Coastal areas are dynamic and changing environments, quickly transformed by anthropogenic factors, especially in the last fifty years. The natural behaviour of coastal ecosystems is in constant evolution because of these intensive activities, and mapping bathymetric changes is a core task of a wide range of applications, such as safety of navigation, research, and coastal zone monitoring. This situation increases the necessity of improved nautical charting as, according to GEBCO, less than 20% of the world’s ocean floor has been mapped. Satellite-derived bathymetry (SDB) using remote sensing techniques has become an extremely cost and time effective tool, especially in remote areas, compared to traditional methods. Sentinel-2 is part of the open policy Copernicus programme for Earth Observation, and since 2015 provides a new opportunity to produce bathymetric charts due to their high standard of image and sensor calibration and revisit time period. Not only ARGANS EO scientists have observed that large Sentinel-2 time series provide better results than those yielded by limited numbers of costly and noisier VHR images, but they concluded that traditional SDB methods must be revisited to eliminate the undesirable depths they often produce. Our new approach dubbed ‘Water Column Parameter Estimator’ (WCPE) is based on the statistical analysis of the distribution of pixels across entire satellite images. It relies on the same Radiative Transfer Equations as previous empiric and physics-based methods but augments them with a statistical clustering tool that allows for automatic determination of the many coastal types of waters in presence along with associated parameters such as sea floor reflectance, water IOPs, DOP, etc. crucial for analysis and exploitation. This methodology was first tested in the challenging Madagascar environment selected for the GEBCO Seabed 2030 project, and has demonstrated its capability for different applications that can be derived from the observed data.



Dr Noelia Abascal-Zorrilla

Noelia Abascal-Zorrilla joined Argans in 2020 after completing a PhD in Geosciences at CNRS-French Guiana. Noelia’s background is within coastal science and she has used remote sensing data for different applications, such as mud banks characterization, litter detection, coastal erosion and SDB.