Joint retrieval of surface reflectance and aerosol optical depth from MSG/SEVIRI observations with an optimal estimation approach: 2. Implementation and evaluation

[1] An original method, based on optimal estimation, was presented in a part one of this paper for the joint retrieval of the mean daily total column aerosol optical depth and the surface Bidirectional Reflectance Factor (BRF) from the daily accumulated Meteosat Second Generation-Spinning Enhanced Visible and Infrared Imager (MSG/SEVIRI) observations in the solar channels. The objective of this paper is to evaluate the benefits of the proposed approach and to document the limits of the algorithm assumptions in the context of its implementation in an operational ground segment. A twofold approach is followed. In a first step, by looking at the posterior correlation error matrix the capability of the so-called Land Daily Aerosol (LDA) algorithm to decouple the surface-atmosphere signal is analyzed. In particular, the impact of the prior information is investigated in detail. In a second step, the results of the algorithm are compared with independent data sets of aerosol optical depth and surface reflectance. In this phase, the accuracy of the algorithm is evaluated against ground observations from the AERONET network. LDA is shown to be in good agreement with these data, especially when the prior update mechanism is activated. Comparisons with the MODIS surface product showed that the bihemispherical reflectance derived from the LDA products is consistent with the equivalent MODIS white-sky albedo. Aerosol spatial distributions are comparable in terms of geographical location and intensity, in particular for aerosol episodes with a limited daily variation.