NEMURO &NEMURO & NEMURO.FISHNEMURO.FISHShin-ichi Ito (Fisheries Research Agency)

Michio Kishi (Hokkaido Univ.)NEMURO Mafia

Today's contentsToday's contents

1. Brief history of ecosystem model(go through Prof. Fei Chai's lecture in the 1st PSS)

2. NEMURO (a lower trophic level EM)3. NEMURO.FISH (a higher trophic level EM)4. hands on

EcosystemEcosystemEcosystem is one component of the earth system.It is not static but dynamic.

Difficulties of Earth System ScienceDifficulties of Earth System ScienceHarte (2002)

1. The global scale of human activities and the historically unprecedented magnitude of human disturbance of the planet mean that past experience is often not a reliable guide to predicting the consequences of our actions.

2. The Earth system is rife with feedback, nonlinear synergies, thresholds, and irreversibilities, that confound our intuition.

3. Ecosystem is dynamically changing.4. Conducting large scale experiments on this system is

impossible.

Ecosystem modelingEcosystem modelingTherefore, we need models to test the ecosystem functions and responses and to understand its structure and functioning.

• Always we meet with tradeoff between "resolution or coverage in biological components" & "spatial resolution or coverage".

• Computational power is still limited.

ProblemsProblems

Strategy for Ecosystem ModelingStrategy for Ecosystem Modeling1. Nesting in spatial resolution2. Rhomboidal approach (deYoung et al., 2004)

increase resolution of target species.

From JCOPE home pagedeYoung et al. (2004)

Functional complexity

Tro

phic

leve

l

FermiFermi approach (simple models)approach (simple models)1. Complex models, which look as inscrutable as nature

itself, have numerous adjustable parameters and are generally unfalsifiable. However, it is impossible to validate the models.

2. Simple models that capture the essence of the problem, but not all the details, might get us farther (Fermi approach).

3. This "Fermi approach" to ESS will only be effective, of course, if decision makers can be weaned from their awe of computer-simulated complexity. Harte (2002)

Another or similar solutionAnother or similar solution

Everything should be as simple as possible, but no simpler.By Albert EinsteinBy Albert Einstein

history of marine ecosystem modelhistory of marine ecosystem modelRiley G. A. (1946) "Factors controlling phytoplankton population on Georges Bank.", J. Mar. Res., 6, 54-73.This is the first paper introduced differential equation to marine ecosystem modeling.

Dots:observationLine:model

Memoirs (G. A. Riley, 1984)Memoirs (G. A. Riley, 1984)

H. Bigelow said "Anybody who thinks he can predict more than 10% of plankton variability is a damn fool, but good luck".

To be sure, I have never pretended that [models] are truly realistic. My only defense has been that they help us to think….

They frequently yield results that are not intuitively obvious, and they teach us caution about drawing conclusions that seem to violate mathematical logic.

That is the way physics and astronomy have grown.

Biological oceanography, messy though it may be, needs the same kind of disciplined thinking.

milestones of NPZD modelsmilestones of NPZD modelsGorden Riley (1946)

Factors controlling phytoplankton population on Georges Bank.J. Mar. Res. 6, 54-73

John Steele (1974)The structure of marine ecosystems

Evans and Parslow (1985)A model of annual plankton cycles, Biol. Oceanogr., 24, 483-494

Fasham, Ducklow, McKelvie (1990)A Nitrogen-based model of plankton dynamics in the ocean mixed layerJ. Mar. Res., 48 (3): 591-639

Fasham (1995)Variations in the seasonal cycle of biological production in sub-arctic oceans - a model sensitivity analysis.Deep-Sea Res. Part I: 42 (7): 1111-1149

courtesy of Fei Chai

NEMURONEMUROAs a science project of PICES, CCCC (Climate Change and Carrying Capacity) was started in 1993.Under CCCC, "Conceptual theoretical modeling studies Task Team" was formed. =>MODEL Task Team

To develop a lower trophic ecosystem model which can be applied commonly to the North Pacific, PICES LTL model workshop was held in Nemuro in 2000.The developed model was names as NEMURO (North Pacific Ecosystem Model for Understanding Regional Oceanography).

NEMURONEMUROAs a science project of PICES, CCCC (Climate Change and Carrying Capacity) was started in 1993.Under CCCC, "Conceptual theoretical modeling studies Task Team" was formed. =>MODEL Task Team

To develop a lower trophic ecosystem model which can be applied commonly to the North Pacific, PICES LTL model workshop was held in Nemuro in 2000.The developed model was names as NEMURO (North Pacific Ecosystem Model for Understanding Regional Oceanography).

NEMURONEMURO(North Pacific Ecosystem Model for (North Pacific Ecosystem Model for

Understanding Regional Oceanography)Understanding Regional Oceanography)

1.1. NutrientNutrient--PhytoplanktonPhytoplankton--Zooplankton Zooplankton modelmodel

2.2. Developed under PICES Model Task Developed under PICES Model Task Team since 2000Team since 2000

3.3. Coupled with fish bioenergetics modelsCoupled with fish bioenergetics models4.4. Published special issue on Ecological Published special issue on Ecological

ModellingModelling in 2007in 20075.5. Published 37 papers on peer reviewed Published 37 papers on peer reviewed

journalsjournals6.6. Used in more than 8 countries (Canada, Used in more than 8 countries (Canada,

Russia, South Korea, China, USA, Japan, Russia, South Korea, China, USA, Japan, Mexico, Greek, etc.)Mexico, Greek, etc.)

NEMURONEMURONorth Pacific Ecosystem Model for Understanding

Regional Oceanography

Kishi et al. (2001), Kishi et al. (2007)

nutrient

phyto

zoo

ZLnGraPSZSnGraPSExcPSnMorPSnPSnResGppPSndt

dPSn 22 −−−−−=

GraPL2ZPnGraPL2ZLnExcPLnMorPLnsPLnRenGppPLdt

dPLn−−−−−=

EgeZSnExcZSnMorZSnZPnGraZSZLnGraZSZSnGraPSdt

dZSn−−−−−= 222

EgeZLnExcZLnMorZLnGraZL2ZPnGraZS2ZLnnZLaPLrGnZLaPSrGdt

dZLn−−−−++= 22

UPWnNitRnewLResPLn)nGppPLRnewSPSnResGppPSndt

dNO++−−−−= ()(3

ExcZPnExcZLnExcZSnDecD2NDecP2NNit

RnewL)ResPLn)(1(GppPLnRnewS)ResPSn)(1(GppPSndt

dNH

+++++−

−−−−−−=4

SEDnDecP2DDecP2NEgeZPn

EgeZLnEgeZSnMorZPnMorZLnMorZSnMorPLnMorPSndt

dPON

−−−+

++++++=

NDecDDDecPnExcPLExcPSndt

dDON 22 −++=

EgeZPnExcZPnMorZPnGraZL2ZPnnZPaZSrGnZPaPLrGdt

dZPn−−−++= 22

Nitrogen equationsNitrogen equations

GraPL2ZPsiGraPL2ZLsiExcPLiMorPLsisiPLRessiGppPLdt

dPLsi−−−−−=

EgeZLsisiZLLParGdt

dZLsi−= 2

EgeZPsisiZPLParGdt

dZPsi−= 2

SiDecPSEDsiEgeZPsiEgeZLsiMorPLsidt

dOpal 2−−++=

SiDecPUPWsiExcPLsiisResPLsiGppPLdtOHdSi 24)(

++++−=

Silicon equationsSilicon equations

ZLnGraPSZSnGraPSExcPSnMorPSnPSnResGppPSndt

dPSn 22 −−−−−=

e.g. small phytoplankton equatione.g. small phytoplankton equation

photosynthesisrespiration

mortalityExtracellular

excretionGrazing from

small zooplankton

Grazing from large

zooplankton

)()exp(

*1exp*)*exp(*

44)*exp(

33*

21

0

0

44

3max

PLnPSnZII

PSndzI

II

ITMPk

KNHNHNH

KNONOVGppPSn

optSH optSGppS

SNHS

SNOS

++=

−=

⎟⎠⎞

⎜⎝⎛ −

⎟⎠⎞

⎜⎝⎛

++Ψ−

+=

∫−

αακκ

SNHS

SNO

SSNO

KNHNHNH

KNONO

NHKNO

NO

RnewS

43

3

44)4*exp(

33

)4*exp(3

3

++Ψ−

+

Ψ−+=

ResPSn = ResPS0 * exp( kResPS * TMP ) * PSnMorPSn = MorPS0 * exp( kMorPS * TMP ) * PSn2

ExcPSn = γPS* GppPSn

{ }[ ]ZSnPSn)ZS*(PSTMPkSpsGRZSnGraPS SGraS * )2exp(-1 *)*exp(* 0,Max2 *max −= λ

{ }[ ]ZLnPSn)ZL*(PSTMPkLGRZLnGraPS LGraLps * )2exp(-1 *)*exp(* 0,Max2 *max −= λ

small phytoplankton equationssmall phytoplankton equations

1

1/2

K C

C C+K Michaelis-Menten

1

Iopt I

Light dependency

1

C* C

Ivlev function

1. Limiting factor for PhotosynthesisMultiplier of nutrient concentration, temp. and light.Another possibility is f(T)*min(f(N), f(L))

2. Natural mortalityProportional to square of plankton densityDue to stabilize the system (no theoretical reason) Insensitive to high frequency variability

(Wainwright et al. 2007)Since predation pressure from higher trophic already acts on planktons, it is possible to modify such as D**1.5.

3. Predation pressureCritical value is defined to avoid extinction

4. Many of parameters were borrowed from other models.PL photosynthesis parameters must be revised.

problemsproblems

NEMURONEMURONorth Pacific Ecosystem Model for Understanding

Regional Oceanography

Kishi et al. (2001), Kishi et al. (2007)

Ont

ogen

etic

ve

rtic

al m

igra

tion

Ecosystem resolution or coverage

Spac

e/tim

e re

solu

tion

or c

over

age

3D-PLANKTON10

0D-NPZDBox-NEMURO.anchovyBox-NEMURO.sardineBox-NEMURO.FISH

NEMPORT

SEAPODYMOSMOSE

NEMURO family (NEMURO family (GR: publishedGR: published, , RD:developingRD:developing))

3D-eNEMURO3D-FeNEMURO

tradeoff between (finer individual species) & (basin scale model)

box-NEMURO

1D-NEMURO

3D-NPZD

NEMUROMSCHOPE-NEMURO3D-NEMURO(1/4)

migrate-NEMURO.salmonmigrate-NEMURO.squid

box-eNEMURObox-FeNEMURO

Sponge-NEMURO

eNEMUROMSCHOPE-eNEMURO

NEMURO.SAN

eNEMURO.SAN

NEMUROMS-SAN

3D-PLANKTON53D-NEMURO

LTL modelsLTL models

Kuroshio path

Log10(Chl) [mg/m-3]SeaWiFS

model

2005/4/22

CHOPE + CHOPE + eNEMUROeNEMUROKomatsu et al. (in prep.)

CHOPE + CHOPE + eNEMUROeNEMURO

High resolution ocean circulation model was assimilated to satellite altimetry data.Therefore, physical part is close to the true.LTL model can capture the variability of phytoplankton well in this case.

(Yoshie et al., in prep.)

NEMURO.FISHNEMURO for Including Saury and Herring

Megrey et al. (2007a), Ito et al. (2004) etc.

Bioenergetics Model for herring and saury

change of weight

[ ]f

z

CALCALPEFSRC

dtWdW

⋅++++−=⋅

)(

E: excretion

C: consumption

R: respiration (loses through metabolism)

S: specific dynamic action(digesting food)

F: egestion

P: egg production

30

35

40

45

130 135 140 145 150 155 160 165

Kuroshio Extension

Oyashio Front

spawning groundin winter

spawning groundin spring & autumn

Mixed water region

Migration

Migration

Feeding groundin summer

Fishing groundin autumn

Life History of Pacific Saury with Oceanographic Features

Ito et al. (2004a)

Table 2. Life stages of Pacific saury in the saruy bioenergetics modelStage regionlarvae Kuroshiojuvenile & young mixed regionsmall Oyashioadult mixed regionadult matured Kuroshioadult mixed regionadult Oyashioadult mixed regionadult matured Kuroshio

3-box version

OyashioMixed Waterregion

Kuroshio

9 life stages

Ito et al. (2004b)Ito et al . (2007)

Mukai et al. (2007)

Simulated wet weight & observed growth (Kurita et al.)

maximum consumption rate multiplied by temperature function (solid line) & water temperature (dotted line).

Terms of the bioenergetics equationblack solid: consumptionred solid: respirationblue solid: egestionblack dotted: excretionred dotted: specific dynamic actiongreen: egg productionopen circle: observed consumption

by Kurita (2002)

Ito et al. (2004b)

Dynamic linkage between LTL & HTL

[ ]f

z

CALCALPEFSRC

dtWdW

⋅++++−=⋅

)(

NH4

Decrease ZS, ZL, ZP

PON

dN/dt = - (F+M) N

Population dynamics model

(b) Coupled

Year0 2 4 6 8 10

Ave

rage

wei

ght

(g w

et w

eigh

t)

0

50

100

150

200

250

(a) Uncoupled

0 2 4 6 8 10

Ave

rage

wei

ght

(g w

et w

eigh

t)

0

50

100

150

200

250

Age-0Age-1Age-2Age-3

Age-4Age-5Age-6Age-7

Age-8Age-9Age-10

Effect of Dynamic Linkage and Predation Pressure

Reduction in large zooplankton

Slower herring growth

(Megrey et al. 2007a)

Ecosystem resolution or coverage

Spac

e/tim

e re

solu

tion

or c

over

age

3D-PLANKTON10

0D-NPZDBox-NEMURO.anchovyBox-NEMURO.sardineBox-NEMURO.FISH

NEMPORT

SEAPODYMOSMOSE

NEMURO family (NEMURO family (GR: publishedGR: published, , RD:developingRD:developing))

3D-eNEMURO3D-FeNEMURO

tradeoff between (finer individual species) & (basin scale model)

box-NEMURO

1D-NEMURO

3D-NPZD

NEMUROMSCHOPE-NEMURO3D-NEMURO(1/4)

migrate-NEMURO.salmonmigrate-NEMURO.squid

box-eNEMURObox-FeNEMURO

Sponge-NEMURO

eNEMUROMSCHOPE-eNEMURO

NEMURO.SAN

eNEMURO.SAN

NEMUROMS-SAN

3D-PLANKTON53D-NEMURO

NEMURO.SANNEMURO.SAN

Held a workshop in Tokyo in 2005 to compare sardine & anchovy in California, Benguera, Funbolt, Oyashio/KuroshioCurrent systems.0

150

300

450

0

15

30

45Japan

0

150

300

450

0

350

700

1050Chile-Peru

0

50

100

150

1920 1930 1940 1950 1960 1970 1980 1990020406080100

Benguela

Sard

ine

catc

h x

104

(tons

)

Anchovy catch x 104(tons)

0

20

40

60

80

0

10

20

30

40

California

0

150

300

450

0

15

30

45Japan

0

150

300

450

0

350

700

1050Chile-Peru

0

50

100

150

1920 1930 1940 1950 1960 1970 1980 1990020406080100

Benguela

Sard

ine

catc

h x

104

(tons

)

Anchovy catch x 104(tons)

0

20

40

60

80

0

10

20

30

40

California

Common features:asynchronicity between sardine &

anchovygeographical expansion and reduction of

distribution

We decided model extensions:Two species (sardine & anchovy)Dynamic predator on sardine & anchovy

Provided by: Salvador E. Lluch-CotaSource: Schwartzlose et al., 1999

California Current SystemCalifornia Current System

Ligh

t (ly

/min

)

0.0

0.1

0.2

0.3

Mix

ed L

ayer

Dep

th (m

)

30

45

60

75

Day of Year0 60 120 180 240 300 360

Tem

pera

ture

(o C)

8

10

12

14

Day of Year0 60 120 180 240 300 360

Nut

rient

Exc

hang

e

0

5

10

15

Environmental VariablesAt West Vancouver Island

Light

Temp.

MLD

Nut. Exc.

Rose et al.(in prep.)

Experiment 1Experiment 1Year 1-10: Spin-upYear 11-20: warm (+2 degC)Year 21-30: cold (-2 degC)Year 31-40: warm (+2 degC)Year 41-50: cold (-2 degC)

Spa

wni

ng S

tock

Bio

mas

s (1

05 M

T)

0

3

6

9 A n c h o v yS a r d in e

Y e a r0 1 0 2 0 3 0 4 0 5 0

Mea

n W

eigh

t

at A

ge-4

(g)

3 0

4 5

6 0

7 5

spawning stock biomass (105MT)

mean weight at age-4 (g)

Rose et al.(in prep.)

Experiment 1Experiment 1

Year 1

5

10

15

20

25

30

35

40

0 1e+9 2e+9 3e+9 4e+9 5e+9 6e+9 7e+9 8e+9 9e+9 1e+10

Year 1

5

10

15

20

25

30

35

40

0.0 1.0e+9 2.0e+9 3.0e+9 4.0e+9 5.0e+9 6.0e+9 7.0e+9 8.0e+9 9.0e+9 1.0e+10 1.1e+10 1.2e+10 1.3e+10

4 8 12 16 20

5

10

15

20

25

30

35

40

0 5000 10000 15000 20000 25000 30000 35000 40000

1 10 20 30 40 50

Anchovy

Sardine

Predator

Rose et al.(in prep.)

NEMURONEMURO & NEMURO.FISH& NEMURO.FISH•LTL-HTL coupled model.•Now interspecies interaction is included.•Therefore, it is possible to apply NEMURO.FISH to ecosystem based management

•However, we must keep the fact in memories that the model parameters are not enough validated and other factors such as adaptability may change the parameters.•Models include errors (uncertainty) because of

lack of dynamics,insufficient determination of parameters,inapposite forcings, etc.

•Make model simple as possible as we can.

NEMURO.FISHNEMURO.FISHVery simple exercisesVery simple exercises

1. One box model and one fish.2. Physical forcing only includes SST.3. Deeper layer temperature is fixed.4. If the SST becomes cooler than DL temp, vertical mixing

ratio is increased by unstable stratification and nutrient is transported to the surface layer.

5. Idealized seasonal light intensity is assumed.6. Fish has two year life history and the parameters are

similar to Pacific saury.7. Fish natural mortality depends on body length.8. Annual fishery mortality is assumed to 30%.9. Reproduction depends on the size and food availability of

spawning stock.

Ex. 1. Climatological forcingEx. 1. Climatological forcing

0.0E+00

5.0E-02

1.0E-01

1.5E-01

2001/1/1 2003/1/1 2004/12/31

light

inte

nsity

(ly/

min

)

051015202530

wat

er te

mp.

(deg

C)

Lint0 TMP

0.0E+00

5.0E-03

1.0E-02

1.5E-02

2.0E-02

2.5E-02

2001/1/1 2003/1/1 2004/12/31

conv

ectio

n tim

e (1

/day

)

0

50

100

150

200

mix

ed la

yer d

epth

(m)

ExcTime MLD

Climatological forcingClimatological forcing

05

101520253035

2001/1/1

2003/1/1

2004/12/31

2006/12/31

2008/12/30

2010/12/30

(um

ol/l)

NO3 NH4 SiOH4

0

0.5

1

1.5

2

2001/1/1

2003/1/1

2004/12/31

2006/12/31

2008/12/30

2010/12/30

(um

ol/l)

PS PL

00.10.20.30.40.50.60.7

2001/1/1

2003/1/1

2004/12/31

2006/12/31

2008/12/30

2010/12/30

(um

ol/l)

ZS ZL ZP

00.10.20.30.40.50.60.7

2001/1/1

2003/1/1

2004/12/31

2006/12/31

2008/12/30

2010/12/30

(um

ol/l)

PON DON Opal

Climatological forcingClimatological forcing

0.0E+00

5.0E-01

1.0E+00

1.5E+00

2.0E+00

2.5E+00

3.0E+00

2001/1/1

2003/1/1

2004/12/31

2006/12/31

2008/12/30

2010/12/30

density(1) density(2) density(3)

0

20

40

60

80

100

120

140

160

2001/1/1

2003/1/1

2004/12/31

2006/12/31

2008/12/30

2010/12/30

2012/12/29

w(1) w(2) w(3)

0.0E+00

1.0E-03

2.0E-03

3.0E-03

4.0E-03

5.0E-03

6.0E-03

7.0E-03

8.0E-03

2001/1/1

2003/1/1

2004/12/31

2006/12/31

2008/12/30

2010/12/30

2012/12/29

biomass(1) biomass(2) biomass(3)

0.0E+00

5.0E-03

1.0E-02

2001/1/1

2003/1/1

2004/12/31

2006/12/31

2008/12/30

2010/12/30

2012/12/29

biomass total

Ex. 2. decadal forcingEx. 2. decadal forcing

0.0E+002.0E-024.0E-026.0E-028.0E-021.0E-011.2E-011.4E-011.6E-01

2001/1/1

2003/1/1

2004/12/31

2006/12/31

2008/12/30

2010/12/30

light

inte

nsity

(ly/

min

)

051015202530

wat

er te

mp.

(deg

C)

Lint0 TMP

1 degC decadal fluctuation is added to idealized seasonal forcing.

decadal forcingdecadal forcing

0

20

40

60

80

100

120

140

160

2001/1/1

2003/1/1

2004/12/31

2006/12/31

2008/12/30

2010/12/30

2012/12/29

w(1) w(2) w(3)

Cooler condition increases wet weight of adult fish because of rich food.

decadal forcingdecadal forcing

0.0E+00

5.0E-03

1.0E-02

1.5E-02

2001/1/1 2003/1/1 2004/12/31 2006/12/31 2008/12/30 2010/12/30

biomass total

0.0E+00

5.0E-01

1.0E+00

1.5E+00

2.0E+00

2.5E+00

3.0E+00

2001/1/1 2003/1/1 2004/12/31 2006/12/31 2008/12/30 2010/12/30

density(1) density(2) density(3)

However, cooler condition increases mortality of larvae and reduces biomass.

Ex. 3. Fisheries impactEx. 3. Fisheries impact

Fisheries pressure is doubled in 2005 only.

In normal year, the annual fisheries catch is 30% of adult biomass.

In 2005, the annual fisheries catch is increased to 60%.

Fisheries impact in 2005 (doubled)Fisheries impact in 2005 (doubled)

0.0E+00

5.0E-03

1.0E-02

2001/1/1 2003/1/1 2004/12/31 2006/12/31 2008/12/30 2010/12/30

biomass total

0.0E+00

5.0E-01

1.0E+00

1.5E+00

2.0E+00

2.5E+00

3.0E+00

2001/1/1 2003/1/1 2004/12/31 2006/12/31 2008/12/30 2010/12/30

density(1) density(2) density(3)

Memoirs (G. A. Riley, 1984)Memoirs (G. A. Riley, 1984)

H. Bigelow said "Anybody who thinks he can predict more than 10% of plankton variability is a damn fool, but good luck".

To be sure, I have never pretended that [models] are truly realistic. My only defense has been that they help us to think….

They frequently yield results that are not intuitively obvious, and they teach us caution about drawing conclusions that seem to violate mathematical logic.That is the way physics and astronomy have grown.Biological oceanography, messy though it may be, needs the same kind of disciplined thinking.