Comparing Operational MSG/SEVIRI Land Surface Albedo Products From Land SAF With Ground Measurements and MODIS

This study is devoted to an assessment of the quality performance of the Satellite Application Facility on Land Surface Analysis (LSA-SAF) albedo products, which are generated from the high frequency of observations now offered by the Meteosat Second Generation (MSG) geostationary satellite series. Land surface albedo products are compared against in situ measurements and the analogous products from the Collection 4 of Moderate Resolution Imaging Spectroradiometer (MODIS) in domains in European and Northern African regions. We find a good consistency with MODIS broadband shortwave and near-infrared directional-hemispherical reflectance (DHR) over midlatitude regions. Comparative statistics indicate biases below 0.01 in absolute and 5% in relative terms, and an average root-mean-square error (RMSE) around 0.03 for all situations, including in snow cases. The comparison for shortwave bihemispherical reflectance (BHR) shows an even better match over herbaceous cover with respective values for bias and RMSE of 0.007 and 0.012. Broadband visible BHR from LSA-SAF overestimates MODIS equivalent albedo product by 20%, which seems to be caused by remaining uncertainties in narrowband-to-broadband conversion, residual aerosol signals, and the use of a different bidirectional reflectance distribution function (BRDF) model. The commendable statistical results obtained over bright desert targets such as Northern Africa are worth noticing. Assuming that the MODIS product is a good reference, a relative bias of 10% can be judged satisfactory for the LSA-SAF snow-free broadband DHR values. In addition, quarter-hourly MSG observations better capture snow episodes, although some discrepancies appear for areas with high vegetation cover.