — Living marine resources protected under the Marine Mammal Protection Act,
Endangered Species Act, or Migratory Bird Treaty Act.
Quota — The maximum amount offish that may be legally landed in a time period. It can apply to
the entire fishery, to an individual fisherman's share under an individual fishing quota (IFQ) system,
or to the size offish.
Recreational fishery — Harvesting fish for personal use, fun, and challenge.
Round (live) weight — The weight offish, shellfish, or other aquatic plants and animals as taken from
the water; the complete or full weight as caught.
Saltonstall— Kennedy (S—K) grant program — A competitive program that provides funds through grants
or cooperative agreements for research and development projects to benefit the U.S. fishing indus-
try. The S-K Act, as amended [15 U.S.C. 713 (c) (3)] is the program's statutory authority.
Stakeholder — One who is expected to receive economic or social benefits from the conservation and
management of living marine resources.
Stock assessment — The biological assessment of the status of the resources. This analysis provides the
official estimates of stock size, spawning stock size, fishing mortalities, recruitment, and other param-
eters.
Strategic stocks — Marine mammal stocks that have a level of human-caused mortality likely to reduce
or keep the stock below its optimum sustainable population (e.g., short-finned pilot whale, Baird's
beaked whale, mesoplodont beaked whales, Cuvier's beaked whales, pygmy sperm whale, sperm
whale, and humpback whale).
glossary
Take-reduction plan — Plans to assist in the recovery or prevent the depletion of strategic marine
mammal stocks by outlining strategies for reducing the number of marine mammals incidentally
taken in the course of commercial fishing operations.
Take-reduction teams — Established by the 1994 amendments to the Marine Mammal Protection Act,
these teams are made up of individuals who represent the span of interests affected by the strategies
to reduce marine mammal takes, including commercial and recreational fishing industries, fishery
management councils, interstate commissions, academic and scientific organizations, environmental
groups, Native Alaskans or other Native American interests if appropriate, and NMFS representa-
tives.
Threatened species — A species is considered "threatened" if it is likely to become an endangered
species; it is considered "endangered" if it is in danger of extinction throughout a significant portion
of its range.
Turtle excluder device (TED) — An implement that has been certified to reduce the likelihood of
capturing turtles.
appendices
Matrix
The National Marine Fisheries Service's Bycatch Team, consisting of representatives from the
five science centers/regions and headquarters, conducted a survey of available information on bycatch
and discards in the nation's fisheries, and efforts to understand and manage the issue. Throughout
the assessment analyses were conducted only on the discard component of bycatch; information on
the unobserved mortality component of bycatch is not quantified in most fisheries. It is intended
that this type of assessment will be updated regularly as new information on bycatch becomes
available and as data collection programs are expanded to include other sources of bycatch mortality.
The survey is intended to represent the latest (1996) information available for each fishery,
defined by gear type and target species or species group. Estimates of discards are presented, where
available, as well as landings levels in each fishery, and associated descriptive data. In many cases,
assessments of variables described below are based upon subjective judgments by knowledgeable
regional representatives of the National Marine Fisheries Service. The intent of this survey is to
update information previously compiled elsewhere, and to serve as a benchmark from which to
judge future efforts in data collection and management efforts to mitigate negative effects of bycatch.
Regional matrices consisting of 28 variables assessed for each defined fishery are presented in this
appendix for six regions: Northeast (Maine— North Carolina), Atlantic Highly Migratory Pelagic
Species (large pelagic species); Southeast (North Carolina— Texas, and Caribbean); Western Pacific
and Pelagics (Western Pacific); Pacific Coastal (Washington— California); and Alaska.
Major Categories for Fisheries Groups
Eleven major categories were addressed for each fishery group. Generally, within regions fish-
eries were grouped after the groupings in NMFS' Our Living Oceans (NMFS 1996a). Within these
major groupings, fisheries were broken down by gear type. Target species and discard species, and
the fishery management plans that manage them, are identified. Additionally, the matrix identifies
each fishery's categorization under the Marine Mammal Protection Act. The volume of fish cap-
tured (metric tons), the value of the fishery (millions of dollars), and the number of vessels or permit
holders are identified where available for target and discard species. The status of the stock is de-
scribed by two factors — the stock's level of utilization and the stock's status relative to its long-term
potential yield. These evaluations were based upon the 1995 Our Living Oceans publication (NMFS
1996a). Significance of the discards is a qualitative identification of the nature of the concern about
bycatch (population, ecosystem, socio-economic) and the level of significance or seriousness of that
concern. The matrix also identifies reasons for discards (regulatory, discretionary, or prohibited species).
Bycatch
Matrix.
Matrix category
Information contained in category
0L0 Fishery
MM PA Cat.
Fishery category as identified in Our Living Oceans (NMFS 1996a)
Fishery category under the Marine Mammal Protection Act
Gear
Primary gear used to capture target species
Retained species
Fish retained in the fishery
Discarded species
Fish discarded in the fishery
FMP (or other)
The management plan that applies to the retained and discarded species
Volume
Value
The volume, in metric tons, offish captured (retained and discard species) in a fishery.
Figures are for 1995 except where noted.
The ex-vessel value of the fishery
# Vessels
The number of vessels or permit holders in the fishery
Utilization
Stock Size
The level of utilization of the fishery resource [based upon Our Living Oceans (NMFS
1996a)]
The size of the stock relative to its long-term potential yield [based upon Our Living
Oceans (NMFS 1996a)]
Level of Concern
The level of concern about the particular bycatch problems of that fishery
Nature of Concern
The nature of concern about bycatch in that fishery [population (p), socio-economic
($), or ecosystem (e)]
Reasons for Discards
The reasons that fish are discarded in that fishery [regulatory (REG); discretionary/
economic/personal considerations (DIS); or prohibited species (PS)]
Seven Steps to Addressing Discards
Based upon the quantitative and qualitative information gathered in the first 1 1 major catego-
ries, each fishery was evaluated under "seven steps to addressing discards."The steps focus on deter-
mining the status of information on the amount and type of discards (Step 1); assessing the current
state of knowledge about the population, ecosystem, and socio-economic impacts of discards (Step
2); evaluating the effectiveness of current bycatch management measures (Step 3); identifying how
extensively alternative management measures have been considered (Step 4); determining whether
the population, ecosystem, or socio-economic effects of those management measures have been
identified (Step 5); determining whether alternative management measures have been implemented
(Step 6); and assessing the capacity of the fishery to monitor the effectiveness of new bycatch
management programs (Step 7). The criteria that determined each score in the seven steps are as
follows.
Step 1: Information on Magnitude of Discards
The quality of discard information was evaluated for each fishery using a 4-point scale where
= no information available; 1 = unverified harvester or incidental observer reports; 2 = isolated
snapshots from observer programs, 3 = estimates of discards possible with limitations on precision
and accuracy; and 4 = estimates available with adequate precision and accuracy.
Step 2: Impact Analyses of Discards
The current status of impact analyses of discards was evaluated for populations, socio-eco-
nomic considerations and ecosystems. Available impact analyses were scored as = no evaluation
made; 1 = qualitative information about impacts; 2 = some quantitative information mixed with
qualitative information; 3 = quantitative information with limitations on precision and accuracy;
and 4 = information on impacts with adequate precision and accuracy
Step 3: Effectiveness of Current Measures
This step evaluated the adequacy of current measures by = current measures inadequate,
identification of alternative management measures needed; x = no discard problem exists; and * =
existing measures adequate to manage the fishery.
Step 4: Identification of Potential Alternatives
Progress in identification of bycatch management alternatives was evaluated as = no alterna-
tives have been identified; 1 = factors affecting discard rates and mortality have been identified; 2 =
input of constituency groups solicited; 3 = management measures have been identified; and 4 =
practicality of proposed alternatives has been assessed in terms of industry acceptability and council
policy.
Step 5: Evaluation of Impacts of Bycatch Mitigation Alternatives
Impact analyses for mitigation alternatives were scored as = no evaluation made; 1 = quali-
tative information about impacts; 2 = some quantitative information mixed with qualitative infor-
mation; 3 = quantitative information with limitations on precision and accuracy; and 4 = informa-
tion on impacts with adequate precision and accuracy.
Step 6: Implementation of Alternative Management Measures
The implementation of alternative measures was quantified by = none of the following; 1 =
Is there a fishery management plan or regulatory amendment for discard regulation?; 2 = Is there a
technology transfer program (if applicable)?; 3 = Is there a discard reduction incentive program (if
applicable)?; 4 = Is enforcement ready to go?; and 5 = Is a monitoring system ready to go?
Step 7: The Adequacy of Monitoring Programs
The adequacy of monitoring programs for evaluating the effectiveness of selected and imple-
mented management efforts was described by = no capacity to monitor and evaluate effectiveness
of implemented measures; 1 = demonstrated commitment to program; 2 = adequate pre- and
postmitigation monitoring; 3 = monitoring sufficiently accurate and precise to quantify effects of
mitigation measures; and 4 = effective communication program to ensure user-group buy-in to
program.
-
J0)IU0|^
=
-
o
o
c
o ©
o <= o c
o o c o
— — o ©
°
© —
© <-.
© ©
— o
■*
©
=
*
©
-r — —
-
vO
)U9U19|dlU|
r,
5
v
£
• 1
&
m
3 E
' c
u Z
(•!
■^
t-i
o
04
"•
«v| rs
~ ^ © =
r»>
© r^
rsi ©
■*
©
oj
*
N
^ ^ -
M
u
1
<
©
"'""
©
T
n
T r± —
M
13
<
•»
*•> U]KUJ.1J| y [BI1U310J
m
©
r* r* — —
"*
© —
© ©
*
"
©
"*
©
T © ©
w
Q,
c
m
S3jrm3(\ iu.ujn_)
o
°
o
o
©
=
= =
©CO©
=
© ©
© ©
_ o
=
©
©
o
=
©
= = =
6
„
o
1
3 S
r*
3 1
i/i
o
© © © ©
© ©
— ©
©
"
=
, = =
°
e.
-
-
-
©
r*t
— — © ©
©
T
N
-
ipiBiA'g jo 9pniiu3B[^
-
1 -3
s a
i/i
Q«
C/3
H9
uo
in
O u
a ■=
5
5
a
Q
5
5
1°
a
o
1/5
O
O
o d
O'gMW
M
c/> o
° n
d
in
O* vi
d
£
s
5
a
a
a a
2 2 o 5
5
5 a
a 5
a.
a-
a
in
a.
5
£ a 5
a
£
m
Ti
■g
Si
"8
s
"8
•Si o
"S "S *S m
s
oO "S
o °°
BO
"?
?
•=
"to
■5, * s
■a
m
a
c
9
5
2
***
s
^
^
s
X S
S S 2 i
x S
-3 x
X
k;
s
X
U
a.
X
x i 3
s
3
M
1
z
a.
Q.
o\
Ch
"5 -|
""
"■
~
—
8 1
> S
% D.
s
—,
l~4
It
^c
M
%c
©
o*
c
M
©
o
fM
t
i
V
JS
s
iZ
*•
>
o
©
©
© ON
©
<
© *S
<
1
o
©
©
©
© V.
i
o
1-. ^
z
r- c7*
z
1 p
o
«
©*
©*
cm' — '
s
OK
^j
r-;
*3 _
>
^
|
S
e
<
a.
b
<
a.
S
f-1
u
< "-
> in
0.
UJ
UJ
UJ
UJ
T J
5 [Z 2 < ^
U
3 3 uj
S
1
UJ UJ
2
<
E-IM1
E-l
E-5 A
3 LU
Z
ii
Z
z
e
3
Z Z
z z z z z
Z
i/) oo Z
Z Z Z
Z
s
1
s
Z Z
z
I
Z
z z z
(« z
|
^
ts
o
£ |
,s
*w
"5 vz
s.
ir
g.
a £ it
a
-a
■9 i- -S _
- "= -rT
m
do'
1
1=
aq
o
-a
5
"c "1
groun
lobsic
dogfi;
skates
=
I!
15
1 1
1
u
|
00
o
5 c o
1 = a
s a e
g
3
s
I
1
K
1
8
5 £
C
■
£
J=
-a
8
■
"o *3
W C
u =
c
3
s
o
■o
i
5
1
s
B
s
1-
Z
3
O
CO
O
CO
CA
^
-
'u
u
■<
U3UJ3|dlU|
1
u
«
£ ■
£
in
II
M
*
- -
-
—
rM ^
r4
"
^ M ^ ^
»N r-i
~
"■
—
~
"~
M
c
T
i
<
■<
" t
»Aii>ujai|v |B")u.->i
»jnna|V iu-ujiij
O
© o
°
=
O O
=
C
© = © C
a ©
°
=
°
=
=
g
H
,.
o o
o
o
_ -
O
n
i^i © © ©
_ _
©
©
a
o
o
I
n
w —
v^
*
© o
o
a
=
„
f-, © o ©
— ©
©
o
©
=
=
ft
"
o ©
o
-
9
-
m r^ © ©
t/i
©
t^
-
-
-
-
q ji-oa'q jo 3pn)!u3«|^[
ji
5
5
5
Q
5
5
a
o a
d
o d
d
O
o w o"
o a"
d
o"
&a
VI
to
CO
Ou
i
a a
J j
5
"8
2
a
5
a a
a
a
■8
Mod RE
Mod PS
High DI!
Mod RE
x 2
2
5
J
5
J
5
J
1
3
£
©
§
!
3
0J
(A (A
o
(A «•!
«■>
«* k! «„ *»
*^ ,j "
(A
o
u
i
z
e£
d. 6.
**
a a.
e£
o. d m a.
d. a.
(£
a.
Sg
e
"Z >
«:
j§ o o
i i
< <
^ 1
< CD
5
o S
_o J
5
3
s ^ $ s
CO* z < cS
_o _5 _S
S o
s s
< CO
O
< CO
g
z
CQ fu CQ
£
00* CO
CO CQ
£
§
co ca m
^
Jf
1-4
> v
s
• &
©
C
0^
^
\A
t-
^>J
p
1
.9
s
E
1
>
^
s
1
8
-
e
s
o
©
i
5 as
8 S
©
o
s
8
8
y
z
&
o\
ii
~-
*
ri
"'
«o
"*
^ —
K
■a
> "
A
|
<
o
7
u
u.
£
s
2
2 v.
•
a.
2
2
«9
<
<
T
T
t
T -■ ^. T
1 T T
7 7
T T
o o
b. u.
S 2
to &o
_
< S
^ u.
*? 2
UJ
ui iii ui
3: ^
UJ UJ
LU
ai uj
LU LU
<
<
<
< S S <
< < <
< <
< <
<
LU (/>
m
z
1
&
8.
z z z
Z Z
z z
z
Z Z
■a
e
c Or
z z
2
2
o
2
f flounder M
i N
Hsh M
2 2 2
t3
2 2
2 Z
1
< <
'E
2
1
z <
30
1
1
■
|
1
1 1
00
e
3
e
1 |
1
E
§
3
I 111
1 1 1 1
■a
a
a. 3
3 p
3
c
3
o
|
'e
C
s
1
13
e
1
e
1
B
C
31
C
of
e
!
c
IS
e
c
8
c
■S
■
1
9
■n
m
s
|
i
£
5
1
o
■a
1
.5
E
e
E
E
3
E
E
3
a.
3
z
b
E
|
<
l
<
2
1
_J
_1
CD
X
o
o
CO
■9
0>
3
'C
o
->
J
S
© ©
©
©
©
© ©
©
s.
■*»
KajmreaiM tuajjn^
© ©
°
•
© ©
© © ©
© ©
©
«•
o
©
o ©
©
e
-^
© c
©
_
-
o
o
©
©
© ©
©
i
5 3
M
©
o
©
© ©
©
c
© —
©
o
©
©
CO
Q
© ©
©
CO
5
-
upinj vg jo .ipnjiuSHiAj
1
1°
(O
VI
co O
W ryj O
„ [/ .
a
a
O
CO
b
a £
EO
o_
a
a s
5 5 £
5 5
S3
a
tj0
a
5 2
i
£
5
a*
o
1
I
11
O (J
a. a.
■8
y
a-
■8
s
J J
S Mod
S Mod
p.$ Mod
"3 "S
J
J
1
u
0-
Q.
■g
2
>
j J
■g
2
M
t
z
ea
e
||
I % *
1 a z
<
3 3
5 .2
S S 3°
3 8 S
_2 3
3
<
S
o
i
<
o
o « o S ^
3
s
a
■*
Is
Z
z z
< m
Z Z Z £
z z z
CD CO
z
s
s
£
a z s s z
z
£
z
CO
V)
*-
3
■
J)
_©
J
5
7 Under
Over
N/A
<
Z
1 1
D ^
i£ i£ 5 6
Under
Under
FulK
1 1
1
e
D
<
z
<5
5
c
6
FulK
Fully
Over
Over
N/A
3
&
6
"3
r-
BD
r*i
n
11
c E
> 5
* £
s
„_
IN
©
-»
o
■*
■>
CO
C
©
©
£■
i
t>
£
y
3
E
>
S
©
©
©
©
©
©
©
i
© o ©
5 © ©
5
M
O
°*
*t
E C
n
N
N
u _
z
< U
Cj
U
<
u
< u
(j U O U ^
u
o u
6
0-
1
MA-I
ASMF
MMP/
2
2
2
MA-1
s ASMF
HMS
MA-1
MA-I
-MA-I
MA-4
MA-1
MA-I
MA-I
s <
2
on
<
<
a.
2
2
2
00
<
i
S <
ASMF
ASMF
ASMF
ASMF
MMP/
I
<
u- u.
2 2
< <
u-
2
OO
<
.s
1
Bfl
•H
-o
3
3
00
'h
CO
X
■
ickerel
river henin
pitol whale
E
a.
o
-a
\ |
"S "2
X o
E £
d
Loiigo squi
butterfish
groundfish
j 1
1
-b
l
1
|
1
1
•5
o
If
3 |
C E
8,
1
E
i
•fl
E
^
a
_
1
E
a
a
a
1
'a
2
'5
a
I
|
i
|
■c
1
!
|
2 '£ 2
I
•c
5
<
x
z
<
o
—
-J
i
<
z
3
8
h
j
5
O
9
o
B
i
O
2
O
o
C/l
j
-
'C
a
<
A
r
.
„
--,
-r a
NO
r.
N
o* ©
-
o
z
-
JO|IUO|^
o
o
o
o
©
o
:
©
©
=
©
©
•o
lU3IU3fdlU|
g
„
•!
a \
3
in
| E
^
©
ni
m
ft
m
r*J
~
IN
f
2
<
c
-
n
n
rl
„
-
-
-
^
«
«
*
UAI|VUJ3||V |»I1U.H0J
o
O
O
o
=
©
©
©
1
»n
!3jnfW3|^ m.ujn }
o
o
O
o
.
-
e
©
.
.
• O
8
s
If
o
^
n
© ©
1
-
3 6
1 1
u —
c
©
-
-
-
-
-
-
-
-
*
© ©
© —
-
ipl« j\g jo ->pniiu3nAj
If
8 %
V)
j_
o
o
d
O
O
1
a
a
a
5
5
a
a
a
£
a s
£
1
s
X
J
j
j
be
J
J
3
J
J J
1
1
1
i=
5
Q.
i
Z
Q.
ft
e
SI
o
%
5
s
%
6
s s
_S _S
S
s
S o
s
Above
N/A
1
£
1
&
£
&
£
z z
co en
3
£
z ^
3
(A
1
3
1
h
fe
*■
h
.
. .
.
.
i-
. .
|
U_
1
1
6
6
6
"5 "5
6 6
<5
6
"3 "3
Li- U.
6
Ill
9
^
s
"
^
©
1
1!
©'
> y
* £
|
_
fS
O
M
r-
©
w
ci
8
00
c-
s
fM
J!
.9
s
E
-a
>
I
S
s
_
|
s
Is
~
i
>o
©
>• *"
~
I
<
U
u.
A
S
< <
*.
u.
u. u_
u.
|
2
2
3
5
7
<
•i <
UJ O
UJ
<
s -
is> UJ
uj
t-rft
U Ul ^
<
< <
z
z
s
s s
Z !•!
Z
3
-< z
Z
Z 2 2
V
•ft
T3
-o
s
e
c
o.
'3i
3
■6
m
.3
&
B
(5
3
|
1 6
J a
o
a
E
1
1
1|
7, c
C
E
E
(0
Q
1
-o
c
1
11
» E
1
1,
E
uf
M
|
i
"3
|
"5
2
£
m
■
•c
5
c
*
m
SI
Q
3
a
Q
o
CO
c
2
|
0!
a
a>
1
'E
<
a»
=
=
=
E
s
'■Ti
1
I
tl
£•
1
+*
1
1
CA
li
t
H
s
o
■
a>
J=
Z
r
j.\ iu>->jjn )
© © ©
o © — — —
©
u
1
Is
W
3 2
M
ri m tn
©
<"M C O O O
— © © ©
M IN (N ft*
— — © o ©
o* rs| — cm r4
f\ t*i t*\ t*\ r*\
H,
-
ipiBM'g JO opn jiu3bjaj
•2-S
11
faO
5
£°
O O O
c/j
O c/>
O V)
o o „
a
saa
5
H £ vi ) >
o: Q o_ a. a.
a 5 £ 2
S S 5 £ £
a
■C
a
1
a*
CD
5
S _M op
"8
"5 - -= - •=
O S M S oo
s * s i
s 6 o oo oo
,
e
y
I
E
E = * S M £
_o e _o _o
2 E E js '£
_o
e
i=
3
U
o.
o
K
I
z
5>
°: O O O
O. Q.
o. •* aid.
D. M M
k!
fa* Q. O. Q.
i/ D. H £
EL
W Q. 6.
Q. o. o. Q.
vf d. m a. cl
M
e
<- o-
1T h
o
S o J2
o
o
^ ^
|
1
£
u
.. .. 2 w CB «
5
u 2 u
CS L.
n i_ k. u i_ cd
o ,5 o
U9
e
3
5
1
2
z
- - s 3 s o
z z i2 z i2 £
3
z
| < j j j
tfi z z z z
ziSz
3 S
fa. g ^ < <
£ z z z z
LuZZZZiZZZZ
5 .2 = 5
e
to
P
o
5
u U C l_
" < < < <
y = ^ ^ ^
M -a ss = = o<<<
6
3
Li-
S
u. u. O u- O O
3
U.
6 z z z z
u. <5 u.
3 3
U. Lt_
6 lS z z z
OU.U.L1.U.OZZZ
6 6 5 6
so
11
•5 -
— .
2
i
> o
D.
£
* £.
Q.
t
C"
1
*
=
«N rs|
^
£•
1
s
«
£
>
^
^ ,
.N ^ « » r-
^
r- ~ ~ ~
^ r- *0 1^- V «ft
u
E
^
^
TS
>
-^
|
Cs|
CM
IN
B
«J
4
4
©
,
"7
" <
<
7 7 7 7 <
7
ft.
CO
00
CO
<" ^ -• S:
(/I
a.
co to to to j, a
to
s
2
£
s
S < < 2
s
to s
s s s s 2 s
X X X X lu z
S
E
I
X
X
X a 3 Z
X
u S
X
.s
I
ra
_
09
c _
2 "S
l l * l
| ! |
■5 -a 1
. 1 -5 s 1
i .1 a i
^
'C
B
s § i
ass
a
billfi
mahi
tunic
birds
mam
s s J i
2 Z - E
!b
j
1
1
-a
I
Efl
'B
■g
c
e
£
■
c
C
%
C
bigeye tuna
albacore tuna
blucfin tuna
yellow fin tuna
blucfin tuna
albacore
■
_ C
S 2
S _f
it
■
e
3
c
IS
i
1 i e
f * s s
i i £ 1
Elf!
blue mirlin
white martin
sailfish
-2
I
-i
u
z
u
Cb
E
a
£
2
a
a.
<*-
— J --
u
3
<
_
s
j
_
a
ft.
1
1
=
•a
c
t
.>.
ea
|
1
E
s
o
M
t «
a t
ill
'•to*
C
1*1
w
Os © *^
■»—
*
<
o
o
o
o
o
©
©
_
J01IUOJAJ
o
o
o
o
o
o
e
o
e
©
©
©
© ©
o o
«
)U3UU|(IU1I
B
w
•g
•J "3
3
w
1!
ko
o
o
o
o
©
©
©
© ©
OB
> t
M
C
U S
Q.
p
o
o
©
o
©
© ©
<
1
■<
e
"*
E3AI1>UJ31|y |«I)U3)0 J
o
o
o
o
o
o
o
©
©
©
©
©
© ©
© ©
»•»
E
fojnsou iu;ujn;)
-j
1
r*
Si
1 E
to) ~
iO
c
n
FM
«
~
fi
~
*N
m
-
ipujxg jo dpniiu3i|^
fc
If
i s
to
(A
tsi
a •*
a
a
a
S a
CO
o
CO
o
l/J
o
5
a
a.
5
^
a.
5
a £
je
a
7)
St
1
"mi
■s
■5
"5)
■g
■s
.
■g "g
©
8
I
s
1
X
2
s
5
s s
S
e
c
n
1
Z
CL
Sd
a
u
a.
k!
M
Eh
Iff*
*
a.
w
d
M
-* Q.
e
|
a
S
i
s
i
e
3
Cfl
5
1
s
a
c
<
z
o
1
3
C
<
Z
|
4
I i
3
e
e
H
|
u
5
5
s
1
^
b
£
g
-
_
i <
o
o
3
z
o
,3
z
«H
a z
II
s
-
©
©
vt
I
1
S
■a
>
ON
§
5
r*
g
9
"5
>
1
<
<
<
D-
Cl.
a.
o
i
f
<
t
<
2
S
*
*
S
6
n
in
to
to
1/3 *+
1
s
i
s
<
s
s
<
s
s
s 5
E
I
i
o
UJ
X
o
UJ
X
X
a uj
I
|
■s
1 ^
■a
9 5
•o
1 i
09
'C
1 1
1 * *
O co UJ
"8 „
a. to
.3
E 3
5.=
O o
| g. "1
O CO UJ
1 S
1 &
£ to
"i 1
O E So S »
h | S 8 S
i s 1 2 s.
O to uj £ to
I
■1
a
a
§
a
3
a
o
11
s
-1 CO
a
u
E x
ii
E 2
J3
"3
5
-J
-1
m
z
a
8
CD
lj«
"a
O
<
1
I
=
=
=
'U
B
t
|
«
1
'-5
a
E
3
O
1
<
jt
f.
3
*
•
*
3
-
JO)jUO[^
o
o
©
©
©
©
©
©
©
©
©
©
©
©
©
o
©
©
©
©
© ©
o ©
o ©
IB
)U3U»|dUl|
ii
JB
_>
S *■
m
| E
*"
©
©
©
©
©
©
o
© -
>•
x &
m
.>; t
M
u S
B
<
&
©
©
©
©
©
o
O
© ©
i
^
O
s,
"*
taAi)iiu3i|v |»r|i»jo j
©
©
©
©
©
©
©
©
o
©
©
o
©
©
© ©
© ©
«*»
c
Kunnrei^ luajjnj
u
E
u —
c
~
-
~
-
-
-
fi
M
-
H j)B3A"g jo apn)iu3i|^
!*
= m
£2
VI
BQ
© w
ji
5
5
5
5
(J
a
a.
to
5
o
a
£
a
o
M
a
"B
P
p
m
5
■ t
t
■g
c
■g
00
1
"S "5
S
3
2
5
■3
5
X
"=>
S it
5
K
IS
3
I
Z
B.
u
a.
M
o.
a
M
Sd
v*
M
a.
w
Ck
m tL
e
i
55
©
a
■ji ST
i
5
1
I
<
z
o
E
P.
c
3
o
1
c
5
Z
P
O
|
s
1 ^
e Z
e
o
03
.a
l_
P
P
^
p
.
o
a
a
1
1
<
z
1
e
a
1
,3
<
Z
1
i <
e z
1 ^
> V
©
* i
©
G"
c-'
*■*
5
fr
J
¥
-
&
>
©
c
n
B
E
a
^
>
■£■
|
<
<
<
a.
a.
a.
la
2
2
S
e.
T
<
2
,
,
<
S
,
< s
a.
V)
VI
(Si
EA
(A
OT -/
s
s
%
<
2
s
s
<
1
2 5
u.
X
O
1U
I
X
o
Hj
X
o !2
.5
0/5
-
1
I
1 1
-a EP
5,=
i 1
ill
a. in
1 ^
1 1
« j |
s x -g
1 s
1!
1
1 |
iilll
|
on
■a
J
2
^
5
e
if
9
m
1
1
to
^2
tf)
a
m
1
0£
E f
"C
X
hi
i/) (/)
a.
a.
fi
«
1
P*
Z
CO
-i
a.
z
2
«*-
j
(3
3
<
—
o
'U
e
c
«
J!
.9
w
u.
S
O
•*—
B
CQ
j_
w-,
-
A
*-
*
5
~*
*
5
^<
^V
o
GO.
-
JO|IUO|^
©
o
o
o
o
o
o
o
=
© o
o o
o o
©
©
©
©
©
©
©
©
©
©
©
©
©
©
©
o
©
M
lUJUJ3|duJ|
B
V
■8
5
£
f
in
11
II
W v
k/1
a.
©
o
9
o
o
©
o
a
© ©
© ©
© ©
o
©
o
o
©
©
©
©
o
©
©
©
©
©
©
©
_
I3A|iaUJ»|V
<
e
imiinioj
a
o
o
=
© ©
=
a
=
=
=
=
*•»
ujnn»|^ iu.ij jn j
E
1
m
ll
uj —
„
N
«
p.
N
-
-
M
-
«
-
fit
—
ipiijVg
io 9Dn|iua*|^|
o _
3 £j
CO
CO
CO
CO
i J?
5
5
5
5
8 2
a
o
a
o
a
o
o
o
2
1
£
1
a s
a
2
V)
0.
i
a
CL
3
11
£
i
■8
■8
■8
■3
"8 "8
00
"8
■8
•8
"8
3
©
1
2
2
2
2
2 2
X
2
2
2
2
2
e
B
3
1
z
u
y
o
u
io
a.
Q. Q_
GL
a.
Q.
a.
Q.
I
a.
a.
1
o
|
d.
i
O
3
(A
i 1
3
<
Z
c
i
3 |
c z
o o
5 Ji
3
E
Z
c e
Z
8
3
3
to
J
j.
j.
„
s
P
-
a
e c
3 3
|
,3
<
z
1
5
1 <
a z
S o
1
z
ifl
|
z
S
3
J!
sO
£
* £
©
^
rvj
IA
fr
I
1
.3
s
E
■a
>
v>
r-
©
sO
II
>
■
E
e
F
c
5
s
1
>
-5
.3
j;
_^ -5
s
,3
en
.3
<
<
o o
<
<
o
J
e
3
c
3
c
z
1
5 1
z
JS J!
Z
s
1 1
F
z
2
£
e
F
55
1
■5
F
s
o
c
.a
5
c
c
O
e
|
a
s
1
<
u
<
5 ~5
5
Jj
o ^
<
5
s
*~
z
O
5 o
z
o o
z
o
o o
z
o
s
!•?
^
> >
5
©^
* Si
s
sO
01
v»
£■
1
£
J
***
£
m
>
©
o
r*»
E P
sO
— ™
?
> w
|
-7
o
<
a.
<
- —
<
- -
<
-
— -7
<
a.
-
-
e
a.
<
s
<
2
2
S
2
< <
i
< <
b0 (/I
s
<
< <
in (A
<
in
<
|
<
o
s s
<
s s
<
S
2 2
<
S
s
m
o
O
UJ
o o
LU
1
a
o
o
£
c
.3
3 K
'3a
B
c
UJ
"8 .
c
M
e
V)
Undersized
Target
Other Reef
Species
Endangered
Prolcctcd
Species
Undersized
Target
Other Reef
Species
i
LU
a. to
Undersized
Targel
Other Reef
Species
Endangered
■5
I!
1
I £
c -
3 (-
8
I 8.
j
I
1
-a
c
e
g-
e
i
5
■
p
|
1
1
-J
11
5^
O en
I
J
X
X
Gfl
01
O
n
o
_
a
jO
fN
3
-D
3
«
•n
*^
•*
©
C0
o
o o
©
c © ©
© ©
©
© ©
©
JOHUO^
o
o
© ©
© ©
©
= © ©
© © ©
© ©
© ©
©
© o
© ©
o
©
*c
ltUUI3|dlil|
|
u
f
if -5
S a
I
■A
CD
V)
11
"
o
© ©
©
a © ©
© ©
©
© ©
©
•o
<
a.
©
o
© ©
© o
©
©
= © ©
© © ©
© o
© ©
o
©
© ©
© ©
*r
u\nvujoi|v
<
e
f
|B.(UlJUj
o
© ©
©
© o ©
© ©
©
s ©
<=
»•»
i,unw»D^ ju.ujnj
8
1
*»«
II
M
C
-
M M
-
- «
M „
„
- -
-
„
qDl«J\fl
10 3finnu3ihl
(/J
v)
to
i s
3*
5
5
a
5
o o
a o
a o
o
o o
CL
a a
a.
a a £
a a
2
s a
a.
M
X
1
1
■8
1 "8
"8
"8 "8 "8
"8 "S
on
§ -g
"8
e
s
2 z
-
2 2 S
s s
X
X S
2
s
t
1=
I
Z
a
O.
o
a.
a
D.
U
a.
ki
K
c£
a. yi
c
CL
C
a. «a"
P
D,
Q. tA
p
o.
e
g
|
g
%
S
TI >
c
C
c
c
11
c
c
e
c
c
.3
.3
,3
s
.S J
5
"J
>
£ ~i
8
a
(A
i
s
a
I 1
z
is J <
i 5
3
5
5 1
<
Z
1
8
J
5
6
c
1
o
1
5
c
1
p
c
p
c
c
5
6
_
= "B
^
- u <
— 1j
t
J;
b fc
<
S
■s s
z
a s z
S g
s
o
S 5
z
II
* —
£
^
2
m
i:-
|
|
S
£
>
•£>
r-i
V
*°.
°\
ii
"
"'
> ~
1
B
2
<
a.
-
- -
<
a.
- i 2
- -
•3
-
- -
<
s
<
< <
2
< < s
< <
<
< <
2
w
z
in
C/> V)
2
(/I <« £
V)
a
m
<
i
-
ft.
1
=
=
~
=
=
4
z
*
I
n
E
5
3
i
*
O
^
<-.
•B
4
•^
3
5
*
-
joi;uo|^i
o o
© ©
© ©
©
©
©
©
©
©
©
©
© ©
© ©
© ©
©
©
©
© ©
© ©
© ©
©
©
©
©
©
o
©
©
©
©
©
©
©
©
©
©
SO
lU3UI3|dUI|
©
u
©
£
t> 'S
« e
o
*fi
3 F
?i
fc«
© ©
o
©
©
© ©
©
© ©
©
©
©
©
©
©
■
1
TO
<
Bj
© ©
© ©
©
©
©
©
©
©
© ©
© ©
©
©
© ©
© ©
©
©
a
©
©
©
o
©
©
©
©
^
S3AI)IUJ31(V
©
<
e
f
|BI»U.->|Ud
© ©
©
©
©
© ©
©
© ©
©
©
©
©
©
»n
funmj^ luajjnj
©
c
u
— —
_
_
_
_ _
_,
_
_
s,
2 c
«M
■a *
tlA
"
"
" "
—
~
~*
—
—
"
B>
-
—
IpJBMfl
|0 .lpiT|tuai!I\|
| Si
(Si
VI
^
^
.a —
o >
55
(A
o o
o
■s
3
C
J
J
5 S
<;
J
3 s
^
c
S
s
<
o o
J
o
3
ii
E
3
3
i if
F
z
1
3 5
F
Z
1
3
e
z
3 is
2
e
F
55
•2
5
§
s
s
5
|
.3
5
c
c
.3
C
c
.3
F
o
c
F
I
_ c
1 1
<
|
' -1
5
6
fe
"2
<
o o
c c
1
c
o o
o
J3
3
1
3 S
z
1
z
o
o
o
z
3 3
o
i-s
1 1
•>.
C^
©
z
o\
Ox*
i
5"
r|
OX
©
©'
M
£■
S
f
y
£
s
>
O
r**
f^
r4
oy
©
«M
ii
1*1
*>
V
o*
> w
a.
c
g
e
**
— —
o
—
_ _
<
_,
_, _,
<
<
s §
1
e
A.
< <
00
E
<
(JO
< <
Pa
s
<
in
< <
(A bO
a.
S
s
<
3
00
a.
1
J3 J5
3 3
00 00
S K
3
i
1
E
2 2
O O
1
<
<
s
a
<
CO
II]
o
<
3
<
a a
a
s
1
[
do'
E
T>
L£ *~-
£ ^r
•^
Undenized
Target
Other Reef
Species
"5.
s
p g
11
(/J CO
Undersized
Target
Other
1 8.
Undersized
Target
Other Reef
Species
Endangered
1 s
II
Q- oo
Undersized
Target
Other Reef
Species
Endangered
1 s
H
.1
it
Other
Groundfish
Species
Endangered
1 .
Undenized
Target
Other
-5
1-5
ell
|
[
BO
J2
.9
s
1
'5
z
m
is
9
?
.j
X
1
E
<
X
1
1
x 1
I
a
3
e
■
X
z
3
CA
O
o
o
o
o
u
U
<
B
_
„
_
_
O
■»
—*
—
=
w
=
.e
t/;
E
J!
15
*-
3
3 -3
n
C
I s
09
3
So
o>
o
(rt
A
**
^.
JO
<--
M
C-
3
■
*"
**
W
O
GO
Gfl
-
JO|IUO|^
O
O
o
o
o
o
O
o
o
o
©
©
o
Q
©
©
a
©
©
©
©
©
©
©
©
©
©
©
©
o
e
©
=
•o
)U3UI3fdlU|
©
8
V
■e
2 -
8
n
11
M
o
©
e
o
=
©
©
©
©
©
©
©
©
m
w E
1
&
o
o
©
o
©
o
o
©
©
©
©
©
©
©
©
©
©
©
©
©
©
o
«t
f3Al!BIU31|V
©
<
o
1
(II) 113)0 J
o
o
o
o
=
©
©
©
o
o
©
©
*•»
ujnvi.i^ iujjjii )
©
S
1
is
-
-
-
-
-
-
-
-
-
-
-
M
n
fN
r*
u —
V)
<^
Q.
r^
<*,
—
ipiuA'a
r^
io anniiuSvw
s
jl
o
5
o
o
to
Q
O
o
5
o
)
■
£
a
2
c/)
a.
c
a
a.
a
£
o
■£.
5
■s
X
5
OB
s
•8
oo
"S
■g
o
■s
t
s
1
■g
*s
O
j
J
2
I
s
s
3
s
s
D
s
5
s
-J
t
,2
a
6
1
z
»
a
k!
D.
ki
p.
«
CL
CL
H
CL
CL
B
c£
CL
a.
a
w
F
c
Q.
<<<
e
B
O
O
o
B
ii £
B
C
J2
c
c
_g
r^
g
E
F
E
3
P
=
a
e
,3
E
X
l J
g
o S
O
^
|
e
s
^
s
B
5
|
5
^
L.
O
^
•
s
z
J! i
i!
z
=
e
3
s
z
C
=
£
z
B
1
z
1 1
i
C
e
F
F
a
s
o
6
i
s
1
c
=
=
c
e
F
c
F
c
g
c
"E
<
z
3 o
p
o*
5
<
z
c
c
a
e
o
z
o
1
5
o
c
1
.3
O
<
z
3
s
<
z
1 s
If
> t
o»
on
ON
E
* °-
*
«q
©
^^
—
M
n
o^
e-
|
.1
S
E
-a
>
o
rs
2
©
ao
*P
rn
«
o
©
"
N
•N*
o*
©'
>
, ,
1
<
e c
o o
•3 "a
c
<
.2
o
c
o
■5
<
1
e
■s
c
Q
•S
<
c
<
o
s
CL
2
J J3
3 3
3
a.
1
*3
"a
3
3
a.
9
3
J9
3
2
3
I
3
ft.
2
00 00
00
s
on
i?
?
S
?
?
s 5
S
u.
00
E
S
(J 00
1
<
a a
2
<
o
u
3
<
a
g
3
<"
s
«j
<
s «
E
to
u
a a
3
3
3
5
t3
s
s
3
LU
<
3
c«
o 5
T
•8
1
■a y>
oo S y>
"S
1,1
1 B
1,
-5 "8
c 6
u "b S ?
sill
1 s
-|
11 |
Q
1
l5 £ on
1 1
C 5 G
Jen
o o t^
III
lei
lu £ )
"i s 1
■5 £ £ |
■g 1
ll
1
3 'ij
ill
111
3 u
Jl
.(
8
1
1
e
Q.
c
4>
I
u
E
3
a
1
M
c
£
1
c
<
-J
X
o
m
X
O
1
1
6
<
E
1
iC
3
s
<
=
s
=
=
=
1
*
i
■
3
]
o
u
3
5
3
'*'
3
»
-
joiiuow
O
O
c
c>
„
n
O
O
O
©
©
©
©
©
©
=
NO
)U3uia|diu]
©
|
H
©
£
£
Is
s
trt
w
O
N
N
r*
f>
n
t^
m
©
r-i
©
©
©
a*
w E
M
C
i
<
a.
O
O
•»
*>■
O
©
©
©
©
©
^
UAUKUJ3)|V
©
■<
o
s.
c
|«l)U3)0j
O
O
O
O
O
©
©
©
»*»
»jnnra|^| ju.ujn.)
rN
-N
M
<-N|
^
«
-
m
n
-
-
«
N
<*i
-
-
:
_
ipw-
[/)
VJ
U)
V)
3 s
2
O
£2
a
a
a
s
a
<2°
BU
5
13
2
QU
C/J
5
VJ
a.
v)
a.
a
2
in
c
jq
S
1
■?
•g
■8
.c
■g
00
■g
H
"g
"g
■5
■g
on
■g
e
8
3
2
5
2
s
X
5
if
2
S
X
s
X
2
5
£
IS
«
I
z
Q.
Q.
id
0.
^
0.
y
Ck
0.
ki
a
M
0.
a.
vi
e£
*
M*
e£
*%
0.
vi
P
P
V4
e
u >
>
C
5
s
S
_=
>
.3 J
C
O
B
c
c
c
.at
e
-5
1
■5
,
1
->
1
^
|
<«
<
<
^
O
<
o
3
Z
e
3
c
z
-
5
5
C
z
c
P
c
z
3
ii
is
s
c
Cn
■
1
1
c
C
c
1
c
c
^
3
<
z
3
.0
<5
z
1
<
z
^
O
.0
J5
5
z
e
e
1
■3 -S
E
s
©
»5>
i
©
»I
00'
P
,«.
r*l
—
*^>
00
Ol'
n
£
i
r^
~
£
is
£
>
o^
t>
■V
O
u
s
s
O
O"'
o r
nO*
*
8
T.
s
§
i
m
>
s
>£>'
c
E
e
e
2
1
<
a-
_o
T
<
■2
<
a.
^
•3
-
•
s
"5
_
2
*3
<
_
s
■5
2
s
"5
<
flu
2
£
<
2
n
u
2
<
s
E
<
Z
^
?
„
■i
E
E
<
UJ
<
1
2
<
3
2
<
s
5
7.
O
s
<
V)
< s
I
j
i
(A
ll 3
|
I
00 - O
"°
^
1
00 £ y>
1 S
u QO
1 s
|
lE
I
oa'
«0 O (j
III
c
1 \
O v>
0.
ill
uj £ vl
ii
ii
g 8 S
UJ it V)
at
8 S
1*
ll
flu
g
J
•sl
Ji
1
p f
is )
e
e
£ c
a
en
,_,
c * 1 -
= 6
E
?
t c
X
1
3
■s *
1^1
1* B
CD w
X
"C.s
£ ^
w - -e
■
1 i I
311
f-
O V)
t-
5
O
O
O
Eft
GQ
CD
CD
CO
4>
^
o
»-
<
_-
_
M
_
4>
0.
—
=
—
—
-C
s
Eft
E
J
1
lb
S 11
■l s
a 1
Q
3
O
-=
™
w
S
•
IS
*
nC
B
o
CZ)
-
JO)IUO|^
O
O
O
O
O
O
O =:
O O
to
|U3IU3|dlll|
8
w
X
v '5
1
w
1!
M
O
O
O O
=>
M
£
O
O
a
O O
a
3
e
!
<
O
O
O O
O
-»
■9Ai(«iua)|V
•
1
|«!)iuioj
O
O
O O
O
C3
*n
■ajnna^j 1u.1j.tn3
6
y
_
r-»
oj
rs
™
O O
a
c
1
n
a z
fad ~
tO
-
-
p,
-
«
O O
=
t
m
r4
O ©
s
_
i|3i«3.
2
%
« %
m
=1*
ll
>
©
=»
m
rn
IS
*6
OX
S In
•r
g
I'
O
c*»
i
1
i
**l
V
•0
of
00 n
O
C7>'
s
s
M
r*i
rs
*■*
1
<
—
„
,0,
<
c/l
■¥
a
"C
u
i
~
=
z
tZ
C
a
1
.1 c
u
CI
s
3
«
•B
r ""
r "
O
n
lU3U13|dlUI
n
»aiiiuj3)|v ftiiuaioj
Mjnn^^j ju.ujti )
* E
ipjm.ig jo opn»iu3R|,\
i
•5 3
II
Ij
11
.22
'u
JS
—
P
a
—
c
u
'5
js
"3
Ou
u
s
-
1°
o o o o o
o o o o o
© o o © o
o © ©
o o ©
© o o o o
© © © © ©
© © © © ©
© © © © ©
© © © © ©
© © © © ©
© © © © ©
o o "5 o S S
§ §
c c S c c "5
3 3 O 3 3 3
&j k! k! k! k!
EL CL 6. EL G.
C e c c E
3 3 3 3 3
s § s s §
o o o o o
C Q Q Q 5.
■ n tt u
X X X X
k; k; kj k:
k: k:
1 1 5
■g 1 -g 1 !
6 b
. II
g e ij-
I i ■ ?
I 1 -=• 1 I ?
| ISsli Hili
X <
II b j g ■ i
ft
;. s -s «
II
+
i '
slS
5
Si
luauiaftluii
1 —
II
WAIJ»UJ.>J| y |»!)U-110J
njnnaw juajjn j
is
qji«3\g jo 3pniiu3*w
.3
si
> w
W5
-
'£
un i^i lo
X S — S « 10 v
Q Q Q Q Q a. a.
3 3 3 3 3 3 3
i_ O
U C
■2 -i = Tj
5 = 3 D 5 D 3
E E s E E E
P" o o 2 5 6 o
t *. =• 8 J i
I &
£1
© © ©
© © ©
© © ©
00 00 « 5
X X X $
II
s *
SI
S? ii
sis*!
= 1
1 1 1 1 1
- o
Ho
11
J0)IUO^]
©
o
o
e
o o
© o
SO
lU3U13|dUJI
•a
—
V,
rs
rsi
fS|
fM
o o
e o
•*»
in
00
M V V
c
M
3 C .1
B
■ s a a ■
1
u < £
* • •
^
o —
o
S3 sl)BUJ3)|V |B))U«1)0J
c
o
o
O
o o
»*»
a,
S
K3Jn£B3|^j )U9Jjnj
1
N
<-4 fN -n
©
o
O
O
c o
s c
I 2
H
m
t/f tf)
GOO
£
a
s
S
□u CL.
J=
■
X
CD
e
5
i
_s _s _s
j
J
2
1 1
1 J
s
s
e
«
1
z
K
a
a
ki
se a
k a.
5a
*■»
*«» w^ irt
Q.
a.
6.
a.
d o.
a. q.
4
©
1
55
o
3
.IP
Cfl
1
l
B
5 o
1 1
La J b
U Q O
c -S c
O O
o
o
o
=2
3
3 8 8^1
2 2 2 Z tfi
n n a < _£
3 *> v --5 3
z Z Z Z tfi
s
c
.2
B9
.5
5
3
s
s t>
6
5
h
(j
= =• =' < b
=' = ^ < 5
£
1
1 1
§ a a
6 6
2
6
<5
6
6
3 3 3 i X
U. Li. U. 2 O
3 3 3 i X
u. u. u. Z O
oe
Os
i
11
—
s
1 £
"3)
> s
M
* i
Si
*
— — o
w
£•
I
1
.s
s
s
>
**>
E C
3 C
1-
I
«*v
^.
Os*
m
^
* * %
£ % s
* » U
*• it —.
Os
f*l
= R »
<>
«n
— rM —
"*'*'*
J!
©
e
a.
s
t/3
V)
H
u.
u
Id
a.
.IS
|
t5
1
M
tz
S o 2 E <•
1 fill
s « 2 o c
1
O
|
o
s.
1 1
CO
e
S
Q-
S
2
2. a. E en
a
o
J> .c
X) XL
1
c/>
e
"5
£
j=
1
u
m
E
a
■tt
e
'3
O
t
e
c
F
■
a.
w
m
e
8
Ck.
s
s
1
1
■ockey
chum
pink
sockey
chum
pink
f-
o
K
5
E =
s
§•3^
2
£
2
a
'C
<
0>
i<3
=
=:
JO
1
V)
a
o
U
IZ
i
•%i
+-
i
tS
t/5
b
O
U
o
3
o
""I s
Slfa
1 B X ■
7- ZL o
m
♦*
-np —
J01IU0|^
, -y_
o
— —
o
a
%*
-
"5-
©
"~
— ^ —
©
|UJUJJ|(|UJ|
o
o
©
= O
o
a
3
o
C9
o
©
—
J
M
«
M
-
-
©
= —
o
o
o
o
=■
O
©
©
M
1 c £
Bj
-
-
o
= ~
o
=
E9
o
e
©
©
o
c
| a 1 f -
1
u ;e
m
o
o
o
O
o
o
r*>
©
■o
■<
e
t.i.\i)*u j.->|| y |»I1I»|0J
o
o
© ©
©
©
©
p.
as
s
1
tajnnai^ linjjrr)
o
=:
©
= o
o
=:
z:
©
_
a
_
_
*s
M
o
o
=
©
=;
©
©
©
»»«
It
W m
=.
n
n
-
r. -
o
a
s
o
=
©
o
©
©
©
©
H
q JiBM'g jo .>|in jiu3u jm
VI
(/>
a £
o
w
o
o a
o
v
. i
<
s
-
„
r-j
o
z
M
£■
|
|
s
£
■a
>
8
o o
o o
S
1
s
s
3
W
«o
9
6 P
vO*
?s -
of
f-f
-
N "
■s e
> ~
|
O
£
u
m.
u.
9
o
o
o
O
a
s
E
s
c
V.
Sd
K
2
e
c
8
3
a
3
■
3
E
jO
—
^
.O
c
jf>
c
■8
2
"5
1
i
|
S
•5
5
p^:
3
1
|
1
»=
E
1 j
1
c
3
1
s
1
B
"S
1
"o
^ O
I
I
i
1
<
1
£
DO
CL
-H o
o
a.
"5
|
i
B9
1
M
■a
.c
|
J
x:
*
&
1
1
:i
I 1
c
§
i
i
c
^
3
E
3
= 3
3
e
*
£
I
\
■
'C
i
E c
fc
-J
H
-j
z
M
O
b
3
VI
■
O
CD
s
S
X
o
o
£
o
CD
0>
*c
<
&
o
o
o
u
u
XI
u
CD
ffl
<
CO
>
o
X)
<
>
o
X)
<
c
o
e
e
D
e
c
c
o
>
<3
o
o
5
"a*
o
X)
u
2
1
e
3
13
c
CQ
<
e
e
CO
p
P
P
09
s
c
V
j=
en
CO
to
0)
o
Gfl
u
•**
s
en
9
09
£
«
O
U
+-
CW
s
£ S3 % 53
a > e >
£ O ^ O
u >» >> >\ > *
c ~5 ~3 ~5 ~B
P U. U. U. Lu
T5 = T3
CSC
D fa D
c
fa c
p
^3 = =
3 O 5 5
&o
s
u
5
c
o.
u
•a
^ S
XT' w
x> p
c 5
S 2
x x
c/2 H
x
u
ad
X
X
*
o
E
O
O
o
E
X
_4J
B
-a
X
>
o
Q
"ob
i-
Xj
£
6
1
c
6
S
X
<0
IS
S
■o
-o
c
3
CO
1
O
c
—J
OS
'{3
a.
-o
s
01
o
o
u
O
X
eh
IB
u
X
VB
u
€
CXJ
'o
c
X
'o
ca
3
ed
Cu
J
c/:
a.
JU3U13f(luj|
II
.JM1«UJJ1|V
I.IIU.11UJ
ujnn>^ )U3Jjn3
II
•a g-
qmxig
lo af»niiu3wn]
I 5
o © o
o o o a o o
— — — LlJ UJ LU UJ LU LU
oao2222<2 2
c/i c/i (/i
■8 = i 5
■8 "8 --
E E .2
k! » k!
8 J
3 o £> y
J! 2
2 R
liS
J S 1 "I
8. I i I
E J
I £
QQQ22222
a o a
1 s a
"8 "8 =
E E 2
U w k!
■= "8 TS "8 = "8 "8
-2 E E E -2 E E
i 3
■i 2
3 | J
2 2
Q\ «n M
a d o
S 2
% r- i*-
© fc o
£ -3
J 5
a i
e s
1 I
III
o © e
<2 i£ !£ £ S 2 S
D Q Q Q od. d 2
"8 "8 s *
E E -S °
*•> 0. a. a- m ^ ««
J .8 3
J 3 2 2
° S °
J! I J!
* I i
I | J
n
ox
c
E
■<
o
p.
«5
B
1
-
jujiuoiv
other salmon 720* full above J,PC low REG 4 2 2 2'
red king crab 4883* full below S.p mod REG 43320 4 33233
other king crab 938* full below S.p low REG 42220 1 22112
Tanner crab 244705* full below S.p mod REG 43320 4 33133
snow crab 46102* full above J mod REG 43320 4 33133
Ill BOT yfiniok AK-I 79.764 33 4 77 under near
Pacific cod 4.942 1 1 under above
pollock 4.086 08 full near
groundfilh 8.451 1.9 under near
ycllowfinsote 21.187 under near PC mod DIS 43320 433223
pollock 21.715 full near PC mod DIS 4 3 3 2 4 3 3 2 2 3
other flatfish 7,622 under above PCS low DIS 43220 222111
Pacific cod 6.464 under above S low DIS 43320 433223
groundfish 9.196 under near PC low DIS 4 3 2 2 1 1 1 1 1 1
Pacific halibut 554 full below S mod REG 4 4 4 3 3 4 4 3 2 3
herring 43 full near J low REG 4 2 3 2*
other salmon 324* full abo\c J.PC low REG 4 2 2 2*
red king crab 8648* full below J,PC mod REG 43320 4 33233
other king crab 1928* full below J.p mod REG 42220 1 22112
Tanner crab 1349275* full below J.p mod REG 43320 4 33133
snow crab 3196459* full above J mod REG 4 3 320 4 33133
*-* «-j r^ — t-> r>» — — —
f-^r-^r-i — -r (•» #M t*l 1*1
pi m in — ■• * « it t*i
©OOOO* • • OOOO
|
C
3
<
a
c
£
o
I-
o
3
«
1U3W3|(JUJJ
„
if
l|
<
-
S3AI)VUJ3}|V
|BI|U3J0J
(•»
tajnsni^i ju.ujn 3
»s
II
J E
u —
Wl
a.
III BOT rockiok AK-I 15.636 6 6 39 under above
yfiniok 4,765 2 under near
Pacific cod 4,563 10 under above
groundfilh 2,139 0.5 under near
rock sole 13,571 under above PC mod DIS 4 3 3.
pollock 6,719 full near PC mod DIS 4 3 3,
Pacific cod 5.159 under above S.PC mod DIS 4 3 3.
groundfish 5,410 under near PC low DIS 4 3 3.
Pacific halibut 741 full below S high REG 4 4 4:
herring 8 full below J.PC low REG 4 2 3,
Chinook 426* full below S,PC low REG 4 2 2.
othersalmon 811* full above J.PC low REG 4 2 2.
red king crab 22675* full below J.p high REG 4 3 3.
other king crab 3549* full below S.p mod REG 4 2 2 i
Tanner crab 3S8878* full below J.p mod REG 4 3 3;
snow crab 363302* full above J mod REG 4 3 3 i
-
tpimtg
10 snmiu^RiAj
O ,,
s s
O U
3 s
■g
3
n
o
U
1
1
■B
1
3
z
■M
e
e
5
3
n
o
V)
3 i
s
.a
'3
3
c c
3 3
11
1 ^
1
s
>
— ©
> w
§ g
Jj
o
m
2
E
.5
to'
.5
I
1
-a
s
M
h
1 1
1
g
<
!
-=
iZ
I
3
O
*
5 S!
B
1
lU3UI3f(IUl]
is
S9AI)IUJ31|V
■3jnfV>|X |u,u jn ■■)
II
ipnnig
io aonnu3itiAi
II
if
oaaQo222222a:a;
5 "S
.2 E
', i % t %
3 S E E E 1
iii:
***>•»*•)
J 8 %
-a -o -a
8 J
s" JS
o o _o
i! 1 £
oo *ri -r
Hill
I ill 1
o o © © o •
— — — — UJ UJ UJ UJ UJ UJ
qqoo222222
J £
S S ** H_ H|
■? S I 8 li V I « J I
S ii -K
J! ii
9 I 1 1 I
111 1 I
5 .! Hi 1 ? J
= 1
3 1 1
© © ©
c/> w > o o a
« » ~ UJ 111 UJ
Q Q Q 2 * 2
S i! i!
a §
S 3 S P]
S |
a
M
B
e
5.
B
BQ
e
V
1
-
joiiuow
red king crib 2599# full below $,p low REG 4 3 3 2*
other king crab 3293# full below S.p low REG 42220 2 22122
snow crab 74* full above $ low REG 43320 4 33133
III BLL Pacific cod AK-I 97,473 53.7 100 under above
groundfish 2.391 1.5 under near
Pacific cod 3 r 988 under above $,PC low DIS 43320 433223
pollock 2.653 full near PC low DIS 43320 433223
— — fi
©©©XXXXXX
ogogg2o;c22
OOgOOOOOO
o o ffj. fit
x> 3"5"53'v"y"oi
ra c x> x> o X' x. _c n
llillllll
o * a
— ' os"
in r»
=tt % =» >r, O
(N % IN O » fl
T OO O C -t srs
_ XI
J g -s S .o
11 1 S S k < fc ! i
f ' § S -S .g -q J S S
1 1
c c
3 3
NO O
— tn
<
1
3
II
1
-j
-i
CO
©
5
■?
E
(J
a.
o
c
3
f>
O
1
s
1
"E
J
— r*i — r->
— ^ — N
— r-i — ^
ooooxxXX
o
1
!c/>vjOOOOO
» — — UJ UJ u in u
oqq22222
.£ .2 .£ E .£ .2 ^ £
CLtWCLtAMMMVt
llllllll
3 5
e c
i-s
o
O
s
a
_S
lis
X) ■ c
■s ■§ =
— ©
© ©'
SO O^ rn —
-s
a
a
i
9
' 1
1 1
O 0.
- n - -
sO
)l»UI3|(llll|
— IN — IN
VI
1
- *s
1
u b
<
— in — m
— m — •«•
-
f3AlllUJ3)|V
|BI)U3|0J
« » - „
m
sajnnaw iu.iJ.inj
© © © © X
"
S 6
= 5
1 1
V
c
m m m -v —
m i*i f> ■* «n
-
tpiaa.tg
io spnuuitoui
Si
DC
S2 S2 £ 2 S
G G Q 5 2
x;
3
pa
o
1
IT
1
35
5
■g i S S S
£ _0 _© _0 _g
e
■
^ U y
M wf CU ifl M
5
09
o
5
GO
e
If
35
1 ill
e
■
5
3(2 - - = = 3 3
— ■ — •»
* IN =» ©
O O — V>
Tl V N -iO
-. X)
1 •§ 6 -
•S -S T5 5 M B j>
5 ■! c ■» ,! .5 ■" 5 u
1 J ■§ es-g-sj i
*2 (3
o
ON
Os ©
E 3
**i —
<
- -s
§ 5
: I
-J
-J
03
9 § 1 s s
11
II
Si
1
s
■a
>
>
•<-> os o> ^
s
1
5
e
CL.
i
I
I
(A
to'
1, la
If Hi
O £L S, 0. o
1
1
M
1
'a
£
i
■r
3
•<
aw
S
S
U
n
-
1!
!
u.
3
o
5
J3
-
8
s
o*
-
joi;uo|*«j
O
O
-
© r-4
© —
© <-4
© —
©
©
©
©
iaj
|U,1UJ3|lluj|
s
p
„
©
x
11
s
«n
^
- t
O
ri - - - -
-
© f-i
© f-1
©
©
©
oa
£ 1
BJ
c
!
<
c.
- -*
O
„ _ _ _ _
~
© rs
© fM
©
©
©
i
m§
i>A|iaujai|v
— r-.
O
■* _ _ — ~,
r*
O r^
© i^J
©
—
—
<
1
imuaioj
X X
O O
O
~
X
X X X X © ©
X
X
© ©
.
=
*•»
tajnna|^ luajjn^
©
6
„
_ _
rv. r-,
_
IS N ■•> N N *|
N
J!
«
3 C
a
- -
-
1*1
-
^ ^ ^ ^ ^ ^
-
-
«-4 «-<
-
N
_
ipiajlg
to annimSa^j
^■a
s= *■
o o
> S
bj w H S B S
O
s a
5 a
5
. _ ' n > in iii n|
O O 2 0: 2 2
|
CL.
i/i c/) c/) c/) 1/) t/>
q. a. a. a. a. a.
£
to
a.
a. a.
C/) CO
a. a.
CL
CO
a.
afj
*J
T>
a*
5
, .,
; s s = ; ■?
■8
>
S is is is 00
j,
■g -5
■3 ■?
"S
■g
o
5
Jl .2
s _s
_S
_o _o _o _o _o £
E
© ° e is
^
_S
E J*
E E
E
E
e
■3
&
z
U U O
a.
a
l fi
n ra n) -2
3 u o o
C c C -G
«r>)
i s
m a. o_ a.
M £ M « V
* £
i
^
—
n e
-t ©
•»
H
©'
t
|
s
c
i
I
E
>
— r- 1^ %o
— ^ *£> «rt ©
"5
li
•O —
en —
S m S P "
1
S »
©*"
c
E
n
> ~
© *o
1
O" T *N
o*
1-4 r- ve
c
S
— — — fi r-j —
■*
— r*»
M r^i
—
|
P
B
e
fr
fr
5
■<
e
<
5
a.
^
«
a.
*
a.
|
Ml
ii
^
IS
Qi
S
01
S
M
"
<
X)
<
<
1 1 1 ! 1 1
-G
IS
c
D
O
2
E
a_ _ C
E "S ^ .So
a 1 1 a 11
f
S
e
3 M
i 3
s
1
I
M
J s
c ■=
2
E E -S « S S!
■S 5
i _
-O u
■0
n ^ c E ^ k.
c E
e
S c
1
CD
1 I
§ 1
11
3
90
s § s ^ s 1
<2 1 I 1 1 i=
n
CQ
•3
LE
northi
alcid
sea bi
mam
Dall's
Sidle
15
3 E
3 E
1 1
1 1
a. ;
4
s
c
11
£ E
|
|
I
■5
«rt
E
]
1
il
|
1 1
1
E
J|
1
"J
|
1
—
5 i
I
1 §
• £
-3
5
c
i
I
1
3
:
£ ot
"
a. m
•E
w
■
M
i-
'£
5
-J
-J
03
1-
2
1
E
|
O
CQ
—I
-J
CQ
*
a
BB
<
■5
QJ
a.
I
t
k.
X
s
=
s
?
^
J5
C
1/2
I
s
■
3
ad
O
3
n
S So
«
*
■■3
-
joiiuo^
O
o
o
o
o
o
o
o
o
o
©
©
O
©
© © © o =
© © © © ©
© © © © ©
© © © ©
© © © ©
o © © ©
©
o
©
© © r->
© © (-4
© © —
©
©
©
©
©
^ _ „ _ _
«
|U9UI9|dUJ|
_ _ „ - -
1
„
— — m — —
f
s •
■i
VI
= E
M
o
o
©
o
©
o © © o ©
© © © ©
©
© © m
©
©
"* ~ ** ~ ~
M
3
<
&
©
o
o
©
©
o o o — =
© © © ©
©
© © r*
©
©
" - * - -
T3
-r
V.1VIJBUJ3JIV
©
_
m _ ^, _ _
<
e
ptiiuaioj
c
o
o
©
©
o e o © —
© o o ©
©
© © ©
©
a
«
»jm.-R3|^| )U3Jjn3
o o o X o o
c
V
„ ^ „ _ _ _
3
•M
If
u —
"
-
«
M
M
M M « ^ M
~ ~ ~ „
M
~ M „
-
M
" " "» - - -
—
l|3)R3.\'e
io 9nn|iu3B|^j
c *
© w
a S
& a
on „, O O O O
OO
a.
a.
00
a.
a.
v) m c/i (/>
a_ a. a. a. a.
c cl q_ a.
Cu
1 o. a.
a.
a.
5 5 a s a a
■a
a
5
"8
■g
"8
■§
1
"8 "g *s ■§ ■§
1 "8 "S "8
■s
"8 "8
o
"8
e
e
E
E
E
E
E
E E E = =
E E E E
E
E E 2
E
£
o o o o o o
£
IS
■
*|
z
c_ a. *« ^ v- w-
in
o.
a.
y
a.
O
a.
O
a.
i. a.
o
.a J
„ ,
VI
<
<
<
<
<
< ^ ^ < <
< < < <
^
S i <
<
<
belo
near
belo
near
near
belo
3bo\
e
3
9
z
Z
Z
Z
Z
Z Z Z Z Z
z z z z
Z
z z z
Z
Z
e
JS
•
.2
a
<
<
^
<
<:
^ ^ < <
< 5 < <
<
5! S <
<
^
z
z
z
z
z
z z z z z
z z z z
z
z z z
z
z
©
d
If
8 !
> u
% e.
1
to
•» ©
£•
■ S
M
s
(£
■a
>
^ — r~ - ^i ac
£ ^ r^ r-^ — -r
© t*i c^ *r
u
IP
vO
■i i
=*t ^
S
« e»
**
« * =tt * «
"^
^
*
M
™ — •*■ r~
M
"
~
— ■*
i
©
a.
£
<
£
s
S
Ut
fc
1
J
C
1 g
"O b
3 S
£
3
1 z
$ * £ E 1
5 = E 8. 2 E
•s = = |
a o 3 3> ?
■a
IS
ds unident
nils M
porpoise
while
1
2
s
<
1
1
pollock
groundfish
Pacific halibut
herring
chinook
other salmon
3
.i §
1 £
O U
1
■a
Shear
unidei
north<
storm
cornu
gull -
3 1 1 1 1 =5
*6 ^
^ i
B 1
E p
>
1
8
•
1
E
1
(A
1
i
a
I
05
n
'C
■
a
<
l
I
4»
n
|
1
E
3
o
J2
-
8 3
2
-
JO)IUO|M
-
— est — ego© — — — —
— — — . M © © — — — —
— N «N - — <"M © © — — — —
«
lU3UI3|tlulI
- ^ _ _
|
u
a
w c
s
V)
3 S
U u
-
-
— rs| — *»o© — — — —
— r>* 4S) — — TOO — — — —
e>
<
c
— (si — iJ-OO — — — —
— (^4^ — — ttOO — — — —
<*i
- - iuai|Y
_
— r"1 — r*s © © (*4 f»4 r-4 rs|
— r**r*i — — mO©«NCM«s)
vunNR.Ti^ ju.ujn )
© © © ©
c
u
„
1-4 (s, ,-,
fN r^ rs. r^j (s, r-i —
r-i
_ N N «,
£
1 s
M
H
trt
N
— r-i/N-r — — — — — —
m
— N M N
U "
rt W M ""
^
lplB>.<8
(-4
e
S
° .2 E S .2 £ E E
ooooogoooogg
J
^ J _S ^S
V
e
9
M
'§»
z
00
, w c_ r . c_ 0. a. c D- a.
u y
u
u
y 0_ O. (j
a_ s^i t> q.
)L
(jO^OO o_ a. 0. 0. a. 0.
£
M
■3 >
b a.
.* -
.a J
«>
00 > u > j. > ci
oo> > « > > p > > > a
> u
^ „
«5
e
5
CA
| J 3 I
Jo .0 3"5.6*5'<3"o-o
«a ne c _c « jS j5 .0 n
ooJa ° £ J n aJoJ°J 5
3 "v XI i
c -c n c =
ills
H A M C
B
So
-a
&
3
§ 1 1 §
O
"1 5 5 5
«
11
-5 ^
> J>
* t
s
_
1^ © m
O IN - ^
— fS —
v%
—
— ' — ©
£•
s
■B
u
J»
to.
>
Vw"irMoo©v>viOi~)C*i
•n •#-. so v.
3 — 00 vi
sO — TT 5
■r- -r — -r
— m r-. *n
m «r» *r 00 *e •» oo
— 00 — ■»
i e
ri
so' — * — ' — — '
—
•is
*
» « « «
> w
»>* r- « v. r-
00 r«. m ** so so
r» © «o 00 oa
«S rj %S ri c» —
I
3
e
a.
rs
fN
N
E
<
w -D
<
<
|
1
1
arrow tooth
flounder
Pacific cod
groundfish
Pacific halibu
chinook
other salmon
red king crab
other king era
Tanner crab
snou crab
other flatfish
pollock
Pacific cod
rock fish
groundfish
Pacific halibu
chinook
other salmon
red king crab
other king era
Tanner crab
snow crab
1-4|11
m
^S
3
3
11 1 i 1
i J 1 1 1
.5
1
i/5
1
■e
1
sZ ~ £
"8 -s
1
fc y %
C w "S -o
3 3 u «
ISS
11! I
0. S a
!s S s S
M
" - s
■s 2 i
■= g s g
S ° a. ••
-1
5
O
-j
n
CD
m
tt
■<
:
til,',-,
.2 .2 E -2 °
«a ul. Q. «A ^ ^,
% 3 J .8 3
C* ■* ^0
3 S
■?
O
c-< r^ ^ ^
■
£ |
1
<
C
O
O
c
CO
O © r^
C
c
_,
t3AtlBUJ3||V
<
SL
pproiod
O
000 =
CO
O OO
O = = X
m
»jnstt3|^ luajjn^
e
u
3
~
1!
u —
M
^ ^ ^ ^ -
c
—
ipiBJ.Cg
10 spniiuSsui
O M
M L.
= -
tn
II
O
OOOOO
C/) (/)
00
a s 2 a s
a.
CL
eu
a.
a. a. a. a.
CL CL CL
a. a. a.
CL
CL
x
■
BB
3
■8
■g
"g
%
*8 -g "8 *8
1 -g "g
T3 "9 •=
1 --.
"g
s
E
E
E
B
E £ E E
E E E
E B 2
E
E
c
■
«
a
z
O
U
U
y
u u y
U u
U y (J
CL i 1
O
u
•>^
CL
CL
CL
a. a. a. a.
a. a. a.
CL
CL M M M M
e
■5 >
1- a.
iS
* ~ * ^
•55
>
^
<
<
<
5 5^5
< < <
< < 1
<
<
s s s si -
J£ iS 2 ■§ Ji £
Z
z
Z
2
Z Z Z Z
z z z
Z Z s
z
2
3
■
e
<35
J
5
<
<
<
<
< < ;< <
< < <
< < <
<
<
z
z
z
z
z z z z
z z z
z z z
z
z
O -0 5
h m oo
* t
J! -5 .5
I ll
s
*:
O **
1*1
£f
1
? E
■B
u
£
3
5
>
||
1 C
"1 s
— *"
3 -s 1 s « 1
■* * »
>. ■*"
*" *fi S 2? = ^
ff. — rM
R
,t
^
^
vO 3fc
*t * :»
^
^
S 3 2 T .£ -S
>fi >n ^* r^i *
— O «r»
1-1 O r* **4 no
1
a.
£
1
2
s
s
^:
u.
2
s
<
•g
a
is
"8
c c
c
1
JC
■0
1 a
1 1
1 S
i 1
3
C
8
£
hearualcr -
nidc.ii
orthern 1 u 1 it
:orm petrel
ull • unidnci
-a tt
= .5
3 E = 3
1:1 J fi
seabirds unid
mammals
harbor seal
Stellersea lie
1
3
!
ing crab
anncrcrab
low crab
ihcr crab
round fish
) "S
■a
■n a c ■ «
-0 _* X oo oO
CL
X
JC l- B O OB
.5
|
5
1
'8
1
a
5 s
S
9
B* ■£
a
I
2
1 i
2
1
J '
■
e/i
<
•>
S
i
a
-
ft
(/)
1
U A
e
3
sli
ill
< H 1=
B9
c;
Sj
*
—
■J
-
joiiuow
r«-i m M ci
"
3 = o
o o o
e o ©
=
I
so
lU9UI3f(JlUI
B
ti
f
«
j
u z
-
« « « «
- - - -
O
=
M
e
J
X
<
c
«-4 CM fvl CM
^S
-,.
■3AI)IUJ3||V
Ci r-i rn M
n m m w
_ _ _
_ _ _
_ — —
— —
©
©
1
(■I1U310J
o o o o X
o o o o X
o o o
o © o
© ©
x
=
X
O
r»»
..ijnkV.i^M 1U9JJH3
e
1
M
II
fad ™
II
-
©
©
©
©
tx
m f^
©
iplMig
to apniiulliui
If
c a
o
5 ■=
5
a
°
o o o o o
o o o o o
o o o
13 13 O
o o o
O O
o
saass
sssss
asa
a a a
a a s
a a
a
5
s
f
X
g
5
w
o o 5 o S
o o S o S
6 o o
6 o o
o o o
J J
J
o
j
S
1
g
IS
3
1
z
ifl
o.
Q.
Ou
&
Q.
„ „
„
w
M
e
M
If
> > > « >
c
t»
O O O > O J;
JS ^ JS -S JS s
5 S J J J n
■S 1 JS ^ JS £
o o i i
■§ 1 5 5
— — — 2
S S S
1
5 S
e
|
J! I J!
J! JS
=
3
c
J5
J
5
1
5 (2 3 £ s
s
* fc
e>
C*
O*
o*
*
©
_
M
o
K
M
fr
|
~™
j
.9
s
E
«
>
o
o
*••*
ae
1 £
1 ll
"
3
-
• w
* r- =«
r* Os %
:» =tt
*
§• s J fl
fS
1-1 *-. ■■>* it « =«t
^p — rn ** — t
O N - - (N
00
00 © 3* ■«• 06
■* © fN — O*
3 s^ i
=
1
•
J
e
<
s
3i
|
f*l
*?
*?
.?
O
f
J
%
a*
tt
&4
£
a
b2
3
<
<
<
<
0)
<
J.
<
<
J
■g
8
rtb-C.
Tanner era
king crab
snow crab
other crab
groundfish
snow crab
Tanner era
king crab
other crab
groundfish
ri
C lanncri
king crab
other crab
C angulatu
king crab
other crab
onchair
horsehair
king crab
C- bairdi
C. opilio
king crab
Tanner era
o.
E
■c
a
■
I
l
i
M
I
c
■
H
1
&
■
■
M
S
B
e
8
||
1 =
|
'5
e
n
o
I
H
H
i-
h-
b
a
a
B
H
£
2
2
2
2
2
2
.
a
C/5
<
■c
*.
=
=
=
=
g
-
-
o>
I
3
O
!
E
«
tl
1
a
•j c
< IZ
<
<
t/j
t -■ ^ r- ac
^
B
«
*
— — — — ~
s
21
£3
r»
JO)IUO|^
•c
lu.-iuio|cluj|
1
y
4)
• 5
3
m
II
H
- - -
fiC?
U w
M
■
1
<
c
_
f3AI)BUJ31|V
■<
e
f
III1U310J
o o o
X
X
•
X
X
.
X
X
.
X
X
•
X
X
•
x'
f»
V3jnc«3|^ lu.ujn/)
E
t,
_ _ —
o
o
_
a
_
©
o
—
©
©
_.
©
o
_
©
1
2 s
3 "
M
— O-
; e
(■J ™
V)
o
O
o
o
©
©
©
©
©
©
&
o
o
o
o
e
©
o
©
o
©
©
©
©
©
©
©
©
©
©
©
©
_
ipjHM'g
io 9pn]iu3i|^[
J -
s 1
© W
§1
5
£
X
8
ft
5
■
"g
©
_o
_o
e
-
—
o
j2
_o
_s
J
^
^
i
J
_s
i
i
£
l£
3
I
z
n
e
■M
if
m
55
*©
3
1 1 1 i
5
£
3 i
5
5
S 5
8
c
s
1 1
1
1
i 3
fS c
|
1
J s
|
M
e
55
J
5
J5 ^ ^ 8
_
_
_ _
__
_
— —
__
—
_. _
\o
""
3
*~
"— ■—
<—
Im
■— i£
*■
e
>^1
o
m
11
N
as
SO
2
2
SO
s i
> .«
* e.
s
—
so
m
©
00
Os
«
M
os
r-.'
s
*o
t
Z
»-.
|
S
£
>
f^
3 s
r*»
vO
so
1
«
■*
■*. £
o\
E P
s
■*■ ""
10
O
ss
s
>5
Ov
3
n
©*
—
s
£
3
> ~
1
R
n.
o
5
sO*
*©
s
«
2
2
c c c
B
c
c
^
*
e
n
c
_ ^
1
'S
©
«>
,3
9
J3
|
J
e
c
o
|
J
I
J3
9
U
<
<
<
<
<
z
(la
"0
*? s
^ a
t a
f a
^ s
M
3 a
u 1
^ I
^ i
^ 5
<
< Co
< C/l
< to
< <«
< (/J
|
[
(A
Q.
D.
d.
Gl
£
* 1 1
B
c
c
c
O
o
aa
halib
sjblc
grout
a
a
j
-
r
M
s
s
a
e
a
c
J
«
a
J
3
I
a
5
&
S
£
a
■M
V
V
I
M
X
£
s
e
e
E
3
E
3
E
3
1
M
■M
e
4
e
|
1
I
£
£
e.
is
0.
ig
(/J
a.
m
X
4
'c.
o
5
o
I
E
a
1
_j
u
CD
z ' ^'
U
CO
z 4
u
CO
Z =J
u
CD
z" =j
U
m
z =;'
o
-j
3
a
3
£§
a
3
2§
s
D
Si
s
Z>
2S
s
to
2g
a
5
CA
<
0>
Z
g
_
•c
=
=.
§
5
3
3
*!l(lui)|V
<
|«!IU310J
}
*•»
njnnrei^ luajjnj
X
X
X
8
V
©
_
©
1
is
1^
u —
V)
°
©
&
©
©
©
_
ipiuig
io »pniiu2ahi
h
£ J
j 5
t
X
I
OB
_o
_o
•
j
If
i
1
9
z
a
^
M
M
•
«s
(A
i
g
8
•
I
C
C
*
1
5)
.u
2
1
1
J3
3
*
If
> t
* °"
i
M
©
0*
i
E
■a
>
1
s
p ~
:»
■0
• —
>
—
a
!_^
1
e
3
o
t/>
a.
J3
|
w
E
<
J
■6
B
ffl
C
2
B
1
M
3
i
e
■c
C
■c
*
1
1
i
1
OQ
3
Z O
22
BQ
3
■
tn
<
O^ Urn Jl
—
o
ai
■a
u
jj:
u
a
o
3
E
o
<
u.
o^
Z ■=
2 e
a. 2
>. j ~
E U
a> — -o -^
■/IE— S •-
1 S I
I g o
i £
M
§ ■§
'•3 "3 -o
C o 53
lii
3
X
<
U-
t>
1
te
p
«3
p
(=
a,
"3
e
i
«
C
L_
Jtt
03
-n
n
a
h
^
n
<-
,
m
i
M
&
3
O
03
3
5b
u
a
<
2
4J
^
-a
S3 i
2 — •- «
.EEC
3 r^, -**
* 2
■c •£
= 3
— V m
■c
^
ra
E
F
^3
Ji
IB
O-
S
|
<
D
<
o ,_
8 I
f g
u c
5 a
a § ■§ f-
5 & S -5 JP e
= T3 T3
S e V
a
8
<
-*
€
o
c
5
o
§
1
^2 J
s s
f 6
a -
9- o
i *
u
"3
Ed
fid
O
(J
41
S
o
«
-^
s
3*
U
ft
^
>/>
3
V
^
C/3
«
3
E
P
JS,
B
3
i
o
■a
B
E
_;
U.
tc
60
c
o
T3
c
C
E
2
c
E
E •£
1 .B
o. ,„
Q 8
Os <
2 8
« *- r;
5 I
00 ~
.s 1 S
E Q
M _ Q. Q.
i -5
i i
t z
> a
-= -= -
n
>
1
2
6
3
A
1
i
c
a-
ft
y
eg
§
o
2
4»
a.
^ £
3 S >
« '3 «
(S.6vOdN^rC S*^^
|2
Q E
t: o
u. F- ^ —
■n
tuMes
A conceptual framework is used in Chapter 2 to explore the nature and source of the multidimen-
sional bycatch problem. This appendix contains a more complete discussion of and conclusions
from that exploration. In addition, empirical assessments, in the form of three case studies, are used
to reinforce some of the conclusions from the conceptual framework and to identify some of the
types of information required to address the bycatch issues.
Conceptual Framework
One way to frame the bycatch issue is to answer the following five questions:What is bycatch?
Why does bycatch occur? When is bycatch a problem? What is the appropriate level of bycatch
mortality? Why is there often excessive bycatch mortality?
What Is Bycatch?
For the purposes of this plan, bycatch is defined as fishery discards and unobserved mortalities
resulting from commercial, recreational, and subsistence fishing. Bycatch mortality is bycatch minus
the discards that survive the rigors of their encounter with fishing gear or of being caught and
released.
Why Does Bycatch Occur?
Bycatch occurs if the fishing method used is not perfectly selective. A fishing method is per-
fectly selective if it results in the catch of the desired size, sex, quality, and quantity of the target
species, without causing other fishing-related mortality. Although bycatch rates often can be de-
creased by changing fishing methods, very few fishing methods are perfectly selective. In a commer-
cial or subsistence fishery, bycatch mortality is a by-product of catching fish that are retained. In a
recreational fishery, bycatch mortality is a by-product either of catching fish that are retained or of
catching and releasing fish. Therefore, bycatch is a by-product and a source of fishing mortality for
the bycatch species.
When Is Bycatch a Problem?
Bycatch mortality is a management problem if a lack of information on the level of bycatch
mortality increases substantially the uncertainty concerning total fishing mortality, or if it precludes
a use that would provide greater overall net benefit to the nation. The precluded uses include: (1)
later harvest as target catch in the same or in a different commercial, recreational, or subsistence
fishery; (2) later harvest as bycatch in another fishery; (3) remaining in the sea to contribute to the
ecosystem; and (4) being available for viewing or other nonconsumptive uses.
In the case of the bycatch of dolphins in the Eastern Tropical Pacific tuna fishery, Congress
acted to ensure that dolphin bycatch mortality would be reduced to an insignificant level. This
action reflects an implicit determination by Congress that the benefit of this reduction, principally
in terms of ecosystem and nonconsumptive uses, would exceed the costs that it would impose on the U.S. tuna
fleet and on U.S. tuna processors and consumers, and that the action would be beneficial to the nation.
Because bycatch mortality is a by-product of fishing, it usually cannot be reduced in a fishery
without either reducing the amount and benefit of the catch or increasing operating and manage-
ment costs in that fishery For example, bycatch reduction devices may reduce catch per unit of
effort and, therefore, may decrease the catch of the target species. This means that operating costs per
unit of catch would increase due to the cost of the gear modifications as well as the decrease in catch
per unit of effort. The net benefit of using fish and other living marine resources as bycatch in a
fishery is determined by the reduction in the benefit of the harvest of that fishery and the increase
in the cost of operating in that fishery that would be required to eliminate bycatch mortality. If
bycatch mortality could be decreased without decreasing the difference between the benefit of the
harvest and the cost of operating in a fishery, bycatch mortality would not be a contentious manage-
ment problem; it would simply be eliminated.
What Is the Appropriate Level of Bycatch Mortality?
From a national perspective, too much bycatch mortality exists in a fishery if a further reduc-
tion in bycatch mortality would increase the overall net benefit of that fishery to the nation through
alternatives uses of the bycatch species. In that case, it is practicable to reduce bycatch mortality and
the excess bycatch mortality is a wasteful use of living marine resources. Conversely, if a reduction in
bycatch mortality would not increase the overall net benefit to the nation, there is not too much
bycatch mortality and bycatch mortality is not precluding better uses of resources. Reducing bycatch
mortality in a cost-effective manner until a further reduction would not increase overall net benefit
to the nation is equivalent to minimizing the cost to the nation of the bycatch problem, which is the
sum of the cost of the bycatch itself and the cost of reducing bycatch. In many cases, it may be
possible but not practicable to eliminate all bycatch and bycatch mortality in a fishery.
Bycatch mortality can be reduced by changing how, when, where, and how much fish are
caught, what is discarded, and how fish and other bycatch species are handled before being dis-
carded. Such changes can have desirable nation undesirable effects for the individual fishermen who
reduce their bycatch mortality and for the Nation as a whole. Those effects determine if a further
reduction in bycatch mortality would increase the overall net benefit of that fishery to the nation.
The effects include the following: (1) changes in the bycatch mortality of the species for which a
reduction is the objective; (2) resulting population effects for the bycatch species; (3) ecological
effects due to changes in the bycatch of that or those species; (4) changes in the bycatch of other
species of fish and the resulting population and ecosystem effects; (5) changes in the incidental
catches of marine mammals and birds and the resulting population and ecosystem effects; (6) changes
in fishing, processing, disposal, and marketing costs; (7) changes in the economic, social, or cultural
value of fishing activities and nonconsumptive uses of fishery resources; (8) changes in the effective-
ness and cost of research, management, enforcement, and information exchange programs; and (9)
the distributional effects of the preceding types of effects.
Case Studies
m
The probability that a further reduction in bycatch will increase net benefit to the nation is
decreased if the methods used to reduce bycatch mortality are not cost-effective. The methods are
not cost-effective if the cost of achieving a given reduction in bycatch mortality can be decreased by
any of the following: (1) having a fisherman use a lower-cost technique to attain a given reduction
in its bycatch mortality; (2) changing the distribution of effort to decrease bycatch mortality among
the vessels in a fishery; and (3) changing the distribution of effort to decrease bycatch mortality
among fisheries.
Cost-benefit analysis (CBA) is the primary tool used to determine net benefits to the nation generated
by the exploitation of domestic living marine resources and their distribution among different user
groups. The economic theory of the firm and consumer theory are the theoretical foundations for CBA.
Using this scientific approach, the fishery manager is faced with the problem of maximizing net benefits
from exploiting a stock or stocks offish by different user groups — e.g., commercial, recreational, and
subsidence fishermen as well as other consumptive and nonconsumptive users of the resource.
Net benefits received by the commercial harvesting sector consist of profits in excess of a normal
return for fishing. Known as producer surplus, these net benefits are the remainder of total revenue
minus harvest costs and a fair return to the owner of the harvesting equipment.
Consumers of commercial fishery products also receive benefits in the form of the difference between
the purchase price of the fish and what they would be willing to pay forthe fish — i.e. consumer surplus.
For example, if an individual were willing to pay $10 for a pound offish but the market price were$1, a
net benefit of $9 per pound in consumer surplus would exist.
Recreational fishermen receive satisfaction from taking fishing trips and presumably catching fish on
these trips. The value of a recreational fishing trip can be measured using nonmarket valuation
techniques based on, for example, the costs of taking the trip. If the number of fish caught on a trip
declines or the cost of taking the fishing trip increases, demand for fishing trips declines and the net
benefits received by recreational fishermen from fishing decline. Similarly, subsistence fishermen
receive satisfaction from consuming fish they have caught for their own use. Although not sold, the
value of those fish can also be determined using nonmarket valuation techniques if the underlying
objectives of the subsistence fishermen are understood — e.g. if the subsistence fishing is for tradi-
tional, cultural, or religious reasons.
Nonconsumptive users also value living marine resources because they know the resources exist or
they enjoy or get satisfaction from viewing them. Scuba divers and snorkelers, for example, realize
greater net benefits if reefs are heavily populated by fish species than if they are devoid of life. Even
people who never see a reef environment in the sea can value the existence of fish species and will
receive satisfaction just from knowing that the resource is being protected.
Cost-effective analysis is more restrictive than CBA, since it does not consider benefits to individuals or
the nation generated by living marine resources. The value that may exist for a threatened or endan-
gered species in terms of its contribution to biodiversity are presently unknown but possible future
benefits to society have caused Congress to enact the Endangered Species Act to protect and improve
the stock of that species.
The conservation and management measures that are used to reduce bycatch mortality as well
as the overall management regimes of the fisheries in which bycatch occurs will determine whether
cost-effective methods are used to reduce bycatch mortality. The management regimes for those
fisheries and for other fisheries are critical in determining which alternative uses of living marine
resources will increase when bycatch mortality is decreased and the net benefit to the nation of such
increases.
Why Is There Often Excessive Bycatch Mortality?
A widespread perception is that greed or lack of concern by fishermen results in excessive
bycatch mortality. This line of reasoning ignores the decision environment in which individual
commercial, recreational, and subsistence fishermen work. Bycatch mortality results from the fish-
ing practices employed by individuals that are in turn conditional on personal preferences and
prevailing regulatory and economic circumstances. Thus, decisions made by individual fishermen
and fishery managers are interdependent and jointly determine the levels of bycatch mortality.
The decisions of individual fishermen (or processors) tend to result in excess bycatch mortality
if they do not consider fully the net benefit to the nation of reducing bycatch mortality. This can
happen for two reasons. First, the information they have understates the overall net benefit to the
nation of a further reduction in bycatch mortality. This could occur either if they are not aware of
lower-cost techniques that are available to reduce bycatch mortality or if they are not aware of all of
the benefits of reducing bycatch mortality. Second, they do not have sufficient incentives to consider
fully the increase in overall net benefit to the nation of a reduction in bycatch mortality. For ex-
ample, an individual fisherman is more likely to consider the costs and benefits that accrue to him
than those that accrue to others; often a substantial part of the benefits of reducing bycatch mortality
is not captured by the individual fisherman who reduces bycatch mortality.
Most fishery management regimes do not create clearly defined and enforceable property
rights for fish in the sea, which would allow the market mechanism to be used to allocate fish
among fishermen and among competing uses. Instead, fish are allocated to fishermen on a first-
come, first-served basis — that is, the "race for fish" is used as the allocation mechanism. This means
that individual fishermen do not pay for the fish and other living marine resources they use. There-
fore, fishermen have an incentive to use too much fish as bycatch, just as they each would have an
incentive to use too much fuel if fuel were free to them or grossly underpriced. Other undesirable
effects of this allocation mechanism often include overfished stocks, overcapitalization, boom-and-
bust fisheries, and hazardous fishing practices. Management actions that have been taken to address
some of these other symptoms of a flawed allocation mechanism often have further increased the
incentive for fishermen to use fish as bycatch mortality. For example, bycatch mortality often has
been increased by species specific trip limits in multispecies fisheries, inconsistent mesh size and
minimum fish size regulations, pot limits, and TACs that decrease season lengths and increase the
intensity of fisheries. Finally, the strategy of treating the symptoms of the bycatch problem and
related management problems rather than eliminating the cause has resulted in a need to constantly
change conservation and management measures. This has prevented more substantive progress in
dealing with bycatch.
The level of bycatch and the methods used to reduce bycatch are determined by individual
fishermen in response to a variety of incentives and constraints that reflect the economic, social,
regulatory, biological, and physical environments in which they operate. The tendency for the deci-
sions of individual fishermen to result in excessive levels of bycatch can be decreased by providing
better information on the techniques to reduce bycatch mortality and on the benefits of decreasing
bycatch. Ensuring that such information is used in making decisions can be done either by increas-
ing incentives fishermen have to fully consider the information or by restricting the decisions they
can make — that is, by making more decisions for them. With adequate information, either method
can be used to improve bycatch management and, therefore, increase the benefits the nation receives
from fisheries. These methods differ in the types of information needed by fishery managers and
fishermen, as well as the costs of obtaining the information. The information requirement differ-
ences are important factors in determining which method or mix of methods will be more effective
in reducing the bycatch problem in a particular fishery. For example, if the monitoring and enforce-
ment cost of making individual fishing operations accountable for their bycatch mortality is too
high, that method is not viable, and restricting the decisions of fishermen is a more viable solution.
Compliance with regulations is an important factor in determining whether a set of regula-
tions designed, at least in part, to reduce bycatch mortality will be effective in doing so. Involving
fishermen in the development and implementation of fishery regulations can have a substantial
positive effect on compliance. It does this by increasing the ownership fishermen have in the regu-
lations and by having the regulations based more on the understandings of fishermen concerning
the fishery and methods to reduce bycatch mortality.
The quality of decisions made by fishery policy makers and managers also depends on the
information that is available to them and their decision-making processes. Costs or benefits of a
fishery that are not fully considered by policy makers and fishermen can lead to poor management
decisions. Information that would decrease the uncertainty concerning the biological productivity
of stocks offish, the impacts of fishing activities on living marine resources, and the economic and
social impacts of alternative management policies would allow better decisions to be made by policy
makers. Public review of the costs and benefits associated with a fishery and a clear identification of
the objectives for a management policy will help improve the overall quality of management
decisions. The increased involvement by the public also increases the need to ensure that public
opinion is based on accurate information.
Conclusions
The Magnuson— Stevens Act defines the term optimum, with respect to the yield from a fishery,
as the amount of fish that will provide the greatest overall benefit to the nation, particularly with
respect to food production and recreational opportunities, and taking into account the protection of
marine ecosystems. The amount offish is the sum of target catch and bycatch; therefore, the opti-
mum bycatch in a fishery is that which will provide the greatest overall benefit to the nation. Thus,
if a reduction in bycatch mortality will increase overall net benefit to the nation, there is excessive
bycatch mortality, and it is practicable to reduce bycatch mortality. The term practicable is not syn-
onymous with the term possible because not all reductions that are possible are practicable. In many
fisheries, it probably is not practicable to eliminate all bycatch and bycatch mortality.
The extent to which it is practicable to reduce bycatch mortality is not static. Examples of
changes that would tend to increase the extent to which it is practicable to reduce bycatch mortality
include the following: (1) the development of lower-cost methods either of avoiding bycatch or of
increasing the survival rates of discarded catch; (2) changes in biological or oceanographic condi-
tions that make it easier to avoid bycatch; (3) changes in market conditions, in population and
ecosystem conditions, or in fishery regulations that increase the value of the uses of living marine
resources made possible by a reduction in bycatch mortality; (4) changes in fishery regulations that
encourage the development and use of lower-cost methods to decrease bycatch mortality; and (5) a
change in the open-access managed common property resource management paradigm to a rights-
based management institution.
Because neither the extent to which it is practicable to reduce bycatch nor the best methods
for reducing bycatch mortality are static, there is a periodic need to evaluate the merits of existing
and alternative conservation and management measures to reduce bycatch. The evaluation should
be in terms of the population, ecosystem, social, and economic effects, which determine whether
they have increased or are expected to increase net benefit to the nation. They should not be
evaluated only in terms of their effects on the levels of bycatch. A mix of quantitative and qualitative
analyses often will be appropriate for such evaluations. The cost of adequately monitoring the catch
and bycatch of individual fishing operations is critical in determining which of the latter two
methods is more likely to increase the overall net benefit to the nation. Similarly, the decisions of
fishery policy makers and managers can be improved by improving the information they have and
by ensuring that they consider all the significant benefits and costs of reducing bycatch mortality.
The allocation of NMFS and Council resources will be critical in determining their success in
increasing net benefits to the nation by decreasing bycatch mortality. In setting priorities to address
bycatch and other fishery management problems, it is important to do the following: (1) recognize
that physical measures of bycatch are of limited use in comparing the magnitude of the bycatch
problem among fisheries because the expected net benefit to the nation of reducing bycatch is not
the same for all species or even for all fish of the same species; (2) recognize that typically it cannot
be determined if a particular use of living marine resources is wise or wasteful compared to another
use without considering all the costs and the benefits to the nation for the two uses; (3) consider the
expected net benefit to the nation of addressing a specific problem; (4) make a distinction between
the sources and symptoms of a bycatch problem; (5) identify the principal concerns in terms of the
population, ecosystem, social, or economic effects of bycatch; (6) recognize that rather than being a
separable fishery management issue, bycatch management is an integral part of fishery management
and that the ability to improve overall fishery management and bycatch management is limited by
similar decision-making process flaws that include the same information gaps and faulty incentives
for individual fishermen, fishery managers, and other participants in the fisheries and fishery man-
agement process; (7) recognize that much of the information necessary to identify and quantify the
effects of a specific set of changes in research, management, enforcement, and information exchange
programs intended to decrease bycatch mortality is also necessary to address other management
problems, such as determining the appropriate levels of exploitation for living marine resources and
determining how to allocate the associated harvest levels among competing uses and users; (8)
recognize that bycatch is a multispecies problem because actions to decrease the bycatch of one
Case, Studies
species can increase or decrease the by catch of other species and because the by catch of one species
can affect the status of other species, through predator/prey or other biological interactions; and (9)
determine if there are common solutions to multiple management problems that may only be
feasible when the commonality is recognized.
The importance of identifying the principal concern about bycatch in a specific fishery is
demonstrated by the following examples. Uncertainty about total fishing mortality can be decreased
by improving the estimate of bycatch mortality or by decreasing bycatch mortality. If the current
level of total fishing mortality for a species threatens either the population of that species or other
components of the ecosystem and if that threat essentially could be eliminated by decreasing other
sources of fishing mortality that a council and NMFS control, the bycatch itself is not a threat. In this
case, bycatch results in an allocation problem, not a population or ecosystem problem. If the bycatch
problem is principally that the bycatch by one group of fishermen decreases the retained catch and
benefits of another group of fishermen, compensation from the former group to the latter group
may be more beneficial to both groups than a reduction in bycatch mortality.
Information requirements and compliance with regulations are two important factors in determining
whether a set of regulations designed, at least in part, to reduce bycatch mortality will be effective in doing so
and will increase net benefit to the nation.The more uncertain fishery managers are about what they know and
the lower the compliance, the less likely it is that the objectives of the regulations will be met.
The overall management regimes for the fisheries in which a species is taken as bycatch and for
the fisheries in which that species is taken as target catch are important in determining the extent to
which it is practicable to reduce the bycatch of that species. For example, increased bycatch mortal-
ity is one of the effects of using the race for fish to allocate fish among competing fishermen.
Bycatch can be reduced by avoiding bycatch to begin with, by increasing the survival of
discards, or by retaining fish that would normally be discarded. The optimum mix of these three
methods depends on the desirable and undesirable effects of each method. The effects and, therefore,
the optimum mixes are case-specific.
Empirical Assessments
There are three principal reasons why it can be very difficult to determine if a specific set of
actions to reduce bycatch mortality will be beneficial to the nation. First, there can be significant
uncertainty concerning the direction or magnitude of each type of effect. The uncertainty is gener-
ally the result of a limited understandings of the relevant biological, ecological, economic, and social
relationships. Second, even when the effects can be quantified, not all of them can be measured in
common units, such as discounted present value. For example, the distributional effects cannot be
summarized in terms of discounted present value because the assessment of alternative distributions
requires value judgments. Similarly, although the population and ecosystem effects are important
due to their effects on net national benefits, the change in national benefits generated by a specific
population or ecosystem change is difficult to identify. Third, for the effects that can be measured in
terms of discounted present value, the quality of the valuation techniques can differ substantially by
type of effect. For example, better estimates of the change in value may be available for market
goods than for non-market goods.
Some of the types of information required to estimate whether a change in bycatch manage-
ment will tend to increase or decrease the net benefit to the nation are described in three case
studies that were completed as part of the process for developing this plan. The case studies of the
net benefit of bycatch reductions for the Alaska groundfish fishery, the Gulf of Mexico shrimp
fishery, and Southern New England trawl fishery are intended to:
• Provide examples of the types of information that are needed to estimate the net benefit of
reducing bycatch in a specific fishery.
• Emphasize the need to consider the desirable and undesirable effects of reducing bycatch
mortality and indicate that the result of decreasing bycatch can be an increase, decrease, or no
change in the net benefit to the nation.
• Note the importance of both the methods used to reduce bycatch and the overall manage-
ment regime in determining the direction and magnitude of the change in the net benefit to
the nation.
• Identify information gaps.
• Emphasize that rather than being a separable fishery management issue, bycatch management
is an integral part of fishery management, and that the ability to improve overall fishery
management and bycatch management is limited by similar decision-making process flaws
that include the same information gaps and faulty incentives.
Case Studies
Halibut Bycatch in the Alaska Groundfish Fishery
J.M.Terry
Alaska Fisheries Science Center
Seattle, WA
Summary
Halibut bycatch mortality in the Alaska groundfish fishery has been a contentious issue for
many years because it decreases the halibut fishery quota and because it cannot be decreased with-
out imposing costs on groundfish fishermen. This case study compares the benefits of reducing
halibut bycatch mortality in the Alaska groundfish fishery and the estimated cost to groundfish
fishermen of three methods of decreasing groundfish catch to reduce halibut bycatch mortality.
The most selective reduction in groundfish catch considered was the most likely to result in
the benefits to the halibut fishermen exceeding the cost to the groundfish fishermen. With the most
selective of the three methods considered, that happened if the ex- vessel value net of fishing costs is
less than 44% of the ex-vessel value of groundfish. However, with the least selective method, that
happened only if the ex-vessel value net of fishing costs is less than 3% of the ex-vessel value of
groundfish. Generally as more selective methods are used to reduce halibut bycatch mortality, the
cost to groundfish fishermen is decreased and larger reductions in bycatch mortality are practicable.
Market-based solutions to the halibut bycatch problem that provide bycatch accountability by indi-
vidual fishing operations, such as individual bycatch quotas, tend to result in the most selective
methods being used. However, the monitoring and enforcement costs may be substantially greater
for such programs. The existence of an extensive at-sea observer program for the groundfish fishery
makes such a solution more feasible.
Background
Until recently, the bycatch issue in the Alaska groundfish fishery that received the most atten-
tion was the bycatch of halibut, salmon, crab, and herring. These are relatively high-valued species
for which the domestic fisheries had been fully developed well before the domestic groundfish
fishery replaced the foreign and joint- venture groundfish fisheries. This bycatch decreased the catch
that was available to domestic halibut, salmon, crab, and herring fishermen. Retention of these
species is prohibited in the groundfish fishery (i.e., these are "prohibited species" in the groundfish
fishery), and a variety of other measures have been used to control the bycatch of these species. This
case study compares the benefit to the halibut fishery and the cost to the groundfish fishery of three
method of reducing halibut bycatch mortality by decreasing groundfish catch.
Benefit of Decreasing Halibut Bycatch Mortality
The use of halibut that would be increased by a decrease in halibut bycatch mortality in the
groundfish fisheries has been made explicit by the International Pacific Halibut Commission (IPHC),
which establishes the annual quotas for the halibut fisheries. In recent years, the IPHC has estab-
lished quotas for total halibut removals in all fisheries combined, then subtracted expected bycatch
removals and removals in all fisheries except the commercial halibut fishery, and set the quota for
that fishery equal to the residual. The expected halibut bycatch mortality in the groundfish fishery
was set equal to the estimated bycatch mortality from the previous year. Therefore, for each metric
ton (nit) of estimated bycatch mortality in the groundfish fishery one year, there was a 1-mt reduc-
tion in the halibut fishery quota the next year. This reduction in the halibut fishery quota was made
in an attempt to prevent the bycatch from adversely affecting the long-term biological productivity
of the halibut stocks.
In addition to this immediate adjustment to the halibut fishery quota, the IPHC estimated the
long-term yield loss in the halibut fishery per metric ton of bycatch mortality in the groundfish
fishery. The current estimate is 1.8 mt. That is, based on a population dynamics model and the
current exploitation rate strategy, it is estimated that the cumulative effect of each metric ton of
halibut bycatch mortality in the groundfish fishery is a 1.8-mt reduction in halibut fishery
quotas over a 20-year period. Although new quota adjustment methods and yield loss esti-
mates are being prepared by the IPHC, the current quota adjustment and yield loss esti-
mate are used below in estimating the benefit of reducing halibut bycatch mortality in the
groundfish fishery.
The IPHC estimates provide two important pieces of information concerning the benefit of
reducing halibut bycatch mortality. They indicate what alternative use would increase and by how
much. Specifically, catch in the halibut fishery would increase by 1 mt the next year and by 1.8 mt
over the next 20 years. When a discount rate of 5% is used to account for the fact that a 1-mt
increase in the halibut quota several years from now is not comparable to an immediate 1-mt
increase, the discounted yield loss is about 1.5 mt per 1 mt of halibut bycatch mortality.
Since 1995, the halibut fishery off Alaska has been managed under an individual transferable
quota program. The price that fishermen are willing to pay for 1 mt of halibut quota for one year
provides an estimate of the net benefit to halibut fishermen of additional halibut quota. The quota
price is about $1 per pound or about $2,205 per metric ton net weight, which is 75% of the round
weight. The quota price is about half of the ex-vessel price of halibut, which suggests that the
marginal harvesting cost is about half of the ex-vessel value of halibut. Therefore, using $2,205 as the
net benefit per metric ton of net weight, the round- to net-weight conversion factor of 0.75, and a
discounted yield loss of 1.5, the estimated increase in net benefit to halibut fishermen per 1-mt
reduction in halibut bycatch mortality in the groundfish fishery is $2, 481. Therefore, given that the
halibut bycatch mortality in the Alaska groundfish fishery was about 6,720 mt in 1995, the potential
benefit to halibut fishermen of the elimination of halibut bycatch in the groundfish fishery would
have been about $16.7 million.
Developing this estimate of the potential benefit of eliminating halibut bycatch mortality in
the Alaska groundfish fishery was greatly simplified by (1) the use of individual transferable quota
management of the halibut fishery, which provides a market-based estimate of the net benefit of
additional halibut quota; (2) a halibut allocation system, which makes it clear what the alternative
use of halibut will be; (3) a halibut stock and management model that provides an estimate of the
halibut fishery yield loss due to halibut bycatch mortality in the groundfish fishery; and (4) extensive
at-sea observer and groundfish catch reporting systems, which provide a generally well-accepted
estimate of halibut bycatch mortality.
This estimate tends to overstate the net benefit to commercial halibut fishermen because it
does not allow for the downward ex-vessel price adjustment that would accompany an increase in
halibut landings, but it tends to understate the benefits to the nation as a whole because it excludes
any benefits beyond the ex-vessel level. An estimate of the ex-vessel demand for halibut could be
used to account for the price effect and benefits beyond the ex-vessel level. In the absence of an
estimate of the ex-vessel demand function, it is not known whether the estimate of $2,481 is higher
or lower than the actual benefit to the nation of a 1-mt reduction in halibut bycatch mortality in the
Alaska groundfish fishery.
Net Benefit of Decreasing Halibut Bycatch Mortality
Assuming that the benefit of reducing halibut bycatch mortality by 1 mt is $2,481, net benefits
can be increased by decreasing bycatch as long as the cost of reducing bycatch is less than $2,481 per
metric ton. Unfortunately, with the exception of the extreme case in which halibut bycatch mortal-
ity is reduced by decreasing groundfish catch, the cost of reducing bycatch is difficult to estimate. In
the extreme case, halibut bycatch mortality can be decreased with a proportionate decrease in the
catch of all groundfish species. Given the halibut bycatch mortality of 6,720 mt and a groundfish ex-
vessel value of $585 million, this would result in about a $87,000 reduction in groundfish ex-vessel
value per 1-mt reduction in halibut bycatch mortality. Therefore, if the ex-vessel value net of fishing
costs is less than 3% of the ex-vessel value of groundfish, the benefit of the reduction in halibut
bycatch mortality would exceed the cost through the ex-vessel level.
The cost of reducing halibut bycatch mortality can be decreased with a more selective reduc-
tion in groundfish catch. For example, the 1995 Bering Sea cod trawl fishery had an ex-
vessel value of about $28 million and accounted for about 1,512 mt of halibut bycatch mortal-
ity. Therefore, with a proportionate decrease in the catch of all cod trawl fishing operations, there
would be about a $18,500 reduction in ex- vessel value for each 1-mt reduction in halibut bycatch
mortality. If the ex- vessel value net of fishing costs is less than 13% of the ex-vessel value of ground-
fish, the benefit of the reduction in halibut bycatch mortality would exceed the cost through the ex-
vessel level.
An even more selective reduction in the catch in the Bering Sea cod trawl fishery would
produce a greater reduction in the cost of decreasing halibut bycatch mortality. For example, if
halibut bycatch mortality is reduced by 20% by eliminating the catch of the fishing operations with
the highest ratios of halibut bycatch mortality to retained cod catch, halibut bycatch mortality
would be reduced by about 302 mt, and the ex- vessel value of the cod fishery would be reduced by
about $1.7 million or about $5,600 per 1-mt of reduction in halibut bycatch mortality. Therefore, if
the ex- vessel value net of fishing costs is less than 44% of the ex-vessel value of groundfish, the
benefit of the reduction in halibut bycatch mortality would be more than the cost through the ex-
vessel level.
Conclusions
Among these three examples, the expectation that the benefit of reducing halibut bycatch
mortality would exceed the cost increased as the method of reducing bycatch became more selec-
tive. Given this example in which the benefit of a 1-mt reduction in bycatch mortality is $2,481, the
optimum situation is that in which each fishing operation reduces its halibut bycatch mortality to
the point at which its cost to reduce bycatch mortality by another 1 mt is also $2, 481. When that
condition is met, the net benefits from the fishery cannot be increased by either changing the total
level of bycatch or by changing the distribution of bycatch among fishing operations. This most
selective method of reducing bycatch could be attained if each fishing operation had to pay $2,481
per metric ton of halibut bycatch mortality. Other mechanisms for inducing reductions in bycatch
are generally less selective, will have higher costs to the groundfish fishery, but may have substantially
lower monitoring and enforcement costs.
The preceding discussion of the benefits and costs of decreasing halibut bycatch mortality in
the groundfish fishery excludes any discussion of the changes in the distribution of benefits and
costs. Such changes are clearly important in determining if a reduction in halibut bycatch mortality
will benefit the nation. If the use of halibut as bycatch is justified in terms of net benefits, but results
in an undesirable change in the distribution of benefits and costs, the gain in net benefits must be
weighed against the adverse change in their distribution. The determination of whether a specific
change in the distribution of benefits and costs is in itself desirable or undesirable for the nation and
the determination as to whether an increase in net benefits is more than offset by an undesirable
distribution change require value judgments. The value judgments used are implicit in the decisions
of those who determine whether a specific management measure will be implemented.
Case, Studies
m
Economics of Bycatch: The Case of Shrimp and
Red Snapper Fisheries in the U.S. Gulf of Mexico
R. C. Raulerson
Southeast Regional Office
St. Petersburg, FL
J.R. Waters
Southeast Regional Office
Beaufort, NC
Summary
One of the more challenging fishery management issues in the Southeast concerns the inci-
dental bycatch of juvenile red snapper by shrimp trawlers in the U.S. Gulf of Mexico. In the absence
of bycatch reduction, directed red snapper catches will continue at their current small annual levels
and the stock will not recover.
Analyses of potential economic outcomes of alternative ways to reduce the bycatch required
information primarily on the current harvest sector costs for shrimp and red snapper and on the
demand for the target and bycatch species. The analyses indicated that conservation of the red
snapper resources via a reduction in bycatch will result in significant losses in producer surplus
generated by shrimp trawling. Furthermore, while a reduction in red snapper bycatch could lead to
recovery of the red snapper stock, the potential benefits would not be realized unless the commer-
cial red snapper fishery was managed in an optimal fashion and unless the recreational catches could
be constrained within their quotas.
An economic and technical evaluation of alternatives for reducing red snapper bycatch con-
cluded that if bycatch reduction was to be accomplished, it would be via the use of bycatch reduc-
tion devices (BRDs) in shrimp trawls. Further analyses determined the particular BRD design that
would meet the bycatch requirements at the lowest cost relative to other BRD designs. The out-
come was that the cumulative discounted net value from shrimp harvesting would decrease by $1 17
million for the period from the inception of required BRD use until a new harvest equilibrium in
the shrimp fishery was reached. The evaluation also concluded that benefits from red snapper stocks
could approach the level of the estimated shrimp losses, but only if the red snapper stocks are
managed under an individual transferable quota program or some other management regime that
would produce the same results.
The case of red snapper and shrimp in the Gulf of Mexico provides an excellent illustration
that bycatch reduction does not necessarily provide for increases in net economic benefits and also
highlights the critical role that the management regimes for the target and bycatch species will play
in the overall, long-term economic changes that may result from reductions in bycatch for any
fishery.
Economic Consequences
A major finding of the RIR was that although a reduction in red snapper bycatch had the
potential to allow for recovery of the red snapper stock, the potential benefits would not be realized
unless the commercial red snapper fishery was managed in an optimal fashion and unless the recre-
ational catches could be controlled. Optimal management of the red snapper fishery was portrayed
in the RIR as an ITQ management system, and it is important to note that the Council developed,
and the Secretary of Commerce approved, an ITQ system for red snapper. In addition, the Council
has constrained future recreational catches by implementing an overall recreational quota.
While there was an a priori theoretical basis upon which to predict that benefits would not
accrue to the red snapper fishery in the absence of an optimal management system for red snapper,
and that the effects on the shrimp fishery would be negative, the RIR proceeded on the assumption
that the ITQ system for red snapper would be implemented and that recreational catch would be
controlled. Given these assumptions, the overall outcome of the RIR was dependent on empirical
estimates that compared the level of shrimp losses to the potential gains to the red snapper fishery.
While there may be instances where the economic outcome of bycatch reduction is so certain that
declarations can be made as to whether net benefits will increase, decrease, or even remain about the
same, such is not the case for this example, and this example is probably not particularly unique.
Impacts on Shrimp Fishery
Management alternatives considered under the amendment included area closures, seasonal
closures, and the use of bycatch reduction devices (BRDs) in shrimp trawls. Although the alternative
ways to reduce bycatch have different cost and revenue implications for the shrimp harvest industry,
they have roughly identical implications for the directed fisheries for the bycatch species because all
the alternatives must reduce bycatch by 44%. It is important to note that this criterion was based
strictly on biological grounds, without knowledge of the magnitude of the economic consequences
of a bycatch reduction of this level.
As a step toward determining whether to implement bycatch reduction, the Council con-
ducted an economic and technical evaluation of alternatives. This process resulted in a decision that
if bycatch reduction was to be accomplished, it would be via the use of BRDs in shrimp trawls. This
outcome was based on a finding that the alternative of requiring BRDs was superior to the alterna-
tives of area or seasonal closures in terms of the costs to shrimpers and the feasibility of meeting the
44% bycatch reduction criterion while the effects on the red snapper stocks would be similar to
other alternatives. For this reason, the balance of the economic analysis focused on the BRD alternative.
Even though that alternative was chosen as the preferred alternative on the basis of being the
least costly, a complicating factor is that several BRDs met the technical criteria of a bycatch reduc-
tion of 44%. Because it was known from the outset that they would perform differently in terms of
the economics of shrimp harvesting, a separate economic analysis was conducted for all three BRDs
that were deemed to meet the bycatch reduction technical requirement. The model results clearly
indicated that the major factor providing for an economic differentiation among the BRDs is that
they lose differing amounts of shrimp and hence have significantly different outcomes in terms of
short-term effects on shrimp harvest, on the resources used to harvest shrimp, on harvesting costs,
and therefore, on profitability.
Case, Studies
• ■■ m Wm&b..
The long-term differences are more complex, because the long term depends largely on the
reaction of overall shrimp harvesting effort in response to the short-term effects of management.
Understanding that the shrimp harvesting industry is open access, and further understanding that
the year-to-year variation in the shrimp resource is independent of previous harvesting effort, the
models used indicate that overall long-term effort reductions result in long-term industry benefits
that tend to offset part of the short-term losses. It should be noted that this finding does not imply
an absence of "losers" when shrimp effort declines. Indeed, the offsetting long-term benefits are
attributed mainly to the exit of marginal firms.
The RIR determined that one particular BRD would meet the bycatch requirements at a
minimal cost relative to other BRD designs. The outcome was that the cumulative discounted net
value from shrimp harvesting would decrease by $1 17 million for the period from the inception of
required BRD use until a new harvest equilibrium in the shrimp fishery was reached. There was an
additional refinement in the BRD analysis could not be considered and that was the possible con-
sideration of a number of BRDs being used by different fisherman or in different areas.
Although it is clear that the introduction of BRDs will result in varying outcomes according
to which style of BRD is chosen, it is highly likely that several BRDs will be used. Since there is no
information available to suggest which approved BRD will be chosen by individual shrimpers (they
will make individual choices according to their particular shrimping strategy and geographical area),
the indicated refinements to the analysis are not possible. Using only one style of BRD as an
example, the shrimp models indicated that a device called the "30-mesh fisheye" would reduce
overall shrimp catches and, hence, revenues by 3% if all shrimpers used that style of BRD. Consid-
ering the costs of purchasing the BRDs, and the fact that profits as a percent of revenues are small for
shrimp harvesting firms, it was fairly straightforward that the average shrimp vessel would incur an
annual short-term profit loss of significantly over 3%.
The economic models also considered the reaction of the effort response of the shrimp har-
vesting industry to the loss in short-term profits. This information came from an entry-exit model
and a within season effort model that in combination described the expected change in shrimp
harvesting behavior. The result was that total effort, in terms of overall fleet size, will decline in
response to the decreased shrimp catches and increased costs of purchasing and maintaining BRDs.
The models indicated that when a new harvesting equilibrium was reached, then the overall reduc-
tion in effort tended to reduce overall costs and, hence, tended to produce long-term economic
gains that offset, to some degree, the short-term costs to shrimpers.
Impacts on Red Snapper Fishery
The RIR also concluded that benefits from red snapper stocks could approach the level of the
estimated shrimp losses, but only if the red snapper stocks were managed under the approved ITQ
program or some other management regime that would produce the same results. Since the
Magnuson— Stevens Act has subsequently imposed a moratorium on ITQs until the year 2000, and
since equally effective management regimes have not been discovered, the hypothesized benefits
cannot be realized at this time. It is worth repeating that the case of red snapper and shrimp in the
Gulf of Mexico provides an excellent illustration that bycatch reduction does not necessarily pro-
vide for increases in net economic benefits and also highlights the critical role that the management
regimes for the target and bycatch species will play in the overall, long-term economic changes that
may result from reductions in bycatch for any fishery.
Given the result regarding potential losses to the shrimp harvesting industry, there would need
to be a larger positive change in net benefits to the commercial and recreational users of the red
snapper resource if the RIR test of benefits exceeding costs is to be met. A problematical issue on
the benefits side is that the red snapper fishery is managed on a constant catch basis, versus a constant
fishing mortality rate basis, so some of the gains cannot begin accruing until the year 2019, which is
the projected stock recovery time. In the interim, the red snapper stock size will be increasing, so
there may be a tendency for costs per unit of catch to fall somewhat and that would seem to signal
an increase in benefits. However, as a counter situation, the fishery operates under a quasi license
limitation program with an overall quota, trip limits, and other restrictions. This management sys-
tem produces the traditional derby fishery, and as red snapper stocks increase, there will be expected
decreases in revenues for a given level ofTAC.The revenues decrease because TAC (quota) does not
increase, but the season would become shorter due to increased catchability.The average annual red
snapper price will be expected to fall as it has in other recent years when the derby fishery intensi-
fied. Hence, in the period preceding optimal management of the red snapper fishery, even the short-
term overall effects on benefits to the commercial users of the resource are uncertain or negative.
There is a quota in effect for the recreational users, and the recreational fishery is to be closed
when the quota is reached. From an economics standpoint, the truncation of recreational fishing
years would create losses in the red snapper recreational fishery and likely would create additional
problems as the recreational effort moves to other species. Furthermore, red snapper are part of a
mixed catch, and a continuing mortality of red snapper from discarded recreational bycatch will be
present for the balance of the fishing year. The possibility of a recreational closure would also likely
create something like a derby recreational fishery, especially since red snapper are a highly sought
species and are pursued in particular by the for-hire recreational sector. The Council could elect to
reduce bag limits or take other actions in an attempt to ensure that the fishery does not close during
any given season, but such actions would also tend to decrease the values obtained from recreational
fishing. Hence, short-term benefits to recreational users will not exist.
For commercial and recreational users alike, the red snapper TAC would be increased to some
sort of optimum yield level in the year 2019, the expected time when the fishery would be biologi-
cally recovered. Since the new, higher TAC level is not known at this time, and since the recovery
period is 23 years in the future, the net present value of benefits to the commercial and recreational
users that would accrue starting in 2019 cannot be forecast and in any event would be lower than
many might suppose because of the influence of discounting a benefits stream that does not start for
23 years.
Importance of Management Regime
The case of red snapper and shrimp in the Gulf of Mexico strongly emphasizes that the man-
agement regime in place will have a dominant role to play in the overall, long-term economic
changes that will result from reductions in bycatch for any fishery. For example, theory and empiri-
cal evidence support the notion that if a fishery operates as a totally unmanaged fishery, then at-
tempts to rebuild stocks via such devices as bycatch reduction, habitat restoration, or other means
C Mb Studies
designed to enhance stocks will not be successful. This outcome is predicated on conditions whereby
total effort will increase, such that the fishery reaches a long-term equilibrium that stabilizes catches
at some level that is lower than maximum economic yield. This outcome will occur if the demand
for the species under consideration is large enough to encourage the additional effort.
Under other scenarios, there can be an open-access management regime that features quotas
and a variety of other restrictions, like trip and size limits, area closures, and gear restrictions. This
situation helps preserve the biological status of the stocks, but problems of bycatch mortality and
inefficient production methods will still preclude the attainment of all the potential biological and
economic benefits. Hence, the overall outcome under this situation is also subject to speculation.
A third class of management features market-driven effort controls, such as individual transfer-
able quotas for the commercial sector. In this case there is a possibility to forecast an increase in
overall benefits even without a great deal of information for management purposes.
The current and future management regime has particularly important consequences in the
case of the red snapper and shrimp fisheries and indeed on a number of target/bycatch fishery
combinations throughout the Southeast and the United States in general. As indicated earlier, at
about the time it became clear that an amendment to reduce shrimp bycatch was imminent, Con-
gress indicated the intent to impose a moratorium on new ITQ or similar management approaches
until the year 2000. Subsequently, this intent was written into law in the form of the Magnuson Act
as amended. As a direct result, the benefits that would have resulted from the simultaneous imple-
mentation of bycatch reduction and effective management of the red snapper resource for commer-
cial purposes have largely been put off at least until the year 2001. Under certain circumstances,
benefits could still accrue to recreational fishermen, but such benefits are not guaranteed because
the bycatch reduction is necessary merely to maintain recreational and commercial catches at cur-
rent levels.
How Much Should Bycatch Be Reduced?
From an economic perspective, in those cases where it is economically rational to reduce
bycatch to some degree, the optimum reduction in bycatch would be determined by comparing the
marginal benefits and marginal costs of each additional reduction in bycatch, and the bycatch should
be reduced as long as the marginal benefit exceeds that marginal cost of doing so. In the case of
shrimp and red snapper, the marginal costs refer to the extra cost that would be incurred by shrimp-
ers and consumers from each additional reduction in bycatch, and includes the present value of
current and future losses that would be incurred.
If it could be assumed that fishery managers and shrimpers are economically rational, the
easiest, least-cost methods of reducing bycatch would be required and adopted first. Additional
reductions in bycatch that could be achieved technically, but only with increasingly restrictive regu-
lations on shrimping activity and concurrent increases in the cost of shrimping, would be adopted
next.
At the same time when costs are being determined for the first units of bycatch reduction, the
marginal benefits from the reduction should be determined. Marginal benefits refer to benefits that
would be received by harvesters and consumers of red snapper that result from a reduction in
bycatch by the shrimp fishery.These values include the present value of the extra current and future
benefits that would be generated with each additional reduction in bycatch. It should be recognized
that even if the first units of bycatch reduction are expected to increase marginal benefits to com-
mercial and recreational red snapper fishermen, marginal benefits from successive increments of
bycatch reduction would decline for several reasons. For example, each additional 10% reduction in
bycatch probably would yield successively smaller additions to adult red snapper stocks due to the
existence of other environmental factors that tend to limit stock growth. Additions to adult red
snapper stocks also would probably yield successively smaller additions to profits of commercial
fishermen as they increase their investments in fishing effort to harvest additional quantities, and
would yield successively smaller additions to enjoyment of recreational fishermen due to the eco-
nomic principle of diminishing marginal utility. For example, the first five fish caught per trip by
recreational fishermen would yield more enjoyment than the second five if bag limits were less
restrictive.
Epilogue
In summary, biologists have determined that the red snapper resource in the Gulf of Mexico is
depleted for several reasons, including the application of too much fishing effort by commercial and
recreational red snapper fishermen and the incidental bycatch of juvenile red snapper by the shrimp
trawl fleet. The ensuing debate about how best to restore the red snapper population to desirable
levels involves numerous technological, political, biological, and economic factors. Among them are:
technological interaction in which shrimping gear inadvertently harvests juvenile red snapper; man-
agement interaction between the Reef Fish and Shrimp Fishery Management Plans; competition
between commercial and recreational fishermen and among fishermen with different gear types
within each group; economic trade-offs over time among various harvesting groups and between
different groups of consumers; the current uncertainty regarding whether the commercial manage-
ment structure for red snapper will shift to an ITQ-based system; a lack of current biological
information to determine the desirable size of the red snapper stock and the size of future yields; and
the possibility of effort controls on the recreational fishery. For all these reasons, the interaction
between the shrimp and red snapper fisheries of the U.S. Gulf of Mexico constitutes a management
problem that is controversial, challenging, and, as yet, unresolved.
References
Gulf of Mexico Fishery Management Council. 1995. Draft Amendment 8 and Environmental Assessment (Effort
Management Amendment) to the Reef Fish Fishery Management Plan for the Reef Fish Resources of the Gulf of
Mexico. Lincoln Center, Suite 331, 5401 West Kennedy Boulevard, Tampa, Florida, May, 151 pp.
Hendrickson, Holly M., and Wade L. Griffin. 1993. "An Analysis of Management Policies for Reducing Shrimp
Bycatch in the Gulf of Mexico." North American Journal of Fisheries Management 13:686-697.
Keithly Walter R., Kenneth J. Roberts, and John M.Ward. 1993. "Effects of Shrimp Aquaculture on the U.S. Market:
An Econometric Analysis." Chapter 8 in Aquaculture, Models and Economics, Upton Hatch and Henry Kinnucan (eds.).
Westview Press, Boulder, Colorado.
McConnell, Kenneth E., and Jon G. Sutinen. 1979 "Bioeconomic Models of Marine Recreational Fishing. "Journal of
Environmental Economics and Management 6:127-139.
Ward, John M. 1994. "The Bioeconomic Implications of A Bycatch Reduction Device as a Stock Conservation
Management Measure." Marine Resource Economics 9(3):227-240.
Ward, John M., and Seth Macinko. 1996. "Static and Dynamic Implications of a Gear Modification Designed to
Reduce Bycatch in a Stylized Fishery." Tire Southern Business and Economic Journal 19(4):273— 292.
Ward, John M., and Jon G. Sutinen. 1994. "Vessel Entry-Exit Behavior in the Gulf of Mexico Shrimp Fishery."
American Journal of Agricultural Economics 76(4):916-923.
Ward, John M.,Teofilo Ozuna, andWade L. Griffin. 1995. Cost and Revenues in the Gulf of Mexico Shrimp Fishery.
NOAA Technical Memorandum NMFS-SEFSC-371, National Marine Fisheries Service, Southeast Regional Office,
Economics and Trade Analysis Division, 9721 Executive Center Drive, North, St. Petersburg, Florida, May, 76 pp.
Bycatch of Yellowtail Flounder in the
Southern New England Trawl Fishery
S.F. Edwards, S.A. Murawski, and E. Thunberg
Northeast Fisheries Science Center
Woods Hole, MA
Summary
The Southern New England (SNE) stock of yellowtail flounder has been important to New
England groundfish fisheries for several decades, but the stock has been depleted to a record low
level. During 1988 to 1994, most of the catch, including the exceptionally large 1987 year class, was
discarded by trawlers because the majority offish were either too small for market or smaller than
the legal size limit.
A comparison was made of the 1988—1994 outcome with simulations of other scenarios in-
volving lower rates of fishing mortality and discarding. Analyses indicated that when compared to a
fishing strategy that maximizes yield per recruit for this fishery, discarding cost the industry and
consumers about $15 million in income (profit and crew share) and $11 million in consumer
benefits that could have manifested in lower prices for more fish. It was not possible to identify the
economically efficient optima for this fishery for two principal reasons. First, managers have not
articulated a specific bycatch policy. Although zero discards would most likely be a prohibitively
expensive option, not having a policy to evaluate makes it impossible to assess the costs of discard
reduction. Second, the single-species analysis fails to account for the benefits derived from addi-
tional species caught jointly with yellowtail flounder in the multispecies trawl fishery, including
winter flounder, ocean pout, goosefish, and several others.
The study concluded that most of the variation in landings (and discards) was attributable to
reductions in fishing mortality rates, implying that to characterize the Southern New England
yellowtail flounder trawl fishery as primarily a bycatch issue misidentifies the management problem.
The consequences of problem misidentification may have broad implications for developing man-
agement strategies for the SNE yellowtail flounder trawl fishery in particular and other fisheries in
general. In addition, while the economically optimum level of bycatch could not be determined,
analyses indicated that, when compared to a fishing strategy that maximizes yield per recruit for this
fishery, the economic optimum is probably associated with a lower level of discard, a lower exploi-
tation rate, and a higher level of catch for the seven-year period as a whole.
Background
The Sustainable Fisheries Act amendment to the Magnuson— Stevens Fishery Conservation
and Management Act requires fishery management plans (FMPs) and international agreements to
adopt regulations and measures that, to the extent practicable, minimize bycatch in U.S. waters.
Congress defined bycatch as "fish which are harvested in a fishery, but which are not sold or kept for
personal use," and distinguished between "economic discards" and "regulatory discards." Economic
discards are "fish which are the target of a fishery, but which are not retained because they are of an
undesirable size, sex, or quality, or for other economic reasons." In contrast, "fish harvested in a
Case, Studies
fishery which fishermen are required by regulation to discard whenever caught, or are required by
regulation to retain but not sell" are known as regulatory discards. 1
Both regulatory and economic discarding have historically occurred in the Southern New
England (SNE) otter trawl fishery where the ratio of discards to landings has been estimated to be as
high as 3-to-l by weight. 11 This trawl fishery has been regulated by minimum size limits (12 and 13
inches) on yellowtail flounder (Pleuronectes ferrugineus), the principal target species, and by mesh size
controls on the trawl net (4—6 inches) . Of particular interest for this case study was the fate of the
exceptionally large, 1987 year class of which 60% was discarded, by weight, throughout the cohort's
seven-year life span (Rago et al. 1993). Apart from the issues of why such discards occurred and the
responses by industry and managers, we focus on the potential economic value of this fishery during
1988-1994.
Conceptual Framework
Before reporting our findings, it is important to characterize discarding in the SNE trawl
fishery in more clear economic terms. First, discarding of undersized yellowtail in the SNE trawl
fishery is a form of economic growth overfishing because fishing mortality occurs before optimum
economic value is achieved. Not only are the sublegal fish prevented from growing larger and
possibly reproducing (virtually all discards are dead), but the "large" market category of this flounder
averages about 30 cents per pound more than "small" or "medium" fish. 111 Such premature harvest is
a chronic feature of "non-exclusive," open-access fisheries, including fisheries regulated by total
allowable catch (TAC) limits. Its extent in limited-access fisheries is being researched for Congress
by the National Academy of Sciences. 1 "
Second, the "single-species" management approach, which characterizes FMPs, including the
New England Fishery Management Council's Multispecies Groundfish Fish Management Plan, does
not accurately account for fishing practices in multi-product fisheries/ Yellowtail flounder is jointly
targeted and harvested by trawl gear in SNE along with winter flounder (Pleuronectes dentatus) and
Atlantic cod (Gadus morhua) in large mesh fisheries; whiting (Merluccins bilinearis) in small mesh
fisheries; and spiny dogfish (Squalus acanthias), skates (Family Rajidae), and monkfish (Lophius americanus),
which are not currently managed by a FMP, to name a few. The latter three species were, until
recently, biologically "underutilized" because of weak market demand, but have gained in economic
importance since the 1980s as either substitutes for depleted groundfish resources, including yel-
lowtail, or because of foreign demand. In addition, scores of species captured by the trawl gear are
not targeted (e.g., pelagic species such as long-finned squid (Loligo spp.) and Atlantic mackerel (Scomber
scombrus), which are managed by the Mid-Atlantic Fishery Management Council's Squid, Mackerel
and Butterfish Fish Management Plan). Although not targeted, this incidental catch contributes none-
theless to the financial viability of trawlers." 1 Finally, in other cases, such as for the sea raven (Hemitripterus
americanus), there is no market for the incidental catch. To base a bycatch reduction policy solely on
the consequences for the SNE stock of yellowtail flounder would imply that other stocks have no
value, that the yellowtail stock is by far most valuable, or that all parameters defining the biology
(i.e., recruitment) and economics (i.e., prices) of the resource and fishery vary in direct and constant
proportion to yellowtail. None of these implications is correct. However, an analysis of joint-pro-
duction and of discarding of other species is beyond the scope of this inquiry.
Economically Efficient Level of Discarding
Figure- 1&.
Total Net Value and Cost
Discards are shown to vary in direct opposite proportion to landings (i.e., high levels of sustainable landings
require low discarding, and high levels of discarding result in low sustainable landings). The (hypothetical)
amount of total net economic value is drawn to increase with landings, but at a decreasing rate due to consumer
preferences. In contrast, the (hypothetical) total cost of managing sustainable fisheries, including discard manage-
ment, is drawn to grow at an increasing rate (i.e., it would take increasing amounts of costly, ' 'command- and-control"
regulations to manage for low discards).
t
^ — ]
i
V)
03
8
<- Discards
Landings - *
Marginal net value of landings
Marginal costs of management
Case, Studies
Having provided a broad economic context for the yellowtail flounder bycatch problem, it is
time to impart an economic way of thinking about the efficiency of regulating fishermen. Put
simply, society is better off from an efficiency standpoint whenever the economic value from the
regulation outweighs the costs of imposing the regulation. This reasoning is illustrated in Figure la,
where the distance between the total net economic value of landings (net of fishing costs) and the
total costs of management, including discard reduction, is greatest. (Notice that landings and discards
vary inversely on the abscissa because high landings require low discards, but high discards cause low
landings.) Put another way, discard reduction improves efficiency as long as the "marginal benefit"
of the action is greater than the "marginal cost" (Figure lb). Applying this decision rule, the eco-
nomically optimal level of discards coincides with D*on Figure lb where marginal benefits equal
marginal costs (i.e., where the curves intersect) . To see the logic of this result, note that left of D*,
where discards are relatively high and landings are relatively low, the marginal benefits of reducing
discards (moving left to right on the abscissa) exceed marginal costs. In contrast, continuing to
reduce discards beyond D* is perhaps too costly for society. That is, at discard levels less than D*, the
resources used to reduce bycatch (e.g., the labor, managerial skill, and physical capital used to imple-
ment, manage, and enforce discard reduction) are more valuable in the production of other goods
and services that are valued by consumers than the gains from greater landings.
Findings
Discarding of SNE yellowtail flounder was investigated using a simulation model of stock
dynamics and dockside pricing.The stock assessment model quantified resource conditions, by age,
during the seven-year period 1988—1994, while the 1987 year class was vulnerable to trawl gear
(Rago et al. 1993). Other cohorts recruited to the fishery during this period were modeled similarly.
Starting stocks (numbers at age) from Rago et al.'s (1993) most recent resource assessment were
"fished" in a simulation model that varied the fishing mortality rate (F), proportions of the age-
structured stock selected by trawl gear, and age-specific proportions of the catch that were retained
or discarded, including all cohorts recruited to the fishery during 1988—1994. Fishing mortality,
recruitment, and discard rates were varied at random within conceivable ranges in 1,000 iterations
of the model, yielding 1,000 "observations" on landings and discards for the large and small market
categories/"
The stock assessment model was complemented with price equations for the large- and small-
market categories of yellowtail flounder. Prices were predicted from total yellowtail landings, in-
cluding landings from Georges Bank, which were held constant at their reported levels during
1988— 1994.Yellowtail flounder revenues were generated from the predicted market prices and SNE
landings, and the price models were integrated for an estimate of value for consumers."" 1 Consumers
"profit" whenever market prices are lower than what they would be willing to pay for seafood,
thereby leaving more income to spend on other goods and services. However, it is important to
account for changes in consumption as prices vary. Together, dockside revenues and consumer
"profit," or what economists prefer to call consumers "surplus," are an estimate of the total gross
value of SNE yellowtail landings. ,x Revenues and consumer surplus were standardized to constant
1994 dollars (an attempt to control for inflation), and then discounted to a present value in 1988
using the Office of Management and Budget (OMB) mandated rate of 7%. This procedure would
be required of managers who, in 1988, might have asked "What impact would bycatch reduction on
the 1987 and subsequent year classes have on dockside benefits during 1988—1994?"
Finally, the dearth of cost data on fisheries is a bane of regulatory impact analysis, but costs were
roughly approximated from available information as follows. Vessel operating costs were set equal to
30% of baseline (i.e., observed) revenues and, therefore, held constant for each iteration. This as-
sumes that total fishing effort measured in terms of days is constant regardless of policies to control
discards. In contrast, shares paid to captain and crew were calculated as 40% of gross revenues.
Figure 2 compares dockside revenues from SNE yellowtail flounder under baseline (F. „ =
2.29), or observed, conditions to the biological optimum, which maximizes yield-per-recruit (F
= 0.48). Annual dockside revenues drop sharply after reaching $15 million in three years in the
baseline case. In contrast, revenues climb steadily to nearly $16 million in five years and level off
under the F alternative. Throughout this arbitrarily short, seven-year period, revenues for the F
scenario are nearly double those for the baseline case when simply summed across years ($71 mil-
lion compared to $38 million, respectively) and nearly 70% greater when present values are com-
pared ($54 million versus $32 million, respectively). Part of the reason for the superiority of the F
bycatch scenario is that cumulative baseline landings were less during these years (37 million pounds
versus 63 million pounds), and discards were correspondingly higher (50 million pounds versus 25
million pounds) . However, landings of the higher-priced large yellowtail were also relatively greater
for F (Figure 3). It is important not to overlook the added benefit offish size.
Figure, 2.
Comparison of Annual Southern New England
Yellowtail Flounder Revenues, 1998-1994
The depleted, baseline case corresponds to actual conditions, including revenues obtained from land-
ings of the exceptionally large 1987 year-class. In contrast, revenues that might have been earned
from management at the biological optimum (F—0.48) are estimated to be considerably greater.
Biological Optimum (F=0.48)
Baseline (F=2.29)
1988 1989 1990 1991 1992 1993
1994
Case- Studies
Figure- 3.
Comparison of Annual Southern
New England Yellowtail Flounder
Revenues, 1988-1994
Added across this period, revenues from landings of large, high-priced yellowtail flounder comprised the
vast majority of total revenues in the scenario where biological yield was optimized. In contrast, revenues
from this component of the catch comprised less than half of total revenues in the depleted, baseline case.
Baseline (F=2.29)
Biological Optimum
Scenario
(F=0.48)
Next, turn your attention to the optimum amount of discard reduction from an economics
perspective. This inquiry is compromised by not having a clearer understanding of the potential
contribution of other species harvested with yellowtail in the SNE trawl fishery and by the absence
of a specific policy or set of alternatives intended to reduce discarding. We also cannot identify the
point of maximum net economic value because we do not understand the opportunity costs of
trawl vessels, captains, and crew. However, we can indicate the approximate neighborhood of an
economic optimum and regions that are clearly losers.
In Figures 4a and 4b, each graph selected gross and net economic benefits as a function of
landings (left to right) and discards (right to left) . Going from top to bottom, we have gross value
(revenues plus consumer surplus), net benefits (here, gross benefits net of vessel operating costs),
revenues, and profit. Profits and net benefits are upper estimates of their value counterparts, pro-
ducer surplus and net economic value, respectively/ Specifically, net benefits are here comprised of
consumer surplus, profit, and income to crew. To derive net economic value, however, the opportu-
nity costs of the vessel and of labor should be subtracted from net benefit, and the opportunity cost
of the vessel should be subtracted from profit. Figure 4a reports totals, and Figure 4b reports marginals
(i.e., first derivatives). In both cases, numerical values have been discounted to their present values, as
mentioned above. Due to the assumption of constant vessel operating costs, the marginal gross and
net benefits curves in Figure 4b are identical.
Looking first at Figure 4a, you see that cumulative total benefits during 1988—1994, measured
in terms of the present value of the cumulative gross benefits, net benefits, revenues, or profits, climb
from low amounts in the neighborhood of the overfished, baseline case of low landings and high
discards, but taper and then decline at some point in excess of 70 million pounds and less than 17
Benefits of Discard Reduction
figure, 4a,.
Total Benefits of Discard Reduction
Annual values were discounted to single, present values. Benefits peak before landings are maximized
because it would take too long to achieve higher landings (i.e., distant values are worth too little in
present value). Values corresponding to the depleted and biological optimum are marked.
30 40 50 60 70 80 90
Cumulative Landings, 1988-1994 (million pounds)
^™^™ Total Gross Benefits
— • - Total Net Benefits
Total Revenues
Total Profit
Figure, 4b.
Marginal Benefits of Discard Reduction
Annual values were discounted to single, present values. Marginal
benefits become negative after peaks in the corresponding total benefits.
40 50 60 70 80 90
Cumulative Landings, 1988-1994 (million pounds)
Marginal Gross Benefits
Marginal Revenues
Marginal Profit
Case, Studies
.
■fa '^m.
million pounds of discards, depending on the curve. The decline results for two reasons. First, there
is a point where further increases in landings fail to compensate for the resultant lower price. Sec-
ond, greater cumulative landings are achieved over the seven-year period by reducing landings
sharply early on. However, the benefits of future landings are being discounted to their present value
to properly evaluate the economic investment in discard reduction.
Also notice that by adding consumer surplus and crew incomes to profit, the economic case
for discard reduction is bolstered. This is apparent on Figure 4b where the marginal benefits become
zero (coincident with the peaks in total benefits on Figure 4a). Profit is maximized at about 74
million pounds of SNE yellowtail and 13 million pounds of discards. Here, profit is 50% greater than
for the baseline where discards are 50 million pounds. The point of maximum profit practically
coincides with F , but this is merely coincidental. However, by factoring in consumer surplus and
income for crew, maximum net benefits coincide with about 84 million pounds of landings and
negligible discards. Here, some profit and crew share is sacrificed to increase consumer surplus by a
greater amount. Be aware, however, that if opportunity costs could be accounted for, net economic
value probably would be maximized somewhere between these points.
Discard reduction is not a costless policy, however. The regulatory process and regulations
themselves typically involve costs that are paid by society and/or the fishing industry. To illustrate
the point, the marginal cost of discard reduction is assumed to increase in the direction of lower
discards (i.e., from left to right; Figure 5). Optimum discarding from an economic efficiency stand-
by ure- S.
Efficient Discarding of Southern
New England Yellowtail Flounder
Marginal net benefits and marginal profit are from Figure 4b. The marginal cost
of discard reduction (implicit in increasing landings) is hypothetical for illustra-
tive purposes. Due to the effect of landings on prices, profits are maximized at
lower cumulative landings than are net benefits, which include consumer surplus.
.5 <" o
■o E •«
ED km -
—I (D
03
.S to
_o> 55
a 5
e o
40
$2
..''
$1
- ">c"
$0
I ""^J \
N \
\ \
\ \
$1
-
\ \
Marginal Net Benefits
\
• Marginal Profit
\
$2
Marginal Cost of Discarding
\
I
i i i i
i i
50 60 70 80 90
Cumulative Landings (million pounds)
100
point is found where the marginal benefit of discard reduction and the marginal cost of discard
reduction curves intersect. Ignoring these costs would result in too little discarding. Some bycatch —
along with fuel, crew, vessel insurance, etc. — is part of the cost of trawling for yellowtail.
Exactly where marginal benefit and cost curves intersect depends, of course, on which discard
reduction policy is adopted and the costs of the resources used to reduce bycatch. The possibilities
include (1) conservation engineering and input management, (2) area and/or time management, (3)
effort reduction, and (4) property rights. These possibilities are not necessarily mutually exclusive,
and each would involve a degree of costs for management, monitoring, and enforcement.
Conservation engineering involves increased mesh size and new designs, such as the Nordmore
grate now used in the Gulf of Maine shrimp fishery, which are intended to facilitate escapement of
undersized yellowtail flounder and possibly other bycatch species from the trawl. To management,
monitoring, and enforcement, we should add the costs of research and development and production
of new gear purchased by fishermen. The effectiveness of conservation engineering alone is
suspect, however, because it is easy for fishermen to undermine its intended purpose with-
out a complete and costly at-sea enforcement program, which heretofore has not been
implemented.
Area and/or time management often involve closures of fishing grounds during times when
undersized fish aggregate. In the yellowtail case, a closure in the SNE area was designed to coincide
with where age-2 fish were historically discarded. Area/time management is a substitute for conser-
vation engineering, the costs of which, measured in terms of foregone revenues from jointly caught
species, could be high. Figures 4a and 4b do not reflect this option because the "opportunity costs"
of foregone revenues are not depicted.
By reducing the time that trawl gear is fished, effort reduction can be an important avenue to
bycatch reduction in overcapitalized fisheries, such as the SNE trawl fishery. Reducing effort via
reductions in vessel numbers would also greatly reduce dissipation of resource rents." 1 However,
resource rent would not result if vessel owners could not economize on fleet sizes. Allocations of
days at sea (DAS) to individual vessels in the Northeast trawl fishery are scheduled for a reduction to
50% of pre-1994 levels in 1997, but at this time consolidation is not permitted.
Finally, there has not been sufficient discussion of the conservation benefits, including bycatch
reduction, of property rights policies, including individual transferable effort quotas (ITEQs) or the
more common individual transferable landings quotas (ITQs). Such property rights policies are
potentially discard-friendly for at least two reasons. First, fishing effort and, therefore, discards will
decline in ITQ fisheries as fishermen economize on the amount of capital they require to catch
their quota. Second, and more controversial, by creating a valuable asset to yield in the fishery,
fishermen have more incentive to fish in ways that are less damaging to the resource, depending on
the degree that their fishing right is attenuated and their influence on management decisions. For
example, mobile-gear fishermen in the Atlantic Canada ITQ fishery for cod have voluntarily in-
creased their mesh size and have researched using grates to reduce by catch. x " Also, ITQ fishermen in
New Zealand's Hoki fishery twice urged their government not to raise TACs as planned for 1993
and 1994. X1 " Some researchers have also highlighted stock enhancement projects funded by associa-
tions of New Zealand ITQ fishermen. x,v "Highgrading," which is a form of economic discarding, is
often cited as problem for ITQ fisheries, but its extent might be exaggerated. 1 " Congress has re-
Case,S\
tudles
cently ordered a comprehensive review of ITQ fisheries throughout the world, including their
possible conservation benefits.
Conclusions
Discarding of sublegal (regulatory discards) and, to an extent, unmarketable (economic dis-
cards) yellowtail flounder in the Southern New England trawl fishery during 1988— 1994, including
recruits from the exceptionally large 1987 year class, resulted because the overcapitalized trawl
fishery used a mesh size which, in practice, captured too many small fish. Although the minimum
fish size was increased from 12 inches to 13 inches, the trawl gear continued to capture sub- 13-inch
fish in large quantities.
Discarding SNE yellowtail was costly for the Northeast region. Compared to the biological
optimum that maximizes yield-per-recruit, discarding cost the industry and consumers approxi-
mately $15 million in income (profit and crew share) and $11 million in consumer surplus (present
value estimates). However, discard reduction is always an economically costly task. Therefore, the
biological target does not necessarily coincide with what is economically optimal. Furthermore, a
single-species approach to discard reduction ignores the contribution made by other jointly har-
vested species to benefits.
An economic optimum cannot be identified without data on the costs of discarding, including
the management and enforcement costs of specific bycatch reduction policies. To completely elimi-
nate discarding is clearly too costly (see Figure 5), but ignoring it is economically wasteful (see
Figure la). Contrary to popular thinking, traditional notions of conservation and economic effi-
ciency are allies.
Controlling yellowtail fishing mortality somewhere within the economically relevant range
for bycatch reduction by correcting the mismatch between fish size and trawl selectivity, including
use of time or area closures, could have been a win-win policy from traditional economic and
conservation perspectives because of the greater industry incomes and consumer benefits just noted
and because of less discarding (e.g., less than 13 million pounds where the marginal benefit and cost
of discard reduction are equal compared to 50 million pounds for the baseline) . Amendment 7 to
the Multispecies Fishery Management Plan, which seeks to reduce DAS in Northeast groundfish trawl
fisheries to 50% of pre-1994 levels in 1997, might contribute to bycatch reduction. However, addi-
tional necessary measures involving conservation engineering and incentive-compatible property
rights are not part of Amendment 7.
The costs of various options to curtail discarding will place a lower bound on what is eco-
nomically sensible to do. Unlike on fish farms where size-selectivity can be precisely practiced, at
least some regulatory discarding of undersized yellowtail flounder will have to be accepted as part of
the cost of prosecuting wild fisheries. However, the task is not necessarily limited to searching for a
least-cost alternative that minimizes yellowtail discards. That is, if ITQs are found to be bycatch-
friendly and their use by regional fishery management councils is permitted and implemented, the
fishing right will take on value in proportion to reductions in vessel numbers and growth of the
demersal resources. In any case, reducing discards of juvenile yellowtail flounder in the SNE trawl
fishery will frustrate managers until they devise a system whereby those who discard suffer the cost,
and those who conserve reap the benefits.
Endnotes
' Devising a mutually exclusive taxonomy for bycatch is a daunting task that is complicated by the behaviors of
fishermen, managers, consumers and processors. For example, a 400-pound trip limit in the "general" permit
category of the Atlantic sea scallop (Placopecten magellanicus), fishery can result in regulatory discards. How-
ever, by imposing a landings constraint, the trip limit might simultaneously cause fishermen to "high-grade"
scallop catches due to the higher price per pound paid for larger scallop meats. Thus, a scalloper could simul-
taneously be engaged in regulatory and economic discarding. Another difficulty arises because markets are
changeable.Therefore, what was an economic discard one year might only be partially discarded the following
year after the market is fully supplied, and after several years become a regulatory discard after the resource is
overfished and a management plan put in place. For example, incidental catches of monkfish (Lophius americanus)
were considered a nuisance by groundfish and scallop fishermen until markets for tails and livers were discov-
ered in Europe and Asia during the 1980s. Now they are overexploited, and an amendment to the Multispecies
Groundfish Plan proposes adding monkfish to the management unit and to impose minimum size and trip limits.
" Rago et al. (1993) calculated this estimate using 1990 sea sampling data.
'" Gates and Norton (1974) examined growth overfishing in the SNE yellowtail fishery during the 1970s, and
Gates (1976) estimated the influence offish size on yellowtail dockside prices.
" See Gordon's (1954) seminal work on overcapitalization that results from open access, but see Cheung (1970)
for waste on the revenues side of the equation when resources are non-exclusive, including what has come to
be called growth overfishing. Congress used the Magnuson-Stevens Act to call for an independent review of
limited access systems, including whether ITQs create sufficient incentives for fishermen to minimize bycatch.
* Although called the Multispecies Fishery Management Plan, overfishing definitions and practices reveal a
single-species approach to groundfish management. Also, the effects of resource, market, or regulatory condi-
tions on fishermen's behaviors and vice versa-e.g., fishing effort and input substitution (Squires and Kirkley
1991)-are not appropriately accounted for in management plans.
v ' The FAO defines bycatch as the sum of discards and incidental catch (Alverson et al. 1994).
v " The simulation model was run using Microsoft Excel 5.0 and @RISK software. Fishing mortality (F) and
partial recruitment (PR) and discard rates (DR) for age-1, age-2, and age-3 flounder were varied within
uniform distribution functions as follows: (1) 0.000 to 3.000 for F ; (2) 0.000 to 0.050 for PR n t ; (3) PR a e , to
0.200 for PR ,; (4) PR , to 1 .000 for PR • (5) DR , to 1 .000 for DR , ; (6) DR , to 0.900 for DR
age-2 v ' ^ge-2 age-3 v ' age-2 age-1 x ' age-J age-
,; (7) 0.300 to 1 .000 for Dr t .
v '" The average value of consumers surplus from total landings was used in these estimates because it would be
arbitrary to assign SNE landings as being either the first or last supplies.
a This is not the place to delve further into notions of economic value. See Edwards (1991) for a discussion
related to fisheries and for more references.
* See Copes (1972) for definitions and graphical relationships.
X1 See Gordon (1954).
1 This information was reported by Jean Guy d'Entremont, a small-boat fishermen, at the New England Aquarium's
forum on "Establishing an Agenda for Responsible Fishing," held in Boston on December 3, 1996.
*"' Paper presented by Eric Barratt, then past president of the New Zealand Fishing Industry Association and
General Manager of the Sanford South Island Limited fishing business, at the Fishery Council of Canada's
1994 Annual Convention on "Building a Fishery that Works: A Vision for the Atlantic Fisheries," held in
Fredericton, New Brunswick.
x,v Pearse and Walters (1992).
Mv Arnason (1994).
&to Comments
In March 1997 NMFS published a Notice of Availability in the Federal Register announcing the
availability of the draft bycatch plan, Managing the Nation's Bycatch: Priorities, Programs and Actions for
the National Marine Fisheries Service. Comments were received from 36 organizations or individuals
representing a range of interests from conservation organizations to commercial and recreational
fishing associations to the fisheries management councils and state management agencies. The com-
ments were very helpful in revising the plan and preparing the final document. This appendix
summarizes the comments and addresses each major comment. Many commenters also identified
inconsistencies in the draft plan and suggested editorial or textual changes. These specific comments
are not addressed here; however, the final plan document incorporated these suggestions as much as possible.
Inconsistent Definition of Bycatch
Comment: Several commenters stated that the definition of bycatch used in the bycatch plan is
inconsistent with that in the Magnuson— Stevens Act. Specifically, commenters questioned the in-
clusion of retained incidental catch and unobserved mortality in the definition of bycatch. Several
stated that the definition used in the plan deviated from the accepted definition of bycatch.
Response: The retained incidental component was removed from the definition of bycatch in the
final version of this plan in order to make the NMFS national bycatch goal consistent with the
MSFCMA and National Standard 9. The inclusion of unobserved mortality is essential to meeting
NMFS' responsibility to assess total fishing mortality and to base management decisions on the best
scientific information available.
The question of whether retained incidental catch should be included in the definition of
bycatch is difficult. There are situations where suboptimal use of species as incidental catches in one
fishery may adversely affect the catch and revenue to fisheries for which the same species is the
primary target. In such instances, the by-product (bycatch) of one fishery may have very great
consequences on our ability to maximize the biological (yield or spawning) or economic potential
of such shared resources, and may very well require their reduction or elimination in some fisheries.
However, inclusion of retained incidental catch is not consistent with the MSFCMA s defini-
tion of bycatch. The MSFCMA defines bycatch as "fish which are harvested in a fishery, but which
are not sold or kept for personal use." The MSFCMA goes on to require that this bycatch be
minimized to the extent practicable. Including retained incidental catch in the definition of bycatch
would conflict with this mandate because, in many cases, retained incidental catch is a vital compo-
nent of the overall economic activity generated by a fishery.Thus, it may be desirable to preserve this
component of the catch, even as fishermen strive to eliminate or greatly reduce the discard and
unobserved mortalities.
Adopting a broader operational definition of bycatch, NMFS recognized that mortality associ-
ated with fisheries is greater than retained catch. Other components of fishing mortality, such as
unobserved fishing mortality due to encounters with gear, may be critical elements affecting the
sustainability of fisheries and marine ecosystems.
Unclear Prioritization of Minimization vs. Increased Utilization
Comment: Several commenters suggested that the bycatch plan does not clearly prioritize bycatch
minimization over increased utilization of retained incidental catch. Commenters expressed con-
cern that while the bycatch "scorecard" could be improved by increasing utilization of catch that is
currently discarded, this would not result in the decrease in bycatch mortality implicit in National
Standard 9.
Response: In resolving bycatch issues first priority must be given to avoiding bycatch to the extent
practicable. To the extent that it is not practicable, then priority must be given to minimizing
bycatch mortality. The goal of the bycatch plan is to "implement conservation and management
measures for living marine resources that will minimize, to the extent practicable, bycatch and the
mortality of bycatch that cannot be avoided." Inherent in this goal is the need to avoid bycatch over
creating new ways to utilize bycatch. The practicability of reducing bycatch is not a static concept.
It is expected that, over time, technological innovations and changing demands of fisheries may
expand the practicability of reducing bycatch and bycatch mortality. In such cases, bycatch manage-
ment, too, would respond by taking steps to further reduce bycatch as intended by National Stan-
dard 9.
Need for a Framework for Determining Priorities
Comment: Several commenters suggested that the bycatch plan include a framework for determin-
ing priorities among objectives and recommendations.
Response: Determining priorities for bycatch minimization is, in many cases, a regional- and Council-
level issue, and NMFS feels strongly that this is appropriate. For national and intra-regional objec-
tives and recommendations, the bycatch plan establishes a framework by which priorities can be
determined. The framework is flexible yet consistent, designed to meet evolving needs of scientists
and managers in a predictable, consistent fashion. This flexible approach is intended to encourage
innovative approaches to bycatch management while establishing a deliberative framework by which
management decisions are made.
The purpose of this plan is to assess what is known about bycatch in the nations fisheries and
what steps should be taken to address bycatch through science and management. Decisions regard-
ing funding and allocation of the Agency's resources is outside the scope of this document.
More Detailed Discussion of Ecosystem Effects
Comment: Several commenters remarked that the discussion of ecosystem effects of bycatch lacked
sufficient detail to be indicative of NMFS' ecosystem-related bycatch planning.
Response: The full range of ecosystem effects of all three components of bycatch (discarded catch,
retained incidental catch, and unobserved mortality) on living marine resource populations, preda-
Response, to
CoHmvettts
tor/prey relationships, detrital food webs, and essential fish habitat is not well understood. The
inclusion of ecosystem-level effects in all stages of bycatch planning as outlined in the document
emphasizes its importance to research planning and management decision-making. While a body of
literature has begun to develop on the effects of bycatch on the functioning of components of
marine ecosystem, a detailed discussion of specific ecosystem effects of bycatch is not possible at this
time.
Insufficient Attention to Recreational Fisheries
Comment: Commenters expressed concern that the draft bycatch plan focused on commercial
fisheries bycatch to the exclusion of recreational fisheries.
Response: Bycatch is an issue that affects nearly every fishing operation, both commercial and
recreational. The bycatch plan is intended to be used to guide NMFS' bycatch-related research and
management, for commercial and recreational fisheries bycatch. As in many commercial fisheries,
the magnitude of bycatch mortality in recreational fisheries is not fully documented and post-
release survivability is of great interest. The discussion of bycatch has been expanded at several places
in the document to more fully address bycatch issues in recreational fisheries.
Overemphasis of Economic Considerations
Comment: Several commenters stated that the bycatch plan overemphasized economic consider-
ations related to bycatch management in relation to population, socio-economic, and ecosystem
considerations.
Response: Economic factors must be considered in bycatch management. However, they must be
balanced with other concerns. Full consideration of economic incentives and disincentives can help
managers determine how current management may contribute to total fishing mortality, including
bycatch, and how management can be designed to most effectively minimize bycatch. However,
these economic considerations should be viewed in context with considerations of bycatch as it
affects ecosystems and populations of living marine resources as well as fishery participants and
fishing-dependent communities. The bycatch plan attempts to give balanced consideration to each
of these factors in order to meet the national goal of minimizing bycatch and bycatch mortality.
The economic consequences of dealing with bycatch is one of the factors that determines the
extent to which it is practicable to reduce bycatch or bycatch mortality in a particular fishery. The
determination must be based on the net benefit to the nation resulting from particular management
measures. The net benefit to the nation includes, but is not limited to, reductions in negative impacts
on affected stocks; short- and long-term incomes accruing to participants both in the fisheries in
which the bycatch is taken and in the fisheries that target the bycatch species; environmental conse-
quences; non-market values of bycatch species, including non-consumptive uses of bycatch species
and existence values; recreational values; and impacts on other marine organisms.
Inconsistent Use of Bycatch and Discards
Comment: Several commenters noted that the terms bycatch and discards are used interchangeably or
inconsistently throughout the bycatch plan.
Response: Discards and unobserved mortality, are the components of bycatch. There is currently
very little quantitative information available about unobserved mortality, and, by necessity, much of
the bycatch plan's discussion about what is known about bycatch is based solely on discards. How-
ever, bycatch planning must acknowledge and incorporate and unobserved mortality. Furthermore,
while retained incidental catch is not part of bycatch, estimates of this measure should be included
in assessments of total fishing mortality and the total economic activity of a fishery. In the bycatch
plan, discussion of long-term strategies and objectives generally focuses on bycatch, including im-
proving current methods of estimating discards, developing new methods to estimate unobserved
mortality, and developing management systems to address all components of bycatch. Sections of
the bycatch plan that discuss what is currently known about bycatch, such as the National Assess-
ment, focus on discards since that is generally the only component of bycatch mortality for which
quantitative information is available. Corrections have been made to various places in the text
where the usage of bycatch and discards was unclear or inaccurate.
Insufficient Attention to Public Concerns
Comment: Several commenters stated that in parts of the bycatch plan NMFS appeared dismissive of
public concerns about bycatch and assumed a condescending tone regarding bycatch management.
Response: Public concern about bycatch has been critical to establishing bycatch as a global fisher-
ies concern. NMFS, with its partners in industry, the recreational and conservation communities,
state fishery management agencies, the fishery management councils, and tribal and international
management organizations has listened to this public concern and is responding, in part, with initia-
tives like this bycatch plan. NMFS recognizes and appreciates the importance of two-way dialogue
with all of its constituents, particularly in addressing an issue as complex as bycatch.
More Direct Discussion of the Waste Issue
Comment: Several commenters suggested that NMFS address the issue of waste more directly in the
bycatch plan.
Response: Reports of large quantities of fishery resources being dumped at sea or of large numbers
of marine mammals being taken in fishing operations have resulted in a mounting public concern
that valuable marine resources are being wasted or needlessly killed. In part, the inclusion of Na-
tional Standard 9 in the Magnuson— Stevens Act is the most recent response to this growing public
concern. As NMFS, working in cooperation with regional fishery management councils and con-
stituents, implements measures to minimize bycatch and the mortality of bycatch that cannot be
avoided, a concerted effort must be made to consider measures that result in real reductions of
bycatch, rather then new ways to use existing bycatch.
Objection to the Term Adequacy of Current Measures
Comment: Several commenters objected to the use of the term adequacy of current measures to describe
current management programs.
Response: The determination of adequacy of current measures was made in the context of developing
the Bycatch Information Matrix found in Appendix A. This determination was solely part of an
Response* to
Co*
exercise to determine what is known about the magnitude and type of bycatch in the nation's
fisheries and to identify those fisheries that have some bycatch management measures in place.
Adequacy in this context is not meant to serve as a justification to fail to explore other management
options nor was the determination made in view of National Standard 9 criteria.
Copies of this report may be obtained from:
OFFICE OF SCIENCE AND TECHNOLOGY, F/ST
NATIONAL MARINE FISHERIES SERVICE, NOAA
1315 EAST-WEST HIGHWAY
SILVER SPRING, MD 20910
1
A