Futures 73 (2015) 86–99

Contents lists available at ScienceDirect

Futures

journal homepage: www.elsevier .com/ locate / fu tures

Multi-scale participatory scenario methods and territorialplanning in the Brazilian Amazon

Ricardo Theophilo Folhesa,b, Ana Paula Dutra de Aguiarc,*, Emilie Stolla,d,Eloi Lennon Dalla-Norac, Roberto Araújoa,c, Andrea Coelhoe, Otávio do Cantoa

aUniversidade Federal do Pará (UFPA), rua Augusto Corrêa 1, Guamá, 66075-110 Belém, Pará, BrazilbUniversité Paris 3 Sorbonne Nouvelle, 28 rue Saint Guillaume, 75007 Paris, FrancecBrazilian National Institute for Space Research (INPE), av. dos Astronautas 1758-Jd Granja, 12227-010 São José dos Campos, São Paulo,BrazildUniversity of Aberdeen/Fyssen Foundation, 24 High street, Aberdeen, Scotland, United Kingdome Instituto de Desenvolvimento Econômico, Social e Ambiental do Pará, Rua Municipalidade 66050-350, Umarizal, 66050350 Belém, Pará,Brazil

A R T I C L E I N F O

Article history:

Received 28 November 2012Received in revised form 31 July 2015Accepted 25 August 2015Available online xxx

Keywords:ScenariosParticipatory methodsTerritorial planningBrazilian AmazonLand use and cover changeDeforestation

* Corresponding author at: Av dos AstroE-mail address: ana.aguiar@inpe.br (A.P

http://dx.doi.org/10.1016/j.futures.2015.08.0016-3287/$ – see front matter ã 2015 Els

nautas 1.D.d. Agu

005evier Ltd

A B S T R A C T

In recent years, discussions regarding scenariosmethods for the Brazilian Amazon have beenmostly explored from the sole perspective of deforestation concerns. This paper proposesan original approach using participatory scenarios as a method that may be able to put inperspective, at different levels of decision-making, a specific action of territorial planningin Pará State, Brazil. The method allows the dialogue between stakeholder representatives,government organizations and communities involved in that territorial process. Thescenarios produced substantial and sometimes contradictory data: while they can beconsidered as a way of empowerment for the local communities, participatory scenariosalso have their limits andmay reveal structural forms of authority or dominationwithin theproject-promoting institutions and local communities. The information collected allowsscientists of various areas (Modeling, Social Sciences), community leaders andmanagers toelaborate a reflection upon the levers that may condition the implementation of effectiveactions and public policies in territorial units.

ã 2015 Elsevier Ltd. All rights reserved.

1. Introduction

In recent decades, scenarios approaches came to be considered a valuable tool in the environmental area, due to concernsrelated to climatic changes, water availability, ecosystem functioning, air quality and land use change (Wilkinson & Eidinow,2008). Scenarios are plausible, challenging, and relevant stories about how the futuremight unfold, which can be expressedin both words and numbers. Scenarios are not projections, predictions, or recommendations. They are about envisioningfuture pathways and accounting for critical uncertainties (Raskin, Monks, Ribeiro, Van Vuuren, & Zurek, 2005).Environmental scenarios have been built up on global, regional and local scales, with different objectives and degrees ofsocial participation (Alcamo & Ribeiro, 2001; Kok, Patel et al., 2006; Kok, Rothman, & Patel, 2006). Scenario-buildingexperiences carried out in different parts of theworld indicate that involving stakeholders in the reflectionprocessmay favor

758, Jd Granja, 12227 010, São José dos Campos, São Paulo, Brazil.iar).

. All rights reserved.

R.T. Folhes et al. / Futures 73 (2015) 86–99 87

understanding of diverging points of view, and the attainment of consensus, facilitating the collective decision-makingprocess (Wollemberg, Edmunds, & Buck, 2000).

Up to now, scenario discussions for the Brazilian Amazon has beenmostly limited to future deforestation trends at broad-scale (Buarque, 2003; Lapola et al., 2010; Laurance et al., 2001; Soares-Filho et al., 2006). The Brazilian Amazon rainforestcovers an area of approximately 4 million km2. Due to the intense and increasing occupation process in the last decades,approximately 18% of the forest has already been converted into pastures and croplands. The biodiversity loss and thegreenhouse gas emissions derived from the deforestationprocess drew international attention to the region. However, underthe deforestation process lies a dynamic of occupation, historically built on top of territorial and agrarian conflicts, for landpossession and ownership (Araújo & Léna, 2010; Becker, 1997, 2009; Bunker, 1985; Schmink & Wood, 1992).

In the last decade, along with deforestation control policies (Dalla-Nora, Aguiar, Lapola, & Woltjer, 2014), severalgovernmental programs focusing on land tenure regularization and state territorial planning were implemented in theAmazon region. The aim was to regularize the occupation of public land (since most of Amazonian peasants are not legalowners of their lands),mediate conflicts of interest and lay down rules of use of the land and its natural resources. As a result,today, the Brazilian Amazon looks like a mosaic of different territorial units, including indigenous lands, protected areas ofdifferent types (national forests, ecological reserves, etc.), and agrarian projects to settle farmers. Such units act as a barrieragainst the deforestation progression and contribute to biodiversity conservation purposes. But they also have the socialfunction of granting land tenure rights to a diverse local population. In this context, the State conceived new modalities ofagrarian projects and conservation units, based on the sustainable exploitation of natural resources. Such is the case of theAgro-extractivist Settlement Project (PAE) modality, a kind of unit aiming at regularizing the land situation of the so-called“traditional populations” (a legal category that designate historical and riverine peasantries). Currently, the mosaic ofterritorial units covers approximately 60% of the region. Therefore, the future of the region largely depends on thesustainability of each unit.

The use of scenarios in the Brazilian Amazon as a tool for local population empowerment and collective decision-makingis an incipient process (de Aguiar et al., 2014; Folhes, 2010). In this paper, we propose to adapt participatory scenarioapproaches to explore how this method could contribute to strengthen local population political awareness, whicheventually would fortify the creation process of territorial units and their subsequent sustainability. Our hypothesis is that asimple and replicable normative andmulti-scalemethod (Folhes, 2010;Wollemberg et al., 2000) could facilitate the dialogueacross levels, mainly between local populations and governmental agencies, and help to promote the empowerment of thefirst. To explore this idea, we chose to focus our research in the Agro-extractivist Settlement Project PAE LagoGrande, locatedin the rural area of the municipality of Santarém, West of Pará State, along the Amazon River (Fig. 3).

PAE Lago Grande is an agrarian project that was created in 2005 by the National Institute for Colonization and AgrarianReform (INCRA), in a region characterized by a precolonial human occupation process as it is attested in chroniclers’ registers(e.g. Acuña, 1994; Bettendorff, 2010) and recent archaeological works (e.g. Gomes, 2007). PAE Lago Grande illustrates thesituation of many other units in the Amazon: even after being officially created, it still does not receive public policies’benefits, such as those aiming at regularizing the land-tenure status and promoting socioeconomic development throughthe implementation of infrastructures (roads, electricity network, etc.), credits and technical assistance. Beyond theseproblems, local populations are being highly pressured by economic interests connected to the lumber, mineral andagricultural sectors attracted by PAE Lago Grande land and its natural resources.

In the following section, wewill introduce a theoretical dialog between the scenariomethod that we adopted in our workand the other potentially relevant scenario approaches. In Section 3, we will present a detailed description of the researcharea (socio-environmental characterization, social organization, socio-environmental conflicts, and communities’characterization) and of the research protocol. In Section 4, we will expose the results and discuss them at the communityand settlement levels, including convergences and divergences between scales. Finally, in Section 5, we will synthesize ourmain conclusions.

2. On scenario methods

2.1. Background: main scenario types

Various typologieshavebeenproposedtoclassifyscenarioapproaches (Aguiaretal., 2014;Alcamo&Ribeiro,2001 [267_TD$DIFF];Börjeson,Höjer, Dreborg, Ekvall, & Finnveden, 2006; Bradfield, Wright, Burt, Cairns, & Van Der Heijden, 2005; Godet, 2000; Kok, 2009;Miller, 2007; Swart, Raskin, & Robinson, 2004; van Notten, Rotmans, van Asselt, & Rothman, 2003). In this section, in order tocontextualize the approachwe adopted in relation to theworksmentioned above,we summarize themainvariations found onenvironmental scenarios through an organization in four topics: (1) qualitative and quantitative scenarios; (2) normative andexploratory scenarios; (3) participatory or developed by specialist scenarios; (4) single or multi-scale scenarios.

2.1.1. Qualitative and quantitative scenariosScientists developed two distinct approaches for scenarios’ construction during the period betweenWorldWar II and the

mid-90s: (a) quantitative modeling and (b) qualitative narratives (Raskin et al., 2005). However, these approaches did notdialog between them. That dualism reflects two challenges of equal importance in terms of scenarios: to provide systematicand replicable representations, on one hand, and to contrast social and non-quantitative visions, on another. According to

88 R.T. Folhes et al. / Futures 73 (2015) 86–99

Swart et al. (2004), quantitative analysis often relies on formal models, using mathematical algorithms to represent keyfeatures of human and environmental systems. While the narrative scenarios gives voice to important qualitative factors,shaping development such as values, behaviors and institutions, providing a broader perspective than it is possible frommathematical modeling alone (Alcamo, 2008; Swart et al., 2004). Narratives provide texture, richness, and insight, whilequantitative analysis offers structure, discipline and rigor (Raskin et al., 2005). In the environmental area, the combination ofboth types of scenarios is commonly practiced (Alcamo, 2008).

2.1.2. Normative and exploratory scenariosAnother way of classify scenarios is related to their perspective (Alcamo & Ribeiro, 2001; Alcamo, 2008; Raskin et al.,

2005; Swart et al., 2004): from the present to the future (exploratory) or from a vision of the future to the present (normative/backcasting). Exploratory scenarios are those ones, which begin in the present and seek to explore different trends to thefuture. This type of scenario is closer to the originalmeaning of theword “scenario” as they consist of a sequence of emergentevents. Normative scenarios, on the other hand, begin with a prescribed overview of the future (optimistic, pessimistic orneutral). Backcasting can be defined as “first creating a desirable (sustainable) future vision or normative scenario, followedby looking back at how this desirable future could be achieved, before defining and planning follow-up activities anddeveloping strategies leading towards that desirable future” (Quist & Vergragt, 2004, p. 410). Normative/backcastingscenarios are built backwards linking the established future to the present. Exploratory scenarios are, theoretically, “valuefree”; in general, they are conceived to analyze the possible course of future events on a particular topic (for instance, wateravailability, CO2 atmospheric concentration) to facilitate the positioning and decision-making. Although environmentalscenarioswere largely exploratory in the past decades, some recentworks combine both perspectives, for instance, theWordWater Vision scenarios (Cosgrove, Rijsberman, & Council, 2000).

Following Vergragt and Quist (2011), we postulate that normative approaches “gradually becamemore popular and havebeen more widely applied over the last decade, due to the rising popularity of the strongly normative concept ofsustainability [ . . . ]. For instance, [268_TD$DIFF]Giddens (2009) has suggested backcasting as a sustainable alternative to traditionalplanning and considers backcasting a tool formoving toward alternative futureswhen dealingwith climate change” (p. 747).

2.1.3. Participatory or developed by specialist scenariosSeveral authors have discussed the benefits of broader participatory approaches in the scenario-building process (Dufva &

Ahlqvist, 2015; Patel, Kok, & Rothman, 2007; Vliet, Kok, & Veldkamp, 2010;Wollemberg et al., 2000). The benefits include thecreation of scenarios that are relevant and credible to the end-users/beneficiaries, and that synthesize better information asthey include local knowledge. Besides, participation provides a learning arena for all participants involved (van Vliet et al.,2010). Patel et al. (2007) argue that “interactive participation, i.e. where the participants can engage directly with each otherand the researchers or decision-makers, is regarded as being fundamental for achieving ’empowerment’ [ . . . ]. [It can]enhance the confidence of stakeholders, so to enable them to define, express and analyze their reality, and not to reflect theopinions of the stronger, more dominant voices” (p. 550). This point is also stressed in Dufva and Ahlqvist (2015). On theother hand, difficulties such as the lack of reproducibility and the result’s dependence on participants’ selection and solidmediation should also be noted (Kok, Rothman et al., 2006; van Vliet et al., 2010).

In general, normative scenarios require greater stakeholders’ participation on the scenarios formulation process(although exploratory scenarios can also be developed using participatorymethods). According toWollemberg et al. (2000),scenarios are more efficient to create new knowledge when they involve stakeholders. At global scale, scenarios tend to beexploratory, and focus on quantitative analysis based on experts’ knowledge (Parson, 2008). Stakeholders and decision-makers are usually involved on reviewing steps at the end of the process.

2.1.4. Single or multi-scale scenariosFinally, scenarios can be built for a single or multiple scales. Environmental scenarios have often been built at different

scales of time and space, from global to local scales. In general, global scenarios cover larger time horizons, 50–100 years,while regional and local scenarios focus on shorter periods, 20–30 years. Alcamo (2008) recommend multi-scale scenariosdevelopment as a way to increase the consistency and relevance of the scenarios. An important issue regarding multi-scalescenarios is how the interactions between scales are considered in the scenario-building process (Alcamo, 2008; Folhes,2010). The scale interactionsmay run in a top-down (global/regional scenarios contextualizing local scenarios) or bottom-upmanner (local scenarios feeding regional/global scenarios). Relations between scales can be “loosely linked”, where thescenarios developed independently (or partially independently), or “tightly coupled” when prospects, uncertainties, andfactors of one scale inform the construction of scenarios in the other scales. This decision depends again of the purpose of thestudy, taking into account, for example, the necessity to compare results andmaintain consistency among scales—versus thefact that excessive coupling can restrict the creativity in local scales, affecting the outcome scenarios, especially the feeling of“ownership” of the scenarios by stakeholders.

In summary, the widespread use of scenarios followed the emergence of a variety of techniques and methodologies,which resulted in a complex network of concepts and definitions. Different approaches have potentials and restrictions, andtheir choice depends on the research question we want to answer. In the scope if this paper, our goal is to explore howexisting scenario approaches could contribute to the strengthening of territorial units in tropical forest areas. For territorialplanning purposes, political and institutional issues are core parts of the problem. In this case, stakeholders’ engagement at

R.T. Folhes et al. / Futures 73 (2015) 86–99 89

different levels is essential on the scenarios formulation as well as to promote participatory planning. Therefore, we proposea qualitative, participatory, normative and multi-scale approach, as detailed in the next section.

2.2. How did we apply the scenarios in this study?

Our approach is participatory, as the scenarios were produced with the contributions of representatives of different socialsegments, interest groups and decision-takers, by means of workshops where participatory methods are employed. Onceconducted through participatory methods, scenario initiatives can promote commitment among the people in the localcommunity scale and enhance the governance quality by creating processes that are more democratic and equitable (Patelet al., 2007; van AsseltMarjolein & Rijkens-Klomp, 2002). Also, participatorymethodsmay ensure that eventual changeswillhave more credibility throughout the segments of the community because they were discussed by a group representing allsegments of the community (Kok, Rothman et al., 2006).

Amulti-scale strategy has also to be taken explicitly. Workshops were carried out within the community level and in thesettlement level, through several meetings over a period of time (2 years) that permitted a relevant dialog. Threecommunities with distinct socio-economic and biophysical characteristics were selected, as detailed in Section 3. The scaleswere “loosely-coupled” (Biggs et al., 2007; Kok, Biggs, & Zurek, 2007): scenarios were independently conceived in eachcommunity and at the settlement scale, but the workshops followed always the same structure (presented below). Later inthe process, we organized integrative workshops where representatives from both scales (community, institutions andstakeholders) analyzed the results and identified convergences/divergences. We chose this approach, as we did not want toimpose pre-defined ideas about the future upon the participants. The goal was to allow the scenarios to emerge according toeach context, and then compare the results.

The final products of the scenarios arewritten narratives of the future up to 2020, in the form of text and summary tables.Therefore, our scenarios are purely qualitative. We used maps, images and tabular quantitative information to support theworkshops and the analysis.

Finally, the key decisionwas to select the normative approach to scenarios construction. The process we adopted consistsof defining visions of a desirable and a non-desirable future, and the actions needed to build a trajectory linking the future tothe present, using backcasting methods (Kok, Patel et al., 2006). We based our choice on the understanding that discussingcommon desirable future could provide the ground to engagemore directly the stakeholders in the scenario process than anexploratory exercise, while at the same time, it would facilitate the dialogue across scales and interest groups. Fig. 1illustrates the structure adopted in all workshops, independently of the scale.

[(Fig._1)TD$FIG]

Fig. 1. Overall structure of the proposed method.

90 R.T. Folhes et al. / Futures 73 (2015) 86–99

During each workshop, the discussions were split into four stages: (1) past/present; (2) present/near Future; (3) futuredesired/undesired; (4) trajectory present/desired future. As our goal was to limit each workshop up to two days, and to makesure that all stages were covered, we further enforced the structure by defining a set of focal themes. In each stage, the samefocal themes were discussed. In PAE Lago Grande, for instance, based on previous knowledge of the main issues in the area(Folhes, 2007), we selected the following themes as focal points during the workshops: (i) social organization, (ii)infrastructure, (iii) projects and land use, (iv) socio-environmental conflicts. A team of mediators had the role of making surethat the time frame and the themes were respected, nevertheless avoiding interfering with the content of the discussions.The four stages of the workshops, repeated in all communities and scales, are briefly described below:

1.

Past/present– Data pertaining to the past and present, previously systematized into the selected discussion themes byresearchers, were presented and discussed providing a basis of reflection on the history and present status of thecommunity and the settlement.

2.

Present/near future – Short-term expectation (probable futurewithin 2/3 years), some of which had already been foreseenin the previous stage, are debated by participants and summarized per theme.

3.

Desired/undesired future – The group debates on desired and non-desired future, for each theme, taking as a reference atime horizon, which – in the case of PAE Lago Grande –was the year 2020. Initially, the desired and undesired futures aresummarized, using keywords, as in the previous stages. Thereafter, researchers request participants – divided into groups– to make up narrations as texts, synthesizing a vision of a desired and a non-desired future.

4.

Elaboration of trajectories – Participants are ultimately required to discuss which plausible paths they should take in orderto reach the desired future. They have to decide which authority is the most relevant to tackle each issue previouslydebated by using a backcasting approach (Kok, Patel et al., 2006). During the elaboration of the trajectory, we solicited theparticipants to define which institutions could support them to complete each step. At this point, we offered acomplementary method: we invited the participants to define their real power of interference regarding each one of theseissues. This proposition is based on the Structural Matrix Analysis as discussed in Buarque (2003) according to whom, foreach variable, the degree of interdependency and power of interference of the players involved is discussed. Thismodification of the backcastingmethod allowed us tomeasure the perceived level of decision-making of every workshopparticipant. It was evaluated on a scale from 1 to 3; one representing no decision-making capacity; two, a low or limiteddecision-making capacity; and three, possible decision-making capacity.

In the Section 3, we detail the complete methodological approach devised to build the scenarios for a given area. Theapplication consists of several activities aiming at the preparation of the workshops, their execution, data organization andstakeholder feedback.

3. Case study: scenario process in PAE Lago Grande, Brazilian Amazon

3.1. The applied research process

The application of the process outlined in the previous section for PAE Lago Grande consisted of the following activities:(a) primary data collection; (b) secondary data collection; (c) workshop planning and participant selection; (d) execution ofthe workshops; (e) handling over preliminary results at each scale; and (f) final integrative debate.

The primary data collection consisted of undertaking fieldwork within the communities and social organizationsdeveloping activities in PAE Lago Grande. Members of our teammade direct observation of the social dynamics in differentcommunities in the area. They would then select the communities where wewould organize the workshops, as well as theycould understand the regional planning actions at the settlement level. The secondary data-collecting phase aimed atproviding support to the analyses and understanding the processes under way involving the PAE Lago Grande. In this phase,we built a geographic database with spatially explicit information about soils, vegetation, rivers, relief, communities,infrastructure, land use and social conflicts. Besides, we organized information on socioeconomic and demographicindicators, rural and environmental legislation, official documents related to the area, and minutes of civil societyorganization meetings.

Once the communities were defined, we actually started planning the workshops and selecting the participants. Wecarefully defined which and howmany of the participants would attend the workshops, and what would be the best way tomobilize them, in order to ensure that all social segments and interest groups were represented in each one of the studylevels (Table 1). For each workshop, on each level, we planned to receive about 30 participants. In the communities’workshops, field researchers had previously determined interest groups and existing community organizations.Subsequently, inside each of them, participants were chosen by the community leaders. We scheduled the workshopsmore than a month in advance, and days before the scheduled date, a research teammember would go into the community,and with the help of the community association, set up preliminary meetings to explain better the objectives and stages ofthe proposedworkshops. At the settlement level, we invited the main governmental and civil society organizations active inthe settlement, FEAGLE (Federation of Lago Grande Settlement Communities Associations) directors, as well as some localleaders fromother settlements areas selected by FEAGLE itself. At the settlement level, the communities’ representatives andFEAGLE compiled together the definitions of desired and non-desired futures. Both governmental and civil society

Table 1List of the workshops carried out.

Workshops Scale Participants/players Date Period

1st Settlement FEAGLE Board of Directors Oct/2009 1 dayPAE NGOs and civil society

Governmental organizations

2nd Settlement FEAGLE Board of Directors Nov/2009 2 daysPAE Representatives of the five regions of PAE Lago Grande: (1) higher,

(2) medium e (3) lower Lago Grande, (4) Arapixuna, (5) Arapiuns

3rd Aracy Community Community representatives (members e leaders) Mar/2010 3 days4th Vila Brasil Community Community representatives (members e leaders) Apr/2010 3 days5th Settlement FEAGLE Board of Directors May/2010 1 day

PAE Representatives of the five regions of PAE Lago Grande

6th Settlement FEAGLE Board of Directors May/2010 1dayPAE Representatives of the five regions of PAE Lago Grande

NGOs and civil society (Table S1)Governmental organizations (Table S1)

7th Retiro Community Community representatives (community members and leaders) Jan/2011 2 days

R.T. Folhes et al. / Futures 73 (2015) 86–99 91

organizations (Table S1) participated in the trajectory construction phase—the steps to attain the desired future. At this level,an initial workshopwas also carried out with the organizations in order to diagnose each of the activities they were engagedin, and confirm their commitment in the scenarios’ process.

As Table 1 shows, we organized seven workshops. The length of each workshop depended on logistics and participantavailability. But in each community and at the settlement level, all the stages were executed, most of them in a singlemeeting, sometimes broken in more than one (past/present, present/near future, desired/undesired future, elaboration oftrajectories). At the beginning of theworkshops, we invited the participants to discuss and approve the structure of the workand the relevance of the selected themes. In PAE Lago Grande, the information presented at the first stage (past/present) wasbased on interviews carried out during the primary data collection stage mentioned above. At the community scale, giventhe difficulty to access to the area (mainly by boat) and the socioeconomic context (no access to supermarket or electricity,for instance), we decided to use simple and inexpensive material to support the workshops in the communities. Therefore,instead of images and collages as proposed in similar participatory scenario studies (Kok, Patel et al., 2006; Kok, Rothmanet al., 2006), we used brown paper stapled on the walls on which participants’ contributions were organized in a syntheticmanner, bymeans of keywords. At the settlement level; theworkshops took place in the city of Santarém. A larger number ofparticipants were expected and laptops and data show resources were available. The participants themselves came with aninteresting addition to the method: in the second community (Vila Brasil); the participants had the idea to stage a theaterplay illustrating the contrasting visions for the rest of the inhabitants; especially for the children; arguing that being aware ofit was important for everyone. The playwas performed at night in front ofmore than a 100 people (that is a lotmore than the30 participants of the workshop). It was a powerful mechanism to increase the awareness and the appropriation of thestories being created; andwe incorporated it into our approach; and successfully repeated it in the other communities. Fig. 2presents pictures of the workshops; illustrating the material organization into tables.

After the workshops, we organized the results into separate reports (one for each community and one at the settlementlevel), containing a brief description of the methodological approach utilized, as well as a synthesis of the results of theprocess (transcription of the tables filled during theworkshops). Afterwards, wewent back to each community to handle thereports during shortmeetings. At the settlement level, FEAGLE directors and all organizations participating in theworkshopsreceived similar reports. By forwarding the reports, we aimed at supporting the execution of debates about the outcome ofscenarios within the organizations existing in the two levels, thus providing a basis for them in final debate. The final debatewas held in Santarém city; it included representatives of all communities and organizations at the settlement level that hadparticipated in the previous workshops. Convergences and divergences of scenarios produced in the different levels wereanalyzed and systematized. Subsequently, communitymembers and FEAGLE directors came upwith questions to be debatedby governmental organizations representatives having decision-making power on the PAE Lago Grande creation process.

3.2. Research area

PAE Lago Grande comprises 290,000ha, corresponding to approximately 15% of the area of the Santarém municipality(Fig. 3a). This land is home to a population of approximately 30,000 people, organized into 143 small villages, denominated“communities”, that emerged in the 1970s from the evangelizationwork of the Catholic Church around a community of faith(Araújo, 2009; Stoll & Folhes, 2014). The communities were reinforced in the 1980’s by the Union of rural workers, whosecharacteristic was to be politicized and able tomobilize the peasant masses for the accomplishment of a fair agrarian reform(Leroy, 1991).

[(Fig._2)TD$FIG]

Fig. 2. Pictures illustrating the workshops at different scales.

92 R.T. Folhes et al. / Futures 73 (2015) 86–99

3.2.1. Socio-environmental characterizationThe Lago Grande region is an old occupation areawith significant ecosystemvariability. It comprises areas of firm ground,

riverside land, várzea (seasonally overflowed low-lands by the Amazon river), and savannah where the human occupationprocess was more intense, particularly after the migration flows occurring as from the 19th century, long before the PAEagrarian project was established. Nevertheless, at the end of the referred century, numerous land titles had already beenproduced (Muniz, [269_TD$DIFF]1909a,b), mainly located in the várzeas’ lands – among which Lago Grande (Fig. 3b) –where the culture ofcacao was prolific.

In the decade 1940–1950, three economic activities were relevant in Lago Grande area: livestock, jute and fishing. Duringthe 1960s therewas an increase in the raising of small bovine herds by family units of riverside and prairie communities. Thecurrent land uses at PAE Lago Grande are mainly extractivism (collection of natural products from animal, vegetable ormineral origin), animal rearing and manioc cultivation for flour preparation. These activities basically aim at auto-subsistence, and small surpluses are sold; cattle breeding andfishing being the activities of prominent economic importance.

The climate in the region is equatorial, characterized by a short drought period varying from August to October and aconcentrated rainy period from February to May. The main access to PAE Lago Grande is by boat on the river. Land access ispossible through State Highway PA 257. However, the poor condition of this road and the high regional rainfall figures lead toconstant blocking of said road during the rainy season. The main vegetation types are rainforests, secondary vegetation,savannas, prairies and floristic contacts between forests and savannas. Recently, forest being converted to pasture graduallyincreased, particularly basedonapartnership systembetween large farmers and communityones. Approximately 22,000haofforest areas (primary and secondary forests) were converted to pasture from 2005 to 2009. Large farmers provide the cattlewhilethesmallonesprovidethe landandthecattlehandling.Thatway, largecattlebreedersmayendupowningover4000headof cattle spread out among the medium and small areas through partnerships, with a great number of community people(Folhes, 2010).

3.2.2. Social organizationFEAGLEwas created in 2005, the same year as the PAE LagoGrande creation. FEAGLE is a civil organization responsible for

representing the settlement in institutional dealings with government agencies and other sectors of civil society. There arebarely three territorial scales reached by the organizations and inhabitants of PAE Lago Grande: community level of whichcommunity associations are the main example of; intercommunity level, including a number of communities gathered in

[(Fig._3)TD$FIG]

Fig. 3. Location of study area (a) and the conflict zones in PAE Lago Grande: mining (b) and land use (c).

R.T. Folhes et al. / Futures 73 (2015) 86–99 93

producers’ and fishermen’s associations, or in parish activities; and the level involving the whole settlement, including theUnion of Santarém RuralWorkers (STTR), the Fishermen Union (Z20) and FEAGLE itself. Each one of those levels represents adecision-making sphere, one not always in agreement with the others.

From the 70s, STTR was one of the first organizations to act in most of the Lago Grande communities. At the beginning,STTR dedicated itself in recruiting local families to join the Union, so that they might enjoy labor law benefits. Thereafter itplayed an important role in social mobilizations, with a view to regularize land tenure in the region. Another organization,Z20, started in the 80s, with the emergence of fishing conflicts, and contributed to organize the fishing activity and thefishermen Union movement. The Catholic Church also had a fundamental role in the social organization process, startingback in the sixties, having influenced the formation of most of the communities (Araújo, 2009; Stoll & Folhes, 2014). Untiltoday, the Church is close to the main leaders of the communities, who gather quite often at religious intercommunity andregional events. Evangelic churches are also present in most communities today, but they are not active in political issues. Inthe villages, public fundamental education is available for all the families. Many of them receive benefits from the federalgovernment for each child enrolled at the school (programa bolsa família). INCRA provides credit for basic infrastructure andhousing, but the issues of land tenure rights and technical assistance had not been solved at the time we organized theworkshops (see Section 3.2.3). Some NGOs promote projects in specific communities, aiming at stimulating alternativesustainable activities.

3.2.3. Socio-environmental conflictsThe PAE Lago Grande sits on abundant deposits of bauxite that were already prospected and which are today included in

the mining firm Alcoa’s production strategy. In 2009, a conflict emerged when the mining company installed a base withworkforce and machines in a certain zone of the settlement (Fig. 3c). A few months later FEAGLE organized publicconsultancies in more than 60 communities of the settlement in order to document their position Alcoa’s activities in thearea. All communities took positions against mining exploitation. At that point, FEAGLE and the communities consistentlydenied Alcoa access to the settlement and authorization to carry out prospectingworks. As a result, in November 2010, Alcoastarted a judicial process that would ensure the mining prospecting activities, such as planned in the Brazilian Mining Code.

94 R.T. Folhes et al. / Futures 73 (2015) 86–99

Besides these external pressures, the intrinsic weakness of the land situation is a weakness of PAE Lago Grande itself.About 6000 individual land title attribution processes exist in the settlement. The majority of these processes are a result ofthe demarcations promoted by INCRA over recent decades. This tricky situation explains that the crucial document thatwould legally secure the lands of the families living in the settlement – the Real Use Right Concession (CRU) – has not yetbeen issued by INCRA.

Another difficulty of the land regularization process is due to a localized case of overlapping between the PAE and anindigenous land demarcation process currently ongoing by the Indian National Foundation (FUNAI). Indeed, recent internalconflicts in communities located near Aracy community induced the active participation in the regional indigenousmovement of local families who claim an indigenous identity and the demarcation of Cobra Grande Indigenous land (Stoll,2014). Due to those uncertainties, the first support-credits (a credit for supporting basic tools to improve the familyproduction) promised by the government were only released in 2010, and scarcely distributed in 2011, along with the“housing credits”.

In summary, thousands of land titles had been historically given and many processes are being judged in local andregional complex contexts of conflicts. Inside the settlement about 140 communities coexist with large cattle breeders,mining and timber companies’ interests, identity claims, social movements, various land situations and divergent interests,including activists of the separatist movement for the creation of the Lago Grande municipality. Yet, five years after thepublication of the ordinance creating PAE Lago Grande, great uncertainties remain about the viability of the settlement, andabove all, about the effective guarantee of territorial rights for the local populations.

3.2.4. Characterization of the communitiesIn this context, the three selected communities are located in different regions of the settlement (Fig. 3) and present

specific socioeconomic and biophysical characteristics.Aracy is located on the borders of LagoGrande. The community is composed of about 90 families and 384 inhabitantswho

occupy solid ground areas and prairies, not having access to primary forest areas. Community secondary forests are often cutdown to be used as an energy source for the brickyards in Vila Socorro, a neighboring community. Activities dedicated tosustenance of basic needs of the families prioritize fishing, animal husbandry, manioc flour production and extractivism.Aracy’s dwellers follow closely the process of creation of the Cobra Grande Indigenous Land, which area, claimed by aneighboring group, overlaps part of their territory.

Vila Brasil is located on the banks of the Arapiuns River. This village is home to 84 families and approximately 400 people.The land cover is a composition of primary and secondary forest areas where manioc fields appear scattered. Manioc flourproduction and extractivism are the main activities. Animal husbandry never consolidated as a major activity, and fishing,although fundamental to the daily diet, has no commercial relevance. In Vila Brasil, a serious land dispute opposes theinhabitants with a former community member who claims ownership of a large part of the vila, or, in other words, thecommunity headquarters. The vila refers to the central place of the village, where the majority of the dwellers build theirhouses. This place is supposedly public and belongs to the “community” (to the collective). The vila hosts the publicinstitutions such as the churches, the school, the health center, the radio center, etc. (Stoll & Folhes, 2014).

Finally, Retiro is located alongside PA 257 road, at thewestern edge of PAE Lago Grande, on the border between Santarémand Juruti municipalities, in an area less populated than Vila Brasil and Aracy. It accounts for 24 families and approximately104 people. More than a half of the community area is covered by primary forest, having rules of use agreed with thecommunity people. Main agricultural crops are manioc, maize and watermelon, while fishing is practically inexistent.However, extractivism and hunting aremuch engaged in. There is no cattle breeding, except for some areas, which are leasedto prairie cattlemen during high flood years. Due to the proximity to Juruti city, and the bauxite-rich subsoil, the communityhas a long history of relationship, often inharmonious, with the mining firm Alcoa.

4. Results of the scenario process

In this section, we present selected parts of the material produced during the workshops to support the discussion aboutthe method. The contrasting narratives in Table 2 provide a general understanding of the desirable and undesirable future atboth scales. Table S2 presents a synthesis of the main issues discussed in Vila Brasil regarding projects/land use andinfrastructure themes. Table S3 summarizes the social organization theme discussions for Aracy and Table S4 presents theConflict theme as discussed at the settlement level.

Debates that took place during the workshops revealed both consensus and divergences inside the communities andacross levels. In some cases, as a result of the trajectory phase, the participants planned meetings with organizationrepresentatives as a first step to the desired future. For many of the participants, the process of elaboration of scenarios wasthe first integrated discussion hold by the five regions of PAE Lago Grande since its creation. The key convergences anddivergences in both levels are summarized below.

4.1. Community level

In spite of the individuality of each community, similar concerns and dreams lead to similar desired and undesiredfutures. At the community level, differences betweendesired and undesired futures liesmainly in the successful development

Table 2[14_TD$DIFF]Example of narratives written by the stakeholders during the workshops ( [15_TD$DIFF]stage 3: desired/undesired future) synthetizing the visions about the desired andundesired future.

Desired [16_TD$DIFF]future (vision 2020) Undesired [17_TD$DIFF]future (vision 2020)

Vila Brasilcommunity

"Vila Brasil, January 20, 2020. Today Vila Brazil has 95 [18_TD$DIFF]ha of landand 160 families living in the community. There is reforestationwith various qualities of native and fruit trees. We have brickhouses, running water and energy of good quality. We have a largeschool from fundamental to high school; residents have theircreation of fish and chickens and the community has its collectivefish farming. We have a community radio with daily programs.Today the morning program radio is announcing that ourcooperative is helping us expand our production of fish and fruitpulp. And telling people not to forget that there will a soccer gamebetween the National and North tonight at the “Kamelão”Community Stadium. In the early evening program, the speakerEdno announces that there will be visited by a group of touristswhowill be staying in one of the hostels. The radio also reports thatthe Doctor Pedro will be in health clinic serving the people ofneighboring communities after eight in the morning. And theprogram "The Voice of the People" will have an interview with thedelegate Luis Carlos who will be talking about safety in ourcommunity. Today, Vila Brazil is a well-organized and structuredcommunity, with a lifestyle better than ten years ago."

"By 2020, Vila Brasil has electric energy 24 h [19_TD$DIFF], but without landtenure regularization of the settlement, one person managed totitle most of the land as private land, removed the others, and afterthat sold the land to Alcoa. With the arrival of Mining, smallfarmers who remained were forced to sell their properties due tothe finance pressure. Thinking about the future of the children, thesmall farmer ended up buying a house in town. Relying on moneythat he actually does not have, without formal work in town, withtime passing he is indebted, and sells everything to pay their debts.The community would no longer exist. The associations woulddisappear."

Retiro community “[....] Several patients were treated at the Maria Dias da SilveiraMedical Center of the community, by doctors specialized indifferent areas. On the same day, a study was carried out in theforest of the community reserve: the trees were found wellpreserved, with many kinds of animals; the sources of the streamspresented crystalline and abundant water; there were differentmedicinal plants. Hardwood trees were protected and no trace ofburnt areas was found inside the reserve.”

“[ . . . ] Community without forest started mining extraction in theCaranã hill. A railroad was built cutting through the village, andthence, the disappearance of the ten main sources, as well as thedisappearance of the animals and the contamination of the waterof the few remaining streams. Due to the implantation of themining company, a migratory process of foreign workers arose,causing a rapid population growth and tensions around landoccupation.”

Settlement “In 2020, communities and their members shall be capacitated toexercise dedicated leadership positions, aware of their real identityof workers, enjoying adequate living conditions [ . . . ], treasuringthe environment and enjoying good health to be able to carry outservices together and without conflict, having the CRU in hand,knowing in depth the territory where they live and work.”

“If our dreams do not materialize, everything, fishes, grass orforests, community dwellers and organizations will cease to exist.FEAGLE [20_TD$DIFF]’s head office will be manipulated by opportunists that willforsake the families, suffering from lack of credit, health, road,transports and enduring poor living conditions, prostitution,alcoholism and drug abuse.”

R.T. Folhes et al. / Futures 73 (2015) 86–99 95

of new economic activities, capable of generating income and keeping the young people in the community. The threecommunities’ desired scenarios envision a greater diversification of land uses, with a verticalization of production (anaggregation of value on the product) and an adoption of adapted technologies, aswell as the increase of activities not directlyconnected with the land, such as fish-farming and tourism. In the desired scenario, they also envision a reduced impact onprimary and secondary forests and, in fact, an increase in forest areas through regeneration and planted species. Therewouldalso have less pressure on the land for agricultural use in 2020, because a larger number of families would be engaged in theservice sector and in the management of the organizations that would have arose.

In the undesired scenario, territorial conflictsmight render unfeasible the existence of communities in 2020, and the lack ofalternatives would lead to the depletion of natural resources and take people away from their land. In this scenario, theexistence of communities is challenged by privatization of the land, due to the property right issues and by the Alcoaminingcompany interest in the area. We also find the feature of the selfish individual from the inside that makes a deal with apowerful outsider. In a broader sense, these narratives echo the ambivalent and painful historical relation betweenAmazonian peasantries, the State (which is seen as a provider of benefits but also as an outsider incapable of – or notinterested in – fulfilling its obligations) and the “rich outsider” which speculate the land (from colonists to miningcompanies).

The trajectory of the desired future, as discussed during theworkshops, relies on a series of innovations needed to face thepublic policies implementation in the settlement and the envisioned context of production diversification. Socialorganization strengthening and a more efficient governmental action to solve the conflicts are seen as the key points tocounteract the insider/outsider forces described above. The participants reinforced the need for improvement in themanagement of community organizations, high-level education and in their interlocution with the outside world. Thecapacitating process, including managerial, technical and educational formation, would avail the community members ofthe diversification of economic activities and development of collective strategies.

96 R.T. Folhes et al. / Futures 73 (2015) 86–99

4.2. Settlement level

At the settlement level, the participants also considered that the strengthening of social organizations was the mostimportant action to reach the desired future. To that end, they discussed strategies, such as capacitating youths and leaders,and increasing their management skills. In the trajectory they constructed, the search for land use alternatives is alsofundamental, and to a great extent based on agrarian reform credits (credits and government subsidies prescribed by theBrazilian law to be assigned to rural settlements such as PAE LagoGande) granting an execution of awide technical assistanceprogram, adapted to interregional differences inside the settlement. Along the way investment in highways infrastructure,power and communication would be fundamental for verticalization and distribution of production, as would be, later, thedevelopment of business plans and management models of collective projects.

An undesired scenario shows that from the stakeholders’ point of view the collapse of PAEwould lead LagoGrande regionto exhaust its natural resources, to deforestation and breakdown of its social organizations.

4.3. Convergences and divergences between scales

Concerning the two levels analyzed, we understand that strategies for attaining the desired future must originate fromthe strengthening of settlement organizationsmanagement capacity. Strategies defined as a prioritywere the investment onhuman capital through improved Primary and High School education in the communities allied to the offer of managerialand technical schooling, as well as university courses, in the main settlement villages. Production diversification,intensification and verticalization also appear as converging strategies in the two levels, aiming at increased productionassociated to preservation of forest areas. However, there was always a sense of “how dowe do this?" at the last stage of theworkshops, due to the low level of “intervention” power of most actors regarding many of the issues. Hence, strengtheningthe social organization arose repeatedly as a priority.

A shared concern existed on both levels about the mining companies’ plans to exploit minerals in the area’s rich subsoil.Both discussed the need for a better organization to prevent the companies from entering the area. However, considering theBrazilian legislation applicable to subsoil, if the largemining company entering is inevitable, participants also discussed theneed to organize better themselves to minimize the mining effects on community life and its impacts on natural resources.

However, in the strategies defined for the resolution of land tenure conflicts, important divergences emerged. Thecomplex agrarian situation of the area precludes INCRA from issuing the document (CRU) to regularize the land tenuresituation, providing legal security to families living in the PAE. One possible solution discussed during the workshop at thesettlement level was to achieve granting of right of use for the conflict-free areas. Although this would seem to be a feasiblepath to solve partially land problems of part of the area initially destined to PAE, Aracy and Vila Brasil, for instance, would beoutside the area included in the use grant.

More broadly, the fear of the end of the communities and of the settlement itself, due to external or internal economicpressure leading to the privatization of the land, was a unifying theme of the undesired scenarios. This theme links the PAELago Grande to the threats to other units in the region (see for instance, Ferreira et al., (2014), and to the process of landconcentration and urbanization in the region and in Brazil (Martinelli, Naylor, Vitousek, & Moutinho, 2010). It is alsoimportant to notice how this issue dialogs with the global land grabbing literature (Borras,[270_TD$DIFF] Hall, Scoones, White, & Wolford,2011 [271_TD$DIFF]), turning the regions into “dead reserves” of natural resources for global capitalism.

4.4. Method discussion

In this section, we stress the main positive aspects and the limits of the scenario method we adopted. One of theinnovations of our method was to initiate a discussion about the future in two stages – near future and distant future –

favoring a distinction between what was more to happen soon, and what might make up the future in 2010, significantlyhelping along the discussion regarding trajectories. Discussion about intervening power of stakeholders enables analysisbased on the very understanding, by stakeholders and players, of the capacity that each organization has to act on the giventheme under discussion, and the institutional and organizational arrangements, which should – in theory – be carried out toreach the common objectives.

Rather than attestingwhether the desired futureswill come true following the trajectories defined during theworkshops,this work focuses on how the whole scenario process provides a structured method to discuss necessary actions to reach acommon desired future, opting for a clear normative approach that favors engagement and buy-in of the scenarios. In thecommunities, after the completion of the last workshop stage, community members scheduled meetings to present anddebate with other community members the results obtained, even though they still did not have in hands the final reportsdrawn up by the researchers. At the settlement level, the last workshop produced new agendas on territorial planning of thesettlement, established among participating organizations. We consider this as one of the most positive outcomes of theproject.

The multi-scale approach allowed representatives of organizations that act at the settlement level to constructtrajectories based on desires and fears of stakeholders from the communities, as both levels were discussed during the lastworkshops. It also allowed similar problems between scales to become visible, such as the urgent need for land tenureregularization. At the same time, it showed that some trajectoriesmay be relevantwith regard to one level but not to another,

R.T. Folhes et al. / Futures 73 (2015) 86–99 97

such as the case involving grant of land use to settlement areas free from land-use conflicts, as discussed in Section 4.3. Themulti-scale approachwe used allowed the incorporation of thesemultiple perspectives to the scenario construction process,increasing their credibility and relevance through the combination of a larger range of visions (Alcamo & Ribeiro, 2001).

Like most participatory approaches, methodology applied to this work reached its most important point during debateshappening in collective workshops. In these participatory exercises, products and implications of narratives produced wereless important than the sharing of knowledge, the creation of visions and the empowerment that occurred during thescenarios development process (Wollemberg et al., 2000). However, the workshops required a team of researchers formediation, annotation and systematization of the debates. Therefore, aside from costs involved in the preparation ofworkshops themselves, expenses of team deployment and upkeep are quite high. In some cases, this may render unfeasiblethe use of workshops in a larger number of communities, and capture better differences inside a given territorial unit.

Mediation and systematization of workshop process refers to the role played by involved researchers. Duringmediations,two basic approaches can be conducted. According to the first, researchers should lead the workshop with a minimum ofcritical intervention as regards the topic being discussed, only trying to give fluidity to the debate, keeping it within theproposed discussion focal points. According to the second, the team intervenes often, giving opinions and introducingthemes to be analyzed by the group. In this research, the team was careful enough during the workshops to intervenecritically as little as possible, unless when requested by stakeholders to offer its opinion. The same concern was appliedduring the compilation of narratives so that the latter do not become charged with the subjective vision of researchers. Tominimize this problem, researchers may carry out together an internal review of the workshop results analyses and of textsgenerated. In this respect, the role of scientists within the normative/propositional objective of scenarios may besummarized as follows: (a) to mediate; (b) to organize results; (c) to report and discuss results; (d) to stimulate so thatactions may have continuity.

It is fundamental that the mediation team should know the specificities of the social groups involved in the scenarioselaboration procedures, as well as the political context in which they evolve. This contributes to a better dialogue betweenresearchers and stakeholders and, above all, to a perception of adaptations to bemade in themethod and in the participatoryinstruments employed in the approach.

In each community, we observed that the internal power structure was based on a non-democratic leadership of therepresentatives of a strong determined familiar segment that acted such as a faction (cf. Stoll & Folhes, 2013, 2014; Stoll,2014). It may represent an obstacle to the progress of the scenarios discussion, and to the proper accomplishment of theactions to be carried out to reach the desired future referred to. Indeed, during theworkshops, not everyone could easily andreadily express himself; there were differences in the degree of participation between those holding leadership positionsand the others. Community leaders aremore accustomed to participate in eventswhere their opinions are required. Becausethey aremore articulated, if due care is not taken, theymay lead communityworkshops so that scenarios becomemore theirown than collective ones. This social configuration increases the workshop mediator’s responsibility to involve allparticipants in an endeavor not to let false consensus come out as a position assumed by the group, and to keep allparticipants motivated in debating the proposed discussion issues (about power asymmetry in scenarios, see also Dufva &Ahlqvist, 2015).

To debate the future, the group should always be questioned regarding which collective trajectoriesmaymake it possibleto reach a desirable future. This made us realize how complex is the relation between the desire of innovations expressed bythe members of the communities and the factional power structures mentioned above that may eventually limit theefficiency of such projects. Therewas actually a great gap between the real action of the leadership inside the community andthe possible conduct of external projects, thought of as democratic and participative and proposed by a network of localpartners (Government, NGOs, Rural Workers Union, etc.). We have stressed here one of the main limits of the scenariomethodwe adopted. This does notmean that the process is not useful, as it brings important issues to be openly discussed ina structured format by heterogeneous actors at different scales. Nevertheless, the actual effectivity of the participatorymethod relies very much on a clear understanding of the local power structures, which may undermine the effort.

5. Conclusions

The outcome of this research showed that the methodology adopted here has great potential for replication in otheragrarian reform settlements and in units of sustainable use and preservation, in both levels studied. Through themulti-scalemethod, the threat of the disappearance of such units and population out-migration – in face of powerful global and regionaleconomic interests – emerged as a unifying theme, linking the local communities’ problems to a much larger global trend.

In the application of the method, it should not be overlooked that historical processes and heterogeneous conjecturesinfluence individual and collective decision-making processes regarding land use and adhesion or rejection to a given publicpolicy. Theworkshops would evolvewithin the pre-established universe of the communities, where the influences of real orclassificatory kinship must be stressed in the covering up of contrary positioning between relatives, neighbors or compadres(relation between one person and his/her godchild’s parents). Therefore, to express disagreements in the presence ofoutsiders (we, researchers) could have questioned and threatened these kinship bounds on which, very often, peasantgroups depend for thematerial and social reproduction of community life, within the ins and outs of deliberations inside the“community”. This reinforces the importance of knowing specificities of social groups involved in scenarios elaborationprocedures, and the importance that a multi-disciplinary vision may have in this process.

98 R.T. Folhes et al. / Futures 73 (2015) 86–99

More broadly, the implications of scenarios studies at the science/decision-making interfacemust be examined in depth.Interesting questions for the future research work include, for example, how does the relation between scenarios anddecision-makers (either involved or not in the building thereof) work? How and under what circumstances do the processand results of scenarios contribute to stakeholders, in terms of social empowerment and learning? How should weunderstand the outcomes of the local power structure in a participatory scenario processes?

Finally, the scenario approachweproposed and applied in this paperwas purely qualitative. Quantitativemodels could beused complementarily to represent (and test the feasibility) of the visions and trajectories. The challenge is to selectmodeling approaches (An, 2012; Auvinen et al., 2015; Becu, Neef, Schreinemachers, & Sangkapitux, 2008; Voinov &Bousquet, 2010) capable of capturing the complex political and institutional context discussed in the participatory multi-scale process.

Acknowledgements

For their supports, wewould like to thank all of PAE LagoGrande communities’members and institutions’ representativesthat took part in the scenarios. We are also grateful for our colleagues at the National Institute of Spatial Research (INPE) inparticular Patricia Pinho, Pedro Andrade and Angelica Giarolla. Special thanks go to Alain Stoll for proof reading the Englishversion of the article. This study was part of a broader research funded by [272_TD$DIFF]FAPESP’s Program on Global Climate Change [273_TD$DIFF]thatwas carried through by the ‘Land Use Change in Amazonia: Institutional Analysis and Modelling at multiple temporal andspatial scales (LUA/IAM)’ project of Brazilian National Institute of Spatial Research (INPE). Finally, we thank the reviewerswho greatly contributed to the clarity and organization of the manuscript.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.futures.2015.08.005

References

Acuña, C. D. (1994). Novo descubrimento do grande rio das Amazonas. Rio de Janeiro: Agir.Alcamo, J. (2008). Environmental Futures: The Practice of Environmental Scenario Analysis. Amsterdam: Elsevier.Araújo, R., 2009. De la Communauté aux populations traditionnelles, Nuevo Mundo Mundos Nuevos. Nouveaux mondes Mondes nouveaux, Novo Mundo

Mundos Novos, New World New Worlds. Available: http://doi.org/10.4000/nuevomundo.56593Alcamo, J., & Ribeiro, T. (2001). Scenarios as tools for international environmental assessments. Environmental issue report No. 24. Germany: European

Environment Agency http://www.google.fr/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CCcQFjAAahUKEwjt4Oqv39rGAhUBVhQKHc55AAA&url=http%3A%2F%2Fwww.eea.europa.eu%2Fpublications%2Fenvironmental_issue_report_2001_24%2Fdownload&ei=9BClVe2WLoGsUc7zAQ&usg=AFQjCNHghDY0BWJ7hHTQBk9W2WNqTPqJOA&bvm=bv.97653015,d.d24, p. 31.

An, L. (2012). Modeling human decisions in coupled human and natural systems: review of agent-basedmodels. Ecological Modelling 229(24), 25–36. http://doi.org/10.1016/j.ecolmodel.2011.07.010.

Araújo, R., & Léna, P. (2010). Da predação à sustentabilidade na Amazônia: a difícil metamorfose. Desenvolvimento Sustentavel e Sociedades na Amazonia.Belém: MPEG pp. 13–55.

Auvinen, H., Ruutu, S., Tuominen, A., Ahlqvist, T., & Oksanen, J. (2015). Process supporting strategic decision-making in systemic transitions. TechnologicalForecasting and Social Change 94, 97–114. http://doi.org/10.1016/j.techfore.2014.07.011.

Börjeson, L., Höjer, M., Dreborg, K.-H., Ekvall, T., & Finnveden, G. (2006). Scenario types and techniques: towards a user’s guide. Futures 38(7), 723–739.http://doi.org/10.1016/j.futures.2005.12.002.

Becker, B. (1997). Amazônia. São Paulo: Ática.Becker, B. K. (2009). Amazonia: geopolıtica na virada do III milenio. Rio de Janeiro: Editora Garamond.Becu, N., Neef, A., Schreinemachers, P., & Sangkapitux, C. (2008). Participatory computer simulation to support collective decision-making: potential and

limits of stakeholder involvement. Land Use Policy 25(4), 498–509. http://doi.org/10.1016/j.landusepol.2007.11.002.Bettendorff, J. F. (2010). Cronica da missao dos padres na Companhia de Jesus no Estado do Maranhao. Brasília: Edições do Senado Federal.Biggs, R., Raudsepp-Hearne, C., Atkinson-Palombo, C., Bohensky, E., Boyd, E., & Cundill, G., et al., (2007). Linking futures across scales: a dialog on multiscale

scenarios. Ecology and Society, 12(1) art. 17.Borras, S. Jr, Hall, R., Scoones, I.,White, B., &Wolford, (2011). Towards a better understanding of global land grabbing: an editorial introduction. The Journal of

Peasant Studies 38(2), 209–216. http://doi.org/10.1080/03066150.2011.559005.Bradfield, R., Wright, G., Burt, G., Cairns, G., & Van Der Heijden, K. (2005). The origins and evolution of scenario techniques in long range business planning.

Futures 37(8), 795–812. http://doi.org/10.1016/j.futures.2005.01.003.Buarque, S. C. (2003). Metodologia e técnicas de construção de cenários globais e regionais. Texto para discussao no. 939. Brasília: IPEA http://www.ipea.gov.

br/portal/images/stories/PDFs/TDs/td_0939.pdf, p. 75.Bunker, S. G. (1985). Underdeveloping the Amazon. Chicago: University of Chicago Press. http://www.press.uchicago.edu/ucp/books/book/chicago/U/

bo3634235.html.Cosgrove, W. J., Rijsberman, F. R., & Council, W. W. (2000). World water vision: making water everybody’s business. London: Earthscan Publications.Dalla-Nora, E. L., de, Aguiar A. P. D., Lapola, D. M., & Woltjer, G. (2014). Why have land use change models for the Amazon failed to capture the amount of

deforestation over the last decade? Land Use Policy: The International Journal Covering All Aspects of Land Use, 39, 403–411.Dufva, M., & Ahlqvist, T. (2015). Knowledge creation dynamics in foresight: A knowledge typology and exploratory method to analyse foresight workshops.

Technological Forecasting and Social Change 94, 251–268. http://doi.org/10.1016/j.techfore.2014.10.007.Ferreira, J., Aragão, L. E. O. C., Barlow, J., Barreto, P., Berenguer, E., & Bustamante, T. A., et al., (2014). Environment and Development. Brazilãs environmental

leadership at risk. Science 346(6210), 706–707. http://doi.org/10.1126/science.1260194.Folhes, R. T. (2007). Uma cartografia da memoria (cartografia participativa). Santarém: Projeto Saude e Alegria.Folhes, R. T. (2010). Cenários de mudanças de uso da terra na Amazônia: explorando uma abordagem participativa e multi-escala no PAE Lago Grande, em

Santarém-Pará. Dissertacao de Mestrado. Belém: UFPA-PPGCA.Giddens, A. (2009). The Politics of Climate Change. Cambridge UK: Polity Press.

God

GomKok

Kok

Kok

Kok

Lap

Lau

J.-PMa

MilMu

Mu

Par

Pat

Qui

Ras

SchSoa

StoSto

Sto

Sw

Ver

Voi

Wil

Wo

de A

van

van

van

R.T. Folhes et al. / Futures 73 (2015) 86–99 99

et, M. (2000). The art of scenarios and strategic planning: tools and pitfalls. Technological Forecasting and Social Change 65(1), 3–22. http://doi.org/10.1016/S0040-1625(99).es, D. M. C. (2007). The diversity of social forms in pre-colonial Amazonia. Revista de Arqueologia Americana, 25, 189–225., K. (2009). The potential of fuzzy cognitivemaps for semi-quantitative scenario development,with an example fromBrazil.Global Environmental Change19(1), 122–133. http://doi.org/10.1016/j.gloenvcha.2008.08.003., K., Biggs, R., & Zurek, M. (2007). Methods for developing multiscale participatory scenarios: insights from Southern Africa and Europe. Ecology andSociety, 13(1) art. 8., K., Patel, M., Rothman, D. S., & Quaranta, G. (2006). Multi-scale narratives from an IA perspective: Part II. Participatory local scenario development.Futures 38(3), 285–311. http://doi.org/10.1016/j.futures.2005.07.006., K., Rothman, D. S., & Patel, M. (2006). Multi-scale narratives from an IA perspective: Part I. European andMediterranean scenario development. Futures38(3), 261–284. http://doi.org/10.1016/j.futures.2005.07.001.ola, D. M., Schaldach, R., Alcamo, J., Bondeau, A., Koch, J., Koelking, C., & Priess, J. A. (2010). Indirect land-use changes can overcome carbon savings frombiofuels in Brazil. Proc. Natl. Acad. Sci. 107(8), 3388–3393. http://doi.org/10.1073/pnas.0907318107.rance,W. F., Cochrane, M. A., Bergen, S., Fearnside, P. M., Delamônica, P., & Barber, C., et al., (2001). The future of the Brazilian Amazon. Science 291(5503),438–439. http://doi.org/10.1126/science.291.5503.438., Leroy (1991). Uma chama na Amazonia. Rio de Janeiro: Vozes/FASE Editora.rtinelli, L. A., Naylor, R., Vitousek, P. M., & Moutinho, P. (2010). Agriculture in Brazil: impacts, costs, and opportunities for a sustainable future. CurrentOpinion in Environmental Sustainability 2(5–6), 431–438. http://doi.org/10.1016/j.cosust.2010.09.008.ler, R. (2007). Futures literacy: A hybrid strategic scenario method. Futures 39(4), 341–362. http://doi.org/10.1016/j.futures.2006.12.001.niz, J. de P. ([276_TD$DIFF]1909a). Indice dos tıtulos de terras expedidos pelo governo do estado na Administracao do Exm. Snr. Dr. AugustoMontenegro: 1901–1908. Belém: Secretária de Estado de Obras Públicas, Terras e Viação.niz, J. de P., & de, J. P. ([278_TD$DIFF][279_TD$DIFF]1909b). Indice geral dos registros de terras: descritivo das freguesias, vol. T.6, Belém: Secretaria de Estadode Obras Publica Terras e Viação.son, E. A. (2008). Useful global-change scenarios: current issues and challenges. Environmental Research Letters 3(4), 45016. http://doi.org/10.1088/1748-9326/3/4/045016.el, M., Kok, K., & Rothman, D. S. (2007). Participatory scenario construction in land use analysis: an insight into the experiences created by stakeholderinvolvement in the Northern Mediterranean. Land Use Policy 24(3), 546–561. http://doi.org/10.1016/j.landusepol.2006.02.005.st, J., & Vergragt, P. J. (2004). Backcasting for industrial transformations and system innovations towards sustainability: relevance for governance?Governance for industrial transformation. Proceedings of the 2003 Berlin conference on the human dimensions of global environmental change 409–437.kin, P., Monks, F., Ribeiro, T., Van Vuuren, D., & Zurek, M. (2005). Global scenarios in historical perspective. Ecosystems and Human Well-being: Scenarios.Findings of the Scenarios Working Group. Washington: Island Press http://www.millenniumassessment.org/documents/document.326 aspx.pdf, pp. 35–44.mink, M., & Wood, C. H. (1992). Contested frontiers in Amazonia. New York: Columbia University Press.res-Filho, B. S., Nepstad, D. C., Curran, L.M., Cerqueira, G. C., Garcia, R. A., & Ramos, C. A., et al., (2006).Modelling conservation in the Amazon basin.Nature440(7083), 520–523. http://doi.org/10.1038/nature04389.ll, E. (2014). Rivalités riveraines: territoires, stratégies familiales et sorcellerie en Amazonie brésilienne. These de doctorat. Paris: EPHE.ll, E., & Folhes, R. T. (2013). Frères ennemis: la participation à l’épreuve des factions en Amazonie brésilienne. Cahiers Des Ameriques Latines, 72–73,141–160.ll, E., & Folhes, R. T. (2014). La (dés) illusion communautaire: de l’ambivalence de la notion de communauté en Amazonie brésilienne. Journal de La SocieteDes Americanistes, 100(2), 73–113.art, R. J., Raskin, P., & Robinson, J. (2004). The problem of the future: sustainability science and scenario analysis. Global Environmental Change 14(2),137–146. http://doi.org/10.1016/j.gloenvcha.2003.10.002.gragt, P. J., & Quist, J. (2011). Backcasting for sustainability: introduction to the special issue. Technological Forecasting and Social Change 78(5), 747–755.http://doi.org/10.1016/j.techfore.2011.03.010.nov, A., & Bousquet, F. (2010). Modelling with stakeholders. Environmental Modelling & Software 25(11), 1268–1281. http://doi.org/10.1016/j.envsoft.2010.03.007.kinson, A., & Eidinow, E. (2008). Evolving practices in environmental scenarios: a new scenario typology. Environmental Research Letters 3(4) 1), 7. http://doi.org/10.1088/1748-9326/3/4/045017.llemberg, E., Edmunds, D., & Buck, L. (2000). Anticipating change: scenarios as a tool for adaptive forest management. Indonesia: Center for InternationalForestry Research. http://www.cifor.org/publications/pdf_files/SCENARIO.pdf.guiar, A. P. D., Assis, T. O., Folhes, R. T., Dalla-Nora, E. L., Araújo, R., & Alves, D. (2014). Modelagem e cenários para a Amazônia Ambiente e Sociedade naAmazônia. Uma abordagem interdisciplinar. Belém: MPEG100–134.Asselt Marjolein, B. A., & Rijkens-Klomp, N. (2002). A look in the mirror: reflection on participation in integrated assessment from a methodologicalperspective. Global Environmental Change 12(3), 167–184. http://doi.org/10.1016/S0959-3780(02)12-2.Notten, P. W. F., Rotmans, J., van Asselt, M. B. A., & Rothman, D. S. (2003). An updated scenario typology. Futures 35(5), 423–443. http://doi.org/10.1016/S0016-3287(02)90-3.Vliet,M., Kok, K., & Veldkamp, T. (2010). Linking stakeholders andmodellers in scenario studies: the use of fuzzy cognitivemaps as a communication andlearning tool. Futures 42(1), 1–14. http://doi.org/10.1016/j.futures.2009.08.005.