Methodology
The SLATS team has developed a rigorous methodology to detect and map land cover change across Queensland.
Acquiring imagery
The SLATS team uses images from the Landsat satellites. Each image or ‘scene’ represents approximately 185km by 185km on the ground, so covering the whole state requires over 80 scenes.
The coverage of specific scenes can be viewed in the SLATS Vegetation Sites and Foliage Projective Cover Sites map. (PDF, 787K)*
Landsat takes multiple images of the same area each year. Where possible, the SLATS team selects images from the dry season. Not only are these images less likely to be affected by cloud, it is also easier to distinguish different vegetation types. Grass tends to dry out while woody vegetation remains greener.
Preparing the imagery
Each image needs preparation before it can be compared with images from previous years.
- Radiometric correction, to address variation in sun angle and atmospheric effects
- Geometric correction and referencing, to ensure individual pixels represent the same location on the ground each year
- Masking, to remove areas of cloud, cloud shadow, smoke, bad data and water from the analysis
Processing and classification
Land cover change is identified through two steps.
- Automated classification based on analysis of the foliage projective cover (FPC) over time
- Manual raster editing to supplement the initial classification and adequately isolate the woody vegetation change
Field verification
Once an initial assessment has been made, scientists check their mapping of woody vegetation change through extensive fieldwork across Queensland:
- Checking cleared areas and areas of uncertainty
- Collecting ground control points for geometric correction
- Estimating the disappearance rate for coarse woody debris, to assist in determining greenhouse gas emissions
Detailed information is collected during a field survey. A modified version of the SLATS field survey method has been trialled and endorsed by all states and territories and the federal government through the Department of Agriculture, Fisheries and Forestry.
Finalising the data
The SLATS team finalises areas of change based on the field observations. The final classification is closely peer-reviewed by senior staff with extensive experience to ensure accuracy and consistency is maintained across all scenes. A final dataset is then prepared, including metadata consistent with Australian Spatial Industry Standards.
More information
Contact the SLATS Principal Scientist: dave.harris@science.dsitia.qld.gov.au
* Requires Adobe Reader
Last updated 3 January 2013
