BUREAU DE RECHERCHES GEOLOGIQUES ET MINIERES (BRGM) (FR)
The BathySent project aims at the development of an automated method for mapping coastal bathymetry (water depths) on the basis of Copernicus Sentinel-2 mission. The interest of using Sentinel-2 data lies on the capacity to cover large areas (National and European scale targeted), while benefiting from the high repeat cycle (5 days) of the mission. The systematic acquisition plan of Sentinel-2 is of major interest for studying and monitoring coastal morphodynamics. The proposed methodology avoids limitation of exiting techniques in terms of dependency on water turbidity and requirement for calibration. The main objective of the project is to propose a method for deriving coastal bathymetry on wide areas (National/European scale) based on Sentinel-2 data and assess its performances.
Today knowledge of near-shore bathymetry is essential for multiple applications such as for the study of submarine morphodynamics. These data are vital for planning sustainable coastal development, coastal risks assessments (including tsunamis) and conservation of submarines ecosystems. Moreover, they represent a crucial input for near-shore navigation and submarine resources exploration.
The reasons why space-borne remote-sensing techniques must play an essential role in retrieving near-shore bathymetry are threefold. First, space-borne imagery makes it possible to access remote areas with wide spatial coverage at high spatial resolution. Second, because space-borne imagery is acquired on a regular basis, a historical data archive is accessible for most sensors, which enables scientists to access information from the past. Third, the cost of the data is relatively affordable compared to airborne or ground missions.
In the BathySent project, we propose to extract bathymetry from a single Sentinel-2 dataset, exploiting the time lag that exists between two bands on the focal plane of the Sentinel 2 sensor. To tackle the issue of estimating bathymetry using two Sentinel 2 images acquired quasi simultaneously, we plan to develop a method based on cross-correlation and wavelet analysis that exploits the spatial and temporal characteristics of the Sentinel 2 dataset to jointly extract both ocean swell celerity (c) and wavelengths (λ). Our team has already started to develop this method based on the French Space Agency’s (CNES) SPOT 5 dataset (Système Probatoire pour l’Observation de La Terre) with promising results (Pourpardin et al., 2015). We called it the CWB method, which stands for Correlation, Wavelets and Bathymetry. Our method combines the direct measurement of c presented in (de Michele et al., 2012) with an original wavelet-based adaptive λ estimate (that we published in Poupardin et al., 2014) to retrieve a spatially dense cloud of (λ, c) couples that are then used to estimate water depth (h) via the dispersion relation presented in equation (1).
The method preferably applies to the zone between the coast and an area of depth less than or equal to half the wavelength of the waves (typically up to a hundred meters deep), with the exception of the wave breaking zone.
Poupardin, A., D. Idier M. de Michele D. Raucoules “Water depth inversion from a single SPOT-5 dataset” IEEE Trans. Geosci. Remote Sens. vol. 54 no. 4 pp. 2329-2342 Apr. 2016.
de Michele M., Leprince S., Thiébot J., Raucoules D., Binet R., 2012, “Direct Measurement of Ocean Waves Velocity Field from a Single SPOT-5 Dataset”, Remote Sensing of Environment, vol 119, pp 266–271.