an introduction to wildlife conservation in the brazilian ...

An Introduction to Wildlife Conservation in the Brazilian Amazon

Carla C. Eisemberg Stephen J. Reynolds

A View from Northern Australia

AN INTRODUCTION TO WILDLIFE CONSERVATION IN THE BRAZILIAN AMAZON

A VIEW FROM NORTHERN AUSTRALIA

Carla C. Eisemberg, Stephen J. Reynolds

Brazilian Amazon Field Intensive Charles Darwin University Darwin, Australia

© Carla C. Eisemberg 2017 All rights reserved. Except under the conditions described in the Australian Copyright Act 1986 and subsequent amendments, no part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, duplicating or otherwise, without the prior permission of the copyright owner.

This work may be cited as: Eisemberg CC, Reynolds SJ. 2017. An Introduction to Wildlife Conservation in the Brazilian Amazon: A View from Northern Australia. Brazilian Amazon Field Intensive, Charles Darwin University, Darwin. Individual contributions may be cited thus: Hunter-Xenie H, Whittaker C, Ghidini AR. 2017. The Amazon Environment. pp. 5-12 in Eisemberg CC, Reynolds SJ (eds) An Introduction to Wildlife Conservation in the Brazilian Amazon: A View from Northern Australia. Brazilian Amazon Field Intensive, Charles Darwin University, Darwin.

National Library of Australia Cataloguing-in-Publication entry Eisemberg, Carla C., Reynolds, Stephen J., authors. An Introduction to Wildlife Conservation in the Brazilian Amazon : A View from Northern Australia / Carla C. Eisemberg, Stephen J. Reynolds 9780646973227 (paperback)

Published by Brazilian Amazon Field Intensive (BAFI) Charles Darwin University, Darwin

Contact: Carla C. Eisemberg Phone: 0401 737 884 Email: [email protected]

Cover design by Carla C. Eisemberg Cover photos by Stephen J. Reynolds, Hmalan Hunter-Xenie, Adam Bean, Sarah Perkins Proofreading by Patricia M Reynolds Printed in Darwin by UniPrint NT, 4 Ellengowan Dr, Brinkin NT 0909

978-0-646-97322-7

CONTENTS

Preface............................................................................................................................................. 1 Introduction ................................................................................................................................... 3 Chapter 1 – The Amazon Environment .................................................................................... 5 Chapter 2 – Amazon Wildlife Ecology .................................................................................... 13 Chapter 3 – The Origins of Amazon Biodiversity .................................................................. 25 Chapter 4 – Threats to Amazon Wildlife ................................................................................. 39 Chapter 5 – Wildlife conservation and management in the Amazon .................................. 51 Chapter 6 – Management of river turtles at the Rio Trombetas Biological Reserve – A case study of wildlife conservation in Amazon ....................................................................... 63 Chapter 7 – Northern Australia and Northern Brazil Connections ..................................... 73

PREFACE Arthur Georges For many of us the astounding diversity of life on earth is an endless source of amazement and wonder. The diversity of plants and animals on our planet is brought to life by the talented oration and spectacular footage of the David Attenborough series, but there is nothing like experiencing it for yourself. Bring in a healthy curiosity and a commitment to science as a path to discovery, and the world opens up – questions answered, more questions emerge like an endlessly unfolding rose. The beauty of life on earth and the complexity that has evolved over time given the long-standing stability of just-right conditions, is for me a never-ending source of awe. Nowhere better exemplifies the diversity of life on earth than the Brazilian Amazon. The Amazonian rainforests is widely recognized for its megadiversity. Remarkably, one in ten known species in the world lives in the Amazon rainforest. Among the most familiar are the caiman, jaguar, cougar, anaconda, electric eels, piranha, poison arrow frogs, night monkeys, day geckoes, vampire bats and sloths, but these are the more obvious tip of the iceberg. There are lots of reasons for biologists to be interested in visiting Brazil and the Amazon. For Australians, there are other reasons for a special interest in Brazil. Five hundred million years ago, the lands on which our two countries reside were part of the supercontinent of Pangaea (all lands). Pangaea included most of the land masses in today's southern hemisphere, including Antarctica, South America, Africa, Madagascar, Australia, New Guinea, New Zealand and India. Over time, this large mass of continental crust broke up to form Laurasia in the north and Gondwanaland in the south, about 200 million years ago. The southern hemisphere biota was

isolated from the northern hemisphere, and evolved its own unique floral and faunal composition – the marsupials flourished, the ratite birds, the platypus, the chelid turtles. Gondwana itself began to break up 167 million years ago to form the southern continents we see today. The links between the land masses of Australia and South America lingered for some time, perhaps as late as 45 million years ago, with Antarctica as an intermediary when climes were much warmer than today. So Brazil and Australia have a connection deep in time, and we see the history of this connection written in the unique fauna of our two countries and indeed, written into the genetic relationships among them as we move into the molecular age of biological science. The connections between Brazil and Australia provide a remarkable foundation for addressing the common challenges our two countries face in conserving our rich and unique heritage and biodiversity. Scientific exchanges of researchers and students allows us to communicate and understand our needs and capitalize on this commonality and foundation for generating new knowledge and acting on the challenges before us. It is in this context that biologist Dr Carla Eisemberg has pursued her passion to bring the Brazilian Amazon within reach of young Australians who have chosen biology as the career they wish to pursue. Carla began her studies working in Amazon on turtle biology, and having cut her teeth on working in remote areas of dense rainforest, came to Australia to undertake PhD studies in the driest inhabited continent on earth. Instead, she was sent to the megadiverse tropical rainforests of the Kikori River, in the lowlands of Papua New Guinea, to work with the local people on the challenges they faced with environmental 1

sustainability, and conservation of the pignosed turtle. There, she focused on environmental education and community-led conservation. Carla's series of children's books and radio plays, in three languages, are a delight. Now Carla, her co-author Stephen Reynolds, and the 2016 Brazilian Amazon Field Intensive students are keen to bring their experiences in Australia and Brazil within reach of a new generation of biologists, ecologists and wildlife managers in Australia, through this new and informative book. It is the perfect guide for Australian students wishing to visit the Brazilian Amazon via student exchanges and

excursions. Such excursions have been organized by the authors. The associated quality of experiential learning simply cannot be delivered in the classroom. These activities, complemented by this book, provide an essential basis for providing the role models and the motivation to further pursue a career in field biology. This book is a good read, providing insight to the unique biota and ecosystems of the Brazilian Amazon, the challenges we face in conserving the biodiversity of this megadiverse region, and the actions that are being taken to achieve this conservation.

Photo: S. Georges

Arthur Georges is Distinguished Professor in Applied Ecology at the University of Canberra and has visited the Amazon on several occasions. 2

INTRODUCTION Carla C. Eisemberg, Stephen J. Reynolds As the largest drainage system and rainforest in the world, the Amazon is constantly attracting national and international attention. Headlines can be both positive thanks to new amazing discoveries, and negative as deforestation has already destroyed over 15% of the Brazilian Amazon, and wildlife is becoming increasingly imperilled due to overharvest. While Amazon has attained an almost mythical status as the “lungs of the planet”, the eyes of the international community turn to the Amazon countries and

demand to hear of actions to protect this vast and irreplaceable environment. However, there are no simple answers to the complex environmental and social issues facing the Amazon. To better understand the past failures and successes of programs targeting the conservation of Amazon and its wildlife it is important to acknowledge the interconnectivity of its physical environment, its wildlife and the anthropological pressures initiated since European Colonization some 500 years ago. Book authors during the 2016 Charles Darwin University Brazilian Amazon Field Intensive, at the Rio Trombetas Biological Reserve.

The Amazon is one of the most biologically diverse regions in the world. Understanding ecological processes in this biome helps to inform management efforts now and into the future. The aim of this book is to provide an introduction to wildlife conservation in the Brazilian Amazon, which comprises 60% of the total Amazon biome. The first chapter of the book introduces the physical environment, providing context for the chapters that follow. The second chapter explores how environmental and biotic factors shape Amazon wildlife ecology, while the third chapter explores the reasons behind the exceptionally high species richness and endemism. Only after the Amazon environment and its ecology is understood, can we comprehend in depth the humaninduced threats to Amazon wildlife, as described in Chapter 4. In response to these

threats, there are many wildlife conservation management efforts within the Brazilian Amazon; some of these are discussed in Chapter 5. Conservation efforts in Brazil are not new, with the country hosting a growing network of National Parks and Reserves. While the first five chapters are very broad and encompass the entire Brazilian Amazon, Chapter 6 provides a specific example of wildlife conservation and management actions at a particular place: the Rio Trombetas Biological Reserve, one of the oldest protected areas in the Brazilian Amazon. Although the focus of this chapter is on the management and conservation of river turtles from the family Podocnemididae, many examples in this book are associated with these turtles. The reason being that the history of Amazon river turtle exploitation and subsequent protection 3

goes hand-in-hand with the birth and development of conservation and management programs in Brazil. Historically, freshwater turtles were not only a significant socioeconomic resource to riverine communities, but also an important part of their culture. However, since European colonization a huge number of adults and more than 210 million eggs have been harvested for food, oil and tools. This resulted in a massive decline in abundance and overall range. In order to avert further decline, 40 years ago the Brazilian Government initiated a long-term and widespread conservation project across the Amazon. Finally, Chapter 7 explores the unexpectedly close connection between northern Australia

and northern Brazil. These links are also mentioned briefly in other chapters. Despite the immense ocean between them, there are many similarities in the ecology and environmental management issues of northern Brazil and northern Australia. Both regions have a rich suite of wildlife, with a surprising number of evolutionarily related groups due to ancient Gondwanan connections. At the same time both areas present challenges in the field of conservation and sustainable development due to their remoteness from major population centres. The parallels between the two regions countenances the possibility of global learning and exchange of ecological knowledge, which will benefit research and management between these regions and tropical countries elsewhere.

To understand the connections between northern Australia and northern Brazil we need to look at the two regions (highlighted in green) from a different perspective.

This book was a result of a partnership between Charles Darwin University (CDU), the National Institute for Amazon Research (INPA), Amazon Society of Ichthyologists and Herpetologists (AIHA) and Chico Mendes Institute for Biodiversity Conservation, Brazilian Ministry of the Environment (ICMBio). It was sponsored by the Australian Government Council on Australia Latin

America Relations (COALAR) and the Turtle Conservation Fund (TCF). Although this book is primarily targeted at undergraduate and post-graduate environmental science students from Brazil and Australia, we believe it is relevant to anyone interested in understanding the complexities of Amazon wildlife conservation.

4

CHAPTER 1 – THE AMAZON ENVIRONMENT Hmalan Hunter-Xenie, Crystal Whittaker, André R. Ghidini Introduction The Amazon is the largest drainage system in the world. The Amazon Basin is around seven million square kilometres in area (Herrera et al. 1978), and covers approximately 40% of the South American continent (Swap et al. 1992). Of the countries through which it extends, Brazil has the largest proportion, at 63%, of the Basin (Villar et al. 2009). The Amazon River, the largest river by discharge on the planet, is 6 500 km in length and drains an area of 7.05 x 106 km2 (Hedges et al. 1986; Subramaniam et al. 2008). The River and its tributaries flow west to east through six countries, the last being Brazil. Near the city of Manaus (at about 15 m above sea level) the Negro River, one of the main effluents of the Amazon River, joins the Amazon (Franzinelli & Igreja 2002). At their confluence, the dark

waters of the Negro travel more than six kilometres alongside the muddy waters of the Amazon, creating the famous “Meeting of the Waters”, one of the most popular tourist attractions in the area (Fig. 1.1). After meandering across the South American continent (Davidson et al. 2012), the Amazon waters enter the Atlantic Ocean just south of the equator. The discharge is so great that the fresh water extends 100 km out to sea (Curtin 1986). The Amazon contributes up to 20% of riverine discharge to oceans globally (Gibbs 1972; Moura et al. 2016). Different parts of the catchment area receives between 1 500 to 3 000 mm of rainfall annually (Davidson et al. 2012), however, rainfall is not homogenous across the landscape (Villar et al. 2009).

Figure 1.1. Meeting of the Waters (encontro das águas), near the city of Manaus. Here the dark waters of the Negro River travel alongside the muddy waters of the Solimões for over six kilometres, before uniting and forming the Amazon River proper. Photo: T. Hanna.

The largest rainforest on earth, referred to as Amazonia or the Amazon rainforest, covers an area of 5.5 million km2 (Ritter et al. 2017). The Amazon is a vital ecosystem providing goods and services to local people such as food, medicine, cultural fulfilment and shelter to sustain their livelihoods. In addition, it

supports a tremendously rich biodiversity of flora and fauna. The Amazon rainforest supports one tenth of the world’s biodiversity, and helps regulate regional and global climate (Fearnside 1999; Foley et al. 2007). Unfortunately the Amazon faces many serious threats, one of which is deforestation, which 5

occurs at a staggering 2 million hectares per annum (Laurance et al. 2001; Chapter 4). Deforestation and hydroelectric dams also greatly alter water flow regimes, which have flow on effects to fauna and flora. For example, widespread deforestation could lead to an increase of runoff and river discharge by 20% even if precipitation remains at the same level (Foley et al. 2007). Understanding the ecological processes of the Amazon helps to inform management efforts now and into the future. Amazon Rivers and Lakes The Amazon River, including its upper part (also known as the Solimões River), has around 1 000 other rivers and streams flowing into it. These tributaries originate from different geological areas (Duncan & Fernandes 2010). The complexities of the hydrological system are exemplified by the fact that the high water mark in Manaus is reached four to six weeks after the maximal rainfall period in the Andes, which contributes a flood wave from the headwaters. Not only does the Amazon Basin have large river systems and streams, but it also contains thousands of lakes which are either interconnected or separated according to flood cycles. During high water, lakes are connected to rivers via channels. The lakes thus fluctuate at the same water level as the rivers, until the dry season, when they become disconnected. At times the lakes have only about 1 m of water. Other lakes that run parallel to the Amazon River are known as “Ria” Lakes. These are flooded valleys and are also connected to the river via channels (Junk 1997). River flow is influenced by several factors. Early research in the Amazon focussed on river ecology to compare and contrast with knowledge of river function in temperate regions. Of particular relevance to ecological processes in the Amazon River and its tributaries are the river continuum and flood

pulse concepts. The river continuum concept argues that energy flows downstream in interconnected river systems. Rivers are classified by order according to the number of streams feeding the main channel. It includes small streams (one to three tributaries), medium size rivers (four to six tributaries) and large rivers (greater than six tributaries). It is assumed that the primary energy source comes from leaf litter, aquatic plants and organic matter. However, the river continuum concept was developed to describe river systems in temperate and forested watersheds, and it fails to incorporate the lateral input of energy and nutrients from wetlands and floodplains (Johnson et al. 1995). This is a limitation when considering the large rivers of the Amazon Basin, with their complex floodplain system. Hence, the flood pulse concept, originally developed for the Amazon (Junk et al. 1989), is more likely to better explain the complexities of the Amazon river systems. Water levels of the Amazon River and main tributaries fluctuate greatly during the year with periods of high rainfall inundating the floodplain (Fig. 1.2). The flood pulse concept, first enunciated by Junk et al. (1989), attempts to explain river flow dynamics by incorporating both lateral and longitudinal mechanisms. According to this concept, the floodplain alternates between aquatic and terrestrial environments and is termed the aquatic – terrestrial transition zone (ATTZ). Floodplains are areas inundated by water from the lateral overflow of rivers and streams, directly from rainfall or rising water tables (Junk et al. 1989). The ATTZ is traversed by a “moving littoral”, which is the inshore edge of the aquatic environment. Large rivers, their catchments and floodplains act as a functional unit in times of flooding and exchange energy and nutrients, organisms and biomass. This process initiates a temporary change in the landscape which may be as severe as a change from a lake to a desert (Junk et al. 1989). 6

Therefore, the littoral community have their life cycle adjusted to this change in the level of inundation. The Amazon Basin is prone to cyclic flooding events. The flooding predictability is very high in the Amazon, as

well as the long inundation period of the flood pulse. The flood pulse is important for many Amazon species, including fish, spawning and migration being triggered when the waters rise (Junk & Furch 1993).

Figure 1.2. Water levels fluctuate greatly during the year in the main rivers. The Negro River has the typical four hydrological periods defined as rising, flood (high water), falling, and dry (low water) phases. However, the pattern of timing and water levels vary greatly among rivers. Figure adapted from Bittencourt and Amadio (2007).

Water types The water types of the Brazilian Amazon derive from different landscapes and differ in their ecologies. Runoff from the highlands plays an important part in determining sediment load and nutrient composition. The different water types are commonly referred to as white water, clear water and black water (Fig. 1.3). Each type has specific characteristics, including distinctive physicochemical properties (Table 1.1; Duncan & Fernandes 2010). An important contributing factor is the soil material at the origin of the water source feeding sections of the system. White water originates from the Andes and is nutrient rich (Junk 1984). These waters contain comparatively high concentrations of dissolved solutes, high sediment load and a pH of alkaline to neutral (Duncan & Fernandes 2010). White water rivers have a muddy appearance, as for example the Amazon and Madeira Rivers (Duncan & Fernandes 2010). White water tends to have higher biomass, dissolved solids and productivity (Putz 1997).

Black water rivers are usually a dark tea colour and have low nutrient content, productivity and plant and animal diversity as well as being acidic (low pH). The waters drain from the central Amazon Basin (Hedges et al. 1986), originating from areas of white sandy soils with depleted nutrients and heath forest. Black water systems usually have low photosynthetic activity due to the darker colour of the water and low light penetration (Sponsel, 1986). Low light penetration is common to all three water types to varying degrees and limits phytoplankton production (Hedges et al. 1986). The dark tea colour stain originates from high concentrations of dissolved organic carbon from vegetation and debris. Due to the conditions there are few macrophytes, which would normally break down decomposed vegetation matter, releasing nutrients (Junk 1997). Unlike the floodplains associated with white water systems, black water systems and adjoining floodplains have low primary productivity (Junk & Furch 1993).

7

Figure 1.3. Main white, clear and black water rivers of the Amazon Basin. White water rivers are rich in nutrients and originate from the geologically recent Andes, whereas black and clear water rivers originate from ancient Precambrian Shields. The area between the shields is known as the Central Amazon Floodplain. Map adapted from Goulding et al. (2003) and Eisemberg et al. (2017). Clear water rivers are the third main type and are mostly found in the centre of the Amazon Basin. Their source is in the northern and southern edges of the Basin (Duncan & Fernandes 2010). From these Precambrian Shield areas, clear water carries low quantities of sediments (Konhauser et al. 1994). These

rivers have a pH of acidic to alkaline (Duncan & Fernandes 2010). Clear water rivers are olive green in colour and more transparent than white and black water. Clear water is found in the Tapajós and Xingu Rivers (Duncan & Fernandes 2010).

Table 1.1. Comparison of water chemistry between the main Amazon water types. Adapted from Duncan and Fernandes (2010). Water type White Black Clear

Conductivity (µS/cm) 44.8 ± 24.8 17.0 ± 15.2 14.4 ± 13.1

pH 6.6 ± 0.2 4.5 ± 0.9 6.5 ± 0.4

Total Dissolved Solids (mg/l) 23.9 ± 17.8 7.1 ± 6.7 7.7 ± 5.6

Examples Solimões, Madeira Negro, Aturama Tapajós, Trombetas

Northern Brazil-Northern Australia connection The flood pulse concept is also important for understanding river ecology in northern Australia. During the wet season in the north of the Northern Territory there is a monsoonal period of around three months that causes lateral overflow from rivers to adjacent floodplains. Many fish species from a tributary of the East Alligator River, for example, reply on the flood pulse to initiate their breeding cycle and many other species migrate downstream to the lower floodplains during this time (Douglas et al. 2005). Fig. 1.4. shows this lateral overflow and inundation of the floodplains.

Figure 1.4. Lateral overflow of the East Alligator River during the 2017 wet season Photo: C. Whittaker

8

Vegetation types Hot and humid conditions make the Brazilian Amazon an ideal biome for tropical forest to develop, however vegetation cover across the landscape is heterogeneous. This is caused by multiple factors including topography, hydrological cycles, climate variation and soil type. At higher altitudes in the north of Brazil, montane forests exist near Pico da Neblina (Pires & Prance 1985). However, the majority of the Brazilian Amazon is under 200 metres above sea level (Pires & Prance 1985). In the plains country, grassland and savanna ecosystems are present (Hedges et al. 1986). Moving eastwards across the country, there are mangrove forests at the mouth of the Amazon (Menezes et al. 2008). Soils and topography are particularly important drivers of vegetation composition and structure. For example, areas on sloping terrain experience lateral movement of water which leaches the clay from the upper soil strata and results in areas of higher sand content and low nutrients. These sandy soils can result in heath vegetation consisting of scrubby trees and shrubs. Two important types of Amazon rainforest vegetation that will be discussed further are unflooded areas (terra-firme forests) and flooded areas (várzea and igapó; Fig. 1.5) (Pires & Prance 1985). The other vegetation types of the Brazilian Amazon are herbaceous shrubs and grasses, floating meadows, extensive grasslands as well as savanna in the upper reaches of the Rio Madeira River (Hedges et al. 1986). Floating meadows are composed of floating macrophytes that colonize the lakes during the rising water and can occupy a large percentage (approximately 30%) of the floodplain during high water (Junk & HowardWilliams 1984; Engle & Melack 1993). Várzea and igapó are flooded forests found at lower elevations, in close proximity to the Amazon River and its tributaries. River levels fluctuate greatly, increasing an average of seven to thirteen metres during high water season. Várzea are forests flooded by white water,

while igapós are forests flooded by black and clear waters (Prance 1979). For this reason várzea have higher net productivity than igapó (Parolin & Ferreira 1998). Approximately 150 000 km2 of the Amazon rainforest is made up of várzea and igapó habitats (Goulding 1993). Plants in these areas are tolerant of seasonal flooding (Worbes et al. 1992). Typical in várzea forests are softwood trees, palms, grasslands and estuaries (Pires & Prance 1985; Klinge et al. 1995). Unlike várzea forests, igapó forests are flooded by nutrient poor waters which result in sandier soils that cannot sustain rich plant diversity. Due to the lower nutrients the trees in igapó flooded forest are stunted and have sclerophyllous leaves, as well as adventitious root systems that provide support in sandy conditions (Prance 1979). The vast areas of the Amazon situated on land, and not subject to flooding, are known as terra-firme and make up the largest percentage of vegetation cover. Small terra-firme forest streams (igarape) are relatively cool (24oC), transparent and the fauna is mostly aggregated in the submerged litter, which may reach considerable densities (Walker 1985; Henderson & Walker 1986). Terra-firme forests usually have higher species richness than flooded forests (Hedges et al. 1986). The main sub-categories of terra-firme are dense forests, open forests without palms, open forest with palms, liana forests and dry forests. Dense forests contain the greatest biomass. Little sunlight reaches the ground so it lacks a grassy understorey. Trees are shallow-rooted because soil-water is available year-round. Nutrients are supplied from the decomposing litter on the ground (Pires & Prance 1985). Species composition differs for other terra-firme rainforest sub-categories. For open forests light availability is somewhat increased permitting shrubs to grow. Open forests with palms have more available moisture. They occur more frequently than open forests without palms. Liana forests are dominated by vines in addition to babaco palms and the 9

Brazil Nut tree. Dry forests occurs where the climate is less humid and rainfall more seasonal. Here the trees are semi-deciduous and this is a relatively small area of the

Amazon Basin (Pires & Prance 1985). Such examples suggest that several factors determine the flora that exists in a particular place.

Figure 1.5. Amazon flooded habitats: (a) igapó (Trombetas River – clear water) and (b) várzea (Amazon River – white water). Areas flooded by black water rivers are also called igapó. Photos: S. Reynolds.

Overview Differences in topography, soil, water level and duration of inundation influence which plants and animals are found in a particular location. The characteristics of the environment determine fauna and flora distribution patterns. Topography and slope, for example, influence water availability in the landscape, which is a limiting factor for many plants. Soil type influences the moisture content of soils depending on soil texture in the soil profile. Runoff in the landscape

contributes to the chemical makeup of water. This has a strong correlation in the water types of the Brazilian Amazon. Runoff from the Andes provides nutrient rich white water but the ancient Guiana shield supplies fewer nutrients, as observed in black water rivers. As such, this creates distinct habitats for particular flora and fauna (Janzen 1974; Henderson & Crampton 1997; Duncan & Fernandes 2010; Lujan et al. 2013). The next chapter will investigate how wildlife ecology in the Amazon is influenced by biotic and abiotic factors and interactions.

Did you know? In black water rivers, as vegetation decays, tannins are leached, which stains the water, making it look like tea in colour. Black water rivers in Australia include the Gordon, Pieman and Davey Rivers in Tasmania, Frankland and Kalgan Rivers in the south-west of Western Australia, and parts of Noosa River in Queensland.

10

Literature Cited Bittencourt MM, Amadio SA. 2007. Proposta para identificação rápida dos períodos hidrológicos em áreas de várzea do rio Solimões-Amazonas nas proximidades de Manaus. Acta Amazonica 37: 303-308. Curtin TB. 1986. Physical observations in the plume region of the Amazon River during peak discharge—II. Water masses. Continental Shelf Research 6: 53-71. Davidson EA, de Araújo AC, Artaxo P, Balch JK, Brown IF, Bustamante MM, Coe MT, DeFries RS, Keller M, Longo M. 2012. The Amazon basin in transition. Nature 481: 321328. Douglas MM, Bunn SE, Davies PM. 2005. River and wetland food webs in Australia’s wet–dry tropics: general principles and implications for management. Marine and Freshwater Research 56: 329-42. Duncan WP, Fernandes MN. 2010. Physicochemical characterization of the white, black, and clearwater rivers of the Amazon Basin and its implications on the distribution of freshwater stingrays (Chondrichthyes, Potamotrygonidae). Pan-American Journal of Aquatic Sciences 5: 454-464. Eisemberg CC, Reynolds SJ, Christian KA, Vogt RC. 2017. Diet of Amazon river turtles (Podocnemididae): a review of the effects of body size, phylogeny, season and habitat. Zoology 120: 92-100. Engle DL, Melack JM. 1993. Consequences of riverine flooding for seston and theperiphyton of floating meadows in an Amazon floodplain lake. Limnology and Oceanography 38: 15001520. Fearnside PM. 1999. Biodiversity as an environmental service in Brazil's Amazonian forests: risks, value and conservation. Environmental Conservation 26: 305-321. Foley JA, Asner GP, Costa MH, Coe MT, DeFries R, Gibbs HK, Howard EA, Olson S, Patz J, Ramankutty N. 2007. Amazonia revealed: forest degradation and loss of ecosystem goods and services in the Amazon Basin. Frontiers in Ecology and the Environment 5: 25-32. Franzinelli E, Igreja H. 2002. Modern sedimentation in the Lower Negro River, Amazonas State, Brazil. Geomorphology 44: 259-271 Gibbs RJ. 1972. Water chemistry of the Amazon River. Geochimica et Cosmochimica Acta 36: 1061-1066. Goulding M. 1993. Flooded forests of the Amazon. Scientific American-American Edition 268: 44-50. Goulding M, Barthem R, Ferreira E. 2003. The Smithsonian Atlas of the Amazon. Smithsonian Books, Washington. Hedges JI, Clark WA, Quay PD, Richey JE, Devol AH, Santos UDM. 1986. Compositions and fluxes of particulate organic material in the Amazon River. Limnology and Oceanography 31: 717-738. Henderson PA, Walker I. 1986. On the leaf litter community of the Amazonian blackwater stream Tarumazinho. Journal of Tropical Ecology 2: 1-16. Henderson PA, Crampton WG. 1997. A comparison of fish diversity and abundance between nutrient-rich and nutrient-poor lakes in the Upper Amazon. Journal of Tropical Ecology 13: 175-198. Herrera R, Jordan C, Klinge H, Medina E. 1978. Amazon ecosystems. Their structure and functioning with particular emphasis on nutrients. Interciencia 3: 223-231. Janzen DH. 1974. Tropical blackwater rivers, animals, and mast fruiting by the Dipterocarpaceae. Biotropica 6: 69-103. Johnson BL, Richardson WB, Naimo TJ. 1995. Past, present, and future concepts in large river ecology. BioScience 45: 134-141. Junk WJ, Furch K. 1993. A general review of tropical South American floodplains. Wetlands Ecology and Management 2: 231-8. 11

Junk WJ. 1984. Ecology of the várzea, floodplain of Amazonian whitewater rivers. Pages 215243 in Sioli H, editor. The Amazon. Dr. W. Junk, Dordrecht. Junk WJ, Howard-Williams P. 1984. Ecology of aquatic macrophytes in Amazonia. Pages 269293 in Sioli H, editor. The Amazon. Dr. W. Junk, Dordrecht. Junk WJ. 1997. The central Amazon floodplain: Ecology of a pulsing system. Ecological Studies, 126. Springer, New York. Junk WJ, Bayley PB, Sparks RE. 1989. The flood pulse concept in river-floodplain systems. Canadian Special Publication of Fisheries and Aquatic Sciences 106: 110-127. Klinge H, Adis J, Worbes M. 1995. The vegetation of a seasonal várzea forest in the lower Solimões river, Brazilian Amazonia. Acta Amazonica 25: 201-220. Konhauser K, Fyfe W, Kronberg B. 1994. Multi-element chemistry of some Amazonian waters and soils. Chemical Geology 111: 155-175. Laurance WF, Cochrane MA, Bergen S, Fearnside PM, Delamônica P, Barber C, D'Angelo S, Fernandes T. 2001. The Future of the Brazilian Amazon. Science 291: 438-439. Lujan NK, Roach KA, Jacobsen D, Winemiller KO, Vargas VM, Ching VR, Maestre JA. 2013. Aquatic community structure across an Andes‐to‐Amazon fluvial gradient. Journal of Biogeography 40: 1715-1728. Menezes MPM, Berger U, Mehlig U. 2008. Mangrove vegetation in Amazonia: a review of studies from the coast of Pará and Maranhão States, north Brazil. Acta Amazonica 38: 403-420. Moura RL, Amado-Filho GM, Moraes FC, Brasileiro PS, Salomon PS, Mahiques MM, Bastos AC, Almeida MG, Silva JM, Araujo BF. 2016. An extensive reef system at the Amazon River mouth. Science Advances 2: e1501252 Parolin P, Ferreira LV. 1998. Are there differences in specific wood gravities between trees in várzea and igapó (Central Amazonia). Ecotropica 4: 25-32. Pires JM, Prance GT. 1985. The vegetation types of the Brazilian Amazon. Pages 109-145 in Prance GT, Lovejoy TE, editors. Key Environments: Amazonia. Pergamon Press, Oxford. Prance GT. 1979. Notes on the vegetation of Amazonia III. The terminology of Amazonian forest types subject to inundation. Brittonia 31: 26-38. Putz R. 1997. Periphyton communities in Amazonian black-and whitewater habitats: Community structure, biomass and productivity. Aquatic Sciences 59: 74-93. Ritter CD, McCrate G, Nilsson RH, Fearnside PM, Palme U, Antonelli A. 2017. Environmental impact assessment in Brazilian Amazonia: Challenges and prospects to assess biodiversity. Biological Conservation 206: 161-168. Sponsel LE. 1986. Amazon ecology and adaptation. Annual Review of Anthropology 15: 67-97. Subramaniam A, et al. 2008. Amazon River enhances diazotrophy and carbon sequestration in the tropical North Atlantic Ocean. Proceedings of the National Academy of Sciences 105: 10460-10465. Villar JCE, et al. 2009. Spatio‐temporal rainfall variability in the Amazon basin countries (Brazil, Peru, Bolivia, Colombia, and Ecuador). International Journal of Climatology 29: 15741594. Walker I. 1985. On the structure and ecology of the micro-fauna in the Central Amazonian forest stream ‘Igarapé da Cachoeira’. Hydrobiologia 122: 137-152. Worbes M, Klinge H, Revilla JD, Martius C. 1992. On the dynamics, floristic subdivision and geographical distribution of várzea forests in Central Amazonia. Journal of Vegetation Science 3: 553-564.

12

CHAPTER 2 – AMAZON WILDLIFE ECOLOGY Michael Kelly, Helen Truscott, Tamara R. Andersen, Akemi Shibuya Introduction The Amazon is one of the most biologically diverse environments in the world, with many rare, endemic and unknown species (AzevedoSantos et al. 2016). It is estimated that the region supports some 40 000 plant, 427 mammal, 1 300 bird, 378 reptile, and more than 400 amphibians and 3 000 fish species (Da Silva et al. 2005; Lewinsohn & Prado, 2005). Wildlife ecology in Amazon is greatly influenced by both biotic interactions and environmental (abiotic) factors (Antunes et al. 2016), such as nutrient availability for primary production which ultimately supports higher trophic levels (i.e. food webs; Ward et al. 2014). It is also important to take into account co-evolutionary relationships between plants and herbivores, and the dispersal of seeds by fauna. Many plants invest heavily in defences against herbivory due to poor nutrient soils and the resulting high cost of growth (Janzen 1974). Central to many Amazonian ecosystems is the annual hydrologic cycle and associated flood pulse (Chapter 1). Within this Neotropical environment the flood pulse, with its varying river discharges, channel morphologies, water chemistry and seasonal variability, influences and further shapes the evolution and ecology of biotic communities (Roque et al. 2012). Understanding how these ecosystems function is crucial, as climate change and deforestation may alter important parameters which could jeopardize the system irreversibly (Röpke et al. 2016). Differences in nutrient availability in water and soils across the Amazon result in differences in primary production. Soils in terra-firme forests (non-flooded forests; Chapter 1) are usually highly weathered and nutrient poor (Nardoto et al. 2008). In these areas light and water availability are thought to be the controlling factors in forest dynamics and

biomass. When water is not a limitation, trees can be spectacularly tall and the biomass high. Over time, competition for light between the trees creates a closed canopy (Pires & Prance 1985). These systems are generally less productive than flooded areas (Myster 2016). Black water rivers and their floodplains (igapó, Chapter 1) typically display low primary productivity with reduced plankton and lack of aquatic plants due to their acidic nature, low light penetration and resulting low oxygen levels. This reduces the carrying capacity of these environments (Janzen 1974). Additionally, trees in the low nutrient white sand soils of black water systems allocate more primary productivity to defence and seed storage than to growth available for consumers (Janzen 1974). White water rivers and their floodplains (várzea, Chapter 1), on the other hand, support macrophyte communities (e.g. Echinochloa polystachya) and attain some of the highest known rates of primary productivity (Castello et al. 2013). Clear water rivers have an intermediate nutrient content, which is usually higher than black water but lower than white water rivers (Sioli 1965). These white water systems support more insect life, which is prey for secondary and tertiary consumers (Silva et al. 2013). Periphyton communities are dominated by green algae in white water and acid-resistant diatoms in black water (Putz 1997). The availability of resources in most Amazonian floodplains fluctuates seasonally with the flood pulse. During high water, increased nutrient inputs from flooded areas stimulate growth of micro-organisms, plants and invertebrates. Given the oscillating nature of trophic resources due to the hydrologic cycle, many secondary and tertiary consumers have developed dietary flexibility to maximise resource utilisation. For example, aquatic 13

fauna increase consumption of animal prey when macrophytes are less available, and vice versa (Mortillaro et al. 2015). Some species of fish change their diet from fruit to fish seasonally (Mérona & Rankin-de-Mérona 2004). Species of ungulates, such as the Red Brocket Deer (Mazama americana) and the Collared Peccary (Tayassu tajacu), switch from a frugivorous diet to a woody browse diet when flood waters are retreating (Bodmer 1990). Habitat heterogeneity The complex and heterogeneous habitats of the Amazon rainforest support a very high faunal diversity. The multiple niches are finely divided, enabling the coexistence of species that are similar in size and morphology. Vertical habitat segregation and resource

partitioning, for example, are believed to reduce interspecific competition (Vieira & Monteiro-Filho 2003). This vertical stratification is evident in terra-firme forest, with animal species utilizing ground level, understory or canopy levels exclusively (Peters et al. 2006). The great majority of the terrestrial fauna are invertebrates that forage in the leaf litter and soil of the forest floor. In contrast to the forest floor, the canopy conditions are characterised by intense sunlight, greater extremes of humidity and wind, higher water stress, and a smaller, pulsesupplied pool of nutrients. However, the forest canopy provides habitat for a diverse range of birds (Fig 2.1) and arboreal animals such as monkeys, small felids, frogs, lizards, snakes and sloths (Sponsel 1986; Nadkarni 1994).

Figure 2.1. Birds observed at the Musa Tower, 30 m above ground level, Adolpho Ducke Forest Reserve: (a) Araracanga or Scarlet Macaw (Ara macao), (b) Tucanuçu or Toco Toucan (Ramphastos toco), (c) Araçari-negro or Guianan Toucanet (Selenidera piperivora), (d) Macuru-de-pescoço-branco or White-necked Puffbird (Notharchus macrorhynchus). Photos: H. Hunter-Xenie, 2016.

14

Ants, one of the most abundant insects in the Amazonian system, can be divided into two main groups: soil ants and canopy ants (Lojka et al. 2010). Bat species show distinct ecological and morphological traits, adapted to their habitat segregation. Canopy specialists have long narrow wings, which are built for speed, but lack the manoeuvrability required by the insectivorous understorey bats. Frugivorous bats are more abundant in disturbed forest or edges, whereas carnivorous bats are rare in such areas, as they need the forest interior to provide food and other resources (Peters et al. 2006). Some mammals use multiple strata based on resource availability, such as the opossum, Didelphis aurita, which can be found in the understory or ground level (Grelle 2003). The density of the forest can also influence fauna and is of particular significance to small mammals, which rely on habitat structure for protection from predators (Lambert et al. 2006). Such species have adapted to overcome the limited visibility of highly dense forest. Arboreal primates use loud vocalisations to communicate over long distances. These are often low frequency vocalisations which attenuate slowly and penetrate the forest further (Bezerra et al. 2012). The fauna itself can also influence the structure of the terrestrial environment. Members of the ant family Formicidae, for example, are important ecosystem engineers. These soil-based ants alter the physical and chemical composition of soils, increasing drainage and aeration and consequently improving conditions for plants (Lojka et al. 2010). Herbivory The vegetation is subject to damage by a diverse array of vertebrate and invertebrate herbivores and, as a consequence, plants have evolved a variety of chemical, developmental, and phenological defences. Plants defend themselves from herbivory by having leaves that are low in nutritional quality, tough and with an array of secondary metabolites (i.e.

chemical defences). To avoid or reduce the impact of herbivores, some plants produce leaves during the dry season when herbivore numbers are low. They can also perform synchronous leaf-flushing, grow and expand leaves rapidly, or delay “greening” until the leaf is mature (Coley & Barone 1996). Plants minimise losses to herbivores by investing heavily in secondary metabolites. Mature leaves are tough, and investment in secondary metabolites is reduced. Young leaves have higher nutritional value and they experience a higher rate of damage by herbivores. As leaf toughness is an effective defence, young leaves rapidly toughen as soon as they reach full size. In black water igapó habitats, plants are constrained by low nutrient availability, and the species present often have high concentrations of toxic alkaloids, phenols and tannins to reduce leaf damage (Coley & Barone 1996). To escape damage by seed predators some plants engage in mast-fruiting (producing fruits or seeds synchronously in mass numbers, e.g. in the family Dipterocarpaceae). Most tropical lowland forest plants either display mast-fruiting or produce toxic seeds or seeds with alternative forms of protection (Janzen 1974). Herbivores have co-evolved with plants in response to their defences (Table 2.1; Coley & Barone 1996). A common response among insect herbivores involves diet specialization. In many cases, insects have developed adaptations to a set of host plant defences, such as tolerance to particular secondary metabolite toxins (Lamarre et al. 2016). Diet supplementation or expansion is a response frequently seen in mammals. Herbivorous mammals consume the highly nutritious young leaves when available. When only mature leaves with low nutritional value are available, these mammals will balance their low nutrient foliage diet with seeds and fruits (Coley & Barone 1996). Some herbivores from the Lepidoptera (butterflies and moths) have physiological 15

adjustments and can feed successfully on lownutrient foliage (Lamarre et al. 2016). Changing digestive physiology and morphology of the alimentary tract is a common adaptation in mammals, which also rely on microbiological symbionts to aid in the detoxification of secondary metabolites (Coley & Barone 1996). Morphological adaptations can also overcome plant defences. Uakari

Monkeys (Cacajao calvus), for example, have dental specialisations designed to break through hard-shelled fruits (Barnett et al. 2013). Behavioural changes are also seen in response to plant defence mechanisms. Many species of Lepidoptera feed predominantly in terra-firme clay forests, where plants invest less in chemical defences, resulting in more palatable foliage (Lamarre et al. 2016).

Table 2.1. Plant defence strategies against herbivory in the Amazon and the resulting adaptations of herbivores to counteract these strategies (based on Coley & Barone 1996; Barnett et al. 2013; Lamarre et al. 2016) Plant strategy

Defence mechanism

Adaptation

Rapidly expanding leaves Developmental

Behavioural Delaying greening

Phenological

Chemical

Physiological

Flushing leaves synchronously in large numbers Producing leaves seasonally when predators are less abundant Producing secondary metabolites as toxins Low nutritional value of leaves Toughness and high fibre content

Seed dispersal Interconnections between the Amazon flora and fauna take various forms. Many plants rely on animals as seed dispersal agents. Lowland Amazonia has the highest diversity of aquatic and terrestrial frugivorous vertebrates and the widest range of morphological fruit types (Hawes & Peres 2014). Frugivorous bats are important agents for terra-firme seed dispersal. It has been suggested they are also critical in forest regeneration following disturbance, as they disperse seeds of pioneer tree species (Peters et al. 2006). Likewise, birds such as

Physiological

Morphological

Herbivore response Diet supplementation or expansion Changes in movement and habitat utilisation Diet specialisation to tolerate plant toxins Diet specialisation to digest vegetation with high fibre content Microbiological symbionts Dental specialisations Digestive tract modifications

guans and curassows play an important role in forest regeneration as seed dispersers (Gorchov et al. 1995). Large primates, tapirs, agoutis (Fig 2.2a) and iguanas (Fig 2.2b) are also critical for the dispersal of seeds of largeseeded tree species (Antunes et al. 2016). Maximiliana maripa, a large-seeded palm, has its seed distributed short distances by fauna including Collared Peccaries (Tayassu tajacu), deer (Odocoileus virginianus and Mazama spp.) and primates, while the Tapir (Tapirus terrestris) disperses the intact palm seed over greater distances (up to 2 km away; Fragoso 1997).

16

The seasonal inundation of the floodplains has resulted in a system of seed dispersal via water and aquatic animals such as fish. Fish are the most important animal seed dispersers on the Amazon floodplains. The primary fruit-eating fish are the characids and catfish. Many plants of the flooded forests synchronise fruit production with the high water phase so their seeds may be distributed through the water (hydrochory; Goulding 1993) or by fishes (ichthyochory; Hurd et al. 2016). The association between inundation, seed dispersal, fish migration, and germination biology suggests that floodplain forests are highly integrated ecological systems that have developed over long periods (Kubitzki & Ziburski 1994).

Fish, the primary consumers of fruits and seeds in the flooded forest, can migrate over large distances carrying seeds in their stomach for up to a week, providing a wide dispersal range. Germination does not typically begin until the waters recede. The dormancy of some seeds is broken by exposure to the hypoxic conditions that are common in the still waters that form after the river level has fallen. These relationships in dispersal and germination demonstrate the long co-evolution of plants and their dispersal agents, and the important role of animals in ecosystem regeneration (Kubitzki & Ziburski 1994).

Figure 2.2. Many Amazon tree species rely on herbivorous species that act as seed dispersers such as the (a) Red-rumped Agouti (Dasyprocta leporina) and (b) Green Iguana (Iguana iguana). Photos: S. Reynolds, 2016.

Influence of water types The Amazon Basin is a mosaic of different water types (white, black and clear waters; Chapter 1) connected by the main stem of the Amazon River. The physicochemical characteristics of the different water types act as a barrier constraining the distribution and dynamics of aquatic habitats and biota (Venticinque et al. 2016). For example, the

distribution of members of the family Potamotrygonidae (a family of freshwater stingrays endemic to South America), is greatly influenced by this hydrological barrier (or hydrological filter). Potamotrygon leopoldi and Potamotrygon henlei are endemic to clear water habitats, while the Cururu Ray (Potamotrygon wallacei) is endemic to the acidic black waters of the Negro River (Carvalho, Rosa & Araújo, 2016). The Cururu Ray has physiological traits 17

and unusual gill epithelial morphology that allow it to tolerate acidic water. In contrast, Potamotrygon motoro and Potamotrygon orbignyi are spread throughout the different water types. Some species of freshwater stingray are widely distributed throughout the Amazon Basin, while others are restricted to a single river and its tributaries (Duncan & Fernandes 2010). Fish are the most diverse and abundant vertebrate group in the Amazon Basin, with over 3 000 species (Da Silva et al. 2005). Some species are widely distributed, occurring throughout all water types, whereas others cannot survive in the acidic black waters. Other fish are specialised in a particular water type, giving them an ecological advantage and affecting distribution patterns (Lima & AraujoLima 2004). A comparative study by Henderson and Crampton (1997) focused on fish diversity and density (abundance and biomass) in nutrient poor (black water - igapó) and richer (white water - várzea) habitats. White water habitats had higher diversity than black water habitats. As the white water environment is richer in nutrients and more productive, it is capable of supporting a more diverse fish assemblage. The relative abundance of fish species differs greatly between white water and black water habitats. For example, the electric fish, Brachyhypopomus brevirostris, is abundant in black water habitats, but is relatively uncommon in white water habitats. The most important determinants influencing fish presence is oxygen availability and the trophic relationships of the water system (Henderson and Crampton 1997).

Role of the flood pulse The Amazon fauna has evolved to respond to the seasonal flood pulse, and possesses a variety of morphological, physiological and ethological adaptations (Junk 1993). The seasonal changes affect dissolved oxygen concentrations, habitat availability, productivity, and predator-prey interactions (Gomez-Salazar et al. 2012). These factors affect species distribution, community composition, migratory behaviour and reproductive strategies. Amazonian river dolphins, for example, have smaller group sizes during high water in response to lower food availability, and display largest group sizes when fish are concentrated and abundant in dry periods (Gomez-Salazar 2012). Nearly all fish migration, reproduction and feeding strategies are synchronised to stages of the hydrologic cycle (Hurd et al. 2016). The flood pulse also imposes a seasonal pattern on the life cycles of terrestrial animals (Junk 1993). Terrestrial invertebrates display a range of responses to the rising waters (Fig. 2.3). Non-migratory invertebrates may become dormant or active underwater during high water levels, while others migrate vertically into differing forest strata or horizontally along the water line or into terra-firme forests (Adis & Junk 2002). Among vertebrates, there are migratory animals such as rodents, ungulates, terrestrial birds and tortoises (e.g. Chelonoidis carbonaria; Fig. 2.4a). In contrast, arboreal and scansorial primates, squirrels, canopy birds, and bats, retain accessibility to the floodplain forests all year round.

18

Figure 2.3. Responses of terrestrial invertebrates from the organic and upper inorganic soil layers to floodplain inundation. These responses enable them to survive during high water periods of the flood pulse. Modified from Adis and Junk (2002). Terrestrial animals are seasonally replaced by frugivorous fish, including characids and catfish, freshwater turtles (e.g. Matamata, Chelus fimbriata; Fig. 2.4b), and aquatic mammals (e.g. manatees, family Trichechidae, and the Boto Dolphin Inia geoffrensis). These species take advantage of the resources in flooded landscapes, including seeds, fruit pulp and arthropods (Hawes & Peres 2014). As the water rises, providing increased habitat and food resources, herbivores and omnivores are at an advantage. Opportunistic fish, as well as other aquatic species including turtles, adjust their diets to align with the newly available food resources. While the forest is flooded, these species supplement their diets with fruits or seeds. However, as the water recedes and floodplains shrink, density dependent processes are intensified. The floodplain environment during low water is advantageous to predators including large fish. Among herbivores, predation pressure increases as well as competition for space and food. Another consequence of the reduced water volume is a decrease in oxygen availability.

Amazonian fish have evolved a range of adaptations that enable survival in hypoxic conditions during low water. Behavioural responses are the most common, with those fish not resistant to hypoxia migrating back into main lakes or channels. There are also anatomical adaptations, such as the expansion of the lower lip to skim the oxygen rich water near the surface in several unrelated fish (e.g. species of Colossoma, Brycon, Triportheus and Mylossoma; Val & Almeida-Val 1995). Physiological and biochemical adaptations enable some fish to survive in anoxic conditions. Some species can alter ventilation rates, circulatory systems and blood cell chemistry (Val et al. 1998a,b). In many cases, changes are temporary; some facultative airbreathers, for example, will only breathe air during periods of aquatic hypoxia. Other changes are permanent. The Pirarucu (Arapaima gigas), a large obligate air-breathing fish, transitions from water to air breathing during its early development and must breathe air regularly for survival (Brauner et al. 2004).

19

Figure 2.4. During high water periods, terrestrial tortoises such as (a) the Red-footed Tortoise (Jabuti-piranga, Chelonoidis carbonaria) migrate to non-flooded areas, while freshwater turtles such as (b) the Mata-mata (Chelus fimbriata) migrate to the floodplain to take advantage of food resources. Photos: A. Bean and M. Kelly, 2016. The Amazon flood pulse has a strong influence on reproductive strategies. Invertebrates in seasonally flooded areas have strategies aligned to the varying hydrology, with larval or dormant periods timed with inundation. These tightly coupled strategies affect food abundance for the higher trophic levels, which in turn influences vertebrate reproductive strategies and timing, thus demonstrating the complex interconnectedness of Amazon systems (Adis & Junk

2002). During the flood season, many fish species utilise highly productive temporary habitats for reproduction and early development of offspring. These favourable conditions permit rapid growth, and a consequent larger body size, by the conclusion of the flood season. This increases offspring survival and chances of successful dispersal under the severe conditions of the subsequent dry season (Espirito-Santo et al. 2013).

Did you know? Side-necked turtles from the family Chelidae occur only in South America and Australasia. They display similar responses to seasonal flooding in the Amazon and Australia (Merrick et al. 2006). Studies on two northern Australian side-necked turtles (Chelodina rugosa and Elseya dentata) revealed that they are opportunistic feeders, changing their diet with the water level and the consequent changes in habitats and resources (Kennett & Tory 1996). Similar behaviour is seen in the omnivorous habits of the Amazonian Red Side-necked Turtle (Phrynops rufipes) (Caputo & Vogt 2008).

The majority of fish species in the Amazon have a distinct spawning season during the rising waters, when floodplains become accessible as offspring feeding grounds (Junk 1993). Many characiforms (e.g. Colossoma, fanily Charicidae, and Prochilodus, family Prochilodontidae) spawn in the main water channel when the water is rising. Their larvae

and juveniles disperse passively downstream and then enter the nursery habitats of the floodplain (Hurd et al. 2016). The nesting biology of podocnemidid river turtles (e.g. the Giant South American Turtle Podocnemis expansa) is also closely synchronised with the flood pulse. The nesting and incubation period occurs during the dry season, while hatching 20

coincides with the rising of the water level (Ferreira Júnior & Castro 2005; Eisemberg et al. 2016b). The unseasonal flooding of sandbanks has major impacts on offspring survival, being the major natural cause of embryo mortality in freshwater turtles throughout the Amazon Basin (Alves et al. 2012). Similarly, flooding is a major cause of nest mortality for the Black Caiman (Melanosuchus niger), another species that lays its eggs in the dry season (Villamarín-Jurado & Suárez 2007). Overview Amazonian wildlife has co-evolved in response to predators, competitors and food resources. This is typified by the arms race between herbivores and plants to utilise the few nutrients available from poor soils (Coley &

Barone 1996), and the mutualistic relationships between seeding plants and fauna dispersal agents (Hawes & Peres 2014). Wildlife interactions with the environment have likewise driven the evolution of varied and unique Amazonian species and ecosystems (Lojka et al. 2010), driven by habitat complexity (Vieira & Monteiro-Filho 2003), water type and the overarching influence of the flood pulse with its seasonal hydrologic variation (Espírito-Santo et al. 2013). The result is a biotic community renowned for its distinctiveness and high diversity (Myster 2016). Ecological interactions and resulting diversity have developed over geological time through changes in landscapes, connections and processes (Lovejoy & De Araújo 2000; Albert et al. 2006). The origins of this incredible biodiversity will be discussed in the next chapter.

Northern Brazil-Northern Australia connection Amazonian fish are not the only ones that have developed the ability to breathe air to survive aquatic hypoxia. Northern Australia’s Tarpon (Megalops cyprinoides) is also an air breather (Daniels et al. 2004). The northern Australian monsoonal tropics experience seasonal inundation of the floodplains. The resulting floodplain lakes are referred to as billabongs. These lakes become cut off from main water sources in the dry season, which can lead to hypoxic conditions and fish mortality. In 1999, hundreds of fish died at Shady Camp lake (Mary River system) as a result of hypoxia. Fish species not adapted to low-oxygen levels were unable to survive. Due to its air-breathing adaptations, the Tarpon was not amongst the dead fish (Townsend & Edwards 2003). Literature Cited Adis J, Junk WJ. 2002. Terrestrial invertebrates inhabiting lowland river floodplains of central Amazonia and Central Europe: a review. Freshwater Biology 47: 711-731. Albert JS, Lovejoy NR, Crampton WGR. 2006. Miocene tectonism and the separation of cisand trans- Andean river basins: evidence from Neotropical fishes. Journal of South American Earth Sciences 21: 14-27. Alves RRN, Vieira KS, Santana GG, Vieira WLS, Almeida WO, Souto WMS, Montenegro PFGP, Pezzuti JCB. 2012. A review on human attitudes towards reptiles in Brazil. Environmental Monitoring and Assessment 184: 6877-6901. Antunes AP, Fewster RM, Venticinque EM, Peres CA, Levi T, Rohe F, Shepard GH. 2016. Empty forest or empty rivers? A century of commercial hunting in Amazonia. Science Advances 2: e1600936.

21

Azevedo-Santos VM, Garcia-Ayala JR, Fearnside PM, Esteves FA, Pelicice FM, Laurance WF, Benine RC. 2016. Amazon aquatic biodiversity imperiled by oil spills. Biodiversity and Conservation 25: 2831-2834. Barnett AA, Ronchi-Teles B, Almeida T, Deveny A, Schiel-Baracuhy V, Souza-Silva W, Spironello W, Ross C, MacLarnon A. 2013. Arthropod predation by a specialist seed predator, the Golden-backed Uacari (Cacajao melanocephalus ouakary, Pitheciidae) in Brazilian Amazonia. International Journal of Primatology 34: 470-485. Bezerra BM, Souto AS, Jones G. 2012. Propagation of the loud “tchó” call of the golden-backed uakaris, Cacajao melanocephalus, in the black-swamp forests of the upper Amazon. Primates 53: 317-325. Bodmer R. 1990. Responses of ungulates to seasonal inundations in the Amazon floodplain. Journal of Tropical Ecology 6: 191-201. Brauner CJ, Matey V, Wilson JM, Bernier NJ, Val AL. 2004. Transition in organ function during the evolution of air-breathing; insights from Arapaima gigas, an obligate air-breathing teleost from the Amazon. The Journal of Experimental Biology 207: 14331438. Carvalho MD, Rosa RS, Araújo ML. 2016. A new species of Neotropical freshwater stingray (Chondrichthyes: Potamotrygonidae) from the Rio Negro, Amazonas, Brazil: the smallest species of Potamotrygon. Zootaxa 4107: 566-586. Caputo FP, Vogt RC. 2008. Stomach flushing vs. fecal analysis: the example of Phrynops rufipes (Testudines: Chelidae). Copeia 2008: 301-305. Castello L, McGrath DG, Hess LL, Coe MT, Lefebvre PA, Petry P, Macedo MN, Renó VF, Arantes CC. 2013. The vulnerability of Amazon freshwater ecosystems. Conservation Letters 6: 217-229. Coley PD, Barone JA. 1996. Herbivory and plant defences in tropical forests. Annual Review of Ecology and Systematics 27: 305-335. Da Silva J, Rylands A, Da Fonseca G. 2005. The fate of the Amazonian areas of endemism. Conservation Biology 19: 689-694. Daniels CB, Orgeig S, Sullivan LC, Ling N, Bennett MB, Schürch S, Val AL, Brauner CJ. 2004. The origin and evolution of the surfactant system in fish: Insights into the evolution of lungs and swim bladders. Physiological and Biochemical Zoology 77: 732-749. Duncan WP, Fernandes MN. 2010. Physiochemical characterization of the white, black, and clearwater rivers of the Amazon basin and its implications on the distribution of freshwater stingrays (Chondrichthyes, Potamotrygonidae). Pan-American Journal of Aquatic Species 5: 454-464. Eisemberg CC, Machado Balestra RA, Famelli S, Pereira FF, Diniz Bernardes VC, Vogt RC. 2016. Vulnerability of Giant South American Turtle (Podocnemis expansa) nesting habitat to climate-change-induced alterations to fluvial cycles. Tropical Conservation Science 9: 1940082916667139. Espírito-Santo HMV, Rodríguez MA, Zuanon J. 2013. Reproductive strategies of Amazonian stream fishes and their fine-scale use of habitat are ordered along a hydrological gradient. Freshwater Biology 58: 2494-2504. Ferreira Júnior PD, Castro PDTA. 2005. Nest placement of the giant Amazon River Turtle, Podocnemis expansa, in the Araguaia River, Goiás State, Brazil. Ambio 34: 212-217. Fragoso JM. 1997. Tapir-generated seed shadows: scale-dependent patchiness in the Amazon rain forest. Journal of Ecology 85: 519-529. Gomez-Salazar C. 2012. Ecological factors influencing group sizes of river dolphins (Inia geoffrensis and Sotalia fluviatilis). Marine Mammal Science 28: 124-142. Gorchov DL, Cornejo F, Ascorra CF, Jaramillo M. 1995. Dietary overlap between frugivorous birds and bats in the Peruvian Amazon. Oikos 74: 235-250. Goulding M. 1993. Flooded forests of the Amazon. Scientific American 268: 44-50.

22

Grelle CEV. 2003. Forest structure and vertical stratification of small mammals in a secondary Atlantic forest, Southeastern Brazil. Studies on Neotropical Fauna and Environment 38: 81-85. Hawes JE, Peres CA. 2014. Fruit-frugivore interactions in Amazonian seasonally flooded and unflooded forest. Journal of Tropical Ecology 30: 381-399. Henderson PA, Crampton WGR. 1997. A comparison of fish diversity and abundance between nutrient-rich and nutrient-poor lakes in the upper Amazon. Journal of Tropical Ecology 13: 175-198. Hurd LE, Sousa RG, Siqueira-Souza FK, Cooper GJ, Kahn JR, Freitas CE. 2016. Amazon floodplain fish communities: Habitat connectivity and conservation in a rapidly deteriorating environment. Biological Conservation 195: 118-127. Janzen DH. 1974. Tropical blackwater rivers, animals, and mast fruiting by the Dipterocarpaceae. Biotropica 6: 69-103. Junk WJ. 1993. Wetlands of tropical South America. Wetlands of the world: Inventory, ecology and management 1: 679-739. Kennett R, Tory O. 1996. Diet of two freshwater turtles, Chelodina rugosa and Elseya dentata (Testudines: Chelidae) from the wet-dry tropics of northern Australia. Copeia 1996: 409419. Kubitzki K, Ziburski A. 1994. Seed dispersal in flood plain forests of Amazonia. Biotropica 26: 30-43. Lamarre GPA, Amoretti DS, Baraloto C, Bénéluz F, Mesones I, Fine PVA. 2016. Phylogenetic overdispersion in Lepidoptera communities of Amazonian white-sand forests. Biotropica 48: 101-109. Lambert TD, Malcolm JR, Zimmerman BL. 2006. Amazonian small mammal abundances in relation to habitat structure and resource abundance. Journal of Mammalogy 87: 766-776. Lewinsohn TM, Prado PI. 2005. How Many Species Are There in Brazil? Conservation Biology 19: 619-624. Lima ÁCD, Araujo-Lima CARM. 2004. The distributions of larval and juvenile fishes in Amazonian rivers of different nutrient status. Freshwater Biology 49: 787-800. Lojka B, Krausová J, Štěpán K, Polesný Z. 2010. Assessment of insect biological diversity in various land use systems in the Peruvian Amazon. Pages 103-121 in Rojas N, Prieto R, editors. Amazon Basin: plant life, wildlife and environment. Nova Science Publishers INC, New York. Lovejoy NR, De Araújo MLG. 2000. Molecular systematics, biogeography and population structure of Neotropical freshwater needlefishes of the genus Potamorrhaphis. Molecular Biology 9: 259-268. Mérona BD, Rankin-de-Mérona J. 2004. Food resource partitioning in a fish community of central Amazon floodplain. Neotropical Ichthyology 2: 75-84. Mortillaro JM, Pouilly M, Wach M, Freitas CEC, Abril G, Meziane T.2015. Trophic opportunism of central Amazon floodplain fish. Freshwater Biology 60: 1659-1670. Myster R. 2016. The physical structure of forests in the Amazon Basin: a review. The Botanical Review 82: 407-427. Nadkarni NM. 1994. Diversity of species and interactions in the upper tree canopy of forest ecosystems. American Zoologist 34: 70-78. Nardoto GB, Ometto JPH, Ehleringer JR, Higuchi N, Da Cunha Bustamante MM, Martinelli LA. 2008. Understanding the influences of spatial patterns on N availability within the Brazilian Amazon forest. Ecosystems 11: 1234-1246. Peters SL, Malcolm JR, Zimmerman BL. 2006. Effects of selective logging on bat communities in the Southeastern Amazon. Conservation Biology 20: 1410-1421. Pires J, Prance G. 1985. The Vegetation Types of the Brazilian Amazon. Pages 110-144 in Prance G, Lovejoy T editors. Key Environments: Amazonia. Pergamon Press, Oxford. 23

Putz R. 1997. Periphyton communities in Amazonian black- and whitewater habitats: Community structure, biomass and productivity. Aquatic Sciences 59: 74-93. Röpke CP, Amadio S, Zuanon J, Ferreira EJG, Deus CPD, Pires THS, Winemiller KO. 2016. Simultaneous abrupt shifts in hydrology and fish assemblage structure in a floodplain lake in the central Amazon. Scientific Reports 7: 40170. Roque FDO, Lima DVM, Siqueira T, Vieira LJS, Stefanes M, Trivinho-Strixino S. 2012. Concordance between macroinvertebrate communities and the typological classification of white and clear-water streams in Western Brazilian Amazonia. Biota Neotropica 12: 83-92. Sponsel LE. 1986. Amazon ecology and adaptation. Annual Review of Anthropology 15: 67-97. Sioli H. 1965. Bemerkungen zur Typologie amazonischer Flusse. Amazoniana 1: 74-83. Townsend SA, Edwards CA. 2003. A fish kill event, hypoxia and other limnological impacts associated with early wet season flow into a lake on the Mary River floodplain, tropical northern Australia. Lakes and Reservoirs: Research and Management 8: 169-176. Val AL, Almeida-Val VMF. 1995. Fishes of the Amazon and their Environments. Physiological and Biochemical Features. Springer-Verlag, Heidelberg. Val AL, Silva MNP, Almeida-Val VMF. 1998a. Hypoxia adaptation in fish of the Amazon: a never-ending task. South African Journal of Zoology 33: 107-114. Val AL, Gonzalez RJ, Wood CM, Wilson RW, Patrick ML, Bergman HL, Narahara A. 1998b. Effects of water pH and calcium concentration on ion balance in fish of the Rio Negro, Amazon. Physiological and Biochemical Zoology 71: 15-22. Venticinque E, Forsberg B, Barthem R, Petry P, Hess L, Mercado A, Cañas C, Montoya M, Durigan C, Goulding M. 2016. An explicit GIS-based river basin framework for aquatic ecosystem conservation in the Amazon. Earth System Science Data 8: 651-661. Vieira EM, Monteiro-Filho ELA. 2003. Vertical stratification of small mammals in the Atlantic rain forest of south-eastern Brazil. Journal of Tropical Ecology 19: 501-507. Villamarín-Jurado F, Suárez E. 2007. Nesting of the Black Caiman (Melanosuchus niger) in Northeastern Ecuador. Journal of Herpetology 41: 164-167. Ward DP, Petty A, Setterfield SA, Douglas MM, Ferdinands K, Hamilton SK, Phinn S. 2014. Floodplain inundation and vegetation dynamics in the Alligator Rivers region (Kakadu) of northern Australia assessed using optical and radar remote sensing. Remote Sensing of Environment 147: 43-55.

Podocnemis sextuberculata hatchlings Illustration: Fernando A. Perini

24

CHAPTER 3 – THE ORIGINS OF AMAZON BIODIVERSITY Adam Bean, Lou Martini, Sarah E. Perkins, Maxley B. Dias Introduction The Amazon Basin has the highest species richness of any freshwater system in the world (IUCN 2016). Covering an area almost the size of Australia, and including the largest rainforest on Earth, the Amazon is home to one in 10 known species. Many more remain to be discovered, with 2 200 new species of vertebrates and plants described between 1999 and 2015 (Charity et al. 2016). This includes the surprising discovery of a new species of tapir, Tapirus kabomani, in 2013 (Cozzuol et al. 2013). The number of vertebrate species are dwarfed by estimates for plants and invertebrates, with over 40 000 species of vascular plants and more than 100 000 invertebrate species described (Da Silva et al. 2005). Similar to other biomes, the present biodiversity of the Amazon is a result of the accumulation of species over millions of years through speciation and immigration, and the loss of existing species through extinction and emigration (Rull 2011). Many hypotheses have been proposed to explain the high species richness and endemism of the Amazonian biota, and it remains an area of ongoing research and debate (Medeiros et al. 1997; Pereira et al. 2004; Kay et al. 2005; Smith et al. 2014). Important concepts Before exploring the origins of Amazon biodiversity, and four key events behind it, it is important to understand some concepts and processes leading to speciation, i.e. the formation of new species. One measure of diversity is simply species richness, i.e. the number of species that occupy an area. Alternatively, various indices exist that take into account the proportional abundance of different species (Hill 1973; Pielou 1977). Speciation is the evolutionary process in which

the mechanisms of reproductive isolation between populations are established, leading to the formation of new species. For this to occur, gene flow between populations must be prevented for long enough to allow sufficient genetic divergence to establish reproductive isolation (Mayr 1942). The three most common modes of speciation are allopatric, parapatric and sympatric (Fig. 3.1). They are based on the extent of geographic isolation between populations in which divergence occurs (Butlin et al. 2008). Allopatric speciation occurs when populations of the same species become isolated geographically and evolve separately (Mayr 1942). This can occur as the result of the separation of continents by plate tectonics, the uplift of mountains or the formation of large rivers, and is known as vicariance (Smith et al. 2014). In the absence of gene flow, the isolated populations diverge as a result of mutation, genetic drift and adaptation to different selective pressures (Hoskin et al. 2005). In parapatric speciation, two partially separated populations may diverge as a result of speciation across a sharp environmental gradient (Endler 1977). For this mode of speciation an ecological niche must be available in order for a new species to evolve. Heterogeneity among edaphic factors such as soil conditions may lead to selective pressure between adjacent populations of tropical trees (Fine et al. 2005). Sympatric speciation occurs when populations occupy the same geographic location with no extrinsic barrier to gene flow (Mayr 1963). In this case, speciation does not require reproductive isolation to be complete, but requires very strong forces of natural selection acting on heritable traits, as there is no geographic isolation to aid in isolation (Butlin et al. 2008).

25

Figure 3.1. Three recognised modes of speciation: (a) allopatric, where an extrinsic barrier exists between populations during divergence; (b) sympatric, where there is no extrinsic barrier; and (c) parapatric, where there is a partial extrinsic barrier. The process of speciation generally extends over multiple generations, in some cases over millions of generations (Butlin et al. 2008). Changes to the Amazon biome occurred over geological time, so it is necessary to understand the various named epochs and how many million years ago (mya) they occurred (refer to Table 3.1). However, speciation can occur rapidly. For example, it has been suggested that the rate of speciation of cichlid fishes in Lake Victoria (Africa) is as rapid as one species every 500 years (Kricher 2011). This rapid diversification by organisms into a multitude of new species is referred to as adaptive radiation (Schluter 2000). Other well-known examples include the radiation of Darwin’s Galapagos finches and the honeycreepers of Hawaii (Kricher 2011). In all three examples, a single ancestral species entering a new environment gave rise to a large number of phenotypically divergent species, each with traits that allow it to efficiently use different resources in their shared environment. The evolution of morphological and functional differences

allows them to exploit different ecological resources, permitting the coexistence of several closely related species in the same habitat (Schluter 2000). Key factors and events A contributing factor for Amazon diversity is its location, straddling the equator. More species are found in the tropics than in temperate zones, a phenomenon known as the latitudinal diversity gradient (Kricher 2011). Whether this is due to tropical regions experiencing higher levels of speciation or lower levels of extinction than other regions is the subject of ongoing debate. The size of the Amazon is another contributing factor, as larger areas consistently have more species (MacArthur & Wilson 1967). Speciation rates are generally higher in large areas as there are more opportunities for isolation and divergence (Rosenzweig 1995; Ricklefs 2004). Though eighty percent of the Amazon is tropical evergreen forest (Charity et al. 2016), it is not a uniform ecosystem.

26

Table 3.1. Summary of geological events from the Palaeozoic, 570 million years ago (mya), until the present (Holocene epoch) that have influenced South American biogeography and the Amazon. The super-continent Gondwanaland, where many ancestors of the Amazon’s extant species evolved, became a sutured landmass in the early Palaeozoic. During the midCarboniferous, Pangaea formed, as the earth’s land masses joined into a single entity. In the Jurassic, Pangaea broke into Laurasia (Northern Hemisphere), and Gondwanaland (Southern Hemisphere). South America broke away from Gondwanaland in the early Cretaceous. The uplift of the Andes towards the end of the Oligocene and into the Miocene, and the consequent formation of the Amazon Basin, was followed by rapid speciation. The formation of a land bridge with North America c.3 million years ago resulted in major movements of species between the formerly isolated continents, leading to significant alterations in the composition of ecological communities. Most recently the impact of Pleistocene climatic oscillations and the arrival of humans in the late Pleistocene have seen rapid speciation, followed by extinction. Eon

Era

Period

Epoch

Time (mya)

Quaternary

Holocene Pleistocene

2–0

Pliocene

5–2

Miocene

23 – 5

Oligocene

37 – 23

Eocene Paleocene

58 – 37 66 – 58

Neogene Cenozoic Tertiary Paleogene

Phanerozoic Mesozoic

Paleozoic

Cretaceous

144 – 66

Jurassic

208 – 144

Triassic Permian Carboniferous Devonian Silurian Ordovician

245 – 208 286 – 245 360 – 286 408 – 360 438 – 408 505 – 438

Cambrian

570 – 505

Allopatric speciation is widely acknowledged as the primary mechanism underlying Amazonian species richness and endemism (Ribas et al. 2011). The region can be considered an “archipelago” of areas of endemism separated by the major rivers (Da Silva et al. 2005). This was first noted by the founder of biogeography, Alfred Russel Wallace, who observed that for monkeys “the

Events Arrival of humans Climatic oscillations Land bridge with North America Expansion of the Amazon Basin Rise of the Andes Mountain range

South America breaks away from Gondwana Pangaea splits into Laurasia and Gondwana

Pangaea forms

Gondwanaland becomes a sutured landmass

species found on one side (of the river) do not occur on the other” with the Amazon, Rio Negro and Madeira forming the borders beyond which some species never passed (Wallace 1854). The large Amazon rivers formed after the Andean uplift are effective dispersal barriers, isolating populations and leading to allopatric speciation, especially for species restricted to terra-firme forests. 27

Birds such as the antbirds, who inhabit the dark understorey of the rainforest, are unlikely to cross large light gaps especially at the lower and wider portions of the Amazon River (Hayes & Sewlal 2004). A likely example of the river forming a barrier between populations and leading to speciation is the Dusky Antbird (Cercomacroides tyrannina) which occurs north of the Amazon, whilst the similar Blackish Antbird (C. nigrescens) is only found south of the river. The formation of the Andes has also created major physical barriers, fragmented previously contiguous populations and restricted gene flow, facilitating the evolution of new species. Baird’s Tapir (Tapirus bairdii), for example, is found only in lowland forest on the west side of the Andes, whilst the similar Brazilian Tapir (Tapirus terrestris) occurs throughout the Amazon Basin, but only on the eastern side of the Andes (Kricher 2011). One example of parapatric speciation is the evolution of tropical trees on nutrient-poor white sand soils in the Amazon. The diversity of forests on white sand soils differs significantly from that found in the surrounding nutrient-rich clay soil forests. It is suggested that speciation may have occurred as a result of strong selection for defence compounds to combat herbivory in white sand forests, sufficient to overcome gene flow from individuals on clay soil (Fine et al. 2006). Suggested examples of sympatric speciation are found in insects that become dependent on different host plants in the same area. Polyploidy, where organisms contain more than two sets of chromosomes, is also considered a form of sympatric speciation and is common in plants, especially grasses, where up to 70% of species are believed to be polyploid (Drès & Mallet 2002). The Amazon has a rich paleo-geographical history and sequential biogeographical changes are important when interpreting the current state of this diverse biome (Woodburne 2010; Rull 2006; Antonelli et al. 2009). Four events that can help to explain the origin of

Amazon’s biodiversity will be explored in this chapter: (1) the Amazon’s Gondwanan heritage, (2) the rise of the Andes, (3) South America’s land bridge with North America, and (4) the glacial cycles of the Pleistocene epoch. Gondwanan heritage The origins of many of the families and genera found today in the Amazon date back to when South America, like Australia, formed part of the great southern continent Gondwanaland (Marshall 1988). Gondwana formed in the early Palaeozoic (< 520 mya), when Australia, India, Madagascar and Antarctica (east Gondwana) merged with Africa and South America (west Gondwana) (Eagles et al. 2008). Gondwanaland combined with the remainder of the earth’s landforms during the midCarboniferous period around 330 mya, forming Pangaea (Eagles & König 2008). During the Jurassic (< 200 mya) Pangaea broke into Laurasia, which stayed in the northern hemisphere, and Gondwana, which steadily drifted into the southern hemisphere. By the beginning of the Cretaceous (< 145 mya) east and west Gondwana had separated. Africa and South America detached from each other around 95 mya, effectively dissolving the western portion of the supercontinent (Fig. 3.2) (Murphy et al. 2009). South America and Australia were connected via Antarctica for over 400 million years. It is not then surprising that South America and Australia share a number of plant and animal families (Woodburne & Case 1996). For example, frogs in the family Hylidae are found in both South America and Australia and share a common Gondwanan ancestor (Duellman et al. 2016). Similarly, the freshwater turtle family Chelidae represent a biological link between the two continents (Smith 2010). In addition, fossil evidence shows that an ancestral monotreme once lived in Argentina (Pascual et al. 1992). This Palaeocene era (< 66 mya) ancestral relative of the modern Platypus (Ornithorhynchus anatinus) 28

has changed perceptions as to the historical distribution of monotremes. It is now accepted that these mammals, although only extant in Australia and Papua New Guinea, were present as far as South America in west Gondwana (Pascual et al. 1992). Other likely Gondwanan relicts include Psittacines (parrots), Ratites (large flightless birds), lungfish, non-biting midges of the family

Chironomidae, and marsupials, which can be found today in South America and Australasia (Upchurch 2008; Wright et al. 2008; Nilsson et al. 2010; Krosch et al. 2011). Gondwanan heritage does not, however, account for Amazonian biodiversity in its entirety, and many more species have ancestral origins elsewhere.

Figure 3.2. The spatial relationships of Gondwanan proto-continents at 150 million years ago and 95 million years ago. Formation of the Amazon Basin Prior to the separation of Africa and South America in the early Cretaceous (< 145 mya), the Amazon River flowed from east to west and drained into the Pacific Ocean (Hoorn 2006). After South America broke away from Africa, the South American tectonic plate moved westwards and collided with the Nazca Plate in the Pacific (c.130 mya; Montgomery et al. 2001). The effects of this collision would continue to be seen for almost 100 million years with the gradual subduction (forcing under) of the Nazca Plate beneath the western rim of the South American Plate (Schellart 2008). In the late Oligocene and early Miocene (< 23 mya) the force and friction of this subduction pushed the continental crust of South America upwards, creating the Andes Mountain range or cordillera (Montgomery et al. 2001). As this occurred, the original exit point

of the Amazon River to the Pacific Ocean on the western side of the continent, was interrupted by the emerging mountain range, and the river reversed its flow, eventually carving out a new mouth on the eastern side of the continent, thus spilling into the Atlantic Ocean (Latrubesse et al. 2005; Latrubesse et al. 2010). Another important event related to the formation of the modern Amazon drainage basin was the early formation of the Amazon Fan (c. 9 mya), which separated the western Amazon from the Orinoco Basin (Fig. 3.3) (Dobson et al., 2001). The Andean uplift created a major continental barrier isolating humid lowland forests east and west of the Andes (Smith et al. 2014). It also created a multitude of new montane and riverine habitats, providing opportunities for colonisation, allopatry and adaptive radiation (Rull 2011). In addition, the tectonic changes 29

led to increased rainfall on the eastern flank of the Andes, followed by increased erosion, which significantly changed water and sediment supply to Amazon rivers. The resulting increase in nutrients and habitat heterogeneity is believed to have accelerated speciation for most taxa, including birds,

mammals, fishes and vascular plants, partially explaining why young areas in Western Amazonia have more species, genera, and families than the much older and less nutrient rich Guianan and Brazilian Shields (Hoorn et al. 2010).

Figure 3.3. Schematic model for the sequential isolation of drainage basins in northern South America due to tectonism. The Andes uplift had created a barrier by the late Oligocene to early Miocene (c. 23 mya). The rise of the Eastern Cordillera (~12 mya) isolated the Magdalena Basin. The initial formation of the Amazon Fan (~9 mya) separated the western Amazon Basin from the Orinoco Basin. Finally, the rise of the Merida Andes (~8 mya) separated the Maracaibo and Orinoco basins. Abbreviations: mya = millions of years ago. Note that time scale is not proportional. Adapted from Albert et al. (2006) and Hoorn et al. (2010). Land bridge with North America During the Great American Biotic Interchange (GABI) in the early to mid-Pliocene (< 4.5 mya), South America became connected to North America via Central America (Stehli & Webb 1985). The land bridge known as the Isthmus of Panama brought about one of the

most significant biotic changes in the history of Amazonian biodiversity. Prior to the connection, South America and North America were separated by ocean (sometimes referred to as the Central American Seaway). During the early Miocene (< 20 mya) the Cocos and Caribbean Plates collided, the 30

Cocos Plate subducting under the Caribbean Plate, generating volcanic activity and sediment upheaval (Webb 2006). A combination of tectonic events and a global drop in sea levels of up to 50 m formed the isthmus and connected the two continents (Marshall et al. 1982). This exchange, first recognised by Wallace (1876), significantly altered ecological communities on both continents (Hoorn et al. 2010). About 2.5 million years ago, 24 North American mammal genera dispersed southward, whilst 12 genera of South American mammals dispersed northward (Marshall et al. 1982). This included the immigration of North American ungulates, including camelids (guanacos and vicunas), tapirs, cervids (deer) and horses; proboscids (gomphotheres); carnivores including felids, canids, mustelids (weasels and otters), procyonids (racoons and coatis) and bears; rabbits, tree squirrels and a variety of murids (rodents; Marshall 1988). South American species that moved north included the now extinct ground sloths and terror birds, as well as the glyptodonts, pampatheres, and the notoungulate Mixotoxodon (Marshall 1988). Some notable species that invaded South America are the Capybara (Hydrochoerus hydrochaeris), the largest rodent in the world, the Giant River Otter (Pteronura brasiliensis), the largest of the mustelids, and the Jaguar (Panthera onca), the only extant ‘true’ big cat (Panthera) in the Americas (Webb 2006; Stehli & Webb 1985). Around half of South America’s mammals are a result of the GABI (Table 3.2), and many of these are found in the Amazon (Bofarull et al. 2008). The success of the southern emigration is highlighted by the number of extant genera of canids and cervids in South America, which is greater than in North America, and in the case of canids more than any other continent (Webb 2008). A number of North American immigrants later became extinct in North America, but survived in South America. As a

result, groups like tapirs and the camelids, ancestors of today’s guanacos and vicunas, show a disjunct distribution in that their only surviving relatives are found in Asia (Marshall 1988). Ranid frogs, elapid snakes and vipers from North America also diversified, with now more than 50 taxa in South America. The two families of turtles that entered the Brazilian Amazon since the GABI are Emydidae (sliders), and Geoemydidae (wood tortoises) (Stehli & Webb 1985). In the case of amphibians, salamanders from the family Plethodontidae have shown a high level of speciation in South America (Pinto-Sánchez et al. 2012). Given that many birds can migrate over oceanic divides, it is hardly surprising that many of the dispersal events of North American birds into the Amazon region may have happened much earlier than the GABI, and more than once. Indeed, northern birds may have invaded South America on up to 25 separate occasions, some as early as the Miocene (23 mya; Cody et al. 2010). The family Thraupidae (tanagers) are of North American origin and yet once in South America, the family diverged into multiple new genera and now constitute some 60% of the total species, with many of these in the Amazon. Swallows (Hirundinidae) and blackbirds (Icteridae) also diversified after reaching South America (Smith & Klicka 2010). Small, brightly coloured hummingbirds, found only in the neotropics, are believed to have radiated dramatically once they reached South America. Related to the eurasian swifts, the original ancestor of today’s hummingbirds migrated across the Bering Strait and North America before entering South America some 22 mya (McGuire et al. 2014). From a single common ancestor they have evolved rapidly with 338 extant species, of which 40% are found in the Andes (McGuire et al. 2014). The variety of habitats and ecological niches available in the Andes has provided 31

opportunities for hummingbirds to finely partition geographical space and radiate rapidly. According to McGuire et al. (2014) some lineages of hummingbirds have filled the available space in their environments, while

others are continuing to develop into new species at a rapid rate, with the potential to double the number of species over the next several million years before reaching an equilibrium.

Table 3.2. Approximate numbers of North American-derived terrestrial vertebrates by class, currently extant in South America. Numbers are approximate due to taxonomic variation, data deficient genera or discrepancies between taxonomists. Mammal numbers exclude marine and aerial species (i.e. bats). % S Am spp = North American-derived species as a percentage of total South American species. Genera

Species

% S Am spp.

Amphibia

5

~ 141

~7%

Reptilia

10

~ 62

~7%

Aves

160

~ 850

~ 24 %

Mammalia

176

~ 486

~ 49 %

Total

351

~ 1 569

~ 20 %

Class

As a result of these migrations, South America lost many of its most ancient mammal taxa including the Sabre-toothed Marsupial (Thylacosmilus), which is believed to have been out-competed by the placental Sabre-toothed Tigers (Smilodon spp.) from North America (Marshall 1988). Similarly, Australia’s marsupial carnivore the Tasmanian Tiger (Thylacinus cynocephalus) became extinct on the mainland after the introduction of the placental dingo. Aboriginal cave paintings in Kakadu, northern Australia, are one of the few remaining reminders of their presence. Though a significant number of South American groups became extinct, overall diversity of land mammals did not change, with the number of families in South America rising from 32 to 39 after the interchange began, and falling back to 35 at present (Marshall et al. 1982).

Examples Common name Scientific name Amazon Lungless Bolitoglossa Salamander altamazonica Bushmaster Lachesis muta Snake Masked Crimson Ramphocelus Tanager nigrogularis Jaguar Panthera onca Short-eared Dog Atelocynus microtis Giant Otter Pteronura brasiliensis Marsh Deer Blastocerus dichotomus

Pleistocene climatic oscillations The Pleistocene, a period of global climatic oscillations from about 2.6 million to 11.7 thousand years ago (Rull 2011), imposed significant changes on the neotropical landscape, thus promoting changes in Amazon ecological communities (Garzón‐ Orduña et al. 2015). It is suggested that glacial and interglacial periods caused relatively rapid allopatric speciation in plants and animals. It is hypothesised that certain clades diverged as a result of differences in local conditions leading to changes in morphology and phenology (Lynch Alfaro et al. 2012; Peterson & Ammann 2013). This is known as the Pleistocene Refugium Hypothesis (PRH; Haffer 1969). According to the PRH, during glacial ‘drying’ periods, many parts of the Amazon rainforest receded and changed structurally towards 32

transitional dry forests and open savanna (Peterson et al. 2013). The remaining rainforest became spatially fragmented (Fig 3.4). During interglacial ‘wetter’ periods, forests expanded, with the newly speciated populations returning to sympatry, which may explain why so many morphologically similar species can be found in Amazonia today. According to Garzón‐Orduña et al. (2014), the PRH is a foundation that helps explain the diversity of extant Amazonian taxa. Evidence from butterflies and birds suggests that most extant species arose during the Pleistocene, including 72% of butterflies. Such evidence has been used by proponents of the PRH to support the model as the major driver of speciation in the Amazon (Garzón‐Orduña et al. 2015). However, the PRH has been challenged over the past 20 years. Opponents of the PRH argue that fossil pollen data used to propose the PRH are limited and do not encompass the vast spatial-temporal variation of the Pleistocene (Haffer 1997; Poelchau et al. 2013). Other arguments against the PRH consider that allopatric vicariance was caused by physiographic changes such as riverine barriers (River-refuge hypothesis) and marine incursions, all of which predate the Pleistocene (Fig. 3.4) (Haffer 1997). Capuchin monkeys (genus Sapajus), for example, show evidence of sympatric speciation as a result of ancestral radiation from the Atlantic forests into the Amazon where, as genetic research shows, they now exist among their founding genus, Cebus (Lynch Alfaro et al. 2012). This

predates the Pleistocene epoch, with clade divergence over four million years earlier. Fossil insight into species diversification during the Pleistocene does, at least for now, show that there is no consensus or a ‘one size fits all’ explanation. While origins of certain clades predate the Pleistocene, some indeed show heightened speciation events during this epoch (Garzón‐Orduña et al. 2015). This period of rapid speciation was followed by a period of major extinction, particularly of larger vertebrates (megafauna), in the late Pleistocene. As in Australia, the cause of the megafauna extinction in South America is hotly debated with supporters of the human over-kill hypothesis suggesting that the demise of megafauna was the result of increased predation and changes of habitat, especially as a result of burning of vegetation by humans. Megafauna were particularly susceptible to over-kill due to their slow breeding and relatively small populations (Stuart 2015). Others argue that climate change was the key factor, with most of South America’s megafauna adapted to the open areas which were prevalent during the dry and cold climate of the mid to late Pleistocene. It is most likely, however, that the extinctions were a result of a combination of factors, with humans arriving at the end of the last glacial period, a time of intensified climate change when the biomass of megafauna was considerably lower, and more vulnerable to over exploitation (Cione et al. 2003).

Did you know? The Pleistocene marked the beginning of the end for much of the world’s megafauna. The megafauna extinctions in South America are considered more extensive than on any other continent, with the loss of over 80% of genera weighing 45 kg or more, including all species of horses (Stuart 2015). Prior to the megafaunal extinctions, South America had some 25 megaherbivore species weighing more than 1 000 kg, consisting of gomphotheres, camelids, ground sloths, glyptodonts and toxodontids. This compares with Africa’s current total of six species over 1 000 kg (Martin 2005). At present, Baird’s Tapir is South America’s largest mammal and weighs 350 kg.

33

Figure 3.4. Schematic representation of three speciation models for Amazonia. According to the Riverine barrier hypothesis (purple), dispersal is relatively uninhibited at the headwaters regions of broad rivers. The River-refuge hypothesis (pink) suggests that rainforest contracted on broad latitudinal fronts during Cenozoic dry climate periods, isolating forest animals between the broad lower courses of rivers (though narrower than at other times they still acted as barriers). The Refuge hypothesis (green) posits that dry climatic periods isolated forest refugia, which existed adjacent to pronounced surface relief in peripheral regions. Wooded savannas, open liana forests and humid gallery forests covered the regions between forest refugia. The Interglacial Amazonian embayment (yellow), formed when the sea-level was raised by several meters, probably also separated many populations in northern and southern Amazonia. Adapted from Haffer (1997). Overview The origins and development of Amazon biodiversity has been a long and complex process, strongly linked to tectonism and climate change and most recently to anthropogenic influences (Hoorn et al. 2010). Scientific explanations are continually developing and it is important to remember that the various theories explaining the development of biodiversity within the Amazon are not necessarily incompatible with each other. Biology is a complex science and speciation is a continual process. It is likely that the evolution and development of such intricate biological communities over geological time occurred as a result of multiple factors. Though a single definitive answer may

never be found, such research is vital to develop our understanding of speciation, adaptation and extinction. This knowledge may be applied to future development to minimize negative human impacts on such communities. The sheer size of the Amazon and its relative remoteness suggests that many more species remain to be discovered. It is estimated that only 2-10 % of insects have been described to date (Hoorn et al. 2011). In addition, the Amazon region may contain as many as 6 000 to 8 000 fish species, approximately double the 3 000 species currently described (Macedo & Castello 2015). With Amazon biodiversity under increasing threat, the need to discover and describe species is now even more important (Elmer & 34

Cannatella 2008). Species declines are often a result of land clearing or other anthropogenic pressures (Serrano-Rojas et al. 2017). The

main threats to Amazon wildlife and their consequences will be discussed in the next chapter.

Northern Brazil–Northern Australia connection Australia experienced a significant loss of megafauna during the Late Pleistocene. It is estimated that 29 mammal species weighing 50 kg or more became extinct (Webb 2008; Stuart 2015). Most of the Australasian megafauna is thought to have gone extinct about 46 000 years ago, significantly earlier than their South American counterparts. At this time the giant wombat-like mammal Diprotodon optatum, weighing approximately 2 000 kg, is considered to have been the largest known Australian mammal. Today the Red Kangaroo (Macropus rufus), at 85 kg, is Australia’s largest mammal, and one of only a few species, along with the Grey Kangaroo, Emu, Cassowary and Saltwater Crocodile, that qualify as megafauna. Literature Cited Albert JS, Lovejoy NR, Crampton WGR. 2006. Miocene tectonism and the separation of cisand trans-Andean river basins: Evidence from Neotropical fishes. Journal of South American Earth Sciences 21: 14-27. Antonelli A, Nylander JA, Persson C, Sanmartín I. 2009. Tracing the impact of the Andean uplift on Neotropical plant evolution. Proceedings of the National Academy of Sciences 106: 9749-9754. Badets M, Whittington I, Lalubin F, Allienne J-F, Maspimby J-L, Bentz S, Du Preez LH, Barton D, Hasegawa H, Tandon V. 2011. Correlating early evolution of parasitic platyhelminths to Gondwana breakup. Systematic Biology 60: 762-781. Bofarull AM, Royo AA, Fernandez MH, Ortiz-Jaureguizar E, Morales J. 2008. Influence of continental history on the ecological specialization and macroevolutionary processes in the mammalian assemblage of South America: Differences between small and large mammals. BioMed Central Evolutionary Biology 8: 97 Butlin RK, Galindo J, Grahame JW. 2008. Sympatric, parapatric or allopatric: the most important way to classify speciation? Philosophical Transactions of the Royal Society of London B: Biological Sciences 363: 2997-3007. Charity S, Dudley N, Oliveira D, Stolton SE. 2016. Living Amazon Report 2016: A Regional Approach to Conservation in the Amazon. WWF Living Amazon Initiative, Brasília and Quito. Cione AL, Tonni EP, Soibelzon LH. 2003. The broken zig-zag: Late Cenozoic large mammal and tortoise extintion in South America. Revista del Museo Argentino de Ciencias Naturales 5: 1-19. Cody S, Richardson JE, Rull V, Ellis C, Pennington RT. 2010. The great American biotic interchange revisited. Ecography 33: 326-332. Cozzuol MA, Clozato CL, Holanda EC, Rodrigues FH, Nienow S, de Thoisy B, Redondo RA, Santos FR. 2013. A new species of tapir from the Amazon. Journal of Mammalogy 94: 1331-1345. Da Silva J, Cardoso M, Rylands AB, Fonseca D, Gustavo A. 2005. The fate of the Amazonian areas of endemism. Conservation Biology 19: 689-694. Dobson DM, Dickens GR, Rea DK. 2001. Terrigenous sediment on Ceará Rise: a Cenozoic record of South American orogeny and erosion. Palaeogeography, Palaeoclimatology, Palaeoecology 165: 215–229. 35

Drès M, Mallet J. 2002. Host races in plant–feeding insects and their importance in sympatric speciation. Philosophical Transactions of the Royal Society of London B: Biological Sciences 357: 471-492. Duellman WE, Marion AB, Hedges SB. 2016. Phylogenetics, classification, and biogeography of the treefrogs (Amphibia: Anura: Arboranae). Zootaxa, 4104: 1-109. Eagles G, König M. 2008. A model of plate kinematics in Gondwana breakup. Geophysical Journal International 173: 703-717. Elmer KR, Cannatella DC. 2008. Three new species of leaflitter frogs from the upper Amazon forests: cryptic diversity within Pristimantis "ockendeni" (Anura: Strabomantidae) in Ecuador. Zootaxa 1784: 11-38. Endler JA. 1977. Geographic Variation, Speciation, and Clines. Princeton University Press, Princeton. Fine PVA, Daly DC, Munoz GV, Mesones I, Cameron KM. 2005. The contribution of edaphic heterogeneity to the evolution and diversity of Burseraceae trees in the western Amazon. Evolution 59: 1464-1478. Fine PVA, Miller ZJ, Mesones I, Irazuzta S, Appel HM, Stevens MHH, Sääksjärvi I, Schultz JC, Coley PD. 2006. The growth-defense trade-off and habitat specialization by plants in Amazonian forests. Ecology 87: S150-S162. Garzón‐Orduña IJ, Benetti‐Longhini JE, Brower AV. 2014. Timing the diversification of the Amazonian biota: butterfly divergences are consistent with Pleistocene refugia. Journal of Biogeography 41: 1631-1638. Garzón‐Orduña IJ, Benetti‐Longhini JE, Brower AV. 2015. Competing paradigms of Amazonian diversification and the Pleistocene refugium hypothesis. Journal of Biogeography 42: 1357-1360. Haffer J. 1969. Speciation in Amazonian forest birds. Science 165: 131-137. Haffer Jr. 1997. Alternative models of vertebrate speciation in Amazonia: an overview. Biodiversity and Conservation 6: 451-476. Hayes FE, Sewlal JAN. 2004. The Amazon River as a dispersal barrier to passerine birds: effects of river width, habitat and taxonomy. Journal of Biogeography 31: 1809-1818. Hill M.O. 1973. Diversity and evenness: A unifying notation and its consequences. Ecology 54: 427-432. Hoorn C. 2006. The birth of the mighty Amazon. Scientific American 294: 52-59. Hoorn C, Wesselingh F, Ter Steege H, Bermudez M, Mora A, Sevink J, Sanmartín I, SanchezMeseguer A, Anderson C, Figueiredo J. 2010. Amazonia through time: Andean uplift, climate change, landscape evolution, and biodiversity. Science 330: 927-931. Hoorn C, Wesselingh F. 2011. Amazonia, Landscape and Species Evolution: a Look into the Past. Wiley-Blackwell, Chichester. Hoskin CJ, Higgie M, McDonald KR, Moritz C. 2005. Reinforcement drives rapid allopatric speciation. Nature 437: 1353-1356. IUCN. 2016. Centres of Species Richness, Available from www.iucnredlist.org/initiatives/freshwater/centralafrica/geographicpatternsca (accessed January 2017). Kay KM, Reeves PA, Olmstead RG, Schemske DW. 2005. Rapid speciation and the evolution of hummingbird pollination in neotropical Costus subgenus Costus (Costaceae): evidence from nrDNA ITS and ETS sequences. American Journal of Botany 92: 1899-1910. Kricher J. 2011. Tropical Ecology. Princeton University Press, Princeton. Krosch MN, Baker AM, Mather PB, Cranston PS. 2011. Systematics and biogeography of the Gondwanan Orthocladiinae (Diptera: Chironomidae). Molecular Phylogenetics and Evolution 59: 458-468.

36

Latrubesse EM, Cozzuol M, da Silva-Caminha SA, Rigsby CA, Absy ML, Jaramillo C. 2010. The Late Miocene paleogeography of the Amazon Basin and the evolution of the Amazon River system. Earth-Science Reviews 99: 99-124. Latrubesse EM, Franzinelli E. 2005. The late Quaternary evolution of the Negro River, Amazon, Brazil: implications for island and floodplain formation in large anabranching tropical systems. Geomorphology 70: 372-397. Lynch Alfaro JW, Boubli JP, Olson LE, Di Fiore A, Wilson B, Gutiérrez‐Espeleta GA, Chiou KL, Schulte M, Neitzel S, Ross V. 2012. Explosive Pleistocene range expansion leads to widespread Amazonian sympatry between robust and gracile capuchin monkeys. Journal of Biogeography 39: 272-288. MacArthur RH, Wilson EO. 1967. The Theory of Island Biogeography. Princeton University Press, Princeton. Macedo M, Castello L. 2015. State of the Amazon: Freshwater Connectivity and Ecosystem Health. WWF Living Amazon Initiative. Brasilia. Marshall LG. 1988. Land mammals and the great American interchange. American Scientist 76: 380-388. Marshall L, Webb D, Sepkoski Jr J, Raup D. 1982. Mammalian evolution and the great American interchange. Science 215: 1351-1357. Martin PS. 2005. Twilight of the Mammoths: Ice Age Extinctions and the Rewilding of America. University of California Press, Berkeley. Mayr E. 1942. Systematics and the Origin of Species from the Viewpoint of a Zoologist. Columbia University Press, New York. Mayr E. 1963. Animal Species and Evolution. Belknap Press, Cambridge. McGuire JA, Witt CC, Remsen J, Corl A, Rabosky DL, Altshuler DL, Dudley R. 2014. Molecular phylogenetics and the diversification of hummingbirds. Current Biology 24: 910-916. Medeiros M, Barros R, Pieczarka J, Nagamachi C, Ponsa M, Garcia M, Garcia F, Egozcue J. 1997. Radiation and speciation of spider monkeys, genus Ateles, from the cytogenetic viewpoint. American Journal of Primatology 42: 167-178. Montgomery DR, Balco G, Willett SD. 2001. Climate, tectonics, and the morphology of the Andes. Geology 29: 579-582. Murphy JB, Nance RD, Cawood PA. 2009. Contrasting modes of supercontinent formation and the conundrum of Pangaea. Gondwana Research 15: 408-420. Nilsson MA, Arnason U, Spencer PB, Janke A. 2004. Marsupial relationships and a timeline for marsupial radiation in South Gondwana. Gene 340: 189-196. Pascual R, Archer M, Jaureguizar EO, Prado JL, Godthelp H, Hand SJ. 1992. First discovery of monotremes in South America. Nature 356: 704-706. Pereira SL, Baker AJ, Fleischer R. 2004. Vicariant speciation of curassows (Aves, Cracidae): a hypothesis based on mitochondrial DNA phylogeny. The Auk 121: 682-694. Peterson AT, Ammann CM. 2013. Global patterns of connectivity and isolation of populations of forest bird species in the late Pleistocene. Global Ecology and Biogeography 22: 596606. Pielou E.C. 1977. Mathematical Ecology. John Wiley & Sons, New York. Pinto-Sánchez NR, Ibáñez R, Madriñán S, Sanjur OI, Bermingham E, Crawford AJ. 2012. The great American biotic interchange in frogs: multiple and early colonization of Central America by the South American genus Pristimantis (Anura: Craugastoridae). Molecular Phylogenetics and Evolution 62: 954-972. Poelchau MF, Hamrick J. 2013. Palaeodistribution modelling does not support disjunct Pleistocene refugia in several Central American plant taxa. Journal of Biogeography 40: 662-675.

37

Ribas CC, Aleixo A, Nogueira AC, Miyaki CY, Cracraft J. 2011. A palaeobiogeographic model for biotic diversification within Amazonia over the past three million years. Proceedings of the Royal Society of London B: Biological Sciences 279: 681-689. Ricklefs RE. 2004. A comprehensive framework for global patterns in biodiversity. Ecology Letters 7: 1-15. Rosenzweig ML. 1995. Species Diversity in Space and Time. Cambridge University Press, Cambridge. Rull V. 2006. Quaternary speciation in the Neotropics. Molecular Ecology 15: 4257-4259. Rull V. 2008. Speciation timing and neotropical biodiversity: the Tertiary-Quaternary debate in the light of molecular phylogenetic evidence. Molecular Ecology 17: 2722-2729. Rull V. 2011. Neotropical biodiversity: timing and potential drivers. Trends in Ecology and Evolution 26: 508-513. Schellart W. 2008. Overriding plate shortening and extension above subduction zones: A parametric study to explain formation of the Andes Mountains. Geological Society of America Bulletin 120: 1441-1454. Schluter D. 2000. The Ecology of Adaptive Radiation. Oxford University Press, Oxford. Serrano-Rojas SJ, Whitworth A, Villacampa J, Von May R, Padial JM, Chaparro JC. 2017. A new species of poison-dart frog (Anura: Dendrobatidae) from Manu province, Amazon region of southeastern Peru, with notes on its natural history, bioacoustics, phylogenetics, and recommended conservation status. Zootaxa 4221:71-94. Smith ET. 2010. Early Cretaceous chelids from Lightning Ridge, New South Wales. Alcheringa 34: 375-384. Smith BT, Klicka J. 2010. The profound influence of the Late Pliocene Panamanian uplift on the exchange, diversification, and distribution of New World birds. Ecography 33: 333-342. Smith BT, McCormack JE, Cuervo AM, Hickerson MJ, Aleixo A, Cadena CD, Pérez-Emán J, Burney CW, Xie X, Harvey MG. 2014. The drivers of tropical speciation. Nature 515: 406-409. Stehli FG, Webb SD 1985. The Great American Biotic Interchange. Plenum Press, New York. Stuart AJ. 2015. Late Quaternary megafaunal extinctions on the continents: a short review. Geological Journal 50: 338-363. Upchurch P. 2008. Gondwanan break-up: legacies of a lost world? Trends in Ecology and Evolution 23: 229-236. Wallace AR. 1854. On the monkeys of the Amazon. Journal of Natural History 14: 451-454. Wallace AR. 1876. The geographical distribution of animals: with a study of the relations of living and extinct faunas as elucidating the past changes of the earth's surface, vol. 1. Macmillan, London. Webb SD. 2006. The Great American Biotic Interchange: Patterns and processes. Annals of the Missouri Botanical Garden 93: 245-257. Webb S. 2008. Megafauna demography and late Quaternary climatic change in Australia: A predisposition to extinction. Boreas 37: 329-345. Woodburne MO, Case JA. 1996. Dispersal, vicariance, and the Late Cretaceous to early Tertiary land mammal biogeography from South America to Australia. Journal of Mammalian Evolution 3: 121-161. Woodburne M. 2010. The Great American biotic interchange: Dispersals, tectonics, climate, sea level and holding pens. Journal of Mammalian Evolution 17: 245-265. Wright TF, Schirtzinger EE, Matsumoto T, Eberhard JR, Graves GR, Sanchez JJ, Capelli S, Müller H, Scharpegge J, Chambers GK. 2008. A multilocus molecular phylogeny of the parrots (Psittaciformes): support for a Gondwanan origin during the Cretaceous. Molecular Biology and Evolution 25: 2141-2156.

38

CHAPTER 4 – THREATS TO AMAZON WILDLIFE Tess Hanna, Ella-Monique Mason, Bruno O. Ferronato

Introduction The connectivity of freshwater and terrestrial ecosystems renders the Amazon highly susceptible to environmental threats (Castello et al. 2013). Aquatic and terrestrial health are dependent on regional hydrology which is being impacted by habitat modification, deforestation and climate change. The Amazon region is at high risk of extensive forest loss. Population growth and urban expansion, hunting, harvest and trade, mining, agricultural production, hydroelectric dams, roads, linear clearing and deforestation are all features of unsustainable human activity. These impacts strain, alter and manipulate Amazon’s biodiversity and ecosystem services which support life, promote aesthetics and help regulate atmospheric carbon (Cook et al. 2008; Laurence et al. 2009). Such humaninduced threats to Amazon wildlife are interconnected and have the potential to lead to population declines throughout the Basin. Human population expansion Tropical rainforests often occur in developing regions and nations with rapid population growth, intense natural resource exploitation, and pressure for economic development (Laurance et al. 2009). A dramatic increase in the human population across the Amazon Basin over the last few decades has had a large impact on natural resources within the area. The region has supported a human population for an estimated 13 000 years. However, recent population growth, expansion and development continues to threaten biodiversity and ecological systems. Although the population density is relatively low compared to other tropical rainforest regions (such as South-east Asia), high deforestation rates are still being observed. The overall

human population has increased from approximately 2.5 million in 1960 to about 28 million in 2016 (FAO 2015; Tritsch & Le Tourneau 2016), with more than 70% of inhabitants residing in urban areas (FAO 2011). This population growth happened mainly within the state capitals of the Amazon, such as Manaus, Belém, Cuiabá, Porto Velho, Boa Vista and Macapá. Brazil’s rural population represents 15 million of its total population of over 200 million (FAO 2017). During the past forty years, the natural environment has been intensively exploited. However, human population density is not considered to have been a major determinant of deforestation within the Amazon in the past 20 years (INPE 2008; Tritsch & Le Tourneau 2016). Most cleared land has been used for low-productivity pastures, which came about during the 1960-70s, when there were strong incentives from the military government for people to migrate to the region and convert forests to agricultural lands (Azevedo-Ramos 2007). Infrastructure projects and the Transamazon Highway (Fig. 4.1) attracted development and thereby increased population density and demand for products both inside and outside Brazil. Part of the government’s rationale was to enhance the quality of living of local communities. However, exploitation and conversion of regional forests have not significantly increased employment opportunities, income distribution or environmental benefits to local people (Azevedo-Ramos 2007). Approximately 45% of the Brazilian Amazon population currently live below the poverty line, and have an accentuated lack of social investments and infrastructure (Azevedo-Ramos 2007; FAO 2015).

39

Figure 4.1. The extent of the Transamazon Highway (route BR-230). After over 10 years of planning and construction, the Transamazon was inaugurated in 1972. Much of the highway remains unpaved and impassable during the rainy season. Total agricultural land use density is related to the percentage of croplands, natural pasturelands, and planted pasturelands in 2010, based on a pixel size of approximately 1 km2. Adapted from Walker (2011) and Dias et al. (2016).

Roads and linear clearing The Amazon is also susceptible to secondary effects associated with population growth, economic development and natural resource exploitation. These factors are driving the expansion of the road network into the core of the region, which has a number of impacts, including increased disturbance to local soils, hydrology and aquatic systems, chemical and nutrient pollution, wildlife road-related mortality and an increased level of illegal activities. Roads, power lines, gas lines, railroads and canals are all examples of anthropogenic constructions which significantly affect organisms and ecosystems, and their impacts are especially noticeable in the Amazon rainforest (Laurance et al. 2009). Roads encourage urbanisation and the spread of agriculture, particularly in remote areas where property rights are unclear or poorly

regulated (Soares-Filho et al. 2006; Hosaka et al. 2014). Paving roads further degrades and impacts the Amazon by causing expansion of existing roads, construction of side roads and tracks, and increased logging, illegal poaching and deforestation (Fearnside 2007). Of the approximately 26 3930 km of roads in the Brazilian Amazon, three-quarters were illegally constructed (RAISG, 2012). Illegal roads are usually local in scale, between 10 km and 200 km in length and built by local social actors to gain access to land, timber and other natural resources (Perz et al. 2007). Removal of vegetation during construction of mineral or seismic exploration utility tracks and corridors, roads and infrastructure are often heavily distributed and create a patchwork-like effect across a landscape (Carthew et al. 2013; de Oliveira et al. 2011). 40

Tropical forests are characterised by a humid, shaded and stable microclimate that supports ecologically specialised forest-dependent species. Due to their preference for such understorey microclimates, these species are highly susceptible to environmental change associated with roads and linear clearing. Many forest specialists will avoid areas of open canopy, including narrow clearings (< 30 m) and forest edges (Laurance et al. 2009). Therefore, construction of roads and linear clearings have a negative effect on forest specialists as it creates artificial barriers, preventing movement, increasing vehicle related mortality and opening the forest to exotic species. Agricultural and pastoral expansion Expansion of large-scale commercial agriculture is one of the main drivers of deforestation in the Amazon Basin (Martinelli et al. 2010). The growing demand for soybean production and cattle ranching has led to Brazil becoming one of the largest beef exporters and the second largest soybean producer worldwide (FAO 2017). In 2008, Brazil’s National Institute for Space Research (INPE) reported that 62% of deforested lands in the Brazilian Amazon were occupied by cattle pastures, and just under 5% was being utilized for crops such as soy (INPE 2013; de Almeida et al. 2016). Cattle and water buffalo are introduced species, whose activities result in changes to forest structure, flora composition and soil and light properties in floodplain areas (Sheikh & Uhl 2002). Soybean yields require large amounts of fertilizer. To satisfy international market demand, pastoralists and agriculturalists apply fertilisers, which offer short-term economic benefits. The excessive use of fertilizers aims to keep crop yields high, however, it often leads to environmental contamination and alteration of soil chemistry (Campo et al. 2009). Surface runoff with pesticides and fertilizers carries bacteria, nitrogen and phosphorus into watercourses and can result

in destructive algal blooms, removing oxygen from the ecosystem. The high aquatic wildlife fatality rates caused by anoxia can alter the structure of food webs (Martinelli et al. 2010; Castello et al. 2013). Similarly to mercury, pesticides can bioaccumulate and harm animals further up the food chain (Castello et al. 2013). Crops and livestock, such as soybean, cattle and buffalo alter floodplain hydrology, cause soil erosion, soil compaction and destruction of wetland vegetation and can be vectors of disease (Sheikh & Uhl 2002; Davidson et al. 2012). Clearing of land for commercial production creates a drier climate and distorts the movement of water vapour passing from the Atlantic Ocean to the Andes. From an agricultural perspective, a drier climate might be more favourable and profitable, however such a climate promotes and facilitates fire (Fernside 2007). Agricultural practices themselves also result in increased frequency of fire. Fires are used to clear regions of land and can drastically alter forest characteristics by reducing overall canopy cover, biomass and species richness. Following the abandonment of pastoral lands, rates of secondary forest regrowth were found to be negatively correlated with the number of fires during the pasture phase (Davidson et al. 2012). Frequent fires can lead to increases in plant species that are fire-adapted and flammable, thus resulting in more-savanna like ecosystems (Davidson et al. 2012). Mining Similarly to Australia, Brazil has a strong, globally-competitive mining industry that has exploited the abundance of mineral deposits found throughout the country. In 2010 there were over 7 900 mining companies operating in Brazil, and many of these mining sites are located in the Amazon (Fig. 4.2). Throughout the region there is a wide variety of naturally occurring minerals, including copper, iron, tin, manganese, nickel, bauxite and gold (IBRAM 2012). Mining land is cleared of vegetation 41

prior to being developed, with resultant loss of fauna and flora habitat (Castello et al. 2013). Whilst mines take up relatively small areas of land, they can have widespread and longlasting environmental impacts. Aside from land clearing, construction of mines and associated worker villages requires resources such as timber, hard metals and charcoal (Fearnside 2007; Laurance et al. 2009). Issues with land rights arise when mines are developed in areas inhabited by indigenous people, and conflict between miners and local communities can be devastating (Leguen 2017). Mineral exploration activities and construction attracts people to the area of development for employment opportunities, increasing population, expanding commercial and residential infrastructure and resulting in additional and often long-lasting environmental degradation. Gold mining is particularly problematic as high levels of mercury (a heavy metal harmful to living organisms), can be released directly into waterways through the extraction process (Junk & Piedade 2005). Lacerda and Pfeiffer

(1992) estimated that between 1 500 to 3 000 tons of mercury have contaminated the Amazon Basin from 1975 to 1990. This has severely impacted local fish populations in the vicinity of gold extraction sites, the fish having been found with high concentrations of mercury in their bodies. In theory at least, large mining corporations must adhere to strict environmental legislation. However, less well known is the impact of small scale or artisanal gold mining on the environment (Leguen 2017). Small-scale mines are a major source of mercury contamination in Amazon rivers. Mercury can enter the environment as vapour during the milling process or through soil erosion and leaching. Mercury percolates into aquatic systems where it is transformed into methylmercury which can bio-accumulate and bio-magnify in the food chain. Top predators may have four times higher mercury levels than noncarnivorous species and it can potentially damage human respiratory and neurological systems (Leguen 2017).

Figure 4.2. Completed and planned dams in 2016 according to their size (power output in megawatts), and mining activity in 2005 for the seven northern states of Brazil. Adapted from Lees et al. (2016) and RAISG (2012).

42

Although major environmental disasters still occur, such as the collapse of the Mariana mine dam in the state of Minas Gerais in 2016, the Brazilian Government’s policies and approaches to mining activities have continued to become more environmentally responsible and strict in recent years. Mining companies are forced to adhere to strict environmental legislation, and therefore have strong environmental rejuvenation programs (ICMM 2013). In many regions across the Amazon, this goes hand in hand with strong community involvement programs. Hydroelectric dams Currently, 191 dams exist throughout the Amazon Basin, and the Governments of Amazonian countries are pushing to develop a further 246 dams (Fig. 4.2; Lees et al. 2016). Whilst being in the renewable energy sector, hydroelectric dams are not truly “green” providers of energy due to their significant environmental and social impacts. Although published data for emissions vary widely, hydroelectric dams are known to produce greenhouse gases, and these are substantial during the first 10 years of dam operation. Most of the emissions occur when water is released from the turbines, however, they are also produced through above-water decay of trees remaining in reservoirs (Fearnside 2015). Submerged decomposed organic matter is deprived of dissolved oxygen, producing methane, which accumulates in the lower water layer of a stratified reservoir. Turbines eventually rotate the stratified water (upper and lower layers), releasing methane downstream. Furthermore, mercury released from logged soils reacts with the anoxic lower waters and transforms into methylmercury, which bio-accumulates proving harmful for fauna and human populations (Kemenes et al. 2007; Castello et al. 2013). Tropical dams produce more greenhouse gases than dams in other climatic regions; studies show that tropical dams can emit between 1 300 to 3 000 g CO2e/kWh (e/kWh = energy in kilowatt-

hour), whilst boreal dams emit 160 to 250 g CO2e/kWh. They also emit more greenhouse gases than thermoelectric plants using natural gas, oil and coal, which produce 400–500, 790–900 and 900–1200 g CO2e/kWh, respectively (Steinhurst et al. 2012; Fearnside 2015). Dams also drastically alter the hydrological cycle and the seasonal pattern of flooding and drying (Chapter 2). Dams are strategically placed on stable channels, and replace sections of river that have lotic (fast-flowing) waters with more lentic (still) waters, whilst also creating reservoirs of water in forested areas (Fearnside 2014). The topography of the Amazon Basin is flat and extremely susceptible to flooding, thus riverside environments and forest are transformed into empty inundated marshes, with inundated stretches of dead trees. These environmental modifications can lead to loss of regional diversity, by impacting on range-restricted and endemic species that require turbulent rivers and rocky outcrops, and working in favour of generalist or invasive species. Dams also impact flood pulse variability by disrupting lateral pathways between river channels, floodplains and riparian zones. The inhibition of sediment and micro-organism movement can also impact aquatic communities and marine processes thousands of kilometres downstream (Lees et al. 2016). The release of thermally stratified waters alters river temperatures and, as a consequence, the structure of biotic communities downstream. These factors can cause destruction of spawning habitats and block long-distance fish migrations, reduce biological production, impact on biogeochemical cycles and result in significant biodiversity loss through the reorganisation of flora and fauna assemblages (including both fish populations and their apex predators; Castello et al. 2013). Terrestrial ecosystems are also subject to change following the construction of dams, when flooding by hydroelectric reservoirs 43

splits regions of forest into smaller islands. This can result in major losses of forest carbon storage due to tree mortality and changes in ecosystem structure (Lees et al. 2016). Socially, the construction of power plants requires work forces, providing short-term

employment and encouraging migration of construction workers and their families to an area. However, on completion, little to no economic benefits are offered in return (Fernside 2016; Winemiller et al. 2016).

Did you know? The Balbina Dam in the State of Amazonas, Brazil, began generating power in 1989. Unfortunately the electrical output is only small, generating 112.2 MWe of electricity annually (less than 30% of the electricity required for Manaus), while flooding over 2 360 km2 of rainforest. The reservoir is shallow, with 1 500 islands and stagnant bays that became full of decomposing vegetation that turned the water acidic and anoxic (Fearnside 1989). The dam resulted in loss of potential use of forested areas and major losses of carbon storage, and saw the displacement of approximately one third of the surviving members of the Waimiri-Atroari indigenous people (Lees et al. 2016). Whilst the dam was initially built to supply the city of Manaus with electricity, during the period of construction the city grew so much that by the time the dam was finished, alternative energy sources were already required (Fearnside 1989). Overharvest Hunting of wildlife and harvesting of natural resources has been practiced by the residents of the Amazon Basin for thousands of years. Whilst it is still necessary in most places to ensure the survival of the local inhabitants, it is considered by many to be one of the most significant drivers of biodiversity loss (Morcatty & Valsecchi 2015; Antunes et al. 2016). In remote rural areas where access to produce is limited, hunting and harvesting not only provide sources of protein, fat and micronutrients, but can enable residents to retain their social and cultural connections. Harvesting wildlife also provides opportunities to earn extra money, with bushmeat sold to urban markets (Morcatty & Valsecchi 2015). However, local inhabitants still retain their traditions and habits even when they are living in urban areas, which adds further pressure on bush meat and illegal trade. Overharvesting has threatened the existence of a range of Amazonian wildlife, including mammals, turtles and fish. Loss of fauna results in modification of energy flows, thus severely impacting on ecosystems (Castello et al. 2015; Antunes et al. 2016). For example, the

loss of an apex fish species (such as Arapaima) due to overfishing can alter food web structures, nutrient cycles and water quality, while the depletion of megaherbivores (such as manatees, Fig 4.3a) in floodplain regions has been associated with overabundance of macrophytes (Castello et al. 2013). Commercial hunting has been banned in Brazil under the Wildlife Protection Act since 1967 (Chapter 5). However, illegal wildlife trade continues in the Amazon. Large-bodied vertebrates have generally persisted in the dense, inaccessible upland forests of terra firme, whereas aquatic species have nearly been wiped out in rivers and floodplains (Antunes et al. 2016). One good example is the Amazonian skin and fur trade, which was at its height from the 1930s to 1960s. This trade had a greater effect on aquatic species, with many species suffering basin-wide population collapse. Conversely, terrestrial species did not suffer such collapse due to adequate refuges. Chelonians, for example, are one of the most hunted species of wildlife in the Amazon (Fig. 4.3b; Morcatty & Valsecchi 2015). The Giant South American Turtle (Podocnemis expansa) has been harvested to the point of becoming 44

locally extinct in many locations. This species is the largest neotropical freshwater turtle and it is of significant cultural and socio-economic value, as it has long been a staple for inhabitants of the Amazon Basin (Edwards 1847; Wallace 1889; Alves et al. 2012). Once the region became colonised, egg harvest grew into a commercial production system which resulted in over 214 million P. expansa eggs being collected from over 400 000 females and

a significant reduction in population numbers (Alves et al. 2012; Cantarelli et al. 2014). Due to P. expansa decline, other smaller podocnemidid species such as the YellowSpotted Amazon River Turtle (Podocnemis unifilis) and Six-tubercled Amazon River Turtle (Podocnemis sextuberculata) started to become targets, and they are now experiencing an increase in exploitation and consequent decline (Caputo et al. 2005; Alves et al. 2012).

Figure 4.3. Animals targeted by illegal poachers: (a) Juvenile Manatee (Trichechus inunguis) at INPA (National Institute for Amazon Research) in November 2016. Most manatees are taken to INPA as juveniles, because their mothers were killed by hunters. (b) Red-footed (Chelonoidis carbonaria) and Yellow-footed (Chelonoidis denticulata) tortoises being released at Rio Trombetas Biological Reserve after being seized from poachers, December 2016. Photos: E. Mason and T. Hanna. Deforestation From 1960 to 2010 the Amazon forest cover was reduced to approximately 80% of its original area. Although deforestation rates have been gradually decreasing (Fig. 4.4), the impact on local and regional ecosystems remains high (Davidson et al. 2012). As outlined above, forests within the Amazon are cleared for a variety of reasons. Although there is a lack of basin-wide data, it is believed that selective logging has already severely impacted many habitats, and illegal logging occurs commonly (Castello et al. 2013). Forest fragmentation and edge effects resulting from logging increase tree mortality, forest

vulnerability to wildfire and carbon emissions (Broadbent et al. 2008). Deforestation has a severe impact on fauna, with reports showing that from 1963 to 2004 the number of endangered animal species in Brazil increased from fewer than 100 to more than 600 (Martinelli et al. 2010). Deforestation opens forest environments to exploration from developers, leading to increased hunting and resource extraction. It also affects species composition, wildlife migrations and ecological interactions (Soares-Filho et al. 2006; Laurance et al. 2009; Hosaka et al. 2014). Deforestation also results in the loss of important ecosystem services, a prime example 45

being the ability of the rainforest to drive the production of rainfall. Forest evapotranspiration has a strong effect on atmospheric dynamics, and plants also produce isoprene, a gas that forms a compound that enhances the formation of cloud condensation nuclei. These mechanisms, which assist in the formation of water droplets and drive precipitation, are being severely impacted by the loss of natural ecosystems within the Amazon (Martinelli et al. 2010). Deforestation results in decreased evapotranspiration and thus increased levels of water runoff and stream discharge. As a consequence of a lack of vegetation cover, there is an increase in soil erosion and terrestrial sediment transportation, ultimately impacting on the biogeochemistry and geomorphology of freshwater systems locally and downstream (Castello et al. 2013). Due to the large amount of carbon stored in the forest biomass, deforestation of the Amazon is also considered to be one of the biggest contributors to atmospheric greenhouse gas emissions globally (Davidson et al. 2012; de Espindola et al. 2012). Climate change The Amazon rainforest plays a significant role in regulating regional and global climate systems, however it is facing major threats due to anthropogenic climate change (Soares-Filho et al. 2006, 2010). The basin acts as a carbon sink, storing over 100 billion tonnes of carbon in soils and biomass (Davidson et al. 2012; Fearnside 2012). At the same time, the Amazon is the second largest source of greenhouse gas emissions in the world, as large amounts of CO2 and methane are released into the atmosphere through deforestation, agriculture, mining and dams (Soares-Filho et al. 2010; Fearnside 2012). As greenhouse gas emissions impact the global climate system, Amazon wildlife will potentially suffer forced migration, extinction, changes in phenology and reduced growth rates, which are likely to lead to reduced levels of biodiversity (Eguiguren-Velepucha et al. 2016).

For example, climate change threatens nesting success and survival of Amazon turtles such as the Giant South American Turtle (Podocnemis expansa) and the Yellow-Spotted Amazon River Turtle (Podocnemis unifilis). Changes in seasonal flooding may affect the timing of sandbank exposure, leading to a decline in the quality of nesting areas and consequent high egg mortality rates (Eisemberg et al. 2016). However, changes in the flood pulse are not the only threat. The sex of many Amazon turtles is determined by the temperature in the nest. Rising temperatures may increase the possibility of entirely female populations, and high mortality as a result of extreme temperatures, effectively driving the species to extinction (Paitz et al. 2010). Another example is the Brazil Nut Tree (Bertholletia excelsa). Although future climate projections predict stable or increased habitat suitability due to its dependence on forest clearings for growth, this species is largely reliant on pollinators, such as large bees, and its seeds are dispersed by agoutis and acouchis (Daysprocta spp. and Myoprocta spp.), which are vulnerable to environmental change (Thomas et al. 2014). Climate models predict temperatures throughout lowland tropical regions worldwide will increase by 3-8 °C during the 21st century, along with increased rates of carbon dioxide (CO2) gas emissions (Eguiguren-Velepucha et al. 2016). Whilst increases in CO2 may lead to a “CO2 fertilization effect” (increased CO2 in the atmosphere increases the rate of photosynthesis), increased temperatures will counteract this with increased rates of evapotranspiration, resulting in plants increasing stomatal closure times and sacrificing productivity levels to lessen water losses. As a consequence of climate change, the dry season will become longer, the annual rainfall threshold required to support seasonal forest areas will increase, and together these factors will considerably intensify the risk of forest areas transforming into regions of savanna (Malhi et al. 2009). 46

Figure 4.4. Annual deforestation rates of the Amazon rainforest for each Brazilian state (y axis = deforested area in km2, x axis = year from 1988 to 2016). Orange area represents the Arc of Deforestation (Source: INPE 2016: http://www.obt.inpe.br/prodes/prodes_1988_2016n.htm).

Overview Increasing loss of tropical rainforests is of great concern. It is clear that anthropological processes are directly impacting Amazon wildlife. Human population expansion, habitat contamination and modification, deforestation and climate change jeopardise Amazon ecology and stress irreversible transformations. The gravity of threats to the Amazon calls for

immediate intervention, conservation and restoration (Tritsch & Le Tourneau 2016). Due to the interconnectedness of ecosystems within the Amazon and the global significance of the region, it is imperative that further conservation efforts be made to ensure that it is protected (Cook et al. 2008; Laurance 2007). Actions and regulations towards the conservation of the Brazilian Amazon wildlife will be discussed in the next chapter.

Northern Brazil-Northern Australia connection Developing the North! Recent interest in potential development has put the construction of dams in northern Australia in the spotlight. However, the detrimental impacts to the environment in tropical regions generally outweigh the economic value of hydroelectric power plants. Dams in northern Australia have the potential to have similar detrimental impacts to ecosystems as observed in northern Brazil. Careful site-selection and planning are imperative to minimize the extent and severity of damage that dams will cause to northern Australian environments.

47

Literature Cited Alves RRN, Vieira KS, Santana GG, Vieira WLS, Almeida WO, Souto WMS, Santana GG, Montenegro PFGP, Pezzuti JCB. 2012. A review on human attitudes towards reptiles in Brazil. Environmental Monitoring and Assessment 184: 6877-6901. Antunes AP, Fewster RM, Venticinque EM, Peres CA, Levi T, Rohe F, Shepard Jr GH. 2016. Empty forest or empty rivers? A century of commercial hunting in Amazonia. Science Advances 2: e1600936. Azevedo-Ramos C. 2007. Sustainable development and challenging deforestation in the Brazilian Amazon: the good, the bad and the ugly. Unasylva 59: 12-16. Broadbent E, Asner G, Keller M, Knapp D, Oliveira P, Silva J. 2008. Forest fragmentation and edge effects from deforestation and selective logging in the Brazilian Amazon. Biological Conservation 141: 1745-1757. Campo RJ, Araujo RS, Hungria M. 2009. Molybdenum-enriched soybean seeds enhance N accumulation, seed yield, and seed protein content in Brazil. Field Crops Research 110: 219-224. Cantarelli VH, Malvasio A, Verdade LM. 2014. Brazil’s Podocnemis expansa Conservation Program: Retrospective and Future Directions. Chelonian Conservation and Biology 13: 124-128. Caputo FP, Canestrelli D, Boitani L. 2005. Conserving the terecay (Podocnemis unifilis, Testudines: Pelomedusidae) through a community-based sustainable harvest of its eggs. Biological Conservation 126: 84-92. Carthew S, Jones K, Lawes M. 2013. Responses of small vertebrates to linear clearings in a South Australian Woodland. Ecological Research 28: 1003-1010. Castello L, McGrath DG, Hess LL, Coe MT, Lefebvre PA, Petry P, Macedo MN, Renó VF, Arantes CC. 2013. The vulnerability of Amazon freshwater ecosystems. Conservation Letters 6: 217-229. Castello L, Arantes CC, McGrath DG, Stewart DJ, Sousa FSD. 2015. Understanding fishinginduced extinctions in the Amazon. Aquatic Conservation: Marine and Freshwater Ecosystems 25: 587-598. Cook KH, Vizy EK. 2008. Effects of twenty-first-century climate change on the Amazon rain forest. Journal of Climate 21: 542-560. Davidson EA, et al. 2012. The Amazon basin in transition. Nature 481: 321-328. de Almeida CA, Coutinho AC, Esquerdo JCDM, Adami M, Venturieri AV, Diniz CG, Dessay N, Durieux L, Gomes AR. 2016. High spatial resolution land use and land cover mapping of the Brazilian Legal Amazon in 2008 using Landsat-5/TM and MODIS data. Acta Amazonica 46: 291-302. de Espindola GM, de Aguiar APD, Pebesma E, Câmara G, Fonseca L. 2012. Agricultural land use dynamics in the Brazilian Amazon based on remote sensing and census data. Applied Geography 32: 240-252. de Oliveira PRR, Alberts CC, Francisco MR. 2011. Impact of road clearings on the movements of three understory insectivorous bird species in the Brazilian Atlantic forest. Biotropica 43: 628-632. Dias LC, Pimenta FM, Santos AB, Costa MH, Ladle RJ. 2016. Patterns of land use, extensification, and intensification of Brazilian agriculture. Global Change Biology 22: 2887-2903. Edwards WE. 1847 A Voyage up the River Amazon, Including a Residence at Para. John Murray, London. Eguiguren-Velepucha PA, Chamba JAM, Mendoza NAA, Ojeda-Luna TL, Samaniego-Rojas NS, Furniss MJ, Howe C, Mendoza ZHA. 2016. Tropical ecosystems vulnerability to climate change in southern Ecuador. Tropical Conservation Science 9: 1940082916668007 Eisemberg CC, Machado Balestra RA, Famelli S, Pereira FF, Diniz Bernardes VC, Carl Vogt R. 2016. Vulnerability of Giant South American Turtle (Podocnemis expansa) nesting habitat to 48

climate-change-induced alterations to fluvial cycles. Tropical Conservation Science 9: 1940082916667139. FAO. 2011. The State of Forests in the Amazon Basin, Congo Basin and Southeast Asia. Food and Agriculture Organization of the United Nations, Rome. FAO. 2015. Amazon Basin. Food and Agriculture Organization of the United Nations, Rome. FAO. 2017. Brazil – Country indicators. Food and Agriculture Organization of the United Nations, Available from http://www.fao.org/faostat/en/#country/21 (accessed February 2017). Fearnside PM. 1989. Brazil's Balbina Dam: Environment versus the legacy of the Pharaohs in Amazonia. Environmental Management 13: 401-423. Fearnside PM. 2007. Brazil’s Cuiaba-Santarem (BR0163) highway: The environmental cost of paving a soybean corridor through the amazon. Environmental Management 39: 601614. Fearnside PM. 2012. Brazil's Amazon forest in mitigating global warming: unresolved controversies. Climate Policy 12: 70-81. Fearnside PM. 2014. Brazil’s Madeira River dams: A setback for environmental policy in Amazonian development. Water Alternatives 7: 256-269 Fearnside PM. 2015. Emissions from tropical hydropower and the IPCC. Environmental Science and Policy 50: 225-239. Fearnside PM. 2016. Environmental and social impacts of hydroelectric dams in Brazilian Amazonia: Implications for the aluminum industry. World Development 77: 48-65. Hosaka T, Niino M, Kon M, Ochi T, Yamada T, Fletcher CD, Okuda T. 2014. Impacts of smallscale clearings due to selective logging on dung beetle communities. Biotropica 46: 720731. IBRAM. 2012. The Strength of Brazilian Mining. Brazilian Mining Institute (IBRAM, Instituto Brasileiro de Mineração), Brasilia. ICMM. 2013. The mining sector in Brazil: building institutions for sustainable development International Concil on Mining and Metals, London. INPE. 2008. Monitoramento da floresta amazônica brasileira por satélite – Projeto PRODES. Instituto Nacional de Pesquisas Espaciais (National Institute for Space Research), São Paulo. INPE. 2013. Metodologia para o Calculo da taxa anual de desmatamento na Amazonia legal. In Coordenadoria Geral de Observaçao da Terra Programa Amazonia - Projeto PRODES. Junk WJ, Piedade MTF. 2004. Status of knowledge, ongoing research, and research needs in Amazonian wetlands. Wetlands Ecology and Management 12: 597-609. Kemenes A, Forsberg BR, Melack JM. 2007. Methane release below a tropical hydroelectric dam. Geophysical Research Letters 34: L12809. Lacerda LD, Pfeiffer WC. 1992. Mercury from gold mining in the Amazon environment: an overview. Química Nova 15: 155-160. Laurance W. 2007. Road to ruin. New Scientist 194: 25. Laurance WF, Goosem M, Laurance SG. 2009. Impacts of roads and linear clearings on tropical forests. Trends in Ecology and Evolution 24: 659-669. Lees AC, Peres CA, Fearnside PM, Schneider M, Zuanon JAS. 2016. Hydropower and the future of Amazonian biodiversity. Biodiversity and Conservation 25: 451-466. Leguen R. 2017. Amazon Mining, World Wide Fund for Nature, available from http://wwf.panda.org/what_we_do/where_we_work/amazon/amazon_threats/other_t hreats/amazon_mining/ (accessed February 2017). Malhi Y, Aragäo L, Galbraith D, Huntingford C, Fisher R, Zelazowski P, Sitch S, McSweeney C, Meir P, Schellnhuber H. 2009. Exploring the likelihood and mechanism of a climatechange-induced dieback of the Amazon rainforest. Proceedings of the National Academy of Sciences 106: 20610-20615. 49

Martinelli LA, Naylor R, Vitousek PM, Moutinho P. 2010. Agriculture in Brazil: impacts, costs, and opportunities for a sustainable future. Current Opinion in Environmental Sustainability 2: 431-438. Morcatty TQ, Valsecchi J. 2015. Social, biological, and environmental drivers of the hunting and trade of the endangered yellow-footed tortoise in the Amazon. Ecology and Society 20: 3. Paitz RT, Clairardin SD, Griffin AM, Holgersson MCN, Bowden RM. 2010. Temperature fluctuations affect offspring sex but not morphological, behavioral, or immunological traits in the Northern Painted Turtle (Chrysemys picta). Canadian Journal of Zoology 88: 479-486. Perz SG, Caldas MM, Arima, Walker RJ. 2007. Unofficial road building in the Amazon: socioeconomic and biophysical explanations. Development and Change 38: 529-551. RAISG. 2012. Rede Amazônica De Informação Socioambiental Georreferenciada, available from www. raisg.socioambiental.org (accessed February 2017). Sheikh P, Uhl C. 2002. The impacts of water buffalo and cattle ranching on the lower Amazon Floodplain: An Ecological and Socio-economic Comparison. PhD Thesis, Department of Ecology, Pennsylvania State University. Soares-Filho B, et al. 2010. Role of Brazilian Amazon protected areas in climate change mitigation. Proc Natl Acad Sci U S A 107: 10821-10826. Soares-Filho BS, Nepstad DC, Curran LM, Cerqueira GC, Garcia RA, Ramos CA, Voll E, McDonald A, Lefebvre P, Schlesinger P. 2006. Modelling conservation in the Amazon basin. Nature 440: 520-523. Steinhurst W, Knight P, Schultz M. 2012. Hydropower Greenhouse Gas Emissions: State of the Research. Synapse Energy Economics, Cambridge. Thomas E, Caicedo CA, Loo J, Kindt R. 2014. The distribution of the Brazil nut (Bertholletia excelsa) through time: from range contraction in glacial refugia, over human-mediated expansion, to anthropogenic climate change. Boletim do Muesu Paraense Emilio Goeldi. Ciencias Naturais 9: 267-291. Tritsch I, Le Tourneau F-M. 2016. Population densities and deforestation in the Brazilian Amazon: New insights on the current human settlement patterns. Applied Geography 76: 163-172. Wallace AR. 1889. Travels on the Amazon and Rio Negro. Ward Lock, London. Walker R, Perz S, Arima E, Simmons C. 2011. The transamazon highway: past, present, future. Pages 569-599 in Brunn SD, editor. Engineering Earth. Springer, Netherlands. Winemiller KO, McIntyre PB, et al. 2016. Balancing hydropower and biodiversity in the Amazon, Congo and Mekong. Development and Environment 351: 128-129.

Rhinemys rufipes hatchling Illustration: Fernando A. Perini 50

CHAPTER 5 – WILDLIFE CONSERVATION AND MANAGEMENT IN THE AMAZON Jaime Marr, Ashley E. Owen, Emma Barrett, Shirley Famelli Introduction The Amazon has increasingly been seen as an area full of unrealised economic opportunities and potential for large-scale development, which has caused a conflicting situation – the desire to modernize and improve the Amazon’s economic standing, while also conserving biodiversity for the future (Lima et al. 2016). Past major projects should provide lessons for today’s policy makers, although even with advances in technology and scientific research, processes threatening wildlife (Chapter 4) are still occurring faster than environmental policies are being created (Fearnside 2016). Therefore, there is a need for an integrated, multi-disciplinary approach to guide wildlife conservation management and provide decision-makers with information on the effects of development on biodiversity. The focus of this chapter is to provide a summary of wildlife conservation management efforts within the Amazon, by including the science behind management and community engagement programs. Conservation efforts in Brazil are not new, with the country hosting a growing network of National Parks and Reserves called Protected Areas (PAs), which were created in efforts to protect biodiversity and control the exploitation of wildlife (Rylands & Brandon 2005). Management programs involving indigenous and other community stakeholders are increasing in number, with scientific and traditional knowledge of wildlife complementing each other (Castello et al. 2009; Marioni et al. 2013). However, scientific research is necessary in order to provide data on the current and potential future status of threatened wildlife,

and to prioritise management efforts (Castello et al. 2011). Evolution of environmental and conservation laws The trajectory of Brazilian Forest Policy effectively started in the 1900s (Figure 5.1). The earliest attempts to establish national parks in Brazil, made in the 1800s, were unsuccessful. Environmental and conservation laws were first introduced by the Brazilian Government in the 1930s with the aim of conserving the Amazon and its threatened species, and have evolved over the following decades. The 1934 Forest Code provided National Parks with legal status. The first National Park was created in 1937 on a remnant of Atlantic Rainforest in Southeast Brazil (Itatiaia National Park; Rylands & Brandon 2005). The first National Park in Amazon, Araguaia National Park (Tocantins state), was created in 1959. It is on the edge of the Amazon at the junction of the Amazon and Cerrado biomes (Ayres et al. 2005). Brazil’s first attempt to create a threatened species list was in 1964. This list was eventually published in 1973 and contained 86 taxa. In 1989 the list was revised in a joint effort between the Brazilian Institute for the Environment and Renewable Natural Resources (IBAMA) and the Brazilian Society of Zoology. In 2002, the IUCN (2001) criteria and categories were used to reassess species; at that time 239 aquatic and 395 terrestrial species were listed. The current legislation concerning land management and use for private properties in Brazil is based on the 1965 Forest Code, which faced the challenge of protecting native vegetation, 53% of which

51

was located on private land at the time (Soares-Filho 2014). The Forest Code has continued to be updated and revised since 1965, creating a vast network of protected natural parks and reserves, many of which have strict wildlife conservation objectives (Rylands & Brandon 2005; SoaresFilho et al. 2014). Laws concerning wildlife were introduced in 1967, with the prohibition of commercial wildlife exploitation (i.e. hunting and poaching) in an effort to slow wildlife declines and allow species to recover (Marioni et al. 2013). Since 2001, the Code requires privately owned properties to maintain an area of forest as legal reserve, as well as preserving permanent protected areas, including areas of riparian vegetation (as protection from flooding) and steep slopes (to prevent landslides; Fearnside 2016). The past 30 years in Brazil has seen the most

progress in terms of establishing organisations and laws aimed at protecting the environment. IBAMA was established in 1989 under the Ministry of Environment to manage areas classed as "public forests", and at the same time to create a presence in areas of forest that are otherwise vulnerable to deforestation (Fearnside, 2016). In 2000, the National System for Protected Areas was made law, creating a unified system for state, federal and Municipal parks (Rylands & Brandon 2005). In 2007, the IBAMA sectors responsible for the management of protected areas were separated from the agency and the ICMBio (Chico Mendes Institute for Biodiversity Conservation) was created. However, the Brazilian Ministry of Environment is not responsible for legislation in practical terms, as highways, dams and other major public works are organised under the Ministry of Mines and Energy through Brazil’s major electric utilities company Electrobras (Fearnside 2016).

Figure 5.1. Timeline summarising key events and institutions in the evolution of Brazilian forest policy up until 2006. Adapted from Banerjee et al. (2009). 52

Protected Areas Protected Areas in Brazil are defined as “all public areas under land use restrictions that contribute to protecting native ecosystems, even if they were created for purposes other than environmental conservation” (SoaresFilho et al. 2010). The current network of Protected Areas is regulated and classified by the National Protected Areas System (SNUC). Created in 2000, the SNUC stems from the 1965 Forest Code and primarily aims to provide legal protection to some of the most species-rich biomes (Rylands & Brandon 2005; Ferreira et al. 2014). The current measures taken by governments to manage PAs are related to international obligations to protect at least 10% of each ecosystem (Phillips 2002; Banerjee et al. 2009; Fearnside 2016). Protected Areas offer varying levels of protection (Table 5.1), which can influence the nature of effects on wildlife and habitat (Naughton-Treves et al. 2003). Currently there are three main categories under the Protected Areas system that offers sustainable use of resources or high level of protection: Strictly Protected Areas, Sustainable Use Areas and Indigenous Land Use Areas. Sustainable Use Areas and Indigenous Land Use Areas were included in the Forest Code to offer environmental protection and to secure the

rights and livelihoods of traditional communities (Rylands & Brandon 2005; Schwartzman & Zimmerman 2005). The most recent data from the World Bank (2013) showed the protected areas system covers over 5 million km² of the legal Amazon region in Brazil (Fig. 5.2). Strictly Protected Areas were established to reduce natural resources exploitation. They allow access for education, research and recreation purposes, but no extractive activities such as hunting or logging are permitted (de Marques et al. 2016). In contrast, Sustainable Use Areas are more flexible and allow the sustainable use of natural resources, controlled extraction of natural materials and land use changes (Nolte et al. 2013). Strictly Protected Areas and Sustainable Use Areas are managed by state and federal governments, with states preferring Sustainable Use Areas, and federal swaying toward more strict conservation areas (Rylands & Brandon 2005). Indigenous lands can overlap Sustainable Use Areas and Strictly Protected Areas. Indigenous Protected Areas have restrictions on the level of resource use that can take place, and are, in theory, managed through joint planning between the communities and government (Nolte 2013). However, the land is still owned by the federal government along with mineral and water rights.

Table 5.1. Summary of the number and categories of Protected Areas of Brazil (Rylands & Brandon 2005; UNEP-WCMC 2017; IUCN 2017). Classification

Land use and restrictions

Strictly Protected Areas (532)

In some cases prevents physical access and discourages consumptive use of resources

Sustainable-Use Conservation Areas (531)

Allows for controlled extraction, human settlements and land use change

Indigenous Land Use Areas (727)

Protection of indigenous rights and livelihoods

Categories Biological Reserve Ecological Station National Park Natural Monument Wildlife Refuge Forest Reserve Sustainable Development Reserve Extractive Reserve Environmental Protection Area Area of Relevant Ecological Interest Indigenous Reserve Indigenous Area

Number 54 87 325 36 30 97 35 90 262 47 19 708

53

Figure 5.2. Brazilian Amazon Protected Areas system which includes Indigenous Land Use, Strictly Protected and Sustainable Use Areas. Polygons compilation produced by the Environmental Ministry (Ministério do Meio Ambiente, 2015).

The effectiveness of Protected Areas is influenced by government resources along with the desire to enforce the laws, the level of pressure for development within that area and access to the land and its resources (Schwartzman & Zimmerman 2005; Nolte 2013). One of the roles of Protected Areas is to reduce carbon emissions due to deforestation and forest degradation, which is the second largest source of carbon emissions behind burning fossil fuels (Soares-Filho et al. 2010). The amount of carbon stored in tropical forests classified as Protected Areas in the Brazilian Amazon, however, is not sufficient to have an impact on global climate. To achieve a significant impact, it is necessary to protect more than 70% of the forest cover within the Amazon (Soares-Filho 2006). Governance As deforestation is a global issue, the United Nations produced the Bali Action Plan in 2007 to provide incentives to reduce emissions from deforestation and forest degradation in developing countries (REDD+: United Nations Programme on Reducing Emissions

from Deforestation and Forest Degradation). In response, Brazil established its own ambitious target at the Copenhagen Climate Conference in 2009; to reduce Amazon deforestation by 80% by 2020 (Scarano et al. 2012). Although Protected Areas have increased over time, managing and enforcing the restrictions applied to these areas has not been a strong point for the Brazilian government (Soares-Filho et al. 2014). The government has historically provided minimal enforcement of environmental laws, because of the lack of adequate staff and resources to manage the parks – most of which cover large areas (Maretti 2003). Despite management issues, all types of Protected Areas were found to be effective at reducing deforestation. Strictly Protected Areas have been shown to have the greatest success in avoiding deforestation, both before and after increased enforcement efforts by the Brazilian government. This suggests that even without strict law enforcement, there is an effective level of compliance to Strictly Protected Areas (Nolte et al. 2013). However, the effectiveness of management of Protected Areas continues to be challenged in the face of illegal activities 54

such as logging and wildlife poaching. Even Biological Reserves, which are given the strictest protection to conserve wildlife, are not immune to these activities; Protected Areas often have inadequate staff or resources to enforce regulations or hold a deterring presence (Maretti 2003). Effective governance cannot be attained with limited infrastructure, financial support and guidance to local management strategies (Banerjee et al. 2009). A number of policies, programs and plans have been introduced to reduce deforestation across Amazonia. For example, the ARPA (Amazon Region Protected Areas) Program was launched by the Brazilian government in 2002 with the aim of supporting 600 000 km2 of new and existing Protected Areas (Soares-Filho 2010). The government established a working group to tackle deforestation which aims to reduce land clearing, administer land tenure, improve environmental monitoring and support sustainable forest developments. Federal and military police field enforcement stations were set up in hot spots, ultimately lowering the rate of deforestation in the surrounding areas (Banerjee et al. 2009). Such actions have achieved positive outcomes. Although gross forest loss is the second highest globally, it had a dramatic policy-driven reduction in deforestation (Hansen et al. 2013), and can be

considered a global exception in regard to percentage of decline of forest cover loss (Fig. 5.3). This was the case until 2013, when an increase in the global price for soy and beef has again driven rising levels of deforestation (Fonseca et al. 2014, 2015). Currently, there is pressure from the federal government to push its Growth Acceleration Program (PAC) across the region (de Marques et al. 2016). Like northern Australia, the Brazilian Government has plans to develop the north of the country with large road and dam infrastructure projects. Pressure from government comes in the form of development plans, some of which were put in place during Brazil’s dictatorship (1964-1985), such as Operação Amazônia (Operation Amazon), with the intention of developing infrastructure, including roads, that encourages forestry projects, mining and agri-business and other business in the region (Albert 1992). Other development plans have been set in place more recently. However, they are often put in place prior to the undertaking of economic viability and environmental impact studies, heavily impacting on conservation efforts. These projects include: “Brazil in Action” (1996-1999), “Forward Brazil” (20002003) “Pluriannual Plan” (2004-2007) and the “PAC” (2008-2011 and 2012-2015) as described above (Fearnside 2016).

Figure 5.3. Annual forest loss trend (tree cover loss – km2/year) from 2000 to 2012. Brazil exhibited the greatest decline in annual loss. However, increased annual loss of other regions more than offset the slowing of Brazilian deforestation. Data from Hansen et al. (2013).

55

Science applied to conservation Science plays an important role in providing information on the effectiveness of management approaches, and can provide indications of expected outcomes to inform researchers, communities and policy makers (Lima 2016). Effective future conservation requires current and ongoing modelling and monitoring of population dynamics, paying closer attention to endemic and threatened species. The application of science in the conservation of wildlife in the Amazon is important in identifying areas of concern and management priorities. Models can predict the effect of threatening factors on populations (Soares-Filho et al. 2006). On a broader scale, models have been used to predict the effects of large-scale threats. For example, a quarter of the 382 mammal species examined by Soares-Filho et al. (2006) are predicted to lose 40% of their Amazon forest habitats if agricultural and clearing practices continue. Data from population dynamics modelling are particularly useful for community-based programs in Sustainable-use Conservation Areas. An example is the study of population dynamics in the conservation of the Arapaima Fish (Arapaima sp.), which is threatened by overfishing (Castello et al. 2011). The Arapaima Fish is one of the world's largest freshwater fish (reaching 3 m in length). This study modelled potential populations by incorporating the size and timing of harvests and the size of the fish caught. The results assisted management by showing the importance of complying with catch minimum size, closed seasons and catch limits. With the Amazon facing increased pressure to develop from the southern Brazilian states and government, there is a need for developmentrelated science to be readily available for decision-makers. Numerous multi-disciplinary studies have been done on long-term effects

of developments, such as hydroelectric dams and increased agriculture and infrastructure, and concluded that there are often negative impacts on human health, socio-economics, hydrology and the environment (Lima et al. 2016). Unfortunately, development has often gone ahead despite the findings of scientific research. Community-based conservation programs Amazon traditional and local communities (Table 5.2) have a history of diverse interactions with wildlife, making them important stakeholders in wildlife conservation efforts. Traditional cultures have historically used many animals for food, medicine and religious uses, yet some are now threatened and face further threats from human activities such as overhunting and increased hunting efficiency due to modernized technology (e.g. guns, more efficient fishing gear; Alves et al. 2012). Despite their threatened status, some endangered Amazon species continue to be harvested unsustainably by local communities (Caputo et al. 2005; Alves et al. 2012). The outcomes of scientific research can be used in community-based education and conservation management programs by identifying threatened species and providing baseline data (Caputo et al. 2005; Shepard et al. 2012). Scientific research and traditional knowledge can complement each other in conservation efforts, with science providing quantitative data, while local knowledge provides empirical data (Castello et al. 2009). It is important to acknowledge local and traditional knowledge, as it is the local communities who are in most frequent contact with wildlife and therefore have one of the strongest influences in conservation efforts. Local knowledge has proven to be invaluable in some studies, by providing reliable and efficient counts and recordings of wildlife catches (Caputo et al. 2005; Castello et al. 2009; Shepard et al. 2012), 56

as well as providing information on wildlife behaviour and ecology based on experience (Marioni et al. 2013). McGrath and Castello (2015) have explored varying ways of integrating scientific knowledge with local

ecological knowledge, specifically with local fisherman in the management of the Arapaima fisheries throughout the Amazon Basin. They suggest that fisherman have a detailed knowledge of species and their interactions.

Table 5.2. Amazon traditional groups according to Diegues and Arruda (2001). The indigenous groups have retained their languages and have a socio-cultural history prior to European colonization. Brazil has 300 000 indigenous people, which correspond to only 0.2% of the Brazilian population, but represent an extensive sociodiversity. Traditional non-indigenous groups were the result of mixing of indigenous and non-indigenous cultures. Non-indigenous groups speak Portuguese, although with several variants. Traditional Groups Indigenous – people who have maintained a historical and cultural continuity since before European colonisation.

Non-indigenous – groups with a strong indigenous influence, which is observed not only in regional terms, but also in the various technologies of food preparation, ceramics, techniques for the construction of hunting and fishing instruments, etc.

Subcategories

Locations

Livelihoods

206 indigenous groups.

Refer to Fig. 5.2

c.180 different languages and societies varying from traditional hunter-gathering to subsistence agriculture and fishing

Caboclo amazônico, Ribeirinhos – Amazon riverine people, usually a mix of indigenous and European background.

On flooded plains (varzéa and igapó) and near rivers.

Caboclo amazônico, Seringueiros and castanheiros – Amazon rubber tapper and Brazil nut collectors, usually a mix of indigenous and European background.

Near the rivers and flooded plains (including varzéa and igapó), but also in nonflooded forests (terra firme) and nearby creeks and streams.

Quilombolas – African slave descendants

Near the rivers and flooded plains (including varzéa and igapó), but also in nonflooded forests (terra firme) and nearby creeks and streams.

Praieiros – Amazon coastal people

Coastal area of the Amazon between Piauí and Amapá states.

In many cases, a lack of formal education becomes problematic when communication barriers arise in land management discussions between communities and controlling authorities or scientists. These communication issues can arise out of the complexity of

Traditional activities that rely mostly on fishing and subsistence agriculture. Traditional activities that rely on forest resources, mainly rubber and Brazil nut. Fishing and subsistence agriculture play a secondary but important role. Survive in communal enclaves, often former slave farms abandoned by landowners (Brazilian slavery period 1532 to 1850). Ensure subsistence with small-scale fishing, hunter-gathering and agriculture. Main activity is fishing, although in many places income is supplemented with small-scale farming and other activities.

working with many ethnic groups that hold their own traditional practices, beliefs and language (Castello et al. 2009). However, targeted approaches can overcome these challenges. The Instituto Socio Ambiental (ISA - Social Environment Institute), for 57

example, overcame obstacles in the A’ukre Kayapo village (part of the Xingu Indigenous Park within the Brazilian Amazon) by providing training to teachers for language education, skills for mapping resources, and studying alternative incomes strategies that are environmentally sustainable (Schwartzman & Zimmerman 2005). With development pressures looming, allowing communities to be key stakeholders, maintaining good structure within communities and maintaining authority over their resources (Schwartzman & Zimmerman 2005; Chacin 2010) will empower them to become effective managers. In such cases, scientific studies can help traditional practices, such as the consumption of turtle eggs, to continue sustainably by advising which eggs

should be taken for consumption from nest sites that will flood and be naturally unsuccessful, and by providing sustainable catch limits (Caputo et al. 2005; Castello et al. 2009). However, community-based conservation efforts (Fig. 5.4a) and education and awareness (Fig. 5.4b) can only go so far if the over-exploitation of wildlife provides more attractive economic benefits. Wildlife conservation efforts within communities need to provide competitive, if not greater, economic benefits than exploitative activities, otherwise they are likely to be unsuccessful. For example, there is still a great demand and high price paid for Giant Amazon Turtle meat (Alves et al. 2012), which often comes from illegally poached mature adult females, an act which threatens the species more seriously than harvesting eggs (Caputo et al. 2005).

Figure 5.4. Community-based conservation efforts to protect the Six-tubercled Amazon River Turtle (Podocnemis sextuberculata) at the Mamirauá Sustainable Development Reserve in 2003. Conservation efforts included (a) nesting beach protection by local rangers and (b) educational activities involving local schools, when students were invited to release the hatchlings grown at the protected nesting site. Photos: C. Eisemberg. For example, in the Rio Trombetas Biological Reserve, locals are allowed to harvest Brazil nuts inside the Reserve, yet the work is difficult and the payment per bag of nuts is low, making the poaching of Giant Amazon

Turtles attractive. On the other hand, conservation programs with economic incentives have shown promising results, such as the payment per hatched Giant Amazon Turtle egg in Ecuador. At the Arapaima 58

fishery at Mamirauá Sustainable Development Reserve, integrated management approaches were effective for two reasons: fishermen were made responsible for counting fish, and this aided in their acceptance of fishing quotas,

hence they respected the limits, while scientific studies showed that fishers’ counts of Arapaima were reliable, resulting in harvest permits being granted by IBAMA officials (Castello et al. 2009).

Did you know? The strictly protected area, the Rio Trombetas Biological Reserve, is a conservation location for giants. Four threatened ‘giant’ species are listed as occurring within the area; they are called giants as they are the largest species within their families (Alves et al. 2012; ICMBio 2017). These are the Giant Anteater (Myrmecophaga tridactyla), the Giant Amazon River Turtle, the Giant Armadillo (Priodontes maximus) and the Giant Brazilian Otter (Pteronura brasiliensis). The future of Amazon conservation Environmental threats are increasing faster than the establishment of environmental management strategies and institutions (Fearnside 2016). There is a need for longterm strategies and programs to balance growing environmental pressures. Thus a system can be implemented that promotes sustainable development and economic incentives. As it stands, Brazil is in the opening stages of the Payment for Environmental Services (PES) system, underfunded and currently focused on the southern states though there is greater potential in the Amazon region. Environmental services are currently ‘free’ to use and limited policy capacity and enforcement inevitably does little to prevent the loss of forests (Hall 2008; Fearnside 2016). Some initiatives already established have enabled rural and small communities to sell carbon credits through reforestation and conservation projects. Companies can purchase these credits as a trade-off for sustainable development. This incentive will provide long term monetary support to ‘forest guardians’ for their efforts in biodiversity maintenance and encourage sustainable development in the Amazon (Fearnside 1997). Forest guardians are key in wildlife conservation efforts and should be

empowered to be included in environmental policy decisions. Indigenous communities are identified as effective actors in balancing deforestation pressures (Nolte et al. 2013), yet they have little inclusion in policy intervention. Brazil attempted to codify traditional and indigenous people’s rights into legislation through the 1988 Convention. The federal government remains owner of indigenous lands, whilst indigenous peoples have ‘original rights’ on the land they traditionally occupy. In 2013, the Amazon region held the greatest concentration of indigenous populations in Brazil, with further unestimated populations of unrecognised indigenous groups. Indigenous people are marginalised in the political arena although large development projects heavily affect them. Local indigenous groups have expressed their concern at their lack of participation in planning of proposed development projects that affect their lives. In 2009 it was estimated that 426 projects were being undertaken on the 400 legally recognised indigenous lands (Becker 2013). Current Brazilian legislation does not cover subsistence hunting effectively, which adversely affects the development of community-based management initiatives for traditional peoples. According to the 1967 Wildlife Protection Act, animals are the property of the state and hunting is prohibited. 59

Although the Act states that hunting is not considered a crime in cases when the hunter relies on the resource to avoid starvation, governmental preparatory courses instruct environmental agents to investigate and fine any individual who is hunting (Pezzuti 2009).

To prevent future socio-ecological problems which stem from these developments, it is important to value the decisions of local people in order to protect their rights to land and identity.

Northern Brazil-Northern Australia connection In Brazil, the importance of community-based wildlife management was recognised in 1996 with the establishment of SDRs (Sustainable Development Reserves), which allowed the management of certain wildlife species for regulated, commercial use (Marioni et al. 2013). Since then, communities have been involved in several successful conservation management roles, such as park rangers and co-managers of commercial harvesting ventures of caiman and arapaima fish (Castello et al. 2009). This allows communities to be self-sustaining and has been proven to increase harvests and decrease poaching (Caputo et al. 2005). The Australian legal system also acknowledges land rights and the interests of indigenous people, allowing them to conduct activities such as hunting, as they have in the past, under traditional laws and customs. For example, indigenous people in the Northern Territory have the right to use their country in accordance with traditional hunting and food gathering practices (Wilson et al. 2010), allowing for indigenous enterprises, such as the Bawinanga Aboriginal Corporation in Arnhem Land. This Corporation sustainably harvests three animal species for commercial sale: Saltwater Crocodile (Crocodylus porosus), Northern Long-necked Turtle (Chelodina rugosa), and Tarantula spiders (Selenotholus sp.) (Fordham et al. 2010). Overview Wildlife conservation in the Amazon faces challenges due to exploitative human activities and lack of government funding, yet it is showing areas of progress and success. This chapter highlighted four main areas that are important for future wildlife management and conservation in the Amazon: the history of protected areas, science that can be applied to conservation in the Amazon, importance of indigenous and community-based engagement, and recommendations for the future of conservation. Scientific research can provide baseline data for decision-makers in regard to future development. The involvement of traditional and local stakeholders in wildlife management programs helps to empower local communities, increase wildlife viability and

decrease exploitative activities. The conservation issues facing wildlife in the Amazon, however, are continually changing, highlighting the need for continued scientific research, increased awareness at local, national and international levels and adaptive community-based conservation management. The way forward involves learning from the past (Fearnside 2016), implementing strategies backed by research and adding to the strategies that are already showing promise – many of which have begun on a small scale in individual communities. The Rio Trombetas Biological Reserve is a good example of slow progress over the years in effective wildlife management. As outlined in the following chapter, rangers are working in partnership with local volunteers at this Reserve to conserve Amazon River turtles.

60

Literature Cited Alves RRN, Vieira KS, Santana GG, Vieira WLS, Almeida WO, Souto WMS, Santana GG, Montenegro PFGP, Pezzuti JCB. 2012. A review on human attitudes towards reptiles in Brazil. Environmental Monitoring and Assessment 184: 6877-6901. Albert B. 1992. Indian lands, environmental policy and military geopolitics in the development of the Brazilian Amazon: the case of the Yanomami. Development and Change, 23: 35-70. Ayres JM, Fonseca GAB, Rylands AB, Queiroz HL, Pinto LP, Masterson D, Cavalcanti RB. 2005. Os Corredores Ecologicos das Florestas Tropicais do Brasil. Sociedade Civil Mamirauá, Tefé. Banerjee O, Macpherson AJ, Alavalapati J. 2009. Toward a policy of sustainable forest management in Brazil a historical analysis. The Journal of Environment and Development 18: 130-153. Becker PB. 2013. Indigenous land right in Brazil: A comparison between the letter of the law and its applications. Ciência & Trópico 34: 47-71 Caputo FP, Canestrelli D, Boitani L. 2005. Conserving the terecay (Podocnemis unifilis, Testudines: Pelomedusidae) through a community-based sustainable harvest of its eggs. Biological Conservation 126: 84-92. Castello L, Viana JP, Watkins G, Pinedo-Vasquez M, Luzadis VA. 2009. Lessons from integrating fishers of arapaima in small-scale fisheries management at the Mamirauá Reserve, Amazon. Environmental Management 43: 197-209. Castello L, Stewart DJ, Arantes CC. 2011. Modelling population dynamics and conservation of arapaima in the Amazon. Reviews in Fish Biology and Fisheries 21: 623-640. Chacín CL. 2010. Towards sustainable harvest of sideneck river turtles (Podocnemis spp.) in the middle Orinoco, Venezuela. Doctoral dissertation. Duke University, Durham. de Marques AA, Schneider M, Peres CA. 2016. Human population and socioeconomic modulators of conservation performance in 788 Amazonian and Atlantic Forest reserves. PeerJ 4: e2206 Diegues AC, Arruda RSV. 2001. Saberes tradicionais e biodiversidade no Brasil. Ministério do Meio Ambiente, Brasilia. Fearnside PM. 1997. Environmental services as a strategy for sustainable development in rural Amazonia. Ecological Economics 20: 53-70. Fearnside PM. 2016. Environmental policy in Brazilian Amazonia: lessons from recent history. Novos Cadernos NAEA 19: 27-46. Ferreira J, et al. 2014. Brazil's environmental leadership at risk. Science 346: 706-707. Fonseca A, Souza Jr C, Veríssimo A. 2014. Deforestation report for the Brazilian Amazon September 2014. Imazon, Belém.ff Fonseca A, Justino M, Souza Jr C, Veríssimo A. 2015. Deforestation report for the Brazilian Amazon -November 2015. Imazon, Belém. Fordham A, Fogarty W, Fordham D. 2010. The viability of wildlife enterprises in remote Indigenous communities of Australia: A case study. ANU, Centre for Aboriginal Economic Policy Research, Canberra. Hall A. 2008. Paying for environmental services: the case of Brazilian Amazonia. Journal of International Development 20: 965-81. Hansen MC, et al. 2013. High-resolution global maps of 21st-century forest cover change. Science 342: 850-853. IUCN. 2001. IUCN Red List Categories and Criteria: Version 3.1. IUCN Species Survival Commission - IUCN, Gland. IUCN. 2017. Protected Areas Categories. International Union for the Conservation of Nature (IUCN), Gland, Available from https://www.iucn.org/theme/protectedareas/about/protected-areas-categories (accessed February 2017). 61

Lima JM, et al. 2016. A social-ecological database to advance research on infrastructure development impacts in the Brazilian Amazon. Scientific Data 3: 1-9. Maretti CC. 2003. Protected Areas and Indigenous and Local Communities of Brazil International Union for the Conservation of Nature (IUCN), Gland, Available from http://cmsdata.iucn.org/downloads/cca_cmaretti.pdf (accessed February 2017). Marioni B, Botero-Arias R, Fonseca-Junior SF. 2013. Local community involvement as a basis for sustainable crocodilian management in protected areas of Central Amazonia: Problem or solution. Tropical Conservation Science 6: 484-92. McGrath D, Castello L. 2015. Integrating fishers’ ecological knowledge and the ecosystem based management of tropical inland fisheries: an Amazon case study. Pp. 127-148 in Fischer J, Jorgensen J, Josupeit H, Kalikoski DC, Lucas CM, editors. Fishers' knowledge and the ecosystem approach to fisheries: applications, experiences and lessons in Latin America. Food and Agriculture Organization of the United Nations, Rome. Ministério do Meio Ambiente. 2015. Brazilian Amazon protected areas system. Available from http://www.mma.gov.br/areas-protegidas/cadastro-nacional-de-ucs/dadosgeorreferenciados (accessed January 2015) Naughton‐Treves L, Mena JL, Treves A, Alvarez N, Radeloff VC. 2003. Wildlife survival beyond park boundaries: the impact of slash‐and‐burn agriculture and hunting on mammals in Tambopata, Peru. Conservation Biology 17: 1106-1117. Nolte C, Agrawal A, Silvius KM, Soares-Filho BS. 2013. Governance regime and location influence avoided deforestation success of protected areas in the Brazilian Amazon. Proceedings of the National Academy of Sciences 110: 4956-61. Pezzuti JCB. 2009. Manejo de caça e a conservação da fauna silvestre com participação comunitária. Universidade Federal do Pará, NAEA 235: 1-13. Phillips A. 2002. Management guidelines for IUCN Category V protected areas: Protected landscapes/ seascapes. International Union for the Conservation of Nature (IUCN), Gland. Rylands AB, Brandon K. 2005. Brazilian protected areas. Conservation Biology 19: 612-8. Scarano F, Guimarães A, da Silva JM. 2012. Rio+ 20: Lead by example. Nature 486: 25-26. Schwartzman S, Zimmerman B. 2005. Conservation alliances with indigenous peoples of the Amazon. Conservation Biology 19: 721-727. Shepard Jr GH, Levi T, Neves EG. Peres CA, Yu DW. 2012. Hunting in ancient and modern Amazonia: rethinking sustainability. American Anthropologist 114: 652-667. Soares-Filho, BS, et al. 2006. Modelling conservation in the Amazon basin. Nature 440: 520-523. Soares-Filho B, et al. 2010. Role of Brazilian Amazon protected areas in climate change mitigation. Proceedings of the National Academy of Sciences 107: 10821-10826. Soares-Filho B, et al. 2014. Cracking Brazil's forest code. Science 344: 363-364. UNEP-WCMC. 2017. Protected Area profile for Brazil from the World Database of Protected Areas. Protected Planet, Available from https://www.protectedplanet.net/country/BR (accessed February 2017). Wilson GR, Edwards MJ, Smits JK. 2010. Support for Indigenous wildlife management in Australia to enable sustainable use. Wildlife Research 37: 255-263. World Bank. 2013. Brazil Protects the Amazon, Available from http://www.worldbank.org/en/results/2013/10/09/Brazil-protects-Amazonincreasing-size-protected-areas (accessed February 2017).

62

CHAPTER 6 – MANAGEMENT OF RIVER TURTLES AT THE RIO TROMBETAS BIOLOGICAL RESERVE – A CASE STUDY OF WILDLIFE CONSERVATION IN AMAZON Carla Eisemberg, Diane Bowman, Richard Vogt, Marcello Silva, Carolina Moura, Virginia Bernardes, Sofia Leão

Introduction The Rio Trombetas Biological Reserve (ReBio-Trombetas) comprises an area of 385 000 ha. It is located on the north-east side of the Trombetas River, Pará, Brazil (Haller & Rodrigues 2006). The Trombetas River is considered a clear water river (Fig 6.1a), originating from the Guiana Shield, with a drainage area of 125 000 km2 (Penã 2002). This area has long been associated with turtle research, particularly regarding the three river species (Family Podocnemididae): the Giant South American Turtle (Podocnemis expansa), the Yellow-spotted River Turtle (Podocnemis unifilis) and the Six-tubercled Amazon River

Turtle (Podocnemis sextuberculata) (Table 6.1a). The Reserve incorporates nesting sandbanks and adjacent seasonally inundated forest habitats (Fig 6.1b). In this region nesting occurs from September to November, when water levels are low and nesting sandbanks are exposed and therefore available (Alho & Pádua 1982; Haller & Rodrigues 2006). The main nesting sandbanks for P. expansa and P. sextuberculata in the region are Leonardo, Farias, Jacaré and Uerana beaches (Fig. 6.2). These sandbanks are large, over a kilometre long and up to 500 m wide in the nesting season. Both P. unifilis and P. sextuberculata also nest in smaller sandbanks scattered over the Reserve.

Figure 6.1. Rio Trombetas Biological Reserve: (a) Tabuleiro Basecamp and (b) Trombetas River, exposed sandbanks and seasonally inundated forest habitats. Photos: C. Eisemberg and ICMBio. There are records of river turtles from the Trombetas River since the 1950s. In 1953, it was estimated that 100 000 P. expansa nested in the area (Alfinito et al. 1973). Protection of the Trombetas River turtle population, before the

Reserve was established, commenced in 1964. The protection actions were initially under the management of the Ministry of Agriculture, which created the Service for Turtle Protection in 1965. In 1967, the Brazilian Institute for 63

Forestry Development (IBDF - Instituto Brasileiro de Desenvolvimento Florestal) was created and together with the Fauna Protection Law (which outlawed wildlife harvest in Brazil), gave legal support for

protection efforts inside the Reserve. From 1976 onwards, the IBDF coordinated the government wildlife protection actions, which resulted in the creation of the Rio Trombetas Biological Reserve.

Table 6.1. Characteristics of the three main protected species at ReBio-Trombetas: the Giant South American Turtle (Podocnemis expansa), the Yellow-spotted River Turtle (Podocnemis unifilis) and Six-tubercled Amazon River Turtle (Podocnemis sextuberculata). Data from Alho and Pádua (1982), Ernst and Barbour (1989), Souza and Vogt (1994), Haller and Rodrigues (2006), Vogt (2008) Ferreira-Júnior and Castro (2010), Pignati et al. (2013), Arraes and Tavares-Dias (2014), Eisemberg et al. (2016), (2017). Photos: C. Eisemberg, J. Marr, S. Perkins. P. expansa

P. unifilis

P. sextuberculata

107

68

34

95.5

89

96

highest points of large sandbanks

generalist, smaller sandbanks

prefers large sandbanks, usually below P. expansa

Average incubation period (days)

45

68

57

Nesting behaviour

large groups 64 95

nests dispersed and usually single 18 29

nests dispersed and in low densities 21 15

October

September–October

September–November

Species Size (maximum carapace length - cm) Diet (volume of vegetable matter - %) Nesting sites

Average nest depth (cm) Average clutch size Nesting season (Trombetas River)

Hatchling

The creation of National Parks and Reserves within the Brazilian Amazon commenced in the 1970s (Chapter 5). The ReBio-Trombetas was created in 1979 (Fig 6.2) and it was the second Biological Reserve created in the Brazilian Amazon. One of the main objectives of the ReBio-Trombetas is the protection of the local freshwater turtle populations (IBAMA 1989). Once the Reserve was created, the management and protection of turtles within the Reserve became part of the Amazon Turtles Project (Projeto Quelônios da Amazônia, PQA), which was initially managed by the IBDF and later by the Brazilian Institute of Environment and Renewable

Natural Resources (Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis, IBAMA). The ReBio-Trombetas is currently under the management of the Chico Mendes Institute for Biodiversity Conservation (Instituto Chico Mendes de Conservação da Biodiversidade, ICMBio). ReBio-Trombetas wildlife conservation stakeholders The ICMBio and IBAMA are the two independent branches of the Ministry of Environment. (Ministério do Meio Ambiente) While IBAMA primarily provides environmental approvals, the ICMBio is in 64

charge of the management of federally protected areas and the promotion of biodiversity conservation through research, education, and appropriate wildlife management practices. Although the ReBioTrombetas is currently under the management of the ICMBio, it relies on partnerships with universities, research institutes, schools, the police and industry (particularly the Bauxite Mine nearby – Mineração Rio Norte) to protect the Reserve and promote research and education (Fig 6.3). Although, in theory, access to a Biological Reserve is restricted to research and educational purposes, traditional groups were living in the area before the implementation of the Reserve. Many of these groups are still living inside the area or enter the area for subsistence activities such as fishing and Brazil

Nut (Bertholletia excelsa) collection. The traditional communities living within the reserve and its surroundings include the local indigenous people and other non-indigenous groups such as the quilombolas (African slave descendants living in communal enclaves) and ribeirinhos (traditional extractive populations that rely mostly on fishing and subsistence agriculture) (Chapter 5; Table 5.2). The ribeirinhos, making up the largest traditional group within the Brazilian Amazon, are from mixed Amerindian, European and African descent. In contrast with the indigenous groups, which speak their own languages, the ribeirinhos are a traditional rural population who speak Portuguese, reside near the river growing cassava, practicing shifting irrigation, fishing and harvesting forest resources to sustain their livelihoods (Oliveira 2006).

Figure 6.2. Rio Trombetas Biological Reserve. Areas protected during the turtle nesting season are highlighted (Moura 2017). Area monitored and protected by ICMBio rangers varies according to the resources allocated in a given year. The Quilombola communities are descendants from the slave period when, during the 16th century, Africans were shipped to Brazil to work on the land, in areas such as sugar

plantations and mines. Their ancestors fled from slavery in the late nineteenth century, thus becoming known as "quilombos", which means “hideout” in Ambundu (Angola's 65

North-West Bantu language). The Trombetas River offered conditions favourable to the Quilombos, as it is difficult to access due to its many waterfalls. The Quilombolas started to be recognized as traditional communities in the 1980s, when the Federal Constitution of Brazil came into force. This determined the territorial regularization of Quilombola communities and protected their cultures.

Unlike other Quilombolas within the Brazilian Amazon, the Quilombolas of Pará state (where the ReBio-Trombetas is located) hold claim to land within their territories, a claim which requires proof that the land was a ‘means of existence’ and that ‘knowledge and practices’ have continued to be passed on through the generations (Soares-Filho 2006).

Fig 6.3. Wildlife conservation stakeholders associated with the Rio Trombetas Biological Reserve. Photos: C. Eisemberg, S. Reynolds, A. Bean, H. Hunter-Xenie, L. Martini, D. Bowman, S. Barrett. Threats From 1981 to 2000 over five million hatchlings of P. expansa, P. unifilis and P. sextuberculata were released in the Reserve. Despite annual conservation actions, a sharp decline in the number of nests and hatchlings of P. expansa was observed over the past 30 years at the protected sandbanks (Fig 6.4). Understanding and stopping this decline has been one of the main goals for wildlife managers, park rangers and researchers (RANIBAMA 2003).

Turtle hunting has long been embedded within the Amazon culture. From traditional hunting practices when turtle meat was a staple diet, to now, when the consumption of turtle products have, in many cases, become a “delicacy”, monetary gains are replacing the need for core dietary requirements. Turtle harvest is illegal in Brazil, and wild turtles are generally sold on the black market, one turtle being able to return a quarter or more of monthly income (Moll et al. 2004; Schneider et al. 2011). That in itself poses a serious ongoing and 66

confronting threat to turtle populations. An illustration of the pressures associated with managing the conservation of turtle populations is seen in the current supply and demand in Amazonia, where wealthy people of the Amazon use turtle meat from P. expansa, P. unifilis, P. sextuberculata and Podocnemis erythrocephala (Red-headed Amazon River Turtle) as a main meal during celebrations, known as a “tartarugada”. This in turn creates a market for poorer communities to supply turtle meat (Alho 1985). Overharvest is still the main cause of P. expansa decline in the Amazon (Fig. 6.5) (Ojasti 1967; Andrade et al. 1988; Kemenes &

Pezzuti 2007; Schneider et al. 2011). Podocnemis expansa is a highly mobile species that can travel over 200 km between feeding grounds and nesting beaches (Hildebrand et al. 1988). Illegal harvest outside and inside the Trombetas Reserve is the main reason behind the population decline. While one side of the River is a protected Reserve, the other side is not protected. The area is inhabited by local communities that invade the Reserve with impunity. Based on radio tracking studies, about 20% of the nesting population is lost to poachers (Pearse et al. 2006). There is also evidence that in the past hatchlings produced from the protected beaches were sold illegally to turtle ranchers downstream (Vogt 2014).

Figure 6.4. Decline in Podocnemis expansa hatchlings recorded at areas protected by ICMBio rangers at the for 30 years, to 2005. Data: SisQuelônios (2016).

Other anthropological disturbances such as hydroelectric dams, roads and dredging can also negatively impact populations of Podocnemis and their nesting habitats (Fig. 6.5) (Alfinito 1975; Smith 1975; Mittermeier 1978). The Trombetas River has been dredged to facilitate boat access to Porto Trombetas, 70 km downstream from the Reserve. Dredging of riverbeds can affect Podocnemis turtles (Moretti 2004; Haller & Rodrigues 2006) since it disturbs fluvial sediment dynamics and destroys nesting sites (Rodrigues 2005). There have also been proposals to build a dam upstream of the Reserve (Cachoeira Porteira hydroelectric dam) since the 1980s (CEDI

1991). A hydroelectric dam would severely modify the Trombetas River flood pulse (Chapter 1) with tragic consequences for the river turtle population (Chapter 3 and 4), destroying both nesting and feeding habitat (Eisemberg et al. 2016). Changes in the water cycle due to climate change could also combine with the key sources of decline and produce a cumulative effect on P. expansa decline by affecting the time of sandbank exposure and increasing nest mortality (Eisemberg et al. 2016). Furthermore, recent studies have demonstrated that P. expansa vocalize to 67

communicate and hatchlings migrate in synchrony with the females from the nesting sites to the feeding grounds (Ferrara et al. 2013, 2014; Bernardes et al. 2014). Therefore,

excess underwater noise from large vessels could also potentially affect the migration of adults and hatchlings (Fig. 6.5).

Fig 6.5. Threats and actions needed for the protection and conservation of river turtle species in the Rio Trombetas Biological Reserve.

Management and conservation The research, management and environmental education activities undertaken at ReBioTrombetas have produced positive outcomes. Since 2006, the number of P. sextuberculata and P. unifilis hatchlings has been increasing (Fig. 6.6). Although the number of P. expansa hatchlings per year is still declining, the decline is not as steep as it was during the 1980s and 1990s, which would have led to local extinction (Fig 6.6). Currently, the River Turtles conservation Project at ReBioTrombetas has three aims (Moura 2017): - Protect and recover the populations of P. expansa, P. unifilis and P. sextuberculata. - Promote community-led management initiatives and grassroots environmental education about the turtle conservation program actions and aims. - Study the population dynamics of the river turtles species at the areas protected within the ReBio-Trombetas.

Since the early 1980s, most of the conservation efforts within the ReBioTrombetas have been focused on monitoring and protection of nesting sites near the Tabuleiro Basecamp and Jacaré Lake (total area of 50 km²; Fig. 6.2) by government rangers (Fig 6.7a). During the past 10 years, a community-based approach has also been initiated at the Erepecu Lake to protect and manage nests of P. sextuberculata and P. unifilis. Under this program, the local community participates actively in the protection and translocation of nests to safe areas (Fig 6.7b). For example, during the 2016-2017 nesting season, 22 park rangers worked at the Tabuleiro and Santa Rosa Basecamps, together with 27 local volunteer families. Two workshops were offered to prepare the volunteers (25 hours) and rangers (35 hours) for the standard monitoring and management procedures. In total, 18 rangers and 75 local volunteers participated in this workshops and conservation actions (Moura 2017). 68

Figure 6.6. Trends for the numbers of P. expansa hatchlings recorded from areas protected by ICMBio rangers at the Tabuleiro Basecamp (blue) and numbers of P. unifilis and P. sextuberculata (sum of both species) hatchlings at the areas managed at Erepecu Lake (orange). Data from Moura (2017). The ReBio-Trombetas has a strong history of research applied to conservation. The nesting ecology of riverine turtles has been studied since the late 1970s (Alho & Pádua 1982; Alho et al. 1985). Since then, many studies on population dynamics, ecology and genetics have been published (Haller & Rodrigues 2006, Ponce de Leão 2015; Freda et al. 2016; Eisemberg et al. 2016).

Innovative technologies, such as the use of satellite and sonic telemetry, bioacoustics and underwater video cameras were applied for the first time on Amazon river turtle species at the ReBio-Trombetas, which led to breakthrough discoveries, such as P. expansa sound communication, social behaviour and post-hatching parental care (Ferrara et al. 2013, 2014, 2015; Bernardes et al. 2014).

Figure 6.7. Conservation efforts to protect the Amazon River turtles at the Tabuleiro Basecamp with employed ICMBio rangers (“beach” agents”) (left) and Erepecu Lake with local volunteer families (right). Photos: C. Eisemberg, L. Martini, C. Whittaker. Grassroots environmental education activities are also vital for increasing awareness about the endangered status of the river turtles at the ReBio-Trombetas. Such activities are usually aimed at the local schools and the Porto Trombetas village school. Students are invited to hatchling release events as well as talks and

research activities. For example, in 2015, secondary students from Professor Jonathas Pontes Athias School (Porto Trombetas Town) visited the large sandbanks at Tabuleiro Basecamp and participated in hands-on activities related to the ReBio-Trombetas research and conservation efforts (Fig. 6.8). 69

Figure 6.8. Secondary students from the Professor Jonathas Pontes Athias school (Porto Trombetas) during their visit to the Tabuleiro Basecamp sandbanks. Photos: C. Eisemberg. A postgraduate field intensive course to teach Brazilian masters and PhD students about the research behind freshwater turtle biology and conservation has also been running intermittently at the ReBio-Trombetas for 20 years. In 2016, a similar course, with a broader scope, was offered for the first time to sixteen international students enrolled in Australian Universities (Fig. 6.9). Environmental Science undergraduate and postgraduate students from

the Brazilian Amazon and northern Australia share many interests and challenges. Both regions are located in the tropics and are considerably remote. Above all, both regions have a rich wildlife, with species surprisingly related to each other due to ancient Gondwana connections. Chapter 7 will discuss in detail this northern Australia and northern Brazil connection.

Figure 6.9. Australian undergraduate and postgraduate students at Re-Bio Trombetas during the Brazilian Amazon Field Intensive, November 2016. Photos: J. Marr, C. Eisemberg. Literature Cited Alfinito J, Vianna CM, Valle RC, Silva MMF. 1973. Preservação da Tartaruga da Amazônia. Simpósio Internacional Sobre Fauna Silvestre e Pesca Fluvial e Lacustre Amazônica. IBDF/SUDEPE/MA, Manaus. Alfinito J. 1975. A preservação da tartaruga da Amazônia. Brasil Florestal 6: 20-23. Alho CJR, Pádua LFM. 1982. Reproductive parameters and nesting behavior of the Amazon turtle Podocnemis expansa (Testudinata: Pelomedusidae) in Brazil. Canadian Journal of Zoology 60: 97-103.

70

Alho CJ. 1985. Conservation and management strategies for commonly exploited Amazonian turtles. Biological Conservation 32: 291-298. Alho CJ, Danni TM, Pádua LF. 1985. Temperature-dependent sex determination in Podocnemis expansa (Testudinata: Pelomedusidae). Biotropica 17: 75-78. Andrade PCM et al. 1998. Consumo de produtos da fauna silvetre no Estado do Amazonas. Relatório parcial das atividades do Convênio Universidade do Amazonas e IBAMA na área de fauna silvestre. Faculdade de Ciências Agrárias da Fundação Universidade do Amazonas, Manaus. Arraes DRDS, Tavares-Dias M. 2014. Nesting and neonates of the yellow-spotted river turtle (Podocnemis unifilis, Podocnemididae) in the Araguari River basin, eastern Amazon, Brazil. Acta Amazonica 44: 387-392. Bernardes VCD, Vogt RC, Ferrara CR. 2014. Tracking migrating hatchlings of giant Amazon River Turtles Podocnemis expansa in the Rio Trombetas, Brazil with sonic transmitters. 12th Annual Symposium - Turtle Survival Alliance, Orlando. CEDI. 1991. Povos Indígenas no Brasil 1987/1990. Centro Ecumênico de Documentação de Informação, Sao Paulo. Eisemberg CC, Balestra RAM, Famelli S, Pereira FF, Bernardes VCD Vogt RC. 2016. Vulnerability of Giant South American Turtle (Podocnemis expansa) nesting habitat to climate-change-induced alterations to fluvial cycles. Tropical Conservation Science 9: 1940082916667139. Eisemberg CC, Reynolds SJ, Christian KA, Vogt RC. 2017. Diet of Amazon river turtles (Podocnemididae): A review of the effects of body size, phylogeny, season and habitat. Zoology 120: 92-100. Ernst CH, Barbour RW. 1989. Turtles of the World. Smithsonian Institution Press, Washington. Ferrara CR, Vogt RC, Sousa-Lima RS. 2013. Turtle vocalizations as the first evidence of posthatching parental care in chelonians. Journal of Comparative Psychology 127: 1-9. Ferrara CR, Vogt RC, Sousa-Lima RS, Tardio BM, Bernardes VCD. 2014. Sound communication and social behavior in an Amazonian river turtle (Podocnemis expansa). Herpetologica 70: 149-156. Ferrara CR, Vogt RC, Bernardes VCD. 2015. A glimpse into the social life of freshwater turtles from their eyes: a study case of 10 Podocnemis expansa underwater. 13th Annual Symposium - Turtle Survival Alliance, Tucson. Ferreira-Júnior PD, Castro PTA. 2010. Nesting ecology of Podocnemis expansa (Schweigger, 1812) and Podocnemis unifilis (Troschel, 1848)(Testudines, Podocnemididae) in the Javaés River, Brazil. Brazilian Journal of Biology 70: 85-94. Freda FP, Bernardes VCD, Eisemberg CC, Fantin C, Vogt RC. 2016. Relationship between multiple paternity and reproductive parameters for Podocnemis sextuberculata (Testudines: Podocnemididae) in the Trombetas River, Brazil. Genetics and Molecular Research 15: gmr.15017335 Haller ECP, Rodrigues MT. 2006. Reproductive biology of the Six-Tubercled Amazon River Turtle Podocnemis sextuberculata (Testudines: Podocnemididae), in the Biological Reserve of Rio Trombetas, Pará, Brazil. Chelonian Conservation Biology 5: 280-284. Hildebrand P, Saenz C, Pehuela M, Caro C. 1988. Biologia reproductiva y manejo de la tortuga Charapa (Podocnemis expansa) en el bajo rio Caqueta. Colombia Amazonica 3: 89-102. IBAMA 1989. Projeto quelônios da Amazônia 10 anos. Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis, Brasília. Kemenes A, Pezzuti JCB. 2007. Estimate of trade traffic of Podocnemis (Testudines, Podocnemididae) from the middle of Purus River, Amazonas, Brasil. Chelonian Conservation Biology 6: 259-262. Mittermeier RA. 1978. South America's river turtles: Saving them by use. Oryx 14: 222-230.

71

Moll D, Moll EO. 2004. The Ecology, Exploitation, and Conservation of River Turtles. Oxford University Press, Oxford. Moretti R. 2004. Biologia reprodutiva de Podocnemis erythrocephala (Spix, 1824), Podocnemis expansa (Schweigger, 1812) e Peltocephalus dumerilianus (Schweigger, 1812) (Testudinata, Podocnemididae) na bacia do Rio Trombetas, Pará (Master Thesis). Universidade de São Paulo, São Paulo. Moura CMM. 2017. Projeto Quelônios do Rio Trombetas, Relatório Anual – Temporada 2016. Instituto Chico Mendes de Conservação da Biodiversidade - ICMBio, Porto Trombetas. Ojasti J. 1967. Consideraciones sobre la ecologia y conservacion de la tortuga Podocnemis expansa (Chelonia, Pelomedusidae). Atas Simposio Sobre Biota Amazônica 7: 201-206. Oliveira NFB. 2006. The Political Significance of Non-tribal Indigenous Youth’s Talk on Identity, Land, and the Forest Environment: An Amazonian Case Study from the Arapiuns River, Brazil. PhD Thesis. Australian National University, Canberra. Pearse DE, Arndt AD, Valenzuela N, Miller BA, Cantarelli V, Sites JW. 2006. Estimating population structure under nonequilibrium conditions in a conservation context: continent-wide population genetics of the giant Amazon river turtle, Podocnemis expansa (Chelonia; Podocnemididae). Molecular Ecology 15: 985-1006. Peña AP. 2002. Floresta Nacional Saracá-Taquera (Relatório Técnico Anual). Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis, Oriximiná. Pignati MT, Fernandes LF, Miorando PS, Ferreira PD, Pezzuti JC. 2013. Effects of the nesting environment on embryonic development, sex ratio, and hatching success in Podocnemis unifilis (Testudines: Podocnemididae) in an area of várzea floodplain on the lower Amazon River in Brazil. Copeia 2013: 303-311. Ponce de Leão SEGM. 2015. Movimentação e uso do espaço por adultos de Podocnemis unifilis Troschel, 1848, na Reserva Biológica do Rio Trombetas, Pará, Brasil. Master Thesis. Instituto Nacional de Pesquisas da Amazônia, Manaus. RAN-IBAMA. 2003. Quelônios da Amazônia – PQA-PA/Trombetas. Centro Nacional de Pesquisa e Conservação de Répteis e Anfíbios, Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis, Goiás. Rodrigues MT. 2005. The conservation of Brazilian reptiles: Challenges for a megadiverse country. Conservation Biology 19: 659-664. Schneider L, Ferrara CR, Vogt RC, Burger J. 2011. History of turtle exploitation and management techniques to conserve turtles in the Rio Negro basin of the Brazilian Amazon. Chelonian Conservation Biology 10: 149-157. SisQuelônios. 2016. SisQuelônios - Sistema de Gestão e Informação dos Quelônios Amazônicos. Instituto Chico Mendes de Conservação da Biodiversidade – ICMBIO, Centro Nacional de Pesquisa e Conservação de Répteis e Anfíbios – RAN. http://www.ibamanet.ibama/sisquelonios (accessed June 2016). Smith NJH. 1975. Destructive exploitation of the South American river turtle. Chelonia 2: 1-9. Soares-Filho BS, Nepsted DC, Curran LM, Cerqueira GC, Garcia RA, Ramos CA, Voll E, McDonald A, Lefelovre P, Schlesinger P. 2006. Modelling Conservation in the Amazon Basin. Nature 440: 520-523. Souza RR, Vogt RC. 1994. Incubation temperature influences sex and hatchling size in the neotropical turtle Podocnemis unifilis. Journal of Herpetology. 28: 453-464. Vogt RC. 2008. Tartarugas da Amazônia. Gráfica Biblos, Lima. Vogt RC. 2014. Chattering turtles of the Rio Trombetas. The Tortoise 1: 118-127.

72

CHAPTER 7 – NORTHERN AUSTRALIA AND NORTHERN BRAZIL CONNECTIONS Sarah Sutcliffe, Storm Barrett, Larissa Schneider, Keith Christian Introduction Despite the immense oceans between them, there are many similarities between the ecology and environmental management issues of northern Brazil and northern Australia. These parallels began when the two landmasses were part of the supercontinent of Gondwana over 180 million years ago (Chapter 3). The Gondwana link explains many similarities in a number of familial groups, including marsupials and side-necked turtles (family Chelidae). In addition to this ancient geological link, northern Australia and northern Brazil share similarities in climate. Both regions are located in the tropics, and are characterised by seasonally high rainfall, which drives the distinct ecologies of the two regions. The environmental similarities, in combination with the vastness, remoteness and relatively intact nature of the two regions, means that they share unique management challenges. Consequently, drawing comparisons between the regions and understanding what we may learn from each other could help to improve future wildlife management in both areas. This chapter highlights the connections and parallels between northern Brazil and northern Australia and links issues in wildlife management to provide lessons and insights concerning sustainable use and management of the natural resources of these two regions. Northern Australia and northern Brazil regions The North of Brazil is the largest region of Brazil, comprising 45% of the national territory. This region is the least inhabited by humans and consists primarily of Amazon rainforest and, to a lesser extent, of cerrado ecoregions to the north and south-east. It therefore comprises the largest and most

species-rich tract of tropical rainforest in the world (Turner et al. 2001). The states of northern Brazil are Acre, Amapá, Amazonas, Pará, Rondônia, Roraima and Tocantins. Their economies are essentially based on plantations for latex, açaí, wood and nuts, and mineral extraction of gold, precious stones, cassiterite (tin oxide) and tin. Other mining activities include iron in the Carajás Mountain Range and manganese in the Navio Mountain Range (de Paula et al. 2016). The term northern Australia includes parts of Queensland, the Northern Territory north of about Newcastle Waters and Western Australia north of Broome (18°S); areas which are dominated by savanna vegetation (Fox et al. 2001). Similar to northern Brazil, northern Australia comprises nearly half of the total area of Australia, but it includes only about a quarter of the population. The soil is mostly relatively poor, and dense forests (monsoon forests or vine thickets) establish themselves only in sheltered places in higher rainfall zones. This has created a type of tropical savanna environment in which fires play a crucial role in influencing nutrient levels and growth responds dramatically to wet season rains (Williams et al. 1999). As for northern Brazil, northern Australia’s economy is based on exports such as cattle, limited irrigated agriculture, coal from the Great Dividing Range in Queensland and mining (Woinarski et al., 2007). Wildlife ecology As with northern Brazil, northern Australian flora and fauna are adapted to the environmental conditions of the region. Both regions are influenced by the location of the Intertropical Convergence Zone (ITCZ), where northeast and southeast trade winds 73

converge, and this accounts for some of their climatic similarities. In northern Australia, most rainfall occurs from December to March (the wet season, Fig. 7.1), and is highly variable between years (Taylor & Tulloch 1985; Warfe et al. 2011). This marked seasonality in rainfall creates flood pulse events similar, but lesser in scale, to those in the Brazilian Amazon (Chapter 1), and is fundamental to the region’s ecology. Both regions support globally significant wetlands. However, northern Brazil is dominated by rainforest communities with smaller patches of savanna (cerrado) and northern Australia is dominated by savanna with patches of riparian forests associated with waterways and rainforests in sheltered sites (Fig. 7.2). The long period of isolation of the Australian continent has resulted in high levels of endemism and a unique biotic character (Braithwaite 1990). Similar to the flood pulse responses in Amazon (Chapter 2), many northern Australian vertebrates have reproductive strategies that are adapted to the highly seasonal and variable nature of the rainfall and flood pulse dynamics of the region. These adaptations include mechanisms to exploit or buffer the effects of stochastic variation in wet season rainfall. Some reproductive adaptations include strategies such as using direct rainfallrelated cues to synchronise breeding, nesting on higher ground, and flexible developmental trajectories (Shine & Brown 2008). For example, like some Brazilian freshwater turtle species, such as the Scorpion Mud Turtle (Kinosternon scorpioides scorpioides) in the northern state of Maranhão (Pereira et al. 2007), the Northern Snake-necked Turtle (Chelodina rugosa) exhibits adaptations to the wet-dry seasonal environment. They spend the dry season aestivating buried in the soil (Grigg et al. 1986), where they survive by conserving energy by depressing their metabolic rates (Kennett & Christian 1994). Many northern Australia fish species, such as Barramundi (Lates calcarifer), breed during the wet season, increasing the chances of juvenile survival by

providing access to habitats which have fewer predators and less competition for resources (Castello et al. 2011; Stewart‐Koster et al. 2011). Reproduction in the Northern Snake-necked Turtle has evolved in response to wet season flooding whereby eggs are laid underwater in seasonally flooded areas (Kennett et al. 1993). Another physiological mechanism, developmental arrest, allows the embryos to survive in a suspended state until the floodwaters recede and development begins (Seymour et al. 1997). It is likely that some species of freshwater turtle from Amazon terra firme creeks, such as the Red Side-necked Turtle (Rhinemys rufipes), also lay their eggs underwater (Eisemberg 2006). The Australian Pig-nosed Turtle (Carettochelys insculpta) has also developed unique adaptations to deal with the high natural variability of wet season onset and magnitude (Doody et al. 2003). Similar to the riverine Amazon turtles from the family Podocnemididae (Eisemberg et al. 2016), timing of C. insculpta nesting (Fig. 7.1) is correlated with the size of the preceding wet season and appears to be related to food acquisition. After yolk internalisation, Pignosed Turtle embryos enter developmental arrest for up to 59 days until eggs are exposed to anoxia from inundation and hatch explosively within minutes, communicating through vibrations to ensure synchronised emergence (Doody et al. 2012). Timing emergence with the onset of the wet season maximises food availability for the offspring and may also reduce mortality from predation as water clarity and visibility is reduced. The wetland and floodplain vegetation of northern Australia shares similar adaptations with northern Brazil in response to the seasonal flood pulse. Many floodplain species have developed aerenchyma (spongy tissue that forms spaces or air channels in the leaves, stems and roots of some plants), adventitious roots, air cavities and corky tissues to assist in gas exchange and buoyancy during inundation 74

(Junk & Piedade 1993; Parolin & Wittmann 2010). Despite long dry periods, the majority of dispersal mechanisms in northern Australian floodplains involve water, either by seed buoyancy mechanisms, trailing stems, or floating vegetative parts (Cowie et al. 2000).

Vegetation community composition is largely determined by the historical flooding frequency and magnitude (Finlayson 2005). Persistence of these vegetation communities is therefore highly dependent on natural flooding events.

Figure 7.1. Monthly rainfall and interannual variation in monthly rainfall for the Daly River (Douglas Daly Research Farm from 1968 to 1997). Horizontal bar shows the nesting season for the pig-nosed turtle (Carettochelys insculpta) population In the Daly River (Doody et al. 2004). Rainfall averages are given with their 95% confidence limits (vertical boxes) and ranges (vertical bars). Adapted from Eisemberg et al. (2015). Fire is a dominant feature of northern Australian savannas and northern Brazilian cerrado. Rainfall in the wet season promotes rapid growth rates of C4 grasses, which become cured during the dry season, providing abundant fuel loads. As a result, savanna communities are dominated by fire-adapted species. These adaptations include pronounced bark thickness, resprouting by lignotubers and seed dormancy (Russell-Smith et al. 2012; Simon & Pennington 2012). Land tenure Northern Australia and northern Brazil also share historical similarities in land use. Much of northern Australian savanna is used for low intensity pastoralism (75%), with other uses

including indigenous lands (15%), conservation reserves (6%) and military training areas (1%; Woinarski et al. 2007, 2013). Less than five percent of the region is cleared and, as a result, the native vegetation remains largely intact (aside from weed invasion), creating a unique opportunity for conservation planning (Fig. 7.3). Recognition of indigenous land rights occurred relatively recently in both Brazil and Australia, beginning in the 1970s. Following land rights and native title legislation, aboriginal-owned land now accounts for approximately 80% of northern Australia’s coastline, and 50% of the land (Franklin et al. 2008). In Australia 365 000 km2 has been declared as Indigenous Protected Areas, 40 000 km2 of which is in the Northern 75

Territory (Hancock 2014; Commonwealth of Australia 2017). In contrast, 1 million km2 of the Amazon is officially recognized as

indigenous lands, but this accounts for only 23% of the Brazilian Amazon (Schwartzman & Zimmerman 2005; PIB 2014).

Figure 7.2. Comparison of broad vegetation types between Amazon and key bioregions of northern Australia. Data from PROBIO (2004) and Kutt et al. (2009). Rainforests cover around one million hectares of the wet-dry tropics of northern Australia, however, this is less than 1% of the land area (Woinarski et al. 2007).

Northern Brazil–Northern Australia connection The Ramsar Convention is an international treaty, signed in 1971, for the conservation and sustainable use of wetlands. Both the Brazilian Amazon and northern Australia have Ramsar sites, which are recognised as wetlands of international importance. As of April 2015, there are over 2 100 designated Ramsar Sites with 65 sites in Australia (eight in northern Australia including Cobourg Peninsula, Kakadu National Park, Roebuck Bay and Lakes Argyle and Kununurra) and 13 sites in Brazil (three in Amazon: Mamirauá, Cabo Orange National Park, and Reentrâncias Maranhenses) (RAMSAR 2017). Traditional people have shaped and managed northern Australian landscapes for centuries, however, multiple factors have seen much of the remote area become increasingly depopulated. This depopulation, combined with resource restrictions, has meant that traditional practices of land management that support biodiversity (particularly fire management and pest and weed control) are increasingly difficult to implement (Woinarski

et al. 2007). Unlike Brazil’s extensive protected areas network (Chapter 5), which contains 54% of the remaining forests of the Brazilian Amazon (Soares-Filho et al. 2010), just 6% of northern Australia is managed as conservation reserves (Fig 7.3) (Franklin et al. 2008; Woinarski et al. 2013). On the other hand, Australian reserves are typically more actively managed than Brazilian reserves. Many of these reserves, including Kakadu National 76

Park, are jointly managed with traditional owners, and contribute significantly to biodiversity values (Woinarski et al. 2013). Given the small proportion of conservation land and the dynamic nature of savanna ecology, there is justification for management of biodiversity across all land tenures despite political boundaries (Woinarski et al. 1992, 2007, 2013). Environmental issues, such as the spread of invasive species and fire, are rarely restricted to particular land tenures, which is a great source of conflict in northern Australian land management. Land management practices differ significantly depending on the culture, background, knowledge and specific objectives of the land manager, creating considerable room for misunderstanding. This has engendered antagonistic relationships between indigenous groups and pastoralists relating to fire management, Aboriginal people often being met with hostility and ‘blamed’ for their seemingly random use of fire (Martin 2013). In Australia, conflicts between indigenous groups and pastoralists also arise as a result of competing land management objectives. For example, in northern Queensland, conservation objectives aimed at protecting rainforest through fire suppression has led to rainforest encroaching on culturally important open sclerophyll forest, and a loss of traditional resources used for shelter and weaving (Hill et al. 1999). In the same region, large and hot fires used for weed control and improving green-pick for cattle have been known to burn out of control and destroy areas of fire-sensitive rainforest. In comparison, conflicts over land in the Brazilian Amazon are usually more violent and can often result in death. Such conflicts are multifaceted and involve numerous actors such as indigenous people, miners, loggers, ranchers and small farm holders. The Brazilian government, intending to bring economic and social development to the region, promoted

development strategies that led to a relative land scarcity, a prime ingredient for agrarian problems and land conflict. Many indigenous reserves, unlike conservation units, are located near areas of settlements, and there have been numerous violent incursions onto indigenous lands (Simmons 2004). Successful management of these regions rests upon improving communication and understanding between a diverse range of land managers, thereby building the capacity of the community to accomplish environmental goals. Threats to wildlife Many of the threatening processes found in northern Brazil (Chapter 4) are also occurring in Australia, although the causes, scale and management of these threats are not always the same. One of the main drivers of degradation in both regions is pressure for economic development. The pressure to ‘develop the north’ is occurring in both regions, while the majority of the population occupies the southern regions. In both cases, development has been a fairly recent phenomenon due to historical difficulties with access and implementation of projects in these remote regions. The implementation of the Manaus Free Trade Zone in the 1960s with an industrial centre, as well as an agricultural centre, and construction of major highways to allow for regional development in order to lift the Amazon out of economic isolation, triggered intensified and accelerated deforestation in northern Brazil in the 1970s and 1980s (Malingreau & Tucker 1988; Moran 1993; Kirby et al. 2006; Davidson et al. 2012). Although rates of deforestation in northern Brazil have declined since 1988, an estimated 2 million hectares of the Brazilian Amazon is cleared annually to support infrastructure, pastoralism, agriculture and urbanisation (Laurance et al. 2001).

77

Figure 7.3. Map of northern Australia showing land tenure for conservation, Aboriginal land, and state forest (adapted from Woinarski et al. 2007). The map also shows areas where (in eucalypt dominated plant communities) less than 10% of the area was cleared across northern Australia (adapted from Franklin and Preece 2014). In Australia, rapid and extensive land clearing and deforestation has occurred since European settlement, mainly in the southern states. However, as with Brazil, the expansion of Australia’s pastoral and agricultural industries in the past 50 years (Fig. 7.4) has seen a push for development in the north, leading to the destruction of over 50% of north Queensland’s primary tropical forest, largely for grazing (Bradshaw 2012). Deforestation and degradation in northern Australia and northern Brazil, driven by increasing pressure for development, has the potential to significantly alter the ecological structure and function of these landscapes, threatening biodiversity, altering soil nutrients and triggering large-scale changes in regional rainfall patterns and global climate, among other impacts (Laurance 1999). The varied and irreversible impacts of land clearing and deforestation make it one of the most serious environmental issues. In addition to clearing, exploitation of natural resources threatens to further degrade these landscapes. Mining is a major industry in northern Australia and Brazil, generating significant income for both regions. Although mining is a temporary land use, it has longterm impacts on the environment which are

often extremely difficult, if not impossible to reverse. Rehabilitation of mined land and associated wastes is a legal requirement in Australia, yet success has been varied. For example, the 1950s uranium and copper project at Rum Jungle in the Northern Territory caused widespread environmental pollution. Attempts to rehabilitate the area in the 1980s were largely unsuccessful, and significant groundwater pollution still exists (Mudd & Patterson 2010). Mining in the Amazon may not cause deforestation on the same scale as logging and mass agriculture, but it can affect the environment in the vicinity of the mining site and downstream. The Amazon is considered to have great potential for minerals, namely copper, tin, nickel, bauxite, manganese, iron ore and gold (Barreto, 2001). As a result, governments are providing tax incentives to promote large-scale projects. In Brazil, power from hydroelectric dams accounts for 77% of the energy supply (Ferreira et al. 2014). Pressure to develop northern Brazil has resulted in unprecedented investment in hydroelectric dams, largely to provide electricity for aluminium smelting (Fearnside 2016). As of 2015, 15 large dams (>30 MW) were operating in the Legal Amazon region, with 37 planned or under 78

construction. These dams have had widespread social and environmental consequences, displacing communities, altering hydrological dynamics, blocking fish migration and causing deforestation and mercury contamination (Fearnside 2016). Although there is only a small number of dams in northern Australia to date (Fig 7.4), the abundance of freshwater in the region (over 60% of Australia’s surface run-off) makes this resource increasingly attractive for development, particularly for agriculture (Hart 2004; Warfe et al. 2011; Karim et al. 2015). Limited ecological understanding of these

tropical freshwater systems makes development a risky venture, and there are concerns that uninformed development may lead to repetition of the adverse impacts seen in other tropical systems. On the other hand, the relatively untouched nature of northern Australian freshwater systems provides a unique opportunity to understand the ecological role of natural flow variability in tropical systems, which may benefit future regional water resource development and provide a model for wetland restoration and conservation elsewhere (Warfeet al. 2011).

Figure 7.4. Land tenure for livestock grazing (adapted from Woinarski et al. 2007). Large dams with a crest height greater than 10 metres marked with dots, blue lines delineate drainage divisions. The map also shows areas where (in eucalypt dominated plant communities) more than 50% of the area has been cleared (adapted from Franklin and Preece 2014). Since the 1950s, increases in atmospheric CO2 leading to global warming have been observed at unprecedented rates (IPCC 2014). Climate change is predicted to result in increased temperatures, modified rainfall patterns and a greater number of extreme climatic events, among other effects (Hughes 2003; Junk et al. 2013). For example, some models predict up to a 70% reduction in the extent of the Amazon rainforest by the end of the twentyfirst century as a result of reduced rainfall and modified seasonality (Cook & Vizy 2008). Similarly, northern Australian ecosystems are

vulnerable to climate change and are expected to undergo dramatic changes in vegetation structure and loss of biodiversity. In the tropics, climate change is predicted to increase rainfall variability and dry season severity, as well as increasing frequency of extreme weather events, including cyclones and floods, thereby altering the flow regime that aquatic organisms depend upon (Williams et al. 2003). There are, therefore, growing concerns for the ability of species to adapt to these changes, in combination with other pressures, and the possibility of large scale extinction. 79

Fire is an integral and determining feature of the Brazilian and Australian savannas. Historically, indigenous people inhabiting these biomes used fire for a variety of reasons, creating a mosaic of burnt and unburnt patches (Mistry et al. 2005; Whitehead et al. 2009). As was the case in Australia, before European colonization Brazilian indigenous peoples used fire with well-defined objectives, locations and knowledge of fire behaviour (Pivello 2011). Although these practices were more common in the cerrado, there is evidence that indigenous groups from the Amazon Basin used fire to create a very dark, organic, fertile soil with a high charcoal content. The great majority of current wildfires in northern Brazil are caused by human ignition, aimed at removing the natural vegetation or managing agricultural crops (Pivello 2011). Logging of the Amazon forest (Fig. 7.5a) itself does not require fire. However, extraction of large trees increases the susceptibility of the forest to wildfires. Once logged, the forests are no longer profitable, and the remaining vegetation is usually burned for the land to be used as pasture (Fearnside 2005, Balch et al. 2009). Changing climate and land use has significantly altered fire regimes and led to extensive changes to ecological function of northern Australian savannas (Russell-Smith et al. 2003). In Brazil, Silvério et al. (2013) found that fires associated with long-term drought events promoted grass invasion and converted rainforest areas into grass-dominated habitats. Experiments in Australia have found that changing fire regimes potentially leads to a biome switch between savanna and closed forest (Lehmann et al. 2014). In northern Australia, a shift toward a greater frequency of

late dry-season, high-intensity fires that threaten property, livelihoods, vegetation structure and biodiversity has made fire a major environmental management issue (Whitehead et al. 2009). Looking forward, accelerated climate and land use change on both continents will continue to alter natural fire regimes, and therefore ecological structure and function, as well as contribute significantly to greenhouse gas emissions. There is a clear need for adaptive and well-informed fire management strategies. Management and conservation Environmental management in northern Australia shares many challenges with northern Brazil. The sheer size and remoteness of these regions means that implementation of management activities can be costly and time consuming. Another common difficulty lies in managing environmental issues in a crosscultural setting. To illustrate, efforts to eradicate feral Buffalo (Bubalus bubalis; Fig. 7.5b) as part of the national Brucellosis and Tuberculosis Eradication Campaign in Australia in the 1980s were made extremely difficult and expensive by the vast and rugged terrain. Regional extermination was deemed economically and practically impossible, despite an investment of eight million dollars in the Northern Territory (Bowman & Robinson 2002). Additionally, despite the cultural and economic significance of buffalo to aboriginal people (they are prized and hunted for their meat), and the fact that half of the high-risk area was aboriginal land, little community consultation was undertaken and no compensation was offered for culling these herds (Robinson & Whitehead 2003).

80

Figure 7.5. Threats to northern Brazil and northern Australia. Fire is not as much of a threat in the Amazon rainforest as it is in northern Australia, however, extraction of large trees (a) increases susceptibility to fires. The introduction of (b) Water Buffalo (Bubalus bubalis) has caused extensive damage to the floodplains of the Amazon and northern Australia. Photos: S. Reynolds. Failure to involve these communities led to management decisions with little community support and ultimately an economically unrealistic eradication target. Although the Eradication Campaign reduced feral buffalo populations, failure to implement follow up measures resulted in rapid recovery. As a result, buffalo continue to impact the environment and act as a vector for disease. The Water Buffalo population in the Brazilian Amazon is one of the fastest growing in the world. From 1975 to 2000, it increased 13% per year. Although the use of buffalo is desirable and profitable for many people in the Amazon, there are conflicts between pastoralists and fishermen due to the damage caused to the floodplains, particularly during the wet season (Sheikh et al. 2006). The experience and information gathered in northern Australia can be valuable for future plans to manage Water Buffalo in the Amazon. Community-based management has great potential to overcome some of these management obstacles. As with northern Brazil, the future of northern Australia is dependent on balancing the aims of development, conservation and indigenous wellbeing (Stephens et al. 2015). To address these (sometimes competing) aims,

decentralised management strategies that involve local communities and traditional ecological knowledge are necessary. Such strategies have the capacity to improve the ability to make natural resource management decisions, and also to develop strong community relationships, resolve conflicts and improve livelihoods. Community-based management strategies have had variable success within Australia and elsewhere. Success of these strategies appears to be linked with strong attachment to place, initiation by the communities themselves, well-defined goals and the adaptive capacity of these programs (Armitage 2005; Measham & Lumbasi 2013). The need for environmental management based on research is being increasingly recognised (Laurance et al. 2012). A fundamental difficulty for management in remote regions is a lack of historical data and research compared with more developed temperate regions (Meijaard & Sheil 2007). In many cases, this has translated to poorly informed management decisions. Improved management is therefore dependent on the application of continued research and monitoring. The natural values, vulnerability of indigenous communities, wealth of traditional ecological knowledge and multiple-stakeholder 81

status of environmental issues make northern Australia and northern Brazil good candidates for community-based management programs (Yibarbuk et al. 2001; Altman & Whitehead 2003; Schwartzman & Zimmerman 2005; Chacin 2010; Stoeckl et al. 2013). The wetlands (Erwin 2008), savannas (Grace et al. 2006) and rainforests (Boscolo et al. 1997) of northern Australia and Brazil have the capacity to buffer or slow climate change through carbon sequestration. Therefore, strategies for management and conservation of these ecosystems are important on a global scale. One such strategy in northern Australia, with multiple benefits, is the West Arnhem Land Fire Abatement (WALFA) program, in operation since 2005. This emissions offset program, developed over years of capacity building and research, combines traditional and contemporary fire management practices

to reduce greenhouse gas emissions associated with wildfires, whilst also conserving biodiversity and creating culturally appropriate employment (Russell-Smith et al. 2013). The project, which operates over remote and rugged indigenous lands, reduced mean annual emissions by 37.7% in the first seven years, whilst promoting landscape heterogeneity and savanna biodiversity through mosaic burning (Russell-Smith et al. 2013). In the process, the project has provided employment in aboriginal communities and helped to maintain and strengthen cultural connections with the landscape. In contrast, the Brazilian Amazon emissions offset program is in its infancy. With the Brazilian government’s goal to reduce Amazon deforestation by 80% by 2020, most Amazon states have launched a new line of carbon credits which have already attracted some investment (Nepstad et al. 2014).

Did you know? A community-based program was implemented in the Northern Territory to eliminate the dengue/zika virus mosquito (Aedes aegypti). This program required access to every household in small communities several times over the course of two years, while also promoting awareness around the community to eliminate potential mosquito breeding receptacles. In 2008, the program was declared a success with A. aegypti eradicated from Groote Eylandt (Whelan et al. 2009).

Management and conservation of crocodilians Like many crocodilian species worldwide, the Black Caiman (Melanosuchus niger) of Brazil and Saltwater Crocodile (Crocodylus porosus) of Australia (Fig. 7.6) have a long history of exploitation for their skins and meat. High intensity commercial harvest of Black Caiman in northern Brazil since the 1930s resulted in severe declines and significant range contraction (Plotkin et al. 1983). Despite being officially protected in Brazil since 1967, illegal harvest continues and is considered one of the largest illegal wildlife industries in the world (Mendonça et al. 2016). In the early

1980s, Plotkin et al. (1983) estimated that existing Black Caiman populations represented just 1% of historical populations. There is evidence that some wild populations have recovered substantially (Da Silveira & Thorbjarnarson 1999), and they are now listed as Least Concern on the IUCN Red List. However, there are concerns about how continued hunting will affect these populations, and there is interest in developing sustainable and economically viable harvest strategies (Da Silveira & Thorbjarnarson 1999; Thorbjarnarson 2010). In comparison, the Saltwater Crocodile was nearing extinction in Australia in 1971 82

following years of intense commercial hunting, prompting their protection (Fukuda et al. 2011). Protection of crocodiles was highly successful, monitoring efforts showing that the population recovered comparatively quickly. Despite an egg-harvest program being established in 1983, and the removal of CITES (Convention on International Trade in Endangered Species) restrictions and conditions in 1994, the crocodile population has continued to grow. Saltwater crocodiles are now considered an economically important natural resource in the Northern Territory

through crocodile farming for skins and meat, and tourism. However, the greater number of crocodiles has stimulated increasing concerns for public safety, and there is a need for management strategies that balance conservation, economic development and public safety goals (Leach et al. 2009). These two parallel management scenarios illustrate the complexities of balancing economics with conservation. Sustainable harvest of crocodilians is possible when appropriate management and population monitoring are implemented.

Figure 7.6. The crocodilian connection. (a) Black Caiman (Melanosuchus niger) of Brazil and (b) Saltwater Crocodile (Crocodylus porosus) of Australia. Photos: H. Hunter-Xenie and M. Kelly. Overview Northern Brazil and northern Australia are ecologically important regions, supporting exceptional biodiversity and endemic species, sequestering large amounts of carbon and providing abundant natural resources. Although relatively unspoiled compared to most of the world, these regions are faced with increasing threats as the pressure for development intensifies and climate change continues. Their relatively intact nature, however, also provides a unique opportunity to improve knowledge of the ecological function of tropical environments. In order to manage these threats and impending challenges for sustainable development, it is important that these two regions share and

learn from their experiences and develop management programs based on sound scientific research produced. Shared experiences and management programs could also strengthen the capacity of local populations to protect and benefit from the land to ensure a sustainable existence. The parallels and exchange of ecological knowledge will benefit research and management between these regions and elsewhere.

83

Literature Cited Altman JC, Whitehead PJ. 2003. Caring for Country and Sustainable Indigenous Development: Opportunities, Constraints and Innovation. CAEPR Working Paper No. 20/2003. Centre for Aboriginal Economic Policy Research, Australian National University, Canberra. Armitage D. 2005. Adaptive capacity and community-based natural resource management. Environmental Management 35: 703-715. Balch JK, Nepstad DC, Curran LM. 2009. Pattern and process: fire-initiated grass invasion at Amazon transitional forest edges. Pp. 481-502 in: Cochrane MA, editor. Tropical Fire Ecology: Climate Change, Land Use and Ecosystem Dynamics. Springer Praxis Books, Heidelberg. Barreto ML. 2001. Mineração e desenvolvimento sustentável: desafios para o Brasil. Centro de Tecnologia Mineral (CETEM), Rio de Janeiro. Boscolo M, Buongiorno J, Panayotou T. 1997. Simulating options for carbon sequestration through improved management of a lowland tropical rainforest. Environment and Development Economics 2: 241-263. Bowman DM, Robinson CJ. 2002. The getting of the Nganabbarru: observations and reflections on Aboriginal buffalo hunting in northern Australia. Australian Geographer 33: 191-206. Bradshaw CJA. 2012. Little left to lose: deforestation and forest degradation in Australia since European colonization. Journal of Plant Ecology 5: 109-120. Braithwaite RW. 1990. Australia's unique biota: Implications for ecological processes. Journal of Biogeography 17: 347-354. Castello L, Stewart D, Arantes C. 2011. Modeling population dynamics and conservation of arapaima in the Amazon. Reviews in Fish Biology and Fisheries 21: 623-640. Chacín CL. 2010. Towards sustainable harvest of sideneck river turtles (Podocnemis spp.) in the middle Orinoco, Venezuela. Doctoral dissertation. Duke University, Durham. Commonwealth of Australia. 2017. Indigenous Protected Areas. http://www.environment.gov.au/land/indigenous-protected-areas (accessed May 2017) Cook K, Vizy E. 2008. Effects of twenty-first-century climate change on the Amazon rain forest. Journal of Climate 21: 542-561. Cowie ID, Short PS, Madsen MO. 2000. Floodplain Flora. Parks and Wildlife Commission of the Northern Territory, Darwin. Da Silveira R, Thorbjarnarson JB. 1999. Conservation implications of commercial hunting of black and spectacled caiman in the Mamirauá Sustainable Development Reserve, Brazil. Biological Conservation 88: 103-109. De Paula LF, de Melo MA, Pires MCC. 2016. The Brazilian economy after the global crisis: An assessment of the economic slowdown. 2011–2012. Pp. 43-73 in: Pereira AW, Mattei L, editors. The Brazilian Economy Today. Palgrave Macmillan, London. Davidson EA, de Araújo AC, Artaxo P, Balch JK, Brown IF, Bustamante MM, Coe MT, DeFries RS, Keller M, Longo M. 2012. The Amazon basin in transition. Nature 481: 321328. Doody JS, Georges A, Young JE. 2003. Twice every second year: reproduction in the pig‐nosed turtle, Carettochelys insculpta, in the wet–dry tropics of Australia. Journal of Zoology 259: 179-188. Doody JS, Georges A, Young JE. 2004. Determinants of reproductive success and offspring sex in a turtle with environmental sex determination. Biological Journal of the Linnean Society 81: 1–16.

84

Doody JS, Stewart B, Camacho C, Christian K. 2012. Good vibrations? Sibling embryos expedite hatching in a turtle. Animal Behaviour 83: 645-651. Eisemberg CC. 2006. Aspectos da Biologia Reprodutiva de Rhinemys rufipes (Spix, 1824) (Chelidae, Testudines) na Reserva Florestal Adolpho Ducke, Amazonas, Brasil. Masters Thesis. National Institute for Amazon Research – INPA, Manaus. Eisemberg CC, Rose M, Yaru B, Amepou Y, Georges A. 2015. Salinity of the coastal nesting environment and its association with body size in the estuarine pig‐nosed turtle. Journal of Zoology 295: 65-74. Eisemberg CC, Balestra RAM, Famelli S, Pereira FF, Bernardes DVC, Vogt, RC. 2016. Vulnerability of Giant South American Turtle (Podocnemis expansa) nesting habitat to climate-change-induced alterations to fluvial cycles. Tropical Conservation Science 9: 1940082916667139. Erwin KL. 2008. Wetlands and global climate change: the role of wetland restoration in a changing world. Wetlands Ecology and Management 17: 71-84. Fearnside PM. 2005. Deforestation in Brazilian Amazonia: history, rates, and consequences. Conservation Biology 19: 680-688. Fearnside PM. 2016. Environmental and social impacts of hydroelectric dams in Brazilian Amazonia: Implications for the aluminum industry. World Development 77: 48-65. Ferreira J, Aragão L, Barlow J, Barreto P, Berenguer E, Bustamante M, Gardner T, Lees A, Lima A, Louzada J. 2014. Brazil's environmental leadership at risk. Science 346: 706-707. Finlayson CM. 2005. Plant ecology of Australia's tropical floodplain wetlands: A review. Annals of Botany 96: 541-555. Fox ID, Neldner VJ, Wilson GW, Bannink PJ. 2001. The Vegetation of the Australian Tropical Savannas. Environmental Protection Agency, Brisbane. Franklin DC, Petty AM, Williamson GJ, Brook BW, Bowman DMJS. 2008. Monitoring contrasting land management in the savanna landscapes of northern Australia. Environmental Management 41: 501-515. Franklin DC, Preece ND. 2014. The Eucalypts of Northern Australia: An Assessment of the Conservation Status of Taxa and Communities. A report to Kimberley to Cape and the Environment Centre NT. Environment Centre NT, Darwin. Fukuda Y, Webb G, Manolis C, Delaney R, Letnic M, Lindner G, Whitehead P. 2011. Recovery of saltwater crocodiles following unregulated hunting in tidal rivers of the Northern Territory, Australia. The Journal of Wildlife Management 75: 1253-1266. Grace J, José JS, Meir P, Miranda HS, Montes RA. 2006. Productivity and carbon fluxes of tropical savannas. Journal of Biogeography 33: 387-400. Grigg GC, Johansen K, Harlo P, Beard LA, Taplin LE. 1986. Facultative aestivation in a tropical freshwater turtle Chelodina rugosa. Comparative Biochemistry and Physiology Part A: Physiology 83: 321-323. Hancock D. 2014. Indigenous Protected Areas, available from http://www.australiangeographic.com.au/topics/history-culture/2014/06/indigenousprotected-areas (accessed February 2017). Hart B. 2004. Environmental risks associated with new irrigation schemes in northern Australia. Ecological Management and Restoration 5: 106-111. Hill R, Baird A, Buchanan D. 1999. Aborigines and fire in the Wet Tropics of Queensland, Australia: ecosystem management across cultures. Society and Natural Resources 12: 205223. Hughes L. 2003. Climate change and Australia: Trends, projections and impacts. Austral Ecology 28: 423-443. IPCC. 2014. Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the fifth assessment Report of the Intergovernmental Panel on Climate Change. Geneva. 85

Junk W, Piedade MTF. 1993. Herbaceous plants of the Amazon floodplain near Manaus: Species diversity and adaptations to the flood pulse. Amazoniana. Kiel 12: 467-484. Junk W, An S, Finlayson C, Gopal B, Kvĕt J, Mitchell S, Mitsch W, Robarts R. 2013. Current state of knowledge regarding the world's wetlands and their future under global climate change: a synthesis. Aquatic Sciences 75: 151-167. Karim F, Dutta D, Marvanek S, Petheram C, Ticehurst C, Lerat J, Kim S, Yang A. 2015. Assessing the impacts of climate change and dams on floodplain inundation and wetland connectivity in the wet–dry tropics of northern Australia. Journal of Hydrology 522: 8094. Kennett R. 1999. Reproduction of two species of freshwater turtle, Chelodina rugosa and Elseya dentata, from the wet–dry tropics of northern Australia. Journal of Zoology 247: 457-473. Kennett R, Christian K. 1994. Metabolic depression in estivating long-neck turtles Chelodina rugosa. Physiological Zoology 67: 1087-1102. Kennett R, Christian K, Pritchard D. 1993. Underwater nesting by the tropical freshwater turtle, Chelodina rugosa (Testudinata, Chelidae). Australian Journal of Zoology 41: 47-52. Kirby KR, Laurance WF, Albernaz AK, Schroth G, Fearnside PM, Bergen S, Venticinque EM, da Costa C. 2006. The future of deforestation in the Brazilian Amazon. Futures 38: 432453. Kutt, A, Felderhof L, VanDerWal J, Stone P, Perkins G. 2009. Terrestrial Ecosystems of Northern Australia. Northern Australia Land and Water Taskforce Full Report. CSIRO Sustainable Ecosystems, Canberra. Laurance WF. 1999. Reflections on the tropical deforestation crisis. Biological Conservation 91: 109-117. Laurance WF, Cochrane MA, Bergen S, Fearnside PM, Delamonica P, Barber C, D'Angelo S, Fernandes T. 2001. The future of the Brazilian Amazon. Science 291: 438-439. Laurance WF, Koster H, Grooten M, Anderson AB, Zuidema PA, Zwick S, Zagt RJ, Lynam AJ, Linkie M, Anten NPR. 2012. Making conservation research more relevant for conservation practitioners. Biological Conservation 153: 164-168. Leach G, Delaney R, Fukuda Y 2009. Management Program for the Saltwater Crocodile in the Northern Territory of Australia, 2009-2014. Department of Natural Resources, Environment, the Arts and Sport, Darwin. Lehmann CER, et al. 2014. Savanna vegetation-fire-climate relationships differ among continents. Science 343: 548-552. Malingreau J-P, Tucker CJ. 1988. Large-scale deforestation in the southeastern Amazon Basin of Brazil. Ambio 17: 49-55. Martin RJ. 2013. Sometime a fire: Re-imagining elemental conflict in northern Australia's Gulf country. Australian Humanities Review 55: 67-91. Measham TG, Lumbasi JA. 2013. success factors for community-based natural resource management (CBNRM): Lessons from Kenya and Australia. Environmental Management 52: 649-659. Meijaard E, Sheil D. 2007. Is wildlife research useful for wildlife conservation in the tropics? A review for Borneo with global implications. Biodiversity and Conservation 16: 30533065. Mendonça WCDS, Marioni B, Thorbjarnarson JB, Magnusson WE, Da Silveira R. 2016. Caiman hunting in Central Amazonia, Brazil. The Journal of Wildlife Management 80: 1497-1502. Mistry J, Berardi A, Andrade V, Krahô T, Krahô P, Leonardos O. 2005. Indigenous fire management in the cerrado of Brazil: The case of the Krahô of Tocantíns. Human Ecology 33: 365-386. Moran EF. 1993. Deforestation and land use in the Brazilian Amazon. Human Ecology 21: 1-21.

86

Mudd GM, Patterson J. 2010. Continuing pollution from the Rum Jungle U–Cu project: A critical evaluation of environmental monitoring and rehabilitation. Environmental Pollution 158: 1252-1260. Nepstad D, et al. 2014. Slowing Amazon deforestation through public policy and interventions in beef and soy supply chains. Science 344: 1118-1123. Parolin P, Wittmann F. 2010. Struggle in the flood: tree responses to flooding stress in four tropical floodplain systems. AoB PLANTS 2010: plq003-plq003. Pereira LA, Sousa AL, Cutrim MVJ, Moreira EG. 2007. Características ecológicas do habitat de Kinosternon scorpioides Linnaeus, 1766 (Reptilia, Chelonia, Kinosternidae) no município de São Bento-Baixada Maranhense (Maranhão, Brasil). Boletim do Laboratório de Hidrobiologia 20: 9-14. PIB. 2014. Localização e extensão das Terras Indígenas. Povos Indígenas no Brasil, available from https://pib.socioambiental.org/pt/c/terras-indigenas/demarcacoes/localizacao-eextensao-das-tis (accessed May 2017). Pivello VR. 2011. The use of fire in the Cerrado and Amazonian rainforests of Brazil: past and present. Fire Ecology 7: 24-39. Plotkin MJ, Medem F, Mittermeier RA, Constable ID. 1983. Distribution and conservation of the black caiman (Melanosuchus niger). Pp. 695-705 in: Rhodin A, Mitaya K. editors. Advances in Herpetology and Evolutionary Biology. Museum of Comparative Zoology, Cambridge. PROBIO. 2004. Uso e cobertura da terra na floresta Amazônica. Fundação de Ciência, Aplicações e Tecnologia Espaciais – FUNCATE, São José dos Campos. RAMSAR. 2017. Ramsar sites around the world, available from http://www.ramsar.org/sitescountries/ramsar-sites-around-the-world (accessed May 2017). Robinson CJ, Whitehead P. 2003. Cross-cultural management of pest animal damage: a case study of feral buffalo control in Australia’s Kakadu National Park. Environmental Management 32: 445-458. Russell-Smith J, Cook GD, Cooke PM, Edwards AC, Lendrum M, Meyer CP, Whitehead PJ. 2013. Managing fire regimes in north Australian savannas: applying Aboriginal approaches to contemporary global problems. Frontiers in Ecology and the Environment 11: e55-e63. Russell-Smith J, Gardener MR, Brock C, Brennan K, Yates CP, Grace B. 2012. Fire persistence traits can be used to predict vegetation response to changing fire regimes at expansive landscape scales – an Australian example. Journal of Biogeography 39: 1657-1668. Russell-Smith J, Yates C, Edwards A, Allan GE, Cook GD, Cooke P, Craig R, Heath B, Smith R. 2003. Contemporary fire regimes of northern Australia, 1997–2001: change since Aboriginal occupancy, challenges for sustainable management. International Journal of Wildland Fire 12: 283-297. Schwartzman S, Zimmerman B. 2005. Conservation alliances with indigenous peoples of the Amazon. Conservation Biology 19: 721-727. Seymour RS, Kennett R, Christian K. 1997. Osmotic balance in the eggs of the turtle Chelodina rugosa during developmental arrest under water. Physiological Zoology 70: 301-306. Sheikh PA, Merry FD, McGrath DG. 2006. Water buffalo and cattle ranching in the lower Amazon Basin: Comparisons and conflicts. Agricultural Systems 87: 313-330. Shine R, Brown GP. 2008. Adapting to the unpredictable: reproductive biology of vertebrates in the Australian wet-dry tropics. Philosophical Transactions: Biological Sciences 363: 363373. Silvério DV, Brando PM, Balch JK, Putz FE, Nepstad DC, Oliveira-Santos C, Bustamante MM. 2013. Testing the Amazon savannization hypothesis: fire effects on invasion of a neotropical forest by native cerrado and exotic pasture grasses. Philosophical Transactions of the Royal Society of London B: Biological Sciences 368: 20120427. 87

Simmons CS. 2004. The political economy of land conflict in the eastern Brazilian Amazon. Annals of the Association of American Geographers 94: 183-206. Simon MF, Pennington T. 2012. Evidence for adaptation to fire regimes in the tropical savannas of the Brazilian Cerrado. International Journal of Plant Sciences 173: 711-723. Soares-Filho B, et al. 2010. Role of Brazilian Amazon protected areas in climate change mitigation. Proceedings of the National Academy of Sciences 107: 10821-10826. Stephens A, Oppermann E, Turnour J, Brewer T, O'Brien C. 2015. Identifying tensions in the development of northern Australia: Implications for governance. Journal of Economic and Social Policy 17: 1-21. Stewart‐Koster B, Olden J, Kennard MJ, Pusey BJ, Boone E, Douglas M, Jackson S. 2011. Fish response to the temporal hierarchy of the natural flow regime in the Daly River, northern Australia. Journal of Fish Biology 79: 1525-1544. Stoeckl N, Jackson S, Pantus F, Finn M, Kennard MJ, Pusey BJ. 2013. An integrated assessment of financial, hydrological, ecological and social impacts of ‘development’ on Indigenous and non-Indigenous people in northern Australia. Biological Conservation 159: 214-221. Taylor JA, Tulloch D. 1985. Rainfall in the wet-dry tropics: extreme events at Darwin and similarities between years during the period 1870-1983 inclusive. Australian Journal of Ecology 10: 281-295. Thorbjarnarson JB. 2010. Black caiman Melanosuchus niger. Pp. 29-39 in: Manolis SC, Stevenson C. editors. Crocodiles. Status Survey and Conservation Action Plan. Crocodile Specialist Group, Darwin. Turner IM. 2001. The Ecology of Trees in the Tropical Rain Forest. Cambridge University Press, Cambridge. Warfe DM, et al. 2011. The ‘wet–dry’ in the wet–dry tropics drives river ecosystem structure and processes in northern Australia. Freshwater Biology 56: 2169-2195. Whelan PI, Kulbac M, Bowbridge D, Krause V. 2009. The eradication of Aedes aegypti from Groote Eylandt NT Australia 2006-2008. Arbovirus Research in Australia 10: 188-199. Whitehead P, Russell-Smith J, Cooke P 2009. Culture, Ecology and Economy of Fire Management in North Australian Savannas - Rekindling the Wurrk Tradition. CSIRO, Collingwood. Williams RJ, Cook GD, Gill AM, Moore PHR. 1999. Fire regime, fire intensity and tree survival in a tropical savanna in Northern Australia. Austral Ecology 24: 50-59. Williams SE, Bolitho EE, Fox S. 2003. Climate change in Australian tropical rainforests: An impending environmental catastrophe. Proceedings: Biological Sciences 270: 1887-1892. Woinarski J, Whitehead PJ, Bowman DMJS, Russell-Smith J. 1992. Conservation of mobile species in a variable environment: The problem of reserve design in the Northern Territory, Australia. Global Ecology and Biogeography Letters 2: 1-10. Woinarski J, Mackey B, Nix H, Traill B 2007. The Nature of Northern Australia: Natural Values, Ecological Processes and Future Prospects. ANU Press, Canberra. Woinarski J, Green J, Fisher A, Ensbey M, Mackey B. 2013. The effectiveness of conservation reserves: Land tenure impacts upon biodiversity across extensive natural landscapes in the tropical savannahs of the Northern Territory, Australia. Land 2: 20-36. Yibarbuk D, Whitehead PJ, Russell-Smith J, Jackson D, Godjuwa C, Fisher A, Cooke P, Choquenot D, Bowman DMJS. 2001. Fire ecology and Aboriginal land management in central Arnhem Land, northern Australia: a tradition of ecosystem management. Journal of Biogeography 28: 325-343.

88

Carettochelys insculpta hatchlings Illustration: Fernando A. Perini

89