Exploring perceptions on participatory management of NATURA 2000 forest sites in Greece

Forest Policy and Economics 56 (2015) 1–8

Contents lists available at ScienceDirect

Forest Policy and Economics journal homepage: www.elsevier.com/locate/forpol

Exploring perceptions on participatory management of NATURA 2000 forest sites in Greece Nikoleta Jones a,⁎, Elias Filos b, Eleftherios Fates b, Panayiotis G. Dimitrakopoulos b a b

Department of Geography, Walton Hall, Open University, MK7 6AA Milton Keynes, United Kingdom Biodiversity Conservation Laboratory, Department of Environment, University of the Aegean, 81100 Mytilene, Lesbos, Greece

a r t i c l e

i n f o

Article history: Received 16 May 2014 Received in revised form 19 March 2015 Accepted 23 March 2015 Available online 17 April 2015 Keywords: Conservation policy National Parks Trust Networks Collaborative management

a b s t r a c t Several studies in the biodiversity management literature investigate the potential shift from state-based forest management to participatory management frameworks. A main challenge in this context is to identify social factors influencing the level of public acceptability towards co-management frameworks especially in countries where the state has had traditionally a very strong role in the management of natural resources. The present paper aims to investigate the social factors influencing the level of acceptability for participatory management frameworks in two forest protected areas of Greece differing in the date of their establishment: the Tzoumerka–Peristeri–Arachthos Gorge National Park and the Vikos–Aoos National Park. Specifically, through the distribution of a structured questionnaire to local communities, we explored the influence of social factors (trust in institutions, social trust and social networks) on citizens' perceptions for a shift in participatory management frameworks. Furthermore, we explore the restriction that citizens perceive from the implementation of such frameworks. According to the study, there is clear preference towards a collaborative management framework which is based on the cooperation of local communities with state actors. Social trust, social networks and institutional trust have a significant impact on citizens' perceptions, especially for collaborative management policies which promote the cooperation of local communities with public actors. Finally, the level of restriction that citizens perceive from each proposed policy is also linked with the level of acceptability. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Improving biodiversity management is a main priority in the international field of environmental policy and planning (Rands et al., 2010). Several international policy frameworks have been developed for this purpose, such as the EU Habitat Directive, which are often planned in central level and need to be implemented in a national and local setting (Borrass, 2014). In this context, there is significant discussion in the literature regarding the effectiveness of participatory management frameworks for Protected Areas (PAs) (Berkes, 2004; CBD, 2004; Young et al., 2012) especially in countries with a strong culture of state-based environmental management (Dimitrakopoulos et al., 2010; Randulovic, 2008; Petrova et al., 2009; Petrova, 2014a,b; Cellarius, 2004). This discussion is triggered by a global shift to move from state-based policies to community-based management frameworks (Robson and Kant, 2007; Blomley et al., 2008; UNESCO, 1996). In such frameworks, local communities are encouraged to participate in the management of a protected area involved both in decisionmaking processes and management activities (Klooster and Masera, ⁎ Corresponding author. Tel.: +44 1908652972. E-mail addresses: [email protected] (N. Jones), [email protected] (P.G. Dimitrakopoulos).

http://dx.doi.org/10.1016/j.forpol.2015.03.010 1389-9341/© 2015 Elsevier B.V. All rights reserved.

2000) and to collaborate with civil society organizations, private actors and the state (Humphreys, 2006). The shift to participatory management frameworks is accompanied with significant benefits both for ecosystems and the society (Klooster and Masera, 2000). From an environmental perspective, there is evidence that involvement of a community in the management of a PA can result to a more effective management of biodiversity (e.g. Blomley et al., 2008). This is mainly because participatory management frameworks promote sustainable management practices (Berkes et al., 2003). From a socio-economic perspective, a major benefit is the increase of social acceptability levels for specific protection frameworks (Stoll-Kleemann and Welp, 2008). Furthermore, the application of ‘softer’ and participatory management tools may assist in local economic development through, for example, eco-tourism activities and minimization of social conflicts (Misra and Kant, 2004; Chowdhury and Koike, 2010; Nuggehalli and Prokopy, 2009; Nayak and Berkes, 2008; Sandstrom and Widmark, 2007; Matose, 2006; Jones et al., 2012a). In addition, participatory management allows the use of local values and knowledge for the management of a specific area of high biodiversity value in combination with scientific information (Berkes, 2004). Despite the numerous benefits of participatory management for areas of high biodiversity value, its' success depends significantly on

2

N. Jones et al. / Forest Policy and Economics 56 (2015) 1–8

finding a good balance between state involvement and the level of participation of civil society (Young et al., 2012). Furthermore, the shift towards participatory management frameworks may culminate in a significant challenge as several social factors may obstruct its' application (Jones et al., 2012b). For example, when the level of trust in management institutions is low and the density of social networks in a community is weak, management of natural resources becomes more difficult (Pretty and Smith, 2004). In such social circumstances, no strong common norms exist which could assist members to act in a collective manner in order to protect biodiversity (Jones et al., 2012b). However, it should be noted that facilitation techniques, in order to overcome such difficulties, have been identified, such as role-playing games and agent-based simulations (Briot et al., 2007). Consequently, it is necessary to explore social factors influencing the effectiveness of participatory management frameworks (Berkes, 2004) and it is even more important to identify and explore these social barriers during the policy planning stage (Thomas and Middleton, 2003; Oikonomou et al., 2011). In this context, the present paper aims to investigate social factors influencing citizens' perceptions on three alternative management scenarios. The research was conducted in order to explore differences on citizen perceptions from state-based to more co-management frameworks and explanatory factors for these perceptions. Regarding the explanatory factors, we focus on three social aspects which have been recently identified as very important for the effective management of natural resources (Pretty, 2003; Jones et al., 2012b; Adger, 2003). These are: trust in institutions, social trust and social networks. Trust in institutions influences communities' perceptions for biodiversity management (Gong et al., 2010). Especially in the context of collaborative frameworks, where local communities have to collaborate with state and non-state actors, the level of trust that local communities have towards these entities will significantly affect the level of cooperation between them. Furthermore, management institutions are usually responsible for the effectiveness of control mechanisms for illegal exploitation of natural resources (Nayak and Berkes, 2008). Thus, weak institutional trust often results in irresponsible environmental behavior obstructing the management of a PA (Jones et al., 2012b). Second, social trust refers to the level of trust which is developed among members of a community (Putnam, 2000). This type of trust is strongly connected to the response of communities in collaborative frameworks (Pretty, 2003; Jones et al, 2012b). When high levels of social trust exist, citizens believe that their fellow citizens will act in a collective manner supporting the common good and promoting environmental responsible behavior (van Laerhoven, 2010; Pretty, 2003). Under such circumstances, citizens tend to be more positive towards collaborative management frameworks. Finally, social networks, refer mainly to the formal and informal organizations which exist in a community (Putnam, 2000). Social networks are important for the flow of information concerning environmental management (Nuggehalli and Prokopy, 2009; Jones et al., 2012a). As a result they influence the level of participation in management frameworks (Djamhuri, 2008), the level of environmental awareness in a community and also environmentally responsible behaviors (Jones et al., 2012b; Cramb, 2005; Wakefield et al., 2006). In order to explore the role of these factors on citizens' responses to participatory management frameworks, we conducted an empirical study in two forest PAs in Greece. The selection of the specific country was mainly based on the fact that Greece has a strong tradition in managing natural resources in a centralized way (Jones et al., 2012c; Dimitrakopoulos et al., 2010; Papageorgiou and Vogiatzakis, 2006). An attempt to change this trend has been initiated in the past decade as several management authorities for protected areas were created by the national government promoting the participation of different actors, mainly state actors, local authorities, members of NGOs, members of local communities and environmental scientists (Vokou et al., 2014). Despite these efforts the absence of institutional consolidation of

protected areas, the lack of regular funding in order to assist in the management of protected areas along with the lack of support from the state has resulted in the questioning of the effectiveness of the specific policy initiative (Vokou et al., 2014). 2. Methods 2.1. Selection and description of research areas The Tzoumerka–Peristeri–Arachthos Gorge National Park (Tzoumerka hereafter) and the Vikos–Aoos National Park were selected as study areas. Both areas are located in the Pindos mountain range, the largest in size mountain range of Greece. Tzoumerka is located in the central part of Pindos, while Vikos–Aoos in the northern part. A main difference between the two areas is their date of establishment: the Vikos–Aoos National Park was founded in 1973 (Presidential Decree 213/73, Official Gazette 198/A/73) and since 2005 is part of the Northern Pindos National Park (Joint Ministerial Decision 23069, Official Gazette 639/D/14.6.2005). Tzoumerka was designated as a National Park in 2009 (Official Gazette 49/D/12.02.2009) and had no official protection status before that. Both parks support rich biodiversity (species and habitats) and include many sites of the EU-wide NATURA 2000 network, i.e. Special Areas for Conservation (designated under Habitats Directive 92/43/EEC) and Special Protection Areas (designated under Council Directive 2009/147/EC) (law 3937/2011). In both areas the main activities are agriculture, farming, forestry and recreation. However, in Vikos–Aoos eco-tourism activities are more developed (Trakolis, 2001) compared to Tzoumerka. 2.2. Sampling A questionnaire was created and distributed in the Tzoumerka site in spring–summer 2011 and in the Vikos–Aoos area approximately a year later. Regarding the local population, in the Tzoumerka area the final sampling frame was estimated to 10,000 inhabitants living in 10 local mountain communities (Hellenic Statistical Authority, 2001). In Vikos–Aoos there are three communities which are in the protection zone and another 6 which are close to the park (Papageorgiou and Kassioumis, 2005) with a total population of approximately 850. Due to the absence of official lists of the total population in the local communities we applied a stratified random sampling based on the population of the local communities (Bryman, 2012). The main aim was to select a specific number of questionnaires from local permanent residents from each community of the research areas while taking into consideration the local demographic characteristics available (gender, education and age) (Hellenic Statistical Authority, 2001). In total 367 questionnaires were collected (response rate 70%), 200 from the Tzoumerka site and 167 from the Vikos–Aoos area. The characteristics of the sample are presented in Table 1. When comparing these characteristics with those of the actual population these are in

Table 1 Demographic characteristics of the sample.

Educational level (%) Up to 6 years 9 years–high school 12 years–secondary Post-secondary Higher education Post-graduate studies Gender (%) Male Female Age (mean) Household members (mean)

Tzoumerka

Vikos–Aoos

Total

20.6 11.6 32.2 12.6 25.1 1.5

3.6 16.2 31.1 10.2 33.5 5.4

12.1 13.7 31.7 11.5 29.0 3.3

66.5 33.5 42.7 4.09

59.3 40.7 44.7 3.76

63.2 36.8 43.7 3.92

N. Jones et al. / Forest Policy and Economics 56 (2015) 1–8

accordance with those of the sampling frame apart from a higher percentage in higher education. However, this can be explained from the fact that the last available data for both areas are from 2001 and higher education graduates have increased significantly in Greece in the past decade (OECD, 2013).

2.3. Description of the questionnaire The questionnaire was divided in five parts. In the first section of the questionnaire, perceptions on environmental issues were explored both for general environmental issues and for more specific matters for the PAs. In particular, the following questions were presented to respondents: Are you interested on environmental issues in general?, How important do you regard the protection of this area which is considered as a national park?, How important is the area of the national park for your community?. All questions were measured on a 10 point Likert scale with higher values revealing higher levels of importance. Regarding the level of knowledge, respondents were first asked if they were aware of the existence of the protected area and also of the specific protection status (whether the area is a National Park and/or is regarded a Natura 2000 site) (answers: Yes/No/don't know). Additional questions were included concerning the knowledge of the existence of protection zones, and also awareness for specific restrictions referring to building and infrastructures (e.g. construction of houses, road networks, industrial establishments), traditional land uses (logging, hunting, grazing) and visitors' activities. We decided to include these specific restrictions in the questionnaire as they were identified as the most important by the Special Environmental Study for each area based on which the national protection regulations were subsequently formed (Northern Pindos National Park: Joint Ministerial Decision 23069, Official Gazette 639/D/14.6.2005; Tzoumerka: Official Gazette 49/D/12.02.2009). The source of information for these restrictions was also explored. Furthermore, participants were asked how frequently they visit the area and the main reason for their visit. A different section of the questionnaire focused on social factors (institutional trust, social trust and social networks). Questions were created in order to examine whether these social factors may facilitate or obstruct the transition to participatory management frameworks. The questions mainly focused on issues of trust and networking in the community. All questions concerning social issues along with their descriptive statistics are presented in Table 2 and were created according to relevant studies (van Oorschot et al., 2006; Jones et al., 2012a,b).

3

A final section of the questionnaire focused on alternative management policies for the PAs. Three hypothetical management scenarios were presented to respondents increasing in each one the role of the local community. Specifically, differences between the three scenarios focused on: a) environmental management, b) financial management (referring also to the use of natural resources) and c) control mechanisms. In the first policy (state-based: SB), all responsibilities are attributed to the Management Actor and the Forest Authority. We decided to include a state-based policy in order to use it as a comparable measure to the other two participatory management scenarios. This policy is very similar to the one which has been applied in Greece for several decades and local communities are familiar with this type of management framework. In this policy, local communities have practically no rights in the management of the forest and their actions are limited by the decisions of the state. In the second policy, collaboration of state actors with local communities is promoted (collaborative management—COL). State actors retain their responsibilities in environmental management and control mechanisms but the financial management of the natural resources is a responsibility of the local community. Thus, local communities are responsible for the trading of local goods and the income from such activities is used for the benefits of the community directly. In this case, local communities can control some aspects of the management of the forest in collaboration with the state, thus increasing their level of participation and involvement in forest management. Furthermore, they are able to be involved in trading and recreational activities and receive income from these activities although that will necessitate time allocation for certain members of the community. In the third policy the role of the local community is stronger in managing the forest area (community-based—COM). Local people are responsible for the financial management of the forest resources. A wide spectrum of activities has to be organized by local communities in order to increase their income, including ecotourism activities, selling of local agricultural products and selling of timber. Through the profits from these activities, local communities will cover the expenses for personnel who will be responsible for control actions and environmental management of the area. Furthermore, citizens will be allowed to some extent to use the forest products for their own use. In this case, there is a high level of freedom for local communities while increased rights to manage the forest are given to them. At the same time, this type of freedom will be accompanied with certain management responsibilities for local communities.

Table 2 Social factors explored in the study along with their descriptive statistics.

Are you satisfied with the actions of the forest authority for the protection of the park (scale 1–5)a Trust forest authority (scale 1–10)b Trust local authority (municipality) (scale 1–10)

b

Trust national government (scale 1–10)b Trust Ministry of Environment (scale 1–10)b Level of trust towards fellow citizens in general (scale 1–10)

b

Level of fairness from fellow citizens (scale 1–10)b Level of trust in locals (scale 1–10)b Compliance of fellow citizens with the current regulations for the park (scale 1–5)a a b

1–5 Likert scale, 5: Highest level of agreement or compliance. 1–10 Likert scale, 10: Highest level of trust.

Research area

Mean

Std. deviation

Differences

Tzoumerka Vikos–Aoos Tzoumerka Vikos–Aoos Tzoumerka Vikos–Aoos Tzoumerka Vikos–Aoos Tzoumerka Vikos–Aoos Tzoumerka Vikos–Aoos Tzoumerka Vikos–Aoos Tzoumerka Vikos–Aoos Tzoumerka Vikos–Aoos

1.90 2.34 4.51 4.31 4.43 4.13 1.74 2.43 2.44 2.81 3.70 3.90 3.38 3.97 4.76 4.32 2.31 2.51

1.05 1.08 3.00 2.50 2.97 2.71 2.51 2.19 2.67 2.28 2.99 2.28 2.83 2.28 2.57 2.34 1.23 1.098

t = −3.955, p b 0.05 t = 0.680, p N 0.05 t = 1.016, p N 0.05 t = −2.784, p b 0.05 t = −1.407, p N 0.05 t = −0.684, p N 0.05 t = −2.167, p b 0.05 t = 1.710, p b 0.1 t = −1.628, p b 0.05

4

N. Jones et al. / Forest Policy and Economics 56 (2015) 1–8

Respondents were asked to declare their agreement for the proposed policies on a 5 point Likert scale with higher values revealing a higher level of agreement. They were also asked to state how restrictive they regard the specific policies (measured on a 5 point scale, with higher values revealing higher restriction). In the context of the third scenario, because it was significantly based on the level of participation of local communities in activities, participants were also asked to state if they would participate in three types of activities: voluntary fire protection activities, cleaning of the forest and patrols in order to check for illegal activities. The specific activities were selected because they could be easily understood by respondents, as they are familiar with them, increasing the reliability of the answers. 3. Results 3.1. Level of importance and awareness From the data analysis, a high level of concern for environmental issues in general was presented. Specifically, in the total sample, the level of interest for the environment and the level of importance of the park for the local community were all evaluated with high mean scores in the Likert scale (Table 3). Higher values were presented for the level of importance of environmental protection of the park. Regarding differences between the two areas, higher levels of interest for the environment in general and the importance of environmental protection of the park were presented in the Tzoumerka site. 75.2% of the participants were aware of the existence of the protected area, 73.5% were aware that it is considered a National Park and 44.3% were aware that it includes Natura 2000 sites. Furthermore, 43.1% were aware that there are established protection zones in the area (Table 4). Regarding knowledge on specific restrictions, the highest level of awareness was presented for restrictions in hunting followed by restrictions in logging. The level of awareness for the protection zones and grazing restrictions was significantly higher in the case of Vikos– Aoos compared to Tzoumerka. If respondents were aware of at least one restriction connected with the existence of the protected area, they were then asked to state what the source of this information was. In the total sample, informal networks of friends and family are the main source of information (68.7%) followed by local authority (33.8%) and mass media (33.2%). Other sources mentioned by respondents were the forest authority (26%), the local community (21.8%) and the hunting association (25.4%). 3.2. Environmental behavior Regarding the frequency of visiting the area, 30.7% responded that they visit the area on a daily base (Table 5). Very few participants mentioned that they never visit the park (Table 5). The main reason for visiting the PA in both sites was for walking (Tzoumerka: 84.1%, Vikos–Aoos: 98.8%). Apart from walking, other activities were mentioned but with significant lower percentages. The most common

activities for the Tzoumerka site were the collection of firewood (26.5%), hunting (16.9%), collection of aromatic plants (16.4%) and collection of mushrooms (15.9%). In the Vikos–Aoos site, frequently mentioned activities were collection of aromatic plants (40.4%), collection of mushrooms (25.9%), collection of firewood (21.7%) and also access to agricultural land (17.5%). 3.3. Local participation Another topic investigated through the study focused on networking and participation connected with community activities. An initial question explored whether citizens are informed about the decisions of the local council. According to the study, over half of the sample was informed of these decisions (Τzoumerka: 69.5%, Vikos–Aoos: 68.9%, X2 = 0.02, p N 0.05). Regarding their actual participation in the local councils only 13.2% from Vikos–Aoos and 9% from the Tzoumerka area stated that they participate (X2 = 1.63, p N 0.05). Lower percentages were presented concerning the level of participation in decisionmaking processes for the management of the park. Only 2.5% of the sample stated a positive answer in Tzoumerka and 1.2% in the Vikos– Aoos site in the specific question (X2 = 0.837, p N 0.05). Finally, in the case of Tzoumerka 16% of respondents mentioned that they participate in voluntary fire watches while the respective percentage on Vikos– Aoos (2.4%) is statistically significantly lower (X2 = 19.04, p b 0.01). 3.4. Perceptions on proposed management policies Concerning citizens' perceptions on the three management frameworks, the highest level of acceptance among the three was presented for the COL scenario (Table 6) which promoted the collaboration of the local communities with the current management actors. The lowest level of acceptance was presented for the community management option, where the management of all aspects in the area would be the responsibility of the local community (Table 6). Regarding the restrictions that accompany these management frameworks according to citizens' perceptions, the level of restriction is very low in all scenarios. The highest restriction was presented for the COM scenario (Tzoumerka: 1.59, Vikos–Aoos: 1.37, t = 2.24, p b 0.05), followed by the SB (Tzoumerka: 1.45, Vikos–Aoos: 1.34, t = 1.34, p N 0.05) while the least restrictive is the COL (Tzoumerka: 1.19, Vikos–Aoos: 1.43, t = −3.66, p b 0.01). The level of acceptability is negatively correlated with the restriction level for each scenario (Spearman coefficient results: SB Rho: −0.27, p b 0.01, COL Rho: −0.17, p b 0.05, COM Rho: − 0.19, p b 0.05). This means that as the level of restriction increases for each scenario the level of acceptability becomes lower. In the context of the COM scenario, respondents were also asked to declare if they would like to participate in certain voluntary activities which would facilitate the functioning of the specific management framework. In general, respondents were positive in participating in such activities. 73.3% mentioned that they would be willing to participate in voluntary fire watches (Tzoumerka: 79.5%, Vikos–Aoos: 65.9%, X2 = 6.85, p b 0.05), 76.8% that they would participate in cleaning

Table 3 Level of interest for the environment (scale 1–10).

Level of interest for the environment Importance of the park for the community Important of environmental protection of the park

Tzoumerka

Vikos–Aoos

Total

Mean (std. dev.)

Mean (std. dev.)

Mean (std. dev.)

8.44 (1.93) 8.38 (2.10) 9.16 (1.77)

7.68 (1.77) 8.35 (1.63) 8.87 (1.51)

8.09 (1.88) 8.37 (1.90) 9.03 (1.66)

Differences

t = 3.90 p b 0.01 t = 0.10 p N 0.05 t = 1.67 p b 0.1

N. Jones et al. / Forest Policy and Economics 56 (2015) 1–8

5

Table 4 Level of awareness for the level of protection (positive answer).

Knowledge for the establishment of the park Aware that the area is a National Park Aware that the areas include NATURA 2000 sites Knowledge for protection zones Knowledge of restrictions for logging Knowledge of restrictions for hunting Knowledge of restrictions for grazing

Tzoumerka

Vikos–Aoos

Total

p-Value

71.5 74.3 49.1 31.5 85.9 97 54

79.6 74.2 39.5 56.9 91.6 96.4 72.5

75.2 73.5 44.3 43.1 88.5 96.7 62.5

X2 = 3.23, p b 0.01 X2 = 304.87, p b 0.01 X2 = 148.75, p b 0.01 X2 = 25.5, p b 0.01 X2 = 2.89, p b 0.1 X2 = 0.09, p N 0.01 X2 = 13.10, p b 0.01

activities in the forest (Tzoumerka: 85.5%, Vikos–Aoos: 66.5%, X2 = 18.52, p b 0.01) and 61.7% that they would participate in voluntary patrols for illegal activities in the park (Tzoumerka: 67.8%, Vikos– Aoos: 54.5%, X2 = 6.85, p b 0.05). Consequently, a relatively high percentage of respondents are willing to volunteer in the context of the participatory framework. 3.5. Factors explaining the level of agreement In this section the analysis on the influence of social factors on citizens' perceptions will be divided in two main parts based on the variables used and the statistical analysis conducted. In the first part links between the level of agreement with environmental perceptions and trust are explored though the use of the Spearman correlation coefficient which is appropriate when trying to investigate the relationship between two ordinal variables. In the second part, mean differences using the Mann–Whitney test are conducted in order to explore the link between the level of agreement and the level of participation in community activities which was measured in a dichotomous scale. Mann– Whitney is appropriate for comparing means in non-parametric data. 3.5.1. Environmental perceptions and trust According to the bivariate correlation analysis, the collaborative management policy is connected with several social factors explored in the present study. Specifically, the level of agreement towards the collaborative scenario is positively correlated with all variables measuring the level of interest and importance of the park along with trust in local authorities, trust in the national government and also trust in members of the local community (Table 7). Some correlations were also presented for the SB and COM policies. SB is positively connected with the level of satisfaction and trust for the Forest Services. Regarding the COM policy, the level of agreement is positively correlated with trust in local authorities and also with perceptions of whether other citizens comply with regulations (Table 7). 3.5.2. Participation Connections between the level of participation and the level of acceptability were identified with the Mann–Whitney test, by exploring differences in the level of agreement between those who replied a positive and a negative answer. Some statistical significant differences

Table 5 Frequency of visiting the park. Site

Not at all 3 times a year 6 times a year Monthly Weekly Everyday

Total

Tzoumerka

Vikos–Aoos

3.1% 13.3% 8.7% 12.8% 27.2% 34.9%

0.6% 9.0% 8.4% 22.8% 33.5% 25.7%

Chi-square test: X2 = 13.12, p b 0.05.

1.9% 11.3% 8.6% 17.4% 30.1% 30.7%

were identified from the analysis. Citizens who declared that they participated in local councils also presented higher levels of agreement for the COL and COM scenario (COL: 4.25, COM: 3.25) compared to those who did not participate (COL: 3.56, COM: 2.34) (p b 0.05). Those who stated that they are informed of the local council decisions were also more positive in the COL scenario (mean: 3.79) compared to those who were not informed (mean: 3.30) (p b 0.05). In addition, those who participated in fire watches were less positive in accepting the SB scenario (mean: 2.25) compared to those who did not participate (mean: 3.13) (p b 0.05). Finally, individuals who participated in other voluntary activities were less willing to accept the SB scenario (mean: 2.86) compared to those who did not participate in such activities (mean: 3.22). However, citizens who participated in voluntary activities were more willing to accept the COL scenario (mean: 3.80) compared to those who did not participate (mean: 3.47) (p b 0.05). 3.5.3. Regression analysis Apart from exploring bivariate relationships we also explored the multivariate relations with the use of ordinal regression models. A major obstacle in applying regression models in the context of the present study was the strong correlations between some variables. For example, all the variables measuring trust in institutions are statistically significant correlated between them and as a result can cause multicollinearity. In order to overcome this problem, we decided to explore the impact of certain variables on the level of acceptability of each scenario which are not correlated between them and do not explore similar topics. In this context, we introduced in the model as independent variables: the level of restriction for each scenario, the level of importance of the PAs, age, educational level, level of social trust towards fellow citizens, trust in the forest authority and participation in the decisions of the local council. Due to the large number of categories we introduced all variables as covariates in the model, apart from the dichotomous ones (gender and participation in local council decisions) (Table 8). According to the regression models, in the state-based scenario, trust in the forest authority and the level of restriction are statistically the most important parameters explaining social acceptability (Nagelkerke R2 = 0.13). Specifically, the level of restriction is linked negatively to acceptance, meaning that individuals who tend to consider the statebased scenario as restrictive were also less willing to accept it. Furthermore, citizens who tended to trust the forest authority were more willing to accept the state-based management framework.

Table 6 Level of agreement in management scenarios. Scenario

Site

Meana

Std. deviation

Differences

State based (SB)

Tzoumerka Vikos–Aoos Tzoumerka Vikos–Aoos Tzoumerka Vikos–Aoos

3.00 3.07 3.57 3.72 2.23 2.69

1.55 1.26 1.58 1.28 1.64 1.38

t = 0.522, p N 0.05

Collaborative (COL) Community based (COM) a

1–5 Likert scale, 5: Highest agreement.

t = 0.347, p N 0.05 t = −2.865, p b 0.05

6

N. Jones et al. / Forest Policy and Economics 56 (2015) 1–8

R2 = 0.12). Similar to the previous policies, the level of restriction was linked negatively with the level of acceptance. Also, respondents with a higher educational level were more willing to accept the community based proposed policy. All other factors included in the model were not statistically significant explanatory variables. Looking at the results of the model for the sample population some interesting findings were observed. Those who considered the park as very important and also trusted the forest authority were less willing to accept the proposed policy. On the other hand, individuals who engaged in decisions of the local council and trusted their fellow citizens presented a higher level of acceptance.

Table 7 Correlations linking perceptions and trust with the level of agreement.

Environmental perceptions Interest of the environment Importance of the park for the local community Importance of protecting the park Institutional trust Level of satisfaction from forest services Trust in forest services Trust in local authorities Trust national government Trust ministry of environment Social trust Trust fellow citizens Trust fellow citizens in the local community Felling of fairness Perception whether fellow citizens comply with regulations

SB

COL

COM

−0.024 −0.004

0.127⁎ 0.128⁎

−0.046 0.005

−0.008

0.177⁎⁎

0.170⁎⁎ 0.158⁎⁎ 0.079 −0.028 0.048

0.054 −0.022 0.044 −0.052 0.212⁎⁎ 0.169⁎⁎ 0.110⁎ 0.067 0.027 −0.022

−0.008 0.008 0.055 0.079

0.081 0.175⁎⁎ 0.082 0.036

0.058

4. Discussion

0.057 0.046 0.074 0.115⁎

Spearman correlation, *p b 0.05, **p b 0.01.

According to the statistical analysis, three more variables were linked with the level of acceptance but with a lower level of statistical significance (p b 0.1). In particular, participants who were not involved in local council decisions were more in favor of this specific policy. Furthermore, older individuals and also those that regarded the park as important for the local community were also more willing to accept the state-based policy. Regarding other variables which were not statistically significant, gender and social trust were linked positively with the level of acceptance of the specific policy whereas educational level had a negative sign. In the collaborative scenario, statistically, the most important parameters were social trust and the level of restriction (Nagelkerke R2 = 0.11). In particular, similar to the state-based scenario, the level of restriction was linked negatively to acceptance, revealing that respondents who considered the specific policy as restrictive were also negative regarding its implementation. Furthermore, individuals who tended to trust their fellow citizens were also more positive for the proposed policy. Age and education were both linked positively with the level of acceptance, but their statistical significance in the regression model was low (p b 0.1). Specifically, people with a higher level of education and also of older age were more positive towards the collaborative policy. Other factors, such as participation in decisions of the local council, the importance of the park for the community and the level of trust in the forest authority were not statistically significant, but they were all positively linked with the level of acceptance in the model. Finally, in the community based scenario, the most important parameters were education and the level of restriction (Nagelkerke

This study aimed to explore citizens' perceptions on three alternative management scenarios and the factors affecting these perceptions. The results presented contribute to the rapidly increasing discussion concerning the need to find a balance between state involvement and public participation in protected area management especially in countries where the state has a dominant role in the management of natural resources. The results from the two case studies revealed that the specific Protected Areas are very important for the local communities and there are a variety of reasons that people visit the area, referring both to recreation activities and use of natural resources. This result is in accordance with previous findings in Greece where PAs are considered an important asset for local communities (Jones et al., 2012b). Furthermore, it was interesting that although a large percentage of the respondents were aware that the specific area is considered as a protected site, there were several participants who were not informed on the specific protection legislations currently in force, with significantly low levels of awareness observed for the Natura 2000 designation and protection zones. In general, the low level of awareness for the existence of a protected area is a widely recognized problem in biodiversity conservation studies (Booth et al., 2009; Dimitrakopoulos et al., 2010). Thus, information dissemination on relevant restrictions and the designation status should be significantly increased in the area by conducting seminars organized both by state and civil society actors and also by increasing the flow of information through local media. Regarding the source of information on issues of the protected sites, it was observed that the role of informal networks was very important in both areas mainly through family and friends. This result highlights the importance of informal networks in Greece which is considered as one of the ‘strongest’ social characteristics of the Greek culture (Jones et al., 2008). It should be underlined that such networks do not necessarily have a positive impact (Putnam, 2000). On the contrary, they may obstruct the application of environmental management initiatives through the dissemination of false information or they may promote

Table 8 Ordinal regression: factors explaining the level of acceptance. State based

Level of acceptance (ref category: 5—high)

Part. in decisions of the loc. council (ref category: yes) Gender (ref category: female) Importance of the park for the local community Age Education Social trust Trust forest authority Level of restriction

1 2 3 4 No Male

Collaborative

Comm. Based

Est

Wald

Sig

Est

Wald

Sig

Est

Wald

Sig

−.819 −.143 .936 2.149 1.252 .045 .097 −.012 −.113 .002 .110 −.666

.677 .021 .885 4.622 2.780 .049 3.285 3.345 2.312 .002 9.294 25.170

.411 .885 .347 .032 .095 .824 .070 .067 .128 .967 .002 b0.001

.136 .906 1.724 2.467 .152 .185 .083 .013 .143 .129 .025 −.536

.019 .831 2.993 6.076 .044 .821 2.332 3.241 3.341 9.628 .449 11.742

.891 .362 .084 .014 .833 .365 .127 .072 .068 .002 .503 .001

−.328 .308 .972 1.679 −.337 .049 −.063 .010 .280 .060 −.037 −.670

.117 .103 1.023 3.030 .235 .055 1.310 1.921 12.499 2.115 1.008 26.218

.732 .748 .312 .082 .628 .815 .252 .166 b0.001 .146 .315 b0.001

N. Jones et al. / Forest Policy and Economics 56 (2015) 1–8

habits which are harmful for the environment retaining of nonenvironmental habits. Other sources, such as state actors and the media, were also mentioned with significantly lower frequencies revealing that further efforts should be conducted in the future by the state and NGOs in order to increase the flow of information towards local communities. The lack of flow of information in the community is also evident from the low level of participation of locals in local council meetings. Apart from these observations a main aim of the study was to explore citizens' level of acceptability on three proposed management frameworks where the state and local communities have different levels of involvement. According to the results of the study the highest level of acceptance was presented for the collaborative scenario which promoted the cooperation of local and state actors with the local community. The least accepted management framework was the community based scenario where the majority of responsibilities would fall on local communities and they would be minimum interference from the state. According to our study the most important factor determining this ranking of the scenarios is the restriction that individuals perceived from each management option. This is supported both by the bivariate and multivariate analysis conducted. In particular, the communitybased scenario was the one which restricted the least individuals although this was not the one which provided the most ‘freedom’ in terms of land use and managing forest resources. On the contrary, participants regarded that the most restrictive scenario was the one which provided the highest levels of freedom to the local community on forest use. One explanation of this finding is that the restriction declared by individuals and the level of acceptability reflects a ‘fear of the unknown’ and also a refusal to undertake additional responsibilities. In the case of the community-based scenario, this framework implies a significant change for local communities and participants may have doubts on how this framework will work and if it is going to be efficient. A community-based scenario could give additional freedoms to local communities to manage the forest land along with employment opportunities, which are key factors for efficient governance of forests (Petrova, 2014a). However, during the implementation of the research it is possible that participants did not perceive the level of these potential benefits. On the contrary, a state-based management policy, which provides local communities with limited use rights to the forest area, is a framework that citizens are more familiar with and proposes fewer changes in the everyday lives of locals. Apart from the level of restriction, the influence of other factors was explored in the study focusing mainly on social trust, institutional trust and social networks. In the collaborative policy, the regression model revealed that social trust was an important explanatory parameter for citizens' level of acceptance. Furthermore, social trust and trust in local authorities and the national government were all statistically correlated in the bivariate analysis with the level of acceptability for this management framework. Trust is a fundamental issue in social theories as it is considered a main parameter that influences collective actions of individuals. It is an important factor for the effective application of these types of policy initiatives as it influences the level and type of collaboration developed between citizens and representatives of management actors (Jones et al., 2012a,b,c). In the case of the community-based framework, which was also the least accepted by respondents, bivariate analysis revealed that citizens who tended to believe that other members of the community do not comply with regulations also tended to disagree with the proposed policy. This is linked to the high reliance of such frameworks on the level of compliance to social norms of a community. In case of recurrent incidents of free-riding and non-compliance with regulations, such frameworks are very difficult to be successfully implemented. It should be noted though, that this finding is not supported by the results of the regression model where trust was not a statistically significant parameter for the specific scenario. Finally, in the state-based scenario, regression analysis revealed that trust in the forest authority was an important parameter explaining the level

7

of acceptability. Thus, respondents who tend to trust certain state institutions are more positive towards policies which incorporate a high level of involvement from the state. Another explanatory factor explored in the study was social networks. When observing the results of the bivariate analysis some links between networks and citizens' perceptions for the participatory management framework were revealed. In particular, individuals who were more active and informed on community issues were also more positive towards the proposed collaborative and community management policies. On the contrary, these respondents were less willing to accept a state-based scenario. This result reveals that participants who are more engaged in community issues are also more willing to accept a policy which would necessitate their collaboration and possibly their active involvement. In the regression models for the collaborative and community based policies, networks were not a statistically significant explanatory variable. Only in the state-based scenario a link between networks and acceptability was observed, although this is not supported by a strong statistical significance (p b 0.1). Specifically, individuals who did not participate in decision-making processes were more willing to accept the state-based policy. This finding may be explained by the fact that citizens who do not engage with collective issues of their community are also more willing to let the state manage the forest area with a minimum level of involvement from locals. The opposite result was observed in the community-based policy where respondents who participated in decision-making processes were more willing to accept it. Although this latter result is not statistically significant, it supports the above assumption; that citizens who are more active in their community are also more open to policies which need a high level of public engagement. From the above discussion it is interesting to note that some variables which were significant in the bivariate analysis, were not significant in the regression models. For example, trust in the forest authority, participation in decisions of the local council and the level of importance of the park for the local community were not significant variables in the collaborative and community based management regression models. Due to the different propositions included in each scenario, differences in the results of the regression models were expected, with some variables having a stronger influence than others. However, this result also underlines the complexity of exploring the simultaneous impact of social variables on the level of acceptability for forest management policies. Thus, we propose that future research should focus on identifying these types of connections and exploring how links between social factors influence the level of acceptability for participatory management frameworks. 5. Conclusions In conclusion, in the present study we aimed to measure and explain local community acceptability for the three management frameworks (state-based, collaborative management and community management) in two of the most important forest protected areas of Greece. A main aim in this context was to contribute in the discussion concerning a potential shift to co-management frameworks in Greek PAs. The most important explanatory factor for the level of acceptability in all three proposed scenarios was the level of restriction that citizens perceived from each management framework. In this context, the study revealed that a community-based scenario, where citizens would have significant freedom in managing forest resources, was considered as the most restrictive by participants and was also the least accepted. This is possibly because the potential new use rights provided to local communities under such frameworks also imply a significant commitment from them and new management responsibilities. The study also revealed a high level of support for a collaborative management framework in both case study areas proposing the collaboration of local and state authorities with local communities. The level of social trust determined significantly the level of acceptability for the specific

8

N. Jones et al. / Forest Policy and Economics 56 (2015) 1–8

management framework. These findings provide an initial indication of how local communities perceive non-state based management frameworks in Greece. Further research is necessary in order to understand the practical implications of a shift in co-management frameworks in countries with similar social characteristics as Greece and identify ways in order to proceed to such initiatives. In addition, further research is necessary in order to identify whether a shift of this kind will actually be beneficial for Greek PAs and will result in biodiversity conservation.

Acknowledgment The authors would like to thank Dr. David Humphreys for his constructive comments in an earlier version of the paper.

References Adger, W.N., 2003. Social capital, collective action and adaptation to climate change. Econ. Geogr. 79, 387–404. Berkes, F., 2004. Rethinking community-based conservation. Conserv. Biol. 18, 621–630. Berkes, F., Anderson, J., Duffield, C., Gardner, J.S., Sinclair, A.J., Stevens, G., 2003. Participatory management and sustainability: evolving policy and practice in a mountain environment. In: Eugene, L., Perl, A. (Eds.), The Integrity Gap: Canada's Environmental Policy and Institutions. UBC Press, Vancouver-Toronto, pp. 134–166. Blomley, T., Pfliegner, K., Isango, J., Zahabu, E., Ahrends, A., Burgess, N., 2008. Seeing the wood for the trees: an assessment of the impact of participatory forest management on forest condition in Tanzania. Oryx 42, 380–391. Booth, J.E., Gaston, K.J., Armsworth, P.R., 2009. Public understanding of protected area designation. Biol. Conserv. 142, 3196–3200. Borrass, L., 2014. Varying practices of implementing the Habitats Directive in German and British forests. For. Policy Econ. 38, 151–160. Briot, J.P., Guyot, P., Irving, M., 2007. Participatory simulation for collective management of protected areas for biodiversity conservation and social inclusion. International Modeling and Simulation Multiconference 2007 (IMSM'07)pp. 183–188 (Buenos Aires, Argentine, février). Bryman, A., 2012. Social Research Methods. 4th edition. Oxford University Press. CBD (Secretariat of the Convention on Biological Diversity), 2004. Guidelines on Biodiversity and Tourism Development: International Guidelines for Activities Related to Sustainable Tourism Development in Vulnerable Terrestrial, Marine and Coastal Ecosystems and Habitats of Major Importance for Biological Diversity and Protected Areas, Including Fragile Riparian and Mountain Ecosystems. Secretariat of the Convention on Biological Diversity, Montreal (29 pp.). Cellarius, B.A., 2004. In the Land of Orpheus: Rural Livelihoods and Nature Conservation in Postsocialist Bulgaria. University of Wisconsin Press, Madison WI. Chowdhury, M.S.H., Koike, M., 2010. An overview on the protected area system for forest conservation in Bangladesh. J. For. Res. 21, 111–118. Cramb, R.A., 2005. Social capital and soil conservation: evidence from the Philippines. Aust. J. Agric. Resour. Econ. 49, 211–226. Dimitrakopoulos, P.G., Jones, Ν., Iosifides, T., Florokapi, I., Lasda, O., Paliouras, F., Evangelinos, K., 2010. Local attitudes on protected areas: evidence from three Natura 2000 wetland sites in Greece. J. Environ. Manag. 91, 1847–1854. Djamhuri, T.L., 2008. Community participation in a social forestry program in Central Java, Indonesia: the effect of incentive structure and social capital. Agrofor. Syst. 74, 83–96. Gong, Y., Bull, G., Baylis, K., 2010. Participation in the world's first clean development mechanism forest project: the role of property rights, social capital and contractual rules. Ecol. Econ. 69, 1292–1302. Hellenic Statistical Authority, 2001. Database for the 2001 national census. Available at, www.statistics.gr Accessed on March 2011. Humphreys, D., 2006. Logjam: Deforestation and the Crisis of Global Governance. Earthscan, London. Jones, N., Malesios, C., Iosifides, T., Sophoulis, C.M., 2008. Social capital in Greece: measurement and comparative perspectives. S. Eur. Soc. Polit. 13, 175–193. Jones, N., Gleridou, C., Dimitrakopoulos, P.G., Evangelinos, I.K., 2012a. Investigating social acceptability for public forest management policies as a function of social factors. For. Policy Econ. 14, 148–155. Jones, N., Clark, J.R.A., Panteli, M., Proikaki, M., Dimitrakopoulos, P.G., 2012b. Local social capital and the acceptance of protected area policies: an empirical study of two Ramsar river delta ecosystems in northern Greece. J. Environ. Manag. 96, 55–63.

Jones, N., Iosifides, T., Evangelinos, I.K., Florokapi, I., Dimitrakopoulos, P.G., 2012c. Investigating knowledge and perceptions of citizens of the National Park of Eastern Macedonia and Thrace, Greece. Int. J. Sustain. Dev. World 19, 25–33. Klooster, D., Masera, O., 2000. Community forest management in Mexico: carbon mitigation and biodiversity conservation through rural development. Glob. Environ. Chang. 10, 259–272. Matose, F., 2006. Co-management options for reserved forests in Zimbabwe and beyond: policy implications of forest management strategies. For. Policy Econ. 8, 363–374. Misra, D., Kant, S., 2004. Production analysis of collaborative forest management using an example of joint forest management from Gujarat, India. For. Policy Econ. 6, 301–320. Nayak, P.K., Berkes, F., 2008. Politics of co-optation: community forest management versus joint forest management in Orissa, India. Environ. Manag. 41, 707–718. Nuggehalli, R.L., Prokopy, L.S., 2009. Motivating factors and facilitating conditions explaining women's participation in co-management of Sri Lankan forests. For. Policy Econ. 11, 288–293. OECD, 2013. Education at a glance: Greece. Country Note. OECD. Oikonomou, V., Dimitrakopoulos, P.G., Τroumbis, A.Y., 2011. Incorporating ecosystem function concept in environmental planning and decision making by means of multi-criteria evaluation: the case-study of Kalloni, Lesbos, Greece. Environ. Manag. 47, 77–92. Papageorgiou, K., Kassioumis, K., 2005. The national park policy context in Greece: Park users' perspectives of issues in park administration. J. Nat. Conserv. 13, 231–246. Papageorgiou, K., Vogiatzakis, I.N., 2006. Nature protection in Greece: an appraisal of the factors shaping integrative conservation and policy effectiveness. Environ. Sci. Policy 9, 476–486. Petrova, S., 2014a. Communities in Transition: Protected Nature and Local People in Eastern and Central Europe. Ashgate, Aldershot. Petrova, S., 2014b. Contesting forest neoliberalization: recombinant geographies of ‘illegal’ logging in the Balkans. Geoforum 55, 13–21. Petrova, S., Bouzarovski, S., Cihar, M., 2009. From inflexible national legislation to flexible local governance: management practices in the Pelister National Park, Republic of Macedonia. GeoJournal 74, 589–598. Pretty, J., 2003. Social capital and the collective management of resources. Science 302, 1912–1914. Pretty, J., Smith, D., 2004. Social capital in biodiversity conservation and management. Conserv. Biol. 18, 631–638. Putnam, R., 2000. Bowling Alone: The Collapse and Revival of American Community. Simon & Schuster Paperbacks, New York. Rands, M.R.W., Adams, W.M., Bennun, L., Butchart, S.H.M., et al., 2010. Biodiversity conservation: challenges beyond 2010. Science 329, 1298–1303. Randulovic, R., 2008. Participation in nature protection: does it benefit the local community? A Triglav National Park case study. Int. J. Biodivers. Sci. Ecosyst. Serv. Manag. 4, 209–218. Robson, M., Kant, S., 2007. Structure of causation and its influence on cooperation: a comparative study of forest management in Ontario, Canada. For. Policy Econ. 10, 70–81. Sandstrom, C., Widmark, C., 2007. Stakeholders' perceptions of consultations as tools for co-management — a case study of the forestry and reindeer herding sectors in northern Sweden. For. Policy Econ. 10, 25–35. Stoll-Kleemann, S., Welp, M., 2008. Participatory and integrated management of biosphere reserves—lessons from case studies and a global survey. GAIA 17 (S1), 161–168. Thomas, L., Middleton, J., 2003. Guidelines for Management Planning of Protected Areas. IUCN Gland, Switzerland and Cambridge, UK. Trakolis, D., 2001. Perceptions, preferences, and reactions of local inhabitants in Vikos– Aoos National Park, Greece. Environ. Manag. 28, 665–676. UNESCO, 1996. Biosphere Reserves: The Seville Strategy and the Statutory Framework of the World Network. UNESCO, Paris. van Laerhoven, F., 2010. Governing community forests and the challenge of solving twolevel collective action dilemmas—a large-N perspective. Glob. Environ. Chang. 20, 539–546. van Oorschot, W., Arts, W., Gelissen, J., 2006. Social capital in Europe. Measurement and social and regional distribution of a multifaceted phenomenon. Acta Sociol. 49, 149–176. Vokou, D., Dimitrakopoulos, P.G., Jones, N., Damialis, A., Monokrousos, N., Pantis, J.D., Mazaris, A.D., and the Natura 2000 Committee (2010–2013) members, 2014. Ten years of co-management in Greek protected areas: an evaluation. Biodivers. Conserv. 23, 2833–2855. Wakefield, S., Elliott, S., Eyles, J., Cole, D., 2006. Taking environmental action: the role of local composition, context, and collective. Environ. Manag. 37, 40–53. Young, J.C., Butler, J.R.A., Jordan, A., Watt, A.D., 2012. Less government intervention in biodiversity management: risks and opportunities. Biodivers. Conserv. 21, 1095–1100.