Abstract
Quantitative relationships are derived among visible (0.58–0.68 μm) and near‐infrared (0.73–1.1 μm) reflectances selected according to three monthly compositing methods of the global area coverage (GAC) data of the advanced very high resolution radiometer (AVHRR) and vegetation indices (normalized difference vegetation index, NDVI, and soil adjusted vegetation index, SAVI) calculated from these reflectances. This analysis has been done using the GAC data over northern Africa (7°N‐20°N, 10°W‐10°E) for January and August of 1987. The three methods of compositing used to obtain the monthly values of the reflectances are: (1) maximum difference of near‐infrared and visible channel counts (DVI compositing), (2) maximum normalized difference vegetation index calculated from reflectances (NDVI compositing), and (3) maximum Channel 4 (10.3–11.3 μm) temperature (T4 compositing). The present analysis shows that biophysical results derived from the DVI and NDVI compositing would be more closely related and comparable than those derived from the T4 compositing. Biophysical results calculated from these composited data also depends upon the vegetation index (NDVI or SAVI) being used to calculate these results. Whereas the NDVI values calculated from the NDVI compositing are generally higher (c. 2–5%) than those from the DVI compositing, the SAVI values calculated from the NDVI and DVI compositing are less biased. Vegetation indices computed from the T4 composited data are 30–40% lower than that computed from DVI or NDVI composited data, suggesting similar differences in the biomass production computed from these composited data. Different methods of compositing from the T4 composited data were, on the average, 30–40% lower than those calculated from the DVI composited data.
