EPA/600/R-16/136 | September 2016 | vwvw.epa.gov/research
Lessons learned in applying ecosystem
goods and services
to community decision making
Office of Research and Development
National Health and Environmental Effects Research Laboratory
-------
EPA/600/R-16/136
September 2016
Lessons learned in applying ecosystem
goods and services
to community decision making
COMMUNITY BASED DECISION SUPPORT
By
R.S. Fulford, R. Bruins, T.J. Canfield, J.B. Handy, J.M. Johnston, P.
Ringold, M. Russell, N. Seeteram, K. Winters, and S. Yee
Office of Research and Development
U.S. EPA
Matthew C. Harwell, Project Lead
Gulf Ecology Division
National Health and Environmental Effects Research Laboratory
-------
Notice
The U.S. Environmental Protection Agency through its Office of Research and Development (ORD)
funded and collaborated in the research described herein. This document has been subjected to the
Agency's peer and administrative review and has been approved for publication as an EPA document.
Any mention of trade names, products, or services does not imply an endorsement or recommendation
for use.
This is a contribution to the EPA ORD Sustainable and Healthy Communities Research Program.
The appropriate citation for this report is:
Fulford, R.S., R. Bruins, TJ. Canfield, J.B. Handy, J.M. Johnston, P. Ringold, M. Russell, N. Seeteram,
K. Winters, and S. Yee. 2016. Lessons Learned in Applying Ecosystem Goods and Services to
Community Decision Making, Community-Based Decision Support. U.S. Environmental Protection
Agency, Gulf Breeze, FL, EPA/600/R-16/136.
-------
Foreword
The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the Nation's
land, air, and water resources. Under a mandate of national environmental laws, the Agency strives to
formulate and implement actions leading to a compatible balance between human activities and the
ability of natural systems to support and nurture life. To meet this mandate, EPA's research program is
providing data and technical support for solving environmental problems today and building a science
knowledge base necessary to manage our ecological resources wisely, understand how pollutants affect
our health, and prevent or reduce environmental risks in the future.
The National Health and Environmental Effects Research Laboratory (NHEERL) within the Office of
Research and Development (ORD) is the Agency's center for investigation of technological and
management approaches for preventing and reducing effects of pollution that threaten human health and
the environment. The focus of the Laboratory's research program is on methods and their cost-
effectiveness for prevention and control of pollution to air, land, water, and subsurface resources;
protection of water quality in public water systems; remediation of contaminated sites, sediments and
ground water; prevention and control of indoor air pollution; and restoration of ecosystems. NHEERL
collaborates with both public and private sector partners to foster technologies that reduce the cost of
compliance and to anticipate emerging problems. NHEERL's research provides solutions to
environmental problems by: developing and promoting technologies that protect and improve the
environment; advancing scientific and engineering information to support regulatory and policy
decisions; and providing the technical support and information transfer to ensure implementation of
environmental regulations and strategies at the national, state, and community levels.
This report is the result of an investigation into the use of models, indicators, and metrics of ecosystem
goods and services as tools for evaluating decision options (community-based decision support) that
affect resource sustainability at the community level. Moreover, this report focuses on place-based
studies (PBS) as a critical research tool for integrating environmental, social, and economic priorities
into a decision framework. While the report mentions numerous PBS examined as a part of this analysis,
the intent is not to focus on them individually, but consider the collective results from all PBS under
several major theme areas associated with ecosystem goods and service (EGS)-based research. The goal
is to develop better and more useful tools for assessing how decisions affect sustainability at the
community level in order to better protect human health and the environment.
The chapters of this report are organized around four major theme areas described in Chapter 1. While
the authors worked collectively on editing the report, each chapter was the primary work of the subset of
authors with particular expertise for that theme area:
Overall report editor/introduction/synthesis - Richard S. Fulford
Stakeholder engagement/decision context - Susan Yee and Timothy J. Canfield
Final ecosystem goods and services - Paul Ringold, Kirsten Winters, Marc Russell
Ecological production functions - Randy Bruins and John B. Handy
Benefits of ecosystem goods and services - John Johnston, Nadia Seeteram
William Benson, Acting Director
National Health and Environmental Effects Research Laboratory
-------
Table of Contents
Notice ii
Foreword i i i
Tables vii
Figures viii
Acronyms and Abbreviations x
Acknowledgments xi
Executive Summary xii
1 Introduction 1
1.1 Conceptual Model of Community-Based Decision Support (CBDS) 2
1.1.1 Decision Context and Stakeholder Engagement 3
1.1.2 Final Ecosystem Goods and Services 3
1.1.3 Ecological Production Functions 4
1.1.4 Measures of Human Benefit 4
1.1.5 Conceptual Model Synthesis 4
1.2 Place-Based Studies 5
1.2.1 Why Focus on Place-Based Studies? 5
1.2.2 Identification of Place-Based Studies 5
1.2.3 Data Gathering from Place-Based Studies 5
1.2.4 Description of PBS Included in the Study 6
1.3 Goals of this Report 10
1.4 Literature Cited 11
2 Decision Context and Stakeholder Engagement 13
2.1 Introduction 13
2.2 Structuring the Decision Process 15
2.2.1 Decision Analysis Frameworks 15
2.2.2 The Role of Ecosystem Services in the Decision Process 19
2.3 Stakeholder and Decision Maker Engagement 20
2.4 Characterizing the Decision Context 24
2.4.1 Elements of Decision Context 24
2.4.2 Visualizing the Decision Problem with Conceptual Models 29
2.5 Characterizing Stakeholder Objectives and Identifying Decision Options 32
2.5.1 Objectives in Place-Based Studies 32
IV
-------
2.5.2 Defining Objectives with Performance Measures 37
2.5.3 Using Objectives to Identify Decision Alternatives 38
2.6 Estimating Consequences of Alternative Decision Scenarios 39
2.6.1 Tools for Estimating Consequences 40
2.6.2 Characterizing Tradeoffs 48
2.7 Lessons Learned 50
2.7.1 Engage Stakeholders and Local Decision Makers Throughout 50
2.7.2 Formally Define the Decision Context 50
2.7.3 Clarify Why Natural Resources Matter 50
2.7.4 Consider Unintended Consequences through Systems Thinking 50
2.7.5 Integrate Multi-Disciplinary Sources of Information 51
2.8 Conclusions 51
2.9 Literature Cited 52
3 Final Ecosystem Goods and Services 58
3.1 Background 58
3.1.1 Definition 58
3.1.2 Classification Systems of Beneficiaries 60
3.2 Methods 61
3.3 Results 61
3.3.1 Critical Evaluation 65
3.4 Conclusion 66
3.5 Literature Cited 68
4 Ecological Production Functions 71
4.1 Definition and Purpose of EPFs 71
4.2 Uses of EPFs in EPA Place-Based Studies 73
4.3 Number of EPFs Used, and Linkage to Decision Support Systems 79
4.4 Model Computational Approach and Specificity of Model Development 79
4.5 Model Variables and Input Data 80
4.6 Modeling Management Alternatives and Final Ecosystem Goods and Services 81
4.7 Representation of EPF Uncertainty, and EPF Transferability 82
4.8 General Observations on EPF Use, and Lessons Learned, in these Place-Based
Studies 83
4.9 Literature cited 85
5 Identifying and Measuring Benefits of FEGS 90
5.1 Differentiating FEGS from Benefits 90
-------
5.2 Communicating Benefits to Stakeholders 92
5.2.1 The Concept of Benefits 92
5.2.2 Terminology to Describe Benefits 92
5.3 Types of Benefits 93
5.4 Tools to Quantify Benefits 93
5.5 Measuring Success 94
5.6 Conclusions 95
5.7 Literature Cited 96
6 Synthesis of Lessons Learned 97
6.1 Components of the Conceptual Model 100
6.2 Literature Cited 102
Appendix A - Full Set of Questions Used in Information Request 103
Appendix B - List of PEGS Beneficiaries Compiled for Participating Place-Based Studies... 114
Appendix C - Models Used or Described by Place-Based Studies 121
VI
-------
Tables
Table 2.1 Examples of structured decision processes or frameworks used or developed by place-based
studies 16
Table 2.2 Examples of decision context inferred from 15 place-based study responses 25
Table 2.3. Approaches, tools, and software used by place-based studies to develop conceptual models. 30
Table 2.4 Example of using HWBI conceptual framework to identify community goals 34
Table 2.5 Examples of approaches, tools, or software used by surveyed studies to characterize
stakeholder objectives 36
Table 2.6 Examples of approaches, tools, or software used by surveyed studies to compare alternative
decision scenarios 42
Table 2.7 Hypothetical example of a consequence table (modified from 3VS, Industrial Economics, in
review) 48
Table 3.1 Initial categorization of responses to the question: "What ecological endpoints did you use [in
your study] relative to your beneficiary list?" into Intermediate or Final Ecosystem Goods and
Services as well as other categories 62
Table 4.1. Overview of place-based studies, ecological production functions or decision support systems
used, and documentation of models or modeling (see Appendix C for details). Citations in italics
apply to the EPFs or DSSs in question, but not specifically to the PBS location 74
Table 4.2. Approximate numbers of input variables per EPF, where "drivers" are variables whose values
change over space or time, and "constants" do not vary within the application. Values are for 75
EPFs from four well-documented case studies 80
Table 6.1 Summary of PBS responses to question, "If I could change something about the project what
would it be?" First seven columns were options in a menu, while the final column was an open
response 99
VII
-------
Figures
Figure 1.1 The conceptual model central to our project focuses on the process of informing decision
making through the use of ecological production functions (EPFs), ecosystem goods and services
(EGS), and indicators of human well-being 3
Figure 1.2 Map showing locations and names for all the place-based studies participating in our
information request on the use of an EGS approach for local decision support 9
Figure 2.1 Survey responses from place-based studies as to whether a structured decision process was
used or developed for their study 15
Figure 2.2 The steps in a generic decision process 17
Figure 2.3 Decision support framework for theDASEES tool 19
Figure 2.4 Integrating PEGS concepts into a generic decision process 20
Figure 2.5 Survey responses from place-based studies as to whether community stakeholders were
involved 21
Figure 2.6 Percentage of place-based studies which engaged or wished they had engaged different types
of stakeholder group. A single study may have engaged more than one type of stakeholder 22
Figure 2.7 Percent of place-based studies out of fifteen that identified having direct involvement with
stakeholders throughout the study 23
Figure 2.8 Percent of place-based studies, with either direct or indirect stakeholder involvement, which
identified certain kinds of stakeholder involvement. A single study may have more than one kind of
stakeholder involvement 23
Figure 2.9 Percentage of the fifteen study responses identifying having used or developed a conceptual
model 29
Figure 2.10 Example conceptual framework for conducting ecosystem services assessments 31
Figure 2.12 Percentage of the fifteen place-based studies which included an identification of stakeholder
objectives. A single study may have used more than one method to characterize stakeholder
objectives 33
Figure 2.13 Reasons identified by place-based studies for characterizing stakeholder goals. A single
study may have identified more than one reason 37
Figure 2.14 Percentage of fifteen place-based studies that included an identification of decision options.
A single study may have used more than one method to identify decision options 38
Figure 2.15 Example means-ends network illustrating how ecosystem services, social services, and
economic services are means to achieving a fundamental objective of improving human well-being.
39
Figure 2.16 Percentage of fifteen place-based studies that included an effort to compare decision
options 40
Figure 2.17 Screenshot from EPA H2O tool illustrating area of interest (dashed black polygon) and
stream connectivity network defining the upstream area of interest (red line) 43
Figure 2.18 Hypothetical BBN describing probabilities of water quality, reef health, tourism economy,
and fishing outcomes under a proposed lagoon restoration decision (grey box) 45
viii
-------
Figure 2.19 Conceptual framework for Envision spatially-explicit decision support tool 46
Figure 2.20 Conceptual framework for 3VS integrated assessment tool 47
Figure 3.1. Total Economic Value categories integrated with FEGS-CS categories of value 61
Figure 4.1. Examples of relationships between ecological productions functions (EPFs) and decision
support systems (DSSs) in the modeling of ecological processes 73
Figure 5.1 Responses on scale of 1-5, where 1= "not at all important" and 5= "extremely important" to
determine how important various sources were in ES identification 91
Figure 5.2 Frequency of use of various terms to describe "natural resources" 93
Figure 5.3 Results depicting frequency of tool use for measuring changes in Ecosystem Services/Human
Health and Human Well-Being 94
Figure 5.4 Responses to "How was success measured in line with project outcomes?" 95
IX
-------
Acronyms and Abbreviations
AOC Area of Concern
BBN Bayesian Belief Networks
BMPs Best Management Practices
BRI Benefit Relevant Indicators
CBDS Community-Based Decision Support
CICES Common International Classification of Ecosystem Services
DASEES Decision Analysis for a Sustainable Environment, Economy, and Society
DSS Decision Support System
DST Decision Support Tool
DP SIR Drivers-Pressures-States-Impacts-Response
EBFs Ecosystem Benefits Functions
EGS Ecosystem Goods and Services
EJSCREEN Environmental Justice and Mapping Screen Tool
EPA U.S. Environmental Protection Agency
EPFs Ecological Production Functions
PEGS Final Ecosystem Goods and Services
GIS Geographic Information System
HIA Health Impact Assessment
HWBI Human Well-Being Index
IAF Integrated Assessment Framework
LEGS Intermediate Ecosystem Goods and Services
IRM Integrated Resource Management
LTER Long Term Ecological Research
LULC Land-Use/Land Cover
MVD Making a Visible Difference community
NESCS National Ecosystem Services Classification System
ORD Office of Research and Development
PBS Place-Based Study
TEV Total Economic Value
VELMA Visualizing Ecosystems for Land Management Assessments
-------
Acknowledgments
We acknowledge the support of the task leads for the 15 place-based studies (PBS) that participated in
our request for information, as well as staff from the Sustainable and Healthy Communities Research
Program that facilitated distribution of the information request. We acknowledge Marilyn ten Brink for
her assistance in finding representative place-based studies for this work. Pat Clinton assisted with
development of figures. Chloe Jackson and Kate Murphy provided editorial support for the final report.
Angelica Sullivan assisted with formatting and distributing our information request in SharePoint. Joel
Hoffman, Steve Balogh, and Lisa Wainger served as reviewers for this report. The following individuals
participated in our information request for a place-based study.
Contributor
Jana Compton
Brian Dyson
Florence Folk
Richard Fulford
Joel Hoffman
Anne, Kuhn
Marisa Mazotta
Bob McKane
Marc Russell
Michelle Simon
Jason Turgeon
Lisa Wainger
Henry Walker
Jeff Yang
Susan Yee
PBS Locations
Southern Willamette Valley, OR
Dania Beach, FL
Suffolk county, NY
Pensacola, FL; Opelousas, LA; Vero Beach, FL; Thibodaux, LA,
Pascagoula, MS; Lewisville, NC; Woodbine, IA, Windsor Locks, CT;
Freeport, NY
Duluth, MN, Superior, WI
Taunton River Watershed, MA
Woonasquatucket River, RI
Blue River Watershed, OR
Tampa, FL
Birmingham, AL
New Bedford, MA
Chesapeake Bay, MD
Narragansett Bay, RI
Lawrence, MA
Guanica Bay, PR
XI
-------
Executive Summary
Human well-being is inextricably connected to the sustainable use of natural and built resources. The
ecosystem goods and services (EGS) concept has become increasingly valuable for identifying and
evaluating important trade-offs among diverse beneficiary groups and by extension has become a central
element of decision support for both public and private institutions. We extend that model here by
referring to final ecosystem goods and services (PEGS) as those goods and services directly linked to a
human beneficiary, as this allows for a direct link to human benefit to be integrated into discussion. The
U.S. Environmental Protection Agency (EPA) has been particularly active in researching methods for
incorporation of PEGS into decision making to protect human health and the environment.
Place-based studies (PBS) are a critical element of FEGS-based research, as they more fully integrate
environmental, social, and economic services into evaluation of decision alternatives. PBS provide a
proving ground for the operationalizing of scientific information for decision making through the
integration of science with social, economic, and environmental characteristics of a place. As such, the
application of PEGS concepts in a PBS-based research program is a valuable step in development of
effective science-based decision support tools.
This report is intended to describe lessons learned from the application of FEGS-based research in a
series of PBS conducted by EPA's Office of Research and Development (ORD) and make this
information available and useful for planning future research into local decision support for
sustainability. A key goal of this report is to break the PEGS concept into a series of steps, called the
"PEGS approach," and examine how each of these steps may, or may not, have been applied in prior
PBS. To begin, we introduce a general model of local decision making based on the EGS concept. This
conceptual model represents a guidance tool for future research.
This report concerns existing and past placed-based research in ORD with the objective of describing
how this research has (or has not) applied the elements of our EGS-based conceptual model for decision
support. The data used for this analysis of EGS use in place-based studies comes from an information
request sent out to project and task leads throughout the EPA Office of Research and Development. A
total of 25 place-based studies participated in the data request. These sites are distributed throughout the
continental US and Puerto Rico and cover a spatial scale from individual municipalities to watersheds
containing multiple communities. The data request was organized into four chapters, each corresponding
to chapters of the report (Decision Context/Stakeholder Engagement, Final Ecosystem Goods and
Services [PEGS], Ecological Production Functions [EPF], and Human Benefits).
Each chapter addresses specific elements of the conceptual model with the goal of explaining how these
specific elements have been used in practice, why they did what they did, what was the result, and what
could have been the result if that element was not a part of the work. Chapter 6 contains a synthesis of
lessons learned across all the model elements and also addresses how the model might be used going
forward to develop tools and approaches for community-based decision support focused on
sustainability of ecosystem goods and services.
XII
-------
Overall the application of FEGS-based decision support across the PBS participating in this study can be
described as pervasive but incomplete. Multiple studies reported use of EGS concepts, stakeholder
engagement, application of environmental models, or direct measures of human benefit linked to
environmental action. Yet, the focus was almost universally on a subset of these elements usually tied to
the specific project objectives. For instance, in one PBS located in the Pacific Northwest, the emphasis
was on the application of quantitative models to address existing land management issues, but because
the issues were well-established, minimal stakeholder engagement occurred during the project and direct
measures of human benefit were not developed. This incomplete application of the core PEGS elements,
as well as the need for integration of these elements into a cohesive approach, represents a major area
for future work in decision support research. More detail specific to the four report chapters are given
below.
Decision Context/Stakeholder Engagement
Many methods, tools, and approaches have been used to integrate ecosystem services into a community
decision process and to engage stakeholders in those decisions. This review was divided into 5 steps
with the first step to identify the degree to which studies have used a structured process to identify five
decision-relevant science needs or facilitate decision making (Chapter 0), and the degree to which
studies involved stakeholders in their process (Chapter 2.3). Second, examples are provided of decision
contexts for which ecosystem goods and service concepts are directly relevant from the place-based
studies (Chapter 2.4). Third, ecosystem services concepts are placed within the larger context of
community social and economic goals (Chapter 2.5). Fourth, the degree to which place-based studies
evaluated changes in ecosystem goods and services under alternative decision scenarios, and some of the
tools and approaches used for scenario analysis, are reviewed (Chapter 2.6). Finally, a synopsis is
provided of the lessons learned from place-based studies when linking ecosystem services concepts to
community decision making (Chapter 2.7).
Structured decision analysis provides one approach for evaluating tradeoffs in a way that encourages
greater public participation, collaborative decision making, and allows consideration of multiple
attributes. A problem with ecosystem services assessments is that they are often aimed at adapting tools
to a problem, rather than allowing the problem to determine the appropriate set of tools. To accomplish
this, more studies are needed that integrate ecological structure and function, ecosystem services, human
welfare, and decisions into a single study. The place-based studies examined here provide more than a
dozen examples of studies that are attempting such an integrated approach, illustrating tools and
approaches with a high potential for transferability to other communities and relevance to decision
makers. Developing guidance that clearly lists the common themes identified in this report will assist
future studies with the incorporation of stakeholder engagement that will lead to more productive
outcomes of the process. Ultimately this will lead to better decision making that promotes more
sustainable approaches to balancing the gives and takes inherent between economic, environmental and
social aspects in decisions communities face every day.
Final Ecosystem Goods and Services (FEGS)
A key to collaboration between stakeholders and natural and social scientists is the identification and
measurement of indicators of final ecosystem goods and services. FEGS indicators measure what
directly affects people's welfare. Figuring out what directly matters to community stakeholders is not as
straightforward as it may seem. There is no crisp definitional test of an "outcome that directly matters."
However, we can apply certain principles in our search for such outcomes. We first need to describe
environmental change in terms of service production and we then need to connect changes in service
xiii
-------
production as directly as possible to human welfare. Critical to the second principal are efforts to define
specific metrics and indicators of PEGS that are linked to specific beneficiaries. Specification of
indicators of PEGS therefore starts with the identification of the beneficiary.
EGS-specific information was provided from four of the information request questions. Responses to
these questions from PBS provided twenty-six suspected intermediate EGS metrics or indicators, fifteen
possible PEGS metrics or indicators, nine for an economic good or service, and three that were
uncertain.
On the basis of inspecting these PBS responses we identified five conclusions:
1. One case study claimed to look at existence values, and none looked at option values.
2. Endpoints listed are often intermediate measures, or single attribute measures, rather than PEGS.
3. Methods for identifying beneficiaries are highly variable.
4. Some beneficiaries previously identified were not human, (e.g., bald eagles).
5. Some areas of refinement of the FEGS-CS system came from a result of this survey.
The primary lesson learned regarding stakeholder engagement in PBS is that place-based study
practitioners need a more detailed implementation of the PEGS approach in future studies.
Ecological Production Functions (EPFs)
Quantifying the production of ecosystem goods and services in natural systems in response to human
impacts is an important method for the use of scientific data for guiding decisions. Accounting
approaches that estimate ecosystem services only as a function of ecosystem areas are useful for
estimating the ecosystem-service impacts of management actions that change land-use/land cover
(LULC), but they are insufficient for estimating impacts of management actions that impose other kinds
of changes. When the environmental management decisions that communities face entail changes in
water or air pollutant delivery, species and habitat abundance, or other ecological processes, predictive
modeling approaches may also be a valuable tool for decision support.
Most of the place-based studies that were considered for this study reported some use of models to
estimate endpoints related to the production of ecosystem services. Of 15 case studies examined, 13
reported the intention to model ecosystem services-related endpoints, seven reported completion of at
least some modeling, and seven reported that modeling was underway or being planned. Results of
modeling have already been published in five cases; in several cases, no documentation was available of
case study modeling, but documentation was available of the models themselves.
Seven of the case studies used, or intended to use, decision support systems (DSS) (e.g., Envision, EPA
H2O) to coordinate multiple models and examine more complex scenarios. For example, the Tampa Bay
study's EPA H2O tool (Russell et al. 2015) uses static land use conditions to drive the following EPFs
for each subwatershed included in the model: stormwater retention, based on an improved soil-specific
calculation of water retention; air pollutant removal and carbon sequestration by vegetation using the
UFORE or i-Tree Eco model (USDA 2012); and an arithmetic summation of geographic features of
ecological interest. The integration of models allowed for the examination of multiple land use changes
at once such as watershed urbanization combined with shoreline restoration. The Guanica Bay case
xiv
-------
study used the Envision decision support platform to link a dynamic coral reef model to 28 other EPFs
corresponding to specific, service-related endpoints (Yee et al. 2012). The use of Envision greatly aided
the consideration of scenarios including both agricultural land use and coastal tourism issues. Decision
support systems help integrate multiple EPFs both in terms of input data and the usefulness of the output
for guiding decisions.
These PBSs were encountered at different stages of execution, and the modeling information obtained
was therefore highly uneven. It should be noted that several issues of importance in EPF selection and
use, such as spatial extent and scale, could not be evaluated very well via the information request
employed for this study. Nor could the level of integration of the modeling process with all aspects of
the decision context, such as stakeholder engagement and benefit estimation, be evaluated, nor whether
trade-off analysis were systematically examined. Nonetheless, in several of the completed and
documented studies, EPF use was relatively well executed and integrated, showing several
characteristics that bode well for the continued improvement of ecosystem-service modeling state-of-the
art and feasibility:
• The frequent use of multiple EPFs to estimate production of multiple services is important for
enabling robust trade-off evaluation and avoiding the problem of unforeseen consequences.
• The practice of linking different EPFs enables the use of existing, relatively complex process
simulation models (off-the-shelf or with adaptation) to model management action influences on
ecosystem processes, to be complemented by relatively simple, logic-based models that are
readily adapted to local conditions of final ecosystem service delivery and use.
• The bundling of these different EPFs within a decision support tool (DST) coordinates the
computation of multiple EPFs for given scenario conditions, and may enable policy simulation or
optimization.
• Many of the models used had been used previously, or were considered by the investigators to be
readily transferable.
At the same time, several deficiencies in how models were chosen or used were noted, or were
suspected:
• The lack of replication of any EPF or DSS across these case studies may speak in part to the
small number of studies involved, the diversity of their settings and the relative newness of
efforts to include ecosystem services, but it may also suggest an investigator-specific approach
has been taken to model selection. Coordination of tools use across communities facing similar
issues would greatly aid the comparability of results and the universality of the solutions. A more
formal approach to model suitability evaluation may be needed. Both practical guidance for
assessing model transferability, and the establishment of communities of practice around specific
EPFs or DSS', or around EGS modeling in general, could help to make model evaluation and
model transfer more systematic.
• Part of that suitability evaluation would require more systematic approaches to the various facets
of uncertainty assessment, and ensuring that uncertainty assessment is built into DSS so as to
become more routine - not only to modeling practitioners but to users of ecosystem service
production estimates.
XV
-------
• In the information available for this analysis, it was not always clear that EPF choice was based
on the decision context, as well as the ecosystem characteristics, of the PBS site. Therefore, a
final issue for increased emphasis is ensuring alignment of EPF capabilities with: (a) the
characteristics of management alternatives (which generally should align with model inputs) and
(b) stakeholder goals and the PEGS that most closely meet them (which should align with model
outputs). This alignment-to-context may be accomplished in a single EPF or via multiple, linked
EPFs.
Human Benefits
As stated in the discussion of stakeholder engagement, it is critical to the application of ecosystem goods
and services to link change in service production directly to human welfare. Comparison of the value
placed on ecosystem services across communities is problematic because of varying level of awareness,
understanding and appreciation of the concept of PEGS benefits as it relates to human health and
welfare. A benefit is something that has an explicit impact on changes in human welfare, such as more
food, better hiking, less flooding, which differs from PEGS, which are the "components of nature,
directly enjoyed, consumed, or used to yield human well-being". As seen in the conceptual model In
Chapter 1, benefit functions represent the link between PEGS and human well-being, and "demonstrates
what people would be willing to pay to achieve a gain or avoid a loss in an ecosystem service or
suggests a relative magnitude of social value when willingness-to-pay is not measurable" (Wainger and
Mazzotta 2011). In other words, benefits can be quantified through economic valuation methods or
through assessing tradeoffs when presented with limited information. Conducting these types of
assessments provides vital information upon the contributions of PEGS towards human well-being.
Identifying PEGS and the resulting benefits can be achieved through a variety of means. Within the
place-based studies, most project leads («= 15) indicated that the stakeholder engagement process
(32.3%) and literature reviews (32.3%) were the most widely used sources for ES identification, but
local expert consultation (25.8%) and other sources including peer groups, social media, news sources
(9.7%) were also utilized.
When asked if stakeholders struggled with understanding the definition of benefits used in a particular
PBS, 64.3% of project leaders stated "no," while 21.4% indicated that stakeholders did struggle with this
concept. Only 14.3% of project leads expressed that stakeholders had a "mixed response" to the
explanation of this concept. In order to increase comprehension around benefits, project leads suggested
that "further explanation of benefits and ES" and "connecting their priorities to the domains of human
well-being" were helpful in reducing misunderstandings.
Project leads were asked to indicate which of the following terms they used during the stakeholder
engagement process to describe "natural resources." Results are presented in Chapter 5 with most
(33.3%) of project leads indicating that they used "ecosystem services" to describe "natural resources,"
followed by 14.8% of project leads indicating they also used "nature's benefits" and "environmental
value," within these discussions. When asked how frequently they used these terms, project leads
expressed that they used "ecosystem services" most (37.5%) followed by "nature's benefits" (18.7%).
Within place-based studies that incorporate ecosystem services into their project framework, success is
dependent on the goals of the study, and there can be several. Complete follow-through on the
identification of benefits would be an incorporation of agreed-upon measures of these benefits (i.e.,
benefits indicators) into the goals of the project. However this was not at all common among the PBS
included in this study. In terms of defining and measuring success, "stakeholder engagement" elicited
xvi
-------
the highest degree of prioritization with 21.7% of respondents, followed by 17.4% of project leads who
selected "publications/dissemination" of results as measure of success. "Other" measures of success was
another frequently selected option. When asked to elaborate on this selection, project leads indicated that
"increased interest, support, and participation in the study," "creating awareness about environmental/
health issues within local communities," and "making an impact on [the] decision" were amongst the
responses. One of the most effective ways of achieving the latter goal includes developing "benefit
relevant indicators" (BRIs) that directly integrates the benefits people receive from enhancement of
EGS. The Woonasquatucket PBS demonstrated how hydrologic and hydraulic simulation model results
can be used to address critical questions regarding EGS provisioning and beneficiaries in order to
develop indicators for a specific decision context (Bousquin et al. 2015). However this PBS was the
exception rather than the rule for integrating benefits into assessment of project success.
A few conclusions can be drawn from the information presented in Chapter 5. First, project leads
indicated that local expert consultation was "extremely important" in identifying ES for use in their
studies. While consulting local experts may have been the most effective source to use in ES
identification, project leads utilized a variety of sources for the most comprehensive understanding of
priority ES. In doing so, project leads were likely well equipped to engage with stakeholders, local
decision makers, and experts. Following this positive conclusion, project leads observed that the
majority of stakeholders did not struggle in understanding the concept of "benefits," even though they
used the term "ecosystem services" to frame the discussion more frequently than "nature's benefits" or
"nature's value." Further studies should be conducted to test the effectiveness of using one phrase over
the other, as terminology choice and framing can substantially alter discussions with stakeholder and
local decision makers. The identification and practical application of benefit indicators is a key gap
observed in the PBS considered for this study and this represents a key area for future research into
decision support and its acceptance at the community level.
Conceptual Model for Application of Ecosystem Goods and Services
In this report, a complete conceptual model for an EGS approach at the community level has been
evaluated in the context of existing and previous place-based studies with an eye towards how this
model has been used, and what gaps exist that might be filled to maximize its successful use in future
PBS. Taken as a whole, the model provides critical linkages across the respective elements that can
bring about a novel integration of science and policy and yield much more effective measures of
decision outcomes. Stakeholders bring an understanding of both potential actions and the desired
outcomes from those actions; and science provides a defendable, robust understanding of how actions
can translate into desired outcomes. Barriers to such an integration of science and policy include a lack
of stakeholder involvement and understanding of an EGS approach, challenges of matching EPFs to the
problem at hand, use of inadequate short-term objectives as measures of benefit, and a need to better
integrate multiple issues into a common decision framework. An EGS-based conceptual model for
decision support, such as the one proposed here, can be highly effective in overcoming these challenges
by linking the decision process together in a clear way, but more work needs to be done. Place-based
studies offer a rich opportunity to explore the application of this conceptual model to real-world issues
and as such are a vital link in EPA research into community-based decision support and the fostering of
sustainable human and environmental health.
XVII
-------
1 Introduction
Human well-being is inextricably connected to the
sustainable use of natural and built resources (Summers et
al. 2012). Yet, the complexity of resource sustainability
yields unavoidable trade-offs, as different decision priorities
have different impacts on resource availability and human
well-being. The ecosystem goods and services (EGS)
concept has become increasingly valuable for identifying
and evaluating these trade-offs (Fulford et al. 2015; Ringold
et al. 2013) and by extension has become a central element
of decision support for both public and private institutions
(Olander et al. 2015). The U.S. Environmental Protection
Agency (EPA) has been particularly active in researching
methods for incorporation of EGS into decision making to
protect human health and the environment (Fulford et al.
2016; Ringold et al. 2013; Smith et al. 2013).
There is a history of success within EPA in applying the
EGS concept at the local level in place-based study (e.g.,
Bousquin et al. 2015; Hopton and Heberling 2012). All
research is done in a place, but place-based studies are
unique in that they integrate important characteristics of the
study location, which may include community stakeholders,
specific local issues of concern, atypical elements of the
ecosystem, or a focal endpoint such as solving an acute
problem (e.g., pollution reduction). What sets a place-based study apart from traditional problem-based
research, and makes them valuable, is the integration of basic research with practical problem solving.
Place-based studies are a critical element of EGS-based research, as they more fully integrate
environmental, social, and economic services into evaluation of decision alternatives. This fuller
integration is the result of using specific local issues and stakeholder priorities to translate environmental
data into viable decision options. Scientific data relevant to an EGS approach to decision making can be
collected in a variety of ways. However, in application these data must be made functional for decision
making through the process of translating the data into both decision alternatives (Hopton and Heberling
2012) and decision criteria (Bousquin et al. 2015). Place-based studies provide a proofing ground for the
operationalizing of scientific information for decision making through the integration of science with
social, economic, and environmental characteristics of a place. As such, the application of EGS concepts
in a PBS-based research program is a valuable step in development of effective science-based decision
support tools.
Photo credit - Nadia Seeteram
-------
This report is intended to describe lessons
learned from the application of EGS-based
research in a series of place-based studies
conducted by EPA's Office of Research
and Development (ORD) and make this
information available and useful for
planning future research into local
decision support for sustainability.
However, rather than focusing solely on
the outcomes of this prior research, a key
goal of this report is to break the EGS
concept into a series of steps and examine
how each of these steps may, or may not,
have been applied in prior PBS. To begin,
we introduce a general model of local
decision making based on the EGS
concept. This conceptual model is itself
based on previous work, but also
represents a guidance tool for future
research. This conceptual model will be
introduced here, and in subsequent
chapters, will be applied to the question of
whether the necessary elements of EGS
based research, as we have defined them,
were present in previous PBS.
Differences between place-based ecosystem studies and
question-based ecosystem studies.
Place-based studies
Focused on unique aspects
of a place as an integrated
part of the study to address
a more complex and
specific problem
Stakeholders either
considered a part of the
study or directly engaged in
project
Research addresses a
current, specific local issue
Question-based studies
Place is either a non-unique
replicate or unique aspects are
minimized to facilitate
generalization of outcomes
Stakeholder involvement
limited usually to development
of the research question and
reporting of outcomes
Issue is generalized to fit a
broader question of interest
Project outcomes consider
both science and policy
outcomes as integrated
Project outcomes largely
limited to science outcomes
usually with reporting to policy
makers
1.1 Conceptual Model of Community-Based Decision Support (CBDS)
Decision support can take on many forms depending on the target audience, the specificity of the
problem, or the ultimate objectives of the decision. Recent guidance highlights three best practices for
integrating ecosystem services into federal decision making (Olander et al. 2015). These include
connecting assessments to both scientific data and stakeholder values, establishing well-defined
measures of success, and including a comprehensive set of services to people. At the local level,
achieving the first practice is greatly aided by a PBS approach to research. The conceptual model of an
EGS-based approach is intended to complete the set of best practices with a focus on a comprehensive
suite of measurable outcomes important to people. In particular, a common theme of sustainable final
ecosystem goods and services (PEGS) is proposed as a decisional objective. An PEGS approach is
defined by two target outcomes: sustainable production of an PEGS (e.g., catchable fish biomass), and
the delivery of that PEGS to an identifiable beneficiary (e.g., angler). Both the goods/services and the
beneficiary must be present to measure PEGS. This is a distinction from simply considering EGS
production and clearly links EGS to human benefit as a measure of decisional outcomes. The advantage
of this approach is that human stakeholders are simultaneously identified as decision makers and
beneficiaries yielding a feedback loop with measurable outcomes. Here, we introduce a conceptual
model (Figure 1.1) of decision support based on PEGS, which will provide structure for the examination
of previous place-based studies. First, we will discuss the components of this conceptual model and then
consider the value of the model as a coherent strategy for comparison of place-based studies.
-------
SHC 2.61 Draft Causal Conceptual Framework
Mgmt
Actions
A Biophysical State of
the Ecosystem
(includes intermediate
EGS)
Information for Decision Support
Figure 1.1 The conceptual model central to our project focuses on the process of informing decision making
through the use of ecological production functions (EPFs), ecosystem goods and services (EGS), and
indicators of human well-being.
1.1.1 Decision Context and Stakeholder Engagement- The initial step in decision support is to
identify the decision under consideration and the stakeholders invested in that decision (Gregory et al.
2012). We need to be clear that beneficiaries are not necessarily stakeholders. There will be stakeholders
impacted by a decision that are not identifiable until decision options are clearly defined. For that reason
stakeholders at this stage are individuals involved in the decision making process, which may or may not
include all beneficiaries. Stakeholder engagement is the process of defining decision context (e.g.,
"What are we trying to do?"), decision alternatives, and measures of consequences that will be used to
evaluate these alternatives. Chapter 2 will introduce stakeholder engagement as the entry point for the
conceptual model and examine how stakeholder engagement has been used in PBS to define the local
decision context and develop measures of success.
1.1.2 Final Ecosystem Goods and Services - In keeping with the best practice of connecting
ecological services to peoples values, PEGS are defined as the joining of EGS to human beneficiary.
These final services (e.g., harvestable fish) can be distinguished from more intermediate services (e.g.,
fish habitat) that are important but require additional steps to reach a human beneficiary. The PEGS
concept represents an important step in decision support by clearly linking environmental change to
benefits enjoyed by people. Chapter 3 will address how and to what extent existing and previous PBS
applied the PEGS concept as we have defined it. This involves the identification of both affected EGS
and beneficiaries for the chosen decision context and suite of decision options considered in the study.
One key question addressed in this chapter is whether an PEGS framework is being applied even in
cases where it is not identified as such. In other words did they use an PEGS approach without the PEGS
label? This will be an important lesson as to whether the application of the PEGS approach to decision
-------
support requires a large change in approach or just an identification and unification of existing pieces.
1.1.3 Ecological Production Functions - Ecological production functions (EPFs) have been
defined as "usable expressions (i.e., models) of the processes by which ecosystems produce ecosystem
services, often including external influences on those processes" (Bruins et al. 2016). These functions
can be as simple as a map that shows increases in greenspace resulting from a community planning
effort and as complicated as a complete hydrological simulation model that links changes in watershed
land use practices to changes in estuarine water quality hundreds of miles away. The connection is a
desire to qualitatively or quantitatively link a decision to an environmental result. These EPFs represent
the definitive use of scientific data in the decision process and the most promising methodology of
achieving the best practice of assessing ecosystems using well-defined measures of change (Olander et
al. 2015). Chapter 4 will describe how EPFs have been used in previous and existing PBS, including a
characterization of these EPFs by complexity, data needs, and outcome. The objective is to learn where
types of EPFs may be most effective. This analysis of EPFs will be necessarily linked to the analysis of
FEGS as a target output, but since not all EPFs use FEGS as an output, the EPF-FEGS connection will
be a useful arena for future study.
Best practices for integrating ecosystem services into federal decision making as outlined by the
National Ecosystem Services Partnership (Olander et al. 2015):
1) "Extend assessments beyond purely ecological measures that are not explicitly tied to people's
values to measures of ecosystem services that are directly relevant to people."
2) "Assess ecosystem services using well-defined measures that go beyond narrative description
and that are appropriate to the analyses, even when data, time, or resources are limited."
3) "Include all important services, even those that are difficult to quantify."
1.1.4 Measures of Human Benefit - Finally, clear measures of human benefit that consider all
important services to people from the ecosystem are a critical end point of best practices for EGS-based
decision making (Olander et al. 2015). This represents the most important element of the conceptual
model as the answer to "What do we get out of it?" when we make restrictive decisions or decisions that
require a high level of investment. It is also the element of the model that is the least developed at the
operational level in that most community decision options are evaluated based on short-term gain (e.g.,
reduced costs of service) or the attainment of clearly definable goals (e.g., recycled x tons of garbage).
Measuring human benefit requires that we identify specific beneficiaries, as well as a meaningful metric
of benefit. This can become quite complicated and has been the subject of extensive research (e.g.,
Johnston and Russell 2011; Landers et al. 2016). Chapter 5 will explore how benefit has been measured
and reported in previous and existing PBS with an eye towards how the PBS approach can be leveraged
to provide the most comprehensive results possible.
1.1.5 Conceptual Model Synthesis - Each element of the conceptual for EGS-based decision
support have been addressed in turn, which will lead to a synthesis of this information and the
compilation of elements into a cohesive model for decision support. The final synthesis chapter of the
report will synthesize lessons learned from previous chapters and also compile the information into a
discussion of the entire conceptual model as a tool for future research. Where are the strong points?
-------
Where are the gaps in use or knowledge? What is needed to move this model forward to application?
This is intended to provide a clear path for the analysis, highlight how the separate elements are linked
together, and also to build a common framework for PBS research into decision support in the future
(Figure 1.1).
1.2 Place-Based Studies
1.2.1 Why Focus on Place-Based Studies?
For the purposes of this report, a place-based study is defined as any study that is connected to unique
aspects of the location of the work. These unique aspects will most often be associated with a target
community, but could also be components of a watershed or other physical feature of the place. We also
make no a priori limitations with respect to spatial scale so as to not limit the definition but rather allow
characteristics of the place to define themselves. The only limitation is that research results are tied to
the locality and its human inhabitants.
Place-based studies serve a unique role in environmental research as case studies for both causes and
effects of human stress on the environment. In contrast to cross sectional studies that are highly focused
on replicating a necessarily limited analysis over multiple sites, place-based studies allow for a fuller
story to be told in a particular place. This can limit generality, but is optimal for longer term ecosystem
studies. The classic EPA application is a case study that approaches a practical problem as an example.
This analysis of EGS applications in place-based studies is intended to inform future research in
community-based decision support through the application of a common EGS-based conceptual model
for decision support. The objectives of this analysis of practices in existing and completed PBS are to: 1)
summarize outcomes of previous research in the context of decision support; 2) examine the utility and
value of individual components of EGS research based on a common conceptual model; and 3) identify
gaps in the application of the conceptual model found in some or all PBS examined. Initially we limited
the scope of our data gathering to ORD projects, mainly to provide a logical stopping place, but also in
acknowledgement of the fact that many place-based studies outside ORD are very action-specific ("the
what") and do not consider goals ("the why") or process ("the how") in sufficient detail. This means that
our primary criteria for consideration of a place-based study was that they considered all three elements
in their presentation and analysis.
1.2.2 Identification of Place-Based Studies
Data gathering began with a general call for identification of place-based studies in ORD through the
ORD Sustainable and Healthy Communities (SHC) Research Program. Facilitators within the SHC
program were asked to circulate a call for information that included not just research in SHC but also in
the other major research areas across ORD. The call described our objectives, asked for point of contact
information and requested that interested parties complete a short questionnaire (six questions; attached
below) to provide an introduction to the project to aid identification of qualifying studies for follow-up.
This effort yielded a list of 52 studies for additional data gathering.
1.2.3 Data Gathering from Place-Based Studies
Once candidate studies were identified, a more detailed questionnaire (39 questions; Appendix A) was
developed to provide summary information about the project decision context and the components of
each project as it aligned with our EGS conceptual model. This data request was organized into four
-------
sections, each corresponding to chapters of the report (Final Ecosystem Goods and Services [PEGS],
Ecological Production Functions [EPF], Human Benefits, and Decision Context/Stakeholder
Engagement). The full set of questions was evaluated for completeness and clarity in one test PBS from
Guanica Bay, Puerto Rico, and minor revisions were made before being declared final. The final
complete set of questions was then adapted to an online tool, and requests to complete it were sent out to
Project Leads for each of the 52 studies identified during Phase I. Once the request was out, multiple
reminder emails were sent out to maximize response rate. No minimum response rate was set as this was
an information request not an analytical survey.
A total of 15 responses were received for our data request covering 25 place-based studies. Those
responses were compiled together, and answers were examined independently for each element of the
conceptual model to address chapter-specific questions. Overall analysis of the responses was limited to
a summary of each place-based study included in the report. Specific chapters may expand their dataset
beyond data from the questionnaire (e.g., project reports), but they will include these results as a
baseline analysis common to all four chapters.
1.2.4 Description of PBS Included in the Study
These 25 PBS sites are distributed among 15 studies described below located throughout the continental
US and Puerto Rico (Figure 1.2) and cover a spatial scale from individual municipalities to watersheds
containing multiple communities. Each PBS is briefly described here including an answer to "Why
here?" as a frame of reference for the presentation of results.
Birmingham, AL, is an urban community in northern Alabama. This PBS is an MVD community,
and research here is focused on control of stormwater effects. This was a model-centric analysis with
stakeholder engagement largely focused on informing community leaders of the results. This site
was chosen based on available resources and an acute need for decision support tools.
Blue River Watershed (BRW), OR, is a tributary of the McKenzie River in central Oregon. The
BRW is dominated by Blue River Lake, contains the HJ Andrews Experimental Forest, and is an
NSF Long Term Ecological Research (LTER) site. The focus of this PBS is prediction of impacts of
forest management approaches on production of ecosystem goods and services. The research is
largely model-based, and the location was chosen because it is relatively data rich and well
positioned for analysis of management tradeoffs. No fixed community is a part of the study, but the
work was conducted in cooperation with state and federal management stakeholders. More involved
stakeholder engagement is forthcoming now that the tool development phase is near completion.
Chesapeake Bay, is located in MD, but the Watershed is spread across 6 states. It is the largest
estuary on the Atlantic coast of the US. This PBS is the site of extensive collaborative work on
nutrient load reduction and research into the effects of nutrients on EGS involving EPA, the National
Oceanic and Atmospheric Administration (NOAA) and the decision authority of MD, VA, WV, DE,
PA, and NY. The focus of this PBS is an analysis of tradeoffs between nutrient load reduction
strategies (e.g., point vs. non-point) including both costs, projected economic and EGS benefits of
each potential action. The goal is development of a decision framework that produces an efficiency
analysis of the most benefit to EGS production for the least cost. This PBS was chosen based on
strong leadership and stakeholder support, data availability, and acute need based on focal legislation
to reduce nutrient load to this ecosystem (Messer et al. 2012).
-------
Dania Beach, FL, is a small community on the Florida Atlantic coast. This community is highly
dependent on coastal natural resources both economically and as a facet of community identity. This
community has shown strong interest and leadership in the area of climate resilience and this PBS
involved an applications of an existing decision tool to the development of a climate action plan to
protect coastal resources for future generations.
The GMeCCS communities were nine specific towns throughout the US: Pascagoula, MS;
Pensacola, FL; Lewisville, NC; Windsor Locks, CT; Freeport, NY; Woodbine, IA; Vero Beach, FL;
Thibodaux, LA; and Opelousas, LA. The focus of this PBS was to solicit stakeholder priorities and
fundamental objectives through direct (workshop) and indirect (document review) methods. The
work was grouped and conducted in a consistent manner in each community so that comparisons
could be made with the resulting information. The objective was to define measures of success for
decision making that transcended the current focal issue and were applicable to multiple potential
decisions the community might have to make. The communities were selected because they were
open to working with EPA, had available resources and personnel, and an interest in sustainability.
Guanica Bay, Puerto Rico, is a small bay on the southwestern coast of the island that drains a
complex agricultural watershed. This place-based study focused on a watershed approach to manage
impacts of land use and water quality on nearshore coral reef health. This effort had a large emphasis
on stakeholder engagement and consideration of stakeholder concerns during implementation of a
watershed management plan (Bradley et al. 2015).
Lawrence, MA, is a medium-sized community north of Boston and is a Making a Visible
Difference (MVD) Community for ORD. This PBS is a collaboration between ORD's National
Exposure Research Laboratory, National Risk Management Research Laboratory, National Health
and Environmental Effects Research Laboratory, and Region 1 with a focus on improving and
sustaining drinking water quality and reducing human health risks through effective monitoring,
decision making, and planning. The work at this PBS is predefined by stakeholder engagement and
the identification of local priorities and needs. There was a focus on model development and spatial
data gathering, but intended to produce functional results for the community.
Narragansett Bay, RI, is located in Rhode Island and includes the communities of Newport and
Providence. This PBS is a collaborative modeling study of the Bay with respect to EGS production
and covers both environmental, ecological, and economic endpoints. This PBS is a large-scale study
within the Safe and Sustainable Water Research program of ORD focused on the development of a
nitrogen roadmap for decision making on Total Maximum Daily Limits. Stakeholder engagement is
largely focused on research to inform state and federal decision makers.
New Bedford, MA, is a coastal community south of Boston located on Buzzards Bay. This PBS was
a RARE project in EPA Region 1 and a selected Healthy Priorities community. The focus of this
PBS was integrated resource management as a method for the use of household and industrial waste
as a community resource. Selection criteria were used to choose the location of this study including a
formal ranking score sheet. Criteria included willingness to work with EPA, proximity to Region 1
office in Boston, and existence of substantial waste production for study. The work involved
application of existing strategies for waste reuse and reclamation including stakeholder involvement
in making decisions and training in novel techniques.
Southern Willamette Valley, OR, is a coastal river valley entirely within the state of Oregon. The
7
-------
key issue in this PBS is management of groundwater nutrient levels as a part of land use
management. This PBS is a partnership with the Southern Willamette Valley groundwater
management area to improve agricultural nutrient use efficiency through the examination of
management plans and involved a cooperative panel stakeholder engagement effort largely centered
on decision options and information exchange. This community was selected as a PBS because of its
history of working with EPA and strong leadership.
St. Louis River, MN, AOC is a community selected by the EPA under a joint US-Canadian Water
Quality Agreement as an Area of Concern (AOC) for special attention. All of the 43 AOC sites are
impaired in some way with acute problems usually related to sediment contamination. The mouth of
the St. Louis River is adjacent to Duluth, MN; Superior, WI; and the Lake Superior Chippewa
Reservation of the Fond du Lac Band. This AOC has a high need for clean-up of sediment and
restoration of ecosystem goods and services. This PBS is also an archetypal AOC system with many
commonalities to other AOC locations.
Suffolk County, Long Island, NY, was greatly impacted by Tropical Storm Sandy in October of
2012. In the wake of this storm a collaborative effort was initiated to examine how to increase
community resilience to coastal storms in the future. This work involves modeling, community
engagement, and is a collaboration between the state of NY, EPA Region 2, and EPA ORD. Suffolk
County was chosen as a PBS because of previous EPA ties in the community and strong leadership
interest in resilience planning. Tools used in this PBS include Triple value modeling (3VS), Human
Impact Assessment (HIA), and Ecosystem Service Assessment (ESA) all with a focus on
improvement of long-term planning for this highly suburbanized region.
Tampa Bay, FL, is a coastal community on the west coast of Florida. This PBS was chosen as a
regional pilot for study of EGS as a base for management planning in cooperation with the Tampa
Bay Estuary Program. The focus was a study of production and delivery of selected EGS in response
to LULC planning and the development of tools to aid in LULC decision making. This PBS included
model development and strong engagement with the TBEP management community. Public
engagement was not attempted but input was leveraged from previous TBEP activities. This PBS
provides an example of leveraging local planning activities to streamline a project.
Taunton Watershed, MA, is located in southern Massachusetts and includes 42 towns. This PBS is
a part of the Healthy watersheds initiative pilot to examine issues of climate change and urban
development. It was chosen based on strong local leadership and being open to working with EPA.
EPA Region 1 has worked in this watershed for several years in cooperation with researchers in the
EPA ORD - Atlantic Ecology Division (AED, NHEERL) and multiple state and private partners
within the watershed. The goal is the development of a decision support system that accounts for
climate change issues.
Woonasquatucket Watershed, RI, is located in northern Rhode Island and drains into Narragansett
Bay in Providence. The focus of this PBS was the examination of flood control benefits from
wetlands with an emphasis on development of indicators of EGS benefits. This was a model-centric
study that utilized this PBS as a case study site. Stakeholders were largely community leaders who
were informed of the results and trained to use the EGS indicators. The site was chosen for its urban
character, proximity to AED, previous research experience in the watershed, and the willingness of
community leaders to work with EPA (Bousquin et al. 2015).
-------
2223
QD
MapJD Site
1 Duluth, MN
2 Chesapeake Bay, VA
3 Narragansett Bay
4 Tampa Bay, FL
5 Woodbine, IA
6 Birmingham, AL
7 Guanica Bay, Puerto Rico
8 Pensacola, FL
9 Vero Beach, FL
10 Thibodaux, LA
11 Opelousas, LA
12 Pascagoula, MS
13 Lewisville, NC
14 Dania Beach, FL
15 Freeport, NY
16 Suffolk County, Long Island, NY
17 New Bedford, MA
18 Woonasquatucket River Watershed, Rl
19 Windsor locks, CT
20 The Taunton Watershed, MA
21 Lawrence, MA
22 Southern Willamette Valley Groundwater Managemett Area, OR
23 Blue River Watershed, OR
24 Superior, Wl
25 Fond du Lac Band of Lake Superior Chippewa reservation, Wl
25Q)
24
'
4 9
00
21
Copyright: © 2013 National Geographic Society
Figure 1.2 Map showing locations and names for all the place-based studies participating in our information request on the use of an EGS
approach for local decision support. USGS base map produced by National Geographic Society © 2009. This work licensed under ESRI master license
agreement. Map service URL: http://services.arcgisonline.com/ArcGIS/rest/services/NGS_Topo_US_2D/MapServer. See Section 1.2.4 for details on each
of these PBS sites.
-------
1.3 Goals of this Report
The integration of the PEGS approach into decision support at the community level has great promise as
a method for both broadening the goals of decision making beyond the immediate rewards of a
particular choice, and making these broader goals measurable and understandable to community
stakeholders. However, for this approach to be successful it must be accepted locally and to properly
investigate the question of acceptability of our conceptual decision support model, we need a baseline
for the application of ecosystem goods and service concepts to decision making in existing and past
PBS. This report will discuss existing and past placed-based research in ORD with the objective of
describing how this research has (or has not) applied the elements of our EGS-based conceptual model
for decision support. The conceptual model will be revisited in each chapter and forms the basis for
synthesis of lessons learned. Chapters 2 to 5 address specific elements of the conceptual model with the
goal of explaining how these specific elements have been used in practice, why they did what they did,
what was the result, and what could have been the result if that element was not a part of the work. Each
of these contain a similar examination of the PBS under consideration but address specific questions
related to the individual elements of the conceptual model. Chapter 6 contains a synthesis of lessons
learned across all the model elements and also addresses how the model might be used going forward to
develop tools and approaches for community-based decision support (CBDS) focused on sustainability
of ecosystem goods and services.
Photo credit - Nadia Seeteram
10
-------
1.4 Literature Cited
Bradley, P., W. Fisher, B. Dyson, S. Yee, J. Carriger, G. Gambirazzio, J. Bousquin, and E.
Huertas. 2015. Application of a Structured Decision Process for Informing Watershed
Management Options in Guanica Bay, Puerto Rico. U.S. Environmental Protection Agency,
Washington, DC, EPA/600/R-15/248.
Bousquin, J., K. Hychka, and M. Mazzotta. 2015. Benefit Indicators for Flood Regulation Services of
Wetlands: A Modeling Approach. U.S. Environmental Protection Agency, Washington, DC,
EPA/600/R-15/191.
Fulford, R.S., M. Russell, and I.E. Rogers. 2016. Habitat restoration from an ecosystem goods and
services perspective: Application of a spatially explicit individual-based model. Estuaries and
Coasts pp 1-15.
Fulford, R.S., D. Yoskowitz, M. Russell, D.D. Dantin, and J. Rogers. 2015. Habitat and recreational
fishing opportunity in Tampa Bay: Linking ecological and ecosystem services to human
beneficiaries. Ecosystem Services 17:64-74.
Gregory, R., L. Failing, M. Harstone, G. Long, T. McDaniels, and D. Ohlson. 2012. Structured decision
making: A practical guide to environmental management choices. Wiley-Blackwell, London,
UK.
Hopton, M.E. and M.T. Heberling. 2012. Proceedings of Quantifying Sustainability in Puerto Rico: A
Scientific Discussion. U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-
12/723.
Johnston, RJ. and M. Russell. 2011. An operational structure for clarity in ecosystem service values.
Ecological Economics 70:2243 -2249.
Landers, D.H., A.M. Nahlik, and C.R. Rhodes. 2016. The Beneficiary Perspective: Benefits and
Beyond. In: Haines-Young, R., M. Potschin, R. Fish, and R.K. Turner (eds.) Routledge
Handbook of Ecosystem Services. Routledge, Oxon, UK, pp 74-87.
Messer, J., L. Wainger, R. Wolcott, and A. Almeter. 2011. An Optimization Approach to Evaluate the
Role of Ecosystem Services in Chesapeake Bay Restoration Strategies. U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-11/001.
Olander, L., RJ. Johnson, H. Tallis, J. Kagan, L. Maguire, S. Polasky, D.L. Urban, J. Boyd, L.A.
Wainger, and M. Palmer. 2015. Best Practices for Integrating Ecosystem Services into Federal
Decision Making: National Ecosystems Partnership, Duke University doi: 10.13016/M2CH07.
Ringold, P.L., J. Boyd, D. Landers, and M. Weber. 2013. What data should we collect? A framework
for identifying indicators of ecosystem contributions to human well-being. Frontiers in Ecology
and the Environment 11:98-105.
Smith, L.M., J.L. Case, H.M. Smith, L.C. Harwell, and J.K. Summers. 2013. Relating ecoystem
services to domains of human well-being: Foundation for a US index. Ecological Indicators
28:79-90.
11
-------
Summers, J.K., L.M. Smith, J.L. Case, and R.A. Linthurst. 2012. A review of the elements of human
well-being with an emphasis on the contribution of ecosystem services. Ambio 41:327-340.
USDA. 2012. i-Tree Eco User's Manual V.5. Online: USDA Forest Service [accessed 20 May 2016].
Tree Manuals & Workbooks.
Yee, S.H., K. Vache, L. Oliver, and W. Fisher. 2012. Development of Envision, A Spatially Explicit
Framework for Modeling Future Scenarios: Assessing Sustainability of Reef Ecosystem
Services under Water Quality Criteria in St. Croix, USVI. EPA Internal Report.
12
-------
2 Decision Context and Stakeholder
Engagement
2.1 Introduction
Sustainable human well-being is
linked to sustainable use of
environmental resources (MEA 2005;
NRC 2011). Implementation of
decisions targeting sustainable use of
resources will inevitably involve
economic, social, and environmental
tradeoffs. Yet, with the complex and
multi-dimensional problems
communities face, it is often difficult
to anticipate the effects of alternative
decisions on the environment,
economy, and society (Knol et al.
2010). The concept of ecosystem
services, in particular, has generated
much research in recent decades
(Daily 1997; MEA 2005) as an approach to better understand and quantify the benefits people obtain
from the environment. However, ecosystem services are also used to generally advocate for
environmental goals alongside social and economic ones, with little reference to specific decisions,
context of decisions, or stakeholder objectives (Laurans et al. 2013).
Photo courtesy of USEPA
Establishing ecosystem services concepts within a
decision context is a crucial step that helps both
scientists and decision makers (Keeney 1992;
Gregory et al. 2012) to:
• Frame the problem
• Bring clarity to the scope and bounds of
decision making capabilities
• Prioritize information needs
• Focus on and more effectively evaluate the
most relevant potential tradeoffs
• Identify alternatives that are directly
responsive to stakeholders and have a
greater probability of acceptance
Two theories of thought have historically
dominated the way we think about the environment
(Miller et al. 2011). Preservationist theory seeks to
protect nature from use by humans as nature has an
inherent value and right to exist apart from humans,
while the conservationist approach seeks to manage
the environment for the proper use by humans.
While the preservationist approach was often
applied in the early part of the last century to areas
like national parks, wildlife preserves and
endangered species preservation efforts, this
approach soon ran into conflict with the economic
desires of society. An effort to incorporate a more
conservationist approach was adopted where there
might be a more balanced approach to protecting
the environment, while still utilizing the natural
resources in a sustainable manner. Many times the
decisions on what approach to adopt and the
implantation of these approaches were developed
13
-------
from the input of a small select group of professionals based on their expertise or their political standing.
This often led to implemented approaches that had minimal support from the local community and were
subject to frequent criticisms. Additionally, with the relatively narrow range of input into development
of these approaches, the potential for unintended consequences, and thus stakeholder dissatisfaction with
the implemented approaches, was greatly increased. The recognition of these short comings, and the
resulting dissatisfaction, has led to a broad consensus that to implement sustainable solutions the
decision process must also engage stakeholders in all aspects of the process from the starting point of
problem definition to the endpoint of solutions application and implementation (U.S. EPA SAB 2000;
Stahl et al. 2001; NRC 2008).
This chapter reviews methods, tools, and approaches that PBS have used to integrate ecosystem services
into a community decision process and to engage stakeholders in decisions. This review was divided
into five steps with the first step to identify the degree to which studies have used a structured process to
identify decision-relevant science needs or facilitate decision making (Chapter 2.1), and the degree to
which studies involved stakeholders in their process (Chapter 2.3). Second, examples are provided from
the place-based studies of decision contexts for which ecosystem services concepts are directly relevant
(Chapter 2.4). Third, ecosystem services concepts are placed within the larger context of community
social and economic goals (Chapter 2.5). Fourth, the degree to which place-based studies evaluated
changes in ecosystem services under alternative decision scenarios are reviewed, and tools and
approaches used for conducting scenario analysis are described (Chapter 2.6). Finally, a synthesis is
provided of the lessons learned from place-based studies when linking ecosystem services concepts to
community decision making (Chapter 2.7).
Place-based studies were asked what they considered the biggest successes of their study or what they
would have done differently. From survey responses, a number of common themes or lessons emerged,
including: 1) engage stakeholders and local decision makers early in the process; 2) take the time to
formally define the decision context; 3) use conceptual models and systems thinking to uncover
unintended consequences; and 4) work with diverse groups of experts to integrate multidisciplinary
sources of information.
Lessons learned about stakeholder engagement in place-based studies:
1) Engage stakeholders and local decision makers throughout the process
2) Formally define the decision context in cooperation with stakeholders
3) Clarify why natural resources matter
4) Consider unintended consequences of decision options through systems thinking
5) Integrate multi-disciplinary sources of information into the decision process through stakeholder
engagement
14
-------
2.2 Structuring the Decision Process
Environmental decisions are often complex, involve unpopular tradeoffs, have multiple stakeholder
types, and require a variety of information from environmental, economic, and social sciences (Bradley
et al. 2015). A reasonably structured decision process can aid in assembling the most important
information, examining decision alternatives, and exposing likely tradeoffs in a way that maintains a
level of transparency and common understanding (Gregory et al. 2012).
Approximately 80% of reviewed place-based studies (Chapter 1, Chapter 1.2.4) identified using a
structured decision framework to guide their study (Figure 2.1). Only 20% of studies did not use a
structured decision process to guide their research effort, and in those cases did not identify that a
structured process would have been useful.
o>
T3
T3
HI
in
as
CO
50
40
30
Was a structured decision process, such as a defined framework, regulatory
guidance, or workbook used or developed for this place-based study?
S 20
*4-
O
+-*
£ 10
No, but a Yes, we used or Yes, development Yes, we both used
structured process adapted an of a process was a existing and
would have been existing process key research effort developed a
useful process
Figure 2.1 Survey responses from place-based studies as to whether a structured decision process was used
or developed for their study.
The majority of studies used or adapted an existing decision process, but for a number of studies, a key
focus of the research effort was development of a process for integrating ecosystem services concepts
into community decision making. Decision frameworks identified by place-based studies are listed in
Table 2.1.
2.2.1 Decision Analysis Frameworks
A number of place-based studies used, adapted, or developed some variant of decision analysis (Keeney
1982). Decision analysis is a decision process that integrates science and fact-based information with
stakeholder-derived values (Gregory et al. 2006; Failing et al. 2007). The EPA and other federal
agencies have been criticized in the past for relying too heavily on technical-based assessments and not
adequately considering stakeholder values in decisions (Arvai and Gregory 2003; U.S. EPA SAB 2001).
Integrating stakeholder values with scientific information can ensure that future research, data gathering,
and model development better support the decision-making process (Maguire 2003). Decision analysis
15
-------
frameworks vary in the specifics, but most include a common set of steps (Figure 2.2; Gregory 2012)
that are consistent both with the conceptual model introduced in Chapter 1 (Figure 1.1) and with expert
recommendations for sustainability assessments (NRC 2011).
Table 2.1 Examples of structured decision processes or frameworks used or developed by place-based
studies.
Process
Structured Decision Making
Citation
Case Study Location
Bradley et al. 2015; Gregory et Guanica Bay Watershed, PR
al. 2012
Decision Analysis for
Sustainable Economy,
Environment, and Society
(DASEES)
Stockton et al. 2011; Gregory et Dania Beach, FL
al. 2012; Keeney 1992; Jacobs et
al. 2013
Health Impact Assessment
None available
Suffolk County, Long Island, NY
Integrated Resource
Management (based on the
Integrated Design Process)
None available
New Bedford, MA
Integrated Assessment
Framework
None available
Taunton Watershed, MA
Forest Management Planning Cissel et al. 1999
(USFS)
Blue River Watershed, OR
Guidance for development of
narrative criteria for water
quality or biotic integrity
Bradley etal. 2010
Narragansett Bay Watershed, Rl
TMDL regulatory guidance
EPA Chesapeake Bay Program,
2000
Chesapeake Bay Watershed,
MD
Area of Concern Blueprint
Guidance for applying non-
monetary indicators of
ecosystem services benefits
LimnoTech 2013
St. Louis River Estuary, MN and
Wl
Bousquin etal. 2015
Woonasquatucket River
Watershed, Rl
16
-------
Implement, Monitor,
and Review
Monitor and adapt to
changing conditions
Clarify Decision
Context
The who, what,
where of a decision
opportunity
Define Objectives
What is valued in the
decision opportunity,
and how to measure it
Evaluate Trade-offs
and Select
Strategy for achieving
some balance across
objectives
Develop Alternatives
Decision choices to
fulfill the objectives
Estimate
Consequences
Potential outcomes
from decisions on the
objectives
Figure 2.2 The steps in a generic decision process (NRC 2011; Carriger and Benson 2012; Gregory et al. 2012).
The first step of a generic decision analysis framework (Figure 2.2) is understanding the context for the
decision, which will define the problem under consideration and the subsequent analysis of the problem.
The next step requires consideration of what is valuable to stakeholders (i.e., stakeholder objectives) and
identifying metrics to evaluate how potential decision options affect stakeholder objectives. Once
decision options for achieving objectives are identified, analysis is needed to compare the potential
outcomes from decisions and explore trade-offs that affect stakeholders differently depending on what
course of action is chosen. The final step is selecting decision options for achieving objectives that are
consistent with values and preferences of stakeholders. For the process to be effective, all steps should
include consideration of whole systems-thinking, long-term consequences of the decision, and include
stakeholder involvement (NRC 2011). One decision-analytic framework approach used by place-based
studies was Structured Decision Making (SDM) (Gregory et al. 2012), which is an integral part of the
DASEES (Decision Analysis for a Sustainable Environment, Economy, and Society) framework
currently being developed by EPA (Stockton et al. 2011). Other frameworks being used by place-based
studies include Health Impact Assessment (HIA), Integrated Assessment Framework (LAP) and
Integrated Resource Management (IRM).
SDM (Gregory et al. 2012) is a collaborative and often times facilitated process where problems are
shared by multiple stakeholders that have multiple objectives, and goals are tackled in an open format
through group input and deliberation. The primary purpose of SDM is to allow stakeholders to share
their perspectives and objectives regarding the problem with a broader group with the end goal of
17
-------
informing the decision makers as opposed to selecting a particular solution to the problem. A good SDM
approach is one that is inclusive, open, and transparent with all input and outcomes, and provides a
record of such discussions so the information is available if there is a need to refer back to the process in
the future.
DASEES is a web-based decision analysis framework that uses an SDM approach that was developed
by EPA through Neptune and Company, Inc. DASEES uses an accepted conceptual approach for setting
decision context and problem framing (Figure 2.3). The conceptual model used as a foundation for
DASEES is the Drivers-Pressures-States-Impacts-Response (DPSIR) conceptual model (OECD 1994;
UNEP 2007; Vega et al. 2009). DASEES is organized into five steps (Figure 2.3). Though the steps may
differ slightly, this approach is consistent with most traditional generic decision analysis approaches
(Figure 2.2) and SDM. The DASEES SDM process seeks to gather information to address the following
questions:
1. What is the decision context?
2. What are the objectives and what indicators should be used to measure success?
3. What options address the objectives and indicators?
4. How well do options achieve objectives, how do trade-offs affect assessments and what are
the important uncertainties?
5. How can the decision be implemented, tracked and modified as we learn?
Photo courtesy of USEPA
18
-------
Goals &
Objectives
Values &
Preferences
Decision
Makers
Scientific
Input
St
ho
Problem
Formulation
Alternative
Management
or Policy
Options
Adaptive
Management
Objectives
Selected
Management
or Policy Action
Response
Options generate
implementation
costs that have
Impacts
sthere an
Option(s| that
meets the
Objectives with
acceptable
certainty?
Drivers are human needs
that generate Pressures
on the Ecological and
Environmental State
•/ Societal Sustainability
^ Economic Sustainability
^ Environmental Sustainability
•/ Meet environmental regulations
^ Acceptable level of uncertainty
Environmental, Economic,
Societal & Health Impacts
are evaluated against the
Objectives to generate a
Response
Impacts
The change in the
Environmental State
generates Impacts
Pressures generate
stresses that change the
Environmental State
Figure 2.3 Decision support framework for the DASEES tool (Vega et al. 2009).
Other case studies also used processes consistent with a generic decision process (Figure 2.2) but were
tailored toward a particular management goal, such as health-related objectives in a HIA, forest
management, or site restoration from specific pollution (Table 2.1). A number of case studies focused on
very specific regulatory issues related to the development and enforcement of Clean Water Act criteria.
In these cases, the place-based studies generally relied on existing regulatory guidance or regulatory
processes. Other case studies were more focused on elements of the decision process specifically related
to ecosystem services assessments (Olander et al. 2015), such as developing guidance on identifying and
applying indicators of ecosystem services benefits (Bousquin et al. 2015).
2.2.2 The Role of Ecosystem Services in the Decision Process
A generic causal framework for an PEGS analysis was presented in Chapter 1 (Figure 1.1). A structured
decision process can help researchers or decision makers to identify the core elements of that analysis
(Figure 2.2).
A component of clarifying the decision context (Figure 2.4) includes identifying who are the key
beneficiary groups being impacted by the decision under consideration to make sure their concerns,
including what benefits they are hoping to maintain or achieve, are represented in the decision process.
Some stakeholder objectives may be directly represented by PEGS, whereas others may be more
appropriately described by social, economic, or health measures that lead to FEGS. Even when
19
-------
ecosystem services are not the ultimate objectives of stakeholders, they may be a means toward
achieving other objectives and can help bring to light new decision alternatives. For a formal ecosystem
services assessment, researchers can then identify which ecological production functions (EPFs) and
ecosystem benefits functions (EBFs) are needed to link the suite of decision alternatives to measures of
FEGS and other stakeholder-relevant endpoints. Ecosystem services assessments can then be used to
evaluate how different beneficiary groups may be differentially impacted by alternate decisions and a
decision ultimately made. Whether the decision leads to measurable benefits can be monitored over time
so that researchers and decision makers can better understand what levels of ecosystem function are
needed for meaningful change and adapt with future decisions.
What levels of
ecosystem function are
needed for meaningful
change & measurable
benefits?
Are beneficiary
groups being
differentially
impacted?
Implement,
Monitor,
and Review
Evaluate
Trade-offs
and Select
Define
Objectives
Develop
Alternatives
Estimate
Consequences
What benefits do we want?
Which objectives are FEGS?
How do we measure them?
What intermediate
ecosystem services
might be means to
achieving broader
health, social or
economic benefits?
What EPFs or EBFs are
needed to estimate
consequences?
Figure 2.4 Integrating FEGS concepts into a generic decision process.
2.3 Stakeholder and Decision Maker Engagement
All fifteen of the surveyed place-based studies included some degree of stakeholder engagement (Figure
2.5). Almost 60% of place-based studies, in retrospect, wished more had been done with stakeholder
engagement. Specific concerns identified by place-based studies included the general public or specific
stakeholder or beneficiary groups being under-represented in discussions, and a lack of involvement by
local government entities which could have brought greater authority to discussions.
20
-------
Were community stakeholders involved in
this place-based study?
in
re
CO
I
0)
u
_rc
a.
>4-
o
4-^
S
OJ
CL
60
50
40
30
20
10
No
Yes
Yes, but wish more had
been done
Figure 2.5 Survey responses from place-based studies as to whether community stakeholders were
involved.
Place-based studies engaged a number of different kinds of stakeholders, including EPA Regions or
Program Offices, non-EPA decision makers, scientific collaborators or other data experts, and public
stakeholders, including local residents or interest groups (Figure 2.5). The most productive
collaborations with stakeholders tended to be with scientific collaborators or other data experts.
Approximately 85% of place-based studies engaged scientists or data experts outside of ORD, the
majority of which reported satisfaction with the degree of engagement (Figure 2.6). By comparison, 85-
90% of place-based studies also reported engagement with non-scientists (decision makers and the
general public), but in almost 30-40% of cases wished more had been done to engage those groups. In
open-ended survey questions to provide any explanation or additional information, several surveyed
studies noted that involvement of such groups could have brought a greater level of buy-in or reality to
the process.
21
-------
Which of the following community stakeholders were involved in this place-based
study, or in retrospect, do you wish had been more engaged - either at all, or a
different/broader group of stakeholders, or more in-depth engagement?
100
80
CD
60
£ 40
8 20
EPA Regions or Program Non-EPA decision- Scientific collaborators General public, local
Offices makers, or those with or other data experts residents, interest
authority to act outside ORD groups
Figure 2.6 Percentage of place-based studies which engaged or wished they had engaged different types of
stakeholder group. A single study may have engaged more than one type of stakeholder.
In approximately 75% of place-based studies, stakeholders were directly involved throughout the study
(Figure 2.7), often through periodic updates of study progress, and were particularly important for
providing information identified as essential to the study (Figure 2.8). Although 25% of the studies
reported on occasional or indirect involvement, there were no studies in the survey that reported they
conducted their work absent of stakeholder involvement. More than 60% of place-based studies
provided a final report back to stakeholders at the end of the study, and slightly over 50% trained
stakeholders in the use of tools, approaches, or models for decision making. For the studies that did not
involve direct stakeholder involvement throughout, stakeholder engagement most likely involved
reporting results back to stakeholders at the end of the study (Figure 2.8).
22
-------
How were community stakeholders involved in
this place-based study?
0)
^
s
•o
01
re
op
S
a.
t—
o
01
u
^
01
CL
80
70
60
50
40
30
20
10
0
Not involved
Directly involved
throughout
Occasionally or
indirectly involved
Figure 2.7 Percent of place-based studies out of fifteen that identified having direct involvement with
stakeholders throughout the study.
How were stakeholders involved in the study?
D
4J
in
03
CQ
i
u
JS
CL
v+-
o
4-1
c
41
CL
80
70
60
50
40
30
20
10
0
I Directly
involved
I Indirectly
involved
Updated Periodically Provided Essential ORD Reported Final Trained by ORD
Information Results
Figure 2.8 Percent of place-based studies, with either direct or indirect stakeholder involvement, which
identified certain kinds of stakeholder involvement. A single study may have more than one kind of
stakeholder involvement.
23
-------
2.4 Characterizing the Decision Context
2.4.1 Elements of Decision Context
In ecosystem services assessments, clarifying the decision context is critical to bring focus to a problem
and define the scope of information that will be needed. Environmental management problems, in
particular, may cross multiple jurisdictions and time scales, both in the source of impacts as well as who
may ultimately benefit.
Questions to be considered when defining the decision context. The decision context is the
problem, issue, or reason for making a decision, and can be defined by elements such as
(Gregory et al. 2012; Rehr et al. 2014):
• What is the decision to be made?
• Who is making the decision?
• What is are the spatial and temporal scales of the decision?
• What is the general range of alternatives being considered?
• What is the general range of objectives or goals being considered?
• What kind of tools or information will be needed?
• What is the current state of knowledge and uncertainty?
• What kind of decision maker, stakeholder, or expert consultation is needed?
• What are the legal bounds on the decisions and their enforcement?
• What is the history, including any past decisions or existing planning documents?
Ecosystem services assessments are applicable for a variety of decision contexts, as illustrated by the
place-based studies (Table 2.2). Community issues for which ecosystem services assessments were
performed included watershed management, climate change resiliency, development of resource
sustainability plans, water quality regulation, and land use development. Communities defined by
studies ranged from a single city to multi-city counties or watersheds to larger multi-watershed areas.
Decision makers included individual citizens, city/county/state agencies, and federal agencies.
Stakeholder groups included many interests, such as fishermen, tourists, and farmers, who were likely to
be direct beneficiaries of ecosystem services. Objectives from studies were also strongly linked to
ecosystem services, including air quality, water quality and quantity, eco-tourism and recreational
opportunities, and flood protection.
24
-------
Table 2.2 Examples of decision context inferred from 15 place-based study responses.
Study
Location
Guanica Bay,
Puerto Rico
What is the What spatial
issue under area is
consideration? covered by
the decision?
Watershed
management to
protect coral reefs
and ecosystem
services
Single
watershed that
impacts a bay
and offshore
area
Who are the decision
makers and stakeholders
involved or impacted by
the decision?
Fishermen; farmers; boaters;
resource-dependent
businesses; local residents;
power suppliers
(hydroelectric); municipal
water suppliers; recreational
(swimmers, kayakers, etc.);
artists; campers/hikers;
tourism industry; EPA Regions;
other federal agencies; local
government agencies
What is the general range of decision
alternatives being considered?
Implementing agricultural best management
practices; dredging and maintenance of reservoirs;
restoring lagoon and marshes; citizen education;
improve wastewater treatment plant
What is the general range of objectives
under consideration?
Integrity of coral reef and mangrove
ecosystems; integrity of terrestrial
environment; economic opportunities (fishing,
eco-tourism, farming); improve water quality
and quantity
Suffolk County,
Long Island, NY
New Bedford,
MA
Linking decision
alternatives to
changes in health
determinants to
health outcomes
Waste
management and
reuse as resources
Single county
Multiple cities
EPA Region 2; residents and
businesses impacted by
Hurricane Sandy; county
officials
Municipal water
suppliers/users; local
government; landfill managers;
WWTF managers; residents;
recyclers; EPA Regions
Linking health determinants to health outcomes,
including ecosystem services; creating awareness
about environmental/health issues within local
communities
Measured enhancement in human well-being
Promote composting; legislation for waste
management; biosolids management options;
Incinerators that generate electricity; WWTF
upgrade
Improve nutrient capture/reuse; improve
sustainability within the communities; reduce
number of landfills; reduce number of biosolids
25
-------
Study
Location
Tampa Bay, FL
What is the What spatial
issue under area is
consideration? covered by
the decision?
Watershed
management to
protect ecosystem
services
Single
watershed that
impacts a bay
and multiple
overlapping
counties
Who are the decision
makers and stakeholders
involved or impacted by
the decision?
Fishermen; all humans; home
owners/residential;
businesses; famers; timber
industry; municipal water
suppliers/users; EPA Regions
What is the general range of decision
alternatives being considered?
Alternative land-use development scenarios
What is the general range of objectives
under consideration?
Integrity of natural systems and wildlife
corridors; improved water quality; pollution
reduction; maintain cultural heritage
(agriculture); preservation and enhancement
of human well-being; bay restoration;
improved land use planning; improved
respiratory health (air quality); improved
human well-being
Dania Beach, FL
St Louis River
Area of
Concern,
including
Duluth, MN;
Superior, Wl
and the Fond du
Lac Band of Lake
Superior
Chippewa
Reservation
Climate change
resiliency
Regulatory
guidance based on
stakeholder input
Single city
River area of
concern that
impacts
multiple cities
Residents; fishermen; tourism
industry; EPA Regions
Fishermen; recreational
(swimmers, kayakers, etc.);
people who care (existence);
industries; boaters;
transporters of goods/people;
hunters/trappers;
campers/hikers; spiritual
participants; resource-
dependent businesses; EPA
Regions
Redesign and redevelopment of community; use
mangrove/wetland area as polder; raise roadways
to act as dikes; establish series of canals; retrofit
historic buildings
Improve stakeholder understanding and education
about factors affecting the environment and
ecosystem services; examine effects of decision
alternatives
Reduce climate change impacts (flooding,
saltwater intrusion, etc.); reduce nutrient
discharge from failing septic systems; Integrity
of natural resources; maintain economic
stability; Increase energy efficiency; minimize
gentrification; improve human health
Maintain and/or improve ecosystem services;
restoration and remediation of area of
concern; improved environmental/habitat
quality (SAV); improved habitat quality for
endangered/threatened species (bald eagle);
improved awareness of ecosystem services and
environmental condition
26
-------
Study What is the What spatial
Location issue under area is
consideration? covered by
the decision?
Who are the decision
makers and stakeholders
involved or impacted by
the decision?
What is the general range of decision
alternatives being considered?
What is the general range of objectives
under consideration?
Southern Management to Groundwater Farmers; irrigators; water
Willamette improve reservoir area subsisters; municipal water
Valley groundwater of concern suppliers/users; local
Groundwater quality and reduce
Management nitrate leaching
Area, OR
Taunton
from farm fields
Promotion of
Single
government; EPA Regions
EPA Regions; State
Improved monitoring program; education about
the connection between management and
Improved stakeholder engagement; improved
groundwater and drinking water quality;
outcomes; land use/management (agriculture); reduced nutrient input; improved human
improved agricultural best management practices
(BMPs)
Improve stakeholder understanding and education
health and well-being; use of NTT in other
agricultural communities
Connection of human health and well-being
Watershed, MA resiliency of watershed that environmental department; | about factors affecting the environment and with ecosystem services to improve
Blue River
Watershed, OR
communities in impacts municipal officials; non-profit ecosystem services; protect/restore management and decision alternatives; protect
the face of climate multiple cities environmental organizations; upland/riparian habitat; mitigate non-point source and restore water quality; reduce flood risks;
change and
development
Effects of
estuary program; planning and stressors (nutrients); restore floodplains/riparian increase aquatic connectivity for habitat and
Single
economic district; other zones; plan for sea level rise and adaptation; flood resistance; optimize economic
federal agencies; EPA Regions
Tribes; recreational
optimize green infrastructure and BMPs
Simulating how changes in climate and land use
alternative land watershed (swimmers, kayakers, etc.); affect ecosystem services
management
options
Lawrence, MA
Risk and exposure
assessments
boaters; fishermen;
municipality/government
Single city
(flood control); municipal
water suppliers; forest
subsisters; food subsisters; EPA
Regions
Environmental justice (EJ)
communities
Remediation; risk reduction
development and multimodal transportation
Information developed will potentially inform
policy divisions for PNW federal forests;
protection of water resources, habitat,
ecosystem services
Water quality; human well-being; human
health;
27
-------
Study
Location
Pascagoula, MS;
Pensacola, FL;
Vero Beach, FL;
Thibodaux, LA;
Opelousas, LA;
Lewisville, NC;
Woodbine, IA;
Windsor locks,
CT; and
Freeport, NY
What is the What spatial
issue under area is
consideration? covered by
the decision?
Sustainability
planning
Multiple
municipalities
Who are the decision
makers and stakeholders
involved or impacted by
the decision?
Residents, city government
officials, environmental
groups, business and industry,
community leaders; EPA
Regions
What is the general range of decision
alternatives being considered?
Land development; greenspace; community
resilience; flood protection
What is the general range of objectives
under consideration?
Sustainable human well-being as an
understandable and measurable objective
Narragansett
Bay, Rl
Chesapeake
Bay, MD
Birmingham, AL
Excessive nutrient
input in fresh and
estuarine waters
TMDL tradeoffs
regarding
when/how to
control upstream
pollutant sources
Stormwater
Single
watershed that
impacts a bay
Bay and
area
Single city
Residents; environmental
protection departments; EPA
Regions; waste water
treatment facility managers
WWTF managers; residents;
local policymakers; farmers;
EPA Regions
Waste Water Treatment Facility permits for point
source control; developing criteria for nitrogen and
phosphorous; multi-scale, multi-media
interventions to improve water quality
Identify a "socially optimal" mix of control options;
green infrastructure; agricultural BMPs
Site developers; landscape
architects/urban planners;
residents; EPA Regions;
Green infrastructure; low-impact development;
Improved water quality; improve management
connectivity; reduce nutrient input from point
and non-point sources; recover some lost
ecosystem services; reduce occurrence of
hypoxia
Reduce pollution from point sources; reduce
costs to policymakers and stakeholders;
Improve and achieve water quality goals
(TMDL); improve ecosystem services
Reduce urban Stormwater runoff; improve
water quality
28
-------
2.4.2 Visualizing the Decision Problem with Conceptual Models
Conceptual models are invaluable as a means to quickly describe a problem and decision approach.
Conceptual models can provide a clear picture of what is being discussed by the stakeholders, as well as
the means to connect the knowledge of experts, decision makers, and stakeholders in developing a
common understanding of the decision at hand (U.S. EPA 1998; ASTM 2009; Yee et al. 2011).
Conceptual models provide one tool for identifying elements of the decision context, including
preliminary decision alternatives and objectives, information needs, potential outcomes, and key
stakeholder groups (Yee et al. 2014). Modern environmental problems are rarely about singular issues,
and conceptual models can help to capture, visualize, and organize the connections among key factors in
a complex system (Joffe and Mindell 2006; Knol et al. 2010; Yee et al. 2011).
More than 90% of the place-based studies reported using a conceptual model, with the exception of one
ongoing study that is not yet complete (Figure 2.9). A small number of studies were able to leverage an
existing conceptual model developed by the community or outside group. However, the majority of
studies reported developing the conceptual model as an integral part of the study's research effort
(Figure 2.9). In about half of these cases the development and use was an internal effort of the study not
shared with stakeholders. The other half developed and used the conceptual models in direct
collaboration with stakeholders.
Was a conceptual model linking decisions to outcomes
developed for this study?
IS)
0)
s
TO
40
30
20
o 10
o>
No
No, the study used Yes, we developed Yes, we developed
an existing one and used one and used one in
developed by an internally collaboration with
outside group stakeholders
Figure 2.9 Percentage of the fifteen study responses identifying having used or developed a conceptual
model.
Surveyed studies used their conceptual models for a variety of reasons, including to elicit information
from stakeholders, encourage systems-thinking by decision makers, identify areas where research was
needed, and identifying beneficiaries or stakeholders that should be included in future consultations
(Yee et al. 2014; Bradley et al. 2015; Russell et al. 2011). In most cases, the conceptual model itself was
not the goal or the product - it was the process that was important. Place-based studies identified a
variety of approaches and tools used for developing conceptual models (
29
-------
Table 2.3).
Table 2.3. Approaches, tools, and software used by place-based studies to develop conceptual models.
Process
Drivers-Pressu res-States-
Impacts-Response framework
Citation
Case Study Location
Yee et al. 2014; Bradley and Yee Guanica Bay Watershed, PR
2015
Cmap Software
Yee etal. 2011
Guanica Bay Watershed, PR
Pathway diagrams
None available
Suffolk County, Long Island, NY
DASEES
Stockton et al. 2011; Gregory, et Dania Beach, FL
al. 2012; Keeney 1992; Jacobs et
al. 2013
APEX model
USDA
Southern Willamette Valley
Groundwater Management
Area, OR
Nutrient Tracking Tool
USDA
Southern Willamette Valley
Groundwater Management
Area, OR
ES tradeoffs for Alternative
Forest Management Scenarios
(SHC 2.1.4.2 EPA report)
McKane et al. 2014; USEPA/ORD Blue River Watershed, OR
Report ORD-010213
Human Well-Being Index
Smith etal. 2012
9 Gulf Coast Municipalities
Conceptual Framework
Collins etal. 2011
Narragansett Bay Watershed, Rl
AOC Blueprint
None available
St. Louis River Estuary, MN and
Wl
There are a number of approaches to building conceptual models, including influence diagrams, concept
mapping, and pathway diagrams. The underlying premise of all of these approaches is to diagram cause
and effect relationships that are important to the system or decision under consideration. In some place-
based studies, conceptual models were drawn by hand during meetings, or ideas were captured with
notes during meetings and integrated into conceptual models afterwards. Other place-based studies used
available software programs, such as CmapTools (http://cmap.ihmc.us), to capture stakeholder
30
-------
development of conceptual models on the fly during the meeting.
For complex systems, unstructured development of conceptual models can result in key concepts being
missed or overlooked by the developers. Building a conceptual model within a structured conceptual
framework can provide a scaffold and common language for building the conceptual model, providing
clarity and helping guide discussions (Yee et al. 2012). A number of place-based studies identified using
existing conceptual framework to guide stakeholder discussion.
Some of the conceptual frameworks focused very specifically on cause-effect relationships in ecosystem
services assessments (Bousquin et al. 2015). Ecosystem assessment frameworks specifically link
ecosystem structure and function to ecosystem services and their benefits and value to people (Figure
2.10). This approach is consistent with the recommendations of best practices for federal agencies that
environmental assessments should extend to measures that are directly relevant to people (Olander et al.
2015).
Biophysical
structure:
Wetland
characteristics -
size, depth,
location
Function:
Retention or
slowing of water
Ecological
Assessment
Functional
Assessment
Service:
Flood regulation
that reduces risk
of flooding
V
Ecosystem
Service
Assessment
Benefit
Reduction In
damages to
property and
infrastructure
Benefit
Assessment
v*.
Value of lost uses;
avoided costs of
repair or
replacement
)
o
Monetary
Valuation
Figure 2.10 Example conceptual framework for conducting ecosystem services assessments (Bousquin et al.
2015).
Other frameworks looked more broadly at cause and effect relationships between interacting
components of social, economic, and environmental systems, such as by using the Drivers-Pressures-
States-Impacts-Response (DPSIR) framework (Yee et al. 2014; Bradley and Yee 2015). Within DPSIR,
Drivers (D) are social and economic forces leading to human activities that create Pressures (P) on the
State (S) of the environment, and Impact (I) the economic, physical, cultural, and social well-being of
humans through a loss or gain in ecosystem goods and services. Decision makers may enact a Response
(R) to reduce the impacts on environmental resources through regulations, policies, and other decisions,
which may alter Drivers (D) or Pressures (P), or directly affect the State (S) of the ecosystem (Figure
2.3). By comparison, an ecosystem services assessment framework (Figure 2.10), focuses on expanding
the cause and effect relationships between State and Impact in more detail.
31
-------
A primary benefit of conceptual frameworks, like DPSIR, is they guide discussions so that key concepts
are not overlooked. The Human Weil-Being Index (HWBI) (
Table 2.3), for example, defines well-being in terms of how ecosystem, economics, and social services
influence eight domains of well-being: social cohesion, living standards, education, leisure time,
connection to nature, safety and security, health, and cultural fulfillment (Smith et al. 2012). The
structure of the HWBI was used as a framework to encourage community stakeholders to think more
broadly beyond economic goals to characterize what really matters to their community (Fulford et al.
2016).
2.5 Characterizing Stakeholder Objectives and Identifying Decision Options
Objectives are statements of what is important to stakeholders within a particular decision context
(Gregory et al. 2012). They can be used to help decision makers focus on what really matters in terms of
outcomes and, if well-defined by performance measures, can become the evaluation criteria for
comparing outcomes.
Importance of characterizing objectives in a structured decision process. In a structured decision
process (Chapter 2.2.1), characterizing objectives is an important early step because they (Gregory et
al. 2012):
• Help decision makers on what really matters about a decision
• Help experts identify the information needs and uncertainties that really matter to a decision
• Guard against decision alternatives being too narrowly defined, by providing a basis for value-
focused, creative decision options
• Reduce confusion about what different stakeholders mean, if they are well-defined by
performance measures
• Form the foundation for evaluating and comparing decision alternatives
2.5.1 Objectives in Place-Based Studies
All of the place-based studies identified a characterization of stakeholder objectives as part of the study.
For all of the studies, the characterization of stakeholder objectives from documents, discussions, or
formal workshops was an active part of the study's research effort (Figure 2.11). For one study, the
stakeholders had already developed their own goals through an independent process, but the EPA study
conducted an additional workshop to characterize specific ecosystem services of concern. Stakeholder
objectives identified by or inferred from place-based studies were strongly linked to ecosystem services,
including water quality and quantity, recreational opportunities, fishing, and flood regulation, as well as
their potential benefits to economics, human health, and human well-being (Table 2.2).
32
-------
Were stakeholder concerns or goals characterized for the study, such as to
identify focal endpoints for the study?
100
5 80
T3
i
S 60
op
i
u
I 40
•s
1 20
No No, they already Yes, they were Yes, they were Yes, they were
existed or effort was inferred from characterized characterized
led by other group existing documents through informal through a formal
conversations stakeholder
engagement process
Figure 2.11 Percentage of the fifteen place-based studies which included an identification of stakeholder
objectives. A single study may have used more than one method to characterize stakeholder objectives.
Often stakeholder objectives are only implicitly considered in decisions, and decisions are focused on
comparing possible alternatives to achieve a narrow set of objectives (Bradley et al. 2015). Value-
focused decision making, in contrast to alternative-focused decision making, places a much stronger
emphasis on fully considering and explicitly defining objectives before defining alternatives, so that the
overall decision process in more transparent, inclusive, defensible, and grounded in achieving common
values (Keeney 1992).
Many of the place-based studies used principals of structured decision making or values-focused
thinking to guide an elicitation of stakeholder objectives (Table 2.5). Often decision makers will use the
word "objectives" to describe a vision narrative, regulatory targets, sets of principles, or lists of things to
do. In structured decision making, however, objectives are defined in a very specific way (Keeney 1992;
Gregory et al. 2012): objectives are concise statements of what matters for the decision at hand and
should be complete, necessary, unambiguous, sensitive to the decision alternatives under consideration,
and separate what fundamentally matters from means to achieve success.
33
-------
Table 2.4 Example of using HWBI conceptual framework to identify community goals (modified from
Fulfordetal. 2016).
Community Goals
Basic educational knowledge and skills
Positive social and emotional development
More advanced knowledge and skills
HWBI Category
Education
Reasonable life expectancy
Physical and mental well-being
Emotional well-being
Health
Good quality healthcare
Healthy lifestyle and behavior
Enough time devoted to leisure activities
Enough time devoted to physical activity and vacation
Leisure Time
Reasonable time spent working and caring for others
Ability to afford basic necessities
Reasonable income
Reasonable wealth
Living Standards (Economics)
Job stability and satisfaction
Being safe
Feeling (and being) safe
Resilience to hazards
Connectedness to nature
Cultural fulfillment
Healthy family bonding
Supportive network of friends and family
Regular participation in community activities
Responsible engagement in our democracy
Satisfaction with others and the community
Approximately 20% of studies characterized stakeholder objectives through inference from existing
documents or informal conversations with collaborators or stakeholders (Figure 2.11). More than 80%
of studies additionally or instead conducted some form of formal stakeholder engagement process in
order to characterize stakeholder objectives, through interviews, community meetings, or stakeholder
workshops (Table 2.5). Ideally objectives should be directly elicited through discussions with
stakeholders and decision makers, but inferring them from decision maker reports or from
representatives familiar with the issue are also considered acceptable (Parnell et al. 1998).
Software such as CmapTools or DASEES were used by some place-based studies to record discussion
of stakeholder objectives during internal working groups or meetings. DASEES, in particular, aids users
34
-------
in the development of objectives hierarchies. Objectives hierarchies are one tool that place-based studies
have used to structure and define stakeholder objectives. Other studies identified using tools such as
social network analysis to characterize stakeholders. All of these approaches are consistent with the
recommendations of best practices for federal agencies.
For some place-based studies, conceptual frameworks were integral in eliciting or structuring objectives
from stakeholders. As noted with the development of conceptual models (Chapter 2.4.2), frameworks
such as DPSIR (Figure 2.3), HWBI (Error! Reference source not found.), or the Millennium
cosystem Assessment typology of ecosystem services categories (MEA 2005) can help structure
conversations with stakeholders, so that potential objectives are not overlooked.
Photo credit - Nadia Seeteram
35
-------
Table 2.5 Examples of approaches, tools, or software used by surveyed studies to characterize stakeholder
objectives.
Approach, Tool, Software
Structured Decision Making (Values Focused
Thinking)
Objectives hierarchy
Guided discussions using DPSIR framework
Community meetings and stakeholder
workshops
Formal stakeholder identification and
interview process
Formal social network analysis
Cmap
Combination of stakeholder priorities,
economic valuation, and bibliometric
analysis of literature
DASEES
MEA typology to identify ecosystem services
Citation
Bradley et al. 2015; Keeney 1992
Carriger et al. 2013
Yeeetal. 2014
Fulfordetal. 2016
Rehretal. 2014
Snjiders 2001
Harvey et al. 2010; Yee et al. 2011
Russell etal. 2011
Stockton et al. 2011; Gregory et al. 2012; Keeney 1992;
Jacobs etal. 2013
MEA 2005
For more than 50% of studies, a key reason for characterizing stakeholder objectives was to assist the
community in identifying their goals, identifying key modeling endpoints, identifying goal-oriented
decision options and identify key research needs for future work (Figure 2.12).
However, many of the studies also identified additional benefits of helping to define what sustainability
means to the stakeholders, developing indicators and measures of success and informing or training
stakeholders on a concept, process or tool. A couple of studies listed an additional reason for
characterizing stakeholder goals: to help the community identify barriers to achieving their goals,
including issues of legal authority and capacity. These barriers may be a key part of a discussion toward
identifying creative decision options.
While almost half of the studies listed informing or training stakeholders on a concept, process or tool,
the effort of capacity building is often times a very tough goal. Training a group on concepts of a
process are typically easier to do than training on the process itself. Often the time the stakeholders have
36
-------
available for the training process is limited (1-2 days), which makes it difficult to learn the process well
enough to be able to do it on their own. This limited training time is also a big factor that impacts the
ability of the stakeholder to develop the capacity to actually use the tool on their own. Without follow-
up interactions close to the time of the actual training, the success rate for stakeholders to retain the
ability to use the tool or process is diminished.
If stakeholder goals were characterized for the study, why was this done?
-------
Several of the surveyed place-based studies had a strong research focus on helping communities or
environmental managers identify indicators (Bousquin et al. 2015; Fulford et al. 2016) by providing
guidance to help decision makers zero in on measures of highest relevance to the decision at hand by
thinking carefully about stakeholders or beneficiaries and their objectives, the relevance and sensitivity
of different indicators to the decision context, and the comfort level of decision makers with uncertainty.
2.5.3 Using Objectives to Identify Decision Alternatives
Defining objectives can help shift the focus from "What can we do?" to "What do we want to achieve?"
(Gregory et al. 2012). In environmental management, there is often a narrow list of alternatives aimed at
achieving a singular objective, and other stakeholder objectives are only considered indirectly if at all
(Bradley et al. 2015). Values-focused thinking, in contrast, focuses on defining what really matters, to
multiple stakeholders and across multiple objectives in order to come up with creative alternatives to
achieve an end result that the stakeholders understand and value. Through this process all participating
stakeholders can provide their input on things that are of value to them while also hearing and learning
about the things that others value which may not be on their list of values. This process of defining
objectives and performance measures is open and transparent and ultimately provides a rigorous and
transparent way to compare potential outcome alternatives. By conducting this process in an open and
transparent manner, stakeholder expectations are clearly addressed and ultimately managed where
misunderstandings of what the final outcome or decision action may be are essentially nullified.
All surveyed case studies, with the exception of one study still in development, included an
identification of decision options under consideration for the community, the majority of which were
developed as part of the study (Figure 2.13). Approximately 60% of surveyed studies identified that they
characterized stakeholder objectives as a preliminary step to characterizing decision options (Figure
2.12).
In a few studies, a suite of decision alternatives were already under consideration, and the place-based
study was assisting the community in identifying the impacts of decisions on stakeholder objectives.
Were one or more decision options identified for the study?
100
5 80
U1
60
E 40
20
No
No, they already
existed
Yes, they were
inferred
Yes, through
informal
conversations
Yes, through a
formal
engagement
process
Figure 2.13 Percentage of fifteen place-based studies that included an identification of decision options. A
single study may have used more than one method to identify decision options.
38
-------
Some studies used means-ends networks as a way to link decision alternatives to stakeholder objectives
(Carriger et al. 2013). Means-ends networks are diagrams that help to clarify the relationships between
what we fundamentally want to achieve (ends) and how we might achieve it (means) (Figure 2.15).
Ecosystem goods and services by themselves are unlikely to be fundamental goals of stakeholders, but
do play a critical role as a means to achieving what matters to stakeholders (Yee et al. 2014). Structured
tools like means-ends networks, influence diagrams, and conceptual frameworks (e.g., DPSIR; HWBI;
Chapter 2.4.2) can help expose decisions that maximize ecosystem services as creative alternatives
alongside more standard economic or social decisions (Figure 2.14). For other decision contexts,
ecosystem services may be a fundamental objective, or reasonable proxy, in and of themselves.
Means Objectives
Maximize Ecosystem Services
• Air quality regulation
• Food provisioning
• Greenspace
• Water quality regulation
• Water quantity provisioning
• Natural hazard protection
Maximize Social Services
* Social activism
* Education initiatives
• Emergency services
• Family services
Healthcare
• Public works
Maximize Economic Services
• Capital investment
• Employment
• Innovative technologies
• Production
• Consumption
• Redistribution
Fundamental Objectives
Maximize Human Well-being
• Connection to nature
• Cultural fulfillment
• Education
• Human health
• Leisure time
• Living standards
• Safety and security
* Social cohesion
Figure 2.14 Example means-ends network illustrating how ecosystem services, social services, and
economic services are means to achieving a fundamental objective of improving human well-being (adapted
from Smith etal. 2014).
2.6 Estimating Consequences of Alternative Decision Scenarios
In a decision analysis process, once objectives have been defined and decision alternatives identified, the
next step is to evaluate the performance of the different alternatives in achieving the objectives (Gregory
etal. 2012).
Regardless of the source, the goal of information collection is the same - to reduce uncertainties
regarding the consequences of proposed actions on stakeholder objectives. Two specific types of
information, ecological production functions (EPFs) (Chapter 4) and ecosystem benefits functions
(EBFs) (Chapter 5), can be used to link decision alternatives that impact ecosystem condition to changes
in ecosystem services and subsequent benefits to stakeholders.
39
-------
Information to estimate consequences of alternative decision scenarios can come from a number of
sources, including:
• Qualitative assessments through group deliberations
• Targeted field or lab studies to obtain empirical data
• Predictive quantitative models
• Expert judgements
2.6.1 Tools for Estimating Consequences
The development or application of approaches or tools to compare alternative decision scenarios was a
focus of almost all of the surveyed studies (Figure 2.15). Decision scenarios varied widely among the
case studies and illustrate the relevance of ecosystem services assessment to a variety of community and
management issues including: implementation of agricultural best management practices, wastewater
treatment facility permitting, future urban development scenarios, and habitat restoration (Table 2.2).
Were any approaches, software, or tools used to
qualitatively or quantitatively compare alternative
decision options?
M 100
0)
80
5 so
I
0)
u
^ 40
g
I
20
No
Yes
Figure 2.15 Percentage of fifteen place-based studies that included an effort to compare decision options.
Studies used a number of decision analysis tools and approaches for comparing alternative decision
scenarios (
40
-------
Table 2.6). Approaches included qualitative assessments based on literature reviews or expert opinion,
quantification of ecosystem services metrics in terms of production or dollar values, spatial mapping of
ecosystem services metrics under alternative scenarios, and integrative tools that link multiple kinds of
models and information.
41
-------
Table 2.6 Examples of approaches, tools, or software used by surveyed studies to compare alternative
decision scenarios.
TOOl
Qualitative assessments, such as literature
reviews
Quantification of ecosystem services production
or dollar values
Ecosystem services mapping (GIS)
InVEST
EPA H2O
EJSCREEN: Environmental Justice Mapping and
Screen Tool
Bayesian Belief Networks (BBNs)
VELMA: Visualizing Ecosystems for Land
Management Assessments
Envision: Integrated Modeling Platform
3VS: Triple Value Simulation Tool
Consequence tables
DASEES: Decision Analysis for a Sustainable
Environment, Economy, and Society
Citation
Russell etal. 2011
Russell and Greening 2015
Angradi et al. In review; Russell et al. 2012
Smith et al., In review
Russell etal. 2015
U.S. EPA 2015
Bousquin et al. 2014; Rehr et al. 2014
McKane etal. 2014
Bolte 2009; Bolte et al. 2011; Yee et al. 2012b
Industrial Economics, In Review; tenBrink et al. 2016
Gregory and Gonzales 2013; Bradley et al. 2015
Stockton et al. 2011; Bradley et al. 2015
Many of the surveyed studies identified mapping of ecosystem services metrics as a primary way by
which they compared decision scenarios. Mapping of baseline measures of relevant ecosystem services
can help decision makers target appropriate land use changes and support broader, regional assessments
(Swetnam et al. 2011; Grossman et al. 2013). Often assessments focus on one or two targeted ecosystem
services, but an assessment of a broader suite of valued ecosystem services (e.g., Raudsepp-Hearne et al.
2010; Maes et al. 2012; Russell et al. 2013; Arkema et al. 2015; Queiroz et al. 2015) will improve the
ability of decision makers to characterize the starting point and potential range of consequences for
decision alternatives.
42
-------
- o
Figure 2.16 Screenshot from EPA H2O tool illustrating area of interest (dashed black polygon) and stream
connectivity network defining the upstream area of interest (red line). EPA H2O Tool developed for Tampa
Bay, FL area (Russell et al. 2015).
Surveyed place-based studies identified a number of tools for mapping ecosystem services (
43
-------
Table 2.6). The general approach is to identify and parameterize ecological production functions (EPFs)
(Chapter 4) for the location of concern and then implement and map them within a GIS system, such as
ArcMap (ESRI2010). Some studies used available software, such as In VEST (Tallis et al. 2013), or
developed software, such as EPA H2O (Russell et al. 2015), which contain an integrated suite of
ecosystem services mapping tools. Because spatial areas that benefit from ecosystem services may be far
distant from areas that provide them (Syrbe and Walz 2012), EPA H2O explicitly considers upstream
areas that may be connected to the area of interest via hydrological connectivity networks (Figure 2.16).
Maps of ecosystem services production or benefits can be useful, but may reflect only one piece of the
overall decision making objectives. Many of the studies identified other stakeholder objectives,
including safety and well-being. Surveyed studies identified other mapping tools, such as EJSCREEN,
that can specifically help map environmental and demographic metrics related to environmental justice
(U.S. EPA 2015).
Because environmental management decisions are complex, covering multiple stressors and multiple
objectives, estimating the consequences of different scenarios may require integrating a number of
different types of information, data, and models. Surveyed studies identified a number of different
integrative decision support tools (
44
-------
Table 2.6).
Bayesian Belief Networks (BBNs) are particularly helpful in problems characterized by high levels of
uncertainty among experts (Kjaerulff and Madsen 2008) (Figure 2.18). Development of BBNs begin
with a simple conceptual model representing cause and effect relationships between variables, and
subsequently quantify the relationships among the variables using probabilities (Bousquin et al. 2015).
Identifying and quantifying uncertainties can bring insight to a decision, can help prioritize information
needs, can help to explore the risk tolerance of decision makers and can identify where additional
information may be needed (Figure 2.17) (Rehr et al. 2014).
Inland Water
Poor 91 .4
Good 8.60
Quality
TTT
0.086 ± 0.28
V
Lagoon
None 100
Poor 0
Good 0
WQ
i i i
1 ! !
1 1 1
I i i
! i i
Bay & Ocean Water Quality
Good 10.31
Poor 89.7
Coral Reef Health
Poor
Good
78.6
21.4
0.214 ±0.41
Tourism Econ. Gr.
Low
High
79.4
20.6
0.206 ± 0.4
Comm. & Rec. Fishing
Poor
Good
73.6
26.4
0.264 ± 0.44
Figure 2.17 Hypothetical BBN describing probabilities of water quality, reef health, tourism economy, and
fishing outcomes under a proposed lagoon restoration decision (grey box) (modified from Rehr et al. 2014).
In contrast to a BBN which is static in nature, other decision support tools are designed to allow
examination of coupled human and natural environmental systems in a way that allows dynamic changes
over time or spatially-explicit representations. VELMA is an eco-hydrological modeling framework for
assessing trade-offs among ecosystem services in response to alternative land use, climate, or other
changes within a watershed (McKane et al. 2014). Envision is a GIS-based tool for scenario-based
community and regional planning and environmental assessments that employs multi-agent decision
making including actor values, behaviors, and policy intentions (Figure 2.18) (Bolte 2009; Bolte et al.
2011; Bradley et al. 2015). Changes in the landscape are described by natural dynamic processes that
45
-------
can be modified by actor decisions, impacting the supply of ecosystem services. An alternative
integrated assessment tool developed for one surveyed study is 3 VS (Triple Value Simulation) (tenBrink
et al. 2016), which captures the dynamic interrelationships among economic, environmental, and social
systems (Figure 2.19), and allows users to explore the impacts of different decision scenarios on
ecosystem endpoints.
Evaluative Models
Generating landscape metrics
reflecting "stuff people care about",
e.g. ecosystem services, jobs
Actors
Decision-makers managing the
landscape by selecting policies
responsive to their objectives
Landscape
Feedbacks
Multiagent
Decision-making
Select policies and generate
land management decision
affecting landscape pattern
Scenario
Definition
Landscape
Spatialcontainerin which
changes in landscapeor
ecosystem services are depicted
Policies
Fundamental descriptors of
constraints and actions
defining land use management
decision-making
Autonomous Process Mod
Models of landscape change processes
(e.g. hydrology, ecological dynamics)
Figure 2.18 Conceptual framework for Envision spatially-explicit decision support tool.
Econonvy
Society
Flows of water, nutrients, pathogens
via land, groundwater, surface water
Interventions
WWTF treat.
CSOtunnels
LID and Gl
ISOS upgrade
Environment
46
-------
Figure 2.19 Conceptual framework for 3VS integrated assessment tool (tenBrink et al. 2016).
47
-------
2.6.2 Characterizing Tradeoffs
Ultimately, the goal of mapping and predictive modelling is to provide information on how measures
reflecting stakeholder objectives perform under alternative decision scenarios. Consequence tables are
one useful tool that some surveyed studies used for succinctly expressing differences among decision
alternatives in terms of defined objectives (Table 2.7). Explicitly linking alternatives to objectives in a
consequence table can help highlight if important objectives are missing or too vaguely defined, or bring
to light new alternatives that might better achieve objectives (Gregory et al. 2012).
Table 2.7 Hypothetical example of a consequence table (modified from 3VS, Industrial Economics, in review).
Objective
Measure
Maximize water
quality
Maximize beach
recreation
Change in summer
monthly nitrogen
loading relative to
status quo
Alternative 1
Additional Waste
Water Treatment
Facilities
-86,000 kg/month
Alternative 2
Aquaculture
Alternative 3
Low Impact
Development
Total additional
visitors each summer
relative to status quo
13,000 visits
-6,600 kg/month -53,000 kg/month
1,900 visits
6,300 visits
Maximize
property values
Change in $ value of
homes relative to
status quo
$130 million
$2.3 million
$53 million
Maximize eel Change in index of eel 15%
grass habitat grass quality
No change
0.4%
Minimize costs to Annual per capita cost $55/person/year
taxpayers
under 30-year
financing
No cost to the
public
$17/person/year
In a decision analysis approach, unless clear win-win scenarios emerge, decision makers will ultimately
have to explore tradeoffs among the objectives - how much are they willing to sacrifice of one objective
to have more of another (Keeney 1992). Methods such as direct ranking and swing-weighting can be
used to quantify tradeoffs, but ultimately their role should be to provide greater insight into the
deliberation process, not prescribe an optimal solution or approach (Gregory et al. 2012). Ultimately any
solution or approach selected will be based not only on the quantitative and qualitative information
brought to the table, but the perceived value of all the endpoints that are being traded off one for the
other. The DASEES system (
48
-------
617
-------
Table 2.6) also helps decision makers to compare alternatives with a consequence table, and provides
tools for assigning values to different stakeholder objectives. Multi-attribute approaches are also
generally more satisfying than strict cost-benefit methods, which emphasize a comparison of options
expressed in monetary terms (Failing et al. 2007).
2.7 Lessons Learned
The ultimate goal of engaging the place-based studies was to gather information on what they
considered as successful outcomes of their study and based on their efforts and experiences what they
would have done different if they had the opportunity to do work over. From the survey responses a
number of common themes emerged.
2.7.1 Engage Stakeholders and Local Decision Makers Throughout
Surveyed place-based studies universally acknowledged the importance of stakeholder engagement in
their studies, with more than half of the studies wishing more had been done. Many studies relied on and
benefitted from partners in the EPA (e.g., Regions) or community (e.g., regional planning council) to
serve as a liaison to public stakeholder groups. Studies with community partners reluctant to public
stakeholder engagement led to difficulties for the study in meeting research objectives, particularly in
identifying relevant ecosystem services endpoints and defining decision scenarios for analysis and tool
development.
Stakeholder engagement was key to:
• Prioritizing research needs
• Developing realistic applications for
developing and testing tools
• Identifying stakeholder-relevant
endpoints
• Defining decision scenarios
• Interactive development to improve
tool usability
• Finding partners who would carry the
approaches and tools forward after
completion of the study
2.7.2 Formally Define the Decision Context
Working with stakeholder and decision makers early in
the process is essential to more formally define the
decision context and initiate a common path forward.
Environmental management problems can easily
become a list of issues, with different people
understanding the nature of the problem and its
possible solutions very differently (Gregory et al.
2012). Taking the time to formally define the decision
context helps to build a shared understanding of the
key elements of the decision, get buy-in, and establish
a common path forward.
2.7.3 Clarify Why Natural Resources Matter
Working with stakeholders to clarify and define what
really matters in the decision process can help connect
natural resources directly to social and economic
outcomes. Many of the communities in surveyed
studies identified human health and human well-being
as key objectives that stood alongside more commonly assessed ecological or economic endpoints.
Engaging stakeholders and decision makers in discussions can help clarify the fundamental objectives of
stakeholders, help decision makers set and define measures of success, and bring to light the role of
ecosystem services as a means to achieve goals.
2.7.4 Consider Unintended Consequences through Systems Thinking
Conceptual models and systems-thinking were widely identified by survey studies as a successful
50
-------
approach. Benefits of ecosystems to human well-being can be overlooked by individuals, businesses, or
regulatory agencies when making economic and social decisions, often due to an inability to foresee the
full range of intended and unintended consequences (O'Connor and McDermott 1997; Knol et al. 2010;
Hodgson 2011). Conceptual models and systems-thinking frameworks provide a useful tool for
capturing and organizing connections among factors in a complex system (Joffe and Mindell 2006;
Meadows 2008; Knol et al. 2010; Yee et al. 2011). Some studies acknowledged the biggest success of
their project was not a model, map, or tool, but simply getting stakeholders and decision makers to think
about the bigger picture, to consider ecosystem services in their thinking, and raise awareness of
unintended consequences (Bradley et al. 2015).
2.7.5 Integrate Multi-Disciplinary Sources of Information
Ecosystem services is a concept that connects environmental conditions to social and economic benefits
(Wainger and Boyd 2009). Furthermore, for a given decision context, stakeholders may identify
ecological, health, economic, safety, or other objectives as likely to be impacted by decisions. As a
consequence, assessments must often integrate many different types of information from many different
sources in order to provide a meaningful comparison of decision scenarios (Gregory et al. 2012). A
majority of surveyed place-based studies relied on scientific collaborators and experts to provide
essential data and information for their study. Studies also benefitted from cooperation among parallel
modeling efforts: conflicts or agreements among experts help bring to light uncertainties in predicting
decision outcomes and ways in which the quality of information could be improved.
2.8 Conclusions
Decision analysis provides one approach for evaluating tradeoffs in a way that encourages greater public
participation, collaborative decision making, and allows consideration of multiple attributes (Liu et al.
2010). A common problem with the decision making process is that solutions are often pre-ordained by
the perceived issue (e.g., poor water quality) and often involve adapting tools to a problem, rather than
allowing the problem to determine the appropriate set of tools. To make the best use of resources,
communities must broaden the scope of their assessments by applying a more integrated approach. The
place-based studies examined here provide more than a dozen examples of studies that are attempting
such an integrated approach, illustrating tools and approaches with a high potential for transferability to
other communities and relevance to decision makers (Turner et al. 2003). Yet, none of these PBS
demonstrate a fully integrated approach. More PBS are needed that integrate ecological structure and
function, ecosystem services, human welfare, and a full suite of decision options into a single study (Liu
et al. 2010). Developing guidance for how concepts such as PEGS can be integrated into the decision-
making process will assist future studies in providing the kinds of information that are most relevant to
stakeholders. Ultimately this will lead to better decision making that promotes more sustainable
approaches to balancing the gives and takes inherent between economic, environmental and social
aspects in decisions communities face every day.
51
-------
2.9 Literature Cited
Angradi, T.R., D.W. Bolgrien, JJ. Launspach, BJ. Bellinge, M.A. Starry, J.C. Hoffman, M.E. Sierszen,
A.S. Trebitz, and T.P. Hollenhorst. In Review. Mapping ecosystem services of a Great Lakes
estuary can support local decision making.
Arkema, K.K., G.M. Verutes, S.A. Wood, C. Clarke-Samuels, S. Rosado, M. Canto, A. Rosenthal, M.
Ruckelshaus, G. Guannel, J. Toft, J. Paries, J.M. Silver, R. Griffin, and A.D. Guerry. 2015.
Embedding ecosystem services in coastal planning leads to better outcomes for people and
nature. Proceedings of the National Academy of Science 112:7390-7395.
American Society for Testing and Materials (ASTM). 2009. E-1689 Standard guide for developing
conceptual site models for contaminated sites. Annual Book of Standards. ASTM International,
Conshohocken, PA.
Arva, J. and R. Gregory. 2003. Testing alternative decision approaches for identifying cleanup priorities
at contaminated sites. Environmental Science & Technology 37(8): 1469-1476.
Bolte, J. 2009. Envision: A Guide to Application Development. Biological and Ecological Engineering
Department, Oregon State University, Corvallis, OR.
Bolte, J.P., R.B. McKane, D.L. Phillips, N.H. Schumaker, D. White, A. Brookes, and D.M. Olszyk.
2011. In Oregon, the EPA calculates nature's worth now and in the future. Solutions 2:35-41.
Bousquin, J., W. Fisher, J. Carriger, and E. Huertas. 2014. A Bayesian Belief Network Approach to
Explore Alternative Decisions for Sediment Control and Water Storage Capacity at Lago
Lucchetti, Puerto Rico. U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-
14/296.
Bousquin, J., K. Hychka, and M. Mazzotta. 2015. Benefit Indicators for Flood Regulation Services of
Wetlands: A Modeling Approach. U.S. Environmental Protection Agency, Washington, DC,
EPA/600/R-15/191.
Bradley, P., L. Fore, W. Fisher, and W. Davis. 2010. Coral Reef Biological Criteria: Using the Clean
Water Act to Protect a National Treasure. U.S. Environmental Protection Agency, Washington,
DC,EPA/600/R-10/054.
Bradley, P. and S. Yee. 2015. Using the DPSIR Framework to Develop a Conceptual Model: Technical
Support Document. U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-
15/154.
Bradley, P., W. Fisher, B. Dyson, S. Yee, J. Carriger, G. Gambirazzio, J. Bousquin, and E.
Huertas. 2015. Application of a Structured Decision Process for Informing Watershed
Management Options in Guanica Bay, Puerto Rico. U.S. Environmental Protection Agency,
Washington, DC, EPA/600/R-15/248.
Carriger, J.F. and W.H. Benson. 2011. Restoring and managing Gulf of Mexico fisheries: A path
toward creative decision-making. In: Estuaries: Classification, Ecology, and Human Impacts;
Jordan, SJ. (ed.). NOVA Science Publishers, Inc., Hauppauge, NY, pp 291-334.
52
-------
Carriger, J.F., W.S. Fisher, T.B. Stockton, and P.E. Strum. 2013. Advancing the Guanica Bay (Puerto
Rico) Watershed Management Plan. Coastal Management 41(1): 19-38.
Cissel, J.H., FJ. Swanson, and PJ. Weisberg. 1999. Landscape management using historical fire
regimes: Blue River, Oregon. Ecological applications 9(4): 1217-1231.
Collins, S.L., S.R. Carpenter, S.M. Swinton, D.E. Orenstein, D.L. Childers, T.L. Gragson, N.B. Grimm,
J.M. Grove, S.L. Harlan, J.P. Kaye, A.K. Knapp, G.P. Kofinas, JJ. Magnuson, W.H.
McDowell, J.M. Melack, L.A. Ogden, G.P. Robertson, M.D. Smith, and A.C. Whitmer. 2011.
An integrated conceptual framework for long-term social-ecological research. Frontiers in
Ecology and the Environment 9(6):351-357.
Crossman, N.D., B. Burkhard, S. Nedkov, L. Willeman, K. Petz, I. Palomo, E.G. Drakou, B. Martin-
Lopez, T. McPhearson, K. Boynova, R. Alkemade, B. Egoh, M.B. Dunbar, and J. Maes. 2013.
A blueprint for mapping and modeling ecosystem services. Ecosystem Services 4:4-14.
Daily, G.C. (ed.). 1997. Nature's Services. Societal Dependence on Natural Ecosystems. Island Press,
Washington, DC.
Failing, L., R. Gregory, and M. Harstone. 2007. Integrating science and local knowledge in
environmental risk management: A decision-focused approach. Ecological Economics 64:47-
60.
Fulford, R., D. Yoskowitz, M. Russell, D. Dantin, and J. Rogers. 2016. Habitat and recreational fishing
opportunity in Tampa Bay: Linking ecological and ecosystem services to human beneficiaries.
Ecosystem Services 17:64-74.
Gregory, R., L. Failing, D. Ohlson, and T.L. McDaniels. 2006. Some pitfalls of an overemphasis on
science in environmental risk management decisions. Journal of Risk Research 9(7):717-735.
Gregory, R., L. Failing, M. Harstone, G. Long, T. McDaniels, and D. Ohlson. 2012. Structured
Decision Making: A Practical Guide to Environmental Management Choices. Wiley-Blackwell,
West Sussex, UK.
Gregory, R. and J. Gonzalez. 2013. Protecting the Guanica Bay Watershed: A Public Values Forum.
Guanica Bay Public Values Forum, Final Workshop Report. Value Scope Research, Vancouver,
Canada.
Harvey, J., M. Russell, D. Dantin, J. Nestlerode, H. Greening, and S. Cooper. 2010. Integrated
Approaches to Estuarine Use and Protection: Tampa Bay Ecosystem Services Case Study. In:
Estuaries: Types, Movement Patterns, and Climatical Impacts. Nova Publishers, Hauppauge,
NY.
Industrial Economics, Inc. (In review) Narragansett Bay 3 VS Report: Model Summary.
Jacobs, S., B. Dyson, W.D. Shuster, and T. Stockton. 2013. A structured decision approach for
integrating and analyzing community perspectives in re-use planning of vacant properties in
Cleveland, Ohio. Cities and the Environment (GATE) 6(1): Article 11. [accessed 7 September
2016]. Available at: Digital Commons.
53
-------
Joffe, M. and J. Mindell. 2006. Complex causal process diagrams for analyzing the health impacts of
policy interventions. American Journal of Public Health 96:473-479.
Keeney, R.L. 1982. Decision analysis: An overview. Operations Research 30(5):803-838
Keeney, R.L. 1992. Value-Focused Thinking: A Path to Creative Decision Making. Harvard University
Press, Cambridge, MA.
Kjaerulff, U. and A. Madsen. 2008. Bayesian Networks and Influence Diagrams: A Guide to
Construction and Analysis. Springer, New York, NY.
Knol, A.B., DJ. Briggs, and E. Lebret. 2010. Assessment of complex environmental health problems:
Framing the structures and structuring the frameworks. Science of the Total Environment
408:2785-2794.
Laurans, Y., A. Rankovic, R. Bille, R. Pirard, and L. Mermet. 2013. Use of ecosystem services
economic valuation for decision making: Questioning a literature blindspot. Journal of
Environmental Management 119:208-219.
LimnoTech. 2013. St. Louis River Area of Concern Implementation Framework: Roadmap to Delisting
(Remedial Action Plan Update), [accessed 7 September 2016]. PC A State PDF.
Liu, S., R. Costanza, S. Farber, and A. Troy. 2010. Valuing ecosystem services e theory, practice, and
the need for a transdisciplinary synthesis. Annals of the New York Academy of Sciences
1185:54-78.
Maes, J., B. Egoh, L. Willemen, C. Liquete, P. Vihervaara, J.P. Schagner, B. Grizzetti, E.G. Drakou, A.
La Notte, G. Zulian, F. Bouraoui, M.L. Paracchini, L. Braat, and G. Bidoglio. 2012. Mapping
ecosystem services for policy support and decision making in the European Union. Ecosystem
Services 1:31-39.
Maguire, L.A. 2003. Interplay of science and stakeholder values in Neuse River total maximum daily
load process. Journal of Water Resources Planning and Management 129:261-270.
McKane, R.B., A. Brookes, K. Djang, M. Stieglitz, A.G. Abdelnour, F. Pan, JJ. Halama, P.B. Pettus,
and D.L. Phillips. 2014. VELMA Version 2.0 User Manual and Technical Documentation. U.S.
Environmental Protection Agency, Office of Research and Development.
Meadows, D.H. 2008. Thinking in Systems: A Primer. Earthscan, London.
Millennium Ecosystem Assessment (MEA). 2005. Ecosystems and Human Well-being: Current State
and Trends, Volume 1. Island Press, Washington, DC.
Miller, T.R., B.A. Minteer, and L-C. Malan. 2011. The new conservation debate: The view from
practical ethics. Biological Conservation 144:948-957.
National Research Council (NRC). 2008. Public Participation in Environmental Assessment and
Decision Making. The National Academies Press, Washington, DC.
National Research Council (NRC). 2011. Sustainability and the U.S. EPA. National Academies Press,
54
-------
Washington, DC.
Olander, L., RJ. Johnston, H. Tallis, J. Kagan, L. Maguire, S. Polasky, D. Urban, J. Boyd, L. Wainger,
and M. Palmer. 2015. Best Practices for Integrating Ecosystem Services into Federal Decision
Making. National Ecosystem Services Partnership, Duke University, Durham, NC, 54 p.
Organization for Economic Co-operation and Development (OECD). 1994. Environmental Indicators -
OECD Core Set. OECD, Paris, France.
Parnell, G. S., H. W. Conley, J. A. Jackson, L. J. Lehmkuhl, and J. M. Andrew. 1998. Foundations
2025: A value model for evaluating future air and space forces. Management Science 44:1336-
1350.
Queiroz, C., M. Meacham, K. Richter, A.V. Norstrom, E. Andersson, J. Norberg, and G. Peterson.
2015. Mapping bundles of ecosystem services reveals distinct types of multifunctionality within
a Swedish landscape. Ambio 44(Suppl. 1):S89-S101.
Raudsepp-Hearne, C., G.D. Peterson, and E.M. Bennett. 2010. Ecosystem service bundles for analyzing
tradeoffs in diverse landscapes. Proceedings of the National Academy of Sciences 107:5242-
5247.
Rehr, A.P., M.J. Small, P.S. Fischbeck, P. Bradley, and W.S. Fisher. 2014. The role of scientific studies
in building consensus in environmental decision making: A coral reef example. Environmental
Systems and Decisions 34:60-87.
Russell, M. and H. Greening. 2015. Estimating benefits in a recovering estuary: Tampa Bay, Florida.
Estuaries and Coasts 38:9-18.
Russell, M., A. Teague, F. Alvarez, D. Dantin, M. Osland, J. Harvey, J. Nestlerode, Johne Rogers, L.
Jackson, D. Pilant, F. Genthner, M. Lewis, A. Spivak, M. Harwell, and A. Neale. 2013.
Neighborhood Scale Quantification of Ecosystem Goods and Services. U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-13/295.
Russell, M., J. Harvey, P. Ranade, and K. Murphy. 2015. EPA H2O User Manual. U.S. Environmental
Protection Agency, Office of Research and Development, Washington, DC, EPA/600/R-15/090
Russell, M., J. Rogers, S.J. Jordan, D. Dantin, J. Harvey, J. Nestlerode, and F. Alvarez. 2011.
Prioritization of ecosystem services research: Tampa Bay demonstration project. Journal of
Coastal Conservation 15:647-658. doi:10.1007/s!1852-011-0158-z.
Smith, L., C. Wade, J. Case, L. Harwell, K. Straub, and K. Summers. 2014. Indicators and Methods for
Constructing a U.S. Tribal Well-being Index for Sustainable and Healthy Communities
Research. U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-12/023, Supl.
Smith, L., L. Harwell, K. Summers, H. Smith, C. Wade, K. Straub, and J. Case. 2014. A U.S. Human
Well-being Index (HWBI) for Multiple Scales: Linking Services Provisioning to Human Weil-
Being Endpoints (2000-2010). U.S. Environmental Protection Agency, Washington, DC,
EPA/600/R-14/223.
Snijders, T.A.B. 2001. The statistical evaluation of social network dynamics. In: Sobel,
55
-------
M. E. and M.P. Becker (eds.). Sociological Methodology 31:361-395. Basil Blackwell, Boston,
MA and London.
Stahl, R.G., R.A. Bachman, A.L. Barton, J.R. Clark, P.L. deFur, SJ. Ells, C.A. Pittinger, M.W. Slimak,
and R.W. Wentsel (eds.) 2001. Risk Management: Ecological Risk-Based Decision-Making.
SET AC Press, Pensacola, FL, 202 p.
Stockton, T., B. Dyson, W. Houghteling, K. Black, M. Buchholtz ten Brink, T.J. Canfield, A.M. Vega,
M. Small, and A. Rehr. 2012. Decision Support Framework Implementation of the Web-Based
Environmental Decision Analysis DASEES: Decision Analysis For A Sustainable Environment,
Economy, And Society. U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-
12/008.
Swetnam, R.D., B. Fisher, B.P. Mbilinyi, P.K.T. Munishi, S. Willcock, T. Ricketts, S. Mwakalila, A.
Balmford, N.D. Burgess, A.R. Marshall, and S.L. Lewis. 2011. Mapping socio-economic
scenarios of land cover change: A GIS method to enable ecosystem service modelling. Journal
of Environmental Management 92:5 63 -5 74.
Syrbe, R.U. and U. Walz. 2012. Spatial indicators for the assessment of ecosystem services: Providing,
benefiting and connecting areas and landscape metrics. Ecological Indicators 21:80-88.
Tallis, H., T. Ricketts, A.D. Guerry, S.A. Wood, R. Sharp, E. Nelson, D. Ennaanay, S. Wolny, N.
Olwero, K. Vigerstol, D. Pennington, G. Mendoza, J. Aukema, J. Foster, J. Forrest, D.
Cameron, K. Arkema, E. Lonsdorf, C. Kennedy, G. Verutes, C.K. Kim, G. Guannel, M.
Papenfus, J. Toft, M. Marsik, J. Bernhardt, R. Griffin, K. Glowinski, N. Chaumont, A.
Perelman, M. Lacayo, L. Mandle, R. Griffin, P. Hamel, and R. Chaplin-Kramer. 2013. In VEST
3.0.0 User's Guide. The Natural Capital Project, Stanford.
tenBrink, M., G. Foley, J. Hunter, H. Walker, and R. Eric. 2016. NARRAGANSETT-3VS MODEL
OVERVIEW (Information Sheet). U.S. Environmental Protection Agency, Office of Research
and Development, Washington, DC, EPA/600/F-16/060.
Turner, R.K., J. Paavola, P. Cooper, S. Farber, V. Jessamy, S. Gerogiou. 2003. Valuing nature: Lessons
learned and future research directions. Ecological Economics 46:493-510.
United Nations Environmental Programme (UNEP). 2007. GEO4: Global Environment Outlook -
Environment for Development. Nairobi.
http://www.unep.org/roap/Portals/96/REEO_AP_Content_Intro.pdf.
U.S. Environmental Protection Agency (USEPA). 1998. Guidelines for Ecological Risk Assessment.
Washington, DC, EPA/630/R-95/002F.
U.S. Environmental Protection Agency Science Advisory Board (USEPA SAB). 2000. Toward
Integrated Environmental Decision-Making. Washington, DC, EPA-SAB-EC-00-001.
U.S. Environmental Protection Agency Science Advisory Board (USEPA SAB). 2001. Improved
Science-Based Environmental Stakeholder Processes. Washington, DC, EPA-SAB-EC-COM-
01-006.
U.S. Environmental Protection Agency (EPA). 2015. EJSCREEN Technical Documentation.
56
-------
Vega, A.M., T. Canfield, B. Faulkner, P. Bradley, H.L. Fredrickson, A. Rehr, M. Small, T. Stockton,
M. tenBrink, V.E. Hansen, A.N. Pilant, K. Black, D.S. Burden, and W.S. Fisher. 2009. Decision
Support Framework (DSF) Team Research Implementation Plan. U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-09/104.
Wainger, L.A. and J.W. Boyd. 2009. Valuing Ecosystem Services. In: McLeod, K. and H. Leslie (eds).
Ecosystem-based Management for the Oceans. Island Press, Washington, DC, pp 92-111.
Yee, S.H., I.E. Rogers, J. Harvey, W. Fisher, M. Russell, and P. Bradley. 2011. Concept Mapping
Ecosystem Goods and Services. In: Applied Concept Mapping: Capturing, Analyzing, and
Organizing Knowledge, Moon, B.M., R.R. Hoffman, J.D. Novak, and AJ. Canals (eds.). CRC
Press, Boca Raton, FL, pp 193-214.
Yee, S.H., P. Bradley, W.S. Fisher, S.D. Perreault, J. Quackenboss, E.D. Johnson, J. Bousquin and P.A.
Murphy. 2012. Integrating human health and environmental health into the DPSIR framework:
A tool to identify research opportunities for sustainable and healthy communities. Ecohealth
9(4):4 H-426.
Yee, S.H., J. Carriger, W.S. Fisher, P. Bradley, and B. Dyson. 2014. Developing scientific information
to support decisions for sustainable reef ecosystem services. Ecological Economics 115: 39-50.
[accessed 7 September 2016]. Science Direct.
Yee, S.H., K. Vache, L. Oliver and W. Fisher. 2012b. Development of Envision, a Spatially Explicit
Framework for Modeling Future Scenarios: Assessing Sustainability of Reef Ecosystem
Services Under Water Quality Criteria in St. Croix, USVI. EPA Internal Report.
57
-------
3 Final Ecosystem Goods and Services
fish & wildlife
Photos courtesy of USEPA
3.1 Background
The goal of this chapter is to examine the use of final ecosystem goods and services (PEGS) as a tool for
decision making at the community level. First PEGS will be defined and place in context and then the
identification and measurement of PEGS will be examined in the place-based studies examined in this
report.
3.1.1 Definition
A key to collaboration between natural and social scientists is the identification and measurement of
indicators of final ecosystem goods and services. Indicators measure what directly affects people's
welfare. Figuring out what affects the welfare of community stakeholders is not as straightforward as it
may seem. There is no universally accepted definition of welfare. However, we can apply certain
principles in our search for such outcomes.1
1 Much of this background section has been taken verbatim or nearly verbatim from (Boyd et al. 2015)
58
-------
One important principle is to describe
Lessons learned from the identification and use of PEGS in ecological outcomes (and their indicators)
place-based studies:
1) Non-use values of ecosystem goods and services
rarely considered in place-based studies
in terms of an ecological production
framework. Placing ecological services in
a production framework is useful because
it allows us to differentiate between drivers
of change in services (inputs) and change
2) Endpoints used in place-based studies are often in the goods and services (outputs), and to
intermediate ecosystem goods and services rather distinguish outputs based on the degree to
than final ecosystem goods and services which they are fmal yersus intermediate
I steps in the pathway to a given human
3 Methods for identifying beneficiaries in place-based ,.,_., ^ , ,
,.,,.,, ,, , benetit. Final outcomes are those closer—
studies are highly variable including a focus on non- .. . , _ „.
, , , . . causally—to a human benefit of interest.
human beneficiaries _, , . . .
For example, nitrogen concentrations in
4) The PEGS classification system may need to be water affect the concentration of algae,
refined to better represent beneficiary groups at whlch affects the concentration of oxygen
the community level m tne water, which affects the number and
kind offish that can live in it. Maintenance
of nitrogen levels, algae concentration, and
oxygen concentration are intermediate services contributing to the final good: fish abundance.
Intermediate goods and services are of great significance not only in manipulating ecosystems to obtain
different levels of PEGS, they are also important in understanding and assessing PEGS in the first
instance.
A second principle is to identify and measure environmental changes that matter as directly as possible
to human welfare. What does it mean to directly matter? An outcome directly matters when it is
understood and valued by stakeholders as an end in itself (Johnston and Russell 2011). An outcome
indirectly matters when it affects, or is a necessary input to, a subsequent outcome that affects welfare.
For example, fish abundance directly matters to anglers. Oxygen concentrations in water, necessary for
fish abundance, indirectly matter to anglers.
Scientists working on ecosystem goods and services have adopted the labels of final versus intermediate
ecological goods and services (e.g., Daniels and Hensher 2000; Boyd 2007; Boyd and Banzhaf 2007;
Fisher et al. 2008; Fisher and Turner 2008; Kontogianni et al. 2010; Johnston and Russell 2011; Ringold
et al. 2013). The distinction between intermediate and final goods comes from economics, where there is
a desire to describe the economy as a complex balance between production and consumption patterns.
Much like an ecological system, an economic system can be viewed as a set of inputs and outputs linked
by production relationships, which can be distinguished based on their perceived importance to
consumers. For instance, bulk iron ore does not directly matter to consumer welfare, but is a necessary
and valuable input to products like cars and washing machines that do directly matter to consumer
welfare. Thus iron ore is an intermediate good, whereas cars and washing machines are final goods. On
this point, we have referred generically to stakeholders and the outcomes that matter to them, which will
be labelled as beneficiaries of ecosystem goods and services. But beneficiaries are a diverse group in
terms of their environmental preferences and the ways in which nature contributes to their welfare.
Beneficiaries also exhibit vast differences in how they value nature based on the ways they interact with
and experience natural systems. Accordingly, indicators of final ecosystem goods and services can and
should be tailored to specific kinds of human beneficiaries.
Efforts to define specific metrics and indicators of final ecosystem goods and services requires that
59
-------
ecosystem goods and services be linked to specific beneficiaries. These efforts started with a series of
workshops to solicit expert opinion (Ringold et al. 2009; Ringold et al. 2011) on specific ecosystems:
streams, wetlands and estuaries. Findings from those interdisciplinary workshops reveal that indicators
for many beneficiaries can be complex and measure multiple things. For example, a recreational angler
may be concerned with fish abundance and desirability, but also the appeal of the place where fishing
occurs (Hunt 2005). Again, in economics this complex method of measuring consumer goods, is well-
known (Green 1990). However, FEGS indicators should be designed with an eye towards the ways in
which people value a natural resource rather than to measure a distinct feature of a natural system such
as "integrity". Ringold and his colleagues (Ringold et al. 2013) illustrate how metrics should be
developed and combined into indicators - See Figure 1 and the six steps listed in that report. In addition,
qualitative research, such as stakeholder surveys, can make a substantial contribution to refining the
scientific data and expert opinions typically used for FEGS indicator development (Weber and Ringold
2012; Weber and Ringold 2015). Thus, development of FEGS indicators requires the identification of
beneficiaries, and the combination of ecological and social science understanding, especially when using
expert knowledge or qualitative research.
Another way to think about FEGS is to describe ecosystem goods and services (and their indicators) in
the context of a linked ecological and social conceptual map, such as the one described in Chapter 1
(Figure 1.1). A concept map allows experts and stakeholders alike to easily see the connections between
decisions, environmental change resulting from those decisions, and changes in human benefit (e.g.,
restore fish habitat, abundance of species, and satisfaction level of anglers), as well as to articulate the
relationship of the outcomes to human experience. Another core principle driving the FEGS approach is
the importance of environmental outcomes to human welfare. Consider the following example regarding
the importance of clean water: nitrogen concentrations in water affect the concentration of algae, which
affects the concentration of oxygen in the water, which affects the number and kind offish that can live
in it. Nitrogen levels are one feature of a system that scientists study to determine levels offish
abundance. Algae and oxygen concentrations are, from the perspective of an angler, intermediate goods
and so less obvious to human experience than the fish they support: It's impossible for someone to sense
nitrogen levels. And people may or may not directly understand or care about oxygen levels; however,
for those people interested in fishing, fish abundance and presence is very important and easy to
observe. In other words, anglers understand that oxygen and nitrogen levels are of great importance
because they affect fish distribution and abundance, not because anglers have a direct interest in them. A
concept map that links intermediate and final ecosystem goods and services can greatly aid in
connecting the intermediate processes of the ecosystem into the final ecosystem good or service of
interest, which, in the example, turns out to be presence and abundance offish rather than nitrogen
levels, oxygen levels, or even algae. The concept map helps link the intermediate processes in the
ecological system (i.e., thing we have to manage) to expressions of stakeholder value and preferences
(i.e., thing they care about). An outcome directly matters when it is valued as an end in itself (Johnston
and Russell 2011).
3.1.2 Classification Systems of Beneficiaries
Specification of indicators of FEGS starts with the specification of the beneficiary (Ringold et al. 2013;
Landers 2015; Landers et al. 2016). Thus analysis of FEGS is assisted by a standardized list of
beneficiaries. Economists have recognized discrete but broad categories of the ways in which people
benefit from ecosystems for decades (e.g., Bishop et al. 1987; Randall 1987). These systems of Total
Economic Value (TEV) were represented in the Millennium Ecosystem Assessment and elsewhere
(Figure 3.1) (e.g., UNEP 2005; Turner et al. 2008). The expert workshops on FEGS mentioned in the
previous chapter used a heuristic approach allowing diverse beneficiaries to identify themselves rather
60
-------
starting from a standardized list of beneficiaries. Landers and Nahlik (2013) then used the outcome of
these workshops to develop a new standardized list. This analysis merges the previously existing TEV
categories (Figure 3.1) with the workshop-based Landers and Nahlik list as the starting point for
identification of PEGS beneficiaries. This merger is represented in a simplified form in Appendix B.
Note that in this merger that the new PEGS beneficiary categories do not fit cleanly under a single TEV
category.
TOTAL ECONOMIC VALUE
USE VALUE NON-USE VALUE
DIRECT-USE INDIRECT-USE OPTION EXSISTENCE
AGRICULTURE COMMERCIAL/INDUSTRIAL OPTION OF INTRINSIC VALUE
COMMERCIAL/INDUSTRIAL GOVERNMENT/MUNICIPAL/RESIDENTIAL FUTURE USE EXISTENCE
GOVERNMENT/MUNICIPAL/RESIDENTIAL COMMERCIAL/MILITARY TRANSPORTATION
SUBSISTENCE RECREATIONAL
RECREATIONAL INSPIRATIONAL
INSPIRATIONAL LEARNING
LEARNING
Figure 3.1. Total Economic Value categories integrated with FEGS-CS categories of value. Tuner et al.
(2008) note that metrics and indicators tend to be more tangible for beneficiaries on the left of this chart than for
those further to the right. This figure is expanded further in Appendix B.
3.2 Methods
The approach of this chapter is to describe how select place-based studies used ecosystem goods and
services concepts—specifically PEGS concepts—in their work. To this end, we developed a series of
questions (Appendix A) that were embedded in an informal information request distributed to people
who had managed place based studies in the past. The place-based projects that responded to this
information request are described in the report introduction. These responses served as the foundation
for our analysis and conclusions.
3.3 Results
The raw results are compiled in Appendix B; and they are summarized in Table 3.1. Of special interest
is the categorization in endpoints listed as Intermediate Ecosystem Goods and Services (LEGS), PEGS,
Economic Goods and Services or unclear. Categorizations are initial because while elements of PEGS
were present in many of the PBS, the PEGS concept as a whole was not available for use when many of
these projects were started. Our goal then is to find PEGS elements, which are often difficult to define.
Beneficiaries associated with endpoints were not listed for any PBS except Tampa Bay, PL. In addition,
the units of measure were often not provided, so it could not be determined if the endpoint listed was a
biophysical measure or not. For example, what does "snorkeling opportunity" mean? Is it a place where
people happen to go snorkeling or is it a place that has biophysical attributes that make it an appealing
location for snorkeling? If it is the former, it would be a social measure, the latter could serve as an
PEGS metric. Similarly, is "hunting benefit" measured in biophysical units (in which case it might be a
an PEGS) or in social units, e.g. number of days of hunting per hectare per year (in which case it would
61
-------
be an economic or social measure dependent upon an PEGS and a number of other features)? Usable
answers were provided for only four of the survey responses. Of the responses provided, 26 were for
suspected LEGS metrics or indicators, 15 for possible PEGS metrics or indicators, nine for an economic
good or service and three for uncertain (?).
On the basis of inspecting these results we identified five conclusions:
1) Only one case study claimed to look at existence values and none looked at option values (Figure
3.1).
2) Endpoints listed were often intermediate measures, or single attribute measures.
3) Methods for identifying beneficiaries were highly variable.
4) Some beneficiaries identified aren't human, e.g., bald eagles.
5) There are some questions for the refinement of the FEGS-CS system as a result of examining this
survey.
Our overall conclusion is that place-based study practitioners need to collect additional data for a more
detailed implementation of the PEGS approach. Whether this requires additional clarification or
guidance or other support is not clear from the questions asked and the responses provided.
Table 3.1 Initial categorization of responses to the question: "What ecological endpoints did you use [in your
study] relative to your beneficiary list?" into Intermediate or Final Ecosystem Goods and Services as well as
other categories.
Case Study
Nitrogen
Endpoint
Nitrate leaching from farm fields
Initial
Classification
IEGS
Chesapeake Bay
Brook trout habitat
IEGS
Chesapeake Bay
Flood risk mitigation
IEGS
Tampa Bay
Carbon sequestration and/or storage for global population IEGS
affected by climate change
Tampa Bay
Tampa Bay
Denitrification rates for people paying waste water
treatment fees
IEGS
Flood water retention using the TR-55 curve number
method for home or business owners downstream
IEGS
Tampa Bay
Local or neighborhood scale removal of atmospheric
pollutants for people breathing
IEGS
Tampa Bay
Pollinator index based on habitat suitability of a proto-
pollinator and location adjacent to crop production for
farmers needing pollinators
IEGS
62
-------
Case Study
Tampa Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Chesapeake Bay
Tampa Bay
Tampa Bay
Tampa Bay
Endpoint
Storm water retention for people getting water from
groundwater sources
% wave energy attenuation by reef
Air pollutant removal
Carbon sequestration
Curio production
Denitrification
Forest habitat
Life expectancy of reservoir and origin of sediment load
Nitrogen fixation
Nutrient retention
Rates of erosion
Retained rainwater
Sediment loading
Sediment retention
Tree canopy for shading
Water yield
Wave energy attenuation
Wave runup
Air quality
Fish of catchable size present based on HSI for 4 targeted
species for recreational anglers fishing in the estuary
Number of species for global population interested in
biodiversity
Presence of mature tree (80 m2 canopy) on property
and/or line of site to water body for Home owners viewing
Initial
Classification
IEGS
IEGS
IEGS
IEGS
IEGS
IEGS
IEGS
IEGS
IEGS
IEGS
IEGS
IEGS
IEGS
IEGS
IEGS
IEGS
IEGS
IEGS
FEGS
FEGS
FEGS
FEGS
63
-------
Case Study
Tampa Bay
Tampa Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Chesapeake Bay
Tampa Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Guanica Bay
Endpoint
green and bluescapes
Presence of park for Home owners in walking distance to
parks
Shading by canopy cover present in the southern direction
from each property center for home owners paying
heating and cooling utility bills
Classes of reef integrity (excellent, good, fair, poor)
Commercial fish biomass
Density of key commercial harvest species
Farmland quality
Potential for pharmaceutical discovery
Reef integrity
Richness of charismatic or endangered species
Hunting benefits
Timber available to be cut for timber producers
Ease of access for recreation
Poor/good commercial and recreation fishing
Probabilities of low/high tourism economic growth
Probabilities of research improving uncertainties in
outcomes
Rates of visitation to dive sites
Value of a dive site
Value of finfish production
Dive site favorability
Quality of a dive/snorkeling site
Snorkeling opportunity
Initial
Classification
PEGS
PEGS
PEGS
PEGS
PEGS
PEGS
PEGS
PEGS
PEGS
Economic G or S
Economic G or S
Economic G or S
Economic G or S
Economic G or S
Economic G or S
Economic G or S
Economic G or S
Economic G or S
?
?
?
64
-------
3.3.1 Critical Evaluation
3.3.1.1 Existence and Option Values Identified in Only One Study
Despite the importance of existence and option values (Arrow et al. 1993; Attfield 1998; Hanemann
2006; Heal et al. 2004; Mendelsohn and Olmstead 2009), only one study identified this benefit category.
This is surprising, because despite the complexity of assigning a weight to these values, most experts
acknowledge the importance of non-use values (Figure 3.1). Confusion about how to account for
existence values in cost benefit analysis for environmental decision making is not new (McConnell
1997; Randall and Stoll 1983). Existence values are: the worth people assign to aspects of the
environment for spiritual, philosophical, or other reasons that don't depend on directly experiencing
those systems (Dietz 2015). Existence values are a non-use value, and related non-use beneficiaries
value resources even though they have no intention or expectation of interacting with the resource
(Freeman et al. 2014). Option values, though not mentioned in any of the place-based studies, can be
used in decision-making processes where future uses of a service are taken into account in methods of
valuation (Pinto et al. 2013). Options values, like existence values, are often difficult to quantify.
With regards to the placed-based studies, two of the projects alluded to existence value as part of their
evaluation (i.e., "people who care," and "anyone who has concerns about the health of the Chesapeake
Bay").
3.3.1.2 Endpoints Listed are often Intermediate Measures, or Single Attribute Measures rather
than Indicators of PEGS.
Place-based studies are using ecosystem goods and services language in their analysis, referring to
endpoints related to ecological production. However, they are more often using examples of
intermediate goods or services rather than PEGS endpoints. To be of use in social analyses, indicators of
intermediate ecological goods and services must be translated to the final ecosystem goods or services
that directly benefit people. This is illustrated in the conceptual model described in Chapter 1 (Figure
1.1). For example, commercial fish biomass, and potential for pharmaceutical discovery could both be
considered as PEGS, and linked to specific beneficiaries such as commercial fishers, and pharmaceutical
companies in an analysis using ecosystem goods and services and the conceptual model (Figure 1.1).
Other examples of possible FEGS used in the place-based studies include the richness of charismatic or
endangered species for those who care (plausibly linked to existence values - although not stated as such
explicitly in response to our questions). The pollinator index could also be considered a measure of the
FEGS: presence of pollinators (ecosystem service) for pollinating the crops of a farmer (beneficiary).
In contrast to the multi-attribute character of many FEGS noted above, many of the endpoints described
are for single attributes, and while some (e.g., farmland quality) may have multiple attributes, it was not
clear from the information provided if they were viewed that way. Is "farmland quality", for example, an
indicator which composites soil quality, water availability, and water quality or is it based on a single
attribute? As a measurable endpoint this may not be important, but as a component of human welfare it
requires more detail.
A shift to an FEGS methodology requires us to be more explicit about distinguishing between
intermediate processes of ecological systems and final ecosystem goods and services, and it requires
focusing on both beneficiaries and translating ecosystem production function endpoints into biophysical
features directly relevant to human experience and values.
65
-------
3.3.1.3 Methods for Identifying Beneficiaries are Variable and do not Appear to Provide a
Complete Listing
Appendix B captures how variable the approaches are in terms of identifying beneficiaries. These range
from stakeholder engagement to focus groups, workshops, interviews, content analysis and extracting
beneficiaries from the FEGS-CS system. No matter the method, the list of beneficiaries is sparse.
Appendix B summarizes responses from 16 place-based studies. These are placed in a table which lists
46 FEGS-CS beneficiaries, thus there are 690 potential entries. Ten of the studies listed no beneficiaries.
Of the remaining six studies the number of beneficiaries listed ranged from three to 14 with an average
of 7.7 beneficiaries listed. Perhaps these were the dominant forms of economic activity in their study
area, and by focusing on these beneficiaries they were able to capture a large fraction of the economic
activity in their study region. In the absence of information on this quality, one can only speculate as to
whether the identification of additional beneficiaries, by whatever means effective, would identify
additional material value or additional equity issues. In any event, the use of a checklist, at any of the
categorical levels shown in Appendix B might be a useful way to ensure that the listing of beneficiaries
is substantially or, in accounting terms, materially, complete.
3.3.1.4 Some Beneficiaries Listed Aren't Human Beneficiaries
One misconception that a study lead might have is "What constitutes a beneficiary?" For example, in
one of the studies a non-human entity—an eagle—was counted as a beneficiary. While changes in
ecosystems might benefit eagle populations, it would be better, depending on the beneficiary (e.g. for a
bird watcher), to count the change in eagle populations as a FEGS rather than as a beneficiary.
3.3.1.5 Potential for Refinement of the FEGS-CS System
Development of the questions for information request and analysis of the responses yielded questions
that may be germane to the refinement of the FEGS-CS as a community decision making tool. First, in
developing Figure 3.1 (and also evident in Appendix B) we found that the FEGS-CS categories did not
have unique links to the TEV categories. For example, FEGS-CS Categories 3. Government and
Municipal, 6. Recreational and 8. Learning could fall in both the direct and indirect use categories.
Whether this is an issue for any or all uses of the classification system is a question to be addressed by
the FEGS-CS team. In addition, examination of the beneficiaries listed in Appendix B show that they are
generally more finely resolved than the finest FEGS-CS categories, i.e., the beneficiaries listed are one
member of the beneficiary category listed by the FEGS-CS system. In addition, in several cases multiple
beneficiaries were listed within single FEGS-CS categories. The question this raises for the FEGS-CS
team is by what criteria is it decided whether there are too many or too few beneficiaries listed in the
FEGS-CS system?
3.4 Conclusion
Given these findings, especially, the lack of clarity about beneficiaries, the lack of discrimination
between FEGS metrics or indices rather than any other class of metric or indicator, and the lack of a
consistent capacity to link specific FEGS with specific beneficiaries, it would seem that additional
attention needs to be paid to communicating these steps and FEGS concepts in general to place-based
studies. The pieces are there but the connection between EGS and human welfare inherent to the FEGS
concept are not completely fleshed out. This is not surprising as the FEGS concept was not a priori a
part of these studies and the progress made in applying EGS concepts in general is very promising.
Stakeholder engagement is a key area for progress in stakeholder understanding and utilization of FEGS
both in goal setting and in identification of indicators for monitoring of progress. Working to identify
benefits and tie them directly to beneficiaries is also discussed in Chapter 5 and is an important gap in
66
-------
stakeholder understanding and acceptance of the link between ecosystem goods and services and human
beneficiaries.
List of questions included in the LLR Survey that served as the foundation for the PEGS chapter of this:
• What beneficiaries (e.g., anglers, irrigators) of ecosystem goods and services did you identify in your
study? Please list all.
• Briefly describe the process or method you used to identify these beneficiaries. If you did not use a
process or method for identifying, please note N/A.
• What ecological endpoints did you use relative to your beneficiary list (e.g., "species
presence and abundance" for "anglers")? Please be specific about the endpoint or provide
citations that describe them (e.g. "wetland habitat" is the number of acres of vegetated
wetland using the definition in xxx). If you did not identify the relationship between ecological
endpoints and beneficiaries, please note N/A.
• What are the temporal and spatial dimensions of your endpoints (e.g., long-term average
biomass of commercial fish in a grab sample taken at a single point once a year)?
• What was your rationale and foundation for linking the indicator used to the beneficiary
considered (e.g., qualitative research, literature review, or expert judgment)?
Photo credit - Nadia Seeteram
67
-------
3.5 Literature Cited
Arrow, K., R. Solow, P.R. Portney, E.E. Learner, R. Radner, and H. Schuman. 1993. Report of the
NOAA Panel on Contingent Valuation.
Attfield, R. 1998. Existence value and intrinsic value. Ecological Economics 24:163-168.
Bishop, R.C., KJ. Boyle, and M.P. Welsh. 1987. Toward total economic valuation of Great Lakes
fishery resources. Transactions of the American Fisheries Society 116:339-345.
Boyd, J. 2007. Nonmarket benefits of nature: What should be counted in green GDP? Ecological
Economics 61:716-723.
Boyd, J. and S. Banzhaf. 2007. What are ecosystem services? The need for standardized environmental
accounting units. Ecological Economics 63:616-626.
Boyd, J.W., P.L. Ringold, AJ. Krupnick, RJ. Johnston, M. Weber, and K. Hall. 2015. Ecosystem
Services Indicators: Improving the Linkage between Biophysical and Economic Analyses. RFF
DP 15-40, Resources for the Future, Washington, DC.
Daniels, R.F. and D.A. Hensher. 2000. Valuation of environmental impacts of transport projects: The
challenge of self-interest proximity. Journal of Transport Economics and Policy pp 189-214.
Dietz, T. 2015. Environmental Value. In: Handbook of Value: Perspectives from Economics,
Neuroscience, Philosophy, Psychology and Sociology, 329 p.
Fisher, B., K. Turner, M. Zylstra, R. Brouwer, R. de Groot, S. Farber, P. Ferraro, R. Green, D. Hadley,
J. Harlow, P. Jefferiss, C. Kirkby, P. Morling, S. Mowatt, R. Naidoo, J. Paavola, B. Strassburg,
D. Yu, and A. Balmford. 2008. Ecosystem services and economic theory: Integration for policy-
relevant research. Ecological Applications 18:2050-2067.
Fisher, B. and R.K. Turner. 2008. Ecosystem services: Classification for valuation. Biological
Conservation 141:1167-1169.
Fisher, B., R.K. Turner, and P. Morling. 2009. Defining and classifying ecosystem services for decision
making. Ecological Economics 68:643-653.
Freeman HI, A.M., J.A. Herriges, and C.L. Kling. 2014. The Measurement of Environmental and
Resource Values: Theory and Methods. Routledge, New York, NY.
Green, P.E. and V. Srinivasan. 1990. Conjoint analysis in marketing: New developments with
implications for research and practice. Journal of Marketing pp 3-19.
Hanemann, M. 2006. The Economic Conception of Water. In: Rogers, P.P., M.R. Llamas, and L.
Martinez-Cortina (eds.). Water Crisis: Myth or Reality? Marcelino Botin Water Forum 2004.
Taylor and Francis, London, 332 p.
Heal, G.M., E.B. Barbier, KJ. Boyle, A.P. Covich, S.P. Gloss, C.H. Hershner, J.P. Hoehn, S. Polasky,
K. Segerson, and K. Shrader-Frechette. 2004. Valuing Ecosystem Services: Toward Better
Environmental Decision-Making. National Academy of Science, Washington, DC.
68
-------
Hunt, L. M. 2005. Recreational fishing site choice models: Insights and future opportunities. Human
Dimensions of Wildlife 10:153-172.
Johnston, RJ. and M. Russell. 2011. An operational structure for clarity in ecosystem service values.
Ecological Economics 70:2243 -2249.
Kontogianni, A., G.W. Luck, and M. Skourtos. 2010. Valuing ecosystem services on the basis of
service-providing units: A potential approach to address the 'endpoint problem' and improve
stated preference methods. Ecological Economics 69:1479-1487.
Landers, D. 2015. National Ecosystem Services Classification System (NESCS): Framework Design
and Policy Application. U.S. Environmental Protection Agency, Washington, DC, EPA/800/R-
15/002.
Landers, D., A. Nahlik, and C.R. Rhodes. 2016. The Beneficiary Perspective - Benefits and Beyond.
In: Potchin, M., R. Haines-Young, R. Fish, and K. Turner (eds.). Routledge Handbook of
Ecosystem Services. Routledge, New York, NY, pp 74-87.
Landers, D. and A. Nahlik. 2013. Final Ecosystem Goods and Services Classification System (FEGS-
CS). U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-13/ORD-004914.
Lopez-Gonzalez, A. A., A. Aguilo, M. Frontera, M. Bennasar-Veny, I. Campos, T. Vicente-Herrero, M.
Tomas-Salva, J. De Pedro-Gomez, and P. Tauler. 2014. Effectiveness of the heart age tool for
improving modifiable cardiovascular risk factors in a Southern European population: A
randomized trial. European Journal of Preventive Cardiology.
McConnell, K.E. 1997. Does altruism undermine existence value? Journal of Environmental
Economics and Management 32:22-37.
Mendelsohn, R. and S. Olmstead. 2009. The economic valuation of environmental amenities and
disamenities: Methods and applications. Annual Review of Environment and Resources 34:325-
347.
Newell, R.G. and J.V. Siikamaki. 2013. Nudging Energy Efficiency Behavior: The Role of Information
Labels. RFF DP 13-17, National Bureau of Economic Research.
Pinto, R., V. de Jonge, J. Neto, T. Domingos, J. Marques, and J. Patricio. 2013. Towards a DPSIR
driven integration of ecological value, water uses and ecosystem services for estuarine systems.
Ocean & Coastal Management 72:64-79.
Randall, A. 1987. Total economic value as a basis for policy. Transactions of the American Fisheries
Society 116:325-335.
Randall, A. and J.R. Stoll. 1983. Existence value in a total valuation framework. Managing Air Quality
and Scenic Resources at National Parks and Wilderness Areas pp 265-274.
Ringold, P., J. Boyd, D. Landers, and M. Weber. 2013. What data should we collect? A framework for
identifying indicators of ecosystem contributions to human well-being. Frontiers in Ecology
and the Environment 11:98-105.
69
-------
Ringold, P.L., J. Boyd, D. Landers, and M. Weber. 2009. Report from the Workshop on Indicators of
Final Ecosystem Services for Streams. U.S. Environmental Protection Agency, Corvallis, OR.
Ringold, P.L., A.M. Nahlik, J. Boyd, and D. Bernard. 2011. Report from the Workshop on Indicators of
Final Ecosystem Goods and Services for Wetlands and Estuaries. U.S. Environmental
Protection Agency, EPA/600/X-11/014.
Turner, R.K., S. Georgiou, and B. Fisher. 2008. Valuing Ecosystem Services: The Case of Multi-
functional Wetlands. Hardcover edition. Routledge, New York, NY.
UNEP. 2005. Concepts of Ecosystem Value and Valuation Approaches. Ecosystems and Human Well-
being: A Framework for Assessment.
Weber, M. and P. Ringold. 2012. Ecosystems and People: Qualitative Insights. Environmental Health
and Society Bulletin pp 2-8.
Weber, M.A. and P.L. Ringold. 2015. Priority river metrics for residents of an urbanized arid
watershed. Landscape and Urban Planning 133:37-52.
70
-------
4 Ecological Production Functions
4.1 Definition and Purpose of EPFs
Steady progress is being made in the definition of
ecosystem goods and services (Boyd and Banzhaf
2007; Fisher et al. 2009; Johnston and Russell 2011;
Munns et al. 2015) in their classification (EEA 2016;
Landers and Nahlik 2013; Landers 2015) and in their
valuation (Costanza et al. 2014; National Research
Council 2004). There is also growing recognition of
the potential value of ecosystem services to decision-
making processes (Daily et al. 2009; de Groot et al.
2010; Executive Office of the President 2015; Maes
etal. 2012).
The problem of quantifying the production of
ecosystem goods and services, especially in
relationship to human impacts on or management of
ecosystems, has also received attention. Accounting
approaches are sometimes used that ascertain
quantities or values of ecosystem services as a
function of ecosystem areas (Costanza et al. 2014; de Groot et al. 2012). These approaches are useful for
estimating the ecosystem-service impacts of management actions that change land-use/land cover
(LULC), but they are insufficient for estimating impacts of management actions that impose other kinds
of changes. When the environmental management decisions that communities face entail changes in
water or air pollutant delivery, species and habitat abundance, or other ecological processes, predictive
modeling approaches may be required for decision support.
Photo courtesy of USEPA
Lessons learned from use of ecological production functions in place-based studies:
1) Place-based studies most frequently used multiple EPFs in a coordinated fashion (via linkage to
one another or execution within decision-support tools), so as to estimate multiple services
2) Many of the EPFs used may be considered transferable to other locations because they have
been transferred in the past and/or they use widely available data
3) Many EPFs that estimate FEGS are simple models (e.g., of 5 or fewer input variables), but these
may require linkage to more complex models for simulation of management alternatives
4) Evaluation and communication of uncertainty must become more systematic
5) Methods are needed for evaluating model suitability - especially alignment to decision context
71
-------
Photo courtesy of USEPA
As outlined in the introduction to this report and as further
explored in Chapter 3, the quantification of final ecosystem
goods and services (PEGS) can play an important role in
informing place-based decisions (Ringold et al. 2013).
Ecological production functions (EPFs) are necessary to the
use of PEGS, as illustrated in the overall conceptual model
(Figure 1.1) Ecological production functions have been
defined as "usable expressions (i.e., models) of the processes
by which ecosystems produce ecosystem services, often
including external influences on those processes" (Bruins et
al. 2016). The most useful EPFs are those which estimate
final ecosystem goods and services (Boyd and Banzhaf 2007;
Nahlik et al. 2012) since the latter are amenable to benefit
estimation. Within a given a decision context, then, the
purpose of using one or more EPFs is to estimate the
influence of decision alternatives on one or more final
ecosystem service endpoints.
By the definition of EPF given above, any model that
estimates one or more endpoints useful for ecosystem
service estimation would be considered an EPF.
However, it is useful to distinguish decision support
systems (DSSs), which link multiple EPFs, from
individual EPFs; this distinction is illustrated in Figure
4.1 and can be explained using examples of EPFs and
DSSs from EPA place-based studies (PBS). That is,
some models focus on one primary ecological process
or relationship and produce endpoints narrowly related
to that process (Figure 4.1 A) - an example would be a
model of the relationship between fish habitat and
fishery populations applied in Tampa Bay, FL (Fulford
et al. 2016b). Others combine multiple ecological
processes as necessary to represent the dynamic
interactions between them - for example, a model of
hydrologic dynamics and plant growth processes
(SWAT) applied in the Guanica Bay, PR Watershed
(Hu and Yuan 2013). These models can estimate
multiple services produced by these inter-related
processes (Figure 4. IB). In several cases, a primary EPF that is capable of connecting ecological
responses directly to management actions is used to provide input to secondary EPFs that relate model
outputs to endpoints of interest, often PEGS; this situation is portrayed in Figure 4.1C&D. Decision
support systems, or DSS, often serve as computational platforms for the estimation of multiple
ecosystem services by linking multiple EPFs. In static DSS implementations (Figure 4.1C), scenarios are
represented by fixed values (e.g., the EPA H2O tool used in Tampa Bay, FL); more sophisticated DSS
(Figure 4. ID) enable feedbacks that can be used to simulate management changes over time (e.g.,
Envision, used in Guanica Bay, PR) or that enable optimization of management alternatives, such as an
economic optimization model used to demonstrate the design of nutrient control policies in Chesapeake
Photo courtesy of USEPA
72
-------
Bay (U.S. EPA 2011; Wainger 2012; Wainger et al. 2013). When, as is often the case, EGS production
is computed in a spatially explicit manner, geospatial mapping of service production - or of inter-
scenario changes in service provision (e.g., Russell et al. 2015; Angradi et al. 2016) - is a useful feature
ofmanyDSS.
A. Single-process EPF
B. Inter-related-process EPF
r__________,
EPF
C. DSS with a primary EPF driving secondary EPFs
D. Dynamic DSS with primary and secondary EPFs
EPF
EPF
EPF
Figure 4.1. Examples of relationships between ecological productions functions (EPFs) and decision
support systems (DSSs) in the modeling of ecological processes.
4.2 Uses of EPFs in EPA Place-Based Studies
This lessons learned effort sought to understand the characteristics and degree of success of EPF uses in
place-based studies (PBSs) recently or currently being conducted by U.S. EPA. Besides PBS leaders'
written responses to a request for information, the following resources were consulted:
• documented and published details of model development or application at the PBS site (e.g., as
reports or journal articles), provided by PBS leaders
• unpublished documentation of model application that was provided in draft form
• other information about the model (such as a user's manual or published description of another
application) obtained from the open literature (this was done when PBS leaders identified a
model that was in use or would be used in the future, but did not provide documentation).
Given these limitations, best efforts were made to summarize model use, or intended use, in each case
(Table 4.1). Full details are listed in Appendix C. The analysis presented in this chapter evaluates these
EPFs according to several characteristics, including: their linkages within each case study;
computational approaches used; numbers of variables; linkage to decision context; representation of
uncertainty; and transferability. The chapter concludes with a brief set of observations on what can be
learned about EPF use in place-based studies.
73
-------
Table 4.1. Overview of place-based studies, ecological production functions or decision support systems used, and documentation of models or
modeling (see Appendix C for details). Citations in italics apply to the EPFs or DSSs in question, but not specifically to the PBS location.
Place-based Overview of ecological production functions (EPFs)
Study (PBS) and decision support systems (DSSs) used or intended
Location for use
Guanica Bay
Watershed, PR
• Dynamic DSS (Envision) provides change
scenarios to a watershed process model (Soil
and Water Assessment Tool, SWAT) and coral
reef process model (Coral Reef Scenario
Evaluation Tool, CORSET)
• SWAT results drive 9 secondary EPFs
• CORSET results drive 28 secondary EPFs
Status of
modeling
Modeling
completed
Status of modeling
documentation
Case study modeling
documented by
published journal articles
and unpublished journal
manuscripts; previous
model applications
documented by
published journal articles
Available
document
citations
Bradley et al.
(2014); Bradley et
al. (2016);
Orlando and Yee
(in review);
Bousquin et al.
(2014); Rehr et
al. (2014); Smith
et al. (in review);
Yee et al. (in
revision); Yee et
al. (2014); Yuan
etal. (2012)
Suffolk County,
Long Island, NY
Modeling is intended, but models have not yet been Modeling not
identified started
Modeling documentation
not available/examined
for this study
None available
New Bedford,
MA
Insufficient information was provided to determine
models used
Progress unclear
based on
information
examined
Modeling documentation
not available/examined
for this study
None available
74
-------
Place-based Overview of ecological production functions (EPFs)
Study (PBS) and decision support systems (DSSs) used or intended
Location for use
Tampa Bay
Watershed, FL
• Watershed process modeling: Static DSS (EPA
H2O model) used to link land use/land cover
scenarios to: stormwater detention model
(composite curve number); air pollutant
removal by vegetation (UFORE = i-Tree Eco);
and a summation of presence of 20 different
ecological features of interest
• Bay process modeling: tidal exchange model
linked to water clarity model
• Fish habitat valuation: fishery value
apportioned to habitat areas for recreational
and commercial species
• Mangrove development: empirical model of
mangrove ecosystem development over time
Status of
modeling
Modeling
completed
Status of modeling
documentation
Case study modeling
documented by
published journal articles,
unpublished journal
manuscripts and a model
website
Available
document
citations
Fulford et al.
(2016a); Jordan
et al. (2012);
Osland et al.
(2012); Russell et
al. (2015);
Reistetter and
Russell (2011);
Rogers et al. (in
prep); USDA
Forest Service
(2012)
Dania Beach, FL
A decision support framework tool (Decision Analysis
for a Sustainable Environment, Economy and Society or
DASEES) was used for identifying objectives, examining
management alternatives and determining their
consequences, but information on procedures for
consequence modeling was not available
Progress unclear
based on
information
examined
Modeling documentation
not available/examined
for this study
None available
75
-------
St. Louis River
Estuary, MN
andWI
Place-based Overview of ecological production functions (EPFs)
Study (PBS) and decision support systems (DSSs) used or intended
Location for use
26 simple, logic-based EPFs were developed that were
capable of evaluating impacts of projected bathymetry
changes as a result of sediment remediation and
restoration in the study area. Some of these required
input from other process models (i.e., denitrification,
fish or bird species habitat suitability, prediction of
submerged aquatic vegetation)
Status of
modeling
Modeling
completed
Status of modeling
documentation
Case study modeling is
documented in an
unpublished journal
manuscript
Available
document
citations
Angradi et al.
(2013; 2016);
Bellinger et al.
(2014)
Southern
Willamette
Valley, OR
Groundwater
Management
Area
Modeling is using the field-scale, soil-process and plant-
growth model APEX (Agriculture Policy Extender) to
examine water quality consequences of implementing
agricultural best management practices; study may also
use the Nutrient Tracking Tool, an online front-end for
APEX
Modeling in
process
Documentation of case
study modeling is not
available. Model websites
and user manuals
available
Texas A&M
Agrilife Research
(2016)
Taunton
Watershed, MA
Blue River
Watershed, OR
Decision support tool is under development to
integrate the computation of multiple watershed
ecosystem services in a geospatial platform;
information was not available on specific modeling
tools being used
Progress unclear
based on
information
examined
Case study modeling is
not documented
The relatively complex ecohydrology model VELMA
(Visualizing Ecosystem Land Management Assessments)
was used to model hydrology and forest growth in
response to climate change scenarios
Modeling
completed
Case study modeling is
documented in published
journal articles
None available
Abdelnour et al.
(2011, 2013);
USEPA(2016)
76
-------
Lawrence, MA
Place-based Overview of ecological production functions (EPFs)
Study (PBS) and decision support systems (DSSs) used or intended
Location for use
Case study will use:
• EJSCREEN tool to map environmental and
demographic indicators
• WTP-ccam (Water Treatment Plant - climate
change adaptation model) to evaluate climate-
change impacts on drinking water treatment
• A nowcasting model to predict bacterial water
quality
Status of
modeling
Progress unclear
based on
information
examined
Status of modeling
documentation
Case study modeling is
not documented; a
website providing and
documenting EJSCREEN is
available
Available
document
citations
USE PA (2015)
9 Gulf Coast The core objective was to describe and rank
communities stakeholder priorities and map them to domains of the
Human Well-Being Index; modeling to evaluate
management alternatives was not undertaken
Modeling was not
planned
Not applicable
Not applicable
Narragansett
Bay Watershed,
Rl
Narragansett 3VS is a relatively simple model of
nitrogen loading and dispersal processes in
Narragansett Bay, and management actions
EFDC (Environmental Fluid Dynamics Code)
models estuarine circulation
WASP (Water Quality Analysis Simulation
Program) uses EFDC outputs to model water
quality in the Bay
Some modeling
completed; some
still underway.
Some difficulty is
reported in
modeling tidal
circulation and
achieving realistic
calibration for
water quality
modeling
Case study modeling is
documented in internal
reports or is as-yet
undocumented;
documentation of several
models is available
Krumholz et al.
(2015); USEPA
(2007; 2013)
77
-------
Place-based Overview of ecological production functions (EPFs)
Study (PBS) and decision support systems (DSSs) used or intended
Location for use
EcoGEM (Gross Exchange Matrix) models Bay
water quality under conditions of climate
change
EcoOBM (Officer Box Model) models water-
sediment interactions in the Bay and their
impact on water quality
Status of
modeling
Chesapeake Bay
Watershed,
parts of 6 states
Chesapeake Bay Program's Phase 5.3 Community
Watershed Model (CBWM) was used in an optimization
framework to evaluate the trade-offs (including
ecosystem services) of different point- and nonpoint-
source control projects/policies
Modeling
completed
Status of modeling
documentation
Case study modeling is
documented in an EPA
report and published
journal articles
Available
document
citations
(USEPA(2011);
Wainger (2012);
Wainger et al.
(2013)
Birmingham, AL
The National Stormwater Calculator uses national
databases to evaluate amount/frequency of runoff
events, and uses locally-supplied information to
evaluate potential impacts of green-infrastructure;
details of the local application were not available
Modeling
completed
Case study modeling is
not documented; a
website providing and
documenting the
National Stormwater
Calculator is available
Rossm an (2014)
78
-------
4.3 Number of EPFs Used, and Linkage to Decision Support Systems
Most of the PBS that were considered for this study reported some use of models to estimate endpoints
related to the production of ecosystem services. Of 15 case studies examined, 13 reported the intention
to model ecosystem services-related endpoints, seven reported completion of at least some modeling,
and seven reported that modeling was underway or being planned. Results of modeling have already
been published in five cases; in several cases no documentation is available of case study modeling, but
documentation is available of the models themselves.
To gain a sense of the level of effort and diversity of ecosystem services examined in each case, it is
informative to consider the number of EPFs used in each study. Models of single ecological process
(Figure 4.1 A) or of multiple but dynamically inter-related processes (Figure 4. IB) were counted as
single EPFs. By contrast, when models of unrelated processes were linked together in a DSS (Figure
4.1C or D), each unrelated model was counted as a separate EPF. By this process the total number of
EPFs per study was estimated to range from 1 to 28, for an approximate total of 89 unique EPFs used in
this study; that is, no instances were found of the same EPF used in more than one of the subject PBS
(although some had been used in previous EPA studies). No attempt was made to estimate the number of
different ecosystem-service endpoints associated with these EPFs.
Seven of the case studies used, or intended to use, DSS to link EPFs or coordinate modeling across
multiple scenarios. The Tampa Bay, FL study's EPA H2O tool uses static land use conditions to drive
the following EPFs for each subwatershed included in the model: stormwater retention, based on an
improved soil-specific calculation of water retention (Reistetter and Russell 2011); air pollutant removal
and carbon sequestration by vegetation using the UFORE or i-Tree Eco model (USDA 2012); and an
arithmetic summation of geographic features of ecological interest (Ranade et al. 2015). The Guanica
Bay, PR study used the Envision decision support platform to link CORSET, a dynamic coral reef
model, to 28 EPFs corresponding to specific, service-related endpoints (Orlando and Yee, in review).
4.4 Model Computational Approach and Specificity of Model Development
The types of models applied varied widely. Numerical simulation models, which use mathematical
formulae to simulate change through time in complex systems, are often the most complex models used
for ecosystem service estimation. Of the models identified in this compilation, about six involved
numerical simulation, most often to predict water movement. Some were applications of existing models
(either off-the-shelf or with adaptation) such as the ecohydrology model VELMA used in the Blue River
Watershed, OR (Abdelnour et al. 2011, 2013); others, such as a model of tidal exchange in Tampa Bay,
FL were developed specifically for the case study. Analytic models, not live simulations, ranged from
simple mathematical functions with few variables (e.g., Guanica Bay, PR relative recreational
opportunity), which were either locally developed or transferred from other studies, to relatively
complex, pre-existing mathematical functions (Tampa Bay application of the UFORE or i-Tree Eco
model).
Simple, logic-based models (e.g., presence/absence) were also used to create predictive maps of 24
different ecosystem service-related endpoints in the St. Louis Bay estuary (MN or WI) (Angradi et al.
2016). Some depended only on single, measured variables (e.g., power boating opportunity is dependent
on water depth) while others depended on other models - for example presence of esocid (pike)
spawning depended on both water depth and a modeled statistical probability of submerged aquatic
vegetation (Angradi et al. 2013). Two kinds of stochastic techniques were used, which estimate outcome
probabilities. An existing fuzzy logic model (based on expert opinion) of bald eagle nesting was applied
79
-------
in the St. Louis Bay estuary (Angradi et al. 2016), and two Bayesian belief network models, describing
sedimentation impacts on reservoir life and coral reef condition respectively, were employed in the
Guanica Bay study (Bousquin et al. 2014; Rehr et al. 2014).
4.5 Model Variables and Input Data
One gauge of the complexity of the modeling task is the number of input variables for each model, and
the specificity of the data requirements for those variables. Input variables can be segregated into
"driving variables," here defined as those whose values vary over space or time within the model, and
constants, which do not vary within the model application. It was possible to estimate the numbers of
input variables required for about 75 EPFs representing most of the models used in the four best-
documented case studies (see Appendix C for EPF input variable totals). As seen in
Table 4.2. Approximate numbers of input variables per EPF, where "drivers" are variables whose values
change over space or time, and "constants" do not vary within the application. Values are for 75 EPFs from
four well-documented case studies.
, in spite of the fact that some models have large numbers of variables (the largest number was
approximately 89 for VELMA), typical numbers of input variables were quite low because of the
inclusion of large numbers of simple relationships in some case studies.
Table 4.2. Approximate numbers of input variables per EPF, where "drivers" are variables whose values
change over space or time, and "constants" do not vary within the application. Values are for 75 EPFs from
four well-documented case studies.
80
-------
Since so many EPFs had few variables, it is reasonable to ask whether complex modeling is necessary at
all, in order to estimate ecosystem service response to environmental management. The answer depends
on the nature of the management alternatives and of the ecosystem services of interest to community
stakeholders. If management alternatives boil down to changes in LULC, for example, and the services
of interest are adequately expressed by features readily reducible to LULC (for example, some services
estimated by the EPA H2O model, described above), then simple models often suffice. Similarly, in the
St. Louis Bay Estuary, the management alternatives entailed differing bathymetries that were related to
most of the services of interest by simple logic models (although some required computational models;
see Table 4.1). However, in the Guanica Bay watershed, because many of the management alternatives
(e.g., coffee cultivation BMPs) were in the upland and the coral reef ES of interest were enjoyed in the
Bay, primary EPFs were needed to model watershed and coral ecosystem processes before the simpler,
secondary EPFs could be used to estimate PEGS. In fact, the modeling of dynamic watershed or
estuarine processes was required in the majority of case studies examined (Table 4.1). Further, because
information limitations prevented estimation of the numbers of variables of many of those dynamic
models, the values in
Table 4.2. Approximate numbers of input variables per EPF, where "drivers" are variables whose values
change over space or time, and "constants" do not vary within the application. Values are for 75 EPFs from
four well-documented case studies.
likely present an overly simplistic picture.
4.6 Modeling Management Alternatives and Final Ecosystem Goods and Services
The definitive purpose of an EPF is to provide estimates of the production of ecosystem services -
ideally, FEGS - under environmental conditions considered likely under management alternatives at
hand. As shown in Figure 4.1, however, some EPFs require linkages to achieve this. Available
information about the decision context of each case study was insufficient to clearly indicate the
management alternatives under consideration or the stakeholder objectives. Thus, it was not possible to
determine whether the EPFs in use, singly or in combination, could fully address the case study goals.
81
-------
However, based on an inspection of model input requirements, one can gauge the types of management
actions each model should be capable of addressing. Similarly, by considering model outputs the
likelihood can be estimated that a model's endpoints corresponded to PEGS. Since the distinction
between an PEGS and an intermediate service is situation- and beneficiary-specific it was not possible to
be definitive in this regard, but one can make judgments about the prevalence of beneficiaries that would
directly value a given endpoint. For example, although retention of water on the landscape could
theoretically prevent flooding, additional modeling would be required to determine the circumstances.
Nearly all the EPFs examined appeared to be capable of modeling management actions of some kind.
The exception was the set of 28 EPFs used to estimate ES from coral reef ecological condition in
Guanica Bay; coral reef condition is largely managed indirectly via upland activities. These required
linkage to watershed and coral reef ecological process models, as explained previously. Similarly, 45
EPFs in nine of the case studies were judged to yield at least one FEGS-indicating endpoint, while 40
EPFs (being used in five of those same nine case studies) yielded only intermediate service endpoints.2
These either would require linkage to another EPF or would require decision-makers to gauge endpoint
relevance for themselves. For example, the St. Louis River Estuary study mapped locations providing
wave-action attenuation based on a simple logic-based EPF (i.e., attenuation present = area with depth
<0.75 m or aquatic vegetation present). This is not a PEGS because the amount of attenuation area
needed to achieve a given amount or frequency of shoreline erosion is not determined, but it may still
inform the judgment of decision-makers.
4.7 Representation of EPF Uncertainty, and EPF Transferability
To make good decisions, decision-makers should understand the dimensions of uncertainty surrounding
consequence estimates for management alternatives (Walker et al. 2003). Whether EPFs are developed
for the study setting at hand or transferred from a different setting, good modeling practice typically
includes: verification that a model's internal logic is correct; examination of output sensitivity to
variations in model inputs; calibration so that outputs align well with available observations; and
validation by comparison with new data, with assessment of remaining uncertainty (J0rgensen 2011).
EPFs are therefore most useful when they quantify the degree of uncertainty in model output. While
uncertainty can be expressed in statistical terms (e.g., as related to: calibration or goodness-of-fit;
validation; uncertainty analysis; sensitivity analysis) or by an explanation of knowledge limitations, PBS
leaders were only asked whether output uncertainty was estimated, and if so to provide a description. Of
11 PBS leads providing a response to that question, nine indicated that uncertainty was represented in
some form, and four of those provided some explanation. Even where the answer was "No" or no
answer was provided, our review of model-specific information suggested that uncertainty was nearly
always accounted for to a certain degree, such as by examining sensitivity associated with plausible
ranges of uncertain inputs.
The transferability of a model from a previous use to a new application site is important to analysts who
will make model use decisions. Transferability considerations include both the ease of transfer (e.g., data
available) and appropriateness of the new application (USEPA 2009). Appropriateness can be assessed,
a posteriori, by model calibration or functional validation, but little information on the use of these steps
in the case studies was provided. However, the ease of model transfer can be examined by way of
several practical considerations. First, most of the survey respondents indicated that the tools used were
2 These determinations of IEGS or PEGS are similar to those shown in Table 3-1; some differences occur because that table
is based on PBS-lead responses to an information request whereas the values given here, compiled from Appendix C, are
based mainly on reviews of model source documents.
82
-------
publicly available - in some cases this meant that model software was available via the web; in others it
merely meant that methods were published and software would be provided by the author. Second, this
assessment determined that more than half of the EPFs identified had been used previously versus
having been newly developed for the case study (i.e., 51 vs. 37); and finally, it was estimated that the
preponderance of EPFs described (54) could be applied3 using data that were widely available, while
fewer (27) required data that were only locally collected.
4.8 General Observations on EPF Use, and Lessons Learned, in these Place-Based
Studies
These PBSs were encountered at different stages of execution, and the modeling information obtained
was therefore highly uneven. It should be noted that several issues of importance in EPF selection and
use could not be evaluated very well, via the information request employed for this study. For example,
it usually was not possible to determine whether model selection was well-matched to, or limited by, the
type and quality of data available. It also was not possible to evaluate how well the spatial extent and
granularity of modeling (see data in Appendix C), or the temporal extent and granularity (not listed),
reflected the scales of ecosystem service production and use, and of ecosystem management. The level
of integration of the modeling process with all aspects of the decision context, such as stakeholder
engagement and benefit estimation, could not be evaluated, nor whether trade-off analysis was
systematically presented. Nonetheless, in several of the completed and documented studies, EPF use was
relatively well executed and integrated, showing several characteristics that bode well for the continued
improvement of ecosystem-service modeling state-of-the art and feasibility:
• The frequent use of multiple EPFs to estimate production of multiple services is important for
enabling robust trade-off evaluation and avoiding the problem of unforeseen consequences.
• The practice of linking primary and secondary EPFs enables the use of existing, relatively
complex process simulation models (off-the-shelf or with adaptation) to model management
action influences on ecosystem processes, to be complemented by relatively simple, logic-based
models that are readily adapted to local conditions of final ecosystem service delivery and use.
• The bundling of these primary and secondary EPFs within DSS coordinates the computation of
multiple EPFs for given scenario conditions, and may enable policy simulation or optimization.
• Many of the models used had been used previously, or were considered by the investigators to be
readily transferrable.
At the same time, several deficiencies were easily noted, or were suspected:
• The lack of repetition of any EPF or DSS across these case studies may speak in part to the small
number of studies involved, the diversity of their settings and the relative newness of efforts to
include ecosystem services, but it may also suggest an investigator-specific approach has been
taken to model selection. A more formal approach for evaluating model suitability may be
needed. Both practical guidance for assessing model transferability, and the establishment of
3The focus here is on availability of data needed for model application, not model development. Some models requiring
specialized local data for development could be applied in another setting based on commonly available data. For example,
some coral reef EPFs developed using detailed, coral condition datasets available near St. Croix were applied in Guanica Bay
using only bathymetry and benthic habitat data.
83
-------
communities of practice around specific EPFs or DSSs, or around EGS modeling in general,
could help to make model evaluation and model transfer more systematic.
Part of that suitability evaluation would require more systematic approaches to the various facets
of uncertainty assessment, and ensuring that uncertainty assessment is built into DSS so as to
become more routine - not only to modeling practitioners but to users of ecosystem service
production estimates.
In the information available for this analysis, it was not always clear that EPF choice was based
on the decision context, as well as the ecosystem characteristics, of the PBS site. Therefore, a
final issue for increased emphasis is ensuring alignment of EPF capabilities with: (a) the
characteristics of management alternatives (which generally should align with model inputs); and
(b) stakeholder goals and the PEGS that most closely meet them (which should align with model
outputs). This alignment-to-context may be accomplished in a single EPF or via multiple, linked
EPFs.
Photo credit - Nadia Seeteram
•T*BC.rf5^-»- 'f
84
-------
4.9 Literature cited
Abdelnour, A, R.B. McKane, M. Stieglitz, F. Pan, and Y. Cheng. 2013. Effects of harvest on carbon
and nitrogen dynamics in a Pacific Northwest forest catchment. Water Resources Research
49:1292-1313.
Abdelnour, A., M. Stieglitz, F. Pan, and R.B. McKane. 2011. Catchment hydrological responses to
forest harvest amount and spatial pattern. Water Resources Research 47:18.
Angradi, T.R., JJ. Launspach, D.W. Bolgrien, BJ. Bellinger, M.A. Starry, J.C. Hoffman, A.S. Trebitz,
M.E. Sierszen, and T.P. Hollenhorst. 2016. Mapping ecosystem service indicators in a Great
Lakes estuarine Area of Concern. Journal of Great Lakes Research in press.
Angradi, T.R., M.S. Pearson, D.W. Bolgrien, BJ. Bellinger, M.A. Starry, and C. Reschke. 2013.
Predicting submerged aquatic vegetation cover and occurrence in a Lake Superior estuary.
Journal of Great Lakes Research 39:536-546.
Bellinger, B.J., T. Jicha, M. Lehto, L.P. Seifert-Monson, L.R. Bolgrien, D.W. Starry, M.A., T.R.
Angradi, M.P. Pearson, C. Elonen, and B.H. Hill. 2014. Sediment nitrification and
denitrification in a Lake Superior estuary. Journal of Great Lakes Research 40:392-403.
Bousquin, J., W. Fisher, J. Carriger, and E. Huertas. 2014. A Bayesian Belief Network Approach to
Explore Alternative Decisions for Sediment Control and Water Storage Capacity at Lago
Lucchetti, Puerto Rico. U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-
14/296.
Boyd, J. and S. Banzhaf. 2007. What are ecosystem services? The need for standardized environmental
accounting units. Ecology Economics 63:616-626.
Bradley, P., W. Fisher, B. Dyson, S. Yee, J. Carriger, G. Gambirazzio, J. Bousquin and E. Huertas.
2016. Application of a Structured Decision Process for Informing Watershed Management
Options in Guanica Bay, Puerto Rico. U.S. Environmental Protection Agency, Office of
Research and Development, Narragansett, RI, EPA/600/R-15/248.
Bradley, P., D. Santav, and J. Gerritsen. 2014. Workshop on Biological Integrity of Coral Reefs,
August 21-22, 2012, Caribbean Coral Reef Institute, Isla Magueyes, La Parguera, Puerto Rico.
U.S. Environmental Protection Agency, Office of Research and Development, Narragansett, RI,
EPA/600/R-13/350.
Bruins, R.J.F., T.J. Canfield, C. Duke, L. Kapustka, A.M. Nahlik, and R.B. Schafer. 2016. Using
ecological production functions to link ecological processes to ecosystem services. Integrated
Environmental Assessment and Management. Accepted Author Manuscript.
doi:10.1002/ieam.l842.
Costanza, R., R. de Groot, P. Sutton, S. van derPloeg, SJ. Anderson, I. Kubiszewski, S. Farber, and
R.K. Turner. 2014. Changes in the global value of ecosystem services. Global Environmental
Change 26:152-158.
Daily, G.C., S. Polasky, J. Goldstein, P.M. Kareiva, H.A. Mooney, L. Pejchar, T.H. Ricketts, J.
Salzman, and R. Shallenberger. 2009. Ecosystem services in decision making: Time to deliver.
85
-------
Front Ecological Environment 7:21-28.
de Groot, R., L. Brander, S. van derPloeg, R. Costanza, F. Bernard, L. Braat, M. Christie, N.
Grossman, A. Ghermandi, and L. Hein. 2012. Global estimates of the value of ecosystems and
their services in monetary units. Ecosystem Services 1:50-61.
de Groot, R.S., R. Alkemade, L. Braat, L. Hein, and L. Willemen. 2010. Challenges in integrating the
concept of ecosystem services and values in landscape planning, management and decision
making. Ecological Complex 7:260-272.
European Environment Agency (EEA). 2016. CICES: Towards a common classification of ecosystem
services (website), [accessed 20 May 2016]. CICES.
Executive Office of the President. 2015. Memorandum for Executive Departments and Agencies:
Incorporating Ecosystem Services into Federal Decision Making. Office of Management and
Budget, Council on Environmental Quality, Office of Science Technology and Policy,
Washington, DC, No. M-16-01
Fisher, B., R.K. Turner, and P. Morling. 2009. Defining and classifying ecosystem services for decision
making. Ecological Economics 68:643-653.
Fulford, R., D. Yoskowitz, M. Russell, D. Dantin, and J. Rogers. 2016a. Habitat and recreational
fishing opportunity in Tampa Bay: Linking ecological and ecosystem services to human
beneficiaries. Ecosystem Services 17:64-74.
Fulford, R.S., M. Russell, and I.E. Rogers. 2016b. Habitat restoration from an ecosystem goods and
services perspective: Application of a spatially explicit individual-based model. Estuaries and
Coasts pp 1-15.
Hu, W. and Y. Yuan. 2013. Evaluation of Soil Erosion and Sediment Yield for the Ridge Watersheds in
the Guanica Bay Watershed, Puerto Rico, Using the SWAT Model. U.S. Environmental
Protection Agency, Office of Research and Development, National Exposure Research
Laboratory, Environmental Sciences Division, Las Vegas, NV.
Johnston, RJ. and M. Russell. 2011. An operational structure for clarity in ecosystem service values.
Ecological Economics 70:2243 -2249.
Jordan, S.J., T. O'Higgins, and J.A. Dittmar. 2012. Ecosystem services of coastal habitats and fisheries:
Multiscale ecological and economic models in support of ecosystem-based management.
Marine and Coastal Fisheries 4:573-586.
J0rgensen, S.E. 2011. Fundamentals of Ecological Modelling: Applications in Environmental
Management and Research. Elsevier, Amsterdam, The Netherlands.
Krumholz, J., J. Vaudrey, M. Brush, and D. Ullman. 2015. A Review of the EcoGEM Modeling
Approach as Applied to Narragansett Bay, RI. National Oceanic and Atmospheric
Administration [accessed 7 September 2016]. Long Island Sound Study.
Landers, D. 2015. National Ecosystem Services Classification System (NESCS): Framework Design
and Policy Application. U.S. Environmental Protection Agency, Washington, DC, EPA/800/R-
86
-------
15/002.
Landers, D. and A. Nahlik. 2013. Final Ecosystem Goods and Services Classification System (FEGS-
CS). U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-13/ORD-004914.
Maes, J., B. Egoh, L. Willemen, C. Liquete, P. Vihervaara, J.P. Schagner, B. Grizzetti, E.G. Drakou,
A.L. Notte, and G. Zulian. 2012. Mapping ecosystem services for policy support and decision
making in the European Union. Ecosystem Services 1:31-39.
Munns, W.R., A.W. Rea, MJ. Mazzotta, L.A. Wainger, and K. Saterson. 2015. Toward a standard
lexicon for ecosystem services. Integrated Environmental Assessment and Management 11:666-
673.
Nahlik, A.M., M.E. Kentula, M.S. Fennessy, and D.H. Landers. 2012. Where is the consensus? A
proposed foundation for moving ecosystem service concepts into practice. Ecological
Economics 77:27-35.
National Research Council. 2004. Valuing Ecosystem Services: Toward Better Environmental
Decision-Making. The National Academies Press.
Orlando, J.L. and S. Yee. Forthcoming in review. Linking Sediment Threat to Declines in Coral Reef
Ecosystem Services. U.S. Environmental Protection Agency, Office of Research and
Development, National Health and Environmental Effects Research Laboratory, Gulf Ecology
Division, Gulf Breeze, FL.
Osland, M.J., A.C. Spivak, J.A. Nestlerode, J.M. Lessmann, A.E. Almario, P.T. Heitmuller, MJ.
Russell, K.W. Krauss, F. Alvarez, and D.D. Dantin. 2012. Ecosystem development after
mangrove wetland creation: Plant-soil change across a 20-year chronosequence. Ecosystems
15:848-866.
Rehr, A.P., MJ. Small, P.S. Fischbeck, P. Bradley, and W.S. Fisher. 2014. The role of scientific studies
in building consensus in environmental decision making: A coral reef example. Environment
Systems and Decisions 34:60-87.
Reistetter, J.A. and M. Russell. 2011. High-resolution land cover datasets, composite curve numbers,
and storm water retention in the Tampa Bay, FL region. Applied Geography 31:740-747.
Rogers, I.E., MJ. Russell, and M.C. Harwell. Forthcoming 2016. Improved Method for Calibration of
Exchange Flows for a Physical Transport Box Model of Tampa Bay, FL, USA. U.S.
Environmental Protection Agency, National Health and Environmental Effects Research
Laboratory, Gulf Ecology Division, Gulf Breeze, FL.
Rossman, L.A. 2014. National Stormwater Calculator User's Guide - Version 1.1. U.S. Environmental
Protection Agency, Office of Research and Development, Cincinnati, OH. [accessed 7
September 2016]. National Stormwater Calculator User's Guide.
Russell, M., J. Harvey, P. Ranade, and K. Murphy. 2015. EPA H2O User Manual. U.S. Environmental
Protection Agency, Office of Research and Development, Washington, DC, EPA/600/R-15/090.
Smith, A., S. H. Yee, M. Russell, J. Awkerman and W. S. Fisher. Forthcoming. Linking Ecosystem
87
-------
Services Supply to Stakeholder Concerns on Both Land and Sea: An Example from Guanica
Bay Watershed, Puerto Rico. U.S. Environmental Protection Agency, Office of Research and
Development, Gulf Ecology Division, Gulf Breeze, FL.
Texas A&M Agrilife Research. 2016. APEX - Agricultural Policy/Environmental extender Model.
[accessed 29 August 2016]. EPIC & APEX Models.
U.S. Department of Agriculture (USDA). 2012. i-Tree Eco User's Manual V.5. Online: USDA Forest
Service [accessed 20 May 2016]. i-Tree Manuals & Workshops.
U.S. Environmental Protection Agency (EPA). 2007. The Environmental Fluid Dynamics Code User
Manual Version 1.01. In: Agency UEP (ed). Tetra Tech, Inc., Fairfax, VA, 231 p.
U.S. Environmental Protection Agency (EPA). 2009. Guidance on the Development, Evaluation, and
Application of Environmental Models. U.S. Environmental Protection Agency, Washington,
DC,EPA/100/K-09/003.
U.S. Environmental Protection Agency (EPA). 2011. An Optimization Approach to Evaluate the Role
of Ecosystem Services in Chesapeake Bay Restoration Strategies. U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-11/001.
U.S. Environmental Protection Agency (EPA). 2013. Water Quality Analysis Simulation Program
(WASP), [accessed 25 April 2016]. WASP.
U.S. Environmental Protection Agency (EPA). 2015. EJSCREEN Technical Documentation, [accessed
2 August 2016]. EJSCREEN Tool.
U.S. Environmental Protection Agency (EPA). 2016. Visualizing Ecosystem Land Management
Assessments (VELMA) Model - 2.0. Online: U.S. Environmental Protection Agency [accessed
26 April 2016]. VELMA Model 2.0.
Wainger, L.A. 2012. Opportunities for reducing total maximum daily load (TMDL) compliance costs:
Lessons from the Chesapeake Bay. Environmental Science and Technology 46:9256-9265.
Wainger, L.A., G. Van Houtven, R. Loomis, J. Messer, R.H. Beach, and M. Deerhake. 2013. Tradeoffs
among ecosystem services, performance certainty, and cost-efficiency in implementation of the
Chesapeake Bay total maximum daily load. Agricultural and Resource Economics Review
42:196-224.
Walker, W.E., P. Harremoes, J. Rotmans, J.P. van der Sluijs, M. van Asselt, P. Janssen, and M. Krayer
von Krauss. 2003. Defining uncertainty: A conceptual basis for uncertainty management in
model-based decision support. IntegratedAssessment 4:5-17.
Yee, S.H., J.A. Dittmar and L.M. Oliver. 2014. Comparison of methods for quantifying reef ecosystem
services: A case study mapping services for St. Croix, USVI. Ecosystem Services 8:1-15.
Yee, S.H., J. Orlando, and K. Vache. Forthcoming. Linking management of sediment runoff to coral
reef ecosystem services in the Guanica Bay Watershed, Puerto Rico.
Yuan, Y., E. Mbonimpa, W. Hu, W. Fisher, and A. Casper. 2012. Quantifying Sediment Contributions
88
-------
to Guanica Bay, Puerto Rico. American Water Resources Association Conference, Jacksonville,
FL
89
-------
5 Identifying and Measuring Benefits of
Photo courtesy of USEPA
5.1 Differentiating PEGS from Benefits
Comparison of the value placed on ecosystem
services (ES) across communities is
problematic because of varying level of
awareness, understanding and appreciation of
the concept of PEGS benefits as it relates to
human health and welfare. Welfare in
particular is a broad benefit category and
potentially includes any number of well-being
aspects such as happiness, aesthetic appeal,
and sense of purpose. As defined by Fisher
and Turner (2008), "a benefit is something that
has an explicit impact on changes in human
welfare, like more food, better hiking, less
flooding," which differs from PEGS, which
are the "components of nature, directly
enjoyed, consumed, or used to yield human
well-being" (Boyd and Banzhaf 2007). The concept of ecological and health "benefits" resonates more
with the general public than the concept of PEGS because benefits are easier to understand and to
connect an ecosystem's contribution to personal well-being.
The benefits that people derive from PEGS can manifest in terms of gains in social, economic, and
health welfare. As seen in the conceptual model (Figure 1.1), "benefit functions," represent the link
between PEGS and human well-being, and "demonstrate(s) how people value a gain or avoid a loss in
an ecosystem service either in monetary terms or as a relative magnitude of social value when
willingness-to-pay is not measurable" (Wainger and Mazzotta 2011). In other words, benefits can be
quantified through economic valuation methods, or through assessing tradeoffs when presented with
limited information. Conducting these types of assessments provides vital information upon the
contributions of PEGS towards human well-being. The nuances in identifying and differentiating
between PEGS and benefits have prompted the development of several ES classification systems such as
PEGS Classification System (Landers andNahlik2013), the National Ecosystem Services Classification
System (NESCS) (Landers 2015) and Common International Classification of Ecosystem Services
(CICES) (Haines-Young and Potschin 2010). Although the approaches within the systems differ, the
main purpose of these classification systems is to identify ecosystem services for assessment and to
facilitate quantification and valuation. These classification systems are, however, not taxonomies for ES.
Instead, they are a method of identifying PEGS of interest (using the classification system) that link ES
to benefits and the relevant beneficiaries (and also industrial classification systems).
90
-------
Identifying PEGS and the resulting benefits
can be achieved through a variety of means.
Usually, a stakeholder engagement exercise is
used combined with a classification system to
guide stakeholders through the process.
Within the place based studies, most project
leads («= 15) indicated that the stakeholder
engagement process (32.3%), and literature
reviews (32.3%) were the most widely used
sources for ES identification, but local expert
consultation (25.8%), and other sources
including peer groups, social media, news
sources (9.7%) were also utilized.
The project leads were also asked to assess
how important each source was in ES
identification. Figure 5.1 displays the results
of this question across a spectrum of
"extremely important" to "not at all
important." 40% of project leads indicate that
"local expert consultation" was "extremely
important" as a source, while 47% of project
leads stated that "stakeholder engagement,"
"literature reviews," and "other sources" were
"Important" to their ES identification process.
Lessons learned from measurement of human benefits
in place-based studies:
1) A benefit is something that impacts human
welfare
2) Place-based studies reported that most
stakeholders did not struggle with
understanding the concept of benefits but
terminology was highly variable
3) Many studies described project success as
something other than human benefits such as
stakeholder education or eco-integrity
4) A stronger connection is needed between
project goals and measures of benefits
5) This connection is greatly facilitated by
stakeholder engagement in the identification
and measurement of human benefits
How important were each of these sources in
determining the PEGS used in your study?
Extremely important
Important
Neither important nor unimportant
Unimportant
Not at all important
012345
I Other (including social media, news sources, peer groups)
I Literature reviews
I Local expert consultation
I Stakeholder engagement
Figure 5.1 Responses on scale of 1-5, where 1= "not at all important" and 5= "extremely important" to
determine how important various sources were in ES identification.
91
-------
On the other end of the spectrum, 20% of researchers indicated that "Other sources" and "local expert
consultation" were "not at all important," while 13% of researchers expressed that "literature reviews"
and "stakeholder engagement" were also "not at all important." However, two of these researchers did
not indicate in the previous question that they used any of these sources to identify ES. As such, it is
very likely that these responses are negligible.
5.2 Communicating Benefits to Stakeholders
In discussion with stakeholders about complex concepts such as PEGS, using terminology that resonates
with the general public is critical for effective communication of scientific information. Generally
speaking, the public-at-large has low levels of understanding of scientific information, which is further
complicated by the dynamic nature of ecological processes. As such, researchers are challenged to
convey appropriate levels of scientific knowledge that the public can readily understand without
compromising the integrity of the information.
5.2.1 The Concept of Benefits
When asked if stakeholders struggled with understanding the concept of "benefits," 64.3% of project
leaders stated "no," while 21.4% indicated that stakeholders did struggle with this concept. Only 14.3%
of project leads expressed that stakeholders had a "mixed response" to the explanation of this concept.
In order to increase comprehension around benefits, project leads suggested that "further explanation of
benefits and ES" and "connecting their priorities to the domains of human well-being" were helpful in
reducing misunderstandings.
5.2.2 Terminology to Describe Benefits
Project leads were asked to indicate which of the following
terms they used during the stakeholder engagement process to
describe "natural resources." Figure 5.2 displays the results
with most (33.3%) of project leads indicating that they used
"ecosystem services" to describe "natural resources,"
followed by 14.8% of project leads indicating they also used
"nature's benefits" and "environmental value," within these
discussions. When asked how frequently they used these
terms, project leads expressed that they used "ecosystem
services" most (37.5%) followed by "nature's benefits"
(18.7%).
Recent research has shown that the public does not readily
understand the term "ecosystem services," and instead
"nature's benefits" and "nature's value" may be more
effective in conveying the link between ecosystems and the
benefits received from them (Metz and Weigel 2010). As
place based studies continue to progress, researchers should
consider using different terminology in order to facilitate
comprehension on this subject matter.
Photo courtesy of USEPA
92
-------
In discussion with stakeholders, did you use any of
the following terms to describe "natural resources"?
Ecosystem services • Nature's benefits • Nature's value PEGS
Natural capital • Environmental Value • Other
Figure 5.2 Frequency of use of various terms to describe "natural resources".
5.3 Types of Benefits
In addition to introducing the idea that enhancements or restoration of ecosystem goods and services can
provide people with many ecosystem benefits (i.e., clean drinking water supplies, raw materials, cleaner
air, etc.), researchers should also enumerate the potential human health benefits that may result along
with ecosystem improvements (e.g., reduced incidence of disease, enhanced quality of life). Project
leads indicated across the board that they were able to connect EGSs to human health and well-being in
discussions with stakeholders. In one specific case, the study was able to link specific ES such as,
"protection against extreme events/floods, water quantity protection, water quality protection, habitat
protection, air quality protection and open space conservation" to human health and well-being. In a
more systematic example of how project leads were able to connect these concepts, the project lead used
a Health Impact Assessment (HIA) to form this connection, and explained that "the intent of the HIA is
to link a decision to health determinants to health outcomes. In this HIA, ecosystem services were
included in the pathway from decision to health outcome."
5.4 Tools to Quantify Benefits
Several tools exist to measure the health and well-being benefits that people receive from changes in the
provisioning of PEGS as a result of proposed management actions. Figure 5.3 depicts the results of
which tools were used within the PBS. In measuring changes in ES, In VEST
(http://www.naturalcapitalproiect.org/invest/) was used most frequently. As for measuring changes in
human health/well-being, "HIA, the Eco-health relationship browser
(https://www.epa.gov/enviroatlas/enviroatlas-eco-health-relationship-browser), risk assessment,
epidemiology studies, and reviews of clinical case studies" were all used at the same rate. However, the
"other" response was the most frequently selected response. "EnviroAtlas, EPA H2O, i-Tree" were
listed amongst the "other tools" used for measuring changes in ES, while the "other tools" listed for
measuring human health and well-being were simply clarifications on tools already listed amongst the
selection.
93
-------
Were any of these tools used to measure changes
in ES/Human Health outcomes, as a result of
proposed management actions ?
Figure 5.3 Results depicting frequency of tool use for measuring changes in Ecosystem Services/Human
Health and Human Well-Being.
5.5 Measuring Success
Within placed based studies that incorporate ecosystem services into their project framework, "success"
is dependent on the goals of the study, and there can be several (Figure 5.4). Complete follow-through
on the identification of benefits would be an incorporation of associated measures of these benefits (i.e.,
benefits indicators) into the goals of a project. However this was not at all common among the PBS
included in this study. In terms of defining and measuring success, "stakeholder engagement" elicited
the highest degree of prioritization with 21.7% of respondents, followed by 17.4% of project leads who
selected "publications/dissemination" of results as measure of success. "Other" measures of success was
another frequently selected option. When asked to elaborate on this selection, project leads indicated that
"increased interest, support, and participation in the study," "creating awareness about environmental/
health issues within local communities," and "making an "impact on [the] decision" were amongst the
responses. One of the most effective ways of achieving the latter goal includes developing "benefit
relevant indicators," or BRIs, that directly integrates the benefits people receive from enhancement of
EGS. The Woonasquatucket PBS demonstrates how hydrologic and hydraulic simulation model results
can be used to address critical questions regarding EGS provisioning and beneficiaries in order to
develop BRIs for a specific decision context (Bousquin et al. 2015). However this PBS was the
exception rather than the rule for integrating benefits into assessment of project success.
94
-------
How was success measured in line with
project outcomes?
I Stakeholder engagement • Creating awareness
I Building decision support networks • Measured change in eco integrity
I Measured enhancement in human well-being • Policy creation/implementation
I Publications/Dissemination • Other
Figure 5.4 Responses to "How was success measured in line with project outcomes?"
5.6 Conclusions
A few conclusions can be drawn from the information presented in this chapter. First, project leads
indicated that local expert consultation was "extremely important" in identifying ES for use in their
studies, while sources like stakeholder engagement, literature review, and other sources were also
"important." While consulting local experts may have been the most effective source to use in ES
identification, project leads utilized a variety of sources for the most comprehensive understanding of
priority ES. In doing so, project leads were likely well equipped to engage with stakeholders, local
decision makers, and experts. Following this positive conclusion, project leads observed that the
majority of stakeholders did not struggle in understanding the concept of "benefits," even though they
used the term "ecosystem services" to frame the discussion more frequently than "nature's benefits" or
"nature's value." Further studies should be conducted to test the effectiveness of using one phrase over
the other, as terminology choice and framing can substantially alter discussions with stakeholder and
local decision makers. The identification and practical application of benefit indicators is a key gap
observed in the PBS considered for this study and this represents a key area for future research into
decision support and its acceptance at the community level.
95
-------
5.7 Literature Cited
Bousquin, J., K. Hychka, and M. Mazzotta. 2015. Benefit Indicators for Flood Regulation Services of
Wetlands: A Modeling Approach. U.S. Environmental Protection Agency, Washington, DC,
EPA/600/R-15/191.
Boyd, J.W. and S. Banzhaf. 2007. What are ecosystem services? The need for standardized
environmental accounting units. Ecological Economics 63(2-3):616-626.
Fisher, B. and R.K. Turner. 2008. Ecosystem services: Classification for valuation. Biological
Conservation 141(5): 1167-1169.
Haines-Young, R. and M. Potschin. 2010. Proposal for a Common International Classification of
Ecosystem Goods and Services (CICES) for Integrated Environmental and Economic
Accounting. Report to the European Environment Agency.
Landers, D. 2015. National Ecosystem Services Classification System (NESCS): Framework Design
and Policy Application. U.S. Environmental Protection Agency, Washington, DC, EPA/800/R-
15/002.
Landers, D. and A. Nahlik. 2013. Final Ecosystem Goods and Services Classification System (FEGS-
CS). U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-13/ORD-004914.
Metz, D. and L Weigel. 2010. Key Findings from Recent National Opinion Research on 'Ecosystem
Services'. The Nature Conservancy, [accessed 7 September 2016]. Conservation Gateway.
Wainger, L. and M. Mazzotta. 2011. Realizing the potential of ecosystem services: A framework for
relating ecological changes to economic benefits. Environmental Management 48(4):710-733.
96
-------
6 Synthesis of Lessons Learned
The application of an EGS approach to community-level decision support is an effective method for
balancing complicated issues and targeting sustainability. Guidelines for use of an EGS approach in
decision making highlight the need for connecting assessments to both scientific data and stakeholder
values; establishing well-defined measures of success, and including a comprehensive set of services to
people (Olander et al. 2015). The PBS considered in this study have made tangible progress towards
achieving these objectives and most PBS reported interest in doing more (
Overall the application of FEGS-based decision support across the PBS participating in this study can be
described as pervasive but incomplete. Multiple studies reported use of EGS concepts, stakeholder
engagement, application of environmental models, or direct measures of human benefit linked to
environmental action. Yet, the focus was almost universally on a subset of these elements usually tied to
the specific project objectives. For instance, in one PBS located in the Pacific Northwest, the emphasis
was on the application of quantitative models to address existing land management issues, but because
the issues and planned remediation actions were both well-established, minimal stakeholder engagement
occurred during the project and direct measures of human benefit were not developed. This incomplete
application of the core PEGS elements, as well as the need for integration of these elements into a
cohesive approach, represents a major objective for future work in decision support research.
97
-------
Table 6.1). Yet, there is a need for a unifying framework that connects goals to measures of success and
allows for the maximum integration of stakeholder input with the best available scientific information.
In this report, a complete conceptual model for an EGS approach at the community level has been
evaluated in the context of existing and previous place-based studies with an eye towards how this
model has been used, and what gaps exist that might be filled to maximize its successful use in future
PBS.
Overall the application of FEGS-based decision support across the PBS participating in this study can be
described as pervasive but incomplete. Multiple studies reported use of EGS concepts, stakeholder
engagement, application of environmental models, or direct measures of human benefit linked to
environmental action. Yet, the focus was almost universally on a subset of these elements usually tied to
the specific project objectives. For instance, in one PBS located in the Pacific Northwest, the emphasis
was on the application of quantitative models to address existing land management issues, but because
the issues and planned remediation actions were both well-established, minimal stakeholder engagement
occurred during the project and direct measures of human benefit were not developed. This incomplete
application of the core FEGS elements, as well as the need for integration of these elements into a
cohesive approach, represents a major objective for future work in decision support research.
Photo credit - Nadia Seeteram
98
-------
Table 6.1 Summary of PBS responses to question, "If I could change something about the project what would it be?''
options in a menu, while the final column was an open response.
First seven columns were
Information
about risks
Mediation
to balance
coalition
interests
Structured
decision
approach
More trust Information
between <_
stakeholders cost/benefit
of options
More
ability or
resources
to enact
good
decisions
Technological
tools or
models for
comparing
guidance
outcomes
Other responses
Total
Percentage
8
53
5
33
10
66
4
27
13
87
8
53
6
40
Focus on public
health; Spatial
information on
supply of EGS;
Environmental
assessment methods;
Data needs on
pollutant migration
from sources; Tools
that promote
resiliency to climate
change and
development;
Networking among
scientists
99
-------
6.1 Components of the Conceptual Model
Stakeholder Engagement - Establishing ecosystem services concepts within a decision context is a
crucial step that helps both scientists and decision-makers. In the PBS examined for this report,
decisions were often made with simple end goals in mind (i.e., economic development) that only
encompassed the input from a small select group of professionals that drove the decisions on the basis of
their expertise. This resulted in limited support from the local community and potentially made the
outcomes vulnerable to criticism. In another example in Puerto Rico, stakeholder engagement was
extensive and demonstrated previously unknown frustration among stakeholders for the decision process
in which they previously had limited engagement. Some common lessons emerged, including:
• Engage stakeholders and local decision-makers throughout the process
• Take the time with stakeholders to formally define the decision context
• Clarify with stakeholders why natural resources matter.
• Use conceptual models and systems thinking to uncover unintended consequences
• Work with diverse groups of experts to integrate multidisciplinary sources of information.
Ecosystem services assessments are often centered on adaptation of tools to a problem, rather than
allowing the problem to determine the appropriate set of tools. More studies are needed that apply a
structured decision making approach to better integrate stakeholder priorities with data on ecosystem
services and human welfare.
FEGS Concepts - Final ecosystem goods and services are those EGS that directly benefit humans. In
Chapter 3, this distinction is used to explore how EGS are defined in the PBS considered for this study
and whether any rise to the level of being a PEG by making the beneficiary link. In practice, FEGS
identification within PBS considered here was limited, largely because the FEGS concept is a novel one,
but the pieces were there nonetheless. For example in one PBS in Massachusetts, wetlands were
identified as providing the EGS of flood control and the economic benefits were estimated, but no
specific beneficiary was directly identified. Beneficiary identification was variable with only a small set
of studies formally linking EGS to human welfare and only one study considering existence and option
value of EGS. Study endpoints were more likely to be intermediate EGS not directly linked to human
beneficiaries. Given these findings, especially the variability in identifying beneficiaries, the mixing of
metrics or indices between final and intermediate EGS, and the need to link FEGS with specific
beneficiaries, it would seem that additional effort will be needed to fully utilize FEGS concepts in place
based studies.
EPFs - Ecological production functions have been defined as "usable expressions (i.e., models) of the
processes by which ecosystems produce ecosystem services, often including external influences on those
processes" (Bruins et al. 2016). These EPFs represent the most tangible application of scientific
information to problem solving and so are a vital link in the conceptual model between public policy and
science. In examining how PBS considered for this study used EPFs, several key conclusions were:
• Place-based studies most frequently used multiple EPFs in a coordinated fashion (via linkage to
one another or execution within decision-support tools), so as to estimate multiple services
100
-------
• Many of the EPFs used may be considered transferable to other locations because they have been
transferred in the past and/or they use widely available data
• Many EPFs that estimate PEGS are simple models (e.g., of five or fewer input variables), but
may require linkage to more complex models for simulation of management alternatives
• Evaluation and communication of uncertainty must become more systematic
• Methods are needed for evaluating model suitability, especially alignment to decision context.
In several cases, different PBS had similar focal issues, such as the PBS in Wisconsin and Rhode Island
that had a similar focus on water quality effects of watershed urbanization but each used different
models to approach these similar questions. Taken in isolation, EPFs were commonly applied although
the information request lacked specificity to examine EPF use in key areas such as spatial/temporal scale
of application, or how well the EPF used was connected to the specific PBS decision context. However,
it can be concluded that more work is needed to standardize EPFs for particular problems and effectively
link EPF outputs to independently derived measures of human benefit.
Benefits - A benefit is something that has an explicit impact on changes in human welfare, which differs
from Final Ecosystem Goods and Services (PEGS). Benefits are derived from PEGS production and
delivery to people. For the PBS considered in this study, project leads observed that the majority of
stakeholders did not struggle in understanding the concept of "benefits," even though they used the term
"ecosystem services" to frame the discussion more frequently than "nature's benefits" or "nature's
value." Yet, many study leads listed their measure of project success as either enhanced stakeholder
engagement (e.g., education) or generation of publications. A small number of PBS reported success as
enhancement of either human well-being or ecological-integrity. Clearly for the PBS considered here, a
stronger connection is needed between project goals and measures of benefit used to estimate project
success. This connection can be greatly facilitated by involvement of stakeholders and the careful
integration of desired outcomes into project indicators and terminology. Future studies should be
conducted to test the effectiveness of using one phrase over the other to describe benefits with
stakeholders, as terminology choice and framing can significantly alter discussions with stakeholder and
local decision makers.
The Conceptual Model in Decision Support - The respective elements of the proposed conceptual
model (e.g., stakeholder engagement/decision context, PEGS, EPFs, and measures of benefit) each
represent significant efforts to support community-level decision making as evidenced by their
successful use in one or more of the PBS considered in this report. Taken as a whole, the model also
provides critical linkages across the respective elements that can bring about a novel integration of
science and policy and yield much more effective measures of decision outcomes. Stakeholders bring an
understanding of both potential actions and the desired outcomes from those actions; and science
provides a defendable, robust understanding of how actions can translate into desired outcomes. Barriers
to such an integration of science and policy include a lack of stakeholder involvement and understanding
of an EGS approach, challenges of matching EPFs to the problem at hand, use of inadequate short-term
objectives as measures of benefit, and a need to better integrate multiple issues into a common decision
framework. An EGS-based conceptual model for decision support, such as the one proposed here, can be
highly effective in overcoming these challenges by linking the decision process together in a clear way,
but more work needs to be done. Place-based studies offer a rich opportunity to explore the application
of this conceptual model to real-world issues and as such are a vital link in EPA research into
community-based decision support and the fostering of sustainable human and environmental health.
101
-------
6.2 Literature Cited
Olander, L., RJ. Johnston, H. Tallis, J. Kagan, L. Maguire, S. Polasky, D. Urban, J. Boyd, L. Wainger,
and M. Palmer. 2015. Best Practices for Integrating Ecosystem Services into Federal Decision
Making. National Ecosystem Services Partnership, Duke University, Durham, NC.
doi:10.13016/M2CH07
102
-------
Appendix A - Full Set of Questions Used in Information Request
1. Please describe or name the focal community of your place-based study.
2. What made you choose the place in your place-based study over another? Choose all that apply.
Availability of resources/personnel
i—
Proximity
Pre-defined by research question
Randomly selected
Specify your own value:
3. In your opinion, is your case study community somehow different than other communities in any
of the following ways? Choose all that apply.
People living there are more open to working with scientists
Acute need for remediation or risk reduction
Strong leadership and interest in consulting with scientists and models
Atypical or archetypal ecological system
Required by law to address an environmental problem
Specify your own value:
r
r
4. In your opinion, what are the community needs of your place-based site? Choose all that apply.
^~-
Information about risks
Mediation or consultation to balance coalition interests
Structured approach to decision making
More trust between community stakeholders, scientists, and/or the federal government
Information about costs and benefits of certain decisions
More power or resources to enact good decisions
Technological tools (hardware) or programs (software) for implementing guidance instruments such as the
PEGS framework
103
-------
Specify your own value:
5. Which of the following non-ORD community stakeholders were involved in this place-based
study, or in retrospect, do you wish had been more engaged - either at all, or a
different/broader group of stakeholders, or more in-depth engagement?
EPA Regions or Program Offices
No
f~
No, but wish more had been done
r Yes
r*-
Yes, but wish more had been done
Non-EPA decision-makers, or those with authority to act
No
No, but wish more had been done
r Yes
Yes, but wish more had been done
Scientific collaborators or other data experts outside ORD
No
r-
No, but wish more had been done
r
Yes
r
Yes, but wish more had been done
General public, local residents, interest groups
No
No, but wish more had been done
Yes
Yes, but wish more had been done
104
-------
Other or additional explanation.
6. How were stakeholders involved in this place-based study? Choose all that apply.
Stakeholders were not involved
Stakeholders were directly involved throughout study development and implementation
Stakeholders were updated periodically throughout the study and/or gave periodic feedback on ORD
progress
Stakeholders provided essential information or data to the ORD study
ORD reported results to stakeholders at the end of the study
Stakeholders were trained by ORD in the use of project approaches, tools, or models for decision-making
Specify your own value:
7. Was a structured decision process, such as a defined framework (e.g. Structured Decision
Making), regulatory guidance, or workbook used or developed for this place-based study? Choose
all that apply.
l~ No
No, but in retrospect a formal or structured process would have been useful
Yes, we used or adapted an existing decision process
Yes, development of a process was a key research effort within our study
Specify your own value:
If YES, please provide a name and/or citation for the process(es) used.
8. Was a conceptual model linking decisions to outcomes developed for this study? Choose all
that apply.
r~ NO
No, one already existed or the effort was led by the community or other group, and it was used but not
developed as part of the ORD study
Yes, we developed and used one internally within ORD for the study
105
-------
Yes, we developed and used one in collaboration with stakeholders
Specify your own value:
If YES, please list any existing or ORD-developed frameworks, software, or tools used in the
development of conceptual models. Please provide key citations.
9. Were stakeholder concerns or goals characterized for the study, such as to identify
focal endpoints for the study? Choose all that apply.
r~ NO
No, they already existed or the effort was led by the community or other group, and it was used but not
developed as part of the ORD study
Yes, they were inferred from existing management or workshop documents for the community
Yes, they were characterized through informal conversations with stakeholders
Yes, they were characterized through a formal stakeholder engagement process, such as workshops, surveys,
or structured interviews
Specify your own value:
If YES, please list any approaches, software, or tools that were developed, used, or adapted for
the study to characterize stakeholder goals or measures of success. Please provide key citations.
10. If stakeholder goals were characterized for the study, why was this done? Choose all that apply.
No, they weren't
To assist the community in identifying their goals
To assist the community in developing indicators or measures of success
To assist the community in identifying goal-oriented decision options
To inform or train the community on a concept, process, or tool
To help the ORD study and/or the community define sustainability
To identify key research needs to be carried out by the study
106
-------
To identify key modeling endpoints for the study
Specify your own value:
11. Were one or more decision options identified for this study? Choose all that apply.
r~ NO
No, they already existed or the effort was led by the community or other group, and it was used but not
developed as part of the ORD study
Yes, they were inferred from existing management or workshop documents for the community
Yes, they were characterized through informal conversations with stakeholders
Yes, they were characterized through a formal stakeholder engagement process, such as workshops, surveys,
or structured interviews
Specify your own value:
If YES, please list any approaches, software, or tools that were developed, used, or adapted for
the study to characterize decision options. Please provide key citations.
12. Were any approaches, software, or tools used to qualitatively or quantitatively COMPARE the
alternative decision options identified in the
previous question? Choose all that apply.
r~ NO
No, they already existed or the effort was led by the community or other group but the results were used by
the ORD study
Yes, ORD developed, used, or adapted approaches, software, or tools to compare alternative scenarios
Specify your own value:
If YES, please list any approaches, software, or tools used to qualitatively or quantitatively
compare alternative decision scenarios, rank outcomes, or assess tradeoffs.
107
-------
13. Did this case study involve the development or use of one or more models (i.e., qualitative or
quantitative estimation tools) to ESTIMATE the outcomes of the decision options?
r Yes
r No
14. Please specify how the tool(s) was selected or developed. Choose all that apply.
An existing quantitative tool(s) was used with changes limited to either input or model parameters
An existing quantitative tool(s) was adapted to the needs of this place-based study involving changes in the
structure or output of the tool(s)
One or more existing tool(s) were adapted for this place-based study in such a way that a new independent
tool was developed
A new quantitative tool(s) or model(s) was developed for this place-based study
Specify your own value:
15. Please identify the type and name of each model/tool developed or used. If more than
one, please number them and continue to use these numbers for questions 15-23.
(Example: 1. Bayesian Belief Network describing sediment impacts on reservoirs; 2. Spatially
explicit dynamic models of landscape/seascape change; 3. Etc...)
16. Please identify a point of contact for each model, who is able to describe its development or
use within the context of your study. (If outside EPA, please include contact information.)
17. Please specify how each tool(s) was selected or developed.
18. Please list the data input needs of each model or tool.
(Example: 1. coral cover, algal cover, fish biomass...; 2. land use cover; precipitation events; soil type...; 3. Etc...)
19. Please list the data output provided by each model or tool.
20. Was output uncertainty estimated for each/any model?
r
Yes
(
No
108
-------
If YES, please describe for each model or tool.
21. Are any of the models or tools publicly available?
Yes
No
If YES, please describe for each tool or model.
22. Is the development and use of the model(s) documented?
Yes
No
If YES, please provide a document name(s) and sources(s).
23. Do you consider the tool(s) used in this place-based study transferable to other places and issues?
r Yes
No
Please comment briefly on tool transferability.
24. Was tool transferability a goal of your place-based study?
r Yes
No
25. What beneficiaries (e.g., anglers, irrigators) of ecosystem goods and services did you identify in
your study? Please list all.
26. Briefly describe the process or method you used to identify these beneficiaries. If you did not
use a process or method for identifying, please note N/A.
27. What ecological endpoints did you use relative to your beneficiary list (e.g., "species
presence and abundance" for "anglers")? Please be specific about the endpoint or provide
citations that describe them (e.g. "wetland habitat" is the number of acres of vegetated
wetland using the definition in xxx). If you did not identify the relationship between ecological
endpoints and beneficiaries, please note N/A.
28. What are the temporal and spatial dimensions of your endpoints (e.g., long-term average
biomass of commercial fish in a grab sample taken at a single point once a year).
109
-------
29. What was your rationale and foundation for linking the indicator used to the beneficiary
considered? (E.g., qualitative research, literature review, or expert judgment).
30. How were the final ecosystem goods and services identified for use in this study
determined? Choose all that apply.
Stakeholder engagement
Local expert consultation
Literature reviews
Other (including social media, news sources, peer groups)
31. Please rate how important each of the following methods of identification were when
determining the final ecosystem goods and services used in the study.
Stakeholder engagement
f~
M f~»+ rt+ nil I *vi t-\ f~iv-f rt 1-1+
Not at all important
("5
Unimportant
r*-
Neither important nor unimportant
Important
r*-
Extremely Important
Local expert consultation
Not at all important
Unimportant
/~
Neither important nor unimportant
Important
Extremely Important
Literature reviews
110
-------
Not at all important
p
Unimportant
Neither important nor unimportant
p
Important
p
Extremely important
Other (including social media, news sources, peer groups)
P
Not at all important
p
Unimportant
Neither important nor unimportant
p
Important
Extremely important
32. Did stakeholders appear to struggle with the concept of benefits?
r Yes
No
P
Specify your own value:
If YES, please explain what steps were taken to increase comprehension of this concept.
33. In discussions with stakeholders, did you use any of the following terminologies to
describe "natural resources"? Choose all that apply.
Ecosystem services
Earth's capital
Natural life support
Nature's benefits
Nature's value
(Final) environmental goods and services
111
-------
Natural capital
Environmental value
Specify your own value:
34. (If applicable) Out of all of the terminology listed above, please indicate which of the above
phrases you used most frequently within stakeholder discussions. Please limit your choice to
three phrases.
Ecosystem services
Earth's capital
Natural life support
Nature's benefits
Nature's value
(Final) environmental goods and services
Natural capital
Environmental value
Specify your own value:
35. Did stakeholders connect human health and/or well-being with one or more ecosystem service?
r Yes
No
If YES, please explain.
36. How was "success" measured in line with project outcomes? Choose all that apply.
Stakeholder engagement
Creating awareness about environmental/health issues within local communities
Building decision support networks for further research
Measured change in ecological integrity
112
-------
Measured enhancement in human well-being
Policy creation/implementation
Publications/Dissemination of results
Specify your own value:
37. Were any of these tools used to measure changes in provisioning in ecosystem services,
such as a result of proposed management actions? Choose all that apply.
Envision
l~ InVEST
l~ ARIES
r LUCI
r EPFs
Specify your own value:
38. Were any of these tools used to measure changes in human health, as a result of proposed management
actions? Choose all that apply.
Health impact assessment
Eco-health relationship browser
Risk assessments
Epidemiology studies
Review of clinical case studies
Specify your own value:
39. If you had to do your study all over, what would you do differently this time? What worked
well the first time that you would likely do again?
113
-------
Appendix B - List of PEGS Beneficiaries Compiled for Participating Place-Based Studies
TEV FEGS-CS FEGS-CSSub- Yee, Guanica, Suffolk Co., New Tampa, FL Dania Beach, Southern Blue River Lawrence, Narragesett GMeCCS
Categories Categories Categories PR NY Bedford, FL Willamette Watershed, MA Bay, Rl communities
MA Valley, OR OR
o>
ro
o>
on
^
i .
o
o>
00.01
Agricultural
00.02
Commercial /
Industrial
1 Irrigators NA NA Irrigators NA NA
_. CAFO
2
Operators
_. Livestock
3
Grazers
. Agricultural
Processors
5 Aquaculturists
6 Farmers Coffee Farmers Farmers Farmers
Timber
/ Foresters
Producers
1 Food Extractors
Timber, Fiber,
^ and
Ornamental
Extractor
3 Industrial
Processors
,, Industrial
A
Dischargers
Electric and
5 other Energy
Generation
Dive/Snorkeling
Boat Operators
Resource- /<-,,•
r Dive/Snorkelmg
h Dependent
Shop Operators
Businesses
Tourism
Industry
114
-------
Pharmaceutical
_ and Food
Supplement
Suppliers
Fur/Hide
8 Trappers and
Hunters
Trapping
00.03
Government,
Municipal, and
Residential
Municipal
1 Drinking Water
Plant Operators
Waste Water
2 Treatment
Plant Operators
,. Military/Coast
Guard
Municipal
Water Suppliers
Municipalities
00.05
Subsistence
1 Water
Subsisters
2 Food Subsisters
Timber, Fiber,
3 and Fur/ Hide
Subsisters
Building
4 Material
Subsisters
Groundwater
Extractors
Well water
drinkers
Artisanal
fishermen
Tribes
Food Pickers
and Gatherers
00.06
Recreational
Hunters
Recreational
hunters
Leisure Deer
hunting
Leisure
Waterfowl
hunting
Hunters
115
-------
o>
_3
(T3
>
O>
on
=>
+J
o
OJ
^
c
00.07
Inspirational
00.08 Learning
00.02
Commercial /
Industrial
Angling from
a boat
. Recreational Recreational Angling from
4 Anglers . Anglers Anglers
fishermen anglers shore or pier
Recreational
anglers
Waders.
r- . , Boaters.
S Swimmers, and
Swimmers
Divers
Spiritual and
Ceremonial .
. . Native
. Participants .
1 , American Tribes
and .
Spirituality
Participants of
Celebration
2 Artists Artists
1 Educators and
Students
2 Researchers
Electric and
5 other Energy Hydroelectricity
Generation
116
-------
00.03
Government,
Municipal, and
Residential
00.04
Commercial /
Military
Transportation
00.06
Recreational
Resource-
6 Dependent
Businesses
Residential
3 Property
Owners
. Military /Coast
Guard
1 Transporters of
Goods
_ Transporters of
People
,. Experiencers
and Viewers
4 Anglers
Waders,
5 Swimmers, and
Diver
6 Boaters
Dive/Snorkeling
Boat Operators
Dive/Snorkeling
Shop Operators
Tourism
Industry
Local Residents
Bird Watchers
Ecotourism
Campers
Hikers
Recreational
fishermen
Beachgoers
Kayakers
Home Property
Owners Owners
Shipping
Seaplane
Operations
Landscape
viewing
Non-
motorized
park and
trail
recreation
Angling from
a boat
Recreational Angling from
anglers shore or pier
Recreational
anglers
Beach
recreation
Recreational
waders/
swimmers
and divers
Power
Boating
Power Boaters
Cruising
Sailing
117
-------
00.07
Inspirational
00.08 Learning
O>
"ro
C 00.09 Non-Use
O
Q.
0
O>
(J
^ OJ
O> _E
•4-- m 00.09 Non-Use
to .i"
'x ^
LU
00.10
Humanity
Other
Beneficiaries
Listed
Spiritual and
Ceremonial
1 Participants
1 and
Participants of
Celebration
2 Artists Artists
1 Educators and
Students
2 Researchers
Native
American
Spirituality
People Who
2 Care (Option /
Bequest)
. People Who
Care (Existence)
1 All Humanity
Anyone who
has concerns
People who about the
care health of the
Chesapeake
Bay
People
Breathing
, Anyone at
Other .
Bald Eagles drinking Flood control ..
6 6 climate
water users
change
Global
Population , , Consumers of
„ , , Colonial
affected by forest and
,-,• t Waterbirds
Climate food products
Change
Dredged Global climate
Sand beneficiaries
118
-------
Process for
identifying
Beneficiaries
Measured
endpoints in
study
Endpoint
Selection
rationale
Focus Groups
Workshops
Interviews
Content
Analysis
Surrogate
metrics
Broad goals
specified by
stakeholders;
often
intermediate
measures
Esocid fishes
Lake
sturgeon
Riparian and
semi-aquatic
wildlife
Shoreline
Protection
Wild rice
Walleye
Stakeholder
MA MA Drawn from
engagement NA NA
FEGS-CS
Lit review
Many
intermediate
measures,
e.g. habitat,
floodwater
retention, Presence of
NA denitrificatio service in
n rates, pixels
carbon
sequestration
Some
possible PEGS
Metrics
Economic Expert
Valuation judgment
Methods and
from extensive
literature local
and experience
model applied to
review FEGS-CS
Stakeholder-
defined from
previous
work
Nitrate
leaching
from farm
fields
Regulatory-
the drinking
watsr
standard of
10 mg/L
nitrate-N
and the state
of Oregon's
action level
of 7 mg/L
nitrate-N.
Discussions
with USFS
stakeholders,
region staff,
and general
knowledge
gleaned from
scientific and
news sources
Used common
sense
approach
ratherthan a
formal means
of quantifying
this
relationships
(as through
HWBI or
similartools)
All of these:
qualitative
research,
literature
review, expert
judgment of
clients,
stakeholders
and EPA
Regional staff
?Hal
Walker's Review of
answers economic
don't seem literature
to address and models
the questions
Hunting
benefits,
Check 3vs flood risk
project for mitigation,
details air quality,
brook trout
habitat
Economic
analysis
119
-------
How
determined
Stakeholder
engagement-
Stakeholder Local Expert Literature Local expert Stakeholder
engagement Consultation Review consultation; engagement
Literature
reviews
Stakeholder
engagement,
local expert
consultations,
literature
reviews, other
(including
social media,
news sources,
peer groups)
Stakeholder
engagement,
local expert
consultation,
and
literature
reviews
Literature
reviews
120
-------
Appendix C - Models Used or Described by Place-Based Studies
The following table provides details on ecological production functions (EPFs) and decision support systems (DSS) used, or intended for use,
in the PBSs as listed in Table 4.1. Information is based either on questionnaire responses provided by PBS leads, on examination of
references provided by PBS leads, or on existing information in the literature about models identified by the PBS leads. Information is color-
coded by PBS location; within a given PBS, if there are multiple rows, color is alternated by row with no-color so that rows can be more
easily followed. The code "ins-inf' means that insufficient information was available. For full references of studies cited, see Chapter 4
reference list.
Study
Location
Guanica Bay,
Model name or
description
Bayesian Belief
Network
describing
sediment
impacts on
reservoirs
Bayesian Belief
Network
describing
impacts of
decisions on
coral reef ES
Spatially
explicit
dynamic
models of
landscape/
seascape
change:
Envision with
CORSET (coral
reef dynamic
model plug-in)
and coral reef
ecosystem
services plug-in
EPFs linking
coral condition
to coral reef ES
Modeling
completion or
documentation
Published EPA
report
Published
journal article
Draft journal
article
Published
journal article
and draft
documentation
Citation or
URL
Bousquin et
al.2014
Rehretal.
2014
Yee et al. in
revision
Yee et al.
2014;
Orlando et
al., in review
Model outputs
Lucchetti Reservoir
life expectancy
[yrs]
Benefits
Simplified Reef
Health Index
(SIRHI)
PEGS?
Yes
No
-
Yes
Model new
or existing?
Newly
developed
for this
study
Newly
developed
for this
study
Application
of existing
model
Existing
(Turneret
al. 2000
cited in Yee
et al.2014)
Model
computational
approach
Bayesian belief
network;
Bayesian belief
network;
stochastic
Analytic;
deterministic;
statistical
Spatial
extent
modeled
45.09 kmA2
Coral reef
tract of
Puerto Rico is
"2,000 kmA2
Area of
Puerto Rico =
9,100 kmA2
Ins-inf
Spatial
grain type;
size
Ins-inf; Ins-
inf
EPF not
spatially
explicit
Grid; 500 m
x 500 m
[2 km x 2
km reef
cell]
Grid; 50 m
x50m
Input data
typical
availability
Some locally
collected, not
available
everywhere
Some locally
collected, not
available
everywhere
Generally
available
Summary of data
input needs
Land use landcover;
precipitation events;
soil type; slope;
water storage
capacity of
reservoirs; water
outflow from
reservoirs
User defined
uncertainties (expert
opinion) on effects
of decisions on coral
reefs & eco- services
Coral cover, algal
cover, fish biomass;
historic frequency of
hurricane &
bleaching events;
sediment & nutrient
distribution offshore;
fishing pressure
Total no.
predictors
9
10
Ins-inf
5
Management
actions model
could directly
represent
Dredging sediment
from the reservoir;
altering coffee
farming practices
Establishing marine
protected areas
(MPA); Restoring
the lagoon;
Reducing loadings
from agriculture;
from sewage or
from development
(each by 40%, 70%,
90%)
Fishing pressure
Does not directly
model
management
actions (except via
Output
uncertainty
estimated?
Bayesian belief
networks describe
the probabilities of
outcomes, based
on probabilities of
precipitation
events or the
likelihood of
decisions being
implemented
Bayesian belief
networks describe
the uncertainty in
expert opinion -
probabilities of
certain events
occurring
Variability in model
predictions based
on uncertainty in
model parameters.
as well as
stochastic
hurricane and
bleaching events
Model
transferable?
Yes, Almost all
were re-
modifications, or
applications of
modelsortools
that had
previously been
applied in other
locations, and
would be
expected to be
highly
transferable
121
-------
Study
Location
Model name or
description
Modeling
completion or
documentation
Citation or
URL
Model outputs
State of the Reef
Index (SOR)
Wave energy
dissipation
Coral Reef
Protection Index
Wave height
attenuation [%]
Wave energy
attenuation [%]
Wave energy
attenuation [%] by
reef type
Wave energy
attenuation [%] by
reef height
Decrease in
erosion [%]
Decrease in wave
runup [%]
PEGS?
Yes
No
No
No
No
No
No
No
No
Model new
or existing?
Existing
(van
Beukering
and Cesar
2004 cited
in Yee et al.
2014)
Existing
(Mumby et
al. 2008
cited in Yee
etal.2014)
Existing
(Burke et al.
2008 cited
in Yee et al.
2014)
Existing
(Sheppard
et al. 2005
cited in Yee
etal.2014)
Existing
(Sheppard
et al. 2005
cited in Yee
etal.2014)
Existing
(Sheppard
et al. 2005
& van
Beukering
etal.
2011 cited
in Yee et al.
2014)
Existing
(Sheppard
et al. 2005
& Lowe
2005 cited
in Yee et al.
2014)
Existing
(CETS 1987
& Dean and
Galvin 1976
cited in Yee
etal.2014)
Existing
(FEMA2007
& Hughes
2004 cited
Model
computational
approach
Spatial
extent
modeled
Spatial
grain type;
size
Input data
typical
availability
Summary of data
input needs
Coral cover, algal
cover, fish biomass;
historic frequency of
hurricane &
bleaching events;
sediment & nutrient
distribution offshore;
fishing pressure
Total no.
predictors
10
3
4
5
3
6
3
6
3
Management
actions model
could directly
represent
linkage to other
models)
Output
uncertainty
estimated?
Model
transferable?
122
-------
Study
Location
Model name or
description
Modeling
completion or
documentation
Citation or
URL
Model outputs
Total ease of
access for
recreation
Total sand
generation
Total density of
Nassau grouper.
Epinephelus
stnatus
Total opportunity
forsnorkeling
Value of a dive
Visitation to dive
sites [boats siteA-l]
Visitation to dive
sites with
topography [boats
siteA-l]
Visitation to dive
sites with
topography [boats
siteA-l]
Total density of
Caribbean spiny
lobster, Panulirus
argus
Total density of
conch, Strombus
gigas
PEGS?
Yes
No
Yes
Yes
No
No
No
No
Yes
Yes
Model new
or existing?
in Yee et al.
2014)
Existing
(Mum by et
al. 2008
cited in Yee
etal.2014)
Existing
(Mum by et
al. 2008
cited in Yee
etal.2014)
Existing
(Mum by et
al. 2008
cited in Yee
etal.2014)
Existing
(Mumby et
al. 2008
cited in Yee
etal.2014)
Existing
(Wielguset
al. 2003
cited in Yee
etal.2014)
Existing
(Pendleton
1994 cited
in Yee et al.
2014)
Existin
(Pendleton
1994 cited
in Yee et al.
2014)
Existin
(Pendleton
1994 cited
in Yee et al.
2014)
Existing
(Mumby et
al. 2008
cited in Yee
etal.2014)
Existing
(Mumby et
al. 2008
Model
computational
approach
Spatial
extent
modeled
Spatial
grain type;
size
Input data
typical
availability
Summary of data
input needs
Total no.
predictors
3
3
3
3
6
5
7
7
3
3
Management
actions model
could directly
represent
Output
uncertainty
estimated?
Model
transferable?
123
-------
Study
Location
Model name or
description
EPFs linking
terrestrial
environment to
ecosystem
services
Modeling
completion or
documentation
Draft
manuscript-
work
progressing
Citation or
URL
Smith etal. in
review
Model outputs
Overall presence of
food algae
Eucheuma spp.
Total value of
finfish
Total production of
curios and jewelry
Mangrove
connectivity
Total key
commercial fish
biomass
Bioprospecting
potential
Sponge richness
Total
pharmaceutical
products
Dive site
favorability
Relevant pollutant
removal
Total
denitrification
PEGS?
Yes
No
No
Yes
Yes
Yes
Yes
Yes
No
No
Model new
or existing?
cited in Yee
etal. 2014)
Existing
(Mumby et
al. 2008
cited in Yee
etal. 2014)
Existing
(Mumby et
al. 2008
cited in Yee
etal. 2014)
Existing
(Mumby et
al. 2006
cited in Yee
etal. 2014)
Existing
(HRI2010
cited in Yee
etal. 2014)
Existing
(Principe et
al.2012
cited in Yee
etal. 2014)
Existing
(Hunt and
Vincent
2006 cited
in Yee et al.
2014)
Existing
(Mumby et
al. 2008
cited in Yee
etal. 2014)
Newly
created for
this study
Application
of off- shelf
model
Model
computational
approach
Analytic;
deterministic
Spatial
extent
modeled
Ins-lnf
Spatial
grain type;
size
Grid; 100 m
x 100 m
Grid; 30 m
x 30 m
Input data
typical
availability
Some locally
collected, not
available
everywhere
Generally
available
Summary of data
input needs
Landuse landcover
maps; soil types;
slope; precipitation;
management
practices on
landtypes
Total no.
predictors
3
3
3
2
10
2
3
2
5
3
Management
actions model
could directly
represent
LULC, land
management
practices
LULC, land
management
practices
Output
uncertainty
estimated?
Model does not
represent
uncertainty
Model does not
represent
uncertainty
Model
transferable?
124
-------
Study
Location
County, Long
Model name or
description
Hydrologic
models
describing
sediment
distribution in
Guanica
Watershed
Biological
condition
gradient
qualitatively
describing
thresholds of
coral condition
This is an
ongoing study
so all
tools/models
and
approaches
Modeling
completion or
documentation
Published
report
Published
report
Planned
Citation or
URL
Yuanetal.
2012; Hu et
al. submitted
Bradley et al.
2014
Not yet
available
Model outputs
Water yield
(average runoff)
retention [mmA3
mmA-2]
Nutrient retention
Sediment
Measure of soil
quality and
agricultural
productivity; urban
carbon
Important, prime,
or potentially
prime farmland
area [%]
Measure of forest's
importance;
threatened and
endangered
Sediment yield to
stream network
[mt]; sediment
deposition;
Sediment
degradation; final
amount of
Biological condition
level
Ins-lnf
PEGS?
No
No
No
No
Yes
Yes
Yes
No
Yes
-
Model new
or existing?
Application
of off- shelf
model
Application
of off- shelf
model
Ins-lnf
Model
computational
approach
Numeric;
deterministic
Qualitative;
analytic;
deterministic
Ins-lnf
Spatial
extent
modeled
10 - 100
kmA2
[Yahuecas
and Adjuntas
Watersheds]
Ins-lnf
Ins-lnf
Spatial
grain type;
size
Catchment-
Catchment
Grid; 30 m
Grid; 30 m
x 30 m
Grid; 30 m
x 30 m
Hydrologic
response
u n its;
Variable
EPA coral
reef
stations;
EPA coral
Ins-lnf; Ins-
lnf
Input data
typical
availability
Generally
available
Some locally
collected, not
available
everywhere
Ins-lnf
Summary of data
input needs
Landuse landcover
maps; soil types;
slope; precipitation;
management
practices on
landtypes
coral reef condition
attributes (rugosity,
cover, abundance
diversity, fish
biomass, fish
Ins-lnf
Total no.
predictors
9
5
6
8
11
6
8
19
9
Ins-lnf
Management
actions model
could directly
represent
LULC.Iand
management
practices
LULC.Iand
management
practices
LULC.Iand
management
practices
LULC.Iand
management
practices
LULC.Iand
management
practices
LULC.Iand
management
practices
LULC.Iand
management
practices
Scenarios: All land
changed to: 1)
forest; 2) grass; 3)
coffee. Sub-
scenarios: 1) All
coffee is sun-
grown; 2) All coffee
is shade- grown; 3)
No coffee is grown
Does not directly
model
management
actions
Ins-lnf
Output
uncertainty
estimated?
Model does not
represent
uncertainty
represent
uncertainty
Model does not
represent
uncertainty
Model does not
represent
uncertainty
Model does not
represent
uncertainty
Model does not
represent
uncertainty
Model does not
represent
uncertainty
Model does not
represent
uncertainty
Model does not
represent
uncertainty
Ins-lnf
Model
transferable?
Yes (no further
information)
125
-------
Study
Location
New Bedford
Tampa Bay
Dania Beach,
St. Louis River
Estuary Area
of Concern
Model name or
description
have not been
identified
Social network
analysis tool
and estimates
of the GHG and
job impacts of
some high level
scenarios
EPA H20
Scenario
analysis tool
Curve number
model
application
(linked to H20
model)
UFORE model
algorithm
(linked to H20
model)
Tidal exchange
model
developed in
SIMILE
Water Clarity
model
developed in
SIMILE
Habitat and
Recreational
Fishing
Opportunity
Model
Sediment and
development
after mangrove
restoration
model
Fishery
Production
model
DASEES
Nitrogen
processing
Modeling
completion or
documentation
Planned
Published EPA
report
Published
journal
manuscript
Draft
documentation
Draft
documentation
Draft
documentation
Journal
manuscript
submitted or in
Published
journal article
Published
journal article
Planned
Published
journal article
Citation or
URL
Not yet
available
EPA H20
Reistetter
and Russell
2011
Tool is
available at
i-Tree tools
Not yet
available
Not yet
available
Fulford et al.
2016
Osland et al.
2012
Jordan et al.
2012
Not yet
available
Bellinger et
al. 2014;
Model outputs
Ins-lnf
Water retention
value (feature);
usable water value
(nitrogen removal
value)
Composite curve
number
Carbon
sequestration, air
pollutant removal
Ins-lnf
Ins-lnf
Fish habitat
specific value
Soil bulk density;
percent sand;
percent soil
moisture; soil
organic matter; soil
submerged aquatic
vegetation habitat;
change in exvessel
value
Ins-lnf
Denitrification
cycling rate
PEGS?
Yes
No
No
No
Yes
No
No
No
No
Model new
or existing?
Application
of off- shelf
model
Newly
developed
Newly
developed
Application
of off- shelf
model
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Ins-lnf
Newly
developed
Model
computational
approach
Ins-lnf
Analytic;
conventional
(frequentist)
'
Analytic;
statistical;
deterministic
Numerical
simulation
Numerical
simulation
Numerical
simulation
conventional
(frequentist)
statistics;
.
conventional
(frequentist)
statistics;
Numeric;
conventional
(frequentist)
statistics;
Ins-lnf
Analytic;
Conventional
Spatial
extent
modeled
Ins-lnf
Ins-lnf
6342 kmA2
Ins-lnf
Ins-lnf
Ins-lnf
1000 - 10,000
kmA2
100 - 1000
kmA2
100 - 1000
kmA2
Ins-lnf
72 kmA2
Spatial
grain type;
size
Ins-lnf; Ins-
lnf
Grid; 30 m
x 30 m
Grid; 30 m
x 30 m
Topographi
c Index
block;
variable
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Grid; 1 km x
1km
Grid; 1 mx
1m
Grid; 55.2
kmA2
Ins-lnf; Ins-
Ins-lnf; Ins-
lnf
Input data
typical
availability
Ins-lnf
Generally
available
Generally
available
Some locally
collected, not
available
everywhere
Ins-lnf
Ins-lnf
collected, not
available
everywhere
Generally
available
Generally
available
Ins-lnf
Some locally
collected, not
Summary of data
input needs
Ins-lnf
Land use, canopy
cover, impervious
surfaces, soil type,
denitrification rates
per land use
Soil type, impervious
surface %, 2yr
average storm event
precipitation, land
use
Many- refer to
UFORE user manual
Many - refer to draft
publication
Many- refertodraft
publication
HSI by fish and
season for salinity,
temperature,
dissolved oxygen,
bottom type
Time since
restoration
Habitat specific
fishery production
rates, habitat areas
Ins-lnf
Bathymetry;
denitrification
predictive models
Total no.
predictors
Ins-lnf
40
6
34
Ins-lnf
Ins-lnf
10
1
12
Ins-lnf
12
Management
actions model
could directly
represent
Ins-lnf
LULC changes,
canopy cover
changes
LULC changes,
impervious cover
changes
LULC changes,
canopy cover
changes
Ins-lnf
Ins-lnf
Habitat (especially
SAV) area lessor
restoration
Mangrove
restoration
Habitat area loss or
restoration
Ins-lnf
"Using planned
bathymetric
alterations as
Output
uncertainty
estimated?
Ins-lnf
No model
uncertainty
analysis performed
Yes, model
uncertainty
analysis performed
Ins-lnf
Average error and
reliability index
were used during
calibration
Ins-lnf
No model
uncertainty
analysis performed
Yes, model
uncertainty
analysis performed
No model
uncertainty
analysis performed
Ins-lnf
Yes, The relative
uncertainty of the
map boundaries
Model
transferable?
Yes, SNA analysis
and high level
Yes, While many
of these tool
were applied in
the Tampa Bay
our best to use
nationally
available
datasets or
approaches that
could be applied
Yes
Yes, The
approach is
applicable to
126
-------
Study
Location
Model name or
description
Walleye
spawning
Esocid
spawning
Lake-sturgeon
spawning
Fishing from
boat or ice
Fishing from
shore
Power cruising
Sailing
Power boating
Human-
powered
boating
Modeling
completion or
documentation
Citation or
URL
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2013; Angradi
etal. 2016
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Model outputs
Walleye spawning
(present/absent);
total walleye
spawning area
[kmA2]
Esocid spawning
(present/absent);
total esocid
spawning area
[kmA2]
Lake-sturgeon
spawning
(present/absent);
total lake-sturgeon
spawning area
[kmA2]
Fishing from boat
(present/absent);
total fishing from
boat or ice area
[kmA2]
Fishing from shore
(present/absent);
totalfishingfrom
shore area [kmA2]
Power cruising
(present/absent);
total power
cruising area
[kmA2]
Sailing
(present/absent);
total sailing area
[kmA2]
Power boating
(present/absent);
total power
boating area
[kmA2]
Human-powered
boating
(present/absent);
total human-
powered boating
area [kmA2]
PEGS?
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Model new
or existing?
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Model
computational
approach
(frequentist)
statistics;
Presence-
absence
criterion
Analytic;
Conventional
(frequentist)
statistics;
Presence-
absence
Presence-
absence
en enon
Presence-
absence
criterion
Presence-
absence
criterion
absence
Presence-
absence
criterion
Presence-
absence
criterion
Spatial
extent
modeled
72 kmA2
72 kmA2
72 kmA2
72 kmA2
72 kmA2
72 kmA2
72 kmA2
72 kmA2
72 kmA2
Spatial
grain type;
size
Ins-lnf; Ins-
lnf
Grid; 10 m
x 10 m
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Input data
typical
availability
available
everywhere
Some locally
collected, not
available
everywhere
Some locally
collected, not
available
everywhere
Some locally
collected, not
available
everywhere
collected, not
available
everywhere
Some locally
collected, not
available
everywhere
Some locally
collected, not
available
everywhere
Some locally
collected, not
available
everywhere
Some locally
collected, not
available
everywhere
Some locally
collected, not
available
everywhere
Summary of data
input needs
Location of walleye;
walleye spawning
habitat
characteristics
Bathymetry;
submerged aquatic
vegetation predictive
models; emergy
aquatic vegetation
Location of lake
sturgeon spawning
habitats; Lake
sturgeon spawning
habitat
characteristics
Bathymetry;
restricted boating
areas
Bathymetry;
restricted boating
areas; developed or
traditional shoreline
fishing access point
locations; shoreline
land
Bathymetry;
restricted boating
areas; bridge
location
Bathymetry;
restricted boating
areas; bridge
location
Bathymetry;
restricted boating
areas; bridge
location
Bathymetry;
restricted boating
areas; slip locations
Total no.
predictors
3
3
4
2
3
3
4
3
4
Management
actions model
could directly
represent
scenarios... we
determined the
changes in
ecosystem service-
providing
area resulting from
each project.
...Three scenarios:
1) Current
conditions (no R2R
project); 2)
Predicted
distribution of
services after
that includes
created shoals; and
3) after habitat
Output
uncertainty
estimated?
for each mapped
final ecosystem
service was stated
in the manuscript
describing this
research
Model
transferable?
other Great
Lakes coastal
wetlands.
embayments
and tributaries,
but the details
applytothe St.
Louis River, with
the exception of
the Bald Eagle
habitat model
127
-------
Study
Location
Model name or
description
Beach
recreation
Park and trail
recreation
Aquatic-
terrestrial
connectivity
Residential
property views
Waterfowl
hunting
Semi-aquatic
fur-bearer
trapping
Deer hunting
Native-
American
spirituality
Modeling
completion or
documentation
Citation or
URL
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Model outputs
Beach recreation
(present/absent);
total beach
recreation area
[kmA2]
Park and trail
recreation
(present/absent);
total park and trail
recreation area
[kmA2]
Aquatic-terrestrial
connectivity
(present/absent);
total aquatic-
terrestrial
connectivity area
[kmA2]
Residential
property views
(present/absent);
total residential
property views
area [kmA2]
Waterfowl hunting
(present/absent);
total waterfowl
hunting area
[kmA2]
Semi-aquatic fur-
bearertrapping
(present/absent);
total semi-aquatic
fur- bearer
trapping area
[kmA2]
(present/absent);
total deer hunting
area [kmA2]
Native-American
spirituality
(present/absent);
total Native-
American
spirituality area
[kmA2]
PEGS?
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Model new
or existing?
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Model
computational
approach
Presence-
absence
criterion
Presence-
absence
criterion
Presence-
absence
criterion
absence
Presence-
absence
criterion
Presence-
absence
criterion
Presence-
absence
criterion
Presence-
absence
criterion
Spatial
extent
modeled
72 kmA2
72 kmA2
72 kmA2
72 kmA2
72 kmA2
72 kmA2
72 kmA2
72 kmA2
Spatial
grain type;
size
Ins-lnf; Ins-
Inf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Input data
typical
availability
Some locally
collected, not
available
everywhere
Some locally
collected, not
available
everywhere
Generally
available
Generally
available
Generally
available
Some locally
collected, not
available
everywhere
Generally
available
Some locally
collected, not
available
everywhere
Summary of data
input needs
Beach location
Land cover; park and
trail locations;
shoreline land
ownership
Shoreline type;
ecosystem services
of shorelines
Shoreline land
ownership; digital
surface model;
viewsheds
Land cover; shoreline
land ownership
Bathymetry; land
cover; shoreline land
ownership
Roads; shoreline
land ownership;
Duluth city archery
hunting areas
boundaries; Superior
archery hunting
areas; Wisconsin and
Minnesota hunting
regulations.
Spiritual areas;
spiritual area
locations
Total no.
predictors
2
3
1
1
2
3
2
1
Management
actions model
could directly
represent
Output
uncertainty
estimated?
Model
transferable?
128
-------
Study
Location
Model name or
description
Sand for
industrial reuse
Seaplane
operations
Protection
from fluvial
floods
Wave-energy
attenuation
Bald-eagle
nesting
Wild rice
harvesting
Waterbird
nesting
Shipping
Modeling
completion or
documentation
Citation or
URL
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016; ArcGIS-
bald eagle
Angradi et al.
2016
Angradi et al.
2016
Angradi et al.
2016
Model outputs
Sand for industrial
reuse
(present/absent);
total sand for
industrial reuse
area [kmA2]
Seaplane
operations
(present/absent);
total seaplane
operations area
[kmA2]
Protection from
fluvial floods
(present/absent);
total protection
from fluvial floods
area [kmA2]
Wave -energy
attenuation
(present/absent);
total wave-energy
attenuation area
[kmA2]
Bald eagle nesting
(present/absent);
total bald eagle
nesting area
[kmA2]
Wild rice
harvesting
(present/absent);
total wild rice
harvesting area
[kmA2]
Waterbird nesting
(present/absent);
total waterbird
[kmA2]
Shipping
(present/absent);
total shipping area
[kmA2]
PEGS?
Yes
Yes
No
No
Yes
Yes
Yes
Yes
Model new
or existing?
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Newly
developed
Model
computational
approach
Presence-
absence
criterion
absence
Presence-
absence
criterion
Presence-
absence
criterion
Fuzzy logic
absence
Presence-
absence
absence
Spatial
extent
modeled
72 kmA2
72 kmA2
72 kmA2
72 kmA2
72 kmA2
72 kmA2
72 kmA2
72 kmA2
Spatial
grain type;
size
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Input data
typical
availability
Some locally
collected, not
available
everywhere
Generally
available
Some locally
collected, not
available
everywhere
Some locally
collected, not
available
everywhere
Generally
available
Some locally
collected, not
available
everywhere
Generally
available
Some locally
collected, not
available
everywhere
Summary of data
input needs
Bathymetry;
Sediment
contamination
Runway boundaries
Bathymetry; channel
locations; land cover;
bridge location
vegetation;
submerged aquatic
vegetation predictive
models; depth
criteria
Forest tree cover,
distance from human
activity, distance to
Bathymetry;
traditional SLRE
harvest areas; wild-
rice depth range
waterbird services;
locations of nesting
habitat in SLRE
Bathymetry; slip
locations
Total no.
predictors
2
1
2
2
3
2
1
3
Management
actions model
could directly
Output
uncertainty
estimated?
Model
transferable?
129
-------
Study
Location
Southern
Willamette
Valley
Groundwater
Management
Area
S
Willamette
Valley
Taunton
Watershed
Blue River
Watershed,
Oregon
Lawrence,
MA
Model name or
description
USDA's APEX
model
(Agriculture
Policy
Extender)
USDA's
Nutrient
Tracking Tool
VELMA
(Visualizing
Ecosystem
Land
Management
Assessments).
VELMA
represents the
interaction of
hydrological
and
biogeochemical
processes in
watersheds,
enabling it to
simulate how
changes in
climate and
land use affect
the capacity of
ecosystems to
simultaneously
provide food
and fiber, clean
water, flood
control, climate
(GHG)
regulation.
sources and
sinks of
reactive
nitrogen, and
habitat for fish
and wildlife
EJSCREEN:
Environmental
Justice analysis
tool
Modeling
completion or
documentation
Validation and
testing ongoing
(in cooperation
with
Willamette
Partnership,
USDAARS
modelers and
USDA Office of
Environmental
Markets)
Planned
Modeling in
progress
Peer-reviewed,
published
Planned
Citation or
URL
Not yet
available
Not yet
available
Not yet
available
Abdelnour et
al. 2013;
Abdelnour et
al. 2011; Tool
is available at
EPA VELMA
Tool is
available at
EPA
EJSCREEN
Model outputs
Crop yield, N loss
Ins-lnf
Simulated
streamflow (total
catchment
discharge)
[mm/day];
greenhouse gas
regulation (C02,
N20, NOx); net
ecosystem carbon;
timber harvest (gC
m2/year) ; Losses
of dissolved
inorganic and
organic nitrogen
and carbon (g
m2/day); wildlife
habitat (forest age
class categories)
Demographic and
environmental
indicators
PEGS?
No
Yes
-
Model new
or existing?
off -she If
Ins-lnf
Application
with
substantial
adaptation
Application
of off- shelf
model
Model
computational
approach
Simulation
Ins-lnf
Numeric
systems-based
simulation;
Conventional
statistics;
deterministic
Analytic;
stochastic [?];
statistical
Spatial
extent
modeled
GW Mgmt
area is 600
kmA2
(unclear how
much will be
modeled)
Ins-lnf
Blue River
Watershed
area is 230
kmA2; Tool
development
scale is 0.1
Ins-lnf
Spatial
grain type;
size
By ag field?;
ag field?
Ins-lnf; Ins-
lnf
Grid; 30 m
x 30 m
Census
block group
Input data
typical
availability
Generally
available
Ins-lnf
Generally
available
Generally
available
Summary of data
input needs
Weather data, soils
data, crop practices,
management such as
irrigation, cover
crops, fertilizer rates,
timing
Ins-lnf
All or most data are
publically available:
-Climate drivers
(daily precip& avg
temp)
- Digital elevation
model (DEM)
- Land use and land
cover spatial data
- Calibration and
validation data for
streamflow, stream
chemistry.
vegetation and soils
Total no.
predictors
Ins-lnf
Ins-lnf
79
Management
actions model
could directly
represent
Agricultural
conservation
practices
Ins-lnf
Scenario drivers:
stand age; old-
growth (pre-
harvest),and
harvested
(post harvest)
Ins-lnf
Output
uncertainty
estimated?
Yes, Calibration
process using
existing data;
validation step is
forthcoming in
future work
Yes, Calibration
process using
existing data;
validation step is
forthcoming in
future work
Ins-lnf
Yes, but in a
rudimentary way
based on estimates
of the sources of
uncertainty in
calibration data.
Methods for
quantifying, and
controlling for,
model uncertainty
has since improved
with incorporation
of a genetic
algorithm for
calibratingthe
model.
Yes (no further
information)
Model
transferable?
Yes, This is the
hope and the
reason for the
design isthat the
tool (NTT) and
approach be
used in other
agricultural
communities
Ins-lnf
Yes, VELMA is
currently being
used or
considered for
use by other
groups at other
Yes
130
-------
Study
Location
Nine
municipalities
: Pascagoula,
MS
Narragansett
Bay, Rl
Model name or
description
CREAT: Climate
Resilience
Evaluation and
Awareness Tool
Real-time
water quality
now casting for
river system (to
be developed)
WTP-ccam for
drinking water
treatment
plant analysis
None used
Narragansett
3VS
Modeling
completion or
documentation
Planned
Planned
Planned
Modeling is not
planned
Modeling in
progress
Citation or
URL
Tool is
available at
EPA CREAT
Not yet
available
Not yet
available
Not yet
available
Model outputs
Lists of drinking
water and
wastewater utility
assets that climate
change could
Ins-lnf
Ins-lnf
Ins-lnf
PEGS?
No
Yes
-
Model new
or existing?
Application
of off- shelf
model
Ins-lnf
Ins-lnf
Application
of off- shelf
model
Model
computational
approach
Analytic;
statistical;
deterministic
Ins-lnf
Ins-lnf
Ins-lnf
Spatial
extent
modeled
Ins-lnf
Ins-lnf
Ins-lnf
Narragansett
Bay
Watershed =
4543 kmA2;
Narragansett
Bay = 380
kmA2
Spatial
grain type;
size
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Ins-lnf; Ins-
lnf
Input data
typical
availability
Ins-lnf
Ins-lnf
Ins-lnf
Ins-lnf
Summary of data
input needs
Ins-lnf
Ins-lnf
Ins-lnf
Information on
nitrogen loading
from a variety of
sources, patterns of
dispersal and
nitrogen removal in
Narragansett Bay,
expected response
relationships
associated with
changes in [TN], and
economic
information related
how a variety of
interventions to
reduce nutrient
loading, or enhance
sinks ( e.g. shellfish
filtration) could
affect aspects of the
bay's ecology
Total no.
predictors
Ins-lnf
Ins-lnf
Ins-lnf
Management
actions model
could directly
represent
Climate at local
scales to support
identification of
Ins-lnf
Ins-lnf
Output
uncertainty
estimated?
Yes (no further
information)
Yes (no further
information)
Ins-lnf
Model
transferable?
Yes
Yes
Yes
Yes
131
-------
Study
Location
Model name or
description
EPA Water
Quality Analysis
Simulation
Program
(WASP)
Environmental
Fluid Dynamics
Code (EFDC) -
Water Quality
EcoGEM [BH:
The doc listed
is for linking
ROMS to
EcoGEM]
Modeling
completion or
documentation
Modeling in
progress.
Although we
are mimicking
aspec so
surface [Chl-a]
and duration of
hypoxia in
bottom waters,
we have not
achieved a
calibration that
compares
favorably to
variation in
water column
diel cycles
Modeling in
progress; we
are still bogged
down in
aspects of
model
calibration and
validation
Modeling in
progress
Citation or
URL
Tool is
described at
EPA WASP
Tool is
described in
US EPA 2007;
approach is
described in
US EPA 2015
Approach is
described in
Krumholz et
al.2015
Model outputs
Water quality
metrics
The model focuses
primarily on
internal estuarine
nutrient and water
quality dynamics
Phytoplankton
reduction [%];
oxygen - alleviation
of hypoxia
[gO2 mA-3];
nitrogen reduction
[gN mA-3]
PEGS?
Yes
No
Yes
Model new
or existing?
Application
of off- shelf
model
Application
of off- shelf
model
Application
of off- shelf
model
Model
computational
approach
Numeric;
deterministic;
Analytic;
deterministic
Numeric;
deterministic;
statistical [?]
Spatial
extent
modeled
Unknown
Narragansett
Bay = 380
kmA2
Narragansett
Bay = 380
kmA2
Spatial
grain type;
size
Grid; Ins-lnf
Grid; Ins-lnf
Grid; Ins-lnf
Input data
typical
availability
Generally
available
Unknown
Some locally
collected, not
available
everywhere
Summary of data
input needs
WASP requires
physical forcing of
estuarine circulation.
and we are using the
EFDC modelforthis.
needs nutrient
loading from point
and non-point
sources, and a
number of choices
related to WASP
model set up, and
parameterization,
too numerousto list
here. For
Narragansett Bay,
model calibrations
can be compared to
data from moored
instrumentation.
that record a variety
of factors associated
with fluctuations in
surface and bottom
[DO]
EFDC -WQ. A series
of reports from
Mohamed
Abdelrhman
(NHEERLAED)
document data input
needs, and model
calibration
EcoGEM requires
extensive
hydrodynamic
calculations that
involved ROMS, and
which has not been
run for as many
years as EcoOBM
considered next.
(Uncertain whether
EcoGEM has run
nutrient reduction
scenarios)
Total no.
predictors
Management
actions model
could directly
represent
"This model helps
users interpret and
predict water
quality responses
to natural
phenomena and
man-made
pollution for
various pollution
management
decisions." (WASP
Fact Sheet)
Used to explore
effects of nutrient
reductions and
temperature
changes associated
with climate
change
Output
uncertainty
estimated?
Ins-lnf
Ins-lnf
Ins-lnf
Model
transferable?
Yes
Yes. "EFDC has
been applied to
over 100 water
bodies ... in
support of
environmental
assessment and
management and
regulatory
requirements"
(p. 3, EFDC User
Manual)
Yes
132
-------
Study
Location
Chesapeake
Bay
Model name or
description
EcoOBM
Economically-
based
optimization
model used for
scenario
analysis.
Spatially
detailed
information
was used.
Modeling
completion or
documentation
Modeling in
progress
Modeling
completed
Citation or
URL
Not yet
available
US EPA 2011
Model outputs
Plankton
community
respiration [mg Chi
mA-3 mA-l];
phytoplankton NPP
[mg C mA-2 mA-l];
carbon flux to
sediments [gC mA-
2 yA-l]'
denitrification;
surface layer Chl-a;
surface DIN; water
column respiration
[g C mA-2 mA-l];
sediment
respiration [g C
mA-2 dA-l]; surface
DO [mg IA-1] ;
bottom DO [mg IA-
1]; annual hypoxia
Index; and many
more [likely]
Annual N reduction
[million lbsyA-l];
annual P reduction
[million lbsyA-l];
annual sediment
reduction [billion
lbsyA 1]; Annual
control costs [$
millions yA-l];
bonus ecosystem
services [$ millions
yA- 1]; annual NET
costs [S millions
yA-l]; ecosystem
service indicators
(i.e., reductions in
nutrient and
sediment inputs.
water storage, fish
habitat, animal
habitat, waterfowl
habitat and GHG
mitigation)
PEGS?
No
Yes
Model new
or existing?
Application
of off- shelf
model
Newly
developed
for this
study
Model
computational
approach
Analytic [PL-
deterministic;
statistical
Analytic;
deterministic;
empirical/
statistical
Spatial
extent
modeled
Narragansett
Bay = 380
kmA2
Chesapeake
Bay
Watershed =
166,000
kmA2
Spatial
grain type;
size
Officer box
model
segments;
14
segments
or arr
Bay
Basin; Ins-
lnf
Input data
typical
availability
Some locally
collected, not
available
everywhere
Generally
available
Summary of data
input needs
EcoOBM involves the
same simplified
ecology equations
used in EcoGEM, but
with Officer Box
modeling used to
estimate water
column exchanges in
"14 segments of
Narragansett Bay.
Both EcoGEM and
EcoOBM model a
surface and bottom
layer in Narragansett
Bay. The EPA ORD
WASP and EFDC
modeling efforts
have not modeled
interannual
variations in [DO],
but EcoGEM and
EcoOMB has.
EcoOBM has
modeled over 10
years of interannual
variations in the
duration of hypoxia
recorded at moored
instrumentation sites
in the bay.
Not available
Total no.
predictors
Management
actions model
could directly
represent
Simulates
responses to
nutrient reductions
and climate change
Many scenarios are
listed, each with a
varying amount of
BMP transact ion
costs and
agricultural and
stormwater BMP
costs
Output
uncertainty
estimated?
Ins-lnf
Yes, sensitivity
analyses were used
for several highly
uncertain inputs
Model
transferable?
Yes
Ins-lnf
133
-------
Study
Location
Birmingham,
Model name or
description
National
Stormwater
Calculator
Modeling
completion or
documentation
Planned
Citation or
URL
Tool is
described at
National
Stormwater
Calculator
Model outputs
Average annual
runoff [in]; days
per year with
rainfall; days per
year with runoff-
days retained;
smallest rainfall w/
runoff [in]; largest
rainfall w/o runoff
retained [in]
PEGS?
Yes
Model new
or existing?
Application
of off- shelf
model
Model
computational
approach
Numeric;
deterministic;
statistical [?]
Spatial
extent
modeled
Ins-lnf
Spatial
grain type;
size
Site; Ins-lnf
Input data
typical
availability
Generally
Summary of data
input needs
Actual cost of green
Total no.
predictors
Ins-lnf
Management
actions model
could directly
represent
Model compares
scenarios to
baseline values to
differences in
runoff generation
Output
uncertainty
estimated?
Yes, per Users
Guide,
"...uncertainty
predicting future
climate
conditions..." is
captured by the
creation within
GREAT of" three
scenarios that span
results..."
Model
transferable?
Yes
134
-------
&EPA
United States
Environmental Protection
Agency
Office of Research and
Development (8ioiR)
Washington, DC 20460
EPA/600/R-16/136
------- |