1 of 39

Ecosystem modelling:�Representing Mass Balance using Ecopath

Sean Lucey

​

Northeast Fisheries Science Center

Models for Marine Ecosystem-based Management – October, 20, 2022

2 of 39

Whole Ecosystem Models

Phytoplankton, detritus

Zooplankton, filter-feeders

Clupeoids, demersals etc

Marine mammals, sharks etc

TROPHIC LEVEL

Biophysical

Extended Single Species

Minimum Realistic Models

Modified from Plagányi 2007

3 of 39

Mass balance models

  • Identify and quantify major energy flows in the ecosystem
  • Describe ecosystem resources and their interactions
  • Evaluate the ecosystem effects of fishing or environmental changes
  • Explore management policy option

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 3

Plagányi 2007

4 of 39

Mass Balance

  • Snapshot of the ecosystem state
  • Trophic model
  • Balanced but not necessarily steady-state

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 4

5 of 39

www.theguardian.com

www.gma.org

www.voanews.com

Consumption (Q)

Production (P)

Unassimilated Food (U)

Respiration (R)

Production (P)

Biomass Accumulation (BA)

Predator Mortality (M2)

Other Mortality (M0)

Emigration (E)

Fishery Yield (Y)

6 of 39

Mass Balance Master Equations

Production = fishery removals + predation + emigration + biomass accumulation + other mortality

​

​

Consumption = production + unassimilated food + respiration

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 6

 

 

7 of 39

Dynamic simulations (Next Lecture)

  • Uses mass-balance for initial state
  • Utilizes ‘Foraging Arena Theory’
  • Can includes both biomass and size-structure dynamics

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 7

8 of 39

Applications in the Northeast US Continental Shelf

  • Studies of energy budgets go back to the 1940s (Clarke et al. 1946)
  • Expanded in 1980s (Cohen et al. 1982, Sissenwine et al. 1984)
  • Full food web in late 1990s (Link 1999)

​

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 8

From Link et al. 2008

From Link 1999

9 of 39

Ecopath with Ecosim

  • Ecopath - Food web modeling framework developed by Polovina (1984) and refined by Christensen and Pauly (1992)
  • Ecosim – Dynamic simulation developed by Walters et al. (1997)
  • Ecospace – Spatial dynamic simulations developed by Walters et al. (1999)

​

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 9

10 of 39

Highlights

  • > 1000 publications
  • Multiple International Symposium

35th Anniversary in 2019

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 10

  • Top 10 scientific break through by NOAA
  • Used sparingly in management

11 of 39

Best use

  • Identify and quantify major energy flows in the ecosystem
  • Describe ecosystem resources and their interactions
  • Evaluate the ecosystem effects of fishing or environmental changes
  • Explore management policy option

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 11

Plaganyi 2007

12 of 39

Ecopath applications

13 of 39

Rpath – R implementation of EwE

  • Complement and expand the open source possibilities of Ecopath with Ecosim
  • Use a platform widely used by ecologists, R
  • Utilize the built-in statistical and graphical capabilities of R

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 13

14 of 39

Building a Mass Balance model

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 14

15 of 39

Before you begin: Define your boxes

  • Consider what question you are asking
  • Use functional ecological groupings with niche overlap rather than taxonomy to aggregate
  • Leaving out a group because of poor data is worse than guessing
  • Include all trophic levels
    • Careful with bacteria
  • Must have at least one detrital group

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 15

16 of 39

More on boxes

  • Be sure to include top predators
  • Consider age-specific stanzas that capture ontogenetic shifts in diet or exploitation patterns
    • Used in dynamic silulations

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 16

science-facts.top5.com

17 of 39

Model time frame

  • Models typically use average values for the year
  • Seasonal changes are not incorporated
  • Important to note large scale changes in ecosystem structure

​

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 17

18 of 39

What you need: Data requirements

  • Biomass (B)
    • t km-2
  • Production to Biomass (PB)
    • year-1
  • Consumption to Biomass (QB)
    • year-1
  • Ecotrophic Efficiency (EE)
    • proportion
  • Diet Composition (DC)
    • proportion
  • Catch (landings/discards)
    • t km-2 year-1

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 18

19 of 39

Biomass (B)

  • Obtained from standard assessment methodologies
  • Important to use local information (scale of model)

19

20 of 39

Production to Biomass Ratio (PB)

  • Production = mortality + ∆B
  • Estimate of total mortality (Z)
    • PB = Z = F + M
  • Can calculate F from estimates of biomass and catch (F = C/B)
  • Get M from assessments or other tools
    • FishBase
    • Pauly’s (1980) empirical natural mortality equation
  • Also calculate Z directly using Beverton-Holt (1957) equation

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 20

21 of 39

  • Pauly (1980)

​

​

​

  • Beverton-Holt (1957)

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 21

 

 

22 of 39

22

FishBase.org

23 of 39

More on PB

  • Good rule of thumb
    • reciprocal of PB ≈ mean life span
    • Unless growth is linear (primary producers) then the reciprocal ≈ mean age
  • Example
    • Whale – PB = 0.03 year-1, BP = 33 years
    • Herring – PB = 0.4 year-1, BP = 2.5 years

​

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 23

24 of 39

Consumption to Biomass (QB)

  • Primarily from literature values (FishBase)
  • Varies with PB therefore it can be calculated using a PQ ratio

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 24

25 of 39

Production to Consumption Ratio (PQ)

  • Typically varies between 0.05 and 0.30
    • Lower for baleens and higher for very small organisms
  • Smaller individuals generally more efficient (higher PQ)
  • If PQ entered then PB or QB is calculated

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 25

kidsaquariumsquotes.wordpress.com

26 of 39

Ecotrophic Efficiency (EE)

  • Defined by Ricker (1969) as the fraction of a prey species’ annual production that is consumed by predators
  • Basically proportion of production explained in the system
  • Other mortality is 1 – EE
  • Typically close to 1
    • Unexploited top predators close to 0
    • Phytoplankton in bloom closer to 0.5

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 26

27 of 39

Diet Composition (DC)

  • Use proportions by either weight or volume
  • May need to aggregate/ disaggregate data based on box structure
  • Often necessary to tweak DC to ensure mass-balance

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 27

28 of 39

Catches

  • Include any number of fleets/gears
  • Possible to include economic variables
  • Advisable to create a ‘discards’ detrital group
    • Discards assigned to it via discard fate

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 28

29 of 39

Other parameters

  • Biomass accumulation (BA) – t km-2 year-1
    • If data shows change in B over model domain
  • Migration (E) - t km-2 year-1
    • Can be used by Ecospace
  • Detritus fate
    • All living groups produce detritus – U and M0
    • Necessary to specify where it is going if more than one

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 29

30 of 39

Respiration (R) and Unassimilated food (U)

  • Rearrange consumption equation

​

​

  • Q and P are estimated first
  • U is input parameter than effects R
  • Default value for U is 0.2
    • Herbivores and Detritivores ~ 0.4 leads to more reasonable R/B ratios

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 30

 

tcmmg.org

31 of 39

Multi-stanza groups

  • More important for Ecosim
  • Additional parameters
    • Length of stanzas (months)
    • Z per stanza
    • Von Bertalanffy K
    • Relative weight at maturity (Wmat/Winf)

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 31

32 of 39

From Parameters to Mass Balance

  • Can’t possibly have estimates for all parameters
  • Set of linear equations
  • Number of unknowns less than the number of equations
  • Order things are solved
    • QB
    • PB
    • EE
    • B
  • Can not estimate DC

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 32

33 of 39

Solving unknowns

  • Rearrange the production master equation

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 33

 

 

 

34 of 39

Mass Balance

  • Solve series of equations
  • Typically several EE > 1 (Not physically possible)
  • Tweak input parameters

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 34

35 of 39

Keys to fixing unbalanced models

  • Explore the mortalities on problem species
    • Fishing
    • Predation
  • Identify big problems initially
  • Change most uncertain parameters first

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 35

36 of 39

Balancing procedure – PREBAL (Link 2010)

  • Biomass across trophic levels
  • Biomass ratios
  • Vital rates across trophic levels

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 36

  • Vital rate ratios
  • Total production and removals

37 of 39

Balancing

  • Two ways to reduce EE
    • Increase a groups production
      • Increase biomass
      • Increase PB
    • Decrease their mortality
      • Decrease predator B or QB
      • Decrease fishing pressure

​

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 37

38 of 39

Mass balance in Rpath

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 38

39 of 39

Questions?

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 39

https://github.com/NOAA-EDAB/Rpath

​