How Critical Stakeholders Interact for Multiple-Use Marine Protected Area Implementation

Ocean & Coastal Management 91 (2014) 5e16

Contents lists available at ScienceDirect

Ocean & Coastal Management

journal homepage: www.elsevier.com/locate/ocecoaman

Collaboration and knowledge networks in coastal resources management: How critical stakeholders interact for multiple-use marine protected area implementation

P. Francisco Cárcamo a,b, Rosa Garay-Flühmann c, Carlos F. Gaymer d,e,* a Departamento de Biología Marina, Universidad Católica del Norte, Programa de Doctorado en Biología y Ecología Aplicada, Larrondo 1281, Coquimbo, Chile b Departamento de Medio Ambiente, Instituto de Fomento Pesquero, Doctor Marín 340, Coquimbo, Chile c Centro de Aprendizaje, Universidad Santo Tomás, Sede La Serena, Ruta 5 Norte 1068, La Serena, Chile d Departamento de Biología Marina, Universidad Católica del Norte & Centro de Estudios Avanzados en Zonas Áridas (CEAZA), Larrondo 1281, Coquimbo, Chile e Instituto de Ecología y Biodiversidad (IEB), Chile article info abstract

Article history: Recent studies have shown that social networks and their properties are key in the search for expla- Available online nations for the success in the governance and management of natural resources. We investigated the structure and properties of inter-organizational social networks involved in the use and management of natural resources in a coastal marine ecosystem in northern Chile proposed as a possible marine protected area. We explored two network configurations: i) relations of collaboration and ii) relations of transfer and exchange of scientific knowledge and information useful to the management of natural resources and the decision-making process involved. Both networks showed little cohesion, with low values of centralization and density indicating a low flow of collaborative and exchange relations among different stakeholders. The knowledge network achieved greater levels of centralization than the collaboration network. National government agencies and fishermen organizations were the most powerful stakeholders in the collaboration network. National government agencies and universities were the most powerful stakeholders in the knowledge network. We found a disconnected network when analyzing the flows of collaboration and knowledge between different administrative operation levels that potentially would hamper the governance of this area. On the other hand, we identified stakeholders that would help to connect the network (bridging stakeholders). Additional analysis of stakeholders using influence-capacity matrix allowed us to identify key stakeholders for planning and implementing the new marine protected area, compare results with the network analysis, and propose network interventions. We propose combining Social Network Analysis with other methods of stakeholder analysis to produce more practical and implementable results. Our results are relevant for future interventions aimed at improving or implementing management and governance of coastal areas. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction place-based approach that considers the entire ecosystem and the connections between its various components, including the strong Ecosystem-based management (EBM) is promoted as an alter- connections between social and natural systems, and focuses on native to traditional approaches for managing natural resources maintaining the ecosystem in a healthy, productive, and resilient (Kidd et al., 2011; McLeod and Leslie, 2009). EBM corresponds to a condition so that it can provide the services that humans want and need (McLeod and Leslie, 2009). An adaptive and ﬂexible management network has been suggested as necessary for implementing EBM. This means adopting organizational changes aiming * Corresponding author. Departamento de Biología Marina, Universidad Católica to achieve: i) interdisciplinary collaboration across multiple levels; del Norte & Centro de Estudios Avanzados en Zonas Áridas (CEAZA), Larrondo 1281, ii) involving stakeholders and considering their knowledge and Coquimbo, Chile. Tel.: þ56 51 2209969; fax: þ56 51 2209812. ﬁ E-mail addresses: [email protected] (P.F. Cárcamo), rosagaray@ opinions; iii) incorporating and sharing relevant scienti c infor- santotomas.cl (R. Garay-Flühmann), [email protected] (C.F. Gaymer). mation; iv) integrating science and policies; and v) making

0964-5691/$ e see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ocecoaman.2014.01.007 6 P.F. Cárcamo et al. / Ocean & Coastal Management 91 (2014) 5e16 collaborative decisions (Carollo and Reed, 2010; Granek et al., potential collaboration networks, key stakeholders to lead the 2010). Any region to be managed contains a range of stakeholders implementation, thus providing inputs for the design or redesign of who are involved, interested, or affected by the decisions regarding management plans. natural resources management (Pomeroy and Douvere, 2008). This article investigates the structure and properties of inter- Indeed, many management and conservation initiatives fail organizational social networks associated to the use and manage- because the knowledge, relations, interests, and characteristics of ment of natural resources in a highly productive and biologically stakeholders are poorly incorporated into the equation (Prell et al., diverse coastal marine ecosystem hotspot in northern Chile. We 2008). On the other hand, there is enough evidence to suggest that explored two configurations of networks: i) relations of collabo- the acceptance, involvement, and early participation of the affected ration and ii) relations of transfer and exchange of scientific community in planning and decision-making activities are key for knowledge and information on natural resources, which is useful to effective implementation of marine conservation and management their management or the decision-making processes. By describing measures (Pomeroy and Douvere, 2008; Rodríguez-Martínez, how organizations connect with each other (i.e., whole network 2008). structure, groups, positions and roles), we analyzed the potential of Next, to achieve good and effective governance in protected these networks to support the implementation and operation of a areas, institutional capacity and supportive contexts are needed new marine protected area (MPA). Additionally, and to contrast (Lockwood, 2010). Therefore, it is a crucial issue to investigate and with the results of SNA, we conducted an SA using influence and develop the capacity amongst local resource users to sustainably capacity of stakeholders as variables to identify and categorize key manage natural resources and the capacity amongst relevant gov- stakeholders for an MPA planning and implementation process. ernment authorities to enforce management decisions (Jones and Finally, we discuss the value and applicability of these tools in the Burgess, 2005). context of governance and management of natural resources in Stakeholder analysis (SA) is understood as a process that: i) coastal areas. defines components of a social and natural system affected by a decision or action; ii) identifies and describes individuals and groups who are affected by or can affect different components of 2. Methods the system (e.g., relations, interests, objectives); and iii) ranks and prioritizes these individuals and groups for involvement in the 2.1. Study area and target population decision-making process (Pomeroy and Douvere, 2008; Reed et al., 2009). Therefore, given the importance of the human system in The study area corresponds to a system of coastal islands in EBM, SA is a key issue to be carried out prior to its implementation. northern Chile (28 540Se29 450 S; Fig. 1) belonging to the tran- Social Network Analysis (SNA) adds to SA the study of social sition region of the temperate Humboldt Current System (Thiel structure and the relations between key actors (Freeman, 2004; et al., 2007). Their fishing and ecological characteristics have Reed et al., 2009). SNA has been widely used by social scientists made it an area of great interest for both fisheries and conservation to explain various social phenomena (Borgatti et al., 2009). How- (Luna-Jorquera et al., 2012; Thiel et al., 2007). Several marine and ever, empirical research in natural resources management and terrestrial areas have been established for biodiversity conservation environmental governance networks applying quantitative analysis and management and exploitation of benthic resources (Cárcamo is recent and scarce (Bodin and Crona, 2009; Bodin and Prell, 2011). and Gaymer, 2013). Fishing is the main source of employment for Social networks among the different stakeholders may facilitate the residents of the small coastal villages, however, in recent years or obstruct collaboration and problem solving related to natural due to tourism growth, fishermen have ventured into wildlife resources (Bodin and Prell, 2011; Grafton, 2005; Marín and Berkes, watching activities (Cárcamo and Gaymer, 2013). The possible 2010; Pretty and Ward, 2001). Recent research has shown that installation of coal-fired power plants near these protected areas social networks and their structural properties affect collaborative generated social conflicts and controversies for their location and processes and, therefore, should be considered as key variables in incompatibility with existing uses and biodiversity relevance the search for explanations for the success or failure in natural (Cárcamo et al., 2011). Nowadays, potential conflicts are foreseen resources management (Bodin and Crona, 2009; Bodin and Prell, with mining ports projected to be built on the same coastline. In 2011; Cohen et al., 2012; Marín and Berkes, 2010; Ramirez- 2009 this area was defined as priority site for biodiversity conser- Sanchez and Pinkerton, 2009; Weiss et al., 2012). Identifying the vation by the Corporación Nacional Forestal (CONAF, National structural properties of social networks is useful for understanding Forestry Service) and the Servicio Nacional de Pesca (SERNAPESCA, the power dynamics, knowledge and information exchange (e.g., National Fisheries Service), and in 2010 a non-governmental or- traditional or scientific knowledge for management), and ganization (NGO) proposed the Chilean Government to evaluate improving co-management arrangements to increase the adaptive the declaration of this area as a Multiple-Use Coastal Marine Pro- capacity and resilience of social-ecological systems (Bodin and tected Area (MU-CMPA). Prell, 2011; Hughes et al., 2005; Weiss et al., 2012). Thus, SNA of Regarding the political-administrative division of Chile, the management and governance of natural resources also becomes a study area encompasses the coastal areas of two administrative valuable tool to examine the relations between stakeholders, to regions: the Freirina commune (Huasco Province, Atacama Region) determine how different stakeholders are positioned within the and La Higuera commune (Elqui Province, Coquimbo Region). network, and to reveal how relations are structured within the Recently, Cárcamo et al. (2013) studied the same area and evaluated whole network. SNA can also be a tool to identify, select, and the functional fit between a conceptual ecosystem defined by engage stakeholders in participatory processes associated with stakeholders (and based on ecosystem services, threats and uses/ natural resources management (Bodin and Prell, 2011; Prell et al., activities) and the Chilean legislation related to coastal and marine 2009; Vance-Borland and Holley, 2011). resources planning and management. Results of this study revealed In the context of the transition of current governance systems a low functional fit between stakeholders’ definitions and the towards approaches such as EBM, stakeholders and social network current legal framework. However, they noted that some existing analysis can contribute to its implementation in particular areas instruments in the current legislation, such as MU-CMPA, can be identifying, among others, patterns and networks of scientific used as a basis for short-term implementation of a management knowledge exchange and transfer, power networks, current and strategy based on the principles and criteria of EBM. P.F. Cárcamo et al. / Ocean & Coastal Management 91 (2014) 5e16 7

Fig. 1. Research area indicating the existing conservation and management areas. The dotted line indicates the area proposed as MU-CMPA.

We focused our research on institutional interactions, consid- coastal marine area of La Higuera and Freirina communes. Based on ering individual representatives of organizations or institutions three information sources (conservation participatory committees that are involved, interested or affected by initiatives of use, man- meetings, protected areas advisory boards and operating commit- agement and/or conservation of natural resources that occur in the tees, and names request to key informants) we triangulated 8 P.F. Cárcamo et al. / Ocean & Coastal Management 91 (2014) 5e16

(individuals or organizations cited in two or three sources) and from very low to very high), and to determine the annual fre- defined an initial group of twenty key informants to which we quency of contacts (e.g., meetings or workshops per year). In the administered a questionnaire. Next, we asked informants to indi- third dimension, we asked respondents to indicate which cate any other stakeholder they considered relevant to be surveyed stakeholders they considered key to participate in the planning considering their importance in terms of management of natural and implementation of MU-CMPA and to indicate the perception resources at the study area. This allowed us to validate the initial on their influence and capacity in these activities (5-point scale sample and, on an iterative process (Ritchie et al., 2003), add new ranging from not at all to extremely influential or capable). In- stakeholders. The criterion for including new stakeholders in the fluence was defined as the power a stakeholder has to facilitate or sample was that they had to be mentioned by at least two other hamper the planning and implementation. Capacity was defined stakeholders. The final sample was 44 stakeholders (Table 1, full list as a measure of available resources by an organization or insti- available in Supplementary Material). tution (e.g., human, financial, knowledge, engagement systems) that potentially enables them to be part of planning and imple- 2.2. Questionnaire design and data collection mentation (Lockwood, 2010). The questionnaires were administered face to face during We designed a questionnaire to collect information from November 2011 and December 2012. From 44 relevant stake- stakeholders and to conduct the SA and SNA analyses. The first holders identified, 42 completed the questionnaire (95% response section of the questionnaire included background information of rate). In two cases, by request of fishermen’s leaders, the survey was each respondent and its organization. This information was ar- responded in presence of other fishermen from their organization. ranged in an attribute matrix and permitted us to categorize stakeholders in three ways: Organization type, administrative 2.3. Data analysis operation level, and participation on public-private management or conservation committees or boards. The second section had 2.3.1. Decision-making three dimensions: perceptions on decision-making process, social Ordinal data from stakeholders’ self-perceptions regarding networks, and stakeholders in MU-CMPA planning. In the first participation in decision-making were tabulated to show the dimension, using a Likert scale statements (5-point scale ranging percentage of responses to each question for each scenario from not at all important to extremely important) we asked re- (current and hypothetical) and each stakeholder group. Wilcoxon spondents to evaluate the participation of their organizations in Signed Rank test for dependent samples was performed to the decision-making of natural resources management under two compare both scenarios. Differences among stakeholders groups situations: i) in the current scenario, and ii) in the hypothetical were tested with a non-parametric KruskaleWallis test using scenario of implementing an MU-CMPA. In the second dimension, rank sums instead of means (Zar, 1996). Statistical analyses were we asked respondents to indicate stakeholders with whom they carried out with R environment version 2.15.2 (RCore have links in two fields: i) collaboration (hereinafter collabora- Development Team, 2012). tion network), and ii) exchange and/or transfer of scientificin- formation and knowledge (hereinafter knowledge network) in 2.3.2. Social networks measures terms of use and management of natural resources. For both The SNA for the two network configurations were focused on fields, we asked respondents to evaluate the importance level of individual level in complete networks, subgroups level and collaboration and knowledge exchange (7-point importance scale network level (Prell, 2012).

Table 1 Surveyed stakeholders, stakeholders named in social networks, and key stakeholders in MU-CMPA planning and implementation. Classiﬁcation was made by type of organization, administrative operation level and belonging to public-private committees associated to management or conservation of natural resources. For MU-CMPA planning and implementation, stakeholders were named in general way without specifying a particular organization (e.g., ﬁshermen organizations) or administrative operation level.

Abbreviation Surveyed Collaboration network Knowledge network MU-CMPA planning

Type of Organization Civil Society Organization CSO 1 3 1 2 Diving Organization DOR 1 1 1 1 Local elementary school EDUC e 11e Artisanal Fishermen Organization AFOR 10 13 13 1 Land Owners LAND_OWN 2 2 2 1 Local Government (Municipality) LOG 2 2 2 1 National Gov. Funding Agencies NAG_FUND e 821 National Government Agencies NAG 12 22 13 11 Environmental NGO NGO 2 9 7 2 National Police POLICE e 1 e 1 Public-Private Committee PPC 1 4 4 2 Regional Government Agencies REG 3 4 4 3 University or Research Institute RES 4 6 7 2 Technical Consultant TCO 1 3 1 1 Tourism enterprise TOUR 3 4 3 1 Industrial Fishermen Organization IFOR ee 11 Administrative Operation Level Atacama Region 9 18 12 Coquimbo Region 28 46 38 National or more than one region 5 19 12 Public-Private Committee Yes 18 19 19 No 24 64 43 Total 42 83 62 31 P.F. Cárcamo et al. / Ocean & Coastal Management 91 (2014) 5e16 9

To assess how well the networks were connected and how the 3. Results power and exchange was distributed among network actors we calculated the following network cohesion measures: i) density 3.1. Participation in decision-making (the proportion of all possible links present in a network), ii) diameter (the longest number of steps between any two actors), iii) Statistically significant differences were detected between the average path length (average number of steps between any two decision-making responses in both scenarios (Wilcoxon Signed actors), iv) in/out degree centralization (the extent to which one Rank test, Z ¼ 3.2, P < 0.01). In the current scenario, 88% of re- actor is holding all the links in the network), and v) betweenness spondents considered that participation of their organization in centralization (the variation in the number of times that actors in decision-making is moderate, very or extremely important. This the network lie on paths between other actors) (Hanneman and percentage increased to 95% in the hypothetical scenario of Riddle, 2005; Prell, 2012; Vance-Borland and Holley, 2011). To implementing an MU-CMPA. 90% of the respondents considered test the degree of association between collaboration and knowl- that their participation would remain unchanged or increase in edge relations (i.e., the collaboration relations existing between terms of importance in the hypothetical MU-CMPA scenario, and two actors correlate significantly with exchange of scientific in- only three stakeholders considered that their participation would formation between them) we used a correlation test (UCINET decrease. For both scenarios, we found no statistical differences in Quadratic Assignment Procedure Correlation, QAP) to develop participation in decision-making for the variables “type of organi- standard errors to test for the significance of association zation or stakeholder” (Fig. 2) and “administrative operation level” (Hanneman and Riddle, 2005). The same test was used in both (KruskaleWallis test, P > 0.01). When analyzed “belonging to networks to evaluate whether the importance level assigned to public-private committee” as a grouping variable, in the current relations was positively associated with the annual frequency of scenario, the stakeholders involved in committees had a greater contacts. participation than those not involved (KruskaleWallis test, Using type of organization and administrative operation level as H ¼ 10.7, P < 0.01). However, in the hypothetical MU-CMPA sce- grouping attributes, we calculated the E-I index (UCINET E-I Index nario, we found no statistical differences between both groups function) that simply captures how much tendency to homophily (KruskaleWallis test, H ¼ 2.25, P > 0.01). there is in a network (i.e., tendency of an actor to interact with others who are similar) (Hanneman and Riddle, 2005). This index 3.2. Collaboration and knowledge network analyses takes the number of links of group members to outsiders, subtracts the number of links to other group members, and divides it by the The collaboration network had 83 stakeholders and the total number of links. This index ranges from 1 (all links are in- knowledge network 62 stakeholders, being National Government ternal to the group) to þ1 (all links are external to the group). With Agencies, Artisanal Fishermen Organizations, and Environmental a 5 000-iteration permutation test we assessed whether the NGOs, the most represented (Table 1). The number of links was observed index value was significantly higher or lower than greater for the collaboration network than for the knowledge expected. network. Stakeholders in the knowledge network matched in 92% To identify single stakeholders that were potentially more with stakeholders in the collaboration network. Diameter, average strategic, influential and active than others, we calculated actor path length, and density showed similar values for both networks betweenness centrality and brokerage scores for each stakeholder. (Table 2). In-degree centralization showed similar values between Betweenness centrality measures the importance of an actor by its networks and out-degree and betweenness centralization were proportion of paths between other actors (Hanneman and Riddle, greater for the knowledge network (Table 2) indicating that the 2005). An actor connecting more actors than others indicate his knowledge flow was centralized in fewer stakeholders than in the potential control over the relations or information flow in the collaboration network. The Quadratic Assignment Procedure Cor- network (Hanneman and Riddle, 2005; Prell, 2012). Brokerage relation (QAP) correlation was significant between both networks scores can be used as a measure of bridging social capital (Vance- (Pearson correlation, r ¼ 0.522, P < 0.01), suggesting a positive Borland and Holley, 2011). To calculate brokerage scores we applied an analysis for brokerage according to Gould and Fernandez (1989), focusing on between-groups broker roles (UCINET G&F Brokerage Roles function). Using the administrative operation level as grouping attribute, this procedure allowed us, for each actor, to find the number of times they connected two actors belonging to different administrative operation levels, who would not otherwise be connected. All network analyses were performed using UCINET 6.0 software (Borgatti et al., 2002). Graph visualizations were performed using NETDRAW 2.0 (Borgatti, 2002).

2.3.3. SA according to their influence and capacity A code was assigned to each actor. We then calculated its relative frequency and the score assigned for capacity and influence. For categorizing stakeholders and identifying key stakeholders we constructed an influence-capacity matrix using the mean score achieved by each stakeholder. We used this matrix as an alternative Fig. 2. Stakeholders’ perceptions on the importance of their participation in the to commonly used interest-influence matrix, where stakeholders decision-making process for natural resources management at the study area. Values represent mean standard error in 5-point Likert scale by type of organization or are mapped according to their relative interest and influence and stakeholder (ranging from “1, not at all important” to ’’5, extremely important”), black can be categorized as supportive or unsupportive (Reed et al., bars indicate the current scenario; white bars indicate the hypothetical scenario of 2009). implementing an MU-CMPA. 10 P.F. Cárcamo et al. / Ocean & Coastal Management 91 (2014) 5e16 relation between the matrices that is unlikely to have occurred by Table 3 chance, which indicates that the existence of collaboration re- Whole networks E-I index, using type of organization and administrative operation level as stakeholder’s grouping attributes. lations between two stakeholders enhances the probability of knowledge exchange between them. Also we found a strong cor- Observed values Expected value* relation between importance level of relations and their annual Internal External E-I index E-I index P observed frequency (r ¼ 0.973, P < 0.01 in the collaboration network; Collaboration network ¼ < r 0.968, P 0.01 in the knowledge network) indicating that the Type of 0.168 0.832 0.665 0.757 0.031 valuation of the relation importance between two stakeholders was organization positively associated with the relation frequency. Administrative 0.653 0.347 0.305 0.201 0.000 The E-I index results are shown in Table 3. When groups were operation level fi Knowledge network de ned by type of organization, the collaboration and knowledge Type of 0.153 0.847 0.693 0.773 0.103 networks showed similar observed values (0.665 and 0.693, organization respectively), indicating that the proportion of external links (be- Administrative 0.642 0.358 0.284 0.139 0.000 tween groups) was higher than the proportion of internal links operation level (within groups). However, observed values were not statistically *A 5 000-iteration permutation test was used to assess whether the observed index different from expected values (P ¼ 0.031 and 0.103, respectively). value was significantly higher or lower than expected. When groups were defined by administrative operation level, again the collaboration network and knowledge network showed similar observed values (0.305 and 0.284, respectively), indicating that Both networks, matched only in a 50% when we compared stake- the proportion of internal links was higher than the proportion of holders holding top 10 betweenness values. external links. Observed values were significantly different to ex- The stakeholders holding the best brokerage values using pected values (P < 0.01 for both networks), indicating a strong administrative operation level as a grouping attribute are shown in within-groups cohesion. Table 5. In the collaboration network, the highest scores (i.e., Stakeholders holding best betweenness centrality values are stakeholders that more times connected two actors from different shown in Table 4. In the collaboration network, higher scores (i.e., administrative operation levels) matched in 60% with stakeholders stakeholders that more times lay between two others who are holding top 10 betweenness values. The Fisheries Zonal Council themselves disconnected) were achieved by National Government achieved the first place. Compared to the betweenness analysis, we Agencies, corresponding to the regional offices of SERNAPESCA, could notice the incorporation of the Fisheries Development followed by CONAF (both regional offices), one Artisanal Fishermen Institute, a private Technical Consultant, and an International Organization located at La Higuera commune (Coquimbo Region), Environmental NGO, and the absence of SERNAPESCA Coquimbo and one Artisanal Fishermen Organization located at Freirina regional office. In the knowledge network, only six stakeholders commune (Atacama Region). In the lower part of the table we achieved brokerage scores, being one a private Technical Consul- found one Regional Government Agency, the Fisheries Zonal tant, three Universities or Research Institutes, one Artisanal Fish- Council that operates in both administrative regions, one Tourism ermen Organization of Freirina commune, and the CONAF Atacama Enterprise that operates at La Higuera commune, one Artisanal regional office. In both networks, National Government Agencies Fishermen Federation (grouping several fishermen organizations), were not the main bridging stakeholders (Table 5). and one Regional University located at the Coquimbo Region (Fig. 3). 3.3. Influence-capacity matrix In contrast to the collaboration network, in the knowledge network the highest scores were achieved by SERNAPESCA Respondents indicated 31 stakeholders (regardless the opera- Coquimbo regional office, a private Technical Consultant associated tion region) as key to participate in planning and implementation to fisheries management, the Instituto de Fomento Pesquero (IFOP, of the MU-CMPA, the most cited being those belonging to eleven Fisheries Development Institute, a national institute dedicated to national government agencies (Table 1). Artisanal Fishermen Or- research in fisheries and aquaculture), and one Regional University ganizations and SERNAPESCA were mentioned by 86% of the sur- with emphasis in marine research, located at the Coquimbo Region veyed stakeholders. 50e60% of the stakeholders mentioned the (Table 4, Fig. 4). Showing a lower score appeared CONAF Coquimbo national government agencies of the Dirección General de Terri- regional office, SERNAPESCA Atacama regional office, and another torio Marítimo y Marina Mercante (DIRECTEMAR, General Direc- Regional University located in the Coquimbo Region but with tion of Maritime Territory and Merchant Navy), the Environment emphasis in terrestrial research. In the lower part of the table we Ministry and CONAF, the Regional Governments, Municipalities, found a regional Environmental NGO that mainly works at La and Universities. When we placed stakeholders in the matrix ac- Higuera commune, the Coquimbo Regional Undersecretary of cording to the mean score obtained, and when we incorporated the Environment depending of Environment Ministry, and one Arti- times that each actor was mentioned by another as a third score, sanal Fishermen Organization located at La Higuera commune. these stakeholders were perceived as the most relevant in terms of influence and capacity in the area to do the planning and implementation of the MU-CMPA. The exceptions were Municipalities Table 2 Measures of cohesion for whole networks. that obtained a lower score on capacity (Fig. 5). Of the total stakeholders, 65% achieved high scores of influence and capacity. Measure Collaboration network Knowledge network Network size 83 62 3.4. Comparison of stakeholders selected by different analysis Number of ties 410 207 Diameter 8 8 Average path length 5.39 5.41 When comparing the top 10 most named stakeholders as key in Density (%) 6 5.5 the planning and implementation of the MU-CMPA with the top 10 Indegree centralization (%) 21.32 18.06 highest betweenness scores stakeholders, we found that matching Outdegree centralization (%) 23.82 31.61 was higher with the collaboration network than with the knowl- Betweenness centralization (%) 6.80 12.60 edge network (60 and 50%, respectively, Table 4, Fig. 5). However, P.F. Cárcamo et al. / Ocean & Coastal Management 91 (2014) 5e16 11

Table 4 Normalized betweenness centrality scores by individual level. List of top 10 betweenness scores stakeholders. See Table 1 for stakeholder’s names and abbreviations. In organization name, Romans numerals indicate administrative operation level, III and IV correspond Atacama Region and Coquimbo Region respectively; Arabic numerals indicate a speciﬁc type of institution or organization.

ID Organization name Collaboration ID Organization name Knowledge network network

NAG_IV_2 SERNAPESCA IV 7.611 NAG_IV_2 SERNAPESCA IV 13.860 NAG_III_2 SERNAPESCA III 6.146 TCO_1 Technical Consultant 1 13.777 AFOR_4 Artisanal Fishermen Organization 4 5.585 RES_1 IFOP 9.102 NAG_IV_3 CONAF IV 5.350 RES_5 Regional University 5 6.775 AFOR_1 Artisanal Fishermen Organization 1 5.118 NAG_IV_3 CONAF IV 5.918 NAG_III_3 CONAF III 4.527 NAG_III_2 SERNAPESCA III 5.847 REG_III_IV Fisheries Zonal Council 4.367 RES_3 Regional University 3 4.977 TOUR_IV_3 Tourism Enterprises IV 3.558 NGO_1 Regional Environmental NGO 1 4.035 AFOR_10 Artisanal Fishermen Federation 10 2.761 NAG_IV_1 Undersecretary of Environment IV 4.007 RES_5 Regional University 5 2.495 AFOR_4 Artisanal Fishermen Organization 4 3.806 organizations such as Municipalities, Regional Governments, Civil perception can be found in the organization and collaborative work Society Organizations and DIRECTEMAR which were highly cited developed in recent years by some Chilean ﬁshermen organization for MU-CMPA planning did not appear among the best between- around management areas for benthic resources (Gelcich et al., ness and brokerage scores. 2005; Schumann, 2007). The hypothetical declaration of an MU- CMPA in the study area generates a wide interest to participate in 4. Discussion the decision-making process, which could be seen as a positive valuation of the participation in the MPA and a recognition of the ﬁ 4.1. Perceptions on participation in decision-making potential bene ts derived from this (Cárcamo et al., 2014).

The self-perception regarding participation in decision-making 4.2. Collaboration and knowledge network analyses was high, even in groups that have historically been negatively discriminated in terms of citizen participation, such as artisanal In our study, SNA as a diagnostic tool provided two main con- ﬁshermen and civil society organizations. An explanation for this tributions: 1) the structure and properties of collaboration and

Fig. 3. Collaboration network representing collaboration relations among stakeholders, for natural resources management at the study area. Graph based on multidimensional scaling (MDS) ordination, relations are represented by lines (not considering relation directionality), nodes were colored depending on the administrative operation level of each stakeholder (white circle: Atacama Region, grey circle: Coquimbo Region, and black circle: National or more than one region), size of each node positively correlates with the betweenness centrality value for each stakeholder (larger node indicating higher betweenness). Nodes were labeled according to organization type. See Table 1 and Supplementary Material for stakeholder’s names and abbreviations. Arabic numerals indicate a speciﬁc type of institution or organization. A same Arabic numeral and different Romans numerals indicate the same institution but in different administrative regions. 12 P.F. Cárcamo et al. / Ocean & Coastal Management 91 (2014) 5e16

Fig. 4. Knowledge network representing relations of exchange and/or transfer of scientific information and knowledge for natural resources management among stakeholders at the study area. Graph based on multidimensional scaling (MDS) ordination, relations are represented by lines (not considering relation directionality), nodes were colored depending on the administrative operation level of each stakeholder (white circle: Atacama Region, grey circle: Coquimbo Region, and black circle: National or more than one region), size of each node positively correlates with the betweenness centrality value for each stakeholder (larger node indicating higher betweenness). Nodes were labeled according organization type. See Table 1 and Supplementary Material for stakeholder’s names and abbreviations. Arabic numerals indicate a specific type of institution or organization. A same Arabic numeral and different Romans numerals indicate the same institution but in different administrative regions. knowledge networks in the area, and 2) the extent to which these more advantageous for long-term planning and solving complex networks matched with the spatial scale of the natural ecosystem tasks (Janssen et al., 2006; Prell et al., 2009). Additionally, in to intervene. As whole networks, both showed little cohesion, with decentralized networks the participation can be perceived as more low values of centralization and density, indicating the existence of fair and open, and they are more resilient when actors disappear or a low flow of collaborative and exchange relations among different are removed from the network (Janssen et al., 2006). stakeholders. Some disadvantages resulting from low density are The asymmetry in the number of stakeholders per region (i.e., higher transaction costs associated to exchange and decision- greater at the Coquimbo Region) can be explained by the greater making capabilities, low capacity for collective action, and limited number of management and conservation measures and of in- resilience of the system (Bodin and Crona, 2009; Janssen et al., stitutions dedicated to management and research (e.g. regional 2006; Weiss et al., 2012). A decentralized network can be ineffi- universities) existing in this part of the study area. While both cient in coordination to solve simple tasks, forming groups, and networks share some similar cohesive properties (e.g., density, building support for collective action in initial phases, however, it is diameter, in-degree centralization), the number of nodes of the

Table 5 Un-normalized Brokerage scoresa by individual level in collaboration and knowledge networks, using administrative operation level as stakeholder’s grouping attribute. List of top 10 brokerage scores stakeholders. See Table 1 for stakeholder’s names and abbreviations. In organization name, Romans numerals indicate administrative operation level, III and IV correspond Atacama Region and Coquimbo Region respectively; Arabic numerals indicate a speciﬁc type of institution or organization.

ID Organization name Collaboration ID Organization name Knowledge network network

REG_III_IV Fisheries Zonal Council 43 TCO_1 Technical Consultant 1 32 RES_1 IFOP 24 RES_1 IFOP 22 TCO_1 Technical Consultant 1 16 AFOR_1 Artisanal Fishermen Organization 1 4 AFOR_1 Artisanal Fishermen Organization 1 12 RES_3 Regional University 3 2 RES_5 Regional University 5 9 RES_5 Regional University 5 2 NGO_2 International Environmental NGO 2 6 NAG_III_3 CONAF III 1 AFOR_4 Artisanal Fishermen Organization 4 5 NAG_III_2 SERNAPESCA III 5 NAG_III_3 CONAF III 4 PPC_1 National Reserve Advisory Council 3

a Based on binary and directed data. P.F. Cárcamo et al. / Ocean & Coastal Management 91 (2014) 5e16 13

Fig. 5. Inﬂuence-capacity matrix. Stakeholders were indicated as key to participate in the MU-CMPA planning and implementation. Positions corresponded to the mean score (5- point scale ranging from very low to very high) assigned by surveyed stakeholders to each stakeholder for capacity level to perform the planning and inﬂuence level in the area and on other stakeholders. Terms size is proportional to the number of times that each stakeholder was named by another stakeholder. NFS: SERNAPESCA (National Fisheries Service), AFOR: Artisanal Fishermen Organizations, DMT: DIRECTEMAR (General Direction of Maritime Territory and Merchant Navy), MUN: Municipalities, MIN-E: Environment Ministry, NFC: CONAF (National Forestry Service), UN: Universities, RG: Regional Government, TE: Tourism Enterprises, LCOM: Local communities, FZC: Fisheries Zonal Council, R_NGO: Regional NGOs, L_OWN, Land Owners, NA: Neighborhood Associations, NTS: SERNATUR (National Tourism Service), MIN_NA: Ministry of National Assets, ALS: SAG (Agriculture and Livestock Service), DPW: Direction of Port Works, RCCB: Regional Commission of Coastal Border, DC: Diving Centers, FDI: IFOP (Fisheries Development Institute), TCO: Technical Consultants, MIN_H: Housing Ministry, I_NGO: International NGOs, MPA_B: Marine Protected Areas Boards, PA_B: Terrestrial Protected Areas Boards.

collaboration network is greater than those of the knowledge power of SERNAPESCA in collaborative networks in marine envi- network, probably explained by the nature of the production, ronments is also recognized in other regions of Chile (Marín and dissemination and exchange of knowledge, which is generally more Berkes, 2010). Government agencies that work with natural re- restricted and requires a higher level of expertise and technical sources are not direct producers of scientific knowledge, but they skills (Weiss et al., 2012). The latter can also be corroborated by the achieve relevant positions in the knowledge network, indicating a greater level of centralization in the knowledge network, which potential role as bridging actors connecting producers and con- indicates that the knowledge exchange is brokered by fewer sumers of knowledge. The local governments (i.e., municipalities) stakeholders. were not positioned as key or relevant in either network, revealing The high presence of fishermen organizations in the collabora- their secondary role in decision-making and in the resources flow tion network and their high scores of betweenness centrality in the area. In the last decades, Chile has experienced various indicated the social and economic importance of the fishing and regionalization and decentralization reforms (Mardones, 2006), but management areas for benthic resources managed by them. In the possibilities for intervention in territorial planning and in the knowledge network, they decreased their presence in the high coastal border zoning are scarce for communes (i.e., local govern- scores, and in contrast universities, research institutes and tech- ment), and there is still a strong dependence on national or regional nical consultants (private institutions which advise technically and governments (Fløysand et al., 2010). scientifically to fishermen and have played a relevant role in Chil- When we used type of organization as stakeholder’s grouping ean co-management regime for shellfish (Schumann, 2010)) attribute (i.e., E-I index), there were no disconnected groups. In become important. contrast, when we used the administrative operation level as SERNAPESCA and CONAF showed key positions according to stakeholder’s grouping attribute, we found a high within-group betweenness centrality and in both networks and administrative cohesion, meaning that the interaction and flow was very low be- regions (Atacama and Coquimbo). Although their original function tween different regions stakeholders (even between counterpart was more associated with the enforcement of fishing and forestry government agencies), condition that was expected given the cur- laws, they are presently national agencies responsible of managing rent Chilean political-administrative division. From the perspective marine reserves and terrestrial protected areas, respectively. The of the governance of social-ecological systems, we could point out 14 P.F. Cárcamo et al. / Ocean & Coastal Management 91 (2014) 5e16 that there is a spatial and operational mismatch between in- consolidating the exchange circuit, from the acquisition of new stitutions or management regimes and natural ecosystems (Paavola information and scientific (e.g., university) and traditional (e.g. et al., 2009), where current governance systems represented in part fishermen) knowledge to the implementation (e.g., fishermen, by some government agencies and their regional offices do not enforcement agencies, decision makers), including the feedback match with the extent of the coastal ecosystem characterized by processes. This is very relevant if we considered that one of the the islands system. Another finding worth mentioning is that main elements of EBM is science-based management (McLeod and certain types of organizations, such as fishermen organizations and Leslie, 2009). The existing knowledge network and the possibility government agencies achieved low levels of within-group cohe- of improving it will have a strong influence in the management sion. Also, Marín and Berkes (2010) found little horizontal exchange outcomes. Weiss et al. (2012) pointed out that bridging organiza- and cooperation among fishermen organizations of central Chile. In tions can be most effective if they exist at the appropriate admin- the case of sectoral government agencies, the traditional focus on istrative scale and if their role as facilitators is stated in their single resources has created institutional structures with com- institutional mandate. partmentalized decision-making processes, leading to instruments Stakeholders’ perceptions regarding who should be involved in and policies that undermine sustainability. An example of this is MU-CMPA planning and implementation, pointed to the need of indicated by Andrade et al. (2008): the Chilean territorial planning incorporating stakeholders such as Municipalities, regional gov- is not integrated but compartmentalized, as dictated by sectoral ernments and local civil organizations (e.g., Neighborhood Associ- agencies which do not necessarily work in a coordinated way. As ations); however, those stakeholders are currently in peripheral pointed by Crowder et al. (2006) laws and government agencies on positions in both networks. Networks are not static and their multiple levels and sectors are generally not designed to solve properties can change; indeed, stakeholders and their relations can inter-sectoral conflicts, thus they do not adequately support the sometimes be more or less active. Maintaining the activation implementation of strategies such as EBM. capability is an important property that can contribute to the system resilience and to face up crisis (Janssen et al., 2006). 4.3. What kind of network do we need to manage an MU-CMPA? Theoretically, institutions such as the government can help or Selecting stakeholders for their implementation promote the construction or activation of networks with favorable structural properties to the management, but there are stories of An area proposed as MU-CMPA will require a governance and relations and conflicts that should not be ignored. Bodin and Crona management approach that copes with new economic challenges (2009) give some tips to achieve this: provide arenas for interac- (e.g., ports, mining, touristic development) and environmental tion, develop participatory processes, and provide funding for co- threats (e.g., species mortality, pollution, climate change), but also ordinators or facilitators. with traditional activities such as artisanal fishing (Cárcamo et al., Implementing an MPA under EBM approach requires early and 2011; Cárcamo and Gaymer, 2013). The search for a better effective participation of different stakeholders (e.g., to define approach to ensure sustainability has to acknowledge the MU- vision, targets and goals). However, we first need to identify who CMPA as a complex dynamic system and to address the existing the stakeholders are and next, which and when stakeholders spatial and operational mismatch between the potential gover- should participate. This study identified potential key stakeholders nance system and the natural ecosystem. Governance approaches for the planning and implementing processes and also found fea- such as adaptive management and adaptive co-management are tures and properties of networks which, if not addressed, could suggested for the implementation of EBM (Arkema et al., 2006; hinder the governance and management actions of the future MU- Berkes, 2012; Österblom et al., 2010). Co-management has CMPA (e.g., disconnected network between administrative re- evolved from dyadic conception (e.g., state and local community) to gions). This study is a first attempt to explore social networks in the a complex network of actors from different organizations and hi- area and offers clues on key stakeholders that should be present at erarchical levels connected through institutional arrangements, the starting of a formal process of MU-CMPA planning. resources dependence and information exchange (Bodin and Prell, In Chile, MU-CMPAs have been established under a combination 2011; Carlsson and Berkes, 2005). For this purpose, when thinking of legal attributions of different ministries and its administration is in the governance and management of an area comprising different intended to be public-private, involving regional governments and communes and administrative regions is crucial to improve the NGOs (Sierralta et al., 2011). However, it is important to note that, flow and exchange of knowledge and collaboration among stake- unlike the Chilean marine parks and marine reserves that have a holders of different administrative operation levels, between regulatory legal status supported by the Chilean Fisheries Law, groups and within-groups. In this context, the role of bridging ac- Chilean MU-CMPAs are not regulated by a legislative body, and tors (e.g., Fisheries Zonal Council, IFOP, technical consultants) in therefore, the responsibilities and roles of agencies for planning, linking disconnected segments will be crucial. Bodin and Crona managing, administrating, and/or financing, among others, are not (2009) indicated that the positive effect of bridging links in natu- completely clear (Praus et al., 2011). This regulatory gap offers an ral resource governance extends beyond the provision and ex- opportunity for early identification and categorization of stake- change of resources of various kinds, fostering trust among holders in the primary stages of the planning process, in addition to previously unconnected groups and facilitating collective actions. suggesting new institutional and governance arrangements to meet In our study, stakeholders such as some National Government the conservation and management goals and objectives defined by Agencies (e.g., SERNAPESCA and CONAF) were important in both the stakeholders for this area. networks, indicating that they can connect the two networks (i.e., collaboration and knowledge), probably reducing costs of collabo- 4.4. Using SA in the real world ration and provision of information and knowledge between groups. In the other hand, more central and powerful stakeholders Our research used three approaches to stakeholder analysis: i) did not necessarily act as bridging actors between the different Self-perception of stakeholders on their importance in current and administrative scales. Indeed, bridging stakeholders were generally future decision-making; ii) stakeholders’ perceptions on their not National Government Agencies, indicating the important role of peer’s capacity and influence to develop a planning and imple- private and non-governmental organizations and regional mentation process; and iii) measurement of stakeholders’ impor- agencies. These latter stakeholders can play a crucial role in tance derived from their positions and roles in networks. SNA goes P.F. Cárcamo et al. / Ocean & Coastal Management 91 (2014) 5e16 15 beyond an academic exercise and has been suggested to assist in Crowder, L.B., Osherenko, G., Young, O.R., Airamé, S., Norse, E.A., Baron, N., Day, J.C., the decision-making process for natural resources (Bodin and Prell, Douvere, F., Ehler, C.N., Halpern, B.S., Langdon, S.J., McLeod, K.L., Ogden, J.C., Peach, R.E., Rosenberg, A.A., Wilson, J.A., 2006. Resolving mismatches in U.S. 2011; Vance-Borland and Holley, 2011). Our SNA characterized the ocean governance. Science 313, 617e618. existing situation in the MU-CMPA area in terms of relations pat- Fløysand, A., Barton, J.R., Román, Á., 2010. La doble jerarquía del desarrollo eco- terns and we found strategic actors or groups as central and nómico y gobierno local en Chile: El caso de la salmonicultura y los municipios chilotes. EURE (Santiago) 36, 123e148. bridging stakeholders. Also, this characterization could be used to Freeman, L.C., 2004. The Development of Social Network Analysis: a Study of the assess the long-term evolution of networks. To categorize stake- Sociology of Science. Empirical Press, Vancouver. holders and investigate relations between them, we recommend to Gelcich, S., Edwards-Jones, G., Kaiser, M.J., 2005. Importance of attitudinal differ- fi fl ences among artisanal shers toward co-management and conservation of combine SNA with other tools of SA such as interest-in uence marine resources. Conserv. Biol. 19, 865e875. matrices, stakeholder-led stakeholder categorization, among Gould, R.V., Fernandez, R.M., 1989. Structures of mediation: a formal approach to others (Reed et al., 2009). This combination of tools can produce brokerage in transaction networks. Sociol. Methodol. 19, 89e126. fi Grafton, R.G., 2005. Social capital and fisheries governance. Ocean. Coast. Manag. signi cant and more implementable results in the context of in- 48, 753e766. terventions aimed to improving or implementing governance of Granek, E.F., Polasky, S., Kappel, C.V., Reed, D.J., Stoms, D.M., Koch, E.W., multiple-use areas under EBM principles. Kennedy, C.J., Cramer, L.A., Hacker, S.D., Barbier, E.B., Aswani, S., Ruckelshaus, M., Perillo, G.M.E., Silliman, B.R., Muthiga, N., Bael, D., Wolanski, E., 2010. Ecosystem services as a common language for coastal ecosystem-based management. e Acknowledgements Conserv. Biol. 24, 207 216. Hanneman, R.A., Riddle, M., 2005. Introduction to Social Network Methods. Uni- versity of California, Riverside, CA published in digital form at. http://faculty.ucr. We thank fishermen leaders and all organizations and agencies edu/whanneman/. for agreeing to be surveyed. We also thank R.H. Sampieri, J. Porras, F. Hughes, T.P., Bellwood, D.R., Folke, C., Steneck, R.S., Wilson, J., 2005. New paradigms for supporting the resilience of marine ecosystems. Trends Ecol. Evol. Pers. Ed.) Squeo, E. Sala, M. Thiel and R. Hucke-Gaete for helpful suggestions 20, 380e386. on an early version of this paper. This work was part of the partial Janssen, M.A., Bodin, Ö., Anderies, J.M., Elmqvist, T., Ernstson, H., McAllister, R.R.J., requirements for FC to obtain a PhD in Applied Biology and Ecology. Olsson, P., Ryan, P., 2006. Toward a network perspective on the resilience of socialeecological systems. Ecol. Soc. 11 (1), 15 [online] URL: http://www. FC was supported by a CONICYT doctoral scholarship (Comisión ecologyandsociety.org/vol11/iss1/art15/. Nacional de Investigación Científica y Tecnológica). Additional Jones, P.J.S., Burgess, J., 2005. Building partnership capacity for the collaborative support was obtained from CEAZA (Centro de Estudios Avanzados management of marine protected areas in the UK: a preliminary analysis. e en Zonas Áridas) and IEB (Instituto de Ecología y Biodiversidad). J. Environ. Manag. 77, 227 243. Kidd, S., Plater, A., Frid, C., 2011. The Ecosystem Approach to Marine Planning and Management. Earthscan, Washington, D.C. Lockwood, M., 2010. Good governance for terrestrial protected areas: a framework, Appendix A. Supplementary data principles and performance outcomes. J. Environ. Manag. 91, 754e766. Luna-Jorquera, G., Fernández, C., Rivadeneira, M., 2012. Determinants of the di- Supplementary data related to this article can be found at http:// versity of plants, birds and mammals of coastal islands of the Humboldt current systems: implications for conservation. Biodivers. Conserv. 21, 13e32. dx.doi.org/10.1016/j.ocecoaman.2014.01.007. Mardones, R., 2006. Descentralización y transición en Chile. Revista de ciencia política (Santiago) 26, 03e24. Marín, A., Berkes, F., 2010. Network approach for understanding small-scale fish- References eries governance: the case of the Chilean coastal co-management system. Mar. Policy 34, 851e858. Andrade, B., Arenas, F., Guijón, R., 2008. Revisión crítica del marco institucional y McLeod, K.L., Leslie, H.M., 2009. Ecosystem-based Management for the Oceans. legal chileno de ordenamiento territorial: el caso de la zona costera. Rev. geogr. Island Press, Washington, D.C. Norte Gd., 23e48. Österblom, H., Gårdmark, A., Bergström, L., Müller-Karulis, B., Folke, C., Arkema, K.K., Abramson, S.C., Dewsbury, B.M., 2006. Marine ecosystem-based Lindegren, M., Casini, M., Olsson, P., Diekmann, R., Blenckner, T., Humborg, C., management: from characterization to implementation. Front. Ecol. Environ. Möllmann, C., 2010. Making the ecosystem approach operational e Can regime 4, 525e532. shifts in ecological- and governance systems facilitate the transition? Mar. Berkes, F., 2012. Implementing ecosystem-based management: evolution or revo- Policy 34, 1290e1299. lution? Fish. Fish. 13, 465e476. Paavola, J., Gouldson, A., Kluvánková-Oravská, T., 2009. Interplay of actors, scales, Bodin, Ö., Crona, B.I., 2009. The role of social networks in natural resource gover- frameworks and regimes in the governance of biodiversity. Environ. Policy Gov. nance: what relational patterns make a difference? Glob. Environ. Change 19, 19, 148e158. 366e374. Pomeroy, R., Douvere, F., 2008. The engagement of stakeholders in the marine Bodin, Ö., Prell, C., 2011. Social Networks and Natural Resource Management: spatial planning process. Mar. Policy 32, 816e822. Uncovering the Social Fabric of Environmental Governance. Cambridge Uni- Praus, S., Palma, M., Domínguez, R., 2011. La Situación Jurídica de las Actuales Áreas versity Press, Cambridge, UK. Protegidas de Chile. Proyecto GEF-PNUD-MMA “Creación de un Sistema Borgatti, S.P., 2002. NetDraw: Graph Visualization Software. Analytic Technologies, Nacional Integral de Áreas Protegidas para Chile: Estructura Financiera y Harvard. Operacional". Fondo del Medioambiente, Santiago. Borgatti, S.P., Everett, M.G., Freeman, L.C., 2002. Ucinet for Windows: Software for Prell, C., 2012. Social Network Analysis: History, Theory & Methodology. SAGE Social Network Analysis. Analytic Technologies, Harvard, MA. Publications Ltd, London. Borgatti, S.P., Mehra, A., Brass, D.J., Labianca, G., 2009. Network analysis in the social Prell, C., Hubacek, K., Quinn, C., Reed, M., 2008. ‘Who’s in the network?’ when sciences. Science 323, 892e895. stakeholders influence data analysis. Syst. Pract. Action Res. 21, 443e458. Cárcamo, P.F., Cortés, M., Ortega, L., Squeo, F.A., Gaymer, C.F., 2011. Crónica de un Prell, C., Hubacek, K., Reed, M., 2009. Stakeholder analysis and social network conflicto anunciado: Tres centrales termoeléctricas a carbón en un hotspot de analysis in natural resource management. Soc. Nat. Resour. Int. J. 22, 501e518. biodiversidad de importancia mundial. Rev. Chil. Hist. Nat. 84, 171e180. Pretty, J., Ward, H., 2001. Social capital and the environment. World Dev. 29. Cárcamo, P.F., Garay-Flühman, R., Gaymer, C.F., 2013. Opportunities and constraints R Core Development Team, 2012. R: A Language and Environment for Statistical of the institutional framework for the implementation of an ecosystem-based Computing. R Foundation for Statistical Computing, Vienna, Austria. management: the case of the Chilean coasts. Ocean. Coast. Manag. 84, 193e203. Ramirez-Sanchez, S., Pinkerton, E., 2009. The impact of resource scarcity on Cárcamo, P.F., Garay-Flühman, R., Squeo, F.A., Gaymer, C.F., 2014. Using stakeholders’ bonding and bridging social capital: the case of fishers’ information-sharing perspective of ecosystem services and biodiversity features to plan a marine networks in Loreto, BCS, Mexico. Ecol. Soc. 14 (1), 22 [online] URL: http:// protected area. Environ. Sci. Policy in revision. www.ecologyandsociety.org/vol14/iss1/art22/. Cárcamo, P.F., Gaymer, C.F., 2013. Interactions between spatially explicit conserva- Reed, M.S., Graves, A., Dandy, N., Posthumus, H., Hubacek, K., Morris, J., Prell, C., tion and management measures: implications for the governance of Marine Quinn, C.H., Stringer, L.C., 2009. Who’s in and why? A typology of stakeholder Protected Areas. Environ. Manag. 52, 1355e1368. analysis methods for natural resource management. J. Environ. Manag. 90, Carlsson, L., Berkes, F., 2005. Co-management: concepts and methodological im- 1933e1949. plications. J. Environ. Manag. 75, 65e76. Ritchie, J., Lewis, J., Elam, G., 2003. Designing and selecting samples. In: Ritchie, J., Carollo, C., Reed, D.J., 2010. Ecosystem-based management institutional design: Lewis, J. (Eds.), Qualitative Research Practice. A Guide for Social Science Stu- balance between federal, state, and local governments within the Gulf of dents and Researchers. SAGE Publications Ltd., London, pp. 77e108. Mexico Alliance. Mar. Policy 34, 178e181. Rodríguez-Martínez, R.E., 2008. Community involvement in marine protected Cohen, P.J., Evans, L.S., Mills, M., 2012. Social networks supporting governance of areas: the case of Puerto Morelos reef, México. J. Environ. Manag. 88, coastal ecosystems in Solomon Islands. Conserv. Lett. 5, 376e386. 1151e1160. 16 P.F. Cárcamo et al. / Ocean & Coastal Management 91 (2014) 5e16

Schumann, S., 2007. Co-management and “consciousness”: Fishers’ assimilation of Manriquez, P.H., Marin, V., Munoz, P., Navarrete, S.A., Perez, E., Poulin, E., management principles in Chile. Mar. Policy 31, 101e111. Sellanes, J., Sepulveda, H.H., Stotz, W., Tala, F., Thomas, A., Vargas, C.A., Schumann, S., 2010. A tenuous triumvirate: the role of independent biologists in Vasquez, J.A., Vega, J.M.A., 2007. The Humboldt Current System of Northern and Chile’s co-management regime for shellﬁsh. Mar. Policy 34, 133e138. Central Chile. In: Oceanography and Marine Biology, vol. 45, pp. 195e344. Sierralta, L., Serrano, R., Rovira, J., Cortés, C., 2011. Las Áreas Protegidas de Chile. Vance-Borland, K., Holley, J., 2011. Conservation stakeholder network mapping, Antecedentes, Institucionalidad, Estadísticas y Desafíos. Ministerio del Medio analysis, and weaving. Conserv. Lett. 4, 278e288. Ambiente, Chile, Santiago. Weiss, K., Hamann, M., Kinney, M., Marsh, H., 2012. Knowledge exchange and policy Thiel, M., Macaya, E.C., Acuna, E., Arntz, W.E., Bastias, H., Brokordt, K., Camus, P.A., inﬂuence in a marine resource governance network. Glob. Environ. Change 22, Castilla, J.C., Castro, L.R., Cortes, M., Dumont, C.P., Escribano, R., Fernandez, M., 178e188. Gajardo, J.A., Gaymer, C.F., Gomez, I., Gonzalez, A.E., Gonzalez, H.E., Haye, P.A., Zar, J.H., 1996. Biostatistical Analysis, third ed. Prentice Hall, Saddle River, New Illanes, J.E., Iriarte, J.L., Lancellotti, D.A., Luna-Jorquerai, G., Luxoroi, C., Jersey.