Case 2 Regional Coast Colour: a neural network-based framework for atmospheric correction and in-water retrievals across multiple ocean colour satellite sensors

Since the 1990s, Doerffer and Schiller have been developing physics-based neural network algorithms for analyzing ocean colour in satellite imagery of optically complex coastal waters. At its core, the approach uses neural networks to solve the inversions in various aspects of solar radiative transfer in both the atmosphere and water, including atmospheric correction, towards the estimation of inherent optical properties (IOPs) of the water constituents. Empirical bio-optical models are then applied to derive constituent concentrations from these IOPs. Over the years, this algorithm has evolved significantly and is now widely recognized as Case-2 Regional CoastColour (C2RCC), a trusted tool within the ocean colour research community. Originally designed for the MERIS sensor aboard ENVISAT, C2RCC is now the operational ground segment processor for generating Case-2 (complex) water products from Sentinel-3 OLCI data and from Sentinel-2 MSI data in the Copernicus Marine High Resolution Ocean Colour Service. Adaptations of the algorithm have also been developed for other satellite missions, including SeaWiFS, MODIS, VIIRS, Landsat OLI, and Sentinel-2 MSI. The C2RCC processor is freely accessible through the Sentinel Application Platform (SNAP). This article provides an overview of the background and evolution of the C2RCC algorithm, presenting validation results at coastal sites and in land waters alongside user performance evaluations analyzing the influence of system vicarious calibration gains. It highlights cases where the algorithm delivers reliable results as well as its limitations and areas for future improvement. In its current iteration for Sentinel-3 OLCI, C2RCC performs effectively, particularly in moderately absorbing or scattering Case-2 waters.