Economic Values of Coastal Erosion Management: Joint Estimation of Use and Passive Use Values with Recreation and Contingent Valuation Data

Home , Beach house

Craig E. Landry Department of Agricultural and Applied Economics University of Georgia Athens, GA 30602 [email protected]; 706-542-0747

John C. Whitehead Department of Economics and Finance Appalachian State University Boone, NC

Selected Paper prepared for presentation for the 2015 Agricultural & Applied Economics Association and Western Agricultural Economics Association Annual Meeting, San Francisco, CA, July 26-28.

Copyright 2015 by Craig E. Landry and John C. Whitehead. All rights reserved. Readers may make verbatim copies of this document for non-commercial purposes by any means, provided this copyright notice appears on all such copies.

Economic Values of Coastal Erosion Management: Joint Estimation of Use and Passive Use Values with Recreation and Contingent Valuation Data

The coastal zone is a dynamic and recalcitrant ecological system. Management problems stemming from coastal erosion, storms, and sea level rise are exacerbated by development along the coast and, especially, by development at the water’s edge. More than 52% of the U.S. population lives in coastal counties (Joint Ocean Commission 2012). Focusing on U.S. East Coast barrier island beaches, approximately 86% percent of the shoreline has exhibited significant erosion in the past 100 years (Galgano, et al. 2004), averaging an annual rate of 1.6 feet of shoreline recession (USGS 2010). A 2000 study by the Federal Emergency Management Agency (FEMA) and the Heinz Center estimates that 25% of homes within 500 feet of the coast could be lost by 2060, at a potential cost of $530 million per year (Heinz Center 2000)! This vulnerability has serious implications for economic welfare of coastal households, viability of private and public insurance programs, recreation and tourism along the coast, regional economies adjacent to the coast, and ecological sustainability of coastal systems.

Options for management of shoreline erosion include shoreline hardening (seawalls and other structures designed to protect land and preserve property boundaries), beach replenishment (adding sand to the beach and dune system), and coastal retreat (moving buildings and infrastructure away from eroding shorelines as necessary). The effects of erosion management strategies on the configuration and value of the coastal housing units have been studied by a number of scholars (e.g., Hamilton 2007; Pompe 2008; Landry and Hindsley 2011; Gopalakrishnan, et al. 2011; Ranson 2012; Landry and Allen 2015). Management of erosion, however, also has important implications for the local tourism economy, the quality of coastal resources, and ecological sustainability over the long term. These aspects of management have received considerably less attention in the literature. Coastal property owners have been perhaps the most vocal proponents of shoreline hardening and beach replenishment, with other voices and options receiving little consideration.1

The project employs a survey of North Carolina households in order to evaluate the welfare effects of beach erosion management alternatives on the general population. The survey gathers data on use (and non-use) of coastal beaches, perceptions of coastal resource quality, knowledge of coastal processes, and stated preference referendum votes for programs to manage coastal erosion. With a 61% response rate, our online panel data compares favorably to U.S. Census decriptive statistics for North Carolina, with slightly greater representation for more educated and wealthier households.

1 Until recently, North Carolina was one of only two states that prohibited shoreline hardening. In 2012, the NC legislature passed a bill (unsigned and unvetoed by the Govnernor, thus becoming law) that allows terminal groins along inlets, and there are significant pressures to allow other types of shoreline armoring.

By combining information on beach users and non-users, we are able to represent diverse groups of stakeholders and conduct a comprehensive and comparative analysis of economic welfare and empirical tests on the magnitude of use and non-use values. We collect information on recreational visitation under current and hypothetical future conditions in order to assess the effects of coastal erosion and erosion managmenent policies on tourism in the coastal zone. Our experimental deisgn provides for both within- and between-subject tests of the economic effects of increasing beach width and negative environmental impacts on visitation patterns and economic value. This permits us to assess the effects of environmental quality on recreation value and tourism, and to conduct tests for sequencing effects (for each of the three erosion management strategies we study).

In addition, we implement contingent valuation to assess households’ willingess to pay (both users and non-users) for different approaches to shoreline erosion management. Treatment effects relating to beach width and the presence of negative environmental impacts are incluced within the CV-design. Following modern best-practices in stated preference analysis, we implement a consequential survey design (highlighting policy relevance and application of study findings) and measure a number of important individual perceptions, including response certainty, perceived consequences of survey responses, and efficacy of government management actions.

We build on the microeconomic model of Eom and Larson (2006) to jointly estimate parameters of recreation demand and passive use values. Our model does not impose weak complementarity (typically invoked in welfare analysis of recreation demand), but rather can test for its existence. By combining contingent valuation and contingent behavior data, we employ a consistent behavioral model that permits analysis of co-existing use and passive use values and how these values are affected by beach width, erosion management strategy, and the presence of environmental impacts engendered through management.

Preliminary results indicate that shoreline retreat recieves the highest level of support among NC households – both in terms of expected future beach trips and affirmative responses in the contingent valuation exercise. Beach replenishment is a close second, while shoreline armoring receives the lowest level of support among the general population of North Carolina.

Projected beach visitation under beach replenishment is approximately the same as the RP status quo (3 trips/year), while the shoreline armoring scenarios exhibits are average treatment effect of 0.5 trips per year. Contingent trips under the shoreline armoring scenario decrease by 0.75 trips per year. Negative environmental impacts of management reduce recreation demand by at least one trip per year (with the greatest effect for shoreline retreat). Contingent recreation demand decreases in scenarios that describe the beach as severelly eroded, but only increases slightly in scenarios that inmprove beach width. Preliminary analysis of contingent valuation data indicates similar patterns as the recreation demand results.

Background

Residential and commercial development in the coastal zone facilitates access to and enjoyment of coastal amenities, such as beaches, estuaries, fisheries, and cultural resources. Dynamics of the shore have resulted in various patterns of erosion and accretion, with an overwhelming majority of shorelines along the east coast exhibiting net erosion in recent decades (Galgano, et al. 2004; USGS 2010); the North Carolina coastline, our study site, is no exception (Riggs and Ames 2003). Driven by wind, waves, storms, and sea level, erosion of beaches, bluffs, and estuarine shoreline can threaten the viability of residential and commercial development, as well as public infrastructure, by undermining foundations and exposing structures to wave attack, currents, and tidal fluctuations. Waterfront structures face the greatest risk, and owners have strong incentives to lobby for public projects that protect their property. The management of coastal erosion, however, affects environmental quality and the overall appeal of beaches and estuaries as tourist destinations. Thus, erosion control policies influence environmental quality and recreation & tourism patterns for the public at large, but policy decisions can be more heavily influenced by coastal residents that face greater personal risk.

The objective of this paper is an analysis of public support and willingness to pay for coastal erosion control and the impact of coastal erosion control policies on recreation and tourist behavior in order to fill a gap in the existing literature. While much has been learned about coastal residents’ value of beach quality through analysis of property value data (e.g., Pompe 2008; Landry and Hindsley 2011; Gopalakrishnan, et al. 2011; Ranson 2012; Landry and Allen 2015), comparatively little is known about general public support and economic value for erosion management policies or how erosion and erosion control policies influence recreation behavior, experience, and value. Coastal erosion can result in diminished beach and dune quality and may give rise to situations where the beach transformed into a construction zone or is littered with debris and dilapidated buildings. Policies to manage erosion can include beach replenishment, shoreline armoring, or coastal retreat, each of which may garner different levels of support among the general public and impact choice, experience, and value of recreational users.

Beach replenishment involves periodically replacing eroded beach and dune sand; while this can improve beach width and dune height while augmenting beach habitat, it may have negative impacts on other beach quality measures (texture, color, etc.) and can induce negative environmental impacts at the sand mining or placement sites. Shoreline armoring can provide protection to coastal development and affect the distribution of sand, but often has negative environmental and aesthetic impacts (including destruction of habitat, disruption of sand sharing across parts of the beach, and loss of beach width in some areas). Along sandy beaches, replenishment is usually conducted in conjunction with armoring in order to ameliorate some of these negative effects. In contrast to active management of beach resources, coastal retreat is a passive management approach that entails moving structures and infrastructure to adapt to an evolving coastline; while this approach attempts to allow a natural barrier island system to persist, it can restrict access and enjoyment by limiting the extent and quality of residential & commercial development and public infrastructure that supports beach recreation and leisure activities. The process of retreat, moreover, can be unsightly, as buildings and infrastructure are moved or demolished.

Each of these policies induces an array of benefits and costs that may have different impacts on concerned parties. The interests of coastal private property owners are perhaps more obvious than the preferences and concerns of recreational users and the greater public. Since erosion control projects typically entail use of public funds and can have significant impact on the quality of natural resources (beaches, water quality, ecological habitat, etc.), the overall support for different approaches to coastal erosion management, their impact on recreation and tourism patterns, and the benefits and costs engendered by the approaches should be evaluated. Our objective in this paper is to assess economic benefits and costs of coastal erosion management policies to provide for a better understanding of tradeoffs that relate to the impact of erosion control on the recreation and tourism sectors and the public at large.

Literature

Early reviews of economic values related to beach resources and management of coastal erosion include Freeman (1995) and Bin, et al. (2005). We thus focus primarily on the literature since. Of particular note is the introduction of dynamic optimization models designed to identify optimal rotation times for beach replenishment – identifying efficient quantities and scheduling of sediment restoration activities (Landry 2008, 2011; Smith, et al. 2009), exploration of spatial externalities among communities engaging in beach replenishment (Slott, Smith, and Murray 2008; McNamara, Murray and Smith 2011; Lazarus, et al. 2011), and political economy models of coastal development and abandonment (McNamara and Keeler 2012). A recent paper by McNamara, et al. (2015) examines the effects of stochastic coastal storms, replenishment costs, erosion rates, and federal replenishment subsidies on optimal beach rotation and the resulting property values. They find that subsidized replenishment projects can inflate property values 9% - 34% (McNamara, et al. 2015)!

In addition to information on geomorphology and atmospheric conditions, dynamic optimization models require inputs on the economic value of beach sediments and the economic costs of manipulating the sediment budget to provide more beach and dune space. The hedonic property price valuation method has been used to estimate marginal willingness to pay for increments in beach width (Pompe 2008; Landry and Hindsley 2011; Gopalakrishnan, et al. 2011; Landry and Allen 2015). Estimates suggest that homeowners are willing to pay $22 to $1400 for an additional foot of beach width, and economic values are influenced by levels of overall beach width and baseline property values. Gopalakrishnan, et al. (2011) explore the potential role of endogeneity in marginal values under the assumption that they play a role in benefit-cost analysis of beach replenishment. Their findings suggest downward bias in conventional hedonic price estimates of beach width; corrected estimates can be as large as $8800 for an additional foot of beach width.

Landry and Allen (2015) explore the proximity effect of beach quality capitalization in coastal home prices, finding that homes within the bounds of 1600 to 2200 feet from the coast exhibit evidence of beach width capitalization, with inverse-distance and discrete-distance (i.e. dummy variable) interaction terms showing the best fit to the proximity effect. Further, they find little evidence of errors-in-variables (i.e. endogeneity) when estimating spatial regression models, though their data does not include complications arising from beach replenishment (i.e. potential for reverse causation) because their study site was not actively managed. Ranson (2012) analyzes a large set of transactions in Florida, where beach replenishment is widespread. His results suggest that marginal increases in beach width have at best minor influences on property value, unless the beach is extremely narrow.

Recreation demand models have been used to analyze preferences for beach trips and the influence of trip attributes on economic value. These measures can also play a prominent role in management and optimization models. Lew and Larson (2008) use information on individual labor market choices and trips to San Diego, CA beaches to estimate a repeated nested logit model of whether and where to go to the beach, while permitting an endogenous and jointly estimated shadow value for travel time. They find positive utility associated with the beach length, and estimate compensating variation around $22 per day. Whitehead, et al. (2008, 2010) utilize the single-site demand equation approach, but combine data on revealed and stated trips in order to assess economic measures of welfare stemming from North Carolina beach visitation and how values are influenced by beach access and beach width. They estimate consumer surplus of around $90 per trip, increasing by $25 with improvements in beach access and $7 with wider beaches (2008). Estimates of economic value, however, are sensitive to the choice of recreation demand model (2010). Using contingent valuation, Oh, et al. (2008) estimate visitors to South Carolina beaches are willing to pay $6.60/day for additional beach access points and parking spaces.

To develop a model of individual choice among participation, activity, and beach site in California, Pendleton, et al. (2012) use a nested discrete choice framework. Importantly, they find that the value of beach width varies systematically by activity, with sand-based activities exhibiting the largest value relative to water-based or pavement-based activities. Increasing beach width enhances economic value, but only up to moderate width levels. They estimate consumer surplus ranges between $13 and $74 per trip, and aggregate welfare measures for increasing beach width in the Los Angeles area by 50% are over $3 million per year. Most similar to our study, Parsons, et al. 2013 combine revealed and stated preference on beach use in Delaware to assess changes in beach width. They estimate the value of Delaware beach visits at $81/ trip for those that stay overnight and $33/ trip for single-day trips. Welfare losses from narrowing of the beach width by one-quarter its current with are about $5 per day, and doubling the current beach width increases economic value by about $3 per day.

Landry (2011) outlines the components of an economic cost function for beach replenishment, but there has been little research on this topic (despite the existence of extensive archival data at Western Carolina University’s Program for the Study of Developed Shorelines (PSDS)). An important component of beach replenishment is the potential for external costs imposed upon the surrounding environment (Speybroeck, et al. 2006). These have received little attention in the literature, with a notable exception being Huang, Poor, and Zhao (2007), which examines economic costs of wildlife impacts associated with beach replenishment; residents of New Hampshire and Maine are willing to pay on the order of $4 to $6 per household to prevent these impacts.2 Likewise, there has been little research on households’ willingness to pay or support for the difference ways that coastal erosion can be managed – shoreline armoring, beach replenishment, and shoreline retreat. These distinct responses can be viewed as a subset of adaptive measures that can be employed to manage sea-level rise – each invoking unique profiles of costs and benefits over time, accruing to different classes of property owners, public resources, beach visitors, and concerned non-users.

Our project seeks to provide information on household preferences for the different approaches to managing coastal erosion, while incorporating potential negative environmental impacts and variations in beach width. We explore the influence of variability in beach width, environmental impacts, and individual characteristics (e.g., environmental attitudes, political ideology, education, income, etc.) on willingness-to-pay for beach replenishment, shoreline armoring, and coastal retreat by beach users and non-users. Moreover, we test whether coastal erosion policies and environmental conditions affect intended visitation, particularly focusing on the effects of beach width, presence of shoreline armoring structures, and related negative environmental impacts.

Data

We collect data on visitation to and management of the North Carolina ocean beaches. Shoreline armoring has been proscribed in the state since the 1980s, though some oceanfront properties facing high erosion risk use large sand bags to protect their homes. The North Carolina legislature, however, recently approved the use of terminal groins,3 and some suspect that they may permit more hardening of the shoreline in the near future. Beach replenishment is conducted along many parts of the North Carolina coast; some are a part of ongoing federal projects that were approved decades ago (e.g. Wrightsville Beach) and others funded locally (e.g. Carteret County). Like most parts of the U.S. coast, shoreline retreat is typically only employed as a measure of last resort, but there are parts of the North Carolina shore that have had to embrace retreat (e.g. South Nags Head).

Our data were collected through contract with Online Sampling Solutions, Inc. in late fall of 2014. The data provider gave us access to their Research Now panel,4 which is designed for research purposes, actively managed, and recruited using standard marketing research techniques. These sorts of large Internet panels provide for much better response rates than mail or phone surveys and have known characteristics (e.g. income, education level) that allow

2 On the other hand, Shivlani, Letson, and Theis (2003) find that willingness to pay for additional beach width that provides additional habitat to sea turtle in Florida increases willingness to pay by about 25% - from $2.22 v. $2.78 per household, per visit. 3 Groins are shore-perpendicular structures designed to protect downdrift land and/or trap sand. Terminal groins are found at the end of the beach, often adjacent to waterways. 4 www.researchnow.com for analysis of sample response bias. We received a 61% response rate, and our data compares favorably to the population based on observable collected via U.S. Census (more on this below).

The survey questionnaire included a short pre-amble to describe the resource problem under study:

This survey is about North Carolina beaches.

Beaches provide for storm protection and coastal recreation. Wind, waves, currents, storms, and changing sea levels have contributed to the erosion of coastal beaches. About 75% of North Carolina beaches have eroded an average of 2.7 feet per year in the past 20 years. Between 1% and 2% of the North Carolina coastline has no dry sand at high tide.

Our survey instrument collects information on subjects’ knowledge of coastal processes (trends in seal level and damage due to coastal storms), beliefs about coastal erosion, and attitudes toward federal, state, and local governments’ management of erosion. All respondents are asked to describe their level of concern (using Likert scales) over beach width, protection of coastal properties, and the use of public money for erosion management. Our stated preference survey is designed to assess a cooperative state program that would pursue a concerted strategy to manage erosion on all North Carolina beaches. The proffered strategies are beach replenishment, shoreline armoring in conjunction with beach replenishment, and shoreline retreat. Each of these was to be evaluated relative to the status quo, which was described as ”No state program - coastal communities would pursue limited beach nourishment and continued erosion. Overall coastal beach width would continue to decline, leading to loss of beach area for recreation and tourism and loss of natural beach habitat.”

Table 1: Details of Beach Erosion Management Alternatives Beach Erosion Description of Policy Management Scenarios Beach Replenishment Public funding for replenishment of North Carolina beaches. This policy will improve beach width, providing for improved recreation, storm protection, and ecological habitat. Shoreline Armoring (with Public funding for shoreline armoring in conjunction with replenishment) replenishment of North Carolina beaches. This policy will protect oceanfront development and improve beach width, providing for improved recreation, storm protection, and ecological habitat. Coastal Retreat Public funding for coastal retreat along North Carolina beaches. This policy will provide funds for relocating building and infrastructure which will improve beach width, providing for improved recreation, storm protection and ecological habitat. Note: Each program is evaluated relative to the status quo of limited beach nourishment and continued erosion. “Overall coastal beach width would continue to decline, leading to loss of beach area for recreation and tourism and loss of natural beach habitat.”

For each possible strategy, a brief description of the approach was provided, as well as potential negative environmental impacts. Table 1 provides succinct summaries. (See the Appendix for additional details.) Each respondent was asked to describe their level of support for each of the coastal management strategies and the status quo (ranging from strongly oppose to strongly support).

As we are interested in both use and non-use values, we make a distinction between beach users and non-users. We classify respondents as beach users if they reported at least one trip to East Coast beaches in previous 24 months. While the distinction is somewhat arbitrary, this screening question permits us to gather detailed trip information only from those that indicate they’ve recently been to an east coast beach. For these subjects, we collect information on trips to North Carolina Beaches to other East Coast beaches in previous 12 months. While 12- month recall is demanding of subjects, we provide a detailed map of the North Carolina coast to help jog memory and our data collection period (late fall) was chosen to help respondents book-end the peak summer season when most trips are taken. In addition to revealed preference trips, we inquire about stated preferences. Each respondent is asked to state their expected future trips to North Carolina and other East Coast beaches over next 12 months under current conditions (average NC beach width of about 100 feet), and expected trips under a scenario in which North Carolina beaches are heavily eroded to an average width of 30 feet. Each beach width is depicted with color photos that include a single-person for scale. (See the Appendix.)

Table 2: Experimental Design for Contingent Valuation Study Environmental Impacts Minimal Negative I. Wide beaches (e.g., 50 foot II. Wide beaches (e.g., 50 foot Wide increase); minimal environmental increase); negative impacts associated with environmental impacts Beach Width management strategy associated with management strategy III. Wider beaches (e.g., 150 foot IV. Wider beaches (e.g., 150 foot Wider increase); minimal environmental increase); negative impacts associated with environmental impacts management strategy associated with management strategy

In assessing each specific erosion management strategy, we utilized a ‘between’ research design, wherein approximately one-third of sample was allocated to ‘nourish’, ‘armor’, and ‘retreat’. For each scenario, we inquire about households’ stated preference trips using the experimental design in Table 2. Each respondent is systematically assigned to one of the cases I – IV for their first SP trips estimation. A follow-up SP trip estimation is also collected moving horizontally or vertically in the table (thus adding a ‘within’ dimension to the trip data). For example, a respondent assigned to assess stated preference trips for ‘nourish’ Case I (wide beaches with minimal environmental impacts) would receive a follow-up question requesting them for stated trips under Case II or III (either including negative environmental impacts or assessing a wider beach, respectively). Thus, we collect a quasi-panel dataset with revealed preference trips under current conditions, stated preference trips under current conditions, and stated preference trips under scenarios involving changes in beach width and presence or absence of negative environmental impacts associated with erosion management. Since our data include beach users and non-users, some respondents exhibit zero demand for all trips.

In addition, we include a series of contingent valuation (CV) questions to assess willingness to support beach erosion management. Each respondent is asked a dichotomous choice question (Yes/No/Undecided) regarding their willing to vote in support of the erosion management plan to which they were assigned (‘nourish’, ‘armor’, or ‘retreat’). The payment vehicle was described as an increase in beach property taxes (2 cents per $100 value) accompanied by an increase in overall state income tax. The inclusion of property taxes was incorporated to account for perceived inequity that was revealed during pre-testing. Our data includes information on beach homeowners, so we are able to control both sources of payment in our analysis. The randomly assigned bid levels were $4, $28, $49, $81, and $114 per household, per year, and the survey included the following text to enhance incentive compatibility (Landry and List 2006):

Imagine that you have the opportunity to vote on the proposed coastal erosion management plan, ______. If more than 50% of North Carolina households vote for the plan then it would be put into practice.

Sometimes when people are asked to evaluate a proposed policy like this one, it is easy for them to say they support a policy either because they are not being asked to pay at the same time, or they don’t think they will have to pay based on their response.

We want you to only respond with what you actually think you would do given the beach impacts and the estimated cost to your household.

Also consider your personal income and current payment obligations. If you vote for the policy then you would have ______less to spend on other things each year. If you pay property taxes on a beach house you would have even less to spend on other things each year.

There is no right or wrong answer but results from this study will be shared with North Carolina coastal policy makers.

The first blank was filled with their assigned policy scenario (‘nourish’, ‘armor’, or ‘retreat’), and the second blank was filled with their randomly assigned bid.

The beach erosion management scenario that they initially evaluated corresponded with their initial SP trip estimation from Table 2. Immediately following their dichotomous choice, they were asked to state their level of certainty in their response, after which they were asked to evaluate a table of all bid levels (including a lower bid of $2 and a greater bid of $163) and indicate their likelihood of paying all the possible tax increases. Two following CV questions inquired about their likelihood of voting to support the proffered program with a different level of beach width and with or without negative environmental impacts (effectively assessing three of the four cells in Table 2).

Lastly, respondents were asked to indicate their level of agreement with statements regarding the suitability of voting referenda to influence coastal erosion policy, the likelihood that results of the survey would be shared with North Carolina policy makers, and the likelihood that the results of the survey would influence policy makers. We also measured the self-assessed understanding in the information contained in the survey and their confidence in the ability of North Carolina State Government to achieve the goals of the proffered beach erosion policy. Lastly, standard demographic data were collected.

Descriptive statistics are presented in Table 3. Eighty-four percent of our respondents met our classification as a beach user (having visited an East Coast beach within the previous 24 months). About 80 percent planned to make a trip to an East Coast beach in the next 12 months under current conditions, and only 52 percent would make a trip to a North Carolina beach if beach width were eroded from an average of 100 feet to 30 feet. When queried about the support for the various beach erosion management strategies, similar proportions (around 46%) supported or strongly supported the three options identified – beach replenishment, shoreline armoring, and shoreline retreat - while only 16% supported the status quo (limited beach replenishment and continued erosion). The SP visit indicator variables identify the proportions of the data allocated to each treatment and the effects of negative environmental impacts on visitation for each treatment. Sixty-six percent of the 335 households allocated to the replenishment treatment indicated that they would visit a North Carolina beach in the next 12 months under this erosion management strategy, but only 38% would visit if there were negative environmental impacts (impacts on benthic communities and short-term effects where sand is placed). Fifty-nine percent of the 338 household allocated to shoreline armoring intend to visit a North Carolina beach, but this proportion drops to 23% when there are negative environmental impacts (i.e. loss of beach sand). For shoreline retreat, 69% of the 335 households would make a beach trip, falling to 34% with negative environmental impacts (debris on the beach).

Table 3 Descriptive Statistics Variable | Obs Mean Std. Dev. Min Max ------+------visitor | 1000 .834 .3722668 0 1 sp_visit | 1000 .797 .4024338 0 1 sp_visit_30 | 1000 .52 .4998498 0 1 ------+------favor_n | 1000 .4666667 .499136 0 1 favor_a | 1000 .4616915 .4987785 0 1 favor_r | 1000 .4517413 .4979134 0 1 favor_sq | 1000 .160199 .3669731 0 1 ------+------sp_visit_indicators nourish_env | 335 .6656716 .4724608 0 1 nourish_impact | 327 .3761468 .48516 0 1 armor_env | 338 .5946746 .4916828 0 1 armor_impact | 338 .2307692 .4219497 0 1 retreat_env | 335 .6925373 .462133 0 1 retreat_impact | 335 .3432836 .4755154 0 1 ------+------male | 1000 .426 .4947411 0 1 hh_size | 1000 2.321 1.083112 1 10 children | 1000 .341 .792052 0 8 h_school | 1000 .065 .2466492 0 1 college | 1000 .359 .4799472 0 1 ------+------grad | 1000 .327 .469352 0 1 income | 1000 78.4275 59.92004 0 275 (NOTE: median is 62.5) white | 1000 .872 .334257 0 1 env | 1000 .073 .2602667 0 1 liberal | 1000 .17 .3758208 0 1 ------+------moderate | 1000 .395 .4890953 0 1 conservative | 1000 .344 .4752787 0 1 beachhouse | 1000 .05 .218054 0 1 ------+------Table 3 also includes demographic variables. Forty-two percent of respondents were male, with a household size of 2.3 persons. U.S. Census 2010 indicates 49% males and average household size of 2.5 persons for North Carolina. Whereas 68% of hour sample has a bachelor’s degree or greater educational attainment, the NC average is 26.5% (U.S. Census 2010). Median income for our sample is $62,500 (mean = $78,427), while the NC median is $46,291 (U.S. Census 2010). Eighty-seven percent of our sample reported being white (NC average is 72%). Over 7% of respondents indicated membership in an environmental organization. Seventeen percent self-identified the political view as liberal, while 39% (34%) reported moderate (conservative) political views; the remaining 10% indicated ‘none of the above’ for political view. Lastly, 5% of our respondents owned a home or mobile home (either outright or as a time-share) in a beach/coastal community in North Carolina.

Table 4 Trip Statistics . sum nc_trips ec_trips if rp==1

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 1000 3.588 10.19766 0 150 ec_trips | 1000 1.38 6.685679 0 200

. sum nc_trips ec_trips if sp==1

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 1000 3.507 10.32049 0 150 ec_trips | 1000 1.281 6.693492 0 200

. sum nc_trips ec_trips if sp_30==1

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 1000 3.202 10.01616 0 150 ec_trips | 1000 1.503 6.694087 0 200

. sum nc_trips ec_trips if sp_width==1 & nourish==1 & env_improve==1 & bw150==0

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 169 5.195266 18.26533 0 200 ec_trips | 169 1.349112 2.445115 0 24

. sum nc_trips ec_trips if sp_width==1 & nourish==1 & env_improve==1 & bw150==1

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 158 4.177215 13.46972 0 150 ec_trips | 158 1.310127 2.071652 0 15

. sum nc_trips ec_trips if sp_width==1 & nourish==1 & env_improve==0 & bw150==0

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 169 4.325444 18.26021 0 200 ec_trips | 169 1.260355 2.012675 0 15

. sum nc_trips ec_trips if sp_width==1 & nourish==1 & env_improve==0 & bw150==1

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 158 2.316456 6.287272 0 50 ec_trips | 158 1.303797 2.270579 0 15

. sum nc_trips ec_trips if sp_width==1 & armor==1 & env_improve==1 & bw150==0

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 166 2.512048 5.465859 0 50 ec_trips | 166 1.427711 2.785931 0 20

. sum nc_trips ec_trips if sp_width==1 & armor==1 & env_improve==1 & bw150==1

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 172 3.680233 12.66976 0 150 ec_trips | 172 1.377907 1.800643 0 11

. sum nc_trips ec_trips if sp_width==1 & armor==1 & env_improve==0 & bw150==0

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 166 2.198795 9.201594 0 100 ec_trips | 166 1.46988 2.756095 0 20

. sum nc_trips ec_trips if sp_width==1 & armor==1 & env_improve==0 & bw150==1

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 172 2.52907 12.5061 0 150 ec_trips | 172 1.27907 1.880346 0 11

. sum nc_trips ec_trips if sp_width==1 & retreat==1 & env_improve==1 & bw150==0

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 168 4.053571 11.37981 0 100 ec_trips | 168 2.309524 15.49733 0 200

. sum nc_trips ec_trips if sp_width==1 & retreat==1 & env_improve==1 & bw150==1

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 167 4.035928 10.62272 0 100 ec_trips | 167 1.389222 3.525894 0 42

. sum nc_trips ec_trips if sp_width==1 & retreat==1 & env_improve==0 & bw150==0

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 168 2.660714 9.758663 0 100 ec_trips | 168 2.303571 15.50382 0 200

. sum nc_trips ec_trips if sp_width==1 & retreat==1 & env_improve==0 & bw150==1

Variable | Obs Mean Std. Dev. Min Max ------+------nc_trips | 167 1.916168 5.049649 0 36 ec_trips | 167 1.299401 2.685335 0 20

Table 4 depicts detailed beach trip statistics. The average RP trips for the previous 12 months are 3.6 for North Carolina beaches and 1.4 for other East Coast beaches. SP trips for the next 12 months under current conditions are 3.5 for NC beaches and 1.3 for East Coast beaches. The SP erosion scenario (reducing NC beach width from 100 feet to 30 feet) has only a modest effect on the average of trips – 3.2 NC trips and 1.5 East Coast trips. NC SP trips under the replenishment scenario are generally greater (4.2 to 5.2 trips per year), unless the beach is not as wide and there are negative environmental impacts (2.3 trips per year). Planned trips under the shoreline armoring scenario are generally lower (around 2.5 trips per year), unless negative environmental impacts (loss of beach sand) can be ameliorated. Planned trips under shoreline retreat are around 4 per year, unless the beach is sometimes littered with debris, in which case annual trips are lower (2 – 2.5 trips).

Table 5 Contingent Valuation Statistics . sum nc_trips ec_trips if rp==1

Variable | Obs Mean Std. Dev. Min Max

Table 5 presents descriptive statistics for the contingent valuation data. Proportions of respondents that support or strongly support the hypothetical management plan they are asked to evaluate are 67% for beach replenishment, 58% for shoreline armoring, and 71% for shoreline retreat. Across all treatments, about 40% are concerned or very concerned about the costs of the erosion management plan. Across all three CV referenda, about 29% voted to support the erosion management policy, and 32% were certain or very certain about their vote (whether it was ‘yes’ or ‘no’). Seventy percent of respondents support the idea of a referendum to decide on coastal erosion management, and 67% believed that the results would be shared with policymakers. Fifty-four percent perceive that the results of the survey could have consequences for what policies are actually implemented, and a similar percentage have confidence in the state government and local communities to effectively implement the policy. Almost 80% of respondents indicate that they understood all information presented in the survey.

Methods and Results

This section presents an initial analysis of the data. For the recreation demand data, we have a quasi-panel composed of stacked RP/SP responses and treatment effects (Landry and Liu 2009). We estimate the following Multivariate Poisson-Gamma (AKA Random Effects Poisson) travel cost model: ⎛⎞J −⎜⎟y +α JJ⎜⎟∑ ij ⎛⎞⎛⎞α ⎝⎠j=1 Γ⎜⎟⎜⎟yij+αα µ ij +α y ∑∑ J µ ij fy(|) x ⎝⎠⎝⎠jj==11 ij , ii= ∏ Γ()α j 1 y ! = ij where yij is NC beach trips for household i under conditions j, μij = exp(β’xij) is the conditional mean of trips, α is a scale parameter, and xij includes travel cost to NC beach, travel costs to substitute sites (Virginia Beach and Myrtle Beach), management treatment effects (armor, retreat; nourish = excluded), cubic function of beach width [Beach width, (beach width)2, (beach width)3], environmental impacts of management indicator variables, income and demographics. The results are presented in Table 6.

Table 6 Random Effects Poisson Regression Results for NC Beach Trips Variable Coefficient Standard Error Marginal Effect nc_tc -0.0065353 0.0006785 -0.01425

vb_tc 0.0020399 0.0005736 0.004447

mb_tc 0.0019237 0.0007749 0.004194

state_p -0.0242954 0.0242669 ------

armor -0.28332 0.1006445 -0.59186

retreat 0.1995368 0.1009107 0.45031

bw 0.0241212 0.0023993 0.052582

bw2 -0.0002376 0.0000226 -0.00052

bw3 5.84E-07 5.62E-08 1.27E-06

env_impact -0.3991037 0.0414034 -0.98736

enva -0.2314674 0.0497889 -0.43498

envr -0.1120093 0.045588 -1.23721 ------linc 0.0975267 0.0787062

Statistically significant covariates are in bold. Travel costs to NC beaches are negative and statistically significant, while travel costs to substitutes sites are positive and statistically significant. Relative to beach replenishment, respondents take fewer trips under the shoreline armoring scenario and more trips under the shoreline retreat scenario. The cubic function of beach width is statistically significant. Negative environmental impacts reduce the number of trips, and this effect is greater for armoring and retreat (relative to beach replenishment). The natural log of income and the SP indicator are not statistically significant. Marginal effects can be explored by examining the effect on the fitted mean. RP baseline trips = 3.12, and Beach Replenishment trips = 2.99, which are reduced by 0.98 when there are negative environmental impacts (habitat disruption, turbidity, burial of organisms). The conditional mean of trips is 2.25 under shoreline armoring, which is further reduced by 0.43 when beach width is lost and habitat disrupted. The mean of trips under shoreline retreat is 3.65, but goes down by 1.24 if the beach is littered with debris for some period of time. Turning to economic welfare measures, RP / SP baseline WTP is $477 per household/year; WTP under beach replenishment is $457 per household/year (reducing to $306 per household/year when there are negative environmental impacts); WTP under armoring is $344 per household/year ($291 with negative environmental impacts); WTP for shoreline retreat is $558 per household/year ($334 with negative environmental impacts). For the CV data, we use the exponential willingness-to-pay formulation with a random effects Probit model. For out preliminary results, we consider the first bid only (not payment menu), and only count ‘yes’ votes that are declared ‘somewhat certain’ or ‘very certain’. Our specification includes tax level, treatment effects, interactions, political ideology, and demographic variables.

Table 7 Random Effects Probit Model of Exponential Willingness to Pay for Beach Erosion Control Policies Variable Coefficient Standard Error

ln(tax) -0.7535466 0.084933 armor -0.5396919 0.277094 retreat 0.1634383 0.273948 bw150 0.0365666 0.151742 env_impact -0.5801119 0.179738 enva 0.3060154 0.215893 envr 0.3770403 0.21346 linc 0.4100704 0.163629 liberal 1.419038 0.435158 moderate 1.002995 0.402589 conservative 0.3131694 0.408564

Results indicate that the probability of voting for coastal erosion control policies is decreasing in the natural log of the tax amount. Shoreline armoring receives lower support, while shoreline retreat is not statistically different from beach replenishment. Negative environmental impacts decrease the probability of voting for an erosion control program; the negative environmental effects of shoreline armoring are not significantly different from replenishment, but negative effects associated with shoreline retreat have less of a negative impact on the probability of voting to support erosion management. The probability of supporting erosion control is increasing in the natural log of income and higher for those that view themselves as liberal or moderate. Assessing economic welfare stemming from CV votes implies that the average household is WTP $18.84 per household/year for beach replenishment (decreasing to $8.77 if there are negative environmental impacts). WTP for armoring is $9.20 per household/year ($6.40 with negative environmental effects), while WTP for shoreline retreat is estimated at $23.41 per household/year ($17.88 with negative environmental impacts). The next step in analysis is to combine the recreation demand and contingent valuation data in a joint model of visitation and voting support for erosion management. We will use the framework of Eom and Larson (2007) to estimate this model and test for weak complementarity. We expect that beach erosion management can affect non-use values for some portion of stakeholders, and our sampling protocol will capture information from users and non-users of North Carolina coastal resources. As such, we adopt a research design that can account for use and non-use values. We specify individual utility (i.e., happiness) as u = u( y,z,q ), where y is a vector of recreation trip counts for North Carolina beaches and other East Coast beaches, z is a numeraire good, and q represents erosion management regime and environmental quality for beach recreation sites (which cannot be freely selected at the individual level). The budget constraint is given by m = yʹc + z , where, m is income, c is a vector of travel costs – a combination of explicit (gas and vehicle wear-and-tear) and implicit (opportunity cost of time) costs of travel to a site – and z is a vector of market goods with price normalized to unity. We rewrite the utility function u(y, z, q) = G(u*(y, z, q), q) to indicate that environmental quality may influence utility through two avenues – the consumption of complementary goods (recreation trips, y) and a pure non-use benefit, independent of the consumption decisions for other goods and services (Eom and Larson 2006; McConnell 2011). For CVM data, response patterns will be analyzed using discrete choice panel data models, such as fixed effects logit or random effects probit (Haab and McConnell 2002). These models account for individual level unobserved heterogeneity and can be estimated with standard econometric software packages. Our models will be specified so that we can estimate mean and median willingness to pay for each of the coastal erosion management programs specified in table 1, and covariates will be included to account for beach width, environmental impacts, and individual level factors (such as knowledge, perceptions, political ideology, and demographic factors). Tests of model parameters will provide evidence of the influence of each of these factors on provisional voting patterns and economic value (willingness to pay). We specify the demand function for recreation trips to NC beaches as yNC = f(c, q, m). With revealed preference and contingent behavior data, we can estimate joint revealed and stated preference recreation demand models. Pooling trips to North Carolina sites, we can estimate stacked revealed and stated preference single-site demand models and utilize information on regional visitation to other locations (other East Coast beaches) to help identify the effect of substitute sites. Analysis of the combined revealed preference and contingent behavior data will inform expectations of beach management impacts on recreational visitation, how these impacts vary across user types, and will allow for estimation of economic benefits and costs. Landry and Liu (2011) review a number of econometric approaches, each exhibiting varying degrees of flexibility and accounting of individual heterogeneity, that can be used to analyze these type of data, which are commonly referred to as ‘quasi-panel’ and ‘stacked frequency demand’. To different extents, these models are robust to mis-specification and can enable testing of critical hypotheses about individual preferences and model assumptions. Model parameters can be used to estimate the net economic value (willingness to pay) of beach visitation under status quo conditions and how economic value changes with management regimes, beach width, and negative ecological impacts. By combining contingent valuation and contingent behavior data, we can employ the same behavioral models, but make use of consumer theory to relate observed demand to potential non-use value. We can integrate the NC beach demand equation (yNC = f(c, q, m)) over travel cost to recover the quasi-expenditure function, E( c,q,G -1 ( q,u)), which gives the level of money necessary to achieve utility level u, as a function of prices, beach erosion management regime, and environmental quality. Integration introduces a constant term, which can be theoretically defined to reflect potential non-use value, and information from the contingent valuation and contingent behavior data can be used to estimate the parameters of demand and non-use value and test for certain preference restrictions (e.g. weak complementarity) (Eom and Larson 2006). Utilizing the stacked revealed and stated preference single-site demand model, we can combine data responses to estimate demand and non-use parameters for North Carolina beaches using a similar approach to Eom and Larson (2006) (albeit with additional information from non-users and variability in management approach, beach width, and environmental impacts in the pseudo time-series dimension). We will also explore other possibilities for modeling beach recreation demand, such as the generalized corner solution model (Phaneuf, Kling, and Herriges 2000). Our analysis will examine convergent validity of different econometric models of recreation behavior (Whitehead, et al. 2010). In addition, these models will provide a baseline for possible future tests of predictive validity of contingent behavior data in comparison with actual visitation changes (Whitehead 2005).

Conclusions

For general citizenry, preliminary estimates indicate welfare losses stemming from armoring, but gains from shoreline retreat. Economic costs of erosion management don’t appear to be negligible: $150 for benthic habitat, turbidity, burial (nourishment); $45 for loss of beach at high tide (armor); $220 for loss of habitat, debris on beach (retreat). The results suggest that organized adaptation to shoreline change has considerable support amongst voters and deserves more study. Benefit-cost analyses regarding allocation of coastal resources typically include information on the benefits that accrue in markets for tradeable goods and services – such as real estate and tourism. Nonmarket benefits and costs, including recreation values over and above tourist expenditures and environmental values held by the general public, are more difficult to measure. Oftentimes, nonmarket values go unmeasured and are not included in policy analyses giving more weight to market values. This project will provide relevant and timely information on the broader impacts of erosion management polices by examining the preferences, concerns, and behavior of NC citizens and beach visitors. Information on benefits and costs accruing to distinct groups will allow for a more comprehensive assessment of erosion management policies. The benefit-cost analysis will provide a ranking of the net benefits of erosion management policies while considering their long term environmental and economic sustainability. Important categories of benefits and costs will be measured for the first time (Landry 2011). Empirical estimates from this project can play a central role in dynamic optimization models for coastal erosion management of the North Carolina shoreline (Landry 2008; Smith, et al. 2009; Landry 2011).

References

Bin, Okmyung Craig E. Landry, Chris Ellis, and Hans Vogelsong. 2005. “Some Consumer Surplus Estimates for North Carolina Beaches” Marine Resource Economics 20(2): 145-61. Carson, R. and T. Groves, “Incentive and informational properties of preference questions,” Environmental and Resource Economics 37(1):181-210, 2007. Chang, L. and J.A. Krosnick, “National Surveys Via RDD Telephone Interviewing Versus the Internet: Comparing Sample Representativeness and Response Quality,” Public Opinion Quarterly, 73(4):641-678, 2009. Cummings, R.G., and L.O. Taylor, “Unbiased Value Estimates for Environmental Goods: Cheap Talk Design for the Contingent Valuation Method,” American Economic Review 89(3):649–65, 1999. Cummings, R.G. and L.O. Taylor, “Does Realism Matter in Contingent Valuation Surveys?” Land Economics 74(2):203-15, 1998. Eom, Y.S. and D.M. Larson, “Improving environmental valuation estimates through consistent use of multiple information,” Journal of Environmental Economics and Management, 52: 501–16, 2006. Freeman, A.M. 1995. The Benefits of Water Quality Improvements for Marine Recreation: A Review of the Empirical Evidence. Marine Resource Economics 10(4): 385-406. Gopalakrishnan, S., M.D. Smith, J.M. Slott, and A.B. Murray, “The Value of Disappearing Beaches: A Hedonic Pricing Model with Endogenous Beach Width,” Journal of Environmental Economics and Management 61(3): 297-310, 2011. Haab, T.C., and K.E. McConnell, Valuing Environmental and Natural Resources: The Econometrics of Non-Market Valuation, Edward Elgar Press, 2002. Hamilton, Jacqueline. 2007. “Coastal Landscape and the Hedonic Price of Accommodation” Ecological Economics 62: 594-602. Landry, C.E., 2008. Optimal Erosion Management on Developed Barrier Island Beaches, Working Paper East Carolina University: Greenville, NC. Landry, Craig E. 2011. “Coastal Erosion as a Natural Resource Management Problem: An Economic Perspective” Coastal Management 39(3): 259-78. Landry, C.E. and P. Hindsley, “Valuing Beach Quality with Hedonic Property Models” Land Economics 87(1): 92-108, 2011. Landry, C.E. and J.A. List, “Using Ex Ante Approaches to Obtain Credible Signals for Value in Contingent Markets: Evidence from the Field,” American Journal of Agricultural Economics 89(2):420-29, 2007. Landry, C.E. and H.Liu, “Econometric Models for Joint Estimation of Revealed and Stated Preference Site-Frequency Recreation Demand Models,” Chapter 7 in Whitehead, John C., Timothy C. Haab and Ju-Chin Huang, editors, Preference Data for Environmental Valuation: Combining Revealed and Stated Approaches, Routledge, March 2011. Lazarus, E.D., D.E. McNamara, M.D. Smith, S. Gopalakrishnan, and A.B. Murray. 2011. “Emergent Behavior in a Couple Economic and Coastline Model for Beach Nourishment” Nonlinear Processes in Geophysics 18: 989-99. Lew, Daniel K. and Douglas M. Larson. 2008. Valuing a beach day with a repeated nested logit model of participation, site choice, and stochastic time value. Marine Resource Economics 23 (3): 233-252. McConnell, K.E., “Joint Estimation of Stated and Revealed Welfare Measures: The Conceptual Basis,” Chapter 2 in Whitehead, John C., Timothy C. Haab and Ju-Chin Huang, editors, Preference Data for Environmental Valuation: Combining Revealed and Stated Approaches, Routledge, March 2011. NcNamara, D.E., Sathya Gopalakrishnan, Martin D. Smith, and A. Brad Murray. 2015. “Climate Adaptation and Policy-Induced Inflation of Coastal Property Value” PLOS One (May 25) DOI:10.1371/journal.pone.0121278 Oh, Chi-Ok, Anthony W. Dixon, James W. Mjelde, and Jason Draper. 2008. Valuing visitors' economic benefits of public beach access points. Ocean and Coastal Management 51 (12): 847-53. Parsons, George R., Zhe Chen, Michael K. Hidrue, Naomi Standing, and Jonathan Lilley. 2013. Valuing beach width for recreational use: Combining revealed and stated preference data. Marine Resource Economics 28(3): 221-41. Pendleton, Linwood, Craig Mohn, Ryan K. Vaughn, Philip King, and James G. Zoulas. 2012. Size matters: The economic value of beach erosion and nourishment in southern California. Contemporary Economic Policy 30 (2): 223-37. Phaneuf, D.J., C.L. Kling, and J.A. Herriges, “Estimation and Welfare Calculations in a Generalized Corner Solution Model with an Application to Recreation Demand,” Review of Economics and Statistics 82:83-92, 2000. Pompe, J. 2008. The Effect of Gated Community on Property and Beach Amenity Valuation, Land Economics 83(4): 423-33. Ranson, Matthew, “What Are the Welfare Costs of Shoreline Loss? Housing Market Evidence from a Discontinuity Matching Design.” Discussion Paper 2012-07, Cambridge, MA: Belfer Center for Science and International Affairs, May 2012. Ready, R.C., P.A. Champ, and J.L. Lawton, “Using Respondent Uncertainty to Mitigate Hypothetical Bias in a Stated Choice Experiment.” Land Economics 86: 363-81, 2010. Riggs, S.R. and D.V. Ames, Drowning The North Carolina Coast: Sea-Level Rise and Estuarine Dynamics, North Carolina Sea Grant, 2003. Shivlani, M.P., D. Letson, and M. Theis, 2003. Visitor Preferences for Public Beach Amenities in South Florida, Coastal Management 31: 367-85. Slott, J.M., M.D. Smith, and A.B. Murray, 2008. Synergies between Adjacent Beach-Nourishing Communities in a Morpho-Economic Coupled Coastline Model, Coastal Management 36: 374-391. Smith, M.D., D. McNamara, J.M. Slott, and A.B. Murray. 2009. Beach Nourishment as a Dynamic Capital Accumulation Problem, Journal of Environmental Economics and Management 58(1): 58-71. Speybroek, J., Bonte, D., Courtens, W., T. Gheskiere, P. Grootaert, J.P. Maelfait, M. Maths, S. Provoost, K. Sabbe, E.W.M. Stienen, V. Van Lancker, M. Vincx, and S. Degraer. 2006. Beach Nourishment: An Ecologically Sound Coastal Defence Alternative? A Review, Aquatic Conservation: Marine and Freshwater Systems 16: 419–35. Whitehead, J.C., “Environmental Risk and Averting Behavior: Predictive Validity of Revealed and Stated Preference Data,” Environmental and Resource Economics, 32(2):301-16, 2005. Whitehead, J.C., T.C. Haab and J.C. Huang, “Measuring Recreation Benefits of Quality Improvements with Revealed and Stated Behavior Data,” Resource and Energy Economics, 22, 339-54, 2000. Whitehead, J.C., T.C. Haab and J.C. Huang, editors, Preference Data for Environmental Valuation: Combining Revealed and Stated Approaches, Routledge, March 2011. Whitehead, J.C., C.F. Dumas, J. Herstine, J. Hill, and B. Buerger, 2008. Valuing Beach Access and Width with Revealed and Stated Preference Data, Marine Resource Economics 23(2): 119-135. Whitehead, J.C., D. Phaneuf, C.F. Dumas, J. Herstine, J. Hill, and B. Buerger, 2010. Convergent Validity of Revealed and Stated Recreation Behavior with Quality Change: A Comparison of Multiple and Single Site Demands, Environmental & Resource Economics 45: 91-112.

Appendix:

A. Descriptions of Management Scenarios

1. Beach nourishment involves adding sand to the beach to increase beach width and build dunes. The beach will still erode over time so sand must be added about every 3 to 5 years to maintain width. Beach nourishment would be paid for by a public program that provides funds to periodically place sand on North Carolina beaches.

Beaches that have been nourished provide for greater storm protection for coastal communities and provide for recreational and leisure benefits for all visitors.

Potential negative environmental impacts from beach nourishment include:

• disruption of ocean-bottom habitats, which can affect turtles, fish communities, and small creatures like clams and crabs • increased cloudiness and alterations to sediment movement in coastal waters • burial of beach organisms and alteration of sand texture on the beach, which can affect plants, turtle nesting, shorebird habitat, and small creatures like clams and crabs Most of these effects are short term, but can present longer term problems if incompatible sand is mined from the wrong areas offshore.

Once complete beach nourishment can provide additional beach habitat for coastal animals and plants, such as sea turtles, shorebirds, dune mice, dune plants, and small creatures like clams and crabs.

2. Shoreline armoring uses concrete and rocks to protect coastal buildings, roads, and utilities. This process often erodes the beach and can increase erosion at other nearby beaches. Beach nourishment can be used to maintain beaches where this erosion occurs. Shoreline armoring and beach nourishment would be paid for by a public program that provides funds to construct and maintain shoreline armor and periodically place sand on North Carolina beaches.

Potential negative environmental impacts from shoreline armoring and beach nourishment include:

• disruption of ocean-bottom habitats, which can effect turtles, fish communities, and small creatures like clams and crabs • increased cloudiness and alterations to sediment movement in coastal waters • burial of beach organisms and alteration of sand texture on the beach, which can affect plants, turtle nesting, shorebird habitat, and small creatures like clams and crabs Most of these effects are short term, but can present longer term problems if incompatible sand is mined from the wrong areas offshore.

The negative impacts of shoreline armoring are more permanent and include:

• disruption of continuous beach • loss of beach habitat. Beaches would be wide, but could be interrupted by seawalls, rocks, and concrete. These negative impacts can be minimized through beach nourishment and careful project management.

3. Retreat and adaptation involves moving vulnerable buildings, roads and utilities out of harm’s way to adapt to the changing shoreline. This approach allows the beach to evolve naturally. Some debris can be temporarily left on the beach during the retreat. Parts of the beach may be inaccessible at times as buildings and infrastructure are moved. Shoreline retreat and adaptation would be paid for by a public program that provides funds to move buildings, roads, and infrastructure out of harm’s way and to offer partial compensation to those that lose their land. Retreat would only be implemented when shoreline erosion gets severe enough to require adaptation.

Potential negative environmental impacts of shoreline retreat and adaptation:

• temporary disruption of continuous beach • temporary loss of beach habitat • temporary presence of debris on some parts of the beach. Most of these effects appear to be short term. There are no negative impacts after the debris is removed. Negative environmental impacts can be minimized through careful project management.

B. Visual Aids

Description of Baseline: Along the North Carolina coastline, the average beach width is around 100 feet. The pictures below indicate what the current beach width looks like:

Erosion scenario (SP1): If nothing is done to combat beach erosion, beaches along the North Carolina coast could face significant erosion that could reduce beach width to 30 feet. The pictures below indicates an average beach width of 30 feet.

The picture below indicates the increase in beach width to 150 feet.

The picture below indicates the increase in beach width to 250 feet.