Smuggling snakes: a study on the global trade of green pythons (Morelia viridis)

Transcription

1 Smuggling snakes: a study on the global trade of green pythons (Morelia viridis) Jessica A. Lyons A thesis submitted for the degree of Master of Science University of New South Wales School of Biological, Earth and Environmental Sciences Facaulty of Science August 2012

2 UNIVERSITY OF NEW SOUTH WALES THESIS/DISSERTATION SHEET Surname or Family name: Lyons First name: Jessica Other name/s: Ann Abbreviation for degree as given in the University calendar: MSc School: Biological, Earth and Environmental Sciences Faculty: Science Title: Smuggling snakes: a study on the global trade of green pythons (Morelia viridis) ABSTRACT The wildlife trade is a multi-billion dollar industry that operates on local, national and international levels. In recent years the trade in wildlife for pets has increased considerably and demands species that are brightly coloured, attractively marked and rare. Green pythons (Morelia viridis) are one of the most highly sought after reptiles in the pet trade. Indonesia is the only country that permits the commercial export of green pythons, but all individuals must be the progeny of captive-bred stock. There are, however, a number of reports that suggest that many of the green pythons exported from Indonesia are in fact illegally harvested from the wild and laundered through wildlife breeding farms to be exported as captive-bred. I surveyed reptile collectors, traders, middlemen and farm owners in Indonesia, and examined the harvest demographic of green pythons collected for trade to determine the extent of illegal trade and effects of harvest on this species. In total, 4227 illegally harvested wild green pythons were recorded during surveys and high levels of harvest were found to have depleted and skewed the demographics of some island populations. Wild green pythons were traced from their point of capture to wildlife breeding farms in Jakarta where they are to be exported for the pet trade, confirming reports of laundering. Extrapolation of monthly collection estimates provided by wildlife traders revealed that at least 5337 green pythons are collected each year, suggesting that at least 80% of the green pythons exported from Indonesia annually are illegally harvested from the wild. The results of examination of 139 eggshells from five python species suggest that reptilian eggshell morphology may be used as proof of provenance for each individual reptile exported. Advertisements selling green pythons and consumer questionnaire surveys were used to test the Anthropogenic Allee Effect (AAE), a concept that shows rare species are disproportionally valued (and exploited) by humans, which reduces their population size and increases their value. I propose that an AAE may not only occur between species, but between populations within a species. Consumers were found to place disproportionate value on rare populations of green pythons. Populations that exhibited a high relative frequency of abnormal colouration, and were relatively small (such as those on islands), were found to be suffering from the effects of over harvesting for the pet trade. The threats faced by green pythons are discussed and recommendation and directions for future research are offered to mitigate illegal harvests and ensure the survival of this species, and its populations, into the future. DECLARATION RELATING TO DISPOSITION OF PROJECT THESIS/DISSERTATION I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis/dissertation in whole or in part in the University libraries in all forms of media, now or hereafter know, subject to the provision of the Copyright Act I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis/dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only). 30/08/2012 SIGNATURE WITNESS DATE The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restrictions for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research. FOR OFFICE USE ONLY DATE OF COMPLETION OF REQUIREMENTS FOR AWARD:

3

4 ORIGINALITY STATEMENT I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged. Parts of this thesis have been published as articles in peer-reviewed scientific journals (as listed in Chapter six). Jessica Ann Lyons 30 th August 2012 i

5 COPYRIGHT STATEMENT I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350-word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). Where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation. I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format. Jessica Ann Lyons 30 th August 2012 ii

6 AUTHENTICITY STATEMENT I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format. Jessica Ann Lyons 30 th August 2012 iii

7 ACKNOWLEDGMENTS This study would not have been possible without the inspiration, assistance and support of Daniel Natusch. Thanks to my supervisor, Michael Archer, and co-supervisor, Sue Hand, for backing this study and providing valuable comments. Thanks to Chris Shepherd, Vincent Nijman and David Natusch for providing insightful discussions, and to Barry Brook and Grahame Webb for their reviews that improved this thesis. Thanks to Ben Osborne and Lachlan McIntyre for their assistance in the field. Thank you to Vladimir Odinchenko and Yuri Kukin for allowing access to python eggshells, generously providing their unpublished data on green python reproduction and for their kindness. Большое спасибо друзья. Thanks to my family for their constant guidance, support and love. I can never thank you enough for all you have sacrificed and the opportunities you have provided me. Nothing is impossible. I would like to thank the many people who assisted me in the field. Despite the sensitive nature of this study, never were people in Indonesia unfriendly or unwilling to assist me. Although the Indonesian provinces of West Papua and Papua are resource rich, the people residing there are among the nation s poorest. I often found that those who had little to give were the most generous. Terima kasih banyak. Thank you also to the large number of green python keepers and breeders around the world that have shared information about the trade in green pythons. I also want to thank you, the reader, for having an interest in, and reading about my study. This study was conducted in accordance with the University of New South Wales Human Research Ethics protocol (Permit Number 08/2011/33) and Animal Research Ethics protocol (Permit Number 10/90A). iv

8 TABLE OF CONTENTS ORIGINALITY STATEMENT i COPYRIGHT STATEMENT ii AUTHENTICITY STATEMENT iii ACKNOWLEDGEMENTS iv TABLE OF CONTENTS v LIST OF FIGUES viii LIST OF TABLES xi LIST OF ACRONYMS AND ABBREVIATIONS xii ABSTRACT xiii ORGANISATION OF THESIS xiv CHAPTER ONE: Introduction The wildlife trade Convention on the International Trade in Endangered Species of Wild Flora and Fauna (CITES) History and purpose The pet trade The green python (Morelia viridis) Conservation status and relevant legislation International National Indonesia Australia Papua New Guinea Aims Study region Study sites 15 CHAPTER TWO: The illegal trade of green pythons in Indonesia Materials and methods Provincial trader identification and interviews Examination and morphometrics Results Trade dynamics 20 v

9 Trade chain Harvest levels Variation in harvest demographic among localities Final destination Discussion Effects of illegal harvest on wild populations of green pythons The threat of breeding farms 32 CHAPTER THREE: Effects of rarity and consumer preferences on the global trade of green pythons Materials and methods Harvest data from Indonesia Advertisements Survey questionnaire Analysis of data Results Provincial traders in Indonesia Survey responses and advertisements Rarity, demand, harvest, and price Consumer preferences and attitudes toward green pythons Discussion Rarity and consumer preferences Consumer preferences and attitudes 46 CHAPTER FOUR: What comes first, the snake or the egg? A method for regulating the export of green pythons Materials and methods Results Discussion Potential loopholes Using the eggshells of other species The transfer of eggshells 54 CHAPTER FIVE: Threats, recommendations, future research directions, and conclusion Threats 57 vi

10 The Anthropogenic Allee Effect (AAE) Human-induced habitat degradation and loss Logging and mining Agriculture Roads Harvest and trade Subsistence Commercial Recommendations Acknowledging there is a problem Educating the consumer market Proving provenance of green pythons with the eggshell method Reducing corruption within conservation Enforcing current legislation Adequate monitoring and regulation Determination of economic viability Husbandry training Allowing legal harvest Conclusions 66 CHAPTER SIX: Resulting publications and editorials Publications Editorials Pernetta (2012): Effective and sustainable farming of green pythons requires a sound chain of custody and conservation taxation of end consumers Lyons and Natusch (2012): Consumer driven conservation of green pythons is possible if the price is right: A reply to Pernetta (2012) 72 LITERATURE CITED 76 APPENDICES 93 vii

11 LIST OF FIGURES Fig Fig Fig Fig Fig The distribution of the green python (Morelia viridis) (in red). The island of New Guinea is divided politically between Papua New Guinea (eastern half of New Guinea) and Indonesia (western half of New Guinea, represented by the provinces of West Papua and Papua). The Aru Islands are biogeographically part of the island of New Guinea, but politically part of Indonesia s Maluku province (Allison 2007). Green pythons have been found in suitable habitat throughout the island, however, their specific distribution, particularly in the highlands, is unknown. The dashed line represents the approximate border between the Indonesian provinces of West Papua (left) and Papua (right). 8 The three colour morphs of the green python: (a) red juvenile, (b) yellow juvenile, (c) green adult, and (d) an example of a designer green python. Note: the four images are not to scale. Rico Walder provided the image of a designer green python. 10 Twenty-eight study sites visited in Indonesia ( ), and the five corresponding localities from which green pythons are sourced (enclosed in boxes). Note: The Aru Islands are biogeographically part of the island of New Guinea, but politically part of Maluku province (Allison 2007). 16 Three trade chain routes for green pythons collected in New Guinea to reach the consumer market. The dashed line indicates where wildlife is sent from New Guinea by plane to Jakarta, and therefore where targeted enforcement can take place. The trade chain does not end at the consumer; green pythons can be captive-bred in consumer countries and sold to other consumers. 23 Numbers of green pythons collected from Biak between August 2009 and April Black columns represent the numbers counted in this survey and white columns represent the number recorded by a single provincial trader. Missing columns indicate months when viii

12 Fig Fig Fig Fig Fig Fig Fig Fig no data were gathered. 25 The percentage of green pythons that were red juveniles ( ), yellow juveniles ( ) and green adults ( ), collected from the Aru Islands, Merauke, Vogelkop and Raja Ampat, and Jayapura. 26 The percentage of green pythons that were yellow juveniles, red juveniles and green adults, collected from Biak in 2009 (black columns) and 2011 (white columns). Sample sizes are presented above each column. 26 Green pythons found in the possession of provincial traders: (a), (b), and (c) present with serious health issues, while (d), still coiled on its perch was dead. 28 Mass vs. SVL of green pythons from the Aru Islands measured the day of collection ( ), compared to individuals recorded further along the trade chain ( ). 29 Four reptile species exported from Indonesia in the highest numbers as captive-bred between the year 2000 and 2009 (CITES Trade Database, 2011). Morelia viridis ( ), Varanus timorensis ( ), Varanus prasinus ( ), Varanus indicus ( ). 32 Relationships between trade variables for the green python. (a) Mean commonness of green pythons from six localities as ranked by respondents vs. mean number of green pythons that provincial traders claimed to have harvested per month from the same localities. (b) Mean commoness of green pythons from eight localities as ranked by respondents vs. mean price of green pythons from the same localities. (c) Mean price of green pythons from seven localities vs. mean number of green pythons that provincial traders claimed to have harvested per month from the same localities. North America ( ), Europe ( ), and Asia ( ). 42 Percentage of responses given for attributes looked for in a quality green python by respondents from five regions. Note: Percetages may sum to more than one hundred because respondents were allowed to select more than one attribute. 43 Mean price that provincial traders in New Guinea and respondents ix

13 Fig Fig in North America would pay for the different colours of green python. 44 Length vs. width of pythonid eggshells, showing the distinctive shape of Morelia viridis ( ), Morelia spilota ( ), Apodora papuana ( ), Morelia tracyae ( ), and Python curtus ( ). Error bars indicate the extremes of length and width for each species. 52 Used Pythonid eggshells showing the distinctive shape of Python curtus, Morelia spilota, Morelia tracyae, Morelia viridis and Apodora papuana. 53 x

14 LIST OF TABLES Table 2.1. Trade data for green pythons collected from five localities in Indonesia. N = number of provincial traders. 21 Table 3.1. Number of green pythons collected per month by provincial traders in Indonesia. 39 Table 3.2. Summary data for survey questionnaire and advertisements. 40 Table 4.1. Eggshell measurements of the five species of python. N = sample size, SD = standard deviation. 51 xi

15 LIST OF ACRONYMS AND ABBREVIATIONS AAE ASEAN-WEN BPS CI CITES DEC DEHP ha Rp IRATA IUCN LIPI MA MoF N NDF NGO OCC PHKA SA SVL TL TRAFFIC Undang Undang UNEP-WCMC USD WCO WWF Anthropogenic Allee Effect Association of Southeast Asian Nations Wildlife Enforcement Networks Badan Pusat Statistik [Indonesian Central Bureau of Statistics] Conservation International Convention on International Trade in Endangered Species of Wild Fauna and Flora Department of Environment and Conservation (Papua New Guinea) Department of Environment and Heritage Protection (was previously known as the Department of Environment and Resource Management, DERM) (Australia) Hectare Indonesian Rupiah Indonesian Reptile and Amphibian Trade Association International Union for Conservation of Nature Lembaga Ilmu Pengtahuan Indonesia [Indonesian Institute of Sciences, Indonesia] Management Authority Ministry of Forestry (Indonesia) Number of individuals Non-detriment finding Non-governmental organisation Ontogenetic colour change Perlindungan Hutan dan Konservasi Alam [Directorate General of Forest Protection and Nature Conservation under the Ministry of Forestry, Indonesia] Scientific Authority Snout-vent length Tail length Trade Records Analysis of Flora and Fauna in Commerce Law United Nations Environment Programme - World Conservation Monitoring Centre United States Dollar World Customs Organisation World Wildlife Fund xii

16 ABSTRACT The wildlife trade is a multi-billion dollar industry that operates on local, national and international levels. In recent years the trade in wildlife for pets has increased considerably and demands species that are brightly coloured, attractively marked and rare. Green pythons (Morelia viridis) are one of the most highly sought after reptiles in the pet trade. Indonesia is the only country that permits the commercial export of green pythons, but all individuals must be the progeny of captive-bred stock. There are, however, a number of reports that suggest that many of the green pythons exported from Indonesia are in fact illegally harvested from the wild and laundered through wildlife breeding farms to be exported as captive-bred. I surveyed reptile collectors, traders, middlemen and farm owners in Indonesia, and examined the harvest demographic of green pythons collected for trade, to determine the extent of illegal trade and effects of harvest on this species. In total, 4227 illegally harvested wild green pythons were recorded during surveys and high levels of harvest were found to have depleted and skewed the demographics of some island populations. Wild green pythons were traced from their point of capture to wildlife breeding farms in Jakarta where they are to be exported for the pet trade, confirming reports of laundering. Extrapolation of monthly collection estimates provided by wildlife traders revealed that at least 5337 green pythons are collected each year, suggesting that at least 80% of the green pythons exported from Indonesia annually are illegally harvested from the wild. The results of examination of 139 eggshells from five python species suggest that reptilian eggshell morphology may be used as proof of provenance for each individual reptile exported. Advertisements selling green pythons and consumer questionnaire surveys were used to test the Anthropogenic Allee Effect (AAE), a concept that shows rare species are disproportionally valued (and exploited) by humans, which reduces their population size and increases their value. I propose that an AAE may not only occur between species, but between populations within a species. Consumers were found to place disproportionate value on rare populations of green pythons. Populations that exhibited a high relative frequency of abnormal colouration, and were relatively small (such as those on islands), were found to be suffering from the effects of over harvesting for the pet trade. The threats faced by green pythons are discussed and recommendation and directions for future research are offered to mitigate illegal harvests and ensure the survival of this species, and its populations, into the future. xiii

17 ORGANISATION OF THESIS This thesis is arranged into six chapters. The first chapter provides an introduction to the wildlife trade and study species, the green python (Morelia viridis), and specifies the aims and significance of the study. Chapter two examines the trade of green pythons in Indonesia and the effect this trade has on wild populations of green pythons. Chapter three determines the effect rarity and consumer preferences have on the harvest and trade of green pythons. Chapter four proposes a simple, reliable and cost effective method for potentially regulating the laundering of wild reptiles through breeding farms and uses the green python as a case study species. Chapter five examines threats to green pythons in Indonesia and discusses the management implications and conclusions related to the findings of this study. Finally, Chapter six lists the publications, and presents the editorials, that have resulted from the study. xiv

18 CHAPTER ONE The green python (Morelia viridis). 1

19 Jessica A. Lyons Introduction 1.1. The wildlife trade The trade in wildlife, that is the sale and exchange by people of flora and fauna and their derivatives, is a complex and perpetually evolving industry that generates employment and income for rural communities, the business sector and national economies (Oldfield 2003). However, the trade in wildlife is also a major contributor to biodiversity loss and has been recognised as a key conservation concern (Grieser-Johns and Thomas 2005; Sutherland et al. 2009). Driven mainly by economics, wildlife is traded for use in medicines, luxury goods, food, and as pets (Nijman 2010). The trade chain is complex, involving numerous actors that span local, national and international levels (Broad et al. 2003). In 2005, the legal trade in wildlife was estimated to be worth USD 300 billion (Roe 2008); however, this does not represent the entirety of trade because very little is known about the extent of illegal trade (Rosen and Smith 2010). Nonetheless, illegal trade continues to flourish and is the second largest black market worldwide, after narcotics (Toledo et al. 2012). When wildlife is traded illegally, conservation efforts and sustainable harvests can be seriously undermined (Schoppe 2009; Zhou and Jiang 2005). Additionally, illegal trade can involve criminal organisations, which can foist violence and corruption on communities and impair efforts of developing nations to manage their natural resources (Warchol et al. 2003; Zimmerman 2003). Few studies have attempted, or been able to, determine the scale of illegal trade (Gavin et al. 2009), the effects of illegal harvests (Schoppe 2009; Smith et al. 2011) or the mechanisms by which the trade operates (but see Wutty and Simms 2005). As the demand for wildlife increases, additional strain is placed on wild populations, and unsustainable harvesting practices, such as overexploitation, may result in extensive biodiversity loss and ecosystem degradation (Broad et al. 2003; Lenzen et al. 2012; Roe 2008). 2

20 Introduction 1.2. Convention on the International Trade in Endangered Species of Wild Flora and Fauna (CITES) History and purpose The international trade in wildlife is regulated by CITES, a multilateral institution. CITES (also referred to as the Convention ) originated from the 1963 resolution of the World Conservation Union (now formally known as the International Union for the Conservation of Nature [IUCN]), as there was a need for an international agreement on the trade in wildlife. The CITES treaty entered into force on 1 July 1975 and is one of the most important, yet controversial, of the international conservation treaties (Hutton and Dickson 2000). The aim of CITES is to conserve biodiversity and ensure that wild species do not go biologically extinct through international trade. Currently, CITES regulates the international trade in more than 34,000 species and has reduced threats associated with over-harvest of imperilled species (Phelps et al. 2011). However, the Convention has no power to ensure that trade is sustainable at national and local levels (Abensperg-Traun 2009). One hundred and seventy-six countries are signatories to CITES, agreeing to ensure that international trade is not detrimental to the survival of wild species listed within the CITES Appendices. CITES-listed species are subject to varying degrees of regulation and protection from over-exploitation according to their listing in these Appendices (from Winjnsteker 2011): I. Appendix I species are threatened with extinction, for which trade must be subject to strict regulation and only authorised in exceptional circumstances. Both import and export permits are required for international trade. II. Appendix II species are not necessarily now threatened with extinction but may become so unless trade is strictly regulated. Appendix II also contains look-a alike' species, whose trade is controlled because of their similarity in appearance to other unregulated species. An export permit is required for international trade. 3

21 Jessica A. Lyons III. Appendix III species are subject to regulation within the jurisdiction of a Party (i.e. the signatory country) and for which the co-operation of other Parties is needed to control their trade. Each Party must designate one or more Management Authorities (MA) responsible for issuing permits, subject to the advice from one or more Scientific Authorities (SA) designated for that purpose. Another important role of SA is to conduct non-detriment findings (NDF) for Appendix II listed species to evaluate whether a proposed trade action will be detrimental to the survival of those species in the wild (CITES Article III and IV; see CITES Secretariat 2008). The SA will determine whether the export of a species should be limited, in order to maintain that species throughout its range at a level consistent with its role in the ecosystems in which it occurs and well above the level at which that species might become eligible for inclusion in Appendix I (CITES Article IV). The SA then advises the MA of suitable measures to be taken to limit the grant of export permits for that species. In the early nineties a checklist was devised to provide guidance to the SA when making a NDF. This checklist addresses important factors, such as population status and trends, distribution, threats, and harvest levels of wild populations (see Rosser and Haywood 2002). However, it appears that in some countries, appropriate NDF have not been undertaken for numerous species. This may be because it is very difficult to conduct thorough analyses on these factors, and often SA making a NDF are under-resourced, under-staffed and in some cases, non-existent or marginalised (Murphy 2006). CITES rarely evaluates the adequacy of studies made to inform a NDF, and such studies can be problematic and logistically difficult for various species found in trade (Natusch and Lyons 2012b). For example, biological and ecological information for many wild species in trade is extremely variable and in most cases deficient; illegal or undeclared trade can also make it difficult to quantify harvest and trade levels (see Jenkins 2009). Each Party is required to provide an annual trade report and a biennial implementation report, which is submitted to the CITES Secretariat. Annual reports contain information on permits granted, the countries with which trade occurred and details of CITES-listed species traded (Reeve 2006). This report also provides the basis for monitoring trade in 4

22 Introduction CITES-listed species and enables assessment of compliance with quotas by highlighting discrepancies between reported imports and exports. A biennial report contains information on legislative, regulatory and administrative measures taken to enforce CITES (Reeve 2006). The credibility of CITES is dependent on the quality of these reports and trade data, as this informs decisions on the conservation of species and amasses political will among Parties (Gehring and Ruffing 2008). However, the incentives are high for biased analyses and misreporting, as CITES relies primarily on Parties self-reporting (Courchamp et al. 2006; Phelps et al. 2010). Collection and quality control of such data is challenging (Tomas and Albert 2006), and it has been documented that there are large discrepancies between officially reported import and export figures and actual imports and export figures (Blundell and Mascia 2005; Wood et al. 2012). The CITES Trade Database, which currently holds over 10 million records of trade in wildlife, is free to access and managed by the United Nations Environmental Programme World Conservation Monitoring Centre (UNEP-WCMC) ( Data on the trade in wildlife is also supplied by inter-governmental organisations, such as Interpol, the Association of Southeast Asian Nations Wildlife Enforcement Networks (ASEAN-WEN), the World Customs Organisation (WCO) and the Lusaka Agreement Task Force (LATF), as well as non-governmental organisations (NGOs), such as Trade Records Analysis of Fauna and Flora in Commerce (TRAFFIC), World Wildlife Fund (WWF) and Conservation International (CI) The pet trade The growing trade of wildlife for pets, both legal and illegal, has received little attention from conservation scientists (but see Auliya 2003; Nijman and Shepherd 2007; Shepherd 2006; Yuwono 1998). In this trade, species in high demand are often brightly coloured, rare, difficult to breed in captivity, restricted to isolated geographical regions, and/or newly described species or subspecies (Auliya 2003; Stuart et al. 2006). 5

23 Jessica A. Lyons Although the pet trade has been a cause of overexploitation of some wild populations in the past (O Brien et al. 2003; Shepherd and Ibarrondo 2005; Stuart et al. 2006; Zhou and Jiang 2004), it also provides many benefits. For example, the pet trade has fostered an appreciation among the general public for wild species (Prokop and Tunnicliffe 2010) and provides an alternative means of income for people in rural communities. There is a growing demand for reptiles as pets, many of which are sourced from Southeast Asia, an area with high species richness and endemism (Sodhi et al. 2010a). Approximately CITES-listed reptiles are exported live from Southeast Asia annually (CITES 2012). One country in particular, Indonesia, is known for harvesting and exporting reptiles, both legally and illegally, for the international pet trade (Hoover 1998; Pernetta 2009a; Samedi and Iskandar 2000; Shepherd 2000; Shepherd 2006; Shepherd and Ibarrondo 2005; Soehartono and Mardiastuti 2002; Nijman and Shepherd 2007; Nijman and Shepherd 2009). In Indonesia, reptiles listed under Appendix II of CITES are allocated an annual harvest quota (i.e. a specific number of individuals that can be harvested from the wild), while others may only be exported if bred in captivity. Several Indonesian wildlife farms supposedly breed and export large numbers of reptiles that have no or very low harvest quotas. However, an increasing number of reports (Auliya 2003; Nijman and Shepherd 2009) suggest that many reptiles exported from Indonesia are in fact wild-caught and laundered through breeding farms under the guise of being captive-bred, and this is well known within the pet trade. Currently there is no easy method for differentiating between wild-caught and captive-bred reptiles destined for export The green python (Morelia viridis) The green python is a small (<2 m) arboreal snake species restricted to tropical rainforests in Cape York Peninsula, Australia, and the island of New Guinea, which is divided politically between the independent nations of Papua New Guinea (PNG) (eastern half of New Guinea) and Indonesia (western half of New Guinea, represented 6

24 Introduction by the provinces of West Papua and Papua) (Natusch and Natusch 2011; O Shea 1996; Fig. 1.1). The green python is also found in the Aru Islands, which is bio-geographically part of the island of New Guinea, although it is politically part of Maluku province, Indonesia (Allison 2007). On the island of New Guinea, green pythons are found in lowland and montane rainforests up to 2000 m above sea level, as well as in secondary re-growth areas (O'Shea 1996). In Australia, they are most frequently recorded in evergreen to semi-deciduous notophyll vine forest, and have not been located in woodlands, swamp, heath or grasslands (Natusch and Natusch 2011). The green python is a nocturnal, ambush predator that typically hunts close to the ground (<1 m above ground level). Green pythons are relatively generalist in their diet and display an ontogenetic shift in prey, with juveniles preying on small lizards and adults on small mammals (Natusch and Lyons 2012a). Little is known of the reproductive biology of green pythons in the wild, though in captivity, adult females produce an average of one clutch of eggs each year (Maxwell 2005). Clutch size for green pythons in captivity varies from eggs. The average clutch size of wild individuals is 14 eggs (Natusch pers. comm.). Populations of green pythons have been distinguished using mitochondrial DNA, suggesting that future genetic studies using SNPs or microsatellites may be useful for determining provenance (Rawlings and Donnellan 2003). 7

25 Jessica A. Lyons Fig The distribution of the green python (Morelia viridis) (in red). The island of New Guinea is divided politically between Papua New Guinea (eastern half of New Guinea) and Indonesia (western half of New Guinea, represented by the provinces of West Papua and Papua). The Aru Islands are biogeographically part of the island of New Guinea, but politically part of Indonesia s Maluku province (Allison 2007). Green pythons have been found in suitable habitat throughout the island, however, their specific distribution, particularly in the highlands, is unknown. The dashed line represents the approximate border between the Indonesian provinces of West Papua (left) and Papua (right). 8

26 Introduction Green pythons are highly recognisable due to their distinctive colouration and pattern, with juveniles born either a lemon yellow or maraschino red colour, and change to lime green at approximately 65 cm in length (Wilson et al. 2007; Natusch and Lyons 2012a; Fig a,b,c). This phenomenon is known as ontogenetic colour change (OCC), and can occur over a period as short as four days or as long as several years (Barker and Barker 1994). It is this unique and striking colouration that has captured people s imaginations and made this species highly prized in the pet trade. Throughout their range, green pythons exhibit subtly different locality specific colours and patterns. As a result, individuals from certain localities are more desirable than others, and can demand higher prices (Maxwell 2005). In addition, green pythons that display atypical variations in colour also generally obtain a higher price, and are often referred to as designer snakes (Fig d). 9

27 Jessica A. Lyons Fig The three colour morphs of the green python: (a) red juvenile, (b) yellow juvenile, (c) green adult, and (d) an example of a designer green python. Note: the four images are not to scale. Rico Walder provided the image of a designer green python. 10

28 Introduction 1.5. Conservation status and relevant legislation International All but one species of python (Python molurus molurus, Appendix I) has been listed in Appendix II of CITES since This listing restricts their international trade by requiring the prior grant and presentation of an export permit (from the national MA) to export any individuals. All countries where green pythons are distributed (that is Australia, PNG and Indonesia) are Party to CITES and therefore legally bound to uphold the Convention. In 2009, the green python was evaluated by the IUCN for the Red List of Threatened Animals and classified as Least Concern. This list is widely recognised as the most comprehensive, objective global approach for evaluating the conservation status of flora and fauna. This status was justified owing to its large distribution (Auliya et al. 2009) National Indonesia Indonesia became a Party to CITES in 1979 and the trade of green pythons is monitored by the Indonesian CITES Management Authority, Perlindungan Hutan dan Konservasi Alam (the Directorate General of Forest Protection and Nature Conservation [PHKA]). The green python is fully protected under Government Regulation No. 7/1999 (Dilindungi Peraturan Pemerintah Nomor 7 Tahun 1999), which derives from Conservation Act No. 5/1990 (Undang Undang Nomor 5 Tahun 1990), and states that no utilisation of protected species (i.e. the green python) in any form is permitted without prior permission. Indonesia is currently the only country that permits the 11

29 Jessica A. Lyons commercial export of green pythons, but they must be captive-bred, meaning that individuals must be produced using techniques that are demonstrably capable of producing offspring of second generation (F2). Persons found smuggling and/or misdeclaring trade that is not in accordance with the provision of Government Regulation No. 7/1999 are liable to imprisonment (in accordance with the Customs and Excise Law) and/or a maximum fine of Rp 250 million (approximately USD 27,000). However, permission to take green pythons from the wild for an approved, specific use, such as research or captive breeding, may be granted by the Minister of Forestry (MoF) and under the consent of the Indonesian CITES SA, Lembaga Ilmu Pengetahuan Indonesia (the Indonesian Institute of Sciences [LIPI]) (Samedi and Iskandar 2000) Australia The green python is protected under the Nature Conservation Act (1992) and currently listed as Near Threatened under the Queensland Nature Conservation (Wildlife) Regulation (2006). Freeman and Borsboom (2006) justified this listing with the finding that the population size of green pythons was <3000 mature individuals, or, area of occupancy <40 km 2, or very small number of locations, 10 or fewer. However the data presented by Wilson and Heinsohn (2007) and Natusch and Natusch (2011) suggest that this listing is unnecessary, as both studies found population and range sizes to be considerably greater than previously reported. Under the Nature Conservation Act (1992), a person may not take green pythons from the wild unless authorised to do so (Section 88, 2). The taking, keeping or use of green pythons for display may be authorised by the Department of Environment and Heritage Protection (DEHP) only if it is for: (1) an approved captive breeding program, and is likely to result in a benefit to the species in the wild, or (2) authorised under a conservation plan for the species (Section 30). 12

30 Introduction Papua New Guinea Green pythons are not protected in PNG and can be harvested from the wild for consumption by nationals. They are not currently exported for commercial purposes, but any export of individuals requires a permit issued by the Management Authority (the Department of Environment and Conservation [DEC]) under the Papua New Guinea International Trade (Fauna and Flora) (Amendment) Act Aims This study is one of few to examine the wildlife trade in the Indonesian provinces of Maluku, West Papua and Papua. Certainly, it is the first to examine the illegal trade of green pythons and the entire trade chain, from source to destination. The specific aims of the study were: i. to quantify the scale of illegal trade in green pythons in Indonesia and gain an understanding of harvest demographics, ii. to evaluate the effects of current harvest levels on wild populations, iii. to determine where established trade routes and trade chains operate, iv. to survey the consumer market driven and serviced by this trade, and test the recently proposed concept termed the AAE, whereby species that are rare are disproportionally exploited, reducing their population size and thereby increasing their value, v. to devise a means to regulate the laundering of green pythons through breeding farms, and vi. to examine threats to green pythons in Indonesia and use the above data as a foundation for future strategies to maximise the long-term conservation of this species. 13

31 Jessica A. Lyons 1.7. Study region Papua (formally known as Irian Jaya) informally refers to the western half of the island of New Guinea. It comprises the Indonesian provinces of West Papua and Papua and encompasses 421,981 km 2 (Beehler 2007). The capital city of West Papua province is Manokwari and the capital city of Papua province is Jayapura. New Guinea harbors a diverse range of habitats from lowland swamps and savanna grasslands to montane marshes, and tropical and alpine rainforests. New Guinea supports the largest tract of old growth tropical forest in the Asia-Pacific region and has a high level of biodiversity and endemism. Papua alone has 602 bird species (52 endemic), 125 mammal species (58 endemic), and 223 reptile species (35 endemic) with many species yet to be formally described (Marshall 2007; Marshall and Beehler 2007; EarthTrends 2003). For nearly half a century Papua was virtually inaccessible to all but a few foreign scientists, however, it is now recognised as one of the most biologically important regions on earth (Supiatna 1999; CI 1999; WWF 2000). According to Badan Pusat Statistik (the Indonesian Central Bureau of Statistics [BPS]), in 2005 the population of Papua reached 2.5 million people, more than 61% of who live in rural areas (BPS 2006). Papua is home to ca. 250 tribes, each with their own language and traditional culture, many of which are forest dwelling (Taime 2002). Agriculture is of central importance to indigenous peoples, who cultivate a variety of crops for subsistence, such as sago (Metroxylon sagu), cassava (Manihot esculenta), and banana (Musa sp.). The majority of tribes also hunt and gather as supplementary activities (Boissière and Purmanto 2007; Pattiselanno 2006). Located along the Ring of Fire, where the Indo-Australian and Pacific plates converge, Papua s geography is defined by mountains. The Central Cordillera range extends east to west, and dominates the island of New Guinea, dividing it into isolated southern and northern lowland regions (Frazier 2007). Papua is rich in minerals, oil, and natural gas, and natural resource extraction currently generates approximately 75% of Papua s Gross Domestic Product (GDP); 60% of which is derived from the world-famous Grasberg 14

32 Introduction Mine operated by PT Freeport-McMoRan Copper and Gold, Inc. (Anggraeni 2007). In 2001 Papua was granted decentralisation (under the Special Autonomy Law) from the Indonesian central government, which provided the opportunity for the provincial and regency governments to manage their own resources to develop the two provinces (Mollet 2011) Study sites During fieldwork, twenty-eight sites were visited in the Indonesian provinces of Maluku, West Papua and Papua between August 2009 and April 2011 (Fig. 1.3). These sites were selected upon the basis of known consumer demand for locality specific green pythons, and therefore areas where trade in green pythons was likely to occur. Study sites were grouped into five localities based on geography (Fig. 1.3). In addition, markets and breeding farms that claimed to keep, breed and export green pythons were visited in the Indonesian capital, Jakarta. 15

33 Jessica A. Lyons Fig Twenty-eight study sites visited in Indonesia ( ), and the five corresponding localities from which green pythons are sourced (enclosed in boxes). Note: The Aru Islands are biogeographically part of the island of New Guinea, but politically part of Maluku province (Allison 2007). 16

34 CHAPTER TWO Above: An array of wild-caught green pythons in a local collector s house in Papua province, Indonesia. Below: Inside a breeding farm in Jakarta, a shipment of green pythons waits to be packed for international export.

35 Jessica A. Lyons The illegal trade of green pythons in Indonesia Indonesia currently exports more than 160 reptile species for the international pet trade (Anon. 2011a; Anon. 2011b). A small number of these reptile species are protected under Indonesian legislation and may be traded legally only if bred in captivity. In the early 1990s, in response to recommendations from the CITES Secretariat and high demands from consumer countries, the Indonesian government encouraged captive breeding of selected species for export (Siswomartono 1998). This was intended to aid conservation by: (1) breeding foundation stocks for re-release into the wild and (2) protecting species seriously threatened by commercialisation (Siswomartono 1998). While farming has resulted in reduced pressure on some wildlife populations (Revol 1995), it is feared that commercial breeding may result in increased demand for wild founder stock and be used to launder illegally wild-caught animals (Bulte and Damania 2005; Mockrin et al. 2005). For example, in Indonesia, nationally protected wildlife can be traded under permit if captive-bred, promoting the mis-declaration of animals that are in fact wild-caught (Engler and Parry-Jones 2007). Globally, there are an increasing number of reports suggesting that for many species, this may very well be the case (Auliya 2003; Engler and Parry-Jones 2007; Nijman and Shepherd 2009; Brooks et al. 2010; Vinke and Vinke 2010). Nijman and Shepherd (2009) found large discrepancies between the number of reptiles exported annually from Indonesia and the number of reptiles capable of being produced by Indonesian breeding farms. Their study provided strong evidence for spurious captive breeding in Indonesia (Nijman and Shepherd 2009). The CITES-listed species exported in largest numbers from Indonesia as captive-bred is the green python (CITES Trade Database 2011). Green pythons are keenly sought after by reptile keepers due, largely, to their distinctive and unique colouration. Indonesia is the only range state that allows export of captive-bred green pythons for commercial purposes (CITES source code C ; see CITES 1992). Reptile enthusiasts have 18

36 The illegal trade of green pythons in Indonesia recognised subtle differences in adult and juvenile colouration of green pythons and as such, have designated each colour morph as a specific locality type (Kivit and Wiseman 2005; Maxwell 2005). This has resulted in the search for new colour types. There are numerous reports suggesting that illegal harvesting of green pythons is occurring and that some populations are in decline (Auliya et al. 2009). There is however, no direct evidence of the existence of an illegal trade in wild-caught specimens. Indeed, a recent report submitted by LIPI for the CITES Asian Snake Trade Workshop (2011) stated that illegal trading of snakes in Indonesia was non-existent. Most importantly, there is currently no easy method of differentiating between wild-caught or captive-bred reptiles destined for export (Auliya 2003). In this chapter, the scale of illegal trade in green pythons is quantified the effects of current harvest levels on wild populations are evaluated. The evidence for laundering of green pythons through breeding farms is examined and the role that commercial breeding plays in the conservation of wild animals is discussed Materials and methods Provincial trader identification and interviews Green python provincial traders were identified mainly through anonymous informants, and additional provincial traders, local collectors and villagers were located using snowball (non-probability) sampling (which uses recommendations from provincial traders to establish contact with others; Bryman 2004). Trade data were gathered by conducting semi-structured interviews with traders at each site and included the number of snakes collected, collection trends, and trade history. The information given in interviews was ground-truthed using direct counts of individual green pythons and by crosschecking with others within the trade chain. The average numbers given by provincial traders at each locality were combined to determine the total number of green pythons collected each month. These data were then extrapolated to estimate the total 19

37 Jessica A. Lyons number collected annually from each locality (Table 2a). Companies registered to export reptiles internationally were identified using lists provided by the Indonesian Reptile and Amphibian Trade Association (IRATA) Examination and morphometrics A large number (N = 701) of green pythons in the possession of provincial traders and local collectors was measured to determine the harvest demographic for each locality. The measurements recorded were: (1) snout-to-vent length (SVL), measured with a steel measuring tape to the nearest 0.5 cm; (2) mass, to the nearest 1 g using Pesola spring scales, and (3) sex, determined by insertion of a blunt probe into the cloaca and recording probe depth. Probing is a fast and reliable metric for sex determination (Reed and Tucker 2012), and subsequent dissections of probed pythons (Broghammerus reticulatus and Python breitensteini) reveal 100% accuracy using this method (Lyons unpubl. data). In addition, the colour of snakes was recorded as red, yellow or green and their condition of health was noted. Finally, each snake was given a unique ventral scale clip so that it could be identified if relocated (Brown and Parker 1976). Because of the short duration of this study, and specific use of marks in this instance, this method of marking is preferential to PIT tagging because PIT tags are (a) expensive, (b) clearly visible on snakes, (c) can be easily removed from snakes Results Trade dynamics In total, 13 provincial traders were located and visited a total of 94 times between August 2009 and April 2011 (Table 2.1). Several different individuals are involved in 20

38 The illegal trade of green pythons in Indonesia the trade of green pythons, and collectively they form the trade chain. Villagers working in, or in close proximity to, rainforest during the day opportunistically collect green pythons that are captured by hand and kept in a plastic bottle or bag for a variable period of time. Depending on the ease of access, snakes are sold either directly to a provincial trader situated in a major centre in New Guinea, or via a local collector (Fig. 2.1). Seventy-six per cent (10/13) of provincial traders stated that collecting wildlife was not their only source of income. On average, provincial traders had been dealing in wildlife for about 14 years (range 4 27; 9/13). Only one provincial trader (1/13) harvested green pythons only; the others stated that they also traded in other amphibian and reptile species (see Appendix III). None of the provincial traders (0/13) had attempted to breed green pythons, although 15% (2/13) expressed interest in doing so in the future. Sixty per cent (6/10) stated that they had been approached by foreigners, who had purchased green pythons from them directly. Ninety-two per cent (12/13) of provincial traders reported that they could easily circumvented laws and regulations by paying off officials. Finally, all provincial traders (13/13) were clearly aware that trading wildcaught green pythons was illegal. Table 2.1. Trade data for green pythons collected from five localities in Indonesia. N = number of provincial traders. Locality N Green Times Collected Collected pythons visited per month per year recorded Aru Biak Jayapura Merauke Vogelkop and Raja Ampat Total

39 Jessica A. Lyons Trade chain The wildlife trade in New Guinea is carried out by a chain of actors who collect, buy, transport, and on-sell reptiles, birds and mammals, both legally and illegally. Three common trade chain routes for green pythons are illustrated in Figure 2.1 and transport distances largely determine their use. Regional collectors, local collectors and villagers are based in New Guinea, while middlemen and farms/exporters are based in Java and Bali. Trade in protected species, or those without harvest quotas, is facilitated either by concealing them within shipments of legally harvested species or by provision of rewards to officials and transport providers. Provincial traders claimed that all green pythons collected were destined for licenced breeding farms in Jakarta or Bali, where animals are either exported (as captive-bred) or sold to pet shops for the domestic pet trade (Fig. 2.1). The trade of green pythons does not stop at the consumer. Commercial captive breeding of green pythons is common in consumer countries and has an impact on all levels of trade (e.g. by providing an additional supply of captive-bred green pythons to consumers in consumer countries, which may reduce the demand for wild green pythons from New Guinea; this in turn affects villagers harvesting green pythons from the wild). 22

40 The illegal trade of green pythons in Indonesia Fig Three trade chain routes for green pythons collected in New Guinea to reach the consumer market. The dashed line indicates where wildlife is sent from New Guinea by plane to Jakarta, and therefore where targeted enforcement can take place. The trade chain does not end at the consumer; green pythons can be captive-bred in consumer countries and sold to other consumers Harvest levels In total, 4229 illegally harvested green pythons were recorded between August 2009 and April Most were collected from Biak (including both Biak and nearby Supiori Islands but together referred to hereafter as Biak), with a smaller number being collected from the four other localities (Table 2.1). Seventy-six per cent (10/13) of provincial traders provided information on the average number of green pythons collected each month. These figures were corroborated by the surveys and through interviews with others along the trade chain. It should be noted that two known green 23

41 Jessica A. Lyons python collectors from Vogelkop and Raja Ampat and one from Jayapura did not provide information on the number of green pythons collected. Consequently, the total numbers collected from these localities are likely to be higher. The number of green pythons that are claimed to be collected each month differed significantly among localities (χ 2 = 384, df = 4, P = <0.001). Ground-truthing showed that the numbers provided by traders were consistent, although they did depend on the number and timing of visits. For example, despite only two visits to the trader from the Aru Islands, large numbers of snakes were recorded on both occasions due to a backlog of unsent shipments. One trader from Biak provided us with written records of the number of green pythons collected between January and September 2010 and surveys conducted in this locality indicated that the numbers claimed were consistent with the numbers recorded (Fig. 2.2). The trader from Biak indicated that during the 10 years green pythons had been collected, they had become less abundant. Similarly, collectors on the island of Kofiau (located in the Raja Ampat Archipelago) reported that green pythons had become increasingly difficult to find. According to traders and local people this was due to intensive harvests driven by a high demand for green pythons from this island, which apparently retain their yellow juvenile colouration into adulthood. The single trader on Kofiau estimated that only one snake was collected per month. During two visits I recorded five snakes on each occasion, which were apparently collected over 4-month periods. Further, one long-time trader based in Sorong on the Vogelkop Peninsula and another based in Jayapura reported declines in mainland populations. 24

42 The illegal trade of green pythons in Indonesia 600 Number of individuals Date Fig Numbers of green pythons collected from Biak between August 2009 and April Black columns represent the numbers counted in this survey and white columns represent the number recorded by a single provincial trader. Missing columns indicate months when no data were gathered Variation in harvest demographic among localities There was a significant difference in the demographic composition of green pythons collected at each locality (χ 2 = 22.86, df = 4, P = 0.001; Fig. 2.3 and Fig. 2.4). Adults significantly outnumbered juveniles at all localities except Biak (Fig. 2.3 and Fig. 2.4). In Biak, juveniles were collected more often than adults in both years surveyed (2009, χ 2 = 10.99, df = 1, P = 0.001; 2011, χ 2 = 87.67, df = 1, P = <0.001; Fig. 2.4). The single provincial trader on Biak indicated that when harvesting first began more than 10 years ago, a substantial number of adults were collected. More recently however, juveniles are most commonly encountered. The larger number of juveniles from Vogelkop and Raja Ampat was due primarily to the inclusion of two clutches of recently hatched snakes (Fig. 2.3). 25

43 Jessica A. Lyons % of snakes Aru Islands Merauke Vogelkop and Raja Ampat Jayapura Locality Fig The percentage of green pythons that were red juveniles ( ), yellow juveniles ( ) and green adults ( ), collected from the Aru Islands, Merauke, Vogelkop and Raja Ampat, and Jayapura. % of snakes Yellow Red Green Colour of snake Fig The percentage of green pythons that were yellow juveniles, red juveniles and green adults, collected from Biak in 2009 (black columns) and 2011 (white columns). Sample sizes are presented above each column. 26

44 The illegal trade of green pythons in Indonesia There was no significant difference between the number of red and yellow juveniles collected from all localities (χ 2 = 3.405, df = 3, P = 0.333), or from Biak in either year surveyed (2009, χ 2 = 2.32, df = 1, P = 0.128; 2011, χ 2 = 0.47, df = 1, P = 0.49; Fig. 2.4). No red juveniles were recorded from the Aru Islands or Merauke where this juvenile colour phase does not occur (Natusch and Lyons 2012a). There was a significant change in the proportion of juvenile to adult green pythons collected from Biak between 2009 and 2011 with the proportion of green individuals dropping from 37% to 21%, respectively (χ 2 = 26.77, df = 1, P = <0.001; Fig. 2.4). The numbers of each sex collected at all localities were equal (χ 2 = 1.561, df = 4, P = 0.816). The general health of green pythons being traded was poor. Approximately 80% of snakes were malnourished, showing symptoms of disease and/or infection, or were dead (Fig. 2.5). A small number of green pythons were measured within a day of collection at villages in the Aru Islands and their mass was compared with those of snakes from the Aru Islands encountered further along the trade chain. A significant difference was observed between the mass of each (one-way analysis of co-variance with time since capture as factor, SVL as the covariate, and ln weight as dependent variable, F (1,24) = 72.4, P = 0.001), indicating that the stress of trade conducted in this manner adversely affects the health of green pythons (Fig. 2.6). 27

45 Jessica A. Lyons Fig Green pythons found in the possession of traders: (a), (b), and (c) present with serious health issues, while (d), still coiled on its perch was dead. 28

46 The illegal trade of green pythons in Indonesia Mass (g) SVL (cm) Fig Mass vs. SVL of green pythons from the Aru Islands measured the day of collection ( ), compared to individuals recorded further along the trade chain ( ) Final destination Unique scale clips enabled green pythons to be traced as they moved through the trade chain. After leaving Papua, 60 snakes clipped by us were found again in Jakarta for sale in the Barito Market, in the possession of two middlemen and at a single breeding farm. According to the middlemen, all green pythons in their possession were destined for breeding farms. This confirmed the information given by the provincial traders surveyed, of which 76% (10/13) reported sending green pythons, and other wildlife, to many different middlemen who sell the snakes to breeding farms in Java and/or Bali. 29

47 Jessica A. Lyons 2.3. Discussion Effects of illegal harvest on wild populations of green pythons Evaluating the extent to which harvesting threatens wild populations is important for designing adequate regulatory and management strategies. Other studies have found that harvesting techniques or selection for specific attributes have altered population demographics (Fitzgerald and Painter 2000; Fenberg and Roy 2008). The results presented herein, however, indicate that the opportunistic nature of green python collection at most sites does not result in collection biases based on sex, size or colour. On Biak the large number of green pythons collected allowed determination of the harvested demographic at this site. The large number of juveniles collected from Biak suggests that over-harvesting may have skewed the age composition as has been observed in populations of other harvested snakes (Webb et al. 2002; Sasaki et al. 2008; Means 2009) and, to a lesser degree, in other green python populations (Natusch and Natusch 2011). In support of this view, the single trader on Biak indicated that when harvesting first began more than 10 years ago, a substantial number of large green snakes were collected. More recently however, juveniles are most commonly encountered. Furthermore, the proportion of adult green pythons collected from Biak decreased significantly between 2009 and 2011 (Fig. 2.3). Although the absolute number of adults collected remained the same between sampling periods, field time differed between 2009 and 2011 (two month vs. five months, respectively), suggesting that the smaller proportion of adults indicates a continued decline in the reproductive potential and perhaps even the sustainability of that population. The serendipitous nature of collection means it is unlikely that attention was focused on the collection of juveniles, which may have also explained this difference in Adding to these concerns, the green python population in Biak appears to be distinct, possibly specifically, from mainland populations (Natusch pers. comm.). Intensive, prolonged depredation of this population caused by exploitation may lead to earlier maturation of individuals as well as genetic changes that could increase the risk of extinction and reduce the ability of the population to recover even if this depredation is 30

48 The illegal trade of green pythons in Indonesia brought under control (Congdon et al. 1993; Allendorf et al. 2008). Further, it is unknown whether this level of harvest is sustainable. It may be that current harvest levels are sustainable, but that collection practices have merely skewed the population demographic towards the harvest of juvenile snakes. Additional work is needed to understand the harvest demographic of this site. Trade was also found to occur at other sites, with small numbers of snakes being taken from the islands of Numfor and Yapen and the central highlands. In addition, traders reported that they occasionally obtained snakes from neighbouring PNG, as had been found with the cross-border trade of other wildlife products (Georges et al. 2006; Hitchcock 2006). The scale of trade from these areas was not assessed in the present study, but due to the relatively small numbers of snakes encountered in the trade cycle, it is suspected that the distance of these areas from major trading hubs, combined with the small numbers actually collected, has resulted in minimal impact on those populations. It is widely known that a large number of animals suffer and die in the pet trade every year (Bulte and Damania 2005; Herbig 2010). The results presented herein confirm that the husbandry skills of traders are very poor and snakes lose condition as they are transported from their collection location (Fig. 2.5, 2.6). Data on mortality rates of reptiles transported is sparse, however, Honegger (1974) found that between collection and the final sale to the consumer the number of deaths is extremely high. It is estimated that up to 50% of the green pythons collected at some of the localities will die before they are transported to Jakarta. This rough figure is based upon the total number of snakes collected by traders per month, and the percentage of those individuals that were either dead, or too sick to be sold. Captive-bred specimens are often more attractive for the hobbyist market as they are more resistant to health complications and hence easier to keep than wild-caught animals (Auliya 2003). As such, the finding that many of the green pythons traded were in poor health suggests that trade based on wild-caught individuals is economically counter-productive because provincial traders must carry the cost of being unable to sell sick or dead individuals. 31

49 Jessica A. Lyons The threat of breeding farms The finding that most of the green pythons harvested are destined for breeding farms confirms previously held suspicions that most of the green pythons exported annually as captive-bred are in fact wild-caught. The number of companies in Indonesia registered to breed reptiles for export to supply the pet trade increased from 11 in 2006 to 19 in 2008 (Nijman and Shepherd 2009). Similarly, the number of green pythons exported annually as captive-bred has increased dramatically since the year 2000 (Fig. 2.7) Number of individuals Year Fig Four reptile species exported from Indonesia in the highest numbers as captive-bred between the year 2000 and 2009 (CITES Trade Database, 2011). Morelia viridis ( ), Varanus timorensis ( ), Varanus prasinus ( ), Varanus indicus ( ). Because visits were irregular, the number of green pythons collected per unit time could not be determined. However, the large number of snakes recorded during surveys and the extrapolation of numbers reported by provincial traders each month (Table 2.1) is 32

50 The illegal trade of green pythons in Indonesia similar to the annual number of green pythons exported (Fig. 2.7). This indicates that wild-caught green pythons could make up at least 80% of the annual export of this species from Indonesia. It is also possible that provincial traders have underestimated the numbers claimed to be collected each month because they are aware that such collection is illegal. This, in addition to the lack of monthly harvest data from three provincial traders means that the number collected annually may far exceed the 5337 estimated (Table 2.1). Nijman and Shepherd (2009) found that many Indonesian breeding operators do not have the knowledge to successfully breed many reptile species. Surveys by TRAFFIC discovered that some facilities do not have parent stock and others do not even have premises from which to run a successful breeding operation (C.R. Shepherd, pers. comm. 2011). For breeding farms to be successful they should provide a cheaper, more acceptable, product to the consumer than wild-caught individuals (Bulte and Damania 2005). Not-surprisingly, however, breeding, raising and feeding large numbers of many species over successive years reduces profit margins and is more time consuming than directly selling wild-caught animals. This was confirmed by one farm owner who stated that all other farms launder wild-caught green pythons, so to remain competitive; he had to do the same thing. As well as citing economic motivation, this farm owner stated that the high year-round demand for pets meant that snakes could not be bred fast enough to meet demand a common Achilles heel of species that have relatively long reproductive cycles (Mockrin et al. 2005; Vinke and Vinke 2010). For this reason, it may be that breeding and selling many Indonesian reptile species is not economically viable given that many are slow to mature and have long reproductive cycles. I was told that foreigners who wanted to select green pythons personally had approached the majority of provincial traders surveyed. Provincial traders reported that the snakes chosen were then sent to breeding farms where they are claimed to be captive-bred, given CITES export permits, and shipped to the individual concerned in the importing country. Interestingly, provincial traders I interviewed said they had not collected green pythons prior to the time foreign dealers had approached them offering to pay for snakes. A small number of provincial traders claimed that they currently had orders from foreign clients. Because reptile enthusiasts seem to prefer captive-bred over wild-caught animals (Auliya 2003) I assumed that many who purchase green pythons 33

51 Jessica A. Lyons are not aware of the provenance of that individual. Indeed, there are a number of dealers who knowingly import wild-caught green pythons and other species and sell them as captive-bred, relying on the difficulty of differentiating between the two in order to mislead unsuspecting buyers and enforcement authorities in both Indonesia and the importing countries. It is also likely that other dealers are unaware that they are receiving wild-caught green pythons, relying on the word of the Indonesian exporter that they are captive-bred. Finally, the finding that green pythons were present in Jakarta s Barito Market raises the need for further research into Indonesia s growing domestic pet trade. Regardless of whether the laundering of green pythons for the international trade is stopped, demand from the Indonesian domestic market may continue to place stress on wild populations and prove to be much harder to regulate (Pires and Moreno 2011). 34

52 CHAPTER THREE Above: A designer green python. Keepers prize individuals that retain their juvenile colouration. Below: A provincial trader holds a wild-caught green python that he sold for Rp 4,000,000 (approximately USD 450).

53 Jessica A. Lyons Effects of rarity and consumer preferences on the global trade of green pythons An increasing number of studies are beginning to show that the inherent rarity of a species, whether actual or perceived, can have a negative effect on that species because it can lead to increased rates of unsustainable wild harvest (Brook and Sodhi 2006; Hall et al. 2008; Angulo and Courchamp 2009). Economic theory predicts that the exploitation of a species alone is unlikely to result in extinction because of the escalating cost involved in finding increasingly rare individuals of a declining species (Gault et al. 2008). However, Courchamp et al. (2006) hypothesised that if people place disproportionate value on a rare species, this may result in a cycle whereby increased exploitation reduces the population size, which in turn increases its value and ultimately leads to its extinction in the wild - a concept termed the Anthropogenic Allee Effect (AAE). This concept is founded on two assumptions: (1) there is a positive correlation between species rarity and its value and (2) this fuels sufficient demand to ensure that the market price exceeds the escalating costs of finding and harvesting a declining species (Courchamp et al. 2006). A growing number of studies have suggested that some people do place disproportionate value on rare species (see Slone et al. 1997; Cheung and Dudgeon 2006; Gault et al. 2008; Johnson et al. 2010; Palazy et al. 2011; Tournant et al. 2012). However, there have been no studies that have shown the direct effect consumer preferences for rarity can have on the harvest of a species. Understanding the response of consumers to a species perceived rarity is vital for predicting the impact of intervention strategies that seek to minimize extinction risk (Hall et al. 2008). Because many reptile species exhibit significant intraspecific variation (e.g. colour forms specific to particular regions), I hypothesised that an AAE could occur not only at the species level, but also among populations within a single species. Surveys 36

54 Effects of rarity and consumer preferences on the global trade of green pythons questionnaires, advertisements and harvest data were used to determine the effect of consumer preferences for rarity on the harvest of wild populations of green pythons (Morelia viridis) and test this theory Materials and methods Harvest data from Indonesia At each site (see Section 1.7.1), interviews were conducted with provincial traders who were identified through anonymous informants. Interview questions focused on the number of green pythons collected, prices, collection trends, and trade history. The information given in interviews was ground-truthed using direct counts of individual green pythons and by crosschecking with others within the trade chain Advertisements Data were obtained from advertisements selling green pythons from sixty well-known reptile websites and online classifieds between May 2010 and July The Internet was used to search for advertisements haphazardly. Searches were conducted in English, French, German, and Bahasa Indonesian. For each advertisement the following information was gathered: (1) colour of green python (yellow, red, green, or designer ), (3) price (in USD), and (4) locality Survey questionnaire A structured survey questionnaire was created using an online survey site 37

55 Jessica A. Lyons (SurveyMonkey 2010) and made available on twenty international reptile forums between 29 th May 2010 and 26 th April 2011 (see Appendix I). In addition, a number of well-known green python keepers and breeders (mostly in Europe and the USA) were identified and sent the survey questionnaire via . Each respondent was asked to pass the survey questionnaire to other green python keepers to facilitate a wide coverage of responses. Survey questionnaires were provided in four languages (English, French, German, and Bahasa Indonesian) depending on where the survey questionnaire was offered. Answering questions was not mandatory and respondents could exit the survey questionnaire at any time. Each respondent s IP address was recorded to permit only one response per machine, and once the survey questionnaire was exited, the respondent was not able to re-enter the survey questionnaire. Although Internet surveys such as these are a relatively new means of gathering social demographic data they often provide much larger sample sizes than other survey methods and can be used to target specific groups (in this case herpetocultralists). Furthermore, Internet surveys have been shown to be comparable to more typical methods of survey (Fleming and Bowden 2009). The survey questionnaire consisted of ten close-ended, categorical and multiplechoice questions designed to assess consumer preferences and attitudes towards the trade of green pythons (see Appendix B) Analysis of data Contingency table analysis was used to test for significant differences between what respondents look for in green pythons and Fisher Exact tests to determine differences in the proportion of responses among regions. Only advertisements from North America provided detailed information on the colour of green pythons. Therefore, I excluded colour analysis to this region. A two-way analysis of variance (ANOVA) was used with the log 10 -price as the dependent variable and region and colour as factors, to test for differences between the price of green pythons of different colours between New Guinea and North America. Repeated linear regressions were performed to determine significant relationships between the numbers of green pythons harvested, how common they are in consumer markets and price. Data were log 10 -transformed in order to meet 38

56 Effects of rarity and consumer preferences on the global trade of green pythons the assumptions of normality and homogeneity of variance. Respondents from Australia and Africa indicated that very few locality specific green pythons were available in trade and therefore had difficulty ranking the commonness of the localities listed within the survey questionnaire. Thus, Australia and Africa have been included in the examination of consumer preferences, but omitted from rarity analysis Results Provincial traders in Indonesia As stated in Section , 4229 green pythons were recorded in Indonesia. Seventy-six per cent (10/13) of provincial traders provided information on the mean number of green pythons collected each month. The mean numbers given by provincial traders at each locality were combined to determine the total number of green pythons collected each month (Table 3.1). Table 3.1. Number of green pythons collected per month by provincial traders in Indonesia. Locality Aru Biak Jayapura Kofiau Manokwari Merauke Sorong Mean # collected per month

57 Jessica A. Lyons Surveys responses and advertisements Overall, 410 survey responses were collected from 28 countries. The majority of respondents were from the United States of America (USA) (41.7%), Australia (22.1%) and the United Kingdom of Great Britain and Northern Ireland (UK) (10.4%). Respondents from each country were grouped into five geographic regions (Table 3.1). Although the summary data appear biased as to the geographic origin of respondents (i.e. North America and Europe), the sample fits with the popularity of reptile keeping in these countries. Furthermore, the vast majority of green pythons exported from Indonesia are imported into the USA and Europe (CITES Trade Database 2011). Table 3.2. Summary data for survey questionnaire and advertisements. Region Number of Most desirable Number of Number of respondents that locality of green advertisements respondents keep green python examined pythons Asia Kofiau 37 Australia Australian N/A Europe Biak 51 North America Biak/Kofiau 225 Africa Biak N/A Total Preliminary analysis revealed no significant difference in responses to all questions between respondents that kept green pythons and those that did not. Thus, I pooled these groups together for subsequent analysis. Respondents from Australia and Africa indicated that trade in locality specific green pythons is limited. These respondents stated that very few locality specific green pythons were available and therefore had difficulty ranking the commonness of the localities listed within the survey 40

58 Effects of rarity and consumer preferences on the global trade of green pythons questionnaire. Thus, Australia and Africa were omitted from rarity analysis. Information obtained from advertisements selling green pythons is summarised in Table Rarity, demand, harvest, and price Not surprisingly, there was a strong positive correlation between green python localities ranked as most common on all markets and the mean number harvested from those localities (F 1,5 = 124, P = <0.001; Fig a). There was a strong negative correlation between the advertised price of green pythons from specific localities, and both the commoness (Fig b) and numbers of green pythons harvested (Fig c) from those localities (all P < 0.05). In other words, green pythons from localities where little harvest occurred and were ranked as the least common (i.e. rare ), also obtained the highest prices. Accroding to respondents, the two most sought after localities of green python are from Kofiau and Biak islands (Table 3.1). Provincial traders of green pythons from Kofiau Island claimed that despite snakes becoming increasingly rare, their high price meant they continue to be targeted. There were signficant differences among what respondents looked for in a quality green python, with colouration and pattern being the most sought after traits (χ , df =20, P = 0.008). Proportions of these attributes were almost identical among regions, with the exception that respondents from Europe placed less emphasis on colouration and more emphasis on the locality of green python than respondents from other regions (Fishers Exacts test; Z = -3.04, P = 0.002; Fig. 3.1). 41

59 Jessica A. Lyons Log 10 mean commoness Log 10 mean harvest level Log 10 mean price Log 10 mean harvest level Log10 mean price Log 10 mean commonness Fig Relationships between trade variables for the green python. (a) Mean commonness of green pythons from six localities as ranked by respondents vs. mean number of green pythons that provincial traders claimed to have harvested per month from the same localities. (b) Mean commoness of green pythons from eight localities as ranked by respondents vs. mean price of green pythons from the same localities. (c) Mean price of green pythons from seven localities vs. mean number of green pythons that provincial traders claimed to have harvested per month from the same localities. North America ( ), Europe ( ), and Asia ( ). 42

60 Effects of rarity and consumer preferences on the global trade of green pythons Percentage of respondents (%) South Africa North America Europe Australia Asia 0 Colouration Markings Health Locality Reputable breeder Attribute High market value Fig Percentage of responses given for attributes looked for in a quality green python by respondents from five regions. Note: Percetages may sum to more than one hundred because respondents were allowed to select more than one attribute. Provincial traders in Indonesian New Guinea sold yellow, red and green snakes for the same price. Similarly, the price of yellow, red, and green snakes were similar in North America (Fig. 3.3). However, the mean price of designer green pythons in both New Guinea and the USA were sold for considerably more than those that exhibited normal colouration (two-way ANOVA; F 3,113 = 80.5, P = 0.001; Fig. 3.2). This appears to have had a flow on effect to green pythons harvested in New Guinea, as small numbers of snakes exhibiting rare colours were also sold at a much higher mean price than their normal counterparts (Fig. 3.3). 43

61 Jessica A. Lyons Mean price (USD) $2, $2, $2, $1, $1, $1, $ Red Yellow Green Designer $ $10.00 New Guinea North America Region Fig Mean price that provincial traders in New Guinea and respondents in North America would pay for the different colours of green python Consumer preferences and attitudes toward green pythons Forty eight per cent of respondents (200/410) knew how to acquire a wild green python and 23% (98/410) would purchase a wild green python. ANOVA showed that between geographic regions, there was a statistically signficant difference in the willingness of repondents to purchase wild-caught green pythons (F 409 = 8.79, P = 0.001). On average, 75% of respondents from Asia were willing to purchase a wild-caught green python, compared to an average of 19% of respondents from all other geographic regions. Of those respondents that would not purchase a wild green python, 60% (231/381) would not do so because they considered them to be inferior to a captive-bred green 44

62 Effects of rarity and consumer preferences on the global trade of green pythons python. The main reasons given by these respondents, for their view about the inferior status, were that wild green pythons are more aggressive, have more diseases and parasites (and therefore would be more likely to infect a respondent s reptile collection), harder to acclimatise to captive conditions and to breed and feed. In addition, 8% of these respondents also stated that they would prefer to purchase from a reputable and trusted local breeder that provided captive history records (i.e. pedigree information) of a green python Discussion There has been relatively little research conducted on the AAE and consumers preferences for species in the pet trade, and few studies have related these preferences back to the actual harvest of a species from the wild. Herein, an example of how consumers place disproportionate value on rare populations within a species is presented. Those populations perceived to be rare may be exploited at a higher rate, and this may ultimately lead to their extinction in the wild if harvest is left unchecked Rarity and consumer preferences Green pythons from localities with the lowest harvest levels were ranked as least common by respondents from all regions (i.e. these localities were perceived to be the rarest), and therefore obtained the highest prices on their respective markets. The two most sought after localities were Biak and Kofiau, despite green pythons from Biak being relatively common and inexpensive (Table 3.2). Both of these localities are oceanic islands where individual anomolies in colour and timing of OCC in green pythons are present (Natusch and Lyons 2012a). Colour anomolies in certain green python populations have resulted in sustained consumer demand and thus harvest of these populations for international pet markets. This suggests that rarity is not 45

63 Jessica A. Lyons exclusively related to the abundance, but also to other attributes that confer the perception of rarity (such as colour or time since scientific description). This is further corroborated by consumer demands for designer green pythons and those that are colourful and well patterned. The demand for designer green pythons and colour morphs of other reptiles has recently skyrocketed in the international pet trade with many keepers breeding solely for this purpose (Bartlett and Bartlett 2005; Berry 2010; Maxwell 2005). Indeed, provincial traders in Indonesia showed me photographs from popular literature (Maxwell 2005) which depicted designer green pythons. It was claimed that, if harvested, these snakes fetched the highest price and were therefore most sought after by provincial traders and local collectors. Chapter two explains how green pythons from both Biak and Kofiau were found to exhibit changing size structures of harvested snakes due to collection for trade. Provincial traders claimed that collection of green pythons from these localities was becoming increasingly difficult. As with many reptile species, there is little data on the population biology and demographics of green pythons in the wild. Populations suffering from the AAE may exhibit negative growth rates at low densities, driving them to even lower densities and ultimately to extinction (Courchamp et al. 2006) Consumer preferences and attitudes A quarter of all respondents reported that they were willing to buy wild-caught green pythons. Given the large number of survey respondents this figure is surprisingly high. The motivations behind willingness to pay for illegally wild-caught wildlife are currently many and varied (TRAFFIC 2008). A small number of keepers stated that although they did not want to buy wild-caught green pythons, if spectacularly coloured animals were made available they would purchase them. Others claimed that captivebred green pythons have low genetic diversity and were therefore willing to buy wildcaught snakes in order to increase the diversity within the captive gene pool. Numerous studies have shown that genetic variation does decline over time in captive populations 46

64 Effects of rarity and consumer preferences on the global trade of green pythons (Briscoe et al. 1992; Earnhardt et al. 2004; Lacy 1997). The large number of respondents from Asia who were willing to buy wild-caught snakes suggests apathy about the well being of the natural environment, a view that has been previously suggested (Sodhi 2008; Sodhi et al. 2010a). However, this interpretation may be biased because survey questions did not make it clear that trade in wild-caught green pythons equalled illegal trade. Because of the sensitivity surrounding the illegal trade of wildlife, I attempted to make the survey questions as anonymous as possible. However, the survey questionnaire would have benefited from a representative test of the population sample of respondents (age, education level, income), thus providing valuable information on the socioeconomic correlates of consumer demands for green pythons. Because environmental monitoring and enforcement in countries like Indonesia is poor, gauging how consumers perceive a species (i.e. whether rare or common and/or attractive to collectors) is vital and should be factored into conservation plans. Conservationists and environmental managers should consider monitoring early warning indicators for species becoming popular in trade. This includes having greater awareness of the species (and populations) whose numbers, price and popularity are increasing within the pet trade and might present a potential AAE. Additional studies should be carried out on the links between awareness raising in consumer countries and actual changes in the attitudes and behavior of consumers. In the case of green pythons, educating consumers and raising awareness about how purchasing preferences can directly affect this trade is needed and conservationist should increasingly engage popular media to highlight these issues. As a concluding point, high consumer demands for rare species or populations may not be all bad. Consumers and the public in general may be more willing to financially support programs aimed to protect rare species or populations compared to common ones. Thus, rare species (and populations) can be promoted as flagships for conservation of themselves and sympatric species and habitats (Angulo et al. 2009). However, conservationists are divided as to whether flagship species that act as umbrellas for sympatric species have any conservation merit (Andelman and Fagan 2000; Caro et al. 2004; Munoz 2007). 47

65 CHAPTER FOUR Above: Used green python eggshells and a yellow hatchling. Below: A red hatchling emerges from an egg. 48

66 What comes first, the snake or the egg? A method for regulating the export of green pythons What comes first, the snake or the egg? A method for regulating the export of green pythons In many developing nations, inadequate allocation of resources, a low level of political will, and corruption (Laurance 2004) make adequate regulation and monitoring of wildlife breeding farms difficult. In 2006, Nijman and Shepherd (2009) surveyed several Indonesian breeding farms and noted that only two of these were run professionally as commercial facilities. They also found that the majority of these farms showed no evidence of breeding activities. Currently, there is no easy method for differentiating between wild-caught or captive-bred reptiles destined for export (Auliya 2003). A simple and cost effective method for reliably regulating the laundering of certain protected reptiles from Indonesia and elsewhere is presented herein. It is proposed that captive breeding farms should provide an original eggshell for each individual reptile that is to be exported. If the animal has indeed been bred in captivity, there should be an associated eggshell to prove it Materials and methods The eggshells of five species of snake from the family Pythonidae were measured to determine whether the eggshell size could be used to identify the species involved (Table 4.1). Eggshells from a number of different clutches were measured to avoid intra-clutch homogeneity and to encompass the variation in eggshell size and mass for each species. Eggshells were obtained from a breeding farm in Indonesia and had been kept in a dry and sheltered room between 0 and 2 years. The length and width of an 49

67 Jessica A. Lyons eggshell was determined to the nearest 0.1 mm using a dial caliper. Mass of the eggshell was measured to the nearest 0.1 g using 30 g Pesola spring scales. The eggs of pythons approximate closely to a prolate spheroid. As such, the following formula was used to calculate the volume of each egg in cm 3 : V = 4/3 π a 2 b where a is the width of the egg divided by 2 and b is the length of the egg divided by 2. 50

68 What comes first, the snake or the egg? A method for regulating the export of green python? Table 1.1. Eggshell measurements of the five species of python. N = sample size, SD = standard deviation. Length (mm) Width (mm) Mass (g) Volume (cm 3 ) Common name Scientific name N Mean (SD) Extremes Mean (SD) Extremes Mean (SD) Extremes Mean (SD) Extremes Green python Morelia viridis (3.1) (2.2) (0.1) (2.8) Carpet python Morelia spilota (3.5) (2.6) (0.1) (5.1) Halmahera scrub python Morelia tracyae (3.5) (2.6) (0.1) (11.7) Papuan olive python Apodora papuana (5.3) (3.3) (0.1) (15.6) Sumatran blood python Python curtus (3.6) (2.8) (0.2) (18.3)

69 Jessica A. Lyons 4.2. Results A total of 139 eggshells from five species of snake in the family Pythonidae were measured (Table 4.1). Each species has a distinctively shaped eggshell (Fig. 4.1). The largest eggshells were those of the Papuan olive python (Apodora papuana) and Sumatran blood python (Python curtus). Although both eggshells are large, they differ significantly in their shape, mass and volume (Fig. 4.1; Table 4.1). The eggshells of the green python were the smallest and lightest of those measured (Fig. 4.1; Table 4.1). Figure 4.2 shows the distinctive shape of used Pythonid eggshells of Python curtus, Morelia spilota, Morelia tracyae, Morelia viridis and Apodora papuana Eggshell length (mm) Eggshell width (mm) Fig Length vs. width of pythonid eggshells, showing the distinctive shape of Morelia viridis ( ), Morelia spilota ( ), Apodora papuana ( ), Morelia tracyae ( ), and Python curtus ( ). Error bars indicate the extremes of length and width for each species. 52

70 What comes first, the snake or the egg? A method for regulating the export of green pythons Fig Used Pythonid eggshells showing the distinctive shape of Python curtus, Morelia spilota, Morelia tracyae, Morelia viridis and Apodora papuana Discussion In order to regulate the illegal export of green python, and, indeed, other reptile species, from Indonesia, breeding farms should be required to keep eggshells from the reptiles that are bred and to export them with each individual reptile as evidence of their provenance. Upon reaching their destination, the importing country can then easily and accurately identify eggshells against a reference guide, after which point eggshell can be discarded. The eggshells of reptiles are leathery in texture and, despite compression and dis-colouring, keep their general shape and are consistent in dry mass. Further discussion on proving provenance of green pythons with the eggshell method is provided in Section and Section

71 Jessica A. Lyons Potential loopholes There are several ways in which breeding farms might try to circumvent the proposed eggshell identification method. However, in each case there are ways to minimise the risks of corrupt behaviour Using the eggshells of other species On a few occasions measurement ranges for the different python species examined overlapped. However, never more did a single measurement do so (Table 4.1; Fig. 4.1) and, when used in conjunction with eggshell mass and volume, each species has unique eggshell metrics. Python eggs are roughly spherical or oval in shape and closely approximate a prolate spheroid. The eggs of other snake families such as Elapidae and Colubridae are generally smaller and more elongate. Consequently, confusion between families is highly unlikely. Further, Indonesia does not harvest or breed large enough numbers of suitably sized elapids or colubrids to substitute these shells for those of green pythons. Also, given the complexity and cost of production, it is unlikely that surrogate eggshells could be produced by artificial means The transfer of eggshells Individual breeding farms may produce only small numbers of green pythons each year. The snakes that are bred may produce eggs but not a juvenile of export quality. This could, potentially enable its shell to be attributed to what is in fact a wild-caught individual. Data compiled from captive breeding records of 109 breeding pairs of green pythons in Indonesia in the years 2009 and 2010 showed that each female produced an average of 17 eggs. On average, four snakes in each clutch died, leaving about

72 What comes first, the snake or the egg? A method for regulating the export of green pythons empty eggshells over the 2-year period. It is possible that these eggshells could be kept and attributed to wild-caught individuals intended for export. However, these eggshells represent only a small percentage of the green pythons that can currently be exported from Indonesia each year, and would have a relatively small impact. Reference measurements have been provided to easily and accurately differentiate the eggs of green pythons from those of other python species commonly exported from Indonesia. These results suggest that the eggshell method could be very effective in reducing the laundering and export of wild-caught green pythons through Indonesian breeding farms. There are a number of other python species exported in small numbers by Indonesian breeding farms. These include the black python (Morelia boeleni), water python (Liasis fuscus), New Guinea scrub python (Morelia amethistina) and whitelipped python (Leiopython albertisii). Although a small number of average egg measurements for these species were available (Charles et al. 1985; Barker and Barker 1994; Flagle and Stoops 2009), data sets that encompass the variation between individual hatched egg cases are lacking. Without this information, differentiating between the egg cases could be difficult. As a result, although further studies could turn this limitation around, at present the eggshell method can only be used confidently to monitor trade in the green python. I suggest that this method be trialed in relation to green pythons, with the onus falling upon both the exporting and the importing countries to monitor the eggshells. Gathering comparable detailed information about the size and shape of the eggs of other reptiles, both lizard and snakes, combined with capacity-training on the part of authorities responsible for checking on the legitimacy of imports, may well help to diminish the illegal harvest and trade of many reptile species and be an important step towards conserving Indonesia s reptilian diversity. 55

73 CHAPTER FIVE Above: A green python crossing the road at night. Below: Green python habitat is cleared for agricultural development in Papua province.

74 Threats, recommendations and conclusions Threats, recommendations and conclusions Green python populations are subject to a number of threats, but often the direct impacts of these threats are unknown. The extent that green pythons are distributed on the island of New Guinea, both geographically and spatially, in terms of habitat use, relative abundance and population demography also remain unknown. In addition, little value is placed on species conservation in Indonesia, and especially in Papua, where a conservation strategy or plan exists for only one species (crocodiles, see Whittaker et al. 1985). Papua also faces a number of unresolved management issues, such as unclear jurisdiction between local and central governments, limited fiscal resources and staff allocation, and frequent disputes between the government and traditional communities over land claims (De fretes 2007; Mollet 2011). There is no easy fix to these complex issues, meaning that management of the green python remains somewhat challenging. Nonetheless, several recommendations are suggested in the following section to mitigate and reduce threats to the green python Threats The Anthropogenic Allee Effect (AAE) The AAE hypothesises that if consumers value rarity there may be no economic constraint to the exploitation of a species (or population) at low density (Courchamp et al. 2006). Fifty per cent of Indonesia s population lives on less than USD 2 per day (World Bank 2011). The costs associated with harvesting a declining green python population would often be a meager amount compared to the final market price for increasingly rare individuals thus providing incentive for sustained harvesting. Island 57

75 Jessica A. Lyons populations of many snake species frequently occur at lower densities than their mainland conspecifics (Purvis 2000; Boback 2005) and may be particularly vulnerable to the effects of over-exploitation (Dodd 1993). The isolation and inherent rarity of these island endemics makes them prized and vulnerable targets for overexploitation by the pet trade with many-documented harvest (Shepherd and Ibarrondo 2005). The particular threat that AAE poses for rare species is sufficiently disturbing for scientists to use caution when disclosing rarity, both within the scientific community and the general public. Disclosure of locality information for the endangered Broad-headed snake (Hoplocephalus bungaroides) on the Internet and in popular herpetolocultural magazines resulted in the illegal collection and associated declines of that species in the wild (Webb et al. 2002) Human-induced habitat degradation and loss Indonesia is of great conservation concern because it has one of the highest rates of habitat loss in the world (Sodhi et al. 2010b). Despite being relatively new provinces, West Papua and Papua have been targeted by the government and private investors for their rich natural resources. Human-induced habitat degradation and loss has resulted for several reasons, as listed below Logging and mining Although tropical forest cover is extensive in Indonesia, and in particular in Papua, it is declining (USAID 2008). Corruption, which is defined by Transparency International (2010) as the abuse of entrusted power for private gain, has resulted in high levels of illegal logging in Indonesia (Sodhi et al. 2007). The MoF (2003) found that the rate of forest loss and degradation in Indonesia was approximately 2.1 million ha per year. In total, 10 million m 3 of timber is being illegally exported from Indonesia each year, with 58

76 Threats, recommendations and conclusions Papua contributing 72% of that total. There have also been regular contradictions between provincial and central government decrees relating to logging. For example, the West Papuan Governor issued a decree allowing the export of logs from West Papua, despite the national export ban (Currey et al. 2002). The effects of this logging, both legal and illegal, on wild populations of green pythons are unknown. Nonetheless, the green python is an obligate rainforest species restricted to tropical rainforest habitat; when this habitat is destroyed, populations will be affected Agriculture Clearing rainforest to use the land for agricultural plantations has major deleterious effects on rainforest ecosystems that are highly specialised. The green python is restricted to tropical rainforest habitat, and is yet to be found in other habitats. During surveys of traders in Indonesia, all claimed that green pythons were never found in agricultural plantations, such as oil palm. Indeed, the development of oil palm in Papua has caused deforestation, resulting in significant secondary external impacts such as water pollution, soil erosion and air pollution (Obidzinski et al. 2012). An increase in the number of migrants (especially under the transmigration programme ) to Papua has also seen an increase in exotic crops, which Wilcove and Koh (2009) argue is the greatest immediate threat to biodiversity in Southeast Asia Roads Roads are a critical determinant of most forest loss globally (Vijayakumar et al. 2001; Laurance et al. 2009). The length of existing roads in Papua is approximately 2,700 km, with planned roads to increase this number by approximately 1,250 km (Cannon 2007). In the Indonesian province of Papua, the government of the Sarmi District is planning to construct the Mamberamo Basin Road (Laurance et al. 2009). To complete the project the government is soliciting tenders from logging companies who will be permitted to 59

77 Jessica A. Lyons log approximately 5 km either side of the road (known as the Logs-to-Roads deal) (Murdiyarso and Kurnianto 2008). This will amount to approximately 1000 ha of forest logged for every kilometre of road constructed, and will have a significant impact on biodiversity and habitats (Murdiyarso and Kurnianto 2008). Indonesia has the right to expand its road systems, however, any development should compare the real value of natural systems against any intended man-made replacement before any development occurs (Secretariat of the CBD 2004). Roads are also considered one of the major factors affecting fauna survival (Garriga et al. 2012). Vehicles traveling on roads pose a threat to snakes and are a major cause of mortality (Bonnet et al. 1999; Row et al. 2007). In addition to clear felling habitat where green pythons are known to inhabit, this species is at risk of being killed and/or injured by traffic. During fieldwork in Indonesia, several road-killed adult green pythons were observed. In Australia, Natusch (2011) also observed road-killed adult and juvenile green pythons. In addition, roads may also facilitate the trade in wildlife by providing permanent year-round access through suitable forest habitat where green pythons, and other rainforest species, can be easily collected Harvest and trade Subsistence Many indigenous peoples depend on wild animals and plants for their survival (Roe et al. 2002; Boissière and Purmanto 2007; Roe et al. 2008). Throughout New Guinea, indigenous peoples are known to consume wild green pythons for subsistence (Wilson et al. 2006); however, this practice was not observed during fieldwork in Indonesian New Guinea. Because many rural peoples are turning to alternative, less traditional, sources of protein, the effect of subsistence hunting on green python populations is likely to be negligible. 60

78 Threats, recommendations and conclusions Commercial The green python is a colourful, charismatic snake species that is in high demand as a pet in captive collections worldwide. Chapter two identified the key factors that facilitate the illegal harvest and trade of wild green python in Indonesia. Although green pythons are still relatively common in most of the areas in which they occur, noticeable declines have occurred in island populations and, to a lesser degree on the mainland. The main threat posed by this illegal harvest and trade is over-exploitation. As there is no regulation, monitoring or control of wild harvests, unsustainable trade of green pythons for commercial use is occurring. The high levels of trade observed in this study should act a warning, because in conjunction with other threatening factors such as habitat degradation and destruction, further population declines will occur. For some populations of green pythons, this may result in extirpation Recommendations The following initiatives are recommended to mitigate and reduce the threats to green pythons in Indonesia. These initiatives focus mainly on the illegal trade and the laundering of wild green pythons through Indonesian breeding farms Acknowledging that there is a problem Indonesia needs to recognise that illegal harvest and trade of wild species is occurring. This is the first stage in addressing issues of corruption within conservation and begins the process of rectifying the many issues at hand. Aid from the international community is paramount to ensure that if wild species are utilised, trade is tightly regulated and conducted in a sustainable manner. 61

79 Jessica A. Lyons Educating the consumer market One would assume that with so many green pythons in foreign collections, there would be a large population of healthy, captive-bred animals. It may be, however, that the difficulty of keeping wild green pythons, many of which may have been in poor health when purchased, has resulted in the mortality of many individuals within a short time after reaching their destination. As discussed in Chapter three, educating the consumer market and highlighting the issue of illegal wildlife trade may help to reinforce the need to know the provenance of any animal purchased. This may, in turn, reduce the number of dealers and consumers knowingly or unknowingly buying animals that may be wildcaught. Educating consumers about the impacts of the wildlife pet trade, and how their purchasing preferences can directly affect this trade, is needed. This may be achieved through a number of initiatives, such as: mass media awareness raising campaigns, educational material and actions for schools, exhibits at airports and other major border points, information displayed at zoos, information for travellers (i.e. brochures and in-flight information), and internet web sites Proving provenance of green pythons with the eggshell method Even when corruption is not involved, it is clear that, despite the best efforts of Indonesian authorities, determining whether an individual is wild-caught or captivebred is difficult. In Chapter four, a method for regulating the export of reptiles is examined which suggests that the unique size and shape of the eggs of many reptile species would enable the empty eggshell to be used as proof that any individual snake had been captive-bred. This would work by not permitting a breeding farm owner to export any animal unless they could prove, with its accompanying eggshell, that it had 62

80 Threats, recommendations and conclusions been captive-bred at their farm. Breeding farms should be required to keep eggshells from the reptiles that are bred and to export them with each individual reptile as evidence of their provenance. The results presented in Chapter four shows that green python eggshells are considerably smaller and lighter, in terms of length, width, mass and volume, by comparison with the four other python species examined in this study (Table 4). A single extreme measurement of length overlapped slightly with that of a single egg of the closely related carpet python (Morelia spilota). However, in all other respects the measurements of this eggshell matched those for other green pythons. This species has the smallest egg of all commercially bred python species in Indonesia. With a little knowledge, and the aid of a reference guide, identifying the eggs of green pythons would be a relatively simple task Reducing corruption within conservation Corruption was observed throughout all parts of the trade chain and ranged from local transportation providers to government and conservation authorities, which helped facilitate trade and laundering of green pythons through breeding farms. Reducing corruption within agencies tasked to monitor exports may be the most important step towards deterring laundering of wildlife through breeding farms (Laurance 2004) Enforce current legislation Monitoring adherence to rules and agreements and punishing infringements when detected is an essential part of successful conservation and natural resource management (Walsh et al. 2003; Rowcliffe et al. 2004). Any weakness in enforcement, even at the community level, can foster both corruption and overexploitation of natural resources (Gibson et al. 2005). The current punishment for illegally trading wildlife in Indonesia is very weak. If punishments for the illegal trade in wildlife were greater, it 63

81 Jessica A. Lyons should act as a sufficient deterrent to undertake such illegal activities. In the absence of strong regulatory mechanisms in Indonesia, and given the large monetary gains involved for importers and exporters (more than US one million dollars annually) the demand for green pythons, and indeed other species, will continue to be fulfilled, placing considerable pressure on those populations Adequate monitoring and regulation Currently, breeding farms are inadequately monitored. Regular inspections by individuals trained in the identification of species of interest should be pursued. Disincentives are needed for farm owners who illegally trade in wildlife and these should be appropriate to the scale of illegal trade involved. Registrations should be revoked if farm owners are found to not be complying with national laws (Keane et al. 2008; Wellsmith 2011). Mandatory provision of tissue samples by breeding farms may wish to be considered. These samples could be used for genetic testing to confirm the parentage of stock. Although the costs associated with this undertaking may be practicably prohibitive, such an initiative may deter exporters from conducting illegal activity by highlighting that this technology is available and can be employed if necessary. Established trade routes should also be monitored. There are no breeding farms in Papua, and therefore it can be assumed that any green pythons found leaving Papua, en route to major cities were illegally collected from the wild. Trade routes for green pythons extend from the towns of Manokwari (West Papua province), Jayapura (Papua province), and Merauke (Papua province) to the islands of Bali and Java by ship and plane. The restricted transport options for leaving Papua suggests that monitoring shipments at a few key ports and airports would substantially reduce the illegal trade of green pythons. 64

82 Threats, recommendations and conclusions Determination of economic viability To provide incentives for farm owners not to trade in illegal wildlife, the economic viability of farming should be determined based primarily on the biology of the species involved in the trade (Mockrin et al. 2005). Specifically, account should be taken of the receptiveness of a species to farming, rather than based primarily on consumer demand. The economic feasibility of establishing and maintaining a commercial captive breeding operation for the green python in Indonesia should be examined Husbandry training As discussed in Chapter two, at least two traders wanted to breed green pythons, but did not have the knowledge or capacity to do so. Assistance and training in the methods required to breed this, and other species, may offer economic incentive to those at the bottom of the trade chain not to trade in wild-caught animals Allowing legal harvest Wildlife conservation can sometimes be enhanced through allowing and even promoting the harvesting of wildlife (Webb 2002). Legalising the harvest of green pythons under a quota system may help to improve trade monitoring and provide economic benefit to a wider range of people and local communities. Indeed, Hutton and Leader-Williams (2003) believe that rural communities are of critical importance for the successful implementation of CITES, and argue that they can be seen as an additional enforcement and implementation instrument, supporting national efforts. One issue, however, is that it is currently unknown what level of harvest different populations of green pythons can sustain and would require strict enforcement by Indonesian authorities to ensure harvest quotas are not over-stepped. Ultimately, the sustainable 65

83 Jessica A. Lyons management of wild resources is vital for the future of human and ecological communities Conclusions This is one of few studies to report on the wildlife trade in the Indonesian provinces of Maluku, West Papua and Papua. Despite being illegal in Indonesia, collection of wild green pythons is occurring. Further, the results presented in Chapter two indicate that most of the green pythons exported from Indonesia each year are actually wild-caught and laundered through Indonesian breeding farms under the guise of being captive-bred. One of the main threats to the green python is commercial harvest and trade. Results indicate that this process may be having serious deleterious effects on some wild populations, and, in particular, those on isolated islands. Of most concern is that green pythons from Biak, which may prove to be a distinct species, are experiencing the highest levels of population depletion. It is hoped that by quantifying the level and impacts of illegal trade, identifying the mechanisms by which it operates and publishing this information, international and Indonesian national regulatory agencies will focus on enforcing laws, already in place, that were intended to control this debilitating activity. Gauging how consumers perceive a species, i.e. whether rare or common and/or attractive to collectors, is vital and should not be underestimated. This information should be factored into strategies and plans designed to conserve the species in question. It is important that consumers are made aware of, and have an understanding of, the potential effect that their actions may have on the very species they appreciate. Being identified as a rare species can also be advantageous for conservation. For example, some populations of green pythons may be used as keystone species to conserve and protect rainforest habitat in New Guinea. The results presented in Chapter four suggest that the eggshell method would be effective in reducing the laundering and export of wild-caught green pythons through Indonesian breeding farms. I suggest that this method be trialed with green pythons, 66

84 Threats, recommendations and conclusions with the onus falling upon both the exporting and the importing countries to monitor eggshells. The feasibility of breeding farms for producing wildlife to alleviate harvest of wild animals needs to be re-evaluated. It appears that breeding green pythons is currently not a cheaper alternative to laundering wild-caught animals and is therefore not fulfilling the conservation objectives that led to the establishment of farms in the first place. Adequate monitoring and serious disincentives for illegal activity by farm owners should be introduced and enforced. Despite the growth of wildlife farms and their promotion by governments in Southeast Asia, a thorough assessment of the economic and biological viability of reptile farming is needed. In addition to the eggshell method, and other recommendations offered to regulate trade, future research directions should focus on follow-up trade studies to determine the future extent of harvest and illegal trade levels. Population field studies, particularly of the heavily harvested island populations, are needed to understand the demographic effects of harvest at these locations. If the trade wild of green pythons in Indonesia is to be legalized and promoted, then population demographic models underpinned by robust field data are essential. Ongoing research into the adaptive management and monitoring of green python populations will be fundamental for such a system to provide maximum economic returns, adequate income distribution among rural communities, and ensure that trade does not jeopardies the future survival of this species. 67

85 CHAPTER SIX

86 Resulting publications and editorials Resulting publications and editorials 6.1. Publications Natusch, D. J. D., and Lyons, J. A. (2011). The harvest of Antaresia maculosa from West Papua, Indonesia. Herpetological Review 42: Lyons, J. A., and Natusch, D. J. D. (2011). Over-stepping the quota? The trade in Sugar Gliders Petaurus breviceps in West Papua, Indonesia. TRAFFIC Bulletin 24(1): 5. Lyons, J. A., and Natusch, D. J. D. (2011). Wildlife laundering through breeding farms: Illegal harvest population declines and a means of regulating the trade of green pythons (Morelia viridis) from Indonesia. Biological Conservation 144: Natusch, D. J. D., and Lyons, J. A. (2012). Distribution, ecological attributes and trade of the New Guinea Carpet Python (Morelia spilota) in Indonesia. Australian Journal of Zoology 59(4): Natusch, D. J. D., and Lyons, J. A. (2012). Ecological attributes and trade of the whitelipped pythons (Genus Leiopython) in Indonesian New Guinea. Australian Journal of Zoology 59(5): Natusch, D. J. D., and Lyons, J. A. (2012). Exploited for pets: The harvest and trade of amphibians and reptiles from Indonesian New Guinea. Biodiversity and Conservation 21(11): Lyons, J. A., Natusch, D. J. D., and Shepherd, C. R. (2013). The harvest of freshwater turtles (Chelidae) from Papua, Indonesia, for the international pet trade. Oryx 47(2):

87 Jessica A. Lyons Lyons, J. A., and Natusch, D. J. D. (in press). Effects of rarity on the harvest of wild populations within species. Ecological Economics Editorials Pernetta (2012): Effective and sustainable farming of green pythons requires a sound chain of custody and conservation taxation of end consumers The recent article by Lyons and Natusch (2011) provided much needed evidence of the illegal harvest and subsequent laundering of green pythons by commercial breeding farms in Indonesia. In addition, the paper details results highlighting the unique morphology of green python eggshells, in comparison to four sympatrically occurring python species. The authors conclude that this difference can allow eggshells retained by commercial farming operations to be used as provenance of their captive bred origins for subsequently traded pythons and rather worryingly, argue it be trialled as a method to regulate exports of green pythons. Whilst it is encouraging to see researchers suggesting novel forensic methods to attempt to monitor and regulate the trade in green pythons I believe that eggshells as a sole tool will not prevent further unsustainable harvest of wild snakes. Of particular concern is the fact that the authors fail to highlight one major loophole that might be exploited by breeding farms selective harvesting of gravid females from wild populations. Even with a proposed improvement in monitoring regimes for breeding facilities it is not inconceivable that farm owners could pass off incubating eggs taken from wild caught females, as being produced by breeding stock registered to the premises. Shifting illegal harvest from the current indiscriminate collection of encountered snakes, to focused extraction of mature reproducing females could result in drastic declines in population recruitment and male biased sex ratios in wild populations, as has been seen in selectively harvested shovelnose sturgeon populations (Tripp et al. 2009). Further 70

88 Resulting publications and editorials sustained removal of reproductive females of the more desirable colour variants located on the smaller islands mentioned in their article could lead to localised extinction events. As Lyons and Natusch (2011) state, effective long-term management of green python trade requires effective enforcement of Indonesia s pre-existing wildlife conservation laws. I would argue that in order for this to occur we require a focused approach to establishing a comprehensively monitored and internationally certified chain of custody for traded snakes. Green pythons do represent a good opportunity to establish such a scheme, as a result of the relatively small number of individuals traded internationally on an annual basis when compared with other globally traded snakes and the fact that they are able to command high market prices from end buyers in European and North American markets. Through an international certification scheme end consumers could be charged a conservation tax as a percentage of the purchase price, which could be directed towards improving monitoring and regulation of commercial farming operations. Part of such a monitoring scheme could then include the use of microsatellite genotyping and parentage assignment techniques as a comprehensive forensic technique to reduce the risks of wild caught snakes entering the trade. Given the fact that a library of polymorphic markers is already known for Morelia viridis (Jordan et al. 2002) and the accuracy of parentage assignment techniques is well understood this approach could be put in place with immediate effect, through collaborations between Indonesian authorities and national or international laboratories with relevant expertise. Employing genotyping approaches would provide a fool proof method to prevent further laundering wild caught animals, with definitive results unlike the use of eggshells and would provide end consumers with confidence that they were not contributing to further declines in wild populations. Given all we know about the consequences of overexploitation and the need for new approaches to prevent future declines in species, radical approaches are needed in regulating the future wildlife trade (Pernetta 2009b). As a result, I believe that while the approaches suggested by Lyons and Natusch (2011) are laudable in their intentions, the goals they purport will not be achieved without creating consumer driven demand for ethically and sustainably sourced animals, backed by a certified chain of custody, even if it means the costs of certification are passed on to end buyers. For sustainable 71

89 Jessica A. Lyons exploitation of wildlife to be achieved the conservation community needs to rethink its approach to change consumer attitudes from being price-based to conservation focussed Lyons and Natusch (2012): Consumer driven conservation of green pythons is possible if the price is right: A reply to Pernetta (2012) Preventing unsustainable and illegal trade of wildlife is vital for protecting biodiversity, yet finding ways to monitor and minimise such trade cost-effectively remains a significant challenge. Recently, Pernetta (2012) criticised our suggestion to use eggshells as a means of regulating the laundering of illegally wild-caught green pythons (Morelia viridis) through breeding farms in Indonesia viz. if a snake was bred at a farm it must have an accompanying eggshell (Lyons and Natusch 2011). Pernetta (2012) suggested that such a method may create demand for gravid females from wild populations and argued that it should not be used as the sole tool for preventing laundering. Instead, Pernetta (2012) suggested that a chain of custody should be set up whereby microsatellite genotyping and parentage assignment techniques would provide a fool-proof method of unequivocally identifying wild-caught snakes, the funding for which could be procured via a conservation tax levied from end consumers. We agree with Pernetta (2012) that the eggshell method should not be used as the sole tool, but one part of a multi-tiered, integrated and holistic approach to combating illegal harvest of green pythons. Despite being omitted from Pernetta s (2012) review, our paper also suggested several other means of minimising illegal collection; educating the consumer market, husbandry training, determining the economic viability of green python farming and allowing limited legal collection of wild individuals. The beauty of the eggshell method, however, lies in its simplicity; it is cost effective and can be implemented immediately. We argue, based on our fieldwork, that unless demand for green pythons (and therefore market price) significantly increases, the time involved in searching for gravid females makes it impractical. Indeed, of the 1733 green pythons collected by provincial traders and thoroughly examined by us, 0.4% (N = 7) 72

90 Resulting publications and editorials were gravid females. While it may be naive to assume that intensive searching for selective harvests will not occur in the future, currently every snake encountered by collectors is taken for the pet trade gravid or not. Although microsatellite genotyping and parentage assignment techniques may be a valid means of regulating illegal trade, they will take a considerable time to introduce and have significant associated cost. These techniques are currently not done for any species in Indonesia, not even those of high commercial value or those that are globally threatened, and it is a stretch to expect Indonesian authorities to introduce this system for green pythons. Further, many of the farms regularly exporting green pythons from Indonesia do not have parent stock from which to allow parentage assignment. As an initial step in monitoring and enforcement, which can be implemented immediately at minimal cost, proof of breeding (in the form of eggshells) is sufficient to identify which farms are providing fallacious breeding records. Simply the added request of enclosing eggshells (and not even measuring them) would curtail much of this trade. We suggest, therefore, that while the application of genetic forensic techniques may be more reliable than the eggshell method, the significant cost involved currently makes them impractical for regulating farming operations in Indonesia. Indonesian law has strict requirements for the quarantine and transport of wildlife domestically (Decree of the Ministry of Forestry No. 447/Kpts-11/2003). Because nearly all illegally collected wildlife originating from West Papua and Papua arrives at Jakarta s Soekarno-Hatta Airport, adequate training in identification and regulation of transported wildlife could be targeted at a few airport bio-security officials and would go a long way toward stifling this trade. Despite consumers preferring captive-bred green pythons over wild-caught ones (Auliya 2003) Indonesian farm-bred snakes are attractive due to their lower market price (USD 225 compared to USD 500, respectively). Indonesian exporters currently sell snakes at these prices because of the reduced cost involved in collecting and exporting wild-caught individuals. By comparison, housing, feeding and breeding green pythons at farms may not be economically feasible if prices are not increased. Raising the price 73

91 Jessica A. Lyons of farm-bred individuals, however, and imposing a conservation tax on their exportation will inevitably reduce demand for some snakes from Indonesia. Because Indonesia is entitled to benefit from its biological resources, such a tax may be applicable to a program whereby green pythons are sustainably harvested from specific localities. There is a demand for locality specific wildlife from Indonesia and consumers may be more willing to pay a small levy, and therefore higher price, for sustainably harvested locality specific animals than for certified farm-bred individuals. Such an initiative would provide a greater number of individuals and communities with a source of revenue than is currently the case and may, ideally, promote habitat conservation. For breeding farms to be successful, and in the presence of proper regulatory measures for preventing illegal harvests, they should provide a cheaper, more acceptable product to the consumer than wild-caught animals. Thus, for sustainable farming of green pythons, changing consumer attitudes from being price-based to conservation focussed can be achieved but only if the price is right. 74

92 Locals from the village of Deer, Kofiau (West Papua), look at a book on the green python.