-74
-68
-66
42.5
41.2
41.2
39.9
39.9
Oct 1999
38.6
38.6
Oct 1999
37.3
36.0
-76
37.3
-74
-72
-70
Longitude
-68
36.0
-66
Latitude
Latitude
-76
42.5
Longitude
-72
-70
Area II Reporting Locations: Weeks 41-42
41.5
> 2500 lb/day
1850-2500 lb/day
1500-1850 lb/day
1100-1500 lb/day
< 1100 lb/day
41.4
41.3
41.2
41.1
41.0
-67.50
-67.25
-67.00 -66.75
Longitude (deg)
-66.50
-66.25
Latitude (deg)
Area II Reporting Locations: Before Oct 1
41.5
> 2500 lb/day
1850-2500 lb/day
1500-1850 lb/day
1100-1500 lb/day
< 1100 lb/day
41.4
41.3
41.2
41.1
41.0
-67.50
-67.25
-67.00 -66.75
Longitude (deg)
-66.50
-66.25
Latitude (deg)
Area II Reporting Locations: After Oct 1
41.5
> 2500 lb/day
1850-2500 lb/da
1500-1850 lb/da
1100-1500 lb/da
< 1100 lb/day
41.4
41.3
41.2
41.1
41.0
-67.50
-67.25
-67.00 -66.75
Longitude (deg)
-66.50
-66.25
The After Effects of the 1999 fishery:
Catch and Discard Rates in the 2000 fisheries in
the re-opened closed areas
Area
Tows
Contact
Hrs
Meat Wt
(lb)
Catch lbs / Scallop
hr contact Discard
Area 1
3,104
2,076
866,146
417
14%
5,504
5,707
472,135
83
24%
914
568
447,832
789
10%
Area 2
NLS
Bases for Tradeoffs
•
Habitat and Bycatch Issues
– Better information on habitat
implies less impacts on nonscallop habitats
– When total harvest weight is
constrained, fishing on
higher concentrations of
scallops implies less bottom
contact time.
– Less contact time implies
less habitat impact
– Less contact time implies
less chance of bycatch
Multi-Objective Linear Programming
A relatively simple way to compare tradeoffs among
objectives
Key Elements:
Quantifiable Objective,
Decision Variables,
Constraints
Di,j = Decision variable for area i, j
where Di,j = 1 if area is open to fishing, else =0
Vs,i,j = Value of species s in area i, j.
where Vs,i,j = f(biomass, impact potential, etc…)
Defining Objectives and Constraints
Objective Function for the set {E} of species or attributes
that are enhanced by fishery,
  D V
i, j
sE
i
s ,i , j
j
Objective Function for the set {I} of species or attributes that are
dimished/degraded/impacted by fishery.
  (1  D )V
sI
i, j
i
j
s ,i , j
Evaluating Multiple Objectives
It is not necessary for the two objective functions to have commensurate values. Each
objective function is weighted by an arbitrary value m such that m = 1. For a simple
problem with two objectives, the optimization model can be written as:
Maximize { 
sE
 D V
i, j
i
j
s ,i , j
+ (1-)
  (1  D )V
sI
Subject to: 0< Di,j <1, and other constraints
i, j
i
j
s ,i , j
}
Evaluating All Possible Alternatives
• It is not necessary to derive the relative value or merit of each
objective function component. This is the subject of endless
and divisive debate and source of amusement to outsiders.
• Instead, one examines the value of the objective function over
the full range of relative values of  between 0 and 1.
• The resulting set of optimal solutions define the Pareto
optimality frontier, a boundary that separates feasible from
infeasible solutions, and a benchmark against which specific
solutions can be compared.
• The solution set corresponding to a point on the Pareto
boundary can be used as starting points for the development of
a particular solution in which non-quantifiable or difficult to
quantify factors are incorporated
Bycatch Reduction
Classic Economic Choices:
Guns vs Butter—Swords vs Plowshares—
Scallops vs Bycatch
Optimal
Solutions
Feasible Solutions
Scallop Yield
Infeasible
Solutions
Bycatch Reduction
Solutions that approach the boundary are
better than those near the origin because
more of one or more of the objectives is
attained
Best Solution
Better Solution
Good Solution
Poor Solution
Scallop Yield
Infeasible
Solutions
Solutions on the boundary represent the set
of possible weighting of the objective
function
Bycatch Reduction
P=1
P=0.7
P=0.5
Infeasible
Solutions
P=0.2
P=0
Scallop Yield
Solutions on the boundary represent the set of possible weighting of the
objective function and a particular pattern of open and closed areas.
Bycatch Reduction
P=1
'
'
' '
'
'
'
'
'
' ' '
' ' '
' '
'
' ' '
' '
P=0.5
'
'
'
'
' ' ' ' '
' ' '
'
' '
'
' '
'
' ' '
' ' ' '
P=0
Scallop Yield
'
'
'
'
'
'
'
'
' ' ' '
'
' '
' '
'
'
' '
'
'
'
'
' '
'
' ' '
'
'
' '
' '
'
'
' ' '
' ' '
' ' '
Bycatch Reduction
Alternative Solutions can be evaluated with respect to
the attainment of maximum values that would be
possible in the absence of additional objectives.
Acceptable solutions are those that are acceptable to all
parties
Bycatch Reduction
Scallop Yield
Yield Loss
“Steeper” Solutions on the boundary represent the
ideal situation: Both objective functions are near
their maximum values and little has to be given up.
Bycatch Reduction
Bycatch Reduction
Scallop Yield
Yield Loss
Some Conclusions
• Spatial patterns of fishing have important
implications for bycatch, habitat, and
fishing mortality
• Each pattern of fishing has different
consequences for each species.
• Managing at the margin poses risks to the
resources, industry and ecosystem
• Tradeoffs are an essential aspect of
fisheries resource management.