In the first step, the LER filter and a priori CH4 cloud filter modules of the pre-processor are launched in order to discard the ground pixels with too little LER value and too much cloud contamination.
In the second step, the SICOR full physics retrieval approach is used to infer CO columns from the spectral window 2324-2338 nm. Here, accurate a priori knowledge of the atmospheric CH4 abundance (from TM5 CTM) is used to retrieve effective cloud parameters from CH4 absorption bands simultaneously with atmospheric CO and H2O abundances. The algorithm uses two highly efficient numerical modules.
The inversion infers the trace gas column information by scaling a CO reference profile. This method, which from a theoretical point of view is identical to a regularized retrieval using infinite regularization strength, allows also to calculate the column averaging kernel. The solution is obtained by a non-linear least square minimization.
The retrieval employs a two-stream radiative transfer model, which is linearized with respect to the parameters to be retrieved. The two-stream approximation is the simplest approximation to account for multiple light scattering in atmospheric radiative transfer and its numerical implementation comes with low computational cost.
Fitting window: 2324-2338 nm (spectral band 7).