The growth of satellite-based earth observation data, combined with increases in the capability of drones, potentially provides sufficient data to answer questions around complex coastal and oceanographic phenomena. However, this deluge of imagery requires a revolution in data analysis. Big data is no use if it cannot be converted to knowledge. As remote sensing data floods in, machine learning will have a critical role to play in turning data into oceanographic insight. Machine learning opens up a new world of nonlinear models and rapid hypothesis testing, helping reduce highly dimensional remote sensing datasets into forms useful for analysis. The community is responding by applying machine learning approaches across disciplines. This session welcomes examples of how machine learning can be used to answer a breadth of coastal and oceanographic questions across scales and data streams; including but not limited to retrieval algorithms, anomaly detection, habitat mapping, prediction, and model parameterization.