The objective of this work was to assess the impact of the better spatial and radiometric resolutions of the Ikonos-II panchromatic image, for the identification of coffee (Coffea arabica) planting areas, in comparison with a Landsat/ETM+ image. The area of study is situated in the city of Pedregulho, State of São Paulo, Brazil, where 50 coffee fields were selected. Information about plants features, like height, age, spacing and variety were collected. Images allowed the identification of coffee areas with different field features, and the calibrated Ikonos-II image showed the best results. Considering the areas with similar field features, images from both satellites were not efficient in the coffee identification. The atmospheric and radiometric corrections applied on the Ikonos-II image did not improve the analyses results. More than half of the identified areas in the Ikonos-II image could be found in the Landsat/ETM+ image (68%). The significant correlation between Landsat/ETM+ band 4 and Ikonos-II panchromatic channel shows a link between both satellite images.
Coffea arabica; remote sensing; panchromatic image
