Difference between the mean EE spectra of the 300 end-to-end simulations and the noise-free simulation; both share the same input CMB and foreground maps. The difference between the two spectra, shown in blue, is smoothed in Δℓ = 20 flat bins. The blue shaded area gives the variance between the 300 simulations. The green line is the beam-leakage correction model shown for scale; it is cancelled in the difference between the mean end-to-end simulation and the noise-free spectra. For scale as well, the grey line presents 5% of the input simulation power (e.g., CMB plus foreground from the Planck Sky Model). From top to bottom the panels are the 100 × 100,143 × 143, 100 × 143, and 217 × 217 spectra. Apart from the lowest bin (ℓ = 20, which is not used for the likelihood analysis), the 100 × 143 spectra do not show any significant residual, while all of the others show signs of possible correlated noise. We do not show the other cross-spectra that behave similarly. In the auto-frequency spectra plot, we show (with a red line) the empirical end-to-end EE correlated noise model, produced by fitting smoothing splines to the blue lines, after removing strongly varying trends at low ℓ and high ℓ.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.