Image interpretation and data integration
Baseline data
Land use is mapped primarily on the interpretation of Landsat imagery and aerial photography. An extensive library of Landsat imagery with annual coverage from 1988 to 2006 for Queensland is managed by the Remote Sensing Centre.
Other datasets are used to assist the mapping process:
- National Park boundaries identify conservation land uses
- State forests and timber reserves assist in identifying timber production areas
- The Digital Cadastral Database (DCDB) is used in identifying property and land tenure boundaries;
- The Queensland Valuation and Sales System (QVAS) is linked to the DCDB and provides a spatial view of land use information collected by valuation staff within the department. This is particularly useful for differentiating urban land uses, such as residential, commercial and industrial areas,
- Any additional project-specific datasets relevant to the catchment, which may be available within the department or from local shire councils, catchment groups, industry or from other sources of expert knowledge; and
- Other ancillary datasets
All mapping and editing is performed in ERDAS Imagine raster format (.img). Raster format preserves the spatial and attribute integrity of the land use layer, and is simpler to edit a than a complex vector polygon coverage.
Raster pixels are assigned to land-use classes according to the Australian Land Use and Management classification (ALUM):
- ALUM version 5 (February 2002) was used for 1999 mapping;
- ALUM version 6 (November 2005) is used for 2004 mapping
The raster image is converted to a vector coverage in preparation for verification and field checking.
Figure 3 - Land use information is derived by interpreting aerial photographs and satellite images. For example, areas where sugar cane is growing can be easily delineated from the surrounding woodland in the above images.
Mapping Change
The 2004 land use dataset is created by mapping changes to the 1999 baseline land use dataset using the Australian Land Use and Management Classification (ALUM), version 6, November 2005. Changes are identified, delineated and attributed through the interpretation of the above datasets.
The images below show how change is identified and mapped, in this instance using aerial photos and Landsat imagery. The final land use map for each year is shown at the bottom. In this example the land use change is from irrigated sugar to residential.
Aerial photo (left: 1999, right: 2004)
Landsat imagery (left: 1999, right: 2004)
Land Use mapping (left: 1999, right: 2004)
All updates and edits are performed in ERDAS Imagine using a raster format (.img). Once all updates and edits have been made, the raster image is converted to a vector coverage
Last updated: 20 August 2008