[Paper Review] Imprints of fermionic and bosonic mixed dark matter on the 21-cm signal at cosmic dawn

Sambit K. Giri, Aurel Schneider|arXiv (Cornell University)|Jan 6, 2022

Radio Astronomy Observations and Technology117 references32 citations

TL;DR

This paper investigates how mixed dark matter scenarios—featuring either fermionic (sterile neutrino) or bosonic (ultralight axion) non-cold dark matter—leave distinct imprints on the 21-cm signal during cosmic dawn. Using a halo-model-based simulation of the 21-cm global signal and power spectrum, the authors forecast that the Square Kilometre Array (SKA) will constrain both the mass and fraction of non-cold dark matter, yielding stringent limits: for fermionic DM, mnCDM > 15 keV (thermal mass) at 95% CL when fnCDM ≈ 1, and for bosonic DM, mnCDM > 2 × 10⁻²⁰ eV under the same condition.

ABSTRACT

The 21-cm signal from the epoch of cosmic dawn prior to reionization consists of a promising observable to gain new insights into the dark matter (DM) sector. In this paper, we investigate its potential to constrain mixed (cold + non-cold) dark matter scenarios that are characterised by the non-cold DM fraction ($f_{ m nCDM}$) and particle mass ($m_{ m nCDM}$). As non-cold DM species, we investigate both a fermionic (sterile neutrino) and a bosonic (ultra-light axion) particle. We show how these scenarios affect the global signal and the power spectrum using a halo-model implementation of the 21-cm signal at cosmic dawn. Next to this study, we perform an inference-based forecast study based on realistic mock power spectra from the Square Kilometre Array (SKA) telescope. Assuming inefficient, yet non-zero star-formation in minihaloes (i.e. haloes with mass below $10^8$ M$_{\odot}$), we obtain stringent constraints on both $m_{ m nCDM}$ and $f_{ m nCDM}$ that go well beyond current limits. Regarding the special case of $f_{ m nCDM}\sim 1$, for example, we find a constraint of $m_{ m nCDM}>15$ keV (thermal mass) for fermionic DM and $m_{ m nCDM}>2 imes10^{-20}$ eV for bosonic DM. For the opposite case of dominating cold DM, we find that at most one percent of the total DM abundance can be made of a hot fermionic or bosonic relic. All constraints are provided at the 95 percent confidence level.

Motivation & Objective

• To investigate the imprints of mixed dark matter (cold + non-cold) on the 21-cm signal during cosmic dawn.
• To assess the potential of the Square Kilometre Array (SKA) to constrain fermionic (sterile neutrino) and bosonic (ultralight axion) non-cold dark matter components.
• To provide quantitative forecasts on the mass and fractional abundance of non-cold dark matter using realistic mock power spectra from SKA-Low.
• To explore how suppressed small-scale power from non-cold DM affects the 21-cm global signal and spatial power spectrum.

Proposed method

• Uses a halo-model implementation to compute the 21-cm global signal and power spectrum in mixed dark matter scenarios.
• Parametrizes non-cold DM via the fraction fnCDM and particle mass mnCDM for both fermionic (WCDM) and bosonic (FCDM) models.
• Applies a realistic star formation model in minihaloes (M < 10⁸ M⊙) with non-zero, inefficient star formation to compute source contributions.
• Performs a Bayesian inference-based forecast using mock SKA-Low power spectra to derive constraints at 95% confidence level.
• Employs a modified version of the 21-cm signal code tools21cm to simulate the 21-cm signal across redshifts z ≈ 10–25.
• Uses emcee for Markov Chain Monte Carlo sampling to perform parameter inference on the simulated data.

Experimental results

Research questions

• RQ1 How do fermionic and bosonic non-cold dark matter components alter the 21-cm global signal and power spectrum during cosmic dawn?
• RQ2 What constraints can the Square Kilometre Array (SKA) place on the mass and abundance of non-cold dark matter in mixed dark matter scenarios?
• RQ3 How do the suppression of small-scale power and free-streaming effects from non-cold DM influence the 21-cm signal's amplitude and scale dependence?
• RQ4 What are the implications of non-zero star formation in minihaloes for the detectability and interpretation of the 21-cm signal in mixed DM models?

Key findings

• For a fermionic non-cold dark matter component with fnCDM ≈ 1, the SKA forecasts a 95% confidence level lower limit of mnCDM > 15 keV (thermal mass).
• For a bosonic non-cold dark matter component with fnCDM ≈ 1, the SKA forecasts a 95% confidence level lower limit of mnCDM > 2 × 10⁻²⁰ eV.
• In the limit of dominant cold dark matter (fnCDM ≪ 1), the SKA constrains the non-cold component to at most 1% of the total dark matter abundance.
• The 21-cm power spectrum exhibits a scale-dependent suppression due to non-cold DM, with steeper cutoffs for bosonic (fuzzy) DM than for fermionic (warm) DM.
• The inclusion of inefficient star formation in minihaloes enhances the sensitivity of the 21-cm signal to non-cold DM, improving constraint power.
• The study demonstrates that SKA-Low observations will provide significantly stronger constraints on non-cold dark matter than current observational limits, especially for ultralight axion-like particles.