DLCD classification (Lymburner et al., 2011)

Submitted by johnathan.kool… on Fri, 26/05/2017 - 14:46
Short Description

The DLCD classification algorithm provides a basis for reporting on change and trends in vegetation cover and extent. Information about land cover dynamics is essential to understanding and addressing a range of national challenges such as drought, salinity, water availability and ecosystem health.

The DLCD analysis uses an innovative technique which reduces each time series into 12 coefficients based on the statistical, phenological and seasonal characteristics of each pixel. These coefficients were then clustered using a support vector clustering algorithm and the resultant classes were labelled using agreed National data supplied from catchment scale land use mapping and the National Vegetation Information System (NVIS).

The DLCD classification methodology and process along with major algorithms used are described in detail in the National Dynamic Land Cover Dataset - Technical Report (Lymburner et al, 2011).