Hemispherical Asymmetry from Parity-Violating Excited Initial States
Authors: Amjad Ashoorioon and Tomi Koivisto Preprint number: UUITP-14/15 We investigate if the hemispherical asymmetry in the CMB is produced from parity-violating excited initial condition. We show that in the limit where the deviations from the Bunch-Davies vacuum is large and the scale of new physics is maximally separated from the inflationary Hubble parameter, the primordial power spectrum is modulated only by dipole and quadrupole terms. Requiring the dipole contribution in the power spectrum accounts for the observed power asymmetry, $A=0.07\pm0.022$, we show that the amount of quadrupole terms is roughly equal to $A^2$, which is still consistent with the bounds from the CMB...

Bildtext
We investigate if the hemispherical asymmetry in the CMB is produced from parity-violating excited initial condition. We show that in the limit where the deviations from the Bunch-Davies vacuum is large and the scale of new physics is maximally separated from the inflationary Hubble parameter, the primordial power spectrum is modulated only by dipole and quadrupole terms. Requiring the dipole contribution in the power spectrum accounts for the observed power asymmetry, $A=0.07\pm0.022$, we show that the amount of quadrupole terms is roughly equal to $A^2$, which is still consistent with the bounds from the CMB. The {\it mean} local bispectrum which gets enhanced for the excited initial states is within the $1\sigma$ bound of Planck 2015 results, $f_{\rm NL}\simeq 4.17$, but reachable by future CMB experiments. The amplitude of the local non-gaussianity modulates around this mean value, approximately depending on the angle that the short wavelength mode makes with the preferred direction. The amount of variation maximizes for the configurations that are coplanar with the preferred direction. For counterclockwise oriented configurations, maximum and minimum values for the non-gaussianity will occur for the ones in which the short wavelength mode is, respectively, antiparallel and parallel with the preferred direction. The difference of non-gaussianity between these two configurations is as large as $\simeq 1.2$ which can be used to distinguish this scenario from other scenarios that try to explain the observed hemispherical asymmetry. Such modulation in non-gaussianity is minimized for the configurations that are in the equator plane orthogonal to the preferred direction.