1 of 72

Making marine ecosystem-based�management work�Trust, Trophic models, and Tradeoffs

USM Gulf Coast Research Laboratory, 10 October 2019

Gavin Fay

Acknowledgements: Geret DePiper, Mark Dickey-Collas, �Mike Fogarty, Beth Fulton, Sarah Gaichas, Robert Gamble, �Catalina Gomez, Bec Gorton, Amanda Hart, Isaac Kaplan,�Scott Large, Jason Link, Sean Lucey, Erik Olsen, Jamie Tam, �Howard Townsend, Robert Wildermuth

gfay@umassd.edu

www.thefaylab.com

@gavin_fay

Decisions about human uses of marine ecosystems have largely focused on a single species, sector, or industry. Choices for management are generally not stress-tested, and in many cases stakeholders don’t believe that science can support decision-making. Because human-natural systems are linked, the choices made about one piece affect the others. Management approaches that take a systemic view of human-environmental linkages and speak to a diversity of objectives have become known as Ecosystem-Based Management. For these approaches to be effective, decision support tools are needed that capture “what ifs” associated with management choices, so managers, scientists, industry, and other stakeholders can collectively find solutions. In this talk I will discuss two available tools, with applications in the Northeast US: 1) rapid qualitative modeling to provide a lens for exploring ecosystem questions, and 2) trade off analysis to explore the feedbacks and consequences of options.

�We used qualitative models of the Georges Bank fishery system to integrate physical and ecological processes and human dimensions. We compared environmental scenarios, assessing relative changes in modeled management outcomes. Although these models sacrifice complexity, their simplicity means they can be applied in regions lacking substantial ecosystem and economic modeling resources. Importantly, because the models are quick to construct and evaluate, they can be co-created with stakeholders, building trust and understanding about how these tools work, and how they help address questions.

�Coupled ecological-economic modeling, where models for ecological and economic subsystems are linked through their inputs and outputs, can quantify ecosystem tradeoffs. We linked the harvest output from fishery management scenarios in an end-to-end ecosystem model (Atlantis) to a regional economic model for the Northeast United States to calculate changes in socio-economic indicators, including consequences of management action for regional sales, wages, and employment. Systematic indicators revealed different ecological and economic outcomes, with clear tradeoffs among catch and biomass of species groups. Our results demonstrated that consequences of management may be felt disproportionately through the region and across different fisheries sectors. These two case studies represent part of a spectrum of available modeling solutions. These tools are able to assess the effects of management across a broader suite of indicators that have relevance for policymakers across multiple objectives, providing a structured framework for ecosystem-based decision-making.

2 of 72

3 of 72

SMAST mission emphasizes �interaction with regional �industry, government & �non-governmental agencies

Research Areas:

biogeochemical cycling
coastal ecosystem dynamics and restoration
computational modeling
fisheries science and management
marine renewable energy
ocean observing/remote sensing
ocean physics

4 of 72

The Fay Lab

Quantitative decision support for managing our oceans

www.thefaylab.com � @thefaylab

(image: NOAA NEFSC)

5 of 72

Empirical and mechanistic tools for

Population assessment
Quantifying tradeoffs among management objectives
Evaluating robustness to uncertainty and performance of decisions
Risk analysis for ecosystem consequences of climate change

Statistics & Simulations

6 of 72

7 of 72

Ecosystem-Based Management:�Balancing human activities & environmental stewardship in a multiple use context

M. Dickey-Collas

8 of 72

9 of 72

Evaluating Tradeoffs

To get more of one thing, you have to give something up.

10 of 72

Coupled social-ecological systems

11 of 72

A spectrum of tools, a spectrum of uses

Stock/Single Species

Ecosystem

Aggregate Biomass

Single stock models

Gadids

Flatfish

Pelagics

Multiple stock assessments integrated

Stock assessments with add-ons: explicit M2 or habitat or climate considerations

Multi-species assessments

Functional group models

Whole system models

Integrated ecosystem assessments

Multi-species

Economic assessments, social impacts

Distinguish between models

for TACTICAL, STRATEGIC, and HEURISTIC use / advice.

12 of 72

Integrated Ecosystem Assessment as framework for doing Ecosystem-Based Management

NOAA Fisheries

13 of 72

NOAA Fisheries

14 of 72

Characterizing ecosystem responses to change

Qualitative�ecosystem�modeling

Coupled ecological-economic quantitative systems tradeoff analysis

15 of 72

Challenges of ecosystem (or any) models:�finding the sweet spot

(Collie et al. 2014. Fish & Fisheries)

Bias-variance tradeoff

16 of 72

Challenges of ecosystem (or any) models:�finding the sweet spot

(Collie et al. 2014. Fish & Fisheries)

Bias-variance tradeoff

17 of 72

Can simple be useful?

Simple models are easy to interpret AND explain.

Despite not being realistic, are they robust?

18 of 72

ICES Working Group on the� Northwest Atlantic Regional Sea (WGNARS)

19 of 72

Conceptual ecosystem models

Highlight links between species and key ecosystem drivers, components, and goals.

Understand how human well-being is affected by changing conditions.

NOAA NEFSC

What is most important for your system?

20 of 72

What can you do with a conceptual model?

Formalize dialogue & discussion about objectives, system linkages.
Include relationships across disciplines
Give everyone a common understanding of the system (better graphics)
Do basic what-if analyses
Address uncertainty

21 of 72

Gaichas et al. 2016 http://journal.frontiersin.org/article/10.3389/fmars.2016.00105/full

22 of 72

Modules built with discipline-specific groups

Human dimensions

Food web

Physical environment

23 of 72

Then merged into the working�conceptual model: Georges Bank

All links in the network are documented

24 of 72

Qualitative Modeling to address uncertainty

Only specify the directions and signs of the interactions between system components.
Seeks qualitative �predictions of the system’s �response to a perturbation.
Useful for limited or �inequities in data.
Can be built in real time �with stakeholders.

Dambacher et al. 2003

25 of 72

Build the rest! bit.ly/GCRLqnm

Seabird

Coastal Habitat

Zooplankton

Fish

Fishery

Phytoplankton

Manager

-

Qualitative

Modeling

Example

26 of 72

Georges Bank Conceptual Model

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

27 of 72

What-if analysis: different model structures

Wildermuth et al 2018. CJFAS.

28 of 72

Council/stakeholder process

Specifies MSE objectives,

Performance measures,

Range of strategies

Scientists

develop tools

Council Decision Support:

Tradeoffs between objectives
Potential management strategy performance considering

key interactions
risks
uncertainties

S.Gaichas

29 of 72

Why include stakeholders at all?

Help stakeholders identify objectives

are the ‘right questions’ are being addressed?

Facilitate interactions, increased awareness of what tools can(‘t) do.
Provide stakeholders outside standard management process chance to participate.
Break down siloing.
Clarify roles.

Scientists:

Represent objectives of the decision makers quantitatively.
Identify factors that could be used in management.

Stakeholders / decision makers

Identify management objectives
Identify candidate management strategies.
Make decisions.

Thebaud et al. 2017. ICES JMS.

30 of 72

Characterizing ecosystem responses to change

Qualitative�ecosystem�modeling

Coupled ecological-economic quantitative systems tradeoff analysis

31 of 72

Evaluating Management Strategies

Compare relative effectiveness of management strategies for achieving multiple management objectives, and to quantify tradeoffs.
Identify sensitivity of management performance to system drivers and key uncertainty
Stress-testing options via �simulations where:

Truth is known
No real negative �consequences �of poor options

32 of 72

Coupling economic and ecological models�Approaches in the Northeast US

Whole of System Modeling: Linking Atlantis ecosystem model to Input-Output model for Northeast US economy.

33 of 72

Coupling economic and ecological models�Approaches in the Northeast US

Run Management Scenarios in Atlantis.
Fishery yield from Atlantis fed to regional economic model for Northeast US.
Consequences of management action for regional economy and employment.

34 of 72

Atlantis Northeast US “sunlight to dinnerplate”

0

50

120+

300+

Sediment

Epibenthic

Pelagic

Link JS, Gamble RJ, Fulton EA. 2011. NEUS – Atlantis: Construction, Calibration, and Application of an Ecosystem Model with Ecological Interactions, Physiographic Conditions, and Fleet Behavior. NOAA Tech Memo NMFS NE-218 247 p. Available at http://www.nefsc.noaa.gov/nefsc/publications/.

45 Functional Groups

18 Fishing fleets

35 of 72

Economic model: �Northeast region input-output model (NERIOM)

Developed from IMPLAN Pro System
Translates seafood sector revenue to supporting industries’ sales, income, and employment.
Additional societal indicators than Atlantis.

Three simple scenarios

Maintain current level of fishing effort
Halve current fishing effort
Double current fishing effort

Steinback & Thunberg. 2006. NOAA Tech Memo NMFS NE 188

Fay et al. 2019 Front. Mar. Sci.

36 of 72

Systemic indicators reveal disproportionate ecological and economic outcomes

Large ecological responses.
Tradeoffs in catch and biomass of species groups.

Economic indicators responded similarly to �fishery yield.
Changes in Sales did not�match those in landed catch.
Little change to average income, but lower under both alternative effort scenarios.

Fay et al. 2019 Front. Mar. Sci.

37 of 72

Aggregate indicators can mask variable outcomes for individual sectors

Fishing sectors

Fay et al. 2019 Front. Mar. Sci.

System scale metrics

Sector

metrics

38 of 72

Developing & Testing Fishery Ecosystem Plans

Multiple models, management feedback�Council-driven process to define objectives and performance

39 of 72

Fishing with threshold-based limit on exploitation:�ecological benefits with little economic cost

Fay et al. In Prep.

40 of 72

Operationalizing EBM:�Strategies must address tradeoffs among species and balance management objectives.

Conceptual models formalize dialogue & discussion about objectives, system linkages.
Knowledge co-creation increases trust in process.
Coupled ecosystem models retain detail of both economic and ecological systems.
Need to further show value of these approaches to translate to effective uptake.

Ecosystem-based control rule better balanced tradeoff between ecological and economic goals.

Challenge is to do this without creating burden of additional complexity and management cost.

Input-output models offer considerable industry detail to determine linkages and to generate estimates of indirect and induced multiplier effects.

Spatial resolution of the NERIOM model clearly provides information at scales relevant to fisheries management decision-making in the Northeast US, although we largely focused on super-regional scales here.

Our approach lacks feedback mechanisms from the economic back to the ecological sub-model.

Input-output analysis is static. Effects should therefore be viewed as the immediate/short-term impacts of the change being analyzed.

simple scenarios provide an easy way to demonstrate economic effects at the regional level

Applied in a multi-model inference context, our worked example provides a framework for evaluating the tradeoffs between management simplicity, risk, yield, and biomass status at both multispecies fisheries and whole-of-system scales.

41 of 72

Thank you!

Email: gfay@umassd.edu

Twitter: @gavin_fay

Web: www.thefaylab.com

Code, course materials:

github.com/gavinfay

github.com/thefaylab

Thebaud et al. 2017. ICES JMS.

Papers referenced in this talk:

http://bit.ly/2B1e5GA

42 of 72

Additional slides

43 of 72

Dealing with Tradeoffs

Some tradeoffs may be easier than others.

Objective 1

Objective 2

Gain on one, lose on the other.

Don’t have to give up much to gain.

Strong tradeoff, not able to gain on one without giving up a lot on the other.

44 of 72

Holsman et al. 2017. Ecosys. Health & Sustainability. https://doi.org/10.1002/ehs2.1256

45 of 72

Dolan et al. 2016. ICES J Mar Sci.

46 of 72

Georges Bank submodels

NOAA NEFSC

47 of 72

Developing an ecosystem-based fishery management procedure

48 of 72

Mid-Atlantic Fishery Management Council�Stock assessment support: black sea bass

Council-funded project to evaluate spatial stock assessments

Gavin Fay

gfay@umassd.edu

49 of 72

New England Fishery Management Council�Herring Management Strategy Evaluation

Infographics and Decision Analysis �text for public engagement process

Amanda Hart ahart2@umassd.edu

Gavin Fay

gfay@umassd.edu

50 of 72

NOAA Climate Program Office�New England groundfish management under climate change

Gavin Fay

gfay@umassd.edu

51 of 72

End to end models allow for evaluation of tradeoffs across broader suite of objectives at system level.

Fulton EA, Smith ADM, Smith DC, Johnson P (2014) An Integrated Approach Is Needed for Ecosystem Based Fisheries Management: Insights from Ecosystem-Level Management Strategy Evaluation. PLoS ONE 9(1): e84242. doi:10.1371/journal.pone.0084242

http://127.0.0.1:8081/plosone/article?id=info:doi/10.1371/journal.pone.0084242

52 of 72

Atlantis Ecosystem Model Summit

December 7^th-11^th, 2015

Honolulu, Hawaii

53 of 72

Tradeoffs, winners, and losers : broad and consistent patterns globally ?

If ocean acidification directly impacts calcifying species, are there indirect effects in the food web, and are there winners as well as losers?

Marine Protected Areas: encourage species recovery, but at the cost of other objectives related to economics,non-target species, and biodiversity?
Shifts in fisheries targeting: How do strong increases (or decreases) in effort by particular fishing sectors affect the ecosystem, versus broader scale alterations in fishing by all fishing sectors.

54 of 72

Common scenarios applied to 8 ecosystem models worldwide

Two alternate parameterizations of fisheries dynamics for NE USA and SE Australia.

*Olsen et al. In Review. Frontiers Mar. Sci.

55 of 72

5th

25th

Median

Kernel density of biomass responses across all individual functional groups in all models

Example: Ocean Acidification

56 of 72

Response of Eight Ecosystem Models to 50 Year Projections of Ocean Acidification

Direct Impact

Indirect Impact

57 of 72

Worm et al. (2009)

Quantify tradeoffs�between biodiversity, �catch, and �employment.

Able to gain �biodiversity with�small loss in fishery�yield.

58 of 72

Response to 50 Year Projections of 2x current fishing mortality rate on small pelagic fish

Direct – and Indirect + Impact

Indirect Impact

“Losers” (lower quartile) in pelagic fish decline by only 6%.
Median response per guild is <1% in all cases, even for pelagic fish guild
Minor impacts on mammals, birds, and sharks (1-3% declines for “losers” in lower quartile)
Some instances of compensatory increases in non-harvested small pelagic fish, but rare (even 95% percentile has only 4% gain)
Some instances of + and – response for demersal fish, but rare.

59 of 72

Response to 50 Year Projections of 2x current fishing mortality rate on invertebrates

Direct Impact

Very minor indirect effects throughout most simulated food webs (<1%, even on invertebrates).
Strongest “losers” (direct effect) is 8% decline in epibenthos (lower quartile)
Note coarse aggregation of invertebrate groups in most models.

60 of 72

Response to 50 Year Projections of 0.5x current fishing mortality rate on demersal fish

Direct Impact +

Indirect Impacts

Slight increases in shark, no median increase in demersal fish guild
Winners (3^rd quartile): shark 11% increase & demersal fish 15% increase)
Winners (3^rd quartile): seabirds 4% increase, largely only in NE USA, SE Australia

61 of 72

Responses & Tradeoffs

Effects of ocean acidification negative.
MPAs led to tradeoffs (winners & losers)
Fishing pressure scenarios had smaller effects
Compensatory effects within guilds lead to weaker average effects.

Sometimes large changes for individual groups

62 of 72

Is this useful?

Synthetic approach at system level
Scenario analysis broad-scale, tradeoffs at ‘big-picture’ level.
Not possible to do statistical analysis of parameter space for each system
Comparative analysis: different systems help represent uncertainty in marine systems, with caveats.

63 of 72

Simulations: balancing objectives

Gaichas et al. 2012 MEPS

Yield maximizing biodiversity is ~95% of MMSY

64 of 72

Sensitivity of thresholds to climate impact

Reduction in groundfish growth rate.

Thresholds of response to fishing can be dependent on other system drivers.

65 of 72

Northeast US: Indicators sensitive to effects of acidification, magnitude depends on scenario

Effects of acidification transfer to species not directly vulnerable.
Can see large negative effects on fisheries yield.
Dependent on detail of modeling acidification effects. Future versions of our Atlantis model will model functional responses to pH and ocean chemistry.
Results from the Northeast US compare with findings elsewhere.

Fay et al. 2017. Ecol. Mod.

66 of 72

Minimal Realistic Models�a.k.a. Models of Intermediate�Complexity for Ecosystems (MICE)

Objective of these models is to consider the interactions among a small number of species and components.

Focus on ‘key’ system properties/components relevant to the research or management questions.
May include a suite of model types for each component.
Perhaps statistically estimate model parameters using data similar to single-species models.

67 of 72

Gulf of

Maine

Sarah Gaichas, NOAA NEFSC

68 of 72

Ecosystem-based management procedures:�Georges Bank finfish fishery

Goosefish

Silver Hake

Dogfish

WinterSkate

Herring

Mackerel

Winter Fl

Haddock

Yellowtail Fl

Cod

Predation

Competition

10 interacting species

Multispecies operating model

Limit on total removals for the ecosystem / aggregate groups.

Thresholds in Ecological Indicator responses to drivers
Aggregate modeling – multispecies MSY

Minimum stock size thresholds for individual species

Gaichas et al. 2016. ICES J. Mar. Sci.

Fay et al. 2015. ICES J. Mar. Sci.

Hart and Fay, in prep.

69 of 72

Ceilings on system catch based on indicator thresholds

Improved performance with respect to catch and biodiversity objectives.

Set ceilings on system catch based on the median values for the thresholds determined from simulations in the previous slide.

Then ran simulation model given different fishing patterns using these ceilings.

When predicted catch to be higher than ceiling, Fishing exploitation rates on the species groups are reduced to obtain required ceiling.

Gray ‘None’ box represents distribution when no ceiling on catch is imposed.

Demonstrate that ceilings on saystem catch can constrain values for indicators.

Using ceilings based on Pelagic:Demersal Ratio and pPred and MTL appears to result in improved performance

- higher biomass

- higher yield

- lower proportion of species that are overfished (according to single-species criteria).

Indicator based control rules that use ceilings on system catch do seem to be effective means of controlling system behavior, though noted that the modelled system is fairly simple in this case (multispecies production model).

Deomsntrates utility of approach.

70 of 72

71 of 72

Ecosystem Indicators as reference points

Indicator-based control rules improved performance over single-species with respect to system catch and biodiversity objectives.

Fay et al. 2015. ICES J Mar Sci 72: 1285-1296.

72 of 72

A spectrum of tools, a spectrum of uses

Economic assessments, social impacts

Stock/Single Species

Ecosystem

Aggregate Biomass

Single stock models

Gadids

Flatfish

Pelagics

Multiple stock assessments integrated

Stock assessments with add-ons: explicit M2 or habitat or climate considerations

Multi-species assessments

Functional group models

Whole system models

Integrated ecosystem assessments

Multi-species

Add to the EBFM toolbox

Conceptual models and Qualitative Network Models