Land Cover Dynamics in Miombo Ecosystems

Characterizing the dynamics of landscape spatial patterns in the Miombo is fundamental to understanding the impacts of changes in landscape structure on ecological processes and functions. Land Cover Dynamics in Miombo Ecosystems examines methodologies for computer-aided classification of Landsat TM data for mapping land use and land cover in Miombo ecosystems in Southern Africa. The work compares supervised and unsupervised methods for classifying Landsat TM and developed a hybrid supervised-unsupervised classification method Stepwise Thematic Classification (STC). STC substantially improved both the accuracy and reliability of the classification process. Spatial pattern analysis was used to quantify changes in landscape metrics as a result of land cover transitions under different land tenure systems at locations in Malawi and Zimbabwe.