Remote Sensing of Coastal and Inland Waters

The new generation of satellites (e.g. Landsat 8, HyspIRI, Sentinel 2 and Sentinel 3 …) contain sensors that enable monitoring at increased spatial and/or spectral resolution. This opens a wide range of new opportunities, amongst others improved observation of coastal and inland waters. Algorithms for the pre-processing of these images and the derivation of Level 2 products for these waters need to take into account the specific nature of these environments, with adjacency effects of the nearby land and complex interactions of the optially active substances with varying degrees of turbidity. Here a new atmospheric correction algorithm, OPERA, is presented which can deal with these highly complex environments and which is sensor generic. OPERA accounts for the contribution of adjacency effects and provides surface reflectances for both land and water targets. OPERA is extended with a level 2 water algorithm providing TSM and turbidity estimates for a wide variety of water types. The algorithm is based on a multi wavelength switching approach using shorter wavelengths in low turbid waters and long NIR and SWIR wavelengths for highly and extremely turbid waters. Results are shown for Landsat-8, Sentinel-2 and MERIS for a variety of scenes, validated with field aeronet and turbidity data.