Abstract
Extremely precise cosmic background radiation (CMB) data from Planck satellite confirmed the cosmological model λCDM and established tight constraints for several features of the primary and secondary CMB temperature fluctuations. Possible non-Gaussian (NG) contributions to the CMB field could be originated during (or soon after) primordial cosmic inflation, where the types, scale dependences, and intensities expected depend on the inflationary models. For this, the robust detection -or not- of primordial NG in the CMB signify a unique probe to the early universe, allowing to distinguish between competing models. Recent analyses from Planck CMB data strongly limit the level of NG and show consistency with the Gaussian hypothesis although they do not exclude the presence of weak Gaussian deviations. A problem inherent with the confidence of a positive detection is the possibility that any non-primordial contamination could be mixed with primary contributions leading to qualitative and/or quantitative imprecise detections. A variety of methods are being used to search for different NG signals in CMB data because one does not expect that a single statistical tool can be able to identify all possible forms of Gaussian deviations. Using two directional large-angle NG indicators based on skewness and kurtosis statistical momenta of patches of the CMB sphere, we analyze the three nearly full sky foreground-cleaned Planck maps: smica, nilc, and sevem. Our results show that these foreground-cleaned Planck maps exhibit different levels of NG at large angles, depending on the cut-sky mask used (all of them released by the Planck collaboration). The separation component minimum mask, termed M82, and the U73 mask appear to be equally efficient to Gaussianize all these CMB Planck maps.
