Unveiling the Non-Market Value of a Fragile Coastal Wetland: A CVM Approach in the Amvrakikos Gulf, Greece

Abstract

Wetlands are highly productive ecosystems with multidimensional value and significant social and economic impacts. Estimating the economic value of their non-marketed goods and services—benefits not traded in conventional markets—can provide essential insights to guide protection, restoration, and sustainable management strategies for these sensitive ecosystems. The present study employs environmental economics tools, specifically the Contingent Valuation Method (CVM), to assess the value of the Amvrakikos Gulf in northwestern Greece. This semi-enclosed wetland system is particularly fragile due to its low water renewal rate, while being a primary source of income and an integral component of local cultural identity. Despite its high ecological importance, the Amvrakikos Gulf has experienced substantial environmental degradation stemming from its geomorphological characteristics and external anthropogenic pressures. This investigation was designed to explore residents’ perceptions of the wetland’s value and its correlation with the need for restoration. In total, 383 coastal area residents participated in this study. Data analysis was conducted using appropriate econometric methods based on both parametric and non-parametric models. Approximately 46.2% of respondents expressed willingness to pay, and the environmental restoration of the Amvrakikos Gulf was valued at EUR 715,968.36. Additionally, this study examined potential associations between willingness to pay and various socio-cultural and demographic factors recorded during the interviews. In conclusion, the need for the restoration and preservation of the Amvrakikos Gulf’s natural wealth was made evident, affirming the contribution of the CVM in valuing wetlands and enriching the existing literature, while explicitly recognizing the subjectivity inherent in WTP assessments.

1. Introduction

Water resources worldwide are under mounting pressure from population growth and anthropocentric development, emphasizing the need for innovative management strategies. In this context, legal and policy frameworks have been introduced to facilitate their restoration, conservation, and protection. However, despite the globally recognized ecological and socio-economic importance of wetlands, their inclusion in water protection and integrated management strategies often remains inadequate.

Historically, wetlands were inaccurately perceived as ecosystems of limited utility, low productivity, and sources of disease transmission [1]. This misconception led to widespread drainage projects aimed at facilitating tourism, agriculture, and industrial exploitation. Over time, this perspective has been largely abandoned, with the impacts of degraded wetlands providing substantial evidence of their ecological, social, and economic value.

Wetlands exhibit significant variability in size, climatic conditions, geological and hydrological characteristics, and biodiversity, complicating efforts to develop standardized approaches to their study and management. Each wetland is distinct, necessitating a thorough understanding of its biotic and abiotic attributes as a fundamental step toward designing effective restoration and conservation strategies [2].

Coastal wetlands, in particular, function as biodiversity hotspots and complex ecological systems characterized by high productivity [3]. They serve as critical breeding and nesting grounds for fish, birds, and marine organisms [4], forming unique ecosystems that provide an array of ecosystem services, including hydrological cycle regulation, water quality improvement, flood mitigation, coastal erosion prevention, carbon sequestration, climate regulation, nutrient recycling, fisheries production, recreation, tourism, and others.

Economic activities such as fisheries, aquaculture, recreation, and tourism within coastal wetlands play a crucial role in local economic development, contributing to the reduction in regional inequalities and the demographic revitalization of coastal communities. Nevertheless, these ecosystems face intensifying pressures from anthropogenic activities and natural factors that degrade ecological integrity, cause pollution, and threaten local livelihoods. Furthermore, coastal wetlands are particularly vulnerable to the effects of climate change [5,6,7]. Alterations in precipitation patterns, temperature fluctuations, and rising sea levels disrupt the physical, chemical, and biological processes within these ecosystems, amplifying the risks of eutrophication and hypoxia [8,9].

Raising awareness of the societal challenges and potential losses associated with wetland degradation is essential for driving the development of conservation strategies and actions. Coastal wetlands are among the most productive ecosystems globally, and assessing their value provides a tangible measure of the benefits at stake. Integrating ecological valuation into economic frameworks can thus ensure the continued provision of these ecosystem services, prioritizing the welfare of future generations [10,11]. As Pearce and Turner [12] note, divergences between private and social costs and benefits can lead to suboptimal outcomes for society. Wetlands often become arenas for conflicting interests [4,13], and valuing their services helps reconcile individual economic motives with broader social benefits.

Although valuation methods for environmental goods have advanced, coastal wetlands remain under-represented in the literature despite their substantial ecological and socio-economic significance [3]. This study addresses this gap by focusing on the Amvrakikos Gulf, a semi-enclosed wetland in Greece that features low water renewal rates and significant anthropogenic pressures. Although internationally recognized under the Ramsar Convention and the Natura 2000 network, the Amvrakikos Gulf has continued to suffer from agricultural runoff, fisheries overexploitation, and climate-induced stresses.

This research employs the Contingent Valuation Method (CVM) to estimate the restoration value of the Amvrakikos Gulf by measuring local residents’ willingness to pay (WTP). While several CVM studies in Greece have investigated freshwater wetlands or other environmental contexts [14,15,16,17,18,19,20,21,22,23,24], a thorough review of the literature reveals no previous applications specifically targeting coastal wetlands. This gap is striking, given the ecological fragility and socio-economic importance of these transitional ecosystems, where marine and terrestrial influences converge. By focusing on the Amvrakikos Gulf—a semi-enclosed, climate-vulnerable coastal wetland—our study aims to fill this void and highlight how local perceptions, socio-cultural factors, and demographic attributes shape the WTP for restoration. In the sections that follow, we outline the materials and methods employed, present the survey findings, and discuss how these insights can inform policy and management strategies in the face of ongoing ecological challenges.

2. Materials and Methods

2.1. Study Area

The Amvrakikos Gulf, one of Greece’s most significant coastal wetlands, is located in the northwestern region of the country and spans an area of 411.4 km² [25] (Figure 1). Its semi-enclosed nature, characterized by limited connectivity with the Ionian Sea via a narrow channel of approximately 6 km long, has a profound impact on its hydrodynamic conditions [25]. The Gulf’s distinctive geomorphology and rich biodiversity have earned it recognition under international conservation frameworks, including the Ramsar Convention, the Montreux Protocol, and the Natura 2000 network [26].

Despite its ecological importance, the Amvrakikos Gulf is subject to significant environmental pressures arising from both natural characteristics and anthropogenic activities [27]. The Louros and Arachthos rivers, which discharge into the northern coastline, play a critical role in shaping the Gulf’s water quality [28,29,30,31]. Extensive agricultural activity in the Arta plain, heavily dependent on these rivers for irrigation, has made the Gulf a primary recipient of agricultural runoff. Since the 1970s, the transition from cereal to fruit and vegetable cultivation has led to a sharp increase in fertilizer use, resulting in elevated concentrations of heavy metals and nutrients throughout the aquatic system [26].

In addition to agricultural pressures, the Gulf endures further stress from aquaculture, livestock farming, and untreated domestic wastewater discharge from coastal settlements [25,32,33,34]. The intensification of these activities has disrupted the nutrient balance of the Gulf [25,31,35]. Moreover, dams on the Louros and Arachthos rivers, combined with dredging operations in the Preveza Strait, have significantly altered the Gulf’s natural hydrology and hydrodynamics, thus affecting water flow and renewal [26,36].

These anthropogenic and natural changes have resulted in the emergence of hypoxic and anoxic conditions and frequent eutrophication events. The development of anoxic environments in the Gulf’s deeper layers has caused the collapse of aquatic ecosystems, resulting in substantial biodiversity loss. This ecological degradation has severely impacted local fisheries, due to declining fish stocks, and has diminished the region’s appeal for tourism, further compounding the socio-economic consequences.

Climate change has further exacerbated these challenges, contributing to rising water temperatures and lower vertical mixing, which intensifies stratification and decreases dissolved oxygen levels [27]. These compounded pressures highlight the critical need for integrated management practices aimed at biodiversity conservation, climate resilience, pollution mitigation, and nutrient input regulation. Such practices are essential to restore and safeguard the Gulf’s ecosystems from further degradation [37,38].

Implementing a comprehensive management approach would not only preserve the Gulf’s natural environment but also benefit local communities by supporting key economic sectors such as fisheries and tourism. These efforts would contribute to the long-term socio-economic resilience and sustainability of the region, ensuring benefits for future generations [39].

2.2. The Contingent Valuation Method

The present study utilizes the Contingent Valuation Method (CVM), a well-established technique for estimating the value of environmental goods and services that lack a direct market price but provide substantial economic and social benefits [40,41]. In contexts where the value of a good cannot be determined through observed consumer behavior, the CVM creates a hypothetical market framework based on individuals’ stated preferences [42]. Respondents are asked to indicate their willingness to pay (WTP) or willingness to accept (WTA) compensation for enabling or preventing changes to a public good, thereby quantifying the societal value assigned to it [43,44,45,46].

A key advantage of the CVM is its ability to capture both use and non-use values, including the “option value”—the value individuals place on preserving a resource for potential future use [47,48,49]. This aspect is particularly relevant in the context of ecosystem conservation, as individuals often seek to safeguard access to future benefits derived from environmental resources [50,51]. Furthermore, valuing non-use goods, such as biodiversity conservation, is critical for informing public policy. The CVM provides quantitative evidence that aids policymakers in understanding the economic significance of preserving natural resources [43,52,53,54].

Despite its strengths, the CVM is subject to certain limitations, including potential biases in respondents’ answers, such as hypothetical bias or strategic bias, and challenges in designing robust questionnaires [55,56]. These biases are primarily attributed to the reliance on hypothetical scenarios and self-reported data, which may lead to deviations from the actual WTP [52]. However, careful survey design, along with clear and detailed information provided to respondents, can reduce these biases and improve the validity and reliability of the results [57,58,59,60,61].

An essential component of the CVM is the format of the WTP question. Common elicitation techniques include open-ended questions, bidding games, payment cards, and dichotomous choice formats [62]. The choice of the most suitable approach depends on the nature of the environmental good and the target population [63].

For this study, which focuses on the Amvrakikos Gulf, the use of payment cards and bidding games was deemed inappropriate due to challenges in establishing representative amounts and the potential for respondent fatigue. While dichotomous choice formats are often considered user-friendly [64], they can introduce “yea-saying” bias, where respondents tend to agree regardless of their actual WTP [65,66,67]. Furthermore, existing evidence does not clearly indicate the superiority of dichotomous choice over open-ended formats, as outcomes are often context-dependent [68,69,70]. To address these considerations, an open-ended format was selected, allowing respondents to freely express their WTP without constraints. This approach ensures flexibility in responses and reduces potential bias, thereby enhancing the accuracy and robustness of the valuation results.

2.3. Survey Design and Implementation

The target population for this study comprised residents of the coastal areas surrounding the Amvrakikos Gulf, selected for their in-depth familiarity with the ecosystem, its benefits, income-generating potential, and developmental prospects. Individuals from non-coastal areas were excluded to minimize the influence of emotional or pro-environmental biases on the results [43].

The survey was conducted using structured questionnaires administered through face-to-face interviews in public spaces within the study area. Although personal interviews are resource-intensive and time-consuming, they are recognized for eliciting more reliable responses. This approach is recommended by organizations such as the National Oceanic and Atmospheric Administration for environmental valuation studies [52]. Additionally, in-person interviews build trust and facilitate participation among individuals who may lack access to digital technologies [71].

Based on the 2021 census data from the Hellenic Statistical Authority, the population of the study area was 36,411 individuals. Data collection was carried out in January 2024, yielding 383 completed questionnaires. Responses with missing data or those reporting a willingness to pay (WTP) exceeding 2% of the respondent’s annual income were excluded as outliers [45].

The questionnaire adopted the WTP approach rather than the willingness to accept (WTA) compensation for the valuation of the Amvrakikos Gulf’s restoration. This choice was informed by evidence that income constraints limit the maximum amount individuals are willing to pay, resulting in lower estimates compared to the WTA, which is not subject to the same financial restrictions [44]. Additionally, behavioral factors such as “loss aversion” and the “endowment effect”, as outlined in Prospect Theory, contribute to the divergence between the WTP and WTA. People typically assign greater value to goods they already own and respond more strongly to potential losses than to equivalent gains [72,73]. Consequently, WTA estimates tend to be significantly higher than the WTP, as participants demand greater compensation for the loss of a good than they are willing to pay to acquire it.

The potential for hypothetical bias, where respondents might report exaggerated WTA or understated WTP values due to uncertainty or strategic behavior, was also considered [74]. Furthermore, individuals are generally more accustomed to the role of a consumer paying for a good than to that of an owner requiring compensation for its loss. This familiarity makes the WTP a more intuitive and practical approach in environmental valuation surveys [75].

The questionnaire explored various dimensions, including the region’s environmental challenges, the current ecological condition of the Amvrakikos Gulf, key factors driving its degradation, and the associated consequences. Respondents’ WTP for the Gulf’s restoration was assessed, and for those indicating zero WTP, the reasons for refusal were investigated. This allowed responses to be categorized as either “protest responses” or “true zero payments”. Demographic data were also collected, and the quality of each interview was evaluated.

A pilot study with 40 respondents was conducted to identify potential issues or ambiguities in the questionnaire. As no significant problems were observed, the questionnaire was finalized without modifications.

Data analysis was performed using SPSS 20.0, focusing on identifying factors influencing respondents’ WTP and estimating the mean WTP value. This approach provided a robust framework for understanding public preferences and deriving policy-relevant insights.

2.4. Data Analysis Methods

Prior to analyzing the data and estimating the average willingness to pay (WTP), the independence between nominal survey variables was evaluated using the Chi-square test to detect potential associations. A robust analytical framework was employed, combining non-parametric and parametric models to ensure comprehensive insights. The progression from non-parametric to parametric approaches introduced additional covariates and nonlinear response functions, enhancing the precision and interpretability of the results [76].

2.4.1. Kaplan–Meier

The Kaplan–Meier non-parametric estimator was utilized to analyze the survival function of respondents’ WTP amounts. This method is particularly advantageous for calculating mean and median WTP values. The median WTP represents the payment amount at which the probability of willingness to pay is 50%, often referred to as the “survival threshold” in valuation studies [77]. The reported WTP values ($C_j$) were ordered in ascending sequence, and for each amount, the number of respondents ($n_j$) willing to pay more than $C_j$ was calculated to estimate the survival probabilities stepwise [78]. The survival function was expressed as follows:

$$\widehat{S}\left(C_j\right)={\displaystyle \frac{n_j}{N}},$$

(1)

where $N$ denotes the total sample size. This stepwise calculation is critical in survival analysis, particularly when handling censored data, such as zero responses [79]. The average WTP $\left(\widehat{C}\right)$ was estimated using the following formula:

$$\widehat{C}={\sum }_{j=0}^j\widehat{S}\left(C_j\right)\cdot \left[C_{j+1}-C_j\right],$$

(2)

This method incorporated all WTP values, including zero payments. The final average WTP was adjusted by multiplying it by the proportion of respondents who expressed a willingness to pay, in line with established contingent valuation practices [45].

2.4.2. Spike Model Without Covariates

A parametric spike model was applied to estimate the WTP, accounting for zero responses. Originally developed for dichotomous choice contingent valuation studies by Kriström [80] and Hanemann [81], this model has been adapted for open-ended data to integrate zero responses while maintaining analytical accuracy [82].

Respondents were divided into two groups: those with zero WTP and those with a positive WTP, assuming a continuous WTP distribution for the latter group. Following the approach by Reiser and Shechter [83], the probability distribution of WTP ($w$) was expressed as follows:

$$P\left(\mathrm{W}\mathrm{T}\mathrm{P}<w\right)=\left\{\begin{array}{c}0 , w<0\hfill \\ p , w=0\hfill \\ p+\left(1-p\right)F\left(x\right), w>0\hfill \end{array}\right.,$$

(3)

where $p$ represents the proportion of respondents with zero WTP, and $F\left(x\right)$ is the cumulative distribution function (CDF) of the WTP for respondents with a positive WTP. Assuming a lognormal distribution for positive WTP values, the likelihood function was expressed as follows:

$${\prod }_{i=1}^n{p}^{{\delta }_i}[\left(1-p\right)f\left(w_i\right){]}^{1-{\delta }_i},$$

(4)

where $f\left(w_i\right)$ is the probability density function (PDF) derived from $F\left(x\right)$, and δ_i = 1 if w_i = 0, otherwise δ_i = 0. For positive WTP values, the lognormal CDF was defined as follows:

$$F_{\left(w\right)}=\varphi \left({\displaystyle \frac{\mathit{log}w-\mu }{\sigma }}\right),$$

(5)

where $\varphi$ denotes the standard normal distribution function. The median and mean WTP values were calculated as follows:

$$Median=\left\{\begin{array}{c}\left(1-p\right)e^{\mu } , p<{\displaystyle \frac{1}{2}}\hfill \\ 0 , p\ge {\displaystyle \frac{1}{2}}\hfill \end{array}\right. ,$$

(6)

$$Mean=\left(1-p\right)e^{\mu +\sigma \raisebox{1ex}{$2$}\!\left/ \!\raisebox{-1ex}{$2$}\right.},$$

(7)

Maximum likelihood estimation (MLE) was used to derive the parameters μ and σ [78].

2.4.3. Spike Model with Covariates

To examine factors influencing both the likelihood of zero WTP and the declared WTP amounts, a spike model incorporating covariates was employed. This approach aligns with established contingent valuation methodologies for handling zero responses while accounting for covariate effects [82,83,84]. Key covariates, such as age, gender, and income, were integrated to evaluate their impact on the parameters $p$ (zero WTP probability) and $f\left(w_i\right)$ (WTP distribution) [85,86]. The likelihood function was adjusted to incorporate covariate effects as follows:

$${\prod }_{i=1}^n{p_i}^{{\delta }_i}{\left(1-p_i\right)}^{1-{\delta }_i},$$

(8)

where $p_i$ represents the zero WTP probability for respondent i, modeled using a logit transformation:

$$\mathit{logit}{p}_i=\mathit{log}\left(p_i/1-p_i\right)=a_0+a_1{x}_1,$$

(9)

where α₀ and α₁ are regression coefficients estimated using logistic regression. Positive WTP values were modeled using a lognormal distribution with the mean of the log-transformed WTP ($ln WTP$) expressed as follows:

$${\mu }_i={\beta }_0+{\beta }_1{x}_1,$$

(10)

where β₀ and β₁ are coefficients estimating the influence of covariates $\left(x_i\right)$ on the WTP. Outliers were identified and excluded by comparing WTP values to respondents’ annual household income, resulting in the removal of two cases from the analysis.

By combining parametric and non-parametric approaches and integrating covariate effects, this analytical framework provides robust insights into factors influencing the WTP while ensuring precise and comprehensive estimation of WTP distributions.

3. Results

• Interview Quality and Respondent Demographics

The quality of the interviews was evaluated as satisfactory for over 99.5% of the sessions, with 94% rated as “good”, indicating effective implementation of the survey and reliability of the data collection process. The demographic characteristics of respondents are summarized in

Table 1. Respondent demographics.

Variable	Types	Relative Frequency (%)
Gender	Female	50.9
	Male	49.1
Age Group	18–25	2.3
	26–35	23.4
	36–45	26.3
	46–65	31.6
	>65	16.4
Employment	Employee	79.8
	Unemployed	2.9
	Retired	10.5
	Housekeepers	1.0
	Student	4.2
	Other	1.6
Education	No formal schooling	0.5
	Primary school	7.1
	High school	36.0
	Bachelor’s degree	41.5
	Master’s/Doctorate degree	15.0
Household Income	<EUR 10,000	22.6
	EUR 10,000–19,999	42.5
	EUR 20,000–29,999	19.7
	EUR 30,000–39,999	8.4
	>EUR 40,000	3.4
	I prefer not to answer	3.4

.

• Awareness of Amvrakikos Gulf Environmental Issues

Participants were first briefed on the purpose and scope of the survey, emphasizing anonymity and confidentiality. Introductory questions regarding the Amvrakikos Gulf (AG) aimed to capture respondents’ perceptions of its condition and importance (

Table 2. Perceptions of Amvrakikos Gulf.

Variable	Types	Proportion (%)
Main benefits of AG	Special landscape	22.8
	Recreation/swimming	11.0
	Fish/seafood for consumption	24.6
	Attraction for tourists/visitors	13.3
	Maintains a special ecosystem	27.8
	Other	0.6
Swimming in AG	Attractive destination	23.9
	No longer attractive	60.6
	None of the above	12.1
	Other	3.4
Fish/seafood consumption from AG	Fearlessly consume	75.9
	Hesitate to consume	18.9
	None of the above	2.9
	Other	2.4
Environmental condition	Very Good	4.7
	Good	18.6
	Moderate	45.4
	Poor	17.8
	Very poor	10.0
	Don’t know/prefer not to answer	3.41
Factors affecting AG’s environmental condition	Agricultural practices	23.5
	Livestock/poultry waste	21.1
	Urban wastewater	19.7
	Dam construction (Louros and Arachthos)	9.2
	Aquaculture	16.0
	Natural causes	7.2
	Other	1.5
	Don’t know/prefer not to answer	1.9
Consequences of environmental degradation	Aesthetic decline	21.8
	Fewer tourists	7.1
	Lower quality of fish/seafood	18.7
	Losses in aquaculture/fisheries	22.3
	Environmental harm/loss of species	28.8
	Other	1.1
	Don’t know/prefer not to answer	0.2
Importance of environmental protection	Extremely important	79.3
	Very important	15.7
	Important	3.4
	Minor importance	0.5
	Not important at all	0.3
	Don’t know/prefer not to answer	0.8
Membership in environmental organizations	Yes	39.9
	No	47.0
	Don’t know/prefer not to answer	13.1
Are you a member of any environmental organization?	Yes	3.1
	No	96.3
	Don’t know/prefer not to answer	0.5

).

• Willingness to Pay for the Restoration of Amvrakikos Gulf

Participants were asked to respond to a hypothetical market scenario that highlighted the need for the environmental restoration of the Amvrakikos Gulf. The scenario emphasized the significant financial and human resources required. Respondents were asked if they would be willing to contribute a one-time payment per household to support an independent and reliable management body. This body would oversee and implement the restoration actions with transparency and efficiency. The primary goal of this scenario was to capture participants’ willingness to financially support the environmental restoration of the Gulf.

The findings revealed that 46.9% of respondents expressed a willingness to pay (WTP). Among them, the following was observed:

• A total of 50% cited ensuring the Gulf’s restoration for future generations as their primary motivation;
• A total of 33.5% expressed a desire to contribute to ecosystem protection;
• A total of 8.5% highlighted the wish to continue enjoying the Gulf’s benefits as in the past;
• A total of 8.0% recognized the Gulf’s economic importance through tourism and fisheries.

Conversely, 48.56% of participants declined to pay, while a smaller proportion (5.25%) chose not to answer. The main reasons for refusal were the following:

• A total of 70% believed that funding such actions is the responsibility of the state;
• A total of 19.5% cited a lack of trust in the proper use of funds;
• A total of 9.2% reported limited financial capacity due to low household income;
• A total of 1.1% mentioned other reasons.

Analysis using the Chi-square test identified two variables significantly associated with the WTP:

• Age: respondents aged 26–45 exhibited a higher willingness to pay compared to individuals over 45 years of age;
• Environmental awareness: participants who were aware of mass fish mortality incidents in aquaculture facilities demonstrated a greater likelihood of willingness to pay. In contrast, those unaware of these incidents predominantly refused to contribute financially.

• Non-parametric assessment: Kaplan–Meier

The Kaplan–Meier survival function was used to analyze the distribution of the WTP across the sample. Outliers were excluded from the analysis. The survival curve, shown in Figure 2, depicts the WTP distribution. After a series of estimation steps, the mean WTP was calculated at EUR 58.03 (

Table 3. Kaplan–Meier estimation (non-parametric model) results.

Kaplan–Meier Estimation Results: Total Sample
Sample size	Mean WTP	Std. Error	Lower Bound (95% Confidence Interval)	Upper Bound (95% Confidence Interval)
381	58.031	7.679	42.981	73.082

).

• Parametric assessment: Spike model-No covariate information

For the parametric assessment using the Spike model without the inclusion of covariates, the mean and standard deviation of the lognormal distribution for the willingness to pay (WTP) were estimated. After applying natural logarithmic transformation, the data approximated a normal distribution (Figure 3). The calculated mean of the log-transformed WTP was 4.20, the median was 3.91, and the standard deviation was 1.055. Considering these values and the proportion of the sample that accepted the payment scenario, the mean WTP for the entire sample was estimated at EUR 53.75.

This parametric approach provides a detailed analysis of the WTP, leveraging the properties of the lognormal distribution to capture the underlying variation in respondents’ willingness to financially support environmental restoration. It complements the non-parametric findings and highlights the robustness of the estimation process.

• Parametric assessment: Spike model covariate information

Using a spike model incorporating covariate variables, factors influencing respondents’ willingness to pay (WTP), and the payment amount were examined (

Table 4. Logistic regression results.

Logistic Regression Results: Total Sample
Variables	αi	Standard Error	Wald	p-value	Exp (B)/ Odds ratio	Mean (Variable)
Constant: Acceptance of the payment	3.095	0.613	25.466	0.000	22.086	-
Location	−0.070	0.026	7.397	0.007	0.933	8.88
How important do you consider the environmental protection of the Amvrakikos Gulf?	−0.501	0.195	6.634	0.010	0.606	1.29
Are you aware of the incidents of mass fish mortality that have occurred at fish farms?	−0.523	0.169	9.584	0.002	0.593	1.73
Age	−0.445	0.136	10.673	0.001	0.641	2.48

** Model Summary: • -2 Log likelihood = 490.847 ▪ Cox and Snell R² = 0.088 • Nagelkerke R² = 0.118.

). The results indicated that key factors significantly affecting WTP acceptance included the survey location, the importance attributed to the environmental protection of the Amvrakikos Gulf, awareness of mass fish mortality incidents in aquaculture, and age. The WTP amount was influenced by respondents’ professional status (

Table 5. Linear regression results.

Linear Regression Results
Variables	βi	Standard Error	Beta	t-value	p-value	Mean (Variable)
Constant: lnWTP	3.910	0.128	-	30.628	0.000	-
Professional status	0.194	0.067	0.215	2.903	0.004	1.250

** Model Summary: R = 0.215 • R² = 0.046 • Adjusted R² = 0.041 • Standard Error of the Estimate = 1.033.

). By applying the model coefficients and solving the corresponding equations, dependent variables were calculated. According to the logistic regression model, the probability of accepting payment was estimated at 45.5%. Considering this probability and the results of the linear regression model, the mean WTP was estimated at EUR 49.31.

• Aggregated WTP Estimation

This study targeted a total population of 36,411 residents. According to the most recent data from the Hellenic Statistical Authority, the average household size in Greece is approximately 2.6 individuals per household. Based on these figures and the mean WTP estimates from the different models, the total value of restoration and preservation efforts for the Amvrakikos Gulf was calculated as follows (

Table 6. Willingness to pay for the preservation of Amvrakikos Gulf.

Total Sample
Approach Method	Average WTP:	Aggregated Value Based on Permanent Population
Non-parametric assessment: Kaplan–Meier	EUR 58.03	EUR 812,652.12
Parametric assessment: Spike model—No covariate information	EUR 53.75	EUR 752,715.00
Parametric assessment: Spike model—Covariate information	EUR 49.31	EUR 690,537.24

).

4. Discussion

The participants identified ecosystem conservation, the provision of fish and seafood, and the aesthetic value of the landscape as the Amvrakikos Gulf’s main benefits, thereby highlighting the wetland’s dual ecological and economic significance. Notably, 60.6% of respondents reported that the Gulf is no longer attractive for swimming, whereas approximately 75.9% continue to consume fish and seafood—suggesting either trust in their quality or a strong cultural and/or economic reliance on local fisheries. Furthermore, the overwhelming majority (98.4%) of participants deem the Gulf’s preservation “important” to “extremely important”, reflecting a widespread understanding of its pivotal role in local development and biodiversity protection.

Approximately half of the respondents (46.9%) expressed a willingness to pay (WTP) for the restoration of the coastal wetland, focusing primarily on safeguarding it for future generations. This observation aligns with previous studies indicating that non-market valuation methods effectively capture citizens’ preferences for environmental goods, encompassing both tangible (e.g., fisheries and tourism) and intangible (e.g., cultural identity and aesthetics) attributes [45,47,87]. Age—specifically the 26–45 group—and heightened awareness of environmental issues, such as mass fish mortality, emerged as key factors enhancing the willingness to contribute financially. Such findings underscore how information dissemination and firsthand experiences of ecological disruptions reinforce environmental sensitivity, a trend also corroborated by global valuation studies [22,88,89].

In contrast, a significant proportion of respondents—exceeding one-third—expressed unwillingness to pay, either because they perceived the responsibility to lie solely with the state or due to concerns regarding the transparent allocation of funds. These so-called “protest zeros” are commonly observed in CVM applications and often do not indicate a genuine unwillingness to pay but rather reflect ideological beliefs about the fairness of public goods financing or distrust toward managing authorities and skepticism about the effectiveness of the proposed intervention. Such tendencies are particularly prevalent in societies where trust in public administration is limited or where the institutional framework lacks transparency and accountability [90,91]. To avoid inflating the mean WTP, such protest responses were included as valid zero payments, indicating that the ethical dimension of resource protection constitutes a point of tension between viewing natural assets as public goods and the notion of individual monetary contributions. From a policy standpoint, this prevalence of “protest zeros” reinforces the need for more transparent, participatory governance processes, which promote co-management and strengthen the role of local communities in decision-making [1,92]. International evidence suggests that resident involvement in planning or monitoring committees can alleviate distrust and enhance the effectiveness of conservation measures.

Moreover, the CVM itself faces inherent constraints regarding its “snapshot” nature and the innate subjectivity of participants: the stated WTP values may not fully align with actual behavior, particularly under rapidly shifting economic, cultural, or climatic conditions [88]. Recent interdisciplinary approaches emphasize that an ecosystem’s value arises from the dynamic interplay between its objective functions and the informational or symbolic representations held by its users [22]. In the case of the Amvrakikos Gulf—facing pressures from climate change (sea-level rise and eutrophication) and intensive agriculture—it is paramount to consider its cultural dimension, rather than focusing exclusively on its “market” or ecological value. Combining quantitative tools (the CVM and travel cost) with qualitative methods (focus groups and participatory mapping) can yield more holistic insights into the mechanisms underpinning residents’ cultural attachment and their willingness to support restoration actions.

Finally, regarding sample demographics, a relatively high proportion of participants had higher education degrees and were actively employed, potentially limiting the generalizability of these findings. Nevertheless, their lack of prior survey experience may have introduced some degree of uncertainty or hesitation, yet it also facilitated more spontaneous and non-strategic responses, possibly resulting in more reliable and representative data. Future research could focus on the following: (a) broader and more diverse samples (e.g., tourists, professional fishers, and farmers), (b) a longitudinal examination of the WTP, and (c) the correlation of additional factors (e.g., membership in environmental organizations and attitudes toward climate change) that may reveal new opportunities and challenges for the sustainable management of the Amvrakikos Gulf.

5. Conclusions

This study highlights the multifaceted interplay between ecological, socio-cultural, and economic factors that influence public support for the restoration of semi-enclosed coastal wetlands affected by anthropogenic and climate-induced pressures.

The willingness to pay (WTP) for the restoration of the Amvrakikos Gulf, while significant among local residents motivated by intergenerational equity and environmental awareness, is strongly contingent on transparency in governance and institutional trust. The findings emphasize that successful restoration efforts should not rely exclusively on economic or ecological criteria but must also integrate local socio-economic factors, residents’ values, and motivations through active community engagement.

The Gulf’s role in supporting local economic stability and mitigating rural depopulation further emphasizes the socio-economic significance of coastal wetlands. Sustainable economic activities driven by restoration efforts can foster regional resilience, reduce urban–rural disparities, and enhance community cohesion. Policymakers must integrate local perceptions into decision-making processes to align conservation goals with the socio-economic and cultural fabric of the community.

While the specific valuation results may not be directly applicable to other regions, the underlying principles—trust, cultural relevance, and demonstrable benefits—are transferable. Methodologically, the dynamic nature of the WTP under evolving socio-economic or climatic conditions highlights the value of longitudinal CVM studies, complemented by qualitative approaches such as focus groups and participatory mapping.

In conclusion, this study provides compelling evidence for the importance of local engagement, transparent governance, and robust economic justifications in restoration efforts. The Amvrakikos Gulf exemplifies how interdisciplinary approaches can secure public commitment, ensuring the long-term viability of ecosystems and their contributions to socio-economic development. Offering a transferable framework, this research underscores the global relevance of fostering inclusive and sustainable restoration practices for coastal wetlands.