Land Theme Report

Australia State of the Environment Report 2001 (Theme Report)
Prepared by: Ann Hamblin, Bureau of Rural Sciences, Authors
Published by CSIRO on behalf of the Department of the Environment and Heritage, 2001
ISBN 0 643 06748 5

Accelerated erosion and loss of surface soil (continued)

Pressures (continued)

Grazing animal density in the Extensive Land-use Zone [L Indicators 1.2A and 1.2B]

  • General implications of grazing for erosion and land cover [L Indicator 1.2]
  • -->

    The arid and semi-arid regions that make up what is often described as the Extensive Land-use Zone (ELZ) (Graetz et al. 1996) occupy nearly 6 million km2, of which nearly two-thirds carries domestic stock for commercial grazing. Animal densities are very low throughout, but range from extremely low (0.5 cattle units/km2) in the less fertile western half of the continent, to low (1-2 cattle units/km2) in the more fertile eastern rangelands.

    Domestic grazing animal numbers in this vast region account for less than 13% of total sheep and 25% of total cattle numbers in Australia (Figure 10). These differences in density are associated with broad differences in productive capacity. In the western spinifex sandplains and chenopod shrublands of the west and south it takes 10-20 hectares to support one sheep (1 cattle unit or 8-10 DSE per km2), compared with the savanna woodlands and Mitchell grass plains of the north-east where the safe stocking rate is 4-6 cattle unit per km2. As a general rule, sheep predominate in the southern parts of the rangelands and cattle in the north. However, there are significant areas of overlap in the Channel Country, and in the mulga lands of north-western New South Wales and south-western Queensland.

    Figure 10: Domestic stock densities in 1997.

     Domestic stock densities in 1997

    Source: NLWRA 2001

    In the period 1990-1998 the total number of sheep declined steadily from 173 million to 121 million, whereas the number of cattle fluctuated only 1-2 million around the average of 23 million. The steady decline in sheep numbers during this period can be safely attributed to the continuing low wool prices and unprofitability of sheep-based enterprises over much of the decade (Ha and Chapman 2000). Live cattle exports to South East Asia and fattening enterprises directing stock into high-value markets in Japan and north America have influenced the higher numbers of cattle in the tropical, northern parts of the rangelands (ABARE 1999).

    There is considerable evidence that the worst pressure on much of the pastoral rangelands in recent decades occurred in the 1970s, when stock numbers were about twice the numbers in the 1990s, and before the Landcare movement was established in 1989. In many areas these pressures and earlier periods of overgrazing led to massive vegetation loss and accelerated erosion that have left landscapes permanently degraded.

    There has also been more intensive action to control feral cattle through the brucellosis and tuberculosis eradication programs in northern Australia. Feral pigs, goats, donkeys, horses, and buffalo have all been subjected to control programs coordinated through national strategies, and the rabbit calicivirus disease (RCD) has had its greatest effect in reducing rabbit populations in the more arid regions. All told it is estimated that these may have reduced grazing pressure to between 10 and 90% of the levels experienced in the 1980s.

    The National Land and Water Resources Audit (NLWRA) rangelands theme recently estimated the long-term changes in stocking numbers (and other vertebrate herbivore numbers) in the Extensive Land-use Zone (ELZ). The project will later relate these to modelled biomass production based on climate statistics and the 'Aussie-grass' pasture growth model (McKeon et al. 1990). Figure 11 shows the changes in stocking density for two selected years: 1996 and 1999. Some regions have experienced very large changes over the period.

    Figure 11: Reductions and increases in stocking densities (DSE/km2) in the Extensive Land-use Zone, 1996 and 1999.

    Sources: Hall et al. (2001), NLWRA.

     Reductions and increases in stocking densities in the Extensive Land-use Zone, 1996 and 1999
     Reductions and increases in stocking densities in the Extensive Land-use Zone, 1996 and 1999
     Reductions and increases in stocking densities in the Extensive Land-use Zone, 1996 and 1999
     Reductions and increases in stocking densities in the Extensive Land-use Zone, 1996 and 1999

    In 1993 a very thorough review was undertaken to assess the economic viability and ecological sustainability of the rangelands (Wilcox and Cunningham 1994), following a comprehensive survey of the record the condition, productivity and sustainability of pasture lands of northern Australia by Tothill and Gillies (1992). Both reports identified widespread vegetational degradation and other ecological changes, such as extensive weed invasion, rill and gully features, soil salinisation, and bare scald extension linked to overgrazing.

    Part of the degradation in the Northern Territory and Western Australia, such as that in Victoria River district and Ord River catchments, is largely the result of earlier periods of mismanagement. Eradication campaigns against bovine brucellosis and tuberculosis in the 1980s and 1990s have controlled animal numbers, and reduced feral animal populations. Unfortunately, recent surveys (e.g. ABARE 1999) suggest that the higher-productivity regions of the central and eastern rangelands are still being used unsustainably, with consequent continuation of vegetation and land degradation. This is occurring particularly where the more productive regions are fragmented topographically within larger areas of low productivity.

    Before the National Land and Water Resources Audit, the principal aim of land condition surveys was to assess rangelands from the perspective of grazing industry utilisation, rather than ecosystem integrity. Nevertheless these surveys show clearly that land, where there are the signature indicators of severe soil erosion, such as gullying, bare salt scalds, complete loss of topsoil ('Class 3' land) and large tracts of land where there is often massive weed invasion largely through overgrazing of more palatable herb and grasses ('Class 2' land) cannot support continuing 'business-as-usual' pastoralism, from an economic as well as ecological perspective (Tothill and Gillies, 1992, Johnston et al. 1996, Water and Rivers Commission 1997).

    Table 5 compares some of the features from the combined Tothill and Gillies (1992) and Wilcox and Cunningham (1994) studies with those undertaken for the National Land and Water Resources Audit and other recent studies.

    Table 5: Comparison of animal numbers and estimates of rangeland land degradation, early and late 1990s.
    Region Major vegetation types 1993 A estimated area of land degraded 1993 B stock animal numbers Change in stock animal numbers 1996-1999 C 1997-1999 estimated area of land degraded
    KimberleyPilbara Eucalypt woodland, various grasslands, Sorghum australiense, spinifex with bare ground 1 6% severe gully/sheet erosion, 20% degraded vegetation 643,000 cattle, some sheep in central-south, (WA Bushlands) Cattle densities unchanged between 0.4-5.4 /100 ha. Sheep few, densities have declined to 0.5/100 ha 20-25% overgrazed in Timor Sea drainage division: 50% overgrazed in Indian Ocean division D
    Darwin-Gulf Kangaroo-grass, perennial sorghum grass, spinifex, some Mitchell grass patches 232,000 cattle Cattle no change (0.4-3/100 ha). Sheep declined from 5-0.5 /100 ha No more recent information available
    Victoria River and Barkly Tableland Mitchell grass and bluebush, soft spinifex 2.5-5% severely affected. Lack of data elsewhere 384 000 cattle (Victoria River)457 000 cattle (Barkly Tableland) Cattle increase from about 5-10 /100 ha.Sheep down to Death of some Mitchell grass due to drought in NW F
    Alice Springs Spinifex predominates; some mulga with annual grasses Severe bare areas in Todd River plains; Eragrostis degraded 297 000 cattle No change in cattle densities (0.4-1.5 /100 ha). Sheep static or less No more recent information available
    Cape and Gulf Low eucalypt woodland with various grass types Declining grassland condition, cover and type 586 000 cattle Cattle no change (average 3/100 ha). Sheep decreased from 5-0.5/100 ha No more recent information available
    North East Uplands, Queensland Eucalypt woodland, brigalow, bluegrass treeless plains. Black speargrass, Aristida spp. , some Mitchell and soft spinifex 8 out of 10 grass types more degraded over 1980-1990s 1.2 million cattle and 2.2 million sheep Cattle some increase (6-10/100 ha) big range in sheep but increasing, from 6-34, to 10-50/100 ha Aristida sp. communities declined 1992-1997; prickly acacia (A. nilotica) now covers 28% of Mitchell grass shires F
    Western Plains Queensland Extensive grasslands, mulga, gidgee, other Acacia woodlands, spinifex 18 pasture types, with 6 degraded, 5-50% degraded by erosion, parthenium and prickly Acacia weeds 2 million cattle and
    8 million sheep
    Cattle no change (1.5-3), sheep decreasing, from 2-10/100 ha to 0.5-6/100 ha No more recent information available
    Western Division NSW (Far Western Plains) Mulga woodland, box and cypress pine, mallee woodlands, floodplain grasses, saltbush and bluebush 11 systems, all showing soil erosion, weed invasion, rabbit warrens and vegetation depletion D 137 000 cattle and
    5 million sheep
    Cattle have doubled in number, sheep down by 25% E Kangaroos more than monitored. G50 000 ha woody weed cleared
    (20 000 per year) I , 202 000 ha cleared of warrens. ELess than 5% vegetation types in reserves I
    WA Bushlands (south and east of Pilbara to Nullarbor) NW-SE tussock grasses and acacias, mulga, mallee eucalypts, chenopod shrubs 2.5 million sheep,
    few cattle
    Low densities and few cattle, sheep no change (0.5-3.5/100 ha) D
    Northern South Australia Low shrublands and mulga to north, saltbush, bluebush, grasses and acacias to south Up to 30% of historically degraded vegetation, but most regions better in 1990 than in 1970 from controls 137 000 cattle north of dog fence, 1.3 million sheep south of dog fence Cattle increase (0.4- 1.4/100 ha), sheep increase, from 0.5-2.1 to 2.1-4/100 ha 20-30% still degraded from past effects. Only slow recovery despite destocking because poor seasons H

    A Wilcox and Cunningham (1994).
    B Tothill and Gillies (1992).
    C Hall et al. (2001), NLWRA rangeland grazing pressures project (unpublished).
    D Waters and Rivers Commission (1997).
    E Kerin and Hyder Consulting (2000).
    F State of Queensland (1999).
    G Pople and Grigg (1999).
    H SA Department of Environment, Heritage and Aboriginal Affairs (1998).
    I EPA NSW (2001).

    Cattle and sheep densities show the general trend of increased cattle densities in the northern, more fertile regions, but decreasing sheep densities in most areas. Low-productivity regions such as the Western Australian Bushlands and northern South Australia have maintained low or declining densities. More recent information on the condition of rangelands in the north and eastern parts of the rangelands will be welcome when the NLWRA rangeland project on historical trends in grazing density and pressure is completed, as little additional information has been forthcoming from State of the Environment reports for Queensland (State of Queensland 1999) or New South Wales (EPA NSW 2001).

    The effect of RCD has been very great in the southern half of the rangelands where rabbits have been reduced to less than 10% of their former numbers. The challenge now is to sustain this improvement and to extend the effect into the higher rainfall regions.

    The issue of how vegetation type affects the susceptibility to degradation and erosion pressure has been the focus of concern in the animal grazing industries and among conservation ecologists, particularly in north-eastern Australia, where woody weeds are a serious concern to pastoralists (see Introduction of novel biota into native habitats and communities). Many grazing properties in this area have experienced a large increase in woody weeds over the past two to three decades; 50% of cattle properties in Queensland have an average of 2000 hectares of woody regrowth (ABARE 1999). On these properties more than 90% of the land grazed is woodland of some type or other and is described either as forest or woodland by the National Forest Inventory (see Agricultural best practice).

    Pastoralists consider it essential to clear some timbered land to maintain grazing productivity, and 63% considered their productivity has declined as a result of woody regrowth in ABARE's (1999) survey. Nevertheless, woody regrowth in many cases takes place as the result of current or past grazing practices, and occurs predominantly in native woodlands that are grazed. Whether such regrowth is regarded in a negative or positive light is a matter of value judgement, reflecting the diversity of opinion that exists on this matter.

    Regeneration of trees is restricted where animals graze in woodlands, as seedlings are grazed out. A reduction in grazing pressure has been associated with rapid regrowth of shrubs and trees when sheep have been removed (Cooney 1995). The Regional Forest Agreement process is redefining grazing lease areas in some areas as old growth forests in NSW and Queensland, leading to strong adverse reaction from graziers (The Land 2000). This provides a telling example of the tensions that arise from different views about how these forests and woodlands should be used.

    General implications of grazing for erosion and land cover [L Indicator 1.2]

    Vegetation and land cover increase when total grazing pressure is reduced by a combination of low commodity prices, vertebrate pest control and good seasons. The increase in cropping area during the 1990s was less than many believe, and there has been some substitution of cattle for sheep in some regions. On the whole, adjustments have affected the degree to which more marginal land within each SLA or farm is used or not, rather than reducing grazing pressure equally everywhere.

    One of the main reasons that we cannot assess the impact of grazing animals adequately is the uncertainty of ABS statistical data on pastures.

    Grazing management represents the single biggest, and least costly method of managing vegetation cover. It is probably the most important management tool in combating most land degradation problems. Well-controlled grazing greatly assists in increasing tree and shrub regeneration, maintaining a balanced assemblage of grasses and herbs in the understorey, and controlling weeds. Unfortunately, the level of animal control needed (particularly through fencing off remnant vegetation, watercourses and overgrazed areas) is frequently not available.

    The current debate about grazing and land management hinges on whether total grazing pressure is a useful management tool when compared to concepts of safe thresholds for specific ecosystems (McLeod 1997).

    In rangeland ecology around the world, the concept of carrying capacity is often disputed, because it leads to a more static management approach than is required in regions of highly erratic rainfall and patchiness of vegetation. The 'state-transition' see-saw is regarded as a more representative analogy of the way in which rangelands operate. This concept has led to the development of an 'adaptive management' approach of action-learning and partnership between pastoralists and scientists that is proving very successful. Nevertheless, in recent government inquiries into rangeland condition, both pastoralists and scientists have continued to use 'safe carrying capacity' to evaluate whether stocking densities are leading to land degradation or loss of productive capacity (Johnston et al. 1996).

    In more reliable rainfall regions, pasture management includes inputs such as fertiliser, herbicide control of weeds, sowing of exotic species of high palatability and nutritional value, and use of animal grazing to control pasture composition. In the mixed farming regions, pastures are rotated with crops to maintain soil fertility and break cycles of disease. In high rainfall districts, pastures are permanent and inputs vary according to the roughness of terrain, value of land and animals, and proportion of off-farm income. In all these environments, the stocking rate is highly dependent on the level of inputs and sophistication of the grazing management (Sustainable Grazing Systems 1998).

    Queensland and Northern Territory specialist cattle properties have performed more profitably than other specialist cattle producers in mixed farming or high rainfall regions in recent years (ABARE 1999). This has occurred because of their geographical advantage in the live cattle export trade to Indonesia and Singapore, which survived the vicissitudes of the Asian economic crisis. However, apart from the large corporate farms, average farm business profits have been negative in all areas through much of the decade (ABARE 1999). In the most difficult cases, where pastoralists have been unprofitable for over a decade, structural adjustments have been proposed in which leaseholders would become environmental stewards, with a shift in emphasis to enterprises such as ecotourism and conservation. Consultations with pastoralists have demonstrated their interest in these proposals, but land title acts are often restrictive and new legislation may be required to alter the primary activities on many rangeland leases.

    Large amounts of money and volunteer effort have gone into land restoration using labour-intensive tree and shrub planting schemes, contouring, and weed control. But until recently very little has been done to destock, fence out and encourage vegetation regeneration through reduction in grazing pressure. Some Catchment Management Authorities in Victoria are now discussing the possibility using a variety of market-based mechanisms to encourage landholders to manage land for environmental outcomes, such as trading salinity credits for destocking (DNRE Victoria, pers comm.). As yet there are few other regions where this option is being taken seriously.

    Nationally, the impact of grazing pressure continues unabated except on lands that have been converted either to conservation reserves, or to more intensive uses not involving stock, such as viticulture and horticulture. While there have been some recent proposals to destock vast tracks of the rangelands (Archer et al. 1997), the feasibility of this is questioned.

    In Western Australia the Department of Conservation and Land Management has fenced off the Peron Peninsula across its 3 km width and systematically cleared it of nearly all foxes and most cats with a four-year control program. Now the reintroduction of native fauna is possible. Similar exercises have been undertaken over small areas in South Australia, New South Wales and Queensland. However, the cost of such an exercise makes it prohibitive for large areas.

    It is widely accepted in all pest, weed and disease management strategies that total eradication is an almost impossible and highly expensive operation that cannot be undertaken other than in the most threatening situations. What we must consider is where the most damage is occurring, both from a land degradation point of view and a biodiversity point of view. This issue is discussed in more detail in the Biodiversity Theme Report.