Upload
catelia-kulman
View
213
Download
0
Embed Size (px)
Citation preview
7/29/2019 Session5b
1/45
Biological Models Used to
Evaluate Medium-Term Impacts
9-January-2004
7/29/2019 Session5b
2/45
Stock Assessment
Objectives of an assessment:
Stock status
Uncertainties
Projections
7/29/2019 Session5b
3/45
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)
7/29/2019 Session5b
4/45
Types of Assessments
A nonlinearprogression 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 simplebiomass-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 surveyindices to minimize measurement error.
Statistical Catch-at-Age Analysis: Forward-projection ofstock numbers at age using age distribution of the catch,calibrated with survey indices or other auxiliary informationin a likelihood-based framework.
7/29/2019 Session5b
5/45
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
Year
7/29/2019 Session5b
6/45
VPA
Reconstruction of allyearclasses gives a
total population
estimate.
Input Data:
catch at age
estimate of natural
mortality
initial guess aboutabundance 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
Year
7/29/2019 Session5b
7/45
VPA Calibration
Initial guesses are replacedwith estimates:
oldest age of historical
yearclasses estimated by
assuming that age-7 fish havethe 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
Year
( )tZttt
teF
ZCN
=
1
7/29/2019 Session5b
8/45
VPA Calibration
Abundance oflivingyearclasses in2002 areestimated using a
predictiverelationshipbetween
historical VPAabundance andsurvey indices.
0
5
10
15
20
25
0 10 20 30 40 50 60 70
Age-3 Abundance from VPA (millions)
Age-3SpringSurvey
Index(#/tow)
observed
predicted
2002
1990
(1987 yearclass)
7/29/2019 Session5b
9/45
VPA Estimates
Informative
Assessment:
Example: SNEyellowtail
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
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
FishingMortality(4-6)
F40%
0
5
10
15
20
25
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
SSB Year; Recruitment Yearclass
Spawning
Biomass('000smt)
0
20
40
60
80
100
120
140
Age-1Abundance(millions)
recruitment
SSB
7/29/2019 Session5b
10/45
Groundfish Stock Status - 1996
Biomass 1996 / B-MSY
0.0 0.3 0.5 0.8 1.0
F1996/F-MS
Y
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.05.5
6.0
GM Cod
GB Cod
GM HadGB Had
CC YT
GB YT
SNE YT
Witch
Plaice
GM WinGB Win
SNE Win
W Hake
Pol
RedfishPout
N Wind
S Wind
overfishingnot overfished
no overfishing
overfished
no overfishing
not overfished
overfishingoverfished
F-MSY
1/2 B-MSY
7/29/2019 Session5b
11/45
Groundfish Stock Status - 2002
Biomass 2002 / B-MSY
0.0 0.3 0.5 0.8 1.0 1.6 1.8 2.0
F2002/F-MS
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
PoutN Wind
S Wind
overfishingnot overfished
no overfishingoverfishedno overfishingnot overfished
overfishing
overfished
F-MSY
1/2 B-MSY
7/29/2019 Session5b
12/45
VPA Uncertainty
Similar to productionmodel: surveymeasurement errors arereshuffled many timesto estimate precision.
(bootstrapping). The estimate of 2001
SSB is 1850mt, with a
80% confidence limitof 1500 to 2500mt.
SNE Yellowtail
0
10
20
30
40
50
60
70
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
2500
2600
2700
2800
2900
3000
3100
3200
3300
2001 SSB (mt)
Freq
uency
7/29/2019 Session5b
13/45
Projections
"It is far better to
foresee evenwithout certainty
than not toforesee at all"
Poincare, The
Foundations ofScience
7/29/2019 Session5b
14/45
Short-Term Projection
Fluke (SAW35):
Landings in 2003would need to be
10,580 mt (23.3
million lbs) tomeet the target F
rate of Fmax =
0.26 with 50%probability.
7/29/2019 Session5b
15/45
Long-Term Projection
GeorgesBank Cod
the stock isexpected tohaveapproximately a 50%chance ofrebuilding to
SSBMSYby2026 iffished at anF of 0.18.
Year
2002 2006 2010 2014 2018 2022 2026
Spawn
ing
Biomass
(000sm
t)
0
50
100
150
200
250
300
75th percentile
Median Spawning Biomass
25th percentile
BMSY
7/29/2019 Session5b
16/45
Age-Structured Model
Population Numbers, Survival, Spawning Biomass
Catch, Landings, and Discards
Population Harvest
7/29/2019 Session5b
17/45
Population Numbers at Age
( )
( )( )
( )
( )
N t
N tN t
N t
N t
R
R
R
A
=
+
+
1
2
M
7/29/2019 Session5b
18/45
Survival by Age Class
N t N t e
for a R to A
a aM t F ta( ) ( ) ( ) ( )=
= +
11 11
1 1
7/29/2019 Session5b
19/45
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 11
7/29/2019 Session5b
20/45
Spawning Biomass
[ ]
SSB t W FM N t eS a a aZ t M t F t
a R
A
PROJ a
( ) ( ),( ) ( ) ( )= +
=
7/29/2019 Session5b
21/45
Catch Numbers at Age
[ ]C tF t
M t F t e N taa
a
M t F t
a
a
( )
( )
( ) ( ) ( )
( ) ( )=+
1
7/29/2019 Session5b
22/45
Landings
[ ]L t C t DF t Waa
A
a L a( ) ( ) ( ) ,= =1 1
7/29/2019 Session5b
23/45
Discards
D t C t DF t Wa a D aa
A
( ) ( ) ( ) ,= =1
7/29/2019 Session5b
24/45
Population Harvest
Input fully-recruited
fishing mortality F(t)
Input partial recruitment
vector PR(t) and ZPROJ(t)
Input discard fraction atage vector DF(t) if
applicable
Input landings quota Q(t)
Input partial recruitment
vector PR(t) and ZPROJ(t)
Input discard fraction at
age vector DF(t) ifapplicable
Solve for F(t)
7/29/2019 Session5b
25/45
Fishing Mortality at Age
F t F t PR ta a( ) ( ) ( )=
7/29/2019 Session5b
26/45
Catch Numbers at Age as a Function
of Fishing Mortality
[ ]C F
PR t F
M t PR t Fe N t
a
a
a
M t PR t F
a
a( )( )
( ) ( )( )( ) ( )=
+ 1
7/29/2019 Session5b
27/45
Landings as a Function of F
[ ]L F C F DF t Waa
A
a L a( ) ( ) ( ) ,= =1 1
7/29/2019 Session5b
28/45
Solve for Fishing Mortality to Harvest
Landings Quota
Q L F =( ) 0
7/29/2019 Session5b
29/45
Age-Structured Model
Stock-Recruitment Relationship
Initial Population Abundance
Abundance and Fishing Mortality Thresholds
7/29/2019 Session5b
30/45
Stock-Recruitment Relationship
Deterministic component
Stochastic component
( ) ( )N t f SSB t R tR ( ) ( ) , ,=
7/29/2019 Session5b
31/45
Recruitment Models Dependent on spawning biomass (n = 10)
Independent of spawning biomass (n = 5)
Uncorrelated stochastic component (n = 10)
Correlated stochastic component (n = 5)
7/29/2019 Session5b
32/45
Beverton-Holt Curve Lognormal Error
( )
n t a ssb t R
b ssb t R
e
where w N
R
w
w
( ) ( )
( )
~ ,
= +
0 2
7/29/2019 Session5b
33/45
Empirical Cumulative
Distribution Function
( )
N t T R R US
TR
where S U T
R S S S( ) ( )
( )
=
+
= +
+1
1
1
1 1
1
7/29/2019 Session5b
34/45
Georges Bank yellowtail flounder recruitment CDF
Recruitment (000,000s age-1 fish)
0 20 40 60 80 100
Frequency
R