A new large CMB non-Gaussian anomaly and its alignment with cosmic structure

TL;DR

This paper reports the detection of a significant non-Gaussian anomaly in the CMB aligned with a large quasar group, with statistical analysis suggesting it is unlikely due to primordial Gaussian fluctuations.

Contribution

It introduces a new large-scale non-Gaussian anomaly in the CMB, aligned with a large quasar group, and assesses its statistical significance using Monte Carlo simulations.

Findings

The anomaly has a temperature difference of approximately 43 μK.

The anomaly's significance exceeds 2.3σ, indicating it is unlikely from Gaussian fluctuations.

The anomaly's alignment suggests a possible Rees-Sciama effect origin.

Abstract

We provide evidence of the detection of a new non-Gaussian anomaly in the cosmic microwave background (CMB) radiation which has larger statistical significance than the Cold Spot (CS) anomaly and comparable size. This temperature anomaly is aligned with a huge large quasar group (HLQG), and for this reason we call it HLQG anomaly. There are different physical phenomena by which the HLQG could have produced the observed temperature anomaly, such as for example the Sunyaev Zeldovich (SZ), the integrated Sachs Wolfe (ISW) or the Rees Sciama (RS) effect. The goal of this paper it is not to explain the observed alignment in terms of these effects, but to show the shape and position of the HLQG anomaly, and estimate its statistical significance, i.e. the probability that it could be just the result of primordial Gaussian fluctuations. We analyze the CMB Planck satellite temperature map of the region of sky corresponding to the angular position of the HLQG, and compute an inner and an outer temperature by averaging the CMB map over respectively the region subtended by the HLQG on the sky, and over a region surrounding it. It turns out that the inner region is warmer than the outer one, with a measured temperature difference of $\Delta T^{obs} \approx 43\mu K$. The temperature excess is then compared with the results of Montecarlo simulations of random Gaussian realizations of the CMB map, indicating with at least a $2.3\sigma$ confidence level, that the measured $\Delta T^{obs}$ cannot be attributed to primordial Gaussian fluctuations. Its alignment with HLQG and suggests that this could be due to the Rees-Sciama effect. Our results are stable under the choice of different simulations methods and different definitions of the inner and outer regions.