Citation
Using Participatory Decision Support to Improve Coral Reef Management

Material Information

Title:
Using Participatory Decision Support to Improve Coral Reef Management
Creator:
Fletcher, Pamela J
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (160 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Soil and Water Science
Committee Chair:
LI,YUNCONG
Committee Co-Chair:
KIKER,GREGORY A
Committee Members:
CLARK,MARK W
SPRANGER,MICHAEL S
HENDEE,JAMES C
Graduation Date:
8/9/2014

Subjects

Subjects / Keywords:
Climate change ( jstor )
Climate models ( jstor )
Coral reefs ( jstor )
Corals ( jstor )
Ecology ( jstor )
Ecosystem models ( jstor )
Ecosystems ( jstor )
Marine resources ( jstor )
Needs assessment ( jstor )
Oceanic climates ( jstor )
Soil and Water Science -- Dissertations, Academic -- UF
decision -- participatory -- socio-ecological
Miami metropolitan area ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Soil and Water Science thesis, Ph.D.

Notes

Abstract:
Participatory decision support systems are a promising approach for planning for and responding to climate change due to the ability to integrate interdisciplinary datasets transparently to help bridge the gap between science and management. These systems identify risk, uncertainty, and stakeholder preferences to improve the understanding of dynamic, socio-ecological systems for informed decision-making. Coral reef ecosystems, especially when located adjacent to an urbanized landscape, present an opportunity to explore the value and utility of participatory decision support for improving the understanding of the coupled system for informed decision making. Three research phases were used to explore participatory decision support in southeast Florida, USA: needs assessments, integrated ecosystem assessments, and Bayesian belief networks. Each step relies upon the previous to build, construct, and evaluate participatory decision support with emphasis on the importance of stakeholder collaboration for improved management strategy alternatives for coral reef ecosystems in a changing climate. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2014.
Local:
Adviser: LI,YUNCONG.
Local:
Co-adviser: KIKER,GREGORY A.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-02-28
Statement of Responsibility:
by Pamela J Fletcher.

Record Information

Source Institution:
UFRGP
Rights Management:
Copyright Fletcher, Pamela J. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
2/28/2015
Resource Identifier:
968131431 ( OCLC )
Classification:
LD1780 2014 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

USING PARTICIPATORY DECISION SUPPORT TO IMPROVE CORAL REEF MANAGEMENT By PAMELA J. FLETCHER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2014

PAGE 2

2014 Pamela J. Fletcher

PAGE 3

To my family

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my committee members for guiding me in the pursuit of this degree for several years and seeing it through completion. Attainment of this degree would not have been possible without their encouragement and patience. Dr. Yuncong Li, Dr. Greg ory Kiker, Dr. Michael Spranger, Dr. Mark Clark , and Dr. James Hendee provided ex pertise in their field of science, allowing me an opportunity to gain insight from a variety of scientific perspectives and real world experiences they chose to share with me during this academic journey. I thank my colleagues at NOAA’s Atlantic Oceanogra phic and Meteorological Laboratory and the University of Miami’s Rosenstiel School for Marine and Atmospheric Science. Their knowledge, data, and willingness to experiment with new concepts w ere vital to the success of my research. I thank S acheen Tavares Lei ghton and Chris Ellis from NOAA’s Coastal Services Center for their help in developing the needs assessment and pre/post survey instruments. I extend my appreciation to those who participated in this research, providing input throughout the project peri od. I recognize the University of Florida’ s Institutional Review Board for approving permit #2013U 913 allowing me to conduct interviews with resource managers . In addition, I acknowledge funding support from NOAA’s Center for Sponsored Coastal Ocean Research under award NA08OAR4320889, NOAA’s Coral Reef Conservation Program under award C085ID647, NOAA’s National Sea Grant College Program, and UF/IFAS Florida Sea Grant College Program. Finally, I thank my family for their support and encouragement over the past four years, and especially my children for their endless curiosity and fascination for South Florida’s marine and coastal environment . They were an ins piration for me to complete this degree so we can spend time outside exploring this amazing ecosystem together.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 8 LIST OF FIGURES ........................................................................................................ 10 LIST OF ABBREVIATIONS ........................................................................................... 11 ABSTRACT ................................................................................................................... 12 CHAPTER 1 CORAL REEF MANAGEMENT .............................................................................. 13 Background and Problem Statement ...................................................................... 13 Why Corals, Climate, and Water Quality? ........................................................ 14 What Is Participatory Decision Support? .......................................................... 16 Why Is Participatory Decision Support Important to Coral Reef Ecosystems in Florida? ..................................................................................................... 16 How Was Participatory Decision Support Used in Southeast Florida? ............. 18 Phase I: Needs assessment of climate information for use by marine resource and coral reef managers in southeast Florida .......................... 19 Phase II: Using the Integrated Ecosystem Assessment (IEA) process to transfer information to managers ............................................................ 19 Phase III: Using participatory decision support to develop tools to improve the understanding of coral reef ecosystems for informed decision making ...................................................................................... 20 Research Objectives ............................................................................................... 20 Organization of the Dissertation .............................................................................. 21 2 DECISION TOOLS FOR CORAL REEF MANAGERS: USING PARTICIPATORY DECISION SUPPORT TO INTEGRATE POTENTIAL CLIMATE IMPACTS AND INFORMED DECISION MAKING ................................. 23 Introduction ............................................................................................................. 23 Methods .................................................................................................................. 26 People .............................................................................................................. 26 End users ................................................................................................... 27 Researchers ............................................................................................... 28 Coordinator ................................................................................................ 29 Process ............................................................................................................ 29 Stating the issue and intended audience ................................................... 30 Establishing a planning team ..................................................................... 30 Information and literatur e search ............................................................... 31 Characterizing the audience ...................................................................... 31

PAGE 6

6 Establishing goals and objectives of the needs assessment ..................... 32 Selecting data collection methods .............................................................. 33 Tools ................................................................................................................ 34 Results .................................................................................................................... 37 People .............................................................................................................. 37 Process ............................................................................................................ 38 Tools ................................................................................................................ 39 Implications ............................................................................................................. 42 People .............................................................................................................. 43 Process ............................................................................................................ 44 Tools ................................................................................................................ 45 3 USING THE INTEGRATED ECOSYSTEM ASSESSMENT FRAMEWORK TO BUILD CONSENSUS AND TRANSFER INFORMATION TO MANAGERS ........... 59 Introduction ............................................................................................................. 59 Materials and Methods ............................................................................................ 61 Scoping ............................................................................................................ 62 Indicat ors .......................................................................................................... 66 Risk Analysis .................................................................................................... 67 Evaluating Management Scenarios .................................................................. 67 Communications ............................................................................................... 7 3 Printed materials ........................................................................................ 73 Internet based products ............................................................................. 74 Interpersonal communications ................................................................... 74 Results .................................................................................................................... 75 Scoping ............................................................................................................ 75 Indicators .......................................................................................................... 76 Risk Analysis .................................................................................................... 77 Bayesian Belief Network ................................................................................... 77 Communications ............................................................................................... 78 Discussion .............................................................................................................. 79 Implications ............................................................................................................. 84 4 CONSTRUCTING DECISION SUPPORT TOOLS TO IMPROVE CORAL REEF ECOSYSTEM MANAGEMENT .................................................................. 104 Introduction ........................................................................................................... 104 Materials and Methods .......................................................................................... 107 Defining the Problem Statement ..................................................................... 107 Constructing Preliminary Networks ................................................................ 109 Ecosystem scale BBN ............................................................................. 110 Habitat scale BBN .................................................................................... 114 Consulting Stakeholders to Assess the Utility of the Networks ....................... 114 Results .................................................................................................................. 117 Defining the Problem Statement ..................................................................... 117 Constructing Preliminary Networks ................................................................ 117

PAGE 7

7 Consulting Stakeholders to Assess the Utility of the Networks ....................... 121 Implications ........................................................................................................... 122 Defining the Problem Statement ..................................................................... 122 Constructing Preliminary Network .................................................................. 123 Consulting stakeholders to assess the utility of the networks ......................... 124 5 CONCLUSIONS ................................................................................................... 142 LIST OF REFERENCES ............................................................................................. 145 BIOGRAPHICAL SKETCH .......................................................................................... 160

PAGE 8

8 LIST OF TABLES Table page 2 1 Managers who participated in the CHAMP pre/post survey. .............................. 51 2 2 Marine resource data currently used by managers ............................................. 52 2 3 Manager c limate information needs. ................................................................... 53 2 4 Climate information that is difficult to locate or interpret ..................................... 54 2 5 Recommendations for the design of a climate information decision tool. ........... 55 2 6 Recommendations for the delivery of a climate information decision tool. ......... 55 2 7 List of recommendations to improve the climate data product prototypes .......... 56 2 8 Pre test survey to identify manager responsibilities for each respondent . .......... 57 2 9 Survey responses related to the mapbased query tool ..................................... 58 3 1 Indicator development criteria. ............................................................................ 93 3 2 Prototype Bayesian Belief Network indicators, units, and states for the southeast Florida coral reef and hardbottom conceptual ecological model . ....... 94 3 3 Conditional probability table for “Aesthetics” ...................................................... 98 3 4 Science communications developed to for the Integrated Ecosystem Assessment ....................................................................................................... 99 3 5 Sensitivity analysis for the “Aesthetics” node in the proof of concept Bayesian Belief Network .................................................................................................. 102 3 6 Sensitivity analysis for the “GrowthSurvivalReproRecr uitment” node in the proof of concept Bayesian Belief Network ........................................................ 103 4 1 Agency perspectives and goals. ....................................................................... 127 4 2 Problem statement identification ....................................................................... 128 4 3 Integrated conceptual ecosystem model nodes, descriptors, and values used to develop the Bayesian network for the southeast Florida coast. .................... 131 4 4 Conditional probabilities assigned for the “State_Index” node in the EBM DPSER integrated conceptual ecosystem model for the southeast Florida coastal ecosystem. ........................................................................................... 134

PAGE 9

9 4 5 Managers who participated in the BBN pre/post survey . .................................. 136 4 6 Sensitivity analysis for the “Aesthetics Environments” node in the proof of concept Bayesian Belief Network . .................................................................... 139 4 7 Conditional Probability Table used to run sensitivity analysis of Drivers based on Management Response Situation A and Situation B. .................................. 139 4 8 Scenario analysis for Situation A and Situation B for the ecosystem scale proof of concept Bayesian Belief Network ........................................................ 139 4 9 Sensitivity to findings for the “ Aesthetic Environments ” node when Management Response is set to Yes for Situation A and Situation B. ............. 140 4 10 Sensitivity to findings for Aesthetic Environments node when Management Response is set to No for S ituation A and Situation B. ..................................... 140 4 1 1 Survey responses ............................................................................................. 141

PAGE 10

10 LIST OF FIGURES Figure page 1 1 South Florida’s coastal marine ecosystem ......................................................... 22 2 1 Map of the Florida Reef Tract. ............................................................................ 47 2 2 Timeline of needs assessment proces s components and activities used to complete climate information tool development for the Florida Reef Tract. ........ 48 2 3 Sample of the Coral Health and Monitoring Program mapbased query tool tuto rial used during the pre/post test. ................................................................. 49 2 4 Cognitive map of the decision making process used in one management agency in South Florida. ..................................................................................... 50 3 1 Map of the South Florida I ntegrated Ecosystem Assessment project . ................ 86 3 2 South Florida Integrated Ecosystem Assessment process and products.. ......... 87 3 3 Conceptual diagrms of the South Florida ecosystem . ........................................ 88 3 4 I ntegrated Conceptual Ecosystem Models and the DPSER framework. ............ 89 3 5 Water Quality Integrated Conceptual Ecological Submodel. .............................. 90 3 6 Climate change impacts of the coral reef and hardbottom habitats .................... 91 3 7 Bayesian Belief Networks and southeast Florida reefs . ...................................... 92 4 1 EBM DPSER Integrated Conceptual E cosystem M odel for the southeast Florida coastal ecosystem. ............................................................................... 126 4 2 Bayesian belief network for the Integrated Conceptual Ecosystem Model for the southeast Florida coast. ............................................................................. 130 4 3 B ayesian belief network f or the coral reef and hardbottom C onceptual Ecological Model for the southeast Florida coast. ............................................ 135 4 4 Tutorial for Bayesian Belief Network decision support tool ............................... 137

PAGE 11

11 LIST OF ABBREVIATIONS AOML Atlantic Oceanographic and Meteorological Laboratory BBN Bayesian Belief Network CEM Conceptual Ecological Model CERP CHAMP Comprehensive Everglades Restoration Plan Coral Health and Monitoring Program CRCP Coral Reef Conservation Program CSC E mail EBM Coastal Services Center Electronic mail Ecosystem based management EBM DPSER Ecosystem based management Drivers, Pressures, State, Ecosystem Services, Response FKNMS Florida Keys National Mari ne Sanctuary ICEM Integrated Conceptual Ecosystem Model IEA Integrated Ecosystem Assessment MARES Marine and Estuarine Goal Setting for South Florida NCCOS National Centers for Coastal Ocean Science NOAA National Oceanic and Atmospheric Administration NSG National Sea Grant

PAGE 12

12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy USING PARTICIPATORY DECISION SUPPORT TO IM PROVE CORAL REEF MANAGEMENT By PAMELA J. FLETCHER August 2014 Chair: Yuncong Li Cochair: Gregory Kiker Major: S oil and Water Science Participatory decision support is a promising approach for planning for and responding to risk and uncertainty when managing dynamic socioecological systems. The process integrates interdisciplinary knowledge transparently through stakeholder consultation to help bridge the gap between science and management . Coral reef ecosystems, especially when located adjacent to an ur ban landscape, present an opportunity to determine the value and utility of participatory decision support for building an understanding of the coupled socio ecological system for informed decision making. Three phases of research were used to explore part icipatory decision support for coral reef management in southeast Florida, USA: needs assessment of stakeholder preferences , the integrated ecosystem assessment process framework to transfer information to managers , and decision support tool development an d evaluation using Bayesian belief networks. Each phase built upon the previous, emphasizing stakeholder collaboration to improve the understanding the South Florida coral reef ecosystem for informed decision making.

PAGE 13

13 CHAPTER 1 CORAL REEF MANAGEMENT Background and Problem Statement Managing natural resources is a complex socioecological problem, sometimes called a “wicked ” problem ( Batie , 2008; Brown et al., 2010; Churchman, 1967; Conklin and Weil,1997; R itt el and Webber, 1973). Participatory decision support is one approach to begin unravelling the complexities of these problems. This method emphasizes the importance of stakeholder collaboration through effective science communications and the transformation of data into information and knowledge. In many cases, a combination of people, processes, and tools can be used t o improve the understanding of a complex situation to move towards informed decision making (Kiker et al., 2005) . This dissertation is an exploration of participatory decision support inte grating people, processes, and tools to improve the understanding of impacts to coral reef ecosystems for informed decision making. It is helpful to understand that the research project was closely tied to the University of Florida/Florida Sea Grant Colle ge Program extension efforts to link the National Oceanic and Atmospheric Administration ( NOAA) research to the broader community of stakeholders in South Florida. The role of the Sea Grant Regional Extension Coordinator located at NOAA’s Atlantic Oceanogr aphic and Meteorological Laboratory (AOML) in Miami, Florida is to liaise am ong resource managers, decision makers, researchers, educators, and stakeholders to improve the exchange of information for informed decision making. Interactions with these user groups formed the motivation for much of this research and the coordination of and participation in 60 meetings, workshops, focus group sessions, and interpersonal communications from 20092014. Participation as the “ honest broker" in these interactions for med the

PAGE 14

14 foundation of the participatory decisi on support research with over 213 individuals contributing to the output of this dissertation. It is also important to mention that funding for the trans disciplinary research carried out by the Regional Sea Gr ant Extension C oordinator was sought and secured from several NOAA offices including N ational S ea G rant College Program (NSG), Atlantic Oceanographic and Meteorological Laboratory (A OML ) , C oral Reef Conservation Program (C RCP) , and N ational Centers for Coastal Ocean Science (N CCOS) . The funding opportunities were pursued with the explicit intention of addressing components of the National Ocean Policy through the R egional Sea Grant Extension Coordinator . All activities sought to improve the understanding of the marine ecosystem for informed decision making within the objectives of the University of Florida/Florida Sea Grant College Program liaison position . Why C orals, C limate , and W ater Q uality ? Coral reef ecosystems worldwide are in decline, largely due to changes in water quality (e.g., temperature, nutrients, sea water chemistry) ( Bellwood et al. , 2004; Carpenter et al. , 2008; Donahue et al. , 2008; Gardner et al. , 2005 ; Keller et al., 2009; Lirman et al., 2014; Nuttle and Fletcher , 2013a,c; Wagner et al., 2010; Wilkinson, 2000) . This is significant because of the importance of coral reefs in supporting ecosystem services upon which humans depend. These services include but are not limited to habitat for reef dependent fish that often end up on dinner tables, tourism opportunities that are a major economic driver in many coastal communities, and storm protection for coastal areas. The second longest bank barrier reef system in the world is located off of the southeast coast of the Florida peninsula a nd is ca lled the Florida Reef Tract (Figure 1 1) (Walker et al., 2008) . In an effort to protect and preserve the southernmost portion of the reef ecosystem, NOAA in partnership with local, state, and federal offices

PAGE 15

15 established the Florida Keys National Marine Sanctuary (FKNMS) in 1990. In 1994 , the Water Quality Protection Program was created, and two years later an action plan for the sanctuary was developed by the Environmental Protection Agency and the State of Florida. In 2002, a science plan and conceptual ecosystem model outlining pressures impacting the reef ecosystem, data gaps, and management responses to address these impacts were produced (FKNMS, 2002). Two pressures identified in the plan were 1) upland water management within the Greater Everglades Ec osystem, and 2) climate variability. From 20092013 , another model was developed to identify impacts, data gaps, status and trends, and management responses for the Florida Reef Tract . R esults illustrate upland influences and climate as the major drivers impacting the marine and coastal ecosystem in southeast Florida (Nuttle and Fletcher, 2013a, c ). These efforts, along with products developed for the South Florida Ecosystem Restoration Task Force and the Comprehensive Everglades Restoration Plan (CERP) pr ovide an overview of the ecosystem and models and monitoring data reflecting an overall decline in its condition (Nuttle and Fletcher, 2013a, b, c; RECOVER, 2012). it is recognized that management challenges persist even with the plethora of robust monitor ing data and complex models available to managers (Walters, 2007). Challenges to the decisionmaking process are further exacerbated by the need to include human dimensions science (e.g., social well being, socioeconomic data, etc.) and climate impacts. D ecision support systems designed to achieve the integration of biophysical, institutional, and human dimensions sciences at varying time and spatial scales are critical to guide managers and stakeholders in deciphering appropriate responses to ecosystem ch anges and ecological events.

PAGE 16

16 What I s Participatory Decision S upport ? Participatory decision support refers to a process of bringing people, sometimes called stakeholders, together to develop a solution, idea, or concept in which there will be some action taken in the future ( Cain et al., 2003; deGroot et al., 2010; Raymond and Cleary, 2013; Senge, 1990; Sheppard, 2005; Slo combe, 1993a ; Slocum et al., 1995; van de Kerkhof 2001; Walker et al., 2002). Pa rticipatory decision support is a valuable tool for applied, ecosystem based planning due to the robustness of the data and transparency of the approach that reaches across disciplines (Martin et al., 2011; Pimbert and Pretty, 1997; Pomeroy and Douvere, 2008; Ramirez, 1999; Raymond and Cleary, 2013; Szaro et al, 1998). This method has the potential to bridge the highly technical knowledge of scientists and the varied knowledge and values of nonscientists to make decisions, provide reasonable assurance to justify those decisions, or create policies that respect livelihoods and maintain ecological integrity (Kiker et al., 2005; Lynam et al., 2007). Participatory decision support is useful to bring diverse stakeholders together to synthesize the state of the ecosystem and apply this knowledge to develop and implem ent management priorities. Why Is P articipatory Decision S upport I mportant to C oral R eef E cosystems in Florida? Participatory decision support helps identify and prioritize the competing tradeoffs among the varied goals and values that stakeholders have f or the coastal environment using an open discussion format that informs decision making. The synthesis of knowledge is important for sustaining the benefits provided by marine ecosystems and their value to the regional economy for both ecosystem and human health. The resulting information can then be used to construct management alternatives .

PAGE 17

17 Participatory decision support is one approach to address the state and federal mandates to achieve ecosystem based management . R ecent regulatory advances in marine r esource management at the state and federal level include an ecosystem approach and participatory decision support. On July 19, 2010, the U.S. National Ocean Policy was established by Executive Order 13547, and an implementation plan was released on April 16, 2013 (Executive Order 1357, 2010) . These actions resulted in guidance for the creation of a National Ocean Council comprised of 27 Federal agencies and an outline for pursuing policy initiatives and establishing conditions for the development and implementation of a National Ocean Policy. The following three components of the Implementation Plan directly relate to the research presented in this dissertation: coastal and ocean resilience, local choices, and science and information. These sections touch upon the need for a “comprehensive, adaptive, integrated, ecosystem based, and transparent spatial planning process, based on sound science, for analyzing current and anticipated uses of ocean, coastal and Great Lakes areas” (Executive Order 13547, 2010). F urthermore, the Florida state legislature passed the Oceans and Coastal Resources Act in 2005 that calls for active management aimed at “restoring, rehabilitating, and maintaining the quality and natural function of [Florida’s] oceans and coastal resources .” The act calls for managers to take an ecosystem based approach supported by the development of regional goals and improved monitoring and assessment. In this way, coral reefs are included in a larger ecosystem that includes near field and far field impa cts affecting their health and management. In 2009, the Florida Coral Reef Protection Act was created to include the northern portion of the Flo rid Reef Tract in managing and protecting coastal resources beyond the Florida Keys (Figure 1 1). This action ex panded the management authority of the State of Florida and

PAGE 18

18 reinforced the concept of ecosystem based management that encompasses a broader geographic range. P articipatory decision support was used to synthesize knowledge about the marine and coastal envir onment in this area using an integrated ecosystem assessment process model. How Was Participatory Decision Support Used in Southeast Florida? Management of South Florida’s coastal environment is fraught with conflicting user group interests, multiple juri sdictional boundaries, and sometimes competing goals and management mandates. The Florida Reef Tract is one such area that poses a wicked environmental problem due to its regional geography and proximity to a major metropolitan area, impacts from upstream and upland water quality, impacts from climate variability and ecological events, and multiple management authorities. Participatory decision support was used to engage resource managers and decision makers in the development of the first comprehensive ass essment of South Florida’s marine ecosystem . From 20092014, 213 individuals contributed to a synthesis of the biophysical, institutional, and human dimensions science of the marine and coastal ecosystem that include d ecosystem services and societal needs (Nuttle and Fletcher, 2013a, b, c) . Concurrently, another study gathered manager’s perspectives on the content, design, and delivery of climate information as a decision support tool to manage coral reefs (Fletcher et al., 2014 in preparation) . Participato ry decision support was also used to construct and evaluate ecosystem scale and habitat scale scenario models. In all instances mentioned above, s takeholder involvement was required to gather data and information for managing this dynamic socioecological system. Addressing South Florida’s wicked environmental problem of managing marine and coastal resources comprised of multiple stakeholders, a complex humannatural

PAGE 19

19 system, and competing interests within the highly developed urban area requires a complex management response (Churchman, 1967; Rittel and Webber, 1973). The research approach described in this document integrated stakeholder participation in three phases: manager needs assessment, integrated ecosystem assessment (IEA) process framework , and decision support tools using Bayesian belief network s (BBNs) . Phase I: Needs assessment of climate information for use by m arine resource and coral reef manager s in southeast Florida Participatory decision support was used to identify researchers, managers, and nongovernmental organizations interested in improving the understanding of coral reef ecosystems for informed decision making. The Coral Reef and Marine Resource Manager Climate Information Needs Assessment for the Florida Reef Tract was conducted fr om 20112014 with funding support from NOAA Coral Reef Conservation Program grant award C085ID647 and NOAA/AOML in Miami, Florida (Fletcher and Hendee, 2012) . Informational interviews, a formal needs assessment, and analysis was completed to outline climate data desired by reef managers and the preferred delivery of that information . The participants in this assessment provided continued support as the test bed of endusers throughout the dissertation research. Phase II: Using the Integrated Ecosystem Assessment (IEA) process to transfer information to managers The South Florida IEA, also known as the Marine and Estuarine Goal Setting for South Florida project ( MARES) was a NOAAfunded project to improve the understanding of South Florida’s ecosystem. From 20092013, 125 stakeholders participated in the development of a synthesis of the ecosystem using the EBM DPSER framework (Kelble, et al., 2013) . Participants developed pictures of the ecosystem, integrated conceptual ecosystem models, and conceptual ecol ogical models. The IEA

PAGE 20

20 framework presented an opportunity for shared learning and an open dialogue to achieve consensus on the fundamental regulating processes of a healthy, sustainable South Florida coastal marine ecosystem (Nuttle and Fletcher, 2013a, b, c ). Phase III: Using participatory decision support to develop tools to improve the understanding of coral reef ecosystems for informed decision making Decision support tools are developed for assisting in analyzing decisions. Bayesian belief networks ( BBNs) are one type of decision tool based on probabilities or the likelihood of an event using past knowledge or information. The development of the models requires participation from experts and endusers. The product is a visual representation of the ecosystem, based on information gathered from phase I and II of this research. BBNs were developed by translating the cognitive maps from stakeholders’ minds from Phase II into interactive models to facilitate exploration of potential futures to stimulate hea lthy dialogues for planning and response strategies. The BBN s w ere designed, constructed, and reviewed to determine management scenarios for coral reefs . Research O bjectives The goal of this dissertation was to support interdisciplinary applied research f or an improved understanding of coral reef ecosystem s for informed decision making. The research included people, processes, and tools. Participatory decision support was the cornerstone of the research necessary to engage experts and endusers in achieving the goal. The hypothesis of this dissertation research was to asses s if participatory decision support using scenario modelling increases knowledge in the decisionmaking process . To address this hypothesis , th ree research objectives were designed:

PAGE 21

21 Obj ective 1. F acilitate p articipation of resource managers. Activities include d : 1) conducting a climate information needs assessment of coral reef and marine resource manager s, 2) coordinating participation in a South Florida IEA, and 3) maintaining communic ations with resource managers to promote an iterative process for developing decision tools . Objective 2. C onstruct a BBN decision support tool . Activities include d : 1) perform ing a literature review of existing BBNs used in natural resource management, a nd 2) designing , develop ing , and test ing the network. Objective 3. Determine the utility of decision support tools. Activit y include d 1) assessing decision tools for improving knowledge for informed decision making. Organization of the Dissertation This P h.D. dissertation has 5 c hapters, including the present introductory chapter (Chapter 1). Chapter 2 outlines participatory decision support and a coral reef and marine resource climate information needs assessment process and findings . Chapter 3 describes the stakeholder engagement activities used to characterize the coral reef ecosystem in southeast Florida and the creation of a prototype decision support tool, a BBN of the climate impacts to coral reef and hardbottom habitats . Chapter 4 expands the BBN mo del to an ecosystem scale and includes an evaluation of the BBN to obtain stakeholder input on the utility of the models to contribute to an improved understanding of the ecosystem for informed decision making. Chapter 5 summarizes the project results from the previous chapters including insights and recommendations for continued research in this field. References are included at the end of this document.

PAGE 22

22 Figure 11. South Florida’s coastal marine ecosystem is comprised of several habitat types and i s influenced by upstream oceanic processes and upland features (e.g., Everglades). (Adapted from Kruczynski and Fletcher, 2012) .

PAGE 23

23 CHAPTER 2 DECISION TOOLS FOR CORAL REEF MANAGERS: USING PARTICIPATORY DECISION SUPPORT TO INTEGRATE POTENTIAL CLIMATE IMPACTS AND INFORMED DECISION MAKING Introduction Coral reefs throughout the world are in significant decline (Wilkinson, 2000; Bellwood , et al. 2004; Carpenter et al. , 2008). Climate change, landbased sources of pollution, and fishing impacts have been the major drivers of change to the reef ecosystem (NOAA, 2009). In Florida, events, such as hurricanes have resulted in physical damage to reefs and coldwater events have caused coral to die resulting in overall declines in stony coral cover. The Florida Reef Tract located adjacent to the urbanized corridor along the southeastern portion of the Florida peninsula is no exception to this deterioration (Figure 2 1) (Gardner et al. , 2005 , Donahue et al. , 2008). While managers have developed strategies to monitor changes in reef health, challenges exist within the complex integration of biophysical, institutional, and human dimensions science to take action to reverse these declines (Donahue et al. , 2008; Keller and Donahue, 2006; Keller et al., 2009; NOAA, 1996 and 2007; Nuttle and Fletcher , 2013a, c; ONMS, 2011). These challenges include managing for far field drivers and pressures, such as global climate change that negatively impact the reef ecosystem (Wagner et al., 2010; Nuttle and Fletcher, 2013a, b, c; Lirman et al., 2014). Managing natural resources in a highly developed, urbanized area coupled with global climate change is a wicked environmental problem (Batie, 2008; Balint et al. , ___________ R eprint ed with permission from Fletcher, PJ, Spranger, MS, Hendee, JC, Li, Y, Clark, M, Kiker, G. in preparation. Decision tools for coral reef managers: Using p articipatory decision support to integrate potential climate impacts and informed decision making. Coral Reefs.

PAGE 24

24 2011). These problems are complex, are hard to define, have no easy solutions, involve multiple stakeholders, and are constantly evolving within a coupled socioecological system (Churchman , 1967; Rittel and Webber , 1973). Resource managers must respond to diverse stakeholder needs and rely upon the best available science to make sound management decisions, all the while recognizing that stakeholder needs can be driven by special interests, values, and desires (Conklin and Weil , 1997). One approach to address these dynamic socioecological systems is to incorporate participatory decision support tools into the management process. These tools can be both trans disciplinary and adaptive and aid managers in designing and evaluating soci o ecological management alternatives (Kiker et al. , 2008; Goosen et al. , 2007). Participatory decision support refers to a process of bringing people, sometimes called stakeholders (i.e. those with a stake in the outcome), together to develop a solution, idea, or concept to take action in the future (Raymond and Cleary, 2013; Sheppard, 2005). Stakeholders are individuals who have an interest in an issue or resource and are critical components of decision support project s (Ramirez , 1999). Pomeroy and Douvere (2008, p. 816) describe the importance of stakeholder engagement in the decisionmaking process within the context of marine spatial planning as: “understanding of the complexity of the ecosystem; understanding of the human influence on the ecosystem and its management; examining the compatibility and/or (potential) conflicts of multiple use objectives; identifying, predicting and resolving areas of conflict; and discovering existing patterns of interaction.” As such, participatory decision support is a valuable tool for planning and managing resources due to the robustness of data and information shared during the process, and the

PAGE 25

25 transparency of the approach that reaches across disciplines and helps bridge the gap in applying science to management decisions. When carefully designed, the participatory decision support approach can be useful to achieve agreement amongst diverse stakeholder groups. The method has the potential to integrate highly technical knowledge of scientists with the varied knowledge and values of nonscientists in order to make informed decisions, and to pr ovide reasonable assurance to justify those decisions (Lynam et al. , 2007; Lynam et al. , 2010). The technique presented in this chapter illustrates a process to gather technical information from a vari ety of sources and synthesize it into a format that has relevance to decision makers. This paper describes a participatory decision support methodology used to guide the development of climate information tools for coral reef managers in southeast Florida within the construct of three components: people, proc ess, and tools (Kiker et al., 2005). Each component is critical to obtain stakeholder input and to build a sense of ownership and commitment to the use of the product or results from the effort. The methodology consisted of several concurrent efforts to m atch stakeholder needs with climate researcher capabilities. From January 2010 to May 2014, a project coordinator identified stakeholders, conducted a reef manager needs assessment of climate information, guided the exchange of information between the researchers and managers, and facilitated a pre/post survey to evaluate the utility of climate information tools developed during this project. Accordingly, the chapter addresses the following objectives: 1. Review the participatory decision support methodology used within the context of people, process, and tools described by Kiker et al., 2005 2. Identify climate information and decision support needs of resource managers

PAGE 26

26 3. Survey and assess resource managers’ reactions to prototype decision support tools. Consequent ly, this chapter is divided into three parts. The first section outlines the participatory decision support methodology within context of people, process , and tools. The second highlights the climate information needs that emerged from the cooperative effort and participant reactions to the prototype tools. Finally, a discussion section sums up the lessons learned and highlights the next steps in this participatory decision support process. Methods Participatory decision support uses a bottom up iterative assessment that allows for stakeholder input, good research science, action, reflection, and communication (Sheppard, 2005; Raymond and Cleary , 2013). Needs assessments are a process that can be used to encourage participatory decision support to identify interests, capabilities, and evaluation of outputs. The needs assessment methodology consists of planning, data collection, data analysis and reporting as described by Witkin and Altschuld (1995), Alts chuld and Witkin (2000), and the National Oceanic and A tmospheric Administration’s Coastal Services Center training manual for Project Design and Evaluation (NOAA/CSC, 2003). The coral reef and marine resource manager climate information needs assessment in south Florida was achieved over the course of four years (Figure 2 2 ) by integrating three components: people, process, and tools (Kiker et al., 2005). People People are an essential component of participatory decision support research. In this study there were three key actors. First, coral reef and marine resource managers in southeast Florida (end users) who would be vested in decision support products;

PAGE 27

27 second, the researchers of prototype climate information tools; and third, a coordinator to liaise between the experts and end users. End users Coral reef and marine resource managers in southeast Florida were identified as the target enduser audience for the development of climate information tools. The deliberate selection of this segment of stakeholders was due to their decision making role in managing resources and resource use within the region. The individuals are responsible for taking management actions to study, protect, conserve, and allow the use of marine resources including the potential negative impacts from climate. In addition, the publicat ion of the National Coral Reef Conservation Program Goals and Objectives 20102015 identified climate impacts as a priority theme as this research project was being initiated and provided momentum to develop tools for managers to use in addressing climate impacts to reef resources (NOAA, 2009). Overall, 19 managers representing 15 resource management offices participated in the project between October 5, 2011 and February 25, 2014. The first step in this process was to identify these specific coral reef managers (the end users). This was accomplished by coordinator meet ing with staff from several of the resource management offices in the regi on. Five state and local agency offices were vi sited between October 5, 2011 and October 21, 2011. From these interact ions, key informants were identified and then contacted to confirm interest in participating in an assessment to examine climate information needs for managing the Florida Reef Tract. The second step in the process was to conduct a climate information needs assessment. Between December 16, 2011 and May 17, 2012, 15 individuals provided information and insights regarding managing resources along the reef tract and climate

PAGE 28

28 information needs and the format for the delivery of that information. From January 8, 2014 and February 25, 2014, 11 resource managers participated in a pre/post survey to evaluate climate information tools. Throughout the project period, interactions with end users consisted of facilitated focus group sessions, workshops, and oneonone m eetings with resource managers and researchers. The purpose of all interactions was to clearly articulate climate information needs, to identify recommendations for the design and delivery of that information, and to evaluate prototype climate information decision support tools. These meeting are described in greater detail in the Process section. Researchers Researchers developed prototype climate information decision support tools for decision makers during the project period. This was initially an independent effort being conducted by a group of researchers in early 2010. The project coordinator was informed of this group and secured permission to share the prototype tools with managers to obtain feedback on the content and delivery of the information. Four researchers were identified as potential contributors. These individuals had participated in climaterelated workshops held in Miami, Florida, USA on January 1114, 2010, and August 1516, 2011. Consultation with these researchers in early 2012 result ed in the development of three prototype climate information decision support tools, where two of the individuals worked collectively on one product. Prototypes were produced and shared with managers to gather feedback on the usefulness of these products f or improving the understanding of climate impacts to reef and marine resources for informed decision making.

PAGE 29

29 Coordinator The Sea Grant College Program integrates research, education, and extension for coastal issues (Balcom et al., 2013). Throughout the project period, a Regional Extension Coordinator was located at NOAA’s Atlantic Oceanographic and Meteorological Laboratory in Miami, Florida, USA. The coordinator served as a program liaison to the various governmental, nongovernmental, and academic insti tutions interested in marine resources in southeast Florida. The coordinator acted as a boundary organization linking end users with researcher capabilities throughout the project (Batie, 2008). Interactions between and among researchers and managers were organized in part, or entirely by the coordinator. The purpose of these meetings was to establish a conduit between those with data and information, and those end users desiring this information. The coordinator was responsible for carrying out informational conversations, developing and conducting a needs assessment, gathering researcher information, documenting and sharing findings, evaluating the prototype decision support tool, and maintaining a dialogue to link researcher capabilities with manager needs in the region. Process The needs assessment process was used to guide the participatory decision support project (Altshuld and Witkin, 2000; NOAA/CSC, 2003; Witkin and Altschuld, 1995). The approach included: stating the issue and intended audience, es tablishing a planning team, conducting an informati on and literature search, characterizing the audience, establishing goals and objectives of the needs assessment, and selecting data collection methods (Figure 22 ).

PAGE 30

30 Stating the issue and intended audience Preliminary information was gathered during a climate and marine protected area workshop that was held on January 1114, 2010 in Miami, Florida, USA. The purpose was to bring together researchers, managers, and academic partners who would identify interest in collaborating to develop a suite of climate information tools for resource managers. The workshop was attended by 28 individuals. Meeting notes were used to identify researcher capabilities, consistent terminology (e.g., defining what constitutes cl imate data), and manager perspectives. No formal summary report was produced, but the need for climate information that is easy to understand and access was mentioned on several occasions throughout the workshop. Establishing a planning team Following the January workshop, several exchanges among a smaller group of participants continued. On August 1516, 2011, a group session consisting of 14 individuals from government offices and academic institutions met in Miami, Florida, USA to outline current resear ch and identify opportunities to integrate this information into existing marine protected area management activities. Participants from these meetings worked together in smaller groups to develop proposals to fund research and outreach activities to implement the development and delivery of climate information tools for coral reef and marine resource managers in southeast Florida. Funding to conduct a climate information needs assessment was secured in 2012. Following the assessment, a planning team was f ormed that consisted of four members. These individuals were responsible for reporting results, refining decision support tools, and for writing proposals to continue funding climate tool development and climate literacy activities.

PAGE 31

31 Information and literat ure search Coral reef researchers and managers periodically come together to share information about the biophysical, ecological, and management of the Florida Reef Tract at workshops, conferences, or meetings. The project coordinator’s involvement in thes e events was strategically planned to maximize the opportunity to interact with the greatest number of potential participants interested in contributing to this effort. During these meetings, individuals were asked if they would be willing to partake in a preliminary assessment to identify climate information needs and to assist decision support tool development. For those interested, follow up meetings were scheduled at the stakeholder’s office or by phone when there was no option to meet in person. Indivi duals at these meetings provided examples of the information and literature being used by resource managers to improve their understanding of climate impacts to coral reef and marine resources. Characterizing the audience Informational conversations were c onducted with 14 individuals from five management offices in the region between October 5, 2011 and October 21, 2011. The coordinator scheduled meetings with a single representative or group of individuals from each office. The purpose was to document the current use of climate information within each office and to identify who would be the most appropriate person to complete the needs assessment, and most likely to use climate data in their decision making. The format for discussion began with introductions to present the project concept, to learn about current climaterelated activities in each office, and to provide a key informant or point of contact in each office for the upcoming needs assessment. There was no formal agenda to allow for openended conversation and the exploration of ideas and

PAGE 32

32 informatio n. The discussions lasted for one to two hours. Specific individuals from each office were identified by their peers, as well as some working in other offices. The key informants were then contacted to confirm: 1) they have a role in managing coral reef and marine resources; 2) to identify resource protection mandates their agency is responsible for implementing; and, 3) their interest in participating in the needs assessment. Researchers were also consi dered part of the audience in this project because of their involvement in not only developing tools, but also because they would receive feedback about their products from the resource managers. Researchers were selected using two criteria: 1) the individual was conducting climatefocused oceanic or atmospheric research, and 2) the researcher’s study site included a portion of, or the entire Florida Reef Tract. The researchers were also interested in the application of their science in management. Establi shing goals and objectives of the needs assessment A front end needs assessment tool was conducted in late 2011 and reviewed by human dimensions researchers at NOAA/CSC (Altshuld and Witkin, 2000; NOAA/CSC , 2003 and 2009; Witkin and Altschuld, 1995). These tools are specifically designed to gather information about stakeholders prior to better characterize the situation at the start of a project. T he tool consisted of 23 openended questions , or questions that allow for narrative responses with three exampl es of prototype decision support tools. The questions were divided into two categories. The first set of questions was designed to gather general marine resource management information and background on the resources within their jurisdiction. The second s et centered on climate information needs and the preferred delivery of that information. Prototypes were included with a

PAGE 33

33 limited description of each to gather a first impression of the types of tools being developed to obtain feedback to refine or improve the products and ease in interpretation. Selecting data collection methods Data collection consisted of informational conversations, interviews, and a pre/post survey conducted with coral reef and marine resource managers in the region. The conversations w ere used to gather information about climate information needs and to gather input for the design and delivery of climate tools. Several months later , the first of several climate information tool s were developed. A pre/post survey was designed to gather attitudes and beliefs about climate information decision support tools where participants could choose between five responses (e.g., 1=not useful, 2, 3=moderately useful, 4, 5=very useful). The basis for the use of this format was to assess the participant’s level of agreement or disagreement related to a series of questions about a prototype climate information tool. Data collection was carried out inperson, by phone, or using Internet conferencing (GoToMeeting, 2014). Questions were asked and responses recorded on paper, or were submitted electronically through E mail. Each interaction took an average of one to two hours to complete. Informational conversations and interview responses were transferred into a spreadsheet and ranked with those most frequently mentioned placed at the top of the list and those mentioned the least at the bottom of the list. Pre/Post survey data were entered into a spreadsheet and a median was calculated to show the central tendency of responses (Boone and Boone, 2012).

PAGE 34

34 Tools Three prototype decision support tools were identified for use in the needs assessment during the August 1516, 2011 workshop (Fletcher and Hendee, 2012). The coordinator shared these prototype products with managers to obtain feedback for further development following the informational conversations phase of the project. The tools were shown to managers during the needs assessment from December 16, 2011 to May 17, 2012. One of the products was a sea surface temperature map that provided a spatial visuali zation of the potential for coral reef bleaching based on increased temperatures (van Hooidonk, 2014). The second prototype was a weather typing map developed to identify relationships between weather patterns and the presence of elevated chlorophyll levels in the water column as detected by satellites (Sheridan et al., 2013). The third was a monthly onshore reef flux report created to explain the movement of nutrient rich deep water onto the nearshore shelf or reef system (Gramer et al., 2008). This repor t provided background information on the monthly onshore flux and an automated system was created to provide electronic (E mail) alerts detailing changes in water quality along the southern portion of the Florida Reef Tract. Further tool development was c arried out beyond the development of the three tools described prior due to feedback gathered during the needs assessment. More specifically, managers stated a desire for information beyond climate and wanted something that included a map. In response, res earchers redesigned the Coral Health and Monitoring Program (CHAMP) data portal website to meet the format and delivery specified by managers ( http://www.coral.noaa.gov/champportal/ ). CHAMP is a comprehensive website that includes research, data, and resources to improve the

PAGE 35

35 understanding of coral reef ecosystems around the world. The CHAMP portal provides in situ and remotely sensed data being collected at coral reef locations around the globe. The content of the portal included a user’s guide, a mapbased query tool showing coral reef monitoring stations, and a feature to download data (Figure 2 3 ). A formal evaluation of the CHAMP portal was conducted from January 8, 2014 to February 25, 2014. A pre/post survey was conducted with 11 resource managers in southeast Florida; all but two had participated in the initial needs assessment. Ten of the surveys were completed inperson and one using online conferencing (GoToMeeting, 2014). One session included t hree individuals working for the same agency with varying degrees of responsibility at the local and regional levels. The survey was designed to obtain a “first impression” of the CHAMP website and to assess the utility of this as a decision support tool f or resource managers. Each participant was asked to listen to a project overview, complete a five question pretest survey, work through a hands on tutorial of the CHAMP website, and complete a seven question post test survey. The pretest survey began wit h a question on the decision making process at the respondent’s home institution or agency. The purpose of this question was to obtain insight about how decisions are made, to identify the respondent’s role in the process, and to explain where and how CHAM P data can be applied to decisionmaking. The second question requested that the respondents identify the type of manager they are (e.g., program manager, project manager, manager, researcher, or other) as a direct response to listening to the varied responsibilities of the individuals who participated in the needs assessment. The third question asked how often managers accessed climate information to complete work related responsibilities. The final three questions on the

PAGE 36

36 survey were tailored to identify: 1) if managers thought decision support tools would be useful for making management decisions related to climate impacts to marine resources and coral reef ecosystems; 2) if managers would consider using decision support tools detailing pressures and stres sors to marine resources and coral reef ecosystems in future management decisions; and, 3) if a manager would recommend using a decision support tool to other marine resource managers within their management portfolio. Optional responses consisted of yes o r no, or a fivepoint scale response (1=not useful, 2, 3=moderately useful, 4, 5=very useful, or 1=No, 2=Doubtful, 3=Maybe, 4=Likely, 5=Absolutely). Respondents were encouraged to elaborate on their responses to each question to share additional informat ion about their responses. The tutorial included a brief overview of the CHAMP website and a series of steps for managers to take to explore the page (Figure 2 3 ). Managers were asked to look at a particular coral monitoring station and the data contained on the website. Then they were instructed to look at temperature data for a particular date and time, and to download the dataset associated with that information. They were also asked to explore the user’s guide. A post test survey was administered after the tutorial. The post test consisted of seven questions; three were identical to the pretest with specific mention of the CHAMP web page as the decision support tool. Two of the questions were designed to conduct a longterm assessment of the decision s upport tool and will be reported in another research paper. The final two questions on the post test were structured to identify if the tool increased the respondent’s knowledge of climate impacts to coral reefs and marine resources, and if the manager bel ieved they would use the tool in the future. A fivepoint scale was used for the former (1=not useful, 2, 3=moderately useful, 4, 5=very useful), and a yes/no response for the latter.

PAGE 37

37 Results People Informational conversations with the broader group of individuals working in resource management or climate research provided general information about the breadth and extent each management office or individual was involved climaterelated activities. The informational conversations provided a plethora of information about manager interests, concerns, data needs and preferred delivery format. General information about the need for more monitoring, research, and synthesis was consistent among offices. In total, 15 managers participated in a coral reef and mar ine resource manager climate information needs assessment for the Florida Reef Tract within the project boundary stretching from St. Lucie Inlet to Key West, Florida (Fletcher and Hendee, 2012). Three of the managers provided a single response from their office, so the total number of assessments received was 13. However, the responses on this survey were recorded individually because each manager was responsible for a different spatial domain consisting of a state park, state recreation area, and a regiona l management program responsible for a total of 19 managed areas within the study site. The results contain a listing of climate information currently used (Table 2 2), needs identified by managers (Table 2 3), information that is difficult to obtain or interpret (Table 2 4), and recommendations for designing an Internet based discovery prototype for climate decision support tools (Table 2 5) and its delivery (Table 2 6). In addition, ancillary information was reported to help guide the development of climate information tools. The top three data needs that the managers categorized as difficult to obtain and interpret consisted of sea lev el rise/tides/storm surge modeling, temperature data, and

PAGE 38

38 Gulf Stream current movements (Table 2 7). This information was identified as useful for making climaterelated decisions. Process The needs assessment process resulted in a systematic structure to characterize the situation and obtain information from stakeholders. A stepwise approach outlined by the NOAA/CSC (2003) Project Design and Evaluation training manual was used to guide the development and implementation of group sessions, informational conversations, interviews, and a pre/post survey. Inefficiencies in the process were tied to operational aspects related t o the difficulties in securing time to meet with managers who often had limited time for conversations to provide an indepth introduction to the topic and sufficient time to conduct the surveys. For this reason, some of the post surveys were completed aft er an inperson meeting and sent to the coordinator shortly afterwards. It must also be noted that the coordinator had been involved in reef related activities for 16 years in this region and was familiar to several of the managers partaking in this projec t. These longterm relationships that had been built often aided the willingness of managers to meet to participate in this project. The information shared during conversations far exceeded the survey questions. In several instances, this information was recorded, but not included in the survey reports or publications. However, it was helpful to form the foundation for future activities and exploration of additional insights to the management of marine and coastal resources. These include a broader approach to managing coral reef resources beyond climate, looking at the multiple lines of pressures and stressors by which managers can respond and intervene. One such instance is the development of a new cognitive map for the decision making process at one of the regional management

PAGE 39

39 agencies (Figure 24 ). Gaining insight to the decision process can help identify where managers can intervene and if coupled with appropriate tools can aid in decision making. The needs assessment process was a fundamental variable in describing the varied level of stakeholder’s responsibility in managing resources. It is a critical element in understanding the type of information needed by managers to support their decision making. Insights from listening to remarks during the infor mational conversations and the needs assessment led to the inclusion of a question in the pre/post survey that asked managers to define their management role(s). The purpose of doing so was to recognize manager motivations for needing or using climate information within the broader understanding of managing marine and coastal environments. During the pre/post survey, verbal comments such as “I will not use this because it does not relate to my primary responsibilities”, “I have more accurate data at my off ice”, or “I will definitely use this” provided further awareness about the varied roles which managers serve and the level of involvement in implementing decisions. Tools Three prototype climate information tools were shared with managers to obtain feedbac k and comments on their utility for conveying data for decision making during the needs assessment (Fletcher and Hendee, 2012). First, a map showing Degree Heating Weeks (DHW) and Sea Surface Temperature (SST) in a marine protected area (Papahanaumokuakea Marine National Monument, Hawaii, USA) were developed to illustrate areas at risk of coral bleaching in a map format that could be developed for southeast Florida (van Hooido nk and Huber, 2012). Second, a weather typing product resulted in the identificati on of ten weather patterns for South Florida and the relation to

PAGE 40

40 chlorophyll levels in coastal waters as reflected in satellite imagery (Sheridan et al., 2013). Third, the monthly onshore reef flux report was created and was comprised of a suite of ecological forecasts that illustrate the movement of water in the region and the possibility of changes in water quality based on satellite derived products and in situ monitoring stations along the reef tract in the Florida Keys (Gramer et al., 2008). All products were shared with participants, although not all commented on them during the informational conversation. Information needs ranged from more general information about resources in the region, to specific datasets tied to climate. Many data needs were r eported that were not climate specific, but were broader data needs that could help assess climate impacts and other stressors to the ecosystem. The most requested climate data was data from peer reviewed literature, monitoring information (delivered via t he Internet), and temperature data, both temperatures at the sea surface and on the reef crest (Table 2 3). The multitude of information and content requested during this initial assessment clearly supported the need for a suite of decision support tools s ince no single tool could meet the information needs. In response to this need, the CHAMP webpage was updated and evaluated as a decision support tool for resource managers in southeast Florida. The CHAMP website pre/post survey resulted in several new ins ights and quantitative results. The pretest question to identify the decisionmaking process illustrated that the process is varied in each institution. However, a general observation was that managers provide sciencebased recommendations to upper level decision makers. As such, when respondents were asked to identify their own management role, they described these roles as primary, secondary, tertiary, and quaternary roles

PAGE 41

41 showing the dynamics of responsibilities of resource managers surveyed (Table 2 8) . They were also asked to state the frequency at which they access climate information for work related purposes: four did this weekly, three did it daily, one did it once a year, one did it six times a year, one did it monthly, and one did it every other week between May and November. Table 2 9 presents the results of the pre/post survey questions using a five point scale as well as one quest ion with a yes/no response. The median response for managers stating if decision support tools would be useful for m aking management decisions related to climate impacts to marine resources and coral reef ecosystems was a four , or between three (moderately useful) and five (very useful) for both the pretest and post test. Likewise, the response was the same when asked if they would consider using a mapbased tool and/or an ecosystem model detailing pressures and stressors to marine resources and coral reef ecosystems in future management decisions responses. The only change in the median value of responses occurred when asked if they would recommend using a mapbased tool and/or an ecosystem model to other marine resource managers within their management portfolio. The median value changed from a four (likely) to a five (absolutely) in the pretest and post test respecti vely. The post test survey asked two additional questions. The median response for the degree to which the CHAMP mapbased query tool increased knowledge about climate impacts to marine resources and coral reef ecosystems was three (moderately useful). The final question asked if they believed they would use the CHAMP tool in their decision making, ten individuals said yes, and one stated no. The one individual who stated they would not use the tool was due to their ability to obtain raw data from research being conducted within their office.

PAGE 42

42 The pre/post surveys were useful to gather an overall view of the utility of the CHAMP website by the target audience. The manager’s responses changed somewhat, for example one manager may have changed their response fr om a four to a three, but the median responses stayed the same for the sample size. Beyond the survey, comments such as “wow,” “that’s neat,” “I like this” were noted, but not recorded in the responses. In addition, informational conversations and discussi ons after the pre/post survey were helpful to gain a better sense of how managers thought they could use the tools to carry out work related responsibilities. These interactions led to further research in developing the tools that was ongoing beyond the dissertation research. Implications Wicked environmental problems are complex and thus require complex solutions. They are volatile and evolving, involve multiple stakeholders, possess dynamic constraints and often there is no definitive solution to address the difficult, complex problem (Conklin and Weil, 1997). Managers, specifically those in Florida’s densely populated southeast coast , are dealing with very wicked environmental problems. They must balance the needs of multiple stakeholder groups and resources taking into account the varied stakeholder needs while maintaining livelihoods and ecosystem integrity. This project identified the climate information needs of managers working in a coupled socioecological system, and provided a participatory decis ion support approach to assess what information is needed to begin to tackle the wicked problem of managing resources impacted by climate variability and a multitude of other factors. The people, process, and tools methodology helped structure a discussion for the wicked problem and a starting point to begin to understand the complex uncertainties tied to

PAGE 43

43 managing climate impacts to coral reef ecosystems and connectivity to other systems, and the tools currently available (Batty, 2013). People Participatory decision support was used to identify researchers, managers, and nongovernmental organizations interested in improving the understanding of coral reef ecosystems for improved management. The project developed a foundation for a common knowledge domain for future actions. The informational conversations and assessments were most valuable for translating ideas from stakeholders’ minds into a list of needs and guidance for decision support tools. The interactions foster the exploration of new tools that can stimulate healthy dialogues for planning and response strategies. Participatory decision support is a process that brought awareness to the value of multiple ideas and shared knowledge. Goosen et al. (2007) emphasized the value of a demanddriven versus a supply driven process where tool development guided by endusers can stimulate knowledge and learning among the participants. The project coordinator was invaluable in designing, developing, and implementing stakeholder involvement and the development of a collaborative manager researcher climate information tool. Much time was spent by the coordinator liaising among offices and maintaining contact with managers to illustrate a vested interest in the outcome of the project by all those involved. Liaison wor k is important to facilitate the process acting as a honest broker interested in the value of the outcome between the expert and endusers and serving as a boundary between the two interest groups. Liaison work is also valuable to provide a single point of contact to field questions and filter responses, however the time commitment must be established to ensure timely responses and

PAGE 44

44 individual attention to the stakeholders involved in the project. In addition, managers spent several hours with the coordinator during one or all of the project interactions consisting of informational conversations, the needs assessment, and the pre/post survey. Weichselgartner and Kasperson’s (2010) stated that both good information and a coordinated process for sharing are requirements of knowledge and initiative for moving science to action. The term ‘manager’ had a broad range of meaning in this study. Managers may manage people, programs, or projects. They may make decisions about resource management, or feed into the deci sion making process by providing science or interpreting results of scientific studies to their colleagues or superiors. In some cases, they may perform one or a combination of these roles. All of those involved have some responsibility to manage coral or marine resources and their information needs varied based upon their job description. Ultimately, what each manager is responsible for is different and consequently, there are varying needs for an end product. These may range general information to spreads heets of data, or real time data to historical or event driven data. Process Curtice et al. (2012) identified the failings of ecosystem based management as a result of a lack of resources, funding support, or inappropriate tool developers. This effort focu sed on a front end assessment to characterize the types of climate information and the delivery of that information and matched those needs to researcher capabilities. It is noteworthy to mention that the involvement of endusers in the development of products can assist in unlocking integrated, dynamic issues fraught with uncertain information. This can lead to progress in learning and opportunity driven

PAGE 45

45 solutions that would otherwise not be identified in a confined approach where tool developers did not i nteract with stakeholders (Conklin and Weil , 1997), Informational conversations and the formal needs assessment provided information beyond the scope of the study, but will be useful in providing context for tool development tied to this project and future activities. The first round of meetings at the manager’s office provided an opportunity to meet several staff and learn about agency programs. Discussions and comments were invaluable to learn more about manager perceptions of climate information and immediate and longterm needs at a local level. These interactions also helped develop a relationship with endusers that is needed to fully develop the suite of decision support tools. Tools Recommendations for developing prototype climate decision support tools from several individuals illustrated that the participatory decision support process proved useful in gathering manager needs and researcher capabilities for use in developing climate information products for endusers. Changes and enhancements to the three examples of climate information products in this study served local management needs. Managers stressed the importance to have more information about specific areas of the reef, those that fall into a more local jurisdiction. In addition, several expressed the need for both general reef ecosystem and climatespecific information. At the time of publication, it was unclear if these prototype tools would continue to be further refined. However, new tools and the improved delivery of information as descr ibed by managers would continue. In summary, this research described a participatory decision support methodology consisting of people, process, and tools that guided the design of climate

PAGE 46

46 information for coral reef managers in southeast Florida. All three components were required to gather input to characterize the information needed and to begin the discussion about how decisions are made and the use of climate and other information in the decision process. While the prototype tool is still in its infancy , it shows promise for use by this enduser group and its use will be monitored to assess its utility (e.g., Google Analytics) over time and to obtain additional input during future updates. This project benefitted greatly from the concurrent efforts to develop climate information tools and the longterm interactions among the reef manager community in southeast Florida. Additionally, the ability to identify appropriate resource managers, researchers, and a coordinator and their willingness to participate in this exploratory activity were invaluable in completing the project objectives within four years. The value of stakeholder input was critical to obtain stakeholder input, to gather a “first impression” of the ease of access and use of the tool, and to begin to build a sense of ownership and commitment to the use of the product. It was also critical to understand that while managing for climate impacts is needed, it is a far field influence that may not have actionable interventions. For this reason, the broader suite of information about the reef ecosystem in CHAMP proved useful in assessing and determining potential management actions tied to near field and far field influences on the reef. The wicked issue of managing biophysical, institutional, and human dimensions of coral reefs will be revised and evaluated periodically as issues and technology evolve.

PAGE 47

47 Figure 21. Map of the Florida Reef Tract ( adapted from Kruczynski and Fletcher, 2012).

PAGE 48

48 Figure 22 . Timeline of needs assessment process components and activities used to complete climate information tool development for the Florida Reef Tract.

PAGE 49

49 Map based query tool First Impressions The Coral Health and Monitoring Program (CHAMP) is a web presence that highlights the integration of satellite, in situ, modeled, and other sources of meteorological and oceanographic data in near real time, for the purpose of eliciting ecological forecasts for coral bleaching and oceanographic events with the capability to extend to other marine environmental events. These ecological forecasts, or "ecoforecasts" are aimed at informing poli cy and management decisions. CHAMP developers are interested in updating the mapbased query tool for resource managers in southeast Florida. For the next series of questions, please explore the CHAMP website using the Molasses Reef site as an example of what the site contains and its layout. Let’s imagine we want to see the SST at Molasses Reef on 2 specific days. Select the Molasses Reef location from the Station/Sensors section of the page. Select monitoring data for that site by highlighting parameters (hold shift and select several) and then right mouse click to display the last 180 days of data. Examine the Plots to the right of the map. There are 2 tabs, one is User Guide the other Plots . Scroll down until you see the SST plot. Put your cursor over t he blue line to see dates and temperatures representing peaks and valleys in the graph. Locate the high temperature for September 12, 2013 (30.46 C). At the bottom of the page you will find the Data display showing all of the monitoring information for that site. Download the data in commaseparated values (CSV) by clicking on the Download tab and saving to your computer. Open the file in Excel and look for the SST value on July 23, 2013 (29.88 C). If you need assistance, select the User Guide located on the right panel of the page. Figure 23 . Sample of the Coral Health and Monitoring Program mapbased query tool tutorial used during the pre/post test.

PAGE 50

50 Fig ure 24 . Cognitive map of the decision making process used in one management agency in So uth Florida.

PAGE 51

51 Table 21. Managers who participated in the CHAMP pre/post survey. Name Title Agency Ken Banks Natural Resource Specialist IV Broward County Environmental Protection and Growth Management Department Rene Baumstark Associate Research Scient ist Center for Spatial Analysis Information Science and Management, Fish and Wildlife Research Institute, Florida Fish and Wildlife Conservation Commission Karen Bohnsack NOAA Coral Fellow Consultant ECS, Florida Department of Environmental Protection, Coral Reef Conservation Program Billy Causey Regional Director Southeast Atlantic, Gulf of Mexico and Caribbean Region, NOAA’s Office of National Marine Sanctuaries Ernest M. Cowan Chief Biologist Bureau of Parks District 5, Florida Park Service Steven R. Dale Park Manager III John U. Lloyd Beach State Park, Florida Park Service Lou Fisher Retired Natural Resource Specialist III Marine Section, Broward County Environmental Protection and Growth Management Department Kathy Fitzpatrick Coastal Engineer Mar tin County Coastal Engineering Department Charles Jabaly District Biologist Bureau of Parks District 5, Florida Park Service Janet Phipps Coral Reef Ecologist Palm Beach County Department of Environmental Resources Management Erik Stabenau Oceanographe r Department of Interior, Technical Lead, National Park Service

PAGE 52

52 Table 22. Marine resource data currently used by the 15 managers participating in this study. Numerical totals show most frequent mention of data used by managers. A “1” was assigned for each occurrence, “0.1” was assigned for ancillary mention of the listing (i.e., mention of State technical reports=State Guidance documents.) Marine Resource data currently used Frequency of use p eer reviewed journals 6 s tate guidance documents/Best Ma nagement Practices 5.2 listserv/forums (topics: harmful algal blooms, turtles, corals) 5 visuals/resource mapping/benthic habitats 5 GIS/GoogleEarth/Everglades Restoration/database (geodatabase) 5 office committees/regional teams/personal communication s (focus on topics) 4 damage minimization procedures or recommendations 3 NOAA protocols/guidance documents 2.5 personal observations 2 real time data 2 u niversity r esearch 1.5 Southeast Florida Coral Reef Initiative 1.2 white papers 1.1 American A ssociation of Underwater Scientists newsletter 1 Florida Reef Resilience Program 1 on the ground programs producing information (data and delivery process) 1 web based reporting (e.g., NOAA Integrated Coral Observing Network , Bleachwatch , etc. ) 1 monit oring protocol manuals 1 c onferences 1 s ea surface temperature data from NOAA web page 1

PAGE 53

53 Table 23. Climate information needs identified by 15 managers participating in this study. Numerical totals show the frequency each was mentioned during the as sessment. A whole number “1” was assigned for each occurrence, “0.1” was assigned for ancillary mention of the listing. Climate information data needs (Q11) Frequency of mention d ata from peer reviewed literature 5 monitoring information delivered via we b 5 temperature (sea surface and on reef crest ) 4.1 s cience recommendations (being discussed in “4 County compact”) 3 current research findings (quick, easy access) 3 b est management practices 3 damage minimization procedures or recommendations 3 c ar bon dioxide in atmosphere 2 tide/sea level rise 2 s cenario/predictive tools to get concepts across to audiences 2 weather (including air temperature; patterns of rainfall from climate change and impacts on water supply and delivery to the coast) 2 agen cy website data 1.1 online videos about climate science 1 Impacts to marine resources 1 Intergovernmental Panel on Climate Change 1 finer scale/spatial resolution compatibility (Gulf of Mexico to Florida Keys) to get loc al and regional perspectives 1 reef crest data 1 Gulf stream movements (eddies and currents) 1 peer reviewed data 1 visualizations (e.g., CanViz) for South Florida 1 science based information for sharing with th e public 1 present data with a balanced view 1 data that has verified and good references 1 Florida Reef Resilience Program data 1 Precipitation 1 data to help manage resources (e.g., artificial reefs as managem ent do/not help address climate change concerns) 1 analysis and trends in wildlife data reports sho w ing increases or decreases in habitat etc. 1 US Geological Service groundwater data (salt water intrusion) 1 NOAA long term data sea le vel records 1 N ational Park Service data records 1 long term salinity 1 pH 1

PAGE 54

54 Table 24. Climate information that is difficult to locate or interpret as identified by resource managers during formal needs assessment. A whole number “1” was assigned for each occurrence, “0.1” was assigned for ancillary mention of the listing. Climate information that is difficult to locate or interpret (Q12) Frequency of mention sea level rise /tides/slosh 5.1 temperature (downscaling) 3 .1 Gulf Stream information/patterns (e . g ., event maps) 3.1 w eather stations 2.1 downscable to this region of the world 2 rainfall patterns 2 more meteorological data 2 water quality (nutrient levels , detection limits , microbial activity) 1.1 DBhydr o (water quality database housed at regional water management office) 1.1 physical oceanography 1.1 subset of Coastal Marine Automated Network (C MAN) station information 1.1 current/accurate/regularly examined 1 changes in reef species and other biolo gical data 1 a tmospheric carbon dioxide 1 benthic mapping (expansion of Dendrogyra ) 1 ocean acidification 1 linking human activity to impacts 1 integrating what we know to impacts 1 vulnerability analysis (e . g . , sea level rise and scenarios ) 1 a irpo rt weather data 1 fisheries data of all life stages (younger year classes) to help make management decisions 1 all physical measurements 1 Water Management District data 1 in situ observations 1 pH 1

PAGE 55

55 Table 25. Recommendations for the design of a climate information decision tool. Recommendations for the design of climate information (Q21) Frequency of mention v isuals (appropriate graphs and maps) 6 t ime scales (10yrs, 15yrs, events ) 5 r egional scale (entire Florida Reef Tract) 4 i mprove data (more water quality monitoring data) 2 w ebinars 2 a pp for climatology 1 t hink o f being the end user (children, boaters, etc. ) 1 c artoons for illustrative purposes (e.g., persistent blooms) 1 r eal time Gulf Stream location 1 s ea surface temperature of Gulf Stream water 1 r eal time satellite data 1 f act sheet with climate impacts 1 i ntegrate multidisciplinary informati o n 1 Add “what can you do” for managers 1 Table 26. Recommendations for the delivery of a climate information decision tool. Deliv er of climate information Frequency of mention l istservs 11 u se (well designed) web based interface 10 n ewsletter 5 P e er reviewed publications 4 s ocial media 1 s ynthesis of information (don’t really need more information, just improve delivery) 1 p u blic service a nnouncements 1 u se a map in the product 1

PAGE 56

56 Table 27. List of recommendations to improve the climate data product prototypes developed during the project period. make smaller/finer resolution grid size on reef flux report u se macroalgae versus chlorophyll a in the weather typing tool Email these [alerts] with messages need more interpretation to fully understand and apply to management decision making d esign these similar to harmful algal bloom report, send via email and attachment r egional scale too large for county analysis potential t o discuss why there are changes (e.g., Gulf Stream meanders, water quality , etc. ) expand to northern portion of the reef tract did not use graph [in the weather typing] too much make newsletters an d ask managers how they incorporate this new science integrate fisheries data give status report with days, last month, last 10,20,30 y ears include Southeast Environment al Research Center reports add info on h ow you get from one weather pattern to anot her and what is normal to extreme use these for hindcasting to identify management strategies for the future (look for precursors) good to help understand climate science, not sure how to use for decision making at larger scale level

PAGE 57

57 Table 28. A pr e test survey was conducted to identify the type of management responsibility for each respondent. Eleven managers were asked to describe their management role in their institution. They were provided with a list of management types with descriptions and asked to select those that apply, and to rank the order of their responsibilities in each role if they selected more than one, or provided an “other” management role. Management type Description Primary Role Secondary Role Tertiary Role Quaternary Role Pro gram Manager one who manages a program that responds directly to organization 3 Project Manager one who manages projects that directly support program goals 1 4 1 Manager one whose primary role is to manage staff to complete programs and projects 2 1 1 Science Manager one who manages the collection, analysis, and reporting of organizational science 2 1 1 Researcher one who is technically engaged in scientific investigations and inquiry 2 1 1 Other Coastal engineer 1 Other Retired reef resou rce manager 1

PAGE 58

58 Table 29.Survey responses to questions relating to a marine and coastal resource decision support tool (n=11). Questions Alternate responses Pre survey Post survey (Map based query tool) Response(%) Median Response(%) Median Do y ou think decision support tools would be useful for making management decisions related to climate impacts to marine resources and coral reef ecosystems? 1 Not useful 2 3 Moderately useful 4 5 Very useful 0 1 (9.0) 3 (27.2) 3 (27.2) 4 (36.3) 4 0 0 2 ( 18.1) 4 (36.3) 5 (45.4) 4 Would you consider using a map based tool and/or an ecosystem model detailing pressures and stressors to marine resources and coral reef ecosystems in future m anagement decisions? 1 No 2 Doubtful 3 Maybe 4 Likely 5 Absolutely 0 0 2 (18.1) 5 (45.4) 4 (36.3) 4 1 (9.0) 0 1 (9.0) 4 (36.3) 5 (45.4) 4 Would you recommend using a map based tool and/or an ecosystem model to other marine resource managers within their management portfolio? 1 No 2 Doubtful 3 Maybe 4 Likely 5 Absolutely 0 0 2 (18.1) 4 (36.3) 5 (45.4) 4 0 0 2 (18.1) 2 (18.1) 7 (63.6) 5 Post survey Questions (Map based query tool) Alternate responses Alternate responses Median Rank the degree to which the tools increased your knowledge about climate impacts to marine resources and coral reef ecosystems? 1 No 2 Doubtful 3 Maybe 4 Likely 5 Absolutely 1 (9.0) 2 (18.1) 3 (27.2) 4 (36.3) 1 (9.0) 3 Do you believe you will use one of these tools in your decision making? Yes No 10 1

PAGE 59

59 CHAPTER 3 USING THE INTEGRATED ECOSYSTEM ASSESSMENT FRAMEWORK TO BUILD CONSENSUS AND TRANSF ER INFORMATION TO MANAGERS Introduction Ecosystem based management (EBM) can be defined simplistically as an operational strategy to manage the ecosystem in a holistic manner and is w idely accepted as a method to improve natural resource management (McCleod et al., 2005; Szaro et al., 1998; Christensen et al., 1996). A primary goal of EBM is to achieve a balance between the needs of society, the environment, and institutional arrangements (Slocombe, 1993b ). To achieve such a balance requires knowledge regarding the complexities and intersection of biophysical science, human dimensions, and governance mechanisms (Altman et al., 2011; Kelble et al., 2013; Rosenberg and McLeod, 2005; Sloco mbe, 1993b ; Szaro et al., 1998). Implementing EBM poses numerous challenges to decision makers. Many of these challenges are tied to understanding the interconnectedness of the dynamic environment and the coupled socioecological system, often when scient ific information and management processes do not efficiently serve one another. These challenges become pressing concerns to address in a timely manner as coastal populations grow and declines in the health of marine and coastal ecosystems continue (Keller et al., 2009; Leslie and McLeod, 2007; MEA, 2005; POC, 2003; USCOP, 2004; WHCEQ, 2010). The primary challenge facing managers is assimilating complex information, drawing substantive inferences from this information, and then translating these _________ __ R eprint ed with permission from Fletcher, PJ, Kelble, CR, Nuttle, WK, Kiker, G. 2014. Using the integrated ecosystem framework to build consensus and transfer information to managers. Ecological Indicators .

PAGE 60

60 inferences into concrete and defendable managem ent actions. Integrated Ecosystem Assessments (IEAs) are a process designed to overcome or, at least, minimize this challenge (Levin et al. 2009; Reiter et al. 2013). IEAs provide ‘a formal synthesis and quantitative analysis of information on relevant na tural and socioeconomic factors, in relation to specified ecosystem management objectives’ (Levin et al., 2009). The IEA approach includes scoping and stakeholder engagement to aid in defining ecosystem goals, targets, and indicators. Thus, the IEA proces s provides opportunities to bring together stakeholders for developing consensus designed to inform EBM implementation and decisionmaking (Tallis et al., 2010; deReynier et al., 2010; Levin et al., 2009; Levin et al. 2013). Because the role of these inter actions in the IEA framework is not explicit, a modification of the IEA process has been proposed to increase the interactions with management and provide tangible products to support management decisionmaking (Reiter et al., 2013). IEA and other scientif ic processes designed to inform EBM should in their early stages address the need for building a sciencebased consensus of the current state of the ecosystem (Harwell et al., 1996). The consensus building process is reliant on active, two way stakeholder engagement. Scientists, resource managers, and stakeholders need to be engaged at the onset to present their scientific knowledge of the ecosystem and develop a consensus of not just how the ecosystem functions, but also how humans interact with the ecosys tem. After developing the sciencebased consensus, management and stakeholder entities must be engaged to communicate this consensus in an effective manner to further their missions.

PAGE 61

61 In South Florida, the MARine and Estuarine goal setting project (MARES) sought “to reach a sciencebased consensus about the defining characteristics and f undamental regulating processes of a South Florida coastal marine ecosystem that is both sustainable and capable of providing the diverse ecosystem services upon which our society depends.” This goal is directly aligned with the IEA process in that it seeks to develop scientific consensus through the development of EBM goals and indicators. Moreover, MARES and IEAs both have the specific aim of conducting this consensus buil ding to increase the use of science in resource management decisionmaking. We will discuss how MARES applied the established IEA steps for defining EBM goals, developing indicators, conducting risk analysis, and evaluating scenarios and supplemented thes e with continuous and ongoing engagement with managers in an attempt to improve the transfer of scientific knowledge into resource management decision making in south Florida. Materials and Methods South Florida’s coastal marine ecosystem is located at the southern tip of the Florida peninsula. The MARES domain spanned from the Caloosahatchee River Estuary on the west coast, through the southernmost region of the Florida Keys and Dry Tortugas, and to the southeast coast northward to the St. Lucie River Estuary (Figure 3 1). To better engage stakeholders and resource managers, we divided South Florida into three subregions: Southwest Florida Shelf, Florida Keys and Dry Tortugas, and Southeast Florida Coast. MARES focused on four steps in the IEA process 1) s coping, 2) indicator development, and 3) risk analysis, and 4) evaluating scenarios. However, MARES had as its central hypothesis that developing consensus through integration between natural

PAGE 62

62 systems scientists, human dimensions scientists, resource managers, and nongovernmental organizations (NGOs) would improve the use of science in resource management decisionmaking. As such, MARES had a leaders group with representatives from Everglades National Park, Florida Keys National Marine Sanctuary, University of Miami, Florida International University, the National Oceanic and Atmospheric Administration, Florida Department of Environmental Protection, Florida Fish and Wildlife Conservation Commission, and Audubon of Florida to ensure the methods being employed were appropriate for building consensus in a manner that would improve resource management. Moreover, significant focus was placed upon communications and engagement with stakeholders and resource managers to keep them involved in the process and promote the transfer of this scientific knowledge to management. Scoping The South Florida project employed consensus building throughout the scoping process. Ralph and Poole (2003) define consensus building as “an interpersonal and political process designed to facilitate decisionmaking in the divisive and contentious political environment that surrounds the development of natural resource management policies.” The multidisciplinary nature of EBM underscored the importance of consensus building and illustrated t he value of including it as an internal component when integrating natural and human dimensions science (Fox et al., 2006; Slocombe, 1993b ; Weichselgartner and Kasperson, 2010). The scoping process included several components intended to synthesize current knowledge and systems viewpoints. The process included information exchanges and briefings, conceptual diagram development, and integrated conceptual ecological model (ICEM) development.

PAGE 63

63 Throughout the project, multiple streams of communication were used to encourage information sharing to improve the synthesis of knowledge (Figure 3 2). Exchanges occurred in person at workshops or through social media, E mail, online meetings (GoToMeeting, 2013), or interpersonal communications. The crux of the MARES sc oping process was four twoday facilitated workshops, one in each subregion and one for the total South Florida coastal marine ecosystem. All workshops were announced publicly via the project website, public listservs, and during public meetings related t o resource management in the region (e.g., U.S. Coral Reef Task Force). Moreover, we specifically targeted NGOs to represent the viewpoints of their constituency. Each workshop was attended by 40 to 60 participants from federal, state, and local agencies, academic researchers, NGOs, and an elected official. The facilitated workshops were used as a forum to present ideas and to build consensus regarding the structure and function of the coastal ecosystem, including humans. In most instances, consensus was achieved by general agreement. These workshops first aimed to develop a conceptual diagram that depicted the ecosystem, including relevant human activities. During regional workshops, participants were split into groups and asked to create a diagram of the ecosystem. Group discussion was used to create and refine these diagrams representing the biophysical and human dimensions of the ecosystem in a visual format for science communications with policy makers and nontechnical audiences (Dennison, 2007; Likens, 2010). Sometimes, one or two participants led the effort, and in other instance several individuals joined in the exercise. The group reviewed their drawing and when consensus on the content of the sketches was achieved, a photograph was taken, and the drawings were recreated in Illustrator CS6 software (Adobe, 2012) and shared with participants for review (Figure 3 -

PAGE 64

64 3). Diagrams were added to subsequent reports, presentations, and fact pages to share a visual representation of the complexities of the ecosystem with nontechnical audiences. The conceptual diagrams were then used as the basis for developing ICEMs that described cause and effect relationships in the ecosystem, and identified focal ecosystem components for both biophysical and human dimensions. A modified Drivers Pressures States Impacts Responses (DPSIR) framework (O ECD, 2003; UN, 1996) termed EBM DPSER was used to develop the ICEMs. The primary modification in the EBM DPSER framework was to emphasize the importance and explicit inclusion of ecosystem services by replacing the I mpacts module in DPSIR with E cosystem Services in EBM DPSER (Kelble et al., 2013) (Figure 3 4). Ecosystem services in the South Florida project were defined as the benefits that humans derive from the ecosystem (MEA, 2 005). They linked people to the state of the ecosystem using the terminology “Ecosystem attributes that people care about” (Kelble et al., 2013). The attributes were carefully selected to capture services that had value for both people who live in the Sout h Florida ecosystem and people who do not. The value of the ecosystem service was related to environmental conditions, and this value was measured and reported in a monetary, cultural, or social context (Loomis and Paterson, 2014). The state module of EBM DPSER ICEM was further described by developing state models for each of the focal ecosystem components in the state module. Prior to each regional workshop, a subset of project participants were asked to draft a conceptual ecological model (CEM) for state s of the ecosystem (Gentile et al., 2001). These individuals were regarded as experts in their field of science. Draft models were

PAGE 65

6 5 presented for further comment during the regional workshops. Final drafts of the diagrams were put into a standardized format for reporting. The CEMs were used to impart the knowledge about the ecosystem with the appropriate target audience with emphasis on the linkages among the components. CEMs provided a more technical view of the ecosystem than the conceptual diagrams and helped characterize and assess an ecosystem using a systematic approach to identify drivers, stressors, and desired endpoints with considerations of risk and uncertainty tied to alternative futures (Figure 3 5). Gentile et al. (2001) and Reiter (2004) illustrated the utility of these models as management tools that aid in identifying societal preferences and ecological states and the linkages among these. Work groups used the consensus built during the scoping process to create reports describing focal ecosystem components within the EBM DPSER model allowing one report to reflect the consensus knowledge of all workshop participants. Work groups presented a synthesis of scientific knowledge on a particular component of the ecosystem followed by an open discussion of how to move that science into the IEA framework. In some instances, workshop attendees ranked and prioritized features of the environment by placing adhesive dots on a poster listing each of those features, in other cases, they used audience response systems (Turning Technologies, 2013). These methods characterized areas of convergence and divergence within the expert knowledge base in the room. Divergent topics were discussed in a facilitated session to gain consensus from participants. Workshop r eports were shared for review and additional comment as a final opportunity to illustrate acceptance of the synthesis and findings.

PAGE 66

66 Indicators Indicators are quantitative measures that assess the status of the ecosystem (Doren et al., 2009). They are most useful when assessing the status relative to a predefined goal. Because MARES stated goal was an ecosystem that is sustainable and delivering societally desired levels of ecosystem services, selected indicators were required to be scientifically defensible and able to assess the ecosystem relative to this goal. To accomplish this objective a set of 11 criteria for ecosystem indicators were described based upon previously applied indicator criteria in the literature and input from workshop participants (Table 3 1). Indicators were developed based on these criteria for both the biophysical and human dimensions of the ecosystem. This led to a large number of indicators that reduced the ability to quickly communicate the status of the ecosystem. Thus, MARES ag gregated indicators into indices providing a hierarchy of information relevant to a broad array of audiences. The hierarchy of indicators enabled a range of users from those only interested in the broad status of the ecosystem and thus only the most aggregated of indices to those interested in a specific sector within the ecosystem to assess the current status of a specific ecosystem metric (Loomis et al., 2014). The focal ecosystem components identified in the EBM DPSER models became the full suite of highlevel composite indices required to assess the status of the ecosystem. However, the underlying indicators had to be developed to calculate these indices. Work groups identified at the scoping workshops developed these underlying indicators. They lead indicator development for each state component as well as all relevant human dimensions components of the ecosystem. These work groups were comprised of 215 people including topical experts and, where possible, relevant

PAGE 67

67 resource managers. Each work group was tasked with selecting indicators based on the MARES indicator criteria and writing a manuscript reporting on the indicator selection process and results (c.f. Lirman et al., 2014; Loomis and Paterson, 2014; Marshall et al. 2014; Ogden et al., 2014a, b; W ingard and Lorenz, 2014). Risk Analysis The third step in the IEA process was to conduct risk analysis on the ecosystem components and their relationships to one another. This was performed through a twoday scoping workshop at the end of the MARES process that sought to elicit expert opinion to rank the strength of connections between pressures, states, and ecosystem services identified in the EBM DPSER model for Florida Bay, the Florida Keys, and the Dry Tortugas. The workshop was held on August 2223, 2012 and 386 questions were posed to 25 regional scientists, managers, and an elected official using keypad polling systems (Cook et al., 2014; Turning Technologies, 2013). The responses were used to gather, analyze, and illustrate consensus on the strengt h of the model linkages between ecosystem pressures, states, and ecosystem services, and the relative risk of selected ecosystem states and services due to the cumulative impacts from pressures. For a detailed description of this methodology see Cook et al . (2014). Evaluating M anagement S cenarios The evaluation of alternatives has been found to be a key feature in the use of conceptual models to support resource management decision making (Tscherning et al., 2012). There are a number of methods that can be applied to evaluate management scenarios with conceptual models (c.f. Harwell et al., 2010; Reiter 2004; Elmer and Riegl, 2014); however, Bayesian Belief Networks (BBNs), or probabilistic models that employ adaptive management and can incorporate consensus based approaches, were

PAGE 68

68 selected as a method to test the utility of translating the MARES science to evaluate management scenarios (Nyberg, et al., 2006). BBNs were chosen due to their relative ease of use, ability to incorporate qualitative and quantitati ve data, and visualization features. In addition, BBNs are themselves focused on building consensus consistent with the overall MARES goals. Moreover, BBNs can develop a broader understanding of impacts and range of outcomes, sometimes with limited inform ation from stakeholder perceptions (Cain, 1999; Cain, 2001; Lynam 2003). Growing interest in the use of BBNs to develop consensus based management alternatives, or “reasonable assurance” in the decision making process provoked their use for assessing appli cability to evaluating management scenarios and increasing the use of science by resource managers within MARES (Uusitalo, 2007; Lynman, 2010; Nyberg et al., 2006, van Dam et al., 2013). Using responses from the IEA process, a proof of concept BBN was cons tructed using Netica (Norsys, 2013) software to represent the relationship of ICEM variables and the usefulness of the network in decision making. This exploratory network was constructed using the coral reef and hardbottom CEM from the Southeast Florida C oast (SEFC), with a subset of climate change impacts that were determined to be the most prevalent global driver in the region (Figure 3 6) (Nuttle and Fletcher, 2013c; Riegl et al., 2013). The prototype was designed to mimic the CEM model structure with input from MARES participants, coral researchers, and the literature. The prototype was shown to potential endusers (i.e. resource managers) in small group settings to obtain a first impression of the BBN as a decision support tool. The far field drivers and pressures selected were “pressures related to climate change and the rising concentration of carbon dioxide in the atmosphere, including the effects of ocean acidification and accelerated sealevel rise” (Nuttle and Fletcher, 2013c). While it was

PAGE 69

69 recognized that far field pressures were typically beyond the jurisdictional boundary of regional management entities, the climate change driver was selected for use in the BBN to investigate what pressures (e.g., sea level, temperature, acidification, etc.) may have the most impact on coral reefs and hardbottom habitats, and which management intervention (e.g., reef restoration, water management, etc.) might be used to address the impacts. Thus, BBN scenarios were built for use in promoting awareness of the cascade of impacts to reef ecosystems caused by climate change when communicating with the general public and decisionmakers. The BBN was completed using a stepwise process focused on stakeholder engagement (Cain, 2001). In this way, each activity from the South Florida IEA was incorporated into building the network. Scoping identified the network purpose and provided a comprehensive overview, the ICEM of the region that proved most useful for building the BBN structure, indicators were used to define the nodes in the BBN, and the risk assessment confirmed through expert opinion that the greatest strengths of relationships from ecosystem pressures to states impacting ecosystem services were freshwater delivery to the coastal ecosystems and climate change suggesting the use of climate change in the proof of concept model (Cook et al., 2014). The BBN presented climate change impacts to the SEFC in a visual format consisting of 18 nodes (boxes) and 25 lines connecting those nodes as detailed in the CEM coral and hardbottom diagram (Figure 3 7). The nodes were connected by arrows to illustrate parent child relationships. These links supported the BBN mathematical model calculating the probabilities of the condition of nodes (e.g., high or low diversity of corals). Working from the bottom of the diagram upwards, and following the EBM DPSER model, two nodes described the drivers: climate change and storms and low

PAGE 70

70 pressure systems. These nodes lacked parents, but collectively had 7 parent child relationships with nodes representing pressures in the coral and hardbottom CEM. Node values could be manipulated (e.g., frequency of storms being many or few per year) to measure the cascade of impacts to the network ultimately ending with “aesthetics” being the selected desirable ecosystem attribute for this exercise. Moving upwards in the model, there were seven pressures (sea level rise, depth distribution change, temperature, physical damage, acidification (measured by pH was combined with aragonite saturation state from the CEM), salinity, and turbidity and light attenuation were joined together). Sea level rise was a parent node using a 2010 reference level, and regional projections of 3 to 7 inches (2030) and 9 to 24 inches (2060) (Nuttle and Fletcher, 2013c; SFRCCC, 2011) . Sea level rise also fed into one other pressure node (depth distribution change). Depth distribution change was associated with rising sea levels and the ability of corals to grow upwards, retain their vitality at deeper depths, or to see a shift in coral species at deeper depths. Qualitative values (weak, moderate, strong) were assigned to this node because of the uncertainties tied to this change. Temperature parameters selected were determined using one of the protected species of coral, staghorn coral ( Acropora cervicornis ) having quantitative temperature thresholds (Shinn, 2012). Physical damage was generically categorized as high, moderate, or little based on acreage impacted. Ocean acidification was represented by combining aragonite saturation stat e and pH. The mineral aragonite is available in seawater for use by organisms such as corals to build their skeletons. The saturation state of aragonite affects the availability of the mineral impacting coral growth rates. Lower seawater pH levels result i n lower amounts of aragonite available for coral growth. The resulting node reflects qualitative values based on saturation state

PAGE 71

71 ranges derived from research: no mineral available for coral growth (no growth (0 to 2)), same growth rate as today (3 to 4), and faster growth rate (5+) (Langdon, 2012). Salinity values of 34, 35 to 39, or greater than 40 ppt were determined from Riegl et al. (2013). Turbidity and light were not consistent in their impacts to corals. Riegl et al. (2013) noted several studies in which coral growth rates both increased and decreased with changes in light penetration to the sea floor associated with turbidity. This node remained part of the BBN because of its inclusion in the CEM with values assigned as good or bad light levels. It served as another example of how to deal with uncertain variables in the network. States were addressed next following the structure of the CEM. Bleaching, the expulsion of symbiotic algae from coral tissue, was represented by the presence or absence of t his condition in coral colonies. Calcification rates ( high or low) represented the ability of corals to produce calcium skeletons. Growth, survival, reproduction and recruitment were grouped together in the CEM and remained so for the BBN to summarize the overall condition of corals. This node fed into four more states (diversity, resilience, and reef structure, and the grouping of species abundance, spatial extent and distribution). These nodes represent measureable attributes that can be used as indicator s of reef condition, and are therefore valuable to the BBN in providing qualitative and quantitative information that can be used and updated in scenarios. Responses, or management interventions identified were included (fishery regulations, water management, reduced coastal development, climate change coping strategies, reef restoration, invasive species management, and lifestyle response by individuals). These were identified by the CEM authors and vetted during document review. The values of the responses were manipulated during scenarios to identify if

PAGE 72

72 there were changes to coral reef and hardbottom state or desirable ecosystem attributes. Desirable ecosystem attributes was the topmost node. Aesthetics was chosen to encapsulate all of the other nodes and values in the network. Riegl et al. (2013) note the importance of aesthetics for recreational and tourism in the region. It was used in the prototype BBN to explore probability values one might place on the state of the ecosystem and effects of managem ent responses. A table (Table 3 3) was created to assign conditional probabilities for every node and combination of nodes. This task was performed using knowledge from the development of the CEM reports and input from multiple stakeholder meetings. The pr obabilities associated with each node in the toy model were assigned as estimates and were not vetted for accuracy. Probabilities reflected a general outcome where ‘high’ or ‘true’ conditions of coral growth, survival, reproduction, and recruitment would r esult in higher aesthetic values, and ‘low’ or ‘false’ conditions resulted in lower aesthetic values. The outputs were intended to illustrate the capabilities and limitations of the BBN and scenario evaluation outputs but require further input for use in m anagement decision making. Once the table was completed, a series of sensitivity analyses were run using the Netica software (Norsys, 2013). This analysis illustrated the influence of other nodes within the network on the selected node. Four analyses were performed, two for aesthetics and two for growth, survival, reproduction and recruitment. Scenarios examined changes resulting from climate change ( yes or no) in combination with storms and low pressures systems ( many or few ) using the reef restoration management response.

PAGE 73

73 Communications As MARES central goal included the improved use of science by resource management, communications that would aid in the transfer of science to management were an integral component of the program. Communications were developed using stakeholder input from both technical experts and resource managers. Products consisted of printed materials, Internet sites, and interpersonal interactions. Project participants and the resource management community in South Florida confirmed these types of products were appropriate mechanisms to convey information to managers during a scoping exercise. As a result, this communications strategy was pursued to transfer science to managers. Social media was added to the Internet based products and stronger emphasis placed on developing fact pages to highlight regional and topical components of the ecosystem. Printed materials Printed materials were created for use in a variety of settings with the overarching goal to inform the target audience about the IEA effort and its findings. Reports, fact pages, diagrams, and posters were created to meet target audience needs. Fact pages in particular were developed on a regional and topical basis for decisionmaking and resource managers. First, the projec t planning team identified a variety of manager types. Then, the team outlined several kinds of printed materials each manager type would be most likely to use. The materials were created by committee and incorporated data, images, and summary text. The pr inted materials were then refined through engagement with the targeted management type. Printed materials were identified at the beginning and throughout the project to meet the needs identified by the target audience and as

PAGE 74

74 opportunities to share informat ion. Materials were distributed inperson by project partners at meetings throughout South Florida. Internet based products Internet based products were developed to share project updates and information. A webpage (www.soflamares.org), document managem ent system, Facebook, and blog were produced and electronic correspondence using listservs was maintained throughout the project period. The use of Internet based products was undertaken to encourage interactions with managers and stakeholders and also to reach a broader audience. The website created an Internet presence and established a consistent location for access to meeting announcements, reports, and project contacts. A document management system was designed as a repository for all project documents and reference materials and to share ideas among principal investigators and partners. Two listservs, or email discussion groups were created, one for principal investigators, the other was a moderated list for the public. As the project evolved, a Facebook page, and a blog site were built to improve the awareness of the project and to further discussions among the researchers, managers, and interested public. Google Analytics were used to determine the frequency of visits to the website, Facebook page, and blog. Interpersonal communications Formal and informal interactions during workshops, focus groups, meetings, and lunchtime sessions helped established new professional relationships. Many of the principal investigators, especially those from the Florida Keys and Dry Tortugas subregion, worked together since the start of an extensive regional restoration project in the late 1990s. Interpersonal communications occurred among this network of

PAGE 75

75 individuals during the multitude of restorationrelated meetings and continued with the introduction of MARES in late 2009. In addition, conferences and regional meetings afforded the opportunity for colleagues to interact and get to know one another. Results Scoping MARES conceived and assembled the first comprehensiv e, sciencebased consensus for the marine and coastal ecosystem of South Florida through the development of conceptual diagrams, models, and indicators. Scoping served as a platform to conduct this synthesis by building consensus on ecosystem structure and function, diagnosing participant needs (some of whom were managers and NGOs), and creating a shared responsibility for learning among project leaders and participants (Etling, 1993; Kleis et al., 1973). In the case of South Florida, workshops elicited par ticipant opinion and generated a scientific consensus on the fundamental regulating processes of South Florida’s marine and coastal ecosystem (Nuttle and Fletcher 2013a, b, c). Group learning opportunities materialized beyond initial expectations and inclu ded exploration for integrating ecosystem components, improved comprehension of specific elements of the ecosystem from specialists, and most prominently with the introduction of human dimensions science into the project (Kelble et al., 2013). Hungerford and Volk (1990) illustrated the process for “environmental citizenship behavior” and how it evolved over time beginning with individual awareness, and then leading to ownership and empowerment (NOAA/CSC, 2007). In this way, each participant, as a scientist, manager, or interested party, engaged in shared learning and the open dialogue that contributed to achieving the project goal.

PAGE 76

76 Indicators Indicators were developed to determine and convey whether an ecosystem is approaching sustainability and delivering desired levels of ecosystem services that capture the status of both the natural and societal aspects of the system. Consistent terminology for both natural system and human dimensions indicators were defined using an iterative process during workshops and work group meetings, and vetted with the larger participant audience during document development. A hierarchical approach was used to identify a range of indicators. The hierarchical approach to developing consistent indicators is described in detail in L oomis et al. (2014). At the lowest level were data or measurements of one or more variables. The highest level was comprised of an aggregated composite index used to define overall ecosystem status. The highest level indices were developed using ICEMs and were generally either broad natural science system components (e.g., water column, coral and hardbottom) or ecosystem services (e.g., recreation quality) that were themselves composed of a number of lower level indicators. Indicators were developed or alr eady existed for economic human dimensions (Johns et al., 2014), noneconomic human dimensions (Loomis and Patterson, 2014), coastal wetlands (Wingard and Lorenz, 2014), corals (Lirman et al., 2014), marine birds (Ogden et al., 2014a,b), fish (Ault et al., 2014), oysters (Volety et al., 2014), seagrass (Madden et al., 2009), and the Water Column (Kelble et al., In Prep). Although beyond the scope of the MARES project, further efforts can incorporate these indicators into ecosystem report cards that document trajectories towards (or away from) a sustainable condition that delivers the desired levels of ecosystem services.

PAGE 77

77 Risk Analysis Risk analysis using the conceptual models developed during MARES and participant input are described in detail in Cook et al. (2014). Participant opinion gathered during a workshop resulted in fish and shellfish, protected species, and marine birds ranking as the most at risk of the ecosystem states. The highest ranked pressures that contributed to the largest degradation of ecosystem services were freshwater delivery, temperature effects of climate change, and impacts of climate change on weather. The most at risk ecosystem services were existence of a natural system, pristine wilderness experience, and nonextractive recreati on. The risk analysis workshop results were the first quantitative assessment of the ICEM components and further analysis to their utility in management are being explored. Bayesian Belief Network The prototype BBN captured EBM DPSER components using quant itative and qualitative information obtained during the IEA process to examine causal probabilities and sensitivity analysis among selected components. Overall, more intense climate impacts resulted in lower quality coral reef and hardbottom aesthetics. For periods with more intense climate impacts, the probabilities of “aesthetics” being high quality was 22.5%, and low quality 77.5%, and for periods with lower climate change impacts, probabilities for high quality were 50.4%, and low quality were 49.6%. Sensitivity analysis showed that the nodes with the most influence on the “aesthetics” outcome were parent nodes: diversity, resilience, reef structure and the grouping of species abundance, spatial extent and distribution (Table 3 5). For high climate change impacts the probabilities for the state indicator “growth, survival, reproduction, recruitment” being were good 1.26%, moderate 39.7%, and low 59% showing low coral health and growth

PAGE 78

78 in these conditions. For periods with less climate change impacts, the probabilities for this indicator changed to good 12.1%, moderate 71.7%, and low 16.2% validating reef conditions would be better during times with less climate impacts. Sensitivity analysis for both high and low climate impacts showed that child nodes within the state element of the CEM had the most influence on this indicator: reef structure, species abundance, extent and distribution, aesthetics, diversity, resilience and bleaching (Table3 6). In general, the network validated MARES discussions about cli mate influencing ecosystem attributes that people care about, the limitations of management responses to the global climate change driver, and the network provided numeric values to evaluate indicators. Ten individuals responsible for resource management or supporting management decisions viewed the tool and most expressed interested in continuing to build and refine the network, because they felt it could help inform their decisions. This is evidence that the BBN can be used for further stakeholder discuss ions and evaluating scenarios to transfer scientific knowledge into management. Communications Thirty five science communications were produced to document and synthesize information for managers and stakeholders (Table 3 4), and numerous information exch anges took place in the form of workshops, meetings, writing sessions, electronic exchanges, and conversations over the four year project period. Six conceptual diagrams, 3 ICEMS, 19 state CEMS, greater than 50 indicators, 20 indices, and one risk analysis were completed with input from 124 participants, many of those repeatedly engaged in workshops and writing activities.

PAGE 79

79 The IEA process resulted in the formation of an informal social network of scientists, resource managers , and interested parties located throughout South Florida that is used to exchange information and promote the use of science to inform decision making. While no formal evaluation was performed to document the utility of the different communication strategies, observations by staff made during workshop participation, information requests, and project partner feedback reflected positively on the outputs. The website had 2582 visits from 17 November 2011 to 17April 2013. The Facebook page had 658 ‘likes’ from 4 March 2012 to 20 April 2013. The greatest use of the Facebook page occurred prior to workshops and when preliminary findings were posted. The blog also experienced highest usage prior to workshops, meetings and project team updates and had 1939 visits from 29 March 2012 to 18 April 2013 (Table 3 4). Discussion MARES was a resource intensive four year project that made significant progress toward implementing an IEA in South Florida. However, this IEA still has not assessed the current status of the ecosystem to determine if it is sus tainable and delivering desired levels of ecosystem services and has yet to fully evaluate management scenarios. Thus, it has yet to complete a single iteration of the IEA framework and as such is still in its infancy. The scoping, indicator development, and risk analysis represent a consensus developed from participant opinion, resource managers, and NGOs accomplishing one of MARES primary objectives. While this effort was comprehensive and inclusive, it forms only the beginning of an adaptive process necessary to provide the scientific support for EBM of South Florida’s coastal and marine system. Despite being in its infancy, the south Florida IEA process has

PAGE 80

80 already begun to increase the use of science by resource managers. It has increased communication between scientists, NGOs, and resource managers as exemplified by the development of consensus conceptual diagrams, models, and indicators. Moreover, the risk analysis and BBN were both developed in close partnership with the resource management agencies, which had input on the design, hypotheses, and implementation from their conception to their completion in hopes that this would increase their utility to relevant resource management decision makers. At this stage, monitoring and integrative tools emerge to inform, challenge and adapt the current system understanding into adaptive and responsive management options. Following with our example BBN, scenarios were used for illustrative purposes to show how the network can provide useful opportunities for further dialogue and to evaluate potential future scenarios including pending management decisions. The BBN demonstrated that even with simplistic yet plausible initial data, the results of the network model are consistent with our scientific understanding and help spotlight relationships and potential challenges in adaptive management. The proof of concept BBN proved useful in examining the ICEM components in an iterative format and assessing management interventions in an adaptive capacity. The network lac ked any new information or insights compared to the ICEM from which it was designed and should be taken into considered when reviewing the results. Further exploration of the BBN as a tool for transferring science to management is warranted to identify the value of BBNs as a tool for transferring science to resource managers. It is unclear if other scenario evaluation tools could further the IEA process since this was the only method used during the MARES project. However, the network shows promise as a met hod to assimilate and address uncertainties in environmental decision making,

PAGE 81

81 screen management or policy options, and identify knowledge gaps as described by Cain (2001). The South Florida IEA network developed by MARES was essential to successfully testi ng 1) if a BBN could be designed from the ICEMs, and 2) the utility of the model for endusers. This preliminary BBN met the categories of evaluation criteria developed by Lynam et al. (2007) which state “the tool must 1) support communication and learning between the insiders and outsiders who are using the tools; 2) be adaptable for implementation in various decisionmaking contexts for use by diverse users, including those at the local level; and 3) produce data and information that are useful and valid as a basis for decisionmaking or can be used for further analyses.” The results of the experimental BBN showed that little could be done to improve reef aesthetics as a result of intense climate impacts. The sensitivity analyses performed helped identify factors that may be most useful to managers in deciding what action to take to reduce or enhance the aesthetics of coral reefs and hardbottom in southeast Florida. In addition, the network can be used to test indicators to determine which indicators are more sensitive in assessing the aesthetics of the ecosystem. These concepts can be further explored by manipulating the structure of the prototype BBN, for example, management interventions might link directly to multiple drivers, pressures, and states, and expanding the BBN to include the human population elements of the coral and hardbottom CEM. MARES provided several new insights for organizing and implementing long term ecosystem management activities and IEAs. The participatory consultative approach was successful in catalyzing new ideas for research, monitoring, and management actions that emphasize human dimensions science, including well being (Loomis and Paterson, 2014). Input from knowledge actors (e.g., researchers, planners, and

PAGE 82

82 managers) was instr umental in all aspects of the project and their commitment to participate in the multi year effort critical to the success of the IEA. The process helped develop new products, such as diagrams of the South Florida ecosystem that assist in conveying information about the form and function of the environment. These concepts can be used to develop new ways of managing that truly incorporate stronger integration of biophysical and human dimensions science. Lastly, the development of a coordinated network of res ource managers, scientists, and NGOs that have developed a consensus on ecosystem structure and function cannot be overstated. This network has already shown an increased ability to transfer scientific knowledge to management as demonstrated by the BBN des cribed herein, and other related efforts described by Elmer and Riegl (2014) and Cook et al. (2014). The development of additional communications in MARES to help transfer this knowledge to management improved the utility of IEA products and has resulted in the use of IEA products in ongoing management activities, such as a scoping exercise for southeast Florida and science communications in the Florida Keys and southwest Florida. Additional insights from the project included the following: Plan for evaluation . Print, Internet, and information exchanges in general appear to be valuable to the multitude of managers involved in the project, but reporting for these was not consistent. Understanding how effective these products are is difficult without conducti ng a formal needs assessment that includes an evaluation method at the beginning of the project (NOAA/CSC, 2003 and 2009; Witkin and Altschuld, 1995). This process aids in explicitly identifying a target audience and their needs, and a process to evaluate measurable outcomes of the effort.

PAGE 83

83 Build upon existing networks. Informal communications are difficult to evaluate, but often are critical in building a network of partners who work together to achieve a common goal. The South Florida IEA project is no exception to this concept. In 1993, a regional water monitoring and management initiative began and in part, evolved into a larger ecosystem restoration effort that included frequent interactions among researchers and managers attending workshops and meetings. The network of individuals formed over the past 20 years appears to foster a level of respect or trust for colleagues with a common interest in South Florida’s ecosystem that benefitted the MARES project. Plan for Adaptation. Wicked environmental problems are complex, and often require complex solutions (Churchman, 1967; Rittle and Webber, 1973; Xiang, 2013). The solutions are often adaptive as new information is added into the decisionmaking process. Fluidity in project planning must be recognized and accounted for in the adaptive process. Pictures tell stories. Conceptual diagrams proved useful in synthesizing science into a visual format for multiple purposes. The process of developing the conceptual diagrams resulted in shared vision with participants coalescing around one easy to interpret representation of a complex ecosystem. Use of diagrams in presentations, print materials, and the number of requests to obtain images exceeded that of any other products from the project. Consensus building helps explore new ideas and concepts in a safe place. Participants with varied backgrounds observed and discussed the causeandeffect relationships found in the ecosystem during several steps in the IEA process. The result was the development of a single document synthesizing the environment using the

PAGE 84

84 Driver, Pressure, State, Ecosystem Services, and Response for the ecosystem, and 19 state ecological submodels (Nuttle and Fletcher, 2013a, b, c). Workshop discussions of human dimensions science, especially the noneconomic values, brought to the forefront a realization that a broader view of this field of science was necessary in the project, and a work group was created to further this element of the project. Integrative tools are required to move forward. Gol dston (2009) states that science is one factor in the decisionmaking process, noting that typically, the problem is not a lack of information, rather difficulty in separating the policy questions from the science questions, and the interplay of values and risk in decision making. BBNs were helpful in integrating ICEM components. The ability to build a network with qualitative and quantitative information is useful to begin adaptive management discussions and to explore optimal management responses. The BBN is a promising decision support tool that will be expanded for further analysis and use in the IEA process. Implications The IEA process used in South Florida was helpful in framing the interconnected multidisciplinary components of the system. The use of tools, such as the conceptual diagrams in building consensus and describing the ecosystem to nontechnical audiences and BBN management alternatives, can be helpful in engaging the management community and stakeholders in the EBM discussion. The IEA process is useful and all three parts: people, process and tools, needs consideration, this effort supported the importance of consensus building process to truly implement the EBM science into decisionmaking (Kiker et al., 2005).

PAGE 85

85 The IEA process was success ful in building consensus focused on a comprehensive synthesis of the regional marine and coastal ecosystem as well as in contributing to the transfer of science to a variety of target audiences including managers as evidenced by the vast array of products produced and the number of visitations to project websites. While there was no single decisionmaking outcome that was directly attributable to the project, several benefits were identified by participants: a foundation to incorporate marine and coastal elements into regional restoration and a larger ecosystem assessment; the importance of a consensus based approach to integrate human dimensions science in what many consider ecologically focused projects; and, it forged relationships to include complex, multivariate results into the decision making process among multi agency, nonprofit, and academic institutions in the region. In addition, an informal EBM IEA network was created and was valuable to obtain feedback on the needs and uses of ecosystem information by managers. The EBM IEA foundation and MARES network continues as part of the Gulf of Mexico IEA and management planning.

PAGE 86

86 Figure 31. The South Florida IEA is comprised of the Marine and Estuarine Goal Setting for South Florida (MARES) project domain. It encompasses the marine and coastal components of the southern tip of the Florida peninsula.

PAGE 87

87 Figure 32. South Florida IEA process and products. Project participants, staff and representatives from regional academic, government, and nongovernmental offices, engaged in scoping, indicator development, and risk assessment activities. Throughout the process, multiple streams of communications were developed to transfer information to managers and stakeholders.

PAGE 88

88 Figure 33 During a scoping ex ercise, workshop participants sketched their view of the ecosystem, reviewed the drawings, and then staff recreated the sketch in Adobe Illustrator software for inclusion in reports and project communications.

PAGE 89

89 Figure 34. ICEMs are based on the DPSER f ramework which describes the ecosystem in terms Drivers, Pressures, State, Ecosystem Services, and Response (Kelble et al., 2013).

PAGE 90

90 Figure 35. Water Quality Integrated Conceptual Ecological Submodel developed to capture the details of the drivers , pressures, ecosystem services, and responses in the ecosystem (Nuttle and Fletcher, 2013a, b, c). The figure illustrates the interconnectedness of the ecosystem and humans as a part of it. Expert opinion was used to develop a diagram representing the ecosystem form and function during workshops.

PAGE 91

91 Figure 36. Climate change impacts of the coral reef and hardbottom habitats of the Southeast Florida Coast Conceptual Ecological Model, highlighted here in yellow, were used to create the Bayesian Belief Net work (see Figure 3 7).

PAGE 92

92 Figure 37. Bayesian Belief Networks are used to incorporate stakeholder knowledge, values, and preferences in the decision making process. An example was developed to test the utility for applying this method to complement the S outh Florida IEA process. Climate change impacts to coral reefs and hardbottom habitats were drawn from the Conceptual Ecological Model for the Southeast Florida Coast subregion (see Figure 3 6). Drivers, pressures, states, attributes that people care about and management responses were incorporated into the network to illustrate the potential to create and use these networks to test climate impacts and management interventions.

PAGE 93

93 Table 31. Criteria to guide development of indicators from attributes that we can measure. Primary Criteria 1. Does the indicator provide an integrative measure of the overall status of the ecosystem or of essential ecosystem structures, functions or processes? (Doren et al. 2009, Dale & Beyeler 2001, Luckey 2002) 2. Does t he indicator relate to ecosystem service(s)? (modified from Feld et al. 2009) 3. Is the indicator relevant to management goal(s)? (Bradley et al. 2010) 4. Is the indicator sensitive to system Drivers and Pressures? (Doren et al. 2009, Dale & Beyeler 20 01, ICES 2002) Data/Analysis Criteria 5. Is the indicator based upon data that can be generated with accuracy and precision relatively easily and for which there is sufficient existing data to evaluate change going forward? (Doren et al. 2009, ICES 2002, Dale & Beyeler 2001, Rice & Rochet 2005) 6. Is it possible to predict how the indicator will respond to changes in the ecosystem (including societal changes) over management -relevant time scales? (Feld et al. 2009, Dale & Beyeler 2001) 7. Does the indicator have a response that is easily detectable above the background variability to make it useful in measuring response to management actions or a change in a Pressure that may or may not be a result of management action(s)? (This also means the response signal should be attributable to a change in management or pressure.) (ICES 2002, Bradley et al. 2010) Communication 8. Is the indicator understood by managers and the public? (Rice & Rochet 2005) 9. Does the indicator respond to stress earlie r than the rest of the system (i.e. is it a leading indicator?)? (Dale & Beyeler 2001) 10. How long will it take for the indicator to show a response to possible management actions? (Dale & Beyeler 2001) 11. Has the indicator been employed effectively either in south Florida or elsewhere? (NRC 2000)

PAGE 94

94 Table 32. Prototype BBN indicators, units, and states for the southeast Florida coral reef and hardbottom conceptual ecological model. Node Indicator Unit State Reference Storms_LowPressureSystems Num ber of storm events per year based on average from 1996 2009 Number of storms 13+ 12 <=11 NOAA, 2013; Neuman et al. 1999; Nuttle and Fletcher, 2012c ClimateChange Presence or absence of climate change impacts qualitative High Low Sea level rise Sea lev el based on 2010 regional assessment and projections for 2030 and 2060 inches 0 to 2 3 to 7 9 to 24 SFRCCC, 2007 DepthDistributionChange Incremental change based on current sea level and presence/absenc e of stony corals qualitative Weak incremental change Moderate incremental change Strong incremental change Riegl et al., 2012 pH_AragoniteSaturationState Saturation state of aragonite (carbonate mineral) in seawater tied to rates of calcification based on level of pH of oceans caused by uptake of anthropog enic CO2 from atmosphere Qualitative (state) No growth (No available 0 to 2) Same rate as today (3 to 4) Faster growth rate (aragonite available (5+)) Kleypas et al., 1999; Langdon, 2012 Temperature Survival temperature thresholds for staghorn coral (Acro pora cervicornis) F 0 to 56 57 to 98 99+ Shinn, 2012; Mayor, 1914

PAGE 95

95 Table 32. c ontinued Node Indicator Unit State Reference PhysicalDamage Based on the extent of damage caused by natural (storms) or human (boats) activities qualitative High impa ct reef structure Moderatemix of Little and High Little superficial Ault et al., 1997; Chiappone et al., 2005; Lirman et al., 2010; Riegl et al., 2013 Light_Attenuation_Turbidity Based on clear, turbid free water allowing light to penetrate to depth for coral growth qualitative Good Bad Dodge and Aller, 1974; Dodge et al., 1974; Riegl et al., 2013 Salinity Average seawater content (36ppt) and stony coral thresholds ppt 0 to 20 25 to 40 40+ Edmondson, 1928; Jokiel and Coles, 1977; Rigel, et al. 2013 Ble aching Presence or absence of bleaching, expulsion of symbiotic dinoflagellates caused by warm or cool temperatures Boolean True False Van Oppen and Lough, 2008; Riegl et al., 2013 Calcification Current acidity rates of seawater and the ability of coral t o grow qualitative Low High Kleypas et al., 1999; Andersson et al., 2005; Cohen and Holcomb, 2009 Riegl et al., 2013 Growth, survival, reproduction, recruitment Current baseline in Atlantic/Caribbea n reef systems ~3 6% qualitative High Moderate Low Riegl et al., 2013

PAGE 96

96 Table 32. c ontinued Node Indicator Unit State Reference Management interventions Current options identified for responding to changes in coral reef and hardbottom habitats Fishery Regulations Water Management Reduced Coastal Develo pment Climate Change Coping Strategy Reef Restoration Invasive Species Management Lifestyle Response by Individuals Riegl et al., 2013 SpeciesAbunExtentDist Estimated percent cover of stony coral over southeast Florida substratum qualitative High Low Gard ner et al., 2005 ; Riegl et al., 2013 Diversity Number of species/taxa and genetic level qualitative High Low Connell, 1978; Riegl et al., 2013 Resilience Ability of system to absorb, resist or recover from disturbances or to adapt to change while continuing to main essential functions and processes qualitative High Low Riegl et al., 2013

PAGE 97

97 Table 32 . C ontinued Node Indicator Unit State Reference ReefStructure Presence of three dimensional geomorphic features Boolean True False Riegl et al., 2013 A esthetics Diverse, productive, healthy coral reef habitats Qualitative High Low Riegl et al., 2013

PAGE 98

98 Table 33. Conditional probability table (CPT) for the output node “Aesthetics” in relation to parent nodes “SpeciesAbunExtentDist,” “Diversity,” Resilience,” and “ReefStructure.” SpeciesAbunExtentDist Diversity Resilience ReefStructure High Low High High High True 100 0 High High High False 90 10 High High Low True 90 10 High High Low False 80 20 High Low High True 80 20 High Low High False 70 30 H igh Low Low True 70 30 High Low Low False 60 40 Low High High True 40 60 Low High High False 20 80 Low High Low True 10 90 Low High High Low False 5 95 Low Low High True 5 95 Low Low High False 5 95 Low Low Low True 5 95 Low Low Low False 0 100

PAGE 99

99 Table 34. Science communications products developed to transfer science to management and to promote stakeholder engagement. Science communication product Description Usage 1 Project brochure Two page overview of the project goals, partners, and process used to complete the effort. Provide project overview with background information; Reference to access Web page Observations reflect use during workshops and meetings in the region. 3 Posters Themed posters illustrating the overall project, process, an d science Observations reflect the posters gained the attention of conference participants and interested public; limited audience, but effective in establishing a presence at conferences and building awareness that led to continued interactions with proje cts outside of the geographic scope of south Florida; used during ecosystem restoration and human dimensions conferences helpful in the development of the socio ecological element of the project 3 Fact pages Two page overview with colorful imagery and limited text describing individual components of the ecosystem designed for use by specific segments of the management community and geographic scope. Observations reflect fact pages were shared (most often in electronic format) among the project partners wi th limited use at the time of this report. The fact pages in two geographic subregions could be used more frequently in the future as part of stakeholder engagement and a management plan review. 10 White papers Topical summaries and findings written in a standardized format Observations reflect white papers were used during workgroup sessions and as reference materials in preparation for workshops and meetings. 3 Technical memorandums ICEM workshop reports Observations reflect the memorandums serve a s reference materials in the decisionmaking process and science synthesis. 15 Journal articles Peer reviewed publications authored by project investigator Under development at the time of this publication. Intended to be used to disseminate MARES findi ngs to a broad audience and science managers Document Management System (DMS) Web based clearinghouse for all project documents. Computer software to manage documents tied to project Used by project team to share and archive documents.

PAGE 100

100 Table 34. C o ntinued Science communication product Description Usage Website Web page ( www.sofla mares.org ) released on Thursday, November 17, 2011 and revised on May 1, 2012. Serves as a Web presence and resource to post meeting announcements and documentation for the worldwide audience. The majority of web users downloaded documents (Google Analytics, 2013). From the release date until April 17, 2013, the total visits to the page was 2,582. The majority of visits were from M iami, Florida USA (15%) and Ottawa, Canada (5%), Followed by Gainesville, Florida (4%), Silver Spring, Maryland (4%), and Tallahassee, Florida (3%). Those users looked on average at three or four pages on the website with 77% of those visiting the home page, followed by 4% visiting the document downloads page and organization page. Listserv E mail service for group announcements and exchanges: principle investigators ( mares_pi list@lists.rsmas.miam i.ed u ) and interested public ( mares list@lists.rsmas.miami.ed u ) Served as a one way mechanism to deliver meeting announcements with very little ancillary information sharing. The MARES PI list contai ned 61 participants, and the MARES project list 186. All participants were added by request or due to their affiliation as a PI on the project. From August 5, 2009 there were 43 Emails on the MARES PI list, and 25 on the general list starting on November 1 2, 2009. The majority of Emails were sent by project leaders, with only three messages being sent by list members during the project period. Facebook Social media site released on March 4, 2012 ( http://www.facebook.com /South.Florida.MARES) Used as information exchange and for presence in social media. Initial usage reflects ‘likes’ from three Florida cities Miami, Fort Lauderdale, and Key Largo (JuneJuly 2012), and then briefly changed to Miam i, Florida, and two cities in Mexico, Ciudad Victoria, Tamaulipas, and Mexico City, Distrito Federal (July August 2012). From August 2012 to present, the greatest number of visitor home locations were Miami, Florida (10%), Milwaukee, Wisconsin (9%), and Me xico City, Distrito Federal, Mexico (6%). On April 20, 2013, the page had 658 ‘likes’ and the page was most active between July 2012 and September 2012. During this time, MARES announced the August 22 23 manager’s workshop, posted preliminary results of that meeting, and also placed an advertisement on Facebook to encourage use of the site (Facebook Analytics, 2013).

PAGE 101

101 Table 34. C ontinued Science communication product Description Usage Blog Social media site released on March 29, 2012 and is revised fr equently with updated reports, diagrams, and summaries. ( http://www.maresblog.or g/ ) Created to foster further online communications and interactions among interested groups and individual viewers. The majority of users visited pages with project documents (Google Analytics, 2013). From the release date until April 18, 2013, the total visits to the page was 1,939. The majority of visits were from Ottawa, Canada (8%), Gainesville, Florida, USA (8%) and Miami, Florida USA (7%). Those users looked on average at one to two pages on the blog with 52% of those visiting the blog entering via the main page, followed by 8% visiting the Florida Keys/Dry Tortugas Integrated Ecological Model page, 7% visiting workshop read ahead materials (Everglades National Park meeting), and 6% viewing the introduction to the MARES project page. The highest use days (page views) occurred prior to MARES workshops and meetings, and project team update announcements to ‘check the blog.’

PAGE 102

102 Table 3 5. Proof of concept BBN sensitivity analysis results “Aesthestics” node based on scenarios using high and low “ClimateChange,” many and few “Storms_LowPressureSystem,” and reef restoration “Intervention_Mgnt_Response” based on Figure 3 7. The nodes are l isted as a percent value in order of greatest to least influence on “aesthetics” with high climate impacts (left column) and low climate impacts (right column). Node % Node % SpeciesAbundExtentDist 61.1 GrowthSurvivalRepro 49.1 GrowthSurvivalRepro 48.5 D iversity 28.6 Diversity 22.6 ReefStructure 12.9 ReefStructure 21.9 Bleaching 9.8 Resilience 19.1 Resilience 9.36 Bleaching 5.35 Temperature_Staghorn 3.66 Temperature_Staghorn 3.13 DepthDistributionChange 2.78 Salinity 1.14 Light_Attenuation_Turbidity 0.567 DepthDistributionChange 0.77 SeaLevelRise 0.156 Light_Attenuation_Turbidity 0.1 PhysicalDamage 0.135 SeaLevelRise 0.08 Calcification 0.0977 PhysicalDamage 0.007 pH_AragoniteSaturation 0.000137 Calcification 0.003 SeaLevelRise 0 Interventions_M gnt 0 ClimateChange 0 Storms_LowPressureSys 0 Storms_LowPressureSystems 0 pH_AragoniteSaturation 0 Temperature_Staghorn 0 ClimateChange 0 Interventions_Mgnt 0

PAGE 103

103 Table 36. Proof of concept BBN sensitivity analysis results “GrowthSurvivalReproRecruitm ent” node based on scenarios using high and low “ClimateChange,” many and few “Storms_LowPressureSystem,” and reef restoration “Intervention_Mgnt_Response” based on Figure 3 7. The nodes are listed as a percent value in order of greatest to least influence on “GrowthSurvivalReproRecruitment” with high climate impacts (left column) and low climate impacts (right column). Node % Node % ReefStructure 40.4 ReefStructure 26 SpeciesAbunExtentDist 40.4 SpeciesAbunExtentDist 26 Aesthetics 35.2 Aesthetics 25.1 D iversity 32.8 Diversity 24.8 Resilience 32.8 Resilience 24.8 Bleaching 9.4 Bleaching 9.32 Temperature_Staghorn 5.45 Salinity 7.2 Salinity 2.01 DepthDistributionChange 1.7 DepthDistributionChange 1.55 Light_Attenuation_Turbidity 0.5 Light_Attenuation_ Turbidity 0.17 PhysicalDamage 0.48 SeaLevelRise 0.13 Calcification 0.3 PhysicalDamage 0.01 pH_AragoniteSaturation 0.0004 Calcification 0.01 Temperature_Staghorn 0 Interventions_Mgnt 0.007 SeaLevelRise 0 Storms_LowPressureSys 0 Interventions_Mgnt 0 pH _AragoniteSaturation 0 Storms_LowPressureSys 0 ClimateChange 0 ClimateChange 0

PAGE 104

104 CHAPTER 4 CONSTRUCTING DECISION SUPPORT TOOLS TO IMPROVE CORAL REEF ECOSYSTEM MANAGEMENT Introduction Coral reefs worldwide are in decline and impacts to these ecosystems associated with climate change have been well documented (Baker et al., 2008; Carpenter et al., 2008; Collier et al. 2008; Cook et al., 2014 ; Donahue et al. , 2008; HoeghGuldberg et al., 2007; Keller, et al., 2009; Lirman et al., 2014; Marshall and Schutt enberg, 2006; Wilkinson, 2000 ). In the Florida Keys, coastal and coral reef ecosystems have seen declining water quality indicators since the establishment of the Florida Keys National Marine Sanctuary (FKNMS) in 1990 (NOAA, 1996). Subsequently, the Water Quality Protection Program in the Keys was created in 1994 along with an action plan in 1996 (EPA, 1996; EPA, 1999). In 2002, the FKNMS established a conceptual model and science plan to identify major information gaps and to formulate adequate management responses to external stresses of their coral reef ecosystems (FKNMS, 2002). Additionally, Keller and Causey (2005) highlighted the importance and connectivity of the human and biophysical systems in managing the FKNMS alongside landbased activities in t he adjacent watershed. In 2008, Donahue et al. wrote the state of the coral reef ecosystem of the Florida Keys and in 2011 the Office of National Marine Sanctuaries produced condition reports calling for an ecosystem approach to manage marine and coastal r esources (Donahue et al., 2008; ONMS 2011). EBM is one strategy to integrate and balance the varied biophysical, institutional, and human dimensions of an ecosystem in a holistic manner in order to sustain or improve resources ( Christensen et al., 996; McC leod et al., 2005; Slocombe, 1993b; ___________ R eprint ed with permission from Fletcher, PJ, Li, Y, Kiker, G. submitted. Constructing decision support tools to improve coral reef ecosystem management. Ecology and Society.

PAGE 105

105 Szaro et al., 1998). I EAs are one method to achieve EBM. In 2009, a South Florida IEA began to formalize through the Marine and Estuarine Goal Setting for South Florida project (Fletcher et al., 2014; Nuttle and Fletcher 2013a, b, c; Ortner et al., 2014). Over 124 stakeholders contributed to a four year effort to identify, integrate, and synthesize the complexities of the South Florida marine and coastal environment resulting in the first comprehensive assessment of the total South Florida coastal marine ecosystem that incorporates socioec ological science with emphasis on ecosystem services that benefit the ecosystem and the humans living in the region. Project summary reports emphasize the significance of the socioecological ecosystem to the local and regional economies, notably, the $2.2 billion tied to coastal recreation in Monroe County (Florida Keys) and the FKNMS, as well as $8.5 billion asset value from reef visitation in southeast Florida ( Johns et al., 2001 ; Leeworthy and Ehler, 2010a, 2010b). However, even with these significant economic benefits, few interdisciplinary decision support tools exist to aid decision makers and stakeholders in planning for and responding to the pressures affecting these ecosystems. Decision support has expanded to include both computational and social constructs to aid interested managers and stakeholders in deciphering uncertain information within diverse scientific disciplines (Fletcher et al., in pr ep. ). BBNs are one decision support tool that shows promise for integrating multidisciplinary science, uncertainty, and where information is lacking (Lynam et al . , 2003 Lynam et al., 2007; Martin et al., 2012 ; McCann et al., 2006; Nyberg et al . , 2006; van Dam et al., 2013). BBNs are probabilistic models that aid in structuring a decision problem using the best available science and stakeholder knowledge. The BBN model is a visual, transparent representation of an ecosystem using quantitative and/or qualitative information. The process relies on participatory decision support to achieve

PAGE 106

106 consensus, and perhaps more importantly, identify areas of convergence and divergence in stakeholder opinions within the decisionmaking process (Cain et al., 1999; Heckerman, 1996; Kelly et al., 2013; Lynam et al, 2003 Lynam, et al., 2007 Nyberg et al., 2006; Uusitalo, 2007). BBNs are one method to assimilate and address uncertainties in environmental decision making, seek alternative management options, and identify knowledge gaps as described by Cain (2001). The BBN process reveals that stakeholder engagement can be equally as important as the model building and that collective learning is critical in managing natural resources (Lynam et al., 2003). Cain’s (2001) 12step methodology emphasizes stakeholder participation for building BBNs as outlined in his guidebook illustrat ing that engagement efforts are a necessary component of developing a decision support tool. Lynam et al. (2003, pg.57) stated the “most important lesson was the development of a common understanding of a problem.” Emphasis is placed on the importance of s takeholder engagement and the value of developing a model alongside practitioners. The utility of BBNs in creating transparent networks with explicit strength or weaknesses presents a useful tool for socioecological resource management. The transparency allows end users to interpret uncertain and certain information and to evaluate scenarios and management alternatives in a safe place. In this way, we begin to see the value of BBNs as a tool to guide discussions for participatory decision support in comple x socio ecological systems. This chapter describes our experience with developing Bayesian networks in the context of improving the understanding of the coral reef ecosystems in a changing climate for informed decision making. Accordingly the paper address es the following objectives: Describe the steps taken to develop and test two BBNS Animate and assess the utility of BBNs to for use in decision making

PAGE 107

107 Summarize stakeholder opinions for using the BBN s Consequently, this chapter is divided into three parts showing the methodology used to address the objectives, a results section to highlight important features that emerged from the cooperative effort, and a discussion section summarizes the results and lessons learnt in ecosystem based management . Materials and Methods T wo BBNs were constructed using a combination of processes and procedures outlined in Cain (2001) and Chen and Pollino (2012). One network was an ecosystem scale model and the other a habitat scale network. Participatory decision support was used to desi gn and test the prototype models . The networks integrated the knowledge of experts and endusers resulting in a product that was designed both by and for managers to use in a decision making realm. The process spanned fr om September 2 009 to February 2014 with input from 2 13 r esource managers, researchers, academics, nongovernmental organizations, and agency staff at the local, state, and federal level working in South Florida. The project consisted of three phases: defining the proble m statement, constructing the prototype model s, and consultation with stakeholders to assess the utility of the model. Defining the P roblem S tatement The first step in the construction of the BBN s was to identify a problem statement that would guide the development of the network structure. This initial stage was achieved by bringing varied stakeholders together in a variety of formats (e.g., workshops, meetings, conferences, etc.) for the purpose of characterizing the ecosystem and identify drivers and pressures impacting the system (Fletcher et al., 2014). Forty four participants attended a facilitated workshop held in Fort Lauderdale, Florida on March 2930, 2011 to develop a framework for identifying the defining

PAGE 108

108 characteristics of a sustainable southeas t Florida coast al ecosystem, sources of change impacting the ecosystem, key attributes of the ecosystem, ecosystem services, and management responses (Nuttle and Fletcher, 2013c). Agency perspectives and goals were presented to illustrate the varied mandat es and interests of each of the entities involved in the regional management effort (Table 41) (Mitchell, 2011). Participants also reviewed a modified a preliminary list of impact s defined as categories that affect the health or condition of South Florida ’s socio ecological system and described those categories further to include variables to monitor changes in the system (Table 42). This information was used to develop a characterization of the ecosystem. Participants comprised of scientists, managers, and decision makers used the Ecosystem Based Management Drivers Pressures Stressors Ecosystem ServicesResponse (EBM DPSER) framework described by Keble et al. (2013) to construct an integrated conceptual ecosystem model (ICEM) (Figure 41). The collective knowledge of the participants was captured, and later refined through follow up meetings and document review (Nuttle and Fletcher, 2013c). The resulting I CEM was used to assemble more detailed conceptual ecological models (CEMs) of the state of the ecosystem (e.g. water column, fish and shellfish, and marine dependent people) and habitat s (e.g., mangroves, seagrasses, beaches, coral and hardbottom). These models depicted physical processes, impacts, and characteristics of the ecosystem . Over the next two y ears, i ndividuals from the workshop volunteered or were selected to work in small writing teams comprised of experts in their respective fields of science. Participants used an iterative process intended to foster a continuum of learning, with each meeting or workshop building upon the previous to obtain an improved understanding of South Florida’s socioecological system (Hungerford and Volk, 1990). Authors developed problem statements as drivers or mechanisms of change to the state

PAGE 109

109 of the ecosystem habit at and ecosystem services using the ICEMs ( Ault et al., 2013; Browder et al., 2013; Carsey, 2013; Lorenz, 2013; Marshall and Banks, 2013; Riegl et al., 2013) . Constructing Preliminary N etworks Netica software was used to construct decision problems for t wo BBNs, one being an ecosystem scale network and the other a habitat scale network (Norsys, 2014) . The networks consist ed of nodes, links and Conditional Probability Tables (CPTs). Nodes described ecosystem variables (e.g., seagrasses, beaches, mangroves ) and had descriptive (e.g., good/ bad) or quantitative states (e.g., 0 to 56 degrees, 57 to 98 degrees, 99+ degrees) as defined by coral reef scientists or managers. L inks were drawn between nodes and show ed the cause and effect parent child relationships b etween nodes flowing from a parent to a child (Cain et al., 1999) . CPTs are matrices populated with values defining the parent child relationship as a percentage of the likelihood of a specified outcome, or probability that an event will occur . The probabi lities were based on posterior knowledge, or knowledge gained from past events and experiences that comprise the Bayes belief system (Bayes, 1763) . For example, the probability there was a degraded marine ecosystem given declining water quality and a set of pressures ranging from climate impacts to physical damage caused by coastal construction could be high er , or closer to 100% as estimated through published literature or expert judgment . Likewise, if there were water quality improvements and reduced press ures from climate or physical damage the probability that the marine ecosystem was degraded could be low er , or closer to 0% . The belief system simply stated is the likelihood of an event occurring based on ones prior experiences . This can be used to explor e scenarios when dealing with uncertainty or where there is a lack of data (Castelletti and Soncini Sessa, 2007; Kelly et al., 2013; van Dam et al., 2013) .

PAGE 110

110 Ecosystem scale BBN The ecosystem scale prototype BBN was constructed using 19 nodes and 30 links fo llowing the structure of the southeast Florida coastal ecosystem I CEM with the addition of three nodes representing index values for habitats and states in the model (Figure 42) (Nuttle and Fletcher, 2014c) . The addition of the three index nodes was made to simplify the network in preparation for assigning probability values in the CPTs and to more appropriately address stakeholder needs (Cain, 2001; Chen and Pollino, 2012). The ICEM components and three new indices were represented in the Netica software as nodes and each one defined using the descriptions developed in the ICEM process (Table 43) . The ICEM structure was followed to illustrate the transferability of the ICEM to the BBN. The network consisted of the following nodes : Drivers, Pressures, States and Habitats, Desirable Ecosystem Attributes , and Management Responses. Drivers, located at the bottom of the model represented the biophysical, human, or institutional actions that caused change to the ecosystem. Pressures were the result of Drivers and were considered those biological, chemical, or physical processes occurring within the ecosystem. Responses were those management actions that could be taken to address changes to State (including Habitat) and Desirable Ecosystem Attributes. The States component of the model was modified with the addition of three nodes: Habitat Index, Other States Index and an overall State Index. The original nodes were “divorced” from linking directly to Desirable Ecosystem Attributes and were linked to the new index n odes (Cain, 1999). The State Index node had three child nodes: Management Responses, Other States, and Habitat States. The Other States node had three child nodes: Water Column, Fish and Shellfish, and Marine Dependent People. The parent Habitat States node had four children: Mangroves, Beaches, Seagrasses, and Coral

PAGE 111

111 and Hardbottom. Management Response was a parent node to Drivers, Pressures, State Index and Aesthetic Environments. This was done to illustrate to the enduser audience the potential influence for management interventions at multiple levels (global, regional, local) on ICEM components. The new index nodes presented endusers with a “snap shot” of the conditions for ecosystem State s and Habitat . One other modification to the ICEM consisted of l isting each of the Desirable Ecosystem Attributes identified by workshop participants in March 2011 as individual nodes . This aligned with the emphasis of ecosystem services within the EBM DPSER model framework , ICEM, and the ability of endusers to detect the significance of individual parent child node relationships within the network. The Desirable Ecosystem Attributes nodes included the following ICEM components : Aesthetic Environments, Beach Activities, Wildlife Viewing, Fishing Opportunities, Boating Opportunities, and S norkeling/Diving Opportunities. Nodes values assigned to the ecosystem scale network were q ualitative . Values consisted of the following groupings: condition improving, condition same, condition worse; increasing level of use, same lev el of use, decreasing level of use; more, same less; many, few; decreasing impacts, same amount, increasing impacts; or yes and no. The node values were structured to contain the optimal condition first (condition improving), the same condition second (con dition same), and the declining condition last (condition worse). This was useful for consistently populating the CPTs later on in the network development with the more desirable conditions being presented at the top of the table and less desirable conditi ons at the bottom . Definitions of each node and its values were included in the Netica software system table properties feature to allow endusers a convenient way to access descriptions and intentions of the node as

PAGE 112

112 defined during the workshop and subsequent stakeholder consultations that followed (Table 43). The c onditional probabilities for this initial model were assigned by the network developer using information gathered during the EBM DPSER project . The probability values were allocated using a step wise process examining each of the parent child node relationships oneby one and assigning values using EBM DPSER project reports and information obtained during workshops . T he table option for each node was opened in the Netica software and then populated with a percent value that reflected the likelihood of an outcome. For example, the State Index probabilities were assigned by asking the following question for each parental node linked to it “ what is the percent probability that the State Index=C ondit ion Improving if Habitat State=conditions improving, Other States=condition improving, and Management Response=Yes.” The process continued for all entries in the table, and for every child node in the network. Once the CPTs were complete, t hree network sc enarios were run to test the prototype model output. The automated probability inference feature in Netica software was used to animate the network by selecting different node values. In this case, t he Drivers node was manipulated to reflect a value of 100% for each of the three conditions in the node (Decreasing Impacts, Same Amount, and Increasing Impacts). The Netica software performed an auto update animating the network with posterior beliefs and delivering the results immediately in the network visual ization. The network was further explored by establishing two management scenarios, Situation A and Situation B (Table 46) . The situations were created to test the validity of the network node values . Aesthetic Environments was tested by manipulating the values for Management Response and Drivers. The Aesthetic Environments node was selected because of the EBM DPSER project emphasis to integrate human dimensions

PAGE 113

113 science and biophysical science into ICEMs and CEMs. The ICEM project described Ecosystem Servi ces within the context of those “benefits that people receive from the ecosystem that link people to the State of the ecosystem, through “attributes [of the environment] that people care about. They can be measured in a monetary, cultural, or social contex t, and the value of Ecosystem Services depends on conditions in the environment” ( Farber et al., 2006; MEA, 2005 as referenced in Nuttle and Fletcher, 2013c, page 25). The ICEM defined Aesthetic Environments as those things that “provide aesthetic quality of aquatic and terrestrial environments (visual, olfactory, and auditory), therapeutic benefits, pristine wilderness for future generations” (Nuttle and Fletcher, 2013, page 26). In Situation A, the CPT for Drivers was set to 50% for a Decrease or Same Amount of influence if Management Responses were set to Yes, and a 50% probability that Drivers will be the Same or Increase if there is No Management Response. The CPT for Situation B reflects relatively equal probabilities (33%) that there will be Decreasing, Same Amount, or Increasing impacts with a Yes or No Management Response. Thus , Situation A reflected a more active management response that positively impacted the ecosystem, and Situation B presented a management response that did little to address impacts to the ecosystem from Drivers. In the two scenarios, the Yes and No values for Management Responses node were manipulated to 100%. Four sensitivity analyses w ere run using Netica’s automated sensitivity to findings feature for Aesthetic Environments n ode. A ll other Desirable Ecosystem Attributes were deleted from the network to avoid influencing outputs of the analysis. The sensitivity analyses w ere set up using S ituation A and S ituation B that were dev eloped for the management scenarios . Therefore, th e BBN could be used by managers to identify the potential outputs for management interventions based on far -

PAGE 114

114 field drivers impacting the southeast Florida ecosystem. The Management Response node was set to Yes/No in Situation A and B, and the Netica automat ed feature for sensitivity to findings run for the Aesthetic Environments node. Habitat scale BBN The habitat scale network for coral reef and hardbottom s was constructed with 14 nodes and 17 links (Figure 43) . The network used only those components of the coral reef and hardbottom CEM related to climate impacts (Fletcher et al., 2014; Nuttle and Fletcher, 2013c; Riegl et al., 2013) . The determination to use only climate related information selected from the CEM was made based on results from the EBM DPSER documentation (Riegl et al., 2013; Lirman et al., 2014) and a climate information needs assessment conducted with coral reef and marine resource managers in southeast Florida and the Florida Keys (Fletcher and Hendee, 2012). The March 2011 workshop and interactions with coral reef researchers were used to gather comments and background documentation to develop and test the network with endusers. Greater details concerning this testing with coral reef managers is found in Fletcher et al. (2014) (in preparation). Consulting Stakeholders to Assess the Utility of the Networks Stakeholder consultations were carried out to evaluate the usefulness of the BBNs as a decision support tool . An evaluation survey w as designed and then administered to 11 coral reef and marine resource managers in southeast Florida. Analysis of the responses was completed using both percentages and a median response rate. The evaluation instrument was developed with input from a social scientist and a project design and evaluation special ist from the National Oceanic and Atmospheric Administration’s Coastal Services Center (NOAA/CSC 2003, 2009). The instrument

PAGE 115

115 consisted of a project introduction, consent form, a five question presurvey, a tutorial and exercise for using the network, and a seven question post survey. The survey was designed to be administered using a consistent format to avoid biasing responses. Each component of the survey was completed prior to moving on to the next. The project introduction and consent forms were used to present background information and to ensure stakeholders were willing participants in this effort. The presurvey consisted of five questions. The first question was used to gather information about the decision making process at the stakeholders office. The second question asked the stakeholder to categorize their role(s) in coral reef and marine resource management to characterize their involvement in the decision process . The next three survey questions were identical in the pre/post survey in order to measure a before and after impression of the networks as a decision support tools. The questions were: 1) whether managers believed the BBN decision support tool would be useful for making management decisions related to climate impacts to marine resources and coral reef ecosystems; 2) whether managers would consider using BBN decision support tools detailing pressures and stressors to marine resources and coral reef ecosystems in future management decisions; and, 3) if a manager would recommend using a BB N decision support tool to other marine resource managers within their management portfolio. The responses were a 5 point scale response (1=not useful, 2, 3=moderately useful, 4, 5=very useful, or 1=No, 2=Doubtful, 3=Maybe, 4=Likely, 5=Absolutely). The tutorial portion of the survey was used as a guide to introduce BBNs and to encourage stakeholders to manipulate the prototype BBNs (Figure 44 ) . A brief paragraph presented the BBN concept and illustrations depict ed how the EBM DPSER model , ICEM , and CEM were used to design the ecosystem scale and habitat scale

PAGE 116

116 networks . Participants followed instructions to manipulate the node values in the ecosystem model to see how the network changed when probability values were altered. Then they followed directions t o run sensitivity to findings in the habitat model. After the tutorial, a post survey consisting of seven questions was administered. Three of the questions were identical to the presurvey. Two questions were structured to conduct a longterm assessment o f using the decision support tool that will be part of ongoing research. One question asked participants to rank the degree to which the tool increased their knowledge about climate impacts to marine resources and coral reef ecosystems. The last question asked the stakeholder if they will use the tool in their decision making. The survey instrument was administered to 11 stakeholders responsible for managing, conducting research, o r synthesizing science for management of the marine and coastal resources i n southeast Florida between January 8, 2014 and February 25, 2014 (Table 45) . The se individuals were identified as stakeholders through their involvement in one or more of the following activities : 1) the EBM DPSER project from September 2009 – December 2013 (Fletcher et al., 2014; Nuttle and Fletcher, 2013a,b,c), 2) taking part in informational conversations regarding climate information at natural resource and/or marine resource management offices in October 2011 (Fletcher and Hendee, 2012), or 3) parti cipation in a coral reef and marine resources manager climate information needs assessment in November 2011 May 2012 (Fletcher et al., in preparation ; Fletcher and Hendee, 2012). One survey was conducted online using GoToMeeti ng software (GoToMeeting, 2014) due to travel limitations and all of the others were held inperson. One survey session was attended by three managers from the same agency and all others were conducted as oneonones. All of the presurveys were conducted inperson or using GoToMeeti ng. Four post surveys were completed in

PAGE 117

117 person and seven were completed after the inperson meeting, scanned, and submitted electronically. The survey responses were entered into a spreadsheet for analysis. Percentages of the frequency of the response were calculated. The median was also determined for each of the survey answers. Results Defining the P roblem S tatement The problem statement for the development of the BBN (Cain, 1999) was expressed in terms of drivers and pressures impacting the ecosystem during facilitated discussions at the EBM DPSER workshop. The EBM DPSER drivers and pressures that negatively impacted the southeast Florida coastal ecosystem were “far field drivers [that] include global climate change and climatic extremes (e.g., sea level rise, high and low temperatures, storms, acidification), diseases , and invasive species ” (Lirman et al, 2014, page 58). These terms were transformed into the ICEMs and CEMs , and subsequently into the BBNs . The BBNs were then used to explore scenarios of far field d rivers on the South Florida ecosystem with results described below in this paper. Constructing Preliminary N etworks The two prototype BBNs included all EBM DPSER components with the addition of t hree index nodes (State Index, Habitat Index, and Other States) in the ecosystem scale ICEM , and the exclusion of all nonclimate related nodes in the coral and hardbottom CEM . The ICEM and CEM frameworks were used in the placement of nodes in the Netica software with relative ease. Divorcing the State c omponents to three indices was helpful for both computational purposes in the Netica software and for management discussions. Limiting the number of parent child links to a single node was useful for CPT review and interpreting the contents of the table. Too many entries required additional time to assign probabilities and slowed the stakeholder consultation

PAGE 118

118 process. Adding the three index nodes reduced the number of CPT entries and aligned with the concept of developing the BBNs for the stakeholder audienc e who were interested in exploring ways to assess and report ecosystem health (e.g., ecosystem status report or ecosystem report cards) ( Loomis et al., 2014; RECOVER, 2012) . In the ecosystem scale model, stakeholders explored causal probabilities among se lected components. T he node Drivers was set at 100% for three different intervals (Decreasing Impacts, Same, Increasing Impacts) to examine changes in the Aesthetic Environments node ( Table 46 ) . In general, the results were expected with Aesthetic Environ ments having the highest probability for improving when Drivers were decreasing 100%, and becoming worse when Drivers were increasing 100%. One surprise was the Condition Same value for Same Amount (12.9) having a lower probability than Condition Improving (29.6). The model was revisited and CPT values altered to identify if there was threshold for having the Condition Same output higher than Condition Improving. None was observed illustrating that more weight was assigned to the value of nodes with Conditi on Improving and the cascade of impacts through the model. One practical consideration for the result was the lack of a timestep in the network and a network void to include resiliency of the States and Habitats in the ecosystem. Management scenarios wer e created to test the network outcomes for determining the influence of Management Response on Aesthetic Environments in relation to Drivers (Table 48) . In Situation A, Management Responses were taken (Yes) and were set to influence the likelihood of Driv ers having a 50% probability of either Decreasing Impacts or having the Same Amount of impact on the ecosystem. In Situation B , probabilities were set to illustrate that Management Response (Yes) had no significant change on the likelihood of Drivers improving or decreasing the ecosystem

PAGE 119

119 with values for Yes/No set to 3334% for node values . These values were set within the context of far field impacts (global climate change) that managers may not be able to locally manage based on stakeholder input during I CEM development . Again, there was no surprise for Situation A where Management Interventions were set to have a stronger influence on the likelihood of outcomes improving or a lack there of. However, there was again a lower rate for Condition Same in every instance with the exception of a slight (0.4) increase when the Management Response No was used in Situation A. The more significant finding in the model is the difference in the probabilities when the influence of Management Responses was set to the Yes/ No values with 50 or 0% for Situation A and 3334% in Situation B. Comparing the Yes Management Response outputs, the difference between S ituation A and B is a12.3% probability that conditions can improve with stronger management, and a 14.8% probability that conditions will be worse if there is lower effort placed on management actions. Comparing the No Management response outputs, the difference in the Situation A and B was 7.2% probability that conditions will improve and a 10.6% that conditions will be worse if there is no management action. Four sensitivity analyses were run for the Aesthetics Environments node to more closely examine how management responses in Situation A and B influence the selected desirable ecosystem attribute Aesthetic Environments (Table 49 and 410). Three nodes consistently had the highest influence on the situations: State Index, Other States, and Habitat States. This was no surprise because of the parent child relationship established in the network. For the same reason, t h e State Index had the highest percent influence on Aesthetic Environments. In all four analyses, the percentages of the Pressures, Mangroves, FishAndShellfish, WaterColumn, Seagrasses, CoralAndHardbottom, and Beaches were within a range of 60.948.4%. All

PAGE 120

120 of these nodes with the exception of Pressures were included in the State box in the ICEM model and therefore in the network. In Situation A there was a small percentage range (1.2%) in which these were listed. In Situation B, the range was 1.3%. The inc lusion of Pressures in this grouping of responses was no surprise because of the parent child relation of Pressures to the three habitats. In all instances, t he nodes with the least influence were consistently ranked in the following order: Drivers, Marin eDependentPeople, and Management Responses. The Drivers node was not a surprise because of the more distant location within the network. MarineDependentPeople was a surprise because within the range of activities associated with this node there are extract ive and nonextractive activities that can impact the aesthetic environment (e.g., physical damage caused by or biological impacts caused by extractive uses such as fishing) . The sensitivities for these three nodes were lower than expected. The Management Response of a 0% value in all instances was unexpected. The network was further manipulated to investigate these results. The Management Response link to Aesthetic Environments removed and the sensitivity remained the same for the management intervention. Next the State Index CPT was reviewed to identify the strength of the Management Response in this node. Contained within was the justification for the low ranking of the management node. The probabilities assigned to the network for the indices (Habitat St ates and Other States) when there was a Yes/No Management Response were always the same. Probabilities were adjusted in the table and it became clear that populating the CPT for these indices would require more explicit instructions to improve the understanding of the topical focus of the ecosystem model by all stakeholders . Additionally, more detailed models for each habitat and state could be used to clarify the condition of the ecosystem component (improving, same, worse) and feed into the ICEM ecosystem model .

PAGE 121

121 The habitat scale network for coral reef and hardbottom habitats was developed using a portion of the CEM designed during the EBM DPSER workshop in March 2011. Only the climaterelated components were placed into the Netica software as nodes and followed the links depicted in the CEM. Q uantitative and qualitative information was used to populate the network using a literature review and through expert opinion with coral reef researchers . In general, the network showed there would be lower quality coral reef and hardbottom when climate impacts were greater. Parent nodes of the coral Aesthetics had the highest sensitivity to findings as expected in the network. A complete analysis of the results can be found in Fletcher et al. (2014). Consulting Stakeholders to Assess the Utility of the Networks Overall, managers commented that the ecosystem network was intuitive and not very useful; however, it was a visually transparent format to comprehend the cascade of impacts through the ecosystem and a good introduction to BBNs. Some managers began to explore different components of the networks and asked for additional information about the network structure. They were encouraged to continue manipulating the network by changing values of other nodes, to examine the CPTs, and to use the sensitivity analysis feature of the Netica software. Results of the pre/post survey assessing three aspects of the BBN using a five point scale and one question with a yes/no response are reported in Table 412 . The median response for managers stating if decision support tools would be useful for making management decisions related to climate impacts to marine resources and coral reef ecosystems was a 4, or between 3 (moderately useful) and 5 (very useful) for both the pre/post surveys. The median response was 4 (likely) for the pre/post survey when asked if they would consider using decision support tools detailing pressures and

PAGE 122

122 stressors to marine resources and coral reef ecosystems in future management decisions. The third pr e/post survey question also had a consistent median, 4 (likely) when asked if they would recommend using a decision support tool to other marine resource managers within their management portfolio. The post survey asked two additional questions. The median response for the degree to which the BBN tool increased knowledge about climate impacts to marine resources and coral reef ecosystems was 4 (likely). The final question asked if they believed they would use the BBN tool in their decision making, 9 individ uals said yes, and 2 stated no. The two no responses were due to more accurate data being available at one individuals office, and the other individual stated that this type of analysis in not directly related to their work responsibilities. Implications D efining the Problem Statement The problem statement to drive the development of BBNs was defined through a series of facilitated workshops and exploration of the socioecological impacts and activities affecting the South Florida marine and coastal ecosyst em in the EBM DPSER process . The identification and definitions for the Drivers, Pressures, State (habitat), Ecosystem Services and Responses were necessary to obtain a clear understanding of the ecosystem and those factors influencing its condition. The r esult ing ICEM and CEM and supporting documentation were used to construct the BBN in Netica for further exploration in the decision making realm. The relative ease of using BBNs versus other more complex and sometimes expensive mathematical models proved useful in transferring the EBM DPSER biophysical and human dimensions science to managers. However, the network purpose should be clearly defined (e.g., ecosystem scale, level of

PAGE 123

123 detail, importance of one ecological state over another) as early as possible in the process to ensure CPTs values are appropriately assigned. Constructing Preliminary Networ k Bayes Theorem when applied to a BBN contributes to decision making because the nodes, links and probabilities reflect a decision problem that can be examined and refined using expert opinion, the literature, or other inputs (e.g., model). The inputs, prior knowledge, become outputs, posterior knowledge and inform decision making in a visually transparent manner in Netica software. The BBNs help ed validate opini ons and explore management alternatives in a flexible, “group thought” mode that can validate collective knowledge and contribute to management strategy alternatives (Cain, 2001; Lynman, et al., 2003; Lynam et al. , 2010; Nyberg et al., 2006; Uusitalo, 2007; van Dam et al., 2013). BBN software runs more efficiently by following rules (e.g., using no more than 3 parent child relationships) and should be noted that these models truly are simplified representations of the world that do make not make decisions themselves, but provide insights to the decision process and expert elicitation. While no other method was compared during this research, the prototype networks were developed in a twoweek period following years of ICEM and CEM development, followed by several weeks of stakeholder consultation and refinement. Conditional probabilities are useful for unraveling of the complexities associated with wicked environmental problems because they are good for understanding perceptions (inference) and adaptive lear ning. The conditional probabilities placed on the likelihood of an event occurring provide the mechanism to include uncertain information in the reasoning process in some form (e.g., quantitative or qualitative) based on prior knowledge. One important note from the testbed of participants was the ability to downscale and upscale analysis of information to obtain a “pulse” of the

PAGE 124

124 ecosystem and to also evaluate in more detail changes occurring within specific areas of the ecosystem. The relative ease of us e of the BBNs, flexibility and transparency of the models was useful to assess risk and uncertainties tied to the results with both quantitative and qualitative information. The concept of managers being “the first to know” what is happening on the reef, or the possibility of conditions on the reef is invaluable (Fletcher and Hendee, 2012). Using Netica resulted in a limited number of variables being used, or in this case an “index” for variables, for example, t he “State I ndex” representing several ecologic al states in the EBM DPSER model. This can affect the sensitivity analysis and perceptions of the strength of one particular node on other, so the network must be designed appropriately (Lynam , 2007). In addition, the lack of a timestep to evaluate changes in the ecosystem was one drawback of the software. However, the software was helpful for addressing missing data and most importantly for visualizing the model outputs. Several stakeholders commented on the feature and began to explore other areas of the model beyond the tutorial provided. This opened the discussion of BBNs abilities to incorporate expert knowledge that can be combined with data when no data exists (Marcot et al., 2001), and the belief that the networks themselves have value in obtaining expert opinion, but tend to provide more accurate estimates than direct assessments of probabilities (Armstrong et al.,1975). Consulting stakeholders to assess the utility of the networks The research tested the utility of a decision support tool developed with stakeholder consultation. The project success was dependent upon the willingness of stakeholders to explore their role in the transfer of information to knowledge and the application of knowledge innovations. Stakeholders were made aware of the purpose of

PAGE 125

125 the prototype BBN project and process and expectations that they were to contribute to building knowledge from information over a longer period of time. In summary, this research outlined the steps to build BBNs to visually and transparently deliver interactive ecosystem models to stakeholders. The ecosystem scale network consisted of a prototype for characterizing and presenting socioecological components of the ecosystem to enhance stakeholder’s understanding of the system and adaptive management t hrough scenarios. The coral reef ecosystem BBN integrated products from several research efforts to formulate a process for adaptive exploration of complex environmental challenges. Resource managers, many of who were seeing BBNs for the first time were able to animate the models and assess them following a short tutorial. The BBNs were also useful to explore t he role of tacit knowledge for building a mathematical model that uses beliefs or representations of the world around us to frame our reasoning (Pear l and Russell , 2000). This is particularly useful when dealing with uncertainty in managing natural resources. The relative ease of using BBNs versus other more complex and sometimes expensive mathematical models proved useful in transferring biophysical and human dimensions science to managers. The stakeholder responses recorded in the pre/post surveys present the potential for continued development of the BBNs as d ecision support tools .

PAGE 126

126 Figure 41. Integrated conceptual ecosystem model for the southeas t Florida coast al ecosystem (Nuttle and Fletcher, 2013c) .

PAGE 127

127 Table 41 . Agency perspectives and goals were identified as the first step in the problem statement identification to explicitly recognize the varying mandates and goals of each agency involved i n developing the EBM DPSER model to represent the south Florida marine and coastal ecosystem (Mitchell, 2011). Agency Perspectives and Goals N OAA/Florida Keys National Marine Sanctuary The FKNMS has a keen interest in the MARES objectives of developing goals for and indicators of a restored sustainable South Florida coastal system. The Conceptual Ecological Models will be used to revise Sanctuary science and management plans, especially in expanding the scope of the Sanctuary’s science plan to more explicitly address external influences on the con dition of Sanctuary resources. NOAA/National Marine Fisheries Service The NOAA Southeast Fisheries Science Center has particular interest in the total system Conceptual Ec ological Model (CEM) for South Florida, because of its mandate to provide scientific data for the longterm management of marine resources that utilize coastal areas as feeding and nursery grounds. The total system CEM, by linking fresh water flows, northern and southern estuarine ecosystems, adjacent coastal habitats and the South Florida shelf areas will contribute significantly to SEFSC’s understanding of how coastal ecosystem dynamics, including human influences, affect federally managed resources. U. S. National Park Service The South Florida Natural Resources Center provides science directed toward the management needs of Everglades and Dry Tortugas National Parks. The goals and Conceptual Ecological Models for the southwest Florida shelf and the F lorida Keys/Dry Tortugas will be of particular interest to these Parks in refining resource management and science plans. The Keys/Dry Tortugas workshops will provide a forum for exchange of information regarding the interagency NPS/FFWCC program to asses s the efficacy of the Dry Tortugas NP Research Natural Area. U.S. Fish & Wildlife Service The MARES Conceptual Ecological Models will assist the Service’s Coastal Program to meet its primary goal to conserve and restore habitat for fish and wildlife spe cies, by assisting the Coastal Program in developing geographic focus areas in South Florida. The establishment of feasible and quantifiable goals and targets for South Florida will aid the Service’s efforts in the recovery and protection of trust resourc es, including Federally listed threatened and endangered species. The MARES project will benefit the Service, which is a CERP partner, by contributing to the refinement of CERP performance measures in the Southern Coastal Systems region, and perhaps other regions. Also, the Service anticipates that MARES will help in the development of critical elements of our Strategic Habitat Conservation initiative.

PAGE 128

128 Table 41. Continued Agency Perspectives and Goals U.S. Env ironmental Protection Agency EPA works to ensure that environmental decision making in South Florida is based upon sound science and that progress is measurable by producing positive environmental and/or socioeconomic outcomes. Scientific findings gained through research and monitoring, combined with the collaborative process, increases the probability of reaching desirable environmental outcomes. The MARES Conceptual Ecological Models will be based upon current knowledge of the south Florida marine ecosy stem and will result in an improvement to our understanding of the components and their interactions. The manager/stakeholder workshops will help connect scientific findings to the user community. South Florida Water Management District A goal of the Di strict is the protection and restoration of estuarine ecosystems along the South Florida coastline. Environmental science provides a basis for defining the environmental needs of these systems and identifying targets for improved water management through actions such as the CERP and water allocation policies. The MARES project is a contribution to a comprehensive integrated framework for evaluation of South Florida coastal systems, and water management effects on them. Florida Fish & Wildlife Conservat ion Commission The Florida Fish and Wildlife Conservation Commission’s mission is to m anage Florida’s fish and wildlife resources for their long term well being and the benefit of people. Accomplishing this mission requires the synthesis of the physical and natural sciences with better understanding of social issues combined with stakeholder involvement . The MARES approach of incorporating human dimensions into Conceptual Ecosystem Models will be a substantial step forward in their usefulness as a broadbased tool for fish and wildlife management in south Florida. Florida Department of Environmental Protection/Coral Reef Conservation Program A goal of the FDEP CRCP is to promote and contribute to the management of the Florida Reef Tract as a holistic s ystem. Through the Southeast Florida Coral Reef Initiative (SEFCRI), FDEP CRCP is generating data and working with stakeholders to develop alternatives for the SEFCRI goal of establishing a managed area and management plan for northern extension of the Florida Reef Tract. The MARES integrated CEM, goals and indicators will support the implementation of these projects.

PAGE 129

129 Table 42. Problem statement identification was achieved by listing categories of things impacting the condition of marine and coastal res ources with examples of the variables that could be used to measure and monitor change. Category Variables Management objectives things wis h to affect through management resources Productivity, resilience, protection, education/research, aesthetics/cultur al value, recreation Interventions things want to implement in order to achieve objectives (management objectives) Climate change coping strategies, reduced coast impacts, water management, fishery harvesting regulations, reef restoration of physical damage, invasives management, lifestyle, response by individuals, artificial reef placement (reduced diver pressure) Intermediate factors factors which link objectives and interventions (one off or over long period of time) Sea level rise depth distribution change, storms physical damage, temperature (coral bleaching), salinity, turbidity (light), pollution Controlling factors factors which cannot be changed by intervening at the scale you are considering but control the environmental system at that scale in some way Global climate change greenhouse gases (population), rainfall, economy, policies Implementation factors factors which directly affect whether the intervention can be successfully implemented both immediately and in the future (depending on whether the intervention is implemented as a oneoff or longer period) Research/monitoring funding, community support/awareness Additional factors factors which are changed as a result of interventions that do not affect anything else in the environmental syst em Increased fish and protected species populations, increased opportunities for recreation

PAGE 130

130 Figure 4 2. Bayesian belief network for the ecosystem scale Integrated Conceptual Ecosystem Model for the southeast Florida coast (Nuttle and Fletcher, 2013c) .

PAGE 131

131 Table 43. Integrated conceptual ecosystem model nodes, descriptors, and values used to develop the Bayesian network for the southeast Florida coast. Node Descriptor Values Drivers Changes resulting from humans on the current condition of the ecos ystem (e.g., population, waste management, water management, global climate change (includes sea level rise)) Decreasing changes Same amount Increasing changes Pressures Impacts from Drivers on the ecosystem (e.g., land based sources of pollution, fishing, diving, coastal construction, natural phenomenon, lack of awareness) Many Few CoralAndHardbottom Represents reef and hardground ridges along SEFL coast. Consists of hardbottom areas, patch reefs, worm reefs that are colonized by octocoral, stony coral, macroalgae, and sponge assemblages. This area is adjacent to a densely populated area (greater Miami. Condition improving Condition same Condition worse Seagrasses Benthic areas colonized by seagrasses and macroalgae.7 Species of grass are found in South Florida and help support fisheries and maintain water quality. Condition improving Condition same Condition worse Beaches Landscapes comprised of sand located along the marine interface of the SEFL coast. Condition can mean areal extent, contamination, etc. Condition improving Condition same Condition worse Mangroves Landscapes comprised of one or more of the mangrove tree species. Condition can mean aerial extent, health. Condition improving Condition same Condition worse HabitatStates Summary node ca pturing the ecological state of the habitats identified for this model (e.g., Coral and hardbottom, Seagrasses, Beaches, and Mangroves) Condition improving Condition same Condition worse WaterColumn Encompasses all aspects of water quality. Physical, che mical and biological characteristics of the water column along the SEFC. Includes: suspended benthic sediment, phytoplankton, and zooplankton. Condition improving Condition same Condition worse

PAGE 132

132 Table 43. C ontinued Node Descriptor Values FishAndShellfi sh Includes recreational and commercial fishing activities targeting coral reef and hardbottom communities. Condition improving Condition same Condition worse MarineDependentPeople The "people" part of the ecosystem comprised of three levels of users based on the degree to which their activities take place in or near the coastal marine environment: primary, secondary, and tertiary. Increasing level of use Same level of use Decreasing level of use OtherStates Summary node capturing ecological “state” of o ther components of the ecosystem (e.g., Water Column, Fish and Shellfish, and Marine Dependent People) Improving Same Worse StateIndex Represents the summary of the State conditions for the SEFC. Considered as the "report card" of the condition of the St ate. More detailed information about the “states” can be obtained by constructing a BBN for each node (e.g., water column and coral and hardbottom habitats). Condition improving Condition same Condition worse ManagementResponses Responses people can take related to changes in the State of the environment of Ecosystem Services. The SEFC workshops identified these as options. Protection: limiting use of resources through permits or existing laws (coral protection act) Research and Monitoring: study the res ources (including human dimensions) to understand the conditions of the Resources Management: various management, policy, and outreach activities that can be used to carry out agency mandates Fishery Regulations: focus on fish harvesting and methods Hydr ologic Restoration: focus on water quality and water delivery to coastal/marine areas Southeast Florida Climate Compact: focus on actions and awareness about the impacts of climate on resources and human activities Yes No

PAGE 133

133 Table 43 . C ontinued Node Des criptor Values AestheticEnvironments Aesthetics refer to the conditions found in the southeast coastal marine ecosystem. Examples consist of the clarity of the water, or water quality. Condition improving Condition same Condition worse BeachActivities De fined as those activities occurring on or near the beach (e.g., beach walking, sun bathing, fishing, wildlife viewing (birds, sea turtle nesting)). More Same Fewer WildlifeViewing Opportunities to see wildlife at the marine and coastal areas (e.g., manatees, sea turtles, reef fish). More Same Fewer FishingOpportunities Recreational opportunities to catch fish (e.g., availability of fish) Improved Same Less BoatingOpportunities Recreational opportunities to go boating (e.g., opportunities for access t o marine/coastal resources) Good Moderate Poor DivingSnorkelingOpportunities Opportunities to SCUBA dive and snorkel in the region. Consideration to "satisfaction" of trips, but relies more heavily on the ability to dive or snorkel in clean, clear waters. Good Moderate Poor

PAGE 134

134 Table 4 4. Conditional probabilities assigned for the “State_Index” node in the EBM DPSER integrated conceptual ecosystem model for the southeast Florida coast al ecosystem . Habitat_States Other_States Management Responses Conditi on Improving Condition Same Condition Worse Condition Improving Condition Improving Yes 100 0 0 Condition Improving Condition Improving No 100 0 0 Condition Improving Condition Same Yes 50 50 0 Condition Improving Condition Same No 50 50 0 Condition I mproving Condition Worse Yes 33 34 33 Condition Improving Condition Worse No 33 34 33 Condition Same Condition Improving Yes 50 50 0 Condition Same Condition Improving No 50 50 0 Condition Same Condition Same Yes 0 100 0 Condition Same Condition Same No 0 100 0 Condition Same Condition Worse Yes 0 50 50 Condition Same Condition Worse No 0 50 50 Condition Worse Condition Improving Yes 33 34 33 Condition Worse Condition Improving No 33 34 33 Condition Worse Condition Same Yes 0 50 50 Condition Wors e Condition Same No 0 50 50 Condition Worse Condition Worse Yes 0 0 100 Condition Worse Condition Worse No 0 0 100

PAGE 135

135 Figure 43. BBN for the coral reef and hardbottom CEM for the southeast Florida coast . BN_Coral_ClimateChange_MARES_UsingIEAs_Manuscript_v6TESTING_cleane d.neta

PAGE 136

136 Table 45. Managers who participated in the BBN pre/post survey. Permission for interviews was given through University of Florida Institutional Review Board permit #2013U 913. Name Title Agency Ken Banks Natural Resource Specialist IV Broward County Environmental Protection and Growth Management Department Rene Baumstark Associate Research Scientist Center for Spatial Analysis Information Science and Management, Fish and Wildlife Research Institute, Florida Fish and Wildlife Conservation Commission Karen Bohnsack NOAA Coral Fellow Consultan t ECS, Florida Department of Environmental Protection, Coral Reef Conservation Program Billy Causey Regional Director Southeast Atlantic, Gulf of Mexico and Caribbean Region, NOAA’s Office of National Marine Sanctuaries Ernest M. Cowan Chief Biologist Bu reau of Parks District 5, Florida Park Service Steven R. Dale Park Manager III John U. Lloyd Beach State Park, Florida Park Service Lou Fisher Retired Natural Resource Specialist III Marine Section, Broward County Environmental Protection and Growth Mana gement Department Kathy Fitzpatrick Coastal Engineer Martin County Coastal Engineering Department Charles Jabaly District Biologist Bureau of Parks District 5, Florida Park Service Janet Phipps Coral Reef Ecologist Palm Beach County Department of Envir onmental Resources Management Erik Stabenau Oceanographer Department of Interior, Technical Lead, National Park Service

PAGE 137

137 Figure 4 4. Tutorial for Bayesian Belief Network decision support tool based on the integrated conceptual ecosystem model and integrated ecological model for the Southeast Florida Coast and the coral reef and hardbottom habitats in southeast Florida and the Florida Keys.

PAGE 138

138 Figure 4 4. C ontinued

PAGE 139

139 Table 46 . BBN probabilities for Aesthetic Environments when Driver values are manipulated to three different values. Aesthetic Environments % Decreasing Impacts100% Condition Improving 80.8 Condition Same 17.6 Condition Worse 1.53 Same Amount 100% Condition Improving 29.6 Condition Same 12.9 Condition Worse 57.4 Increasin g Impacts 100% Condition Improving 0 Condition Same 5.29 Condition Worse 94.7 T able 4 7 . Conditional Probability Table used to run sensitivity analysis of Drivers based on Management Response Situation A and Situation B. M anagement Response Drivers D ecreasing Impacts Drivers Same Amount Drivers Increasing Impacts Situation A Yes 50 50 0 No 0 50 50 Situation B Yes 33 33 34 No 33 33 34 Table 48 . Scenario analysis for Situation A and Situation B for the ecosystem scale BBN showing the beli ef network probabilities for Aesthetic Environments when Management Response is manipulated from Yes to No using two situation analyses (see Table 46 for CPTs ). Situation A Management Response % Situation B Management Response % Yes 100% Condition Improving 60.6 Condition Improving 48.3 Condition Same 14.5 Condition Same 12.7 Condition Worse 24.8 Condition Worse 39.1 No 100% Condition Improving 13.5 Condition Improving 23.7 Condition Same 5.8 Condition Same 6.2 Condition Worse 80.7 Conditio n Worse 70.1

PAGE 140

140 Table 49. Sensitivity to findings for Aesthetic Environments node when Management Response is set to Yes for Situation A and Situation B. Situation A Node-Decreasing Impacts 100% % Situation B Node-Decreasing Impacts 100% % State_Index 76 .4 State_Index 82.2 Other_States 54.5 Habitat_States 60.9 Habitat_States 53.0 Other_States 60.9 Pressures 50.7 Pressures 58.5 Mangroves 50.0 Mangroves 57.9 FishAndShellfish 49.5 FishAndShellfish 57.2 WaterColumn 49.5 WaterColumn 57.2 Seagrasses 48.7 Seagrasses 56.2 CoralAndHardbottom 48.6 CoralAndHardbottom 56.0 Beaches 48.4 Beaches 55.8 Drivers 18.9 Drivers 20.5 MarineDependentPeople 8.04 MarineDependentPeople 9.1 ManagementResponses 0.0 ManagementResponses 0.0 Table 410. Sensitivity to fin dings for Aesthetic Environments node when Management Response is set to No for Situation A and Situation B. Situation A and B Node-Same Amount 100% Situation A Node-Same Amount 100% State_Index 86.8 State_Index 82.2 Habitat_States 61.4 Habitat_States 62.3 Other_States 59.0 Other_States 60.2 Mangroves 57.0 Mangroves 59.0 FishAndShellfish 54.7 Pressures 57.9 WaterColumn 54.7 FishAndShellfish 57.3 Pressures 54.4 WaterColumn 57.3 Seagrasses 52.5 Seagrasses 55.8 CoralAndHardbottom 52.1 CoralAndHardb ottom 55.4 Beaches 51.9 Beaches 55.2 Drivers 14.2 Drivers 14.3 MarineDependentPeople 6.74 MarineDependentPeople 8.0 ManagementResponses 0.0 ManagementResponses 0.0

PAGE 141

141 Table 4 1 1 . Survey responses to questions relating to a marine and coastal resource decision support tool (n=11). Questions Alternate responses Pre test Post test Response(%) Median Response(%) Median Do you think decision support tools would be useful for making management decisions related to climate impacts to marine resources and coral reef ecosystems? 1 Not useful 2 3 Moderately useful 4 5 Very useful 0 1 (9.0) 3 (27.2) 3 (27.2) 4 (36.3) 4 0 1 (9.0) 2 (18.1) 4 (36.3) 4 (36.3) 4 Would you consider using a map based tool and/or an ecosystem model detailing pressures and str essors to marine resources and coral reef ecosystems in future management decisions? 1 No 2 Doubtful 3 Maybe 4 Likely 5 Absolutely 0 0 2 (18.1) 5 (45.4) 4 (36.3) 4 0 0 2 (18.1) 5 (45.4) 4 (36.3) 4 Would you recommend using a map based tool and/or an ecos ystem model to other marine resource managers within their management portfolio? 1 No 2 Doubtful 3 Maybe 4 Likely 5 Absolutely 0 0 2 (18.1) 4 (36.3) 5 (45.4) 4 0 1 (9.0) 3 (27.2) 2 (18.1) 5 (45.4) 4 Questions specific to the BBN tool Alternate responses Responses Median Rank the degree to which the tools increased your knowledge about climate impacts to marine resources and coral reef ecosystems? 1 No 2 Doubtful 3 Maybe 4 Likely 5 Absolutely 0 3 (27.2) 2 (18.1) 2 (18.1) 4 (36.3) 4 Do you be lieve you will use one of these tools in your decision making? Yes No 9 2

PAGE 142

142 CHAPTER 5 CONCLUSIONS The hypothesis of this research, participatory decision support using scenario modelling improves the understanding of the ecosystem for informed deci sionmaking h as been supported by resource managers survey responses . Par ticipatory decision support was needed to improve the understanding of the ecosystem and to synthesize science . Stakeholders were engaged in an iterative process to characterize the S outh Florida ecosystem and explore new perspectives for interdisciplinary applied research. The process resulted in an improved understanding of the biophysical, institutional, and human dimensions of the system . Scenario models were constructed from the knowledge innovations realized during stakeholder engagement. The improvements were made through collaborative lear ning between “experts” and “end users . ” Visualizations and mathematical representations helped to examine knowledge interfacing as a process of relating what we know to what we expect (preferences, uncertainty, risk). Before and after surveys quantified the usefulness of these representations in making management decisions. The methodologies included a structured systematic process to integrate science and management in decision making using stakeholder input. In Chapter 2, the objective to facilitate participation of resource managers was met within the constructs of people, process, and tools. Participatory decision support was applied to translate information from stakeholders’ minds into a format to design, construct, and evaluate decision support tools . A test bed of end users w as identified and maintained throughout the project period to obtain a clear picture of coral reef and marine resource manager climate information needs and the preferred delivery of that

PAGE 143

143 information. In this way, the group promoted a mutual understanding the ecosystem and what was needed to manage it through data indexing, integration, and visualizations. The desired outcome, to develop a common knowledge domain to access and understand scientific information for future management actions was obtained and used to design, construct, and evaluate decision support tools . In Chapter 3, the IEA framework facilitated stakeholder engagement and provided the science needed to develop a prototype coral reef BBN. The integrated ecosystem and conceptual ecological models were constructed using an iterative process. Multiple lines of communication were employed to design and refine the biophysical and human dimensions science characterizing the South Florida ecosystem. The products represented the first comprehensive synthesis of the marine and coastal environment located at the southern tip of the Florida peninsula. The synthesis, when coupled with the management agency perspectives clarified the requirements for developing a prototype BBN. In Chapter 4, the objective to construct and assess decision support tools was accomplished. Two BBNs were constructed and then evaluated by s takeholders. The methodology described the development of an ecosystem scale model and a habitat scale model for coral reef and hardbottom. The models were built using information realized during the needs assessment and the IEA. Analysis of the BBNs was completed with a test bed of stakeholders responsible for managing marine and coral reef resources in southeast Florida. The median response from s takeholder s indicated that the models improved their knowledge of climate impacts to reef ecosystems and that most would recommend use of the model to their colleagues.

PAGE 144

144 The research tested the utility of a participatory decision support for c omplex environmental planning. The project deliver ed a geographically transferable prototype to characterize the ecosystem and enhance stakeholder capacity and their understanding for improved management through participatory decision support . It also presented a unique opportunity for scientists and managers to participate in developing an applied prototype to address the unknowns tied to climate variability and ecological events within marine and coastal systems. Project success was dependent upon the willingness of stakeholders to explore their role in the transfer of information to knowledge and the application of knowledge innovations. This encompassed their ability to maintain interest in the process that includes unfamiliar terms, methodologies, and the potential for scrutiny of their research and/or management. Nevertheless, the project outcomes, partial or complete, i dentify the strengths and weaknesses of knowledge exchanges related to climate variability (events) and resource management in marine and coastal environments to advance future iterations of decision support systems using participatory decision support .

PAGE 145

145 LIST OF REFERENCES Adobe Systems, 2012. www.adobe.com Altman, I., Blakeslee, A.M., Osio, G.C., Rillahan, C.B., Teck, S.J., Meyer, J.J., Byers, J.E., Rosenberg, A.A., 2011. A practical approach to implementation of ecosystem based management: a case study using the Gulf of Maine marine ecosystem. Frontiers in Ecology and the Environment 9, 183189. Altschuld, J.W., Witkin, R. , 2000. From needs assessment to action: Transforming needs into solution strategies. Sage. Thousand Oaks, California. Armstrong, J.S., Denniston, W.B., Gordon, M.M. , 1975. The use of the decomposition principle in making judgments. Organizational Behavior and Human Performance, 14(2), 257263. Ault, J.S., Browder, J., Nuttle , W.K., 2013. Fish and shellfish. In Integrated Conceptual Ecosystem Model Development for the Southeast Florida Coastal Marine Ecosystem , W.K. Nuttle and P.J. Fletcher (eds.). NOAA Technical Memorandum, OARAOML 103 and NOSNCCOS163. Miami, Florida, pp. 5362. Ault, J.S., Se rafy, J.E., Diresta, D., Dandelski, J., 1997. Impacts of commercial fishing on key habitats within Biscayne National Park. Biscayne National P ark Report, Homestead, Florida. Ault, J.S., Smith, S.G., Browder, J., Nuttle, W., Franklin, E.C., Dinardo, G.T., B ohnsack, J.A., 2014. Indicators for Assessing the Ecological and Sustainability Dynamics of Southern Florida’s Coral Reef and Coastal Fisheries. Ecological Indicators . DOI: 10.1016/j.ecolind.2014.04.013 Baker, A. C., Glynn, P. W., Riegl, B. , 2008. Climate change and coral reef bleaching: An ecological assessment of longterm impacts, recovery trends and future outlook. Estuarine, Coastal and Shelf Science, 80(4), 435471. doi: http://dx.doi.org/10.1016/j.ecss.2008.09.003 Balcolm, N., Liffmann, M., Spranger, M., 2013. History and philosophy of sea grant extension: New York Sea Grant College Program. (second edition). NYSGI H 13001. Cornell. In Fundamentals of sea grant extension (2nd edition). 2nd ed., 9 15. Ithaca, NY: New York Sea Grant College Program, pp. 100. Balint, P.J., Stewart, R.E., Desai, A., Walters, L.C. , 2011. Managing wicked environmental problems. In Wicked environmental problems . Springer, pp. 207217. Bammer, G., Smith son, M. , 2009. Uncertainty and risk: Multidisciplinary perspectives. London, UK: CSIRO (Earthscan) , p p. 382.

PAGE 146

146 Batie, S. S. , 2008. Wicked problems and applied economics. American Journal of Agricultural Economics 90(5), 11761191. Batty, M., 2014. Great planning disasters. In Tewdwr Jones, M., Phelps, N., Freestone, R. (Eds.). The Planning Imagination: Peter Hall and the Study of Urban and Regional Planning. Routledge. Oxon, London, pp. 2839. Bayes, T. 1763. An essay towards solving a problem in the doctrine of chances. Phil. Trans. Roy. Soc. 53, 370418; 54, 296325. (Reprinted in Biometrika 45 (1958), 293315.) Bellwood, D., Hughes, T., Folke, C., Nystrm, M. , 2004. Confronting the coral reef crisis. N ature 429(6994), 827833. Boone, H. N., Boone, D. A. , 2012. Analyzing likert data. Journal of Extension, 50(2). Browder, J.A., Fourqurean, J., Lirman, D., Nuttle, W.K. , 2013. Benthic Habitat: Seagrasses. In Integrated Conceptual Ecosystem Model Development for the Southeast Florida Coastal Marine Ecosystem , W.K. Nuttle and P.J. Fletcher (eds.). NOAA Technical Memorandum, OAR AOML 103 and NOS NCCOS163. Miami, Florida, pp. 8493. Brown, V. A., Har ris, J., Russell, J., ( Eds. ) , 2010. Tackling wicked problems: through the transdisciplinary imagination. Earthscan, pp. 312. Cain, J., 2001. Planning improvements in natural resources management: guidelines for using Bayesian networks to support the planning and management of development programmes in the water sector and beyond. Centre for Ecology and Hydrology, Wallingfor d, UK, pp. 124. Cain, J., Batchelor, C., Waughray, D., 1999. Belief networks: a framework for the participatory development of natural resource management strategies. Environment, development and sustainability 1 (2), 123133. Cain, J.D., Jinapala, K., Maki n, I.W., Somaratna, P.G., Ariyaratna, B.R., Perera, L.R., 2003. Participatory decision support for agricultural management. A case study from Sri Lanka. Agricultural Systems , 76(2), 457482. Carpenter, K. E., Abrar, M., Aeby, G., Aronson, R. B., Banks, S., Bruckner, A., . . . DeVantier, L. , 2008. Onethird of reef building corals face elevated extinction risk from climate change and local impacts. Science 321(5888), 560563. Carsey, T.P. , 2013. Water Column. In Integrated Conceptual Ecosystem Model Developm ent for the Southeast Florida Coastal Marine Ecosystem , W.K. Nuttle and P.J. Fletcher (eds.). NOAA Technical Memorandum, OAR AOML 103 and NOSNCCOS163. Miami, Florida, 4152 pp.

PAGE 147

147 Castelletti, A., SonciniSessa, R. , 2007. Bayesian Networks and participatory modelling in water resource management. Environmental Modelling & Software, 22(8), 10751088. Chen, S. H., Pollino, C. A. , 2012. Good practice in Bayesian network m odelling. Environmental Modelling & Software, 37, 134 145. Chiappone, M., Dienes, H., Swans on, D.W., Miller, S.L., 2005. Impacts of lost fishing gear on coral reef sessile invertebrates in the Florida Keys National Marine Sanctuary. Biol. Conserv. 121, 221230. Christensen, N., Bartuska, A., Brown, J., Carpenter, S., D’Antonio, C., Francis, R., Franklin, J., MacMahon, J., Noss, R., Parsons, D., Peterson, C., Turner, M., Woodmansee, R., 1996. The report of the Ecological Society of America committee on the scientific basis for ecosystem management. Ecol. Appl. 6, 665– 691. Churchman, C. W. , 1967. Guest editorial: wicked problems. Manage. Sci. 14 (4), 141 – 142. Collier C, Ruzicka, R, Banks, K, Barbieri, L, Beal, J, Bingham, D, Bohnsack, J, Brooke, S, Craig, N, Dodge, R, Fisher, L, Gadbois, N, Gilliam, D, Gregg, L, Kellison, T, Kosmynin, V, Lapointe, B, McDevitt, E, Phipps, J, Poulos, N, Proni, J, Quinn, P, Riegl, B, Spieler, R, Walczak, J, Walker, B, Warrick, D. , 2008. The state of coral reef ecosystems of southeast Florida, in: Waddell, J.E., C larke, A.M. (eds), the state of coral reef ecosystems of the United States and Pacific Freely Associated States. NOAA Technical Memorandum NOS NCCOS 73, 161187, NOAA/NCCOS Center for Coastal Monitoring and Assessment's Biogeography Team, Silver Spring, Maryland, pp. 131159. Conklin, E. J., Weil, W. , 1997. Wic ked problems: Naming the pain in organizations. Available from: http://www.accelinnova.com/docs/wickedproblems.pdf (accessed 05.25.2014) Cook, G., Fletcher, P., Kelble, C., 2014. Towards marine ecosystem based managem ent in South Florida: Investigating the connections among ecosystem pressures, states, and services in a Complex Coastal System. Ecological Indicators Special Issue: South Florida Ecosystem Based Management. Cortner, H., Moote, M. A. , 1999. The politics of ecosystem management . Island Press. Location, pp. 179. Curtice, C., Dunn, D. C., Roberts, J. J., Carr, S. D., Halpin, P. N. , 2012. Why ecosystem based management may fail without changes to tool development and financing. Biosc ience, 62(5), 508515.

PAGE 148

148 Dennison, W.C., Lookingbill, T.R., Carruthers, T.J., Hawkey, J.M., Carter, S.L., 2007. An eyeopening approach to developing and communicating integrated environmental assessments. Frontiers in Ecology and the Environment 5, 307314. deGroot, R. S., Alkemade, R., Braat, L., Hein, L., Willemen, L. , 2010. Challenges in integrating the concept of ecosystem services and values in landscape planning, management and decision making. Ecological Complexity , 7 (3), 260272. deReynier, Y.L., Levin, P.S., Shoji, N.L., 2010. Bringing stakeholders, scientists, and managers together through an integrated ecosystem assessment process. Mar. Policy 34, 534540. Dodge, R., Aller R.C., 1974. Coral growth related to resuspension of bottom sediments. Natu re (London) 247, 574577. Donahue, S., Acosta, A., Akins, L., Ault, J., Bohnsack, J., Boyer, J., Callahan, M., Causey, B., Cox, C. and others. , 2008. The state of coral reef ecosystems of the Florida Keys, in: Waddell, J.E., Clarke, A.M. (eds), the state of coral reef ecosystems of the United States and Pacific Freely Associated States. NOAA Technical Memorandum NOS NCCOS 73. Silver Spring, Maryland, pp. 569. Doren, R.F., Trexler, J.C., Gottlieb, A.D., Harwell, M.C., 2009. Ecological indicators for syste m wide assessment of the greater everglades ecosystem restoration program. Ecol. Ind. 9, S2S16. Doren, R.F., Trexler, J.C., Gottlieb, A.D., Harwell, M.C., 2009. Ecological indicators for system wide assessment of the greater everglades ecosystem restorat ion program. Ecol. Ind. 9, S2S16. Edmondson, C.H., 1928. The ecology of a Hawaiian coral reef. Bulletine Bernice P Bishop Museum 164. Elmer F., Riegl, B., 2014. A discrete mathematical extension of conceptual ecological models: Application for the sout heast Florida Shelf. Ecological Indicators Special Issue: South Florida Ecosystem Based Management. EPA (Environmental Protection Agency), 1996. Final water quality protection program document for the FKNMS. Atlanta, GA: U.S. Environmental Protection Agenc y , pp. 167. EPA (Environmental Protection Agency) , 1999. Water quality concerns in the Florida Keys: Sources, effects, and solutions. No. EPA 904 R 99005. Atlanta, GA: US Environmental Protection Agency, pp. 65. Etling, A., 1993. What is nonformal educati on? Journal of Agricultural Education 34, 7276.

PAGE 149

149 Executive Order, 2010. Stewardship of the Ocean, Our Coasts, and the Great Lakes. Executive Office of the President. July 19, 2010. Available from: https://www.federalregister.gov/articles/2010/07/22/201018169/stewardship of theoceanour coasts andthegreat lakes (accessed 06.21.2014) Facebook Analytics, 2013. www.facebook.com . Farber, S., Costanza, R., Childers, D.L., Erickson, J., Gross, K., Grove, M., ... and others. , 2006. Linking ecology and economics for ecosystem management. Bioscience , 56(2), 121133. FDEP (Florid a Department of Environmental Protection) , 2014. Southeast district state parks. Vero Beach, Florida. Available from: http://www.floridastateparks.org/findapark/di strict southeast.cfm (accessed 05.25.2014) FKNMS (Florida Keys National Marine Sanctuary), 2002. Comprehensive Science Plan. Marathon, Florida, pp. 33. Available from: h ttp://floridakeys.noaa.gov/research_monitoring/fknms_science_plan.pdf (accessed 5.25.2014) Fletcher, P. Li, Y. Kiker, G., 2014 in preparation. Participatory decision support tools for marine and coastal resource management: stakeholder perceptions of using Bayesian Belief Networks to marine and coastal ecosystems in southeast Florida. Ecology and Society. Fletcher, P., Hendee, J. , 2012. Coral reef and marine resource Manager Climate information needs assessment for the Florida Reef Tract. NOAA Technical Me morandum OAR AOML 100. National Oceanic and Atmospheric Administration. Office of Oceanic and Atmospheric Research. Atlantic Oceanographic and Meteorological Laboratory. Miami, Florida, pp. 20. Fletcher, P., C.K. Kelble, W.K. Nuttle and G. Kiker. , 2014. Us ing integrated ecosystem assessments to transfer science to managers. Ecological Indicators Special Issue: South Florida Ecosystem Based Management. Fletcher, P., Spranger, M., Hendee, J., Li, Y., Clark, M. Kiker, G. , 2014 in preparation. Decision Tools for Coral Reef Managers: using participatory decision support to improve the understanding of climate impacts to coral reefs for informed decision making. Coral Reefs. Fox, H.E., Christian, C., Nordby, J.C., Pergams, O.R., Peterson, G.D., Pyke, C.R., 2006. P erceived barriers to integrating social science and conservation. Conserv. Biol. 20, 18171820. Gardner, T.A., Gill, J.A., Grant, A., Watkinson, A.R., Ct, I.M., 2005. Hurricanes and Caribbean coral reefs: Immediate impacts, recovery trajectories, and contribution to longterm decline. Ecology 86, 174– 184.

PAGE 150

150 Gentile, J., Harwell, M., Cropper Jr, W., Harwell, C., DeAngelis, D., Davis, S., Ogden, J., Lirman, D. , 2001. Ecological conceptual models: a framework and case study on ecosystem management for South Florida sustainability. Sci. Total Environ. 274, 231253. Goldston, D., 2009. Science, policy, and the US Congress. Cell 139, 647648. Google Analytics, 2013. http://www.google.com/analytics . Go osen, H., Janssen, R., Vermaat, J. E. , 2007. Decision support for participatory wetland decisionmaking. Ecological Engineering 30(2), 187199. GoToMeeting. 2014. www.gotomeeting.com . Gramer, L.J., Johns, E.M., H endee, J.C., Hu, C. , 2008. Characterization of biologically significant hydrodynamic anomalies on the Florida Reef Tract. Proceedings of the 11th International Coral Reef Symposium, Ft. Lauderdale, Florida, 711 July 2008 Session number 15. Harwell, M. a, Gentile, J.H., Cummins, K.W., Highsmith, R.C., Hilborn, R., McRoy, C.P ., Parrish, J., Weingartner, T., 2010. A Conceptual Model of Natural and Anthropogenic Drivers and Their Influence on the Prince William Sound, Alaska, Ecosystem. Hum. Ecol. Risk Assess. 16, 672– 726. Harwell, M., Long, J., Bartuska, A., Gentile, J., Harwell, C., Myers, V., Ogden, J., 1996. Ecosystem Management to Achieve Ecological Sustainability: The Case of South Florida. Environ. Manage. 20, 497– 521. Heckerman, D. , 1996. Bayesian networks for knowledge discovery. Advances in Knowledge Discovery and Data Mining, 11, 273305. HoeghGuldberg, O., Mumby, P.J., Hooten, A.J., Steneck, R.S., Greenfield, P., Gomez, E., and others. 2007. Coral reefs under rapid climate change and ocean acidifica tion. S cience, 318(5857), 17371742. Hungerford, H.R., Volk, T.L., 1990. Changing learner behavior through environmental education. J. Environ. Educ. 21, 822. Johns, G., Lee, D.J., Leeworthy, V.R., Boyer, J.N., Nuttle, W.K., 2014. Developing economic indices to assess the human dimensions of the South Florida coastal marine ecosystem services. Ecological Indicators Special Issue: South Florida Ecosystem Based Management. Johns, G.M., Leeworthy, V.R. Bell, F.W., Bonn, M.A.. 2001. Socioeconomic study of reefs in southeast Florida. Final Report to the Broward County Department of Planning and Environmental Protection, Fort Lauderdale, Florida, pp. 348. Available from: http://coastalsocioeconomics.noaa.gov/core/reefs/0201.pdf ) (Accessed 06.21.2014).

PAGE 151

151 Jokiel, P., Coles, S., 1977. Effects of temperature on the mortality and growth of Hawaiian reef corals. Marine Biology 43, 201208. Kelble, C.R., Loomis, D.K., Lovelace, S., Nuttle, W.K., Ortner, P.B., Fletcher, P., Cook, G.S., Lorenz, J.J., Boyer, J.N., 2013. Th e EBM DPSER Conceptual Model: Integrating Ecosystem Services into the DPSIR Framework. PloS one 8, e70766. Keller, B. D., & Causey, B. D. , 2005. Linkages between the Florida Keys National Marine Sanctuary and the South Florida ecosystem restoration initiative. Ocean & coastal management , 48(11), 869900. Keller, B. D., Donahue, S. , 2006. 200203 sanctuary science report: An ecosystem report card after five years of marine zoning. Marine Sanctuaries Conservation Series NMSP0612. Silver Spring, Maryland. U.S. Department of Commerce, National Oceanic and Atmospheric Administration, National Marine Sanctuary Program pp. 360. Keller, B. D., Gleason, D. F., McLeod, E., Woodley, C. M., Airam, S., Causey, B. D., . . . Miller, S. L. , 2009. Climate change, coral reef ecosystems, and management options for marine protected areas. Environmental Management 44(6), 10691088. Kelly (Letcher), R.A., Jakeman, A.J., Barreteau,O., Borsuk, M.E., Elsawah, S., Hamilton, S.H., Henriksen, H.J., ... and others., 2013. Selecting among five common modelling approaches for integrated environmental assessment and management. Environmental Modelling & Software, 47 , 159181. Kiker, G. A., Bridges, T. S., Varghese, A., Seager, T. P., Linkov, I. , 2005. Application of multicriteria decision analysis in environmental decision making. Integrated Environmental Assessment and Management 1 (2), 95108. Kiker, G. A., Muoz Carpena, R., Wolski, P., Cathey, A., Gaughan, A., & Kim, J. , 2008. Incorporating uncertainty into adaptive, transboundary wat er challenges: A conceptual design for the Okavango river basin. International Journal of Risk Assessment and Management 10(4), 312338. Kleis, J., Lang, L., Mietus, J.R., Tiapula, F., 1973. Toward a contextual definition of nonformal education. Nonformal education discussion papers, East Lansing, MI: Michigan State University, 36. Kleypas, J.A., Buddemeier, R.W., Archer, D., Gattuso, J., Langdon, C., Opdyke, B.N., 1999. Geochemical consequences of increased atmospheric carbon dioxide on coral reefs. Sci ence 284, 118120. Kruczynski, W.L., Fletcher, P.J. (eds)., 2012. Tropical Connections: South Florida’s marine environment. IAN Press, University of Maryland Center for Environment al Science, Cambridge, Maryland, pp. 492.

PAGE 152

152 Langdon, C., 2012. Ocean acidific ation may be a threat to the marine ecosystem. In Tropical Connections: South Florida's Marine Environment, Kruczynski, W. and Fletcher P., editors. IAN Press, Cambridge, MD, pp. 40 41. Leeworthy, V.R., and R. Ehler. , 2010a. Linking the economy and environment of the Florida Keys/Key West: Economic contribution of recreating visitors to the Florida Keys/Key West, 20072008. NOAA/ National Ocean Service, Silver Spring, MD, pp. 19. Available from: http://sanctuaries.noaa.gov/ science/socioeconomic/floridakey s/pdfs/economic08.pdf (assessed 6.21.2014) Leeworthy, V.R., and R. Ehler. 2010b. Linking the economy and environment of the Florida Keys/Key West: Importance and satisfaction ratings by recreating visitors to the Florida Keys/Key West, 20072008. NOAA/Nati onal Ocean Service, Silver Spring, MD, pp. 20. Available from: http://sanctuaries.noaa. gov/science/socioeconomic/floridakeys/pdfs/importance08.pdf (assessed 6.21.2014) Leslie, H.M., McLeod, K.L., 2007. Confronting the challenges of implementing marine ec osystem based management. Frontiers in Ecology and the Environment 5, 540548. Levin, P.S., Fogarty, M.J., Murawski, S.A., Fluharty, D., 2009. Integrated ecosystem assessments: developing the scientific basis for ecosystem based management of the ocean. P LoS Biology 7, e1000014. Levin, P.S., Kelble, C.K., Shuford, E., Ainsworth, C., deReynier, Y.L., Dunsmore, R., Fogarty, M.J., Holsman, K., Howell, E., Monaco, M., Oakes, S., Werner, F., 2013. Guidance for implementation of integrated ecosystem assessments : a US perspective. ICES Journal of Marine Science doi:10.1093/icesjms/fst112. Li kens, G.E., 2010. The role of science in decision making: does evidencebased science drive environmental policy? Frontiers in Ecology and the Environment, e1e9. Lirman, D. , Formel, N., Schopmeyer, S., Ault, J., Smith, S., Gilliam, D., Riegl, B., 2013. Percent recent mortality (PRM) of stony corals as an ecological indicator of coral reef condition. Ecological Indicators Special Issue: South Florida Ecosystem Based Management. Lirman, D., Gracias, N., Gintert, B., Gleason, A., Deangelo, G., Dick, M., Martinez, E., Reid, P.R., 2010. Damage and recovery assessment of vessel grounding injuries on coral reef habitats using georeferenced landscape video mosaics. Limnology and Oceanography 88 97. Loomis, D.K., Ortner, P.B., Kelble, C.R., Paterson, S.K., 2014. Developing integrated ecosystem indices. Ecological Indicators Special Issue: South Florida Ecosystem Based Management.

PAGE 153

153 Loomis, D.K., Paterson, S.K., 2014. Human dimensions indicators of coastal ecosystem services: A hierarchical perspective. Ecological Indicators Special Issue: South Florida Ecosystem Based Management. Loredo, T. J., 1990. From laplace to supernova SN 1987A: Bayesian inference in astrophysics. Maximum entropy and bayesian methods . Springer, pp. 81142. Lorenz, J. . 2013. Shoreline Habitat: Mangroves. In Integrated Conceptual Ecosystem Model Development for the Southeast Florida Coastal Marine Ecosystem , W.K. Nuttle and P.J. Fletcher (eds.). NOAA Technical Memorandum, OAR AOML 103 and NOS NCCOS163. Miami, Florida, pp. 109119. Lynam, T., Bousquet, F., Le Page, C., d'Aquino, P., Barreteau, O., Chinembiri, F., Mombeshora, B., 2003. Adapting science to adaptive managers: spidergrams, belief models, and multi agent s ystems modeling. Integrated natural resource management [electronic resource]: linking productivity, the environment, and development, 157. Lynam, T., de Jong, W., Shei l, D., Kusumanto, T., Evans, K., 2007. A review of tools for incorporating community knowledge, preferences, and values into decision making in natural resources management. Ecology and Society 12(1), 5. Lynam, T., Drewr y, J., Higham, W., Mitchell, C., 2010. Adaptive modelling for adaptive water quality management in the Great Barrier Reef r egion, Australia. Environmental Modelling & Software 25(11), 12911301. doi: http://dx.doi.org/10.1016/j.envsoft.2009.09.013 Madden, C. J., Rudnick, D. T., McD onald, A. A., Cunniff, K. M., Fo urqurean, J. W. , 2009. Ecological indicators for assessing and communicating seagrass status and trends in Florida Bay. Ecological Indicators 9 (6), S68S82. Marcot, B.G., Holthausen, R.S., Raphael, M.G., Rowland, M.M., Wisdom, M.J., 2001. Using b ayesian belief networks to evaluate fish and wildlife population viability under land management alternatives from an environmental impact statement. Forest Ecology and Management, 153(1), 2942. Marshall, F.E., Banks, K., 2013. Shoreline Habitat: Beaches. In Integrated Conceptual Ecosystem Model Development for the Southeast Florida Coastal Marine Ecosystem , W.K. Nuttle and P.J. Fletcher (eds.). NOAA Technical Memorandum, OARAOML 103 and NOS NCCOS163. Miami, Florida, pp. 94108. Marshall, F.E., Banks, K., Cook, G.S. 2014. Ecosystem indicators for southeast Florida beaches. Ecological Indicators Special Issue: South Florida Ecosystem Based Management. Marshall, P.A., Shuttenberg, H.Z. , 2006. A Reef Manager's Guide to Coral Bleaching . Great Barrier Reef Mar ine Park Authority, Townsville, Qld, Australia , pp. 163.

PAGE 154

154 Martin, T.G., Burgman, M.A., Fidler, F., Kuhnert, P.M., LOW CHOY, S., McBride, M., Mengersen, K. , 2012. Eliciting expert knowledge in conservation science. Conservation Biology , 26(1), 29 38. Mayor, A.G., 1914. The Effects of Temperature upon Tropical Marine Animals. Carnegie Institution of Washington. McCann, R. K., Marcot, B.G., Ellis, R., 2006. Bayesian belief networks: applications in ecology and natural resource management. Canadian Journal of Forest Research, 36(12), 30533062. McLeod KL, Lubchenco J, Palumbi SR, and Rosenberg AA. , 2005. Communication Partner ship for Science and the Sea scientific consensus statement on marine ecosystem based management. Available from: http://www.compassonline.org/sites/all/files/document_files/EBM_Consensus_St atement_v12.pdf. (Accessed 06.21.2014) MEA (M illennium Ecosystem Assessment), 2005. Ecosystems and Human Well Being: Our Human Planet: Summary for Decision Makers. Synthesis Report, 1155. Mitchell, C. , 2011. Relevance of MARES to Agency and Organizational Participants. MARES Whitepaper. Version 6 June 2011. Available f rom: http://www.sofla mares.org/docs/MARES_WhitePaper3_AgencyPerspectives_20110606.pdf (accessed 06.21.2014) NOAA (National Oceanic and Atmospheric Administr ation), 1996. Florida Keys National Marine Sanctuary management plan/environmental impact statement, vol II. Silver Spring, Maryland. U.S. Dept. of Commerce, National Oceanic and Atmospheric Administration, National Ocean Service, pp. 251. NOAA (National O ceanic and Atmospheric Administration), 2007. Florida Keys National Marine Sanctuary revised management plan. Silver Spring, MD.: US Department of Commerce, National Oceanic and Atmospheric Administration, National Ocean Service, pp. 382. NOAA (National Oc eanic and Atmospheric Administration), 2009. NOAA coral reef conservation program goals & objectives 20102015. Silver Spring, MD: US Department of Commerce. National Oceanic and Atmospheric Administration. Cora l Reef Conservation Program, pp. 40. NOAA/CSC (National Oceanic and Atmospheric Administ ration/Coastal Services Center), 2003. Product design and evaluation. Course training manual. Updated periodically. Available from: csc.noaa.gov/training/ (accessed 05.25.2014) NOAA/CSC (National Oceanic and Atmos pheric Administ ration/Coastal Services Center), 2007. Coastal Community Planning and Development. Course Training Manual. Charleston, SC. URL: http://.csc.noaa.gov/training/ (accessed 04.22.13)

PAGE 155

155 NOAA/CSC (National Oceanic and Atmospheric Administ ration/Coastal Services Center), 2009. Planning for meaningful evaluation. Updated periodically. Available from: csc.noaa.gov/training/ (accessed 05.25.2014) Norsys, 2013. www.norsys.com . Nuttle, W. K., Fletcher, P., 2013a. Integrated conceptual ecosystem model for the Florida Keys/Dry Tortugas coastal marine ecosystem. NOAA technical memorandum OAR AOML 101. Miami, Florida, pp. 91. Nuttle, W. K., Fletcher, P., 2013b. In tegrated conceptual ecosystem model for the southwest Florida shelf coastal marine ecosystem. NOAA technical memorandum OAR AOML 102. Miami, Florida, pp. 1 08. Nuttle, W. K., Fletcher, P., 2013c. Integrated conceptual ecosystem model for the southeast Flori da coast marine ecosystem. NOAA technical memorandum OAR AOML 103. Miami, Florida, pp. 125. Nuttle, W.K., Bergh, C. , 2013. MarineDependent People. In Integrated Conceptual Ecosystem Model Development for the Southeast Florida Coastal Marine Ecosystem , W.K . Nuttle and P.J. Fletcher (eds.). NOAA Technical Memorandum, OARAOML 103 and NOSNCCOS163. Miami, Florida, pp. 120125. Nyberg, J.B., Marcot, B.G., Sulyma, R., 2006. Using Bayesian belief networks in adaptive management. Canadian Journal of Forest Research 36, 31043116. O E C D (Organisation for Economic Development and Cooperation), 1993. Core set of indicators for environmental performance reviews. Paris, France, pp. 83. Ogden, J.C. Bal dwin, J.D. Bass, O.L., Browder and others, 2014. Waterbirds as indi cators of Ecosystem Health in the Coastal Marine habitats of southern Florida 1: Conceptual Ecological Models. Ecological Indicators Special Issue: South Florida Ecosystem Based Management. Ogden, J.C. Baldwin, J.D. Bass, O.L., Browder and others, 2014. Wa terbirds as indicators of Ecosystem Health in the Coastal Marine habitats of southern Florida 2: Conceptual Ecological Models. Ecological Indicators Special Issue: South Florida Ecosystem Based Management. ONMS (Office of National Marine Sanctuaries), 2011. Florida Keys National Marine Sanctuary condition report 2011. Silver Spring, Maryland. U.S. Department of Commerce, National Oceanic and Atmospheric Administration, Office of National Marine Sanctuaries, pp. 105.

PAGE 156

156 Oregon Sea Grant. , 2000. A mandate to engage coastal users: A review of the National Sea Grant College Extension Program and a call for greater national commitment to engagement. Corvallis, Oregon. ORESU Q 01001. Available from: http://seagrant.oregonstate.edu/sgpubs/onlinepubs/q01001.html (accessed 6.10.2014) Oregon Sea Grant. , 2003. Implementing the National Oceanic and Atmospheric Administration’s mandate to engage coastal users: Opportunities for National Sea Grant outreach growth. Corvallis, Oregon. ORESU Q 03002. Available from: http://seagrant.oregonstate.edu/sgpubs/onlinepubs/q03002.pdf (accessed 6.10.2014) Pearl, J. and Russell. , 2000. Bayesian networks. Department of Statistics, UCLA . Available from: http://escholarship.org/uc/item/53n4f34m (accessed 06.21.2014) Pimbert, M. P., Pretty, J. N., 1997. Parks, people and professionals: Putting participation into protected area management. Social change and conservation 297330. POC (Pew Oceans Commission), 2003. America’s Living Oceans: Charting a Course for Sea Change. Summary Report. Location, pp. 40. Pomeroy, R., Douvere, F. , 2008. The engagement of stakeholders in the marine spatial planning process. Marine Policy 32 (5), 816822. Ralph, S.C., Poole, G.C., 2003. Putting monitoring first: designing accountable ecosystem restoration and management plans. Restoration of Puget Sound R ivers.D Montgomery, S Bolton, D Booth, L Wall (eds.) University of Washington Press, Seattle, 226247. Ramirez, R. , 1999. Stakeholder analysis and conflict management. In Buckels, D (ed) Cultivating Peace: Conflict and Collaboration in Natural Resource Management The World Bank Institute, Washington, D.C. , pp. 101126. Raymond, C. M., Cleary, J., 2013. A tool and process that facilitate community capacity building and social learning for natural resource management. Ecology & Society 18(1) . RECOVER (Resto ration Coordination Verification Program), 2012. System Status Report Interim Update. US Army Corps of Engineers, Jacksonville, Florida and South Florida Water Management District, West Palm Beach, Florida. Available from: http://www.evergladesplan.org/pm/ssr_2012/ssr_2012_pdfs/2012_ssr_full_web.p df (accessed 6.10.2014) Reiter, M. A., 2004. A simple fuzzy logic approach to analyze integrated interdisciplinary “four component” conceptual resource management models. Interdisc . Environ. Rev. 6(2):1 13.

PAGE 157

157 Reiter, M. a., Matlock, G.C., Gentile, J.H., Harwell, M. A ., Kelty, R., Barko, J.W., Baker, S., Scott, G., 2013. An integrated framework for informing coastal and marine ecosystem management decisions. J. Environ. Assess. Policy Manag. 15. Riegl, B., Gilliam, D., Lirman, D., 2013. Coral reefs and hardbottom habitats. In Integrated Conceptual Ecosystem Model Development for the Southeast Florida Coastal Marine Ecosystem , W .K. Nuttle and P.J. Fletcher (eds.). NOAA Technical Memorandum, OAR AOML 103 and NOS NCCOS163. Miami, Florida, pp. 6383. Rittel, H. W., Webber, M. M. , 1973. Dilemmas in a general theory of planning. Policy Sciences 4 (2), 155169. Rosenberg, A.A., McLeod, K.L., 2005. Implementing ecosystem based approaches to management for the conservation of ecosystem services. Marine Ecology Progress Series, 270274. Senge P. , 1990. The fifth discipline: the art and practice of the learning organization. Dou bleday, N ew York, New York, USA, pp. 423. SFRCCC (Southeast Florida Regional Climate Change Compact Technical Ad hoc Work Group), 2011. A Unified sea level rise projection for southeast Florida. A document prepared for the Southeast Florida Climate Change Compact S teering Committee. 24 July 2013, pp. 27. Sheppard, S. R. , 2005. Landscape visualisation and climate change: The potential for influencing perceptions and behaviour. Environmental Science & Policy 8 (6), 637654. Sheridan, S. C., Pirhalla, D. E., Lee, C. C. , Ransibrahmanakul, V. , 2013. Evaluating linkages of weather patterns and water quality responses in south Florida using a synoptic climatological approach. Journal of Applied Meteorology & Climatology 52(2). Shinn, E., 2012. Corals are a potential tool in measuring climate change. In Tropical Connections: South Florida's Marine Environment, Kruczynski, W. and Fletcher P., editors. IAN Press, Cambridge, Maryland, pp. 213215. Slocombe, D. S. , 1993a. Environmental planning, ecosystem science, and ecosystem approaches for integrating environment and development. Environmental management , 17(3), 289303. Slocombe, D.S., 1993b . Implementing ecosystem based management. Bioscience 43, 612622. Slocum, R., L. Wichart, D. Rocheleau, and B. Thomas Slayter (Eds), 1995. Power, process and participation: tools for change. Intermediate Tec hnology Publications. Exter, UK, pp. 251.

PAGE 158

158 Spiegelhalter, D. J., Dawid, A. P., Lauritzen, S. L., Cowell, R. G. , 1993. Bayesian analysis in expert systems. Statistical Science, 8 (3), 2192 47. Szaro, R.C., Sexton, W.T., Malone, C.R., 1998. The emergence of ecosystem management as a tool for meeting people's needs and sustaining ecosystems. Landscape Urban Plann. 40, 17. Tallis, H., Levin, P.S., Ruckelshaus, M., Lester, S.E., McLeod, K.L., Fluharty, D.L., Halpern, B.S., 2010. The many faces of ecosystem based management: making the process work today in real places. Mar. Policy 34, 340348. Tscherning, K., Helming, K., Krippner, B., Sieber, S., 2012. Does research applying the DPSIR framewo rk support decision making? Land Use Policy 29, 102110. Turning Technologies, 2013. www.turningtechnologies.com . USCOP ( US Commission on Ocean Policy), 2004. An ocean blueprint for the 21st century. Final Report. ISBN#0 9759462O X., 1 676. URL: http://govinfo.library.unt.edu/oceancommission/documents/full_color_rpt/000_oce an_full_report.pdf (accessed 08.26.13) Uusitalo, L., 2007. Advantages and challenges of Bayesian networks in environmental modelling. Ecol. Model. 203, 312318. van Dam, A.A., Kipkemboi, J., Rahman, M., Gettel, G.M., 2013. Linking Hydrology, Ecosystem Function, and Livelihood Outcomes in African Papyrus Wetlands Using a Bayesian Network Model. Wetlands 117. van de Kerkhof, M. , 2001. A survey on the methodology of participatory integrated assessment. IIASA Working paper IR 01014, Laxenburg, Austria. van Hooidonk, R., Huber, M., 2012. Effects of modeled tropical sea surface temperature variability on coral reef bleaching predictions. Coral reefs , 31(1), 121131. van Hooidonk, R., Maynard, J. A., Manzello, D., Planes, S. , 2014. Opposite latitudinal gradients in pr ojected ocean acidification and bleaching impacts on coral reefs. Global Change Biology, 20 (1), 103112. Van Oppen, M. J., Lough, J. M. (Eds.), 2008. Coral Bleaching. Springer Verlag . Heidelberg, Germany , pp. 192. Volety, A, Haynes, L., Goodman, P., 2014. Ecological Condition and Value of Oyster Reefs of the Southwest Florida Shelf Ecosystem. Ecological Indicators Wagner, D. E., Kramer, P., Van Woesik, R. , 2010. Species composition, habitat, and water quality influence coral bleaching in southern Florida. Marine Ecology Progress Series 408, 6578.

PAGE 159

159 Walker, B., Carpenter, S., Anderies, J., Abel, N., Cumming, G., Janssen, M., Lebel, L., Norberg, J, Peterson, G.D., Pritchard, R., 2002. Resilience management in social ecological systems: a working hypothesis for a participatory approach. Conservation ecology , 6 (1), 14. Walker, B. K., Riegl, B., Dodge, R. E. , 2008. Mapping coral reef habitats in southeast Florida using a combined technique approach. Journal of Coastal Research, 11381150. Walters, C. J., 2007. Is adaptive management helping to solve fisheries problems?. AMBIO: A Journal of the Human Environment , 36(4), 304307. Weichselgartner, J., Kasperson, R., 2010. Barriers in the sciencepolicy practice interface: Toward a knowledgeactionsystem in global envir onmental change research. Global Environ. Change 20, 266277. Westmacott, S. , 2001. Developing decision support systems for integrated coastal management in the tropics: Is the ICM decisionmaking environment too complex for the development of a useable and useful DSS? Journal of Environmental Management, 62(1), 55 74. WHCEQ (The White House Council on Environmental Quality), 2010. Final recommendations on the Interagency Ocean Policy Task Force. URL: www.whitehouse.gov/files/documents/OPTFFinalRecs.pdf (accessed 07.24.13) Wilkinson, C. , 2000. Status of coral reefs of the world: 2004. Chapter 18, 473491. Wingard, L., Lorenz, J., 2014. Integrated Conceptual Ecological Mo del and Habitat Indices for the Southwest Florida Coastal Wetlands. Ecological Indicators Special Issue: South Florida Ecosystem Based Management. Witkin, B. R. , Altschuld, J.W., 1995. Planning and conducting needs assessments: A practical guide. Sage. Thousand Oaks, California, pp. 302. Xiang, W., 2013. Working with wicked problems in socioecological systems: Awareness, acceptance, and adaptation. Landscape Urban Plann. 110, 14.

PAGE 160

160 BIOGRAPHICAL SKETCH Pamela Fletcher is a University of Florida/ Florida S ea Grant College Program liaison to the National Oceanic and Atmospheric Administration’s (NOAA) Atlantic Oceanographic and Meteorological Laboratory in Miami, Florida. She is currently involved in projects centered on translating and delivering NOAA research to managers , decision makers , and interested stakeholders . Her efforts are focused on improving the understanding of marine ecosystems for informed decision making with an emphasis on human dimensions science. She has conducted field research, implemented policy and management actions, and is committed to translating science in a useable format for informal and formal education. She is a contributor to and an editor of “Tropical Connections” a book about South Florida’s marine environment published in 2 012. She has led or contributed to a series of NOAA Technical Memorandums and white papers, science communications, and peer reviewed journals. She holds a B achelor of A rts in Marine Affairs and B achelor of S cience in Fisheries Science and Technology from the University of Rhode Island, a M aster of S cience in Environmental Management from Rensselaer Polytechnic Institute , and a Doctorate of Philosophy in Soil and Water Science from the University of Florida.