Retrieval of ozone profiles by using GOME measurements and a neural network algorithm

GOME is the first instrument of ESA capable of monitoring global distribution of ozone and other trace gases. To extend the use of GOME products further towards more operational applications, it is needed to complement the off-line service with a product delivery which should be as fast as possible. The neural network-based approach proposed in this study gives a contribution to this direction. The retrieval algorithm, which also includes an appropriate normalization of the measured quantities, is presented and the results discussed.