AFPA EPA/600/R-17/266
uTtedOs August 2017
Environmental Protection Agency http://WWW.epa.gov/si
Practical Strategies for Integrating
Final Ecosystem Goods and Services into
Community Decision-Making
U.S. Environmental Protection Agency
Office of Research and Development
-------
EPA/600/R-17/266
August 2017
Practical Strategies for Integrating
Final Ecosystem Goods and Services into
Community Decision-Making
By
Susan Yee1, Justin Bousquin1, Randy Bruins2, Timothy J. Canfield3, Theodore H. DeWitt4,
Rebeca de Jesus-Crespo1, Brian Dyson5, Richard Fulford1, Matthew Harwell1, Joel Hoffman6,
Chanda J. Littles7, John M. Johnston8, Robert B. McKane9, Lauri Green7, Marc Russell1, Leah
Sharpe1, Nadia Seeteram10, Arik Tashie11, Kathleen Williams6
Office of Research and Development
U.S. Environmental Protection Agency
1. Gulf Ecology Division, National Health and Environmental Effects Research Laboratory, Gulf
Breeze, FL 32561
2. Systems Exposure Division, National Exposure Research Laboratory, Cincinnati, OH 45268
3. Ground Water and Ecosystems Restoration Division, National Risk Management Research
Laboratory, Ada, OK 74820
4. Western Ecology Division, National Health and Environmental Effects Research Laboratory,
Newport, OR 97365
5. Land and Materials Management Division, National Risk Management Research Laboratory,
Cincinnati, OH 45268
6. Mid-Continent Ecology Division, National Health and Environmental Effects Research Laboratory,
Duluth, MN 55804
7. Oak Ridge Institute for Science Education (ORISE) Postdoctoral Fellow, Western Ecology Division,
National Health and Environmental Effects Research Laboratory, Newport, OR 97365
8. Computational Exposure Division, National Exposure Research Laboratory, Athens, GA 30605
9. Western Ecology Division, National Health and Environmental Effects Research Laboratory,
Corvallis, OR 97333
10. ORISE Fellow, Computational Exposure Division, National Exposure Research Laboratory, Athens,
GA 30605
11. ORISE Fellow, Environmental and Public Health Division, National Health and Environmental
Effects Research Laboratory, Research Triangle Park, NC 27709
-------
Notice and Disclaimer
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 citation for this report is:
Yee, S., J. Bousquin, R. Bruins, T.J. Canfield, T.H. DeWitt, R. de Jesus-Crespo, B. Dyson, R. Fulford,
M. Harwell, J. Hoffman, C.J. Littles, J.M. Johnston, R.B. McKane, L. Green, M. Russell, L. Sharpe, N.
Seeteram, A. Tashie, and K. Williams. 2017. Practical Strategies for Integrating Final Ecosystem Goods
and Services into Community Decision-Making. U.S. Environmental Protection Agency, Gulf Breeze,
FL, EPA/600/R-17/266.
Acknowledgments
We greatly appreciate the efforts of reviewers who took the time to read the report: Cristina Carollo,
Bruce Duncan, Chris Kelble, and Paul Ringold. Chloe Jackson provided editorial and formatting
assistance.
Cover photo credits:
Lake Carl Blackwell, Oklahoma: https://www.flickr.com/photos/gmeador/7146586311
San Juan, Puerto Rico: U.S. EPA
Duluth, Minnesota: https://commons.wikimedia.Org/wiki/File:Lakewalk-Duluth-2006.jpg
Tillamook Coast, Oregon: https://www.flickr.com/photos/misserion/2411644789
Mobile Bay, Alabama: U.S. EPA
-------
Abbreviations and Symbols
Throughout this report, the term "ecosystem goods and services" is often abridged to "ecosystem
services", and may include either intermediate or final ecosystem goods and services.
This symbol is used throughout this report to highlight practical strategies for
integrating ecosystem goods and services into community decision-making.
Acronyms and abbreviations used in this report include the following. Additional words in the
abbreviated title, but not specified by the acronym letters, are given in brackets.
3VS
Triple Value Simulation
AM
Adaptive Management
AOC
Areas of Concern
ARIES
Artificial Intelligence for Ecosystem Services [Modeling System]
BBN
Bayesian Belief Network
BenMap
Benefits Mapping and Analysis Program
BRI
Benefit-Relevant Indicators
BUI
Beneficial Use Impairment
CDC
Centers for Disease Control and Prevention
CEQ
Council on Environmental Quality
CGEM
Coastal Generalized Ecosystem Model
CHSI
Community Health Status Indicators
CICES
Common International Classification of Ecosystem Services
CMAQ
Community Multi-scale Air Quality [Modeling System]
CMECS
Coastal and Marine Ecosystem Classification Standard
DASEES
Decision Analysis for Sustainable Economy, Environment, and Society
DCE
Discrete Choice Experiment
DPSIR
Drivers-Pressures-State-Impact-Response [Framework]
DSS
Decision Support System
EBF
Ecological Benefit Function
EGS
Ecosystem Goods and Services
EM
Ecological Models
EEA
European Environment Agency
EFDC
Environmental Fluid Dynamics Code [Modeling System]
ELAPP
Environmental Lands Acquisition and Protection Program
Envision
[An integrated modeling platform]
EPA
U.S. Environmental Protection Agency
EPA H20
[EPA's Ecosystem Services Scenario Mapping Tool]
EPF
Ecological Production Function
ESML
EcoService Models Library
iv
-------
FEGS
Final Ecosystem Goods and Services
FEGS-CS
Final Ecosystem Goods and Services - Classification System
FGDC
Federal Geographic Data Committee
GIS
Geographic Information System
GLWQA
Great Lakes Water Quality Agreement
HGS
Human Goods and Services
HIA
Health Impact Assessment
HT
Hypothesis Testing
HWBI
Human Weil-Being Index
IEGS
Intermediate Ecosystem Goods and Services
IL
Iterative Learning
InVEST
Integrated Valuation of Ecosystem Services [and Trade-offs]
i-Tree
[Tools for Assessing and Managing Forests and Community Trees]
MEA
Millennium Ecosystem Assessment
NAS
National Academy of Sciences
NEP
National Estuary Program
NESCS
National Ecosystem Services Classification System
NO A A
National Oceanographic and Atmospheric Administration
NRC
National Research Council
ORD
[EPA's] Office of Research and Development
PAC
Public Advisory Committee
PNW
Pacific Northwest
R2R2R
Remediation to Restoration to Revitalization
RBI
Rapid Benefits Indicators
SDM
Structured Decision Making
SPA
Service Providing Area
SWAT
Soil and Water Assessment Tool
TEV
Total Economic Value
USDA
U.S. Department of Agriculture
VELMA
Visualizing Ecosystem Land Management Assessments [Model]
VRD
Variable Relationship Diagram
WTP
Willingness to Pay
V
-------
Executive Summary
The concept of Final Ecosystem Goods and Services (FEGS) explicitly connects ecosystem
services to the people that benefit from them. This report presents a number of practical
strategies for incorporating FEGS, and more broadly ecosystem services, into the decision-
making process. Whether a decision process is in early or late stages, or whether a process
includes informal or formal decision analysis, there are multiple points where ecosystem services
concepts can be integrated.
This report uses Structured Decision Making (SDM) as an organizing framework to illustrate the
role ecosystem services can play in a values-focused decision-process, including:
Clarifying the decision context: Ecosystem services can help clarify the potential
impacts of an issue on natural resources together with their spatial and temporal extent
based on supply and delivery of those services, and help identify beneficiaries for
inclusion as stakeholders in the deliberative process.
Defining objectives and performance measures: Ecosystem services may directly
represent stakeholder objectives, or may be means toward achieving other objectives.
Creating alternatives: Ecosystem services can bring to light creative alternatives for
achieving other social, economic, health, or general well-being objectives.
Estimating consequences: Ecosystem services assessments can implement ecological
production functions (EPFs) and ecological benefits functions (EBFs) to link decision
alternatives to stakeholder objectives.
Considering trade-offs: The decision process should consider ecosystem services
objectives alongside other kinds of objectives (e.g., social, economic) that may or may
not be related to ecosystem conditions.
Implementing and monitoring: Monitoring after a decision is implemented can help
determine whether the incorporation of ecosystem services leads to measurable benefits,
or what levels of ecosystem function are needed for meaningful change. An evaluation of
impacts on ecosystem services from past decisions can provide a learning opportunity to
adapt future decisions.
Each chapter of this report details one of these steps, and each chapter is paired with a set of
appendices providing examples of tools and approaches that decision makers can use for that
step. This report also presents a number of case study examples that illustrate the ecosystem
services concepts, approaches, and tools for a variety of community decision processes, such as
resiliency planning or sustainability planning, watershed or coastal management, habitat
restoration, risk assessments, or environmental impact assessments. Advantages of integrating
ecosystem services concepts into community decision-making through values-focused thinking
include: improved information collection, improved communication, expanded stakeholder
engagement, creative development and evaluation of alternatives, interconnected decisions, and
strategic thinking.
vi
-------
Table of Contents
Front Matter
Notice and Disclaimer iii
Acknowledgments iii
Abbreviations and Symbols iv
Executive Summary vi
Table of Contents vii
List of Tables x
List of Figures xii
List of Strategies xvii
1. Introduction 1
1.1. Need for an Operational Framework 1
1.2. Integrating Final Ecosystem Goods and Services into Decision-Making 2
1.3. Benefits and Barriers of a PEGS Approach 3
1.4. Structured Decision Making Framework 6
1.5. Integrating Ecosystem Services Assessments within a Structured Decision Making
Framework 8
1.6. Flexibility to Work with Alternative or Existing Processes 9
1.7. Overview of this Report 13
2. Clarify the Decision Context 15
2.1. Clarifying the Decision Context 15
2.2. Incorporating Ecosystem Services into Decision Context Characterization 15
2.3. Elements of Decision Context 16
2.4. Establishing a Decision Context 17
2.5. Using Conceptual Models to Identify Ecosystem Services 20
3. Define Objectives and Performance Measures 23
3.1. Defining Objectives 23
3.2. Distinguishing Means Objectives from Fundamental Objectives 24
3.3. Performance Measures 25
3.4. Integrating Ecosystem Services when Decision Objectives are Pre-defined 28
4. Develop Alternatives 30
4.1. Creating Alternatives 30
4.2. Leveraging Ecosystem Services to Achieve Economic and Social Objectives 31
4.3. Integrating Ecosystem Services When Decision Alternatives are Pre-defined 33
5. Estimate Consequences 34
5.1. Estimating Effects of Decision Alternatives on Performance Measures 34
5.2. Conceptual Models as the Foundation for PEGS Mathematical Models 35
5.3. Ecological Production Functions 37
5.4. Ecological Benefits Functions 40
5.5. Data Availability and Model Transferability 43
5.6. Using Decision Support Systems to Integrate Information and Compare Scenarios 44
5.7. Comparing Scenarios with Consequence Tables 46
6. Evaluate Trade-offs and Take Action 48
vii
-------
6.1. Evaluating Trade-offs 48
6.2. Implementing a Decision 49
6.3. Monitoring the Outcomes and Adaptive Management 49
7. Applications using Case Studies 52
7.1. Case Study Examples 52
7.2. Strategies from Case Studies 52
8. Synthesis 57
8.1. Final Ecosystem Goods and Services Facilitate Values-focused Thinking 57
8.2. Stakeholder Engagement 59
8.3. Guiding Information Collection 59
8.4. Improving Communication 60
8.5. Alternative Development and Evaluation 61
8.6. Interconnecting Decisions and Guiding Strategic Thinking 61
8.7. Conclusions 62
References 63
Glossary 72
Appendix A. Approaches to the Decision Process 75
Al. Using a DASEES Approach 76
A2. Working within an Existing Process 80
Appendix B. Tools and Approaches for Clarifying the Decision Context 82
Bl. Using a DASEES Approach to Characterize and Understand Decision Context 83
B2. Applying FEGS-CS to Identify Key Beneficiary Groups 88
B3. First Steps toward Assessing Coastal FEGS Vulnerabilities from Environmental Change 91
B4. Working with Stakeholders to Build Conceptual Models 94
Appendix C. Tools and Approaches for Objectives and Performance Measures 97
CI. Using a DASEES Approach to Develop Objectives Hierarchies 98
C2. Applying the FEGS-CS to Identify and Measure Ecosystem Services Objectives 101
C3. Using the EnviroAtlas to Identify and Map Potential Measures of FEGS 104
C4. Rapid Benefits Indicators (RBI) as Performance Measures 107
C5. Applying the HWBI Framework to Structure and Measure Community Objectives Ill
Appendix D. Tools and Approaches for Creating Alternatives 114
Dl. Developing Alternatives using Means-Ends Objectives Hierarchies in a DASEES
Approach 115
D2. Using Networks to Link Decision Alternatives to FEGS and Human Well-being 118
Appendix E. Tools and Approaches for Estimating Consequences 120
El. Modeling Ecosystem Services Production using EPFs 121
E2. Modeling Ecosystem Services Production using SPA 124
E3. Using the Modeling Tool VELMA to Compare Scenarios 126
E4. The EcoService Models Library (ESML) 129
E5. Non-Market Economic Valuation for Ecosystem Services 133
E6. Measuring Health Outcomes 136
E7. Health Impact Assessment and Ecosystem Services 140
E8. Estimating Consequences of Ecosystem Services Changes on Human Well-being 143
viii
-------
E9. Methodology to Assess whether to Transfer Existing Measurements or Models to New
Sites 147
E10. Evaluating Consequences of Decision Options using a DASEES Approach 151
Appendix F. Tools and Approaches to Evaluate Trade-offs and Monitor Outcomes 155
Fl. Measures Preferences and Trade-offs using a DASEES Approach 156
F2. Evaluating Trade-offs with Rapid Benefits Indicators (RBI) 160
F3. Working with Stakeholders to Weight Components of Well-being 163
F4. Using a DASEES Approach to Implement Adaptive Management and Define Triggers to
Monitor Achievement of Objectives 168
Appendix G. Case Studies 170
Gl. Great Lakes Areas of Concern Case Study 171
G2. Mobile Bay, Alabama Case Study 176
G3. Pacific Northwest Case Study 179
G4. San Juan, Puerto Rico Case Study 185
G5. Southern Plains, Oklahoma Case Study 189
ix
-------
List of Tables
Table 1.1. Examples of how FEGS-oriented questions can help bridge intermediate ecosystem
goods and services (IEGS) with PEGS (italics), the relevant environmental context
(underline), and their beneficiaries (bold) 4
Table 1.2. Examples of meeting/workshop planning schedules for integrating FEGS into decision
processes of varying budget or timeline (Gregory et al. 2012, Bradley et al. 2016) 10
Table 1.3. Comparison of a variety of other decision processes to SDM, illustrating where FEGS
concepts could be integrated 12
Table 2.1. Examples of FEGS related questions that could be asked to help clarify the decision
context 16
Table 2.2. Classes and sub-classes of Environments, Beneficiaries, and FEGS in the FEGS-CS
(from Landers and Nahlik 2013), and example questions that might help identify them 19
Table 3.1. Key types of objectives (Carriger and Benson 2012; Bradley et al. 2016). Note here
"maximize water quality" is a means to improving human health, however in other contexts
water quality may be a fundamental objective. Similarly, "maximize human well-being" here
is a strategic objective, but in other contexts it may be a fundamental objective 24
Table 3.2. Example of using HWBI conceptually in community-scale workshops to identify and
measure community goals (modified from Fulford et al. 2016b; Smith et al. 2012) 28
Table 5.1. Desired attributes of ecological production functions. Source: Bruins et al. (2017) 37
Table 5.2. Hypothetical example of a consequence table (from 3VS, Tenbrink et al. 2016) 46
Table 7.1. Description of major environmental issues, and ecosystem services and well-being
endpoints under consideration in each of the five case studies 53
Table 7.2. Summary of strategies being implemented in each case study to integrate ecosystem
services concepts within the steps of a community decision-making process (for details, see
Appendix G). Key tools used by each case are in bold 55
Table 8.1. Example tools and approaches from this report and how they could be used to integrate
FEGS into a decision process 57
Table 8.2. Example strategies from this report that can be used to engage stakeholders 59
Table 8.3. Example strategies from this report that can be used to guide information collection 60
Table 8.4. Example strategies from this report that can be used to improve communication 60
Table 8.5. Example strategies from this report that can be used to create and evaluate
alternatives 61
Table 8.6. Example strategies from this report that can be used to interconnect decisions and
guide strategic thinking 62
Appendix Tables
Table C4.1. Datasets used and their source in the example RBI analysis 110
Table E4.1. Key concepts used by ESML to assist the user in comparing and fitting models to an
intended use 130
Table E5.1. Descriptions of Non-Market Economic Valuation Methods 133
Table E5.2. Choice Card Example, modified from Johnston et al. 2011 134
Table E5.3. WTP results from Johnston et al. (2011). WTP estimates are per household, per year.
For all variables, except access, estimates represent WTP ($) for a one percent increase in that
variable. P-values are two-tailed for the null hypothesis of zero WTP 135
Table E6.1. Examples of potential steps in measuring community level health outcomes. Not all
steps need to be followed, and the list is not comprehensive 139
x
-------
Table E8.1. EPFs are applied to translate attributes of ecosystem condition into measures of
ecosystem services (Russell et al. 2013; Tallis et al. 2013; Smith et al. 2017) 144
Table F3.1. Results from mapping exercise used to connect stakeholder identified decision
opportunities in Vero Beach, FL to the domains of HWBI. Checks indicate participants
identified a given domain as relevant to why the decision opportunity was being proposed.
Decision opportunities most directly related to ecosystem services are in bold (from Fulford et
al. 2016) 165
Table F3.2. Percent of participants in three Vero Beach workshop exercises identifying a given
domain as important in the exercise mapping priorities, as a group by dot voting, and
individually by ranking. Top three domains by each method are in bold 166
Table Gl.l. Final ecosystem service by management action matrix for the St. Louis River Area of
Concern (AOC). Example actions anticipated or otherwise realistic projects for the AOC.
Expected effects of management actions on SPA of each FEGS are rows in the table; trade-
offs among FEGS resulting from habitat restoration or other management actions are in
columns. For riparian confined sediment disposal facility (CDF), the assumption is that no
aquatic habitat is lost. Responses: zero = no effect; + = more area of the service is created; - =
area of the service is lost; ?= response depends on context (Angradi et al. 2016) 174
Table G3.1. PNW Case Study stakeholders/partners and benefits addressed 181
xi
-------
List of Figures
Figure 1.1. PEGS based approach conceptual model of decision support 2
Figure 1.2. Illustration of the three elements needed to define PEGS 3
Figure 1.3. The steps in a decision analysis process (NRC 2011; Carriger & Benson 2012;
Gregory et al. 2012) 7
Figure 1.4. Points where FEGS concepts (white boxes) could be integrated into a decision process
(yellow boxes). FPF = ecological production function, EBF = ecological benefit function 8
Figure 2.1. Example questions asked during a generic decision sketching process (yellow boxes),
and where ecosystem services questions could be integrated (white boxes). Green arrows
illustrate the process can be highly iterative as information from later steps leads to previous
steps needing to be revisited 18
Figure 2.2. Questions to guide development of a DPSIR causal-chain conceptual model (Bradley
and Yee 2015) 20
Figure 2.3. Example conceptual model for conducting ecosystem services assessments (Bousquin
et al. 2015) 21
Figure 2.4. FEGS based approach conceptual model for decision support 21
Figure 4.1. Illustration of how production of Final Ecosystem Goods and Services can influence
economic production 31
Figure 4.2. Ecosystem services values typology within the TEV paradigm. Adapted from Pascual
et al. (2010) 32
Figure 4.3. Illustration of how ecosystem services, alongside social and economic services, can be
means to improving components of human well-being (from Smith et al. 2014) 33
Figure 5.1. Generic conceptual model showing the use of FEGS in decision support analyses for
community well-being. EGS = ecosystem goods and services 36
Figure 5.2. Generic depiction of a Decision Support System. Dashed lines emphasize the role of
scenario modeling 44
Figure 6.1. Important aspects of monitoring outcomes (dashed arrows) and AM principles of
iterative learning and hypothesis testing ("IL & HT"; dashed oval) mapped onto the FEGS-
based conceptual model for structured ecosystem-services decision-making 50
Figure 7.1. Locations of the five case studies 52
Appendix Figures
Figure Al.l. Illustration of a disorganized process, where all the pieces are present but do not fit
together 76
Figure A1.2. The five steps in the DASEES process, and benefits of a DASEES approach 77
Figure A1.3. Information describing the Guanica Bay decision context (from Bradley et al. 2016) 78
Figure A1.4. Objectives hierarchy elicited from stakeholder workshop discussions for the
Guanica Bay watershed (from Bradley et al. 2016). Values functions are for illustration only 78
Figure A1.5. Illustration of how a Bayesian Belief Network could be used to predict effects of
decision options on measured objectives (from Bradley et al. 2016) 79
Figure A2.1. Differences between the land and water sides of Areas of Concern with respect to
changes, policy mechanisms, results, and goals (from Williams and Hoffman 2017) 81
Figure B1.1. The components of the DPSIR framework 84
Figure B1.2. An example of a DPSIR created through Guanica Bay Puerto Rico workshop
discussions on impacts of agricultural practices on coral reefs (see Appendix Al). Red text
xii
-------
illustrates how objectives about ecosystem services benefits can be inferred from conceptual
model development (from Bradley et al. 2016) 84
Figure B1.3. Screenshot of SystemSketch, illustrating the category panel (left) which can be
browsed for over 800 distinct DPSIR nodes, and the links display (right), which visualizes the
causal linkages between nodes 85
Figure B1.4. Example social network analysis, illustrating well-connected and isolated groups of
decision-makers and stakeholders 86
Figure B1.5. The Decision Landscape (from Rehr et al. 2012) 86
Figure B2.1. Image of the St. Louis River Estuary National Water Trail Master Plan 88
Figure B3.1. Conceptual model demonstrating the link between coastal habitats and potential
FEGS delivery 91
Figure B3.2. Example chart of the type of output that will be generated once literature evidence
for FEGS beneficiary:habitat linkages is compiled 92
Figure B3.3. The map depicts actual habitat classes (from Clinton et al. 2007) for Tillamook Bay,
Oregon (top inset boxes), though beneficiary (FEGS) linkages (bottom left inset) are
completely hypothetical 93
Figure B4.1. Preliminary conceptual framework for the use of ecosystem service trade-off
analysis to support decision-making in an estuarine Great lakes AOC. Small arrows indicate
flow of decisions 95
Figure B4.2. Conceptual framework for the use of ecosystem service mapping and associated
analysis to support decision-making in an estuarine Great Lakes AOC (Angradi et al. 2016).
R2R = remediation to restoration; FES = final ecosystem services; BUI = beneficial use
impairment; AOC = area of concern; SPA = service providing area; SWF = social welfare
function (i.e., ecological benefit function [EBF]) 95
Figure Cl.l. Example of developing an objectives hierarchy for Guanica Bay watershed
management (see Appendix Al) to protect coral reefs (original from Carriger et al. 2013) 99
Figure C2.1. The structure of the new community organization in Duluth 101
Figure C2.2. Revealed values related to a trail project 102
Figure C3.1. General FEGS classification structure, from Landers and Nahlik 2013 104
Figure C3.2. Conceptual model showing production function linkages between IEGS, FEGS, and
HGS 105
Figure C3.3. An example of how the FEGS Environmental Sub-Classes in a study area could be
mapped and aggregated into a profile. From Landers and Nahlik, 2013 105
Figure C4.1. Benefits previously assessed using Rapid Benefit Indicators (Mazzotta et al. 2016) 107
Figure C4.2. Benefits can be received where EGS are produced (left) when EGS flow out to
people from where they are produced (middle) or when EGS flow through specific pathways
to reach people 108
Figure C4.3. Where a benefit is available (blue area), beneficiaries (houses) benefit less from that
service (green overlap area) if additional sources of the service are available (yellow area) 109
Figure C5.1. Community participants at workshop Ill
Figure C5.2. Flip chart illustrating use of colored dots to prioritize objectives 112
Figure C5.3. Community participants at workshop 113
Figure Dl.l. Example Means-Ends Network in DASEES illustrating how a specific management
option (dredging the reservoir) can be identified to accomplish a means objective (maximizing
reservoir volume) in support of a fundamental objective (maximizing power generation) 116
Figure D1.2. Specification of decision scenarios in DASEES as different combinations of
management options 116
xiii
-------
Figure D2.1. Diagram describing the eight domains of the Human Well-being Index (HWBI;
Smith etal. 2013) 118
Figure D2.2. Diagram showing linkages between decisions, ecosystem services, and domains of
human well-being (from HWBI). Example from text of investing in Greenspace is linked to
HWBI via the Social Cohesion domain. This is not the only connection between Greenspace
and HWBI and each one can be visualized in a similar way in the interactive tool 119
Figure El.l. Example of using the EPA H20 to generate a new scenario by changing the
highlighted row crop land parcel to freshwater marsh 121
Figure El.2. Example output from EPA H20 for the Hillsborough County, FL 122
Figure El.3. The EcoServices Model Library (ESML; Appendix E4) variable typology showing
that FEGS are the break point between EPFs and EBFs 123
Figure E2.1. Composite SPA map for the St. Louis River AOC showing the number of final
ecosystem services for each 100-m2 map pixel. Insets show riparian detail (from Angradi et
al. 2016) 124
Figure E3.1. VELMA conceptual model 126
Figure E3.2. FEGS based approach conceptual model 127
Figure E4.1. ESML is an online database that enables users to rapidly find and compare
information on >100 ecological models or model applications. Beta users may register at
https://esml.epa.gov/epf_l/public/signup 129
Figure E4.2. Data map for ESML. Circled numerals indicate the number of descriptors in the
database for each model aspect 130
Figure E4.3. ESML provides a variable relationship diagram (VRD) for each included model or
application. Pictured is the VRD for an application of the In VEST water provision model for a
river in Spain (Marques et al. 2013). PD = Predictor, Driver; PC = Predictor, Constant/Factor;
RC = Response, Computed. Arrows denote that one variable (or variables, if gathered within a
box) is required for computation of the other. Asterisk indicates that data for multiple runs of
this EM are present in ESML; the value of a variable that is marked with an asterisk changes
to define run conditions. Double dagger denotes a variable whose value is constant with
respect to a driving class variable (such as when derived from a lookup table) 131
Figure E6.1. The Eco-Health Browser visualizes linkages between ecosystems and human health,
and provides literature. Urban Ecosystems ~ Obesity, used in an example below, is
highlighted 136
Figure E6.2. Descriptive analysis of obesity trends in relation to tree cover at the neighborhood
scale. Data on tree cover comes from EnviroAtlas. Obesity data comes from the 500 Cities
Project 137
Figure E6.3. County level analysis. Data on Green Spaces comes from EnviroAtlas. Data on
Obesity and Physical Inactivity comes from CHSI 138
Figure E7.1. The six steps of a Health Impact Assessment Process, from Mecklenburg County,
NC Health Department 140
Figure E7.2. Illustration of how decisions can lead to health impacts by modifying the benefits
received from the natural environment 141
Figure E7.3. Simplified pathway diagram from HIA illustrating how adoption of decision
alternatives may lead to changes in human health (from Johnston et al. 2017) 141
Figure E8.1. Human Well-Being Index conceptual model (simplified from Summers et al. 2016).
Human well-being domains (HWd) are a function of ecosystem (Se), social (Ss), and
economic (Sec) services 143
xiv
-------
Figure E8.2. Applying EPFs and EBFs to link alternative land-use scenarios to effects of human
well-being objectives. Ecosystem services (Se) are functions of various attributes of
ecosystem condition (xi) 144
Figure E8.3. Hypothetical land-cover scenarios based on decisions that influence population
growth and resource protection (FORE-SCE, USGS) 145
Figure E9.1. Major steps in the methodology to assess the transferability of measurements or
models to new sites 148
Figure E9.2. Locations of 13 sites evaluated for model transferability in this example 148
Figure E9.3. CART analysis of thirteen sites 149
Figure E10.1. Example of a consequence table in DASEES, displaying the effects of different
scenarios (e.g., Strobl Marsh, Schlepps, etc.) on fundamental objectives (e.g., minimizing
recontamination from flooding) where variables are categorical measures. Value functions for
each objective are combined to develop a composite prioritization score for each scenario
(Appendix Fl) 152
Figure E10.2. Example of a consequence table in DASEES, displaying the effects of different
scenarios (e.g., status quo, selective implementation, dredge) on fundamental objectives (e.g.,
minimizing coral reef impacts, maximize power generation) where variables are continuous
measures. Value functions for each objective are combined to develop a composite
prioritization score for each scenario (Appendix Fl) 152
Figure E10.3. Example of a Bayesian Network in DASEES to evaluate the probabilistic impacts
of decision options (e.g., dredging, subsidizing shade grown coffee) on fundamental objectives
(e.g., power generation, coral cover) 153
Figure Fl.l. Screenshot from DASEES indicating equal preference weights on three objectives 156
Figure F1.2. Screenshot from DASEES indicating different preference weights on three
objectives 157
Figure F1.3. Screenshot from DASEES indicating calculated preference weights on three
objectives 157
Figure F1.4. Screenshot from DASEES comparing prioritization scores across three scenarios
(status quo, selective implementation, or dredge) for each of three objectives - maximizing
sediment removal (green), minimizing coral reef impacts (blue), and maximizing power
generation (red) 158
Figure F1.5. Screenshot from DASEES comparing effects of three alternative scenarios on
measured objectives 158
Figure F2.1. Indicator summary results page for 4 of the 8 restoration sites assessed in Tampa
Bay 161
Figure F3.1. Eight domains of the Human Well-being Index (HWBI, Smith et al. 2013) 163
Figure F3.2. Stakeholders discussing issues and opportunities at the beginning of a workshop 164
Figure F4.1. Screen shot from DASEES showing a trigger point (green) that observational data
(orange) indicates was attained within the prescribed time frame 168
Figure F4.2. Screen shot from DASEES showing a trigger point (green) that observational data
(orange) indicates was not attained within the prescribed time frame 169
Figure Gl.l. US and US-Canada Great Lakes Areas of Concern indicated by current status with
respect to the Great Lakes Restoration Initiative. (Source: US EPA, updated October 2014) 171
Figure G1.2. Conceptual framework for the use of ecosystem service mapping and associated
analysis to support decision-making in an estuarine Great Lakes AOC (Angradi et al. 2016).
R2R = remediation to restoration; FES = final ecosystem services; BUI = beneficial use
impairment; AOC = area of concern; SPA = service providing area; SWF = social welfare
function (i.e., ecological benefit function [EBF]) 172
xv
-------
Figure G2.1. Map showing sub-watersheds adjacent to Mobile Bay, AL. Source: Mobile Bay
National Estuary Program 176
Figure G2.2. Example screen shot of the H20 tool intended for mapping of select ecosystem
goods and services at the watershed scale (Russell et al. 2015). Example is from Tampa Bay,
FL 177
Figure G3.1. Pacific Northwest Case Study sites in the Mashel and Tolt River watersheds in
Washington's Puget Sound Basin, and in Oregon's Tillamook Bay estuary and contributing
coastal watersheds. [Note: Seattle watersheds are being studied under a separate but related
EPA research program focused on urban water quality modeling] 180
Figure G3.2. A hypothetical example of ecosystem service trade-offs associated with alternative
watershed management scenarios in Puget Sound 183
Figure G4.1. San Jose Lagoon, San Juan, PR 185
Figure G4.2. Conceptual model merging DPSIR and benefits assessment to diagram cause-effect
relationships for San Juan case study (courtesy S. Balogh) 186
Figure G4.3. Ecosystem services related to San Juan NEP objectives, derived from existing
documents 187
Figure G5.1. Map showing the Stillwater Creek Watershed, Lakes McMurtry and Carl Blackwell,
and the Cities of Perry and Stillwater 190
Figure G5.2. Map showing the Stillwater Creek Land Use - Land Cover 2015 190
xvi
-------
List of Strategies
$5$ Apply PEGS concepts to explicitly connect EGS to people 5
Mi
%&ป<* Apply principles of SDM to emphasize flexible approaches to PEGS 7
^5$ Incorporate FEGS concepts at any point in the decision process 11
Use FEGS to identify beneficiaries as potential stakeholders 16
JWfj
Use conceptual models as a scaffold to visualize cause and effect 20
Jilt
iSM" Use objectives hierarchies to define what is important about FEGS 23
Use structured systems as a starting point to identify measurable objectives 26
Jf||
Consider FEGS as means to achieve other objectives 30
Use structured paradigms to link EGS alternatives to broader objectives 32
Jill
IS-,# Prioritize information and analysis to what is actually needed 34
Use conceptual models to visualize relationships 35
Jfl|
O-" Quantify FEGS with ecological production functions 38
Let objectives drive choice of methods for FEGS benefits analyses 40
Jill
vM Use Decision Support Systems to organize and link FEGS analyses 45
Compare alternatives and gain insights with consequence tables 47
Consider tradeoffs in FEGS benefits relative to other kinds of objectives 48
jl?lj
Monitor impacts to FEGS benefits after a decision to inform future decisions 51
xvii
-------
1. Introduction
1.1. Need for an Operational Framework
Sustainable human well-being depends upon sustainable management of environmental
resources (MEA 2005; NRC 2011). Decisions targeting sustainable use of resources will
inevitably involve economic, social, and environmental trade-offs. 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). One approach
in particular, the ecosystem goods and services concept, often simplified to "ecosystem
services", has generated much research in recent decades (Daily 1997; MEA 2005; Potschin and
Haines-Young 2011; Portman 2013; Bagstad et al. 2014; Boyd et al. 2015; Posner et al. 2016) as
an approach to better understand and quantify the benefits people obtain from the environment,
and to more generally advocate for consideration of environmental goals alongside social and
economic ones.
There is growing awareness at federal levels that benefits from nature contribute to economic
and social well-being, and integrating ecosystem services into planning can lead to better
outcomes, fewer unintended consequences, and more efficient use of limited funds (CEQ 2015).
However, integrating ecosystem services into environmental management or community
decisions is challenging because ecosystem services assessments (Bradley et al. 2016):
require significant information from environmental, economic, and social sciences;
are saddled with the inherent uncertainty of natural systems;
rarely involve a single issue, but instead are nested within trade-offs among other
social and economic resources;
inherently couple science-based and values-based information; and
are embedded in a decision environment with multiple stakeholder perspectives and
limited resources.
Ecosystem services conceptual models help organize our thinking around how decisions and
human actions lead to changes in ecosystem state and function, availability of ecosystem goods
and services, and ultimately the human well-being benefits derived from them (MEA 2005).
Multiple ecosystem services conceptual models are available, each tailored to specific
practitioner needs, such as conducting ecosystem service assessments (Kelble et al. 2013,
Verburg et al. 2016), linking ecosystem changes to economic valuation and cost benefit analysis
(Wainger and Mazzotta 2011, Boyd et al. 2015), assessing social impacts (Harper and Price
2011, Vanclay and Esteves 2011), or completing cumulative trade-off analysis on multiple
components of human well-being (Cowling et al. 2008, Yee et al. 2014). Boyd and Banzhaf
(2007) assessed a number of ecosystem services conceptual models for their capacity to
operationalize the linkage between ecological information and social analysis, but found most
had muddled typologies in need of refinements. Even with refinements, conceptual models that
rely on ecological and human well-being terminology to connect decisions to an array of impacts
and stakeholder groups, have proven difficult to implement into practice (Karrasch 2014, Van
Wensem et al. 2016, Verburg et al. 2016). Communities need practical strategies for
operationalizing conceptual models of ecosystem services within their decision processes.
1
-------
1.2. Integrating Final Ecosystem Goods and Services into Decision-Making
There is a need for scientifically sound and user-friendly approaches, tools, and methods to
support community decision-making. Support can assume 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 goods and services into federal
decision-making (Olander et al. 2015), including
connecting assessments to both scientific data and stakeholder values;
establishing well-defined measures of success; and
including a comprehensive set of services to people.
In concordance with this guidance, we introduce a simplified conceptual model based on the
Final Ecosystem Goods and Services (FEGS) approach (Fig. 1.1) and propose tools and methods
for operationalizing that model within a decision framework. In general, decisions alter social
and economic services (e.g., built infrastructure, health care) to impact human well-being.
However, decisions can also impact the state of the natural environment leading to changes in
human well-being via changes in the production of FEGS (Wainger and Mazzotta 2011). Impacts
of decisions may need to be considered within the context of external drivers (e.g., climate
change) that may also affect the state of the environment.
A Human
Well-beins
& Final EGS
Decision
Alternatives
Social & Economic
Services
A Ecosystem State
& Intermediate EGS'
nformation for Decision Support
Figure 1.1. FEGS based approach conceptual model of decision support.
In keeping with the best practice of connecting ecological services to people's values, FEGS are
defined as the "components of nature, directly enjoyed, consumed, or used to yield human well-
being" (Fig. 1.2; Boyd and Banzhaf, 2007). FEGS are biophysical qualities or features of the
environment that need minimal translation to be relevant to human well-being (Landers and
Nahlik 2013). Along a continuum of ecological production, FEGS are the last step before a user
interacts with the ecosystem. These final goods and services are distinguished from intermediate
ecosystem goods and services (e.g., fish habitat) that require additional steps to reach the
2
-------
ecological features (e.g., harvestable fish) that are directly experienced by beneficiaries. The
FEGS approach represents an important contribution to decision support by explicitly linking
environmental changes to changes in benefits.
Final Ecosystem Goods and Services (FEGS)
"[biophysical] components of nature,
directly enjoyed, consumed, or used to
yield human well-being" (Boyd&Banzhaf2007)
Beneficiary
+
Environmental
Context
Final Ecosystem
Good or Service
ft- M
il
- x*t yi iflPIW! JLlj
Recreational Birdwatchers
Mangroves
Charismatic bird species
Figure 1.2. Illustration of the three elements needed to define FEGS.
The FEGS-based approach (Fig. 1.1) and associated tools and methods presented here are
intended to provide practical strategies for focusing on a comprehensive suite of measurable
outcomes that are important to people. A FEGS approach has two target outcomes: production of
a FEGS (e.g., catchable fish biomass), and the delivery of that FEGS to an identifiable
beneficiary (e.g., angler). The FEGS-based approach requires the identification of both the
goods/services and the beneficiary (Fig. 1.2). FEGS should also be tied to the specific ecosystem
or environmental context (e.g., anglers in streams versus anglers in open coastal waters). This is
a distinction from simply considering ecosystem goods and services (EGS) production more
generally because it clearly links EGS to human benefit. The advantage of a FEGS approach is
that stakeholder values, with measurable outcomes directly linked to beneficiaries, can be clearly
incorporated into decisions. FEGS also provides a consistent and comprehensive approach to
ecosystem services that harmonizes across the multitude of approaches in the literature.
1.3. Benefits and Barriers of a FEGS Approach
Community decision-making is traditionally a problem-focused or opportunity-focused exercise.
Stakeholders naturally coalesce around focal issues of interest and this momentum frequently
drives the decision process. The criteria used to make a decision should be chosen by, and
meaningful to, the communities themselves. From a FEGS perspective, a community should
include the set of beneficiaries, and their geographic locations, potentially impacted by a given
decision.
Although decision-makers are increasingly following recommendations to consider ecosystem
services metrics as decision criteria (MEA 2005), many of the more commonly referenced
intermediate ecosystem services (e.g., production of fertile soils, air quality regulation, regulation
3
-------
of water quality and quantity, habitat for fauna or flora) may not be particularly relevant to
communities. A FEGS approach can facilitate increased understanding of how biophysical
processes integrate with social and economic concerns to give decision-makers a way to
communicate with stakeholders using relatable language.
A FEGS approach takes these intermediate ecosystem services and answers the questions "For
what" or "For who?". For example, "Production of fauna for what or for who?" (Table 1.1). In a
FEGS approach, the answers to these questions should include both a beneficiary and ecosystem
services metrics that are most directly relevant to that beneficiary (Table 1.1). Moreover, asking
"By what?" or "Where?" can bring additional clarity by specifying the environmental context, if
different ecosystems are providing services. The answers to these questions can affect which
specific metrics or decision alternatives should be under consideration, and multiple answers
may expose multiple objectives or potential tradeoffs that might need to be considered. A
community thus needs to identify appropriate FEGS metrics that are meaningful to the
community itself, based on the suite of beneficiaries and ecosystem types within the scope of the
decision being made.
Table 1.1. Examples of how FEGS-oriented questions can help bridge intermediate
ecosystem goods and services (IEGS) with FEGS (italics), the relevant environmental
context (underline), and their beneficiaries (bold).
IEGS
FEGS-oriented Question
FEGS-oriented Answer
Habitat for
fauna
Production of fauna for
what, by what, and for
who?
A specific type of fauna that is important to recreational
hunters visiting forested areas in the region
A specific species of charismatic bird that draws
birders to wetlands in the region
Water
quality
regulation
Water quality for what,
where, and for who?
Water salinity in groundwater for minimizing saltwater
intrusion impacts on local residents dependent on
drinking water aquifers
Water turbidity in coastal waters DODular for snorkelers
Contaminants in edible fish tissue in lakes DODular for
commercial fishermen
Water
quantity
regulation
Water quantity for what,
where, and for who?
Availability of fresh water in streams used bv municipal
drinking plant operators
Height of flood waters overtopping coastal sand dunes,
affecting coastal home owners during periodic storm
events
This report proposes that incorporation of FEGS into decision-making could lead to a more
inclusive process with improved outcomes, and specifically discusses
What are potential advantages of incorporating FEGS into community decision-making?
What are potential barriers to incorporating FEGS into community decision-making?
What practical strategies can help communities incorporate FEGS into their decision-
making processes?
4
-------
Incorporating FEGS thinking into community decision-making has a number of potential
advantages. By explicitly connecting ecosystem services to those that benefit from them, FEGS
allow for ecosystem services metrics that are directly relatable and most meaningful to
community stakeholders. A FEGS approach keeps the focus on what the community really
wants, avoiding potentially misleading or uninterpretable intermediate surrogates that are less
likely to resonate with a community or decision-makers. A FEGS approach also potentially
makes the decision process more inclusive, by explicitly connecting to beneficiary groups that
may otherwise be overlooked as stakeholders.
Strategy:
Apply FEGS concepts to explicitly connect EGS to people
The concept of Final Ecosystem Goods and Services (FEGS) explicitly connects
ecosystem services to the people that benefit from them, leading to identifying
biophysical metrics that are more meaningful to a community and what they care
about.
FEGS requires a broader consideration of the factors that might be influencing their production,
including potential feedbacks, interactions, or short versus long-term consequences. For
example, sport fish available in a lake can be a FEGS criterion for evaluating alternative lake
management options. However, the abundance of that fish is dependent on a number of broader
ecosystem conditions such as prey availability or nutrients, and how those may change through
space and time. Moreover, social and economic factors, such as the availability of boat access
points or alternative fishing sites, can influence whether potential benefits of FEGS are achieved.
Systems-thinking like this can help bring to light previously unconsidered factors or unintended
consequences associated with proposed decisions, and improve the likelihood of achieving
successful and sustainable outcomes (Bradley et al. 2016). It can also increase awareness across
stakeholders, as different "competing" priorities may ultimately connect to similar underlying
ecological processes, increasing the opportunity to implement decisions with multiple benefits.
For communities to take advantage of a FEGS approach, however, the science of FEGS must be
made operational. The use of FEGS requires taking a broader ecosystem view that captures
multiple issues, multiple interacting factors, and the impacts on multiple beneficiary groups. This
means that the identification of appropriate and relevant FEGS can be a complex process. In
addition, the use of FEGS metrics as decision criteria requires data to calculate those metrics and
methods to predict how they may be impacted by decisions under consideration.
Various methodologies, tools, and approaches exist that can facilitate incorporating FEGS into
community decision-making. Community issues and their decision processes are highly diverse,
often at various stages of implementation, and variable in their knowledge and experience with
ecosystem services. To be practical, a wide array of adaptable tools and approaches are needed
that can be implemented at various levels of depth, stages of the decision process, and levels of
experience. This report presents a number of these as practical strategies for incorporating FEGS
concepts, and more broadly ecosystem services, into the decision-making process.
For this report, we apply principles of Structured Decision Making (SDM) (Gregory et al. 2012)
as an operational framework for identifying and organizing strategies for integrating FEGS into
community decision-making. This framework has previously been applied to develop scientific
5
-------
information to support decisions for sustainable reef ecosystem services (Yee et al. 2014), and
recently been suggested as an approach for integrating ecosystem services into applied resource
management (Maseyk et al. 2017).
1.4. Structured Decision Making Framework
Both a review of previous ecosystem services assessments (Fulford et al. 2016a) and
recommendations for sustainability assessments (NRC 2011) suggest that a structured process
grounded in decision analysis can help provide an organizational framework for integrating the
fact-based scientific information and stakeholder-derived values needed for ecosystem services
assessments. Such a process can aid in assembling the most important information, identifying
tools and approaches to examine options, and ensuring that decisions are consistent with
stakeholder values, cognizant of trade-offs, and account for risks and uncertainties (Gregory et
al. 2012; Bradley et al. 2016).
Although variations of decision analysis processes exist (Keeney 1982; Hammond et al. 1999;
Goodwin and Wright 2004; Edwards et al. 2007; Gregory et al. 2012), most include a similar set
of steps (Fig. 1.3):
1. Clarify the context;
2. Define objectives and performance measures;
3. Develop alternatives;
4. Estimate consequences;
5. Evaluate trade-offs and select alternatives; and
6. Implement, monitor, and review.
The first step consists of understanding the context for decisions, which frames the focus of the
problem and the subsequent analysis. The next step requires consideration of what is valuable to
stakeholders through clarification of objectives, including how to measure them. Once options
for achieving objectives are identified, analysis can be done to compare the potential outcomes
from each alternative on the identified objectives, and explore trade-offs that stakeholders might
be willing to accept. The final step calls for implementing the selected strategy, and monitoring
to compare the results against the stated objectives. For the process to be effective, all steps
should involve stakeholders and include consideration of whole systems-thinking and long-term
consequences of the decision (NRC 2011).
This report highlights Structured Decision Making (Gregory et al. 2012) as a framework to
operationalize ecosystem services or FEGS assessments. SDM is a broad organizing framework,
grounded in decision analysis, but borrowing theory and practice from other fields. SDM, like
other decision analysis frameworks, provides a flexible approach for working through the steps
of a decision process (Fig. 1.3). The DASEES (Decision Analysis for Sustainable Economy,
Environment, and Society) approach, for example, follows a similar set of steps to SDM, and
provides a workspace, guidance, and suite of tools for stepping through and implementing a
decision analysis process (EPA 2012; Appendix Al).
6
-------
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
Estimate
Consequences
Potential outcomes
from decisions on the
objectives .
Develop Alternatives
Decision choices to
fulfill the objectives
Figure 1.3. The steps in a decision analysis process (NRC 2011; Carriger & Benson 2012;
Gregory et al. 2012).
SDM emphasizes developing better alternatives that are grounded on the values of the
stakeholders that decisions will ultimately impact. Quantitative analyses and evaluation of
alternatives may be done, but are not required. SDM instead places the strongest emphasis on
problem structuring: defining the issue, identifying objectives, and defining measures. Without
this structure, practitioners may end up applying their limited resources to collect the wrong
information for the wrong problem, leading to irrelevant, biased, or misleading assessments
(Carriger and Benson 2012). SDM incorporates Values-Focused Thinking (Keeney 1992), the
philosophy that clearly establishing what is important upfront will lead to more creative,
effective, defensible, and robust outcomes that have a higher probably of being accepted and
supported by stakeholders.
Strategy: Apply principles of SDM to emphasize flexible approaches to FEGS
&
Principles of Structured Decision Making (SDM) can provide a philosophy for
integrating FEGS into decision-making by emphasizing flexible approaches to
develop creative alternatives that are responsive to what stakeholders care about.
SDM has a number of advantages that make it a useful and transferable framework for
integrating FEGS into decision-making (Keeney 1992; Bradley et al. 2016):
Guiding information collection: Resources to gather information, collect data, or
develop and apply quantitative analysis are usually limited. An emphasis on structuring
stakeholder objectives early in the process can help to target resources on information
most directly relevant to comparing the effects of alternatives on what is important to
stakeholders.
7
-------
Improving communication: An emphasis on objectives keeps the discussion on what is
important to the group as a whole, and not on specific, technical aspects of decision
alternatives. Stakeholders may be suiprised to have common shared values, even if they
differ in magnitude.
Involving stakeholders: Decisions are often multi-faceted involving multiple
stakeholder viewpoints. SDM provides a process by which all stakeholders can express
and communicate what is important to them. Stakeholders are more likely to accept
decisions grounded in common values, even if they disagree in the degree of importance
of certain values over others.
Developing creative alternatives: SDM encourages the development of new alternatives
that are directly responsive to stakeholder objectives. Taking the time to characterize
what is important to stakeholders creates an environment for fostering creative options
with better prospects for acceptance and success.
1.5. Integrating Ecosystem Services Assessments within a Structured Decision
Making Framework
SDM provides an operational framework for integrating FEGS concepts into community
decision-making and environmental management (Fig. 1.4).
Clarify
Decision
Context
Estimate
Consequences
Define
Objectives
Develop
Alternatives
Evaluate
Trade-offs
and Select
Implement,
Monitor,
and Review
What EPFs or EBFs are
needed to estimate
consequences?
Which objectives are
ecosystem services?
How do we measure
them?
What ecosystem services may
be impacted (when & where)?
Who are the beneficiaries?
What intermediate
ecosystem services
might be means to
achieving broader
health, social or
economic benefits?
How much loss or eain in
ecosystem services are
we willing to accept?
Are beneficiary groups
being differentially
impacted?
Did the decision lead to
measurable change in
ecosystem services?
Were there unforeseen
impacts to be considered
going forward?
Figure 1.4. Points where FEGS concepts (white boxes) could be integrated into a decision
process (yellow boxes). EPF = ecological production function, EBF = ecological benefit
function.
8
-------
The first step is to clarify the decision context, including identifying the key stakeholders being
impacted by an issue under consideration and the benefits the community is hoping to achieve. In
this early stage, a consideration of ecosystem services may be particularly important - to identify
potential impacts on natural resources, key beneficiary groups to include in the deliberative
process, and the temporal and spatial scope of potential impacts. Some stakeholder objectives
may be directly represented by ecosystem services, whereas others may be more appropriately
described by social, economic, or health measures. Even when ecosystem services are not the
ultimate objectives of stakeholders, they may be a means toward achieving other objectives and
can help bring new alternatives to light. To estimate consequences, practitioners can then
identify ecological production functions (EPFs) and ecological benefits functions (EBFs) needed
to link the suite of alternatives to measures of FEGS and other stakeholder-relevant endpoints.
Such FEGS assessments can help evaluate how different beneficiary groups may be differentially
impacted by alternative options. Once a decision is made, researchers and practitioners can
monitor over time to determine whether the final decision leads to measurable benefits.
Monitoring will also help practitioners better understand what levels of ecosystem function are
necessary for meaningful change and to apply that knowledge to adapt with future decisions.
1.6. Flexibility to Work with Alternative or Existing Processes
Scientists, communities, and environmental managers interested in integrating FEGS into
decision-making may express concerns that a formal process sounds expensive and time-
consuming (Gregory et al. 2012; Bradley et al. 2016). In reality, most decisions made go through
the steps one way or another (Fig. 1.3), even if not formally called "decision analysis" or
"structured decision making". For less complex decisions, this often takes place informally or
even unconsciously. For example, when deciding what to have for lunch, explicitly clarifying the
context, developing performance measures, and carrying out the remaining steps is unnecessarily
burdensome. If, however, you were to explain to someone else how you decided your lunch
choice, you would see how you rapidly went through each step in the SDM process:
It was lunchtime and I was hungry (clarifying the context);
I wanted something that was healthy and nearby (defining objectives and
performance measures);
There were two restaurants in walking distance (developing alternatives);
Only one had salads (evaluating trade-offs);
Sol went there (selecting alternatives).
For more complex decisions, such as those encountered in communities or environmental
management, explicitly working through the steps of a decision analysis process has many
benefits, but even without a formal process, these steps will be taken one way or another.
The SDM framework is used in this report to demonstrate how FEGS thinking can be
incorporated into a decision-making process. Realistically, practitioners might incorporate
ecosystem services thinking in one, some, or all of these steps (Fig. 1.4). Moreover, the steps
may or may not occur within the context of formal SDM, or any other decision analysis process.
Numerous examples exist of SDM processes ranging from one day to two years (see Gregory et
al. 2012), and ecosystem services can be incorporated with varying levels of involvement, from a
single to multiple meetings, depending on budgets and timelines (Table 1.2).
9
-------
Table 1.2. Examples of meeting/workshop planning schedules for integrating FEGS into
decision processes of varying budget or timeline (Gregory et al. 2012, Bradley et al. 2016).
Timeline or
Budget
Meeting Plan
Fast timeline
(<6 months)
Small budget
1. Decision sketching: Quickly sketch through decision context, objectives, alternatives,
consequences, trade-offs, including potential impacts on FEGS and their beneficiaries;
identify low-hanging fruit that can be initiated quickly
Medium
timeline
(6-18
months)
Medium
budget
1. Clarify decision context, objectives, & alternatives: Develop work plan; identify
representative stakeholders to participate in the process, including beneficiaries of FEGS;
confirm objectives to use as evaluation criteria, including any FEGS benefits; develop
menu of alternatives, including whether FEGS could be means to achieve other objectives
2. Review existing information with technical experts: Identify key uncertainties where
more information, including information on FEGS, is needed
3. Work with technical experts to conduct assessment of alternatives: Identify
performance measures, including indicators of FEGS; conduct evaluations (expert opinion,
quantitative analyses using EPFs and EBFs) to compare alternatives
4. Review consequences of alternatives and make draft recommendations: Uncover
trade-offs, including across FEGS and other objectives; identify areas of agreement
5. Make a decision and develop an implementation strategy: Make and implement a
decision; develop a plan to monitor FEGS and other measurable objectives, and address
critical uncertainties moving forward
Slower
timeline
(1-5 years)
Larger budget
1. Decision process: Develop a work plan; identify representative stakeholders to
participate in the process, including beneficiaries of FEGS;
2. Decision sketching: Quickly sketch through context, objectives, alternatives,
consequences, trade-offs, including potential impacts on FEGS and their beneficiaries;
build a common understanding of the scope of the problem
3. Define objectives: Confirm objectives for evaluation criteria, including FEGS benefits
4. Specify performance measures: Define the evaluation criteria with help from expert
judgment, including indicators of FEGS benefits
5. Develop alternatives: Develop a preliminary menu of alternatives, including whether
FEGS could be means to achieve other objectives
6. Identify information needs: Identify key uncertainties where more information,
including information on FEGS, is needed
7. Technical working group: Identify technical needs and process
8. Technical field work and analysis: Conduct field work, valuation studies, development
& application of EPF/EBFs, or rely on expert groups as needed to evaluate consequences
9. Round 1 alternatives & consequences: Review outcomes from technical evaluations;
identify potential trade-offs; revise objectives, performance measures, or alternatives as
necessary; conduct additional technical evaluations as necessary to revise
10. Round 2 alternatives & consequences: Review outcomes and any revisions from
technical evaluations; Explore trade-offs; identify areas of agreement
11. Make a decision and develop an implementation strategy: Make and implement a
decision; develop a plan to monitor FEGS and other measurable objectives, and address
critical uncertainties moving forward
12. Monitor and review: Conduct technical monitoring, field work to evaluate success of
plan, including changes in FEGS; adapt as necessary moving forward
10
-------
For scientists or technical experts not directly involved in the decision process, organizing
research within an SDM framework may help improve the relevance of scientific information
collection to decision-making (Yee et al. 2014). It can also help in the development of data
quality objectives and quality assurance plans to document what will be measured and why, and
what level of uncertainty is considered acceptable.
When decision-makers already have their own process, this presents a substantial challenge to
make an FEGS assessment both credible and relevant within the existing context (Posner et al.
2016). The barriers are numerous, and likely include the lack of inclusion in regulatory or
programmatic language (Portman 2013), a lack of an effort to make the information usable by
decision-makers (Dilling and Lemos 2011, Posner et al. 2016), and the scale of ecosystem
services assessments (Chan et al. 2006, Wainger et al. 2010, Koschke et al. 2014).
Strategy: Incorporate FEGS concepts at any point in the decision process
Ecosystem services concepts can be integrated at multiple points in a
decision process, whether that process is in early or late stages, or whether
that process includes informal or formal decision analysis.
Ecosystem services can be incorporated into decision-making even if there is a previously
existing or on-going process. For example, the EPA Area of Concern (AOC) program was
developed in the 1980s to address severe pollution impairments in Great Lakes communities.
The program has pre-existing programmatic targets agreed upon through the AOC governance
structure. Although an SDM-type process to identify new objectives may not be feasible,
information on how decisions affect ecosystem services can still be provided in a way that
preserves the existing targets and governance structure (Angradi et al. 2016; Appendix A2). This
can be done with a three-fold strategy:
1. Tailor ecosystem services assessments to fit the needs or requirements of the
established governance and regulatory structure,
2. Co-produce data, maps, and models iteratively with community partners through
participatory science, to make ecosystem service research products more useful
for the existing established structure; and
3. Generate data, maps, and ecosystem services models at a scale that meets the
needs of local decision-makers.
Even if a formal SDM process is not being used, the general steps of SDM (Fig. 1.4) can still be
used to identify where FEGS concepts can be integrated into the existing process (Table 1.3).
For example, ecological risk assessments (EPA 1998), sustainability planning (CDC 2016), or
developing environmental impact statements under the National Environmental Policy Act
(NEPA; CEQ 2007) are all decision processes communities might be involved with. Each has a
unique set of steps (Table 1.3), but ecosystem services concepts can still be integrated at many
places within each process. Using FEGS endpoints in ecological risk assessments, for example,
may provide more relevant assessment endpoints and better communication of risk to humans
than conventional biophysical endpoints (Munns et al. 2015b).
11
-------
Table 1.3. Comparison of a variety of other decision processes to SDM, illustrating where FEGS concepts could be integrated.
SDM Framework
EGS Concepts
Ecological Risk Assessment3
Community Sustainability
Planningb
NEPA Process0
1. Clarify Decision Context
What FEGS might be impacted?
What FEGS beneficiaries should
be included as stakeholders?
1. Planning and Scoping
Scope and complexity
2. Problem Formulation
Prepare conceptual model
1. Create a Shared Understanding
2. Create a Work Plan
3. Position Coalition Efforts
4. Look at the Current Picture and
Pending Items
1. Scoping
Define purpose and
needs
Identify issues to be
considered
2. Define Objectives and
Performance Measures
Which objectives are FEGS or
their benefits?
How will they be measured?
Evaluate goals
Select assessment endpoints
5. Develop Criteria
Identify criteria based on
sustainability objectives
Determine what effects
will be analyzed for
impacts
3. Develop Alternatives
Are any FEGS means to achieve
objectives?
Develop analysis plan, including
stressors & levels to be evaluated
based on management options
Identify ongoing and pending
efforts
Identify alternatives
under consideration
4. Estimate Consequences
What EPFs or EBFs are needed
to estimate consequences?
3. Analysis Phase
Evaluate exposure to stressors
Evaluate relationships between
stressors and assessment endpoints
Evaluate performance of efforts
on criteria
2. Assessment
Analyze impacts of
alternatives
5. Evaluate Trade-offs
How much loss or gain in FEGS
benefits is acceptable ?
Are FEGS beneficiaries
differentially impacted?
4. Risk Characterization
Make comparisons based on
exposure, stressor-response profiles
5. Report to Risk Managers
Consider social, political, economic,
legal factors, as well as risk analysis
6. Prioritize Efforts based on
Evaluation Criteria
3. Public Review and
Comment
4. Decision on Impact bv
Reviewing Agency
6. Implement, Monitor, Review
Did the decision lead to
measurable change in FEGS?
Were there unforeseen impacts on
FEGS to consider next time?
Select a course of action
Monitor results
7. Create Options for Priority Efforts
8. Synthesize Options in a Plan
9. Implement the Plan
10. Evaluate Outcomes and Revise
5. Implement with
Monitoring or Mitigation
as Provided in Decision
a) EPA 1998; b) CDC 2016; c) CEQ 2007
12
-------
1.7. Overview of this Report
This report is intended to provide practical strategies, approaches, and tools to integrate FEGS
concepts into a decision-making process:
Chapter 2 covers the SDM step of clarifying the decision context. This step aims to bring focus
to a problem and defining the types and scope of information needed. Approaches such as the
development of conceptual models and characterization of stakeholders benefit from the
incorporation of FEGS thinking by providing step-by-step guidance to link human activities and
environmental stressors to impacts on ecosystem goods and services and loss or gain to potential
beneficiaries. Appendix B provides details on tools and approaches to clarify the context.
Chapter 3 describes characterizing stakeholder objectives, and explicitly defining them with
performance measures so there is a clear understanding of what endpoints should be evaluated
(qualitatively or quantitatively). In many cases, ecosystem services may be fundamental
stakeholder objectives. In other cases, they may be means to achieving other economic or social
objectives. This chapter reviews a number of approaches that can help communities identify and
measure ecosystem services objectives, and potential benefits of ecosystem services. Appendix
C provides additional details on tools and approaches to characterize objectives and performance
measures.
Chapter 4 covers the development of decision alternatives, and highlights how ecosystem
services can lead to the creative development of new alternatives. In other cases, the alternatives
may be pre-defined, and the focus is more on the impact of those options on ecosystem services
objectives. Appendix D provides additional details on tools and approaches to develop
alternatives.
Chapter 5 describes approaches for estimating the consequences of the decision alternatives on
the objectives, including those related to FEGS. This chapter describes tools and approaches for
modeling how changes in stressors or environmental conditions may affect the production of
ecosystem services (through EPFs) and the delivery of ecosystem services (through EBFs) to
benefit human well-being. Methods for identifying models and characterizing their
transferability across locations are also discussed. If ecosystem services are not considered when
estimating the consequences of the alternatives, decisions made to maximize economic or social
benefits can have unintended consequences to the environment and lead to unexpected outcomes.
Appendix E provides additional details on tools and approaches to estimate consequences.
Chapter 6 explores the last two steps of an SDM process: evaluating trade-offs and taking
action, and monitoring outcomes. This chapter describes approaches for weighing different
stakeholder objectives, and characterizing acceptable thresholds for losses and gains in
ecosystem services. Because ecological, social, and economic systems are interrelated, it is
important to include ecosystem services trade-offs alongside the more commonly considered
economic and social trade-offs when making decisions. Methods are identified for monitoring
changes in ecosystem services, and other objectives, once a decision is implemented. Appendix
F provides additional details on tools and approaches to evaluate trade-offs.
13
-------
Chapter 7 presents the different ways that five example case studies are using principles of
Structured Decision Making to integrate FEGS into a community decision process. Appendix G
provides additional details on case studies.
Finally, Chapter 8 synthesizes some of the advantages of using principles of Structured
Decision Making, particularly values-focused thinking, to integrate FEGS into community
decision-making.
14
-------
2. Clarify the Decision Context
2.1. Clarifying the Decision Context
The decision context is the problem, issue, or reason for making a decision, and defines the
scope of information needed (Gregory et al. 2012). Whether or not this step is taken deliberately,
every decision begins with a context. Taking the time to clarify the context can help bring focus
to a problem and define the types and scope of information needed. Failing to fully characterize
the context may lead to limited resources being applied to solving the wrong problem (Carriger
and Benson 2012). Furthermore, decisions are often made within a very narrow context defined
by the mission or objectives of the decision-maker, but may fail to consider unintended
consequences on social, economic, and environmental values (Bradley et al. 2016). Structured
Decision Making (SDM) places a strong emphasis on problem structuring early in the process
(Gregory et al. 2012). Having a clear understanding of the full scope under consideration allows
decision-makers to move forward with confidence that they are gathering the needed information
and involving the appropriate stakeholders.
A decision context can be narrow (e.g., reducing nitrate leaching from farm fields) or broad (e.g.,
improving resiliency of communities to climate change), and ecosystem services is often not the
central focus (Fulford et al. 2016a). Taking the time to explore the role of ecosystem services in
the context can help ensure that important values, novel alternatives, or key participants in the
process are not overlooked.
2.2. Incorporating Ecosystem Services into Decision Context Characterization
All community decisions can, in one way or another, be related back to the broad goal of
increasing the well-being of community members. Human well-being encompasses the three
components of sustainability: social, economic, and environmental (Summers et al. 2014). While
economic and social concerns are often concrete and tangible (i.e., cost of the project, job
creation or loss), environmental concerns can be more ephemeral and less obviously connected
to community well-being, and therefore easy to neglect or ignore. This omission can lead to an
incomplete understanding of the context that can, in turn, lead to unintended outcomes when a
final decision has been made. Unintended consequences can be further exacerbated because
social, economic, and environmental variables are inextricably intertwined with one another
(Summers et al. 2016) so changes to one will have an impact on another. In order to have a
complete understanding of the context, decision-makers must have a firm grasp of the
environmental aspects of the problem, as well as the social and economic.
By connecting ecosystem services to human beneficiaries of those services, Final Ecosystem
Goods and Services (FEGS) thinking is a useful approach for making environmental concerns as
concrete and relevant to decision-makers as social and economic ones. This thinking can also be
helpful in identifying connections between the environmental, social, and economic elements.
For example, while it may be difficult to connect water quality to job creation or loss, it is easy
to see how fish, the abundance of which are dependent on water quality, could connect to job
creation in a coastal town. Consideration of FEGS when characterizing decision context allows
for a more complete understanding of the problem and a more successful decision process.
15
-------
Strategy: Use FEGS to identify beneficiaries as potential stakeholders
&
Final Ecosystem Goods and Services (FEGS) is a useful construct for ensuring
potential benefits and costs of environmental impacts are under consideration,
and identifying beneficiaries to engage as stakeholders in the decision process.
2.3. Elements of Decision Context
The decision context should include a consideration of (Gregory et al. 2012; Rehr et al. 2012):
What is the issue or problem being addressed, and how does it relate to other issues,
programs, or decisions already in motion?
Who is making the decision?
What is the spatial scale of the decision?
What is the temporal scale of the decision?
What is the general range of objectives being considered?
What is the general range of alternatives being considered?
What outcomes are likely for various alternatives?
What kind of information will be needed?
What kind of tools will be needed?
What are the key areas of uncertainty?
What kind of decision-maker consultation is needed?
What kind of stakeholder consultation is needed?
What kind of expert consultation is needed?
What are the policy and legal bounds on the decisions and their enforcement?
What is the history or culture of past decisions?
What is the trajectory of the current decision, including any related prior decisions or
existing planning documents?
Many of these components can be informed by consideration of FEGS (Table 2.1).
Table 2.1. Examples of FEGS related questions that could be asked to help clarify the
decision context.
Decision Context Element
Example Ecosystem Services Questions
What is the issue or problem
being addressed?
Are potential impacts on ecosystem services under consideration?
Who is making the decision?
Is a multi-agency effort relevant, including the involvement of
agencies with a mission or mandate to consider ecosystem services?
What is the general range of
objectives being considered?
Have ecosystem services and their benefits been considered as
potential objectives? Are objectives linked to specific beneficiaries
(i.e. FEGS)?
What is the spatial scale of the
decision?
Where are ecosystem services being impacted?
Should the spatial scale be expanded to include the area supplying
ecosystem services or the area where benefits are received, or both?
What is the temporal scale of
the decision?
Should either short-term or long-term impacts on ecosystem services
be considered, or both?
16
-------
Decision Context Element
Example Ecosystem Services Questions
What is the general range of
alternatives being considered?
Might ecosystem services be means to achieving other objectives?
What outcomes are likely for
various alternatives?
Might ecosystem services be impacted by alternatives?
What benefits could be derived from altering ecosystem services?
What kind of information will
be needed?
Which ecosystem services or benefits need to be quantified?
What ecological production functions (EPFs), ecological benefits
functions (EBFs), or monetary valuations might be needed?
What kind of tools will be
needed?
What ecosystem services assessment tools (e.g., In VEST1) might be
useful?
What are the key areas of
uncertainty?
Which ecosystem services or benefits are lacking information?
Where can we rely on existing knowledge vs. target new research?
What kind of decision-maker
consultation is needed?
How can ecosystem services consideration be integrated into an
existing decision process?
What other agencies could be brought in to implement ecosystem
services related alternatives?
What kind of stakeholder
consultation is needed?
What beneficiary groups are potentially going to be impacted by loss
or gain in ecosystem services (i.e., PEGS)?
Are they being represented as stakeholders in the process?
What kind of expert
consultation is needed?
Are social scientists, economists, ecologists, or other experts needed
to quantify ecosystem services and their benefits?
What are the legal bounds on
the decisions and their
enforcement?
Does the agency have a legal mandate to consider ecosystem
services?
Are the impacts or alternatives related to ecosystem services within
the decision-maker's scope?
If the decision is tied to an existing planning document, did it
include a consideration of ecosystem services?
What is the history or culture
of past decisions?
How have past decisions impacted ecosystem services?
Have ecosystem services related goals been considered in the past?
What is the trajectory of the
current decision?
If the current decision is building off existing planning documents,
do those documents include ecosystem services?
Have prior decisions had desired impacts on ecosystem services, and
can we learn from them?
If a decision is already in motion, is there flexibility to pause and
consider the potential role of ecosystem services?
1 https://www.naturalcapitalproject.org/invest/
2.4. Establishing a Decision Context
The decision context may initially be established with a small group of stakeholders, but may be
iteratively refined through ongoing discussions as the process develops. Initial attempts to
characterize the context typically involve a small group with factual knowledge about the
decision at hand (Bradley et al. 2016). In the early stages, it is important to allow the context to
be sufficiently broad (Brown 1984); limiting the context to exclude issues or values that some
stakeholders consider important is setting the stage for disagreements (Gregory and Keeney
2002). The context can be iteratively adapted and refined throughout the process.
17
-------
A useful first step in an SDM process is to quickly "sketch" through the steps (Fig. 2.1) to
identify preliminary information on objectives and possible measures* a range of possible
management actions and their consequences, and key pieces of information and their
uncertainties (Gregory et al. 2012). The process may be iterative in that information at any one
step may lead to changes in the previous steps, including redefining the initial context. Decision
sketching could also include an explicit consideration of FEGS (Fig. 2.1).
Could the decision lead to
measurable change in
ecosystem services?
Can benefits be monitored?
What ecosystem services may be
impacted (when & where)?
Who are the beneficiaries?
How much loss or
gain in ecosystem
services are we
willing to accept?
Are beneficiary
groups being
differentially
impacted?
Implement. Monitor, and
Review
How might implementation
be monitored and impact
future decisions? J /
. t
Evaluate Trade-offs
and Select
What are the critical
trade-offs across
alternatives?
Clarify Decision
Context
The who, what,
where of a decision
opportunity
J}
Estimate
Define Objectives
What is valued in the
decision opportunity,
and how to measure it?
Consequences
What are potential
outcomes from
decisions on the
objectives?
Develop Alternatives
What creative decision
choices could fulfill the
objectives?
Which objectives
are ecosystem
services?
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 2.1. Example questions asked during a generic decision sketching process (yellow
boxes), and where ecosystem services questions could be integrated (white boxes). Green
arrows illustrate the process can be highly iterative as information from later steps leads to
previous steps needing to be revisited.
Numerous tools and approaches exist for helping decision-makers sketch the context, a number
of which are included in DASEES (Decision Analysis for Sustainable Economy, Environment,
and Society; Appendix Bl). The FEGS Classification System (FEGS-CS), in particular, can
provide a structured set of questions and keywords for identifying relevant ecosystems
(Environmental Class), potential ecosystem services they provide (FEGS), and their beneficiaries
(Beneficiary Class) (Table 2.2; Landers and Nahlik 2013). An example of using the FEGS-CS to
expand stakeholder engagement to include beneficiaries of ecosystem services is provided in
Appendix B2. Another application of FEGS-CS uses a weight of evidence approach, based on
the growing body of literature evaluating coastal ecosystem services, and considers the linkages
between coastal habitats and FEGS-CS beneficiaries (Appendix B3). This information might
then aid decision-makers in prioritizing restoration activities or in assessing vulnerability to
stressors.
18
-------
Table 2.2. Classes and sub-classes of Environments, Beneficiaries, and FEGS in the FEGS-
CS (from Landers and Nahlik 2013), and example questions that might help identify them.
Class
Sub-Class
Elicitation questions
Environmental
1. Aquatic 11. Rivers and Streams
What ecosystems are relevant
Class
12. Wetlands
to the decision?
13. Lakes and Ponds
14. Estuaries, Near coastal, Marine
What ecosystems might be
15. Open ocean and sea
impacted by decisions?
16. Groundwater
2. Terrestrial 21. Forests
What ecosystems might be
22. Agroecosystems
providing benefits?
23. Created Greenspace
24. Grasslands
25. Scrubland/Shrubland
26. Barren / Rock and Sand
27. Tundra
28. Ice and Snow
3. Atmospheric 31. Atmosphere
Beneficiary
01. Agricultural
For each environmental class,
Class
02. Commercial, industrial
which beneficiaries are present?
03. Government, municipal, residential
04. Commercial, military transportation
Who might be benefitting from
05. Subsistence
these ecosystems?
06. Recreational
07. Inspirational
08. Learning
09. Non-use
10. Humanity
FEGS
01. Water
How is each beneficiary class
02. Flora
benefitting from the
03. Presence of the environment
environment?
04. Fauna
05. Fiber
What do they use or care about
06. Natural materials
that is directly provided by the
07. Open space
environment?
08. Viewscapes
09. Sounds and scents
10. Fish
11. Soil
12. Pollinators
13. Depredators and (pest) predators
14. Timber
15. Fungi
16. Substrate
17. Land
18. Air
19. Weather
20. Wind
21. Atmospheric phenomena
19
-------
2.5. Using Conceptual Models to Identify Ecosystem Services
Conceptual models help to capture, visualize, and organize the connections among factors in a
complex system (Joffe and Mitchell 2006; Knol et al. 2010; Yee et al. 2011). There are a number
of approaches to building conceptual models, including influence diagrams, concept mapping,
logic models, and pathway diagrams (Qiu 2016). Conceptual models can range from simple
models showing the connections between different factors, to complex models showing cause
and effect information. Many models are available that describe the connections and feedbacks
among economic, environmental, and social elements (e.g., Triple Value Simulation [3VS];
Tenbrink et al., 2016), and may ultimately form the foundation for predictive models used
subsequently to estimate consequences of alternatives. When built with the involvement of
stakeholders, conceptual models can help to unite different areas of knowledge and build a
common understanding of the decision at hand (EPA 1998; NRC 2011). An example of working
with stakeholders to integrate ecosystem services concepts into a conceptual model for the St.
Louis River Area of Concern (AOC) is provided in Appendix B4.
Strategy: Use conceptual models as a scaffold to visualize cause and effect
Conceptual models can help visualize cause and effect between decisions,
stressors, FEGS, and benefits. They help provide a common language, guide
discussions, and elicit information, especially when built from structured
generic model as an underlying scaffold.
A structured conceptual model can provide a scaffold and common language for describing a
system and helping guide discussions. They ensure that key concepts, particularly those relevant
to ecosystem services, are not overlooked. Some conceptual models look more broadly at cause
and effect relationships between interacting components of social, economic, and environmental
systems, such as with the Drivers-Pressures-State-Impact-Response (DPSIR) model (Fig. 2.2;
EEA 2005; Yee et al. 2014; Bradley and Yee 2015).
STATE
What are the
environmental
consequences?
RESPONSE
What options are
under consideration?
IMPACT
Why do we care?
What benefits
could be lost?
PRESSURES
Why is this action
proposed?
What stressors will
it reduce?
DRIVERS
What economic
sectors may be
impacted by decisions?
Figure 2.2. Questions to guide development of a DPSIR causal-chain conceptual model
(Bradley and Yee 2015).
Ecosystem assessment models provide greater detail specifically linking ecosystem structure and
function to ecosystem services and their benefits and value to people (Fig. 2.3; Wainger and
Mazzotta 2011; Olander et al. 2015; Bousquin et al. 2015).
20
-------
Supply
Demand
Biophysical
structure:
Wetland
characteristics -
sue, depth,
location
v 7
if"
Ecological
Assessment
r
^v,
Function:
Retention or
slowing of water
1j"
Functional
Assessment
r
Service:
Flood regulation
that reduces risk
of flooding
1j"
Ecosystem
Service
Assessment
Benefit
Reduction in
damages to
property and
Infrastructure
i
\ I
Benefit
Assessment
Value
Value of lost uses;
avoided costs of
repair or
replacement
v_
{y
Monetary
Valuation
Figure 2,3. Example conceptual model for conducting ecosystem services assessments
(Bousquin et al. 2015).
The Final Ecosystem Goods and Services (FEGS) conceptual model distinguishes direct benefits
provided by ecosystems from intermediate supporting services that may indirectly contribute to
human well-being benefits (Fig. 2.4; Fulford et al. 2016a), as well as emphasizing ecosystem
services alongside economic and social services that also contribute to components of human
well-being (Smith et al. 2014).
Benefit
Functions
Impact
Functions
Decision
Alternatives
A Final EGS
Social & Economic
Services
A Ecosystem State
(& Intermediate EGS)
Information for Decision Support
Figure 2.4. FEGS based approach conceptual model for decision support.
21
-------
The use of structured generic conceptual models (e.g., Figs 2.2,2.3,2.4) to develop a
community-specific conceptual model can help to elicit and clarify many elements of the
decision context (Table 2.1) that might otherwise overlook ecosystem services, including (Yee et
al. 2014; Bradley et al. 2016):
Preliminary objectives: Conceptual models help to identify benefits to society,
including FEGS and human well-being, which may be or be related to fundamental
stakeholder objectives. Visualizing the system as a whole may help stakeholders identify
novel objectives or see other points of view.
Preliminary decision alternatives: Conceptual models can help to identify creative
intervention points by visualizing the system as a whole, including interacting
components. Intermediate ecosystem services may be means to achieving other
objectives.
Key areas of uncertainty and information needs: In attempting to illustrate cause and
effect relationships in a conceptual model, key relationships where there is more or less
certainty can emerge and be captured. Relationships directly or indirectly linking
decisions to stakeholder objectives should be a high priority for information gathering.
These could include ecological production functions, benefits functions, or monetary
valuation, but should focus on information necessary to link alternatives to objectives.
Potential outcomes: Because conceptual models illustrate cause and effect relationships,
they inherently relate decision alternatives to potential changes in benefits. Conceptual
models can form the foundation for predictive quantitative analysis. Many times the
process of developing the conceptual model itself may be informative enough to move
forward with a decision.
Key stakeholder groups: Conceptual models that specifically identify social drivers,
economic drivers, or FEGS beneficiaries can help to identify key stakeholder groups for
inclusion in the process going forward.
22
-------
3. Define Objectives and Performance
Measures
3.1. Defining Objectives
Objectives are statements of what is important to stakeholders within a particular decision
context. Taking the time to characterize objectives early in the process is important because well-
defined objectives (Keeney 1992; Gregory et al. 2012):
focus in on what really matters about a decision;
help experts target information needs and uncertainties to what is most relevant
for a decision;
shift the focus from "what can we doT to "what do we want to achieve'1.":
provide a basis for creative options to achieve what really matters (value-focused
vs. alternative-focused thinking);
promote inclusive decision-making, if a broad range of stakeholder objectives are
fully considered;
promote transparency and reduce confusion about what is really meant by
different stakeholders, if they are well-defined by performance measures;
form the foundation for evaluating and comparing alternatives; and
ground the process toward achieving common values, promoting more defensible
decisions that are more likely to gain broad acceptance.
While most decision processes characterize objectives, often including ecosystem goods and
services (Fulford et al. 2016a), objectives can be a complex mix of means and ends, vision
statements, or lists of things to do (Keeney 1992; Gregory et al. 2012), or may fail to consider
the full range of what stakeholders care about (Gregory and Keeney 2002). In decision analysis,
objectives are concise statements of what matters for the decision at hand, and should be
complete, necessary, unambiguous, sensitive to the alternatives under consideration, context-
specific, and separate what fundamentally matters from the means to achieve it.
Strategy:
Use objectives hierarchies to define what is important about FEGS
i
Depending on the context, FEGS may be fundamental objectives or means to
achieving other social or economic objectives, such as better health or more
jobs. Objectives hierarchies can help clearly define what is important about
ecosystem services (intermediate or final), and the means to achieve it.
Formal decision analysis tools, such as objectives hierarchies (Appendix CI), can help elicit,
structure, and define objectives (Merrick et al. 2005). Ideally objectives should be directly
elicited through discussions with stakeholders and decision-makers, but inferring them from
management plans, community reports, or from representatives familiar with the issue is also
considered acceptable (Parnell et al. 1998).
23
-------
Like all the steps of Structured Decision Making, development of objectives can be an iterative
process (Gregory et al. 2012). Early assessments may be useful in identifying critical missing
stakeholders who may need to be brought into the conversation (Bradley et al. 2016).
Development of performance measures, decision alternatives, or estimates of consequences on
objectives may bring to light new objectives or uncertainties where objectives need to be further
refined.
3.2. Distinguishing Means Objectives from Fundamental Objectives
Objectives should define what fundamentally matters about a decision (Keeney 1992) and are
usually described with a verb indicating a preferred direction of change for something that
matters (Carriger and Benson 2012).
Fundamental objectives can be determined through an iterative process of asking "Why is this
important?". If the answer is to achieve something else, then it is likely not a fundamental
objective, but rather a means to achieving another objective (Table 3.1). Progress toward a
fundamental objective is achieved when the answer to "Why is this important?" is "Jus! because
it is". Means objectives also can be identified by working backwards from fundamental
objectives to answer "ffow clo I achieve this?". The distinction is important because:
fundamental objectives are used to evaluate alternatives;
alternatives that improve means objectives may or may not improve fundamental
objectives;
analyses based on means objectives may be measuring the wrong thing; and
focus on means objectives may impede the development of creative alternatives to
achieve what really matters.
Table 3.1. Key types of objectives (Carriger and Benson 2012; Bradley et al. 2016). Note
here "maximize water quality" is a means to improving human health, however in other
contexts water quality may be a fundamental objective. Similarly, "maximize human well-
being" here is a strategic objective, but in other contexts it may be a fundamental objective.
Objective
Type
Definition
Example
Fundamental
The outcomes you really care about, regardless of how
they are achieved
Maximize human health
Means
Particular ways of achieving fundamental objectives
Maximize water quality
Process
Describe how something should be done, generally to
improve the decision process itself
Maximize the use of
credible scientific data
Strategic
Broad objectives that describe higher-level or longer-term
objectives, particularly for an organization or agency
Maximize human well-
being
24
-------
Depending on the decision context, ecosystem goods and services may be fundamental
objectives or means to achieving other objectives. Taking the time to consider FEGS can:
bring to light economic, social, or human well-being objectives that might be
affected if alternatives impact ecosystem condition;
broaden the objectives to be more inclusive of beneficiary concerns; and
open the door to creative alternatives that leverage ecosystem services to achieve
other objectives.
When a community is making a decision, economic and social objectives are often the focus
(Fulford et al. 2016b) and considered separately from environmental objectives. However,
economic, social, and environmental objectives are intertwined, and the inclusion of ecosystem
services can help make they are being fully considered. For example, if a community is making a
decision about whether or not to deepen a harbor to allow for larger cruise ships to enter, a key
objective may be to increase revenue from tourism. Looked at independently, decision-makers
may simply weigh the decrease in water quality from the dredging and construction against the
increase in tourism revenue as more cruise ship travelers come to the port and spend money in
local businesses. Looked at from an FEGS perspective, however, decision-makers can see how a
decrease in water quality may impact fish populations on which local businesses depend. This
could impact a range of beneficiaries from local restaurants to snorkelers to fishing charters.
With an FEGS perspective, decision-makers are able to see that increasing the capacity of the
harbor to bring in additional tourists, may not lead to a simple and direct increase in tourism
revenue for the community. Ecosystem services thinking can help decision-makers to more
comprehensively define and evaluate impacts on their objectives. In this example, looking at
both an intermediate ecosystem service (water quality) and a final ecosystem service (fish)
allows decision-makers to incorporate a broader system perspective into their thinking.
In the harbor expansion example above, FEGS thinking encourages a more comprehensive view
of identifying and evaluating impacts to objectives. That same thinking can also expand a
decision-maker's understanding of the decision, leading to an expanded set of objectives that
includes stakeholder groups that may not have been considered otherwise. Without considering
ecosystem services in the harbor expansion example, decision-makers would likely consider the
port authority, shipping and cruise companies, and local tourism dependent businesses as their
key stakeholders. However, with a FEGS perspective, potentially impacted ecosystem services
such as attractive views, noise associated with the increased harbor traffic, and potential
degradation of air or water quality in nearby recreational areas are all potentially identified. Once
identified, decision-makers are able to include those who benefit from those services (e.g., local
residents, park associations) as additional key stakeholders in the decision-making process and
ensure that their perspectives are also represented in the objectives.
3.3. Performance Measures
Performance measures are attributes that define the fundamental objectives and are used to
assess how well the alternative decision options fare at achieving stakeholder objectives (Keeney
1992; Gregory et al. 2012). Performance measures help reduce uncertainty amongst stakeholders
about what is really meant by objectives. For example, vague objectives like "maximize water
25
-------
quality" can mean different things to different people. FEGS metrics, in contrast, explicitly
define the biophysical measures of the ecosystem tied to specific beneficiary (see Table 1.1),
reducing ambiguity. They help precisely define the exact information that needs to be collected
or modeled to compare alternatives. Performance measures should be (Keeney 1992; Keeney and
Gregory 2005; Bradley et al. 2016):
clearly defined to avoid ambiguous interpretation;
comprehensive in reflecting the concerns specified in the objectives;
direct, covering the range of potential outcomes from the decision alternatives;
operational, having attainable data or models to assess the impacts; and
understandable to all parties.
Ideally, performance measures should be as directly representative of stakeholder objectives as
possible, otherwise they can obscure important relationships or trade-offs (Gregory et al. 2012).
Practically speaking, however, many decision processes rely on proxies because they can be
easier to operationalize, depending on available data or models (Fulford et al. 2016a). In
ecological risk assessments, for example, assessment endpoints describe what is cared about and
why, but measurement endpoints are reasonable surrogates of what actually can be measured
(Suter 1990). Guidance can help decision-makers zero in on measures of highest relevance to the
decision at hand by thinking carefully about stakeholders and their objectives, the sensitivity of
different indicators to the particular context, and the comfort level of decision-makers (Bousquin
et al. 2015; Fulford et al. 2016b).
Strategy: Use structured systems as a starting point to identify measurable objectives
A number of tools and approaches, such as the FEGS Classification System (Table 2.2;
Appendix C2) and the EnviroAtlas (http://www.epa.gov/enviroatlas; Appendix C3) can provide
a starting point for identifying ecosystem services related objectives and example indicators for
how to measure them. The EnviroAtlas contains more than 300 metrics, many of which can be
directly linked to FEGS and their beneficiaries (Appendix C3), organized into seven categories
of benefits:
Clean Air;
Clean & Plentiful Water;
Biodiversity Conservation;
Food, Fuel, & Materials;
Natural Hazard Mitigation;
Climate Stabilization; and
Recreation, Culture, and Aesthetics.
Structured approaches to indicator development, such as the FEGS Classification
System, Rapid Benefits Indicators, and Human Well-being Index, can provide a
starting point for clarifying objectives and how to measure them in ways that
reduce ambiguity.
26
-------
In general, FEGS define the production of ecosystem goods and services, but need human inputs
and human interaction to be realized as actual benefits. Indicators of ecosystem services benefits
should account for a number of factors that influence value (Mazzotta et al. 2016; Appendix
C4), including:
Can people actually benefit from an ecosystem service, including whether there is
demand, sufficient quality and quantity, and necessary complementary inputs
(e.g., access points)?
How many people benefit?
How much do people benefit, including whether there are available substitutes
(e.g., an alternative recreational area), the quality of the ecosystem service and
any complementary inputs, and the preferences of the local community?
Are there social equity implications?
Will the ecosystem service be provided reliably over time?
The answers to these questions may help elicit performance measures that more precisely define
what stakeholders really care about, so that decision alternatives can be meaningfully evaluated.
People often think about the more obvious economic benefits of ecosystem services, such as
harvestable fish or timber, but ecosystem services can often have more subtle benefits, that are
equally important to community well-being. Benefits of ecosystem services can include
monetary benefits (e.g., income, property values), health benefits (e.g., reduced risk of illness,
greater physical activity), or benefits to general well-being (e.g., social connections, happiness,
learning). Examples include psychological benefits of feeling connected to nature (Van Den
Berg et al. 2007), cultural and spiritual value of ecological landscapes (Schaich et al. 2010),
visual impacts of non-degraded natural ecosystems in reducing mental stress (Hendryx and
Innes-Wimsatt 2013), improvements in children's educational skills and experience through the
use of natural areas for teaching (Luov 2005), improvements in physical health with access to
outdoor areas (Bell et al. 2008), and promotion of positive social interactions in greener urban
environments (Kuo and Sullivan 2001).
The Human Weil-Being Index (HWBI) allows for a multi-factor perspective by defining well-
being in terms of how ecosystem, economic, 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). A detailed example of using the
HWBI to elicit stakeholder objectives more broadly beyond economic goals is provided in
Appendix C5. The more than 70 metrics composing the HWBI can provide a starting point for
identifying measures to define well-being objectives.
27
-------
Table 3.2. Example of using HWBI conceptually in community-scale workshops to identify
and measure community goals (modified from Fulford et al. 2016b; Smith et al. 2012).
HWBI Domain
Goals identified by participants
Measure
Connection to Nature
Feeling a connectedness to nature
Connection to life
Cultural Fulfillment
Cultural fulfillment
Performing arts attendance
Education
Basic educational knowledge and skills
Positive social and emotional development
More advanced knowledge and skills
Adult literacy rate
Rates of bullying
College graduation rate
Health
Reasonable life expectancy
Physical and mental well-being
Emotional well-being
Good quality healthcare
Healthy lifestyle and behavior
Life expectancy
Life satisfaction
Happiness
Satisfaction with care
Smoking rates
Leisure Time
Enough time devoted to leisure activities
Enough time devoted to vacation
Reasonable time spent working
Activity participation
Vacation days
Working hours
Living Standards
Ability to afford basic necessities
Reasonable income
Reasonable wealth
Job stability and satisfaction
Home affordability
Median income
Mortgage debt
Fear of job loss
Safety and Security
Being and feeling safe
Resilience to hazards
Crime rate
Social Vulnerability Index
Social Cohesion
A sense of place
Engagement in the community
City satisfaction score
Voter turnout
3.4. Integrating Ecosystem Services when Decision Objectives are Pre-defined
Although stakeholders generally perceive that consideration of ecosystem services can help
make more informed decisions (Koschke et al. 2014), a decision process may already be well
under way by the time an ecosystem services perspective is brought in. In such cases, ecosystem
services may be a low priority for consideration or may not be within the set of objectives at all.
Additionally, decisions that occur within regulatory or programmatic frameworks may have
existing environmental management goals, such as the need to meet a specific water quality
criterion or air quality standard. Ecosystem services practitioners must appreciate that the
regulatory or programmatic structure may require a decision, with or without ecosystem services
information (Granek et al. 2010).
Within such contexts, ecosystem services may or may not be specified as objectives. However,
ecosystem services can still complement other project objectives, especially specific regulatory
or programmatic goals. In this case, the question becomes whether ecosystem services or well-
being objectives (or both) can be optimized alongside the regulatory or programmatic goals
under consideration. Furthermore, ecosystem services can provide stakeholders with means for
achieving their existing goals. For example, there may be alternatives for achieving water and air
quality standards beyond source reduction, such as restoring riparian buffers, conserving land for
source protection, or reforesting urban centers.
28
-------
Still, a credibility gap can arise if ecosystem services are not already under consideration as
objectives or if decisions are occurring within an established regulatory or programmatic
framework. In these situations, it is of upmost importance for practitioners and researchers to
produce usable, credible, timely information to decision-makers (Portman 2013). Techniques to
produce this kind of information include the following:
Scientists and practitioners must communicate ecosystem services in terms that are
comprehensible to stakeholders, and sufficient for them to express a relative value
(Wainger and Mazzotta 2011).
The context should be carefully considered by both the stakeholder and decision-makers
so that both groups understand how ecosystem services information will be used or
weighed against other project goals, particularly regulatory goals.
Development and application of ecosystem services models should be transparent,
otherwise decision-makers may lack confidence in the ecosystem services assessments
provided by experts (Townsend et al. 2014).
Across a range of decision contexts and ecosystem services types, there is broad
consensus that co-producing knowledge and data regarding ecosystem services is critical
for stakeholder trust, ecosystem services legitimacy, diverse valuation considerations, and
thus the quality of information provided to decision-makers (Druschke and Hychka 2015;
Posner et al. 2016). The advantage of co-production is that scientists learn more about
what is important to stakeholders, while stakeholders learn more about evidence and its
application, thus building both trust and legitimacy.
Participatory mapping of ecosystem services, especially to include cultural ecosystem
services and values (Scholte et al. 2015), can provide decision-makers information
regarding how stakeholders understand bundling of ecosystem services and relationships
between ecosystem services and place (Klain and Chan 2012).
Stakeholders have expressed that ecosystem services assessments are helpful because it broadens
the scope of the project impact and improves stakeholder commitment (Koschke et al. 2014).
This suggests ecosystem services information can improve the quality of decision-making, even
if the project objectives are already established.
29
-------
4. Develop Alternatives
4.1. Creating Alternatives
Objectives and performance measurements (Chapter 3) help concisely define what matters
about a decision and drive the search for creative alternatives. Decision alternatives describe the
different options available to achieve stakeholder objectives (Gregory et al. 2012; Bradley et al.
2016). Alternatives should only be created after objectives have been fully considered and
understood.
As with development of objectives, an inclusive process to brainstorm alternatives to achieve
those objectives can lead to innovative ideas. Community members often have a strong sense of
connection to their community and a desire to improve it that motivates participation. As
stakeholders, they also often have a strong local sense of what is being threatened, who is
creating the threat, what is feasible in the community, and who or what might be affected
(Gregory et al. 2012; Bradley et al. 2016).
Means objectives (Section 3.2) can provide an initial step toward identifying paths to achieve
fundamental objectives. Decision analysis tools like means-ends networks can help diagram the
relationships between decision options and objectives by asking "ffow can we achieve that?".
Means-ends networks are similar to logic models, which diagram the relationships between
inputs, activities, outputs, and outcomes. If alternatives are not achieving the objectives, they
may need to be iteratively reconsidered or refined (Bradley et al. 2016).
Strategy:
Consider FEGS as means to achieve other objectives
Depending on the decision context, ecosystem goods and services may be
means to achieving other economic, social, health, or general well-being
objectives, and may provide an opportunity for developing creative
alternatives alongside more typical social or economic initiatives.
Decision alternatives have a variety of characteristics that influence their feasibility (Bradley et
al. 2016):
complexity to implement;
effectiveness for achieving a singular objective;
consequences to other objectives, positive or negative;
how big the project is or might become, including spatial scale, time, and cost;
number and types of decision-makers involved;
funding requirements; and
acceptability by the community.
Some decision alternatives may be quickly ruled out because they simply are not feasible in their
spatial scale, cost, or complexity. The remaining alternatives can be evaluated for their
effectiveness by estimating which alternative, or combination of alternatives, is more likely to
30
-------
achieve objectives. Most alternatives will affect more than one objective, and involve trade-offs
across objectives that will need to be considered.
A list of decision alternatives can be organized and merged to create scenarios as appropriate. An
example of using DASEES to link means to objectives, and identify and organize alternatives
into scenarios is provided in Appendix Dl. DASEES helps explicitly document these scenarios
and rank them based on their impacts on stakeholder values.
4.2. Leveraging Ecosystem Services to Achieve Economic and Social Objectives
Just as consideration of FEGS can lead to a broader and more inclusive set of objectives
(Chapter 3), so too can it lead to development of a broader and more creative set of decision
options (Carriger et al. 2013; Fulford et al. 2016a).
Consideration of ecosystem services can give decision-makers a broader perspective on how to
achieve their objectives. In the harbor expansion example (Chapter 3.2), economic development
was a key objective. By ignoring ecosystem services, decision-makers could take a very narrow
view of economic development, looking strictly at the impact of a larger harbor on local business
revenue. With an ecosystem services perspective, decision-makers are able to take a much
broader view of the ways changes to the ecosystem can provide economic benefit to the
community, such as aesthetic value to local residents and water recreational opportunities.
Economic development may still be a fundamental objective, but now in addition to harbor
expansion, improvements in viewsheds, buffering of noise pollution, and maintenance of water
quality for recreation may be additional means to maximizing economic value.
The FEGS concept links intermediate ecosystem services provided by nature to economic
benefits through ecological and economic production functions (Fig. 4.1; Landers and Nahlik
2013). Decision alternatives could target improving the production of FEGS, which have direct
linkages to economic benefits.
Ecological
Production
Function
Environment
Economic
Production
Function
Total
Economic
Value
Figure 4.1. Illustration of how production of Final Ecosystem Goods and Services can
influence the total economic value of an ecosystem.
Total economic value (TEV) aggregates all values, both use and non-use, for all benefit flows
provisioned by an ecosystem both at present and into the future (Fig. 4.2). Use values refer to
benefits gained from ecosystems by beneficiaries using them or interacting with them in some
way. Use values include both consumptive (e.g., raw materials, catchable fish) and non-
consumptive goods (e.g., recreation), as well as indirect uses, which include regulating and
31
-------
supporting ecosystem services (e.g., water purification, pollination, health benefits). Non-use
values refer to those values that do not necessarily involve a person ever interacting with an
ecosystem. Non-use values include both bequest values, where a person derives value from the
knowledge that future generations will have access to an ecosystem, and existence values, where
a person values that the ecosystem will continue to exist. Option values encompass the future use
of known and unknown benefits (Pascual et al. 2010).
Option value
Bequest value
Use value
Altruism to
biodiversity
Consumptive
Non
consumptive
Actual Value
Indirect use
Non-use
value
Existence
value
Philantropic
value
Direct use
Altruist value
Total Economic Value
Figure 4.2. Ecosystem services values typology within the TEV paradigm. Adapted from
Pascual et al. (2010).
Another approach, the Human Weil-Being Index (HWBI), provides a generic means-ends
diagram that allows for an even broader perspective. Elements of well-being (e.g., health, living
standards, cultural fulfillment) are the fundamental objectives (see Chapter 3.3), and ecosystem
services provide a means to achieving them that could be considered alongside other economic
or social services (Fig. 4.3). A detailed example of using the HWBI as a skeleton to link decision
alternatives to FEGS and human well-being is in Appendix D2.
Structured paradigms like FEGS or HWBI can help identify alternatives that take advantage of
an ecosystem services viewpoint. For example, decision options aimed at increasing non-
consumptive or indirect uses of ecosystem services can help maximize the economic value of
ecosystems (Fig. 4.1, Fig. 4.2); greenspace may be an alternative to education initiatives or
technology for improving educational outcomes (Fig. 4.3). Not all the endpoints in these
paradigms may be relevant to a particular community or context. For example, bequest or
existence values may have little relevance to a decision where an objective is to increase
economic revenue. For decision-makers or stakeholders that are largely unfamiliar with
ecosystem services, strawman lists can provide a starting point for brainstorming potential
ecosystem services as alternatives alongside economic and social alternatives (Appendix D2).
Strategy: Use structured paradigms to link EGS alternatives to broader objectives
yH Structured paradigms, such as FEGS or the Human Well-being Index (HWBI) can
l T' provide a starting point for identifying alternatives that leverage ecosystem services
(intermediate or final) to achieve economic or well-being objectives.
32
-------
Means Objectives
Fundamental Objectives
Maximize Economic Services
Capital investment
Employment
Innovative technologies
Production
ป Consumption
Redistribution
Maximize Human Well-being
Connection to nature
ฆ Cultural fulfillment
Education
Human health
Leisure time
Living standards
Safety and security
Social cohesion
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
Figure 4.3. Illustration of how ecosystem services, alongside social and economic services,
can be means to improving components of human well-being (from Smith et al. 2014).
4.3. Integrating Ecosystem Services when Decision Alternatives are Pre-defined
In some cases, decision alternatives may have already been defined by the time an ecosystem
services perspective is brought into the process. Examples might include an ecological risk
assessment for a particular risk management scenario, or development of an environmental
impact statement for a limited set of planned alternatives. In such cases, a consideration of
ecosystem services is largely aimed at comparing existing options or evaluating the benefits of a
chosen option to inform future decisions (Fulford et al. 2016a).
For ecosystem services practitioners working on a project where decision options are already
defined, there are additional considerations. It is important to understand the process by which
stakeholders provide information to decision-makers. Practitioners and researchers should
integrate ecosystem services assessments into the existing process to maximize trust and
credibility (Posner et al. 2016). These processes vary in many dimensions. For example,
stakeholders may be asked whether they prefer one or many among a set of fixed alternatives.
Alternatively, the process might allow for public comment periods, where stakeholder feedback
could allow the alternatives to be modified in response to ecosystem services assessments.
33
-------
5. Estimate Consequences
5.1. Estimating Effects of Decision Alternatives on Performance Measures
Once objectives (Chapter 3) and potential alternatives to achieve them (Chapter 4) are
identified, the next step in the decision process is to estimate the potential consequences of
alternatives on each of the multiple objectives identified by stakeholders, as defined by the
performance measures (Gregory et al. 2012).
In time or budget-limited processes, decision sketching or the development of a conceptual
model may be sufficient to characterize the consequences of alternatives on objectives and make
a decision (See Chapter 2.4). In other cases, information and data may need to be collected,
analyses may need to be conducted, or experts may need to be consulted.
Information collection should be prioritized to what is actually needed to make a decision - that
is, needed to estimate consequences of alternatives on performance measures. Time and
resources could be spent collecting data that measures the wrong thing, or developing complex
models when a simple model would suffice. Some contexts or decision-makers may be able to
tolerate greater uncertainty, and need less intense data collection or less complex models to make
a decision. Uncertainties can arise from a number of different places including lack of data, data
measurement error, model availability and assumptions, or even from ineffective communication
(Gregory et al. 2012). This is particularly true for ecosystem services assessments, as
environmental processes are inherently characterized by high uncertainty (NAS 2013).
Strategy:
Prioritize information and analysis to what is actually needed
i
&
Information collection and application of tools should be prioritized to what is
needed to estimate consequences of alternatives on measurable objectives, and to
reflect the uncertainty decision-makers are able to tolerate. Complex FEGS
assessments or economic valuations may or may not be needed.
Information about consequences can come from three main sources (Gregory et al. 2012;
Bradley et al. 2016):
Group deliberations: Consequences may be evaluated qualitatively through stakeholder
deliberations. Influence diagrams or graphical conceptual models can be used to describe
the likely effects of decisions on objectives through changes in intermediate variables
(Marcot et al. 2012; Harvey et al. 2016).
Targeted studies and predictive modeling: Influence diagrams or conceptual models
can be turned into predictive quantitative assessments based on empirical data and/or
computational models. Studies may be conducted to collect data (e.g., surveys, field
monitoring, literature reviews) as needed.
Expert judgments: In some cases, group deliberations may raise questions where
empirical data or models are unavailable to provide answers. In such cases, analysts may
rely on expert judgments to reduce uncertainties in predicting consequences.
34
-------
The goal of each approach is ultimately the same: to estimate the consequences of each
alternative on the performance measures. Both the conceptual and mathematical models used to
estimate the consequences of alternatives on FEGS tend to be complex. The complexity requires
a more detailed technical approach to incorporating ecosystem services at this decision step than
are necessary at other steps.
5.2. Conceptual Models as the Foundation for FEGS Mathematical Models
FEGS conceptual models are hypotheses about the flow of cause-and-effect linking the decision
alternatives to changes in stakeholder objectives, as reflected by the performance measures. They
form the basis for any attempt to estimate the consequences of alternatives. In some cases,
qualitative approximations of those outcomes may be sufficient to allow comparison of the
alternatives. For example, measured objectives may be clearly higher under one alternative
versus another, and a more precise estimate would not change the decision. Qualitative or semi-
quantitative network modeling approaches can be useful for examining socio-ecological impacts,
engaging stakeholders, and evaluating tradeoffs (Gray et al. 2012; Cook et al. 2014; Reum et al.
2015; Harvey et al. 2016). In other cases, a greater level of precision is needed or desired to
accurately assess tradeoffs. In those cases, the conceptual links pictured in the diagram are
converted to mathematical expressions. Those expressions should reflect current, quantitative
evidence about the underlying causation. Mathematical models must be found, or developed, that
adequately reflect the conceptual model. Adherence to the conceptual model - including the
management actions, stressors, ecosystem characteristics and FEGS-related indicators that it
comprises - helps to avoid the fallacy of using models selected solely for reasons of familiarity
or convenience.
Strategy: Use conceptual models to visualize relationships
Conceptual models or influence diagrams visualize cause and effect or
information flow between decisions, ecosystems, ecosystem services, and
benefits. They provide the foundation for assembling data, local knowledge,
expert opinion, or mathematical models needed to estimate consequences.
But before selecting analysis models, the conceptual model itself should be re-examined. First, a
combination of expert judgment and stakeholder consensus should identify any pathways for
which mathematical modeling is unnecessary - for example, those judged to have negligible
impacts. Next, some pathways may be excluded from the modeling effort because information
about them is insufficient; these may need to be treated qualitatively. It should be noted,
however, that fuzzy cognitive mapping or Bayesian belief networks can enable computational
analysis even when information is limited (Bagstad et al. 2014; Bousquin et al. 2014; Gray et al.
2015). Finally, critical sources of variability that may not have been recognized during earlier
stages of analysis - because they were not considered to be part of the decision - may need to be
added to the conceptual model and included in mathematical modeling. These may include long-
term cycles of environmental variability (e.g., periodic drought), predicted trends (such as
climatic warming or land-use change), or feedback loops (e.g., changing fish availability
influencing fishermen behavior). If not accounted for, these influences could invalidate modeling
efforts and lead to unintended consequences.
35
-------
While conceptual models of environmental decision-making using FEGS tend to be complex,
they can be summarized by the generic conceptual model shown in Figure 5.1. The generic
diagram uses three arrows to distinguish three types of estimation processes typically required:
the influence of management actions on the state of the natural environment; Wainger
and Mazzotta (2011) called these the impact functions;
changes in the production of FEGS by the natural environment (ecological production
functions or EPFs); and
the relationship between changes in FEGS and changes in well-being (benefit functions).
An optional fourth arrow shows the potential intervention of important external drivers (e.g.,
climate change) as a source of variability that may also require estimation.
A Human
Well-being
Final EGS
Decision
Alternatives
Social & Economic
Services
A Ecosystem State
(& intermediate EGS]
Information for Decision Support
Figure 5.1. Generic conceptual model showing the use of FEGS in decision support
analyses for community well-being. EGS = ecosystem goods and services.
Different researchers have divided these generic stages differently (Munns et al. 2015 a; Potschin
and Haines-Young 2011; Wainger and Mazzotta 2011), and computational models also vary in
their approach to these phases. In many cases, ecological production functions include impact
functions (Bruins et al. 2017). Integrated modeling systems may comprehensively simulate all
ecological and social dimensions of a decision problem, including their dynamic interactions
over time, within a single computational system (Turner et al. 2016). These integrated systems
may have the characteristics of Decision Support Systems (DSS). The use of DSS to manage the
evaluations of scenarios and trade-offs is discussed in Section 5.6. In other cases, modeling is
more piecemeal; different models address different aspects of the estimation problem, with
varying degrees of computational integration between the models.
The distinctive aspect of a FEGS approach is that it requires specific consideration of the nature
of each final good or service, in the context of its use by a specific beneficiary in a specific
environmental setting. Computational models estimate expected changes in the metrics or
indicators selected to correspond to these FEGS. Whether this occurs within a comprehensive
36
-------
ecological-economic model or at a point of linkage between different models, quantitative values
of these FEGS metrics or indicators can be estimated.
5.3. Ecological Production Functions
Selected computational models should correspond to the ecological production functions (EPFs)
indicated in the conceptual model. 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. 2017). To be useful for comparing multiple
objectives, EPFs would rarely be used in isolation, but instead coupled in systems models that
account for relationships and feedbacks among variables. Desired attributes of EPFs are listed in
Table 5.1.
Table 5.1. Desired attributes of ecological production functions. Source: Bruins et al. (2017).
Desired attributes of ecological production functions
1. Estimate Indicators of Final Ecosystem Services: Understanding of intermediate services is
useful, but EPFs that estimate final services (i.e., those directly meaningful to human beneficiaries)
are most valuable to decision-makers.
2. Quantify Ecosystem Service Outcomes: EPFs that yield qualitative outcomes are sometimes
useful for scoping and mapping, but quantification can be more informative for analysis of
ecosystem services trade-offs.
3. Respond to Ecosystem Condition: Since delivery of ecosystem services may vary with
ecosystem condition, EPFs should change when ecosystem conditions change.
4. Respond to Stressor Levels or Potential Management Scenarios: EPFs should include
variables necessary for evaluating stressor impacts and predicting the outcome of management
scenarios.
5. Appropriately Reflect Ecological Complexity: EPFs must reflect critical complexities (e.g.,
nonlinearities and feedbacks affecting ecosystem services provision) while remaining simple
enough to be understandable.
6. Rely on Data with Broad Coverage: EPFs must be able to perform using typical data (i.e. those
available for most geographic areas) to maximize their transferability.
7. Are Shown to Perform Well: Because EPFs are used to evaluate hypothetical scenarios, it is
important to consider whether EPF performance has been evaluated for situations similar to those
facing the decision-maker.
8. Are Practical to Use: EPFs should run on conventional personal computers, produce usable
results with modest data input, and be usable by people other than trained modelers.
9. Are Open and Transparent: EPFs should be thoroughly documented and codes should be
publicly available, although well-documented proprietary models may be useful in some situations.
As the desired attribute list makes clear, an ideal EPF would have the ability to both evaluate
management alternatives (thus directly incorporating the impact function) and estimate indicators
of FEGS. In practice, several models may be needed to address all ecosystem impacts of the
alternatives as well as all affected FEGS.
37
-------
Aligning computational EPFs to the conceptual model requires three kinds of alignment:
1. alignment of processes represented by the computational model to the ecosystem
processes affected by the management actions;
2. alignment of model predictor variables to key factors affected by the
management actions; and
3. alignment of model response variables to metrics or indicators of FEGS.
In addition to these conceptual model alignments, it is necessary to:
align model data requirements to the available data; and
align model complexity to the scope and scale of decisions.
The complexity of models needed will depend both on the nature of the ecological processes
affected by the management actions and by management scale (Appendix El). In some cases,
processes are relatively simple and can be represented by simple models, such as when service
provision is a function of the presence/absence of key assets. These simple models are readily
adapted to correspond to locally relevant definitions of FEGS. For example, in an estuary subject
to contaminated sediment removal and ecological restoration, Angradi et al. (2016; Appendix
E2) used bathymetry and several logic-based models to map the provisioning of 23 final services
under different restoration scenarios. As one example, water body areas suitable for sailing were
those > 1.5 m deep, > 100 m offshore and accessible to a marina. Models of recreation provision
by upland areas may have a simple, logic-based character when they are based on
presence/absence or proximity of assets that matter directly to recreators (Casado-Arzuaga et al.
2013; Paracchini et al. 2014). Often, a goal is to produce broad-scale maps of intermediate or
final service distribution, corresponding to given points in time (Costanza et al. 1997; Costanza
et al. 2014; Maes et al. 2012). In such cases, simplified models may reasonably account for
ecological assets at county, regional, or national scales.
Strategy:
Quantify FEGS with ecological production functions
i
A number of mathematical modeling tools, ranging from fairly simple lookup
tables to complex biophysical models, can quantify the effects of alternative
scenarios on provisioning of ecosystem services through the use of ecological
production functions, EPFs.
By contrast, management alternatives sometimes interact with complex ecosystem processes that
resist simplification and are most readily addressed using vetted software packages, adaptable to
user needs. A few examples include models of surface/subsurface hydrologic transport (e.g.,
VELMA; Abdelnour et al. 2013; see Appendix E3), plant-soil interactions in watersheds
(SWAT; Francesconi et al. 2016), circulation in tidal estuaries (EFDC; Devkota and Fang 2015),
atmosphere-biosphere interactions (CMAQ; Cooter et al. 2013) and species population dynamics
(HexSim; Huber et al. 2014; http://www.hexsim.net/).t When the outputs of these packaged
models do not correspond directly to FEGS, linkages to add-on functions may be required. These
linkages may be managed within a DSS, allowing simultaneous scenario estimation for all
t See List of Abbreviations and Symbols for full names of models.
38
-------
FEGS. One study of the impacts of upland activities on coral reef ecosystem services used the
Envision decision support platform to link models of coral reef ecological processes to 28 EPFs
corresponding to specific, FEGS-related endpoints (Orlando and Yee 2017; Fulford et al. 2016a).
Empirical or statistical modeling approaches provide a level of complexity in-between dynamic
process models and simple logic models (Appendix El). Bayesian belief network (BBN) models
rely on probabilistic inference from existing data and knowledge. Bayesian Belief Networks are
readily adaptable to varying levels of data availability and can be updated as knowledge
improves. For example, in a regional study, Bagstad and coworkers (Bagstad et al. 2011; Bagstad
et al. 2014) used available spatial data sets to create BBNs to model scenic viewsheds and open
space proximity for homeowners and flood regulation services for developed land in a 100-year
floodplain. The empirical BBN models took into account spatial locations of ecological
resources providing services, access to resources, and factors that detracted from enjoyment of
the resources. Empirical methods can also be used to build simplified models, having reduced
data and computational demands, from the input and output datasets of process-based models
(e.g., Nedkov and Burkhard 2012). The degree of complexity of many EPFs can be assessed
using the EcoService Models Library (ESML; https://esml.epa.gov/epf_l/public/signup;
Appendix E4).
Besides ensuring alignment to ecological processes, a scan of model predictors (i.e., input
variables) in models under consideration can determine whether they correspond to stressors or
management levers altered by the management alternatives. If actions are intended to reduce a
stressor, then stressor concentration or loading should be a model variable. If actions include
ecosystem restoration, such as wetland or forest re-establishment, the models must include
variables that correspond either to the restoration actions themselves or the proximal expected
effects of the actions. For example, VELMA includes several variables (e.g., time since
disturbance) used to compute a post-clear-cut recovery function, for examining the hydrologic
implications of different forest management practices (Abdelnour et al. 2011; Appendix E3). A
model of alternative trajectories of landscape change in the Puget Sound region, executed using
the Envision platform (Bolte and Vache 2010), enabled scenario-based manipulation of regional
development variables such as percent impervious surfaces, percent nearshore development, and
number of marinas. A model predicting occurrence of submerged aquatic vegetation (Angradi et
al. 2013) included variables (e.g., water depth, bed slope) relevant to removal and repositioning
of riverine sediment. Descriptions of predictor variable for many existing EPFs can be quickly
inspected using ESML (Appendix E4).
Correspondence of response variables to FEGS metrics or indicators can also be examined when
choosing models. As in the case for predictor variables, response variable descriptions for many
existing EPFs can be examined using ESML (Appendix E4). For EPFs entered in ESML,
predictor variables have been evaluated for correspondence with ecological end product
categories and subcategories, as classified by the National Ecosystem Services Classification
System (NESCS; EPA 2015). NESCS end products correspond to FEGS, and were developed
based on the FEGS Classification System (Landers and Nahlik 2013).
39
-------
5.4. Ecological Benefits Functions
In most cases, estimating the consequences of decisions on ecosystem services should include a
direct connection from those services to human beneficiaries (EPFs to EBFs; Figure 5.1). This
connection allows for more direct and meaningful assessments of impacts when production of
FEGS may be affected by a decision. While benefits can be universally understood through a
multi-dimensional concept of well-being (Munns et al. 2015a), impacts of ecosystem services on
well-being may be measured in a number of different ways (see Chapter 4.2), and numerous
approaches have been applied to quantify well-being.
In this section we summarize ecological benefits functions from three dominant benefit
viewpoints:
monetary value;
human health; and
human well-being.
Note, these approaches are not independent and have significant overlap. Monetary values can be
placed on human health (e.g., costs of health care), and environmental economists often consider
Total Economic Value (TEV) as a way to quantify well-being. Human health endpoints (e.g.,
mortality or disease rates) are also commonly assessed as a measure of well-being for outcomes
that are difficult to monetize (Reacher et al. 2004; Gariepy et al. 2014). The final approach,
human well-being, attempts to provide a comprehensive non-monetary measure by integrating
metrics of health and wealth with other factors such as culture, safety, and social cohesion
(Summers et al. 2014; Dyrbye et al. 2016). All three approaches have strengths and weaknesses
worth considering, and could be used individually or in combination when assessing benefits
from FEGS.
Strategy: Let objectives drive choice of methods for FEGS benefits analyses
Choice of methods to estimate ecosystem services benefits (EBFs) should
primarily be driven by 1) the benefits endpoints under consideration as objectives
and 2) the depth of analysis required to make a decision. Tool assumptions and
input data needs also influence whether implementation of a method is practical.
The development of predictive or descriptive tools is similar across all three benefit categories,
but the resulting tools are dependent on different assumptions and input data. Tools in general
can be categorized across a continuum from:
conceptual tools that elucidate connections between ecosystem services and benefits
(Eco-Health Relationship Browser; Jackson et al. 2013), but do not attempt to
estimate outcomes; to
semi-quantitative tools that make predictions based on associative relationships (e.g.,
i-Tree, www.itreetools.org); to
fully quantitative models that predict outcomes (NOAA 2009; Dorn et al. 2014)
optimally with some estimate of uncertainty appropriate for assessment of risk (Miller
et al. 2005).
40
-------
These are similar to ecological production functions discussed in the previous section (Section
5.3), but have human benefit metrics as their output. The type of tool selected will depend on the
specific benefits and management actions under consideration. Here we focus on data needs and
outputs to compare different approaches for benefit functions.
Monetary endpoints are by far the most dominant category in terms of available tools (Birol et al.
2006; Johnston and Wainger 2015; Yoskowitz et al. 2016). This also includes endpoints that
clearly represent economic markets (e.g., commercial fish harvest) from which a monetary value
could easily be calculated or inferred. Aside from the plethora of choices, these tools are also
useful for comparing benefit across disparate issues, as they focus on a common endpoint
measure (e.g., monetary value). However, available valuation tools tend to be biased towards
assessing benefits from consumptive services, as values of market goods are often more
straightforward to quantify (Tuya et al. 2014). In cases of marketable commodities (e.g., natural
resource products) this is a relatively straight forward exercise involving economic models of
supply and demand (Ronnback 1999; Samonte-Tan et al. 2007; Fulford et al. 2015b). Yet, many
ecosystem goods and services, such as habitat or carbon storage, are hard to monetize, as people
do not buy and sell them directly. Instead, they provide non-consumptive services to a set of
beneficiaries that is difficult to define. In the latter case, value is typically estimated based on
preference models (Beaumont et al. 2008; Mangi et al. 2011; Appendix E5) that substitute stated
or revealed preference of stakeholders for a more direct estimate of cost.
An extensive literature exists on the subject of preference models and the application of such
models to general monetary valuation is well-supported (Castro et al. 2016; Irvine et al. 2016;
Romo-Lozano et al. 2017). However, these models can be controversial as a predictive tool,
because of underlying assumptions and the large potential for bias resulting from an often
inadequate beneficiary sampling frame (Brouwer 2006; Christie and Gibbons 2011).
Beneficiaries are asked to convert their perceived preference for a good or service into monetary
terms. However, many unstated or unmeasurable factors can influence the outcome, including
proximity of respondents to the good or service, demographics of respondents, or even state of
mind of respondents during the survey. Despite these issues, stated preference models remain a
common tool for estimating value of ecosystem services to beneficiaries because they allow
straightforward comparison of the results across beneficiary groups and decision contexts.
Measures of health benefits from ecosystem goods and services are another prolific area of
research and policy (Appendix E6). Human health endpoints bridge the gap between
consumptive (e.g. diet) and non-consumptive (environmental quality) services (Reacher et al.
2004; Gariepy et al. 2014). Health outcomes can be divided into two broad categories (Jackson et
al. 2013):
acute or chronic health problems that could be buffered by natural ecosystems
(Reacher et al. 2004; Wade et al. 2014); or
the health promotional role of nature on physical activity, mental health, and
general healthiness (Gariepy et al. 2014; Lee and Duk-Chul. 2014).
Most progress in development of health benefit functions has occurred in the former category
where integrated public health and environmental research has generated abundant data linking
environmental quality to health outcomes. Examples include the relationship between access to
41
-------
green space and reductions in heat morbidity (Harlan et al. 2013). Here, the impact of preserving
shading/transpiration capacity of greenspace can be clearly connected to changes in incidence of
heat morbidity in urban ecosystems. A number of models and tools (e.g., i-Tree, BenMapt) exist
to predict effects of ecosystem services on health outcomes, including respiratory diseases, heat-
related morbidity, cardiovascular diseases, and mortality (reviewed in Oosterbroek et al. 2016).
However, many tools rely on proxies to quantify health impacts, and while some ecosystem
services (i.e., air purification, natural hazard protection) are fairly well represented, there are
few, if any, tools available for other kinds of health outcomes (e.g., mental disorders, injuries,
infectious diseases) or other kinds of health-related ecosystem services, including biological
control of infectious diseases and promotion of social interactions (Oosterbroek et al. 2016).
Promotional benefits of the natural environment on general healthiness may be particularly hard
to quantify (Breslow et al. 2016). Ecosystem services, such as swimmable water, clean air, and
accessible green space, have well-accepted relationships to beneficiary activity level, fitness, and
quality of life (Bryce et al. 2016), but data to effectively link these to measurable health benefits,
particularly given the numerous social, economic, and behavioral correlates (e.g., regional
differences in the popularity of outdoor activities), are challenging to collect. Some models and
tools use proxies, such as reductions in the frequency of doctor visits or reduced healthcare costs,
to quantify such promotional benefits (Oosterbroek et al. 2016). As a result, predictions of
promotional health benefits to evaluate decision options are often based on qualitative
relationships or educated guesses. Their quantification as causal relationships (or atleast robust
correlations) is a promising area of research for development of benefit production functions.
In cases where tools to link ecosystem services to health outcomes are lacking, approaches that
leverage more general evidence-based assessments or expert opinion may be better suited.
Health Impact Assessments (HIA; Appendix E7) are conducted on a community by community
basis, and rely on expert opinions to predict likely health outcomes across a suite of decision
options (Rhodus et al. 2013). A formal analysis, such as a metadata analysis, can be used to
assess the degree of scientific evidence for potential health outcomes (Norris et al. 2011; de
Jesus-Crespo and Fulford 2017).
The final and most general category of ecological benefits functions includes functions that
attempt to predict or quantify human well-being using non-monetary measures of well-being.
Measures of well-being are typically composite indices of multiple metrics across a suite of
categories developed based on expert opinion (Smith et al. 2013; Ferrara and Nistico 2015;
Dyrbye et al. 2016). Measures of well-being are used globally to assess and summarize benefit in
a consistent way across spatial areas or over time (e.g., Gallup Healthways Well-being Index,
http://www.well-beingindex.com/; Canadian Index of Well-being, http://uwaterloo.ca/canadian-
index-well-being).
Human well-being is a broad concept covering multiple benefits, including the monetary and
health benefits previously discussed. The advantage of considering multiple types of well-being
benefits is that it
t See List of Abbreviations and Symbols for full names of models.
42
-------
may be more closely aligned with what stakeholders value about their community
than monetary measures (Fulford et al. 2016b; Appendix C5);
provides a method to quantify well-being holistically as a single objective -
enabling comparisons with other types of objectives (e.g., cost of a project); and
provides a flexible approach for examining trade-offs across multiple objectives
and their metrics of well-being that may not be amenable to standard cost-benefit
analyses (e.g., living standards vs. connection to nature vs. social cohesion).
Several efforts have been undertaken to connect the provisioning of ecosystem services to
measures of well-being, paving the way for a well-being benefit function (Gordon and Folke
2000; Pereira et al. 2005; Pinto et al. 2014). One example is the Human Weil-Being Index
(HWBI, Smith et al. 2013, Appendix E8), a compilation of metrics across eight domains of
well-being. Recently developed models relate measures of economic, social, and ecosystem
services to changes in HWBI, and can be applied to investigate how changes in ecosystem
services might impact different domains of well-being, or well-being as a whole (Summers et al.
2016). Though based on nationally consistent county-scale metrics, these models may be
informative for predicting potential qualitative outcomes for well-being under different decision
scenarios (e.g., an increase in greenspace), and provide a transferable approach for development
of finer scale models, with alternative metrics where needed, if community-scale data is
available.
5.5. Data Availability and Model Transferability
Developing models and techniques for quantifying impacts on FEGS and their benefits can be a
complex, highly technical process that is time-consuming and data-intensive. Often, the timeline
of decisions is faster than the pace of ecosystem services research (Granek et al. 2010). Decision-
makers are often faced with insufficient time to collect data specific to their site or develop
models specific to their needs.
Practitioners and researchers may need to be willing and ready to work with available data on
projects. Appropriate information or technical data may be limited regarding the future
biophysical state of the ecosystem (e.g., preferred land use, topography or bathymetry), and the
ability to control a future biophysical state adds uncertainty to the FEGS assessment. Further,
ecological and decision-scales may be poorly matched (e.g, fish habitat may function on a spatial
scale far larger than the decision scale regarding future ecological restoration; Angradi et al.
2016).
Practitioners can explore the potential of transferring existing measurements or models from the
original site for which they were collected or developed, to their current site. Inappropriately
transferring models or measurements between sites can lead to inaccurate estimates of FEGS,
and consequently, decisions based on inaccurate information. While benefit-transfer methods
exist to transfer economic models and measurements (e.g., Johnston and Rosenberger 2010), a
formal methodology for transferring ecological models and measurements has only recently been
developed. That methodology is detailed in Appendix E9.
43
-------
5.6. Using Decision Support Systems to Integrate Information and Compare
Scenarios
As explained in Section 5.2, mathematical models used to estimate consequences should
correspond to the conceptual model developed for the decision problem. Because multiple
models and model scenarios may be required, it is useful to employ a Decision Support System
(DSS). A DSS is an interactive computer-based system that can aid decision-makers in
identifying and solving problems, and making decisions. These systems may use data from
observations, output from statistical or dynamic models, or rules based on expert knowledge.
They manage input data to define scenarios and may employ multiple linked models (e.g., EPFs
and EBFs) to model outcomes. They may include explicit methods to estimate uncertainties,
such as through specification of probabilities using BBNs, or through estimation of confidence
intervals from multiple model runs that draw parameters from a specified range or incorporate
environmental stochasticity. Some also include the ability to conduct sensitivity analysis, to
investigate which model parameters are having the largest impact on outcomes of interest.
Sensitivity analysis can help target information collection to maximize accuracy of the most
sensitive parameters, or may even lead to decision alternatives being revisited, if unanticipated
factors are discovered to have potentially large impacts.
As illustrated by Figure 5.2, many or all of the functions and informational linkages that were
pictured in the generic conceptual model of a decision problem (Fig. 5.1) can be simulated
within a DSS. As shown in Figure 5.2, however, the selection of decision alternatives is now
substituted by scenario modeling. Besides simulating decisions, scenarios may also simulate
alternative future conditions - such as land-use change, climate extremes or climate change -
that are external to the decision itself but could alter system responses to the decision.
Social & Economic
Services
Decision
Alternatives
Impact
Functions
A Ecosystem State
(& Intermediate EGS)
' V"
' # ฃr
Ecol. Prod
Functions
Benefit
A Final EGS
Functions
A Human
Well-being
Information for Decision Support
Figure 5.2. Generic depiction of a Decision Support System. Dashed lines emphasize the
role of scenario modeling.
44
-------
One DSS can differ from another in features such as comprehensiveness and time dependence.
For example, the EPA H20 tool (Russell et al. 2015; Appendix El) uses static land-use
conditions to drive the following EPFs for each subwatershed included in the model:
stormwater retention, based on a soil-specific calculation of water retention;
excess nutrient removal via denitrification by land use, based on estimates from
peer-reviewed literature
air pollutant removal and carbon sequestration by vegetation using the UFORE or
i-Tree Eco model (USDA Forest Service 2012); and
an arithmetic summation of geographic features of ecological interest.
The EPA H20 Tool also includes ecological benefits functions that assign monetary values to
the outcomes of each of these processes (i.e., flood protection, clean water, clean air, and a
stabilized climate).
The integration of EPFs and benefit functions in EPA H20 allows for a quick preliminary
examination of multiple land-use change scenarios such as watershed urbanization combined
with shoreline restoration (see example in Appendix El) in a simple to run open-source GIS
based software package using nationally available datasets. However, each scenario entails only
a snapshot of land-use conditions at a given point in time. This DSS coordinates the calculation
of changes in FEGS and monetary values but lacks feedbacks to modeling scenarios. In contrast,
Envision (Bolte and Vache 2010), is also driven by land-use changes, but incorporates policy
functions that correspond to decision alternatives in Figure 5.2, and yearly feedback from the
landscape that can modify the application of policy functions. Modeling this kind of feedback is
common in fisheries management models and has broad transferability to terrestrial conservation
(Bunnefeld et al. 2011). After computing EPFs, the Envision system can calculate metrics, which
may include FEGS, to evaluate the acceptability of the landscape at a given time t, and provide
feedback to the policy function, altering land use at time t+1. Envision can thereby simulate the
dynamic process of policy-driven landscape change over time.
Strategy: Use Decision Support Systems to organize and link FEGS analyses
$>
Decision Support Systems (DSS) can help engage stakeholders in a step-by-step
process by organizing information and models linking decisions to ecosystem
services (EPFs) to benefits (EBFs), to facilitate estimation of consequences.
DSS must overcome issues of data-model and model-model compatibility by ensuring semantic
and spatiotemporal alignment at each point of linkage. Envision employs a spatially-explicit
multi-agent construct within a set of polygon-based GIS maps to model the value-based behavior
of decision-makers, policy intentions, landscape change processes, and landscape production via
a series of model plug-ins that conform to the software architecture (Bolte 2014). The developers
of A RTFS (ARtificial Intelligence for Ecosystem Services) are incorporating ontologies that
define a core vocabulary for ecosystem services and what is needed to quantify them, which
facilitates automated model integration. (Villa et al. 2014).
45
-------
5.7. Comparing Scenarios with Consequence Tables
Consequence tables, also known as management option matrices (LoSchiavo et al. 2013), are a
useful tool to display the effects of different decision alternatives across multiple objectives
(Gregory et al. 2012; Appendix E10). Consequence tables are typically represented as a matrix
linking the alternatives explicitly to the performance measures (Table 5.2). Practitioners
populate the cells with information from group deliberations, expert judgements, or quantitative
analyses of data. Unlike a standard cost-benefit analysis where comparisons are made across
monetary values, the entries in a consequence table do not need to be in the same units.
Evaluated performance measures could be monetary, biophysical units, percentages, or even
categories - as long as they are clearly defined to reflect the objectives.
Table 5.2. Hypothetical example of a consequence table (from 3VS, Tenbrink et al. 2016).
Objective
Measure
Alternative 1
Additional Waste
Water Treatment
Facilities
Alternative 2
Aquaculture
Alternative 3
Low Impact
Development
Maximize
water quality
Change in summer monthly
nitrogen loading relative to
status quo
-86,000 kg/month
-6,600
kg/month
-53,000
kg/month
Maximize
beach
recreation
Total additional visitors each
summer relative to status
quo
13,000 visits
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
grass habitat
Change in index of eel grass
quality
15%
No change
0.4%
Minimize costs
to taxpayers
Annual per capita cost under
30-year financing
$55/person/year
No cost to the
public
$17/person/
year
Consequence tables help focus the conversation on the information needed to estimate
consequences (Gregory et al. 2012). In many cases, budget or time limitations may prohibit
consequence tables from being fully populated with models and data. However, the most
important role of a consequence table often is not an extensive quantitative analysis of
alternatives, but instead to identify key uncertainties, expose key trade-offs, and provide a
communication tool for stakeholders and decision-makers (Gregory et al. 2012; Bradley et al.
2016).
Consequence tables facilitate a side by side comparison of multiple kinds of measures, chosen to
best represent what stakeholders care about. This could include intermediate ecosystem services,
FEGS, benefits of FEGS expressed as monetary values, health endpoints, or a well-being index,
as well as other kinds of objectives, such as the cost of a project or tax revenue. Summarizing the
outcomes of decision scenarios in this way may also expose inadequate measures or objectives
that fail to capture what is really important to stakeholders. For example, is a predicted decline in
an intermediate ecosystem service like "water quality" adequately capturing benefits to the
community (Table 5.2), or would a more direct measure tied to specific beneficiaries, e.g., FEGS
46
-------
such as biomass of fishable species or visibility for snorkeling, allow for more meaningful
comparisons? Consequence tables can also display output from alternative model tools (e.g.,
In VEST vs. EPA H20), to investigate to what degree uncertainty in outcomes might or might
not affect a decision.
Strategy:
Compare alternatives and gain insights with consequence tables
i
Consequence tables are a useful tool to display effects of decision alternatives on
stakeholder objectives. They can lead to insights where better models (EPFs,
EBFs) are needed or where ambiguous measures could be improved, such as
with measures that are more directly relevant to beneficiaries (FEGS).
As with other parts of an SDM, estimating consequences can be an iterative process (Gregory et
al. 2012). Attempts to estimate consequences can bring to light key uncertainties, explore the risk
tolerance of decision-makers, and target where additional information may be needed (Rehr et al.
2014). Initial efforts can lead to insights about:
Additional information needs: Is the uncertainty around estimates of
consequences on performance measures too high? Can additional data be
collected or better models be used? Would it affect the decision?
New or combined alternatives: Are any of the proposed alternatives able to
attain the desired effects on performance measures? Can the proposed
alternatives be modified or combined (or new alternatives proposed) to achieve
multiple benefits across performance measures or reduce negative trade-offs?
Refined performance measures: Are the performance measures representing
what was intended? Are they sensitive to the decision options? Are there
alternative performance measures that better represent the intention, risk
tolerance, or help eliminate ambiguity?
Refined objectives: Do the estimated consequences on performance measures
broadly represent what really matters? Or is something missing?
47
-------
6. Evaluate Trade-offs and Take Action
6.1. Evaluating Trade-offs
Once consequences of alternatives on stakeholder objectives have been estimated (Chapter 5),
clear win-win alternatives or low-hanging fruit may emerge (Gregory et al. 2012). Often,
however, the next step is to explore trade-offs that stakeholders are willing to make among the
objectives. In other words, how much of one objective are stakeholders willing to sacrifice to
have more of another (Keeney 1992)? There is often a temptation to rank or prioritize objectives
early in the process, prior to any analysis. However, trade-offs very much depend on the
quantities that are actually at stake with the decision. Stakeholders may be willing to accept a
small loss in something "very important" in order to prevent a large loss in something "less
important".
In a decision analysis approach, a key goal is to optimally combine value-based trade-offs, which
reflect stakeholder opinions, with technical-based trade-offs, which reflect scientific facts and
local knowledge (Gregory et al. 2012). Although facts and knowledge (e.g., information, models,
expert judgment) are used to estimate consequences, the ultimate decision is informed by trade-
offs in what stakeholders value (Bradley et al. 2016).
Decision analysis provides a number of methods to examine trade-offs, including direct ranking
and swing weighting (Gregory et al. 2012). Swing weighting (Appendix Fl) takes the technical-
based information from a consequence table and assigns relative importance to the different
performance measures as they "swing" from the estimated worst-case to best-case condition (von
Winterfeldt and Edwards 1986; Failing et al. 2007).
Strategy: Consider tradeoffs in FEGS benefits relative to other kinds of objectives
Trade-offs across decision alternatives can be quantified with multi-attribute
methods such as direct ranking or swing-weighting that do not rely on monetary
cost/benefit analysis. Losses or gains in a variety of ecosystem services benefits,
including those hard to monetize, can then be considered alongside other
economic or social objectives.
Some methods for examining trade-offs are highly quantitative, but the goal should not be to
obtain a mathematically optimal solution (Gregory et al. 2012). Instead, methods for exploring
trade-offs are valuable because they (Failing et al. 2007; Bradley et al. 2016):
facilitate discussion;
help identify areas of agreement;
help identify areas where further dialogue or information is needed;
ensure value judgements are well-informed by facts; and
make trade-offs explicit and reduce ambiguity.
For these reasons, multi-attribute approaches are generally more satisfying than strict cost-
benefit methods that compare options in monetary terms (Failing et al. 2007). Multi-attribute
48
-------
approaches allow comparisons across various kinds of indicators, including qualitative and
quantitative (Appendix F2).
Even a relative comparison of the importance of different objectives can bring insights into
achieving what really matters to a community. For example, it is unlikely that well-being means
the same thing to all communities (Appendix F3; Fulford et al. 2015a), and static composite
indices diminish local perspectives on what well-being means to a given community. One
solution is to apply weights to different components of well-being, so that trade-offs across
different well-being objectives (e.g., income vs. cultural opportunities) can be explicitly
examined. Weights can also be incorporated into composite indices to better reflect the relative
importance of the individual metrics that compose them (Smith et al. 2013).
6.2. Implementing a Decision
Ecosystem services practitioners should understand external constraints on the decision. These
could include the project budget, or whether the decision is within a regulatory framework that
has certain requirements. Under the NEPA process, for example, environmental impact
statements have certain requirements that need to be followed in order to be approved. In other
situations, certain approaches like environmental risk assessments or logic models, may be
required as part of an approval process. These factors may affect whether the most appropriate
form of trade-off analysis is a cost-benefit approach, a formal optimization, or a simple ranking
(de Groot et al. 2010). Researchers can provide meaningful information to decision-makers
outside of these approaches, as well. For example, heat maps can be used to describe
relationships between place or ecosystem type and ecosystem services (Troy and Wilson 2006),
and qualitative methods can elucidate relationships between important ecosystem services,
benefits, and values, which alone can be important for decision-making (Chan et al. 2012).
The goal of a decision process is not necessarily consensus among the participants (Gregory et
al. 2012). Inevitably, stakeholders will differ in what they value and their willingness to accept
trade-offs. A consideration of trade-offs, however, can help decision-makers document areas of
agreement or disagreement and the rationale for choices. Some alternatives may be quickly
dismissed. Others may be short-listed as acceptable to stakeholders, with or without a clear
winner emerging. Discussion can center around how alternatives could be modified to make
them more acceptable. Decision-makers may not find an optimal consensus solution, but they
should have a better understanding of how stakeholders feel about the trade-offs that inform the
ultimate choice of action.
6.3. Monitoring the Outcomes and Adaptive Management
Criteria for measuring success should be a key element of community decision-making (Fulford
et al. 2016b). As decisions are implemented, evaluation measures for stakeholder objectives
should be monitored to gauge the success of implementation and the need for corrective action or
reassessment of the context (Appendix F4). Success criteria need to include useful measures of
management success. That is, they need to be case specific and able to measure a response signal
of the implemented action from background noise. In general, one suite of metrics should focus
on analyzing whether the outcomes of the decision activity met the objective(s) while another
49
-------
suite may focus on the effectiveness of the process itself (Dean and Sharfman 1996). For
example, did the approach to stakeholder engagement affect the results of the decision?
Adaptive Management (AM) is a structured decision-support process often considered in
resource management (Williams et al. 2009) and can be a practical approach to SDM in
community based decision-making. Traditional AM elements, often centered around testing
hypotheses, are compatible with the SDM framework (Gregory et al. 2012) and include:
examining alternatives, and the predicted outcomes for each alternative, for
hypotheses to test or learning opportunities;
implementing one or more alternatives to test hypotheses;
monitoring results to learn about the outcome of a given action and to analyze
hypotheses; and
making adjustments to current or future management actions using the results.
The AM principles of hypothesis testing and iterative decision-making are highly relevant to the
SDM emphasis on refining learning and defining objectives. Likewise, the SDM focus on
decision processes and criteria can be applied to advancing AM approaches for a resource
management decision. Practitioners are encouraged to learn about elements of both SDM and
AM for developing strategies to approaching a decision process. Lessons learned from
applications of AM (e.g., Williams et al. 2009; LoSchiavo et al. 2013) should be folded into both
AM and SDM thinking processes for future decision activities.
Benefit
Functions
Ecol. Prod.
Functions
Impact
Functions
MONITORING
~ OUTCOMES
a Human
Well-being
Decision
Alternatives
Social & Economic
Services
A Ecosystem State
(& Intermediate EGS)
"~"-N
HT f Information for Decision Support
Figure 6.1. Important aspects of monitoring outcomes (dashed arrows) and AM principles
of iterative learning and hypothesis testing ("IL & HT"; dashed oval) mapped onto the
FEGS-based conceptual model for structured ecosystem-services decision-making.
50
-------
The FEGS-based approach is represented by a structured ecosystem-services conceptual model
(Fig. 6.1). The arrows in the conceptual model feeding into the Information for Decision Support
box represent areas where monitoring information can be applied to evaluate alternatives and/or
monitor the results of a decision. Within the Information for Decision Support box and the arrow
to the Decision Alternatives box, adaptive management principles focused on iterative learning
and hypothesis testing can inform evaluation of whether the objectives for a given resource
management or community decision are being met.
Strategy: Monitor impacts to FEGS benefits after a decision to inform future
decisions
^ * ฆ
Monitoring information can be used to evaluate whether an implemented decision
is achieving desired objectives or impacting ecosystem services benefits, and help
inform future decisions through iterative learning and hypothesis testing.
51
-------
7. Applications using Case Studies
7.1. Case Study Examples
In this chapter, a number of case study examples are presented to illustrate how the FEGS
concepts, approaches, and tools presented in this report could be integrated into a variety of
community decision processes. In many cases, decision-makers may have their own process or
actions may already be underway by the time ecosystem services are considered. Other processes
may have budget or time constraints that prevent them from completing every step. The
principles of SDM can still be used, however, to identify different entry points where FEGS
concepts can be integrated into a community decision process,
A comparison of five case study communities from across the United States illustrates
similarities and differences in how FEGS concepts could be integrated into decision-making
across a variety of community types and across a variety of focal issues (Fig. 7.1). Details on
each ongoing case study are provided in Appendix G.
Mobile Bay, Pacific San Juan, Southern
Alabama Northwest Puerto Rico Plains
Figure 7.1. Locations of the five case studies.
7.2. Strategies from Case Studies
The five case studies are grounded in an exploration of how the availability of Final Ecosystem
Goods and Services (FEGS) is impacted by community decision-making and how this
relationship alters human well-being. Although, the case studies differ in the spatial scale of the
community and the specific issues under consideration (Table 7.1), the five have core
commonalities:
Working with community stake holders to derive transferable measures of
community well-being and link them to the production of FEGS that directly
benefit the community (Chapter 1, Fig. 1.1);
Applying principles of Structured Decision Making (SDM) approaches to
organize and implement scientific research, and integrate FEGS concepts into
decision-making (Chapter 1, Fig. 1.4); and
Developing decision support based on applying tools to evaluate specific actions
associated with objectives in multiple communities.
52
-------
Table 7.1. Description of major environmental issues, and ecosystem services and well-being
endpoints initially under consideration in each of the five case studies.
Case Study
Environmental Issues
Initial Ecosystem
Services Endpoints
Initial Well-being
Endpoints
Great Lakes
Areas of
Concern
(AOC)
Sediment remediation,
aquatic and riparian habitat
restoration, excess nutrient
loads, climate change;
R2R2R (Remediation to
Restoration to
Revitalization) to improve
Great Lakes communities
Fishing, recreation, water
quality, nutrient retention,
spiritual, aesthetic
Economic (non-market)
valuation, public health
Mobile Bay,
Alabama
Long-term degradation of
coastal habitat and water
quality resulting from
suburban development and
industrialization of coastal
areas
Water quality, water
quantity, recreational
opportunities, fishing, and
viewscapes
Human well-being with a
focus on human health,
living standards, and
community identity
Pacific
Northwest
Degradation of salmon and
shellfish habitat; climate
change and land-use
impacts on supplies of
clean water (nutrients,
pathogens, sediments),
flooding, carbon
sequestration, agricultural
and forest products,
recreation
Restored salmon and
shellfish habitats; climate
change
mitigation/adaptation
strategies for sustaining
clean water, flood
protection, commercial &
recreational opportunities
Human health (water borne
illness); economic and
social benefits accruing
from Community Forest
initiative; environmental
justice for tribes and local
communities
San Juan,
Puerto Rico
Urbanization, aquatic
debris, habitat loss, storm
water runoff, sewage
discharges, flooding;
environmental justice
Contaminant and nitrogen
processing; carbon
sequestration; recreational
and aesthetic value of
habitat and biodiversity;
recreational and artisanal
fishing; flood protection -
homes
Human health (asthma,
vector-borne disease,
waterborne illness);
economic benefits
(tourism); social benefits
(safe housing, community
connection to estuary)
Southern
Plains,
Oklahoma
Water supply, flood
control, agricultural runoff,
excess nutrient and
suspended sediment runoff,
fecal bacteria, dermal
exposure (recreational
uses), dissolved oxygen
Drinking water, flood
protection, recreational
use (fishing, swimming,
boating, hiking, camping),
water source for economic
development
Rustic natural areas
(aesthetics-sense of place),
safe drinking water,
economic benefits (flood
protection, tourism,
increased property values,
attracts business/jobs),
social benefits (community
gathering point, fishing,
back to nature experience)
53
-------
Each case study is using different approaches and tools to address each of the steps of SDM.
Table 7.2 illustrates how the various strategies presented in this report can be applied to integrate
ecosystem services into community decision-making. Details of how each case study is
integrating FEGS concepts into community decision-making are presented in Appendix G. In
some cases, intermediate ecosystem goods and services (IEGS) are included to reflect initial
objectives of stakeholders (e.g., water quality), but will be refined as the case studies work to
link them more explicitly to FEGS and human well-being endpoints.
These case studies are all multi-year studies, affording the ability for each to explore relatively
complex tools, conduct targeted field data collection, and try multiple complementary
approaches. Other communities or decisions may be on a faster time scale or have more limited
resources. However, individual elements or approaches may still be relevant. This kind of
coordinated comparison across communities, grounded in the principles of SDM, will provide a
foundation for:
Evaluating transferability and usability of quantitative tools that link delivery of
FEGS and community decisions across a variety of communities;
Examining similarities and differences across communities in available FEGS,
community well-being, and the sustainability of community decisions; and
Assessing the benefits of a FEGS approach to community decision-making.
54
-------
Table 7.2. Summary of strategies being implemented in each case study to integrate ecosystem services concepts within the steps
of a community decision-making process (for details, see Appendix G). Key tools used by each case are in bold.
Case
Study
Clarify Decision
Context
Define Objectives
Develop Alternatives
Estimate Consequences
Evaluate
Trade-offs
Implement,
Monitor, and
Review
Great
Lakes
AOC
Expand existing AOC
process to include
broad consideration of
ecosystem services,
including engaging
larger and different
group of stakeholders
Preserve the current,
previously existing
programmatic targets
agreed to through the
AOC governance
structure, but expand
an explicit
consideration of FEGS
Provide a forum for
stakeholders to discuss
direct and indirect
connections between
remediation/restoration
activities and
ecosystem services
Conduct participatory
mapping and co-
development of spatially-
explicit EPFs (SPA) and
eco-hydrological models
(VELMA) to demonstrate
how removal of Beneficial
Use Impairments can
improve ecosystem services
and human well-being
(HWBI)
Provide
analysis
results to
stakeholders
who provide
comments on
various
trade-offs to
decision-
makers
Moving forward,
better understand
how spatial
provisioning of
ecosystem services
can affect trade-offs
Mobile,
Alabama
Evaluate impacts and
outcomes of National
Estuary Program
(NEP) restoration
activities within the
landscape
Work with
stakeholders to define
broad FEGS-based
objectives and
measures
Decision alternatives,
including restoration
and mitigation
activities, are derived
from the NEP
management plan
Use eco-hydrological
models (VELMA) and
ecosystem services
mapping tools (EPA H20)
to assess effects of
restoration activities on
FEGS production
Provide
information
on trade-offs
across
ecosystem
services to
decision-
makers
Use the results to
inform future
restoration planning
Pacific
Northwest
Identify ecosystem-
based management
solutions that consider
the linkage of
terrestrial and aquatic
systems
Work with
stakeholders to
identify FEGS deemed
essential to community
well-being, and how to
measure them
Identify best
management practices
for restoring
ecosystem services
Use models (VELMA,
CGEM) to simulate how
specified changes in land-
use practices impact
ecosystem goods and
services, and human well-
being (HWBI)
Use
visualization
tools to
communicate
complex
model
outputs to
decision-
makers
Explore development
of simpler, more
accessible models
55
-------
Case
Study
Clarify Decision
Context
Define Objectives
Develop Alternatives
Estimate Consequences
Evaluate
Trade-offs
Implement,
Monitor, and
Review
San Juan,
Puerto
Rico
Use FEGS conceptual
models to understand
linkages between
stressors, ecosystem
services, and
beneficiaries in the
NEP watershed
Review existing
planning documents to
infer objectives,
exploring the FEGS-
CS, RBI, and HWBI
to identify potential
metrics to measure
them
Decision alternatives,
including restoration
and mitigation
activities, are derived
from the NEP
management plan
Conduct field work to fill in
key information gaps
between anthropogenic
stressors, ecosystem
services, and human well-
being impacts, and to
inform development of
EPFs and EBFs (Envision,
HWBI)
Evaluate
benefits of
NEP
management
activities to
ecosystem
services and
well-being
Improve
understanding of
benefits and trade-
offs involved in
watershed
management of
estuaries
Southern
Plains,
Oklahoma
Use DASEES
(Decision Analysis for
a Sustainable
Environment,
Economy, and
Society) to engage
stakeholders
Use DASEES to
develop objectives
hierarchies and
performance measures.
Use DASEES to
develop means,
decision options, and
scenarios to achieve
objectives
Use DASEES, possibly
using influence diagrams
and BBNs to link scenarios
to impacts on ecosystem
services
Use
DASEES to
develop
measures
preferences
and weights
on
alternative
scenarios
Use DASEES to
identify priority
actions and set
triggers for
monitoring success
56
-------
8. Synthesis
8.1. Final Ecosystem Goods and Services Facilitate Values-focused Thinking
Community decision-making rarely involves assessments of isolated issues. Therefore,
ecosystem services should be considered in the context of environmental management or
community decisions and the broad suite of stakeholder objectives. Structured Decision Making
provides a framework to integrate FEGS concepts into values-focused thinking (Keeney 1992;
Gregory et al. 2012; Bradley et al. 2016). Numerous tools and approaches exist for integrating
ecosystem services into decision-making (Table 8.1), many examples of which are provided in
the appendices to this report. Many decisions may use some, many, or none of these tools, and
the tools may be used in a variety of ways. But ultimately, the concept of FEGS can facilitate
decision-making based on what stakeholders value, regardless of what tools are used.
Table 8.1. Example tools and approaches from this report and how they could be used to
integrate FEGS into a decision process.
Example Tools and
Approaches
Description
Clarify Decision
Context
Define Objectives
and Measures
Develop
Alternatives
Estimate
Conseauences
Evaluate
Trade-offs
DASEES
Decision support system provides a
workspace, guidance, and suite of tools
Conceptual models,
influence diagrams, logic
models
Diagram cause effect relationships between
decisions, FEGS, and benefits
FEGS conceptual model
framework, DPSIR
Provide a structured starting point for
diagramming cause/effect relationships
between decisions, FEGS, and benefits
Eco-Health Relationship
Browser
Describes connections between ecosystem
services and health benefits
Objectives hierarchies
Elicit, structure, and define objectives,
including those related to FEGS
FEGS-CS
Provides a structured starting point for
identifying FEGS, the ecosystems that
provide them, and who is benefitting
Means-ends networks
Diagram relationships between objectives
and means to achieve them, including FEGS
Rapid Benefits
Indicators (RBI)
Provides guidance for measuring the value of
ecosystem services
Human Well-Being
Index (HWBI)
Provides a framework and collection of
metrics for quantifying FEGS impacts on
well-being
57
-------
Example Tools and
Approaches
Description
Clarify Decision
Context
Define Objectives
and Measures
Develop
Alternatives
Estimate
Conseauences
Evaluate
Trade-offs
EnviroAtlas
Provides maps and metrics of FEGS benefits
ESML
Model library for identifying and assessing
ecosystem services models
Bayesian Belief
Networks
Diagram and model probabilistic
relationships between decisions and FEGS
benefits
Triple Value Simulation
(3VS)
Diagram and model relationships between
decisions and FEGS benefits, incorporating
feedbacks among variables
EPA H20, InVEST,
SPA, i-Tree, ARIES
Produce maps of ecosystem services and
their benefits, including scenario
comparisons
VELMA, Envision
Produce spatial maps relating decisions to
FEGS benefits, incorporating dynamic
feedbacks among variables over time
Consequence tables,
management option
matrices
Visualize and summarize the effects of the
alternatives on performance measures,
including FEGS benefits
Cost-benefit analysis,
Total Economic Value,
Willingness to Pay
Can be used to calculate the costs and
benefits of a decision, including FEGS
Direct ranking, swing-
weighting
Multi-attribute methods to assess trade-offs
across alternatives
Making decisions based on what is important to stakeholders is the basis of values-focused
decision-making and is fundamentally distinct from the more common alternative-focused
decision-making. Alternative-focused decision-making also includes values, but often implicitly.
Stating values explicitly promotes a more inclusive, transparent, and defensible process, which
creates an environment for fostering options with better prospects for desired outcomes and
minimal negative impacts (Gregory et al. 2012).
Advantages of integrating FEGS concepts into values-focused thinking include (modified from
Keeney 1992; Bradley et al. 2016) expanded stakeholder engagement, improved information
collection and communication, creative development and evaluation of alternatives;
interconnected decisions, and strategic thinking.
58
-------
8.2. Stakeholder Engagement
Community members often have a strong sense of connection to their community and want to
improve it. As stakeholders, they often have a desire to be informed, educated, and engaged in
local decisions. Decisions made outside community involvement can be perceived as having
little transparency or not reflecting local values (Bradley et al. 2016). Values-focused thinking
encourages all stakeholders, including community members, to communicate what is important
to them (Keeney 1992). Moreover, early consideration of FEGS can bring to light beneficiaries
that might otherwise have been overlooked as stakeholders in the decision process (Table 8.2).
Table 8.2. Example strategies from this report that can be used to engage stakeholders.
Strategy
Chapter
Example Tools
Appendix
Use of Decision Support Systems to engage
participants in a step-by-step process
5.6
DASEES
RBI Guidance
Al
C4
Use of decision support tools to identify isolated
or heavily engaged stakeholder groups
2.3
Social networks
B1
Use of conceptual models to engage
stakeholders and elicit feedback
2.5
Conceptual models
DPSIR
HWBI
B1,B4
B1
C5
Use of ecosystem services classification systems
to identify beneficiary groups that may need to
be engaged as participants
2.2, 2.4
FEGS-CS
B2
8.3. Guiding Information Collection
Taking the time to understand stakeholder values can help prioritize limited resources and focus
information gathering on what is important (Keeney 1992). Environmental monitoring often
targets indicators of environmental condition that may or may not be relevant to stakeholders. A
consideration of FEGS can help focus in on attributes of environmental condition or measures of
benefit that are most directly relevant to stakeholders (Table 8.3). EPFs or EBFs may need to be
applied or developed to adequately estimate effects of alternatives on stakeholder objectives. The
complexity of models needed depends on the level of uncertainty that decision-makers are
comfortable with. In some cases, group deliberations using qualitative conceptual models may be
adequate to identify win-win or low-risk alternatives, and spending budget on more detailed
information gathering may not be necessary. In other cases, a greater level of precision is needed
or desired to accurately assess tradeoffs.
A key point in a structured decision making process is that both fact-based information and
values-based information are taken into account (Bradley et al. 2016). Information about
stakeholder values helps to prioritize collection of scientific information based on what is most
relevant to decisions.
59
-------
Table 8.3. Example strategies from this report that can be used to guide information
collection.
Strategy
Chapter
Example Tools
Appendix
Development of conceptual models to describe
cause and effect
2.5,5.2
Conceptual models
B1,B4
Use of structured generic conceptual models
that pre-specify causal linkages between
decisions, stressors, ecosystem services, and
benefits
2.5,4.1
DPSIR
Ecosystem Services Cascade
HWBI
EcoHealth Browser
B1
El
C5
E6
Application of ecological production functions
(EPFs) and ecological benefits functions
(EBFs) to translate information on ecological
condition to stakeholder relevant measures
5.3,5.4
EPA H20
SPA
VELMA
Economic valuation
Eco-Health Browser
Services to HWBI
El
E2
E3
E5
E6
E8
Use of model libraries and information on
model transferability to identify best available
EPFs and EBFs
5.3,5.5
ESML Library
Model transferability
E4
E9
8.4. Improving Communication
In some cases, ineffective communication is itself a major source of uncertainty where more
information is needed (Gregory et al. 2012), and research or engagement efforts should be
focused on clarifying what stakeholders value. This could include reducing ambiguity between
what stakeholders care about versus how to achieve it, or rigorous definition of objectives with
performance measures so that what is meant by objectives is clear to participants. Tools such as
objectives hierarchies and means-ends networks can help clarify what is meant by objectives
(Table 8.4). Structured conceptual models or hierarchies can provide a starting point for guiding
discussions or providing examples.
Table 8.4. Example strategies from this report that can be used to improve communication.
Strategy
Chapter
Example Tools
Appendix
Use of structured generic conceptual models to
guide discussions and elicit information
2.5
DPSIR
HWBI
B1
C5
Use of objectives hierarchies to define what
matters, and distinguish means from ends
3.1,3.2
Objectives hierarchies
CI
Use of structured hierarchies to provide strawman
objectives and example performance measures
3.3
FEGS-CS
HWBI
C2
C5
Use of ecosystem services classification systems
to reduce ambiguity in performance measures by
linking them explicitly to beneficiaries
3.3
FEGS-CS
Rapid Benefits Indicators
C2
C4
60
-------
8.5. Alternative Development and Evaluation
Values-focused thinking supports the development of new and creative alternatives that are
directly responsive to what stakeholders value, and have a better chance of acceptance and
successful outcome (Keeney 1992). A consideration of FEGS may bring to light social and
economic benefits to community well-being provided by ecosystems. In some cases, measures of
FEGS may be useful surrogates for what stakeholders value (Yee et al. 2014). In other cases,
ecosystem services may provide a means to achieving other objectives, and may provide novel
solutions to finding areas of agreement among stakeholders with a variety of different objectives
(Table 8.5).
Values-focused thinking, particularly with the development of well-defined performance
measures, provides a path forward for analyzing the desirability of alternatives and evaluating
trade-offs (Keeney 1992; Bradley et al. 2016). Ecosystem services frameworks inherently
provide a conceptual link between decisions, the environment, and stakeholder-relevant benefits.
Conceptual models can be applied to quantitatively evaluate alternatives through the use of EPFs
and EBFs. Objectives related to ecosystem services should be considered in the broader context
of other stakeholder objectives (Table 8.5). Situations where multiple stakeholder groups have
competing objectives may find areas of agreement when ecosystem benefits are more broadly
considered.
Table 8.5. Example strategies from this report that can be used to create and evaluate
alternatives.
Strategy
Chapter
Example Tools
Appendix
Use of decision analysis tools to link
ecosystem services as means to achieving other
objectives
2.5,3.2
Means-ends networks
Influence diagrams
D1
Bl, E10
Consideration of ecosystem services as means
to achieve health related or well-being
objectives
3.3,5.4
Eco-Health relationships
Services to HWBI
E6
E8
Use of decision analysis tools to compare
alternatives and evaluate trade-offs
5.7, 6.1
Consequence tables
Swing-weighting
Multi-attribute comparisons
E10
F1
F2
8.6. Interconnecting Decisions and Guiding Strategic Thinking
Decisions in one context may affect how a decision will be made in another context (Keeney
1992). Values-focused thinking can interconnect decisions and show how they impact a broader
suite of stakeholder objectives (Bradley et al. 2016). Ecosystem function and ecosystem services
are often overlooked or taken for granted in social and economic decision-making, yet there is
growing recognition that human well-being is linked to the sustainable use of environmental
resources (MEA 2005; NRC 2005). A consideration of FEGS can inform longer term strategic
objectives, and help support interconnecting decisions for the sustainable delivery of social and
economic benefits derived from ecosystems (NRC 2011; Anastas 2012; Yee et al. 2014; Table
8.6).
61
-------
Table 8.6. Example strategies from this report that can be used to interconnect decisions
and guide strategic thinking.
Strategy
Chapter
Example Tools
Appendix
Consideration of well-being objectives as
strategic long-term goals
3.3
HWBI
C5, D2, F3
Using information from past decisions to
guide future decisions
5.5,6.3
Model transferability
Setting triggers
E9
F4
8.7. Conclusions
Community decisions, and the agencies and organizations that make them, are often focused
around a relatively narrow context or singular issues, yet the consequences can range across
economic, environmental, and social aspects. Structured decision analysis provides an approach
for evaluating trade-offs in a way that encourages public participation and collaborative decision-
making, and allows for consideration of multiple attributes (Liu et al. 2010). Integrating the
concept of Final Ecosystem Goods and Services into decision-making can help provide a direct
link from environmental conditions to social and economic benefits, ensuring that key
stakeholders, key objectives, and creative alternatives are not overlooked. Ultimately this will lead
to more inclusive decision-making that promotes more sustainable approaches to balancing the
economic, environmental, and social trade-offs in decisions that communities face every day.
62
-------
References
Abdelnour A., R.B. McKane, M. Stieglitz, F. Pan, 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, R.B. McKane. 2011. Catchment hydrological responses to forest
harvest amount and spatial pattern. Water Resources Research 47:18.
Anastas, P.T. 2012. Fundamental changes to EPA's research enterprise: The Path Forward. Environmental
Science and Technology 46:580-586.
Angradi, T.R., J.J. Launspach, D.W. Bolgrien, B.J. Bellinger, M.A. Starry, J.C. Hoffman, A.S. Trebitz,
M.E. Sierszen, T.P. Hollenhorst. 2016. Mapping ecosystem service indicators in a Great Lakes
estuarine Area of Concern. Journal of Great Lakes Research 42:717-727.
Angradi, T.R., M.S. Pearson, D.W. Bolgrien, B.J. Bellinger, M.A. Starry, C. Reschke. 2013. Predicting
submerged aquatic vegetation cover and occurrence in a lake superior estuary. Journal of Great Lakes
Research 39(4):536-546.
Bagstad, K.J., F. Villa, D. Batker, J. Harrison-Cox, B. Voigt, G.W. Johnson. 2014. From theoretical to
actual ecosystem services: Mapping beneficiaries and spatial flows in ecosystem service assessments.
Ecol Soc. 19(2):64.
Bagstad, K.J., F. Villa, G.W. Johnson, B. Voigt. 2011. Aries artificial intelligence for ecosystem services:
A guide to models and data, version 1.0. The ARIES Consortium.
Beaumont, N.J., M.C. Austen, S.C. Mangi, M. Townsend. 2008. Economic valuation for the conservation
of marine biodiversity. Marine Pollution Bulletin 56:386-396.
Bell, J.F., J.S. Wilson, G.C. Liu. 2008. Neighborhood greenness and 2-year changes in Body Mass Index
of children and youth. American Journal of Preventive Medicine 35(6):547-553
Birol, E., K. Karousakis, P. Koundouri. 2006. Using economic valuation techniques to inform water
resources management: A survey and critical appraisal of available techniques and an application.
Science of the Total Environment 365:105-122.
Bolte, J., K. Vache. 2010. Envisioning puget sound alternative futures: PSNERP Final Report. Puget
Sound Nearshore Ecosystem Restoration Project.
Bolte J. 2014. Envision (Version 7): A Guide to Application Development. Biological and Ecological
Engineering Department, Oregon State University, Corvallis, OR.
Bousquin, J., W.S. Fisher, J. Carriger, E. Huertas. 2014. A Bayesian Belief Network Approach to Explore
Alternative Decisions for Sediment Control and Water Storage Capacity at Lago Lucchetti, Puerto
Rico. US Environmental Protection Agency Office of Research and Development. EPA/600/R-14/296.
Bousquin J, K. Hychka, M. Mazzotta. 2015. Benefit indicators for flood regulation services of wetlands: a
modeling approach. U.S. Environmental Protection Agency, Office of Research and Development,
Atlantic Ecology Division, Narragansett, RI, EPA/600/R-15/191.
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, A.J. Krupnick, R.J. Johnston, M. Weber, K. Hall. 2015. Ecosystem Services
Indicators: Improving the Linkage between Biophysical and Economic Analyses. RFF DP 15-40,
Resources for the Future, Washington, DC.
Bradley, P., W. Fisher, B. Dyson, S. Yee, J. Carriger, G. Gambirazzio, J. Bousquin, 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. and S. Yee. 2015. Using the DPSIR Framework to Develop a Conceptual Model: Technical
Support Document. US Environmental Protection Agency, Office of Research and Development,
Atlantic Ecology Division, Narragansett, RI. EPA/600/R-15/154
63
-------
Breslow, S.J., B. Sojka, R. Barnea, X. Basurto, C. Carothers, S. Charnley, S. Coulthard, N. Dolsak, J.
Donatuto, C. Garcia-Quijano, C.C. Hicks, A. Levine, M.B. Mascia, K. Norman, M. Poe, T. Satterfield,
K. St Martin, P.S. Levin. 2016. Conceptualizing and operationalizing human well-being for ecosystem
assessment and management. Environmental Science & Policy 66:250-259.
Brouwer, R. 2006. Do stated preference methods stand the test of time? A test of the stability of contingent
values and models for health risks when facing an extreme event. Ecological Economics 60: 399-406.
Brown, C.A. 1984. The Central Arizona water control study: A case for multiobjective planning and public
involvement. Water Resources Bulletin 20(3):331-337.
Bruins R.J.F., T.J. Canfield, C. Duke, L. Kapustka, A.M. Nahlik, R.B. Schafer. 2017. Using ecological
production functions to link ecological processes to ecosystem services. Integrated Environmental
Assessment and Management 13(1):52-61.
Bryce, R., K.N. Irvine, A. Church, R. Fish, S. Ranger, J.O. Kenter. 2016. Subjective well-being indicators
for large-scale assessment of cultural ecosystem services. Ecosystem Services 21:258-269.
Bunnefeld, N., E. Hoshino, E.J. Milner-Gulland. 2011. Management strategy evaluation: a powerful tool
for conservation? Trends in Ecology & Evolution 26(9):441-447.
Carriger, J. and W.H. Benson. 2012. Restoring and Managing Gulf of Mexico Fisheries: A Path toward
Creative Decision-Making. Chapter 13 in: Estuaries, Classification, Ecology, and Human Impacts.
Nova Science Publishers, Inc., Hauppage, NY, 291-334.
Carriger, J.F., W.S. Fisher, T.B. Stockton, P.E. Sturm. 2013. Advancing the Guanica Bay (Puerto Rico)
Watershed Management Plan. Coastal Management 41(1): 19-38.
Casado-Arzuaga, I., M. Onaindia, I. Madariaga, P.H. Verburg. 2013. Mapping recreation and aesthetic
value of ecosystems in Bilbao Metropolitan Greenbelt (Northern Spain) to support landscape planning.
Landscape Ecology 29(8): 1393-1405.
Castro, A.J., C.C. Vaughn, M. Garcia-Llorente, J.P. Julian, C.L. Atkinson. 2016. Willingness to pay for
ecosystem services among stakeholder groups in a South-Central US watershed with regional conflict.
Journal of Water Resources Planning and Management 142:05016006
CDC (Centers for Disease Control and Prevention). 2016. A Sustainability Planning Guide for Healthy
Communities. CDC, Division of Community Health.
CEQ (Council on Environmental Quality) 2007. A Citizen's Guide to the NEPA: Having Your Voice
Heard. Executive Office of the President. Washington, DC
CEQ. 2015. Incorporating ecosystem services into federal decision-making. Memorandum for Executive
Departments and Agencies: M-16-01.
https://obamawhitehouse.archives.gov/sites/default/files/omb/memoranda/2016/m-16-01.pdf
Chan, K.M.A., M.R. Shaw, D.R. Cameron, E.C. Underwood, G.C. Daily. 2006. Conservation planning for
ecosystem services. PLoS Biol 4(11): e379. https://doi.org/10.1371/journal.pbio.0040379
Chan, K.M.A., A.D. Guerry, P. Balvanera, S. Klain, T. Satterfield, X. Basurto, A. Bostrom, R.
Chuenpagdee, R. Gould, B.S. Halpern, N. Hannahs, J. Levine, B. Norton, M. Ruckelshaus, R. Russell,
J. Tam, U. Woodside. 2012. Where are cultural and social in ecosystem services? A framework for
constructive engagement. Bioscience 62:744-756.
Christie, M. and J. Gibbons. 2011. The effect of individual 'ability to choose' (scale heterogeneity) on the
valuation of environmental goods. Ecological Economics 70(12):2250-2257.
Cook, G.S., P.J. Fletcher, C.R. Kelble. 2014. Towards marine ecosystem based management in South
Florida: investigating the connections among ecosystem pressures, states, and services in a complex
coastal system. Ecological Indicators 44:26-39.
Cooter, E.J., A. Rea, R. Bruins, D. Schwede, R.L. Dennis. 2013. The role of the atmosphere in the
provision of ecosystem services. Science of the Total Environment 448:197-208.
Costanza, R., R. d'Arge, R. de Groot, S. Farber, M. Grasso, B. Hannon, K. Limburg, S. Naeem, R.V.
O'Neill, J. Paruelo, et al. 1997. The value of the world's ecosystem services and natural capital. Nature.
64
-------
387(6630):253-260.
Costanza, R., R. de Groot, P. Sutton, S. van der Ploeg, S.J. Anderson, I. Kubiszewski, S. Farber, R.K.
Turner. 2014. Changes in the global value of ecosystem services. Global Environmental Change
26:152-158.
Cowling, R.M., B. Egoh, A.T. Knight, P.J. O'Farrell, B. Reyers, M. Rouget, D.J. Roux, A. Welz, A.
Wilhelm-Rechman. 2008. An operational model for mainstreaming ecosystem services for
implementation. Proceedings of the National Academy of Sciences 105:9483-9488.
Daily, G. 1997. Nature's Services. Island Press, Washington, D.C.
de Jesus-Crespo, R. and R. Fulford. 2017. Eco-Health linkages: Assessing the role of ecosystem goods and
services on human health using causal criteria analysis. International Journal of Public Health, doi
10.1007/s00038-017-1020-3.
de Groot, R.S., R. Alkemade, L. Braat, L. Hein, L. Willemen. 2010. Challenges in integrating the concept
of ecosystem services and values in landscape planning, management and decision-making. Ecological
Complexity 7:260-272.
Dean, J.W. Jr., M.P. Sharfman. 1996. Does decision process matter? A study of strategic decision-making
effectiveness. Academy of Management Journal. 39(2):368-396.
Devkota, J., X. Fang. 2015. Quantification of water and salt exchanges in a tidal estuary. Water. 7(5): 1769.
Dilling, L., and M.C. Lemos. 2011. Creating usable science: opportunities and constraints for climate
knowledge use and their implications for science policy. Global Environmental Change. 21(2): 680-
689.
Dorn, M., K. Aydin, D. Jones, W. Palsson, K. Spalinger. 2014. Assessment of the Walleye Pollock Stock
in the Gulf of Alaska. In Stock Assessment and Fishery Evaluation Report for the Groundfish
Resources of the Bering Sea/Aleutian Islands Regions, p. 53-170. North Pacific Fishery Management
Council, 605 W. 4th Avenue Suite 306, Anchorage, AK 99501.
http://www.afsc.noaa.gov/refm/docs/2014/GOApollock.pdf
Druschke, C.G. and K.C. Hychka. 2015. Manager perspectives on communication and public engagement
in ecological restoration project success. Ecology and Society 20:58.
Dyrbye, L.N., D. Satele, T. Shanafelt. 2016. Ability of a 9-item well-being index to identify distress and
stratify quality of life in US workers. Journal of Occupational and Environmental Medicine 58: 810-
817.
Edwards, W., R.F. Miles, Jr. , D. von Winterfeldt. 2007. Advances in Decision Analysis: From
Foundations to Applications. Cambridge University Press, New York.
EEA (European Environment Agency). 2005 Sustainable Use and Management of Natural Resources.
EEA Report No 9/2005, Copenhagen: European Environment Agency, 72 pp.
EPA (U.S. Environmental Protection Agency). 1998. Guidelines for Ecological Risk Assessment.
Washington (DC): USEPA, Risk Assessment Forum. EPA/630/R-95/002F.
EPA. 2012. Decision Support Framework Implementation of the Web-based Environmental Decision
Analysis Application DASEES: Decision Analysis for a Sustainable Environment, Economy, and
Society. US Environmental Protection Agency. EPA/600/R-12/008.
EPA. 2015. National Ecosystem Services Classification System (NESCS): Framework Design and Policy
Application. Washington, DC: U.S. Environmental Protection Agency, Office of Water and Office of
Research and Development. EPA-800-R-15-002.
Failing, L., R. Gregory, M. Harstone. 2007. Integrating science and local knowledge in environmental risk
management: A decision-focused approach. Ecological Economics 64:47-60.
Ferrara, A.R., and R. Nistico. 2015. Regional well-being indicators and dispersion from a
multidimensional perspective: evidence from Italy. Annals of Regional Science 55: 373-420.
Fiksel, J., R. Bruins, A. Gatchett, A. Gilliland, M. ten Brink. 2014. The triple value model: A systems
approach to sustainable solutions. Clean Technologies and Environmental Policy 16:691-702.
65
-------
Francesconi, W., R. Srinivasan, E. Perez-Minana, S.P. Willcock, M. Quintero. 2016. Using the Soil and
Water Assessment Tool (SWAT) to model ecosystem services: A systematic review. Journal of
Hydrology. 535:625-636.
Fulford, R.S., L.M. Smith, M. Harwell, D. Dantin, M. Russell, J. Harvey. 2015a. Human well-being differs
by community type: Toward reference points in a human well-being indicator useful for decision
support. Ecological Indicators 56:194-204.
Fulford, R.S., D. Yoskowitz, M. Russell, D.D. Dantin, J. Rogers. 2015b. Habitat and recreational fishing
opportunity in Tampa Bay: Linking ecological and ecosystem services to human beneficiaries.
Ecosystem Services 17: 64-74.
Fulford, R., R. Bruins, T. Canfield, J. Handy, J.M. Johnston, P. Ringold, M. Russell, N. Seeteram, K.
Winters, S. Yee. 2016a. Lessons Learned in Applying Ecosystem Goods and Services to Community
Decision-Making. U.S. Environmental Protection Agency, Washington, DC. EPA/600/R-16/136.
Fulford, R., M. Russell, J. Harvey, M. Harwell. 2016b. Sustainability at the community level: Searching
for common ground as a part of a national strategy for decision support. U.S. Environmental Protection
Agency, Washington, DC. EPA/600/R-16/178.
Gariepy, G., A. Blair, Y. Kestens, N. Schmitz. 2014. Neighborhood characteristics and 10-year risk of
depression in Canadian adults with and without a chronic illness. Health and Place, 30:279-86
Goodwin, P. and G. Wright. 2004. Decision Analysis for Management Judgement. 3rd Edition. Wiley,
Chichester.
Gordon, L., and C. Folke. 2000. Ecohydrological landscape management for human well-being. Water
International 25: 178-184.
Granek, E. F., S. Polasky, C.V. Kappel, D.J. Reed, D.M. Stoms, E.W. Koch, C.J. Kennedy, L.A. Cramer,
S.D. Hacker, E.B. Barbier, S. Aswani, M. Ruckelshaus, G.M.E. Perillo, B.R. Silliman, N. Muthiga, D.
Bael, E. Wolanski. 2010. Ecosystem services as a common language for coastal ecosystem-based
management. Conservation Biology, 24: 207-216.
Gray, S., A. Chan, D. Clark, R. Jordan. 2012. Modeling the integration of stakeholder knowledge in
social-ecological decision-making: benefits and limitations to knowledge diversity. Ecological
Modelling 229:88-96.
Gray, S.A., S. Gray, J.L. De Kok, A.E.R. Helfgott, B. O'Dwyer, R. Jordan, A. Nyaki. 2015. Using fuzzy
cognitive mapping as a participatory approach to analyze change, preferred states, and perceived
resilience of social-ecological systems. Ecology and Society 20(2): 11
Gregory, R. and R.L. Keeney. 2002. Making smarter environmental decisions. Journal of the American
Water Resources Association 38(6): 1601-1612
Gregory, R., L. Failing, M. Harstone, G. Long, T. McDaniels, D. Ohlson. 2012. Structured Decision-
making: A Practical Guide to Environmental Management Choices. West Sussex, UK: Wiley-
Blackwell.
Hammond, J.S., R.L. Keeney, H. Raiffa. 1999. Smart Choices: A Practical Guide to Making Better
Decisions. Harvard Business School Press, Boston, MA.
Harlan, S.L., J.H. Declet-Barreto, W.L. Stefanov, D.B. Petitti. 2013. Neighborhood effects on heat deaths:
social and environmental predictors of vulnerability in Maricopa County, Arizona. Environmental
Health Perspectives (Online) 121(2): 197.
Harper, G. and R. Price. 2011. Defra Evidence and Analysis Series Paper 3. A framework for
understanding the social impacts of policy and their effects on well-being. A paper for the Social
Impacts Taskforce.
Harvey, C.J., J.C. Reum, M.R. Poe, G.D. Williams, S.J. Kim. 2016. Using conceptual models and
qualitative network models to advance integrative assessments of marine ecosystems. Coastal
Management, 44(5):486-503.
Hendrix, M. and K.A. Innes-Wimsatt. 2013. Increased risk of depression for people living in coal mining
66
-------
areas of Central Appalachia. Ecopsychology 5:179-187
Huber, P.R., S.E. Greco, N.H. Schumaker, J. Hobbs. 2014. A priori assessment of reintroduction strategies
for a native ungulate: Using HexSim to guide release site selection. Landscape Ecology 29(4):689-701.
Irvine, K.N., L. O'Brien, N. Ravenscroft, N. Cooper, M. Everard, I. Fazey, M.S. Reed, J.O. Kenter. 2016.
Ecosystem services and the idea of shared values. Ecosystem Services 21: 184-193.
Jackson L.E., J. Daniel, B. McCorkle, A. Sears, K.F. Bush. 2013. Linking ecosystem services and human
health: The Eco-Health Relationship Browser. International Journal of Public Health 58:747-755.
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^-79
Johnston, R.J., and R.S. Rosenberger. 2010. Methods, trends and controversies in contemporary benefit
transfer. Journal of Economic Surveys 24:479-510.
Johnston, R.J., and L.A. Wainger. 2015. Benefit Transfer for Ecosystem Service Valuation: An
Introduction to Theory and Methods. In Benefit Transfer of Environmental and Resource Values: A
Guide for Researchers and Practitioners, ed. R.J. Johnston, J. Rolfe, R.S. Rosenberger and R. Brouwer,
237-273.
Karrasch L., T. Klenke, J. Woltjer. 2014. Linking ecosystem services approach to social preferences and
needs in integrated coastal land use management - A planning approach. Land Use Policy 38:522-532.
Keeney, R.L. 1982. Decision analysis: An overview. Operations Research 30(5): 803-838.
Keeney, R. 1992 Value-Focused Thinking: A Path to Creative Decision Making. Cambridge, MA: Harvard
University Press.
Keeney RL and Gregory RS. 2005. Selecting attributes to measure the achievement of objectives.
Operations Research 53: 1-11.
Kelble, C.R., D.K. Loomis, S. Lovelace, W.K. Nuttle, P.B. Ortner, P. Fletcher P, et al. 2013 The EBM-
DPSER Conceptual Model: Integrating Ecosystem Services into the DPSIR Framework. PLoS ONE
8(8): e70766.
Klain, S.C. and K.M.A. Chan. 2012. Navigating coastal values: participatory mapping of ecosystem
services for spatial planning. Ecological Economics 82:104-113.
Knol A.B., D.J. Briggs, E. Lebret. 2010 Assessment of complex environmental health problems: Framing
the structures and structuring the frameworks. Science of the Total Environment 408:2785-2794
Koschke L., C. Lorz, C. Furst, T. Lehmann, F. Makeschin. 2014. Assessing hydrological and provisioning
ecosystem services in a case study in Western Central Brazil. Ecological Processes 3:2.
Kuo, F.E., and W.C. Sullivan. 2001. Environment and crime in the inner city, does vegetation reduce
crime? Environment and Behaviour 33:343-367
Landers D. and A. Nahlik. 2013. Final Ecosystem Goods and Services Classification System (FEGS-CS).
Washington, DC: U.S. Environmental Protection Agency. EPA/600/R-13/ORD-004914
Lee J-Y and L. Duk-Chul. 2014. Cardiac and pulmonary benefits of forest walking versus city walking in
elderly women: A randomised, controlled, open-label trial. European Journal of Integrative Medicine
6(1): 5-11.
Liu, S., R. Costanza, S. Farber, A. Troy. 2010. Valuing ecosystem services: theory, practice, and the need
for a transdisciplinary synthesis. Annals of the New York Academy of Sciences 1185:54-78.
LoSchiavo, A., R. Best, R. Burns, S. Gray, M. Harwell, E. Hines, A. McLean, T. St. Clair, S. Traxler, J.
Vearil. 2013. Lessons learned from the first decade of adaptive management in comprehensive
Everglades restoration. Ecology and Society. 18(4):70-85.
Louv, R. 2005. Last Child in the Woods: Saving our Children from Nature-deficit Disorder. Algonquin
Books of Chapel Hill, Chapel Hill, North Carolina, USA
Maes, J., B. Egoh, L. Willemen, C. Liquete, P. Vihervaara, J.P. Schagner, B. Grizzetti, E.G. Drakou, A.L.
Notte, G. Zulian, et al. 2012. Mapping ecosystem services for policy support and decision-making in
67
-------
the european union. Ecosystem Services. l(l):31-39.
Mangi, S.C., C.E. Davis, L.A. Payne, M.C. Austen, D. Simmonds, N.J. Beaumont, T. Smyth. 2011.
Valuing the regulatory services provided by marine ecosystems. Environmetrics 22: 686-698.
Marcot, B.G., M.P. Thompson, M.C. Runge, F.R. Thompson, S. McNulty, D. Cleaves, M. Tomosy, L.A.
Fisher, A. Bliss. 2012. Recent advance in applying decision science to managing national forests. Forest
Ecology and Management 285: 123-132
Maseyk, F.J.F., A.D. Mackay, H.P. Possingham, E.J. Dominati, Y.M. Buckley. 2017. Managing natural
capital stocks for the provision of ecosystem services. Conservation Letters 10(2):211-220.
Mazzotta, M., J. Bousquin, C. Ojo, K. Hychka, C.G. Druschke, W. Berry, R. McKinney. 2016. Assessing
the Benefits of Wetland Restoration: A Rapid Benefit Indicators Approach for Decision Makers. U.S.
Environmental Protection Agency, Narragansett, RI, EPA/600/R-16/084.
MEA (Millennium Ecosystem Assessment). 2005. Ecosystems and Human Well-being: Current State and
Trends, Volume 1. Island Press, Washington, DC.
Merrick, J., G. Parnell, J. Barnett, M. Garcia. 2005. A multiple-objective decision analysis of stakeholder
values to identify watershed improvement needs. Decision Analysis 2: 44-57.
Miller, T.J., S.J.D. Martell, D.B. Bunnell, G. Davis, L. Fegley, A. Sharov, C. Bonzek, D. Hewitt, J.
Hoenig, R. Lipcius. 2005. Stock assessment of Blue Crab in Chesapeake Bay Final Report. Technical
report series No. TS-487-05 University of Maryland Center for Environmental Science.
Munns, W.R., A.W. Rea, M.J. Mazzotta, L.A. Wainger, K. Saterson. 2015. Toward a standard lexicon for
ecosystem services. Integrated Environmental Assessment and Management 11(4):666-673.
Munns, W.R., A.W. Rea, G.W. Suter III, L. Martin, L. Blake-Hedges, T. Crk, C. Davis, G. Ferreira, S.
Jordan, M. Mahoney, M.G. Barron. 2015. Ecosystem services as assessment endpoints for ecological
risk assessment. Integrated Environmental Assessment and Management 12:522-528.
NAS (National Academy of Sciences). 2013. Environmental Decisions in the Face of Uncertainty. The
National Academies Press, Washington, D.C.
Nedkov, S. and B. Burkhard. 2012. Flood regulating ecosystem servicesmapping supply and demand, in
the Etropole municipality, Bulgaria. Ecol Indicators. 21:67-79.
NOAA (National Oceanographic and Atmospheric Administration). 2009. SEDAR 15 Stock Assessment
Report - South Atlantic Red Snapper. South Atlantic Fishery Management Council, North Charleston,
SC, http://sedarweb.org/sedar-15
Norris, R.H., J.A. Webb, S.J. Nichols, M.J. Stewardson, E.T. Harrison. 2011. Analyzing cause and effect
in environmental assessments: using weighted evidence from the literature. Freshwater Science, 31(1),
5-21.
NRC (National Research Council). 2005. Valuing Ecosystem Services: Toward Better Environmental
Decision-Making. The National Academies Press, Washington, DC.
NRC. 2011. Sustainability and the U.S. EPA. The National Academies Press, Washington, DC. 286 pp.
Olander, L., R.J. Johnson, H. Tallis, J. Kagan, L. Maguire, S. Polasky, D.L. Urban, J. Boyd, L.A. Wainger,
M. Palmer. 2015. Best Practices for Integrating Ecosystem Services into Federal Decision Making:
National Ecosystems Partnership, Duke University doi:10.13016/M2CH07.
Oosterbroek, B., J. de Kraker, M.M. Huynen, P. Martens. 2016. Assessing ecosystem impacts on health: a
tool review. Ecosystem Services, 17, 237-254.
Orlando, J. and S. Yee. 2017. Linking terrigenous sediment delivery to declines in coral reef ecosystem
services. Estuaries and Coasts 40(2):359-375
Parnell, G.S., H.W. Conley, J.A. Jackson, L.J. Lehmkuhl, J.M. Andrew. 1998. Foundations 2025: A value
model for evaluating future air and space forces. Management Science 44: 1336-1350
Paracchini, M.L., G. Zulian, L. Kopperoinen, J. Maes, J.P. Schagner, M. Termansen, M. Zandersen M, M.
Perez-Soba, P.A. Scholefield, G. Bidoglio. 2014. Mapping cultural ecosystem services: A framework to
assess the potential for outdoor recreation across the eu. Ecol Indicators. 45:371-385.
68
-------
Pascual, U., R. Muradian, L. Brander, E. Gomez-Baggethun, M. Martin-Lopez, et al. 2010. The Economics
of Ecosystems and Biodiversity Ecological and Economic Foundations. In: Kumar P (ed.) The
Economics of Valuing Ecosystem Services and Biodiversity. London: Earthscan, 183-256.
Pereira, E., C. Queiroz, H.M. Pereira, L. Vicente. 2005. Ecosystem services and human well-being: a
participatory study in a mountain community in Portugal. Ecology and Society 10: 14-article 14.
Pinto, R., V.N. de Jonge, J.C. Marques. 2014. Linking biodiversity indicators, ecosystem functioning,
provision of services and human well-being in estuarine systems: Application of a conceptual
framework. Ecological Indicators 36: 644-655.
Portman, M. 2013. Ecosystem services in practice: challenges to real world implementation of ecosystem
services across multiple landscapes - a critical review. Applied Geography 45:185-192.
Posner, S.M., E. McKenzie, T.H. Ricketts. 2016. Policy impacts of ecosystem services knowledge.
Proceedings of the National Academy of Sciences 113:1760-1765.
Potschin, M.B., R.H. Haines-Young. 2011. Ecosystem services. Progress in Physical Geography.
35(5):575-594.
Qiu, J. 2016. Causal chains linking ecosystem changes and society: from theory to implementation.
Presentation to ACES 2016 (A Community on Ecosystem Services). December 5-9, Jacksonville EL.
Reacher, M., K. McKenzie, C. Lane, T. Nichols, I. Kedge, A. Iversen, J. Simpson. 2004. Health impacts of
flooding in Lewes: a comparison of reported gastrointestinal and other illness and mental health in
flooded and non-flooded households. Communicable Disease and Public Health 7(l):39-46.
Rehr, A.P., M.J. Small, P. Bradley, W.S. Fisher, A. Vega, K. Black, T. Stockton. 2012. A decision support
framework for science-based, multi-stakeholder deliberation: A coral reef example. Environmental
Management 50(6): 1204-1218.
Rehr, A.P., M.J. Small, P.S. Fischbeck, P. Bradley, 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(l):60-87.
Reum, J.C., B.E. Ferriss, P.S. McDonald, D.M. Farrell, C.J. Harvey, T. Klinger, P.S. Levin. 2015.
Evaluating community impacts of ocean acidification using qualitative network models. Marine
Ecology Progress Series 536:11-24.
Rhodus, J., F. Fulk, B. Autrey, S. O'Shea, A. Roth. 2013. A Review of Health Impact Assessments in the
U.S.: Current State-of-Science, Best Practices, and Areas for Improvement. U.S. Environmental
Protection Agency, Cincinnati, OH. EPA/600/R-13/354.
Romo-Lozano, J.L., J. Lopez-Upton, J.J. Vargas-Hernandez, M.L. Avila-Angulo. 2017. Economic
valuation of the forest biodiversity in Mexico, a review. Revista Chapingo Serie Ciencias Forestales Y
Del Ambiente 23: 75-90.
Ronnback, P. 1999. The ecological basis for economic value of seafood production supported by mangrove
ecosystems. Ecological Economics 29: 235-252.
Russell, M., J. Harvey, P. Ranade, K. Murphy. 2015. EPA H20 User Manual. US. EPA Office of Research
and Development, National Health and Environmental Effects Research Laboratory, Gulf Ecology
Division. No. EPA/600/R-15/090.
Samonte-Tan, G.P.B., A.T. White, M.A. Tercero, J. Diviva, E. Tabara, C. Caballes. 2007. Economic
valuation of coastal and marine resources: Bohol Marine Triangle, Philippines. Coastal Management
35: 319-338.
Schaich, H., C. Bieling, T. Plieninger. 2010. Linking Ecosystem Services with Cultural Landscape
Research. GAIA-Ecological Perspectives for Science and Society 19:269-277.
Scholte S.S.K., A.J.A. van Teeffelen, P.H. Verburg. 2015. Integrating socio-cultural perspectives into
ecosystem service valuation: A review of concepts and methods. Ecological Economics 114:67-78.
Smith, L. M., H. M. Smith, J. L. Case, L. Harwell. 2012. Indicators and Methods for Constructing a U.S.
Human Well-being Index (HWBI) for Ecosystem Services Research. U.S. Environmental Protection
69
-------
Agency, Washington, DC, EPA/600/R-12/023.
Smith, L.M., J.L. Case, H.M. Smith, L.C. Harwell, J.K. Summers. 2013. Relating ecoystem services to
domains of human well-being: Foundation for a US index. Ecological Indicators 28: 79-90.
Smith, L.M., L.C. Harwell, J.K. Summers, H.M. Smith, C.M. Wade, K.R. Straub, J.L. Case. 2014. A U.S.
Human Well-being Index (HWBI) for multiple scales: linking service provisioning to human well-
being endpoints (2000-2010). EPA/600/R-14/223.
Summers, J.K., L.M. Smith, L.C. Harwell, J.L. Case, C.M. Wade, K.R. Straub, H.M. Smith. 2014. An
index of human well-being for the U.S.: A TRIO approach. Sustainability 6: 3915-3935.
Summers, J.K., L.C. Harwell, L.M. Smith. 2016. A model for change: An approach for forecasting well-
being from service-based decisions. Ecological Indicators 69:295-309
Suter, G.W., III. 1990. Endpoints for regional ecological risk assessments. Environmental Management
14:9-23.
Tenbrink, M., G. Foley, J. Hunter, H. Walker, R. Eric. 2016. Narragansett-3VS Model Overview
(Information Sheet). US EPA Office of Research and Development, Washington, DC, EPA/600/F-
16/060.
Townsend, M., S.F. Thrush, A.M. Lohrer, J.E. Hewitt, C.J. Lundquist, M. Carbines, M. Felsing. 2014.
Overcoming the challenges of data scarcity in mapping marine ecosystem service potential. Ecosystem
Services 8, 44-55.
Troy, A. and M.A. Wilson. 2006. Mapping ecosystem services: practical challenges and opportunities in
linking GIS and value transfer. Ecological Economics 60:435-449.
Turner, K.G., S. Anderson, M. Gonzales-Chang, R. Costanza, S. Courville, T. Dalgaard, E. Dominati, I.
Kubiszewski, S. Ogilvy, L. Porfirio, et al. 2016. A review of methods, data, and models to assess
changes in the value of ecosystem services from land degradation and restoration. Ecological Modeling
319:190-207.
Tuya, F., R. Haroun, F. Espino. 2014. Economic assessment of ecosystem services: Monetary value of
seagrass meadows for coastal fisheries. Ocean & Coastal Management 96: 181-187.
USDA (U.S. Department of Agriculture) Forest Service. 2012.1-Tree Eco User's Manual v.5. Online:
USDA Forest Service.
Van Den Berg, A.E., T. Hartig, H. Staats. 2007. Preference for nature in urbanized societies: Stress,
restoration, and the pursuit of sustainability. Journal of Social Issues 63(l):79-96
Van Wensem J., P. Calow, A. Dollacker, L. Maltby, L. Olander, M. Tuvendal, G. van Houtven. 2016.
Identifying and assessing the application of ecosystem services approaches in environmental policies
and decision making. Integrated Environmental Assessment and Monitoring 13:41-51.
Vanclay, F., and A.M. Esteves. 2011. Current issues and trends in social impact assessment. In: F. Vanclay
and A.M. Esteves, eds. New directions in social impact assessment: conceptual and methodological
advances. Cheltenham: Edward Elgar, 3-19.
Verburg R., T. Seines, P. Verweij. 2016. Governing ecosystem services: national and local lessons from
policy appraisal and implementation Ecosystem Services, 18:186-197
Villa, F., K.J. Bagstad, B. Voigt, G.W. Johnson, R. Portela, M. Honzak, D. Batker. 2014. A methodology
for adaptable and robust ecosystem services assessment. PLoS ONE. 9(3):e91001.
von Winterfeldt D and Edwards W. 1986. Decision Analysis and Behavioral Research. Cambridge,
England, Cambridge University Press, 604 pages.
Wade T.J., C.J. Lin, J.S. Jagai, E.D. Hilborn. 2014. Flooding and emergency room visits for
gastrointestinal illness in Massachusetts: A case-crossover study. PloS one, 9(10), el 10474.
Wainger, L.A., D.M. King, R.N. Mack, E.W. Price, T. Maslin. 2010. Can the concept of ecosystem
services be practically applied to improve natural resource management decisions? Ecological
Economics 69:978-987.
Wainger, L. and M. Mazzotta. 2011. Realizing the potential of ecosystem services: A framework for
70
-------
relating ecological changes to economic benefits. Environmental Management 48(4):710-733.
Williams, B.K., R.C. Szaro, C.D. Shapiro. 2009. Adaptive Management: The U.S. Department of the
Interior Technical Guide. Adaptive Management Working Group, U.S. Department of the Interior,
Washington, DC.
Yee S.H., J.E. Rogers, J. Harvey, W. Fisher, M. Russell, P. Bradley. 2011. Concept mapping ecosystem
goods and services. In: Applied Concept Mapping: Capturing, Analyzing, and Organizing Knowledge,
Moon BM, Hoffman RR, Novak JD, Canas AJ (editors), Boca Raton, FL: CRC Press, pp 193-214
Yee, S.H., J. Carriger, W.S. Fisher, P. Bradley, B. Dyson. 2014. Developing scientific information to
support decisions for sustainable reef ecosystem services. Ecological Economics 115: 39-50.
Yoskowitz, D.W., S.R. Werner, C. Carollo, C. Santos, T. Washburn, and G.H. Isaksen. 2016. Gulf of
Mexico offshore ecosystem services: Relative valuation by stakeholders. Marine Policy 66: 132-136.
71
-------
Glossary
Glossary definitions were obtained from:
Harwell, M.C., C. Jackson, and J. Molleda. 2017. Managed Vocabulary for use of Ecosystem Goods and
Services in Decision-Making. U.S. Environmental Protection Agency, Gulf Breeze, EL, EPA/600/X-
17/168.
Alternatives: Alternative inputs, policies, or solutions open to a decision-maker.
Area of Concern: Geographic areas within the Great Lakes where significant impairment of beneficial uses
has occurred as a result of human activities at the local level.
Beneficiary: In common usage, a beneficiary is "a person who receives benefits." However, because a single
person with multiple interests can benefit from ecosystems in multiple and distinct ways, the Final
Ecosystem Goods and Services Classification System uses the term beneficiary to refer to the person's
awareness and interests, relative to ES, rather than the person themselves. Therefore, the Final Ecosystem
Goods and Services Classification System defines beneficiary as "The interests of an individual (i.e.,
person, group, and/or firm) that drive active or passive consumption and/or appreciation of ecosystem
services resulting in an impact (positive or negative) on their welfare."
Benefits: A good, service, or attribute of a good or service that promotes or enhances the well-being of an
individual, an organization, or a natural system.
Benefit-cost analysis: A formal quantitative and sometimes qualitative evaluation of the benefits to be
derived from a decision or action compared to the costs that would be, or have been incurred, by
implementing that decision or action. Benefits and costs may include market (monetary) and nonmarket
values.
Benefits transfer: Techniques to estimate values of ecosystem goods and services in a given context or
location, based on relevant valuation studies conducted in a different context or location.
Classification system: A method to group individual elements or features into collections similar in type,
function, affiliation, behavior, response, or ontogeny.
Conceptual model: A written description and/or visual representation of known or hypothesized relationships
among variables in a system (e.g., human or ecological entities), often representing causes and effects,
environmental stressors, and/or potential management strategies.
Consequence table: In decision analysis, a matrix that characterizes the consequences of proposed actions
with respect to each objective. Also, contains information about performance measures, used to more
precisely define the meaning of objectives.
Contingent valuation: An economic valuation technique based on the stated preference of respondents
regarding how much they would be willing to pay for specified benefits.
Decision analysis: The discipline compromising the philosophy, theory, methodology, and professional
practice necessary to address decisions in a formal manner. Decision analysis includes many procedures,
methods, and tools for identifying, clearly representing, and formally assessing important aspects of a
decision, for prescribing a recommended course of action, and providing insight for the decision-maker
and other stakeholders.
Decision context: The environment in which a decision is made, and the environment that will prevail when
the effects of the decision are brought to bear, including the set of values, preferences, constraints, policies,
and regulations that will affect both the decision-makers and those identified as the ultimate beneficiaries.
Decision Support System (DSS): An interactive system to aid decision-makers in identifying and solving
problems, and making decisions. These systems may use data from observations, output from statistical or
dynamic models, and rules based on expert knowledge.
Decision support tool: A tool that provides resources such as analysis methods, models, data sets, maps, etc.
in order to inform one or more types of decision-making processes.
72
-------
Direct use value: The value of ecosystem goods and services that are directly used by human beings.
Ecological benefit function (EBF): Usable expressions to quantify a change in human well-being that results
from an ecosystem change.
Ecological model: A structured description of an ecological system with qualitative or quantitative
components.
Ecological production function (EPF): Usable expressions (i.e., models) of the processes by which
ecosystems produce ecosystem services, often including external influences on those processes.
Ecological risk assessment: A science-based process that evaluates the likelihood that adverse ecological
effects may occur or are occurring as a result of exposure to one or more stressors.
Ecosystem goods and services: Outputs of ecological processes that directly ("final ecosystem service") or
indirectly ("intermediate ecosystem service") contribute to social welfare. Some outputs may be bought
and sold, but most are not marketed. Often abbreviated as ecosystem services, a common descriptor for
non-technical audiences when describing ecosystem goods and services.
Existence value: The enjoyment people may experience simply by knowing that a resource exists even if they
never expect to use that resource directly themselves.
Final Ecosystem Goods and Services: Components of nature, directly enjoyed, consumed, or used to yield
human well-being. The final ecosystem goods or services is a biophysical quality or feature and needs
minimal translation for relevance to human well-being. Furthermore, a final ecosystem good or service is
the last step in an ecological production function before the user interacts with the ecosystem, either by
enjoying, consuming, or using the good or service, or using it as an input in the human economy.
Fundamental objectives: In decision analysis, the basic things that matter, and the outcomes you really care
about regardless of how they are achieved.
Goods: Tangible items that people may consume or purchase to satisfy needs and wants.
Health: A state of complete physical, mental, and social well-being and not merely the absence of disease or
infirmity.
Human well-being: A multidimensional description of the state of people's lives, which encompasses
personal relationships, strong and inclusive communities, meeting basic human needs, good health,
financial and personal security, access to education, adequate free time, connectedness to the natural
environment, rewarding employment, and the ability to achieve personal goals.
Human Well-being Index (HWBI): An index of well-being for the U.S. based on indicators and metrics
derived from existing measures of well-being.
Indicator: An interpretable value or category describing trends in some measurable aspect, often used
singularly or in combination to generate an index.
Indirect use value: The value of ecosystem goods and services that provide benefits outside of the ecosystem
itself.
Intermediate ecosystem goods and services (IEGS): Attributes of ecological structure or processes
(including functions, characteristics, and interactions) that influence the quantity and/or quality of
ecosystem services but do not themselves quantify as final ecosystem goods and services (because they are
not directly enjoyed, consumed, or used by beneficiaries).
Mathematical model: A set of mathematical equations that describe a system.
Means objective: In decision analysis, objectives that provide a means to fulfill the fundamental objectives.
Metric: A [singular] measurable, observable, or interpretable value.
Natural capital: An extension of the economic concept of capital (manufactured means of production) to
ecosystem goods and services. Natural capital is the stock of natural ecosystems that yields a flow of
valuable ecosystem goods or services into the future.
Nonuse value: The value people hold for an ecosystem attribute or service that they do not use in any tangible
way. Sometimes referred to as "passive use value."
73
-------
Objectives: In decision analysis, statements of what is valuable to stakeholders in a certain context.
Performance measures: A specific metric or indicator that can be used to consistently estimate and report the
anticipated consequences of a management alternative with respect to a particular objective.
Policy: A principle or rule to guide decisions and achieve rational outcomes. Policy differs from law. While
law can compel or prohibit behaviors (e.g., a law requiring the payment of taxes on income), policy merely
guides actions towards those that are most likely to achieve desired outcome.
Process objectives: In decision analysis, objectives that are designed to improve the decision process itself
and do not focus on what should be done, but rather how it should be done.
Resilience: The capacity of a system to absorb disturbance and reorganize while undergoing change so as to
still retain essentially the same function, structure, identity, and feedbacks.
Scalability: The degree to which a relationship (such as an ecological model) that applies at a given spatial or
temporal scale tends to hold at different (especially larger) dependent and/or independent scales.
Service providing area (SPA): A spatially explicit area (e.g., map polygons of contiguous pixels where the
service indicator is present) representing the area providing an ecosystem service.
Stakeholder: An individual, group, or organization with an interest in, or potentially impacted by, the
outcome of a policy or management choice.
Stressor: Any chemical, physical, or biological entity whose presence or absence can induce adverse effects
on ecological components (i.e., individuals, populations, communities, or ecosystems).
Structured decision-making (SDM): An organized approach for identifying and evaluating alternatives that
focuses on engaging stakeholders, experts, and decision-makers in productive decision-oriented analysis
and dialogue and that deals proactively with complexity and judgement in decision-making. It provides a
framework that becomes a decision focused roadmap for integrating activities related to planning, analysis,
and consultation.
Sustainability: To create and maintain conditions under which humans and nature can exist in productive
harmony that permit fulfilling the social, economic, and other requirements of present and future
generations.
Swing-weighting: One of the available methods for eliciting weights for the various criteria defined for multi-
criteria analysis. The swing-weighting method requires specifying hypothetical changes (swings) in the
level of performance against different objectives and then obtaining judgments of the relative preferences
for obtaining those swings, typically using a 0-to-100 scale.
Total economic value (TEV): The concept of total economic value involves measuring the value of the sum
of all flows -present or future- of natural capital provisioned by an ecosystem, including both use and
nonuse values.
Tradeoff: An exchange of one thing in return for another, especially relinquishment of one benefit or
advantage for another.
Transferability: The degree to which a relationship that was developed in a given set of circumstances can
validly be applied in another circumstance.
Use value: The value of a good or service derived from its direct or indirect use (as opposed to nonuse value).
This includes direct and indirect values.
Utility: In environmental management, the usefulness of a product, tool, or information source to its intended
user.
Valuation: The process of expressing a value for a particular good or service in a certain context (e.g.,
decision-making), usually in terms of something that can be counted, often money, but also through
methods and measures from other disciplines (sociology, ecology, and so on).
Value-focused thinking: A philosophy to guide decision-makers. It has three major ideas: start with values,
use values to generate better alternatives, and use values to evaluate those alternatives.
Weight of evidence: The process for characterizing the extent to which the available data support a
hypothesis that an agent causes a particular effect.
74
-------
Appendix A
Approaches to the Decision Process
Al: Using a DASEES Approach
A2: Working within an Existing Process
Clarify
Decision
Context
Implement,
Monitor,
and Review
IE
Evaluate
Trade-offs
and Select
^7
1
Define
Objectives
TT
Develop
Alternatives
Estimate
Consequences
ฃ
75
-------
Appendix A1
Using a DASEES Approach
Problem
In many instances societal decisions regarding community growth (both social and economic) are made
primarily on economic and societal considerations, and do not take into account environmental
considerations as equal components of the decision process. Far too frequently decisions either ignore or
do not adequately factor in consideration of the roles that ecosystem services play in a sustainable system.
Most decision-makers do not currently have access to useful or usable methods and approaches presenting
economic and social choices that will have significant ecosystem impacts. Decision-makers often have the
pieces necessary for a decision but no guidance on how they fit together (Fig. Al.l). This leads to
frustration on the part of the stakeholders and ultimately a disengagement from the process.
Figure Al.l. Illustration of a disorganized process, where all the pieces are present but do not fit
together.
Approach
Effective and user-friendly decision methods empower decision-makers to more explicitly, routinely, and
substantively incorporate ecosystem services into their decision-making process. The goal of the
DASEES (Decision Analysis for a Sustainable Environment, Economy and Society) framework is to
provide decision-makers with an understanding of potential outcomes and effects of their planned
decisions on economic, social, and ecological systems in order to promote more balanced and sustainable
solutions (EPA 2012). DASEES is an open-source, web-based decision analysis framework that
integrates guidance and decision support tools to implement a five step iterative Bayesian decision
process designed to:
1. Understand Decision Context;
2. Define Objectives;
3. Develop Options;
4. Evaluate Options;
5. Take Action.
DASEES is being developed with stakeholder and decision-maker input, through case studies, to ensure
the guidance, tools, and templates meet user needs and facilitate the incorporation of economic, societal
and ecosystem services in the decision-making process.
Beliefs
A f
&
76
-------
What is DASEES?
DASEES is a decision analysis web application that provides guidance and tools for implementing a
coherent and comprehensive decision analysis process. The DASEES process is organized in five steps
(Fig. A 1.2). These steps are designed to integrate and coordinate stakeholder and decision-maker
objectives and values, information, data, and modeling, under decision analysis principles to limit
environmental risk, meet regulatory requirements, and optimize sustainability. DASEES provides
guidance and analysis tools for decision-makers and stakeholders so they can make more informed
decisions through a deep understanding of trade-offs between meeting various decision-maker and
stakeholder objectives. Specifically, DASEES will help decision-makers and stakeholders to easily
incorporate and retrieve information and data in a comprehensive and user-friendly manner.
Question
DASEES
Five Step Generalized Decision Cycle
Decision Support Framework Process
Promotes Deliberative Thinking/Stakeholder Involvement
Amenable to Adaptive Management Approaches
Flexible
Land Risk Management Research
ฆEcosystem Services Research
Sustainable Healthy Communities
Web-based
Guidance
Decision Support Tools
Figure A1.2. The five steps in the DASEES process, and benefits of a DASEES approach.
DASEES Example for Guanica Bay, Puerto Rico
An application from the Guanica Bay watershed of southwestern Puerto Rico provides an example of how
DASEES and its associated tools can help facilitate a decision process (Bradley et al. 2016). Guanica Bay
watershed was the focus of a U.S. Coral Reef Task Force research initiative that brought together multiple
agencies to address the effects of land management decisions on coral reefs. From 2010-2012, stakeholder
workshops were held to better understand what stakeholders value in their watershed, and how human
activity may be affecting environmental condition and ecosystem services. The outcomes of these
workshops were to broaden the decision context and objectives beyond coral reef protection, and provide
a clearer understanding of how decision alternatives might support or conflict with stakeholder objectives.
Information about the decision context, including written descriptions and maps and communication
across key decision-makers, were entered into DASEES using tools to characterize the decision
landscape, draw conceptual models, and build social networks (Fig. A1.3; Appendix Bl). Stakeholder
objectives, elicited through workshop discussions, were entered into DASEES as an objectives hierarchy
(Fig. A1.4; Appendix CI). An important feature of DASEES is that the two latter steps of Developing
Options (Appendix Dl) and Evaluating Options (Fig. A1.5; Appendix E10) are each dependent on its
prior step, to ensure that options development and evaluation are directly linked to stakeholder driven
objectives. DASEES employees Bayesian Belief Networks to model effects of decision options on
measured objectives. DASEES provides a common repository for organizing information related to a
management problem in a single place, facilitating transparency and collaboration.
77
Understand
Context
Define
Obtectwes
Take Action
Stakeholder
Engagement
Evaluate
Options
Develop
Options
Test:
Implement. Monitor, Adapt
Framing:
Asking the right
Analysis:
Develop Hypothesis
-------
DASEES Decision Analysis foraSustdnable Environment, Economy, and Sodety
Guaniea 2013
& Understand Context
Overview
Decision Landscape
Current Condition
System Sketch
Social Network
Map
jfj Define Objectives
Overview
^ Objectives
Objective Preferences
Develop Opbons
Overview
W Define Options
Management Scenarios
, 11 Evaluate Options
Overview
Consequence Table
Consequence Model
O Take Action
Overview
Objective Results
Decision Landscape
Adaptive Management
Ej0l Decision Landscape ~ [OJ
L0 Save ฃ Revert I New Delete Rename Help
The Southwest Puerto Rico Project and the Lajas Valley Irrigation System
A component of Operation Bootstrap was to industrialize agriculture along the southern coast of
Puerto Rico by providing irrigation and cheap hydroelectric energy for pumping water onto fields. As
early as 1908, the South Coast Irrigation Service was formed to maximize farming potential, but
under its aegis only small irrigation projects were completed. In 1915 the first reservoir was built at
Carite (southeast Puerto Rico), which fostered sugar cane production and provided hydroelectric
power for water pumps. Similar plans had been prepared for the southwest by the Puerto Rico
"Utilization of the Water Resources" department, but these never matured, usually for lack of
funding.
In 1941, the department was changed to a public-private entity (Puerto Rico Water Resources
Authority) and with better funding planned and implemented the Southwest Puerto Rico Project
(SWP) and the Lajas Valley Irrigation System (LV1S), a series of five dams and an extensive irrigation
canal and drainage system (Fig. 2-7). The intent of these projects, at an anticipated cost of S32
million (1950 dollars), was to improve sugar cane production in the southwest coastal plain and
provide inexpensive hydroelectric power for farmers to pump irrigation water. The dams were
completed from 1951-1956 and the irrigation system, including drainage of a large lagoon (Guฃnica
Lagoon, northwest of the Bay), was completed by 1961.
Figure 2-8. The Southwest Puerto Rico Project (blue arrows)
consisted of five dams, three of which were In watersheds that
would otherwise flow to the north (see ridge line). The Lajas
Valley Irrigation System (red arrows) consisted of a long canal
that diverted water from Lago Loco across the Lajas Valley for
irrigation, with a return ditch for drainage into Cuinica Bay.
The natural flow of Rio Loco and Rio Hoquer6n is shown (gold
arrows).
Guaniea Bay
igure A1.3. Information describing the Guaniea Bay decision context (from Bradley et al. 2016)
DASEES Decision Analysis for a Sustainable Environment, Economy, and Society
I 5 Project Data ' Jฃ* Logout J ฉ
Guaniea 2013
(J) Understand Context
Overview
Decision Landscape
Current Condition
System Sketch
Social Network
Map
$1 Define Objectives
Overview
tf Objectives
Objective Preferences
Develop Opbons
Overview
t* Define Options
Management Scenarios
a|| Evaluate Options
Overview
Consequence Tabic
Consequence Model
O Take Action
Overview
Objective Results
Decision Landscape
Arifirrtiyr HiTOorrrnl
(Jp Objectives
g Save
Fundamental Objectives Hierarchy
I New Objective Delete Objective M H
d i Sustainable Guaniea Watershed
9 CD Social
(8 O Process
3 CD Economy
3 CD Ecology Land
S3t3 Ecology Aquatic
Reduce uncertainty about outcomes of management act.
Restore lagoon natural processes (absorption of nutrient-
ill D Restore shallow water coral reefs
Improve monitoring and feedback of current actions
13 CD Reduce human contamination
Reduce uncertainty about outcomes of management act.
Protect endangered and threatened marine spebes
Protect mangrove habitats
Protect marine habitat for migratory bird
Improve water quality
Protect marine habitat for migratory bird
Improve water quality
Objective Measures
New Measure
* Measure
<5Hj Coliforms
0
Delete Measure
Nutrients
Solids in suspension
Turbidity
Units
unknown
unknown
unknown
unknown
Measure Value Function N
f* Continuous * Worst Case:
0.2 0.4 0.5 0.6 0.7
Conforms (unknown)
Figure A1.4. Objectives hierarchy elicited from stakeholder workshop discussions for the Guaniea
Bay watershed (from Bradley et al. 2016). Values functions are for illustration only.
78
-------
DASEES
Gu4nfcป3913
\% UndC*stปrtS Confess
Owrvttw
Decison Landscape
OiWltCM)iปn
Systofn StoeGch
Social Network
Msp
Dcfซe Otrjcaivw
Ovซซv*w
y Objects
Otgcdrw Preferences
Develop Opbarts
Qwrvwwr
%/ Dc/oe Options
Micagwvt Sซfwu4
J| EvlluซU O0W
Owntar
C0ซซgซrปซ TWซ
)/ Coowqซeocซ Modซ-i
o Uxtebon
OvtJViOwr
Objcdnc ftesute
ttoc&cn Landscape
Mapbv*
Figure A1.5. Illustration of how a Bayesian Belief Network could be used to predict effects of a
decision option on measured objectives (from Bradley et al. 2016).
Lessons Learned
Use of a structured decision-making approach (Gregory et al. 2012), and specifically the web-based
DASEES approach, fostered greater participation of stakeholders who are typically excluded from the
process - either by design of a traditional top down decision process, or by unintentional omission
because they were unknown. Using the DASEES approach supported broader stakeholder inclusion and
enhanced collaboration among all the stakeholders in helping create better solutions to the complex
problems facing the entire group.
For More Information
Bradley, P., W. Fisher, B. Dyson, S. Yee, J. Carriger, G. Gambirazzio, J. Bousquin, 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.
Gregory, R., L. Failing, M. Harstone, G. Long, T. McDaniels, D. Ohlson. 2012. Structured Decision Making: A
Practical Guide to Environmental Management Choices. West Sussex, UK: Wiley-Blackwell.
EPA (Environmental Protection Agency). 2012. Decision Support Framework Implementation of the Web-based
Environmental Decision Analysis Application DASEES: Decision Analysis for a Sustainable Environment,
Economy, and Society. EPA/600/R-12/008.
79
Hba Aiutfffr
-------
Appendix A2
Working within an Existing Process
Problem
The AOC program is considered to be an example of adaptive governance, which is a multidisciplinary
approach that is useful for evaluating ecosystem services trade-offs in the context of environmental
management goals (Gunderson et al. 2016). Areas of Concern were established in 1987 in response to a
crises of legacy contamination of heavy metals, polychlorinated biphenyls (PCBs), and dioxins, as well as
combined sewage overflows and storm water runoff in Great Lakes coastal communities. The governance
structure as established under the bi-national Great Lakes Water Quality Agreement is polycentric,
comprised of federal, state, and local agencies working with local stakeholders (through a Public
Advisory Committee, or PAC). The goal of the AOC program is to remove beneficial use impairments
(BUIs) in those Great Lakes communities with severe pollution-related problems by remediating
sediments and restoring coastal ecosystems (GLWQA 1987).
Approach
Researchers have been developing approaches to incorporate ecosystem services into decision-making by
providing information regarding how AOC decisions affect ecosystem services, but to do so in a way that
preserves the existing programmatic targets agreed to through the AOC governance structure (Angradi et
al. 2016). The strategy was three-fold: 1) adapt work to fit the established governance and regulatory
structure, 2) utilize an iterative data production model designed to make research products usable, and 3)
generate data and ecosystem services models at a scale that meets the needs of local decision-makers.
The case study worked with the St. Louis River AOC, which includes the communities of Duluth, MN,
Superior, WI, and the Fond du Lac Band of Lake Superior Chippewa reservation. A series of workshops,
designed to improve information relevance by producing data with local stakeholders (sensu Beebeejaun
2015, Posner et al. 2016), provided a forum for state agencies and PAC members to discuss direct and
indirect connections between BUIs and ecosystem services. Researchers worked collaboratively with
stakeholders to build a conceptual model that demonstrates how removing BUIs can lead to ecosystem
services and improve human well-being (see Appendix B4). Workshops were also used to conduct
participatory mapping to identify geographic areas of importance for ecosystem services (Klain and Chan
2012), and co-develop geospatial ecosystem services production models.
Lessons Learned
Notably, the creation of geospatial ecosystem services production models required substantial effort by
the research team to scale-down existing geospatial data sets on natural resources and land uses, or create
new data sets altogether. In addition, we encouraged research team members to participate in AOC
technical meetings to familiarize themselves with the AOC goals and terminology, as well as develop
assessment tools (Blazer et al. 2014, Bellinger et al. 2016, Angradi et al. 2017). Participation helped to
build trust between the research team and the decision-makers, and also served to highlight the
researchers' commitment to supporting the AOC governance structure and decision-making process.
To improve community sustainability, researchers should strive to provide information that integrates
distinct but related decision contexts. Researchers need to recognize where the decision context may not
be the same among related decisions. For the AOC program, sediment remediation and aquatic habitat
restoration are the important steps in removing beneficial use impairments - the measure of success for
the AOC. On the other hand, the reclamation of land and enhancement of ecosystem services enables
communities to improve the quality of life - the measure of success for local governments (Fig. A2.1). If
the decision-support model fails to recognize this important distinction, then ecosystem services
80
-------
production model outputs and other recommendations will not necessarily be compatible with both kinds
of decisions and possibly reinforce the divide between decisions for the land and the water.
Changes
Trails
Retail, restaurants
Riverwalks
New manufacturing
Housing
Marinas
Mechanisms
Comprehensive planning
Land use and neighborhood planning
Park and/or transportation plans
Brownfields remediation
Stormwater management
/green infrastructure
Results
Increased use of water
Stormwater reduction
Improved aesthetics
Land reuse
Neighborhood enhancements
Land side of AOC
Cleaner water, soil, air
create opportunities
for activity
Water side of AOC
Changes
Clean sediment
Restored habitat
Increased
ecosystem services
Remedial Action Plans
Public Advisory Councils
Great Lakes Restoration Initiative
Results
Remove BUI
Cleaner water
Safer fish
More habitat
Environmental R2R IS
the desired end
Figure A2.1. Differences between the land and water sides of Areas of Concern with respect to
changes, policy mechanisms, results, and goals (from Williams and Hoffman 2017).
For More Information
Angradi, T.R., D.W. Bolgrien, J.L. Launspach, B.J. Bellinger, M.A. Starry, J.C. Hoffman, M.E. Sierszen, A.S.
Trebitz, and T.P. Hollenhorst. 2016. Mapping ecosystem services of a Great Lakes estuary can support local
decision-making. Journal of Great Lakes Research 42:717-727.
Angradi, T.R., W.M. Bartsch, A.S. Trebitz, and V J. Brady. 2017. A depth-adjusted ambient distribution approach
for setting numeric removal targets for a Great Lakes Area of Concern beneficial use impairment: Degraded
benthos. Journal of Great Lakes Research 43:108-120.
Beebeejaun, Y., C. Durose, J. Rees, J. Richardson, L. Richardson. 2015. Public harm or public value? Towards
coproduction in research with communities. Environment and Planning C: Government and Policy 33:552-565.
Bellinger, B.J., J.C. Hoffman, T.R. Angradi, D.W. Bolgrien, M. Starry, C. Elonen, T. Jicha, L. Lehto, L. Monson,
M.S. Pearson, L. Anderson, and B.H. Hill. 2016. Water quality in the St. Louis River Area of Concern, Lake
Superior: historical and current conditions and delisting implications. Journal of Great Lakes Research 42:28-38.
Blazer, V.S., J.C. Hoffman, H.L. Walsh, R.P. Braham C. Hahn, P. Collins, Z. Jorgenson, and T. Ledder. 2014.
Health of white sucker within the St. Louis River Area of Concern associated with habitat usage as assessed
using stable isotopes. Ecotoxicology 23:236-251.
GLWQA (Great Lakes Water Quality Agreement). 1987. International Joint Commission, United States and Canada.
Gunderson, L.H., B. Cosens, and A.S. Garmestani. 2016. Adaptive governance of riverine and wetland ecosystem
goods and services. Journal of Environmental Management 183:353-360.
Klain, S.C. and K.M.A. Chan. 2012. Navigating coastal values: participatory mapping of ecosystem services for
spatial planning. Ecological Economics 82:104-113.
Posner, S.M., E. McKenzie, and T.H. Ricketts. 2016. Policy impacts of ecosystem services knowledge. Proceedings
of the National Academy of Sciences 113:1760-1765.
Williams, K.C., and J.C. Hoffman. 2017. Remediation to Restoration to Revitalization - A Path Forward for AOCs?
US Environmental Protection Agency EPA/600/R-17/119.
81
-------
Appendix B
Tools and Approaches for Clarifying the
Decision Context
Implement,
Monitor,
and Review
I
Evaluate
Trade-offs
arid Select
1
Bl: Using DASEES to Characterize Decision Context
B2: Applying FEGS-CS to Identify Key Beneficiaries
&
B3: Assessing Vulnerabilities
B4: Building Conceptual Models
Clarify
Decision
Context
V
c
Define
Objectives
I
Develop
Alternatives
Estimate
Consequences
~
82
-------
Appendix B1
Using a DASEES Approach to Characterize and Understand
Decision Context
Formalized Common Sense
Applying the ideas of value-based decision-making to complex environmental management problems
requires a conceptual framework or formalized process to ensure that a decision is consistent with
stakeholder values, cognizant of trade-offs among alternatives, and accounts for associated uncertainties
and risks. Keeney (1982) described the discipline of decision analysis as " a formalization of common
sense for decisions that were too complex for the informal use of common sense." The DASEES
(Decision Analysis for a Sustainable Environment, Economy and Society) process (Appendix Al)
incorporates this thinking and approach to address multifaceted, multi-stakeholder problems in a
transparent and well documented manner (EPA 2012).
Political, Regulatory, Social, and Institutional Setting
The first step in DASEES is to develop an understanding of the scope of the scientific and decision
setting and context of the problem or management decision at hand. This involves identifying the
environmental, economic and social issues that are part of the problem being considered and thus will
likely be affected by any decisions that are made as part of the process. It also involves identifying the
stakeholders, decision-makers and regulations that will be players in the overall process and the ultimate
decisions. The decision context will be characterized through development of a decision diagram,
showing the political, regulatory, social, and institutional setting of the environmental management
problem. This will provide participants with important information defining the context for their studies.
Examples of this are:
Do the stakeholders utilize common sources of information? Are these sources of information
trusted by all or most of the stakeholders? If there are competing studies, are they presented so all
the stakeholders have this information?
Are management options limited to a set of predefined alternatives or is there flexibility to
explore new approaches?
Are decision options constrained or specified by law or prior agreements?
Are mechanisms in place to include ecosystem services and externality costs in the overall
accounting for the project?
Several tools are used in DASEES to develop and understand the decision context. These tools are not
specific to DASEES and can be utilized as standalone tools separate from a DASEES context.
Scientific Setting
The scientific systems context is addressed through the development of a DPSIR information framework
(conceptual model) for describing causal relationships between the environment and society (Fig. Bl.l).
DPSIR shows the relationship between Driving forces (e.g. need for food, demand for flood control),
Pressures (e.g.,urban growth into farmlands, development in flood zones), States (the response of
environmental compartments and ecological variables), Impacts (changes in ecosystem services,
economic systems, social systems), and Responses (actions taken, policy decisions). The DPSIR process
facilitates defining a broad set of responses from which specific potential management options can be
developed. The conceptual model depicts the lines of connection from the initially proposed actions
(Drivers), which bring the stakeholders to the table, to the potential effects those drivers may have on
environmental components (Pressures and States), which potentially lead to effects on the well-being of
people (Impacts) (Fig. B1.2).
83
-------
Driving Forces
Socioeconomic sectors and
cultural factors that drive
human activities (causes)
/
Response
Response of society to the
environmental situation
(policies, decisions)
Pressure
Human activities that
place stress on the
environment (pollutants)
\
State
Condition of the environment
(composition, distribution,
quality)
Impact
Effects of environmental
degradation (changes in
attributes, services)
J
Figure Bl.l. The components of the DPSIR framework.
Need
of hutotx
iTHjalW*
system
Tim* to
firm
(ACoซt)
for Bird*
and Wildlife
Aesthetics
riculture
Tourism
AA/ater
delivery
lftlegr-fy
Cyc*)
Orl-Kfvj
Water
Touป1ปm &
Recreation
Project costs
Responses
Driver = Agriculture
Pressures
Agriculture
Chang
Production
(Quantity of
Coffaa Crop)
State
Impact
Terrestrial
ecosystems
Fisheries
Figure B1.2. An example of a DPSIR created through Guanica Bay Puerto Rico workshop
discussions on impacts of agricultural practices on coral reefs (see Appendix Al). Red text
illustrates how objectives about ecosystem services benefits can be inferred from conceptual model
development (from Bradley et al. 2016).
84
-------
SystemSketch
It is critical for decision-makers to be able to clearly understand the linkages between ecosystem
functions and ecosystem goods and services and the benefits people derive from them in the form of
economic, social and cultural value. Often times these connections are complicated and not intuitively
understood. SystemSketch is a dynamic, graphic visualization tool that can provide stakeholders and
decision-makers a clear visual of how different systems are linked and have access to information
resources to help better understand the system context for the problem at hand (Fig. B1.3). It is built using
the DPSIR framework, and functions both as a stand-alone tool and as a component in DASEES allowing
users to better analyze decision options and trade-offs. It allows decision-makers and stakeholders to not
only visualize the system linkages, but provides a visual of stakeholder values as well. With a clear
picture of these linkages that is developed in an open and transparent manner, decision-makers and
stakeholders arc better positioned to integrate economic, social, cultural and environmental information
into the decision process. This can increase cooperation among stakeholders and decision-makers, while
reducing the potential for creating unintended consequences.
1 SystemSketch
5 O-Ma Import
El
Legend
1*1
Categories
ซ Koef Hobitat Crop Produc lion (tuMnqftl
Homo Construction ft Main tenant: o Aesthetic or Sensory Vah*e Sketch 6
filter:
Search:
Sketch Set -
Q. Q 9- I - - ,T ' .
I
Crwers
Pressures
State
Impacts
dl Ecosystem Sflrvtts
jj ProviscrMng Services
si Regulating Services
j Si^portng Services
Carbon Storage 6 Cycfog
j Cornpte* Habitat S Resources
Contaminant Processing
Ecological Integrity
jj Nutrient Cychng
Primary Product**)
j struc%i-e & Stafcrfeatnn
Water Cyclrig
Watershed integrity
D Human Well-Beng
Re<ซ)or*ses
UbdniJtan $ <ฃ>ป
Sfrwrf # O*
So d W'ซr# frsposs Siซ $ <ฃป
Land Dtvwlapmmnl ฃ *
Impfcan 0
FซflilcซUmง ^
Domestic Annul V.jate ฃ ป
St:*m*alec Management ฃ cnป
ParVmg Ldt 0
-------
NRCS
NOAA NOAA
NIAA
NOAA
UPRM
PRDNER
UPRM
UPRM UPRM
\NOAA
NOAA
EPA
NOAA
NRCS
EPA
CWP
NOAA
PRDNER
PR DNER
NOAA
NOAA NOAA
EPA
Consultant
NOAA
Consultant
EPA
PR DNER
EPA
EPA
Citizen
EPA
NOAA NOAA
CAMARED
PRASA
Citizen
CAP
NOAA
Citizen
PRDNER
Guanica
UPRM
NRCS
PR DNRA
UPRM
NRCS
PR DNER PR DNER
Figure B1.4. Example social network analysis, illustrating well-connected and isolated groups of
decision-makers and stakeholders.
Decision Landscape
Taken together the Political, Regulatory, Social, Institutional and Scientific context provides the Decision
Landscape of the decision. In the diagram below, DASEES contributes to the decision landscape in the
areas with red text (Fig. B1.5).
Decision Makers
have
authority
influence
Regulators &
Enforcers
Interested & Affected
Parties
inform
decide
among
inform
issue
Decision Support
Providers
Mandates,
Rules &
Standards
Decision
Options
constraii
characterize
Science (relationship
between options &
outcomes)
characterize
Outcomes
(Environmental,
Ecosystem
Services, Cost)
influences
External
Variables
predict
influence ~
Valuation by participants
(Utility)
Preferences
& Values
Figure B1.5. The Decision Landscape (from Rehr et al. 2012)
86
-------
For More Information
Bradley P, Fisher W, Dyson B, Yee S, Carriger J, Gambirazzio G, Bousquin J and Huertas E. 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.
EPA (Environmental Protection Agency). 2012. Decision Support Framework Implementation of the Web-based
Environmental Decision Analysis Application DASEES: Decision Analysis for a Sustainable Environment,
Economy, and Society. EPA/600/R-12/008.
Keeney, R. (1982) Decision Analysis: An Overview. Operations Research, Vol. 30, No. 5 (Sep. - Oct., 1982), pp.
803-838.
Rehr, A.P., Mitchell J. Small, Patricia Bradley, William S. Fisher, Ann Vega, Kelly Black, Tom Stockton (2012) A
Decision Support Framework for Science-Based, Multi-Stakeholder Deliberation: A Coral Reef Example.
Environmental Management, December 2012, Volume 50, Issue 6, pp 1204-1218.
87
-------
Appendix B2
Applying FEGS-CS to Identify Key Beneficiary Groups
Background
Great Lakes Areas of Concern have been called a site of innovation in the participatory governance of
water resources (Botts and Muldoon 2005; Grover and Krantzberg 2012; Williams 2015). The
participation of the public and stakeholders in environmental decisions started in the mid-1980s and is
now considered a norm. More importantly, water quality has improved throughout the Great Lakes, and
communities are starting to notice new users of the water resources, as well as new waterfront
developments. Recognizing this change, the Great Lakes National Program Office is working to
understand how communities make decisions about in relation to water resources. Remediation to
Restoration to Revitalization (R2R2R) is used to describe the flow of decisions that starts with the
remediation of contaminated sediments and/or restoration of habitat that eventually results in community
revitalization.
In Great Lakes Areas of Concern (AOC), attention has historically been directed towards restoring the
beneficial uses of the ecosystem (focused on the water) by removing Beneficial Use Impairments through
contaminated sediment remediation and habitat restoration. Expanding the vision to include the adjacent
community has introduced new challenges, including engaging with a larger and different group of
stakeholders. The Final Ecosystem Goods and Services Classification System (FEGS-CS) is a tool that
can be used to identify beneficiaries of environmental goods and services in a particular type of
environment (Chapter 2, Table 2.2; Landers and Nahlik, 2013). Beneficiaries are those individuals or
groups who might directly use or enjoy the environment. Examples of beneficiaries in a river environment
might include recreational paddlcrs, transporters of goods or people, anglers, or energy generators.
Approach
The Great Lakes AOC provides an illustrative example to demonstrate how FEGS-CS can be used to
identify stakeholders and beneficiaries. The City of Duluth has been developing active recreation
resources in response to AOC clean-up efforts. As part of the larger investigation, we analyzed
overlapping governance processes to identify who was participating in AOC and community
revitalization processes. For this example, we focus on the St. Louis River Estuary National Water Trail
Master Plan (Fig. B2.1). The plan was analyzed to identify which stakeholders were participating and
used FEGS-CS as an analysis tool to identify the interest and values of the stakeholder groups.
ST. LOUIS RIVER ESTUARY
NATIONAL WATER TRAIL MASTER PLAN
OULUTH, MINNESOTA & SUPERIOR, WISCONSIN
rMfch <>
DULUTH
Figure B2.1. Image of the St. Louis River Estuary National Water Trail Master Plan.
88
-------
Example: Analysis of the St. Louis River Estuary National Water Trail Master Plan
One indicator of community revitalization in response to AOC environmental restoration is the pursuit of
a National Water Trail designation for the St. Louis River by two municipalities. The Cities of Duluth,
Minnesota and Superior Wisconsin prepared a master plan and identified over fifty project partners,
stakeholders, and outreach partners to take part in the planning (Cities of Duluth, MN and Superior, WI
2016). Using FEGS-CS as a key, the participant list was coded and analyzed to identify beneficiaries. The
document cited the importance of the "outdoor recreation economic engine," signaling a community
economic revitalization motivation to the plan.
The intent of the Cities is to create a regional recreational amenity to contribute to a growing outdoor
recreational-based economy. That intent is reflected in the number of recreational and educational
beneficiaries. An analysis of the stakeholders would indicate that ten of the stakeholder groups could be
identified as resource-dependent recreation-based businesses or non-profit organizations. In addition, user
groups like boating and fishing clubs have taken part in the process. Furthermore, the participation of over
ten educational or research institutions indicated that learning is an important aspect of the water trail and
value of the communities.
The Twin Ports of Duluth and Superior have national and international importance as a bulk cargo port,
thus commercial interests were not ignored. Recognizing the importance of shipping and recreation, the
Seaway Port Authority was also active in the process. In fact, the master trail maps identify where
shipping is the major activity and can be a hazard to recreational boaters.
Take Home Message
FEGS-CS can be used to identify community values through an analysis of existing data. In this case, the
local governments are dedicated to improving access to clean water resources because it is an important
economic activity. An analysis of key documents can reveal the preferences and values of communities
and decision-makers.
The example of the St. Louis Water Trail is an illustrative example because the river is a multi-purpose
waterway. First, the river is an important recreational resource and valued by Duluth and Superior, and
both cities recognized the opportunity to grow the outdoor recreation sector. On the other hand, the Port
of Duluth and Superior is the western-most port on the Great Lakes and one of the largest bulk-cargo
handlers in the US. The Cities recognized the overlapping interests and created an inclusive process to
ensure (sometimes) competing users of the river had a voice in the process. Industrial, motor and human-
powered boating were all represented.
We used FEGS-CS as an analysis tool, however, the St. Louis River Water Trail example also provides a
list of beneficiary groups of river restoration projects.
Beneficiaries in the St. Louis Water Trail planning process and (potential representative stakeholders)
Government, municipal, residential (agencies, local governments, community clubs)
Commercial/Military Transport (intergovernmental transportation planners, port
authority)
Commercial/industrial - resource-based businesses (visitors bureau, tour companies,
marinas, outfitters, environmental nonprofits, fishing charter businesses)
Subsistence food gatherers (tribal agencies)
Recreation - boating (motor boating clubs, paddling groups, sailing organizations)
Recreation - angling and hunting (fishing clubs, Ducks Unlimited)
Recreation - experiencers and viewers (birdwatchers, conservation groups, scenic train)
89
-------
Learning - educators (schools, environmental nonprofits, university outdoors clubs,
American Canoe Association-affiliated groups)
Learning - research (universities, agencies, historical societies, cultural organizations)
People who care - (land trusts, conservation organizations)
Inspiration (tribal members, humanity)
For More Information
Botts, L., and P. Muldoon. 2005. Evolution of the Great Lakes water quality agreement. Michigan State University
Press.
Cities of Duluth, MN and Superior, WI. (2016). St. Louis River Estuary National Water Lrail Master Plan.
Published by the City of Duluth. Accessed online at http://www.duluthmn.gov/media/541652/final-water-trail-
12916.pdf
Landers D.H. and A.M. Nahlik. 2013. Final Ecosystem Goods and Services Classification System (FEGS-CS).
EPA/600/R-13/ORD-004914. U.S. Environmental Protection Agency, Office of Research and Development,
Washington, D.C. https://gispub4.epa.gov/FEGS/FEGS_home.html
Grover, V.I., and G. Krantzberg, eds. 2012. Great Lakes: Lessons in participatory governance. CRC Press.
Williams, K. C. 2015. Building bridges in the Great Lakes: how objects and organization facilitate collaboration
across boundaries. Journal of Great Lakes Research 41:180-187.
90
-------
Appendix B3
First Steps toward Assessing Coastal FEGS Vulnerabilities
from Environmental Change
Background
Numerous coastal Final Ecosystem Goods and Services (FEGS), e.g., property protection, recreational
opportunities, food production, etc., depend on the habitats where they are produced. Land-use planning
efforts within coastal communities can therefore be improved by accounting for the array of potential
human beneficiaries in a given area, an understanding of how they derive benefits from their coastal
environment (e.g., which FEGS they use), and thinking long-term about how the availability of those
FEGS can be affected by altering the coastal environment (Fig. B3.1). A methodology is needed that
allows coastal community stakeholders and leaders to quickly see how local ecosystem services are tied
to FEGS-producing habitats. Linking coastal beneficiaries to habitat classes that can be visually depicted
on a map serves as a first step in strengthening the short and long-term resilience of FEGS in coastal
communities.
Therefore... "
Change in FEGS
Figure B3.1. Conceptual model demonstrating the link between coastal habitats and potential
FEGS delivery.
Approach
Available scientific literature is being reviewed and synthesized in a weight of evidence approach for
characterizing the association of selected FEGS and coastal habitats, with a primary focus on evidence
that links specific human beneficiaries to those habitats.
This approach takes advantage of existing classification schemes for identifying FEGS, beneficiaries, and
coastal habitats. FEGS and beneficiaries are identified based on the guidance developed by Landers and
Nalilik 2013. Habitat classes are incorporated in accordance with the NOAA Coastal and Marine
Ecosystem Classification Standard (CMECS), endorsed by the Federal Geographic Data Committee
(FGDC), currently under various stages of implementation throughout the United States (FGDC 2012).
Habitat(s) most relevant for each FEGS beneficiary group will be identified by systematically scoring
91
-------
literature sources to assess the collective weight of evidence for those linkages, given differences in how
communities may weigh criteria based on their priorities. Criteria used for evaluating the relevant
evidence for a particular location include the scale and proximity of a given source; degree of beneficiary
specificity (i.e., Does the study identify specific beneficiaries, the ecological elements used, or how FEGS
are used?); origin of the data (e.g., Was the study based solely on other literature, opinion surveys, or
direct observation by the authors?); and habitat classification (i.e., How narrow/broadly are habitats
defined and how closely do habitat classes align with CMECS?).
Example Results
Hypothetical results demonstrate how the information might assist community stakeholders. Tillamook
Bay, Oregon is the location of one of EPA's National Estuary Programs and provides an example of the
mapping resources available in a state that has been proactively implementing the CMECS standard and
making map layers publicly available. Hypothetical output for habitats (using the CMECS classification
standard) most relevant for residential property owners are identified based on the cumulative weight of
literature evidence (Fig. B3.2).
Hypothetical Output
FEGS Category - Property Protection
Linkage - Property Owners:Coastal Habitats
i Estuarine Openwater
Rock
i UMS
iBeach
i Dune/ Dune field
i Flat
Reef Biota
I Faunal Bed
Aquatic Bed
Emergent Wetland
i Scrub-shrub Wetland
i Forested Wetland
figure B3.2. Example chart of the type of output that will be generated once literature evidence for
FEGS beneficiary:habitat linkages is compiled.
Mapping these habitats might then assist communities in determining, for example, which habitats seem
to contribute most to the delivery of property protection benefits, in conjunction with maps depicting the
landscape in their area (Fig. B3.3). Having beneficiaries linked to mapped habitat classes makes it easier
to incorporate FEGS considerations into land-use planning efforts and visualize priority habitats, from a
FEGS perspective, alongside other spatial data.
92
-------
Emergent
Wetland
19%
Property
Owners
Aquatic
Bed
Figure B3.3. The map depicts actual habitat classes (from Clinton et al. 2007) for Tillamook Bay,
Oregon (top inset boxes), though beneficiary (FEGS) linkages (bottom left inset) are completely
hypothetical.
i Estuarine Openwater
i Rock
i UMS
i Beach
i Dune/ Dune field
i Flat
i Reef Biota
i Faunal Bed
ป Aquatic Bed
Emergent Wetland
ฆ Scrub-shrub Wetland
ฆ Forested Wetland
] Aquatic Bed (AB)
j AB:Benthic Macroalgae
[ AB:Aquatic Vascular Vegetation
Emergent Wetland (EW)
EW:Emergent Tidal Marsh (ETM)
EW:ETM:Emergent Brackish TM
Scrub-Shrub Wetland (SSW)
SSW:Tidal SSW (TSSW)
SSW:TSSW: Brackish TSSW
Forested Wetland (FW)
FW: Tidal FW
Unclassified
Barriers and Opportunities
The approach outlined for assessing the relationship between FEGS and coastal habitats has certain
limitations. For example, habitat is not an exact proxy for FEGS delivery. Thus, while understanding the
beneficiary:habitat linkages will improve a community's ability to make proactive plans based on their
priorities, this information alone will not allow stakeholders to make specific predictions about the
amount of potential loss in services if habitats are lost or replaced. Nonetheless, assessing the weight of
evidence for beneficiary:habitat relationships serves as a first step toward identifying areas where more
targeted valuation studies might be wan-anted. There are opportunities for input to the development of this
methodology, including how this approach may be extended to terrestrial and freshwater environments.
For More Information
Clinton, P.J., D.R. Young, D.T, Specht, and H. Lee. 2007. A guide to mapping intertidal eelgrass and nonvegetated
habitats in estuaries of the Pacific Northwest USA. U.S. Environmental Protection Agency, Washington, DC,
EPA/600/R-07/062.
FGDC (Federal Geographic Data Committee). 2012. Coastal and Marine Ecological Classification Standard. Federal
Geographic Data Committee, FGDC-STD-018-2012.
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.
93
-------
Appendix B4
Working with Stakeholders to Build Conceptual Models
Background
Defining the socio-ecological context of decision-making is fundamental for characterizing ecosystem
services (Chan et al. 2012). Our goal was to co-develop a conceptual model that relates the production of
ecosystem services to decisions, actions, and goals of the EPA Area of Concern program (within its
programmatic structure), using an iterative, participant-based process (Menzel and Teng 2009).
The EPA Area of Concern (AOC) program began in the late 1970s and is an early example of an
ecosystem-based approach founded on the maintenance of ecosystem integrity and recognition of human
use of and benefits from nature (i.e., Christensen et al. 1996; see Appendix A2). The foundation of the
AOC program is that Great Lakes coastal ecosystems provide beneficial uses for humans such as drinking
water, clean sediment, and fish to eat. Beneficial use impairments were established for environmental
problems such as beach closures, fish consumption advisories, dredging restrictions, and excess nutrients
and sediment (GLWQA 1987). These beneficial use impairments were identified by stakeholders within
Great Lakes coastal communities and are analogous to ecosystem services (Angradi et al. 2016). The goal
of the AOC is to remove BUIs through ecosystem restoration and ultimately de-listing AOCs. The AOC
program is based on stakeholder input and participation, and thus a stakeholder-based process was
paramount for considering ecosystem services (Portman 2013).
Approach
To provide usable ecosystem service assessments for decision-makers, we held four monthly workshops
with the St. Louis River Area of Concern (AOC; one of 27 U.S. AOCs in the Great Lakes) to iteratively
co-produce a conceptual model of how the decision-making structure and beneficial use impairment
(BUI) removal relate to ecosystem services. It was useful to work with the St. Louis River AOC because
community stakeholders, conservation groups, and state agencies had previously developed by consensus
conservation goals for the AOC that "[would] achieve a mix of ecological and social benefits, [and]
presents a new vision of the St. Louis River ecosystem towards which communities, organizations, and
individuals can work in cooperation" (SLRCAC 2002). We then held a second workshop with several of
the federal AOC partners (EPA, USGS, and NOAA) in which we presented the conceptual model
developed in partnership with the St. Louis River AOC to provide an opportunity to further revise and, if
needed, generalize the model.
An initial version created by the researchers and St. Louis River AOC stakeholders described a static and
linear relationship between AOC goal establishment (selecting a project from the AOC implementation
framework that provides a roadmap for delisting; St. Louis River AOC 2013 ), project design (taking into
account how the project contributes to achieving targets for BUI removal), sediment remediation and
aquatic habitat restoration preferred project recommendations, engineering evaluation, and final project
design (Fig. B4.1). Feedback only occurred within different aspects of making technical planning
decisions (design development, evaluation), and only within the AOC leadership team (coordinators) and
associated AOC workgroups. That is, public participation was not recognized. Further, Final Ecosystem
Goods and Services (FEGS) were an input into the design process only. Also, the conceptual model
emphasized the workflow of project planning and did not provide explicit connections to improved
environmental quality, decision-maker information, or community benefits. This is striking given that the
goal of the program is to restore beneficial uses by changing the ecosystem.
94
-------
Project Selection
(Delisting Roadmap)
AOC Coordinators
Ecological
Design +
BUI Targets
Remediation and
Restoration
Recommendations
AOC Coordinators,
Workgroups
jO^l
Engineering
Evaluation
^_7
A FES
Permitting,
Final Design
Figure B4.1. Preliminary conceptual framework for the use of ecosystem service trade-off analysis
to support decision-making in an estuarine Great Lakes AOC. Small arrows indicate flow of
decisions.
The final conceptual model was a substantial improvement with respect to recognizing the role of
decisions-makers and identifying their information needs, as well as defining how AOC programmatic
goals (BUI removal, AOC delisting) relate to EGS ("ecosystem benefits" in Fig. B4.2) and community
benefits ("revitalization" in Fig. B4.2). Importantly, the conceptual model became dynamic and cyclical,
focusing on change over time and recognizing that policies are shaped by community revitalization goals
(Angradi et al. 2016).
AOC delisted
A Biophysical
state of the
ecosystem
Ecosystem
mediated
processes
Community
revitalization
(A) Policies
& decisions
A Ecosystem
benefits
R2R
implementation
Information for decision-makers:
SPA maps, tradeoff analyses, benefit tracking
Figure B4.2. Conceptual framework for the use of ecosystem service mapping and associated
analysis to support decision-making in an estuarine Great Lakes AOC (Angradi et al. 2016). R2R =
remediation to restoration; FES = final ecosystem services; BIJI = beneficial use impairment; AOC
= area of concern; SPA = service providing area; SWF = social welfare function (i.e., ecological
benefit function [EBF])
Lessons Learned
The AOC program is highly-collaborative. Engaging all stakeholders improves the likelihood that
information provided to support decisions will be timely, appropriate, and useful. Developing the
conceptual model with a range of stakeholders, including non-government organization, state agencies,
tribal agencies, and federal partners was valuable for creating a conceptual model that was responsive to
the needs of all these organizations.
95
-------
Model development also had substantial added-value to the stakeholders. Although extensive
documentation exists defining AOC goals and describing how they will undertake BUI removal, the
documentation does not address the working relationships between the various organizations. This
process allowed the stakeholders to consider how they wished to operationalize their processes in the
context of the relationships among the various local, state, tribal, and federal partners.
For More Information
Angradi, T.R., D.W. Bolgrien, J.L. Launspach, B.J. Bellinger, M.A. Starry, J.C. Hoffman, M.E. Sierszen, A.S.
Trebitz, and T.P. Hollenhorst. 2016. Mapping ecosystem services of a Great Lakes estuary can support local
decision-making. Journal of Great Lakes Research 42:717-727.
Chan, K.M.A., A.D. Guerry, P. Balvanera, S. Klain, T. Satterfield, X. Basurto, A. Bostrom, R. Chuenpagdee, R.
Gould, B.S. Halpern, N. Hannahs, J. Levine, B. Norton, M. Ruckelshaus, R. Russell, J. Tarn, andU. Woodside.
2012. Where are cultural and social in ecosystem services? A framework for constructive engagement.
Bioscience 62:744-756.
Christensen, N. L., A.M Bartuska, J.H. Brown, S. Carpenter, C. DAntonio, R. Francis, J.F. Franklin, J.A.
MacMahon, R.F. Noss, D.J. Parsons, C.H. Peterson, M.G. Turner, R.G. Woodmansee. 1996. The Report of the
Ecological Society of America Committee on the Scientific Basis for Ecosystem Management. Ecological
Applications, 6: 665-691
GLWQA (Great Lakes Water Quality Agreement). 1987. International Joint Commission, United States and Canada.
Menzel, S. and J. Teng. 2009. Ecosystem services as a stakeholder-driven concept for conservation science.
Conservation Biology 24:907-909.
Portman, M. 2013. Ecosystem services in practice: challenges to real world implementation of ecosystem services
across multiple landscapes - a critical review. Applied Geography 45:185-192.
SLRCAC (St. Louis River Citizens Action Committee). 2002. Lower St. Louis River Habitat Plan. The St. Louis
River Alliance, http://stlouisriver.org/aoc-documents/
St. Louis River Area of Concern (AOC). 2013. St. Louis River Area of Concern Implementation Framework:
Roadmap to Delisting, https://www.pca.state.mn.us/sites/default/files/wq-ws4-02a.pdf
96
-------
Appendix C
Tools and Approaches for Objectives and
Performance Measures
Clarify
Decision
Context
Estimate
Consequences
Define
Objectives
Develop
Alternatives
Evaluate
Trade-offs
and Select
Implement,
Monitor,
and Review
CI: Using DASEES to Develop
Objectives Hierarchies
C2: Applying FEGS-CS to Identify
and Measure Objectives
C3: Using EnviroAtlas to Identify
and Measure FEGS
C4: Rapid Benefits Indicators as
Performance Measures
C5: Applying HWBI to Structure
and Measure Objectives
97
-------
Appendix C1
Using a DASEES Approach to Develop Objectives Hierarchies
Background
A Values Focused Thinking approach places the decision focus on what the stakeholder values are rather
than focusing on the decision alternatives (Keeney 1992). Stakeholder values are what people really care
about and by identifying the stakeholder values it is possible to develop objectives that will play a central
role in guiding all phases of the decision process. Objectives should also guide development of decision
alternatives and the collection and analysis of information needed to evaluate decision alternatives. This
focus on objectives helps to ensure the decision process produces solutions that are transparent in how
stakeholder and decision-makers needs and desires are satisfied.
Approach
The second step in the DASEES (Decision Analysis for Sustainable Economy, Environment, and Society)
Process (Appendix Al) is to Define Objectives. DASEES contains a number of tools to assist decision-
makers in developing objectives hierarchies and identifying measurable attributes to represent objectives.
DASEES can also assist in distinguishing fundamental objectives from means to achieve them.
An Objectives Hierarchy organizes the things stakeholders care about into a set of layers, or a tree of
objectives with broader objectives at the top, called fundamental objectives, with tiered sub-objectives
that refine and provide more specifics or meaning on the fundamental objective. Fundamental objectives
are short phrases that define what is desired. Some examples of fundamental objectives are:
Protect wetlands
Maximize recreational opportunities
Minimize environmental damage
Minimize net cost
In an Objectives Hierarchy, each objective has a specific definition that specifies what is hoped to be
achieved. This process of thinking through and writing down the fundamental objectives and defining
what they mean with measurable attributes goes a long way towards determining what information to
seek. Objectives hierarchies define what is meant by each objective, help explain decisions to others, and
help determine a decision's importance, including how much time and effort it deserves.
Fundamental Objectives
Identifying the fundamental objectives in DASEES is a five step approach. The steps are as follows:
1. Write down all the concerns you hope to address through the decision
2. Convert these concerns into succinct objectives
3. Separate ends (fundamental objectives) from the means (means objectives) to attain those ends.
4. Clarify what is meant by each objective
5. Test the objectives to see if they capture the interests of the stakeholders.
98
-------
Step 1: Understand Context J Step 2: Define Objectives | Step 3: Develop Options j. Step 4: Evaluate Options jl Step 5: Take Action |
i Getting Started
Objectives
Decision makers' and stakeholders' Fundamental Objectives can be added, deleted, and edited in the
Fundamental Objectives tab. Fundamental Objectives are organized hierarchically from general (fundamental) to
more specific. The aim of this refinement is to help with defining Measurable Attributes that indicate achievement of
an Objective and with defining Means Objectives which help in defining Management Options. Under Step 3
Means Objectives are translated to Management Options. Under Step 4 the Impact of Management Options on
Measurable Attributes can be estimated. Measurable Attributes are translated to Ecosystem Services in Step 4.
Objectives Hierarchy
ฃ5) Expand All r] Collapse to First Level ฃ Add & Delete
a _j Improve the quality of life in GBW
a _j Maximize ecological integrity
!51 Maximize ecosystem connectivity and linkages
_J Maximize the ecological integrity of terrestrial habitats
_j Maximize the ecological integrity of freshwater habitats
l_l Maximize the ecological integrity of estuarine habitats
a Maximize the ecological integrity of marine aquatic habitats
IS1 Maximize the integrity of open ocean habitats
S Maximize the integrity of coral reef ecosystems
i~~l Maximize economic benefit
f 1 Maximize social well-being
i _ Minimize threats to human health
H Maximize learning opportunities
a j Meet political and legislative requirements
jifcl Meet Coastal Zone Management Act regulations
a , Meet Clean Water Act regulations
32 Meet dissolved oxygen standard
__ Measurable Attributes
ฃ Add Delete . " Edit Attribute
Attribute Category
Stony Coral Colony Size State
H"1 Means Objectives
^ Add ^ Delete . * Edit Objective
Objective
Minimize sediment load from agriculture
Minimize total nitrogen load from WWTP
Minimize impact of tourism on corai reefs
igure Cl.l. Example of developing an objectives hierarchy for Guanica Bay watershed
management (see Appendix Al) to protect coral reefs (original from Carriger et al. 2013).
Step 1: Ask the stakeholders what is important to them. There are no boundaries at this time on what a
stakeholder can add to the list. It is important to capture every idea that is said by the stakeholders so
having multiple people recording what is said on large notepads that all stakeholders can see is very
important. Make sure everybody can speak and say what is important to them. No suggestion is out of
bounds at this stage if it is important to a stakeholder. Be thorough in your recording. Make sure
everybody has a chance to add to the list even if that means asking the quiet and shy stakeholders to
specifically comment.
Step 2: Once you have the list from the stakeholders start to convert their concerns and ideas into
succinct statements (objectives). These objectives need to be simple in their statement so they are easily
remembered and the end position(s) of the overall effort is clear. Long statements tend to be confusing for
stakeholders whereas short succinct statements are more easily prioritized.
Step 3: Through this process it will start to become clear that some of what have been proposed as
fundamental objectives (desired ends) are in fact means objectives (approach options to achieve desired
ends). Means objectives are not part of the fundamental objectives hierarchy as they are captured later on
in the process when identifying how to achieve the fundamental objectives. It is important to identify and
exclude means objectives at this point in the fundamental objectives hierarchy since means objectives
typically correspond to a stakeholders' position of how something should be achieved. If means
objectives are confused with fundamental objectives, it can have the effect of cementing positions from
99
-------
one set of stakeholders, thereby giving the impression that there is already a preferred approach to
achieving the ends and ultimately hampering the consensus-building process.
Step 4: This is where you want to ask the stakeholders why that issue or statement is important. As you
are developing these succinct objective statements ask "what do you mean by that" to help provide
exactly what is meant by the objectives. This question will help to identify sub-objectives that clarify
what is meant by the overall fundamental objective. Depending on the decision context any fundamental
objective may have different interpretations depending on the background of the stakeholder. For
example, "protect open space" would likely be quite different for a hiker versus a golfer. Each sub-
objective should have a measure associated with it that helps determine if that sub-objective is being
accomplished.
Step 5: Once you have developed a list of fundamental objectives and ensured there are no means
objectives on the list, take a step back and determine if this final list has sufficiently captured the values
of the stakeholders. If the consensus is yes, then the SDM process can proceed to the defining measurable
attributes and measures preference for each fundamental objective. If the stakeholders are not satisfied
with the list, look to determine what might be missing and include those fundamental objectives in the
final list.
Measurable Attributes
Once the Fundamental Objectives are organized in a hierarchical manner, there has to be a way to
determine if the attainment of those fundamental objectives is being achieved. Developing measurable
attributes that specifically show if the fundamental objective is being achieved is key to the success of the
SDM process. Measurable attributes need to be meaningful, measurable, operational and interpretable.
Without an effective means of measuring progress and success in attaining the project goals, the entire
effort is moot.
Take Home Message
With the fundamental objectives, measurable attributes, and measure preferences defined, it is necessary
to determine how to go about achieving means and decision options to achieve them (Appendix Dl). The
development of an Objectives Hierarchy is critical to taking what stakeholders value and translating it
into clear endpoints they want to achieve. By going through this process it helps the stakeholders to
separate out fundamental objectives (the ends they want to achieve) and the means objectives (the things
they do to achieve the ends). Without developing an objectives hierarchy, it is difficult to keep
fundamental objectives and means objectives straight, making it much harder to make a decision that
actually achieves what is desired.
For More Information
Carriger, J.F., W.S. Fisher, T.B. Stockton, and P.E. Sturm. 2013. Advancing the Guanica Bay (Puerto Rico)
Watershed Management Plan. Coastal Management 41(1): 19-38.
Keeney, R. 1992 Value-Focused Thinking: A Path to Creative Decision Making. Cambridge, MA: Harvard
University Press.
100
-------
Appendix C2
Applying the FEGS-CS to Identify and Measure
Ecosystem Services Objectives
Background
The City of Duluth, Minnesota is creating a new vision for the St. Louis River Corridor, or the
neighborhoods closest to the river. These neighborhoods were once home to wood and steel mills,
foundries, and coal tar plants. The City is trying to capitalize on Areas of Concern sediment remediation
and habitat restoration, attract new families and development, but also repeat the success of another
visitor-destination neighborhood in town.
The expectation is that enhanced trails and access to the river will spur the development of small shops,
restaurants, and related businesses. The City of Duluth demonstrated considerable commitment to
implementing this new vision through creating a new tax to secure bonding to fund the effort. On the
other hand, residents who live in the St. Louis River Corridor have been concerned that the City has not
adequately included them in planning.
Example: Community opposition and trail planning
Citizens responded to the City efforts by creating a new community organization, comprised of residents
representing different neighborhoods throughout western Duluth. Citizens were concerned that the City
was developing parks and trails to attract visitors, while ignoring housing conditions, failing to market the
neighborhoods, and ignoring existing parks.
Community
Organization
Marketing
Housing
Figure C2.1. The structure of the new community organization in Duluth.
The new community organization met monthly and had three committees: Housing, Marketing, and Parks
and Trails (Fig. C2.1). The Parks and Trails Committee was the most active, in part, because of the
amount of park planning activity. The group organized a significant amount of opposition to a perceived
lack of citizen input in proposed projects. The City responded with more opportunities for public input,
including a survey about a particularly contentious trail project.
Using FEGS-CS and HWBI to uncover neighborhood values
Two tools useful in uncovering neighborhood values are the Human Well-being Index (HWBI) and the
Final Ecosystem Goods and Services Classification System (FEGS-CS). FEGS-CS can be used to identify
beneficiaries of environmental goods and services in a particular type of environment (Chapter 2, Table
2.2; Landers and Nahlik, 2013). HWBI is a composite of elements or domains that contribute to our well-
being, including connection to nature, health, social cohesion, and cultural fulfillment (HWBI; Smith et
101
-------
al. 2013, 2014; Appendix D2). We applied the tools to the analysis of a particular data set. The City of
Duluth conducted a survey with the open ended question, "Do you have any comments on the project's
vision?"
We used conventional content analysis of the answers to the open ended question to discern what citizens
valued most about the trail (Hsieh and Shannon, 2005). An analysis of the emergent categories (Fig.
C2.2) shows that the values of citizens align with ecosystem services such as engagement with nature,
and elements of the HWBI such as safety. In particular: safety, existing trail, experience, transportation,
scenic beauty, connection to nature, and connection to other parts of the community.
Project Vision Comments
Figure C2.2. Revealed values related to a trail project.
Results
The most significant considerations for citizens were:
Safety (similar to HWBI): The citizens felt that a trail should provide an off AND away
from the road experience with minimal traffic conflicts.
Munger Trail: The citizens thought that the trail should connect the proposed trail to the
current trail head for a regional multi-use trail.
Experience (similar to EEGS-CS): The respondents supported the City's vision for a
contiguous, non-motorized, multi-use trail and felt that the trail route closer to the river
provided a unique-one of a kind experience.
102
-------
Several of the emergent themes reflected an attachment to ecosystem services, especially aesthetics,
viewscapes, or sounds and scents. Others appreciated the economic opportunities (like outdoor recreation
businesses), as well as the appreciation that trails connect communities, and connect people to nature.
We found that the route that followed an abandoned rail bed closer to the river was the preferred
alignment. This trail would provide off-road experience with the fewest number of traffic conflicts, a
connection to a popular regional trail, and a desirable experience.
Advice for New Users
We started with conventional content analysis, which allowed us to uncover what the participants valued.
The values are the categories on the chart. The open-ended nature of the analysis allowed us to capture
the breadth of values including community connections (similar to social cohesion) and sense of place
(similar to identity). Using FEGS-CS and HWBI as a second layer of analysis revealed what elements
contributed to each value.
For More Information
Hsieh, H.F., and S.E. Shannon. 2005. Three approaches to qualitative content analysis. Qualitative health research,
15(9): 1277-1288.
Landers, D.H., and A.M. Nahlik. 2013. Final Ecosystem Goods and Services Classification System (FEGS-CS).
EPA/600/R-13/ORD-004914. U.S. Environmental Protection Agency, Office of Research and Development,
Washington, D.C.
Smith, L.M., J.L. Case, H.M Smith, L.C. Harwell, J.K. Summers. 2013. Relating ecosystem services to domains of
human well-being: Foundation for a US index. Ecological Indicators 28:79-90.
Smith, L., C. Wade, K. Straub, L. Harwell, J. Case, M. Harwell, K. Summers. 2014. Indicators and Methods for
Evaluating Economic, Ecosystem and Social Services Provisioning: A Human Well-being Index (HWBI)
Research Product. U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-14/184.
103
-------
Appendix C3
Using the EnviroAtlas to Identify and Map
Potential Measures of FEGS
Background
Ecosystems benefit human well-being and economies in myriad ways, but systematic accounting of these
benefits remains beyond our ability. EnviroAtlas, is a web-based, geospatially-explicit collection of tools,
data, and resources centered around the concept of ecosystem services (http://www.epa.gov/enviroatlas;
Pickard et al. 2015). The Final Ecosystem Goods and Services (FEGS), provides a comprehensive
classification system (FEGS-CS) that could allow the ecosystem services data in the EnviroAtlas to be
tied to specific beneficiaries. FEGS-CS is a leaping off point for estimating the economic and social value
of interactions between human communities and the natural environment (Landers and Nahlik 2013). This
is accomplished in part by focusing on human beneficiaries and the specific Final Ecosystem Goods and
Services they directly use, enjoy, or experience. This methodology allows for the explicit definition, and
therefore accounting, of all ecosystem services directly relevant to human well-being and economies. Our
goal is to map data layers developed and hosted by EnviroAtlas onto the FEGS-CS, and to explore the
EnviroAtlas mapping application as a tool for creating a geospatially explicit representation of FEGS.
Approach
The FEGS-CS provides a six-digit code to uniquely identify FEGS (see Figure C3.1). Where data layers
in the EnviroAtlas do directly indicate potential FEGS, we simply cross referenced the FEGS-CS and
applied the relevant code to the data layer in question. However, the majority of the data layers in the
EnviroAtlas describe ecosystem services that are not directly tied to human beneficiaries, meaning those
layers are not FEGS.
FEGS Classification Structure
X
Environmental Class
XX.
Environmental Sub-Class
xx.xx
Beneficiary Category
xx.xxxx
Beneficiary Sub-Category
Figure C3.1. General FEGS classification structure, from Landers and Nahlik 2013.
Instead, many data layers are metrics or indicators of intermediate ecosystem goods and services (IEGS),
i.e., aspects of the natural world that require additional information to directly link to human benefit. For
example, pollinator habitat is not directly enjoyed by humans, but the pollination of crops - which does
directly benefit farmers - is dependent on the existence of habitat. In such cases, we applied a conceptual
production function to identify the potential FEGS most directly dependent upon or affected by the IEGS
in question (see Figure C3.2).
104
-------
Ecological Production
Function Labor and Capital
I J
IEGS
hbGS
HGS
Figure C3.2. Conceptual model showing production function linkages between IEGS, FEGS, and
HGS.
In other cases, data layers in the EnviroAtlas are metrics or indicators of ecological human goods and
services (HGS), i.e. goods and services directly affected by the natural world but also dependent to some
degree on human intervention. For example, fruit crop yields are dependent on human labor and capital,
but are also directly affected by the presence of pollinators and depredators, the quantity and quality of
water available for irrigation, and soil conditions, among other factors. In these cases, we applied a
conceptual production function to work backwards from the HGS in question to find the FEGS on which
it most depends.
FEGS - EnviroAtlas Crosswalk
By this methodology, we were able to identify over 14,000 metrics of IEGS, HGS, and potential FEGS.
These metrics were catalogued into a readily searchable database, which has been used to help identify
data layers in the EnviroAtlas relevant to the development of metrics and indicators of FEGS. We have
also analyzed these metrics in relation to the broader scope of beneficiary sub-categories in the FEGS-CS
to identify both beneficiaries who are especially well served by the EnviroAtlas, and beneficiaries for
whom it is necessary to begin developing targeted data sets. And we have begun to assess the ability of
the EnviroAtlas mapping applications to aggregate beneficiary sub-categories and environmental sub-
classes in a given study area for the development of geospatially explicit FEGS profiles (see Figure
C3.3).
FEGS Environmental Sub-Classes
FEGS Environmental Sub-Class Profile
~ Forests
~Agroecosystems
~ Barren
ฆ Rivers and Stream;
~ Lafcesand Ponds
~ Wetlands
~ Grasslands
Figure C3.3. An example of how the FEGS Environmental Sub-Classes in a study area could be
mapped and aggregated into a profile. From Landers and Nahlik, 2013.
105
-------
Transferability of Approach
We have shown that it is possible to crosswalk the FEGS-CS with a large database designed without the
PEGS approach in mind. Government agencies at the federal, state, and local levels have developed an
extraordinary amount of data available at various scales. However, the ability to discover beneficiary
specific data is currently entirely dependent on the knowledge of the potential user. Therefore, much of
the data relevant to user needs goes unused and under-appreciated. The option to plug extant data from
these various sources into the PEGS approach allows the opportunity to begin to assimilate available
ecosystems services data into an internally consistent, readily searchable database. As the ecosystem
services community - and the data available to them - continues to grow, the ability to search for and
identify data relevant to one's needs should prove to be valuable.
For More Information
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.
Pickard, B.R., J. Daniel, M. Mehaffey, L.E. Jackson, A. Neale. 2015. EnviroAtlas: A new geospatial tool to foster
ecosystem services science and resource management. Ecosystem Services 14:45-55
-------
Appendix C4
Rapid Benefits Indicators (RBI) as Performance Measures
Background
The RBI approach is an easy-to-use process for assessment based on non-monetary benefit indicators. It is
intended to be used in conjunction with existing ecosystem service assessment approaches and tools to
connect changes in the availability of Ecosystem Goods and Services (EGS) to where and how people
benefit from those goods and services. The RBI approach was originally developed for use with urban
freshwater wetland restoration, but the general approach and indicator framework can be adapted to work
with other types of environmental changes or within different ecological systems, and to focus on PEGS
by tying endpoints more explicitly to beneficiaries.
Approach
The approach is a five step process, starting off with identifying stakeholders and their objectives. Based
on these objectives, relevant ecosystem services and resulting benefits are selected to compile indicators
based on five questions:
1. Can people benefit from an ecosystem service, including whether there is demand, sufficient
quality and quantity, and necessary complementary inputs?
2. How many people benefit?
3. How much are people likely to benefit?
a. What is the quality of the service?
b. Are there substitutes for the service or is the service scarce?
c. What is the quality of complementary services?
d. How strong are people's preferences?
4. What are the social equity implications?
5. How reliable are benefits expected to be over time?
Indicators for five benefits (Figure C4.1) have been previously developed and integrated into two tools
that help users more easily apply the Rapid Benefit Indicators approach.
Ecosystem Service How people benefit
Reduced Flood Risk: The risks from
Flood water regulation floods to people and structures are
reduced.
Is!
1
. . , Scenic Views: People can enjoy
Scenic landscapes
scenic views.
I
Si1
Environmental Education: People
Learning opportunities can benefit from studying nature or
from enhanced connection to nature.
Recreation: People can enjoy
Recreational opportunities
recreation
-
Bird Watching: People can watch
Blrt,S U i.i
or hear birds.
Figure C4.1. Benefits previously assessed using Rapid Benefit Indicators (Mazzotta et al. 2016).
107
-------
The Rapid Benefit Indicators (RBI) checklist tool can be used for recording results of manual or field
analysis, while the Rapid Benefit Indicators (RBI) spatial analysis tools can be used to compile indicators
based on spatial datasets.
Example: Performance Measures for Restoration on Tampa Bay Conservation Lands
For this example, we assessed the five benefits provided by wetlands restoration on Hillsborough County
Environmental Lands Acquisition and Protection Program (ELAPP) 2011 land holdings north of Tampa
Bay. This assessment was done solely to demonstrate the application of Rapid Benefit Indicators. An
actual application requires more in-depth restoration feasibility assessment, stakeholder engagement, and
consideration of decision context when choosing relevant benefits and restoration sites to consider.
Determining if people benefit from an Ecosystem Good or Service (EGS) is intended as an initial
screening step. It requires that the following three criteria are met: (1) a good or service is produced, (2)
there are people who want those benefits, (3) complementary inputs required for benefits to be received
are available.
Potential wetland restoration sites were generated using areas classified as having high potential for
restoration in the Potentially Restorable Wetlands on Agricultural Land (PRWA) dataset from the
EnviroAtlas (Pickard et al. 2015). Areas with this classification had drainage and wetness characteristics
to allow for wetland restoration, and, once restored, production of EGS. In addition, these areas present
fewer land-use conflicts because they also had to be classified as agricultural. Restoration sites were
further reduced to avoid double counting of existing benefits by removing portions that overlap with
existing wetlands, as defined by the National Wetlands Inventory (NWI; USFW 2017).
Service
is produced
Figure C4.2. Benefits can be received where EGS are produced (left) when EGS flow out to people
from where they are produced (middle) or when EGS flow through specific pathways to reach
people.
Determining how many people benefit from each EGS involves investigating the ways those EGS flow to
people (Figure C4.2) and quantifying the number of people who could benefit.
How much people benefit from an EGS depends on the quality of the EGS, how much of that EGS is
already available or available through alternative substitutes, the quality of any complementary inputs,
and how much the people who benefit care about those benefits.
108
-------
People benefit more from higher quality EGS than lower. The quality of EGS can be determined through
the indicators developed in the RBI approach, or through a variety of existing alternative EGS
assessments (e.g., those in Appendix C3, El, E2).
For a given beneficiary, the more alternative sources of an EGS that exist, the lower the benefit of
additional sources created through restoration (Figure C4.3). For wetlands restoration, alternative sources
are typically existing wetlands or other green spaces able to provide the same EGS. For indicators, areas
are characterized for each restoration site rather than characterizing areas for each beneficiary, using
double the distance of the EGS delivery to ensure that even if a beneficiary is at the outer extent its
alternative sources of EGS are considered.
Figure C4.3. Where a benefit is available (blue area), beneficiaries (houses) benefit less from that
service (green overlap area) if additional sources of the service are available (yellow area).
The quality of complementary inputsthings that get consumed at the same time as an EGS-can help to
differentiate between benefits of two alternative sites. For example, if in Figure C4.3 the beneficiary in
the green area is choosing between wetlands to go recreate, higher quality trails through the wetland
surrounded by the yellow area might make the benefits in the yellow area better.
Strengths of preferences are the characteristics of the beneficiaries that give benefits from a site more or
less value. For example, schools with curriculums that include wetlands will have stronger preferences
for, and place greater value on, a wetland with environmental education benefits than a school without
such curriculum or requirements. Indicators for preferences are more specific to the local context, but can
often be gleaned from demonstrated interest and/or stakeholder engagement.
Examining social equity implications explores if beneficiaries are part of a particularly vulnerable
population or if they face concerns about environmental justice. For example, populations with fewer
resources to deal with flood events benefit more from flood risk reduction benefits.
The longer EGS are expected to continue providing benefits into the future, the higher the total value of
those benefits. For example, a site that is expected to reliably provide benefits for the next 10 years will
provide more benefits than one that is expected to be drained to plant crops in five years. Since all
potential restoration sites were already located on conservation lands, reliability of benefits would have
required a more in-depth investigation of wetland stressors and was not assessed.
109
-------
Lessons Learned
The example assessment outlined here was a demonstration only. An actual application requires more in-
depth stakeholder engagement and consideration of decision context when choosing relevant benefits and
restoration sites to consider. Results of the example application are explored more in Appendix F4.
Rapid Benefit Indicators (RBI) are a viable option for measuring performance of alternatives based on the
people who benefit from EGS. The RBI approach has the flexibility to integrate results of other types of
analysis and be more or less involved depending on the communities' needs. The example indicators
available are able to be used on potential wetlands restoration sites, meaning performance indicators can
be evaluated a priori, before actions are implemented. The RBI approach can be followed without using
the example indicators, alternatively developing indicators that better fit the decision context, data
availability or ecological system, but still following the RBI approach.
For More Information
Mazzotta, M. J. Bousquin, C. Ojo, K. Hychka, C.G. Druschke, W. Berry, and R. McKinney. 2016. Assessing the
Benefits of Wetland Restoration: A Rapid Benefit Indicators Approach for Decision Makers. U.S.
Environmental Protection Agency, Narragansett, RI, EPA/600/R-16/084.
Pickard, B. R., Daniel, J., Mehaffey, M., Jackson, L. E., & Neale, A. 2015. EnviroAtlas: A new geospatial tool to
foster ecosystem services science and resource management. Ecosystem Services, 14, 45-55
US Fish and Wildlife Service. 2017. National Wetlands Inventory website. US Department of the Interior, Fish and
Wildlife Service, Washington, D.C. http://www.fws.gov/wetlands
Table C4.1. Datasets used and their source in the example RBI analysis.
Parameter
Source
Link
Population Raster
EnviroAtlas National Dasymetric data
www.epa.gov/enviroatlas
Flood Zone
FEMA National Flood Hazard Layer
https://msc.fema.gov
Dams/Levees
FEMA National Flood Hazard Layer
https://msc.fema.gov
Educational Institutions
Homeland Infrastructure Foundation Public Schools
http: //hifld-dhs -
gii. opendata. arcgis .com
Bus Stops
Open Street Maps
www.openstreetmap.org
Trails
Open Street Maps
www.openstreetmap.org
Roads
Open Street Maps
www.openstreetmap.org
Wetland Polygons
US FWS National Wetlands Inventory
www. fws. go v/wetlands
Land-use/Greenspace
EnviroAtlas Communities Land-use data
www. epa. gov/enviroatlas
Social Vulnerability
CDC Social Vulnerability Index
https://svi.cdc.gov/
110
-------
Appendix C5
Applying the HWBI Framework to Structure and Measure
Community Objectives
Background
The application of systems thinking to community decision-making is greatly enhanced by a careful
analysis of the desired community endpoints. These endpoints are usually tied directly to a particular
focal issue (e.g., brownfield restoration endpoint is successful repurposing of the target sites), but
application of a systems approach transcends particular focal issues and so must the desired endpoints.
Yet, new, broader endpoints must also be accepted and understood by stakeholders prior to making the
decisions. This requires a careful examination of stakeholder interests, as well as stakeholder involvement
in the development of new endpoints.
Approach
Structured decision-making (SDM, Gregory et al. 2012) is a well-established technique for organizing
stakeholder thinking around a set of fundamental objectives. Fundamental objectives are defined as those
objectives that do not change from issue to issue and so represent a core set of community targets for
measuring value of different decision options. Systems thinking requires these fundamental objectives be
defined across a broad set of stakeholder interests, which can be confusing, but the concept of human
well-being can be used to organize these objectives into meaningful groups so they can be selectively
applied to particular issues of interest (e.g., downtown redevelopment vs. health and wellness initiatives).
Figure C5.1. Community participants at workshop.
The Human Well-being Index (HWBI, Smith et al. 2013) was recently applied to guide self-identification
of stakeholder fundamental objectives at the community level (Fulford et al. 2016). A series of SDM
workshops were held in nine communities across the U.S. to identify community-specific fundamental
objectives and gauge their relative importance to human well-being (Fig, C5.1). This process is described
in more detail in Appendix F2. Fundamental objectives identified during the workshops were organized
based on the eight domains of the HWBI (Fulford et al. 2015; Fulford et al. 2016). The domains were
then used to define how these fundamental objectives contributed to community well-being. The domains
were also ranked in several different ways to assess their relative importance (Fig. C5.2). Each
111
-------
community applied their results to a particular focal issue, and the collective results across all
communities were used to assess commonalities and to evaluate HWBI as a tool for goal setting at the
community level. The outcome was a transferable approach for identifying community-specific objectives
that can be used to form system-level measures of success for community decision-making.
a rv,e
tlnbaaV
bosic
Figure C5.2. Flip chart illustrating use of colored dots to prioritize objectives.
Example case study: Vero Beach, FL
Vero Beach, FL is a small residential and tourist community on the eastern coast of Florida about 80
miles south of Cape Canaveral, FL. The community was approached about hosting an SDM workshop
and decided to organize it around the focal issue of redeveloping downtown as an arts destination. The
initial phase of the workshop was planning who should attend. The attendees should be a representative
group of community stakeholders and an open meeting format is not optimal for this purpose. A planning
group established a list of stakeholder groups in Vero Beach including churches, citizen groups, business,
education, local government, and the arts community. This list was used to identify representative leaders
and make sure all interests were represented at the workshop.
The workshop was held February 2015 and was attended by 32 people not counting the organizing
committee. It was held at a centrally located, easy to find, neutral location. It was also desired to choose a
workshop location not associated with any particular stakeholder group so as to not inadvertently
influence the outcome. The workshop lasted six hours and was separated into two parts: Development of
community fundamental objectives and application of these objectives to the focal issue. Workshop
outcomes included identification of fundamental objectives, linking of these objectives to domains of
HWBI, and ranking of these domains for Vero Beach.
Vero Beach mapped most of their fundamental objectives to Social Cohesion and Health and the least to
Education. In contrast a ranking of well-being domains placed Education among the most important for
workshop participants. This exercise highlighted the gap between how the community prioritized actions
and broader community priorities, and this gap was a key subject of the discussion during the second part
of the workshop. Attendees considered how downtown development for the arts could support
community education and social cohesion in addition to assisting downtown residents (Fig. C5.3). Such
112
-------
discussions, in the context of specific decisions, help communities better align resources and choices with
all of their priorities as opposed to just the objectives most obviously impacted by a decision.
Figure C5.3. Community participants at workshop.
Lessons Learned
Each workshop in the SDM series was different just as every community has different qualities and focal
issues. The value of the structured decision-making approach was two-fold. First to highlight the
importance of defining fundamental objectives and achieving consensus on how to measure the impact of
decisions on human well-being. Second was for stakeholders to consider similarities and differences
between current decision-making and a systems approach focused on human well-being. This is a lot of
ground to cover in a six-hour workshop so the outcome was only a beginning. Stakeholders expressed a
good understanding of human well-being and how it is connected to their priorities but did not express a
lot of understanding for how to use this information or how it might change "business as usual" for
decision-making. However, this is a step in a longer process and these workshops highlight the need to
engage stakeholders early in the decision process and to gain a consensus on measures of success clearly
tied to community objectives. The challenge going forward is to make use of these lessons to develop
practical strategies for systems-based decision-making.
For More Information
Fulford, R.S., M. Russell, J. Harvey, and M.C. Harwell. 2016. Sustainability at the community level: searching for
common ground as a part of a national strategy for decision support. Gulf Breeze, FL: US EPA EPA/600/R-
Fulford, R.S., L.M. Smith, M. Harwell, D. Dantin, M. Russell, and J. Harvey. 2015. Human well-being differs by
community type: Toward reference points in a human well-being indicator useful for decision support.
Ecological Indicators 56: 194-204.
Gregory, R., L. Failing, M. Harstone, G. Long, T. McDaniels, and D. Ohlson. 2012. Structured decision-making: A
practical guide to environmental management choices, London, UK: Wiley-Blackwell.
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.
16/178.
113
-------
Appendix D
Tools and Approaches for Creating
Alternatives
Clarify
Decision
Context
Estimate
Consequences
Define
Objectives
Develop
Alternatives
Evaluate
Trade-offs
and Select
Implement,
Monitor,
and Review
D2: Using Networks to Link Decision
Alternatives to FEGS and Well-being
Dl: Developing Means-Ends
Objectives Hierarchies in DASEES
114
-------
Appendix D1
Developing Alternatives using Means-Ends Objectives
Hierarchies in a DASEES Approach
Background
Having a clear set of fundamental objectives is essential in order to meet any project goal. But once you
have these fundamental objectives, options need to be developed regarding how you will achieve these
fundamental objectives. Many times one or two options to achieve an end goal are presented as if they are
the only available options. This is often not the case. A range of management alternatives or approaches
need to be developed that reflect as many different approaches to address the achievement of the
fundamental objectives as the stakeholders can articulate. These approaches are typically based on value
perspectives of the stakeholders. It is necessary to develop means objectives (i.e. what might actually be
done) to help identify how to go about achieving the fundamental objectives. Without identifying what
options and approaches are available to implement, it is impossible to achieve the fundamental objectives
in a clear and transparent manner.
Approach
Developing means objectives that describe ways to achieve fundamental objectives and developing
options or alternatives from the means objectives is a part of a values-focused thinking approach. Values-
focused thinking is a bottom up approach where decision alternatives are clearly derived from the
decision context and objectives, which directly emanate from stakeholder values and preferences. Instead
of having a couple of top down driven options, this bottom up approach will routinely produce a greater
number of viable alternatives for consideration by the stakeholders. Good alternatives have the following
characteristics (Gregory et al. 2012):
Complete and directly comparable - Do alternatives have the same scope? Are apples to
apples being compared?
Value-focused - Do they address the fundamental objectives?
Fully specified - Is what is meant by each alternative clear and unambiguous?
Internally coherent - Do they make sense and work together?
Distinct - Are the alternatives really different from one another?
When alternatives have these characteristics they can be rigorously compared and assessed.
Means Objectives
With fundamental objectives, measurable attributes, and preferences defined (Appendix CI), it is
necessary to determine how to go about achieving the fundamental objectives. This is where means
objectives come into play. Each of the final sub-objectives should have an associated means objective.
The means objective helps identify how to achieve fundamental objectives. As with the development of
fundamental objectives, the development of means objectives should follow the same open process. More
time than not there are multiple ways (means) to achieve an end goal. Stakeholders will often provide
multiple options to achieve a sub-objective. All may have viability but through comparison certain
approaches will come to be favored because of factors (e.g., ease of implementation, cost of application,
shorter time frame for measurable progress) deemed important by the stakeholders. Sometimes the means
objective is clear and straightforward such as when it is mandated by a regulatory policy or law. Means
objectives point to options on how to achieve fundamental objectives and ultimately lead to identifying
management options or alternatives for the stakeholders to consider and implement.
115
-------
The third step of a DASEES (Decision Analysis for Sustainable Economy, Environment, and Society)
process (Appendix Al) uses the development of means objectives to create the decision options or
alternatives by which to achieve the fundamental objectives. Each means objective should have a clear
link to the fundamental objective identified in the Objectives Hierarchy (Appendix CI). Means
objectives are created in the Means-Ends Objectives Hierarchy (left-side of figure; Fig. Dl.l).
Define Options
Means - Ends Objectives Hierarchy
Management Options
| ฃ2 Delete Means Objective
New Option Detete Option
- Minimize Sediment Impacts on SustainaWity
Option
Units
- Maxhmixe Power Genera two
O Maximize reservlor volume
. j/] Dredge reservoir
cubic meters
Dredge Lucchetti Reservoir
Mm3
- Minimize Coral Reef Impacts
O Minimize sediment load
Q NPDES TN Limit
Riparian Buffers
Subsidize shade grown coffee
lbs/day
ha
ha
Figure Dl.l. Example Means-Ends Network in DASEES illustrating how a specific management
option (dredging the reservoir) can be identified to accomplish a means objective (maximizing
reservoir volume) in support of a fundamental objective (maximizing power generation).
Management Options
Management options (right side of figure; Fig. Dl.l) are developed that identify the actions to be taken as
part of the means objective to achieve the desired fundamental objective. For example, an acceptable
means to reduce downstream flooding of neighborhoods below a reservoir might be to increase reservoir
volume. Actions to achieve that might be dredge the reservoir or raise the dam height.
Once the decision options are developed, the stakeholders can then develop potential management
scenarios that utilize collections of the established decision options (Fig. D1.2). These management
scenarios are where stakeholders will start to assign actual values to the options they defined earlier.
Having a broad variety of scenarios to consider provides the most opportunities to consider potential
tradeoffs.
(Jl ManagementScenarios
Save ฃ Revert
Management Scenarios
Management Options
Scenarios
Dredge reservoir
New Deteteป
Editป
Dupfecste *
Expand/Col lapse Options
Subsidize shade grown coffee Status Quo ซ Selective Implementation ซ Dredge ซ
Dredge reservoir ฃ X
Dredge reservoir l*J X
Dredge reservoir Ifcl X
0 cubic meters
1000 cubic meters
4000000 cubic meters
Subsidize shade grown coffee ซ X
Subsidize shade grown coffee 55 X
Subsidize shade grown coffee ฃ X
0 ha
100 ha
0 ha
Figure D1.2. Specification of decision scenarios in DASEES as different combinations of
management options.
116
-------
Take Home Message
Clearly identify ends goals that you are trying to achieve and then look to develop as many ways of
accomplishing those end goals as possible. Not all options will be incorporated into the final management
plan or approach. But by fully developing the list of potential options to achieve the end goals, the overall
completeness and utility of the final approach will be much stronger.
For More Information
Gregory, R., L. Failing, M. Harstone, G. Long, T. McDaniels, D. Ohlson. 2012. Structured Decision Making: A
Practical Guide to Environmental Management Choices. John Wiley & Sons
117
-------
Appendix D2
Using Networks to Link Decision Alternatives to FEGS and
Human Well-being
Background
A systems approach to decision-making requires some understanding of how a broad array of issues are
interrelated. This includes how different decisions (e.g. downtown redevelopment or community health
initiatives) are connected, but also how these different decisions may mutually impact ecosystem services
and human well-being. Moreover, these connections must be well-understood by the community
stakeholders involved in decision-making. Here we highlight an interactive visualization tool intended to
assist stakeholders in viewing and understanding the collective relationships between decision
alternatives, ecosystem service production, and human well-being.
Approach
The Eco-wellbeing relationship browser is an interactive tool based on network concepts and intended to
clarify how decisions are connected to domains of human well-being (Fig. D2.1) in a specific community.
These connections are based on both scientific data (Smith et al. 2013) and community input (Fulford et
al. 2016) regarding the relative importance of various ecosystem services to the community.
HWBI
Leisure
Time
Connection
to nature
Cultural
Fulfillment
Safety and
Security
Education
Health
Living
Standards
Social
Cohesion
Figure D2.1. Diagram describing the eight domains of the Human Well-being Index (HWBI; Smith
et al. 2013).
Network-based tools have been used extensively to organize systems of complex relationships and the
interactive portion of the tool allows for a multidimensional network to be visualized all at once, rather
than as a set of separate two-dimensional pieces (i.e. graphs). Such a "birds-eye view" is critical to
communicating systems thinking to community stakeholders in a way that is relevant to the decisions at
hand in a particular community. The visualization can be used to consider the relative impact of decision
options on high priority well-being endpoints (e.g., economic vs. social endpoints), and can be mated to
network-based indices of sustainability allowing for a comparison of suites of decisions that take into
account the characteristics of the specific community making the decisions.
Example
In an example application, Community X is considering multiple options for downtown development that
vary in emphasis between economic enhancement, and social factors such as investment in greenspace
and urban housing. Through a series of stakeholder meetings, they break down their objectives to a suite
of decision options including alternatives such as tax incentives, zoning alterations, investments in
infrastructure such as parks and utilities, and marketing campaigns to attract new businesses and
residents. The stakeholders can use the network tool (Fig. D2.2) to map each decision option (i.e., build
parks) to FEGS production (Greenspace) and then to domains of well-being (Social Cohesion). They can
118
-------
then compare the projected impacts of each decision to stakeholder priorities to assure that actions taken
are those most likely to achieve stated goals for well-being. Finally, they can use these linkages to
establish measures of success tied to well-being endpoints to track how well implemented decision
options fulfill community expectations. The network tool allows for assessment to be done
simultaneously across a full suite of decisions so that the collective outcome can also be evaluated.
Air Quality
Communication
Capital
Investment
Water Quality
Community
and Faith-
based
Consumption
Public works
Social
Cohesion
Education
Services
Production
Labor
Employment
Family
Services
Healthcare
Justice
Greenspace
Figure D2.2. Diagram showing linkages between decisions, ecosystem services, and domains of
human well-being (from HWBI). Example from text of investing in Greenspace is linked to HWBI
via the Social Cohesion domain. This is not the only connection between Greenspace and HWBI
and each one can be visualized in a similar way in the interactive tool.
Lessons Learned
The network tool described here is intended to be both a planning tool and a tool for educating
stakeholders about the value of a systems-approach to decision-making. Multiple decisions are made by
communities, and the overall well-being of the community and its citizens is best served if these decisions
are not made in isolation. Systems thinking enhances understanding of how these decisions are
interrelated and how they collectively impact human well-being. Future work will involve integration of
the network tool with other planning tools such as ENVISION, as well as the development of measures of
community sustainability based on network structure. This research will be conducted in cooperation with
community stakeholders to allow for transferability among different community types.
For More Information
Fulford, R.S., M. Russell, J. Harvey, M.Q Harwell. 2016. Sustainability at the community level: searching for
common ground as a part of a national strategy for decision support. US EPA, EPA/600/R-16/178.
Smith, L.M., J.L. Case, H.M. Smith, L.C. Harwell, J.K. Summers. 2013. Relating ecosystem services to domains of
human well-being: Foundation for a US index. Ecological Indicators 28: 79-90.
119
-------
Appendix E
Tools and Approaches for Estimating
Consequences
Clarify
Decision
Context
Implement,
Monitor,
and Review
Define
Objectives
Eva uate
Develop
Alternatives
Trade-offs
and Select
Estimate
Consequences
El: Ecosystem Services
Production Functions (EPFs)
E2: Modeling Ecosystem
Services Production using SPA
E3: Using the Modeling Tool VELMA
E4: The EcoService Models Library (ESML)
E5: Non-Market Economic Valuation
E6: Measuring Health Outcomes
E7: Health Impact Assessment
E8: Consequences of Ecosystem Services on Human Well-being
E9: Assessing Transferability of Measures and Models
E10: Evaluating Consequences using a DASEES Approach
120
-------
Appendix Ei
Modeling Ecosystem Services Production Using EPFs
Problem
One of the first steps in assessing what ecosystems are providing to human well-being is to estimate the
production of Final Ecosystem Goods and Services (FEGS) and how that production is affected by
changes in ecosystem quantity and quality. These changes may arise from anthropogenic influence (i.e.
caused by humans) such as logging, agricultural production, urban development, transportation,
restoration etc. They can also be caused by less manageable biogeochemical cycles such as those driven
by the changing of the seasons, ecosystem succession, longer term climatic periodicity, tectonic shifts or
major disturbance events such as wildfires, floods, hurricanes and droughts. Here we explore several
approaches for assessing both the baseline production of FEGS and how it might change under decision
alternatives using ecological production functions.
Approaches and Tools
Modeling approaches for estimating the production of FEGS range from relatively simple and static
mapping to more complicated dynamic mechanistic models. The mapping approach usually entails using
literature values to estimate current stocks and production capacity based on ecosystem extents. Maps can
display the potential ecosystem spatial supply of FEGS to humans. With the inclusion of access pathways
such as proximity to residential areas, distance along transportation and hydrological networks,
viewscapes etc. maps can be used to estimate how much of that production is accessible to human use
demands, a pre-requisite for turning ecological structures and functions into ecosystem goods and
services. Mapping tools, such as EPA H20, i-Tree, In VEST, or ARIES, are useful for preliminary
ecosystem goods and services assessments and for comparing FEGS production under alternative future
scenarios. EPA H20, for example, allows users to quickly create land-use scenarios and generate reports
comparing the production and potential value of four ecosystem services (Fig. El.l).
: 00 ปmrfi
an
ฃ
(1-20)#
mm i n
MW ซ1 ฉ
ฆ TKUFUil
Control ron6anng ardor
I iMimii Mfccttd an lay*> So
Scri. 1 ฆ rem- epsg ws' $
Figure El.l. Example of using the EPA H20 to generate a new scenario by changing the
highlighted row crop land parcel to freshwater marsh.
121
-------
Empirical modeling (i.e. using measurements and statistics to model how one ecosystem factor changes
due to a change in another) offers a way to estimate production trends as conditions change within an
ecosystem. These models are useful for assessing more subtle changes in condition, beyond a simple
categorical change from one land-use to another. However, they require sufficient data for the generation
of response curves. Ecosystems are dynamic and interact in multiple non-linear ways, however, so robust
estimates of FEGS production may require a mechanistic systems understanding in order to account for
temporal trends, spatial interactions, and feedbacks within and among ecosystems.
Mechanistic modeling (i.e., specifying the nature of relationships using the underlying biophysical
processes) can be illustrative for exploring unintended consequences to parts of the system not targeted by
management efforts. This can be useful for generating more holistic consequence tables, and for
integrating effects through time and space to estimate net changes to FEGS production and access by
humans. Data requirements for mechanistic modeling are often extensive and certainty in model output
tends to decrease as more interacting factors are added. Once developed, however, mechanistic models
afford users a more realistic and transparent tool for exploring and managing the underlying causal factors
responsible for changes in FEGS production.
Example: The Value of Conservation Programs
Hillsborough County, located to the North East of Tampa Bay, maintains a significant amount of
conservation lands through the Jan K. Piatt Environmental Lands Acquisition and Protection Program
(ELAPP). More than 61,000 acres of environmentally sensitive wildlife habitat and corridors (Fig. 2) are
managed by this voluntary program established for the purpose of providing the process and funding for
identifying, acquiring, preserving and protecting endangered, environmentally-sensitive and significant
lands in Hillsborough County. In 1987, a time when many of the environmental efforts responsible for the
now restored condition of the bay were getting started, The Board of County Commissioners approved an
Environmentally-Sensitive Lands Ordinance. The ordinance provided $21 million for ELAPP over a
four-year period to acquire environmentally-sensitive lands. Since then they have acquired a further $300
million in bonds to secure other land purchases. In 2016, ELAPP piloted the EPA H20 during a
preliminary valuation of ecosystem goods and services production from their conservation lands as one
way of demonstrating the value of the program (Fig. E1.2).
Figure El.2. Example output from EPA H20 for the Hillsborough County, FL.
122
-------
ELAPP conservation land purchases (grey parcels) as of 2016 cost over $300 million, but potentially
provide water treatment through nitrogen removal and reduced respiratory health costs through air
pollution removal worth close to $80 million a year to the region's population. In addition, these lands
produce flood water retention potentially worth over $350 million in avoided storm water infrastructure to
downstream residents and carbon sequestration worth over $26 million a year in avoided future costs of
climate change to the global human population.
Lessons Learned: The Ecosystem Service Cascade
In its simplest form, an ecological production function relates a change in something to an ecological
attribute as a linear function. However, there are many things that could be described as having an effect
on or as being an ecological attribute. Effectors could be changes in ecosystem policy, use or
management, a change in land use or cover, or even changes in ecological structure or function.
Responsive ecological attributes also cover a wide range of things such as ecosystem types, biophysical
characteristics, abiotic conditions, structural complexity, process rates etc. This complexity means that
any qualitative or quantitative relationship describing two attributes that lie on the production pathway
leading towards a final ecosystem good or service can be described as an ecological production function
(Fig. El.3). While it is an artificial break point, PEGS represent the end of the ecological production
function chain and the beginning of what could be described as the ecological benefit function chain. This
is where more traditionally considered goods and services, such as equipment, modes of transportation,
labor, and other capital, are needed to translate EEGS into some form of benefit to humans.
Ecological Model Variable Typology
Ecological Model Variable Typology
r >
'
f >
r a
t 1
r
f \
Position
Position
Position
Position
Position
Position
Position
Position
0
1
2
3
4
5
6
7
Change in
Und
Stressor (or
Ecological
final
f ECS
Social
Economic
Ecosystem
Cover
kruclwt or
Eeosyst,
Litoly to
Sena'it
Value of
Polity, Us*
Procws
Good o*
bi Uttd
Indicator
Social
Of Mgirtl.
Inlffrme-djal#
Strvic#
eซwrn
<ซGS)
' '
Ecological Production Functions Ecological Benefit Functions
Figure El.3. The EcoServices Model Library (ESML; Appendix E4) variable typology showing that
FEGS are the break point between EPFs and EBFs.
For More Information
EPA H20 tool: http://www2.epa.gov/water-research/ecosystem-services-scenario-assessnient-using-epa-h2o
i-Tree: https://www.itreetools.org/
InVEST; http://www.natnralcapitalproject.org/invest/
ARIES: http://aries.integratedmodelling.org/
123
-------
Appendix E2
Modeling Ecosystem Services Production using SPA
Background
Ecosystem services mapping can inform spatial planning, explore relationships between landscape change
and ecosystem services bundling, and provide a means to co-produce ecosystem services data and
valuation with stakeholders (Troy and Wilson 2006, Klain and Chan 2012, Hutchison et al. 2015). One
version of ecosystem services mapping is a service provision map (de Groot et al. 2010), which is a
geospatial expression of the relationship between ecological variables (land cover, primary production)
and ecological production functions.
Approach
Service providing area (SPA) mapping is an extension of this idea, specifically mapping the spatial
distribution of indicators that correspond to a variety of ecosystem services (Fig. E2.1; Angradi et al.
2016). In a Geographic Information System (GIS), a service providing area (SPA) is a polygon of
contiguous pixels where the ecosystem services indicator is present that represents the area providing an
ecosystem service (Fisher et al. 2009, Syrbe and Walz 2012).
Number of Services
Boundary of aquatic
Kilometers
Figure E2.1. Composite SPA map for the St. Louis River AOC showing the number of final
ecosystem services for each 10Q-m2 map pixel. Insets show riparian detail (from Angradi et al.
2016).
124
-------
Service providing area mapping can be used in trade-off analysis to examine how human-caused changes
in environmental quality or land-use change the extent and distribution of ecosystem services. Angradi et
al. (2016) examined how planned sediment remediation or ecosystem restoration projects would affect
ecosystem services in the St. Louis River Area of Concern (AOC), one of 27 US AOCs in the EPA AOC
program. Individual ecosystem services such as boating areas, wild rice beds, and game fish spawning
habitat were mapped and then bundled to evaluate how they collectively changed in response to project
design alternatives using area-based tabulations (e.g., acres of boating area).
A heat map of cumulative ecosystem services was created to identify important areas for ecosystem
services provisioning (Fig. E2.1). The map was used to identify the number of services associated with
specific geographic zones within the St. Louis River AOC, but also specific habitats. Notably, the choice
of ecosystem services to map was dependent on the ecological character and land use of the AOC,
available data, and the decision context.
Lessons Learned
The advantage of the SPA approach is that the ecosystem services models are
occurs entirely within a GIS environment and likely is within the institutional
government. The maps and tabulations are an approach to ecosystem services
to design, plan and evaluate environmental management actions.
For More Information
Angradi, T.R., D.W. Bolgrien, J.L. Launspach, B.J. Bellinger, M.A. Starry, J.C. Hoffman, M.E. Sierszen, A.S.
Trebitz, T.P. Hollenhorst. 2016. Mapping ecosystem services of a Great Lakes estuary can support local
decision-making. Journal of Great Lakes Research 42:717-727.
de Groot, R.S., R. Alkemade, L. Braat, L. Hein, L. Willemen. 2010. Challenges in integrating the concept of
ecosystem services and values in landscape planning, management and decision-making. Ecological Complexity
7:260-272.
Fisher, B., R.K. Turner, P. Morling. 2009. Defining and classifying ecosystem services for decision-making.
Ecological Economics 68:643-653.
Hutchison, L., P. Montagna, D. Yoskowitz, D. Scholz, J. Tunnell. 2015. Stakeholder perspectives of coastal habitat
ecosystem services. Estuaries and Coasts 38 (Supplement l):67-80.
Klain, S.C. and K.M.A. Chan. 2012. Navigating coastal values: participatory mapping of ecosystem services for
spatial planning. Ecological Economics 82:104-113.
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.
Troy, A. and M.A. Wilson. 2006. Mapping ecosystem services: practical challenges and opportunities in linking GIS
and value transfer. Ecological Economics 60:435-449.
transparent, mapping
capacity of local
assessment that can be used
125
-------
Appendix E3
Using the Modeling Tool VELMA to Compare Scenarios
Overview
VELMA - Visualizing Ecosystems for Land Management Assessments - is a spatially distributed, eco-
hydrology model that links a land surface hydrology model with a terrestrial biogeochemistry model to
simulate the integrated responses of vegetation, soil, and water resources to interacting stressors
(Abdelnour et al. 2011 and 2013; McKane et al. 2014a, b). For example, VELMA can simulate how
changes in climate and land use interact to affect soil water storage, surface and subsurface runoff,
evapotranspiration, vegetation and soil carbon and nitrogen dynamics, and transport of dissolved nutrients
and contaminants to streams and estuaries (Fig. E4.1).
VELMA provides a means for synthesizing decades of data describing component parts of ecosystems
that most often have been studied in isolation from the rest of the system. This systems-level synthesis
enables analyses of unanticipated consequences of land management policies and practices, and a deeper
understanding of environmental responses and underlying causal factors. VELMA differs from other
existing eco-hydrology models in its theoretical foundation, transferability, and ability to simulate trade-
offs among many ecosystem services (Fig. E3.1). The model has a user-friendly Graphics User Interface
(GUI) for easy input of model parameter values. In addition, advanced visualization of simulation results
can enhance understanding of results and underlying concepts. VELMA's visualization and interactivity
features are packaged in an open-source, open-platform programming environment (Java / Eclipse).
Climate & Land Use Effects Simulated
Hydrology: streamflow, vertical & lateral
flow, evapotranspiration, soil moisture,
stream temperature
Plants & Soils: uptake, transformation and
transport of carbon, nutrients and toxics
from terrestrial to aquatic systems
Ecosystem Services Simulated
Water quality (nutrients, toxics...)
Water quantity (floods, low flows)
Food & fiber production
Climate regulation (C02, N20, NO J
Fish & wildlife habitat
To
Stream
Figure E3.1. VELMA conceptual model.
Model Validation and Transferability
VELMA has been validated for simulating the effects of climate and land use on water quality and
quantity and ecosystem carbon and nitrogen dynamics, initially for coniferous forest ecosystems in the
Pacific Northwest (Abdelnour et al. 2011, 2013; McKane et al. 2014a,b). VELMA has since been applied
to a wide range of major ecosystem types across North America - Central Plains rangelands; eastern
deciduous forests; agricultural-riparian forest systems draining to Chesapeake Bay; tidal marshes in
Oregon; and arctic tundra in Alaska. Across these biomes, VELMA successfully simulates observed
effects of climate and land use on water quality and quantity, vegetation growth and turnover, and soil
126
-------
carbon and nitrogen cycling. A number of current applications are focused on urban stormwater
remediation.
Importantly for regional-scale applications, VELMA transferability tests demonstrate that parameter sets
developed for data-rich validation sites (e.g., the H.J. Andrews Experimental Forest in western OR) can
be applied without change to locations hundreds of miles distant, without any significant loss in
predictive accuracy. This includes simulation of observed streamflow dynamics and successional trends
in plant and soil carbon and nitrogen dynamics. VELMA has also been successfully applied to the
-10,000 square-mile tallgrass prairie rangelands of eastern Kansas.
Decision Support
The overall goal of VELMA has been to provide comprehensive decision support tools that can help
local, state and federal decision-makers address present needs without compromising the ability of society
and the environment to meet the economic, social and human health needs of current and future
generations. Key decision support goals are to (1) quantify and value the services ecosystems provide for
humans, and (2) assess the effectiveness of natural and engineered green infrastructure for protecting
water quality of streams and estuaries.
VELMA was specifically designed to provide information for quantifying how alternative land use and
policy scenarios affect trade-offs among multiple ecosystem services. Using the terminology of the
Millenium Ecosystem Assessment (2005), VELMA can help quantify supporting services (hydrological
processes, nutrient cycling, etc.), provisioning services (clean water, food, fiber, biofuels, etc.), regulating
services (flood control, carbon sequestration, climate regulation, disease control, etc.) and cultural
services (recreation, spiritual inspiration, etc.).
While VELMA has proven useful for quantifying how such services interact and respond in concert to
environmental changes, it is important to also quantify the economic, social and health impacts or benefits
associated with such changes. In this context, supporting and regulating services are best described as
intermediate ecosystem goods and services, or IEGS. A problem with IEGS is that they are tightly
integrated and prone to double counting in the valuation of different services. Instead, IEGS need to be
linked to Final Ecosystem Goods and Services, or PEGS, for valuation purposes (see Appendix G3). This
is accomplished through an established classification system of the final products of nature enjoyed by
defined groups of beneficiaries (Landers and Nahlik 2013). FEGS can then be less ambiguously translated
into societal benefits that may involve both monetary and non-monetary valuations (Fig. E3.2).
Ecol. Prod.
Functions
Benefit
Functions
Impact
Functions
Decision
Alternatives
A Final EGS
A Human
Well-being
Social & Economic
Services
A Ecosystem State
& Intermediate EGS)
Information for Decision Support
Figure E3.2 FEGS based approach conceptual model.
127
-------
To facilitate the process of translating decision alternatives -> IEGS -> FEGS Benefits (see Figure
E3.2), we have collaborated with Oregon State University to link VELMA with ENVISION, a well-
established decision support tool that integrates landscape GIS layers, stakeholder defined decision
scenarios, ecological models, and various benefits valuation tools (http://envision.bioe.orst.edu/). The
integration of VELMA with Envision is complete and will be tested for various communities. For
example, see Appendix G3 for a Pacific Northwest Case Study where VELMA has been transferred to
tribal and community stakeholders in Washington state. They are using the model to help inform their
decision-making process to improve watershed-scale ecosystem services and their cultural, economic and
general well-being. The linked Envision-VELMA platform will also be transferred to these stakeholders
after proof-of-concept testing has been completed.
Products and Impacts
The major product of this research is a broadly applicable decision support tool that community
stakeholders can use to explore trade-offs in bundled ecosystem goods and services resulting from
alternative decision choices to support sustainable decision-making. Such tools can help balance
environmental, economic and human well-being criteria over timescales relevant to immediate needs and
long-term planning goals.
For More Information
Abdelnour, A., M. Stieglitz, F. Pan, R. McKane. 2011. Catchment hydrological responses to forest harvest amount
and spatial pattern, Water Resources Research, 47, W09521.
Abdelnour, A., R. McKane, M. Stieglitz, F. Pan, Y. Cheng. 2013. Effects of harvest on carbon and nitrogen
dynamics in a Pacific Northwest forest catchment, Water Resources Research, 49.
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, 2013.
McKane, R.B., A. Brookes, K. Djang, M. Papenfus, J. Ebersole, D. Phillips, J. Halama, P. Pettus, C. Burdick, and
M. Russell. 2014a. Sustainable and Healthy Communities Pacific Northwest Demonstration Study. US
Environmental Protection Agency, Report No. ORD-007386.
McKane, R.B., A. Brookes, K. Djang, M. Stieglitz, A. Abdelnour, F. Pan. 2014b. Enhanced version of VELMA
ecohydrological modeling and decision support framework to address engineered and natural applications of GI
for reducing nonpoint inputs of nutrients, contaminants, and sediments. US Environmental Protection Agency,
Report ORD-010080, Safe and Sustainable Waters Research Program.
Millennium Ecosystem Assessment. 2005. A Framework for Assessment. Washington, DC: Island Press.
128
-------
Appendix E4
The EcoService Models Library (ESML)
Background
As our knowledge grows about the ecological processes underlying the provision of FEGS, computational
models of these processes are being developed by scientists in government, academia and business. These
models are highly diverse (see Appendix El), and information about them is scattered throughout
journals, websites and government reports, and may not be available when needed to inform decision-
makers. The EcoService Models Library (ESML) is an online database of EPFs. ESML's online interface
(Fig. E4.1) is designed to help users quickly find and compare models, based on an ecosystem service
that can be estimated or a location where previously applied. ESML is also designed for researchers
interested in improving ecological modeling methods. Ecological models (EMs) described in ESML
range from well-known online tools to quantitative relationships described in the scientific literature.
cvFPA ap-u 1
* ^ Lai m m United SUtes Environmental Protection Agency
I Tieng Vtฃt
I e^oi
Learn Ltie Issues Science & technology Laws & Regulations About EPA
KTffiiUfiT
wm
EcoService Models Library (ESML)
My EMs IQ\ 0 Learn about ESML
~j Contact Us
search E
Learn about the ESML
My EMs
Search Ecological Models (EMs)
Search the ESML for EMs and related variable and source document
information.
Find Source Document Info
ESML Data and Guiding Concepts
Learn about ecological models, the data contained within this tool
and the underlying concepts regarding their use.
Using ESML
Understand how to take advantage of the features available within
this tool.
o
Figure E4.1. ESML is an online database that enables users to rapidly find and compare
information on >100 ecological models or model applications. Beta users may register at
https://esml.epa.gov/epf_l/public/signup
Finding the Right Model
Models selected to estimate FEGS need to fit the conceptual model developed for the decision problem at
hand (see Section 5.2), and they must conform to limitations of user experience and data availability.
ESML provides key information to assist in determining model fit.
ESML is supported by a structured database describing models, variables within each model, and model
source documents (Fig. E4.2). The ESML collection can be searched by environment type, location of
model application, ecosystem service or by keyword. ESML guides the user through the use of several
concepts for comparing and selecting ecological models (EMs) by providing key information about user-
selected EMs relating to each concept (Table E4.1). For each model or model application included,
ESML provides a variable relationsliip diagram (VRD) showing all model variables, their types (i.e.,
predictor, intermediate, response), their units (if available), and the logic relationships between them (Fig.
E4.3).
129
-------
Description
EM Source Document
(uniquely identified
by Document ID)
EM Variable
(uniquely identified
by Variable ID)
Ecological Model (EM)
(uniquely identified by EM ID)
al Aspects
Figure E4.2. Data map for ESML. Circled numerals indicate the number of descriptors in the
database for each model aspect.
Table E4.1. Key concepts used by ESML to assist the user in comparing and fitting models to an
intended use.
To understand this
Key Concept...
... ESML provides the following information for each included model or
model application (to the extent available in model documentation)
Modeling Objective
Summary Description I Policy or Decision Context I Scenario Drivers I
Ecosystem services according to CICES (Common International
Classification of Ecosystem Services) IEEGS according to NESCS (National
Ecosystem Services Classification System) I Model Variables (names, units)
Modeling Context
Modeling Situation (Spatial, Temporal, Ecological) I Modeling Scale
(Spatial, Temporal, Ecological) I Predictor Variable Numerical Values
Feasibility of Model
Use
Indications of wide use and support I Listing of predictor variables and their
data sources I Temporal and Spatial Discreteness of EM Computations, and
Grain Type or Size
Model Certainty
Model goodness of fit I Model validation I Characterization of predictor or
response variable variability I Sensitivity analysis I Uncertainty analysis
Potential for
Linkage to Other
Models
Listing of model variables using an Ecological Model Variable Typology
Model Structural
Information
Time-dependent vs. stationary I Uses "Future time" vs. "Past time" I
Temporally continuous vs. discrete I Spatially distributed vs. lumped
parameters I Analytic vs. numeric computation I Deterministic vs. stochastic
130
-------
PC*: Water demand [mm
watershed2^^r^l]^_^
PD: Land use land cover class
RC: Water yield [mA3 haA-l yrA-l]
RC: Average annual watersupply
(drinking) [mA3 haA-l yrA-l]
PC*: Root depth (max for
vegetated land use classes)
PCt: Evapotranspiration
coefficien^ฃ|ant^^^^
PCI: Effective soil depth
PD*: Evapotranspiration [mm
PC: Zhang coefficient
PD: Topography/DEM
PD*: Precipitation [mm yrA-l]
PD: Plant available water
contentjsoil]_[mm^3]__
PD: Sub-watershed boundary
PC: Watershed bounda
Figure E4.3. ESML provides a variable relationship diagram (VRD) for each included model or
application. Pictured is the VRD for an application of the InVEST water provision model for a
river in Spain (Marques et al. 2013). PD = Predictor, Driver; PC = Predictor, Constant/Factor; RC
= Response, Computed. Arrows denote that one variable (or variables, if gathered within a box) is
required for computation of the other. Asterisk indicates that data for multiple runs of this EM are
present in ESML; the value of a variable that is marked with an asterisk changes to define run
conditions. Double dagger denotes a variable whose value is constant with respect to a driving class
variable (such as when derived from a lookup table).
Associating Ecological Models with FEGS
To show which ecosystem goods and services could potentially be modeled using a given EM, ESML
links each EM to particular ecosystem services using two different classification schemes,
NESCS. The National Ecosystem Services Classification System (EPA 2015) includes only final
ecosystem goods or services, or FEGS. The classification approach for the supply of theseFEGS
(NESCS-Supply, or NESCS-S) focuses first on the type of environment where the EEGS is provided and
second on the biophysical nature of the EEGS produced (not the kind of benefit derived from it). The four
NESCS hierarchical levels used by ESML are Environmental Class, Environmental Sub-Class, End
Product Class and End Product Sub-Class. Finally, a modifier is appended to address the specific aspect
of the FEGS that the EM may be able to estimate (i.e., Stock Indicator, Flow Indicator, Quality Indicator,
Site Indicator, Extreme Event Indicator).
CICES. The Common International Classification of Ecosystem Services (EEA 2016) is organized
according to the nature of the benefit that humans receive; for example, the first (i.e., Section) level of the
CICES hierarchy distinguishes provisioning, regulation-maintenance and cultural services; provisioning
services break down (by Division) into nutrition, materials and energy - and so on. The lour CICES
hierarchical levels used by ESML are Section, Division, Group and Class. The ecosystem goods and
131
-------
services described by CICES include both final and intermediate goods and services.
For More Information
ESML is currently available for beta testing. Users may register at
https://esml.epa.gov/epf_l/public/signup.
References
EEA (European Environment Agency). 2016. CICES: Towards a common classification of ecosystem services.
European Environment Agency.
EPA (Environmental Protection Agency). 2015. National Ecosystem Services Classification System (NESCS):
Framework design and policy application. Washington, DC: U.S. Environmental Protection Agency, Office of
Water and Office of Research and Development. No. EPA-800-R-15-002.
Marques M., R.F. Bangash, V. Kumar, R. Sharp, M. Schuhmacher. 2013. The impact of climate change on water
provision under a low flow regime: A case study of the ecosystems services in the Francoli river basin. Journal
of Hazardous Materials 263(Part l):224-232.
132
-------
Appendix E5
Non-Market Economic Valuation for Ecosystem Services
Methods for Economic Valuation of Ecosystem Services
The interest in non-market valuation for ecosystem services has risen in the last few decades. Providing
monetary values for goods and services provisioned by the environment presents a method to tangibly
convey the value of nature's benefits to the general public and decision-makers at all levels. Methods for
non-market valuation can be divided into revealed or stated preferences (Table E5.1). Revealed
preferences methods includes the hedonic pricing and travel cost methods, which seek to "reveal" the
implied values that preface an individual's choices. Stated preferences methods, like contingent valuation
and choice experiments, ask respondents within studies to directly indicate their willingness to pay (WTP)
for quantities of environmental goods and services.
Table E5.1. Descriptions of Non-Market Economic Valuation Methods.
Method Type
Valuation Method
Description
Benefits
Transfer
Benefits Transfer
A method that applies economic value estimates from one site
to another - through value or function transfer-usually within a
similar geography and context (decision scenario,
socioeconomic, biophysical) when resource/time constraints
hinder local data collection
Revealed
Pre Terence
Travel Cost Study
A study that assesses the costs of recreational travel as a proxy
for the value of site (ecosystem) visited.
Revealed
Preference
Hedonic Analysis
Assesses the influence of an environmental/ ecosystem
attribute (i.e. water quality, air quality) on property values.
Stated
Preference
Contingent
Valuation
Directly asks participants in study to state their willingness to
pay/accept compensation for modifications in ecosystem
services provisioning
Stated
Preference
Discrete Choice
Experiment
Creates a hypothetical market scenario in which participants
are asked to choose between alternatives of enhanced
ecosystem services. Produces marginal WTP values for
indicators of ecosystem services. Useful in assessing
preferences for trade-offs within decision scenarios.
Discrete Choice Experiment Example
Discrete Choice Experiments (DCEs) are especially helpful in trade-off analysis and comparing decision
scenario alternatives. DCEs involve the creation of choice cards, which contain different scenarios- a
status quo scenario and scenarios featuring enhanced ecosystem services - with varying levels of
enhancements within attributes of interest. Johnston et al. (2011) conducted a DCE in order to assess
preferences and WTP for the restoration of migratory fish within the Pawtuxet Watershed in Rhode
Island.
An example of a choice card from the Johnston et al. (2011) study is displayed in Table E5.2. The card
contains six attributes and the payment attribute, along with three scenarios- one status quo scenario and
two scenarios that feature restoration. Within the DCE, respondents were asked to evaluate all three plan
options and choose the plan they would be most willing to pay for with respect to the payment attribute at
the bottom of the card. The three plan options all contain varying levels of the restoration plan attributes,
133
-------
which represent changes of ecosystem services provisioning or benefits resulting from watershed
restoration.
Table E5.2. Choice Card Example, modified from Johnston et al. 2011.
Question 5. Projects A and B are possible restoration projects for the Pawtuxet River, and the Current
Situation is the status quo with no restoration. Given a choice between the three, how would you vote?
Effect of
Restoration
Measure
Current Situation
(no restoration)
Restoration
Project A
Restoration
Project B
Fish Habitat
Percent of river acres
accessible to fish (out of
4347 acres)
0%
5%
20%
Fish Population
Survival Score
Chance of 50-year survival
0%
30%
30%
Catchable Fish
Abundance
Percent of fish found per
hour (out of 145)
80%
70%
70%
Fish-Dependent
Wildlife
Percent of native species
that are common (out of 36)
55%
80%
60%
Aquatic Ecological
Condition Score
Natural condition out of
100% maximum
65%
70%
80%
Public Access
Can public walk and fish in
area
No
No
Yes
Cost to your
Household per year
Dollar increase in annual
taxes and fees
$0
$15
$25
HOW WOULD YOU VOTE?
(CHOOSE ONE ONLY)
~
I vote for NO
RESTORATION
~
I vote for
PROJECT
A
~
I vote for
PROJECT
B
The choice in restoration plan scenario is the foundation for the choice modeling that allows researchers
to estimate WTP. Respondents' choices are modeled through various regression analyses to estimate plan
attribute coefficients to develop marginal WTP (mWTP) values for statistically significant plan attributes.
/wWTP for plan attributes is defined as:
mWTP = 4^
(Pc)
where, |3n is the coefficient of a plan attribute and |3C is the coefficient of the payment attribute (or cost of
the restoration plan; Haab and McConnell 2002). Since mWTP is calculated for individual plan attributes,
we can observe which attributes respondents preferred and how respondents valued trade-offs within
decision scenarios- in addition to monetizing these benefits.
134
-------
The WTP results from Johnston et al. (2011) are displayed in Table E5.3. In this example, respondents
were willing to pay more per percentage point for aquatic condition (measured as an Index of Biological
Integrity, IB I) than for any other attribute. Because survey respondents were provided information about
measurements that form the IBI (e.g., mussel species richness, native species richness, sensitive species
richness), the methods can be used to disentangle the marginal values of intermediate ecosystem services
(e.g., richness of mussel species) that stakeholders do not directly value through the use of a composite
measure (IBI) which may be more meaningful to respondents as a proxy for Final Ecosystem Goods and
Services.
Table E5.3. WTP results from Johnston et al. (2011). WTP estimates are per household, per year.
For all variables, except access, estimates represent WTP ($) for a one percent increase in that
variable. P-values are two-tailed for the null hypothesis of zero WTP.
Variable
WTP
Standard
Deviation
P-value
Fish Habitat
1.0910
0.3523
<0.01
Fish Population Survival Score
0.4136
0.1462
<0.01
Catchable Fish Abundance
0.0688
0.2073
0.72
Fish-Dependent Wildlife
0.6369
0.2088
<0.01
Aquatic Ecological Condition Score (IBI)
1.1879
0.5017
<0.01
Public Access
27.3285
6.0602
<0.01
For More Information
Haab, T.C., and K.E. McConnell. 2002. Valuing environmental and natural resources: the econometrics of
non-market valuation. Edward Elgar Publishing.
Johnston, R.J., K. Segerson, E.T. Schultz, E.Y. Besedin, M. Ramachandran. 2011. Indices of biotic
integrity in stated preference valuation of aquatic ecosystem services. Ecological Economics
70(11): 1946-1956.
135
-------
Appendix E6
Measuring Health Outcomes
Background
Ecosystem services have been linked to a variety of physical and mental health benefits. For example, the
recreation opportunities provided by nature promote greater physical activity and stress reduction, leading
to lower rates of obesity and depression (Bell et al. 2008; Gariepy et al. 2014). The removal of air and
water pollutants by trees and wetlands helps protect urban communities from respiratory, cardiovascular
and gastrointestinal diseases (Yang et al. 2008; Nowak et al. 2013). Trees and wetlands also help mitigate
weather hazards, ameliorating the impact of heat waves and flood events on human health (Reacher et al.
2004; Harlan et al. 2013). Many of these connections have been well studied and as a result, there is a
growing number of resources available for communities to incorporate ecosystem services into decision-
making for the benefit of human health. Here we list some of these resources and provide examples of
how to apply these tools at the community level.
Available Tools
To understand the role of ecosystem services on human health, a good starting point is the Eco-Health
Relationship Browser (Jackson et al. 2013), which compiles the evidence currently available, and displays
linkages in a user friendly interface (Fig.E6.1).
Forests
Wetlands
Stress
Urban
Ecosystems
Vulnerable
Populations
Self-Esteem
ADHD
Respiratory
Symptoms
Anxiety
PTSD
Obesity
Cardiovascular
Diseases
Recreation &
Physical
Activity
Cognitive
Function
Mortality
Confusion
Mental Health
Depression
Low Birth
Weight
High Blood
Pressure
Fatigue
Diabetes
Longevity
Happiness
Aggression
Figure E6.1. The Eco-Health Browser visualizes linkages between ecosystems and human health,
and provides literature. Urban Ecosystems Obesity, used in an example below, is highlighted.
136
-------
Next, communities may be interested in knowing what environmental stressors they are exposed to, and
what natural assets may protect them from these stressors;. This type of information can be found in the
EnviroAtlas at the National Level and for around 50 communities through the US. In addition,
understanding the current health status of your community is key for establishing priorities, and assessing
the impact of management decisions. The CDC's Community Health Status Indicators (CHSI) provides
information about disease trends at the county level, and visualization tools that help define how a
specific place compares to the rest of the nation. For higher resolution, the CDC foundation recently
launched the 500 cities project, which provides census block group health data, allowing communities to
access information specific to their neighborhoods. For neighborhoods not featured in the 500 cities
project, but that need information at a fine resolution, conducting Health Impact Assessments (HIA) may
be the best alternative (see Appendix E7). The Eco-Health Browser and EnviroAtlas as well as CDC's
Healthy Places Initiative provide tools, conceptual models, and educational material to help guide this
process. Lastly, modelling approaches such as i-Tree Eco and BenMap are useful for communities
interested in quantitatively estimating health benefits/impacts at present conditions and under different
scenarios. These modelling tools are mainly available for the most studied Eco-Health linkages such as
the role of trees in preventing respiratory and cardiovascular diseases, as well as heat hazards.
Example: Tree cover in Tampa Florida and Obesity Rate
Tampa Florida, is featured in both the EnviroAtlas Community datasets and the 500 Cities Project, which
allows for low scale visualizations of Eco-Health linkages. In this example, we used block group data
from EnviroAtlas showing the percentage of tree cover within walkable streets. We compared this map
layer with data on obesity rates from the 500 cities project (Figure E6.2).
Neighborhood Obesity and Tree Cover in
Tampa, FL.
o
0 0.01
Walkable Streets with >75% Tree Cover (Area Ratio)
% Tree Cover along Walkable Roads
o' <ฃ> ^3
Figure E6.2. Descriptive analysis of obesity trends in relation to tree cover at the neighborhood
scale. Data on tree cover comes from EnviroAtlas. Obesity data comes from the 500 Cities Project.
137
-------
The resulting scatterplot suggests lower obesity rates on block groups with a high percentage of trees
bordering walkable streets. This trend corresponds to the relationships shown in the Eco-Health browser,
and may be used to set research priorities. For example, a follow up research question would be to ask if
the relationship is driven more by walkability, by greenness or by the interaction of both elements. This
and similar questions would help communities prioritize ecosystem service management for health
promoting services.
Transferability of Approach
Many communities are not included on the high resolution datasets used for the example above. In such
cases, County data may bring to attention larger trends to guide prioritization for research and
management. For example, CHSI data for counties in Tampa, FL suggests lower physical in-activity and
obesity rate for Pinellas vs Hillsborough County (Fig. E6.3). This finding may lead communities in
Hillsborough to investigate potential drivers, such as the difference in proximity to green spaces observed
in the same two counties at lower resolutions (Fig. E6.3).
County Level Obesity, Physical
Green Spaces in Tampa, FL.
Inactivity and Proximity to
ij (
30
25
20
15
10
,, * :
*ซE v%? ฆ' wป
21.2
25
24.3
26.4
%Physical Inactivity % Obesity
i Pinellas Hillsborough
:'f}
Pinellas |TJ~
C1 Hillsborough
r
L
Proximity Green Spaces
Near
Moderate
Far
Figure E6.3. County level analysis. Data on Green Spaces comes from EnviroAtlas. Data on Obesity
and Physical Inactivity comes from CHSI.
Transferability may also be affected by socio-economic and behavioral differences among exposed
populations, which may either exacerbate or ameliorate the impact of ecosystem services on health.
Incorporating data from the U.S census or CDC's behavioral risk factor surveillance system (BRFSS)
could assist in identifying important confounding factors that may alter Eco-Health linkages.
Summary
Practical strategies discussed here for measuring health outcomes associated to Ecosystem Services are
summarized in Table E6.1.
138
-------
Table E6.1. Examples of potential steps in measuring community level health outcomes. Not all steps
need to be followed, and the list is not comprehensive.
Steps
Useful Tools*
Scale
Understand Eco-Health Linkages
Eco-Health Browser
All
Evaluate Exposure to Environmental Stressors
EnviroAtlas
National-HUC/Community
Determine Health Priorities
CDC-500 Cities
HIA**
CDC-CHSI***
Community
Community
National-County
Model Current and Future Impact to Health
BenMap
I-Tree
All
All
'These are just examples of practical tools, and not a comprehensive list
"Health Impact Assessment (HIA)
""Center for Disease Control (CDC) Community Health Status Indicators (CHSI).
For More Information
500 Cities Project https://www.cdc.gov/500cities/
Bell, J.F., J.S. Wilson, G.C. Liu. 2008. Neighborhood greenness and 2-year changes in Body Mass Index of children
and youth. American Journal of Preventive Medicine 35(6):547-553
BenMap Model: https://www.epa.gov/benmap
Centers for Disease Control and Prevention (CDC) Behavioral Risk Factor Surveillance System (BRFSS):
https://www.cdc.gov/brfss/
CDC Community Health Status Indicators (CHSI) https://www.cdc.gov/CommunityHealth/home
Gariepy, G., A. Blair, Y. Kestens, N. Schmitz. 2014. Neighborhood characteristics and 10-year risk of depression in
Canadian adults with and without a chronic illness. Health and Place 30:279-86
Harlan, S.L., J.H. Declet-Barreto, W.L. Stefanov, D.B. Petitti. 2013. Neighborhood effects on heat deaths: social and
environmental predictors of vulnerability in Maricopa County, Arizona. Environmental Health Perspectives
(Online) 121(2): 197.
I-Tree Model: http://www.itreetools.org/
Jackson L.E., J. Daniel, B. McCorkle, A. Sears, K.F. Bush. 2013. Linking ecosystem services and human health:
The Eco-Health Relationship Browser. International Journal of Public Health 58:747-755.
Nowak, D.J., S. Hirabayashi, A. Bodine, R. Hoehn.. 2013. Modeled PM 2.5 removal by trees in ten US cities and
associated health effects. Environmental Pollution 178:395-402.
Reacher, M., K. McKenzie, C. Lane, T. Nichols, I. Kedge, A. Iversen, J. Simpson. 2004. Health impacts of flooding
in Lewes: a comparison of reported gastrointestinal and other illness and mental health in flooded and non-
flooded households. Communicable Disease and Public Health, 7(l):39-46.
U.S. Census: https://www.census.gov/
U.S. Environmental Protection Agency. 2015. Health Impact Assessment & EnviroAtlas: Integrating Ecosystem
Services into the Decision Making Process. Office of Research and Development, National Health and
Environmental Effects Research Laboratory, Research Triangle Park, NC. EPA/600/RR-15/128.
Yang, Q., N.F. Tam, Y.S. Wong, T.G. Luan, W.S. Su, C.Y. Lan, S.G. Cheung. 2008. Potential use of mangroves as
constructed wetland for municipal sewage treatment in Futian, Shenzhen, China. Marine Pollution Bulletin
57(6):735-743.
139
-------
Appendix E7
Health Impact Assessment and Ecosystem Services
Background
Health Impact Assessment (HIA) is a process used to inform community decision-making of potential
well-being outcomes, specifically in regard to human health. HIA is tied to the evaluation of one or more
alternatives facing decision-makers on human health outcomes. The decision may or may not have an
explicit human health goal, as many decisions facing communities do not (examples include community
planning of transportation corridors and green spaces and funding infrastructure projects). Human health
may be affected in obvious or subtle ways, and HIA aims to provide a structured methodology to
document the various human health dimensions of decisions in a way the decision-makers can easily use.
Figure E7.1. The six steps of a Health Impact Assessment Process, from Mecklenburg County, NC
Health Department.
Approach
Mien conducting an HIA, a holistic evaluation is performed of all health impacts that could reasonably
stem from the decisions at hand. If that decision has anything to do with the natural environment (as it
often does) there will be an implicit inclusion of ecosystem services.
Screening
Is HIA needed?
Monitoring &
Evaluation
What are the
changes in health?
What are the
outcomes of an
HIA? i
Health Impact
Assessment ^
Recommendations:
What actions can be
taken to improve health
or manage health
effects?
140
-------
A key challenge is to be able to link Final Ecosystem Goods and Services to benefits to human health.
This can be accomplished by explicitly conceptualizing impacts that decisions (e.g., a new initiative or
policy) will have on the environment (e.g., land-use change) and the various effects on ecosystem services
- that in turn could impact human health (Fig. E7.2).
Figure E7.2. Illustration of how decisions can lead to health impacts by modifying the benefits
received from the natural environment.
The cornerstone of HIA is regular meetings with decision-makers (to better understand the decision
options and issues at hand), stakeholders (to better understand health concerns and provide local
knowledge), and researchers (to provide expert opinion on impacts that are likely or unlikely to occur)
(World Health Organization 2017). Given that data, tools, and models linking ecosystem services to
human health can be difficult to find, this kind of iterative, inclusive process that leverages expert
opinion, literature reviews, and local knowledge can be invaluable.
Example
An HIA was recently conducted in Suffolk County, New York to evaluate how proposed municipal code
changes regarding onsite sewage disposal systems might affect human health (Johnston et al. 2017).
Pathway diagrams describing the potential impacts of decision alternatives on human health were
developed iteratively and interactively with researchers and stakeholders, using expert opinion, models,
and data (Fig. E7.3). The assessment included a consideration of how proposed changes might impact or
be mitigated by ecosystem services benefits.
What we think will happen....
>
->
Frequency
of failure
Wastewater
discharges
Reduction in
wetland cover
Groundwater
quality
Alternative 1
No action
Surface
water
quality
Human illness
from toxin
exposure
Human
ingestion from
drinking water
Human illness
from pathogen
exposure
Harmful algal
blooms in
water used for
recreation
Harmful
pathogens in
water used for
recreation
Illness, injury,
and mortality
From storm
events
Alternative 2
All cesspools
required to be
upgraded
Alternative 3
Cesspools in
high priority
areas required
to be upgraded
Nutrient & pathogen
buffering ability of
natural aquatic and
soil ecosystems
Figure E7.3. Simplified pathway diagram from HIA illustrating how adoption of decision
alternatives may lead to changes in human health (from Johnston et al. 2017).
141
-------
For example, the community was concerned whether nitrogen discharges from sewage disposal systems
were impacting the local estuary, particularly the ability of wetlands to buffer severe storms and
hurricanes that can affect the health and lives of local residents through flooding and erosion. Through a
series of workshops, expert discussions, and literature reviews, the HIA determined that nitrogen and
pathogen discharges from sewage disposal systems were overwhelming the ability of local soil and
aquatic ecosystems to absorb and control them. However, reductions in nitrogen pollution from proposed
changes to sewage disposal were unlikely to have much effect on wetland loss, since it is instead
primarily driven by sea level rise and ocean warming (Wigand et al. 2014). Positive health impacts of
proposed changes were more likely to be through improvements to surface water quality, reducing
illnesses in swimmers and consumers of shellfish, and groundwater quality, reducing risk of illness from
private drinking water wells.
Take Home Message
The goals of HIA and ecosystem service assessments overlap, and both involve systems thinking to
evaluate the causal pathways surrounding a decision. In some cases, decisions may directly alter
ecosystem services (e.g., loss of wetland cover) and lead to potential health impacts; in other cases,
ecosystem services may be a mitigating factor in the background influencing the impacts of decisions on
human health (e.g., ability to buffer nutrient pollution). From the perspective of an ecosystem services
assessment, methods and approaches from HIA can be applied to explicitly connect ecosystem services to
human health outcomes through working groups, expert opinion, literature reviews, and local knowledge.
Alternatively, ecosystem services concepts can be integrated into HIA to ensure that benefits provided or
lost from the natural environment are not overlooked as decision alternatives are evaluated.
For More Information
Johnston, J.M., R. de Jesus Crespo, J. Hoffman, M. Myer, N. Seeteram, K. Williams, and S. Yee. 2017. Valuing
Community Benefits of Final Ecosystem Goods and Services: What Worked (and Didn't) for Valuation. U.S.
Environmental Protection Agency, Washington, D.C. EPA/600/R-17/XXX
Wigand, C., C. Roman, R. Davey, M. Stolt, R. Johnson, A. Hanson, P. Rafferty. 2014. Below the disappearing
marshes of an urban estuary: historic nitrogen trends and soil structure. Ecological Applications, 24(4):952-963.
World Health Organization. 2017. Health Impact Assessment (HIA). http://www.who.int/hia/en/
142
-------
Appendix E8
Estimating Consequences of Ecosystem Services Changes on
Human Well-being
Background
Forecasting approaches that link decision alternatives, including those that impact ecosystem services, to
human well-being may be valuable in estimating potential benefits and trade-offs that are meaningful to
people living in a community (Summers et al. 2016). For example, the Millennium Ecosystem System
Assessment (MEA 2005) used a series of working groups to identify the potential pathways through
which provisioning, regulating, and cultural services could impact human well-being. These qualitative
relationships were used to compare the potential impacts of different future scenarios of ecosystem
services degradation on different constituents of human well-being.
The Human Well-being Index (IIWBI) conceptual model quantitatively defines these relationships in a
way that can be used to forecast the influence of ecosystem service flows on human well-being, alongside
social service flows and economic service flows (Smith et al. 2014a; Summers et al. 2016; Fig. E8.1).
HWd = (Se,Ss,Sec)
Social Services
Activism
Communication
Community
initiatives
Education
Emergency
preparedness
Family services
Healthcare
Justice
Labor
Public works
Ecosystem Services
Air quality
Food, fiber, fuel
Greenspace
Water quality
Water quantity
Economic Services
Capital Investment
Consumption
Employment
Finance
Innovation
Production
Redistribution
Human Well-being
Connection to Nature
Leisure Time
Living Standards
Cultural Fulfillment
Safety and Security
Education
Hea th
Soda Cohesion
Figure E8.1. Human Well-being Index conceptual model (simplified from Summers et al. 2016).
Human well-being domains (HWd) are a function of ecosystem (Se), social (Ss), and economic (Sec)
services.
HWBI Conceptual Model
The HWBI conceptual model was crafted to: 1) describe the relationships between ecosystem services,
social services, economic services, and human well-being (Summers et al. 2016; Fig. E8.1), 2) identify
indicators and metrics to quantify human well-being for the United States at a county level (Smith et al.
2012), 3) identify indicators and metrics to quantify services for the United States at a county level (Smith
et al. 2014a), and 4) utilize this quantitative indicator data to develop predictive models describing the
relationships between services and human well-being (Smith et al. 2014b; Summers et al. 2016).
143
-------
The IIWBI conceptual model by itself can provide a strawman by which ecosystem services can be
identified and their potential consequences on constituents of human well-being discussed through group
deliberations. National data on services metrics and well-being metrics have been used to reduce
uncertainty in the relationships between services and well-being through the development of predictive
regression models describing the most likely outcomes (Smith et al. 2014b; Summers et al. 2016). Once
alternative decisions are linked to changes in the ecological, economic, and social states of communities,
the regression models can be used to forecast potential consequences for human well-being.
Example
As an example, the IIWBI model can be used to evaluate the impacts of alternative land-use scenarios on
components of human well-being (Figure E8.2).
Ecosystem Condition
Landcover
Hydrology
Canopy cover
Impervious surface
Open water
Social
Economic
Services
Services
S = f(x,)
Ecosystem Services
Air quality
Food, fiber, fuel
Greenspace
Water quality
Water quantity
HWd = (Se,Ss,Se,
| Human Well-being |
| | Connection to Nature Leisure Time
ฉ
| | Cultural Fulfillment Living Standards
Ok
Education
Safety and Security
fi\
f \
Health
Social Cohesion
/
Figure E8.2. Applying EPFs and EBFs to link alternative land-use scenarios to effects of human
well-being objectives. Ecosystem services (Se) are functions of various attributes of ecosystem
condition (xi).
To be able to estimate and compare consequences, alternative land-use/land-cover maps must be linked to
changing ecosystem services through the application of ecological production functions (EPFs; Table
E8.1).
Table E8.1. EPFs are applied to translate attributes of ecosystem condition into measures of
ecosystem services (Russell et al. 2013; Tallis et al. 2013; Smith et al. 2017).
Ecosystem
Service
EPF methods to calculate
Air quality
Food, fiber, fuel
Greenspace
Water quality
Water quantity
Biophysical functions that calculate rates of air pollutant removal based on
changes in canopy cover
Fraction of forest cover available for timber harvest; agricultural productivity
based on literature derived rates of carbon into soil or nitrogen fixation by land-
use type
Percent cover of greenspace; bluespace (open water per population); biodiversity
based on estimates of mean richness per land-use type;
Hydrology models of sediment retention, nutrient retention, and fecal coliform
retention based on landcover type, soil type, and precipitation
Curve number method to estimate rainwater retention based on land-use and soil
types
144
-------
The relative changes in ecosystem services can then be linked to changes in human well-being through
ecological benefits functions (EBF), i.e., the application of IIWBI regression models (Summers et al.
2016). Specific changes in social or economic services could also be included directly, or linked
indirectly to changes in ecosystem services based on past data trends through a covariance matrix.
Different land-use scenarios (Fig. E8.3) may lead to different trade-offs across the domains of human
well-being, such as if human health improves but living standards decline, even if overall composite well-
being are similar.
Baseline
Scenario A
High population growth
Low resource protection
Year 2000
Scenario B
Low population growth
High resource protection
~
Water
~
Developed Open
ฆ
Developed
EI
Barren
ฆ
Deciduous Forest
ฆ
Evergreen Forest
~
Mixed Forest
~
Grassland
~
Hay and Pasture
~
Agriculture
~
Woody Wetland
ฆ
Herbaceous Wetland
Year 2050
Year 2050
Figure E8.3. Hypothetical land-cover scenarios based on decisions that influence population growth
and resource protection (FORE-SCE, USGS).
Recommendations
The IIWBI model provides cause-effect relationships for examining impacts of ecosystem services on
multiple objectives that are almost universally important to communities (e.g., health, safety, living
standards, culture). The forecasts from the IIWBI model are not intended to provide precise predictions of
future well-being. Instead, forecasting should be interpreted as a potential direction and magnitude for
highlighting potential trade-offs, and providing a starting point for further discussion (Summers et al.
2016).
For More Information
Millennium Ecosystem Assessment (MEA) 2005. Ecosystems and Human Well-being: Synthesis. Island Press,
Washington, D.C.
Russell, M Teague, A., Alvarez, F., Dantin, D., Osland, M., Harvey, J., Nestlerode, J., Rogers, J., Jackson, L.,
Pilant, D Genthner, F., Lewis, M., Spivak, A., Harwell, M., Neale, A., 2013. Neighborhood scale quantification
of ecosystem goods and services. EPA/600/R-13/295. U.S. Environmental Protection Agency, Office of
Research and Development, Gulf Ecology Division, Gulf Breeze, Florida
Smith, A., S.H. Yee, M. Russell, J. Awkerman, W.S. Fisher. 2017. Linking ecosystem services supply to stakeholder
values in Guanica Bay Watershed, Puerto Rico. Ecological Indicators 74:371-383.
Smith, L.M., H.M. Smith, J.L- Case, L.C. Harwell, J.K. Summers, C. Wade. 2012. Indicators and Methods for
Constructing a U.S. Human Well-being Index (IIWBI) for Ecosystem Services Research. EPA/600/R-12/023
Smith, L.M., C.M. Wade, K.R. Straub, L.C. Harwell, J.L. Case, M. Harwell, and J.K. Summers. 2014a. Indicators
and Methods for Evaluating Economic, Ecosystem, and Social Services Provisioning. EPA/600/R-14/184
145
-------
Smith, L.M., L. Harwell, J.K. Summers, H.M. Smith, C.M. Wade, K.R. Straub, and J.L. Case 2014b. A U.S. Human
Well-being Index (HWBI) for multiple scales: linking service provisioning to human well-being endpoints
(2000-2010). EPA/600/R-14/223.
Summers, J.K., L.C. Harwell, L.M. Smith. 2016. A model for change: An approach for forecasting well-being from
service-based decisions. Ecological Indicators 69:295-309.
Tallis, H.T., Ricketts, T., Guerry, A.D., Wood, S.A., Sharp, R., Nelson, E., Ennaanay, D., Wolny, S., Olwero, N.,
Vigerstol, K., Pennington, D., Mendoza, G., Aukema, J., Foster, J., Forrest, J., Cameron, D., Arkema, K.,
Lonsdorf, E., Kennedy, C., Verutes, G., Kim, C.K., Guannel, G., Papenfus, M., Toft, J., Marsik, M., Bernhardt,
J., and Griffin, R., Glowinski, K., Chaumont, N., Perelman, A., Lacayo, M. Mandle, L., Griffin, R., Hamel, P.,
Chaplin-Kramer, R., 2013. In VEST 3.0.0 User's Guide. The Natural Capital Project, Stanford.
146
-------
Appendix E9
Methodology to Assess whether to Transfer Existing
Measurements or Models to New Sites
Background
Utilizing measurements or models derived from one site and applying them to other sites (i.e., transferring
estimates or models) is frequently performed in economic, environmental and ecological studies. Analysts
substitute existing measurements or models from previous sites and apply them to new site(s) where data
or funds are insufficient to derive site-specific estimates. These site-specific data limitations will be
common for communities or agencies seeking to estimate ecosystem goods and services to support
decision-making. Inappropriate transfers of measurements or models can lead to inaccurate estimates of
ecosystem services (or other environmental or benefit metrics) and expose decisions based on those
estimates to be challenged (McGarity and Wagner 2003). While benefit-transfer methods exist to transfer
economic measurements and models (Johnston and Rosenberger 2010), no similar formal methodology
for transferring ecological (let alone ecosystem services) measurements or models existed until recently.
Here we present an overview of that new methodology.
Approach
The fundamental assumption underlying the transfer of environmental measurements or models is that the
item or process being estimated exists or operates identically at new site as it does at previous sites from
which the data or models were derived. Thus, the basis for determining whether measurements or models
are appropriate to transfer is to ascertain that the new site is identical (or similar enough) to the previous
site with respect to the item or process being estimated. This is accomplished by comparing the values of
biophysical variables that affect the item/process (e.g., the context variables) being estimated at the
previous and new sites, deciding whether the contexts of previous and new sites are similar enough, and
then using the measurements from the most similar previous sites to generate an estimate at the new site
(i.e., by direct application of the data, or by using the data from previous sites to parameterize a model).
This requires that the analyst understands the environmental/ecological processes that are necessary to
produce the item/process of interest (e.g., estimates of ecosystem goods or services), because those
underlying processes drive the selection of the context variables upon which the site similarity assessment
is based.
Prior to starting the transferability assessment, the analyst should determine whether previous
measurements or runs of the model have been made at a sufficient number of sites to compare with the
new site, and determine the feasibility of running the model (i.e., Is it available in a usable format, can the
data requirements be met, is sufficient documentation available?); collectively, these are logistical
constraints. Finally, before starting the assessment, the analyst should determine the acceptability criteria
for the expected accuracy of the to-be-transferred measurements or the performance of the to-be-
transferred model at previous sites similar to the new site. This critical step leads the analyst to decide
whether it is appropriate to use (i.e., transfer) measurements or evaluate alternatives should the expected
accuracy or performance fail the criteria, such as selecting a different metric or model or measuring the
item/process directly.
Example: Estimating Soil Carbon Storage Capacity at a New Site
Consider the need to estimate the carbon storage capacity of soils (CSCS; kg C m"2) in the forested
watershed of Tillamook Bay, OR (assuming there are no existing measurements or modeled estimates
there). One could make that estimate based on CSCS measurements at other locations, or with an
ecological model. But in either case, is it appropriate to use those measurements or models (i.e., to
transfer them) to make such an estimate for Tillamook watershed?
147
-------
Figure E9.1. Major steps in the methodology to assess the transferability of measurements or
models to new sites.
The seven steps of the transferability assessment method outlined in Figure E9.1 are illustrated with an
example. First identify measurements from other sites or select a model. For this example, CSCS
measurements are available at other sites (USDA soil database 2013) or the model CarbOn Management
and Evaluation Tool (COMET; Farm Colorado State University 2013) could be used.
Second, determine that sufficient previous measurements or runs of the model have been conducted (13
sites; Figure E9.2), and that the analyst can run the model; assume these were satisfied for this example.
mi Clinton
Iroquois
ower Arl
UPPer-Che wa
Tillamook^,ackamas. Smith
' T> <>t
South_San 0R Gallatin $
W 10
South_For Summer_La
WY
North_For
NV
UT
1,000
Figure E9.2. Locations of 13 sites evaluated for model transferability in this example.
148
-------
Third, set the acceptability criteria for the expected accuracy/performance of the transferred
measurements or the model. For the performance of COMET-Farm relative to the CSCS measurements,
acceptability criteria were 40% coefficient of variation (CV) for the accuracy for the CSCS
measurements, and an R2 of 0.60 . The accuracy of a model in classifying sites should be greater than
would be expected by chance (>50%).
Fourth, identify context variables for CSCS (e.g., ecotype, soil type, vegetation type, land type, climate
type, forest type, precipitation, temperature, bulk density, soil organic matter, water penetration depth,
available water capacity, canopy cover) and obtain values for each context variable at each previous site
where CSCS measurements were made or COMET-Farm was run (MRLC 2006, PRISM 2017, USGS
2017a, b).
Fifth, evaluate the similarity of previous sites to the new site (Tillamook Bay watershed) based on the
context variables. This can be accomplished using a multivariate classification (classifying sites by
context variables); Classification and Regression Tree Analysis (CART; Death and Fabricius 2000) was
used for this example. Figure E9.3 shows the resulting classification tree revealing Clackamas (Clac),
South Lake Trinity (Sout), and Upper Verde (Uppe) sites as most similar to Tillamook (Till). Similarity
Ecotype (ECO), precipitation (PPT), and air temperature (TMEA) were the main variables driving
similarity in sites.
ECO = HIlJWofl.PIn.WM
ECO = Mon.Pln.W#t
ECO = Mon.Wrf
ECO = W*i
ECO = Mon
TMEA* <7.4
TMEAt >=7.4
TMEAt >= 11
TMEAt >= 8.S
TMEAt < 8.4
II"
Figure E9.3. CART analysis of thirteen sites.
Sixth, collect the CSCS measurements from most similar sites, or run COMET-Farm at those sites, and
determine the accuracy of the measurements (i.e., CV of the CSCS measurements; CV for Clackamas,
South Lake Trinity and Upper Verde=38%, South Lake Trinity= 23%) or the performance of the model
(i.e., R2 of the COMET-Farm runs, relative to validated measurements of CSCS across those sites;
R2=0.99). The accuracy of the CART model in classifying sites was 98%.
Seventh, compare the measurement accuracy or model performance with the pre-defined acceptability
criteria (step 3). For the assessment of CSCS measurement transferability, the CV of measurements was
greater than the acceptability criteria. This indicates that, per the analyst's requirements (e.g.,
149
-------
acceptability criteria), the measurements of CSCS from other sites should not be transferred to Tillamook
Bay. The analyst could then decide to 1) look for more CSCS measurements, particularly from sites that
are ecologically similar to Tillamook Bay watershed, and re-run the assessment, 2) select a model to
estimate CSCS (and run it through a transferability assessment for models), 3) conduct a study to measure
CSCS in Tillamook Bay watershed, 4) relaxing the acceptability criteria (i.e., the study requirements) to
tolerate a greater amount of uncertainty in the extrapolated estimate, or 5) find a different metric that
addresses the project's need and for which more measurements might be available (e.g., soil organic
matter content).
For the assessment of the COMET-Farm model transferability, the R2 of the model at sites similar to
Tillamook Bay watershed was equal or greater than the acceptability criteria R2 (step 3). This indicates
that the COMET-Farm model is likely to perform adequately when applied to Tillamook Bay watershed.
Conclusions and Caveats
Transferability assessment provides a consistent methodology to assess the justifiability for extrapolating
measurements from previous studies to a new site, or to apply a model developed elsewhere to a site of
interest. Furthermore, the methodology compels the analyst to state a priori the study's requirements for
the accuracy of the transferred measurement or the performance of the model. That requirement can be
used to determine whether the transferred measurements or models would be sufficiently accurate or the
model would perform adequately. In so doing, this method provides transparency and consistency in the
decision-making process, whereby the objective basis for deciding how to make an estimate of the
item/process in question can be clearly documented.
It is important to note that transferability assessment only provides an expectation of the accuracy of
transferred estimates or the performance of a model. However, the expectation of acceptable accuracy or
performance is not a guarantee of the accuracy or precision of the transferred measurements or models;
that can only be determined through validation with measures from the new site (Kleijnen 1995).
For More Information
Colorado State University. 2013. COMET-Farm. http://cometfarm.nrel.colostate.edu/
De'ath, G. and K. E. Fabricius. 2000. Classification and regression trees: a powerful yet simple technique for
ecological data analysis. Ecology 81, 3178-3192.
Johnston, R.J., and R.S. Rosenberger. 2010. Methods, trends and controversies in contemporary benefit transfer.
Journal of Economic Surveys 24:479-510.
Kleijnen, J. P. C. 1995. Verification and validation of simulation models. European journal of operational research
82, 145-162.
McGarity, O., W. E. Wagner. 2003. Legal aspects of the regulatory use of environmental modeling. Environmental
Law Reporter News and Analysis 33:10751-10774.
Multi-Resolution Land Characteristics Consortium (MRLC). 2006. www.mrlc.gov
PRISM Climate Group. 2017. http://www.prism.oregonstate.edu/normals/
USDA (U.S. Department of Agriculture). 2013. Forest carbon stocks of the contiguous United States (2000-2009).
Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station.
https://www.fs.usda.gov/rds/archive/Product/RDS-2013-0004
USGS (U.S. Geological Survey). 2017a.
https://water.usgs.gOv/GIS/metadata/usgswrd/XML/muid.xml#Entity_and_Attribute_Information.
US GS. 2017b. https ://catalog. data, gov/dataset/global-ecological-land-units -elus.
150
-------
Appendix E10
Evaluating Consequences of Decision Options
using a DASEES Approach
Background
In a structured decision-making approach, the means objectives can be useful for identifying and
designing management options to achieve the fundamental objectives. Options are the individual actions
that a single decision-maker or group of decision-makers can take. In some contexts, only one type of
option with different categories or values might be evaluated. In environmental management, it is more
often the case that multiple types of options need to be developed for achieving the fundamental
objectives. When multiple options must be considered together in a decision, they can be grouped and
organized into portfolios of options called management scenarios In this case, each option becomes an
element in the scenario to be grouped and combined with other options. Once two or more management
scenarios have been developed from the options, it is important to compare these scenarios by evaluating
each of their potential consequences.
For example, in a wetland restoration context, some useful options might be to transplant wetland
vegetation or to create upstream riparian vegetated waterways for improving incoming water quality (the
latter being a means objective in this case). The different potential levels of these two options (e.g.,
location and acres of vegetated waterways; location, acreage and density of transplanted shoots) can
create a trade-off between restored vegetative function and costs of restoration (the two fundamental
objectives). Combining these potential levels into different management scenarios allows useful cohesive
and quantitative comparisons to select between the two options. This is also known as a portfolio
approach to decision-making.
Approach
The fourth step in a DASEES (Decision Analysis for Sustainable Economy, Environment, and Society)
process (Appendix Al) is to evaluate the effects of decision options on the objectives identified in prior
steps. In situations involving uncertainty, consequences on the fundamental objectives are analytically
examined in DASEES by considering the impacts on the objectives from the options themselves; the
decision recommendations are made in later steps based on the scenarios. In practice, scenarios are a way
of simplifying the thousands of option combinations that must be considered in a complex decision by
developing and evaluating a useful set of options for making a decision. DASEES currently provides two
approaches (tools) for evaluating the potential consequences of the options and for examining
management scenarios: consequence tables and Bayesian networks. The choice of approach is based on
the level of uncertainty on the outcomes and the amount of time and resources the stakeholders are
willing to commit to assess the consequences with each approach.
Consequence Tables
A consequence table is a simple construct where the objectives and performance measures are placed into
rows, the management scenarios are placed into columns and the estimates of the consequences of each
scenario are recorded in the cells. A consequence table allows the direct entry of consequences for each
management scenario. Consequence tables are one of the most frequently used structured decision-
making tools. They allow stakeholders and decision-makers to clearly visualize the impacts of the
scenarios on the objectives. Scenarios that are clearly inferior can easily be identified, along with
scenarios that are the top performers and why (Gregory et al. 2012). Within DASEES, outcomes can be
either categorical measures (Fig. E10.1) or continuous measures (Fig. E10.2).
151
-------
(5P Define Consequences
0 Save Revert Reset Consequences
Strobl Marsh
Schlepps
Lane Marsh
Gleason's Field
Canyon Marsh
Black Rock Slough
0.4 0.5 0.6
Prioritization Score
Clean Water -v River water (Minimize Recontamination from Flooding)
Scenario Low
Strobl Marsh
Schlepps
Lane Marsh
Gleason's Held
Canyon Marsh
Black Rock Slough
| Cleanup Costs (Minimize Cleanup Cost)
H Completion Time (Minimize time to completion)
Feeding Habitat (Maximize Feeding Habitat)
Habitat Sediment Lead (Minimize Habitat Lead Concentration)
H Human Lead Exposure (Minimize Recreational Sediment Exposure)
j Manage water depth (Optimize Wetland Water Depth)
ฆ O&M Cost (Minimize O&M costs)
11 Project efficiency (Maximize project efficiency)
Minimize Rocontamination from Flooding
Minimize Cleanup Cosl
Minimize time to completion
Maximize Feeding Habitat
Minimize Habitat Lead Concentration
Minimize Recreational Sediment Exposure
Optimize Wotland Water Depth
Minimize O&M costs
Maximize project efficiency
Maximize recreational opportunities
Minimize time to start
Ensure technical feasibility
High
Value Function (Canyon Marsh)
1.0
0.9-|
0.8
0.7-
0.6-
ฉ
| 0.5-
0.4
0.3-
0.2
0.1 -
0.0
1.00
0.50
0.00
Clean Water -v River water
Figure E10.1. Example of a consequence table in DASEES, displaying the effects of different
scenarios (e.g., Strobl Marsh, Schlepps, etc.) on fundamental objectives (e.g., minimizing
recontamination from flooding) where variables are categorical measures. Value functions for each
objective are combined to develop a composite prioritization score for each scenario (Appendix Fl).
OP Consequence Table
jSj Save ฃ Revert Reset Consequences
Status Quo
Sfl-iecfcve implementation -
Dredge
Coral cover
I Powe< Geoefabon
o.o o.i o.2 o.:
Coral cover (MinimizeCoral Reef Impacts)
Scenario
Status Quo
Seiectsve Implementation
Dredge (8sflB
H Power Generation (Maxtaitxei Power Generation)
Scenario
Status Quo g
Selective Imptementalaon
Dredge
0.4 0.5 0.6
Scenario Weighted Vakw
0-1000 (ha)
value Function (Selective implementator
lOOQO-lSOOO (1CVA)
14800 "
Powปr Gsrieration (kV
Figure E10.2. Example of a consequence table in DASEES, displaying the effects of different
scenarios (e.g., status quo, selective implementation, dredge) on fundamental objectives (e.g.,
minimizing coral reef impacts, maximize power generation) where variables are continuous
measures. Value functions for each objective are combined to develop a composite prioritization
score for each scenario (Appendix Fl).
152
-------
Bayesian Networks
A Bayesian Network is a probabilistic graphical model that represents the causal relationships between
physical states and their associated conditional dependencies displayed using a directed acyclic graph. In
DASEES, these graphical models capture how decision options can impact physical states, resulting in
changes to fundamental objective measures.
DASEES provides a Bayesian modeling tool that could accommodate and integrate any environmental
modeling approach into the decision analysis framework (Fig. E10.3). This Bayesian approach provides a
formal framework for merging knowledge and data. Advantages of this approach include:
supplies a rigorous framework for including modeling uncertainty in decisions,
promotes efficient allocation of study resources: start with simple models, iteratively
include more complex models as required by the decision; the level of model complexity
is driven by the value of information provided by the inclusion of additional variables,
allows the user to build a causal network with uncertainty characterized through discrete
probability distributions.
l^P Bayesian Network
y Save ฃ Revert ^ Update + Add Node ^ Settings - Q Status Quo ป
a Scenario Comparison
Dredge -
Selective Implementation -
Status Quo
Maximixe Power Generation
Minimize Coral Reef Impacts
1 1 1 1 1 1 1 1
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Scenario Weighted Value
0.9
I
1.0
Subsidize shade grown coffee
0(taj
Dredge reservoir
4000000 (cubic mctcr^)
Catchment Sediment Traoping
Reservoi' Volume
Power Generation
Coral cover
10000.00 to 11250.00
0.C0 to 25>Q.o:
11250-00 to 12500.00
I"" 1
12S00.00 to 13750.00
13/50.00 to 15000.00
Figure E10.3. Example of a Bayesian Network in DASEES to evaluate the probabilistic impacts of
decision options (e.g., dredging, subsidizing shade grown coffee) on fundamental objectives (e.g.,
power generation, coral cover).
153
-------
Currently, the DASEES Bayesian Network tool uses discretized distributions (and potentially loss of
information). This simplification has several advantages:
quicker consequence assessment using expert judgment, and
flexibility in exploring evidence with both forward and backward uncertainty propagation.
The Bayesian Network tool is where the user will see how changes in values of variables that are not
controllable by the decision but are important inputs to the system (e.g., precipitation, water stratification)
along with proposed management scenarios might affect the outcomes on the fundamental objectives
(Fig. E10.3).
Take Home Messages
Not all actions are going to yield the same results. Some management scenarios will provide better results
than others. It is important to clearly understand how these management scenarios compare to each other,
what the up sides and down sides are for each action so the level of risk can be assessed. The management
scenario adopted will depend on what each of the management options in the scenario produces as well as
what level of risk for negative impacts is posed from each action. Stakeholders will weigh their risk
tolerance for the uncertain outcomes that could be realized from pursuing any particular management
scenario. Coupling the values of the stakeholders with the factual information from the predicted
consequences will lead to the selection of the best management approach to support the public well-being
in environmental management decisions.
For More Information
Gregory, R., L. Failing, M. Harstone, G. Long, T. McDaniels, D. Ohlson. 2012. Structured Decision Making: A
Practical Guide to Environmental Management Choices. John Wiley & Sons
154
-------
Appendix F
Tools and Approaches to Evaluate Trade-offs
and Monitor Outcomes
F4: Using DASEES to Implement
Adaptive Management and
Define Triggers
Fl: Measures Preferences and
Trade-offs in DASEES
F2: Evaluating Trade-offs with
Rapid Benefits Indicators (RBI)
F3: Weighting Components of
Well-being
Clarify
Decision
Context
Implement,
Monitor,
and Review
Define
Objectives
Evaluate
Trade-offs
and Select
Develop
Alternatives
Estimate
Consequences
155
-------
Appendix F1
Measures Preferences and Trade-offs using a
DASEES Approach
Background
In a structured decision-making (SDM) approach, objectives and their associated attributes (measures) are
weighted relative to each other for trade-off analysis when competing alternative options are considered
(Gregory et al. 2012). Objective preference is based on the relative importance stakeholders hold for each
objective depending on the decision context. In DASEES (Decision Analysis for Sustainable Economy,
Environment, and Society; Appendix Al), objective preference is performed after establishing the
Objective Hierarchy in Step 2 (Appendix CI) when the reasons for the objectives are still fresh in the
stakeholder's minds.
Measures Preferences
Immediately following the creation of an Objective Hierarchy in DASEES objective preference is
developed (Appendix CI). While all identified objectives bring valuable information to the decision
process, not all objectives are deemed of equal importance or weight. Stakeholders will have views on
which objectives are most important compared to others. Objective are first ranked ordinally as a group,
with each successive measure placed after the most important according to the importance agreed upon by
the stakeholders. The next step is to determine relative preference between two adjacent objectives. This
is important because it cannot be assumed that the relative preference between any two adjacent
objectives is the same.
At this point relative preferences are all of equal value (Fig. Fl.l). This is where weights can be assigned
to these adjacent objectives by doing paired comparisons starting with the two lowest ranked objectives.
The initial ordinal ranking assigns an equal weight to each objective equal to 1/# of objectives. For
example, if there are five objectives the default weight is 0.2. Thus each adjacent pair of objectives is
assumed to have a 1- to -1 ratio of preference.
Figure Fl.l. Screenshot from DASEES indicating equal preference weights on three objectives.
156
-------
Using the preference ranking wizard in DASEES, stakeholders can then adjust that relative preference
from 1-to-l to something higher, such as 2-to-l or something else (Fig. F1.2). Once the relative
preference assignment is done for the bottom two objectives, the process is repeated up the list by pairing
the next two objectives (i.e. second from the bottom paired with third from the bottom) and repeating the
assignment of a relative preferences.
Figure F1.2. Screenshot from DASEES indicating different preference weights on three objectives.
The wizard normalizes the preference weightings to 1 to be consistent with value function normalization
(Fig. F1.3). These weightings are subsequently used in the consequence table (Appendix E10) in
conjunction with value function weights to represent their adjusted importance in the overall consequence
comparisons.
Ofcllwi
(ซWa | U (tMtTlKM
Otfxt >
j Mwkk g M *ป
Cซซcm> wont cm* I NKemiW
* T|fWปr HM Mil (MI
NtKmlMOO
Figure F1.3. Screenshot from DASEES indicating calculated preference weights on three objectives.
Evaluating Trade-offs
Once the consequence table is constructed and all the appropriate weightings have been applied, the effect
of changing values of the measures associated with the objectives can be seen by isolating each of the
measures and looking at them over the scenario options that are being considered. This will provide a
snapshot of how much each measure contributes individually to each scenario and how the measure may
change depending on the scenario that is being considered (Fig. F1.4).
157
-------
fMMttMa Vera
Figure F1.4. Screenshot from DASEES comparing prioritization scores across three scenarios
(status quo, selective implementation, or dredge) for each of three objectives - maximizing sediment
removal (green), minimizing coral reef impacts (blue), and maximizing power generation (red).
When all the measures are combined for all the scenarios you can see the overall priority score for each
scenario as it reflects the impact on each of the measures. This provides a graphic demonstrating the
scenario option that provides the overall best score and how the other scenario options compare to the
overall best score. From the graph presented it is clear that the Status quo scores markedly less than either
of the other two options. So the Status Quo would most likely be dropped from consideration.
But the other two scenario options appear to rank fairly closely in terms of score. This is where the
potential trade-offs become more difficult to make, and a balance between stakeholder values, needs and
wants will be discussed in terms gains and losses (Fig. F1.5). For example, is it more important to gain 40
ha of coral cover at the expense of a 200 NTH (Nephelometric Turbidity Unit) increase in water turbidity
and 3000 kva of power generation, or is it acceptable to sacrifice the 40 ha gain in coral cover to achieve
significantly clearer water and a 25% increase in power generation? Unless, for example, there are legal
regulations that mandate the coral cover increase at no less than the 690 ha which would make the
tradeoff consideration a moot exercise, only the collective stakeholders can decide which approach is best
to meet their values and needs.
Figure F1.5. Screenshot from DASEES comparing effects of three alternative scenarios on
measured objectives.
158
-------
Take Home Message
Weighting objectives and determining what aspects you are willing to give up to gain in other areas is
very much a bottom up, stakeholder values driven exercise. Unless there are legal requirements or
regulations that supersede the stakeholder values, decisions based on what is more important than
something else and what one is willing to give up in one area to gain in another area, will more likely be
successful because all stakeholders have ownership in the outcome.
For More Information
Gregory, R., L. Failing, M. Harstone, G. Long, T. McDaniels, D. Ohlson. 2012. Structured Decision Making: A
Practical Guide to Environmental Management Choices. John Wiley & Sons
159
-------
Appendix F2
Evaluating Trade-offs with Rapid Benefits Indicators (RBI)
Background
The RBI approach is an easy-to-use process for assessment based on non-monetary benefit indicators. It is
intended to be used in conjunction with existing ecosystem service assessment approaches and tools to
connect changes in the availability of Ecosystem Goods and Services (EGS) to where and how people
benefit from those goods and services. The approach was originally developed for use with urban
freshwater wetland restoration, but the general approach and indicator framework can be adapted to work
with other types of environmental changes and within different ecological systems.
Approach
The approach is a five step process, starting off with identifying stakeholders and their objectives. Based
on these objectives relevant ecosystem services and resulting benefits are selected to compile indicators
based on five questions outlined in Appendix C4.
Indicators for five benefits (Appendix C4; Fig. C4.1) have been previously developed and integrated into
two tools that help users more easily apply the Rapid Benefit Indicators approach. The Rapid Benefit
Indicators (RBI) checklist tool can be used for recording results of manual or field analysis, and the Rapid
Benefit Indicators (RBI) spatial analysis tools can be used to compile indicators based on spatial datasets.
Both of these tools summarize results into a similar table with color coding to emphasize trade-offs across
wetland restoration sites. The results table and color coding does not aggregate indicator results into a
single solution or list of priority sites. Instead, weights such as those described in Appendix Fl, or a
discussion based stakeholder engagement process should be used to go through these results and select
the best sites for the benefits that are most important to stakeholders in the decision context. Using this
results table is the emphasis of this appendix.
Example: Evaluating Indicator Results
For this example, we assessed the five benefits provided by wetlands restoration on Hillsborough County
Environmental Lands Acquisition and Protection Program (ELAPP) 2011 land holdings north of Tampa
Bay. This assessment was done solely to demonstrate the application of Rapid Benefit Indicators. The
indicator approach questions and the datasets used for this analysis are both outlined in Appendix C4.
Analysis was performed using the Rapid Benefit Indicators (RBI) spatial analysis tools. For more details
on the specific indicators and their assumptions see Mazzotta et al. 2016.
An example Indicator Summary Table Results page from the assessment is shown in Figure F2.1. Cells
that are Gray were designated as Not Applicable (NA) in the example indicators set. Cells that are black
(e.g. 3.5 Reliability) were not assessed using the spatial analysis tools and lack results to consider in
decision making. Cells in red are less preferable for that indicator metric whereas cells in blue are better.
If all qualitative indicator results were "No" (e.g., dams and levees 2.5 mi downstream) the indicator can't
be used to differentiate the sites and appear in gray. Removing gray and black cells from consideration
reduces the results to 19 quantitative and five qualitative (Yes/No) indicators.
160
-------
Summarize the Indicators
Benefit
Indicators
3.2 now Mam* Benefit
5 mi downstream of site and in flood zone
Area of ressoranoa site (acres)
3.3. A Service Quality
reatures that urease retennoo volume?
Danii and levees 2 5 nn downstream?
3.3.B Sentor
Wetlands within 2_5 mi (percent area)
3.3.C Complements
3.3-D Preferences
Are peopje worried about flood nsk
Number within 160 ft of site
Number within 160o2> ft of site
3.2 How Many Benefit?
Weighted number who benefit
Are there roads or trails within .*25 ft of site
3.3. A Service Quality
Aesthetic features or characteristics?
: : 5 Scanitv
wetlands or water withm 6>0 ft (percent area)
3.3.C Complements
Natural land use types within <550 ft (types)
-V 3 D Preferences
Will people find it aesthetically pleasing
3.2 How Manv Benefit?
Education institutions withm 0.25 mi of site
3.3. A Service Qualitv
- res - > eft ?e f ec ฆ
: ; r> :it*.
Wetlands within 0.5 mi of the site (percent area)
3.3.C Complements
:uxucanonal facilities or infra-trueture on site
; ; I: ?ie:e:ei::e-.
will people prefer charactensacs of the site?
Number within 1. 3 mi of the site
Are there bike paths within 1 3 mi of site?
3.2 How Mam- Benefit"*
.Are there bus stops within 1 3 mi of site?
Number within 0 to 0 ? rni of site
Number within 0 5 ป 6 mi of site
60.S47
3S.549
32.23S
3.3. A Service Quality-
local area of green space around site
Green space within 2/3 mi of site
3.3.B Scarcity
Green space within 1 mi of site
Green space within 12 mi of sate
3.3.C Complements
Infrastructure supporting recreational activities?
-V3.D Preferences
.Are there addmceai features on tae site
Number within 0 2 mi of site
3.2 How Manv Benefit?
Are there roads or trails within 0.2 mi of site
3.3. A Service Quality
Will the site support rare cc unique species?
3.3.B Scarcity
3.3.C Complements
Supporting infrastructure or habitat on site?
33J) Preference:,
Will people be interested in birds at the site?
3.4 Social Equitv
3.5 Reliability
3LACK = No entry*: GRAY=NA; = Above Average Yes; RED = Below Average No (reverse for scarcity)
Figure F2.1. Indicator summary results page for 4 of the 8 restoration sites assessed in Tampa Bay
The color coding makes it easier to make trade-offs for a single benefit indicator across sites. For
example, the number of people who benefit from flood risk reduction is better at sites 1, 4 or 8 than site 2.
When reviewing results, it is important to remember color coding is based on the average of all sites, not
just the sites visible on that page.
Making trade-offs between indicators within an indicator category may be facilitated by color-coding, e.g.
for scenic views there are 4 indicators within the "3.2 Now Many Benefit?" indicator category, but it is
easy to see site 8 is better than site 2 across all these indicators. Another way to facilitate trade-offs within
an indicator category is to use weights, ensuring that more weight is given to indicators that are more
influential. This was done to aggregate the number of houses within 160 ft, and the number within 160-
325ft of the site, into the single "weighted number who benefit" indicator (see Mazzotta et al. 2016 for
details). This makes it easier to tell site 8 is better than site 4. In other cases, such as with "weighted
161
-------
number who benefit" and "are there roads or trails with 325ft of the site", it is more difficult to combine
indicators and stakeholders should be consulted to determine the aspects of the benefit indicator category
that are more important.
Making trade-offs among indicator categories for a single benefit requires weighing the importance of
different indicator categories, based on importance of the different factors to relevant beneficiaries. For
example, for flood risk, site 4 has fewer who benefit, but greater service quality than site 1.
As with trade-offs among indicator categories, making trade-offs among benefit categories requires
weighing the importance of different benefits to the relevant stakeholders. For example, site 8 seems to
provide better scenic view benefits, but site 1 provides better recreation benefits.
Trade-offs may also be made among the groups of people who benefit, for example based on social equity
characteristics. For example, site 1 appears to be the only one of these four sites to provide benefits to
people in highly vulnerable populations.
Lessons Learned
The results shown here were for demonstration purposes only, and an actual application requires more in-
depth stakeholder engagement and consideration of decision context when choosing relevant benefits and
restoration sites to consider. Although the summary report generated using the RBI tools provides useful
color coding, it does not generate aggregated summary scores or a prioritized list. Decision makers can
use the color coding to further reduce indicators and in some cases may be able to make trade-offs
between sites, but additional techniques are required to make final decisions.
For More Information
Mazzotta, M. J. Bousquin, C. Ojo, K. Hychka, C.G. Druschke, W. Berry, R. McKinney. 2016. Assessing the
Benefits of Wetland Restoration: A Rapid Benefit Indicators Approach for Decision Makers. U.S.
Environmental Protection Agency, Narragansett, RI, EPA/600/R-16/084.
162
-------
Appendix F3
Working with Stakeholders to Weight Components of Well-being
Background
A key tool for the use of a systems approach in community decision-making is the development of clear
measurable objectives that allow for the assessment of trade-offs among key priorities. Objectives can be
separated into three general categories: Successful implementation of a decision (e.g., park built),
immediate impact of the decision on services (e.g. target number of people use park), and impact of
decision on community well-being (e.g., wellness/activity indices increase). The impact of decisions on
community well-being is the objective most closely aligned with a systems approach to decision-making.
It is also the hardest objective to define and measure effectively. This is true because every community is
different and while it may be easy to define the basic components of well-being (Health, Sense of Place,
Basic needs met), stakeholders have very different ideas of what is most important, making defining
trade-offs particularly difficult. Here we introduce a structured approach to help stakeholders clarify and
rank well-being domains with the objective of developing measurable outcomes, so that community
decision-makers can make use of these concepts to compare the impacts of different decision options.
Approach
This approach combines two well-established tools: Structured Decision Making (SDM, Gregory et al.
2012) for stakeholder engagement and the Human Well-being Index (HWBI, Smith et al. 2013) for
measuring well-being. We used an SDM workshop approach to define and rank fundamental objectives at
the community level and then tied those fundamental objectives to the domains of the HWBI. The HWBI
is comprised of eight domains of well-being roughly corresponding to social, environmental, and
economic well-being categories (Fig. F3.1; Smith et al. 2013).
HWBI
Leisure
Time
Connection
to nature
Cultural
Fulfillment
Safety and
Security
Education
Health
Living
Standards
Social
Cohesion
Figure F3.1. Eight domains of the Human Well-being Index (HWBI, Smith et al. 2013).
As a part of the workshop stakeholders are asked to list issues and opportunities in their community
(additional details in Appendix C5). These issues and opportunities are then redefined by the ecosystem
services that support them and these services are mapped to the domains of HWBI. These connections
from issue to service to HWBI domain are all based on stakeholder input and minimal effort is made to
guide the discussion in any way. Facilitation is provided to maintain stakeholder focus and maximize
understanding of the reasons for making these connections. Once the connections have been made the
stakeholders are then asked to rank the domains of HWBI in terms of importance to their community.
This ranking is conducted both as a group exercise (dot voting) and as an individual exercise (Ranking
domains 1 to 8).
163
-------
Three key outcomes occur from the SDM workshop. First, stakeholders increase their understanding of
human well-being as a target for decision-making. Second, the relative importance of the individual
domains of HWBI represent a community-specific definition of human well-being. Third, the mapping of
issues to services to well-being supports the identification of community specific indices of success that
can be used to assess potential trade-offs in progress towards well-being objectives. This final outcome is
critical to facilitating a shift to a systems approach to decision-making.
Case Study Example: Vero Beach, FL
As previously described in Appendix C5, the city of Vero Beach is a small beach town on the east coast
of Florida. The residents of Vero Beach were interested in a revitalization of the downtown corridor to
improve pedestrian safety and foster an investment in the arts. Researchers worked with a group of local
community leaders and neutral facilitators to organize a one-day SDM workshop, which was attended by
32 stakeholders. The planning process for the workshop included an examination of stakeholder interests
in Vero Beach and an effort to assure that as many as possible were represented at the workshop (Fig,
F3.2).
Figure F3.2. Stakeholders discussing issues and opportunities at the beginning of a workshop.
The workshop discussion was initiated with a small group exercise centered on the question "How would
you describe your community ?"' The results of this exercise were adapted into a list of strengths and
weaknesses and then into services either provided or desired from the community resources. Once this
was complete the HWBI domains were introduced by the facilitator, followed by another small group
exercise to map the list of services into the HWBI domains. Through both small and large group
discussion the stakeholders sought a consensus on how the individual services contribute to human well-
being. The results of this exercise are shown in Table F3.1.
164
-------
Table F3.1. Results from mapping exercise used to connect stakeholder identified decision
opportunities in Vero Beach, FL to the domains of HWBI. Checks indicate participants identified a
given domain as relevant to why the decision opportunity was being proposed. Decision
opportunities most directly related to ecosystem services are in bold (from Fulford et al. 2016).
Why is this being proposed?
What we care about in a community
Social
Cohesion
Health
Cultural
Fulfillment
Living
Standards
Safety and
Security
Education
Work-life
Balanace
Connection
to Nature
Access to a wide variety of activities, many at no charge
V
V
Access to beach, rivers at night
/
Accessibility to natural resources
V
/
Active philanthropic community
V
V
Aesthetic (beauty, uniqueness, nature)
/
V
V
Airport (availability of)
Alternative modes of transportation
V
Balance of Cultural/Nature
V
V
Better accessibility to multi-modal transportation systems
V
V
V
/
Churches
/
Cultural diversity
V
Diversity of choices - geography, dining, shopping,
entertainment, attractions, residences, job opportunities
(need more)
V
Diversity of demographics
V
Ease of mobility
V
V
V
V
Education
/
V
Environment (clean air/water)
V
V
Expand "clean job" opportunities
V
V
V
Expand higher education
V
/
V
Find out "who we want to be"
V
Friendly town and people - human scale connectivity
V
Gathering places
/
V
Healthcare
V
Healthy environment (air, water)
/
V
History
V
Human-scale development
V
V
V
Interaction between diversities with philanthropic groups
/
/
Jobs for young professionals so children do not move away
/
V
Listening government - open in discussion
V
Medical services
V
More affordable activities
V
V
More job diversity (all service jobs now)
/
V
More sidewalks
V
V
More visual charm
/
V
V
More walkability - less asphalt
/
V
V
No sprawl
/
Off road bike path (along main relief canal)
V
/
V
Open City Government
V
Philanthropy
/
V
V
V
Places to eat, meet, walk and talk
/
/
/
Preserve access to natural resources (e.g., beach)
V
/
V
165
-------
Why is this being proposed?
What we care about in a community
e
#o
"s s
0>
e &
'3 -S
o ฎ
cป O
et
0>
m
a
s 3
r
> 55
J cn
ZJ S
a qj
CZi Xfl
3
w
o -s
s Z
o 3
Preserve beauty, resources
V
V
Public parks and events / safe beaches
V
/
Relaxed lifestyle
/
Responsive and caring local government
Safety
S
Safety (low crime)
V
Slower pace - "we are old school, yet modern"
/
V
Small town feel
Stop sprawling
V
V
V
Traffic control in cultural districts
Walkable, bikeable (need safer bike paths)
V
The second stage of the workshop involved the ranking of domains of HWBI. The domains were listed on
large sheets of paper and the participants were given eight dot stickers each and asked to place their dots
next to the highest priority domains. The dot voting was done as a group so that everyone could see the
outcome. The participants were then asked to compete a ranking sheet individually at their seat by placing
the eight domains in order from most to least important. The outcome of these two ranking exercises are
shown in Table F3.2.
Table F3.2. Percent of participants in three Vero Beach workshop exercises identifying a given
domain as important in the exercise mapping priorities, as a group by dot voting, and individually
by ranking. Top three domains by each method are in bold.
HWBI Domains
Mapping of
priorities
Dot voting
Ranking
Education
8%
16%
13%
Health
17%
21%
17%
Work Life Balance
10%
15%
6%
Living Standards
10%
13%
16%
Safety and Security
13%
8%
22%
Connection to Nature
14%
7%
6%
Cultural Fulfillment
10%
7%
6%
Social Cohesion
18%
14%
14%
Stakeholders connected the most services to the HWBI domains Social Cohesion, Connection to Nature,
and Health. When asked to rank the domains, they gave the highest rank to Education and the second
lowest rank to Connection to Nature. This indicated a disconnect between provided services and
stakeholder priorities that will need to be accounted for in goal setting.
166
-------
The third phase of the workshop involved the application of well-being objectives to the issue of
downtown revitalization. Here the discussion becomes practical as ideas were suggested including a
reorganization of the traffic signals to reduce vehicle speed and favor pedestrian access and safety. Ideas
were suggested for measuring the impact of these suggested actions on the domains of well-being with the
highest stated priorities in the community. The final report for the workshop included a spreadsheet to
guide stakeholders in choosing metrics that best inform human well-being improvements.
Lessons Learned
Both the main objectives of the workshop were achieved: stakeholders gained understanding in
developing measurable objectives based on human well-being, as well as a community specific
examination of such objectives for a specific relevant issue. A one-day workshop does not provide
adequate time to complete the process but does serve as a catalyst for continued effort. The organizing
group for Vero Beach followed up with a series of design charrettes for downtown planning that included
use of well-being concepts as objectives. A key element for success is the involvement of a representative
group from all major stakeholder interests in the community. This was achieved through planning and an
effort to be inclusive at the interest group rather than the individual level. Follow-up is also critical as the
well-being approach must be made useful for decisions of local interest. Finally, the structured approach
is intended to be transferable both across communities and across issues. However, transferability should
not supersede local utility, and measures of success should account for differences in how stakeholders
rank the domains of human well-being.
For More Information
Fulford, R., M. Russell, J. Harvey, M. Harwell. 2016. Sustainability at the community level: Searching for common
ground as a part of a national strategy for decision support. U.S. Environmental Protection Agency, Washington,
DC, EPA/600/R-16/178.
Gregory, R., L. Failing, M. Harstone, G. Long, T. McDaniels, D. Ohlson. 2012. Structured Decision Making: A
Practical Guide to Environmental Management Choices. London, UK: Wiley-Blackwell.
Smith, L.M., J.L. Case, H.M. Smith, L.C. Harwell, J.K. Summers. 2013. Relating ecoystem services to domains of
human well-being: Foundation for a US index. Ecological Indicators 28: 79-90.
167
-------
Appendix F4
Using a DASEES Approach to Implement Adaptive Management
and Define Triggers to Monitor Achievement of Objectives
Background
Once an approach for addressing an issue or problem has been decided upon it is time to implement the
proposed actions. By bringing together a diverse group of stakeholders that include experts as well as lay-
people, the proposed action plan should have target values for measures that if achieved are anticipated to
produce the end results desired by the group. But what if the approach is implemented and the proposed
actions do not seem to be achieving the desired end goals? The process needs to be evaluated, and
adjustments or modifications need to be implemented in an adaptive management approach. Adaptive
management is often described as "learning by doing", and its cyclical structure allows for improving
management based on the outcomes of monitoring and evaluation of decisions and policies (Williams and
Brown 2012). Outcomes can and should be monitored while the management actions are being
implemented. A set of decision points, or triggers, need to be defined where evaluations of the progress
and success of the actions will be evaluated, such that course corrections can be made within a reasonable
time period. The DASEES (Decision Analysis for Sustainable Economy, Environment, and Society;
Appendix Al) approach incorporates an adaptive management strategy that hinges upon setting triggers
as benchmarks. The progress and effectiveness of the selected management actions will be evaluated to
determine if adjustments need to be made to achieve the desired end results.
Approach
In the DASEES approach, adaptive management triggers can be defined for chosen measures and state
nodes. The trigger point (both target date for evaluation and desired level of measurable achievement) is
based on what the stakeholders identify as a reasonable time frame to achieve a desired level of
performance for the monitored measure or state (Fig. F4.1). Once this is set then data for these measures
or states are collected and uploaded to the DASEES platform as they become available and are associated
with the respective measure or state node. Graphical presentations of the data with the dates and the
trigger points are generated to allow progress evaluation.
Tnpfซr
CM*
00/22/2014
CXU
Oatซ
U*i
04717014
1140
OV217014
20040
20000
07/U7414
9
>0000
ttOCC
0412
1*72
7ซH
>j:ซ
tปce
4*0 60
400 00
i 400 00
V
I
u
200 00
CfcM
-------
Triggers can be set both as positive (desired attainment of a measure within the prescribed time frame is
achieved so implemented management option continues to be followed as implemented; Fig. F4.1) or
negative (desired attainment of a measure within the prescribed time frame is not achieved so
implemented management option must be re-evaluated and adjustments should be made going forward;
Fig. F4.2). Pro-actively setting trigger points for a management plan will ensure that the progress and
attainment of the end goals are evaluated on a prescribed time frame. Any needed changes can then be
made efficiently and in a timely manner in order to facilitate the achievement of the Decision Objectives.
(ซrซl tiwr
Drvftptaw lUM^rnrW
C
OtU
9V211
OV220OH
V/2ltt14
9WX 14
W32/20U
icoe
200 40
2*04 CC
WM
*mco
JV3 oc
m
5 20000
100 00
U i >14
HUM
Figure F4.2. Screen shot from DASEES showing a trigger point (green) that observational data
(orange) indicates was not attained within the prescribed time frame.
Take Home Message
When management and action plans are decided upon and set in motion they should not be considered as
static entities that are set in stone. Setting triggers at pre-decided time frames to evaluate the progress of
the plan is necessary to ensure that either the plan is working as intended, or is not achieving the desired
results in the desired time frame. The process of adaptive management should be implemented to evaluate
what is impeding the attainment of the desired measures at the desired times. Without set evaluation
points the management plan is more vulnerable to failure if needed adjustments in the approach are not
recognized and course corrections adopted in a timely manner.
For More information
Williams, B.K., and ELD. Brown. 2012. Adaptive Management: The U.S. Department of Interior Applications
Guide. Adaptive Management Working Group, U.S. Department of the Interior, Washington, DC.
169
-------
Appendix G
Case Studies
l i >33
fir*
mm
M
mm:
9 -
Great Lakes
ZP
Mobile Bay,
Alabama
Implement,
Monitor,
and Review
Pacific
Northwest
San Juan,
Puerto Rico
Southern
Plains
Clarify
Decision
Context
v y
Evaluate
Trade-offs
and Select
Estimate
Consequences
Define
Objectives
Develop
Alternatives
170
-------
Appendix G1
Great Lakes Areas of Concern Case Study
Background
The EPA Area of Concern (AOC) program began in the late 1970s and is an early example of an
ecosystem-based approach founded on the maintenance of ecosystem integrity and recognition of human
use of and benefits from nature. Great Lakes Areas of Concern were established in response to a crises of
legacy contamination of heavy metals, polychlorinated biphenyls (PCBs), and dioxins, as well as
combined sewage overflows and storm water runoff in Great Lakes coastal communities. Annex Two of
the 1987 Protocol to the US-Canada Great Lakes Water Quality Agreement of 1978 set in motion a new
approach to the bi-national clean-up of the Laurentian Great Lakes. In the 1987 protocol, Areas of
Concern were defined as "geographic areas that fail to meet the general or specific objectives of the
agreement where such failure has caused or is likely to cause impairment of beneficial use of the area's
ability to support aquatic life."
The vision was to restore the beneficial uses of the aquatic ecosystem that had been impaired in the most
degraded sites within the Great Lakes, particularly industrial and population centers along the Great
Lakes shoreline. In all, 43 Areas of Concern (AOC) were identified in Canada and the US. Today, 27
AOCs remain on the US side of the Great Lakes (Fig. G 1.1). Superfund sites may be located within an
AOC, and brownfields are commonly located nearby. Federal funds administered by EPA Great Lakes
National Program Office under the Great Lakes Legacy Act (GLLA) and Great Lakes Restoration
Initiative (GLRI) provide funding for sediment remediation and aquatic habitat restoration in AOCs, In
some cases, economic revitalization has been a desired consequence of those activities.
St Louis River
Great Lakes Areas of Concern
Manistique River
St. Marys River
^ Delisted before GLRI
^ Delisted during GLRI
Management actions completed
during GLRI Action Plan i
Management actions targeted for
completion during GLRI Action Plan II
^ Remaining Areas of Concern
Menominee River
Fox River/ ^
Lower Green Bay
Sheboygan River^^
Milwaukee Estuary^
Waukegan Harbor A
Saginaw River and Bay
ite Lake ^
Mu skegon Lake St. Oair River
Clinton River
Kalamazoo River
River Raisin
Grand Calumet River Maumee Ri^
St. Lawrence River
^ Detroit River
^Rouge River ^ Tresque Isle Bay
* ^ Ashtabula River
5L c,
Eighteen Mile Creek
f WOsi
Niagara River 0| Rochester Embayment
Buffalo River
wego River
Black Ri^r Cuyahoga River
October 30,2014
Figure Gl.l. US and US-Canada Great Lakes Areas of Concern indicated by current status with
respect to the Great Lakes Restoration Initiative. (Source: US EPA, updated October 2014).
171
-------
The governance structure of the AOC program as established under the bi-national Great Lakes Water
Quality Agreement is polycentric, comprised of federal, tribal or First Nation, state, and local agencies
working with local stakeholders through a Public or Citizen Advisory Committee. Ultimately, de-listing is
approved by EPA Great Lakes National Program office. The foundation of the AOC program is that Great
Lakes coastal ecosystems provide beneficial uses for humans such as drinking water, clean sediment, and
fish to eat. Beneficial use impairments (BUIs) were established for environmental problems such as beach
closures, fish consumption advisories, dredging restrictions, and excess nutrients and sediment (GLWQA
1987). These beneficial use impairments identified by stakeholders within Great Lakes coastal
communities are ecosystem services (Angradi et al. 2016). Ultimately, 14 possible BUIs were identified.
Most AOCs identified the presence of a subset of those 14 possible BUIs, though a few identified all 14
impairments.
Decision Context
This case study has been working with the EPA Area of Concern (AOC) program to incorporate
ecosystem services into decision-making. The goal of an AOC is to remove identified BUIs through
sediment remediation, water quality improvements, and aquatic habitat restoration. AOCs are responsible
for identifying the management actions that are needed to remove BUIs (e.g., identify sites for
remediation, establish goals for combined sewage overflow reductions or nutrient concentration, or
determine the area and type of aquatic habitat to be restored). Ultimately, once all the identified
management actions to achieve removal of the identified BUIs are completed, the AOC is recognized as
having management actions complete. At this point, it may take multiple years for the AOC to observe
improvements from the management actions. In the last step, after the AOC determines that BUIs have
been successfully removed (i.e., the BUI removal targets have been met), the AOC petitions EPA for de-
listing. The AOC program requires that each step (BUI identification, developing and completing
management actions, removing BUIs, and AOC de-listing) involves stakeholder input and participation.
The goal of the research project was to incorporate ecosystem services into decision-making by providing
information regarding how AOC decisions affect ecosystem services, but to do so in a way that preserves
the current, previously existing programmatic targets agreed to through the AOC governance structure
(Fig. G1.2).
SWFs
Ecosystem
mediated
processes
Community
revitalization
R2R
implementation
SWFs
(A) R2R
project
design
A FES
(A) Policies
& decisions
AOC delisted
A Ecosystem
benefits
A Biophysical
state of the
ecosystem
Information for decision-makers:
SPA maps, tradeoff analyses, benefit tracking
Figure G1.2. Conceptual framework for the use of ecosystem service mapping and associated
analysis to support decision-making in an estuarine Great Lakes AOC (Angradi et al. 2016). R2R =
remediation to restoration; FES = final ecosystem services; BUI = beneficial use impairment; AOC
= area of concern; SPA = service providing area; SWF = social welfare function (i.e., ecological
benefit function [EBF]).
172
-------
Define Objectives and Measures
The EPA Great Lakes National Program Office (GLNPO) uses the term R2R2R or R3 to refer to the
process of Remediating contaminated sediments and Restoring aquatic habitat to foster devitalization in
Great Lakes AOCs. In practice, this means that GLNPO recognizes that there is an implicit connection
between improved ecosystem services and the quality of life in AOC communities. To characterize the
linkage between remedial efforts and revitalization, we proposed using the concept of ecosystem services.
The ecosystem services concept is advantageous because it aligns well with beneficial use impairments
(that is, a restricted use of fish, water or sediment is a loss of ecosystem services), Great Lakes coastal
habitats provide many ecosystem services (Sierszen et al. 2012), and because it is possible to quantify
indicators of change in ecosystem services and benefits from a remedial action (e.g., the increased
recreational use of a site after clean-up and restoration).
Our strategy included a number of elements: 1) adapt our work to fit the established governance and
regulatory structure, 2) utilize an iterative data production model designed to make our research products
usable, 3) generate data and final ecosystem services (PEGS) indicators and models at a scale that meets
the needs of local decision-makers, and 4) undertake studies of how changes in ecosystem services are
affecting perceptions and revitalization activities in and around AOCs. As a case study, we worked with
the St. Louis River AOC, which includes the communities of Duluth, MN, Superior, WI, and the Fond du
Lac Band of Lake Superior Chippewa reservation. Through a series of workshops designed to improve
information relevance by producing data with local stakeholders (sensu Beebeejaun 2015, Posner et al.
2016), we provided a forum for state agencies and PAC members to discuss direct and indirect
connections between BUIs and ecosystem services. Researchers worked collaboratively with stakeholders
to build a conceptual model that demonstrates how removing BUIs can lead to ecosystem services and
improve human well-being (see Appendix B4). Workshops were also used to conduct participatory
mapping to identify geographic areas of importance for ecosystem services (Klain and Chan 2012), and
co-develop geospatial ecosystem services production models (Angradi et al. 2016; see Appendix E2).
Estimate Consequences
We considered how a variety of likely AOC management actions would affect PEGS indicators and the
geospatial distribution of PEGS. Generally, for any given management action, there is a trade-off in
FEGS such that some are increased whereas others are decreased (Table Gl.l). In the context of
decision-making, this information would be supplied to stakeholders, who would then provide comments
on the various trade-offs to decision-makers. There is uncertainty at this level of the analysis; for
numerous FEGS indicators, the response (increase or decrease) will depend on the specifics of the
management action. The analysis demonstrated that some management actions are broadly favorable with
respect to both AOC management and FEGS. For example, FEGS indicators generally are neutral or
increase in response to increased public access to the river or sediment remediation. There are, however,
cases where management action could result in substantial trade-offs among services. For example,
increasing shallow water habitat can increase wild rice production, fish nursey habitat, and wave
attenuation, but will decrease the opportunity for boating, fishing (especially ice fishing), and other
services that depend on deep water.
Ethnographic methods (content analysis) were used to evaluate how changes in ecosystem services are
affecting perceptions and revitalization activities in and around AOCs, particularly new trails that are
designed to follow the river corridor along the restored ecosystem (Appendix C2). Based on the content
analysis, community members felt that the trail alignment should maximize safety (avoid traffic
conflicts), connect to other regional trail systems, remain close to the river, and provide a connection to
nature. We evaluated various trail route alternatives with respect to these findings, and found that the
route following an abandoned railroad bed situated along the river would best match the interests of
community members.
173
-------
Table Gl.l. Final ecosystem service by management action matrix for the St. Louis River Area of
Concern (AOC). Example actions anticipated or otherwise realistic projects for the AOC. Expected
effects of management actions on SPA of each FEGS are rows in the table; trade-offs among FEGS
resulting from habitat restoration or other management actions are in columns. For riparian
confined sediment disposal facility (CDF), the assumption is that no aquatic habitat is lost.
Responses: zero = no effect; + = more area of the service is created; - = area of the service is lost; ?=
response depends on context (Angradi et al. 2016).
Create
Re-slope/
shoals
Create
Increased
soften vertical
Remove
(<0.5m)
islands
public
shorelines
contam-
Create
FEGS
outside
channel
outside
channel
river
access
(restore
littoral)
inated
sediments
riparian
CDF
Boat/ice-caught fish
-
-
0
-
+
0
Shore-caught fish
-
0
+
-
0
0
Lake sturgeon
?
-
0
?
0
0
Esocid fishes
+
-
0
+
0
0
Walleye
?
-
0
?
0
0
Colonial waterbirds
0
+
0
0
0
0
Bald eagles
0
+
-
0
0
?
White-tailed deer
0
?
?
0
0
?
Waterfowl
+
?
?
+
0
?
Semi-aquatic fur-bearers
0
?
?
+
0
?
Riparian & semi-aquatic
wildlife
0
?
?
+
0
?
Wild rice
+
-
0
+
+
0
Human-powered boating
areas
0
0
0
+
0
Power boating areas
-
-
0
-
+
0
Power cruising areas
-
-
0
-
+
0
Sailing areas
-
-
0
-
+
0
Seaplane runway
-
-
0
0
0
0
Shipping channels
0
0
0
-
+
0
Wave energy attenuation
+
+
0
+
0
0
Public parks and trails
0
0
?
0
0
?
Public beaches
0
0
+
?
0
0
Native-American sacred
sites
0
0
0
0
0
0
Natural viewscapes
0
-
0
0
0
-
Impact
Trade-off analysis was a useful tool to evaluate management actions (R2R projects, associated changes to
built infrastructure along the river) in terms of ecosystem services. For example, trade-off analyses
revealed that management actions might not meet the interests of all stakeholders (beneficiaries).
Incorporating metrics to represent relative value, as through content analysis, can help identify which
trade-offs are more or less important as identified by stakeholders. The open-ended nature of the analysis
174
-------
allowed us to capture the breadth of values including community connections (similar to social cohesion)
and sense of place (similar to identity). In addition, mapping the area provided by certain services can
help understand spatial trade-offs in ecosystem services, which may be important with respect to access to
and use of restored sites, as well as the socioeconomic status of adjacent neighborhoods. In the case of the
St. Louis River AOC, long segments of the shoreline lack opportunities to fish from shore, and the
riparian area providing public park and trail is virtually absent in the upriver portion of the AOC.
In our future research to be conducted in the St. Louis River AOC, we plan to merge these two
approaches; our goal is to better understand how spatial provisioning of ecosystem services relates to
stakeholder values and trade-offs, especially in the context of these large-scale clean-up projects.
Literature Cited
Angradi, T.R., D.W. Bolgrien, J.L. Launspach, B.J. Bellinger, M.A. Starry, J.C. Hoffman, M.E. Sierszen, A.S.
Trebitz, T.P. Hollenhorst. 2016. Mapping ecosystem services of a Great Lakes estuary can support local
decision-making. Journal of Great Lakes Research 42:717-727.
GLWQA (Great Lakes Water Quality Agreement). 1987. International Joint Commission, United States and Canada.
Sierszen, M.E., J.A. Morrice, A.S. Trebitz, J.C. Hoffman. 2012. A review of selected ecosystem services provided
by coastal wetlands of the Laurentian Great Lakes. Aquatic Ecosystem Health and Management 15:92-106
175
-------
Appendix G2
Mobile Bay, Alabama Case Study
Background
Mobile Bay is the drainage point for a 43,000 square mile watershed that covers portions of three states.
However, the quality and quantity of services provided by the Bay is greatly determined by urbanization
of land in smaller sub-watersheds (Fig. G2.1) along the edges of Mobile bay itself. Sub-watershed
restoration is a key objective in the management plan for the Mobile Bay National Estuary Program
including improvements in stream water quality and shoreline health. Yet, these efforts are not currently
evaluated with respect to provision of ecosystem goods and services or in the context of land-use change
in the surrounding landscape. Services provided to people is a key measure of success for restoration
projects. The goal of this case study research is to examine how planned and implemented restoration
activities have contributed to production of ecosystem goods and services, and how that contribution is
impacted by changes in land use.
M{ Vernon
Atmore (21
Perdido
Flomalon
Knetfe
ssPowtf
agoula
Figure G2.1. Map showing sub-watersheds adjacent to Mobile Bay, AL. Source: Mobile Bay National
Estuary Program.
Decision Context
Restoration activities are mandated by National Estuary Program goals and described in the
Comprehensive Conservation and Management Plan (CCMP), which is amended and updated every five
years. Overarching goals of restoration are to improve and maintain the quality of natural resources in the
Mobile Bay watershed with a focus on human benefit in six target categories: access, healthy beaches,
fish abundance, preservation of heritage and culture, promote ecosystem resilience, and maintain water
quality. Implementation of the CCMP requires that priorities be set and indicators of desired outcomes be
defined, so to allow for evaluation of resource investments. These indicators of success can be defined
based on ecosystem services production and so tie outcomes more directly to human benefit. In addition,
value of stream and shoreline restoration may be impacted by changes in the surrounding landscape that
are driven not by NEP priorities, but by municipal and county strategic planning. Impacts and outcomes
of NEP restoration activities should be evaluated in the context of landscape changes to allow for the
176
-------
most realistic measure of restoration impacts. What are the best metrics of restoration success and how
should landscape change be incorporated into project evaluation and planning?
Define Objectives and Measures
We are working with MBNEP staff and local municipalities to apply models and tools that relate
restoration changes to ecosystem service production and delivery at the subwatershed scale. These are
largely tools developed in other coastal watersheds (i.e., Tampa Bay, FL, Pensacola Bay, FL, Willamette
river, OR) and a secondary goal of the case study is to evaluate transferability of selected tools between
ecosystems and between target issues. The initial process is to work with stakeholders to define broad
ecosystem service-based objectives and measures of success and this process involved the formation of an
ecosystem services working group sponsored by the MBNEP. This group is working to match ecosystem
service priorities to the priorities laid out in the current CCMP. In congruence with this effort is the
parameterization of key model-based tools for a target subwatershed to be used as a testbed for model
application.
MnHiu c ymfMno Cuu l -twm.mmi I ami I num twom.' 38ฎ
Figure G2.2. Example screen shot of the H20 tool intended for mapping of select ecosystem goods
and services at the watershed scale (Russell et al. 2015). Example is from Tampa Bay, FL.
Initially, work is being conducted in the D'Olive watershed on the eastern shore of Mobile Bay near
Spanish Fort, AL. This initial work involves the parameterization of two models. An eco-hydrological
model (VELMA, Abdelnour et al. 2013) will be used to assess the impact of land cover and land use of
water quality and fish habitat. Second, an ecosystem services mapping tool (H20; Fig. G2.2) will be used
to directly measure ecosystem services production and delivery to beneficiaries in the subwatershed.
Together these two tools will be used to assess impacts of restoration activities, as well as the
interrelationship between in stream restoration and changes in land use/land cover.
177
-------
Identify Decision Alternatives
The target of this case study research is to develop new tools to evaluate existing restoration and
mitigation planning. Decision alternatives were defined by the MBNEP CCMP and to a lesser degree
from municipal and county strategic planning. These activities involve a large amount of stakeholder
engagement including planning meetings, phone and written surveys, and public comment periods.
Alternatives also included competitive proposal development for the specific implementation of
restoration activities. These were evaluated based on a comparison of projected benefits to the six goal
categories outlined by the MBNEP: access, healthy beaches, fish abundance, preservation of heritage and
culture, promote ecosystem resilience, and maintain water quality.
Estimate Consequences
Consequences of planned restoration or mitigation activities will be assessed based on changes in
production or delivery of ecosystem good and services to human beneficiaries. The Ecosystems Working
Group organized within the MBNEP Science Advisory Committee is tasked with identifying high priority
ecosystem services that are linked to the goal categories so that outcomes can be evaluated in the context
of stated goals. Models mentioned above will provide a process for measuring consequences for
ecosystem services, focused primarily on improvement of stakeholder access to resources, water quality,
and habitat.
Evaluate Trade-offs and Take Action
The evaluation of outcomes from planned restoration or mitigation activities with models and tools and
based on measures of success tied to stakeholder priorities will provide information for decision-making.
The objective is to consider trade-offs largely between different ecosystem services rather than between
different actions as the restoration planning occurs in five-year cycles. This is important as a planning
tool, but also provides an approach that can be applied in other sub-water sheds to evaluate projects from
an ecosystem services perspective.
Lessons Learned
Lessons learned from this exercise will directly inform future planning by the MBNEP. In addition, this
approach is being applied to federal restoration under the National Environmental Protection Act (NEPA)
Superfund program as a proof of concept for assessment of ecosystem services in NEPA projects. The
intended outcome is a broader suite of success indicators that can be linked to multiple stakeholder
objectives.
For More Information
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 catchments. Water Resources Research 49: 1292-1313.
Russell M, Harvey J, Ranade P, Murphy K. 2015. EPA H20 User Manual. US. EPA Office of Research and
Development, National Health and Environmental Effects Research Laboratory, Gulf Ecology Division. No.
EPA/600/R-15/090.
178
-------
Appendix G3
Pacific Northwest Case Study
Background
The Pacific Northwest (PNW) is a region of diverse and highly valued natural resources that provide a
variety of ecosystem services vital to human well-being. For example, the PNW's iconic conifer forests
are among the world's most productive forests, and are economically and ecologically important for
providing dependable supplies of clean drinking water, timber products, habitat for a diverse array of
wildlife species, and recreational opportunities for urban and rural populations. Rich alluvial soils in
broad valley bottoms support high value agricultural crops such as vegetables, fruits, grains, and wine
grapes. Abundant freshwater streams, estuaries and coastal habitats support salmon, shellfish and many
other economically important aquatic species.
However, these resources and services are being strained by population growth, land-use change, climate
change, and other stressors. Indeed, the historical pattern of resource use in the PNW has often been one
of boom and bust, with unsustainable management practices leading to severe downturns in major
industries, such as the once thriving salmon fishery and forest products industries. The resulting economic
and social impacts have been particularly damaging to rural communities.
Many PNW communities, tribes and state agencies are seeking assistance for mitigating and/or adapting
to projected changes in land use and climate, particularly with regard to longstanding and emerging water
resource issues. While there is considerable momentum and collaboration amongst stakeholders to design
and implement watershed-scale (crest-to-coast) management plans that address water resource and other
sustainability issues, there is uncertainty about how best to proceed (The Oregonian 2015).
The Pacific Northwest Ecosystem Services Case Study is working closely with community stakeholders
and state and federal agencies in Washington and Oregon to provide systems-based tools for identifying
ecosystem management solutions that consider the linkage of terrestrial and aquatic systems, especially
for coastal watersheds. Our primary goal is to assist local, state and federal jurisdictions seeking to
evaluate how alternative decision options affect ecosystem services and human well-being.
The PNW Case Study includes two major themes in three distinctly different watersheds, two in
Washington's Puget Sound basin and one on the Oregon coast (Figure G3.1). While each of these
involve unique sets of stakeholders and watershed impairment issues, we are finding that community-
based restoration planning goals can be addressed through a common decision support approach that uses
a transferable set of modeling tools.
The following summaries describe of our PNW Case Study watersheds and research themes.
Salmon Recovery Planning Support for Puget Sound Tribes & Communities
Degradation of riparian and stream habitats has been an important contributing factor to steep declines in
salmonid populations in the PNW. Forest and agricultural practices, road construction and other land-use
practices in or near riparian areas have increased stream temperatures and sedimentation, decreased large
woody debris, and accentuated stream peak and low flows. Collectively, these impacts have decreased the
ability of salmon to reproduce, grow and survive to complete their cycle of migration between terrestrial
and ocean habitats (Hawley 2012). For these and other reasons, 18 species of salmon in the state of
Washington are now listed under the Endangered Species Act
(http://www.rco.wa.eov/%5C/salmon recovery/listed species.shtml).
179
-------
Tillamook Bay Case Study Site )
Tillamook
Bay
Tillaml
Netarts
Bay
m
Puget Sound Case Study Sites
Seattle
~ ~
Tolt River
Mashel River
Figure G3.1. Pacific Northwest Case Study sites in the Mashel and Tolt River watersheds in
Washington's Puget Sound Basin, and in Oregon's Tillamook Bay estuary and contributing coastal
watersheds. [Note: Seattle watersheds are being studied under a separate but related research
project focused on urban water quality modeling].
We developed a set of transferable tools to provide decision support for community-based salmon
recovery planning in coastal zone watersheds. These tools have been integrated and are being applied in
collaboration with Puget Sound tribes and community stakeholders to address restoration of hydrological
and ecological processes critical to salmon recovery, and more broadly, to the functioning of entire
watersheds and the ecosystem services they provide.
For case studies in the Nisqually and Tolt River watersheds in central and southern Puget Sound (Figure
G3.1), the VELMA eco-hydrology model (Abdelnour et al. 2011, 2013; McKane et al. 2014a, 2014b) is
being used to quantify effects of various forest and floodplain management scenarios on key salmon
habitat variables, including peak and low flows, in-stream wood, fine sediment in spawning beds, and
riparian condition. Stream temperature is being simulated using Penumbra, a new stream shade and
temperature model integrated within VELMA (Halama 2017). VELMA/Penumbra stream habitat outputs
are being used to drive the Ecosystem Diagnosis and Treatment model, or EDT, fish habitat model to
simulate habitat potential and salmon population responses to the forest management and climate
scenarios (Blair et al. 2009). A 3-D visualization tool, VISTAS (Cushing et al. 2014), is being used to
summarize and communicate model outputs in an intuitive way.
Model results and/or training in the use of these tools is being provided to tribal and community members
to help them identify management practices for mitigating and adapting to climate variability. For
example, where and what kinds of in-stream, floodplain, riparian and upland restoration practices will be
most effective for improving cold water refuges, spawning and rearing habitat, and hydrologic flow
regimes (higher summer flows and lower peak flows)?
Stakeholders are currently using these model results to help address these and other concerns, such as the
establishment of a Nisqually Community Forest that sustainably supports local forest-sector jobs,
recreation and tourism.
180
-------
This work is being conducted in collaboration with the partners and stakeholders listed in Table G3.1.
Table G3.1. PNW Case Study stakeholders/partners and benefits addressed.
Case Study
Community, Tribal and
private stakeholders
State and Federal
partners
Benefits addressed
Puget Sound
Case Study
Nisqually & Snoqualmie
Tribes; Nisqually River
Foundation; Nisqually
Land Trust; Washington
Environmental Council;
Seattle City Light; King
County, WA; Towns of
Eatonville & Carnation,
WA; private industrial
forest owners
EPA Region 10;
Washington Dept of
Natural Resources;
Washington Dept of
Ecology; Puget Sound
National Estuary Program;
Oregon State University;
National Oceanic and
Atmospheric Association,
Oregon Dept of Fish and
Wildlife
Health: clean water for
drinking, swimming;
climate mitigation
Economic: forest &
fishery products; local
jobs; tourism; recreation
Social: Tribal cultural
benefits (salmon); local
self-sufficiency; human
well-being; sense of place
Tillamook Bay
Case Study
Town of Tillamook, OR;
shellfish producers; dairy
producers; private
industrial forest owners;
Tillamook Bay National
Estuary Program; EPA
Region 10; Tillamook
Estuaries Partnership,
Oregon Dept of
Environmental Quality,
Oregon Dept of
Agriculture, and Oregon
Dept of Fish and Wildlife
Health: less seafood
contamination & water-
borne GI disease; climate
mitigation
Economic: fisherv. aซ &
forest products; local
jobs; tourism; recreation
Social: human well-
being; sense of place
Tillamook Bay National Estuary Study, Oregon
The Tillamook Case Study on the Oregon coast (Figure G3.1) focuses on Water Quality, Ecosystem
Services and Environmental Change. Tillamook Estuary and watershed provides many water quality
dependent Final Ecosystem Goods and Services (EEGS), such as provisioning of shellfish and salmon,
clean drinking water from upland forest watersheds, and fertile floodplain soils that support the local
dairy industry. There is a history of interrelated impairments in the Tillamook system that have the
potential to impact FEGS. Recently, there have been improvements in water quality resulting from
management actions. This research is focused on what actions will sustain improvements in water quality
and provision of FEGS in the face of different causes of environmental change.
This research involves developing linked watershed and estuarine models (VELMA and the Coastal
Generalized Ecosystem Model [CGEM]) for assessing effects of changes in land use and climate on
watershed water quantity and quality (nutrients, temperature, pathogens); nutrient enhanced coastal
acidification and hypoxia; fecal bacteria source, transport and fate; and applying these results to FEGS
endpoints (shellfish, seagrass, clean water for drinking, swimming, fishing).
Tillamook Bay case study research is being conducted in collaboration with the partners and stakeholders
listed in Table G3.1.
181
-------
The following describes our structured decision-making approach for working with stakeholders to
identify environmental problems and solutions for these PNW watersheds.
Define Objectives and Measures
We are working with case study stakeholders to identify (1) impairments to intermediate and final
ecosystem good and services deemed essential to community well-being; and (2) methods and measures
for restoring those services at relevant spatial and temporal scales. Methods and measures range from
empirical field and laboratory studies (e.g., sampling and analysis of stream nutrients and pathogens) to
application of systems-based watershed, and estuarine/ocean models (VELMA, CGEM). These methods
and models are being integrated to quantify impacts of land use and climate on a comprehensive suite of
ecosystem services provided by terrestrial, stream and estuarine habitats. Our case study partners include
a wide range of stakeholders (Table G3.1).
Identify Decision Alternatives
Our stakeholder partners are directly involved in developing alternative model-based decision scenarios.
Scenarios are designed to identify best management practices for restoring ecosystem services important
to the economic, human health and cultural goals of communities and tribes. Scenarios typically focus on
specific watershed restoration questions. For Puget Sound, what watershed restoration practices will most
effectively restore populations of endangered salmonids, while also providing clean drinking water and
sustainable local forest sector jobs? For Tillamook Bay, which floodplain, urban and forest management
practices will most effectively reduce inputs of nutrients, sediments and fecal matter to the estuary? And
how can these practices be prioritized to best protect multiple objectives - human health, shellfish
production, sustainable local economies, and recreational pursuits?
Estimate Consequences
VELMA, CGEM and associated models will be used to estimate consequences of the alternative decision
scenarios on delivery of ecosystem goods and services for communities. These models are primarily
designed to simulate how specified changes in land-use practices impact intermediate ecosystem goods
and services (IEGS). IEGS examples include biophysical processes that regulate an ecosystem's capacity
to provide clean water and air, habitat for fish, shellfish and wildlife, and production of plant biomass for
food and fiber. Many of our PNW stakeholders are primarily interested in scientifically defensible
information they can use to formulate restoration decisions that target specific IEGS - for example, water
and air quality (as defined by TMDLs), fish and wildlife habitat (as mandated by the Endangered Species
Act, and tribal, federal and state salmon recovery goals), and many others. Such IEGS are things that
stakeholders engaged in natural resources management (Table G3.1) focus on and carefully monitor, as a
result of pressures from community constituents and state and federal regulators. Obviously, it is
necessary to also demonstrate how positive or negative changes in IEGS translate to Final Ecosystem
Goods and Services (FEGS) and, ultimately, to changes in human well-being (Appendix El). Therefore,
our PNW case study approach will include the end-to-end process of translating impacts of alternative
decision scenarios to consequences for IEGS, FEGS and benefits to human well-being (Appendix E4;
Fig. E4.2). This will require additional tools such as the FEGS Classification System (Landers and Nahlik
2013), Human Well-Being Index (HWBI; Smith et al. 2012) and, perhaps, the Envision decision support
platform (Bolte et al. 2011).
Evaluate Trade-offs and Take Action
Case study model outputs will be evaluated and communicated to stakeholders in ways that are intuitively
useful for their decision-making process. Our models incorporate various visualization tools - charts,
graphs and animations - designed to help communicate complex model outputs for decision-makers.
Figure G3.2 provides a hypothetical example of modeled trade-offs for one of our Puget Sound case
study locations (actual watershed-scale model simulations are in progress). This figure proved effective
182
-------
for engaging community and tribal stakeholders in the design of modeling scenarios now being used to
help inform salmon recovery planning and Nisqually Community Forest management decisions.
Tradeoffs for Alternative Watershed Management Scenarios
Hypothetical Example
Objectives
Scenario 1:
Current Industrial
Forest Practices
(maximize timber
yield & profit)
Scenario 2:
No Forest Harvest
(maximize ecological
benefits)
Scenario 3:
Multi-Stakeholder
Community Forest Plan
(optimize ecologicai
economic S cultural benefits
for tribes, community
stakeholders)
Forest Products
(1.0 = Most Board Feet)
Local Forest Sector Income
(1.0 = Most Local Income)
Salmon Habitat Quality
(1.0 = Most Salmon)
Ecosystem Carbon Stocks
(1.0 = Most Carbon Stored)
Figure G3.2. A hypothetical example of ecosystem service trade-offs associated with alternative
watershed management scenarios in Puget Sound.
Lessons Learned
A major early lesson from the PNW Case Study is that model-based outputs need to be communicated in
ways that connect immediately with the needs of stakeholders - '"How will this information answer our
community's questions, and how can we use it to implement solutions that recognize trade-offs among our
environmental, economic and social goals?" We are finding that simple informational summaries
(animations, graphs, bar charts) can be very effective when presented (1) within a restoration-relevant
narrative (usually Powerpoint-based), (2) in an interactive workshop setting at the stakeholder's site
location, and (3) with a follow-up field site tour that promotes discussion of restoration goals and
solutions. While some stakeholders are interested in learning how to apply our models (which we are
supporting), most prefer to focus on using model output summaries to identify best decision options. This
argues for development of simpler models accessible to most decision-makers. For example, EPA Region
10 has requested development of simplified models (e.g., "VELMA Lite") or tabular summaries (look-up
tables) of model outputs. Such an approach would make model-based assessments accessible to a much
wider range of stakeholders, and we are currently exploring ways to address this need.
References
Abdelnour, A., M. Stieglilz, P. Pan, R. McKane. 2011. Catchment hydrological responses to forest harvest amount
and spatial pattern. Water Resources Research 47:W09521.
Abdelnour, A., R, McKane, M. Stieglitz, F; Pan, Y. Cheng. 2013. Effects of harvest on carbon and nitrogen
dynamics in a Pacific Northwest forest catchment. Water Resources Research 49:1292-1313
Blair, G.R, L.C Lestelle, L,E. Mobrand. 2009. The ecosystem diagnosis and treatment model: a tool for assessing
salmonid performance potential based on habitat conditions. In K.E. Knudsen, & J.H. Michael Jr (Eds.),
American Fisheries Society Symposium (Vol. 71, p. 2008).
Bolte, J., R. McKane, D. Phillips, N. Schumaker, D. White, A. Brookes, D. Olszyk. 2011. In Oregon, the EPA
183
-------
calculates nature's worth now and in the future. Solutions 2(6): 35-41.
Cushing, J.B., N. Molnar, V. Ratanasangpunth, M. Bailey, J. Bolte, A. Brookes, D. Lach, J. Mangue, B. McKane, G.
Orr, Emily Piatt, K. Schmal, S. Stafford, C. Thomas, P. Wingo, K. Winters, D. Witherspoon. 2014. Visualizing
Terrestrial and Aquatic Systems in 3D. VisWeek, November 11-18, 2014, Paris, France.
Halama, J.J. 2017. Penumbra: A spatiotemporal shade-irradiance analysis tool with external model integration for
landscape assessment, habitat Enhancement, and water quality improvement. Ph.D. dissertation, Oregon State
University, Corvallis, OR.
Hawley, S. 2012. Recovering a Lost River: Removing Dams, Rewilding Salmon, Revitalizing Communities. Beacon
Press, Boston, MA.
Landers, D.H., and A.M. Nahlik. 2013. Final Ecosystem Goods and Services Classification System (FEGS-CS). US
Environmental Protection Agency, Western Ecology Division, Corvallis, OR. EPA/600/R-13/ORD-004914
McKane, R.B., A. Brookes, K. Djang, M. Papenfus, J. Ebersole, D. Phillips, J. Halama, P. Pettus, C. Burdick, and
M. Russell. 2014a. Sustainable and Healthy Communities Pacific Northwest Demonstration Study. US
Environmental Protection Agency, Report No. ORD-007386.
McKane, R.B., A. Brookes, K. Djang, M. Stieglitz, A. Abdelnour, F. Pan. 2014b. Enhanced version of VELMA
ecohydrological modeling and decision support framework to address engineered and natural applications of GI
for reducing nonpoint inputs of nutrients, contaminants, and sediments. US Environmental Protection Agency,
Report ORD-010080, Safe and Sustainable Waters Research Program.
McKane, R.B. et al. 2014c. Community-based decision support tool for flexible, interactive (real-time) assessments
that quantify trade-offs in ecosystem good and services for alternative decision scenarios in the Pacific
Northwest. US Environmental Protection Agency, Report ORD-010213, Sustainable and Healthy Communities
Research Program.
Smith, L.M., H.M. Smith, J.L. Case, and L. Harwell. 2012. Indicators and Methods for Constructing a U.S. Human
Well-being Index (HWBI) for Ecosystem Services Research. U.S. Environmental Protection Agency,
Washington, DC, EPA/600/R-12/023.
The Oregonian. 2015.
http://www.oregonlive.com/environment/index.ssf/2015/01/feds_reject_oregons_coastal_po.html
184
-------
Appendix G4
San Juan, Puerto Rico Case Study
Background
The San Juan National Estuary, Puerto Rico, is located in a predominately urban watershed integrated
with a number of freshwater and coastal ecosystems (Fig. G4.1). The estuary and its watershed are facing
a number of pressures, including urbanization, aquatic debris, habitat loss, stormwater runoff, sewage
discharges, and changing climate. Watershed management decisions have been proposed and are being
implemented to target these pressures, and improve the condition of the estuary, as well as associated
terrestrial and coastal ecosystems. Improvements in ecosystem goods and services may have benefits for
human well-being. The goal of this case study is to develop tools and approaches to investigate the
potential impacts of alternative watershed management decisions on social, economic, and ecological
benefits to the greater San Juan community living in the estuary watershed.
Figure G4.1. San Jose Lagoon, San Juan, PR.
Clarify Decision Context
The decision context is largely focused on implementation of the San Juan National Estuary Program
(NEP) management plan, as well as a nested effort focused on the revitalization of an impoverished urban
neighborhood that borders the estuary. The San Juan NEP has extensive stakeholder outreach in
development and implementation of management plans. Consequently, this case study is largely
leveraging reviews of existing documents to characterize the decision context.
The PEGS Classification System (Landers and Nahlik 2013) is being applied to review NEP management
documents to identify Final Ecosystem Goods and Services most relevant to the San Juan case study,
including ecosystems providing the services and key beneficiary groups. The Drivers-Pressures-State-
Impact-Response (DPSIR) conceptual model (Bradley and Yee 2015) is being merged with the benefits
assessment model (Wainger and Mazzotta 2011) to understand interconnections among components of
the system, identify areas where information is needed, communicate with collaborators, and provide the
conceptual foundation for future predictive models (Fig. G4.2).
185
-------
Drivers
Pressures
Responses
State
Intermediate ES
average summer temp as
f(Kveg In upwind fetch)
from Murphy et al. 2011)
Air
temperature
per delta
deg.C
| Ozone
Delta hospitalization
asthma per delta 03
| concentration
Public
perception of
natural spaces
Delta 03 per
delta dee. C
Response:
El Fideicomiso
de la Tierra
Response:
Zika/Dengue
Prevention
Response:
Amt / location
of public trees
Atmospheric
pollution
removal
Population
in CMP
[Tree model algorithms
Tree
Maintenance
C02 seq. per unit area of
Carbon
sequestration
Need for
shelter
Aedes habitat
Walking distance to open / green spaces & Viewable street
Need for
food
Response:
Inaction
Denitrlfication
[gN/m2/yr by land cover]
Russell etal. (2013)
Response:
Width/location
of buffer
Denltrification rate
[gN/m2/yr]
Mangrove
Forest
Erosion control
4, Wave height
<4, Storm surge
Saltmarsh
/ seagrass
Waste
generation
Habitat and
nursery for
fisheries
Response:
Wider Street /
Trash removal
Fish recruitment
per unit
Contaminant
processing
(EC)
Response:
Access to
Channel
Figure G4.2. Conceptual model merging DPSIR and benefits assessment to diagram cause-effect
relationships for San Juan case study (courtesy S. Balogh).
186
-------
Define Objectives and Measures
Objectives relevant to estuary management were characterized by reviewing existing planning documents.
San Juan NEP objectives, comparable to most other NEPs, can be binned into 4 broad categories:
maximizing ecological integrity, maximizing social benefits, maximizing economic opportunities, and
maximizing stewardship (Fig. G4.3). Inferred objectives showed similarity to concepts in the Human
Well-Being Index, including culture, social cohesion, connection to nature, safety, living standards,
health, and education. The HWBI is being explored as a way to measure and monitor successful decision-
making (Orlando et al. 2017). Metrics in the EnviroAtlas, Rapids Benefits Indicators (Appendix C4), or
derived from the FEGS-CS are also being explored as potential performance measures to evaluate
decision alternatives.
Ecosystem Services
Contaminant processing
Mitrogen processing
Carbon sequestration
Habitat & biodiversity
Water quality regulation
Harvestablefish
Flood regulation
Disease regulation
Objectives
Maximize Ecological Integrity
Maximize natural hydrology and flow
ฆ Maximize natural habitat quality and quantity
ฆ Maximize biodiversity (desirable & endangered/threatened)
Maximize Social Benefits
ฆ Maximize beneficial usesforthe public (recreation, culture, open access)
Minimize public health threats (drinking water, natural hazards, seafood
contamination, pathogen contamination, vector-borne illness)
Maximize access to safe housing
Maximize Economic Opportunities
Maximize tourism opportunities
ฆ Maximize industry/shipping/port opportunities
Minimize financial burdens and threats to livelihood
Maximize Stewardship
ฆ Maximize stakeholder engagement (education, volunteering, partnerships)
ฆ Maximize community connectedness to the estuary
ฆ Foster good governance
Figure G4.3. Ecosystem services related to San Juan NEP objectives, derived from existing
documents.
Estimate Consequences
A number of watershed management plans have been proposed through NEP planning efforts and are
being implemented, including canal dredging, restoration of mangroves, sewage discharge interventions,
and climate adaptive strategies. Case study research efforts are largely focused on developing information
to reduce uncertainties when estimating the potential consequences of these decisions on community-
relevant endpoints, including:
Reviewing and applying ecological production functions (EPFs) to quantify ecosystem services
production of stakeholder relevant endpoints
Conducting field work to characterize carbon storage and anthropogenic nitrogen flow through
the estuarine system
Developing ecological benefits functions (EBFs) to link ecosystem services to human health
benefits (Fig. G4.3)
Conducting field work to link flooding and water quality to impacts on asthma and vector-borne
illnesses
Calculating historic and baseline estimates of HWBI in the San Juan watershed, and developing
quantitative relationships between ecosystem services and well-being (Smith et al. 2014)
Integrating information, data, and models into modeling systems (e.g. energy and materials flow,
urban metabolism, Envision) to investigate the impacts of alternative decision scenarios on
priority ecosystem services and associated benefits to human well-being
187
-------
Impact
The San Juan case study will explore impacts of watershed management decisions on the condition of
estuarine ecosystems and associated ecosystem goods and services, with a particular focus on quantifying
potential benefits to human well-being. This case study will improve understanding of the relationships
and trade-offs involved in watershed management of estuaries; identify decisions and management
options that support economically, socially, and environmentally sustainable communities; and explore
approaches for integrating ecosystem services into community decision-making that are scalable and
transferable.
For More Information
Bradley, P., and S. Yee. 2015. Using the DPSIR Framework to Develop a Conceptual Model:Technical Support
Document. EPA/600/R-15/154, 69
Landers D. and A. Nahlik. 2013. Final Ecosystem Goods and Services Classification System (FEGS-CS).
Washington, DC: U.S. Environmental Protection Agency. EPA/600/R-13/ORD-004914
Orlando, J.L., S.H. Yee, L.C. Harwell, and L.M. Smith. 2017. Technical Guidance for Constructing a Human Well-
being Index (HWBI): A Puerto Rico Example. U.S. Environmental Protection Agency, Gulf Breeze, FL,
EPA/600/R-16/363.
Smith, L.M., L. Harwell, J.K. Summers, H.M. Smith, C.M. Wade, K.R. Straub, J.L. Case. 2014. A U.S. Human
Well-being Index (HWBI) for multiple scales: linking service provisioning to human well-being endpoints
(2000-2010). EPA/600/R-14/223.
Wainger, L. and Mazzotta, M. 2011. Realizing the potential of ecosystem services: a framework for relating
ecological changes to economic benefits. Environmental management 48(4), 710-733.
188
-------
Appendix G5
Southern Plains, Oklahoma Case Study
Background
In the Southern Plains area many rural communities have incorporated impoundments of varying sizes to
improve numerous ecosystem services such as flood control, water supply, recreational activities,
irrigation, and wildlife habitat. These small communities have come to rely on these impoundment
systems to provide everyday benefits that help sustain their communities. There are many challenges that
potentially impact the ability of these systems to provide the services that these communities have come
to not only rely on, but in many cases take for granted. Many of these communities have purposely
developed community growth plans to increase the development potential and ultimately the economic
viability of these communities in an effort to attract businesses and people to the community. This
growth, while good for the community, has impacts on the surrounding area and the potential
provisioning of the ecosystem services the community relies upon. Couple this with the frequent
conditions of protracted drought and interspersing flooding events, these communities have a strong need
to develop and implement community sustainability and resilience approaches that look to understand
how economic growth and climatic conditions impact the provisioning of the ecosystem services that
support their community viability.
As communities continue to grow, strains on resources increase. Often times in the southern plains, these
resources are shared by multiple communities that have their own sets of needs and priorities that rarely
take into account the needs and priorities of the other communities. This may not be a problem when
resource use is far below resource provisioning, but as demands on these resources has greatly increased,
tension and disagreement about the use of these resources has created concerns between communities that
share resources. These are extremely difficult problems to address and require approaches that are
designed to deal with the multifaceted, multi-stakeholder planning processes to develop community
sustainability and resiliency plans focused on the needs and values of all the various stakeholder groups.
Decision Context and Approach
The Southern Plains case study will be focused on working with two communities, Perry Oklahoma and
Stillwater Oklahoma, which are located 25 miles apart and share a common water resource. Over time
these communities have taken different development paths where the populations of Perry (approx. 5000)
and Stillwater (approx. 39,000) are markedly different. Much of the growth difference for Stillwater has
been driven by the presence of Oklahoma State University in Stillwater. The focal point in common for
both communities is Lake McMurtry, a reservoir impoundment located in the Stillwater Creek Watershed
(SCW). Stillwater Creek Watershed is located in North Central Oklahoma in the counties of Logan,
Noble and Payne. The SCW has a drainage area of 276 square miles. Lake McMurtry, with a drainage
area of 25.04 square miles, is located in the North Western portion of the Stillwater Creek Watershed and
drains to Stillwater Creek via the North Stillwater Creek tributary (Figure G5.1). The watershed had a
mixed land use - land cover comprised of both urban and rural landscapes (Figure G5.2).
Communities such as Perry, Oklahoma and Stillwater, Oklahoma have needs for assistance in resiliency
planning that provides for shared cooperation for a variety of reasons including changing climatic
conditions (drought, flood), water resource planning and management, contaminated runoff, sediment
impact to water source issues, higher frequency extreme weather events, and expected population
increases with community development expansion to name a few. The problems and challenges are
known, but the approach to address them is complicated. The Southern Plains Case Study is being
developed to incorporate broad stakeholder involvement from the very start of the project effort using the
DASEES (Decision Analysis for a Sustainable Environment, Economy, and Society; Appendix Al)
structured decision making approach to help these two communities plan for future development and
189
-------
address the challenges of shared resources (e.g., water, land use) impacted by drought and floods.
SiIIMwbw C-ae-k Watershed
Figure G5.1. Map showing the Stillwater Creek Watershed, Lakes McMurtry and Carl Blackwell,
and the Cities of Perry and Stillwater.
Figure G5.2. Map showing the Stillwater Creek Land Use - Land Cover 2015
The case study will be working with the Oklahoma Conservation Commission, Oklahoma State
University, the Cities of Perry and Stillwater and other identified stakeholders with interest in developing
these sustainability and resiliency plans. We will be applying the DASEES approach and implementing
the five step iterative decision process designed to:
190
-------
1. Understand the Decision Context (see Appendix Bl);
2. Define Objectives (see Appendix CI);
3. Develop Options (see Appendix Dl);
4. Evaluate Options (see Appendix E10, Fl);
5. Take Action (see Appendix F4).
Define Objectives and Measures
The DASEES process will capture the values of the stakeholder and decision-makers as the foundation
for developing the final management plans for achieving sustainability and resiliency of both
communities. It is through this values elicitation process that the fundamental objectives and measures of
success will be developed for this effort. The ultimate goal of this process will be to have a management
plan that meets the needs of both communities and helps to ensure and facilitate the incorporation of
economic, societal and ecosystem services in the decision-making process for both communities.
Identify Objective Preference Weighting and Decision Alternatives
Through the use of the DASEES process a number of objectives will be developed based on the values of
the stakeholders and experts. These objectives will be ranked from most important to least important and
each will be given a relative preference weighting based on how the stakeholders value the objectives.
Once these weightings are assigned, scenarios will be run based on these weights derived from the
stakeholders and the input from the experts. These scenarios will be evaluated against each other to
determine how they perform.
Estimate Consequences
Consequences of each management scenario will be evaluated and compared by constructing
Consequence tables that clearly show how each scenario performed. In this way the stakeholders and
experts can see the different results and chose an approach that achieves the desired end objectives for
each community. These consequence tables will help the communities to compare and contrast each
approach for the positives and negatives that each presents.
Evaluate Tradeoffs and Take Action
Once the stakeholders and experts have compared and contrasted the pros and cons of each approach,
they will discuss the trade-offs that will have to be made to achieve a mutually beneficial outcome for
each community. Once this is accomplished they can move forward with implementing the management
plans. But if they cannot come to an agreement on an approach to implement, they risk the potential for
litigation that would ultimately be in neither community's best interest.
Impact
Working with through the DASEES structured decision making approach will be used to apply a
management plan to ensure the long term sustainability and resiliency of both communities. Lessons
learned may be relevant to other communities with shared resources that may be facing similar
sustainability and resiliency issues as well.
For More Information
EPA (Environmental Protection Agency). 2012. Decision Support Framework Implementation of the Web-based
Environmental Decision Analysis Application DASEES: Decision Analysis for a Sustainable Environment,
Economy, and Society. EPA/600/R-12/008.
191
-------
&EPA
United States
Environmental Protection Agency
------- |