Abstract

An extensive fauna survey of forest habitats in the Murwillumbah Forestry Management Area of north-eastern New South Wales allowed us to examine and compare the use of Elliott traps, wire cage traps, soil plots, hair tubes, spotlighting, dry pitfall traps, sightings and vocalisations as methods to determine the identity, distribution and abundance of ground-dwelling and arboreal mammals. Transects were established in 10 areas in six State Forests and one National Park.

The aim of this study was to examine each method to satisfy three criteria and to determine the most efficient and effective method or combination of methods to survey arboreal and ground-dwelling mammals in forests. The three criteria were to identify the species present, to determine their distribution and to establish an index of their abundance. In all, 35 species were identified by the seven methods. However, the species recorded, and the usefulness of the method to provide their distribution and/or an index of abundance, differed markedly according to the method used.

Stratified sampling with Elliott traps for small mammals, spotlights for arboreal mammals and soil plots for medium-sized and large ground-dwelling mammals was the most labour-efficient and productive combination for recording species present and their status comprehensively. If needed, other methods such as hair tubes or cage traps may be used locally and intensively to assist in the identification of species. Moreover, randomly gathered sight and vocalisation records accumulated during the survey were found to be very useful in corroborating species present and in helping to identify the tracks recorded to genus or family by the soil-plot method.