Biodiversity Series, Paper No. 4
S.R. Morton, J. Short and R.D. Barker, with an Appendix by G.F. Griffin and G. Pearce
Department of the Environment, Sport and Territories, 1995
It is imperative that we be clear from the beginning about what constitutes a refuge in arid and semi-arid Australia. The Oxford dictionary defines "refuge" as "place of shelter from pursuit or danger or trouble". In ecological or conservation terms, though, this definition needs to be refined in order to encompass several important varieties of refuge springing from quite different sources.
In evolutionary terms, a refuge is a region in which certain types or suites of organisms are able to persist during a period in which most of the original geographic range becomes uninhabitable because of climatic change. The resulting refuge contains high frequencies of endemic species, because the species in it tend to respond to the contraction of range by evolving differences from their original, widespread stock (e.g. Brown and Gibson 1983). Well-known examples include patches of tropical forests in many parts of the world, refugia for plants in the Arctic, and – of importance to us – organisms confined to isolated, moist environments in arid regions. Such species are frequently referred to as relicts. However, this category of refuges in evolutionary time does not include all phenomena that we need to describe. Ecologists frequently use the term refuge in other ways, particularly to mean a region in which a species or suite of species persist for short periods when large parts of their preferred habitats become uninhabitable because of unsuitable climatic or ecological conditions (e.g. drought, flooding or biologically-driven collapses in food supply). Such ecological refugia may operate for less than one generation, or just a few, but may nevertheless be especially significant in arid Australia because of the continent's propensity to drought. Yet another category of refuge comprises regions in which threatened species occur; in this case, a species has retreated because of factors ultimately to do with environmental changes set in train by European settlement. For purposes of description, and particularly for management action, it is vital to identify the correct category of refuge. Our investigations have led us to the following set of categories, as defined below.
Refuges in historical time from introduced predators and competitors and from changing land-use due to European use that has occurred on the adjacent mainland. Often rich in endangered mammals and endemic species of mammals, reptiles and plants. Also offshore islands from Barrow Island in the north-west to Kangaroo Island in the south-east were refuges from predation by Aborigines and dingoes in pre-European times (Abbott 1980).
3.1.2. Mound springs
Localised habitat with Regional endemics of aquatic invertebrates (isopods, ostracods, and hydrobiid molluscs) and fish (where springs form wetlands). Springs are areas with concentrations of species with limited range and specialised habitat requirements; as more taxonomic work is undertaken, more of the biota is being recognised as regionally endemic. They may also provide a localised habitat for waterbirds that is not affected by seasonal variations in rainfall. The Dalhousie Springs are a well-known example.
Provide a refuge in evolutionary time for species that formerly occupied the litter layer of the forest floor of rainforest at a time when the climate of Australia was much wetter and rainforest extended into areas that are now arid or semi-arid. The Nullarbor caves constitute examples of such refuges.
Provide permanent or semi-permanent wetland habitat in arid or semi-arid environments to the benefit of waterbirds and aquatic fauna and flora. Includes long-lived pools in river systems as well as lakes, and can be broadened to incorporate intermittent wetlands such as Lake Eyre.
Provide a specialised microclimate for plants and animals that allow the continued survival of forms that were far more widespread in the evolutionary past when climates were wetter. These areas often provide additional refuge from present-day ecological forces such as fire. A well-known example is Palm Valley in central Australia.
3.1.6. Mountain ranges
Provide a mixture of refuges in evolutionary time through the presence of sheltered environments and geographically isolated habitats in which speciation may take place, but also provide run-off water and plant nutrients which may produce relatively resource-rich areas functioning as refugia in ecological time-frames.
3.1.7. Ecological refugia
As noted in the previous category, the presence of relatively dependable supplies of moisture and nutrients may provide refugia for animals dependent upon regular plant production for persistence in the uncertain Australian climate.
3.1.8. Refuges from exotic animals
These refuges in historical time include sites other than islands. The refuges have effectively been created by human activities, and include environmental zones at the edge of the range of exotic species (e.g. the northern fringes of the ranges of rabbits and foxes) and localised areas where species have been able to evade the impact of these species (e.g. rugged cliff habitat where animals can escape from the effects of introduced predators and competitors).
3.1.9. Refuges from land clearing
Areas where species or suites of species have survived because clearing has been proved uneconomic due to aridity or infertility. This phenomenon is most prominent at the semi-arid fringes of the arid zone.
The arid and semi-arid zones are widely recognised to constitute those areas which receive less than 250 mm annual average rainfall on the southern margin and, in the north where evaporation is greater, less than 500 mm (Williams and Calaby 1985; Figure 3.1). We adopt this definition and the boundaries shown by Williams and Calaby.
Some geographic structure is crucial as a basis for discussion of refugia within the vast area constituted by arid and semi-arid Australia. Rather than create our own sub-divisions, or use older regions such as those of Beard (1969) even though they may be established in the literature, we chose to use the Interim Biogeographic Regionalisation for Australia (IBRA) of Thackway and Cresswell (1994). The advantage of doing so, i.e. of presenting the information in a context in which future discussion of it can readily proceed, far outweighs the slight disadvantage that may spring from forthcoming minor adjustments to the boundaries of the Interim Regions. Nevertheless, it should be borne in mind that these Regions are still subject to modification and refinement, and are best considered as a convenient framework for discussion rather than as absolute environmental categories. The IBRA was derived by compiling the best available data and information about each State and Territory, including field knowledge, published resource and environmental reports, and biogeographic regionalisations, as well as continental data sets. The IBRA was developed specifically for the National Reserves System Cooperative Program, and the focus is on continental landscape assessments. It is acknowledged that such a regionalisation cannot be the sole criterion for allocating conservation priorities. The Biogeographic Regions occurring in arid and semi-arid Australia are shown in Figure 3.1.
3.3. A literature survey designed to identify foci of biological diversity and to determine the location of refugia
We approached the task of describing refugia by scanning as much scientific and technical literature as possible within the time available to us. For each Biogeographic Region, we accumulated references on natural history and ecology and extracted from them information relevant to potential foci of biological diversity. We made particular efforts to obtain material from the "grey literature" represented by publications of very limited distribution but which frequently contain much detailed survey or ecological information. The information sought concerned:
- species listed as endangered (E) or vulnerable (V) by ANZECC (from that shown in Australian Nature Conservation Agency 1994);
- species that are thought to be endemic to the Region under consideration;
- relict populations (whatever the cause of relictuality);
- other populations considered by authors of the material to be significant;
- and geographic areas identified by the authors as constituting refugia.
This material is presented, Region-by-Region and State-by-State, in sections 4-8, in the order Western Australia, South Australia, Northern Territory, Queensland, and New South Wales.
When this material was brought together, we inspected it to determine whether significant records of species or groups of species fell into a pattern that might reflect the presence of any of the categories of refuges specified in section 3.1. These data were then summarised into identification of the refugia occurring in each Region, again grouped according to State or Territory, in sections 10-14. Our assessment has been inclusive rather than critical; instead of excluding doubtful cases, we have tended to list them. This approach has been adopted because we envisage this Report as a starting point for more detailed field assessment of the refuges identified herein purely from a literature survey.
Sections 15 and 16 discuss some general aspects of the refugia and of the management issues emerging from our analysis. Finally, an Appendix (section 18) examines the possibility of using satellite imagery to identify dependably productive areas that may function as short-term refugia for certain types of organisms. If imagery could be analysed in this way, it might expedite the identification of ecologically significant parts of the landscape requiring special management.
The Australian continent showing the boundary of the arid and semi-arid region (heavy line; from Williams and Calaby 1985) and the relevant Biogeographic Regions (from Thackway and Cresswell 1994).