Biodiversity publications archive

Refugia for biological diversity in arid and semi-arid Australia

Biodiversity Series, Paper No. 4
S.R. Morton, J. Short and R.D. Barker, with an Appendix by G.F. Griffin and G. Pearce
Biodiversity Unit
Department of the Environment, Sport and Territories, 1995

15. Summary of refugia in arid and semi-arid Australia

15.1. An overview of the refugia in each category

Our investigations led us to classify refugia into nine categories: islands; mound springs; caves; wetlands; gorges; mountain ranges; ecological refugia; refuges from exotic animals; and refuges from clearing. The following listing shows how the refugia fall into those categories; again, the reference number allows the reader to see which State or Territory the refuge is centred upon. A minority of the refugia were classed as possessing refugial value within more than one category; in those cases, they are listed under the first-mentioned category.

15.1.1. Islands

There are nine examples:

15.1.2. Mound springs

There are four in this category:

15.1.3. Caves

Only three fall into this class:

15.1.4. Wetlands

The most numerous refuges, with 36 examples:

15.1.5. Gorges

There are five in this category:

15.1.6. Mountain ranges

Twelve refugia were identified:

15.1.7. Ecological refugia

There were two classified here:

15.1.8. Refuges from exotic animals

Three such refuges were identified:

15.1.9. Refuges from land clearing

Two refuges:

15.2. An overview of the relative importance of the individual refugia

As explained in section 9, each refuge was scored as to its apparent importance according to the numbers of species identified as ANZECC-listed, endemic, relictual, or otherwise significant, to a maximum score of 9. We emphasise again, as we did in that section, that these scores cannot be considered as concrete, objective or final. They are put forward as a guide only, for readers to consider and to modify in the light of their own experience and interests. In particular, we must mention the matter of spatial scale. It is evident that the ranking system we developed is scale-dependent; that is, it is likely to produce higher scores for larger areas because they will tend to possess more species of significance. This should actively be borne in mind when these rankings are inspected.

Below, we list the refugia in the order in which our scores rank them. The refugia are not ranked within a particular score. The listing suggests that, although wetlands dominate in terms of numbers of identified refugia, they rank lower than most other categories.

Nevertheless, these wetland refugia are of exceptional importance because of the vast numbers of waterbirds that inhabit them, a significance reflected by the fact that many are listed under the Ramsar Convention (Phillips 1993).

15.2.1. Extremely significant refugia

Score=9
Score=8
Score=7

15.2.2. Highly significant refugia

Score=6
Score=5
Score=4

15.2.3. Significant refugia

Score=3
Score=2
Score=1

15.3. Discussion

Our attempt to specify and classify refugia in arid and semi-arid Australia has covered a great deal of ground and raised a number of important issues that require mention here. The most vital of these is the necessity to recognise that several distinct phenomena are likely to be included within any investigation of refugia. The concept of a refuge covers the following important but distinct situations:

The choice of classification depends, of course, on the aims of the investigator. In our case, we have chosen to be inclusive rather than to set aside information that may be relevant in the task of managing this continent’s biota more effectively. Nevertheless, we need to mention several further issues flowing from the realisation that our investigation cuts back and forwards across several conceptual aspects of refugia.

The first of these is the matter of available information. The evolutionary history of the arid Australian biota is hardly well-understood. For only a couple of taxa are there emerging pictures of the flow of lineages throughout the arid zone as it came into existence over the past couple of million years. Numerous patterns are evident in different groups of organisms, and it does not seem possible at present to generalise about centres of evolutionary radiation; indeed, it may never be possible to do so (see Barker and Greenslade 1982; Stafford Smith et al. 1995). Nevertheless, if we did know more about evolutionary history we may be more inclined to specify more refuges than we have done in this study: we would probably emphasise the significance of various foci of biological diversity that are outlined in the early sections of this report but which were passed over when it came time to recognise refugia.

The issue of available data is reflected in ecological considerations as well as evolutionary ones. Several aspects stand out here. One is the ready recognition of many wetlands as refugia. Such systems are probably more obvious to humans than any of our other categories, and support occasional but very impressive aggregations of animals, but our rankings (if reasonable) tend to downplay them. The wetlands are undoubtedly important, but probably because they tend to support a mobile and widespread biota they do not rank highly against other types of refugia.

A further aspect emphasising the significance of data availability concerns our category of Ecological refugia. Elsewhere, we have argued that parts of the landscape rich in moisture and nutrients are likely to support important suites of organisms, which we have suggested are unfortunately liable to disruption by human activities if management is not aware of them (Morton 1990; Stafford Smith and Morton 1990; Morton et al. 1995). The Appendix to this report (section 18) discusses the possibility of identifying such refugia from satellite imagery, and concludes that this technique indeed provides great potential for this purpose. However, we still lack understanding of the possible connections between patterns of productivity at the landscape scale and biological diversity, a gap of substantial importance that is inhibiting research and management (James et al. 1995). Filling this gap will require more biological survey. To date, such surveys have not been uniformly distributed across arid Australia, and much more remains to be understood. This conclusion is applicable to all taxa, but it is particularly pertinent for plants, where there is a wealth of potential understanding that we have barely touched on here, and invertebrates, where our understanding could be greatly enhanced but which suffer from a profound taxonomic impediment.

Yet another aspect concerning data needs to be mentioned. Inspection of section 15.1 shows that the refugia we have identified are not uniformly distributed across the States and Territories. Even if we allow for discrepancies of size of these units, we must ask: is it fundamentally true that Western Australia contains more refugia than any other part of the arid zone? The answer may well be no. Our summary of refugia has without question been greatly swayed by the more extensive knowledge of arid Western Australia. In particular, it is probable that most readers would conclude – as we do – that our lists of refugia do not adequately reflect the ecological diversity of arid and semi-arid Queensland.