Abstract
Currently, global cloud amount estimates from the National Environmental Satellite, Data, and Information Service (NESDIS) experimental cloud remote sensing algorithm, CLAVR ( Cl ouds from AV HR R ) - Phase I, are biased towards 50% cloud cover. This is caused by assigning 50% cloud amount to 2×2 global area coverage (GAC) pixel arrays which have been classified as “mixed”. The classification identifies pixel arrays that are spatially non-uniform in albedo or temperature. To remove this source of bias, a clear-sky radiation data set (CRDS) is produced from “clear” pixels located in half degree latitude/longitude grids. Statistics of each channel's radiation measurements are computed over the 9-day repeat cycle of the polar orbiting satellite. The previous 8-day CRDS record is then used to predict the clear-sky radiation values for the current day (9th day) being processed. These are used to reclassify mixed pixels into clear or cloudy categories.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
