ABSTRACT
ABSTRACT: Remote sensing presents a valuable tool to address the highly complex issue of land degradation in dryland areas. Ten-daily composites of 8 km-AVHRR reflectance data, for band 1 (0.58-0.68 μm) and band 2 (0.725-1.1 μm), were merged to form a complete coverage of North Africa, the Horn of Africa, West and Central Asia regions (the ‘CWANA’ region), and transformed into monthly NDVI composites. This temporal NDVI dataset was converted into a land use/land cover classification, using a hierarchical decision-tree, based on the average values of the mean and maximum NDVI. In a second stage the NDVI thresholds for different agroclimatic zones were adjusted to account for weather variability. Using these procedures 17 annual land cover maps were produced for each year of the period and further condensed into ‘change’ maps, using a procedure of identifying hierarchical levels of change patterns.
