Probing Decaying Dark Matter Using the Post-EoR HI 21-cm signal

TL;DR

This work investigates the potential of the post-reionization HI 21-cm power spectrum as a probe of decaying dark matter (DDM) in a two-body decay scenario, characterized by the energy transfer fraction $\epsilon$ and decay rate $\Gamma$. By modeling the background expansion and small-scale power suppression, and computing the redshift-space anisotropic 21-cm signal with HI bias and an emulator for the matter power spectrum, the authors forecast constraints on $\epsilon$ and $\Gamma$ using a futuristic SKA1-Mid-like intensity mapping setup. They find that, in an ideal foreground-free case, $\epsilon$ and $\Gamma$ can be constrained to about 3.4% and 6.9% respectively for fiducial values, with weaker constraints when foregrounds or wedge contamination are included; a second benchmark with much smaller $(\epsilon,\Gamma)$ yields correspondingly tighter or looser constraints depending on foreground handling. Marginalizing over the HI bias markedly degrades the constraints, highlighting the importance of accurate bias modeling and control of astrophysical systematics. Overall, post-reionization 21-cm intensity mapping offers a promising avenue to test late-time small-scale suppression from DDM, albeit with significant observational and modeling challenges such as foregrounds, bias, and baryonic feedback to be addressed.

Abstract

We propose the HI 21-cm power spectrum from the post-reionization epoch as a probe of a cosmological model with decaying dark matter particles. The unstable particles are assumed to undergo a 2-body decay into a massless and massive daughter. We assume that a fraction $f$ of the total dark matter budget to be unstable and quantify the decay using the life time $Γ^{-1}$ and the relative mass splitting $ε$ between the parent and the massive daughter. The redshift space anisotropic power spectrum of the post-reionization 21-cm signal brightness temperature, as a tracer of the dark matter clustering, imprints the decaying dark matter model through its effect on background evolution and the suppression of power on small scales. We find that with an idealized futuristic intensity mapping experiment with a SKA1-Mid like radio-array, $ε$ and $Γ$ can be measured at $3.37\%$ and $6.86\%$ around their fiducial values of $ε= 0.012$ and $Γ= 0.008 {\rm Gyr}^{-1}$ respectively. When only the foreground-free window is considered in the $k-$ space, these percentage errors in $ε$ and $Γ$ degrade to $9.91\%$ and $12.03\%$ respectively.The forecasts under identical assumptions for a second, literature-anchored benchmark ($ε=1.14\times10^{-4}$, $Γ=1.04\times10^{-3}\,\mathrm{Gyr}^{-1}$) yields projected uncertainties of $6.16\%$ on $ε$ and $8.29\%$ on $Γ$ while, restricting to the foreground-free window increases these to $18.77\%$ and $14.22\%$, respectively.

Figures (9)

• Figure 1: A schematic representation of the 4-momentum $p^{\mu} =(E, {\mathbf p})$ in a two-body decay with a massive parent particle decaying into a massless and massive daughter
• Figure 2: The Hubble parameter $H(z)$ for different $(\bar{\tau} = \tfrac{\Gamma^{-1}}{1 {\rm Gyr}}, \epsilon)$ shows the expected match with $\Lambda$CDM model at high redshifts and departure at low redshifts. The inset shows the evolution of the parent DDM and daughter particles. Here, we have assumed $f=1$.The observational results at $z=0.38$, $0.51$, and $0.61$ are from galaxy clustering data of BOSS SDSS III sdss3. We have also shown some high redshift data from Lyman-$\alpha$ forest at $z=2.3$boss-lydata2.3 and $z = 2.4$boss-lydata2.4.
• Figure 3: The small-scale suppression of power for DDM and other models at three redshifts $z =1.75, 2.00, 2.35$(top to bottom). The suppression of power for DDM may be indistinguishable from the similar effect arising from different models over a range of scales.
• Figure 4: (a),(b) and (c) shows the dimensionless anisotropic 21-cm power spectrum $k^3 P_{\rm HI} /2\pi^2$ for a DDM cosmology at the fiducial redshifts $z=1.75, 2.0$, and $2.35$ respectively at the fiducial values of parameters $(\Omega_m, \epsilon, \Gamma, H_0, f) = (0.3094, ~0.012, ~0.008 {\rm Gyr^{-1}}, ~67.73 {\rm Km/s/Mpc}, ~1.0)$ respectively. The departure from spherical symmetry is seen on small scales.
• Figure 5: (a) Shows the actual antenna distribution. (b) Shows the sampling $u-v$ coverage for $5$hrs synthesis imaging.