|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 D ISENTANGLING THE EFFECTS OF DEFORESTATION AND INDIGENOUS HUNTING ON WILDLIFE IN THE AMAZON By PEDRO DE ARAUJO LIMA CONSTANTINO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2010
2 2010 Pedro de Araujo Lima Constantino
3 To my father
4 ACKNOWLEDGMENTS My family gave me the support needed to start working with indigenous people, even not understanding what I was doing no meio do mato. The support continued during my graduate education. Ana, thanks for courageously accepting the challenge of spending two years in Florida, despite all the fears and difficulties we knew we would face, and still keep staying by my side, my companheira Ana, my sister, you was especially important to me, challenging me to look into myself. Me, I admire you. Pai, thanks for your courage and perseverance. The influence you have in my life and decisions are impressive, stronger than I even realize. Dulce, thanks for all the love you have to my father. V, v, you are the persistence and kindness. Thanks Livia and Hector for helping us whenever we needed! Thanks to our family of friends we discovered in Gaines ville. All the Kaxinaw and Katukina people involved in the monitoring program are co authors of this research since they are collecti ng and analyzing hunting data in the ir own village However, a large scale analysis made it impossible to include all part icipants as a uthors. I would like to give thanks for the support provided by Brian Child, Marianne Schmink, and Robert Fletcher, and valuable comments of E. H. M. Vieria -Neto, J. Lucchetti, and C. Pizano during early drafts. Comisso Pr-ndio do Acre (CPI -AC) and Associao do Movimento dos Agentes Agroflorestais Indgenas do Acre (AMAAI -AC) provided part of the data and logistic support for field work. The School of Natural Resources and Environment, the Tropical Conservation and Development Program and the Amazon Conservation Leadership Initiative at the University of Florida, and the Wildlife Conservation Society financially supported the first authors assistantship and field research and community workshops. The Nature Conservancy
5 and Gordon and Betty Moore Foundation supported CPI AC in the Transfrontier Conservation Project Sierra/Serra del Divisor.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ...................................................................................................... 4 LIST OF TABLES ................................................................................................................ 8 LIST OF FIGURES .............................................................................................................. 9 LIST OF ABBREVIATIONS .............................................................................................. 10 ABSTRACT ........................................................................................................................ 11 CHAPTER 1 INTRODUCTION ........................................................................................................ 13 The Bushmeat Crisis in the Amazon .......................................................................... 13 Wildlife Conservation in Amazonian Protected Areas ............................................... 14 Pano People and Wildlife in Acre ............................................................................... 15 Assessing Wildlife Abundance Using Hunting Dat a .................................................. 15 Objectives ................................................................................................................... 18 Research Questions ................................................................................................... 18 2 METHODOLOGY ....................................................................................................... 19 Study Area .................................................................................................................. 19 Acre Ecological -Economic Zoning ............................................................................. 20 Pano Hunting .............................................................................................................. 20 Data Collection and Sources ...................................................................................... 21 Hunting Data ........................................................................................................ 21 Socioeconomic Data ........................................................................................... 22 Spatial Data .......................................................................................................... 22 Wildlife Status Indicators ............................................................................................ 23 Statistical analy sis ...................................................................................................... 25 Selecting Appropriate Indicator of Distance of Preferred Hunted Animals to the Village ......................................................................................................... 25 Drivers of Wildlife Depletion a cross Pano Communities .................................... 25 Linear multiple regression ............................................................................. 25 Spatial dependency ....................................................................................... 26 Capacity of indigenous lands to protect wildlife against external threats .... 27 3 RESULTS .................................................................................................................... 30 Pano Hunters Prey Profi les ....................................................................................... 30 Information Provided by Indicators of Wildlife Status ................................................ 30 Drivers of Wildlife Depletion across Pano Communities ........................................... 31
7 Spatial Dependence ................................................................................................... 34 Capacity of Indigenous Lands to Protect Wildlife against External Threats ............. 34 4 DISCUSSION .............................................................................................................. 51 Wildlife Status nearby Pano Communities ................................................................. 51 Drivers of Wildlife Depletion ....................................................................................... 52 Effects of Indigenous Hunting Pressure on Game.............................................. 53 Effects of Deforestation around Villages on Game ............................................. 54 Effects of Surrounding Disturbance: the Acre State Zoning ............................... 56 Effects of Roads on Wildlife and Hunters ............................................................ 58 Source -sink Dynamics of Game Populations ............................................................ 60 Indigenous Lands as Protected Areas for Wildlife..................................................... 61 Management Implications and State Conservation Policy ........................................ 62 Considerations on Scale and the Absence of Hunting Effect ................................... 64 5 CONCLUSIONS .......................................................................................................... 70 APPENDIX A KERNEL DENSITY ESTIMATES OF DISTANCE OF PREFERRED ANIMALS HUNTED FROM THE VILLAGE ................................................................................ 71 B WILDLIFE STATUS INDICATORS AND ITS RELATED ECOLO GICAL MODELS AND THEORIES ........................................................................................ 73 C PEARSONS CORRELATION MATRIX OF VARIABLES POSSIBLY INFLUENCING HUNTING IN INDIGENOUS VILLAGES, CONSIDERING 29 VILLAGES ................................................................................................................... 75 LIST OF REFERENCES ................................................................................................... 78 BIOGRAPHICAL SKETCH ................................................................................................ 9 5
8 LIST OF TABLES Table page 2 -1 Defensibility of Pano Indigenous Lands against external threats ......................... 29 3 -1 Wildlife species hunted by Pano communities in Acre, Braz ilian Amazon, during this study ..................................................................................................... 37 3 -2 Relative importance of prey taxa to Pano communities ........................................ 39 3 -3 Characteristics of Pano villages in the State of Acre, Brazi lian Amazon, and absent species ........................................................................................................ 39 3 -4 Pearson correlation matrix between simple indicator variables of wild life depletion in Pano villages ...................................................................................... 41 3 -5 Principal Component Analysis axis of wildl ife abundance simple indicators ....... 41 3 -6 Backward stepwise regression models for all response variables at the village level ............................................................................................................. 42 3 -7 Explanatory variables included in the final regression models after st epwise and supervised selection ....................................................................................... 43 3 -8 Spatial dependence of response variab les and resid uals of final regression models .................................................................................................................... 43 3 -9 Student's t tests on wildlife indicators between villages protected and vulnerable to external threat .................................................................................. 44 3 -10 Pearson's correlations between indicators of wildlife status that were driven by deforestation and the total defor estation within 5km radius area .................... 44 3 -11 Pano Indigenous Lands with total area that would support the average local people harvesting of preferred species in the Amazon ......................................... 45
9 LIST OF FIGURES Figure page 2 -1 Kaxinaw and Katukina Indigenous Lands studied in Acre State, Brazilian Amazon ................................................................................................................... 28 3 -1 Relative contribution o f wildlife species to Pano diet ............................................ 46 3 -2 Significant Pearson's correlation between wildlife status indicators and the proportion of individual species, major taxa, and preference group, after Bonferroni correction .............................................................................................. 47 3 -3 Minimal linear regression models after bac kward stepwise selection .................. 48 3 -4 Examples of non significant Pearson's correlation between hunting pressure, deforestation an d indi cators of wildlife status ........................................................ 49 3 -5 Box -plot of wildlife indicators among Indigenous Lands of different levels of defensibility against external threats to wildlife ..................................................... 50
10 LIST OF ABBREVIATION S AMAAI -AC Associao do Movimento dos Agentes Agroflorestais Indgenas do Acre CPI -AC Comisso Prndio do Acre CPUE Capture-per -unit effort IAA Indigenous Agroforestry Agents IL Indigenous Land PRODES Projeto Prodes Monitoram ento da Floresta Amaz nica Brasileira por Sat lite
11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science D ISENTANGLING THE EFFECTS OF DEFORESTATION AND INDIGENOUS HUNTING ON WILDLIFE IN THE AMAZON By Pedro de Araujo Lima Constantino May 2010 Chair: Brian Child Major: Interdisciplinary Ecology The decline in wildlife populations in the Amazon threatens game species and forest dwell ers subsisting from hunting. Deforestation and overhunting are the most commonly reported causes of depletion. Few studies assessed game drivers at regional scale or they did not consider the effects of hunting and habitat loss. Indigenous people in the B razilian State of Acre largely rely on wildmeat, although many villages experience game depletion. I worked with the Pano people of Acre to understand the drivers of game depletion across villages and Indigenous Lands. I derived four simple indicators and two multivariate indicators of wildlife status from ecological models to analyze 9109 hunting records from 35 Pano indigenous villages of eight Indigenous Lands between 2005 and 2009 in the Alto Juru and Purus valleys. I conducted multiple regression anal yses to identify the drivers of variation in animal dispersion, game populations, richness of sensitive species, and hunting success across villages. I supplemented these regressions with ANOVA of wildlife status in relation to village vulnerability and Indigenous Lands defensibility to external threats. Indigenous hunting pressure was not associated to forest loss in Pano hunting territory. Deforestation explained between 30% and 56% of variation in mean prey weight, richness of sensitive
12 species and capture per -unit effort Indigenous density and road presence explained 30% of variation in the distance of animals from villages. D eforestation and hunting may have affect ed game species at two differe nt spatial scales, the former at a large scale whereas the latter at a local scale. D eforestation and its related disturbances were the probable drivers of game depletion, possibly leading to the extirpation of sensitive species. Pano hunting pressure might only affect wild life close to villages by changing popula tion dispersion, although, not necessarily depleting them. Vulnerable villages and Indigenous Lands with low defensibility against external threats had more depleted wildlife than protected ones. Non -indigenous hunting and deforestation outside Indigenous Lands probably negatively affect ed wildlife hunted by Pano people. Other studies indicate hunting pressure as the main cause of depletion. The regional scale of a heterogeneous landscape intensively hunted for decades might help to explain the difference f ound in the drivers of game depletion across Pano villages. Deforestation can determine game conservation even in a little altered Amazon region. Indigenous Lands may be able to inhibit forest loss but not game depletion. The future scenario of reduced def orestation foreseen by the Acre State Government, depending on the execu tion of recent state policies, m an avoid the current trend of regional wildlife depletion as a result of habitat loss.
13 CHAPTER 1 INTRODUCTION The Bushmeat Crisis in the Amazon The rec ent decline in populations of large -bodied game species in tropical forests worldwide led to the recognition of a global bushmeat crisis", which threat ens fo rest ecosystems and its people (Milner -Gulland et al. 2003; Nasi et al. 2008). Large bodied mammal s provide most of the wildmeat consumed in the tropics (Jerozolimski and Peres 2003; Fa et al. 2005) and their survival is threatened mainly by habitat loss occurring at unprecedented rates (Brook et al. 2008; Davidson et al. 2009) and overhunting to suppl y rural and urban populations (Isaac and Cowlishaw 2004; Robinson and Bennett 2004; Cardillo et al. 2005; Peres and Palacios 2007). In terms of conservation policy, however, the important debate over the relative threats posed by deforestation and hunting pressure remains open, even though these factors may interact (Brook et al. 2008; Laurance and Useche 2009). Hunters in unaltered forests represent the single anthropogenic disturbance to game populations. Even indigenous people hunting for subsistence can locally deplete wildlife (Souza Mazurek et al. 2000; Peres and Nascimento 2006). It is controversial, though, if their impact alone deplete s game species at regional scales, given their centenary widespread use of wildlife (Robinson and Bodmer 1999; Smith 2005; Milner -Gulland and Rowcliffe 2007; Adeney et al. 2009; Levi et al. 2009; Rist et al. 2009). Several Neotropical indigenous groups increasingly hunt in regions that encompass both undisturbed and anthropogenic ally altered habitats (Smith 2005, 2008; Gavin 2007), and are potentially subject to multiple drivers of wildlife depletion. Yet many studies did not properly differentiate between these factors (Peres 2001; Daily et
14 al. 2003; Smith 2005; Fa 2007; Fa and Brown 2009) or were conducted at scales inappropriate to policy formulation, given that the effects of these drivers change depending on the scale of study (Palmer and White 1994; Fragoso 1999; Ferrier 2002; Fisher and Owens 2004; Pautasso 2007; Cardillo et al. 2008; Fritz et al. 2009). Wildlife Conservation in Amazonian Protected Areas Protected Areas in the Amazon are mostly established to prevent and combat threats to biodiversity (Margules and Presley 2000; Brooks et al. 2009). Although titled for different purposes, Indigenous Lands are fund amental for conservation (Schwartzman and Zimmerman 2005; Malhi et al. 2008) because they cover approximately 21% of the Brazilian Amazon (Carneiro Filho and Souza 2009). They figure among the most effective Amazonian Protected Areas in maintaining biodiversity, and inhibiting deforestation and forest degradation (Asner et al. 2005; Oliveira et al. 2007; Adeney et al. 2009) possibly better than other Conservation Units in Brazil (Nepstad et al. 2006). Nonetheless, the consequences of indigenous subsistence hunting ha ve been used to argue against the sustainable use policy inside Protected Areas (Redford and Sanderson 2000; Schwartzman et al. 2000; Peres and Zimmerman 2001; Schwartzman and Zimmerman 2005; Sunderland et al. 2008) in a debate that questions the conservation outcome of Protected Areas of sustainable use (Peres and Zimmerman 2001; Terborgh 2004; Brockington et al. 2006; Sirn 2006; Buck et al. 2007; Hames 2007). The degree to which Indigenous Lands are able to conserve wildlife remains unclear bec ause of the absence of available data at regional scale (Barreto et al. 2006; Nepstad et al. 2006; Adeney et al. 2009).
15 Pano People and Wildlife in Acre The debate and issues presented above is exemplified by the Pano Indigenous Lands in the Brazilian Stat e of Acre, southwestern Amazon, a region characteri zed by increasing deforestation ( yet low in compar ison to other Brazilian states; Acre 2009; Hayashi et al. 2009), elevated hunting pressure (Peres and Palacios 2007) but where most large game species are still hunted (Calouro 2005; Ramos 2005; Constantino et al. 2008). Indigenous Lands are an important component of Acre environmental policy and conservation strategy, covering 14% of the s tate (Iglesias and Aquino 2006; Acre 2008). The Kaxinaw and Katukina are peoples of the Pano linguistic family the most populous group in the region, for whom large game species are the preferred source of meat (Kensinger 1975, 1983; Deshayes 1986; Aquino and Iglesias 1994; Cunha and Almeida 2002; Amaral 2005). The reduced abundance of preferred game species in some Indigenous Lands, however, shifts reliance to alternative meats, purchased in city markets or raised in villages (Lima 2001; Calouro 2007). It has been suggested that the Kaxinaw hunting pressure had extirpated several species ( Peres and Zimmerman 2001), and these data were used to understand hunting impacts on wildlife in the Amazon basin (Peres 2000; Peres and Palacios 2007). Other evidence suggest s however, that some Kaxinaw Indigenous Lands have allowed r ecovery of several depleted game species (Constantino et al. 2008). Assessing Wildlife Abundance Using Hunting Data Human -mediated disturbances can induce three main negative responses of game assemblages: 1) reduction in population abundance, 2) species e xtirpation, and 3) dispersion of individuals and populations. However, t his is not exhaustive (e.g. some
16 species can increase abundance, or animals can differ in stress levels, body conditions, etc.) Surveying vertebrates by direct observation is hard and expensive in tropical forests (Naughton-Treves et al. 2003; Fa et al. 2005; Norris et al. 2008) especially in harvested areas where animals are wary of people (Fitzgibbon 1998), imposing limitations on this method (Carrillo et al. 2000). Consequently, re search on hunter societies and wildlife conservation, often uses hunting data and a variety of indicators to assess population abundance as well as identify its drivers (Appendix A; Juste et al. 1995; Souza Mazurek et al. 2000; Nagaoka 2002; Rowcliffe et al. 2003; Fa et al. 2004; Crookes et al. 2005; Franzen 2006; Albrechtsen et al. 2007; Ohl -Schacherer et al. 2007; Norris et al. 2008; Levi et al. 2009). The central place foraging model links resource depletion to local harvesting intensity. This model assu mes that hunting pressure within a limited territory departing from a fixed settlement first deplete s wildlife close to the settlement subsequently increasing the distance where species sensitive to hunting are caught from the settlement (Sirn et a. 2004 ; Levy et al. 2009). Therefore, the distance that desirable species are hunted from settlements indicates the wildlife status in response specifically to central place hunting pressure (Alvard et al. 1997; Hill et al. 1997; Cannon 2003; Ohl Schacherer et a l. 2007; Smith 2008; Levi et al. 2009). The status of wildlife population can also be assessed by indicators derived from Optimal Foraging Theory. Crudely, this theory predicts that hunters hunt more animals of lower ranked species as the relative cost of hunting them declines compared to higher ranked species (Hames and Vickers, 1982; Smith 1983; Winterhalder and Lu
17 1997; Rowcliffe et al. 2003). The addition of new species to, and increasing frequencies of less preferred species in hunters prey profile is a response to the decline in abundance of preferred species (Smith 1983; Rowcliffe et al. 2005; Parry et al. 2009a). This framework has been applied to assess the effects of indigenous and non indigenous Amazonian hunting (Alvard 1993, 1995; Hames and V ickers 1982, Jerozolimski and Peres 2003). The premises of Optimal Foraging Theory applied to hunters that prefer large-bodied animals allowed researchers to use the mean weight of hunted animals to assess the variation of wildlife status (Jerozolimski and Peres 2003; Fa et al. 2005; Franzen 2006; Ohl Schacherer et al. 2007; Constantino et al. 2008; Parry et al. 2009a). The capture-per unit effort (CPUE) of game species is often used to evaluate variation in populations' abundance, given that hunters have t o spend more time hunting in depleted sites t o have similar return rates compared to non depleted sites (Souza Mazurek et al. 2000; Hill et al. 2003; Sirn et al. 2004; Puertas and Bodmer 2005). While mean prey weight indicates the status of a group of hig h ranked species, CPUE allows the assessment of the abundance of individual species (Sirn et al. 2004; Parry et al. 2009a, b). Nevertheless, CPUE is a measurement of effort and should be carefully analyzed in order to provide estimates of wildlife status (Rist et al. 2008). In the extreme cases of extirpation or drastic reduction in density, the species is no longer present in the hunters prey assemblage. Certain large primates, birds and ungulates in the Amazon are less resilient to habitat disturbance and hunting pressure, and thus, more prone to extirpation (Robinson 1996; Peres 2000; Peres et al. 2003; Peres and Nascimento 2006; Michalski and Peres 2007; Barlow et al. 2007; Peres and
18 Palacios 2007; Boyle 2008; Takahashi 2008; Parry et al. 2009b). These species are often used as indicators of wildlife community status in the Neotropics (Daily et al. 2003; Parry et al. 2007). Given that these species are ranked high among Amazonian hunters, their absence in prey profiles may be a proxy of wildlife depletion close to hunter settlements instead of a consequence of narrowing diet breadth expected by Optimum Foraging Theory in cases of hi gh abundances of preferred prey (Smith 1983). Objectives In this study I analyzed the regional scale effects of deforestati on and hunting on wildlife consumed in Indigenous Lands based on hunting data from Pano villages in Acre. I also related wildlife status to the level of defensibility of Indigenous Lands against nonindigenous disturbances. I developed indicators based on ecological models to assess different perspectives on wildlife responses to disturbances, and used multivariate indicators to combine these responses. Findings are discussed regarding the recent environmental and development political agenda of Acre State. Research Questions 1 What are the drivers of variation of wildlife hunted by Pano people at the village level? 2 Are Pano Indigenous Lands able to conserve wildlife against surrounding threats?
19 CHAPTER 2 METHODOLOGY Study Area The eight Pano Indigenous Lands studied in the Brazilian State of Acre, southwestern Amazon, were titled between the 1980s and 2002. They are located in the Alto Juru and Alto Purus valleys in Acre S t ate, specifically in two tributaries of the Juru river, the Breu and Tarauac, and tw o tributaries of the Tarauac river, the Jordo and Humait rivers, as well as along the Purus river (Figure 2 -1). Approximately 4500 people live in these Indigenous Lands comprising about 50% of titled areas for Pano people in Acre. Villages vary in thei r degree of linkage to markets in the most populous cities of Cruzeiro do Sul, Tarauac and Feij. The small urban centers of Jordo, Santa Rosa do Purus and Marechal Thaumaturgo have some of the lowest HDI in Brazil (PNUD 2003). Villages in Campinas and I garap do Caucho Indigenous Lands have road access to cities, whilst other villages access market by boat (Figure 2 -1). The region is embedded in the Southwestern Amazon Moist Forest ecoregion (TNC 2009), one of the priority ecoregions for Neotropical terr estrial vertebrate conservation, in particular for endemic species (Loyola et al. 2009). Predominant vegetation can be characterized as typical Terra Firme forest, dominated by open canopy with palm trees and natural patches of bamboo in clay soils (Silvei ra et al. 2002; Acre 2006). The exception is the Alto Purus Indigenous Land that has portions of the Iquitos Vrzea ecoregion. Precipitation ranges from 1600 to 2000 mm/year (Sombroek 2001), causing little variation in ungulate abundances (Robinson and Ben nett 2004; Mandujano and Naranjo 2010).
20 Acre Ecological -Economic Zoning The Pano Indigenous Lands are under low threat compar ed to other Indigenous Lands in the Amazon (Carneiro Fil ho and Souza 2009). Within the s tate of Acre, however, Indigenous Lands ar e located in zones of different land use and development planning. The State Ecological Economic Zoning indicated four major zones of planning, three of which are relevant to this study. The area surrounding the highways composes Zone 1, characterized as t he development frontier of the s tate. Zone 2 encompasses all Protected Areas, including Indigenous Lands. Zone 4 comprises urban areas and its surroundings, sometimes including Indigenous Lands (Acre 2000). Beyond this zoning, the studied region can be spatialized as: 1) the development frontier alongside the BR -364, comprising the largest cities and the Campinas, Igarap do Caucho and Praia do Carapan Indigenous Lands, and 2) the international frontier with Peru, comprising most Protected Areas and the ot her Pano Indigenous Lands (Figure 21). The mos aic of Protected Areas along the border with Peru composes the Western Amazon Ecological Corridor (Brackelaire 2005). Pano Hunting The Kaxinaw and Katukina share cultural similarities, general to most Pano pe ople. Hunting is a prestigious male activity and wildlife is fundamental to structure communities through meat sharing, providing meat, and is strongly present in medicine and cosmology (Kensinger 1975, 1995; Almeida et al. 2002; Lima 2002; Aquino and Igl esias 1994; Lagrou 2004). Hunters prefer larger animals (Table 21) such as ungulates, large primates, reptiles, and understory birds, with few exceptions, that are hunted whenever encountered. Smaller species of secondary preference are hunted incidentall y when hunters do not succeed in hunting preferred animals. The most
21 common hunting strategy is the single man one -day search for animals following defined tracks and its surroundings (Deshayes 1986; Navarro 2004; Constantino et al. 2008). The s hotgun has substituted traditional bow andarrows for hunting. The ideal hunting territory of an isolated Pano village has a circular shape smaller than 6km from the center of the village, which is exclusively used by village members (Kensinger 1975; Navarro 2004). Although the actual Pano hunting territory is modified by the presence of nearby villages or Indigenous Lands boundary limits to reduce overlap between adjacent villages, the maximum distance of one day hunts does not go beyond 6km (Bant and Pessoa 2008) One way Pano hunters evaluate wildlife abundance is estimating the straight distance, in minutes walking, to where the highest number of preferred animals is hunted (Kensinger 1975; Constantino et al. 2008). Data Collection and Sources Hunting Data Kaxinaw and Katukina representatives participate in a long-term capacity building program for environmental management led by the local NGO Comisso Pr-ndio do Acre (CPI -AC). These representatives are paid by the State Government to provide extension work as I ndigenous Agroforestry Agents (IAAs) in their communities. The wildlife monitoring program was added to the capacity building program in 2004 and is facilitated by the IAA who engages other community members (Constantino et al. 2008). In Acre, 45 IAAs col lect hunting data in spreadsheets containing hunting effort (date, number of hunters, and time spent hunting) and success (species, weight, sex, approximate age, and straight distance where caught from the village). Between 2004 and 2009 the total number o f hunted animals recorded by these villages was 13,540. Out of this population, I selected 35 villages (33 Kaxinaw and two Katukina) from eight
22 Indigenous Lands, corresponding to approximately 33% of all Kaxinaw and 28% of all Katukina villages in Acre excluding villages according to the following criteria: 1) hunting records only to the first year of monitoring training, 2) less than six months of hunting records and less than 50 hunted animals recorded, 3) only records on preferred species hunted. The 35 selected villages recorded hunting 9109 animals between 2005 and 2009. I surveyed the georreferenced locations and recorded the IAA distance estimation in minutes of 65 hunted animals and landscape features that orient the indigenous hunts in 2006 and 20 09 using GPS Garmin Csx60, and used the relationship to convert the distance of hunted animals from the village from minutes to meters. Socio -economic Data I used socioeconomic data (e.g. population, number of hunters and employees, permanent goods, lives tock etc.) collected from villages in Acre by CPI -AC since 2004. In addition, I updated the information for 17 communities during workshops held in villages between April and July 2009. Spatial Data Deforestation maps were obtained from the Program for Mon itoring Deforestation in the Brazilian Amazon (PRODES; http://www.dpi.inpe.br/prodesdigital/prodes.php). PRODES information estimates a cumulative increase in deforested area, and hence, do es not account for forest regrowth (INPE 2005). Nevertheless, PRODES provides the best dataset available in a regional scale for the area (DeFries et al. 2005) and has been widely used in land use research (Asner et al. 2005; Morton et al. 2006; Arago et al. 2008; Broadbent et al. 2008; Broich et al. 2009) and planning (Acre 2009). The analyses of wildlife status are, thus, related to original forest cover loss in the period compatible to hunting data, and not necessarily to the current non-forest area. I
23 est imated zero deforestati on in the Peruvian border near the four villages located in the Breu Indigenous Land (Figure 2-1) because no geographic data on deforestation was available. This is reasonable since this is one of the most remote and unchanged fores t in Peru (CDC -UNALM 2006; Oliveira et al. 2007; Constantino P.A.L, personal observation). Spatial analyses were performed using ArcMap 9.3. I used a 5km radius buffer to calculate the deforestation and density of indigenous people near each Pano village. This area corresponds to the current ideal Pano hunting territory (Constantino P.A.L, unpublished data) All villages hunt most of their preys within these limits (Appendix A). Other studies estimate the same hunting area for one day hunts of indigenous an d non-indigenous Amazonian groups (Stearman 1990; Bonaudo et al. 2005). Wildlife Status Indicators I assessed wildlife status in response to anthropogenic disturbance across the 35 villages using three simple indicators supported by ecological theories pre sented above: 1) mean prey weight of hunted animals, 2) richness of human-sensitive species, and 3) mode of straight distance at which preferred animals were caught from the village. In addition, I used the capture per unit effort (CPUE) of hunted animals of preferred species, in terms of biomass, as another indication of overall abundance variation for 21 villages (Appendix B ). The distance of hunted animals to the village is the only indicator assuming that hunting pressure is the main driver of wildlife variation (Levi et al. 2009). Although the other indicators attempt to reflect variation in wildlife status through species availability to hunters, they do not assume a priori the underlying cause of wildlife variation (Rist et al. 2009). Hence, the four indicators would reflect the same trend if hunting pressure was the main factor driving wildlife population variation. All
24 indicators were normally distributed. I conducted pairwise Pearsons correlations between the four indicators of wildlife status to a ssess the difference in information provided. None of these indicators were correlated with monitoring effort (all 0.1
25 Peres 2003) Nonetheless, indicators of wildlife status based on ecological models using hunting data are widely reported in the literature. In addition to the simple indicators I created multivariate indicators that capture the joint response of wildlife assemblage. These indicators resulted from a Principal Component Analysis that considered the mean prey weight, number of sensitive species and the mode distance of animals of the 35 villages. Principal components with eigenvectors corresponding to more than 80% cumulative variation in the data were selected. CPUE was not included because of missing data. Statistical analysis Selecting Appropriate Indicator of Distance of Preferred Hunted Animals to the Village The estimation of wildlife distance to the village has been assessed elsewhere using the mean distance of hunted animals (Ohl -Schacherer et al. 2007). Instead, I selected the mode of the distance where animals of preferred species were hunted fro m the village), because data were not normally distributed (Appendix A). This distribution is likely to be a result of hunting pressure that is higher close to the village center (Levi et al. 2009). I estimated mode distance of hunted animals using the Gaussian kernel density estimator, a statistical technique often applied to study animals' use of space (Seaman and Powell 1996; Gitzen et al. 2006). Dri vers of Wildlife Depletion across Pano Communities Linear multiple regression I used multiple linear regressions to identify the drivers of wildlife status across Pano villages. I regressed the six wildlife status indicators (i.e. four simple indicators and two multiple indicators) against explanatory variables related to hunting pressure,
26 land use change and socio economic condition of Pano communities (Appendix C). The explanatory variables were transformed (i.e. log 10 or sq uare root) when needed to achieve normal dis tribution. In order to avoid multicolinearity, I conducted pairwise Pearson's correlations between the 12 continuous explanatory variables before running the regression models selecting for variables with fewer correlations and more accurately measured. The remaining variables used to build the full models were: the density of indigenous people in a 5km radius; village age; a ctual indigenous population in the village (hunting pressure); animal unit (socio economi c condition); and log 10 deforested area (land use). The nominal variable presence of the road was also used in model building. The former three variables were used as proxy to hunting pressure and suggested to drive wildlife to decline elsewhere (Jerozol imski and Peres 2003; Brashares et al. 2001). I selected the minimal model for each wildlife indicator excluding unrelated explanatory variables from the full model (p>0.1) through backward stepwise elimination. I supplemented the backward selection with t he analysis of all possible subset models (Neter et al. 1985) checking for AICc (considering a variation in 2.0 as significant), the consistency in variables entrance in models, and the signals of estimates. I selected all the possible final models for eac h response variables. This additional procedure reduces the selection of spurious patterns that can be present in multiple regressions with low ratio of response/explanatory variable s Spatial dependency The home range of some important game speci es (e g. white -lipped peccary) can be as large as the 5km radius area used to estimate deforestation and density of indigenous people (Fragoso 1998; Reina -Hurtado et al. 2009) As a consequence, t he
27 spatial dist ribution of villages could give rise to spatial depend ence of wildlife status indicators and induce erroneous conclusions regarding drivers of wildlife variation. I used the Morans I test for spatial dependence to analyze of wildlife indicators and residuals of minimal regression models using ArcMap 9.3. Cap acity of i ndigenous l ands to protect wildlife against external threats I analyzed the e ffects of surrounding disturbances on wildlife hunted in Pano villages in two levels: the village level and the Indigenous Land level At the village level, those villages with a portion of the 5km radius out side of Protected Areas were considered vulnerable to external threats, whilst others were considered protected. Student t -tests were conducted to evaluate differences of wildlife indicators between these two groups. Nonetheless, village vulnerability to external threats may be a result of the poor design and location of Indigenous Lands. Therefore, at the Indigenous Land level, I followed Peres and Terborghs (1995) categorization of reserve defensibility to label In digenous Lands according to their design and location in the s tate (Table 2 1). I grouped villages pertaining to Indigenous Lands as low, medium, or high defensibility against external threats. I tested the differences of wildlife status between these grou ps using ANOVA and Tukey Kramer HSD tests. All statistical analyses were performed in the software JMP8. In addition, I compared the size of Pano Indigenous Lands to the minimum area required to maintain sustainable harvest (Peres 2001) of eight preferred species, considering the annual per capita harvest rates of local people throughout the Amazon.
28 Figure 21. Kaxinaw and Katukina Indigenous Lands studied in Acre State, Brazilian Amazon. Label numbers correspond to identificat ion in Table 3 -4. Indigenous Lands studied (yellow), other Indigenous Lands (beige), other Conservation Units (green), cities (crosses). Indigenous Lands labels: a) Kaxinaw do Igarap do Cauho, b) Katukina do Campinas, c) Kaxinaw da Praia do Carapan, d ) Kaxinaw do Baixo Rio Jordo, e) Kaxinaw do Rio Jordo, f) Kaxinaw/Ashaninka do Rio Breu, g) Kaxinaw do Humait, and h) Alto Purus. Close up illustrates village locations in the Kaxinaw do Baixo Rio Jordo and Rio Jordo Indigenous Lands.
29 T able 2 1. Defensibility of Pano Indigenous Lands against external threats. Indigenous Land Size (km 2 ) Defensibility Design features to include in the category Location in State Igarap do Caucho 123.18 Low small size, road presence, unprotected surround, near major city development frontier Campinas 326.24 Low small size, road presence, two access points, partially unprotected surround, near major city development frontier Praia do Carapan 606.98 Low two access point, intense river traffic, unprotected surround d evelopment frontier Baixo Jordo 87.26 Medium small size, partially protected surround western Amazon ecological corridor Humait 1273.83 Medium large size, unprotected surround, watershed headwaters western Amazon ecological corridor Alto Purus 2631.3 Medium large size, partially protected surround, two access points, intense river traffic western Amazon ecological corridor Jordo 872.93 High medium size, totally protected surround, watershed headwaters western Amazon ecological corridor Breu 312.77 H igh totally protected surround, watershed headwaters, remoteness western Amazon ecological corridor
30 CHAPTER 3 RESULTS Pano Hunters Prey Profiles The Kaxinaw and Katukina communi ties hunted 54 wildlife species or taxa (e.g. Dasypodidae). L arge bodied preferred vertebrate s were the most hunted and provided the largest amount of meat (Table 31). The 3375 hunted ungulates provided more meat than other taxa to the whole Pano population ( Table 3 2 ). Ungulates were most hunted in almost all villages with wh ite lipped peccary being the most hunted among them, and provided up to 60% of wildmeat in some villages ( Figure 3 1 ). Rodents and primates were second in terms of animals hunted and meat contribution ( Table 3-2 ), especially because of the presence of larg e -bodie d species. Large birds were frequently hunted but less important in providing meat ( Table 3-2 ). Species of secondary preference that were frequently hunted include paca, agouti, armadillos, capuchin monkey, squirrel s and coati (Table 3-1) but only p rovided 13% of total consumed meat. Human -sensitive species were absent in hunters prey profiles of several villages, particularly in those located in the development frontier (Table 3 3). Villages alongside roads did not hunt large monkeys or birds. On t he other hand, white-lipped peccary and howler monkey were hunted in almost all villages. Information Provided by Indicators of Wildlife Status Three indicators of wildlife status (i.e. mean prey weight, richness of sensitive species, and CPUE of preferred species) were significantly correlated among each other, representing a pattern in which the average weight of preys decreased associated to the loss of sensitive species, reflecting in the decreased CPUE of preferred animals. In turn, the mode distance o f preferred animals to the village was not
31 correlated with other indicators, suggesting a different process causing its variation (Table 34). As suggested by the Optimal Foraging Theory, the variation in mean prey weight represented the positive variation in the proportion of preferred species, specifically ungulates, simultaneously to the negative variation in the proportion of species of intermediate preference and low ranked species, specially primates ( Figure 32 a). Furthermore, the variation in CPUE of preferred species was positively associated to the proportion of hunted animals, in particular the tapir and all ungulate species, and negatively to intermediate and low ranked species, specially rodents ( Figure 32 b). Piping guan, razor -billed curassow woolly and spider monkey, and tapir were frequently absent from Pano catchments. Villages that hunted four or less sensitive species never hunted the piping guan, razor -billed curassow, and tapir, whilst only one hunted the spider monkey and four the wooly monkey. The two Principal Components (PCs), created from the multivariate analysis, correspond to 86.55% of the variation present in simple indicators. PC1 reflects the simultaneous variation in mean prey weight and sensitive species richness, with litt le contribution of mode distance. PC2 reflects the simultaneous variation of mode distance in opposition to sensitive species richness, where villages that have hunted animals farther away and hunted a small number of sensitive species (Table 3 -5). Drivers of Wildlife Depletion a cross Pano Communities Only hunting pressure represented by the density of indigenous people, and road presence composed the single significant minimal regression model, explaining 30% of variation in mode distance of preferred ani mals hunted away from villages (Table 3-6; Table 3 7). According to this model, an addition of 100 Pano people in 5km represented
32 an increase of 8.4 minutes distance (app. 230m) to the sites where preferred animals are more frequently hunted. Hunters in roadside villages hunted animals one hour (app. 1715m) farther away from village centers than riverside villages (Figure 3-3). Deforestation had no association with mode distance of preferred animals from village center (Figure 34). Only deforestation and presence of roads composed the single significant minimal model, explaining 50% of mean prey weight variation (Table 3 -6; Table 3 -7). Clearing the first 10 hectares of forest near villages resulted in a drop of app. 3kg in prey size. The rate of reducing pr ey size decreased as deforestation increased, probably because smaller species are more resilient to habitat change and can continuously provide meat. Hunters in roadside villages harvested animals an average of 6kg larger than riverside villages (Figure 3-3), a wildlife response contrary to the expected. Roads usually cause reduction in wildlife abundances and richness (Laurance et al. 2006, 2008). A possible explanation is that the mean weight of prey hunted in Pano communities alongside roads does not re flect game abundance but rather the access hunters have to alternative meat purchased in city markets. Hunting pressure, in turn, had no association with mean prey weight (Figure 34). The minimal models for CPUE of preferred species and number of sensitiv e species hunted were composed only by deforestation that explained 30% and 56% of variation, respectiv ely (Table 3 6; Table 37). L osing the first 10 hectares resulted in the absence of more than two sensitive species in prey profiles. The rate of species absence decreased as deforestation increased (Figure 3-3). In terms of hunting success, hunters caught 0.65kg/hour less after the loss of the first 10 hectares of forest
33 near villages, a rate that decreased as deforestation increased (Figure 33). Hunting pressure had no relationship with CPUE or sensitive species richness (Figure 34). The minimal model explaining 55% of variation in PC1 (Table 3 6) was composed by deforestation, presence of road, and the density of indigenous people although the latter w 7). According to this model, indigenous hunting and deforestation around villages have additive effect s on wildlife that responded by decreasing the abundance of large animals and richness of s ensitive species, while increasing the distance of preferred animals from the village. Nevertheless, indige nous hunting seems to have less effect on this multiple response. In turn, PC2 was driven by road presence and indigenous density in a model explaining 50% of variation. Thus, indigenous hunting is negatively related to the dispersion of animal s far away from villages and, to a smaller extent the reduction of sensitive species richness, a process exacerbated by the presence of roads (Figure 33). Nevertheless, PC2 represent s a smaller portion of wildlife multiple response. Because the variables describing the two primary drivers of game depletion, deforestation and hunting pressure (i.e. village age, density of indigenous people, and actual village population) were not correl ated, it was possible to assess the importance of the effect of each factor on wildlife status. The regression results suggested that deforestation and indigenous hunting induced different responses of wildlife species and the road had a significant impac t exacerbating effects of both drivers. Hunting pressure affected populations of preferred game species by displacing them away from village. Deforestation is the only factor that significantly led to an increase in the effort needed to obtain meat of pref erred species, induced shifts in large prey abundances,
34 and was the main factor responsible for the decline in the richness of human -sensitive species. Spatial Dependence The actual values of mean prey weight, sensitive species richness and PC1 were autoco rrelated, but not the residuals of the minimal regression models (Table 38). Hence, neighbor villages have wildlife status es more similar than those farther apart Deforestation across villages explained spatial dependency because this was the only variable included in regression models, which residuals were not autocorrelated. Given that the actual values and the residuals of regression model of mode distance of hunted animals were not spatially dependent, indigenous hunting that affects wildlife distribution is apparently a localized process within 5km around villages. It was not possible to test spatial dependence of CPUE due to the reduced data points. Capacity of Indigenous Lands to Protect Wildlife a gainst External Threats The results from regression models clearly showed a difference of wildlife status between protected villages and those vulnerable to threats from outside Indigenous Lands ( Figure 3 -3 ). Hunters in vulnerable villages hunted animals on average 2kg smaller, one sensitive species less, a nd had three times less hunting success than those in protected villages. The distance where most of the animals were hunted was not significantly different between the two groups (Table 39). Deforestation was the only explanatory variable used in regress ion models significantly higher in vulnerable villages than in protected (Table 3 9), suggesting that deforestation is possibly the factor creating differences of wildlife status between protected villages a nd those vulnerable to external threats. The absence of difference in mode distance between the groups of villages provides additional evidence that indigenous hunting pressure has only
35 localized consequences to wildlife. Part of the deforested area affecting game hunted in vulnerable villages is outside the boundaries of Indigenous Lands. Across these villages, deforestation inside Indigenous Lands was not correlated to the variation of any wil dlife status indicator, whereas total deforestation and deforestation near villages but outside Indigenous Lands were (Table 3 -10). Another clear pattern is the spatial distribution of vulnerable and protected villages within Indigenous Lands of different levels of defensibility against external threats. Vulnerable villages were all located in Indigenous Lands with low or medium defensibility whereas protected villages were located in reserves with high defensibility, except for one protected village in a reserve with medium defensibility ( Table 2 1 ; Table 3 -3 ). Hunters in Indigenous Lands of low defensibility hunted prey twice as small (F=3.47, p<0.05) and twice as many less sensitive species (F=15.59, p<0.01) than hunters in reserves of medium or high defensibility. Those hunters also had almost four times lower hunting success than hunters in Indigenous Lands of high defensibility (F=5.68, p<0.05; Figure 35). Hence, Indigenous Lands better protected against external threats have wildlife in a better status. The distance where animals are caught was not different among Indigenous Lands of different levels of defen sibility (F=0.22, p>0.1; Figure 35), suggesting again that the process driving this wildlife response occurs at a smaller spatial scale. In addition, the size of the reserve seems to be of less importance to Indigenous Lands, at least, with partially prot ected surroundings and located out of the development frontier. According to Peres (2001), hunting sustainability is associated to the size of protected areas because these should be able to maintain viable populations of game
36 species at the harvest level of subsistence hunting of Amazonian people. Analyzing Pano Indigenous Lands according to this rationale, Alto Purus is the only Indigenous Land large enough to guarantee alone sustainably harvested populations of all preferred species. If reserve size were the most important factor determining the presence of species, Jordo Indigenous Land would not be able to maintain the harvest of tapirs, while, Breu Indigenous Land would support harvested populations of tapir, howler and woolly monkey and curassow (Table 311). However, these species are still hunted in these reserves (Table 3-3). Although Baixo Jordo Indigenous Land would only harvest collared peccary according to its small size (Table 311), hunters heavily relied on other preferred species (Figure 3 -1). Indigenous Lands size, year of title, and Pano population were not correlated to any of the wildlife status indicators (all 1.5
37 Table 3 1. Wildlife species hunted by Pano communities in Acre, Brazilian Amazon, during this study. Preferred species (1), secondary species (2), low ranked species (3) indicated according to Cunha and Almeida (2002) and Constantino et al. (2008). For some analysis in this study we collapsed groups 2 and 3 in the group of species of secondary preference. Taxa referenced with superscript letters were not always recorded at the species level or identification was not certain. Scientific name Common name # of animals Weight of animals Preference Sensitive species ref. Tayassu pecari white lipped peccary 1335 24455.6 1 1 Pecari tajacu collared peccary 1267 1 6049.0 1 Mazama americana red brocket deer 695 15381.8 1 Tapirus terrestris lowland tapir 68 6786.0 1 2 Alouatta seniculus howler monkey 539 3343.5 1 3 Geochelonia denticulata tortoise 536 3161.3 1 4 Ateles chamek spider monkey 258 1937.5 1 5 Caima n sp. cayman 177 1614.5 1 Lagotrix lagotricha woolly monkey 72 536.2 1 6 Mitu tuberosum razor billed currasow 96 457.1 1 7 Mazama gouazabira gray brocket deer 10 136.0 1 Pipile cujubi piping guan 18 37.5 1 8 Agouti paca paca 524 3933.1 2 Dasypodid ae a armadillo 427 2368.5 2 Dasiprocta fuliginosa agouti 411 1812.2 2 Nasua nasua coati 313 1450.8 2 Cebus apela brown capuchin 266 1031.1 2 Sciurus sp. b squirrel 636 657.5 2 Penelope jacquacu Spix's guan 228 421.7 2 Tinamus guttatus large tinam ous 200 326.1 2 Hydrochaeris hydrochaeris capybara 13 291.0 3 Psophia leucoptera pale winged trumpeter 132 251.3 2 Dinomys branickii pacarana 31 236.0 3 Pithecia sp. c saki monkey 68 203.0 3 Myoprocta pratii acouchi 162 183.8 3 Cebus albifrons w hite fronted capuchin 45 162.5 3 Aotus nigriceps night monkey 88 119.8 3
38 Table 3 1. Continued Scientific name Common name # of animals Weight of animals Preference Sensitive species ref. Priodontes maximus giant armadillo 4 95.0 3 Myrmecophaga trida ctyla giant anteater 4 87.0 3 Ara sp. e macaw 53 74.9 3 Saimiri sciureus squirrel monkey 53 60.7 3 Tinamus tao grey tinamous 24 55.8 2 Callicebus moloch titi monkey 44 53.1 3 Saguinus sp d tamarin 53 44.2 3 Amazona sp. e parrot 34 31.8 3 Cryptu rellus cinereus small tinamous 30 21.6 3 Ramphastos sp. e toucan 26 20.5 3 Ortalis guttata Speckled chachalaca 23 16.8 3 Crypturellus soui little tinamous 25 16.3 3 Rallidae rail 23 15.0 3 Coendou sp. quandu 3 15.0 3 Tinamous major small tinamou s 15 12.5 2 Crypturellus sp.3 small tinamous 15 11.9 3 Crypturellus sp.2 small tinamous 18 11.6 3 Odontophorus sp. e wood quail 11 6.9 3 Jabiru mycteria jabiru 1 6.0 3 Harpia harpyja harpy eagle 1 5.0 3 Columbidae dove 14 4.3 3 Anatidae duck 3 4.0 3 Accipitridae kite 3 2.5 3 Tupinambis teguixin common tegu 2 2.0 3 Psarocolius sp. oropendola 5 1.9 3 Psittacidae macaw 5 1.9 3 Pteroglossus sp. aracari 2 0.9 3 Total 9109 88023.3 aDasypus novemcinctus Dasypus kapplery Cabassous uni cinctus bS. ignitus and S. spadiceus. cP. irrorata or P. monachus. dS. imperator S. fusicolor S. melanoleucus and possibly S. mystax. etaxa containing more than one species but with uncertain identification.
39 Table 3 2. Relative importance of prey taxa to Pano communities. Taxa Number of animals hunted % of offtake Amount of meat % of meat contribution ungulates 3375 37.05 62808.4 71.35 primates 1486 16.31 7491.6 8.51 rodents 1777 19.51 7113.6 8.08 reptiles 715 7.85 4777.8 5.43 birds 1005 11.03 1815.8 2.06 other 751 8.24 4016.3 4.56 total 9109 100 88023.5 100 Table 3 3. Characteristics of Pano villages in the State of Acre, Brazilian Amazon, and absent species. Map ref. Village External threat # months monitored # animals hunted # species hunted Absent sensitive species Igarap do Caucho 1 18 Praia* 1 8 50 8 2, 3, 4, 5, 6, 7, 8 Campinas 2 Samauma* a 1 11 158 18 2, 4, 5, 6, 7 8 3 Varinawa* a 1 11 103 9 2, 3, 5, 6, 7, 8 Praia do Carapan 4 Segredo do Arteso a 1 10 60 19 2, 5, 6, 7, 8 5 Agua Viva 1 10 115 26 1, 2, 5, 6, 7, 8 6 Goiania a 1 9 359 33 2, 4, 5, 6, 7, 8 7 Nova Vida 1 10 210 33 1, 2, 5, 6, 7, 8 8 Mibay 1 16 288 35 2, 5, 6, 7, 8 Baixo Jordo 9 Nova Extrema a 1 22 373 27 2, 7, 8 10 Nova Cachoeira 1 14 219 23 2, 4, 5, 7, 8 11 Nova Empresa a 1 14 196 20 7, 8 12 Nova Mina 1 16 158 19 2, 4, 5, 6, 7, 8 Jordo 13 Morada Nova 0 10 78 20 2, 5, 7, 8 14 Novo Astro 0 13 167 26 5, 7, 8 15 Sacado 0 10 104 22 2, 4, 5, 6, 7, 8 16 Boa Vista a 0 16 244 29 2, 7, 8 17 Nova Fortaleza 0 10 155 23 6, 7, 8 18 Nova Aliana 0 11 83 19 2, 5, 7, 8
40 Table 3 3. Continued. Map ref. Village External threat # months monitored # animals hunted # species hunted Absent sensitive species 19 Novo Natal 0 7 56 18 2, 7, 8 20 Chico Curumim 0 17 279 28 2, 7, 8 21 Bom Jesus 0 21 307 28 8 22 Verde Floresta a 0 22 278 17 6, 8 23 Belo Monte 0 28 529 30 6, 8 24 Paz do Senhor 0 8 126 26 2, 7, 8 25 Novo segredo 0 8 125 21 6 Breu 26 Vida Nova 0 30 927 41 none 27 Cruzeirinho 0 18 321 29 none 28 Japinim 0 45 1423 38 none 29 Jacobina 0 8 492 28 none Humait 30 Boa Sorte a 1 10 99 18 2, 6, 8 31 Boa Vista a 1 9 133 12 2 6 32 Porto Brasil a 0 16 126 17 6, 7 Alto Purus 33 Nova Fronteira a 1 18 373 35 6, 7, 8 34 Dois Irmos a 1 9 149 25 6, 8 35 Porto Rico a 1 9 124 20 6 Numbers refer to village location in Figure 2 1. Villages subject to external threat are N umbers refer to species numbers in column Sensitive species of Table 2 1. *villages accessing city market by roads. avillages excluded from CPUE analysis.
41 Table 3 4. Pearson correlation matrix between simple indicator variables of wildlife depletion in Pano villages (upper) and excluding roadside villages (bottom) in Acre, Brazilian Amazon. Simple indicators Mean prey weight (kg) CPUE preferred species (kg/hour) Number of sensitive species Mean prey weight (kg) n=35 1 CPUE of preferred species (kg/ho ur) n=21 0.83** 0.82** 1 Number of sensitive species n=35 0.44* 0.73** 0.62* 0.59* 1 Mode distance of preferred species hunted from the village (min.) n=35 0.33 0.19 0.00 0.14 0.09 0.18 *p<0.01;**p<0.001 Table 3 5. Principal Component Analysis axis of wildlife abundance simple indicators. PC1 PC2 PC3 Axis Percent (%) 50.45 36.01 13.46 Components Mean prey weight 0.73 0.05 -0.68 Richness of sensitive species 0.57 -0.59 0.57 Mode distance 0.37 0.81 0.46
42 Table 3 6. Backward stepwise regre ssion models for all response variables at the village level. Response variable R 2 adj DF F ratio p SSE final AICc full model AICc minimal model n Simple indicators Mean prey weight (kg) 0.50 2 18.26 <0.001 119.17 51.79 48.88 35 CPUE preferred species (kg/hour) 0.30 1 9.67 0.006 7.63 -11.99 -17.27 21 Richness of sensitive species 0.56 1 44.8 <0.001 60.3 25.53 23.04 35 Mode distance of preferred species hunted from the village (min.) 0.30 2 8.42 0.001 39135.16 251.09 251.68 35 Multivariate i ndicators PC1 0.55 3 14.64 <0.001 21.29 -6.67 -9.39 35 PC2 0.50 2 18.08 <0.001 17.28 15.19 18.7 35
43 Table 3 7. Explanatory variables included in the final regression models after stepwise and supervised selection. Response variable Parameters Estimate SE t Ratio Prob>t Mean prey weight Intercept 18.93 1.68 11.29 <0.001 Log deforested area 3.36 0.64 14.67 <0.001 Road presence 3.00 0.62 5.44 <0.001 C PUE o f preferred species Intercept 2.84 0.46 6.13 <0.001 Log deforested area 0.65 0.21 3.11 0.006 Richness of sensitive species Intercept 10.36 0.89 11.69 <0.001 Log deforested area 2.63 0.39 6.69 <0.001 Mode distance of preferred species to village center Intercept 72.76 16.71 4.35 <0.001 Road presence 31.50 10.71 2.94 <0.01 Density of indigenous people 6.60 2.83 2.33 0.03 PC1 Intercept 4.18 0.73 5.72 <0.001 Road presence 0.97 0.28 3.53 0.001 Density of indigenous peo ple 0.14 0.07 2.00 0.055 Log deforested area 1.80 0.28 6.53 <0.001 PC2 Intercept 0.21 0.35 0.60 0.55 Road presence 1.07 0.22 4.78 <0.001 Density of indigenous people 0.17 0.06 2.81 0.008 Table 3 8. Spatial dependence of response variables and residuals of final regression models. Morans I index reported and its significance given by Z -scores in parenthesis. Wildlife indicators Actual values Residuals of minimal models n Mean prey weight 0.23 (2.57)* 0.04 (0.66) 35 Number of sensitive sp ecies 0.2 (2.3)* -0.04 ( 0.1) 35 Mode distance of preferred species to village center 0.06 (0.85) 0.08 (1.11) 35 PC1 0.24 (2.7)** 0.06 ( 0.35) 35 PC2 0.08 (1.07) 0.02 (0.11) 35
44 Table 3 9. Student's t tests on wildlife indicators between villages protected and vulnerable to external threat. Variables Vulnerable Protected mean (SD) n mean (SD) n t p response variable mean prey weight (kg) 8.11 (3.18) 17 10.05 (1.88) 18 2.17 <0.05 mean prey weight (kg)* 7.30 (2.46) 14 10.05 (1.88) 19 3.46 <0.001 richness of sensitive species 3.53 (1.80) 17 5.67 (1.71) 18 3.58 <0.001 CPUE of preferred species (kg/hour) 0.83 (0.63) 6 1.72 (0.66) 18 2.89 <0.01 mode distance of animals hunted from village center (min.) 80.12 (44. 51) 17 67.13 (39.51) 18 0.91 n.s. PC1 0.51 (1.32) 17 0.48 (0.94) 18 2.57 <0.005 PC1* 0.73 (1.25) 14 0.48 (0.94) 18 3.03 <0.005 PC2 0.43 (1.04) 17 0.40 (0.89) 18 2.53 n.s. mean (SD) n mean (SD) n t p explanatory variables Deforested ar ea (ha) 2.57 (0.49) 17 1.81 (0.41) 18 4.87 <0.001 Density of indigenous people (in 5km radius) 4.18 (2.30) 17 3.94 (2.06) 18 0.31 n.s. Village age (year) 11.47 (8.31) 17 14.61 (8.50) 18 1.10 n.s. Village population 83.53 (54.71) 17 79.17 (29.14) 18 0.29 n.s. *comparisons excluding the roadside villages of indicators subjected to road effects on access to alternative meat. Values of mean prey weight in these villages do not reflect wildlife abundance. Table 3 10. Pearson's correlations between indicators of wildlife status that were driven by deforestation and the total deforestation within 5km radius area, then inside and outside Indigenous Lands (IL) for the group of villages vulnerable to external threats. Roadside villages were excluded. Defor estation Total Inside IL Outside IL N Mean prey weight 0.63 (0.02)* 0.48 (0.09) 0.60 (0.04)* 13 Sensitive species richness 0.38 (0.19) 0.4 (0.20) 0.12 (0.70) 13 PC1 0.58 (0.04)* 0.58 (0.05) 0.43 (0.16) 13
45 Table 3 11 Pano Indigenous Lands with total area that would support the average local people harvesting of preferred species in the Amazon. Species Minimum required area (km2) a Indigenous Lands tapir 2003475 Purus howler monkey 864303 Purus, Humait, Jordo wool ly monkey 443132 Purus, Humait, Jordo, Praia do Carapan currasow 36967 Purus, Humait, Jordo, Praia do Carapan white lipped peccary 29899 all but Igarap do Caucho and Baixo Jordo spider monkey 25864 all but Igarap do Caucho and Baixo Jordo red brocket deer 11244 all but Baixo Jordo collared peccary 6919 all aEstimated meanSD minimum sustainable area for hunted species according to Peres (2001), based on estimated average annual harvest in the Amazon applied in Robinson and Redford (1991) model of hunting sustainability.
46 Figure 31 Relative contribution of wildlife species to Pano diet Black: white -lipped peccary; dark grey: brocket deer; medium grey: collared peccary; light grey: tapir; white: agouti. Column numbers refer to village numbers in Figure 21
47 Figure 32. Significant Pearson's correlation between wildlife status indicators and the proportion of individual species, major taxa, and preference group, after Bonferroni correction. A) Mean prey w eight. Only the proportion of ungulates and preferred species were positively correlated (r=0.69, p<0.000; r=0.67, p<0.000, respectively) and the proportion of primates, low ranked and secondary species (r= 0.46, p=0.02; r= 0.59, p=0.001; r= 0.62, p<0.001, respectively) were negatively correlated. B) CPUE of preferred species.
48 0 0.5 1 1.5 2 2.5 3 3.5 CPUE (kg/hour) 1 1.5 2 2.5 3 3.5 4 Log10 Deforested Area (ha) 0 0.5 1 1.5 2 2.5 3 3.5 CPUE (kg/hour) 1 1.5 2 2.5 3 3.5 4 Log10 Deforested Area (ha) 0 1 2 3 4 5 6 7 8 9 Sensitive species richness 1 1.5 2 2.5 3 3.5 4 Log10 Deforested Area (ha) 0 1 2 3 4 5 6 7 8 9 Sensitive species richness 1 1.5 2 2.5 3 3.5 4 Log10 Deforested Area (ha) 0 50 100 150 Mode distance (min.) 0 1 2 3 4 5 6 7 8 Indigenous density (people/km2) 0 50 100 150 Mode distance (min.) 0 1 2 3 4 5 6 7 8 Indigenous density (people/km2) -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 PC2 1 2 3 4 5 6 7 8 Indigenous density (people/km2) -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 PC2 1 2 3 4 5 6 7 8 Indigenous density (people/km2) 2.5 5 7.5 10 12.5 15 17.5 Mean prey weight (kg) 1 1.5 2 2.5 3 3.5 4 Log10 Deforested Area (ha) 2.5 5 7.5 10 12.5 15 17.5 Mean prey weight (kg) 1 1.5 2 2.5 3 3.5 4 Log10 Deforested Area (ha) Figure 33 Best linear regression models after backward stepwise selection (based on AICc) Protected villages (solid dots), vulnerable villages (open dots), roadside villages re gression line (bottom line), riverside villages regression line (upper line). PC1 could not be represented in two dimensional graph.
49 0 50 100 150 Mode distance (min.) 1 1.5 2 2.5 3 3.5 4 Deforested area (Log10 ha) 0 50 100 150 Mode distance (min.) 1 1.5 2 2.5 3 3.5 4 Deforested area (Log10 ha) 0 0.5 1 1.5 2 2.5 3 3.5 CPUE (kg/hour) 0 5 10 15 20 25 30 Village age (years) 0 0.5 1 1.5 2 2.5 3 3.5 CPUE (kg/hour) 0 5 10 15 20 25 30 Village age (years) 0 1 2 3 4 5 6 7 8 9 Sensitive species richness 0 5 10 15 20 25 30 Village age (years) 0 1 2 3 4 5 6 7 8 9 Sensitive species richness 0 5 10 15 20 25 30 Village age (years) 2.5 5 7.5 10 12.5 15 17.5 Mean prey weight (kg) 0 5 10 15 20 25 30 Village age (years) 2.5 5 7.5 10 12.5 15 17.5 Mean prey weight (kg) 0 5 10 15 20 25 30 Village age (years) 0 1 2 3 4 5 6 7 8 9 Sensitive species richness 0 1 2 3 4 5 6 7 8 Indigenous density (people/km) 0 1 2 3 4 5 6 7 8 9 Sensitive species richness 0 1 2 3 4 5 6 7 8 Indigenous density (people/km) 0 0.5 1 1.5 2 2.5 3 3.5 CPUE (kg/hour) 0 1 2 3 4 5 6 7 8 Indigenous density (people/km) 0 0.5 1 1.5 2 2.5 3 3.5 CPUE (kg/hour) 0 1 2 3 4 5 6 7 8 Indigenous density (people/km) 2.5 5 7.5 10 12.5 15 17.5 Mean prey weight (kg) 0 1 2 3 4 5 6 7 8 Indigneous density (people/km) 2.5 5 7.5 10 12.5 15 17.5 Mean prey weight (kg) 0 1 2 3 4 5 6 7 8 Indigneous density (people/km) Figure 34 Examples of non significant Pearson's correlation between hunting pressure, deforestation and indicators of wildlife status. Vulnerable villages (open dots) and protected villages (solid dots). Mean prey weight and hunting pressure (rvillage age=0.14, p=0.42; rindigenous density=0.04, p=0.83), CPUE and hunting pressure (rvillage age= -0.15, p=0.53, rindigenous d ensity= 0.31, p=0.18), sensitive species richness and hunting pressure (rvillage age= 0.05, p=0.79, rindigenous density= 0.22, p=0.21), and mode distance and deforestation (r=0.09, p=0.89).
50 Figure 35. Box -plot of wildlife ind icators among Indigenous Lands of different levels of defensibility against external threats to wildlife: low (n=8 villages), medium (n=10 villages), and high (n=17). Roadside villages were excluded from mean prey weight analysis. Vulnerable villages (open dots) and protected villages (dark dots). Letters indicate significantly different means (p<0.05). Indigenous Lands within each category are listed in Table 21.
51 CHAPTER 4 DISCUSSION Wildlife Status nearby Pano Communities Most Pano communities still reli ed on preferential animals as a meat source, even after many decades of intense wildlife harvesting and high c urrent hunting pressure in the s tate of Acre. The pattern of game use by Pano communities was similar to those observed in tropical forests at smaller and larger spatial scales. In the Alto Purus region of Peru w here wildlife is less dep leted, ungulates predominate in the diet of Kaxinaw communities (Navarro 2004). Ungulates predominance is observed in other indigenous and non -indigenous unaltered forest settlements in the Amazon (Alvard 1993; Robinson and Bodmer 1999; M ena et al. 2000), as well as commonly at the large scale of the Amazon basin (Jerozolimski and Peres 2003) and other tropical forests (Fa et a. 2005). The transition from hunting pr incipally ungulates to smaller and less preferred species observed across Pano villages agrees with the pattern found in single Amazonian settlements that had wildlife depleted by persistent hunting (Stearman 1990) or by factors other than hunting (Escamil la et al. 2000). The same gradual shift is also observed in the spatial variation of wildlife status across multiple sites in a single Extractive Reserve in the Amazon (Ramos 2005) as well as larger tropical spatial scales (Jerozolimski and Peres 2003; Fa and Brown 2009). Pano hunters' return rate of meat declined in response to decline in abundance of preferred game, particularly ungulates, as observed in local communities in the Peruvian Amazonian where peccaries and tapir declined over 10 years reflecti ng in hunters return rate (Bodmer et al. 1997). The depleted Pano villages disproportionately hunted smaller resilient species, while sensitive species like razor billed curassow, piping guan, spider and woolly monk eys,
52 and tapir were usually absent from t heir prey profiles. In the Neotropics these species are the most sensitive to anthropogenic disturbances and likely to be the first ones to be extirpated locally (Daily et al. 2003; Parry et al. 2007) and across sites in the Amazon basin (Peres 2000). Henc e, wildlife responses in Pano hunting territory refer to change in assemblage structure by reducing abundances of preferred species, particularly ungulates, and composition by extirpation of some specific sensitive species, which resulted in the decline of capture -per unit of effort according to Optimum Foraging Theory. The increase in the distance of animals from village is related to the reduction of abundance of tropical game (Hill et al. 1997; Ohl Schacherer et al. 2007) and meat return rate in other in digenous groups in the Amazon (Alvard et al, 1997; Sirn et al. 2004). In these sites, hunting was indicated the driver of game variation. The pattern of animal distance to the village across Pano communities, in turn, differed from these Amazonian. Pano c ommunities hunting preferred animals farther away were not those with their wildlife assemblage considerably depleted, but were those with higher hunting pressure, conforming to the premises of the central place forager model. Therefore, this indicator might represent a different process driving wildlife response across Pano villages than the others. Drivers of Wildlife Depletion In this study it was possible to disentangle the effects of indigenous hunting and deforestation influencing game hunted by Pano people. Even in one of the less deforested regions in the Amazon, indicators of game depletion across Pano villages was mostly explained by increasing nearby deforestation. Indigenous hunting, in turn, was only associated to the dispersion of animals from the village. How ever, non-
53 indigenous hunting may be responsible for part of game depletion since hunting outside villages and Indigenous Lands was not estimated, and is possibly related to external deforestation. The most disturbed villages, in terms of wi ldlife and forest loss, are those more vulnerable to non-indigenous threats because of their location close to Indigenous Lands boundaries. Moreover, these vulnerable villages are located in Indigenous Lands of low and medium defensibility against external threats which suggest s that factors operating at this level can affect wildlife close to villages. In fact, wildlife is more depleted in villages of Indigenous Lands poor ly designed and located in the s tate development frontier. Effects of Indigenous Hunting Pressure on Game The dispersion of large game species was the only wildlife response possibly related to indigenous hunting pressure across Pano communities, represented by the density of indigenous people in neighbor villages. Hunting is responsible for game depletion at local scales (Robinson and Bodmer 1999) and across sites in the Amazon (Peres and Palacios 2007). High concentration of villages exacerbates the effects of hunting because the source of disturbance has many close origins (Levi et al. 2009). Given that the distance of animals hunted by Pano was not associated to other wildlife responses, I suggest two possible nonexclusive hypotheses One is indigenous hunting caused game depletion only very close to the villages, which is supported b y the absence of spatial dependence of distance values, the lack of difference between villages of different protection to external threat, and lack of difference among Indigenous Lands of varying levels of defensibility. In a similar scale to my study, game depletion was related only to local effects of human density (grain of 3.1km 2 ) across forest fragment of Mexican tropical forests (Urquiza-Hass et al. 2009). Or indigenous
54 hunting just displaced animal populations and did not reduce their abundances. G ame species in hunted areas are more wary of people (Hill et al. 1997; Fitzgibbon 1998) and, given the mobile nature of this resource, may flee and avoid disturbance in the village area (Kensinger 1975; Urquiza-Hass et al. 2009). A previous study in a sing le Kaxinaw Indigenous Land suggested that the history of land use, not indigenous hunting, explained wildlife status (Constantino et al. 2008). The population of local people was not associated to depletion across settlements in Central/West Africa forest s (Fa et al. 2005) and in the Amazon (Jerozolimski and Peres 2003). Although Jerozolimski and Peres (2003) found that the persistence of local hunting over time explained wildlife depletion in the Amazon, the age of Pano villages was not related to game variation. Therefore, there is no clear evidence in this study to support the idea that contemporary and past indi genous hunting severely depletes game species at the regional scale. Effects of Deforestation a round Villages on Game Deforestation was the only factor significantly affecting wildlife across villages with drastic consequences for Pano hunters who responded by spend ing more time hunting to supply their meat needs. At large spatial scales, deforestation is the most important factor depleting wildl ife in the tropics and has been used to project the reduction of populations' range and species extinction (Cuarn 2000; Grelle 2005; Kinnaird et al. 2003; Soares Filho et al. 2006; Vale et al. 2008). At local scales, higher deforestation in hunting terri tory near a highway caused Amazonian hunters to catch smaller prey (Bonaudo et al. 2005). At regional scales, hunters caught more ungulates in areas of primary forests of heterogeneous landscape in eastern Amazon (Parry et al. 2009b). In the Alto Juru Ext ractive Reserve, adjacent to the Pano Indigenous Lands of Jordo and Breu Indigenous Lands (Figure 21), ungulates hunted by local people declined in
55 response to increasing deforestation density (Ramos 2005). Deforestation, in spite of local hunting, caused the loss of many game species across fragments in a regional scale in Mexico (Urquiza-Hass et al. 2009) and the Brazilian Amazon (Michalski and Peres 2005, 2007). Even though d eforestation estimates around Pano villages may include regrown fores ts creati ng mosaics of habitats, there is enough evidence that slight habitat disturbances in tropical forests (i.e. fallows and secondary forests) are related to decline in wildlife abundances and richness (Naughton-Treves et al. 2003; Parry et al. 2009a). Forest specialist species are expected to be affected by light habitat alteration but, depending on the quality of secondary forests, tolerant species are only present in primary forests (Estrada et al. 1994; Parry et al. 2009a, b; Urquiza-Hass et al. 2009). Desp ite the independence of indigenous hunting, deforestation estimates may have other associated factors that deplete game species. Habitat loss and land use practices inside Indigenous Lands are likely to affect wildlife since most deforested patches corresp ond to slashand burn agriculture and small scale pasture bordered by forests, which are prone to effects of fire and fragmentation (Norris et al 2008). Fragmentation, susceptibility to wildfire, and wildlife disease are related to deforestation and have direct (i.e. reducing population size) and indirect (i.e. reducing game resources) impacts on large game, operating independently or in synergy (Laurance et al. 2002; Michalski and Peres 2007; Laurance and Useche 2009). These findings, thus, cause alarm for the negatives consequences of deforestation to game species even in tropical forest regions yet well conserved. The effects of deforestation and associated disturbances were described in highly fragmented
56 landscape (Cullen et al. 2000; Chiarello 1999; Mic halski and Peres 2005, 2007; Urquiza-Hass et al. 2009) but less studied in more conserved regions (but see Ramos 2005). Effects of Surrounding Disturbance: the Acre State Zoning Habitat loss near vulnerable villages could not be disentangled from the effec ts of non-indigenous hunting pressure and land use around Indigenous Lands. In the absence of hunting pressure data, deforestation is used as proxy for both because clearing forest increases hunting intensity (Lewis et al. 2004) and hunters' accessibility to hunting sites (Peres 2001). Near vulnerable Pano villages, these factors are likely to be more important than disturbances originated in Indigenous Lands to affect wildlife. Nevertheless, the surrounding effect may operate at the Indigenous Land level r eflecting on villages within them, since all vulnerable villages were located in reserves with lower defensibility against external threats. The effects of human disturbance (i.e. habitat loss, its associated disturbances, and hunting) outside Protected Ar eas impact game populations within their boundaries because of their mobile nature and population dynamics (Woodroffe and Ginsberg 1998; Hansen and DeFries 2007). Therefore, poorly designed reserves (i.e. small and without surrounding protection) may not s upport viable game populations even in the absence of encroachment e specially f or those species requiring large tracts of forests, such as tapir, white lipped peccary and large monkeys (Peres and Terborgh 1995; Peres 2005). Low defensibility and location of Pano Indigenous Lands is, nonetheless, related to nonindigenous invasion for hunting. Only Pano Indigenous Lands not bordered by Protected Areas report frequent n on indigenous encroachment using more impacting hunting techniques ( Peres 1993; Lima 2001; CPI AC 2005a,b; Bant and Pessoa 2008; Constantino et al. 2008), becoming a
57 major problem in those areas locat ed in the development frontier (i.e. Praia do Carapan ( Bant and Pessoa 2008) and Campinas (Lima 2001; Calouro 2007)). Private lands around Indig enous Lands have a more intensive land use pattern than do Pano people (IBGE 2009), supporting the notion of external threats associated to land cover conversion affecting wildlife hunted in vulnerable villages. Thus, the historical spatial occupancy of w estern Amazon and the current State Ecological -Economic Zoning create a heterogeneous regional landscape with polarized differences, including land tenure, infrastructure, urban development and incentives for deforestation, that have considerable influence on wildlife and indigenous societies using them. Highly defensible Indigenous Lands strategically compose the Western Amazon Ecological Corridor, protected by adjacent Conservation Units and covering the watershed's headwaters along the international fron tier with Peru, hosting less modified wildlife assem blages. There also lie the Conser vation Units of restricted use in the s tate. Development and forest exploitation in this region threat en little the Indigenous Lands (Carneiro Filho and Souza 2009) given their remoteness, tenure, and the fact that considerable deforestation started only after 1995, remaining limited (Franke 2008; Acre 2009). Hunting with shotguns, conversely, occurs since before 1920s everywhere in the state and is likely to have been mor e intensive along the international frontier in the past because of meat and pelt trade route s to Peru until the 1970s (Doughty and Myers 1971). Low defensible Indigenous Lands, in turn, have unprotected perimeter s and small relative size and are scattered in the developmen t frontier zone planned by the s tate government, which hosts the largest cities alongside the BR 364 (Acre 2000), probably resulting in depleted wildlife. People in these Indigenous Lands have received
58 compensation to mitigate social and environmental road impacts (Iglesias and Aquino 2006). Deforestation there was limited until the exponential growth in cattle production during the 1980s and intensification during the 1990s (V alentin et al. 2002; Arima et al. 2005, Oliveira and Bardales 2 006; Franke 2008). In addition, commercial logging, regulated or not, takes place in this region including inside some Protected Areas (Acre 2009). In surroundings of the Antimary State Forest, located in the development frontier in Acre State, game specie s hunted by local people are likely to have been affected by encroachment and deforestation (Calouro 1995, 2005). In that Protected Area, the density of local people decreased 13% in the last 10 years reducing their hunting pressure; however, the region w ent through a rapid development process and forest loss leading to deforestation in the surroundings. Landscape transformation and encroachment, rather than local hunting, could explain the disappearance of some sensitive species even in the less intensiv e hunted areas. At a regional scale, Urquiza Hass et al. (2009) associate non-local peoples hunting to the loss of many species in fragments of Mexican forest. Effects of Roads on Wildlife and H unters Roads had a dual effect on the indigenous hunting syst em. On the one hand they exacerbated the negative effect of indigenous hunting on animal distance from roadside villages, which is expected given the disturbances associated to roads that deplete wildlife in their surroundings (Laurance et al. 2006, 2008) Road effects are expected to negatively affect large mammals and birds in particular because of their biological traits (i.e. large movement range, low reproductive rates, and low natural densities; Fahrig and Rytwinski 2009). Furthermore, roads diminish defensibility of Pano Indigenous Lands since these are often used by poacher to access forest areas inside and close to
59 Campinas Indigenous Land (Lima 2001; Iglesias and Aquino 2006; Calouro 2007), a typical consequence of roads in tropical forests (Peres and Lake 2003; Laurance et al. 2008). On the other hand, it seems that, due to their easy access to cities, roadside Pano communities had different hunting behavior in comparison to communities alongside rivers, by excluding secondary species from prey pro files. Pano roadside communities shifted into a market dependent diet that includes farmed fish, livestock and industrialized meat (Lima E.C., Pessoa, M. and Tavares, R.A. Personal communication). Because of elevated ammunition price s and low return rates for small species hunting, hunters substitute them for easily accessed alternative meat sold in markets and restrict their hunting to large bodied mammals, mainly ungulates, even in situations of depleted wildlife (Kensinger 1975; Redford 1993; Jerozolims ki and Peres 2003). Independent of my study, a contemporary wildlife census and hunting survey in the Campinas Indigenous Land indicates that most sensitive species are absent and ungulates occur in very low densities, although these contribute most to wi ldmeat consumed, while smaller species occur at higher densities but yet are little hunted (Calouro 2007). This substitution pattern was observed in communities that have livestock as an alternative meat source, and therefore avoided hunting certain previously acceptable species (Jerozolimski and Peres 2003) or after an upgrade in hunting weapons and transportation to market (Winterhalder 1981; Hill and Hawkes 1983). Increasing the profits of hunting (i.e. shifting from subsistence to commercial harvest) ch anges the composition of off -take assemblages by focusing on more profitable species (Damania et al. 2005). Other indige nous people intensified their game harvest in certain species with increased access to market as a consequence of the
60 shift from subsist ence to commercial hunting system s ( Sierra et al. 1999). Although this could explain the pattern in these Indigenous Lands, if Pano engage in ill egal commercial hunting it is to a much lesser extent than subsistence hunting, even though illegal wildlife co mmercialization in the region is high. All other Pano communities that had restricted access to markets, because of reduced size of nearby cities (i.e. Jordo and Santa Rosa do Purus) or their distance to them (i.e. Tarauac and Cruzeiro do Sul to Praia do Carapan and Breu Indigenous Lands), relied on a much more diversified fauna shifting to smaller animals when large game is deplete Hence, resulting off -takes and mean prey weight in these villages reflect human processes more than prey abundances (Ling and Miner -Gulland 2006) and should not be used to infer wildlife status. Nonetheless, these results are important to understand the effects of roads in indigenous communities behavior and societies. Source sink Dynamics of Game P opulations The Western A mazon Ecological Corridor has very low deforestation and human density, concentrated around small settlements, being one of the few residences of isolated indigenous groups in the Amazon (Castillo 2002). In such a landscape, game populations in the large t rac ts of forests may source populations to the disturbed areas (Novaro et al. 2000). Source-sink systems might be responsible for the persistence of health y populations of sensitive species hunted for decades at unsustainable rates by indigenous people in a nearby region of Peruvian Amazon (Ohl Schacherer et al. 2007). Peres (2001) suggests that hunting persistence (age of village) do es not deplete most of the targeted species near villages because of the constant animal immigration from adjacent large unde rharvested forested areas that include source populations. This dynamic might also be important to continually provide animals from a same source
61 area for Kaxinaw villages in the Jordo Indigenous Land (Constantino et al. 2008) and extractivist settlements in the Alto Juru Extractive Reserve (Ramos 2005). At a larger scale source -sink dynamics may be important to maintain the high proportion of ungulates and the presence of tapirs and large monkeys in villages located in the headwaters even after many d ecades of intense hunting. Conversely, source -sink dynamics in Indigenous Lands of the development frontier, containing more forest and less hunting pressure than outside, would not function due to reduced habitat or connectivity (Peres 2001; Hansen and De Fr ies 2007) or, paradoxically, would invert the flow, sourcing game to surrounding disturbed sites (Woodroffe and Ginsberg 1998; Naranjo and Bodmer 2007; Hansen, in preparation). In this zone, this process would exacerbate the negative effect of surroundin g disturbances on wildlife hunted by Pano communities. Therefore, even if all Pano villages overharvest game species, depletion would be hidden behind the differential effects of source-sink dynamics. Indigenous Lands as Protected Areas for W ildlife Indige nous Lands avoid forest impoverishment by deforestation, logging, and fire (Nepstad et al. 2006; Adeney et al. 2009; Asner et al. 2009). The same pattern is observed across Indigenous Lands at the regional scale in Acre, where these are responsible for onl y 1% of State deforestation (Acre 2009). The capacity of Indigenous Lands to avoid depletion of other natural resources, however, has not been assessed at large scales using empirical data (Nepstad et al. 2006; Adeney et al. 2009). AzevedoRamos and colleagues (2006) even suggest that Indigenous Lands are key to mammal conservation in the Brazilian Amazon because they overlap the geographic range of more species than other Protected Area categories and their large size and low human
62 population density woul d effectively protect mammal species that easily recover from hunting loss. Pano people s land use and forest conversion but not hunting affected wildlife, principally large mammals. More important, disturbances associated to deforestation outside Indig enous Lands depleted wildlife in vulnerable villages and low defensible reserves whereas their size alone was not associated to wildlife conservation. For instance, most of these areas were smaller than the minimum required to support the average Amazonian sustainable harvest of most preferred mammals and birds (Peres 2001). The mobility of wildlife also makes it more likely to respond to disturbances from surrounding areas (Hansen and DeFries 2007). The high performance of Indigenous Lands in protecting Am azon forests related to their capacity to inhibit disturbance in regions of high anthropogenic impact (Nepstad et al. 2006; AzevedoRamos et al. 2006) may not function for wildlife protection as evidenced by the fact that the most affected Pano villages are located in Indigenous Lands in the Acre development frontier, thus subject to extreme external influence. Therefore, unlike other sources of forest degradation (e.g. fire and deforestation) Pano Indigenous Lands alone are not able to guarantee game conservation and maintenance of indigenous culture. Management I mpli cations and State Conservation P olicy The findings of this study although concerning the Kaxinaw and Katukina people, could be extended to other central place hunter indigenous peoples tar geting large vertebrates with the same hunting techniques in similar heterogeneous environment s This includes most of the non -isolated indigenous people in Acre State. The majority of governmental economic and environmental policies related to Indigenous Lands in Acre are designed and implemented by the State government, which understands them as
63 strategically important for conservation (Acre 2008, 2009). Hunting restrictions have been adopted as management norms (i.e. quotas, sex and age selection) by man y indigenous and local people to guarantee local sustainable wildlife use, including in Acre ( Gavazzi 2007 ; Zapa ta -Ros et al. 2009 ). Yet if these norms are to succeed, decision making and compliance has to occur in groups of neighboring villages, not only within single villages, although the efficacy of these restrictions may just influence wildlife conservation at very local scales. At regional scale, nevertheless, these management regulations probably do not ensure wildlife conservation. Yet, landscapes cale is the most appropriate to natural resource management policies that involve local people and government interaction in the tropics (Frost et al. 2006). In this sense, s tate policies that attempt to maintain indigenous culture and wildlife conservation should target the deforestation within and in the surroundings of Indigenous Lands, their land use associated disturbances, and non -indigenous hunting in the vicinity of Indigenous Lands. Regarding Pano people, there is a recent increase in deforestation inside some Indigenous Lands (SIPAM 2008), and my data indicate that it is positively associated with village linkage to market but negatively associated with monthly income per family (Appendix C), explaining together 63% of deforestation. Although the m echanisms for these associations are not explicit, the government could explore them given that most of Pano income comes from retirement and salaries paid by the government to control deforestation. Indigenous Lands surroundings, in turn, are projected t o be severely deforested in the near future, principally in the development frontier, if the current forest loss persist s (IPAM and Consrcio MABE 2009). In response, the Acre government recently released a new policy to prevent and control deforestation i n the s tate including
64 within Indigenous Lands (Acre 2008, 2009). Indigenous hunting and game could be considerably protected as a consequence of deforestation reduction and reforestation increase according to the government plan (Acre 2009). There is, however, no link between Indigenous Lands and their vicinity, as well as no mention concerning integrated wildlife management. Moreover, t he government recognizes that these goals will only be achieved in a scenario of high governance, which depends upon sever al factors, including fund raising not yet guaranteed, to strength en environmental management capacity and economic sustainability of environmental policies (Acre 2009). At a local scale, the Ethnozoning and Territorial Management Plan of Indigenous Lands currently supported by the State government (Acre 2 008) can be the instruments through which indigenous people negotiate the landscape approach for wildlife management. Considerations on Scale and the Absence of Hunting E ffect The findings in this study di ffered from most of the others studying the drivers of game depletion, since Pano hunting had minor effects (Peres 2000; Peres and Palacios 2007). Here I briefly explore some possible explanations. Using hunting data, instead of direct observation, to infe r wildlife status can be biased due to non random sampling and a possible inability to capture light variation in game populations since hunter s persist in harvesti ng rare preferred species as by -catch hunting products (Rowcliffe et al. 2003). However, this is less likely to be the reason f or dissimilar results, since Pano hunting off -take followed the widespread pattern observed in the Amazon that can be modeled according to the ecological theories used to derive the indicators. Moreover, I restricted my analyses to more than 9000 records from 35 communities of Pano people in a regional scale with small natural environmental variation, with similar hunting
65 preferences and strategies, that had reported hunting on more 50 animals and over a period of more th an six months between 2005 and 2009. By contrast, o ther large extent studies analyzing drivers of game variation using hunter -kill profile in multiple tropical forest sites (>20) were based on meta analysis of studies with decades of difference conducted at continental scale or at intercontinental scale (Robinson and Bennett 2004; Fa and Peres 2001). These studies used lower cutting values to include sites of hunters of different cultures in the sample (i.e. undetermined number of animals in Peres (2001); >30 animals in Jerozolimski and Peres (2003); >40 animals in Fa et al. (2005)). These differences of culture and hunting technology, ecosystem productivity and spatial and temporal variation on wildlife abundances may introduce noise in these results (Emmo ns 1984; Redford and Robinson 1987; Lupo and Schmitt 2005; Peres and Palacios 2007; Nasi et al. 2008). Nevertheless, I acknowledge that the ecological theories supporting the indicators are not expected to explain all the variation in hunters subsistence s ystems (Hawkes et al. 2008) and that neglected environmental differences (i.e. soil fertility and hydrological regime) within the scale of this study (i.e. grain and extent) may influence game populations non randomly. M any other studies do not clearly def ine hunting pressure (Rist et al. 2008), since it is frequently hard to determine the actual number of hunters using a territory and the intensity of hunting pressure (Nasi et al. 2008). For example, the common use of number of people in a settlement to represent hunting pressure (Jerozolimski and Peres 2003; Fa et al. 2005) might not be appropriate as many indigenous communities fraction in response to depleted wildlife, becoming a settlement with a small number of people in a depleted site ( Ferguson 1989 ). Also, a local community hunting territory is
66 often used by neighbors or poachers or is adjacent to areas more intensively harvested, affecting wildlife hunted by locals ( Ferguson 1989; Stearman 1990) Therefore, attributing the responsibility of game depletion to loc al people living within Protected Areas might underestimate the effect of encroachment, animal population dynamics, spread of wildfire, and edge effects: some of the most common external threats to biodiversity within Brazilian Conservation U nits (Peres and Terborgh 1995; Carrillo et al. 2000; Bruner et al. 2001; Laurance et al. 2000; Nepstad et al. 2006). I was able to isolate indigenous hunting pressure from other factors due to extensive field surveys as well as identify the Pano territories exclusive ly used by them. Nonetheless, hu nting from outside may still have some effect on wildlife since I could not disentangle the effects of deforestation from non -indigenous hunting on game hunted by the Pano. These factors may influence wildlife in synergy, which make it more difficult to distinguish between their effects (Nasi et al. 2008; Fa and Brown 2009; Parry et al. 2009b). In addition, many tropical forest species that are sensitive to hunting are also sensitive to habitat loss (Michalski and Peres 2007; Rist et al. 2009) and fire effects (Peres et al. 2003; Barlow and Peres 2006). Therefore, fluctuations in their populations, frequently assumed to be a c onsequence of hunting pressure may be related erroneously to one of the processes if the ot hers are not properly acknowledged. Satellite based information only recently became available, yet for restricted regions, allowing the reg ular use of habitat disturbance variables in scales appropriate to game species, mostly neglected in previous studie s (Redford and Robinson 1987; Peres 2001; Jerozolimski and Peres 2003; Fa et al. 2005).
67 The scale of this study is a plausible explanation for the marked difference in results. Most wildlife and hunting research is conducted either at local or global scale s in the Amazon, mainly restricted by methodological issues ( Nepstad et al. 2006 ). Spatial scale alone can explain much of the variation in the response of wildlife to habitat loss (Hill and Hamer 2004; Dumbrell et al. 2008) and to human occupancy (Pautass o 2007). Robinson and Bodmer (1999) claim that indigenous hunting has been sustainable at the landscape level whereas, locally, many species must have been extirpated since hunting effects operate only around settlements (Lewis et al. 2004). Pano hunting had only local effect s on wildlife, which might have been suppressed by the landscape differences in deforestation. Likewise, most lo cal studies report the hunting-induced depletion trend, although confounded by source -sink dynamics ( Ohl Schacherer et al. 2007) whilst recent studies at similar regional scale to mine have assessed the effects of hunting and habitat disturbance on game animals (Ramos 2005; Peres and Michalski 2005, 2007; Urquiza -Hass et al. 2009), with impressive indications of the importanc e of deforestation over local hunting. In this intermediate spatial scale (grains circa 10km diameter and extent circa 70,000km 2 ) species richness is expected to have no relationship with human presence (Pautasso 2007). Although local and global studies a re relevant they might not be appropriate to orient policies that are usually implemented at regional levels (Ferrier 2002; Fisher and Owens 2004; Frost et al. 2006; Fritz et al. 2009), the scale of decisions regarding Indigenous Lands in Acre, and particu lar by appropriate to study large mammals in the Amazon due to their metapopulation dynamics.
68 In addition, major drops in mammals densities due to hunting are more likely to occ ur in previously unhunted sites than those already exploited (Nasi et al. 2008; Fa and Brow n 2009). Many studies suggest that hunting depletes wildlife, base d on comparisons between unhunted and hunted sites (Carrillo et al. 2000; Peres 2000) or following game population changes in the first years after settlements establishment (Vic kers 1980; Ayres et al. 1991). This is certainly not the case in Pano villages as the region is subject to centenary intense hunting but only recently has been affected by major land conversion, more prominent in the development frontier. A selective filte r may operate after initial hunting pressure, where tolerant species and population abundances remain stable in sites (Balmford 1996; Cowlishaw et al. 2005) until other processes start adding to the negative effect of hunting (Cullen et al. 2000). Frontier s are more prone to witness extirpation of large bodied species (Davis et al. 2008; Fritz et al. 2009) since sensitive species are still present but unprotected from anthropogenic threats (Balmford 1996). The temporal dim ension of this study may also have implications in my findings. In 2005, the southwestern Amazon experienced an abnormal ly severe drought with widespread and intense associated fire (Arago et al. 2007). Although villages all over the s tate were affected by uncontrolled fire, those located in the development frontier were more affected by external disturbance than those protected in the headwaters with several recurrent burns (Carneiro Filho and Souza 2009). Hence, burning patterns overlaps deforestation distribution making the effects of these threats hard to distinguish. Despite our vague understanding of long term response s of wildlife to forest fire, certain species seems to be absent from recurrent burned sites or have abundance
69 reduced (Barlow and Peres 2006). However, if game species in Acre are somehow responding to the extended wildfire that occurred in 2005, climate change must be a major threat to those species because the southwestern Amazon region is particularly affected in cases of North Atlantic warm ing becoming more prone t o forest fires, principally alongside roads (Arago et al. 2007).
70 CHAPTER 5 CONCLUSIONS Pano people s hunting does not cause major impacts in the game they hunt. Even in an Amazon region relatively conserved, deforestation is the main driver of gradual ga me depletion. Forest loss and probably other associated disturbances (i.e. fragmentation and fire) reduce game populations and extirpate most sensitive species inside Indigenous Lands. Indigenous Lands may avoid forest loss but do not alone guarantee wildl ife conservation in Acre. Land use and hunting by non -indigenous neighbors and the Indigenous Lands location and design may contribute considerably to game depletion in Pano villages. Although hypothesized for regional scales, the null indigenous huntin g effect as opposed to habitat loss impact on game had not been reported in the literature, due to the difficulty of disentangling factors, and data and scale restrictions. Roads impacted wildlife negatively by exacerbating hunting effects while restrict i ng hunters' dependence on large game, independence of its abundance, by allowing easier access to alternative meat. It is possible that the anticipated disaster of near future deforestation will not affect game species and Pano people, if the recent Acre S tate policy for deforestation and fire control and prevention is effectively implemented. However, policies that integrate Indigenous Lands and its vicinity are still required.
71 APPENDIX A KERNEL DENSITY ESTIM ATES OF DISTANCE OF PREFERRED ANIMALS HU NTED FR OM THE VILLAGE
73 APPENDIX B WILDLIFE STATUS INDICATORS AND ITS RELATED ECOLOGICAL MODELS AND THEORIES
74 Indicator Model/Theory Expected outcome Comments Mean prey weight Optimum Foraging Theory Average weight of hunted animals decline in depleted sites in response to decline in large bodied vertebrates Provide a proxy for the assemblage of game animals regardless of species, likely to be related to variation in population abundances Capture per Unit of Effort (CPUE) of preferred species Optimum Foraging Theory/ Ecological Economics Hunters spend more time to pursue meat of preferred animals in depleted sites in response to decline in large game abundance Restricted to preferred species since return rates of secondary animals may not reflect abundance; Allows for indication of hunting success on single species or taxa Sensitive species richness Indicator Species Species sensitive to human disturbance of different natures decline to level that are no longer hunt ed, ultimately becoming extirpated, with increasing wildlife depletion Given non random extinction process certain species reflect overall status of wildlife assemblage status Mode distance of preferred species from the village Central Place Forager Model Hunters catch preferred animals farther away from villages where central place hunting depletes game populations Restricted to preferred animals since the distance of secondary animals might not reflect species abundances; given a non parametric distribut ion of distance values, mode was more appropriate to describe the most common distance
75 APPENDIX C PEARSONS CORRELATION MATRIX OF VARIABLES POSSIBLY INFLUENCING HUNTING IN INDIGENOUS VILLAGES, CONSIDERING 29 VILLAGES
76 Hunting pressure Land use Soc io economy Explanatory variables Population size Village age (years) Square Root of number of hunters Density of Indigenous people (people/5km radius buffer) Log10 deforested area in 5km radius buffer (ha) Square root of distance to entrance of Indigenous Land (km) Log 10 deforested area in 5km buffer but inside Indigenous Land (ha) Log 10 deforested area in 5km buffer but outside Indigenous Land (ha) Log10 number of employees Income per family ($/family) Distance to market (km) Log10 Animal Unita Populatio n sizea 1 0.189 (0.33) 0.672 (0.00)*** 0.181 (0.35) 0.187 (0.33) 0.030 (0.88) 0.14 (0.40) 0.04 (0.88) 0.733 (0.00)*** 0.143 (0.46) 0.70 (0.72) 0.02 (0.9) Village agea 1 0.132 (0.49) 0.07 (0.7) 0.076 (0.69) 0.156 (042) 0.03 (0.84) 0.06 (0.81) 0.294 (0.12) 0.209 (0.28) 0.025 (0.90) 0.04 (0.83) Square Root # Hunters 1 0.127 (0.51) 0.576 (0.00)** 0.353 (0.06) 0.50 (0.00)** 0.37 (0.20) 0.598 (0.00)** 0.119 (0.54) 0.353 (0.06) 0.06 (0.76) Density of Indigenous peoplea 1 0.348 (0.06) 0.22 (0.25 ) 0.37 (0.3)* 0.29 (0.27) 0.107 (0.58) 0.029(0.88) 0.258 (0.18) 0.31 (0.1) Log 10 Deforested area (ha) a 1 0.719 (0.00)*** 0.94 (0.00)*** 0.73 (0,00)*** 0.243 (0.20) 0.568 (0.00)** 0.748 (0.00)*** 0.46 (0.07)* Square Root of distance to entranc e of Indigenous Land 1 0.63 (0.00)*** 0.74 (0.00)** 0.071 (0.71) 0.433 (0.02)* 0.887 (0.00)*** 0.37 (0.05)* Log 10 deforested area in 5km buffer but inside Indigenous Land (ha) 1 0.26 (0.32) 0.27 (0.15) 0.54 (0.00)** 0.74 (0.00)*** 0.30 ( 0.11) Log 10 deforested area in 5km buffer but outside Indigenous Land (ha) 1 0.00 (0.99) 0.43 (0.12) 0.20 (0.51) 0.26 (0.40)
77 APPENDIX C. Continued Hunting pressure Land use Socioeconomy Explanatory variables Population size Village age (year s) Square Root of number of hunters Density of Indigenous people (people/5km radius buffer) Log10 deforested area in 5km radius buffer (ha) Square root of distance to entrance of Indigenous Land (km) Log 10 deforested area in 5km buffer but inside Indigenou s Land (ha) Log 10 deforested area in 5km buffer but outside Indigenous Land (ha) Log10 number of employees Income per family ($/family) Distance to market (km) Log10 Animal Unita Log 10 # Employees 1 0.273 (0.15) 0.071 (0.71) 0.21 (0.28) Income per family 1 0.368 (0.05)* 0.13 (0.49) Distance to market 1 0.21 (0.26)
78 LIST OF REFERENCES Acre (2000) Programa Estadual de Zoneamento Ecolgico -Econmico do Estado do Acre. Zoneamento Ecolgico-Econmico: aspectos socioeconmicos e ocupao territorial, vol. 2 SECTMA, Rio Branco. Acre (2006). Governo do Estado do Acre. Programa Estadual de Zoneamento Ecolgico -Econmico do Estado do Acre. Zoneamento Ecolgico Econmico do Acre Fase II: Documento Sntese Escala 1:250.000. SECTM A, Rio Branco. Acre (2008) Brazil: Acre social and economic inclusion project (PROACRE). Executive summary of the environmental assessment D raft report. World Bank, Washington. Acre (2009) Plano Estadual de preveno e controle dos desmatamentos do Acre. Governo do Estado do Acre, Rio Branco. Adeney, J.M., Christensen, N.L. & Pimm, S.L. (2009) Reserves protect against deforestation fires in the Amazon. PLOS One 4 e5014. Albrechtsen, L., MacDonald, D.W., Johnson, P.J., Castelo, R. & Fa, J.E. (2007) Fa unal loss from bushmeat: empirical evidence and policy implications in Bioko Island. Environmental Science and Policy 10 654-667. Almeida, M.B., Lima, E.C., Aquino, T.V. & Iglesias, M.P. (2002) Caar. Enciclopdia da Floresta (eds M.C. Cunha & M.B. Almeida), pp. 311336. Companhia das Letras, So Paulo. Alvard, M.S. (1993) Testing the ecologically noble savage hypothesis: Interspecific prey choice by Piro hunters of Amazonian Peru. Human Ecology 21 355387. Alvard, M.S. (1995) Intraspecific prey ch oice by Amazonian hunters. Current Anthropology 36 789818. Alvard, M.S., Robinson, J.G., Redford, K.H. & Kaplan, H. (1997) The sustainability of subsistence hunting in the Neotropics. Conservation Biology 11 977 -982. Amaral, B.D. (2005) Fisheries an d fishing effort at the indigenous reserves Ashaninka/Kaxinaw, river Breu, Brazil/Peru. Acta Amazonica 35 133144. Aquino, T.V. & Iglesias, M.P. (1994) Kaxinaw do Rio Jordo: histria, territrio, economia e desenvolvimento sustentado. Comisso Pr-n dio do Acre, Rio Branco. Arago, L.E.OC., Malhi, Y. RomanCuesta, R.M., Saatchi, S., Anderson, L.O. & Shimabukuro, Y.E. (2007) Spatial patterns and fire response of recent Amazonian droughts. Geophysical Research Letters 34 L07701.
79 Arago, L.E.O.C., Ma lhi, Y., Barbier, N., Lima, A., Shimabukuru, Y., Anderson, L. & Saatchi, S. (2008) Interactions between rainfall, deforestation and fires during recent years in Brazilian Amazonia. Philosophical Transactions of the Royal Society B, 363 1779 1785. Arima, E., Barreto, P. & Brito, M. (2005) Pecuria na Amaznia: Tendncias e implicaes para a conservao ambiental Instituto do Homen e Meio-Ambiente da Amaznia Belm. Asner, G.P., Knapp, D.E., Broadbent, E.B., Oliveira, P.J.C., Keller, M. & Silva, J.N. (2 005) Selective logging in the Brazilian Amazon. Science 310, 480482. Asner, G.P., Rudel, T.K., Aide, M., DeFries, R. & Emerson, R. (2009) A contemporary assessment of change in humid tropical forests. Conservation Biology 23 1386 1395. Ayres, M., Mag alhes Lima, D., De Sousa Martins, E. & Barreiros, J.L. (1991) On the tract of the road: changes in subsistence hunting in a Brazilian Amazonian village. Neotropical Wildlife Use and Conservation (eds J.G. Robinson & K.H. Redford), pp. 8292. University of Chicago Press, Chicago. Azevedo-Ramos, C., Amaral, B.D., Nepstad, D.C., Soares -Filho, B. & Nasi, R. (2006) Integrating ecosystem management, Protected Areas, and mammal conservation in the Brazilian Amazon. Ecology and Society 11 art. 17. Balmford, A. (1996) Extinction filters and current resilience: the significance of past selection pressures for conservation biology. Trends in Ecology and Evolution 11 193-196. Barlow, J. & Peres, C.A. (2006) Effects of single and recurrent wildfires on fruit production and large vertebrate abundance in a central Amazonian forest. Biodiversity and Conservation 15 985-1012. Barlow, J., Gardner, T.A., Araujo, I.S., vila -Pires, T.C., Bonaldo, A.B., Costa, J.E., Esposito, M.C., Ferreira, L.V., Hawes, J., Hernandez, M.I.M., Hoogmoed, M.S., Leite, R.N., Lo Man-Hung, F.N., Macolm, J.R., Martins, M.B., Mestre, L.A.M., Miranda -Santos, R., Nunes -Gutjahr, A.L., Overal, W.L., Parry, L., Peters, S.L., Ribeiro -Junior, M.A., da Silva, M.N.F., da Silva Motta, C. & Peres, C.A. ( 2007) Quantifying the biodiversity value of tropical primary, secondary, and plantation forests. Proceedings of the National Academy of Science, 104, 18555 -18560. Bant, A. & Pessoa, M. (2008) Levantamento Etnoecolgico das Terras Indgenas do Complexo Bac ia do Rio Juru: Kaxinaw da Praia do Carapan, Kampa do Igarap Primavera e Kulina do Igarap do Pau. FUNAI/PPTAL/GTZ. Brasilia, DF.
80 Barreto, P., Souza Jr., C., Noguern, R., Anderson, A., Salomo, R. & Willies, J. (2006) Human pressure on the Brazilian Amazon forests World Resources Institute Report, Washington. Bird, D.W., Bird, R.B. 7 Codding, B.F. (2009) In pursuit of mobile prey: Martu strategies and archaeofauna interpretation. American Antiquity 74 3 -29. Bodmer, R.E., Aquino, R., Puertas, P., Reyes, R., Fang, T. & Gottbenker, N. (1997) Manejo y uso sustentable de pecaries en la Amazonia Peruana. International Union for the Conservation of Nature and Natural Resources, Quito. Bonaudo, T., Le Pendu, Y. Faure, J.F. & Quanz, D. (2005) The effects of deforestation on wildlife along the transamazon highway. European Journal of Wildlife Research, 51, 199206. Boyle, S.A. (2008) The effects of forest fragmentation on primates in the Brazilian Amazon PhD thesis, Arizona State University. Brackelaire, V. (2005) Inventrio de iniciativas socioambientales en la Amazonia del Noroeste Brasil, Colombia, Venezuela. Cooperacin y Alianza en el Norte y Oeste Amaznico, Bogot. Brashares, J.S., Arcese, P. & Sam, M.K. (2001) Human demography and reserve size p redicts wildlife extinction in W est Africa. Proceedings of the Royal Society of London B 268, 2473 -2478. Broadbent, E.N., Asner, G.P., Keller, M., Knapp, D.E., Oliveira, P.J.C. & Silva, J.N. (2008) Forest fragmentation and edge effect from deforestation and selective logging in the Brazilian Amazon. Biological Conservation 141, 1745 1757. Brockington, D., Igoe, J. & Schmidt -Soltau, K. (2006) Conservation, human rights, and poverty reduction. Conservation Biology 20 250-252. Broich, M., Stehman, S. V., Hansen, M.C., Potapov, P. & Shimabukuru, Y. (2009) A comparison of sampling designs for estimating deforestation from Landsat imagery: A case study of the Brazilian Legal Amazon. Remote Sensing of Environment 113 24482454. Brook, B.W., Sodhi, N.S. & Bradshaw, C.J.A. (2008) Synergies among extinction drivers under global change. Trends in Ecology and Evolution 23, 453460. Brooks, T.M., Wright, S.J. & Sheil, D. (2009) Evaluating the success of conservation actions in safeguarding tropical forest bi odiversity. Conservation Biology, 23 1448 1457.
81 Bruner, A.G., Gullison, R.E., Rice, R.E. & Fonseca, G.A.B (2001) Effectiveness of parks in protecting Tropical biodiversity. Science 291 125-128. Buck, L.E., Shames, S. & Scherr, S.J. (2007) Reframing th e Protected Areas -livelihood debate: Conserving biodiversity in populated agricultural landscapes. Protected Areas and Human Livelihoods (eds K.H. Redford & E. Fearn), pp. 130-134. Wildlife Conservation Society, New York. Calouro, A.M. (1995) Caa de subs istncia: sustentabilidade e padres de uso entre seringueiros ribeirinhos e no-ribeirinhos no Estado do Acre M.S. thesis, Universidade de Brasilia. Calouro, A.M. (2005) Anlise do manejo florestal de baixo impacto e da caa de subsistncia sobre um a comunidade de primatas na Floresta Estadual de Antimary (Acre, Brasil). Ph.D. Thesis, Universidade de So Carlos. Calouro A.M. (2007) Atividades de monitoramento participativo de fauna na Terra Indgena Campinas/Katukina (AC). Projeto Segurana alimentar, produo e gesto territorial: Apoio s comunidades indgenas Katukina das Terras Indgenas Campinas e Gregrio Fase II Governo do Estado do Acre, Rio Branco. Cannon, M.D. (2003) A model of central place forager prey choice and an application to fa unal remains from the Mimbres Valley, New Mexico. Journal of Anthropological Archaeology 22 1 25. Cardillo, M. Jones, K.E., Mace, G.M., Bielby, J., Bininda -Emmons, O.R.P., Sechrest, W., Orme, C.D.L. & Purvis, A. (2005) Multiple causes of high extinction risk in large mammal species. Science 309, 1239 1241. Cardillo, M., Mace, G.M., Gittleman, J.L., Jones, K.E., Bielby, J. & Purvis, A. (2008) The predictability of extinction: biological and external correlates of decline in mammals. Proceedings of the Royal Society B 275 14411448. Carneiro Filho, A. & Souza, O.B. (2009) Atlas de presses e Ameaas s Terras Indgenas na Amaznia Brasileira. Instituto SocioAmbiental, So Paulo. Carrillo, E., Wong, G. & Cuarn, A.D. (2000) Monitoring mammal populations in Costa Rica Protected Areas under different hunting restrictions. Conservation Biology 14, 15801591. Castillo, H.B. (2002) Indigenous people in isolation in the Peruvian Amazon: their struggle for survival and freedom WIGIA, Copenhagen. CDC-UNALM (2006) Implementacin del plan de monitoreo de la salud de la biodiversidad en la Zona Reservada Sierra del Divisor: Monitoreo de las amenazas va sensores remotos Informe final. CDC -UNALM, Lima.
82 Chiarello, A.G. (1999) Effects of fragmentation of the At lantic forest on mammal communities in south eastern Brazil. Biological Conservation, 89 71 82. Constantino, P.A.L., Fortini, L.B., Kaxinawa, F.R.S., Kaxinawa, A.M., Kaxinawa, E.S., Kaxinawa, A.P., Kaxinawa, L.S., Kaxinawa, J.M. & Kaxinawa, J.P. (2008) I ndigenous collaborative research for wildlife management in Amazonia: the case of Kaxinaw, Acre, Brazil. Biological Conservation 141 2718 2729. Cowlishaw, G., Mendelson, S. & Rowcliffe, M. (2005) Evidence of post -depletion sustainability in a mature bushmeat market. Journal of Applied Ecology 42 460468. Crookes, D.J., Ankudey, N. & Milner -Gulland, E.J. (2005) The value of a long-term bushmeat market dataset as an indicator of system dynamics. Environmental Conservation 32 333339. CPI -AC (2005a) I Oficina de etnomapeamento das TIs Kaxinaw do Rio Jordo, Baixo Jordo e Seringal Independncia. Internal report. Comisso Pr-ndio do Acre, Rio Branco. CPI -AC (2005b) I O ficina de etnomapeamento da TI Kaxinaw do Rio Humait. Internal report. Comisso Pr -ndio do Acre, Rio Branco. Cullen, L., Bodmer, R.E. & Valladares -Pdua, C. (2000) Effects of hunting in habitat fragments of the Atlantic forests, Brazil. Biological Conservation 95 49 56. Cunha, M.C. & Almeida, M.B. (2002) Enciclopdia da Florest a 1st ed. Companhia das Letras, So Paulo. Daily, G.C., Ceballos, G., Pacheco, J., Suzn, G. & Snchez -Azofeifa, A. (2003) Countryside biogeography of Neotropical mammals: Conservation opportunities in agricultural landscapes of Costa Rica. Conservation Biology 17 1814 -1826. Da mania, R., Milner -Gulland, E.J. & Crookes, D.J. (2005) A bioeconomic analysis of bushmeat hunting. Proceedings of the Royal Society B 272, 259266. Davidson, A.D., Hamilton, M.J., Boyer, A.G., Brown, J.H. & Ceballos, G. (2009) Multiple ecological pathways to extinction in mammals. Proceedings of the National Academy of Science, 106, 1070210705. Davis, T.J., Frits, S.A., Gryner, R., Orme, C.D.L., Bielby, J., Bininda Emmons, O.R.P., Cardillo, M., Jones, K.E., Gittleman, J.L., Ma ce, G.M. & Purvis, A. (2008) Phylogenetic trees and the future of mammalian biodiversity. Proceedings of the National Academy of Science 105 11556-11563.
83 DeFries, R., Asner, G.P., Achard, F., Justice, C., Laporte, N., Price, K., Small, C. & Townshend, J (2005) Monitoring tropical deforestation for emerging carbon markets. Reduction of tropical deforestation and climate change mitigation (eds P. Moutinho & S. Schwartzman), pp. 127. IPAM, Washington. Deshayes, P. (1986) La manera de cazar de los Huni Kuin: Una domesticacion silvestre. Extracta 5 7 10. Doughty, R.W. & Myers, N. (1971) Notes on the Amazon wildlife trade. Biological Conservation 3, 293 -297. Dumbrell, A.J., Clark, E.J., Frost, G.A., Randell, T.E., Pitchford, J.W. & Hill, J.K. (2008) Cha nges in species diversity following habitat disturbance are dependent on spatial scale: theoretical and empirical evidence. Journal of Applied Ecology 45 1531 1539. Emmons, L.H. (1984) Geographic variation in densities and diversities of non-flying mamm als in Amazonia. Biotropica, 16 210 -222. Escamilla, A., Sanvicente, M., Sosa, M. & Galindo Leal, C., (2000) Habitat mosaic, wildlife availability, and hunting in the tropical forest of Calakmul, Mexico. Conservation Biolog y, 14 1592 1601. Estrada. A., Coates -Estrada, R. & Meritt Jr., D. (1994) Non flying mammals and landscape changes in the tropical rain forest regions of Los Tuxtlas, Mexico. Ecography 17 229241. Fa. J.E. (2007) Bushmeat markets White elephants or red herrings? Bushmeat and livelihoods: Wildlife management and poverty reduction (eds G. Davies & D. Brown), pp. 47-60. Blackwell Publishing, UK. Fa, J.E. & Peres, C.A. (2001) Game vertebrates extraction in African and Neotropical forests: an intercontinental comparison. Conservation of exploited species (J.D Reynolds, G.M. Mace, K.H. Redford, & J.G. Robinson), pp. 203-241. Cambridge Press, UK. Fa, J.E. & Brown, D. (2009) Impacts of hunting on mammals in African tropical moist forests: a review and synthesis. Mammal Review 39, 23126 4. Fa, J.E., Johnson, P.J., Dupain, J., Lapuente, J., Kster, P. & MacDonald, D.W. (2004) Sampling effort and dynamics of bushmeat markets. Animal Conservation 7 409 416. Fa, J.E., Ryan, S.F. & Bell, D.J. (2005) Hunting vulnerability, ecological charac teristics and harvest rates of bushmeat species in afrotropical forests. Biological Conservation 121 167-176.
84 Fahrig, L. & Rytwinski, T. (2009) Effects of roads on animal abundance: An empirical review and synthesis. Ecology and Society 14 art. 21. F erguson, R.B. (1989) Game war? Ecology and conflict in Amazonia. Journal of Anthropological Research, 45, 179206. Ferrier, S. (2002) Mapping spatial pattern in biodiversity for regional conservation planning: where to from here? Systematic Biology, 51, 3 31363. Fisher, D.O. & Owens, P.F. (2004) The comparative method in conservation biology. Trends in Ecology and Evolution, 19 391-398. FitzGibbon, C. (1998) The management of subsistence harvesting: Behavioral ecology of hunters and their mammalian prey Behavioral ecology and conservation biology (ed T.M. Caro), pp. 449-473. Oxford University Press, UF. Fragoso, J.M.V. (1998) Home range and movement patterns of white -lipped peccary (Tayassu pecar i ) herds in the Northern Brazilian Amazon. Biotropica, 3 0 458 -469. Fragoso, J.M.V. (1999) Perception of scale and resource partitioning by peccaries: Behavioral causes and ecological implications. Journal of Mammalogy 80 9931003. Franke, I.L. (2008) Manejo Florestal Madeireiro e Conteno do Desmatamento no Estado do Acre. Secretaria de Estado de Florestas, Rio Branco. Franzen, M. (2006) Evaluating the sustainability of hunting: a comparison of harvest profiles across three Huaorani communities. Environmental Conservation 33 36 45. Fritz, S.A., Bininda -Emmons, O.R.P. & Purvis, A. (2009) Geographical variation in predictors of mammalian extinction risk: big is bad, but only in the tropics. Ecology letters 12 583 -549. Frost, P., Campbell, C., Medina, G. & Usongo, L. (2006) Landscape scale approaches fo r integrated natural resource management in tropical forest landscapes. Ecology and Society 11 art. 30. Gavazzi, R.A. (2007) Plano de Gesto Ashaninka. APIWTXA/AMAAI -AC/CPI AC, Rio Branco. Gavin, M.C. (2007) Foraging in the fallows: Hunting patterns ac ross a successional continuum in the Peruvian Amazon. Biological Conservation, 134, 64 -72.
85 Gitzen, R.A., Millspaugh, J.J. & Kernohan, B.J. (2006) Bandwidth selection for fixedkernel analysis of animal utilization distributions. Journal of Wildlife Manage ment 70, 1334 -1344. Grelle, C.E.V. (2005) Predicting extinction of mammals in the Brazilian Amazon. Oryx 3 347-350. Hames, R. (2007) The ecologically noble savage debate. Annual Review of Anthropology 36 177190. Hames, R.B. & Vickers, W.T. (1982) Optimal diet breadth theory as a model to explain variability in Amazonian hunting. American Ethnologist 9 358378. Hansen, A. (In preparation) Contribution of source -sink theory to Protected Areas science. Source, sinks, and sustainability across lands capes (eds J. Liu, V. Hull, A. Morizillo, & J. Wiens). H. Ronald Pulliam Symposium Proceedings. 2008 US IALE conference. Hansen, A.J. & DeFries, R. (2007) Ecological mechanisms linking protected areas to surrounding lands. Ecological Applications 17 974988. Hawkes, K., Hill, K. & O'Connell, J.F. (2008) Why hunters gather: Optimal Foraging Theory and the Ach of Eastern Paraguay. Environmental anthropology: A historical reader (eds M.R. Dove & C. Carpenter), pp 265283. Blackwell Publishing, MA. Hill, K. & Hawkes, K. (1983) Neotropical hunting among the Ach of Eastern Paraguay. Adaptive responses of native Amazonians (eds R.B. Hames, & W.T. Vickers), pp. 139-187. Academic Press, NY. Hill, J.K. & Hamer, K.C. (2004) Determining impacts of habitat modifi cation on diversity of tropical forest fauna: the importance of spatial scale. Journal of Applied Ecology 44, 744754. Hill, K., Padwe, J., Bejyvagi, C., Bepurangi, A., Jakugi, F., Tykuarangi, R. & Tykuarangi, T. (1997) Impact of hunting on large vertebr ates in the Mbaracayu Reserve, Paraguay. Conservation Biology 11 13391353. Hill, K., McMillan, G. & Farina, R. (2003) Hunting -related changes in game encounter rates from 1994 t0 2001 in the Mbaracayu Reserve, Paraguay. Conservation Biology 17 1312 -1 323. Hyrashi, S., Souza Jr., C., Sales, M. & Verissimo A. (2009) Transparncia florestal da Amaznia Legal (Dezembro 2009 e Janeiro 2010) IMAZON, Par.
86 IBGE (2009) Levantamento e classificao da cobertura e uso da terra. Coordenao de Recursos Naturais e Estudos Ambientais, Diretoria de Geocincias, IBGE, Rio de Janeiro. Iglesias, M.P. and Aquino, T.V. (2006) Gesto Territorial e Ambiental em Terras Indgena. Programa Estadual de Zoneamento Ecolgico -Econmico do Estado do Acre Fase II Governo do Estado do Acre, Secretaria de Estado de Meio Ambiente e Recursos Naturais, Rio Branco. INPE (2005) Prodes: assessment of deforestation in Brazilian Amazon. Instituto Nacional de Pesquisas Espaciais (INPE), So Paulo. IPAM & Consrcio MABE (2009) Identifi cao de reas crticas ao longo das estradas BR 364 e 317, estados do Acre, Rondnia e Amazonas Instituto de Pesquisa Ambiental da Amaznia, Belm. Isaac, N.B. & Cowlishaw, G. (2004) How species respond to multiple extinction threats. Pro ceedings of the Royal Society B 271 11351141. Jerozolimski, A. & Peres, C.A. (2003) Bringing home the biggest bacon: a cross -site analysis of the structure of hunter kill profiles in Neotropical forests. Biological Conservation 111 415-425. Juste, J., Fa, J.E., Pe rez del Val, J. & Castroviejo, J. (1995) Market dynamics of bushmeat species in Equatorial Guinea. Journal of Applied Ecology 32, 454467. Kensinger, K.M. (1975) Studying the Cashinahua. The Cashinahua of eastern Peru: Studies in Anthropology and Materi al Culture, Volume 1 (ed J.P. Dwyer), pp. 985. The Haffenreffer Museum of Anthropology, Brown University. Kensinger, K.M. (1983) On meat and hunting. Current Anthropology 24, 128129. Kensinger, K.M. (1995). How real people ought to live: the Cashinahua of Eastern Peru, 1st ed. Waveland Press, Illinois. Kinnaird, M.F., Sanderson, E.W., O'Brien, T.G., Wibisono, H.T. & Woolmer, G. (2003) Deforestation trends in a tropical landscape and implications for endangered large mammals. Conservation Biology 17 245-257. Lagrou, E. (2004) Sorcery and shamanism in Cashinahua discourse and praxis, Purus river, Brazil. In darkness and secrecy: the anthropology of assault sorcery and witchcraft in Amazonia (eds N.L. Whitehead and R. Wright), pp. 244271. Duke Univers ity Press Laurance, W.F & Useche, D.C. (2009) Environmental synergisms and extinctions of tropical species. Conservation Biology 23 1427 -1437.
87 Laurance, W.F., Vasconcelos, H.L. & Lovejoy, T.E. (2000) Forest loss and fragmentation in the Amazon: implications for wildlife conservation. Oryx 34, 39 -45. Laurance, W.F., Lovejoy, T.E., Vasconcelos, H.L., Bruna, E.M., Didham, R.K., Stouffer, P.C., Gascon, C., Bierregaard, R.O., Laurance, S.G. & Sampaio, E. (2002) Ecosystem decay of Amazonian forest fragment: a 22year investigation. Conservation Biology 16 605-618. Laurance, W.F., Croes, B.M., Tchignoumba, L., Lahm, S.A., Alonso, A., Lee, M.E., Campbell, P. & Ondzeano, C. (2006) Impact of roads and hunting on Central African rainforest mammals. Conservation Biology 20 12511261. Laurance, W.F., Croes, B.M., Guissouegou, N., Buij, R., Dethier, M. & Alonso, A. (2008) Impact of roads, hunting, and habitat alteration on nocturnal mammals in African rainforest. Conservation Biology 22 721732. Levi, T., Sh epard Jr., G.H., Ohl -Schacherer, J., Peres, C.A. & Yu, D.W. (2009) Modeling the long term sustainability of indigenous hunting in Manu National Park, Peru: landscape -scale management implications for Amazonia. Journal of Applied Ecology 46, 804814. Lewi s, S.L., Malhi, Y. & Phillips, O.L. (2004) Fingerprinting the impacts of global change on tropical forests. Philosophical Transactions of the Royal Society B 359, 437462. Lima, E.C. (2001) Erros repetidos: A pavimentao da BR -364 e os Katukina. Campos Revista de Antropol ogia Social 1 203214 Lima, E.C. (2002) Katukina. Enciclopdia da Floresta (eds M.C. Cunha & M.A. Barbosa), pp. 169176. Companhia das Letras, So Paulo. Ling, S. & Milner -Gulland, E.J. (2006) Assessment of the sustainability of bushmeat hunting based on dynamic bioeconomic models. Conservation Biology 20 1294 1299. Loyola, R.D., Kubota, U., Fonseca, G.A.B & Lewinsohn, T.M. (2009) Key Neotropical ecoregions for conservation of terrestrial vertebrates. Biodiversity and Conservation 18 2017 -2031. Lupo, K.D. & Schmitt, D.N. (2005) Small prey hunting technology and zooarchaeological of taxonomic diversity and abundance: ethnoarchaeological evidence from Central African forest foragers. Journal of Anthropological Archaeology 24 33 5 353. Malhi, Y., Roberts, J.T., Betts, R.A., Killeen, T.J., Li, W. & Nobre, C. (2008) Climate change, deforestation and the fate of the Amazon. Science 319 169 -172.
88 Mandujano, S. & Naranjo, E.J. (2010) Ungulate biomass across a rainfall gradient: a co mparison of data from Neotropical and palaeotropical forests and local analyses in Mexico. Journal of Tropical Ecology 26 13 -23. Margules, C.R. & Presley, R.L. (2000) Systematic conservation planning. Nature, 405, 243-253. Mena V.P., Stallings, J R., R egaladob, J. & Cueval (2000) The sustainability of current hunting practices by the Huaorani. Hunting for sustainability in tropical forests (eds J.G. Robinson & E.L. Bennett), 5778. Columbia University Press, NY. Michalski, F. & Peres, C.A. (2005) Anthr opogenic determinants of primate and carnivore local extinctions in a fragmented forest landscape of southern Amazonia. Biological Conservation 124 383-396. Michalski, F. & Peres, C.A. (2007) Disturbance mediated mammal persistence and abundance area relationships in Amazonian forest fragments. Conservation Biology 21 1626 -1640. Milner -Gulland, E.J. & Rowcliffe, J.M. (2007) Conservation and sustainable use: A handbook of techniques. Techniques in Ecology and Conservation Series Oxford University Pres s, UK. Milner -Gulland, E.J. Bennett, E.L. & SCB 2002 Annual Meeting Wild Meat Group (2003) Wild meat: the bigger picture. Trends in Ecology and Evolution 18 351357. Morton, D.C., Defries, R.S., Shimabukuru, Y.E., Anderson, L.O., Arai, E., Espirito San to, F.B., Freitas, R. & Morisette, J. (2006) Cropland extension changes deforestation dynamics in the southern Brazilian Amazon. Proceedings of the National Academy of Science 103 14637-14641. Nagaoka, L. (2002) The effects of resource depression on for aging efficiency, diet breadth, and patch use in southern New Zealand. Journal of Anthropological Archeology 21 419 -442. Naranjo, E.J. & Bodmer, R.E. (2007) Source -sink system and conservation of hunted ungulates in the Lancadon Forest, Mexico. Biologic al Conservation, 138, 412420. Nasi, R., Brown, D., Wilkie, D., Bennett, E., Tutin, C., van Tol, G., & Christophersen, T. (2008). Conservation and use of wildlife -based resources: the bushmeat crisis Secretariat of the Convention on Biological Diversity, Montreal, and Center for International Forestry Research (CIFOR), Bogor
89 Naughton -Treves, L., Mena, J.L., Treves, A., Alvarez, N. & Radeloff, C. (2003) Wildlife survival beyond park boundaries: the impact of slash andburn agriculture and hunting on mamm als in Tambopata, Peru. Conservation Biology 17 11061117. Navarro, J.G.G. (2004) Aprovechamiento de la fauna silvestre en comunidades cashinaua del ro Curanja y Purus Informe Tcnico I. World Wildlife Foundation, Lima. Nepstad, D., Schwartzman, S., Bamberger, B., Santilli, M., Ray, D., Schlesinger, P., Lefebvre, P., Alencar, A., Prinz, E., Fiske, G. & Rolla, A. (2006) Inhibition of Amazon deforestation and fire by parks and Indigenous Lands. Conservation Biology 20 65 -73. Neter, J, Wasseman, W. & Kutner, M.H. (1985) Applied linear statistical models: regression, analysis of variance and experimental designs IRWIN, Illinois. Norris, D., Peres, C.A., Michalski, F. & Hinchsliffe (2008) Terrestrial mammal responses to edges in Amazon forest patches: a study based on track stations. Mammalia 72 1523. Novaro, A.J., Redford, K.H. & Bodmer, R.E. (2000) Effect of hunting in source sink systems in the neotropics. Conservation Biology 14, 713 721. Ohl -Schacherer, J., Shepard Jr., G.H., Kaplan, H., Per es, C.A., Levi, T. & Yu, D.W. (2007) The sustainability of subsistence hunting by Matsigenka native communities in Manu National Park, Peru. Conservation Biology 21, 1174 -1185. Oliveira, H. & Bardales, N.G. (2006) Aptido natural de uso da terra no Esta do do Acre. Zoneamento Ecolgico-Econmico: Fase II SEMA, Rio Branco. Oliveira, P.J.C., Asner, G.P., Knapp, D.E., Almeyda, A., Galvn -Gildemeister, R., Keene, S., Raybin, R.F. & Smith, R.C. (2007) Landuse allocation protects the Peruvian Amazon. Science 317 1233 1236. Palmer, M. & White, P. (1994) Scale dependence and the species area relationship. American Naturalist 144, 717740. Parry, L., Barlow, J. & Peres, C.A. (2007) Large vertebrate assemblage of primary and secondary forest in the Brazilian Amazon. Journal of Tropical Ecology 23 653662. Parry, L., Barlow, J. & Peres, C.A. (2009a) Allocation of hunting effort by Amazonian smallholders: Implications for conserving wildlife in mixed -use landscape. Biological Conservation 142 17771786. P arry, L., Barlow, J., & Peres, C.A. (2009b) Hunting for sustainability in tropical secondary forests. Conservation Biology 23, 1270 -1280.
90 Pautasso, M. (2007) Scale dependence of the correlation between human population presence and vertebrate and plant s pecies richness. Ecology Letters 10 16 24. Peres, C.A. (1993) Biodiversity Conservation by Native Amazonians: a Pilot Study in the Kaxinaw Indigenous Reserve of Rio Jordo, Acre, Brazil World Wildlife Fund US, Washington. Peres, C.A. (2000) Effects o f subsistence hunting on vertebrate community structure in Amazonian Forests. Conservation Biology 14, 240253. Peres, C.A. (2001) Synergetic effects of hunting and habitat fragmentation on Amazonian vertebrates. Conservation Biology 15 1490 -1505. Per es, C.A. (2005) Why do we need megareserves in Amazonia. Conservation Biology 19, 728733. Peres, C.A. & Terborgh, J.W. (1995) Amazonian nature reserves: An analysis of the defensibility status of existing Conservation Units and design criteria for the future. Conservation Biology 9 34 -46. Peres, C.A. & Zimmerman, B. (2001) Perils in parks or parks in perils? Reconciling conservation in the Amazonian reserves with and without use. Conservation Biology 15 793797. Peres, C.A. & Lake, I. (2003) Exten t of non -timber resource extraction in tropical forests: Accessibility to game vertebrates by hunters in the Amazon Basin. Conservation Biology 17 521535. Peres, C.A. & Nascimento, H. (2006) Impact of game hunting by the Kayap of southeastern Amazoni a: implications for wildlife conservation in tropical forest indigenous reserves. Biodiversity and Conservation, 15 2627 -2653. Peres, C.A. & Palacios, E. (2007) Basin wide effects of game harvest on vertebrate population densities in Amazonian forests: I mplications for animal mediated seed dispersal. Biotropica, 39 304-315. Peres, C.A., Barlow, J. & Haugaasen T. (2003) Vertebrate responses to surface wildfires in a central Amazonian forest. Oryx 37 97109. PNUD (2003) Atlas de desenvolvimento humano no Brasil Programa das Naces Unidas para o Desenvolvimento. Version 1.0.0. Available from http://www.pnud.org.br/atlas/ accessed in 02/27/2010. Puertas, P.E. & Bodmer, R.E. (2004) Hunting effort as a tool f or community based wildlife management in Amazonia. People in nature: Wildlife conservation in South
91 and Central America (eds K. Silvius, R.E. Bodmer & J. Fragoso), pp. 123-135. Columbia University Press, NY. Ramos, R.M., 2005. Estratgia de caa e uso da fauna na Reserva Extrativista do Alto Juru, AC Master Thesis. Universidade de So Paulo. Redford, K.H. (1993) Hunting in the neotropics: A subsidy from nature. Tropical forests, people, and food: Biocultural interactions and applications to development (eds C.M. Hladik, A. Hladik, O.F. Linares, H. Pagesy, A. Semple, & M. Hadley), pp. 227246. UNESCO, Paris. Redford, K.H. & Robinson, J.G. (1987) The game of choice: patterns of indians and colonist hunting in the Neotropics. American Anthropologist 89, 650667. Redford, K.H. & Sanderson, S.E. (2000) Extracting human from nature. Conservation Biology 14 1362 -1364. Reyna Hurtado, R., Rojas -Flores, E. & Tanner, G.W. (2009) Home range and habitat preferences of white lipped peccaries ( Tayassu pecari ) in Calakmul, Campeche, Mexico. Journal of Mammalogy 90 1199 -1209. Rist, J., Rowcliffe, M., Cowlishaw, G. & Milner -Gulland, E.J. (2008) Evaluating measures of hunting effort in a bushmeat system. Biological Conservation 141 20862099. Rist, J. Milner -Gul land, E.J., Cowlishaw, G. & Rowcliffe, M. (2009) The importance of hunting and habitat in determining the abundance of tropical forest species in Equatorial Guinea. Biotropica 41, 700710. Robinson, J.G. (1996) Hunting wildlife in forest patches: an ephemeral resource. Forest patches in tropical landscape (eds J. Schelhas & R. Greenberg), pp. 11130. Island Press, Washington. Robinson, J.G. & Bodmer, R.E. (1999) Towards wildlife management in Tropical forests. Journal of Wildlife Management 63 1 -13. R obinson, J.G. & Bennett, E.L. (2004) Having you wildlife and eating it too: an analysis of hunting sustainability across tropical ecosystems. Animal Conservation 7 397 408. Robinson, J.G. & Redford, K.H. (1991) Sustainable harvest of Neo-tropical forest mammals. Neotropical wildlife use and conservation (eds J.G. Robinson & K.H. Redford), pp. 415429. University of Chicago Press, Chicago. Rowcliffe, J.M., Cowlishaw, G. & Long, J. (2003) A model of human hunting impacts in multi prey communities. Journa l of Applied Ecology 40 872 -889.
92 Rowcliffe, J.M., Milner -Gulland, E.J. & Cowlishaw, G. (2005) Do bushmeat consumers have other fish to fry? Trends in Ecology and Evolution 20 274 -276. Schwartzman, S. & Zimmerman, B. (2005) Conservation alliances with indigenous peoples of the Amazon. Conservation Biology 19 721727. Schwartzman, S., Moreira, A. & Nepstad, D. (2000) Rethinking tropical conservation: Perils in parks. Conservation Biology 14 1351 1357. Seaman, D.E. & Powell, R.A. (1996) An evaluati on of the accuracy of kernel density estimators for home range analysis. Ecology 77 2075 2085. Sierra, R., Rodriguez, F. & Losos, E. (1999) Forest resource use change during early market integration in tropical rain forests: The Huaroani of Upper Amazon ia. Ecological Economics 30 107119. Silveira, M., Torrezan, J.M. & Daly, D.C. (2002) Vegetao e diversidade arbrea na regio do Alto Juru. Enciclopdia da Floresta. (eds M.C. Cunha & M.B. Almeida.), pp. 65-75. Companhia das Letras, So Paulo. SIPAM (2008) Notas de Alerta 21, 23, 24, and 26. Programa de Monitoramento de reas Especiais ProAE. Sistema de Proteo da Amaznia. Available at: http://www.sipam.gov.br/proae/acre/2008/ Accessed in 8/2/2010. Sirn, A. (2006) Natural resources in indigenous peoples' land in Amazonia: A tragedy of the commons? International Journal of Sustainable Development and World Ecology 13, 363374. Sirn, A., Hambck, P. & Machoa, J. (2004) Including spatial heterogeneity and animal dispersal when evaluating hunting: a model analysis and an empirical assessment in an Amazonian community. Conservation Biology 18 13151329. Smith, E.A. (1983) Anthropological applications of optimal foraging theory: A critica l review. Current Anthropology 24, 625651 Smith, D.A. (2005) Garden game: shifting cultivation, indigenous hunting and wildlife ecology in western Panama. Human Ecology 33 505537. Smith, D.A. (2008) The spatial pattern of indigenous wildlife use in western Panama: implications for conservation management. Biological Conservation 141, 925-937. Soares -Filho, B., Nepstad, D.C., Curran, L.M., Cerqueira, G.C., Garcia, R.A., Ramos, C.A., Voll, E., McDonald, A., Lefebvre, P. & Schlesinger, P. (2006) Model ing conservation in Amazon basin. Nature, 440 520 -523.
93 Sombroek, W. (2001) Spatial and temporal patterns of Amazon rainfall: Consequences for the planning of agricultural occupation and the protection of primary forests. Ambio 7 388-396. Souza -Mazurek R.R.D., Pedrinho, T., Feliciano, X., Hilrio, W., Gerncio, S. & Marcelo, E. (2000) Subsistence hunting among Waimiri Atroari Indians in Central Amazonia, Brazil. Biodiversity and Conservation, 9 579596. Stearman, A.M. (1990) The effects of settler incursion on fish and game resources of the Yuqui, a native Amazonian society of eastern Bolivia. Human Organization 49, 373-385. Sunderland, T.C.H., Ehringhaus, C. & Campbell, B.M. (2008) Conservation and development in tropical forest landscape: A time to face the tradeoffs? Environmental Conservation, 34 276279. Takahashi, J. (2008) A literature review of the spider monkey, Ateles sp., with species focus on risk of extinction. PhD thesis, Swedish University of Agricultural Science. Terborgh, J. (20 04) Reflections of a scientist on the World Parks Congress. Conservation Biology 18 619-620. TNC (2009) Terrestrial ecoregional boundaries and assessments geodatabase. Available at: http://conserveonline.org/workspaces/ecoregional.shapefile accessed in 02/27/2010. Urquiza-Haas, T., Peres, C.A. & Dolman, P.M. (2009) Regional scale effects of human density and forest disturbance on large bodied vertebrates throughout the Yucatn P eninsula, Mexico. Biological Conservation 142 134 -148. Valentin, J.F., S., C.P., Gomes, F.C.R. & Santos, J.C. (2002) Tendncias da pecuria bovina no Acre entre 1970 e 2000. Embrapa, Rio Branco. Vickers, W.T. (1980) An analysis of Amazonian hunting yi elds as a function of settlement age. Studies in hunting and fishing in the Neotropics; Working on South American Indians (eds R.B. Hames & K.M. Kensinger), pp. 7-30. Bennington College, VT. Winterhalder, B. (1981) Foraging strategies in the boreal environment: An analysis of Cree hunting and gathering. Hunter -gatherer foraging strategies (eds B. Winterhalder & E.A. Smith), pp. 6698. University of Chicago Press, Chicago. Winterhalder, B. & Lu, F. (1997) A forager -resource population ecology model and imp lications for indigenous conservation. Conservation Biology 11 1354 -1364.
94 Woodroffe, R. & Ginsberg, J.R. (1998) Edge effects and the extinction of populations inside protected areas. Science 280, 2126 -2128. Zapata -Ros, G. Urgils, C. & Surez, E. (20 09) Mammal hunting by the Shuar of the Ecuadorian Amazon: Is it sustainable? Oryx 43, 375385.
95 BIOGRAPHICAL SKETCH Pedro A.L. Constantino was born in Rio de Janeiro, Brazil, and received his Bachelor of Science degree in biological s ciences at the Univer sidade Federal do Rio de Janeiro studying insect ecology. His research on gall midge development received Honor Mention from the Brazilian Botanical Society in 2004. Since then he works in collaboration with the local NGO Comisso Pr-ndio do Acre and many indigenous people of Acre. He helped the implementation of the natural resource use monitoring program in Indigenous Lands of Acre and currently collaborates in the Indigenous Agroforestry Agents Capacity -Building program of CPI AC. In 2007, Pedro was an ACLI visiting scholar at the University of Florida. Meanwhile he worked for the Rio de Janeiro Botanical Garden on the conservation of threatened Brazilian orchids