Responsable : Mahmoud Al-Najar, Erwin Bergsma https://www.kaggle.com/c/bathymetry-estimation/ The ability to track the evolution and development of the physical characteristics of coastal areas over time resulting from different natural forces is an important factor in coastal development, planning, risk mitigation, and overall coastal zone management. Traditional bathymetry surveys using echo-sounding techniques are expensive and are conditioned on different site-specific characteristics. Remote sensing tools have recently emerged as reliable and inexpensive data sources for bathymetry estimation using inversion models. In this hackathon, we will use satellite imagery to predict ocean depth. Following work on physical wave modelling, we will use wave information extracted from satellite images to learn an inverse model of wave dynamics which predicts bathymetry. Through a Kaggle competition, students will have access to 2 training sites, St. Louis and French Guiana, and will be evaluated on their models’ performance on a third site in Portugal.