Nesting ecology, harvest and conservation of the pig-nosed turtle (Carettochelys insculpta) in the Kikori region, Papua New Guinea

  • Carla Camilo Eisemberg de Alvarenga

Student thesis: Doctoral Thesis


The pig-nosed turtle (Carettochelys insculpta) is a species of great conservation concern because it is the sole survivor of a once widespread family of turtles (Carettochelyidae); it has a very restricted distribution in a global sense. Moreover, C. insculpta is unusual among turtles in many facets of its morphology, ecology and behaviour. However, it is subjected to intensely high harvest pressure though much of its range. Highly prized as food, it is the most exploited turtle in New Guinea. Both turtle and eggs are collected for trade or consumption by villagers. In this thesis, I investigate aspects of the interactions of C. insculpta with its environment and with humans. I examine the relationship between C. insculpta nesting ecology and its environment within and among different populations. I also demonstrate population decline in the last thirty years and how this decline relates to the patterns and trends of adult and egg harvest in the Kikori population in Papua New Guinea. The specific aims of my thesis were (a) to compare nesting areas and nesting phenology in populations of C. insculpta and to interpret differences in life history in an evolutionary and ecological context; (b) examine potential reasons for the evolution and maintenance of the coastal-riverine nesting dichotomy in the Kikori region of Papua New Guinea; (c) evaluate the ecological traits of coastal and riverine areas by comparing nests and sandbanks attributes and water inundation patterns; (d) report the results of a matched market and village survey of the pig-nosed turtle over almost 30 years in the Kikori Delta; (e) assess the effects of new fishing technologies and compare the proportion of turtle capture methods utilised recently with those common thirty years ago; and (f) evaluate the effect of change from a subsistence economy to a cash economy in different tribes. Throughout its range, C. insculpta nests during the drier months when suitable sandbanks are exposed. Rainfall in the drier seasons and residual flows from the wet season dilutes salinities in the estuaries of the larger rivers in PNG, where C. insculpta nests in upstream, estuarine and coastal areas. In the wet-dry tropics of Australia, high flows are restricted to the wet season and high salinities prevail in estuarine regions during the nesting season. Substantial marine incursion occurs at this time in the estuaries and estuarine salinities are high. Nesting is only recorded from the upstream reaches of the rivers inhabited by C. insculpta in Australia. Pig-nosed turtles nesting on the coast are significantly larger than those nesting in riverine areas of the Kikori and both are significantly larger than riverine populations in Australia. The same pattern is observed in the superfamily Trionychoidea generally. Trends towards larger body sizes between species and within and between populations, suggest that the coastal environment and its challenges have an important influence on the size of species. In the Kikori region riverine areas, flooding is highly unpredictable, often remodelling the distribution of sandbanks after the eggs are laid. In any one year, riverine inundation can last for several days, increasing the risk of egg mortality from hypoxia; or, depending on rainfall, may not happen at all. Riverine inundation is often a result of rain in the upper catchment remote from the nesting area and so there is a spatial mosaic in the impact of flooding across sub-catchments. In contrast, on the coast, inundation is caused by tides and is spatially universal and more frequent, but is predictable and of short duration. In addition, coastal sandbanks are usually located on islands where monitor lizards (Varanus indicus),the major nest predator at riverine sites, are absent. Females nesting on the coast avoid the extremely high nesting predation rates that occur from non-human predators in riverine areas. I propose that the differing climatic regimes between the wet-dry tropics of northern Australia and the wet tropics rivers of southern New Guinea explain the differences in nesting patterns and contribute to the difference in body size and attendant life history attributes between these two landmasses. Coastal nesting exhibited by C. insculpta is probably a response to highly stochastic environments, such as the Kikori region. Coastal nesting is complementary to riverine nesting, increasing the nesting season period and providing predator-free nesting areas. Adaptation to marine environments in Papua New Guinea populations may explain the traits in Australia that otherwise defy explanation such as two clutches every second year and delayed egg emergence. It may be a case of traits evolving in one ecological context (coastal beaches) and being maintained because of different advantage from those that drove their evolution in other parts of its range (wet-dry tropics of Australia). Of course the reverse could be true, but my study provides an example of the caution required when placing evolutionary interpretations on life history traits whose manifestation is studied only within a restricted portion of a species range. I provide, for the first time, concrete evidence of a substantive decline in populations of the pig-nosed turtle. My study combines matched village and market surveys separated by 30 years, trends in nesting female size and assessment of levels of harvest, all of which are essential to making a definitive assessment of population trends. Turtle harvest extended from the headwaters to the coast, but no small juveniles (<15 cm maximum carapace length) were found outside the delta. On the basis of these data, my best estimate of the level of decline is 57.2% since 1981. Hand capture of female turtles during nesting was the most common hunting method in the coastal and riverine areas, whereas in the delta, owing to the lack of nesting sandbanks, the most common method of capturing turtles was by fishing line. Hunting methods in the delta and river in 2007-09 were significantly different from those employed in 1980-82. The proportion of captures using fishing line was greater in 2007-09 (10.3%) and use of nets, nonexistent in 1980-82,was responsible for 8.9% of the captures in 2007-09. The proportion of C. insculpta meat sold at market differed between the two periods of study among language-groups. The impact of the introduction of outboard motors was greatly moderated by the scarcity and cost of fuel. It seems that hunting in distant areas is still greatly moderated by the scarcity and cost of fuel and the lack of fundamental infrastructure to facilitate travel and transport of goods. This lack of availability of fuel and access to transport networks and infrastructure represents a latent threat to C. insculpta in the Kikori. In a sense, the impact of changes to Kikori communities through introduction of modern technologies is dampened by the limitations the community faces in putting those technologies to effective use. Carettochelys insculpta was suspected of dramatic declines over most of its range. My results give substance to those concerns and should result in a re-evaluation of the status of the species in New Guinea and globally. However, the global and local perspectives of the pig-nosed turtle are dramatically different. From one side, local villages consider it a fishery to be managed sustainably. From the other side, the global community will see the pig-nosed turtle as an important relict species to be preserved. Both perspectives are similar in their common desire to see populations of the species persist into perpetuity. However, they differ in their response to declines in abundance and in the perception of what level of population decline is acceptable. This diversity in perspective is an important consideration while creating conservation plans for the pig-nosed turtle in PNG. Without community-led action informed by applied research and environmental education and supported by wildlife protection; the current declines will continue to yield unsatisfactory outcomes for both fisheries and conservation. (NB: partial abstract reproduced here)
Date of Award2010
Original languageEnglish
Awarding Institution
  • University of Canberra
SupervisorArthur Georges (Supervisor) & Nancy Fitzsimmons (Supervisor)

Cite this