EARLY VALIDATION OF MSG VEGETATION PRODUCTS OVER THE IBERIAN PENINSULA

The validation of satellite derived products is an important development for remote sensing since high-quality and consistent data sets of known accuracy with product continuity between instruments and missions are clearly important goals. In this context, the objective of this work is to accomplish a first validation of MSG vegetation products, fraction of vegetation coverage (FVC) and leaf area index (LAI). This goal is achieved by two different validation activities, the direct and indirect validation. For the direct validation, the LandSAF version 1 algorithm, used to derive the FVC and LAI, is applied on VEGETATION SPOT (VGT) due to the MSG available data showed a poor quality in the study area. High-resolution biophysical maps derived from in-situ measurements are used in order to provide us with product uncertainties over a 50×50 km2 area centred at Barrax test site (Albacete, Spain). The indirect validation is performed by means of an inter-comparison between products derived from different satellite sensors and algorithms over the Iberian Peninsula. Different LAI/FVC products derived from data of different spatial resolutions are used, such as VEGETATION/SPOT, SEVIRI/MSG and MODIS/TERRA. Statistical indicators (e.g. rms, bias, standard deviation, etc.) and spatial correlation analysis are applied in order to assess the competitiveness of MSG products with regard to other currently used products. Results show a good performance of the LandSAF version 1 algorithm on the LAI/FVC estimates. The operational version of this prototype algorithm has been significantly improved, which will provide promising quality improvements on MSG products according to user requirements. 1