Biodiversity Theme Report

Australia State of the Environment Report 2001 (Theme Report)
Prepared by: Dr Jann Williams, RMIT University, Authors
Published by CSIRO on behalf of the Department of the Environment and Heritage, 2001
ISBN 0 643 06749 3

Biodiversity Issues and Challenges (continued)

Disturbance Regimes and Biodiversity (continued)

Clearing, Fragmentation, Degradation of Native Vegetation or Marine Habitat

The conservation of native vegetation is vital to biodiversity conservation. Vegetation is a key element of biodiversity in Australia since it comprises tens of thousands of plant species, thousands of vegetation communities and assemblages, and provides habitat to myriads of microorganisms and animal species. Native vegetation is also integral to the functioning of landscapes and ecosystems. As a consequence, the clearance and modification of native vegetation has several effects far beyond losing trees, shrubs, grasses and seagrasses.

Terrestrial Vegetation [BD Indicators 2.1 and 2.2]

European occupation of the Australian continent has resulted in significant changes in the extent and condition of native terrestrial vegetation. Clearance of vegetation for agriculture in the higher-rainfall regions, as well as those with more fertile soils and close to settlements, was promoted by governments. In all coastal regions of eastern Australia, this activity resulted in the removal and major modification of many vegetation communities including grassland, mallee and closed forest ecosystems. In some instances, the clearance of regional vegetation communities approaches 100%. Continental studies of landscape features such as biophysical naturalness (estimate of the extent to which regional plant and animal communities have been disturbed by modern technology), level of disturbance (Figure 7) and wild rivers (Figure 8) indicate that much of continental Australia has been highly modified by human activities, including areas that until recently have remained relatively inaccessible to Europeans.

Figure 7: Extent of land disturbance in Australia
Criteria for assessment include remoteness from access, remoteness from settlement, apparent naturalness and biophysical naturalness. (Refer to Lesslie et al. (1995) for methodology).

 Extent of land disturbance in Australia

Source: Environmental Information Resources Network

Figure 8: The River Disturbance Index
This is the average of the Flow Regime Disturbance Index and Catchment Disturbance Index (from the Australian Rivers and Catchment Condition Database, ARCCD). Wild rivers are shown in blue (dark blue for rivers with no disturbance and light blue for rivers with a disturbance up to 0.01, a threshold set by the Commonwealth Wild Rivers Program). All other rivers are shown with their respective disturbance level (in order of increasing disturbance: green, brown, orange, pink).

 The River Disturbance Index

Source: Environmental Information Resources Network

The broad land use types in Australia have been categorised simply as two regions: the Intensive Land-use Zone (ILZ), and the Extensive Land-use Zone (ELZ). Typically, the nature and extent of vegetation clearance differs significantly between these two zones. Most Australians would be more familiar with the former that includes the intensive agriculture zone of the Murray-Darling Basin where the ecological and biological effects of clearance are increasingly self evident and irreversible. In contrast, the effects of vegetation modification on biodiversity in the ELZ is generally much less obvious to the human eye. Even today, some well-informed scientists and many members of the public think of northern Australia as 'pristine'. This is far from the truth (see Vegetation modification and fragmentation).

The effects of vegetation clearance

The impact of broad-scale vegetation clearance on natural heritage and biodiversity is profound and has been of concern for several decades in Australia. The immediate effect of clearance of terrestrial native vegetation on plant and animal species can be significant. For vertebrate animals, comparative estimates of the population density of woodland birds indicate that between 1000 and 2000 birds permanently lose their habitat for every 100 ha of woodland cleared (Glanznig & Kennedy 2000), while it has been estimated that the clearing of mallee for wheat kills more than 85% of the resident reptiles (Glanznig & Kennedy 2000), on average, more than 200 individual reptiles per hectare. Longer-term effects of native vegetation clearance on species result from habitat loss and fragmentation combined with other threats.

The links between clearance of native vegetation and changes in hydrological cycles are relatively well understood and have serious implications for land management and biodiversity. Vegetation clearance changes the water balance of an area and this may lead to fundamental changes in the local soils and climate, as well as the local water table and its chemical composition (Stirzaker et al. 2000). Extensive clearance of vegetation across a catchment or region may generate 'cascading effects' (e.g. Mac Nally 1999) on the biophysical systems of the area and these changes may be irreversible or difficult to deal with other than through long-term (decade or century) mitigation and restoration strategies (Blackmore et al. 1999).

One recently publicised example of the physical changes that can occur to the hydrological cycle and environment as a result of vegetation clearance is dryland salinity. This has led to the recent development of several state and regional salinity strategies or audits. Western Australia developed their Salinity Action Plan in 1996, and Victoria, New South Wales and the Murray-Darling Basin Commission followed in 1999 to 2000. The most recent assessment of dryland salinity at the national scale was released in early 2001 by the NLWRA (2001). This assessment estimated that 630 000 ha of remnant native vegetation and associated ecosystems were within regions with areas mapped to be 'at risk'. These areas were projected to increase by up to 2 000 000 ha over the next 50 years. Dryland salinity poses a major threat to biodiversity in Australia.

In October 2000, the Commonwealth government released a National Action Plan for Salinity and Water Quality (Commonwealth of Australia 2000) to help manage dryland salinity and deteriorating water quality more systematically in key catchments and regions (Figure 9; A new National Action Plan box). The National Action Plan explicitly includes the conservation of biodiversity as one of its goals, to help ensure that the policy responses introduced to tackle land degradation are also consistent with the requirements for biodiversity conservation at the landscape level.

Figure 9: Salinity and water quality in Australia showing the major areas of concern
The areas shown were identified in the National Action Plan for Salinity and Water Quality (2000).

 Salinity and water quality in Australia showing the major areas of concern

Source: Areas identified in the Prime Minister's launch of the National Action Plan for Salinity and Water Quality (2000)

Salinity, biodiversity conservation and a new National Action Plan

Australia has critical salinity and water quality problems that require urgent attention. At least 2.5 million hectares (5% of cultivated land) are affected by dryland salinity (Commonwealth of Australia 2000) and this could rise to 17 million hectares at the current rate of increase in this type of land degradation (NLWRA 2001). In addition, one-third of Australian rivers are in extremely poor condition (e.g. within 20 years, Adelaide's drinking water is predicted to fail World Health Organization salinity standards in two days out of five). Infrastructure (e.g. buildings and roads) is being severely damaged in many rural urban centres.

Most salinity problems in Australia result from broad-scale land clearing, such as the type of land clearance occurring in Queensland. The clearance of vegetation can fundamentally change the hydrology of an area, and lead to a significant increase in the volume of water draining beneath the normal root zone of the local vegetation. If stored salt is present in the soil or ground water, then hydrological changes may result in salt concentrations dramatically increasing in near-surface soils and/or at ground level.

Dryland salinity can have a serious effect on biodiversity. For example, preliminary findings from a four-year biological survey of the Western Australian agricultural zone indicate that 450 endemic plant species are under threat of extinction from salinity (CALM 2000a). Further, the death of trees and shrubs in many wetlands in the wheat belt as a result of salinity has caused a 50% decrease in the number of waterbirds using them, and without intervention, about 75% of the waterbird species in the region will severely decline. If all wetlands in the wheat belt become saline, well over 200 aquatic invertebrate species will become regionally extinct (CALM 2000a).

In October 2000, 'A National Action Plan for Salinity and Water Quality in Australia' was released by the Commonwealth government. This plan proposes that concentrated action by governments and communities needs to lead to land use change supported by the application of scientific advances in mapping salinity, targeted tree planting and new cropping systems to manage salinity and water quality. The plan identifies high priority, immediate actions to address salinity, particularly dryland salinity, and deteriorating water quality in key catchments and regions across Australia.

The goal of the plan is to:

  • prevent, stabilise and reverse trends in dryland salinity affecting the sustainability of production, the conservation of biodiversity and the viability of our infrastructure
  • improve water quality and secure reliable allocations for human uses, industry and the environment.

The Action Plan builds on the work established under the NHT, the Murray Darling Basin Commission, state and territory strategies and the COAG Water Agreement. The plan will be implemented through targets and standards for natural resource management, particularly for water quality and salinity, including salinity, water quality and associated water flows, and stream and terrestrial biodiversity based on good science and economics.

To ensure that integrated catchment or region management plans contribute to the achievement of nationally agreed outcomes, catchment or region-specific targets for salt, nutrients, water flow regimes, water quality, stream and terrestrial biodiversity will be required. This capability will need to be able to:

  • map salinity hazard using 'ultrasound' technology and assess catchment and region conditions and issues
  • maintain and improve the condition of existing native vegetation
  • establish multiple purpose perennial vegetation (focused on agriculture, forests, biodiversity and carbon credits) in targeted areas, identified through salinity, vegetation and hydrology mapping, and ground water modelling
  • protect and rehabilitate priority waterways, floodplains and wetlands
  • improve environmental flows, where this is beneficial
  • improve stream water quality using engineering works in critical areas (e.g. salt interception devices and ground water pumping, removal of weirs and redundant structures, fish ladders (to assist fish migrate upstream past structures such as dams) and artificial wetlands)
  • install drainage in catchments or regions where agreed by affected land managers, the downstream effects are positive, and the overall benefits of the scheme provide substantial long-term results over other approaches
  • address the harder adjustment and property amalgamation issues
  • address the problems of degradation of rural urban infrastructure (e.g. buildings and roads).

There are some 20 high priority catchments and regions that need attention, including: the Burdekin-Fitzroy (Qld), Lockyer-Burnett-Mary (Qld), Namoi-Gwydir (NSW), Macquarie-Castlereagh (NSW), Murray (NSW), Goulburn-Broken (Vic.); Glenelg-Corangamite (Vic.), Midlands (Tas.), South-East (SA), Avon (WA), Northern Agricultural Region (WA), South West (WA) and Ord (WA-NT).

The plan acknowledges that land clearing in salinity risk areas is a primary cause of dryland salinity. It indicates that effective controls on land clearing are required in each jurisdiction, and states that:

  • any Commonwealth investment in catchment or region plans will be contingent upon land clearing being prohibited in areas where it would lead to unacceptable land or water degradation
  • the Commonwealth will require agreement from relevant states and territories (particularly Queensland, New South Wales and Tasmania) that their vegetation management regulations are effectively used or, where necessary, amended to combat salinity and water quality.

The National Action Plan is intended to promote major systemic improvements in land and water management. The Plan suggests that attention will need to be given to other high priority natural resource management issues such as the broader conservation of biodiversity and preventing productivity decline in other catchments and regions.

Vegetation clearance also results in changes to the physical and chemical composition of soils and may significantly increase the likelihood of soil erosion and nutrient loss (AAS 2000), whereas clearance itself may lead to soil compaction and other physical modifications of the landscape. These changes may affect biodiversity deleteriously, both directly and indirectly (MDBC 1999; Stirzaker et al. 2000).

Vegetation clearance

Concern about the fate of the nation's native vegetation and its associated biodiversity led to the adoption of two key targets in National Strategy for the Conservation of Australia's Biodiversity (1996) so that by the year 2000:

  • Australia will have avoided or limited any further broad-scale clearance of native vegetation, consistent with ecologically sustainable management and bioregional planning, to those instances in which regional biodiversity objectives are not compromised
  • Australia will have arrested and reversed the decline of native remnant vegetation.

As will become clear in the following sections, these targets are not close to being met.

National figures on vegetation clearance: The most widespread quantitative study of land cover change occurred in the ILZ that covers 38% of the continent (Barson et al. 2000). This study estimated that from 1990 to 1995, about 1 212 000 ha of woody vegetation were cleared for agriculture (cropping), grazing and other activities such as urban development. Although not all of this clearance may result in the permanent loss of woody vegetation, and does not include grasslands or sparse woodlands, these data are a compelling reminder of the spatial extent and intensity of vegetation clearance. These data also indicate that the relatively high rates of vegetation clearance recorded during the 1980s have continued into the 1990s. Recent data for 1997 to 1999 indicate that the rate and area of native vegetation cleared remains very high, especially in Queensland (Figure 10), and in some cases the rate of clearance has increased over the past three years.

Figure 10: Area of native vegetation cleared within the ILZ of Queensland between 1997 and 1999 by subregion.
Indicative map only.

 Area of native vegetation cleared within the ILZ of Queensland between 1997 and 1999 by subregion

Source: EPA, Queensland; AGO. Compiled by: NLWRA, Landscape Health Project, Canberra

Gully erosion along a creek in Bathurst, New South Wales

Gully erosion along a creek in Bathurst, NSW.

Source: JE Williams

Information about types of vegetation cleared and the nature of landscape modification and biodiversity depletion resulting from vegetation clearance is vital. The effects of native vegetation clearance on the extent of particular ecosystems and ecological communities are well known. For example, the estimated original extent of the Big Scrub (subtropical rainforest between Lismore and Bangalow in northern NSW) was over 75 000 ha. However, by 1900, it had been reduced to about 300 ha scattered over 10 remnant patches. An assessment of areas in Queensland where the dominant vegetation type is Brigalow (Acacia harpophylla) shows that of the original estimated extent of more than six million hectares, only about 5% remains, and only about 30 260 ha, or 0.5%, were reserved. Most clearance occurred in the 1960s, and by the 1970s a large portion of the brigalow had disappeared (Nix 1994).

Glanznig and Kennedy (2000) reported that nominations for several biodiversity 'hotspots' that are threatened by vegetation clearance for sugar cane were being prepared by conservation organisations to inform governments of high priority areas for conservation protection. These hotspots include lowland forests, freshwater wetlands, grasslands, littoral rainforest and other ecosystems along the eastern seaboard bioregions of Queensland.

Barson et al. (2000) provided another indication of the changes in particular vegetation types in their review of land cover change in the Intensive Land-use Zone for 1990 to 1995. Using the J. Carnahan broad vegetation mapping as a baseline (Commonwealth of Australia 1990), these authors reported that open forest and woodland ecosystems represented much of the cleared area. For example, 147 650 ha and 515 990 ha of these ecosystems, respectively, were cleared in Queensland between 1990 and 1995. About 350 000 ha of low woodland was also cleared across Australia during the same period, with some 90% occurring in Queensland (Barson et al. 2000). On a more positive note, this and related studies (e.g. Graetz et al. 1995) suggested that the extent of clearance of closed forests such as rainforest is lower than estimated previously.

At the end of the 1990s, the total area of native vegetation that has government sanction for clearing remains high, in excess of one million hectares per year. Rates of clearance varied across Australia, with Queensland estimated to have the highest rate of clearance: about 425 000 ha of vegetation removed per year between 1997 and 1999 (Department of Natural Resources 2000) (see also Figure 10).

Various estimates of clearing rates for New South Wales are 14 028 ha per year for 1997 to 2000 (Department of Land and Water Conservation 2001), 30 000 ha in 1999 (AGO 2001) and 100 000 ha in 2000 (ACF 2001). Based on estimates compiled by the ACF (2001), the total area of native vegetation that was cleared in Australia during 2000 was over 564 800 ha; the Australian Greenhouse Office (AGO) estimate for 1999 is 468 844 ha. On available figures, the former area is exceeded by only four other countries in the world: Brazil, Indonesia, DRC (Congo) and Bolivia (Figure 11).

Permits to clear native terrestrial vegetation

Governments are able to influence the rate of native vegetation clearance through enacting and enforcing legislation that requires permits to clear vegetation. There is no uniform legislation of this type in Australia. At the Commonwealth level, in early 2001, the Minister for the Environment, following advice from the Threatened Species Scientific Committee, listed land clearing as a key threatening process to biodiversity under the Commonwealth EPBC Act (see The Environment Protection and Biodiversity Conservation Act 1999). In general, where legislation for land clearance and vegetation management is in place at the state and territory level, it covers woody native vegetation and provides for many exemptions. The legislation rarely applies to all land tenures and is politically controversial in that private property rights, government involvement, economic growth, regional development, land stewardship, Indigenous land rights, land access and biodiversity conservation are contested (Glanznig & Kennedy 2000). In addition, if the legislation is not enforced, it matters little how much is in place.

Number of permits granted [BD Indicator 18.1]

During 1999, Australian governments granted permits for clearing a total area of well over one million hectares of vegetation. The State governments of Queensland and New South Wales, alone, granted permits to clear 713 515 ha. Since the completion of SoE (1996), the total area of land for which vegetation clearance permits have been granted each year has generally remained at a relatively high level or increased. Selected examples of the number of permits granted are given below.

Queensland: Permits on leasehold land in Queensland increased from 496 957 ha on average between 1995 and 1997, to 644 515 ha in 1999. Recent data reveal that there has been a 31% increase in Queensland land clearing permits for the first six months of 2000 (431 781 ha), compared with 1999 (329 714 ha), and a 104% increase compared with the first six months of 1998 (211 199 ha). Old growth vegetation (bushland) comprises 166 194 ha of the 431 781 ha, with over 90 000 ha of this vegetation type approved for clearing in May 2000 alone.

New South Wales: In New South Wales, about 86 000 ha of native vegetation was approved for clearing in 1998, and at least 69 000 ha were approved in 1999. Although few verifiable, concise figures are available, about 40 000 to 50 000 ha per year are reported to be converted to improved pasture or feed crops for cattle, and 5000 to 6000 ha cleared for mixed horticulture (e.g. bananas and mangoes). Numerous proposals exist for major irrigated agriculture and horticulture projects of up to 250 000 ha and some commentators estimate that between 500 000 and one million hectares of native vegetation are earmarked for clearing over the next 10 years (ACF 2000).

Significant areas of native vegetation have been cleared in several regions of New South Wales that support depleted ecosystems. For example, the annual mean clearing rate for 1995 to 1997 in the Cobar Peneplain is estimated to be 13 250 ha. More than one-third of this region's native vegetation has been cleared since European settlement and

Western Australia: Clearance of native vegetation in south-west Western Australia has been historically high, but is now under strict control. However, even low levels of clearance can have major effects on biodiversity, as it is likely to affect a greater proportion of the remaining vegetation. Effect on issues such as salinity, erosion and water quality are also of serious concern.

Tasmania: The Tasmanian government has made a commitment to best practice vegetation management and relies heavily on incentives and cooperation with landowners to achieve this, rather than regulation. On the basis of the available empirical data, this approach is grossly inadequate. The estimated average annual rate of clearance of native vegetation in this State between 1988 and 1994 is 10 429 ha. For 1994 to 1999, the mean rate of vegetation loss declined to slightly under 7000 hectares per year, and vegetation types of conservation significance continue to be cleared (Kirkpatrick & Mendel 2000). This brings the estimated total area of vegetation cleared in Tasmania for 1972 to 1999 (since consistent quantification commenced) to 265 575 ha. The large-scale clearing of native vegetation remains one of the most significant issues affecting Tasmania's natural environment (DPIWE 2000).

Northern Territory: Some 10 000 ha of vegetation are estimated to have been cleared in the Northern Territory during 1999, but precise details of the clearing are limited. Data for the Shire of Litchfield indicate the success of clearance applications. The Shire is about 310 000 ha in area, of which 27 000 ha are zoned as conservation, and removal of vegetation is subject to the Litchfield Area Plan 1992. The Plan requires that where removal of vegetation is proposed for an area exceeding 50% of an allotment, the vegetation be removed in accordance with environmental guidelines. Since 1996, requests to clear over 2500 ha of vegetation have been made and all appear to have been granted. The reasons for clearing have included horticultural production (mainly mangoes). Eucalyptus woodland (48%) and Eucalyptus open woodland (25%) were the predominant vegetation types cleared.

Vegetation clearance: An overview

The estimates for rates of native vegetation clearance given above underestimate the true picture since they do not account for illegal clearing or clearing carried out under legislative exemptions (which covers regrowth and private forestry). In New South Wales, for example, such exemptions may include 'day-to-day farm management', and in Western Australia exemptions include clearing for urban development. Planning and environmental assessment processes are in place for many urban areas, but the expansion of residential areas is still of concern.

Current technical limitations associated with the use of satellite imagery also may introduce errors and inconsistencies in the estimates of vegetation clearance rates, as can the use of different definitions for vegetation communities. Of particular relevance, the continental-scale monitoring does not pick up clearance of native grasslands or areas where tree cover is sparse, including the large areas of southern Australia where individual scattered trees are a dominant, and very important part of the landscape (see The nature of fragmented vegetation).

The imprecision of data sets that are used to estimate rates of native vegetation clearance can be used by governments to delay action to protect biodiversity. Although not perfect, estimates of the rate of vegetation clearance are indicative of the high rate of clearance, its spatial extent and clearance 'hotspots'. The seriousness of the effects of these human activities on biodiversity is clear, and a far more comprehensive response by governments is required urgently.

The rates of vegetation clearance in Queensland have received particular attention. Although the total area of cleared land is extraordinary by Australian and global standards, the effects on biodiversity resulting from land clearance in States such as New South Wales and Western Australia may be as significant or even more significant. This is because the proportion of vegetation cleared versus the amount remaining might be more significant on biodiversity than the absolute amount cleared. Measures to control land clearance in States such as New South Wales, Western Australia and Tasmania are, therefore, essential to minimise further degradation of the biodiversity associated with the vegetation in these jurisdictions.

New remote sensing techniques, a standardised approach to determining land cover change and further evaluation of methods will undoubtedly improve the precision of current estimates. While these improvements are now actively pursued by government agencies and research institutions, it is very telling that Australia is unable to systematically report on the rate of clearance by vegetation type at the national scale.

Vegetation modification and fragmentation

McIntyre and Hobbs (1999) developed a framework describing the range of landscape alteration 'states'. Four landscape alteration states are recognised:

  • intact
  • variegated
  • fragmented
  • relictual.

These are associated with increasing amount of habitat destruction and decreasing levels of habitat connectivity. In intact landscapes (e.g. arid rangelands), less than 10% of the vegetation is destroyed and the landscape mosaic is, therefore, 'habitat' in various states of modification. At the other extreme are relictual landscapes (e.g. cropping or urban areas) where over 90% of the vegetation is destroyed and small areas must survive in a landscape matrix which may be hostile to the continued persistence of the vegetation.

The 'intact' landscapes described by McIntyre and Hobbs (1999) largely coincide with the ELZ, which covers around 60% of monsoonal, semi-arid and arid Australia. New data suggest that the relatively contiguous extent of overstorey vegetation in the ELZ, compared with that for the highly cleared ILZ, is not a reliable indicator of the conservation status of ecosystems. The disproportionate extinction and regional loss of mammal species for the arid zone and related parts of the ELZ is well known. However, recent and ongoing biological surveys suggest that pervasive changes in these ecosystems (e.g. fire regimes and livestock grazing intensity) that are closely linked to land tenure threaten a diverse range of biota (e.g. Franklin 1999; Fraser 2000; Woinarski 2000). In most instances, the conservation status of native vegetation communities is poor, especially in most parts of the rangelands, arid and wet-dry tropics (Figure 12).

Figure 12: Percentage of native vegetation in land tenures associated with conservative land use practices (indicative map only).

 Percentage of native vegetation in land tenures associated with conservative land use practices (indicative map only)

Source: Bureau of Rural Sciences Landuse Grid; State vegetation coverages. Data currency: land use, 1999; vegetation, NSW 1986-1995; Qld 1997; SA 1985-1995; Tas. 2000; Vic. 1987; WA 2000; NT no data. Compiled by: NLWRA, Landscape Health Project, Canberra

In the ILZ (mainly southern and eastern Australia), large areas of vegetation and its associated biota have been heavily modified (Figure 13) and they generally fall into the fragmented and relictual categories of McIntyre and Hobbs (1999). The clearance of native vegetation disrupts ecosystems and habitats and results in the creation of remnant islands or fragmented patches (e.g. near Jandakot airfield, WA) (Figure 14) and linear fragments along roadsides. These have become important reservoirs of plant and animal species that depend on native habitats. The degree of vegetation modification and fragmentation in these regions and beyond has intensified during the past five years (Figure 15). Figures 13 and 15 overstate the quality of the remaining native vegetation as they are based largely on structural information about the overstorey of communities, but this does not necessarily indicate the quality of the mid-storey and understorey which may be significantly modified as a result of grazing by introduced livestock, weed infestation and other agents of physical and biological change.

Figure 13: Current extent of native vegetation by bioregion (IBRA).
The information on this map has been classed according to the percentage of woody vegetation remaining uncleared, and natural grasslands remaining uncultivated (indicative map only).

 Current extent of native vegetation by bioregion (IBRA)

Source: State vegetation coverages. Data currency: land use, 1999; vegetation, NSW 1986-1995; Qld 1997; SA 1985-1995; Tas. 2000; Vic. 1987; WA 2000; NT no data. Compiled by: NLWRA, Landscape Health Project, Canberra

Figure 15: Mapping of the degree of native vegetation fragmentation. Indicative map only.

 Mapping of the degree of native vegetation fragmentation

Source: Expert opinion based on state vegetation coverages. Data currency: land use, 1999, vegetation, NSW 1986-1995, Qld 1997, SA 1985-1995, Tas. 2000, Vic. 1987, WA 2000, NT no data. Compiled by: NLWRA, Landscape Health Project, Canberra

Few of the vegetation remnants remaining after clearing are large enough to sustain ecological processes such as water and nutrient cycling at the rates that existed before disruption. Many continue to be disturbed by threatening processes such as invasion by weeds or feral animals coming from the surrounding cleared land and firewood collection (see Burning the bush).

More insidious threats like rising saline water tables and the inflow of fertilisers from surrounding lands are also of major concern. Despite the major threat that dryland salinity poses to native vegetation (see The effects of vegetation clearance and Salinity, biodiversity conservation and a new National Action Plan), mapping of the distribution of major fragmented vegetation types in selected catchments and their likely response to projected hydrological changes has begun only recently (Figure 16). The issue of clearing by 'ecological action' is also critical. Putting a fence around patches of bush and keeping grazing animals out may not be sufficient to prevent further degradation, and in many cases loss, of remnants. Changes in the condition of remnant vegetation, such as the presence or absence of plant regeneration, therefore need to be assessed to guide and improve management.

Figure 16: Salinity risk to remnant vegetation in south-west Western Australia.

 Salinity risk to remnant vegetation in south-west Western Australia

Source: Map produced from processed Landsat images and DEM by the Land Monitor Project, WA

A roadside remnant with Giant Blue Waterlily

A roadside remnant with Giant Blue Waterlily
(Nymphaea gigantea, foreground) and Lepironia articulata (mid-ground) in a coastal creek near Grafton, NSW, that has, as yet, not been highly modified. Both species are more common in Queensland, but reach their southern limit in northern New South Wales.

Source: JJ Bruhl, The University of New England

Many vegetation types now exist as remnants along roadsides and railway reserves, such as this community near Bathurst, New South Wales

Many vegetation types now exist as remnants along roadsides and railway reserves, such as this community near Bathurst, NSW.

Source: JE Williams

The mapped area in Figure 16 is about 30 by 30 km and shows (in colour) remnant vegetation (green), present mapped salt-affected and low-productivity areas (red), and risk-areas (ghosted in blue and fringed with yellow) that are defined as low-lying areas with the potential for shallow water tables.

The nature of fragmented vegetation

The destruction and modification of native vegetation has left a legacy of patches of native vegetation of various sizes, shapes, connectivity and condition. Many of the ecological values associated with native vegetation (Kirkpatrick & Gilfedder 1999; Lambeck 1999) relate to medium to larger patches, although all native vegetation has some role in the landscape (Williams 2000). For example, individual trees provide shade for stock, nesting and foraging sites for wildlife (Lumsden & Bennett 2000), cycle nutrients, act as a source of seeds and may help to reduce ground water recharge and to recycle cations from depth (Reid & Landsberg 2000). The gradual decline in vigour and eventual death of many of these trees has led to a phenomenon known as 'rural dieback'. There are several different causes of this widespread condition and their relative effects can vary in different parts of the farm, on neighbouring farms, in adjacent districts and catchments and between regions (Reid & Landsberg 2000). These authors have listed several factors that may cause rural dieback including insect damage, large numbers of Noisy Miners, secondary salinisation, pathogens, drought, nutritional disorders and old age. Most of these are made worse, or are associated with, the broad-scale clearing in the areas where rural dieback generally occurs.

Burning the bush: The implications of firewood collection for biodiversity conservation

Most of the firewood supply in Australia comes from stands of remnant native vegetation on private property (Driscoll et al. 2000; Wall 2000). Firewood collection includes the removal of fallen and standing dead trees from the bush, as well as living trees that are sometimes ringbarked for future use. Overall, the harvesting of wood for domestic heaters means that around five and a half million tonnes per year is burnt, a similar volume to the amount of eucalypt woodchips exported each year (ANZECC 2000c ; Williams 2000). Firewood is, therefore, the third largest source of energy used in Australia after electricity and gas. Around 60% of firewood is purchased through small collectors or suppliers rather than firewood merchants with established premises.

The removal of such large amounts of dead and living wood from patches of bush is considered to have a major effect on the whole spectrum of biodiversity, from ecosystems to genes (Driscoll et al. 2000). Fallen timber provides habitat for insects and other invertebrates, reptiles and ground feeding birds such as the Bush Stone-curlew (Traill 2000). Dead, standing timbers, also targeted by firewood cutters, are more likely to have hollows than live trees and are favoured as nesting sites by possums, parrots, bats and other wildlife. It is also a favoured foraging site for some insects and insect-eating species such as the Brush-tailed Phascogale (Traill 2000). The potential loss of highly specialised species of invertebrates and fungi associated with coarse woody debris is of particular concern (Driscoll et al. 2000). The disappearance of these species from native bushland could affect ecosystem services such as nutrient cycling and plant establishment.

Inland forests and woodlands in lower rainfall zones appear to be the ecological communities most threatened by the collection of firewood (Driscoll et al. 2000). In Victoria, 49 ecological communities have been listed as potentially threatened by firewood collection, emphasising the extent of the problem. At the species level, firewood removal has been implicated in the decline of birds at the local and regional level (see Driscoll et al. 2000), while at the national level, Garnett and Crowley (2000) identified 21 bird species threatened by firewood collection. Plants can also be affected. In Tasmania, 13 species have been listed as threatened by firewood collection. Many of these species have very restricted distributions and it is thought that firewood collectors could inadvertently damage a large proportion of the remaining populations of these plants. The spread of fungi such as Phytophthora cinnamomi by firewood collectors is also considered a threat (Driscoll et al. 2000).

Recently at the national level, government and NGOs have paid attention to firewood collection and its effects. For example, in mid-1999, the Victorian National Parks Association held the first national conference on firewood collection, and ANZECC is developing a national approach to firewood collection and use (ANZECC 2000c). The objectives of this strategy are to:

  1. protect remnant native vegetation, threatened ecosystems and habitat for threatened and declining wildlife species
  2. encourage ecologically sustainable firewood collection from native forest, woodlands and plantations
  3. contribute to broader environmental objectives (e.g. improved air quality, dryland salinity, and contributing to carbon sequestration).

The codes of practice proposed for the firewood industry would be voluntary but given the scale of the industry and its significant impact on biodiversity, more concerted and urgent measures are required.

Fragmentation of native vegetation creates new edges between remnants and cleared or disturbed land which leads to 'edge' effects. These include physical changes to the remnant in the border region such as different levels of exposure to the sun and wind and changes in water cycles and the local air temperature (Saunders & Hobbs 1991). Biotic changes include invasion by opportunistic species with good dispersal or colonising abilities such as weeds and feral animals. Fragmentation also isolates and creates barriers between patches of native bush. In most cases, recently isolated remnants can be expected to continue losing species (Saunders & Hobbs 1991). The loss of a population of a species (that has declined to a size that is not viable) may take considerable time if individuals are relatively long lived. For example, it may take several hundred years to lose species of long-lived trees, particularly since adult plants are often less sensitive to changed environmental conditions than plants in their seedling and juvenile stages. This phenomenon also applies to many fauna (e.g. Trapdoor Spiders in the wheat belt of Western Australia may live for at least 23 years (Main 1999)).

The consequences of habitat fragmentation on biodiversity depend on the interaction of many factors that may vary for different species and habitats (see The Living Landscapes Project). To illustrate the implications of habitat fragmentation on a particular group of species, the effects of habitat fragmentation on birds is summarised (Table 9) (see The relationship between habitat fragmentation and bird abundance and range).

Threats are divided into those that are continuing and those that no longer occur, either because the taxa are extinct or because the process has ceased and is no longer affecting the surviving birds (see The relationship between habitat fragmentation and bird abundance and range).

Table 9: Threats to Australian birds
Clearance and fragmentation of habitat Current threats Former threats Total
Confirmed Speculative Confirmed Speculative
Agriculture 32 4 22 9 67
Mineral extraction 2 4 3 - 9
Softwood plantations 2 1 2 - 5
Urban development 4 3 1 - 8
Forestry operations 3 14 - - 17
Total 43 26 28 9 106

Source: Glanznig and Kennedy (2000).

Other impacts on native terrestrial vegetation and associated biodiversity

Significant modification of native vegetation may result from agents of change other than clearing such as grazing by introduced livestock and native herbivores, changes to the hydrological regimes leading to waterlogging and salinity and altered environmental flows, invasion by weeds and changes in fire regimes. Another threatening process that is receiving increasing attention is the effect of firewood harvesting on remnant native vegetation with current rates of extraction estimated at around 5.5 million tonnes per year, similar to the amount of wood that is exported as woodchips (see Burning the bush ).

The pressures on biodiversity in old growth forests were identified as a major issue in SoE (1996). At the time, the logging of native eucalypt forests was receiving considerable attention in the national media, especially since the level of wood-chipping of native forests had been very high throughout the decade and many ecologists were seriously concerned about the effects of intensive forestry practices on forest biodiversity and ecosystem services. For example, Norton (1996) argued that many forestry practices in eucalypt forests in eastern Australia were not ecologically sustainable, and that biologically significant forest ecosystems and many native forest biota were threatened. One response of governments to attempt to resolve the debate over native forest management was to initiate the RFA process. RFAs arose from the ESD process of the early 1990s and were intended, put simply, to take care of the reasonable conservation needs of the forests, and then to facilitate economic development in the remaining forests (Kirkpatrick 1998).

Unfortunately, the RFAs do not provide a comprehensive coverage of the native forest estate as there are important areas that have not been assessed. Further, within the regions where RFAs were undertaken, many important conservation needs have not been adequately addressed. For example, several biologically significant ecosystems and species have not been adequately protected, many additions to the conservation reserve network have not been determined using the best available scientific techniques, and the efficacy of a number of forestry management prescriptions remains to be determined (e.g. Kirkpatrick 1998). The implications of these limitations for biodiversity conservation may be amplified since government quotas on wood-chipping were removed on signing of an RFA. Hence, the potential for the intensification of wood-chipping in these regions on public and private lands has significantly increased.

The Living Landscapes Project: A community based project to develop sustainable landscape management

The Living Landscapes Project is a community-oriented research and strategy development project in the wheat belt of south-west Western Australia, which is attempting to embed biodiversity conservation into catchment and agricultural planning. This pilot planning process considers both agricultural production and broader landscape issues such as nature conservation and ecological health (Frost et al. 2000). The aim is to assist community groups to develop landscape management practices that protect biodiversity within an ecologically viable and sustainable land use system. The Living Landscapes project involved an interdisciplinary team that used experiential learning as an overarching process (Frost et al. 2000). The other key process used was the focal species approach (Lambeck 1999). By combining these two approaches, Living Landscapes has developed a set of guiding principles for nature conservation planning in the context of sustainable land management. The approach is now being considered in several regions in eastern Australia.

The relationship between habitat fragmentation and bird abundance and range: A case study

Studies have investigated the medium-to long-term effects of fragmentation on different groups of species, in particular, birds and mammals. In relation to birds, Table 9 summarises the threats to the 150 taxa described in a Royal Australasian Ornithologists Union report on threatened and extinct birds of Australia (Garnett 1992). The threats are divided into those that are continuing and those that no longer occur, either because the taxa are extinct or because the process has ceased and is no longer affecting the surviving birds. For instance, the clearance of mallee in the Western Australian wheat belt has now almost stopped but the effects on fragmented populations are continuing.

Table 9 highlights the prevalent threat of clearance and fragmentation of habitat, especially resulting from the conversion of land supporting native vegetation to use for agricultural purposes. Numerous studies in many States have concluded that bird abundances are directly related to the degree of habitat loss and fragmentation, and heavily fragmented areas will be accompanied by net losses of species which can continue long after the initial clearing (Recher & Lim 1990). In Western Australia, Saunders (1989) found that a rapid loss of species had occurred in wheat belt reserves since the clearance of the original vegetation 30 to 50 years ago. This finding has been reinforced by a more recent study which showed that 49% (95 of the 195 species) of birds recorded in the wheat belt (excluding vagrants) have declined in range and/or abundance since the region was developed for agriculture. Most of these losses result from loss of habitat and fragmentation of the remainder. This general pattern of regional loss and decline of bird species has been repeated in other States such as New South Wales (Reid 2000a, 2000b) and Victoria (Bennett & Ford 1997).

The degradation of habitat by removal of the understorey of forest and woodland ecosystems, through grazing for example, is also significant since it can simplify these ecosystems and result in the loss of species and genetic variability. For example, a study in the Latrobe Valley of south-east Victoria found that small, heavily grazed patches (less than 10 ha) supported fewer forest birds and an increased number of farmland birds, including Noisy Miners (Manorina melanocephala), which aggressively excluded other species. Few birds ate insects in the canopy of these patches, which showed signs of dieback due to insect damage. Planting understorey species may be the most effective way to exclude Noisy Miners and encourage other native bird species.

Recher (1999) predicted that if immediate action was not taken to reverse the decline of native birds, then Australia would lose half of its terrestrial bird species in the next century. The most urgent actions identified were to end the clearing of native vegetation, remove inappropriate fire regimes, control feral and native animals whose abundance threaten native species and restore functional ecosystems.

Source: after Glanznig and Kennedy (2000).

Climate change is another major threat to native vegetation (see Human-induced climate change).

Effects of grazing on vegetation and biodiversity in Australian semi-arid and arid rangelands have been quantified (Landsberg et al. 1999) using artificial sources of water (tanks and dams, bores, waterholes and wells) as a measure of potential effect by livestock. Most large mammals require regular access to drinking water, and in arid environments its availability determines where they graze. Thus, sources of water become foci of grazing activity. This may result in a zone of accentuated impact around each water point where the vegetation is browsed and perhaps killed, soils are compacted and habitat for various flora and fauna is modified or destroyed (see the Land Report). CSIRO Sustainable Ecosystems has mapped the location of artificial sources of water to evaluate the potential effect on biodiversity (Figure 17), and has concluded that vast areas of the rangelands previously beyond the reach of large grazing animals may now be exposed to sustained grazing pressure. Further, few potential reference areas for determining pregrazing patterns of biodiversity now remain (Figure 18).

Figure 17: Distribution of all water points named on the 1:250 000 and 1:100 000 topographic maps covering mainland Australia.

100 000 topographic maps covering mainland Australia

Source: National Land Disturbance Database, AHC; CSIRO Sustainable Ecosystems.
Figure 18: Proportion of IBRA regions >9 km from a watering point

Figure 18: Proportion of IBRA regions

 Proportion of IBRA regions

Source: CSIRO Sustainable Ecosystems

Surveys of water points for the Biograze project revealed that the effect of grazing by cattle is minimal beyond 9 km and beyond 6 km for sheep. Therefore, the map is a conservative estimate of the impact of grazing for sheep. The water point data are accurate for rangeland regions in central, northern and western Australia. Areas closer to the coast, especially in south-west and south-east Australia, have too many water points to be mapped. Consequently the figures provided in these regions are an underestimate. The areas masked out of the analysis had incomplete datasets.

Declines in temperate woodland ecosystems arise from proximate factors such as population reduction and species extinction, genetic loss, substrate modification and salinity (Sivertsen & Clarke 2000). Clearing for grazing and grazing of the understorey of these ecosystems by introduced livestock and feral animals have been identified as important threats (e.g. Kirkpatrick & Gilfedder 2000; Landsberg 2000; Lunt & Bennett 2000). Typically, grazing and related human-driven disturbances exacerbate the deleterious effects on biodiversity resulting from the extensive clearance of lowland woodland ecosystems. These types of degradation of ecosystems have been conceptualised as part of an underlying slow process of desertification (Bauer & Goldney 2000). In general, as grazing pressure increases, native species become less abundant and are replaced by exotic species (Yates et al. 2000). However, under certain circumstances, grazing animals do not need to be totally excluded from native vegetation and in some instances the presence of grazing has been associated with the maintenance of high conservation values at a site (Williams 2000).

Regrowth vegetation

Conservation of vegetation regrowth is vital for biodiversity conservation. Regrowth is important because it may provide habitat for key elements of biodiversity that have been affected by vegetation clearance and fragmentation. Regrowth is also important in supporting and sustaining biophysical and ecological processes. For example, as vegetation regrows, the structural, floristic and biological composition of areas change and may reduce the extreme nature of habitat fragmentation for resident species, and favourably modify local climate and environmental regimes (e.g. energy, radiation, light and exposure to extremes in ambient temperature) (Saunders & Hobbs 1991). The notion that regrowth vegetation has little or no value for biodiversity seems widespread, and one consequence of this is that regrowth clearance may proceed without adequate planning, or may not be challenged even when it threatens significant components of biodiversity (Kirkpatrick & Gilfedder 1999).

No net loss: The policy of 'no net loss' of native vegetation (e.g. Bushcare's target within Australia by July 2001, as per Department of the Environment and Heritage - Portfolio Budget Statement 2000-2001), and the associated policy of offsets (to negate the negative effect of clearing by separate actions that have positive effects), are being considered by several jurisdictions. There are many issues associated with these policies that could have long-term effects on biodiversity. For example, replacing mature woodland with an equivalent area of saplings could satisfy the no net loss criteria, but these two vegetation types are very different in structure and function. Once the original vegetation disappears from a site, then it is difficult, if not impossible, to recreate it. And although revegetation projects are becoming increasingly sophisticated, it will take decades to develop the characteristics of the original vegetation (i.e. being self sustaining), especially the benefits provided by large trees. These not only provide hollows on which many Australian birds and mammals depend, but they produce a more reliable and extensive source of nectar for a range of fauna (Wilson & Bennett 1999). Consequently, there is general agreement that the first step to sustainable management is to retain existing native vegetation where possible (Williams 2000). The next steps are to protect and manage that vegetation and then, where appropriate, to revegetate cleared areas.

Offset actions for native vegetation clearing could include improving the management of existing native vegetation, revegetating a previously cleared area or establishing a tree plantation. Biodiversity conservation would be undermined by these types of activities if exemptions (e.g. vegetation of high conservation could not offset and thus be cleared) were not comprehensive and legally enforceable. State management agencies in New South Wales are attempting to benchmark the concept of biodiversity credits (biodiversity values in major ecological communities such as woodlands, rangelands and forests) to provide a basis for trading (see The Australian private sector and biodiversity).

In 1997, the Victorian government released Victoria's Biodiversity Strategy. One important goal of this strategy package is to achieve a reversal, across the entire landscape, of the long-term decline in the extent and quality of native vegetation. This is designed to lead to a 'net gain' in vegetation with the first target being no net loss by 2001 (The State of Victoria 1997). In this case, the goal of no net loss was based on the premise that (although 'natural' is best) it is possible to recover the extent and quality of native vegetation by active management intervention (The State of Victoria 1997). Questions relevant to this goal include whether the plant species in the vegetation patch are locally native, whether are there enough of the locally native species remaining in a patch to warrant inclusion of the patch in accounting for native vegetation and when revegetation could be classified as part of the State's native vegetation estate.

Restoring vegetation: Tree replanting schemes and similar activities are now supported by government, industry and community. However, frequently the amount of vegetation replaced may be orders of magnitude less than the loss of plants as a result of clearing, modification, dieback and other threatening processes. For example, the goal of the Bushcare program, which is the largest in the NHT, is to reverse the long-term decline in the quality and extent of Australia's native vegetation cover. This ambitious goal will not be met, at least in the short term, because of the continuing broad-scale clearing in areas such as Queensland and New South Wales.

Paton et al. (2000) reviewed the distribution, status and threats of woodland ecosystems in South Australia and suggest that future revegetation strategies in the State need to include a greater diversity of plants, use locally endemic species and plant these in natural dispersion patterns to maximise biodiversity benefits.

The removal of extensive areas of deep-rooted perennial native vegetation and replacement by shallow-rooted annual crops and pastures has significantly affected the hydrological cycles of many regions. As a result, the water table that was formerly drained by the deep-rooted vegetation is rising to the soil surface (see Salinity, biodiversity conservation and a new National Action Plan) at rates of up to 0.5 m per year and at least 2.5 million hectares are affected by dryland salinity. Up to 80% of some catchments in Western Australia, for example, may need to be replanted with trees to reverse the rising salinity, and some salinised streams may already be beyond recovery in this region (Glanznig & Kennedy 2000). Clearly, where revegetation and restoration is possible, catchment-based strategies need to be consistent with biodiversity conservation.

Vegetation restoration is a relatively young science and until recently in Australia has focused on ecosystems and vegetation communities where mining and intensive forestry operations have been undertaken. Restoration of native vegetation can be attempted by active (e.g. planting and soil inoculation strategies) or passive (e.g. natural regeneration) means. While the ability to restore a cleared vegetation community or highly modified community to its original state in terms of species composition and ecosystem function is largely untested, substantial progress has been made in the development of techniques to restore some components of the biota to mine sites (e.g. Nabalco, Gove bauxite mine in northern Australia).

Area cleared or modified to area revegetated [BD Indicator 18.2]

Increases in woody vegetation (woody plants greater than 2 m in height and 20% cover) in Australia's more intensively used agricultural areas for 1990 to 1995 were found to be 414 000 ha (excluding that due to regeneration following fire and native forest harvesting) (Barson et al. 2000). The estimated loss of woody vegetation for the same period was 1 282 200, including clearing for agriculture, mainly cropping (213 680 ha), grazing (929 280 ha), on-farm tree planting (1 760 ha) and plantation management (90 160 ha). This shows a net loss of native woody vegetation for the period.

Recent work on landscape design principles and guidelines is attempting to enhance the rehabilitation and restoration of (at least) terrestrial landscapes. Examples of guidelines and principles are the perceived thresholds for vegetation cover (i.e. a minimum of 30% tree cover is being recommended in temperate and subtropical regions not experiencing severe salinity) (see Barrett 2000a; Williams 2000) and the focal species approach being used for revegetation of cleared agricultural landscapes (Lambeck 1999).

Plantations: The practice of commercial plantation forestry is expanding across many regions of southern Australia. A recent study based on computer modelling of 12 regions by the Australian Bureau of Agriculture and Resource Economics (ABARE) suggests that some 19 million hectares of cleared agricultural land may be suitable for commercial plantations (Burns et al. 1999). Potential expansions of this scale may have significant implications for regional development and infrastructure, and biodiversity conservation. For example, the preparation of land for plantation establishment may involve a significant input of fertiliser and the use of chemicals to control weeds and poisons to control native vertebrate herbivores. These inputs can directly and indirectly affect biota.

Aquatic systems

Vegetation clearance, vegetation modification and intensive land use also affect aquatic plant ecosystems and freshwater fisheries. Bunn et al. (1999) demonstrated that native vegetation that overhangs streams and rivers is important in moderating water temperature, which in turn affects native fish and other biota.

Acid sulfate soil poses another potential threat to biodiversity. These soils are rich in iron sulfide and are common in coastal areas. They are relatively stable while saturated with water, but the sulfide forms sulfate and sulfuric acid when drained and exposed to air. The sulfuric acid can kill fish, prawns, oysters and other aquatic life when washed into waterways by rain. Iron sulfate soils on which vegetation is regrowing can also cause problems once they become exposed. The Coastal Acid Sulfate Soils Program, which is managed by Environment Australia through the NHT, has provided funding for projects with onground works that demonstrate better options for managing coastal acid sulfate soils.

By area, most stream catchments have been subjected to medium to high levels of disturbance, including significant portions of the Gulf of Carpentaria Drainage Basin, Lake Eyre Drainage Basin and Bulloo Drainage Basin that support a relatively low human population density. Aquaculture also affects aquatic plant ecosystems and related parts of the catchments used for this purpose. Data on the current and potential effects of aquaculture on biodiversity are limited and disparate. The extent of aquatic systems for which cultivation permits have been granted are increasing and suggest this issue needs further investigation. For example, of the 567 lakes in western New South Wales, 70 (74 136 ha) have government permits for cultivation.

Extent of each vegetation type within protected areas [BD Indicator 13.1]

In a Statement on the Environment by the Australian Prime Minister in 1992 (Commonwealth of Australia 1992c), it was announced that 'The Government has adopted a policy that all ecosystems be surveyed and that a comprehensive, adequate and representative system of reserves be established progressively by the year 2000'. The National Reserve System Program was designed to deliver this outcome.

The NRSP and related state and territory programs, the RFAs, the Indigenous Protected Areas (IPA) scheme of Environment Australia, new multitenure management schemes, and the enormous growth in contributions from the non-government sector (e.g. Trust for Nature and Bush Heritage Fund) have helped to increase the spectrum of the nation's system of conservation reserves. Even so, many anomalies exist and very few regions in Australia support a conservation reserve system that meets all of the requirements of comprehensiveness, adequacy and representativeness (CAR) (Figure 19).

Figure 19: Conservation status of Interim Biogeographic Representation for Australia (IBRA) in 2000 showing the percentage area reserved in each region.

 Conservation status of IBRA in 2000 showing the percentage area reserved in each region

Source: Environmental Resources Information Network

The representation of major vegetation types within the Australian conservation reserve estate remains poor despite the long-standing recognition of the need to enhance the reservation and protection of these ecosystems. In Queensland, for example, native vegetation is typically underrepresented in the conservation estate, including in those areas (e.g. around Charleville and Emerald) where rates of land clearing are very high by international standards and threats to biodiversity are extraordinarily high. Sattler and Williams (1999) reported that over half of the 287 regional ecosystems in the Mulga Lands, Desert Uplands and Brigalow Belt bioregions in Queensland are endangered or threatened with extinction (Table 10). Only 63% of these endangered and threatened ecosystems are represented in the conservation reserve system of these bioregions and less than one-third of these ecosystems is found in more than one protected area.

Table 10: Reservation status of regional ecosystems (REs) in Queensland subject to high rates of clearing and their degree of replication in protected areas of over 1000 ha
Bioregion Regional area (ha) Protected areaA No. of REs in bioregion REs endangered or threatened (%) No. of REs in protected areas REs in protected areasB(%) REs in >1 protected areaA(%)
Mulga Lands 19 097 000 464 900 66 41 47 71 39
Desert Uplands 6 882 000 153 800 58 63 25 43 9
Brigalow Belt 35 158 000 730 400 163 43 110 67 39
Total 61 137 000 1 349 100 287 51 182 63 29

AProtected areas current to 28 February 1998;
BCare should be taken in comparing this table with tables previously published (e.g. Sattler 1986), because of the progressive refinement of REs and bioregional boundaries, and additions to the protected area estate.

Source: after Sattler and Williams (1999).

New South Wales is one of the more 'data-rich' parts of Australia, although its detailed biological data sets, like others elsewhere, are localised. To overcome this limitation, Pressey et al. (2000) have developed a new classification of landscapes at a scale of 1:250 000 across the whole 802 000 square kilometres of New South Wales. The classification is derived mainly from abiotic data (e.g. topography, soils and climate) and, in conjunction with new data on native vegetation cover, has allowed the first quantitative state-wide review of protected areas and future priorities at a scale that approaches decisions about land use. Pressey et al. (2000) found that most of the 1486 landscapes in New South Wales are poorly reserved relative to an indicative conservation target of 15% of the total area of each.

In the eastern 60% of New South Wales, gaps in the reserve system are related to the concentration of reserves on land with high ruggedness and low potential for intensive land use. Pressey et al. (2000) mapped the relative priority of landscapes to indicate the urgency of conservation action to prevent conservation targets being compromised (or further compromised) by clearing of native vegetation. Mapping of priorities shows large differences within and between natural regions and land tenures. More than 9% of private land is occupied by high-priority native vegetation and, across the whole State, about 85% of high priority vegetation occurs on private land. This indicates the importance of controlling vegetation clearance on private land if biodiversity effects are to be avoided.