Co-sponsored by: COMPASS, Census of Marine Life, Stellwagen Bank National Marine Sanctuary, MA Ocean Partnership Fund, Department of Fisheries & Oceans Canada, Gund Institute for Ecological Economics, UVM

Meeting Date & Location: October 29, 2007 at the University of New Hampshire

Overview
Approximately 30 individuals from institutions around the Gulf of Maine met for a two-day meeting to discuss the creation of pilot projects to create a scientific basis for using an ecosystem-based management (EBM) approach. This meeting was an initial step in organizing science and management communities to develop an interdisciplinary modeling framework. The modeling framework is intended to:

1. Work at an appropriate and manageable scale to support an EBM approach.
2. Investigate connections between human activities thus fostering better management across sectors.
   
3. Enable exploration of trade-offs among ecosystem services under various management scenarios so that decision-makers can make informed choices.

Why Pilot Projects?
The overarching objective of a pilot project approach is to create working models to inform management processes. By applying a modeling framework on a pilot project basis first, efforts can be focused, new information learned and the framework can be adapted as needed to eventually inform policy development. Engagement with policymakers in the US and Canada will need to occur to understand the information  needs for EBM. As part of this pilot project effort, policy goals will be developed based on existing and new input from managers so that the modeling framework will actually inform management implementation.

Criteria for determining appropriate pilot areas include availability and transferability of data, infrastructure, engaged scientists and managers, and a multiplicity of human interactions to better evaluate trade-offs. Pilot areas do not need to be bound by political jurisdictions.

After much discussion one Canadian and one United States pilot site was chosen (exact boundaries TBD): southwest New Brunswick portion of the Bay of Fundy and Stellwagen Bank National Marine Sanctuary plus the area in-shore to Boston and Cape Ann.

Multi-Integrated Model for Ecosystem Services (MIMES) as a modeling framework
Roel Boumans of the Gund Institute for Ecological Economics at the University of Vermont, gave an overview of MIMES, a multi-scale, integrated suite of models that simulates the generation and value of ecosystem services within an ecosystem. (http://www.uvm.edu/ecosystems  and http://www.uvm.edu/giee/mimes )  In brief, the MIMES framework models processes within the human and natural systems incorporating generating functions of various types to simulate system dynamics.  Ecological interactions, for example, are modeled through as series of Lotka-Volterra relationships among species.  Biophysical models incorporate environmental features of the system and economic input-output models along with valuation methods are used to model components of human well-being.  MIMES is a simulation framework, initialized by available data or plausible assumptions concerning the parameters.  It is not a statistical data fitting procedure, though fitted models could be incorporated into the simulation. 

The development of MIMES sub-models for these pilot sites will allow ecosystem managers to quickly understand the dynamics of ecosystem services, how their services are linked to human welfare, and how their function and value might change under various management scenarios.

The project goals of MIMES are to

• Create a sophisticated and transferable system to elucidate dynamics of ecosystem services through a dynamic spatial model that provides a user interface
• Understand link between ecosystem services and human welfare by developing valuation protocols
• Simulate changes in ecosystem functions and values under various management scenarios

The model includes feedback loops among and between ecosystem services within various systems. Ecosystem services are classified according to spatial/temporal scales over which they’re produced and delivered.

Does the MIMES application fit the pilot project objectives?
The following were considerations for using MIMES for this pilot project initiative:

1. Does MIMES provide important information for EBM? MIMES provides an integrated framework in such a way that it produces a network of simulation outputs to support management decision-making. It is a dynamic modeling interface that provides for collaborative efforts. MIMES should be connected to existing models so that the best available information is used to structure and initiate the simulation the model.
2. MIMES and an integrated GIS mapping program should be brought together to make the output of the simulations accessible to policy-makers and interest groups.
3. The model should assume that there is feedback in the system, that is, that actions or effects within one part of the system impact other parts of the system and ultimately feedback on the generation of ecosystem services.
4. There should be an ecological forecasting function of the model and a way to quantify the uncertainty around each forecast.
5. Expert knowledge and expertise populate the model. Therefore, model should display the uncertainty in the results, based on the underlying studies and opinion.
6. A stakeholder interface should be built into the model so that different functions can be entered and different outcomes can be produced.

Other modeling approaches were also briefly considered such as Atlantis and EcoPath. MIMES is a framework that can draw upon these and other modeling efforts in structuring the simulations.  The group agreed that exploring how different approaches can inform policy making should be a feature of the pilot project effort. MIMES was chosen as a framework because it can be used as a way to simulate economic consequences for a full set of services under a variety of hypotheses and management scenarios. It is a way to project management consequences over the medium and long term.

Summary
    Although several organizations co-sponsored this meeting, none are funding entities. The EBM Regional Pilot Project Work Group must collectively find funding to develop the pilot projects. The group will simultaneously begin work in the following areas:

Four sub-groups have been formed to guide efforts.

1. Data integration and interoperability. What information is needed and available to populate models and how are data packaged into usable formats.
2. Goal setting. Identify an initial suite of ecological goods and services (services for short) that the model should strive to predict or forecast.
3. Technical modeling consortium. Think through technical details to understand how different modeling approaches may fit together when developing the MIMES model and otherwise and figure out what different modeling approaches do to inform management at different spatial and temporal scales; incorporate models that are looking at interactions across services.
4. Funding. Research various funding opportunities for collaborative projects. Some options include:

• GeoConnections.org – US/Canada cross-border regional online mapping systems - $75,000 to one project.  Proposals due 12/18.
• EBM tools – geospatial tools – EBM tool innovation fund
• Sea Grant
• NOAA – (Beth Turner 09)
• Gulf of Maine renewal proposal
• Marine animal populations renewal
• Sloan Foundation
• NSF (coupled human and natural systems)

Action Steps for Work Group

1. Determine set of primary ecosystem services for initial focus (4-5) and look at management plans to support and/or inform understanding of key services in particular pilot areas.
2. Once services are determined, ask what are the underlying human/natural system attributes/interactions that will affect those services – drivers that influence services (i.e., what needs to be modeled to understand whether or not those services are changing). Note: Need to incorporate climate change and adjacent land use patterns as key drivers of change
3. Take a comprehensive look at pilot sites to ensure there are both good contrasts and similarities between them.
4. Start developing the model.
5. Move into management interaction/evaluation once we have further development of modeling approach to let users know what the group/modeling framework can provide. A demonstration will help to get beyond the conceptual.

Specific Work Group Tasks & Timeline
COMPASS staff coordinates communication among the four working sub-groups

November & December:

• Meeting summary is distributed
• Research on funding sources
• Ecosystem services are recommended for pilot sites through a goal-setting process
• A COMPASS regional work group website is created for resources on all work group activities
  
• An online collaboration page for sub-group members to post examples of data sets we believe to be useful for the model will be created
• COMPASS staff works with sub-group leaders to assess progress and support coordination

January – April:

• Model Creation

April:

• Work Group meeting to assess work to date and plan for future steps

Appendix I: Goal Setting Memo (Jud Crawford - Lead)

Modeling the Production of Goods and Services from the Gulf of Maine Ecosystem
The charge of the goal-setting sub-group is to identify an initial suite of ecological goods and services (services for short) that the model should strive to predict or forecast.  In this context, these services are the goals of the model.  A successful model should be able to forecast the delivery of these services by the ecosystem, taking into account the interactions among services and other parts of the ecosystem, and the tradeoffs that will be necessary when extracting a particular service or set of services.  The initial suite of services is not meant to be comprehensive but just a good starting point for this fledgling modeling effort. In order for the model to be useful to managers, the services the model forecasts must relate directly to the roles and responsibilities of current or future managers.  There is a continuum from broad categorization of services (e.g., food) to highly specific services (e.g. a single species of bivalve).  We have sought to define services with enough specificity so that it is both clear how the services are linked to our GOM pilot areas and clear whether or not a given service falls within the domain of a particular manager.

This working group spent two hours developing a preliminary list of services on 30 October 2007, at UNH.  There was considerable discussion within the group about how to interpret our assignment.  In the service tabulation found at http://www.gulfofmaine.org/EBMWorkGroups/ you will see that so far we have not been entirely consistent in our definition of services.   Your feedback on this will be important before we proceed from here.

The following are a few of the points considered as the working group began its work to identify six services relevant to each of the pilot areas (Massachusetts Bay/Stellwagen Bank and Eastern Gulf of Maine/Bay of Fundy).   
In order to complete our charge as we understand it, we will need to do additional work – checking with managers, reviewing the output of various processes that have been undertaken previously, and vetting our list to make sure that it has salience with a broad community of managers. 

We ask that the modeling working group provide some feedback on the points below and our preliminary list of services before we proceed further with this exercise.  Does it seem that we are on track from your perspective?   

• Each service should be a product of the ecosystem that at least one category of manager would identify as a something that he or she is responsible for as a manager.  The relevant managers are listed with the services.
• The services should be among those that have been identified through stakeholder or other processes so that they are not simply a product of our small self-selected working group.  For example, services identified through the Stellwagen Bank Management Review Process would be strong candidates.
• Some middle ground is desirable in the lumping-splitting continuum of service definition.   If the way services are defined is overly general (lumped) then the connection to our particular pilot areas and ecosystem will be obscure.  For example, food (lumped) is a highly aggregated description of marine ecosystem services while fin fish, and ground-fish, are increasingly specific definitions within the broader category of food.  Where do we need to be on this continuum (we will ask managers this question too)?
• Services defined in terms of the biological components of the ecosystem may be more useful to the modelers than higher level constructs that describe human activities that rest upon them – so whales vs. whale watching – where whales are the service produced by the ecosystem, and whale watching depends up this service, for example.
• The services should be organized under several broad categories (e.g., the services sailing, kayaking, and whale watching can be subsumed under recreation).   
• The definitions from the Millennium Assessment should be applied when possible.
• The initial list of services is a critical first step and should be completed as soon as possible so that the modelers will have this as they mover forward.  The group was able to commit to being finished on or before 15 December 2007.
• A clear 1-2 page summary of the whole modeling endeavor is needed.  It needs to explain that the goal here is to develop a forecasting tool that will be an aid to managers as they embrace EBM.  It should be explained that the modelers do not aim to prescribe management actions or management decision thresholds etc.

Appendix II: Data Integration & Interoperability Memo (Michael Thompson – Lead)
The data sub-group met briefly to talk about the types of data that collectively exist and also possible data warehouses that can be tapped into. The extent of each pilot site was also discussed. For the Stellwagen region, we might include Massachusetts Bay and parts of Jeffrey’s Ledge maybe only as far north as the New Hampshire border to help limit coastal data sources from multiple states. The exact location/extent of the Canadian site was not determined. Basic base data layers were discussed such as: bathymetry, backscatter sediment, shellfish habitat suitability areas, land use, whale locations, and herring landing data. 


Appendix III: Sub Work Group Members

Goal Setting Work Group
1. Ben Cowie-Haskell: ben.haskell@noaa.gov
2. Craig MacDonald: craig.macdonald@noaa.gov
3. Susan Farady: sfarady@oceanconservancy.org
4. Gregor Reid: reidgk@mar.dfo-mpo.gc.ca
5. Briana Brown: bkbrown@bu.edu
6. Anita Hamilton: hamiltona@dfo-mpo.gc.ca
7. Beth Turner: elizabeth.turner@noaa.gov
8. Stephanie Moura: smoura.mopf@comcast.net
9. Karen McCleod: overholk@science.oregonstate.edu
10. Verna DeLauer: verna.delauer@unh.edu
 
Technical Modeling Work Group
1. Hassan Moustahfid: hassan.moustahfid@noaa.gov
2. Les Kaufman: lesk@bu.edu
3. Joe Roman: jroman@uvm.edu
4. Roel Boumans: roelof.boumans@uvm.edu
5. Pat Halpin: phalpin@duke.edu
6. Andy Rosenberg: andy.rosenberg@unh.edu
7. Jud Crawford: jcrawford@clf.org
8. Lew Incze: lincze@usm.maine.edu
9. Jason Link: jlink@whsun1.wh.whoi.edu
10: Mike Fogarty: mfogarty@whsun1.wh.whoi.edu

Data Integration Work Group
1. Nick Wolff: nwolff@usm.me.edu
2. Pat Halpin: phalpin@duke.edu
3. Mike Thompson: michael.a.thompson@noaa.gov
4. Stephanie Moura: smoura.mopf@comcast.net
5. Peter Lawton: lawton@mar.dfo-mpo.gc.ca
6. Sara Ellis: sara.ellis@earthlink.net

Funding Work Group (Individuals who will follow-up on potential funding opportunities)
1. Andy Rosenberg: andy.rosenberg@unh.edu
2. Beth Turner: elizabeth.turner@noaa.gov
3. Les Kaufman: Lesk@bu.edu
4. Lew Incze: lincze@usm.maine.edu
5. Pat Halpin: phalpin@duke.edu
6. Peter Lawton: lawton@mar.dfo-mpo.gc.ca
7. Nick Wolff: nwolff@usm.me.edu
8. Stephanie Moura: smoura.mopf@comcast.net