Coastal bathymetry from satellite high resolution monitoring

Abstract

[EN] Bathymetry is traditionally obtained by echo sounding technology. However, bathymetry can be also obtained from satellite imaging, which is much more cheaper than echo-sound measurements. This is obtained by analyzing the waves near to the shoreline. In order so, wave properties such as wavelength and celerity should be measured, after which the bathymetry is estimated using linear wave theory. In this internship a new method based in the continuous wavelet transform has been implemented. In order to obtain the celerity, two images with a time lag are needed. Two data sets are used. On the one hand a video product, with 12 Pléiades images with a time lag between them of 8s. On the other hand a set of Sentinel-2 images. In the latter, a time shift between bands because of a lag in the acquisition is exploited. An application for the extraction and preparation of Sentinel-2 data in a form of a Graphical User Interface has been implemented. The site that has been studied will be the shore of Capbreton, which hosts one of the world’s deepest canyons. The images have been be pre-filtered by using FFT and Radon filters, with several methods that include windowing of fixed and variable size. Those filtering techniques have be implemented and its results compared. Best results are obtained using a variable-size windowing technique. Finally, the wavelet method has been applied to both datasets to achieve wave propagation information.