An MSE for Pacific Sardine

Lecture 15

2

Fisheries Management(or what goes up must…)

Murphy (1966) & Hill et al. (2009)

0

100,000

200,000

300,000

400,000

500,000

600,000

700,000

800,000

1916

1919

1922

1925

1928

1931

1934

1937

1940

1943

1946

1949

1952

1955

1958

1961

1964

1967

1970

1973

1976

1979

1982

1985

1988

1991

1994

1997

2000

2003

2006

Season

Lan

din

gs

(mt)

0%

10%

20%

30%

40%

50%

60%

70%

Har

vest

Rat

e

Landings (mt) Harvest Rate

3

Fishery Development and Management-I

• Fishery developed during World War I• Peak catch over ~700,000t (largest fishery in

the western hemisphere in the 1930s and 1940s).

• Southeast shift in catch as fishery collapsed. Landings stopped in– 1947-48 in the Pacific Northwest– 1953-52 off San Francisco

4

Fishery Development and Management-II

• In the early 1980s, sardines were taken incidentally with Pacific and jack mackerel

• Incidental fishery for sardines ended in 1991• Management authority for the fishery transferred

to the Pacific Fishery Management Council (PFMC) in January 2000

• The Coastal Pelagic Species Management Plan includes:– Pacific sardine, Pacific mackerel, jack mackerel, anchovy,

market squid

Pacific Sardine Recovery and Expansion

1980s: low abundance,

confined to SCA; minor fisheries in SCA & Mexico

1990s: Expansion offshore

and north to Central California;

CCA fishery begins; Pop’n growth = 33%; Sardine in Oregon

Washington and Canada

2000s: Fisheries in PNW Seasonal

movements N-S, inshore/offshore

San Pedro

Ensenada

WashingtonOregon

Monterey

British Columbia

2000s

90s

80s

6

Challenges for Assessment and Management-I

• Multinational fishery:– US, Mexico, Canada

• Time-varying migration• Multiple fisheries in the US (Southern

California, Central California, Pacific Northwest)

• Environmental-related variation in biomass

7

Challenges for Assessment and Management

Sardine scale-deposition in the Santa Barbara Basin (Soutar & Isaacs 1969; Baumgartner et al. 1992).

•Extreme population variability even in absence of fishing;• periods of peak abundance ~ 50-60 years• link to environmental forcing is assumed

•Typical population dynamic for an ‘R-selected’ species:• small body, rapid growth, early maturation, high fecundity, short generation time,

and the ability to disperse offspring widely

NSP Distribution and Fishing Areas

U.S. Department of Commerce | National Oceanic

and Atmospheric Administration | NOAA

Fisheries | Page 8

Northern Subpopulation DistributionFishing Areas, & Modeled Fleets

Summer/Fall (Feeding)

Winter/Spring(Spawning) So. Calif. (SCA)

Ensenada(ENS)

Washington (WA)

Oregon (OR)

Monterey Bay (CCA)

Vancouver Island (BC)

PacNW Fleet

MexCal Fleet

Southern subpopulation

NSP Distribution & General Survey Areas

U.S. Department of Commerce | National Oceanic

and Atmospheric Administration | NOAA

Fisheries | Page 9

So. Calif. (SCA)

Ensenada(ENS)

Washington (WA)

Oregon (OR)

Monterey Bay (CCA)

Vancouver Island (BC)

NWSS Aerial Survey (summer):2009-2012

Summer ATM Surveys:2008, 2012, 2013

SWFSC Spring Survey:DEPM/TEP 1994-2013;

ATM 2006-2013

Estimated Stock Biomass Series from Base Model and Previous Stock Synthesis Models

11

Harvest Control (US)HG2010 = (BIOMASS2009 – CUTOFF) • FRACTION • DISTRIBUTION

To determine an appropriate (sustainable) FRACTION value:

FMSY = 0.248649805(T2)−8.190043975(T)+67.4558326

where T (oC) is the running average sea-surface temperature at Scripps Pier during the three preceding seasons (July-June), and exploitation FRACTION is bounded between 5% and 15%.Maximum catch allowed = 200,000 mt

Stock biomass

(age 1+, mt) Cutoff (mt)Harvest Fraction

U.S. Distribution

U.S. Harvest for 2011 (mt)

537,173 150,000 0.15 0.87 50,526

Pacific Sardine: Management Process(Amendment 8 & 13)

Overfishing Level(OFL)

Harvest Guideline(HG)

Acceptable BiologicalCatch (ABC)

P*1+ biomass

OFL

Increasing SST

Increasing SST

1+ biomass

HG

Calculate OFLOFL=Biomass*Fraction*Distribution

Calculate ABCABC=OFL*P*

Calculate HGHG=(Biomass-Cutoff)*

Fraction*Distribution

Final HG

SST

SSTE M

SY

Select the minimum from these two quantities

The HG is set following this basic process

Risk Assessment Framework

OFL, ABC, HGControl Rules

Remove catch from population

Recruitment

EnvironmentalDriver

Monitoringdata

15

Performance measures

The performance measures are selected to quantify performance relative to [some] management goals.• Average catch (total)• Average population size (1+ biomass)• Probability [total] catch is less than some threshold (e.g. 50,000t)• Probability 1+ biomass is below a threshold.

SIO is the index being used in the current harvest control rule. CalCOFI is the index

that better explains recruitment

Year

SS

T (

C)

1940 1960 1980 2000

14

16

18

20

SIOCalCOFI

SIO and CalCOFI indices have different scales, and are representative of different geographical areas.

SIO

17

Recruitment is related to both environment and spawning biomass

Series AIC R2

SST_CC_ann 44.49 0.76

SIO_SST_ann 56.81 0.61

ERSST_ann 55.3 0.63

The relation between several environmental indices and recruitment was evaluated.CalCOFI SST provides a better fit than SIO or ERSST to the stock-recruitment data for 1984-2008

From PFMC 2013, Table App.E.6

Modeling environmental drivers

1940 1960 1980 2000

17

.01

7.5

18

.01

8.5

19

.01

9.5

20

.0

Year

En

v. v

ari

ab

le

SQ wave, no ACSQ wave, With ACSine wave

Fit the environmental variable to the ERSTT time series

Calibrating the “CalCOFI” HG control rule

13 14 15 16 17 18

0.0

0.2

0.4

0.6

SST (C)

EM

SY

EMSY

E=0.05-0.18

SEMSY

PFMC June Meeting; Costa Mesa, CA, June, 2013

Calibrating the “CalCOFI” HG control rule

13 14 15 16 17 18

0.0

0.2

0.4

0.6

SST (C)

EM

SY

EMSY

E=0.05-0.18SST_CC_ann quartiles

Maximum harvest rate for computing OFL

CalCOFI-based harvest control rule

CalCOFI OFL

0

500

10001500

14.014.5

15.015.5

16.016.5

17.00

100

200

300

BiomassSST

OFL

CalCOFI HG

0

500

10001500

14.014.5

15.015.5

16.016.5

17.00

50

100

150

200

BiomassSST

HG

CalCOFI HG with ABC

0

500

10001500

14.014.5

15.015.5

16.016.5

17.00

50

100

150

200

BiomassSST

HG

OFL control rule HG control rule HG control rule < ABC

Cutoff

Maxcatch

Maximum harvest rate for OFL Maximum harvest

rate for HGPFMC June Meeting; Costa Mesa, CA, June, 2013

The harvest control rule depends on both the 1+ biomass and the CalCOFI SST

Bio

mas

s ('0

00t)

010

0020

0030

0040

00

Bio

ma

ss (

'00

0t)

ERSSTR

ecru

its (

'000

0)

050

010

0015

0020

00

Re

cru

its (

'00

00

t)

Year

SS

T

0 50 100 150 200

1415

1617

1819

20

SS

TCalCOFI

Year

0 50 100 150 200

Year

This is a comparison of model runs using ERSST (left) and CalCOFI (right) as the environmental driver of recruitment, and no catches

PFMC June Meeting; Costa Mesa, CA, June, 2013

23

50 100 150 200 250 300 3500

200

400

600

800

1000

1200

1400

1600

Average Catch ('000t)

Aver

age

1+ P

opul

ation

“current” HG

50 100 150 200 250 300 3500

20

40

60

80

100

120

Mean Catch ('000t)

Prob

(Bio

mas

s 1+

> 40

0,00

0t)

Performance measures aredefined over many cycles of the environmental variable and arenot predictions of short-termcatches / biomass.

45% harvest rate &high CUTOFF (variant 5)

No CUTOFF (variants 3&4)

24

50 100 150 200 250 300 3500

5

10

15

20

25

30

35

40

Mean Catch ('000t)

Prob

abili

ty ze

ro c

atch

50 100 150 200 250 300 3500

10

20

30

40

50

60

Mean Catch ('000t)

Prob

abili

ty C

atch

< 5

0,00

0t

“current” HG

45% harvest rate &high CUTOFF (variant 5)

25

20 25 30 35 40 45 5080

85

90

95

100

Probability catch < 50,000t

Prob

abili

ty 1

+ Bi

omas

s >

400,

000t

Performance depends on the values for the parameters determining, for example, how the environment impacts recruitment.

More extreme variation inthe environment than assumedfor the base-case

Base-case

26

60 80 100 120 140 160 180 200 220 240 2600

100

200

300

400

500

600

700

800

900

1000

Mean Catch

Mea

n 1+

bio

mas

s

“current” HG OFL only

Key Major Sensitivity: The assumption everyone follows the US control rules (for this test Mexico and Canada areassumed to have a constant fishing mortality rate no matter what)

27

Control Rules and MSE-I

• Design of the operating model– Single-species: Workshop with Council

involvement (Council members, staff, SSC, advisory bodies)

– Multi-species: Ongoing – currently being funding through a Packard grant

– Operating model reviewed by the Council Scientific and Statistical Committee (SSC)

• Selection of OFL control rules: the SSC!

28

Control Rules and MSE-II

• Selection of HG control variants:– Initially based on those considered when the

current Harvest Guideline control rule was adopted

– Additional options added by:• The Coastal Pelagic Species management team• The public (via the Council)

– Software made available to the public who provided suggestions based on their analyses.

29

Control Rules and MSE-III

• Performance measures for the MSE:• Initially based on those considered when the current

Harvest Guideline control rule was adopted• Performance statistics added based on input from the:

– Coastal Pelagic Species Management Team, – Coastal Pelagic Species Advisory Subpanel, and– Public (NGOs, industry).