Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Asymmetries in the local curvature of the WMAP data

Frode K. Hansen, Paolo Cabella, Domenico Marinucci, Nicola Vittorio

TL;DR

This study extends a local-curvature, Hessian-based test of Gaussianity to the full-sky CMB data from WMAP, applying hill/lake/saddle classifications to the renormalized temperature map and comparing observed counts to simulations using a chi-squared statistic. While the full-sky results align with Gaussianity, splitting the data into northern and southern hemispheres reveals 2–3$\sigma$ non-Gaussianity on angular scales of $1$–$5$ degrees, with maximal asymmetry near the ecliptic poles, hinting at potential systematic effects. The analysis demonstrates directional dependence of non-Gaussian signals and shows robustness across frequency channels and smoothing scales, albeit with sensitivity to sky cuts. The findings connect to prior reports of hemispherical power and correlation asymmetries, underscoring the need to account for systematics in anisotropy studies of the CMB.

Abstract

We use the local curvature to investigate the possible existence of non-Gaussianity/asymmetry in the WMAP data. Considering the full sky we find results which are consistent with the Gaussian assumption. However, strong non-Gaussian features emerge when considering the northern and southern galactic hemisphere separately, particularly on scales between 1 and 5 degrees. Quite interestingly, the maximum non-Gaussianity is found for hemispheres centered near the ecliptic poles, which might suggest the presence of some systematic effect. The direction of the asymmetry seems consistent with the findings by Eriksen et al. 2004.

Asymmetries in the local curvature of the WMAP data

TL;DR

This study extends a local-curvature, Hessian-based test of Gaussianity to the full-sky CMB data from WMAP, applying hill/lake/saddle classifications to the renormalized temperature map and comparing observed counts to simulations using a chi-squared statistic. While the full-sky results align with Gaussianity, splitting the data into northern and southern hemispheres reveals 2–3 non-Gaussianity on angular scales of – degrees, with maximal asymmetry near the ecliptic poles, hinting at potential systematic effects. The analysis demonstrates directional dependence of non-Gaussian signals and shows robustness across frequency channels and smoothing scales, albeit with sensitivity to sky cuts. The findings connect to prior reports of hemispherical power and correlation asymmetries, underscoring the need to account for systematics in anisotropy studies of the CMB.

Abstract

We use the local curvature to investigate the possible existence of non-Gaussianity/asymmetry in the WMAP data. Considering the full sky we find results which are consistent with the Gaussian assumption. However, strong non-Gaussian features emerge when considering the northern and southern galactic hemisphere separately, particularly on scales between 1 and 5 degrees. Quite interestingly, the maximum non-Gaussianity is found for hemispheres centered near the ecliptic poles, which might suggest the presence of some systematic effect. The direction of the asymmetry seems consistent with the findings by Eriksen et al. 2004.

Paper Structure

This paper contains 6 sections, 2 equations, 1 figure.

Table of Contents

  1. Introduction
  2. Local curvature on the sphere
  3. The method
  4. The results
  5. Conclusions
  6. Conclusions

Figures (1)

  • Figure 1: The discs show the centeres of hemispheres on which the $\chi^2$ has been calculated for the lake counts. The numbers indicate the percentage of simulations with a higher $\chi^2$ for the given hemisphere. In the analysis, the extended Kp2 cut was used with a smoothing scale of $1^\circ$ FWHM. The two small dots indicate the position of the ecliptic poles.