Applying artificial neural network methodology to ocean color remote sensing

Artificial neural networks (ANN) are widely used as continuous models to fi t non-linear transfer functions. In this study we used ANN to retrieve chlo rophyll pigments in the near-surface of oceans from Ocean Color measurement s. This bio-optical inversion is established by analyzing concomitant sun-l ight spectral reflectances over the ocean surface and pigment concentration . The relationships are complex, non-linear, and their biological nature im plies a significant variability. Moreover, the sun-light reflectances are u sually measured by satellite radiometers flying at 800 km over the ocean su rface, which affect the data by adding radiometric noise and atmospheric co rrection errors. By comparison with the polynomial fit usually employed to treat this problem, we show the advantages of neural function approximation like the association of non-linear complexity and noise filtering. (C) 199 9 Elsevier Science B.V. All rights reserved.