Multi-Temporal Classification of Abandoned Agricultural Land in Gadarif Region, Sudan, Using Original Bands, Indices and Feature Components of Spaceborne Imagery

The generation of accurate land-use/land-cover maps for the Gadarif region, Sudan, represents the first step in developing conservation and management plans for natural resources. The objective of this study is to evaluate combinations of original bands and groups of indices and feature components subjected to standardized principal component analysis (SPCA) for a multi-temporal classification of abandoned agricultural land using Landsat ETM (March 2003) and Terra ASTER (February 2005) imagery. The results of the study indicate substantial improvement for multi-temporal classification of abandoned agricultural land, for rapid characterization of vegetation succession. SPCA generated from indices and feature components can be recommended to obtain accurate classifications for areas covered by sparse heterogeneous vegetation in other regions of the Sahel zone.