This paper presents the results of a study related to the definition and

the validation of processing lines for the exploitation of the MERIS

data over terrestrial surfaces. Two algorithms have been developed: the

first one focuses on the aerosol correction scheme and the second one on

the retrieval of vegetation biophysical parameters from the existing

MERIS L2 standard products. The vegetation products are the Leaf Area

Index (LAI), the fraction of Absorbed Photosynthetically Active

Radiation (fAPAR), the canopy chlorophyll content (LAIxCab) and the

fraction of vegetation cover (fCover). The vegetation products are

usually retrieved from Top Of Canopy reflectances. However, the

reflectances provided by the MERIS level 2 land products are Top Of

Aerosol, i.e. they are only corrected of the atmospheric molecular and

gaseous contributions and not of the aerosol effect. Consequently, a

definition and an implementation of an aerosol correction scheme for the

removal of aerosol scattering contributions over land has been proposed

in the frame of this study before the generation of the vegetation

products. The selected algorithm for the aerosol correction scheme is

based on a simultaneously estimation of the aerosol properties (Aerosol

Optical Thickness and Angstrom coefficient) and the correction of the

atmospheric effects. The algorithm to derive the vegetation products is

based on a modelling inversion method using a Neural Network Technique.

This technique needs a training database on which the network is

calibrated. This database has been simulated with the help of radiative

transfer models.