The present work had the proposal of evaluating the identification and mapping of maize areas in the northwestern region of the State of Rio Grande do Sul using multi-temporal data from MODIS (Moderate Resolution Imaging Spectroradiometer) sensor on board of the Earth Observing System satellite - EOS-AM (Terra). The supervised classification algorithm Spectral Angle Mapper (SAM) was successfully applied in a multi-temporal series of pre-processed EVI images. It was verified that the areas classified as maize in the image fully agree with more extensive or continuous maize areas (> 90 ha). Small or hillsides maize areas having close to shrub and wood vegetation were not detected by the classifier mainly due low spatial resolution of images. The main practical utility of maize’s areas digital classification and identification using MODIS images is in its application to isolate or to complement the crop areas mapping, in order to monitor these areas, using different vegetation indices derived from high temporal and low spatial resolution images.
vegetation indices; supervised classification; orbital remote sensing
