Notes on Chelodina mccordi timorensis Biology, Harvest, Current Threats, and Community Perceptions in the Lake Iralalaro Region, Timor-Leste
Abstract
The long-necked turtle Chelodina mccordi is considered Critically Endangered under IUCN Red List criteria. In Timor-Leste, the subspecies Chelodina mccordi timorensis is restricted to a small area of lacustrine habitat near the eastern tip of Timor around Lake Iralalaro in the Lautém District. We collected information on C. m. timorensis biology and harvest and assessed current threats and community perceptions. Data were collected during 2 surveys (February and July 2014) around Lake Iralalaro in Nino Konis Santana National Park. Threats were identified by direct observation, and local perceptions were recorded during expert interviews. Human harvest is the main threat in the area. Animals are captured using fishing line, are located using a bamboo stick in shallow water to probe the mud, or are captured by hand at the edge of the lake or under dry grass. Turtles are captured mainly during the dry season (April to October). Most experts identified C. m. timorensis under 2 different names according to the color (staining) of the plastron (veu = yellow and clear; sepe veu = dark and red). Local perspectives as to the population status of the turtle (stable, declining, or increasing) varied between 2 villages. Factors that may be reducing the capacity of this turtle to survive human harvest include predation by pigs and dogs. Fire and climate change are also likely to be important factors resulting in declines.
Chelodina mccordi is a recently described chelid species (Rhodin 1994) that is considered Critically Endangered under the International Union for Conservation of Nature Red List criteria (IUCN 2000). It is also included on Convention on International Trade in Endangered Species Appendix II (CITES 2004) and listed as one of the world's 25 most endangered freshwater turtles (Turtle Conservation Coalition 2011). The 2 disjunct subspecies are restricted to permanent and semipermanent shallow eutrophic lakes and swamps on Roti Island, Indonesia (Chelodina mccordi; Rhodin et al. 2008), and swamps and rivers in the Lake Iralalaro region of Timor-Leste (Chelodina mccordi timorensis; McCord et al. 2007; Kuchling et al. 2007 [as Chelodina mccordi timorlestensis]). The combination of restricted range, limited habitat, and intense collection of animals in the wild for the international pet trade drove the species to near extinction on Roti Island, where it was originally discovered (Rhodin et al. 2004). The Roti population is now considered Critically Endangered to ecologically Extinct in the Wild (International Union for Conservation of Nature/Species Survival Commission [IUCN/SSC] et al. 2000; Iskandar 2000; Samedi and Iskandar 2000; Rhodin et al. 2004, 2008).
In Timor-Leste, the subspecies C. m timorensis is restricted to a small area of suitable habitat near the eastern tip of the island of Timor (Kuchling et al. 2007). The 2 subspecies from Roti and Timor-Leste are currently recognized as divergent island forms of C. mccordi (Georges and Thomson 2010; Turtle Taxonomy Working Group 2014). The Timor-Leste C. m. timorensis was originally collected in the vicinity of Lake Iralalaro (Los Palos) but also occurs in the lowland of Irabere River to the southwest of Los Palos in the Iliomar subdistrict (Kuchling et al. 2007). A subsequent herpetological survey confirmed the presence of a population of C. m. timorensis in this area (O'Shea 2009).
There are reports of abundant C. m. timorensis in Timor-Leste, and it has been suggested that the villagers of Malahara collect up to 30 turtles/d (EPANZ Services 2004; Middleton et al. 2006). However, Kuchling et al. (2007) revisited this region and suggested that the absence of a rigorous interview protocol by the earlier reports probably produced dramatic exaggerations and reflected neither present abundance nor recent past exploitation patterns. Kuchling et al. (2007) did report local exploitation at subsistence level for food and trade in markets at Los Palos. This exploitation and local trade seems to be opportunistic and unorganized, with prices varying from US$2 to US$4 per turtle. On the other hand, this local exploitation might impact small populations of C. m. timorensis. Indeed, there are anecdotal reports of C. m. timorensis declines close to human population centers (Kuchling et al. 2007).
Although the subspecies is not considered rare within its habitat, its limited range makes Timor-Leste populations of C. m. timorensis prone to the same fate as the C. m. mccordi populations on Roti Island (Kuchling et al. 2007). This relatively recently described new taxon (C. m. timorensis) is at high risk of rapid exploitation and extirpation, as was C. mccordi after it was first described (Stuart et al. 2006). There is an urgent need for studies that determine its ecological parameters and a need to improve our understanding of the threats to its survival (Rhodin et al. 2004, 2008).
This article reports the findings of the 2014 Lenuk Lorosa'e Project at Lake Iralalaro, Lautém District, Timor-Leste. We investigated community perceptions and current threats to C. m. timorensis and collected information on species biology and harvest in the Lake Iralalaro region.
METHODS
Study Area
Interviews with local experts (knowledgeable villagers and park rangers) were undertaken in the sucos (villages) of Mehara (its aldeias, or communities, of Loiquero, Poros, and Porlamano) and Maupitini (aldeias of Vailoru and Malahara). We defined as experts those people identified as such by village chiefs and park managers during community meetings. They were locals from the communities around the lake who have the knowledge to hunt turtles specifically and have daily contact with the fauna of the lake. Only locals with such knowledge were interviewed, and they were not self-selected.
We surveyed 3 areas along the inundated grassland and forest of the Lake Iralalaro edge (Fig. 1). The Agua Nova trap site (Irakusalu/Iramiri; 08°26′58″S, 127°10′05″E) was suggested by park rangers (Guarda Florestal) because locals regard it as a suitable site for turtles. The area was characterized by a seasonally inundated, Nauclea orientalis forest floodplain, with cheese trees (Morinda sp.) present. Water depth was approximately 0.5 m in February, and emergent aquatic plants were present. The Moto Swamp (Mehara; 08°24′48″S, 127°09′14″E) was suggested during an interview with a local expert who noted that turtles were captured in this area in 2012. The area is a seasonally inundated floodplain with N. orientalis forest and water depth of approximately 1 m. The Moto Swamp site was heavily used by buffalo and pigs, and no aquatic plants were present. The Maupiti Site (08°27′47″S, 127°09′42″E) was also suggested by the park rangers as an area well known for having turtles. This area is similar to Agua Nova but with a sparser Nauclea inundated forest.
Citation: Chelonian Conservation and Biology 15, 1; 10.2744/CCB-1138.1
Capture, Measurement, and Interview Techniques
To capture C. m. timorensis, we used 7 hoop nets (Legler 1960), as modified by Georges et al. (2006), which extends trapping sessions up to 36 hrs by adding a breathing column. Traps were baited with rice, worms, vegetables, or fresh fish from the lake (we were not able to use tuna or other marine fish owing to local taboos about saltwater fish in the lake). Traps were checked and relocated to different sites every 1 to 3 d. The total survey period was 9 d in February (10–19) and 3 d in July (8–10), both in 2014. Timor-Leste undergraduate students participated in the surveys during these periods. We placed temperature loggers in the 3 main monitoring sites in close proximity to the traps during the February survey period. The loggers recorded water temperature at 30-min intervals.
Turtles captured in traps or observed as pets in the community were measured to obtain maximum straight-line carapace length and width to the nearest 0.01 mm, weighed to the nearest 0.01 g, and sexed by examination of plastron and tail. The tails of males are longer and have a wider base (Fig. 2). We also measured midline plastron length to tail notch, tail notch to the anterior margin of the cloaca orifice, and maximum head width.
Citation: Chelonian Conservation and Biology 15, 1; 10.2744/CCB-1138.1
Interviews were performed to infer population trends of C. m. timorensis according to local knowledge and to assess the level of awareness about the species. Threats were also identified by direct observation. We conducted semi-structured interviews (Eisemberg et al. 2014) with 16 local experts during February and July. We obtained basic personal information about the respondents (sex, age, years living in the area). Interview questions were made according to the following structure: Are there turtles in this area? How many types? Describe each species.
Once a C. m. timorensis description and its name in the local language (Fataluku) were identified, specific questions were made with regard to species status (very hard to see one; seen almost every week, month, or year), estimation of decline (there are more now, there are less now, numbers are the same; is there anything affecting the number of turtles? If yes, what?), habitat (Where does this species live during the dry and wet season? Do you see them outside the water? If yes, what time of the day and year?), reproduction (Do you catch females with eggs inside? If yes, what time of the year?), nesting behavior (Have you seen a nest of this species? If yes, when, where, and how did you find it?), harvest (How do you catch turtles? When do you catch turtles?), animal fate (What do you do with the turtle—eat it, keep it as a pet, sell it alive, sell it dead? How often do you sell them? How much does a turtle cost?), and local taboos related to the species that could potentially impact future conservation projects (Are there any traditional restrictions in relation to the harvest of this turtle? If yes, explain and define what type, time, gear, turtle sex, turtle size, fishing areas, period of the year, on special occasions. Are there particular people who cannot eat it or other restrictions? Are people still engaged in this restriction? If not, why?).
RESULTS
Biology
From the 7 modified snorkel traps used to survey Agua Nova, Moto, and Maupiti, only one trap was successful in capturing a female C. m. timorensis (Table 1). This animal was captured using a trap donated by the University of Canberra in July, which has a more complex frame than those used during the February survey. We used fresh captured fish from the lake in the trap. Water temperatures at Agua Nova and Maupiti varied from 26.4°C to 32.2°C in February (Table 1). The Moto Swamp temperature logger was lost.
We sexed, measured, and weighed 10 C. m. timorensis from Lake Iralalaro (Table 2). With the exception of CM11, which was captured during the survey, all specimens were captive animals. These were captured in the dry season (April to October) of 2012 (n = 2), 2013 (n = 2), and 2014 (n = 3) and the wet season (November to February) of 2012 (n = 2). Most animals were captured by hand during the day (Table 2). For one turtle (CM07), the hunter was not present; thus, it was not possible to obtain specific data on capture.
When biometric data (relationship between linear carapace length and width) from this project were plotted with the data from Kuchling et al. (2007), we observed similar trends (Fig. 3). It is possible to determine the sex of an individual with a linear carapace length >120 mm. The smallest and largest animals observed had carapace lengths of 82 mm and 232 mm, respectively. Only females were larger than 220 mm. All turtles measured in 2014 were captured after 2011, and it is unlikely we measured the same animals as Kuchling et al. (2007).
Citation: Chelonian Conservation and Biology 15, 1; 10.2744/CCB-1138.1
The 16 local experts interviewed were aged from 19 to 90 years old. Fifteen were male and one was female. Years of formal education ranged from 0 to 12. Of those interviewed, 3 were aldeia chiefs, 8 were farmers, 3 were park rangers, and 1 was a member of the suco council. The number of years these respondents were living in the area varied from 8 to 72 yrs, representing from 11% to 100% of their lives. Those living for less than 50% of their lives in the area were originally from other villages in close proximity to Lake Iralalaro. Chelodina mccordi timorensis was described by these respondents as living in Lake Iralararo, both in the water and on land, and estivating in tall grass adjacent to the lake during the dry season. It also inhabits the rivers, creeks, and swamps of the Los Palos region. Twelve of 16 respondents had seen C. mccordi outside the water (nesting or estivating). Two respondents mentioned the best time to see C. mccordi outside the water was at night during the full moon. According to one respondent, turtles were particularly common on the road during the 24 hrs following heavy rain.
Twelve respondents had caught females with eggs. From those that had observed gravid females, 10 agreed that females nest during the dry season (July to October); 1 suggested that the species has 2 nesting periods (one in the dry season and another at the beginning of the wet season); and 1 described the nesting period as restricted to February at the end of wet season, when water levels begin to fall. Of the 12 respondents that had seen nests of this species, 9 agreed that the nests were found in the dry season, whereas 1 respondent found nests in both seasons, and 2 respondents could not remember the period they found nests. Nests were found in the mud, close to the edge of the water and around swamps.
Harvest
Respondents cited 3 methods to capture C. m. timorensis (Fig. 4). Animals are captured using fishing line or by probing the mud in shallow water with a bamboo stick. According to the respondents, the bamboo makes a characteristic sound when it hits the carapace. Turtles are also captured by hand at the edge of the lake or under dry grass. Hunters are able to find nests by looking for a small hole in the mud. They can see the plastron mark made by the female after nesting and a small mound indicating the location of the nest. Respondents can also identify turtle footprints. In some cases, dogs were used to find nests.
Citation: Chelonian Conservation and Biology 15, 1; 10.2744/CCB-1138.1
All respondents except one admitted having eaten turtles. Keeping captured C. m. timorensis as pets was also a common practice around the Lake; 43% of the respondents kept turtles as pets. Children in both Maupitini and Mehara prized their pet turtles and would not accept money in exchange for their animals. Turtles were kept in buckets, in the toilet cistern, or in seminatural conditions (artificial ponds).
Turtles were captured mainly during the dry season. Capture with fishing line can occur in both seasons. Capture by hand or using the bamboo stick usually occurs during the dry season. According to respondents, during the wet season, the hunter needs to use a canoe, and it is harder to find turtles. Several members of the community gather together and go to Lake Iralalaro with the aim of collecting turtles perhaps once or twice a year during the dry season, when the lake is at its lowest level. During these events, the community may harvest up to 30 turtles at a time. The capture of 30 animals in a day is considered possible but unlikely, and usually fewer than 20 animals are captured during these events.
No respondent reported ever selling turtles or turtle eggs. The answers given for this question may have been influenced by the presence of park rangers during the interviews. Respondents reported having seen turtles being sold, especially before Timor-Leste gained independence from Indonesia in 1995. Prices depended on the size of the turtle and the buyer. Hunters would be willing to sell C. m. timorensis to their (local) neighbors for US$1. However, prices for foreigners would vary according to the time, size, and willingness of the buyer to pay high prices for a specimen. Before independence, it was reported that Indonesians would pay from US$1.5 (small turtle) to US$2.5 (big turtle). Animals were then eaten or sent to Indonesia. Following independence, there are reports of foreigners buying turtles from US$20 to US$50.
Local Names
Respondents identified 1 (n = 1), 2 (n = 13), or 3 (n = 2) types of turtles living in the lake. Eight respondents cited the long-necked turtle veu (yellow and clear plastron) and sepe veu (red and darker plastron). Veu and sepe veu were not considered fundamentally different animals. Differences in color were associated with distinct environments, with sepe veu living in shallower habitats. According to respondents, the shallow water close to the edge of the lake caused the staining of the plastron. Both types of turtles were identified as C. m. timorensis with different plastron colors (Fig. 5).
Citation: Chelonian Conservation and Biology 15, 1; 10.2744/CCB-1138.1
Four people defined veu and sepe veu with the generic name of veu mani lohai (long-neck turtle). Two people defined the red-plastron turtle as sepe veu but provided a different name for the yellow plastron type (pai veu, pig turtle or veu ratunu, queen turtle). One person recognized only sepe veu. Two people described a third turtle type under the name of veu akanara (maid turtle) or veu manikava (short-necked turtle). This third type of turtle is described as being similar to the long-necked type with the exception that it has a short neck. We did not find a specimen of the third described species during this study, although it might represent the invasive short-necked geoemydid turtle, Mauremys reevesii, previously identified from Timor-Leste (Kaiser et al. 2010).
Perceptions of Species Status
Personal perspectives concerning the status of the turtle population varied between different sucos. All respondents in Maupitini agreed that the C. m. timorensis population had been decreasing in recent times, whereas in Mehara, 50% of the respondents thought the population was increasing, 33% believed the population was declining, and 17% did not know. The reasons for population increase or decline also varied between sucos (Fig. 6).
Citation: Chelonian Conservation and Biology 15, 1; 10.2744/CCB-1138.1
Respondents in Mehara cited three main reasons for the inferred population increase. According to two respondents, the community can rely on the life cycle of the turtle that is always producing more eggs and hatchlings to continuously increase the population. Two respondents cited the decline of traditional harvest as the main reason for C. m. timorensis population increase. Two respondents cited the recent increase in the water level and consequent decline of harvest as the main reason for higher numbers of turtles in the lake. Higher water levels inundate traditional harvest areas, which become inaccessible. Among the reasons given for population decline was the increase in the human and crocodile population, overharvest, and fire in the lake surroundings (Fig. 6).
Local Taboos
No respondents identified any type of traditional restriction with regard to turtle harvest or consumption. There was no particular time, capture method, sex, size, area, period, or special occasion that should be avoided while harvesting C. m. timorensis. There are serious restrictions on killing crocodiles and snakes attributable to traditional beliefs, but these beliefs do not extend to turtles. Crocodiles and snakes are seen as ancestors that should be respected. A taboo that should be taken into account while planning future monitoring projects in the Lake Iralalaro area is the restriction on the use of material that has been in contact with the sea. No material from the sea or that has had contact with the sea should be used in freshwater lakes.
One of the interviewees was identified as a traditional owner of Lake Iralalaro. This is a hereditary title and the owner is responsible for setting the rules for fishing and harvesting in the lake and vicinity. Vicente Araújo, the present owner of the lake, has announced in recent years that C. m. timorensis harvest should be banned. However, many community members do not follow his rules.
DISCUSSION
According to respondents, C. m. timorensis can be found away from water while they are nesting and estivating. These remarks are not unexpected for a species of the genus Chelodina. Freshwater turtles are capable of terrestrial movement between wetlands (Gibbons et al. 1990) and many estivate, that is, they undergo periods of terrestrial inactivity during dry periods (Gregory 1982; Kennett and Christian 1994; Peterson and Stone 2000; Ligon and Peterson 2002). Several Australian species of the genus Chelodina are known to estivate, including Chelodina steindachneri (Kuchling 1988), Chelodina longicollis (Kennett and Georges 1990; Graham et al. 1996), and Chelodina oblonga (formerly Chelodina rugosa; Grigg et al. 1986). Estivation in the soil in C. oblonga is facultative, occurring when water bodies dry out (Grigg et al. 1986). Chelodina longicollis is also capable of facultative terrestrial estivation (Chessman 1983), but unlike C. oblonga, it often occupies ephemeral waters that may dry unpredictably in consecutive years (Kennett and Georges 1990). Chelodina longicollis exhibits a high degree of variation in terrestrial behavior among individuals (Roe and Georges 2008a), with the use of terrestrial habitats associated with but not constrained by rainfall (Roe and Georges 2008b). Heavy rainfall was a factor frequently mentioned during the interviews as the time when C. m. timorensis is found outside the water.
This study confirms the suggestion of Kuchling et al. (2007) that earlier reports of the collection of up to 30 C. m. timorensis per day by local people (EPANZ Services 2004; Middleton et al. 2006) probably occur rarely and at particular times, such as at the end of the dry season. Respondents indicated that harvest events, when the community can collect up to 30 turtles, are rare. In years with above-average rainfall, when the water of the lake remains elevated during the dry season, such harvest events will not occur. This was the case in 2013, which was considered a very wet year. Nevertheless, human harvest is the main threat to C. m. timorensis around the lake. Factors that may be reducing the capacity of this turtle to survive in the face of human harvest include predation by pigs and dogs.
Some long-lived reptiles can be sustainably harvested when appropriate management actions are implemented (e.g., crocodiles [Bradshaw et al. 2006] and some turtles [Fordham et al. 2007]). Indigenous communities in northern Australia have traditionally harvested C. oblonga sustainably. This species can compensate for low rates of adult survival by adjusting life-history parameters (Fordham et al. 2007). However, it is important to recognize that long-lived species such as turtles have a restricted capacity to respond to large increases in adult mortality (Frazer 1992). A high level of iteroparity (repeated reproduction) is required in species with a low likelihood of reproductive success (Congdon et al. 1993; Heppell 1998). Traditional levels of indigenous harvest of C. oblonga are not considered sustainable when combined with intense predation by feral pigs (Fordham et al. 2008).
Fire and increased frequency of dry weather conditions under climate change are also likely to be an important threat to C. m. timorensis. During the dry season, turtles reportedly estivate in tall grass or stay in the shallow waters of the lake. According to respondents, human-induced fires around the lake kill the turtles or expose them to predators. Climate change may also play an important role in modifying harvest pressure. Major turtle harvest events, when many turtles are harvested at once, are restricted to extremely dry years when the lake recedes to its minimum level. A drier climate would increase the number of major harvest events and decrease turtle harvest relief years, such as occurred in 2013, when high water levels prevented intense harvest.
There is no consensus concerning the influence of climate change on rainfall in Timor-Leste. Different projections in Timor-Leste and the Lake Iralalaro region predict that the climate could become on average 1.5°C warmer and 10% wetter by 2050 (Molyneux et al. 2012), or hotter, drier, and increasingly variable by 2070 (Barnett et al. 2007). Terrestrial strategies of members of the genus Chelodina probably enable most individuals to survive normal dry spells. However, the duration of estivation is limited and rarely lasts more than seven months (Roe et al. 2008). According to Chessman (2011), it is important to understand the interactions between climate change and other threats to inform effective conservation for this group.
The “Projetu Lenuk Lorosa'e” was initiated in 2014 to study and protect C. m. timorensis in the Lake Iralalaro region. During its first phase, we raised awareness about the Critically Endangered status of C. m. timorensis in local communities, provided environmental educational materials (Eisemberg and Perini 2014) for teachers locally and nationally, and trained park rangers and Timor Leste undergraduate environmental science students from Charles Darwin University (Australia) who can potentially continue monitoring and researching this species. In the future, we recommend a community-based population monitoring program for C. m. timorensis in accordance with local beliefs and sensitivities. New trapping areas should be identified to initiate long-term monitoring programs that would include both dry and wet seasons. The possibility of transforming the major harvest events, when many members of the community go to the lake with the main aim of collecting C. m. timorensis, from a hunting/harvesting activity to a monitoring mark–recapture activity should be explored.
We also recommend that future conservation activities on C. m. timorensis at Lake Iralalaro focus on the identification and protection of dry-season refugia and nesting areas, taking into account pig and dog predation, buffalo and cattle trampling, and fire. Captive-breeding models should be explored in the Lake Iralalaro region. However, a feasibility study with captive-breeding experts is necessary to assess the viability or need of this recommendation. According to Fordham et al. (2008), the fencing of wetlands to exclude predators and the production of additional hatchlings to offset losses from predation are all viable management strategies that would ensure ongoing traditional harvest of C. oblonga. However, more information about Timor-Leste long-necked turtle biology and population ecology is needed to guarantee the success of Projetu Lenuk Lorosa'e.
Chelodina mccordi timorensis survey locations in the Lake Iralalaro region, Timor-Leste. Sucos (villages) are marked as circles, and the 3 trapping sites are marked as crosses.
Ventral view of a female (♀) and a male (♂) Chelodina mccordi timorensis from Lake Iralalaro, Timor-Leste.
Relationship between linear carapace length and width for Chelodina mccordi timorensis measured during the present study and by Kuchling et al. (2007).
Capture methods used to harvest Chelodina mccordi timorensis at Lake Iralalaro according to the number of times mentioned by respondents and hunters in the 2 sucos (Villages), Maupitini and Mehara.
Long-necked turtle (Chelodina mccordi timorensis) with different plastron colors. Most respondents identified (a) sepe veu (dark, red plastron) and (b) veu (clear, yellow plastron) as the 2 types of turtles that live in Lake Iralalaro.
Perspectives on Chelodina mccordi timorensis population status (more or less turtles than in the past, or unknown) and reasons for population change given by respondents from 2 sucos (Mehara and Maupitini) around Lake Iralararo. All respondents in Maupitini regarded the population status as declining (less).
Contributor Notes
Handling Editor: Peter V. Lindeman
