Various attempts have been made to describe and map the vegetation of southern Africa with recent efforts having an increasingly ecologi cal context. Vegetation classification is usually based on vegetation physiognomy and floristic composition, but phenology is useful source of information which is rarely used, although it can contribute functional information on ecosystems. The objectives of this study were to identify a suite of variables derived from time-series NDVI data that best describe the phenological phenomena of vegetation in southern Africa and, secondly, to assess a classification of pixels of the study area based on NDVI variables using a pre-existing map of the biomes that was delimited on the basis of life forms and climate. A number of variables were derived from the satellite data for describing phenological phenomena, which were analysed by multivariate techniques to determine which variables best explained the variation in the satellite data. This set of variables was used to produce a phenological classification of the vegetation of southern Africa, the results of which are discussed in relation to their concordance with the existing biome boundaries.

Abbreviations: AVHRR = Advanced Very High Resolution Radiometer; NDVI = Normalized Difference Vegetation Index; NOAA = National Oceanic and Atmospheric Administration.