Bering Sea Snow Crab Growth and Alternative Model Scenarios
Benjamin J. Turnock
Alaska Fisheries Science Center
April 16, 2013
THIS INFORMATION IS DISTRIBUTED SOLELY FOR THE PURPOSE OF PREDISSEMINATION PEER REVIEW
UNDER APPLICABLE INFORMATION QUALITY GUIDELINES. IT HAS NOT BEEN FORMALLY DISSEMINATED
BY NOAA FISHERIES/ALASKA FISHERIES SCIENCE CENTER AND SHOULD NOT BE CONSTRUED TO
REPRESENT ANY AGENCY DETERMINATION OR POLICY
Mean growth per molt by premolt width was estimated in the September 2012 snow crab
assessment model as a linear function with likelihood penalties on the intercept and slope. The
intercept was estimated as one parameter for male and female crab (as recommended by the
SSC) , while the slope was estimated as two sex specific parameters. There is a likelihood
penalty for each growth parameter in the September 2012 Base model. The parameters in the
likelihood penalty for the intercept and the male slope were based on the growth data collected
by Rugolo (cooperative seasonality study, 14 crab). The parameter for the female slope was
based on growth data from Canadian snow crab. The parameters for the slopes for males and
females were adjusted so that the growth curve with equal intercept was similar to estimates
from the Rugolo data for males and the Canadian curve for females (Table 1).
The September 2012 model estimates of mean growth for males was lower than the growth
based on priors and the female growth was estimated higher (Table 2 and Figure 1).
Somerton’s (2013) estimates on growth for Bering sea snow crab combined several data sets as
well as female and male data. The best model determined by Somerton(2013) included the
following data :
1.
2.
3.
4.
Transit study; 14 crab
Cooperative seasonality study (Rugolo); 6 crab
Dutch harbor holding study; 9 crab
NMFS Kodiak holding study held less than 30 days; 6 crab
Total sample size was 35 crab. Somerton(2013) excluded data from the NMFS Kodiak holding
study with crab held more than 30 days and also for the ADF&G Kodiak holding study where
crab were collected during the summer survey and held until molting in the next spring because
growth was lower significantly lower than the above four data sets.
Some data points were excluded from 1, 2 and 3 above (above has final sample size). Females
molting to maturity (molt increment is smaller) were excluded from all data sets. Crab missing
more than two limbs were excluded due to other studies showing lower growth. Crab from
Rugolo’s seasonal study were excluded that were measured less than 3 days after molting due to
difficulty in measuring soft crab accurately.
Somerton fit each data set starting with (1) above and testing the next data set for significant
difference. Two linear models were fit that joined at 36.1 mm (Figure 2),
For < =36.1mm
Postmolt = -4.0 + 1.46 * Premolt
>= 36.1 mm
Postmolt = 6.59 + 1.17 * Premolt
The fitted parameters with standard errors are shown in Table 3. The uncertainty for intercept
values is higher and for slope parameters lower than the growth priors used in the September
2012 Base model.
The assessment model was revised to include the joined linear growth with parameters equal for
males and females (Scenario 1) and also with separate parameters for males and females (except
a1 which was the same for males and females)(Scenario 2).
For < =35 mm (first two length bins 25-30, 30-35mm)
Postmolt = a1 + b1 * Premolt
>= 35 mm (all length bins >=35mm)
Postmolt = a2 + b2 * Premolt
A Likelihood penalty was added for each parameter:
0.5(
pest  p Som 2
)
sd p
Where pest is the estimated parameter in the model, psom and sdp are the values from Somerton in
Table 3.
A likelihood component was added to constrain the two linear segments of the mean growth to
be similar at 36.1 mm,
0.5(
P1  P 2 2
)
sd
sd = 0.2 (assumed)
P1 is postmolt size at 36.1mm premolt size with a1 and b1 parameters. P2 is postmolt size at
36.1mm premolt size with a2 and b2 parameters.
Mean growth for scenario 1 for male crab was less at small sizes and higher at larger sizes than
the Base model. Estimated growth for female crab is higher than base model.
Scenario 1 has one more parameter and a higher total likelihood (104 higher) than the Base
model (Table 4). Survey length likelihood is 40 points more, discard catch 20 and growth
penalty, 18. Scenario 2 has 49 higher likelihood points than the Base model. 29 more growth
prenalty, 8 survey length and 10 discard catch. Scenario 2 growth for males was very similar to
the base model above the first two size bins. Estimated female growth was lower in the first 2
sizes bins and higher at larger size bins than the base model.
Fits for the September 2012 Base model are in Figures 4-11.
Fits for the Scenario 1 model are in Figures 12-19.
Fits for the Scenario 2 model are in Figures 20-27.
Figures 28 and 29 compare population mature biomass for males and females for each Scenario.
Scenario 1 female mature biomass is higher than the Base 2012 and Scenario 2 biomass due to
lower survey Q (Table 6) and higher growth for females. Male mature biomass is very similar
between the alternative scenarios.
Literature Cited
Somerton, D., S. Goodman, R. Foy, L. Rugolo and L. Slater. 2013. Growth per Molt of Snow
Crab in the Eastern Bering Sea, North American Journal of Fisheries Management, 33:1, 140147.
Table 1. Empirical growth parameters.
parameter Se
intercept male and
female
6.773
0.3
male slope
1.16
0.1
female slope
1.05
0.1
Table 2. Growth parameters estimated in Base model, september 2012.
intercept male and female
7.45743
female slope
1.06336
male slope
1.13306
Table 3. Parameters for best model estimated by Somerton (2013).
param
<36.1mm
Intercept
slope
Se
-4.0 1.36498
1.46 0.05234
>36.1mm
Intercept
slope
6.59 0.73012
1.17 0.01186
Table 4. Likelihood values.
Likelihood Component
no. parameters
Recruitment
Initial numbers old shell males small length
bins
ret fishery length
Base
Model
2012
sept
Scenario 1
New
growth MF
same
Scenario 2
New growth
MF sep
329
330
333
33.19
34.78
33.84
2.38
2.40
-1946.99
794.32
176.42
3530.80
259.43
-78.99
-68.61
4.80
52.02
5.40
15.42
0.16
0.06
3.46
138.31
10.28
62.20
168.43
82.58
0.51
1.97
519.49
-61.01
-74.08
3.81
54.65
2.39
-1953.56
790.52
173.41
3501.52
255.39
-79.43
-70.43
5.92
48.85
6.56
25.11
0.13
0.06
3.32
129.86
9.65
64.87
167.09
80.30
0.51
2.20
515.69
-59.44
-75.72
3.86
54.89
-1954.44
total fish length
788.61
female fish length
175.98
survey length
3493.46
trawl length
256.76
2009 BSFRF length
-79.33
2009 NMFS study area length
-70.34
M prior
6.13
maturity smooth
50.61
growth intercept
2.60
growth slope and join constraint
0.05
2009 BSFRF biomass
0.14
2009 NMFS study area biomass
0.06
retained catch
3.03
discard catch
trawl catch
female discard catch
survey biomass
F penalty
119.83
9.78
65.10
165.24
80.25
2010 BSFRF Biomass
0.54
2010 NMFS Biomass
2.30
initial numbers fit
515.55
2010 BSFRF length
-60.96
2010 NMFS length
-76.88
male survey selectivity smooth constraint
init nos smooth constraint
3.83
54.77
Total
3588.26
3692.03
3637.33
Table 5. Growth parameters.
Base
model
Sept
2012
females
a1
b1
a2
b2
males
a1
Scenario 2
Scenario 1 M, F
M=F
separate
equal male equal male
1.30
equal male
4.64
equal male
1.13
equal male
0.08
1.34
5.61
1.16
b1
a2
b2
-0.86
1.38
7.18
1.14
September
2012 Base
model
Female
a
b
Males
a
b
equal
male
1.063
7.457
1.133
Table 6. Survey Q and natural mortality parameters.
Survey Q 1989-Present
Survey Q 1989-Present
M
M
M
male
female
Immature male and
female
mature female (fixed)
mature male
Base Scenario Scenario
2012 1
2
0.59
0.59
0.59
0.58
0.52
0.57
0.329
0.230
0.273
0.330
0.230
0.268
0.334
0.230
0.272
30
male growth priors
male estimated
Somerton growth
female estimated
25
Somerton crab<36.1mm
Molt Increment (mm)
Female priors
20
15
10
5
20
40
60
80
100
120
140
Pre-molt Width (mm)
Figure 1. Somerton (2013) estimated growth with Base model priors and estimated growth.
35
somerton
30
estimated MF same
model 6 male
Molt Increment (mm)
25
model 6 female
20
15
10
5
0
0
20
40
60
80
100
120
Pre-Molt width (mm)
Figure 2. Somerton (2013) estimated growth with Base model and Scenario 1 estimated growth.
140
35
somerton
30
estimated MF same
model 6 male
Molt Increment (mm)
25
model 6 female
female sep
20
male sep
15
10
5
0
0
20
40
60
80
100
120
140
Pre-Molt width (mm)
Figure 3. Estimated growth for males and females (dashed lines) separate parameters (except a1), same
parameters, estimated in Base model September 2012, and estimated by Somerton(2013) for male and
females together.
Base Model Figures
survey females
1978
40
60
80
100
120
Length(cm)
40
60
80
100
1983
120
Length(cm)
40
60
80
100
60
80
100
120
60
80
100
120
60
80
100
120
60
80
100
40
2003
120
Length(cm)
40
40
1998
Length(cm)
40
40
1993
Length(cm)
40
40
1988
Length(cm)
40
40
40
2008
120
40
1979
60
80
100
120
Length(cm)
60
80
100
1984
120
Length(cm)
60
80
100
80
100
120
80
100
120
80
100
120
80
100
40
2004
120
Length(cm)
60
40
1999
Length(cm)
60
40
1994
Length(cm)
60
40
1989
Length(cm)
60
40
40
2009
120
40
1980
60
80
100
120
Length(cm)
60
80
100
1985
120
Length(cm)
60
80
100
80
100
120
80
100
120
80
100
120
80
100
40
2005
120
Length(cm)
60
40
2000
Length(cm)
60
40
1995
Length(cm)
60
40
1990
Length(cm)
60
40
40
2010
120
40
1981
60
80
100
120
Length(cm)
60
80
100
1986
120
Length(cm)
60
80
100
80
100
120
80
100
120
80
100
120
80
100
40
2006
120
Length(cm)
60
40
2001
Length(cm)
60
40
1996
Length(cm)
60
40
1991
Length(cm)
60
40
40
2011
120
40
60
80
100
120
Length(cm)
60
80
100
120
Length(cm)
60
80
100
120
Length(cm)
60
80
100
120
Length(cm)
60
80
100
120
Length(cm)
60
80
100
120
Length(cm)
60
80
100
120
Figure 4. Base Model. Model fit to the survey female size frequency data. Circles are observed
survey data. Solid line is the model fit.
70
80
Standardized Pearson Residual Range -3.323 5.906
60
*
50
*
*
*
40
*
*
*
30
*
*
1980
1985
1990
* *
1995
2000
2005
2010
Figure 5. Base Model. Residuals of fit to survey female size frequency. Filled circles are
negative residuals.
1978
40
60
80
100
120
Length(cm)
40
60
80
100
1983
120
Length(cm)
40
60
80
100
60
80
100
120
60
80
100
120
60
80
100
120
60
80
100
40
2003
120
Length(cm)
40
40
1998
Length(cm)
40
40
1993
Length(cm)
40
40
1988
Length(cm)
40
40
40
2008
120
40
1979
60
80
100
120
Length(cm)
60
80
100
1984
120
Length(cm)
60
80
100
80
100
120
80
100
120
80
100
120
80
100
40
2004
120
Length(cm)
60
40
1999
Length(cm)
60
40
1994
Length(cm)
60
40
1989
Length(cm)
60
40
40
2009
120
40
survey males 1980
60
80
100
120
Length(cm)
60
80
100
1985
120
Length(cm)
60
80
100
80
100
120
80
100
120
80
100
120
80
100
40
2005
120
Length(cm)
60
40
2000
Length(cm)
60
40
1995
Length(cm)
60
40
1990
Length(cm)
60
40
40
2010
120
40
1981
60
80
100
120
Length(cm)
60
80
100
1986
120
Length(cm)
60
80
100
80
100
120
80
100
120
80
100
120
80
100
40
2006
120
Length(cm)
60
40
2001
Length(cm)
60
40
1996
Length(cm)
60
40
1991
Length(cm)
60
40
40
2011
120
40
60
80
100
120
Length(cm)
60
80
100
120
Length(cm)
60
80
100
120
Length(cm)
60
80
100
120
Length(cm)
60
80
100
120
Length(cm)
60
80
100
120
Length(cm)
60
80
100
120
Figure 6. Base Model. Model fit to the survey male size frequency data. Circles are observed
survey data. Solid line is the model fit.
120
100
80
Length bin
60
*
*
*
40
*
*
*
* *
*
1980
*
1985
1990
*
1995
2000
2005
2010
Year
Standardized Pearson Residual Range -2.101 2.963
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
Figure 7. Base Model. Residuals for fit to survey male size frequency. . Filled circles are
negative residuals (predicted higher than observed).
10
8
6
4
2
0
Sum of Length Proportions Survey
40
60
80
100
120
Carapace Width(mm)
600
400
200
0
Female Mature Biomass (1000 t)
Figure 8. Base Model. Summary over years of fit to survey length frequency data by sex. Dotted
line is fit for females, circles are observed. Solid line is fit for males, triangles are observed.
1980
1985
1990
1995
2000
2005
2010
Year
Figure 9. Base Model. Population female mature biomass (1000 t, dotted line), model estimate
of survey female mature biomass (solid line) and observed survey female mature biomass with
approximate lognormal 95% confidence intervals.
600
400
200
0
Male Mature Biomass (1000 t)
1980
1985
1990
1995
2000
2005
2010
Year
Figure 10. Base Model. Population male mature biomass (1000 t, dotted line), model estimate
of survey male mature biomass (solid line) and observed survey male mature biomass with
approximate lognormal 95% confidence intervals.
Figure 11. Base Model. Model fit to male directed discard catch for 1992/93 to 2010/11 and
estimated male discard catch from 1978 to 1991.
120
0.00 0.10 0.20
0.00 0.10 0.20
120
120
0.00 0.10 0.20
120
120
Length(cm)
Length(cm)
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
120
0.00 0.10 0.20
0.00 0.10 0.20
120
Length(cm)
120
Length(cm)
120
2007
40 60 80
120
Length(cm)
2011
40 60 80
2002
40 60 80
Length(cm)
120
120
Length(cm)
2006
40 60 80
120
1997
40 60 80
Length(cm)
2010
40 60 80
120
2001
40 60 80
Length(cm)
2009
40 60 80
120
2005
40 60 80
Length(cm)
2008
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
40 60 80
40 60 80
Length(cm)
2000
1992
Length(cm)
1996
40 60 80
Length(cm)
2004
40 60 80
Length(cm)
0.00 0.10 0.20
120
Length(cm)
2003
120
Length(cm)
1999
40 60 80
Length(cm)
Length(cm)
40 60 80
120
Length(cm)
1995
120
Length(cm)
1991
40 60 80
120
1987
40 60 80
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
1998
40 60 80
120
Length(cm)
120
0.00 0.10 0.20
40 60 80
40 60 80
Length(cm)
120
1982
Length(cm)
1986
40 60 80
Length(cm)
1994
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
120
Length(cm)
120
1990
40 60 80
Length(cm)
1993
40 60 80
120
120
0.00 0.10 0.20
40 60 80
40 60 80
Length(cm)
1989
1981
Length(cm)
1985
40 60 80
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
120
Length(cm)
40 60 80
120
Length(cm)
1988
40 60 80
1984
40 60 80
120
Length(cm)
0.00 0.10 0.20
120
Length(cm)
40 60 80
40 60 80
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
1983
40 60 80
120
Length(cm)
0.00 0.10 0.20
40 60 80
0.00 0.10 0.20
120
Length(cm)
1980
survey females
0.00 0.10 0.20
40 60 80
1979
0.00 0.10 0.20
1978
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
Scenario 1 Figures
120
2012
40 60 80
120
Length(cm)
Figure 12. Scenario 1. Model fit to the survey female size frequency data. Circles are observed
survey data. Solid line is the model fit.
80
70
60
50
*
*
*
40
*
*
*
*
30
*
*
*
1980
1985
1990
*
*
1995
2000
2005
2010
Figure 13. Scenario 1.Residuals of fit to survey female size frequency. Filled circles are negative
residuals.
40 60 80
Length(cm)
0.04 0.08
0.00
0.04 0.08
0.04 0.08
0.00
0.04 0.08
0.04 0.08
0.00
0.04 0.08
0.04 0.08
0.00
0.04 0.08
40 60 80
Length(cm)
120
0.04 0.08
2002
40 60 80
120
0.04 0.08
Length(cm)
2007
0.00
120
40 60 80
120
Length(cm)
0.04 0.08
2010
120
0.00
120
2006
40 60 80
0.04 0.08
0.04 0.08
120
40 60 80
Length(cm)
0.00
0.04 0.08
120
1997
2011
2012
0.00
40 60 80
Length(cm)
2009
2001
40 60 80
0.04 0.08
0.04 0.08
120
0.00
0.00
Length(cm)
120
2005
120
Length(cm)
0.00
40 60 80
120
Length(cm)
0.00
0.04 0.08
2004
Length(cm)
2008
120
Length(cm)
0.00
0.00
120
Length(cm)
40 60 80
40 60 80
1992
40 60 80
Length(cm)
2000
120
Length(cm)
1996
40 60 80
0.04 0.08
0.04 0.08
120
120
0.00
40 60 80
Length(cm)
2003
120
0.00
0.04 0.08
1999
1991
40 60 80
Length(cm)
0.00
0.00
120
Length(cm)
40 60 80
40 60 80
1987
40 60 80
Length(cm)
1995
120
Length(cm)
0.00
0.04 0.08
120
120
0.00
40 60 80
Length(cm)
1998
120
0.00
0.04 0.08
120
Length(cm)
40 60 80
40 60 80
40 60 80
Length(cm)
1990
1982
Length(cm)
1986
40 60 80
Length(cm)
1994
120
0.00
0.04 0.08
120
0.00
0.00
40 60 80
120
Length(cm)
1989
40 60 80
40 60 80
Length(cm)
1985
40 60 80
Length(cm)
1993
120
0.00
0.04 0.08
120
0.00
0.04 0.08
120
0.00
0.00
0.04 0.08
0.04 0.08
0.04 0.08
40 60 80
Length(cm)
0.04 0.08
1984
Length(cm)
1988
40 60 80
40 60 80
Length(cm)
0.00
0.04 0.08
0.00
120
Length(cm)
0.04 0.08
120
survey males
1981
0.00
40 60 80
Length(cm)
1983
40 60 80
0.04 0.08
0.00
0.04 0.08
120
Length(cm)
1980
0.00
40 60 80
0.04 0.08
1979
0.00
0.04 0.08
0.00
1978
40 60 80
Length(cm)
120
40 60 80
120
Length(cm)
Figure 14. . Scenario 1. Model fit to the survey male size frequency data. Circles are observed
survey data. Solid line is the model fit.
120
100
80
Length bin
60
*
*
*
40
*
*
*
* *
*
*
1980
1985
*
1990
1995
2000
2005
2010
Year
Standardized Pearson Residual Range -2.111 2.775
Figure 15. Scenario 1. Residuals for fit to survey male size frequency. . Filled circles are
negative residuals (predicted higher than observed).
10
8
6
4
0
2
Sum of Length Proportions Survey
40
60
80
100
120
Carapace Width(mm)
Figure 16. Scenario 1.Summary over years of fit to survey length frequency data by sex. Dotted
500
400
300
200
0
100
Female Mature Biomass (1000 t)
600
700
line is fit for females, circles are observed. Solid line is fit for males, triangles are observed.
1980
1985
1990
1995
2000
2005
2010
Year
Figure 17. Scenario 1. Population female mature biomass (1000 t, dotted line), model estimate
of survey female mature biomass (solid line) and observed survey female mature biomass with
approximate lognormal 95% confidence intervals.
600
400
200
0
Male Mature Biomass (1000 t)
1980
1985
1990
1995
2000
2005
2010
Year
Figure 18. Scenario 1.Population male mature biomass (1000 t, dotted line), model estimate of
20
15
0
5
10
Male discard catch
25
30
35
survey male mature biomass (solid line) and observed survey male mature biomass with
approximate lognormal 95% confidence intervals.
1980
1985
1990
1995
2000
2005
2010
Fishery Year
Figure 19. Scenario 1.Model fit to male directed discard catch for 1992/93 to 2010/11 and
estimated male discard catch from 1978 to 1991.
120
0.00 0.10 0.20
0.00 0.10 0.20
120
120
0.00 0.10 0.20
120
120
Length(cm)
Length(cm)
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
120
0.00 0.10 0.20
0.00 0.10 0.20
120
Length(cm)
120
Length(cm)
120
2007
40 60 80
120
Length(cm)
2011
40 60 80
2002
40 60 80
Length(cm)
120
120
Length(cm)
2006
40 60 80
120
1997
40 60 80
Length(cm)
2010
40 60 80
120
2001
40 60 80
Length(cm)
2009
40 60 80
120
2005
40 60 80
Length(cm)
2008
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
40 60 80
40 60 80
Length(cm)
2000
1992
Length(cm)
1996
40 60 80
Length(cm)
2004
40 60 80
Length(cm)
0.00 0.10 0.20
120
Length(cm)
2003
120
Length(cm)
1999
40 60 80
Length(cm)
Length(cm)
40 60 80
120
Length(cm)
1995
120
Length(cm)
1991
40 60 80
120
1987
40 60 80
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
1998
40 60 80
120
Length(cm)
120
0.00 0.10 0.20
40 60 80
40 60 80
Length(cm)
120
1982
Length(cm)
1986
40 60 80
Length(cm)
1994
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
120
Length(cm)
120
1990
40 60 80
Length(cm)
1993
40 60 80
120
120
0.00 0.10 0.20
40 60 80
40 60 80
Length(cm)
1989
1981
Length(cm)
1985
40 60 80
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
120
Length(cm)
40 60 80
120
Length(cm)
1988
40 60 80
1984
40 60 80
120
Length(cm)
0.00 0.10 0.20
120
Length(cm)
40 60 80
40 60 80
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
1983
40 60 80
120
Length(cm)
0.00 0.10 0.20
40 60 80
0.00 0.10 0.20
120
Length(cm)
1980
survey females
0.00 0.10 0.20
40 60 80
1979
0.00 0.10 0.20
1978
0.00 0.10 0.20
0.00 0.10 0.20
0.00 0.10 0.20
Scenario 2 Figures.
120
2012
40 60 80
120
Length(cm)
Figure 20. Scenario 2. Model fit to the survey female size frequency data. Circles are observed
survey data. Solid line is the model fit.
80
70
60
*
50
*
*
*
40
*
*
*
30
*
*
*
1980
1985
1990
*
*
1995
2000
2005
2010
Figure 21. . Scenario 2. Residuals of fit to survey female size frequency. Filled circles are
negative residuals.
Figure 22. . Scenario 2. Model fit to the survey male size frequency data. Circles are observed
survey data. Solid line is the model fit.
120
100
80
Length bin
60
*
*
*
40
*
*
*
* *
*
*
1980
1985
*
1990
1995
2000
2005
2010
Year
Standardized Pearson Residual Range -2.087 2.955
Figure 23. . Scenario 2. Residuals for fit to survey male size frequency. . Filled circles are
negative residuals (predicted higher than observed).
10
8
6
4
0
2
Sum of Length Proportions Survey
40
60
80
100
120
Carapace Width(mm)
Figure 24. . Scenario 2. Summary over years of fit to survey length frequency data by sex.
500
400
300
200
0
100
Female Mature Biomass (1000 t)
600
700
Dotted line is fit for females, circles are observed. Solid line is fit for males, triangles are
observed.
1980
1985
1990
1995
2000
2005
2010
Year
Figure 25. Scenario 2. Population female mature biomass (1000 t, dotted line), model estimate
of survey female mature biomass (solid line) and observed survey female mature biomass with
approximate lognormal 95% confidence intervals.
600
400
200
0
Male Mature Biomass (1000 t)
1980
1985
1990
1995
2000
2005
2010
Year
Figure 26. . Scenario 2. Population male mature biomass (1000 t, dotted line), model estimate
20
15
0
5
10
Male discard catch
25
30
35
of survey male mature biomass (solid line) and observed survey male mature biomass with
approximate lognormal 95% confidence intervals.
1980
1985
1990
1995
2000
2005
2010
Fishery Year
Figure 27. . Scenario 2. Model fit to male directed discard catch for 1992/93 to 2010/11 and
estimated male discard catch from 1978 to 1991.
Female Mature Biomass (1000t)
700
Base 2012
600
Scenario 1
500
Scenario 2
400
300
200
100
0
1970
1980
1990
2000
2010
2020
Survey Year
Figure 28. Population female mature biomass (1000t) for Base 2012, Scenario 1 and Scenario 2.
Male Mature Biomass (1000t)
600
Base 2012
500
Scenario 1
400
Scenario 2
300
200
100
0
1970
1980
1990
2000
2010
2020
Survey Year
Figure 29. Population male mature biomass (1000t) for Base 2012, Scenario 1 and Scenario 2.