The natural history, conservation status and ecology of the eastern pygmy-possum (\u3cem\u3eCercartetus nanus\u3c/em\u3e)

Many populations of Australian wildlife appear to be diminishing as human activities rapidly encroach on natural areas. Modern conservation efforts seek to provide management solutions, focussing particularly on those species that are officially classed as threatened. However, knowledge of the true status of most species is glaringly incomplete, with fundamental information about distribution and abundance yet to be properly assembled. Hence, there is an increasing need for applied research, particularly on the under-studied species, to assist efforts to conserve wildlife. The eastern pygmy-possum (Cercartetus nanus) provides a good example of a species that requires further study, particularly because it is cryptic and elusive, and has been the subject of relatively few field studies. Furthermore, whilst it appears to be uncommon there has been debate about the reliability of different survey methods. Thus, it appears that field surveys that employ a range of methods are needed. Given that it is listed as threatened in New South Wales and South Australia, more ecological information is required to assist its future conservation and management. This includes further description of its population dynamics, breeding patterns, habitat use including use of nest-sites, and response of populations to fire. The overall objective of this thesis was to make an original and significant contribution to knowledge and understanding of C. nanus natural history, conservation status and ecology. Field studies were conducted with the objective of collecting demographic data. In this thesis, I also undertook comprehensive reviews of the scientific literature, museum collections and existing databases and brought together much scattered and relatively inaccessible data on this species‘ biology. This included desktop surveys of fossil sites that produced records from 57 sites and provided valuable perspective on its prehistoric distribution. Conservation status was also reviewed on a state by state basis. For Victoria, this included reviewing data contained in 133 published and unpublished fauna surveys conducted from 1968 to 2003. For Queensland, review of databases and previous reports documented that only 13 modern records exist. Available evidence suggests that C. nanus should be listed as Vulnerable in Victoria and Endangered in Queensland. For Tasmania, records were sparse (only 12 fauna survey reports with C. nanus out of 51 examined) and survey data were deficient for some areas, but known populations appear to be small and potentially at risk from habitat loss, inappropriate fire regimes and firewood collection. More refined information on extent of occurrence and area of occupancy (following IUCN definitions) are needed across the broader Australian range of C. nanus, as well as modelling to estimate population size. These are prerequisites for any nomination to list the species as threatened under Commonwealth legislation. Field studies for C. nanus were needed because data in relation to demography, detectability, movements, habitat use, and reproduction are scarce. Field studies were conducted at Barren Grounds Nature Reserve, Jervis Bay National Park, and Royal National Park, all in south-eastern New South Wales. A total of 330 captures (156 individuals) were made of C. nanus between May 2002 and January 2007. Survey effort included 12,349 Elliott trap-nights, 2,530 pitfall trap-nights, 151 spotlighting hours, and 3285 checks of 130 installed nest-boxes. All methods employed successfully detected C. nanus, but they varied in efficiency. Captures were usually confined to sites that had the most prolific flowering of potential food plants. At several sites this was measured by scoring flowering intensity along transects. Late summer-early autumn proved to be the best time to survey for C. nanus as several important food plants were in flower, including Banksia serrata, and animal activity was relatively high. Ecological studies at Barren Grounds revealed that births were restricted to the late-summer to early-winter months (February to June) as hibernation appeared to be recurrently employed during winter at this high elevation site (600 m asl), whereas at Royal, births occurred in all months except August and September. Litters of six surviving young were found on three occasions in Royal, the first such records for this species. Usually only four young are produced but there have been occasional reports of five young. In winter at Barren Grounds, C. nanus was extremely difficult to capture, and torpor appeared to be extensively used. Peaks in capture rates occurred in late summer, particularly amongst adult and sub-adult males, and coincided with nectar-production by several large flowered plants (e.g. B. serrata). There was high site fidelity and resident animals were re-trapped much more frequently than was expected. A total of 15 animals remained in the trapping record for 12 months or more. This included a male that was caught on 20 separate occasions over a 15–month period, and a female whose capture records extended for more than 3.5 years. These records confirm previously published statements regarding this species‘ longevity. Most captures were centred on an area of ecotonal vegetation (Banksia thicket) between the heathland and forest. At Royal, more C. nanus captures were from \u27recently unburnt‘ (\u3e11 years post fire) vegetation (65 captures; 52 individuals) compared with \u27recently burnt‘ (\u3e3 years post fire) vegetation (45 captures; 36 individuals), although the difference was not significant. No significant difference in numbers captured per site was found between either the two fire ages or two habitat types sampled (forest and heath) based on trapping methods alone (Elliotts and pitfalls). However, C. nanus captures from nest-boxes were significantly influenced by habitat but not fire age. This result is likely to reflect the local abundance of tree hollows with the older heath being largely devoid of tree hollows. Indeed, radio-tracking in this habitat showed that animals often denned in the forest and then moved into the heath to forage at night. This suggests that there is a need for a habitat mosaic of forest and heath and with different age vegetation. This study has provided an extensive review of the literature involving C. nanus which suggests that this species is patchy in its distribution. It also suggests that a range of survey methods may detect it but none is consistently better than another. The field work conducted has revealed that C. nanus may sometimes be locally abundant but this appears to be mediated by floral resources that may only be temporally abundant. It appears that C. nanus may show high mobility as it tracks the local abundance of floral resources. There also appears to be an elevation-induced influence on reproduction. That is, females in tableland populations (i.e. Barren Grounds) produced only one litter per year and during a restricted breeding period (December-February to June), whereas females in coastal populations (Royal National Park) bred throughout the year (almost any month) and could produce two litters. This pattern is consistent with studies of other coastal and tableland populations. An important hypothesis that can be advanced from the studies described here is that C. nanus is closely associated with ecotones which maximise floristic diversity, in order to satisfy its diet that is predominantly comprised of nectar and pollen. Thus, managing its habitat to maintain mosaics of habitat type and successional age will be fundamental to its conservation