Research

INTRODUCTION

Coastal and marine ecosystem degradation can lead to the reduction of fisheries resources, affecting the livelihoods of artisanal fishing communities (Costanza et al. 2014). Because of the social and natural complexities, most conservation strategies adopt co-management to foster good governance of these environments through the participation of multiple actors (Carlsson and Berkes 2005, Gutiérrez et al. 2011, Alexander et al. 2015, Berkes 2021).

Collaboration between different actors seems a necessary condition for good governance of common-pool resources, although this is not enough (Bodin 2017). When and how collaboration effectively solves a collective action problem is still under discussion (Bodin 2017). With whom actors choose to collaborate will depend on the interaction between the attributes of the potential partner and the capacity to fit in a specific collective action problem (Berardo and Scholz 2010, Bodin 2017, Bodin et al. 2020a, Nohrstedt and Bodin 2020).

Paying attention to the network formed by actors engaged in collaboration, that is, who the actors are and how they interact, is one way to elucidate why some kinds of co-management arrangements seem to be more successful than others (Carlsson and Sandström 2008, Bodin and Crona 2009, Sandström and Rova 2010, Bodin 2017). In turn, network structures can evidence the presence of social capital and leadership proposed as necessary preconditions for collective action (Bodin and Crona 2008, Gutierrez et al. 2011, Crona et al. 2017). Social capital refers to the rules and social networks that facilitate collective action (Crona et al. 2017), and leadership to the presence of actors that facilitates collective action through coordination (Gutierrez et al. 2011).

Social capital develops networks characterized by high frequency of interaction, reciprocity, and social proximity. These network structures are related to increased levels of collective action because they promote the development of trust, share views, and mutual understanding, reducing the risk and cost of collaboration (Ostrom 1990, Bodin and Crona 2009). In addition, identifying the position of actors in the network is important to understand their relevance. Leadership forms network structures with weak ties and centralized actors (Bodin and Crona 2009). Actors in a central position can foster collective action if they facilitate collaboration, initiate action, coordinate resources, and bring together other disconnected actors (Bodin and Crona 2009, Bodin et al. 2020a).

The presence of social capital is important for collective action because it can buffer the lack of leadership (Crona et al. 2017). Also, leadership plays an important role because active and engaged leaders can activate social capital for collective action (Bodin and Crona 2008, Gutiérrez et al. 2011). Therefore, a network with structural elements of social capital and leadership is likely more capable of building resilience and increasing adaptability to environmental changes (Bodin et al. 2006). Then, knowing the resultant collaborative network is important to highlight the implications for collective action.

In Brazil, the Marine Extractive Reserves (hereafter MERs) are a conservation model based on resources use that aims to protect the livelihoods and culture of coastal traditional populations, ensuring the sustainable use of natural resources (Cardozo et al. 2012, Santos and Schiavetti 2013, 2014). The participation of traditional populations in the creation and management is essential to grant MERs legitimacy (Nobre and Schiavetti 2013, Dias et al. 2018). Accordingly, MERs adopt a community-based co-management model in which different actors such as government institutions, civil society, and traditional extractive communities share power and decision making (Nobre and Schiavetti 2013, Santos and Schiavetti 2013).

The Canavieiras Extractive Reserve (CER) is a Brazilian MER located on Bahia’s southern coast fostered by traditional fishing communities looking for socio-environmental protection of their territories (Prost 2018). In 2006, CER was founded by federal decree after a bottom-up process involving traditional community organizations with environmental NGOs and university support (Dumith 2018). This mobilization arised as a response to the implementation of shrimp farms, industrial fisheries, and the expansion of luxury hotels in the region that threatened artisanal fishery communities’ livelihood (Dumith 2018). With CER creation, land expropriation of those not belonging to the traditional community took place, granting the resource tenure rights to traditional communities, also called “beneficiaries” because of their access to resources use (Santos and Schiavetti 2013, Dumith 2018). This configuration resembles the Territorial Use Rights in Fisheries observed elsewhere (Aceves-Bueno et al. 2020).

According to the MERs’ co-management model, the Deliberative Council (hereafter DC) is the executive decision-making body, where the discussion and decisions are collectively taken by different actors that belong to it (Nobre and Schiavetti 2013). Traditional extractive communities are the majority in the DC as a way of safeguarding their rights (Nobre and Schiavetti 2013, Santos and Schiavetti 2013). The DC is headed by the Instituto Chico Mendes de Conservação da Biodiversidade (hereafter ICMBio, in Portuguese). This Brazilian federal environmental agency moderates the meetings and takes the last decision in case of tied votes (Prost 2018).

The incorporation of multiple actors in the co-management brought several challenges. Notably, previous studies have stressed the presence of conflictual interests among different sectors (Dumith 2018, Prost 2018). Despite that, CER is characterized by a good decision-making process, highlighting the leading role of the management agency and the cohesiveness and organization capacity of the traditional communities with high levels of participation and engagement (Cardozo et al. 2012, 2019). Simultaneously, it highlights the poor engagement of local governmental institutions and civil society (Cardozo et al. 2012, 2019, Santos and Schiavetti 2014, Dias et al. 2018).

In this context, and considering the CER clear regulatory framework characteristic of the MER model, we ask if DC members have the power to move from discussion to collective action and how they will collaborate between them to solve different collective problems that may arise. Thus, this work analyzes the collaborative networks’ structures to achieve collective action among the institutions involved in the co-management of the CER and the implications that these network structures may have for resilience. We aim to evaluate the structure of the collaborative networks between the DC institutions looking for social capital and leadership necessary to address different problems that may emerge related to (1) biodiversity conservation, (2) governance, and (3) socioeconomic issues. Also, we discuss which implications network structures may have in CER resilience.

METHODS

Study site

Situated in Brazil’s coastal-marine area, specifically in south Bahia, CER comprises the municipalities of Belmonte, Canavieiras, and Una (Fig. 1). CER covers a total area of 100,726.36 ha, of which 83% corresponds to the marine area and 17% to mangroves, rivers, and land areas (Cardozo et al. 2012). This area benefits approximately 2300 families of nine different communities geographically separated: Oiticica, Puxim do Sul, Puxim de Fora, Barra Velha, Canavieiras, Atalaia, Campinhos, Pedras de Una, and Belmonte (Cardozo et al. 2012, Dumith 2018). These traditional families depend on natural resources for their maintenance and income, their main activities being traditional fishing, shellfish collection, and family farming (Cardozo et al. 2012, Dumith 2018).

The CER creation process fosters the organization of traditional populations in associations according to their communities and different extractive activities (Dumith 2018). Because traditional leaders realized that there was a need to create a central entity that would organize and respect the specificities of existing associations, all beneficiary associations joined in the Mother Association of Extractivists (AMEX, in Portuguese) created in 2009 (Dumith 2018). This institution has signed the Territorial Use Rights in Fisheries of CER resources, which granted the right over the fishing resources of the area for 20 years and can be renewed for another equal period. Despite AMEX not being a DC official member (without the right to vote), it has a crucial role in organizing and articulating the discussions between beneficiaries’ associations to define priorities and demands to present in the DC meetings (Dumith 2018, Cardozo et al. 2019).

The DC was established in 2009, aiming to contribute to implementing CER objectives (Brasil 2009). According to its bylaws, the DC comprises members representing institutions and is renewed every two years. Each institution chooses its representatives for the DC. In the case of the beneficiary traditional communities, the representatives are elected by local assemblies (MMA 2010). To be part of the DC, an institution needs the approval of the DC and must have compatible objectives or develop activities related to the CER (MMA 2010).

All the discussions about CER management and sustainable resources use occur in the DC meetings. The DC decisions are made by simple majority vote of members present, without the president’s vote. In cases of a tie, the decision rests with the president. It is also up to the president of the DC to enforce the decisions of the Plenary (MMA 2010).

In 2018, the DC approved the management agreement, developed in a participatory process with the traditional communities, containing the rules for natural resources use. This management agreement considers the activities traditionally practiced and respects the existing area management (Brasil 2018).

Network survey

A survey was conducted to identify the collaborative networks among CER institutions (Appendix 1) to face biodiversity, governance, and socioeconomic issues that may arise. The members of the DC that responded to the interview provided a response based on what they believed to be the institution that has the capacity and/or responsibility for the subject in question. This relationship is not necessarily based only on trust in the institution’s role, but rather on the recognition of its function within the DC with the purpose of promoting better management of the CER. As we set network boundaries following the nominalist approach (see Laumann et al. 1989), we pre-defined networks’ nodes taking into account the actual list (2020–2022 period) of DC official members (Appendix 1) as our universe sample to be interviewed.

The interview (Appendix 2) had 12 questions, each addressing one potential problem based on the main topics discussed in the DC meetings, and was developed based on the analysis of the DC meeting minutes respective to the 2010–2019 period. Using a closed-ended list form of DC institution members, for each question, we asked participants to nominate, in order of priority, the institutions they would count on to deal with different problems or situations that may appear related to biodiversity conservation, governance, and socioeconomic issues. A pilot survey revealed the need to adjust the network boundaries. As a result, we included AMEX in our study, in the closed-ended list form, because participants frequently mentioned this institution.

A representative of each institution belonging to the DC of CER was invited to participate in the research. In cases where the representative refused to participate, another member of the same institution (usually the president) was interviewed. In this case, it was considered that the respondent should know about the DC’s work in order to be a suitable actor to answer the questions (Robins et al. 2011).

The survey was carried out from November to December 2020. The interviews were previously scheduled and held at the respondent’s institution (when possible) or the ICMBio office. The interview was conducted via online meetings for respondents who did not live in the CER region. Before each interview, a written consent was presented to the participants explaining the study’s objectives, steps, duration (~45 min), data use, and the right to refuse participation. This was made according to the Universidade Estadual de Santa Cruz (UESC) Research Ethics Committee (CEP) requirements, under approved protocol number CEP: 27566819.2.0000.5526.

Data analysis

Of 33 institutions, 30 involved in the CER co-management agreed to answer the interview. That gives a response rate of 90.91%, which allows us to get reliable estimates of network-level statistics (Berardo et al. 2020). Of the 30 representatives who responded to the questionnaire, 14 were women and 16 men. For the study of network structure, we used the social network analysis (SNA; Degenne and Forse 1999), which has a broad adherence among researchers (Ramirez-Sanchez and Pinkerton 2009, Cohen et al. 2012, Marín et al. 2012, Alexander and Armitage 2015, Alexander et al. 2016, 2018, Bodin 2017, Crona et al. 2017, Kluger et al. 2020).

We built a total of four potential networks: the general network (formed by all survey questions) and also the biodiversity, governance, and socioeconomic sub-networks created based on the topics of the potential problems presented in the survey (see Appendix 2). Network nodes represented an institution involved in the CER co-management. Institutions were categorized as “institution types” according to the sector they represent: the category “beneficiaries” includes traditional communities; “civil society” includes universities and NGOs; “territory users” includes organizations of non-traditional activities performed in the area, such as tourism and rural producer sector; and “government” category includes different government level agencies. Network edges represent the potential collaboration relationships between one institution to deal with another in biodiversity conservation, governance, and socioeconomic issues. Edges weights were set based on the priority rank given by the interviewee relative to the total number of institutions. Then, all the networks were defined as a multiple (network with more than one edge between two nodes), weighted (network with weight on the edges that denote the priority of relation; Margerum 2008), and directed (the direction of network edges was indicated, in each case, with an arrow coming out from the institution that points to the relation; Robins et al. 2011).

The average edge weight was calculated as a first step in the analysis. Only edges equal to or greater than the average weight were considered (Appendix 3) because the reported information’s accuracy is improved with the use of stronger ties (Freeman 1977, Wasserman and Faust 1994, Sandström and Rova 2010). The visual representation of the networks was adjusted using the force-directed Fruchterman and Reingold layout algorithm (Fruchterman and Reingold 1991), denoting that nodes closer to each other share more connections among them.

Three network level metrics were calculated to analyze the network structure: density, reciprocity, and transitivity indexes. Network density reflects how well connected a network is and was calculated as the number of observed connections divided by the number of possible connections (Prell 2011). Reciprocity measures estimates, in a directed network, the proportion of mutual connections at the dyadic level (between two institutions; Robins et al. 2011). Structural network closure was denoted with a transitivity index that measures the probability that two actors tied to a common third will also be tied (Robins et al. 2011).

We calculated two centrality indexes to measure the positional relevance of each institution (Wasserman and Faust 1994). The in-degree index measures the number of edges directed to a node (Wasserman and Faust 1994) and will denote the most cited institution for collective action. Betweenness centrality measures the extent to which a node sits between pairs of other nodes in the network, indicating how important a node is in connecting other nodes (Freeman 1977, Freeman et al. 1979).

Also, we used the optimal community structure algorithm (Brandes et al. 2008) to detect cohesive subgroups, which are subsets of nodes well connected among themselves but sparsely connected to nodes from other subgroups, giving an idea of network fragmentation (community structure, see Newman and Girvan 2004). Modularity evaluates the goodness of partitions of a network into subgroups (Newman and Girvan 2004). Therefore, large positive values of modularity (considering that the maximal value adopted is 1) indicate good partitions. This subgroup detection algorithm uses link weights but treats the directed edges as undirected. All network analyses were performed using the package iGraph (Csardi and Nepusz 2006) implemented in R version 3.6.0 (R core team 2019).

RESULTS

Network structure

All networks comprised 33 nodes. The network’s statistics estimates are shown in Table 1. The general network had 1972 edges distributed in three different topics. The network density was 0.156, which indicates that about 15% of the potential edges were present in the network. The network had 306 pairs of reciprocal edges, giving a reciprocity index of 0.313. The network transitivity was high (0.767), indicating that about 77% of the adjacent nodes of a node were connected.

The biodiversity, governance, and socioeconomic sub-networks presented very similar values of estimated statistics (Table 1). Among them, the biodiversity sub-network with 632 edges showed the lowest value for density (0.150) and the highest value for reciprocity (0.259) and transitive triad (0.639). The socioeconomic sub-network with 678 edges had the highest value for density (0.161). The governance and socioeconomic sub-networks had the lowest value of reciprocity (0.251).

In general, the relative low levels of reciprocity and the high levels of transitivity found suggest that the presence of central actors could be more relevant in shaping the networks under study. In this sense, a central institution that better fits a specific collective action problem will receive more interactions, but will not necessarily reciprocate the ties in the same proportion for the other institutions.

Key institutions in collaborative networks

Centrality indices denote CER’s relevant institutions in the networks for solving collective action problems that may arise (Fig. 2). Node sizes are proportional to in-degree centrality values. All networks show few institutions with high and low centrality values, and many with intermediate centrality values (Appendix 4). In the general network (Fig. 2A) ICMBio was the most central institution having the highest in-degree value; owing to the fact that ICMBio is the agency in charge of CER management. Following an order of priority, the most important institutions are AMEX and the Women Network of Extractive Fishing Communities in the South of Bahia (REDE, in Portuguese), representing the beneficiaries’ sector. Particularly, REDE gives voice to the demands and needs of women of this sector and values the extractive work that they perform (Carmo et al. 2016, Dumith 2018). The priority of these institutions makes sense because both AMEX and REDE work as transversal institutions, as their objectives represent fishers independently of each particular community within the CER.

In contrast, results show that institutions representing governmental agencies locally (municipalities and city councils of Canavieiras, Belmonte, and Una) had an important connectivity with all the territory users institutions (nodes 26, 27, and 28). Also, they are the more peripheral institutions in the network, having the lowest in-degree values. The insufficient engagement in collaborative action of some institutions such as Belmonte government (node 6) and institutions representing non-traditional activities such as tourism (node 26) and rural producer sector (node 27), could be attributed to the fact that these were new institutions that got involved in the DC in the 2020–2022 period, when this study was carried out. Therefore, a substantial amount of collaborative relationships with other DC members had not yet been established.

Nodes representing beneficiaries’ institutions are displayed closer to each other, indicating that they share more connections. Despite having an intermediate in-degree value, institutions representing civil society had a position near the highest priority institutions, namely the universities (see node 31 and 32).

Interestingly, a similar pattern was observed in almost all sub-networks. In the biodiversity sub-network (Fig. 2B), the three main institutions that accompany ICMBio were AMEX, the National Commission for the Strengthening of Coastal and Marine Extractive Reserves (CONFREM, node 10), and the universities (UFBA/UESC, node 32). The governance sub-network (Fig. 2C) presents the most cited institutions AMEX, REDE (node 9), and the Z-20 Fisher’s Colony (node 16). The Z-20 Colony is one of the oldest associations of fishers in the region that defends the rights and interests of the artisanal fishing sector, being the institution responsible for registering fishers who can receive the benefit of seguro defeso (a period in which fishing for certain species is banned during the reproductive period and fishers can receive a salary). In contrast, in the socioeconomic sub-network (Fig. 2D) ICMBio and AMEX have very similar values of degree centrality, both being the highest priority institutions, followed by REDE.

Some institutions received more importance according to the evaluated sub-network. For example, universities (UFBA/UESC) have more relevance in biodiversity sub-networks, being more designated to monitoring contamination and pollution of mangroves and waters and carrying out the population monitoring of species of socioeconomic interest (Appendix 5). The navy institution (node 2) was nominated more in the governance sub-network than in other sub-networks, having a relevant role in surveillance (Appendix 5). Local government municipalities of Canavieiras, Belmonte, and Una cities (nodes 5, 6, and 8, respectively) received more attention in the socioeconomic sub-network, being addressed more often to collaborate with tourist activities implementation and quality of life and income improvement of traditional communities. This could be considered as evidence that actors choose the institutions to collaborate with, based on their perceptions of the attributes of their potential collaborators (such as the capacity to provide human, financial, and technical resources) that will provide a better fit for solving a specific collective action problem.

Figure 3 shows network node size proportional to the betweenness centrality values (Appendix 4). In the general network (Fig. 3A), the top six ranking institutions with the higher values were AMEX, CONFREM, ICMBio, UFBA/UESC, REDE, and Municipal Council for the Canavieiras Environment Defense (CONDEMA: civil society institution that proposes programs for environment protection, and decides the approval of all projects that involve an environmental decision). High betweenness centrality values indicates that all these institutions have the potential to act as a bridge, connecting other institutions. The same institutions also have an important role in biodiversity, governance, and socioeconomic sub-networks. In addition, the Segment of Artisans, Art makers and Fishing Gear (Artesãos) and Colônia Z-20 also have high relative values of betweenness centrality in the governance network (Fig. 3C), while the same happens for Association of Shellfishes and Fishers of Pedras de Una (AMEPEDRAS) in the socioeconomic network (Fig. 3D).

Subgroups structure

Subgroup detection identifies the main groups of institutions that interact more between them, in the networks, than with the rest of the institutions (Fig. 3). In all cases, the identified subgroups were not well defined because of a great degree of overlap, indicated by the low modularity values obtained (Table 2). This weak network partition in subgroups can be explained by the presence of bridging institutions that connect other institutions that do not interact, avoiding entirely fragmented subgroups.

In the general network (Fig. 3A), two not well-defined subgroups of nodes were identified. One corresponds to a subgroup (in light blue) of almost all the beneficiary institutions and universities (civil society institutions) sharing more connections between them. The other subgroup (in red) represents government and territory users’ institutions. It seems that homophily could be one factor shaping the subgroups. That is, the institutions may prefer to collaborate with others with similar interests and objectives, and thus are perceived as knowledgeable and trustworthy. The same pattern was observed for the socioeconomic sub-network (Fig. 3D).

Four overlapping subgroups were detected in the biodiversity sub-network (Fig. 3B). One of them (in light blue) included beneficiary institutions. This subgroup may show the formation of a “block of beneficiaries” composed, notably, of beneficiary institutions. The beneficiaries indicate another beneficiary institution as a potential partner, seeking to join forces to achieve common objectives in this specific aspect. The other subgroups clustered different types of institutions. For example, one subgroup (in purple) includes universities, and four beneficiary institutions (nodes 13 to 16). These interactions may reflect the potential partnerships between these traditional community institutions and the academy, aiming to establish specific studies involving different species of interest. For example, in the case of guaiamum (Cardisoma guanhumi), the CER is one of the areas in the country where the capture of this endangered species is allowed, as the Local Management Plan establishes the conditions of use, the minimum size, and the gear allowed in the catch, etc.

Three overlapping subgroups were detected in the governance sub-network (Fig. 3C). One group (the blue one) comprised most beneficiaries’ institutions along with one university. The other two subgroups aggregate different types of institutions: one (in red) is more represented by the government environmental agency, civil society, and territory user institutions, and the other (in green) is more represented by local government and civil society institutions. These results suggest that the governance sub-network is more complex, involving a broad spectrum of institutions, ranging from beneficiaries to government institutions, territory users, and civil society.

DISCUSSION

Through the statement of collective action problems related to biodiversity, governance, and socioeconomic issues we studied the collaborative networks between CER’s Deliberative Council institutions. We identified network structures that can promote the presence of social capital and leadership necessary to address the collective action problems that may arise.

Collaborative governance is one of the most promising ways to address environmental problems (Bodin 2017). Actors’ collaboration will determine network structures that would shape different co-management processes and outcomes (Bodin et al. 2006, Carlsson and Sandström 2008, Sandström and Rova 2010). Then, the success or failure of co-management network performance will be associated with how this network is structured (Bodin and Crona 2009, Alexander et al. 2015). However, there is no “ideal” network structure, instead, there is a tradeoff between certain characteristics supporting different processes (Bodin and Crona 2009, Alexander et al. 2015). Here, we discuss how the network structures might affect collective action and what implications might have on the resilience of the co-management of the CER.

Given that the government environmental agency responsible for CER co-management, ICMBio, was the most sought institution to deal with the different problems presented, we demonstrated the importance of this institution for collective action. In resource governance, it has been shown that a high degree of network centralization appears positively correlated with collective actions, mainly through central actors’ abilities to prioritize and coordinate activities (Carlsson and Sandström 2008). Because of the study’s methodological limitations we cannot discern why ICMBio has this central position. This could be because other institutions are obligated to interact with the ICMBio because of its attributions that other institution cannot fulfill, or as a result of ICMBio effective performance of its role fostering quality relationships of trust. Regardless, the study by Cardozo et al. (2019) found that ICMBio’s leadership was perceived as positive by the rest of the DC members, allowing the understanding of ICMBio’s central position as a coordinator facilitating collective action. Thus, future studies can deepen these issues and verify which factors determine the establishment of relationships within the DC and the strength of these interactions.

However, considering that ICMBio was the most demanded institution to solve any collective action problem that may arise highlights a certain degree of dependency on the state’s human, financial, and technical resources. This may have some negative implications for co-management dynamics related to this institution’s vertical and highly bureaucratic structure. Even if ICMBio was considered as nonhierarchical in this study, it has several levels, and previous studies highlight the cross-jurisdictional mismatch between local and national jurisdictions (Prado et al. 2020). Also, ICMBio depends on the Brazilian Ministry of Environment, so the federal government’s political agenda directly influences it (Prado et al. 2020). Then, situations such as changes in the environmental agencies and slow response time of ICMBio can interfere with the level of trust, reducing the degree of participation of traditional communities (Dumith 2018, Cardozo et al. 2019). An example of that was the environmental decisions of the last federal government (2019–2022) with budget cuts and the dismantling of federal agencies (including ICMBio; Thomaz et al. 2020, Barbosa et al. 2021), which jeopardized the role played by this central actor in the CER network. This situation is being reversed in the current federal government because environmental policies became a priority, bringing an encouraging scenario for protected areas in general (Peres et al. 2023).

Remembering that institutions are constituted by people, the role of ICMBio local manager gains relevance for its hybrid position between the state and the communities (Prado et al. 2020). Nevertheless, often because of different pressures and disagreements, many ICMBio CER’s local managers quit this job (Dumith 2018, Prado et al. 2020). Some studies warn that when an institution leader changes, the relational ties with other institutions could disappear, having consequences for the co-management (Alexander et al. 2015). Even more cautiousness is required if this institution occupies a central position in the network (Bodin et al. 2006). Then, this high dependency on ICMBio can have implications for the dynamics of adaptive co-management, compromising the resilience of the social-ecological systems because collective action initiatives may encounter obstacles due to the rigidity of the central structure, delaying or hampering the passage from discussion to action.

Traditional communities showed a relevant role in collaborative networks, AMEX being the second most important institution sought to work with, even without being an official DC member. The fact that AMEX decided not to belong to the DC is a political decision that aims to strengthen the voice and vote of small fishers and shellfish associations in the region, enabling new leaders to emerge. The implication of not being on the DC does not reflect AMEX’s active participation in governance. Several works address that legitimate and well-engaged institutions are critical for successful co-management, providing resilience to changes in the governance (Gutiérrez et al. 2011, Crona et al. 2017). Also, the network’s proximity between AMEX and ICMBio showed that these key institutions shared many connections and perceived each other as priority institutions to work with. This supports the recognition that traditional communities and government need to act together, watching for the success of the co-management (Alexander et al. 2016).

AMEX, CONFREM, ICMBio, UFBA/UESC, REDE, and CONDEMA are relevant institutions acting like a bridge connecting other DC members. In this sense, a decentralization of the bridging role can be evidenced in which multiple institutions representing different sectors (government, civil society, and beneficiaries) acquire relevance providing functional redundancy. The network acquires resilience to potential disturbances (e.g., political, economic, etc.) that may occur to some key institutions. In this way, the network can absorb this impact and not lose its underlying configuration (Jones et al. 2013). The presence of bridging institutions also can avoid the network partition in subgroups. The interaction between actors of different and heterogeneous groups is necessary to create common understanding for collective action in small-scale fisheries (Bodin et al. 2006). These factors are critical for the success of co-management as they provide resilience to changes in the governance of the socio-ecosystem (Bodin and Crona 2009).

Local government institutions, like municipalities and city councils, and some territory user institutions were clearly peripheral in the networks. The low engagement that local government institutions and the Association of Canavieiras Shrimp Producers (ACCC, a non-traditional activity) have could indicate that old conflicts related to the CER creation process are still latent (Dias et al. 2018, Cardozo et al. 2019). This also can be made evident by the low participation of these institutions in decision making (Cardozo et al. 2012, 2019, Dumith 2018). These conflicts have their roots in the lack of a common vision and objectives, and in the presence of different interests and concerns about natural resources access and management (Dumith 2018). Caution should be taken in affirming this because we focused only on positive relationships (e.g., collaboration) and did not evaluate negative relationships (e.g., conflict). Further investigation is needed to examine if cooperative or conflicting relations predominate shaping the CER’s networks, mainly because neglecting underlying conflicts among actors can have significant implications in collective action network outcomes (Robins et al. 2011, Bodin et al. 2020b).

Moreover, ignoring this conflict of interests could intensify power imbalances that could hinder collective action (Bodin et al. 2020b). In this context, local governments are power institutions with the capacity to mobilize different types of resources, and their network’s peripheral position does not mean that they are not considered key institutions to work with. Most people interviewed manifested the importance and the need to work with these institutions in collective actions, but they also highlighted the lack of support and recognition to the traditional extractive activities that Canavieiras and Belmonte government shows. This also was found in the works of Dumith (2018) and Cardozo et al. (2019). This lack of support translates into a lack of engagement of these institutions, which can end up obstructing collective action attempts. This shortage of support was made explicit years ago: shortly before the creation of the CER, there were demonstrations by sectors opposed to its establishment (business people from the shrimp farming sector, from the hotel sector, and from the local trade); even years after the creation of the CER, there was a political movement to try to reclassify the area as an Environmental Protection Area (APA, in Portuguese). An APA is a more permissive category of protected area for certain developments. This attempt was carried out through a bill sent to the National Congress in 2015. In 2016, this bill was considered inconsistent and was archived in 2019 (Aguiar et al. 2022). This demonstrates the existence of long-lasting conflicts of interest in the CER region, which may explain the absence of certain relationships within the DC.

In all networks a subgroup formed by almost all the beneficiary institutions can be distinguished. This denoted the social capital existent among traditional communities (Bodin and Crona 2009). This type of close bonding structure could foster mutual trust, commitment, and exertion of social control (Berardo and Scholz 2010). It is of huge importance because the absence of self-organization, limited participation, and a lack of a sense of community are among the factors that cause failures in community-based resource management (Alexander et al. 2015). The organization of the beneficiary institutions and the cohesion of their goals and purposes were emphasized in previous studies (Prost 2018, Cardozo et al. 2019). The beneficiaries’ self-organization was evident in the socioeconomic dimension. Traditional communities are organized to formulate different community projects related to improving quality of life, diversifying income sources, and adding value to the production chain. It seems that this experience gives them the ability to learn and deal with changes and difficulties that CER faced in a short time interval, like the oil spill on the Canavieiras coastlines and mangroves (in 2019), and the SARS-CoV-2 Pandemic (Silva et al. 2022). Issues addressed in the governance sub-network relate to information sharing between DC members and beneficiary communities, surveillance, participation, and access to natural capital. These factors connect, to a greater or lesser extent, with principles of natural resource governance. For example, information sharing (from DC members to communities and vice versa) is related to the principles of transparency, inclusiveness, and adaptability (Lockwood et al. 2010). In this sense, the interactions of this sub-network highlight the relevant role involving AMEX, REDE, and the fishers association Colônia Z-20, which, together, enable a more equitable participation, giving opportunity for speech and participation to a wide range of actors and recognize the role of women in the decision-making process, this being another fundamental aspect for community-centered conservation initiatives (Armitage et al. 2020).

In relation to the latter, it is important to highlight the strong female role in the DC (of the 30 interviewees, 14 were women), unlike others MERs, in which women’s role in DC meetings was restricted and subordinate to men (Di Ciommo and Schiavetti 2012). This suggests that the participation of REDE in the DC can encourage women to give their opinions and defend their interests. This, in turn, can strengthen more equitable power relationships that facilitate more harmonious dynamics between genders to promote collective actions in the management of natural resources. Therefore, the promotion of knowledge and experience exchanges between women of different MERs could be fruitful. Thus, the REDE can play an important political role in female empowerment not only in the CER, but in other areas of collective decision making. The existence of exchange platforms or networks can be of fundamental relevance in adaptive learning and resource management, increasing social-ecological resilience (Berkes and Turner 2006).

Universities are grouped with beneficiaries, indicating that these institutions work together and know the importance of integrating traditional with technical and scientific knowledge to address management problems (Bodin et al. 2006). This close relationship must be due to the existence of previous collaboration on certain matters regarding the universities’ scope. The previous experiences of collective action seem to be an important attribute for the choice to collaborate with another institution (Nohrstedt and Bodin 2020). This joint work was evident in formulating the guaiamum (Cardisoma guanhumi) management program (MMA and ICMBio 2020). It is important for the system’s resilience that heterogeneous actors (with different educational backgrounds, roles, and occupations) work together because it facilitates learning about complex problems (Bodin et al. 2006). On the contrary, actors interacting only within their subgroups could reinforce current perceptions, hindering the possibilities of emerging new ideas, which could be detrimental to management problems resolution (Bodin 2017).

Our findings provide guidance, through social network analysis, to consider relationships in a systematic way. Because we identified attributes in the collaborative networks that could compromise collective action, this allows proposed interventions and direct efforts to improve social capital and leadership. The collaborative networks in CER reveal the practical implementation of a formal institutional co-governance framework. Even if CER co-management is well established in the decision-making arena (with stable decision forums and clear rules ensuring the participation of different actors), it seems like, in the process of moving from discussion to collective action, almost everything remains in the ICMBio hands. This can have negative implications for the dynamics of the co-management process, compromising the resilience of the social-ecological system. This is attributable to the state’s bureaucratic structure, delaying decisions taken locally, but implemented late (if at all). In this sense, the power structure of the DC (effectively led by a state institution) does not always favor the active participation of users in resource management. In such manner, we understand that the most appropriate way of strengthening the collaborative network is through greater decentralization of power, that is, local representative institutions (such as AMEX and CONFREM) should have greater influence on final decisions, instead of strengthening and empowering more the figure of the manager, as suggested by Prado et al. (2020). One way to achieve this, could be through the restructuring of national legislation aimed at the co-management of natural resources, so that the political structure could be more appropriate for decision making at the local level. Examples of changes in laws and policies at different levels are seen in other places where small-scale and subsistence fishing are relevant (Caillaud et al. 2004). However, we recognize that putting this into practice is challenging because it implies a paradigm shift within the state bureaucratic structure, streamlining and respecting the decisions made locally by the DC.

The network peripheral position of some institutions representing local governments, civil society, and territory users would be diminishing DC potential for collective action. Therefore, different ways should be explored to achieve a real integration of the DC. This could be done through the training of DC members, aiming to understand how DC works and how institutions participate. Also, the generation of activities that promotes institutions to work together interacting and communicating more is important; this could help to find a shared vision that can build some type of consensus or synergy between the different interests.

CONCLUSION

The collaborative networks analyzed denote the presence of network structures that can promote collective action. ICMBio was the most demanded institution to solve the different problems presented, and this was partly explained because it is the environmental agency responsible for CER co-management. Even if ICMBio can promote collective action, as it has the experience and capacity to provide different types of resources and services, its centralization denotes a high degree of dependence on this institution. Therefore, we believe that there must be a political effort (or a paradigm shift) so that the bureaucratic structure of the DC, led by the state, is reformulated and that the legislation on protected areas, such as MERs, is re-discussed with the purpose of giving greater autonomy for users to make decisions at the local level.

Nevertheless, social capital was high among beneficiaries’ institutions indicating that they are organized and have recognized AMEX as their leader. Also, AMEX was the second most important institution sought to work with, having a solid relationship with ICMBio, reinforcing the idea that traditional community and government need to work together to ensure the success of co-management. In addition to ICMBio and AMEX, CONFREM and REDE played the role of bridge actors, connecting other institutions essential for the success of co-management, as they provide resilience to changes in the governance of the system. Furthermore, the role of the REDE must be highlighted as a relevant element for equity in power relations between genders within the DC. Their role must be strengthened in a way that could induce greater female participation not only in CER, but in other collective spaces.

It should be stressed that the networks are static representations of a dynamic social process. However, collaborative networks are not static and continually evolve as actors adjust to different endogenous and exogenous drivers of change (Alexander et al. 2016, Bodin et al. 2017); then, long-term studies would be necessary for understanding these dynamics.

Furthermore, the findings of our study have to be seen in light of some limitations. First, the study could not determine the reasons why each institution selected (or not) another to work with. In addition, the strength of relationships between institutions was not possible to establish. In this sense, future research should incorporate these elements and investigate the way in which interactions (or the lack of them) can compromise collective actions aimed at the management of common-pool resources.

Future studies are needed to determine statistical validity of network structures. Moreover, it would be interesting moving beyond collaborative relationships, via the examination of both, facilitating and hindering ties with the attempt to elucidate whether there are cooperative or conflicting relations that predominate shaping collective action networks. Further investigation into which factors, whether an actor’s attributes or the nature of the collective action problems, influence the choice of an institution engaging with another, will be necessary for understanding the patterns found.

LITERATURE CITED

Aceves‐Bueno, E., S. J. Miller, J. Cornejo‐Donoso, and S. D. Gaines. 2020. Cooperation as a solution to shared resources in territorial use rights in fisheries. Ecological Applications 30(1):e02022. https://doi.org/10.1002/eap.2022

Aguiar, P. C. B., M. M. Pires, C. C. Profice, and N. L. Bruno. 2022. Origin and mosaic of the landscape of the Canavieiras Marine Extractive Reserve, Bahia, Brazil. Research, Society and Development 11(7):e41311730240. https://doi.org/10.33448/rsd-v11i7.30240

Alexander, S. M., M. Andrachuk, and D. Armitage. 2016. Navigating governance networks for community-based conservation. Frontiers in Ecology and the Environment 14(3):155-164. https://doi.org/10.1002/fee.1251

Alexander, S. M., and D. Armitage. 2015. A social relational network perspective for MPA science. Conservation Letters 8(1):1-13. https://doi.org/10.1111/conl.12090

Alexander, S. M., D. Armitage, and A. Charles. 2015. Social networks and transitions to co-management in Jamaican marine reserves and small-scale fisheries. Global Environmental Change 35:213-225. https://doi.org/10.1016/j.gloenvcha.2015.09.001

Alexander, S. M., Ö. Bodin, and M. L. Barnes. 2018. Untangling the drivers of community cohesion in small-scale fisheries. International Journal of the Commons 12(1):519-547. https://doi.org/10.18352/ijc.843

Armitage, D., P. Mbatha, E. K. Muhl, W. Rice, and M. Sowman. 2020. Governance principles for community-centered conservation in the post-2020 global biodiversity framework. Conservation Science and Practice 2(2):e160. https://doi.org/10.1111/csp2.160

Barbosa, L. G., M. A. S. Alves, and C. E. V. Grelle. 2021. Actions against sustainability: dismantling of the environmental policies in Brazil. Land Use Policy 104:105384. https://doi.org/10.1016/j.landusepol.2021.105384

Berardo, R., M. Fischer, and M. Hamilton. 2020. Collaborative governance and the challenges of network-based research. American Review of Public Administration 50(8):898-913. https://doi.org/10.1177/0275074020927792

Berardo, R., and J. T. Scholz. 2010. Self-organizing policy networks: risk, partner selection, and cooperation in estuaries. American Journal of Political Science 54(3):632-649. https://doi.org/10.1111/j.1540-5907.2010.00451.x

Berkes, F. 2021. Toward a new social contract: community-based resource management and small-scale fisheries. TBTI Global, St. John’s, Newfoundland and Labrador, Canada.

Berkes, F., and N. J., Turner. 2006. Knowledge, learning and the evolution of conservation practice for social-ecological system resilience. Human Ecology 34:479-494. https://doi.org/10.1007/s10745-006-9008-2

Bodin, Ö. 2017. Collaborative environmental governance: achieving collective action in social-ecological systems. Science 357(6352). https://doi.org/10.1126/science.aan1114

Bodin, Ö., J. Baird, L. Schultz, R. Plummer, and D. Armitage. 2020a. The impacts of trust, cost and risk on collaboration in environmental governance. People and Nature 2(3):734-749. https://doi.org/10.1002/pan3.10097

Bodin, Ö., and B. Crona. 2008. Management of natural resources at the community level: exploring the role of social capital and leadership in a rural fishing community. World Development 36(12):2763-2779. https://doi.org/10.1016/j.worlddev.2007.12.002

Bodin, Ö., and B. Crona. 2009. The role of social networks in natural resource governance: what relational patterns make a difference? Global Environmental Change 19(3):366-374. https://doi.org/10.1016/j.gloenvcha.2009.05.002

Bodin, Ö., B. Crona, and H. Ernstson. 2006. Social networks in natural resource management: what is there to learn from a structural perspective? Ecology and Society 11(2):r2. https://doi.org/10.5751/ES-01808-1102r02

Bodin, Ö., M. M. García, and G. Robins. 2020b. Reconciling conflict and cooperation in environmental governance: a social network perspective. Annual Review of Environment and Resources 45:471-495. https://doi.org/10.1146/annurev-environ-011020-064352

Bodin, Ö., A. Sandström, and B. Crona. 2017. Collaborative networks for effective ecosystem-based management: a set of working hypotheses. Policy Studies Journal 45(2):289-314. https://doi.org/10.1111/psj.12146

Brandes, U., D. Delling, M. Gaertler, R. Görke, M. Hoefer, Z. Nikoloski, and D Wagner. 2008. On modularity clustering. IEEE Transactions on Knowledge and Data Engineering 20(2):172-188. https://doi.org/10.1109/TKDE.2007.190689

Brasil. 2009. Portaria no- 71, de 3 de setembro de 2009. Cria o Conselho Deliberativo da Reserva Extrativista de Canavieiras. Brasília, Brasil.

Brasil. 2018. Portaria nº 313, de 12 de abril de 2018. Aprova o Acordo de Gestão da Reserva Extrativista de Canavieiras. Brasília, Brasil.

Caillaud, A., S. Boengkih, E. Evans-Illidge, J. Genolagani, P. Havemann, D. Henao, E. Kwa, D. Llewell, A. Ridep-Morris, J. Rose, R. Nari, P. Skelton, R. South, R. Sulu, A. Tawake, B. Tobin, S. Tuivanuavou, and C. Wilkinson. 2004. Tabus or not taboos? How to use traditional environmental knowledge to support sustainable development of marine resources in Melanesia. SPC Traditional Marine Resource Management and Knowledge Information Bulletin 17:14-35.

Cardozo, L. S., D. T. Alarcon, S. Campiolo, and A. Schiavetti. 2019. Governança ambiental e percepção sobre processos participativos na Reserva Extrativista de Canavieiras, Bahia, Brasil. Desenvolvimento e Meio Ambiente 50:170-191.

Cardozo, L. S., M. F. Porto, P. C. Pimentel, J. S. Rodrigues, A. Schiavetti, and S. Campiolo. 2012. Discussões do Conselho Deliberativo da Reserva Extrativista de Canavieiras, Bahia, Brasil: da gestão pesqueira à ambiental. Revista de Gestão Costeira Integrada 12(4):463-475. https://doi.org/10.5894/rgci367

Carlsson, L., and F. Berkes. 2005. Co-management: concepts and methodological implications. Journal of Environmental Management 75(1):65-76. https://doi.org/10.1016/j.jenvman.2004.11.008

Carlsson, L. G., and A. Sandström. 2008. Network governance of the commons. International Journal of the Commons 2(1):33-54. https://doi.org/10.18352/ijc.20

Carmo, J. C., M. D. Pires, G. D. Jesus, A. L. Cavalcante, and S. Trevizan. 2016. Voz da natureza e da mulher na Resex de Canavieiras-Bahia-Brasil: sustentabilidade ambiental e de gênero na perspectiva do ecofeminismo. Revista Estudos Feministas 24:155-180. https://doi.org/10.1590/1805-9584-2016v24n1p155

Cohen, P. J., L. S. Evans, and M. Mills. 2012. Social networks supporting governance of coastal ecosystems in Solomon Islands. Conservation Letters 5(5):376-386. https://doi.org/10.1111/j.1755-263X.2012.00255.x

Costanza, R., R. de Groot, P. Sutton, S. van der Ploeg, S. J. Anderson, I. Kubiszewski, S. Farber, and R. K. Turner. 2014. Changes in the global value of ecosystem services. Global Environmental Change 26:152-158. https://doi.org/10.1016/j.gloenvcha.2014.04.002

Crona, B., S. Gelcich, and Ö. Bodin. 2017. The importance of interplay between leadership and social capital in shaping outcomes of rights-based fisheries governance. World Development 91:70-83. https://doi.org/10.1016/j.worlddev.2016.10.006

Csardi, G., and T. Nepusz. 2006. The igraph software package for complex network research. InterJournal, Complex Systems 1695(5):1-9.

Degenne, A., and M. Forse. 1999. Introducing social networks. SAGE, London, UK. https://doi.org/10.4135/9781849209373

Dias, M., R. Gomes, S. Batista, S. Campiolo, and A. Schiavetti. 2018. Participação popular na criação de unidades de conservação marinha: o caso da Reserva Extrativista de Canavieiras. Revista Direito GV 14(3):912-936. https://doi.org/10.1590/2317-6172201834

Di Ciommo, R. C., and A. Schiavetti. 2012. Women participation in the management of a Marine Protected Area in Brazil. Ocean & Coastal Management 62:15-23. https://doi.org/10.1016/j.ocecoaman.2012.02.010

Dumith, R. D. C. 2018. Dez anos de r-existência da Reserva Extrativista de Canavieiras (BA): análise dos conflitos inerentes à reprodução social e política das suas comunidades tradicionais. Desenvolvimento e Meio Ambiente 48:367-391.

Freeman, L. C. 1977. A set of measures of centrality based on betweenness. Sociometry 40(1):35-41. https://doi.org/10.2307/3033543

Freeman, L. C., D. Roeder, and R. Mulholland. 1979. Centrality in social networks: II. Experimental results. Social Networks 2(2):119-141. https://doi.org/10.1016/0378-8733(79)90002-9

Fruchterman, T., and E. M. Reingold. 1991. Graph drawing by force-directed placement. Software: Practice and Experience 21(11):1129-1164. https://doi.org/10.1002/spe.4380211102

Gutiérrez, N. L., R. Hilborn, and O. Defeo. 2011. Leadership, social capital and incentives promote successful fisheries. Nature 470(7334):386-389. https://doi.org/10.1038/nature09689

Jones, P. J. S., W. Qiu, and E. M. De Santo. 2013. Governing marine protected areas: social-ecological resilience through institutional diversity. Marine Policy 41:5-13. https://doi.org/10.1016/j.marpol.2012.12.026

Kluger, L. C., P. Gorris, S. Kochalski, M. S. Mueller, and G. Romagnoni. 2020. Studying human-nature relationships through a network lens: a systematic review. People and Nature 2(4):1100-1116. https://doi.org/10.1002/pan3.10136

Laumann, E. O., P. V. Marsden, and D. Prensky. 1989. The boundary specification problem in network analysis. Pages 61-87 in L. C. Freeman, D. R. White, and A. K. Romney, editors. Research methods in social network analysis. George Mason University Press, Fairfax, Virginia, USA.

Lockwood, M., J. Davidson, A. Curtis, E. Stratford, and R. Griffith. 2010. Governance principles for natural resource management. Society and Natural Resources 23(10) 986-1001. https://doi.org/10.1080/08941920802178214

Marín, A., S. Gelcich, J. C. Castilla, and F. Berkes. 2012. Exploring social capital in Chile’s coastal benthic comanagement system using a network approach. Ecology and Society 17(1):13. https://doi.org/10.5751/ES-04562-170113

Margerum, R. D. 2008. A typology of collaboration efforts in environmental management. Environmental Management 41(4):487-500. https://doi.org/10.1007/s00267-008-9067-9

Ministério do Meio Ambiente (MMA). 2010. Conselho Deliberativo da Reserva Extrativista de Canavieiras. Regimento Interno. MMA, Brasília, Brasil.

Ministério do Meio Ambiente and Instituto Chico Mendes de Conservação da Biodiversidade (MMA and ICMBio). 2020. Portaria Nº 1.076, de 13 de novembro de 2020 Aprova o plano de gestão local do guaiamum (Cardisoma guanhumi) na Reserva Extrativista Marinha de Canavieiras e dá outras providências (02125.001081/2019-01). MMA and ICMBio, Brasília, Brasil.

Newman, M. E. J, and M. Girvan. 2004. Finding and evaluating community structure in networks. Physical Review E 69(2):026113. https://doi.org/10.1103/PhysRevE.69.026113

Nobre, D., and A. Schiavetti. 2013. Fisheries agreements, governance and Deliberative Council of Extractive Reserve: the case of Resex Cassurubá, Caravelas, Bahia. Boletim do Instituto de Pesca 39(4):445-455.

Nohrstedt, D., and Ö. Bodin. 2020. Collective action problem characteristics and partner uncertainty as drivers of social tie formation in collaborative networks. Policy Studies Journal 48(4):1082-1108. https://doi.org/10.1111/psj.12309

Ostrom, E. 1990. Governing the commons: the evolution of institutions for collective action. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/CBO9780511807763

Peres, C. A., J. Campos-Silva, and C. D. Ritter. 2023. Environmental policy at a critical junction in the Brazilian Amazon. Trends in Ecology & Evolution 38(2):113-116. https://doi.org/10.1016/j.tree.2022.11.011

Prado, D. S., F. De Castro, and C. S. Seixas. 2020. The dramas of the Marine Extractive Reserve Managers in Brazil: navigating through multiple institutions. Society and Natural Resources 33(5):651-668. https://doi.org/10.1080/08941920.2019.1690722

Prell, C. 2011. Some basic structural characteristics of networks. Pages 29-43 in Ö. Bodin and C. Prell, editors. Social networks and natural resource management: uncovering the social fabric of environmental governance. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/CBO9780511894985.003

Prost, C. 2018. Reservas Extrativistas Marinhas: avanço ou retrocesso? Desenvolvimento e Meio Ambiente 48:321-342.

R Core Team. 2019. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.

Ramirez-Sanchez, S., and E. Pinkerton. 2009. The impact of resource scarcity on bonding and bridging social capital: the case of fishers’ information-sharing networks in Loreto, BCS, Mexico. Ecology and Society 14(1):22. https://doi.org/10.5751/ES-02841-140122

Robins, G., L. Bates, and P. Pattison. 2011. Network governance and environmental management: conflict and cooperation. Public Administration 89(4):1293-1313. https://doi.org/10.1111/j.1467-9299.2010.01884.x

Sandström, A., and C. Rova. 2010. Adaptive co-management networks: a comparative analysis of two fishery conservation areas in Sweden. Ecology and Society 15(3):14. https://doi.org/10.5751/ES-03531-150314

Santos, C. Z., and A. Schiavetti. 2013. Marine extractive reserves of Brazil: contradictions of legal order, sustainability and ecological aspect. Boletim do Instituto de Pesca 39(4):479-494.

Santos, C. Z., and A. Schiavetti. 2014. Assessment of the management in Brazilian Marine Extractive Reserves. Ocean and Coastal Management 93:26-36. https://doi.org/10.1016/j.ocecoaman.2014.03.007

Silva, F. R., A. Schiavetti, A. C. M. Malhado, B. Ferreira, C. V. P. Sousa, F. P. Vieira, F. R Pinto, G. B. G. Souza, G. Olavo, J. B. Q. Santos, J. V. Campos-Silva, J. G. C. Oliveira Júnior, L. T. Messias, M. L. V. Barbosa Filho, M. C. Accioly, N. N. Fabré, P. R. Abdallah, P. F. M. Lopes, R. K. P. Kikuchi, S. F. Câmara, V. S. Batista and M. O. Soares. 2022. Oil spill and socioeconomic vulnerability in marine protected areas. Frontiers in Marine Science 9:859697. https://doi.org/10.3389/fmars.2022.859697

Thomaz, S. M., L. G. Barbosa, M. C. S. Duarte, and R. Panosso. 2020. Opinion: the future of nature conservation in Brazil. Inland Waters 10(2):295-303. https://doi.org/10.1080/20442041.2020.1750255

Wasserman, S., and K. Faust. 1994. Social network analysis: methods and applications. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/CBO9780511815478