Biological Models Used to

Evaluate Medium-Term Impacts

9-January-2004
Stock Assessment
Objectives of an assessment:
Stock status
Uncertainties
Projections
Assessment Input Data
Basic Biology:
lifespan
growth
movements
Fishery Information:
maturity rate
historical development (areas, gears)
past and current regulations (size limits, gear restrictions).
catch (landings, discards, age/size distribution)
effort (catch rates)
Surveys
distribution
relative abundance and biomass over time
age/size structure
life history (growth, maturity)
Types of Assessments
A nonlinear progression from data-poor to data-rich situations.
Index Methods (n = 7):
Descriptive assessment of catch and survey data.
Biomass Dynamics Methods (n = 1):
Combined analysis of catch and survey data with a simple
biomass-based population model.
Age-Structured Methods (n = 11, 8 of 11 include discards):
Virtual Population Analysis: Back-calculate stock numbers at
age using age distribution of the catch, calibrated with survey
indices to minimize measurement error.
Statistical Catch-at-Age Analysis: Forward-projection of
stock numbers at age using age distribution of the catch,
calibrated with survey indices or other auxiliary information
in a likelihood-based framework.
Age-Based
Methods
Age distribution of
the catch.
From census of total
catch biomass and port
samples.
SNE yellowtail
example:
1987 yearclass
dominated the catch in
the late 80s early 90s.
Southern New England Yellowtail Catch
-2003
-1998
-1993
-1988
-1983
-1978
-1973
0 1 2 3 4 5 6 7 8
Age
Y
e
a
r
VPA
Reconstruction of all
yearclasses gives a
total population
estimate.
Input Data:
catch at age
estimate of natural
mortality
initial guess about
abundance of
survivors at the
oldest age.
Southern New England Yellowtail Abundance
-2003
-1998
-1993
-1988
-1983
-1978
-1973
0 1 2 3 4 5 6 7 8
Age
Y
e
a
r
VPA Calibration
Initial guesses are replaced
with estimates:
oldest age of historical
yearclasses estimated by
assuming that age-7 fish have
the same vulnerability to the
fishery as ages 4-6:
yearclasses that are alive now
require more information.
need an independent index of
relative abundance over time.
Southern New England Yellowtail Abundance
-2003
-1998
-1993
-1988
-1983
-1978
-1973
0 1 2 3 4 5 6 7 8
Age
Y
e
a
r
( )
t
Z
t
t t
t
e F
Z C
N

=
1
VPA Calibration
Abundance of
living
yearclasses in
2002 are
estimated using a
predictive
relationship
between
historical VPA
abundance and
survey indices.
0
5
10
15
20
25
0 10 20 30 40 50 60 70
Age-3 Abundance from VPA (millions)
A
g
e
-
3

S
p
r
i
n
g

S
u
r
v
e
y

I
n
d
e
x

(
#
/
t
o
w
)
observed
predicted
2002
1990
(1987 yearclass)
VPA Estimates
Informative
Assessment:
Example: SNE
yellowtail
Estimates of stock size
and F,
But also age
distribution,
recruitment, mature
biomass, etc.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
1
9
7
3
1
9
7
5
1
9
7
7
1
9
7
9
1
9
8
1
1
9
8
3
1
9
8
5
1
9
8
7
1
9
8
9
1
9
9
1
1
9
9
3
1
9
9
5
1
9
9
7
1
9
9
9
2
0
0
1
F
i
s
h
i
n
g

M
o
r
t
a
l
i
t
y

(
4
-
6
)
F40%
0
5
10
15
20
25
1
9
7
2
1
9
7
4
1
9
7
6
1
9
7
8
1
9
8
0
1
9
8
2
1
9
8
4
1
9
8
6
1
9
8
8
1
9
9
0
1
9
9
2
1
9
9
4
1
9
9
6
1
9
9
8
2
0
0
0
SSB Year; Recruitment Yearclass
S
p
a
w
n
i
n
g

B
i
o
m
a
s
s

(
'
0
0
0
s

m
t
)
0
20
40
60
80
100
120
140
A
g
e
-
1

A
b
u
n
d
a
n
c
e

(
m
i
l
l
i
o
n
s
)
recruitment
SSB
Groundfish Stock Status - 1996
Biomass 1996 / B-MSY
0.0 0.3 0.5 0.8 1.0
F

1
9
9
6

/

F
-
M
S
Y
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
GM Cod
GB Cod
GM Had
GB Had
CC YT
GB YT
SNE YT
Witch
Plaice
GM Win
GB Win
SNE Win
W Hake
Pol
Redfish
Pout
N Wind
S Wind
overfishing
not overfished
no overfishing
overfished
no overfishing
not overfished
overfishing
overfished
F-MSY
1/2 B-MSY
Groundfish Stock Status - 2002
Biomass 2002 / B-MSY
0.0 0.3 0.5 0.8 1.0 1.6 1.8 2.0
F

2
0
0
2

/

F
-
M
S
Y
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
GM Cod
GB Cod
GM Had
GB Had
CC YT
GB YT
SNE YT
Witch
Plaice
GB Win
SNE Win
W Hake
Pol
Redfish
Pout
N Wind
S Wind
overfishing
not overfished
no overfishing
overfished
no overfishing
not overfished
overfishing
overfished
F-MSY
1/2 B-MSY
VPA Uncertainty
Similar to production
model: survey
measurement errors are
reshuffled many times
to estimate precision.
(bootstrapping).
The estimate of 2001
SSB is 1850mt, with a
80% confidence limit
of 1500 to 2500mt.
SNE Yellowtail
0
10
20
30
40
50
60
70
1
2
0
0
1
3
0
0
1
4
0
0
1
5
0
0
1
6
0
0
1
7
0
0
1
8
0
0
1
9
0
0
2
0
0
0
2
1
0
0
2
2
0
0
2
3
0
0
2
4
0
0
2
5
0
0
2
6
0
0
2
7
0
0
2
8
0
0
2
9
0
0
3
0
0
0
3
1
0
0
3
2
0
0
3
3
0
0
2001 SSB (mt)
F
r
e
q
u
e
n
c
y
Projections
"It is far better to
foresee even
without certainty
than not to
foresee at all"
Poincare, The
Foundations of
Science
Short-Term Projection
Fluke (SAW35):
Landings in 2003
would need to be
10,580 mt (23.3
million lbs) to
meet the target F
rate of Fmax =
0.26 with 50%
probability.
Long-Term Projection
Georges
Bank Cod
the stock is
expected to
have
approximate
ly a 50%
chance of
rebuilding to
SSB
MSY
by
2026 if
fished at an
F of 0.18.
Year
2002 2006 2010 2014 2018 2022 2026
S
p
a
w
n
i
n
g

B
i
o
m
a
s
s

(
0
0
0
s

m
t
)
0
50
100
150
200
250
300
75th percentile
Median Spawning Biomass
25th percentile
BMSY
Age-Structured Model
Population Numbers, Survival, Spawning Biomass
Catch, Landings, and Discards
Population Harvest
Population Numbers at Age
( )
( )
( )
( )
( )
N t
N t
N t
N t
N t
R
R
R
A
=

(
(
(
(
(
(
+
+
1
2
M
Survival by Age Class
N t N t e
for a R to A
a a
M t F t
a
( ) ( )
( ) ( )
=
= +

1
1 1
1
1 1
Survival of Plus Group
N t N t e
N t e
A A
M t F t
A
M t F t
A
A
( ) ( )
( )
( ) ( )
( ) ( )
=
+

1
1
1 1
1
1 1
1
Spawning Biomass
| |
SSB t W FM N t e
S a a a
Z t M t F t
a R
A
PROJ a
( ) ( )
,
( ) ( ) ( )
=
+
=

Catch Numbers at Age

| |
C t
F t
M t F t
e N t
a
a
a
M t F t
a
a
( )
( )
( ) ( )
( )
( ) ( )
=
+


1
Landings
| |
L t C t DF t W
a
a
A
a L a
( ) ( ) ( )
,
=
=

1
1
Discards
D t C t DF t W
a a D a
a
A
( ) ( ) ( )
,
=
=

1
Population Harvest
Input fully-recruited
fishing mortality F(t)
Input partial recruitment
vector PR(t) and Z
PROJ
(t)
Input discard fraction at
age vector DF(t) if
applicable
Input landings quota Q(t)
Input partial recruitment
vector PR(t) and Z
PROJ
(t)
Input discard fraction at
age vector DF(t) if
applicable
Solve for F(t)
Fishing Mortality at Age
F t F t PR t
a a
( ) ( ) ( ) =
Catch Numbers at Age as a Function
of Fishing Mortality
| |
C F
PR t F
M t PR t F
e N t
a
a
a
M t PR t F
a
a
( )
( )
( ) ( )
( )
( ) ( )
=

+


1
Landings as a Function of F
| |
L F C F DF t W
a
a
A
a L a
( ) ( ) ( )
,
=
=

1
1
Solve for Fishing Mortality to Harvest
Landings Quota
Q L F = ( ) 0
Age-Structured Model
Stock-Recruitment Relationship
Initial Population Abundance
Abundance and Fishing Mortality Thresholds
Stock-Recruitment Relationship
Deterministic component
Stochastic component
( ) ( )
N t f SSB t R t
R
( ) ( ) , , =
Recruitment Models
Dependent on spawning biomass (n = 10)
Independent of spawning biomass (n = 5)
Uncorrelated stochastic component (n = 10)
Correlated stochastic component (n = 5)
Beverton-Holt Curve Lognormal Error
( )
n t
a ssb t R
b ssb t R
e
where w N
R
w
w
( )
( )
( )
~ ,
=

+

0
2

Empirical Cumulative
Distribution Function
( )

N t T R R U
S
T
R
where S U T
R S S S
( ) ( )
( )
=

|
\

|
.
| +
= +
+
1
1
1
1 1
1
Georges Bank yellowtail flounder recruitment CDF
Recruitment (000,000s age-1 fish)
0 20 40 60 80 100
F
r
e
q
u
e
n
c
y

R

<

R
e
c
r
u
i
t
m
e
n
t
0
5
10
15
20
25
Population Abundance and
Fishing Mortality Thresholds
Abundance
Spawning biomass
Mean biomass of USER-selected age range
Total biomass
Fishing mortality
Fully-recruited fishing mortality
Fishing mortality weighted by biomass
Probability of Achieving Threshold
( )
Pr ( )
( )
( )
SSB t SSB
K t
K t
THRESHOLD
THRESHOLD
TOTAL
=
Initial Population Abundance
No uncertainty for estimate of N(1)
Uncertainty for estimate of N(1)
Distribution of bootstrap replicates of N(1)
Nonparametric
Parametric
Samples from posterior distribution of N(1)
Bootstrap Algorithm
George Bank Haddock
2001 Spawning Biomass Distribution
Precision of 2001 SSB Estimate
SSB (thousand, mt)
50 55 60 65 70 75 80 85 90 95 100 105 110
F
r
e
q
u
e
n
c
y
0
20
40
60
80
100
120
140
160
C
u
m
u
l
a
t
i
v
e

P
r
o
b
a
b
i
l
i
t
y

(
%
)
0
10
20
30
40
50
60
70
80
90
100
Input/Output Schematic Diagram
INPUT FILENAME
OUTPUT FILENAME
SYSTEM DATA
AGEPRO
SIMULATION DATA
BIOLOGICAL DATA
FISHERY DATA
INPUT FILE
PROJECTION
DESCRIPTION
HARVEST STRATEGY
SSB TRAJECTORY
SSB THRESHOLD*
RECRUITMENT
TRAJECTORY
CATCH-AT-AGE
INDEX*
LANDINGS
TRAJECTORY*
MARKET CATEGORY
TRAJECTORIES*
DISCARD
TRAJECTORY*
REALIZED F
TRAJECTORY*
OUTPUT FILE
MARKET CATEGORY FILE*
REALIZED LANDINGS
BY CATEGORY
*If applicable
NO
YES
NO
YES
NO
NO
YES
BEGIN
END
*If Applicable
Flowchart for AGEPRO software
DONE
WITH IC
LOOP?
DONE
WITH SIM
LOOP?
SET INITIAL
POPULATION VECTOR
DONE
WITH TIME
LOOP?
SET VARIABLES THAT
VARY BY TIME PERIOD*
COMPUTE
CATCH AT AGE
IS
QUOTA
FEASIBLE
?
RECORD THAT
SIMULATION IS
INFEASIBLE
COMPUTE LANDINGS
COMPUTE SSB
GENERATE RANDOM
RECRUITMENT
COMPUTE FUTURE
POPULATION VECTOR
FOR NEXT TIME PERIOD
COPY CURRENT SSB TO
PREVIOUS SSB
COPY FUTURE
POPULATION VECTOR
TO CURRENT
George Bank Haddock F
REBUILD
for 1999-2009 Time Horizon
Year
2002 2006 2010 2014 2018 2022 2026
F
i
s
h
i
n
g

M
o
r
t
a
l
i
t
y
0.00
0.05
0.10
0.15
0.20
0.25
0.30
Fishing Mortality
FMSY
George Bank Haddock F
REBUILD
Spawning Biomass Distribution
Year
2002 2006 2010 2014 2018 2022 2026
S
p
a
w
n
i
n
g

B
i
o
m
a
s
s

(
0
0
0
s

m
t
)
0
50
100
150
200
250
300
350
75th percentile
Median Spawning Biomass
25th percentile
BMSY
George Bank Haddock F
REBUILD
Landings Distribution
Year
2002 2006 2010 2014 2018 2022 2026
L
a
n
d
i
n
g
s

(
0
0
0
s

m
t
)
0
5
10
15
20
25
30
35
40
45
50
55
60
65
75th percentile
Median Landings
25th percentile

Prediction is very difficult

Prediction is very difficult
Prediction is very difficult
especially about the future
especially about the future
especially about the future
Niels
Niels
Niels
Bohr
Bohr
Bohr