Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Coastal green infrastructure as a sea level rise adaptation measure: assessing environmental, local,… Conger, Tugce 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2019_february_conger_tugce.pdf [ 27.61MB ]
JSON: 24-1.0375779.json
JSON-LD: 24-1.0375779-ld.json
RDF/XML (Pretty): 24-1.0375779-rdf.xml
RDF/JSON: 24-1.0375779-rdf.json
Turtle: 24-1.0375779-turtle.txt
N-Triples: 24-1.0375779-rdf-ntriples.txt
Original Record: 24-1.0375779-source.json
Full Text

Full Text

COASTAL GREEN INFRASTRUCTUREAS A SEA LEVEL RISE ADAPTATION MEASURE:ASSESSING ENVIRONMENTAL, LOCAL ANDINSTITUTIONAL CONTEXTSbyTugce CongerB.S., Gazi University, 2008M.URP., University of Colorado Denver 2011A THESIS SUBMITTED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinThe Faculty of Graduate and Postdoctoral Studies(Resource Management and Environmental Studies)The University of British Columbia(Vancouver)December 2018c Tugce Conger 2018The following individuals certify that they have read, and recommend to the Faculty of Graduateand Postdoctoral Studies for acceptance, the dissertation entitled:Coastal green infrastructure as a sea level rise adaptation measure: Assessing environmentallocal and institutional contextssubmitted by Tugce Conger in partial fulfillment of the requirements forthe degree of Doctor of Philosophyin Resource Management and Environmental StudiesExamining Committee:Stephanie Chang, SCARP & IRESSupervisorMaged Senbel, SCARPSupervisory Committee MemberMark Johnson, IRES & EOASSupervisory Committee MemberJordi Honey-Roses, SCARPUniversity ExaminerCynthia Girling, SALAUniversity ExamineriiAbstractWith the acceleration of climate change impacts, adaptation is no longer a matter of choice formost communities. There has been a growing interest in coastal green infrastructure (CGI), nat-ural and nature-based adaptation measures, due to its role in flood and erosion protection, andprovision of multiple environmental, social and economic benefits. However, there remains agap in understanding the context-dependency of CGI as an adaptation measure. In this disserta-tion, I empirically investigate the environmental, local and institutional contexts in which CGIcan be used as a sea level rise adaptation measure through three distinct studies, focusing on thecoastal regions of British Columbia (BC) and Washington State (WA).First, I conduct a regional study. Using climatic and environment indicators, I investigate whereCGI has the highest coastal protection potential while taking into account its vulnerability. Iconclude that CGI in the large population centers in BC and most of the communities in WAmay not provide high coastal protection benefits, where CGI in the smaller communities havea higher potential. Second, I undertake a local study in BC, investigating community trade-offs between CGI and other adaptation strategies. I incorporate local perspectives to developadaptation scenarios and create an evaluation framework using the literature and expert inputs.Applying the framework to the scenarios, I conclude that there are important trade-offs betweenthe local implications of different strategies. I find that the CGI scenario had the highest positiveimpacts, but displayed institutional drawbacks compared to others. Third, I undertake an insti-tutional study comparing the barriers to and facilitators of CGI implementation in BC and WA.I conclude that besides barriers and facilitators common to adaptation, factors specific to CGI,such as coastal jurisdiction and ownership; financial variation and flexibility; vision; organiza-tion efficiency and access to resources; partnerships and collaborations; NGOs; and communityadvocacy are also influential.Ultimately, this dissertation concludes that CGI’s context dependency influence its potentialbenefits, its applicability as a local adaptation measure, and implementation within the existingiiiinstitutional arrangements. Considering contextual factors can support more successful imple-mentation of CGI, and therefore can increase adaptation to sea level rise.ivLay SummaryAcceleration of climate change necessitates communities to adapt to sea level rise (SLR). Coastalgreen infrastructure (CGI) has the potential to help communities adapt to SLR and to increaseoverall community resilience. However, little is known about in which conditions CGI can beused as a meaningful adaptation action. Therefore, I investigated the environmental, local, andinstitutional contexts of CGI to provide insights to support the wider implementation of CGIand to increase adaptation to sea level rise. Using three distinct studies in the coastal regionsof British Columbia (BC) and Washington State (WA), I concluded that CGI (1) has differentcoastal protection benefits in different environments, (2) has positive trade-offs compared toother adaptation actions but it not well established in the current regulatory frameworks, and (3)is influenced by different institutional barriers and facilitators in BC than in WA.vPrefaceThis dissertation is my original and independent work. I identified the research questions, de-veloped the research design, defined the methodologies, collected and analyzed the data, andwrote the manuscripts for each chapter. My supervisor and advisory committee (Drs. StephanieChang, Maged Senbel, Mark Johnson) provided guidance and feedback at each step. My super-visor also provided edits on the manuscripts.All empirical chapters of this dissertation are stand alone papers, intended for journal publica-tion, which results in some repetition of the content, particularly the literature review/backgroundsections, across the chapters.Chapter 2 was conducted with support from the Pacific Institute for Climate Solutions (PICS)and the Marine Environmental Observation, Prediction and Response Network (MEOPAR). Aversion of Chapter 2 has been submitted for publication with Stephanie Chang as co-author.I collected and analyzed the data and wrote the manuscript. Dr. Chang provided guidance,comments and edits.Chapter 3 was conducted with support from the Pacific Institute for Climate Solutions (PICS)and the UBC Public Scholar’s Initiative (PSI). A version of Chapter 3 has been submitted forpublication with Stephanie Chang as co-author. The research conducted in Chapter 3 was ap-proved by the Behavioral Research Ethics Board (ethics certificate number H16-01408). I col-lected and analyzed the data and wrote the manuscript. Dr. Chang provided guidance, commentsand edits.Chapter 4 was conducted with support from the University of British Columbia (UBC) PublicScholar’s Initiative. A version of Chapter 4 will be submitted for publication with StephanieChang as co-author. The research conducted in Chapter 4 was approved by the BehavioralResearch Ethics Board (ethics certificate number H17-03556). I collected and analyzed the dataand wrote the manuscript. Dr. Chang provided guidance, comments and edits.viTable of ContentsAbstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiiLay Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viTable of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viiList of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiList of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiiiList of Abbreviations and Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . xviAcknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xviiiDedication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xx1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Rationale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Goals of the dissertation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31.3 Situating the research in the literature . . . . . . . . . . . . . . . . . . . . . . 41.3.1 Sea level change: causes, impacts, and projections . . . . . . . . . . . 41.3.2 Sea level rise adaptation and adaptation strategies . . . . . . . . . . . . 71.3.3 Protect strategy measures . . . . . . . . . . . . . . . . . . . . . . . . . Traditional Hard Structures . . . . . . . . . . . . . . . . . . Coastal Green Infrastructure . . . . . . . . . . . . . . . . . . 111.3.4 CGI in the adaptation context . . . . . . . . . . . . . . . . . . . . . . 151.3.5 The context-dependency of CGI . . . . . . . . . . . . . . . . . . . . . 161.4 Overarching and subsequent research questions . . . . . . . . . . . . . . . . . 18vii1.5 Research design, data, and methods . . . . . . . . . . . . . . . . . . . . . . . 191.6 Dissertation synopsis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 Developing indicators to identify coastal green infrastructure potential: incorpo-rating coastal protection benefits and vulnerability to changing environmental con-ditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262.2 Coastal green infrastructure (CGI) . . . . . . . . . . . . . . . . . . . . . . . . 282.2.1 CGI coastal protection benefits . . . . . . . . . . . . . . . . . . . . . . 302.2.2 CGI vulnerability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322.3 Study Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342.4 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362.4.1 Literature review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372.4.2 Content analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372.4.3 Development of the CGI indices . . . . . . . . . . . . . . . . . . . . . 392.4.4 Synthesizing the indices . . . . . . . . . . . . . . . . . . . . . . . . . 422.5 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422.5.1 CGI coastal protection index . . . . . . . . . . . . . . . . . . . . . . . 422.5.2 CGI vulnerability index . . . . . . . . . . . . . . . . . . . . . . . . . 462.5.3 Synthesizing the indices . . . . . . . . . . . . . . . . . . . . . . . . . 502.6 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 532.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563 Resilience-based evaluation of the local trade-offs between coastal green infrastruc-ture and other sea level rise adaptation strategies . . . . . . . . . . . . . . . . . . 583.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 583.2 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 603.2.1 Considering the resilience perspective in adaptation . . . . . . . . . . . 603.2.2 Sea level rise adaptation strategies . . . . . . . . . . . . . . . . . . . . 613.2.3 Adaptation strategies evaluation methods and concepts . . . . . . . . . 633.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 643.3.1 The study area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 663.3.2 Step 1 - The expert meetings and participatory workshop . . . . . . . . 683.3.3 Step 2 - Sea level rise adaptation strategies evaluation framework . . . 70viii3.3.4 Step 3 - Application of the evaluation framework to sea level rise adap-tation strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 733.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 733.4.1 The expert meetings and participatory workshop . . . . . . . . . . . . 733.4.1.1 The first expert meeting . . . . . . . . . . . . . . . . . . . . 733.4.1.2 The participatory workshop . . . . . . . . . . . . . . . . . . 743.4.1.3 The second expert meeting . . . . . . . . . . . . . . . . . . 763.4.2 The final evaluation framework . . . . . . . . . . . . . . . . . . . . . 803.4.3 Application of the evaluation framework to the adaptation strategies . . 903.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 923.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 974 Institutional barriers to and facilitators of coastal green infrastructure implemen-tation: A comparative study in British Columbia and Washington State . . . . . 994.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 994.2 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1014.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1064.3.1 Study Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1074.3.2 Research activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1104.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1134.4.1 CGI project review . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1134.4.1.1 Summary of the CGI project review results . . . . . . . . . . 1174.4.2 Review and synthesis of the institutional arrangements . . . . . . . . . 1184.4.2.1 Governance systems and the corresponding authority distri-bution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1184.4.2.2 Coastal jurisdiction and ownership . . . . . . . . . . . . . . 1214.4.2.3 Coastal and environmental regulations and programs . . . . . 1254.4.2.4 Summary of the review and synthesis of the institutional ar-rangements results . . . . . . . . . . . . . . . . . . . . . . . 1314.4.3 Semi-structured interviews . . . . . . . . . . . . . . . . . . . . . . . . 1324.4.3.1 Summary of the semi-structured interview results . . . . . . 1474.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1484.5.1 Institutional barriers and facilitators . . . . . . . . . . . . . . . . . . . 1504.5.2 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1574.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158ix5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1605.1 Summary of findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1605.2 Broad conclusions and policy implications . . . . . . . . . . . . . . . . . . . . 1645.3 Strengths and limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1665.4 Future research directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1695.5 Reflections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212A Appendix to Chapter 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212A.1 CGI coastal protection index . . . . . . . . . . . . . . . . . . . . . . . . . . . 213A.2 CGI vulnerability index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217B Appendix to Chapter 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221B.1 First expert meeting feedback form . . . . . . . . . . . . . . . . . . . . . . . . 222B.2 Pre-workshop survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228B.3 Post-workshop survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234B.4 Sea level rise adaptation strategies evaluation framework . . . . . . . . . . . . 239B.5 Sea level rise adaptation strategies evaluation framework spreadsheet . . . . . . 253B.6 The evaluation framework application to four sea level rise adaptation scenarios 254C Appendix to Chapter 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256C.1 The data sources for the CGI project reviews . . . . . . . . . . . . . . . . . . . 257C.2 The protocole for the semi-structured interviews . . . . . . . . . . . . . . . . . 258C.3 The coastal and environmental regulations in Canada and BC . . . . . . . . . . 261C.4 The coastal and environmental regulations in the United States and WA . . . . 262C.5 The coastal and environmental programs in Canada . . . . . . . . . . . . . . . 264C.6 The coastal and environmental programs in BC . . . . . . . . . . . . . . . . . 266C.7 The coastal and environmental programs in the United States . . . . . . . . . . 267C.8 The coastal and environmental programs in WA . . . . . . . . . . . . . . . . . 269xList of Tables1.1 Types of coastal protection measures from THSs to CGIs, adapted from (Morriset al., 2018; Sutton-Grier et al., 2018; Borsje et al., 2017; Narayan et al., 2016;Bridges et al., 2015) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102.1 CGI themes and indicators, and their data sources . . . . . . . . . . . . . . . . 402.2 CGI coastal protection potential matrix . . . . . . . . . . . . . . . . . . . . . . 422.3 CGI coastal protection index . . . . . . . . . . . . . . . . . . . . . . . . . . . 432.4 CGI Vulnerability Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 473.1 Summary of changes in the evaluation framework components. . . . . . . . . . 723.2 Wave effects zone estimates for the selected strategies. . . . . . . . . . . . . . 763.3 The final evaluation framework modules and components. The three-point (+1/0/-1) scoring system and the evaluation criteria for each components can be foundin Appendix B.4. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 824.1 The institutions of the semi-structure interview participants . . . . . . . . . . . 132A.1 CGI coastal protection index indicator values, scores and classesLowest score: 4.74, Highest score: 82.16Very low: 4.74 - 23.91; low: 23.91- 36.16; medium: 36.16 - 51.55; high: 51.55- 82.16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216A.2 CGI vulnerability index indicator values, scores and classesLowest score: 3.21, highest score: 26.35Very low: 3.21 - 8.85; low: 8.85- 10.48; medium: 10.48 - 14.43; high: 14.43 -26.35 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220C.1 The CGI project data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . 257C.2 The regulatory actions that may be required for CGI projects in BC . . . . . . . 261xiC.3 The regulatory actions that may be required for CGI projects in WA . . . . . . 263C.4 The federal coastal and environmental programs in Canada. . . . . . . . . . . . 265C.5 The provincial coastal and environemtal programs in BC. . . . . . . . . . . . . 266C.6 The federal coastal and environmental programs in the United States. . . . . . . 268C.7 The coastal and environemtal programs in WA. . . . . . . . . . . . . . . . . . 269xiiList of Figures1.1 Causes of sea level changes: Changes in the relative elevation of the land (greenboxes), changes in the relative elevation of the ocean (red boxes), and glacialisostatic uplift (blue box). Ocean properties include salinity and density. Modi-fied after (IPCC, 2014). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51.2 The global mean sea level rise projections of IPCC reports (red) and publishedresearch (blue) for 2100 (Sriver et al., 2018). It should be noted that the IPCCreports are based on a consensus model of reporting, therefore the reports’ sealevel rise projections are relatively conservative (Horton et al., 2014). . . . . . . 71.3 The sea level rise adaptation strategies: protect, accommodate, avoid, retreat,do nothing and offense. a. CGI, b. THS, c. Raising structures, d. Dry or wetproofing structures. Modified after Emily Underwood’s illustrations . . . . . . 91.4 Examples of dynamic coastal forms: a. Dunes, b. Barrier islands, and c. Sand,gravel, and rocky beaches (images from BC Stewardship Center; the NationalOceanic and Atmospheric Administration (NOAA); United States GeologicalSurvey (USGS)) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121.5 Examples of coastal vegetation: a. Salt marshes, b. Mangroves, and c. Seagrasses, d. Dune vegetation, e. Kelp forest, and f. Riparian vegetation (imagesfrom BC Stewardship Center; NOAA National Marine Sanctuaties; Sea TurtleConservation; Dahdouh-Guebas et al. 2005; Koch et al. 2009. . . . . . . . . . . 131.6 Examples of reefs: a. Mussel beds, b. Oyster reefs, and c. Coral reefs (imagesfrom NOAA National Ocean Service Education; Southern California CoastalWater Research Project) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131.7 The impacts of sea level rise, associated hazardous events, coastal processesCGI interacts with, and CGI’s coastal protection role. Modified after Morriset al. (2018). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14xiii1.8 Research chapters and the contextual concepts investigated in each chapter. . . 181.9 Highlighted areas are the coastal regions of British Columbia and WashingtonState . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201.10 Research design of the dissertation chapters . . . . . . . . . . . . . . . . . . . 212.1 The Salish Sea and the study area communities. Numbers are in no particularorder. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352.2 Themes used to measure CGI coastal protection benefits and vulnerability . . . 382.3 Distibution of the CGI coastal protection benefits in the Salish Sea . . . . . . . 462.4 Distibution of the CGI vulnerability in the Salish Sea . . . . . . . . . . . . . . 492.5 Distribution of the CGI coastal protecton potential in the Salish Sea . . . . . . 523.1 Methods diagram showing the research steps and outputs. . . . . . . . . . . . . 653.2 The location of the study area. . . . . . . . . . . . . . . . . . . . . . . . . . . 673.3 Sea level rise adaptation scenarios conceptual visual illustration. Drawings andillustrations are not to scale. Please see the disclaimer at the end of this chapter. 783.4 Diagram showing the results of the evaluation framework application to four sealevel rise adaptation scenarios. . . . . . . . . . . . . . . . . . . . . . . . . . . 904.1 The map of the study area showing the coastal regions of British Columbia andWashington States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1084.2 The number of CGI projects in BC and WA between 2008-2018 . . . . . . . . 1134.3 The objectives of the CGI projects . . . . . . . . . . . . . . . . . . . . . . . . 1144.4 The CGI project types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1144.5 The institutions where the project funding was allocated from . . . . . . . . . . 1154.6 The CGI project leads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1164.7 The average CGI project funding amount allocated by institutions . . . . . . . 1164.8 The average CGI project size in acres by the lead institution . . . . . . . . . . 1174.9 The graphic illustration of the coastal jurisdiction in BC and WA. Distances arenot to scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1224.10 The timeframe of the coastal & environmental programs in Canada, BC, theUSA, and WA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1304.11 The continuity of the coastal and environmental programs available in WA andBC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1314.12 The financial barriers and facilitators in BC and WA . . . . . . . . . . . . . . . 134xiv4.13 The jurisdiction and ownership barriers and facilitators in BC and WA . . . . . 1364.14 The regulations barriers and facilitators in BC and WA . . . . . . . . . . . . . 1374.15 The capacity and resources barriers and facilitators in BC and WA . . . . . . . 1404.16 The vision and leadership barriers and facilitators in CGI . . . . . . . . . . . . 1424.17 The collaborations barriers and facilitators in BC and WA . . . . . . . . . . . . 1444.18 The community and knowledge barriers and facilitators in BC and WA . . . . . 146B.1 Summary of the sea level rise projection timelines . . . . . . . . . . . . . . . . 222B.2 Summary of the baseline data for water levels . . . . . . . . . . . . . . . . . . 224xvList of Abbreviations and AcronymsBC British ColumbiaBEACH The Beaches Environmental Assessment and Coastal HealthCGI Coastal Green InfrastructureCSD Canadian Census SubdivisionCZM Coastal Zone ManagementDFL Designated flood levelsDFO Fisheries and Oceans CanadaDoE Washington State Department of EcologyECC Environment and Climate Change CanadaEEZ The Exclusive Economic ZoneEPA The Environmental Protection AgencyFCL Flood Construction LevelsFLNRRD The Ministry of Forests, Lands, Natural Resource Operations and RuralDevelopmentFWS The U.S. Fish and Wildlife ServiceGIS Geographic Information SystemsIPCC Intergovernmental Panel on Climate ChangeLWM The low water markMEOPAR Marine Environment and Marine Environmental Observation Predictionand Response NetworkMLLWL The mean lower low water lineMoECCS The Ministry of Environment and Climate Change StrategyNGOs Non-governmental organizationsNOAA The National Oceanic and Atmospheric AdministrationOCPs Official Community PlansPICS Pacific Institute for Climate SolutionsPSI Public Scholars InitiativeSMA Shoreline Management AreaTC Transport CanadaTHSs Traditional hard structuresUBC University of British ColumbiaxviUGA American Census City/Urban Growth Areas (UGA)USACE The U.S. Army Corps of EngineersUSGS United States Geological SurveyWA Washington StateWDFW Washington State Department of Fish and WildlifexviiAcknowledgementsThe dissertation process can be all-consuming. Here, I would like to thank a number of personsand organizations who in tangible and intangible ways contributed to this dissertation.To begin, I want to thank my supervisor Prof. Stephanie Chang for supporting me in more waysthan I can count. She has been a patient, forbearing, and compassionate mentor while providingconstructive criticisms and guidance over the past years. She helped me navigate many twistsand turns during my research, and supported me when I faced obstacles. Besides her academicexcellence, she has been a role model through her mentorship style and professionalism. Iconsider it to be my absolute privilege to have had her as my Ph.D. supervisor. I am alsoextremely grateful to my committee members, Prof. Maged Senbel and Prof. Mark Johnson fortheir mentorship and guidance, and sharing their expertise to support my research.I would like to thank my funders the Marine Environmental Observation, Prediction, and Re-sponse (MEOPAR) Networks Center for Excellence, Pacific Institute for Climate Solutions(PICS), and UBC’s Public Scholars Initiative for their scholarship and other financial supportthat made this research happen.The research in this dissertation had greatly benefited from time, support and expert knowledgeprovided by the British Columbia Ministry of Environment and Climate Change’s Climate Ac-tion Secretariat and Fraser Basin Council. While I cannot mention them by name, I am alsograteful to all of the experts in British Columbia and Washington State who participated in myresearch, sharing their time and expertise with me in Chapter 4. I have learned so much fromthem. I also would like to thank the District of North Saanich for the in-kind support and theircooperation, and all the workshop participants in Chapter 3.Special thanks to my labmates Jackie Yip, Michelle Marteleira, Ryan Reynolds, Krista Cawley,Cher King-Scobie, David Righter, Alexa Tanner, Christopher J. Carter and Rebecca Chaster. Itxviiihas been great to learn from and work with them. I am also grateful to the IRES department staffLisa Johannesen, Bonnie Leung, Stefanie Ickert, Kyla Hicks, Gillian Harris, and Linda Stewardfor the administrative support and always being kind.It may be unusual, but I also want to note my appreciation for this beautiful province, BritishColumbia, my home for the last six years. Exploring the fjords, rain forests, and majesticmountains of this magical place has been one of the most rewarding experiences of my life andhas certainly improved my life quality during the challenging times of my PhD life.Many friends had shared this journey with me, and have always been there when I needed themmost. Ghazal Ebrahimi, Jackie Yip, Nancy Silva, Alejandra Echeverri, Lucy Rodina, PoushaliMaji, and Elisabeth Wiliams have each touched my life in ways I will forever be grateful for.I thank you for all the endless hours of companionship, lifting me up when I was down, andmaking me laugh until I cry. Guillaume, Stephen, Golnoush, Natalia, Daniele, Flora, Alex, andEric, Manuel as well as many other friends all around the world who are too numerous to name- thank you for your support and friendship.I am deeply thankful for the love, understanding, support, and encouragement I have receivedfrommy family all around the world. Emel Aksu, Selcuk Conger, Onur Conger, Ezgi Karasozen,Aleida Reitberg, Ruud Peters, Nikki and Volkan Turan, Halime Conger, and Marijke, Jan-Jaapand Max van der Wal, thank you for always being there for me and reminding me the otherimportant things in life.Finally, I would like to express the depth of my love and gratitude to my husband Bas Petersfor his continued love and support. There is nothing more precious than sharing your life witha joyful person, who loves to do the things you love, supports you to pursue your dreams, andkeeps you happy and sane in the meantime. Thank you for sharing day and night, rain and sun,and oceans and mountains with me.xixTo my mother,for your support, encouragement, and unswerving love.Annem’e,desteg˘in, cesaretlendirmen, ve tu¨kenmeyen sevgin ic¸in.xxChapter 1Introduction1.1 RationaleOver the years, science has left no space for dispute on the acceleration of climate change im-pacts throughout the 21st-century (IPCC, 2014). The last few decades have already witnessedan escalating number of extreme climate events linked to the changes in air and water temper-atures; precipitation, storm, and circulation patterns; and mean sea levels (IPCC, 2014; Carsonet al., 2016; Ekwurzel et al., 2017). Particularly, the rapid changes in sea levels have been oneof the most pressing impacts of climate change (Azevedo de Almeida and Mostafavi, 2016), asthe global mean sea levels are expected to rise around 1m by 2100 (IPCC, 2014). This rise isthreatening the integrity of built and natural environments, as well as the social and economicwell-being of coastal communities around the world (Nicholls, 2015; Devoy, 2015). This is be-cause sea level rise affects coastal communities by expanding water’s reach at coasts, bringingfloodwaters and wave action closer to developed areas, and exacerbating erosion at the edgeof critical assets and infrastructure (Glavovic, 2014; Bronen, 2015). These impacts of sea levelrise amplify the existing risks and vulnerabilities of coastal communities to flooding and erosion(Hinkel et al., 2014; Felsenstein and Lichter, 2014; Neumann et al., 2015) due to the decliningenvironmental conditions, and increasing number of people affected and economic losses asso-ciated with such events (Nicholls, 2011; Renaud et al., 2013). The magnitude of these impacts,however, is expected to vary greatly depending on regional and local characteristics of coastalcommunities, and their adaptation to sea level rise (Michener et al., 1997).Until recently, the emphasis on adaptation to flood and erosion events has been on protectingcoastal areas using hard structures such as dikes, seawalls, bulkheads, and other engineeredmeasures. However, there has been a growing debate on the economic, environmental, and1adaptability deficits associated with the use of the traditional1 hard structures (THSs) for sealevel rise adaptation (Tyler, 2016). The costs of building, maintaining, and eventually rebuildingTHSs are very high (Temmerman et al., 2013). THSs have various unwanted effects on fish andwildlife habitat, and sedimentation patterns (Sutton-Grier et al., 2018). Moreover, THSs arenot flexible enough to respond to changing environmental and climatic conditions as well asnew engineering and planning regulations (Borsje et al., 2011). The increased understanding ofTHSs’ limitations on effectively protecting coasts has led to growing recognition of the need fora cost-effective, multifunctional and adaptable coastal protection measure (Narayan et al., 2016;Sutton-Grier et al., 2018).In this context, coastal green infrastructure (CGI), natural or nature-based systems that protectcoasts from flood and erosion, has started to gain attention due to its adaptability to change, andability to provide multiple social, environmental and economic benefits to communities (Arkemaet al., 2017; Sutton-Grier et al., 2018). CGI not only reduces the wave energy and overtopping atthe coast (Arkema et al., 2017), and sediment drift from and along the coast (Silva et al., 2016);but it also provides habitat for fish and wildlife, sequesters carbon, and increases social andeconomic opportunities (Sutton-Grier et al., 2015). In addition, CGI responds to changes in theenvironmental and climatic conditions and can be altered relatively easily (Borsje et al., 2017).Consequently, there has been a growing interest in and demand for developing CGI areas for sealevel rise adaptation (Sutton-Grier et al., 2018). Communities around the world have started toinvestigate where and how they can implement CGI to increase their adaptation to sea level rise(Ruckelshaus et al., 2016).The literature, however, frequently recognizes that CGI may not be applicable to all coastal ar-eas, or may not yield its full potential benefits in all settings (Catenacci and Giupponi, 2013;Narayan et al., 2016). The functionality of CGI and thus the array of benefits that can be ob-tained depends heavily on the environmental, social, economic, and institutional contexts inwhich CGI is implemented. Therefore, the use of CGI as a sea level rise adaptation measurerequires an understanding of the context-dependency of CGI. However, little is known about thecontext-dependency of CGI. Even though the degrees to which CGI can provide coastal protec-tion benefits have frequently been investigated, in which contexts CGI can be used as a sea level1In this dissertation, the term “traditional” refers to conventional coastal engineering practices, rather than the tra-ditional practices of the Indigenous peoples of North America. Indigenous peoples have observed the changes inclimate for a long time, and have developed traditional practices to address these impacts. There are specific cli-mate change assessment frameworks such as Black et al. 2017 and Justice Institute of British Columbia 2015 tohelp Indigenous communities to understand their vulnerability and develop appropriate adaptation actions.2adaptation measure has been largely absent from the CGI research. Considering the adaptationneed is urgent, understanding the contexts in which CGI can be utilized is essential to protectcommunities from the impacts of sea level rise (Langridge et al., 2014), as well as to ensurethat the natural processes, critical habitats, and social and recreational opportunities at coastsare preserved and enhanced.Therefore, in this dissertation I aim to understand the environmental, local and institutionalcontexts in which CGI can be used as a sea level rise adaptation measure. In the next sections,I outline my research goals (Section 1.2); situate my research in the broader literature (Section1.3); highlight the overarching research question, the subsequent research questions for eachof the three empirical studies (Section 1.4); describe the research design, data and methods ofthe three studies (Section 1.5); and describe the chapters of the dissertation via brief synopsis(Section 1.6).1.2 Goals of the dissertationAdaptation to sea level rise is not a choice but a requirement for most coastal communities.CGI overcomes the short-comings of THSs and provides important coastal protection and otherbenefits to communities. In addition, there has been a growing interest in the use of CGI toadapt to sea level rise. However, there remains a gap in identifying when and where CGI canbe implemented as a sea level rise adaptation measure. Therefore, the overarching goal of thisdissertation is to understand the contexts in which CGI can be used as a sea level rise adaptationmeasure.To achieve this goal, I undertake three empirical studies within the coastal regions of BritishColumbia (BC) and the Washington State (WA), an area with high projected sea level rise im-pacts exacerbated by subsidence, and increased population exposure and community sensitivityto these impacts (Okey et al., 2012; Adelsman et al., 2012). I organize these studies into threerespective themes: Theme 1 - Environmental, Theme 2 - Local, and Theme 3 - Institutional. Inthe first theme, I aim to develop a methodology to understand CGI’s potential coastal protectionbenefits in the Salish Sea region. In the second theme, I seek to highlight the trade-offs betweenCGI and other sea level rise adaptation strategies in a local community. In the third theme, Iseek to better understand the institutional factors that disable or enable the implementation ofCGI in BC and WA.31.3 Situating the research in the literatureI situate my dissertation in the fields of coastal green infrastructure (CGI) and sea level riseadaptation. I draw from a number of different disciplinary perspectives such as urban planning,environmental science and geography, and sociology.Therefore, this dissertation builds on key CGI and adaptation literatures, including CGI termi-nology and overall benefits (i.e. Sutton-Grier et al. 2015; Narayan et al. 2016; Morris et al.2018); CGI’s flood (i.e. Koftis et al. 2013; Anderson and Smith 2014; John et al. 2015) and ero-sion (i.e. Silva et al. 2016; Borsje et al. 2017; Mendoza et al. 2017) protection benefits; CGI’svulnerability to changing environmental conditions (i.e. Feagin et al. 2015; Osland et al. 2015;Kirwan et al. 2016); sea level rise adaptation (i.e. Adger et al. 2007; Matthews et al. 2015;Hagen et al. 2018); resilience and resilience perspectives in adaptation (i.e. Gersonius et al.2016; Cinner et al. 2018); sea level rise adaptation strategies (i.e. Catenacci and Giupponi 2013;Cooper 2016), and evaluation methods (i.e. Cutter 2016; Gersonius et al. 2016; Azevedo deAlmeida and Mostafavi 2016); and institutional barriers to and facilitators of adaptation (i.e.O’Donnell et al. 2017; Hopkins et al. 2018; Sutton-Grier et al. 2018). In each empirical chap-ter, I review and synthesize the relavant and up-to-date literature. In this section, I present anoveraching review of the literature as they relate to the goals of this dessertation.1.3.1 Sea level change: causes, impacts, and projectionsSea level change refers to the temporal and spatial changes of the “height of the ocean surface”,and is measured as relative sea level2 or geocentric sea level3 (IPCC, 2014, pg.1142).Historic and anthropogenic causesChanges in natural processes related to earth’s surface, atmosphere and oceans have been oc-curring due to long-term changes in the climate before the pre-industrial period4 (IPCC, 2014;Devoy, 2015). For example, the global mean sea level was approximately 120 meters lower inthe last glacial period, than its present level (Tamisiea, 2011). However today, anthropogenicfactors are known to be the main drivers of these changes (Shayegh et al., 2016).2Relative sea level is the height of the ocean surface relative to the surface of Earth (IPCC, 2014).3Geocentric sea level is the height of the ocean surface relative to the center of the Earth (IPCC, 2014).4Around 1850s (Nicholls et al., 2007).4Today, anthropogenic factors are known to be the main drivers of climate change (IPCC, 2014).In the post-industrial period, increasing levels of CO2 and other greenhouse gases are affectingatmospheric and oceanic processes (Michener et al., 1997), such as changes in air and watertemperatures; storm intensity, duration, frequency and location; precipitation patterns; oceanicand atmospheric circulation; and altered timing of seasons (Okey et al., 2012; Devoy, 2015).These processes affect global and regional sea level change in the following ways (Figure 1.1):• Changes in the relative elevation of the land (i.e., subsidence or uplift) due to naturalresource extraction and compaction of soil;• Changes in the relative elevation of the ocean due to (1) growth or decay of ice sheetsand their redistribution, (2) increasing air and ocean temperatures that causes thermalexpansion, (3) melting of land-based glaciers and ice sheets due to increased air andocean temperatures, (4) low atmospheric pressure that causes the swell of ocean’s surface(i.e., storm surges), and (6) hydrological cycles that cause evaporation and rain events,(7) total river discharge that traps fresh water floating on salt water, and (8) salinity levelsthat changes water density and thus, volume;• Glacial isostatic uplift caused by the loading and/or unloading on the earth’s crust due tothe changes in the ice sheet mass (Devoy, 2015; Nauels et al., 2017; Piecuch et al., 2018);Figure 1.1: Causes of sea level changes: Changes in the relative elevation of the land (green boxes),changes in the relative elevation of the ocean (red boxes), and glacial isostatic uplift (blue box). Oceanproperties include salinity and density. Modified after (IPCC, 2014).5The combined effect of these three factors and the characteristics of the coastal profile (i.e.,coastal geomorphology), determine local sea level changes and impacts.ImpactsThe impacts of sea level rise are flooding, erosion, ecosystem health and integrity, salinizationof coastal lands and waters, and water table levels (Nicholls, 2015). The changes in waterlevels influence flooding and erosion by altering the patterns (direction, zonation, extent, andimpacts) of tides, storm surges, wave action, currents, and subsequent sediment deposition rates(Davidson-Arnott, 2010). Particularly in gently sloping coasts, the magnitude of these impactsis much higher. The changes in water levels cause alterations in ecosystem health and integritydue to changes in inundation patterns, sediment concentrations, and salinity levels (Devoy, 2015;Passeri et al., 2015). The salinization of coastal lands and waters are caused by the extensionof the flooded areas inland (Devoy, 2015). Lastly, the changes in water levels influence thelevel of the water table, leading to the creation of new wetlands and inundated areas inland orcompaction of soil layers (Rotzoll and Fletcher, 2013).The impacts of sea level rise are expected to increase in future because the magnitude of climatechange and the increases in sea levels are projected to accelerate throughout the 21st-century(Nicholls, 2011; Devoy, 2015; Nauels et al., 2017).ProjectionsSea level rise has been one of the most pressing impacts of climate change (Azevedo de Almeidaand Mostafavi, 2016). The global mean sea levels rose about 1.7 mm per year between 1900-2010 compared to 3.2 mm per year between 1993-2010 (IPCC, 2014). However, it is verychallenging to predict future sea levels because the model-based assessments have to accountfor the gaps in scientific knowledge, they employ different methodologies, and they are oftenbuilt upon different assumptions (Devoy, 2015; Nauels et al., 2017; Sriver et al., 2018).Intergovernmental Panel on Climate Change (IPCC), which brings scientist together to synthe-size the recent literature on the impacts of climate change, is considered to be the most prominentand comprehensive report for sea level rise projections. However, even though the latest IPCCreport 2014 estimates an average global increase in sea levels by 1 meter by 2100, there hasbeen significant disparity amongst the projections of the previous IPCC reports and the studiesIPCC reports are based on. The disparity over the amount of sea level rise expected by 2100 canalso be seen in the findings of the other published research (Figure 1.2).6Figure 1.2: The global mean sea level rise projections of IPCC reports (red) and published research (blue)for 2100 (Sriver et al., 2018). It should be noted that the IPCC reports are based on a consensus model ofreporting, therefore the reports’ sea level rise projections are relatively conservative (Horton et al., 2014).In BC and WA, communities are preparing for 1 meter of sea level rise following their nationaland regional guidelines (The Arlington Group et al., 2013; Adelsman et al., 2012).The uncertainties over the amount of sea level rise expected by 2100 calls for diverse and flexibleadaptation options to account for these uncertainties.1.3.2 Sea level rise adaptation and adaptation strategiesIn this dissertation, I focus on the local level adaptation to sea level rise and define adaptation asthe processes of adjusting community responses to reduce the impacts of sea level rise that are7observed at the local level. These processes are often in the form of planning, policy, design, andengineering strategies (Nicholls, 2015). There are six main sea level rise adaptation strategies,reflecting different motivations (Cooper, 2016). These strategies are “protect”, “accommodate”,“avoid”, “retreat”, “do nothing” and “offense” (Figure 1.3).The Arlington Group et al. (2013), Glavovic and Smith (2014), Cooper (2016) and Manuel et al.(2016) describe these adaptation strategies as follows:• The protect strategy refers to the use of THS and CGI measures to mitigate the impactsof sea level rise.• The accommodate strategy describes adjustments and changes that are implemented onthe existing structures to mitigate the impacts of sea level rise.• The avoid strategy refers to the prevention of development from the high-risk areas.• The retreat strategy describes a phased-out relocation from the high-risk to low-risk areas.• The do nothing strategy describes the absence of a coordinated and planned action toaddress the impacts of sea level rise.• The offense strategy refers to the designation of new development areas on the reclaimedcoastal lands.Each of these adaptation strategies includes different measures. The adaptation measures referto the actions that vary in their nature (i.e., design, material, and method) but serve the samemotivation as the strategy they represent (i.e., a, b, c and d in Figure 1.3). They define the waysin which the strategies are operationalized.8Figure 1.3: The sea level rise adaptation strategies: protect, accommodate, avoid, retreat, do nothing andoffense. a. CGI, b. THS, c. Raising structures, d. Dry or wet proofing structures. Modified after EmilyUnderwood’s illustrations aa 91.3.3 Protect strategy measuresA range of coastal protection measures exist, from grey/structural to green/natural, to defendcoasts from flooding and erosion. Table 1.1 summaries the coastal protection measures, charac-teristics, and types.Coastal protection measures Characteristic and types• Characteristic: Structural elementsSea walls and bulkheadsDikes and super dikesRevetments and leveesEmerged breakwaters, ripraps, and jettiesSubmerged breakwatersGroynes• Characteristic: Combined structural & natural elementsVegeteted revetments and sillsBeach nourishmentTextured and/or natural material breakwatersSubmerged reefballsConstructed dunes• Characteristic: Natural elementsLogs and woody debris placementCoastal and riparian vegetationWetlands and estuariesBarrier islands and coastal dunesCoastal reefsTable 1.1: Types of coastal protection measures from THSs to CGIs, adapted from (Morris et al., 2018;Sutton-Grier et al., 2018; Borsje et al., 2017; Narayan et al., 2016; Bridges et al., 2015). Traditional Hard StructuresTraditional hard structures (THSs) is an adaptation measure of the protect strategy. Typically,communities have prioritized the protection of coasts from flooding and erosion using THSs,10such as dikes, seawalls, groins, and other built structures (Airoldi et al., 2005; Borsje et al.,2011; Renaud et al., 2013). However, there has been a growing debate on economic and envi-ronmental deficits of THSs. THSs have detrimental impacts on sensitive habitat, erosion, andsedimentation patterns (Cheong et al., 2013). They are static, immobile and not flexible enoughto respond to changing conditions such as urban sprawl, climate change and new engineeringregulations (Borsje et al., 2011). They require constant maintenance, and the cost of building,maintaining, and eventually rebuilding hard structures is very expensive (Mo¨ller et al., 2001). Agrowing amount of studies suggest that even with substantial funding, coastal communities arenot always protected effectively by THSs (Klein et al., 2001).Besides the negative implications of THSs, they have been used widely and consistently due toseveral important reasons. First and foremost, they protect private and public properties, andsustain the ownership, development, and occupation of the land (Tyler, 2016). Second, THSsare more tangible and easy to relate to (French, 2006), therefore, they appeal more to decision-makers. This is because they are easy to be visualized and communicated (Klein et al., 2001).Third, compared to CGIs the implementation of THSs often takes considerably less land space.Considering the high land values in coastal areas and the need and desire to utilize these valuableareas, THSs can be implemented much easier than their CGI counterparts in “in intertidal andshallow subtidal environments” (Airoldi et al., 2005, pg.1074). Coastal Green InfrastructureCoastal green infrastructure (CGI) is an adaptation measure of the protect strategy. CGI in thisstudy is used as an umbrella term to describe natural or nature-based systems and processes thatmimic dynamic coastal landforms, coastal vegetations, and reefs systems. CGI provides coastalprotection services as well as ecosystem health, maintenance of natural processes and biodiver-sity, and social and economic benefits to communities. CGI protects coasts from flooding anderosion through reducing wave energy and height, attenuating floodwater, trapping sedimentover soil, binding soil, and mitigating debris movement (Chenoweth et al., 2018; Morris et al.,2018). The following paragraphs refer to three main types of CGI.Dynamic coastal landformsDynamic coastal landforms are formed together with vegetation and sand transported by waves,wind, currents, and tides (Davidson-Arnott, 2010). Examples include sand/gravel/rocky beaches,11barrier islands, and dunes (Barbier et al., 2011; Sutton-Grier et al., 2015; The Horinko Group,2015; Ruckelshaus et al., 2016). They reduce wave energy and trap sediments transported bywater and wind. They respond rapidly to changes in sediment supply, wave action, flooding, andsea level changes (Davidson-Arnott, 2010; Feagin et al., 2010). Due to their unique position atthe edge of coastlines5, these dynamic coastal landforms have provided humans and nature withimportant ecosystem services (Keijsers et al., 2014).Figure 1.4: Examples of dynamic coastal forms: a. Dunes, b. Barrier islands, and c. Sand, gravel, androcky beaches (images from BC Stewardship Center; the National Oceanic and Atmospheric Administra-tion (NOAA); United States Geological Survey (USGS))Coastal vegetationCoastal vegetation is typically characterized by the presence of macrophytes6. In this study, thedefinition of the coastal vegetation also includes the dune vegetation and coastal riparian vegeta-tion. Situated at the interface between land and ocean (IPCC, 2014), coastal vegetation reduceswave energy and trap sediments (Figure 1.5). They are also well adapted to deal with naturalstressors such as saline water, high tides, extreme temperatures, strong winds, and anaerobicsoils (Kathiresan and Bingham, 2001; Davidson-Arnott, 2010; Duarte et al., 2013). Examplesinclude kelp forests, eelgrasses, salt marshes, mangroves, and dune and riparian vegetations(Davidson-Arnott, 2010; Gutie´rrez et al., 2011; Arkema et al., 2013; Sutton-Grier et al., 2015;Ruckelshaus et al., 2016; Sandi et al., 2018). Coastal vegetations are known as ecosystem engi-neers, as they physically modify environments and impact species, coastal processes, and overallecosystem functioning (Bouma et al., 2009; Borsje et al., 2011; Duarte et al., 2013; Sandi et al.,2018).5Coastlines are not stable therefore their exact location cannot be determined. The high water line, the visible lineat the coast where the highest water level reaches, is typically used in the literature as reference (Klemas, 2011;Strauss et al., 2012; Weiss et al., 2011).6Emergent, submergent or floating aquatic plants (Craft et al., 2009; Koch et al., 2009).12Figure 1.5: Examples of coastal vegetation: a. Salt marshes, b. Mangroves, and c. Sea grasses, d.Dune vegetation, e. Kelp forest, and f. Riparian vegetation (images from BC Stewardship Center; NOAANational Marine Sanctuaties; Sea Turtle Conservation; Dahdouh-Guebas et al. 2005; Koch et al. 2009.Reef systemsReef systems are dynamic ecosystems that constantly evolve and change in response to dis-turbances (Davidson-Arnott, 2010). They play roles in reducing wave energy at the coast bycreating drag friction by providing surface roughness (Gutie´rrez et al., 2011). Some examplesinclude mussel beds, oyster reefs and coral reefs (Sutton-Grier et al., 2015; The Horinko Group,2015; Ruckelshaus et al., 2016). Mussel beds and oyster reefs are located in the intertidal zoneor low subtidal zone of coastlines (Scyphers et al., 2011).Figure 1.6: Examples of reefs: a. Mussel beds, b. Oyster reefs, and c. Coral reefs (images from NOAANational Ocean Service Education; Southern California Coastal Water Research Project)13Even though CGI is equipped with mechanisms to deal with environmental stressors, it can alsobe vulnerable to changes in the land use, water levels, and storm intensities and frequencies(Osland et al., 2015). This vulnerability can alter the extent of CGI’s coastal protection benefitsby interfering with CGI’s movement along the coastal profile and its interaction with naturalprocesses (Khattabi and Bellaghmouch, 2009). Figure 1.7 schematizes the SLR impacts and theresulting hazardous events, the natural processes CGI interacts with, and its protection role.Figure 1.7: The impacts of sea level rise, associated hazardous events, coastal processes CGI interactswith, and CGI’s coastal protection role. Modified after Morris et al. (2018).141.3.4 CGI in the adaptation contextSea level rise adaptation is a major challenge for coastal communities not only because of therapid increase in the water levels, associated hazards, and the sensitivity and exposure of com-munities. The ways communities have been responding to changes (such as long-range andrigid planning practices and static engineering designs) and the increasing number of essentiallocal services (such as mental health, social and housing, wastewater, flood management, anddrinking water (Duffy et al., 2014, pg.4)) make it difficult for communities to address adaptationneeds. Therefore, CGI has been a desirable adaptation measure to address these challenges, aswell as other concerns coastal communities are dealing with.Besides its ability to deal with the impacts of sea level rise by mitigating flooding and ero-sion impacts, the value of CGI in adaptation comes into play due to its multi-functionality,cost-effectiveness, and adaptability (Naumann et al., 2011; Narayan et al., 2016; Arkema et al.,2017). This is because the multifunctional, cost-effective, and adaptable nature of CGI helpcommunities deal with some of the other concerns they are facing on a day to day basis. There ismounting evidence in the literature that CGI provides carbon sequestration, water filtration, bio-diversity, fish and wildlife habitat, and recreational benefits (Matthews et al., 2015; Chenowethet al., 2018; Morris et al., 2018). To start with, the multifunctionality of CGI contributes to theglobal efforts on greenhouse gas emissions reduction, reduces water pollution, contributes to thehealth and wellbeing of ecological systems, as well as enhancing human health and social life.Moreover, a growing number of studies and reports suggest that CGI is significantly cheaper toimplement and to maintain than THSs (Lamont et al., 2014; Vineyard et al., 2015; Narayan et al.,2016; Onuma and Tsuge, 2018). By implementing CGI and reducing the implementation andmaintenance costs, communities can allocate resources to other services. Lastly, large uncer-tainties exist over the rate and height of the sea level rise communities will experience (Carsonet al., 2016). CGI’s adaptable nature provides a dynamic and flexible adaptation measure thatcan deal with and adapt to these uncertainties (Nessho¨ver et al., 2017).Ultimately, CGI not only helps communities adapt to sea level rise, but it also increases commu-nity resilience to it by accounting for uncertainties, multi-functionality, adaptability, communityperspectives, and knowledge mobilization.151.3.5 The context-dependency of CGIThe context-dependency of CGI has been a recent but growing field of study. The literatureshows that there are various ways to investigate this context-dependency including but not lim-ited to investigating the seasonal and temporal effectiveness (Koch et al., 2009; van Proos-dij et al., 2006; van Proosdij and Townsend, 2006), cost-effectiveness (Narayan et al., 2016),flood and erosion reduction capabilities (Ruckelshaus et al., 2016), institutional arrangements(Matthews et al., 2015), social acceptability and public perspectives (Chaffin et al., 2016), andtrade-offs (Catenacci and Giupponi, 2013) of CGI in different areas. In this dissertation, I focuson the following three contexts: (1) the locality of CGI, (2) the trade-offs of CGI, and (3) theinstitutional arrangements of CGI.Locality of CGICGI is not independent of its environment. The ways in which CGI functions and providescoastal protection and other benefits are geographically relevant (Langridge et al., 2014). TheCGI benefits depend on CGI’s interactions with the coastal processes and the built environments(Hanley et al., 2014). For example, the biophysical conditions of an area may be right to imple-ment CGI, but there may be other factors influencing its functionality. The rate of sea level riseexpected in that area may be very high, leading to the drowning of CGI (Johnson et al., 2012).Alternatively, there may be urban development immediately after CGI at the coast, creating a“coastal squeeze” and preventing its natural migration along the coastal profile (Osland et al.,2015; Kirwan et al., 2016). There may be a situation where the sea level rise rate in an areais low, and the development does not create a coastal squeeze, but the wave action at the coastmay be too high for the implementation of CGI (Ghosh and Chaudhuri, 2015). Alternatively,the sediment deposition rate at an area may be low due to the human interventions up or downthe coast. The benefits of CGI and its role in adaptation, therefore, vary greatly depending onthe environmental context of CGI’s location (Ruckelshaus et al., 2016).Tradeoffs of CGIEven if all the environmental factors are favorable, there may be key built environment, social,economic and institutional trade-offs between CGI and other adaptation strategies at the locallevel (Catenacci and Giupponi, 2013; Oddo et al., 2015). Therefore, CGI may not be the bestadaptation measure for all communities. In the built environment, CGI may lead to significantchanges in the local land use, resulting in the loss of important agricultural, industrial or resi-16dential areas (Pramanik, 2017; Shayegh et al., 2016). Socially, there may be significant publicpushback for CGI implementation (Flynn et al., 2018) or CGI may lead to social inequalities orexacerbate the existing ones within communities (Leichenko and Silva, 2014; Pramanik, 2017).The social trade-offs of CGI may reduce the public and political support for the use of CGIfor sea level rise adaptation. Economically, other sea level rise adaptation strategies such asaccommodate, avoid, or retreat may be less costly for local governments or homeowners toimplement, or they may have more positive impacts on the local economy than CGI. Depend-ing on who will pay for adaptation in the community, CGI may not be economically desirable(Gibbs, 2015). Institutionally, the local governments may not be equipped with the necessarystaff, technological, and resources to undertake the implementation of an emerging adaptationmeasure. In the absence of the necessary institutional capacities, communities may resort to theimplementation of other strategies (Patterson, 2018; Cinner et al., 2018). Having the right envi-ronmental conditions does not always lead to having the right social, economic and institutionalsettings for CGI. Considering that communities have different environmental, social, economicand institutional settings than each other, CGI will likely to have different trade-offs in differentlocal contexts. Therefore, the trade-offs of CGI vary in the local community settings.Institutional arrangements of CGIEven if all the environmental factors and the local trade-offs are favorable, the implementationof CGI may still be hindered due to the institutional arrangements in place (Biesbroek, 2014).The implementation of CGI and any other adaptation measures rely on the institutional arrange-ments and the corresponding social, cultural, political and regulatory environments (Lo¨f, 2013;Mguni et al., 2015). The reason is that the institutional arrangements can enable the governance,jurisdictional, and regulatory shifts or adjustments necessary to implement new adaptation mea-sures such as CGI (Nessho¨ver et al., 2017). The institutional arrangements influence allocationof government responsibilities, jurisdictional boundaries, regulations, and programs (Barnettet al., 2013). In addition, they influence the financial resources that can be allocated, politicalmotivations, organizational capacities, and communication strategies for CGI implementation(Ziervogel and Parnell, 2014). Communities in different regions and counties abide by differentinstitutional arrangements than each other. The implementation of CGI in one place may not beeasily replicated in another due to different local, regional, and national governments’ responsi-bilities, jurisdiction, regulations, and programs in place. Therefore, the implementation of CGIis influenced by the institutional contexts (Eisenack et al., 2014).171.4 Overarching and subsequent research questionsThe dissertation follows a manuscript format. The overarching research question is as follows:• In which contexts CGI can be used as a sea level rise adaptation measure?In the attempt to answer this question, I develop three empirical studies investigating environ-mental, local and institutional contexts of CGI. Each of these studies corresponds to a chapterthat is intended to stand alone and address the three research themes outlined in Section 1.2. Fig-ure 1.8 shows the research chapters and the contextual concepts investigated in each of them.Figure 1.8: Research chapters and the contextual concepts investigated in each chapter.18In Chapter 2 (Theme 1 - Environmental), I investigate whether CGI’s potential coastal protectionbenefits are influenced by its vulnerability to changes in environmental conditions. I develop amethodology to identify areas where CGI has more (or less) significant promise for coastalprotection, using environmental indicators. The main research question of this chapter is: Howcan the CGI coastal protection potential be identified, while taking into account its vulnerabilityto changing environmental conditions?In Chapter 3 (Theme 2 - Local), I investigate whether different sea level rise adaptation strate-gies have different local trade-offs due to local characteristics of communities. I incorporatelocal perspectives to create sea level rise adaptation scenarios and develop an evaluation tool toidentify the trade-offs between CGI and other adaptation strategies. The main research questionof this chapter is: How can the local tradeoffs between CGI and other adaptation strategies beevaluated?In Chapter 4 (Theme 3 - Institutional), I investigate whether the institutional arrangements ofdifferent regions influence the implementation of the CGI projects. I explore the institutionalbarriers to and facilitators of CGI implementation by comparing the CGI projects, institutionalarrangements, and practitioners’ perspectives in BC and WA. The main research question of thischapter is: What are the institutional barriers to and facilitators of CGI implementation?1.5 Research design, data, and methodsAs described above, this research consists of three empirical studies, aiming to contribute to theoverarching research question - in which contexts CGI can be used as a sea level rise adaptationmeasure? There are, of course, various ways in which one can answer this research question.In the design of this research, I deliberately choose to employ three different studies becauseeach study underpins different contributions to our understanding of the context dependency ofCGI. In each chapter, I deliberately draw on a mix of methodological approaches because themixed-method approaches can add insights and perspectives that may be otherwise missed, thestrengths of one approach can help overcome the shortcomings of another one, and the resultsof different methods can be used to validate each other, thus can provide a stronger foundationfor the findings.19Another deliberate decision of the research design is to conduct the research activities in thecoastal regions of BC and WA in the United States. Situated at the west coast of Canada, andthe northwest coast of the United States (Figure 1.9), these regions share a long border, his-tory, and culture, and often exchange ideas and policies with each other (Simeon and Radin,2010). They have similar yet diverse environmental, climatic and ecological characteristics andconcerns (James et al., 2014). Their natural environments and the processes that affect theseenvironments are connected and not bounded by the political boundaries. Moreover, the envi-ronmental (Okey et al., 2014), economic (Withey et al., 2015) and social (Binder et al., 2010)signs of the impacts of sea level rise are already apparent in these regions. There has alsobeen a growing interest in and expertise with CGI in the region. In WA, many local govern-ments, multi-organizational partnerships, and environmental non-profit organizations have beenremoving THSs and exploring CGI alternatives. In BC, there has been an increasing numberof federal and provincial initiatives and grant programs for CGI projects in the last two years.Therefore, the coastal regions of BC and WA provide unique opportunities to investigate the useof CGI as an adaptation measure under diverse but comparable contexts.Figure 1.9: Highlighted areas are the coastal regions of British Columbia and Washington StateEach chapter includes detailed descriptions of the methods, research activities, and analysis.Here, I explain how each chapter fits into the overarching research design. Figure 1.10 schema-tizes the flow and summary of the research design of this study, including the organization ofthe themes and chapters, study areas, background driver(s) of the chapters, inputs and outputs,types of data, and methods that are used to achieve the goals outlined in Section 1.2.20Figure 1.10: Research design of the dissertation chapters21In the first part of this dissertation (Theme 1 - Chapter 2), I undertake a regional level study,investigating where in the Salish Sea region CGI has the highest potential coastal protectionbenefits using environmental (climatic, natural and built environment) indicators. To start with,I conducted a review of the primary, secondary and grey literature to identify the various pa-rameters used to measure CGI’s coastal protection benefits and vulnerability (N=151). Next, Isystematically reviewed the content of the references that contained specific parameters relatedto CGI’s role in coastal protection and vulnerability (N=77). The results of the content analysisinformed the inputs of this study. I selected coastal communities with populations over 4000in the study area for data collection, which resulted in 74 communities in total (44 in BC and30 in WA). I collected the input data from the publicly available BC provincial and WA statedatabases. Using rule-based methods and the literature, I organized the data into two indices:CGI coastal protection index, and CGI vulnerability index. I synthesized the results of the twoindices using a 2x2 matrix. Lastly, I mapped the results to highlight the areas where CGI’spotential coastal protection benefits are high.In the second part of this dissertation (Theme 2 - Chapter 3), I undertake a local level study,investigating the trade-offs between CGI and other adaptation strategies while incorporatinglocal perspectives. The District of North Saanich was selected to apply the research methodsconceptually. I first organized an expert meeting (N=18) to review local data on sea level rise,tidal range, storm surge, and associated wave effects. Using the survey results and notes fromthe first expert meeting, I organized and facilitated a participatory workshop (N=38) to gathercommunity preferences on sea level rise adaptation strategies. Using the literature and the feed-back from the workshop, I developed narratives and visuals for four sea level rise adaptationscenarios. To review these scenarios, I organized a second expert meeting (N=9), where theparticipants reviewed the scenarios and gave feedback. Next, I developed a draft sea level riseadaptation strategies evaluation framework by reviewing the academic and grey literature. Iconducted an expert elicitation survey with the experts in the region (N=3) to get their feedbackon the evaluation framework. I incorporated the expert feedback on the evaluation framework.Lastly, I applied the evaluation framework to the adaptation scenarios to identify the trade-offsbetween CGI and other adaptation strategies I used descriptive qualitative and quantitative anal-ysis to analyze the results of different research activities in the study.In the third and the last part of this dissertation (Theme 3 - Chapter 4) undertake an institutionalcomparison study, investigating the institutional barriers to and facilitators of CGI implementa-22tion in BC and WA. To start with, I first reviewed the CGI projects that were implemented (orare in the implementation process) between 2008-2018 (N=235). I use descriptive quantitativeanalysis to understand the state of the CGI projects and the institutional roles in BC, comparedto WA. Next, I reviewed the literature, government websites and documents to identify gov-ernment structures and decision-making authority distributions, coastal jurisdiction boundaries,and coastal and environmental regulations and programs. I used document review and synthesisto understand the institutional arrangements in BC and WA and compare them. In the last step, Iconducted semi-structured interviews with the practitioners that have been involved in the CGIimplementation processes in BC and WA (N=5 in each region). Using descriptive quantitativeand qualitative analysis, I aimed to provide a more in-depth understanding of the institutionalbarriers and facilitators and to identify CGI specific barriers and facilitators practitioners face inBC, compared to WA.1.6 Dissertation synopsisIn Chapter 2, Developing indicators to identify coastal green infrastructure potential: incor-porating coastal protection benefits and vulnerability to changing environmental conditions, Iargue that CGI’s coastal protection benefits are influenced by its vulnerability to environmentalconditions and vary by location. Therefore, I suggest that investigations into where CGI pro-vide coastal protection benefits need to include CGI’s vulnerability and location-specific naturalprocesses and built environment characteristics. I ask the question how can the CGI coastalprotection potential be identified in a region while incorporating its vulnerability to changingenvironmental conditions? What parameters can be used to identify CGI coastal protection ben-efits and vulnerability? What criteria can be used to organize these parameters into CGI coastalprotection and vulnerability indices? How can these indices inform where CGI has greater (orlesser) promise for coastal protection on the basis of the degree of potential coastal protectionbenefits and the degree of vulnerability? Drawing on Gornitz (1991)’s methodology, I developindicators to assess CGI’s coastal protection benefits and vulnerability. I synthesize them usinga 2x2 matrix to identify areas with the highest coastal protection potential. Findings suggest thatCGI in the large population centers in BC, and most of the communities in WA may not providehigh coastal protection benefits. CGI in the smaller communities surrounding large populationcenters have a higher potential to provide coastal protection benefits. Besides the findings of this23study, the CGI specific indices developed, and the methodology used in Chapter 2 contribute tothe broader CGI literature.In Chapter 3, Resilience-based evaluation of the trade-offs of coastal green infrastructure insea level rise adaptation strategies, I argue that the adaptation and resilience contributions ofCGI vary significantly with the biophysical, environmental (natural and built), economic, in-stitutional and social environments of communities. Therefore, I suggest that there are localtrade-offs between the implications of CGI and other adaptation strategies. I ask the ques-tion what are the tradeoffs of CGI as a sea level rise adaptation measure, compared to others?How can the local expert and stakeholder perspectives be incorporated in the development ofthe local sea level rise adaptation scenarios? What resilience concepts can be used to developan evaluation framework to identify the local trade-offs of sea level rise adaptation strategies?Drawing on the resilience and adaptation planning literature (i.e., Picketts et al. 2012; Bar-ron et al. 2012), I collaborate with a local coastal community to understand the local impactsof sea level rise, to identify the community’s preferred adaptation strategies, and to developcommunity-specific adaptation scenarios. Next, drawing on the evaluation of adaptation strate-gies literature (i.e. Catenacci and Giupponi 2013; CAP and ICLEI 2015; Lockwood et al. 2015;Azevedo de Almeida and Mostafavi 2016; Cutter 2016; Gersonius et al. 2016), I develop a 30component evaluation framework addressing coastal processes, natural environment, built envi-ronment, economic, institutional, and social concepts of sea level rise adaptation and communityresilience. In the last step, I apply the evaluation framework to the local adaptation scenarios toidentify the trade-offs between CGI and other strategies. Findings suggest that there are localtrade-offs between the implications of different adaptation strategies. In this particular localcontext, the CGI scenario performed the best amongst all scenarios. The CGI scenario had thehighest score for the coastal processes, natural environment, built environment, economic, andsocial components, and the lowest for the institutional component. In addition to the findings ofthis study, the main contributions of Chapter 2 include the methods and processes of developingsea level rise adaptation scenarios and the evaluation framework.In Chapter 4, Institutional barriers to and facilitators of coastal green infrastructure imple-mentation: a comparative study in British Columbia and Washington States, I argue that in-stitutional arrangements of communities influence the implementation of the CGI projects. Isuggest that common barriers to adaptation such as governance structures, decision-making au-thorities, coastal jurisdiction, and ownership, regulations and programs, as well as other CGI24specific barriers impede the implementation of CGI. I ask the question what are the institutionalbarriers to and facilitators of CGI implementation? What are the differences between the in-stitutional roles in CGI implementation in BC and WA? What are the differences between theinstitutional arrangements of different levels of governments in BC and WA?What are the prac-titioners’ perspectives on the institutional barriers to and facilitators of CGI implementation inBC and WA? I first look at the CGI projects implemented in BC and WA between 2008-2018,to compare the institutional roles and support the CGI projects received. Drawing on the bar-riers to climate change and sea level rise adaptation literature (i.e. Adger et al. 2007; Moserand Ekstrom 2010; Mozumder et al. 2011; Carlsson-Kanyama et al. 2013; Lo¨f 2013; Hansenet al. 2013; Hamin et al. 2014; Ziervogel and Parnell 2014; Hamin and Gurran 2015; Mguniet al. 2015; Reckien et al. 2015), I review the governance structures, distribution of decision-making authorities, coastal jurisdiction and ownership, and regulations and programs in BCand WA. Drawing on the same literature, I conduct semi-structured interviews with the practi-tioners in BC and WA. The findings suggest that CGI implementation is indeed influenced bysimilar barriers to and facilitators of adaptation. These are grouped under the following cate-gories: governance structure and authority distribution; regulations; financial assistance; seniorlevel support; organizational capacity; leadership and political will; and public knowledge andcommunication. The findings also suggest that several CGI specific barriers and facilitatorsimpede or drive the implementation of the CGI projects, categorized under coastal jurisdictionand ownership; financial variation and flexibility; vision; organization efficiency and access toresources; partnerships and collaborations; NGOs; and community advocacy. Besides these re-sults, the findings of the study contribute important details to each category, providing additionalinsights into the institutional barriers to and facilitators of adaptation and CGI implementation.In Chapter 5, Conclusion, I synthesize the summary of findings for each chapter. Based onthese findings I discuss broad conclusions and policy implications. I highlight the strengths andlimitations of the dissertation. Next, I identify potential future research directions. Lastly, Iconclude by noting personal reflections.25Chapter 2Developing indicators to identifycoastal green infrastructure potential:incorporating coastal protectionbenefits and vulnerability to changingenvironmental conditions2.1 IntroductionTraditionally, coastal adaptation to climate change has prioritized the protection of coasts usinghard structures such as dikes, seawalls, spillways, groins, and other built structures (Klein et al.,2001; Airoldi et al., 2005; Renaud et al., 2013; Feagin et al., 2015; Schubert et al., 2017). How-ever, there has been a growing debate on coastal protection roles and benefits of hard structures,as well as their environmental, social and economic implications (Sutton-Grier et al., 2015).These human-made structures are static and unable to respond to changing conditions such as ur-ban sprawl, climate change, and new planning and engineering regulations (McGranahan et al.,2007; Borsje et al., 2011). They are often expensive to build, and they require regular mainte-nance (Temmerman et al., 2013; Onuma and Tsuge, 2018). They provide an inflated sense ofprotection and security, resulting in increases in development in flood-prone areas (Tyler, 2016;Schubert et al., 2017). Moreover, they cause significant damages to ecosystems and sensitivehabitats by disrupting natural processes and preventing migration of habitat and species alongto the coastal profile (Hanley et al., 2014; Onuma and Tsuge, 2018; Sutton-Grier et al., 2018).Consequently, they may have adverse implications for the coastal tourism sector, and local andregional fisheries (Onuma and Tsuge, 2018). In addition, parts of these hard structures can26become loose with strong wave activity or through time and can cause significant damages toinfrastructure, assets, and human lives (Tyler, 2016; Sutton-Grier et al., 2018). These unwantedeffects of traditional hard structures have led to growing recognition of the need for a dynamic,safe and multi-functional ways of coastal protection (Cheong et al., 2013; Temmerman et al.,2013; The Horinko Group, 2015; Narayan et al., 2016; Sutton-Grier et al., 2018).In response, the role of coastal green infrastructure (CGI), natural or nature-based systems thatprovide coastal flood and erosion protection as well as multiple social, economic and environ-mental benefits, have started to gain attention (Gedan et al., 2010; Arkema et al., 2017; Sutton-Grier et al., 2018). Studies of Coops et al. (1996); Mol (2003); Feagin et al. (2005, 2009); Kochet al. (2009); Borsje et al. (2011); Anderson and Smith (2014); Mo¨ller et al. (2014); Spaldinget al. (2014); Wu and Cox (2015); Narayan et al. (2016); Ruckelshaus et al. (2016) and others in-dicate that CGI is an effective practice of protecting coasts from flooding and erosion. Althoughthese studies did not use the CGI terminology, their research concluded that CGI practices pro-tect coasts from flooding through reducing the wave energy by drag friction, reducing waveovertopping by eliminating vertical barriers, and absorbing floodwaters in soil (The HorinkoGroup, 2015; Ruckelshaus et al., 2016; Narayan et al., 2016; Arkema et al., 2017); and fromerosion through reducing wave transmission, increasing soil elevation through vertical accre-tion and binding soil properties (Shepard et al., 2011; Hettiarachchi et al., 2013; Spalding et al.,2014; The Horinko Group, 2015; Silva et al., 2016). Besides its coastal protection benefits, CGIenhances natural coastal processes, sequesters carbon, provides habitat for wildlife, increaseseconomic activities such as fishing and tourism, creates recreation opportunities, and improvesaesthetics of coastal communities (Davidson-Arnott, 2010; Barbier et al., 2011; Barnhill andSmardon, 2012; Sutton-Grier et al., 2014, 2015).Although CGI is now widely recognized for its value in coastal protection and providing variousother benefits, it is also considered to be vulnerable to changes in the environments they interactwith (Osland et al., 2015). For example, changes in the land use, water levels, and storm inten-sities and frequencies can have significant implications on the health and integrity of CGIs andthe services they provide (Khattabi and Bellaghmouch, 2009). Therefore, CGI’s coastal protec-tion benefits are not solely affected by the physical characteristics of and changes in the naturaland built environments, and the intensity of the hazards they are exposed to. CGIs adaptabilityto these changes can also significantly affect their coastal protection benefits. Especially withclimate change, and the associated increases in the global sea levels and acceleration of storm27intensities and frequencies, the vulnerability of CGI can potentially reduce, if not eliminate, itscoastal protection benefits (Duarte et al., 2013; Langridge et al., 2014; Osland et al., 2015).Hence, the type and extent of benefits acquired from CGI depend on the location it is imple-mented (Ruckelshaus et al., 2016). Because CGI and its interactions with coastal processes andbuilt environments are context-depended (Hanley et al., 2014) and vary spatially and temporally(Koch et al., 2009; Barbier et al., 2011; Feagin et al., 2015). Yet, despite this place-specificdependency, identifying where to utilize CGI in coastal areas has been largely absent from theliterature (Ruckelshaus et al., 2016). Filling this gap entails understanding where along thecoasts CGI can provide protection benefits, and where it is likely to be vulnerable to chang-ing environmental conditions. Considering CGI vulnerability, identifying coastal areas with thehighest potential CGI protection benefits is important because it allows for prioritizing restora-tion efforts and new CGI implementation. Such an approach can also make it easier for localand regional level governments to allocate funds and other resources to areas with the highestpotential of CGI benefits.Therefore, this study aims to answer the following research questions. What parameters can beused to identify CGI coastal protection benefits and vulnerability? What criteria can be used toorganize these parameters into CGI coastal protection and vulnerability indices? How can theseindices inform where CGI has greater (or lesser) promise for coastal protection on the basis ofthe degree of potential coastal protection benefits and the degree of vulnerability? The studyuses the Salish Sea region as a case study area to demonstrate the methods of this research.Chapter 2 is organized as follows. Section 2.2 provides a literature review of CGI, its coastalprotection benefits, and vulnerability. Section 2.3 describes the study area. Section 2.4 explainsthe methodological approach of this study. Section 2.5 provides and explains the results. Section2.6 discusses the implications of the findings and the limitations of the methodology. Section2.7 offers concluding remarks.2.2 Coastal green infrastructure (CGI)The green infrastructure terminology has been used by natural resource professionals (i.e.,coastal zone management) (Cooper and McKenna, 2008), planners, and engineers in the last40 years (Liquete et al., 2015). There are three main groups of green infrastructure: urban28green infrastructure, which deals mainly with water and stormwater management in urban ar-eas; watershed-based green infrastructure, which protects, fosters and connects networks ofgreen spaces and forests; and coastal green infrastructure, which refers to the practices that aimsto deal with flooding and erosion (The Horinko Group, 2015). This study focuses solely on thecoastal green infrastructure (CGI).Although CGI is a growing practice, its definition is not uniformly accepted and depends on thecontext that it is studied. Some of the most common definitions are as follows:Benedict and McMahon’s definition of CGI was “an interconnected network of natural areasand other open spaces that, (...), provides a wide array of benefits to people and wildlife”(2006, pg.1). Tzoulas defined CGI as “all natural, semi-natural and artificial networks of multi-functional ecological systems, (...), at all spatial scales” (2007, pg.1). Edwards et al. (2013)uses the term “blue infrastructure” to define coastal and riparian habitats that maintain coastalprocesses and ecological functions, and provide various ecosystem services. The EuropeanCommission (2013) defines CGI as natural and semi-natural areas that are strategically planned,designed and managed to provide various ecosystem services. The Conservation LeadershipCouncil’s 2015 report refers to CGIs as “nature-based systems and processes” (2015, pg.1).The Environmental Protection Agency (USEPA, 2015) and the US Army Corps of Engineers(Bridges et al., 2015) in the USA define CGI as natural areas and processes, or nature-based(designed and engineered) systems that mimic natural processes that provide coastal protection,habitat and other services. Recently, Soz et al., defined CGI as an approach that uses naturalprocesses to manage flooding while providing other ecosystem services (2016, pg. 1). Narayanet al., uses the term “nature-based defenses” and defines it as “existing coastal habitats withinwhich wave reduction has been measured” (2016, pg.1).Driven from these definitions, CGI in this study refers to natural or nature-based systems andprocesses that mimic dynamic coastal landforms, vegetations, and reefs to provide coastal pro-tection services as well as various ecosystem health, maintenance of natural processes and bio-diversity, and social and economic benefits to communities.Common types of CGI are:(1) dynamic coastal landforms such as sand/gravel/rocky beaches, barrier islands, and dunes(Barbier et al., 2011; Sutton-Grier et al., 2015; Ruckelshaus et al., 2016), including beach nour-ishment practices (Brown et al., 2016);29(2) coastal vegetation such as mangroves, eelgrasses, dune vegetation, salt marshes and kelpforests (Davidson-Arnott, 2010; Arkema et al., 2013; Duarte et al., 2013; Sutton-Grier et al.,2015; The Horinko Group, 2015; Ruckelshaus et al., 2016); and(3) reef systems such as mussel beds, oyster reefs and coral reefs (Gutie´rrez et al., 2011; Cheonget al., 2013; Sutton-Grier et al., 2015; Ruckelshaus et al., 2016).2.2.1 CGI coastal protection benefitsCGI’s coastal protection benefits are mainly two folds: flood protection and erosion protection.CGI protects coastal areas from flooding by reducing the fetch for the wind to form waves,absorbing wave energy through surface roughness and drag friction, leading to a reduction inwave height and velocity, and absorbing floodwaters through impermeable natural or engineeredsurfaces (Davidson-Arnott, 2010; Cheong et al., 2013; Spalding et al., 2014). Erosion protectionis achieved by maintaining or increasing surface elevation by trapping sediments and build-upfrom decaying vegetation on soil, and binding soil particles through vegetation roots (Barbieret al., 2011; Bryant et al., 2017).Flood protectionFonseca and Cahalan (1992) studied seagrasses and their role in reducing wave energy overa one-meter test section. They concluded that the percentage of average wave reduction byfour species of seagrasses was approximately 40% and wave energy reduction was significantlycut when water depths become higher or equal to twice the mean leaf length. Coops et al.(1996) found that through a 4m wide profile of wave tank, wave heights over areas without saltmarshes ranged from 71% to 129% more of the height of waves for areas with salt marshes.Mo¨ller et al. (2001) found that salt marshes reduced wave heights in all observations at rates27% to 98%. They also found that the total energy dissipation rates were on average 82% oversalt marshes compared to on average 29% on sand flats. In another study Mo¨ller and Spencer(2002) suggested that on average 92% wave height attenuation was obtained over a 310 mcoastal profile of salt marshes, and the first 10 m of the canopy provided the most rapid reductionin wave height. Later, Mo¨ller (2006) suggested that salt marshes could effectively reduce up to33% of wave height over a 10m coastal profile, depending on the salt marsh canopy height anddensity, and water depth conditions.30Loder et al. (2009) used numerical models for their study and found that a 400-km2 salt marsharea is effective to reduce less than 2 m of storm surge (35% to 70% surge decrease). Bradleyand Houser (2009) concluded that over a one m water depth range, wave heights wave heightdecreased on average by 30% after the first 5 m over 39 m, and decreased exponentially over theremainder of the bed. Similarly, Koftis et al. (2013) suggested a 35% reduction in wave heightsby seagrasses in an extensive experimental study. Anderson and Smith (2014) suggested thatartificial seagrasses attenuated the wave energy in all wave frequency conditions. John et al.(2015) suggested an exponential reduction in wave height over a 50 m long wave flume ofsubmerged artificial seagrass vegetation.Studies after Indian Ocean Tsunami of 2004 and Hurricane Katrina of 2005 suggested that pres-ence of wetlands and mangroves reduced the infrastructure damage, injuries and fatalities causedby extreme events (Chang et al., 2006; Barbier et al., 2013). Supporting the abovementionedstudies, Narayan et al. (2016)’s meta-analyses of CGI’s wave reduction suggested that CGI playkey roles in reducing wave heights, on average 35% and 71% reduction in wave heights, and theextent of coastal protection achieved varies with the location and local conditions.Erosion protectionCGI reduces erosion directly by reducing wave transmission, capturing sediments and indirectlythrough stabilizing soil properties (Feagin et al., 2009; Silva et al., 2016; Bryant et al., 2017;Mendoza et al., 2017). Coops et al. (1996) suggested that different types of CGI reduced erosionat different rates. Examining different types of salt marshes, they noted that P. austrulis saltmarshes showed higher aboveground biomass and reduced erosion by 82%, while S. lucustrissalt marshes reduced only 33%, compared to not vegetated areas. Piazza et al. (2005) foundthat coastal erosion was 0.08 ±0.02m/month at oyster reef present sites as compared to 0.12±0.01m/month at oyster reef absent sites. In another study area, Scyphers et al. (2011) foundthat oyster reefs successfully mitigated erosion by more than 40% over two years. Levin et al.(2007) suggested that the sand deposition, thus accumulation was much higher in areas withtaller coastal vegetation, compared to shorter vegetation.More recently, in a study investigating the dune erosion, KobayashiI et al. (2013) concludedthat wooden dowels and wide dune vegetation reduced the dune erosion by decreasing the waveovertopping and overwash rates. In a large-scale implementation and monitoring study, Stiveet al. (2013) suggested that sand nourishment will increase the width of the beach and will in-31crease the beach area approximately 200 ha in 20 years after implementation. In physical modelexperiments on the effects of vegetation on dune erosion, Figlus et al. (2014) suggested that thevegetation reduces eroded dune volumes and concluded that the vegetation root maturity plays asignificant role on reducing erosion. Martinez et al. (2016) found that vegetation effectively re-duces coastal erosion in three different storm conditions, through 24 experiments of two coastalprofiles. Silva et al. (2016) investigated the role of vegetation cover on the sediment movementalong two coastal profiles and concluded that vegetation reduces net coastal erosion regardlessof different wave and morphological conditions, and prevents shoreline retreat. Borsje et al.(2017) suggested that coastal erosion can be prevented and functions of the coastal areas andprocesses can be maintained with beach nourishment. Mendoza et al. (2017) stated that vegeta-tion effectively contributes to the resistance of the coastal profile during storms through trappingsediments and strengthening soil.2.2.2 CGI vulnerabilityThere is now numerous evidence on CGI’s role in coastal protection (i.e., Koftis et al. 2013;Anderson and Smith 2014; John et al. 2015; Silva et al. 2016; Arkema et al. 2017. At the sametime, there is a growing interest in using CGI for coastal protection (Ruckelshaus et al., 2016).It is widely recognized that CGI is more effective in specific contexts than others that its coastalprotection benefits can vary in depending on the context they are utilized. It is also recognizedthat CGI itself can be vulnerable depending on the context it is located in, and the rate andmagnitude of the environmental changes in that environment become too high for CGI to adaptto. Thus, this vulnerability can lessen CGI coastal protection benefits.In the past 50 years, around 25% to 50% of the coastal areas with essential landforms, vegeta-tions, and reef systems have been lost globally due to changing environmental conditions suchas increased pollution, changes in coastal land use, and climate change (Duarte et al., 2013;Feagin et al., 2015). Notably, the impacts of climate change such as increased frequency, dura-tion and intensity of storms, and rising sea levels have intensified the hazards CGI is exposed to(Johnson et al., 2012; Ghosh and Chaudhuri, 2015). However, CGI has response mechanismssuch as accretion or migrating along the coast to deal with these hazards (Feagin et al., 2015).What makes CGI vulnerable, however, are the external interventions that disable these responsemechanisms. These interventions are often in the form of extensive occupation of coastal areas,32where CGI has no or insufficient physical space to move and adjust (‘coastal squeeze’) (Feaginet al., 2005; Osland et al., 2015; Kirwan et al., 2016); construction of human-made structuresor human activities that intervene with the natural sediment flow to the coasts (Feagin et al.,2010); or rapid acceleration of sea level rise due to climate change, or local water levels due toengineering activities that change coastal profile (Duarte et al., 2013).Feagin et al. (2005), investigated the vulnerability of dune vegetation to rising sea levels in aspacial model on Galveston Island, Texas, USA. The study concluded that in the low sea levelrise scenario (0.09m by 2100), dune vegetation was able to fully develop and cope with theincrease in sea level. However, in the high sea level rise scenario (0.88m by 2100), only a thinstrip of dune vegetation was developed due to high stressor level. Moreover, in the high-risescenario, the thin dune vegetation was neither able to block winds nor accumulate new soil lay-ers. Feagin et al. (2005) also stressed that human developments along the coasts which restrictmigration of CGI landwards are key determinants of vulnerability. Khattabi and Bellaghmouch(2009) found in a simulation study that with a 0.5 m sea level rise a loss of 478.7 ha and witha 2 m sea level rise a loss of 1400 ha may be observed by 2100 in a wetland in the North Eastcoastal zone of Morocco.Similarly, Kirwan and Temmerman (2009) and Kirwan et al. (2010), investigated CGI’s surfaceaccumulation while facing sea level rise in modeling studies. Their results concluded that at sealevel rise rates more than 20mm per year; salt marshes in medium to high tidal ranges and sedi-ment concentrations survive, while salt marshes in low tidal ranges and sediment concentrationsfully submerge. They suggested that full submergence occurs approximately 30-40 years afterthe threshold sea level rise rates are exceeded. Later, Kirwan et al. (2016) suggested a thresholdrate of 10 to 50 mm/yr for the relative sea level rise and marsh survival, and vegetation areaswith less than 1 m tidal range and less than 20 mg/L sediment concentrations will be vulnerableto even the moderate rate of sea level rise. Kirwan et al. (2016) also suggested that gently slop-ing coastal profile enhances and fosters marsh expansion. They stated that even at high sea levelrise rates, marsh survival could be achieved, only if the vegetation is able to migrate inland andis not limited by natural or human-made barriers.Thorne et al. (2013) investigated the ability of a 309 ha tidal marsh area to increase its surfaceelevation over 13 years to keep up with the rising water levels using two elevation surveys. Theirresult concluded that 63% of the salt marsh area did not accrate at a rate that exceeded sea levelrise, and therefore drowned. They also stated that the long distance to sediment source was the33most significant factor in the lack of accretion. Mariotti and Carr (2014) also supported previousfindings, stating vegetation is likely to drown when sea level rise is fast, and access to sedimentis low. They suggested a 0.3 (mm/yr)/(mg/L) ratio of sea level rise and sediment concentrationis a threshold for vegetation drowning, and a sediment concentration of at least 50 mg/L wouldbe needed to prevent vegetation retreat even in cases where sea levels are stable.Literature shows that the extent of CGI benefits varies significantly depending on the geomor-phological, wave and CGI related characteristics of their environment. Similarly, the degree ofCGI vulnerability differs depending on the geomorphological, land use and CGI related charac-teristics as well as environmental change types and rates.2.3 Study AreaThe Salish Sea region is selected to identify areas with highest potential CGI. The Salish Sea isa body of water that encompasses south of British Columbia in Canada and north of WashingtonStates in the United States. The area includes the Strait of Georgia, the Puget Sound, and theStrait of Juan the Fuca (Figure 2.1).The Salish Sea region is an important international body of water that is home to many coastalcommunities with diverse economic, and built and natural environment characteristics. Thecoastal communities of the Salish Sea range from small towns with only a few hundred residentsand single sector economy, to large metropolitan areas such as Vancouver and Seattle withcomplex economies. The coasts range from low-lying sand flats to coasts with cliffs, from saltmarsh fields to dunes and rocky beaches. The variation of the characteristics of Salish Seacommunities provides a diverse setting for this study and demonstrates the spatial variation inCGI benefits and vulnerability in the region.Coastal communities with populations over 4000 were identified from both British Columbiaand the Washington States. 44 Coastal communities from British Columbia and 30 coastal com-munities fromWashington State were selected for this study. The unit of analysis of this regionallevel study is the Canadian Census Subdivision (CSD) units for British Columbia communities,which is the equivalent to a municipality, and the American Census City/Urban Growth Areas(UGA) for the Washington States communities, which is the incorporated city boundaries andunincorporated Urban Growth Areas. These units are comparable, and publicly accessible data34are available for both geographic units in the federal and provincial/state databases.Figure 2.1: The Salish Sea and the study area communities. Numbers are in no particular order.Besides its importance and diversity, the Salish Sea region is an important area for this studybecause the governments and communities in this region have a high level of interest in CGIprojects. Particularly in the Washington State, numerous county governments (i.e., Skagit andKitsap Counties), multi-organizational partnerships (i.e., Puget Sound Partnership) and environ-35mental non-profits (i.e., Northwest Straits Foundation) have been initiating projects that removehuman-made structures such as bulkheads and restore habitats and beaches. Besides, alreadyseveral important CGI initiatives such as Green Shores BC and Green Shores Program of Wash-ington State exist in the region and educate local governments, community groups, and water-front property owners in ways they can utilize CGI. Even though such initiatives are limited inBritish Columbia compared to Washington State, recent federal and provincial initiative such asgreen infrastructure grant program creates an increased push for CGI projects.2.4 MethodsThis study uses an indicator-based approach to identify areas with the highest CGI potential inthe study area. Indicator-based approaches were used, amongst others, in fields such as disasterrisk reduction (i.e. Cutter et al. 2003), emergency management (i.e. Flanagan et al. 2011), sus-tainable development (i.e. Tanguay et al. 2010) and coastal zone management (i.e. Martı´ et al.2007). CGI related indicators were often used along with other social, economic and institu-tional indicators to assess or compare vulnerability of places (i.e., Chang et al. 2015), or alongwith other natural indicators to assess the environmental vulnerability of coastal communities(i.e., Shaw et al. 1998; Gornitz 1991; Tibbetts and van Proosdij 2013).Using the indicator-based approaches for understanding CGI’s coastal protection benefits andvulnerability is relatively new in the CGI research as studies typically field studies or lab exper-iments. This approach has been mainly applied in cost-benefit analyses of CGIs (i.e., Narayanet al. 2016; Capotorti et al. 2017), rather than understanding coastal protection benefits of CGIsor their vulnerability to changes. However, the indicator approach provides a high-level frame-work that can be mapped and applied in other regions, and its findings can be easily understoodand interpreted by decision-makers (Hinkel, 2011; Chang et al., 2018).The methodology of this study comprises of the following four systematic steps: (1) literaturesearch to identify CGI studies; (2) content analysis on the selected studies to identify parametersused in the CGI literature; (3) developing indices based on the parameters identified in Step 2 toassess CGI coastal protection benefits and vulnerability to changing environmental conditions,and mapping; and (4) synthesizing the indices developed in Step 3 into a classification systemto identify areas where CGI can potentially be used as a coastal protection tool. Step 1 and 236address the research question “What parameters can be used to identify CGI coastal protectionbenefits and vulnerability?” Step 3 answers the research question “What criteria can be used toorganize these parameters into CGI coastal protection and vulnerability indices?” Lastly, step 4answers the research question “How these indices can inform where CGI has greater (or lesser)promise for coastal protection on the basis of the degree of potential coastal protection benefitsand the degree of vulnerability?”2.4.1 Literature reviewCGI studies have used various parameters in the way they measure CGI’s role in coastal pro-tection and its vulnerability. To identify these parameters, a literature search using the Webof Science, Google scholar and Jstor databases (1970-2015, cut off date December 2015) wasconducted to target references on CGI’s coastal protection benefits and vulnerability. Numerouskeywords were used in the literature review. These keywords were determined through the initialreview of the CGI literature and include the following: coastal protection, (coastal) green infras-tructure, nature-based protection, wave attenuation, shore stabilization, flood protection, erosionprotection, vegetation accretion, coastal habitats, coastal defenses, soft engineering, bioshields,and nature-based solutions.In total, 151 primary, secondary and grey literature references matched the search criteria andwere selected for review. Amongst the 151 references reviewed, 77 references were identifiedthat contain specific parameters related to CGI’s role in coastal protection and vulnerability, andtherefore were selected for this study.2.4.2 Content analysisA content analysis was conducted on the 77 references to identify the parameters used to mea-sure the CGI’s role in coastal protection and its vulnerability. These parameters were recordedand coded in groups to reflect larger themes. For example, parameters related to wave height,frequency, direction, and other wave features were grouped under ‘wave characteristics’ theme.Figure 2.2 displays the themes and the frequency of use in the 77 selected CGI references.The content analysis revealed that a large variety of parameters had been used in the CGI re-37search. For example, Mo¨ller et al. (2001); Mo¨ller and Spencer (2002); Mo¨ller (2006); Mo¨lleret al. (2014) studies used water depth, wave characteristics, and CGI characteristics to inves-tigate whether or not vegetation can reduce wave heights. On the other hand, Costanza et al.(2008) used CGI cover and monetary damages associated with storm surges to investigate floodprotection from CGI. Similarly, Kirwan and Temmerman (2009); Kirwan et al. (2010) investi-gated erosion reduction rates provided by CGI through sediment concentration, sea level riserate, tidal range, and CGI characteristics parameters; while Feagin et al. (2005, 2009) used sed-iment concentration, accretion, CGI characteristics, and wave/storm characteristics parameters.Figure 2.2: Themes used to measure CGI coastal protection benefits and vulnerabilityAs seen in Figure 2.2, the most commonly used theme is the CGI characteristics. 60 out of 77references reviewed included an indicator related to CGI characteristics. The CGI characteris-tics theme refers to CGI types, vegetation densities and length, surface roughness provided bydifferent types of CGIs and other specific features of CGI. The second most used theme is thewave characteristics, which includes measures such as wave length, height, frequency, period,direction and other wave characteristics. Both the CGI characteristics and wave characteristicsthemes have been primarily used to determine the degree of the interaction there could be be-tween waves and CGIs. The third indicator theme is the CGI cover. This theme refers to thearea or percentage of the CGI that exists at the coast where the CGI habitat exists, or the CGIpractices are implemented (i.e., beach nourishment area). This indicator theme was separatedfrom the green infrastructure characteristics because it does not directly refer to the propertiesof CGI; instead, it indicates the space they cover at the coast. This theme determines the stretchof the CGI that interacts with the coastal processes.38The sedimentation theme, which indicates the sediment concentration or the amount of sedi-ments deposited at the coast, was also commonly used in the CGI studies. They are mainly usedto investigate CGI’s vulnerability to changing environmental conditions because the amount ofsediment deposited at coast impacts CGI’s ability to accrete. Another important theme that isused in the CGI studies is the water depth. Kirwan et al. (2010), Koch et al. (2009), Mo¨lleret al. (1999, 2001); Mo¨ller and Spencer (2002); Mo¨ller (2006); Mo¨ller et al. (2014), Stone et al.(2005) and others suggested that the water depth is an important indicator for CGI’s coastal pro-tection benefit as it determines the depth CGI can interact with the wave and reduce its energy.These studies suggested that when the threshold CGI height/water depth is passed, CGI’s abilityto attenuate wave energy decreases because CGI can no longer interact with the waves.Following the water depth theme is the site morphology theme such as relief, slope and typeof coast (i.e., sand, gravel, and rock), and the accretion theme such as changes in soil elevationand organic decomposition. The site morphology theme is a crucial part of CGI studies sincethe features and position of the coasts can provide protection from wave action. The accretiontheme is often used to determine the vulnerability of CGI because CGI can accrete and increasetheir soil elevation to keep up with the rising water levels. Commonly, the sea level change andtidal range themes were used along with the accretion theme to investigate CGI vulnerability,as the speed and rate of the sea level change and associated changes in the tidal range impactCGI’s ability to accrate.Although not used widely in the literature in the past, soil properties, coastal land use, andmonetary damages themes also emerged through the literature review. The soil properties themedetermines how much water can be held in the soil in occasional flooding events and the risk oferosion at the coast. The coastal land use theme is used to indicate the extent of developmentthat needs protection, and how much space is available on the coast for green infrastructure toadapt to various water levels. The monetary damages after extreme events can also indicate thebenefits CGI provides as a defense mechanism at the coast. And lastly, the damages on humanlives after an extreme event can be used to measure the effectiveness of CGI.2.4.3 Development of the CGI indicesBuilding the CGI databaseACGI database was created using the themes identified above. The British Columbia andWash-39ington State spatial data databases were searched to gather publicly available data on the themesand corresponding indicators. Of the 13 themes, data were available for the following ninethemes: CGI characteristics, wave characteristics, site morphology, sedimentation, water depth,accretion, tidal range, sea level change, and coastal land use. These themes, 12 correspondentindicators, and their data sources are shown in the Table 2.1.Themes Indicator(s) SourcesCGIcharacteristics- Coastal vegetation (i.e., kelp,sea grass, salt marsh, dune vegetation)BC-PSZMSa and WA-SZIbWavecharacteristics- Wave exposure (i.e., exposed to veryprotected coasts)- Max. wave height (m)- Max. Wave fetch (km)BC-PSZMS, WA-SZI andThe Geomorphology ofPuget Sound BeachescSite morphology - Relief (m)- Coastal types (i.e., estuaries, flats,beaches, cliffs, and human-made coasts)CanCoastd, BC-PSZMS,WA-SZI, and WA Dep. ofTransportationeSedimentation - Sediment concentration at coast.A relative index of sediment abundance(abundant, moderate and sparse)BC-PSZMS and WA-SZIWater depth - Habitat zone (i.e., subtidal, lower tidal,intertidal, mid/high tidal, supratidal)BC-PSZMS and WA-SZIAccretion - Erosion/change (m/y) CanCoast and WA-SZITidal range - Tidal range (m) CanCoast and WA-SZISea level change - Sea level changes (cm) in the past 100yearsCanCoast and WA Depof EcologyfCoastal land use - Coastal land use. Green (5) to gray(1) scale refering to the use of the coastwhere green refers to mostly agriculturalor natural uses, and gray refers to mostlycommercial and infrastructure usesDMTI Spatial Inc g and WADep. of Natural ResourceshTable 2.1: CGI themes and indicators, and their data sourcesaPhysical Shore-Zone Mapping System for British Columbia, 2009bThe Washington State Shore Zone Inventory, 2006cTechnical Report (Finlayson et al., 2006).dCanCoast: A National-scale Framework for Characterising Canada’s Marine Coasts, 2013e Level Rise in the Coastal Waters of Washington State, 2008g each indicator, the coastal segments and corresponding data were aggregated to the unit ofanalysis of this study (CSDs and UGAs). The dominant features for each community were iden-tified and recorded. Although the data for sediment concentration was available it was incom-plete; thus it was not included in the study. The remaining 11 indicators collectively representthe composition of the coastal areas and the interaction between landforms, built environments,and coastal processes.Formatting and organizing the indicesThe indicators (n=11) from the CGI database were used to create the CGI coastal protectionand vulnerability indices. First, the indicators from Table 2.1 were assigned to either the CGIcoastal protection index and/or the CGI vulnerability index. Second, the rule-based method -if a, then b, where a is the property of an observation and b is the group it is assigned to - wasapplied to the indicators to rank them from very low (1) to very high (5), as shown in Table 2.3and Table 2.4. Third, the CGI coastal protection and CGI vulnerability indices were computed.The methodology used to compute the indices was adopted from that of Gornitz (1991), whichcombines data on indicators and ranks them from 1 to 5 to compute the “Coastal VulnerabilityIndex (CVI)”. The CVI method is defined by Gornitz (1991) as the square root of the geometricmean divided by the total number of variables.CVI =ra1⇥a2⇥a3⇥ ...⇥annFourth, the total range of computed values was divided into normal distribution quantiles, pro-viding four range groups. These groups were assigned to very low, low, medium and highcategories. The four range groups and corresponding categorical groups for both indices areprovided in the (Appendix A.1 and A.2). And lastly, the spatially linked data were mappedusing Geographic Information Systems (GIS).The CVI methodology has been used in many other studies investigating environmental vulner-ability and sensitivity such as Shaw et al. (1998), Gornitz (1991), Gornitz et al. (1992, 1994)as well as Thieler and Hammar-Klose (1999) but it has not been applied to the CGI coastalprotection research.412.4.4 Synthesizing the indicesAfter the CGI coastal protection and vulnerability indices were created, computed, and mapped,they were synthesized using a 2x2 matrix (Table 2.2) to identify coastal protection potential inthe study area. CGIs with low coastal protection benefits and high vulnerability; low coastalprotection benefits and low vulnerability, high coastal protection benefits and high vulnerability,and high coastal protection benefits and low vulnerability were grouped. The spatially linkeddata was mapped using GIS.CGI Coastal protectionVery low/low High/mediumCGI Vulnerability Very low/low Low potential High potentialHigh/medium Very low potential Medium potentialTable 2.2: CGI coastal protection potential matrix2.5 Results2.5.1 CGI coastal protection indexQuantitative and qualitative data on eight indicators were used to create the CGI coastal protec-tion index. These indicators are the relief, coastal types, coastal vegetation, habitat zone, waveexposure, maximum wave height, maximum wave fetch, and coastal land use. Each variable forthese indicators was assigned a rank from 1 to 5, where 1 represents very low, and 5 representvery high existing coastal protection benefits (Table 2.3). The coastal protection scale here doesnot represent an absolute very low to high protection benefits; rather it represents the relativeCGI protection benefits in the study area. In addition, these indicators assess the existing fea-tures of coasts, wave action, and land use, therefore the potential benefits claimed through thisindex is based on the existing conditions.As discussed previously, the wave energy is attenuated through the surface roughness of thecoasts and drag friction provided by the properties of coastal profile and coastal vegetation. The42indicators that address this interaction through CGI properties and therefore used in the coastalprotection index are the relief, coastal types, and coastal vegetation. Habitat zone indicatoraddresses the magnitude of this interaction. While wave indicators such as wave exposure,maximum wave height, and maximum wave fetch reflect the wave energy that is to be attenuatedat the coast. Lastly, coastal land use reflects the extent of the development that needs protection.Very low Low Moderate High Very highINDICATORS 1 2 3 4 5Relief (m) 0-5 6-10 11-20 21-30 >30Coastal types Sand,gravel andmudflatsHuman-madeEstuaries,beaches,and dunesRockybeachesRockycliffs andplatformsCoastalvegetationNovegetationKelpforestsSea grasses Marsh ordune veg.MixedvegetationHabitat zone Sub-tidal Lower tide Inter-tidal Higher tide Supra-tidalWave exposure Exposed Semi-exposedSemi-protectedProtected Very-protectedMaximum waveheight (m)> 6.1 5.1 - 6.0 4.1 - 5.0 2.1 - 4.0 < 2.0Maximum wavefetch (km)> 200 200 - 150 150 - 100 100 - 50 < 50Coastalland useMostlygrayMixed greenand gray(commercial)Mixed greenand gray(residential)Mostlygreen withagricultureMostlygreenTable 2.3: CGI coastal protection indexThe relief indicator in the CGI coastal protection index represents wave attenuation throughcoastal slope. Coops et al. (1996), Nicholls (2004), Wamsley et al. (2009) and Barbier et al.(2011), highlight the importance of coastal slope as a controlling factor for wave attenuationat coast. High relief increases the wave attenuation; therefore provide more coastal protectionbenefits. Coastal types attenuate wave energy through providing rough surfaces for the waveto go over (Mo¨ller, 2006; Loder et al., 2009; Wamsley et al., 2010; Mo¨ller et al., 2014). This43indicator shows the dominant coastal types for the study area communities. Coasts with sand,gravel, and mudflats provide less roughness compared to estuaries, beaches, dunes, and rockyplatforms; therefore, provide less coastal protection. Surface roughness increases with slopedbeaches, rocky beaches, and cliffed coasts. Human-made structures, although provide a degreeof protection, often cause more damage due to their vertical alignment, wave over-topping, andparts of the structure becoming loose over time or with wave energy.Coastal vegetation attenuates waves through drag friction the vegetation stems and leaves pro-vide. Different vegetation types provide different drag friction because of density, length, andother structural differences (Koch et al., 2009; Anderson and Smith, 2014; Mo¨ller et al., 2014).Kelp forests are dense, and the vegetation has long stems, yet their interaction with waves areoften limited because they are often located deeper than the wave break zones. As discussed inthe literature review, seagrasses can effectively provide wave attenuation but their ability is of-ten limited because of their short stems and less dense cover, and the wave height and the waterdepth ratio discussed previously(Mork, 1996). Salt marshes and dune vegetation provide higherdegrees of wave attenuation because they have higher degrees of interaction with the waves(Bradley and Houser, 2009; Manca et al., 2012). The presence of multiple types of vegetationincreases the drag friction at the coast and improves wave attenuation (Mo¨ller et al., 2014).Habitat zone refers to the water depth that the dominant CGI is located. The literature suggestsdecreases in vegetation elevation or increases in depth results in less attenuation (KobayashiIet al., 1993; Nicholls, 2004; Mo¨ller, 2006; Loder et al., 2009; Anderson and Smith, 2014; Johnet al., 2015). CGIs at the inter-tidal and supra-tidal zones are more efficient in dissipating waveenergy than CGIs at the sub-tidal zones due to the relationship between vegetation height andwater depth.Wave exposure indicates the frequency and intensity of wave action at the coast. This indicatorrefers to the tear stress CGI is exposed to. When the wave exposure increases, CGI’s coastalprotection benefits decrease. The maximum wave height indicator refers to the one-year highestwave height. The literature suggests that CGI is more effective in dampening the energy of small(0-2m) to moderate height waves (2-4m) (Wamsley et al., 2010; Duarte et al., 2013; Mo¨ller et al.,2014; John et al., 2015). Therefore increases in wave height lower CGI’s coastal protectionbenefits. The maximum wave fetch indicator refers to the water surface area available for thewind to form waves. High wave fetch indicates higher and stronger waves (Bradley and Houser,2009; Shepard et al., 2011), therefore reduces CGI’s coastal protection benefits.44Lastly, the coastal land use indicates the extent of development at the coast that is at risk andneeds protection. More development and infrastructure at coast increases the assets that areexposed to flooding whereas green and agricultural lands reduce the exposure and can furtherhelp CGI with attenuating wave energy and absorbing floodwater.Based on this classification, CGI’s coastal protection benefits are expected to be high where thecoastal relief is high; the coastline consists of material that provides high roughness, and vege-tation that attenuates wave energy; the habitat zone is in the higher sections of the tidal range;wave exposure, wave height and fetch are low; and the land use at the coast consists mostly ofgreen spaces. After the indicators were ranked and applied to the study area communities, theCGI coastal protection benefits index was computed and mapped.Figure 2.3 shows that the distribution of CGI’s coastal protection benefits at present is not homo-geneous among the Salish Sea communities. The CGI in British Columbia has higher coastalprotection benefits in general than the CGI in the Washington States. CGI in the 59% of thecommunities in BC and 37% of the communities in WS have medium to high coastal protectionbenefits. This is partially the result of the more dense and intensive occupation of the coastalareas in the Washington State, which impacts the coastal types and coastal land use indicators.Approximately 17% of the total Washington State coastline is hardened with human-made struc-tures (Gittman et al., 2015), where only about 3% of the British Columbia coastline has human-made structures, reflecting the differences in degree of human intervention in both coastal areas.Examining Figure 2.3 shows some interesting patterns in the region. For example, CGIs in bigurban centers such as Vancouver, Seattle, Victoria, and Tacoma have different coastal protectionbenefits: very low, medium, low and high, respectively. The results of the coastal protectionindex show that even in communities with extensive areas of CGIs such as Squamish and PortTownsend, the implications of high wave fetch, height and exposure reduces CGI’s coastal pro-tection benefits (Appendix A.1). Similarly, in communities with low wave exposure such asPoulsbo, coastal types undermines the role of the CGI in providing coastal protection benefits.Coastal protection benefits of CGI is one part of identifying areas with the highest potentialcoastal protection benefits. As mentioned in the previous sections of this chapter, the vulnera-bility of CGI to changes in the environment needs to be considered as well.45Figure 2.3: Distibution of the CGI coastal protection benefits in the Salish Sea2.5.2 CGI vulnerability indexQuantitative and qualitative data on seven indicators were used to create the CGI vulnerabilityindex. These indicators are the relief, tidal range, habitat zone, sea level change, erosion change,46wave exposure, and coastal land use. Each variable is assigned a rank from 1 to 5. Differentfrom the CGI coastal protection index, here 1 represents very low, which is a positive constructand 5 represent very high existing vulnerability of CGI (Table 2.4). The vulnerability scale alsodoes not represent an absolute very low to high vulnerability; rather it represents the relativeCGI vulnerability in the study area.Very low Low Moderate High Very highINDICATORS 1 2 3 4 5Relief (m) >30 21-30 11-20 6-10 0-5Tidal range >6.0 4.1-6.0 2.0-4.0 0.5-1.9 <0.50Habitat zone Supra-tidal Higher tide Inter-tidal Lower tide Sub-tidalSea level change(cm/100 years)<-50 -50 to -20 -19 to +20 21 to 40 >40Erosionchange (m/y)>+0.1 0 -0.1 to -0.5 -0.6 to -1.0 >-1.0WaveexposureVeryprotectedProtected Semi-protectedSemi-exposedExposedCoastalland useMostlygreenMostlygreen withagricultureMixed greenand gray(residential)Mixed greenand gray(commercial)Mostly grayTable 2.4: CGI Vulnerability IndexIt has been discussed in the literature review section that factors such as coastal characteristics,coastal land use, and risks induced by climate change such as sea level rise (Duarte et al., 2013)are associated with CGI vulnerability. The indicators that address the coastal characteristicsfactors are the relief, tidal range, and habitat zone. Sea level rise, erosion and wave exposureindicators reflect the risks induced by climate change. The coastal land use reflects the intensityof the development at the coast which limits the amount of space CGIs have to migrate upland.The relief indicator was used in the coastal protection index to reflect the slope and steepness ofthe coastal areas, which can help to attenuate wave energy. In the vulnerability index, the reliefindicator reflects inundation risks throughout coastal slope. Low relief indicates larger areaswith low elevation that is under inundation risk, therefore increases the vulnerability. The tidal47range indicator shows the zone of the coast that is frequently inundated. It impacts the habitatzone and sediment deposition zone at coasts. The literature suggests that vegetation at the mesoand macro-tidal range can accrete and deal with rising sea levels better due to the availabilityof this sediment deposition zone (Morris et al., 2002; Fitzgerald et al., 2008; Craft et al., 2009).The habitat zone indicator is related to the zone CGIs are in the tidal range. CGIs at the lowertidal zones are more vulnerable to the changing conditions because of the availability of thesediments throughout the tidal range is lower in the lower-tidal zones (Craft et al., 2009; Kirwanand Temmerman, 2009; Davidson-Arnott, 2010).The sea level change indicator shows the changes in the water levels over the past 100 years.It includes sea level rise and vertical land movement adjustments. The negative values indicateland uplift, thus decrease in the sea levels, where the positive values indicate increases in thesea levels. Many studies discussed in the literature review section suggested that rapid rates ofsea level rise will likely to cause CGI drowning (Feagin et al., 2005; Kirwan and Temmerman,2009; Kirwan et al., 2010; Mariotti and Carr, 2014). The erosion change indicator refers to thestability of coastlines. Accretion is one of the most critical mechanisms CGIs use to deal withenvironmental stressors such as rising water levels (Feagin et al., 2015). The positive valuesindicate accretion, therefore low CGI vulnerability where the negative values indicate coastalerosion, thus high vulnerability. The wave exposure indicator in the CGI vulnerability indexrefers to the frequency and duration of inundation. High wave exposure can result in the loss ofCGI due to tear stress, therefore increases vulnerability (Gedan et al., 2010).The coastal land use indicator in the vulnerability index reflects the human development andactivities at the coast that can confine CGI to a small zone, affecting its ability to move andadapt to changing conditions (Feagin et al., 2009). Greener, less developed coastal areas providemore room for CGI to migrate upland, therefore reduce CGI vulnerability. Gray and denselydeveloped coastal areas create a ‘coastal squeeze’ (Osland et al., 2015; Kirwan et al., 2016),therefore increase CGI vulnerability.Based on this classification, CGI’s vulnerability is expected to be high if the coastal relief andtidal range are low; the habitat is located at the low parts of the tidal range; sea level changeover the 100 years is high; the coast is erosional and exposed to high wave action; and land useat the coast is densely occupied with mostly of commercial and residential structures. After theindicators were ranked and applied to the study area communities, the CGI vulnerability indexwas computed and mapped.48Figure 2.4: Distibution of the CGI vulnerability in the Salish SeaFigure 2.4 shows that the distribution of CGI vulnerability at present also varies significantlyin the Salish Sea. CGI in British Columbia is less vulnerable in general than the CGI in theWashington States. CGI in the 61% of the communities in BC and 33% of the communitiesin WS have low to very low vulnerability to changing environmental conditions. Relativelylower relief values of Washington State communities have a significant contribution to higher49CGI vulnerability. Besides, higher rates of net sea level rise in the Washington State, which ismainly due to the subsidence of the land, contributes to the CGI vulnerability. Moreover, themost developed and dense coastline of Washington State also increases the vulnerability of CGI.Examining Figure 2.4 also shows some interesting patterns in the region. Amongst the largeurban centers, the CGI in Vancouver and Seattle have a high and medium vulnerability, respec-tively. CGI vulnerability in Vancouver is due to tidal range, wave exposure, and coastal landuse, while in Seattle CGI vulnerability is caused by sea level rise and coastal land use. In otherlarge urban centers such as Victoria and Tacoma, CGI has a low vulnerability (Appendix A.2).2.5.3 Synthesizing the indicesThe results of the CGI coastal protection and vulnerability indices were synthesized to identifyCGI’s coastal protection potential in the region. A 2x2 matrix (Table 2.2) was used to organizethe indices into four categorical groups. These groups were defined and their descriptions areprovided as follows.• “Very low potential”: CGI vulnerability is high, and coastal protection benefits are low.The “very low potential” area can be defined as the worst area for CGI in the study area. CGImay not be the best course of coastal protection and climate change adaptation action for thecommunities in this area. Rather, hybrid uses of hard structures and CGIs can be explored.Where applicable, adjusting the hard structures in a way to accommodate new CGI productioncan help to utilize other benefits of CGI other than coastal protection (i.e., adding texture toconcrete seawalls can foster vegetation and oyster population). Also, communities in the “verylow potential” area can focus on limiting coastal development to non-essential uses.• “Low potential”: CGI vulnerability is low but coastal protection benefits are low as well.The communities in the “low potential” area have low potential to utilize their CGI in the studyarea. Therefore, they can explore ways to increase their CGI coastal protection benefits sincethe CGI vulnerability is low. These communities can focus their efforts on actions such asrehabilitating coastal vegetation or creating new habitats in the riparian areas to increase coastalprotection opportunities. Where there are already hard protection structures, the hybrid uses ofCGI and adjusting existing structures to allow the creation of new CGI can be explored.50• “Medium potential” : CGI vulnerability is high, but coastal protection benefits are highas well.The communities in the “medium potential” area have a higher potential to utilize their CGIcompared to the very low and low potential areas, but their efforts should focus more on reducingthe CGI vulnerability. This could be done by limiting land use and developments and removingbarriers at the coast to ensure that CGI has space to move upland and adjust to changes. Also,beach nourishment and other rehabilitation strategies such as replanting vegetation can be usedto reduce CGI vulnerability. Site-specific investigations can look into replacing existing hardstructures with CGI, where it is not safe or feasible to replace hard structures, hybrid uses ofCGIs and hard structures can be explored.• “High potential”: CGI vulnerability is low, and protection benefits are high.The “high potential” area can be defined as the best area for CGI in the study area. The com-munities in the “high potential” area have the greatest potential to utilize their CGI for coastalprotection as the CGI vulnerability is low. These communities can focus their efforts into site-specific investigations to outline how and where in their coastlines CGIs can be incorporatedin the community’s sea level rise adaptation strategies and flood management practices. Thecommunities in the “high potential” area can undertake habitat enhancement, hard structureremoval, and CGI replacement projects. Also, they can hold public workshops to inform water-front homeowners on how to implement nature-based solutions in their properties.Figure 2.5 shows that the big population centers in British Columbia, such as Vancouver, Rich-mond, Victoria, and Saanich fall under “very low potential” and “low potential” areas, sug-gesting CGI may not be the most appropriate tool for coastal protection in these communities.Hybrid CGI solutions with hard structures can be used to mitigate the environmental impacts ofhards structures. Communities surrounding the large urban centers mostly fall under “mediumpotential” and “high potential” areas in British Columbia, indicating that the British Columbiacommunities in the Salish Sea region have high potential to utilize CGI for coastal protection.About 59% of the communities in British Columbia are either “medium potential” (18%) or“high potential” (41%) areas. On the other hand, only about 36% of the communities in theWashington States fall under “medium potential” (16%) or “high potential” (20%) areas. Com-munities in Washington State, particularly in the south of the Puget Sound, have low potential toutilize their CGI. Unlike in Vancouver, the CGI in Seattle falls under “medium potential” area,51and communities surrounding Seattle has lower CGI potentials.Figure 2.5: Distribution of the CGI coastal protecton potential in the Salish Sea522.6 DiscussionHard structures have been the preferred coastal protection method for centuries. Klein et al.(2001) suggests that the more tangible and easy to visualize the nature of hard structures makesthem more appealing to decision makers. Yet, hard structures provide a false sense of security astheir vertical interaction with waves create wave over-topping, resulting in damage and flooding.The static and maladaptive nature of hard structures contributes to their structural inefficiencyfor coastal protection. The costs associated with the implementation and maintenance of hardstructures create a significant economic burden on communities. In addition, the environmentalimplications of hard structures on coastal processes, ecosystem health and integrity, and wildlifehave been significant. There have been growing concerns over the economic, physical andenvironmental implications of hard structures.As an alternative, CGI has been considered as a viable coastal protection method for the last fewdecades, and a growing number of studies have provided evidence of CGI’s coastal protectionbenefits. Although, it should be noted here that most of these studies were conducted beforethe use of natural assets as coastal protection methods were called CGI. These studies havepaved the way to wide-range consideration of CGIs both in research and practice. Especiallyafter disasters such as Hurricane Katrina and Sandy, there has been a growing interest fromgovernments to invest in nature-based coastal protection methods (Sutton-Grier et al., 2015).However, even though CGI has started to gain popularity as a coastal protection method in theliterature, compared to the hard structures the extent of CGI benefits in different environmentalconditions has not yet well established. Also, how vulnerability can limit, if not dimish CGIcoastal protection benefits has not been studied.Increasingly, federal, state/provincial and regional government documents on sea level rise adap-tation and coastal flood management, in general, have started to include the use of CGI ratherthan or in addition to traditional hard structures. These high-level guidance documents do notalways foster the implementation of CGI because they do not outline where and where not CGIcan be potentially useful in providing coastal protection benefits. This study fills this gap byproviding a methodological approach and a regional level assessment of CGI coastal protec-tion benefits and vulnerability. This paper uses the 74 most populated coastal communities inthe Salish Sea region to demonstrate the methodological framework. The contributions of thisstudy can be summarized in four main points.53First, investigating where CGI can yield high coastal protection benefits while considering itsvulnerability is a novel and actionable approach. Traditional CGI studies study specific sites toestimate coastal protection benefits of the CGI at that location. This approach is useful to set anunderstanding of the rate of wave attenuation or accretion that can be achieved by CGI in thatspecific environment. Most of these studies often disregard the vulnerability CGI may be expe-riencing or could experience in future, yet his vulnerability can drastically alter potential coastalprotection benefits. Including a vulnerability framework helps to identify CGI that has potentialto provide coastal protection benefits but is facing high vulnerability. For example, communitiesunder “medium potential” area are characterized as where the potential CGI coastal protectionbenefits are high, but vulnerability is high as well. These communities should use caution indetermining where to implement CGI, and whether they can provide environmental conditionsor interventions that reduce CGI vulnerability. A misplaced CGI for coastal protection purposescan be a costly mistake communities should avoid.Second, besides the CGI coastal protection potential areas, the indices developed in this studycan be used independently and still be useful. The methodology and the indices used in thisstudy can be used to investigate CGI in multiple scales, from individual property level to na-tional level, and in various coastal areas around the world. The indicators of the indices can becustomized to reflect specific features of different study areas. For example, mangroves and/orreef systems can be incorporated in the coastal vegetation indicator, if the study area is in thetropical and the lower latitudes of the sub-tropical coastlines. In addition, more indicators canbe added to the indices, if data is available. For example, sediment concentration indicator canbe included in the vulnerability index, since it has been deemed very important for determiningCGI vulnerability but was not included in this study due to missing data.Third, the results of a regional level CGI assessment can be a valuable contribution for multiplelevels of governance in the region. The relative nature of scale used in both indices and thefinal synthesis of the indices could inform management and operational issues, and resourceand funding allocation. These actions can aim to reduce vulnerability and increase coastal pro-tection benefits through projects such as coastal rehabilitation, beach nourishment, and others.This level assessment would also help to prioritize such decisions by highlighting areas withhigh CGI coastal protection benefits but high vulnerability (“medium potential”), or areas withlow CGI vulnerability but low coastal protection benefits (“low potential”). Moreover, a re-gional level CGI assessment can help protect essential coastal habitat in some places and can54help the creation of new ones in others, therefore contributes to the overall regional environ-mental sustainability. It should be noted that the management and operational recommendationsprovided in this study are preliminary. More detailed studies would be needed to explore specificprioritization and conservation and rehabilitation actions.Fourth, this study can help regional knowledge sharing and collaboration amongst communities.The study area consists of two countries and state/provincial governments, making it difficult toimplement similar measures. However, communities in similar CGI typologies can neverthelessshare knowledge, resources and best practices on their coastal flood management strategies andsea level rise adaptation programs. For example, “very low potential” and “medium potential”communities can learn ways to reduce CGI vulnerability from “low potential” and “high po-tential” communities. Information sharing on habitat rehabilitation or relocation to change theexisting habitat zone, beach nourishment and planting more vegetation to reduce erosion, andland use practices to ensure suitable habitat for CGI can be beneficial for communities seekingto reduce their CGI vulnerability. Besides, neighboring communities with similar CGI typolo-gies can collaborate on projects. Especially for smaller communities with limited resources,knowledge sharing and opportunities for collaboration can provide various economic, socialand institutional benefits.Besides its contributions, the limitations of this research can be summarized as follows. It hasbeen argued that a significant contribution of this research is the consideration of CGI vulnera-bility while investigating its coastal protection benefits. However, this work does not explain theextent of the vulnerability impact on the protection benefits, nor the magnitude of the impact.The vulnerability of CGI can have distinct manifestations depending on the existing natural andbuilt environment conditions, and the type(s) of CGI present. Therefore, CGI vulnerability canhave various implications for the coastal protection benefits of CGI in different locations. In-vestigation of the extent and magnitude of CGI vulnerability on coastal protection benefits isbeyond the scope of this study.Besides, the data availability and gaps, common issues of indicator-based models, have re-stricted the indicators used in the creation of the indices. Amongst the 13-parameter themesidentified, only eight themes and 11 correspondent indicators were used in this research. Someimportant themes such as sedimentation or CGI cover were not included in the analysis becauseof missing or incomplete data. This limitation has direct impacts on the results as the numberof indicators and values of indicators are properties of the equation used to compute the indices.55However, it should be noted here that Gornitz et al. (1994) suggested that the formula used inthis paper was relatively insensitive to variations and was able to produce usable results whenchanges occur in the variables.The benefits of a regional level CGI assessment were discussed above. It is important to notethat the high-level nature of regional assessments makes it challenging to capture variationsat the local level. While there is consistent evidence in the literature on the positive coastalprotection role of CGI, for a given environmental, morphologic and biological condition, CGImay not be the best course of action (Wamsley et al., 2010) for every segment of a community’scoastline. The local variations influence CGI’s functionality, and the regional scale of this studyprevents understanding the local variations in the study area. For example, this study identifiedlow coastal protection potential of CGI in WA; however, a number of governmental and non-governmental initiatives that have been developing small to large scale CGI projects in WAcoastlines over the last two decades. The project objectives may be different from the provisionof coastal flood and erosion protection, but the disparity between the results of this chapter andthe current CGI practices in WA highlights that local variations are significant to understandwhere CGI projects can be implemented. More detailed studies focusing on coastal segmentsthat can identified through common features (i.e., bare portions of a beach can be a coastalsegment, when the vegetation starts, the vegetated portion can be treated as a new segment)are needed in order to assess the localities and highlight where in each communities CGI canyield the highest protection benefits. Lastly, the findings of this study should be considered inthe relative context of the Salish Sea study area as they do not reflect an absolute low or highpotential areas.2.7 ConclusionThis study develops a methodology to identify areas where CGI has more significant (or lesssignificant) promise for coastal protection. It does so by incorporating potential coastal protec-tion benefits and vulnerability to changing environmental conditions, using an indicator-basedapproach. This study provides a methodology that could help regional and local governments indecision-making for flood management and sea level rise adaptation. Moreover, it can facilitateknowledge sharing and collaboration within a region.56The methodological approach presented in this study and the findings of the research highlightsthe need for further research. A potential area of research includes identifying alternative ap-proaches to indicator-based methods. Lying between policy, practice, and research what othermethods can produce outcomes that are easy to understand and implement? In addition, more re-search is needed to deal with data gaps effectively. Moreover, comparative studies investigatingthe results of different computation methods can help in understanding both the shortcomingsand benefits of the Gornitz’s CVI methodology used in this paper. Another research area in-cludes the local level application of the methods and indices of this work. For example, can asimilar approach be used in the local scale to identify areas of highest CGI potential in a com-munity’s coastline? Advancing the CGI research in various scales and perspectives would bevaluable for its implementation for coastal protection.57Chapter 3Resilience-based evaluation of the localtrade-offs between coastal greeninfrastructure and other sea level riseadaptation strategies3.1 IntroductionThe consequences of sea level rise pose significant threats to coastal communities (Muis et al.,2015). The local impacts of sea level rise, however, vary greatly depending on the characteristicsof coastal areas, and the sea level rise adaptation strategies in place (Michener et al., 1997). Theadaptation strategies (protect, accommodate, avoid, (managed) retreat, do nothing and offense)aim to reduce or eliminate the impacts of sea level rise and community exposure (Catenacci andGiupponi, 2013). Under the protect strategy, coastal green infrastructure (CGI), the natural andnature-based processes that protect coasts from flooding and erosion (Narayan et al., 2016), haverecently started to gain significant attention as an adaptation measure (Ruckelshaus et al., 2016).However, it is recognized that CGI may not always be the most appropriate adaptation optionfor a community depending on the local characteristics. This is because CGI has different ben-efits, impacts and resilience contributions at different local contexts (Ruckelshaus et al., 2016).It entails different local trade-offs. The local trade-offs are not unique to CGI or the protectstrategy, but are shared amongst all strategies (Catenacci and Giupponi, 2013). Thus far, thereis limited research on how CGI and other adaptation strategies fit in the biophysical, environ-mental, economic, institutional and social environments of communities. The local trade-offsbetween CGI and other adaptation strategies are typically not considered, and decision-makersare often in the dark when deciding on one course of action over another one.58CGI provides a sustainable, adaptable and multi-functional adaptation to sea level rise. Itplays important roles in coastal protection, provision of ecosystem services, and contributingto the overall community resilience and well-being (Naumann et al., 2011; Narayan et al., 2016;Arkema et al., 2017; Sutton-Grier et al., 2018). CGI’s effectiveness, benefits, and contribution tolocal resilience vary significantly with the biophysical, environmental, economic, institutionaland social environments of communities (Langridge et al., 2014).Besides CGI, the effectiveness and benefits of other adaptation strategies also depend on thelocal characteristics of communities (Catenacci and Giupponi, 2013) because each communityfaces different levels of risks, and have different capacities to undertake adaptation actions.There are local trade-offs between CGI and other adaptation strategies (Catenacci and Giup-poni, 2013; Oddo et al., 2015) and decision-makers need to consider these trade-offs to makeinformed decisions on adaptation actions. Most of the existing assessment frameworks focuson strategy impacts on one aspect of the local communities such as the cost implications or theenvironmental impacts (i.e., French 2006; Hino et al. 2017; Schubert et al. 2017; Onuma andTsuge 2018). There remains a gap in holistically assessing the local biophysical, environmental,economic, institutional and social trade-offs between different strategies.There is an increasing demand for understanding in which contexts CGI is a meaningful andsustainable sea level rise adaptation measure (Langridge et al., 2014). Consequently, there isa growing need to understand the local trade-offs between CGI and other adaptation strategies(Catenacci and Giupponi, 2013). In addition, there is a gap in the methods and tools that wouldhelp decision-makers assess these trade-offs. Facilitating a sustained adaptation needs toolssuch as frameworks, indices, or scorecards that would help decision-makers understand andevaluate a wide range of impacts of different strategies and prioritize community values (Nelsonet al., 2007; Little and Lin, 2017; Garner and Keller, 2018). Therefore, this chapter presentsan evaluation framework developed to investigate the local trade-offs between CGI and othersea level rise adaptation strategies, applied in a case study community in British Columbia.It provides a methodological process and an assessment tool to help decision-makers identify‘win-win’ solutions and understand how different strategies can interact with coastal processes,natural and built environments, and economic, institutional and social factors.The chapter is organized as follows: Section 2 provides a background literature review includ-ing the role of the resilience perspective in adaptation, sea level rise adaptation strategies andCGI as a sea level rise adaptation measure, and adaptation strategies evaluation methods. Sec-59tion 3 explains the methodological approach and the corresponding research activities. Section4 presents the results of this study, following the format of the methods section. Section 5discusses the research findings, limitations and contributions to the literature, and Section 6provides concluding remarks.3.2 BackgroundThis section provides a review of the key concepts and literature that are used to shape theresearch design, methods, and research activities of this chapter. These concepts and literatureare the resilience perspective in adaptation, sea level rise adaptation strategies, CGI as a sealevel rise adaptation measure, and adaptation strategies evaluation methods and concepts.3.2.1 Considering the resilience perspective in adaptationAdaptation to sea level rise refers to the planning, policy and engineering strategies and mea-sures that are in place to minimize, if not eliminate, the impacts of sea level rise. Adaptationallows communities to identify development trajectories to prepare for the impacts of sea levelrise (Glavovic, 2014; Gregg et al., 2015) and to exploit opportunities posed by changing environ-mental conditions (Glavovic, 2014; Shayegh et al., 2016). The resiliency of coastal communitiesrefers to community capacity to cope with and respond to the adverse impacts of sea level rise(Dolan and Walker, 2006), and to self-organize and learn to preserve natural, economic, socialand institutional functions (Klein et al., 2003). The resilience perspective refers to the shift inapproaches from a static response to a disturbance, to dynamic management of the system as awhole before, during and after a disturbance (Folke, 2006). Using the resilience perspective asan objective of adaptation makes communities more robust in dealing with the implications ofsea level rise (Gersonius et al., 2016).As an objective for adaptation, the resilience perspective addresses the dynamism and complex-ity of coastal areas. Despite significant scientific advancements, there is still considerable un-certainty in the sea level rise projections (Little and Lin, 2017). Coupled with complex coastalprocesses such as tides, waves, currents, storm surges, and sedimentation, predicting the ex-act implications of sea level rise, and adaptation to it are challenging (Cazenave et al., 2014;60Shayegh et al., 2016). Sea level rise adaptation, therefore, has to account for uncertainty andbe adaptable to change (Glavovic and Smith, 2014). Besides its impacts on built and naturalenvironments, other factors are also affected when sea levels are rising, such as land use, envi-ronmental integrity, local economy, and evacuation and emergency responses (Gersonius et al.,2016). Therefore, sea level rise adaptation has to address diverse factors and be integrated. Theresilience perspective brings multi-functionality to sea level rise adaptation. Strategies and mea-sures used for adaptation are diverse and have different motivations (Cooper, 2016). They canprovide various additional environmental, economic, institutional and social benefits (Cooperand Pile, 2014; Wamsler et al., 2016). Therefore, sea level rise adaptation has to consider theco-benefits of strategies and measures. Lastly, the resilience perspective adds a learning domainto the adaptation (Wardekker et al., 2010; Cinner et al., 2018). The field and practice of adapta-tion are growing, and new knowledge is being created rapidly. Communities around the worldare implementing various adaptation actions and monitoring their effectiveness while dealingwith rapid change and uncertainty. Communities’ ability to learn from each others’ experiencesand to absorb new knowledge on climate change and adaptation are critical for improving re-silience (The William D. Ruckelshaus Center, 2017; Cinner et al., 2018). Thus, a continuouseffort for learning has to be a part of sea level rise adaptation.3.2.2 Sea level rise adaptation strategiesThere are no one-size-fits-all approaches available for adaptation (Eisenack et al., 2014). Nonethe-less, actions for sea level rise adaptation are commonly grouped under several diverse strategiesreflecting various motivations (Cooper, 2016). The British Columbia Sea Level Rise AdaptationPrimer (2013) identifies protect, accommodate, avoid, retreat, do nothing and offense as mainadaptation strategies and defines them as follows. The protect strategy uses the THSs or CGI toprotect people, assets, and infrastructure from sea level rise. The accommodate strategy imple-ments adjustments in the existing infrastructure and retrofits structures to adapt to changes. Theavoid strategy prevents new development from flood-prone areas. The retreat strategy phasesthe withdraw from flood-prone areas and relocates private or public assets in low-risk areas. Thedo nothing strategy understands the risks but does not propose subsequent adjustments in com-munity responses. The offense strategy reclaims land from the sea for development purposes.To date, the protect, accommodate, do nothing and offense strategies have been the most com-61mon choices for coastal communities, as they sustain the economic and social activities thatcoastal communities rely on. The avoid and retreat strategies, however, have been discussedmore recently as the sea level rise projections show that the continued occupation of coastal ar-eas is no longer a viable or safe option for some coastal communities. However, the implemen-tation of these strategies has been slow due to the economic, social and institutional difficultiesassociated with limiting the development and relocation of communities (Hino et al., 2017).The adaptation strategies entail different trade-offs. For example, the protect and accommo-date strategies can prevent flooding damages to buildings and infrastructure, but their economicimplications may be very high for property owners and local governments, or they may causesignificant damages to the natural environment. Similarly, the retreat strategy can provide long-term solutions for flood-prone coastal areas, but the social and economic implications for com-munities may be devastating. However, these trade-offs are not well understood yet.Coastal green infrastructure (CGI) as a sea level rise adaptation measureCGI, also known in the literature as ‘nature-based solutions’, ‘soft protection measures’, and‘green or soft shores’, is now widely considered as a viable and sustainable sea level rise adap-tation measure of the protect strategy. CGI has started to gain attention due to the recognitionof the Traditional Hard Structures’ (THSs)1 degrading impacts on coastal processes, built andnatural environment and lack of long-term social and economic benefits (Wild et al., 2017).Different from the other green infrastructure practices, such as urban green infrastructure (i.e.,stormwater management) and water-shed based green infrastructure (i.e. networks of greenspaces and forests), CGI refers to the natural and nature-based systems and processes that mimicdynamic coastal landforms (i.e., dunes, barrier islands, and beaches), coastal and riparian veg-etation (i.e., salt marshes, eelgrasses, kelp and mangroves), and reef systems (i.e., mussel andoyster beds) at coasts (The Horinko Group, 2015).As an adaptation measure, CGI contributes to the community resilience by providing multipleessential ecosystem services to humans and nature, such as increasing recreational and educa-tional opportunities; reducing implementation and maintenance cost of adaptation; improvingbuilt environment aesthetics; providing raw materials and food for humans and animals; habitatfor primary and secondary production; filtering and storage of water; reducing nitrogen input1Here, the term “traditional” refers to conventional coastal engineering practices such as dikes, seawalls, and othertypes of engineered structures62to estuaries and aquifers; and sequestring carbon (Barbier et al., 2011; Lafortezza et al., 2018).Moreover, CGI plays a critical role in coastal flooding and erosion protection through the waveand floodwater attenuation, accretion, binding soil particles, and the mitigation of debris move-ment (Hettiarachchi et al., 2013; Spalding et al., 2014; Chenoweth et al., 2018). Models andfield research investigating CGI suggest that approximately 33% to 80% erosion reduction canbe obtained through appropriate implementation of CGI (Feagin et al., 2009; Kirwan and Tem-merman, 2009; Kirwan et al., 2010). Similarly, 7% to 96% wave attenuation can be achieved,depending on the site-specific conditions (Barbier et al., 2013; Wu and Cox, 2015).Despite its benefits, it is recognized that some coastal areas may not offer the appropriate bio-physical conditions (Narayan et al., 2016), or the built environment may not be suitable for CGI(Ruckelshaus et al., 2016). Similarly, the social and institutional context may not provide thebest environment for sustained and successful implementation of CGI. The local contexts inwhich CGI is considered, including the coastal processes, natural and built environment condi-tions, and socio-economic and institutional factors are recognized to be the key determinants ofCGI’s value as a resilient adaptation measure (Narayan et al., 2016; Ruckelshaus et al., 2016).3.2.3 Adaptation strategies evaluation methods and conceptsMost communities have limited guidance in their decision-making processes to choose adapta-tion strategies, and little is known about how communities make these decisions (Brody et al.,2010). Most of the existing assessment tools focus on the cost-effectiveness of different adap-tation measures, such as traditional hard structures and coastal green infrastructure (Naumannet al., 2011; Byrne et al., 2015; Narayan et al., 2016; Wild et al., 2017; Schubert et al., 2017;Onuma and Tsuge, 2018), or different strategies, such as the cost of managed retreat (French,2006; Hino et al., 2017). The lack of structured ways of assessing the local trade-offs betweendifferent strategies may influence the ability of local governments to undertake adaptation pro-cesses (Bronen, 2015), but also to choose strategies that increase community resilience to sealevel rise.Although limited, there has been a recent acceleration of studies attempting to (1) developmethodologies that evaluate adaptation strategies, and (2) identify concepts that need to becaptured in the evaluation. However, there is hardly any consensus on the methods as wellas the resilience and adaptation concepts that are appropriate or sufficient for evaluation of63adaptation strategies (Cutter, 2016). For example, Little and Lin (2017) developed a decisionsupport framework that targets three elements of adaptation: objectives that highlight the needfor management decisions, an array of adaptation options, and system characteristics. Cutter(2016) identified economic, social, institutional, information/communication, infrastructure andenvironmental attributes as the components of resilience. Gersonius et al. (2016) developed ananalytical framework and used a scorecard to assess effectiveness, side-effects, cost-efficiency,and institutional feasibility of adaptation strategies. Azevedo de Almeida and Mostafavi (2016)used the coastal protection, transportation, water, wastewater, and energy infrastructures fac-tors to evaluate impacts on sea level rise and adaptation strategies. CAP and ICLEI (2015)developed an extensive list of forty indicators covering four sectors: coastal management, floodmanagement, infrastructure, and health. Lockwood et al. (2015) used social capital, human,financial, and physical capital, management approaches, and governance components to mea-sure adaptation. Catenacci and Giupponi (2013) developed a flexible framework that includesclimatic, physical, ecological and socio-economic components to help decision-makers in theiradaptation strategy assessment processes. Lastly, Plummer and Armitage (2007) proposed ananalytical framework to evaluate adaptation through three components: ecological, livelihoods,and process and institutional conditions. Even though these studies have developed variousframeworks, there remains a need for an evaluation framework that includes a comprehensivelist of resilience and adaptation concepts, and allows for comparison of different adaptationstrategies.Drawing from these recent studies, this research provides an alternative framework for assessingsea level rise adaptation strategies to identify local trade-offs. This evaluation framework con-sists of the following concepts: coastal processes, natural and built environment, and economic,institutional and social factors.3.3 MethodsThe objective of this study is to evaluate the local trade-offs between CGI and other adaptationstrategies. It aims to understand how CGI and other adaptation strategies interact with the localcharacteristics of communities. To be able to achieve this objective, a coastal community inBritish Columbia was selected to conceptually illustrate the methods of this study. Even thoughthis community was chosen for illustrative purposes and represented a conceptual coastal area,64real elevation, land use, and water levels data were kindly provided by the study area community.This study uses a mix of research methods, incorporating local perspectives and expertise whenpossible. This is because adaptation to sea level rise concerns a wide array of sectors, exper-tise, levels of governments, organizations, and local businesses and residents. Therefore, mostof the reported successful adaptation approaches have been achieved through engagement withlocal experts and stakeholders (Picketts et al., 2012), and understanding local characteristics ofcommunities (Barron et al., 2012). Studies at the local level engaging with local experts andstakeholders can improve the understanding of the context-specific social, environmental andeconomic conditions; increase local knowledge on climate change impacts and adaptation op-tions; and enable transparency and local participation in the decision-making processes (Zhanget al., 2008; Picketts et al., 2012; Manuel et al., 2016).This study operationalizes three research activities to achieve its objective. The three researchactivities, labeled as Step 1, 2 and 3, are illustrated in Figure 3.1.Figure 3.1: Methods diagram showing the research steps and outputs.65In the first step, the research objectives were to identify the sea level rise adaptation strategiesthat are preferred by the study area community and to develop corresponding adaptation scenar-ios. The participatory methods and public outreach activities such as meetings and participatoryworkshops are recognized as essential parts of the adaptation planning (Flynn et al., 2018). Forthis reason, a set of two expert meetings and a participatory workshop were organized with thepublic and experts in the study area over a period of 15 months. These meetings enabled the re-view of the local data on sea level rise, tidal range, storm surge and associated wave effects, andidentification of the community preferences on sea level rise adaptation strategies. The inputsand feedbacks from these meetings are used to develop adaptation scenarios. Scenarios pro-vide structured ways to describe narratives of future conditions based on designated adaptationstrategies (Barron et al., 2012). They provide storylines that cover social, economic and institu-tional implications of adaptation actions and decisions. Therefore, four sea level rise adaptationscenarios were developed for the selected strategies.In the second step, the research objective was to develop an evaluation framework to assess thetrade-offs between sea level rise adaptation strategies. Assessment tools such as frameworks,indices, or scorecards help the evaluation of a wide range of impacts of adaptation decisions(Nelson et al., 2007; Little and Lin, 2017; Garner and Keller, 2018). Therefore, an extensivereview of the academic and grey literature was conducted to outline concepts that are consideredin the other evaluation frameworks but also to identify new concepts that are relevant to coastalcommunity resilience and adaptation to sea level rise. Based on the findings of these reviews,a draft evaluation framework was developed. An expert elicitation survey was conducted togather feedback on the draft evaluation framework from the experts in the field. At the end ofthe second step, the expert feedback was incorporated, and the final version of the evaluationframework was completed.In the third step, the research objective was to apply the evaluation framework developed in thesecond step to the sea level rise adaptation scenarios developed in the first step. The results werecalculated and illustrated to assess the trade-offs between CGI and other adaptation strategies.3.3.1 The study areaThe study area is located on the west coast of the District of North Saanich, a coastal communityin the province of British Columbia (Figure 3.2). This area was selected to implement the66research methods and conceptually illustrate the scenario models. The study area is on a smallbay and is primarily single-family residential. Some other land uses include marinas, a school,and green spaces. It has a low-lying coast that is mostly protected from wave events, yet it hasbeen experiencing occasional coastal flooding during the king tide and heavy storm events.Municipal experts in the District of North Saanich participated in the expert meetings and partic-ipatory workshop. They also generously shared their expertise, data (i.e., the Geographic Infor-mation Systems (GIS) layers and maps), and the local Flood Construction Levels (FCL) study.The District of North Saanich did not participate in the application of the evaluation frameworkto the four adaptation scenarios developed for the study area. Please see the disclaimers at theend of this chapter.Figure 3.2: The location of the study area.673.3.2 Step 1 - The expert meetings and participatory workshopThe research objectives of this step were to identify the sea level rise adaptation strategies thatare preferred by the study area community and to develop corresponding scenarios. Over aperiod of 15 months (from June 2016 to September 2017), an expert meeting, a participatoryworkshop, and another expert meeting were organized to achieve the research objectives.The first expert meetingThe first expert meeting was organized to prepare for the participatory workshop. 18 expertsfrom the regional district, the provincial government, and environmental organizations; coastalengineer; municipal staff (planners, engineer, and emergency manager); and local politiciansattended the meeting.The primary objectives of the first expert meeting were to come to a consensus on the desig-nated flood level (DFL) elevation for the study area and to identify relevant stakeholders for theparticipatory workshop. The DFL elevation refers to the still water levels, which include thesea level rise allowance, maximum high tide, and the estimation for 1 in 500-year storm surgelevels. It does not include the wave effects (SNC LAVALIN, 2016) and the associated waveeffects zone at the coast. For the first objective, participants were presented with the provincialand regional relative sea level rise projections, local tidal range, 1 in 500 storm surge levels, andthe wave effects estimates. They were asked to review the information and answer the questionslisted in Appendix B.1. For the second objective, the participants were asked to join the opendiscussion on identifying the relevant stakeholders in the region and the workshop format.The participatory workshopThe participatory workshop aimed to understand community knowledge on climate change im-pacts and preferences on sea level rise adaptation strategies. The information about the work-shop was posted on the community’s official website, and invitations were sent to two mainresident groups approximately 45 days prior to the workshop date. 38 people from the resi-dents, community groups, businesses owners, municipal staff, and politicians; provincial andregional governments, transportation authorities, a coastal engineer, and environmental law andNGOs attended the workshop.The participants were asked to fill out a pre-workshop survey (Appendix B.2) before the eventbegan, and the materials were presented. The survey included questions on personal informa-68tion (such as age and education), knowledge on climate change impacts, perceived hazard risks,and perceptions of the effectiveness of different adaptation strategies and measures. Then, in-formation on the processes affecting sea level rise, and the local sea level rise projections werepresented. The impacts of sea level rise on water levels in the study area were presented us-ing data from the Flood Construction Levels study (SNC LAVALIN, 2016). These impactswere presented using conceptual visuals. The main adaptation strategies recommended by theprovincial government and the regional district were presented.After the presentation, workshop participants were asked to form groups to discuss their prefer-ences on the adaptation strategies and tools they preferred to see implemented in the study area.The participants were provided with color-coded moveable cards and maps to discuss differ-ent adaptation strategies and tools, and their potential implications beyond coastal flooding. Apost-workshop survey was administered at the end of the meeting (Appendix B.3). This surveyincluded questions aiming to measure changes in the knowledge, perception, and preferences.The second expert meetingThe second expert meeting was organized after the participatory workshop. Nine experts fromthe regional district and the provincial government; a coastal engineer consulting firm; and localmunicipality (planners and engineers) attended the meeting. The objective of the meeting wasto review the wave effects zone estimates and the sea level rise adaptation scenarios developedbased on the strategies selected at the participatory workshop.The wave effects zone in this study refers to the conceptual illustration of the inland expansion ofwave action (wave overtopping) at the coast. The wave effects zone depends on the interactionsof the waves with the human-made and natural features at the coast. Conceptual methods thatcan be used to represent the wave estimate zones were discussed in the meeting. Participantsagreed on a method that would be appropriate to use for the purposes of this study.The participants were also presented with the visual illustrations of the scenarios that showed thephysical attributes of the community, such as roads, structures, land use, and the DFL elevationthat was agreed in the first expert meeting. The visual illustrations of the scenarios aimed toconceptually communicate the potential physical implications of strategies such as the highestwater levels, high flooding risk areas, and changes in built and natural environment. To visuallyillustrate the scenarios, the local elevation data was converted from GIS to a 3D Sketch Upmodel. Structures, roads, and other infrastructure in the study area were conceptually placed69on the 3D models using google maps and site visit images. Characteristics of each strategywere incorporated in the 3D models. For example, if the strategy includes the use of the THSmeasures, then a seawall was added to the edge of the property lines. Next, the DFL was addedto the 3D models. Lastly, the wave effects zones were added to the corresponding models.Participants were asked to discuss the iterations of the strategies into scenarios and the associatedwave effects zones for each scenario. These discussions were used to improve the scenarios andtheir visual illustrations.3.3.3 Step 2 - Sea level rise adaptation strategies evaluation frameworkThe research objective of this step was to develop an evaluation framework that can be usedto assess the local trade-offs of sea level rise adaptation strategies. To achieve this objective, aliterature review was conducted to develop a draft evaluation framework, an expert elicitationsurvey was administered to get feedback on the draft evaluation framework, and lastly, the expertfeedback was incorporated to complete the final evaluation framework.Literature review and the draft evaluation frameworkAn extensive review of the published academic and grey literature was conducted to develop thedraft evaluation framework. This review focused on the attributes of resilience, community re-silience to sea level rise, and sea level rise adaptation. The literature review highlighted specificconcepts that were commonly considered in the studies and identified new ones. The conceptsdrawn from the literature were developed into components of the evaluation framework.The draft evaluation framework included 35 components. Criteria for each of the thirty-fivecomponents were developed based on the literature review and organized to provide a three-point scoring system. These components were then grouped into six modules. These modulesare coastal processes, natural environment, built environment, economic factors, institutionalfactors, and social factors.A three-point scoring system, from +1 to -1, is used in the evaluation framework to provide astraightforward structure. The -1 score indicates a negative impact, and the +1 score indicatesa positive impact on resilience, while the 0 score indicates minimal to no impact, compared tobaseline conditions. An N/A option is provided to be used when the criteria are not applicableto an area or the strategy. For example, if an area does not have any commercial or industrial70structures, the impacts of a strategy on these structures cannot be scored; therefore the N/Aoption should be selected.Expert elicitation surveyAn expert elicitation survey was designed using the draft evaluation framework to gather feed-back on the draft evaluation framework’s modules, components and criteria used for scoring. Inaddition, the expert elicitation survey sought guidance on the weighting system of the frame-work.The survey was sent to the experts from the provincial government, regional government andorganizations, and local governments. The participants were asked to mark their agreement withthe framework’s modules, components and criteria, and to provide comments and suggestions.Three experts who are involved in the climate change adaptation field in the region, two fromthe provincial government and one from a regional organization, participated in the survey.Final evaluation frameworkThe experts provided detailed and comprehensive feedback on the evaluation framework mod-ules, components, the criteria of the components, the scoring system and the weighting of themodules and components.The experts agreed with the framework modules and the criteria used to assess each componentand recommended minor changes to some of the components. These changes were recom-mended to improve the comprehensibility of the evaluation framework and to ensure the three-point score can be applied meaningfully. The expert elicitation survey feedback was reviewedand incorporated into the final evaluation framework.The feedback from the expert elicitation survey was incorporated to modify the framework con-tents and to develop a weighting system. Based on the recommendations, two new componentswere added, five components were removed, and four components were merged into two. In theend, the final framework contained 30 components. The summary of the changes made after theexpert feedback are listed in Table 3.1. Since the recommended changes were minor, only thefinal evaluation framework is presented in the results section and in Appendix B.4.71Module Component ChangesBuilt Environment Specific land use loss due to strategy implemen-tationAddedBuilt Environment Private property/land loss due to strategy imple-mentationRemovedBuilt Environment Flooding risks on public areas at the coast RemovedEconomic Factors Economic benefits to communitiesEconomic benefits to local governmentsMergedEconomic Lifetime of strategy RemovedInstitutional Factors Municipal authority AddedInstitutional Factors Usability with and support to other documents RemovedSocial Factors Opportunities for public educationOpportunities for community involvementMergedSocial factors Benefits to vulnerable populations RemovedTable 3.1: Summary of changes in the evaluation framework components.Using the expert feedback, a two-step weighting system was developed, giving the users flexi-bility to prioritize their objectives, while limiting exclusion of any modules or components. Inthe first step, users need to allocate 60 points to six modules. The minimum point that can beallocated to a module is set at six, and the maximum is set at 30 so that no module receivesmore than 50% and less than 10% of the points. If one module is allocated 30 points, then theother modules have to be allocated six points, so that the sum of all the module weighting is60 points. Users cannot give 0 or 60 to any module. If all the modules are weighted equally,then each module would be allocated 10 points. In the second step, users need to allocate 25points to five components within each module. The minimum point that can be allocated to acomponent is set at 2.5, and the maximum is set at 12.5 so that no component receives morethan 50% and less than 10% of the points. If one component is allocated 12.5 points, then theother components have to be allocated 2.5 points. Users cannot give 0 or 25 to any component.If all the components are weighted equally, then each component would be allocated 2.5 points.The calculation of the framework can be found in Appendix B.5.723.3.4 Step 3 - Application of the evaluation framework to sea level riseadaptation strategiesIn this last step, the final evaluation framework was applied to the sea level rise adaptationscenarios that are developed using the community inputs and expert feedback. This step of theresearch consisted of two main activities.First, the final evaluation framework was applied to the four adaptation scenarios. The 3D mod-els, GIS land use maps, document reviews (such as the Official Community Plans (OCPs)), andthe researcher’s knowledge of the study area were used to complete the evaluation framework.Second, the results of the application were calculated and illustrated to assess the benefits anddisadvantages and to highlight the trade-offs of the different adaptation strategies.3.4 Results3.4.1 The expert meetings and participatory workshop3.4.1.1 The first expert meetingThe first expert meeting was organized to identify the DFL elevation of the study area andthe list of community groups, stakeholders, and First Nations communities to be invited tothe workshop. The expert meeting participants decided the DFL by reviewing and discussingthe existing information on the baseline data. The participants came to a consensus that theDFL should include a 1m relative sea level rise allowance. They also suggested that a 1.5 mhighest high tide level and a 1.3 m storm surge estimate, which are both based on the localbuoy measurements, should be included in the DFL. Based on these recommendations, the DFLconsisted of the following:DFL = Sea level rise (1 m) + Highest high tide level (1.5 m) + Storm surge (1.3 m) = 3.8 mThe discussion on the upcoming workshop provided a detailed list of community groups, stake-holders, and First Nations communities to be invited to the workshop. This list included theadjacent municipalities and First Nations municipal councils; the regional district and provin-cial government; local school districts; transportation authorities (i.e., ferries, highways, and73airports); local NGOs and other organizations; environmental and engineering consultants; andlocal businesses, residents, and residents associations. The participatory workshopThe participatory workshop was organized to understand community knowledge on climatechange impacts and preferences on adaptation strategies. Most of the workshop participants,65%, were homeowners, where 41% of the participants owned waterfront properties. Partic-ipants indicated that the general concerns over sea level rise impacts and being prepared forthose impacts were the main reasons they attended the workshop. The consequences of sealevel rise adaptation on the rights of waterfront property owners was another important reasonthey attended the meeting.A pre-workshop survey (Appendix B.2) was administered at the beginning of the workshop. Atotal of 29 people completed the survey. The results are as follows:• 34% of the participants have observed increases in the storm intensity over time, where27% observed increases in the storm frequency and 25% in the wave heights.• 85% of the participants perceived 0-40% chance of extreme coastal flooding events overthe next fifteen years (due to the combination of sea level rise, high tides, and heavystorms) in their community. The risk perception increased for the next fifty years. 61% ofthe participants perceived more than 60% change of extreme coastal flooding events.• 41% of the participants thought the potential impacts of climate change would be signifi-cant on sea levels, storm/wave effects, real estate values, and damage to private property.• 44% of the participants indicated that they consider THS (sea walls & dikes), and 41%CGI (coastal vegetation) as effective coastal erosion and flooding protection measures.• 76% of the participants rated avoid as an effective sea level rise adaptation strategy while55% rated retreat, and 45% rated protect and accommodate strategies.• 78% of the participants thought the community participation in the sea level rise adapta-tion planning is very important.74A post-workshop survey (Appendix B.3) was administered at the end of the workshop. A totalof 25 people completed the survey. The results are as follows:• 60% of the participants reported increased knowledge on climate change and its implica-tions in the local context.• Particularly, their knowledge on sea level rise and associated storm and wave impacts,potential damages to private property and infrastructure, and implications on local realestate values increased the most.• 52% of the participants thought sea walls and dikes are effective measures to protectcoastal areas, while 58% groins, 45% revetments and soft structures, and 50% thoughtcoastal vegetation are effective coastal protection measures, but when asked which mea-sures they want to see implemented in the study area, the results were very close.• The accommodate (52%) and protect (47%) strategies were the most favored for the studyarea, even though the avoid and retreat strategies were rated the most effective strategiesin the pre-workshop survey. The retreat and avoid strategies were much less preferred,rated 35% and 30%, respectively.• There was about 4% increase in the participants’ views (up to about 82%) on the impor-tance of the community participation to adaptation planning and policies.• Participants stated that the workshop provided an environment for people to make newconnections, and an inclusive and transparent place for raising their concerns and dis-cussing adaptation options with other stakeholders and municipal staff.The workshop brought together a wide diversity of participants. It appeared to increase par-ticipants’ knowledge on climate change implications, and their views on the importance of thecommunity participation to adaptation processes. The workshop also showed that the partici-pants favored the accommodate and protect strategies, even though they initially selected theavoid and retreat strategies as most effective options. The workshop also enabled an open andtransparent environment for conversation of sea level rise concerns, options, and preferences.Based on the results of the workshop, the protect, accommodate, and a joint retreat & avoidstrategies were the selected for the scenario development. The protect strategy included twoseparate options; one included a THS protection measure, and the other included a CGI measure.75Lastly, a do nothing strategy was also added to illustrate the baseline conditions and to comparethe scenarios. As a result, a total of four sea level rise adaptation strategies, and a baselinestrategy were selected for the scenario development. The second expert meetingThe second expert meeting was organized to review the wave effects zone estimates and the sealevel rise adaptation scenarios developed based on the selected strategies.The wave effects zone estimatesAs described in the methods section, the wave effects zone in this study conceptually reflectsthe inland expansion of wave action at the coast. The estimation of the wave effects zone is verycomplex and requires engineering expertise to assess but moreover, it is beyond the objectivesof this study. Therefore, the participants, which included municipal engineers and an externalcoastal engineer, agreed that the following simple method would be appropriate for purposes ofthis research.Wave effects zone = Wave effects x Wave effects factor (f)The participants agreed that a 0.9 m wave effects should be included in the wave effects zoneestimates in each scenario. This wave effects value was suggested in the local flood construc-tion levels report for this study area. The wave effects factors were discussed separately foreach of scenario, as they illustrate a different type of intervention at the coast. Based on thesediscussions, the wave effects zone for each strategy were estimated as described in Table 3.2.Strategies Wave effects Wave effects factor Wave effects zoneDo nothing 0.9 m 3 2.7 mProtection with THS 0.9 m 10 9 mProtection with CGI 0.9 m n/a 0.9 mAccommodate 0.9 m 3 2.7 mAvoid & Retreat 0.9 m 3 2.7 mTable 3.2: Wave effects zone estimates for the selected strategies.For the protect strategy, two different wave effects factors were used. For the traditional hard76structure (THS) protection measure f=10 was decided, due to the increases in the wave action onthe vertically hardened shorelines. For the coastal green infrastructure (CGI) protection mea-sure, the participants decided that there wouldn’t be a wave effects factor, because the waveenergy would not be amplified by the coastal green infrastructure. For the do nothing andaccommodate strategies f=3 was decided, due to the existing partially hardened shoreline. Sim-ilarly, for the avoid & retreat strategy f=3 was decided because no further action was proposedat the shoreline. In addition to the expert meeting participants, an independent water resourcesengineer also confirmed the method used here to illustrate the wave effects zone conceptually.Sea level rise adaptation scenariosA total of four adaptation scenarios and a baseline scenario were developed using the inputgathered at the participatory workshop and the second expert meeting. They were developed forpurposes of illustrating how the framework could be applied to evaluate alternative adaptationscenarios. The illustrative visuals for each scenario are provided in Figure 3.3.77Figure 3.3: Sea level rise adaptation scenarios conceptual visual illustration. Drawings and illustrationsare not to scale. Please see the disclaimer at the end of this chapter.These conceptual scenarios included a number of quantitative (i.e., DFL and the wave effectszone) assumptions that were explained in the previous sections. They also included variousqualitative assumptions that are described below.78Baseline scenario - Do nothingThe ‘Do nothing’ scenario depicts a future where no adaptation actions were taken to deal withthe sea level rise, and illustrates the corresponding implications. This scenario is the extensionof the baseline conditions over time. In this scenario, there are no physical interventions onthe coastline. So the coast will still be partially protected with seawalls in front of some of theresidential and commercial properties. In this future, the sea level rise, coupled with the hightide levels, threatens several private properties, all of the commercial properties, and some localroads and underlying infrastructure. A 1 in 500-year storm surge, however, impacts most of thestructures, roads, and infrastructures in the study area.Scenario 1 - Protection with traditional hard structure (THS)The ‘Protection with THS’ scenario depicts a future where the protect strategy is adopted, andall of the coastline is hardened with seawalls. This scenario extends the existing partial seawallsthroughout the community’s coastline. In this future, the seawalls are built and maintained bythe private property owners on the landward side of their property boundaries, as it is currentlydone. The extended seawalls are expected to protect the community from the still waters, suchas high tides, sea level rise and storm surges, assuming that the seawalls are structurally in stablecondition. However, the interaction of the storms and waves with the seawalls is expected toamplify the wave effects zone and to increase the associated flooding and damages (The HorinkoGroup, 2015).Scenario 2 - Protection with coastal green infrastructure (CGI)The ‘Protection with CGI’ scenario also depicts a future where the protect strategy is adopted.However, in this future, the eelgrass in the foreshore will be restored; the sandy beach betweenthe foreshore and the private properties will be nourished, and large woody debris will be placed,and the riparian shore (the edge of the private property boundaries) will be restored using nativeplants and shrubs. The provincial government will implement the foreshore restoration, andbeach nourishment as the land is under the provincial jurisdiction. The private property own-ers will implement the riparian restoration. The drag friction gain from the eelgrasses, beachslope, woody debris, and the riparian vegetation is expected to reduce to wave energy, thusthe wave effects zone. Howes et al. (1994)’s ‘The Physical Shore-Zone Mapping System forBritish Columbia’ identifies the soil conditions of the study area as stable and sediment supplyas abundant. Therefore, the restoration of habitat in this study area can help to accumulate soilat the cost and prevent the rising water levels from reaching further in the coastal areas (Houston79and Dean, 2016). Thus, the CGI intervention is expected to help with the accretion processes,increasing the coastal soil elevation (Kirwan et al., 2010).Scenario 3 - Built environment accommodationThe ‘Built environment accommodation’ scenario depicts a future where structures, roads, andinfrastructure are retrofitted by elevating, and/or dry and wet-proofing structures. In this sce-nario, the flooding risks are known, but the primary objective is to either completely avoid theflooding impacts or to minimize damages. Elevating structures, although costly, can preventflood waters from entering structures. Dry proofing involves using waterproof sealants to pre-vent floodwaters from entering through the walls and windows; while wet proofing involvesmodifying the lower floors of the structures to allow occasional flooding without significantdamage to the property and materials (FEMA, 2007). Roads and infrastructure will also needto be floodproofed by elevating on top of a bank or floating roads (ECAP, 2015). This strategywill be implemented and maintained by private owners and local governments. Local govern-ments may build partnerships with other levels of governments to create cost-sharing solutions.This strategy does not impact the area that will be impacted by the sea level rise but reduces thenumber of structures and infrastructures that may be damaged due to flooding.Scenario 4 - Retreat from the coastThe ‘Retreat from the coast’ scenario depicts a future where the avoid and retreat strategies areadopted. In this scenario, no future development is allowed in the DFL and wave effects zones.Moreover, the community is planning for the strategical and phased retreat from these zones.This retreat is complex, and there is no precedence to help guide the process. In this future, thelocal government is responsible for implementation and maintenance of the strategy. Therefore,this scenario involves the collaboration of many levels of governance. The community also hasto develop partnerships to finance the project because the avoid and retreat strategies are veryexpensive as they involve large areas of land acquisition and loss of economic activities (Gray,2017). The DFL area will not change in this scenario since there are no interventions at thecoast, but the flooding risks on the built environment will be eliminated.3.4.2 The final evaluation frameworkThe evaluation framework developed in this study aims to assess sea level rise adaptation strate-gies for the purposes of identifying the local trade-offs of sea level rise adaptation strategies.80It can be used to review the benefits, challenges, and trade-offs between different adaptationoptions to help to prepare recommendations for adaptation actions.The evaluation framework is designed to be implemented by a planning or project team. Thisteam should include municipal planners, engineers, emergency managers, sustainability coordi-nators, utility managers, and other relevant municipal departments, as well as regional, provin-cial, and federal experts. Depending on the community’s adaptation planning objectives, theevaluation framework can be implemented either before or after a public stakeholder workshop.For example, it can be used as an initial screening tool to review various adaptation optionsprior to a public workshop. The planning/project team then can use this tool to help identify fewadaptation strategies that can be brought up in a public stakeholder meeting to discuss with theparticipants in greater detail. Alternatively, the implementation of the tool can follow a publicstakeholder workshop, where the public preferences on adaptation strategies are obtained. Theplanning/project team then can use this tool to review the benefits, challenges, and trade-offseach option to help to prepare recommendations for adaptation actions.As described in Section 3.3, a three-point scoring system, +1 (positive impact), 0 (minimal tono impact), and -1 (negative impact), is used in the evaluation framework to provide a straight-forward structure. An N/A option is provided to be used when the criteria are not applicable toan area or the strategy.Also described in Section 3.3 is the weighting system of the evaluation framework. The evalua-tion framework includes a two-step weighting system: in the first step, users need to allocate 60points (minimum 6 and maximum 30 points) to six modules; and in the second step, users needto allocate 25 points (minimum 2.5 and maximum 12.5 points) to five components within eachmodule.The final evaluation framework consists of six modules, reflecting varying concepts of com-munity resilience and adaptation to sea level rise (Cutter, 2016). These modules are coastalprocesses, natural environment, built environment, economic factors, institutional factors, andsocial factors. There are a total of 30 components, five in each module, aiming to identify thedifferent aspects of the module. Each module includes criteria to assist with scoring. Table 3.3displays the evaluation framework modules and components. The final evaluation framework,which also includes the three-point scoring system and the corresponding evaluation criteria canbe found in Appendix B.4.81Modules ComponentsCP1 Protection from short-term high energy eventsCoastal Processes CP2 Protection from long-term low energy events(CP) CP3 Protection from coastal erosion caused by drag frictionCP4 Protection from coastal erosion caused by disruption to sed-iment processesCP5 Protection from secondary threatsNE1 Habitat migration along the coastal profileNatural Environment NE2 Coastal habitat and biodiversity(NE) NE3 Inland habitat and biodiversityNE4 Natural movement of organic and inorganic sedimentsNE5 Regionally important and sensitive coastal habitatsBE1 Specific land use loss due to strategy implementationBuilt Environment BE2 Flooding risks on private property(BE) BE3 Flooding risks on businesses and other commercial or in-dustrial structuresBE4 Flooding risks on roads and infrastructureBE5 Visual aesthetics of the strategyEF1 Economic benefits to communitiesEconomic Factors EF2 Implementation costs to local governments(EF) EF3 Implementation costs to waterfront property ownersEF4 Maintenance cost to local governmentsEF5 Maintenance cost to waterfront property ownersIF1 Support in planning documentsInstitutional Factors IF2 Municipal authority(IF) IF3 Availability of information, data and toolsIF4 Availability of technical expertiseIF5 Adaptability to changing environmental conditionsSF1 Acceptability of the strategySocial Factors SF2 Opportunities for community involvement(SF) SF3 Access to coastal areasSF4 Preserving ecological and cultural valuesSF5 Displacement of populationsTable 3.3: The final evaluation framework modules and components. The three-point (+1/0/-1) scoringsystem and the evaluation criteria for each components can be found in Appendix B.4.82The importance of the natural and built environment, and economic factors for coastal commu-nity resilience to sea level rise have been well covered in the published academic and grey lit-erature (Gibbs, 2015; Working Group on Adaptation and Climate Resilience, 2016; Cook et al.,2016; Azevedo de Almeida and Mostafavi, 2016; Sa´nchez-Arcilla et al., 2016; Wild et al., 2017;Hagen et al., 2018). The evaluation framework developed in this study contributes to the threeother important concepts: coastal processes, and institutional and social factors. These con-cepts have often been overlooked or dismissed either because they were deemed unimportant,irrelevant, or difficult to assess. The coastal processes module reflects protection from the phys-ical changes that occur between the built environment and natural processes such as floodingand erosion, and the interaction between the proposed adaptation measures and these processes(Frazier et al., 2013; Burton, 2014; Yoon et al., 2015). The institutional factors module reflectsthe overall capacity of communities’ decision-making organizations. There are now numerousstudies citing the vital role of institutions in dealing with change and uncertainty, and adap-tation to sea level rise (Bronen, 2015; Gersonius et al., 2016; Oulahen et al., 2017; Patterson,2018). The social factors module reflects the implications of sea level rise adaptation decisionson communities’ perceptions, involvement, and development. Adaptation to sea level rise willhave different implications on communities’ social fabric, values, and equality (Sa´nchez-Arcillaet al., 2016). The descriptions of each module and component are provided below.Coastal Processes (CP)The main consequence of sea level rise is the increased risk of coastal flooding and erosion(Azevedo de Almeida and Mostafavi, 2016; Passeri et al., 2015). Therefore, the coastal pro-cesses module addresses how different adaptation strategies would affect the processes that oc-cur at the coast.The CP1 & CP2 components deal specifically with the interaction of the adaptation strategieswith processes that result in flooding events. Coastal flooding events are separated into twoparts based on their impact time and impact force. The short-term high-energy events describeflooding caused by low-pressure weather systems that lead to storm surges and powerful waves(Flather, 2001). These events last hours or days at most, and are responsible for most of the dam-age coastal communities suffer (Cox et al., 2018). The long-term low-energy events describeflooding caused by the changes in tidal range and mean water levels over time (Wesselmanet al., 2018). These events are difficult to observe as significant changes are manifested overdecades. However, they can have significant impacts on low-lying coastal areas (Hagen et al.,832018) because they can increase the land area that is affected.The CP3 & CP4 components deal specifically with the impact of the adaptation strategies oncoastal processes that result in erosion events. The coastal erosion events are also separatedinto two parts. First is the erosion caused by drag friction, which is the result of the backwashof the wave, current and tidal oscillations that transports loose sediment, soil, rock, and vege-tation patches away from coasts (Siegle et al., 2008; Ruckelshaus et al., 2016). Second is theerosion caused by disruption to sediment deposition processes, which leads to the decreased ac-cretion at the coast (Sandi et al., 2018). Accretion is the process that increases surface elevationthrough organic and inorganic sediment trapping, root accumulation, primary production, andplant decay (Kennish, 2001), and is an essential natural defense mechanism of coastal areas.The CP5 component addresses the secondary threats from loose materials that break apart fromthe protection measures due to strong wind and wave action. Storm surge and waves can tearparts of the coastal protection structures (such as sea walls, ripraps or breakwaters), and theseloose parts can result in unintended damages (Klein et al., 2001), and can pose significant threatsto human lives (Perdok, 2002).Natural Environment (NE)Sea level rise impacts and human action (or inaction) can alter the resilience of the naturalenvironments (Sa´nchez-Arcilla et al., 2016; Devoy, 2015) by disrupting habitat area, habitatmigration, biodiversity, and sediment transportation through the coastal profile (Comer et al.,2012). Therefore, the natural environment module addresses the effects of the different adapta-tion strategies on habitats, biodiversity, and coastal functionality.The NE1 component targets the impacts of the adaptation strategies on the habitat migrationalong the coastal profile. This migration enables habitats to survive changes in the mean waterlevels by migrating landward or seaward (Doody, 2013; Pontee, 2013). Rapid or drastic changesin the coastal slope and water levels can alter the ability of the habitat to migrate along the coast.In addition, human interventions such as hard structures or dense development at the coast canlimit this migration (Doody, 2004).The NE2 & NE3 components deal with the impacts of the adaptation strategies on coastal andriparian habitat and biodiversity. The NE2 addresses explicitly the coastal habitat and biodi-versity. Coastal habitats are amongst the most valuable and productive areas in the world asthey provide numerous ecosystem services (Wamsler et al., 2016; Sandi et al., 2018). Decreases84in habitat and associated changes in biodiversity can have social, economic and environmentalconsequences (Nicholls, 2004; Sandi et al., 2018). The NE3 component addresses the riparianand inland habitat and biodiversity. Riparian and inland habitats are also essential for sustain-able communities as they provide various environmental, social, health and economic benefits(Li et al., 2015). Preserved and enhanced riparian and inland areas can help promote habitatand biodiversity while serving as a buffer during inundation events and absorbing floodwater(Eastern Research Group, 2013).The NE4 component addresses the movement of sediments throughout the coastal profile. It isdifferent from the CP3 & CP4 components, which are related to the sediment transport to thecoastal areas. This process is affected by the geomorphology, currents, winds, tides, and humaninterventions at the coast (Temmerman et al., 2013). Therefore, strategies can alter the rate ofsediment transferred and the location of the deposition throughout the coastal profile (Siegleet al., 2008). Lastly, the NE5 component targets the regionally important and sensitive coastalhabitats. The ecosystem services of some coastal habitats exceed the boundaries of their localcommunities. Alterations to these coastal habitats would have much broader social, economicand environmental impacts in the region (Burcharth et al., 2007).Built Environment (BE)Population density is typically higher in coastal areas (Neumann et al., 2015), which increasesthe number and type of land, assets, and infrastructures that are at risk of flooding due to sealevel rise. The resilience of the built environment is essential for the safety and functioning ofdaily life in coastal communities (Azevedo de Almeida and Mostafavi, 2016). Therefore, thebuilt environment module deals with the effects of the different strategies on land use, assetsand infrastructure, and the aesthetics of the built environment.The BE1 component addresses the impacts of strategies on local land use. Some land uses maybe disproportionally affected by strategies, and this may lead to the loss of essential economic orsocial activities and livelihoods in the community. For example, changes in the areas dedicatedto food production, rental housing units, or industrial areas (CAP and ICLEI, 2015; Shayeghet al., 2016; Pramanik, 2017) can alter the built environment fabric of communities.The BE2 & BE3 & BE4 components deal the strategy impacts on the flooding risks of assetsand infrastructure. The BE2 component focuses on private properties as the physical damageto homes in coastal areas are expected to increase with rising sea levels and changing storm85patterns (Burcharth et al., 2007). Private property owners can face threats to personal safetyand experience significant losses and damages if the risk of flooding is not reduced (CAP andICLEI, 2015). The BE3 component focuses on commercial and industrial structures. Floodingof the commercial/industrial structures can result in disruptions of economic activities, thus canhave short and long-term financial impacts on coastal communities (CAP and ICLEI, 2015;Azevedo de Almeida and Mostafavi, 2016). The BE4 component focuses on the flooding riskson roads and infrastructure. Roads and infrastructure are critical systems as they connect peopleand services; therefore, disruptions to roads and infrastructure systems can lead to cascadingimpacts on the safety of residents, rescue, clean up and rebuilding and efforts, and functioningof services (Azevedo de Almeida and Mostafavi, 2016)The BE5 component deals with the aesthetic consequences of strategies. Coastal areas are oftenprime real estate locations offering residents and visitors scenic coastal views. Strategies canblock or significantly alter these coastal views, therefore may have significant impacts on theaesthetics and scenic quality of the built environment, and the real estate prices (Berte andPanagopoulos, 2014; Wamsler et al., 2016).Economic Factors (EF)Coastal areas have always been economic hubs due to the benefits of logistical, cultural, andrecreational activities at the coasts (Neumann et al., 2015). Coastal communities rely on theseactivities for their tax revenues. However, flooding and erosion risks associated with the risingwater levels pose threats to coastal communities’ ability to sustain these activities (Hinkel et al.,2014). The adaptation strategies can provide various benefits and challenges to coastal commu-nities’ economic resilience. The decision-makers should understand the cost and benefits of thedifferent strategies (Eastern Research Group, 2013; Fu and Song, 2017) both on private prop-erties and local governments. Therefore, the economic factors module addresses the impactsof the adaptation strategies on local economic activities, and life-cycle costs to private propertyowners and local governments.The EF1 component deals with the economic benefits of strategies. Tax revenues from propertiesare the main contributors to local governments’ budgets; therefore, the provision of municipalservices. However, sustaining and improving economic activities, while providing safety andprotection is a challenging task for local governments (Duffy et al., 2014). Understanding theimplications of strategies on the coastal economic activities is therefore essential.86Coastal areas consist of a mixture of private and public ownership and different governmentjurisdictions. This ownership and jurisdictional arrangements creates different responsibilitiesand economic capacities to pay for the adaptation strategies (Gibbs, 2015). Therefore, the EF2& EF3 & EF4 & EF5 components address the life-cycle cost of the strategies on private propertyowners and local governments. The EF2 & EF4 components focus on the implementation andmaintenance costs of the strategies on local governments. Local governments are responsiblefor the strategies under their jurisdictions. However, the local governments in British Columbiahave direct access to only property tax, and they are restricted in their ability to raise revenue,and access financial resources (Dewing et al., 2006). Therefore, the financial burden of strategieson local governments can be significant. However, the cost of some strategies (i.e., accommo-date) may qualify for federal and provincial cost-sharing programs, grants and partnerships,which may lessen local governments’ financial responsibilities (Frazier et al., 2010; CAP andICLEI, 2015). Maintenance cost, however, is often neglected, and not covered in most of thecost-sharing programs. On the other hand, the EF3 & EF5 components focus on the costs ofthe strategies on private properties. Even though flood insurance is now becoming available, atof the time this research was conducted, private property owners did not have access to coastalflood insurance coverage (see Oulahen 2014. Besides the coastal flood insurance, there is hardlyany financial assistance available for small-scale projects to help them adapt to rising water lev-els. Although there are some technical and conceptual reports available, waterfront propertyowners are often alone in financing their intervention at the coast. Therefore, the economicimplications of strategies should be considered.Institutional Factors (IF)Institutions influence the decision to adopt one strategy or policy over another one (Patterson,2018). The institutional attributes of communities, such as the capacity, access to resources, andregulations in place influence how communities adapt to sea level rise (Oulahen et al., 2017;Chang et al., 2018). However, institutions are facing significant challenges to improve theirability to assess and deal with change and uncertainties, risks and knowledge requirements (Pat-terson, 2018). Therefore, the institutional factors module deals with the impact of the adaptationstrategies on institutional attributes such as plans and policies, jurisdictional authority and de-pendency, and access to knowledge, tools, and data.The IF1 component targets the planning documents. Local governments use various planningdocuments to plan and regulate within their boundaries. The regional districts provide vari-87ous services and support to local governments, and provincial governments provide regulatorycontext, information, and instructional materials (MoMAH, 2018). The degree to which thestrategies are aligned with the goals of the local and regional planning documents can indicatehow easily a strategy can be implemented. The IF2 component targets the municipal author-ity of local governments in implementing the adaptation strategies. Coastal jurisdiction andownership are therefore significant in determining which strategies can be undertaken by localgovernments. The consideration and adoption of a strategy can be more challenging if the localgovernments do not have the authority to implement it.The IF3 & IF4 components focus on the availability of resources such as information, dataand tools, and technical expertise required to undertake the adaptation processes. The strategyimplementation depends on reliable and continuous access to these resources (Working Groupon Adaptation and Climate Resilience, 2016). The availability of these resources within thelocal government or through regional, provincial and federal sources can effectively supportdecision-making.The IF5 component deals with the adaptability of the strategies and the institutional attributesto change and uncertainties. Change and uncertainty are expected parts of social and ecologicalsystems. Therefore strategies that are adaptable to changing and unexpected environmentalconditions, and can be amended/modified easily within the existing institutional arrangementsenhance community resilience (Cinner et al., 2018; Patterson, 2018).Social Factors (SF)The consequences of sea level rise and adaptation to it will have various social implications(Sa´nchez-Arcilla et al., 2016). Providing safety and continued occupation of coasts may re-quire significant efforts and resources from local governments (Working Group on Adaptationand Climate Resilience, 2016), and may result in the unequal access to services in the commu-nity. In some coastal areas, the risk of flooding and erosion may force the relocation of someneighborhoods or the entire community (Working Group on Adaptation and Climate Resilience,2016). Therefore, the social factors component of the evaluation framework deals with the socialimplications of the strategies.The SF1 component addresses to the acceptability of the strategies. Local governments mayface challenges in communicating and discussing some strategies due to the unequal social andeconomic implications of strategies, and the emotional toll of imagining a future where people88may lose their homes, livelihoods and communities (Rijsberman and van Velzen, 1996; Barronet al., 2012). Although challenging, the acceptability of the strategies should be included in theadaptation processes (Flynn et al., 2018). The SF2 component targets the additional social op-portunities and community benefits strategies may provide during the development, implemen-tation and maintenance phases. These benefits may include increased community awareness andeducation on the impacts of climate change, natural and sensitive environments, and social andecological relationships. The additional benefits may also include increases in social networksand trust within the community. Strategies can foster opportunities for knowledge sharing andtransfer within the community members, contributing to the community’s willingness to act,capacity, preparedness, and overall resilience (The William D. Ruckelshaus Center, 2017).The SF3 component focuses on the public access to coastal areas. As the coastal populationsgrow and impacts of sea level rise increase, public access to coastal areas decrease (Higgins,2008). Providing and improving safe and continuous public access to coasts is vital for pro-moting tourism and recreational opportunities, and contributes to individual and communityhealth (Kim and Nicholls, 2017). The SF4 component addresses the preservation of the eco-logical and cultural values. Sea level rise will impact the cultural and ecological heritage sitessuch as archaeological sites, historical monuments, and structures threatened by coastal erosion(Marzeion and Levermann, 2014). They can be barriers of the adaptation strategies (Burcharthet al., 2007; Graham et al., 2013); therefore the decision-makers should consider the implica-tions on strategies on these significant sites.Lastly, the SF5 component focuses on the displacement of populations. As the flooding and ero-sion risks increase, the socio-economically disadvantaged populations will likely be unevenlyimpacted by the impacts of such events and may be forced to relocate (Leichenko and Silva,2014; Pramanik, 2017). The decision-makers should carefully consider the implications of dif-ferent strategies on the displacement of populations and specific population groups.The evaluation framework assesses the effects of an adaptation strategy across a range of com-munity resilience concepts: coastal processes, natural and built environments, and economic,institutional and social factors. Using the framework, decision-makers can identify the differenttypes of advantages and disadvantages an adaptation strategy may offer and can highlight thetradeoffs between them.893.4.3 Application of the evaluation framework to the adaptation strategiesThe evaluation framework was applied to the ‘Protection with THS’, ‘Protection with CGI’,‘Built environment accommodation’, and ‘Retreat from the coast’ scenarios. In the evaluationframework, the +1, 0, and -1 evaluation scale is based on how the strategy’s impact would com-pare (increases, no impact or decreases) to the baseline scenario. Therefore, the ‘Do nothing’scenario was used to compare other scenarios to the baseline conditions. The literature and plan-ning document reviews, illustrative scenario visuals, and researcher knowledge of the study areawere used to complete the evaluation framework. For the purposes of demonstration, modulesand components were weighted equally. The completed evaluation framework and scores of thescenarios can be found in Appendix B.6 and summarized in Figure 3.4.Figure 3.4: Diagram showing the results of the evaluation framework application to four sea level riseadaptation scenarios.90The results show that overall, the ‘Protection with CGI’ scenario scored the highest followed bythe ‘Built environment accommodation’, and ‘Retreat from the coast’ scenarios. The ‘Protectionwith THS’ scenario scored the lowest overall. However, the scores for individual modules weredifferent.• For the coastal processes and natural environment modules the results were similar to theoverall scores: the ‘Protection with CGI’ scenario scored the highest and the ‘Protectionwith THS’ scored the lowest. The ‘Retreat from the coast‘ scenario also scored high forthe natural environment module.• For the built environment module, the ‘Protection with CGI’ scenario scored better thanthe other scenarios. The ‘Retreat from the coast’ scenario had the lowest score, but wasvery close to the ‘Protection with THS’ and ‘Built environment accommodation’ scenar-ios.• For the economic factors module, both of the protection scenarios scored the highest, eventhough the overall score for the economic factors module was very low compared to othermodules. The ‘Built environment accommodation’ scenario scored the lowest.• For the institutional factors module, the ‘Built environment accommodation’ scenario hadthe highest score, followed by the ‘Protection with THS’ scenario. The ‘Retreat from thecoast’ scenario had the lowest score in this module.• For the social factors module, the ‘Protection with CGI’ scenario scored better than theother scenarios, and the ‘Retreat from the coast’ scenario scored the worst.The results indicated that overall CGI is a valuable adaptation measure with high benefits tocoastal processes, natural and built environment, and economic and social factors, but low insti-tutional benefits. The results also showed that the ‘Built environment accommodation’ scenariohad significant economic and institutional trade-offs. The trade-offs of the ‘Retreat from thecoasts’ scenario were between the coastal processes and natural environments, and the eco-nomic, institutional, and social factors. Lastly, the results showed that even though the ‘Protec-tion with THS’ scenario had positive impacts on the institutional and economic factors, it hadsignificant negative implications on the local coastal processes and natural environment.913.5 DiscussionThis study incorporated public perspectives and expert inputs to develop place-appropriate sealevel rise adaptation scenarios and a resilience-based evaluation framework to assess benefitsand trade-offs of different adaptation strategies. The expert participation in this study pooleddiverse expertise from local, regional and provincial governments. It allowed for a more robustreview of the technical data used in the study. The public participation helped to enhance knowl-edge on the local impacts of sea level rise, and the adaptation strategies available to deal withthese impacts. Besides, it helped to identify the local perspectives on the adaptation strategiesand the ones that are most preferred.The incorporation of the community’s concerns and desires improves local adaptation (Barronet al., 2012), but it may not always reflect the community’s best options to reduce risks associ-ated with sea level rise. As seen in the participatory workshop, the managed retreat and avoidstrategies were rated as the most effective adaptation strategies, yet when asked which strategiesthe participants would like to see implemented in their community, the protect and accommodatestrategies were rated higher than the managed retreat and avoid strategies. Considerations suchas the emotional and cultural connections to places, desire to have waterfront properties, fears ofdecreases in real estate prices, and group or individual agendas, can influence decisions on sealevel rise adaptation (Burch et al., 2010). Even when some coastal areas are more likely to expe-rience flooding events regardless of the adaptation strategies in place, the continued occupationof coastal areas is often preferred due to these considerations.Understanding the full implications of the adaptation strategies, therefore, is very important fordecision-making. The evaluation framework developed in this study provides a comprehensive,structured and straightforward method for understanding the broader implications of the adap-tation strategies, using a resilience lens. It provides a high-level assessment of the strategiesin a local context, combining the individual characteristics of communities and the site-specificattributes of strategies. Different than other assessment tools, the evaluation framework high-lights impacts to whom, by separating local governments, and property owners. It incorporatessocial and institutional factors, which are often left out of or overlooked in such assessmentsdue to the difficulties in measuring and quantifying them (Oulahen et al., 2017). The +1 to-1 scoring system, although coarse, allows the evaluation of strategies with minimal technicalknowledge or reliance on technical studies, making it accessible to communities with varying92levels of resources. Moreover, this scoring system and the comparison of the proposed strate-gies to existing baseline conditions help inclusion of social and institutional components. Theweighting system of the evaluation framework provides flexibility for communities to adjust theassessment to address their priorities (equal weights were used in the application). Allocating60 points to six modules (6 to 30 points per module) and 25 points to five components (2.5 to12.5 points per component) requires communities to consider all of the evaluation frameworkmodules and components. Besides providing a holistic and integrated approach, this weightingsystem prevents the dismissal of some modules and components that may not be consideredotherwise.The evaluation framework can help foster a transparent and integrated decision-making processfor adaptation. Therefore, the public participation must be recognized as an integral part of theevaluation framework, and public events should be organized before or after the completion ofthe evaluation framework. In addition, to the integration of the public in the planning process,incorporation of the expert input while completing the evaluation framework can provide trans-parency to and justification of the decisions. Therefore, a team of local experts, which bringstogether diverse perspectives and expertise should complete the evaluation framework.Although the evaluation framework is comprehensive and provides a straightforward structure,it has several limitations. Compromises have to be made so that the users can use this tool with-out the need for detailed economic or technical studies. Therefore, not all important resilienceconcepts of adaptation were included. For example, the magnitude of the economic implicationsof strategies can be a deal-breaker when considering adaptation options. However, this wouldrequire detailed site assessments and cost-benefit studies. Communities would have to completesuch studies to be able to use the tool, which was not the intended purpose of the evaluationframework. Moreover, it was also challenging to format concepts into a +1 to -1 scale and iden-tify the potential changes compared to the baseline conditions. This is because some conceptsoccur independently from the proposed strategies or the baseline conditions, thus predicting theimplications of strategies would not be possible. For example, while building functional col-laborations with other communities and jurisdictions is acknowledged to be a significant factorfor resilience, it may exist (or not) independently from the strategies. In fact, they depend heav-ily on the leadership and local government will, and capacity to undertake appropriate publicengagement events.The scenarios developed in this study conceptually illustrate the adaptation strategies and their93implications in a simple and visual way. The expert meetings and participatory workshop pro-vided the background material and data for the scenarios. Four adaptation scenarios were de-veloped to illustrate the future conditions of the study area when four different strategies wereadopted. A baseline scenario was also developed to understand future conditions if no actionswere taken. The evaluation framework was applied to the four scenarios using equal weightsfor each of the framework’s modules and components. The results highlight the site-specificbenefits and disadvantages of the adaptation strategies and the trade-offs between them.First, the results supported the growing body of literature, indicating that overall the CGI is avaluable adaptation measure with various co-benefits. Indeed, the ‘Protection with CGI’ sce-nario performed the best amongst all scenarios. It scored the highest positive benefits for thecoastal processes, natural and built environment, and economic and social factors modules. Thisconfirms the findings of the previous studies citing the wide range of CGI benefits. Even thoughthe score of the economic factors module was low compared to other modules, CGI performedrelatively high. It should be noted that this score does not reflect the actual cost of implementingCGI, but indicates the increases and decreases in the costs that private owners and local govern-ments have to cover. Here, the coastal jurisdiction and ownership play an important role. Theeconomic costs on private owners will likely to increase with a ‘Protection with CGI’ scenariobecause the improvements in the riparian areas will fall under private property boundaries. Onthe other hand, the economic costs on local governments will likely decrease because the fore-shore falls under provincial jurisdiction; thus the improvements in the foreshore area become aprovincial responsibility. The results also indicate that there are institutional trade-offs of thisscenario. The low institutional score can be explained by the lack of experience, expertise, anddata on CGI in this region. The CGI is a relatively new field of practice in British Columbia, andthe expertise and resources are limited at the moment. If the study area was in an area with moreestablished CGI practices, for example in Washington State, then the results of the institutionalfactors module would likely be much higher.The results indicated that the ‘Built environment accommodation’ scenario was the second bestoption for the study area. It performed average or above average in most modules. However, theresults highlighted that there were significant trade-offs between the economic and institutionalimplications of this scenario. On the one hand, the score of the institutional factors modulewas the highest amongst all scenarios. This is due to the existing institutional capacity of thecommunity to undertake measures for the accommodate strategy. This is because the structural94changes to buildings and upgrades in roads and infrastructure are services that are readily pro-vided by the local government. On the other hand, the score of the economic factors module wasthe worst because actions related to the built environment retrofitting, and the associated imple-mentation and maintenance costs fall under both the private owners’ and local governments’responsibility. Even though there were some benefits through sustained development and busi-nesses in the coastal area, this scenario will likely to have significant economic implications forthe community.One of the most interesting results was the trade-offs of the ‘Retreat from the coasts’ scenario.The retreat strategy is known to be a long-term solution for dealing with the sea level riseimpacts. Indeed, the benefits of this scenario on the coastal processes and natural environmentswere very high. Allowing rising water levels to reclaim flood-prone areas and preventing newdevelopment from taking place enhances the sediment and wave processes, and the health ofnatural environments. However, the results showed that besides these high positive benefits,this scenario may have significant negative implications on the economic, institutional, andsocial components. Even though the avoid and retreat strategies used in this scenario do nothave maintenance costs, the discontinuation of the development and economic activities willlikely to have significant implications on the revenue collected by the local government, as wellas impacts on the community’s livelihoods. Besides, the cost of land acquisition will createa significant economic burden, especially in British Columbia where the property prices arevery high, and there are insufficient federal and provincial financial resources to cover this cost.Institutionally, the implementation of this scenario will likely be extremely challenging becausethere are hardly any local or regional precedence, guidance, or resources for local communities.Moreover, the existing examples are typically from small coastal areas with low populationsand assets. This scenario will also be very complicated to implement. A number of jurisdictionshave to be involved in the development, design and decision-making processes but the rolesof these jurisdictions are often unclear. Socially, the relocation of the community will be verydifficult to communicate and the social acceptability of it will be very low compared to otherstrategies because this scenario is initially based upon the displacement of communities. Andnot surprisingly people do not want to leave their homes and communities they have known foryears and move somewhere else.The ‘Protection with THS’ scenario, which extended the existing seawalls throughout the coast-line, performed the worst overall. It has similar build environment, institutional and social95component scores with the ‘Built environment accommodation’ scenario. However, the coastalprocesses and natural environment implications of this scenario were the worst amongst all.This result also supports the literature citing the false sense of security and significant degrada-tion on coastal processes and natural environments caused by the THI. However, this scenarioperformed relatively high for the institutional factors module. This is mainly due to existinginstitutional capacity to implement such measures. As described in the previous sections, someparts of the study area are already protected using sea walls. Thus the expertise, data, andother resources to implement and monitor such measures already exist in the community’s in-stitutional capacity. This scenario had the same the economic factors module score as the CGIscenario. It had similar negative economic costs on the private property owners due to the re-sponsibility to implement and maintain seawalls within their property boundaries. However, ifan another THS measure were chosen rather than the sea walls, such as breakwaters, groins orjetties, the economic implications of this scenario would be very different as the responsibilityof implementation and maintenance would be under another jurisdiction. Similar to the CGIscenario, the economic activities at the coasts were maintained, and there was no increase in thecosts for local governments.Each component and module were weighted equally in this study. However, the results showedthat the place and community-appropriate weighting is critical to help decision-making, es-pecially when the trade-offs of the strategies are extreme. This is seen with the institutionaltrade-offs of the ‘Protection with CGI’ scenario; the economic trade-offs of the ‘Built environ-ment accommodation’ scenario; the economic, institutional and social trade-offs of the ‘Retreatfrom the coasts’ scenario; and the coastal processes and natural environment trade-offs of the‘Protection with THS’ scenario.Low overall economic scores of the scenarios suggested that any adaptation action will likelyincrease the economic burden on coastal communities as a whole. Economic considerationshave been typically the most important factor while choosing an adaptation strategy, but theyshould not be the only ones. Other factors also can be barriers to or facilitators of successfulimplementation of actions and sustained improvements to them. By weighting the importance ofthe modules and components, communities can manage these trade-offs. In addition, the resultsalso pointed out to the actionable modules that can weaken the differences in the trade-offs.For example, the trade-offs of the built environment module can be lessened by the changes inthe building codes and zoning plans. Similarly, the differences between the institutional factors96module scores can be reduced by the changes in the local and provincial institutional frameworksand arrangements. However, these changes have been proven to be very difficult to achieve.The importance of the site-specific adaptation has already been recognized widely in the liter-ature. The results of this study highlight that the place-appropriate adaptation options are notonly essential to identify who and what needs to adapt, and the local capacity to undertake ac-tions; but also to determine which levels of governments are responsible for implementation,and the roles of the private property owners in adaptation processes. As shown in Chapter 4,the beach and the foreshore are often owned by private owners in Washington State; thereforethe economic and institutional implications of strategies would be different than in jurisdictionswhere the private ownership of coastal areas is not the case. In British Columbia, the jurisdic-tional boundaries of local and provincial governments played a significant role in determiningthe economic and institutional implications of strategies.3.6 ConclusionThis study develops a comprehensive but straightforward tool to evaluate the sea level rise adap-tation strategies. It does so by incorporating local perspectives and diverse technical expertise,and by operationalizing the resilience concept in adaptation. The methodological approach ofthis study, the tool developed for strategy evaluation, and the demonstration on four adapta-tion scenarios provide an in-depth understanding of (1) how the resilience framework can beincorporated in the adaptation planning, (2) the factors influencing adaptation and communityresilience, and (3) the implications and trade-offs of adaptation strategies, particularly CGI’s.This study demonstrates that CGI can be a viable adaptation measure if communities choose toprotect their coastlines. It highlights the positive trade-offs of CGI in the interaction with andprotection from coastal processes, the conservation and enhancement of natural environments,and the protection of the assets in built environments. The study also shows that communitiesmay not always have the capacity or the organization structure to undertake CGI actions. Theimprovements and changes have to be made to the institutional frameworks and arrangementsso that the CGI’s negative institutional implications can be weakened.When dealing with the sea level rise impacts, it is likely that communities will choose a combi-nation of different strategies, rather than a single one. This way they will be able to address var-97ious concerns and get the best out of each strategy. However, regardless of the type of strategy,communities’ adaptation processes, the use of assessment tools, the successful implementationof strategies, and the continuous updating of the plans and policies will heavily rely on the activeleadership of the knowledgeable and skilled practitioners and politicians in the community.Disclaimer: This research is not commissioned by the District of North Saanich. The techni-cal aspects of this work were not completed or supervised/reviewed by a professional engineerof the District. The work completed and any conclusion drawn as part of this academic re-search does not necessarily represent the position of the District. The analysis completed andinformation collected as part of this research is not intended to direct or influence future activi-ties/initiatives/studies completed by the District vis-a-vis sea level rise.98Chapter 4Institutional barriers to andfacilitators of coastal greeninfrastructure implementation: Acomparative study in British Columbiaand Washington State4.1 IntroductionCommunities have long found ways to adapt to changing environmental conditions (Adger,2003; Lo¨f, 2013). They have done so by using different ways and capacities to adapt (Berrang-Ford et al., 2011; Smith et al., 2011). Institutions, from local to federal governments and or-ganizations, and their arrangements have played important driver roles in adaptation (Bettiniet al., 2015). They have paved the way for key changes in the way communities deal with ad-verse environmental events, such as implementing coastal green infrastructure (CGI) measuresto protect coasts from rising sea levels. However, research on the institutional arrangements thatimpede CGI implementation or drives it has been very limited (Matthews et al., 2015).CGI refers to the natural or nature-based (designed or engineered) systems that protect coastsfrom flood and erosion and provides numerous essential ecosystem services (Barbier et al., 2011;Bridges et al., 2015). CGI is often implemented in the form of coastal dunes, intertidal vegeta-tions, barrier islands, sloped beaches (beach nourishment) with large woody debris, coastal andriparian habitats, and reef systems. Over the past few decades, CGI has gained increased atten-tion (Sutton-Grier et al., 2018) because decision-makers have started to recognize the maladap-99tive nature of the traditional hard structures and their implications on coastal habitats (Hewitsonet al., 2014). Decision makers also started to acknowledge the adaptive, multi-functional andlow-cost and low-maintenance nature of CGI (USEPA, 2015).Although CGI’s recognition has increased around the world, its implementation has gained mo-mentum at various levels in different places. The mainstreaming of CGI has been limited insome parts of the world due to the context-specific institutional barriers, amongst others (TheHorinko Group, 2015; Sutton-Grier et al., 2018). For example, Washinton State has startedto include CGI related coastal and environmental regulations, develop programs to fund CGIimplementation, and increase education and knowledge on CGIs as early as 1975 (U.S, 1972;Houle and Macdonald, 2011). However, the progression of CGI during the same time has beenrelatively limited in British Columbia. British Columbia and the Washington States share a bor-der, history, and culture, and have similar environmental and ecological characteristics and con-cerns (Simeon and Radin, 2010). Yet, the presence of two different countries (Canada and theUnited States), and corresponding institutional arrangements and regulatory frameworks havecontributed to the differences in the implementation of CGI (Naumann et al., 2011; Carlsson-Kanyama et al., 2013).The literature suggests that the institutional arrangements play a significant role in the slow pro-gression of CGI implementation in some places, and the rapid diffusion in others. Specifically,constraints and opportunities related to the governance structure and corresponding authoritydistributions; financial resources dedicated for adaptation; leadership; regulations and mandatein place; senior government support; political will; organizational capacity; and communicationhave been reported to influence adaptation (Adger et al., 2007). However, there has been insuf-ficient research looking into the CGI specific institutional barriers and facilitators. Identifyingand understanding the institutional factors contributing to these barriers and strategies helpingto these facilitators provide opportunities to increase the implementation of CGI.Therefore, this study investigates the barriers to and facilitators of CGI implementation that arerooted in the institutional arrangements of British Columbia and Washington State. It aims toanswer the following research question: what are the institutional barriers to and facilitators ofCGI implementation? It operationalizes three methods, a review of the CGI projects, a reviewand synthesis of the institutional arrangements, and semi-structured interviews. Chapter 4 is or-ganized as follows. Section 4.2 provides a background review of the relevant literature. Section4.3 explains the methodological approach of this study, including the description of the study100area and specific research activities. Section 4.4 provides and explains the results of the researchactivities. Section 4.5 discusses the findings, highlights the barriers and facilitators identifiedthroughout this research, and identifies the limitations of the research methods and findings.Lastly, Section 4.6 provides concluding remarks.4.2 BackgroundThe rapid acceleration of climate change impacts such as sea level rise throughout the 21st cen-tury, driven both by natural and anthropogenic forces (IPCC, 2014), and increased complexityand diversity of communities (Bauer et al., 2012) have made adaptation no longer an option, buta requirement for most coastal communities. Owing to the mitigation efforts that were unableto sufficiently to cut global greenhouse gas emissions and the urgency of action, adaptation hasgained significant academic and political interest (Bauer and Steurer, 2014; Berrang-Ford et al.,2011). One adaptation measure, coastal green infrastructure (CGI) has been praised not only forits ability to protect coasts from flooding and erosion but also to provide an adaptive and a rela-tively low cost and low maintenance alternative to traditional hard coastal protection structures(Barbier et al., 2011).It has been long discussed that adaptation, and therefore CGI should be a local level responsi-bility (Nicholls, 2011) because the climate change impacts are observed, felt and dealt with atthe local level (Adger et al., 2005). While this is often the case, the notion that adaptation hasto be dealt with at the local level has led to the absence of the regional and national level coor-dination, authority distribution, regulations, partnerships, programs, and capacity and resourcesin Canada and the United States (Hassol and Udall, 2003; Lorenzoni and Pidgeon, 2006; Burch,2010; Giest and Howlett, 2013). This absence is problematic because operationalizing adapta-tion actions, especially for emerging measures such as CGI, relies heavily on these institutionalarrangements and the resulting social, cultural, political and regulatory environment (Lo¨f, 2013;Mguni et al., 2015).Institutional arrangements, therefore, are significant pieces of mechanisms fostering CGI imple-mentation. Institutional arrangements in this context refer to the hierarchies, structures, orga-nizational features and functioning of the formal governments and organizations (Glavovic andSmith, 2014; Oulahen et al., 2017). They define the roles of different levels of governments and101their decision-making authorities. They influence regulations and policies, and impact adoptionof one strategy or policy over another one (Murphy, 2014). They provide structure, finan-cial assistance, technical guidance and expertise, tools and other essential resources to fosterCGI policies and practices (Farber, 2009; Lo¨f, 2013; Reckien et al., 2015). Therefore, institu-tional arrangements are recognized to be both barriers to and facilitators of CGI implementation(Mozumder et al., 2011; Uittenbroek et al., 2012; Reckien et al., 2015).The institutional barriers to and facilitators of adaptation have been considered as context-dependent (Biesbroek, 2014). This is because they have displayed variation across differentlevels of governments (Reckien et al., 2015). Countries, regions, and communities reported thatthey experience different barriers to CGI implementation; so do local, regional and nationalgovernments. Barriers in this context refer to the obstacles (Moser and Ekstrom, 2010) andchallenges resulting from our social, cultural, historical and intellectual constructs (Biesbroek,2014). Therefore, they are not immutable (Hamin and Gurran, 2015); they can be changed andreorganized (Glavovic and Smith, 2014). Facilitators are opportunities and pathways to over-come barriers. They lead ideas to policies and policies to actions. Many barriers reported inthe literature are also facilitators if the conditions creating these barriers can be overturned oreliminated (Uittenbroek et al., 2012).The institutional barriers and facilitators of adaptation have only started to be recognized in themid-2000s with Adger et al. (2007) (Barnett et al., 2013). Yet, the existing literature has pri-marily focused on the climate change and (sea level rise) adaptation (Biesbroek, 2014). Thereis a long list of institutional barriers to and facilitators of adaptation reported in the literature,covering a range of institutional factors (Biesbroek, 2014; Hamin et al., 2014). The most com-mon barriers include the governance and decision-making authority; financial; leadership; reg-ulations and mandate; senior government support; political will; organizational capacity; andcommunication.Arguably, the most common and important reported barrier to and facilitator of adaptation isthe governance structures and allocation of decision-making authorities (Moser and Ekstrom,2010; Barnett et al., 2013). The ambiguity over which levels of governments and which depart-ments are or should be responsible for implementing adaptation measures remain to be unclearin many places (Hansen et al., 2013; Ziervogel and Parnell, 2014; Mguni et al., 2015). In somecases, these responsibilities may be overlapping, leading to increased complexity and bureau-cracy (Lemmen et al., 2008; Keeley et al., 2013). Moreover, even in places where the roles of102local, regional and national governments’ are clearly defined, problems related to the fragmen-tation of key decision-making authority, and regulatory roles of different levels of governmentscreate another barrier (Keeley et al., 2013; Chaffin et al., 2016; Hopkins et al., 2018). In addi-tion, the lack of communication and interaction between different institutions creates inefficientintergovernmental coordination and partnerships (Lo¨f, 2013; O’Donnell et al., 2017).Another common barrier and facilitator in the literature is financial (Adger et al., 2007; O’Donnellet al., 2017). The reported financial barriers of adaptation are as follows. One important issueis the lack of or insufficient budget within the organizations to understand risks, and develop,assess and implement solutions (Burch, 2010; Mozumder et al., 2011; Hansen et al., 2013).The next one is the limited high governmental level financial assistance, grants and programsto address adaptation adequately (Uittenbroek et al., 2012; Ziervogel and Parnell, 2014; Reck-ien et al., 2015). Another issue is the restrictions in financial resources to allocate spending todesign, construction, or monitoring effectiveness (O’Donnell et al., 2017; Sutton-Grier et al.,2018). Without the financial mechanisms and resources in place, overcoming other barriers isless meaningful for CGI implementation (Keeley et al., 2013).Leadership is also one of the most common barriers and facilitators cited in the literature (Moserand Ekstrom, 2010; Mozumder et al., 2011; Carlsson-Kanyama et al., 2013; Reckien et al.,2015). Leaders or champions with decision-making authorities can propose policies for adap-tation and advocate for new approaches. Particularly in the absence of senior level guidanceand regulatory support, adaptation depends on the characteristics of the local decision-makers(Ziervogel and Parnell, 2014). The lack of such leadership (Hamin et al., 2014; O’Donnell et al.,2017) or incompetent leadership (Uittenbroek et al., 2012) creates many barriers to adaptation,while the presence of it will increase the momentum to it (Ziervogel and Parnell, 2014).The literature suggests that regulations and organizational mandates also hinder or drive progresson adaption (Moser and Ekstrom, 2010; Barnett et al., 2013; O’Donnell et al., 2017). Institutionsthat are keen to implement adaptation measures are often limited in their ability to do due to thelack of regulations and government programs that support adaptation (Houle and Macdonald,2011; Lo¨f, 2013; Hamin et al., 2014), and regulations that do not integrate climate change im-pacts and adaptation measures (Mozumder et al., 2011; Carlsson-Kanyama et al., 2013). Thelack of space and flexibility in the current regulatory environment to accommodate adaptationmeasures are considered to be an important regulatory issue (Sutton-Grier et al., 2018). In addi-tion, conflicting codes and rules within the existing regulations also hamper progress (US EPA,1032015; O’Donnell et al., 2017).The senior government support, usually in the form of guidance, regulations, programs, partner-ships, tools, and data is another critical barrier and facilitator noted in the literature (Moser andEkstrom, 2010; Mozumder et al., 2011; Barnett et al., 2013; Reckien et al., 2015). The absenceof the senior level support forces other levels of governments to rely solely on their organiza-tional capacities. This absence leads to inequalities due to the institutions’ ability to undertakeadaptation actions by themselves (Dickinson and Burton, 2011). On the other hand, the pres-ence of the senior level support can prevent disparate approaches to adaptation and promotes aharmonious regional adaptation response (Kittinger and Ayers, 2010).Another frequently reported barrier and facilitator is political will. The political will barrierrefers to the politicians’ lack of motivation to consider long-term implications of adaptationmeasures due to the reelection concerns (Ziervogel and Parnell, 2014). It is considered to be akey barrier (Burch et al., 2010; Reckien et al., 2015), because the absence of political supportleads to policies and actions focusing on short-term gains (Uittenbroek et al., 2012; Ziervogeland Parnell, 2014). It also prevents the long-term vision necessary for CGI implementation.The organizational capacities of institutions, including having the right expertise in the orga-nization (Moser and Ekstrom, 2010; Keeley et al., 2013), enough staff to work on variety ofissues (Mozumder et al., 2011), and the availability and access to resources such as technolo-gies and data (Uittenbroek et al., 2012; Barnett et al., 2013) is reported to be an important driverof adaptation (Ziervogel and Parnell, 2014; O’Donnell et al., 2017; Hopkins et al., 2018). Theabsence or limitations to these organizational capacities create significant constraints to the gov-ernments’ ability to implement adaptation (Keeley et al., 2013; Hamin et al., 2014; Hamin andGurran, 2015), even if the other institutional barriers are removed.Lastly, communication is reported to be a significant barrier of adaptation. It is because com-munication issues influence perceived risks, public perceptions on the effectiveness of differ-ent measures, community buy-in, and advocacy (Moser and Ekstrom, 2010; Uittenbroek et al.,2012; Keeley et al., 2013; Ziervogel and Parnell, 2014; Reckien et al., 2015). Communicat-ing the relevant, evidence-based, and up-to-date information with public increases awarenessand reduces incorrect perceptions on effectiveness and cost of adaptation measures (Mozumderet al., 2011). The language used and the approach taken for the interactions with commu-nities improves institution-to-institution and institution-to-community communication (Moser104and Ekstrom, 2010). Communication fosters transparency and trust in relationships. Ensuringthat the relevant information is available to and accessible by public fosters local advocacy foradaptation (Hamin et al., 2014; Measham et al., 2011).Since 2013, the barriers to and facilitators of adaptation measures, particularly in the urbanstormwater management context, has also started to gain attention. Studies of Keeley et al.(2013), Matthews et al. (2015), Mguni et al. (2015), Thorne et al. (2015), Chaffin et al. (2016),O’Donnell et al. (2017), and Hopkins et al. (2018) were the first attempts to characterize thefactors impeding the use of urban green infrastructure for managing urban stormwater. Findingsof these studies show that urban green infrastructure implementation is constrained by similarbarriers to adaptation in general, but it also facing unique and specific challenges (Chaffin et al.,2016). Thorne et al. (2015) suggested that ensuring social equality, while delivering essentialurban services is a barrier of urban green infrastructure implementation. Chaffin et al. (2016)added that functionality of land use changes due to green infrastructure implementation is an-other important barrier. O’Donnell et al. (2017) concluded that negative past experiences ofmunicipal practitioners, low priority of urban stormwater issues compared to other essential ser-vices, and lack of available space act as barriers. Matthews et al. (2015); Hopkins et al. (2018)suggested that the path dependency such as long-term investments in grey infrastructure limitswill to change. Lastly, Keeley et al. (2013); O’Donnell et al. (2017); Hopkins et al. (2018) con-cluded that the complexity and scale of the stormwater management systems and consequentmanagement issues could prevent implementation of urban green infrastructure.The barriers to and facilitators of adaptation described in this section are reported to be rela-tively common across different places. There are undoubtedly other factors that impact adapta-tion implementation, but not listed here. The barriers detailed here are typically challenging toovercome because institutional change is complicated and takes a long time to achieve; yet canbe accomplished. Facilitators, often overshadowed by the barriers, can provide key opportuni-ties and strategies to achieve institutional change and overcome related barriers (Glavovic andSmith, 2014).This review showed that the investigation into the institutional barriers to and facilitators of dif-ferent adaptation measures has finally started. The early studies indicated that besides the sim-ilarities, specific adaptation actions face unique challenges and influenced by different drivers.The review indicated that urban green infrastructure is influenced by specific barriers and fa-cilitators due to the complexities of urban systems, competing interests and investments, and105integration of new technologies into existing city plans and building codes. Similarly, CGI islikely to be influenced by CGI-specific barriers and facilitators due to its unique location at theinterface between land and ocean (IPCC, 2014), the overlapping jurisdictions that regulate andmanage the coastal areas, the lack of knowledge on its coastal protection role, and the unknownssurrounding its integration in the existing regulatory frameworks. This review highlighted thatthere are currently no studies investigating the specific institutional barriers to and facilitatorsof CGI. Therefore, this research aims to fill this gap using a comparative approach and threedifferent and complementary methods.4.3 MethodsThis study operationalizes a mixed-method approach to identify the institutional barriers to andfacilitators of CGI implementation. The study compares CGI projects, institutional arrange-ments and practitioner perspectives from British Columbia (BC), Canada and Washington State(WA), the United States. The main hypothesis of this research is that the implementation ofCGI is influenced by common barriers to and facilitators of adaptation as well as CGI-specificdisablers and enablers. The methodology of this study consists of the following three system-atic steps, each aiming to contribute to the overall research question of what are the institutionalbarriers to and facilitators of CGI implementation, using three sub-research questions.The first step is the review, analysis, and comparison of the CGI projects implemented in thestudy area, between 2008-2018. This step aims to answer the sub-research question what are thedifferences between the institutional roles in CGI implementation in BC andWA? It investigatesthe number, type, and area of the projects implemented in each region, as well as the rolesand capacities of different levels of institutions in project implementation. By comparing thequantitative data in BC and WA, this step aims to provide an understanding of how often, inwhat capacity, and through what resources the CGI projects were implemented in the studyarea.The second step is the review and synthesis of the institutional arrangements. It includes theliterature and document review of the governance structures and consequent distribution of thedecision-making-authorities, coastal jurisdiction, and coastal and environmental regulations andprograms in BC and Canada, and in WA and the United States. This step poses the sub-questionwhat are the differences between the institutional arrangements of different levels of govern-106ments in BC and WA? This step investigates the published academic and grey research as wellas the official federal and provincial government websites for acts and programs. By compar-ing the qualitative data in BC and WA, this step aims to highlight the institutional barriers andfacilitators through identifying relevant responsibilities of the different levels of governments,jurisdictional and ownership boundaries, and regulations and programs.The last step is the semi-structured interviews with the practitioners from different governmen-tal and non-governmental organizations (NGOs) in the study area. It poses the sub-researchquestion what are the practitioners’ perspectives on the institutional barriers to and facilitatorsof CGI implementation in BC and WA? Semi-structured interviews are conducted in this step toobtain qualitative (the types of barriers and facilitators) and quantitative data (i.e., the numberof people that mentioned the same barriers and/or facilitators). The qualitative data is used tocheck whether the practitioners mentioned the barriers and facilitators similar to the findings ofthe CGI project review and the review and synthesis of the institutional arrangements steps. It isalso used to identify the CGI specific institutional barriers and facilitators and to provide morenuanced and detailed factors to the common barriers and facilitators. The quantitative data isused to highlight the differences between BC and WA.The comparative design of the research methods helps to understand the broad patterns, simi-larities and differences (Lachapelle et al., 2012) in the organizational, structural and functionalaspects of institutions in these two regions. Using the comparative research approach, this studyaims to provide insights into the barriers and facilitators that are experienced in each region dueto the differences in their institutional arrangements. While an in-depth study focusing on oneregion may have yielded a more in-depth understanding of the institutional arrangements, andcorresponding jurisdictional issues and regulatory frameworks, it would not have provided a wayto understand how these arrangements act as barriers to and facilitators of CGI implementation.4.3.1 Study AreaThe coastal regions of British Columbia (BC) in the Pacific coast of Canada and WashingtonState (WA) in the Pacific Northwest of the United States were selected to for the study area ofthis research (Figure 4.1).107Figure 4.1: The map of the study area showing the coastal regions of British Columbia and WashingtonStatesThis study area provides unique opportunities to identify institutional barriers to and facilitatorsof CGI implementation in a comparative way because of the following reasons. The Cana-dian and American governments are both liberal democracies and federations (Lachapelle et al.,2012). They have relatively similar organizational structures: the federal government, FirstNations and Tribal governments; provincial, territorial, and state governments; and local gov-ernments. Also, the study area includes an important body of water, the Salish Sea, which hostsmost of the BC and WA populations and economic activities. In addition, the coastal populationin both regions is very high compared to the rest of the regions. It is estimated that around70% of Washington State’s 7.2 million resident (2016 census), roughly 5 million people live incoastal counties (Adelsman et al., 2012), where in BC, around 78% of BC’s of 4.6 million resi-dent (2016 census), roughly 3.6 million people live in coastal municipalities (Statistics Canada,1082016). Moreover, both coastal regions have similar geomorphological, ecological and oceaniccharacteristics (Demarchi, 2011). These characteristics are not uniform but are similar in theirdiversity throughout the study area. Therefore, the adaptation measures implemented in oneregion can be considered in the other one. Lastly, Canada and the United States, and BC andWA often experience the flow of ideas, culture, and policies (Simeon and Radin, 2010), furtheradvancing the knowledge sharing for adaptation in the region.BC and WA also have several significant differences that should be mentioned here. For exam-ple, the length of the WA coastline (5,310 km) is about five times shorter than the BC coastline(25,725 km). Considering BC’s population is smaller than WA’s population, the BC coasts arenot as densely developed and occupied as the WA coasts. Moreover, BC and WA face similarrates of projected sea level rise rates (1m) for 2100, but the coastal subsidence due to sedimentcompaction and groundwater extraction in parts of WA, and uplift due to geological and tectonicprocesses in parts of BC cause spatial variations of coastal flooding and erosion risks (Mazzottiet al., 2008). The variation in the risks they are exposed to may lead to differences in adaptationneeds.Despite their differences, BC and WA have been subject to numerous other comparative stud-ies. These studies included investigating various issues such as coastal zone management (Dayand Gamble, 1990), environmental management (Norman and Melious, 2004), adaptive man-agement (Halbert, 2008) and governance models (Wolman, 2017). The similarities these tworegions share, overtake the differences they have and provide a unique opportunity to investigatethe institutional barriers to and facilitators of CGI implementation.There has also been a growing interest in and expertise with CGI in the region. In Canada, CGIhas started to be recognized as the part of the solution to deal with climate change. Restoringecosystems, particularly coastal ecosystems that sequester 50 times more carbon than the samearea in Canada’s boreal and temperate forests has started to become a priority (CPAWS, 2016).There has been an increasing number of federal and provincial initiatives and grant programsfor CGI projects in the last two years such as Natural Resources Canada’s Green Infrastruc-ture programs, Federation of Canadian Municipalities’ Green Municipal Fund, and the GreenInfrastructure Environmental Quality Program in BC. In the United States, EPA and NOAAhave been providing guidance and programs for CGI projects over two decades. In WA, manymunicipal and county governments, state departments, multi-organizational partnerships, andenvironmental NGOs have been removing THSs and exploring CGI alternatives.1094.3.2 Research activitiesReview and analysis of the CGI projectsThis step aims to answer the sub-research question - what are the differences between the insti-tutional roles in CGI implementation in BC and WA? To answer this question, the CGI projectsthat were implemented or are in the implementation process in the last ten years (2008-2008)were collected. These projects were reviewed and if the information was available, projectsdetails, such as project objectives, types, area size, lead institutions, funding institutions, andfunding amount were recorded. The recorded information was analyzed to understand how of-ten, in what capacity, and through which resources the CGI projects were implemented, and toidentify the differences between the institutional roles in CGI implementation in BC and WA.The projects were gathered from various publicly available databases from federal, provincial,state, and NGOs’ websites. The list of the data sources can be found in Appendix C.1. Theproject selection criteria included the following rules:• Projects have to be (approximately) within 100m of the coastline. Projects on watersheds,streams, or urban parks were not included.• Project implementation has to be completed or in the process of completion. Projects thatare in the design and development stages were not included. If the permit applicationswere submitted for review, then the project was included. If this information was notavailable, then the project was not included.• Projects have to include an on-site intervention. Land acquisition, design, feasibility,research, and mapping projects were not included.• Projects have to be on public land (federal, First Nations, Tribal, provincial or state) dueto data availability and access. Projects on private properties were not included.Once the projects were located and the selection criteria were applied, information on the projectname, type, objective, funding amount, lead institution, institutions where the funding was al-located from were systematically recorded. Information on the initial funding sources and thedetails of partnerships was not available in all cases; therefore they were not included. Thepast ten years (2008-2018) $CAD to $USD currency rates were gathered and averaged. Thefunding amount of the projects in BC were exchanged to USD. This exchange allowed for the110comparison of the funding amount of the projects in BC and WA.Review and synthesis of the institutional arrangementsThis step aims to answer the sub-research question - what are the differences between the institu-tional arrangements of different levels of governments in BC and WA? To answer this question,the review and synthesis of the institutional arrangements in Canada, BC, the United States andWAwere conducted to identify the institutional barriers and facilitators. Desk research was con-ducted to locate, review and synthesize the barriers and facilitators identified in the CGI projectreview step. These concepts included the governance systems and authority distribution, coastaljurisdiction and ownership, and the coastal and environmental regulations and programs.In the first part, the governance systems of Canada and the United States were reviewed, andthe responsibilities allocated to the provincial, state and local governments were identified. Inthe second part, coastal jurisdictions in the study area were identified using published govern-ment documents. In the third part, coastal and environmental regulations were researched onthe federal, provincial and state governments’ and departments’ websites. The contents of theseregulations were reviewed, and actions required for CGI implementation (such as permits, com-pliance, and approvals) were noted. Lastly, the federal, provincial, and state programs relatedto coastal and environmental issues were identified through a web search. If available, the startdate, the specific time period of the programs, and the amount of financial assistance (such asgrants, fundings, and others) were recorded.The differences in the study area were identified and compared to highlight the barriers to andfacilitators of CGI implementation.Semi-structured interviewsThis step aims to answer the sub-research question - what are the practitioners’ perspectives onthe institutional barriers to and facilitators of CGI implementation in their region? To assesswhat the practitioners involved in the CGI projects consider as institutional barriers to and facil-itators of CGI implementation, interviews were held with the selected practitioners in the studyarea.The criteria used to select the participants included:• The participants should be representing the provincial/state, local and NGOs that are in-volved in the regulatory, implementation, guidance and coordination of the CGI projects;111• The participants should have been actively involved in the abovementioned roles in thelast ten years and be aware of the concepts discussed; and• The participants should be willing to be interviewed as part of the research.After the participants were identified, they were contacted through their publicly available emailaddresses. After the initial contact and explanation of the research, the consent forms were sentif they agreed to participate in the study.The semi-structured interview format was selected to allow for a range of responses, and toenable clarifying questions and additional discussions. The interviews were conducted either inperson or on the phone, shortly after receiving the consent forms. The interviews were audiorecorded for future analysis. The interview protocol included five parts. The first part was aninitial introduction of the researcher, the research, and the format of the interview. After theintroduction, in the second part, the participants were asked questions about their role of theirorganization and their role in the organization. In the third part, the participants were askedabout jurisdiction, decision-making authority, and mandate of their organizations; regulationsthey are bounded by when dealing with the CGI projects; the funding sources for their organi-zations and the CGI projects; if applicable, how they use or distribute the funding; and otherlevels and types of organizations they commonly build partnerships with. In the fourth part,they were asked about what they would consider as the main barriers to and facilitators of CGIimplementation. In the last part, they were presented with the prompted questions, which in-cluded a brief introduction of the results of the CGI project review, and the review and synthesisof the institutional arrangements. They were asked whether they agree with the findings of theprevious research activities, and elaborate on their answers. The detailed interview protocol canbe found in Appendix C.2.After the semi-structured interviews were completed, the verbatim transcriptions of the audiofiles were created. These transcripts were coded using the NVIVO software. The initial cod-ing procedure followed the themes discussed in the interviews. After the initial coding of theoverarching themes, sub coding categories were created to get depth and context on the broaderthemes. The coded text was analyzed by the respondents’ region (BC or WA) and level ofgovernments or organizations they represent (provincial, state, local or NGO).1124.4 Results4.4.1 CGI project reviewA total of 235 CGI projects were gathered from the publicly available federal, provincial, state,and NGO sources listed in Appendix C.1. Out of the 235 CGI projects, 47 projects (20%) werefrom BC, and 188 projects (80%) were from WA (Figure 4.2).Figure 4.2: The number of CGI projects in BC and WA between 2008-2018The significant majority of the projects did not have adaptation as their objective (Figure 4.3).Fish habitat restoration was the main driver of the projects in BC (79%) and in WA (47%).The CGI projects with adaptation objective made only about 8.5% of the total projects in BCand 20% of the total projects in WA. The projects with the indirect adaptation objective, suchas projects restoring sediment flows and natural processes, made up about 12.5% of the BCprojects and 33% of the WA projects.113Figure 4.3: The objectives of the CGI projectsThe projects did not show significant variety in types (Figure 4.4). Most of the projects wereinitiated to restore habitat and/or to remove human-made structures such as bulkheads and cul-verts. 83% of the CGI projects in BC were habitat restoration projects, where only 17% wereboth habitat restoration and human-made structure removal. In WA, 74% of the CGI projectswere both habitat restoration and human-made structure removal projects, and only 21% of theprojects were habitat restoration projects. Different than BC, 5% of the CGI projects were alsohazard protection and climate change adaptation projects.Figure 4.4: The CGI project typesThe institutions where project funding was allocated from were very different. The state gov-ernment in WA was the main institution where the CGI project funding was allocated. 79% of114the projects received their funding from a state source (different state departments). It should benoted here that this does not mean that the funding was originated from a state department. Theorigins of the projects’ funding were difficult to locate because the initial funding sources aretypically nested in several layers of programs, and the funds often go through various channelsof bureaucratic steps before being allocated. In BC, the federal government played an importantrole in funding allocation. 47% of the CGI projects in BC received their funding from a federalsource (usually different ministers or directly from the government of Canada). In BC, localgovernments, NGOs, and partnerships also played roles in allocating funding for CGI projects.Figure 4.5: The institutions where the project funding was allocated fromDifferent institutions acted as the project leads (Figure 4.6). These organizations included thelocal governments, NGOs, provincial and state departments, partnerships, First Nations, IndianTribes, and the federal governments. The project leads undertook various roles such as projectinitiation, coordination, and implementation. The NGOs played significant roles as the projectleads in BC andWA, 51% and 42% respectively. In WA, the local, state and Tribal governments,and partnerships had more lead roles, compared to their counterparts in BC.115Figure 4.6: The CGI project leadsThe minimum funding amount of the CGI projects was $5,375 in BC and $500 in WA, whilethe max funding amount was $8,081,345 and $64,707,676, respectively. The average fundingof the CGI projects in BC was $764,427, and $1,802,752 in WA. The average funding amountallocated by different institutions was consistently and significantly higher in WA compared toBC, except for the local governments (Figure 4.7). The average funding allocated by the Federalgovernment was the highest amongst all level of institutions in BC and WA.Figure 4.7: The average CGI project funding amount allocated by institutionsThe minimum project area in BC and WA was 0.01 acres, while the max project area was 250acres and 762 acres, respectively. The average project area in WA was 51.37 acres, about 20acres larger than the average project area in BC (30.34 acres). The area size of the projects116lead by the state government and partnerships in WA was significantly higher than those inBC, the size of the projects lead by the local governments and NGOs in BC was higher thanthose in WA. In WA, the project area size increased with the level of the lead institution (Figure4.8). The CGI projects with the Tribal governments, state governments, and multi-institutionalpartnerships as the lead had larger average project areas than of the NGOs and local governmentsas the leads. In BC, the projects with the local governments and NGOs as the lead had largeraverage project areas than of the provincial governments, multi-institutional partnerships, andthe federal government.Figure 4.8: The average CGI project size in acres by the lead institution4.4.1.1 Summary of the CGI project review resultsThe project review showed that there were notable differences in the CGI projects in BC andWA. These differences were in their project numbers, objectives, types, area size, leads, fund-ing, and the institutions where the project funding was allocated. There were significantly moreprojects implemented or are in the implementation process in WA than in BC. In both regions,the project objectives were mainly fish habitat health, continuity, and access, rather than adapta-tion. The CGI project types were mainly habitat restoration projects in both places, but human-made structure removal projects were also a common type in WA. The funding for the CGIprojects was allocated primarily from the state government in WA and the federal governmentin BC. The NGOs played significant roles as the project leads both in regions. In addition, thelocal, state and Tribe governments also took important lead roles inWA, compared to their coun-117terparts in BC. The project areas varied in size but overall, the projects in WA were larger thanthose in BC. The CGI projects in WA received much more funding, compared to the projects inBC. In WA both the federal government and the state governments were the main institutionsto allocate funding for the CGI projects, where in BC the funding was allocated mainly by thefederal government and local governments.The project review demonstrated that there were clear distinctions between the roles differentgovernments in the study area played. The results show differences in the arrangements andfunctioning of the institutions in BC and WA, particularly for the institutions where the fundingwas allocated from, organizations with the lead roles, the funding availability, and the size of theprojects. Based on the literature, these differences can be defined by the governance structure,and corresponding decision-making authority allocation, jurisdictional issues, and regulatoryenvironment and programs that provide funding and other resources, and therefore were selectedto be reviewed in the next section.4.4.2 Review and synthesis of the institutional arrangements4.4.2.1 Governance systems and the corresponding authority distributionGovernance systems influence the authority distribution in different levels of governments throughthe allocation of regulatory and financial responsibilities. These systems define the structure ofthe institutional arrangements, and institutions’ decision-making authority to apply laws andimplement programs. Therefore, understanding the governance systems and the correspondingdecision-making authority distribution is significant when investigating the institutional barriersto and facilitators of CGI implementation.Federal governmentsCanada and the United States are both liberal democracies and federations (Be´langer, 2005;Simeon and Radin, 2010). However, Canada’s governance system has started as a relativelycentralized federation and has become decentralized over time, where the United State’s gov-ernance system has started as a relatively decentralized, and has become relatively centralizedover time (Be´langer, 2005; Thomas and Biette, 2014; Government of Alberta, 2015). Thesecontrasting natures of the high-level governance systems and the differences in political tradi-tions (Hamilton, 2013) have led to unique provincial and state, and local government structure,118coordination, and cooperation in Canada and the United States (Field, 1992; Simeon and Radin,2010; Thomas and Biette, 2014).Provincial and state governmentsThe influence of the different high-level governance systems in Canada and the United Statescan be seen in the decision-making authorities allocated to the Canadian provinces and theUnited States’s states. For example, the United States constitution initially allocates the statesall powers not explicitly assigned to the federal government (Be´langer, 2005; Taylor, 2005).These responsibilities are often characterized as imprecise and poorly defined but importantnevertheless. They include justice, education, and environmental protection, amongst others(Patmore, 2009; Thomas and Biette, 2014). However, over time the federal government has usedits authority to expand its jurisdiction and intervened with the broad responsibilities given to thestates (Government of Alberta, 2015). As a result, there are only a few state-only responsibilitiesleft, as most of them are now shared with the federal government. On the other hand, theprovincial responsibilities are initially limited compared to the states, but they are carefully andclearly defined in the Canadian constitution. These responsibilities include taxation, naturalresources, education, and health (Be´langer, 2005; Taylor, 2005). Yet, the division of theseresponsibilities is not always apparent in practice either. Similar to its neighbor in the South,the Canadian federal government has also used its authority to influence provincial policies anddecisions (Government of Alberta, 2015).Overall, the distinctions between provinces and states are not very clear. They depend pre-dominantly on provinces’ and states’ financial autonomy and dependency on federal regulationsfor specific subject areas such as natural resources, marine habitat, and environment. For ex-ample, the provinces are larger and financially more independent than the states. They do notdepend heavily on the federal transfers, as they raise large proportions of the provincial revenue,compared to the states (Simeon and Radin, 2010). On the other hand, states have their own con-stitutions, which do not conflict with but go beyond the federal constitution and allow states toshape their institutional structures (Arnold, 2004). In Canada, only BC has its own constitution,through the BC Constitution Act of 1996, which is not the equivalent of the state constitutionsand can be easily amended (Morton, 2004). In general, there is a consensus in the literature thatthe provinces have more decision-making authority than their American counterparts, becausethey are financially more independent (Hamilton, 2013).119Local governmentsLocal governments are not recognized under the Canadian and American constitutions (Shah,2006). Therefore, the decision-making authorities of the local governments are defined by theprovincial and state governments (Dewing et al., 2006; Hamilton, 2013). However, there aremajor differences between the authorities allocated to the local governments in Canada and theUnited States. For example, the state governments have allocated greater authorities to theirlocal governments compared to the provinces, leaving local governments in Canada with hardlyany real power (Duffy et al., 2014). In fact, the Canadian local governments are restricted intheir ability to raise revenue, and access and allocate financial resources as they are subject toa significant number of provincial rules and regulations (Dewing et al., 2006). On the contrary,the United States local governments have access to a broad range of financial resources. Thereis also a significant variation in municipalities’ access to revenue tools (Kitchen, 2004). Forexample, the Canadian local governments have direct access to only property tax, which makesup more than 90% of all local government tax revenue (Kitchen, 2004). In the United States,however, local governments may have direct access to one or all of the income tax, sales tax,and property tax (Kitchen, 2004, 2002). As a result, the Canadian local governments rely moreheavily on the federal and provincial funding sources for infrastructure projects and develop-ment and implementation of specific programs such as adaptation, compared to the Americanlocal governments.The local governments in Canada and the United States have experienced a significant increasein their responsibilities, starting around the 1980s and 1990s (Berkes, 2010; Kousser, 2014).This increase was due to the transfer of the regulatory and financial responsibilities from seniorlevels of governments to local governments, which is called “the downloading (also known asoffloading)” in Canada and “the devolution” in the United States (Hamilton, 2013; Duffy et al.,2014). In Canada, the downloading was carried first through the significant cuts in transferpayments from the federal government to provincial governments. Next, it was carried throughthe transfer of the essential responsibilities, such as flood protection, from provincial govern-ments to local governments without providing sufficient funding or additional revenue sources(UBCM, 2011; Duffy et al., 2014). While resulting in significantly more autonomous provincialand local governments, the downloading created significant concerns over local governments’ability to carry out these responsibilities (UBCM, 2011). In the United States, the devolutionaimed to reduce the size and role of the federal government and was carried through the provi-sion of the block grants with overarching federal goals and guidelines to the states, which was120then distributed to the local governments (Rodrı´guez-Pose and Gill, 2003; Hamilton, 2013). Asa result, the states were allocated more responsibilities in developing and administering state-specific programs to achieve the federal goals (Hamilton, 2013).In Canada, the downloading influenced the environment and the related infrastructure most(Duffy et al., 2014, p.6). In the United States, the devolution mainly impacted the welfare andhealthcare areas, rather than the environment (Rodrı´guez-Pose and Gill, 2003; Kousser, 2014).Consequently, adaptation, and particularly flood management have become the main regulatoryand financial responsibility of the local governments in Canada (UBCM, 2011; Duffy et al.,2014). The decentralization of some key services has essentially increased the local governmentexpenditures in BC and WA (Hamilton, 2013; Duffy et al., 2014). However, the impact has beenstronger in Canada, where the local governments have already had limited access to the revenuetools (UBCM, 2011). In a survey of the 133 municipalities, Duffy et al. (2014) found that 83.6%of the municipalities agreed that the downloading had been a major concern and challenge fortheir local government. The downloading of the senior government responsibilities to the localgovernments, without providing direct (through grants or programs) or indirect (through accessto financial tools) have prevented local governments in BC to bridge the gap between what isneeded to provide services and what is available (Duffy et al., 2014).The decentralization of the senior level responsibilities has fundamentally transformed the au-thority distribution and roles of federal, provincial, state and local governments in Canada andthe United States. Although the initial approach to bring regulatory responsibilities closer tothe people affected is meaningful and breaks the one size fits all thinking (Hamilton, 2013);the lack of financial support and change in the existing financial mechanisms, and inequalityover the ability of different local governments to handle these new responsibilities (Duffy et al.,2014) have led to the new debates on the roles of senior governments (Berkes, 2010). Coastal jurisdiction and ownershipBesides the differences in their governance systems, Canada and the United States also havesignificant differences in their coastal jurisdiction and ownership (Figure 4.9). Jurisdiction hererefers to the official decision-making authority to interpret and apply laws (Blair, 2009), andto regulate and manage issues through permits and programs in a specified spatial boundary.The jurisdictional boundaries, corresponding responsibilities, and conflicts due to ownership121are not well-understood (Bauer et al., 2012). This misunderstanding is because the jurisdictionalboundaries do not always reflect ownership, and ownership does not always translate into havingjurisdiction. Therefore, understanding which aspects of the coastal areas falls under which levelof governments’ jurisdiction, and relates to which regulations (Becklumb, 2013) are significantwhen investigating the barriers to and facilitators of CGI implementation.Figure 4.9: The graphic illustration of the coastal jurisdiction in BC and WA. Distances are not to scale.Canada and BCIn Canada, the federal government and provincial governments have jurisdictions over lands,waters, and submerged waters they own. The local governments have jurisdiction over the landthey own. However, there are several important overlaps (Giest and Howlett, 2013) as seen inFigure 4.9.The federal government’s jurisdiction and ownership extends from the low water mark (LWM)to 12 nautical miles in the Territorial Sea1, to 200 nautical miles in the Exclusive Economic1The Territorial Sea extends from the low water mark up to 12 nautical miles out to sea (Fisheries and Oceans122Zone2 (Fisheries and Oceans Canada, 2011). This jurisdiction can also extend over the coastalriparian areas if the land is a federal Crown land (Blair, 2014). Within its jurisdiction, the fed-eral government regulates environmental issues related to federally owned properties; coasts andfisheries; navigation and shipping; marine pollution and interprovincial water pollution; crimi-nal law; boundary waters; migratory birds; and First Nations and First Nations lands (Becklumb,2013, p.1-2).The First Nations governments are sovereign nations and have jurisdiction to use and manageterrestrial and aquatic lands within their territories. This jurisdiction is subject to the nation tonation treaty negotiations between the First Nations governments and the federal governmentoutside of their territories (Blair, 2014; McLeod et al., 2015). The federal and provincial gov-ernments have to consult with the First Nations governments on issues related to the “wildlifemovement, supply and access; decisions with respect to pollution from construction or use thatmay affect flora or animal populations; change in regulation or policy that may restrict land use;federal life cycle of land management that may affect legal obligations and relationships withAboriginal groups; or decisions with respect to use of natural resources that may limit supplyand use by Aboriginal groups” (Minister of the Department of Aboriginal Affairs and NorthernDevelopment Canada, 2011, p.11).The provincial jurisdiction and ownership at the coast extends throughout the foreshore (alsoknown as the intertidal zone), which is the area between the low water mark3 and the naturalboundary4 (Nature, 2002). This jurisdiction can extend over the coastal riparian areas if theland is a provincial Crown land (Blair, 2014). In BC, the provincial jurisdiction and ownershipexpand further from the low water mark towards inland seabeds such as the Strait of Geor-gia (Blair, 2014), and “submerged lands between major headlands such as bays, estuaries andfjords” (Fisheries and Oceans Canada, 2009, p.2). Within their jurisdiction, provincial govern-ments regulate environmental issues related to the property and civil rights; regulation of mosttypes of business, natural resource industries, and emissions; management of provincial Crownlands; and regulations related to municipal institutions (Becklumb, 2013, p.2). The provincialCanada, 2011; USCOP, 2004a)2The Exclusive Economic Zone (EEZ) refers to the zone between the Territorial Sea to a maximum of 200 nauticalmiles (Fisheries and Oceans Canada, 2011; USCOP, 2004a).3The low water mark is the level reached by water at low tide (Fisheries and Oceans Canada, 2018).4The natural boundary (high water mark) is the level reached by water at high tide (Fisheries and Oceans Canada,2018).123governments do not regulate marine pollution as it is under the federal jurisdiction (Becklumb,2013). In addition, the fish habitat in the coastal areas falls under the federal jurisdiction.Local governments own and have jurisdictions over the land starting from the natural boundaryand extending over their municipal boundaries (Blair, 2014). There are also exceptions if theland is federal and provincial Crown land. Within their jurisdiction, local governments regu-late land use, building permits, development, waste management, and drinking and wastewater(Becklumb, 2013).The United States and WAIn the United States, lands, waters and submerged waters can be privately owned or owned by agovernment institution, yet regulated by another.Traditionally, the state governments had jurisdiction to govern coastal areas, but this responsibil-ity had shifted to the federal government in the 1970s (Paddock, 1990). The federal jurisdictionat the coast extends from the mean lower low water line (MLLWL)5 to 12 nautical miles inthe Territorial Sea, and to 200 nautical miles in the EEZ (USCOP, 2004a). The federal gov-ernment gives authority to the states to manage and regulate the zone known as the ShorelineManagement Area (SMA)6. Within its jurisdiction, the federal government regulates environ-mental issues, which include coasts and fisheries; navigation and shipping; commerce; powergeneration; national defense; and international affairs (USCOP, 2004b).The Indian Tribes are sovereign nations and assert their right to govern their members and lands,and control decision-making within their territories (Arnold, 2004; Kalt and Singer, 2004). Thisarea also includes the intertidal lands if the land tenure is owned by the Tribes (NOAA, 2018a).Within their territories, the Tribal governments have rights to regulate matters such as economicdevelopment; natural resources; land use; religious and spiritual sites; wildlife habitat; and sen-sitive environmental areas, amongst others (Logsdon, 2001). Outside of their territories, theTribal governments have special rights such as fishing, due to nation-to-nation treaty negotia-tions between the Tribal governments and the federal government (Miller, 2001). The Tribesgovernments co-manage the fishery resources with the states governments and the federal gov-ernment (NOAA, 2018a).5The mean lower low water line (MLLWL), refers to the line that represents the elevation of mean lower low water(NOAA, 2018b).6The Shoreline Management Area in WA includes the area between the state waters (3 nautical miles in the territorialsea) and 200 feet inland.124The state governments have jurisdiction within the SMAs (State of Washington Departmentof Ecology, 2009). Although the federal government has jurisdiction over the states waters,the state governments were allocated the authority to govern individuals, property, and enter-prises within the SMA boundaries (USCOP, 2004b) and “manage, develop, and lease resourcesthroughout the water column and on and under the seafloor.” (USCOP, 2004a, p.71).The counties and municipalities have “regulatory, administrative, and taxing authorities” as de-termined by their state governments (Arnold, 2004, p.25). Within their jurisdiction, they areresponsible for local recordkeeping and elections; creating and updating shoreline managementplans, zoning, building codes; construction and maintenance of local and rural roads; and lawenforcement (Arnold, 2004). Contrary to BC, beaches and tidelands can be privately ownedby individuals in WA. The privately owned lands are still subject to local, state and federalregulations. Coastal and environmental regulations and programsVarious coastal and environmental regulations are applied within the federal, provincial andstate jurisdictions described above. The CGI projects in BC and WA may have to comply withvarious regulations and obtain permits from a number of different institutions. The completelist of the relevant CGI regulations, administering institutions, and the required actions can befound in Appendix C.3 and C.4. These regulations influence actions, policies, and programsrelated to CGI; therefore can be barriers to and facilitators of CGI implementation.Regulating the coasts- CanadaFisheries and Oceans Canada (DFO), previously known as the Department of Fisheries andOceans, is delegated as the main federal institution to govern oceans and marine habitats underthe Department of Fisheries and Oceans Act (1985) and Oceans Act (1996). The DFO was alsoallocated jurisdiction to regulate and manage fisheries, the quality of fish-bearing waters andhabitat, and marine plants and marine mammals under the Fisheries Act (1985); as well as thesafety of coastal areas through the Canadian Coast Guard (Blair, 2014; Becklumb, 2013).Transport Canada (TC) is another important federal institution involved in the management ofcoastal areas. Under the Canada Shipping Act (2001), the TC regulates emissions, sewage,125oil discharges, as well as shipping routes and safety (Becklumb, 2013). Under the NavigationProtection Act (1985), the TC regulates in areas where there are navigable waters7 (Blair, 2014)to reduce hazardous conditions to navigation (Sheffield, 2013). Under the Canada Marine Act(1998) TC regulates and manages harbor and shipping facilities that are in the federal Crownlands such as port authorities and major harbors (Becklumb, 2013).Parks Canada regulates and protects marine areas through the Canada National Marine Conser-vation Areas Act (2002), and natural areas of national significance through the Canada NationalParks Act (2000). Environment and Climate Change Canada (ECC) protects endangered orthreatened species and their habitats through the Species at Risk Act (2002). The EEC creates,protects and regulates wildlife areas for research and conservation through Canada Wildlife Act(1985). It also regulates and protects migratory birds and their habitats through the MigratoryBirds Convention Act (1994).Under the Canadian Environmental Protection Act (1999), the ECC and Health Canada collab-oratively assess the environmental and human health impacts of projects, particularly the risksfrom pollution (Becklumb, 2013). The Canadian Environmental Assessment Act, managed bythe Canadian Environmental Assessment Agency, necessitates federal departments to assess theimpacts of the federal projects on the environment and human health (2012).- British ColumbiaIn BC, provincial laws are administered to activities on and related to natural resources, marineresources, and subsurface resources (Fisheries and Oceans Canada, 2009).TheMinistry of Environment and Climate Change Strategy (MoECCS) is the main provincial in-stitution responsible for the “protection, management and conservation of” provincial resourcessuch as land, water, air, and living resources under the Environmental Management Act, EMA(Ministry of Environment, 2016, pg.1). The EMA regulates environmental issues related to pol-lution, hazardous waste, waste discharges, and contaminated sites; and air quality (Sheffield,2013; Ministry of Environment, 2016). The MoECCS also regulates all actions “constructed,assembled or installed to prevent the flooding of land” under the Drainage, Ditch and Dike Act(1996) or primarily by appointing an Inspector of Dikes under the Dike Maintenance Act (1996)(Fisheries and Oceans Canada, 2009). This responsibility includes approval of the construction,design, and changes to dikes; determining standards and design criteria for flood protection in-7Navigable waters refer to the waters that can be passed even with a vessel126frastructures; working with the local diking authorities to monitor the management and assessthe safety of flood protection infrastructure (The Arlington Group, 2010). The MoECCS alsomanages and administers all matters concerning parks, conservancies and recreation areas underthe Park Act (1996) (BC Parks, 2018); designates Crown land for conservation under the En-vironment and Land Use Act (1996); and designates parks, recreation areas and conservanciesunder the Protected Areas of British Columbia Act (Fisheries and Oceans Canada, 2009).TheMinistry of Forests, Lands, Natural Resource Operations and Rural Development (FLNRRD)regulates seabed lands such as Strait of Georgia, and provincial Crown lands through the LandTenure Branch under the Land Act (1996) (Becklumb, 2013). This responsibility includes al-location of land for industrial, private dock and commercial marina uses; permits; and landtenures (licenses or leases) to the federal government, First Nations, and local governments(Blair, 2014). The FLNRRD also protects riparian areas by requiring a Qualified Environmen-tal Professional to assess proposed residential, commercial, and industrial development beforelocal government approval under the Riparian Areas Protection Act (2016), formerly Fish Pro-tection Act (The Ministry of Forests, Lands, Natural Resource Operations and Rural Develop-ment, 2016). In addition, the FLNRRD designates Wildlife Management Areas to manage andconserve fish, wildlife and their habitats under the Wildlife Act (1996) (Fisheries and OceansCanada, 2009).The BC Environmental Assessment Office regulates the Environmental Assessment Act (Envi-ronmental Assessment Office, 2018), except for the projects with implications on matters underthe federal jurisdiction (Becklumb, 2013). The Land Title and Survey Authority of BritishColumbia manages “registration of land titles and the subdivision of lands.” under the LandTitle Act (Fisheries and Oceans Canada, 2009, p.13).Under the Local Government Act (1996) and the Community Charter Act (2003) the local gov-ernments in BC govern and regulate, coastal riparian lands using bylaws, ordinances, zoningregulations, building permits, and specific plans such as Liquid Waste Management Plans.- The United StatesThe National Oceanic and Atmospheric Administration (NOAA), under the Department ofCommerce, was delegated as the main federal institution to manage coastal areas (Day and Gam-ble, 1990). Arguably, the most important regulation related to the management of the coastalareas in the United States is the Coastal Zone Management (CZM) Act (1972). This is because127through CZM, NOAA delegates state governments’ responsibilities including the protection ofnatural resources (estuaries, beaches, dunes, barrier islands, and wetlands) and minimizing theloss of life and property from hazards, and evaluates state Coastal Zone Management programs’performance to approve or withhold federal funding and approval (USCOP, 2004b).Under the Magnuson-Stevens Fishery Conservation and Management Act (1976), NOAA alsoregulates fishing activities in federal waters. The U.S. Army Corps of Engineers (USACE)regulates the nation-wide permit and development in the state waters (USCOP, 2004b) to preventobstructions to navigation, under the Rivers and Harbors Act (1899).NOAA and the U.S. Fish and Wildlife Service (FWS) regulate “the conservation of threatenedand endangered plants and animals” and the habitats, under the Endangered Species Act (1973)and the Fish and Wildlife Act (1956). They also regulate the marine mammal species andpopulation stocks under the Marine Mammal Protection Act (1972).The Environmental Protection Agency (EPA) under the Clean Air Act (1970) regulates air emis-sions and protects public health (USCOP, 2004b). Under the Clean Water Act (1972), the EPAregulates activities affecting water quality, and pollutant discharges (USCOP, 2004b). EPA ad-ministers the National Environmental Policy Act (1969), which requires the assessment of theimpacts of federal projects and decisions on the environment by the corresponding federal agen-cies. Under the Beaches Environmental Assessment and Coastal Health (BEACH) Act (2000),the EPA manages program developments for states, territories, and tribes to increase their waterquality standards for public use. The EPA also develops ocean dumping criteria and evalu-ates permit applications under the Marine Protection, Research, and Sanctuaries Act (1988).EPA and the Coast Guard, “regulates the transportation of municipal and commercial wastesin coastal waters”, under the Shore Protection Act (1988). Lastly, under the EPA, the Officeof Pollution Prevention and Toxics manages programs aimed to reduce pollution and increaseresource efficiency under the Pollution Prevention Act (1990).- Washington StateWA has jurisdiction to regulate coastal issues in the state waters, but also has to comply with theabovementioned federal regulations (USCOP, 2004b).The Department of Natural Resources manages the natural and living resources and human-made structures in and above of the state-owned aquatic lands under Title 332 of the StateLegislature (1889). The Department of Ecology (DoE) regulates all private, local government,128and state government actions on lands privately owned or owned by the local and state gov-ernments(State of Washington Department of Ecology, 2009) under the Shoreline ManagementAct (1971). The DoE also has regulatory control over the formation and co-management ofthe flood control districts (State of Washington Department of Ecology, 2004), under the FloodControl Act (1935) and the Diking and Drainage Act (1985. The DoE implements these acts inpartnership with the local governments (Adelsman et al., 2012). In addition, under the ClimateLeadership Act, the DoE oversees the integrated climate response strategy and the use of thisstrategy in the planning and designing policies and programs (Ziff, 2017)The Department of Fish andWildlife (WDFW) conducts a review process of permit applications,plans and policies to assess their environmental impacts under the State Environmental PolicyAct, (1971). Under the State Hydraulic Code, the WDFW also regulates construction projectsor activities such as works on bulkheads, culverts, bridges, and sediment dredging projects in ornear state waters through the Hydraulic Project Approval. The Department of Archaeology andHistoric Preservation regulates all alterations to an archaeological site under the ArchaeologicalResources Protection Act (1979).The local governments in WA govern coastal riparian lands using by-laws, ordinances, zoningregulations, building permits under the Title 35 Cities and Towns (1969), and Title 36 Counties(1969). Under the state’s ShorelineManagement Act, local governments regulate and administer“the substantial development permits, conditional use permits, and variance permits”.Coastal and environmental programsCoastal and environmental programs (i.e., grants, projects, or services) that are provided bydifferent levels of governments to assist provincial, state and local governments, organizations,and individuals in matters related to coasts and environment. These programs provide incentivesfor actions such as habitat restoration, adaptation, and land conservation, amongst others. Forexample, the National Wetland Conservation Fund operated by the Environment and ClimateChange Canada provides funding to projects that conserve and protect wetlands (Environmentand Climate Change Canada , 2018). Similarly, the Estuary and Salmon Restoration programoperated by NOAA provides funding and guidance to protect estuary habitats (National Oceanicand Atmospheric Administration, 2018).Coastal and environmental programs in the study area were searched in the official governmentand NGO websites. The review identifies diverse federal, provincial and state level programs129that are available for CGI implementation. This review shows that there have been significantlymore programs in WA and the United States. The programs in the United States have startedaround three decades earlier than in Canada (Figure 4.10). In addition, the programs in WA havealso started much earlier than the provincial programs in BC. Particularly since 2010, there hasbeen a significant increase in the number of Canadian federal and BC provincial programs.Figure 4.10: The timeframe of the coastal & environmental programs in Canada, BC, the USA, and WABesides the number of programs, a significant portion of the programs in the United States andWA have been continuous, where most of the programs in Canada and BC have been initiatedfor a limited period of time (Figure 4.11).Unfortunately, the data on federal, provincial, and state grants and funding are not comparablebecause the information was not reported in a uniformed way across the study area. The fundinginformation was sometimes available in the maximum amount available per project format, andother time in the total project funding per year format. However, the availability of the numberof grants and funding in WA has been higher than in BC. The list of programs, lead institutions,time frames, and total or max. per project grant information can be found in Appendix C.5,C.6,C.7, and C.8.130N=32 N=45WA & USABC & CanadaFigure 4.11: The continuity of the coastal and environmental programs available in WA and BC4.4.2.4 Summary of the review and synthesis of the institutional arrangements resultsThe results of the review and synthesis of the institutional arrangements indicate that thereare no clear authority distinctions in BC and WA. The BC provincial government is financiallymore independent than the WA government. The decentralization of the regulatory and financialresponsibilities has altered the federal, provincial, state and local governments’ roles in Canadaand the United States. As a result, the local governments in BC has significantly less access tofinancial and other resources than the local governments in WA.The results also show that the coastal jurisdiction in BC has multiple important overlaps. Theseoverlaps are between the local, provincial, and federal jurisdictions. Overall, the provincialjurisdiction in BC is limited to a small strip at the coast. The BC provincial jurisdiction widenswhen there are inland seabeds such as Strait of Georgia. In WA, the state government wasallocated the jurisdiction of the SMAs, which is the area between 200 feet inland to 3 nauticalmiles in the Territorial Sea.A long list of coastal and environmental regulations are identified in this review. The resultssuggest that the CGI projects in BC and WA may have to deal with the similar number of131regulations. In general, the contents of these regulations are similar across BC and WA. Theydeal with similar issues related to the air and water pollution, the safety of the navigable waters,marine and land habitats, fish populations, habitat, and passages, and other relevant issues. Themost significant difference between BC and WA in coastal and environmental regulations is theCZM Act in the United States, which allocates states authority to regulate a large portion ofthe coastal zone that includes landward and seaward sides of the WA’s coastline. In BC, theprovincial government’s role is limited to a narrow strip in the intertidal zone.Lastly, there are numerous coastal and environmental programs in BC andWA. The results showthat in the past 50 years there have been significantly more federal and state programs availablefor the CGI projects in WA than in BC. These programs have been predominantly continuous,and have been operating for longer than the programs in BC.4.4.3 Semi-structured interviewsA total of 10 semi-structured interviews, five in each region, were conducted during April andMay 2018. In each region, two participants were from the local governments, two participantswere from the local NGOs, and one participant was from the provincial or state government.Table 4.1 shows the institutions of the interview participants.British Columbia Washington StateNon-governmentalorganizations1. Peninsula Streams Society2. Fraser River Estuary Manage-ment Program1. Northwest Straits Foundation2. Puget Sound PartnershipLocal governments 1. City of Surrey2. Town of Qualicum Beach1. Kitsap County2. Skagit CountyProvincial-Stategovernments1. Ministry of Environment andClimate Change1. Department of Recreation andConservation OfficeTable 4.1: The institutions of the semi-structure interview participantsThe interviews were conducted on the phone and in person. Interviews lasted between 40 min-utes and 70 minutes. The interviews were audio recorded for analysis purposes. Each interview132followed the protocol described in the Section 4.3.2 and can be found in Appendix C.2. Afterthe interviews were completed, the verbatim transcriptions of the audio files were created. Thetranscripts were first coded to large themes such as “Barriers”, second to sub-themes such as“Financial” and last to specific themes such as “Funding amount” using the NVIVO software.The findings of the semi-structured interviews were organized into seven main themes to providea structured way to explain the results and to compare BC and WA. These themes are not meantto be mutually exclusive. The barriers and facilitators under each theme can be relevant to someof the other themes. The seven themes are as follows:• Financial,• Jurisdiction and ownership,• Regulations,• Capacity and resources,• Vision and leadership,• Collaborations, and• Community and knowledge.The analysis of the semi-structured interviews has identified a number of institutional barriersto and facilitators of CGI implementation. The barriers that were most common both in BCand WA were financial, the funding amount and continuity, and timeline limitations of fundingsources; and regulations, permitting processes and the wait time to obtain approvals. In BC,the regulations, financial, capacity and resources, and jurisdictional barriers were mentionedby most of the participants. In WA, the financial and regulations barriers were mentioned bymost of the participants. In both BC and WA, the vision and leadership, financial, capacity andresources, collaborations, and community and knowledge facilitators were mentioned by mostof the participants.Each of the barrier and facilitator themes is described below. Each section also includes agraphic showing the different concepts of each theme. The scale bar from 0-5 indicates howmany respondents mentioned the concept as a barrier and/or a facilitator. The pink and bluereflects the respondents from BC and WA, respectively.133FinancialAs mentioned above, financial issues were recognized to be one of the most common barriersin BC and WA (Figure 4.12). Particularly, the funding amount and continuity, and the timelinelimitations of funding sources were frequently mentioned by all of the participants.Figure 4.12: The financial barriers and facilitators in BC and WAThe participants in BC suggested that the downloading of the senior level government respon-sibilities on the local governments, without providing means to finance them, have created evenlarger needs for funding. The participants acknowledged that the funding amount in WA hasbeen significantly higher and has been available for a longer time than in BC. However, theystill described a continuous need for more funding to achieve local, regional and national goals.One participant from a WA state department explained the funding need this way:“In the beginning projects tended to be somewhat opportunistic. They were kind ofrelatively easy to address and implement in a very short amount of time and didn’tcost an incredibly large amount of money. And what we are finding now that theprojects are more complex, they involve more partners, they are being implementedin phases, and the funding may be available only for certain phases.”The participants described that even they receive funding for a project, it is typically for a limitedtwo-three year period, which is not long enough to cover all the project steps such as planning,design, consultation, permit applications, waiting for approvals, waiting for fish windows, con-struction and monitoring. They indicated that by the time they are ready for implementation,their funding cycle expires, creating a significant barrier for the CGI implementation. One par-134ticipant from an NGO in BC described the time limitations of fundings as follows:“[The funding agency says] you have got 12 months to do it and you go. But ourwork window is between July 15th to September 15th. They do not understand thatin Ottawa, right.”The participants also described how the funding is used and on what, as a barrier. Notably, theavailability of the funding only for the site assessment, planning, and design, or constructionphases of the projects life cycle has frequently been reported. In addition, the lack of fundingfor specific actions such as land acquisition to create and rehabilitate CGI was noted as a barrier.The participants also acknowledged funding as a key facilitator. From small seed funds to helpwith project applications or initial assessment of habitats, to large funds for complex and multi-objective projects, the funding amount and types create new opportunities for communities ofall sizes and capacities to implement CGI projects. One participant from a WA NGO describedhow they use the small seed funds this way:“We are working right now to identify a stable funding source to provide loansto people to remove their shoreline armoring (...) a very successful program hasbeen giving workshops and giving people a five thousand dollar grant to look intoremoving their shoreline armoring and to get the feasibility done on those. A highpercentage of those small loans have resulted in the removal of armors.”The variation in funding sources, particularly in politically challenging times, reduces depen-dencies on one funding source and ensures that there are continuous funding sources for the CGIprojects. The cost-saving opportunities, such as bringing landowners together for one projectapplication is also reported as a facilitator of CGI implementation. Lastly, the availability of thespecific funding sources for land acquisition was defined as a driver of the CGI projects.Jurisdiction and ownershipJurisdiction and ownership is another major theme that has been described as a barrier and facili-tator (Figure 4.13). The common concepts identified during the interviews include jurisdictionalboundaries, organizational mandate/authority, organizational structure and involved parties, andprivate ownership of coastal areas.135Figure 4.13: The jurisdiction and ownership barriers and facilitators in BC and WAThe participants in BC stated that the coastal jurisdiction is not always apparent, and often cited“the layers of overlapping jurisdictions” on the coastline. They also described the complexitiesthat result from the way the jurisdictional boundaries are structured. As described in the previoussection, the jurisdiction of coastal British Columbia has a number of overlaps below and abovethe natural boundary, which cause ambiguities regarding identifying institutions with decisionmaking authority. They stated that operating within this jurisdictional context, and finding theright government departments or other involved parties pose significant challenges for the CGIprojects. A participant from a local government in BC has explained the jurisdictional issue theyexperience this way:“The jurisdictional overlap on the foreshore, I would say is one of the biggesthurdles to effective decision making because our jurisdiction overlaps with theprovince’s ownership of the land on the foreshore. (...) [The province] owns theCrown land but they do not always have the ability to bring things into too muchdetail as local governments do.”On the other hand, the participants in WA described challenges related to the private owner-ship of the coastal areas, particularly the tidelands, rather than the jurisdictional issues. Theydescribed that even though there are many jurisdictions involved in the coastal areas, the juris-dictional boundaries, and the roles of different institutions are relatively clear. However, theydescribed the private ownership of the coastal areas as a significant challenge for CGI imple-mentation. The participants cited that the private ownership of coastal areas in WA has resultedin the fragmented and uncontrolled development of different coastal protection measures. Aparticipant from a WA state government described it this way:136“In Washington State, almost 70 percent of the tidelands are not in governmentownership and not in public ownership. So the land that private landowners mayown are tidelands which has led to all the development going on along the shorelinekind of uncontrolled for a long time.”The participants from both sides, but mainly from BC, described the clarification of jurisdic-tional boundaries as a facilitator of the CGI implementation. One participant from an NGO inBC stated that “it is important to have those jurisdictions very clearly outlined and have all theplayers at the table when you are trying to make progress”.Most of the participants in BC suggested that the limitations in their decision-making authorityor lack of mandate of their organization were significant barriers to the CGI projects. Theyoften cited the bureaucracy they have to go through and the challenges related to “how differentauthorities might react to some of the bold approaches required” for adaptation (Participant froma local government in BC). Some of the participants recognized the decision-making authorityand a structured organizational mandate as facilitators of the CGI projects. For example, oneparticipant from an NGO in BC suggested that “ governments need to be stronger in terms ofpreventing walls from being built unless there is absolutely no alternative”.RegulationsThe participants in BC and WA described frustration over the regulations and regulatory pro-cesses related to CGI projects (Figure 4.14). Particularly issues related to the number of permitapplications they have to go through, the review processes, and the time they have to wait toobtain approvals.Figure 4.14: The regulations barriers and facilitators in BC and WA137Some of the participants in BC stated that they are not entirely sure what specific rules and per-mits they would need in certain areas. Most of the participants in BC perceived the challengeswith the permitting processes as direct results of the jurisdictional ambiguity and the disconnec-tion between departments and government levels. They described the existing regulations in BCthat permits are set up to favor traditional hard solutions over CGI. One participant from a localgovernment in BC stated:“The governance structure is actually set up to default to the worst solution [hardsolutions]. And that is one of our biggest challenges right now. The approvalsprocess, resulted from that jurisdictional gray area, (...) is cumbersome and doingit the right way is harder than doing it the wrong way.”This participant explained that it is easier to build a seawall within the municipal boundary,rather than to go through the lengthy process of permit applications with the provincial andfederal governments within the current regulatory environment.One participant from a local government in BC explained the permitting barriers as contra-dictions between different regulations, and disconnections between different departments andgovernment levels:“A lot of times we find ourselves in the local government being sort of caughtbetween two contradictory regulations. One says do X the other says Z, and theyare not mutually exclusive. Then it makes it very difficult to control which oneis before or how to work through that difference. (...) The contradictions oftencome within the province. Different departments or ministries I should say, andthen between the different levels of government. There’s very little coordinationbetween the two. To the extent where it might take eight months to hear back fromone group at the province for an application and from the federal government youmay find out sooner, but the whole point is that it has been completely siloed.”Although the permitting processes were described as a significant barrier in WA, the participantsalso recognized the value of these regulations. They stated that the permits ensure the coastalareas are protected. The participants in WA viewed stated that the permitting processes as ”bighurdles”, yet they stated that they are used to them, and know what to do and whom to call. Themain issue they experience was related to the burden of the permitting processes on individual138homeowners, as the processes can be costly and confusing. A participant from an NGO in WAexplained it this way:“We go to the full permitting process like any other project be required to do, andI would not say that I am bothered by any single regulation. I think I agree withthe majority of the regulations that are out there. The challenge that we have and Ithink probably other organizations have, is really the amount of time that it takes togo through the permitting process here. (...) If there were fewer permits, then youwould get more private property landowners that would be willing to engage in the[CGI] process.”On the other hand, the participants suggested that a streamlined permitting process would bea significant driver of the CGI projects. Both in Bc and WA, the participants agreed that thestreamlined processes would reduce confusion, complexities and the wait time to receive ap-provals. A participant from a local government in WA described the way they try to reducebarriers related to permitting as follows:“For the permitting barrier, we put we put together permitting packages and then wealso offer what we call “a restoration site consultations” where we kind of coordi-nate with local permit reviewers. So we will go up to the site with the homeowner,the engineers, or other involved parties. We talk about the site and the homeowners’objectives and then determine what permits would be needed in the case, what stud-ies and plans that needs to be part of the permitting process. That really streamlinedthe process in the way that everybody involved knew from the very beginning whatwas required and what was not required so that they could save time and money ingetting the permits through the process.”The wait time for the permit applications was characterized as a significant regulations barrierfor the CGI projects. The participants from both side often cited “ the workload” and “thelengthy review processes” as challenges. One participant from an NGO and then potentiallylose your funding if your grant is only for a specific amount of time. So the permit process canbe cumbersome.”The lack of proactive climate change and adaptation training for professional and building stan-dards, such as engineering or planning certificates were also characterized as barriers in BC.Particularly one participant from the provincial government in BC described the professional139reliance model used in BC and its short-comings this way:“It runs on the assumption that engineers will be prompted by their [engineering]association to acquire appropriate training to be capable of advising on decisionsrelated to future climate. (...) But in reality, we are not there yet. We have manycompetent engineers who have all the necessary qualifications that engineers needto have. But they do not really understand yet how they should consider futureclimate information”.One participant from a local government in BC described these standards as an important poten-tial driver of CGI projects. Lastly, the land use demands and regulations were considered to beboth a barrier and a facilitator in BC. For example, balancing between the type of coastal accessindustries need and creating habitat was reported as a barrier.Capacity and resourcesMain barriers of the capacity and resources theme are senior level support, precedent, and staffexpertise; while the main facilitators are senior level support, and processes regarding meetings,reviews and decision making (Figure 4.16).Figure 4.15: The capacity and resources barriers and facilitators in BC and WAThe participants in WA did not indicate significant capacity and resources related barriers, ex-cept for challenges related to ineffective meetings, project reviews and decision-making pro-cesses. They noted that these barriers influence the time it takes to get things done. Providingstaff with training in facilitation and hosting efficient meetings were seen as facilitators that help140overcome these barriers. The frequent staff change in key organizations and positions were alsoidentified as a barrier. The participants described that where it was time-consuming to find theright person and build functioning relationships. These barriers were also reported in BC.The constant struggle to show the value of partnerships and effectiveness of project implemen-tation were described to be a barrier in BC, and a facilitator in both regions. A participant froman NGO in BC described the issue as follows:“You continually had to show, check in, and make sure you are showing value.Because the minute [partners] felt like they were not getting value, they would startto question their [monetary] contribution and not send staff to the meeting.”The lack of the necessary staff expertise was characterized as an important barrier in BC andthe presence of it as an important facilitator. For example, the continuous empowerment andeducation of local planners and engineers were reported as important facilitators.The lack of precedents such as case study examples and demonstration projects were cited asan important barrier to and facilitator of developing new CGI projects. A participant from alocal government in BC stated “I do not think there are many examples or sort of a procedurefor how to work through those issues, and it is a big problem”. The presence of these examples,and learning from them were identified as an important facilitator. A participant from a localgovernment in WA explained it this way:“We did a demonstration site on a public county park for people to see and kindof know that (...) [the shoreline] is supposed to be natural this way. So I thinkhaving these demonstration sites and having actual homeowners carry out projectsand having their neighbors see these projects really be successful in building a lotof momentum.”The most important barrier and facilitator in this theme was the senior level technical, regulatory,and data support. In BC, the lack of the senior level government support was regularly cited bythe participants. One participant from a local government in BC expressed their frustration withthe lack of senior-level support this way:“When you think about where innovation is coming from, it is coming from localgovernments. It is coming from NGOs. It is not coming from the province. We aredragging them along with us.”141However, besides being a significant barrier, the presence of the senior level support was alsoconsidered to be an important facilitator. In WA, the participants frequently voiced the impor-tance of the senior level support they have received over the years. One participant from a localgovernment in WA explained:“I think the amount of support that we have gotten from our local elected officialfrom the first day and the federal government in providing monetary resources hasreally helped this program and move the ball. [The program] wouldn’t exist other-wise.”Vision and leadershipThe leadership, political will, vision and planning, and projects with multiple objectives wereidentified as the components of the vision and leadership theme (Figure 4.16).Figure 4.16: The vision and leadership barriers and facilitators in CGIThe participants in BC characterized the lack of leadership as a significant barrier. Even in thelocal government level, the absence of a “keener” has been described as an important barrier toCGI implementation. The participants often cited “navigating the field” alone without guidance,due to the lack of leadership from the different levels of governments. A participant from anNGO In BC described the issue this way:“Here, there is not really any leadership. Leadership is not just throwing this vol-untary stewardship program like they have. They have to say “OK this is the GreenShores program, we are going to educate people and then we are going to start reg-ulating it.” (...) The leadership has all the resources, and they do not want to getin a situation where they are pushing municipalities around and telling them what142to do. Well yes, they should be told what to do because the shoreline of the oceandoes not belong to the property and it certainly does not belong to the municipality,it belongs to all of us.”On the other hand, having the right leadership has also been seen as a significant facilitator. “Theout of the box thinkers” and “innovators” in local, provincial, state and federal governments havefrequently been noted by participants from both sides.Political will, similar to leadership, has been characterized both as a barrier and a facilitator. Theparticipants described the lack of political will as the result of the political cycle dependencies,and he concerns over the electorates’ perceptions. The participants indicated that particularlyin BC, the local governments are susceptible to what their electorate is concerned about. Theysuggested that the electorate is mostly concerned about the government spending, provision ofservices, real estate values, rather than the future changes in the climate. One participant from aprovincial government in BC explained it this way:“When local governments feel that their electorate does not see the justification forthem trying to implement CGI. I think that is a significant barrier.”When discussing the differences between the political will in BC and WA, one participant froman NGO in BC described the difference as follows:“There is something in the [WA] political process that seems to work better thanours. The issue is political; not so much the people are different. The people arevery similar to the U.S., and the goals on the environment are the same. But when itcomes down to getting things done on a political basis, they seem to have the abilityto advocate and get stuff done where we do not.”Developing and adopting an institutional vision and plan were described as significant facil-itators. In partnerships, having a shared vision, implementing practices that are developedusing this vision, and sharing information across the partnership were noted to be successfulapproaches in WA. One participant from an NGO in WA described their vision, the collectiveimpact model, as follows:“A model we use a lot is the collective impact model. The idea is that you havea backbone an organization that holds the shared vision, and together we can getourselves organized with bringing in different groups. Collectively we will make a143bigger impact than if we all just keep doing our piecemeal.”Lastly, developing project proposals with multiple objectives and providing various ecosystemservice benefits has been noted as a significant selling point of the CGI projects. In general,most of the participants indicated that the general understanding and knowledge on CGI havebeen increasing over time. Participants described that in most cases, the primary focus of CGIprojects has still been on fish habitat rehabilitation or creation. However, there has been ashift from an opportunistic and single focus projects to complex and multi-objective projects,where the CGI benefits are highlighted as significant additional benefits. One participant from astate department in WA suggested that complex and competitive proposals with more ecologicalservices in one project have a better chance of being funded. The participant suggested that inthe cases where the primary objective is the restoration of fish habitat, including other objectiveshelp diffuse the CGI projects.CollaborationsCollaborations were mainly considered as important facilitators of CGI implementation in BCand WA. The concepts identified during the interviews include multiple partnerships, institu-tional relationships, NGOs, and private owner and community buy-in (Figure 4.17).Figure 4.17: The collaborations barriers and facilitators in BC and WAEnsuring respectful and fruitful relationships with all the parties involved were noted as facili-tators. The participants from both sides described the diverse partnerships they have developedover the years involving various federal, provincial, state, and local departments; academic andscientific institutions; NGOs; politicians; businesses; consultant; land developers; private wa-terfront owners; and community residents and volunteers. Although the participants recognized144the value of bringing all involved parties together and working collaboratively towards the sameobjective, they also identified working with multiple partners as a barrier. One participant froman NGO in WA described the challenge they face in this way:“The challenges are geographic. It is really hard to get people in the same roomwhen they live everywhere. Getting to drive across the Puget Sound, getting themall in the same room, and asking them to volunteer their time (...) It is very chal-lenging.”.Besides the geographic challenges, the participants also frequently noted that often partnersmight have different agendas or interests. Therefore, they described making everybody happyand meeting everyone’s expectations as difficult tasks. Moreover, they explained that the work-load of managing the partnerships and working with different governments was very time-consuming.The most significant facilitator in the collaborations theme was the NGOs. The participants fre-quently reported the important roles NGOs play in the CGI projects. They described that NGOsoften applies to the funding sources that are not typically available to governmental organiza-tions, or they build partnerships with the private owners, they are more flexible as they are notbounded by governments’ rules. A participant from an NGO in BC described their roles andopportunities provide as follows:“The NGOs can do can do things that government agencies cannot do. They buildbridges to the property owners that the governments cannot do. They can be moreflexible in terms of their interactions with the other partners that the governmentscannot do. For example, let’s say we want to do a stream restoration work onsomebody’s property. If we go and say “look we would like to do a little streamrestoration on your property. What do you think?” and they ask “how much will itcost?”. We say “it will not cost you anything because we already got funding for it”.They will say “Well that sounds great!”. Where if I was a government agency andI came to you and say “ We have got a mandate to fix the creek on your property”,they will ask you “Well, what do I get out of it?” (...) We found that people will notallow the municipality on their property. Never mind do any work on the propertywithout getting paid for. They want something back from them. (...) We do nothave that problem because when we go, they know that we cannot give them what145they want (a tax break or a better driveway). So the NGOs have a huge opportunityto bridge the gap between levels of governments and property owners and also thesenior level of governments and local and local governments as well. So that isextremely important.”Lastly, the participants from both sides reported that for “a project to proceed, it is important tohave buy-in from the landowners” and identify “the willing landowners”. Therefore, the privateowner and community buy-in are described as essential parts of facilitating the CGI projects.Community and knowledgeAdvocacy, knowledge on natural processes, CGI’s functionality, and costs associated with CGIimplementation are considered as important barriers to and facilitators of CGI implementationin BC and WA (Figure 4.18).Figure 4.18: The community and knowledge barriers and facilitators in BC and WAThe advocacy of the local and environmental community was characterized as a barrier to anda facilitator of the CGI projects. The lack of advocacy for the CGI projects can influence theattention CGI receives, and reduce the resources being put towards. At the same time, thepresence of coordinated advocacy can lead to the creation of new partnerships, and allocationof resources that can drive CGI implementation. A participant from an NGO described the roleof the advocates as follows:“When we go back to when [the organization] was created, the drivers for it beingcreated came a lot from the community.”The lack of or insufficient knowledge of natural processes, associated risks, the effectivenessof CGI and the perceptions of the implementation costs of CGI have been identified as key146barriers. The presence of such knowledge, and science and evidence-based understanding ofCGI were described as effective facilitators. The fears of flooding and the drive to protectpeople and assets have frequently been noted as significant drivers of adaptation. Increasing theunderstanding and the knowledge on the natural processes, the implications of climate change,and local risks associated with these events, therefore, has also been reported to be important.One participant from the provincial government in BC explained the role of the knowledge andpublic perception this way:“I think that is something that people do not necessarily understand. Sometimespeople get really scared, and they end up reacting more on an emotional basis ratherthan actually having considered some evidence. They do not know what the evi-dence is. They do not know whom to talk to about and where they can get the kindsof figures. However, basically, there are things that have a lot to do with how peopleperceive [the implementation cost of CGI].”Effective communication strategies, such as workshops for specific groups or topics; and pro-viding easy access to resources have been reported as an essential part of increasing communityknowledge and changing misperceptions. This includes developing different strategies, tools,and methods of communications and knowledge sharing. One participant from a state depart-ment in WA described the effect of knowledge sharing this way:“We need to be reaching out to other groups that maybe haven’t been as supportiveor even as knowledgeable of the issues as some people that are involved in. (...) Itis information sharing, and it is engaging with folks and answering questions notin a paternalistic way or condescending way. Also working with kids and doingsomething that kids get excited about and (...) they can help bring that back home.” Summary of the semi-structured interview resultsBesides the funding amount, the significant finding of the interviews included the roles of thefunding continuity, funding timeline, and the timeline limitations of the funding sources asbarriers. More flexible funding structures and diverse funding sources were suggested as facil-itators. The results highlighted that the jurisdictional issues were mainly experienced in BC.The participants in BC described their frustrations and the challenges they experience related147to the overlapping jurisdictional boundaries. In WA, the main issue was related to the privateownership of the tidelands, rather than the coastal jurisdiction.The participants all agreed that the regulatory processes are work extensive and time-consuming.A significant finding of the interviews was the role of streamlined permitting processes can playin reducing some of the regulations barriers. The development of the regulations, and landscape,building and infrastructure designs that incorporate climate change and adaptation measures inthe professional standards was another important finding. The capacity and resources theme wasa major barrier in BC, rather than in WA. Notably, the lack of senior government support, casestudy examples, and staff expertise were frequently reported in BC. The capacity and resourceswere also seen as important facilitators of the CGI projects in BC and WA.The participants in BC reported that they have suffered from the lack of leadership and politicalwill. In WA, these were not reported as significant barriers. A significant finding was the roleof having a formal organization vision as a facilitator. Participants in BC reported that the lackof this vision was an important barrier. The results show that the presence of the leadership,political will, vision, and developing projects with multiple objectives were facilitators in BCand WA. Except for the barriers related to the coordination and management of the partnerships,the participants identified the collaborations theme as another important facilitator of the CGIprojects. An important finding here was the role of the NGOs in bridging people and institutions,and in accessing to the resources that are typically not available to governments.Lastly, the main commonly reported barrier under the community and knowledge theme wasthe lack of knowledge and misperceptions on issues related to natural processes, risks, the ef-fectiveness of CGI, and the implementation cost of CGI. In BC, the lack of local advocacy wasalso a significant barrier. All the factors related to the community and knowledge theme werealso recognized as important facilitators.4.5 DiscussionThe literature on the barriers to and facilitators of adaptation recognizes the institutional ar-rangements as significant contributors. Several common and important barriers and facilitatorswere identified in the literature, focusing predominantly on adaptation as a general concept.Recently, studies have started to emerge, investigating the adaptation challenges and drivers of148urban green infrastructure. These studies cited that the implementation of urban green infras-tructure suffers from similar issues as the adaptation barriers, but identified several urban greeninfrastructure specific challenges. However, there remained a gap on the specific institutionalchallenges and drives of CGI implementation. This study provides the first attempt to identifythe CGI specific institutional barriers and facilitators through a comparative study of the BC andWA institutional arrangements.In the first step of this research, a review of the 235 CGI projects was carried out to understandthe level of CGI implementation in BC and WA. This review showed that over the last ten yearsthere were about four times more CGI projects implemented (or are in the process of imple-mentation) in WA, compared to BC. A reason for this disparity may be due to the populationand geographical differences between BC and WA. For example, WA’s coasts have been moredensely occupied and modified than the BC coasts, yet the length of the WA coastline is aboutfive times shorter than the BC coastline. In addition, most parts of the BC coasts are not easilyaccessible, protecting coasts from human-made modifications, thus reducing habitat degrada-tion. However, looking at the details of the projects yielded differences in the CGI projectobjectives, levels and types of institutions allocating funding, lead institutions, funding amountand project size in BC and WA.The findings of the project review indicated that the geographical and population differenceswere not the only causes of the disparity between the CGI project implementation in BC andWA. Various institutional factors were influencing the mainstreaming of CGI projects, some ofwhich were already frequently reported in the literature. These factors included the governancestructure, and corresponding decision-making authority; jurisdictional issues; regulations; se-nior government support; and financial and other resources.Therefore, the second step of this research included an extensive review and synthesis of theinstitutional arrangements to understand how these factors influence the CGI projects in BCand WA. The results indicated that there is more continuity in the federal goals, states’ accessto financial, regulatory and other resources, and allocation of these resources within the states’coastal zone. The Coastal Zone Management Act plays a significant role in the functioning ofthis chain. The management of the coastal zone as a whole empowered the WA governmentto oversee and work to ensure that the local needs are met, and regional and national goals areachieved. In addition, there have been numerous national programs and state programs available149for the adaptation and CGI projects over the last 50 years. In BC, the provincial government isfinancially more independent than the WA state government. This independence has resulted inthe expectation that provinces should start their own programs to address adaptation and CGI.This independence also means that the provincial government is less restricted to follow federalobjectives. Even though the provincial government is relatively independent, its jurisdiction isvery limited in the coastal zone. This limited jurisdiction and the absence of a formal provin-cial coastal zone management practice are preventing any holistic adaptation action from beingimplemented in BC.In the third and last step, semi-structured interviews were conducted to get a deeper under-standing of the barriers and facilitators reviewed in the CGI project reviews and the review andsynthesis of the institutional arrangements steps. It also aimed to gain new insights into otherless common but important factors and CGI specific barriers and facilitators. The results of theinterviews confirm the barriers and facilitators identified in the previous steps and contributenew and important nuances to each. The results also show that there are new and specific barri-ers and facilitators influencing CGI implementation in BC and WA such as coastal jurisdictionand ownership; financial variation and flexibility; vision; organization efficiency and access toresources; partnerships and collaborations; NGOs; and community advocacy. In addition, theyhighlight that the CGI projects in BC are affected by different institutional barriers and facilita-tors than the CGI projects in WA.4.5.1 Institutional barriers and facilitatorsThis research highlights 14 major institutional barriers to and facilitators of CGI implementationin BC andWA. Seven of them are commonly cited in the adaptation barriers and facilitators liter-ature such as governance structure and authority distribution; regulations; financial assistance;senior level support; organizational capacity; leadership and political will; and public knowl-edge and communication. The other seven are CGI specific barriers and facilitators, and theyinclude coastal jurisdiction and ownership; financial variation and flexibility; vision; organiza-tion efficiency and access to resources; partnerships and collaborations; NGOs; and communityadvocacy.The governance structure is a crucial factor in determining the roles and responsibilities ofdifferent levels of governments. This research suggests that the main differences in BC and150WA can be seen in the decision-making authorities allocated to their local governments. InBC, after the transfer of several key provincial roles to the local governments, the planning andimplementing of adaptation has become a local government responsibility. This downloadingclarified the ambiguities related to who is responsible for adaptation. However, without anychanges to the local governments’ access to already limited financial means and resources, andin the absence of provincial and federal financial programs, local governments are restrictedin their ability to address adaptation and implement measures like CGIs. In WA, although thelocal government responsibilities are similar to the BC local governments, they have accessto significantly more financial sources, such as tax revenues, government programs, and otherresources. For example, through the Flood Control Act and the Diking and Drainage Act, thestate government can create special purpose governments, for flood management districts. Thesedistricts operate with a mandate and can allocate federal and state funding to flood protectionprojects within their boundaries, as well as provide policy oversight and other technical andcapacity related resources.Arguably, the main barrier impeding the CGI projects in BC is the coastal jurisdiction and own-ership. The first issue is that the coastal jurisdiction in BC has several major overlaps betweenthe local governments and the provincial and federal governments. Therefore, any holistic ap-proach to adaptation requires multiple jurisdictions to work together and various regulations toadhere. The second issue is that the current coastal jurisdiction arrangement prevents any re-gional level management for adaptation. The provincial jurisdiction extends from the low waterline until the natural boundary, limiting actions that can be implemented within its jurisdictionto the foreshore. And the third issue is that there is no regional oversight to local governmentactions above the natural boundary. The lack of regional oversight results in the lack of contin-uous and coherent adaptation actions and implementation of specific measures. The adaptationmeasures are implemented in fragmented pieces the absence of a coordinated management prac-tice that sees the entirety of the coast as one connected system. On the contrary, the presence ofthe 1972 CZM Act in the United States and the corresponding SMA program in WA allocatesregulatory authorities to the state government in the SMA zone. This mandate and the cleardivision of regulatory responsibilities enable a structured regional vision, objectives, programs,partnerships, and plans. While WA has clear coastal jurisdiction and associated institutionalroles, it faces barriers related to the ownership of the coastal areas. The fact that beaches andtidelands are privately owned significantly impacts where the local governments and the stategovernment can lead CGI projects. CGI implementation in these areas relies on the private151owners’ initiatives, leading to fragmented pieces of CGI implementation.Regulations, resulting from the distribution of the decision-making authority and coastal juris-diction is both a driver that facilitates CGI implementation and a barrier that is work inten-sive, complicated and time-consuming. This research suggests that there are many contributingfactors to the regulations barriers and facilitators. In BC, the jurisdictional context and thefollowing regulatory environment do not favor CGI implementation. Due to the overlappingjurisdictions and the lack of coordination between these jurisdictions, it is challenging to findthe information on the regulations that the CGI projects need to comply with or obtain permitsfor. In addition, preparing and applying to various permits are time-consuming and work ex-tensive processes. The wait time to receive approvals is different depending on the jurisdictionreviewing the applications and can take a very long time. The wait time to receive approvals candelay the CGI project implementation, especially if the project funding is for a short period oftime. Moreover, CGI’s regulatory framework is not established extensively, and its integrationto the professional standards such as engineering and planning certificates have been limited.The regulatory gaps and the lack of integration to the professional standards make implementa-tion of CGI much more difficult than the traditional measures. In addition, the existing land usedesignations that do not provide space or category for CGI. Consequently, the current regula-tory environment has room for only small and simple CGI projects, as larger and more complexprojects likely face conflicting regulations, and experience significant challenges obtaining nec-essary permits.Financial assistance was an important barrier and facilitator of the CGI projects. The numberand availability of financial resources have increased the number of the CGI projects imple-mented in WA. In addition, all of the interview participants recognized financial assistance asa key factor. They described that the absence of financial assistance is a crucial barrier and thepresence of it is a significant driver.Besides the amount of the financial assistance available for the CGI projects, this research sug-gests that the continuity and diversity of financial sources, the flexibility of funding conditions,opportunities for cost saving arrangements, and financial assistance for specific purposes areessential contributing factors to the financial barriers and facilitators. The continuity of finan-cial resources ensures that there are long-term funding sources available for the CGI projects.In BC, the number of financial resources has increased over the two years, yet most of these152resources are available only for a limited period of time. The diversity of the financial resourcesensures availability of funding for the CGI implementation, even if one of the funding sourcesis cut. The funding amount for the CGI projects in WA was considerably higher in WA. Theproject review shows that funding from different state and federal institutions were allocated forthe CGI projects, where the funding for the projects in BC was allocated main from the federaland local institutions, showing the absence of the provincial role in CGI implementation. Theincrease in the number of diverse funding sources in BC over the last two years is, therefore, anencouraging sign for fostering more CGI projects. The flexibility of funding conditions ensuresthat projects can use the funding in a way that is appropriate to complete their objectives, andover a period of time that takes the lengthy permit applications and fish windows into account.In addition, providing flexible funding structures that provide cost-saving opportunities throughpartnerships, collaborations and merging projects can make more people and institutions to con-sider CGI. Lastly, developing specific funding programs targetting specific adaptation actionscan reduce competition in other funding sources, provide diversity in funding structure and in-crease CGI implementation. The projects in BC and WA can both benefit from the changes inthe funding structures.The findings of this research emphasize that the senior level government support is one of thekey barriers to and facilitators of CGI implementation. The technological and policy innovation,programs, tools and data, and training provided by provincial governments facilitates CGI im-plementation and impedes it when absent. In WA, the senior level support through continuousprograms, have helped the diffusion of the CGI projects. During the interviews participants inWA frequently praised the support they have received from their state and federal governments.On the contrary, there has been significantly less number of programs in BC. Some local gov-ernments and organizations feel as they have been left alone to tackle adaptation with limitedguidance and support from their senior governments. The findings of this research also suggestthat providing precedent, in the form of case studies and demonstration projects, illustrates thedesign, regulatory and construction processes. It sets an example and fosters the developmentof the CGI projects.The organizational capacity of institutions has frequently been cited in the adaptation and ur-ban green infrastructure literature. The CGI projects typically have to go through complicatedand competitive grant applications, permit applications, and other work intensive and lengthyprocesses. Small organizations and local governments with limited resources usually do not153have enough staff, staff time, or the right expertise to manage these processes. In addition, thefrequent changes in staff in key positions and departments delay progress.Besides the organizational capacity, CGI implementation also suffers from organizational inef-ficiency and lack of access to resources. An important barrier is the lack of access to data, tools,technologies, and dedicated room in the budget to monitor implemented projects. Especiallywhen governments are trying to deliver other essential services, allocating resources from pri-ority services to monitor the project effectiveness is challenging. The lack of efficiency in themeetings, project development, and review processes delay progress. In WA, the formal trainingof staff with facilitation techniques was reported to increase efficiency in the decision-makingprocesses.The research suggests that leadership and political will is an important barrier of and facilitatorto CGI implementation. Having leadership and political will have repeatedly been reported inthe adaptation literature. This is because leaders can influence CGI implementation by propos-ing and advocating for adaptation policies for the use of THSs or adaptive measures like CGI.The political will may be affected by reelection concerns or conflicting political interests, re-sulting in strategies with short-term gains to be preferred.In addition to the leadership and political will, having a formal regional and an organizationalvision was noted to be significant for CGI implementation. In WA, the use of a formal regionaland an organizational vision, called the collective impact model, has helped to bring differentlevels of governments, organizations, and stakeholder groups together to develop proceduresto work towards shared-objectives. Even though this partnership operates without a mandate,it is now considered to be one of the most successful partnerships in the region. On the otherhand, the lack of a formal vision in a similar partnership operating without a mandate in BCwas dissolved due to the lack of a formal vision leadership, and political will (semi-structuredinterview notes). Having a vision can also influence the type of CGI projects developed andapproved. Preparing proposals that provide multiple benefits can help sell the CGI projects, orgiving preference to projects with multiple objectives can lead the way for more CGI imple-mentation. For example, a project that aims to improve fish passages by removing culverts andrestoring fish habitat also stabilizes coastal edges and river mounts, thus unintentionally reduceserosion. The project reviews show that the fish habitat rehabilitation has been the primary ob-jective of the projects in BC and WA. Therefore, demonstrating the CGI’s role in fish habitatrehabilitation as well as restoring natural processes and flood and erosion protection can open154more opportunities for its implementation.Not commonly referenced in the literature, partnerships and collaborations are also importantbarriers to and facilitators of CGI. Most CGI projects involve multiple partners and collabora-tions with different jurisdictions. Bringing the right people in the project development, con-sultation, and implementation steps is not just important for CGI implementation but is alsoessential to obtain permits. Investing in the partnerships with community and private ownerscan increase the public interest and community in-in the CGI projects. In WA, most of theNGOs and local governments have developed protocols and programs to support public-privatepartnerships and collaborations (semi-structured interview notes). Yet, managing these partner-ships is usually time-consuming and difficult. As mentioned above, partners or collaboratorsmay have conflicting interests and agendas. There may also be issues related to the past andpresent relationships between different partners. Therefore, healthy and efficient partnershipswould entail the development of common objectives and good relationships amongst partners.A significant finding of this research is the roles NGOs play in facilitating CGI implementation.The results of the project review and the semi-structured interviews showed that the NGOs havean important lead role in initiating, developing and implementing CGI projects. Both in BCand WA, the NGOs were able to access financial resources that are typically not available togovernments, build stronger relationships with their communities and private owners, and brideinstitutions. Empowerment of such organizations, and increasing funding sources they can applyto increases the implementation of CGI.The local level advocacy for CGI is important for fostering political will and increasing thenumber of small-scale demonstration projects. The lack of local advocacy influences the atten-tion CGI receives from politicians, while the presence of it can promote funding sources to beallocated to CGI implementation and creation of partnerships.Lastly, the findings of this research emphasize the importance of public knowledge and commu-nication as barriers and facilitators for CGI implementation. Improved local knowledge on thenatural processes and associated risks, the effectiveness of CGI to reduce these risks, and the im-plementation costs of CGI are considered as important factors. This is because they contributeto the advocacy for CGI, and therefore increase the implementation of CGI. The communityperceptions built over false information damages the mainstreaming of CGI. A way to eliminatethese knowledge gaps and change the misperceptions is to develop robust, diverse and continu-155ous communication strategies, and to make sure that the scientific information is available andaccessible in a way that is understandable by the interested parties. For example, it has beenreported that the waterfront property owners and community workshops on coastal protectionhosted in various communities in WA have promoted small-scale CGI projects.This research contributes to the literature on adaptation and CGI implementation. The barriersand facilitators highlighted above provide new insights into what factors impede CGI imple-mentation and what strategies drive it. These institutional barriers make CGI implementationmore time consuming to go through and challenging to accomplish, now and in the future. Thefacilitators help developing strategies to overcome these barriers.The findings of this research suggest that adaptation and adaptation measures such as CGI neednew approaches to governance rather than trying to squeeze them in the existing institutionalroles and responsibilities. Due to its unique location at the interface between land and ocean(IPCC, 2014), CGI implementation requires a holistic approach to the jurisdictions and corre-sponding regulatory environment, rather than siloed actions and disconnected regulations. Insti-tutions with clear mandates and support from senior levels of governments are needed to addressadaptation adequately and holistically. Developing and implementing small to large scale CGIprojects require financial assistance, diversity in financial sources, and flexibility in fundingterms. Engineers, planners, emergency managers, and other key government staffs need to betrained in climate change adaptation and CGI’s efficiency and cost, as well as other tasks suchas funding application writing and meeting facilitation. Communities need to identify their dataknowledge and technology needs and look for ways in which they can acquire or access theresources that are required to undertake adaptation actions.Facilitating a holistic and resilient adaptation needs the development of a shared vision. Fol-lowing that vision, new strategies for building horizontal (between communities) and vertical(between different levels of institutions, property owners, and residents) partnerships and col-laborations need to be identified. To change public misperceptions over CGI’s effectivenessand costs, innovative communications and knowledge transfer strategies need to be developed.Moreover, communities need to take advantage of the unique roles NGOs have been playingin facilitating partnerships and collaborations, as well as improving knowledge of CGI and lo-cal advocacy. It should be acknowledged that most of the changes listed above are difficult toachieve and time-consuming, but not impossible. Political will that is separated from short-term156gains because of the re-election concerns and effective leadership that can put forward policiesand regulations for a holistic and resilient adaptation can help drive the change step by step.4.5.2 LimitationsThis research has several limitations. The project reviews of the CGI projects may not show thecomplete picture of all the CGI projects in the study area. Access to data has been a challengethroughout the project review, particularly in BC. A thorough review of the publicly accessibleproject databases was conducted. However, access to the projects and relevant information onFirst Nations and Indian Tribes territories was limited. This limitation is because the reportsof the projects on the First Nations and Indian Tribes territories were not always publicly ac-cessible. Therefore, the CGI projects that were not reported in the public databases were notincluded. Even though the qualitative information obtained throughout the interviews supportedthe findings of the project CGI reviews, the 235 projects reviewed may not be complete. There-fore, they may not represent the complete picture of the CGI projects in BC and WA.As discussed in the paper, the disparity of the number of projects in BC and WA may be thedirect result of the more densely developed nature of WA’s coastlines and untouched nature ofthe BC’s coastlines. InWA, more people live at the coast and are exposed to flooding and erosionrisks compared to BC. In addition, a higher portion of the coastline is hardened in WA than inBC. Historically, there have been many modifications toWA coastlines in the form of bulkheads,seawalls, culverts, dikes, and fillings (State of Washington Department of Ecology, 2010). Thistype of modification is less common in BC and exist mainly in the southern part of the province(Howes et al., 1994). Therefore the need for the CGI projects and the allocation of the state andfederal resources are likely to be higher in WA. Although, the project review showed that theresignificant institutional factors influencing the CGI projects, the differences in the populationdensity and naturalness of the coastal areas should be acknowledged as a limitation.The document review and synthesis section of this research did not detail each regulation andprogram available in BC and WA. It did list the relevant regulations that have been adoptedover time, and how they may relate to the CGI practices. However, merely the presence of aregulation may not always translate into a program that provides funding and other resources.The objectives of the semi-structured interviews were to understand whether the practitioners157were experiencing barriers and facilitators that were found during the previous steps and to iden-tify if they have been experiencing different and CGI-specific barriers. Although the selectedparticipants equally represented practitioners in BC and WA and different levels of institutions,the small sample size reduced the generability of the findings of the interviews. Moreover, theinterview participants were selected from the practitioners who have been involved in the CGIproject implementation in the last ten years. However, this selection excluded some of the otherimportant actors that are involved in CGI implementation, such as First Nations and IndianTribes communities, local politicians, lawmakers, and private owners. This exclusion furtherrestricted the generability of the findings of the interviews.It should also be noted that even though some of the findings of this research may be expe-rienced in other parts of the world, such as constraints related to regulations, financial issues,organizational capacity and efficiency, leadership and political will, or public knowledge, theresults should be considered within the context of the study area. The barriers and facilitatorsidentified in this study are the results of the past and present social and institutional constructs ofBC, WA, Canada and the United States. Therefore, some of the barriers and facilitators outlinedhere may not be present in other places.4.6 ConclusionThis research investigates the institutional barriers to and facilitators of CGI implementation.It does so by using a comparative approach to the institutional arrangements in BC and WA.It operationalizes this approach using three different but complementary methods; CGI projectreviews, document review and synthesis, and semi-structured interviews.The findings of this research indicate that CGI implementation is constrained by the CGI-specific institutional barriers and facilitators, and shares some common ones with the adaptationliterature. The results of this research show that the CGI implementation in BC has been limitedcompared to WA. Besides the geographical and density differences, this limitation is due to thespecific barriers in the institutional arrangements. This research provides 14 major institutionalbarriers to and facilitator of CGI implementation: governance structure and authority distribu-tion; regulations; financial assistance; senior level support; organizational capacity; leadershipand political will; public knowledge and communication; coastal jurisdiction and ownership;158financial variation and flexibility; vision; organization efficiency and access to resources; part-nerships and collaborations; NGOs; and community advocacy.The findings of this research suggest many complementary future directions. The developmentof a provincial and state CGI project database that details relevant project information would behandy for future research. A more comprehensive qualitative work using participatory methodslike focus groups and workshops with different actors (First Nations and Indian Tribes, electedlocal politicians, state lawmakers, local governments, and private owners) can yield more re-fined barriers and facilitators. It would also be helpful to consider a quantitative approach andinvestigate correlations between the CGI projects and the barriers and facilitators outlined inthis research. A qualitative or quantitative measurement of the effectiveness of the some of thefacilitators outlined in this research would help prioritization of the strategies. Studies inves-tigating the influence of the recent financial incentives in Canada and BC on the progressionof CGI implementation would be helpful. Lastly, future research can investigate what types ofinstitutional barriers and facilitators are experienced in other parts of the world, and comparethe results from different regions.159Chapter 5ConclusionThis dissertation is comprised of three empirical studies aiming to understand the environmen-tal, local and institutional contexts in which CGI can be used as a sea level rise adaptationmeasure within the coastal regions of British Columbia (BC) and the Washington State (WA).This chapter will provide the summary of findings, broad conclusions and policy implications,strengths and limitations, future research directions, and personal reflections.5.1 Summary of findingsIn this dissertation, I sought to contribute to the scholarship on coastal green infrastructure’scontext-dependency as a sea level rise adaptation measure. As I discussed at several points inthis dissertation, the literature on CGI has frequently pointed to the context-dependency of CGI(i.e., Catenacci and Giupponi 2013; Langridge et al. 2014; Biesbroek 2014; Oddo et al. 2015).However, the influence of different contexts on the functionality, adaptability, and implemen-tation of CGI has been greatly understudied. Therefore, this dissertation aimed to understandthe contexts in which CGI can be used as a sea level rise adaptation measure within the coastalregions of BC and WA, and three empirical studies were designed to achieve the research goal.Achieving this goal in Chapter 2 involved the development and synthesis of CGI indices toidentify CGI’s potential coastal protection benefits in an environmental context. I hypothesizedthat CGI’s vulnerability to changing environmental conditions influences its protection bene-fits. I argued that both CGI’s protection benefits and vulnerability vary by the environmentalcontexts in which CGI is considered. I suggested that incorporating CGI’s vulnerability andusing spatially linked environmental indicators can provide a holistic approach to identifyingCGI areas with high coastal protection potential. Therefore, I developed two indices: the CGI160coastal protection index and the CGI vulnerability index. I organized and calculated the indicesusing Gornitz (1991)’s CVI methodology that has been used in Canada and the United Statesfor coastal sensitivity assessments. I applied the indices to the Salish Sea region to demonstratethe methodology. Lastly, I synthesized the indices to achieve the goal of the chapter.The results of Chapter 2 showed that the CGI’s coastal protection potential varies by the en-vironmental contexts. The results indicated that the big population centers in the Salish Searegion may not have the ideal environmental conditions for CGI to provide coastal protectionbenefits. On the contrast, the smaller communities in the study have higher potential to utilizeCGI for coastal protection. Overall, the results showed that CGI in the north of the study area,which is the southwest of BC, have higher coastal protection potential than the CGI in the southof the study area (north of WA). This is mainly due to the high vulnerability of CGI in WA.This vulnerability is caused primarily by the low-lying slopes, sea level change trends that showsubsidence, thus higher sea level rise, and denser developments at the WA’s coasts. In addition,the differences in coastal types also contributed to these findings, where approximately 3% ofthe BC’s coastline and 17% of the total WA’s coastline is hardened with human-made structures.This is because human-made structures at the coast both reduces CGI’s vulnerability by reduc-ing habitat and increases the wave impacts by amplifying wave spray. The results of this chapterare informative for the environmental policy and management practices in the study region butnot as informative for other regions. The results are relative to the study area and should beinterpreted within the contexts of this study area as the CGI in the low potential zone in thisstudy area may fall under a higher or lower zone in other regions.In Chapter 3, accomplishing the research goal involved the development of the local sea levelrise adaptation scenarios and an evaluation framework to highlight the trade-offs between CGIand other adaptation strategies in a local context. I hypothesized that the adaptation and re-silience contributions of CGI vary by the characteristics and capacities of the local commu-nities. I suggested that the coastal processes, natural and built environments, and economic,institutional and social implications of the adaptation strategies should be considered to under-stand the tradeoffs between CGI and other adaptation strategies in local contexts. I employeda case study with the District of North Saanich in BC to conceptually apply the research meth-ods. I first developed four local sea level rise adaptation scenarios, using public perspectiveson the preferred adaptation strategies and expert feedback. Second, I developed an evaluationframework that incorporates coastal processes, natural and built environments, and economic,161institutional and social components of community adaptation and resilience. Finally, I appliedthe framework to the scenarios to achieve the goal of the chapter.The conceptual application of the methods to the case study community showed that there aretrade-offs between CGI and other adaptation strategies due to the local characteristics of thecommunity. The results of Chapter 3 are consistent with the findings of other studies, suggest-ing that overall CGI is a valuable adaptation measure with co-benefits and that THSs have neg-ative implications for coastal processes and natural environments. The “protection with coastalgreen infrastructure” scenario performed the best in the evaluation framework, followed by the“built environment accommodation”, “retreat from the coast”, and “protection with traditionalhard structures” scenarios. The analysis demonstrated that CGI provides benefits for the coastalprocesses, natural and built environment, economic and social modules, but has important in-stitutional trade-offs compared to other strategies. The institutional trade-offs were caused bythe lack of integration of CGI into the existing environmental, urban planning and engineeringregulations, as well as the lack of staff expertise and data, tools and other resources that couldsupport CGI strategies. The results also highlighted the trade-offs between the economic andinstitutional implications of the “built environment accommodation” scenario; the economic,institutional and social implications of the “retreat from the coasts” scenario; and lastly, thecoastal processes and natural environment implications of the “protection with traditional hardstructure’ scenario. The findings of this study need to be considered within the local the study isapplied. Different sea level rise adaptation strategies may perform better than the CGI strategies,and the local trade-offs of strategies may vary in different local contexts.Lastly, achieving the research goal in Chapter 4 involved comparing the CGI project implemen-tation, institutional arrangements and practitioners’ perspectives in BC and WA to identify theinstitutional barriers to and facilitators of CGI implementation. I hypothesized that the imple-mentation of CGI depends on the institutional contexts. I discussed that CGI implementationis influenced by the barriers to and facilitators of the general adaptation concept, but also byCGI-specific barriers and facilitators. I searched the publicly available databases to collect infor-mation on the CGI projects implemented between 2008-2018 in BC and WA. Next, I reviewedthe governance structures, distribution of decision-making authorities, coastal jurisdiction andownership boundaries, and regulations and programs in BC and WA. Lastly, I conducted semi-structured interviews with the practitioners in BC and WA to achieve the goal of the chapter.162The findings of Chapter 4 showed that the implementation of CGI is influenced by the differ-ences in institutional arrangements in BC andWA. The results showed that there are significantlymore CGI projects implemented in WA than in BC and that there are important differences inways and capacities governments organizations have been involved in the CGI project imple-mentation. The findings indicated that the coastal zone is managed and regulated in silos inBC and more holistically in WA. The downloading of senior-level responsibilities to local gov-ernments without providing adequate resources have had significant implications in BC than inWA. The results showed that different barriers and facilitators influenced the CGI project imple-mentation in BC and WA, owing to the differences in their institutional arrangements. Overall,the general adaptation barriers and facilitators, as well as several key CGI-specific barriers andfacilitators influence CGI implementation.A total of 14 barriers and facilitators were identified and grouped under governance structureand authority distribution; regulations; financial assistance; senior level support; organizationalcapacity; leadership and political will; public knowledge and communication; coastal juris-diction and ownership; financial variation and flexibility; vision; organization efficiency andaccess to resources; partnerships and collaborations; NGOs; and community advocacy. Mostimportantly, the amount, diversity, and continuity of federal and state programs in WA haveprovided the financial opportunities for the CGI projects. Moreover, the holistic managementof WA coastlines through the clear distribution of responsibilities and regulations, as well asthe federal and state level political, technical and guidance have empowered different levels ofgovernments and organizations to undertake CGI projects. In addition, organizational capacityand access to resources have enabled action in WA. In BC, the overlapping coastal jurisdictionhas created a restricted environment for CGI implementation. Financial, regulations and se-nior support barriers have limited CGI’s progression in BC. The lack of vision, leadership andpolitical will have further restricted action in BC. NGOs have been very effective in BC andWA to mobilize resources, create partnerships and reach out to residents because of their non-government status and ability to build relationships with residents. Similar to the other chaptersof this dissertation, the findings of Chapter 4 need to be considered within the geographical re-gion in which the research was conducted. The barriers and facilitators identified in Chapter 4may not be present in other countries and geographical regions.1635.2 Broad conclusions and policy implicationsBy investigating CGI’s environmental, local and institutional contexts, I have identified a num-ber of broad conclusions and policy implications.This dissertation shows that environmental, local and institutional contexts influence the waysCGI provides coastal protection benefits, contributes to the local adaptation and resilience, and islimited or driven by institutional arrangements. This finding highlights that communities needto undertake context-specific investigations prior to considering, deciding and implementingCGI as a sea level rise adaptation measure. Communities need to confirm if they have theappropriate environmental conditions to obtain coastal protection benefits from CGI. They needto verify if the local tradeoffs of CGI are more appropriate than some of the other adaptationstrategies. They need to explore if any institutional barriers may prevent or discourage themfrom implementing CGI.Each chapter of this dissertation provides important contributions to the management, regula-tion, prioritization, decision-making, and knowledge sharing practices in the study area. Theuse of CGI can help protect critical ecosystems, natural processes, habitats, and biodiversity,even in a case where the coastal protection benefits of CGI may not be high, or other strategiesmay have higher adaptation and resilience benefits. Therefore, hybrid or alternative ways ofCGI implementation should be considered.The findings of this dissertation do not only contribute to the CGI literature but also to thebroader adaptation literature. Adaptation also depends on the contexts that it is considered andone size fits all adaptation approaches fail to deal with the local challenges communities face(Bierbaum et al., 2012). Adaptation is not, and should notbe just about protection from thephysical impacts of sea level rise. Considering that the coastal communities are coupled socialand ecological systems (Kittinger and Ayers, 2010), the local level economic, environmental,social, and institutional adaptation to sea level rise should also be regarded.The economic module of the evaluation framework developed in Chapter 3 and the findingsof other studies (i.e., Fankhauser 2009; Hinkel et al. 2014; Lu et al. 2018) remind that anyactions for sea level rise adaptation will be costly. Some adaptation measures will likely bemore costly than the others (Haer et al., 2017). Besides, the cost of adaptation will likely notbe shared equally amongst levels of governments and within communities (Salzmann et al.,1642016). Moreover, an increasing number of studies predict that the cost of adaptation will behigher in the future (i.e., Hallegatte et al. 2013; Nicholls 2015). The cost of adaptation wasalso a theme brought up multiple times by the interview participants in Chapter 4. The lack offinancial assistance, and variations and flexibility of financial programs were frequently statedas important barriers to CGI implementation. In addition, the Canadian local governments’limited access to financial resources such as taxation is recognized as a key factor impeding theimplementation of adaptation actions. Therefore, communities and all levels of governmentsneed to determine who pays for adaptation, and how. Financial mechanisms and financingmodels need to be developed and studied to reduce the current and future costs of adaptation.The inclusion of public perspectives in Chapter 3 was an essential part of the research designand provided valuable insights into the concerns and expectations of the residents in the studyarea. However, the participants responded differently to the questions on what strategies areeffective to protect coasts from flooding and erosion, and what strategies they would like tosee implemented in their community’s coastline. This disparity showed that there might bedifferent motivations behind public’s adaptation strategy choices. The differences in public’sand decision-makers’ motivations were also highlighted in Chapter 4. Practitioners in BC andWA described challenges regarding collaborating with partners with different agendas or in-terests. Therefore, it is essential for the researchers, planning teams and decision-makers torecognize these differences and understand the underlying causes and rationales behind publicand decision-maker opinions. Sea level rise impacts on local real estate prices and the fear oflosing a portion of the private lands to strategy implementation are real concerns influencingpublic opinions over adaptation strategies.The results of Chapter 4 showed a clear difference between the CGI projects implemented in BCand WA. WA had significantly more number of CGI projects implemented than BC. However,the results of Chapter 2 showed that CGI in WA had less coastal protection potential than CGIin BC. The differences in CGI’s coastal protection potential and the implemented CGI projectsin the study area emphasize two important points. First, CGI projects can be implemented evenif the objective of the implementation is not flood and erosion protection. CGI’s wide array ofsocial, environmental, and economic benefits attract investments from all levels of governments.The CGI project review in Chapter 4 showed that most of the projects were implemented torestore habitats, rather than to provide flood and erosion protection. Second, the differences inChapter 2 and 4 findings highlighted the importance of accounting for local variations. CGI165can be implemented on small properties or large regional coastlines, and the local variations canplay important roles in determining where CGI can be used.Lastly, the scientific evidence shows that communities will experience more rapid changes infuture due to accelerated impacts of climate change (Devoy, 2015; Nicholls, 2015; Nauels et al.,2017). However, our planning and engineering practices, as well as the national and regionalpolicies and regulations are not yet well-equipped to deal with and respond to rapid changes.Therefore, institutional adaptability, flexibility, and creativity are needed for meaningful andsustainable sea level rise adaptation.5.3 Strengths and limitationsA key strength of this dissertation is that it investigates three contexts that influence the use ofCGI as a sea level rise adaptation measure: environmental, local and institutional. In addition,it employs multiple quantitative and qualitative methodologies, incorporates public perspectivesand opinions, and advises with expert knowledge and feedback in various parts of the research.Moreover, the methods and tools developed in this dissertation can be used independently andapplied to different community scales and regions. Besides the strengths, the dissertation has anumber of limitations. Each chapter of the dissertation includes specific limitations of the meth-ods, research activities, and analysis. Here, I provide an overview of the limitations identified.I would argue that the greatest weakness of this dissertation is related to the data and samplesize. As with many other research projects, this dissertation is constrained by the availabilityof and access to data, as well as the willingness of the public, experts, and practitioners toparticipate. In Chapter 2, this limitation inevitably affected the number of indicators used forthe development of the CGI indices. Some of the key themes used to measure CGI coastalprotection benefits and vulnerability (i.e., CGI cover, sedimentation, and soil properties) wereexcluded from the development of the CGI indices due to the lack of availability and access tothe spatially-linked data. In addition, there were differences in the scale of the data collectedfor the study area. For example, the coastal morphology data were in fine resolution and wereavailable for small coastal segments, where land use data were in much coarser resolution).Therefore, when indices were calculated, the finer resolution data were aggregated to largersegments to match the coarser resolution data. This approach was suitable to the regional scale166and the unit of analysis of this study. However, the use of the finer resolution data wouldhave been more beneficial to assess the implications of the biophysical and geomorphologicalcharacteristics and the built environment on CGI’s coastal protection benefits and vulnerability.The approach used in Chapter 2 overlooked the local variations that could have been captured ifthe finer resolution data were to be used or the coarser resolution data were to be dissipated tothe finer resolution.In Chapter 3, the data and sample size limitation resulted in the low sample size for the expertelicitation survey and the lack of participation of the local community in the application of theevaluation framework to the adaptation scenarios. Even though an extensive literature reviewwas conducted for the development of the evaluation framework and a considerable effort wasput into recruiting experts, higher participation in the expert elicitation survey would have madethe framework more robust. In addition, the local community’s participation in the applicationof the framework would have been ideal. However, research depends on the mutual understand-ing, and functioning and healthy relationships between the researchers and the participants. Itis important to understand communities’ concerns and explore ways to work through the sensi-tivities and issues related to communities’ objectives and fears over public perceptions. In caseswhere the research objectives can no longer be achieved within the existing relationship with thestudy area community, researchers would need to find alternate pathways. Therefore, alterationsto the research activities were needed in Chapter 3 to accommodate for the changes regardingthe community’s participation in the project.In Chapter 4, the data and sample size limitation led to the gaps in the project information andthe number of recorded CGI projects in BC. I should note here that based on my conversationswith the practitioners and knowledge of the region, I believe the number of CGI projects in BCwould not go up by much if I would have had more access to data. The number of the CGIprojects in BC would still be significantly lower than the ones in WA. Nevertheless, access toinformation on more CGI projects in BC would have helped to portray a more accurate picturein the study area.There are also limitations related to the goal of the dissertation. First, even though it was nota research goal, it should be acknowledged that this dissertation did not explain the magnitudeof influence the environmental, local, and institutional contexts have on the use of CGI. Forexample, the extent of protection benefits or vulnerability of CGI in Chapter 2, the extent ofeconomic impacts of strategies on local communities in Chapter 3, and the extent of the barriers’167or facilitators’ impacts on CGI implementation in Chapter 4 were not explained.Second, a number of different ways could have been used to develop the CGI indices in Chapter2 and the evaluation framework in Chapter 3. For example, a combined CGI coastal protectionand vulnerability index could have been developed, or different organizational and computa-tional methods for the indices could have been used in Chapter 2. I decided to use two separateindices because they can be used independently. Practitioners and researchers might be inter-ested in using only one of the indices rather than two together. For example, the CGI vulnera-bility index can be used as a stand-alone index to explore CGI’s vulnerability to environmentalchange, independent from CGI’s coastal protection benefits in that location. The results caninform priority locations for habitat restoration/rehabilitation actions. I also chose to employGornitz’s CIVI methodology to organize and compute the indices because of its compatibilitywith the coastal sensitivity assessments in Canada and the United States. These deliberate de-cisions, therefore, eliminated the consideration of some other potential methods by default. InChapter 3, in an attempt to develop a straightforward evaluation framework, I chose to use a+1 to -1 scoring system and a criteria format that compares the proposed strategies to baselineconditions. This decision excluded some important concepts (such as political leadership, col-laborations with neighboring communities, and so on) from the framework because they occurindependently from the strategies and therefore, could not be compared to the baseline condi-tions and could not be scored using the +1 to -1 scoring system.Third, some other aspects of CGI’s context dependency were left out of the scope of this disserta-tion. The temporal variability of CGI influences its coastal protection benefits and vulnerabilityto environmental change due to the changes in CGI density, plant thickness, and area throughoutseasons (Koch et al., 2009). However, the temporal aspects of CGI’s coastal protection benefitsand vulnerability were not included in Chapter 2. In Chapter 3, each scenario included onlyone strategy. However, it is likely that communities will implement more than one strategy totackle the multi-faceted challenges of sea level rise. In Chapter 4, I focused on the institutionalarrangements to explain the differences in CGI implementation in BC and WA. Other importantfactors, such as the progression of climate change adaptation and the use of CGI terminology inpolicies and regulations in Canada and the United States, were left out of scope.Lastly, the tools developed in each chapter and the results of the dissertation are to some extentlimited to the geographical region where the research was conducted. For example in Chapter1682, CGI types such as mangroves and reefs were not included in the indices because the studyregion does not have such CGI types. The local trade-offs identified in Chapter 3 are specificto the study area community and may not be present in other places. Lastly, Canada and theUnited States are both developed western countries: therefore, the results of Chapter 4 may notrepresent the institutional barriers to and facilitators of CGI implementation in other parts ofthe world. However, the methodologies used in each chapter could be applied elsewhere, andtools/frameworks can be amended to be used in other studies.5.4 Future research directionsSince the CGI research is relatively new, there are many fruitful research directions. The lim-itations discussed above suggest some general and priority areas for further research in orderto advance understanding of the context-dependency of CGI and to support local and regionaldecision-making regarding the use of CGI as a sea level rise adaptation strategy. These researchdirections are grouped under the following categories: general directions, methodological ap-proaches, expanding the concepts, application to different scales and places, and non-researchdirections that could help future research.General directionsOther contexts that influence the use of CGI as a sea level rise adaptation measure can be in-vestigated to improve understanding of CGI’s context-dependency. These contexts may includethe economic/financial context, where various mechanisms to pay for the CGI projects and thedivision of the costs amongst homeowners and taxpayers can be investigated. Alternatively, acoastal urban context can be explored, where hybrid THS/CGI measures are considered. So-cial contexts of CGI can also be investigated to better understand the public knowledge on andacceptability of CGI as an adaptation measure.As I mentioned in the previous section, this dissertation did not explain the magnitude of theinfluences that the environmental, local, and institutional contexts have on the use of CGI. There-fore, future research can investigate the magnitude of influence different contexts have on CGI.This involves posing different research questions such as what are the impacts of CGI’s vul-nerability on its coastal protection potential; what are the coastal processes, natural and builtenvironment, economic, institutional and social implications of different adaptation strategies;or, to what degree the institutional barriers and facilitators influence CGI implementation?169Methodological approachesFuture research can test the use of different methodological approaches. For example, alternativemethods to incorporate CGI vulnerability to its coastal protection benefits can be explored.Different organizational and computational approaches can be tested for the CGI indices ratherthan Gornitz’s CIVI approach. Rather than the indicator-based methods, different approachescan be employed to measure CGI coastal protection benefits and vulnerability.In addition, quantitative approaches can be employed to measure the coastal processes, nat-ural and built environment, economic, institutional and social impacts of different adaptationstrategies. Similarly, quantitative and qualitative methods of evaluating the contributions of theevaluation framework to the local decision-making processes can be explored.Furthermore, quantitative approaches can be considered to investigate correlations between theCGI projects and the institutional barriers and facilitators. Besides the semi-structured inter-views, other participatory methods such as focus groups or workshops can be employed to ob-tain practitioners’ perspectives. Lastly, qualitative or quantitative methods can be employed tomeasure the influence of the institutional barriers to and facilitators of the CGI implementation.Expanding the researchFuture research can also expand the context and design of the evaluation framework developedin Chapter 3. Workshops or focus groups with experts from different government levels and or-ganizations can be organized to review and enrich the evaluation framework. This may includeadding other modules or components or finding alternative scoring systems to account for vari-ations in the implications of strategies. For example, an ecological/biodiversity component canbe added, or a {1 to 3}, {1 to 5} or an alternative point system can be considered. In addition,how multiple strategies can be evaluated under different future scenarios can be explored.Similarly, future research can expand the semi-structured interview method in Chapter 4. Amore comprehensive qualitative research that incorporates diverse perspectives on the institu-tional barriers to and facilitators of CGI implementation can be conducted. Practitioners fromthe federal government, First Nations and Indian Tribes, elected local politicians, state lawmak-ers, local governments, and private owners can be included. Future research can also track anddocument how the recent financial incentives in Canada drive (or not) the progression of CGIimplementation.170Application to different scales and placesFuture research can explore the application of the indices developed in Chapter 2 at a local scaleto identify coastal segments where CGI yields higher coastal protection potential. In addition,the application of the evaluation framework developed in Chapter 3 to different communitiescan improve the understanding of the trade-offs between strategies. Similarly, the applicationof the methods in Chapter 4 to different regions around the world can bring more light into theinstitutional barriers to and facilitators of CGI implementation.Non-research directions that can help future researchThere are also some important non-research directions that can help future research. For exam-ple, the evaluation framework can be operationalized. It can be formatted as an online interactivedocument, where planning teams can customize, fill and download the document. This way, theframework can be easily implemented and tracked for future assessments.Alternatively, a region-wide CGI project database can be established that systematically collectsand details the information on the implemented CGI projects. Such a comprehensive databasecan reduce data gaps; track the progression of the projects; contribute to knowledge sharing andspread of best practices; and provide valuable data for various research projects looking intoenvironmental, economic, institutional and social factors influencing CGI implementation.5.5 ReflectionsI discussed the significance and limitations of the research findings at length in the previoussections. Here, I want to conclude by noting some of my reflections on the research process andsome of the findings.To start with, I find some findings of this dissertation surprising. Based on my knowledge of thestudy area, I was aware of the numerous governmental and non-governmental initiatives in WAthat have been investing in CGI projects. Therefore, the low CGI coastal protection potential inWA was an unexpected result of Chapter 2. But as I mentioned in the broad conclusions sec-tion, this finding highlights the regional scale of the study and its limitations in capturing localvariations. Therefore, I believe the results be considered to provide guidance for further inves-tigations into the specific coastal segments of communities’ coastlines. These coastal segmentscan be identified through coastal types using the geomorphological or ecological features such171as unvegetated mud flats can be separated from the sloped beaches and rocky coastlines.I was also surprised by the built environment, economic, institutional and social trade-offs ofthe managed retreat strategy. I believe this surprise was due to the wide recognition of managedretreat as the most sustainable and long-term strategy for sea level rise adaptation in the literature(i.e., Pethick 2002; Abel et al. 2011; Alexander et al. 2012). However, the findings of Chapter3 suggest similar trade-offs to Hino et al. (2017)’s findings, which suggested that the socio-policial implications of managed retreat strategy are important barriers for its implementation.These unexpected findings pushed me to use a more critical lens during the research activitiesand interpretation of the results.On a different note, this research challenged me to think about the difficulties of conductingresearch with local communities and other participatory approaches. I have found that the chal-lenges around research collaborations with local communities are often understated. I feel thatthere is a disconnection between the common academic research objectives and methods, andlocal governments’ political and public-related concerns. In my observation, the political con-cerns are related to election cycles and re-election desires and maintaining key relationshipswith the prominent members of public or strong community groups. The public-related sensi-tivities are often around the (perhaps, premature) exposure of the public to information on sealevel rise impacts, local risks, and adaptation options. Both of these concerns seemed to befueled by the fears of potential implications of sea level rise adaptation on local property values.I think researchers should be aware of these concerns and understand that even academic re-search exploring adaptation options can contribute to such fears. In cases where the completionof research depends on the continuous participation of local communities, the prior understand-ing of these concerns and having backup plans can ensure researchers to achieve their researchobjectives.This research also made me think about the role of the communities’ demographics, economicwealth, and social status in their adaptation planning processes. I think the research on adapta-tion commonly focuses on high-risk communities (either due to high exposure and sensitivity,or low capacity) with low social and economic status. However, research on wealthy communi-ties’ adaptation planning processes seems to be very limited. I think that the existing economicwealth and social status of communities, and the age group of the residents play important rolesin communities’ attitudes towards adaptation needs. A wealthy community is likely to considera wider variety of adaptation options, even the expensive ones because the local government172and property owners can likely afford the implementation and maintenance costs. On the otherhand, a community with low social and economic status may have to exclude some adaptationoptions from public debate. Therefore, the processes followed and conclusions driven from theadaptation research on high-risk communities with low social and economic status may not beobserved in other types of communities.As adaptation occurs in an extended period, communities with older populations may not bevery concerned with the impacts of sea level rise. During the workshop in Chapter 3, someparticipants repeatedly stated that they are not concerned with sea level rise impacts because oftheir age (about 53% of the study community is 55 years and older (Statistics Canada, 2017),and 70% of the workshop participants were 55 years and older). They stated that they will notbe around to experience the impacts of sea level rise and that they are mainly concerned withthe effects of the local decisions on their property values. These statements were not reflected asstrongly in the surveys they filled during the workshop. I found myself contemplating these con-cerns and the literature on sea level rise impacts and objectives adaptation for a long time. I havecome to the conclusion that the full extent of the implications of sea level rise and adaptationactions are not entirely understood yet.Lastly, adaptation is not an easy task and requires time for planning, implementation, and evalua-tion. Communities need to identify the data gaps and expertise needs for undertaking adaptationprocesses, and they need to work towards to fill those gaps and needs urgently. In addition, theyneed to start transparent and meaningful engagement processes with their residents to under-stand their concerns and desires, to increase public knowledge on adaptation strategies and theirpotential implications, and to reduce misperceptions over THSs and CGI.CGI provides a critical resource for communities, not just for adaptation to sea level rise butalso for the greater community resilience. It is my hope that this dissertation will contribute tothe shift from using THSs to a more beneficial, adaptable, flexible and multifunctional way ofprotecting our coasts using CGI. I hope that this dissertation will improve the public, politicaland practitioner understanding of CGI and will lead to meaningful change in practices andpolicies.173BibliographyAbel, N., Gorddard, R., Harman, B., Leitch, A., Langridge, J., Ryan, A., and Heyenga, S.(2011). Sea level rise, coastal development and planned retreat: Analytical framework, gov-ernance principles and an Australian case study. Environmental Science & Policy, 14(3):279–288.Adelsman, H., Ekrem, J., Binder, L. W., Cook, K., Cooper, K., Helbrecht, L. M., Jamison, R.,Phillips, J., Pickett, P., Roalkvam, C. L., Salisbury, S., Siemann, D., and Speaks, P. (2012).Preparing for a Changing Climate: Washington State’s Integrated Climate Response Strategy.Department of Ecology, Publication No. 12-01-00404, Olympia, WA. Accessed on April 2018.Retrieved from:, W. N. (2003). Social Capital, Collective Action, and Adaptation to Climate Change.Economic Geography, 79(4):387–404.Adger, W. N., Agrawala, S., Mirza, M., Conde, C., O’Brien, K., Pulhin, J., Pulwarty, R.,Smit, B., and Takahashi, K. (2007). Assessment of adaptation practices, options, constraintsand capacity. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution ofWorking Group II to the Fourth Assessment Report of the Intergovernmental Panel on ClimateChange, Cambridge University Press, Cambridge, UK. Accessed on August 2018. Retrievedfrom:, W. N., Arnell, N. W., and Tompkins, E. L. (2005). Successful adaptation to climatechange across scales. Global Environmental Change, 15(2):77–86.Airoldi, L., Abbiati, M., Beck, M. W., Hawkins, S. J., Jonsson, P. R., Martin, D., Moschella,P. S., Sundelo¨f, A., Thompson, R. C., and A˚berg, P. (2005). An ecological perspective onthe deployment and design of low-crested and other hard coastal defence structures. CoastalEngineering, 52(10-11):1073–1087.174Alexander, K. S., Ryan, A., and Measham, T. G. (2012). Managed retreat of coastal commu-nities: understanding responses to projected sea level rise. Journal of Environmental Planningand Management, 55(4):409–433.Anderson, M. E. and Smith, J. M. (2014). Wave attenuation by flexible, idealized salt marshvegetation. Coastal Engineering, 83:82–92.Archaeological Sites and Resources Act of 1979 (1979). Chapter 27.53 RCW.Arkema, K. K., Griffin, R., Maldonado, S., Silver, J., Suckale, J., and Guerry, A. D. (2017).Linking social, ecological, and physical science to advance natural and nature-based protectionfor coastal communities. Annals of the New York Academy of Sciences, 1399(1):5–26.Arkema, K. K., Guannel, G., Verutes, G., Wood, S. A., Guerry, A., Ruckelshaus, M., Kareiva,P., Lacayo, M., and Silver, J. M. (2013). Coastal habitats shield people and property fromsea-level rise and storms. Nature Climate Change, 3(10):913–918.Arnold, P. A. (2004). About America: How the United States is Governed? U.S. Depart-ment of State, Bureau of International Information Programs, Herndon, Virginia. Accessedon April 2018. Retrieved from: 122709/US Governed.pdf.Azevedo de Almeida, B. andMostafavi, A. (2016). Resilience of Infrastructure Systems to Sea-Level Rise in Coastal Areas: Impacts, Adaptation Measures, and Implementation Challenges.Sustainability, 8(11):1115–28.Barbier, E. B., Georgiou, I. Y., Enchelmeyer, B., and Gutierrez, J. L. (2013). The value of wet-lands in protecting southeast louisiana from hurricane storm surges. PLoS ONE, 8(3):e58715.Barbier, E. B., Hacker, S. D., Kennedy, C., Koch, E.W., Stier, A. C., and Silliman, B. R. (2011).The value of estuarine and coastal ecosystem services. Ecological Monographs, 81(2):169–193.Barnett, J., Waters, E., Pendergast, S., and Puleston, A. (2013). Barriers to adaptation tosea level rise. National Climate Change Adaptation Research Facility, Gold Coast. Accessedon August 2018. Retrieved from: filespublications/Barnett 2013 Barriers to adaptation to sea level rise.pdf.175Barnhill, K. and Smardon, R. (2012). Gaining ground: Green infrastructure attitudes andperceptions from stakeholders in Syracuse, New York. Environmental Practice, 14(1):6–16.Barron, S., Canete, G., Carmichael, J., Flanders, D., Pond, E., Sheppard, S., and Tatebe, K.(2012). A Climate Change Adaptation Planning Process for Low-Lying, Communities Vulner-able to Sea Level Rise. Sustainability, 4(9):2176–2208.Bauer, A., Feichtinger, J., and Steurer, R. (2012). The Governance of Climate Change Adap-tation in 10 OECD Countries: Challenges and Approaches. Journal of Environmental Policy& Planning, 14(3):279–304.Bauer, A. and Steurer, R. (2014). Multi-level governance of climate change adaptation throughregional partnerships in Canada and England. Geoforum, 51:121–129.BC Parks (2018). Provincial Protected Area Boundary Adjustment Policy, Process andGuidelines. Accessed on July 2018. Retrieved from: Environmental Assessment and Coastal Health (BEACH) Act of 2000 (2000). 33U.S.C. §1313 et seq.Becklumb, P. (2013). Federal and Provincial Jurisdiction to Regulate Environmental Is-sues. Economics, Resources and International Affairs Division Parliamentary Informationand Research Service, Publication No. 2013-86-E. Accessed on July 2018. Retrieved from:´langer, C. (2005). Comparing Canadian and American federalism. Studies on the CanadianConstitution and Canadian Federalism. Accessed April-2018. Retrieved from:, M. A. and McMahon, E. T. (2006). Green Infrastructure. Linking Landscapes andCommunities. Island Press, Washington, DC.Berkes, F. (2010). Devolution of environment and resources governance: Trends and future.Environmental Conservation, 37(04):489–500.Berrang-Ford, L., Ford, J. D., and Paterson, J. (2011). Are we adapting to climate change?Global Environmental Change, 21:25–33.176Berte, E. and Panagopoulos, T. (2014). Enhancing city resilience to climate change by meansof ecosystem services improvement: A SWOT analysis for the city of Faro, Portugal. Interna-tional Journal of Urban Sustainable Development, 6(2):241–253.Bettini, Y., Brown, R. R., and de Haan, F. J. (2015). Exploring institutional adaptive capacity inpractice: Examining water governance adaptation in Australia. Ecology and Society, 20(1):1–47.Bierbaum, R., Smith, J. B., Lee, A., Blair, M., Carter, L., Chapin, F. S., Fleming, P., Ruffo,S., Stults, M., McNeeley, S., Wasley, E., and Verduzco, L. (2012). A comprehensive reviewof climate adaptation in the United States: More than before, but less than needed. Mitigationand Adaptation Strategies for Global Change, 18(3):361–406.Biesbroek, G. R. (2014). Challenging barriers in the governance of climate change adaptation.PhD thesis, Graduate School for Socio- Economic and Natural Sciences of the Environment(SENSE), Wageningen University, Laxenburg, Austria. Accessed on July 2018. Retrievedfrom:, L. C. W., Barcelos, J. K., Booth, D. B., Darzen, M., Elsner, M. M., Fenske, R., Graham,T. F., Hamlet, A. F., Hodges-Howell, J., Jackson, J. E., Karr, C., Keys, P. W., Littell, J. S.,Mantua, N., Marlow, J., McKenzie, D., Robinson-Dorn, M., Rosenberg, E. A., Sto¨ckle, C. O.,and Vano, J. A. (2010). Preparing for climate change in Washington State. Climatic Change,102(1-2):351–376.Black, R. A., Bruce, J. P., and Egener, I. D. M. (2017). Managing the Risks of Climate Change:A Guide for Arctic and Northern Communities Volume 1. Centre for Indigenous Environmen-tal Resources. Accessed on January 2018. Retrieved from: Adaptation Guide Volume 1.pdf.Blair, D. G. (2009). Coastal Shore Jurisdiction in British Columbia. The Stewardship Cen-tre for British Columbia. Accessed on April 2018. Retrieved from:, D. G. (2014). Green Shores Background Report: Shoreline Regulations andPermitting Processes in BC. The Stewardship Centre for British Columbia. Accessedon July 2018. Retrieved from: docs/greenshores/ResourcesShorelineRegulationsandPermittingProcessesBCReport.pdf.177Borsje, B. W., de Vries, S., Janssen, S., and Living, A. L. (2017). Building with nature ascoastal protection strategy in the Netherlands. In Bilkovic, D. M., Mitchell, M. M., La Peyre,M. K., and Toft, J. D., editors, Living Shorelines: The science and management of nature-basedcoastal protection. CRC Press, New York.Borsje, B. W., van Wesenbeeck, B. K., Dekker, F., Paalvast, P., Bouma, T. J., van Katwijk,M. M., and de Vries, M. B. (2011). How ecological engineering can serve in coastal protection.Ecological Engineering, 37(2):113–122.Bouma, T. J., Olenin, S., Reise, K., and Ysebaert, T. (2009). Ecosystem engineering andbiodiversity in coastal sediments: Posing hypotheses. Helgoland Marine Research, 63(1):95–106.Bradley, K. and Houser, C. (2009). Relative velocity of seagrass blades: Implications forwave attenuation in low-energy environments. Journal of Geophysical Research-Earth Sur-face, 114(F01004):1–13.Bridges, T. S., Burks-Copes, K. A., Bates, M. E., Collier, Z. A., Fischenich, J. C., Piercy,C. D., Russo, E. J., Shafer, D. J., Suedel, B. C., Gailani, J. Z., and others (2015). Use ofNatural and Nature-Based Features (NNBF) for Coastal Resilience, volume 15 of ERDC SR.U.S. Army Engineer Research and Development Center, Environmental Laboratory, Coastaland Hydraulics Laboratory.Brody, S. D., Kang, J. E., and Bernhardt, S. (2010). Identifying factors influencing floodmitigation at the local level in Texas and Florida: The role of organizational capacity. Naturalhazards, 52(1):167–184.Bronen, R. (2015). Climate-induced community relocations: Using integrated social-ecological assessments to foster adaptation and resilience. Ecology and Society, 20(3):36.Brown, J. M., Phelps, J. J. C., Barkwith, A., Hurst, M. D., Ellis, M. A., and Plater, A. J. (2016).The effectiveness of beach mega-nourishment, assessed over three management epochs. Jour-nal of environmental management, 184:400–408.Bryant, D. B., Bryant, M. A., and Grzegorzewski, A. S. (2017). Erosion of Coastal Foredunes:A Review on the Effect of Dune Vegetation. U.S. Army Engineer Research and DevelopmentCenter, Environmental Laboratory, Coastal and Hydraulics Laboratory, ERDC/CHL CHETN-178I-94,. Accessed on May 2018. Retrieved from:, S. (2010). Transforming barriers into enablers of action on climate change: Insightsfrom three municipal case studies in British Columbia, Canada. Global Environmental Change,20(2):287–297.Burch, S., Sheppard, S. R. J., Shaw, A., and Flanders, D. (2010). Planning for climate changein a flood-prone community: Municipal barriers to policy action and the use of visualizationsas decision-support tools. Journal of Flood Risk Management, 3(2):126–139.Burcharth, H. F., Hawkins, S. J., Zanuttigh, B., and Lamberti, A. (2007). Chapter 3 - objectivesand target effects of lcss. In Burcharth, H. F., Hawkins, S. J., Zanuttigh, B., and Lamberti, A.,editors, Environmental Design Guidelines for Low Crested Coastal Structures, pages 11 – 14.Elsevier Science Ltd, Oxford.Burton, C. G. (2014). A Validation of Metrics for Community Resilience to Natural Hazardsand Disasters Using the Recovery from Hurricane Katrina as a Case Study. Annals of theAssociation of American Geographers, 105(1):67–86.Byrne, J. A., Lo, A. Y., and Jianjun, Y. (2015). Residents’ understanding of the role of greeninfrastructure for climate change adaptation in Hangzhou, China. Landscape and Urban Plan-ning, 138:132–143.Canada Marine Act of 1998 (1998). S.C. 1998, c. 10.Canada National Marine Conservation Areas Act of 2002 (2002). S.C. 2002, c. 18.Canada National Parks Act of 2000 (2000). S.C. 2000, c. 32.Canada Shipping Act of 2001 (2001). S.C. 2001, c. 26.Canada Wildlife Act of 1985 (1985). R.S.C., 1985, c. W-9.Canadian Environmental Assessment Act of 2012 (2012). S.C. 2012, c. 19, s. 52.Canadian Environmental Protection Act of 1999 (1999). S.C. 1999, c. 33.CAP and ICLEI (2015). Are We There Yet? Applying Sustainabil-ity Indicators to Measure Progress on Adaptation. ICLEI—Local Govern-179ments for Sustainability and the Clean Air Partnership. Accessed on April2018. Retrieved from:, G., Ortı´, M. M. A., Copiz, R., Fusaro, L., Mollo, B., Salvatori, E., and Zavattero,L. (2017). Biodiversity and ecosystem services in urban green infrastructure planning: A casestudy from the metropolitan area of Rome (Italy). Urban Forestry & Urban Greening.Carlsson-Kanyama, A., Carlsen, H., and Dreborg, K.-H. (2013). Barriers in municipal climatechange adaptation: Results from case studies using backcasting. Futures, 49:9–21.Carson, M., Ko¨hl, A., Stammer, D., A Slangen, A. B., Katsman, C. A., W van de Wal, R. S.,Church, J., and White, N. (2016). Coastal sea level changes, observed and projected during the20th and 21st century. Climatic Change, 134(1-2):269–281.Catenacci, M. and Giupponi, C. (2013). Integrated assessment of sea-level rise adaptationstrategies using a Bayesian decision network approach. Environmental Modelling & Software,44:87–100.Cazenave, A., Dieng, H.-B., Meyssignac, B., von Schuckmann, K., Decharme, B., andBerthier, E. (2014). The rate of sea-level rise. Nature Climate Change, 4(5):358–361.Chaffin, B. C., Shuster, W. D., Garmestani, A. S., Furio, B., Albro, S. L., Gardiner, M., Spring,M., and Green, O. O. (2016). A tale of two rain gardens: Barriers and bridges to adaptivemanagement of urban stormwater in Cleveland, Ohio. Journal of Environmental Management,183(Pt 2):431–441.Chang, S. E., Adams, B. J., Alder, J., Berke, P. R., Chuenpagdee, R., Ghosh, S., and Wabnitz,C. (2006). Coastal ecosystems and tsunami protection after the December 2004 Indian Oceantsunami. Earthquake Spectra, 22:863–887.Chang, S. E., Yip, J. Z. K., Conger, T., Oulahen, G., and Marteleira, M. (2018). Communityvulnerability to coastal hazards: Developing a typology for disaster risk reduction. AppliedGeography, 91:81–88.Chang, S. E., Yip, J. Z. K., van Zijll de Jong, S. L., Chaster, R., and Lowcock, A. (2015).Using vulnerability indicators to develop resilience networks: A similarity approach. Naturalhazards, 78(3):1827–1841.180Chenoweth, J., Anderson, A. R., Kumar, P., Hunt, W. F., Chimbwandira, S. J., and Moore, T.L. C. (2018). The interrelationship of green infrastructure and natural capital. Land Use Policy,75:137–144.Cheong, S.-M., Silliman, B., Wong, P. P., van Wesenbeeck, B., Kim, C. K., and Guannel, G.(2013). Coastal adaptation with ecological engineering. Nature Climate Change, 3(9):787–791.Cinner, J. E., Adger, W. N., Allison, E. H., Barnes, M. L., Brown, K., Cohen, P. J., Gelcich,S., Hicks, C. C., Hughes, T. P., Lau, J., Marshall, N. A., and Morrison, T. H. (2018). Buildingadaptive capacity to climate change in tropical coastal communities. Nature Climate Change,8(2):117–123.Cities and Towns (1969). Title 35 RCW.Clean Air Act of 1970 (1970). 42 U.S.C. §7401 et seq.Clean Water of 1972 (1972). 33 U.S.C. §1251 et seq.Coastal Zone Management Act of 1972 (1972). 16 U.S.C. §§ 1451 et seq.Comer, P. J., Young, B., Schulz, K., Kittel, G., Unnasch, B., Braun, D., Hammerson, G.,Smart, L., Hamilton, H., Auer, S., Smyth, R., and Hak, J. (2012). Climate Change Vulnerabil-ity and Adaptation Strategies for Natural Communities: Piloting methods in the Mojave andSonoran deserts. Technical report, Report to the U.S. Fish and Wildlife Service. NatureServe,Arlington, VA. Accessed on July 2018. Retrieved from: HCCVI Report.pdf.Community Charter (2003). [SBC 2003] Chapter 26.Cook, J., Frankel-Reed, J., Bruzgul, J., Helmuth, M., Hurley, B., Mack, C.,Schultz, P., Strange, E., and Anderson, G. (2016). Climate Vulnerability As-sessment: An Annex to the USAID Climate-Resilient Development Framework.United States Agency for International Development (USAID), Washington, DC.Accessed on January 2018. Retrieved from:, J. A. G. (2016). Adaptation or Resistance: A classification of responses to sea-levelrise. American Geophysical Union, Ocean Sciences Meeting 2016, abstract #EC31A-02. Ac-cessed on August 2018. Retrieved from:, J. A. G. and McKenna, J. (2008). Working with natural processes: The challenge forcoastal protection strategies. Geographical Journal, 174(4):315–331.Cooper, J. A. G. and Pile, J. (2014). The adaptation-resistance spectrum: A classificationof contemporary adaptation approaches to climate-related coastal change. Ocean & CoastalManagement, 94:90–98.Coops, H., Geilen, N., Verheij, H. J., Boeters, R., and van der Velde, G. (1996). Interactionsbetween waves, bank erosion and emergent vegetation: An experimental study in a wave tank.Aquatic Botany, 53(3-4):187–198.Costanza, R., Perez-Maqueo, O., Luisa Martinez, M., Sutton, P., Anderson, S. J., and Mulder,K. (2008). The value of coastal wetlands for hurricane protection. AMBIO: A Journal of theHuman Environment, 37(4):241–248.Counties (1969). Title 36 RCW.Cox, R., Jahn, K. L., Watkins, O. G., and Cox, P. (2018). Extraordinary boulder transport bystorm waves (west of Ireland, winter 2013–2014), and criteria for analysing coastal boulderdeposits. Earth-Science Reviews, 177:623–636.CPAWS (2016). Nature for Climate Change. Accessed on September 2018., C., Clough, J., Ehman, J., Joye, S., Park, R., Pennings, S., Guo, H., and Machmuller, M.(2009). Forecasting the effects of accelerated sea-level rise on tidal marsh ecosystem services.Frontiers in Ecology and the Environment, 7(2):73–78.Cutter, S. L. (2016). The landscape of disaster resilience indicators in the USA. Naturalhazards, 80(2):741–758.Cutter, S. L., Boruff, B. J., and Shirley, W. L. (2003). Social Vulnerability to EnvironmentalHazards. Social Science Quarterly, 84(2):20.182Dahdouh-Guebas, F., Jayatissa, L. P., Di Nitto, D., Bosire, J. O., Lo Seen, D., and Koedam,N. (2005). How effective were mangroves as a defence against the recent tsunami? CurrentBiology, 15(12):R443–R447.Davidson-Arnott, R. (2010). Introduction to Coastal Processes and Geomorphology. Cam-bridge University Press, Cambridge, UK ; New York.Day, J. C. and Gamble, D. B. (1990). Coastal zone management in British Columbia: Aninstitutional comparison with Washington, Oregon, and California. Coastal Management,18(2):115–141.Demarchi, D. (2011). An introduction to the ecoregions of British Columbia. Ecosys-tem Information Section Ministry of Environment, Victoria, British Columbia. Ac-cessed on June 2018. Retrieved from: introduction to theecoregions of british columbia.pdf.Department of Fisheries and Oceans Act of 1985 (1985). R.S.C., 1985, c. F-15.Devoy, R. J. N. (2015). Chapter 8 - Sea-Level Rise: Causes, Impacts, and Scenariosfor Change. In Shroder, J. F., Ellis, J. T., and Sherman, D. J., editors, Coastal and MarineHazards, Risks, and Disasters, pages 197–241. Elsevier, Boston.Dewing, M., Young, W. R., and Tolley, E. (2006). Municipalities, the Constitution, and theCanadian Federal System. Parliamentary Information and Research Service. Accessed on April2018. Retrieved from:, T. and Burton, I. (2011). Adaptation to Climate Change in Canada: A Multi-level Mosaic. In Climate Change Adaptation in Developed Nations, pages 103–117. Springer,Dordrecht, Dordrecht.Diking and Drainage Act of 1985 (1985). Title 85 RCW.Dolan, A. H. and Walker, I. J. (2006). Understanding vulnerability of coastal communities toclimate change related risks. Journal of Coastal Research, III(39):1316–1323.Doody, J. P. (2004). ’Coastal Squeeze’: An Historical Perspective. Journal of Coastal Con-servation, 10(1/2):129–138.183Doody, J. P. (2013). Coastal squeeze and managed realignment in southeast England, does ittell us anything about the future? Ocean & Coastal Management, 79:34–41.Duarte, C. M., Losada, I. J., Hendriks, I. E., Mazarrasa, I., and Marba`, N. (2013). The roleof coastal plant communities for climate change mitigation and adaptation. Nature ClimateChange, 3(11):961–968.Duffy, R., Royer, G., and Beresford, C. (2014). Who’s Picking Up the Tab?: Federaland Provincial Downloading Onto Local Governments. Columbia Institute and Centre forCivic Governance. Accessed on April 2018. Retrieved from: Research Group (2013). What Will Adaptation Cost? An Economic Frameworkfor Coastal Community Infrastructure. National Oceanic and Atmospheric Administration(NOAA) Coastal Services Center. Accessed on September 2018. Retrieved from: (2015). Adaptation options - Floating or elevated roads. Climate - ADAPT - Eu-ropean Climate Adaptation Platform. Accessed on September 2018. Retrieved from:, P. E. T., Sutton-Grier, A. E., and Coyle, G. E. (2013). Investing in nature: Restoringcoastal habitat blue infrastructure and green job creation. Marine Policy, 38:65–71.Eisenack, K., Moser, S. C., Hoffmann, E., Klein, R., Oberlack, C., Pechan, A., Rotter, M., andTermeer, C. J. A. M. (2014). Explaining and overcoming barriers to climate change adaptation.Nature Climate Change, 4(10).Ekwurzel, B., Boneham, J., Dalton, M. W., Heede, R., Mera, R. J., Allen, M. R., and Frumhoff,P. C. (2017). The rise in global atmospheric CO2, surface temperature, and sea level fromemissions traced to major carbon producers. Climatic Change, 144(4):579–590.Endangered Species Act of 1973 (1973). 16 U.S.C. §1531 et seq.Environment and Climate Change Canada (2018). National Wetland Conservation Fund. Ac-cessed on September 2018. Assessment Office (2018). Legislation.European Commission (2013). Green Infrastructure (GI) — Enhancing Europe’s Natural Cap-ital. COM (2013) 249 Final. Accessed on May 2018. Retrieved from:, S. (2009). The costs of adaptation. Climate Change, 1(1):23–30.Farber, D. A. (2009). Climate Adaptation and Federalism: Mapping the Issues. 1 San DiegoJ. Climate & Energy L.259.Feagin, R. A., Figlus, J., Zinnert, J. C., Sigren, J., Martı´nez, M. L., Silva, R., Smith, W. K.,Cox, D., Young, D. R., and Carter, G. (2015). Going with the flow or against the grain? Thepromise of vegetation for protecting beaches, dunes, and barrier islands from erosion. Frontiersin Ecology and the Environment, 13(4):203–210.Feagin, R. A., Lozada-Bernard, S. M., Ravens, T. M., Mo¨ller, I., Yeager, K. M., and Baird,A. H. (2009). Does vegetation prevent wave erosion of salt marsh edges? Proceedings of theNational Academy of Sciences, 106(25):10109–10113.Feagin, R. A., Sherman, D. J., and Grant, W. E. (2005). Coastal Erosion, Global Sea-LevelRise, and the Loss of Sand Dune Plant Habitats. Frontiers in Ecology and the Environment,3(7):359–364.Feagin, R. A., Smith, W. K., Psuty, N. P., Young, D. R., Martinez, M. L., Carter, G. A.,Lucas, K. L., Gibeaut, J. C., Gemma, J. N., and Koske, R. E. (2010). Barrier islands: Couplinganthropogenic stability with ecological sustainability. Journal of Coastal Research, 26(6):987–992.Felsenstein, D. and Lichter, M. (2014). Social and economic vulnerability of coastal commu-nities to sea-level rise and extreme flooding. Natural hazards, 71(1):463–491.FEMA (2007). Selecting Appropriate Mitigation Measures for Floodprone Structures. FederalEmergency Management Agency (FEMA). Accessed on May 2018. Retrieved from: 551.pdf.Field, M. A. (1992). The differing federalisms of Canada and the United States. Law andContemporary Problems, 55(1):107–120.185Figlus, J., Sigren, J. M., Armitage, A. R., and Tyler, R. C. (2014). Erosion of vegetated coastaldunes. Coastal Engineering Proceedings, 1(34):20.Finlayson, D. P., Program, W. S. G., and Partnership, P. S. N. (2006). The Geomorphology ofPuget Sound Beaches. Washington Sea Grant Program, University of Washington. Accessedon September 2018. Retrieved from: papers/geomorphology.pdf.Fish and Wildlife Act of 1956 (1956). 16 U.S.C. §§ 742a et seq.Fisheries Act of 1985 (1985). R.S.C., 1985, c. F-14.Fisheries and Oceans Canada (2009). The Role of the Provincial and Territorial Governmentsin the Oceans Sector . Accessed on July 2018. Retrieved from: and Oceans Canada (2011). Defining Canada’s Maritime Zones. Accessedon July 2018. Retrieved from: 2012/mpo-dfo/Fs23-571-2011-eng.pdf.Fisheries and Oceans Canada (2018). Glossary - Tides, Currents, and Water Levels. Accessedon July 2018. Retrieved from:, D. M., Fenster, M. S., Argow, B. A., and Buynevich, I. V. (2008). Coastal ImpactsDue to Sea-Level Rise. Annual Review of Earth and Planetary Sciences, 36(1):601–647.Flanagan, B. E., Gregory, E. W., Hallisey, E. J., Heitgerd, J. L., and Lewis, B. (2011). A SocialVulnerability Index for Disaster Management. Journal of Homeland Security and EmergencyManagement, 8(1):Article 3.Flather, R. A. (2001). Storm Surges. In Encyclopedia of Ocean Sciences, pages 530–540.Elsevier.Flood Control Act of 1935 (1935). Title 86 RCW.Flynn, M., Ford, J. D., Pearce, T., and Harper, S. L. (2018). Participatory scenario planningand climate change impacts, adaptation and vulnerability research in the Arctic. EnvironmentalScience & Policy, 79:45–53.186Folke, C. (2006). Resilience: The emergence of a perspective for social–ecological systemsanalyses. Global Environmental Change, 16(3):253–267.Fonseca, M. S. and Cahalan, J. A. (1992). A preliminary evaluation of wave attenuation byfour species of seagrass. Estuarine, Coastal and Shelf Science, 35(6):565–576.Frazier, T. G., Thompson, C. M., Dezzani, R. J., and Butsick, D. (2013). Spatial and temporalquantification of resilience at the community scale. Applied Geography, 42:95–107.Frazier, T. G., Wood, N., and Yarnal, B. (2010). Stakeholder perspectives on land-use strategiesfor adapting to climate-change-enhanced coastal hazards: Sarasota, Florida. Applied Geogra-phy, 30(4):506–517.French, P. W. (2006). Managed realignment – The developing story of a comparatively newapproach to soft engineering. Estuarine, Coastal and Shelf Science, 67(3):409–423.Fu, X. and Song, J. (2017). Assessing the Economic Costs of Sea Level Rise and Benefits ofCoastal Protection: A Spatiotemporal Approach. Sustainability, 9(8):1495–14.Garner, G. and Keller, K. (2018). Using direct policy search to identify robust strategies inadapting to uncertain sea-level rise and storm surge. Environmental Modelling & Software,107:96–104.Gedan, K. B., Kirwan, M. L., Wolanski, E., Barbier, E. B., and Silliman, B. R. (2010). Thepresent and future role of coastal wetland vegetation in protecting shorelines: answering recentchallenges to the paradigm. Climatic Change, 106(1):7–29.Gersonius, B., van Buuren, A., Zethof, M., and Kelder, E. (2016). Resilient flood risk strate-gies: institutional preconditions for implementation. Ecology and Society, 21(4):28.Ghosh, S. and Chaudhuri, P. (2015). A Review of Threats and Vulnerabilities to MangroveHabitats: With Special Emphasis on East Coast of India. Journal of Earth Science & ClimaticChange, 06(04):1–10.Gibbs, M. T. (2015). Pitfalls in developing coastal climate adaptation responses. Climate RiskManagement, 8:1–8.187Giest, S. and Howlett, M. (2013). Comparative Climate Change Governance: Lessons fromEuropean Transnational Municipal Network Management Efforts. Environmental Policy andGovernance, 23(6):341–353.Gittman, R. K., Fodrie, F. J., Popowich, A. M., Keller, D. A., Bruno, J. F., Currin, C. A.,Peterson, C. H., and Piehler, M. F. (2015). Engineering away our natural defenses: an analysisof shoreline hardening in the us. Frontiers in Ecology and the Environment, 13(6):301–307.Glavovic, B. (2014). Chapter 3 - On the frontline in the Anthropocene: Adapting to climatechange through deliberative coastal governance . In Glavovic, B., Mick Kelly, P., Kay, R., andTravers, A., editors, Climate Change and the Coast: Building Resilient Communities, page596. CRC Press.Glavovic, B. C. and Smith, G. P. (2014). Adapting to Climate Change. Lessons from NaturalHazards Planning. Springer Science & Business.Gornitz, V. (1991). Global coastal hazards from future sea level rise. Palaeogeography, Palaeo-climatology, Palaeoecology, 89(4):379–398.Gornitz, V. M., Daniels, R. C., White, T. W., and Birdwell, K. R. (1994). The Development ofa Coastal Risk Assessment Database: Vulnerability to Sea-Level Rise in the U.S. Southeast.Journal of Coastal Research, 12(12):327–338.Gornitz, V. M., White, T. W., and Daniels, R. C. (1992). A coastal hazards data base for theUS East Coast. Technical report, Oak Ridge National Laboratory (ORNL), Oak Ridge, TN.Accessed on June 2018. Retrieved from: of Alberta (2015). Canadian and American Governance: A Com-parative Look. Government of Alberta. Accessed on September 2018. Re-trieved from:, S., Barnett, J., Fincher, R., Hurlimann, A., Mortreux, C., and Waters, E. (2013). Thesocial values at risk from sea-level rise. Environmental Impact Assessment Review, 41:45–52.188Gray, E. C. (2017). Making the Case for Coastal Buyouts in Canada. Master’s thesis, Schoolof Community and Regional Planning, Faculty of Applied Science. University of BritishColumbia, British Columbia, Canada.Gregg, R., Kershner, J., and Hansen, L. (2015). Strategies for climate change adaptation: Asynthesis. In Reference Module in Earth Systems and Environmental Sciences. Elsevier.Gutie´rrez, J., Jones, C., Byers, J., Arkema, K., Berkenbusch, K., Commito, J., Duarte, C.,Hacker, S., Lambrinos, J., Hendriks, I., Hogarth, P., Palomo, M., and Wild, C. (2011).7.04 physical ecosystem engineers and the functioning of estuaries and coasts. In Trea-tise on Estuarine and Coastal Science, pages 53–81. Academic Press. Exported from on 2018/09/14.Haer, T., Botzen, W., Zavala-Hidalgo, J., and Cusell, C. (2017). Economic evaluation ofclimate risk adaptation strategies: Cost-benefit analysis of flood protection in Tabasco, Mexico.Atmo´sfera, 30(2):101–210.Hagen, S. C., Passeri, D. L., Bilskie, M. V., DeLorme, D. E., and Yoskowitz, D. (2018). Sys-tems Approaches for Coastal Hazard Assessment and Resilience, volume 1 of Systems Ap-proaches for Coastal Hazard Assessment and Resilience. Oxford University Press.Halbert, C. L. (2008). How adaptive is adaptive management? Implementing adaptive manage-ment in Washington State and British Columbia. Reviews in Fisheries Science, 1(3):261–283.Hallegatte, S., Green, C., Nicholls, R. J., and Corfee-Morlot, J. (2013). Future flood losses inmajor coastal cities. Nature Climate Change, 3(9):802–806.Hamilton, D. K. (2013). Measuring the Effectiveness of Regional Governing Systems: A Com-parative Study of City Regions in North America. Springer Science + Business Media.Hamin, E. and Gurran, N. (2015). Climbing the Adaptation Pl