Estimation of Coastal Bathymetry Using Wavelets

When waves propagate in coastal areas at depths lower than one half the wavelength, they exhibit a different signature at the sea surface and the observed wavelength pattern enables inferring bathymetries. Commonly, a spectral analysis using the fast Fourier transform (FFT) is employed to derive wavelength and wave direction of swell waves, in nearshore regions. Nevertheless, it is recognized that this method presents limitations, particularly regarding depth inversion limits that do not allow obtaining bathymetric data close to the shoreline. This work explores a wavelet spectral analysis to obtain bathymetric data. This new imaging methodology is applied over different seafloors with 2D and 3D features such as longshore bars or headlands. The synthetic images of the water surface are generated from a numerical Boussinesq-type model that simulates the propagation of both regular and irregular waves. The spectral analysis is carried to estimate the water depths, which are validated with the model's bathymetry. Wavelet image processing methodology shows very positive results, revealing the capabilities of this new methodology to map shallow marine environments.