Multi-Sensor Data Fusion for Geomorphological and Environmental Sensitivity Index Mapping in the Amazonian Mangrove Coast, Brazil

This research emphasizes the potential of space-borne optical and synthetic aperture radar (SAR) data for coastal geomorphological and environmental sensitivity index (ESI) mapping for oil spill in the macrotidal Amazon Region. Optical (Landsat-5 TM) and SAR (RADARSAT-1 Wide 1) imageries were evaluated regarding geomorphological information content. Particular interest was placed on the Selective Principal Component-SAR (SPC-SAR) product, an alternative scheme for the fusion of TM bands with Wide Mode data through the combination of Principal Component and Intensity-Hue-Saturation Transforms. TM images have contributed to enhance vegetation and sedimentary environments based on the optical response, while Wide Mode data has allowed enhancing differences between inundation frequency and distinct moisture content. Results of this investigation have revealed significant geomorphological information with 12 units mapped in the study area. The lowest ESI was assigned to cliffs (ESI 1) and the highest was marsh and mangroves, respectively (ESI 10A, 10C). The SPC-SAR product represents a powerful tool for geomorphological mapping, providing relevant information about geobotany, emerge and submerge coastal geomorphology. An annual monitoring of both natural and human-induced changes would provide information for the construction of multi-temporal ESI maps to oil spills in different seasons and tidal cycles in the Amazon coastal zone.