Adjacency effects in Sentinel-2 data: application of SIMEC (SIMilarity Environment Correction)

With its high spatial resolution, and suitable spectral bands, Sentinel-2 data can support the implementation of the Water Framework Directive by providing systematic observations over (small) inland waters, estuaries and coastal waters. However before remote sensing can be used in a fully operational way, the correction for atmospheric scattering over waters close to land surfaces needs to be satisfactorily addressed. Such "adjacency effects" modify the spectral signature of the observed pixel, particularly in the near infrared.Here we show the results of the SIMEC adjacency correction method applied to an airborne dataset spectrally and spatially resampled in order to mimic S-2 data. First, a detection of the magnitude of the environment effects is performed based on the deviations from the NIR similarity spectrum. Secondly, the SIMEC algorithm is applied to the data to correct for adjacency effects