Distribution of observables before and after each step of pre-processing from the mock input data with 2 populations of galaxies (Ellipticals+Spirals) described in Sect. 7.4. The dark red, orange and yellow areas in the contour plots are the regions that enclose 99%, 95% and 68% of the points respectively. Top left panel: scatter plot of the FLUX_AUTO of extracted sources (in ADUs) in filters uiK_s and their covariances. Top right panel: same plot, but with the dynamic range of the FLUX_AUTO distributions reduced via Eq. (16). Bottom panel: same plot, after whitening of the reduced distribution of observables. The latter distribution is uncorrelated, centered on the mean of the distribution and rescaled, allowing for a much more efficient binning process than on raw fluxes, and a more practical comparison with the simulated observables.