Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

1. Australia's arid zone: Geographical setting

D.N. Parkes

General climate

"In its sense of desolate or uninhabited, a desert expresse a climatic limitation on life, mainly through aridity" (Mabbut 1978, p. 113). A number of factors combine to produce what Mabbutt has called a "moderate desert regime" in Australia, one of moderate aridity that is apparent in the general mantle of vegetation extending over areas far removed from the watercourses (Mabbutt 1978, p. 113). Seasonal rainfall contrasts in arid Australia are smaller than in most deserts. However, in central Australia within a 500-km radius of Alice Springs, summer rainfall is dominant, and in some years this can be quite extreme, especially when associated with southward penetration of monsoonal flows from North Australia.

FIG. 1.1. Average Annual Rainfall of Australian Arid Zone, in mm (adapted from Gentilli 1978 and Australian Bureau of Meteorology). (The arid zone is delimited by the thick line and follows the delimitation adopted in Slatyer and Perry 1969.)

A study of human thermal climates of Australia has recently been published by Auluciems and Kalma (1981).

The central Australian arid zone has a mid-winter afternoon heat stress (HS) factor which is among the lowest in Australia. In effect, it is a potentially attractive climate. In summer, however, the arid region experiences some of Australia's highest heat stress values. However, although very high heat stress "may be reached during summer . . . providing that garments may be freely discarded, homeothermy in average warm weather conditions may be maintained" (Auluciems and Kalma 1981, p.10).

As Auluciems and Kalma demonstrate, most of Australia's arid zone is a high energy-dependent area. Growing numbers of urban people live in this region but only with a high level of technologically advanced, fossil-energy-consuming support, whereas traditional occupation of Australia's arid zone probably never exceeded 50,000. The absence of large permanent settlements and the use of simple tools and human energy prior to European occupation are testament to the need at that time for small-scale, highly mobile social groups able to respond quickly to the changing circumstances of life imposed by the desert climate, in effect, to be adaptable-and one mechanism of adaptation was movement. There still are alternatives to dependence on "energy-intensive, technological solutions" to the problems of growing urban settlement in the arid zone. They include, inter alia, passive design housing and commercial structures, use of natural vegetation, and appropriate timing of activities (Auluciems and Kalma 1981, p.22). In places such as Alice Springs the better timing of activities and improved housing designs may well prove to be essential if the rapid growth in the population is not to be allowed to escalate demands for energy and water. Gibbs (1969) and Gentilli (1978) are recommended as general references on the climate of Australia's arid zone.

General Physiography and Drainage Setting

Figure 1.2 illustrates the six main physiographic types of desert identified by Mabbutt (1969, p.22).

The sand desert type dominates, and apart from the shield desert of the Lander-Barkly plains to the north-east of the central Australian mountain and piedmont type, sand deserts more or less completely surround the region in which Alice Springs is located, further complicating construction and increasing the cost of road and rail transportation. More or less centred on Alice Springs, the central Australian ranges are one of a number of mountain and piedmont desert regions in Australia. "This desert type is the source of most of [the] desert drainage and of its water supplies and was a favoured habitat for the desert aboriginal. Unfortunately, no Australian desert mountain rises high enough to yield significantly greater rainfall nor to store precipitation as snow, and so provide an exception to the ephemeral character of Australian desert rivers" (Mabbutt 1969, p.23).

Apart from the Warrago and Darling rivers, which more or less mark the south-eastern border of the arid zone (excluding semi-arid), and the much shorter Gascoyne, Ashburton, Fortescue, de Gray, and Fitzroy rivers of the Indian Ocean drainage division in Western Australia, there are no perennial rivers throughout this vast region. Uncoordinated, intermittent drainage pervades the environment, the principal source of water being from ground-water bores. Groundwater resources are in the aquifers of the sedimentary basins (Fisher 1969, p.63).

For Alice Springs, the central Australian ground-water source is based on the Amadeus Basin to the south of the town. The gneissgranite hills of the Arunta block to the north of Alice Springs separate the Amadeus from the Ngalia Basin. According to MacQueen (1980), enough low-yield bores would tap a large water potential. Most of the recharge for aquifers such as those of the Amadeus Basin appears to "take place over the fractured sandstone and limestone outcrops . . . [whereas] recharge resulting from the rivers appears to be much less." River basin recharge is probably local and short-lived (MacQueen 1980).

Heathcote (1973, p.17) draws attention to the recency of drought studies in Australia and suggests a number of reasons related to problems of definition, measurement, and accuracy of reporting or registration. He also stresses the importance of perception factors. In drawing attention to Saarinen's work (1966) in the Great Plains of North America, he demonstrates, however, that "as aridity increased, so did the farmer's estimates of the possibilities of drought occurrence" in contrast to an alternative hypothesis that socalled "folk droughts" are perceived to occur with less frequency due to the old adage that "familiarity breeds contempt." Recent though the careful scientific study of drought may have been, proposals for intervention-for instance in those for adjusting the flow of Australia's rivers-have been around for two generations.

About 50 years ago (1930) Mr. J.J. Bradfield proposed a massive scheme for water conservation and irrigation in Queensland and central Australia. Serious deliberations over the scheme were pursued in the early post-war years, but a report by Mr. H. N. Warren, the Director of Meteorological Services, on rainfall and climatological aspects of the scheme concluded that "concerning the Finke River scheme, (central Australia) considerable doubt is felt . . . The comparative infrequency of high rainfalls which would cause streams to flow, and the brief periods during which intense rain with consequent 'flow' are experienced, suggest that this scheme may, upon close survey, be found impracticable."

FIG. 1.2. Desert Physiographic Types (Mabbutt 1969, p. 23). The arid zone limit, as shown on this map, differs from that used elsewhere in this report, in particular because it excludes the eastern and semi-arid regions of Queensland and New South Wales. The arid zone limit, in northern Australia, is also drawn further south than the limit used elsewhere (cf. figs. 1.3 and 1.4, for example).

During the years 1941-1945 the scheme was thought by many to have considerable national importance, but the majority opinion of the committee of meteorologists was that even if it were possible from a hydrological engineering point of view to fill Lakes Eyre and Frome, Bradfield's claim of climate modification was not acceptable. This and other decisions not to proceed with the scheme followed comparative study of research in Szechwan Province, the People's Republic of China, the Nile Valley, and Lake Victoria-Nyanza, of proposals to fill the Qattara Depression in North Africa, and evaluation of the climatic effects of the Salton Sea on South California. In 1944 the Rural Reconstruction Commission also examined Bradfield's proposals and recommended that they should not be implemented.

Concern with Australia's inland water supply is not over. In December 1979, a leading article in a national Australian newspaper (The Australian, 15-16 December 1979) advocated "taking life to the Dead Centre" and proposed the re-establishment of an authority such as the Snowy Mountains Authority. The critical thrust of the article was that the real shortage the world will face in the future is not fossil fuels-for which there are substitutes-but water. The article concluded that "the development of The Great Interior is the biggest challenge of the 80's for Australia." On 25 May 1981, Sir Barton Pope (in The Australian) wrote, "Our national life line is being strangled while politicians idly stand by and watch its death pangs. The Murry [River] has ceased to flow. Health is at risk." He advocates reappointment of inland irrigation schemes and development of salt-resistant grains, and points to the paradox that "we gloat over the forthcoming wealth of our new found mineral resources while we neglect the resources in shortest supply-WATER."

FIG. 1.3. Major Sedimentary Basins for Ground-water Supply

A key factor militating against the realization of this challenge is the lack of a ministry responsible for arid zone affairs and the politico-geographic fact that each of the mainland states has a border contiguous with other state borders in the arid region. Rational, integrated policy is hard to achieve. In association with small absolute population numbers, the political facts of contemporary society do not suggest promise of much change for a very long time. Nobody traveling in central Australia goes any distance from a settlement without a good supply of water. But how long will it be before there is no longer a supply of good water in the settlements?

Land Tenure, Mining, and Transport

At present Australia's political and economic energies appear to be directed to exploitation of mineral resources and the associated mining process. The existing geographical facts of mineral resource distribution therefore become the principal determinants of the use which will be made of Australia's arid lands in the foreseeable future. Almost every ore site is potentially a human settlement of considerable scale, say 20,000 people. The continuing dominant context, however, is still pastoralism.

One of the most awkward issues to be confronted is that of land tenure, including, inter alia, the rights of traditional owners, the pastoralists, the capital-intensive mining operations and their employees, and the private and public sector support activities that form the infrastructure necessary to satisfy the wants and needs of contemporary urban residents (chapter 6). Heathcote (1969) has presented a valuable contribution to the better understanding of this important aspect of the geographical setting of Australia's arid lands. His classification of land tenure types is still of value, but there is now a need for a comprehensive study of the existing situation with a view to facilitating arid land management in the years ahead, especially as the mineral discoveries of the past 15 years move towards full production and a responsive settlement system begins to take shape.

Recently, Young (1982, CSIRO, Deniliquin) has discussed relationships between land tenure and arid zone land use. In this excellent paper he draws attention to the current attitude prevailing "throughout Australia [that] there is a general belief that all land users should act as stewards of the land" and develops arguments against nine prevalent misconceptions about land tenure. Four of the nine misconceptions are summarized below.

1. Perpetual leases are 99-year leases. There are in fact no 99-year leases in arid Australia. In New South Wales, for instance, perpetual leases are issued to the "lessee, his heirs and assigns forever." In Western Australia and in the Northern Territory most pastoral leases are issued for 50 years.

2. Multiple use should be encouraged in the arid zone. Young states that "multiple use involves the simulataneous use or management of a single area of land for two purposes by one person. This rarely occurs in the arid zone. However, joint land use, the simultaneous use of land by different people is common. Multiple use requires the exclusive right to control all land uses within an area to be vested in one person."

3. The attachment of covenants to leases has enabled governments to effectively administer arid lands. Young demonstrates that there are many anomalies. Effective convenants, obliging a lessee "to do or not to do something" and therefore enabling effective control of land use, are not consistent across the states. Synchronized review of covenants every 15 years only occurs in Western Australia; elsewhere inter-review periods may be anywhere between 20 and 30 years. The presence of lease covenants is not sufficient to enable effective land administration.

4. Term leases retain more future options than perpetual leases. The retention of "future options" is usually taken to mean that the Government should retain the right to change the ownership of rights to use land. However, it is the provisions within each lease and not their term that determine the degree to which future options are retained. It is not important whether the lease is called a term lease or a perpetual lease: the use made of the land is bound up in the provisions of the lease.

The cost of waiting until leases can be reviewed and costs of compensation can be very high; resumption is thus likely to be cheaper than waiting for termination. Young concludes, "If Governments are serious in their attempts to prevent land degradation . . . the people who design these systems must expect land users to put their interests first and those of the land second."

The relevance of this brief excursion into land tenure is to demonstrate that, with increasing urbanization of the Australian arid zone, new factors will need to be taken into account in the determination of covenants and the synchronization of reviews. Appropriate and farsighted legislation regarding future use of arid lands is also critical.

There is "an ultimate potential for mineral discovery that would far exceed that of most countries, but the handicap of distance where kilometres must be measured across a harsh and inhospitable environment has discouraged the search for all but the most valuable minerals in the past, and the development of all but the most extensive in reserves at present" (Duncan 1978, p. 435).

The central Australian region around Alice Springs occupies about 603,500 kmē Gold and copper at Tennant Creek and tungsten, tin, and bismuth at Hatches Creek have been mined for many years, but recent closure of some mining operations in Tennant Creek clearly illustrates the potential ephemerality of settlements. At Mereenie-Palm Valley, about 200 km southwest of Alice Springs and part of the Amadeus sedimentary basin, a small oil and gas field has been in operation for some years. The 1980 Northern Territory Assembly elections included an Australian Labor Party (ALP) policy proposal to buy a quarter share in the Mereenie partner, Magallen Petroleum, with a view to converting "most of the Territory's electricity generation to natural gas fuel" (John Isaacs, Opposition Leader in the Northern Territory Assembly). The intention was that Alice Springs would be the first place to benefit; an A$10 million pipeline would be constructed. Oil production could continue on the basis of present proven reserves for 30 years, and Magellan Petroleum had "plans to process 5,000 barrels a day in an Alice Springs refinery . . . [and such plans were] in line with Labor's policy" (John Isaacs, Report in Centralian Advocate, 10 April 19801. However, natural gas is seen as the key to the energy demands of central Australia's arid zone, with solar energy being used increasingly to convert publicoffice-building air-conditioning systems, and also incorporated into a scheme for solar hot-water systems in private dwellings. At 1980 prices, the solar water heating scheme would cost each purchaser A$35 a quarter until the cost of the plant was paid. In spite of the climatic potential for solar energy technology in the Australian arid zone, utilization to date has been very small.

FIG.1.4. Population Mean Centres, 1971 and 1976, and Major Settlements in Australia's Arid Zone

Transportation difficulties remain an important component of the growing arid zone urban economy, and increases since 1974 in the price of oil fuels have exacerbated the problems. North and south between Darwin and Adelaide there is neither a continuously sealed highway nor a through railway link. The only east-west rail link skirts the southern limits of the continent between Perth and Adelaide across the Nullabor Plain. A sealed road parallels the main route. In 1980 the Federal Government committed itself to spending A$400 million on the construction of a rail link north, between Alice Springs and Darwin. The very low population densities and large distances have also caused the rise and fall of many small airlines. Recently in the Northern Territory (January 1981) a key transportation network serviced by Northern Airlines ceased operating. Such interruptions to movement further emphasize the small resident population potentials (fig. 1.4), akin to those in western desert areas of the United States over 100 years ago.

Population and Settlement Settings

About 97 per cent of the Australian population live on the humid coastal fringe, and the majority probably have negative attitudes to living in the hot, dry desert areas of Australia. But possibly more significant than these negative attitudes is the likelihood that the majority would have little interest in what does or should happen there. The political consequences are far reaching.

Adopting a definition of the arid zone used by the CSIRO (Slatyer and Perry 1969) and after correction for population census underenumeration, the estimated population resident in the Australian arid region was 478, 417 in 1976 (fig. 1.5). The arid and semi-arid zone occupies about 75 percent of the continental area of Australia, that is, about 5.7 million kmē out of a total 7.6 million kmē Between 1971 and 1976 this population grew by about 2.8 per cent. If this population of 478, 417 people were evenly distributed across the region, any one person would be about 3.2 km away from a nearest neighbour. One's eight nearest neighbours would be distributed over a subregion of about 110 kmē This means that the average density of people over the 5.7 million kmē is about one person (including children of all ages) for every 10 kmē and this may be compared with about 272 people for every 1 kmē in the Sydney statistical division (1976) of 3.02 million people and 11,710 kmē

The mean centre of the population of Australia's arid region is in South Australia at the apex of a triangle which has its other two base points at Oodnadatta and Coober Pedy (fig. 1.5), but its geographical centre lies approximately at Papunya, 200 km west of Alice Springs. Between 1971 and 1976 the population mean centre had shifted north and west, reflecting the growth of population in the north-west coastal and Pilbara region of Western Australia and the growth of Alice Springs.

FIG. 1.5. Population Potentials and Mean Centres for Total Australian Population, 1947 and 1976

Another characteristic of the population and settlement setting can be illustrated by the pattern of population potentials. The population potential at a point is a measure of the proximity or "accessibility" of that point to all other points (or places) in the system, weighted by population size. The model is rather crude, but has certain value in illustrating the changing relative accessibility of points within a single system (nation) over a number of years.

Figure 1.5 illustrates Australia's population potentials for 1947 (Neft 1966), and values have been computed and mapped for 1976. The original calculations for 1947 potentials were based on Imperial distances (miles); for 1976 metric units have been used. To assist comparison, values for 1947 have been converted to their approximate metric equivalent. Also shown in each of the population potential maps is the mean centre of the total Australian population. Again, there is north-west drift. Allowing for different methods of interpolating the lines of equal potential from the different centroid and area bases used, there seems to be sufficient comparability from one period to another to accept that this direction of drift deeper into the arid zone is essentially correct.

Central Australia, as shown in figure 1.6, includes two of the environmental regions that have recently been proposed in a joint study by the CSIRO and the Federal Department of Environment and Science (taut et al. 1980). This region extends over 603,559 kmē and within it there was an estimated resident population of about 25,800 people in 1976. Approximately 66 per cent of this population resides in the two urban centres of Tennant Creek (2,327) and Alice Springs (14,729). Together these two urban populations occupy less than 0.3 per cent of the central Australian region.


Road/Region Conditions and Advice
Stuart Highway South: Open to all vehicles with care
North: Passable with care
Kulgera All roads passable
Plenty Highway Passable with care to plenty crossing
East of plenty crossing not recommended
Glen Helen Haast Bluff: Impassable
Ross River Passable with care to Ross River
Beyond not recommended
Trephina Gorge Passable with extreme care
Ormiston Gorge Closed due to flood and road damage
Finke Gorge Closed
Altunga Road 4-wheel drive only
Papunya Main road impassable at the Derwent
Dirt roads 4-wheel drive only
Ti Tree Bitumen passable with care
Dirt roads 4-wheel drive only
Yuendemu Dirt roads passable with care
Ayers Rock Road Passable with care
Hermannsburg All roads passable except Palm
Valley road
Lake Nash Not recommended: Georgina River 2.5 metres
Warrabri Entrance road, passable with care
  All other roads 4-wheel drive only

FIG. 1.6. The Central Australian Region

FIG. 1.7. Northern Territory Southern Region Road Report, February 1981 and Associated Synoptic Weather Chart

The region is not only remote but susceptible to periodic disruption of its road and rail communications, especially during a summer "wet." The copy of the Southern Regions Road Report, made available by the Northern Territory Department of Transport and the Territory Emergency Services in Alice Springs (February 1981), and the associated synoptic weather chart (fig. 1.7) illustrate the periodic susceptibility of the region to extreme dislocation. Remoteness is therefore a periodically variable attribute.

While the distributional characteristics of the population are an important contextual factor in their own right, there are also demographic and other ecological attributes of that population which act to construct the context of life in the Australian arid zone. They are the source of a range of attitudes and behaviours which of themselves influence the use and perceptions of the arid lands. In order to find out whether there is any territorial variability in the demographic and ecological structure of the arid zone population, 25 variables were selected from the 1976 Australian census. Table 1.1 presents the 25 variables and their associated values. Local government and equivalent areas for the Northern Territory were used as an areal base for a preliminary screening of demographic and (implied settlement) ecological structure. Cluster analyses of various sorts and a principal components analysis (table 1.2) suggested five underlying dimensions and clusters of human ecological region types or "settlement." Figures 1.8 (a) and (b) shows maps of the distribution of scores on the first and second components.

TABLE 1.1. Summary Statistics for Arid Zone Settlement in 1976 (Alice Springs values are shown in parentheses)

Variable Mean   Co-efficient of variation Smallest value Largest value
%ABOR 1 11.89 (11.13) 1.50 0.00 80.10
%NOTMARI 19.83 (18.40) 0.30 10.70 61.00
%DEPKIDS 30.24 (33.20) 0.16 3.80 43.60
%NOTMOVED 58.36 (44.40) 0.22 19.90 82.50
%INCOME 4.65 (9.70) 0.76 0.00 19.20
%PENSIONS 17.10 (9.90) 0.38 2.70 32.80
%SELFEMPE 10.56 (4.00) 0.69 0.20 32.10
%UNEMPLOY 1.92 (1.30) 0.54 0.00 5.20
%IMPROV IS 4.19 (1.00) 2.06 0.00 60.80
%NO WATER 4 50 (2.30) 1.62 0.00 56.80
%NO CAR 14.13 (9.90) 0.73 0.70 56.60
MF RATIO 1.33 (1.06) 0.90 0.93 14.03
AGE/EMP 14.66 (7.20) 0.56 1.04 50.19
%SOLAR 3.18 (0.22) 2.68 0.00 80.00
%FOR SALE 2.81 (3.20) 1.51 0.00 26.70
%GOVEMP 24.40 (39.30) 0.52 0.00 76.50
%MINING 7.49 (0.40) 1.87 0.00 66.00
%CONSTRUC 8.14 (12.10) 0.56 0.00 20.60
%TRANSPRT 4.73 (9.50) 1.05 0.00 34.00
%COMMUNIC 1.68 (1.00) 0.64 0.00 4.80
%FINANCE 2.04 (5.50) 0.86 0.00 10.40
%HOTELREC 4.44 (7.00) 0.49 0 00 12.50
%WALLMAT 42.84 (9.80) 0.52 0.40 85.66
%SHARED 2.44 (1.90) 1.47 0.00 26.90
%METALMER 2.39 (1.00) 3.47 0.00 66.40

Key to variable names in same order:

ABOR I Aboriginals and Islanders
NOTMARI Not married over 15 years of age
DEPKID Dependent children
INCOME Income over A$12,000
PENSIONS All types of pension
SELF EMP Self-employed
UNEMPLOY Unemployed
IMPROVIS Improvised dwellings
NO WATER No piped water
NO CAR No car
MF RATIO Ratio of males to females
AGE/EMP Ratio over 65 to total employed
SOLAR Solar energy in private dwelling
FOR SALE Private dwellings for sale
GOVEMP Government employees
MINING Mining employees
CONSTRUC Construction employees
TRANSPRT Transportation employees
COMMUNIC Communication employees
FINANCE Finance employees
HOTELREC Hotels, recreation, restaurant employees
WALLMAT Fibro and metal dwellings
SHARED Shared facilities, dwellings
METALMFR Metal manufacture employees

It is not necessary to go into more detail here. These two components alone suggest that the Australian arid zone is in fact structured along a number of demographic and ecological dimensions. Further study of these and other dimensions will lead to a better understanding of the patterns of settlement structure. The first component to emerge in both the cluster and the principal components analyses was an Aboriginal-housing-low-mobility component, indicating "arid-poor regions." The second suggests a mining-higherincome-population-change component, indicating "arid-rich regions."

This chapter has provided a description of some of the key components among the physical and human attributes of the Australian arid zone. The following chapters emphasize the human ecological features and focus on Alice Springs.

TABLE 1.2. Principal Components of 25 Variables, Australian Arid Zone Settlements, 1976

Variable code and number Sorted Rotated Factor Loadings
F 1 F2 F3 F4 F5
% ABOR/IS 3 0.90        
% NO CAR 13 0.89        
% IMPROVIS 11 0.88        
% NO WATER 12 0.82        
% SHARED 26 0.64        
% MINING 19   0.80      
% NOT MOVED 6   -0.74 -0.28 0.38  
% INCOME 7   0.73      
% SELF EMP 9 -0.44 -0 57     0 37
ME RATIO 14     0.89    
% NOTMARI 4     0.85    
DEPKIDS 5     -0.78    
% METALMFR 27     0.65 -0.29  
AGE/EMP 15   -0.39   0.75  
% PENSIONS 8   -0.37   0.75  
% COMMUNIC 22 -0.26     0.61  
% HOTELREC 24   0.39   0.52  
% GOVEMP 18         0.88
% TRANSPORT 21         0.78
% CONSTRUC 20         0.65
% FINANCE 23          
% FOR SALE 17          
% UNEMPLOY 10          
% WALLMAT 25 0.47        
% SOLAR 16 0.49        
Variance   4.45 2.86 2.83 2.41 2.27

The above factor loading matrix has been rearranged so that the columns appear in decreasing order of variance explained by factors. The rows have been rearranged so that for each successive factor, loadings greater than 0.5000 appear first. Loadings less than 0.2500 have been replaced by a blank.

Variable code is explained in table 2.4.

FIG 1.8(a). Upper Quartile of Subareas in the Australian Arid Zone: Dimension l; Arid-Poor Zone: Aboriginal Population Concentrations, Low Status Housing and Limited Mobility

FIG. 1.8(b). Upper Quartile of Subareas in the Australian Arid Zone: Dimension Il; Arid-Rich Zone: Mining-Based Resource and Population Growth

Contents - Previous - Next