Land Cover Change Project Overview

 

Back to Home Page

 

Introduction

 

The primary aim of this project is to quantify the disturbance history at the Injune study site.  The following stages are implemented to achieve this aim:

  1. Land use is assessed across the study area from existing data and checked during field visits
  2. The likely land cover history for various sites is developed through interviews with landholders and calibration with historical records, including air photos and satellite imagery
  3. Land management data is gathered from land holders.  This information can then provide insight into how different management methods may influence the carbon dynamics of this environment
  4. A generalised model of clearing practices, vegetation type and structure, and terrain form is developed to allow assessment of change in areas not directly visited.

 

1. Land use

 

Land use was based on the Australian Land Use mapping procedure for catchment scale land use mapping. Three sets of input data were used:

i                              Cadastre (property boundaries) were obtained from Queensland Department of Natural Resources and Mining (QDNRM) with tenure information attached.

ii                            Ancillary data was collected from 1;100,000 map series including rivers, roads, farms buildings, dams and pumps.

iii                           Landsat TM for 2000 was used to delineate cleared and remnant woody vegetation and to check location of the ancillary data items.

 

The final classification was then coded to the Australian Land Use Management version 5.  The results for the study area are shown in Table 1 below.

 

Table 1. Land use by area in the Injune study area

 

 

2. Land cover history

 

The land cover history layer is based on the combination of 18 Landsat satellite images from 1972-2000 (supplied by Australian Greenhouse Office and Queensland Department of Natural Resources and Mining).  Data from the Queensland SLATS program were interpreted for land cover change, and validated for 5 farms and by State Forests for 4 farms.  Land cover history was based on raw Landsat TM data series and historical air photos from the Queensland Archives.

 

Land cover from the 2000 Landsat data was interpreted first, then we worked backwards in time to ensure integrity.  The dates of clearing were collected through farmer interviews for most of the farms where clearing had taken place and for the State forests.

 

There are 13 farms in the area, and a range of tenure arrangements exist.

v      1 farm in the area has freehold tenure, which means that clearing of remnant vegetation, which is not endangered, can take place under the Vegetation Management Act 1999 between.  Virtually the entire farm, Mount Owen, has been cleared in 1999.

v      10 farms are leasehold, which means that clearing of remnant vegetation requires a permit under the Land Act 1994.  On these farms substantial clearing has taken place in Billin, Mt Owen and Derbyshire Downs and, to a lesser extent, Hillside.

 

Although 18 images were used in the compilation of the land cover history, only one of these images was captured in the 1970s.  Aerial photos were also used to fill in the gaps along with local knowledge from the land holders.  The farmers at Mt Owen and Hillside have only been managing since the late 80s. Figure 1 (below) shows the final land cover history map for the study area.

 

 

 

Figure 1. Land cover history for the study area

 

In order to estimate rates of clearing the data were converted into raster format and each pixel was assigned a likely annual sequence of land cover state using a Markov probability.  Figure 2 (below) shows that two periods of clearing occurred in the mid-70s and late 1980s reaching nearly 20% of the study area.

 

Figure 2. Clearing rates over time, for the study area

 

 

3. Land management history

 

The land management history is based on farmers’ interview as to when, what and why clearing or cultivation took place.  There is a whole range of land management options that the land holders use to improve their land for cattle grazing or cropping.  These include the following, and pictures are shown below:

v      Clearing

o        Chaining

o        Blade ploughing

o        Stem injection

o        Ring-barking

o        Burning

o        Selective harvesting (includes forestry operations)

v      Ploughing

v      Planting pastures

v      Planting crops

 

 

Chaining / pulling

 

Stick rake

 

 

Stem injection

 

Ring Barking

 

 

Land management history was also accessed from the farmer’s interviews and, to a certain extent, from the aerial photos where chaining, ploughing and burning were readily apparent.  By far the most common practice is chaining of these predominantly Eucalypt woodlands.  Figure 3 presents the results of the land management history assessments.

 

Billin has a longer clearing history that the other farms and has moved onto offset ploughing and planting of oats as a fodder crop.  Ring-barking has largely been superseded with stem-injection and other chemical means in the 1960s.

 

These data nonetheless have several drawbacks, particularly the short history available for much of the study area and that prior to 1960s is unknown.  This is one of the objectives of field trips to collect more data on land management decisions.

 

Figure 3. Land management practises for the study area

 

 

4. Land clearing model

 

There are strong relationships between clearing practices used and the terrain characteristics and vegetation type and structure.  In the light of the vegetation management legislation in Queensland, clearing has taken place due to incentives such as the Brigalow development schemes in the 1960s and 70s, restrictions on leasehold land from 1994, and freehold land from 1999 and the phasing out of permits since March this year.  Future clearing events will be restricted to existing permits and the clearing of regrowth.

 

On top of that, how does and did the farmer select what vegetation should be cleared and how it should be carried out?

Current wisdom suggests that:

v      Clearing on clay soils will yield a greater grass biomass but may lead to problems with regrowth.

v      Steep slopes restrict mechanical techniques

v      Labour intensive techniques such as ring-barking and stem-injection are limited to sparser vegetation

v      Burning is only carried out when there is sufficient grass biomass to maintain the burn.

 

This leads us to develop a land clearing model as shown below.  Here the property tenure, vegetation cover and the terrain all influence the management options.  Because we have digital data for all these model layers, we can investigate, and then perhaps predict what has occurred in the past (where we don’t have historical information) and what may happen in the future…

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


Further information

 

Dr Lucy Randall
Landscape Management Sciences
Bureau of Rural Sciences
PO Box 858, Kingston

ACT 2601
AUSTRALIA

 

Tel: +61(0)2-6272 4901
Fax: +61(0)2-6272 5827

Email: Lucy.Randall@brs.gov.au

 

 

Click here for a list of Conference papers on the Publications page