The forest-savanna dynamics from multi-date Landsat-TM data in Sierra Parima, Venezuela

Two Landsat-TM images are used for the identification, characterisation and spatialisation of vegetation changes. In this approach a methodology using a differential image from the vegetation index (MSAVI) is employed. The MSAVI vegetation type relation enabled the identification of vegetation changes. Finally, spatialisation of the transformation was obtained by multi-date classification. The evolutive vegetation trends were interpreted in conjunction with relief, vegetation and land use data. The results show that human activity not only induced degradation but also the regeneration of abandoned areas. The overall ecological balance shows degradation outweighed regeneration. The method is appropriate to monitoring and interpreting of vegetation dynamics in tropical rain forests.