Radiance spectra classification from the ocean color and temperature scanner on ADEOS

Multispectral information from the ocean color sensors of remote sensing sa tellites can be used to classify the ocean surface waters into a number of classes. Nine areas distributed over the Pacific Ocean have been used to de monstrate this approach. Unsupervised neural networks were used to separate water pixels from land and cloud pixels and classify water into a variety of ocean colors. Self organizing feature maps chose radiance spectra by min imizing least square differences amongst multichannel pixels. Pixels with s imilar radiance spectra were coded with similar colors, It has been shown t hat radiance spectra, after correction for the atmospheric absorption of a "standard atmosphere" for varying sun and satellite viewing angles, could b e classified into a single set of radiance spectra that apply over the whol e ocean. No ground truth data was required to make this classification. Examinations of the classified images showed that the method could extract a large number of ocean color categories and provide a basis to separate ca se 1 waters from the case 2 and ocean radiances with a high influence of th e atmosphere. Also, areas of high pigment, inappropriately masked out by th e Conventional routine, were correctly classified. This opens the possibili ty that in the future a robust global algorithm for chlorophyll estimation might be constructed.