ERS, ENVISAT AND SENTINEL DATA APPLICATION FOR CHANGE DETECTION IN AMAZON COASTAL ENVIRONMENTS

Abstract

The Amazon coast is marked by a high discharge of sediment and fresh water under the influence of the Amazon river. A broad continental shelf, extensive flood flats and higher plateaus, formed in older sedimentary rocks, characterize this coast. Coastal environments are naturally vulnerable to morphodynamic processes, such as erosion and sediment deposition, and this condition intensifies in wet tropical conditions. This study presents an approach for morphological mapping and change detection in coastal zone, based on C-band Synthetic Aperture Radar (SAR), in areas with intense coastal dynamics and systematic cloud cover. The method is based on the amplitude SAR processing chain for a temporal profile of ERS-1/2, ENVISAT and SENTINEL-1A data, acquired between 1992 and 2015 and provided by the European Space Agency (ESA). The backscattering coefficient primary information, as well as auxiliary data extracted from the Haralick texture and shape indices were used to describe coastal environments. The total area of coastal changes detected between 1992 and 2015 was 646.15 km(2).year(-1), with erosive sedimentary balance in -2.32 km(2) under the rate of -0,10 km(2).year(-1). The time profile of coastal changes is in equilibrium and presented an erosive condition in the period from 1992 to 1993 and from 1996 to 1999, while in the period from 1993 to 2006 and 2006 to 2015 the accretion condition occurred. The accuracy of the maps of environments and coastal changes, respectively obtained an overall accuracy of 78.35% and 48.54%, and Kappa coeffi cient of 0.65 and 0.32. The ERS, ENVISAT and SENTINEL data allowed a satisfactory mapping of the morphological changes in the flat and dynamic relief of the Amazonian coast due to the availability of a historical series of C-band SAR images with parallel vertical polarization and steeper to shallow incidence. Such application was essential for the systematic monitoring of the shoreline on a regional spatial scale (1: 250,000) and eventual temporal scale (annual to decadal).