Distinguishing the nature of dark matter by mapping cosmic filaments from Lyman-alpha emission

The standard $Λ$CDM cosmological model predicts that cosmic filaments are highly clumpy, whereas warm dark matter – invoked to address small-scale challenges in $Λ$CDM – produces filaments that are noticeably smoother and less structured. In this work, we investigate the potential of Lyman $α$ (Ly$α$) emission to trace cosmic filaments at redshifts $z=2.5$ and $z=4$, and assess their potential for constraining the nature of dark matter. Our analysis shows that Ly$α$ filaments provide a promising observational probe of dark matter: at $z=4$, differences in filament smoothness and surface brightness serve as distinctive signatures between models. Looking ahead, the upcoming generation of 30-meter class telescopes will be critical for enabling these measurements, offering a compelling opportunity to distinguish the nature of dark matter by mapping the structure of cosmic filaments.

💡 Research Summary

This paper investigates whether Lyman‑α (Lyα) emission from the intergalactic medium can be used to map cosmic filaments and thereby discriminate between cold dark matter (CDM) and warm dark matter (WDM) scenarios. The authors perform two high‑resolution zoom‑in hydrodynamical simulations of a Milky Way‑like halo using the same initial random phases but different dark‑matter power spectra: a standard ΛCDM spectrum and a truncated spectrum corresponding to a 1.5 keV thermal relic (WDM). Both runs employ GADGET‑3, include cooling, photo‑heating from a Haardt & Madau UV/X‑ray background, and a simple star‑formation prescription (gas converts to stars when n_H > 0.1 cm⁻³ and overdensity > 2000). Reionization is set at z = 6, and the simulations have gas particle mass 5.16 × 10⁴ h⁻¹ M⊙ and dark‑matter particle mass 2.35 × 10⁵ h⁻¹ M⊙. No explicit feedback processes are modeled.

Because the simulations lack self‑shielding, the authors post‑process each gas cell with the Rahmati et al. (2013) self‑shielding suppression factor, recompute neutral‑hydrogen (n_HI), ionized‑hydrogen (n_HII), and electron (n_e) densities, and then calculate Lyα emissivity from two channels: case‑B recombination (ε_rec = α_B(T) f_rec n_e n_HII E_α) and collisional excitation (ε_coll = q_coll(T) n_e n_HI E_α). Coefficients are taken from established atomic physics references. The total Lyα luminosity per cell is projected onto a 2‑D grid using a Cloud‑In‑Cell (CIC) scheme, attenuated by a redshift‑dependent transmission factor (T_r = 0.92 at z = 2.5, 0.52 at z = 4), and converted to surface‑brightness units (erg s⁻¹ cm⁻² arcsec⁻²).

The resulting maps reveal clear morphological differences. At both redshifts the simulations show a network of filaments feeding a central halo of mass ≈10¹¹ h⁻¹ M⊙. In the WDM case the HI column density is generally higher but distributed more smoothly, especially at z = 4. The CDM filaments are clumpier, with numerous small‑scale overdensities. Correspondingly, Lyα surface‑brightness maps (Fig. 3) show that WDM filaments are brighter (by ≈0.5 dex) and smoother than CDM filaments at z = 4. By z = 2.5 the contrast diminishes as small Lyα clumps appear along the WDM filaments, reducing the discriminating power.

To assess observational feasibility, the authors generate mock Lyα images at z = 4 for three instrument configurations: MUSE on the Very Large Telescope (VLT), a hypothetical MUSE‑like instrument on a 30 m Extremely Large Telescope (ELT), and a next‑generation integral‑field spectrograph (IFS) with larger field‑of‑view (FOV). Gaussian noise consistent with the quoted sensitivities (3.5 × 10⁻²⁰ erg s⁻¹ cm⁻² arcsec⁻² for VLT‑MUSE, 0.8 × 10⁻²⁰ for ELT‑MUSE) is added. Even with noise, WDM filaments remain smoother and brighter, and are marginally detectable with a VLT‑class instrument after ≈150 h exposure. CDM filaments require the higher sensitivity of an ELT‑class instrument to be distinguished. The authors emphasize that the combination of surface‑brightness level and morphological smoothness provides a two‑dimensional diagnostic capable of separating the two dark‑matter models.

The paper discusses several modeling assumptions and their impact. The Lyα emissivity model is deliberately conservative, providing a lower bound on intrinsic brightness; more sophisticated radiative‑transfer treatments could raise absolute values but are unlikely to erase the relative CDM‑WDM differences. The lack of feedback is acknowledged as a limitation; however, the authors argue that feedback would affect CDM and WDM differently because of their distinct small‑scale structure, so the contrast should persist. They also note that the 1.5 keV WDM case is an extreme scenario chosen to maximize observable differences, and that similar filament‑smoothness signatures are expected for fuzzy dark‑matter (FDM) models, which also suppress small‑scale power.

In conclusion, the study demonstrates that high‑redshift (z ≈ 4) Lyα filaments are sensitive probes of the underlying dark‑matter free‑streaming scale. The smoother, brighter filaments predicted in WDM (and FDM) models can be distinguished from the clumpier CDM filaments using deep, wide‑field integral‑field spectroscopy on upcoming 30 m‑class telescopes. Existing deep Lyα surveys (e.g., MUSE Ultra Deep Field, MXDF) may already contain candidate smooth filaments, but larger samples and higher‑sensitivity observations will be required to place robust constraints on the nature of dark matter.