c/EPA
EPA/600/R-09/104
September 2009
U.S. EPA Office of Research and Development
Ecosystem Services Research Program (ESRP)
Decision Support Framework (DSF) Team
Research Implementation Plan
-------
Team Owner:
Herb Fredrickson, Associate Director of Ecology
USEPA/ORD/NRMRL, 26 W. Martin Luther King Dr., Cincinnati, OH, 45268; 513-569-7402;
fredrickson.herbert@epa.gov
Team Lead:
Ann Vega, Physical Scientist
USEPA/ORD/NRMRL/LRPCD, 26 W. Martin Luther King Dr., Cincinnati, OH, 45268; 513-
569-7635; vega.ann@epa.gov
Alternate Team Lead:
Tim Canfield, Ecologist
USEPA/ORD/NRMRL/GWERD, 919 Kerr Research Drive, Ada, OK 74820; 580-436-8535;
canfield.tim@epa.gov
Framework (conceptual model, land and resource use options)
Ann Vega, NRMRL
Pat Bradley, NHEERL
Dave Burden, NRMRL
Tim Canfield, NRMRL
Verle Hansen, NRMRL
Ken Reckhow, Duke University
Amanda Rehr, Carnegie Mellon University
Mitch Small, Carnegie Mellon University
Neptune and Company, Inc.
Stakeholder Involvement SubTeam:
Marilyn Tenbrink (Lead), NHEERL/MIT
Walter Berry, NHEERL
Pat Bradley, NHEERL
Walt Galloway, NHEERL
Norma Lewis, NRMRL
Sue Schock, NRMRL
Tools Database SubTeam:
Bart Faulkner (Lead), NRMRL
Bill Barrett, NRMRL
Dave Burden, NRMRL
Heidi Paul sen, OEI
Joe Retzer, OEI
Shaw Environmental
Information Technology SubTeam:
Bill Barrett (Lead), NRMRL
Dave Burden, NRMRL
Mark Judson, EnvMSI (partner)
Jacques Kapuscinski, ORMA
Rajbir Parmar, NERL
-------
Heidi Paul sen, OEI
Joe Retzer, OEI
Kurt Wolfe, NERL
ESRP Project Liaisons:
Walter Berry, NHEERL - Coastal Carolinas, Wetlands
Pat Bradley, NHEERL - Coral Reefs
Tim Canfield, NRMRL - Wetlands
Curtis Cooper, NRMRL - Future Midwestern Landscapes
Verle Hansen, NRMRL - Tampa Bay, Coastal Carolinas
Linda Harwell, NHEERL - Tampa Bay
Drew Pilant, NERL - Coastal Carolinas
Norma Lewis, NRMRL - Coastal Carolinas, Coral Reefs
Betsy Smith, NERL - Future Midwestern Landscapes
Additional Team Members:
Gordon Evans, NRMRL - Assistant Lab Director for Ecology
Joe Williams, NRMRL - Acting Assistant Lab Director for Water
Albert D. Venosa, NRMRL/LRPCD - Division Director
Randy Parker, NRMRL/LRPCD/RRB - Branch Chief
-------
1. INTRODUCTION 5
2. STRATEGIC OVERVIEW 8
2.1. DECISION-MAKERS 8
2.2. EPA AUTHORITY 9
2.3. VISION AND GOAL 10
3. PHASE 1 - LESSONS LEARNED BY DOING 10
3.1. RATIONALE FOR A FRAMEWORK (NOT A PLATFORM) 10
3.2. LESSONS LEARNED FROM OE/COASTAL CAROLINAS WORKSHOP 11
3.3. CONCEPTUAL MODEL 12
3.3.7. Analytic-Deliberation 14
3.3.2. Adaptive Management 15
3.4. LESSONS LEARNED FROM DSF/CORAL REEFS WORKSHOP 15
4. PHASE 2 - CURRENT APPROACH 18
4.1. EVOLUTION OF A DSF SCHEMATIC 18
4.2. APPLICATION OF DSF SCHEMATIC FOR ADDITIONAL DEVELOPMENT 20
4.3. IMPORTANCE OF CONTINUOUS INVOLVEMENT WITH STAKEHOLDERS AND DECISION-MAKERS 21
4.3.1. Types of Stakeholder/Decision-Maker Interaction 22
4.4. DSF ECOSYSTEM SERVICES TOOLS DATABASE 22
4.4.1. Description, Purpose, and Intended Audience 23
4.4.2. Current status 23
4.4.3. Partnerships for the DSFEcosystem Services Tools Database 24
4.4.4. Future Plans 25
4.5. SOCIAL NETWORK ANALYSIS/TOOLS 25
4.5.1. Social Network Analysis - description, tested use, potential future use 25
4.5.2. Social Networking Tools 27
4.6. DECISION ANALYSIS AND VALUE OF INFORMATION (VOI) 28
4.7. QUALITY ASSURANCE FOR DSF 28
5. PHASE 3 AND BEYOND - FUTURE PLANS 29
6. LIMITATIONS AND BOUNDS 29
7. MEASURES OF SUCCESS 30
APPENDIX 1 - RESPONSE TO COMMENTS FROM SAB 31
APPENDIX 2 - DEFINING THE DECISION PROBLEM AND THE DECISION LANDSCAPE (CONTEXT)
34
APPENDIX 3 - HYPOTHETICAL APPLICATION OF THE DSF TO ADDRESS NUTRIENT LOADS IN
THE FLORIDA KEYS - SIMPLE EXAMPLE 36
APPENDIX 4-SUSTAINABLE LAND AND RESOURCE USE PLANNING CRITERIA 50
APPENDIX 5 - GLOSSARY OF TERMS IN THE ECOSYSTEM SERVICES TOOLS DATABASE 52
REFERENCES 53
-------
1. Introduction
In 2000, an EPA Science Advisory Board (SAB) published a report entitled "Toward
Integrated Environmental Decision-Making" (U.S. Environmental Protection Agency 2000)
and pointed to EPA's outstanding need "to assess cumulative, aggregate risks; to consider a
broader range of options for managing or preventing risks; to make clear the role of societal
(public) values in deciding what to protect; to clarify the trade-offs (including costs and
benefits) associated with choosing some management scenarios and not others; and to
evaluate progress toward desired environmental outcomes". The SAB suggested a
Framework for Integrated Environmental Decision-Making (see Figure 1) that "adopts an
interdisciplinary approach that combines deep understanding of environmental science with
theory and empirical methods in behavioral and decision science".
PHASE I
PROBLEM FORMULATION
- Information
- Expert fud^mem
- V.dues '
- Loyal and
Institutional Milieu
PHASE II
ANALYSIS AND DECISION-MAKING
(Wh.\t *ire the best risk reduction opportunities?
How can we achieve our goals and objectives?)
REPORT
CARD
(Is the nature
of the problem
changing?)
REPORT
CARD
(Are we meeting
our objectives?)
PHASE III
IMPLEMENTATION and
PERFORMANCE EVALUATION
(Ha
Figure 1: Framework for Integrated Environmental Decision-making (U.S. Environmental
Protection Agency 2000)
In its 2000 report the SAB asked EPA to begin to make major changes to the way the Agency
framed and addressed environmental problems. However, they pointed out that this new
-------
decision-making framework would build upon several previous efforts. These include the
risk assessment/risk management model described by the National Research Council
(National Research Council 1983), the update to that report (National Research Council
1994), the ecological risk assessment framework (U.S. Environmental Protection Agency
1982), the report of the Presidential/Congressional Commission on Risk Assessment/Risk
Management (United States 1997) (see also Figure 2) and the risk characterization process
described by the NRC (National Research Council 1996), which focused on the interaction
between analytic and deliberative processes in decision-making.
Risk Management
1
Monitor! nc
Figure 2: The logical model for EPA's analysis and management of risk continues to evolve. Left
panel: Initial models were largely designed around single pollutant-single exposure pathways.
Risk assessments and much of the risk analyses were conducted by scientists. Although effective
for the purpose for which it was designed, this framework was less effective for assessment and
analysis of risk in more complex problems. Right panel: The 1997 Presidential/Congressional
panel proposed a systematic, comprehensive framework that can address various contaminants,
media, and sources of exposure, as well as public values, perceptions, and ethics, and that keeps
the focus on the risk management goal (United States 1997).
In 2003 the U.S. Environmental Protection Agency (EPA) and the National Science
Foundation (NSF) requested the National Research Council (NRC) to help set research
priorities for the social and behavioral sciences as these relate to several different kinds of
environmental problems. Their specific task was to identify a manageable number of
promising research questions, the answers to which were believed to contribute to improved
environmental decision making. In the report "Decision Making for the Environment: Social
and Behavioral Science Research Priorities" (National Research Council 2005), the authors
recommended 5 science priorities.
Federal agencies should support a program of research in the decision sciences addressed
to improving the analytical tools and deliberative processes necessary for good
environmental decision making.
Federal scientific and environmental agencies should support a concerted effort to build
scientific understanding needed for designing and evaluating institutions for governing
human activities that affect environmental resources.
Federal agencies should substantially expand support for research to understand the
influence of environmental considerations in business decisions.
-------
Federal agencies should support a concerted research effort to better understand and
inform environmentally significant decisions by individuals.
To strengthen the scientific infrastructure for evidence-based environmental policy, the
federal government should pursue a research strategy that emphasizes decision relevance.
The first priority of the 2005 NRC report is the DSF team's first priority and we quote its
recommendations in this paragraph. Good environmental decision making requires not only
good environmental science, but also improved understanding of human-environment
interactions and development and implementation of decision-making processes. These
processes must integrate scientific understanding with deliberative processes to ensure that
the science is judged to be decision relevant and credible by the range of parties interested in
or affected by the decisions. Three needs were identified to achieve this goal.
Developing criteria of decision quality. Research is needed to define decision quality for
practical environmental decisions. It would consider such questions as: Which
characteristics of decision processes are associated with judgments of decision quality or
acceptability by decision participants and observers? Do different groups of decision-
makers and stakeholders apply different criteria of decision quality? To what extent does
increased attention to ideals of good public decision processes yield more positive
assessments of actual decision quality? Are decisions of higher normative quality
associated with preferred social and environmental outcomes? How can research results
on such questions best be disseminated to their potential users?
Developing and testing formal tools for structuring decision processes. Research is
needed to refine and apply tools from the decision sciences for helping decision-makers
better approximate ideals of good decision processes. The research might address such
questions as: How can formal methods of value elicitation be applied effectively in real
world decision settings? How can judgments about the nature and likelihood of a range of
outcomes be made more routine and workable through the use of information
technologies? What systematic methods for arriving at collective preference can be
applied in realistic environmental decision settings that can complement those of social
benefit-cost analysis and that do not adopt problematic assumptions typical of that
approach? How can learning be built into decision procedures to allow for updating over
time? How can risk communication methods be used to make issues of preference and
uncertainty intelligible and useful to key decision-makers and affected parties? How can
decision-aiding approaches help individuals by structuring the values, uncertainties, and
broader implications of their choices?
Creating effective analytic-deliberative processes. Research is needed to strengthen the
scientific base for organizing processes, such as are now being used with increasing
frequency in government, in which a broad range of participants take important roles in
environmental decisions, including framing and interpreting scientific analyses. The
recommended research would address such questions as: What are good indicators for
key attributes of success for analytic-deliberative processes, such as decision quality,
legitimacy, and improved capacity for future decision making? How are these outcomes
affected by the ways the processes are organized, the ways they incorporate technical
information, and the environmental, social, organizational, and legal contexts of the
decision at hand? How can decision processes be organized to ensure that all sources of
relevant information, including the local knowledge claims of nonscientists, are gathered
7
-------
and appropriately considered? How can these processes be organized to reach closure,
given the challenges of diverse participants and perspectives? How can decision-analytic
techniques be used to improve these decision processes? How can technical analyses be
made transparent to decision participants who lack technical training?
With this assessment of the evolution of the guidance ORD has received and continues to
receive from external advisory panels, the ESRP Decision Support Framework (DSF) team is
developing an implementation strategy. The DSF team, in developing this implementation
plan, recognizes the current lack of Agency knowledge about environmental decision-making
processes and decision science, and the limited resources currently available to develop this
capability within the ESRP, ORD and the Agency. The plan results from the selection of key
goals and a realistic assessment of what is currently possible within the context of the ESRP.
This plan is designed to allow the DSF team to adapt to new knowledge, resources and
acceptance of an evolving environmental management mandate for the Agency.
2. Strategic Overview
The ESRP DSF team has determined, after over a year of preliminary information gathering,
that it needs to continue learning about analytic-deliberation processes (including
participatory decision making and decision analysis) and collecting and organizing existing
data and information. The DSF team has also determined, based on the results of two
workshops (Coastal Carolinas and Coral Reefs - more information in Sections 3.2 and 3.4),
literature reviews, and discussions with ESRP project leads that our decision support efforts
need to focus on the ecosystem services impacts of land and resource use decisions.
2.1. Decision-Makers
Figure 3 depicts one of the major technical challenges faced by the DSF team, that of
scale. Land and resource use decisions are typically made by individuals, towns,
counties, tribes, states and sometimes multiple states (regions) to increase economic
viability of an area. However, decision-makers frequently fail to consider or weigh the
long term effects on human health and the environment in local-regional decision-making
processes. Improved decision-making includes awareness of the cumulative (and
incremental) impacts of multiple local decisions (bottom-up) and the local consequences
and opportunities of regional/national environmental policy (top-down). Individuals and
groups who typically make land and resource use decisions do not all currently have the
capability to evaluate the impact that their decisions have on ecosystem services and
socio-cultural needs. The ESRP DSF will improve this capability.
-------
DRIVES
Federal Decisions, Policy, and Laws
e.g., Clean Water Act
Impacts
Federal Lands
Drives decision-
Resource requirement
Political and economic
environment
Public health
Science
^B
Regional/State/Tribal
Government
Decisions, Policy, and Laws
Influences
Stakeholders who
Influence decision-
Non profit groups,
Citizens, Congress,
Lobbying groups,
Industry, Scientists,
Academia,
Media
Impacts
State Lands
3
T3
05
a
V)
0)
O
Influences
DRIVES
Local Government Decisions
(e.g., Counties, Townships,
Individuals)
Many decisions/choices are
ultimately made locally but
have huge and cumulative
impacts on regional.
national, and global delivery
of ecosystem services
Figure 3: Decision Making Occurs at Multiple Levels
2.2. EPA Authority
EPA lacks strong and explicit regulatory authority to protect ecosystem services. The
majority of lands and almost all freshwater wetlands in the US are private property.
Given this, what mechanisms does EPA possess or can develop to influence
environmental decisions and their impact on ecosystem services? Salzman (2007)
provided a perspective on this issue which the DSF workgroup used to help define the
research goal in terms of what the DSF should be and how it should be used:
Penalties - Up to now a major mechanism has been for the EPA to prescribe
environmental regulations and penalize those who are proven to be noncompliant
to the regulations.
Property rights - Some land rights are viewed as property (e.g., mineral rights)
and are marketed. Environmental land rights might also be classified as property
and made marketable.
Payments - Payment for services rendered could result in payments to farmers for
removing reactive nitrogen in wetland riparian zones before the field runoff enters
streams.
-------
Persuasion - In many cases persuasion may be the most effective. If people are
aware of consequences to environmental services many will make environmental
management decisions that will minimize impacts. This is particularly true when
financial advantages revealed by life cycle analyses are made apparent.
All these mechanisms are important and their application depends on the context of the
specific problem. A successful DSF will help clarify which mechanisms are most
appropriate for a specific problem and enhance the effectiveness of any chosen
mechanism.
2.3. Vision and Goal
Vision: To create a flexible yet robust framework of knowledge and information that
decision-makers can use to help structure environmentally-impacting decision processes.
These processes will enable stakeholders to identify and understand relevant technical
information and balance the value of ecosystem services with economic and social values
in a transparent way. We will initially focus on providing decision-makers with an
understanding of the implications of land and resource use alternatives. This
understanding will enable decision-makers to make decisions consistent with present and
future desires of the community while maintaining and sustaining functional natural
systems and the services they provide.
Goal: By 2016, the Decision Support Framework (DSF) team will provide an analytic-
deliberative DSF for land and resource use decision-makers at the local, state, tribal and
regional scales.
The ESRP DSF is a structured decision analysis framework with associated tools that will
enable land and resource use decision-makers to make better informed decisions
incorporating ecosystem services and desired environmental quality. The DSF will
provide decision-makers an understanding of probable effects of their planned decisions
on social, economic and ecological systems - thus enabling our planet to sustain society
and nature.
Supported decisions include environmental stressors and land use choices of national
significance such as: pollutant discharges, the built environment, agriculture,
transportation, and energy use on local, state/tribal, and regional scales.
3. Phase 1 - Lessons Learned by Doing
The following subsections provide a chronology of events occurring in Year 1 of DSF team
activities, including the changes made to the direction of the DSF team as a result of lessons
learned.
3.1. Rationale for a Framework (not a Platform)
In spring/summer 2008, the ESRP Multi-Year Plan (MYP) was reviewed by the Science
Advisory Board (SAB). The draft SAB report then underwent a quality review in fall of
10
-------
2008. One of the SAB quality reviewers had grave concerns with the plans for the
decision support platform (DSP) as described in the MYP. This reviewer took issue with
the development of an on-line system that would integrate the tools and models from the
other ESRP teams without first understanding what decision-makers and stakeholders
needed and wanted. The DSP team re-wrote the draft implementation plan to address
these concerns and the draft implementation plan was sent out for review in the winter of
2008.
In January 2009, after attending a joint Outreach and Education (OE)/Coastal Carolinas
workshop, the DSP team determined that an on-line platform that fully integrated all
products from the other ESRP teams was not necessarily what decision-makers needed or
wanted. The team was renamed the Decision Support Framework (DSF) team to reflect
the change in focus from an on-line platform to collecting information and understanding
what decision-makers and stakeholders needed/wanted.
Since fall of 2008, the DSF team has begun focusing more on what decision-makers and
stakeholders need to facilitate decisions related to land and resource use. Ecosystem
services are NOT routinely considered in such decisions and this reduces the chances of
their protection. Decisions are being made without:
identifying and including stakeholders impacted by the decision
understanding stakeholder and decision-maker values and preferences
fully examining the problem
creating goals and objectives for the desired outcomes
determining attributes to measure effectiveness of how well an alternative meets
an objective
investigating insights needed to create multiple alternatives that can be evaluated
with respect to the objectives
understanding potential impacts and the tradeoffs that need to be made, and
understanding the uncertainties of predictions.
Basically, decisions are being made outside of a structured decision analysis framework
and therefore decisions aren't including many of the important elements needed to make
an informed decision (Gregory and Keeney 2002; Reckhow 2003).
The DSF team has therefore selected behavioral decision research and decision analysis
to form the basis of the DSF described above. We have renamed the team and refocused
our efforts on the development of a decision support framework to emphasize this shift in
thought. In this implementation plan, the DSF team has also addressed SAB comments,
SAB quality reviewer comments, and comments received on the draft implementation
plan. Direct responses to the SAB comments can be found in Appendix 1.
3.2. Lessons Learned from OE/Coastal Carolinas Workshop
In January 2009, several members of the DSF team participated in an OE/Coastal
Carolinas workshop. The summary below was produced from a series of public
meetings in Charleston, Litchfield and Bluffton, South Carolina. The following excerpt
is quoted from the unpublished summary:
11
-------
"The most prominent concern, nearly universally expressed, is the broad spectrum of
adverse effects resulting from rapid, inadequately planned and regulated, ecologically
disharmonious anthropogenic development.
Based on a combination of the participants' emphasis and frequency of mention, these
specific issues emerged as the most prominent, significant or urgent:
Uncontrolled, poorly planned and ecologically inappropriate anthropogenic
(human) development degrades and threatens a broad spectrum of environmental
quality, and quality of life values.
Inadequate policy, regulatory and enforcement mechanisms fail to prevent or
constrain inappropriate development, or provide incentives for low impact
development. By law, if a permit application meets established guidelines the
permit must be issued. Additional impacts, such as loss of ecosystem services,
are not established criteria for consideration.
Degradation of water quality (e.g., fecal coliforms, eutrophication, lowered
dissolved oxygen, sedimentation/turbidity, pharmaceuticals, metals, invasive
species) impacts aquatic habitats and commercial and recreational fishing.
Aesthetics have deteriorated.
Terrestrial and aquatic wildlife habitat is no longer healthy due to loss,
fragmentation, and impairment.
The overall ecosystem functioning and health has degraded.
Residents have lost their intrinsic ' sense of place'."
The conceptual model (see Section 3.3) was developed in response to what was learned
during these public meetings.
3.3. Conceptual Model
The DSF team developed a conceptual model in March 2009 based on literature reviews
and preliminary information received from other ESRP teams. The conceptual model is
depicted in Figure 4.
12
-------
A Conceptual Model for Making Land and Resource Use Decisions
Define region, map
current uses,
capacity of
services, and
prerequisites for
maintaining
healthy ecosystem
services.
Identify Stressors
to ecosystem
services and
human health
Develop sustainable
land and resource
use options to
protect ecosystem
services, meet
human
needs/values, and
manage risk
Stakeholders
negotiate inputs,
evaluate, determine
path forward, begin
to implement
Carry out plan with
adaptive
management
Figure 4: Conceptual Model for Making Land and Resource Use Decisions
The draft conceptual model for the DSF has now evolved into a schematic of the DSF
(see Section 4.1), but the five modules and two overarching themes in Figure 4 are still
represented in the schematic:
Determine Services (define area of interest, services of interest, condition/status
and carrying capacity of services of interest, current land and resource use, etc.)
Identify Stressors (including type, magnitude, spatial and temporal effects of
stressor/driver)
Develop Options (create desirable, feasible, and realizable land and resource use
options to protect ecosystem services, meet human needs, and manage risk; use
scientific data, computer based models, values of stakeholders, etc.)
Evaluate Options (stakeholders use agreed-upon measurement rules to score
options; consider uncertainty, value of collecting additional information; evaluate
tradeoffs, risks, opportunities, consequences)
Take Action (determine next steps and implement them; revisit periodically and
adapt as needed).
In each process module, all of the most appropriate technical information must be
understood by stakeholders and decision-makers and environmental values weighed
against economic and social values in a participatory analytic-deliberative process. Once
13
-------
a decision is implemented, adaptive management requires periodic evaluation of the
results of the action and, as warranted, a determination of whether or not changes in data,
models, or ecosystem conditions demonstrate the need for revisiting any steps in the
decision-making process.
3.3.1. Analytic-Deliberation
The ESRP Decision Support Framework will be used to encourage decision-makers
to incorporate scientific information about ecosystem services into their planned
land and resource use decisions. A component in development of the DSF will be
the use of analytic-deliberation. Analytic-deliberation can best be described as a
structured discussion among scientists, decision-makers, and parties with an interest
in a policy or decision. The goals of the discussion are to define the problem, to
identify the values and outcomes of concern, to distinguish disagreements that must
be addressed through compromise and tradeoff from those that might be resolved
with better information, and to agree on appropriate ways to collect and interpret the
needed information. Analytic-deliberation emphasizes people's ability to process
language and develop mutual understanding (Dietz 1994; Renn 1999; 2001; 2006).
While analysis uses rigorous, scientific methods to obtain factual information, the
focus of deliberation is on discussion, reflection, and striving to understand other
points of view (National Research Council 1996). Figure 5 illustrates use of
analytic-deliberation in a risk decision process.
Learning and Feedb
3QOOC
Ht ;i!Kt 1-2. A schemata ivfin-»-n1.«iiin ut rhv rink iliMsmfi p
Figure 5: Analytic-Deliberation as presented in a risk decision process. (National Research
Council 1996).
14
-------
3.3.2. Adaptive Management
Adaptive management is a strategy for managing environmental systems with
highly uncertain responses to alternative actions by monitoring, interpreting, and
using these responses to adjust policies in an iterative manner, providing ongoing
improvements in knowledge and resource productivity (Holling 1978; Lee 1999;
National Research Council 2003; Walters 1986; 1997; Walters and Holling 1990).
As noted by Lee (1999), "adaptive management is learning while doing; it does not
postpone action until 'enough' is known but acknowledges that time and resources
are too short to defer some action." As such, adaptive management provides a
structured approach for making decisions in the face of uncertainty and seeking to
improve these decisions by actively acquiring the knowledge necessary to reduce
the uncertainty. Adaptive management is also enhanced by formal analysis and
optimization methods, e.g., (Williams 2001) and by broad stakeholder participation
(Schindler and Cheek 1999), as illustrated by the DSF conceptual model (see Figure
4).
Adaptive management has been applied primarily to wildlife and ecosystem
management (see, for example, its incorporation into the U.S. Forest Service's Land
and Resources Management Plans (U.S. Forest Service 2009), decisions by the
Alaska Department of Fish and Game (Alaska Department of Fish & Game 2001)
and the EPA/Environment Canada-sponsored Lake Superior Lakewide Management
Plan (U.S. Environmental Protection Agency 2006)). An NRC review of the
adaptive management program for ecosystem resources in the Grand Canyon
(National Research Council 1999) suggests the application of (Ojeda-Martinez et
al.) a long-term monitoring program and (2) a strategy for scientific evaluation of
policy alternatives in terms of ecological and stakeholder valuation outcomes. The
authors note that an effective adaptive management program will require tradeoffs
among objectives preferred by different stakeholders and methods for fairly
weighting these objectives. Similarly, an NRC report (National Research Council
2002) supports using adaptive management to advance scientific inquiry and policy
formulation for the Missouri River ecosystem. The authors suggest an approach that
includes (Ojeda-Martinez et al.) programs to maintain and restore ecosystem
resilience; (2) recognizing and adapting to uncertainty; (3) interdisciplinary
collaboration; (4) models to support collaboration and decisions; (5) meaningful
representation of a wide array of interest groups; and (6) ecosystem monitoring to
evaluate the impacts of management actions. Adaptive management is also applied
outside of the US (e.g., (McClanahan et al. 2006; Mostert et al. 2007; Olsson et al.
2004; Wells 2006).
3.4. Lessons Learned from DSF/Coral Reefs Workshop
After development of the initial conceptual model, the DSF and Coral Reefs teams
organized a joint workshop in June 2009, Key West, FL at the Florida Keys National
Marine Sanctuary. The workshop goals were to:
15
-------
Explore a collaborative vision among decision-makers, scientists and other coral
reef stakeholders for sustainable coral reefs
Initiate a systematic, deliberative process to analyze coastal and watershed
decisions that impact coral reefs
Advance an integrative framework to incorporate the ecological, social,
economic and legal consequences of alternative decisions.
A major focus of the workshop was the Drivers, Pressures, States, Impacts, Response
(DPSIR) Framework. The DPSIR Framework characterizes causal relationships among
categories labeled as Driving forces, Pressures, State, Impact and Response (Pierce
1998; Smeets and Weterings 1999). This model was adopted by the European
Environmental Agency and has been used by the United Nations to organize information
about the state of the environment in relation to human activities (UNEP 2007).
Examples of DPSIR in use include a decision support system for evaluation of wetland
ecosystem management (Turner et al. 2000) and evaluation of ecosystem-based
management alternatives for Marine Protected Areas (Ojeda-Martinez et al. 2009).
Figure 6 provides a visual of the DPSIR Framework. DPSIR definitions are:
Driving forces: Socio-economic sectors that describe basic needs of human
society such as food, water, fuel and shelter, and secondary needs such as
recreation, cultural heritage and sense of place
Pressures: Driver-related human activities that affect the environment
State: status of the environment and ecological resources, including attributes
that provide services; state is altered by changes in pressure
Impacts: changes in coral reef condition, persistence and delivery of services as a
consequence of changes in ecological state; changes can be valued
Response: societal reactions to changes in ecosystem services, values and
sustainability
16
-------
D
P
S
I
Driving Forces Pressure
Socioeconomic sectors and Human activities that
cultural factors that drive , - ' place stress on the
human activities (causes) - ' environment (pollutants)
Response State
Response of society to __ Condition of the
the environmental ^^=^ environment
situation (policies, (composition, distribution,
decisions) quality)
Impact
Effects of environmental
degradation (changes in
attributes, services)
Figure 6: Conceptual relationships among DPSIR sectors.
The workshop provided a significant amount of information for both the Coral Reefs
team and the DSF team. Based on a compilation of notes during discussions and small
group sessions, the following key points relative to the DSF team were made:
A holistic, integrated decision-making framework is needed for the entire reef
ecosystem (not just stony corals) and everything that impacts that ecosystem (e.g.,
land use). Impacts from a larger region, including 25 coastal states, Mississippi
River, and the Everglades, need to be incorporated into the big picture. Once
documented, research and resources need to be coordinated to focus on addressing
problems within the holistic plan. In other words, we must determine needs first,
and then collect the right data.
Public support for coral protection is mixed. This is due to mixed messages being
sent by scientific papers, the media, and a general lack of a comprehensive
understanding of the issues. It is important for the public to believe in science and
management solutions. The public wants scientific integrity. It is important to
include the public in all steps of the decision process so they have direct access to
information and can ask questions to enhance their understanding.
Confidence in data for management decisions needs to be considered. In some
cases, a high level of accuracy is needed, but in other cases, not as much. It is
necessary to balance resources and additional research with the level of accuracy
needed.
Both management-based science and science-based management are needed. In
other words, scientists need to understand what decisions managers need to make,
and do the research that informs those decisions. Likewise, managers need to
communicate with scientists about what research is needed to address a problem.
Scientists and managers must work together continuously.
17
-------
There is a perception that much of the research has been completed, and no
decisions have been made. Decisions need to be made often with incomplete
data. An approach is needed to address this perception of lack of action.
We need to focus on benefits and not just costs of environmental protection. The
DPSIR framework can help with this.
We must be proactive in addition to reactive. We must work to prevent problems
from happening in the first place.
The public needs to be better educated with the right information. Common
misperceptions need to be addressed. The complexity and interrelationship of
things need to be communicated and understood.
Decision science/analysis is needed.
The DSF team began to consider how to evolve the conceptual model into a decision
support framework (DSF) that would address many of the needs/issues identified in the
OE/Coastal Carolinas workshop and the DSF/Coral Reefs workshop. The first lesson
learned is that we are still in a very formative phase of the DSF research. New knowledge
gained still has relatively large impacts on the direction and implementation of the DSF
effort. On the other hand the DSF team realized that many of the needs/issues could be
extrapolated beyond these two specific projects so that the eventual DSF could be applied
more broadly. The approach we are taking is adaptive.
4. Phase 2 - Current Approach
4.1. Evolution of a DSF Schematic
A draft schematic of a DSF for supporting land and resource use decisions is provided
below.
18
-------
Goals ft
Objectives
Values &
Preferences
-
Decision
Makers
Scientific
Input
1
Stake-
holders
Alternative
Management
or Policy
Options
Problem
Formulation
Drrversare human needs
that fienerjte Pressures
on
FnvironmontAf State
Options can be
used lo manage
Pressures & States
--
MMtt
I X^ \w input
, Pressures
Selected
Management
or Policy Action
r.iViMal» Health Impacts
tire evaluated against Ihc
Objectives to generate a
Response
The chance in the
Environmental State
gpnpratps Impacts
Societal Sustainability
Economic Sustainability
Environmental Sustainability
Meet environmental regulations
Acceptable level of uncertainty
Figure 7: Draft DSF Schematic
This draft schematic combines a decision analysis framework (generally following the
decision analysis steps in (Reckhow 2003)) with the DPSIR framework (UNEP/GRID-
Arendal 2002). In the Figure, we start with the "Objectives" box. This is where the
decision problem and decision landscape (context) are defined. Because of the
importance of this piece, additional discussion is provided in Appendix 2.
To develop objectives, one must first define the problem. This includes identifying
decision-makers, stakeholders and groups/individuals affected by the decision. Problem
formulation results from an iterative and deliberative process involving stakeholders and
decision-makers. The stakeholders and decision-makers work to establish:
A clear understanding of the problem and its context(s) including spatial and
temporal scales
Goals and objectives for each decision-maker/stakeholder group
Values and preferences for each decision-maker/stakeholder group.
Once the problem is well formulated and bounded, decision consequences are discussed
leading to the identification of objectives and attributes (measureable endpoints). The
decision problem is then structured to enhance communication among decision-
maker/stakeholder groups.
19
-------
Management and/or policy options are developed in order to address the problem. The
DPSIR framework is used to organize the quantitative evaluation of the consequences of
options with respect to the objectives and attributes, ecosystem services in particular.
Models, maps, monitoring data and other tools are used to do the evaluation itself.
Finally - the decision-makers determine the action they wish to take (response).
A hypothetical example of how the framework could be used to inform a wastewater-
related decision can be found in Appendix 3. This is a very simplistic example intended
only to illustrate how decision analysis and the DPSIR framework can be used to address
a specific problem.
By 2016, we envision a comprehensive, systems level framework that will allow
decision-makers and stakeholders to evaluate planned land and resource use
management options to determine their impacts on ecological sustainability (using
bundled ecosystem services and production functions), social sustainability
(including human well-being, quality of life and sense of place), and economic
sustainability.
A list of sustainable land use planning criteria (compiled by an EPA community planner)
will be used, in addition to objectives, values, and preferences, to inform land and
resource use management options as part of the evaluation. A description of the
planning criteria is provided in Appendix 4.
4.2. Application of DSF Schematic for Additional Development
During 1970-2005, approximately 53% of the nation's population lived in coastal areas of
the U.S. (http://www.census.gov/Press-Release/www/emergencies/coast_areas.html).
NOAA's National Ocean Service study of growth in coastal counties (Crossett et al.
2004) and the Millennium Ecosystem Assessment Report both project increasing
development and stress on coastal ecosystems, which are among the most productive yet
highly threatened systems in the world (Dayton et al. 2005). Population densities in U.S.
coastal areas are triple those in non-coastal areas (mean 305 vs. 57, median 104 vs. 36,
respectively: http://www.census.gov/Press-Release/www/emergencies/coast_areas.html).
Economic interests and needs for housing, water, and transportation exacerbate stresses
on the coastal ecosystem from habitat loss, excess nutrients, invasive species,
contamination, and climate change. The ESRP matrix of stressor, habitat, and place is
structured to allow in-depth, targeted research to be extrapolated to additional stressors,
habitats, and places. Similarly, the development and application of our DSF requires
pilot projects that target decision-making venues and scales where stakes are high,
possible returns are large, and study questions can be clearly defined. Together, the
Coral Reefs and the Coastal Carolina projects meet these needs, and also allow us to take
advantage of pre-existing interagency andNGO collaborations.
The DSF team, in conjunction with the Coral Reefs and Coastal Carolinas teams
(including stakeholders and decision-makers), will use the schematic above (Figure 7) to
evolve a land and resource use decision support framework (DSF) to support land and
20
-------
resource use decisions. Working with Coral Reef and Coastal Carolinas' decision-
makers and stakeholders, the DSF team will develop and apply a framework that enables
identification of all the elements needed to evaluate multiple land and resource use
options for a defined area. The Coral Reefs and Coastal Carolinas teams (with the
assistance of the Nitrogen, Wetlands, Mapping, Monitoring, Modeling, and Human Weil-
Being teams) will bring together (and often develop) the data, information and models
necessary to allow decision-makers and stakeholders to evaluate those land and resource
use options. In other words, the DSF team will develop and apply the framework while
the data and models needed to use the framework will be developed by other ESRP
teams. Representatives from all of the above mentioned teams will participate in the
development, application, and use of the DSF.
The focus on Coral Reefs and Coastal Carolinas will enable us to support decision-
makers who are in immediate, critical need of support. Resource constraints (both human
and financial) only allow us to work with two teams in this phase. Additionally, the
Coral Reefs and Coastal Carolinas projects are in the early stages of research, making it
an ideal time for us to begin collaboration.
As noted above, members of the ESRP theme projects (mapping, monitoring, modeling,
outreach and education, valuation, and human-well being) are also embedded in both the
Coral Reefs and Coastal Carolinas projects. Wetlands and Nitrogen issues are a primary
focus in the Coastal Carolinas project.
4.3. Importance of Continuous Involvement with Stakeholders and Decision-makers
In keeping with the newly coined principle of "pervasive responsibility", every
member of the ESRP and every partner and client is part of the process of designing,
developing, and implementing the ESRP DSF. We plan to update the familiar phrase:
"If you build it, they will come" and operate under the motto "If you invite them to
build it with you, they're already there."
This means that DSF team members become members of the other ESRP place-based
projects, ecosystem based projects and pollutant based project. This enables ongoing
interactions with stakeholders and decision-makers to obtain their perspectives,
guidance, and partnership in the design, development and implementation of the ESRP
DSF.
If we do our work well, ESRP products will support national, regional and local efforts
to create a sustainable balance of growth and development with protection and
conservation so the needs of people and nature can be realized. Yet this will not be easy
for reasons beyond the scientific challenge. While it is important to develop the science
that allows us to better understand how human activities and behaviors deplete or
preserve earth's natural capital, knowledge alone does not translate into action. We must
tailor our products and share scientific results in a form that activates behavioral change.
To accomplish this we must understand and influence the types, diversity, and context of
relevant decisions being made, the perspectives of the stakeholders/decision-makers,
their conceptual understanding of ecosystem services, and their motivations in the
decision process that would enhance the acceptance and use of ecosystem services and
21
-------
human well-being in the decisions they make. Scientific understanding, public will, and
community action combined offer decision-makers the tools for protecting and
enhancing ecosystem services and thus help ensure their integrity and productivity in the
face of ever increasing human pressures.
Due to reasons listed above, the DSF team will be imbedded initially in two teams:
Coral Reefs and Coastal Carolinas (also includes Nitrogen and Wetlands components).
4.3.1. Types of Stakeholder/Decision-Maker Interaction
There are a myriad of approaches available for ensuring stakeholder/decision-maker
involvement in the development of the DSF. A tool within SMARTe (Sustainable
Management Approaches and Revitalization Tools - electronic) provides
information for 66 different community involvement tools:
http://www.smarte.org/smarte/tools/PublicParticipati on/index.xml?mode=ui&topic=
publicinvolvementaction. The EPA public involvement website:
http://www.epa.gov/publicinvolvement/ is another excellent resource for
information related to interactions with the general public (typically stakeholders in
land and resource use decisions).
The Coral Reefs team is using a variety of stakeholder/decision-maker approaches
to obtain information on their needs and preferences. The Coral Reefs team has
already formed 5 focus groups around the DPSIR framework to assist them in the
development of an integrated research program. Joint Coral Reef/DSF
stakeholder/decision-maker workshops are planned for Southeast Florida, Puerto
Rico, and the U.S. Virgin Islands. A workshop in the Florida Keys (at the Florida
Keys National Marine Sanctuary) took place in June 2009. At this workshop, the
Coral Reefs and DSF teams obtained a good understanding regarding the needs of
the Sanctuary and good feedback regarding the workshop structure - both of which
will be used to improve future workshops.
The workshops incorporate several information elicitation procedures including:
flow charting decision-making processes, a value of information exercise, a social
networking analysis exercise, small group breakouts to obtain input on the DPSIR
and decision landscape framework, and a Q&A discussion regarding existing tools
and models. Future workshops will incorporate lessons learned from the first.
Stakeholder interviews and follow-up interactions will include a more specific
elicitation of values and preferences to support Bayesian belief nets as part of the
DSF.
The Coastal Carolinas Team will use many of these same techniques to interact
with their stakeholders and decision-makers. The DSF team members will provide
expertise and support of these interactions.
4.4. DSF Ecosystem Services Tools Database
An ongoing theme for the DSF Team is that planners and decision-makers are challenged
to consider not only direct market costs, but also ecological externalities. There is an
22
-------
increasing emphasis on ecosystem services in the context of human well-being, and
therefore the valuation and accounting of ecosystem services is becoming an integral
component of economic efficiency. It is recognized that organizations and researchers
continue to expand currently available tools that will be of value to the decision-maker
for evaluating alternative approaches and outcomes of management actions, thus
providing the decision-maker with greater levels of confidence that ultimate management
actions will produce the desired outcomes. The Ecosystem Services Tools Database will
maintain a collection of ecosystem services-related tools, approaches, models, etc., for
categories of ecosystem services and decision type. The database is intended to bridge
user needs with existing or planned tools that can be of direct use to the decision-maker.
4.4.1. Description, Purpose, and Intended Audience
Depending on the type of decision to be made, associated ecosystem services may
be quantified by using a variety of approaches that could consider deterministic
physical and chemical processes, known empirical relationships, and/or
socioeconomic valuation methods. Existing lists and directories emphasize process
modeling to evaluate results of water resources decisions, changes in mass and
energy budgets, and other direct physical manipulations. These can be found on
several governmental and non-governmental websites. In the context of decisions
that affect ecosystem services in the more general sense, ecological externalities
may be quantified using process models, but other tools and techniques may
consider broader measures. The Ecosystem-Based Management Tools (EBMTools)
Network (NatureServe 2008) has developed a database of tools that consider
bundled ecosystem services emphasizing coastal and marine systems. The ESRP
DSF Ecosystem Services Tools Database presented herein augments the scope of
the EBMTools Network database by including non-coastal and marine systems and
by including ecosystem services in the broad sense of decision support related to the
USEPA's Ecosystem Services Research Program
(http://www.epa.gov/ord/esrp/index.htm).
The DSF Ecosystem Tools Database is currently a collection of 235 tools that are
designed to provide information across a wide range of disciplines to assist the
decision-maker in making decisions that have the potential to impact ecosystem
services. This database is designed to be flexible and expand or contract as new
tools become available and older tools become obsolete. The database is being
developed to provide the decision-maker with a suite of tools that may be useful in
providing information that will have direct bearing on the management questions
under consideration. The purpose is to provide an evolving searchable database of
tools, approaches, and techniques that can be applied in analytic-deliberative
decision support processes for improving decisions that may affect ecosystem
services. This database is intended for users of varying levels of expertise. It will
provide information about tools and will assist in the decision process. It is not
intended to take the place of the decision-maker and their associated expertise.
4.4.2. Current status
23
-------
As of June 2009, the Ecosystem Services Tools database contains approximately
235 records, and this number is increasing. Figure 8 shows the fractions of the total
list by tool category (pie chart on the left). The Decision Support System Category
is further broken out in the pie chart on the right. Additional information can be
found in Appendix 5.
Probabilistic Model. 2
Game, 2-
Landscape Model, 3
Valuation. 4 _
Government Program, 6
Conceptual Modeling, 6
Model Development
Tools, 8
Maps, 10
Database, 10
Mapping Tool, i;
Social ^
Networking, 14
Emprical Model, 15
Economic Model, 15
Decision Support
System, 52 *
Social Leveraging,:
Real-time Observation, 2 _,
Prloritization, 2
Optimization,
Database, :>
Process Based
Mod el, 42
Guidance, 25
Science
Reinforcement, 3
Adaptive
Management, 8
uation, 12
Figure 8. Listing of the categories of tools for the Ecosystem Services Tools Database.
As part of the DSF/Coral Reefs workshop in Key West, FL with clients and
partners, the DSF Ecosystem Services Tools database was demonstrated as part of
the Phase I development process. The demonstration used a subset of database tools
that were relevant to Coral Reefs. During the workshop, we solicited comments on
ways to improve the database to better suit the needs of Coral Reefs managers and
decision-makers. These comments are currently being incorporated into the
database. We intend to use the workshops as a key means to evolve the DSF
Ecosystem Services Tools database to tighten its usability. The Tools database will
serve as the link between DSF Team clients and partners and the tools that may
assist them in decision-making.
4.4.3. Partnerships for the DSF Ecosystem Services Tools Database
The EBMTools database mentioned in Section 4.4.1 contains over 400 tools related
to coastal and marine ecosystems. Only about 15 of these are common to the current
ESRP DSF Tools database. Therefore, we see an opportunity for collaboration with
the EBMTools group to bring the two tools databases together and thus expand the
respective databases without having to engage in duplicative efforts. The EBMTools
group is very welcoming of this collaboration, and we have begun discussions with
the group on how best to do this.
The DSF team has also developed a relationship with EPA's Office of
Environmental Information (EPA OEI). EPA OEI has established contracts with
Lockheed Martin and subcontractors to establish the Environmental Modeling and
Visualization Laboratory (EMVL). Among the strengths of this group is expertise in
24
-------
the design of databases and query interfaces. The DSF team has secured several
hundred work hours for FY09 for our Phase I database development efforts and
EMVL is making adjustments and "fine-tuning" the current MS Access database to
prepare for a migration to the enterprise level MySQL database management
system. It is expected that EMVL will incorporate comments received in external
peer reviews and in DSF/Coral Reef workshops with clients and partners. We plan
to extend our use of EMVL in FY10 and beyond.
We have used the services of the EPA Center for Subsurface Modeling and Support
(CSMoS) to assist in populating fields of the database as tools have been added. We
plan to continue to utilize CSMoS on a continuing basis as tools continue to be
developed or discovered by the DSF Team.
4.4.4. Future Plans
The Ecosystem Services Tools database is scheduled to be migrated into the
MySQL database management system in September, 2009. In FY 2010, the schema
will continue to be improved and refined. It will be modified to allow greater
expandability in the database structure. Also in FY2010, a web-based user interface
will be developed to allow users to build a query to find a list of tools that can help
meet their decision support needs, based on a series of questions. These will include
questions about the type of decision to be made, the category (Matties et al.) of
tool(s) needed, the temporal and special scales of interest, amount and type of data
available, the user's scientific background, and the type of ecosystem being
considered. The user interface will be vetted with participants in the decision-
maker/stakeholder workshops and will continue to evolve.
In the next year, we will also focus on identifying additional research and model
development needs by using the database to determine what already exists and what
still needs to be developed. This will likely be done in partnership with the ESRP
modeling, mapping, and monitoring teams using the problem statements developed
by other ESRP projects.
The database will ultimately be coupled with the decision support framework to
allow users to find tools that can be used for specific parts of the framework. This
will begin in FY2010 and will be ongoing.
4.5. Social Network Analysis/Tools
4.5.1. Social Network Analysis - description, tested use, potential future use
"Social network analysis is the mapping and measuring of
relationships and flows between people, groups,
organizations, computers, or other information- and
knowledge-processing entities." (Krebs 2008)
25
-------
Social Network Analysis (SNA) has been used in the business world since the
1930s. Its intent is to improve productivity and organizational structure. With
new software and analytical tools, it has gained wider use in studies of knowledge
transfer, communication, collaboration, and decision science. It is a tool that can
be used to support strategic collaboration, facilitate knowledge creation and
transfer, and increase our capability to manage ecosystems and resources.
SNA enables users to determine direction of information/knowledge flow; task
flow; and trust or energy flow. One can determine if a person is overly central or
loosely connected and under-utilized. Divisive subgroups can also be identified.
SNA can be used to increase societal capacity to manage ecosystems and
resources. It can:
Identify and support leadership functions and gaps
Increase participation by bringing in isolated teams or individuals
Detect information bottlenecks.
Identify opportunities for improving the flow of knowledge
Accelerate the flow of knowledge and information across functional and
organizational boundaries
Improve the effectiveness of formal communication channels
Target opportunities through which increased knowledge flow will have
the most impact
Raise awareness of existing informal networks
Identify types of information that are communicated or not.
To use SNA, one must identify study questions and bounds; map network nodes
and connections; analyze the network structure, content and flows; and apply new
understanding to utilize, strengthen, or intervene. Figure 9 shows an example of
an SNA developed for the DSF/Coral Reefs workshop in June 2009. Workshop
participants were given an exercise where they identified:
With whom they communicated frequently and infrequently
The frequency with which they communicated (scale of 0-8, once a year
to many times per day)
The types of information they received from each person
The value of information received (scale of 0-8 from no value to critical)
The types of information they give.
Figure 9 allowed workshop participants to look at clusters of communication
frequency between individuals (actual data included names, but only
organizations were presented for sensitivity purposes). A clustering algorithm
was run on the network. The thickness in arrows represents how often
communication occurred (the thicker the arrow, the more often the
communication). The colors of the nodes represent how the network clusters. In
other words, the colors show the groups of people that are most connected to each
other. For example, the blue cluster at the bottom of the figure gives us
information about a person from Nova Southeastern University. This individual
is connected to other Nova representatives, Broward County representatives, and
26
-------
a NOAA representative, but is disconnected from the main network. The diagram
also shows that the Broward County individuals are not in communication with
the larger network from the larger network's perspective. It would be interesting
to investigate this perceived lack of communication. One way to do this would be
to give the exercise to the Broward County individuals who were identified in the
bottom cluster. With the figure, we can look for patterns and answer the
following questions:
Does the communication network serve needs well?
Are any individuals or clusters of individuals poorly connected?
Is critical information held outside the information network?
Does the network support learning?
Reef Relief' IFKNMS] <\
^* m '
JKBy West Citizen |
Dive Operation
Nova^^|Broward Co DEP
Figure 9: Social Network Analysis for the Florida Keys National Marine Sanctuary
The DSF team will continue to build its understanding and capability in the area
of social network analysis.
4.5.2. Social Networking Tools
A variety of social networking tools (Facebook, MySpace, Linkedln, etc.) are
being investigated for their potential to bring ecosystem services stakeholders and
decision-makers together in a social network to discuss common issues, learn
from each other, and direct them to more robust, scientific websites. Different
27
-------
applications are being considered as well such as "send your friend a fish." The
application would contain fun facts about a particular fish, the ecosystem to which
it belongs, and information about the type of ecosystem service it provides.
Because knowledge and education often impact decisions people make, social
networking tools are being examined by the DSF and OE teams as a way to
educate the general public about ecosystems and the services they provide. This
is a very minor part of the DSF team's activities.
4.6. Decision Analysis and Value of Information (VOI)
A key part of our research efforts will be to explore and test the applicability of decision
analysis methods for actual environmental decision problems. In particular, we will
advance methods for identifying the most valuable new information, data collection, and
research needed to support management decisions involving complex scientific issues in
the context of a multi-stakeholder, deliberative process. Using the conceptual models and
background information developed in our decision landscape and social networking
efforts for the case study sites, we will formulate decision models that include both
scientific and stakeholder input. The decision models will consider available
management options, perceived relationships between management options and
environmental outcomes (and the uncertainty present in these relationships), and
valuations for alternative outcomes derived from economic studies and stakeholder
elicitation. We will then estimate the potential value-of-information (VOI) that could be
provided by further studies designed to reduce uncertainty and clarify the decision
options. Our study will consider both traditional VOI measures from the decision
analysis literature (based on a single decision maker) and a new proposed measure that
addresses the probability that the study result will allow stakeholders who initially
disagree on the preferred management option to reach agreement once the study is
completed and the results are in. For this assessment, stakeholders are elicited for their
beliefs regarding the accuracy and reliability of the proposed scientific studies. This will
provide useful feedback to the ORD and other agencies as to the types of studies that
should be pursued and the measures needed to ensure that stakeholders will support and
trust their outcomes. In this way the proposed decision analysis/VOI studies will support
environmental management plans that include ongoing monitoring, research, adaptive
scientific learning, and deliberative participation.
4.7. Quality Assurance for DSF
The DSF will be a web-based structure for organizing and delineating components within
a land and/or resource use decision. The components will use Bayesian analysis
approaches and must be able to be integrated to allow decision-makers to understand and
evaluate different land and resource use options. This will require statistical software to
be used and/or developed. For any software development effort, a Quality Assurance
Project Plan (QAPP) must be prepared. The DSF team will prepare a QAPP in
accordance with the National Risk Management Research Laboratory's (NRMRL's) QA
requirements and EPA information security requirements. The QAPP will be reviewed
and approved by QA staff and information security experts. Once approved, the QAPP
will be reviewed annually to determine if any modifications are needed.
28
-------
A QAPP for the tools database has already been reviewed and approved by a NRMRL
QA Manager. This QAPP will be reviewed annually to determine if it requires
modification.
Models, data, and analysis tools developed by other ESRP personnel are the
responsibility of the tool developer.
5. Phase 3 and Beyond - Future Plans
As indicated in Section 4.1 above:
By 2016, we envision a comprehensive, systems level framework that will allow
decision-makers and stakeholders to evaluate planned land and resource use management
options to determine their impacts on ecological sustainability (using bundled ecosystem
services and production functions), social sustainability (including human well-being,
quality of life and sense of place), and economic sustainability.
After Phase 2 is completed, we will work with other ESRP and non-ESRP projects to
continue to develop and refine the DSF so that by 2016, the framework will be usable
across the nation to evaluate a variety of land and resource use management options.
6. Limitations and Bounds
EPA needs in the area of decision science are large, and ORD's human resources in
decision science are currently limited. Current and projected ORD budgets through 2010
severely limit ORD's ability to contract outside experts in decision science. We will have
to make the best use of existing ORD, Regional and Program Office expertise in decision
science, use postdoctoral positions to acquire expertise, set decision science as a high
priority in ORD staffing plans, and find potential partners. In this context our approach is
to demonstrate the Agency's recognition of the need for decision science research and to
set yearly objectives toward this goal. These objectives are obtainable given the current
constraints, which include:
Lack of resources. We lack extramural funds to support workshops and other
face-to-face interactions for all ESRP projects (only a few can be supported). We
lack travel money for in-house personnel to travel to workshops and/or meet with
other ESRP stakeholders and decision-makers. We lack dedicated, full time in-
house personnel.
Bounding our efforts: The ESRP is a very broad program. We have attempted to
bound our efforts by focusing on land and resource use decision-makers;
however, this is still very broad and will remain a challenge as the program
evolves. This is discussed in more detail below.
Integrating science and human values/judgments: This is a common decision
science challenge that we too must meet.
Limitations imposed by EPA: Currently, EPA restricts computer software that
can be used. EPA also requires Information Collection Request approval to
29
-------
perform surveys, a tool that is typically used by social scientists to collect
information from decision-makers and stakeholders. Survey approval can take up
to 3 years to receive.
Lack of in-house expertise in social and decision sciences: ORD is attempting to
increase its capability in these areas, but it will take time.
With respect to determining the bounds and limitations on the DSF research itself, under
the current vision, it is necessary to answer the following questions:
What do decision-makers (both internal and external to ESRP) need? The DSF
needs to include information and tools that are desired by decision-makers from a
local to a regional level.
What has already been done well? The DSF will identify existing tools that meet
user needs and identify where gaps exist. We will inform other ESRP theme and
project leads of these gaps so research can be focused to meet these.
What are the technical constraints? The DSF needs to be designed within the
constructs of today's information technology, yet be flexible enough to mature
and grow as technology improves. The DSF team has initiated an IT sub-team to
help address this question.
What are the legal constraints? The DSF needs to provide information and tools to
help decision-makers make decisions. Appropriate disclaimers will need to be
developed and maintained. The DSF team will also need to work with a variety of
partners (both internal and external to EPA) to accomplish its goal and develop
partnership agreements. The DSF team will work with EPA's Office of General
Council and ORD's Office of Science Policy to address these issues.
7. Measures of Success
The fully functional DSF is planned to be released in 2016. Web-statistics and testimonials will
be collected for an additional period of time yet to be determined. The DSF team will attempt to
collect information related to how the DSF:
Informs the protection, restoration, and enhancement of ecosystem services
Enables decision-makers to evaluate management options inclusive of the value of
ecosystem services and human health and well-being
Encourages the consideration/incorporation of ecosystem services in decisions at the
national/policy level scale and at the local/regional/tribal scales.
ESRP will also attempt to measure additional positive outcomes, such as:
An increase in the number of decisions that include traditionally non-market costs and
benefits
Increased availability of ecosystem services
Increased resilience of natural systems
30
-------
Appendix 1 - Response to Comments from SAB
The SAB provided several comments with regard to the Decision Support efforts put forth in
the original MYP. These comments were in six primary areas: (Ojeda-Martinez et al.) Lack
of needed in-house expertise; (2) combining the DSF with OE; (3) adequately describing
how the DSF would work; (4) concerns about feasibility of developing the DSF; (5)
developing connections and utilizing outside partners; and (6) adequately defining potential
clients. The DSF team considered all of the comments made by the SAB and we have
addressed each area as described below.
EPA's Office of Research and Development (ORD) recognized early that we did not
have all the expertise in-house to accomplish all that was needed in the development
and implementation of Long Term Goal 1 and especially the DSP(F). The SAB
comments also pointed this out. The DSF team has been working diligently to bring
in outside expertise to fill in the gaps that exist and impede the development of the
DSF. The ESRP has brought on two expert hires form Carnegie Mellon University
and one from Duke University to assist. ORD and the DSF team have organized a
workshop and a series of webinars from outside experts in the field of decision
science/analysis to bring their perspectives to the table. Two divisions in the National
Risk Management Research Laboratory (NRMRL) are building up their capabilities
in the area of decision analysis. The DSF team continues to identify gaps still unfilled
in this process and search for experts to fill those gaps.
The SAB suggested that the DSF and OE groups be combined into one team. The
ESRP originally had these two groups combined but quickly found that the amount of
work to be done in each of these areas necessitated the need for two full teams. The
DSF team needs to work directly with decision-makers and stakeholders to develop
the DSF while the broader ESRP team needs to ensure that decision-makers and
stakeholders are included in ESRP efforts and to educate people about ecosystem
services in general. The ESRP recognizes that the DSF and OE teams still need to be
closely linked, and to that end the two teams have several overlapping team members.
The teams will continue the strong ties and collaboration to accomplish the ESRP
goals.
The SAB identified the need to provide greater detail on how the DSF would work.
As indicated above, the "Decision Support Platform" team is now the Decision
Support Framework team. The DSF team has refocused its efforts to concentrate not
on an on-line platform, but on collecting information and understanding what
decision-makers and stakeholders need or want. The intent of this revised
implementation plan is to detail the current approach for developing a DSF.
The SAB raised concerns regarding the feasibility of accomplishing Long Term Goal
(LTG) 1 (see the ESRP Multi-Year Plan at: http://epa.gov/ord/htm/multi-
yearplans.htm). This was based on the relatively short time for this goal to be
completed, the lack of available expertise, the lack of resources allocated to this
effort, and that this goal, being dependent on much of the other work being conducted
concurrently, should be re-classified as a long-term objective. The DSF team agrees
with this assessment and we have discussed these concerns with our upper
31
-------
management. This final decision is still being considered at that level. The DSF
team has suggested pushing back the final deliverable of this effort to coincide with
the intent expressed by the SAB. The DSF team is looking to partner with outside
groups in an effort to leverage our resources with these groups and further the
development of the DSF.
The DSF Team has been working on developing interactions and connections with
potential outside partners. These efforts have focused on academics, private sector
companies, other governmental agencies, professional societies, and international
professionals working in the area of ecosystem services and decision analysis. These
were all areas identified by the SAB where the DSF team could do a better job in
broadening the reach and expertise of the DSF effort. Examples include:
Through interactions with SETAC, DSF team members are putting forth sessions
at their annual meetings focused on ecosystem services, developing a Global
Science Advisory Committee to provide a platform for researchers across the
globe to share and exchange ideas and information regarding ecosystem services,
and working with the steering committee to set up special symposia to discuss
ecosystem service concepts both in Europe and the United States.
Through interactions with the German Helmholtz Centre for Environmental
Research (UFZ), DSF team members are learning how others apply integrated
multi-disciplinary research (IMDR) to solve problems of broad national
significance. Since 2006, the UFZ has been applying IMDR to develop a tool for
managing contaminated "mega-sites." The lessons learned are universal and can
be applied to ESRP's integrated multi-disciplinary efforts.
This is a dynamic and ever evolving process that will continue throughout the life of
the DSF effort.
The process to adequately identify potential clients is a constant challenge. The DSF
Team plans, over the next several years, to conduct workshops with the place-based
areas and the coral reefs, nitrogen and wetlands groups to identify and incorporate
these clients into the development process of the DSF. The DSF team participated in
a joint OE/Coastal Carolinas workshop in January 2009. A joint Coral Reef/DSF
workshop was held in mid June 2009. Out of these workshops the DSF team has a
better understanding regarding who the specific clients of the DSF efforts may be.
While we can certainly identify groups that will help develop and use the products
delivered by the ESRP and especially the DSF, it is more difficult to specifically
identify names and individuals that will use these products. This can only be
accomplished at this level by conducting these types of workshops with other ESRP
teams and engaging these individuals and groups face to face. The DSF team will
rely on the OE team to help us reach clients beyond the ESRP.
The DSF team is working diligently to not only address the letter of the SAB comments but
the spirit of these comments as well. This process will continue through the life of this
effort with the goal of providing a top notch DSF that will support the decision needs of our
potential clients.
32
-------
The DSF team also received review comments from other implementation plan reviewers.
The DSF team has addressed these comments in this revised implementation plan and has
prepared a point by point response to these comments back to the reviewers.
33
-------
Appendix 2 - Defining the Decision Problem and the Decision
Landscape (Context)
The process of environmental decision making requires use of good science, stakeholder
participation, and learning, and it must be transparent. Traditionally, the science has been the
focus of this process, but often, the process has failed because of the other less emphasized
parts of this decision making process -the lack of transparency, lack of stakeholder
participation, and the lack of learning or understanding (not just of the science but of the
various stakeholder perspectives). Therefore, "getting the science right" may be a necessary
component but, alone, it is not sufficient for environmental decision making. The process of
environmental decision support requires the application of appropriate environmental
databases, mapping tools, models, and economic valuation methods. This may be referred to
as "getting the science right."(National Research Council 1996). However, prior to this,
critical steps occur in the identification of assessment participants, the framing of issues for
evaluation, enumeration of decision alternatives, and the determination of appropriate metrics
for comparing projected outcomes. NRC (1996) characterizes this as "getting the right
science." It may also be viewed as providing a proper understanding of, and tools responsive
to, the decision landscape of an environmental issue. The proposed DSF research will seek to
provide a framework that enables understanding of the role of different participants in the
decision problem, and the need for an adaptive assessment that responds to their decision
support needs.
Characterization of a decision landscape is made recognizing that various groups and
individuals provide inputs to environmental decisions in different ways, and that these
inputs evolve over time along with the state and quality of the ecological system.
Examples of key participants in an environmental decision include:
Industrial producers that discharge contaminants that affect the resource of interest;
Direct users and beneficiaries of the ecosystem, such as commercial and recreational
fishers, hunters, farmers with fields within the study zone ("upstream" farmers would
be included in group 1), loggers, biofuel producers, hydropower interests, and park
visitors;
Information gatherers and providers regarding the state of the ecosystem and its
services, including scientists in government agencies such as the USGS, NOAA,
USD A, EPA, and private, nonprofit, and academic researchers;
Government agencies charged with resource stewardship and regulating activities
affecting the environmental system;
Public groups or organizations with advocacy positions regarding the environmental
resource and its various uses.
The general public and its representatives who may bear the costs (and receive the
broader benefits) of various economic and regulatory decisions regarding the
protection and management of the resource.
While environmental decision support has traditionally been viewed as being provided by
those in bullet 3, for those facing decisions in bullet 4, the broader potential for decision-
support efforts to support improved decision making across the range of problem participants
is increasingly recognized. It is also evident that effective prediction of the outcomes of
34
-------
alternative management decisions requires an informed consideration of the manner in which
the activities of each of the six groups of participants might co-evolve along with the state of
the natural ecosystem.
Analytic-deliberative approaches that facilitate transparency, stakeholder inclusiveness and
learning provide a better process for informed decision making. Choosing among the
decision tools, methodologies and approaches is difficult because it requires users to be
aware of methodological nuances that may only be known to the designer or scholarly user.
For example, use of cost-benefit tools or approaches are extremely useful to better
understand monetary tradeoffs, but if the user is not careful, he/she may be monetizing
decision criteria that should not be monetized or that don't make sense when monetized.
Other aspects of decision tools can be even more nuanced. Recognition (and documentation)
of the decision landscape can provide a basis for a better-informed, better-focused set of tools
in the proposed DSF. Are the parties to a decision and their roles clearly delineated by law or
regulation? Does an agency with decision authority seek a consensus among various
stakeholders? Are decisions by many disaggregate decision-makers (e.g., homeowner,
farmers, or consumers) critical to the environmental outcome - is effective risk
communication a part of the planned management strategies? Are decision metrics specified
by law or prior agreement? Are management options limited to a set of predefined
alternatives, or is there the flexibility to propose new approaches? Do the various
stakeholders trust and utilize common sources for data and scientific assessment, or are there
competing studies financed by two or more parties?
Each of these elements of the decision landscape has implications for choosing an
appropriate decision-support methodology and tool. Consider the case of a single decision-
making agency with the requirement to choose a "cost-beneficial" management option.
Needs include estimates of the economic costs of various options and the environmental
benefits projected to accrue from implementation of these options. To make the projected
costs and benefits commensurate and directly comparable, valuation of improved (or lost)
ecosystem services will be needed. However, the distribution of costs and benefits among
different segments of the population may not be important - it may be sufficient to estimate
only the aggregate (societal) net benefits. In contrast, consider an environmental problem
where multiple conflicting stakeholders must reach a consensus on an effective management
strategy. Disagreements could include which of numerous metrics for evaluation should be
considered and how they should be weighted, the suite of possible management alternatives
that should be considered as possible options, and maintaining conflicting scientific views of
the current and projected state of the ecosystem and its services. Decision support for this
type of problem is much more complex, requiring methods for multiattribute,
multistakeholder tradeoff analysis; generation of alternative management options and future
scenarios; and sensitivity and uncertainty analysis to show the importance of different views
of the science and the potential value of information for reducing uncertainty and resolving
conflicts. Many decision problems will lie between the two cases considered here - the DSF
will support the range of decision-support tools needed to address them. The DSF will also
provide a framework for decision-makers, interested parties, and assessors to explicitly
delineate the decision landscape applicable to their problem and to choose the suite of tools
best suited for their subsequent decision-support needs.
35
-------
Appendix 3 - Hypothetical Application of the DSF to Address Nutrient
Loads in the Florida Keys - Simple Example
In this hypothetical example, nutrient loading to near-shore waters in the Florida Keys needs
to be reduced. We will use the DSF schematic to demonstrate how management options can
be evaluated by decision-makers. Note that this example is for illustrative purposes only.
Goals &
objectives
values &
Preferences
Decision
Makers
Scientific
Input
Stake-
holders
Alternative
Management
or Policy
Options
Adaptive
Management
Decision Makers
Government Regulatory Agencies
QFederal(EPA)
QState (FDEP, FDCA, FDOH)
O Local (counties, districts, cities, etc.)
Opttomcanbe
used to manage
resS Slates
Driwersarp human aped*.
that gwiprulp Pressures
on the Ecological arid
Environmental State
Selected
Management
or Policy Action
Response
Societal Sustainability
Economic Sustainability
Environmental Sustainability
Meet environmental regulations
Acceptable level of uncertainty
Environmental. Economk,
Societal & Heakh Impacts
rlp.rVnV.t thf
A3-Figure 1: Decision Makers
The decision-makers for this problem are largely federal, state, and local government
regulatory agencies.
36
-------
Goals & Values &
Objectives Preferences
1 1
Decision
Makers
Scientific
Input
Stake-
holders
>
Adaptive
Management
Drlversare human tteein
that generate Pressures
on the Ecological and
Environmen
Selected
Management
or Policy Action
Option^) th.il
meets the
acceptable
certainty?
Pressures p,ftw:ite
sCresses that change the
Environments) State
Environmental. Economic,
Societal & Health Impacts
thr
to
Response
1 he change in the
Lnvironmcntal Stale
Irapacts
^ Societal Sustainability
^ Economic Sustainability
^ Environmental Sustainability
^ Meet environmental regulations
v' Acceptable level of uncertainty
A3-Figure 2: Stakeholders
Stakeholder groups include:
Environmental groups that represent both
o local interests and
o the values of citizens not located in the Florida Keys, some of whom may
never visit the Keys, but still value healthy coral reefs
Business groups that represent a wide variety of interests including fisheries,
snorkeling operations, hotels, etc.
Residents whose interests include recreation, income, taxes, and services
Tourists whose interests include healthy coral, clear water, affordability
37
-------
Goals ft
Objectives
Values &
Preferences
-
Decision
Makers
Scientific
Input
1
Stake-
holders
Alternative
Management
or Policy
Options
Problem
Options can be
Drivers are human needs
that generate Pressures
Adaptive.
Manage m^
Select!
Manage!
or Policy T
Respor
Problem Formulation
Q Increased nutrient levels lead to algal growth which may smother corals or cause deadly
coral disease
r_! Wastewater treatment facilities in the Florida Keys do not comply with either Advanced
Wastewater Treatment or Best Achievable Technology standards
Q The capacity of existing wastewater operations needs to be increased to account for existing
residential septic tanks
Q Coral reef ecosystems provide regulating, provisioning (seafood), cultural benefits (tourism),
and supporting habitat value at between $100k $600k per km2/yr
/ Societal Sustainability \
^ Economic Sustainability ^
^ Environmental Sustainability
^ Meet environmental regulations
/ Acceptable level of uncertainty
TiiViMsl» Health Impacts
tire evaluated aesinil the
Objectives to Rcneratc a
Response
Pressureseenerste
stresjei that change the
FnvLtrmmctiul State
The chance in the
Environmental State
gpnpratps Impact;
A3-Figure 3: Problem Formulation
Interaction between decision-makers and stakeholders help to define the various components
of the issue.
Increased nutrient levels lead to algal growth which may smother corals or cause
deadly coral disease
Wastewater treatment facilities in the Florida Keys do not comply with either
Advanced Wastewater Treatment or Best Achievable Technology standards
The capacity of existing wastewater operations needs to be increased to account for
existing residential septic tanks
Coral reef ecosystems provide regulating, provisioning (seafood), cultural benefits
(tourism), and supporting habitat value at between $100k-$600k per km2/yr.
38
-------
Goals &
Obje dives
Values &
Preferences
Decision Scientific
Makers Input
I
Stake-
holders
Alternative
Management
or Policy
Options
Problem
Formulation
Adaptive
Management
Objectives
Drivers are human needs
that generate Pressures
on thr Ff.olfjgicat Aftd
Fnv iron mon tAf State
Options generate
Selected
Management Y
or Policy Action
Response
^ Societal Sustainability
S Economic Sustainability
^ Environmental Sustainability
^ Meet environmental regulations
/ Acceptable level of uncertainty
Goals & Objectives
imote ecosystem (clear water, health'
momic health through control of point
Q Sustainable Tourism (Industry, Citizens)
Q Sustainable Fishery (Industry, Environmenta
n ("oral Reef Recovery (Environmental Groups)
Measureable Attributes
utrient loads
LJ Algal population
Q Coral Cover
The chance in the
Environmental State
gpnpratps Impact;
A3-Figure 4: Goals and Objectives
A collaborative process between decision-makers and stakeholders is used to define values
and preferences, help formulate the problem, and develop a set of objectives that can then be
used as the target for a set of plausible management options. Measureable attributes are also
defined.
Goal: To promote ecosystem (clear water, healthy fish, healthy coral) and economic health
through control of point source nutrient loads.
Objectives:
Sustainable Taxation Plan (Industry, Citizens)
Sustainable Tourism (Industry, Citizens)
Sustainable Fishery (Industry, Environmental Groups)
Coral Reef Recovery (Environmental Groups)
Measureable Attributes:
Nutrient loads
Algal population
Coral Cover
39
-------
Goals ft
Objectives
Values &
Preferences
-
Decision
Makers
Scientific
Input
1
Stake-
holders
Alternative o^om
Management j manage
Or Policy ^V Drivers
Options
L
Problen
Adaptive
Selected
Management
or Policy Action
Y
I Obj(
Alternative Management Options
instruction or upgrading community wastewater plants to AWT
ndards
2.Construction or upgrading community wastewater plants to AWT
standards + reduce number of tourists 10%
3.Construction or upgrading community wastewater plants to BAT
standards
4.Construction or upgrading community wastewater plants to BAT
standards + reduce number of tourists 10%
5.Re ' ' r' ' ' '
nput
Response
/ Societal Sustainability \
/ Economic Sustainability ^
S Environmental Sustainability
S Meet environmental regulations
/ Acceptable level of uncertainty
r.iViMal» Health Impacts
tire evaluated agsinil the
Objectives to ecneratc a
Response
Pressureseenerste
The chance in the
Environmental State
gpnpratps Impact;
A3-Figure 5: Alternative Management Options
Five management options were identified that could potentially meet the objectives defined
above. These management options include a combination of building or upgrading
wastewater treatment plants to Advanced Wastewater Treatment (AWT) or Best Available
Technology (Turner et al.), and reducing the volume of wastewater by reducing the number
of tourists visiting the Florida Keys.
1. Construction or upgrading community wastewater plants to AWT standards
2. Construction or upgrading community wastewater plants to AWT standards and
reducing the number of tourists by 10%
3. Construction or upgrading community wastewater plants to BAT standards
4. Construction or upgrading community wastewater plants to BAT standards and
reducing the number of tourists by 10%
5. Reducing the number of tourists by 20%
40
-------
Goals ft
Objectives
Values &
Preferences
-
Decision
Makers
Scientific
Input
1
Stake-
holders
Alternative
Management
or Policy
Options
Option V. No change
Option 1: Wintcuwter J, 'M II
Opiion 3: No change
Option -V. Whitewater ,\, 5% ±1
Option 5; Wastewater 4-10% 12
Selected
Management
or Policy Action
Response
Cost Lcosvstem icrvic
Option 1: 5 WM13 ? 10% 13
Option2-$18M±5 "MlWlS
v- Societal Sustains
^ Economic Sustain^ Opt.on* J15M W
^ Environmental Sustaifls»^"p
^ Meet environmental regulafions~^
^ Acceptable level of uncertainty
110% 17
-
/ Maent.fic
x*^ V mput
>*-
1 Pressures
Opsioni: Nutrient loads i 2S%±1
Option2: Nutrient loads J. J0%±7
optinn.4: Nutrirnt loads 4,1 'j% M
Opnonl: nutrient load^ Jr .''1% '8
Options:Nutrientioadi 4-5S±l
Impacts
Option 1: AiSJlflroivtli -i- 15W +7
Option 2: AigJlfirawlh 4.18K.19
option i: Algal growth i-')')(. i 'v
option 4: Algal growth j, is1* i»
Option 5: Aigalgrowth i-J%±0.1
A3-Figure 6: Evaluation through DPSIR
The 5 management options are now evaluated through the DPSIR process.
1. Construction or upgrading community wastewater plants to AWT standards
2. Construction or upgrading community wastewater plants to AWT standards and
reducing the number of tourists by 10%
3. Construction or upgrading community wastewater plants to BAT standards
4. Construction or upgrading community wastewater plants to BAT standards and
reducing the number of tourists by 10%
5. Reducing the number of tourists by 20%
The major Driver in this example is sewage disposal. The management option of "reducing
tourists" can reduce Drivers by reducing the volume of wastewater being generated. In this
example, WWTPs do not impact the overall volume of wastewater being discharged. The
two wastewater treatment technologies in combination with tourist reduction reduce the
Pressure of increased nutrient loads. The percentage decrease in nutrient loads is a function
of both the volume of wastewater and the ability of each of the treatment types to reduce the
nutrient concentrations in the effluent.
41
-------
The reduction in nutrient loads generates a change in State of the algal population. This
change in State is estimated using a model relating algal growth to nutrient concentrations.
The complexity of this model can range from a simple empirical model to a 3D process
model with tidal hydrodynamics. The model complexity needed is determined by the
uncertainty level required for the decision-makers to make a decision. The basic process is to
start with the simplest model and add complexity through a value of information analysis that
provides a measure of the value of reducing uncertainty by increasing complexity and the
resources required to support this increase in complexity (including data requirements).
The Impact generated by the change in State (change in the algal population) generated by
each decision option is then evaluated. Our measure of Impact is based on changes in
ecosystem services. The value of this change in ecosystem services is based on the Values
and Preferences elicited from Stakeholders in the initial Objectives hierarchy development.
The costs of implementing each of the management options are compared with the changes
in ecosystem value generated by each option. The decision-maker is then left to choose the
option that best meets the Objectives.
Note that the figure above includes a measure of uncertainty for each of the measureable
attributes.
42
-------
Decision Makers
Cost
$15M
Option 5 ;°*
Ecosystem Services
Option 4
Option 3
$i7M Option 2
Option 1
10%
8%
r 11%
[ 10%
A3-Figure 7: Evaluating Trade-offs
Option 5: Reducing the number of tourists by 20%
Option 4: Construction or upgrading community wastewater plants to BAT standards and
reducing the number of tourists by 10%
Option 3: Construction or upgrading community wastewater plants to BAT standards
Option 2: Construction or upgrading community wastewater plants to AWT standards and
reducing the number of tourists by 10%
Option 1: Construction or upgrading community wastewater plants to AWT standards
This figure depicts, on the left-hand side, the cost of each option including the calculated
uncertainty (denoted by the oval length). The right-hand side depicts the increase in
ecosystem services with associated uncertainty (again denoted by the oval length). The
figure illustrates the trade-off the options provide between increasing ecosystem services and
cost. For example, Option 3 is expected (based on the mean cost of $7M) to be the least
expensive while providing an expected 8% average increase in ecosystem services. The
uncertainty for Option 3 is low relative to the other options. Option 2 has a large cost
uncertainty and a relatively high expected increase in ecosystem services with an
accompanying high uncertainty.
43
-------
Coals & Values &
Objectives Preferences
1 1
Decision
Makers
Scientific
Input
Stake-
holders
Alternative oplioIB
Management manage
or Policy \. Drivers
Options
Drlversafe human tteein
that generate Pressures
on the Ecological and
Environment
Problem
Formulation
Options \-.An hf
jripd In m,in,i
Pressures & Slates
Adaptive
Management
Options generate
implementation
[hat havr
Impacts
Selected
Management
or Policy Action
Oplion|i| llv.l
p meets the
that change the
bnmenlji State
I he change in the
Lnvironmcntal State
generates Impacts
^ Societal Sustainability
^ Economic Sustainability
^ Environmental Sustainability
^ Meet environmental regulations
v' Acceptable level of uncertainty
A3-Figure 8: Selected Management Option
The decision-makers selected Management Option 1. This option included construction or
upgrading community wastewater plants to AWT standards. This option met the basic
objectives developed by the stakeholders and decision-makers. As can be seen in the
previous figure, Options 3 and 5 were lower cost than Option 1 but did not provide a level of
ecosystem services that the decision-makers felt met the objectives. Option 2 potentially
could provide higher ecosystem services but also could provide lower ecosystem services
because of the level of uncertainty. The decision-makers felt more comfortable with the level
of uncertainty associated with Option 1.
44
-------
Goals ft
Objectives
Values &
Preferences
Decision Scientific
Makers Input
I
Stake-
holders
Alternative
Management
or Policy
Options
Adaptive
Selected
Management
or Policy Action
Problem
Adaptive Manaeemen
The management actions taken should be
monitored and periodically reevaluated by the
decision makers as new data are collected,
new assessment techniques are developed,
and as new technologies become practical.
y optional that
-------
Role of Scientific Input in Decision Problem:
Characterizing Current and Future Resources
Environmental
Variables
Population Growth
Land Use
Economic Activity
industry
agriculture
recreation/tourism
Water use, diversion
WW discharge rates
- N, P, BOD, TSS, toxics
NPS loading rates
Impingement
- boating, diving, etc.
Freshwater flow rates
Ambient WQ
-N, P, Algal, DO, TSS, toxics
Coral Cover/Health
Fish Species Presence
and Abundance
Ecological
Current Knowledge
Census data
USGS land use/CIS data
BEA/Census economic
data
-water, energy, material
use (e.g., fertilizer)
Inventories
-cesspits, OSS's, package
plants, municipal plants
NPDES permit data
Compliance monitoring
NFS studies
USGS flow monitoring
Fed/State WQ data
Special studies
-e.g., FL International U
Coral reef monitoring
- FWRI, NOAA
Volunteer monitoring
- Ocean Conservancy
Research Needs
Scenario Development:
- Future population
- Future economic activity
- Land use/land cover
projection model
Scenario Development:
- water use
- wastewater loading rates
- NPS loading rates
- impingement projections
Statistical analysis
- further data as needed
Statistical analysis
- further data as needed
Marine Health Monitoring
- e.g., Scripps Inst.
A3-Figure 10: Characterizing Current and Future Resources
Acronym List (by column):
WW: Wastewater
N: Nitrogen
P: Phosphorus
BOD: Biochemical Oxygen Demand
TSS: Total Suspended Solids
NPS: Non-Point Source
WQ: Water Quality
DO: Dissolved Oxygen
USGS: United States Geological Survey
GIS: Geographic Information System
BE A: Bureau of Economic Analysis
OSS: On-site Septic Systems
NPDES: National Pollutant Discharge
Elimination System
FWRI: Fish and Wildlife Research Institute
NOAA: National Oceanic and Atmospheric
Administration
This and the following chart provide an illustrative example of the role of scientific information
in the Florida Keys wastewater decision problem. Both charts draw from scientific research
studies described in the Florida Keys National Marine Sanctuary (FKNMS) management plan
and from the greater scientific literature. The chart above describes the role of scientific input in
characterizing current and future resource conditions. It is organized according to the DPSIR
framework (first column).
The second column includes the variables of interest
The third column includes the state of current scientific knowledge on the variables
The last column includes research needed for decision support
46
-------
Drivers
Variables may include population growth, land use, and economic activity
Existing data may be obtained from the Census, USGS, and BEA/Census
Research needs may include scenario development for each of the three variables for use
in modeling
Pressures
Variables may include water use, wastewater discharge rates, and pollutant loading rates
Existing data may include inventories, NPDES permit data, and compliance monitoring
Research needs may include scenario development for each of the variables for use in
modeling
Environmental State
Variables may include freshwater flow rates and ambient water quality and parameters
Existing data may include flow monitoring and water quality data
Research needs may include statistical analysis
Ecological State
Variables may include coral cover, fish species presence and fish species abundance
Existing data may include coral reef monitoring data, and volunteer monitoring data
Research needs may include statistical analysis and long term marine health monitoring
47
-------
Role of Scientific Input in Decision Problem:
Projecting Future Impacts of Management Options
Impacts
Relationship
Relationship between
population, economic
output, land use, and
water use & loadings
Relationship between
loadings & ambient WQ
-N, P, Algal, DO, TSS, toxics
Relationships between
ambient WCt coral
health and fish
presence & abundance
Relationship between
coral, fish & ecosystem
services
(e.g., fisheries, recreation)
Relationship between
alternative
management options
and ecosystem services
Current Knowledge
Increased population and
output leads to increased
water use & loadings
Excessive nutrients (N and P)
lead to algal blooms &
depressed O2
Depressed O2 can lead to a
decrease in coral cover and
an imbalance in fish number
& diversity
Socioeconomic Monitoring
Program (NOAA, etc)
Recreation and Tourist
Uses, Values, Attitudes and
Perceptions study (NOAA)
No integrated model
currently available
Research Needs
Economic input-output models
Model relating population to
water use and wastewater gen.
Scenario evaluation
General ambient WQ model
(USEPA, FLDEP)
Scenario evaluation
Linked WQ-ecological model
Coral health/fisheries model
Scenario evaluation
Economic model to predict
value of services from corals and
fisheries (USEPA and FL DEP)
Scenario evaluation
- Stakeholder valuation
Development of integrated
model
A3-Figure 11: Projecting Future Impacts of Management Options
Acronyms not previously defined:
62: Oxygen
USEPA: U.S. Environmental Protection Agency
FL DEP: Florida Department of Environmental Protection
This chart describes the role of scientific input in projecting the impacts of management options.
This requires models to relate the different components of DPSIR, such as from Drivers to
Pressures, Pressures to State, State to Impact, and Impact to Response.
The first column includes the relationship of interest
The second column includes the state of current scientific knowledge on the relationship
The third column includes research needs for decision support
For example, if we take a narrow view of this problem and look closely at a limited piece:
The DRIVER of Population Growth generates the PRESSURE of increased Wastewater
(volume) and greater Nitrogen (N) loadings. A planning model is needed that predicts wastewater
volumes given a population level. The Wastewater (volume) and greater N loadings generates a
change in STATE of the ambient N and subsequently the ambient algal population. A
WQ/ecological model (could be one model or separate models for each component) is needed
that takes the Wastewater (volume) and N loadings and provides predictions of ambient N >
48
-------
algal population > O2 > cora/ cover. The change (e.g., decrease) in the STATE of coral cover
leads to an IMPACT on the snorkeling business (e.g., decrease). An econometric model is
needed that relates coral cover to snorkeling business.
For this example, Management Options can attempt to control population growth (DRIVER) or
the level of N treatment at the WWTP (PRESSURE). With costs of these management options
compared to the benefits from the snorkel business.
The goal of the DSF is to help decision-makers look broadly across all stakeholder values,
objectives, potential decision options, DRIVERS, PRESSURES,STATES, and IMPACTS to
RESPOND with actions that meet the Objectives.
49
-------
Appendix 4 - Sustainable Land and Resource Use Planning Criteria
The following essential attributes of natural systems are our current best estimate that will maintain
ecosystem integrity and keep natural systems functioning. Because functioning natural systems are
related to functioning social and economic systems, these criteria should be the basis for land-use
decisions and evaluation.
System
Component
Productivity
Biodiversity
Soils
Water
Air/Atmosphere
Energy
No
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
Precondition of Intact Natural System
Productive biomass of any land area is at near-natural levels.
Native plants predominate the ecosystem
Growing trees and plants bring nutrients from deep soils to form cellulose at the surface where
they decompose.
Native coastal mangroves, wetlands, sea grass beds, and coral reefs are intact.
Water chemistry of sea- water is sufficient to maintain photosynthesizing plankton.
Genetic diversity exists.
Native and non-native species are isolated from each other.
Fragments of truly native environments remain intact.
Natural disturbance regimes exist or are simulated when they can not exist.
Distribution of redundant species is maintained across multiple time and space scales.
Habitats exist in configurations, sizes, and quality that meet physiological and behavioral needs of
native populations and communities.
Habitats are refreshed/renewed with clean water.
Native spawningMrthing/hatching sites continue to exist in useful condition.
Connectivity between spawning/birthing/hatching sites and maturation areas and return is open and
accessible (including migration).
Individual species and communities are widely dispersed beyond the range of any disturbance
regime.
Connectivity between habitats is redundant and grain is appropriate for native species.
Unique environments remain intact.
Soil minerals are renewed.
Adequate moisture exists to make nutrients soluble.
Soil chemistry and ph sustains native soil bacteria, microorganisms, and plants.
Organic natural wastes are abundant.
Ground water recharges ( withdrawals.
Surface water recharge ( all combined water uses.
Wetlands exist to purify waters.
Avenues for groundwater recharge are clean.
Air and water must be clean enough for autotrophs to live.
Water quantity and speed of surface flows meet historic cycles, durations, and intensities.
Soil compaction/impermeability and soil cover do not increase runoff above near-natural levels.
Trees/plants break the force of falling rain and loosen soil to allow absorption and slow runoff.
Sufficient forests exist to generate Hydroxyl radicals to process pollutant levels in the atmosphere.
New deciduous forests and crops exist in higher latitudes and old forests exist to consume CO2.
Forests exist in sufficient contiguous sizes to translate and moderate energy influx.
A4 Table 1 - Criteria for Sustainable Natural Systems
50
-------
System
Component
Social
No
33
34
35
36
37
38
39
40
41
Precondition of Intact Social System
A history and progression of how people faced problems is evident and transparent.
Places that provoke spiritual feelings remain intact.
Plant and animal taxonomy is documented.
People are able to freely interact and share ideas, labor, and resources.
Individuals have a voice in matters that affect them.
Risks to human life/health are known.
Human life is isolated from stochastic events.
Institutions exist to serve collective society.
Health risks are monitored and potential risks are made public.
A4 Table 2 - Criteria for Sustainable Social Systems
System
Component
Economic
No
42
43
44
45
46
47
48
49
50
Precondition of Intact Economic System
Materials are efficiently used and reused as much as possible.
Waste is attenuated by environmental processes.
Resource use is linked with investment in resource renewal.
Qualitative community resources are improved.
Net economic effects > costs incurred to natural systems.
Net economic effects > costs incurred to social systems.
Consumption of natural resources is counted as a cost.
All costs are calculated before being incurred.
Financial resources are sufficient to maintain community infrastructures, institutions, and services.
A4 Table 3 - Criteria for Sustainable Economic Systems
51
-------
Appendix 5 - Glossary of Terms in the Ecosystem Services Tools
Database
Tool Type
Description
Decision Support System
Process Based Model
Guidance
Economic Model
Empirical Model
Social Networking
Mapping Tool
Database
Maps
Model Development Tool
Conceptual Modeling
Government Program
Valuation
Landscape Model
Game
Probabilistic Model
Workshop
Multi Media Model
Presentation
Search
See right side, Figure 8
Model that uses physical or chemical principles
Synthesis of information to aid decision making
Model that focuses on the interaction between the environment, the
humans, and our use of goods and services
Model that use and test hypotheses through observation and
experimentation
Tool that measures interactions among individuals or groups in
decision making
Tool or application that builds maps from external information, such
as remote sensing images
An organized compilation of data
Portal for distribution of existing maps
Development environment for constructing models
Tool for building concept maps and models
A government agency or program that produces outputs useful for
ecosystem services decision support
Tool or methodology for quantifying value for economic analysis and
decision support
Models which use landscape metrics for data reduction
Role-playing tool
Model which uses elements of probability theory
Product that came from a workshop
Model that considers fate and transport among different
environmental media (e.g., soil, surface water, air)
Presentation that serves as a decision support tool
Search engine
52
-------
References
Alaska Department of Fish & Game. (2001). "ADF&G Commissioner Endorses Wildlife Plan
for Nikolai, McGrath, Telida, Takotna Area " In: http://www.adfg.state.ak.us/new s/99-
02/3-2-0 l.yhy.
Crossett, K. M., Culliton, T. I, Wiley, P. C., and Goodspeed, T. R. (2004). "Population Trends
Along the Coastal United States: 1980-2008." NOAA's National Ocean Service, 54 pp.
Dayton, P., Curran, S., Kitchingman, A., Wilson, M., Catenazzi, A., Restrepo, J., Birkeland, C.,
Blaber, S., Saifullah, S., Branch, G., Boersma, D., Nixon, S., Dugan, P., Davidson, N.,
and Vorosmarty, C. (2005). "Chapter 19: Coastal Systems." In: Millennium Ecosystem
Assessment: Ecosystems and Human Well-Being: Current State and Trends. Findings
of the Condition and Trends Working Group., J. Baker, P. M. Casasola, A. Lugo, A. S.
Rodriguez, and L. D. Ling Tang, eds., Island Press, Washington, DC, 948 pp.
Dietz, T. (1994). "What should we do? Human ecology and collective decision making." Human
Ecology Review, 1, 301-309.
Gregory, R. S., andKeeney, R. L. (2002). "Making Smarter Environmental Management
Decisions" Journal of the American Water Resrouces Association, 38(6), 12.
Rolling, C. S., ed. (1978). Adaptive Environmental Assessment andManagement, McGraw-Hill,
New York.
Krebs, V. (2008). "Social Network Analysis, A Brief Introduction." www.orgnet.com.
Lee, K. N. (1999). "Appraising adaptive management." Conservation Ecology, 3(2), 3.
Matties, M., Giupponi, C., and Ostendorf, B. (2007). "Environmental decision support systems:
Current issues, methods and tools." Environmental Modelling & Software, 22(2), 123-
127.
McClanahan, T. R., Marnane, M. J., Cinner, J. E., and Kiene, W. (2006). "A comparison of
Marine Protected Areas and alternative approaches to coral reef management." Current
Biology, 16, 1408-1413.
Mostert, E., Pahl-Wostl, C., Y., R., Searle, B., D., T., and Tippett, J. (2007). "Social learning in
European river-basin management: Barriers and fostering mechanisms from 10 river
basins." Ecology and Society, 12(1), 19.
National Research Council. (1983). "Risk Assessment in the Federal Government: Managing the
Process." National Academy Press, Washington, DC.
National Research Council. (1994). "Science and Judgement in Risk Assessment." National
Academy Press, Washington, DC.
National Research Council. (1996). "Understanding Risk: Informing Decisions in a Democratic
Society." P. C. Stern and H. V. Fineberg, eds., National Research Council, Washington,
DC.
National Research Council. (1999). "Downstream: Adaptive Management of Glen Canyon Dam
and the Colorado River Ecosystem." E. Commission on Geosciences, and Resources, ed.,
National Academy Press, Washington, DC.
National Research Council. (2002). "The Missouri River Ecosystem: Exploring the prospects for
recovery." W. S. a. T. Board, ed., National Academy Press, Washington, DC, 175.
National Research Council. (2003). "Environmental Cleanup at Navy Facilities: Adaptive Site
Management. Report of Committee on Environmental Remediation at Naval Facilities."
National Academy Press, Washington, DC.
National Research Council. (2005). "Decision Making for the Environment, Social and
Behavioral Science Research Priorities." G. D. Brewer and P. C. Stern, eds., National
Academy Press, Washington DC.
53
-------
NatureServe. (2008). "Ecosystem-Based Management Tools Network." In: Factsheet,
www.ebmtools.org.
Ojeda-Martinez, C., Gimenez-Casalduero, F., Bayle-Sempere, J. T., and co-authors, a. (2009).
"A conceptual framework for the integral management of marine protected areas." Ocean
& Coastal Management 52, 89-101.
Olsson, P., Folke, C., and Hahn, T. (2004). "Social-ecological transformation for ecosystem
management: the development of adaptive co-management of a wetland landscape in
southern Sweden." Ecology and Society, 9(4), 2.
Pierce, M. (1998). "Computer-Based Model in Integrated Environmental Assessment." European
Environmental Agency (EEA).
Reckhow, K. H. (2003). "A Primer on Decision Analysis " In: International Water Management
Course, EAWAG (Swiss Federal Institute of Technology), Zurich, Switzerland.
Renn, O. (1999). "A Model for an Analytical-Deliberative Process in Risk Management."
Environmental Science & Technology, 33(18), 3049-3055.
Renn, O. (2001). "The need for integration: risk policies require the input from experts,
stakeholders and the public at large." Reliability Engineering & System Safety, 72(2),
131-135.
Renn, O. (2006). "Participatory processes for designing environmental policies." Land Use
Policy, 23(1), 34-43.
Salzman, J. (2007). "The Law and the Policy of Payments for Ecosystem Services." In: Accessed
at: http://www.nicholas.duke.edu/institute/presentations/esw salzman.pdf.
Schindler, B., and Cheek, K. A. (1999). "Integrating citizens in adaptive management: a
propositional analysis." Conservation Ecology, 3(1).
Smeets, E., and Weterings, R. (1999). " Environmental Indicators: Typology and Overviews."
European Environmental Agency (EEA), Copenhagen.
Turner, R. K., Brouwer, R., Georgiou, S., and Bateman, I. J. (2000). "Ecosystem functions and
services: an integrated framework and case study for environmental valuation." The
Centre for Social and Economic Research on the Global Environment.
U.S. Environmental Protection Agency. (1982). "Framework for Ecological Risk Assessment."
EPA/630/R-92/001, ed., Office of Research and Development.
U.S. Environmental Protection Agency. (2000). "Toward Integrated Environmental Decision-
Making." EPA-SAB-EC-00-011, ed., Office of Research and Development, Washington,
DC.
U.S. Environmental Protection Agency. (2006). "Lake Superior Lakewide Management Plan
(LaMP)2006."
U.S. Forest Service. (2009). "National Environmental Policy Act (NEPA) ",
.
UNEP. (2007). "Global Environment Outlook GEO4." Nairobi & Valletta.
UNEP/GRID-Arendal. (2002). "DPSIR framework for State of Environmental Reporting." In:
Available at:
http://maps.grida.no/go/graphic/dpsir_framework_for state of environment reporting,
UNEP/GRID-Arendal Maps and Graphics Library.
United States. (1997). "The Presidential/Congressional Commission on Risk Assessment and
Risk Management Framework for Environmental Health Risk Management. Final
Report." V. 1, ed., United States.
Walters, C. J. (1986). Adaptive Management of Renewable Resources, MacMillan, New York.
Walters, C. J. (1997). "Challenges in adaptive management of riparian and coastal ecosystems.
Conservation Ecology." Conservation Ecology, 1(2), 1.
54
-------
Walters, C. J., and Rolling, C. S. (1990). "Large-scale management experiments and learning by
doing." Ecology, 71, 2060-2068.
Wells, S. (2006). "Assessing the effectiveness of marine protected areas as a tool for improving
coral reef management." In: Coral Reef Conservation, I. M. Cote and J. D. Reynolds,
eds., Cambridge University Press, Cambridge, 314-341.
Williams, B. K. (2001). "Uncertainty, learning, and the optimal management of wildlife."
Environmental and Ecological Statistics, 8, 269-288.
55
------- |