Sensitivity toward dark matter annihilation imprints on 21-cm signal with SKA-Low: A convolutional neural network approach

TL;DR

The paper addresses whether spatial inhomogeneities from dark matter annihilation can leave detectable imprints on the 21-cm signal during the pre-reionization era. Using DM21cm/21cmFAST simulations and realistic SKA-Low noise, it trains convolutional neural networks to classify maps as produced by inhomogeneous versus homogeneous energy deposition, focusing on $e^+e^-$ and $\gamma\gamma$ channels across $m_{\rm DM}$ and $\langle \sigma v\rangle$. It finds that CNNs robustly distinguish inhomogeneous from homogeneous heating for the $e^+e^-$ channel within a broad range of parameters (e.g., $m_{\rm DM}=1$ MeV with $\langle \sigma v\rangle \gtrsim 5\times10^{-30}$ cm$^3$/s and reasonable noise), while the $\gamma\gamma$ channel remains challenging due to photon mean free paths that homogenize energy deposition. The results highlight the potential of spatially-resolved 21-cm observations, aided by ML techniques, to probe exotic DM energy injection scenarios with SKA-Low in the Cosmic Dawn epoch.

Abstract

This study investigates the sensitivity of the radio interferometers to identify imprints of spatially inhomogeneous dark matter annihilation signatures in the 21-cm signal during the pre-reionization era. We focus on the upcoming low-mode survey of the Square Kilometre Array (SKA-Low) telescope. Using CNNs, we analyze simulated 3D 21-cm differential brightness temperature maps generated via the DM21cm code, which is based on 21cmFAST and DarkHistory, to distinguish between spatially homogeneous and inhomogeneous energy injection/deposition scenarios arising from dark matter annihilation. The inhomogeneous case accounts for local dark matter density contrasts and gas properties, such as thermal and ionization states, while the homogeneous model assumes uniform energy deposition. Our study focuses on two primary annihilation channels to electron-positron pairs ($e^+e^-$) and photons ($γγ$), exploring dark matter masses from 1 MeV to 100 MeV and a range of annihilation cross-sections. For $γγ$ channel, the distinction across dark matter models is less pronounced due to the larger mean free path of the emitted photons, resulting in a more uniform energy deposition. For $e^+e^-$ channel, the results indicate that the CNNs can effectively differentiate between the inhomogeneous and homogeneous cases. Despite observational challenges, the results demonstrate that these effects remain detectable even after incorporating noise from next-generation radio interferometers, such as the SKA. We find that the inhomogeneous dark matter annihilation models can leave measurable imprints on the 21-cm signal maps distinguishable from the homogeneous scenarios for the dark matter masses $m_{\rm DM}=1$ MeV and the annihilation cross-sections of $\geq 5 \times 10^{-30}~{\rm cm^3/sec}$ ($\geq 5 \times 10^{-29}~{\rm cm^3/sec}$ for $m_{\rm DM}=100$ MeV) for moderate SKA-Low noise.

Figures (7)

• Figure 1: 21-cm differential brightness temperature lightcone from redshift 45 to 10 for dark matter annihilation and no-annihilation scenario. The lower panel represents the conventional $\Lambda$CDM model without dark matter annihilation. The middle and the top panel represent the annihilation scenarios with the dark matter particle mass $m_{{\rm DM}} = 1$ MeV, and cross section $\langle \sigma v\rangle=1\times10^{-29}~{{\rm cm}^3/{\rm s}}$ for the electron/positron channel. The upper panel illustrates inhomogeneous energy injection and deposition, while the middle panel depicts the scenario with homogeneous energy injection and deposition.
• Figure 2: 21-cm differential brightness temperature snaps at redshifts 12 for dark matter particle mass $m_{\rm DM}=1$ MeV, and cross-section $\langle \sigma v\rangle=1\times10^{-29}~{\rm cm}^3/{\rm s}$ for electron/positron channel. Left: Inhomogeneous energy injection/deposition into the medium. Middle: homogeneous energy injection/deposition into medium. Right: Difference between 21-cm differential brightness temperature snaps of inhomogeneous and homogeneous energy injection/deposition cases.
• Figure 3: Mean 21-cm power spectrum from 1000 realization having different seeds at a redshift of 12 as a function of inverse length-scale (wavenumber-- $k$) for inhomogeneous (navy-blue solid line) and homogeneous (red dashed line) energy injection/deposition into medium for dark matter particle mass $m_{\rm DM}=1$ MeV, and cross-section $\langle \sigma v\rangle=1\times10^{-29}~{\rm cm}^3/{\rm s}$ annihilating to electron/positron channel. In Fig. \ref{['fig:pk_samplevariance']}, the error bars represent the covariance from 1000 realizations--- sample variance. In Fig. \ref{['fig:pk_totalerror']}, the error-bars represent the total error including sample variance and the thermal noise associated with Square Kilometre Array Low-frequency (SKA-Low) for $f_{\rm Noise}=0.1$.
• Figure 4: Histogram for 21-cm signal at a redshift of 12 for inhomogeneous (blue solid line) and homogeneous (red dashed line) energy injection/deposition into medium for dark matter particle mass $m_{\rm DM}=1~{\rm MeV}$, and cross-section $\langle \sigma v\rangle=1\times10^{-29}~{\rm cm}^3/{\rm s}$ annihilating to electron/positron channel. The histogram has been obtained after subtracting the mean from each cell. The $x$-axis represents the 21-cm signal, and the $y$-axis represents the repetition of 21-cm signal values in a particular bin.
• Figure 5: Signal-to-noise ratio variation with redshift ($z$) and inverse length-scale ($k$). The signal consists of the absolute difference of 21-cm power spectrum between inhomogeneous and homogeneous energy injection/deposition scenarios for dark matter particle mass $m_{\rm DM}=1$ MeV, and cross-section $\langle \sigma v\rangle=1\times10^{-29}~{\rm cm}^3/{\rm s}$ annihilating to electron/positron channel. The 21-cm power spectrum is calculated by averaging the power spectrum for 1000 realizations for each $k$ bin. The noise is comprised of the covariance of the difference between the 21-cm power spectra of the inhomogeneous and homogeneous energy injection/deposition scenarios, also taking 1000 realizations, along with the thermal noise associated with the Square Kilometre Array Low-frequency (SKA-Low), assuming a noise frequency of $f_{\rm Noise} = 0.1$.