Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

The cosmic web's Lyman-$α$ glow at $z \approx 2.5$; varying hydrodynamic models, dust, and wide-field, narrow-band imaging detection

Oleksii Sokoliuk, John K. Webb, Kenneth M. Lanzetta, Michael M. Shara, Stefan Gromoll, James S. Bolton, Robert F. Carswell, Gaspar Galaz, Cédric Ledoux, Gaspare Lo Curto, Alain Smette, David Valls-Gabaud, Anja von der Linden, Frederick M. Walter, Joris Witstok

TL;DR

The paper addresses the challenge of directly imaging the cosmic web in Lyα at $z \approx 2.5$ by post-processing five state-of-the-art hydrodynamic simulations to generate synthetic Lyα surface-brightness maps. It employs semi-analytic Lyα emissivities from recombination and collisional excitation, tracks hydrogen species with self-shielding, and applies a dust attenuation model before projecting into narrowband slices to mimic Condor's observations. The results show general consistency with the Condor NM UV excess and demonstrate that diffuse Lyα emission from the cosmic web should be detectable with current and future wide-field instruments, while also highlighting sensitivity to dust treatment and subgrid physics. The study also characterizes the HI column density distribution, phase-diagram relations, and the impact of observational noise, providing a framework for future cartographic mapping of the cosmic web and informing follow-up observations.

Abstract

The diffuse glow of the cosmic web in Lyman-$α$ emission has long been predicted, yet remained elusive to direct wide field detection. We present theoretical calculations that, when compared with recent observations made using the Condor Array Telescope in New Mexico reported in Lanzetta et al. 2024, point to its discovery at $z \approx 2.5$. Synthetic Lyman-$α$ surface brightness maps are constructed from five state-of-the-art hydrodynamic simulations (Illustris-TNG, SIMBA, EAGLE, CROCODILE, and Sherwood), incorporating dust attenuation, star formation, collisional excitation, and recombination physics. Our cosmic web Lyman-$α$ surface brightness predictions are consistent with the UV excess detected at high significance in the recent deep, wide field, narrow-band imaging Condor data. The calculations presented here thus demonstrate that diffuse Lyman-$α$ emission is observable with current (and next-generation) wide field low surface brightness facilities, opening the path to direct cartographic mapping of the cosmic web. These findings mark a turning point: for the first time, cosmology moves beyond inference from absorption and high-density peaks, into panoramic imaging of the faint intergalactic scaffolding that underpins structure formation in the Universe.

The cosmic web's Lyman-$α$ glow at $z \approx 2.5$; varying hydrodynamic models, dust, and wide-field, narrow-band imaging detection

TL;DR

The paper addresses the challenge of directly imaging the cosmic web in Lyα at by post-processing five state-of-the-art hydrodynamic simulations to generate synthetic Lyα surface-brightness maps. It employs semi-analytic Lyα emissivities from recombination and collisional excitation, tracks hydrogen species with self-shielding, and applies a dust attenuation model before projecting into narrowband slices to mimic Condor's observations. The results show general consistency with the Condor NM UV excess and demonstrate that diffuse Lyα emission from the cosmic web should be detectable with current and future wide-field instruments, while also highlighting sensitivity to dust treatment and subgrid physics. The study also characterizes the HI column density distribution, phase-diagram relations, and the impact of observational noise, providing a framework for future cartographic mapping of the cosmic web and informing follow-up observations.

Abstract

The diffuse glow of the cosmic web in Lyman- emission has long been predicted, yet remained elusive to direct wide field detection. We present theoretical calculations that, when compared with recent observations made using the Condor Array Telescope in New Mexico reported in Lanzetta et al. 2024, point to its discovery at . Synthetic Lyman- surface brightness maps are constructed from five state-of-the-art hydrodynamic simulations (Illustris-TNG, SIMBA, EAGLE, CROCODILE, and Sherwood), incorporating dust attenuation, star formation, collisional excitation, and recombination physics. Our cosmic web Lyman- surface brightness predictions are consistent with the UV excess detected at high significance in the recent deep, wide field, narrow-band imaging Condor data. The calculations presented here thus demonstrate that diffuse Lyman- emission is observable with current (and next-generation) wide field low surface brightness facilities, opening the path to direct cartographic mapping of the cosmic web. These findings mark a turning point: for the first time, cosmology moves beyond inference from absorption and high-density peaks, into panoramic imaging of the faint intergalactic scaffolding that underpins structure formation in the Universe.

Paper Structure

This paper contains 22 sections, 44 equations, 7 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Five hydrodynamic simulations
  4. Intrinsic Lyman-$\alpha$ luminosities
  5. Decoupling neutral, molecular and ionised hydrogen
  6. H$_2$ formation
  7. Deriving the electron number density
  8. The H I column density distribution function
  9. Escape of Lyman-$\alpha$ photons from an optically thick medium; the impact of dust.
  10. Removing high density particles - hydrogen number density upper limit
  11. Final surface brightness maps
  12. Results
  13. The H I column density distribution function
  14. Physical insights from the $\log T$-$\log n_{\rm H}$ and $\log\mathcal{S}$-$\log N_{\rm H\textsc{\scriptsize I}{}}$ phase diagrams
  15. $\log T$-$\log n_{\rm H}$ relation
  16. ...and 7 more sections

Figures (7)

  • Figure 1: Comoving H I Column Density Distribution Function and its probability distribution for each simulation.
  • Figure 2: Left column (a): Temperature vs. hydrogen number density relation. Note the abrupt $n_{\rm H}$ cutoff for some simulations, a consequence of the switch to a stochastic Kennicutt–Schmidt star formation law. Right column (b): H I column density vs. Lyman-$\alpha$ surface brightness. Power-law fits for this relation are shown as dotted lines, offset from the image data for visualisation. See Sections \ref{['sec:T-lognH']} to \ref{['sec:insights']}. The colour scale in the left column ranges (approximately) from $0-10^6$ for the left column and $0-10^4$ for the right.
  • Figure 3: Lyman-$\alpha$ surface brightness map for each simulation. Middle and right panels show 5$\times$ and 10$\times$ zoom-ins on arbitrarily selected regions. See Section \ref{['sec:SB']}. Each simulation has its own colour map, i.e. the numerical display ranges are not the same for each simulation. Instead, each colour map is set by the minimum and maximum count in each image.
  • Figure 4: Lyman-$\alpha$ narrow band surface brightness in each simulation. Each panel shows the surface brightness probability distribution. These panels represent simulations of, and may thus be be compared with, figure 2 in Lanzetta2024. The left column corresponds to the dust models described in Section \ref{['sec:dust']}. The standard deviation $\sigma$ of the Gaussian noise added is shown in each panel. The top row ($\sigma = 0$) has no noise. The middle and right panels correspond to the particle threshold method described in Section \ref{['sec:hireject']}. The left column (from which Sherwood is excluded, since metallicity information is not available) illustrates that the combined intergalactic and circumgalatic narrow band Lyman-$\alpha$ is detectable by all simulations for $\mathcal{S} \sim 10^{-18}$ erg s$^{-1}$ cm$^{-2}$ arcsec$^{-2}$. The middle column illustrates that detecting the low density component of the cosmic web narrow band Lyman-$\alpha$ requires a background noise level below $\sigma \sim 5 \times 10^{-20}$ erg s$^{-1}$ cm$^{-2}$ arcsec$^{-2}$. Fig. \ref{['fig:Asquared']} shows the Anderson-Darling statistical text applied to these data, to evaluate detection thresholds.
  • Figure 5: Surface brightness map from the EAGLE simulation made using the dust calculation (Section \ref{['sec:dust']}), for various Gaussian noise models (Section \ref{['sec:addnoise']}), emulating (in part, Section \ref{['sec:diffs']}) real observational data. The standard deviation $\sigma$ of the Gaussian noise added is shown in each panel. The panels in this figure correspond to the left hand panels in Fig. \ref{['fig:SB_PDFs']}.
  • ...and 2 more figures