Evaluation of the spectral shape matching method (SSMM) for correcting the atmospheric effects in the satellite VIS/NIR imagery

An effort has been made to evaluate spectral shape matching method (SSMM) for correcting atmospheric effects in the Landsat VIS/NIR imagery, in order to estimate the desired water-leaving radiance spectra and therefore the water-constituents´, particularly in optically complex Case-2 waters off the Korean coast in the southern peninsula. This method is compared with in-situ data and other classical atmospheric correction algorithms such as the 6S radiative transfer model and standard SeaWiFS atmospheric algorithm. Assuming a standard atmosphere with constant aerosol loading and a uniform, Lambertian surface carries out the atmospheric correction through 6S radiative transfer code, whereas the standard SeaWiFS atmospheric correction algorithm assumes that water-leaving radiance values at the two near-infrared (NIR) bands are negligible to enable retrieval of aerosol reflectance in the correction of ocean color imagery. The atmospheric correction through SSMM is accomplished by assuming the path signal, that is obtained from the total signal recorded at the top of the atmosphere (LTOA) by the use of match-up in-situ water-leaving radiance spectra for typical clears, to be spatially homogeneous contribution of the scattered signal from aerosols and Raleigh particles and the sea surface. The results indicate that the overall shape and magnitude of retrieved spectra with SSMM appear to be in good agreement with in-situ spectra collected from both clear and turbid waters. Because of the standard atmosphere with constant aerosols and models adopted in 6S radiative transfer code, a large error is possible between the retrieved and in-situ spectra. Further extension of SSMM to SeaWiFS ocean color imagery reveals that accurate estimation of water-leaving radiance spectra is feasible with SSMM but not with standard SeaWiFS atmospheric correction algorithm that yields unrealistic water-leaving radiance values in turbid waters, primarily because the water-column reflectance interferes with the atmospheric correction based on the 765 nm and 865 nm spectral bands.