Remote sensing of the seasonal variability of vegetation in a semi-arid environment

This case study assesses the use of satellite data to monitor the vegetation cover in two types of semi-arid environments - sand dune and rocky - located in the Negev Desert of Israel. In this study, satellite images acquired by NOAA/AVHRR over a time period of 2 years were analysed. The phenological spectral characteristics of vegetation in the two environments are very similar. Comparison between the AVHRR-derived NDVI (Normalized Difference Vegetation Index) values and rainfall data shows dependence of the NDVI values on the sum of the amount of rainfall during the concurrent month and the two previous months. Field observations show that the vegetation components in the two semi-arid environments respond to rainfall with a time lag. The satellite-observed peak of NDVI occurs at the same time as the peak of the delayed response of annuals and perennials to rainfall. (C) 2000 Academic Press.