MUSEQuBES: Investigating the Physical and Chemical Properties of Circumgalactic Gas Around Ly$α$ Emitters at $z \approx 3.3$

TL;DR

This work probes the physical state of circumgalactic gas around Ly$\alpha$ emitters at $z\approx3.3$ by combining high-resolution quasar spectroscopy with MUSE-detected LAEs. Using Voigt-profile measurements and a Bayesian Cloudy photoionization framework, it derives $n_{ m H}$ and $[X/H]$ for 75 absorbers, finding a median $\log(n_{ m H}/\mathrm{cm^{-3}}) \approx -2.7$ and a median metallicity $[X/H] \approx -1.6$, with a significant fraction ($\sim15\%$, rising to $\sim40\%$ including upper limits) of gas having IGM-like metallicities ($[X/H] \lesssim -2.8$). The high-$N({\rm HI})$ subset exhibits a three-component metallicity distribution, ranging from pristine inflows to possible solar-metallicity outflows, suggesting a chemically inhomogeneous, multiphase CGM around LAEs. No strong correlations are found between absorber metallicity or density and host galaxy properties such as impact parameter, SFR, or Ly$\alpha$ luminosity, indicating a complex baryon cycle with localized enrichment and mixing. The results underscore a CGM around reionization-era galaxies shaped by inflows, outflows, and mixing, and motivate future cloud-by-cloud modeling to recover component-level physics like sizes and temperatures.

Abstract

We present a detailed study of the physical properties of absorbers associated with $z\approx3.3$ \lya\ emitters (LAEs) from the MUSEQuBES survey. Using \HI\ and ionic column densities derived from Voigt profile fitting, we determine the density and metallicity of 75 absorbers associated with 59 LAEs through a custom-built Bayesian framework. The overall sample shows a median density $\log\,(n_{\rm H}/{\rm cm^{-3}})=-2.7\pm0.7$ and metallicity \met$=-1.6^{+1.2}_{-0.9}$ with $\approx15\%$ ($\approx 40\%$ including upper limits) of systems showing metallicities consistent with the IGM at these redshifts (\met$\lesssim -2.8$). Intriguingly, absorbers with $\log\, N(\rm HI)/{\rm cm^{-2}} \gtrsim 16.5$, corresponding to an overdensity of $\sim 100$ at $z=3$, exhibit a bimodal metallicity distribution with peaks at \met$=-3.8\pm 0.2$ and $-1.8\pm 0.6$. The latter are observed at large impact parameters ($\gtrsim150$~pkpc) and often exhibit low-ionization species (e.g., \SiII, \AlII). We interpret the former as pristine inflowing gas from the cosmic web, while the latter likely traces metal-enriched CGM associated with outflows from faint galaxies not detected in Ly$α$ emission. We find no significant correlations between absorber metallicity or density and host galaxy properties, including redshift, impact parameter, SFR, Ly$α$ luminosity, or environment. Absorbers separated by $\lesssim500$~\kms\ show $\sim$1~dex metallicity and $\sim$0.5~dex density variations, indicating physical and chemical inhomogeneity of the medium around these LAEs.

Figures (17)

• Figure 1: Distribution of $n_{\rm H}$ for the absorbers: maroon denotes values estimated using a flat prior in the Bayesian sampling, while gray corresponds to those with a Gaussian prior of $\log\,($$n_{\rm H}$$/{\rm cm^{-3}})= -3.5\pm 1.0$. The median $n_{\rm H}$ of the systems modeled by a flat (Gaussian) prior is indicated by the vertical dashed (dotted) line.
• Figure 2: The total hydrogen number density as a function of impact parameter (left), SFR (middle), and Ly$\alpha$ line luminosity (right). Only systems where $n_{\rm H}$ could be constrained using flat priors are considered here. The $p$-values from the Kendall-$\tau$ correlation tests are reported outside (within) parentheses for group (isolated) LAEs. The results indicate a strong correlation between $n_{\rm H}$ and impact parameter for the group subsample, and anti-correlation between $n_{\rm H}$ and luminosity for the isolated subsample. However, these trends are largely driven by systems with detected low-ion transitions (shown as diamonds). This is further discussed in the main text.
• Figure 3: Metallicity distribution of the absorbers associated with MUSEQuBES LAEs. Systems with well-constrained metallicities ("Measured": purple), Upper limits ("Uplims": yellow); and lower limits ("Lowlims": gray), are shown separately. The vertical arrows indicate the metallicities of two DLAs, that were derived from Si ii abundance, unlike the rest of the systems for which the Bayesian method was employed.. The gray hatched region indicates typical IGM metallicities at these redshifts Schaye_2003. Nearly $15\%$ of the absorbers lie in this region.
• Figure 4: The top panel shows the normalized H i column density distribution of MUSEQuBES absorbers for which the metallicities could be constrained while the bottom panel shows their metallicities vs. $N({\rm H \textsc{i}}$). The diamond symbols in bottom panel mark systems with detected low-ion transitions; arrows indicate upper or lower limits on metallicity ($\downarrow/\uparrow$) and $N({\rm H \textsc{i}}$) ($\leftarrow/\rightarrow$). The red dashed line marks the metallicity detection threshold set by the sensitivity of the quasar spectra (see text for details). No significant correlation is observed between the two parameters, as confirmed by the Kendall$-\tau$ test ($p-$value $= 0.2$). The gray hatched band indicates the typical IGM metallicity at $z \approx 3$. The extremely metal-poor absorber associated to the nebula from Banerjee2025_filament is highlighted with a star symbol.
• Figure 5: Top: Metallicity distribution of the 22 absorbers associated with pairs/groups. The measured metallicities (purple), upper limits (yellow) and lower limits (gray) are shown separately. The black vertical dashed line marks the median metallicity from the Kaplan–Meier estimator, ignoring the lower limits. Bottom: Same as the $\tt Top$ but for the 53 absorbers associated with isolated LAEs. No significant difference in the metallicity distributions is observed between these sub-samples.