$Λ$CDM or self-interacting neutrinos? - how CMB data can tell the two models apart

TL;DR

The paper investigates whether current CMB data distinguish ΛCDM from a self-interacting neutrino model characterized by a coupling strength $G_eff$ that delays neutrino free-streaming. Using a 9-parameter cosmology and Bayesian model comparison with Planck and BAO data, it finds a robust bimodality in the posterior for $G_eff$ driven by high-multipole measurements, yielding a ΛCDM-like weak island and a strongly interacting island with distinct cosmological parameters. The authors map the spectra along the line between the two islands, showing a pronounced dip in fit quality at intermediate $G_eff$ and illustrating that small-scale TT data and TE polarization carry the most discriminating power. They forecast that the Simons Observatory will dramatically sharpen constraints, likely distinguishing the two modes and potentially ruling out the non-favored one, thereby clarifying whether new neutrino interactions are needed. Overall, the work highlights how high-multipole CMB information constrains beyond-ΛCDM neutrino physics and motivates future data-driven tests of neutrino interactions in the early Universe.

Abstract

Of the many proposed extensions to the $Λ$CDM paradigm, a model in which neutrinos self-interact until close to the epoch of matter-radiation equality has been shown to provide a good fit to current cosmic microwave background (CMB) data, while at the same time alleviating tensions with late-time measurements of the expansion rate and matter fluctuation amplitude. Interestingly, CMB fits to this model either pick out a specific large value of the neutrino interaction strength, or are consistent with the extremely weak neutrino interaction found in $Λ$CDM, resulting in a bimodal posterior distribution for the neutrino self-interaction cross section. In this paper, we explore why current cosmological data select this particular large neutrino self-interaction strength, and by consequence, disfavor intermediate values of the self-interaction cross section. We show how it is the $\ell \gtrsim 1000$ CMB temperature anisotropies, most recently measured by the Planck satellite, that produce this bimodality. We also establish that smaller scale temperature data, and improved polarization data measuring the temperature-polarization cross-correlation, will best constrain the neutrino self-interaction strength. We forecast that the upcoming Simons Observatory should be capable of distinguishing between the models.

Figures (7)

• Figure 1: Probability distributions for parameters from a nine-parameter model ($\mathrm{\Lambda}$CDM plus neutrino self-interaction strength $\textit{G}_{\rm eff}$, effective neutrino number, and neutrino mass), using the WMAP and Planck CMB data combined with BAO and Planck lensing data. The parameters derived using Planck are consistent with previous results kreisch and show the clear bimodality in the neutrino self-interaction strength. The 'strong' and 'weak' distributions show the marginalized posteriors when considering each of the bimodal islands separately. For the unseparated distribution, the strong mode has a lower marginalized posterior relative to the weak mode. The distribution using just WMAP data is not bimodal.
• Figure 2: Illustration of the line connecting the best-fitting models in each mode that we use to compute spectra and likelihoods. The orange stars are the locations of the 4 points in parameter space sampled for Figure \ref{['fig:spectra']} and Figure \ref{['fig:difflmax']}
• Figure 3: $-\chi^2$ values for the Planck$\ell>30$ data along the path shown in Figure \ref{['fig:path']} to show the clear bimodality and the likelihoods between the two modes. The two modes have the same likelihood; the difference in posterior distribution for $\textit{G}_{\rm eff}$ is then due to the volume of well-fitting models in our chosen parameterization.
• Figure 4: CMB power spectra (for TT, left, and TE, right) at the points shown in Fig. \ref{['fig:path']}, shown as residuals compared to the best-fitting $\mathrm{\Lambda}$CDM model from planckparams:2018. The Planck error bars are shown, and forecasted SO errors are indicated on the left-hand plot. The 'weak' and 'strong' modes both fit $\ell<2000$ data but diverge at smaller scales and differ in TE. The intermediate values for $\textit{G}_{\rm eff}$ have a lower TT power at $\ell>1000$, so are excluded by Planck data.
• Figure 5: The goodness of fit ($-\chi^2/{\rm dof}$ for $\ell>30$ from the 'plik-lite' Planck likelihood) as a function of $\ell_{\mathrm{max}}$ for the models shown in Fig. \ref{['fig:spectra']}. This shows how models between the two well-fitting modes are poor fits to the Planck data at small scales.