|
Waldir Renato Paradella, INPE (National Institute for Space Research) (Brazil)
Arnaldo de Queiroz Silva , INPE (National Institute for Space Research) (Brazil)
Sheila Soraya Alves Knust, INPE (National Institute for Space Research) (Brazil)
Athos Ribeiro Santos, INPE (National Institute for Space Research) (Brazil)
Cleber Gonzales Oliveira, INPE (National Institute for Space Research) (Brazil)
Tiago Nunes Rabelo, INPE (National Institute for Space Research) (Brazil)
|
|
Orbital SAR images have been successfully used for geological applications in Brazil, particularly in the moist tropical environments. However, for the Brazilian semi-arid environment (almost 900,000 km2 of the country), research dealing with geological applications based on imaging radar are extremely rare. The Cu-rich district of Curaçá Valley located on the northern border of Bahia State is characterized by a flat topography associated with residual soils, closely related to bedrocks, and sparse steppe-type vegetation (Caatinga). These conditions favor a radar investigation since is often difficult to make a direct relationship between rock types/ alteration products with SAR data. Geologically, the Curaçá Valley is part of the Salvador-Curaçá belt, a Paleoproterozoic N-S trending, granulite-amphibolite belt in the São Francisco Craton. The lithological units in the region can be grouped as Archean gneisses and granulites interbedded with ferruginous rocks, quartzites, amphibolites, mafic-ultramafic intrusives and Upper Proterozoic marbles/limestones, schists, and phyllites. The gneissic-granulitic complex of the Valley is cut by Cu-mineralized intrusives with reserves exceeding 150 million tonnes of ore (0.5 to 1.0% average Cu grade).
Therefore, the objective of the investigation was to evaluate the influence of field surface roughness (microtopography) on RADARSAT-1 and ALOS/PALSAR backscattering responses acquired over the test-area under distinct viewing geometry regarding look-azimuth (ascending and descending passes) and variable incidence (shallow and steep incidences). Four RADARSAT-1 scenes, represented by one Fine mode (F2, ascending) and three Standard modes (S2 and S7 ascending, S7 descending) and two ALOS/PALSAR images, represented by two FBD (Fine Beam Dual, ascending and descending) modes were available for the investigation. The analysis of the relationship between images backscatter and surface roughness expressed by rms-heights revealed that, excluding the influence of the dielectric constant (moisture), microtopography was not enough to explain the SAR backscatter behaviors and the spatial variations of local slope linked to the large-scale surface topography have probably played a key role controlling the SAR responses.
|
