Griffith School of Environment, Griffith University, Nathan QLD 4111, AUSTRALIA.
The south east mainland sub-species of spotted-tailed quoll (Dasyurus maculatus maculatus) was the subject of the present research with a focus on populations persisting in southern Queensland and the northern section of the New England Tableland bioregion. D. m. maculatus is the largest marsupial carnivore remaining on the mainland of Australia. The sub-species is listed as Endangered under national legislation and since European settlement has experienced a range retraction of 50-90%. Estimates indicate that only 10,000 or fewer individuals are now surviving in the wild.
To contribute to the growing body of knowledge of wild spotted-tailed quoll populations the study investigated the distribution, detection, diet and population dynamics of quolls using hair funnels, live trapping and scat analysis. Population viability analysis modelling was used to predict the fate of a population.
A search for quoll populations in the Southern Downs region of southern Queensland confirmed that the sub-species is now largely absent from the area with only one thriving population located at Cullendore. Extension of the search area into northern New South Wales indicated that a widely dispersed metapopulation potentially survives in the New England Tableland bioregion with its northern limits in southern Queensland and its southern limits south of Armidale. An initial assessment of the comparative abundance of quolls and introduced carnivores indicated that competitive and predatory interactions may outweigh benefits quolls derived from the presence of extensive areas of high quality habitat.
The detection of quoll populations has proved problematic as the species has an extremely large home range, naturally occurs at low density, and has cryptic and nocturnal habits. To improve the ability of researchers to detect quolls, a hair funnel survey method using Faunatech funnels was trialled and a new detection unit was produced. Results indicated that using widely spaced hair funnels over long transects was an appropriate design that could allow researchers to adequately cover a large survey area. Testing of various detection units using captive quolls indicated that the design of the Faunatech funnel may not be appropriate for detecting quolls with hair samples retained during only 20% of trials when captive animals were observed investigating the funnels. The new Quoll Tube (V4) successfully retained hairs during all of the small number of captive trials suggesting that field trials of this unit are now warranted.
Field work conducted as part of the present research at a site in southern Queensland between 2004 and 2006 was combined with results from other trapping undertaken at the site between 2002 and 2005 to produce five years of demographic data, the longest time series dataset available for the sub-species. Analysis of this data set found the following. Initial trapping at a site is generally male biased but it is likely that the sex ratio of wild populations is close to parity. Extensive trapping, both in terms of the spatial organisation of traps and the longevity of trapping, is required to successfully trap the majority of females at a site. The species is highly sexually dimorphic but morphometric measurements can not be used to estimate the age of newly trapped quolls as substantial individual variation within age classes for both sexes is apparent. Long term trapping results demonstrate that the lifespan of quolls is generally only three years with very few individuals surviving into their fourth year. The percentage of females successfully producing litters in a population is high as is the average number of pouch young born but juvenile mortality, both pre and post weaning, is naturally substantial. Despite high mortality rates juveniles account for approximately 50% of the trappable population during autumn when newly independent juveniles are present in the population. Population turnover is high with 66% of individuals captured during only one year of the study and the proportion of individuals captured in two successive years ranging from just 29% to 47%.
The diet of spotted-tailed quolls at the Cullendore site was investigated using scats collected from live trapped individuals during 2005 and 2006. Consistent with other dietary accounts, this study found the sub-species to be hyper-carnivorous and largely dependent on mammals. Common brushtail possums (Trichosurus vulpecula), swamp wallabies (Wallabia bicolor) and eastern grey kangaroos (Macropus giganteus) dominated the diet of both sexes while small mammals, birds and reptiles contributed little to the diet of both sexes. Invertebrates contributed significantly more to the diet of females than males and a trend was also observed that indicates that invertebrates are particularly important in the diet of juveniles and smaller individuals. As with most other studies, quolls were reliant on only a small number of species with the majority of detected species occurring comparatively infrequently. This trend, however, appeared to be more pronounced at Cullendore than has been reported elsewhere. The diet of quolls at Cullendore also differed from most other studies in the following ways: (1) a negligible occurrence of rabbits and glider species (Petaurus sp. and Petauroides sp.) (2) a heavy dependence on large macropod species; and, (3) a negligible contribution from small terrestrial mammals.
The potential impact on the resident quoll population of a proposed large hotel and villa complex development at the Cullendore site, which would result in a landscape change from remnant quoll habitat to an environment dominated by human landuse, was investigated using population viability analysis (PVA) modelling. The VORTEX model was parameterised with site specific demographic data collected during the five years of quoll research undertaken at Cullendore. The model indicated that the survival of the population was not assured even if the proposed development of the site did not go ahead indicating that even current levels of disturbance may be too high. Any increases in juvenile mortality in particular resulted in the likelihood of the survival of the population decreasing dramatically along with the mean time to first extinction. A combined increase in juvenile and adult mortality and a reduction in carrying capacity reflecting the loss of habitat resulted in the almost assured extinction of the population.