Testing the running of non-Gaussianity through the CMB<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mml:mi>μ</mml:mi></mml:math>-distortion and the halo bias

Matteo Biagetti, H. Perrier, Antonio Riotto, Vincent DesjacquesPhysical 2013 · DOI · arXiv

Abstract

The primordial non-Gaussianity parameters $f_{\text{NL}}$ and $\tau_{\text{NL}}$ may be scale dependent. We investigate the capability of future measurements of the CMB $\mu$ distortion, which is very sensitive to small scales, and of the large-scale halo bias to test the running of local non-Gaussianity. We show that, for an experiment such as PIXIE, a measurement of the $\mu$-temperature correlation can pin down the spectral indices $n_{f_{\text{NL}}}$ and $n_{\tau_{\text{NL}}}$ to values of the order of 0.3 if $f_{\text{NL}}=20$ and $\tau_{\text{NL}}=5000$. A similar value can be achieved with an all-sky survey extending to redshift $z \sim 1$. In the particular case in which the two spectral indices are equal, as predicted in models where the cosmological perturbations are generated by a single field other than the inflaton, the $1\sigma$ error on the scale dependence of the nonlinearity parameters goes down to 0.2.

Introduction

Standard "single-field" inflation models predict that initial density fluctuations were almost perfectly Gaussian. However, more complex multi-field models suggest a degree of primordial non-Gaussianity (NG). While NG parameters like $f_{\text{NL}}$ and $\tau_{\text{NL}}$ are often treated as constants, theoretical models suggest they might "run," meaning their strength changes across different physical scales. Detecting this scale-dependence requires a "long lever arm"—data spanning from the largest observable scales down to the very small. This paper explores how next-generation cosmological probes can bridge this gap.

Methodology

The study employs a Fisher Information Matrix analysis to forecast the sensitivity of two primary observables:

  1. CMB $\mu$-distortions: Tiny deviations in the CMB's blackbody spectrum caused by the dissipation of acoustic waves at small scales ($50 \lesssim k \text{ Mpc} \lesssim 10^4$).
  2. Halo Bias: The clustering of dark matter halos relative to underlying matter, which exhibits a unique scale-dependent signature in the presence of primordial NG.

The authors assume specifications for future experiments such as the PIXIE (Primal Inflation Explorer) satellite mission and large-scale galaxy surveys like EUCLID or BigBOSS reaching redshift $z \sim 1$.

Probing the Running of Non-Gaussianity

$\mu$-distortions are uniquely sensitive to scales far smaller than those accessible via standard CMB temperature maps. The correlation between these distortions and temperature anisotropies ($\mu-T$) effectively probes the bispectrum ($f_{\text{NL}}$), while the $\mu-\mu$ self-correlation probes the trispectrum ($\tau_{\text{NL}}$).

1-sigma errors on the spectral index $n_{f_{NL}}$ as a function of $f_{NL}^*$, derived from CMB $\mu$-distortion correlations ($C^{\mu\mu}$ and $C^{\mu T}$) for a PIXIE-like experiment.

Forecasted Sensitivity and Results

The analysis reveals that for a local model with $f_{\text{NL}} = 20$ and $\tau_{\text{NL}} = 5000$, PIXIE could measure spectral indices with an accuracy of approximately 0.3. Similarly, all-sky surveys like EUCLID can achieve a precision of $\sim 0.3$ using halo bias measurements.

1-sigma errors on the spectral indices $n_{f_{NL}}$ (left) and $n_{\tau_{NL}}$ (right) as a function of the non-linearity parameters, derived from halo bias measurements in a galaxy survey.

In models where a single field (other than the inflaton) generates perturbations, the symmetry $n_{f_{\text{NL}}} = n_{\tau_{\text{NL}}}$ can be exploited. This constraint significantly tightens the results, reducing the $1\sigma$ error on the scale-dependence to 0.2. For scenarios with higher non-Gaussianity ($f_{\text{NL}} = 50$), the error via halo bias could drop as low as 0.15.

Conclusion

Testing the running of non-Gaussianity offers a powerful new way to distinguish between inflationary theories. By combining the small-scale sensitivity of CMB $\mu$-distortions with the large-scale reach of halo bias, future experiments like PIXIE and EUCLID will provide a comprehensive test of the early universe's dynamics across a vast range of scales.

Summary updated

Abstract

The primordial non-Gaussianity parameters ${f}_{\mathrm{NL}}$ and ${\ensuremath{\tau}}_{\mathrm{NL}}$ may be scale dependent. We investigate the capability of future measurements of the CMB $\ensuremath{\mu}$ distortion, which is very sensitive to small scales, and of the large-scale halo bias to test the running of local non-Gaussianity. We show that, for an experiment such as PIXIE, a measurement of the $\ensuremath{\mu}$-temperature correlation can pin down the spectral indices ${n}_{{f}_{\mathrm{NL}}}$ and ${n}_{{\ensuremath{\tau}}_{\mathrm{NL}}}$ to values of the order of 0.3 if ${f}_{\mathrm{NL}}=20$ and ${\ensuremath{\tau}}_{\mathrm{NL}}=5000$. A similar value can be achieved with an all-sky survey extending to redshift $z\ensuremath{\sim}1$. In the particular case in which the two spectral indices are equal, as predicted in models where the cosmological perturbations are generated by a single field other than the inflaton, the $1\ensuremath{\sigma}$ error on the scale dependence of the nonlinearity parameters goes down to 0.2.
BibTeX
@inproceedings{biagetti2013testing,
  author = {Matteo Biagetti and H. Perrier and Antonio Riotto and Vincent Desjacques},
  title = {{Testing the running of non-Gaussianity through the CMB<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mml:mi>μ</mml:mi></mml:math>-distortion and the halo bias}},
  booktitle = {Physical},
  year = {2013},
  doi = {10.1103/physrevd.87.063521},
}
Figure