This is the MODIS/Aqua subset along MLS field of view track. The goal of the subset is to select and return MODIS data that are within +-100 km across the MLS track. I.e. the resultant MODIS subset swath is sought to be about 200 km cross-track. Geolocations in the original product, however, are subsampled at 5-km, and thus the crosss-track width of the subset geolocations is 41 pixels. The subset Cloud Mask has 201 pixels across-track. However, some of the Cloud Mask information is at bit level, thus allowing storing actually 250-m information in seemingly 1-km pixels. This is achieved by reserving 2 bytes (the last two out of six) of every 1-km pixel as 16 yes/no-cloud bits. Each one of these 16 bits flags a corresponding 250-m tile inside the 1-km pixel. Their state is described in the local attributes to the Cloud_Mask HDF data set, and accordingly must be interprated as 0=YES, 1=NO. Thus the effective cross-track width, for these two bytes only, is 804 tiles of 250-m denomination.
Along-track, all MODIS pixels from the original product are preserved.
In the standard product, the MODIS level-2 cloud mask product is a global product generated for both daytime & nighttime conditions at 1-km spatial resolution (at nadir) and for daytime at 250-m resolution. The algorithm employs a series of visible and infrared threshold and consistency tests to specify confidence levels that an unobstructed view of the Earth's surface is observed. An indication of shadows affecting the scene is also provided. The 250-m cloud mask flags are based on the visible channel data only. Radiometrically accurate radiances are required, so holes in the cloud mask will appear wherever the input radiances are incomplete or of poor quality. The shortname for this Level-2 MODIS cloud mask product is MYD35_L2 and the principal investigator for this product is MODIS scientist Dr. Paul Menzel (paulm@ssec.wisc.edu).
(The shortname for this product is MAM35S0).