[Paper Review] The two-mode puzzle: Confronting self-interacting neutrinos with the full shape of the galaxy power spectrum

David Camarena, Francis-Yan Cyr-Racine+1 more|arXiv (Cornell University)|Sep 7, 2023

Cosmology and Gravitation Theories4 citations

TL;DR

This paper investigates self-interacting neutrinos that delay free streaming until near matter-radiation equality, using the full shape of the galaxy power spectrum to constrain cosmological models. It finds a modest preference for this scenario over standard ΛCDM, driven by mildly non-linear galaxy clustering data, suggesting new neutrino interactions may be consistent with large-scale structure observations independent of CMB data.

ABSTRACT

A cosmological scenario in which the onset of neutrino free streaming in the early Universe is delayed until close to the epoch of matter-radiation equality has been shown to provide a good fit to some cosmic microwave background (CMB) data, while being somewhat disfavored by Planck CMB polarization data. To clarify this situation, we investigate in this paper CMB-independent constraints on this scenario from the Full Shape of the galaxy power spectrum. Although this scenario predicts significant changes to the linear matter power spectrum, we find that it can provide a good fit to the galaxy power spectrum data. Interestingly, we show that the data display a modest preference for a delayed onset of neutrino free streaming over the standard model of cosmology, which is driven by the galaxy power spectrum data on mildly non-linear scales. This conclusion is supported by both profile likelihood and Bayesian exploration analyses, showing robustness of the results. Compared to the standard cosmological paradigm, this scenario predicts a significant suppression of structure on subgalactic scales. While our analysis relies on the simplest cosmological representation of neutrino self-interactions, we argue that this persistent - and somehow consistent - picture in which neutrino free streaming is delayed motivates the exploration of particle models capable of reconciling all CMB, large-scale structure, and laboratory data.

Motivation & Objective

• To test whether self-interacting neutrinos delaying free streaming until near matter-radiation equality can be constrained using large-scale structure data.
• To assess whether the galaxy power spectrum provides independent evidence for such a scenario, separate from CMB data.
• To explore the robustness of the delayed free-streaming scenario using both profile likelihood and Bayesian methods.
• To evaluate whether this model can reconcile with Big Bang Nucleosynthesis (BBN) and large-scale structure observations.
• To motivate the development of more complex neutrino interaction models that may reconcile CMB, LSS, and laboratory constraints.

Proposed method

• Uses the full shape of the galaxy power spectrum (FS) as a cosmological observable to constrain neutrino self-interaction models.
• Employs an effective four-fermion interaction model to represent neutrino self-interactions, parameterized by a coupling strength $ G_{\mathrm{eff}} $.
• Applies a phenomenological model where neutrino free streaming onset is delayed, altering the linear matter power spectrum.
• Combines FS data with BBN constraints to improve cosmological parameter estimation.
• Performs both profile likelihood and Bayesian model comparison to assess evidence for the self-interacting neutrino (SI ν) scenario.
• Fixes the sum of neutrino masses at $ \Sigma m_{\nu} = 0.06 $ eV but validates results with a free-mass analysis.

Figure 1: Modes crossing the horizon during the neutrino tight-coupling era, $k^{\rm tc}_{\rm h}$ , the onset of free streaming, $k^{\rm fs}_{\rm h}$ , and well after the self-decoupling, $k_{\rm h}$ , as functions of the self-interaction coupling $G_{\mathrm{eff}}$ . The gray gradient represents th

Experimental results

Research questions

• RQ1 Does the full shape of the galaxy power spectrum provide evidence for delayed neutrino free streaming due to self-interactions?
• RQ2 How does the SI ν scenario compare to the standard ΛCDM model in fitting galaxy clustering data?
• RQ3 Is the preference for delayed free streaming robust across different statistical methods, such as profile likelihood and Bayesian inference?
• RQ4 Can the SI ν model be reconciled with both BBN and large-scale structure data without conflicting with CMB polarization constraints?
• RQ5 What are the implications of this scenario for the primordial power spectrum amplitude $ A_{\rm s} $ and tilt $ n_{\rm s} $?

Key findings

• The galaxy power spectrum data show a modest preference for the self-interacting neutrino (SI ν) scenario over the standard ΛCDM model, driven by mildly non-linear scales.
• The preference is robust across both profile likelihood and Bayesian model comparison, indicating consistency and reliability of the result.
• The SI ν model predicts significant suppression of structure on subgalactic scales compared to ΛCDM.
• The model yields a slightly lower amplitude $ A_{\rm s} $ and tilt $ n_{\rm s} $ in the primordial power spectrum compared to ΛCDM, though the difference is small.
• The assumption of fixed $ \Sigma m_{\nu} = 0.06 $ eV does not bias the constraints on $ G_{\mathrm{eff}} $, $ N_{\mathrm{eff}} $, $ H_0 $, or $ \sigma_8 $, validating the analysis choice.
• The results suggest that the SI ν signal may not be captured by standard phase-shift templates based on constant free-streaming fractions, indicating a need for time-varying models.

Figure 2: Linear matter power spectrum (top panel) for the $\Lambda\mathrm{CDM}+N_{\mathrm{eff}}$ cosmology and the self-interacting neutrino model with different values of $G_{\mathrm{eff}}$ . The ratios of the latter with the $\Lambda\mathrm{CDM}+N_{\mathrm{eff}}$ model are shown in the bottom pan