<%BANNER%>

Conservation and livelihood development in Brazil nut-producing communities in a tri-national Amazonian frontier

Permanent Link: http://ufdc.ufl.edu/UFE0041209/00001

Material Information

Title: Conservation and livelihood development in Brazil nut-producing communities in a tri-national Amazonian frontier
Physical Description: 1 online resource (214 p.)
Language: english
Creator: Duchelle, Amy
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: amazon, certification, communities, conservation, livelihoods, ntfps, tenure
Forest Resources and Conservation -- Dissertations, Academic -- UF
Genre: Forest Resources and Conservation thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: This dissertation examines the relationship between forest conservation, forest dependence, and livelihood development of extractive communities in the tri-national frontier region of Madre de Dios, Peru, Acre, Brazil, and Pando, Bolivia in Western Amazonia. While the focus of the dissertation is on Brazil nut (Bertholletia excelsa), this regionally-important non-timber forest product (NTFP) is used to illustrate broader issues related to community-based forest management. This study assessed community-based forest management from three different perspectives: (1) an analysis of land use land cover change in community-managed forests through the use of satellite and survey data; (2) an exploration of the causes and effects of Brazil nut thefts, and resolution of such thefts; and, (3) a comparison of the environmental and economic outcomes of organic, Fair Trade, and Forest Stewardship Council (FSC) certification of Brazil nuts. The first set of results demonstrated minimal deforestation and extremely high forest income dependency in extractive communities. In 2000 2005, deforestation occurred in already fragmented areas and along roads, with most deforestation in Acre and least in Pando. From 2002 to 2007, higher agricultural income was positively correlated with reported forest clearing at the household level in Pando and Madre de Dios, whereas higher Brazil nut income was correlated with less forest cleared. In Acre, government aid, larger households, and higher value of livestock assets were correlated with forest clearing. The second set of results highlighted more Brazil nut thefts in Pando when compared with Acre, likely due to land titling processes in Pando that disregarded traditional forest use, settlement patterns that disconnected producers from their forest resources, and higher nut dependence. Both threat of theft and resource dependence affected nut harvest regimes in Pando. The third set of results showed that organic and Fair Trade Brazil nut certification schemes were associated with better post-harvest practices and higher prices, while producers certified by the FSC in Peru adopted practices related to pre-harvest planning and tree health. These findings contribute to our understanding of the role of community-based forest management in tropical forest conservation and livelihood development.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Amy Duchelle.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Kainer, Karen A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041209:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041209/00001

Material Information

Title: Conservation and livelihood development in Brazil nut-producing communities in a tri-national Amazonian frontier
Physical Description: 1 online resource (214 p.)
Language: english
Creator: Duchelle, Amy
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: amazon, certification, communities, conservation, livelihoods, ntfps, tenure
Forest Resources and Conservation -- Dissertations, Academic -- UF
Genre: Forest Resources and Conservation thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: This dissertation examines the relationship between forest conservation, forest dependence, and livelihood development of extractive communities in the tri-national frontier region of Madre de Dios, Peru, Acre, Brazil, and Pando, Bolivia in Western Amazonia. While the focus of the dissertation is on Brazil nut (Bertholletia excelsa), this regionally-important non-timber forest product (NTFP) is used to illustrate broader issues related to community-based forest management. This study assessed community-based forest management from three different perspectives: (1) an analysis of land use land cover change in community-managed forests through the use of satellite and survey data; (2) an exploration of the causes and effects of Brazil nut thefts, and resolution of such thefts; and, (3) a comparison of the environmental and economic outcomes of organic, Fair Trade, and Forest Stewardship Council (FSC) certification of Brazil nuts. The first set of results demonstrated minimal deforestation and extremely high forest income dependency in extractive communities. In 2000 2005, deforestation occurred in already fragmented areas and along roads, with most deforestation in Acre and least in Pando. From 2002 to 2007, higher agricultural income was positively correlated with reported forest clearing at the household level in Pando and Madre de Dios, whereas higher Brazil nut income was correlated with less forest cleared. In Acre, government aid, larger households, and higher value of livestock assets were correlated with forest clearing. The second set of results highlighted more Brazil nut thefts in Pando when compared with Acre, likely due to land titling processes in Pando that disregarded traditional forest use, settlement patterns that disconnected producers from their forest resources, and higher nut dependence. Both threat of theft and resource dependence affected nut harvest regimes in Pando. The third set of results showed that organic and Fair Trade Brazil nut certification schemes were associated with better post-harvest practices and higher prices, while producers certified by the FSC in Peru adopted practices related to pre-harvest planning and tree health. These findings contribute to our understanding of the role of community-based forest management in tropical forest conservation and livelihood development.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Amy Duchelle.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Kainer, Karen A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041209:00001


This item has the following downloads:


Full Text

PAGE 1

CONSERVATION AND LIVELIHOOD DEVEL OPMENT IN BRAZIL NUT-PRODUCING COMMUNITIES IN A TRI-NATION AL AMAZONIAN FRONTIER By AMY ELEANOR DUCHELLE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Amy Eleanor Duchelle 2

PAGE 3

To Vanessa Sequeira (1970-2006) A bright light in the world of Amazon conservation and development, and a friend and colleague who enriched many lives. 3

PAGE 4

ACKNOWLEDGMENTS First and foremost, I thank my academic advisor, Karen Kainer, for seeing me through this experience and always provid ing outstanding support and feedback. She has been a wonderful role model as a scientis t and as a person. I am also very appreciative of how she enabled and encouraged my involvement in complementary learning experiences along the way, whic h have been fundamental to my professional and personal growth. I am grateful to the other members of my doctoral committee, Marianne Schmink, Jack Putz, Jane Southworth and Brian Child for their unique contributions to the process and products of this doctoral experience. Faculty members, Bob Buschbacher, Gr enville Barnes, Jon Dain, and Stephen Perz have also provided important in sights and support. This work would not have been possible without the support and friendship of many colleagues in the MAP r egion during more than two year s of fieldwork. I am grateful for my regional research collabor ators, especially Lcia Wadt of EMBRAPA Acre, Peter Cronkleton and Marco Antonio Al bornoz of CIFOR, El sa Mendoza of IPAM, Angelica Almeyda and Eben Broadbent of Stanford Univ ersity, Juan Fernando Reyes and Sissy Bello of Herencia, and Luz Marina Velarde of CAMDE Peru. I thank the participants of the Trem do Conhecimento for reinforcing the importance of returning research results in innovative ways to r egional stakeholders. I am deeply indebted to Gladys Guanacoma, Marciane de Araujo, Juceli Bezerra da Souza, and Peter Groenendjik for their outstanding and long-term field assistance. Most importantly, I thank the many Brazil nut producers in Pando, Acre and Madre de Dios who let me and my research team into their lives with openness and enthusiasm. My fieldwork 4

PAGE 5

expenses were covered by the Rainforest Alliance Kleinhans Fellowship, a William J. Fulbright Grant, and the Center for International Forestry Research (CIFOR). I appreciate my friends in Acre, Brazil Silvia Brilhante, Cazuza Borges, Foster Brown, Vera Reis, Monica de los Rios, Willia n Flores, Elaine Silva, Ana Euler, Elektra Rocha, Augusto Nagy, Giuliano Less, Dande Tavares, and Malu for their passion and commitment to Amazonian conservation and dev elopment. I look forward to future collaborations with new colleagu es at the Federal University of Acre, especially Vernica Passos, Cleber Salimon, Lisandro J uno, Patricia Peruquetti, and the Masters students in Ecologia e Manejo dos Recursos Naturais and Desenvolvimento Regional I feel lucky to be part of CIFORs Pove rty and Environment Network and have learned what it means to try to understand and measure livelihoods. A special thanks to Arild Angelsen, Sven Wunder, William Sunderlin Marty Luckert, and Ronnie Babigumira for their mentorship and s upport, and to all the PEN resour ce people and partners who make up this fantastic research group. I tr uly look forward to our global results! At the University of Florida, I am very grateful to the School of Forest Resources and Conservation (SFRC) and the Tropica l Conservation and Development (TCD) Program. The NSF-IGERT Working Forests in the Tropics Program, linked with both programs, not only funded severa l years of my doctoral progr am, but provided important encouragement and support for interdisciplinary and applied resear ch. I thank the terrific staff of SFRC, especia lly Christine Housel, Dawnette Lauramore, Cherie Arias, Winnie Lante, Sherry Tucker, Cindy Love, and others, who were always available to help. I also thank Hannah Covert and Patric ia Sampaio for their support of the TCD program. The NSF-HSD proj ect Infrastructure change human agency, and resilience 5

PAGE 6

in social-ecological systems allowed me to access satellite im agery and other spatial datasets, and I appreciate the Land Use and Environmental Change Institute (LUECI) for use of their lab space. A special t hanks goes to Natalia Hoyos and Matt Marsik for their GIS and remote sensing assistance. I am also grateful to Salvador Gezan from the Department of Statistics for his cheerfulness and aptitude in helping with statistical analyses. I thank the Biotropica group and the WFT 2008 conf erence committee who were so enjoyable to work with. I thank my lab mates, Tita Alvira for being my week end partner in the office, Marina Londres, Marlene Soriano, Joanna Tucker and Vivian Zeidemann for their friendly faces day-to-day, and Shoana Humphr ies, Cara Rockwell, Jennifer Arnold, Christie Klimas, and Rosa Cossio. What a gr eat group of scholarly and fun women! So many other friends at UF have made this a wonderful experience: Maria Digiano, Wendy-Lin Bartels, Claudia Stickler, Gaby Stocks, Miriam Wym an, Chris Baraloto, David Buck, Ellie Harrison, Anna Prizzia, Fr anklin Paniagua, Lin Cassidy, Amy Daniels, Meredith Wagner, Kelly Keefe, Joe Veldm an, Geraldo Silva, Simone Athayde, JG Collomb, Christine Lucas, Camila Pizano, Margo Stoddard, Christine and John Engels, Santiago Espinosa, Richard Wallace, Mary Menton, Alfredo Rios, Leo Martinez, Cynthia Gomez, Pablo Solano, Andrea Chavez Paulo Brando, Leonardo Pacheco, Paula Pinheiro, Ane Alencar, Lucim ar Souza, and the whole galera dos brasileiros ; Amy Waterhouse, Kathleen McKee, Jeff Hoelle Shravanthi Reddi, Hollie Hall, Kelly Biedenweg, Deb Wojcik, Allie Shenkin, Matt Palumbo, Cerian Gibbes, Tracy Van Holt, Jaclyn Hall, Rosana Resende, Omaira Bolanos Kato Caas, and many others. I am deeply appreciative of Valrio Gomes for his love, support, and great sense of humor. 6

PAGE 7

Finally, this experience would never have been possible without the support and patience of my family. I cant even begin to express gratitude fo r my parents, Richard and Anne Duchelle, who prioritized my educ ation and managed to ground me while allowing me total freedom to pursue my path. I thank my brother s and sisters: Liz and John Schuller, Kate Duchelle and Doug Sm ith, Mike and Laura Duchelle, and Marty and Nicole Duchelle for being tolerant of my travels, providing enc ouragement along the way, and reminding me of the im portant things in life. I am so proud of my nieces and nephews Lauren, Nora, Connor, Sergey, Reil ly, Polina, Aidan, and Liam. Thank you all for everything. 7

PAGE 8

TABLE OF CONTENTS page ACKNOWLEDG MENTS .................................................................................................. 4LIST OF TABLES .......................................................................................................... 12LIST OF FI GURES ........................................................................................................ 14LIST OF ABBR EVIATION S ........................................................................................... 16ABSTRACT ................................................................................................................... 18 CHAPTER 1 INTRODUC TION .................................................................................................... 20Statement of t he Problem ....................................................................................... 21Sample of Co mmunities .......................................................................................... 23Land Use Land Cover Change in Co mmunity Managed Forests ............................ 24Property Rights Security of Brazil Nut Producer s ................................................... 25Certification of Brazil Nu ts Promotes Better Management and Increases Income .. 26Importance of the Study .......................................................................................... 262 CONSERVATION IN AN AMAZON IAN TRI-NATIONAL FRONTIER: PATTERNS OF LAND COVER CHA NGE IN COMMUNITY-MANAGED FOREST S ............................................................................................................... 28Introducti on ............................................................................................................. 28History of Settlement in the MAP R egion ......................................................... 31Recent Poli cy Change s .................................................................................... 33Regional Road Development ..................................................................... 33Madre de Dios, Peru .................................................................................. 35Acre, Brazil ................................................................................................. 37Pando, Boliv ia ............................................................................................ 39Methods .................................................................................................................. 40Sample Comm unities ....................................................................................... 41Remote Sensing Analysis ................................................................................. 41Survey Data ...................................................................................................... 44Modeli ng ........................................................................................................... 45Results .................................................................................................................... 47Land Cover Change ......................................................................................... 47Patterns of Deforestation .................................................................................. 48Household-level Variables and Reported Forest Clearin g ................................ 49Comparative household characteri stics, income and assets ..................... 49Explaining reported forest clear ing ............................................................. 50Discussio n .............................................................................................................. 51 8

PAGE 9

Forest Cover Patte rns and Tr ends ................................................................... 51Reported Forest Cleari ng ................................................................................. 54Differences in Observed vs. Reported Defore station ....................................... 58Conclusi on .............................................................................................................. 593 FORMALIZING PROPERTY RIGH TS OF TROPICAL FOREST COMMUNITIES: IMPLICATIONS FOR NON-TIMBER MANAGEMENT, LIVELIHOODS AND CO NSERVATION .................................................................. 74Introducti on ............................................................................................................. 74Study R egion .......................................................................................................... 78Brazil Nut Bridging Forest Conserva tion and Livelihood Development ............ 78History of Forest Extrac tion .............................................................................. 79Rubber Industry Sets the Stage for Brazil Nut Extraction in Western Amazonia ...................................................................................................... 81Recent National Policies Designed to Devolve Rights to Extractivist Communiti es ................................................................................................. 82Methods .................................................................................................................. 85Field Evaluation of Brazil Nut Ha rvest and Management Practices .................. 85Quantification of Ru ral Liveli hoods ................................................................... 86Data Anal ysis ................................................................................................... 86Potential quantitative explanations fo r theft ............................................... 87Potential quantitative im plications of theft on harvest and management .... 87Results .................................................................................................................... 88Comparative Household Characterist ics, Land Tenure, Access and Income Data .............................................................................................................. 88Frequency of Reported Br azil Nut Thefts ......................................................... 89Financial Value of Br azil Nut T hefts ................................................................. 90Quantitative Explanations of Brazil Nut Thefts ................................................. 90Quantitative Implications of Theft on Brazil Nut Harvest Regimes ................... 91Resolution of Thefts ......................................................................................... 92Community Perceptions of Partic ipatory Mapping in Pando ............................. 92Discussio n .............................................................................................................. 93Comparative Timing and Process of Property Rights Formalization ................ 94Settlement Patterns and Resource Dependence Infl uence The ft ..................... 95Implications of Theft Threat and Economic Dependence on Brazil Nut Harvest and Management ............................................................................. 97Participatory Mapping to Promot e Property Rights Security ............................. 99Conclusion ............................................................................................................ 1004 DOES CERTIFICATION LEAD TO BETTER MANAGEMENT AND IMPROVED INCOMES? A COMPARATIVE ANALYSIS OF THREE CERTIFICATION SCHEMES APPLIED TO BRAZIL NUTS IN WESTERN AM AZONIA ................... 116Environmental, Social and Economic Certification of Fore st Products ................. 116Brazil Nut Ecology and Best Management Prac tices ............................................ 118 9

PAGE 10

Diverse Brazil Nut Certification Systems with Disti nct Goals ................................ 120Organic and Product Quality Certificat ion....................................................... 120Fair Trade Certif ication ................................................................................... 122Forest Management Ce rtification ................................................................... 125Methods ................................................................................................................ 127Communities Sa mpled ................................................................................... 127Variables M easured ....................................................................................... 128Data Analys is ................................................................................................. 128Comparative management practices and a ssociations with certification 128Comparative economic vari ables and associations with certification ....... 129Results .................................................................................................................. 129Comparative Management Practice s ............................................................. 129Associations Between Brazil Nut Certification and Management ................... 131Income Derived From Organically-Certifi ed and Non-Certified Brazil Nuts .... 133Producers Perceptions of Br azil Nut Certif ication .......................................... 134Discussion and Conc lusion ................................................................................... 135Certification and Best Br azil Nut Managem ent ............................................... 135Economic Benefits of Certificat ion .................................................................. 137Partnerships for Certif ication are Key ............................................................. 139Benefits and Challenges Associated wit h Three Certificat ion Schemes ......... 1415 CONCLUSION ...................................................................................................... 151 APPENDIX A ANNUAL VILLAGE SURV EY 1 (PORTUGU ESE) ................................................ 157B ANNUAL VILLAGE SURV EY 2 (PORTUGU ESE) ................................................ 163C ANNUAL HOUSEHOLD SURVEY 1 (PORTUGUESE) ......................................... 165D ANNUAL HOUSEHOLD SURVEY 2 (PORTUGUESE) ......................................... 170E QUARTERLY HOUSEHOLD SU RVEY (PORTU GUESE) .................................... 173F HOUSEHOLD ATTRITION S URVEY (PORTUGU ESE) ....................................... 185G HOUSEHOLD BRAZIL NUT MANAGEME NT SURVEY 2006 (PORTUGUESE) .. 186H HOUSEHOLD BRAZIL NUT MANAGEME NT SURVEY 2007 (PORTUGUESE) .. 190I LANDSAT SCENES USED FOR LAND USE LAND COVER CHANGE ANALYSIS IN BRAZIL NUT-PRODUCING COMMUNITIES IN THE MAP REGION ................................................................................................................ 193J DISTANCE FROM ROAD AND DEFO RESTATION RATES FOR MADRE DE DIOS, PERU (GREEN), ACRE, BRAZ IL (PINK) AND PANDO, BOLIVIA (ORANGE) ........................................................................................................... 194 10

PAGE 11

K RESULTS OF UNIVARIATE REGRESSIONS FOR POTENTIAL EXPLANATORY VARIABLES ON LOG FOREST CLEARED FOR COMMUNITIES IN EACH COUNTRY IN THE TIME PERIOD 2002-2007. .......... 195LIST OF REFE RENCES ............................................................................................. 196BIOGRAPHICAL SK ETCH .......................................................................................... 213 11

PAGE 12

LIST OF TABLES Table page 2-1 Policy events relevant to each land use land cover change time series in Madre de Dios, Peru, Acre, Brazil, and Pando, Bolivia over the 1986-2005 study peri od. ....................................................................................................... 662-2 Description of cash and subsiste nce income categories. Inputs were discounted from net inco me calculat ions. ........................................................... 672-3 Regional land cover in select Braz il nut producing communities 1986-2005. ..... 682-4 Regional change trajectories in select Brazil nut producing communities. ......... 692-5 Binomial logit models of deforestation in Brazil nut communities during period 2000-2005. ......................................................................................................... 702-6 Non-forest patch metrics in 2000 and 2005: NP (number of patches); AREA (size); CIRCLE (circularity index); CONTIG (contiguity index); ENN (Euclidean distance to nearest nei ghbor) ........................................................... 712-7 Descriptive statistics of household characteristics, income, assets and reported forest clearing in Brazil nut -rich communities in Madre de Dios, Acre, and P ando. ................................................................................................ 722-8 Results of multivariate regression model with the most important measured variables that explained log amou nt of forest cleared from 2002-2007 (R2=0.585) .......................................................................................................... 733-1 Important historical events and relevant forest policy changes in Acre, Brazil, and Pando Bolivia ............................................................................................. 1123-2 Descriptive statistics of household c haracteristics, land tenure, access and income in Pando and Acre .............................................................................. 1133-3 Total volume and value of Br azil nuts reported stolen in 2006 and 2007 combined .......................................................................................................... 1143-4 Results of a stepwise logistic regr ession to identify the most important measured variables that explained incide nce of Brazil nut thefts in 2006 and 2007 in Pando and Acre (n=235) ...................................................................... 1153-5 Mean and SE of Brazil nut harvest start date (expressed in Julian days where 1=November 8th), quantity of Brazil nuts collected (kg) and harvest length (days) in three Brazil nut harvest contexts in Bo livia and Brazil in 2006 and 2007. ................................................................................................................ 115 12

PAGE 13

4-1 Best management practices for Brazil nut as outlined in regional extension literature, along with justifications and supporting research for the performance of each practice. .......................................................................... 1474-2 Brazil nut certification initiatives in the tri-national region of Western Amazonia ......................................................................................................... 1484-3 Odds ratios (OR) and p-values from Pearson Chi-square tests of association between selling certified nuts and bes t management prac tices. ....................... 1494-4 Odds ratio (OR) and p-value from P earson Chi-square test of association between nut certification and no debt or payments in advance ........................ 150 13

PAGE 14

LIST OF FIGURES Figure page 2-1 Map of Madre de Dios, Peru, Acre Brazil and Pando, Bolivia (MAP region) with most recent portion of Interoceani c Highway and 100 m buffer in red. ....... 612-2 Map of communities sampled in Madre de Dios, Acre and Pand o. .................... 622-3 Land cover change trajectories 19862005 in select co mmunities. .................... 632-4a Area deforested in pai red-date time periods in Brazil nut producing communities in MAP. .......................................................................................... 642-4b Area reforested in paired-date ti me periods in Brazil nut producing communities in M AP ........................................................................................... 642-5 Income share distribution among samp led households in the three countries. .. 653-1 Map of Madre de Dios, Peru, Acre Brazil and Pando, Bolivia (MAP region) with Interoceanic Highway and 10 0 m buffer in red. ......................................... 1033-2 Google Earth images from July 2007 of forest dwelling communities in Acre, Brazil (left) and Pando, Bolivia (ri ght) ............................................................... 1033-3 Map of communities sampled in Acre and Pando. ............................................ 1043-4 Share of total househol d income derived from the forest (%) and from Brazil nuts alone (%) .................................................................................................. 1053-5 Where Brazil nuts were reportedly stolen (% of tota l incidents) ........................ 1063-6 Suspected perpetrators of Brazil nut thefts in 2006 and 2007 (% of total incidents) .......................................................................................................... 1073-7 Modal start and end dates of Brazil nut harvest, and harvest duration, in Pando and Acre in 2006 and 2007 .................................................................. 1083-8 Percentage of households using specif ic Brazil nut harvest methods among 3 groups: (1) Bolivia without nut thefts; (2 ) Bolivia with nut thefts; and (3) Brazil without nut t hefts .............................................................................................. 1093-9 Free-listed benefits of participatory mapping in P ando, Bolivia (45 producers; 55 benefits given). ............................................................................................ 1103-10 Example of a participatory map of i ndividual Brazil nut tr ees in Pando ............ 1114-1 Map of sampled communities .......................................................................... 143 14

PAGE 15

4-2 Brazil nut producers at the door of their storage units in Bolivia (top left) and Brazil (bottom left) ............................................................................................ 1444-3 Percent and SE of Brazil nut pr oducers practicing best management practices in Pando, Acre and Madre de Dios.................................................... 1454-4 Producer prices received when selli ng Brazil nuts to middlemen versus certified cooperativ es in 2007 ........................................................................... 146 15

PAGE 16

LIST OF ABBREVIATIONS ACCA Association for the Conservation of the Amazon Watershed ACERM Association of Extracti vists of the Manuripi Reserve ACEBA Association of Extractivists of Bolpebra ASCART Association of Brazil nut Producers of the Tahuamanu Reserve ASECAMB Association of Brazil nut Extractivists of Madre de Dios AOPEB Bolivian Association of Organic Ecological Producers CAPEB Agroextractivist and Agropasto ral Cooperative for Producers of Brasilia and Epitaciolndia CAEX Agroextractivist and Agropasto ral Cooperative for Producers of Xapuri CIRCLE Patch circularity index COINACAPA Integral Cooperative of Rural Agroextractivists in Pando COOPERACRE Cooperative of Acre CONTIG Patch contiguity index CIFOR Center for Internat ional Forestry Research ENN Euclidean Distance to Nearest Neighbor FAO Food and Agriculture Organization FLO Fair Trade Labelling Organizations FSC Forest Stewardship Council GIS Geographical Information Systems IBNORCA Bolivian Quality and Normalization Institute IDB Inter-American Development Bank IFOAM International Federation of Organic Agriculture Movements IMAFLORA Institute of Forest Man agement and Certificat ion and Agriculture IMO Institute for Marketecology 16

PAGE 17

ISEAL International Social and Enviro nmental Accreditation and Labeling LULCC Land Use Land Cover Change MAP Madre de Dios, Peru, Acre, Brazil, Pando, Bolivia NGOs Non-Governmental Organizations NP Number of patches NSF-HSD National Science Foun dation Human Systems Dynamics NTFPs Non-Timber Forest Products REDD Reduced Emissions from Deforestation and Degradation RIL Reduced Impact Logging RONAP Organic Collectors of Amazon Nuts of Peru PEMD Special Project of Madre de Dios PEN Poverty and Environment Network WHO World Health Organization WWF World Wildlife Fund 17

PAGE 18

Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for t he Degree of Doctor of Philosophy CONSERVATION AND LIVELIHOOD DEVEL OPMENT IN BRAZIL NUT-PRODUCING COMMUNITIES IN A TRI-NATION AL AMAZONIAN FRONTIER By Amy Eleanor Duchelle December 2009 Chair: Karen A. Kainer Major: Forest Resources and Conservation This dissertation examines the relationship between forest conservation, forest dependence, and livelihood development of extractive communiti es in the tri-national frontier region of Madre de Dios, Peru, Ac re, Brazil, and Pando, Bolivia in Western Amazonia. While the focus of th e dissertation is on Brazil nut ( Bertholletia excelsa ), this regionally-important non-tim ber forest product (NTFP) is used to illustrate broader issues related to community-based fo rest management. This study assessed community-based forest management from three different per spectives: (1) an analysis of land use land cover change in community-managed forests through the use of satellite and survey data; (2) an exploratio n of the causes and ef fects of Brazil nut thefts, and resolution of such thefts; and, (3) a comparison of the environmental and economic outcomes of organic, Fair Trade, and Forest Stewardship Council (FSC) certification of Brazil nuts. The first set of results dem onstrated minimal deforestation and extremely high forest income dependency in extractive commu nities. In 2000 2005, deforestation occurred in already fr agmented areas and alo ng roads, with most deforestation in Acre and least in Pando. From 2002 to 2007, higher agricultural 18

PAGE 19

19 income was positively correlated with reported forest cleari ng at the household level in Pando and Madre de Dios, wher eas higher Brazil nut income was correlated with less forest cleared. In Acre, government aid, larger households, and higher value of livestock assets were correlated with forest clearing. The second set of results highlighted more Brazil nut thefts in Pando when compared with Acre, likely due to land titling processes in Pando that disregarded traditional forest use, settlement patterns that disconnected producers from their forest re sources, and higher nut dependence. Both threat of theft and resource dependence affected nut harvest regimes in Pando. The third set of results showed that or ganic and Fair Trade Brazil nut certification schemes were associated with better post-har vest practices and higher prices, while producers certified by the FSC in Peru adopted practices related to pre-harvest planning and tree health. These findings contribute to our understanding of the role of community-based forest management in tropic al forest conservation and livelihood development.

PAGE 20

CHAPTER 1 INTRODUCTION This dissertation examines the relationship between forest conservation, forest dependence, and livelihood developm ent of extractive communities in Western Amazonia. While the focus of th e dissertation is on Brazil nut ( Bertholletia excelsa ), this regionally-important non-tim ber forest product (NTFP) is used to illustrate broader issues related to community-based forest management. The research is presented as three separate papers prepared fo r publication in academic journals. Each paper can be read as an independent document, which addresses a unique aspect of the overarching study. The firs t paper entitled Conservation in an Amazonian tri-national frontier: patterns and trends of land cove r change in community managed forests provides an overview of forest cover change in road accessible communities from 1986 and explores the patterns and drivers of recent deforestation. The second paper entitled Property rights devolution to tr opical forest communities: implications for non-timber management, income generation and forest conservation investigates the effects of property rights devolution, settl ement patterns, and resource dependence on Brazil nut thefts, along with t he harvest and management implicat ions of such thefts. The third paper entitled Does certification lead to better management and improved incomes? A comparative analysis of three ce rtification schemes applied to Brazil nuts in Western Amazonia compares the environm ental and economic benefits associated with different nut certification schemes. Each paper is the result of collaborations with colleagues at the University of Florida and with research partn ers in Western Amazonia. While this dissertation represents my or iginal work, select collaborators will be recognized for their contributi ons as co-authors on the indivi dual publications. For that 20

PAGE 21

reason, the first person plural (we) is us ed in this dissertation to recognize these collaborators supporting roles. Statement of the Problem Tropical forests are renowned for their biol ogical diversity and role in regional hydrology and global climate regulation (Laurance, 1999; Fearnside, 1997) as well as for their contributions to rural livel ihoods (Wunder, 2001; Sunderlin et al., 2005). Community-based forest management is c onsidered an important strategy for promoting conservation and livelihood developm ent in the tropics. Rural communities now own or manage one-quarter of the worlds tr opical forests due to recent devolution of government-owned lands and formal recognition of customary rights (White and Martin, 2002; Sunderlin et al., 2008a). The goals of community-based forest management on national and international levels are to promote ecologicallysustainable management practi ces, increase the socioeconom ic benefits gleaned from tropical forests, and promote greater access to forest resources (Charnley and Poe, 2007). Recent studies have shown that community-managed forests can be equally, if not more, effective in maintain ing forest cover than strict pr otected areas (Carrera et al., 2004; Nepstad et al., 2006, Ellis and Porter-Bo lland, 2008; Bray et al., 2008). That said, many researchers have highlighted the tremendous challenges associated with community-based forest management, includi ng the oversimplified definition of community (Agrawal and Gibson, 1999), lack of attention to the multiple levels of governance that affect community management of natural resources (Berkes, 2007), discrepancies between formal and customar y resource rights (Fortmann and Bruce 1988; Meinzen-Dick and Mwangi, 2008), difficult ies in linking communities to markets 21

PAGE 22

(Schmink, 2004; Scherr et al., 2005), and cha llenges in balancing conservation with livelihood development through fo rest-based activities (Kusters et al., 2006). Although forestry has long focused on tim ber-dominated models, in the late 1980s, community management of non-timber fore st products (NTFPs) began to gain widespread attention as a way to reconcil e seemingly contradict ory conservation and development goals for forest-dwelling communiti es in the tropics. NTFP extraction was thought to be less ecologically destructive than other land-use practices, such as clearing land for agriculture and pasture, or harvesting timber (Arnold and Ruz-Prez, 2001). Enthusiastic studies presented ove r-inflated potential re venues from NTFPs compared with more environmentally-destructi ve land uses (Peters et al., 1989). Notably, more recent comparative studies on an integrated global set of NTFP cases show that rarely are NTFPs able to meet both conservation and development goals, and that an inverse relationship between the two is us ually more reflective of reality (Belcher et al. 2005; Kusters et al., 2006). While NT FPs may be debunked as silver bullets for reconciling conservation and development go als, they are still considered important safety nets for rural communities (Angelsen and Wunder, 2003) and essential components of diverse forest-based livelihood systems, which span multiple scales from households to communities to landscapes (Ehringhaus, 2006). Non-indigenous, extractive communities, whose principal livelihood activity is collection of forest products, occupy nearly one-third of forests in the MAP tri-national frontier region in Western Amazonia, which is comprised of the stat es of Madre de Dios, Peru, Acre, Brazil, and Pando, Bolivia (Z EE, 2006; INRA, 2009; SPDA-INRENA, 2003). We focus our comparison on these three adjacent areas, because while resident 22

PAGE 23

communities have a similar natural resour ce base, their property rights systems, specific livelihood strategies, and forest management regimes are different. Also, construction of the Interoceanic Highway, an extension of the Brazilian BR-317 into Peru and Bolivia, is changing conditions in this formerly remote region, by providing regional access to Pacific ports. Many ex tractive communities in the MAP region produce Brazil nuts, which are co llected primarily from mature forests. The combined ecological and economic characteristics of Bertholletia excelsa make it a species with the potential to promote forest conservation wh ile contributing to rural livelihoods. In this dynamic context, it is essential to examine the role that Brazil nuts play in communities conservation of forests and livelihood development, illuminate the institutional challenges associated with this products shortand long-term management, and examine how market-based mechanisms, such as certification, can add value to Brazil nut-rich forests to promote their conservation. Sample of Communities This research focuses on communities in the Brazil nut-rich MAP region in Western Amazonia. In collaboration with the Center for Internati onal Forestry Poverty and Environment Network (CIFOR PEN) ( http://www.cifor.cgiar.or g/pen/_ref/home/index.htm) and fellow PEN partner Angelica Almeyda from Stanford University, tw o annual village and hous ehold-level socioeconomic and environmental questionnaires were applied to a total of 453 households in 44 communities in Madre de Dios, Acre, and Pando from June 2006 through October 2007 (Appendices A, B, C, and D). These communities r epresented both extractive communities and colonist farme r associations. Additionally, over a 12-month period, four quarterly household questionnaires were conducted with 244 households and three 23

PAGE 24

trimestral surveys conducted with 209 hous eholds (Appendices E and F). These questionnaires, which were designed to quantitatively assess the role of forest income in rural livelihoods, were based on the work of Cavendish (2000). In addition to the PEN questionnaires, in 15 of the extrac tive communities, Brazil nut management practices were evaluated from January 2006 through September 2007 (Appendices G and H). We also evaluated nut management practices in two additional extractive groups, the Chico Mendes Agro-Extractive Settlement Project (commonly known as Cachoeira) in Acre and ASCART (Associacin de Castaeros de la Reserva Tambopata) in Madre de Dios, which were not included in the PEN study, but had unique Brazil nut production histories. In the extractive communities, we collected 216 training samples in the dry season (June-Sept ) of 2006, and 236 in 2007, in forest and non-forest (including agriculture, pasture and re sidential) areas for use in satellite image analysis. Based on the unique objectives of each paper, along with available data, we used different subsamples of extractive co mmunities from the larger set and provided clear justifications fo r our selection. Land Use Land Cover Change in Community Managed Forests The first paper (Chapter 2) provides a comparative land use land cover change (LULCC) analysis in community-managed Brazil nut-rich forests along roads in Madre de Dios, Acre, and Pando. While these thre e adjacent areas have a similar natural resource base, we explore how historical processes, policy changes, and livelihood strategies have differentially affected forest conservation by communities. Our overall research question in this chapter is: What is the relationship between forest clearing and sources of livelihood income in extractive co mmunities of Brazil nut-rich forests in the MAP region? 24

PAGE 25

This study highlights the utility and challenge of combining spatial and survey data to understand the patterns and trends of observed deforestation and potential drivers of reported forest clearing in community-managed forests. The specific objectives of this study are: (1) to compare trends of deforestation and reforestation in extractive communities in neighboring Madre de Dios, Acre, and Pando over a 20-year period (1986); (2) to analyze the spatial patterns and causes of deforestation at the pixel level (i.e., 30 x 30 m) in these communities during the final time period (2000); and, (3) to analyze the socioeconomic causes of reported forest clearing at the household level in 2002. This research illustrates broader issues related to the role of rural communities in tropi cal forest conservation. Property Rights Security of Brazil Nut Producers The second paper (Chapter 3) examines t he phenomenon of Brazil nut thefts that occurs in extractive communities in Pando, but not in Acre. Our overall research question in this chapter is: W hat are the causes and effects of Brazil nut thefts, and how can such thefts be resolved? In this chapter, we focus our comparis on on the adjacent areas of Pando and Acre because, while they have a similar natural re source base, their property rights systems, settlement patterns, and depend ence on Brazil nuts by local communities differ. We first introduce the study region and ecological and economic characteristics of Brazil nut. We next highlight impor tant historical processes and policy changes that have shaped settlement patterns and property rights for communities in Pando and Acre. We then present results of field studies on repor ted thefts of Brazil nuts and explore the causes of such thefts, along with thei r harvest and management implications. Our specific objectives are to evaluate: (1) the ex tent that Brazil nut thefts occur in forest25

PAGE 26

dwelling communities in Pando and Acre; (2 ) the financial value of these thefts; (3) the causes of such thefts; (4) how threat of theft affects Brazil nut harvest and management practices; (5) how thefts are resolved; and, (6) producers perceptions of participatory mapping as one way to resolve Brazil nut thefts. Certification of Brazil Nuts Promotes Better Management and Increases Income The third paper (Chapter 4) compares organi c, Fair Trade and Forest Stewardship Council (FSC) certification, and addresses the challenges associated with the different schemes as applied to Brazil nuts in Pando, Acre, and Madre de Dios. Our overall research question in this chapter is: What are the environmental and economic benefits associated with different Brazil nut certification schemes? In this chapter, we first provide an ov erview of Brazil nut ecology and best management practices for this product, as outlined by regional government and nongovernmental organizations (NGOs). We next present the th ree main nut certification schemes available to regional producers. We then present results of fieldwork that highlight associations between the differ ent Brazil nut certif ication schemes and management and income generation outcomes for producers. Our specific objectives are: (1) to identify performance of Brazil nut best management practices by producers in the three countries; (2) to analyze di fferences in management practices between certified producers and noncer tified producers in each country; (3) to quantify economic benefits associated with produc ing certified nuts; and, (4 ) to explore producers perceptions of the benefit s and costs of Brazil nut certification. Importance of the Study Together these papers contribute to a br oader understanding of the conservation and livelihood benefits associ ated with community-based forest management in 26

PAGE 27

Western Amazonia. Theory and methods from the fields of forestry, geography, and economics were used to develop this research in response to pertinent conservation and development concerns. The compar ative approach between three countries allowed for constant reflection on how differi ng historical processes, policy changes, and livelihood strategies shaped three unique community forest management contexts. This study highlights the relationship betwe en forest conservation and forest income dependence by communities in a global context. The results of this research also have application for government agencies and NGOs in Western Amazonia interested in supporting community-based forest managemen t, and can provide new insights to Brazil nut producers through t he tri-national comparison. Most importantly, this dissertation highlights how community managem ent of a high-value NTFP can promote conservation and livelihood development on dy namic tropical forest frontiers. 27

PAGE 28

CHAPTER 2 CONSERVATION IN AN AMAZONIAN TR I-NATIONAL FRONTIER: PATTERNS OF LAND COVER CHANGE IN COMMUNITY-MANAGED FORESTS Introduction Community-based forest management is c onsidered an important strategy for promoting conservation and livelihood development in the tropics. Rural communities now own or manage one-quarter of the worlds tr opical forests due to recent devolution of formerly government-claim ed lands and formal recognition of customary rights (White and Martin, 2002; Sunderlin et al., 2008a). The goals of community-based forest management on national and international levels are to promote ecologicallysustainable management practi ces, increase the socioeconom ic benefits gleaned from tropical forests, and promote greater access to forest resources (Charnley and Poe, 2007). Recent studies have shown that di fferent types of community-managed forests can be equally, if not more, effective in main taining forest cover than strict protected areas (Nepstad et al., 2006; Ellis and Porter -Bolland, 2008; Bray et al., 2008). That said, many researchers have highlighted the tremendous challenges associated with community-based forest management, includi ng the oversimplified definition of community (Agrawal and Gibson, 1999), discrepancies between formal and customary resource rights (Fortmann and Bruce, 1988; Meinzen-Dick and Mwangi, 2008), difficulties in linking communities to mark ets (Schmink, 2004; Scherr et al., 2005), and challenges in balancing conservation with livelihood development through forest-based activities (Kusters et al., 2006). Communities roles in tropical forest conservation are largely based on the social, political and economic contexts in which t hey are embedded. Much of the literature on deforestation by smallholders has focused on its multi-scalar and interacting drivers 28

PAGE 29

(e.g., Wood, 2002; Schmink, 1994). Proximat e socioeconomic causes of deforestation include infrastructural development, agricul tural expansion, and timber extraction; whereas underlying drivers include demographic, economic technological, policy, institutional, and cultural fa ctors (Geist and Lambin, 2002). Of 152 case studies on the drivers of tropical deforesta tion, economic factors were present in 81% (Geist and Lambin, 2002). Economic returns associated with different land uses often affect small producers decisions about w hether to maintain standing forest or convert it for agricultural or pastoral uses (Angelsen 20 06 in Chomitz 2007). Economic models of deforestation have shown that tropical defor estation is higher when land is more accessible, prices are higher for agricul tural products and tim ber, off-farm income opportunities are lower, and long distance tr ade opportunities are higher (Kaimowitz and Angelsen, 1998; Pfaff, 1999; Lambin et al., 2001; Angelsen 2006 in Chomitz 2007). It is essential to understand the br oader context in which forest-based communities are embedded, along with their diverse livelihood st rategies, to appreciate their decisionmaking in regards to forest conservation. The Amazon is the largest tract of conti guous tropical forest in the world, alone containing half of global biological diversit y and one-fourth of its primary productivity (Soares-Filho et al., 2004). In the approximately 300,000 km2 border region of Madre de Dios, Peru, Acre, Brazil, and Pando, Boliv ia (MAP region) in Western Amazonia, communities own or manage nearly one-third of regional forests (ZEE, 2006; INRA, 2009; SPDA-INRENA, 2003). Many of these communities base their livelihoods on Brazil nuts ( Bertholletia excelsa ), the most important non-timber forest product (NTFP) in the region. These Brazil nut-producing communities share the landscape with 29

PAGE 30

recently settled colonist farmers, cattl e ranchers, and loggers. Although the entire region is characterized by lowland wet tropical forest vegetation, settlement histories, patterns of deforestation, public policy, and socio-economic development vary considerably from one country to the next (www.map-amazonia.net ). Construction of the Interoceanic Highway, an extension of the Brazilian BR-317 into Peru and Bolivia, is changing conditions in this formerly remote region, by providing regional access to Pacific ports (Figure 2-1). In Acre, forest conversion has been rapid, extensive, and largely driven by establishment of cattle ranches (Souza et al., 2006). In Madre de Dios, the deforestation process has been slower and patchier than in Acre, and dominated by small farms, al though recent paving has concentrated land ownership and increased deforestation (Chavez, 2009). In Pando, deforestation has been minimal, with most land conversion occurring in close proximity to populati on centers and along the Brazilian border (Marsik et al., in review). Our overall research question is: What is the relationship between forest clearing and sources of livelihood income in extractive communities of Brazil nut-rich forests in the MAP region? To address this question, we present a comparative land use land cover change (LULCC) analysis in community-managed forests along roads in Madre de Dios, Acre, and Pando. While these thre e adjacent areas have a similar natural resource base, we explore how historical processes, policy changes, and livelihood strategies have differentially affected forest conservation by these communities. Through combining spatial and so cioeconomic analyses, the spec ific objectives of this study are: (1) to compare trends of deforestation and reforestation in extractive communities in the MAP region over a 20year period (1986); (2) to analyze the 30

PAGE 31

spatial patterns and causes of deforestation at the pixel level (i.e., 30 x 30 m) in these communities during the final time per iod (2000); and, (3) to analyze the socioeconomic causes of reported forest clearing at the household level in 2002. This study illustrates broader issues related to the role of rural communities in tropical forest conservation. History of Settlement in the MAP Region The history of colonization and settlement by mostly non-indigenous extractive populations in the MAP region began in the la te 19th century during the first boom of natural rubber (mostly Hevea brasiliensis but also Castilla spp. ) Immigration to rubber estates in the Western Amazon, into Pando and what is now Acre (which at the time was officially Bolivian territory; Barham and Coomes, 1996) greatly expanded in the 1880s after Edwin Heath, a North American doctor, discovered the connection between the upper and lower Beni River (Fifer, 1970). Immigration to Madre de Dios for rubber began a bit later (1893-1895) when Fermn Fit scarrald, a Peruvian explorer who had his rubber base in Iquitos, discovered connections between the Urubamba and ManMadre de Dios Rivers (Fifer, 1970). While the rubber industry in Pando and Acre focused on Hevea brasiliensis most rubber production in Peru came from Castilla spp. locally called caucho (Morcillo, 1982). Caucho was a lesser quality rubber and individual Castilla trees could only be tapped once, but its advantage was that it coul d be extracted during any season. This flexibility resulted in a highly mobile labor force, of prim arily indigenous people in debt peonage, collecting caucho in areas away from rivers that were granted to rubber companies by the Peruvian government (Morcill o, 1982; Chibnik, 1994 ). In contrast, rubber was collected on private estates in Acre and Pando, requiring generally more 31

PAGE 32

stable relations between tapper and patron, as well as investments in estate infrastructure and tapper supplie s (Barham and Coomes, 1996). The tri-national boundarie s between Brazil, Bolivia, and Peru were contested during the rubber boom. The 1903 border war between Brazil and Bolivia, in which the Brazilian rubber tappers in Acre had a major role in Brazils victory, reestablished boundaries between these two countries, and the territory of Acre was ceded to Brazil (Fifer, 1970). In 1909, there were armed conflicts among Pe ruvian and Bolivian rubber tappers, which later resulted in a treaty t hat determined the Peru-Bolivia border equidistant along the Manuripi Rive r from San Lorenzo (Peru) and Illampu (Bolivia). At this same time, Madre de Dios was creat ed as a department with Puerto Maldonado as its capital (Morcillo, 1982). When the price of rubber fell in 1912 in response to the high productivity of established rubber plantations in Mala ysia, Amazonian rubber tappers began to diversify their livelihood strategies to in clude Brazil nuts and agric ulture (Fifer, 1970; Barham and Coomes, 1996; Stoian, 2000). In 1931, the Bolivian Suarez Company introduced Brazil nut shelling, supported by a labor force dominated by women as the Chaco War (1932-35) drained male labor in Bolivia (Fifer, 1970) In Acre, a small Brazil nut factory was established in t he town of Xapuri in 1933 (Kainer et al., 2003). In Madre de Dios, in addition to Brazil nuts, gold was exploited along the Inambari River (Morcillo, 1982). During World War II, there was a second, smaller rubber boom in which the U.S. partnered with Brazil through the Washington Accords, and recruited a second wave of Brazilian rubber soldiers to Acre (Sobrin ho, 1992). Although Pando and Acre were legally separated, the border was porous, and more than 2000 Brazilian rubber tappers 32

PAGE 33

were employed by Suarez and Hermanos, who at the time controlled 80% of rubber production in the Brazil-Boli via border region (Sobrinho, 1992). The only remaining rubber company in Madre de Dios was sold to the U.S. during Worl d War II (Morcillo, 1982), and Brazil nuts continued to be a comple mentary seasonal activity to rubber tapping until the 1950s when nut exports surpa ssed rubber production. The first large price increase in the Brazil nut sector occurred in the 1990s (Stoian, 2000) when this product became the most economically-i mportant NTFP in Amazonia. Recent Policy Changes Several important policy changes occurred at the national level in Peru, Brazil, and Bolivia, which created three unique contexts for community-based forest management. These include road development that has affected the entire region and diverse policies that are more specific to each country (Table 2-1). Regional Road Development The MAP region remained relatively inaccessible until the construction of major roads on all three sides of t he border, which were designed to link these remote states to the rest of their res pective countries (Figure 21). In the mid-1960s, an unpaved highway was constructed from the Andean hi ghlands to Puerto Maldonado, the capital of Madre de Dios, to promot e migration of the landless poor (Dourojeanni, 2006), and in 1979, Peru signed a contract with Brazil to c onstruct an Interoc eanic Highway (CTAR, 1998), which would link southern Brazil to por ts on the Pacific. This Interoceanic highway has become one axis of the Andean Promotion Co rporations (CAF) massive Initiative for the Integration of Infrastruc ture in the South American Region (IIRSA) to facilitate exports to global markets leading toward regional development (IIRSA, 2005; Perz et al., 2008). With funding from Perus Special Project of Madre de Dios (PEMD), 33

PAGE 34

construction of the highway from Puerto Maldonado to the town of Iapari on the border with Brazil began in 1981 (Chavez, 2009). In the 1990s, segments of the unpaved highway were leveled and maintained, and ten years later, it was compacted and prepared for paving, which began in 2006 (Chavez, 2009). The origins of the Interoceanic highway on the Brazilian side also occurred in the mid-1960s with construction of the BR-364 to the state of Rondnia from southern Brazil as part of a national military strategy to maintain control over the Amazon (Cowell, 1990). By 1984, the highway was paved to Porto Velho (t he capital of Rondnia) through the Polonoroeste proj ect funded by the World Bank (Cowell, 1990). The severely negative environmental and social consequences of this project caused the World Bank to suspend its funding in 1985 (S hankland, 1993). Despite initial successes of Acres rubber tapper social movem ent to temporarily stop Inter-American Development Bank (IDB) funds for extension of this highway into Acre, the newly-paved BR-364 from Porto Velho in Rondnia to Rio Branco (the capital of Acre) was opened in 1992 (Shankland, 1993). In 1993, the Brazili an Congress voted to fund the paving of the next critical piece of the Interoceanic highway: the road that became known as the BR-317 from Rio Branco to the border of Peru. Its paving occurred piecemeal until completion in 2000 (Figure 2-1). Recent international attention has focus ed on paving of the Peruvian portion of the Interoceanic highway, which was finally f unded by CAF and the Brazilian and Peruvian governments with a budget of US $890 million (Perz et al., 2008). In 2006, the Brazilian government completed construction of a hi gh-grade steel bridge across the Acre River to connect the Brazilian and Peruvian sides of this transportation corridor. As of 2009, 34

PAGE 35

pavement of the Interoceanic Higway from Iapari to Puerto Maldonado was nearly complete and a long-awaited bridge over Perus Madre de Dios River was under construction. While Pando currently has only 30 km of paved road, national and international integration plans are underway. The terrestrial connection between Cobija, the capital of Pando, and La Paz is via the city of Riberal ta in the department of Beni (Figure 2-1). The 1414 km segment from Riberalta to La Paz was constructed in the 1980s, and the 310 km segment from Riberalta to Cobija completed in 1992 (UND P, 2003). There are four major river crossings along this route, most navi gated by barge, but in 2006 a bridge was built across the Tahuamanu River in Pando. Another road in Pando, from Cobija through Manuripi National Wildlife Reserve to the Madre de Dios River, was improved in 1998 from a former horse trail (J. Rojas, pers. comm.; M. Zentano, unpublished data). Although this road was temporarily considered the most direct highway route to La Paz, it was never developed as such. A final road in Pando called the Bioceanica, which was completed in 2001, runs east from 19 km south of Cobija to Peru where it joins the Interoc eanic highway (J. Rojas, pers. comm .). In 2004, Brazil funded construction of a bridge across the Acre River to Cobija, and as of 2009, the 86 km stretch of road from Riberalta to Guayaramrin (UNDP, 2003), which borders Rndonia, Brazil, was being paved. Madre de Dios, Peru In Peru, recent agrarian and economic po licies influenced settlement in Madre de Dios (Table 2-1). In addition to road dev elopment in the mid-1960s, an agrarian development program in the late 1980s led to thousands of colonist s from the highlands arriving in Madre de Dios in 1985-1990 to cu ltivate small landhold ings (Alvarez and 35

PAGE 36

Naughton-Treves, 2003; Naughton-Treves, 2004). This program promoted access to rural credit from the Agrarian bank, secured land tenure, and led to the creation of farmers cooperatives (Coomes, 1996). As infl ation rates rose rapidly, colonists who benefited from agrarian credit began to raise ca ttle as a more secure investment than annual crops (Coomes, 1996). With new structural adjustment policies, implemented in 1990, agricultural credit dried up and lands along the Interoceanic highway were subsequently abandoned (Naughton-Treves, 2004) Throughout these periods, the Special Project of Madre de Dios (PEMD), wh ich began in 1980, contributed to land use change through its support of infrastructure development, along with agricultural expansion and technical support (Dour ojeanni, 1990; Chavez, 2009). Beginning in 2000, forestry concessions and protected areas were consolidated in Madre de Dios with the development of the Forestry Law and formalization of Tambopata National Reserve and Bahuaja Sonene Na tional Park. Prior to the Forestry Law of 2000, forest policies were limited and unregulated timber activity was the norm (Chavez, 2009). The laws implementation in 2002 established long-term concessions for both timber (5,000-10,000 ha management units) and Brazil nuts (~500 ha units), both of which required forestry management plans (SPDA-INRENA, 2003). Brazil nut concessions are managed by small produc ers, and most are located along the Interoceanic highway. Under the new protec ted area policy for Madre de Dios, a group of Brazil nut harvesters was allowed acce ss to the buffer zone of Tambopata National Reserve during the nut harvest as long as they abided by strict regulations (SPDAINRENA, 2003). Despite these radical chan ges in the forestry sector, long-term governance failures in Peru have been blamed for continued illegal logging activities in 36

PAGE 37

concessions, since producers norms were unchanged when the new Forestry Law was implemented (Smith et al. 2006). Acre, Brazil In Acre, extractive communities have been affected by recent policies for forestbased development (Kainer et al., 2003). In res ponse to the rubber tappers struggle for secure land tenure and international pressu re to halt Amazonian deforestation, the Brazilian government established the first official Extractive Reserve in Acre in 1990 (Schwartzman, 1989; Hall, 1997) (Table 21). Extractive Reserves have been championed as a viable and sustainable alte rnative to widespread deforestation in the Amazon (Allegretti, 1989; 1990). An important difference between Extractive Reserves and other Amazonian protected areas is that they were created not despite but because of people (Ehringhaus, 2006). Acre s rubber tappers used international environmental concerns about the Amazon to their advantage, portraying themselves as forest stewards (Schmink and Wood, 1992). The Extractive Reserve policy guaranteed usufruct rights to people engaged in tr aditional livelihoods based largely NTFP extraction (e.g. rubber, Brazil nuts, various fruits, and palms; Ehringhaus, 2006), while requiring residents to maintain at least 90% of their landholdings in forest cover (Fearnside, 2003). In 1990, t he almost one million hectare Chico Mendes Extractive Reserve (CMER) was created in Acre from 42 former rubber estates. Households are spread throughout the mostly forested l andscape, and the unique tree tenure legacy from the rubber era is st ill honored among extractivists. Reserve residents define individual property holdings by the number and distribution of rubber trails through the forest (Ankerson and Barnes, 2005), even though Brazil nut has replaced rubber as the most important commercial forest product (Wallace, 2004; Ehring haus, 2006). 37

PAGE 38

The Acrean state government, known as the Forest Government, assumed power in 1998 and is credited with several policy changes toward supporting forest-based communities. Acres government has work ed to re-stimulate the rubber economy, which declined after federal price supports were removed (Hall, 1997). They have accomplished this in two ways: (1) implem entation of the Chico Mendes Law in 1999, which subsidized rubber tappers per kilogram of rubber produced (Kainer et al., 2003); and, (2) creation of a condom factory in Xapuri in 2006 to stimulate demand for locallyproduced natural rubber. The government al so performed Ecological-Economic Zoning (ZEE, 2000; ZEE, 2006), created two new ag encies (SEFE and SEAPROF) responsible for timber management and smallholder production systems, experimented with payments for environmental services (Gomes et al., 2008; Bartels, 2009; SEMA, 2009), and pursued sustainable forest management initiatives at both the industrial and community scale (Kainer et al ., 2003). Communities in Acre have been at the forefront of integrating sustainable tim ber extraction into their live lihoods (Rockwell et al., 2007) and are distinguished by being some of the first communities in the Brazilian Amazon to attain Forest Stewardship Council (FSC) certification (Humphries and Kainer, 2006). There are clear challenges to forest -based development in Acre. Despite elimination of federal cattle subsidies in 1991, Acres cattle economy has continued to expand with one of the highest growth rates in the Brazilian Amazon (Valentim et al., 2002); as of 2007, the state had approximately 2.7 million head of cattle (IDAF, 2007). An increase in small-scale cattle ranching has been observed in the Chico Mendes Extractive Reserve, even among rubber ta ppers who initially fought against cattle ranchers to maintain access to their forested landholdings (Gomes, 2009; Vadjunec et 38

PAGE 39

al., 2009). The viability of an Extractive Reserve model centered on NTFP-based livelihoods has been heavily critiqued (Browder, 1990; 1992). While timber management could potentially bolster forestbased livelihoods in Extractive Reserves, policy constraints, technical and gover nance complexity, and a polemic fit with conservation goals have delayed this process. Although Acres Forest Government has been in power for more than ten years, the future of forest -based development in Acre is uncertain. Pando, Bolivia In Bolivia, the Forestry Law and the Agra rian Reform Law, passed in 1996, helped recognize forest-dwelling communities in Pando (Table 2-1). Although the Forestry Law focused on timber production by concessionaires and large landholders, it had several important implications for community-bas ed forest management. In particular, establishment of an area-based land ta x discouraged timber companies from maintaining large landholdings, making more forested land available to communities (Contreras and Vargas, 2001). Forest access was democratized by recognition of indigenous subsistence right s and creation of several avenues through which communities could participate in commercial forestry an activity previously prohibited (Ruiz, 2005). In 2000 and 2004, modifications of the Agrarian Reform Law in northern Bolivia gave forest-dwelling communities le gal rights to 500 ha per family, with the ultimate spatial area of a communal title det ermined by the official number of resident families (Ruiz, 2005; Cronkleton and Pacheco, in press). The Bolivian Agrarian Reform Agency (INRA) began implementing t he titling procedure with little consultation from the communities themselves (Cronkleton and Pachec o, in press). While owners of rubber estates in Brazil went bankrupt or were forc ed to sell their lands in the 1970s, owners of 39

PAGE 40

private forested estates in Pando tried to maintain thei r privileged position (Stoian, 2000). Throughout the 1990s and early 2000s there was a struggle between these large landholders and peasant an d indigenous communities to control forest resources and lands (Cronkleton et al., 2007). Complementing this communally-held land in Pando, the 1.8 million hectare Manuripi National Wildlife Reserve was o fficially created in 1973 for biodiversity conservation, although it was not officially managed as such until 1999 (Miserendino et al., 2003; Khne, 2004). In t he Reserve, most land is held in large (1,000-80,000 ha) private estates (Khne, 2004). Although cattle ranching by communities is still rare in Pando and found mostly closer to the urban center of Cobi ja, both large-scale cattle ranching and illegal logging o ccur within these estates (K hne, 2004). There are nine communities in the Reserve, eight along ma in roads and the ninth located along the Madre de Dios River. Similar to communiti es outside Manuripi Reserve, families within the Reserve live in clustered settlements and only during the Brazil nut harvest season do they relocate to their oft en remote forest landholdings. While communities within the Reserve have received more government aid and forestry extension than other communities in Pando, their use of forest resources has been more heavily regulated by Reserve rules that prohibit timber harvest, limit hunting, and manage forest clearing. Methods This paper combines two analytical appr oaches to examine land use land cover change (LULCC) in Brazil nut-producing communities in close proximity to roads in the MAP region. The first approach is based on remote sensing methods using Landsat imagery data and spatial logist ic regression to assess the patterns and trends of observed deforestation in 11 communities (3 in Madre de Dios, 4 in Acre, and 4 in 40

PAGE 41

Pando), representing all three countries. The second approach is based on regression analysis of household-level data collected in the same communities to assess the potential drivers of forest clearing reported by nut producers. The LULCC analysis allows for a comparative c ontextual understanding of land cover change, which is complemented by a finer-scale analysis at the household level to illum inate the land use choices of individual Brazil nut producers that affect forest cover. Sample Communities The study sample included four communi ties (63 households) in Pando, four communities (59 households) in Acre, and three communities (28 households) in Madre de Dios (Figure 2-2). We selected these communities based on their close proximity to major roads, dominance of Brazil nut producer s, and >90% cloud-free imagery available for the locations over the entire time se ries. The communities sampled in Pando comprise an area of 812 km2. While two of the four communities are in Manuripi National Wildlife Reserve and the other two are outside, all have access to one of the three main roads in Pando. In Acre, the four sampled comm unities comprise an area of 686 km2. All are located in t he Chico Mendes Extractive Reserve where Brazil nut production is part of community livelihood systems (Wallace, 2004; Ehringhaus, 2006; Vadjunec et al., 2009) and are accessed by f eeder roads of BR-317. In Madre de Dios, the three communities sampled comprise an area of 543 km2. These communities form part of the group of Brazil nut-producing communities found along the Interoceanic Highway, which is known as the Castaa Corri dor for conservation (ACA, 2008). Remote Sensing Analysis To examine land use land cover change in these communities, we acquired a set of land cover classification and change trajec tory maps for the MAP region from 1986 41

PAGE 42

2005, which were produced through the Univ ersity of Florida NSF-HSD project, Infrastructure change, human agency, and resi lience in social-ecological systems, following protocols described in Marsik et al. (in review). Each image date consisted of eight Landsat images that had been mosaick ed together. To hone in on the 11 study communities, we clipped community polygons out of each classification and trajectory image using digital cadastral la yers from the Agrarian Refo rm Agency in Pando, Bolivia and the Ministry of the Envi ronment in Madre de Dios, Peru. In the Chico Mendes Extractive Reserve in Acre, Brazil, communi ties are defined as associations, and there are no official geographic data available that define associ ation boundaries. Thus, for the four study communities in Acre, we cr eated polygons using a derived buffer distance from the center of the community, based on the fa rthest household known to be affiliated with the association. A total of 25 images from the original classification mosaics were required to encompass the 11 co mmunities studied in the three countries (Appendix I). For the community subsets, we re-perfo rmed the 1996 classification due to visible inaccuracies in the mosaic classification at the local scale, since 1996 was a particularly wet year. Of the more than 200 training sample points gener ated from fieldwork in the dry season of 2006 (June through October), appr oximately 120 that could be calibrated to 1996 based on detailed field data on class age, were split based on a stratified random sample for respective use in t he development of classification rules and accuracy assessment in the community subsets. Following the methods of Marsik et al. (in review), classification of forest and non-forest for each image year was performed using ENVI remote sensing software combin ing half of the training sample data with a 42

PAGE 43

stacked image (bands 457 from the masked im age, local Morans I spatial statistic applied to bands 457 at a spatial lag of one pixel [30 m], tasseled cap indices and a mid-infrared index [B5-B7/B5+B7 ]) for each year. The forest class included all mature forest and secondary regrowth older than 3-4 years of age. This inclusion of early secondary regrowth in the forest class was based on high leaf reflectance of this land cover, since secondary forest grows quickly in the region, especially in the first three years of succession (Uhl et al., 1987; Broadbent, unpublished data). Non-forest included agricultural, cleared, and residential areas. There was an acceptable level of accuracy in the 1996 classification with 86. 61% overall accuracy and a Kappa statistic of 75.3%. We recreated the 1991 and 1996 c hange trajectories for the community subsets, along with the cumula tive change trajectory, based on the new 1996 classification. Trajectory classes were defined as stabl e forest (F-F), stable nonforest (NF-NF), deforested (F-NF), and re forested (NF-F). A cumulative change trajectory from 1986 to 2005 for the communi ty subsets was created to assess overall change. We also used the classification images to produce landscape metrics to quantify the patterns of recent deforestation in the community areas. After applying a filter to include only non-forest patches of an area equal to or gr eater than the normal area cleared by producers for agriculture (~1ha), we used Fragstats 3.3 (McGarigal et al., 2002) to generate non-forest patch metrics for 2000 and 2005. These metrics included patch area, circularity, contiguity and Euclidean distance to the nearest non-forest patch neighbor. Patch area and Eu clidean distance to nearest neighbor are commonly used 43

PAGE 44

in LULCC studies (e.g., Southworth et al., 2002; Nagendra et al., 2004) as basic measures of patch size and patch distribution in the landscape, respectively. Circularity is an index that provides a good measure of patch elongation; this metric has a value of 0 for circular patches and approaches 1 as patches become more elongated (McGarigal et al., 2002). Contiguity provides a meas ure of patch shape based on connectedness of pixels within that patch; the contigui ty index equals 0 for a one-pixel patch and approaches 1 as patch connectedness increa ses (McGarigal et al., 2002). We generated descriptive statistics for these pat ch metrics in non-forest patches in 2000 and 2005 and evaluated differences ac ross dates and between countries Survey Data To understand the relationship between live lihood systems and forest clearing, two annual village and househol d-level questionnaires were appli ed in all 11 communities in Pando, Acre, and Madre de Dios from June 2006 through October 2007 in collaboration with the Center for International Forestry Research Poverty and Environment Network (CIFOR PEN; http://www.cifor.cgiar.org/pen/_ref/home/index.htm ). These questionnaires enabled us to measure village access to markets and land tenure, along with household characteristics and assets. In the second annual household survey, we also asked households to report the amount (ha) of forest (m ature and secondary regrowth > three years old) cleared wit hin the previous five years. Additionally, over a 12-month period, four quarterly household questionnaires were conducted with 63 households in Pando and 59 hous eholds in Acre, and three trimestral surveys conducted with 28 households in Madr e de Dios, which allowed us to quantify annual income associated with different land us es. Quarterly/trimestral interviews focused on all cash and subsistence income derived from forests and other onand off44

PAGE 45

farm activities, including agric ulture, livestock, wage labor, as well as other sources of external economic support (See Table 2-2 fo r a more detailed explanation of income sources). Prices for subsistence products which were never bought or sold, were estimated in community meetings where a willingness-to-pay price was determined. Based on all these data, we generated descriptive statistics for household characteristics, income and asset variables along with amount of forest cleared, and evaluated between-country differences. Modeling Through use of the different datasets, two modeling approaches were used to explain deforestation in Brazil nut producing communities in the MAP region. Based on findings that most regional deforestation occurred recently (2000), particularly along roads (Southworth et al., in prep; Appendi x J), we chose to focus on the causes of deforestation since 2000. To explain deforestati on at the pixel level, we used a binomial logit model in which the response variable was land cover trajectory class (FF and F NF) for the 2000 time per iod. Following Geoghegan et al. (2001) and Chowdhury (2006), the response variable was categorized as y equal to 1 if a forested pixel in the first date was deforested by the second date of a given time period and y equal to 0 if a pixel remained forested between the paired dates. Explanatory variables for this model were chosen based on previous tropical land use land cover change research that showed higher rates of deforestation closer to roads and in already fragmented landscapes (Kaimowitz and Angelsen, 1998; Chomitz and Gray, 1996; Mertens and Lamb in, 2000). We used a r oad overlay for the region (Marsik et al., in review) to identify primar y roads and digitized tw o additional roads in Pando that were not included in the initia l overlay. Following methods outlined in 45

PAGE 46

Mertens and Lambin (2000), we generated pixe l-level variables in ArcGIS, including distance to road, distance to forest/non-fo rest edge, and a Matheron index, which is a measure of forest fragmentat ion. In our case, Mather on index values were log transformed to express a more normal distribution of residuals Following Geoghegan et al. (2001) and Chowdhury (2006), a pseudo R2 was reported, since a traditional R2 measure of fit would not be not easily ca lculated in a categorical regression. Socioeconomic variables were not included in this model, because individual household parcels were not spatially defined in t he study communities and socioeconomic data aggregated at the communi ty level provided very little information in the pixel-level analysis. To explore the socioeconomic drivers of deforestation at the household level, we used a multivariate regression model wher e the parameters were estimated using ordinary least squares (OLS) regression. In this model, the response variable was total amount of forest reported to have been clear ed by individual nut producers from 2002 2007. The response variable of forest clear ed (ha) was log transformed to express a more normal distribution of residuals. Po tential explanatory variables were chosen based on findings in the literat ure that show the influence of household characteristics, income sources, and assets on defor estation (Kaimowitz and Angelsen, 1998; Chowdhury, 2006; Perz et al., 2006; Caldas et al., 2007; Godoy et al., 2009; Wyman and Stein, in press). Hous ehold characteristics include d size, life cycle location (number of adults, number of children, number of elders, age of household head, and length of residency in the community), acce ssibility (distance to village center and distance to nearest city), and years of educati on. Income and asset variables included 46

PAGE 47

annual cash and subsistence income derived from different land uses in 2006-2007, along with material and livestock assets. We first searched for univariate relationships between potential explanatory variables and t he log of the forest reported cleared, considering a country factor together with its interaction. This process guided our selection of more relevant variables (p 0.10) to be included in the multivariate OLS regression to explain reported forest cl earing from 2002 with country as a fixed effect. Results Land Cover Change Forests dominated community-held lands in Madre de Dios, Acre, and Pando in 1986, 1991, 1996, 2000, and 2005 (Table 2-3). Of the total 543 km2 area of sampled Brazil nut-producing communities in Madre de Dios, forests covered 98.0% of the area in 1986 and declined slightly to 95.1% by 2005. Of the 686 km2 total sampled area in Acre, 99.3% was covered by forests in 1986, 98.2% in 1991, and 97% in 1996. It remained stable in 2000 at 97.1%, but in 2005 declined to 94.4% the lowest percentage of forest cover among all countries in the time series. In Pando, the percentage of forest cover was the highest of all countries in all years. Of the 812 km2 sampled in Pando, 99.6% was forested in 1986 and declined only to 97.7% by 2005. Examination of land cover change trajec tories allowed for a more dynamic understanding of forest cover change from 1986 to 2005 (Figures 2-3) and during the individual time periods (Figure 2-4a,b and T able 2-4). Overall, while forest cover remained fairly stable from 1986 to 2005, there was a tendency toward deforestation in all countries (Figure 2-3). Over the 20year period, 3%, 5%, and 2% of land tended toward deforestation in communities samp led in Madre de Dios, Acre, and Pando, 47

PAGE 48

respectively (Table 2-4). In Madre de Dio s, the area deforested increased slightly between the first (1986) and second ( 1991) time periods and remained at more or less the same level in subsequent time periods (1996). This finding contrasts with Acre and Pando where defores tation also increased between the first and second periods, but decreased in 1996 and increased again in 2000 (Table 2-4, Figure 2-4a). In this most recent period, the area deforested in communities in Acre nearly tripled to 2,360 ha from 844 ha in the previous period (Figure 2-4a). Reforestation was also observed in all th ree countries throughout the time series (Table 2-4, Figures 2-3 and 24b), although the total area t hat underwent reforestation was always less than the area deforested in eac h country and time period. In Madre de Dios, the reforested area steadily increased fr om one time period to the next. This trend contrasts with Acre and Pando, where the area reforested increased steadily prior to 2000, but then decreased in 2000. Patterns of Deforestation According to the results of binary logit m odels for each country, all three pixel-level variables (distance to road, distance to forest/nonforest edge, and Matheron index) explained the presence or absence of defores tation (p<0.001) and showed consistent coefficient signs across the three countries (Table 2-5). While it seemed that these models performed poorly as seen by the relatively low pseudo -R2 values across countries (0.22 0.26), pseudo -R2 values of approximately 0.25 are considered acceptable for this type of analysis (Geoghegan et al., 2001; Chowdhury, 2006). Differences in the spatial pattern of patches were observed across years and between countries, even though patch areas did not differ significantly. In 2000 and 2005, there were relatively fewer non-forest patches in Pando when compared to Acre 48

PAGE 49

and Madre de Dios, especially given the larger area sampled in P ando (Table 2-6). In 2000, non-forest patches in Pando were mo re circular (p=0.017) and less contiguous (p=0.019) than those in Acre and Madre de Dios, and patches in Madre de Dios were much closer together (p=0. 009). From 2000 to 2005, the number of non-forest patches increased in Pando and Acre, whereas in Madre de Dios, the number decreased. In this period, non-forest patches in Madre de Dios became less contiguous (p<0.001) and more isolated (p<0.001), whereas patches in Acre became more contiguous (p<0.001) and closer together (p=0.020). Household-level Variables a nd Reported Forest Clearing Comparative household characteristics, income and assets Despite a similar natural resource base, household characteristics in the communities in the three countries were qui te different. Household size and number of adults were the only variables in this ca tegory that did not differ between countries (Table 2-7). In Madre de Dios, households were more advanced in their life cycles than in Pando and Acre: there were more elder s (p<0.001) and some evidence of fewer children (p=0.06) in the Peruvian househo lds, household heads were older (p=0.003), and they had been residents of the community longer (p<0.001). In terms of accessibility, there was a between-country difference in household distribution (p<0.001) with households in Madre de Dios most clustered in village centers along roads, households in Pando relatively close to village centers, and households in Acre spread throughout the forest. There was also a difference in years of education (p<0.001) between countries with people sampled in Acre the least educated and those sampled in Madre de Dios the most educated. 49

PAGE 50

As expected, forest income in all thr ee countries comprised the greatest share of total income, followed by income from crops (Figure 2-5). In Pando and Madre de Dios, wage income was third, while in Acre, third place was other income, which is largely comprised of government aid. There were also substantial differences among countries in terms of Brazil nut income, business in come, and other income (Table 2-7). While there was no difference in livestock income between countries, livestock assets differed (p=0.044) with Acrean producer s having the greatest value of livestock assets, largely from cattle, producers in P ando the second-most (largely due to a few households in one community with large numbers of cows), and producers in Madre de Dios the least. There was also a between-country difference in amount of land (p=0.025); the largest landholdings were in Acre and the smallest in Madre de Dios. In the end, total cash and subsistence incomes did not differ among the three countries. Explaining reported fo rest clearing Households in all three countries reported clearing relatively little forest from 2002 to 2007 (Table 2-7). Producers in Acre r eported clearing the mo st and producers in Madre de Dios, the least, although none of t hese differences were statistically significant (p=0.676). In t he exploratory analysis of each potentially relevant variable regressed on the amount of fo rest cleared, several variables correlated with reported amount of forest cleared at p 0.10 (Appendix K). The mult ivariate regression model revealed that while only a few variables pr edicted the reported amount forest cleared, this model performed well with a R2 of 0.585 (Table 2-8) bas ed on similar goodness-offit values observed in other studies (e.g., Wa lker et al., 2000; Godoy et al., 2009). In Madre de Dios, income deriv ed from Brazil nuts (p=0.002) was negatively correlated with forest clearing. More income derived from crops (p=0. 008) and larger household 50

PAGE 51

sizes (p=0.051) were positive predictors of fo rest clearing in Peru. In Acre, the most important predictors of forest cleared were larger household size (p=0.024) and more income from other sources (p=0.039), follo wed by less crop income (p=0.048). There was some evidence that households with more livestock assets (p=0.062) and younger household heads (p=0.087), and those that were located closer to cities (p=0.092) cleared more forest in Acre In Pando, the strongest pos itive predictors of forest clearing were income from cr ops (p<0.001) and income from livestock (p=0.004). More income derived from Brazil nuts also explained less forest cleared in Pando (p=0.007). Discussion This study provides insights into the patterns and trends of land cover change in Brazil nut-producing communities in the MAP region, and explores spatial and socioeconomic explanations for recent defor estation. As we demonstrated through a land use land cover change analysis, ther e was minimal deforestation in studied communities from 1986 to 2005. That said, a deforestation trend occurred in all three countries, with a higher probability of defor estation in already fragmented areas along roads. In this tropical forest frontier w here road infrastructure is being developed rapidly, it is worth exploring the patterns and drivers of deforestati on, however minimal, since many communities are highly dependent on Brazil nut-rich forests for their livelihoods while conservationists emphasiz e the biodiversity and carbon benefits of these forests. Forest Cover Patterns and Trends In general, our LULCC results in Brazil nut communities from 1986 to 2005 reflect those of the region-wide study for the same 20-year period. Southworth et al. (in preparation) also observed mostly stabl e forest cover from 1986 with most 51

PAGE 52

deforestation along roads, as well as higher defor estation in Acre relative to Madre de Dios and Pando. There is one impor tant difference in our study for the 20-year period in Acre: we observed a much higher perc entage of stable forest from 1986 in extractive communities in Acre (95%) when compared to the entire regional landscape where approximately only 82% of forests remained (Southworth et al., in preparation). This finding complements other studies, which highlight the role of the Chico Mendes Extractive Reserve as a buffer for deforestation and fire (Brown, 2004), and substantiates the claim that reserve resi dents have generally not exceeded their legal 10% limit on deforestation (Vadjunec et al., 2009). More interesting patterns emerge through an examination of separ ate time periods within each country and comparison between c ountries. In comparing our findings from individual time periods with other land use land cover change studies in Madre de Dios, we noted several important differences. Fo r instance, our finding that deforestation increased between the first (1986) an d second (1991) time periods contrasts with several other studies in M adre de Dios (Alvarez and Naughton-Treves, 2003; Naughton-Treves, 2004; Chavez, 2009), in which greater deforestation along the Interoceanic highway was observed during 1986, due to agricultural credits, and slowed in 1991 when Peruvian structur al adjustment policies were implemented (Naughton-Treves, 2004). This difference coul d be due to the fact that Alvarez and Naughton-Treves study area was in the province of Tambopata, which is closer to the city of Puerto Maldonado wher e farmers had access to greater agricultural credits, and the Chavez study did not focus on Brazil nut producers, but rather was focused in an area north of Iberia, which was a nexus of the PEMDs agricultural extension work. 52

PAGE 53

Also, our observation that deforestation in Peruvian communities did not drastically increase during 2000 contrasts with the r egion-wide LULCC study (Southworth et al., in preparation) and could be explained by several factors. Firs t, landholdings of sampled communities are mostly compris ed of Brazil nut concessions, which are relatively large areas reserved for forest pr oduct extraction, when compared to the small agriculturally-zoned landholdings also found al ong the Interoceanic Highway. Second, the five-year trajectory likely masked annual changes in forest cover in concession areas. For instance, preceding the implem entation of the Forestry Law of 2002 in Madre de Dios, there was much insecurity surrounding fate of fore st access. This insecurity, combined with leve ling of the Interoceanic Highw ay in 2001, promoted land invasions and deforestation (Chavez, 2009). After implementat ion of the law, deforestation slowed within timber and Brazil nut concessions from 2002 to 2004 with regrowth occurring in previously deforested areas (Chavez, 2009; Oliveira et al., 2007). An annual land cover change analysis would be needed to tease out these changes, especially in our community sample where Brazil nut concessions are abundant. Our finding that the number of non-forest patches in Madre de Dios decreased from 2000 to 2005 likely corresponds to abandonment of many settlements. The increased distance between the fewer non-forest patches in 2005, however, likely indicates a few more recent clearings farther in the concessions. In Acre, the drop in deforestation dur ing 1996 and subsequent increase in 2000 were also observed in other l and use land cover change studies in the Chico Mendes Extractive Reserve (Ludewigs, 2006; Vadjunec et al., 2009). While the drop could reflect a lag from the creation of the Reserve and an end of land conflicts 53

PAGE 54

between rubber tappers and rancher s, the subsequent increa se in deforestation can certainly be attributed to the adoption of cattle by rubber tappers (Gomes, 2001; Gomes, 2009; Vadjunec et al., 2009). Fore st conversion to pasture was seen in the increased number of non-forest patches from 2000 to 2005, decreased distance between them, and greater contiguity of these patches. The decrease in reforestation in Acre in the final time period was a sign that rubber tappers in the Chico Mendes Extractive Reserve are less likely to allow successional re-growth of fallows as they increasingly convert early secondary forest to pasture despite the efforts of Acres Forest Government to add value to standing forest (Gomes, 2009; Vadjunec et al., 2009). In Pando, while deforestation was minimal over the 20-year period, the observed increase in deforestation between the fi rst (1986) and second (1991) time periods was likely due to leve ling of the road from Cobija to El Sena in 1992. This infrastructure project has been associated wi th increased migration to community areas and clearing along both primary and secondary roads (J. Rojas, pers. comm.). The much later increase in deforestation (2000) was likely due to the completion of the Bioceanica road and creation of pastures in the one community in the study near the city of Cobija where multiple hous eholds had invested in cattle. Reported Forest Clearing The household-level economic drivers of fo rest clearing in the studied Brazil nutproducing communities reflect ce rtain patterns observed in other tropical forest frontiers, in general, and in the Amazon in particular. Agricultural expansion which includes permanent cropping, cattle ranching, shifti ng cultivation, and colonization has been considered the most important cause of tropical deforesta tion (Geist and Lambin, 2002). 54

PAGE 55

In Pando, crop income was the main positive predictor of reported forest cleared from 2002 to 2007. Among smallholders in the Amazon, conversion of forest for planting of subsistence annual crops is a common non-forest land use, especially early on in a households life cycle (Perz et al., 2006). Crop income was also the main positive predictor of forest clearing in Madre de Dios where households were older and had resided for longer on their land holdings. In Madre de Dio s, crops included perennials (e.g. fruit trees), which was expected am ong households that were more advanced in their life cycles (Perz et al., 2006). In Acre, the negative correlation between crop income and forest cleared was possibly a chance finding due to the fact that forest cleared for pasture in that context is much greater than forest converted for swidden agriculture. Income from livestock was an important pr edictor of forest clearing in Pando, whereas in Acre, there was some evidence that livestock assets and not income from livestock predicted forest clearing. As househol ds build capital, purchase of cattle has been considered an important way to reduce risk through diversification (Faminow, 1998; Mertens et al., 2002; Perz et al ., 2006), and accumulation of cattle has been positively correlated with forest clearing in Western Amazonia (Vosti et al., 2003; Caviglia-Harris and Sills, 2005). The finding in Pando makes sense as higher amounts of deforestation were reported by households that generated substantial income from the sale of cattle. In Acre, livestock income was only minimally important, likely because cattle are not consumed or sold regularly, but rather used as a form of savings among residents of the Chico Mendes Extractive Reserve; a producer may sell a calf or 55

PAGE 56

two when members of the household are sick, and livestock assets provide ongoing sources of milk, cheese, and eggs, which diversify rural diets (Wallace, 2004). In Acre, government aid (other income) was an important economic predictor of forest clearing. Interestingly, in current discussions in Brazil on Reduced Emissions from Deforestation and Degradation (RED D), Brazilian governm ent aid programs, specifically family welfare ( bolsa da familia ), are used as models for planned environmental service payments to small holders (SEMA, 2009). If indeed such government aid is currently being used by extractive households to accumulate livestock and clear more forest, similar pay ment structures under REDD could have a comparable effect, depending on how they are stru ctured and monitored. Interestingly, while compensation for avoided deforestation has been considered an economicallyviable alternative to cattle raising in Panam a (Coomes et al., 2009), the results of a linear program model in Acre showed that additional income from payments for environmental services could lead to increas ed deforestation by small producers, unless accompanied by agroforestry assistance (DiGiano, 2006). In terms of forest conservation, the fi nding that greater income from Brazil nuts was correlated with less forest cleared in Madre de Dios and Pando quantifies the potential of this product to promote both forest conservation and livelihood development. That said, this relationship may only hold true in areas where Brazil nut comprises a large share of the total income. In Acre where the proportion of income from Brazil nuts was much lo wer and livelihood systems were more diverse, there was little association between nut income and fore st cleared. In scenario modeling in a colonist site in the Western Brazilian Amazon, adding value to forests through Brazil nut 56

PAGE 57

income was not considered a viable way to dissuade producers from converting Brazil nut-rich forests to pasture, since higher financial returns associated with non-forest land uses dominated those fr om traditional extracti ve forestry (Vosti et al., 2003). Also, notably, forest income aside from that earned from Brazil nuts (e .g. fuelwood, game) was not correlated with forest cleari ng in any of the three countries. Many studies have highlighted relationshi ps between household characteristics and forest clearing (Kaimowit z and Angelsen, 1998;, Chowdh ury, 2006; Perz et al., 2006). In our study, in both Madre de Dios and Acre, household size was positively correlated with reported forest clearing. This finding reflects work in the Eastern Amazon (Perz et al., 2006) where larger hous eholds were found to deforest more based on subsistence needs and greater labor avail ability. That said, more specific household-level demographic life cycle variables were not correlated with reported forest cleared. For instance Eastern Amazonian households with more children and elderly members allocated more land to cu ltivation of annual crops based on increased subsistence needs (Perz et al., 2006), whic h was not observed in our study. While higher education levels have been found to decrease deforestation through greater involvement in off-farm wage activities (Irwin and Geoghegan, 2001; Chowdhury, 2006; Godoy et al., 2009), in our context where wa ge earning opportunities were minimal, education did not explain forest clearing. That said, producers in Madre de Dios who were more educated than their counterparts in Acre and Pando did glean more income from entrepreneurial business activities (T able 2-8). In the exploratory analysis, business income was inversely correlated with forest clearing (p=0.025; Appendix K), possibly following similar logic to wage inco me above, but did not remain a strong 57

PAGE 58

explanatory variable in the multivariate model and was dropped from the final analysis. Finally, household accessibility, in terms of dist ance to village center or distance to city did not explain forest clearing in the multiv ariate model. This fi nding may be due to the fact that these communities have minimal connections to formal markets; they are primarily engaged in subsistence activities with t he sale of forest or agricultural products generally performed thr ough intermediaries. Differences in Observed vs. Reported Deforestation An important caveat to this study is t hat we used two separate models based on different datasets to understand recent obser ved (via satellite imagery) and reported (via household surveys) deforestation in the sampled communities. There are advantages and limitations associated with combining these approaches. The main advantage is that the remote s ensing analysis provided an im portant spatial context for understanding and comparing t he patterns and trends of def orestation between the three countries, while the household-level analysis allowed us to go deeper in our exploration of the socioeconomic drivers of deforestation. Although we believe these approaches complement each other, ther e was some discrepancy associated with observed versus reported deforestation data at the household level. For instance, for Pando, while mean reported fo rest cleared was 6.5 ha from 2002 to 2007, observed deforestation from the 2000 trajectory image was 8.34 ha per household. In Acre, this discrepancy was larger with a difference of 6.6 ha of reported fo rest cleared versus 24.84 in the country samples. Through a closer examination at the community level in Acre, we saw that most of this differenc e was in the southern portion of one community, where some extractive households were inte rspersed with colonist farmers; when the community boundary was estimated, inevitably some of these colonist lands were 58

PAGE 59

included in the community area. In Madre de Dios, while the mean reported forest clearing was 5.27 ha, observed forest clearing was actually lower at 2.68 ha per household. Notably, in the Peruvian context, the number of households attributed to each community was an approximation, since community associations are not as well defined there as they are in Acre and Pando, which could be the reason for this lower observed deforestation. Discrepancies between observed and reported deforestation could also be due to the fact that there was a two-year difference in the examined time periods. Future planned research that combines the 2002 survey data with an annual land cover change analysis from the sa me years (using newly available Landsat imagery) will allow for better comparison of the satellite and survey data to understand the rates, patterns, and causes of deforesta tion in extractive communities in MAP. Conclusion This study illuminates the relationship between forest conservation and forest dependence in extractive communities in West ern Amazonia. Overall, we found that extractive communities in the region were extremely forest income dependent and had deforested only a minimal portion of their landholdings from 1986 to 2005. While this general land cover change pattern mirrored t he findings of a region-wide study for the same 20-year period (Southworth et al., in preparation), there was much less deforestation in the extractive communi ties in Acre when compared with the surrounding landscape. There was also less deforestation in the final time period (2000) in communities in Madre de Dios when compared with the region-wide study. Although these communities may cu rrently be located beyond the agricultural frontier, given current and pl anned infrastructural exp ansion throughout the Amazon 59

PAGE 60

(Fearnside, 2002; Nepstad et al ., 2002; Soares-Filho et al., 2006; Perz et al., 2008), deforestation in and around these communities will likely increase in the future. A regional approach to forest conserva tion hinges on promoting the livelihood development of forest-dwel ling communities (Kaimowitz and Sheil, 2007; Sunderlin et al. 2008b; Chhatre and Agrawal, 2009). Alt hough we observed that agricultural income was positively correlated with forest clear ing in Pando and Madre de Dios, and some evidence that increased livestock assets expl ained forest clearing in Acre, these land uses were performed on a small scale, generally for subsistence purposes. That said, the trend of not allowing swidden fallows to regenerate and instead converting them to pasture for steady accumulati on of cattle, as seen in the Acre, could have negative conservation implications in t he future. Such implications are especially relevant for the future of Brazil nut, since fallows have been observed to be favorable regeneration sites for this long-lived species (Cotta et al ., 2008). The inverse relationship observed between Brazil nut income and forest cleari ng in Madre de Dios and Pando reinforces the need to continue bolstering the Brazil nut sector to promote fore st-based livelihoods and the conservation of Brazil nut-rich forests in Western Amazonia. 60

PAGE 61

Figure 2-1. Map of Madre de Dios, Peru, Acre, Brazil and Pando, Bolivia (MAP region) with most recent portion of Interoceani c Highway and 100 m buffer in red. 61

PAGE 62

Credit: N. Hoyos Figure 2-2. Map of communities sampled in Madre de Dios, Ac re and Pando. 62

PAGE 63

63 Figure 2-3. Land cover change trajectories 1986-2005 in select communities.

PAGE 64

0 500 1000 1500 2000 2500 3000 3500 4000 HectaresDeforestation Madre de Dios Acre Pando Communities Figure 2-4a. Area deforested in paired-date time periods in Brazil nut producing communities in MAP. 0 100 200 300 400 500 600 700 800 900 1000 HectaresReforestation Madre de Dios Acre Pando Communities Figure 2-4b. Area reforest ed in paired-date time periods in Brazil nut producing communities in MAP. Note different yaxis scale in comparison with Figure 24b, reflecting more deforestation when compared with reforestation. 64

PAGE 65

65 0.0010.0020.0030.0040.0050.00Aquaculture Other Business Livestock Wage Crop Forest Income Share (%) Madre de Dios Acre Pando Figure 2-5. Income share distribution among sampled households in the three countries.

PAGE 66

Table 2-1. Policy events relevant to each land use land cove r change time series in Madre de Dios, Peru, Acre, Brazil, and Pando, Bolivia over t he 1986-2005 study period. Landsat image time series Madre de Dios, Peru Acre, Brazil Pando, Bolivia 1986 Agricultural credits under Agrarian Reform (1985-1990) Federal cattle subsidies (1980s-1991) Creation of Alto Jurua (1989) and Chico Mendes Extractive Reserves (1990) Opening of Cobija-Riberalta road through Pando (1980) 1991 Structural adjustment (1990-2000) Interoceanic highway leveled, maintained (dry season only) (1990s) Completion of paved BR-364 from Porto Velho to Rio Branco (1992) Paving of BR-317 (1993-2000) Leveling of Cobija-Riberalta road through Pando (1992) 1996 Structural adjustment (1990-2000) Interoceanic highway leveled, maintained (dry season only) (1990s) Paving of BR-317 (1993-2000) Acre Forest Government (1998Present) Chico Mendes Law (1999) Agrarian Reform Law (1996) Forestry Law (1996) Opening of Cobija-El Chive road through Manuripi Reserve (1998) Re-designation of Manuripi National Reserve (1999) 2000 Forestry Law (2000; implementation of concessions in 2002) Formalization of Bahuaja Sonene National Park and Tambopata Reserve (2000) Interoceanic highway compacted and prepared for paving (20032006) Acre Forest Government (1998Present) Ecological-Economic Zoning (2002) Proambiente (2004) Construction of Bioceanica Road (2001) Community land titling by INRA (2004-2009) 66

PAGE 67

Table 2-2. Description of cash and subsistence income categories. Inputs were discounted from net income calculations. Income category Description Forest All raw and processed products collected in forests, including wild plants, fruits, seeds, game (mammals, fish and insects), fuelwood and timber. Crop Grains, fruits and vegetables cultivated in swidden-agriculture plots and home gardens. Livestock All livestock (chickens, pigs, sheep, cattle) slaughtered or sold. Livestock assets, which are domestic animals owned, are not included in this category. Aquaculture Fish raised in ponds. Wage Payments for both onand off-farm labor. Business Earnings from onand off-farm businesses, including transport services for forest products. Other Remittances, government payments (pensions, en vironmental service) and non-governmental donations. 67

PAGE 68

68 Table 2-3. Regional land cover in select Brazil nut producing communities 1986-2005. Land cover area (km2) Percent of coverage in parentheses 1986 1991 1996 2000 2005 Madre de Dios, Peru Forest 532.42 (98.0) 527.92 (97.2) 521.37 (96.0) 517.02 (95.2) 516.68 (95.1) Non-forest 1.26 (0.2) 5.75 (1.1) 12.30 (2.3) 16.65 (3.1) 17.00 (3.1) No data 9.44 (1.7) 9.44 (1.7) 9.44 (1.7) 9.44 (1.7) 9.44 (1.7) Acre, Brazil Forest 681.12 (99.3) 673.55 (98.2) 665.26 (97.0) 666.14 (97.1) 647.35 (94.4) Non-forest 3.20 (4.7) 10.78 (1.6) 19.07 (2.8) 18.18 (2.7) 36.97 (5.4) No data 1.30 (0.2) 1.30 (0.2) 1.30 (0.2) 1.30 (0.2) 1.30 (0.2) Pando, Bolivia Forest 808.25 (99.6) 806.60 (99.4) 800.80 (98.7) 799.54 (98.5) 792.86 (97.7) Non-forest 0.75 (0.09) 2.40 (0.3) 8.20 (1.0) 9.46 (1.2) 16.14 (2.0) No data 2.72 (0.3) 2.72 (0.3) 2.72 (0.3) 2.72 (0.3) 2.72 (0.3)

PAGE 69

Table 2-4. Regional change trajectories in select Brazil nut producing communities. Change trajectory area (km2) Percent of cover class in date 1 that transitioned (or not) in date 2 in parentheses 1986 1991 1996 2000 1986-2005 Madre de Dios, Peru Stable forest (F-F) 527466 (99.1) 519758 (98.5) 512132 (98.2) 508839 (98.4) 516248 (97.0) Deforestation (F-NF) 4951 (0.9) 8162 (1.5) 9238 (1.8) 8180 (1.6) 16169 (3.0) Stable Non-forest (NF-NF) 801 (63.8) 4141 (72.0) 7416 (60.3) 8814 (52.9) 824 (65.7) Reforestation (NF-F) 455 (36.2) 1611 (28.0) 4887 (39.7) 7840 (47.1) 431 (34.3) No data 9437 (n/a) 9437 (n/a) 9437 (n/a) 9437 (n/a) 9437 (n/a) Acre, Brazil Stable forest (F-F) 672206 (98.7) 660700 (98.1) 656817 (98.7) 642548 (96.5) 646346 (95.0) Deforestation (F-NF) 8917 (1.3) 12852 (1.9) 8438 (1.3) 23595 (3.5) 34777 (5.0) Stable Non-forest (NF-NF) 1853 (57.9) 6215 (57.7) 9742 (51.1) 13377 (73.6) 2195 (68.6) Reforestation (NF-F) 1347 (42.1) 4555 (42.3) 9326 (48.9) 4803 (26.4) 1005 (31.4) No data 1302 (n/a) 1302 (n/a) 1302 (n/a) 1302 (n/a) 1302 (n/a) Pando, Bolivia Stable forest (F-F) 806050 (99.7) 799618 (99.1) 794771 (99.2) 789535 (98.7) 792347 (98.0) Deforestation (F-NF) 2200 (0.3) 6987 (0.9) 6029 (0.8) 10009 (1.3) 15903(2.0) Stable Non-forest (NF-NF) 196 (26.1) 1213 (50.6) 3427 (41.8) 6134 (64.9) 240 (32.0) Reforestation (NF-F) 554 (73.9) 1183 (49.4) 4773 (58.2) 3322 (35.1) 509 (68.0) No data 2722 (n/a) 2722 (n/a) 2722 (n/a) 2722 (n/a) 2722 (n/a) 69

PAGE 70

Table 2-5. Binomial logit models of defor estation in Brazil nut communities during period 2000-2005. Unit of observation: the pixel (30 x 30 m) (n=888,323 in Pando, n=740,224 in Acre, and n=573,233 in Madre de Dios). All variables significant at p<0.001. Madre de Dios, Peru Acre, Brazil Pando, Bolivia Variable Coeff. Z-stat Coeff. Z-stat Coeff. Z-stat Pixel level Dist. to road -0.00006588 -17.64 -0.00003133 -26.83 -0.00006901 -13.43 Dist. to F/NF edge -0.0002818 -13.71 -0.0016679 -67.44 -0.0006416 -42.94 Matheron index (log) 0.9851 97.70 0.63409 96.32 0.78888 89.63 Pseudo R2 0.26 0.24 0.22 70

PAGE 71

Table 2-6. Non-forest patch metrics in 2000 and 2005: NP (number of patches); AREA (size); CIRCLE (circularity index); CONTIG (contiguity index); ENN (Euc lidean distance to nearest neighbor). Wi thin country significance levels between years are provided in the table (detailed p=values for significant variables in text). Significant differences between countries in each year ( 2000 and 2005) are described in the text only. Location Year NP AREA (ha) CIRCLE (0-1) CONTIG (0-1) ENN (m) Madre de Dios 2000 367 4.11 (0.62) 0.52 (0.01) 0.53 (0.01) 213.73 (18.29) 2005 343 4.00 (1.11) 0.50 (0.01) 0.46 (0.01)*** 356.06 (36.51)*** Acre 2000 392 4.19 (1.17) 0.51 (0.01) 0.51 (0.01) 307.04 (21.28) 2005 494 7.13 (1.48) 0.52 (0.01) 0.57 (0.01)*** 248.17 (14.86)* Pando 2000 216 3.81 (0.93) 0.54 (0.01) 0.47 (0.02) 260.35 (34.33) 2005 251 4.72 (1.14) 0.53 (0.01) 0.46 (0.02) 206.09 (17.75) Notes : *** (p 0.001); ** (p 0.01); *(p 0.05); +(p 0.10). 71

PAGE 72

Table 2-7. Descriptive statistics of househol d characteristics, income, assets and repo rted forest clearing in Brazil nut-rich communities in Madre de Dios, Acre, and Pando. All Madre de Dios, Peru Acre, Brazil Pando, Bolivia Variable Mean (SE) Mean (SE) Mean (SE) Mean (SE) p-value Outcome Cleared forest (ha) 6.30 (0.57) 5.27 (0.67) 6.60 (0.66) 6.50 (1.17) 0.676 Household characteristics Size 5.37 (0.22) 5.07 (0.439) 5.36 (0.37) 5.49 (0.33) 0.785 Life cycle location Number adults (age 15-65) 3.03 (0.13) 3.14 (0.38) 3.12 (0.21) 2.90 (0.18) 0.686 Number children (age <15) 2.20 (0.15) 1.50 (0.23) 2.21 (0.25) 2.46 (0.23) 0.06 Number elders (age >66) 0.18 (0.36) 0.43 (0.12) 0.03 (0.02) 0.19 (0.58) <0.001 Age of household head (yrs) 43.70 (1.20) 51.57 (2.98) 40.29 (1.66) 43.37 (1.85) 0.003 Length of residency (yrs) 18.87 (1.24) 29.68 (3.13) 16.86 (2.01) 16.21 (1.61) <0.001 Accessibility Distance to village center (min) 39.84 (3.86) 2.86 (1.79) 73.59 (6.81) 27.04 (4.39) <0.001 Distance to city (min) 225.40 (8.81) 117.32 (3.06) 260.83 (17.68) 239.08 (9.47) <0.001 Education (sum yrs) 20.52 (1.28) 34.29 (3.42) 11.72 (1.01) 22.30 (1.89) <0.001 Household income (USD per capita) Forest income 542.07 (38.15) 547.87 (113.95) 488.09 (43.54) 592.34 (64.38) 0.461 Brazil nut income 278.24 (35.54) 290.58 (110.74) 137.68 (26.76) 418.49 (60.62) 0.001 Crop income 256.36 (36.24) 142.92 (42.55) 290.97 (66.11) 274.91 (56.61) 0.303 Livestock income 66.95 (24.56) 36.03 (29.13) 44.26 (33.60) 103.53 (48.27) 0.454 Aquaculture income 3.52 (1.64) 0.00 (0.00) 6.09 (3.96) 2.63 (0.94) 0.361 Wage income 225.34 (32.13) 249.95 (65.29) 181.28 (60.02) 257.20 (42.60) 0.525 Business income 72.34 (24.95) 202.13 (60.65) 13.35 (9.08) 70.15 (51.96) 0.020 Other income 113.00 (22.10) 14.39 (8.64) 238.39 (49.83) 35.60 (7.48) <0.001 Total income 1279.58 (82.24) 1193.3 (163.58) 1262.44 (131.60) 1336.36 (135.81) 0.808 Assets (USD per capita) Total land (ha) 496.22 (32.77) 371.21 (71.72) 602.95 (64.05) 459.36 (39.80) 0.025 Material assets 417.43 (51.24) 243.62 (37.74) 488.12 (60.13) 427.36 (99.26) 0.246 Livestock assets 489.47 (97.66) 115.91 (45.22) 769.81 (170.69) 402.53 (157.96) 0.044 72

PAGE 73

Table 2-8. Results of multivariate r egression model with the most important measured variables that explained log amount of forest cleared from 2002-2007 (R2=0.585). All variables were measured using country as a fixed effect. Households were the unit of observation (n=125). Actual p-values for significant variables included in text. Madre de Dios, Peru Acre, Brazil Pando, Bolivia Variable Coeff. t -Statistic Coeff. t -Statistic Coeff. t -Statistic F-Stat country 1.31 1.29 2.159 5.94*** 1.404 3.70*** <0.001 country*crop income 0.001893 2.69** -0.000416 -2.00* 0.000822 3.55*** <0.001 country*Brazil nut income -0.000699 -3.21** 0.000041 0.10 -0.000685 -2.73** 0.001 country*hh size 0.1124 1.98* 0.0775 2.29* 0.0038 0.10 0.002 country*livestock income 0.00226 0.86 0.000360 0.77 0.00342 2.95** 0.012 country*livestock assets -0.00111 -1.00 0.0001555 1.89+ -0.0000825 -0.87 0.157 country*distance to city -0.00827 -1.09 -0.001177 -1.70+ -0.00098 -0.83 <0.001 country*hh head age 0.01151 1.44 -0.01316 -1.73+ 0.00835 1.40 0.016 country*other income -0.00033 -0.12 0.000558 2.09* 0.00043 0.29 0.296 73

PAGE 74

CHAPTER 3 FORMALIZING PROPERTY RIGHTS OF TROPICAL FOREST COMMUNITIES: IMPLICATIONS FOR NON-TIMBER MANAGEMENT, LIVELIHOODS AND CONSERVATION Introduction Local communities may help determine tropic al forest fates. Communities now own or manage one-quarter of the worlds tropi cal forests due to recent devolution of government-owned lands and formal recognition of customary rights (White and Martin, 2002; Sunderlin et al., 2008a). Such transfe rs of forest owner ship and management rights to local control have the potential to promote both conservation and livelihood development in remote tropical regions (S underlin et al., 2005) where rural people often depend on forest resources for their economic we ll-being (Belcher et al., 2005). Secure property rights are considered an important base for allowing communities to attain desired economic benefits fr om land and resources (de Soto, 2000). Although devolution of property rights to community control may not always promote forest conservation (Gould, 2006; Tacconi, 2007), forest-dependent communities with secure rights may be more likely to place greater emphasis on long-term forest management and conservation (McKean, 2000; Colfer, 2005). The process by which property rights are formalized can affect the security of communities rights to forest resources. A common way to designate public lands for community use is to grant co mmunal land titles, creating a sense of order in contested tropical forests. Much caution is needed, however, during the process of formalizing communities often complex customary right s to land and resources. Communities manage natural resources both as common-p ool and private assets (Ostrom, 2003), with rights governed by local institutions often adapting over time (Gibson et al., 2000). 74

PAGE 75

Such complex property rights can be perceived as webs of interest, made up of overriding, overlapping, complementary, or contested intere sts between multiple actors. Formalizing communal tenure may be akin to cutting the web (Meinzen-Dick and Mwangi, 2008) if customary ru les are overturned, and can result in unsustainable resource use if community members are unsure which rules apply and are insecure about long-term ownership (Fitzpatrick, 2006). Furthermore, granting titles to land often ignores traditional tree tenure systems common in tropical forests (Fortmann and Bruce, 1988), resulting in discrepancies between formal and customary rights. These arguments are not meant to diminish the importance of land titling, communal or otherwise, but rather highli ght the need to involve local people in developing formal tenure rules and regulations that promote effe ctive local management of land and resources (Hayes, 2007). Shifts in property rights, such as formal transfers to community control, combined with rapidly changing values of forest resource s can result in forest-based conflicts (de Jong et al., 2006). While rising resource prices may increase demand for property rights security (Demsetz, 1967), costs of enforcing rights to increasingly valuable land and resources may become prohibitive (Fitz patrick, 2006); legitimate users may be excluded while others gain control (Mein zen-Dick and Mwangi, 2008). During such transitions, contradictory policies may also promote conflict. In the Eastern Brazilian Amazon, violent conflicts between landowners and squatters ensued when incompatible, legal rights were granted to each of these groups on lands that were rapidly increasing in value (Schmink and Wood, 1992; Alston et al., 2000). In northern Bolivia, overlapping land titles were grant ed to communities, timber concessions, and 75

PAGE 76

private individuals, which resulted in conflicts (Ruiz, 2005). Such conflicts have indirectly led to deforestation and resource overexploitation, and adversely affected the livelihood benefits obtai ned from contested resources (Als ton et al., 2000; Ruiz, 2005; de Oliveira, 2008, Yasmi and Shanz, in press). A simple, yet common manifestation of forest-based conflicts is theft of forest resources, especially when demand for fore st products is high and local access and exclusion rights are weakened (Koning et al., 2007) In criminology, theft is defined as the taking of another person s property without their cons ent (Allen, 2005). In the context of property rights transitions in tr opical forests, resource thefts could be the result of malicious intent or simple confus ion over the new property rights systems in place. In either case, t he proprietor, as defined by Schlager and Ostrom (1992), is one who can effectively exclude others from their property, wh ile maintaining withdrawal and management rights to re sources. Given the uncertainties and conflict asso ciated with property rights transitions, participatory mapping is one strategy for em powering local communities to negotiate and strengthen customary tenure systems (Peluso, 1995; Alco rn, 2000; Chase Smith et al., 2003; Chapin et al., 2005). Partici patory mapping is often used by nongovernmental and research organizations to fa cilitate individuals or groups of people in spatially representing real-w orld features through hand draw n sketch maps (Lynam et al., 2007) and/or computerized maps using GIS and remote sensing (Chapin et al., 2005). Documentation of customary rights through mapping may help rural communities: (1) clarify boundaries, which ma y enhance their power of exclusion and decrease conflicts; and, (2) improve pl anning and access to government support 76

PAGE 77

(Cronkleton et al., 2007). Working with rural communities to understand and negotiate their property rights through participatory mapping may also help communities manage their forests better. Through comparative research in forest -dwelling communities in neighboring Pando, Bolivia, and Acre, Brazil, we examined the phenomenon of Br azil nut thefts. Brazil nut ( Bertholletia excelsa ) is the most important nontimber forest product (NTFP) in Western Amazonia, and individual trees are considered key livelihood assets by forest dwellers in both countries. We focused our comparison on these two adjacent areas, because while they have a similar natu ral resource base, their property rights systems and dependence on the nuts by local co mmunities are different. Our overall research question in this chapter is: What ar e the causes and effects of Brazil nut thefts, and how can such thefts be resolved? We first introduce the study region and ec ological and economic characteristics of Brazil nut. We next highlight important historical proc esses and policy changes that have shaped settlement patterns and property rights for communities in Pando and Acre. We then present result s of field studies on reported thefts of Brazil nuts, and explore the causes of such thefts, al ong with their managemen t implications. Specifically, we ask: (1) To what extent do Brazil nut thefts occur in forest-dwelling communities in Pando and Acre? (2) What is the financial value of these thefts? (3) What are the causes of such thefts? (4) Does threat of theft affect Brazil nut harvesting and management practices? (5) How are thefts resolved? and, (6) What are producers perceptions of participatory mapping as one way to resolve such thefts? This 77

PAGE 78

comparative study illustrates broader issues related to the role of property rights security, livelihoods, and management of tropical forests by rural communities. Study Region In the approximately 220,000 km2 border region of Acre, Brazil and Pando, Bolivia in Western Amazonia, many non-indigenous extractive communities whose principal livelihood activity is collection of forest products were granted fo rmal rights to their traditional landholdings through recent devol ution of public and private lands. These communities share the landscape with indi genous groups and more recently settled farmers, cattle ranchers, and loggers. Although the entire r egion is characterized by lowland wet tropical forest vegetation, settlem ent histories, patterns of deforestation, public policy, and socio-economic development va ry considerably from one country to the next ( www.map-amazonia.net ). Construction of th e Interoceanic Highway, an extension of Brazilian BR-317 into Peru and Bolivia, is changing conditions in this formerly remote region, by pr oviding regional access to Pacific ports (Figure 3-1). In Acre, forest conversion has been rapid, exte nsive, and largely driven by establishment of cattle ranches (Souza et al., 2006). In Pando, in contrast, deforestation has been minimal, with most land conversion occurri ng in close proximity to population centers and along the Brazilian border (Marsik et al., in review). Brazil Nut Bridging Forest Conser vation and Livelihood Development Brazil nut is currently the most important non-timber forest product (NTFP) in Western Amazonia. It is collected primar ily from mature forests and its combined ecological and economic characteristics make it a species with the potential to promote forest conservation while contributing to the livelihoods of rural communities. At maturity, Brazil nut trees are giants in upland forests of this region, emerge above the 78

PAGE 79

forest canopy, attain up to 3 m in diamet er, and live for centuries. Because of its massive size and high relative densities, th is species provides important ecological structural and functional roles at the loca l and landscape scale (Zuidema, 2003) Its large, woody fruits fall to the ground during the wet season and retain the approximately 25 seeds (here referred to as nuts) inside. The scatterhoarding agouti ( Dasyprocta spp.), one of the few animals that can gnaw through the hard fruits, plays an instrumental role in seed dispersal and buria l. People access the nuts by gathering the heavy fruits in large piles and using machetes to break them open. The nuts are put in large sacks and carried out of the forest. Although Peres et al. (2003) suggested that decades of commercial harvesting may leave in sufficient juvenile recruitment to ensure future generations, populations in the We stern Amazon appear to be viable over the medium-term under a range of harvest intensit ies (Zuidema and Boot, 2002; Wadt et al., 2008). Harvested nuts can be counted on as a s easonal contribution to local livelihoods as there is little variability in fruit produc tion at the population level across years (Kainer et al., 2007). The nuts fetch relatively high prices on local, national, and international markets, particularly because of increased demand and consequent higher prices since 2003. Western Amazonia is the current cent er of the Brazil nut economy, employing tens of thousands of familie s during the primary collect ion season (JanuaryMarch; Bojanic, 2001). Because of Brazil nuts regional economic importance, Brazilian, Bolivian, and Peruvian legislat ion prohibit felling the trees. History of Forest Extraction The history of colonization and settl ement by non-indigenous extractive populations in Acre and Pando began in the late 19th century during the first boom of natural rubber ( Hevea brasiliensis ), and has since been shaped by distinct changes in 79

PAGE 80

policy and market demands (Table 3-1). Immi gration to rubber estates in the Western Amazon, into Pando and what is now Acre (whi ch at the time was officially Bolivian territory; Barham and Coomes, 1996) explo ded after the 1870s, which is also when Bolivian entrepreneur Nicolas Suarez began his dominance of the Bolivian rubber industry (Fifer, 1970). The border conflict between Brazil and Bolivia (1899-1903), in which the Brazilian rubber tappers in Acre had a major role in Brazils victory, reestablished the boundaries betw een the two countries; the territory of Acre was ceded to Brazil in 1903 (Fifer, 1970). When the pric e of rubber fell in 1912 in response to establishment of rubber plantations in Ma laysia, Brazilian and Bolivian rubber tappers began to diversify their livelihood strategies to include Braz il nuts and agriculture (Fifer, 1970; Barham and Coomes, 1996; Stoian, 2000). In 1931, the Suarez Company introduced Brazil nut shelling, dominated by a female labor force because Bolivian men were engaged in the Chaco War (1932-35) (Fifer, 1970). During World War II, there was a second, smaller rubber boom in whic h the U.S. partnered with Brazil through the Washington Accords, and recruited a second wa ve of Brazilian rubber soldiers to Acre (Sobrinho, 1992). Although Pando and Acre we re legally separated, the border was porous, and more than 2000 Brazilian r ubber tappers were employed by Suarez and Hermanos who at the time controlled 80% of rubber production in the Brazil-Bolivia border region (Sobrinho, 1992). Brazil nuts continued to be a complementary seasonal activity to rubber tapping, although in the 1950s, nut exports surpassed rubber production, and the first official boom of the Brazil nut sector occurred in the 1990s (Stoian, 2000). 80

PAGE 81

Rubber Industry Sets the Stage for Braz il Nut Extraction in Western Amazonia Although the rubber industry did not promote long-term economic development in the Amazon (Weinstein, 1983; Ba rham and Coomes, 1996), it wa s critical in setting the stage for current Brazil nut pr oduction in terms of both har vester settlement patterns and forest product market chai ns. In Acre and Pando, comm unities were formed by families of rubber tappers that continued to li ve in the forests even after the price of rubber dropped. In Acre, these people maintai ned their traditional isolated distributions throughout the forest even as the lands beneath them were either sold off or abandoned by the traditional rubber estate owners to cattle ranchers (Sobrinho 1992). In contrast, families in Pando tended to congregate as rubber lost value. This settlement pattern was reinforced as municipal governments pr ovided infrastructure and services in response to Bolivias 1994 Popular Participation Law. The formation of concentrated communities in Pando meant that forest landholdings were occupied only during the Brazil nut harvest season (Cronkle ton et al., 2007; Figure 3-2). In extractive communities in Acre and Pando, customary rights to land were largely based on tree tenure. In Acre, these continued to be defined by the location of rubber trees and the trails le ading to them, whereas in Pando, proprietorship was increasingly defined by the di stribution of Brazil nut trees (Ankerson and Barnes, 2005; Cronkleton et al., 2007). In both places, Braz il nut market chains from the forest to middle men to processors to in ternational markets paralleled t hose of the rubber trade. During the 1990s, two important property rights policy changes occurred at the national level in Brazil and Bolivia: establishing Extrac tive Reserves in Brazil, and passing a new Forestry Law and Agrarian Reform Law in Bolivia, designed to formally recognize property rights of communities. The origin of these policy changes and how they were 81

PAGE 82

implemented in Acre and Pando have important consequences for extractive activities in the region today. Recent National Policies Designed to Devo lve Rights to Extractivist Communities Extractive Reserves have been cham pioned as a viable and sustainable alternative to widespread deforestation in the Amazon (Allegretti 1989; 1990). An important difference between Extractive Reserves and other Amazonian protected areas is that they were created not despite but because of people (Ehringhaus, 2006). In 1990, the Brazilian government established the first official Extractive Reserves in Acre in response to the rubber tappers social movement to secure land tenure coupled with international pressure to halt Amazoni an deforestation (Schwartzman, 1989; Hall, 1997). Acres rubber tappers used internat ional environmental concerns about the Amazon to their advantage, por traying themselves as forest stewards (Schmink and Wood, 1992). The Extractive Reserve po licy guaranteed usufruct rights to people engaged in traditional livelihoods based largely on collection of NTFPs (e.g. rubber, Brazil nuts, various fruits, and palms; Ehringhaus, 2006), while requiring residents to maintain at least 90% of their landholdings in forest cover (Fear nside, 2003). NTFP extraction is considered less ecologically dest ructive than other land-use practices, such as clearing land for agriculture and pasture or harvesting timber (Arnold and RuzPrez, 2001), thereby embedding conservati on goals within the Extractive Reserve model (Fearnside, 2003). Critiques of this c onservation model focus on the possibility of NTFP over-exploitation; the uncertain economic viability of NTFPs (Browder, 1990; 1992); NTFP intensification in plantations or replacement by synthetic substitutes (Homma, 1992; Dove, 1994). These factors could eventually cause people in reserves to use the land in more ecologically-destr uctive ways, such as for cattle ranching and 82

PAGE 83

commercial agriculture (Gomes, 2001; Salis bury and Schmink, 2007; Gomes, 2009). Importantly, certain NTFPs, such as Braz il nut, have the potential to contribute to livelihood development. Whol esale Brazil nut prices in the United States more than doubled from 2000 to 2005 ($2.04 to $4.38/kg dry weight, respectively; Red River Foods, unpublished data), suggesting that this NTFP could contribute to rural poverty alleviation, but only if producers secure an equitable share of ma rket value. In 1990, the one million hectare Chico Me ndes Extractive Reserve (CMER) was created in Acre from 42 former rubber estate s. Households are spread throughout the mostly forested landscape, and the unique tree tenure legacy from the rubber era is still honored among extractivists. Even though Brazil nut has replaced rubber as the most important forest product (Wallace, 2004; Ehringhaus, 2006), reserve residents still define individual property holdings by the nu mber and distribution of rubber trails through the forest (Ankerson and Barnes, 2005). In Bolivia, the Forestry Law and the Agra rian Reform Law, both passed in 1996, affected forest-dwelling communities in Pando. Although the Forestry Law was focused on timber production, it had se veral important implications for community management of NTFPs. In particular, establishment of an area-based land tax discouraged timber companies from maintaining la rge landholdings, making more forested land available to communities (Contreras and Vargas, 2001). Forest access was democratized by recognition of indigenous subsistence ri ghts and creation of several avenues through which communities could participate in comme rcial forestry an activity previously prohibited (Ruiz, 2005). In 2000 and 2004, modi fications of the Agrarian Reform Law in northern Bolivia gave forest-dwelling communities legal rights to 500 ha per family, with 83

PAGE 84

the total area of the communal title determined by the official number of resident families (Ruiz, 2005; Cronkleton and Pacheco, in press). The Bolivian Agrarian Reform Agency (INRA) began implementing these refo rms and titling procedures with little consultation from the communiti es themselves. While owners of rubber estates in Brazil went bankrupt or were forced to sell their la nds in the 1970s, owners of private forested estates in Pando tried to maintain their pr ivileged position (Stoian, 2000). Throughout the 1990s and early 2000s there was a str uggle between these large landholders and peasant and indigenous communities to cont rol forest resources and lands (Cronkleton et al., 2007). Complementing this communally-held l and in Bolivia, the 1.8 million hectare Manuripi National Wildlife Reserve was o fficially created in 1973 for biodiversity conservation, although it was not officially managed as such until 1999 (Miserendino et al., 2003; Khne, 2004). While most land in the reserve is held in large (1,000-80,000 ha) private estates (Khne, 2004) there are also nine reserve communities, eight along main roads and the ninth located along t he Madre de Dios River. Similar to communities outside the Manuripi Reserve, fam ilies within the reserve live in clustered settlements, and only during the Br azil nut harvest season do they relocate to their often remote forest landholdings. Recently, local non-governmental and research organizations in Pando have supported communities in participatory mappi ng of Brazil nut stands. This mapping is seen a way to empower rural people thr ough documentation and clarification of customary resource rights, and in many ca ses comply with the technical norms for Brazil nut management in Bolivia (Ministrio de Desarrollo Sostenible, 2006). Individual 84

PAGE 85

Brazil nut trees are mapped and numbered, their proprietor is identified, and final maps are returned to communities to allow t hem to visualize and negotiate customary tree tenure with other harvesters and government officials. Methods Field Evaluation of Brazil Nut Harvest and Management Practices Brazil nut harvest and management practices were evaluated in eight communities in Pando and four in Acre. We interview ed 189 households (131 in Pando, 58 in Acre) and participated in the 2006 and 2007 Brazil nut har vests. In Pando, communities were chosen to represent differences in market access (river vs. road) and distance to major market centers, and were located both within (4) and outside (4) of the Manuripi National Wildlife Reserve. In Acre, the four communities sampled differed in their distance to central markets, but all were in the Chico Mendes Extractive Reserve and had access to secondary roads (Figure 33). Large distances between households in Acre extended field survey time, precl uding an equal sample size between the two countries. In communities with <30 families, all avail able families participated in the study; in larger communities (>30 families), represent ative samples were chosen at random from lists of total households generated through consultation with community leaders. Harvest and management practices were cat egorized by: (1) initia l harvest date; (2) harvest method and overall har vest duration; and, (3) m anagement practices designed to (a) promote regeneration (e.g ., protection of seedlings from fire), (b) enhance fruit yield (e.g., vine cutting), and (c ) meet certification standards (e.g., nut drying). These management variables were chosen from lo cal literature on best management practices (e.g., Wadt et al., 2005) and the first authors participation in the 2006 Brazil nut harvest. 85

PAGE 86

We also recorded apparent lack of management, as well as intentional or unintentional practices that may have adverse effects on t he species (e.g., cutting into the trees vascular tissue to increase fruit production) During the 2006 and 2007 harvests, we collected data on reported nut thefts based on informant accounts of where and how many nuts were stolen by whom, and what, if anything, was done to resolve the matter. Finally, in three communities from our samp le in Pando that had mapped their Brazil nut stands in 2005 or 2006, nut harvesters freelisted perceived costs and benefits of the mapping exercise. Quantification of Rural Livelihoods In collaboration with the Center for Inte rnational Forestry Research Poverty and Environment Network (CIFOR PEN; http://www.cifor.cgiar.or g/pen/_ref/home/index.htm), we conducted two annual village and household-level socio-economic and environmental questionnaires and four quarterly household questionnaires with the same 189 households from June 2006 through August 2007. We quantified land tenure and access variables through the village and household-level annua l surveys. Quarterly interviews over a 12-month period focused on all subsistence and cash income derived from forests and other onand off-farm activities, including agriculture livestock, wage labor, and other sources of external economic support. We used this detailed evaluation to calculate total household income and analyze the relative contribution of Braz il nut to household livelihoods in Pando and Acre. Data Analysis To understand basic quantitative diffe rences between Brazil nut-producing households in Pando and Acre, we generated de scriptive statistics for: (1) household 86

PAGE 87

characteristics, land tenure, access and inco me; (2) incidence, type and financial costs of nut theft; and, (3) nut harvest and management practi ces. We then performed several analyses to evaluate variables that predicted, or were predicted by, presence or absence of theft. For these analyses, the da taset included measurements from 171 of the total 189 households (125 in Pando and 51 in Acre) that had robust records over two consecutive years (2006 and 2007). Ther efore, this dataset corresponds to repeated measures with two time points. Potential quantitative explanations for theft In the first analytical stage, we sear ched for relationships between reported incidence of theft and possible predictor vari ables for both countries with the two years of data combined. For discret e variables, such as road or river access, Brazil nut tree mapping status, and perceptions of property rights security, chi-square tests were used. For continuous variables, such as distance to market, distance from households to individual nut stands, and income derived from Brazil nuts, logistic regressions were fitted. This process guided our selecti on of potentially relevant variables (p 0.10) to be included in a more complete, single model to analyze theft. For this, we developed a stepwise generalized linear model based on a binomial distribution and a logit link where model response was presence or ab sence of theft. Autocorrelation of the errors was evaluated because these dat a contained repeated measures, but was not statistically signifi cant and therefore dropped from the analysis. Models were fitted using the software SPSS Statistics GradPack 17.0 (2008). Potential quantitative im plications of theft on ha rvest and management In the second analytical stage, we searc hed for relationships between reported incidence of theft and possible harvest and management response variables across four 87

PAGE 88

subgroups comprising country x year combinati ons. Again, for discrete variables, such as Brazil nut harvest method and specific m anagement practices, chi-square tests were used. For continuous explanatory variables such as length of harvest and amount of nuts collected, ANOVA tables we re obtained. Again, this process guided our selection of potentially relevant variables (p 0.10) to be included in a model that combined data from all countries and years to investigate presence or absence of theft as explaining harvest and management practices. We modeled each individual selected res ponse variable considering a year and group factor, together wit h its interaction. Year and group were assumed to be fixed effects. The group factor was a categoric al variable formed by three classes: (1) observations with theft in Pando; (2) observations without theft in Pando; and, (3) observations without theft in Acre. Thefts in Acre were so rare (only 11 reported over a 2-year period) that a fourth potential class (i.e., observations with theft in Acre) was not statistically viable for inclusion in the model. In this analysis, the presence of autocorrelation among observations was relevant and therefore incorporated into the model as a random effect. The continuous and discrete responses were fitted by using a linear mixed model and a generalized linear mixed model, respectively, as implemented in GenStat v. 11 (Payne et al., 2007). Results Comparative Household Characteristics, Land Tenure, Access and Income Data Overall, while households in Pando and Acre were similar in terms of size, land area, and time lived in the community, they differed significantly in their access to markets and forest resource s (Table 3-2). While more than half of households sampled in both places had access to roads from v illage centers, communities in Pando were 88

PAGE 89

located much farther from market centers (p<0.001). Also, in Pando, distance from households to individual Brazil nut stands wa s greater than in Acre (p<0.001). This reflects the concentrated nature of communi ties in Pando, which contrasts with the more diffuse spread of Acrean forest-based households. While total income was similar, income derived from forests and Brazil nuts was greater in Pando (Table 3-2). Forest income made up 63% of the total share in Pando, whereas in Acre, 42% of household income came from forests (Figure 3-4). Similarly, Brazil nut income, as a subset of forest-bas ed income, contributed significantly more to household livelihoods in Pando than in Acre (Table 3-2). In Pando, Brazil nuts alone contributed 43% of the total in come share, whereas in Acre, Brazil nuts contributed just 14% (Figure 3-4). Report ed mean (and SD) volume of nuts per household in Pando was 6187() kg in 2006 (n=125) and 4508() kg in 2007 (n=115). In Acre it was much less at 1807() kg in 2006 and 1812 () kg in 2007. Frequency of Reported Brazil Nut Thefts Theft of Brazil nuts was reported much more frequently in Pando than in Acre (p<0.001). In Pando, 61% of 125 househol ds and 45% of 115 households reported nut thefts in 2006 and 2007, respectively. In Acre, only 14% of 51 households and 9% of 42 households reported nut t hefts in 2006 and 2007 harvests, respectively. While we may have expected fewer thefts in Acre due to the smaller sample size, our findings were reflected in data gathered in a separate extractive community in Acre with only 4% of 26 households reporting nut thefts in 2007 (Duchelle, unpublished data). In both Pando and Acre, most thefts were of nuts we re from the forest floor, rather than of sacks or storage areas (Figure 3-5), suggesting that some reported thefts were due to confusion over customary tree tenure, rather than outright stealing. 89

PAGE 90

The people who reportedly stole nuts we re members of the same community, members of neighboring communities, or, at least in Pando, temporary workers employed in private estates or communities dur ing the harvest season (Figure 3-6). In Pando, half of all thefts were reputedly co mmitted by members of the same community (45% in 2006, 52% in 2007). In Acre, the 11 cases of thefts were reportedly perpetrated by other harvesters from t he same and neighboring communi ties. Thefts by hired migrant workers, such as those observed in Pando, did not occur in Acre where there was no wage labor for nut collection or neigboring private esta tes with migrant workers. Financial Value of Brazil Nut Thefts In Pando, reported volumes of Brazil nut thefts were substantial (Table 3-3), comprising 22% of the total combined harvest in 2006 and 2007. This annual loss, which averaged US$719 (range $11-$5750), is considerable given that mean annual combined subsistence and market income per household in Pando at this time was US$5394 (SD$4764); mean income loss was 13%. Also, in Pando, 69 kg of nuts is equivalent to a days labor (paid upon sale of the nuts), so the esti mated mean theft of 1498 kg per year represents the loss of approximately 22 days of household labor. In Acre, with minimal reported thefts and propor tionately less income from Brazil nuts, financial consequences of thefts were far less important. Quantitative Explanations of Brazil Nut Thefts Results of the stepwise generalized linear model showed that access variables were the most important quantitative determinants of Brazil nut theft (Table 3-4). Households with nut stands farther away fr om the home tended to experience increased theft of nuts (p<0.001), as did households in communities further from markets (p=0.001). While there may have been a slight tendency (p=0.132) for households with 90

PAGE 91

greater income derived from Brazil nuts to be more vulnerable to t heft, total landholding (p=0.311), Brazil nut mapping (p =0.524), and river versus road access (p=0.602) were not significant in explaini ng incidence of nut theft. Quantitative Implications of The ft on Brazil Nut Harvest Regimes Brazil nut harvest regimes we re distinctly different in Pando and Acre. Harvest began much earlier and lasted much longer in Pando than in Acre (Figure 3-7). The year x group analysis to compare harvest and management patterns across countries and levels of theft showed a significant di fference between both harvest start date and duration of harvest in Pando versus Acre (p<0 .001), as well as quantity of nuts collected (p=0.007). No country-level differences were detected within Bolivia, however, when comparing households that direct ly experienced theft with those that did not (Table 3-5). Also, harvesters reported the use of three different collection methods: (1) collect as many fruits as possible in one day, br eaking them open for immediate transport of nuts from the forest; (2) collect and group fruits from one trail into piles, leaving them in the forest for one to several days and then returning to open them; or, (3) over the course of several days or weeks, collect and group all fruits from ones landholding and then return to open them later. Notably, 96% of Bolivian collect ors, whether they experienced theft or not, gathered, opened and transported nuts in the same day, whereas Brazilian producers engaged in a variety of collection methods that did not depend on rapid transport of nuts out of the forest (Figure 3-8). Most Bolivian producers indicated threat of theft as the main reason they transported nuts out of the forest early in the season and as quickly as possible. With the exception of the collection practices described above, no correlations were found between incidence of theft and implementation of any other specific 91

PAGE 92

management practices (e.g., protecting seedlings from fire, vine cutting) in Pando and Acre. This suggests that despite threat of theft, people felt enough proprietorship over their Brazil nut stands to manage for future fruit production and tr ee growth. Resolution of Thefts Brazil nut thefts were gener ally unresolved, meaning vi ctims of thefts never received compensation for their losses. In Pando, only 17% of the 76 cases of theft in 2006 and 6% of 52 cases in 2007 were resolved. Of the 11 cases repor ted in Brazil, in 2006 and 2007 combined, only two were resolved. Resolution usually involved discussions between the two conflicting parties, with full or partial success in the return of stolen nuts. In some ca ses, community officials medi ated the disputes. In one extreme case in Pando, where workers of a former private estate owner stole an estimated 4600 kg from a newly-titled community, a municipal official helped resolve the issue. The officer attended a community m eeting, and exercising his authority, ensured that the money from the st olen Brazil nuts was returned and re-distributed to the whole community. Community Perceptions of Pa rticipatory Mapping in Pando Among producers who mapped their Brazil nut stands with local non-governmental organizations, most of those interviewed perceived mapping as moderately to highly beneficial (76%, n=45). The main perceived benefit was new knowledge of the stands (35%; i.e. number of trees, lim its), followed by perceived reduction in theft (29%), and increased efficiency of nut collection becaus e trails and trees were more clearly identified (12%; Figure 3-9). Twenty percent felt that mapping had no effect on securing tree tenure or reducing Brazil nut thefts, and only a few producers thought their mapping experience was negative (4%). Common negat ive perceptions t hat emerged, even 92

PAGE 93

among those who generally categorized the proc ess as beneficial or neutral, included the idea that newly cleaned and more visible trails from the mappi ng process made it easier for hired collectors, generally from out side the community, to enter into Brazil nut stands to steal nuts, actually increasi ng the incidence of thefts as opposed to decreasing it. There was also criticism of the use of nails to secure numerical identification markers into tr ees; many producers felt this reduced fruit production. Finally, producers who had customary rights to mo re trees and/or forest feared that by making their relative wealth transparent th roughout the community, they risked losing a portion of their area to those with less, if property rights were to be made more equitable. Discussion We found a much greater incidence of reporte d Brazil nut thefts in Pando, Bolivia, than in the adjacent state of Acre, Brazil. Whereas it is di fficult to confirm causality, distinct historical events and policy changes that led to different country and harvest contexts, combined with the results of our quantitative analyses, suggest that two main differences may explain the variat ion. First, there was a clear contrast in the timing and process of formally recognizing property rights of rural communities in Acre and Pando. Second, different settlement patterns and degrees of economic dependence on Brazil nuts may have either discouraged or inadver tently encouraged nut thefts in the forest, as supported by our quantitative results. As a result, both threat of theft and economic dependence on Brazil nuts have important implications for management and conservation of Brazil nut rich forests in Western Amazonia. 93

PAGE 94

Comparative Timing and Process of Property Rights Formalization Whereas Acre and Pando have similar forest ecosystems and extractive histories, the differing outcomes of land tenure reform s in the two places are based on when and how property rights were devolved to local co mmunities. A clear distinction has been made between top-down decentra lization and bottom-up approac hes that are motivated by social movements and local governments (Larson et al., 2007) with demand from below considered an important component of effective decentralization measures (Larson and Soto, 2008). In Acre, the Extrac tive Reserve model was a response to a bottom-up, successful political struggle by ru bber tappers to secure property rights. As a result, customary tenure of dispersed household settl ements throughout the forest was recognized when land rights were forma lly devolved. The inner boundaries of individual landholdings continued to be as we ll defined among extractivists as they were during the rubber era. The outer boundary of the Chico Mendes Extractive Reserve was relatively undisputed when the reserve was created, because claims contested by large landholders were mostly settled by the federal government and heavily regulated by the Brazilian environmental protection agency (C.V. Gomes, pers. comm.). At the time of this field research (2005), re sidents of the Reserve had more than 15 years to adjust to the proper ty rights supported by reserve creation. Reserve comanagement with the Brazilian Government also helped secure their rights to land and resources. Both bottom-up demand for land reform from the social movement and the relative longevity of the Ex tractive Reserve are key to relatively secure land tenure within the Reserve today. In contrast to the well-established Ch ico Mendes Extractive Reserve in Acre, communities in Pando only recently emer ged from a struggle with private estate owners 94

PAGE 95

to have their customary property rights recogni zed by the government. Additionally, this formalization of rights was hande d down from the state in older communities that had managed their Brazil nut groves communally for many year s via internal norms and rules. In many cases, co mmunal titles granted by the government were incongruent with traditional boundaries of t he communities seasonal forest use (Cronkleton et al., 2008), and maps presented to communities lacked geographic features that were familiar to residents, causing boundary pr oblems to go unnoticed (Cronkleton et al., 2007). Our data showed the greatest proportion of thefts in Pando was presumed to be by members of the same community. Th roughout Pando, such intra-community conflicts occurred as the titling process created the expectation that the government would rearrange internal resource access so that everyone could have a 500 ha plot, undercutting the traditional tree tenure system. Also, in newly-formed communities that only recently had land rights devol ved from a large landholder rights to specific Brazil nut trees were ill-defined among residents. Additionally, there was often abrupt incorporation of landless people into co mmunities to access land and municipal services; these newcomers were probably unaware of customary management systems. In sum, the more recent, top-down process of formalizing communal tenure in Pando may have resulted in greater property rights insecurity. Settlement Patterns and Resource Dependence Influence Theft The distinct spatial distribution of fore st-dwelling communities in Acre versus Pando contributed to observed differences in Brazil nut thefts. This finding was supported by our quantitative re sult that distance of househol d to Brazil nut stand most strongly explained theft. In Acre, through creation of the Chico Mendes Extractive Reserve, households maintained their tr aditional dispersed spatial distribution 95

PAGE 96

throughout the forest, allowing them close access to and daily interactions with their forested landholdings (Allegretti, 1990). These nut collectors walk their properties to hunt, tap rubber, work their agricultural fi elds, and travel to neighboring households. They constantly monitor their land and res ources, dealing with disputes quickly. In Pando, in contrast, the 1994 Popular Particip ation Law encouraged the concentration of rural families in settlements to access muni cipal government education, health, water, and electricity services. This policy had the unintended result of removing families from day-to-day monitoring of their forest resources. Those a llotted harvest areas far from village centers experienced mo re nut thefts, likely by members of neighboring communities and migrant workers. In Pando, disputes might only manifest themselves during the Brazil nut harvest period, which contrasts markedly with the year-round patrols of Acrean extractivists of their forested landholdings. Additionally, households in communities farther from markets were more vulnerable to theft. This re sult may be due to a combination of three factors: (1) weak enforcement of local rights by gov ernment agencies may be exaggerated in communities distant from main markets w here these organizations are based; (2) the most remote, riverine community in the Pando sample had t he most internal conflicts, due to recent de facto devolution of rights to communi ty members from the resident estate owner; and (3) due to more infras tructure in Acre, sampled Brazilian communities, where there was le ss reported nut theft overall, were closer to markets than those sampled in Pando. In addition to differences in spatial settl ement patterns, higher relative dependence on Brazil nuts in Pando (versus Acre) may al so have influenced theft of this high-value 96

PAGE 97

resource. Economic dependenc e on Brazil nuts was one of the main quantitative differences observed between Pando and Acre. Due to more livelihood options in Acre, Brazil nuts were relatively less important economically, and considered just one of many seasonal income and subsistence activities Although we found only weak statistical evidence that increased income from Brazil nuts explained theft, th e relatively high value of this one product in Pando is undeniabl e. Indeed, when a dramatic rise in nut prices in 2005 occurred, violent conflicts ensued over Brazil nuts t hat resulted in deaths of several people in Pando (El Deber, 2005); no such response occurred in neighboring Acre. Implications of Theft Threat and Economic Dependence on Brazil Nut Harvest and Management The difference in threat of Brazil nut t heft seemed to be reflected in the use of different harvest practices in Pando and Acre The low incidence of nut thefts in the Chico Mendes Extractive Reserve in Acre allows harvesters to be flexible in their collection method. Acrean harve sters typically are less conc erned with potential theft of nuts, and thus begin their harvest after most fruits have fallen to the ground. Harvesting later in the fruit-fall season allows them to concentrate t heir efforts and collect all nuts within a few weeks. They then transport nuts out of the forest all at once using draft animals, which overall is an easier and more effi cient practice. In c ontrast, harvesters in Pando gather, open, and transport nuts from a subset of thei r trees in one day, and then repeat this process throughout their stand until all trees have been visited. They are also compelled to collect nuts early in the season, which is a dangerous and sometimes fatal practice as the heavy fruits are still falling from trees up to 50 m tall. This collection method is also relatively i nefficient. Nuts must be carri ed on the harvesters shoulders, 97

PAGE 98

and then often by motorcycle to a storage area or temporary hiding place. During our field research, they consistently reported that these proc edures diminished the possibility that their gathered fruits coul d be stolen. Although our data showed no differences in harvest methods between ho useholds that did or did not experience thefts, we conclude that the general climate of resource insecurity in Bolivia forces all harvesters to manage their stands as if thefts could occur at any time. The extended Brazil nut harvest in P ando can also be attributed to rural households heavy economic dependence on th is one product; they need all the cash they can get from their nut stands. Two ecological stud ies that measured Brazil nut harvest intensity by humans in Acre and Pando, respectively, support this difference in economic dependence. Wadt et al. (2008) reported nut colle ction intensities of 45% and 71% of fallen fruits harvested at two site s in the Chico Mendes Extractive Reserve in Acre, which is much lower than the esti mated 93% collection intensity reported at sites in Pando by Zuidema and Boot (2002). Producers in P ando visit the same trees many times over the harvest season because fruit fall naturally occu rs over a period of weeks to months. Although speculative, the Brazilian harvest method may allow a longer period for the scatterhoarding agout i to both consume and disperse nuts, potentially influencing long-term sustainabilit y of the overall Br azil nut production system. Also, nut prices are generally high er later in the season as middlemen and companies attempt to fill their yearly quot as. This dependence on Brazil nuts in Pando is likely not only a reason for nut thefts, but also enhances the negative effects of thefts on households as their vulnerability incr eases with economic dependence on this one product. 98

PAGE 99

Ecological and economic trade-offs exist between Brazil nut harvest systems in Pando and Acre. Although extending the nut harvest season in Pando to collect as many fallen fruits as possible may not prom ote optimal regeneration over the long term, clearing fallows for pasture, as observed in Acre, has its own implications, since these fallows provide favorable regeneration sites for B. excelsa (Cotta et al. 2008). Different harvest systems in Acre and Pando al so have market consequences. A main impediment to international Br azil nut sales is contaminat ion by aflatoxins, which are both toxic and carcinogenic. When Europe an importers raised their quality standards for Brazil nuts in 1998, access to the European market was imperiled (Newing and Harrop, 2000). Nuts transported out of the forest just after fr uit fall, as in Bolivia, are less likely to be contaminated because aflatoxins are caused by the fungus Aspergillus which thrives under hot and humid condition s (Hudler, 1998). In Acre fruits that fall in December and are not collected until February or March, appear to have higher risk of Aspergillus contamination (Souza lvares et al., 2009 ). Organic and Fair Trade certification of Brazil nuts, through the sale of an aflatoxi n-free product and affiliation with cooperatives, have been important in stabi lizing and even increasing the price producers receive for nuts and encouraging better management throughout the Western Amazon (See Chapter 4). Interestingly, co llection earlier in the season is now being promoted as a best management practice in Acre as one step towards attaining certification, which may al ter harvest regimes. Participatory Mapping to Promote Property Rights Security Participatory mapping of Brazil nut st ands may be a good way to help promote local property rights security. In one case in Pando, community members worked together to determine agreed upo n customary ownership of trees and trail networks, 99

PAGE 100

which allowed mediati on over internal resource disputes to begin during the mapping process itself (Figure 3-10). Such negotia tion of customary ri ghts can be considered akin to that undertaken by harvesters in Acre on a daily basis as their day-to-day relationship with the forest and with each other in the forest setting is much stronger. Final map products allowed communities to vi sualize hot spots of conflict as the proprietors of individual trees and trails we re clearly identified. In one case, when a Bolivian producer realized that most of his Brazil nut stand fell within the bounds of another community, he became an official me mber of the neighboring community to resolve this issue. In another situation, new community lands were leveraged based on traditional collection areas delineat ed through the mapping process. Producers perceptions of distinct m apping experiences can help guide regional expansion and adaptation of mapp ing initiatives. For instance, although the mapping processes in Pando were considered benefic ial overall, a common negative perception was that nailing of metal identification markers in adult Brazil nut trees prior to 2006 was to blame for the low fruit production and ev en the drying up and death of some; in reality, there was a natural decrease in Braz il nut production region-wide in the harvest season of 2006 (Kainer et al., 2007) that became even more pronounced in 2007. Because natural production rose to extremel y high levels in 2008 (L.H.O. Wadt, pers. comm.) this misconception may be debunked. Nevertheless, such misconceptions must be addressed for mapping to be effective in helping bridge the gap between formal and customary tenure. Conclusion Several important lessons can be learned fr om this comparative study in Western Amazonia. First, the outcome of land tenure reforms depends on the timing and 100

PAGE 101

process of property rights fo rmalization in local communiti es. In Pando, as demand for clear property rights increased wit h rising prices of Brazil nuts, granting communal titles using a top-down approach sometimes had the unanticipated consequence of enhancing property rights insecurity in any particular community. Bolivian government agencies were unable or unwilling to resolve tenure conflicts that ar ose from competing interests either within communities or bet ween communities and other, more powerful actors (Cronkleton et al., 2007). In the case of Pando, such insecurity appeared to take the form of a high incidence of nut theft. In contrast, in Ac re, land conflicts and resource thefts are currently rare among extractive communities. The ex ternal Extractive Reserve boundary was strongly enforced and cu stomary internal boundaries left flexible to accommodate traditional forest use. T hese Acrean harvesters were considered comanagers of land and resources with the Brazilian federal government as they were central to the bottom-up reforms that resulted in creation of the Extractive Reserve system. That said, property ri ghts security does not necessarily translate to forest conservation as producers in Acre exercise their rights to convert small portions of their forested landholdings to pasture. Second, different settlement patterns and market forces appear to have either discouraged or inadvertently encouraged nut thefts in forests. In the case of Pando, people ar e much less connected with their forested landholdings except during the nut harvest. This lack of physical connection combined with a product that is in high demand may increase the likeliho od of conflictive situations. The threat of Brazil nut theft in Pando, along with high economic dependence on this one product, locks collectors into a man agement regime that is clearly dangerous 101

PAGE 102

and perhaps detrimental to long-term viability of future Brazil nut tree populations. Additionally, given the relati vely high proportion of rural household income derived from Brazil nuts in Pando, forest-dwelling communities are more vulnerable to the negative effects of a stolen harvest. Collectors unable to glean suffi cient income from nuts are likely to turn to other land uses, which may involve converting Brazil nut-rich forests to agriculture or pasture. Nonetheless, earlier collection already practiced in Pando may have the benefit of allowing acce ss to certified markets, as seen in a recent push in Acre for producers to adopt such practice. As further research expl ores the effects of collection on Brazil nut r egeneration and aflatoxin contam ination, property rights security is an important base from which communities can choose whether or not to incorporate such recommended practices in their harvest and management regimes. Participatory mapping is a potentially importa nt tool for communities in dealing with tenure conflict, visualizing traditional forest use systems, and leveraging integration of traditional practices into formal land titling processes and decision-making. This activity may allow governmental and non -governmental organizations to play an important supporting role in smoothing prop erty rights transitions. As trends continue to designate forests for legal use by local communities it is essential to understand and address resource conflicts, their causes and pot ential solutions to promote long-term management of the worlds tr opical forests. 102

PAGE 103

Figure 3-1. Map of Madre de Dios, Peru, Acre, Brazil and Pando, Bolivia (MAP region) with Interoceanic Highway and 100 m buffer in red. Figure 3-2. Google Earth images from July 2007 of forest dwelling communities in Acre, Brazil (left) and Pando, Bolivia (right). Note contrast between dispersed pattern of household clearings in Acres Chico Mendes Extractive Reserve (households circled) and a concentrat ed community settl ement in Pando (community circled). 103

PAGE 104

Figure 3-3. Map of communities sampled in Acre and Pando. Credit: N. Hoyos 104

PAGE 105

Figure 3-4. Share of total household income derived from the forest (%) and from Brazil nuts alone (%). Income includes both cash and subsistence values. Values reflect households in Pando, Bolivia (June 2006-June 2007) and Acre, Brazil (August 2006-August 2007). 105

PAGE 106

Figure 3-5. Where Brazil nuts were reportedly stolen (% of total incidents). Sample includes only those households that ex perienced Brazil nut thefts in 2006 and 2007. 106

PAGE 107

Figure 3-6. Suspected perpetrators of Brazil nut thefts in 2006 and 2007 (% of total incidents). Sample includes only t hose households that experienced thefts. 107

PAGE 108

Figure 3-7. Modal start and end dates of Brazil nut harvest, and harvest duration, in Pando and Acre in 2006 and 2007. Prim ary harvest period (dark gray) is when most producers report the begi nning and end of harvest. Extended harvest period (light gray) is indicated by earliest and latest harvest dates. 108

PAGE 109

Figure 3-8. Percentage of households using s pecific Brazil nut harvest methods among 3 groups: (1) Bolivia without nut thefts ; (2) Bolivia with nut thefts; and (3) Brazil without nut thefts. In ever y case the proportion of households implementing these methods in Brazil was statistically different when compared to proportion of households in Bolivia (p<0.001). There were no differences observed in harvest methods employed by the Bolivian groups (with or without Braz il nut thefts). 109

PAGE 110

Figure 3-9. Free-listed benefits of parti cipatory mapping in Pando, Bolivia (45 producers; 55 benefits given). 110

PAGE 111

Figure 3-10. Example of a participatory map of individual Brazil nut trees in Pando. Note that official polygons are often incongruent with traditional use areas. Each color represents the Brazil nut stand of an individual household.. Courtesy of Cronkleton et al. 2008. 111

PAGE 112

112 Table 3-1. Important historical events and rele vant forest policy changes in Acre, Brazil, and Pando Bolivia. These events conti nue to affect property rights, the diversity of forest products currently exploited, and ultimately, the frequency of Brazil nut thefts. Acre, Brazil Pando, Bolivia 1876-1910: Rubber boom (migration to region) 1903: Acre ceded from Bolivia to Brazil 1870s: Nicolas Suarez founds Suarez Hermanos rubber company 1910-1940: Decline in rubber economy; land-use diversification Former rubber estates diversify production to include Brazil nut harvesting and agriculture 1931-35: Suarez Co. introduces a Brazil nut shelling company by a mostly female labor force 1940-1945: Renewed demand for rubber 1942: Brazil-U.S. Washington Accords to recruit Brazilian rubber tappers to Amazon Suarez and Hermanos control 80% of rubber production in Brazil-Bolivia border 1950-90s: Brazil nuts replace rubber as main forest product 1986: Removal of Brazilian subsidy for rubber that had been extended to Bolivian producers 1990s: New policies for extractivist communities 1990: Extractive Reserves 1996: Forestry Law and Agrarian Reform Law

PAGE 113

Table 3-2. Descriptive statistics of household characteristics, land tenure, access and income in Pando and Acre. Pando, Bolivia Acre, Brazil Variable N Mean (SD) N Mean (SD) p-value Household characteristics Size (# people) 125 6.0 (2.8) 51 5.4 (2.9) 0.159 Land area (ha) 125 587 (470) 51 638 (499) 0.519 Head of household born in community 125 0.20 (0.40) 51 0.20 (0.40) 0.670 Time lived in community (yrs) 125 17 (15) 51 15 (11) 0.238 Schooling (yrs) 125 3.5 (2.3) 51 2.2 (1.4) <0.001 Land tenure Protected area 125 0.50 (0.50) 51 1.00 (0.00) <0.001 Perceived land tenure security (Rank 1-3; 3 most secure) 113 2.27 (0.85) 38 2.26 (0.76) 0.943 Mapped Brazil nut stand 125 0.42 (0.50) 51 0.39 (0.49) 0.772 Access Road access 125 0.65 (0.48) 51 0.78 (0.42) 0.078 Distance to market from village center (hrs) 125 9.5 (7.1) 51 2.4 (1.4) <0.001 Distance to Brazil nut stand from household (min) 124 81 (93) 51 6 (4) <0.001 Household income 2006-2007 Total income (USD) Forest-based income (USD) Brazil nut income (USD) 107 107 107 5394 (4764) 3423 (3604) 2304 (1837) 47 47 47 5460 (2759) 2319 (1295) 765 (973) 0.930 0.043 <0.001 113

PAGE 114

114 Table 3-3. Total volume and value of Brazil nuts reported stolen in 2006 and 2007 combined. Values in USD based on mean price/year/paid to collectors by middlemen during the 2006 ($0.46/kg in Pando; $0.40/kg in Acre) and 2007 ($0.50/kg in Pando; $0.44/ kg in Acre) harvests. Pando, Bolivia Acre, Brazil Variable N Mean (SD) Range N Mean (SD) Range Amount stolen (kg) 104 1498 (1810) 23-11500 9 216 (158) 58-483 Percent of harvest (%) 104 22 (19) 0.82-100 9 10 (7) 0.5-23 Value (USD) 104 719 (875) 11-5750 9 90 (67) 23-202

PAGE 115

Table 3-4. Results of a stepwise logistic regression to identif y the most important measured variables that explained incidence of Brazil nut thefts in 2006 and 2007 in Pando and Acre (n=235). No significant interactions between variables were detected. Variable Estimate Wald P-value Odds Ratio Distance to Brazil nut stand from household (min) 0.001 12.371 >0.001 1.001 Distance to market from village center (hrs) 0.023 10.981 0.001 1.023 Income from Brazil nuts (USD) 0.00002 2.274 0.132 1.000 Total landholding area (ha) 0.00006 1.027 0.311 1.000 Mapped Brazil nut stand 0.035 0.406 0.524 Road access -0.48 0.272 0.602 Table 3-5. Mean and SE of Brazil nut harve st start date (expressed in Julian days where 1=November 8th), quantity of Brazil nuts collected (kg) and harvest length (days) in three Braz il nut harvest contexts in Bolivia and Brazil in 2006 and 2007. Variable 1. Bolivia: Thefts 2. Bolivia: No thefts 3. Brazil: No thefts Harvest start date 60.53 (1.98) 56.37 (2.08) 86.44 (2.48)*** Quantity 163.12 (13.16) 153.53 (13.08) 114.338 (12.06)** Harvest duration 94.74 (3.66) 91.92 (3.83) 43.25 (4.43)*** Notes: Group 3 differed from Groups 1 and 2 for all variables: ***p <.001; **p<.01. Year x group interactions were not significant. 115

PAGE 116

CHAPTER 4 DOES CERTIFICATION LEAD TO BETTER MANAGEMENT AND IMPROVED INCOMES? A COMPARATIVE ANALYSIS OF THREE CERTIFICATION SCHEMES APPLIED TO BRAZIL NUTS IN WESTERN AMAZONIA Environmental, Social and Economic Certification of Forest Products In recent years, a variety of certif ication and labeling sch emes have emerged to address issues of environmental degradation, social injustice, and consumer health. Certification is a market-based tool used to improve the quality, safety, or management of certain products against a set of defi ned standards through third-party auditing (Bass et al., 2001). The innovation of certification is the use of the market to transform the market (Taylor, 2005a). On one end of the commodity chain, there is a consumer who is willing to pay more for a product that is l abeled as environmentally-friendly or sociallyjust, and on the other, there is a producer who seeks price premiums and better market access from use of superior practices (Bass et al., 2001). Certification has also been conceptualized as a regulatory mechanism that alters the rules of market governance (Taylor, 2005b), a signal mechanism that illumi nates organizational characteristics and practices, and a learning mechanism that promotes knowledge transfer and experiential learning (Overdevest and Rickenbach, 2006). Certification in the forestry sector is considered the most advanced of the environmental certification schemes (Auld et al ., 2008). A review of forest certification to date shows that large-sized operations on public or in dustrial lands, production of high-value goods, exports to Europe and North America, support by producer associations, NGO pressure, and government support positively influence producers decisions to adopt certification (Auld et al ., 2008). While forest certification has traditionally focused on timber, certification of non-timber forest products (NTFPs) is 116

PAGE 117

viewed as one way to bolster livelihood benefi ts of these products for rural communities and promote sustainable harvest over the long-term. According to Walter (2003), there are four main certification systems relevant to NTFPs: (1 ) Organic certification, which emphasizes agroecosystem health on the pr oduction end and consumer safety on the consumption end; (2) Product qua lity certification, which includes product consistency and safety standards; (3) Social certificati on, including Fair Tr ade, which promotes equal benefit sharing among NTFP producer s and decent labor conditions; and (4) Forest management certification, which fo cuses mostly on the ec ological aspects of NTFP harvest and management, but also in cludes social and economic standards; While the different certification schemes have distinct goals, there is harmony among them. Organic and product quality certifications ar e best suited for foodand pharmaceutical-based NTFPs, which have the additi onal criteria of certification for food safety and quality. Fair Trade certification is socially-o riented in seeking to encourage and recognize operations that ensure better work conditions fo r small producers. Forest management certification for NTFPs pr omotes better management and forest conservation, and provides greater loca lized benefits for community-based products (Shanley et al., 2002). Combinations of t hese certification schemes may help promote environmental, social, and economic sustainability of NTFP management by communities. This research compares the managem ent and economic benefits of Brazil nut certification systems for communities in nei ghboring Pando, Bolivia, Acre, Brazil, and Madre de Dios, Peru. We focus on this Brazil nut-rich border region in Western Amazonia (Figure 4-1) because it is the cu rrent center of the Brazil nut economy, 117

PAGE 118

employing tens of thousands of families dur ing the primary collection season (January March; Bojanic, 2001). Because of Br azil nuts regional economic importance, Brazilian, Bolivian, and Peruvi an legislation prohibit felling the trees. At the producer level, individual Brazil nut trees are consi dered key livelihood assets by forest dwellers in all three countries. Our overall research question is: What are the environmental and economic benefits associated with differ ent Brazil nut certification schemes? We address this question through research with certified and non-certified nut harvesters in the trinational region of Bolivia, Brazil and Peru. In this context, certified participants are mostly organic and Fair Trade producers, with a small group in Peru having also attained FSC certification. Specific objecti ves of the research are: (1) to identify adherence to Brazil nut best management practices by producers in the three countries; (2) to analyze differences in management practices between certified producers and noncertified producers in each country; (3) to quantify the economic benefits associated with supplying or marketing certified nuts; and, (4) to explore producers perceptions of the benefits and costs of Brazil nut certification. To our knowledge, this is the first study that compares organic, Fair Trade, and FSC certification for a given product using field-based data on the environmental and economic returns of certification. Brazil Nut Ecology and Best Management Practices Brazil nuts are collected primar ily from mature forests, and the species combined ecological and economic charac teristics give it the potential to promote forest conservation and contribute to the livelihoods of rural communities. Brazil nut trees are long-lived emergents in regional upland forests and attain up to 3 m in diameter. Due to 118

PAGE 119

its very large size and dense distribution, this species provides important local and landscape-level ecological structural and f unctional roles (Zuidema, 2003). The large, woody fruits fall to the ground during the rainy season and hold approximately 25 seeds (here referred to as nuts) inside the har d pericarp. The scatterhoarding agouti ( Dasyprocta spp .) plays an instrumental role in s eed dispersal and buria l as one of the few animals that can gnaw through the pericar p. To collect the nuts, people gather the heavy fruits in large piles, use machetes to break them open, and put the nuts in large sacks to carry them out of the forest. Although one study suggest that persistent harvest over decades may result in insufficient juvenile recruitment, populations in Western Amazonia appear to be viable over the medium-term under a range of harvest intensities (Zuidema and Boot, 2002; Wadt et al., 2008). Harvested nuts are a reliable seasonal contribution to local live lihoods, since there is little variability in fruit production at the population level across years (Kainer et al., 2007). The nuts have a relatively high value on local, national, and internati onal markets, particu larly since 2003 with increased demand and consequ ent higher prices. Government and non-governmental organi zations in Western Amazonia have outlined a series of best management practi ces to support Brazil nut production (Table 4-1). These practices are rooted in forestry legislation, certificat ion standards, research on Brazil nut ecology and management, and producers practical knowledge. They can be categorized as: (1) pre-harvest practi ces, such as mapping adult trees and developing a management plan; (2) silvicultura l practices, including those that promote regeneration (e.g. enrichment planting) and enhance fruit yield (e.g. vine cutting); and (3) harvest and post-harvest practices, such as nut drying, to improve quality. These 119

PAGE 120

practices have been disseminated in the regiona l extension literature (Wadt et al., 2005; Card, 2000) and promoted through Brazil nut cooperatives and at producers meetings. Diverse Brazil Nut Certification Systems with Distinct Goals Brazil nut producers access certified market s largely through affiliation with local cooperatives or producers associations that promote best management practices. While the primary nut certif ication option is a combinat ion of organic and Fair Trade certification, FSC standards for Brazil nuts ex ist for all three countries, and we present one case of FSC certification in Madre de Dios (Table 4-2). Organic and Product Quality Certification Organic certification is focused on product quality for export to specialized markets in Europe and the United States The International Federat ion of Organic Agriculture Movements (IFOAM) is the umbrella organization for organic certification with more than 750 member organizations throughout the worl d (IFOAM, 2009). Third-party organic certification began in the 1970s, and standards focus primarily on barring the use of agrochemicals and genetically-modified organisms (GMOs), al ong with the promotion of additional environmental criter ia. Related to organic cert ification, the FAO/WHO Food Standards Program was created in 1962 to protect consumer s from unhealthy foods. It works through the Codex Alimentarius Commission, which developed international codes for product quality (Soldn, 2003). In the case of Brazil nuts, organic certification by communities is achieved through region al adaptations of IFOAM standards. It is important to note t hat under current managem ent and harvest regimes, all Brazil nuts meet organic standards in that they are collected from natura l forests and not cultivated; the focus of organic certification is thus on product quality. The main requirement for organic certification is a clean and dry, aflatoxin-free, pr oduct. Aflatoxins are both a 120

PAGE 121

toxin and carcinogen, and caused by the fungus Aspergillus under hot and humid conditions (Hudler, 1998). They are attributed with causing liver disease in adults and stunted growth and delayed development in children (Abbas, 2005). Based on these health risks, the U.S. Food and Drug Administ ration set an action level of 20 Ppb for aflatoxins in Brazil nuts, at or above which t hey will take legal action to remove these products from the market (FDA, 2000). In 1998, the European Union decreased acceptable levels of aflatoxins in Brazil nuts to 4 Ppb (Newing and Harrop, 2000), and in 2003 banned Brazil nut imports fr om Brazil due to unacceptably high levels of these toxins. While this ban has since been lift ed, European Union legislation requires that acceptable aflatoxin levels re main at < 4 Ppb. This legislation is currently under discussion, however, based on a recent study by the European Food Safety Authority (EFSA) that shows little effect on public health if aflatoxin levels were to increase to 10 Ppb (EFSA, 2009). In Bolivia, the Bolivian Association of Organizations of Ecological Producers (AOPEB), a member organization of the IFOAM, was esta blished in 1991. In 1995, AOPEB founded Bolicert, which is the local or ganization currently responsible for the inspection and certification of Bolivian Braz il nuts (Soldn, 2003). Of the many large processing plants in Bolivia, only three cu rrently produce organically-certified Brazil nuts. One of these, Tahuamanu, also attained certificati on for Brazil nut quality (Soldn, 2003). This latter certificate is managed by the Bolivian Quality and Normalization Institute (IBNORCA), which reflects the nor ms of the FAO Codex Alimentarius. In Brazil, organic certification of Brazil nuts is attained through the Instituto Biodinmico (IBD), also a member organiza tion of IFOAM. In Peru, the Brazil nut company, Candela 121

PAGE 122

Per, attained organic certification in 2001 through the internati onal Institute for Marketecology (IMO). By 2005, Fast Trade S.A., another Peruvian nut company, also attained organic status through IMO. Organic certification is monitored by c ooperative norms that require members to perform a set of harvest and post-harvest management practices. During harvest, producers are encouraged to remove the fruit placental tissue (locally known as ombligo ) and damaged (cut or rotten) nuts, as these can contaminat e the rest of the harvest. Immediately after harvest, produc ers are encouraged to transport nuts from the forest to storage units that meet t he following specifications: covered, wooden structures built one meter off the ground where nuts avoid contact with potential contaminants, such as petroleum-based fuels, batteries, and live animals (Figure 4-2). Certified producers empty the nuts from sacks to dry in the storage units, stirring them several times a week to assure thorough dryi ng. In some cases, collectors construct open air drying racks in addition to their storage units, which are also elevated off the ground to promote even faster drying. Local NGOs, government agencies, and the cooperatives themselves have financed construction of these structures for both individuals and groups, since a clean, dry pr oduct is the basic requirement of organic certification in the case of Brazil nuts. Fair Trade Certification Fair Trade is a socially-focused certificat ion system that aims to promote economic development among disadvantag ed producers. Fair Trade ce rtification emerged from the Alternative Trade move ment (ATO) in the 1970s, and since 1997, Fair Trade standards have been set by the Fairtrade Labelling Organization (FLO). These standards emphasize terms of sale that favo r small producers through price premiums, 122

PAGE 123

safe working conditions, strong social organization, and more direct links between producers and markets (FLO, 2009). With Fair Trade, the buyer, and not the producer, pays the costs associated with certificat ion (Taylor, 2005a). While Fair Trade certification mostly focuses on agricultu ral products, such as coffee, bananas, and cotton, there are exampl es of Fair Trade certification fo r NTFPs, including wild-collected honey, Brazil nuts, and cashews. Fair Trade certification for Brazil nuts requires producers to be organized in cooperatives. The principal certified Braz il nut cooperatives in Pando are COINACAPA and ACERM. COINACAPA (Cooper ativa Integral Agroextractivistas Campesinos de Pando) formed in 1998 as an offshoot of th e oldest producers cooperative in northern Bolivia, CAIC, which began in 1979 in the city of Riberalta. As of 2007, COINACAPA had 300 members, and by 2009 had 456 members from more than 40 communities. ACERM (Associacin de Campesinos Extrativ istas de la Reserva Manuripi) includes only harvesters from the nine communities within the Reserves bounds. ACERM became accredited in 2005, and as of 2007 main tained its original 56 members. By 2009, this number had grown to more than 200 members. Both cooperatives receive extensive financial support from national and international foundations and nongovernmental organizations. As of 2008, a newly-formed cooper ative in Pando, ACEBA, had attained both organic and Fair Trade status (A. Chamas, pers. comm ). In Acre, at the time of field research in 2007, CAPEB (Cooperat iva Extrativista e Agropecuria de Produtores de Brasilia e Ep itaciolndia) was the leading Brazil nut cooperative for producers within the Ch ico Mendes Extractive Reserve with 130 members. From 2001 to 2004, CAPEB received extensive financial and technical 123

PAGE 124

support from a group of Braz ilian governmental and non-gover nmental organizations to attain organic, Fair Trade, and FSC Brazil nut certification for its members. In 2005, CAPEB attained Fair Trade certif ication, but it is questionable whether or not they ever achieved organic certification, although the administrators informed their members that they were selling organically-certified nuts. By 2007, CAPEB was bankrupt apparently due to poor administration. The governm ental and non-governmental resources are now being directed toward COOPERACRE, which is located in the capital city of Rio Branco with new hope for production of org anically-certified and Fair Trade-certified Brazil nuts. In Madre de Dios, the first certif ied Brazil nut cooperative was RONAP (Recolectores Orgnicos de la Nuez Amaznica del Per), which is aligned with the large Brazil nut company, Candela Per. RONAP accessed the organic nut market through Candela and attained Fair Trade certification in 2004. In 2007, RONAP had 62 members, which grew to 72 as of 2009. There were also several certified nut producers associations in Madre de Dios, one of which, ASCART (Associacin de Castaeros de la Reserva Tambopata), sold organic nuts through Fast Trade S.A. and attained Fair Trade st atus in 2005. A relatively new initiative has emerged to connect certified nut producers in Pando, Acre, and Madre de Dios called the Tri-national Agroextractivist Cooperatives Association. The principal objective of the association is information exchange. Through meetings held 3-4 time s per year, this group has had a measure of success in banding together to negotiate minimum prices with the large international buyer, Equal Exchange. 124

PAGE 125

Forest Management Certification Forest management certification for ti mber really began in 1993 through the independent, non-profit organizati on, the Forest Stewardship Council (FSC). While the FSC is still considered the gold standard in fo rest management certification (Butler and Laurence, 2008), other producerbased competitors have emerged in the United States and Europe (Auld et al., 2008), as well as in Indonesia and Malaysia where national certification schemes have been developed (Dennis et al., 2008). In 1996, the FSC formed a NTFP working group. Mexican chicle was the first product certified, and by 2008, products such as Brazil nuts ( Betholletia excelsa ), copaiba oil ( Copaifera spp.) jarina seeds ( Phytelephas spp.) and another approximately 50 commercial NTFPs were certified under t he FSC label (Shanley et al., 2008). Brazil is a leader in certification for NTFPs, largely due to initiatives by IMAFLORA (Instituto de Manejo e Certificao Florestal e Agrcola ), the primary FSC a ccredited certifying body in Brazil (Azevedo and Freitas, 2002) There are ongoing discussions about whether species-specific guidelines for FSC certification of NTFPs should be created or whether annexes to existing standards are sufficient, especially since the development of species-based standards goes against the spirit of the FSC in certifying forests and not products (Shanley et al., 2008). While forest management certification for NTFPs has grown in recent years, specialized markets for these products remain small in comparison to certified timber (Shanley et al., 2008). Even in timber certif ication, where FSC certificate holders have had to adopt more sustainable practices (N ewsom et al. 2006), the imbalance toward temperate and boreal producers, difficulties of integrat ing smallholders, and the limitations of certification in promoting forest conservation remain major challenges 125

PAGE 126

(Auld et al., 2008). NTFPs are considered ev en more challenging to certify than timber because of the diversity of products in cluded in this broad category and varying certification standards, along with insufficient information on the ecological and social aspects of NTFPs, and, ther efore, lack of consumer demand for these products (Pierce and Laird, 2003). A challenge remains to i dentify specialized markets for NTFPs certified for their promotion of sound forest management. Nonetheless, FSC species-specific standards for certification of Brazil nuts were developed in Bolivia (CFV, 2006), Brazil (C BMF, 2003), and Peru (CP-CFV, 2005). These standards generally follow FSCs 10 Principles, which address environmental, social, and economic aspects, while focu sing on additional criteria for NTFPs (SmartWood, 2002). ASCART in Madre de Dios, with 32 me mbers, was the first and only producers group in the region to achieve FSC certification for Brazil nuts in 2003 through the support of a Peru vian NGO, the Asociacin para la Conservacin de la Cuenca Amaznica (ACCA). The FSC standards for Brazil nuts in Madre de Dios were created through a cooperative effo rt between ACCA, the Asociacin of Extractivistas de Castaa de Madre de Dios (ASE CAMD), and Candela Per (CP-CFV, 2005). In addition to respect for national forestry laws FSC certification for Brazil nuts in Peru calls for worker safety precautions during ha rvest, protection of regeneration, limits on hunting, clean storage areas, and a detail ed management plan that includes a tree inventory and documentation of silvicultural practices and annual fruit production (CPCFV, 2005). To date, no produc ers in Pando have attempted to certify their products through this particular certification system, lar gely due to issues of land tenure insecurity (Pacheco and Cronkleton, 2008a; 2008b). In Brazil, the prim ary cases of FSC 126

PAGE 127

certification of Brazil nuts are the cooperative COMARU in the state of Amap and the Kayap indigenous group in Par state (Pinto et al., 2008). Methods Communities Sampled We evaluated Brazil nut management prac tices and income derived from nuts among certified and noncertified producers in 17 communities in the tri-national region of Pando, Bolivia (n=8), Acre, Brazil (n=5), and Madre de Dios, Peru (n=4) (Figure 4-1). In Pando, communities were chosen to represent differences in market access (river vs. road) and distance to major market center s, and were located both within (n=4) and outside (n=4) the Manuripi National Wildlife Re serve. In Acre, the five communities sampled differed in their access to roads and their distance to central markets. Four were located in the Chico Mendes Extractive Reserve and the fifth in the Chico Mendes Agro-Extractive Settlement Project (commonly known as Cachoeira). In Madre de Dios, three of the four communities sampled were located along the Interoceanic Highway. The fourth was comprised of producers who lived in the town of Puerto Maldonado, but had permission to enter the buffer zone of t he nationally-protected Tambopata Reserve to collect Brazil nuts during the harvest seas on. Most communities represented a mix of certified and noncertified producers, although in several remote communities, no producers sold certified nuts. From January 2006 to August 2007, we a ccompanied seasonal Brazil nut harvest and conducted structured interviews with 231 households (125 in Pando, 76 in Acre, and 29 in Madre de Dios). Of the 125 households in Pando, 36 sold certified Brazil nuts (29%). In Acre, 28 of the 76 households sold certified nuts (37%). In Madre de Dios, 15 of the 29 households sampled sold certifi ed nuts (52%). In sm all communities (<30 127

PAGE 128

families) in Pando and Acre, all available fa milies participated in the study; in larger communities (>30 families), participants were chosen randomly from lists of total households in each community. Variables Measured Management variables were based on local best management practices literature (e.g. Wadt et al., 2005; Card, 2000; 4-1) and observations made while participating in the 2006 Brazil nut harvest. Economic variable s included: (1) price per kilogram of Brazil nuts sold to middlemen versus certified cooperatives; (2) frequency and amount of outstanding debt from the 2005-06 harvest ; and, (3) frequency and amount of money advanced for the 2006-07 harvest. We also a sked certified nut producers to rank the benefits of certificati on on a 5-point Likert scale and free-list the perceived benefits and disadvantages of certification. Data Analysis Comparative management practices and associations with certification To understand basic quantitative differenc es between Brazil nut management in Pando, Acre, and Madre de Dios, we first generated descriptive statistics for management practices. We used a generalized linear model (binary logit) to test differences in practices between countries us ing a 95% Wald confidence interval. We then searched for relationships between cert ification and best management practices for each country separately through the use of chi-square tests. Odds ratios were calculated to measure the relations between the pres ence of certification and performance of management practices. 128

PAGE 129

Comparative economic variables and associations with certification We compared economic variables related to sale of nuts in Pando, Acre, and Madre de Dios and measured relationships with type of buyer (middleman versus certified cooperative) or cert ification in general. We firs t generated descriptive statistics for these variables and then searched for relationships with certification for each country separately through the use of chi-square te sts (for discrete variables) and generalized linear models at 95% Wald confidence interval s (for continuous variables). All analyses were performed using the software SPSS Statistics GradPack 17.0 (2008). Results Comparative Management Practices Comparative results between Brazil nut m anagement practices at the pre-harvest stage in Madre de Dios, Acre, and Pando rev ealed between-country differences (Figure 4-3). In Madre de Dios, a greater percentage of producers (p<0.001) had mapped their Brazil nut stands and develo ped management plans when co mpared to producers in Pando and Acre. In Madre de Dios, 79% ( 8.4 SE) of nut producers mapped their stands, compared with 41% (.9 SE) of pr oducers in Pando and 24% (.4 SE) in Acre. While these same 79% Peruvian producers developed nut management plans, no producers in Pando or Acre had such plans. Management for regeneration and to promote fruit production (i.e., silvicultural interventions) showed generally similar patte rns between the three countries (Figure 43). While more than two-thir d of producers in the three c ountries reported that they would not clear and burn agricultural plots in areas where there were Brazil nut seedlings (>1 m high), there was minimal enrichment planting or clearing around seedlings. Relatively few producers in the three countries planted seeds or seedlings 129

PAGE 130

(23% .9 SE in Pando, 21% .9 in Acre and 38% .5 in Madre de Dios) and more than half of these had planted less than 10 seedli ngs each. That said, there were a few cases in which more than 100 seedlings had been planted in agricultural areas and in individual Brazil nut collection outposts in t he forest. Also, relatively few producers liberated seedlings through clearing around t hem; 35% (.8 SE) in Pando, 26% (.7 SE) in Acre, and 42% (.9 SE) in Madre de Dios. Among those who performed this practice, the main reasons given for cleari ng around seedlings were to enhance growth, to protect them from fire, and to make them easier to find. The most common management practice overall in the three countries was vine cutting to enhance fruit production; almost all producers reported conducting this practice. The only statistically-significant between-country di fference among silvicultural practices was found in bleeding trees (p<0.001). This prac tice entails cutting into the inner bark until the red latex is released. While no producers interviewed in Madre de Dios bled adult Brazil nut trees, 32% (.8 SE) in Pando and 34% (.2 SE) in Acre reported bleeding trees that were either minimally or nonproductive to increase fruit yield. In examining post-harvest management practices, the majority of producers removed the fruit placental tissue and damaged nuts during collection after opening the fruits and before placing them in the collect ion sacks (Figure 4-3). That said, a lower number of producers removed the placental tissue in Ac re when compared with Madre de Dios and Pando (p<0.001). Also, a lower percentage of producers in Madre de Dios removed cut and damaged nuts when compared wit h Pando and Acre (p=0.046). Most producers in Pando (85%, .2 SE) transported their nuts from the forest immediately after collection, whereas fewer producers in Madre de Dos (62%, .2 SE) and Acre 130

PAGE 131

(33%, .5 SE) did so (p<0.001) Nearly two-thirds of produc ers in Acre (63%, .5 SE) and Madre de Dios (61%, .8 SE) were careful to keep stored nuts away from potential contaminants. This was higher (p=0.037) than t he 43% (.3 SE) in Pando who performed this practice. While not articulated as a best management pr actice in the trinational literature, the presence of storage units and drying stru ctures allow for post-harvest management. In our sample, more than one-third of produ cers in Pando (36%, .3 SE), Acre (43%, .7 SE), and Madre de Dios (39%, .4 SE) had their own stor age units. Of these producers, 18% (.5 SE) in Acre and 38% (.2 SE) in Madre de Dios also had access to separate drying facilities, whereas in Pando, nuts were dried in the same structures in which they were stored (p<0.001). Associations Between Brazil Nut Certification and Management In Madre de Dios alone, all certified produc ers, regardless of ce rtification type, had mapped their nut stands and dev eloped management plans. They were 1.8 times more likely to have conducted these practices t han noncertified Peruvian producers (p=0.004; Table 4-3). While no associations were observed between certification and mapped stands in Pando, in Acre certified producers we re 8.6 times more likely to have mapped their stand than noncertified Br azilian producers (p<0.001). In all three countries, production of certified Brazil nuts was not associated with most silvicultural practices (Table 4-3). T he main exception was t hat certified producers in Pando were 1.7 times more likely (p=0. 031) and in Acre were 2.1 times more likely (p=0.050) to clear around seedlings when co mpared to noncertifie d producers. In Madre de Dios, although non-intu itive, noncertified producers were 2.1 times more likely (p=0.009) to protect seedlings in agricultural clearing. 131

PAGE 132

The strongest associations between nut certification and management were observed in post-harvest practices specifically required of organic certification, such as drying nuts and storing them away from potential contaminants. Certified producers in Pando were 4.3 times more likely (p<0.001) and in Acre were 15.4 times more likely (p<0.001) to dry nuts than their noncertified c ounterparts. In Pando, certified producers were also 3.4 times more likely (p<0.001) to own their own storage unit and 3.8 times more likely (p<0.001) to keep nuts away from contaminants. This same pattern of positive associations between certification and these particu lar post-harvest practices was also evident in Acre where certified pr oducers were 6.4 times more likely (p<0.001) to own their own storage unit, 1.6 times more likely to separate nuts from contaminants (p=0.009), and infinitely more likely (p<0.001) to own their own dryer. No such associations were observed in the case of certified and noncertifi ed producers in Madre de Dios. Additionally, there was a significant association in all three countries between certification and awareness of aflatoxins as contaminants of Brazil nuts. Certified producers were 1.4 times more likely to be aware of aflatoxins in Pando (p=0.003), 1.8 times more likely in Acre (p=0.002), and 2.0 times more likely in Madre de Dios (p=0.002) than non-certified pr oducers. In Madre de Dios, all certified producers were aware of aflatoxins, independent of certification type. Incidence of hunting was also analyzed, since the FSC Brazil nut standards specifically mention that hunting should be limited in Braz il nut stands (CP-CFV, 2005). Overall, most nut producer s reported hunting during the nut harvest: 75% (.9 SE) in Pando; 78% (.8 SE); in Acre, and 78% (.2 SE) in Madre de Dios, and no differences were observed between cert ified and noncertified producers in each 132

PAGE 133

country. Even though the Peruvian sample size was low, we also noted no difference in reported hunting (p=0.609) between producers affiliated with FSC and those not affiliated. Income Derived From Organically-Cert ified and Non-Certified Brazil Nuts In Pando, producers who sold raw, unshe lled nuts to certified cooperatives as opposed to middlemen received nearly double the price for their sales (Figure 4-4). Of the 257 sales recorded in Pando, the mean (SE) price (USD) per kilogram for sale to cooperatives ($0.98, .02) wa s significantly higher (p<0.001) than the mean price of certified nuts to middlemen ($0.50, .01), a 96% difference. This additional income was generally received in two payments; a portion at the time of the sale, which was comparable to the price paid by middlem en, and a second payment after Brazil nuts were processed, which was received before the start of the s ubsequent harvest. In Acre, there was no statistical difference (p =0.123) for income gleaned from sale to middlemen when compared with sale of raw, unshelled, certified nuts to cooperatives (Figure 4-4). This can be attributed to admin istrative failures of CAPEB, which when it went into bankruptcy in 2007 resulted in most producers never receiving their promised second payment after nut processing. In Madre de Dios, the sale of organic shelled nuts to Candela Per through the Fair Tradecertified cooperative RONAP ($5.55) was significantly higher (p=0.012) than the price gleaned from sale to middlemen ($4.01) a 38% difference. Overall, the price earned fr om nuts in Madre de Dio s was much higher than in Pando or Acre, because Peruvian producers shelled their own Brazil nuts before sale to middlemen or cooperatives. In Pando, certified producers were 2.5 ti mes (p=0.037) less likely to have debt or have accepted an advance payment at the beginning of the harvest than non-certified 133

PAGE 134

producers, and 1.2 times more likely in Acre (p=0.037; Table 4-4). There were no differences between the amount of money owed as debt or taken out as an advance between certified and noncertified producers, with the exception of debt in Acre (p=0.034) where no certifi ed producers had debt from t he previous harvest. Producers Perceptions of Brazil Nut Certification There were differences in perceptions of certification among producers in Pando and Acre who sold certified Br azil nuts. In Pando, most certified producers interviewed perceived certification as moderately to highly beneficial (92%, n=27). The most commonly listed benefit was the better price gai ned from the sale of certified nuts (68%), followed by the health care benefits that cooperative members received through Fair Trade certification (16%). Some of the negative percepti ons included the need to pay transportation costs to the cooperatives ( 36%) and that certification was too much work and responsibility (36%). Conversely, in Acre, while 49% (n=14) found that certification was moderately to highly beneficia l, 22% felt that it had no effect, and 29% reported that their experience was somewhat to extremely negative. The most common complaint was that there was too much work involved in producing certified nuts (33%) and that the promised payment upon nut processing either a rrived late or not at all (34%). Poor cooperative adm inistration and lack of support for producers were also listed as negative aspects of nut certification in Acre (15%). That said, in Acre, some perceived benefits of certificat ion were better price (47% ), higher quality of nuts produced (35%), better transpor t (12%) and better management (6%). In Madre de Dios, half of the certified producers interview ed (n=15) felt that organic and Fair Trade certification was neither positive nor negative, and the other half felt that certification was moderately beneficial. The most comm only listed benefit was better price (67%) 134

PAGE 135

followed by higher nut quality (33%). Negative aspects in cluded too much work (50%) and complaints that certified companies cheated producers by not counting the full value of peeled nuts per bag (50%). The main complaint articulated by Peruvian producers specific to FSC certification was the lack of financial benefit given the tremendous amount of extra work required. Discussion and Conclusion Our overall objective was to understand if di fferent Brazil nut certification schemes were associated with different management and income generation outcomes for nut producers in Western Amazonia. We explored these relationships through a comparative study focused on best management practices, nut prices, and debt among producers in Pando, Bolivia, Acre, Brazil, and Madre de Dios, Peru. Organic and Fair Trade certification were clearly associated with better post-harvest practices and higher prices, while those producers certified by the FSC in Peru appeared to have adopted more practices related to pre-harvest planning and to a certain ext ent, individual tree health. While most producers commented on t he extra work involved in certification, regardless of certification type, certification was viewed most positively in Pando, where producers were able to glean financial and soci al benefits, and less positively in Madre de Dios and Acre where financia l benefits of certific ation were lower or non-existent. Certification and Best Brazil Nut Management Our data showed a positive association between certification and the pre-harvest practices in Madre de Dios in particular, and to a lesser extent, in Acre. Brazil nut management plans were much more advanced in Madre de Dios than in Pando or Acre. Peruvian law requires producers to invent ory their stands and create management plans for legal access to nut concessions, an d FSC certification ex plicitly requires 135

PAGE 136

compliance with national forestry laws. Impo rtantly, all certified producers in Madre de Dios had completed their Brazil nut mapping or management plans, and the noncertified producers who had not would eventually be requir ed to do so. In Brazil, while forestry legislation does not require nut management plans, the handful of mapping initiatives that occurred in Acre were undertaken as a first step towards FSC and organic certification, which likely explains the positive association observed between certification and mapped stands. In Pando, we did not observe any association between certification and mapped stands, mostly because mapping is an activity that occurs where community interest and outside technical support exis t. Initial mapping projects were primarily used to help communiti es visualize their traditional forest uses and negotiate resource conflicts (Cronklet on et al., 2008), not necessarily to pursue certification. Although legally required within the Bolivian technical norms for Brazil nut production (Ministrio de Desarrollo Sostenible, 2006), of the 159 extractive communities in Pando, to date only seven had nut management plans (W. Suarez, pers. comm. ). Elaboration of a managem ent plan is an important component of sustainable Brazil nut management and a basis for FSC certif ication, but most communities lack the resources or capacity to develop these plans on their own. While we observed few associations between certification and silvicultural practices, there was a higher incidence of clearing around seedlings by certified producers in Pando and Acre. This may be the result of cooperative messages that promote seedling liberation to enhance growth. That said, enrichment planting is also promoted as a way to increase Brazil nut densities, and there was no association between certification and performance of this practice. 136

PAGE 137

Our results clearly showed that certific ation influenced harvest and post-harvest practices that are directly related to product quality. This makes sense as guidelines for an aflatoxin-free product are much more strict when compared to pre-harvest and silvicultural practices, which though presented as part of the best management package are not requirements fo r organic and Fair Trade cert ification the types of certification that most of our study participants had acquir ed. Harvest practices, such as removing the placental tissue and rotten an d damaged nuts, were not associated with certification in Pando, likely because almost everyone performs these practices during the harvest. While producers in Pando tr ansported their nuts from the forest immediately after collection to avoid the risk of nut theft (Duchelle et al., in review), the positive association between certification an d immediate transport in Acre was likely because certified producers hav e learned that immediate tr ansport is an important way to deter infection by Aspergillus (Souza lvares et al., 2009) even though this practice may be less efficient (See Chapter 3). Certified producers in all three places were much more aware of aflatoxins when com pared to noncertified producers, which demonstrates advancement in socializing this concept towards promotion of a healthy product. Economic Benefits of Certification The strengthening of producer organization s is considered an important way to facilitate small-scale producers participation in forest mark ets (Sherr et al., 2004) and reduce transaction costs (Mayers and Vermeul en, 2002). Our data showed that organic and Fair Trade certification for Brazil nuts through cooperatives can clearly provide producers with socioeconomic benef its. In addition to the hi gher price for organic nuts, the Fair Trade price premium received by the cooperatives had important social 137

PAGE 138

benefits. Price premiums attained from Fair Trade certification were to be invested back into the cooperative. For inst ance, in the case of COINACA PA in Pando, the Fair Trade premium was used to provide members wit h harvest materials (machetes, sacks), health care benefits, and training opportunities. In Peru, the premium covered member costs of producing annual operational plans for harvest in Br azil nut concessions. The economic success of these certification systems, however, was highly dependent on the cooperatives themselves. As we observed in the case of Acre, if a cooperative is not administratively sound, affiliated producers can easily lose the benefits associate with these specialized markets and become disenchant ed with the certificatio n process. There were some downsides to affiliation with even successful certified cooperatives. For instance, COINACAPA has a norm that members were not allowed to manage and harvest commercial timber on t heir landholdings. While multiple-use forest management is not officially barr ed under the organic standards of Bolicert, according to COINACAPA, Bolicert express ed concern about potentia l contamination of nuts through mechanized timber extraction. This creates a dilemma: while producers may want to engage in integrat ed management of timber and Brazil nuts, if they chose to do so, they risk losing organic and Fair Trade benefits associated with sale of certified nuts. One of the main justificati ons for production of Fair Tra de-certified Brazil nuts is helping producers break the cycle of debt wit h middlemen. While middlemen are often essential actors in NTFP production chains especially for remote communities with limited means of transport to markets (Padoch, 1992), financial dependence on these buyers can be problematic. In the Western Amazon, middlem en arrive in communities 138

PAGE 139

at the end of the dry season when there are few other in come-generating opportunities. Producers who are short on cash will incur debt to stock up on basic food and clothing items, and oftentimes purchase larger items such as motorcycles, which they will then pay back during the nut harvest at a predetermined price/volume set by the buyer. If Brazil nut production is less than expected, producers can easily find themselves with outstanding debt. In one community in Boliv ia, a producer complained of a debt owed from the 2000-01 harvest when pr ices were at an all-time low. In 2007, the producer was still receiving the 2000-01 price for nuts so ld to this buyer even though current market prices were much higher. The benefit of affiliation with certified cooperatives was that the second payment from processi ng nuts from the previous harvest also arrived at the end of the dry season when pr oducers were most vulnerable. While our data from Pando and Acre show ed that certification ind eed helped some producers avoid debt and money advances, the amount owed was no different between indebted certified and noncertified producers. This finding was largely because certified producers often sold a portion of their Braz il nuts to middlemen, while maintaining the agreed-upon portion for sale to cooperatives. This demonstrates the importance of middlemen in the production chain as producers minimize risk by diversifying their sales. Partnerships for Certification are Key Multiple actor partnerships are essent ial to the success of community-based certification projects. Our st udy highlights the im portant role of the different actors involved in supporting community Brazil nut certification in Western Amazonia, including cooperatives, companies, government and non-governmental organizations, and national and internat ional donors. These organization s promoted information sharing 139

PAGE 140

and facilitated producers access to specialized markets through: trainings in best management practices and administrative matters, nut stand inventories and creation of management plans, construction of nut st orage units and drying structures, and communication with accredited cert ifying bodies. In studies of other certified products, such as coffee, the strong NGO base (Rayno lds et al., 2007) and technical farmer-tofarmer training (Bray et al., 2002) have been ess ential to marketing success. Also, the role of producer organizations, NGO invo lvement and pressure, and governments promotion of both the supply and demand of certified products are considered essential to producers uptake of forest certification, highlighting the importanc e of involvement by a broader scope of actors (Carrera et al., 2004; Auld et al., 2008). That said, different actors will have their own reasons for engagement in product-oriented forestry partnerships; while communities may seek livelihood benefits, companies may wish to profit from niche markets, and NGOs and donors may have specific conservation or development objectives (Ros-Tonen et al., 2008) These interests may clash, especially when certification projects are not successful. In the case of Acre we can clearly see the negative effect that bankruptcy of the main certified c ooperative had on affiliated nut producers. They did not financially benefit from certification as promised, and many lost trust in the broader network of support organizations, which may preclude their future involvement in certification initiatives. Also RONAP, the Peruvian producers cooperative, is currently seeking i ndependence from Candela Per, because of perceived economic losses from the way in which Candela determines shelled nut weights. While cooper atives in Pando continue to benefit from extensive donor support, an end to that funding for financial or ideol ogical reasons would certainly have negative 140

PAGE 141

implications. Partnerships for certification have enabled Brazil nut producers in Western Amazonia to access these specialized market s; however, their future success will be based on encouraging fairly negotiated objectives and prices, empowering nut producers and promoting ecological su stainability of the resource. Benefits and Challenges Associated with Three Certification Schemes There are unique benefits and challenges associated wi th the three different certification schemes exam ined in this paper. Organic and Fair Trade certification clearly had some success as market-based me chanisms. Additionally, we can argue that these certification schem es acted as regulatory mechanisms by influencing postharvest management practices, as signal me chanisms of cooperatives administrative successes or failures, and as learning mechanisms that enhanced producers knowledge of product quality. Th e most important challenges for the future of organic and Fair Trade nut certificati on are continued supply of a high-quality product, sound cooperative administration, and c ontinued internati onal demand. Unfortunately, forest management certific ation is limited as a market-based mechanism by lack of consumer demand, even in the seemingly clear case of Brazil nut where regional species-specific FSC standards have been developed. This absence of monetary reward is the main reason that FS C certification of Brazil nuts has not yet taken off in Pando or Acre, and that the one ca se of FSC certification in Madre de Dios ended after five years. Such inconsisten cy between supply and demand of certified products is considered the main limiting fa ctor to the growth of these specialized markets (Overdevest, 2004; RosTonen, 2005; Wilsey, 2008). Of the three certification types, FSC certification is the only one that specifically addr esses pre-harvest, silvicultural and post-harvest management practices in its standards, which seemed to 141

PAGE 142

be present among certified producers in Peru. Although pre-harvest mapping and management plans are time consuming, these practices allow for compliance with federal regulations, greater collection e fficiency in reaching trees once maps are generated and evaluated, and identif ication of previously ov erlooked fruit producing trees. Also, almost no Pe ruvian producers bled their Brazil nut trees. Even though wounding can stimulate fruit production in t he short term, this practice can negatively impact tree health over time with wounds se rving as invasive routes for pathogens (Kramer and Kozlowski, 1979). Given the benef its associated with FSC certification, as Taylor (2005b) suggests, linking FSC with Fair Trade certification (and we argue also organic) may be a way to promote the FSC label, and its added management benefits, within a certification system that is better est ablished for small produc ers. FSC, IFOAM, and Fairtrade Labelling Organizations Intern ational are already unified under the international umbrella ISEAL (International Social and Envi ronmental Accreditation and Labeling) (Auld et al., 2008), which could fa cilitate the creation of complementary certification schemes. Organic, Fair Trade and FSC certification may have benefits for tropical forests and the people who live in them. While our study focused on Brazil nuts, the findings are relevant to products such as cac ao, coffee, chicle, and other consumable nontimber forest products. Finding ways to overcome the challenges associated with certification and maximize the management and economic benefits for producers may help promote tropical forest conser vation and livelihood development. 142

PAGE 143

Credit: N. Hoyos Figure. 4-1 Map of sampled communities. 143

PAGE 144

144 Figure 4-2. Brazil nut producer s at the door of their storage units in Bolivia (top left) and Brazil (bottom left). Separate structures for drying nuts in Peru (top right) and Brazil (bottom right).

PAGE 145

Figure 4-3. Percent and SE of Brazil nut producers practicing best management pr actices in Pando, Acre and Madre de Dios. Significance of between-country comparisons (***p<0.001, **p<0.05); actual p-values given in text. 145

PAGE 146

0.00 0.20 0.40 0.60 0.80 1.00 1.20Sold to middlemanSold to certified cooperativeUS $ / Kg raw nutsPando, Bolivia 0.00 0.20 0.40 0.60 0.80 1.00 1.20Sold to middlemanSold to certified cooperativeUS $ / Kg raw nutsAcre, Brazil 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00Sold to middlemanSold to certified cooperativeUS $/ Kg shelled nutsMadre de Dios, Peru Figure.4-4. Producer prices received when selling Brazil nuts to middlemen versus certified cooperatives in 2007. Prices are based on number of sales recorded (Pando, n=257; Acre, n=98; Madre de Dios, n=37). Cooperative prices were significantly higher than those offer ed by middlemen in Pando and Madre de Dios, but not in Acre (p<0.05). Pric e per kilogram was higher in Madre de Dios, because weights were for shell ed nuts versus raw nuts sold in Pando and Acre. 146

PAGE 147

Table 4-1. Best management practices for Brazil nut as outlined in regional extension literat ure, along with justifications and supporting research for the performance of each practice. Best management practice Justification Supporting research Pre-harvest practices Map productive trees Plan harvest, negotiate conflicts Cronkleton et al., 2008 Develop management plan Promote sustainable management, monitor annual production, comply with national forestry legislation and certification standards Silvicultural practices Enrichment plant Increase density and number of trees Kainer et al., 1998; Kainer et al., 1999; Zuidema, 2003 Protect seedlings when clearing agric. plots Maintain regeneration in swidden fallows Kainer et al., 1998; Cotta et al., 2008 Liberate seedlings through clearing Enhance seedling growth Avoid bleeding trees Potential tree mortality in long-term Kramer and Kozlowski, 1978 Cut vines Increase fruit production Zuidema, 2003; Kainer et al., 2006, 2007 Harvest and post-harvest practices Remove fruit placental tissue Prevent infection by Aspergillus Souza lvares et al. 2009 Remove cut/damaged nuts Prevent infection by Aspergillus Souza lvares et al. 2009 Transport same day as collection Prevent infection by Aspergillus Leite 2008, Souza lvares et al. 2009 Dry nuts Prevent infection by Aspergillus Labuza 1980, Souza lvares et al. 2009 Separate stored nuts from contaminants Prevent infection by Aspergillus Souza lvares et al. 2009 147

PAGE 148

Table 4-2. Brazil nut certificat ion initiatives in the tri-national region of We stern Amazonia. Dashes signify that although standards were created, no companies or cooperatives were certified. Place Certification 1st Year Attained Accredited certifier Certified nut companies Certified nut cooperatives Supporting organizations Status in 2009 Pando, BOLIVIA Organic 1995 Bolicert, IMO Control LA El Campesino, Tahuamanu, Manutata COINACAPA, ACERM (2005) CARE, Fundacion Puma, Herencia New cooperative (ACEBA) certified Product Quality 2003 IBNORCA Tahuamanu ----Maintained Fairtrade (FLO) 2001 Trans Fair International El Campesino COINACAPA, ACERM (2005) CARE, Fundacion Puma, Herencia New cooperative (ACEBA) certified FSC 2003 SmartWood ----CFV, PROMAB No advance Acre, BRAZIL Organic 2005 IBD --CAPEB WWF, EMBRAPA, Ecoamazon, SEBRAE, SEAPROF, CTA COOPERACRE (in process) Fair Trade (FLO) 2005 FLO --CAPEB WWF, EMBRAPA, Ecoamazon, SEBRAE, SEAPROF, CTA COOPERACRE (in process) FSC 2003 IMAFLORA ----WWF, FASE, IMAZON, among others No advance Madre de Dios, PERU Organic 2001 IMO Candela Per, Fast Trade S.A. (2005) RONAP (via Candela Per), ASCART (via Fast Trade S.A.) WWF, ACCA, CAMDE Peru, FONDEBOSQUE New producers associations certified Fair Trade 2004 FLO --RONAP, ASCART (2005) Candela Per Maintained FSC 2003 SmartWood --ASCART ACCA, Candela Per, ASECAMD Ended in 2008 148

PAGE 149

Table 4-3. Odds ratios (OR) and p-values from Pearson Chisquare tests of association be tween selling certified nuts and best management practices. Pando, Bolivia Acre, Brazil Madre de Dios, Peru Variable N OR p-value N OR p-value N OR p-value Pre-harvest practices Mapped stand 122 0.863 0.555 76 8.571 <0.001 29 1.75 0.004 Management plan ------29 1.75 0.004 Silvicultural practices Enrichment plant 122 1.405 0.331 76 1.333 0.519 29 1.633 0.316 Protect seedlings in agric. clearing 66 1.044 0.793 61 1.082 0.634 26 0.468 0.009 Liberate seedlings via clearing 121 1.700 0.031 76 2.095 0.050 26 0.80 0.781 Avoid bleeding trees 106 1.226 0.159 67 1.169 0.395 20 1.000 -Cut vines 122 1.041 0.491 76 1.067 0.177 29 2.5 0.129 Harvest and post-harvest practices Remove placental tissue 123 1.020 0.616 76 1.322 0.011 28 1.300 0.049 Remove damaged nuts 123 1.004 0.929 76 0.985 0.696 27 1.333 0.04 Transport same day 122 1.080 0.311 76 2.571 0.003 29 1.467 0.196 Dry nuts 122 4.330 <0.001 76 15.43 <0.001 29 0.560 0.089 Separate nuts from contaminants 117 3.789 <0.001 76 1.577 0.009 28 0.770 0.390 Other Own storage unit 121 3.413 <0.001 76 6.367 <0.001 28 1.040 0.934 Own dryer ---76 infinite <0.001 29 0.778 0.597 Knowledge of aflatoxins 123 1.356 0.003 76 1.792 0.002 29 2.000 0.002 149

PAGE 150

150 Table 4-4. Odds ratio (OR) and p-value from Pearson Chi-square test of association between nut certification and no debt or payments in advance. Correlations between certific ation and amount of outstandi ng debt from the 2005-06 harvest and amount of advanc ed payments received before the 2006-07 harvest. Pando, Bolivia Acre, Brazil Madre de Dios, Peru Variable N Mean (SE) OR p-value N Mean (SE) OR p-value N Mean (SE) OR p-value No debt or advance payment Non-certified 79 0.10 (0.03) 46 0.46 (0.07) 12 0.50 (0.15) Certified 36 0.25 (0.07) 2.469 0.037 17 0.53 (0.13) 1.160 0.037 8 0.25 (0.16) 0.500 0.264 Amount debt (2005-06) Non-certified 78 310 (65) -0.553 46 239 (58) -0.034 12 124 (74) -0.414 Certified 35 239 (97) -17 0 (96) -7 24 (97) -Amount advance (2006-07) Non-certified 78 678 (145) -0.74 19 399 (102) -0.483 6 527 (361) -0.22 Certified 36 763 (213) -8 267 (158) -8 1113 (313) -

PAGE 151

CHAPTER 5 CONCLUSION Taken together, the three papers in this dissertation allow for a comparative understanding of the conservation and liv elihood benefits associated with communitybased forest management in Western Amazonia. The first paper demonstrated extremely high levels of forest income dependency and minimal deforestation in extractive communities along roads in the tr i-national frontier region of Madre de Dios, Peru, Acre, Brazil, and Pando, Bolivia. That said, there was a def orestation trend from 1986 to 2005 in these communities, with most deforestation having occurred in Acre, followed by Madre de Dios and Pando. While this general pattern mirrored the findings of a region-wide study for the same time period (Southworth et al., in preparation), there was much less deforestation in the extractive communities in Acre when compared with the surrounding landscape (1986) There was also less deforestation in the final time period (2000) in extractive commu nities in Madre de Dios when compared with the region-wide study. Fragmentation and close proximit y to roads helped explain observed deforestation (via satellite) in a ll countries in the final time period (2000 2005), and economic variables were the stronges t predictors of repor ted forest clearing at the household level (via surveys) from 200 2 to 2007. For instance, income derived from agricultural sources was positivel y correlated with forest clearing among households in Pando (crops and livestock) and Madre de Dios (livestock), and higher Brazil nut income predicted less forest clear ed. In Acre, larger households that had received more government aid reported cleari ng more forest, with some evidence that higher value livestock assets also predicted forest clearing. 151

PAGE 152

The second paper highlighted the lack of pr operty rights security in Pando, Bolivia when compared to Acre, Brazil, shown by a much greater incidence of reported Brazil nut thefts in Pando than in Acre. The lack of property rights security observed in Pando was partially due to a community titling proce ss that disregarded tradi tional forest use. Also settlement patterns in Pando, where households were more clustered together, created a situation where peopl e were disconnected from th eir forested landholdings except during the Brazil nut harvest. This lack of physi cal connection combined with a product that was in high demand set people up fo r conflictive situations during the nut harvest. This paper drew attention to parti cipatory mapping as a potentially important tool for communities in dealing with tenure conflict, visualizing traditional forest use systems, and leveraging integrat ion of traditional practices into formal land titling processes and decision-making. The third and final paper demonstrated t hat organic, Fair Trade, and FSC Brazil nut certification schemes were associat ed with different management and income generation outcomes for nut producers in We stern Amazonia. Organic and Fair Trade certification was clearly linked to better pos t-harvest practices and higher prices, while producers certified by the FSC in Peru appeared to have adopted better pre-harvest and silvicultural management practices. Whil e most producers commented on the extra work involved in certification, regardless of certification type, certification was viewed most positively in Pando, where producer s were able to glean financial and social benefits, and less positively in Madre de Dios and Acre where financial benefits of certification were lower or non-existent. Importantly, partnerships with cooperatives, 152

PAGE 153

government and non-governmental organizations, and donors were essential components of these cert ification schemes. Significance : As trends continue to legally designate forests for use by local communities, it is critical to understand their role in tropical forest conservation and address the factors that prom ote or inhibit forest-based livelihood development. In addition to its broader academic significance, the results of this dissertation can inform regional policy and management of Brazil-nut rich forests in the rapidly-changing trinational frontier of We stern Amazonia. This dissertation highlights the impressive role of forests, in general, and Brazil nuts, in particular, in local livelihood systems of extractive communities in the MAP region. In Chapter 2, we observed that forest income comprised approximately 40-45% of the total cash and subsistence income in Brazil nut-producing communities along roads in all three countries. In Chapt er 3, where both roadand river-accessible communities in Pando were incl uded, this figure exceeded 60%. Preliminary results of the CIFOR PEN study revealed that households in Pando had the highest forest income share among 20 sites in tropical countries worldwide (CIFOR PEN, unpublished data), which was largely based on the contribution Braz il nuts. The finding that greater income derived from Brazil nuts predicted less forest clearing in Pando and Madre de Dios, also underscores the role of this NTFP in pr omoting tropical forest conservation. Importantly, the governments of Peru, Brazil, and Bolivia have already recognized the value of this NTFP through legislation that prohi bits the felling of Braz il nut trees. In the face of massive infrastructural development, however, it is also important to consider how to maintain the forests standing around them. 153

PAGE 154

Given the tremendous contribution of Brazil nut s to rural livelihoods, it is essential to ensure communities access to this resource. Both the creation of the Chico Mendes Extractive Reserve in Acre and Brazil nut c oncessions in Madre de Dios were important advances toward securing property rights for Brazil nut pr oducers within the Interoceanic highway corridor. While we did not examine incidence of nut thefts in Madre de Dios, the situation there before implem entation of the concessions has been compared to what we observed in Pando. Legal tenure security through creation of the concessions, complemented by Brazil nut mapping and creation of management plans by local NGOs, helped secure the resour ce rights of Peruvian nut producers (L.M. Velarde, pers. comm.). Although Bolivia has advanced in formalizing property rights for forest-based communities, the climate of insecurity in Pando combined with an extremely high dependence on Braz il nuts, set people up for conflictive situations during the harvest and left them extr emely vulnerable to the negative effects of a stolen product. It is also important to increase the value of Brazil nut-rich forests for communities to promote their conservation. The findings in Chapter 4 demonstrate that organic and Fair Trade certification of Brazil nuts are promising strategies for encouraging better post-harvest management, increasing the in come gleaned from Brazil nuts, and alleviating dependence on middlemen. The mo st important challenges associated with these certification schemes are conti nued supply of a high-quality product, sound cooperative administration, s upport from industry partners, and continued international demand. Since adoption of FSC certification is limited by lack of consumer demand, 154

PAGE 155

pursuit of FSC certification in conjunction with an organic or Fair Trade label may be a way to maximize the distinct environmental and economic benefits of these schemes. Although Acre has lagged behind Pando and Madre de Dios in Brazil nut certification, Acres Forest Government is an important regional m odel for investment in other aspects of community-based forest management, which has included revitalizing the rubber economy (Kainer et al., 2003), prom oting FSC certificat ion for smallholders (Humphries and Kainer, 2006), and experim enting with payments for environmental services (Gomes et al., 2008; Bartels, 2009; SEMA, 2009). Additionally, the Brazilian government recently approved a new federal po licy to establish minimum prices for key Amazonian NTFPs, including Brazil nuts (M DA/MMA/MDS, 2009). Despite certain policy barriers and cooperative norms that currently discourage integrated timber and Brazil nut management throughout the region, reduced-impact logging (RIL) can be compatible with conservati on goals and management of NTFP s (Putz et al., 2001; Garca-Fernndez et al., 2008; Guariguata et al., 2008). An initial study in Pando showed minimal logging damage to Brazil nut trees in certified concessions, but also recommended including pre-harvest marking of Brazil nut trees in RIL guidelines (Guariguata et al. 2009). In a separate study at the community scale in Pando, observations of increased Brazil nut regener ation along logging roads and in tree fall gaps, likely due to greater light availability are being tested. Multiple-use forest management is an important strategy for diversifying forest-based livelihoods. Finally, this dissertation highlights the importance of partner ships with multiple actors in supporting community-based fo rest management. These actors include cooperatives, industries, government and nongovernmental organizations, research 155

PAGE 156

156 organizations, and national and in ternational donors. In P ando, research organizations and NGOs played an important supporting role in smoothing propert y rights transitions in community-managed forests thr ough participatory mapping of Brazil nut stands. In all three countries, certification would not hav e been possible without the extensive support of governmental agencies, NGOs and donors, as seen by the varying degrees of success. In conclusion, this dissertation shows that the future of community-based forest management in Wester n Amazonia will likely depend on secure property rights for extractive communities, access to specialized markets, and partnerships with a diversity of actors to bolster forestbased livelihoods and promote cons ervation of Brazil nut-rich forests.

PAGE 157

APPENDIX A ANNUAL VILLAGE SURVEY 1 (PORTUGUESE) Levantamento da vila/comunidade 1 (V1) Nota: Veja o guia tcnico para obter in formao apropriada e entrevistados/informantes para as vrias questes de levantamento para a comunidade. Informao de controle Tarefas Data(s) Quem fez? Est OK? Se no, faa comentrios Encontros com autoridades Encontros com grupos de interesse/comunitrios Outras entrevistas Verificar o questionrio Codificar o questionrio Digitalizao de dados Verificao e aprovao da digitalizao de dados A. Variveis climticas e geogrficas 1. Qual o nome da vila/comunidade? 1. (nome) 2. (comunidade ##) 2. Quais so as coordenadas de GPS no centro da vila ou comunidade? (formato UTM) 3. Qual a latitude da comunidade? graus 1. Qual a longitude da comunidade? graus 2. Qual a altitude (metros acima do nvel do mar) da comunidade? metros 3. Qual tem sido a precipitao mdia anua l (mm/ano) no dist rito durante os ltimos 20 anos (ou menos, veja os guias)? ms/Ano 4. Qual o coeficiente de variao da precipitao nos ltimos 20 anos? (Nota: Preencher se os dados estiverem prontamente disponveis) B. Demografia 1. Em que ano a vila/comunidade foi criada/estabelecida? 2. Qual o tamanho actual da populao da vila/comunidade? habitantes/pessoas 3. Quantas famlias vivem actualmente na vila/comunidade? famlias 4. Qual foi o tamanho total da populao na vila/comunidade 10 anos atrs? habitantes/pessoas 5. Quantas famlias viveram na vila/comunidade 10 anos atrs? famlias 6. Quantos habitantes (aprox.) vivem na comunidade agora vindos de outras aldeais nos ltimos 10 anos (imigrao)? habitantes/pessoas 7. Quantas pessoas (aprox.) deixaram a comunidade nos ltimos 10 anos (emigrao)? habitantes/pessoas 5. Quantos grupos diferentes (grupos tnicos, tribos ou castas) vivem na comunidade? C. Infra-estrutura 1. Quantas famlias (aprox.) na vila/comunidade tm acesso a electricidade (de fornecedores pblicos ou privados)? Famlias 2. Quantas famlias (aprox.) na vila/comunidade tm acesso a (= utilizam) gua canalizada? Famlias 3. Quantas famlias (aprox.) tem acesso ao crdito formal (bancos governamentais ou privados operando na vila/comunidade)? Famlias 157

PAGE 158

4. H instituies de crdito informal para poupanas (clubes de poupana) e pessoas que emprestarem dinheiro na vila? (1-0) 5. H algum centro de sade na vila/comunidade? (1-0) 6. A vila/comunidade possui no mnimo uma estrada que poder ser utilizada por carros durante todas as estaes do ano? Se sim veja pergunta 8. (1-0) 7. Se No: Qual a distncia em quilme tros para a estrada prxima utilizvel durante todas as estaes (todo o ano)? km 8. H algum rio dentro dos limites da vila/comunidade navegvel durante todas as estaes? Se sim, veja pergunta 10. (1-0) 6. Se no : qual a distncia ao rio mais prximo navegvel durante todas as estaes do ano? km 7. Qual a distncia do centro da vila/comunidade para o prximo (em km e minutos pelos meios de transporte mais comuns) 1. km 2. min 3. cdi go tran sport e 1. Mercado distrital 2. Maior mercado para bens de consumo 3. Mercado onde os produtos agrcolas so vendidos 1. Mercado onde os produtos florestais so vendidos D. cobertura/uso da terra e florestal 1. Categorias de terra na vila/comunidade/comunidade ( rea aproximada em hectares) (Amaznia: 1ha=4 tarefas; 2ha= 1 alqueire). Nota: Veja o guia tcnico para a definio de categorias de terra e propriedade. Categoria de terra 1. Total rea (h) Propriedade (ha) 2. Estado 3. Comunidade 4. Privada 5. Acesso livre ( de facto ) Floresta: 1. Floresta Natural castanhal) Floresta Natural (inundado) 2. Floresta Manejada 3. Plantaes Terra Agrcola: 4. reas de cultivo 2. Pasto (natural ou plantado) 3. Sistemas agroflorestais 4. Silvipastoreio 5. Pousio/Capoeira Outras categorias de terra: 6. Arbustos/Cerrado/Campina 7. Capinzal 8. reas Residncias, infra-estrutura 9. Zonas de Pntanos/Gap/Vargem 10 Outro, especifique: 11. Total de terra 158

PAGE 159

2. Quais so os principais tipos florestais, usurios e produtos na vila/comunidade? Nota: O objetivo ligar tipos florestais, usurios e produtos. Veja o guia tcnico para detalhes. Nota: A rea total florestal dever ser igual a indicada na tabela acima. 1. Tipo florestal (cdigofloresta) 2. Propriedade (cdigo-terra) 3. rea aprox. (ha) Usurios principais1) (max. 3) Produtos principais (max. 3) (cdigo-produto) 4.Ordem1 5.Ordem2 6.Ordem3 7.Ordem1 8.Ordem2 9.Ordem3 1) Por usurios principais so aqueles que tm adquirido o valor mais alto dos produtos florestais (subsistncia e dinheiro) a partir de um dado tipo florestal nos ltimos 12 meses. Cdigos: Seleccione o mais apropriado entre os seguintes grupos (alguns se sobrepem): 1 = Comunitrios que so membros do GUF; 2 = Comunitrios que no so membros do GUF; 3 = Usurios de subsistncia nas comunidades; 4 = Usurios comerciais de pequena escala na vila/comunidade; 5 = Usurios comerciais de grande escala na vila/comunidade; 6 = Usurios de subsistncia de fora da vila/comunidade; 7 = Usurios comerciais de pequena escala de fora da vila/comunidade; 8 = Usurios comerciais de grande escala de fora da vila/comunidade; 9 = Outros, especifique: 3. A comunidade pratica alguma forma activ a ou deliberada de manejo florestal? Tipo de manejo Cdigo1) 1. Plantio de rvores 2. Abate/corte/derrubada de rvores no desejadas (competidoras) 3. Proteo de determinadas rvores nas flores tas (grupos de) para promover a regenerao natural dessas espcies 4. Proteo de reas florestais para servios am bienteis particulares, como bacia hidrogrfica 5. Estabelecer direitos de uso claros para um nmero limitado de pessoas para produtos especficos (Por exemplo rvores melferas) 9. Outras, especifique 1) Cdigos: 0=no, no por completo; 1=Sim, mas somente para certo limite; 2=sim,eles so comuns. E. Base de recursos Florestais Nota: As perguntas devem ser feitas num encontro na comunidade em entrevistas de grupo para cada categoria (i.e. coluna por coluna, e no linha por linha). 1. Lenha ou carvo 2. Madeira 3. Alimentos da floresta 4. Medica -mentos da floresta 5. Pastagem florestal 6. Outros1) 1. Qual o Produto Mais Importante (PMI) para o bem estar das pessoas na vila/comunidade (nesta categoria)? 2) (nome) 2. (cdigo-produto) 3. Como que a disponibilidade dos produtos mais importantes mudou nos ltimos 5 anos? Cdigo: 1= diminuiu; 2= Constante; 3= aumentou 159

PAGE 160

4. Se a disponibilidade de PMI nesta categoria diminuiu quais so as razes? Por favor ordene os motivos mais importantes, max. 3 (Deixe o resto em branco). Razes Ordene 1-3 Ordene 1-3 Ordene 1-3 Ordene 1-3 Ordene 1-3 Ordene 1-3 1. Reduo da rea florestal devido a abertura de roados em pequena escala para agricultura 2. Reduo da rea florestal devido a projetos de grande escala (plantaes, novos assentamentos, etc.) 3. Reduo da rea florestal devido a compra de terra por pessoas no locais e a restrio de acesso 4. Aumento no uso de PMI porque as pessoas locais (comunitrios) coletam mais 5. Aumento no uso de PMI porque as pessoas de outras comunidades coletam mais 6. Restries de uso pelo governo central ou provincial (p. ex., para conservao florestal) 7. Restries locais de uso (p.ex., regras comunitrias) 1. Mudanas climticas, p.ex., seca e menos precipitao 9. Outras, especifique: 5. Se a disponibilidade de PMI nesta categoria aumentou quais so as razes? Por favor ordene os motives mais importantes, max. 3. Razes Ordene 1-3 Ordene 1-3 Ordene 1-3 Ordene 1-3 Ordene 1-3 Ordene 1-3 1. Menos desmatamento florestal para agricultura (incluindo a criao de animais) 2. Menos pessoas locais (comunitrios) coletando menos 3. Menos pessoas de outras comunidades coletando menos 4. Uso reduzido de usurios de grande escala comercial/projetos 5. Mudanas no manejo das florestas 6. Mudanas climticas, p. ex.., mais chuva 2. Outras, especifique: 6. Qual ser o mais importante aumento de benefcios (uso ou rendimentos) dos PMI? Por favor ordene as razes mais importantes, max. 3. Aco Ordene 1-3 Ordene 1-3 Ordene 1-3 Ordene 1-3 Ordene 1-3 Ordene 1-3 1. Melhor acesso a floresta /PMI, i.e., mais direitos para os comunitrios 2. Melhor proteo das florestas /PMI (evitar uso excessivo) 3. Melhores habilidades e conhecimentos sobre como colectar e usar 160

PAGE 161

4. Melhor acesso a credito/capital e equipamento/tecnologia 5. Melhor acesso aos mercados e reduzido risco de baixa de preo 9. Outras, especifique: 1) Selecione o produto mais importante da vila/comunidade que no cai dentro de qualquer das cinco categorias. 2) Muito importante definido como o mais importante para o bem estar na vila/comunidade, quer seja atravs do uso domstico ou atravs da venda para dinheiro ou ambas. F. Instituies florestais Nota: As perguntas devem ser feitas nos encontros locais ou grupos de interesse para cada categoria (i.e., coluna por coluna, e no linha por linha). Nota: O Produto Mais Importante (PMI) em cada ca tegoria dever ser idntico ao da tabela abaixo. 1. Lenha ou carvo 2. Madeira de lei ou outra madeira para uso 3. Alimen tos da floresta 4. Medica -mentos da floresta 5. Rao de animais vindo da florestas 6. Outros1) 1. Qual o produto mais importante (PMI) para o bem estar das pessoas na comunidade (nesta categoria)? (nome) 2. ( cdigo-produto) 3. Em que tipo de mata/floresta voc obtm o PMI? (cdigo-florestal) 4. Q uem dono desta mata/floresta? (cdigo-posse) 5. Na comunidade h regras locais/habituais que regulam a utilizao de PMI? Cdigos: 0=nenhum/muito pouco; 1=sim,mas vago/no claro; 2=sim, existem regras claras Se o cdigo for 0, dirija-se para 7. 6. Se sim : as regras locais so aplicadas/respeitadas pela populao da comunidade?1) 7. H regras governamentais de regulamentao do uso da floresta ? Cdigos: 0=nenhum/muito pouco; 1=sim, mas vago/no claro; 2=sim, existem regras claras Se o cdigo for 0, dirija-se para 9. 8. Se sim (cdigo 1 ou 2 acima): as regras governamentais so respeitadas pelos membros da comunidade?1) 9. Os comunitrios/moradores necessitam de autorizao/licena para explorar os PMI? Cdigos: 0=no; 1=sim, Usurios tem de informar as autoridades; 2=sim, necessrio permisso por escrito se cdigo 0, dirija-se para a prxima seco. 10 Se sim (cdigo 1 ou 2 acima): o usurio dever pagar pela permisso? (1-0) (1-0) (1-0) (1-0) (1-0) (1-0) 11 Se sim : quem d a autorizao/licena? Cdigos: 1=chefe da comunidade; 2=GUF; 3= 161

PAGE 162

funcionrio florestal (departamento florestal); 4=outro funcionrio governamental; 9=outro, especificar: 1) Cdigos: 0=no /muito pouco; 1=em certa medida por alguns grupos de comunitrios; 2=em certa medida por todos; 3=sim, mas somente por alguns grupos de comu nitrios; 4=sim, por todo s; 9=no existem regras particulares/especificas. G. Grupos de usurios florestais/grupos de interesse (GUF) (grupos de produtores) 1. Existncia de grupos de usurios (grupos de interesse) florestais (GUF). Nota: Veja guia tcnico para definio. 1. Quantos grupos de usurios florestais (GUF) existem na comunidade? 2. Informao sobre cada GUF (usar uma coluna por GUF). 1. GUF1 2. GUF2 3. GUF3 1. Quando que o grupo se formou? (ano) 2. Como que o grupo se formou? Cdigos: 1=iniciativa local; 2=initiativa de uma ONG; 3=iniciativa governamental, p. ex., Departamento florestal; 4=outra, especifique: 3. O principal objetivo de GUF est relacionado com o manejo de uma rea florestal particular ou produto(s) florestal em particular? Cdigos: 1=rea; 2=produto(s); 3=ambos 4. Se for o produto (cdigo 2 ou 3 acima), Qual o produto (principal)? (cdigo-produto) 5. Quantos membros fazem parte/participam do grupo? 6. Quantas vezes por ano o GUF tem encontros? 1. O grupo possui um plano de manejo escrito? (1-0) (1-0) (1-0) 2. Quais so as principais tarefas do GUF? Selecione quantas for apropriado: 1-0 cdigo 1. Estabelecer regras de uso (1-0) (1-0) (1-0) 2. Monitorar e fiscalizar (1-0) (1-0) (1-0) 3. Silvicultura & manejo (1-0) (1-0) (1-0) 4. Explorao de produtos florestais (1-0) (1-0) (1-0) 5. Venda de produtos florestais (1-0) (1-0) (1-0) 9. Outras, especificar: (1-0) (1-0) (1-0) 3. Algum projeto de desenvolvimento foi implementado na comunidade nos ltimos 5 anos usando os lucros ou retornos dos GUFs?Outra maneira de formular: Houve alguma bemfeitoria com unitria que veio dos lucros/retornos do GUF? (Dificilmente os benefcios sero entendidos como projeto) (1-0) (1-0) (1-0) 4. Algum na comunidade ter violado as regras do GUF nos ltimos 12 meses? se no, dirija-se para 14. (1-0) (1-0) (1-0) 5. Se sim : o GUF imps alguma penalizao para quem no cumpriu das regras? Se no, dirija-se para 14 (1-0) (1-0) (1-0) 6. Se sim : Qual foi o tipo de penalizao? Cdigos: 1=multa (pronto pagamento); 2=Devolver os produtos coletados; 3=trabalho (trabalho extra); 4=excluso do grupo; 9=outr o, especificar: 7. Quais so os grupos de usurios que mais frequentemente violaram as regras nos ltimos 5 anos? Cdigos: 1=membros do GUF; 2= no membros do GUF na comunidade; 3=pessoas de outras comunidades; 9=outros, especificar: 8. De um modo geral, na escala de 1-5 (1 altssimo, 5 mais baixo) quanto efetivamente poder dizer que os GUF esto assegurando o uso sustentvel e e a diviso justa dos benefcios da floresta? Nota: Qualquer GUF na comunidade dever ser discutido na narrativa da comunidade. 162

PAGE 163

APPENDIX B ANNUAL VILLAGE SURVEY 2 (PORTUGUESE) Levantamento da comunidade 2 (V2) Informao de controle Tarefas Data(s) Quem fez? Est OK? Se no, faa comentrios Encontros com autoridades Encontros com grupos de interesse/comunitrios Outras entrevistas Verificar o questionrio Codificar o questionrio Digitalizao de dados Verificao e aprovao da digitalizao de dados A. Variveis climticas e geogrficas 1. Qual o nome da comunidade? *(nome) (comunidade ##) 2. Qual foi a precipitao na comunida de nos ltimos 12 meses? mm/ano 3. Se os dados de precipitao no estiverem dis ponveis (pergunta 2): Como foi a precipitao nos ltimos 12 meses comparando com um ano normal (= Media dos ltimos 20 anos)? Cdigos: 1= Muito abaixo do normal (< 50 %); 2= Abaixo do normal (50-90%); 3= Normal (90110%); 4= Acima do normal (110-150%); 5= Muito acima do normal (> 150%) B. Risco 1. A comunidade sofreu alguma crise nos ltimos 12 meses? Cdigos: 0= No; 1=sim, crise moderada; 2=sim, crise severa 1. Enchente/Alagao e/ou excesso de chuva 2. Seca 2. Incndios/Queimadas (nas culturas/ florestais/capinzal etc.) 3. Ataque de bicho/Peste generalizada na poca da colheita/doena e/ou doena animal 4. Epidemias humanas (doenas) 5. Conflitos polticos /civis 6. Crises macro-econmicas 7. Refugiados ou imigrantes 8. Outras, especifique: C. Salrios e preos 1. Qual foi a diria tpica para mo-de-obra agrcola no treinada/casual para um adulto masculino/feminino durante as estaes de alta/baixa (escassez) na comunidade nos ltimos 12 meses? (Moeda nacional/dia) Homem Mulher poca de alta 1. 2. poca de baixa 3. 4. 2. Qual a principal alimentao na comunidade? (cdigo-produto) 3. Qual foi o preo por kg do principal alimento durante os ltimos 12 meses antes e depois da principal colheita agrcola? (Moeda nacional/kg) 1. Antes da colheita/safra 2. Depois da colheita/safra 1. Qual o preo de um hectare de terra com boa produo agrcola na comunidade (i.e., no degradada, no muito inclinada, e prpria para culturas comuns, e dentro de 1km da estrada principal ou assentamento) (Lc$/hectare) 163

PAGE 164

D. Servios florestais 1. A comunidade (como comunidade ou indivduos na comunidade) recebeu qualquer benefcio direto (em numerrio (?no sei: talvez em forma de equipamentos/bemfeitorias) ou em espcie/dinheiro) relativo aos servios florestais nos ltimos 12 meses? Cdigos: 0=no; 1=sim, directamen te para as famlias; 2=sim, directamente para a comunidade (p.ex., projectos de desenvolvimento); 3=sim, ambos para famlia e comunidade 2. Se a comunidade tem recebi do pagamentos (cdigos 2 ou 3 abaixo), por favor indique a quantia que a comunidade tem recebido. Pagamentos relacionados com: Quantia 1. Turismo 2. Seqestro de carbono 3. Bacia hidrogrfica 4. Conservao de biodiversidade 9. Outros, especifique: 3. A comunidade tem recebido qua lquer apoio florestal externo (assistncia tcnica, insumos de graa, etc.) a partir do governo, doadores, ONGs) nos ltimos 12 meses? (1-0) Nota: Se qualquer destes pagamentos ou assistncia tenha sido recebido, dever ser mais elaborado na narrativa da comunidade. 164

PAGE 165

APPENDIX C ANNUAL HOUSEHOLD SURVEY 1 (PORTUGUESE) Levantamento (da unidade) familiar a nual 1 (A1) (no Acre: colocao) Informao de controle Tarefa Data(s) Por quem? Est OK ? Se no, faa comentrios Entrevista Verificar o questionrio Codificar o questionrio Digitalizao de dados Verificar e aprovar a digitalizao de dados A. Identificao 1. Identificao e localizao da famlia. 1. Nome da da unidade familiar e cdigo *(nome) (FID) 2. Comunidade *(nome) (comunidade ##) 3. Distrito/Municpio *(nome) (DID) 4. Nome e PID (Veja seco B. ) entrevistado primrio *(nome) (PID) 5. Nome e PID (Veja seco B. ) entrevistado secundrio *(nome) (PID) 6. Ponto de referncia com base em GPS da famlia (formato UTM) 7. Distncia da famlia do centro da comunidade (em minutos andando a p e em km) 1. min 2. km B. Composio da famlia 1. Quem so os membros da famlia? Nota: Recorde-se da definio de famlia no guia tcnico. 1. Nmero de identificao pessoal (PID) Nome do agregado familiar 2. Parentesco com o chefe da famlia1) 3. Ano de nascimento (yyyy) 4. Sexo (0=masculino 1=femenino) 5. Educao (nmero de anos que completou) 1 Chefe da famlia = cdigo 0 2 3 4 5 6 7 8 9 10 11 12 13 14 165

PAGE 166

1) Cdigos: 1=esposa; 2 filho/filha; 3=enteado/enteada; 4= neto(a); 5=Mae/pai; 6=sogra/sogro; 7=irmo ou irm; 8=cunhado/cunhada; 9=tio/tia; 10=sobrinho/sobrinha; 11=filho/filha adoptado(a); 12=outra familiar; 13=no parente. 2) Algem pode perguntar sobre idade e depois calcular o ano nascimento quando entrando os dados. 2. Gostaramos de fazer pergun tas sobre o chefe da famlia. 1. Qual o estado civil do chefe da famlia? Cdigos: 1=casado ou conjugal; 2=casado mas o marido trabalha longe; 3=viuvo/viuva; 4=divorciado;; 5=nunca foi casado; 9=outros, especifique: 2. A quanto tempo que a famlia se formou (veja a definio de famlia) Anos 3. O chefe da famlia nasceu na comunidade? Se sim, dirija-se para 5. (1-0) 4. Se no: A quanto tempo o chefe da famlia vive nesta comunidade? Anos 5. O chefe da famlia pertence ao maior grupo tnico/social da comunidade? (1-0) C. Terra 1. Por favor indique o tamanho de terra/rea (em hectares) que possui e tem arrendado ou arrendou. Nota: Veja as definies de categorias de terra no guia tcnico. Categoria 1. rea (ha) 2. Propriedade (cdigoposse) Principais culturas agrcolas plantadas/colhidas nos ltimos 12 meses Max 3 (cdigo-produto) 3. Ordene1 4. Ordene2 5. Ordene3 Floresta: 1. Floresta natural 2. Floresta manejada 3. Plantaes Terra agrcola: 4. Culturas 5. Pastos (naturais ou plantados) 6. Sistemas agroflorestais (SAFs, quintal) 7. Silvipastoreio 8. Pousio/Capoeira 9. Outros tipos de vegetao/usos da terra (residencial, floresta, capinzal, pntanos, etc.) 10 Total da terra poss uda (1+2+3++9) 11 Terra alugada/arrendada algum (includa em 1-9) 12 Terra alugada/arrendada de algum (no includa em 1-9) D. Bens e poupanas 1. Por favor indique o tipo de casa que possui? 1. Tem casa prpria? 1) 2. Qual o tipo principal (mais comum) de material das paredes? 2) 3. Qual o tipo principal (mais comum) de material no seu telhado/cobertura? 3) 4. Quantos m2 aproximadamente tem a casa? m2 1) Cdigos: 0=no; 1=casa prpria; 2=casa prpria pertencente a mim e outra pessoa(s); 3=aluga a casa sozinho; 4=aluga a casa com outra famlia(s); 9=outros,e especifique: 166

PAGE 167

2) Cdigos: 1=barro/areia; 2=Madeira; 3=peas metlicas ; 4=tijolos ou cimento; 5=paxiuba (tronco de palmeiramuito comum na Amaznia) 9=outro, especifique: 3) Cdigos: 1=capim; 2=Madeira (tbuas); 3=peas metlicas ; 4=telhas; 9=outros, especifi que: (No foi traduzida) 2. Por favor indique o nmero e valor dos instrumentos e outros grandes itens que a famlia possui. 1. No. de unidades possuda 2. Valor total (valor corrente de venda de todas as unidades, no preo de compra) (Lc$. Se o bem no for prprio, coloque ) 1. Carro/caminho 2. Trator 3. Moto 4. Bicicleta 5. Telefone fixo/ celular 6. TV 7. Radio 8. Fita cassette/CD/ VHS/VCD/DVD/ 9. Fogo (s a gs ou eltrico) 10 Geladeira/congelador 11 Barco de pesca e motor de barco 12 Moto serra 13 Arado 14 Carroa / Atrelado para bois 15 Espingarda/Arma de fogo 16 Carroa ou carrinho de mo 17 Movis 18 Agua 19 Panel solar 99. Outros (preo de compra maior que aprox. 50 USD ) 3. Por favor indique as poupanas e dvidas que a famlia possui. 1. Quanto que a famlia possui de poupanas nos bancos, associaes de crditos ou em clubes de poupana? Lc$ 2. Quanto que a famlia possui em poupanas em bens no produtivos como ouro e jias? Lc$ 3. Quanto que a famlia tem em dvidas no pagos? Lc$ E. Base de recursos florestais 1. Qual a distncia entre a casa e a margem da mata/floresta natural ou manejada mais prxima a qual tem acesso e pode usar? 1. medida em termos de distncia (linha recta?) km 2. medida em termos de tempo (minutos a caminhar)? min 2. A famlia coleta lenha? Se no, dirija-se para 8. (1-0) 3. Se sim : quantas horas por semana os membros da famlia gasta na coleta de lenha para uso familiar? (horas) 4. Na sua casa, vocs gastam mais ou menos te mpo para coletar lenha, comparando com 5 anos atrs? Cdigos: 1=mais; 2=mais ou menos mesmo tempo; 3=menos 5. Como que mudou a disponibilidade de lenha nos ltimos 5 anos? Cdigos: 1=diminuiu; 2=mais ou menos a mesma; 3=aumentou se for cdigo ou 3, dirija-se para 7. 6. Se tiver reduzido (cdigo da Resposta Ordene 1-3 167

PAGE 168

pergunta acima), como que vocs reagiram frente ao declnio da disponibilidade de lenha? Por favor ordene as respostas mais importantes, mximo 3. 1. Aumentou o tempo de colheita (ex. o local de colheita mais distante de casa) 2. Plantio de rvores em terra prpria/particular 3. Aumentou o uso de resduos da agricultura como combustvel 4. Compra (mais) lenha e/ou carvo 5. Compra (mais) combustveis comerciais (petrleo, gs ou eletricidade) 6. Reduziu a necessidade de uso de combustveis, p.ex. usando foges melhorados 7. Usa a lenha numa maneira mais conservativa para cozinhar e esquentar a casa 8. R eduziu o nmero de MEALS cozinadas 8. Usa melhor tecnlogia 9. Aumentou o uso de produtos no madereiros (ej. capim) 10 Restrita o acesso/uso de possuir a floresta 11 Conserva os arvores para o futuro 12 Fazer carvo 9. Outro, especifique: 7. A famlia ter plantado rvores na sua colocao/terra nos ltimos 5 anos? Se no, dirija-se para a prxima seco. (1-0) 8. Se sim: qual foi o principal propsito(s) do plantio destas rvores? Por favor ordene os propsitos mais importantes max 3. Propsito Ordene 1-3 1. Lenha para uso domstico 2. Lenha para venda 3. Forragem para uso prprio 4. Forragem para venda 5. Madeira/estacas para uso prprio 6. Madeira/estacas para venda 7. Outros usos domsticos 8. O utros produtos para venda 9. Seqestro de carbono 10. Outros servios ambientais 11. Demarcao da terra 19. Outros, especifique: F. Grupos de usurios florestais (GUF) Nota: O entrevistador deve primeiro explicar o que quer dizer por GUF, cf. o guia tcnico. (Evite o nome GUF na entrevista) 1. Voc ou outro membro da sua famlia so membros de um grupo de usurios florestais (GUF)? Se no, dirija-se para 11. (1-0) 2. Algum na sua famlia participa normalmente/regularmente nos encontros dos grupos de usurios florestais (GUF)? Se no, dirija-se para 5. (1-0) 3. Se sim : na sua famlia, quem normalmente participa nos encontros dos grupos de usurios florestais e participa em outras actividades dos GUF? Cdigos: 1=somente a esposa; 2=ambos, mas principalmente a esposa; 3=ambos participam da mesma maneira; 4=ambos, mas principalmente o marido; 5=somente o marido; 9=outros esquemas no descritos anteriormente. 4. Quantas pessoas dia (= dia completo de trabalho) os membros da famlia gastaram no total em atividades do GUF (encontros, fiscalizao, trabalho conjunto, etc.) nos ltimos 12 meses? dias 5. A famlia faz pagamentos em dinheiro ou contribuies para o os grupos de usurios 168

PAGE 169

florestais (GUF)? Se no, dirija-se para 7. (1-0) 6. Se sim: Quanto que pagou nos ltimos 12 meses? (Lc$) 7. A famlia recebeu qualquer pagamento em dinheir o dos grupos usurios da floresta (p. ex., diviso das vendas/receitas) nos ltimos 12 meses? Se no, dirija-se para 9. (1-0) 8. Se sim: Quanto que recebeu nos ltimos 12 meses? (Moeda nacional) 9. Quais foram as razes para vocs se juntar ao GUF? Por favor ordene as razes mais importantes, max 3. Razes Ordene 1-3 1. Aumentou o acesso aos produtos florestais 2. Melhor manejo florestal e mais benefcios no futuro 3. Acesso a outros benefcios, p.ex., apoio governamental, programas de doadores 4. obrigao proteger a floresta para a comunidade e para o futuro 5. Para ser respeitado e considerado como pessoa responsvel na comunidade 6. Aspectos sociais (encontrar-se com outras pessoas, trabalhar em grupo, receio de excluso, etc.) 7. Forado pelo Governo/lderes/vizinhos 8. Preo mais alto para o produto florestal 10. Melhor qualidade do produto florestal 9. Outros, especifique: 10 De um modo geral, o que diria sobre como a existncia do GUF tem afetado os benefcios que a famlias obtm da floresta? Cdigos: 1=efeito grande negativo; 2=pequeno efeito negativo; 3=nenhum efeito; 4=Pequeno efeito positivo; 5=grande efeito positivo. 11 Se participa em nenhum GUF, Porque ? Por favor ordene as razes mais importantes, max 3. Razoes Ordene 1-3 1. No h GUF na comunidade 2. Sou novo na comunidade 3. Os membros dos GUF de um modo geral pertencem a outros grupo(s) (etnia, partido poltico, religio, etc.) diferentes do meu 4. No possui tempo disponvel 5. No possui o recurso/dinheiro requerido para pagar 6. Os membros do GUF iriam restringir/empatar o meu uso da floresta, e eu pretendo usar a floresta em funo das minhas necessidades 7. No acredito que GUF efetiv o no (ou sabe fazer o) manejo da floresta 8. Falta dos produtos florestais 10. No tem interes nas atividades feitas pelos GUFs 11. Corrupo no GUF 12. Tem interes em afiliar-se mas precisar mais informao 13. GUF existe na comunidade mas unidade familar no sabe da presena 14. Autoridades florestais 9. Outros, especifique: 169

PAGE 170

APPENDIX D ANNUAL HOUSEHOLD SURVEY 2 (PORTUGUESE) Levantamento (da unidade) familiar anual 2 (A2) Informao de controle Tarefa Data(s) Por quem? Esta OK? Se no, faa comentrios Entrevista Verificar o questionrio Codificar o questionrio Digitalizao dos dados Verificao e aprovao da digitalizao dos dados A. Identificao 1. Nome da da unidade familiar *(nome) (FID) 2. Comunidade *(nome) (comunidade ##) 3. Distrito *(nome) (DID) 4. Nome e PID do respondente primrio *(nome) (PID) 5. Nome e PID do respondente secundrio *(nome) (PID) B. Crises e despesas inesperadas 1. A famlia tem enfrentado qualquer falta de rendimentos financeiros ou despesas grandes inesperadas nos ltimos 12 meses? Evento 1. Como sria?1) Como que compensou a perda de rendimento ou custos? Ordene max. 32) 2. Ordem1 3. Ordem2 4. Ordem3 1. Sria falha de cultura agrcola/roado 2. Doena grave na fam lia (grupo de idade do adulto incapaz de trabalhar por mais de um ms durante os ltimos 12 meses devido doena ou por tomar conta de algum doente) 3. Morte de um adulto em idade produtiva 4. Perda de terra (expropriao, etc.) 5. Perda grande de gado (roubo, seca, etc.) 6. Outra perda grande de um bem (fogo, roubo, enchente, etc.) 7. Perda de salrio/emprego 8. Casamento 9. Outro, especifique: 1) Cdigos: 0=No; 1=Sim, crise moderada; 2= Sim, crise sria. Veja o guia tcnico para definies. 2) Cdigos compensar: 1. Explora mais produtos florestais 2. Explora mais produtos silvestres no florestais 3. Planta e colhe mais produtos agrcolas 4. Gastou as poupanas de dinheiro 5. Venda de bens (terra, gado, etc.) 6. Faz trabalho casual/servio prestado/por diria 7. Assistncia de amigos ou parentes 170

PAGE 171

8. Assistncia de uma ONG, organizao comunitria, organizao religiosa ou similar 9. Obteve emprstimo de um emprestador de dinhe iro, associao cr edora, banco etc. 10 Tentou reduzir as despesas familiares 11 No fez nada em particular 19. Outros, especificar: C. Servios florestais 1. A famlia recebeu, nos ltimos 12 me ses, algum dinheiro ou pagamento relacionado com os seguintes servios florestais? Principal propsito 1. Recebeu? (1-0) 2. Se sim, quantidades (valores) recebidas (moeda nacional) (Se no tiver nada, coloque ) 1. Turismo 2. Projetos de carbono 3. Projetos de conservao de gua, bacia hidrogrfica 4. Conservao da biodiversidade 5. Outras, especifique: D. Abertura da floresta 1. A famlia derrubou/cortou qualquer rea florestal nos ltimos 12 meses? Se no, dirija-se para 9. (1-0) Se sim: 2. Qual foi a rea derrubada? ha 3. Para que fim foi usada a rea derrubada? Cdigos: 1=roado; 2=plantio de arvores; 3=pastoreio; 4=usos no agrcolas (Ordene max 3) 1.Ordem1 2.Ordem2 3.Ordem3 4. Se usado para roado (cdigo na pergunta anterior), Quais foram as principais culturas plantadas? (cdigo-produto) Ordene max 3 1.Ordem1 2.Ordem2 3.Ordem3 5. Qual foi o tipo de floresta que derrubou? (cdigo-florestal) 6. Se foi floresta secundria, qual foi a idade da floresta? anos 7. Quem era o dono da floresta derrubada? (cdigo posse) 8. A que distncia da casa estava a terra derrubada? km 9. A famlia nos ltimos 5 anos desbravou a floresta? Se no, dirija-se para 11. 1-0 10 Se sim: que superfcie (aprox.) foi desbravada nos ltimos 5 anos? ha 11 Que quantidade de terra utilizada pela famlia nos ltimos 5 anos foi abandonada (deixada para ser convertida para vegetao natural capoeira)? ha E. Percepo de bem-estar e capital social 1. Considerando todas as coisas, o quanto o senhor esta satisfeito com sua vida durante os ltimos 12 meses? Cdigos: 1=muito insatisfeito; 2=insatisfeito; 3=nem insatisfeito nem satisfeito; 4=satisfeito; 5=muito satisfeito 2. Tm sido a renda da famlia e a produo de comida suficiente durante os ltimos 12 meses para cobrir as necessidades da famlia? Cdigos: 1=no; 2=razovel (justo o suficiente); 3=sim 171

PAGE 172

3. Em comparao com outras famlias na associao (comunidade), como o senhor considera que sua famlias est? Cdigos: 1=pior; 2=ao redor da media; 3=melhor 4. Como esta sua famlia hoje em comparao com sua situao h 5 anos atrs ? Cdigos: 1=pior agora; 2=quase igual; 3=melhor agora Se 1 ou 3, ir a 5. Se 2, ir a 6. 5. Se est pior o melhor qual a principal razo da mudana? Por favor faa um ordem de prioridade s respostas mais importantes. Maximo 3. Razo: Mudana em Range 1-3 1. emprego fora da parcela 2. extenso da parcela (p.ex., terra vendida/comprada) 3. recursos florestais 4. preos da produo (florestal, agrcola,) 5. apoio externo (governo, ONG,) 6. remessas 7. custo da vida (p.ex.,inflao alta) 8. guerra, conflito civil, intranqilidade 9. conflitos na associao (no violentos) 10 situao familiar (p. ex., perda de um membro da famlia que contribua ao sustento) 11 doena 12 acceso (p.ex., ramal novo,) 19. outros (especificar): 6. O senhor considera que sua associao (com unidade) um bom lugar para viver? Cdigos: 1=no; 2=parcialmente; 3=sim 7. O senhor em geral confia nas pessoas em sua associao (comunidade)? Cdigos: 1=no; 2=parcialmente, confio em alguns e no em outros; 3=sim 8. O senhor consegue obter ajuda das pessoas em sua associao (comunidade) em caso da necessidade, por exemplo se precisar de dinheiro extra por a doena de um membro da famlia? Cdigos: 1=no; 2= as vezes posso obter a ajuda mas no sempre; 3=sim F. Avaliao da pesquisadora sobre a famlia Nota: Esta seco dever ser completada pelo pesquisador (a) e/ou a contraparte PEN. Se a pesquisadora que fez A2 (e Q4) no quem tem feito os questionrios trimestrais prvios, aquelas pesquisadoras que tivessem maior relao com as famlias devero responder as perguntas 2-5. 1. Durante a entrevista, o entrevistado sorriu ou gargalhou? Cdigos: 1 = nunca gargalhou nem sorriu (cara feia), 2 = s sorriu, 3 = sorriu e gargalhou, 4 = sim, gargalhou freqente e abertamente 2. Baseado em suas impresses e em o que tem visto ( casa, objetos, etc.), voc acha que esta famlia e rica comparada com outras famlias na associao? Cdigos: 1= pobre, 2 = media, 3 = rica 3. O quanto confivel a informao geralmente fornecida por esta famlia? Cdigos: 1=pobre; 2=razoavelmente confivel; 3=muito confivel 4. O quanto confivel a informao sobre uso/coleta florestal fornecida por esta famlia? Cdigos: 1=pobre; 2= razoavelmente confivel; 3= muito confivel 5. Se a informao sobre o uso e coleta florestal n o to confivel (cdigo 1 acima), pensa que a informao fornecida sobreestima ou sube stima o uso real da floresta? Cdigos: 1=subestima; 2=sobreestima; 3= no h sobre ou subestimao sistemtica; 4=no sei 172

PAGE 173

APPENDIX E QUARTERLY HOUSEHOLD SURVEY (PORTUGUESE) Levantamento familiar trimestral (Q1-Q4) Nota: Todos os re ndimentos so levantados para o ltimo ms (ltimos 30 dias ), exceto para a ltima seco de culturas, gado e outras fontes de rendimentos aonde o perodo para relembrar de 3 meses. Nota: O investigador dever listar os produtos mais comuns em vrias tabelas, com base em RRAs e pr-teste do questionrio. Aps ter perguntado sobre a pr-lista de produtos, o entrevistador dever perguntar se h algum outro produto no mencionado que a famlia explorou/coletou no ltimo ms ou 3 meses. Informao de controle Tarefa Data(s) Por quem? OK? Se no, faa comentrios Entrevista Verificar o questionrio Codificar o questionrio Digitalizao dos dados Verificao e aprovao da digitalizao dos dados A. Identificao 1. Nmero da famlia 2. C omunidade *(nome) (comunidade ##) 3. Distrito 4. Nome e BI do respondente primrio *(nome) (BI) 5. Nome e BI do respondente secundrio *(nome) (BI) B. Renda direta da floresta (rendimentos de produtos florestais no processados 1. Quais so as quantidades e valores de produtos florestais brutos (no processados) que os membros da sua famlia coletaram para uso domstico e venda no ms passado ? Nota: Respostas nas colunas 3 e 4 devem ser consistentes com as categorias de terra reportadas no questionrio da comunidade (V1D01) e no questionrio anual da famlia (A1C). 1. Produto Florestal (cdigoproduto) 2. Cole tado por que m?1) Aonde foi coletado? 5. Quant idade coleta da (7+8) 6. Uni dad e 7. Uso prprio (incl. present e/dado) 8. Vend a(incl. trocas ) 9. Pre o por uni dad e 10. Tipo de mer cado (cdi gomerc ado) 11. Va lor br ut o (5* 9) 12. Custo s de trans porte/ venda (total) 13. Compra de insumos/ materiais & pagament o de mode-obra 14. Rendi mento liquid o (1112-13) 3. Tip o de terr a (cd igoterr a) 4. Propri edade (cdig oposse) Castanha Borracha FRUTO S Aa (fruto) Pato (fruto) Bacaba 173

PAGE 174

Buriti Ouricuri Tucum (fruto) Jatob (fruto) Juta (fruto) Bacur Jac Cacau da mata Doce Palmito CASCAS Adubo vegetal Castanha Copaiba Jatob Juta Pau darco roxo Catuaba Quina quina Canelo Unha de gato Breu Cedro Cerejeira Aguano Assac Sucuuba Balso Garrafada OLEOS Copaiba Andiroba Pato Jatob Mel de abelha FIBRAS Cip amb Cip timb 174

PAGE 175

Cip titica Arum Envira de toar Envira outros Jarina Jac Uricuri SEMEN TES Jarina Tucum Aa Pato Paxiubo Paxiubin ha Murmur MADEI RA Amarelo Angelim Aroeira Blsamo Breu vermelho Cambar Castanhei ra Catuaba Cedro Cerejeira Copaiba Cumaru cetim/cu marurana Cumaru ferro Guariba Imbirindi ba amarela Intaba Jatob Marup Massaran duba 175

PAGE 176

Mogno/A uguano Mulateiro Pau darco roxo Pau darco amarelo Paxiubo Paxiubin ha Preciosa / canelo Quariqua ra Samaba branca Samaba preta Toari/Tau ari ANIMAI S Porco do mato Queixada Veado roxo Veado capoeiro Paca Cutia Cutiara Tatu Capelo / guariba Macaco prego Macaco aranha Capivara Jabuti Anta Jac Mutum Papagaio Namb Jacam Tucano Arara 176

PAGE 177

1) Cdigos: 1=S/principalmente pela esposa e mulher es adultas membros da famlia; 2=ambos adultos homens e mulheres participando de igual forma; 3=S/principal mente pelo marido e homem adulto membro da famlia; 4=S/principelmente por meninas (<15 anos); 5=S/principalmente por rapazes (<15 anos); 6=S/principalmente por crianas (<15 anos), e rapazes e meninas participando de forma igual; 7=todos os membros da famlia participam de igual forma; 8=Nenhuma das alternativas anteriores C. Renda derivada da floresta (renda de produtos processados) 1. Quais so as quantidades e valores dos produtos florestais processados que os membros das sua famlia produziram no ms passado? 1. Produ to (cdig oprodut o) 2. Quem na famlia faz o trabalh o?1) 3. Quantid ade produzi da (5+6) 4. Unid ade 5. Uso prprio (incl. presente/d ado) 6. Ven da (incl troc as) 7. Preo por unida de 8. Tipo de merca do (cdig omerca do) 9. Val or bru to (3* 7) 10. Compra de insumos/mat eriais & mo-de-obra assalariada 11. Custos de transpo rte/ venda 12. Rendime ntos lquidos excl. custos de insumos florestais (9-10-11) Vinho Vasso ura Paneir o Cesta Madei ra 1) Cdigos: 1=S/principalmente pela esposa e mulher es adultas membros da famlia; 2=ambos adultos homens e mulheres participando de igual forma; 3=S/principal mente pelo marido e homem adulto membro da famlia; 4=S/principalmente por meninas (<15 anos); 5=S/principalmente por rapazes (<15 anos); 6=S/principalmente por crianas (<15 anos), e rapazes e meninas participando de forma igual; 7=todos os membros da famlia participam de igual forma; 8=Nenhuma das alternativas anteriores 2. Quais so as quantidades e valores dos produtos florestais no processados usados como insumos para produzir produtos florestais processados na tabela anterior? Nota: os produtos na coluna 1 tero de ser exatamente os mesmos que os da coluna 1 na tabela anterior. 1. Produtos processa dos (final) (cdigoproduto) 2. Produtos florestais no processa dos usados como insumo (cdigoproduto) 3. Quantida de usada (5+6) 4. Unida de 5. Quantida de comprad a 6. Quantida de coletada pela famlia Aonde foram coletados? 9. Quem na famlia coleta dos produtos florestais ?1) 10. Preo por unida de 11. Valo r (3*1 0) 7. Tipo de terra (cdig oterra) 8. Proprieda de (cdigoposse) 177

PAGE 178

1) Cdigos como na tabela acima. Nota: Colunas 7,8,9 devero ser deixadas em branco se a famlia no co lecta. Coluna 10 (preo) devera ser perguntado mesmo se s fazem a colecta, mas se no estiv er disponvel, veja guia t cnico na valorizao. Nota: Resposta nas colunas 7 e 8 devem ser consistentes com as categorias de terra reportadas no questionrio das comunidades (V1D01) e no questionrio anual das famlias (A1C). D. Pesca e piscicultura 1. Quanto peixe a sua famlia pescou exclusivamente selvagem (rios, lagos, mar) durante o ms passado ? *Tipo de peixe (listar os nomes locais) Onde foi coletado? 3. Total pescad o (kg) (4+5) 4. Uso prprio (incl. presente/dad o) 5. Vend a (incl. trocas ) 6. Pre o por kg 7. Valo r brut o (3*6) 8. Custos (insumos, mode-obra assalariada, venda/transport e) 9. Rendiment o liquido (7-8) 2. Tipo de terra (cdigo -terra) 3. Propriedad e (cdigoposse) Tucunare Tambaqu i Piranha Boado Piau Curimat Maturix Pirarucu Filote Surub dorado Piaba Manj Nota: as respostas nas colunas 2 e 3 devem ser consistentes com as categorias de terra reportadas no questionrio das comunidades (V1D01) e no questionrio anual das famlias (A1C). 2. Quanto peixe a sua famlia pescou dos audes (piscicultura) no ms passado ? Tipo de peixe (listar os nomes locais) 1. De onde? 1) 2. Total pescado (kg) (3+4) 3. Uso prprio (incl. presente/dado) 4. Venda (incl. trocas) 5. Preo por kg 6. Valor bruto (2*5) 7. Custos (insumos, mode-obra assalariada, venda/transporte) 8. Rendimento liquido (6-7) Tilapia 1) Cdigos: 1=Tanque propriedade da famlia; 2= Tanque propriedade de um grupo que a famlia membro; 3=Tanque propriedade da comunidade/comunidade; 4=Tanque propriedade de outros e as pessoas podem comprar direitos de pesca (incluindo custos na coluna 7); 9=Outros,especifique: 178

PAGE 179

E. Rendimentos ambientais no florestais 1. Em cima dos produtos florestais e piscatrios includos nas tabelas anteriores, quanto de outros produtos selvagens (p. ex., de savanas, terras em pousio, etc.) a famlia colectou no ms passado ? 1. Tipo de produto (cdigoproduto) Aonde foi colectado? 4. Quanti dade colecta da (6+7) 5. Unidad es 6. Uso prprio (incl. present e/dado) 7. Venda (incl. trocas) 8. Preo por unidade 9. Valor bruto (4*8) 10. Custos (insumo s, mode-obra assalari ada, venda/t ranspor te) 11. Rendim ento liquido (9-10) 2. Tipo de terra (cdigo -terra) 3. Propri edade (cdigo -posse) Nota: As respostas nas colunas 2 e 3 devero ser consistentes com as categorias de terra reportadas no questionrio das comunidades (V1D01) e com o questionrio anual das famlias (A1C). F. Rendimentos salariais 1. Algum membro da famlia foi pago pelo trabalho realizado no ms passado? Nota: Uma pessoa poder ser indicada mais do que uma vez para diferentes trabalhos. 1. Membro da famlia (PID) 2. Tipo de trabalho (cdigotrabalho) 3. Dias de trabalho no ms passado 4. Pagamento dirio 5. Rendimento salarial total (3*4) G. Rendimento do negcio prprio (no florestal ou agricultura) 1. Est envolvido em algum tipo de negcio, e se sim, qual o rendimento bruto e os custos relacionados com o negcio no ms passado? Nota: Se a famlia estiver en volvida em diferentes tipos de negcios deve preencher cada coluna para cada negcio. 179

PAGE 180

1. Negcio 1 2. Negcio 2 3. Negcio 3 1. Qual o tipo de negcio?1) 2. Rendimento bruto (vendas) Custos: 3. Compra de insumos/materiais 4. Insumos prprios, no incluindo mo-de-obra (valor equivalente de mercado) 5. Mo-de-obra assalariada 6. Custos de transporte e venda 7. Custos de reparao, manuteno, etc. 8. Outros custos 9. Rendimento liquido (2 itens 3-8) 10 Valor corrente do capital armazenado 1) Cdigos: 1=loja/comrcio; 2=processamento agrcola; 3=artesanato; 4=carpintaria; 5=outro baseado em floresta; 6=outro mode-obra treinada; 7=transporte (carro, barco,); 8=acomodao/restaurante; 19=outra, especifique: H. Rendimento a partir da agricultura culturas 1. Quais so as quantidades e valores das culturas que a famlia colheu nos ltimos 3 meses ? 180

PAGE 181

1.Culturas (cdigoproduto) 2. rea de produo ( m2) 3. Produo Total (5+6) 4. Unidades (para produo) 5.Uso prprio (incl. presente/dado) 6. Vendas (incl. trocas) 7. Preo por unidade 8.Valor total (3*7) Arroz Feijo Milho Macaxeira* *Farinha Caf Abacaxi Cupuau Amendoim Cana de azucar Banana Pimenta (do reinho) Maracuja Laranja Tangarina Limo Graviola Acerola Goiaba Caj Melancia Guaran Mamo Manga Ing Abacate Batata Cebolha Abobera Tabaco Pepino Tomate Pupunha 181

PAGE 182

2. Quais so as quantidades e valores dos insumos da produo das culturas nos ltimos trs meses (isto refere-se a despesas dinheiro da agricultura)? Nota: tomar em considerao todas as culturas na tabela anterior. Insumos/Materiais 1. Quantidade 2. Unidades 3. Preo por unidade 4. Custo total (1*3) 1. Sementes 2. Fertilizantes 3. Pesticidas/herbicidas 4. Estrume/Adubo animal 5. Trao animal/Animal de trabalho 6. Mo-de-obra assalariada 7. Aluguel de maquinrio 8. Transporte/venda 19. Outros, especifque a. foisa b. tesado c. lima d. limato e. corrente f. gasolina g. oleo 20. Pagamento pelo aluguel da terra I. Rendimentos a partir do gado 1. Qual o nmero de animais ADULTOS que a famlia possui, e quantos vendeu, comprou, matou ou perdeu durante os ltimos 3 meses ? 1. Nmero inicial (3 meses atrs) 2. Venda (incl. trocas), vivos ou mortos 3. Abatidos para uso prprio (ou presente dado) 4. Perdidos (roubados, mortos,..) 5. Comprados ou presente/dado recebidas 6. Cuantos tornaramse adultos? 7. Nmero agora (1-2-34+5+6) 8. Preo por animal adulto 9. Valor total final (7*8) 1. Bovino Boi transporte Touro Novilho Garrote Bezerro Vaca corte Vaca leite 2. Bfalos 3. Cabrras 4. Ovelhas 5. Porcos 6. Burros 7. Patos 8. Galinhas Adultos 182

PAGE 183

Frangos Pintos 9. CavaloEgua 10. Capote 11. Outros 2. Quais so as quantidades e valores de produtos animais e servios produzidos nos ltimos 3 meses ? Produto/servio 1. Produo (3+4) 2. Unidades 3. Uso prprio (incl. presente/dado) 4. Venda (incl. trocas) 5. Preo por unidade 6. Valor Total (1*5) 1. Carne 1) 2. Leite 3. Manteiga 4. Queijo 5. Manteiga de bfalo 6. Ovos 7. Peles 8. L 9. Estrume 10 Trao animal 11 Rapadura 12. Outros 1) Garanta que corresponde com as tabelas anteriores de vendas e consumo familiar de animais. 3. Quais so as quantidades e valores de insumos/ materiais utilizados na produo de gado durante os ltimos 3 meses (despesas em pronto pagamento)? Nota: O ponto chave obter os custos totais, em vez de unidades de insumos. Insumos 1. Unidade 2. Quantidade 3. Preo por unidade 4. Custo total (2*3) 1. Alimentao/pasto Sal Vitaminas 2. Aluguel da terra para pastagens 3. Medicamentos, vacinas e outros servios veterinrios Vacinas aftose Vacinas carbuncro (s para bezerros) Vacinas raiva 4. Custos de manuteno estbulos, cercas, currais, etc. 5. Mo-de-obra assalariada 6. Outra, especifique: 4. Por favor indique aprox. a diviso de alimentao dos animais, quer para pastos de animais prprios ou trazidos para a casa/estbulo/colocao por membros da famlia. 183

PAGE 184

Tipo de terra de pastag em ou fonte de forragem 3. Diviso aproximada (%) 1. Tipo de terra (cdigo-terra) 2. Propriedade (cdigo-posse) Total 100% J. Outras fontes de rendimento 1. Por favor descreva qualquer outra fonte de rendimento que as famlias vm recebendo nos ltimos 3 meses Tipo de rendimento Quantia total recebida nos ltimos 3 meses 1. Remessas (apoio financeiro por parentes/amigos que moram em outro lugar) 2. Apoio governamental, ONG, organizaes ou similar a. FUNRURAL b. Soldado da borracha c. Salario de maternidade d. Auxilio da saude e. Auxilio escolar f. Bolsa famila 3. Ofertas/apoio de amigos e parentes 4. Penso 5. Pagamento por servios florestais 6. Pagamento pelo aluguel da terra (se for em servios e bems, indique o equivalente em valores monetrios) 9. Outros, especifique: 184

PAGE 185

APPENDIX F HOUSEHOLD ATTRITION SURVEY (PORTUGUESE) Levantamento de abandono e ausencia temporria Informao de controle Tarefa Data(s) Por quem? OK? Se no, faa comentrios Entrevista Verificar o questionrio Codificar o questionrio Digitalizao dos dados Verificao e aprovao da digitalizao dos dados A. Identificao 1. Identificao e local da unidade familiar 1. Nome da unidade familiar e cdigo *(nome) (NIF) 2. Comunidade e cdigo *(nome) (comunidade ###) 3. Distrito e cdigo *(nome) (NID) 4. Quem foi entrevistado?1) 5. A famlia saiu temporariamente (s um trimestral) ou permanentemente? (1=temporrio; 2=permanente; 3=ainda no sei)2) 1) Cdigos: 1 = membro da famlia; 2 = vizinhos; 3 = parentes; 4 = lder ou representante da comunidade; 9=outros, especifique: ____________ 2) Cdigo 3 dever ser usado s temporariamente; usa 1 ou 2 na entrada dos dados finais. B. Razes para no participar 1. Qual foi a razo para a famlia no participar neste levantamento trimestral? Razo 0-1 (quest. 1) ou cdigo 1. M udou/migrou permanentemente 2. Temporariamente fora da comunidade (trabalho, visita, ) 3. Se divorciu 4. Se casou 5. Morreu 6. Enfermidade 7. Nascimento de uma criana 8. No participou porque est ocupado demais 9. No participou porque no quer revelar informao da famlia 10 No participou porque est cansado de responder as perguntas 11 No localizou a casa 19. Outro 2. Se se mudou/migrou (resposta 1), para onde? Cdigos: 1=dentro da comunidade; 2=comunidade vizinha; 3=outra comunidade mais longe; 4=a vila mais perto; 5=a cidade mais longe; 9=outro:_________ 3. Se se mudou/migrou (resposta 1), porque foi embora? Cdigos: 1=trabalhar ou procurar trabalho; 2= ( governo) servio, incl. militar; 3=estudar; 4= estar mais perto a um esposo(a)/famlia; 5=casamento; 6=separao/divorcio; 7= usar herena; 8= procurar tratamento mdico; 9=conflitos na comunidade; 19=outro, ___________ 4. Se entrevistado faleceu (resposta 5), colocar o NIP: 5. Se o entrevistado faleceu, qual foi a causa da morte? Cdigos: 1=doena; 2=velie; 3=acidente; 4=violncia; 5=suicidio; 9=outro:_________ 185

PAGE 186

APPENDIX G HOUSEHOLD BRAZIL NUT MANAGE MENT SURVEY 2006 (PORTUGUESE) Comunidade: ____________________________________________ Famlia: ________________________________________________ # _______ Data: ___________________________ Pesquisador(a): ___________ 1. Quantas rvores de castanheira tm em sua propriedade? rvores 2. Quantas rvores nunca produzem ourios? rvores 3. Que data entrou na safra da castanha este ano? 4. At que data colectou castanha na safra? 5. Quando foi a safrinha este ano? 6. Nmero(#) de dias que voc trabalha na safra por semana Dias 7. Nmero(#) de dias na safrinha? Dias 8. Descreve suas atividades associadas com a safra: (1=juntar e quebrar o mesmo dia, 2=juntar uma estrada e depois quebrar, 3=juntar tudo y depois quebrar, 4=quando tm pouco, faz metodo #1, quando tm mais, faz metodo #2, 5=juntar um dia, quebrar o dia seguinte) 9. Quem da familia trabalha con Vc. na safra? Membro da familia (NIP) Quantos dias por semana? Paga esta pessoa? (10) Quanto paga? 10. Contrata mo de obra em sua propriedade durante a safra? Si a resposta no, passa ao nmero # 13 (1-0) 11. Quantas pessoas pagou esta safra? Pessoas Nome Sexo (0=masc, 1=fem) Ano do nascimento (aaaa) Relao ao produtor 186

PAGE 187

12. Como pagou? (1=dinheiro, 2=parte da safra, 3=troca dos dias, 4=outro )? E, quanto? 13. Voc scio de uma cooperativa? Se a resposta no, passa ao nmero # 15 (1-0) 14. Qual cooperativa? 15. Quantas latas voc coletou no ano passado (2005)? Latas 16. Quantas latas este ano (2006)? Latas 17. A quem voc vendeu sua castanha Nome Tipo de comerciante (1=marrateiro, 2=empresa, 3=cooperativa; outro produtor) Quantas latas? Quanto pagou por lata? 18. Qual foi o preo da lata no ano passado (2005)? 19. Voc usa algumas das prticas seguintes no manejo de seu castanhal? a Mapeamento das castanheiras (1-0) Se sim, qual instituio ajudou com o mapeamento? Quanto tempo gastou na prtica? das b Plano de manejo (1-0) Se sim, desde quando (aaaa) ? c Plantio de mudas (1-0) Se sim, quantas mudas foram plantadas? plntulas Onde? (1=roado; 2=capoeira, 3=mata, 4=jardim na casa, 5=outro) Qual instituio ajudou? d. Manuteno das mudas nas capoeiras e roados (0=no, 1=s nas roados, 2=s nas capoeiras 3=s na mata, 4=em todos os lugais, 5=outro lugar) e. Limpeza ao redor das mudas (0=no, 1=s nas roados, 2=s nas capoeiras 3=s na mata, 4=em todos os lugais, 5=outro lugar) Se sim, porque faz a limpeza? (1=crescer melhor, 2=previr fogo, 3=ver a muda melhor, 4=outro) Rango 1: Rango2: Rango 3: f. Protege as mudas nascidas contra fogo? (0=no, 1=s nas roados, 2=s nas capoeiras 3=s na mata, 4=em todos os lugais, 5=outro lugar) g Corte de cips? (1-0) Se sim, porque corta? (o rvore produce mais frutas, 2=cips matam o rvore, 3=o Rango 1: 187

PAGE 188

rvore crescerz melhor, 4=as ramas caen com o peso dos cips, 5=parte das Normas Tcnicas, 6=limpar a copa, 7=outro) Rango2: Rango 3: h. Sangrar os rvores (0=nunca, 1=s aos rvores que nunca producem, 2=a maioria dos rvores) (1-0) i. Quebra os ourios no mesmo dia da coleta (1-0) Mximo nmero dos dias que Vc. deixara a castanha juntada na mata dias j. Transporte das castanha s a um armazm no mesmo dia da coleta (1-0) Se sim, porque no mesmo dia? (1=a castanha fica ruin na mata, 2=os animais comen, 3=a ruin contamina ao resto, 4=eficiencia, 5=previr robos, 6=vende imediatamente, 7=costume, 8=outros) Rango 1: Rango2: Rango 3: Se no, mximo nmero de dias deixara a castanha quebrada na mata? das Metodo de transporte da mata (1=espalda, 2=moto, 3=animal, 4=trator, 5=bicicleta, 6=outro) k Secagem das castanhas (1-0) Tem seu propio armazm? (0=no, 1=sim, 2=sim, com um grupo) Tem seu propio secador? (0=no, 1=sim, 2=sim, com um grupo) l. Uso de sacos para transportar la castaa (0=no, 1=sim, tem que comprar suas bolsas,, 2=sim, os compradores do as bolsas) (1-0) 20. Como transporta Vc. as castaas ao mercado? (1= Seu mesmo, 2=comerciante, 3=cooperativa/empresa (camino), 4=coopera tiva/empresa (barco), 5=amigo/familia, 6=alugar transporte) 21. Voc sabe que so as aflatoxinas (fungo/veneno)? (1-0) Se sim, como aprendeu sobre as aflatoxinas? 22. Voc usa algumas das prticas seguintes: a. Remover o umbligo (1-0) b. Remover a castanha cortada (1-0) c. Secar a castanha bem (1-0) d. Separar a castanha dos animais y combustivl (1-0) e. No deixar os cocos por muito tempo na mata (1-0) f. No coletar a castanha do ano anterior (1-0) 23. Vc. caza durante a safra? (1-0) Se sim, Vc. caza mais ou menos que durante a safra que durante outras pocas do ano? (1=mas, 2=menos, 3=igual) 24. Seu castanhal (est ou est sendo) certificado? (1-0) 188

PAGE 189

Se sim, qual tipo da certificao? (1=orgnica, 2= FSC, 3=mercado justo) Desde quando? (aaaa) 25. Quais atividades so necessrias para ser certificado? Rango 1-3 a. Remover a castanha ruin b. Secar a castanha c. Construir un armazm para secar a castanha d. Usar bolsa novas e. No contato com animais, combustivl etc. f. Mapeamento do castanhal g. Empezar a safra cedo h. Asistir as reunies i. Outro: 26. Tem tido conflitos/ pelejas na poca da safra? (1-0) Conflitos de que tipo? (1=Menhor; sobre os arvores j resolvido, 2= Roubos da castanha (outros comunarios), 3= Roubos por vizinhos (comunarios), 4=Roubos por vizinhos (empresas), 5=Roubos por pessoal da cidade, 6=Roubos por extranjeiros, 7=Roubos da castanha nas bolsas, 8=Empresas enganhando) Rango 1: Rango2: Rango 3: Descreve em detalhe os conflitos, incluido quantas latas eram perdidas y como resolviu o conflito? 27. Vc. teve problemas con o fogo entrando em seu castanhal no ano passado? (1-0) Se sim, quantas hetares da mata queimou? ha 28. Daqui 10 anos, Vc. pense que a parcela que tem agora vai ser sua? (1=muito seguro, 2=seguro, 3=inseguro) Porque? 29. Daqui 10 anos, Vc. pense que vai a ter (1=menos, 2=mais, 3=o mesmo nmero de) rvores da castaa que tem agora? Porque? 30. O que fez com a renda da safra de castanha este ano? APPENDIX H 189

PAGE 190

HOUSEHOLD BRAZIL NUT MANAGE MENT SURVEY 2007 (PORTUGUESE) Comunidade: ____________________________________________ Famlia: ________________________________________________ # _______ Data: ___________________________ Pesquisador(a): ___________ 31. Que data comencou a safra da castanha este ano? At quando o senhor pretende coletar a castanha? 32. Quantas latas colect ou este ano (2007)? latas Quantas mais latas o senhor pretende coletar? latas 33. Pagou trabalhadores para ajudar com a safra este ano (2007)? (1-0) Quantas pessoas? Como lhes pagou? Quantas latas do total tiraram? 34. A quem o senhor vendeu sua castanha este ano? Nome Tipo de comercante (1=marreteiro, 2=empresa direita, 3=comunario, 4=cooperativa) Quantas latas? Quanto lhe pagou por lata? 35. O senhor entrou nesta safra com uma dvida da safra de 2005-06? (1-0) Quanto? A quem? 36. O senhor tirou um adiante do dinheiro antes desta safra (2006-07) da castanha? (1-0) Quanto? A quem? 37. O que comprou com o adiante e/o a renda desta safra? (p. ex. comida, remedio, moto, gado) 38. O senhor scio da CAPEB/CAEX? Desde quando? 190

PAGE 191

(1-0) 39. O senhor vende castanha orgnica? (1-0) a. Como dira o senhor que vender a castanha orgnica tem afeitado os beneficios que a famlia obtene da floresta? ( 1=efeito negativo grande; 2=efeito negativo pequeno; 3=nenhum efeito; 4=efeito positivo pequeno; 5=efeito positivo grande) b. Quais so os beneficios de produzir castanha certificada ( 1=melhor preo, 2=outro) Rango 1: Rango2: Rango 3: c. Quais so os aspetos negativos de ser certificado? (1=muito trabalho, 2=outro) Rango 1: Rango2: Rango 3: 40. O senhor participou num mapeamento de seu castanhal? (1-0) a. Como dira que o mapeo dos castanhais tem afeitado sua familia? ( 1=efeito negativo grande; 2=efeito negativo pequeno; 3=nenhum efeito; 4=efeito positivo pequeno; 5=efeito positivo grande) b Quais so os beneficios do mapeo ( 1=reduzir conflitos dentro da comunidade, 2=reduzir conflitos com os vizinhos, 3=fazer a safra mais eficente, 4=melhor manejo, 5=seguir as Normas Tcnicas, 6=aprender novas tcnicas, 7=outro) Rango 1: Rango2: Rango 3: c. Quais so os aspetos negativos? (1=muito trabalho, 2=nada mudou, 3=mais roubos depois do mapeo, 4=outro) Rango 1: Rango2: Rango 3: 41. Em sua opinio, seus direitos a sua colocao so (1=muito inseguro s, 2=inseguros, 3=no seguro,no inseguro, 4=seguros, 5=muito seguros) 42. Tm tido conflitos/roubos este ano na poca da safra da castanha (2006-07)? (1-0) Se sim, conflictos de qual tipo? (1=Menhor sobre os arvres j resolvido, 2= Roubos por comunarios, 3= Roubos por vizinhos (comunarios), 4=Roubos por vizinhos (empresas), 5=Roubos por gente da ciudade, 6=Roubos por extranjeiros, 7=Roubos da castanha em sacos, 8=Empresas enganando, 9=outro) Rango 1: Rango2: Rango 3: Descreve en detalhe os conflitos, incluindo quantas latas eram perdidas y como resolviu o conflito? 43. O senhor quemou sua terra no ano passado (2006)? (1-0) a. Em que mes? b. Que tipo da terra foi quemado? (1=mata bruta, 2=capoeira (>15 aos), 3=capoeira (<15 aos), 4=outro) c. Quantas hectres eram queimadas? ha d. Porque o senhor quemou? ((1=fazer roado, 2=queimar pasto, 3=otro) 191

PAGE 192

44. O senhor teve problemas con o fogo entrando en seu castanhal por accidente no ano passado (2006)? (1-0) Se sm, quantas hectres da floresta quemaram? ha 45. O qu era/ o enfoque do uso da tierra por sua famlia? (1=castanhaa, 2=madeira, 3=borracha, 4=agricultura, 5=animais pequenos, 6=pecuaria, 7=mineiria, 8=outro) a. Faz 10 anos (1997) Rango 1: Rango2: Rango 3: b. Agora (2007) Rango 1: Rango2: Rango 3: c. Daqui 10 aos (2017) Rango 1: Rango2: Rango 3: 46. Como o senhor acha que asfaltar o ramal aqu afectara a vida da pessoal? ( 1=efeito negativo grande; 2=efeito negativo pequeno; 3=nenhum efeito; 4=efeito positivo pequeno; 5=efeito positivo grande) Porque? 47. O que sera a melhor maneira em apoiar para a produo da castanha no futuro? Y quem pode apoiar? APPENDIX I 192

PAGE 193

LANDSAT SCENES USED FOR LAND USE LAND COVER CHANGE ANALYSIS IN BRAZIL NUT-PRODUCING COMM UNITIES IN THE MAP REGION Platform Path Row Year Month Day TM 2 67 1986 8 6 TM 2 68 1986 8 6 TM 2 69 1986 9 7 TM 3 67 1986 7 28 TM 3 68 1986 7 12 TM 3 69 1986 7 12 TM 2 67 1991 7 27 TM 2 68 1991 7 27 TM 2 69 1991 7 27 TM 3 68 1991 10 14 TM 3 69 1991 10 14 TM 3 67 1992 6 18 TM 2 67 1996 8 1 TM 2 68 1996 7 16 TM 2 69 1996 8 17 TM 3 67 1996 7 23 TM 3 68 1996 7 23 TM 3 69 1996 7 23 ETM+ 2 67 1999 8 2 TM 2 67 2000 7 27 ETM+ 2 68 2000 11 24 TM 2 69 2000 7 27 ETM+ 3 67 2000 5 23 ETM+ 3 68 2000 7 26 ETM+ 3 69 2000 5 23 TM 2 67 2005 8 10 TM 2 68 2005 6 7 TM 2 69 2005 8 10 TM 3 67 2005 6 30 TM 3 68 2005 9 18 TM 3 69 2005 9 18 193

PAGE 194

194 APPENDIX J DISTANCE FROM ROAD AND DEFORESTATION RATES FOR MADRE DE DIOS, PERU (GREEN), ACRE, BRAZIL (PINK) AND PANDO, BOLI VIA (ORANGE). Courtesy of Southworth et al., in preparation.

PAGE 195

APPENDIX K RESULTS OF UNIVARIATE REGRESSIONS FOR POTE NTIAL EXPLANATORY VARIABLES ON LOG FOREST CLEARED FOR COMMUNI TIES IN EACH COUNTRY IN THE TIME PERIOD 2002-2007. Madre de Dios, Peru Acre, Brazil Pando, Bolivia Variable Coeff. t -Statistic Coeff. t -Statistic Coeff. t -Statistic F-Stat Household characteristics Size 0.0592 0.90 0.0385 1.09 0.0632 1.77+ 0.167 Distance to village center (min) 0.0220 1.47 -0.00044 -0.23 0.00716 2.84** 0.019 Distance to city -0.00013 -0.01 -0.000260 -0.35 -0.00372 -3.04** 0.028 Education (sum yrs) 0.00710 0.84 0.0061 0.46 0.00955 1.52 0.362 Age of household head (yrs) 0.01430 1.60 0.00002 0.00 0.01384 2.23** 0.062 Number adults (age 15-65) 0.0406 0.55 0.0746 1.17 0.1373 2.02* 0.130 Number children (age <15) 0.007 0.06 0.0360 0.67 0.0401 0.78 0.788 Number elders (age >66) 0.298 0.231 0.013 0.02 0.164 0.87 0.492 Length of residency (yrs) 0.01081 1.23 -0.00099 -0.15 0.00190 0.26 0.662 Household income (USD per capita) Forest income (w/out Brazil nuts) 0.000458 0.91 -0.000083 -0.22 0.000524 1.16 0.531 Brazil nut income -0.000899 -3.61*** 0.000123 0.26 -0.000953 -4.57*** <0.001 Aquaculture income 0 --0.00129 0.719 0.0359 2.43* 0.104 Wage income 0.000444 0.94 -0.000093 -0.38 -0.000148 -0.44 0.748 Business income -0.001127 -2.27* -0.00034 -0.22 -0.000437 -1.64 0.205 Crop income 0.001889 2.89** -0.000267 -1.36 0.001041 4.63*** <0.001 Livestock income -0.00060 -0.57 -0.000376 -0.89 0.000533 1.84+ 0.217 Other income -0.00156 -0.44 0.000102 0.35 -0.00369 -1.97** 0.246 Total income -0.000328 -1.75+ -0.000099 -0.91 -0.000002 -0.02 0.278 Assets Total land (ha) -0.000604 -1.56 0.000166 0.80 0.000061 0.22 0.378 Material assets (USD per capita) -0.000109 -0.14 0.000028 0.12 -0.000055 -0.45 0.971 Livestock assets (USD per capita) 0.000277 0.45 0.0000715 0.94 0.0002111 2.82** 0.033 Notes: All variables were measured using country as a fixed e ffect. Households were the unit of observation (n=125). 195

PAGE 196

LIST OF REFERENCES Abbas, H.K., 2005. Aflatoxin and Food Safety CRC Press, Taylor and Francis Group, Boca Raton, Florida. ACA (Amazon Conservation Association) 2008. Mitigation Strategy for the Interoceanic Highway. Amazon Conservati on Association, Washington DC. Agrawal, A., Gibson, C., 1999. Enchant ment and disenhantm ent: the role of community in natural resource conser vation. World Development 27, 629. Alcorn, J.B., 2000. Borders, Rules and Gove rnance: Mapping to Catalyse Changes in Policy and Management. IIED, Gate keeper Series SA91, London, UK. Allegretti, M.H., 1989. Reserv as extrativistas: uma pro posta de desenvolvimento da floresta Amaznica. Par Desenvolvimento 25, 2. Allegretti, M., 1990. Extracti ve reserves: an alternative for reconciling development and environmental conservation in Amazonia. In: Anderson, A.B. (Ed). Alternatives to Deforestation: Steps Toward Sustainable Use of the Amazon Rainforest. Columbia University Press, New York, USA, pp. 252. Allen, M., 2005. Textbook on Criminal Law. Oxford Universi ty Press, Oxford, England. Alston, L.J., Libecap, G.D., Mueller, B. 2000. Land reform policies, the sources of violent conflict, and implications for deforestation in the Brazilian Amazon. Journal of Environmental Ec onomics and Management 39, 162. Alvarez, N.L., Naughton-Treves, L., 2003. Linking national agrarian policy to deforestation in the Peruvian Amazon: a case study of Tambopata, 1986-1997. Ambio 32, 269. Angelsen, A., 2006. A stylized model of incent ives to convert, maintain, or establish forests. In: Chomitz, K.M., At Loggerheads ? Agricultural Expansion, Poverty Reduction, and Environment in the Tropical Forests. The International Bank for Reconstruction and Development. The World Bank, Washington D.C., USA, pp. 70. Angelsen, A., Wunder, S., 2003. Ex ploring the forest-poverty link: key concepts, issues and research implications. CIFOR Occasional Paper No. 40, Bogor, Indonesia. Ankersen, T., Barnes, G., 2005. Inside the Polygon: emerging community tenure systems and forest resource extraction. In: Zarin, D., Putz, F.E., Schmink, M., and Alavalapati, J. (Eds.), Working Forests in the Tropics: Conservation Through Sustainable Management ? Columbia University Press, New York, USA, pp. 156 177. 196

PAGE 197

Arnold, J.E.M., Ruz-Prez, M., 2001. Can non-timber forest products match tropical forest conservation and development obj ectives? Ecological Economics 39, 437 447. Auld, G., Gulbrandsen, L.H., McDermott, C. L., 2008. Certification schemes and the impacts on forests and forestry. Annual Review of Environment and Resources 33, 187. Azevedo, T.R., de Freitas, A.G., 2003. Fore st certification in Brazil: the parallel evolution of community forest man agement in the Braz ilian Amazon and FSC Certification. In: Molnar, A., 2003, Fore st Certification and Communities: Looking Forward to the Next Decade, Forest Tr ends, Washington D.C., USA, Annex 1. Barham, B., Coomes, O., 1996. Prosperity's Promise: The Amazon Rubber Boom and Distorted Economic Development. Westview Press, Boulder, Colorado, USA. Bartels, W.L., 2009. Participatory land use planning in the Brazilian Amazon: creating learning networks among farmers non-governmental organizations, and government institutions. Ph.D. Dissertation, University of Florida, Gainesville, USA. Bass, T., Markopoulos, R., Grah, G., 2001. Certifications Impacts on Forests, Stakeholders and Supply Chain Instrum ents for Sustainable Private Sector. Forestry Series, International Institute for Environm ent and Development, London, UK. Belcher, B., Ruz-Prez, M., Achdiawan, R ., 2005. Global patterns and trends in the use and management of NTFPs: implicatio ns for livelihoods and conservation. World Development 33, 1435. Berkes, F., 2007. Community-based conserva tion in a globalized world. Proceedings of the National Academy of Sciences of the United States of America, 104, 15188. Bojanic, A., 2001. Balance is Beautiful: Assessing Sustainable Development in the Rainforests of the Bolivian Amazon. PROM AB scientific series 4. Riberalta, Beni, Bolivia. Bray, D.B.B., Plaz Snchez, J.L., Murphy, E.C., 2002. Social dimensions of organic coffee production in Mexico: lessons for eco-labeling initiatives. Society and Natural Resources 15, 429-446. Bray, D.B., Duran, E., Ramos, V.H., Mas, J.F., Velazquez, A., McNab, R.B., Barry, D., Radachowsky, J., 2008. Tropical deforesta tion, community forests, and protected areas in the Maya Forest. Ec ology and Society 13, 56. 197

PAGE 198

Browder, J.O., 1990. Extractive reserves will not save the tropics. Bioscience 40, 626. Browder, J.O., 1992. The limits of extractivism: tropical forest strategies beyond extractive reserves. Bioscience 42, 174. Brown, I.F., 2004. Significant results of L BA-MAP. Presentation, January 20, 2004, Rio Branco, Acre. Butler, R.A., Laurance, W.F., 2008. New stra tegies for conserving tropical forests. Trends in Ecology and Evolution 23, 469. Caldas M., Walker, R. Arima, E., Perz, S., Aldrich, S., Simmons, C., 2007. Theorizing land cover and land use change: The peasant economy of Amazonian Deforestation. Annals of the Associ ation of American Geographers 97, 86. Card. A.A, 2000. Manejando Bien Tu Cast aal. Asociacin para la Conservacin de la Cuenca Amaznica (ACCA), Puerto Maldonado, Peru. Carrera, F., Stoian, D., Campos, J.J., Morales, J., Pinelo, G., 2004. Forest certification in Guatemala. Proceedings of the Sy mposium on Forest Certification in Developing and Transitional Societies, Yale School of Forestry and Environmental Studies, New Haven, pp. 363. Cavendish, W., 2000. Empirical regularities in the poverty-environment relationship of rural households: evidence from Zimbabwe. World De velopment 11, 1979. Caviglia-Harris, J.L., Sills, E.O., 2005. Land use and income diversification: comparing traditional and colonist populations in the Brazilian Amazon. Agricultural Economics 32, 221. CBMF (Conselho Brasileiro de Manejo Flor estal), 2003. Padres de certificao do FSC-Forest Stewardship Council (Consel ho de Manejo Florestal) para o manejo e explorao de populaes naturais de castanha (Bertholletia excelsa), Documento Verso 4.0. IMAFLORA (Insti tuto de Manejo e Cert ificao Florestal e Agrcola), Piracicaba, SP, Brazil. CFV (Certificacion Forestal Voluntar ia), 2006. Estandares bolivianos para la certificacin forestal de la castaa (Ber tholletia excelsa). Santa Cruz, Bolivia. Chapin, M., Lamb, Z., Threlkeld, B., 2005. Mapping indigenous l ands. Annual Review of Anthropology 34, 619. Charnley, S., Poe, M.R., 2007. Community fo restry in theory and practice: where are we now? Annual Review of Anthropology 36, 301. 198

PAGE 199

Chase Smith, R., Benavides, M., Pariona, M., Tuesta, E., 2003. Mapping the past and the future: geomatics and indigenous territo ries in the Peruvian Amazon. Human Organization 62, 357. Chavez, A., 2009. Public policy and spatial variation in land use and cover in the southeastern Peruvian Amazon. Ph.D. Dis sertation, University of Florida, Gainesville. Chhatre, A., Agrawal, A., 2009. Trade-offs and synergies between carbon storage and livelihood benefits from fo rest commons. Proceedings of the National Academy of Sciences of the United States of America 106, 17667. Chibnik, M.,1994. Risky Rivers: The Econom ics and Politics of Floodplain Farming in Amazonia. University of Arizona Press, Tucson, Arizona, USA. Chomitz, K.M., 2007. At Loggerheads? Agricultural expans ion, poverty reduction, and environment in the tropical forests. The International Bank for Reconstruction and Development. The World Bank Washington D.C., USA. Chomitz, K.M. Gray, D.A., 1996. Roads, la nd use and deforestation: a spatial model applied to Belize. World Bank Economic Review 10, 487. Chowdhury, R.R., 2006. Landscape change in the Calakmul Biosphere Reserve, Mexico: modeling the driving forces of smallholder deforestation in land parcels. Applied Geography 26, 129. Colfer, C., 2005. The Equitable Fores t. Diversity, Community, and Resource Management. Resources for the Future Press, Washington, D.C., USA. Contreras, A., Vargas, M. T., 2001. Social, Environment al and Economic Dimensions of Forest Policy Reforms in Bolivia. Fo rest Trends, Center for International Forestry Research, Washington, D.C., USA. Coomes, O.T., 1996. State credit programs and the peasant ry under populist regimes: lessons from the ARPA experience in the Peruvian Amazon. World Development 24, 1333. Coomes, O.T., Grimard, F., Po tvin, C., Sima, P., 2009. The fa te of the tropical forest: carbon or cattle? Ecological Economics 65, 207. Cotta, J.N., Kainer, K.A., Wadt, L.H.O., St audhammer, C.L., 2008. Sh ifting cultivation effects on Brazil nut (Bertholletia exce lsa) regeneration. Forest Ecology and Management, 256, 28. Cowell, A., 1990. The Decade of Destruct ion: The Crusade to Save the Amazon Rainforest. Anchor Books, New York, USA. 199

PAGE 200

CP-CFV (Consejo Peruano para la Certificacin Forestal Voluntaria), 2005. Estndar para la certificacin del manejo fore stal con fines de produccin de castaa (Bertholletia excelsa) en Per. WWF-Peru, Puerto Maldonado, Peru. Cronkleton, P., Goenner, C., Ev ans, K., Haug, M., De Jong, W., Albornoz, M.A., 2007. Supporting forest communities in times of tenure uncertainty: participatory mapping experiences from Bolivia and Indonesia. RECOFT International Conference, Poverty reduction and forest s: tenure, market and policy reforms, Bangkok, Thailand. Cronkleton, P., Evans, K., Albornoz, M.A., De Jong, W., 2008. Towards well-being: helping local governments respond to forest dependent people; experiences from the Northern Bolivian Amazon. Center for International Forestry Research, Bogor, Indonesia. Cronkleton, P., Pacheco, P. Changing Poli cy Trends in the Emergence of Bolivias Brazil nut sector. In: Laird, S. McLain, R. and Wynberg, R. (Eds.). Non-Timber Forest Products Policy: Framewor ks for the Management, Trade and Use of NTFPs. Earthscan Publications Ltd., London, England, in press. CTAR, 1998. Carretera Interoceanica Peru-Brasil. Consejo Transitorio de Administracion Regional, Puerto Maldonado, Peru. De Jong, W., Ruiz, S., and Becker, M., 2006. Conflicts and communal forest management in northern Bolivia. Fore st Policy and Economics 8, 447. De Oliveira, J., 2008. Property rights, land conflicts and deforestation in the Eastern Amazon. Forest Policy and Economics 10, 303. De Soto, H., 2000. The Mystery of Capital: Why Capitalism Triumphs in the West and Fails Everywhere Else. Basic Books, New York, USA. Demsetz, H., 1967. Toward a theory of property rights. American Economic Review Papers and Proceedings 57, 347. Dennis, R.A., Meijaard, E., Na si, R., Gustafsson, L., 2008. Bi odiversity conservation in Southeast Asian timber concessions: a cr iticial evaluation of policy mechanisms and guidelines. Ecology and Society 13, 25. DiGiano, M., 2006. Assessing the potential impacts of policy interventions on household income and land use among extr activists and colonists in Brazil's western Amazon. M.S. Thesis, Universi ty of Florida, Gainesville, USA. Dourojeanni, M., 1990. Amaz onia: que hacer? Centro de Estudios Teolgicos de la Amazonia (CETA), Iquitos, Peru. 200

PAGE 201

Dourojeanni, M., 2006. Estudio de caso sobre la carretera Interocenica en la Amazonia sur del Per. SERV IGRAHEIRL, Lima, Peru. Dove, M.R., 1994. Marketing the rainfores t: 'green' panacea or red herring? Analysis from the East-West Center 13, 1. Duchelle, A., Cronkleton, P. Kainer, K., Guanacoma, G ., and Gezan, S. Property rights devolution to communities in Western Amazonia: implications for sustainable forest management. Ec ology and Society, in review. EFSA (European Food Safety Authority), 2009. Effects on public health of an increase of the levels for aflatoxin total from 4 g/kg to 10 g/kg for tree nuts other than almonds, hazelnuts and pistachios. T he EFSA Journal, 1168, 1. Ehringhaus, C., 2006. Post-victory dilemmas: land use, development, and social movement in Amazonian Extractive Reserves. Ph.D. Dissertation, Yale University, New Haven, Connecticut, USA. El Deber, 2005. Conflicto por tierras y cast aa deja tres muertos. July13, 2005. Santa Cruz de la Sierra, Bolivia. Ellis, E., Porter-Bolland, L., 2008. Is community-based fore st management more effective than protected areas? A compar ison of land use/land cover change in two neighboring study areas of the Central Yucatan Pe ninsula, Mexico. Forest Ecology and Manage ment, 256, 1971. Faminow, M.D., 1998. Cattle, Deforestati on and Development in the Amazon: An Economic, Agronomic and Environmental Perspective. CAB International Oxford, UK. FDA (Food and Drug Administration), 2000. Gu idance for Industry: Action Levels for Poisonous or Deleterious Substances in Human Food and Animal Feed. College Park, Maryland, USA. Fearnside, P.M., 1997. Greenhouse gases from deforestation in Brazilian Amazonia: net committed emissions. Climatic Change 35,321. Fearnside P.M., 2002. Land-tenur e issues as factors in environmental destruction in Brazilian Amazonia: the case of S outhern Par. World Development 29,1361 1372. Fearnside, P.M., 2003. Conservation policy in Brazilian Amazonia: understanding the dilemmas. World Development 31, 757. Fifer, V.J., 1970. The empire builders: a history of the Bolivian rubber boom and the rise of the house of Surez. Journal of Latin American Studies 2, 113. 201

PAGE 202

Fitzpatrick, D., 2006. Evolution and chaos in property rights systems: the third world tragedy of contested access. The Yale Law Journal 115, 996. FLO (Fairtrade Labelling Organizations International), 2009. FLO homepage. Accessed on September 22, 2009 at www.fairtrade.net Fortmann, L., Bruce, J., 1988. Introduction. In: Fortmann, L., Bruce, J. (Eds.). Whose trees? Proprietary dimensions of forest ry. Westview Press, Boulder, Colorado, USA, pp. 1. Garca-Fernndez, Ruiz-Prez, C.,M., Wunder, S. 2008. Is multiple-use forest management widely implementable in t he tropics? Forest Ecology and Management 256, 1468. Geist, H.J., Lambin, E.F., 2002. Proximate causes and underlying driving forces in tropical deforestation. Bioscience 52, 143. Geoghegan, J., Villar, S. C., Klepis, P., Mendoza, P.M., OgneveHimmelberger, Y., Chowdhury, R.R., Turner II, B.L., Vance, C. 2001. Modeling tropical deforestation in the southern Yucatn peninsular region: comparing survey and satellite data. Agriculture, Ecosystems and Environment 85, 25. Gibson, C., Ostrom, E., Mc Kean, M.A., 2000. Forests, people and governance: some initial theoretical lessons. In: Gibson, C. McKean, M.A. and Ostrom, E. (Eds.). People and Forests: Communities, Institutions and Governance. Massachusetts Institute of Technology, Cambri dge, Massachusetts, USA, pp. 227. Godoy, R., Reyes-Garcia, V., Vadez, V. Leonard, W.R., Tanner, S., Huanca, T., Wilkie, D., TAPS Bolivia Study Team, 2009. The relation between forest clearance and household income among nat ive Amazonians: Results from the Tsimane Amazonian panel study, Bolivia. Ecological Economics 68, 1864. Gomes, C.V., 2001. Dynamics of land use in an Amazonian extractive reserve: case of Chico Mendes Extractive Reserve in Acre Brazil. M.Sc. thesis. University of Florida, Gainesville, Florida, USA. Gomes, C.V.A., Bartels, W.L., Schmink, M., Duarte, A.P., Ar cos, H.D.S., 2008. Planejando Futuros Sustentveis co m Pequenos Produtores: Programa Proambiente Plo Alto Acre. In: Bensusan, N., Armstrong, G. (Eds.). O Manejo da Paisagem e a Paisagem do M anejo, Braslia: IIEB, 121. Gomes, C.V.A, 2009. Twenty years afte r Chico Mendes: Extractive Reserves expansion, cattle adoption and evolving self -definition among rubber tappers in the Brazilian Amazon. Extractive Reserve Ph.D Dissertation. University of Florida, Gainesville, Florida, USA. 202

PAGE 203

Google Earth, 2007. Accessed on July 22, 2007 at http://earth.google.com Gould, K., 2006. Land regulariz ation on agricultural frontiers : the case of northwestern Petn, Guatemala. Land Use Policy, 395. Guariguata, M., Cronkleton, P., Shanley, P., Taylor, P.L., 2008. The compatibility of timber and non-timber forest product extraction and management. Forest Ecology and Managem ent 256, 1477. Guariguata, M.R., Licona, L. C., Mostacedo, B., Cronklet on, P., 2009. Logging damage to Brazil nut trees (Bertholletia excelsa) in industrial timber concessions in Northern Bolivia. Forest Ecology and Management, 258, 788. Hall, A., 1997. Sustaining Am azonia: Grassroots Action for Productive Conservation. Manchester University Pre ss, Manchester, England. Hayes, T.M., 2007. Does tenure matter? A comparative analysis of agricultural expansion in the Mosquitia Forest Corridor. Human Ecology 35, 733. Homma, A.K.O., 1992. The dynamics of ex traction in Amazonia: a historical perspective. In: Nepstad, D.C. and Sc hwartzman, S. (Eds.). Non-Timber Products from Tropical Forests: Evaluation of a Conservation and Development Strategy. The New York Botanical Garden, New York, USA, pp. 23. Hudler, G.W., 1998. Magical Mu shrooms, Mischievous Mold s: The Remarkable Story of the Fungus Kingdom and its Impact on Hu man Affairs. Princeton University Press, Princeton, New Jersey, USA. Humphries, S., Kainer, K.A., 2006. Local per ceptions of certification for communitybased enterprises. Forest Ecology and Management 235, 30. IDAF (Instituto de defesa agropecuria e florestal do Acre), 2007. Campanha de Vacinao da Febre Aftose. Accessed on October 1, 2009 at http://www.ac.gov.br/index.php?option= com_content&task=view&id=1621&Itemid =116 IFOAM (International Federation of Organi c Agriculture Movem ents), 2009. IFOAM homepage. Accessed on September 22, 2009 at www.ifoam.org IIRSA, 2005. Integracin Suramericana. See www.caf.com.view/index.asp INRA, 2009. Resultados del Saneamiento, Titulacin y Distribucin de Tierras Fiscales. Instituto Nacional de Reforma Agraria, Cobija, Bolivia. 203

PAGE 204

Irwin, E.G., Geoghegan, J., 2001. Theory, data, methods: dev eloping spatially explicit economic models of land use change. Agri culture, Ecosystems and Environment, 85, 7. Kaimowitz, D., Angelsen, A., 1998. Econom ic Models of Tropical Deforestation: A Review. Center for International Fo restry Research, Bogor, Indonesia. Kaimowitz, D., Sheil, D., 2007. Conserving what and for whom? Why conservation should help meet basic human needs in the tropics. Biotropica, 39, 567. Kainer, K.A., Duryea, M.L., Costa de Ma cedo, N., Williams, K., 1998. Brazil nut seedling establishment and autoecology in Ex tractive Reserves of Acre, Brazil. Ecological Applications 8, 397. Kainer, K.A., Duryea, M.L., Ma lavasi, M.M., Silva, E. R ., Harrison, J., 1999. Moist storage of Brazil nut seeds for improved germination and nursery management. Forest Ecology and Management 116, 207. Kainer, K.A., Schmink, M., Lei te, A.C.P., Fadell, M.J.S., 2003. Experiments in forest based development in Western Amazonia. Society and Natural Resources 16, 869. Kainer, K.A., Wadt, L.H.O., Gomes-Silva, D.A.P., Capanu, M., 2006. Liana loads and their association with Bertholletia excels fruit and nut production, diameter growth and crown attributes. Journal of Tropical Ecology 22, 147. Kainer, K.A., Wadt, L.H.O., Staudhammer, C ., 2007. Explaining variation in Brazil nut fruit production. Forest Ecol ogy and Managem ent 250, 244. Koning, R., Capistrano, D., Yasmi, Y., Cerutti, P., 2007. Forest-Related Conflict: Impact, Links, and Measures to Mitigat e. Rights and Resources Initiative, Washington D.C., USA. Kramer, P.J., Kozlowski, T.T., 1978. Physiol ogy of Woody Plants. Academic Press, Inc., Orlando, Florida, USA. Knhe, R., 2004. Conflictos Entre Uso y Proteccin de los Recursos Naturales, Reserva Nacional de Vida Silvestre Am aznica Manuripi (RNVSAM). In: De Jong, W. (Ed.). Retos y Perspectivas del Nuevo Rgimen Forestal en el Norte Amaznico Boliviano. Center for Inter national Forestry Research, Bogor, Indonesia, pp. 123. Kusters, K., Achdiawan, R., Belcher, B., Pe rezl, M.R., 2006. Balancing development and conservation? An assessment of livelihood and environmental outcomes of nontimber forest product trade in Asia, Africa, and Latin America. Ecology and Society 11, 20. 204

PAGE 205

Lambin, E.F., Turner, B.L, Geist, H.J., Agbola, S.B., Angelsen, A., Bruce, J.W., Coomes, O.T., Dirzo, R., Fischer, G., Fo lke, C., George, P.S., Homewood, K., Imbernon, J., Leemans, R. Xiubin, Li, Moran, E.F., Mortimore, M., Ramakrishnan, P.S., Richar ds, J.F., Skanes, H., Steffen, W., Stone, G.D., Svedin, U., Veldkamp, T.A., Vogel, C., Jianchu, X., 2001. The causes of land-use and land-cover change: Moving beyond th e myths. Global Environmental Change 11, 261. Larson, A.M., Pacheco, P., Toni, F., Valle jo, M., 2007. The effects of decentralization on access to livelihoods assets. Journal of Environmental Development 16, 251 268. Larson, A.M., Soto, F. 2008. De centralization of natural res ource governance regimes. Annual Review of Environment and Resources 33, 213. Laurance, W.F., 1999. Reflections on the tr opical deforestation crisis. Biological Conservation 91, 109. Ludewigs, T., 2006. Land-use decision making, uncertainty and effectiveness of land reform in Acre, Brazilian Amazon. Ph.D Dissertation, Bloomington, Indiana. Lynam, T. De Jong, W., Sheil, D., Kusumant o, T., Evans, K., 2007. A review of tools for incorporating community knowledge, preferences and values into decision making in natural resources managem ent. Ecology and Society 12, 5. Marsik, M., Stevens, F., Southworth, J. Patterns and rates of land cover change in Pando, northern Bolivia from 1986 to 2005. Progress in Physical Geography, in review. McGarigal, K., Cushman, S. A., Neel, M.C., Ene, E., 2002. FRAGSTATS: Spatial Pattern Analysis Program for Categorical Maps. University of Massachusetts, Amherst, USA. McKean, M.A., 2000. Common proper ty: what is it, what is it good for, and what makes it work? In: Gibson, C., Mc Kean, M.A., Ostrom, E. (E ds.). People and Forests: Communities, Institutions and Governance. Massachusetts Institute of Technology, Boston, Massa chusetts, USA, pp. 27. Meinzen-Dick, R., Mwangi, E., 2008. Cutting the w eb of interests: pitfalls of formalizing property rights. Land Use Policy 26, 36. Mertens, B., Lambin, E.F., 2000. Land-co ver-change trajectories in southern Cameroon. Annals of the Associat ion of American Geographers 90, 467. 205

PAGE 206

Mertens, B., Poccard-Chapuis, R., Piketty, M.G., Lacques, A.E., Venturieri, A. 2002. Crossing spatial analyses and livestock economics to understand deforestation processes in the Brazilian Amazon: the case of So Flix do Xing in South Par. Agricultural Ec onomics 27, 269. Ministrio de Desarrollo Sostenible, 2006. Norma Tcnica para aprovechamento comercial sostenible de recursos forestal es no maderables en bosques y tierras forestales naturales. Resolucin Mi nisterial no. 22/2006. La Paz, Bolivia. Miserendino, R., Aguape, R., Arellano, A., G onzles, L., Torrico, A., Torrez, L., Yunoki, T., Yagami, T., 2003. Biodiv ersidad de la Reserva Nacional de Vida Silvestre Amaznica Manuripi. Herencia, Cobija, Bolivia. MDA/MMA/MDS, 2009. Plano Nacional de Pr omoo das Cadeias de Produtos da Sociobiodiversidade. Ministrio do Des envolvimento Agrrio MDA / Ministrio do Meio Ambiente MMA / Ministrio do Desenvolvimento Social e Combate a Fome MDS. Governo do Brasil, Braslia, DF. Morcillo, J.G., 1982. Del caucho al oro: el proceso colonizador de Madre de Dios. Revista Espaola de Antropol oga Americana 12, 255. Nagendra, H., Munroe, D.K. Southworth, J., 2004. From pattern to process: landscape fragmentation and the analysis of Land Use/Land Cover Change. Agriculture, Ecosystems and Environment, 101, 111. Naughton-Treves, L., 2004. Defore station and carbon emissions at tropical frontiers: A case study from the Peruvian Amazon. World Development 32, 173. Nepstad D., McGrath D., Alencar, A., Barros, A.C., Carvalho, G., Santilli, M., Vera Diaz, C.V. 2002. Frontie r governance in Amaznia. Science 295, 629. Nepstad, D., Schwartzman, S. Bamberger, B., Santilli, M., Ray, D., Schlesinger, P., Lefebvre, P., Alencar, A., Prinz, E., Fi ske, G., Rolla, A., 2006. Inhibition of Amazon deforestation and fire by par ks and indigenous lands. Conservation Biology 20, 65. Newing, H., Harrop, S., 2000. European health regulations and Brazil nuts: implications for biodiversity conservation and sustainable rural livelihoods in the Amazon. Journal of International W ildlife Law and Policy 3, 109. Newsom, D., Bahn, V., Cashore, B., 2006. Does forest certification matter? An analysis of operation-level changes required during t he SmartWood certification process in the United States. Fore st Policy and Economics 9, 197. 206

PAGE 207

Oliveira, P.J.C., Asner, G.P., Knapp, D.E ., Almeyda, A., Galvn-Gildemeister, R., Keene, S., Raybin, R. F., Smith, R.C ., 2007. Land-use allocation protects the Peruvian Amazon. Science 317, 1233. Ostrom, E., 2003. How types of goods and property rights jointly affect collective action. Journal of Theoretical Politics 15, 239. Overdevest, C., 2004. Codes of conduct and standard setting in the forest sector constructing markets for democracy? Relati ons Industrialles / Industrial Relations 59, 172. Overdevest, C., Rickenbach, M., 2006. Fore st certification and institutional governance: an empirical study of Forest Stewardship Council ce rtificate holders in the United States. Forest Policy and Economics 9, 93. Pacheco, P., Cronkleton, P., 2008a. Box 8 Developing standards for Brazil nuts in Bolivia. In: Shanley, P., Pierce, A., Lai rd, S., Robinson, D. (Eds.), Beyond Timber: Certification and Management of Non-Timber Forest Products. Center for International Forestry Re search, Bogor, Indonesia, pp. 41. Pacheco, P., Cronkleton, P., 2008b. Box 17 Land tenure struggles one of the barriers to Brazil nut certific ation in Bolivia. In: Shanley, P., Pierce, A., Laird, S., Robinson, D. (Eds.), Beyond Timber: Certification and Management of NonTimber Forest Products. Center for Inte rnational Forestry Research, Bogor, Indonesia, pp. 88. Padoch, C., 1992. Marketing of non-timber forest products in Western Amazonia: general observations and research priorities. Advances in Economic Botany 9, 43. Payne, R.W., Murray, D.A., Harding, S.A., Baird, D.B., Soutar, D.M., 2007. GenStat for Windows (10th Edition) Introduction. VSN International, Hemel Hempstead, UK. Peluso, N., 1995. Whose woods are these? Counter-mapping forest territories in Kalimantan, Indonesia. Antipode 27, 383. Peres, C.A., Baider, C., Zu idema, P.A., Wadt, L.H.O., Kainer, K.A., Gomes-Silva, D.A.P. Salomao, R.P., Simoes L.L, Franciosi, E.R.N., Va lverde, F.C., Gribel, R. Shepard, G.H., Kanashiro, M., Coventry P., Yu, D.W., Watkinson, A.R., Freckleton, R.P., 2003. Demographic threat s to the sustainabi lity of Brazil nut exploitation. Science 302, 2112. Perz, S.G., Walker, R.T., Caldas, M.M. 2006. Beyond population and environment: household demographic life cycl es and land use allocation among small farms in the Amazon. Human Ecology, 34, 829. 207

PAGE 208

Perz, S., Brilhante, S., Brown, F., Caldas, M., Ikeda, S., Mendoza, E., Overdevest, C., Reis, V., Reyes, J.F., Roja s, D., Schmink, M., Souza, C., Walker, R., 2008. Road building, land use and climate change: pr ospects for environmental governance in the Amazon. Philosophical Transacti ons of the Royal Society, 363, 1889 1895. Peters, C.M., A.H. Gentry, Mendelsohn, R.O., 1989. Valuatio n of an Amazonian rainforest. Nature 339, 655. Pfaff, A., 1999. What drives deforestati on in the Brazilian Amazon? Evidence from satellite and socioeconomic data. Journal of Environment Economics and Management 32, 26. Pierce, A., Laird, S., 2003. In search of comprehensiv e standards for non-timber forest products in the botanicals trade. International Forestry Review 5, 138. Pinto, L.F.G, Shanley, P. Go mes, A.P.C., Robinson, D., 2008. Experience with NTFP certification in Brazil. Forest s, Trees and Livelihoods 18, 37. Putz, F.E., Blate, G.M., Redford, K.H., Fimbel, R., R obinson, J.G., 2001. Tropical forest management and conservation of biodiversity: an overview. Conservation Biology 15, 7. Raynolds, L., 2007. Regulating sustainabilit y in the coffee sector: a comparative analysis of third-party environm ental and social certification initiatives. Agriculture and Human Values 24, 147. Rockwell, C., Kainer, K.A., Ma rcondes, N., Baraloto, C., 2007. Ecological limitations of reduced-impact logging at the smallh older scale. Forest Ecology and Management 238, 365. Ros-Tonen, M., Wiersum, K.F., 2005. T he scope for improving rural livelihoods through non-timber forest products: an evolving research agenda. Forests, Trees, and Livelihoods 15, 129. Ros-Tonen, M.A., Andel, T.V., Mo rsello, C., Otsuki, K., Rosend o, S., Scholz, I., 2008. Forest related partnerships in the Brazili an Amazon: there is more to sustainable forest management than reduced-im pact logging. Forest Ecology and Management 256, 1482. Ruiz, S., 2005. Rentismo, conflictos y bosques en el norte amaznico boliviano. Center for International Forestry Research, Bogor, Indonesia. Salisbury, D., Schmink, M., 2007. Cows versus rubber: changing livelihoods among Amazonian extractivists. Geoforum 38, 1233. 208

PAGE 209

Schlager, E., Ostrom, E., 1992. Property-rights regimes and natural resources: a conceptual analysis. Land Economics 68, 249. Schmink, M., 1994. The socio-economic matrix of deforestation. In: Arizpe, L., Stone, M.P., Major, D.C. (Eds.) Population and Environment: Rethinking the Debate. Westview Press, Boulder, pp. 253. Schmink, M., 2004. Communities forests, markets and conservation. In: Zarin, D., Putz, F.E., Schmink, M., Al avalapati, J. (Eds.) Work ing Forests in the Tropics: Conservation Through Sustainable Management? Columbia University Press, New York, pp. 119. Schmink, M, Wood, C., 1992. Contested Frontiers in Amazonia. Columbia University Press, New York, USA. Schwartzman, S. 1989. Extractive rese rves: the rubber tappers strategy for sustainable use of t he Amazon rainforest. In J.O. Br owder (Ed.). Fragile Lands of Latin America, Strategies for Sustai nable Development. Westview Press, Boulder, Colorado, USA, pp. 150. SEMA (Secretaria do Meio Ambiente do Acre), 2009. REDD ACRE: Plano de Ao para Desenvolvimento do Program a de Reduo das Emisses do Desmatamento e da Degradao Flores tal do Estado do Acre. Governo do Estado do Acre, Rio Branco, Brasil. Shankland, A., 1993. Brazils BR-364 Highway. T he Ecologist, 23, 141. Shanley, P., Pierce, A.R., Laird, S.A., Guilln, A. (Eds .), 2002. Tapping the Green Market: Certification and Management of Non-Ti mber Forest Products. Shanley, P., Pierce, A., Lai rd, S., Robinson, D., 2008. Be yond Timber: Certification and Management of Non-Timber Forest Products. Center for International Forestry Research, Bogor, Indonesia. Sherr, S., White, A., Kaim owitz, D. 2004. A Ne w Agenda for Forest Conservation and Poverty Reduction: Making Markets Work for Low-Income Producers. Washington, DC, Forest Trends, CIFOR, IUCN. SmartWood, 2002. Non-Timber Forest Products Certification Standards Addendum. Richmond, Vermont, USA. Smith, J., Colan, V., Sabog al, C., Snook, L., 2006. Why po licy reforms fail to improve logging practices: the role of governanc e and norms in Peru. Forest Policy and Economics 8, 458. 209

PAGE 210

Soares-Filho, B., Alencar, A., Nepstad, D. Cerqueira, G., Diaz, M.D.V., Rivero, S., Solorzano, L., Voll, E., 2004. Simulating the response of land cover changes to road paving and governance along a majo r Amazon highway: the SantaremCuiaba corridor. Global Change Biology 10, 745. Soares-Filho, B.S., Nepstad, D., Curran, L., Voll, E., Cerqueira, G., Garcia, R.A., Ramos, C.A., McDonald, A., Lefebvre, P., Schlesinger, P., 2006. Modeling conservation in the Amazon Basin. Nature, 440, 520. Sobrinho, P.V.C., 1992. Capital e Trabalho na Amaznia Ocidental. Universidade Federal do Acre, Rio Branco, Brazil. Soldn, M., 2003. The impact of certification on the sust ainable use of Brazil nut (Bertholletia excelsa) in Bolivia. Food and Agriculture Organiza tion, Rome, Italy. Southworth, J., Nagendra, H. Tucker, C., 2002. Fragmentation of a landscape: incorporating landscape metrics into satellite analyses of land-cover change. Landscape Research 27, 253. Southworth, J., Marsik, M., Stev ens, F., Duchelle, A., Qiu, Y. Perz, S., Rocha, K. Roads as drivers of change: MAP region land cover change trajectories from 1986 to 2005: dominant trends and comparis ons across the tri-national frontier, in preparation. Souza lvares, V., Leite, F.M. N. Madruga, A.L.S., Souza, J. M.L. Costa, D.A., 2009. Monitoramento da cadeia produtiva de Castanha-do-Brasil quanto contaminao por coliformes e fungos em trs castanhais no Acre. In: Duchelle, A.E., Passos, V.T.R. (Eds.) Anais do VII Seminrio Anual de Cooperao UFACUF 2009. Universidade Federal do Acre, Rio Branco, Brazil. Souza, C., Verssimo, A., da Silv a Costa, A. Reis, R.S., Bali eiro, C., Ribeiro, J., 2006. Dinmica do desmatamento no estado do Acre. IMAZON, Belm, Brazil. SPDA-INRENA., 2003. Compendio de Legislacin Forestal y de Fauna Silvestre. SPDA-INRENA, Lima, Peru. SPSS Statistics GradPack 17.0. 2008. Rel. 17.0.0. SPSS Inc., Chicago, Illinois, USA. Stoian, D., 2000. Shifts in forest producti on extraction: the post-rubber era in the Bolivian Amazon. International Tree Crops Journal 10, 277. Sunderlin, W.D., Angelsen, A., Belcher, B., Bu rgers, P, Nasi, R., Santoso, L., Wunder, S., 2005. Livelihoods, forests, and cons ervation in developing countries: an overview. World Development 33, 1383. 210

PAGE 211

Sunderlin, W., Hatcher, J., Little, M. 2008a. From Exclusion to Ownership? Challenges and Opportunities in Advanc ing Forest Tenure Reform. Rights and Resources Initiative, Wash ington D.C., USA. Sunderlin, W.D., Dewi, S., Puntodewo, A., Muller, D., Angelsen, A., Epprecht, M., 2008b. Why forests are important for gl obal poverty alleviation: a spatial explanation. Ecology and Society 13, 24. Tacconi, L., 2007. Decentralization, forest s and livelihoods: Theory and narrative. Global Environmental Change 17, 338. Taylor, P.L., 2005a. In the market but not of it: Fair Trade coffee and Forest Stewardship Council certification as market-based social change. World Development 33, 129. Taylor, P.L., 2005b. A Fair Trade approach to community forest certification? A framework for discussion. Journal of Rural Studies 21, 433. UNDP (United Nations Development Progra m), 2003. La transformacin en la regin amaznica. In: Informe de Desarrollo Humano en el Norte Amaznico Boliviano. Programa de las Naciones Unidas para el Desarrollo, La Paz, Bolivia, pp. 43. Vadjunec, J.M, Gomes, C.V., Ludewigs, T., 2009. Land-use/land-cover change among rubber tappers in the Chico Mendes Extracti ve Reserve, Acre, Brazil. Journal of Land Use Science 4, 1. Valentim, J.F., S, C.P., Go mes, F.C.R., Santos, J.C., 2002. Tendncias da pecuria bovina no Acre entre 1970 e 2000. Boleti m de Pesquisa e Desenvolvimento n 38. Rio Branco, Acre, Brazil, EMBRAPA-Acre. Vosti, S. A., Muoz Braz, E., Carpentier, C.L., DOliveira, M. V.N., 2003. Rights to forest products, deforestation and sma llholder income: Evidence from the Western Brazilian Amazon. Wo rld Development 31, 1889. Wadt, L.H.O, Kainer, K.A., Cart axo, C., Nunes, G.M, Leite, F.M, Souza, J.M., GomesSilva, D.A.P, Sousa, M. M., 2005. Manejo da castanheira (Bertholletia excelsa) para produo de castanha-do-bras il. Documento Tcnico Seprof 03. Governo do Estado do Acre, Rio Branco, Brazil. Wadt. L.H.O., Kainer, K.A ., Staudhammer, C.L., Serrano, R.O.P., 2008. Sustainable forest use in Brazilian extractive reserv es: natural regeneration of Brazil nut in exploited populations. Biologi cal Conservation 141, 332. Walker, R., Moran, E., Anselin, L., 2000. Deforestation and cattle ranching in the Brazilian Amazon: external capi tal and household processes. World Development, 28, 683. 211

PAGE 212

Wallace, R., 2004. The effects of wealth and markets on rubber tapper use and knowledge of forest resources in Acre, Br azil. Ph.D. dissertation. University of Florida, Gainesville, Florida, USA. Walter, S., 2003. Certificati on and benefit-sharing mechanisms in the field of non-wood forest products An Overview. Food and Agriculture Organiza tion, Rome, Italy. Weinstein, B., 1983. The Amazon rubber boom 1850-1920. Stanford University Press, Stanford, California, USA. White A., Martin, A., 2002. Who owns the wo rld's forests? Forest tenure and public forests in transition. Forest Trends, Washington D.C., USA. Wilsey, D., 2008. Nontimber forest product certification consi dered: the case of Chameadorea Palm Fronds (Xate). Ph.D. Di ssertation, University of Florida, Gainesville, USA. Wood, C., 2002. Introduction. In: Wood, C., Porro, R. (Eds .). Deforestation and Land Use in the Amazon. University Pre ss of Florida, Gainesville, USA. Wunder, S., 2001. Poverty alleviation and tropi cal forests what scope for synergies? World Development 29, 1817. Wyman, M., Stein, T. Mode ling social and land-use/land -cover change data to assess drivers of smallholder deforestation in Beli ze. Applied Geography, in press. Yasmi, Y., Schanz, H., Conflicts in natural resource management: toward conceptual clarity. Environmental Management, in press. ZEE (Zoneamento Ecolgico-Econmico do Acre), 2000. Ministrio do MeioAmbiente; Cooperao Brasil-Alemanha. Vol. 1-3. Rio Branco, Brazil. ZEE (Zoneamento Ecologico-Ec onmico do Acre) Fase II, 2006. Documento Sintese Escala 1:250.000. Secretar ia do Meio Ambiento do Ac re, Rio Branco, Brazil. Zuidema, P.A., 2003. Ecology and management of the Brazil nut tree (Bertholletia excelsa). PROMAB Scientific Se ries 6, Riberalta, Bolivia. Zuidema, P.A., Boot, R.G.A., 2002. Demography of the Brazil nut tree (Bertholletia excelsa) in the Bolivian Amazon: impact of seed extraction on recruitment and population dynamics, Journal of Tropical Ecology 18, 1. 212

PAGE 213

BIOGRAPHICAL SKETCH Amy Duchelle has experience in a br oad range of tropical forest management issues. She received a Bachelor of Arts in Biology from Colorado College, a Master of Science in Conservation Biol ogy and Sustainable Development from the University of Wisconsin, Madison, and completed her Doctor of Philosophy in the School of Forest Resources and Conservation at the University of Florida. Am ys first engagement with tropical forest ecology was through participati on in Boston Universitys yearlong study abroad program in Ecuador, which included in tensive field study in the Ecuadorian Amazon. After her undergraduate experience, she was employed as Education Coordinator at Olbrich Botanical Gardens in Madison, Wisconsin, where she was able to facilitate connections between people and plants through educational programming and exhibits. For her masters thesis research in Ecuador, she assessed how members of one remote Shuar community in the Amazonian Cordillera del Cndor were managing woody plants in mature and se condary forest, and evaluated a capacitybuilding program in conservation biolog y for Shuar and Aw indigenous groups. Amy strongly believes in collaborative research. During her doctoral work in Western Amazonia, she worked closely with regional partners in the development and implementation of her dissertation, incl uding EMBRAPA-Acre in Brazil, CIFOR and Herencia in Bolivia, and the Asociacin para la Conservacin de la Cuenca Amaznica (ACCA) in Peru. She was also a research partner in the Center for International Forestry Research Poverty and Environment Network (CIFOR PEN), and her fieldwork contributed to a global database on the role of forests in community livelihood systems. Amys field team included undergraduate student s affiliated with regional universities, and she mentored two students in the development of their thesis projects within her 213

PAGE 214

214 larger research umbrella. She made a co mmitment to retuning pr eliminary research results to the communities with whom s he worked and led trainings for undergraduate biology students (research design) and for communities (spatial data collection through participatory mapping). Amy also had the opportunity to participate on a scientific advisory team with regional par tners toward reforming national policy for Brazil nut management in Bolivia. During her Ph.D., she also worked as a Consultant with CIFOR and as an Environmental Auditor for SmartWood. Amy currently serves as a post-doctoral re searcher with the University of Floridas Amazon Conservation Leadership Initiative, based at the Federal University of Acre, Brazil (UFAC). She works to enhance institutional scientific capacity through training and mentoring UFAC masters students in the Ecologia e Manejo dos Recursos Naturais and Desenvolvimento Regional programs in multi-discip linary research that is problem-inspired and engages stak eholders on multiple scales. This research focuses on multiple-use forest management and assessm ent of Amazonian REDD pilot projects, through collaboration with CIFO R, and socio-ecological forest monitoring with WWF Brazil. Amy is pleased to be on a career pat h that is filled with interesting people and places and allows her to contribute to tr opical forest management and policy decisions through research and outreach.