Introduction
Wild species are traded in large numbers and the international trade is believed to be worth billions of dollars annually (Roe, Reference Roe2008; Smith et al., Reference Smith, Behrens, Schloegel, Marano, Burgiel and Daszak2009). Demand is diverse, and includes the markets for food, luxury goods, tourist curios, medicines and live animals for the pet trade (Nijman, Reference Nijman2010). Overexploitation is one of the main threats to species because wild populations are often harvested unsustainably or illegally (Shepherd et al., Reference Shepherd, Sukumaran and Wich2004; Warchol, Reference Warchol2004).
Basic biological and ecological data are lacking for many of the species traded and it is difficult to determine if the trade is sustainable. One group of animals heavily exploited for the pet trade are the turtles (Lau et al., Reference Lau, Chan, Crow and Ades2000; Lee et al., Reference Lee, Gorog, Dwiyahreni, Siwu, Riley and Alexander2005; Cheung & Dudgeon, Reference Cheung and Dudgeon2006; Shepherd, Reference Shepherd2007; Shepherd & Nijman, Reference Shepherd and Nijman2007; Gong et al., Reference Gong, Chow, Fong and Shi2009; Nijman et al., Reference Nijman, Shepherd, Mumpuni and Sanders2012). Large numbers of turtles are sourced from Asia and collectively their decline has been dubbed the Asian turtle crisis (van Dijk et al., Reference Van Dijk, Stuart and Rhodin2000). The decline in wild turtle populations is driven mainly by overexploitation, together with habitat degradation and loss (Thirakhupt & van Dijk, Reference Thirakhupt and Van Dijk1994; Klemens & Thorbjarnarson, Reference Klemens and Thorbjarnarson1995; Das, Reference Das and Abbema1997; Gibbons et al., Reference Gibbons, Scott, Ryan, Buhlmann, Tuberville and Metts2000; van Dijk et al., Reference Van Dijk, Stuart and Rhodin2000; Moll & Moll, Reference Moll and Moll2004; Cheung & Dudgeon, Reference Cheung and Dudgeon2006; Krishnakumar et al., Reference Krishnakumar, Raghavan and Pereira2009; Horne et al., Reference Horne, Poole and Walde2012).
The freshwater turtle family Chelidae is restricted to South America, Australia, Indonesia and New Guinea (Georges & Thomson, Reference Georges and Thomson2010). The island of New Guinea (divided politically into Papua New Guinea and Indonesia) has the highest diversity of chelid turtles in the Australasian region, with at least 13 species currently recognized, 10 of which are endemic (Rhodin & Genorupa, Reference Rhodin and Genorupa2000; Allison, Reference Allison2006). Despite a number of recent studies on the ecology of these turtles, further research into their trade has been prompted by large harvests by indigenous peoples for food, carapaces (fashioned into curios such as masks) and for the pet trade (Rhodin & Genorupa, Reference Rhodin and Genorupa2000; Samedi & Iskandar, Reference Samedi and Iskandar2000; Georges et al., Reference Georges, Doody, Eisemberg, Alacs and Rose2008; Eisemberg et al., Reference Eisemberg, Rose, Yaru and Georges2011).
Currently, none of the known 13 species of New Guinea chelid turtle are listed in the appendices of CITES (CITES, 2011), meaning that their international trade is not monitored or regulated. In Indonesia chelid turtles are traded under a quota system that is set annually by a meeting of various stakeholders, which includes the Indonesian CITES Management Authority, the Directorate General of Forest Protection and Nature Conservation (PHKA) and CITES Scientific Authority, the Indonesian Institute of Sciences (LIPI), relevant NGOs, and licensed wildlife traders. Harvest and export quotas are set for all non-protected species, including CITES and non-CITES listed species. Little work on wildlife trade has been conducted in Indonesian New Guinea (but see Rhodin & Genorupa, Reference Rhodin and Genorupa2000; Georges et al., Reference Georges, Guarino and Bito2006), which incorporates the provinces of Papua and West Papua. It is suspected that significant trade in chelid turtles is occurring to supply international pet markets (Samedi & Iskandar, Reference Samedi and Iskandar2000) and unsustainable trade may already be a significant threat for Indonesian freshwater turtle populations (Rhodin & Genorupa, Reference Rhodin and Genorupa2000; Shepherd & Nijman, Reference Shepherd and Nijman2007). The present study provides data on the trade of six chelid turtle species from Indonesian New Guinea.
Methods
We (JAL, DJDN) visited the town of Merauke in the Indonesian province of Papua between December 2010 and March 2011 (Fig. 1). Data on the chelid turtle trade were gathered opportunistically while conducting other work on wildlife trade in the region (Natusch & Lyons, Reference Natusch and Lyons2012a). We identified wildlife traders through anonymous informants and using snowball sampling, which used recommendations from wildlife traders to establish contact with other traders (Bryman, Reference Bryman2004). We gathered data continuously during this period for one wildlife trader and we visited another once in December 2010 and once in March 2011. Both traders mentioned there was an additional dealer who apparently traded solely in turtles; however, we were unable to contact him.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary-alt:20160713135241-11927-mediumThumb-S0030605312000932_fig1g.jpg?pub-status=live)
Fig. 1 The town of Merauke and the surrounding villages of Kumbe, Muting, Salor and Okaba in the Indonesian province of Papua. The black line approximates the border between the provinces of West Papua and Papua. The inset map indicates the location of the Indonesian capital, Jakarta, in relation to the island of New Guinea (shaded).
We identified turtles using an identification key (Georges et al., Reference Georges, Guarino and Bito2006) and classified them in accordance with recent taxonomic revisions (Georges & Thomson, Reference Georges and Thomson2010). We recorded maximum carapace length (MCL), measured in a straight line to the nearest 0.5 cm with a steel measuring ruler, body mass, to the nearest 1 g using Pesola spring scales, and sex, determined using external dimorphic characteristics, particularly tail length, which is much greater in mature males (Georges et al., Reference Georges, Guarino and Bito2006). Sex could not be determined for small or obviously immature individuals. We grouped turtles into three arbitrary size categories according to MCL: small (< 12 cm), medium (12–20 cm) and large (> 20 cm).
Results
We recorded 264 chelid turtles, of six species, between December 2010 and March 2011 (Table 1). Most had been collected from rivers and swamps by local people in the villages of Kumbe and Salor, near Merauke, but turtles were also harvested from more remote areas such as Okaba and Muting (Fig. 1). Both villagers and wildlife traders claimed that turtles were also collected from the nearby Wasur National Park.
Table 1 The six species of chelid turtle found to be traded in Papua Province, Indonesia, with number located (N), mean maximum carapace length (MCL) and mean body mass, by sex.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary-alt:20160713135241-31442-mediumThumb-S0030605312000932_tab1.jpg?pub-status=live)
The number, MCL and mean body mass of the six species of chelid turtle encountered is listed in Table 1. The New Guinea painted turtle Emydura subglobosa was most frequently recorded, followed by the northern snake-necked turtle Chelodina (Macrochelodina) rugosa (Table 1). Wildlife traders stated that juveniles or subadults are preferred to adults because their smaller size and weight reduced shipping costs; size class distributions of each chelid turtle traded in Merauke are shown in Fig. 2. A two-way analysis of covariance with sex and species as factors, log 10MCL as the covariate and log 10body mass as the dependent variable revealed no significant difference in relative body mass between species (F (3,127) = 0.73, P = 0.54), or between males and females (F (1,127) = 0, P = 0.95). Contingency table analysis revealed no bias towards the sex of the turtles collected (P > 0.05), except for Reimann's long-necked turtle Chelodina reimanni, but this could be because of the small number recorded (Table 1).
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary-alt:20160713135241-11973-mediumThumb-S0030605312000932_fig2g.jpg?pub-status=live)
Fig. 2 Size class distributions of the six species of chelid turtle (Table 1) traded in Merauke.
Over the 4-month study period turtles were sold to wildlife traders on a continuous basis, with peaks occurring when middlemen delivered shipments of turtles from surrounding villages (Fig. 3). Most traders kept turtles for an extended period of time (1–2 months) until a substantial number of individuals had been accumulated (represented by the peaks at the beginning and end of Fig. 3), before sending them by plane to breeding farms and exporters in Jakarta. We observed > 30 turtles that had died while awaiting shipment.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary-alt:20160713135241-83584-mediumThumb-S0030605312000932_fig3g.jpg?pub-status=live)
Fig. 3 Timeline of the total number of six species of chelid turtle (Table 1) traded between December 2010 and March 2011.
Of the six chelid species recorded, none are listed in the CITES Appendices. Two species are categorized on the IUCN Red List as Vulnerable, two as Least Concern, one as Lower Risk and one has not been assessed (Table 2). The pig-nosed turtle Carettochelys insculpta, which is highly sought-after for the pet trade, was not being traded despite 60 wild individuals being offered for sale by a local fisherman. Because they are protected, the trader refused these turtles, possibly because illegal shipments of C. insculpta are regularly seized by officials (Anon, 2010; So, Reference So2011; TRAFFIC, 2012). New Guinea snapping turtles Elseya branderhorsti were, however, found to be traded despite not having an allotted harvest quota (which makes their collection and trade illegal). New Guinea spotted turtles are fully protected in Indonesia but this species is allocated a harvest quota under an older synonym, E. schultzei, and is still collected for trade. The remaining four species of turtle observed in trade have quotas that are divided equally between the provinces of West Papua and Papua, even though none of them are known to occur in West Papua (Georges & Thomson, Reference Georges and Thomson2010; Natusch & Lyons, Reference Natusch and Lyons2012b). In addition, Parker's turtle Chelodina parkeri is known only from the Fly River and its tributaries in Papua New Guinea and it is possible that the individuals we observed were brought across the Indonesian–Papua New Guinea border to be sold (Rhodin & Genorupa, Reference Rhodin and Genorupa2000).
Table 2 IUCN Red List category (IUCN, 2012) and criteria (IUCN, 2001), and 2011 annual harvest quotas for the six species of chelid turtle found traded (Table 1).
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20160311072515745-0284:S0030605312000932_tab2.gif?pub-status=live)
1 All species are exported for ‘pets’ as specified in the Indonesian quota system, except the New Guinea snapping turtle, which it is illegal to harvest or export
2 Referred to by the synonym Chelodina siebenrocki within the Indonesian quota system
3 This species is fully protected in Indonesia, yet is traded via the quota system under the older synonym Elseya schultzei
Discussion
This study shows that large numbers of chelid turtles are being exported from Indonesian New Guinea for the international pet trade. Such trading is of conservation concern as little is known about the distribution or population status of these species. Moreover, it is difficult to assess the sustainability of this trade because none are CITES-listed and export and domestic trade levels are unknown. Population declines have already occurred for another species of chelid turtle, the Roti Island snake-necked turtle Chelodina mccordi, which has been overexploited for the pet trade in Indonesia (Shepherd & Ibarrondo, Reference Shepherd and Ibarrondo2005; Rhodin et al., Reference Rhodin, Ibarrondo and Kuchling2008).
Many overexploited turtle species are vulnerable to extinction because of their slow maturation rates and low fecundity, making them unable to reproduce quickly enough to sustain trade sourced from wild populations (Congdon et al., Reference Congdon, Durnham and Van Loben Sels1993). This also makes commercial farming unattractive as turtles are costly to breed, house and feed for extended periods. The result is that wild-caught animals are frequently laundered through farms under the guise of being captive-bred (Nijman & Shepherd, Reference Nijman and Shepherd2009; Vinke & Vinke, Reference Vinke and Vinke2010; Lyons & Natusch, Reference Lyons and Natusch2011).
Indonesia has adequate legislation to control the harvest and trade of wildlife from point of capture to export. However, harvest and export quotas are often arbitrarily set and sustainability estimates are not based on scientifically sound research (Soehartono & Mardiastuti, Reference Soehartono and Mardiastuti2002; Schoppe, Reference Schoppe2009). Furthermore, harvest quotas are often inadequately monitored and enforced because of a lack of human resources and funding (Lee et al., Reference Lee, Gorog, Dwiyahreni, Siwu, Riley and Alexander2005), undermining efforts to ensure that species are not overexploited (Shepherd, Reference Shepherd2007; Shepherd & Nijman, Reference Shepherd and Nijman2007). Existing laws and regulations concerning trade in freshwater turtles need to be rigorously enforced in Indonesia and punishments for violation should be enforced to the maximum extent of the appropriate laws (Horne et al., Reference Horne, Poole and Walde2012).
Clearly, there is an urgent need to assess both the levels of exploitation of chelid turtle species in Indonesia and to determine a sustainable level of harvest for each. This should include exploration of appropriate proxies for CITES non-detriment findings. However, the general lack of knowledge of the distribution, ecology and biology of these turtle species makes it difficult to identify practices for sustainable exploitation. For example, Indonesia's current harvest quotas for chelid turtles apply to the provinces of both West Papua and Papua (Table 2). With the exception of Elseya novaeguineae (and possibly E. subglobosa), however, these turtle species do not occur in West Papua province (Natusch & Lyons, Reference Natusch and Lyons2012b). Further studies of the trade in chelid turtles in Papua are needed, together with enforcement of domestic and international laws and conventions, to ensure population of these species do not decline. We also suggest that Indonesian authorities recognize recent taxonomic changes, to avoid confusion relating to trade using older synonyms.
The information presented here supports the recommendations of van Djik et al. (Reference Van Dijk, Stuart and Rhodin2000) and Horne et al. (Reference Horne, Poole and Walde2012) for inclusion of New Guinea's chelid turtle species in the Appendices of CITES. This is an important first step, and would enable the international community to assist Indonesian authorities in monitoring exports and enforcing harvest and export restrictions to ensure that chelid populations are being sustainably harvested.
Acknowledgements
We thank the Indonesian traders who allowed us to examine turtles in their care, and two anonymous reviewers for their helpful comments. This work was conducted in accordance with the University of New South Wales Animal Ethics protocols.
Biographical sketches
Jessica Ann Lyons is studying the status, conservation and trade of reptiles in Indonesia. Daniel James Deans Natusch currently works on the evolutionary biology, ecology, conservation and sustainable use of reptiles in Australia, Melanesia and South-east Asia. Chris Shepherd has an interest in wildlife trade issues and specifically in the exploitation of Asian turtles.