The Brazilian Amazon covers an area of 5 millions square kilometers, concentrates a large portion of the biodiversity, and plays an important role for the biochemical cycles of the Earth Systems. However, this region suffers an intense process of deforestation. This fact requires a systematic data acquisition in order to evaluate the current state of this important ecosystem. In this context, radar remote sensing can be considered an important monitoring tool. The fact that microwaves almost do not suffer from atmospheric interference, allows a constant climate independent image data acquisition. This is particularly important in areas of tropical forest, like Amazônia, where the cloud covering is intense during most part of the year. Moreover, the way radar signals interacts with the forest allows a relation with biophysical forest parameters, and the discrimination of some land cover patterns, including deforestation. Polarimetry is a new field of innovation in the SAR technology, which explores the polarized nature of the electromagnetic waves to extract additional information of the targets, many times critical for certain tasks. From the amplitude and phase between polarizations, statistical parameters and other object dependent electrical attributes can be extracted, allowing the quantitative characterization of the scattering properties of the observed targets. The additional information offered by the polarimetric systems can improve the capability of discriminating different land cover patterns. In this way, the objective of this project is to analyze the potential of the polarimetric radar images from ALOS/PALSAR, recently launched, for the classification of the land use and land cover in Amazônia, which is a basis for more complete and complex studies of the evolution of this ecosystem, patrimony of the humanity, so seriously endangered.
News published in Agência FAPESP Newsletter about the scholarship: