I'm trying to determine, day or night, whether or not a GOES image pixel represents a clear view of the land, or if it's obscured by cloud. I believe the relevant channels are CMI_C13, CMI_C08, and CMI_C10, but I'm not sure what function to use to combine them. Does someone have a heuristic for this, or does it require something more complex (like a decision tree or neural network)?
(I'm using Google Earth Engine and the https://developers.google.com/earth-engine/datasets/catalog/NOAA_GOES_16_MCMIPC dataset.)
