High-resolution optical remote sensing for coastal benthic habitat mapping: A case study of the Suape Estuarine-Bay, Pernambuco, Brazil

摘要

The benefits of high resolution remote sensing techniques and data for the mapping of coastal environments include their ability to capture the spatial variability of the rich and diverse sedimentary landforms and habitats of coastal systems as well as gather and manage different types of data and information using the same platform for multi-scale and multi-layer analysis. This work analyses the efficacy of high resolution satellite imagery for mapping benthic habitat and the seascape characteristics of a coastal lagoon (Suape Bay - 40 km south of Recife, Pernambuco, Brazil) subjected to high anthropogenic impact and exposed to different conditions of tides, suspended sediment regime, and wave dynamics, using different types of processing and classification methods. We analyzed two World View 2 images acquired in 2010 and 2015 in similar tidal conditions and low water turbidity. Preprocessing included the application of radiometric and atmospheric correction, sunglint removal, and land/sea areas masking. To extract the depth information, a Band Ratio and Simple Radiative Transfer Model were employed. The model was corrected and calibrated by using a set of 150 control points collected by echo sounding surveying. A benthic terrain model was developed to identify and map topographic features at broad and fine scales. Supervised Pixel and Object-Based classification methods were applied to visible bands of WV2 and bathymetric data to produce benthic habitat maps of the lagoon area. The analyses were validated by ground truthing based on bottom sediment sampling, underwater photographs, and side scan sonar surveys.