Dunwoody, John Ernest (2009) Detection of Greater Bilby (Macrotis lagotis) habitat in South-West Queensland using remote sensing imagery and geographic information systems. Coursework Masters thesis, University of Southern Queensland. (Unpublished)
|
Text (Whole Thesis)
Dunwoody_2009_whole.pdf Download (95MB) |
Abstract
The Greater Bilby (Macrotis lagotis) was once one of Australia’s most widespread marsupials covering seventy percent of the continent before white settlement. It is seldom seen now and is classed as Vulnerable under Federal legislation and Endangered under various State Acts.
Unfortunately, its fate is similar to the fate of many other small body-weight marsupials that have become extinct or are in serious decline in Australia. Their collective contribution to the biodiversity of Australia’s ecosystems remains poorly understood and appreciated despite the
level of resources committed to exploiting their habitat for human benefit. The causes of their decline are well understood but not remediated. Rather than address these causes to ensure the species survival and recover biodiversity, the national and state policy response has been to rely on enclosed and captive populations for the survival of the species.
In this study, the microhabitat features used by a population of bilbies in the Bilby Enclosure
at Currawinya National Park were studied by GPS tracking and aerial photography to develop detailed knowledge about the bilbies habitat requirements in order to construct a habitat suitability model. The bilbies in the enclosure live in a semi-wild environment being protected from predators by a predator proof fence. There is no other human intervention. Aerial photographs, taken in 2003 and prior to the release of the fist bilbies (2005), were classified into five landcover classes. Soil tests were taken throughout the enclosure and these formed
the basis for mapping its soil types. Signs of tracks made by the bilbies such as their burrow entrances, foot prints, feed scrapes and scats were recorded by relative GPS. This information was used in a GIS to extract activity frequencies for each landcover class and soil type. The
frequencies were used in a weighted sum overlay model in Arc GIS to predict the suitability of habitat in the enclosure with 84% accuracy for burrowing sites and 80.5% accuracy for feeding sites. Burrowing and feeding habitat priority areas were combined for the whole enclosure to produce an overall estimate of the amount of suitable habitat in the enclosure for bilbies.
The results provide a quantitative measure of the amount of bilby activity in each soil type and landcover class. They also show that the amount of microhabitat suitable for bilby activity is only a small percentage (about 5%) of the total area available to them. The areas utilised by
the bilbies were widespread throughout the enclosure. These measures of microhabitat suitability provide a basis for comparing other actual or potential bilby habitats for use as future bilby release sites.
The reliability of the results is limited by the use of only two predictor layers, the small number of burrows in the model sample (6) and the use of aerial imagery that did not have an infra-red sensory band. It is strengthened by the use of accurate empirical landcover and soil type distribution data. Future work to correct the shortcomings in the model is recommended.
Statistics for this ePrint Item |
Actions (login required)
Archive Repository Staff Only |