A method of automatically compensating a multi- or hyper-spectral, multi-
pixel image for atmospheric effects, comprising resolving a plurality of spectrally-diverse pixels from the image, determining a spectral baseline from the spectrally-diverse pixels, determining a statistical spectral deviation of the spectrally-diverse pixels, normalizing the statistical spectral deviation by applying a scale factor, and compensating image pixels with both the spectral baseline and the normalized spectral deviation. Another embodiment features a method of automatically determining a measure of atmospheric
aerosol optical properties using a multi- or hyper-spectral, multi-
pixel image, comprising resolving a plurality of spectrally-diverse pixels from the image, determining a statistical spectral deviation of the spectrally-diverse pixels, correcting the statistical spectral deviation for non-
aerosol transmittance losses, and deriving from the statistical spectral deviation one or more
wavelength-dependent
aerosol optical depths. A final embodiment features a method of automatically determining a measure of atmospheric gaseous optical properties using a multi- or hyper-spectral, multi-
pixel image, comprising resolving a plurality of spectrally-diverse pixels from the image, determining a statistical spectral deviation of the spectrally-diverse pixels, and deriving from the statistical spectral deviation
wavelength-dependent gaseous optical depths.