|Paper title||Evaluation of the EarthCARE BBR Unfiltering product (BM-RAD)|
|Form of presentation||Poster|
The Broad-Band Radiometer (BBR) instrument on the future EarthCARE satellite (to be launched in 2023) will provide accurate outgoing solar and thermal radiances at the Top of the Atmosphere (TOA) obtained in an along-track configuration in three fixed viewing directions (fore, nadir and aft).
The BBR will measure radiances filtered by the spectral response of the instrument in two broad-band spectral channels; SW (0.25 to 4µm) and TW (0.25 to > 50µm). These radiances need to be corrected in the unfiltering process in order to reduce the effect of a limited and non-uniform spectral response of the instrument.
The unfiltering parametrization is based on a large simulated database of fine spectral resolution SW and LW radiances convolved with the spectral responses of the BBR channels. In practice, the SW and TW measurements of the BBR must be converted into solar and thermal (unfiltered) radiances. First, the LW radiance is estimated from the SW and TW measurements. Secondly, the inter-channel contaminations, i.e., the parts of the LW signal due to reflected solar radiation and of the SW signal due to planetary radiation, are accounted for. Finally, multiplicative factors are computed in order to estimate the unfiltered solar and thermal radiances from the SW and LW channels, respectively.
Regarding the algorithm, two unfiltering algorithms have been developed for the SW: stand-alone and MSI-based, and one stand-alone for the LW. The stand-alone algorithms aim to enable the unfiltering of the BBR if the MSI measurements are unavailable or degraded and it is done according to the measured broadband radiances and land use classiﬁcation. The MSI-based algorithm makes use of the MSI cloud mask and cloud phase in the unfiltering process.
The study presented here shows an evaluation of the BBR unfiltered radiance estimation using the three synthetic test scenes (Halifax, Baja and Hawaii) created by the EarthCARE team from the Environment Canada and Climate Change’s Global Environmental Multiscale (GEM) model and radiative transfer data derived from them.
It is worth noting that the unfiltering is a crucial part in the BBR processing, as errors in the unfiltering will be propagated to the flux (BMA-FLX product). To this end, the unfiltering performances have been confirmed not only using the test scenes (RMSE ~ 0.5 W m-2 sr-1 for SW and LW) but also using an independent validation database for both SW and LW (RMSE < 1 Wm-2sr-1 for the SW and < 0.2 W m-2 sr-1 for the LW).