Cloud Screening Algorithms
Carsten Brockmann(1) and Kerstin Stelzer(1)
(1) Brockmann Consult, Max-Planck-Str. 2, 21502 Geesthacht, Germany
The identification of the presence of clouds in a pixel is critical for the successful higher level processing. The definition of “cloud” depends on the purpose: the retrieval of cloud physical parameters becomes meaningful only if the signal is totally resulting from radiance scattered by a cloud of an optical thickness of 4 or more. However, the retrieval of geophysical parameters over land or the ocean can be significantly degraded if only a fraction of a pixel is covered by a cloud, or if a semi-transparent cirrus cloud is present in the pixel. Level 2 processing has also different requirements for cloud screening than Level 3 processing.
Clouds are characterised by just the following characteristics: they are bright, white, high and cold. Both, MERIS and AATSR provide spectral information to assess some of these characteristics, but only both instruments together permit the exploitation of all of them at once, which is the precondition for an optimal cloud screening.
A review of the current cloud screening methods of MERIS and AATSR processing as well as of methods applied for both instruments developed in the scope of application projects is presented. These applications include methods using synergistic usage of MERIS and AATSR bands: improved thresholding techniques, cloud probability estimation, feature extraction and spectral unmixing techniques as well as clustering algorithms. A comparison of the different methods is made with respect to the different requirements as described above. Conclusions are drawn for an optimal cloud screening algorithm, providing a graduated scale which indicated the cloudiness of a pixel. The implications of such an algorithm for Sentinel 3 land and ocean processing are discussed.
Keywords: ESA European
Space Agency - Agence spatiale europeenne,
observation de la terre, earth observation,
satellite remote sensing,
teledetection, geophysique, altimetrie, radar,
chimique atmospherique, geophysics, altimetry, radar,