Start Submission Become a Reviewer

Reading: Local Stakeholders Understand Recreational Fisheries as Social-Ecological Systems but Do Not...


A- A+
Alt. Display

Research articles

Local Stakeholders Understand Recreational Fisheries as Social-Ecological Systems but Do Not View Governance Systems as Influential for System Dynamics


Jacob P. Ziegler ,

McGill University, CA
X close

Stuart E. Jones,

University of Notre Dame, US
X close

Christopher T. Solomon

Cary Institute of Ecosystem Studies, US
X close


Recognition that there are often social and ecological components to problems that arise from management of shared resources has led to a dominant paradigm among academics that natural resource management should consider coupled social-ecological systems. For academic theory to have real-world impact it must be understood and acted upon by stakeholders at a local scale. However, it is unclear if stakeholders view their systems as coupled social-ecological systems. We interviewed key stakeholders in an inland recreational fishery to solicit their mental models of system dynamics in the context of Ostrom‘s Social-Ecological Systems Framework (SESF). We found that stakeholders in aggregate considered all components of the SESF (actors, resource systems, environmental settings, and governance systems) in their view of recreational fisheries. However, researchers viewed governance system and environmental setting components as less diverse than actor and resource system components, while anglers and managers viewed the actor component as more diverse than all other components. In addition, all stakeholders viewed governance system and environmental setting components as less influential than actor and resource system components. Given strong empirical evidence of positive relationships between the number and diversity of governance system attributes and successful fisheries outcomes, our results suggest that governance systems that prevent free riding, enforce rules through graduated sanctions, and address large scale problems at the local scale through nested institutions could improve social-ecological outcomes in inland recreational fisheries.

How to Cite: Ziegler, J. P., Jones, S. E., & Solomon, C. T. (2019). Local Stakeholders Understand Recreational Fisheries as Social-Ecological Systems but Do Not View Governance Systems as Influential for System Dynamics. International Journal of the Commons, 13(2), 1035–1048. DOI:
  Published on 30 Oct 2019
 Accepted on 24 Jul 2019            Submitted on 15 Nov 2018

1. Introduction

A social-ecological approach to natural resource management has become the dominant paradigm among academics. In a critical review of natural resource management goals Holt and Talbot (1978) suggested that the primary goal of management should be to maintain a resource system in a desirable state despite environmental and socio-political changes. However, during the 1980’s there was recognition that solutions to natural resource management problems were often composed of ecological, economic, and social components (Mangel et al. 1996, Mace 2014). Since the 1980’s a new form of natural resource management has emerged within academic circles, which focuses on understanding the many complex interactions among social and ecological systems (Kates et al. 2001, Mace 2014). These interactions among system components produce dynamics, like feedbacks and thresholds, that can only be explained by considering coupled social-ecological systems (Costanza et al. 1993). Considering social-ecological interactions has improved our understanding of real world outcomes. For example, considering institutional arrangements can help explain how “tragedy of the commons” type outcomes may be avoided in shared resource use (Berkes et al. 1989, Dietz et al. 2003).

An example of the shift in natural resource management paradigm can be seen in inland recreational fisheries, where research was initially focused on fish and their immediate environment, but has since moved towards a social-ecological perspective. Prior to the 1980’s inland recreational fisheries research was within the realm of ecological and physical sciences (Arlinghaus et al. 2008). However, since the 1980’s there have been increasing calls to correct for the lack of social research and embrace a social-ecological paradigm within recreational fisheries (Fulton et al. 2011, Beard et al. 2011, Arlinghaus et al. 2013, Arlinghaus et al. 2017). Researchers have begun to understand inland recreational fisheries as coupled social-ecological systems (Johnston et al. 2010, Hunt et al. 2013, Ziegler et al. 2017) but so far studies have largely focused on ecological dynamics and individual human behaviors and have not focused on understanding other components of social-ecological systems, like governance systems, and the outcomes of complex interactions among social-ecological components (Arlinghaus et al. 2013, Arlinghaus et al. 2017).

The complexity of interacting social and ecological systems poses a challenge for understanding and defining social-ecological systems (Ludwig et al. 1993, Ludwig 2001). However, empirical data and theory have suggested that social-ecological complexity can be understood and defined by a handful of controlling processes and variables (Ostrom 1990, Holling 2001, Walker et al. 2006). Ostrom (2009) formalized findings from extensive fieldwork on social-ecological systems into a Social Ecological Systems Framework (SESF) designed for understanding local social-ecological systems and for use in natural resource management. She identified variables for understanding social-ecological systems that she organized into system components. The components she defined included: (1) actor – defines the social entities that act within the system, (2) resource system – defines the ecological and biological context of the resource, (3) governance system – defines the rules and rights of actors, and (4) environment – defines related systems that affect the resource and actors. Later, McGinnis and Ostrom (2014) added the (5) Action Situation, which is composed of interactions among the SESF components that create outcomes. The SESF has become one of the most widely used frameworks for understanding complex social-ecological systems (Ostrom 2009 has been cited over 2,000 times, see also past and more recent versions of the SESF in Ostrom 2007, Ostrom 2009, McGinnis and Ostrom 2014, Hinkel et al. 2014).

While the SESF has been applied at local scales by researchers, including in inland recreational fisheries, it is unclear if stakeholders – those managing and affecting natural resources – understand their systems in a similar manner as the SESF. To date, studies applying the SESF at a local scale have focused on determining which of the variables proposed by Ostrom (2009) were present in a given system. These studies have ranged from no stakeholder involvement in applying the SESF by relying on researcher knowledge and literature reviews (Meinzen-Dick 2007, Basurto and Ostrom 2009, Santos and Thorne 2010, Madrigal et al. 2010, Bal et al. 2011, Blanco 2011, Gutiérrez et al. 2011, Amblard 2012, Schlüter and Madrigal 2012, Nagendra and Ostrom 2014) to including stakeholders in applying the SESF through questionnaires, interviews, and focus groups (Blanco and Fedreheim 2011, Dumyahn and Pijanowski 2011, Begossi et al. 2012, Cinner et al. 2012, Falk et al. 2012, Baur and Binder 2013, Risvoll et al. 2014, Naiga et al. 2015). However, studies applying the SESF rarely allow stakeholders to define their understanding of the system and which variables they view as important (although see Delgado-Serrano and Ramos 2015).

While a social-ecological paradigm in natural resource management has become dominant in academic circles, there is little real-world impact of academic theory if those carrying out and informing management at local scales do not share a similar understanding. Shared mental models among diverse stakeholders is often necessary for successfully implementing natural resource management (Biggs et al. 2011). But individual perceptions of social-ecological systems are often diverse (Otto-Banaszak et al. 2011), therefore local stakeholders may not share the same SESF mental model as those in academic circles. While there is strong empirical evidence that fisheries outcomes are often dependent on SESF components like governance systems and environmental and socio-political settings (Pollnac et al. 2010), past focus within academia on ecological dynamics and individual human behaviour, as observed in fisheries management (Salas and Gaertner 2004, Fulton et al. 2011, Murray and Ings 2015), may result in stakeholder understanding not capturing the full extent of social-ecological systems.

Our objective was to understand how stakeholders conceptualized a shared resource, specifically how important they perceived Ostrom’s SESF components to be relative to and interacting with each other. We characterized stakeholder understanding of an inland recreational fishery landscape using mental models that we then represented within the SESF. We compared the importance of SESF components in stakeholder (researchers, managers, and anglers) mental models and discuss how previous studies applying the SESF can help inform inland recreation fisheries management.

2. Methods

2.1. Representing stakeholder mental models within the SESF

Fuzzy cognitive maps (FCMs) are an effective method of representing individuals’ mental models of how a complex system operates (Özesmi and Özesmi 2004, Vasslides and Jensen 2016). They depict key system variables (things that take on specific values at different time points) and concepts (groupings of variables with similar attributes) along with their direct relationships to one another (Bernard and Bernard 2012, see Supplemental Information Fig. S1 for an example). When FCMs are conducted with local experts they can provide detailed depictions of local social-ecological systems (Vasslides and Jensen 2016).

We used FCMs to characterize stakeholder understanding of an inland recreational fishery landscape and coded FCM variables and concepts into Ostrom’s Social-Ecological Systems Framework. While the SESF was developed to classify concepts and variables to allow for generalizations among diverse social-ecological systems, its application is often difficult and inconsistent (Thiel et al. 2015). Hinkel et al. (2014) improved ease and consistency of application of the SESF at the local scale by formalizing methods for adding concepts into the SESF. While Hinkel et al. (2014) did not highlight the action situation, we consider it in our study as it is featured prominently in an updated version of the SESF (McGinnis and Ostrom 2014). Each variable and concept that arose in stakeholder FCMs was categorized as a feature of the actor, resource system, governance system, or environment, and the direction and strength of relationships between concepts was used to describe the action situation using the methods described in the following sections.

2.2. Study region and Fuzzy Cognitive Map collection

We focused on recreational fisheries in Vilas County Wisconsin USA, a 2,600 km2 sparsely populated area in Northern Wisconsin where recreational fisheries represent prominent coupled social-ecological systems (Liu et al. 2007). Vilas County’s landscape is 13% open water, which supports a tourism and hospitality industry that is a major component of the region’s economy (Peterson et al. 2003). In 2011 freshwater anglers in Wisconsin spent 1.4 billion US dollars (US Census Bureau 2011), with Vilas County being a popular sport fishing location. Historically, fishing was culturally important to Native American tribes, like the Lac du Flambeau band of the Lake Superior Chippewa in Vilas County, whose name comes from harvesting fish by torch light. In treaties in 1837 and 1842 the Lake Superior Chippewa tribes ceded their territory to the United States in exchange, in part, for fishing rights that they maintain today (Peterson et al. 2003). Currently, the county contains the Lac du Flambeau Reservation (Figure 1) and many lake front vacation and retirement homes particularly from nearby urban centres of Chicago, Milwaukee, and Minneapolis. Fisheries in Vilas County are co-managed by the Wisconsin Department of Natural Resources (WDNR) and Chippewa tribes of Wisconsin. The state requires all anglers other than tribal anglers to have a fishing licence, use only authorized gear, and respect bag limits, while tribal anglers are able to harvest fish using high efficiency methods up to a “safe harvest level” set by the WDNR.

Figure 1 

Vilas County is a lake-rich region of northern Wisconsin that supports recreational fishing opportunities, which are an important component of the region’s economy.

Based on our knowledge working within this study area we selected 15 individuals who had expert knowledge of the four components of the SESF (actor, resource system, governance system, and environment). We interviewed experts on the actor and governance system components of the SESF (n = 8), who were heads of lake organizations, state fisheries managers, and avid anglers in the county. We collectively refer to this group as managers & anglers. We interviewed experts on the resource system and environment components (n = 7), who were fisheries and aquatic ecology academic researchers in in the study region. We collectively refer to this group as researchers. We used a standard method of accumulation curves to ensure that we conducted enough sampling to thoroughly represent the concepts deemed important by stakeholders for recreational fisheries in our study region (Vasslides and Jensen 2016). We computed the accumulation curve of the number of new concepts with additional FCMs (added at random) using the specaccum() function in the vegan package in R (Oksanen et al. 2018).

We followed standard methods for creating FCMs through one-on-one, in-depth interviews with informants (Özesmi and Özesmi 2004, Vasslides and Jensen 2016). The same interviewer conducted all interviews during the summers of 2016 and 2017. Interviews began with an overview of the project and an explanation of how to draw a FCM using a simple unrelated example of traffic flow on a road (Supplemental Information Fig. S1). The interviewer then asked informants to list which variables, concepts, or things came to mind when thinking about recreational fisheries in the study region and the interviewer recorded this list. Informants then diagramed the relationships among the concepts listed by drawing each as a node and connecting the nodes with arrows to represent directional relationships between concepts. Informants scored the direction of each relationship (positive or negative) and its strength (high, medium, or low). Once all concepts and variables from the list were diagramed, informants were given the opportunity to review and revise their maps until they confirmed that it accurately depicted their understanding of the system. Informants were then asked to rank their involvement in fisheries management, fisheries research, and freshwater angling on a scale from 1 to 10. They were then asked if they lived in Vilas County, if they were a member of a lake organization, and if they lived on a lake. Interviews ranged from 45 min to 105 min, with a mean of 60 min. In accordance with federal regulations, this research was reviewed and approved by an Institutional Review Board and the interviewer received “Protecting Human Research Participants” certification from the National Institutes of Health (IRB no. 130-2016).

2.3. Coding fuzzy cognitive map concepts into the Social Ecological Systems Framework

We used the methods of Hinkel et al. (2014) to add concepts from FCMs into the SESF using attribution and subsumption relationships. Attribution and subsumption relationships are sometimes referred to as “has-a” and “is-a” relationships respectively. For example, a concept X has an attribution relationship with a variable Y if the sentence “X has a Y” is meaningful for all instances of X and Y (e.g. an angler has a preference for fishing locations). Conversely, a concept X has a subsumption relationship with Z if the sentence “all X’s are Z’s but not vice versa” is true (e.g. all anglers are lake users but not all lake users are anglers). We followed these methods when coding concepts from FCMs into the four nested components of the SESF that Hinkel et al. (2014) developed. We provide an overview of how all concepts present in FCMs were classified into the SESF using Hinkel et al.’s (2014) visual representation of the SESF with attribution and subsumption relationships (Figure 3). We also provide a list of the original terms used in FCMs and how they were classified (Supplemental Information Table S1).

2.4. Testing stakeholder perceived importance of SESF components

We tested stakeholder perceived importance of the SESF components in our study system using the percentage and frequency of concepts in FCMs that we categorized within the four components of the SESF. We used Analysis of Variance to test for differences in the mean percentage and frequency of concepts in FCMs categorized into the four SESF components (R Core Team 2017). We then tested for pairwise differences among means using Tukey Honest Significant Differences. Given our a priori selection of stakeholder groups to ensure we had even coverage of the SESF, we expected that researchers would have higher importance of resource system and environment components in their FCMs and managers and anglers would have higher importance of actors and governance systems. Therefore, we fit models that did and did not allow means to differ by stakeholder group (i.e. managers and anglers vs. researchers). To identify potential differences in perceptions among individuals we also tested models that allowed mean percentage and frequency of concepts to vary by self-reported involvement in fisheries management, research, angling, residency in the county, membership in a lake association, and ownership of a lakeside property. We present results from the best performing model as judged by small sample size corrected Akaike Information Criteria (AICc).

Although Hinkel et al.’s (2014) enhanced SESF did not focus on the action situation, FCMs have standard quantitative methods that allowed us to compare differences in the action situation among SESF components. The action situation of the SESF is composed of interactions among SESF components and the resulting outcomes (McGinnis and Ostrom 2014).

Interactions in the SESF are defined as process relationships where one concept influences another (Hinkel et al. 2014, Schlüter et al. 2014). Fuzzy cognitive maps explicitly describe these relationships with the size and direction of effects between concepts. Two standard metrics used to describe these relationships in FCMs are the outdegree and indegree of a given concept. An outdegree is the cumulative strength (absolute value) of the effects a concept has on others, while an indegree is the cumulative strength of the effects other concepts have on a given concept (Özesmi and Özesmi 2004, Vasslides and Jensen 2016).

Outcomes, the other component of the action situation, are the result of all the interactions within a social-ecological system (Hinkel et al. 2014). Outcomes in the context of FCMs can be quantitatively calculated by expressing FCM concepts and relationships among concepts in an adjacency matrix and iterating out the relationships to an equilibrium state from initial conditions (see Dickerson and Kosko 1994). We made these calculations following the methods of Vasslides and Jensen (2016), who used an initial value of one for each state variable and a logistic transformation to bound state variable output between zero and one before each iteration. We expressed outcomes as the percent change in state variables at equilibrium from initial conditions.

We tested for differences in interactions and outcomes among the SESF components using hierarchical linear models where indegree, outdegree, and percent changes from initial conditions were our response variables, we controlled for pseudo replication by nesting the response variables within informants. We tested models that allowed means to differ by interviewee type (i.e. managers and anglers vs. researchers) against models that did not. We also tested for effects of self-reported involvement in fisheries management, research, angling, residency in the county, membership in a lake association, and ownership of a lakeside property on indegree, outdegree and percent change from initial conditions. We present results from the best performing model as judged by AICc.

3. Results

The average number of new concepts among maps declined with additional informants (Figure 2). There were 60 distinct concepts among the 15 maps (Figure 2A); this included both concepts (bolded) and their attributes (non-bolded) in Figure 3 but did not include the four SESF components themselves (i.e. actor, resource system, governance system, and environment) as these were never explicitly included in FCMs. The rate at which new concepts and variables were added to our understanding of the social-ecological system with additional informants interviewed declined from ~10 to 1 by 15 informants, as most concepts were already represented in previous FCMs (Figure 2B).

Figure 2 

The average number (A) and rate (B) of new variables added to our understanding of the social-ecological system declined as the number of informants interviewed increased. The lines represent means from a rarefaction analysis that added informants at random.

Figure 3 

All concepts and variables present in stakeholder mental models coded into the Social Ecological Systems Framework (SESF) using attribution (closed arrows) or subsumption (open arrows) relationships (see Hinkel et al. 2014). See Table S1 for how concepts from FCMS were coded in the SESF.

In aggregate, stakeholder mental models captured the four components of the Social Ecological Systems Framework (Figure 3). Individually, just over two thirds of mental models captured all four SESF components but to varying degrees (, Figure 4). All concepts and variables present in FCMs fit into the four components of the SESF through either attribution or subsumption relationships (closed and open arrows, respectively, in Figure 3). The majority of nodes present in FCMs were attributes of a concept in the SESF (e.g. the ability of an angler, Figure 3) and not subsumption relationships.

Figure 4 

Stakeholders consistently included more actor concepts in their mental models compared to governance and environment concepts and researchers included more resource system concepts than managers and anglers. Mean values highlighted with asterisks indicates that they were significantly different from those without asterisks.

Both stakeholder groups emphasized actor concepts in their mental models, and researchers emphasized resource system concepts more than managers and anglers did (Figure 4). As expected from our a priori group choices, the best preforming model allowed means to vary by group type and indicated that managers and anglers emphasized actor concepts and did not identify many features of the resource system or the environment in their FCMs (p-values <0.001 for all pairwise comparisons, Figure 4), while researchers emphasized resource systems compared to managers and anglers (p-value <0.01, Figure 4). However, what differed from our expectations was that all stakeholders emphasized actor concepts over environment and governance system concepts. Further, researchers did not identify many features of the environment compared to the resource system (p-value <0.01, Figure 4). On average, researchers’ maps included 5 actor concepts (38% of all concepts in the average researcher map), 4 resource system concepts (35%), 3 governance system concepts (17%), and 2 environment concepts (10%). For managers and anglers, the average map included 7 actor concepts (49%), 3 resource system concepts (16%), 2 governance system concepts (17%), and 3 environment concepts (17%). The second-best model indicated that a unit increase in self-reported involvement in angling resulted in a 5% decrease in importance of resource system concepts in FCMs (interaction effect = –0.05 ± 0.03 95% confidence interval). This suggests that anglers brought the combined average of the importance of resource system concepts down for managers and anglers.

All stakeholders emphasized the influence of actors and resource systems in the social-ecological system compared to the environment and governance system (Figure 5A). On average, the cumulative effect of an actor or resource system concept on other concepts was significantly larger than the cumulative effect of a governance system or environment concept (p-values <0.05 for all pairwise comparisons, Figure 5A). The average effect of a resource system concept on other concepts was larger than that of actors but they were not significantly different from one another (Figure 5A). None of the informant self-reported metrics were present in top models nor were they significant in any of the candidate models.

Figure 5 

(A) Stakeholders consistently reported larger influences of actor and resource system variables on other variables (outdegree) compared to governance and environment variables. However, (B) The cumulative effects that other variables had on a given variable (indegree) did not significantly differ among the SESF components. Mean values highlighted with asterisks indicates that they were significantly different from those without asterisks.

Even though actor and resource concepts were viewed as more diverse and influential within FCMs they did not have disproportionate effects on any one SESF component, therefore, equilibrium outcomes were similar among SESF components. The cumulative effects that other concepts had on a given concept (indegree) did not significantly differ among the SESF components (Figure 5B). Consequently, the larger representation of actor and resource system concepts and their influence on other concepts did not lead to significant differences in equilibrium outcomes among SESF components (Figure 6). All equilibrium outcomes of mental models converged on a steady state within 25 iterations of the model.

Figure 6 

The larger representation of actor and resource system concepts and their influence on other concepts did not lead to significant differences in equilibrium outcomes among Social Ecological System Framework components. Average equilibrium outcomes were calculated as the average percent change of state variables within the four Social Ecological System Framework components at equilibrium from initial conditions. Equilibrium conditions were determined from adjacency matrices that iterated out the relationships described in fuzzy cognitive maps to an equilibrium state.

4. Discussion

Our results suggest that stakeholders do view inland recreational fisheries as social-ecological systems but that actor and resource system components of the SESF are emphasized more than governance system and environment components in stakeholder mental models. This may have been because there were fewer governance and environment concepts that informants viewed as important in our study region, or because fuzzy cognitive maps underrepresented government and environment concepts. We discuss these two possibilities below and their implications for inland recreational fisheries management.

4.1. Hypothesis 1: Governance and environment concepts were less important in our region

There is strong empirical evidence that the success of local fisheries management is positively related to the number of attributes of well-functioning governance systems. There has been a long history of research into the effects of governance systems on management of shared resources that have highlighted property rights (Horan et al. 2011), collective choice rights (Dietz et al. 2003, Ostrom 2007), co-management (Worm et al. 2009, Berkes 2009), and strong institutions as having large effects on the success of shared resource management (Hilborn et al. 1995). In a global meta-analysis of 130 locally managed fisheries Gutiérrez et al. (2011) found that the social-ecological success of a fishery was more positively correlated to attributes of well-functioning governance systems than actor attributes. Social-ecological success increased linearly above eight governance system attributes but when fisheries had eight or less attributes their social-ecological success was near zero; informants in our study identified 9 attributes of the governance system in aggregate. Leslie et al. (2015) scored the importance of governance systems in local fisheries based on attributes of rules within local management and access rights of anglers and found that fisheries with higher governance system scores had greater fish abundance. Whereas, Cinner et al. (2012) found no effect of the governance system on fish abundance but strong effects on fisheries livelihood and compliance outcomes.

Evidently, the number and diversity of governance system attributes are important for successful fisheries management yet stakeholders reported less diversity and influence of the governance system in our study region compared to actor and resource system components. Incorporating attributes of the governance system that are known to improve outcomes for shared resources could aid inland recreational fisheries management. One promising approach would be to incorporate a diversity of attributes of governance systems that are likely to lead to social-ecological success of shared resource management (Ostrom 1990, Anderies et al. 2004). Based on empirical studies of local institutions managing a shared resource, Ostrom (1990) hypothesized a set of design principles for governance systems that lead to successful management of a resource. Of the eight design principles proposed by Ostrom there were five that were not explicitly included in the mental models of stakeholders in our system.

Stakeholders did not include the first three of Ostrom’s design principles in their mental models. These three design principles help prevent free riding by: identifying who should receive benefits from and pay costs for a resource, distributing benefits in proportion to the costs people pay for the resource, and allowing for collective choice to set rules of resource use (Anderies et al. 2004). In the context of inland recreational fisheries these design principles could be incorporated by rewarding user groups, like lake associations, who invest in stewardship of fish populations by granting them higher catch quotas to match the investments they make. While community based management rights were included in one FCM, none of the mental models contained rules that defined community management rights or their collective choice rights (Figure 3).

Stakeholders did not include rule enforcement through graduated sanctions, which establishes a feedback about the state of the system into user actions (Anderies et al. 2004). Stakeholders mentioned frequent monitoring of fish populations by state government, which informs management actions like harvest rules. They also mentioned communication and education, which can help improve user knowledge on the status of the fish stock. However, terms like “self-policing” and “norms” were the only reference to enforcement of rules. Norms are informal rules that do not include sanctioning mechanisms (Hinkel et al. 2014). Enforced graduated sanctions could help ensure that information on the state of the fish stock is translated into user actions. Interestingly, Pollnac et al. (2010) found a correlation between community led monitoring programs and compliance with regulations but did not find a correlation between enforcement of regulations and compliance.

Like the governance system, stakeholders identified fewer environment concepts compared to actor and resource system concepts. The environment concepts they did identify tended to be larger scale concepts like climate (global to national level) and social and political settings (national to state level). In addition, all rules in use identified by stakeholders were operational rules set by the state government (Figure 3). Ostrom (1990) suggested that nested institutions could help ensure that large scale problems are considered and addressed at the local scale. Allowing local institutions like lake districts and lake associations to set regulations through devolved management (Berkes 2010) might allow state government to consider and address larger scale issues and take actions at the local scale to offset variables they cannot control at larger scales (Carpenter et al. 2017). Allowing for multiple organizations to set regulations on a landscape of recreational fisheries can also improve the social and ecological resilience of the system (Carpenter and Brock 2004).

4.2. Hypothesis 2: FCMs underrepresented government and environment concepts

Our method of using fuzzy cognitive maps may have underrepresented governance system and environment concepts if FCMs can only capture variables that have clear agency that act within the system but not features of the governance system and environment, which informants may have viewed as fixed contextual settings. Governance and environmental concepts tend to be slower moving processes than actor or resource system concepts (e.g. regulations and climate change vs. fish harvest and fish reproduction). The temporal scale of drivers within recreational fisheries may have influenced the way stakeholders viewed the system; fast moving processes are often viewed as variables and can be captured by FCMs, while slower moving processes are often viewed as parameters or contextual settings that are fixed and may not be as easily captured by FCMs (Carpenter and Turner 2000, Cumming et al. 2006). However, our FCMs did capture several slow-moving governance and environment concepts (e.g. political setting, climate change, regulations, and legislation; Figure 3 and and informants consistently reported these components as having less influence within the social-ecological system (Figure 5).

Similar results have been observed in previous academic applications of the SESF to recreational fisheries that have not used FCMs. In an application of the SESF to inland recreational fisheries Hunt et al. (2013) emphasized the complexity within the actor component of the SESF. Similarly, Hinkel et al. (2014) applied the SESF to a recreational fishery and categorized most variables and concepts in the actor component (11), followed by the resource system (7), the environment (4), and the governance system (1). Our approach of using stakeholder mental models in application of the SESF resulted in a more complex view of recreational fisheries with more concepts and variables identified but the distribution of those concepts among SESF components was similar.

Use of mixed methods for elucidating if FCMs underrepresent government and environment concepts and promoting a view of actor agency within inland recreational fisheries represent future directions for research. Use of mixed methods like semi-structured interviews (Cinner et al. 2012, Leslie et al. 2015) and content analysis from local management reports (Gutiérrez et al. 2011), could help test if FCMs underrepresent governance and environmental concepts in stakeholder mental models. Promoting a view of actor agency that includes active participation in shaping the governance system and environment components could reduce the perception that these components are fixed and could expand their role in inland recreational fisheries management (Larsen et al. 2011).

4.3. Understanding social-ecological systems with stakeholder mental models

Our use of stakeholder mental models can aid in applications of the SESF at a local level. The advantages of using FCMs to help apply the SESF at a local scale are that they follow standardized methods that allow for quantitative testing, they are easy to understand and conduct for both the interviewer and the interviewee (Papageorgiou and Kontogianni 2012), and they involve stakeholders, which often promotes policy relevant results (Walker et al. 2002, Posner et al. 2016, Bennett 2017). Fuzzy cognitive maps may be particularly useful for local level SESF applications because social-ecological systems are notoriously hard to delineate spatially, temporally, and institutionally (Carpenter et al. 2009) and guidance on applying frameworks to specific systems or research questions is often scarce or non-existent (Binder et al. 2013). In a review of empirical applications of the SESF, Thiel et al. (2015) found that authors were not consistent in their application of the framework. Fuzzy cognitive maps represent one tool for helping to consistently apply the SESF among studies. However, the use of FCMs in combination with other methods like semi-structured interviews and content analysis of management reports could aid in revealing potential biases associated with FCMs. Our application of FCMs in combination with the SESF highlighted the opportunity for the governance system and local level actions addressing larger scale environment concepts to play an increased role in recreational fisheries management.

Even with a shared perception among stake holders of recreational fisheries as coupled social-ecological systems, the uncertainty caused by system complexity should be integrated into governance systems using design principles for sustainable shared resource use through strategies like the precautionary approach and adaptive management (Anderies et al. 2004, Kundzewicz et al. 2018). Incorporating academic paradigms and past research into on the ground management is likely to improve shared resource outcomes but uncertainty in the past, present, and future state of a social-ecological system ought to be incorporated into governance systems. The precautionary approach attempts to minimize adverse outcomes under the worst possible scenario, while adaptive management allows for strategies that are better suited for certain situations to be implanted quickly as those situations arise and continued learning to occur within the system (Kundzewicz et al. 2018).

Additional Files

The additional files for this article can be found as follows:

Figure S1

Example of a fuzzy cognitive map depicting traffic flow. Arrows represent direct relationships between concepts and variables that have a positive or negative correlation with a subjective strength of high, medium, or low. DOI:

Table S1

Original terms used in Fuzzy Cognitive Maps and how they were coded into the SESF. DOI:


We thank the University of Notre Dame Environmental Research Center (UNDERC) for hosting our research. We are exceptionally grateful to all the stakeholders who gave their time to be interviewed for this research. This work was supported by the Natural Sciences and Engineering Research Council of Canada under grant numbers 475586-2015 and 402530-2011 to J. P. Ziegler and C. T. Solomon respectively. This material is based upon work supported by the U.S. National Science Foundation under grant number 1716066 to C. T. Solomon and S. E. Jones. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.

Competing Interests

The authors have no competing interests to declare.


  1. Amblard, L. (2012). The potential of collective action for the control of nonpoint source pollution in European rural areas. Presented at: Design and Dynamics of Institutions for Collective Action: A Tribute to Prof. Elinor Ostrom, Second Thematic Conference of the IASC, 29 November–1 December, 2012. 

  2. Anderies, J. M., Janssen, M. A., & Ostrom, E. (2004). A framework to analyze the robustness of social-ecological systems from an institutional perspective. Ecol. Soc., 9, 18. DOI: 

  3. Arlinghaus, R., Alós, J., Beardmore, B., Daedlow, K., Dorow, M., Fujitani, M., Hühn, D., Haider, W., Hunt, L. M., Johnson, B. M., et al. (2017). Understanding and Managing Freshwater Recreational Fisheries as Complex Adaptive Social-Ecological Systems. Rev. Fish. Sci. Aquac., 25, 1–41. DOI: 

  4. Arlinghaus, R., Cooke, S. J., & Potts, W. (2013). Towards resilient recreational fisheries on a global scale through improved understanding of fish and fisher behaviour. Fish. Manag. Ecol., 20, 91–98. DOI: 

  5. Arlinghaus, R., Johnson, B. M., & Wolter, C. (2008). The Past, Present and Future Role of Limnology in Freshwater Fisheries Science. Int. Rev. Hydrobiol., 93, 541–549. DOI: 

  6. Basurto, X., & Ostrom, E. (2009). Beyond the Tragedy of the Commons. Econ. Delle Fonti Energ. E Dell’ambiente. 

  7. Baur, I., & Binder, C. (2013). Adapting to socioeconomic developments by changing rules in the governance of common property pastures in the Swiss Alps. Ecology and Society, 18(4). DOI: 

  8. Beard, T. D., Arlinghaus, R., Cooke, S. J., McIntyre, P. B., Silva, S. D., Bartley, D., & Cowx, I. G. (2011). Ecosystem approach to inland fisheries: research needs and implementation strategies. Biol. Lett., 7, 481–483. DOI: 

  9. Bennett, E. M. (2017). Research frontiers in ecosystem service science. Ecosystems, 20(1), 31–37. DOI: 

  10. Berkes, F. (2009). Evolution of co-management: Role of knowledge generation, bridging organizations and social learning. J. Environ. Manage., 90, 1692–1702. DOI: 

  11. Berkes, F. (2010). Devolution of environment and resources governance: trends and future. Environ. Conserv., 37, 489–500. DOI: 

  12. Berkes, F., Feeny, D., McCay, B. J., & Acheson, J. M. (1989). The benefits of the commons. Nature, 340, 91. DOI: 

  13. Bernard, H. R., & Bernard, H. R. (2012). Social Research Methods: Qualitative and Quantitative Approaches. Thousand Oaks: SAGE. 

  14. Biggs, D., Abel, N., Knight, A. T., Leitch, A., Langston, A., & Ban, N. C. (2011). The implementation crisis in conservation planning: could “mental models” help? Conserv. Lett., 4, 169–183. DOI: 

  15. Blanco, E., & Fedreheim, G. E. (2011). National parks in Norway as socio-ecological system: wildlife, conflict in use, and community-based management. Presented at the 13th Biennial Conference of the International Association for the Study of the Commons (IASC), Hyderabad, India. 

  16. Carpenter, S. R., & Brock, W. A. (2004). Spatial complexity, resilience, and policy diversity: fishing on lake-rich landscapes. Ecol. Soc., 9, 8. DOI: 

  17. Carpenter, S. R., Brock, W. A., Hansen, G. J. A., Hansen, J. F., Hennessy, J. M., Isermann, D. A., Pedersen, E. J., Perales, K. M., Rypel, A. L., Sass, G. G., et al. (2017). Defining a Safe Operating Space for inland recreational fisheries. Fish and Fisheries, 18, 1150–1160. DOI: 

  18. Carpenter, S. R., Mooney, H. A., Agard, J., Capistrano, D., DeFries, R. S., Díaz, S., Dietz, T., Duraiappah, A. K., Oteng-Yeboah, A., Pereira, H. M., et al. (2009). Science for managing ecosystem services: Beyond the Millennium Ecosystem Assessment. Proc. Natl. Acad. Sci., 106, 1305–1312. DOI: 

  19. Carpenter, S. R., & Turner, M. G. (2000). Hares and Tortoises: Interactions of Fast and Slow Variablesin Ecosystems. Ecosystems, 3, 495–497. DOI: 

  20. Cinner, J. E., McClanahan, T. R., MacNeil, M. A., Graham, N. A. J., Daw, T. M., Mukminin, A., Feary, D. A., Rabearisoa, A. L., Wamukota, A., Jiddawi, N., et al. (2012). Comanagement of coral reef social-ecological systems. Proc. Natl. Acad. Sci., 109, 5219–5222. DOI: 

  21. Costanza, R., Wainger, L., Folke, C., & Mäler, K.-G. (1993). Modeling Complex Ecological Economic Systems Toward an evolutionary, dynamic understanding of people and nature. BioScience, 43, 545–555. DOI: 

  22. Cumming, G., Cumming, D. H., & Redman, C. (2006). Scale mismatches in social-ecological systems: causes, consequences, and solutions. Ecology and society, 11(1). DOI: 

  23. del Mar Delgado-Serrano, M., & Ramos, P. A. (2015). Making Ostrom’s framework applicable to characte-rise social ecological systems at the local level. International Journal of the Commons, 9(2), 808–830. DOI: 

  24. Dickerson, J. A., & Kosko, B. (1994). Virtual worlds as fuzzy cognitive maps. Presence: Teleoperators & Virtual Environments, 3(2), 173–189. DOI: 

  25. Dietz, T., Ostrom, E., & Stern, P. C. (2003). The Struggle to Govern the Commons. Science, 302, 1907–1912. DOI: 

  26. Dumyahn, S. L., & Pijanowski, B. C. (2011). Beyond noise mitigation: managing soundscapes as common-pool resources. Landsc. Ecol., 26, 1311. DOI: 

  27. Fulton, E. A., Smith, A. D. M., Smith, D. C., & van Putten, I. E. (2011). Human behaviour: the key source of uncertainty in fisheries management. Fish and Fisheries, 12, 2–17. DOI: 

  28. Gutiérrez, N. L., Hilborn, R., & Defeo, O. (2011). Leadership, social capital and incentives promote successful fisheries. Nature, 470, 386–389. DOI: 

  29. Hilborn, R., Walters, C. J., & Ludwig, D. (1995). Sustainable Exploitation of Renewable Resources. Annu. Rev. Ecol. Syst., 26, 45–67. DOI: 

  30. Hinkel, J., Bots, P., & Schlüter, M. (2014). Enhancing the Ostrom social-ecological system framework through formalization. Ecology and Society, 19(3). DOI: 

  31. Holling, C. S. (2001). Understanding the Complexity of Economic, Ecological, and Social Systems. Ecosystems, 4, 390–405. DOI: 

  32. Holt, S. J., & Talbot, L. M. (1978). New principles for the conservation of wild living resources. Wildlife Monographs, 59, 3–33. 

  33. Horan, R. D., Fenichel, E. P., Drury, K. L. S., & Lodge, D. M. (2011). Managing ecological thresholds in coupled environmental–human systems. Proc. Natl. Acad. Sci., 108, 7333–7338. DOI: 

  34. Hunt, L. M., Sutton, S. G., & Arlinghaus, R. (2013). Illustrating the critical role of human dimensions research for understanding and managing recreational fisheries within a social-ecological system framework. Fish. Manag. Ecol., 20, 111–124. DOI: 

  35. Johnston, F. D., Arlinghaus, R., & Dieckmann, U. (2010). Diversity and complexity of angler behaviour drive socially optimal input and output regulations in a bioeconomic recreational-fisheries model. Can. J. Fish. Aquat. Sci., 67, 1507–1531. DOI: 

  36. Kates, R. W., Clark, W. C., Corell, R., Hall, J. M., Jaeger, C. C., Lowe, I., McCarthy, J. J., Schellnhuber, H. J., Bolin, B., Dickson, N. M., et al. (2001). Sustainability Science. Science, 292, 641–642. DOI: 

  37. Kundzewicz, Z. W., Krysanova, V., Benestad, R. E., Hov, Ø., Piniewski, M., & Otto, I. M. (2018). Uncertainty in climate change impacts on water resources. Environ. Sci. Policy, 79, 1–8. DOI: 

  38. Larsen, R. K., Calgaro, E., & Thomalla, F. (2011). Governing resilience building in Thailand’s tourism-dependent coastal communities: Conceptualising stakeholder agency in social–ecological systems. Glob. Environ. Change, 21, 481–491. DOI: 

  39. Leslie, H. M., Basurto, X., Nenadovic, M., Sievanen, L., Cavanaugh, K. C., Cota-Nieto, J. J., Erisman, B. E., Finkbeiner, E., Hinojosa-Arango, G., Moreno-Báez, M., et al. (2015). Operationalizing the social-ecological systems framework to assess sustainability. Proc. Natl. Acad. Sci., 112, 5979–5984. DOI: 

  40. Liu, J., Dietz, T., Carpenter, S. R., Alberti, M., Folke, C., Moran, E., Pell, A. N., Deadman, P., Kratz, T., Lubchenco, J., et al. (2007). Complexity of coupled human and natural systems. Science, 317, 1513–1516. DOI: 

  41. Ludwig, D. (2001). The Era of Management Is Over. Ecosystems, 4, 758–764. DOI: 

  42. Ludwig, D., Hilborn, R., & Walters, C. (1993). Uncertainty, Resource Exploitation, and Conservation: Lessons from History. Science, 260, 17–36. DOI: 

  43. Mace, G. M. (2014). Whose conservation? Science, 345, 1558–1560. DOI: 

  44. Mangel, M., Talbot, L. M., Meffe, G. K., Agardy, M. T., Alverson, D. L., Barlow, J., Botkin, D. B., Budowski, G., Clark, T., Cooke, J., et al. (1996). Principles for the Conservation of Wild Living Resources. Ecol. Appl., 6, 338–362. DOI: 

  45. McGinnis, M., & Ostrom, E. (2014). Social-ecological system framework: initial changes and continuing challenges. Ecology and Society, 19(2). DOI: 

  46. Meinzen-Dick, R. (2007). Beyond panaceas in water institutions. Proc. Natl. Acad. Sci. 104, 15200–15205. DOI: 

  47. Murray, G. D., & Ings, D. (2015). Adaptation in a time of stress: A social-ecological perspective on changing fishing strategies in the Canadian snow crab fishery. Mar. Policy, 60, 280–286. DOI: 

  48. Naiga, R., Penker, M., & Hogl, K. (2015). Challenging pathways to safe water access in rural Uganda: From supply to demand-driven water governance. International Journal of the Commons, 9(1), 237–260. DOI: 

  49. Ostrom, E. (1990). Governing the commons: the evolution of institutions for collective action. Cambridge: Cambridge University Press. DOI: 

  50. Ostrom, E. (2007). A diagnostic approach for going beyond panaceas. Proc. Natl. Acad. Sci., 104, 15181–15187. DOI: 

  51. Ostrom, E. (2009). A General Framework for Analyzing Sustainability of Social-Ecological Systems. Science, 325, 419–422. DOI: 

  52. Otto-Banaszak, I., Matczak, P., Wesseler, J., & Wechsung, F. (2011). Different perceptions of adaptation to climate change: a mental model approach applied to the evidence from expert interviews. Reg. Environ. Change, 11, 217–228. DOI: 

  53. Özesmi, U., & Özesmi, S. L. (2004). Ecological models based on people’s knowledge: a multi-step fuzzy cognitive mapping approach. Ecol. Model., 176, 43–64. DOI: 

  54. Papageorgiou, E., & Kontogianni, A. (2012). Using fuzzy cognitive mapping in environmental decision making and management: a methodological primer and an application. INTECH Open Access Publisher. DOI: 

  55. Peterson, G. D., Beard, T. D., Jr., Beisner, B. E., Bennett, E. M., Carpenter, S. R., Cumming, G. S., Dent, C. L., & Havlicek, T. D. (2003). Assessing future ecosystem services: a case study of the Northern Highlands Lake District, Wisconsin. Conserv. Ecol., 7(3), 1. DOI: 

  56. Pollnac, R., Christie, P., Cinner, J. E., Dalton, T., Daw, T. M., Forrester, G. E., Graham, N. A. J., & McClanahan, T. R. (2010). Marine reserves as linked social–ecological systems. Proc. Natl. Acad. Sci., 107, 18262–18265. DOI: 

  57. Posner, S. M., McKenzie, E., & Ricketts, T. H. (2016). Policy impacts of ecosystem services knowledge. Proc. Natl. Acad. Sci., 113, 1760–1765. DOI: 

  58. Risvoll, C., Fedreheim, G. E., Sandberg, A., & BurnSilver, S. (2014). Does Pastoralists’ participation in the management of national parks in northern Norway contribute to adaptive governance? Ecol. Soc., 19. 

  59. Salas, S., & Gaertner, D. (2004). The behavioural dynamics of fishers: management implications. Fish and Fisheries, 5, 153–167. 

  60. Schlüter, M., Hinkel, J., Bots, P., & Arlinghaus, R. (2014). Application of the SES framework for model-based analysis of the dynamics of social-ecological systems. Ecology and Society, 19(1). DOI: 

  61. Thiel, A., Adamseged, M. E., & Baake, C. (2015). Evaluating an instrument for institutional crafting: How Ostrom’s social–ecological systems framework is applied. Environ. Sci. Policy, 53(B), 152–164. DOI: 

  62. Vasslides, J. M., & Jensen, O. P. (2016). Fuzzy cognitive mapping in support of integrated ecosystem assessments: Developing a shared conceptual model among stakeholders. J. Environ. Manage, 166, 348–356. DOI: 

  63. Walker, B. H., Carpenter, S., Anderies, J., Abel, N., Cumming, G., Janssen, M., Lebel, L., Norberg, J., Peterson, G. D., & Pritchard, R. (2002). Resilience Management in Social-ecological Systems: a Working Hypothesis for a Participatory Approach. Conserv. Ecol., 6, 14. DOI: 

  64. Worm, B., Hilborn, R., Baum, J. K., Branch, T. A., Collie, J. S., Costello, C., Fogarty, M. J., Fulton, E. A., Hutchings, J. A., Jennings, S., et al. (2009). Rebuilding Global Fisheries. Science, 325, 578–585. DOI: