Australian Land Cover Change data 1990-1995

One hundred and fifty eight Landsat Thematic Mapper scenes for 1990
(1991 in the Australian Capital Territory, New South Wales and
Queensland) and 1995 were used in conjunction with vegetation and a
range of ancillary data sets by State agencies and BRS to establish:

?the extent of woody vegetation (native and exotic vegetation
greater than or equal to two metres tall with greater than or equal
to 20 percent crown cover) in Australia’s more intensively used
agricultural areas in 1990/91;

?the rates of clearing of this woody vegetation and tree planting
1990/91-1995, and the reasons for clearing and planting; and

?the types and biomass of vegetation cleared.

Remote sensing has provided a spatially accurate and reliable estimate
of the rates of clearing 1990-1995 for the Northern Territory, South
Australia, Tasmania, Victoria and Western Australia and for 1991-1995
for the Australian Capital Territory, New South Wales and
Queensland. This information has enabled the revision of preliminary
estimates of annual rates of clearing for woody vegetation for the
early 1990s from around 517,800 ha to 308,000 ha.

The results show that in 1990/91, 38 percent of the study area was
still covered by woody vegetation. Comparison with Carnahan’s
Natural Vegetation (1780s) data set indicates that around 70 percent
of the study area was covered by woody vegetation at the time of
European settlement; almost half of this woody vegetation (92.5
million ha) has been cleared since this time. This estimate does not
include the clearing of grasslands, heathlands and sparse woodlands.

Analysis of the satellite data showed that about 1,212,620 ha were
cleared for agriculture (cropping), grazing and other activities,
principally urban and infrastructure development over the study
period. Clearing for grazing accounted for more than 50 percent of the
total land cover change and 77 percent of the clearing which
contributes to greenhouse gas emissions in the land use change sector
of the National Greenhouse Gas Inventory. This clearing of 1,212,620
ha of native woody vegetation reduced the 1990/91 area of native woody
vegetation cover by 1.1 percent. Annual rates of decrease and
increase in woody vegetation were calculated using results from the
satellite data for use in the National Greenhouse Gas
Inventory. Estimated annual rates of decrease in woody vegetation for
the continent due to clearing for agriculture and grazing total
292,030 ha. Clearing for development of roads and urban infrastructure
added another 16,100 ha.

Increases in woody vegetation (excluding that due to regeneration
following fire and native forest harvesting) totalled 414,010 ha in
the study period. Comparison of the estimated annual rates of loss of
woody vegetation through clearing for agriculture, grazing and
cropping and increases due to on-farm tree planting and plantation
establishment show there was a net loss of native woody vegetation
over the period 1990/91-1995.

For the study area as a whole, Woodland (M2 — 47 percent), Low
woodland (L2 — 28 percent) Open forest (M3 — 16 percent)
and Tall shrubland (S2 — 7 percent were the major types of
native woody vegetation cleared for agriculture, grazing and
development 1990/91 - 1995.

Estimated annual losses of aboveground biomass for the study area due
to clearing for agriculture, grazing and development total 18.5
million tonnes of dry matter. Most of these losses were due to
clearing in Queensland (88 percent). Clearing in New South Wales
contributed seven percent, while clearing in Western Australia
contributed three percent. Other States contributed one percent or
less each.

All areas of land cover change identified through analysis of the
satellite data have been checked extensively by State agencies against
aerial photography, other satellite or ancillary thematic data or
through field work.

An assessment of the land cover change results has been undertaken by
independent consultants. It demonstrated that there were no
significant differences between the States' results and those obtained
by the consultants undertaking the assessment. A comparative analysis
of the States' and consultant's results at the continental, State and
scene levels showed that although the States used different image
analysis approaches, their mapping of land cover change was very
consistent.

The four datasets available as ARC/INFO Grids in Albers Equal Area
projection, namely:

Land cover 1990/91 (LCT_90/LCT_91)
Land cover change 1990/9-1995 (CHANGE) including TYPE
and CAUSE of change
Land cover 1995 (LCT_95)
Structural and floristic vegetation 1990 (SV_90)
including dominant SPECIES, Growth Form and CARNAHAN
structural code

The original 25m pixels have been resampled to 100m and 250m as State
and national datasets respectively.

The data are available on CD-ROM and from the Australian Natural
Resources Data Library (&http://adl.brs.gov.au/anrdl/php/&).

Documents and reports are also available for download in PDF format
from the Greenhouse Sciences Web page under
&http://www.daff.gov.au/brs/land&