干旱区生态环境知识资源中心

The Normalised Difference Vegetation Index (NDVI) can be calculated from satellite imagery and is generally recognised as a reliable index of ground vegetation cover. A simple, parsimonious model is presented to relate observed levels of the NDVI to decadal (10-day) rainfall over the preceding seven months. The model was calibrated at four stations in the North East Arid Zone of Nigeria over ten years, and the values of the empirical constants in the model were shown to be independent of location and annual rainfall. Using the mean calibrated values of the constants, the model was then validated at two further sites over four years and a good fit was observed. Such a model, based on a single input variable, has applications in predicting NDVI (and hence vegetation) patterns in response to changing climatic conditions and for extending the NDVI record to periods before satellite imagery was available. (C) 1996 Academic Press Limited