A A multi-species multi-species population assessment population assessment model for the Gulf of model for the Gulf of Alaska Alaska Kray F. Van Kirk, SFOS, UAF, Juneau Terrance J. Quinn II, SFOS, UAF, Juneau Jeremy S. Collie, GSO, URI, Narragansett ftkv@uaf.edu
27
Embed
A multi-species population assessment model for the Gulf of Alaska Kray F. Van Kirk, SFOS, UAF, Juneau Terrance J. Quinn II, SFOS, UAF, Juneau Jeremy S.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
A A multi-species population multi-species population
assessment model for the Gulf assessment model for the Gulf of Alaskaof Alaska
Kray F. Van Kirk, SFOS, UAF, Juneau Terrance J. Quinn II, SFOS, UAF, Juneau Jeremy S. Collie, GSO, URI, Narragansett
Jointly funded by AK and RI Sea Jointly funded by AK and RI Sea GrantGrant
Also collaboration with North Sea Also collaboration with North Sea colleagues: Lewy, Vinther (Denmark)colleagues: Lewy, Vinther (Denmark)
Project started February, 2004Project started February, 2004 Species groupsSpecies groups
Gulf of AlaskaGulf of Alaska Georges BankGeorges Bank North SeaNorth Sea
Sea GrantSea Grant
JustificationJustification
Crisis in Fisheries ManagementCrisis in Fisheries Management Single species analysis does not Single species analysis does not
answer questions of multi-species answer questions of multi-species interactionsinteractions
Ecosystem approaches desiredEcosystem approaches desired Natural mortality is not constant!Natural mortality is not constant! Harvest strategies cannot evolve Harvest strategies cannot evolve
until models dountil models do
Previous MSVPA ModelsPrevious MSVPA Models Predation mortality estimated from Predation mortality estimated from
gut studiesgut studies Single predation coefficientSingle predation coefficient Works backwards from oldest agesWorks backwards from oldest ages Catch-at-age is measured without Catch-at-age is measured without
errorerror Applied to North, Baltic and Bering Applied to North, Baltic and Bering
Seas, Georges Bank, and a few Seas, Georges Bank, and a few tropical areastropical areas
Model OverviewModel Overview Catch-age Catch-age
measured with measured with errorerror
Forward time Forward time progression from progression from age 1 recruitmentage 1 recruitment
Cohort abundance Cohort abundance subject to fishing,subject to fishing,flexible predation,flexible predation,and residual and residual naturalnaturalmortalitymortality
taiZtaitai eNN ,,
,,1,1,
)(,1,1
,tiSBtiti eSBN
Age-structured AnalysisAge-structured Analysis
Age
3 4 5 6 7 8 ... Total
1989
1990
1991
1992
1993
1994
Year 1995
1996
1997
1998
1999
2000
Recruitment
Natural MortalityFishing MortalityGrowthMovement
Progression of aYear-class or Cohort
Spawner-RecruitRelationship
Modeling Predation MortalityModeling Predation MortalityAnnual Ingestion of Predator Age (grams)Annual Ingestion of Predator Age (grams)
= consumption of modeled prey of interest= consumption of modeled prey of interest
+ consumption of non-modeled ‘other food’.+ consumption of non-modeled ‘other food’.
Where annual ingestion rate is given by:Where annual ingestion rate is given by:
bj
bjbj GE
GII
,
,, 5.0*1
11.0
inf,
,,
j
bjbj W
wGE
Growth Increment
Growth Efficiency Hall et al. in press
Components of Modeled Prey (1)Components of Modeled Prey (1)
Predator Predator j j of size of size bb preference for prey of preference for prey of size size aa.. (from Ursin 1971, where size-preference is (from Ursin 1971, where size-preference is
a function of predator-prey weight-at-age a function of predator-prey weight-at-age ratio)ratio)
Predator Predator jj preference for prey of species preference for prey of species ii.. (from AFSC data, as a ratio of prey (from AFSC data, as a ratio of prey ii
consumed relative to total prey consumed; consumed relative to total prey consumed; changes over time)changes over time)
g~
Components of Modeled Prey (2)Components of Modeled Prey (2)
Suitability of prey Suitability of prey i,ai,a to predator to predator j,b j,b ::
Abundance at age for year 1(SAFE)Abundance at age for year 1(SAFE) Fixed overall F (Mean 1981 – 2001)Fixed overall F (Mean 1981 – 2001) Fixed R (0.1)Fixed R (0.1) Spawner-Recruit alpha and betaSpawner-Recruit alpha and beta Prey Preference (Mean 1981 – 2001)Prey Preference (Mean 1981 – 2001)
Changes in Prey PreferenceChanges in Prey PreferenceProportion of Walleye Pollock in
Pacific Cod diet by Decade
0.185
0.146
0.0620.000
0.050
0.100
0.150
0.200
0.250
1980 1990 2000
Proportion of Walleye Pollock inArrowtooth Flounder diet
0.3970.391
0.285
0.000
0.100
0.200
0.300
0.400
0.500
1980 1990 2000
Proportion of Walleye Pollock inWalleye Pollock diet by Decade
Next StepsNext Steps Parameter EstimationParameter Estimation
Rho (species preference)Rho (species preference) Elements of size-preference functionElements of size-preference function ‘‘Other food’ multiplierOther food’ multiplier
Objective FunctionsObjective Functions Fishery catch at ageFishery catch at age Survey catch at ageSurvey catch at age Stomach contents (predation rates)Stomach contents (predation rates)
Addition of Pacific halibut and Steller Addition of Pacific halibut and Steller Sea LionSea Lion
Areas for ExplorationAreas for Exploration Predation between age classes Predation between age classes
and connection with fishing and connection with fishing pressurepressure
Complexity of multi-species Complexity of multi-species model vs. robustness of single model vs. robustness of single species modelsspecies models
Estimation of recruitment Estimation of recruitment parameters once other parameters once other parameters are setparameters are set