LAKE MICHIGAN STAKEHOLDERS’ PERCEPTIONS OF COASTAL RISK AND MOTIVATIONS FOR COASTAL HABITAT STEWARDSHIP By Julia H. Whyte Michigan State University in partial fulfillment of the requirements for the degree of Fisheries and Wildlife—Master of Science A THESIS Submitted to 2019 ABSTRACT LAKE MICHIGAN STAKEHOLDERS’ PERCEPTIONS OF COASTAL RISK AND MOTIVATIONS FOR COASTAL HABITAT STEWARDSHIP By Julia H. Whyte Lake Michigan communities have already begun to feel the effects of climate change, and research suggests that these areas will experience many phenomena that will negatively impact the ecosystem and human livelihoods (GLISA, 2014). While agencies exist to generally guide coastal management, Michigan lacks institutions that establish regulations or requirements for managing the Great Lakes coastal region (Norton et al., 2018). As a result, Michigan’s coastal communities have the responsibility of preparing for an uncertain future under climate change. I compared risk perceptions between different resident groups, as well as between different communities, varying by county, size, and presence of a previous coastal resiliency program. I used a four-wave tailored design for data collection (Dillman, 2009) in six Michigan communities along Lake Michigan from December 2018 to April 2019. I found communities with resiliency programs are less concerned about coastal risk than other communities and lake residents are more concerned about coastal risk than municipal officials. I also found that previous experience with environmental risk and gender are predictors of concern about coastal risk. I suggest that future outreach materials focus on lake residents and that community-engaged work to create more robust coastal resilience plans are beneficial to mitigating risk perceptions. The results from this research can also be used to inform future planning and zoning policies, as well as other coastal resilience policies. Copyright by JULIA H. WHYTE 2019 ACKNOWLEDGEMENTS I would like to thank everyone who offered me guidance and support throughout my master’s program at Michigan State University. First, I would like to thank my major professor, Dr. Heather Triezenberg, for her help to develop my thesis project and all the other opportunities she gave me that have helped me grow as a researcher and professional. I would also like to thank my graduate committee members, Dr. Meredith Gore and Dr. Alan Arbogast, for their time, guidance, and support throughout the development and implementation of my research. I am also thankful for all members of the Coastal Resiliency Team who were extremely helpful throughout all stages of my thesis project. I also thank the municipal officials and staff, residents, and County GIS employees of Allegan, Ottawa, and Muskegon Counties for their participation and for providing me with data necessary to complete my research. Thank you to the Michigan Sea Grant and Department of Fisheries and Wildlife staff for their support, especially Vanessa Pollok and Jill Cruth who helped me accomplish many tasks. I also owe thanks to the Department of Fisheries and Wildlife and the Department of Community Sustainability faculty for their invaluable teachings and support. I also thank MSU CSTAT for their statistics consultation services and Foresight Group for their survey invitation services. A huge thank you to all my fellow graduate students who supported me with friendship, laughter, and guidance throughout my masters – I couldn’t have done it without you! Finally, I will be forever grateful for the unwavering love and encouragement from my parents who have always been there for me and have always supported me. Thank you so much! v In addition, thank you to the College of Agriculture and Natural Resources for granting me the Graduate Academic Achievement Award that financially supported the first year of my master’s. Financial assistance for this project was provided, in part, by the Coastal Management Program, Water Resources Division, Michigan Department of Environment, Great Lakes, and Energy, under the National Coastal Zone Management Program, through a grant from the National Oceanic and Atmospheric Administration, U.S. Department of Commerce. Financial assistance provided by the Coastal Zone Management Act of 1972. The statements, findings, conclusions, and recommendation in this thesis are those of the author, Julia Whyte, and do not necessarily reflect the view of the Michigan Department of Environment, Great Lakes, and Energy and the National Oceanic and Atmospheric Administration. vi TABLE OF CONTENTS LIST OF TABLES ..................................................................................................................... viii LIST OF FIGURES ....................................................................................................................... x CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW .............................................. 1 Physical Dynamics of the Great Lakes ............................................................................................ 2 Challenges to Coastal Resilience ....................................................................................................... 3 Mitigation vs. Adaptation ................................................................................................................... 5 Risk Perception ..................................................................................................................................... 6 Values-Beliefs-Norms Theory ............................................................................................................ 8 Risk Information Seeking and Processing Framework ............................................................... 9 Research Questions ............................................................................................................................ 15 CHAPTER 2: METHODS ........................................................................................................... 16 Study Sample ....................................................................................................................................... 16 Survey Design ...................................................................................................................................... 19 Analysis ................................................................................................................................................. 24 CHAPTER 3: PERCEPTIONS OF COASTAL RISK ................................................................ 29 Results ................................................................................................................................................... 29 Discussion ............................................................................................................................................. 39 CHAPTER 4: MOTIVATIONS FOR COASTAL HABITAT STEWARDSHIP ....................... 42 Results ................................................................................................................................................... 42 Discussion ............................................................................................................................................. 49 CHAPTER 5: CONCLUSIONS .................................................................................................. 52 Limitations ........................................................................................................................................... 52 Recommendations ............................................................................................................................... 53 APPENDICES .............................................................................................................................. 54 APPENDIX A: INSTITUTIONAL REVIEW BOARD APPROVAL LETTER ....................... 55 APPENDIX B: SURVEY INVITATION NUMBERS AND RESPONSE RATES ................. 56 APPENDIX C: QUALTRICS SURVEY ......................................................................................... 57 APPENDIX D: CODEBOOK ........................................................................................................... 75 APPENDIX E: SOCIO-DEMOGRAPHIC SURVEY AND CENSUS DATA ......................... 91 APPENDIX F: ALLEGAN AND OTTAWA COUNTY RECRUITMENT MATERIALS .... 96 APPENDIX G: MUSKEGON COUNTY RESIDENT RECRUITMENT MATERIALS ..... 100 APPENDIX H: MUNICIPAL OFFICIALS EXAMPLE RECRUITMENT MATERIALS .. 104 REFERENCES .......................................................................................................................... 105 vii LIST OF TABLES Table 1. Community matrix displaying community, county, county size type, total population, previous coastal resilience policy or program, and shoreline type. ............................................... 17 Table 2. Constructs describing different facets of risk perception, the mean response, and standard deviation for the Combined Model. ................................................................................ 20 Table 3. Number and frequency of respondents who said they were involved in a program or organization whose focus is on Great Lakes habitat preservation, conservation, or management. ....................................................................................................................................................... 21 Table 4. Socio-demographic variables hypothesized to predict an individual's risk perceptions. Descriptive statistics displaying mean (x̄ ) or frequency of respondents in each category with range or standard deviation (s) in parentheses. ............................................................................. 22 Table 5. Summary of goodness-of-fit statistics for baseline models and final structural models of Concern and Knowledge facets of risk perception. ....................................................................... 27 Table 6. Environmental concerns in 10 years. Mean responses from a 1 (not at all concerned) to 5 (extremely concerned) Likert-type-scale are displayed with standard deviations in parentheses. Highest average concerns are noted in bold. ................................................................................. 29 Table 7. Environmental concerns in 50 years. Mean responses from a 1 (not at all concerned) to 5 (extremely concerned) Likert-type-scale are displayed with standard deviations in parentheses. Highest average concerns are noted in bold. ................................................................................. 30 Table 8. Responses to the question "Who do you think owns coastal shoreline in your community?" broken down by Resident Type. ............................................................................. 31 Table 9. Responses to the question "Who do you think is responsible for managing coastal shoreline in your community?" broken down by Resident Type. ................................................. 32 Table 10. What is the best way to manage a receding shoreline? Frequency of response by Resident Type (Lake, Near-lake, Inland, and Municipal officials). .............................................. 34 Table 11. Regression model of the final structural models for Concern (Model 1 and 2) and Knowledge (Model 3 and 4). Model 1 and 3 look at characteristics of the communities and counties, Model 2 and 4 include socio-demographic factors. ....................................................... 36 Table 12. Regression models of the final structural models for three counties (Allegan County, Ottawa County, and Muskegon County) of Concern and Knowledge. ......................................... 38 Table 13. Logistic regression predicting involvement in program for Combined Model. Odds ratios are presented with standard errors in parentheses. .............................................................. 43 Table 14. Logistic regression predicting involvement in a program or organization by Concern and Knowledge. Odds ratios are presented with standard errors in parentheses. .......................... 44 viii Table 15. Logistic regression by county predicting involvement in a program or organization. Model 1 includes logistic regression controlling for resident locations. Model 2 includes Model 1 and socio-demographic predictors. Odds ratios are presented with standard errors in parentheses. ....................................................................................................................................................... 46 Table 16. Logistic regression by county predicting involvement in a program or organization for Concern and Knowledge factors. Model 1 includes logistic regression controlling for resident locations. Model 2 includes Model 1 and socio-demographic predictors. Odds ratios are presented with standard errors in parentheses. .............................................................................. 47 Table 17. Range of motivations for stewardship concepts determined by emergent coding. ...... 49 Table 18. Survey invitation numbers and response rates for Pooled Counties. ............................ 56 Table 19. Survey invitation numbers and response rates for Allegan County. ............................ 56 Table 20. Survey invitation numbers and response rates for Ottawa County. .............................. 56 Table 21. Survey invitation numbers and response rates for Muskegon County. ........................ 56 Table 22. Codebook. ..................................................................................................................... 75 Table 23. Socio-demographic survey data compared to census data for gender. ......................... 91 Table 24. Socio-demographic survey data compared to census data for gender by county. ........ 91 Table 25. Socio-demographic survey data compared to census data for age. .............................. 92 Table 26. Socio-demographic survey data compared to census data for age by county. ............. 92 Table 27. Socio-demographic survey data compared to census data for highest level of education. ....................................................................................................................................... 93 Table 28. Socio-demographic survey data compared to census data for highest level of education by county. ...................................................................................................................................... 93 Table 29. Socio-demographic survey data compared to census data for race. ............................. 94 Table 30. Socio-demographic survey data compared to census data for race by county. ............ 94 Table 31. Socio-demographic survey data compared to census data for annual income before taxes. .............................................................................................................................................. 95 Table 32. Socio-demographic survey data compared to census data for annual income before taxes by county. ............................................................................................................................. 95 ix LIST OF FIGURES Figure 1. Map of shoreline recession risk in Michigan (from Section 309 Assessment and Five- Year Strategy for Coastal Zone management Enhancement Fiscal Years 2016-2020). ................. 4 Figure 2. Risk Information Seeking and Processing model, adapted from Yang et al. 2014 and Griffin et al. 1999. Dotted arrows represent theorized causal relationship that were not supported by Yang et al. (2014) who applied the RISP model to climate change mitigation policy. ........... 10 Figure 3. The Risk Seeking Information and Processing model as a precursor to the Theory of Planned Behavior, adapted from Griffin et al. 1999. Variables in the blue box are part of the Theory of Planned Behavior, variables in the grey box are part of the RISP framework. ............ 12 Figure 4. Conceptual mode of "Concern" as a facet of risk perception. ...................................... 13 Figure 5. Conceptual model of self-reported “Knowledge" as a facet of risk perception. ........... 14 Figure 6. Conceptual model of "Concern" as a facet of risk perception with factor loadings. .... 25 Figure 7. Conceptual model of self-reported "Knowledge" as a facet of risk perception with factor loadings. .............................................................................................................................. 26 Figure 8. Response of the participants to the question of who owns coastal shoreline in their community (n=907). ...................................................................................................................... 31 Figure 9. Response of the participants to the question of who is responsible for coastal shoreline in their community (n=919). .......................................................................................................... 32 Figure 10. Response of the participants to the question of what is the best way to manage a receding shoreline (n=844). ........................................................................................................... 33 x CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW The Great Lakes coastal region offers a rich and diverse habitat for millions of people in North America. From 2000 to 2010 there was a 5.94% increase in migration to Michigan’s coastal floodplain (Crossett, Ache, Pacheco, & Haber, 2013). The picturesque views and capacity to support a variety of livelihoods makes these areas highly sought-after real estate. However, Great Lakes coastal habitat is a very dynamic and constantly changing system, making management strategies and regulation more complicated. Furthermore, climate models project that these fluctuations will continue in the future and may increase in intensity (GLISA, 2014; 2017). It is estimated that the Great Lakes region will also experience increases in extreme storms, increases in precipitation, more precipitation as rain than snow, reduced ice cover on the Great Lakes, more flooding events with the risk of erosion, and increases in extreme temperatures. Although agencies and programs exist to generally guide coastal management in Michigan, these institutions are not officially considered active regulatory bodies (Norton, David, Buckman, & Koman, 2018a). As a result, Michigan’s coastal communities have the responsibility of preparing for an uncertain future, especially under a climate-change scenario. Michigan’s local government structure also makes planning more challenging because its townships, which are within county bounds, are regulated by the adjacent cities or villages. In Michigan, 41 of its 83 counties are coastal and 1241 civil townships (67 called “townships). Therefore, there is certainly a need to understand community members’ perceptions of coastal risk because they are ultimately the actors who make coastal management decisions (Norton et al., 2018). If we can understand community members’ perceptions of risk and motivations for stewardship, we can create a foundation of evidence-based best practices to assist community 1 decision-making, particularly in communities lacking coastal resilience policies or educational or other capacity training deficits. This research seeks to help fill the existing knowledge gap surrounding stakeholders’ perceptions of risk and the implications for coastal resilience policy. Physical Dynamics of the Great Lakes The Great Lakes over a region of almost 95,000 mi2 and hold about 21% of the world’s freshwater (EPA, 2018). They are also relatively young, formed about 10,000 years ago at the end of the most recent ice age. The Great Lakes have a highly variable shoreline, consisting of flat coast, dunes, bluffs, and coastal wetlands, all which offer a diverse ecosystem. Compared to most other Michigan landscapes, coastal sand dunes are especially dynamic systems that constantly change over the long and short term (Lovis, Arbogast, & Monaghan, 2012). The dunes, which in many places are great than 30-m high, began to form during the Nipissing high stand (~5500 cal. years BP) and have grown episodically since that time (Arbogast, Hansen, & Van Oort, 2002). The eastern shore of Lake Michigan likely holds the largest number of freshwater dunes in the world (Peterson & Dersch, 1981).This geographical feature of Michigan makes living on the coastal zones of Lake Michigan very attractive. However, for these sand dunes to continue to form and grow, natural sand movement by wind must be maintained. In general, the average lake levels also fluctuate over the years depending on the amount of water entering the lake basin, compared to the amount of water leaving through natural processes and human intervention (Gronewold et al., 2013). Increased temperatures due to climate change are projected to cause higher rates of evaporation and less ice cover on the lakes during the winter (Bai, Wang, Sellinger, Clites, & Assel, 2012; Mason et al., 2016). Fluctuating lake levels greatly impact the shoreline, causing erosion of the coastline. Lake levels partially control the Great Lakes coastal dynamics, which can negatively impact communities primarily through soil erosion and flooding. Although the lake levels are 2 currently relatively high (2018), there are also long periods of time when lake levels are lower, allowing for the formation of more dunes and beaches. Lake Michigan residents often seem to forget about higher water levels and develop in areas that are at a higher risk for flooding and coastal erosion because of the beautiful sandy beaches (Norton, David, Buckman, & Koman, 2018b). These natural fluctuations in lake level alone are reason to encourage communities to engage in coastal resiliency management practices. However, climate models project that these fluctuations will continue in the future and are likely to increase in intensity (Byun & Hamlet, 2018; Hayhoe, VanDorn, Croley, Schlegal, & Wuebbles, 2010; Pyror et al., 2014). Rising average temperatures cause less ice coverage on the Great Lakes, which leads to higher rates of evaporation. Climate change is also predicted to cause more extreme weather events which pose a risk for coastal communities. Leveraging social science insight to aid coastal communities in adapting to the diverse, dynamic systems in which they live can help better prepare them for a future of resilience under extreme climate change scenarios. Challenges to Coastal Resilience Although federal and state agencies and programs, such as the Coastal Zone Management Program established by Coastal Zone Management Act of 1972, exist to generally guide coastal management in Michigan, these institutions do not act as active regulatory bodies (Norton et al., 2018b). As a result, Michigan’s coastal communities have the jurisdictional responsibility of preparing for an uncertain future under climate change. The few legal regulations that exist include the High Risk Erosion Area program (HREA), the Ordinary High Water Mark (OHWM), and the Public Trust Doctrine (Norton, Meadows, & Meadows, 2011a). The HREA designates coastal areas as “high risk” that are receding more than a foot per year, over a minimum of 15 years. These areas have an additional permitting process and regulatory standards enforced by the Michigan Department of Environment, Great Lakes, and Energy (EGLE, formally (MDEQ). 3 Currently 6.1% of Michigan’s coastline falls under the “high risk” designation. In Michigan, the coast of Lake Michigan has the highest risk of coastal recession compared to any other Great Lakes shorelines (Figure 1). Figure 1. Map of shoreline recession risk in Michigan (from Section 309 Assessment and Five-Year Strategy for Coastal Zone management Enhancement Fiscal Years 2016-2020). The OHWM, updated in 1985 to 581.5’ elevation on Lake Michigan by the U.S. Army Corps of Engineers, dictates the elevation at which a permit is required for construction, such as dredging, seawalls, revetments, and other structures. Lastly, the Public Trust Doctrine, upheld by the Michigan Supreme Court in Glass vs. Goeckel, ensures that the state of Michigan is a trustee 4 for the public and protects the land from the water’s edge to the OHWM for public use. Regulating the public trust area under the OHWM and Public Trust Doctrine is complicated because of the natural fluctuations in lake levels and continuous shoreline erosion (Norton, Meadows, & Meadows, 2011b). Mitigation vs. Adaptation There are currently two options decision-makers can choose from when determining the best course of action to reduce impact from environmental issues, such as coastal risk: mitigation or adaptation. Mitigation refers to creating actions of plans, often involving technological changes, that actively decrease harmful activities or processes (Portman, 2016). However, these coastal dynamic trends are arguably inevitable, regardless of climate change, which negates the benefits of mitigation policy (Arbogast et al., 2002). Furthermore, while people may agree it is necessary to change one’s lifestyle, few are actually willing to engage in actions required by mitigation strategies (Leiserowitz, 2005; Lorenzoni & Pidgeon, 2006). On the other hand, adaptation is focused on “actions taken to prepare for and adjust to new conditions to reduce harm and take advantage of new opportunities” (Portman 2016). In general, adaptation strategies contribute to a community’s overall resilience to environmental issues. Because it is difficult to predict future coastal dynamic trends, building up a community’s resilience to better deal with unexpected events is critical (Tompkins, Adger, & Adger, 2004). While agencies, such as the Coastal Zone Management Program (CZMP), or the Michigan Hazard Mitigation Plan (MHMP), or programs, such as the high-risk erosion program with CZMP, exist to generally guide management, neither of these institutions establish regulations or requirements for managing the Great Lakes coastal region (Norton et al., 2018b). Therefore, the responsibility for preparing for an uncertain future falls onto 318 townships, cities, and villages along Lake Michigan (Norton et al., 2018a). Community leaders are only beginning 5 to use adaptive approaches deal with the projected impacts climate change (Stults & Larsen, 2018). Communities need robust information and effective resiliency techniques if they hope to achieve sustainability goals. There is certainly a need to understand community members’ perceptions of coastal risk because they are ultimately the actors who make resilient policy decisions (Norton et al., 2018b). If we can understand community members’ perceptions of risks and motivations for stewardship, we can determine the best approaches to assist communities with educational or capacity and training needs that are lacking in coastal resiliency policy (Feltman, Norris, & Batanian, 2017). Therefore, it is important to understand stakeholders’ perceptions of risk and motivations for stewardship of these coastal habitats, and the implications for coastal resiliency policy. This research is part of a community-engaged scholarship effort conducted in partnership with the Michigan Coastal Resiliency Team (CRT). Together with the Michigan Department of Natural Resources (MDNR), Office of the Great Lakes (OGL), Coastal Zone Management Program (CZMP), Michigan Association of Planning (MAP), University of Michigan (UM), Michigan Technological University, Michigan Environmental Council, Michigan Sea Grant (MSG), and Michigan State University Extension, the CRT aims at providing coastal communities with information and methods to improve resiliency. Risk Perception There are two leading theories used to explain risk perception, the Cultural Theory of risk and the psychometric model (Sjoberg, 2000). The Cultural Theory, originally proposed by (Douglas & Wildavsky, 1983), suggests that risk perception is heavily influenced by the social context in which individuals find themselves. The theory describes four types of people (egalitarian, individualistic, hierarchic, and fatalistic) who differ based on their concern for types of hazards (Dake, 1991). Egalitarians are concerned with technology and the environment, 6 individualists focus on war and the economy, hierarchists care most about law and order, while fatalists care about none of the previously mentioned topics (Sjoberg, 2000). While extremely influential, Sjoberg (2000) notes that this theory is generally lacking in empirical evidence and the scales and concern measurements only explained about 5-10% of the variance. On the other hand, the psychometric model argues that risk perceptions are based on the individual factors of the risks themselves. This model focuses on three exploratory factors to study risk: dread, “new-old”, and number of exposed individuals (Fischhoff, Slovic, Lichtenstein, Read, & Combs, 1978). Fischhoff et al. (1978) studied women voters and their spouses, and found that activities associated with the higher levels of dread and negative consequences were in most need of risk reduction. “New-old” refers to the finding that higher levels of risk are more tolerated for “old, voluntary activities with well-known and immediate consequences.” Fischhoff et al. (1978) also suggest that higher level of risk perception is positively related to the number of people who are potentially at risk. Sjoberg & Drottz-Sjoberg (1993) also argue that there is a fourth factor, morality, that is missing from risk perception research. However, morality can be thought of as a factor that influences and individual’s risk perception instead of a characteristic of the risk itself (Slimak & Dietz, 2006). The psychometric model also includes factors such as perceived behavioral control and knowledge about a risk, as well as attitudes towards risk management. The quantitative nature of this model allows for comparisons between different risks as well as different participant groups. Although the psychometric model was originally used in markets and cost-benefit analysis, many studies have used it to explain risk perceptions of natural hazards. Applications in flood risk research suggest that exposure to the risk and cultural differences influence flood risk perception (Kellens, Terpstra, & De Maeyer, 2013). Specifically, individuals who have been previously exposed to flooding seem to be less willing to adopt mitigation policies than the 7 general public (Ho, Shaw, Lin, & Chiu, 2008; Lin, Shaw, & Ho, 2008). In their study comparing different types of risk, Ho et al. (2008) found that while perceived control over landslides mitigated the risk perception, the opposite was true for flood victims. The authors suggest this is due to the financial cost that is more difficult to mitigate, while landslides usually result in human death which is easier to avoid. Although the psychometric model has its limitations (Sjoberg, 2000), I use it to explore the characteristics of the risks themselves, rather than the social contexts that may form risk perceptions. My research is also focused on perceived risk rather than assessed risk, as it is an individual’s risk perception that actually influences their behavior (Slovic, 1987). (Kettle & Dow, 2014) point out that the “nature and magnitude of the perceived risk” influences individual’s actions to reduce risk. Understanding community members’ perceptions of risk is important to mitigate environmental hazards because people act on their perceptions, rather than the objective, quantifiable risk itself. Values-Beliefs-Norms Theory Resilience is often thought of in the context of social-ecological systems. Social- ecological systems are based on the assumption that human behavior and social structures are directly related to natural systems and the two should not be thought of separate entities (Folke, 2006; Folke, Biggs, Norstrom, Reyers, & Rockstrom, 2016). In hazard research, resilience refers to the ability of a community to respond to and deal with environmental hazards, including “the capacity to reduce or avoid loses, contain effects of disasters, and recover with minimum social disruptions” (Cutter et al., 2008). Cutter et al. (2008) suggest six indicators to measure community resilience: ecological, social, economic, institutional, infrastructure, and community competence. 8 Enhancing a coastal community’s resilience capacity will help ensure the community will be sustained in the future, especially with the uncertainty of environmental change on the Great Lakes. Relating community members’ perceptions of coastal risk to resilience policy (as a proxy for behavior) will provide us with more insight for this area of study. The value-belief-norm theory (VBN) suggests that individual’s values and beliefs can help explain individual’s actions and behaviors (P. C. Stern, 2000; P. Stern et al., 1999; Stern, Paul C., Dietz, 1994). For example, those whose values align with pro-environmental values, form beliefs about the environment, and those beliefs lead to norms about taking action to reduce threats to the environment. For the purposes of this research, risk is measured by community member’s perceptions of risk characteristics. Risk perception can be thought of as the belief, while participating in resilience programs or policies can be thought of as the action or behavior. The VBN theory connects coastal risk perceptions to behaviors, such as creating policy that increases community resilience. Risk Information Seeking and Processing Framework The questionnaire was developed based on risk and resilience concepts, as well as the Risk Information Seeking and Processing (RISP) framework (Appendix B). The RISP framework, developed by Griffin et al. (1999), combines the Heuristic Systematic Model (HSM) and the Theory of Planned Behavior (TPB) to attempt to explain how people come to seek and attend information, and resulting risk-related behaviors. HSM focuses on three factors, information sufficiency, perceived information gathering capacity, and relevant channel beliefs that are driven by individual characteristics to explain information seeking and processing (Figure 2). 9 Figure 2. Risk Information Seeking and Processing model, adapted from Yang et al. 2014 and Griffin et al. 1999. Dotted arrows represent theorized causal relationship that were not supported by Yang et al. (2014) who applied the RISP model to climate change mitigation policy. Information gathering capacity is a measurement of how much an individual can seek and process information about risk. Research suggests that individuals with higher information gathering capacity are better able to understand and seek out information about risk, and therefore process risk systematically rather than heuristically (ter Huurne, Griffin, & Gutteling, 2009; Yang & Kahlor, 2013). Informational subjective norms are the perceived social normative influences that dictate how much an individual should know about an issue. In other words, those who perceive more social pressure to be knowledgeable about issues, such as coastal risk, will believe they need to know more about the issue, and are therefore more likely to process information systematically rather than heuristically. (Griffin, Dunwoody, & Neuwirth, 1999) suggest that one’s demographic characteristics and political party affiliation dictate one’s subjective norms about environmental issues. However, (Yang et al., 2014) found no causal relationship between perceived information gathering capacity and systematic processing. Individual characteristics are described as environmental values (Yang et al. 2014), relevant hazard experience, and demographics. In their reappraisal of the RISP model, Griffin et al. 10 (2012) suggest that gender, ethnicity, income, political party affiliation, and religion are the most important demographic factors. These individual characteristics impact “perceived hazard characteristics,” which then lead to an “affective response.” Griffin et al. (1999) list seven different perceived hazard characteristics that address estimates and assessments of risk, level of personal control, and trust in management. However, in their paper relating the RISP model to climate change policy, Yang et al. (2014) suggest that “perceived salience,” or the likelihood of an individual to attend to relevant or important issues, is more applicable to climate change policy. The authors find that perceived salience and individual characteristics influence risk perceptions, which in turn impact an individual’s affect response. The model predicts that positive moods and emotions about an issue lead one to process information heuristically, while negative moods and emotions lead to systematic processing. However, studies suggest that extreme negative moods and emotions lead to heuristic processing. The RISP model attempts to provide an explanation for why some information is processed heuristically, which requires less cognitive effort and resources, and why other information is processed systematically, requiring more comprehensive effort to analyze and understand an issue. One’s subjective norms, which are the perceived social normative influences that dictate how much an individual should know about an issue, influence both the information sufficiency threshold and information processing. In other words, those who perceive more social pressure to be knowledgeable about issues, such as climate change, will believe they need to know more about the issue, and are therefore more likely to process information systematically rather than heuristically. Griffin et al. (1999) suggest that one’s demographic characteristics and political party affiliation dictate one’s subjective norms. 11 However, Yang et al. (2014) found no causal relationship between perceived information gathering capacity and systematic. Figure 3. The Risk Seeking Information and Processing model as a precursor to the Theory of Planned Behavior, adapted from Griffin et al. 1999. Variables in the blue box are part of the Theory of Planned Behavior, variables in the grey box are part of the RISP framework. The Theory of Planned Behavior, or the idea that all behaviors are completely voluntary, is also incorporated into the RISP model (Figure 3). One’s attitude toward performing a behavior is influenced by salient behavioral beliefs that are weighed against each other. For example, if an individual’s attitude towards climate change information is positive, then one would expect that individual to process policy information systematically, rather than heuristically. The RISP model is a useful tool to investigate the connection between coastal communities’ perceptions of risk and climate change policy. Using constructs from the RISP model and the Values-Beliefs-Norms theory will allow for comparison within and between coastal communities on Lake Michigan. In this research I explore constructs of risk, namely 12 “Concern” and self-reported “Knowledge.” Constructs of severity, susceptibility, dread, and concern for the health of the Great Lakes, private property, and public spaces were closely related and therefore grouped into the general category of “Concern” (Figure 4). The constructs of informational subjective norms, information gathering capacity, and perceived behavioral control were grouped into the category self-reported “Knowledge” as a facet of risk perception (Figure 5). Figure 4. Conceptual mode of "Concern" as a facet of risk perception. 13 Figure 5. Conceptual model of self-reported “Knowledge" as a facet of risk perception. 14 Research Questions The research questions are: 1. What are the differences in perceptions of coastal risks and hazards in Lake Michigan communities? a. What factors influence concern for coastal risks and hazards in Lake Michigan communities? b. What factors influence self-reported knowledge of coastal risks and hazards in Lake Michigan communities? 2. What are Lake Michigan stakeholders’ motivations for coastal habitat stewardship? a. What community characteristics and socio-demographic factors predict motivation for coastal habitat stewardship? b. What factors of coastal risk perception (e.g., concern, self-reported knowledge) predict motivation for coastal habitat stewardship? 15 Study Sample CHAPTER 2: METHODS Coastal communities were identified based on their county, population size, shoreline type, and level of active coastal resilience policies or practices. I determined the level of coastal resilience policies or practices based on prior interventions since 2016 by Resilient Michigan or by presence of policies or programs that reflected obvious prioritization of coastal resilience. Resilient Michigan is an organization funded by the Coastal Zone Management Program that has taken a community-engaged approach to helping Michigan communities update their master plans with a focus on coastal resilience. Three Michigan counties met these criteria and were selected to represent the diversity of communities on Lake Michigan. A total of 8 communities were chosen (Table 1). The counties were Allegan County, Ottawa County, and Muskegon County. In Allegan County, I surveyed City of Douglas, Saugatuck City, and Saugatuck Township. In Ottawa County I surveyed Ferrysburg City, Grand Haven City, and Grand Haven Charter Township. In Muskegon County I surveyed City of Muskegon and City of Norton Shores. Allegan County was considered a county with small population sizes, communities in Ottawa County had mixed population sizes, and communities in Muskegon County had large population sizes. Grand Haven City, Grand Haven Charter Township, and City of Muskegon all had coastal resiliency policies or programs at the time of the study. 16 Table 1. Community matrix displaying community, county, county size type, total population, previous coastal resilience policy or program, and shoreline type. Community County County Size Type Total Populationa Resiliency Policy? Shoreline Typeb City of Ferrysburg Grand Haven City Grand Haven Charter Township Ottawa Ottawa Ottawa Allegan Allegan Allegan Muskegon Muskegon Mixed Mixed Mixed Small Small Small Large Large 2,992 10,929 16,266 993 915 3,113 38,131 23,994 City of Douglas Saugatuck City Saugatuck Township City of Muskegon City of Norton Shores adata from American Community Survey (5-year estimates, 2013-2017) conducted by the U.S. HB HB HB HD HD Census Bureau bHD = high dunes; HB = high bluffs; HD/HB = high dunes and high bluffs HD HD/HB HD/HB No Yes Yes No No No Yes No I identified four strata for my sampling frame. The four categories are: (1) “Lake Residents” who are people who own land parcels immediately adjacent to Lake Michigan, (2) “Near-lake Residents” who were people who own land parcels within a quarter-mile from Lake Michigan, not including Lake Residents, (3) “Inland Residents” who were people who own land parcels more than a quarter-mile from Lake Michigan, and (4) “Municipal Officials” who were both elected and appointed municipal officials and staff people. Although people who were only part-time residents or simply owned property on the coast were included in the survey, for the purposes of this thesis they are all termed “residents.” Municipal officials were identified based on their position in the community and how directly they worked with issues related to coastal resilience. Municipal officials include, but are not limited to, members of the Planning Commission, Zoning Board, and Parks and Recreation Board. The first three categories were established based on interest from agency partners, who hypothesized that there would be 17 differences in risk perception between Lake, Near-lake, and Inland Residents. I was also interested in looking at potential differences between the first three categories compared to municipal officials who ultimately have decision-making power. The contact information for Lake, Near-lake, and Inland Residents was obtained through the relevant county Geographic Information System (GIS) service’s department. ArcGIS was then utilized to determine residents who owned land parcels on the lake, a quarter-mile from the lake, and more than a quarter-mile from the lake. Necessary survey sample sizes were based on the total number of residents in each community. A total of 8420 survey invitations were mailed to the three counties (Appendix A). A four-wave tailored design methodology was used for data collection (Dillman, 2009). Lake, Near-lake, and Inland Residents were mailed four invitations (letter, postcard, letter, postcard) to participate in the online Qualtrics survey. The invitation included an explanation of the study, a link to the survey, an individual code to ensure participants only completed the survey once and to make sure I could identify responses by resident type, and relevant contact information (Appendix F). The survey invitation materials were altered slightly for Muskegon County due to explicitly mention participants could take the survey over the phone and to clarify who to contact with questions (Appendix G). I expected a response rate of about 40% for the email-based invitation (Greenlaw & Brown-welty, 2009) and a response rate of about 10-12% for the mailed invitations (Dykema, Stevenson, Klein, Kim, & Day, 2013; Kaplowitz, Lupi, Couper, & Thorp, 2012). I received 924 total responses for a response rate of 11.0% To reach the appropriate municipal officials, invitations to participate in the online survey were emailed (Appendix H). In Ottawa County, a community contact voluntarily emailed county listservs to the appropriate officials and staff people. In Allegan County and Muskegon County, 18 officials were emailed individually, and administrative assistants were also asked to further distribute the invitation to relevant officials and staff people. This study was approved and determined as exempt by The Michigan State University Institutional Review Board on October 22, 2018 (STUDY00001557) (Appendix E). Allegan and Ottawa County residents were contacted from December 6, 2018 to December 28, 2018. Allegan and Ottawa County municipal officials were contacted from December 12, 2018 to December 20, 2018. Muskegon County residents were contacted from April 2, 2019 to April 30, 2019. Muskegon County municipal officials were contacted from April 26, 2019 to June 2, 2019. The delay in contacting Muskegon County was due to logistical issues. Survey Design I asked participants questions related to environmental coastal risk to better understand how and why community residents form their risk perceptions (Sjoberg, 2000) (Table 2). For the purposes of this predominantly exploratory study, environmental coastal risk was not explicitly defined in order to keep participant responses as unbiased as possible. First, I asked participants to indicate perceived levels of severity, susceptibility, and dread about coastal risk (Yang, Rickard, Harrison, & Seo, 2014). To measure severity, I asked participants to indicate how serious they think coastal risks are to their community. Severity was measured on a 5-point Likert-type scale from 1 (not at all serious) to 5 (very serious). To measure susceptibility, I asked participants to indicate how much they think coastal risks will harm their local community in the next ten and 50 years. Susceptibility was measured on a 5-point Likert-type scale from 1 (not at all susceptible) to 5 (extremely susceptible). To measure dread, I asked participants to indicate how much information about coastal risks makes them feel worried. Dread was measured on a 5- point Likert-type scale from 1 (strongly disagree) to 5 (strongly agree). 19 Table 2. Constructs describing different facets of risk perception, the mean response, and standard deviation for the Combined Model. Model Description How serious of a threat are coastal risks Construct Severity Mean 3.43 SD 1.13 to your community? How much do you think coastal risks will harm your community in the next TEN years? How much do you think coastal risks will harm your community in the next FIFTY years? How worried do you feel about coastal risks? How concerned are you about the health of your community’s shoreline? How concerned are you that coastal risks could affect private property? How concerned are you that coastal risks could affect public spaces? I understand information about coastal risk. When it comes to information about coastal risk, I can separate facts from fiction. My family expects that I know something about coastal risk. My friends expect that I know something about coastal risk. I can easily locate information about coastal risk. I am personally able to take action to manage coastal risk. Susceptibility Concern Dread Stewardship Information gathering capacity Knowledge Informational subjective norms Perceived behavioral control 3.13 1.11 3.77 1.19 3.44 3.80 3.19 3.64 3.99 1.26 1.07 1.35 1.17 0.96 4.05 0.97 3.68 3.60 3.36 2.63 1.07 1.02 1.15 1.17 I also asked participants about characteristics that influence risk perceptions, such as information gathering capacity, informational subjective norms, and perceived behavioral control (Table 2). Information gathering capacity was measured using two items on 5-point Likert-type scales ranging from 1 (strongly disagree) to 5 (strongly agree), in which participants were asked to indicate how much they understand information about coastal risk and how much they feel 20 they can separate facts from fiction. Informational subjective norms were also measured using two items on 5-point Likert-type scales ranging from 1 (strongly disagree) to 5 (strongly agree), in which participants were asked to indicate how much their family and how much their friends expect them to know something about coastal risks. Perceived behavioral control was measured using two items on 5-point Likert-type scales ranging from 1 (strongly disagree) to 5 (strongly agree), in which participants were asked to indicate how easily they can locate information about coastal risk and how able there are to take action to manage coastal risk. To understand motivations for coastal habitat stewardship, I asked participants several questions that measured their levels of concern about the health of the Great Lakes coastal region (Table 2). I also asked participants if they have even been involved in a program or organization whose primary goal was Great Lakes coastal zone management (Table 3). Participants were given several involvement options and were asked to mark all that applied. I also asked participants to list any programs or organizations and briefly explain why they become involved. Table 3. Number and frequency of respondents who said they were involved in a program or organization whose focus is on Great Lakes habitat preservation, conservation, or management. Muskegon County Yes No Allegan County 100 (38.7%) 140 (61.3%) Ottawa County 142 (33.7%) 278 (66.3%) Pooled 307 (35.2%) 565 (64.8%) 57 (34.7%) 132 (65.3%) Finally, participants were asked to self-report socio-demographic information, including: race and ethnicity, highest level of education, annual income before taxes, birth year, gender, and zip code (Table 4). Participants were also asked about their resident status (full-time, part-time, visitor, or other), what best describes their resident location (lake-front property, private beach access but not lake-front property, or neither), and how long they or their family has owned property in their coastal community. Previous experiences have been shown to impact risk perception, so I asked participants if they had experienced the following events in the last five 21 years: coastal flooding, inland flooding, coastal erosion, severe storm events (Melillo, Richmond, & Yohe, 2014; Owens & Driffill, 2008; Weber, 2011). Participants were also asked to estimate how many times they have experienced any of the previously mentioned events in the last five years. Table 4. Socio-demographic variables hypothesized to predict an individual's risk perceptions. Descriptive statistics displaying mean (x̄ ) or frequency of respondents in each category with range or standard deviation (s) in parentheses. Variable name Measurement Continuous (centered) Continuous Ordinal 1 (< 5 years) 2 (6-10 years) 3 (11-20 years) 4 (21-49 years) 5 (> 50 years) Ordinal 1 (< $20,000) 2 ($20,000 ≤ income ≤ $34,999) 3 ($35,000 ≤ income ≤ $49,999) 4 ($50,000 ≤ income ≤ $74,999) 5 ($75,000 ≤ income ≤ $99,999) 6 (> $100,000) 1 if year-round; 0 if not year-round 1 if community has a resiliency policy or program; 0 if no resiliency policy or program 1 if male; 0 if non-male some college 1 if Bachelor’s degree 1 if Master’s degree or higher 1 if resident of Allegan County (small) 1 if resident of Ottawa County (mixed) 1 if resident of Muskegon County (large) 1 if lake-front property 1 if 0.25 miles from Lake Michigan 1 if more than 0.25 miles from Lake Michigan 1 if municipal official or staff Descriptive statistics x̄ =60.87 (19-91) x̄ =3.11 (0-111) x̄ =3.33 (s=1.38) x̄ =5.04 (s=1.27) 73.29% 57.63% 59.62% 18.34% 37.13% 44.53% 26.60% 50.86% 22.54 % 17.13% 25.99% 51.07% 5.81% Age Previous experiences Property ownership time Income Year-round resident Resiliency policy Gender Education Bachelor’s degree Graduate degree (ref.) County size Small (ref.) Mixed Large Resident location Lake (ref.) Near-lake Inland Municipal officials 22 Associate’s degree or less 1 if high school degree, Associate’s degree, or 23 Analysis Research Question #1: What are the differences in perceptions of coastal risks and hazards in Lake Michigan communities? Descriptive statistics were measured for the seven different environmental coastal issues under climate change and for questions related to governance. One-way analyses of variance (ANOVA) were first used to determine appropriate reference categories and for data reduction. A Tukey post-hoc test was conducted to determine the individual differences between counties (proxy for county size categories), resident types (lake, near-lake, inland, and municipal officials), and education levels. A two-way analysis of variance was also conducted to determine if there was an interaction between county and resident type, however there were no significant findings so the results are not included in this chapter. Confirmatory factor analysis first using structural equation modeling and then a structural regression was conducted to understand differences in risk perception within and between Lake Michigan coastal communities. This method is appropriate because the observed variables are all facets of risk perception and reducing the data is much preferred for analysis and interpretation. A general model for one score of risk perception was attempted, but the model fit was not adequate. Therefore, one model was created based on variables related to concern about coastal risk (“Concern Model”) (Figure 6), while the other model was based on variables related to self- reported knowledge and behaviors about coastal risk (“Knowledge Model”) (Figure 7). A baseline model was created based on risk perception theory and was adjusted based on the goodness-of-fit statistics and modification indices (Table 5). All variables had factor loadings greater than .30 (Hair, Black, Babin, Anderson, & Tatham, 2006). 24 Figure 6. Conceptual model of "Concern" as a facet of risk perception with factor loadings. 25 Figure 7. Conceptual model of self-reported "Knowledge" as a facet of risk perception with factor loadings. 26 Table 5. Summary of goodness-of-fit statistics for baseline models and final structural models of Concern and Knowledge facets of risk perception. Model c2 df p RMSEA (90% CFI TLI SRMR CI, lower bound) .138 (.123) .039 (.018) .187 (.169) .046 (.024) .041 Baseline CFA model (Concern) .013 Final structural model (Concern) .083 Baseline CFA model (Knowledge) Final structural model (Knowledge) .019 CFA = confirmatory factor analysis; df = degrees of freedom; RMSEA = Steiger-Lind Root Mean Square Error of Approximation; CFI = Bentler Comparative Fit Index; TLI = Tucker Lewis Index; SRMR = standardized root mean square residual. 238.32 14 < .05 24.89 11 < .05 299.64 9 < .05 7 < .05 20.54 .948 .997 .857 .993 .921 .994 .761 .986 Research Question #2: What are Lake Michigan stakeholders’ motivations for coastal habitat stewardship? To better understand what factors predict whether or not individuals are involved in a program or organization focused on Great Lakes habitat stewardship, I used logistic regressions. Let Y represent whether or not participants are involved in a program or organization and let Xi represent independent variables described in Table 4. logit(Yi) = b0 + Xibi + ei Analysis is broken into two parts: (1) Combined Model that combines all three counties and (2) county models that look at counties individually. For the Combined Model, logistic regressions were run to determine the effect of county characteristics (i.e. size and presence of resiliency policy), resident type (lake, near-lake, inland, municipal officials), and various socio- demographic factors. Four models were run to determine the effect of resident type, county characteristics, and various socio-demographic factors on Great Lakes habitat stewardship. Model 1 compared Lake Residents to the remaining three resident types (Near-lake, Inland, and Municipal Officials). Model 2 included a comparison between resident types, as well as the 27 existence of resiliency policy. Model 3 included county size. Model 4 included various socio- demographic factors. For the individual county models, logistic regressions were run to determine the effect of resident type and various socio-demographic factors. Model 1 is a comparison between Lake Residents and the remaining three residents, and Model 2 includes the various socio-demographic factors. For all models, factors predicting concern for coastal risk and self-reported knowledge of coastal risk were also used to determine the effect on program involvement. The Combined Model includes an additional model that investigates all potential factors influencing involvement in one logistic regression. Effect of shoreline type was also investigated but resulted in no significant findings, and is therefore removed from results. All analysis and descriptive statistics were conducted using statistical package Stata 14.2 and R 3.5.1. To determine participants motivations for coastal habitat stewardship, I used emergent coding to look for common themes in responses to “What motivated you to become involved with the program(s) or organization(s) [whose main goal is Great Lakes coastal region management, conservation, or preservation]?” Through this method, codes are inductively determined, meaning they come from the response data rather than the literature or the researcher’s previous experiences (Miles & Huberman, 1994). This method is preferred for this type of research, as the motivations for stewardship of this specific area and system have not been previously studied. Participants’ responses could be coded into multiple categories, depending on their response and the range of topics they mentioned. 28 CHAPTER 3: PERCEPTIONS OF COASTAL RISK Results Environmental concerns In general, residents were more concerned about seven environmental concerns related to climate change in the next 50 years than in the next 10 years (Table 6, 7). Residents are most concerned about coastal erosion (3.84 (1.17)) and least concerned about increases in precipitation (2.72 (1.23)). Residents in all three counties also reported they are most concerned about coastal erosion and least concerned about increases in precipitation. In general, residents of Allegan County were more concerned about all seven risks than Ottawa or Muskegon County. For all risks and for all counties, the average concern for climate change risks was higher when asked about the next 50 years when compared to the next 10 years. Table 6. Environmental concerns in 10 years. Mean responses from a 1 (not at all concerned) to 5 (extremely concerned) Likert-type-scale are displayed with standard deviations in parentheses. Highest average concerns are noted in bold. Combined More frequent and severe storms Increases in precipitation More precipitation as rain than snow Lakes Reduced ice coverage on the Great More flooding events Coastal erosion Increases in extreme temperatures Model 2.93 (1.28) 2.72 (1.23) 2.82 (1.27) Allegan County 3.17 (1.25) 3.05 (1.25) 3.15 (1.24) Ottawa County 2.86 (1.26) 2.58 (1.19) 2.76 (1.22) Muskegon County 2.85 (1.34) 2.68 (1.25) 2.65 (1.35) 3.12 (1.32) 3.43 (1.27) 3.08 (1.27) 2.94 (1.39) 3.13 (1.30) 3.84 (1.17) 3.29 (1.36) 3.38 (1.25) 4.11 (1.07) 3.50 (1.31) 3.04 (1.28) 3.71 (1.22) 3.21 (1.34) 3.10 (1.34) 3.82 (1.11) 3.27 (1.41) 29 Table 7. Environmental concerns in 50 years. Mean responses from a 1 (not at all concerned) to 5 (extremely concerned) Likert-type-scale are displayed with standard deviations in parentheses. Highest average concerns are noted in bold. Combined More frequent and severe storms Increases in precipitation More precipitation as rain than snow Lakes Reduced ice coverage on the Great More flooding events Coastal erosion Increases in extreme temperatures Model 3.29 (1.41) 3.17 (1.39) 3.21 (1.39) Allegan County 3.61 (1.36) 3.51 (1.36) 3.54 (1.35) Ottawa County 3.22 (1.40) 3.05 (1.37) 3.15 (1.37) Muskegon County 3.18 (1.43) 3.10 (1.42) 3.02 (1.44) 3.42 (1.42) 3.73 (1.33) 3.37 (1.42) 3.24 (1.45) 3.48 (1.37) 3.94 (1.23) 3.54 (1.43) 3.73 (1.33) 4.20 (1.12) 3.79 (1.35) 3.42 (1.38) 3.83 (1.30) 3.45 (1.45) 3.39 (1.38) 3.90 (1.18) 3.51 (1.41) Governance Participants were asked who they think owns coastal shoreline and who they think is responsible for managing the coastal shoreline in their communities in a “mark all that apply” question (Figure 7, Figure 8). Their responses were grouped into four categories: (1) “Private” which included response options “you” and “your neighbor,” (2) “Government” which included response options “Local Government,” “State Government,” and “Federal Government,” (3) “Public” which included responses that mentioned “the general public” or “everyone” having ownership of the shoreline, and (4) “Mix” which included two or more of the categories previously listed. Almost half of the mentions (43.9%) regarding ownership of coastal shoreline was either local, state, and/or federal government (Figure 8). This was closely followed by a mix of government, private landowners, or publicly owned shoreline (41.0%). Private landowners were mentioned 13.0% of the time and “the general public” was explicitly mentioned or alluded to 2.1% of the time. When broken down by Resident Types, Lake Residents indicated that they think a mix of private and government entities own the coastal shoreline (Table 8). While a majority Near-lake Residents, Inland Residents, and Municipal Officials said government owns 30 coastal shoreline. There were no significant differences between the four Resident Types (F(3,881)=1.06, p>.05). Private 13.0% Government 43.9% Mix 41.0% Public 2.1% Figure 8. Response of the participants to the question of who owns coastal shoreline in their community (n=907). Table 8. Responses to the question "Who do you think owns coastal shoreline in your community?" broken down by Resident Type. Resident types Lake Residents Near-lake Residents Inland Residents Municipal officials “Private” included response options “you” and “your neighbor.” “Government” included response options “Local Government,” “State Government,” and “Federal Government.” “Public” included responses that mentioned “the general public” or “everyone” having ownership of the shoreline. “Mix” included two or more of the categories previously listed. Private 26.0% 14.5% 8.5% 6.0% Public 2.0% 2.7% 2.1% 0.0% Government 10.0% 47.1% 53.6% 52.0% Mix 62.0% 35.8% 35.9% 42.0% When asked who they think is responsible for managing coastal shoreline in their community, respondents said a mix of government, private landowners, or the general public 49.2% of the time (Figure 9). This was closely followed by either local, state, and/or federal 31 government (44.8%). Private landowners (“you” or “your neighbor”) was only mentioned 4.1% of the time and “the general public” was explicitly mentioned or alluded to 1.9% of the time. When broken down by Resident Type, all groups seemed to agree that either government or a mix are responsible for managing coastal shoreline (Table 9). There were no significant differences between the four Resident Types (F(3,881)=1.89, p>.05). Mix 49.2% Private 4.1% Government 44.8% Public 1.9% Figure 9. Response of the participants to the question of who is responsible for coastal shoreline in their community (n=919). Table 9. Responses to the question "Who do you think is responsible for managing coastal shoreline in your community?" broken down by Resident Type. Resident types Lake Residents Near-lake Residents Inland Residents Municipal officials “Private” included response options “you” and “your neighbor.” “Government” included response options “Local Government,” “State Government,” and “Federal Government.” “Public” included responses that mentioned “the general public” or “everyone” having ownership of the shoreline. “Mix” included two or more of the categories previously listed. Private 6.7% 4.5% 3.4% 1.9% Public 2.0% 0.9% 2.5% 1.9% Government 34.0% 42.4% 50.6% 38.5% Mix 57.3% 52.2% 43.6% 57.7% 32 Respondents were also asked what they think is the best way to manage a receding shoreline (Figure 10). A majority of respondents (41.1%) said a natural shoreline is the best way. This response was closely followed by those who said both a man-made and natural shoreline are about equal (33.5%). Only 9.7%% said they thought a man-made shoreline was the ideal management strategy. 6.1% of respondents said neither was an effective way to manage a receding shoreline. When responses are broken down by Resident Type, Lake Residents were split between man-made structures or man-made structures and natural structures being about equal (Table 10). Near-lake Residents, Inland Residents, and Municipal Officials all preferred natural structures over any of the other options. Lake Residents had significantly different preferences about management strategies compared to Inland Residents and Municipal Officials (F(3,881)=7.53, p<.001) Other 9.7% Man-made 9.7% Neither 6.1% About equal 33.5% Natural shoreline 41.1% Figure 10. Response of the participants to the question of what is the best way to manage a receding shoreline (n=844). 33 Table 10. What is the best way to manage a receding shoreline? Frequency of response by Resident Type (Lake, Near-lake, Inland, and Municipal officials). Resident types Lake Residents Near-lake Residents Inland Residents Municipal officials “Private” included response options “you” and “your neighbor.” “Government” included response options “Local Government,” “State Government,” and “Federal Government.” “Public” included responses that mentioned “the general public” or “everyone” having ownership of the shoreline. “Mix” included two or more of the categories previously listed. Natural 33.7% 48.7% 46.9% 50.0% Neither 9.6% 7.6% 5.4% 6.5% Man-made 26.9% 7.0% 8.0% 10.9% About equal 29.8% 36.8% 39.6% 32.6% Comparisons between Counties, Resident Types, and Education Levels The results from the ANOVAs revealed that between counties, there was a significant difference in concern for severity of risks (F(2,855)=6.86, p<.001), harm in 10 years (F(2,851)=11.51, p<.001), harm in 50 years (F(2,828)=9.29, p<.001), dread (F(2,876)=7.05, p<.001), health of the Great Lakes (F(2,876)=7.65, p<.001), concern for public spaces (F(2,874)=8.40, p<.001). The results from the Tukey post-hoc tests suggest that there are no differences in risk perception between Ottawa and Muskegon Counties, therefore Allegan County is used as a reference group for remaining analyses. There was also a significant difference in concern for severity of risks (F(3,854)=6.94, p<.001), harm in 10 years (F(3,850)=4.96, p<.01), expectation from friends (F(3,872)=3.88, p<.01) and family (F(3,873)=8.99, p<.001), ability to personally manage risks (F(3,877)=3.61, p<.05), health of the Great Lakes (F(3,875)=4.77, p<.01), and concern for private property (F(3,870)=30.88, p<.001) between Resident Types. The results from the Tukey post-hoc tests revealed that Lake Residents are an appropriate reference group for remaining analyses. Lastly, respondents of different education levels also varied in their concern for severity of coastal risks (F(2,855)=7.75, p<.001), harm in 10 years (F(2,851)=4.61, p<.01) and 50 years (F(2,828)=6.29, p<.01), health of the Great Lakes (F(2,876)=3.15, p<.05), and concern for private property (F(2,871)=3.42, p<.05). 34 Respondents also had significantly different self-reported knowledge scores based on their highest level of education in their ability to separate fact from fiction (F(2,876)=4.77, p<.01) and expectations from their family (F(2,874)=4.69, p<.01). The results from the Tukey post-hoc tests suggest that respondents with graduate degrees should be the reference group for all following analyses. Facets of Risk Perception – Combined Model In general, Land Residents are less concerned about coastal risk than Lake Residents (b=- .45, p<.001) (Table 11). The results from the regression analysis also suggest that communities with resilience policies or programs in place are less concerned about coastal risk than their counterparts (b=-.28, p<.01). Furthermore, as the amount of experience with coastal hazards increased over the last five years, so too did respondents’ concern about coastal risk (b=-.04, p<.001). Education levels also seemed to have an effect on concern about coastal risk. Both participants with less than an associate’s degree (b=-.33, p<.01) and bachelor’s degree (b=-.17, p<.05) were less concerned about coastal risk than those with a graduate degree. Males were also less concerned about coastal risk than females (b=-.34, p<.001). There was no significant difference in concern between counties (as a proxy for county size), amount of time respondents owned property in the coastal community, if they are a year-round resident or not, or respondents age. In general, there were not many significant differences between or within communities’ self-reported knowledge related to coastal risk and hazards. Regression analysis suggests that as previous experience with coastal risks increases, respondents’ self-reported knowledge about coastal risk also increases (b=-.22, p<.05). It also appears that individuals with bachelor’s degrees indicate less knowledge about coastal risk than those with graduate degrees (b=-.19, p<.01). The results from Model 3 suggest that Near-lake Residents (b=-.22, p<.05) and Inland 35 Residents (b=-.23, p<.01) are less knowledgeable about coastal risk than Lake Residents, however this trend disappears in the final model (Model 4). In the final model, there was no significant difference in knowledge about coastal risk between resident types, counties, amount of time individuals have owned property in a coastal community, year-round residents, income, gender, or age. Table 11. Regression model of the final structural models for Concern (Model 1 and 2) and Knowledge (Model 3 and 4). Model 1 and 3 look at characteristics of the communities and counties, Model 2 and 4 include socio-demographic factors. Variables County size (ref: Small) Knowledge Concern Mixed Large Resident location (ref: Lake) Near-lake Inland Municipal officials Resiliency policy Previous experiences Property ownership time Year-round resident Education (ref: Graduate degree) Associate’s degree or less Bachelor’s degree Model 1 -.09 (.11) -.06 (.13) -.27 (.11)* -.45 (.19)*** -.35 (.19) -.28 (.10)** Model 2 -.03 (.12) .04 (.14) -.20 (.12) -.32 (.12)** -.13 (.20) -.32 (.10)** .04 (.01)*** -.03 (.03) -.16 (.09) -.33 (.11)** -.17 (.08)* -.05 (.03) -.34 (.08)*** .00 (.00) Model 3 .01 (.09) .04 (.10) -.22 (.09)* -.23 (.08)** -.26 (.14) .06 (.08) 16467.31 Model 4 -.03 (.10) .00 (.12) -.10 (.11) -.15 (.10) -.17 (.17) .05 (.09) .02 (.01)** .04 (.02) .10 (.08) -.13 (.10) -.19 (.07)** .01 (.03) -.04 (.07) .00 (.00) 30833.89 Income Male Age AIC *p < .05; **p < .01; ***p <.001 Regression coefficients listed with standard errors in parentheses. 16276.96 29642.81 Facets of Risk Perception – By County Inland Residents in Allegan County are less concerned about coastal risk than Lake Residents (b=-.60, p<.001) (Table 12). The results from the regression also suggest that residents with more experience with previous coastal risks are more likely to be concerned about coastal risk (b=-.03, p<.01). Furthermore, those with an Associate’s Degree or less are not as concerned 36 about coastal risk when compared to those with a graduate degree (b=-.79, p<.01). Males are also less concerned about coastal risk than females (b=-.43, p<.01). Interestingly, there were no significant predictors for self-reported knowledge about coastal risk in Allegan County. In Ottawa County, as amount of previous experience with coastal risk increased, so too did individual’s concern about coastal risk (b=-.05, p<.001) and self-reported knowledge about coastal issues (b=-.03 p<.01) (Table 12). Males were also less concerned about coastal risk than females (b=-.29, p<.01). As the amount of time respondents have owned property in their communities increased, so too did their self-reported knowledge about coastal risk (b=-.07, p<.05). Individuals with a bachelor’s degree also indicated they were less knowledgeable about coastal risk than those with a graduate degree (b=-.20, p<.05). Near-lake Residents in Muskegon County reported they are less concerned about coastal risk than Lake Residents (b=-.61, p<.05) (Table 12). Males are also less concerned about coastal risk than females (b=-.37, p<.05). As with the previous counties, as amount of previous experience with coastal risk increases, individuals report more concern (b=-.03, p<.05) and more knowledge (b=-.04, p<.01) about coastal risk. 37 Table 12. Regression models of the final structural models for three counties (Allegan County, Ottawa County, and Muskegon County) of Concern and Knowledge. Allegan County Ottawa County Muskegon County Variables Resident location (ref: Lake) Near-lake Inland Municipal officials Previous experiences Property ownership time Year-round resident Education (ref: Graduate degree) Associate’s degree or less Bachelor’s degree Model 1: Concern -.39 (.20) -.60 (.17)*** -.39 (.38) .03 (.01)** -.03 (.05) -.11 (.13) -.79 (.23)** -.17 (.13) -.04 (.06) -.43 (.13)** .00 (.01) .23 8393.22 Model 2: Knowledge -.20 (.17) -.18 (.14) -.01 (.33) .00 (.01) .08 (.04) .15 (.12) -.08 (.20) -.20 (.12) .01 (.06) -.17 (.12) .01 (.01) .10 8347.251 Model 3: Concern .05 (.16) -.15 (.16) .03 (.22) .05 (.01)*** -.04 (.04) -.20 (.13) -.28 (.15) -.11 (.11) -.03 (.04) -.29 (.11)** .01 (.00) .14 15015.59 Model 4: Knowledge .12 (.13) -.09 (.13) -.15 (.18) .03 (.01)** .07 (.03)* .08 (.10) -.01 (.13) -.20 (.09)* .01 (.04) -.01 (.09) .00 (.00) .08 15776.81 Model 5: Concern .61 (.28)* -.55 (.28) .23 (.51) .03 (.01)* .11 (.06) -.32 (.19) .12 (.20) -.23 (.17) -.06 (.06) -.37 (.16)* -.00 (.01) .19 6905.819 Model 6: Knowledge -.07 (.24) -.05 (.24) .35 (.47) .04 (.01)** -.02 (.05) .06 (.17) -.09 (.17) -.12 (.16) -.01 (.05) .03 (.14) -.00 (.01) .10 7165.321 Income Male Age R2 AIC *p < .05; **p < .01; ***p <.001 Regression coefficients listed with standard errors in parentheses. 38 Discussion In general, Lake Residents are more concerned about coastal risk than Near-lake and Inland Residents. This is most likely because Lake Residents have private property on the lake that is at a much higher risk of flooding and coastal erosion than Inland Residents. This finding supports previous research that suggests direct experience with environmental risks increases an individual’s risk perception and concern of those risks (Whitmarsh, 2008). In fact, I found that as the number of previous experiences with coastal risks increased, so too does an individual’s risk perception. However, there was no significant difference in self-reported Knowledge about coastal risk, this suggests that although Lake Residents are more concerned about coastal risk, they don’t necessarily have more or less knowledge about how to manage coastal risks. The results of the Tukey post-hoc test further show that Lake Residents feel they have less ability to take action to manage coastal risk than Municipal Officials. Furthermore, Lake Residents also think there are less policies in place to help manage coastal risk when compared to Inland Residents and Municipal Officials. This suggests that it is important to focus on decreasing the disconnect in risk perceptions between Lake Residents and Municipal Officials, either by putting in place more beneficial policies or by better educating coastal residents. Although there were no significant differences in county size, it is noteworthy that the presence of resiliency policy or program had an effect on risk perceptions. It is possible that the community-engaged work done by Resilient Michigan helps residents feel more confident in their abilities to handle current and future unpredictable coastal dynamics. However, it is also possible that residents in these areas are simply not as aware of coastal risk and their community’s focus on resilience is mainly the accomplishment of a few municipal officials who prioritize coastal risk. 39 Individuals who had an education level less than a bachelor’s degree (some high school, high school graduate, some college, or associate’s degree) were less concerned about coastal risks than those with a graduate degree (master’s degree, professional degree, or doctorate degree). This finding also follows previous literature that has found a positive relationship between education level and concern about environmental issues, such as climate change (Dietz, Dan, & Shwom, 2007; Semenza et al., 2008). These previous studies have suggested individual’s with higher levels of education are more likely to be open to new ideas and are more likely to change their behavior to help mitigate the impacts of climate change. It may also be that those with higher levels of education have been formally introduced to the complex factors related to the negative impacts of climate change, and therefore are more aware of the issue (Feltman et al., 2017). However, while individuals with less than a bachelor’s degree appear to be more concerned about coastal risk, there was no difference in self-reported knowledge about coastal risk. On the other hand, individuals with a bachelor’s degree had a self-reported knowledge score that was less than those with a graduate degree. Perhaps this is because those with a bachelor’s degree have had enough higher education to better understand the intricacies of climate change, and therefore realize there are a lot of factors they are not equipped to fully understand. I also found that males are significantly less concerned about coastal risk than non-male respondents. Previous research on climate change perceptions has suggested that men are less concerned about climate change most likely because men self-report being more informed about climate change (Gifford & Comeau, 2011). In other words, males seem to be less concerned about environmental risks because they are more confident in their understanding of the issue, and therefore their ability to deal with negative outcomes. Interestingly, the only other predictor of Knowledge scores is the amount of time individuals have owned property in the coastal community. As the amount of time individuals 40 have lived in their coastal communities increases, so too does their self-reported knowledge about coastal risk. This finding supports place-based research that suggests the longer an individual has lived in an area, the more likely they are to develop meaningful relationship to that place (Lewicka, 2011). Perhaps those who have lived in their communities for longer are more invested in the health and resilience of the shoreline. Additionally, people may also feel they know more about their community and the coastline simply because they have lived there for longer and experienced more of the natural lake dynamics. In the context of Lake Michigan residents, variables including Resident Type, resiliency policy, previous experiences, education levels, and gender all influence Concern about coastal risk. However, only previous experiences and education levels influenced self-reported Knowledge about coastal issues. Furthermore, when looking at the counties individually, only previous experiences explains Concern and Knowledge risk perceptions. As the Knowledge model is explained by less of the independent variables in this study, focusing more on the Concern constructs might be more useful for future research. It is also likely that the Knowledge constructs are explained by other variables that were not investigated for this study. These results add to a growing body of research aimed at understanding the underlying mechanisms that influence coastal risk perceptions. 41 CHAPTER 4: MOTIVATIONS FOR COASTAL HABITAT STEWARDSHIP Results Combined Model Based on the results of the Combined Model, it seems that resident type, amount of previous experience with coastal risk, and time of ownership are all predictors for involvement in a program or organization focused on Great Lakes stewardship (Table 13). In the initial model (Model 1), the odds of involvement in a program or organization are lower for both Near-lake (OR=.55, p<.05) and Inland Residents (OR=.43, p<.001) than for Lake Residents. Although this trend stays consistent for Inland Residents in the final model (Model 2), any significant differences disappear for Near-lake Residents. In the final model, the odds of involvement are 5% higher as the amount of previous experience with coastal risk increases (p<.01). The odds of involvement are also 26% higher the longer participants owned property or lived in a coastal community (p<.01). There are no effects of county size, resiliency policy presence, year-round residency status, or socio-demographic factors on involvement in resiliency programs or organizations. 42 Table 13. Logistic regression predicting involvement in program for Combined Model. Odds ratios are presented with standard errors in parentheses. Model 1 1.02 (.25) .85 (.24) .55 (.13)* .43 (.09)*** .66 (.26) .73 (.16) .02 994.415 Model 2 1.09 (.31) .89 (.29) .58 (.16) .41 (.11)** .87 (.39) .68 (.17) 1.05 (.02)** 1.26 (.09)** 1.28 (.27) .65 (.18) .85 (.17) 1.03 (.08) .74 (.14) .99 (.01) .06 807.037 Variables County size (ref: Small) Mixed Large Resident location (ref: Lake) Near-lake Inland Municipal officials Resiliency policy Previous experiences Property ownership time Year-round resident Education (ref: Graduate degree) Associate’s degree or less Bachelor’s degree Income Male Age Pseudo R2 AIC *p < .05; **p < .01; ***p <.001 Odds ratios listed with standard errors in parentheses. None of the factors related to Concern for coastal risk were significant predictors for involvement in a program or organization (Table 14). However, the odds of involvement for respondents who indicated their friends think it is important they understand something about coastal risk are 44% higher than respondents who indicated their friends do not think it is important (p<.01). Additionally, the odds of involvement for respondents who indicated they are more able to personally manage coastal risk is 20% higher than those who indicated they do not feel they are able to personally manage risk (p<.01). These trends were also seen in the final model (Model 3). 43 Table 14. Logistic regression predicting involvement in a program or organization by Concern and Knowledge. Odds ratios are presented with standard errors in parentheses. Health of the GL Private property Public spaces Severity Harm (10 years) Harm (50 years) Dread Understand information Separate facts from fiction Family Friends Ability to locate information Ability to manage risk Pseudo R2 AIC *p < .05; **p < .01; ***p <.001 Odds ratios listed with standard errors in parentheses. Model 1: Concern 1.14 (.14 1.06 (.08) 1.24 (.14) 1.03 (.12) 1.07 (.15) 1.02 (.14) 1.05 (.08) .05 989.1966 Model 2: Knowledge 1.11 (.12) .96 (.11) 1.25 (.15) 1.44 (.18)** .96 (.07) 1.20 (.08)** .06 1053.186 Model 3: Combined 1.14 (.15) 1.06 (.09) 1.13 (.13) 1.01 (.12) 1.03 (.15) 1.06 (.16) 1.04 (.09) 1.01 (.12) 1.02 (.12) 1.20 (.15) 1.40 (.19)* .95 (.08) 1.26 (.09)** .10 939.7509 Individual County Models When looking at the Resident Type model (Model 1) for Allegan County, it seems that the odds of involvement is lower for both Near-lake (OR=.43, p<.05) and Land Residents (OR=.38, p<.01) than Lake Residents (Table 15). However, this trend disappears when socio- demographic factors are included. In the final model (Model 2), as the length of time individuals have lived in the community increases the odds of involvement increase by 37% (p<.05). The odds of involvement are also 2.78 times higher for year-round residents than non-year-round residents (p<.01). In Ottawa County, the odds of involvement are only lower for Land Residents (OR=.49, p<.05) compared to Lake Residents; this trend is also seen in Model 4. Furthermore, as the number of experiences increases the odds of involvement increase by 12% (p<.001). In Muskegon County, the odds of involvement are lower for both Near-lake Residents (OR=.32, p<.05) and Land Residents (OR=.28, p<.05) than Lake Residents. None of the factors related to 44 concern for coastal risk or self-reported knowledge of risk were significant predictors for involvement in any of the counties (Table 16). 45 Table 15. Logistic regression by county predicting involvement in a program or organization. Model 1 includes logistic regression controlling for resident locations. Model 2 includes Model 1 and socio-demographic predictors. Odds ratios are presented with standard errors in parentheses. Resident location (ref: Lake) Muskegon County Allegan County Ottawa County Model 5 Model 4 Model 6 Model 1 .43 (.18)* .38 (.13)** .27 (.24) Model 2 .54 (.27) .37 (.16)* .25 (.24) .99 (.02) 1.37 (.17)* 2.78 (1.02)** Model 3 .72 (.22) .49 (.14)* 1.02 (.43) .71 (.25) .44 (.16)* 1.25 (.64) 1.12 (.03)*** 1.18 (.11) .96 (.29) .32 (.17)* .28 (.14)* .82 (.76) .43 (.27) .30 (.19) 1.00 (--) 1.04 (.04) 1.11 (.17) 1.53 (.81) .02 542.1897 .99 (.37) 1.05 (.29) 1.11 (.12) .76 (.20) 1.00 (.01) .07 439.976 .03 231.834 .38 (.21) 1.02 (.45) .84 (.12) .67 (.28) .98 (.02) .09 193.4373 Near-lake Land Municipal officials Previous experience Ownership time Year-round resident Education (ref: Graduate degree) Associate’s degree or less Bachelor’s degree .03 325.060 Household income Male Age Pseudo R2 AIC *p < .05; **p < .01; ***p <.001 Odds ratios listed with standard errors in parentheses. .47 (.28) .73 (.25) .94 (.15) .50 (.18)* 1.02 (.02) .12 263.2128 46 Table 16. Logistic regression by county predicting involvement in a program or organization for Concern and Knowledge factors. Model 1 includes logistic regression controlling for resident locations. Model 2 includes Model 1 and socio-demographic predictors. Odds ratios are presented with standard errors in parentheses. Allegan County Ottawa County Model 2: Knowledge Model 1: Concern 1.05 (.27) 1.31 (.19) 1.33 (.26) .69 (.17) 1.36 (.36) 1.16 (.33) .91 (.13) .93 (.20) 1.21 (.25) Health of the GL Private property Public spaces Severity Harm (10 years) Harm (50 years) Dread Understand information Separate facts from fiction Family Friends Ability to locate information Ability to manage risk Pseudo R2 AIC *p < .05; **p < .01; ***p <.001 Odds ratios listed with standard errors in parentheses. 1.15 (.15) .07 312.813 1.22 (.26) 1.47 (.33) 1.00 (.15) .07 296.624 Model 1: Concern 1.26 (.24) .94 (.11) 1.14 (.20) 1.20 (.21) .82 (.18) 1.32 (.27) 1.19 (.15) Model 2: Knowledge 1.22 (.18) .97 (.16) Muskegon County Model 2: Knowledge .95 (.25) .76 (.18) Model 1: Concern .97 (.24) .97 (.17) 1.46 (.37) 1.15 (.27) 1.30 (.37) .57 (.17) 1.06 (.18) .08 466.712 1.37 (.25) 1.37 (.25) .91 (.10) 1.18 (.12) .07 517.024 .03 224.303 1.05 (.28) 1.67 (.48) 1.09 (.19) 1.31 (.20) .07 227.9206 47 Motivations for Stewardship When individuals were asked what motivated them to become involved with a program or organization that prioritizes Great Lakes coastal region management, conservation, or preservation, the most common response was regarding environmental concerns (23.4%, n=81) (Table 17). Many respondents simply mentioned general “environmental concerns,” but others mentioned issues such as, “the loss of the beach,” “concern for water quality,” or “to preserve the dunes.” Participants also mentioned the fact that they live near the Great Lakes (9.0%, n=31), concerns related to coastal development (8.1%, n=28). Also common were comments about desire to remove pollution or maintain the aesthetics of the coastline (6.9%, n=24), or a moral obligation or responsibility (6.6%, n=23). Slightly less common were comments about personal connections to the area (5.6%, n=20), careers or educational backgrounds related to coastal risk (4.9%, n=17), a desire for knowledge about coastal issues (4.9%, n=17), and those who joined because a family member or friend encouraged them or those who joined for the sense of community (4.9%, n=17). 48 Table 17. Range of motivations for stewardship concepts determined by emergent coding. Concept Environmental concerns Proximity to lake Issues with development Pollution concerns/Desire for aesthetics Moral obligation Personal connection Desire for community Desire for knowledge Career/Educational background Personal experience with risks Lacking government Love for area Good organization Future generations Need to protect coastline Community concerns General concern Passive Other Percentage (n) 23.4% (81) 9.0% (31) 8.1% (28) 6.9% (24) 6.6% (23) 5.8% (20) 4.9% (17) 4.9% (17) 4.9% (17) 3.8% (13) 3.2% (11) 2.9% (10) 2.6% (9) 2.0% (7) 2.0% (7) 0.9% (3) 4.3% (15) 2.0% (7) 1.7% (6) Discussion Based on the results of both the Combined Model and County Models, it seems that Lake Residents are more involved in a program or organization that is focused on Great Lakes region preservation, conservation, or management than Near-lake or Inland Residents (Table 13). This suggests that proximity to the Great Lakes is related to individual’s motivations for stewardship (Whitmarsh, 2008). In general, residents who live closer to the lake, have more previous experiences with coastal risk, and have property in a coastal community are more likely to be involved in a program or organization. This trend is supported by many respondents self- reporting that proximity to the lake is their main motivation for involvement. This suggests that targeting Lake Residents when developing materials communicating or educating local stakeholders to help increase awareness and resilience to coastal risks. It also appears that respondents who have lived in the coastal community for longer periods of time are more 49 involved. Understandably, a personal connection to the lake, either by proximity or experiences, may be an explanation for involvement in programs or organizations that prioritize coastal habitat stewardship. That the odds of involvement are also higher as self-reported knowledge factors increase supports the primary tenets of Value-Belief-Norms Theory, suggesting that those who care more about coastal risk and protecting the shoreline will engage in behaviors that align with these values (P. C. Stern, 2000; P. Stern et al., 1999; Stern, Paul C., Dietz, 1994). In other words, it makes sense that people who have coastal environmental values will seek out more information and therefore self-report high knowledge scores about coastal risk. Many participants also mentioned concerns about development and insufficient government action to mitigate coastal risk as a main motivation for involvement. This suggests that there are opportunities for municipal officials and natural resource managers to focus on residents who are concerned about development or lax regulations. Results portend motivations for involvement in programs or organizations that prioritize coastal resilience are stemming from residents’ development and environmental concerns, as well as their internal sense of obligation to protect their coastlines. Programs that speak to this sense of connection and responsibility and that also provide residents with actual methods to mitigate coastal risk are more likely to be successful and organizations that simply try to provide residents with more information. In the case of coastal risk on Lake Michigan, it is likely that people who are already aware of the risks and are concerned are also seeking out knowledge and engaging in resilient behaviors. In all counties, Land Residents are less involved in programs and organizations when compared to Lake Residents. However, ownership time and year-round resident status are only factors that predict involvement in Allegan County, while previous experiences as a predictive 50 factor is only seen in Ottawa County. This suggests that resiliency programs need to be specifically tailored to different counties, because motivation for stewardship is different across counties. 51 Limitations CHAPTER 5: CONCLUSIONS The goal of this study was to better understand coastal stakeholders’ perceptions of risk and motivations for coastal habitat stewardship, however it should be noted that the sample population only consisted of residents willing to participate in the survey. Furthermore, the survey could only be completed online, which may have been an obstacle for some interested residents. Although residents had the opportunity to take the survey over the phone if they preferred, this maybe still have dissuaded some people from participating in the research. As the survey results were anonymous, I could not conduct a non-response survey to investigate potential biases in the sample population. This was an unfortunate side effect of choosing to attempt to reach a larger number of potential participants from multiple communities and counties. However, the survey results should only be confidently described for the socio- demographic sample population and one should proceed with caution when generalizing the results to the coastal region as a whole. Furthermore, Allegan and Ottawa Counties were surveyed in December 2017, directly following a mid-term election which may have made certain issues more salient. In addition, potential municipal official and staff turnover may have also impacted the response rates and responses of this Resident Type. It should also be noted that the survey was conducted during a period of relatively higher water levels and coastal risk perceptions may vary during periods of lower water levels. I also only focused on a few Lake Michigan communities, but coastal shorelines and physical dynamics vary in other Great Lakes. As such, preliminary studies such as this one should be first conducted for the other Great Lakes before making management decisions. It should also be noted that although “climate change” or 52 “Combined warming” was never explicitly mentioned in the survey, some survey respondents still mentioned negative comments related to their opinions regarding environmental issues. Recommendations This research provides insight into the risk perceptions and motivations of Lake Michigan coastal stakeholders. This information is useful to municipal officials and government agencies that hope to target educational programs for a variety of coastal residents to increase overall community resilience. Municipal officials in the specific counties can better understand what their constituents believe and how their beliefs shape their decisions about coastal zone management. While state government organizations can better tailor their educational materials to different Resident Types or communities of varying knowledge about coastal risk. This research also adds to a growing body of literature regarding environmental issues and risk perception, specifically the RISP framework. Future research could investigate what motivates coastal stakeholders to become involved in a program or organization that prioritizes coastal resilience. As a few respondents mentioned interest in helping beyond participating in the survey, there may be further opportunities to discuss motivations for sustainable management and community needs. 53 APPENDICES 54 APPENDIX A: INSTITUTIONAL REVIEW BOARD APPROVAL LETTER 55 APPENDIX B: SURVEY INVITATION NUMBERS AND RESPONSE RATES Table 18. Survey invitation numbers and response rates for Pooled Counties. Invitations mailed Responses Response rate % Lake Residents 602 168 27.9% Near-lake Residents 1,816 255 14.0% Inland Residents 6,002 501 8.3% Municipal official responses (n=56) Table 19. Survey invitation numbers and response rates for Allegan County. Invitations mailed Responses Response rate % Lake Residents 203 54 26.6% Near-lake Residents 387 49 12.7% Inland Residents 1750 151 8.6% Municipal official responses (n=7) Table 20. Survey invitation numbers and response rates for Ottawa County. Invitations mailed Responses Response rate % Lake Residents 350 88 25.1% Municipal official responses (n=41) Inland Residents 2,750 226 8.2% Near-lake Residents 1,078 135 12.5% Table 21. Survey invitation numbers and response rates for Muskegon County. Invitations mailed Responses Response rate % Lake Residents 49 21 42.9% Near-lake Residents 351 70 19.9% Inland Residents 1502 119 7.9% Municipal official responses (n=8) 56 Total 8,420 924 11.0% Total 2,340 254 10.9% Total 4,178 449 10.7% Total 1,902 210 11.0% APPENDIX C: QUALTRICS SURVEY Start of Block: Informed Consent Q1 - INFORMED CONSENT - What is the purpose of this research study? The purpose of this study is to understand Lake Michigan community members’ coastal risk perceptions and motivations for coastal habitat stewardship. We hope to understand what community members think, why they have these perceptions, and the implications for coastal community resiliency. Why am I being asked to participate? You are being asked to participate because you are a resident of a Lake Michigan coastal community and therefore have valuable perspectives and experiences that are important to this study. You are being asked to participate in an online survey, with multiple choice and open-ended questions, that takes about 15 minutes or less to complete. Are there any benefits to participating? You will not benefit directly from participating in this research study. However, this study will be used to inform topics of interest for future educational materials that can be more closely tailored to community members’ needs. Are there any risks associated with participation? We believe the risks associated with your participation in this research study are low, but may be psychological or social because you will be asked to reflect upon your perceptions and experiences as a coastal community member. However, please be assured that your participation in this research study is completely voluntary and the survey results are anonymous. If at anytime you do not wish to answer a question or you wish to discontinue the survey there are no economic or social repercussions. Are there any costs or compensation for this study? There is no costs to your for participating in this study other than your time to complete the interview. You will not be compensated for participating in this study. What if I have a question? This research study (STUDY100001557) was approved by the MSU Institutional Review Board on October 22, 2018. If you have concerns or questions about this study, such as scientific issues, how to do any part of it, or to report an injury, please contact the researcher Heather Triezenberg via email vanden64@msu.edu or phone (517)-353-5508, or regular mail Michigan Sea Grant/MSU Extension/Fisheries & Wildlife Department, 1405 S. Harrison Road, Suite 305, East Lansing, MI 48824. If you have questions or concerns about your role and rights as a research participant, would like to obtain information or offer input, or would like to register a complaint about this study, you may contact, anonymously if you wish, the Michigan State University's Human Research Protection Program at 517-355-2180, Fax 517-432-4503, or e-mail irb@msu.edu or regular mail at 4000 Collins Road, Suite 136, Lansing, MI 48910. By completing this survey, I agree to participate in this evaluation study and confirm that I am 18 years or older. COMMUNITY is defined as your local Lake Michigan coastal city, township, or village in Michigan, where you own property. End of Block: Informed Consent 57 Start of Block: Access code Q2 Please enter your access code, provided at the top of your address label. ________________________________________________________________ End of Block: Access code Start of Block: Areas of concern Q3 Please indicate how concerned you are about the following risks to your community in the next TEN years. (Please choose one option per statement.) Moderately concerned Extremely concerned Slightly concerned Not at all concerned concerned Very Unsure (15) (1) (3) (4) (10) (11) (3) More More frequent precipitation (2) precipitation as rain than snow and severe storms (1) Increases in o o o o o o temperatures (7) o Reduced ice cover on the Great Lakes (4) More flooding events (5) Coastal erosion Increases in extreme (6) o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o 58 Q4 Please indicate how concerned you are about the following risks to your community in the next FIFTY years. (Please choose one option per statement.) Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned (1) (2) (3) (4) (5) (3) More More frequent precipitation as rain than snow and severe storms (1) Increases in o precipitation (2) o o o o o o Reduced ice cover on the Great Lakes (4) More flooding events (5) Increases in extreme temperatures Coastal erosion (6) (7) o o o o o o o o o o o o o o o o o o o o o o o o o o o o Unsure (8) o o o o o o o Q5 Are there other environmental coastal issues you believe are a risk to your community? (Please type answer below.) ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ 59 one option.) o Not at all serious (10) o Slightly serious (4) o Moderately serious (5) o Very serious (6) o Extremely serious (7) o Unsure (8) Q7 How much do you think coastal risks will harm your community? (Please choose one option.) Not at all (1) Slightly (3) Moderately (4) Very much so (5) Extremely so (2) Unsure (6) End of Block: Areas of concern Start of Block: Risk Perception Q6 How serious of a threat are environmental coastal risks to your community? (Please choose How much do you think coastal risks will harm your community in the next TEN years? (1) How much do you think coastal risks will harm your community in the next FIFTY years? (2) o o o o o o o o o o o o 60 End of Block: Risk Perception Start of Block: Concern/Knowledge/Behavioral Control Q8 The following statements are related to your perception of environmental coastal risks. Please indicate the degree to which you agree or disagree with the following statements. Neither agree nor disagree (3) Slightly agree (4) Strongly agree (5) o o o o o o o o o o o o o o o o o o o o o o o o Strongly disagree (1) fiction. (5) feel worried. (1) I understand My family expects that I know something Information about coastal risks makes me information about coastal risks. (4) When it comes to information abut coastal risks, I can separate facts from o o o about coastal risks. (6) o about coastal risks. (7) o o o o I can easily locate information about coastal risks. (8) My friends expect that I am personally able to take action to manage coastal risks. (9) There are policies in place that allow my community to manage coastal risks. (11) I know something Slightly disagree (2) o o o o o o o o End of Block: Concern/Knowledge/Behavioral Control Start of Block: Motivations for Stewardship 61 Q9 Please indicate your level of concern for the Great Lakes coastal region. (Please choose one option per question.) Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned (4) (5) How concerned are you about the health of your community's coastal shoreline? (1) How concerned are you that coastal risks could affect private property (eg. your home, your land, etc.)? (2) How concerned are you that coastal risks could affect public spaces (eg. public parks, schools, etc.)? (3) (1) (2) (3) o o o o o o o o o o o o o o o Unsure (6) o o o 62 Q10 How much do you think human activities (eg. economic development, recreational activities, etc.) will increase coastal risks? (Please choose one option.) o Not at all (1) o Slightly (2) o Moderately (3) o Very much so (4) o Extremely so (5) o Unsure (6) Q11 Have you ever been involved in a program or organization whose primary goal was Great Lakes coastal region management, conservation, or preservation? (Please mark all that apply.) ▢ ▢ ▢ ▢ ▢ ▢ ▢ ▢ No (2) Unsure (3) Yes, dues paying member (4) Yes, volunteer (8) Yes, followed on social media (eg. Facebook, Instagram, Twitter, etc.) (5) Yes, donated money (6) Yes, attended a meeting or event (7) Other (1) ________________________________________________ 63 Q12 If yes, what program(s) or organization(s) have you been involved with? (Please type answer below.) ________________________________________________________________ Q13 If yes, what motivated you to become involved with the program(s) or organization(s)? (Please type answer below.) ________________________________________________________________ End of Block: Motivations for Stewardship Start of Block: Previous Experience Q14 Has your community experienced any of the following events in the last 5 years? Please indicate yes or no. If yes, how many times? Yes (1) No (2) # (1) 64 o o o o o o o Severe coastal flooding event (1) Severe inland flooding event (5) Coastal erosion event (2) Extreme winter storm event (3) Other (4) Other (6) Other (7) o o o o o o o End of Block: Previous Experience Start of Block: Owns shoreline 65 that apply.) ▢ ▢ ▢ ▢ ▢ ▢ ▢ ▢ ▢ The public (9) Private property that I own (1) Private property owned by another (2) Local government (3) State government (4) Federal government (5) Non-governmental organization (8) Other (6) ________________________________________________ Q15 Please identify who you think owns coastal shoreline in your community. (Please mark all Unsure (7) End of Block: Owns shoreline Start of Block: Responsible for shoreline 66 Q16 Who do you think is responsible for managing your community's shoreline? (Please mark You (1) Your Neighbor (2) Local Government (3) State Government (4) Federal Government (5) all that apply.) ▢ ▢ ▢ ▢ ▢ ▢ ▢ ▢ Non-governmental organization (8) Other (6) ________________________________________________ Unsure (7) End of Block: Responsible for shoreline Start of Block: Manage shoreline Q17 Which is the best way to manage a receding shoreline? (Please choose one option.) o Man-made, physical barrier (eg. seawall, etc.) (15) o Natural shoreline (eg. plants, etc.) (16) o About equal (20) o Neither (17) o Other (18) ________________________________________________ o Unsure (19) 67 End of Block: Manage shoreline Start of Block: Resident Q18 Are you a year-round resident, part-time resident, or visitor of this community? (Please choose one option.) o Year-round resident (1) o Part-time resident (eg. snowbird) (2) o Visitor (eg. renter, tourist) (4) o Other (3) ________________________________________________ Q19 What type of resident do you most identify with in your Lake Michigan community? (Please choose one option.) o Lake-front property (1) o Private beach access, but not lake-front property (2) o Coastal community resident, but no private beach access or lake-front property (3) o Other (4) ________________________________________________ 68 Q20 How long have you or your family owned a property in your Lake Michigan community? (Please choose one option.) o 5 years or less (1) o 6-10 years (2) o 11-20 years (3) o 21-49 years (4) o 50 years or more (5) End of Block: Resident Start of Block: Socio-Demographic Characteristics Q21 Are you of Hispanic, Latinx, or Spanish origin? o Yes (1) o No (2) 69 Q22 What is your race? (Please mark all that apply.) American Indian or Alaskan Native (1) Arab or Middle Eastern (2) Asian or South Asian (3) ▢ ▢ ▢ ▢ ▢ ▢ White (Non-Hispanic) (6) ▢ Black or African American (4) Native Hawaiian or Pacific Islander (5) Other (7) ________________________________________________ Q23 What is your highest level of education? (Please choose one.) o High school or less (1) o Some college (2) o Associate's degree (3) o Bachelor's degree (4) o Master's degree (5) o Professional degree (6) o Doctorate degree (7) 70 Q24 What is your annual household income before taxes? (Please choose one.) o Less than $20,000 (1) o $20,000-$34,999 (2) o $35,000-$49,999 (3) o $50,000-$74,999 (4) o $75,000-$99,999 (5) o Over $100,000 (6) Q25 What is your zip code of your Lake Michigan community property? (Please type below.) ________________________________________________________________ Q26 What is your birth year? (Please type below.) ________________________________________________________________ Q27 What is your gender? o Male (1) o Female (2) o Other (4) ________________________________________________ End of Block: Socio-Demographic Characteristics Start of Block: NEP 71 Q28 New Ecological Paradigm. This scale is used to assess people's basic beliefs about humanity's relationship with nature. Please indicate the degree to which you agree or disagree with the following statements. 72 Strongly disagree (1) Slightly disagree (3) Neither agree nor disagree (4) Slightly agree (5) Strongly agree (8) We are approaching the limit of the number of people the earth can support. (1) Humans have the right to modify the natural environment to suit their needs. (2) When humans interfere with nature it often produces disastrous consequences. (3) Human intelligence will ensure that we do NOT make the earth unlivable. (4) Humans are severely abusing the environment. (5) The earth has plenty of natural resources if we just learn how to develop them. (6) Plants and animals have as much right as humans to exist. (7) The balance of nature is strong enough to cope with the impacts of modern industrial nations. (8) Despite our special abilities humans are still subject to the laws of nature. (9) o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o 73 The so-called "ecological crisis" facing humankind has been greatly exaggerated. (10) The earth is like a spaceship with very limited room and resources. (11) Humans were meant to rule over the rest of nature. (12) The balance of nature is very delicate and easily upset. (13) If things continue on their present course, we will soon experience a major ecological catastrophe. (14) End of Block: NEP o o o o o o o o o o o o o o o o o o o o o o o o o Start of Block: Anything else? Q29 Is there anything else you would like us to know? (Please type answer below.) ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ ________________________________________________________________ End of Block: Anything else? 74 Response options Proportion (n) Table 22. Codebook. APPENDIX D: CODEBOOK Variable name id_new id_orig q2 Survey question New ID Original ID Access code county County loc Resident Type storm10 How concerned are you about more frequent and severe storms in the next 10 years? 75 - - - Allegan County Ottawa County Muskegon County Unknown Lake Resident Inland Resident Land Resident Municipal Official Other Unknown Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Code - - - 1 2 3 97 1 2 3 4 96 97 1 2 3 4 5 98 99 - - - 26.04% (262) 49.80% (501) 22.07% (222) 2.09% (21) 16.70% (168) 25.35% (255) 49.80% (501) 5.67% (57) .89% (9) 1.59% (16) 17.40% (175) 19.38% (195) 26.44% (266) 22.07% (222) 12.62% (127) 0.80% (8) 1.29% (13) Table 22 (cont’d) Variable name Survey question precip10 How concerned are you about increases in precipitation in the next 10 years? rainsnow10 How concerned are you about more precipitation as rain than snow in the next 10 years? icecov10 How concerned are you about reduced ice cover on the Great Lakes in the next 10 years? flood10 How concerned are you about more flooding events in the next 10 years? 76 Response options Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Code 1 2 3 4 5 98 99 1 2 3 4 5 98 99 1 2 3 4 5 98 99 1 2 3 4 5 98 99 Proportion (n) 20.52% (206) 21.22% (213) 29.68% (298) 16.93% (170) 8.86% (89) 1.59% (16) 1.20% (12) 18.54% (186) 20.34% (204) 27.42% (275) 18.44% (185) 10.97% (110) 3.49% (35) 0.80% (8) 15.84% (159) 14.64% (147) 25.10% (252) 24.20% (243) 17.03% (171) 2.39% (24) 0.80% (8) 13.77% (138) 18.36% (184) 24.05% (241) 24.85% (249) 17.07% (171) 1.10% (11) 0.80% (8) Table 22 (cont’d) Variable name Survey question eros10 How concerned are you about coastal erosion in the next 10 years? temp10 How concerned are you about increases in extreme temperatures in the next 10 years? storm50 How concerned are you about more frequent and severe storms in the next 50 years? precip50 How concerned are you about increases in precipitation in the next 50 years? 77 Response options Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Code 1 2 3 4 5 98 99 1 2 3 4 5 98 99 1 2 3 4 5 98 99 1 2 3 4 5 98 99 Proportion (n) 4.49% (45) 9.77% (98) 19.94% (200) 27.22% (273) 36.89% (370) 0.70% (7) 1.00% (10) 13.36% (134) 15.25% (153) 23.33% (234) 21.24% (213) 24.33% (244) 1.69% (17) 0.80% (8) 14.89% (149) 14.59% (146) 21.08% (211) 19.78% (198) 26.57% (266) 1.70% (17) 1.40% (14) 16.08% (161) 15.18% (152) 23.18% (232) 18.78% (188) 22.28% (223) 3.10% (31) 1.40% (14) Table 22 (cont’d) Variable name Survey question rainsnow50 How concerned are you about more precipitation as rain than snow in the next 50 years? icecov50 How concerned are you about reduced ice cover on the Great Lakes in the next 50 years? flood50 How concerned are you about more flooding events in the next 50 years? eros50 How concerned are you about coastal erosion in the next 50 years? 78 Response options Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Code 1 2 3 4 5 98 99 1 2 3 4 5 98 99 1 2 3 4 5 98 99 1 2 3 4 5 98 99 Proportion (n) 15.18% (152) 15.48% (155) 21.08% (211) 19.98% (200) 22.68% (227) 4.10% (41) 1.50% (15) 13.99% (140) 12.39% (124) 18.88% (189) 20.18% (202) 30.17% (302) 3.20% (32) 1.20% (12) 11.88% (119) 12.77% (128) 19.96% (200) 21.46% (215) 30.64% (307) 1.90% (19) 1.40% (14) 5.69% (57) 9.08% (91) 15.57% (156) 21.76 (218) 44.71% (448) 2.10% (21) 1.10% (11) Table 22 (cont’d) Variable name Survey question temp50 How concerned are you about increases in extreme temperatures in the next 50 years? oth1050 Are there other environmental issues you believe are a risk to your community? serious How serious of a threat are coastal risks to your community? harm10 How much do you think coastal risks will harm your community in 10 years? 79 Response options Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing [Open-ended question] Not at all serious Slightly serious Somewhat serious Very serious Extremely serious Unsure Missing Not at all Slightly Moderately Very much so Extremely so Unsure Missing Code 1 2 3 4 5 98 99 - 1 2 3 4 5 98 99 1 2 3 4 5 98 99 Proportion (n) 12.42% (124) 12.93% (129) 16.73% (167) 18.64% (186) 35.47% (354) 2.51% (25) 1.30% (13) - 5.52% (53) 14.57% (140) 27.16% (261) 30.28% (291) 18.21% (175) 2.60% (25) 1.66% (16) 7.37% (71) 20.33% (196) 29.88% (288) 26.97% (260) 10.17% (98) 3.73% (36) 1.56% (15) Table 22 (cont’d) Variable name Survey question harm50 How much do you think coastal risks will harm your community in 50 years? dread Information about coastal risks makes me feel worried. unstd I understand information about coastal risks. factfict When it comes to information about coastal risks, I can separate fact from fiction. 80 Response options Not at all Slightly Moderately Very much so Extremely so Unsure Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly agree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly agree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly agree Missing Code 1 2 3 4 5 98 99 1 2 3 4 5 99 1 2 3 4 5 99 1 2 3 4 5 99 Proportion (n) 5.09% (49) 10.07% (97) 17.96% (173) 27.00% (260) 32.09% (309) 6.02% (58) 1.77% (17) 11.98% (115) 8.44% (81) 22.29% (214) 33.96% (326) 20.31% (195) 3.02% (29) 1.87% (18) 6.76% (65) 13.53% (130) 43.29% (416) 31.43% (302) 3.12% (30) 1.46% (14) 6.35% (61) 15.10% (145) 36.88% (354) 37.19% (357) 3.02% (29) Table 22 (cont’d) Variable name Survey question fam My family expects that I know something about coastal risks. fr My friends expect that I know something about coastal risks. infoloc I can easily locate information about coastal risks. permang I am personally able to take action to manage coastal risks. 81 Response options Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly agree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly agree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly agree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly agree Missing Code 1 2 3 4 5 99 1 2 3 4 5 99 1 2 3 4 5 99 1 2 3 4 5 99 Proportion (n) 4.38% (42) 6.36% (61) 30.76% (295) 29.51% (283) 25.86% (248) 3.13% (30) 4.49% (43) 5.85% (56) 32.25% (309) 35.28% (338) 18.89% (181) 3.24% (31) 6.46% (62) 18.02% (173) 21.77% (209) 35.00% (336) 15.63% (150) 3.13% (30) 20.17% (194) 25.47% (245) 25.78% (248) 20.69% (199) 4.78% (46) 3.12% (30) Table 22 (cont’d) Variable name Survey question policies There are policies in place that allow my community to manage coastal risks. health How concerned are you about the health of your community’s coastal shoreline? prv How concerned are you that coastal risks could affect private property (eg. your home, your land, etc.)? pub How concerned are you that coastal risks could affect public spaces (eg. public parks, schools, streets, etc.)? 82 Response options Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly agree Missing Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Not at all concerned Slightly concerned Moderately concerned Very concerned Extremely concerned Unsure Missing Code 1 2 3 4 5 99 1 2 3 4 5 98 99 1 2 3 4 5 98 99 1 2 3 4 5 98 99 Proportion (n) 17.40% (167) 20.31% (195) 38.54% (370) 18.02% (173) 2.50% (24) 3.23% (31) 2.90% (28) 9.54% (92) 20.02% (193) 34.44% (332) 28.84% (278) 0.62% (6) 3.63% (35) 13.41% (129) 17.78% (171) 22.45% (216) 20.48% (197) 21.31% (205) 1.04% (10) 3.53% (34) 5.09% (49) 11.84% (114) 22.43% (216) 28.66% (276) 27.41% (264) 0.93% (9) 3.63% (35) Table 22 (cont’d) Variable name Survey question humact How much do you think human activities (eg. economic development, recreational activities, etc.) will increase coastal risks? inv prgorg_txt motivate_txt cflood_exp iflood_exp eros_exp storm_exp Have you ever been involved in a program or organization whose primary goal was Great Lakes coastal region management, conservation, or preservation? If yes, what program(s) or organization(s)? If yes, what motivated you to become involved with the program(s) or organization(s)? Has your community experienced coastal flooding events in the last 5 years? Has your community experienced inland flooding events in the last 5 years? Has your community experienced coastal erosion events in the last 5 years? Has your community experienced extreme winter storm events in the last 5 years? 83 Response options Not at all Slightly Moderately Very much so Extremely so Unsure Missing Yes No Other Unsure Missing [Open-ended response] [Open-ended response] Yes No Yes No Yes No Yes No Code 1 2 3 4 5 98 99 1 0 96 98 99 - - 1 0 1 0 1 0 1 0 Proportion (n) 5.29% (51) 15.15% (146) 19.71% (190) 29.25% (282) 25.31% (244) 1.66% (16) 3.63% (35) 32.90% (307) 60.56% (565) 0.21% (2) 3.22% (30) 3.11% (29) - - 14.20% (115) 85.80% (695) 32.85% (273) 67.15% (558) 65.88% (583) 34.12% (302) 60.38% (512) 39.62% (336) Table 22 (cont’d) Variable name Survey question own Please identify who you think owns coastal shoreline in your community. Mark all that apply. resp Who do you think is responsible for managing your community’s coastal shoreline? Mark all that apply. mang_shore Which is a better way to manage a receding shoreline? res_time Are you a full-time resident, part-time resident, or visitor of your Lake Michigan community? 84 Response options Private Government Public Mix Other Unsure Missing Private Government Public Mix Other Unsure Missing Man-made physical barrier (eg. seawall, etc.) Natural shoreline (eg. plants, etc.) About equal Neither Other Unsure Missing Year-round resident Part-time resident Visitor Other Missing Code 1 2 3 4 96 98 99 1 2 3 4 96 98 98 1 2 3 4 96 98 99 1 2 3 96 99 Proportion (n) 11.72% (113) 39.63% (382) 1.87% (18) 37.03% (357) 0.83% (8) 3.42% (33) 5.50% (53) 3.73% (36) 40.98% (395) 1.76% (17) 45.02% (434) 0.41 (4) 3.01% (29) 5.08% (49) 8.09% (78) 34.44% (332) 28.11% 5.08% (49) 8.09% (78) 11.51% (111) 4.67% (45) 70.02% (675) 21.58% (208) 1.04% (10) 2.90% (28) 4.46% (43) Table 22 (cont’d) Variable name Survey question res_loc What type of resident do you most identify with in this community? own_time How long have you or your family owned property in the community where you received this survey? hisp Are you of Spanish, Hispanic, or Latinx origin? race What is your race? Mark all that apply. Response options Lake-resident Private beach access, but not lake-front property Coastal community resident, but no private beach access or lake-front property Other Missing 5 years or less 6-10 years 11-20 years 21-49 years 50 years or more 5 years or less Yes No Missing American Indian or Alaskan Native Arab or Middle Eastern Asian or South Asian Black or African American Native Hawaiian or Pacific Islander Mixed race White (non-Hispanic) Other Missing 85 Code 1 2 Proportion (n) 24.69% (238) 13.90% (134) 3 96 99 1 2 3 4 5 99 1 0 99 NA 1 2 3 NA 4 5 96 99 51.14% (493) 5.08% (49) 5.19% (50) 15.15% (146) 11.83% (114) 16.80% (162) 28.63% (276) 22.51% (217) 5.08% (49) 96.36% (874) 1.54% (14) 2.09% (19) 0.0% (0) 0.21% (2) 0.10% (1) 0.21% (2) 0.0% (0) 1.24% (12) 85.79% (827) 3.11% (30) 9.34% (90) Table 22 (cont’d) Variable name Survey question educ What is your highest level of education? inc What is your annual household income before taxes? zip birthyr Lake Michigan community zip code? What is your birth year? gend What is your gender? nep1 We are approaching the limit of the number of people the earth can support. 86 Response options High school or less Some college Associate's Bachelor's Master's Professional's Doctorate's Missing Less than $20,000 $20,000 to $34,000 $35,000 to $49,000 $50,000 to $74,000 $75,000 to $99,000 More than $100,000 Missing [Open-ended response] [Open-ended response] Male Female Other Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing Code 1 2 3 4 5 6 7 99 1 2 3 4 5 6 99 - - 1 2 96 99 1 2 3 4 5 99 Proportion (n) 2.29% (21) 9.68% (89) 6.09% (56) 36.56% (336) 26.12% (240) 9.47% (87) 8.27% (76) 1.52% (14) 1.30% (11) 4.38% (37) 7.10% (60) 15.15% (128) 16.80% (142) 51.36% (434) 3.91% (33) - - 56.48% (536) 37.83% (359) 0.42% (4) 5.27% (50) 12.97% (125) 13.38% (129) 23.13% (223) 2614% (252) 18.46% (178) 5.91% (57) Table 22 (cont’d) Variable name Survey question nep2 Humans have the right to modify the natural environment to suit their needs. nep3 When humans interfere with nature it often produces disastrous consequences. nep4 Human intelligence will ensure that we do NOT make the earth unlivable. nep5 Humans are severely abusing the environment. 87 Response options Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing Code 1 2 3 4 5 99 1 2 3 4 5 99 1 2 3 4 5 99 1 2 3 4 5 99 Proportion (n) 20.75% (200) 30.91% (298) 15.87% (153) 21.58% (208) 4.77% (46) 6.12% (59) 4.88% (47) 8.09% (78) 13.17% (127) 36.10% (348) 31.85% (307) 5.91% (57) 15.87% (153) 26.53% (254) 22.61% (218) 21.58% (208) 6.95% (67) 6.64% (64) 6.54% (63) 9.13% (88) 7.68% (74) 28.53% (275) 42.12% (406) 6.02% (58) Table 22 (cont’d) Variable name Survey question nep6 The earth has plenty of natural resources if we just learn how to develop them. nep7 Plants and animals have as much right as humans to exist. nep8 The balance of nature is strong enough to cope with the impacts of modern industrial nations. nep9 Despite our special abilities humans are still subject to the laws of nature. 88 Response options Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing Code 1 2 3 4 5 99 1 2 3 4 5 99 1 2 3 4 5 99 1 2 3 4 5 99 Proportion (n) 12.24% (118) 18.36% (177) 17.95% (173) 29.56% (258) 15.04% (145) 6.85% (66) 6.12% (59) 7.88% (76) 12.86% (124) 23.44% (226) 43.46% (419) 6.22% (60) 39.83% (384) 30.91% (298) 11.62% (112) 8.20% (79) 3.01% (29) 6.43% (62) 1.56% (15) 1.45% (14) 4.77% (46) 29.88% (288) 56.12% (541) 6.22 (60) Table 22 (cont’d) Variable name Survey question nep10 The so-called “ecological crisis” facing humankind has been greatly exaggerated. nep11 The earth is like a spaceship with very limited room and resources. nep12 Humans were meant to rule over the rest of nature. nep13 The balance of nature is very delicate and easily upset. 89 Response options Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing Code 1 2 3 4 5 99 1 2 3 4 5 99 1 2 3 4 5 99 1 2 3 4 5 99 Proportion (n) 43.26 (417) 17.53% (169) 11.93% (115) 13.17% (127) 7.68% (74) 6.43% (62) 7.99% (77) 14.32% (138) 22.72% (219) 27.49% (265) 20.64% (199) 6.85% (66) 37.03% (357) 18.15% (175) 18.78% (181) 11.93% (115) 7.37% (71) 6.74% (65) 3.11% (30) 10.58% (102) 11.93% (115) 34.13% (329) 34.13% (329) 6.12% (59) Proportion (n) 8.30% (80) 8.71% (84) 14.73% (142) 28.01% (270) 34.13% (329) 6.12% (59) - Response options Strongly disagree Slightly disagree Neither agree nor disagree Slightly agree Strongly disagree Missing [Open-ended question] Code 1 2 3 4 5 99 - Table 22 (cont’d) Variable name Survey question nep14 If things continue on their present course, we will soon experience a major ecological catastrophe. last_txt Is there anything else you would like us to know? 90 APPENDIX E: SOCIO-DEMOGRAPHIC SURVEY DATA AND CENSUS DATA Census data comes from the American Community Survey (5-year estimates, 2013-2017) conducted by the U.S. Census Bureau. Table 23. Socio-demographic survey data compared to census data for gender. Male Female Other General Survey 59.9% 39.7% 0.4% General Census 49.6% 50.4% - Table 24. Socio-demographic survey data compared to census data for gender by county. Allegan Survey Allegan Census Ottawa Survey Ottawa Census Male Female Other 63.6% 36.4% 0.0% 49.9% 50.1% - 61.7% 38.2% 0.0% 49.3% 50.7% - 91 Muskegon Survey 52.2% 45.9% 1.9% Muskegon Census 49.7% 50.3% - Table 25. Socio-demographic survey data compared to census data for age. 18-24 25-34 35-44 45-64 65-84 85+ General Survey 0.7% 3.4% 6.9% 45.6% 41.6% 1.9% General Census 14.7% 15.8% 15.6% 34.6% 16.7% 2.5% Table 26. Socio-demographic survey data compared to census data for age by county. Allegan Survey Allegan Census Ottawa Survey Ottawa Census 18-24 25-34 35-44 45-64 65-84 85+ 0.4% 0.4% 4.5$ 51.6% 41.0% 2.0% 10.6% 15.3% 15.9% 38.0% 18.0% 2.1% 18.3% 15.5% 15.5% 32.5% 15.6% 2.7% 0.9% 4.0% 7.7% 42.8% 42.3% 2.3% 92 Muskegon Survey 0.5% 5.6% 8.1% 45.5% 39.9% 0.5% Muskegon Census 11.5% 16.6% 15.7% 35.9% 17.7% 2.6% Table 27. Socio-demographic survey data compared to census data for highest level of education. High school or less Some college or Associate’s degree Bachelor's degree Graduate or professional degree General Survey 2.2% 16.2% 37.4% 44.2% General Census 41.6% 35.0% 16.2% 7.2% Table 28. Socio-demographic survey data compared to census data for highest level of education by county. Ottawa Survey 2.3% 15.5% 35.4% 46.8% Ottawa Census 36.2% 35.3% 19.8% 8.7% Muskegon Survey 3.8% 22.6% 37.5% 36.1% Muskegon Census 45.9% 37.1% 11.8% 5.1% High school or less Some college or Associate’s degree Bachelor's degree Graduate or professional degree Allegan Survey 0.7% 12.3% 41.3% 45.6% Allegan Census 48.2% 31.3% 14.0% 6.5% 93 Table 29. Socio-demographic survey data compared to census data for race. American Indian or Alaskan Native Arab or Middle Eastern Asian or South Asian Black or African American Native Hawaiian or Pacific Islander White (Non-Hispanic) Two or more races Other General Survey 0.0% 0.2% 0.1% 0.2% 0.0% 94.4% 1.5% 3.5% General Census 0.5% - 1.6% 5.1% 0.0% 88.0% 3.0% 1.8% Table 30. Socio-demographic survey data compared to census data for race by county. American Indian or Alaskan Native Arab or Middle Eastern Asian or South Asian Black or African American Native Hawaiian or Pacific Islander White (Non-Hispanic) Two or more races Other Allegan Survey 0.0% 0.0% 0.0% 0.4% 0.0% 94.7% 2.0% 2.8% Allegan Census 0.5% - 0.6% 1.4% 0.02% 94.4% 2.3% 0.9% Ottawa Survey 0.0% 0.2% 0.0% 0.0% 0.0% 95.6% 0.9% 3.3% 94 Ottawa Census 0.3% - 2.7% 1.4% 0.0% 90.0% 2.5% 3.0% Muskegon Survey 0.0% 0.5% 0.5% 0.5% 0.0% 91.0% 2.5% 5.0% Muskegon Census 0.6% - 0.5% 13.6% 0.0% 80.7% 4.1% 0.5% Table 31. Socio-demographic survey data compared to census data for annual income before taxes. Less than $20,000 $20,000-$34,999 $35,000-$49,999 $50,000-$74,999 $75,000-$99,999 More than $100,000 General Survey 1.4% 4.5% 7.2% 15.9% 17.6% 53.5% General Census 19.5% 9.8% 14.5% 21.1% 13.5% 21.7% Table 32. Socio-demographic survey data compared to census data for annual income before taxes by county. Less than $20,000 $20,000-$34,999 $35,000-$49,999 $50,000-$74,999 $75,000-$99,999 More than $100,000 Allegan Survey 0.5% 2.7% 6.7% 11.2% 12.6% 66.4% Allegan Census 17.8% 9.2% 15.2% 21.9% 14.9% 21.0% Ottawa Survey 1.4% 4.2% 5.3% 15.0% 20.1% 53.9% Ottawa Census 15.7% 8.6% 13.5% 21.0% 14.8% 26.3% Muskegon Survey 2.6% 7.4% 11.6% 23.8% 18.0% 36.5% Muskegon Census 26.2% 11.9% 15.5% 20.8% 10.4% 15.2% 95 APPENDIX F: ALLEGAN AND OTTAWA COUNTY RESIDENT RECRUITMENT MATERIALS 96 97 98 99 APPENDIX G: MUSKEGON COUNTY RESIDENT RECRUITMENT MATERIALS 100 101 102 103 APPENDIX H: MUNICIPAL OFFICIALS EXAMPLE RECRUITMENT MATERIALS 104 REFERENCES 105 REFERENCES Arbogast, A. F., Hansen, E. C., & Van Oort, M. D. (2002). Reconstructing the geomorphic evolution of large coastal dunes along the southeastern shore of Lake Michigan. Geomorphology, 46(3–4), 241–255. https://doi.org/10.1016/S0169-555X(02)00076-4 Bai, X., Wang, J., Sellinger, C., Clites, A., & Assel, R. (2012). Interannual variability of Great Lakes ice cover and its relationship to NAO and ENSO. Journal of Geophysical Research: Oceans, 117(3). https://doi.org/10.1029/2010JC006932 Byun, K., & Hamlet, A. F. (2018, January 11). Projected changes in future climate over the Midwest and Great Lakes region using downscaled CMIP5 ensembles. https://doi.org/10.1002/joc.5388 Crossett, K., Ache, B., Pacheco, P., & Haber, K. (2013). National Coastal Population Report: Population Trends from 1970 to 2020 (pp. 1–22). Retrieved from National Oceanic and Atmospheric Administration website: https://aamboceanservice.blob.core.windows.net/oceanservice-prod/facts/coastal- population-report.pdf Cutter, S. L., Barnes, L., Berry, M., Burton, C., Evans, E., Tate, E., & Webb, J. (2008). A place- based model for understanding community resilience to natural disasters. Global Environmental Change, 18(4), 598–606. https://doi.org/10.1016/j.gloenvcha.2008.07.013 Dake, K. (1991). Orienting dispositions in the perception of risk: An analysis of contemporary worldviews and cultural biases. Journal of Cross-Cultural Psychology, 22(1), 61–82. Dietz, T., Dan, A., & Shwom, R. (2007). Support for Climate Change Policy: Social Psychological and Social Structural Influences*. Rural Sociology, 72(2), 185–214. https://doi.org/10.1526/003601107781170026 Douglas, M., & Wildavsky, A. (1983). Risk and culture: An essay on the selection of technological and environmental dangers. University of California Press. Feltman, B. C., Norris, P. E., & Batanian, J. L. (2017). Gauging concerns about climate change in Great Lakes coastal communities. Journal of Great Lakes Research, 43(4), 670–677. https://doi.org/10.1016/J.JGLR.2017.05.003 Fischhoff, B., Slovic, P., Lichtenstein, S., Read, S., & Combs, B. (1978). How safe is safe enough? A psychometric study of attitudes towards technological risks and benefits. Policy Sciences, 9(2), 127–152. https://doi.org/10.1007/BF00143739 Folke, C. (2006). Resilience: The emergence of a perspective for social-ecological systems analyses. Global Environmental Change, 16(3), 253–267. https://doi.org/10.1016/j.gloenvcha.2006.04.002 106 Folke, C., Biggs, R., Norstrom, A., Reyers, B., & Rockstrom, J. (2016). Social-ecological resilience and biosphere-based sustainability science. Ecology and Society, 21(3). https://doi.org/10.5751/ES-08748-210341 Gifford, R., & Comeau, L. A. (2011). Message framing influences perceived climage change competence, engagement, and behavioral intentions. Global Environmental Change, 21, 1301–1307. The Great Lakes Integrated Sciences and Assessments Center (GLISA). (2014). Climate Change in the Great Lakes Region. Retrieved from http://glisa.umich.edu/media/files/GLISA_climate_change_summary.pdf. The Great Lakes Integrated Sciences and Assessments Center (GLISA). (2017). 2017 Annual Climate Trends and Impacts Summary for the Great Lakes Basin. Retrieved from http://glisa.umich.edu/media/files/2017-Climate-trends-and-impacts-summary.pdf. Gronewold, A. D., Fortin, V., Lofgren, B., Clites, A., Stow, C. A., & Quinn, F. (2013). Coasts, water levels, and climate change: A Great Lakes perspective. Climatic Change, 120(4), 697–711. https://doi.org/10.1007/s10584-013-0840-2 Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006). Multivariate data analysis (Vol. 6). Upper Saddle River, NJ: Pearson. Hayhoe, K., VanDorn, J., Croley, T., Schlegal, N., & Wuebbles, D. (2010). Regional climate change projections for Chicago and the US Great Lakes. Journal of Great Lakes Research, 36, 7–21. https://doi.org/10.1016/j.jglr.2010.03.012 Ho, M.-C., Shaw, D., Lin, S., & Chiu, Y.-C. (2008). How Do Disaster Characteristics Influence Risk Perception? Risk Analysis, 28(3), 635–643. https://doi.org/10.1111/j.1539- 6924.2008.01040.x Kellens, W., Terpstra, T., & De Maeyer, P. (2013). Perception and Communication of Flood Risks: A Systematic Review of Empirical Research. Risk Analysis, 33(1), 24–49. https://doi.org/10.1111/j.1539-6924.2012.01844.x Kettle, N. P., & Dow, K. (2014). The Role of Perceived Risk, Uncertainty, and Trust on Coastal Climate Change Adaptation Planning. Environment and Behavior, 48(4), 579–606. https://doi.org/10.1177/0013916514551049 Leiserowitz, A. A. (2005). American risk perceptions: Is climate change dangerous? Risk Analysis, 25(6), 1433–1442. https://doi.org/10.1111/j.1540-6261.2005.00690.x Lewicka, M. (2011). Place attachment: How far have we come in the last 40 years? Journal of Environmental Psychology, 31(3), 207–230. https://doi.org/10.1016/j.jenvp.2010.10.001. Lin, S., Shaw, D., & Ho, M. C. (2008). Why are flood and landslide victims less willing to take mitigation measures than the public? Natural Hazards, 44(2), 305–314. https://doi.org/10.1007/s11069-007-9136-z. 107 Lorenzoni, I., & Pidgeon, N. F. (2006). Public views on climate change: European and USA perspectives. Climatic Change, 77(1–2), 73–95. https://doi.org/10.1007/s10584-006- 9072-z. Lovis, W. A., Arbogast, A. F., & Monaghan, G.W. (2012). The Geoarchaeology of Lake Michigan Coastal Dunes (J. A. Robertson, Ed.). East Lansing, MI: Michigan State University Press. Mason, L. A., Riseng, C. M., Gronewold, A. D., Rutherford, E. S., Wang, J., Clites, A., … McIntyre, P. B. (2016). Fine-scale spatial variation in ice cover and surface temperature trends across the surface of the Laurentian Great Lakes. Climatic Change, 138(1–2), 71– 83. https://doi.org/10.1007/s10584-016-1721-2 Miles, M., & Huberman, A. M. (1994). Qualtiative Data Analysis (2nd ed.). Thousand Oaks: Sage Publications. Norton, R. K., David, N. P., Buckman, S., & Koman, P. D. (2018a). Overlooking the coast: Limited local planning for coastal area management along Michigan’s Great Lakes. Land Use Policy, 71, 183–203. https://doi.org/10.1016/J.LANDUSEPOL.2017.11.049 Norton, R. K., David, N. P., Buckman, S., & Koman, P. D. (2018b). Overlooking the coast: Limited local planning for coastal area management along Michigan’s Great Lakes. Land Use Policy, 71, 183–203. https://doi.org/10.1016/J.LANDUSEPOL.2017.11.049 Norton, R. K., Meadows, L. A., & Meadows, G. A. (2011a). Drawing lines in law books and on sandy beaches: Marking ordinary highwater on michigan’s great lakes shorelines under the public trust doctrine. Coastal Management, 39(2), 133–157. https://doi.org/10.1080/08920753.2010.540709 Norton, R. K., Meadows, L. A., & Meadows, G. A. (2011b). Drawing lines in law books and on sandy beaches: Marking ordinary highwater on michigan’s great lakes shorelines under the public trust doctrine. Coastal Management, 39(2), 133–157. https://doi.org/10.1080/08920753.2010.540709 Peterson, J. M., & Dersch, E. (1981). A Guide To Sand Dune and Coastal Ecosystem Functional Relationships. Michigan State University Cooperative Extension Service. Portman, M. E. (2016). Current Issues: Coastal Adaptation to Climate Change. In Environmental Planning for Oceans and Coasts: Methods, Tools, and Technologies (pp. 191–210). https://doi.org/10.1007/978-3-319-26971-9_11. Pryor, Sara C., Donald Scavia, Charles Downer, Marc Gaden, Louis Iverson, Rolf Nordstrom, Jonathan Patz, and G. Phillip Robertson. (2014). Midwest. Climate change impacts in the United States: The third national climate assessment. In: Melillo, JM; Richmond, TC; Yohe, GW, eds. National Climate Assessment Report. Washington, DC: US Combined Change Research Program: 418-440., 418-440. 108 Semenza, J. C., Hall, D. E., Wilson, D. J., Bontempo, B. D., Sailor, D. J., & George, L. A. (2008). Public Perception of Climate Change. American Journal of Preventive Medicine, 35(5), 479–487. https://doi.org/10.1016/j.amepre.2008.08.020 Sjoberg, L. (2000). Factors in Risk Perception. Risk Analysis, 20(1), 1–11. https://doi.org/10.1111/0272-4332.00001 Slimak, M. W., & Dietz, T. (2006). Personal values, beliefs, and ecological risk perception. Risk Analysis. https://doi.org/10.1111/j.1539-6924.2006.00832.x Slovic, P. (1987). Perception of risk. Science, 236(4), 280+. Stern, P. C. (2000). New Environmental Theories: Toward a Coherent Theory of Environmentally Significant Behavior. Journal of Social Issues, 56(3), 407–424. https://doi.org/10.1111/0022-4537.00175 Stern, P., Dietz, T., Abel, T., Guagnano, G. A., Kalof, L., Human, S., … Winter, N. (1999). Society for Human Ecology A Value-Belief-Norm Theory of Support for Social Movements: The Case of Environmentalism Published by: Society for Human Ecology Stable URL : http://www.jstor.org/stable/24707060 A Value-Belief-Norm Theory of Support for Social. 6(2), 81–97. Stern, Paul C., Dietz, T. (1994). The value basis of environmental concern. Journal of Social Issues, 50, 65–84. Stults, M., & Larsen, L. (2018). Tackling Uncertainty in US Local Climate Adaptation Planning. Journal of Planning Education and Research. https://doi.org/10.1177/0739456X18769134 Tompkins, E. L., Adger, W. N., & Adger, N. W. (2004). Does Adaptive Management of Natural Resources Enhance Resilience to Climate Change ? Ecology and Society, 9(2), 10. Whitmarsh, L. (2008). Are flood victims more concerned about climate change than other people? The role of direct experience in risk perception and behavioural response. Journal of Risk Research, 11(3), 351–374. https://doi.org/10.1080/13669870701552235 U.S. Environmental Protection Agency. (2018). Great Lakes facts and figures. Retrieved from https://www.epa.gov/greatlakes/great-lakes-facts-and-figures. 109