Bathymetric digital elevation model generation from L-band and X-band synthetic aperture radar images in the Gulf of Naples, Italy: innovative techniques and experimental results

The present Ph.D. Thesis investigates remote sensing bathymetric techniques in coastal area based on the use of Synthetic Aperture Radar (SAR) data. Since SAR signal is not able to pass sea surface and to reach sea bottom, indirect processes with sea floor morphology sensed through the effects it may have on sea surface are investigated. The conventional bathymetric techniques are briefly analyzed and the main advantages related to the use of remote sensing SAR-based bathymetric techniques highlighted. Proper sea surface modeling is performed and its interaction with SAR is investigated. Two different bathymetric techniques are proposed: the Wave-Based Approach (WBA) and the Current-Based Approach (CBA). Several theories for proper modeling of the involved phenomena and of their retrieval from SAR images are presented and the most promising are identified. An error budget analysis is performed and potential performance, specifications and requirements of the considered bathymetric approaches are described. With specific reference to the area of the Gulf of Naples, Italy, the Wave-Based Approach is selected because of current velocity being too slow or too irregular both in time and in intensity for proper application of the Current-Based Approach. Therefore, a WBA algorithm for sea surface spectrum retrieval from SAR data is proposed. The proposed WBA algorithm is applied on both L-band ALOS PALSAR images and X-band COSMO-SkyMed SAR images covering the area of the Gulf of Naples, Italy. The experimental results and the achieved improvements with respect to the approaches found in the available literature are presented. Digital Elevation Models (DEMs) of seabed are obtained and a correlation analysis with the retrieved water depth values from the official nautical charts provided by the Italian Navy Hydrographic Institute is performed.