A COMBINED COLOR AND WAVE-BASED APPROACH TO SATELLITE DERIVED BATHYMETRY USING DEEP LEARNING

Abstract

Knowledge of the evolution of the littoral zone over time is paramount for coastal science and coastal zone management. However, traditional bathymetric surveys using echo-sounding techniques are unsuitable for large-scale applications due to a variety of constraints. On the other hand, remote sensing data such as satellite imagery allow for the development and application of bathymetry inversion models on a large scale. Deep learning is a growing field of artificial intelligence that allows for the automatic construction of models from data and has been successfully used for various Earth Observation and model inversion applications. In this work, we develop and apply a deep learning-based depth inversion model combining wave kinematics and water color information from Sentinel-2 satellite imagery. We present two different satellite image processing methods to augment wave kinematics and color information as inputs to the proposed deep learning-based models. We show competitive results with a state-of-the-art physical inversion method for satellite derived bathymetry, Satellite to Shores (S2Shores), demonstrating a promising direction for the use of deep learning models in Satellite Derived Bathymetry (SDB) and Earth observation in general.