This thesis investigates the process of habitat fragmentation and the spatial and temporal scales at which it occurs. Fragmentation has become an important topic in biogeography and conservation biology because of the impacts it has upon species??? distributions and biodiversity. Various definitions of fragmentation are available but in this research it is considered to be the disruption of continuity, either natural or human-induced in its origins and operative at multiple spatial scales. Using the distribution of the southern brown bandicoot Isoodon obesulus as a case study, three spatial scales of fragmentation were analysed. At the continental scale, the Australian distribution of the subspecies I. o. obesulus was examined in relation to climate, geology and vegetation cover at the time of European settlement of Australia and two centuries later. Using archived wildlife records and Geographic Information Systems (GIS) analyses, habitat suitability models were created to assess natural and human-induced fragmentation of the distribution of I. obesulus in 1788 and 1988. At the regional scale, a study was made of the distribution of I. obesulus in the south-east of the State of South Australia. Again, natural and human-induced patterns of habitat fragmentation were modelled using GIS with climate, soil and vegetation data for the time of European settlement and at present. At the local scale, the distribution of I. obesulus was the subject of a detailed field survey of 372 sites within 29 remnant patches of native vegetation in south-eastern South Australia in order to understand the variables that cause habitat fragmentation. Geographic information systems were used again but in a different way to carefully stratify the field survey by overlaying maps of topography, vegetation and past fires. The large dataset collected from the surveys was described using six generalized linear models which identified the significant variables that fragment the distribution of I. obesulus at a local scale. From the results of the field surveys, a subset of four remnants was chosen for further GIS spatial modelling of the probability of I. obesulus occurring within remnants in response to fire via a controlled burning programme put in place to reduce accumulating fuel loads. These investigations show that habitat fragmentation can be caused by different factors at different spatial scales. At the continental scale, it was found that climate played a dominant role in influencing the fragmented distribution of I. obesulus but vegetation change during the past two centuries has also had a profound impact on the availability of habitat. Within south-eastern South Australia, the species??? regional scale distribution is constrained by climate and also by soil and vegetation patterns. Dramatic change to its regional distribution occurred in the 20th century as a result of the clearance of native vegetation for planting pastures, crops and pines. Fragmentation at the regional scale has resulted in the remaining habitat being reduced to small, isolated, remnant patches of native vegetation. At the local scale it was found that variables which disrupt the continuity of I. obesulus habitat within remnants include vegetation cover in the 0-1 m stratum, abundance of Xanthorrhoea australis and soil texture. For a subset of sites located in one landsystem of the study area, named Young, the age of vegetation since it was last burnt was also found to be a significant variable, with vegetation 10-14 years old since burning providing the most suitable habitat. Spatial modelling of two scenarios for prescribed burning over 15 years revealed that the use of fire as a habitat enhancement tool will be complicated and require a detailed understanding of the factors that cause natural fragmentation in the distribution of I. obesulus at the local scale. A further conclusion of the study was that ecological relationships between species and their habitats require careful interpretation of multi-scaled datasets and conservation plans for endangered species ought to be made at multiple spatial scales. Future research directions are identified including the linking of multi-scaled habitat fragmentation models to genetic studies of the species throughout its range.
| Identifer | oai:union.ndltd.org:ADTP/257910 |
| Date | January 2003 |
| Creators | Paull, David James, Physical, Environmental & Mathematical Sciences, Australian Defence Force Academy, UNSW |
| Publisher | Awarded by:University of New South Wales - Australian Defence Force Academy. School of Physical, Environmental and Mathematical Sciences |
| Source Sets | Australiasian Digital Theses Program |
| Language | English |
| Detected Language | English |
| Rights | Copyright David James Paull, http://unsworks.unsw.edu.au/copyright |
Page generated in 0.0026 seconds