A social and ecological assessment of
tropical land uses at multiple scales:
the Sustainable Amazon Network
Toby A. Gardner1,2, Joice Ferreira3, Jos Barlow2, Alexander C. Lees4, Luke Parry2,
rstb.royalsocietypublishing.org Ima Célia Guimarães Vieira4, Erika Berenguer2, Ricardo Abramovay5,
Alexandre Aleixo4, Christian Andretti6, Luiz E. O. C. Aragão7, Ivanei Araújo4,
Williams Souza de Ávila8, Richard D. Bardgett2, Mateus Batistella9, Rodrigo
Anzolin Begotti10, Troy Beldini11, Driss Ezzine de Blas12, Rodrigo Fagundes Braga13,
Research Danielle de Lima Braga13, Janaı́na Gomes de Brito6, Plı́nio Barbosa de Camargo14,
Cite this article: Gardner TA et al. 2013 A Fabiane Campos dos Santos11, Vı́vian Campos de Oliveira6, Amanda Cardoso
social and ecological assessment of tropical Nunes Cordeiro15, Thiago Moreira Cardoso3, Déborah Reis de Carvalho13, Sergio
land uses at multiple scales: the Sustainable
5 16 10
Amazon Network. Phil Trans R Soc B 368: André Castelani , Júlio Cézar Mário Chaul , Carlos Eduardo Cerri , Francisco de
20120166. Assis Costa17, Carla Daniele Furtado da Costa15, Emilie Coudel3,12, Alexandre
http://dx.doi.org/10.1098/rstb.2012.0166 Camargo Coutinho18, Dênis Cunha16, Álvaro D’Antona19, Joelma Dezincourt4,
Karina Dias-Silva20, Mariana Durigan10, Júlio César Dalla Mora Esquerdo18,
One contribution of 18 to a Theme Issue
José Feres21, Silvio Frosini de Barros Ferraz10‘Ecology, economy, and management of an , Amanda Estefânia de Melo Ferreira
4,
agroindustrial frontier landscape in the south- Ana Carolina Fiorini22, Lenise Vargas Flores da Silva11, Fábio Soares Frazão13,
east Amazon’. Rachel Garrett23, Alessandra dos Santos Gomes4, Karoline da Silva Gonçalves4, José
Benito Guerrero24, Neusa Hamada6, Robert M. Hughes25, Danilo Carmago Igliori5,
Subject Areas:
26 17 11
ecology, environmental science Ederson da Conceição Jesus , Leandro Juen , Miércio Junior , José Max Barbosa
de Oliveira Junior27, Raimundo Cosme de Oliveira Junior3, Carlos Souza Junior28,
Keywords: Phil Kaufmann29, Vanesca Korasaki13, Cecı́lia Gontijo Leal13, Rafael Leitão6,
tropical forests, land use, sustainability, Natália Lima15, Maria de Fátima Lopes Almeida15, Reinaldo Lourival30,
trade-offs, interdisciplinary research,
Júlio Louzada13 31 16 3social–ecological systems , Ralph Mac Nally , Sébastien Marchand , Márcia Motta Maués ,
Fátima M. S. Moreira13, Carla Morsello32, Nárgila Moura4, Jorge Nessimian22,
Author for correspondence: Sâmia Nunes28, Victor Hugo Fonseca Oliveira13, Renata Pardini33, Heloisa
Toby A. Gardner Correia Pereira19, Paulo Santos Pompeu13, Carla Rodrigues Ribas13,
e-mail: tobyagardner@gmail.com
Felipe Rossetti10, Fernando Augusto Schmidt13, Rodrigo da Silva11, Regina Célia
Viana Martins da Silva3, Thiago Fonseca Morello Ramalho da Silva5,
Juliana Silveira13, João Victor Siqueira28, Teotônio Soares de Carvalho13, Ricardo
R. C. Solar2,17, Nicola Savério Holanda Tancredi17, James R. Thomson31, Patrı́cia
Carignano Torres33, Fernando Zagury Vaz-de-Mello34, Ruan Carlo Stulpen Veiga35,
Adriano Venturieri3, Cecı́lia Viana4, Diana Weinhold36, Ronald Zanetti13
and Jansen Zuanon6
1Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2 3EJ, UK
2Lancaster Environment Centre, Lancaster University, Lancaster LA1 4YQ, UK
Electronic supplementary material is available 3Embrapa Amazônia Oriental, Travessa Dr. Enéas Pinheiro s/n, CP 48, Belém, Pará 66.095-100, Brazil
at http://dx.doi.org/10.1098/rstb.2012.0166 or 4MCTI/Museu Paraense Emı́lio Goeldi, CP 399, CEP 66040-170, Belém, PA, Brazil
5
via http://rstb.royalsocietypublishing.org. Faculdade de Economia, Administração e Contabilidade, Av. Prof. Luciano Gualberto, Cidade Universitária,
05508-010, São Paulo, SP, Brazil
& 2013 The Author(s) Published by the Royal Society. All rights reserved.
6INPA. Instituto Nacional de Pesquisas da Amazônia. Av. André Araújo, 2.936 -
present human use; (ii) a nested sampling design to aid 2
Petrópolis, 69080-971, Manaus, AM, Brazil
7College of Life and Environmental Sciences, University of Exeter, EX4 4RJ, UK comparison of ecological and socioeconomic conditions
8Universidade Rural da Amazônia, Rodovia PA 256, km 06, Bairro Nova Conquista, s/n, associated with different land uses across local, landscape
68625-000, Paragominas, PA, Brazil and regional scales; and (iii) a strong engagement with a
9Embrapa Monitoramento por Satélite, Av. Soldado Passarinho, 303, Fazenda wide variety of actors and non-research institutions.
Chapadão, 13070-115, Campinas, SP, Brazil Here, we elaborate on thesekey features, and identify
10Universidade de São Paulo, Escola Superior de Agricultura "Luiz de Queiroz", the ways in which RAS can help in highlighting those pro-
Esalq/USP, Avenida Pádua Dias, 11, São Dimas, Piracicaba, SP, Brazil blems in most urgent need of attention, and in guiding
11Universidade Federal do Oeste do Pará, Rua Vera Paz, S/N, Bairro Salé, 68040-250, improvements in land-use sustainability in Amazonia
Santarém, PA, Brazil
12 and elsewhere in the tropics. We also discuss some ofCentre de coopération internationale en recherche agronomique pour le développement - the practical lessons, limitations and realities faced
CIRAD. Campus International de Baillarguet. 34398 Montpellier Cedex 5, France
13 during the development of the RAS initiative so far.Universidade Federal de Lavras, Campus Universitário, CP 3037, 37200-000, Lavras,
MG, Brazil
14Centro de Energia Nuclear na Agricultura, CENA/USP, Universidade de São Paulo,
Av. Centenário 303, São Dimas, Piracicaba, SP, Brazil
15Universidade do Estado do Pará, Rodovia PA-125, s/n, Bairro: Algelim, 68625-000,
Paragominas, PA, Brazil 1. Introduction
16Universidade Federal de Viçosa, Av. P. H. Rolfs, s/n, Centro, 36570-000, Viçosa, MG, Brazil
17Universidade Federal do Pará, Rua Augusto Corrêa, s/n, Campus Profissional II, Land-use and land-cover change associated with agricul-
Guamá, 66000-000, Belém, PA, Brazil tural expansion and intensification is the most visible
18Embrapa Informática Agropecuária, Av. André Tosello, 209, Barão Geraldo, indicator of thehuman footprint on the biosphere [1–3].
13083-886, Campinas, SP, Brazil Ongoing land-use change is most acute in the tropics [4],
19Universidade Estadual de Campinas, Faculdade de Ciências Aplicadas, Rua Pedro with ca 50 000 km2 p.a. of native vegetation being cleared
Zaccharia, 1300, Cidade Universitária, 13484-350, Limeira, SP, Brazil [5]. These changes are driven by increasing resource demands
20Universidade Federal de Goiás, Campus II, 74001-970, Goiânia, GO, Brazil from a larger and wealthier human population, coupled with
21Instituto de Pesquisa Econômica Aplicada, Avenida Presidente Antônio Carlos, 51, the effects of increasing economic globalization and land
178 andar, Centro, 20020-010, Rio de Janeiro, RJ, Brasil
22 scarcity [6]. The creation and strengthening of more sustain-Universidade Federal do Rio de Janeiro, CP 68501, 21941-972, Rio de Janeiro, RJ, Brazil
23 able development trajectories in the twenty-first centuryStanford University, Energy and Environment Building, 4205, 473 Via Ortega,
Stanford, CA 94305, California, USA depends on our ability to balance risingdemands for food,
24The Nature Conservancy, Av. Nazaré, 280, Bairro Nazaré, 66035-170, Belém, PA, Brazil energy, natural resources and the alleviation of hunger and pov-
25Amnis Opes Institute and Department of Fisheries & Wildlife, Oregon State erty with the protection and restoration of natural ecosystems,
University, 200 SW 35th St., Corvallis, OR 97333, USA and the critical ecosystem services they provide [7,8].
26Embrapa Agrobiologia, BR 465, km 7, 23891-000, Seropédica, RJ, Brazil Amazonia represents a major sustainability challenge: as
27Universidade do Estado de Mato Grosso (UNEMAT), Br 158, Km 148, 78690-000, well as being the world’s largest remaining tropical forest,
Nova Xavantina, MT, Brazil the entire Amazon biome is home to more than 30 million
28IMAZON, Rua Domingos Marreiros, 2020, 66060-160, Belém, PA, Brazil people and provides locally, regionally and globally signifi-
29U.S. Environmental Protection Agency Office of Research and Development, 200 S.W. cant human-welfare benefits, including economic goods
35th St., Corvallis, OR 97333, USA
30 (e.g. timber and agricultural products) and non-market eco-Ministério de Ciência, Tecnologia e Inovação, Esplanada dos Ministérios, Bloco E, system services, such as climatic regulation and biodiversity
70067-900, Brası́lia, DF, Brazil
31Australian Centre for Biodiversity, School of Biological Sciences, Monash University, conservation [4,9,10]. Rapid social and ecological change
Victoria 3800, Australia has left the future of the Amazon region uncertain [11–13].
32Escola de Artes, Ciências e Humanidades, Universidade de São Paulo, Rua Arlindo In the Brazilian Amazon, in particular, recent reductions in
Bettio 1000, 03828-000 São Paulo, Brazil the rate of deforestation, expansion of protected areas,
33Instituto de Biociências, Universidade de São Paulo, Rua do Matão, Travessa 14, 101, increasedmarket-based demand for more responsible land-
05508-090 São Paulo, Brazil use practices, and a strengthening of local and regional
34Universidade Federal Mato Grosso, Av. Fernando Correa da Costa, s/n, Coxipó, governments and civil society organizations provide some
78060-900, Cuiabá, MT, Brazil cause for guarded optimism that the Amazon economy can
35Universidade Federal Fluminense, Rua Miguel de Frias, 9, Icaraı́, 24220-900, Niterói, be set on a sustainable footing [14–16]. However, we need
RJ, Brazil
36 to ensure the right choices are made as soon as possible,Department of International Development, London School of Economics, Houghton thereby reducing the likelihoodof costly or potentially irre-
Street, WC2A 2AE, London, UK
versible damageto both social and ecological systems in
the region [12,17]. Science can help this process by identifying
Science has a critical role to play in guiding more sustain-
the problems that need to be addressed first, and assessing the
able development trajectories. Here, we present the
long-term social and ecological implications ofland-use
Sustainable Amazon Network (Rede Amazônia Sustentável,
alternatives in planning for both regional development and
RAS): a multidisciplinary research initiative involving
ecological conservation [2,18,19].
more than 30 partner organizations working to assess
While there is already a substantial body of social and eco-
both social and ecological dimensions of land-use sustain-
logical knowledge on the Amazon [11,20–22], scientists are
ability in eastern Brazilian Amazonia. The research
often criticized for failing to deliver the evidence most
approach adopted by RAS offers three advantages for
needed to foster sustainability [23]. Criticisms include the frag-
addressing land-use sustainability problems: (i) the collec-
mentedanddisciplinary nature of many research projects, a
tion of synchronized and co-located ecological and
narrow focus on specific ecological or social problems and
socioeconomic dataacross broad gradients of past and
spatial scales, and a weak connection to local actorsand
rstb.royalsocietypublishing.org Phil Trans R Soc B 368: 20120166
institutions that are ultimately responsible for implementing The remainder of this paper focuses on describing the 3
changes in land-use policy and management [22–25]. key methodological components and novel features of our
Here, we present the work of the Sustainable Amazon Net- research design. We highlight some of the practical lessons
work (RAS; Rede Amazônia Sustentável in Portuguese), which is and realities faced during the development of the RAS initiat-
a multidisciplinary research initiative involving more than 30 ive so far, and identify the possible ways in which RAS could
research institutions and partner organizations. The overall have a lasting impact in guiding improvements in land-use
aim of this paper is to present the conceptual andmethodologi- sustainability in Amazonia and elsewhere in the tropics.
cal basis of the RAS initiative while also discussing many
fundamental challenges that confront research on land-use sus-
tainability across the tropics. Building on the work of a number 2. The Sustainable Amazon Network: research
of earlier and groundbreaking interdisciplinary assessments in
the Amazon, including the LBA (Programa de Grande Escala design
da Biosfera-Atmosfera na Amazônia) and GEOMA (Pesquisas (a) A conceptual framework for assessing land-use
de Desenvolvimento de Métodos, Modelos e Geoinformação
para Gestão Ambiental) research programmes [11,21,26], RAS sustainability
seeks to address some of the limitations listed above by asses- RAS is inspired by the now well-established paradigm of
sing the sustainability of land-use systems in two dynamic ‘sustainability science’—ascience that is focused explicitly
regions of eastern Brazilian Amazonia. The research approach on the dynamic interactions between nature and society
adopted by RAS offers three advantages for addressing this and is committed to place-based and solution-driven research
overarching goal: (i) the collection of synchronized and across multiple scales [27,28]. Making explicit our under-
co-located ecological and socioeconomic data across broad gra- standing of the interactions among and between social and
dients of past and present human use and exploitation of ecological phenomena, and their relationship to an overarch-
natural resources; (ii) a nested sampling design that allows ing sustainability agenda is critical to the effectiveness and
comparisons of the ecological and socioeconomic conditions transparency of such a research programme.
associated with different land uses to be made across local, The challenge of realizing a more sustainable development
landscape andregional scales; and (iii) a strong engagement trajectory for the Amazon region lies in identifying, protecting
with a wide variety of actors and non-research institutions. and restoring the balance of ecological and socioeconomic
Drawing upon the strengths of our approach, RAS aims to values necessary to maintain the flow of critical ecosystem
make important advances in understanding the sustainability services and adapt to changing conditions, while also safe-
challenges facing Amazonia with regards to four broad objec- guardingthe ability to exploit new opportunities for human
tives. First, we aim to quantify and better understand the development. The starting point for any research programme
ecological consequences of forest clearance, forest degradation on sustainability is the selection of a set of socio-ecological
and exploitation, and agricultural change (including cattle values that canprovide a basis for assessment. Our focus in
farming and silviculture) at several spatial scales. We are par- RAS is on the conservation of forest-dependent biodiversity (ter-
ticularly interested in assessing the relative importance of restrial and aquatic), the conservation and enhancement of
local- and landscape-scale variables, as well as the extent to carbon stocks, soil and water quality, the provision of agricul-
which past human impacts can help explain observed patterns tural, silvicultural, timber and non-timber forest products, and
in current ecological condition. Our measures of ecological the protection and betterment of human well-being.
condition include changes in terrestrial and aquatic biodiver- From this basis, the RAS research process can then
sity, carbon stocks, soil chemical and physical condition and address our primary objectives in helpingto quantify and
aquatic condition. Our second objective is to examine the fac- understand some of the social and ecological problems
tors that determine patterns of land use, management choice, and trajectories faced by the Amazon region, examine
agricultural productivity and profits (and hence opportunity interactions and the potential for costly or potentially irre-
costs for conservation) and patterns of farmer well-being. versible impacts, and evaluate the social and ecological
Beyond input cost, geophysical (e.g. soil type, topography) costs, benefits and trade-offs associated with proposed man-
andlocation (e.g. road and market access) factors, we recog- agement interventions. We view the transition towards
nize the potential importance of social–cultural factors in sustainability as a guiding vision for continuous improve-
influencing land-use behaviours, including geographical ments in management practices rather than a search for a
origin, technical support, credit access, social capital and the static blueprint of best practice techniques. Within this frame-
importance of supply chains. Third, we plan to use our multi- work, we see the role of research as providing both an
disciplinary assessment to evaluate the relationships between ongoing measure of management performance and a labora-
conservation and development objectives and identify poten- tory for testing new ideas for positive change.
tial trade-offs and synergies. Here, we are interested in the Building on earlier work by Collins et al. [19], we present a
relative ecological and socioeconomic costs and benefits of simple framework of how we view the interacting components
alternative land-use and management choices, and the of our social–ecological study system, and the hypothesized
potential for feedbacks, multiple scale interactions and depen- cause–effect relationships, assumptions and feedbacks that
dencies and unintended (‘perverse’) outcomes. Last, RAS provide a foundation for setting specific research objectives
seeks to help enable future research initiatives to maximize (figure 1). Outcomes measures (i.e. changes in valued attributes,
their cost-effectiveness by examining the implications of such as native biodiversity, ecosystem service provision and
choices made with respect to variable selection, sampling human well-being) are captured in both the social and the eco-
design, prioritization of research questions and analyses, and logical dimensions, and through changes in the stocks and
approaches for engaging with local actors and institutions flows of ecosystem services. Effects on these measures are felt
and disseminating results. through the cascading effects of changes in human behaviour
rstb.royalsocietypublishing.org Phil Trans R Soc B 368: 20120166
global and regional drivers 4
background: climate, population, policy and income
potential management and policy levers: zoning
policies, environmental regulation and compliance,
responsible farming approaches, climate and
biodiversity finance
social dimension synthesis and interactions ecological dimension
(institutions, organizations, (past and present) (soil, biogeography, climate)
economics)
environmental impacts or
human behaviour stressors biodiversity outcomes
land-use, migration, plants, birds, fish,
participation and forest loss, land-cover change, terrestrial and aquatic
values fire, logging, multiple degradation invertebrates
events, hunting
access, biotically
information, mediated
incentives, ecosystem
constraints processes
changes in ecosystem services
human outcomes ecosystem functionprovisioning: agricultural and silvicultural and habitat services
demography, production, extraction of timber and non-
timber forest products primary productivity,
development, equity maintenance of soil
regulating: carbon sequestration, water condition, water quality 
quality and stream flow and nutrient cycling
cultural: species conservation, ecotourism
and scientific discovery
social–ecological landscape properties
land cover and condition, management systems
multiple scales of interaction (property/site | catchment | region)
Figure 1. Conceptual model of study system under investigation by the Sustainable Amazon Network. Adapted from a generic framework presented in Collins et al.
[19] to illustrate how we view the interacting components of our social– ecological study system, and the hypothesized cause– effect relationships, contexts (social
and ecological dimensions and social–ecological interactions), assumptions and feedbacks between outcome measures (e.g. related to human well-being, bio-
diversity and ecosystem service provision), impacts and social and ecological processes, which together provide a foundation for setting specific research
objectives. Not all influences and feedbacks are of equal importance and no attempt is made in the model to distinguish relative effect sizes. Social– ecological
landscape properties are emergent and dynamic changes in landscape features that mediate relationships between social and ecological phenomena. System
dynamics play out across multiple spatial scales. Variables listed are those that have been studied by RAS.
and associated environmental impacts on landscape properties of a set of potential management and policy levers on the
and ecosystem functions. Each one of the influence arrows in long-termdynamics and outcomes of the study system (figure 1).
figure 1 encompasses a set of specific, disciplinary research ques-
tions. The importance of diverse human impacts (both faster
dynamics (such as fire and logging) and slower dynamics (b) Key RAS design features
(such as cumulative land-use change and repeated degradation RAS is an example of a research initiative that collects
events)) in determining changes in outcome variables is matched social and ecological data at multiple scales and of
examined using a space-for-time substitution across a highly relevance to multiple sustainability problems (see also [29]).
replicated network of sampling locations and landholdings, A number of features of the research design adopted by
coupled with detailedremotely sensed time-series analysis of RAS offer clear advantages for addressing questions about
past land-cover change and forest degradation. A focus of our land-use sustainability and management.
work is understanding the extent to which landscape properties
(often measurable from satellite and secondary data alone and (i) Spatial scale of assessment
used to compare multiple landscapes) can provide adequate Much of theexisting social and ecological research in the
proxies for understanding changes in the sustainability trajectory Amazon (and elsewhere) has not been conducted at the most
of the system as a whole. As much as possible, we try to ensure relevant spatial scales for assessing and guiding the develop-
that the interpretation of our results takes account of the spatial ment of more sustainableland-use strategies. Research has
scale of observation, and unmeasured factors, including the concentrated either on the entire Amazon basin, which often
effects of external drivers such as climate change andglobalmar- depends upon very coarse-scale data and obscures critically
kets, on the study system. Last, we seek to characterize the effects important inter- and intra-regional processes andinteractions
rstb.royalsocietypublishing.org Phil Trans R Soc B 368: 20120166
[30], or on detailed work on a few intensively studied research study regions, in contrast to Mato Grosso, the majority of prop- 5
sites, which captures only a tiny fraction of the variability in erties are less than 1000 ha. Moreover, local and regional urban
environmental and land-use gradients that drive much social centres still provide significantmarkets forcattle, and landscapes
and ecological change(see [10] in the case of biodiversity are interspersed with adiverse array of densely populated
research). While both large- and small-scale research is necess- small-holder colonies and agrarian reform settlements.
ary, muchmore work is neededat the ‘mesoscale’ level (i.e. Both Santarém and Paragominas have recently embarked
spanning hundredsof kilometresand coincident with the upon high-visibility, multi-sectoral sustainability initiatives;
scale of individual municipalities in Brazil). The RAS assess- specifically, a moratorium onexpansion of soya bean from
ment was conducted in two study regions in the Brazilian deforested areasin Santare ´m, and the foundation of the
state of Pará: the municipality of Paragominas (1.9 million hec- Municı́pio Verde (Green County) initiative for promoting sus-
tares) and part of the municipalities of Santarém and Belterra tainable land-use systems in Paragominas. These processes
(ca 1 million hectares) (figure 2). There are several important have strong support from non-governmental organizations,
advantages to working at this spatial scale. The socioeconomic farmer’s unions and local government, and have facilitated
and ecological data collected by RAS cover broad gradients of the development of RAS by helping us gain trust with local
change in both ecological (e.g. natural factors, such as soil type actors and institutions, tailoring the research planning and
and the extent of forest loss, degradation and land-use intensi- design towards local priorities and needs, and increasing
fication) and socioeconomic variables (e.g. rural population receptivity towards project results and recommendations.
density, property size, wealth and market access), thereby It is not viable to repeat the scale of assessment of the RAS
affording more confidence in the general relevance of the pat- initiative in every tropical forest region around theworld. How-
terns, drivers and trade-offs inferred from sample data [31]. ever, by working at multiple scales and in two differing
In addition, a focus at the mesoscale facilitates assessment of municipalities that encompass many characteristics of eastern
the importance of both local (farm) and regional (state and Amazonia and elsewhere, such as large areas of extensive
biome) processes andobjectives in a way that work focused cattle pasture, emergent mechanizedagriculture anda popu-
on either smaller or larger scales cannot readily achieve. Finally, lation that is highly mobile and dominated bysmall-holder
municipalities (or the equivalent scale of administration else- farmers, we believe that our results provide a suitable
where) are also the administrative unit with arguably the laboratory for better understanding many of the risks and
greatest awareness of local pressures on natural resources and opportunities facing the development of moresustainable
social services, and the greatest responsibility for institutional landscapes across the wider region. By concentrating our
linkages between local communities and states or regions [30]. efforts in two regions that have received particular attention
from existing initiatives in sustainable landuse, our results
almost certainly will receive greater exposure to, and engage-
(ii) Choice of study regions ment with, a wide range of decision makers. Last, a key focus
The RAS study regions of Paragominas and Santarém– of our work is to employ our uniquely comparable and
Belterra differ both biophysically and in their histories diverse datasets to identify a subset of cost-effective ecologi-
of human occupation and use. By collecting data from two cal and social indicators that can help guide applied research
distinct regions of eastern Amazonia, we have a rare oppor- and monitoring work in other study regions.
tunity to better understand the extent to which inferences
derived from one region can be generalized to another.
The modern city of Santarém, once a centre of pre-Colom- (iii) Sampling design
bian civilization, was founded in 1661, whereas Paragominas The RAS sampling design is based on a sample of 18 third- or
was founded as recently as 1959. Recent development of both fourth-order hydrological catchments (ca 5000 ha)in each
regions has been closely associated with the construction of fed- region. Catchments are distributed over agradient of forest
eral highways. Northern Santarém and neighbouring Belterra cover in 2009 (10–100% in Santarém; 6–100% in Paragominas;
have been densely settled by small-scale farmers for more than figure2), with detailed ecological and socioeconomic infor-
a century. By contrast, Paragominas had a very low population mation being collected from study transects and individual
density prior to its colonization by cattle ranchers from southern farms within each catchment (figure 2; electronic supplemen-
Brazilian states in the 1950s and 1960s, andthe boom in the tarymaterial). Advantages to this nested design include the
timber industry during the 1980s and 1990s. Both regions are potential for determining therelativeimportance of drivers
relatively consolidated, with decreasing rates of deforestation and constraints that operate at different spatial scales, and the
of primary vegetation, although on-goingpaving of thehighway capacity to make connections between local/individual (farm)
means southern Santarém will probably experience both and larger scale/public (municipality and state) conservation
increased human colonization and agricultural expansion in and development objectives (table 1). Sampling at the catch-
the near future. Large-scale, mechanized agriculture became mentscale also permits the integration of terrestrial and
established in both regions only in the early2000s and has aquatic information, and the assessment of changes in ecologi-
increased rapidly in recent years (usually at the expense of cal and socioeconomic variables that are highly correlated at
both pastures and secondary forest), currently occupying local scales, such as cumulative deforestation, economic activi-
approximately 40 000 and 60 000 hain Santare ´m and Para- ties and human population density. The 36 study catchments
gominas, respectively. Paragominas has also witnessed a rapid (figure 2; electronic supplementary material, figures S1and
recent expansion of silviculture (mostly Eucalyptus spp. and S2) were selected to capture the full deforestation gradient,
Schizolobium amazonicum). Both regions are distinct from the while incorporating priority areas identified by members of
agro-industrial frontier in Mato Grosso which is dominated by the municipal governments and farming communities (e.g.
large-scale mechanized farming primarily for export [32,33]. agrarian reform settlements, traditional rural communities
Although mechanized farming is expanding rapidly in both and areas of recent agricultural expansion and development).
rstb.royalsocietypublishing.org Phil Trans R Soc B 368: 20120166
6
a catchments 
b 
(a) Santarém- (b) Paragominas 
      Belterra
catchments 
Figure 2. The Sustainable Amazon Network nested sampling design. Distribution of study catchments (white) is shown within both Paragominas (a) and Santarém-
Belterra (b). Black circles show location of streams sampled during the aquatic assessment. Black bar charts show distribution of remnant forest cover across catch-
ments. (c) The distribution of study transects (black lines) and the principal household of producer landowners (triangles) in the catchment of Boa Esperanca in
Santarém. Land-use classification derived from Landsat 2010 image, showing primary forest (grey), secondary forest (light grey), deforested areas (white) and major
water bodies (dark grey). (Online version in colour.)
Ecological datawere collected from a sample of 300 m study alonga gradient of prior human impact based primarily on
transects in every catchment, distributed using a stratified- the amount of remnant forest cover in the upstream catchment
random sampling design, where a standard density of transects (and not constrained to terrestrial study catchments).
(1per 400 ha) was distributed across the catchment in pro- Socioeconomic data were collected from all rural properties
portion to the percentage cover of total forest and production with an ecological study transect. Owing to the stratified
areas (encompassing agriculture, pasture, fruiticulture and silvi- design, transects tended to be in larger properties and under-rep-
culture; figure 2). For example, if half of the landscape was resent smaller farms. Therefore,wemappedall ruralproducers in
covered by forest, then half of the transects were allocated to each catchment and sub-sampled a maximum of 20 randomly
forest. In catchments with very low levels of forest cover we selected properties (with at least 1 ha and producing in 2009).
sampled additional forest transects to ensure a minimum Given our focus on the producer community, this sample
sample of three transects in all catchments. Within each of excluded urban and periurban areas, but could include some of
these two land-use categories (forest and non-forest), sample the same farms in the transect-based sample. This combination
transects were distributed randomly witha minimum separ- of sampling techniques enables us to describe the dominant
ation of 1500 m to minimize spatial dependence. The use of socioeconomic and demographic characteristics of different pro-
this stratified-random sampling design provided a balance ducers, and to provide a detailed socioeconomic profile of the
between the need for: (i) proportional sampling of forest and farming population in each catchment (figure 2). Where rural
non-forest areas, and a sufficient density and coverage of properties had more than one household (e.g. where there are
sample points to capture major differences in landscape workers or relatives living on the property), additional surveys
structure and composition among different catchments; and on household demography, origins and well-being were made
(ii) a well-dispersed set of sampling points across forest and according to the total number of residences (table 1).
non-forest areas that captured important environmental
heterogeneities within each catchment and across the region
as a whole, helping to minimize problems of pseudo-replica- (iv) Social and ecological field sampling
tion. Aquatic sampling was conducted across 50 stream sites, RAS project members conducted a detailed assessment of
each 150 m long in each region, with samples distributed ecological and socioeconomic patterns and processes in
% forest cover 
% forest cover 
rstb.royalsocietypublishing.org Phil Trans R Soc B 368: 20120166
Table 1. Remote-sensing, socioeconomic and environmental data sampled by the Sustainable Amazon Network. 7
summary characteristics
variable type variables Paragominas Santarém
remote sensing biannual land-use classification (since 1988 in Paragominas and 1990 in Santarém-Belterra); age of deforestation; frequency
and timing of forest degradation events; age and frequency of secondary forest regeneration; mapping of fire and logging
scars; indices of deforestation and forest regeneration trajectories; cover of mechanized agriculture since 2000 (MODIS
images); land-use intensity by hydrological distances between stream networks and forest remnants
socioeconomic property sizes in socioeconomic survey number area number area
surveyed surveyed
(ha) (ha)
0–25 ha 44 936 150 1656
25–100 ha 47 3030 110 7587
100–300 ha 20 3577 20 3837
300–1000 ha 16 9222 21 12 397
over 1000 ha 44 238 979 16 62 978
total number of properties 171 255 744 317 88 455
total number of households 223 400
survey modules property characteristics; household characteristics, demography and
well being; productivity and inputs of different production
systems; fire use and impacts; forest use (and hunting)
soil physical structure, soil fertility, total C and N, d13C 3120 and 2580 soil samples from Paragominas and Santarém,
and d15N, phospholipid fatty acids (PLFA) respectively. Five replicates from each transect and at three
analysis of soil microbes, microbial biomass, soil depths (0–10, 10–20, 20–30 cm). Microbial and PLFA data,
water soluble nutrients, soil emissions of CO2, soil water soluble nutrients and soil gases emissions for
NH4, N2O selected catchments from Santarém only.
vegetation and biomass and vegetation structure (including dead 44 359 stems measured and 38 584 stems measured
carbon stocks wood, leaf litter and structural measurements) identified and identified
tree, liana and palm diversity 1052 species 1118 species
disturbance observations of fire and logging scars and other damage on all
stems
terrestrial fauna birds 364 species 377 species
dung beetles 85 species 99 species
53 113 specimens 40 664 specimens
ants ca 300 species 430 species
orchid bees 28 species 34 species
ecosystem functions n.a. dung removal, soil
turbation, and seed
dispersal by dung
beetles, and seed
predation by ants
aquatic system physical habitat 237 measurements relating to channel morphology, substrate,
habitat complexity and cover, riparian vegetation, channel–
riparian interactions and disturbance
aquatic quality physical and chemical parameters of water (dissolved oxygen,
conductivity, pH, temperature, nitrate and ammonia)
(Continued.)
rstb.royalsocietypublishing.org Phil Trans R Soc B 368: 20120166
Table 1. (Continued.) 8
summary characteristics
variable type variables Paragominas Santarém
fish 112 species 71 species
18 669 individuals 7990 individuals
Ephemeroptera, Plecoptera and Trichoptera 49 genera 54 genera
14 113 individuals 7937 individuals
Heteroptera 9 genera 14 genera
1847 individuals 543 individuals
Odonata 97 species 68 species
1990 individuals 1849 individuals
both study regions between April 2010 and August 2011 effectively exploit economies of scale in shared resources and
(table1 and figure 2; electronic supplementary material). technical expertise, recognize and make explicit interconnec-
Choices of sample variables and methods were based on tions and feedbacks among sub-disciplines, andincrease the
our research priorities, cost-effectiveness and the need to col- temporal andspatial scale of existing studies [22] . However,
lect a large number of representative samples [34] (table 1). building effective multi-sector and interdisciplinary research
Sampling of terrestrial biodiversity focused on trees and programmes at large spatial scales remains one of the most
lianas,birds, dung beetles, ants,orchid bees and soil difficult challenges facing sustainability science [37].
microbes. In a subset of catchments, additional measure- One of the greatest challenges of the RAS project has been
ments were made of ecosystem functions mediated by developing and maintaining engagement with partners from
beetles andants (including dung burial, seed dispersal and multiple sectors, institutions, local governments, civil society
seed predation). Aquatic biodiversity (and metrics of aquatic organizations and farmer associations. More than half of the
condition) consisted of fish and macroinvertebrate assem- remaining forest in the Amazon lies within private land [25],
blages (table 1). Ecosystem service supply was measured and one of the novel aspects of RAS is the collection of data
for carbon stocks (above- and below-ground) and the from complex landscapes with multiple owners that encom-
maintenance of soil condition (physical and chemical proper- pass a broad spectrum of culture, wealth and education.
ties). The habitat structure of both terrestrial and aquatic Establishing contact, building a minimum level of trust, and
environments was assessed using a combination of measures securing permissions from more than 200 private landowners
of canopy openness, vegetation structure, dead wood and across the 36 study catchments incurred significant costs in
leaf litter, and the morphology and substrate of stream chan- time and resources. Thiswas especially difficult in areas
nels. Socioeconomic data were collected on the characteristics of with a legacy of conflict over deforestation and the exploitation
study properties (such as land cover, legal status) and producer of natural resources. Such ‘transaction costs’ are rarely factored
households (including household demography, producer into or supported by funders of major research programmes.
origins, income, access to services, subjective measures of Despite the challenges, most landowners recognized the
well-being), costs and productivity of different production value of research in strengthening the evidence basisfor
systems (livestock, arable and perennial crops, silviculture what are otherwise largely rhetorical and highly politicized
and timber harvesting), fire use and effects, and the benefits debates regarding the effects and drivers of land-use change.
and costs of maintaining forest reserves (including the extrac- The diversity of institutional partners that make up RAS,
tion of timber andnon-timber forest products, and risks of including local organizations, and those directly concerned
invasion and theft) (table 1). with agricultural development and local conservation initiat-
Legacyeffects of past human impacts are known tobe impor- ives, was critically important in building trust. While the
tant for both ecological and social systems, but have been poorly establishment of meaningful partnerships with very different
studied to date [35,36]. Remote-sensing analyses were based on a types of landowners (including some of the poorest and richest
22-year time series and provide information on changes in land farmers in the study regions) was critical for the success of
use, forest extent, timing andfrequency of forest degradation RAS, it was also important to avoid over-promising and
and age of regeneration (see the electronic supplementary over-committing on the benefits to individual land owners
material, table S2). These data provide the basis for validating from project outcomes.Considerable care was taken to
remotely sensed indicators of ecological andland-use change manage expectations by distinguishing clearly the purpose of
with direct field observations (e.g. retention and loss of forest research from rural development and agricultural extension,
biodiversity, forest fires and land-mechanization). and presenting realistic timetablesfor project participation
and the dissemination of results.
Maintaining a meaningful level of engagement with our
network of local partners is critical to help maximize the rel-
3. Practical lessons and realities from the field evance of our analyses of project data to local sustainability
The acquisition of extensive and reliable knowledge about the problems [23]. We are keenly aware that the difficulties inherent
Amazon is dependent on research networks that can in giving adequate attention to the needs and problems facing
rstb.royalsocietypublishing.org Phil Trans R Soc B 368: 20120166
local communities can increase the chance of drawing inap- understanding the appropriate locations, scale, starting con- 9
propriate conservation and development recommendations ditions and potential constraints associatedwith any future
from our work. We are wary of presenting and interpreting changes in management actions [40]. Such basic information
trade-offs too simply, and weacknowledge that simplified is still lacking for much of the Amazon region.
quantitative analyses and narratives that only take account of RAS datasets can help reconcile social–ecological objec-
a limited set of attributes can obscure important dynamics tives and reveal trade-offs between farming and conservation
and dimensions of value, often resulting in the marginalization at multiple spatial scales by combining data on socioeconomic
of some interest groups [38]. Although commonplace in and ecological values. One prominent debate concerns the
research projects such risks are rarely made explicit. effectiveness of alternative approaches for attempting to bal-
Within the RAS research network, we encountered many ance conservation and agricultural activities through changes
of the problems faced by other multidisciplinary projects, in agricultural productivity and farming techniques, often
including the need to overcome differences in values, referred to as land-sparing versus land-sharing [41]. Under-
language andmodes of thinking among disciplines [22,24]. standing of this general problem is limited by a lack of data
There are no easy answers to such challenges, though we on the conservation value of areas of remaining native veg-
have found that co-location of researchers from different dis- etation available for conservation investment that are in
ciplines within the same field teams, use of a shared online differing stages of degradation or regeneration, farm-scale
management platform and group exercises (such as partici- differences in agricultural productivity and other socioeco-
pation in conference symposia and writing this paper) have nomic variables related to human well-being and poverty,
all helped promote constructive dialogue. RAS has its origins and landscape-scale influences on local ecological and socio-
in three previously independent research projects that were economic properties. RAS data can make a potentially
amalgamated together with more partners and funding important contribution to the development of Reducing Emis-
sources into a single initiative with shared goals, budget sions from Deforestation and Degradation (REDDþ) initiatives
and management structure. While this historical trajectory [42], recognizing that we currently have a very poor under-
led inevitably to amore complexfunding and communi- standing of the relative ecological and socioeconomic costs
cation system, the resulting strong sense of ownership and benefits of alternativeforest conservation policies (e.g.
shared by many project members often led to a more open, avoided deforestation versus avoided degradation and forest
interactive and democratic decisionmaking process during restoration activities) and the interaction between such policies
project planning and execution. and the agricultural sector [43].
Many of the greatest challenges in developing RAS arose Data and results from RAS ultimatelyaim to contribute
from mundane problems of coordinating the collection, proces- towards more sustainable land-use systems in Amazonia in
sing and analysis of data. There is a need forcontinual five overlapping areas, namely the development of: (i) best
reassessment of the value and purpose of new measurements practice recommendations for sustainable intensification and
or additional samples, and the extent to which more data are responsible agriculture, particularly in the cattle-ranching
necessary to address the priority questions. Cost-effectiveness sector; (ii) cost-effective approaches to achieving compliance
in time and resources are often ignored in conservation research with environmental legislation, especially in Brazilian Forest
(e.g. in biodiversity surveys [34,39]), yet the effectiveness of Law; (iii) strategies for investment in forest conservation and
research would be significantly improved if these considerations restoration through payment for ecosystem service schemes,
were consistently takeninto account in project planningand and particularly carbon finance; (iv) strategies for promoting
development. We suggest that complex projects such as RAS fire-free agriculture; and (v) municipal-level ecological–
establish ‘stopping rules’, both in the collection of more field economic zoning processes. We seek to identify potential
samples and in cutting losses in areas where progress is slow opportunities and motivations for more sustainable develop-
ornegligible. The marginal costs of more field data may ment strategies in eastern Amazonia and elsewhere by
appear to be little, but theymust take account the costs of labora- combining the quantitative foundation of our sustainability
tory and analysis work, and the transaction costs of managing assessment with input from stakeholders and work in the
increasing project complexity. political and social sciences [44].
Wehope that our data will be helpful to assess how
changes in management incentives or regulatory conditions
4. Next steps: guiding improvements in land-use will influence relative ecological and socioeconomic costs
and benefits. However, we also recognize that win–win
sustainability solutions are rare and often misleading. Given this, our
Work to address our first two objectives is ongoing in work seeks to give explicit consideration to possible conflicts,
many disciplines in RAS to assess and better understand compromises and synergies among multiple objectives, unex-
the ecological and socioeconomic consequences of land-use pected interactions and feedbacks, andthe broader political
and landscape changes, with synthesis analyses of trade- and institutional context [45].
offs and scenarios scheduled from 2013. We hope that the Ensuring that the work being undertaken byRASgoes
outcomes from RAS can help guide improvements in land- beyond science and successfully bridges the science–policy
use policy and management in several ways. At the simplest divide is both extremely challenging and unpredictable.
level, the quantification of deleterious trends in valued attri- There are at least three areas where we hope that our approach
butes (e.g. declines in forest biodiversity, ecosystem service can help to increase opportunities for informing development
production and socioeconomic values) and the identification and conservation decision makers. First, our interdisciplinary,
of key stressors can both help to identify management mesoscale and place-based research approach increases the
priorities. A clearer understanding of spatial patterns of eco- likelihood that our results are relevant and applicable to
logical and socioeconomic condition is fundamental for regional problems. Second, we believe that to be most effective
rstb.royalsocietypublishing.org Phil Trans R Soc B 368: 20120166
the process of knowledge exchange should occur across as different sectors and contributing not only to the delivery of 10
broad and diverse set of actors as possible. Here, the partici- policy-relevant research outputs as outlined in this paper, but
pationof such a large group of (mostly Brazilian) students also to broader efforts to build the capacity and understanding
and researchers on the one hand, with a large and diverse necessary to create a more sustainable development trajectory
array of non-research partners and associates (including for the Amazon region. We hope that the work of RAS can
conservation organizations, farmers groups, government make a small contribution towards this enormous challenge.
agencies and individual landowners) on the other has pro-
vided thebasis for multiple ongoing dialogues about our This paper is dedicated to the late Manoel Aviz do Nascimento (‘Nego’)
research objectives and preliminary findings. Knowledge whose assistance to all aspects of RAS work in Santarém was so invalu-
exchange should not be limitedto high-level executive sum- able. We are grateful to the following for financial support; Instituto
maries for policy makers but must exploit opportunities for Nacional de Ciência e Tecnologia—Biodiversidade e Uso da Terra na
shared learning and dissemination of ideas at all levels. Last, Amazônia (CNPq 574008/2008-0), Empresa Brasileira de Pesquisa
Agropecuária—Embrapa (SEG: 02.08.06.005.00 and 01.05.01.003.05),
we are developing an impact strategy that can help to target the UK government Darwin Initiative (17-023), The Nature Conser-
the presentation and discussion of key results through appro- vancy, Natural Environment Research Council (NERC) (NE/
priate media to specific audiencesand demands at local, F01614X/1, NE/G000816/1, NE/F015356/2 and NE/l018123), Con-
regional and national levels. selho Nacional de Desenvolvimento Cientı́fico e Tecnológico (CNPq)
Sustainability science needs to balance the often-conflicting (477583/2009-1), the Fulbright Commission (RH), São Paulo Research
Foundation (FAPESP) (2011/19108-0), Fundação de Amparo à Pesquisa
timetables of researchand policyprocesses. As scientists we do estado de Minas Gerais (FAPEMIG), and the Brazilian Coordenação
strive to ensure the reliability, intellectual credit and indepen- de Aperfeiçoamento de Pessoal de Nı́vel Superior (CAPES). R.M. and
dence of our work; a process that often requires a lot of time. J.R.T. were supported by Australian Research Council grant
However, to influence the policy process effectively, our experi- DP120100797. We also thank the farmers and workers unions of San-
ence is that the research process also needs to be able to tarém, Belterra and Paragominas and all collaborating private
landowners and local government officials for their support. We are
respond to limited andoften unpredictable opportunities for grateful to Paulo Brando, Jamila Haider and two anonymous reviewers
contributing to decisions on management and policy. Engaging for suggestions to improve the manuscript. More information about
in this process requires innovative methods for interacting with RAS can be found at www.redeamazoniasustentavel.org.
References
1. Rockstrom J et al. 2009 A safe operating space for 10. Peres CA, Gardner TA, Barlow J, Jansen J, Michalski institutional responses to climate shocks. In
humanity. Nature 461, 472–475. (doi:10.1038/ F, Lees AC, Vieira ICG, Moreira FMD, Feeley K. 2010 Addressing tipping points (eds T O’Riordan,
461472a) Biodiversity conservation in human-modified T Lenton, I Christie). Oxford, UK: Oxford University
2. Foley JA et al. 2011 Solutions for a cultivated Amazonian forest landscapes. Biol. Conserv. 143, Press.
planet. Nature 478, 337–342. (doi:10.1038/ 2314–2327. (doi:10.1016/j.biocon.2010.01.021) 18. Brussaard L, Caron P, Campbell B, Lipper L, Mainka
nature10452) 11. Davidson EA et al. 2012 The Amazon basin in S, Rabbinge R, Babin D, Pulleman M. 2010
3. Nkonya E, Karsenty A, Msangi S, Souza CM, Shah M, transition. Nature 481, 321–328. (doi:10.1038/ Reconciling biodiversity conservation and food
vom Braun J, Galford G, Park S. 2012 Sustainable nature10717) security: scientific challenges for a new agriculture.
land use for the 21st century. New York, NY: United 12. Gardner TA. In press. The Amazon in transition: the Curr. Opin. Environ. Sustainability 2, 34–42. (doi:10.
Nations Department of Economic and Social Affairs. challenge of transforming the world’s largest tropical 1016/j.cosust.2010.03.007)
Department for Sustainable Development. forest biome into a sustainable social-ecological 19. Collins SL et al. 2011 An integrated conceptual
4. FAO. 2011 The State of Forests in the Amazon Basin, system. In Addressing tipping points (eds T O’Riordan, T framework for long-term social– ecological research.
Congo Basin and Southeast Asia. Rome, Italy: Food LentonI Christie). Oxford, UK: Oxford University Press. Front. Ecol. Environ. 9, 351–357. (doi:10.1890/
and Agriculture Organization of the United Nations. 13. Andersen L, Granger C, Reis E, Weinhold D, Wunder 100068)
5. Hansen MC, Stehman SV, Potapov PV. 2010 S. 2002 The dynamics of deforestation and economic 20. Becker BL. 2011 Amazônia: Geopolı́tica na virada do
Quantification of global gross forest cover loss. Proc. growth in the Brazilian Amazon. Cambridge, UK: III milênio. Rio de Janeiro, Brazil: Garamond.
Natl Acad. Sci. USA 107, 8650–8655. (doi:10.1073/ Cambridge University Press. 21. Keller M, Gash J, Dias PS, Bustamante M (eds)
pnas.0912668107) 14. Nepstad D et al. 2009 Environment: the end of 2009 Amazonia and global change: a synthesis of
6. Lambin EF, Meyfroidt P. 2011 Global land use deforestation in the Brazilian Amazon. Science LBA research. Washington: American Geophysical
change, economic globalization, and the looming (New York N.Y.) 326, 1350–1351. (doi:10.1126/ Union.
land scarcity. Proc. Natl Acad. Sci. USA 108, 3465– science.1182108) 22. Barlow J et al. 2010 Using learning networks to
3472. (doi:10.1073/pnas.1100480108) 15. Nepstad DC, McGrath DG, Soares-Filho B. 2011 understand complex systems: a case study of
7. Steffen W et al. 2011 The anthropocene: from Systemic conservation, REDD, and the future of the biological, geophysical and social research in the
global change to planetary stewardship. AMBIO 40, Amazon Basin. Conserv. Biol. 25, 1113–1116. Amazon. Biol. Rev. Camb. Phil. Soc. 86, 457–474.
739–761. (doi:10.1007/s13280-011-0185-x) (doi:10.1111/j.1523-1739.2011.01784.x) (doi:10.1111/j.1469-185X.2010.00155.x)
8. Godfray HCJ et al. 2010 Food security: the challenge 16. Hecht SB. 2011 From eco-catastrophe to zero 23. Lahsen M, Nobre CA. 2007 Challenges of connecting
of feeding 9 billion people. Science (New York N.Y.) deforestation? Interdisciplinarities, politics, international science and local level sustainability
327, 812–818. (doi:10.1126/science.1185383) environmentalisms and reduced clearing in efforts: the case of the large-scale biosphere–
9. Malhi Y, Roberts JT, Betts RA, Killeen TJ, Li W, Nobre Amazonia. Environ. Conserv. 39, 4– 19. (doi:10. atmosphere experiment in Amazonia. Environ. Sci.
CA. 2007 Climate change, deforestation and the fate 1017/S0376892911000452) Policy 10, 62–74. (doi:10.1016/j.envsci.2006.10.005)
of the Amazon. Science (New York N.Y.) 319, 17. Boyd E. In press. Managing global tipping points: 24. Perz SG et al. 2010 Crossing boundaries for
169–172. (doi:10.1126/science.1146961) exploring the adaptive cycle metaphor and environmental science and management:
rstb.royalsocietypublishing.org Phil Trans R Soc B 368: 20120166
combining interdisciplinary, interorganizational and 31. McIntire EJB, Fajardo A. 2009 Beyond description: 25, 259–264. (doi:10.1111/j.1523-1739.2010. 11
international collaboration. Environ. Conserv. 37, the active and effective way to infer processes from 01608.x)
419–431. (doi:10.1017/S0376892910000810) spatial patterns. Ecology 90, 46–56. (doi:10.1890/ 39. Hughes RM, Peck D. 2008 Acquiring data for large
25. Ferreira J, Pardini R, Metzger JP, Fonseca CR, 07-2096.1) aquatic resource surveys: the art of compromise
Pompeu PS, Sparovek G, Louzada J. 2012 Towards 32. DeFries R, Herold M, Verchot L, Macedo MN, among science, logistics, and reality. J. North
environmentally sustainable agriculture in Brazil: Shimabukuro Y. 2013 Export-oriented deforestation American Benthol. Soc. 27, 837–859. (doi:10.1899/
challenges and opportunities for applied ecological in Mato Grosso: harbinger or exception for other 08-028.1)
research. J. Appl. Ecol. 49, 535–541. (doi:10.1111/j. tropical forests? Phil. Trans. R. Soc. B 368, 40. Tscharntke T, Klein AM, Kruess A, Steffan-Dewenter
1365-2664.2012.02145.x) 20120173. (doi:10.1098/rstb.2012.0173) I, Thies C. 2005 Landscape perspectives on
26. Toledo de PM, Vieira ICG, Câmara G, Nobre CA. 2007 33. VanWey LK, Spera S, de Sa R, Mahr D, Mustard JF. agricultural intensification and biodiversity:
Integrating environmental and social agendas: the 2013 Socioeconomic development and agricultural ecosystem service management. Ecol. Lett. 8, 857–
experience of the Amazonian networks LBA and intensification in Mato Grosso. Phil. Trans. R. Soc. B 874. (doi:10.1111/j.1461-0248.2005.00782.x)
GEOMA. In Communicating global change science to 368, 20120168. (doi:10.1098/rstb.2012.0168) 41. Phalan B, Balmford A, Green RE, Scharlemann JPW.
society (eds H Tiessen, M Brklacich, G Breulmann, 34. Gardner TA et al. 2008 The cost-effectiveness of 2011 Minimising the harm to biodiversity of
RSC Menezes), pp. 109–118. New York, NY: Island biodiversity surveys in tropical forests. Ecol. Lett. 11, producing more food globally. Food Policy 36,
Press. 139–50. (doi:10.1111/j.1461-0248.2007.01133.x) S62–S71. (doi:10.1016/j.foodpol.2010.11.008)
27. Clark WC, Dickson NM. 2003 Sustainability science: 35. Gardner TA, Barlow J, Chazdon R, Ewers RM, Harvey 42. Angelsen A, Brockhaus M, Sunderlin WD, Verchot LV
the emerging research paradigm. Proc. Natl Acad. CA, Peres CA, Sodhi NS. 2009 Prospects for tropical (eds) 2012 Analysing REDDþ : challenges and
Sci. USA, 100, 8059–8061. (doi:10.1073/pnas. forest biodiversity in a human-modified world. Ecol. choices. CIFOR: Bogor, Indonesia.
1231333100) Lett. 12, 561–582. (doi:10.1111/j.1461-0248.2009. 43. Beddington JR et al. 2013 What next for agriculture
28. Carpenter SR et al. 2012 Program on ecosystem 01294.x) after Durban? Science 335, 290–298. (doi:10.1126/
change and society: an international research 36. Brondizio ES, Ostrom E, Young OR. 2009 science.1217941)
strategy for integrated social– ecological systems. Connectivity and the governance of multilevel 44. Le Tourneau F-M, Marchand G, Greissing A, Nasuti
Curr. Opin. Environ. Sustainability 4, 134–138. social-ecological systems: the role of social capital. S, Droulers M, Bursztyn M, Léna P, Dubreuil V.
(doi:10.1016/j.cosust.2012.01.001) Annu. Rev. Environ. Resour. 34, 253–278. (doi:10. 2013 The DURAMAZ indicator system: a cross-
29. Kareiva P, Tallis H, Ricketts TH, Daily GC, Polasky S. 1146/annurev.environ.020708.100707) disciplinary comparative tool for assessing
2011 Natural capital: theory and practice of mapping 37. Carpenter SR et al. 2009 Science for managing ecological and social changes in the Amazon. Phil.
ecosystem services. Oxford, UK: Oxford University ecosystem services: beyond the millennium Trans. R. Soc. B 368, 20120475. (doi:10.1098/rstb.
Press. ecosystem assessment. Proc. Natl Acad. Sci. USA 2012.0475)
30. Brondizio ES, Moran EF. 2012 Level-dependent 106, 1305–1312. (doi:10.1073/pnas.0808772106) 45. McShane TO et al. 2011 Hard choices: making trade-
deforestation trajectories in the Brazilian Amazon 38. Hirsch PD, Adams WM, Brosius JP, Zia A, Bariola N, offs between biodiversity conservation and human
from 1970 to 2001. Popul. Environ. 34, 69–85. Dammert JL. 2011 Acknowledging conservation well-being. Biol. Conserv. 144, 966–972. (doi:10.
(doi:10.1007/s11111-011-0159-8) trade-offs and embracing complexity. Conserv. Biol. 1016/j.biocon.2010.04.038)
rstb.royalsocietypublishing.org Phil Trans R Soc B 368: 20120166