Abstract
This paper presents an evaluation of local efforts to manage Great Lakes coastal shorelands through master plans, focusing on Michigan localities. We framed the analysis around the concepts of capacity, knowledge, and commitment. We conducted plan content evaluations, structured surveys of local officials, and multiple unstructured interviews of local officials and citizens through a participatory action research (PAR) program. We analyzed those data, along with census data, using descriptive statistics, correlations, regression analyses, and triangulation of observations. We found that Michigan’s coastal localities are largely failing to consider their coastal areas in their planning, or to adopt meaningful plan policies to manage them, for at least four reasons: damaging erosion and storm events have been relatively infrequent; localities rely on the state to address coastal issues; insurance programs effectively indemnify them when a storm does happen; and—to some extent—shoreland owners push back against proactive local management. To the extent localities are planning, higher overall plan quality is associated with having in-house planning staff (a measure of both capacity and knowledge) and development pressure (knowledge and commitment). To the extent plans address their coastal areas specifically, the adoption of plan policies advancing coastal area management is associated directly with having higher median house values (capacity), in-house planning staff (capacity and knowledge), and development pressure (knowledge and commitment). Focus on coastal management is inversely associated, however, with the use of planning consultants. Higher plan quality is correlated significantly with the adoption of more robust plan policies overall. In sum, having knowledge about coastal dynamics appears important in explaining local planning efforts, but having the capacity to act on that knowledge and the commitment to do so are equally or more important.
Keywords: Coastal area management, Plan content evaluation, Participant action research, Great Lakes
1. Introduction
Near-shore coastal zones are popular places to build. They are also subject to high-risk hazards like erosion, storm surges, and inundation. There are federal programs in the U.S. that influence development within coastal areas given those values and hazards, including programs emanating from the federal Coastal Zone Management Act of 1972 and the Disaster Mitigation Act of 2000. Despite these programs, however, local governments ultimately play the key role in managing where and how shoreland development takes place. In doing so, they act primarily through their state-enabled local master planning and development management programs (Beatley et al., 2002).
Paradoxically, while coastal communities face the greatest losses should a natural disaster occur, studies have shown that they mostly refrain from managing coastal zone development so as to reduce risk (Burby, 2006; Berke & Lyles, 2013). While these studies have been conducted in a variety of settings, those settings have included mostly ocean coastlines. We set out to evaluate whether this phenomenon is true for Great Lakes coastal communities as well.
In addition to its oceans, the United States enjoys inland freshwater seas, the five Laurentian Great Lakes: Superior, Michigan, Huron, Erie, and Ontario (Fig. 1).1 The lakes extend some 750 miles and cover a combined surface area of about 95,000 square miles, roughly the size of the United Kingdom (see generally Gronewold et al., 2013; US GLERL, 2017; MDEQ, 2017a; US EPA, 2017a, 2017b). The U.S. portion of Great Lakes shorelines, including connecting waters, totals 4530 miles, making it almost as long as its Pacific, Gulf of Mexico, and Atlantic coastlines combined (Gronewold et al., 2013).
Fig. 1.
The Great Lakes Basin, showing the lakes, connecting rivers, and adjacent states and provinces.
The Great Lakes are large enough to provide ocean-like amenities and to generate substantial hazards to coastlands. Great Lakes coasts are different from ocean coasts both physically and institutionally, however, as discussed more below. As a first step for evaluating systematically local efforts to manage Great Lakes coastal shorelands, we undertook a cross-sectional and longitudinal baseline assessment of those efforts. In doing so, we focused on the State of Michigan.
Michigan consists of two peninsulas, upper and lower, surrounded by waters of four of the five Great Lakes (Superior, Michigan, Huron, and Erie), with the smaller Lake St. Clair connecting Lakes Huron and Erie just north of Detroit. Virtually the entire state’s land area drains to the Great Lakes basin, and the state enjoys 3288 miles of Great Lakes shoreline—about 72 percent of the U.S. Great Lakes shoreline (GLIN, 2017). As with most states, Michigan has delegated most of its land development management authorities and responsibilities—including those for coastal areas—to its local governments through broad planning, zoning, and other enabling acts (Fisher et al., 2012; MDEQ, 2017a, 2017b).
This article presents an analysis of local efforts to plan for the management of Michigan’s Great Lakes coastal areas through master plans. We document the extent to which coastal localities are indeed incorporating coastal shoreland management into their local planning efforts, and then explain those outcomes. We first review the literatures on plan content analysis and coastal area management and present a conceptual framework for analyzing local coastal shoreland management generally. We then describe the physical attributes that make the Great Lakes unique and the institutional arrangements that structure planning by Michigan’s coastal localities. We report findings from the content analysis of local master plans conducted for selected localities in the mid 2000 s and again in the mid 2010s; a survey of local officials conducted in 2008; results from statistical analyses and modeling of those data; and findings from ongoing participant action research (PAR) efforts. We conclude with a brief discussion of the applicability of these findings from Michigan to other Great Lakes and ocean coastal settings more broadly.
2. Evaluating local planning by coastal communities
2.1. Conceptualizing and evaluating local planning for coastal management
The research questions addressed here include first documenting systematically the extent to which coastal localities in Michigan are addressing Great Lakes shoreland management through their master plans, and second identifying the key factors that explain their planning efforts. The study sits at the intersection of several distinct literatures, including well-established work on evaluating the quality and policy content of local master plans (e.g., Burby & May, 1997; Berke & Godschalk, 2009; Lyles & Stevens, 2014) and planning for coastal area management (e.g., Godschalk & Cousins, 1985; Beatley et al., 2002; Norton, 2005a, 2005b, 2005c; Tang, 2008; Beatley, 2009; Lloyd et al., 2013).
Following seminal work by Baer (1997) and others, scholars have moved beyond merely describing the plan-making process in order to evaluate the quality and policy content of plans actually produced, along with conceptualizing and evaluating the implementation of those plans (see, e.g., Berke et al., 2006; Norton, 2008; Berke & Godschalk, 2009; Lyles & Stevens, 2014; Lyles et al., 2016). Following seminal work of Berke et al. (1996), Burby and May (1997), and others, much of that early plan quality and plan implementation work was conducted in the context of natural hazards mitigation (see, e.g., Brody, 2003; Lyles et al., 2014b; Horney et al., 2017), although it has since expanded to address other contexts.
From that work taken altogether, consensus has formed around a general set of criteria that should be used to construct index measures of plan quality. Those criteria include the plan’s statement of issues and vision, its fact base, its policy content, and its implementation provisions. They also include documentation of the extent to which public participation was folded into the plan making process and assessment of the communicative efficacy (e.g., readability) of the plan document itself (Berke et al., 2006; Berke & Godschalk, 2009; Horney et al., 2017).
Beyond that basic consensus, different scholars have characterized and organized content evaluation criteria in different ways, depending on the purposes for which the plans under study were produced (e.g., analysis of hazard mitigation planning) or the purposes of the plan evaluation study itself (e.g., analysis of the relationships between plan quality and plan implementation). Berke et al. (2006), for example, distinguish between ‘internal dimensions’ of plan quality (fact base, goals, policies, implementation), and ‘external dimensions’ (conformance to unique local conditions and responsiveness to stakeholder needs and values). Alternatively, Horney et al. (2017) distinguish between ‘direction-setting principles’ (goals, fact base, policies) and ‘action-oriented principles’ (implementation, coordination, participation).
Similarly, some scholars have focused on the effects of state-level planning mandates or incentives on local plan quality and policy content (e.g., Burby & May, 1997; Pendall, 2001); others have focused on the communicative, argumentative, or persuasive attributes of plans, in conjunction with or in addition to the role played by mandates (e.g., Norton, 2008; Bunnell & Jepson, 2011); and yet others have addressed a variety of substantive policy goals, again either in conjunction with or addition to other factors. Examples of the latter include planning for sustainable development (e.g., Berke & Manta Conroy, 2000), improved environmental management (e.g., Brody & Highfield, 2005), smart growth (Edwards & Haines, 2007), and climate adaptation and mitigation (e.g., Basset & Shandas, 2010).
Given that general background, our approach was to evaluate systematically the content of local master plans following this now well-established work on conceptualizing and evaluating plan content just described. In doing so, we build most directly on work by Norton (2008) in several respects. Foremost, we distinguish between the concepts of plan quality and plan policy focus (see Fig. 2).
Fig. 2.
Two dimensions of plan content analysis, including plan quality and plan policy focus, showing attributes of each dimension measured for this study.
Plan quality speaks to the plan as an analytical document, or the extent to which a locality’s plan-making did the following: engaged the community through public participation processes; gathered information on relevant topics; conducted adequate and appropriate analyses of those topics; identified objectives and policies flowing from its stated community goals and analyses; and finally conveyed all of this through coherent and thorough presentation in the plan document itself. For the purposes of this study, and drawing from the plan evaluation and coastal area management literatures (e.g., Beatley et al., 2002; Norton, 2005a, 2005b, 2005c; Beatley, 2009), we focus both on this broader concept of overall plan quality and on a constituent or sub-type measure of coastal area management analysis, as described in the methods section below.
In contrast to plan quality, plan policy focus speaks to the extent to which a plan evinces a clear policy focus on a given policy area, such as coastal area management, by adopting an appropriate array of policies designed to advance that policy. If a locality seeks to advance its coastal area management efforts through its master planning, for example, it might adopt policies calling for the establishment of setbacks from high-risk erosion zones or high-value coastal habitat (Beatley, 2009). We measured coastal management policy focus by identifying a variety of policies that Great Lakes coastal localities might adopt and constructing index measures to assess the presence and strength of that policy focus. We also assessed plans for their policy focus on promoting vital urban centers, conserving rural areas, and managing water quality in order to situate the localities’ relative emphasis on coastal area management in comparison.
2.2. Explaining local planning for coastal area management
A well-developed body of work now offers a variety of theoretical frameworks for evaluating the factors that explain plan-making outcomes (see, e.g., May et al., 1996; Berke et al., 1996; Burby & May, 1997; Berke et al., 1999; Laurian et al., 2004; Norton, 2005a, 2005b; Brody et al., 2010). These various approaches have collectively identified explanations that can be characterized ultimately as speaking to three primary concepts—capacity, knowledge, and commitment—and we organize our conceptual framing around them. There are several challenges that arise in taking this approach, however.
The first difficulty is that researchers have identified a remarkably broad array of independent variables might help to explain the quality of a plan and/or its policy focus. Some are reflected by direct measures of a concept, such as assessment of local officials’ commitment to hazard mitigation provided through surveys (e.g., Berke et al., 1996). Others are reflected more by proxy measures, such as development pressure in high-hazard zones, which provides (among other things) a proxy for commitment to planning for hazard mitigation that comes with the need to regularly address development in those zones (e.g., May et al., 1996; Burby & May, 1997; Norton, 2005b). The former measure has the benefit of being direct, but it is subjective; the latter has the benefit of being objective, but it is a proxy.
A second difficulty is that researchers typically operationalize the various measures they employ in ways that confound and thus introduce measurement error into efforts to distinguish between the larger concepts of capacity, knowledge, and commitment, thereby making it harder to detect true relationships. Brody et al. (2010), for example, conceived and measured “organizational capacity” more broadly than merely administrative capacity, as typically conceived by others, by incorporating measures of commitment into that concept as well. At the same time, a third and related difficulty is that various potential measures of any one of these concepts might actually reflect more than just one of them. It is arguably the case, for example, that having professional in-house planning staff represents a good measure of both local administrative capacity to plan, as that variable is typically specified, and a good measure of at least some aspects of knowledge, given the expertise that in-house planners bring (e.g., Lyles et al., 2014a).
For this study, we framed our analysis in an attempt to distinguish between and provide conceptual clarity to these broader notions of capacity, knowledge, and commitment. We believe that each of these concepts is meaningfully distinct from the others and that it is useful to isolate them for analytical purposes accordingly. At the same time, we also recognize that at least some operational constructs for any one of those concepts might in fact measure different dimensions of multiple concepts. Fig. 3 illustrates our larger conceptual framework. We explain that overarching framework here and then operationalize it in our discussion of methods below.
Fig. 3.
Overarching conceptual model specified for this analysis, distinguishing between the primary concepts of capacity, knowledge and commitment and relating them to secondary concepts, along with key setting factors.
Capacity refers to the ability of a locality to acquire knowledge and to act upon it (see, e.g., Burby & May, 1997; Norton, 2005b; Brody et al., 2010). As we employ the concept here, it consists of at least three different dimensions, including enabled, financial, and technical capacity. Enabled capacity is a locality’s legal and policy authority—enabled by the state—to engage in planning and to adopt actions to advance its goals (see, e.g., Berke et al., 2006; Norton, 2011; Juergensmeyer and Roberts, 2013). It is important to account for this aspect of capacity because, while a locality may have the financial resources and technical capacity to develop plans for advancing various goals, it may lack the legal authority to do so. Or it may be constrained in doing so by state or federal law (e.g., through the preemption of its ability to adopt wet-lands regulations more stringent than parallel state regulations). While important to acknowledge, all local governments in Michigan have been similarly enabled to plan in a way that could incorporate coastal area management, such that enabled capacity has essentially been controlled for by the study’s research design.
Financial capacity encompasses the wealth of the community or, more relevant for local government, the aggregated real property values upon which the locality might impose local property taxes to fund its planning and other governmental functions. It speaks conceptually to the ability of the community to invest in efforts to study local planning problems and take steps necessary to address them (Berke et al., 1996; Norton, 2005b). In contrast, technical capacity can take the form of in-house planning staff, planning consultants engaged as needed, or some combination of both (see Loh & Norton, 2015). This form of capacity speaks to the locality’s ability to study, understand, and generate approaches for responding appropriately to planning problems as they arise.
In addition to capacity, we conceptualize knowledge to encompass two important dimensions. First, it speaks to the extent to which local officials and/or citizens understand the substantive aspects of the planning issue at hand (i.e., substantive knowledge), such as understanding what nonpoint source pollution is and how it affects water quality. Second, it also encompasses an understanding of the systems and processes that are employed to make public planning and policy decisions (i.e., procedural knowledge), such as knowing the reach and limits of local planning authorities, the uses and limitations of planning analyses like land suitability analysis, and the basic procedural steps and best practices localities follow when planning.
While clearly important to understanding why and how localities undertake planning efforts, the concept of ‘knowledge’ has generally not been characterized directly as such in the literature. Rather, it has been conceptualized and measured indirectly in several ways. One common approach has been to account for it in terms of experiences that raise awareness of a problem or threat, or ‘experiential knowledge.’ Burby and May (1997), for example, addressed experiential knowledge in terms of a locality’s experience with agenda-setting events like hurricanes. Experiential knowledge has also been characterized in terms of development pressure within high-risk settings, where higher levels of development generally—and in those settings particularly—increases local knowledge both of the nature of the risks at hand (substantive knowledge) and the role that planning can play in addressing them (procedural knowledge—see, e.g., Berke et al., 1996; Burby & May, 1997; Norton, 2005b).
Another secondary conceptualization speaks to ‘learned knowledge,’ or that knowledge gained by local officials and/or citizens through observation, study, or efforts like educational campaigns. Berke et al. (1996, pg. 86), for example, addressed this kind of knowledge in terms of the locality’s “understanding [of the] threat” through public educational campaigns and hazards disclosure requirements. A related but distinct secondary conceptualization of knowledge speaks to ‘technical knowledge,’ or knowledge gained through focused education or training, typically on the part of local officials. This knowledge is provided by professional planners, for example, who presumably have received extensive technical training through their education as planners. It can also come from specialized training programs undertaken by local officials like planning commissioners, such as training on the planning process or on special topics like coastal area management.2 Again, technical knowledge can encompass both substantive knowledge—issues and challenges related to coastal areas—and procedural knowledge—how local planning might address those issues.
Finally, commitment refers to the willingness to act that local public officials demonstrate toward planning as a governmental function and toward the use of planning to meaningfully address particular policy agendas, given their knowledge and capacity to act (May et al., 1996; Burby & May, 1997; Norton, 2005b). The concept is straightforward and tracks to the same distinction noted above regarding plan quality—that is, it includes commitment to substantive objectives, such as coastal area management, and commitment to the idea that planning and related activities are indeed processes that government appropriately engages to advance those objectives.
The need to account for commitment as distinct from capacity or knowledge is important because, while capacity and knowledge are both necessary for effective coastal area management, neither alone nor taken together is sufficient. Local officials may have adequate capacity (e.g., tax revenues) and knowledge (e.g., educated planning staff) to study, identify, and adopt effective coastal area management policies, for example, but they may not be inclined to do so for a variety of reasons, such as the fear of prompting litigation.
While straightforward in theory, the difficulty with the concept of commitment comes in measuring it, particularly in light of its relationship to knowledge and capacity, as noted above. It is not clear, for example, that simply having a better understanding of Great Lakes shoreline dynamics and coastal area hazards tends to prompt greater commitment toward managing coastal areas. There is some evidence that greater exposure to hazardous events themselves, along with higher development pressures within high-hazard areas—experiential knowledge—might prompt such commitment, but that appears to be in contrast to learned knowledge about the potential for such events (see, e.g., May et al., 1996; Norton, 2005b). Similarly, it is not clear to what extent a locality’s decision to engage more professional planners—a measure of technical capacity—might also reflect a higher degree of commitment to engaging in good planning in the first place, possibly confounding efforts to discern the extent to which having a trained planning staff indicates capacity, knowledge, or commitment, or all three (see, e.g., Loh and Norton, 2013, 2015; Lyles et al., 2014a).
Finally, there may be attributes of coastal localities that are beyond the ability of those localities to influence but that may substantially determine planning outcomes. For our analysis, we considered one such physical attribute—the amount of shoreland area subject to local management—and one institutional attribute—jurisdiction type. We also anticipated that while overall plan quality would not necessarily have a relationship to a plan’s policy focus generally, having a strong analysis of issues related to coastal area management in a plan might. That is, we framed the provision of analysis on development-related issues in coastal areas as a sub-type of overall plan quality, one reflecting both increased knowledge of coastal issues and increased commitment to planning for them. As such, we expected that higher plan quality in terms of coastal analysis specifically would be associated positively with a greater level of plan policy focus on coastal area management.
Before presenting how we operationalized this broad conceptual model for evaluating local planning in the Great Lakes, we first summarize briefly some key physical and institutional attributes of the Great Lakes and of Michigan that shaped our approach to doing so.
3. Michigan’s Great Lakes shores
3.1. Physical dynamics
While similar to oceans in many respects, several attributes of the Great Lakes’ physical system make their coasts unique compared to ocean coasts, making coastal area management along Great Lakes shores uniquely challenging (Norton et al., 2013). The Great Lakes are geologically young, having been created by the retreat of glaciers at the end of the last ice age some 10,000 years ago (Meadows et al., 1997). As a result, Great Lakes shores are mostly erodible sands and gravels, and substantial stretches of Great Lakes shoreline are eroding on average about one foot per year, particularly along the southern edges of the basin. In this respect, shoreline movement landward along Great Lakes coasts from erosion is similar to shoreline movement landward along the ocean coasts from climate-change-induced inundation as sea levels rise, posing similar challenges regarding coastal area management.
The Great Lakes are unique from the oceans, however, because of the ways in which standing lake water levels change over time. While the lakes are not large enough to experience diurnal tides, their standing water levels oscillate naturally on roughly seasonal, decadal, and multi-decadal timeframes as a result of changes in precipitation, evaporation, river outflow, and groundwater inflow (Keillor, 2003pp. 2–4). Since records have been kept for about the past century and a half, documented standing water levels for Lakes Michigan and Huron, hydrologically the same lake, have fluctuated up to six feet vertically (Gronewold et al., 2013).3 The time spans between trough and peak fluctuations occur about every five to 10 years. As a result, the incidence of lake and land on Great Lakes shores can vary horizontally up to several hundreds of feet for extended periods of time, depending on the standing water level at any given time and the slope of the near-shore terrain. Water level fluctuations are amplified by seasonal variations in precipitation and evaporation, along with short-term fluctuations caused by basin seiching (sloshing back and forth) during storms, which together can cause additional fluctuations by multiple feet on top of averaged standing water levels (Keillor, 2003pp. 4–5).
At least two important phenomena occur because of these dynamics. First, as standing water levels drop and stay low for long periods of time, the lakes push sand up onto the beach, resulting in wider and higher stretches of dry-sand beach during those extended low-water periods than would otherwise occur (Meadows et al., 1997; Komar, 1997). Second, because lake water levels inevitably rise following periods of low water, and given the intensity of coastal storms and the highly erodible composition of the low-water beaches, the wide and seemingly accreting beaches that appear during low-water periods inevitably prove to be ephemeral, quickly eroding away as water levels rise, sometimes during single storm events (Meadows et al., 1997).
Finally, fluctuations in Great Lakes water levels have always been difficult to predict (International Joint Commission, 2012), a challenge further complicated by global climate change. While several modeling studies released in the early 2000 s suggested that climate change might result in significant declines in Great Lakes water levels, more recent and sophisticated models indicate the potential for stable to slightly decreasing levels, or even higher levels, depending on the lake and the model scenario used (Gronewold et al., 2013; Angel, 2013; Angel & Kunkel, 2010). More importantly, no evidence suggests that lake level fluctuations will be dampened by climate change, even if long-term water levels decline. Indeed, variability on many scales appears to be increasing, including the frequency and severity of storms, suggesting that near shore coastal hazards on Great Lakes shores from seasonal and storm-induced flooding and wave action will continue to be significant, if not more so, given climate change.
3.2. Institutional dynamics
The institutional issues confronting Great Lakes states are much the same as those confronting the ocean coastal states. These stem from the fragmented structure of land and coastal resource management initiatives across levels of government, such as the Coastal Zone Management Program (CZMP) as it is structured and administered by federal National Oceanic and Atmospheric Administration (NOAA) and corresponding state and local agencies (Hershman et al., 1999; Beatley et al., 2002; Beatley, 2009). Institutional challenges stem as well from the extent to which land management authorities, including those addressing coastal zones, have been delegated by the states to their local governments throughout the U.S. (Platt, 2014; Juergensmeyer & Roberts, 2013).
The first thing to note about Great Lakes coastal area planning in Michigan is that the State of Michigan’s federally approved CZMP does not mandate that Great Lakes coastal communities engage in shoreland area management planning, nor does it require that coastal area management be addressed by local master plans (Wuycheck, 2017).4 Similarly, the state’s CZMP does not establish coastal area management policies with which localities must be consistent through their local planning. Nor does it review local master plans to determine consistency with state policies, or review state policies to ensure that they are consistent with local master plans (Meadows et al., 1997).
The second thing to note is that the State of Michigan has adopted a state-wide Michigan Hazard Mitigation Plan (MHMP) in compliance with the federal Disaster Mitigation Act of 2000 (MDSP, 2014; see Berke et al., 2012). That plan, however, merely documents generally the planning authorities that localities in Michigan enjoy, and it references the state’s CZMP to describe efforts that localities can currently take to address Great Lakes coastal hazards. It does not establish freestanding requirements for advancing local efforts to address coastal hazards specifically, criteria for evaluating those efforts, or an administrative process for doing so.5
These federal-state programs aside, the State of Michigan holds title to lands submerged by the Great Lakes through its public trust doctrine (Slade et al., 1997; Norton et al., 2011). Under that doctrine and its general police powers, it has established regulatory programs that address the dredge and fill of submerged lands, along with programs that regulate development activities within state-designated high-risk erosion areas, coastal dunes, and sensitive environmental areas (see MDEQ, 2017b). These areas are spatially limited, however; something less than 10% of Michigan’s coastline, for example, is regulated through the state’s high-risk erosion area program. Coastal localities may choose to adopt ordinances to administer some of those regulations voluntarily, but few do, as discussed more below. Finally, while the Michigan CZMP provides grant funding to support local planning by Great Lakes coastal communities, it neither mandates that specific policies be adopted by localities receiving funds nor evaluates the plans produced for policy content, and it has not issued formal guidelines to direct local planning efforts (Wuycheck, 2017). As a result of these institutional arrangements, any Great Lakes coastal area planning activities undertaken by coastal localities in Michigan are undertaken through their own initiative under their general planning authorities, not as a means of administering state or federal coastal zone management or hazard mitigation policies.
Beyond that disconnect, an additional attribute of Michigan—and indeed all of the Great Lakes states—complicates yet further comprehensive and coordinated coastal area management: local hyper-fragmentation. All eight of the Great Lakes states (Minnesota, Wisconsin, Illinois, Indiana, Michigan, Ohio, Pennsylvania, and New York) are so-called “civil-township” states (however labeled by a particular state, see Platt, 2014). The civil township is a general-purpose unit of local government that exists between the county and municipality. All Great Lakes states are subdivided into civil townships entirely or within their coastal regions. In contrast, only three of the 23 ocean coastal states—New York, New Jersey, and Washington—are thus organized (Meadows et al., 1997).
The State of Michigan, for example, has 83 counties (4% of all local units of government), 273 incorporated cities (15%), and 262 unincorporated villages (14%), like much of the rest of the U.S. (CRC, 1999). Unlike those states, however, Michigan also has 1241 civil townships (67%—called townships in Michigan), for a total of 1859 units of local government statewide. Of these local governments, 318 touch Great Lakes waters or waterways that connect those lakes, similarly apportioned (see Appendix A). Like cities and villages, each township within a civil-township state resides within a county, but its jurisdictional territory is exclusive of adjacent city or village jurisdictional territories (CRC, 1999). As a result of this structure, local government—which is generally considered to be fragmented throughout the U.S. to begin with—can be thought of as hyper-fragmented in the 16 civil-township states in the U.S., including the eight Great Lakes states.
3.3. Unique challenges in planning for Great Lakes coastal management
In sum, over the long term and on average, Great Lakes coastal shorelines are highly dynamic much like ocean coastlines, presenting similar risks in terms of coastal hazards. In the short term when lake levels are relatively high and rising, the risks of substantial damage from storms, as well as shoreline loss to erosion, substantially increase. But those periods are interspersed with long periods of quiescence when lake levels are falling and low. To that extent, the need to address impending damage from high rates of erosion and threatening storms is correspondingly diminished. The unique challenge this phenomenon poses is that it encourages both shoreland property owners and local officials to become forgetful of the long-term water level fluctuations the lakes experience, along with the implications of those dynamics with regard to near-shore structures.6
Further complicating that challenge, most of the public management initiatives for Great Lakes coastal areas in Michigan are undertaken by numerous local governments, most of which are quite small (see Appendix A), all acting through their broadly enabled local master planning, zoning, and other land use management authorities but not responding to coastal management mandates or even strong incentives. All four types of Michigan’s local governments have been authorized to plan and zone, although counties generally do so only in lieu of townships and villages that choose not to do so (CRC, 1999; Fisher et al., 2012),7 such that townships, cities, and villages do most of the local planning and zoning. Evaluating Great Lakes coastal area management in Michigan therefore requires accounting for a diverse set of very numerous and typically small local governments.
4. Methods
4.1. Research design and sampling
We present here initial empirical findings from a longer-term and ongoing research program on planning for coastal area management by Michigan’s Great Lakes localities. We undertook our efforts in the early 2000s, and our work has evolved into an ongoing participatory action research (PAR) program involving researchers from multiple universities and a variety of state and local stakeholders.
For the work reported here, we began by first identifying Michigan localities that qualify as Great Lakes coastal localities (i.e., touching Great Lakes waters, including connecting rivers—see Appendix A). We then selected a stratified sample for initial data collection, focusing on areas with unique physical features across the lakes (i.e., different shoreline conditions). We then drew a random sample of coastal localities statewide for more in-depth study to augment our initial site selection. In making our initial sample selection, we adapted our strategy to the demographics of the state. Specifically, the Upper Peninsula in Michigan represents about one-third of the state’s total land area, similarly apportioned in terms of jurisdictions (28% of coastal jurisdictions are found in the Upper Peninsula), while the Lower Peninsula encompasses about two-thirds of the land area (72% of coastal jurisdictions). The vast majority of the state’s population and developed land areas, however, are found in the Lower Peninsula. For this reason, we weighted our initial identification of sample sites more toward Lower Peninsula jurisdictions in order to capture those areas where planning is both more needed and more likely occurring.
Using this sampling approach, our initial target sample included 139 coastal localities. Of those, we were able to collect sufficient data to conduct quantitative analyses for a final sample of 70 coastal localities (22% of the 318 coastal locality population8), as described more below. The final sample is relatively robust, including 7 counties (17% of all coastal counties), 45 townships (23% of coastal townships), 13 cities (23% of coastal cities), and five villages (21% of coastal villages). It is also roughly proportional by jurisdiction type as between the sample and coastal population (counties − 10% of sample and 13% of all coastal localities; townships − 64% of sample, 62% of all coastal; cities − 19% of sample, 18% of all coastal; villages − 7% of sample, 8% of all coastal). In terms of location, 6% of the final 70 sample sites are found in the Upper Peninsula, with 94% found in the Lower Peninsula.
Given these proportions, our sample is generally representative of coastal localities in terms of counts by jurisdiction type statewide, although they capture relatively more Lower than Upper Peninsula jurisdictions for the reasons noted above. We also conducted t-tests for differences in means across population density, housing unit density, and mean housing value by jurisdiction type to assess representativeness for analytical purposes. These tests confirmed that the sample is representative on those measures with regard to counties, cities, and villages, with no statistically significant differences in means for study sites versus non-study sites. However, the township study sites proved to be larger on all three measures compared to non-study sites to a statistically significant degree.9 That outcome reflects again the sampling strategy we employed, focusing on the more developed Lower Peninsula jurisdictions. It indicates that the statistical findings presented below are valid specifically for the more-developed portions of the state but may not be as robust for the less-developed portions, particularly with regard to townships in the Upper Peninsula.
Finally, in addition to conducting plan content evaluations, collecting data through structured surveys, and analyzing those data quantitatively, we have gained insight on local planning in coastal Michigan through an ongoing PAR program engaged since 2013. We undertook that program for the purpose of learning by doing in a community-engaged, pragmatic endeavor and to provide direct benefit to the non-researcher participants (see, e.g., Lake & Zitcer, 2012; Du Toit et al., 2016). Working with a multi-university and multi-disciplinary team of researchers, we have engaged stakeholder participants from a non-profit planning firm, staff with the Michigan Coastal Zone Management Program (MCZMP), and local officials from selected coastal communities, including primarily professional planning staff, planning commission members, and local legislative body members.
The practical goal of this PAR effort has been to develop cost-effective planning methods that small Great Lakes coastal localities might incorporate into their master plans using readily available data and analytical tools (e.g., basic GIS), while simultaneously studying how those methods are perceived and adopted by the study localities. Most of our efforts have involved collaborating with localities situated on Lake Michigan, including Hamlin Township, City of Ludington, Pere Marquette Township, Mason County, City of Grand Haven, Grand Haven Charter Township, and City of St. Joseph.
4.2. Data collection and cleaning
We first attempted to collect master plans of record from all of the 139 sample localities between roughly 2004 through 2007. We succeeded in obtaining plans from 60 localities (43% of the sample). Subsequent web searches indicated that, as of mid 2016, 90 of the 139-locality sample (65%) had adopted master plans. This suggests that, statewide, only about two-thirds of Michigan’s coastal localities are planning. We evaluated the 60 plans using coding protocols consistent with methods well established in the plan evaluation literature as described above (e.g., Norton, 2008; Tang, 2008; Berke & Godschalk, 2009; Stevens, 2013). Each plan was evaluated independently by at least two trained evaluators and each set of evaluations was then reconciled to produce a single set of scores for all evaluation items.10 Index scores were standardized on a scale of 0–10, consistent with general practice. Appendix B presents the indices used to evaluate plans and the items comprising them.
Following standard methods for surveying public officials (e.g., Dillman, 2000), we surveyed local officials from the study sample in 2008 to document a variety of local attributes related to planning, focusing our recruiting efforts on the localities for which we had successfully obtained a plan. We attempted to survey the most senior administrator (e.g., city manager), planning director, and planning commission chair from each locality, using the response from the most general respondent (i.e., in that order) when we received multiple responses from a given locality. Unfortunately, we did not receive sufficient multiple responses to discern any differences among respondents by position. We received full responses for 37 localities, including 32 for which we were also able to complete a plan evaluation. The survey covered several topics, including descriptive attributes of the locality (e.g., numbers of employees), characteristics of the legislative body and planning commission (e.g., whether council members were knowledgeable or had received training specifically on master planning), and the legislative body’s collective views on selected policies such as coastal area management, as perceived by the survey respondent.
To assess change in planning efforts over time, we conducted a second short survey of local officials with coastal localities in Michigan in 2014, focusing on our original study sites. We used the survey to determine which localities had adopted formal updates of their plans since our first-round assessment. We identified 12 localities that had updated their plans, then collected and coded those documents using the same process used for the first round of analysis.
As noted above, we were ultimately able to collect a plan, conduct a survey, and/or compile additional information from web-based sources (e.g., census data and basic geospatial data such as length of Great Lakes coastline within the jurisdiction) for a total of 70 jurisdictions. Sufficient data to conduct regression analyses were collected for 32 jurisdictions. Table 1 lists the data collected and analyzed for this article, including the variable label used for each, a description of the concept, and the data source(s).11
Table 1.
Variable labels, descriptors, concepts, and sources for selected data evaluated.
| Label | Descriptor | Concept | Source |
|---|---|---|---|
| Descriptive/Geographic | |||
| CO | County (categorical) | Type of local unit of government | Various |
| TWP | Civil township (categorical) | Type of local unit of government | Various |
| CTY | City (categorical) | Type of local unit of government | Various |
| VLG | Village (categorical) | Type of local unit of government | Various |
| UP | Upper Peninsula (categorical) | A unit of government located north of the Mackinac Straights | Various |
| LP | Lower Peninsula (categorical) | A unit of government located south of the Mackinac Straights | Various |
| LAKE | Lake (categorical) | Great Lake adjacent to which the locality is situated (Lakes Superior, Michigan, Huron, St. Clair, Erie) | Various |
| AREA | Total land area (sq. miles) | Total jurisdictional land area | U.S. Census Data |
| GLSA | Great Lake shoreland area (sq. miles) | Measure of GL shoreland area within jurisdictional boundaries subject to local management authority (i.e., not within state/federal parkland). Calculation: Approximate linear lake shoreline in miles * 0.1 mile (about 550 ft.) | Google maps measuring tool |
| %WTR | Proportion of jurisdiction under water | Measure of predominance of surface waters situated within locality’s jurisdictional boundaries | U.S. Census Data |
| Community Demographics | |||
| TOTPOP | Total population | Total year-round residents as a measure of community size demographically | U.S. Census 2010 |
| POPDNS | Population density (sq. miles) | Population per square mile as a relative measure of community size | U.S. Census 2010 |
| TOTHU | Total housing units | Total housing units as a measure of community buildout | U.S. Census 2010 |
| HUDNS | Housing density (sq. miles) | Housing units per square mile as a relative measure of community buildout | U.S. Census 2010 |
| MHV | Median home value | Median value of owner occupied homes as a measure of the taxable wealth (real property) of the community | U.S. Census 2010 |
| External Support for Planning and Coastal Area Management | |||
| CZM$ | Coastal zone management funding | Locality received funding support for a planning effort from the MI Coastal Zone Management Program (MCZMP) prior to 2007 | MCZMP staff |
| CPPT | Citizen planner training | One or more public officials from the locality participated in Michigan State University’s “Citizen Planner Program” training prior to 2007 | MSU CPP staff |
| Governmental Characteristics Related to Planning | |||
| LUAYR | Land use actions per year | Approximate number of land use actions addressed by the locality per year (e.g., rezonings, site plan approvals) | Survey of local officials (Survey) |
| LB# | Legislative body (size) | Number of members on local legislative body | Survey |
| PC# | Planning commission (size) | Number of members on planning commission | Survey |
| PZS# | Planning/zoning staff (size) | Number of in-house professional planning and zoning staff | Survey |
| PZS% | Planning/zoning staff (percent) | Proportion of local staff who are planning/zoning staff (0–1) | Survey |
| PCNSLT | Planning consultant (categorical) | Planning consultant used in preparation of master plan evaluated | Plan |
| PUBPTN | Public participation | Index measure of public participation efforts used in preparing master plan evaluated (0 = none, 10 = extensive) | Plan |
| LOTCAM% | Percent officials trained coastal | Percent local officials trained in coastal management (0–1) | Survey |
| LBCPLN Legislator commitment planning Legislative body commitment to planning (0 = low, 4 = high) Survey | |||
| Local Master Plan Attributes | |||
| PAGE | Plan age | Number of years since the plan was adopted on the date of evaluation, measuring currency of the plan document | Plan |
| PQUAL | Overall plan quality | Index measure of the overall analytical and presentational quality of the master plan evaluated (0–10, see appendix) | Plan |
| PQCAA | Plan quality — coastal area analysis | Index measure of the analytical quality of the master plan evaluated regarding coastal areas (0–10, see appendix) | Plan |
| PPFCAM | Plan policy focus — coastal area management | Index measure of the plan’s policy focus on coastal area management (0–10, see appendix) | Plan |
| PPFVUC | Plan policy focus — vital urban centers | Index measure of the plan’s policy focus on promoting vital urban areas (0–10, see appendix) | Plan |
| PPFCRA | Plan policy focus — conserved rural areas | Index measure of the plan’s policy focus on promoting conserved rural areas (0–10, see appendix) | Plan |
| PPFWM | Plan policy focus — water management | Index measure of the plan’s policy focus on promoting water quality management generally (0–10, see appendix) | Plan |
Finally, with regard to PAR component of our work, we made a variety of presentations to and conducted a variety of meetings with stakeholder participants on more than 40 separate occasions between February 2014 and February 2017 (see Appendix C). Through those meetings, we explored various concerns and issues raised by coastal residents and officials, discussed the usefulness of various planning approaches, and discussed possible explanations for local planning and coastal area management outcomes. Because these meetings were not structured as formal focus-group or interview-based meetings with standardized questions, and because of concern that formally recording the meetings by audio or video technology would discourage participation and honest feedback, researcher participants took notes on the key issues discussed and responses provided, but the meetings were not recorded electronically or transcribed.
4.3. Data analysis
Based on simple pairwise correlations, local adoption of a plan in the first place was not significantly associated with any factors that might explain that initial decision, given the quantitative data collected for this study. Similarly, given the conceptual framework modeled for this study in terms of quantitative analysis (i.e., designed primarily to explain variation in emphasis on coastal management), our assessment of why some localities are not addressing coastal management at all through their plans is based primarily on insights gained through the PAR component of our study. Analysis of the data collected through those efforts involved both triangulating the information collected across researcher participants and consulting with stakeholder participants to confirm the validity of the observations made and the conclusions drawn.
For the localities that have adopted plans, we used SAS version 9.4. (Cary, NC) to conduct statistical analyses in order to assess the factors that might explain higher levels of plan quality and stronger plan policies. Using the sample of 70 study sites, we performed basic descriptive statistics on study localities and findings from the content analyses, along with simple correlations and tests of means to gauge differences across locality types, geographic regions, and timeframes. After reviewing scatterplots and other descriptive statistics, we assumed a linear relationship between predictors and key outcome variables of interest. Because the sample was originally stratified by jurisdiction type, we used the SAS Proc SURVEYMEAN and SURVEYREG procedures to calculate means and to evaluate associations between those outcome, control, and potential explanatory variables. Those procedures require specifying inverse weightings. For the descriptive statistics of plan quality using Proc SURVEYMEAN, we used the following weightings (n = 70): 7/41 for counties, 44/195 for township, 13/57 for cities, and 6/24 for villages. For the regression analyses on all outcomes using Proc SURVEYREG, we used the following weightings (n = 32): 2/41 for counties, 25/195 for townships, 3/57 for cities, and 2/24 for villages. We performed sensitivity analyses using Proc REG and performed model diagnostics, which confirmed that Proc SURVEYREG was valid and the preferred technique for these analyses.
Fig. 4 presents the conceptual models specified for the statistical analyses, building on the overarching conceptual framework provided by Fig. 3 and integrating separate analyses to illustrate the larger conceptual whole. As discussed above, while we believe it is useful to conceptualize factors that might explain planning efforts in terms of the larger concepts of capacity, knowledge, and commitment, measuring those concepts empirically is challenging because of relationships between them and given the pros and cons of using objective versus subjective measures. Given those considerations, we attempted to develop and specify variable constructs that both effectively and efficiently measure the primary and secondary concepts of interest, indicating where a construct likely captures multiple concepts. In doing so, we hypothesized the most compelling relationships between concept and construct, recognizing the many overlapping relationships at play. For example, we expect that having in-house planning staff best measures local capacity and knowledge, recognizing that it may also reflect local commitment to planning to some extent. Given the very small sample size, we specified only simple OLS regressions using models as parsimonious as possible.
Fig. 4.
Operationalized conceptual model specified for this analysis, indicating variable constructs that measure multiple concepts.
Drawing from the extensive literature on plan content evaluation, and controlling for enabled capacity through the research design, we specified median house value as a measure of financial capacity (e.g., Berke et al., 1996). We similarly specified planning staff as a measure of both technical capacity and knowledge (i.e., both substantive and procedural knowledge), employing measures of both in-house staff and the use of planning consultants but not formally hypothesizing differences between them in terms of potential effect (see e.g., Berke et al., 1996; Brody et al., 2010; Lyles et al., 2014a).
We specified receipt of MCZMP funding support for planning-related activities prior to 2007, at the front end of our plan and survey data collection efforts, as a measure of both increased financial capacity and increased commitment to coastal area management, given the source of the support. We did not expect, however, that this funding would necessarily result in increased knowledge of coastal management issues, given that no substantive requirements or guidance accompanied it. We similarly specified voluntary participation in a ‘citizen planner’ training program for public officials as a measure of both increased procedural knowledge of planning, presuming the training was successful, and local commitment to the planning process, presuming that the dedication of officials’ time to that training indicated such commitment.
We were unable to collect data on development pressure within high hazard zones specifically, but we asked survey respondents to report the number of land use actions taken per year within their jurisdiction. Based on prior work suggesting that increased development pressure generally corresponds to both increased familiarity with planning issues (knowledge) and increased commitment to planning as a way to address those issues (see, e.g., Burby & May, 1997; Norton, 2005b), we specified land use actions per year as a proxy measure of both of those concepts. We similarly specified the use of one or more public participation techniques during plan making as a measure of both local knowledge of and commitment to planning, and we included a direct measure of local officials’ commitment to planning as perceived by survey respondents. Finally, as discussed above, we included two variables as controls, including the amount of shoreland area subject to local management and the type of local governmental unit (1 = county, 0 = other).
5. Findings
5.1. Describing planning outcomes
Table 2 presents basic demographic data for the 70 study sites, taking into account the survey design. Counties, which encompass the territories of all the other local jurisdiction types, also encompass the largest land areas, populations, and housing units. In general, coastal cities and villages have the highest population densities and housing unit densities, followed by townships.
Table 2.
Descriptive statistics by jurisdiction type for selected variables, organized by concept(s) measured by the variable and showing number of plans evaluated and surveys completed by jurisdiction type, including county (CO), township (TWP), city (CTY), and village (VLG).
| Variable | Concept | All Jurisdictions | Mean Values by Jurisdiction | ||||||
|---|---|---|---|---|---|---|---|---|---|
| CO | TWP | CTY | VLG | ||||||
| Plans evaluated n = 70 | 7 | 44 | 13 | 6 | |||||
| Surveys completed n = 32 | 2 | 25 | 3 | 2 | |||||
| Mean | Std Err | Min | Max | Mean | Mean | Mean | Mean | ||
| Plan Evaluation Outcomes | |||||||||
| PAGE | Plan age (years since adopted) | 8.2 | (4.8) | 1 | 23 | 5.5 | 8.6 | 8.1 | 11 |
| PQUAL | Overall plan quality (0–10) | 5.60 | (0.19) | 2.78 | 8.52 | 7.04 | 5.48 | 5.51 | 3.15 |
| PQCAA | Plan quality: coastal analysis (0–10) | 1.81 | (0.25) | 0 | 8.10 | 4.05 | 1.56 | 1.88 | 0.64 |
| PPFCAM | Policy focus: coastal management (0–10) | 0.61 | (0.19) | 0 | 6.25 | 0.63 | 0.64 | 0.45 | 0.85 |
| PPFVUC | Policy focus: vital urban centers (0–10) | 2.93 | (0.34) | 0 | 10 | 3.75 | 2.77 | 3.30 | 2.02 |
| PPFCRA | Policy focus: rural conservation (0–10) | 3.54 | (0.31) | 0 | 10 | 5.25 | 4.05 | 1.64 | 0.97 |
| PPFWM | Policy focus: water management (0–10) | 1.62 | (0.26) | 0 | 10 | 3.33 | 1.69 | 0.98 | 0 |
| Community Demographics/Controls | |||||||||
| AREA | Land area (sq. mi.) | 74.1 | 163.0 | 0.6 | 721.0 | 532.0 | 31.7 | 4.3 | 1.4 |
| GLSA | Great Lake shoreland area (sq.mi.) | 1.1 | (0.2) | 0 | 7.1 | 4.2 | 0.8 | 0.3 | 0.2 |
| %WTR | Proportion of jurisdiction water (0–1) | 0.22 | (0.03) | 0.01 | 0.94 | 0.52 | 0.16 | 0.22 | 0.19 |
| TOTHU | Total housing units (thousands) | 5.35 | (1.12) | 0.28 | 72 | 22.37 | 2.86 | 3.68 | 0.60 |
| HUDNS | Housing units per square mile | 262.8 | (29.6) | 5.80 | 1717 | 39.9 | 136.6 | 747.6 | 513.5 |
| Capacity (Financial) | |||||||||
| MHV | Median house value (thousands) | 190.6 | (11.8) | 77 | 760 | 155.5 | 194.9 | 155.0 | 297.8 |
| CZM$ | Jurisdiction received CZMP funding | 0.11 | (0.04) | 0 | 1 | 0.14 | 0.11 | 0.08 | 0.16 |
| Capacity (Technical) and Knowledge (Technical) | |||||||||
| PZS# | Planning zoning staff number | 3.4 | (0.55) | 0 | 18 | 11.0 | 2.8 | 3.0 | 1.5 |
| PZS% | Planning zoning staff proportion (0–1) | 0.29 | (0.05) | 0 | 1 | 0.07 | 0.34 | 0.04 | 0.42 |
| PCNSLT | Planning consultant used for plan | 0.73 | (0.06) | 0 | 1 | 1.00 | 0.76 | 0.55 | 0.68 |
| CPPT | Official(s) received citizen planner training | 0.32 | (0.06) | 0 | 1 | 0.14 | 0.39 | 0.38 | 0 |
| Knowledge (Experiential) and Commitment | |||||||||
| LUAYR | Land use actions undertaken per year | 16.09 | (2.28) | 0 | 57 | 13.00 | 15.69 | 25.25 | 7.00 |
| PUBPTN | Public participation in planning (0–10) | 4.99 | (0.45) | 0 | 10 | 8.13 | 5.00 | 4.09 | 2.50 |
| Commitment | |||||||||
| LBCPLN | Legislative body commitment planning (1–4) | 3.41 | (0.14) | 1 | 4 | 3.50 | 3.48 | 3.25 | 2.50 |
All of the median home values by jurisdiction type for the study localities were higher than the statewide average of $112,000 in 2010, likely reflecting the higher values that come with coastal property. The six villages sampled for this study enjoyed the highest median home value on average, $297,800. This finding is probably not representative of the rest of the state but appears to reflect unusually high property values in several of the coastal villages selected for study. Beyond that, townships enjoy somewhat higher median home values ($194,900) than those for cities ($155,000) and counties overall ($155,500—the mean for townships, cities, and villages combined). This reflects a pattern typical statewide and nationally, where cities have experienced a decline in residential property values as wealthier residents have moved out into the suburbs, located predominantly in townships in Michigan.
In terms of planning capacity and knowledge, and focusing on townships, cities, and villages, cities tend to have both the largest numbers of planning staff in house and the largest number of land use actions considered per year, followed by townships and then villages on both accounts. Townships, in contrast, tend to have the largest in-house planning staffs as a proportion of total governmental staff (reflecting the very small governmental staffs that townships tend to have in Michigan in general), and they also tend to use planning consultants more often, followed by villages and then cities.
As noted, we analyzed overall plan quality scores, comprised of separate plan quality attributes as shown in the appendix; plan quality for coastal area management analysis as a separate attribute; and policy focus using four focus outcomes. We also prepared sub-indices of overall plan quality following recent work in the plan evaluation literature to see if contemplating ‘internal’ versus ‘external’ (Berke et al., 2006) or ‘direction-setting’ versus ‘action-oriented’ (Horney et al., 2017) attributes of plan quality provided any additional insight with regard to plan-making outcomes, as discussed more below. All indices are reported as standardized scores on a scale of 0.0 (not present) to 10.0 (present, extensive, and well documented). For the initial plans evaluated (n = 70), we performed analysis of variance (ANOVA) tests to determine if any of the means were significantly different from each other by jurisdiction type. We observed no statistically significant differences at the 95% confidence level in mean overall plan quality scores (R2 = 0.10, p = 0.15) or coastal area management analysis scores (R2 = 0.126, p = 0.07).
The most striking finding here is that the standardized scores for coastal area management analysis in the plans evaluated is remarkably lower for all jurisdiction types save counties, scoring on average between 0 to less than 2, as compared to the overall plan quality scores, which range on average from about 4–7. Indeed, 25 of the 60 plans evaluated (42%) failed to provide any analysis of coastal area management issues at all. While all of the localities assessed for this study encompass Great Lakes waters and shorelands, they are almost uniformly overlooking their coastal areas in terms of analyzing opportunities for and constraints to development within those areas.
Regarding plan policy focus, the only statistically significant outcome among all of policy focus scores is the low score for conserved rural areas (CRA) for cities (m = 1.64, sd = 1.21). The mean value for this variable for villages was lower, but there were not enough observations to determine whether that value was statistically significant. Regarding the other government types, differences for CRA were observed (p = 0.0011) between counties and cities (3.61 quality units (95% CI 0.114, 7.113)) and between townships and cities (2.42 quality units (95% CI 0.359, 4.476)) based on Scheffe’s test. No differences were observed for vital urban centers (p = 0.778, R2 = 0.021), water management (p = 0.115, R2 = 0.109), or coastal area management (p = 0.97, R2 = 0.004).
Beyond those observations, it is noteworthy that the mean scores for all of the plan policy focus indices is uniformly low, with none scoring above 5 on a 10-point scale either by jurisdiction or by lake, except for counties’ focus on conserved rural areas (5.25). Moreover, while all of the policy scores are low, the scores for both water management and especially coastal areas management policy are notably lower than both vital urban centers and conserved rural areas. Indeed, mean scores for coastal area management policy were below 1 out of 10 for all jurisdiction types, and 44 out of the 60 plans evaluated (73%) failed to include any meaningful coastal area management policies in their plans at all. As with their lack of attention to coastal areas analytically, coastal localities in Michigan are largely overlooking management of their coastal shorelands through their plan policies.
Regarding relationships between plan quality and plan policy focus, all of the plan policy focus scores generally appear to be positively correlated with overall plan quality to a moderate and statistically significant extent, based on simple pairwise correlations (see Table 3). Similarly, all of the plan policy focus scores were correlated with the sub-index scores generated for internal, external, direction-setting, and action-oriented attributes of plan quality, and specification of those sub-indices provide no additional insight with regard to relationships between plan quality and policy focus for purposes here. Accordingly, we conducted all additional statistical analyses using the single measure of overall plan quality alone.
Table 3.
Pairwise correlations between selected measures of plan quality and plan policy focus outcomes, showing coefficients and p-values.
| Pearson Correlation Coefficient | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Prob > |r| under H0:Rho = 0 60 Observations | ||||||||||
| PQUAL | PQ INT | PQ EXT | PQ DIR | PQ ACT | PPF CAM | PPF VUC | PPF CRA | PPF WM | ||
| PQUAL | Overall plan quality | 1.00 | ||||||||
| PQ INT | PQ — Internal | 0.84 | 1.00 | |||||||
| 0.00 | ||||||||||
| PQ EXT | PQ - External | 0.90 | 0.64 | 1.00 | ||||||
| 0.00 | 0.00 | |||||||||
| PQ DIR | PQ — Direction setting | 0.93 | 0.76 | 0.76 | 1.00 | |||||
| 0.00 | 0.00 | 0.00 | ||||||||
| PQ ACT | PQ — Action oriented | 0.64 | 0.73 | 0.74 | 0.36 | 1.00 | ||||
| 0.00 | 0.00 | 0.00 | 0.00 | |||||||
| PPFCAM | Policy Focus — Coastal | 0.37 | 0.29 | 0.44 | 0.23 | 0.40 | 1.00 | |||
| (coastal area mgt) | 0.00 | 0.02 | 0.00 | 0.08 | 0.00 | |||||
| PPFVUC | Policy Focus — Urban | 0.40 | 0.41 | 0.42 | 0.27 | 0.49 | 0.43 | 1.00 | ||
| (vital urban centers) | 0.00 | 0.00 | 0.00 | 0.04 | 0.00 | 0.00 | ||||
| PPFCRA | Policy Focus — Rural | 0.43 | 0.49 | 0.44 | 0.34 | 0.53 | 0.22 | 0.34 | 1.00 | |
| (conserved rural area) | 0.00 | 0.00 | 0.00 | 0.01 | 0.00 | 0.08 | 0.01 | |||
| PPRWM | Policy Focus — Water | 0.48 | 0.51 | 0.48 | 0.34 | 0.56 | 0.50 | 0.31 | 0.31 | 1.00 |
| (water management) | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.02 | 0.02 | ||
Note: Plan quality measures for ‘internal’ and ‘external’ attributes were calculated consistent with those attributes as described by Berke et al. (2006). Measures for ‘direction-setting’ and ‘action-oriented’ attributes were calculated consistent with those attributes as described by Horney et al. (2017).
These findings taken altogether suggest that localities focusing on stronger policy development also tend to provide stronger analyses and justifications to support those policies, and that they tend to address a wide range of policy domains uniformly. Even so, we observed no significant correlation between the plan quality/coastal area management analysis scores with any of the policy focus scores. That suggests that even if localities develop higher quality plans analytically to support stronger urban and rural policies, they apparently do not do so in terms of analyzing coastal systems for the sake of supporting coastal area management policy.
In terms of measuring changes in planning over time, we observed somewhat more variability in coastal area management analysis than overall plan quality, but there was only a small but not statistically significant improvement from period 1 to period 2 in the quality of coastal area management analysis provided (p-value 0.35). We similarly observed no statistically significant changes in plan policy focus values between the two periods of observation for the plans that were updated. For these reasons, we concluded that planning efforts in coastal localities have not been changing substantially over the last decade, particularly with regard to coastal area management, perhaps reflecting the 5–8 years or more on average that localities take to update their plans. We therefore focused on the first round of plan content evaluations and local official surveys for the remainder of the analyses presented here.
Finally, comparing plan quality and policy focus across jurisdiction types, the descriptive findings presented here make sense. Counties generally scored higher on all accounts, perhaps reflecting that while counties in Michigan can engage in planning, they normally do not implement those plans through regulations like zoning, making it somewhat easier politically to conduct more rigorous analyses and promote more ambitious policies. Similarly, as would be expected, the more urban cities tend to adopt stronger policies promoting vital urban centers than do townships, while the more rural townships tend to adopt stronger policies promoting the conservation of rural areas than do cities. Townships, which generally have more greenfield land available for development than do cities, also appear to be focusing more on managing water quality overall.
5.2. Explaining limited planning for coastal management
Before explaining the outcomes of local planning efforts to the extent they are undertaking planning, it is important to explain why they are largely failing to address coastal shoreland area management through those efforts. Our findings here come primarily from our PAR efforts that have entailed repeated conversations with a diverse array of state and local coastal officials and citizens regarding issues surrounding Great Lakes shoreland development. From those ongoing conversations, four key reasons are regularly discussed and widely accepted as offering the best explanations.
First, given the nature of Great Lakes water level fluctuations, and especially the relatively low lake levels that existed during the late 1990s through the early 2010s (i.e., the period during with plans were collected and surveys conducted for this study), little damage was being done to shoreland properties from coastal storms. That lack of extensive shoreline storm damage appears to have resulted in very low levels of concern by coastal localities regarding hazards or other management needs and thus little commitment to undertaking their own shoreland area planning or development management initiatives (Fedewa, 2017; Howland, 2017).
Second, to the extent that local officials are contemplating potential coastal hazards, they appear to be relying largely on the state—and especially its shoreline regulatory programs—to address them. As discussed above regarding the state’s coastal zone management program, for example, coastal localities in Michigan have the option of adopting and administering the state’s high-risk erosion setback regulations. Only 10 of the 121 localities (8%) that have state-designated high-risk erosion areas have taken on that role since the early 1990s, however, and only two townships (less than 2%) currently do so. Feedback provided by local officials to state officials suggest that localities are not taking on this task primarily because they lack the staff and technical capacity to do so and they see no need to add to their capacity given the existence of state administrators (Lederle, 2017; Warner, 2017).
Third, even if coastal localities are inclined to undertake such initiatives, they encounter resistance (or fear they might encounter resistance) when doing so. In general, the wealthiest and most politically active residents of coastal communities tend to be near-shore property owners. As a group, these residents tend to raise the greatest pushback politically when localities attempt to adopt policies or regulations on shoreland development, especially regulations more stringent than those imposed by the state. For example, reflecting the interests and political influence at play, in 2012 Michigan’s legislature amended state law that regulates development within designated Great Lakes critical sand dunes areas (CDAs). Those amendments now compel coastal localities to permit certain development within CDAs, and they prohibit localities from adopting regulations more stringent than the state’s. The amendments were made in response to lobbying pressure from local realtor and homebuilder associations representing nearshore coastal property owners (Arbogast et al., 2015).
Finally, local officials report that current institutional arrangements for anticipating and mitigating potential damage from coastal storms effectively disincentivize them from adopting regulations that would minimize the costs of cleaning up and restoring damaged properties. Aside from relying on homeowners to secure private insurance, public officials anticipate that sufficiently damaging storms will prompt state if not federal disaster declarations and corresponding funds to cover cleanup costs. Indeed, officials from one study locality noted that they had been informed by municipal insurers of the availability of bridge loans, should a coastal storm occur, to cover the costs of cleanup before state and/or federal funds become available. Coastal localities, therefore, are able to secure the tax and other benefits of allowing near-shore coastal development to occur without incurring the liabilities for cleaning up after the storm (Knizacky, 2014).
In sum, while local citizens’ and officials’ knowledge of Great Lakes shorelines dynamics may be somewhat limited, residents—especially longer-term residents—generally appear to understand at least the basics of those dynamics. Rather than a lack of knowledge, therefore, the limited attention that coastal localities are paying to shoreland management through their planning appears to be more a function of limited technical capacity to act upon that knowledge and little commitment to doing so. Their limited commitment stems in turn from there being little pressing need within recent memory to address coastal hazards, a willingness to let the state take the lead in regulating near-shore development, current insurance-related incentives that effectively indemnify local governments should a severe storm occur, and some concern about shoreland owners’ potential objections to local regulation.
5.3. Explaining variation in local planning for coastal management
Despite these reasons for not addressing coastal area management through their plans in general, a substantial number of coastal localities do engage in local planning, and some of those localities do address coastal management through their plans, at least to some extent. Fig. 5 summarizes the findings from the regression analyses we conducted in terms of the conceptual model framed for this study, again integrating the separate analysis into a conceptual whole.
Fig. 5.
Findings from separate statistical analyses integrated to illustrate the conceptual model framed for this study. CZMP funding support and citizen planner training are not shown as potential explanatory variables because neither proved significant for any model specified.
Table 4 presents the results from the regression analysis conducted for overall plan quality. Recognizing the limitations of conducting regression analysis on such a small sample size, these findings suggest that higher quality plans tend to be significantly associated with, first, a relatively larger in-house planning staff and, second, having to address a relatively larger number of various land use actions per year. Consistent with the theoretical framing presented above, these findings suggest that higher plan quality is indeed positively associated with, in part, having higher levels of capacity for planning (technical if not financial), knowledge about planning issues and processes (both technical—in the form of easy access to planning experts—and experiential—in the form of regular experience with planning-related activities), and commitment to planning (in the form of commitment that comes with higher development pressure).
Table 4.
Results from regression analysis on overall plan quality.
| Regression on PQUAL — Overall plan quality (index scale from 0 to 10) | |||||
|---|---|---|---|---|---|
| R2 = 0.64 | |||||
| n = 28 | |||||
| Variable | Concept | Estimate | SE | t value | p valuec |
| CO (county) | Demographic/control | 0.42 | 1.25 | 0.335 | 0.740 |
| GLSA (GL shoreland area) | Demographic/control | 0.15 | 0.19 | 0.779 | 0.442 |
| MHVa (median house value) | Capacity (Financial) | 0.31 | 0.31 | 1.007 | 0.323 |
| PZS# (planning/zoning staff) | Capacity (Technical)/Knowledge (Technical) | 0.12 | 0.04 | 2.67 | 0.012 |
| PCNSLT (planning consultant) | Capacity (Technical)/Knowledge (Technical) | 0.44 | 0.59 | 0.746 | 0.462 |
| CPPT (citizen planner training) | Knowledge (Technical) | 0.72 | 0.70 | 1.025 | 0.314 |
| LUAYRb (land use actions/year) | Knowledge (Experiential)/Commitment | 0.06 | 0.01 | 4.372 | 0.000 |
| PUBPTN (public participation) | Knowledge (Experiential)/Commitment | 0.12 | 0.08 | 1.506 | 0.143 |
| LBCPLN (legislative commitment) | Commitment | − 0.26 | 0.22 | − 1.195 | 0.242 |
| Intercept | 4.75 | 0.85 | 5.60 | 0.000 | |
In thousands, estimate centered on the mean.
Estimate centered on the mean.
Outcomes with a p-value reflecting a confidence level of 90% or greater are in bold as noteworthy given the small sample size.
Whether for true lack of relationship or limited degrees of freedom, none of the other explanatory variables were statistically significant at the 90% confidence level. In addition, none of the models specified for analyzing plan quality in terms of coastal area analysis, including the final set of variables specified for the full array of models presented here, yielded statistically significant results.
Table 5 presents the results from the regression analysis conducted for plan policy focus on coastal area management, while Table 6 presents the results from the analyses of plan policy focus on vital urban centers, conserved rural areas, and water management. Regarding coastal area management, and consistent with plan quality, higher numbers of in-house planning staff and land use actions per year appear to be associated with a greater plan policy emphasis on coastal management, again reflecting higher levels of technical capacity, knowledge (technical and experiential), and commitment (from experience) as measured by those variables. In addition, higher levels of median home value also appear to be associated with a greater emphasis on coastal management. In contrast, having received funding support from the Michigan CZMP had no statistically discernable effect on plan policy focus regarding coastal area management, acknowledging again the very small sample size.
Table 5.
Results from regression analysis on plan policy focus on coastal area management.
| Regression on PPFCAM — Coastal area management (index scale from 0 to 10) | |||||
|---|---|---|---|---|---|
| R2 = 0.66 | |||||
| n = 28 | |||||
| Variable | Concept | Estimate | SE | t value | p valuec |
| CO (county) | Demographic/control | − 0.54 | 0.95 | —0.561 | 0.579 |
| GLSA (GL shoreland area) | Demographic/control | − 0.13 | 0.12 | —1.063 | 0.297 |
| MHVa (median house value) | Capacity (Financial) | 0.50 | 0.25 | 1.975 | 0.058 |
| PZS# (planning/zoning staff) | Capacity (Technical)/Knowledge (Technical) | 0.07 | 0.03 | 2.25 | 0.033 |
| PCNSLT (planning consultant) | Capacity (Technical)/Knowledge (Technical) | −1.27 | 0.43 | − 2.939 | 0.007 |
| CPPT (citizen planner training) | Knowledge (Technical) | − 0.61 | 0.48 | —1.270 | 0.215 |
| LUAYRb (land use actions/year) | Knowledge (Experiential)/Commitment | 0.08 | 0.02 | 5.037 | 0.000 |
| PUBPTN (public participation) | Knowledge (Experiential)/Commitment | 0.10 | 0.07 | 1.442 | 0.160 |
| LBCPLN (legislative commitment) | Commitment | − 0.03 | 0.25 | —0.139 | 0.891 |
| Intercept | 1.44 | 0.91 | 1.59 | 0.122 | |
In thousands, estimate centered on the mean.
Estimate centered on the mean.
Outcomes with a p-value reflecting a confidence level of 90% or greater are in bold as noteworthy given the small sample size.
Table 6.
Results from regression analyses on plan policy focus comparative outcomes.
| Regression on Plan Policy Focus on Urban, Rural, and Water Policies (index scale from 0 to 10) | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Variable | Concept | Vital Urban Centers | Conserved Rural Areas | Water Management | |||||||||
| R2 = 0.31 | R2 = 0.45 | R2 = 0.60 | |||||||||||
| n = 28 | n = 28 | n = 28 | |||||||||||
| Est | SE | t val | p val | Est | SE | t val | p val | Est | SE | t val | p valc | ||
| CO | Control | − 1.29 | 2.47 | − 0.52 | 0.06 | − 2.98 | 3.55 | − 0.84 | 0.41 | 2.17 | 0.94 | 2.31 | 0.03 |
| GLSA | Control | 0.18 | 0.28 | 0.64 | 0.53 | 0.57 | 0.36 | 1.58 | 0.12 | − 0.31 | 0.16 | − 1.94 | 0.06 |
| MHVa | Capacity | 0.97 | 0.33 | 2.92 | 0.01 | 0.75 | 0.48 | 1.56 | 0.13 | 0.58 | 0.27 | 2.12 | 0.04 |
| PZS# | Capacity/Know | 0.04 | 0.12 | 0.33 | 0.74 | 0.23 | 0.18 | 1.27 | 0.21 | 0.04 | 0.06 | 0.78 | 0.44 |
| PCNSLT | Capacity/Know | 0.34 | 1.27 | 0.27 | 0.79 | 0.38 | 0.92 | 0.42 | 0.68 | −0.80 | 0.81 | − 0.98 | 0.34 |
| CPPT | Know | −0.51 | 1.21 | − 0.42 | 0.68 | 0.42 | 0.85 | 0.50 | 0.62 | −0.11 | 0.62 | − 0.18 | 0.86 |
| LUAYRb | Know/Commit | 0.10 | 0.04 | 2.52 | 0.02 | 0.04 | 0.03 | 1.31 | 0.20 | 0.10 | 0.02 | 5.08 | 0.00 |
| PUBPTN | Know/Commit | 0.14 | 0.16 | 0.90 | 0.37 | 0.30 | 0.14 | 2.14 | 0.04 | 0.22 | 0.11 | 1.93 | 0.06 |
| LBCPLN | Commitment | − 0.13 | 0.56 | − 0.24 | 0.81 | − 0.78 | 0.41 | − 1.89 | 0.07 | 0.14 | 0.37 | 0.37 | 0.71 |
| Intercept | 2.83 | 2.64 | 1.074 | 0.292 | 3.06 | 1.47 | 2.08 | 0.05 | 0.76 | 1.40 | 0.55 | 0.59 | |
In thousands, estimate centered on the mean.
Estimate centered on the mean.
Outcomes with a p-value reflecting a confidence level of 90% or greater are in bold as noteworthy given the small sample size.
Median home value in particular was specified as a measure of financial capacity to engage in planning, and it is not clear why it would be having an effect on plan policy focus but not on overall plan quality. It could be that the desire to protect coastal properties through stronger coastal management provisions is higher when property values are higher. Similarly, to the extent that the higher-value residential properties that coastal localities enjoy tend to be located within their coastal shoreland areas, and that wealthier shoreland property owners may be more educated about potential shoreland risks, those residents may also be pushing for more attention to shoreland management in order to better maintain their property values. Alternatively, however, to the extent that those same property owners might be more inclined and able to dissuade a coastal locality from adopting strong shoreland area policies that would limit development potential (i.e., dampening local commitment), one could expect that higher median home value might be associated inversely with a plan policy focus on shoreland area management. The robustness and implications of this finding in particular warrant further study.
Similarly, the results presented in Table 5 suggest that use of a planning consultant appears to be inversely associated with plan policy focus on coastal area management to a substantial extent. That is, having a consultant prepare some or all of a locality’s master plan will likely yield a plan that focuses less on coastal shoreland issues relative to having the plan prepared without use of a consultant. Moreover, and in contrast, use of a consultant has no statistically significant discernable effect in terms of the plan’s focus on the other more conventional policy areas of interest to localities—urban centers, rural conservation, and water management (see Table 6).
This finding regarding the influence of consultants is consistent with other literature on the use of in-house planning staff versus consultants in general (see, e.g., Loh & Norton, 2013, 2015; Lyles et al., 2014a), and with regard to coastal area management in particular (Norton, 2005b). That work suggests that planning consultants tend to focus on policy issues of broad concern to localities (e.g., vital urban centers, conserved rural areas), such that their contribution by itself would not likely affect the level of attention given to those policies. Consultants also tend to take a standardized or template approach to their efforts, however, which does not allow for much tailoring to the unique specifics of a locality. Coastal Michigan localities are different from most of the other local clients that planning consultants serve in that they have unique coastal shoreland areas to address. The consultants’ standardized approaches may not include attention to coastal management issues, therefore, and the plans they prepare may similarly provide less policy focus on the unique challenges regarding issues of coastal shoreland area management.
Finally, the results from the analysis of other policy focus areas provided for comparison are consistent and suggest that the modeling presented here is sufficiently robust. As noted above, the more dense urban cities in the study sample tend to experience more land use actions per year than the other jurisdictions, and that factor along with higher median home values appear to be significantly related to a heightened focus on promoting vital urban centers. Median home value was specified as a measure of capacity, and it may be capturing commitment by wealthier urban dwellers to enhanced urban revitalization as well.
Higher numbers of land use actions per year similarly are positively associated with greater emphasis on water management. Increased use of public participation in planning is associated with both increased attention on conserving rural areas and water management, and a higher median home value is positively associated with greater focus on water management. The categorical variable ‘county’ has an inverse relationship to a policy focus on vital urban centers, but this is better interpreted to suggest that ‘not county’ (i.e., more urban centered) has a positive relationship. All of these factors are consistent with the expectation that commitment to water management and open space preservation come from increased experiential knowledge of the impacts from development, along with broader community involvement, while increased attention to urban vitality is associated with more urban jurisdictions. In contrast, it appears that higher levels of legislative body commitment to planning, perhaps expressed especially by local officials in more urban areas, is associated with lower levels of emphasis on rural conservation. In addition, having greater relative shoreland area within the jurisdiction appears to be associated with lower emphasis on water management. Both of these latter findings merit additional study.
6. Conclusions and next steps
The need to plan for improved coastal area management, especially in light of global climate change, is a growing imperative, one that follows from calls for sustainability that emerged several decades ago. That need speaks especially to hazards avoidance and mitigation in near-shore areas given threats from high-energy waves and flooding during storms. In a sense, Great Lakes coastal communities have had to plan for these kinds of hazards for a long time given the nature of Great Lakes shoreline dynamics, especially as they relate to lake level fluctuations over time, although the degree of uncertainty and potential for extreme events are both increasing because of climate change. Even so, standing water levels for all of the Great Lakes have returned to mean elevations in the past several years following an extended period of low-water elevations, a period encompassing residents’ short-term memories and the master planning efforts for most of the localities evaluated for this study.
A key finding from this study, largely consistent with expectations from the literatures on local plan evaluation and coastal area management in ocean coastal settings, is that little is happening in Michigan’s coastal communities in terms of planning for coastal area management through local master plans. Great Lakes coastal localities in Michigan are largely overlooking their coastal zones, at least when it comes to master planning. Indeed, the most striking findings from our analysis are that almost half of the plans we evaluated fail to evaluate their coastal areas at all, and almost three-quarters fail to adopt any meaningful coastal area management policies.
These findings appear to be attributable largely to low capacity and commitment by local officials to local-level coastal area management. Nonetheless, in terms of actions that states and local governments can pursue, the results from this study suggest that finding ways to increase localities’ abilities to hire in-house planning staff, and possibly provide additional technical training, may have some effect on improving both the quality of the plans those localities are able to produce and the policy focus they give to coastal shoreland area management, along with other planning goals. Moreover, and perhaps more importantly, our results suggest that providing more training and support for the planning consultant community in particular, addressing specifically issues of shoreland area management and local policies that can be used to address them, might be an especially effective way to enhance coastal localities’ efforts to plan for their coasts.
Finally, we have found no evidence of studies similar to this on coastal area planning in the other Great Lakes states. However, nothing about the institutional structures of those states, and no studies regarding planning in those states more broadly, suggest that other Great Lakes states are likely doing anything substantially different from what we are observing in Michigan (see, e.g., Isely & Pebbles, 2009; Norton et al., 2011). Next steps include continuing our efforts to identify potentially fruitful planning methodologies and technical training programs for enhancing local shoreland area management, and extending our work more intentionally into other Great Lakes states.
Funding
This work was supported in part by grants from the University of Michigan (UM) Center for Local, State, and Urban Policy; UM Taubman College of Architecture and Urban Planning; UM Graham Sustainability Institute; Michigan Coastal Zone Management Program, Office of the Great Lakes; and U.S. National Oceanic and Atmospheric Administration, U.S. Department of Commerce. None of the funders directed: the research design employed for this study; the collection, analysis, or interpretation of data; the writing of this article; or the decision to submit this article for publication. The findings and conclusions presented here are those of the authors and do not necessarily reflect those of the funders.
Appendix A
See Table A1.
Table A1.
Distribution of local general purpose units of government in Michigan in total and those touching Great Lakes waters (lakes or connecting rivers)
| a. Counta | ||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Jurisdiction Type | State-wide | Touching Great Lakes Waters | ||||||||||||||||
| Peninsula | Touching a Lake | Touching a River | River/Lake | |||||||||||||||
| Lower | Upper | Superior | Michigan | Huron | St. Clair | Erie | Touching a Lake | Island | Houghtonb | St. Marys | St. Clair | Detroit | River Only | R & L | R or R & L | Total Lake or River or Both | ||
| County | 83 | 29 | 12 | 9 | 18 | 13 | 3 | 1 | 41 | 3 | 1 | 1 | 1 | 1 | 0 | 4 | 4 | 41 |
| Township | 1241 | 134 | 61 | 30 | 87 | 57 | 4 | 5 | 183 | 6 | 4 | 7 | 4 | 2 | 12 | 5 | 17 | 195 |
| City | 273 | 48 | 10 | 2 | 22 | 13 | 6 | 2 | 44 | 1 | 2 | 1 | 5 | 7 | 14 | 1 | 15 | 58 |
| Village | 262 | 19 | 5 | 3 | 13 | 7 | 0 | 1 | 24 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 1 | 24 |
| Twp, City, or Village | 1776 | 201 | 76 | 35 | 122 | 77 | 10 | 8 | 251 | 7 | 6 | 9 | 9 | 9 | 26 | 7 | 33 | 277 |
| Total | 1859 | 230 | 88 | 44 | 140 | 90 | 13 | 9 | 292 | 10 | 7 | 10 | 10 | 10 | 26 | 11 | 37 | 318 |
| b. Jurisdiction/category as a percent of total jurisdictions statewide | ||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Touching Great Lakes Waters | ||||||||||||||||||
| Peninsula | Touching a Lake | Touching a River | River/Lake | |||||||||||||||
| Jurisdiction Type | State-wide (Count) | Lower (%) | Upper (%) | Superior (%) | Michigan (%) | Huron (%) | St. Clair (%) | Erie (%) | Touches a Lake (%) | Island (%) | Houghton (%) | St. Marys (%) | St. Clair (%) | Detroit (%) | River Only (%) | R & L (%) | R or R & L (%) | Total (%) Lake or River or Both |
| County | 83 | 35 | 14 | 11 | 22 | 16 | 4 | 1 | 49 | 4 | 1 | 1 | 1 | 1 | 0 | 5 | 5 | 49 |
| Township | 1241 | 11 | 5 | 2 | 7 | 5 | < 1 | < 1 | 15 | < 1 | < 1 | 1 | < 1 | < 1 | 1 | < 1 | 1 | 16 |
| City | 273 | 18 | 4 | 1 | 8 | 5 | 2 | 1 | 16 | < 1 | 1 | < 1 | 2 | 3 | 5 | < 1 | 5 | 21 |
| Village | 262 | 7 | 2 | 1 | 5 | 3 | 0 | < 1 | 9 | 0 | 0 | < 1 | 0 | 0 | 0 | < 1 | < 1 | 9 |
| Twp, City, or Village | 1776 | 11 | 4 | 2 | 7 | 4 | 1 | < 1 | 14 | < 1 | < 1 | 1 | 1 | 1 | 1 | < 1 | 2 | 16 |
| Total | 1859 | 12 | 5 | 2 | 8 | 5 | 1 | < 1 | 16 | 1 | < 1 | 1 | 1 | 1 | 1 | 1 | 2 | 17 |
| c. Average land area, population size, and population density of coastal localities by jurisdiction type (based on a random sample of 70 coastal localities, not including Detroit) | |||
|---|---|---|---|
| Jurisdiction Type | Average Land Area (sq. mile) | Average Total Population | Average Population Density (persons per sq. mile of land area) |
| County | 532.0 | 42,961 | 74 |
| Township | 31.7 | 6058 | 297 |
| City | 4.3 | 7597 | 1501 |
| Village | 1.1 | 527 | 962 |
Counts for individual columns may sum to more than the total column because some jurisdictions touch multiple water bodies.
The Houghton river bisecting the Keweenaw Peninsula in the Upper Peninsula is technically a constructed waterway.
Appendix B. Items used for plan content evaluation of plan quality and plan policy focus indices
| a. Overall Plan Quality and Plan Quality for Coastal Area Management Analysis | |
|---|---|
| Item | Scoring Rubric |
| Overall Plan Quality | |
| General Presentation | |
| Table of content | Provided? 1 = yes; 0 = no |
| Sources in text/tables | Provided? 1 = yes; 0 = no |
| Use/quality of maps | 0 = none; 1 = limited/poor; 2 = standard; 3 = extensive, clear, usable |
| Use/quality of tables/figures (readability, etc.) | 0 = none; 1 = limited/poor; 2 = standard; 3 = extensive, clear, usable |
| Readability of text | 0 = poor; 1 = average; 2 = high |
| Articulation of Policies, Goals, and Purpose of Planning | |
| Clear statement of goals, policies, and objectives | Provided? 1 = yes; 0 = no |
| Explanation of planning process provided | 0 = not present; 1 = presented, not detailed; 2 = present and detailed |
| Discussion of planning, plan’s purpose | 0 = not present; 1 = presented, not detailed; 2 = present and detailed |
| Public Participation Used for Plan Preparation | |
| Description of public participation process | 0 = not present; 1 = presented, not detailed; 2 = present and detailed |
| Community visioning session, workshop, etc. | Discussed? 1 = yes; 0 = no |
| Survey of public participation | Discussed? 1 = yes; 0 = no |
| Fact Base | |
| Summary of data collection and analysis process | Provided? 1 = yes; 0 = no |
| Land development/land use change trends | Discussed? 1 = yes; 0 = no |
| Trends/problems regarding environment | Discussed? 1 = yes; 0 = no |
| Existing land uses | Provided? 0 = no; 1 = discussed; 2 = discussed and mapped |
| Land Suitability Analysis | |
| Physical limitations for development | Discussed/analyzed? 1 = yes; 0 = no |
| Floodplains | Identified? 0 = no; 1 = yes, not detailed; 2 = yes, detailed; 3 = mapped |
| Steep slopes | Identified? 0 = no; 1 = yes, not detailed; 2 = yes, detailed; 3 = mapped |
| Fragile natural areas | Identified? 0 = no; 1 = yes, not detailed; 2 = yes, detailed; 3 = mapped |
| Infrastructure Capacity Analysis | |
| Auto/roadway system quality | Discussed/analyzed? 1 = yes; 0 = no |
| Drinking water supply | Discussed/analyzed? 1 = yes; 0 = no |
| Stormwater management system capacity | Discussed/analyzed? 1 = yes; 0 = no |
| Wastewater management | Discussed/analyzed? 1 = yes; 0 = no |
| Police and fire protection | Discussed/analyzed? 1 = yes; 0 = no |
| Greenways/green spaces (e.g., trails) | Discussed/analyzed? 1 = yes; 0 = no |
| Active recreation facilities (e.g., soccer fields) | Discussed/analyzed? 1 = yes; 0 = no |
| Community facilities | Discussed/analyzed? 1 = yes; 0 = no |
| Analysis: population and infrastructure | Provided? 0 = no; 1 = present, not detailed; 2 = present, detailed |
| Vertical, Horizontal, and Internal Consistency | |
| Discussion of intergovernmental coordination | 0 = not present; 1 = present, not detailed; 2 = present, detailed |
| Consistency with other plans, policies, ordinances | 0 = not present; 1 = present, not detailed; 2 = present, detailed |
| Consistency: goals, policies, objectives | 0 = inconsistent; 1 = not inconsistent; 2 = consisted and supported |
| Implementation | |
| Implementation responsibilities | Provided? 1 = yes; 0 = no |
| Implementation mechanisms | Discussed generally? 1 = yes; 0 = no |
| Land use/subdivision regulations, and/or zoning | Discussed? 1 = yes; 0 = no |
| Capital improvement plans and/or facilities plans | Discussed? 1 = yes; 0 = no |
| Plan Quality — Coastal Area Management Analysis | |
| Constraints: coastal zones | Identified? 0 = no; 1 = yes, not detailed; 2 = yes, detailed; 3 = mapped |
| Constrains: erosion | Identified? 0 = no; 1 = yes, not detailed; 2 = yes, detailed; 3 = mapped |
| Natural areas: coastal | Identified? 0 = no; 1 = yes, not detailed; 2 = yes, detailed; 3 = mapped |
| b. Plan Policy Focus | |
| Item | Scoring Rubric |
| Vital Urban Centers | |
| Walkable communities | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| High density concentrated with urban services | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Investment/reinvestment in developed areas | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Growth directed to existing urban areas | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Conserved Rural Areas | |
| Low density expansion controlled or limited | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Growth directed away from resource areas | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Protection of natural areas, open spaces | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Tools for agr/OS preservation (e.g., TDR, PDR) | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Cluster development | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Water Quantity and Quality Management | |
| Buffer zones near sensitive/unique natural areas | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Surface water protection, including wetlands | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Groundwater protection | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Floodplain development restrictions | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Development setbacks/vegetative buffers | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| On-site stormwater management systems | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Coastal Area Management Policy | |
| Coastal setbacks | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Shoreline protection overlay district | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Shoreline erosion best management practices | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
| Public access/use restrictions | Specificity: 0 = not present; 1 = present, not detailed; 2 = detailed |
Appendix C
a. Key non-academic stakeholders in participatory action research (PAR) program (noting positions and affiliations at the time research/work contributions were made):
Joe VanderMeulen, Executive Director, Land Information Access Association (LIAA).
Harry Burkholder, Senior Planner and Executive Director, LIAA.
Claire Karner, Staff Planner, LIAA.
Katie Sieb, Staff Planner, LIAA.
Ronda Wuycheck, Chief, Michigan Coastal Zone Management Program (MCZMP).
Matt Smar, Great Lakes Program Manager, MCZMP.
Matt Warner, Coastal Hazards Specialist, MCZMP.
John Hodgson, Community Development Director, City of St. Joseph.
Jennifer Howland, Community Development Manager, City of Grand Haven.
Stacey Fedewa, Community Development Director, Grand Haven Township.
b. Meeting dates and purposes of key stakeholder-participant presentations and meetings conducted through PAR program, from February 2014 through February 2017.
February 19, 2014–Project kick-off meeting with Ludington community (Ludington City, Pere Marquette Township, Hamlin Township) planning officials and public.
March 19, 2014–Project meeting with Ludington community planning officials.
April 17, 2014–Project kick-off meeting with St. Joseph City planning officials and public.
April 24, 2014–Project meeting with Ludington community planning officials.
April 24, 2014–Project kick-off meeting with Grand Haven City and Grand Haven Township planning officials and public.
May 7, 2014–Project meeting with City of St. Joseph planning officials.
May 16, 2014–Project presentation and meeting with Ludington community public and citizen leaders.
May 21, 2014–Ludington community planning meeting.
June 18, 2014–Ludington community action team (CAT) meeting.
July 16, 2014–Ludington community planning meeting.
July 28, 2014–Project update and discussion with MCZMP officials.
August 27, 2014–Ludington CAT meeting.
September 17, 2014–Project presentation with Ludington community planning commissions and public.
October 1, 2014–Project presentation with Grand Haven City and Township public and citizen leaders.
October 14, 2014–Project update and discussion with MCZMP officials.
November 12, 2014–Project update and discussion with Ludington community planning officials.
November 12, 2014–Project update and discussion with Grand Haven City and Township planning officials.
November 13, 2014–Project update and discussion with the St. Joseph City planning officials.
February 25, 2015–Project update and discussion with the Grand Haven City and Township Planning Commissions.
March 25, 2015–Project update and discussion with Grand Haven City and Township Planning Commissions.
April 22, 2015–Project update and discussion with Grand Haven City and Township Planning Commissions.
June 4, 2015–Project meeting with St. Joseph City planning officials.
September 24, 2015–Final master plan project presentation to St. Joseph City Planning Commission and public.
October 20, 2015–Final master plan project presentation to Grand Haven City and Township Planning Commissions and public
November 23, 2015–Project update and discussion with MCZMP officials.
January 25, 2016–Project meeting with Grand Haven City and Township Planning Commissions and planning officials to introduce integrated assessment project.
March 14, 2016–Project update and discussion with MCZMP officials.
May 10, 2016–Project update meeting with Grand Haven City and Township Planning Commissions and planning officials on integrated assessment project.
May 16, 2016–Project meeting with Grand Haven City and Township planning officials to discuss integrated assessment and gather input.
June 6, 2016–Project meeting with Grand Haven Township Planning Commission to provide project update and discuss integrated assessment scope.
July 18, 2016–Project update meeting and discussion with Grand Haven Township Planning Commission and planning officials on integrated assessment project.
July 19, 2016–Project update meeting and discussion with Grand Haven City Planning Commission and planning officials on integrated assessment project.
August 10, 2016–Project meeting with Grand Haven Township Fire Chief and Department of Public Works Director to obtain input on resiliency concepts.
August 26, 2016–Project stakeholder participant team meeting (including academic researchers, LIAA staff, MCZMP staff) to discuss coastal resiliency and future work.
November 9, 2016–Project update meeting and discussion with Grand Haven City Planning Commission and planning officials on integrated assessment project.
November 21, 2016–Project update meeting and discussion with Grand Haven Township Planning Commission and planning officials on integrated assessment project.
January 17, 2017–Final integrated assessment presentation to Grand Haven City Planning Commission and planning officials.
January 26, 2017–Final integrated assessment presentation to Grand Haven City Planning Commission and planning officials.
February 8, 2017–Project stakeholder participant team meeting (including academic researchers, LIAA staff, MCZMP staff) to discuss coastal resiliency and future work.
Footnotes
Declaration of conflicting interests
The authors have no conflicts of interest to declare.
These five Great Lakes comprise the “Laurentian” Great Lakes because they drain to the Atlantic Ocean through the St. Lawrence River and Seaway Basin. They are international waters, bordering the U.S. and Canada along the Provinces of Ontario and Quebec. Several additional smaller lakes exist within the system as well, most notably Lake St. Clair separating Michigan from Ontario, but those lakes are not labeled “great” lakes. Even so, we include jurisdictions bordering Lake St. Clair and its connecting rivers for analytical purposes here.
In a sense, the analytical components of a master plan also essentially represent an institutionalized kind of knowledge, and part of an assessment of plan quality correspondingly measures the quality of that knowledge in terms of how robust and well conveyed it is.
The elevations of water levels on the Great Lakes are measured by a standardized reference elevation above sea level, the International Great Lakes Datum (IGLD), which was last adjusted in 1985 (noted as IGLD 1985). Since 1860, recorded water levels for Lakes Michigan and Huron, for example, have fluctuated between roughly 576.5 feet to roughly 582.5 feet above sea level (IGLD 1985) (see http://www.glerl.noaa.gov/data/now/wlevels/dlbd/). The enormity of these fluctuations in standing lake water levels are especially evident when compared to the anticipated rise of standing sea level because of climate change, predicted to be on the order of inches over decades rather than feet.
The original Coastal Management Program and corresponding Final Environmental Impact Statement for Michigan’s CZMP was adopted in 1978, and it has not been amended since. The original document can be retrieved from the Michigan CZMP web page at: http://www.michigan.gov/deq/0,4561,7-135-3313_3677_3696-,00.html (August 2017). MCZMP staff are currently working with several of the authors of this paper and other stakeholders through a PAR program to develop potential changes to the state’s coastal management program that would enhance local shoreland planning and development management efforts.
The state may be moving in this direction, but we have found no evidence that the state’s hazard mitigation planning efforts were having any discernable effects on local efforts to address coastal hazards through their master planning at the times that plans were collected and local officials were surveyed for this study.
As it turns out, the period during which we collected master plans and conducted surveys came toward the end of an exceptional long period of very low lake water levels, and the most recent storms of record occurred back in 1986 (MDSP, 2014). Since completing those data collection efforts, the lakes have come back up to long-term average levels and beyond, rising dramatically starting in about 2013 (US GLERL, 2017). No significant storms of record have occurred under those high-water conditions, yet, such that the effects of these increased risks cannot yet be gauged.
Planning and zoning authorities are delegated primarily to townships, cities, and villages in Michigan, which together encompass the entire land area of the state. The vast majority of that land area is controlled by townships and cities, as opposed to villages. Counties are enabled to plan for their jurisdictions, but those plans have no legal effect (see 2008 33 110, MCL 125.381 et seq.). They are similarly enabled to zone for areas outside of cities, but only when their constituent townships have declined to do so (see 2006 PA 110, MCL 125.3101 et seq.). For the most part, counties in Michigan only plan and zone in more rural areas, doing so on behalf of their constituent townships that lack the capacity or desire to adopt their own local plans and codes.
For this study, we include all 318 coastal localities, including those abutting connecting rivers but not necessarily lakefronts.
For townships, the mean population density for study sites was 306 persons per sq. mile, compared to 91 persons per square mile for non-study sites (two-tailed P(T ≤ 1) = 0.00); mean housing unit density was 154 units/sq. mi. for study sites compared to 52 units/sq. mi. for non-study sites (two-tailed P(T,=1) = 0.00); and mean housing value was $196,741 for study sites compared to $150,509 for non-study sites (two-tailed P (T ≤ 1) = 0.00).
The methodology engaged here, and specifically the practice of not calculating and reporting inter-rater reliability scores for paired evaluations reconciled, was undertaken before a paper was published recently by Stevens et al. (2014), in which the authors argue compellingly that such statistics should be reported even when scores are reconciled. Unfortunately, while all of the paired evaluations were reconciled for this study, inter-rater reliability scores were not calculated and are not available for reporting here.
With regard to counties in particular, note that variables drawn from census data and other sources (e.g., land area, population) reflect the totality of each county, inclusive of the townships, cities, and villages encompassed by that county, while data sources drawn from plans and surveys reflect the county governmental unit only (i.e., the 11 planning and zoning staff members employed by counties on average capture only the county government itself and do not reflect the numbers of planners employed by townships, cities, and villages within the county).
References
- Angel JR, Kunkel KE, 2010. The response of Great Lakes water levels to future climate scenarios with an emphasis on Lake Michigan-Huron. J. Great Lakes Res 36, 51–58. [Google Scholar]
- Angel JR, 2013. The response of Great Lakes water levels and potential impacts of future climate scenarios. In: Pryor SC (Ed.), Climate Change in the Midwest: Impacts, Risks, Vulnerability, and Adaptation Indiana University Press, Bloomington. [Google Scholar]
- Arbogast AF, Cabala T, Davis CF III, DeVries-Zimmerman S, Yurk B, Garmon B, van Dijk D, VanHorn J, 2015. Bringing the Latest Science to Management of Michigan’s Coastal Dunes. Mich. Env. Council, Lansing [Google Scholar]
- Baer WC, 1997. General plan evaluation criteria: an approach to making better plans. J. Am. Plan. Assoc 63 (3), 329–344. [Google Scholar]
- Basset E, Shandas V, 2010. Innovation and climate action planning: perspectives from municipal plans. J. Am. Plan. Assoc 76 (4), 435–450. [Google Scholar]
- Beatley T, Brower DJ, Schwab AK, 2002. An Introduction to Coastal Zone Management, 2nd ed. Island Press, Washington, D.C. [Google Scholar]
- Beatley T, 2009. Planning for Coastal Resilience: Best Practices for Calamitous Times Island Press, Washington, D.C. [Google Scholar]
- Berke PR, Godschalk DR, 2009. Searching for the good plan: a meta-analysis of plan quality studies. J. Plan. Lit 23 (3), 227–240. [Google Scholar]
- Berke P, Lyles W, 2013. Public risks and the challenges to climate-change adaptation: a proposed framework for planning in the age of uncertainty. Cityscape 181–208. [Google Scholar]
- Berke P, Manta Conroy M, 2000. Are we planning for sustainable development? An evaluation of 30 comprehensive plans. J. Am. Plan. Assoc 66 (1), 21–32. [Google Scholar]
- Berke PR, Roenigk D, Kaiser E, Burby R, 1996. Enhancing plan quality: evaluating the role of state planning mandates for natural hazard mitigation. J. Environ. Plan. Manage 39 (1), 79–96. [Google Scholar]
- Berke PR, Crawford J, Dixon JE, Ericksen NJ, 1999. Do cooperative environmental planning mandates produce good plans? Empirical results from the New Zealand experience. Env. Plan. B: Plan. Des 26, 643–664. [Google Scholar]
- Berke PR, Godschalk DR, Kaiser EJ, Rodriguez DA, 2006. Urban Land Use Planning, 5th ed. University of Illinois Press, Urbana. [Google Scholar]
- Berke P, Smith G, Lyles W, 2012. Planning for resiliency: evaluation of state hazard mitigation plans under the Disaster Mitigation Act. Nat. Hazards Rev 13 (2), 139–149. [Google Scholar]
- Brody SD, Highfield WE, 2005. Does planning work? Testing the implementation of local environmental planning in Florida. J. Am. Plan. Assoc 71 (2), 159–175. [Google Scholar]
- Brody SD, Kang JE, Bernhardt S, 2010. Identifying factors influencing flood mitigation at the local level in Texas and Florida: the role of organizational capacity. Nat. Hazards 52 (1), 167–184. [Google Scholar]
- Brody SD, 2003. Are We Learning to Make Better Plans?: A longitudinal analysis of plan quality associated with natural hazards. J. Plan. Educ. Res 23, 191–201. [Google Scholar]
- Bunnell G, Jepson EJ, 2011. The effect of mandated planning on plan quality. J. Am. Plan. Assoc 77 (4), 338–353. [Google Scholar]
- Burby RJ, May PJ, 1997. Making Governments Plan: State Experiments in Managing Land Use Johns Hopkins University Press, Baltimore. [Google Scholar]
- Burby RJ, 2006. Hurricane Katrina and the paradoxes of government disaster policy: bringing about wise government decisions for hazardous areas. Ann. Am. Acad. Pol. Soc. Sci 64 (May), 171–191. [Google Scholar]
- Citizens Research Council (CRC), 1999. A Bird’s Eye View of Michigan Local Government at the End of the Twentieth Century Livonia Citizens Research Council (CRC), MI. [Google Scholar]
- Dillman DA, 2000. Mail and Internet Surveys: The Tailored Design Method J. Wiley, New York. [Google Scholar]
- Du Toit J, Boshoff N, Mariette N, 2016. Normative versus actual methodologies in planning research: a hybrid picture. J. Plan. Educ. Res 1–11 (xx, x,). [Google Scholar]
- Edwards MM, Haines A, 2007. Evaluating smart growth: implications for small communities. J. Plan. Educ. Res 27, 49–64. [Google Scholar]
- Fedewa S, 2017. Community Development Director, Grand Haven Charter Township, MI. Personal communication to authors, January, 2017. [Google Scholar]
- Fisher GA, Galvin JF, Greene AM, Need GK, Rosati CA, 2012. Michigan Zoning, Planning, and Land Use Institute for Continuing Legal Education, Ann Arbor. [Google Scholar]
- Great Lakes Information Network (GLIN), 2017. Great Lakes Facts and Figures http://www.great-lakes.net/lakes/ref/lakefact.html.
- Godschalk DR, Cousins K, 1985. Coastal management: planning on the edge. J. Am.Plan. Assoc 51 (3), 263–265. [Google Scholar]
- Gronewold AD, Fortin V, Lofgren B, Clites A, Stow CA, Quin F, 2013. Coasts, water levels, and climate change: a Great Lakes perspective. Clim. Change 120, 697–711. [Google Scholar]
- Hershman MJ, Wood JW, Bernd-Cohen T, Goodwin RF, Lee V, Pogue P, 1999. The effectiveness of coastal zone management in the United States. Coast. Manage 21, 113–138. [Google Scholar]
- Horney J, Nguyen M, Salvesen D, Dwyer C, Cooper J, Berke P, 2017. Assessing the quality of rural hazard mitigation plans in the Southeastern United States. J. Plan. Educ. Res 37 (1), 56–65. [Google Scholar]
- Howland J, 2017. Community Development Manager, City of Grand Haven, MI. Personal communication to authors, January 2017. [Google Scholar]
- International Joint Commission (IJC), 2012. Lake Superior Regulation: Addressing Uncertainty in Upper Great Lakes Water Levels: The International Upper Great Lakes Study Final Report International Joint Commission (IJC), Washington, D.C. [Google Scholar]
- Isely ES, Pebbles V, 2009. U.S. Great Lakes policy and management: a comparative analysis of eight States’ coastal and submerged lands programs and policies. Coast. Manage 37, 197–213. [Google Scholar]
- Juergensmeyer JC, Roberts TE, 2013. Land Use Planning and Development Regulation Law Thompson West, St. Paul. [Google Scholar]
- Keillor P (Ed.), 2003. Living on the Coast: Protecting Investments in Shore Property on the Great Lakes U.S. Army Corp of Engineers, Detroit Division, and University of Wisconsin Sea Grant Program. [Google Scholar]
- Knizacky F, 2014. County Administrator, Mason County, MI. Personal communication to authors, September 2014. [Google Scholar]
- Komar PD, 1997. Beach Processes and Sedimentation, 2nd ed. Prentice-Hall, New York. [Google Scholar]
- Lake RW, Zitcer AW, 2012. Who says? Authority, voice, and authorship in narratives of planning research. J. Plan. Educ. Res 32 (4), 389–399. [Google Scholar]
- Laurian L, Day M, Berke PR, Ericksen N, Backhurst M, Crawford J, Dixon J, 2004. Evaluating plan implementation: a conformance-based methodology. J. Am. Plan. Assoc 70 (4), 471–480. [Google Scholar]
- Lederle K, 2017. Environmental Quality Specialist Great Lakes Shorelands Unit, Michigan Department of Environmental Quality; (Personal communication to authors, August 2017). [Google Scholar]
- Lloyd MG, Peel D, Duck RW, 2013. Towards a social-ecological resilience framework for coastal planning. Land Use Policy 03, 925–933. [Google Scholar]
- Loh CG, Norton RK, 2013. Planning consultants and local planning: roles and values. J. Am. Plan. Assoc 79 (2), 138–147. [Google Scholar]
- Loh CG, Norton RK, 2015. Planning consultant’s influence on local comprehensive plans. J. Plan. Educ. and Res 35 (2), 119–208. [Google Scholar]
- Lyles W, Stevens M, 2014. Plan quality evaluation 1994–2012: Growth and contributions, limitations, and new directions. J. Plan. Educ. Res 34 (4), 433–450. [Google Scholar]
- Lyles W, Berke P, Smith G, 2014a. A comparison of local hazard mitigation plan quality in six states, USA. Landsc. Urban Plan 122, 89–99. [Google Scholar]
- Lyles LW, Berke P, Smith G, 2014b. Do planners matter?: Examining factors driving incorporation of land use approaches into hazard mitigation plans. J. Environ. Plan. Manage 57 (5), 792–811. [Google Scholar]
- Lyles W, Berke P, Smith G, 2016. Local plan implementation: assessing conformance and influence of local plans in the United States. Environ. Plan. B: Plan. Design 43 (2), 381–400. [Google Scholar]
- Michigan Dept. of Env. Quality (MDEQ), 2017a. Coastal Management http://www.michigan.gov/deq/0,1607,7-135-3313_3677_3696—,00.html.
- MDEQ, 2017b. Shoreland Management http://www.michigan.gov/deq/0,4561,7-135-3313_3677_3700—,00.html.
- Michigan Dept. of State Police (MDSP), 2014. Michigan Hazard Mitigation Plan. MSP/EMHSD Pub. 106
- May PJ, Burby RJ, Ericksen NJ, Handmer JW, Dixon JE, Michaels S, Smith DI, 1996. Environmental Management and Governance Routledge, London. [Google Scholar]
- Meadows GA, Meadows LA, Wood WL, Hubertz JM, Perlin M, 1997. The relationship between Great Lakes water levels, wave energies, and shoreline damage. Bull. Am. Meteor. Soc 78 (4), 675–683. [Google Scholar]
- Norton RK, Meadows GA, Meadows LA, 2011. Drawing lines in law books and on sandy beaches: marking ordinary high water on Michigan’s Great Lakes shorelines under the public trust doctrine. Coast. Manage 39 (2), 133–157. [Google Scholar]
- Norton RK, Meadows GA, Meadows LA, 2013. The deceptively complicated ‘elevation ordinary high water mark’ and the problem with using it on a Great Lakes shore. J. Great Lakes Res 39, 527–535. [Google Scholar]
- Norton RK, 2005a. More and better planning: state-mandated local planning in coastal North Carolina. J. Am. Plan. Assoc 71 (1), 55–71. [Google Scholar]
- Norton RK, 2005b. Local commitment to state-mandated planning in coastal North Carolina. J. Plan. Educ. Res 25 (2), 149–171. [Google Scholar]
- Norton RK, 2005c. Striking the balance between environment and economy in Coastal North Carolina. J. Environ. Plan. Manage 48 (2), 177–207. [Google Scholar]
- Norton RK, 2008. Using content analysis to evaluate local master plans and zoning codes. Land Use Policy 25, 432–454. [Google Scholar]
- Norton RK, 2011. Who decides, how and Why? Planning for the judicial review of local legislative zoning decisions. Urb. Lawyer 43 (4), 1085–1105. [Google Scholar]
- Pendall R, 2001. Municipal plans, state mandates, and property rights. J. Plan. Educ. Res 21, 154–165. [Google Scholar]
- Platt RH, 2014. Land Use and Society: Geography, Law, and Public Policy, 3rd ed. Island Press, Washington, D.C. [Google Scholar]
- Slade DC, Kehoe RK, Stahl JK, 1997. Putting the Public Trust Doctrine to Work, 2nd ed. Coastal States Organization, Washington, D.C. [Google Scholar]
- Stevens MR, Ward L, Berke PR, 2014. Measuring and reporting intercoder reliability in plan quality evaluation research. J. Plan. Educ. Res 34 (1), 77–93. [Google Scholar]
- Stevens MR, 2013. Evaluating the quality of official community plans in Southern British Columbia. J. Plan. Educ. Res 33 (4), 471–490. [Google Scholar]
- Tang Z, 2008. Evaluating local coastal zone land use planning capacities in California. Ocean Costal Manage 51, 544–555. [Google Scholar]
- U.S. Environmental Protection Agency (EPA), 2017a. The Great Lakes: An Environmental Atlas and Resource Book http://epa.gov/greatlakes/atlas/index.html.
- EPA, 2017b. About the Great Lakes National Program Office (GLNPO) http://www2.epa.gov/aboutepa/about-great-lakes-national-program-office-glnpo.
- U.S. Great Lakes Environmental Research Laboratory (GLERL), 2017. Great Lakes Water Level Observations http://www.glerl.noaa.gov/data/now/wlevels/levels.html.
- Warner M, 2017. Shoreline Hazards Specialist, Michigan Coastal Zone Management Program Personal communication to authors, August 2017.
- Wuycheck R, 2017. Chief, Michigan Coastal Zone Management Program (Personal communication to authors, August 2017.