APPENDIX 2. Methodological details of the conceptual modeling and participatory scenario analysis.

Here, scenarios are interpreted as alternative futures that are neither predictions nor forecasts, but stylized and contrasting desirable or alarming images of how the future might unfold. Drawing on participatory research (Reed et al. in press), the scenario development in this study has some innovative merits, such as combining explorative and anticipatory methods. At the core of the framework was an iterative, two-way learning cycle between researchers and stakeholders for formulating a portfolio of environmental-management options and policy proposals for adaptation to change. The scenario analysis specifically addressed uncertainties and surprises, by incorporating alternative and potentially conflicting perspectives, values, and interests and by encouraging participation through negotiated deliberation processes. The conceptual modeling exercise and scenario analysis were developed in three key phases: (1) exploring narratives of the agropastoral system’s historical and current structure and functions, (2) envisioning desirable and adverse visions for the future, and (3) backcasting to discuss how these futures could emerge and what policy options could be implemented to achieve them.

Conceptual Modeling Exercise

Once the key stakeholders’interests and relevance had been characterized (Ravera et al. 2009), their early participation was vital to ensure representative, dynamic, and durable decision making throughout the process. The historical analysis of trends and drivers of vulnerability of livelihoods in the studied area was obtained by a triangulation of participatory methods that included key informant interviews (N=5), a focus group with village elders (N=12), and more classical research methods such as aerial photographs and satellite-image interpretation (1954, 1971, 1988, 1996, and 2008), literature review, and archive material study. Secondly, a perception analysis was carried out to explore conflicting concerns regarding environment and development issues and perceptions, and representation of changes in vulnerability. Two series of in-depth and semi-structured interviews were conducted (N=23 and N=41 respectively) within categories of local stakeholders (landless people, small agropastoral farmers, medium-scale semi-rural cattle raisers, traditional large-scale landowners and commercial entrepreneurs, women as single parents, and youth). They were selected through snowball sampling. Four focus groups were then involved in a collective discussion: small agropastoral male farmers (N=15), women (N=20), youth (N=12), and landless people (N=13). We also interviewed representatives of institutions interested or involved in natural resource management in the area (N=13) (e.g., local authorities, government agencies, local administration, trade unions, NGOs, private enterprises) and we organized a focus group of local experts from NGOs and research institutes (N=12). A mix of methods was then used to code and represent local narratives, such as visual representations and grounded theory analysis applied to transcripts and combined with literature and field observations. The final decisions on how to visualize the narratives as a conceptual model were taken to two series of meetings with experts. These meetings included Nicaraguan and Spanish researchers on agroeconomy, agronomy, ecological economics, and ecology, Nicaraguan teachers, and environmental technicians. The experts also decided how to present uncertain and conflicting visions. A simplified version of the conceptual model was discussed in in-depth interviews with key informants (N=12) and presented back to local stakeholders during a series of extended meetings: four meetings with small farmers and landless people, two meetings with large-scale commercial traditional landowners and entrepreneurs, and a meeting with representatives of local institutions. The conceptual model was cross-validated with researchers from system analysis, ecological economy and ecology (N=5), and then developed in the VENSIM program.

Visioning Exercise

In a complementary exercise stakeholders, were asked to envision the connections between components and drivers to changes and future vulnerability. To account for different future visions and to discuss potential uncertainties and surprises, the focus-group participants were divided into mixed subgroups and were guided to construct a set of desirable and undesirable scenarios. To engage participants without formal education, illustrations such as collages from magazines, photographs, sketched maps of the region, etc., were used to create an image of the future and discuss associated storylines on drivers and changes in the multiple dimensions of vulnerability, based on Fraser’s work (2007). A complementary series of in-depth interviews (N=23) was used to explore metaphors that captured stakeholders’ expectations about the future. Quite independently of the details, the metaphors dramatized the inner significances of the situation and alluded to the kind of world within which stakeholders belonged. Titles and the final storylines of future scenarios were then re-elaborated by the research staff.

Backcasting Exercise

A second series of focus groups was conducted with male small farmers and landless people (N=13), women (N=15), commercial landowners and medium semi-rural ranchers (N=6), and local authorities and representatives of institutions (N=10). Here, the conceptual model inputs and scenario narratives were used as the basis for a backcasting exercise. Starting from the future scenarios, the participants were asked to go back to the present time, identifying obstacles and opportunities that might emerge on the way. For each scenario, the likeliness of factors that might influence the vulnerability was inferred. To converge conflicting interests, the likelihood and desirability of different scenarios were discussed, resulting in a compromise about a “sustainable scenario.” The participants started by identifying what changes in land-use allocation, land-management practices, and socioeconomic and institutional arrangements were to be implemented at the present time, to lead to the respective future scenarios. The support of a 3D landscape model helped to ground the discussion in the current context and landscape, and to heuristically anticipate measures to avoid undesirable futures. Participants were asked to respond to a list of key policy questions, derived from the analysis of assumptions and components for the four scenarios. Then, a set of plausible pathways to achieve desirable states was created, and adaptive management strategies were discussed. Throughout the process, feedback and dissemination of information to stakeholders allowed for a dynamic participatory learning process experience, and a set of different tools were useful for overcoming language barriers and preventing misunderstandings. In the future, further research steps will use dynamic computer-based modeling to examine vulnerability indicators and empirically monitor and simulate future changes in livelihood vulnerability under each scenario and options.