The algorithm has the arduous task of separating 'good' pixels from 'cloudy' pixels. The standard M?D35 cloud mask includes using the brightness in the visible channels to identify clouds. This procedure will mistake heavy aerosol as 'cloudy', and miss retrieving important aerosol events over ocean. On the other hand, relying on IR-tests alone permits low altitude, warm clouds to escape and be misidentified as 'clear', introducing cloud contamination in the aerosol products. Thus, our cloud mask over ocean combines spatial variability tests (e.g. Martins et al., 2002) along with tests of brightness in visible and infrared channels. Underwater sediments have proved to be a problem in shallow water (near coastlines) as the sediments can easily have land-like surface properties. Thus, the sediment mask of Li et al. (2003) is used in addition to the cloud mask.
The algorithm sorts the remaining pixels that have evaded all the cloud masks and the sediment mask according to their r0.86 value, discards the darkest and brightest 25%, and thereby leaves the middle 50% of the data. The filter is used to eliminate residual cloud contamination, cloud shadows, or other unusual extreme conditions in the box. For the 10 km retrieval, at least 10 of the 400 pixels in the original box must remain after masking and filtering. If the minimum number of pixels is not met, no retrieval is attempted and fill values are given for the aerosol products in the retrieval box.
The ocean algorithm was designed to retrieve only over dark ocean, (i.e. away from glint). Glint angles of < 40˚ are masked. There is a special case where heavy dust can be retreived over glint, and the retrieval will be assigned a poor quality.