Near-Real-Time Estimation of Spectral Surface Albedo from GOME-2/METOP Measurements

Abstract

An algorithm was developed for estimation of spectral surface albedo from top-of-atmosphere (TOA)-radiances measured by the Global Ozone Monitoring Experiment GOME-2 flying on-board MetOp-A. The algorithm estimates Minimum Lambert-Equivalent Reflectivity (MLER) for a moving time window of variable width. The time window gets smaller for varying surface conditions with time and larger for periods with a homogeneous surface. Different methods are used for computation of the MLER dependent on the amount of data available within the time window. The surface albedo is an important input for GOME-2 trace gas retrievals, both for the near real time and the off-line operational products. Trace gas retrieval algorithms usually take surface albedo from a suitable climatolgical database, but the actual surface albedo can vary considerable, especially for snowing conditions. The first version of the spectral surface albedo algorithm has been integrated into the operational Universal Processor for Atmospheric Spectrometer (UPAS) system. Examples of surface albedo will be presented demonstrating the impact on the retrieved GOME-2 trace gases, specially on the ozone and NO2 column products. Further developments of the surface albedo algorithm are finally discussed.