Abstract
Deep spectroscopic surveys with the Atacama Large Millimeter/submillimeter Array (ALMA) have revealed that some of the brightest infrared sources in the sky correspond to concentrations of dusty star-forming galaxies (DSFG) at high redshift. Among these, the SPT2349-56 protocluster system at $z = 4.304$ is amongst the most extreme examples due to its high source density and integrated star formation rate. We conducted a deep Lyman-$α$ line emission survey around SPT2349-56 using the Multi-Unit Spectroscopic Explorer (MUSE) at Very Large Telescope (VLT) in order to characterize this uniquely dense environment. We identify only three LAEs at $z = 4.3$ in this field, in concordance with expectations for blank-fields, and an extended Lyman-$α$ structure spatially associated with the core of the protocluster. All the previously-identified DSFGs in this field are undetected in Lyman-$α$ emission, consistent with the conspicuous dust obscuration in these systems. We find an extended Lyman-$α$ structure, about $60 \times 60$ kpc$^2$ in size, located 56 kpc west of the protocluster core. Three DSFGs coincide spatially with the location of this structure. We conclude that either the three co-spatial DSFGs or the protocluster core itself are feeding ionizing photons to the Lyman-$α$ structure.
Introduction
Protoclusters are the "construction sites" of the universe, where massive groups of galaxies congregate and eventually merge into the giant galaxy clusters observed today. SPT2349-56 is a particularly extreme example: it is incredibly dense and boasts a star formation rate thousands of times higher than our Milky Way. However, studying these systems is challenging because most member galaxies are Dusty Star-Forming Galaxies (DSFGs). These galaxies are shrouded in cosmic dust that absorbs ultraviolet (UV) light and re-emits it as infrared, making them bright to submillimeter telescopes like ALMA but nearly invisible to optical telescopes. This study investigates whether any ionizing radiation—specifically the Lyman-$α$ (Ly$α$) line of hydrogen—escapes this dusty environment to illuminate the surrounding intergalactic medium.
Methods and Observations
The team employed the Multi-Unit Spectroscopic Explorer (MUSE) on the Very Large Telescope (VLT) in Chile. MUSE is an Integral Field Spectrograph that captures a full spectrum for every pixel in its field of view, creating a 3D "datacube." To find elusive Ly$α$ signals, the researchers used three approaches: a blind search using the LSDCat tool, narrowband imaging centered on the protocluster's redshift, and a targeted search at the coordinates of 24 previously identified ALMA galaxies.
The primary dataset consists of 10 hours of deep MUSE spectroscopic observations. This was complemented by high-resolution ALMA data ([CII] and CO lines), HST/WFC3 near-infrared imaging, and Spitzer/IRAC mid-infrared data to identify stellar populations and dusty regions. Radiative transfer modeling using the tlac code was also performed to simulate Ly$α$ photon scattering.
Key Results and Insights
The survey identified 3 secure and 5 tentative Ly$α$ emitters (LAEs) at $z \approx 4.3$. Surprisingly, this count is consistent with a "blank" field, suggesting that the dust in SPT2349-56 is exceptionally efficient at trapping UV photons. While individual galaxies were scarce, the team discovered a large, diffuse structure of Ly$α$ emission—a "Lyman-$α$ blob" (LAB).
The Lyman-$α$ Blob (LAB)
The LAB measures approximately $60 \times 60$ kpc$^2$ and is offset from the main cluster core by about 56 kiloparsecs. It has a Ly$α$ luminosity of $1.32 \times 10^{42}$ erg/s. If powered solely by star formation, this would represent a star formation rate of only ~1.3 $M_\odot$/yr—a tiny fraction of the total 60,000 $M_\odot$/yr estimated for the whole system, confirming that over 99% of the UV light is hidden by dust.
Kinematics and Power Source
The Ly$α$ emission is consistently redshifted by 300–400 km/s relative to the [CII] gas, which traces the galaxies' actual motion. This is a classic signature of galactic outflows—powerful winds driven by star formation or black holes pushing gas into space. Modeling suggests the gas has a relatively low column density ($log(N_{HI}) \approx 16$), allowing photons to escape after scattering through an expanding shell. The LAB coincides spatially with three DSFGs, suggesting it is powered either by ionizing photons escaping from these galaxies or by radiation from the protocluster's core leaking through "holes" in the surrounding dust.
Conclusion
The extreme dust obscuration in SPT2349-56 means that even deep VLT observations provide only a limited view. It remains difficult to definitively distinguish between star formation and an Active Galactic Nucleus (AGN) as the primary power source for the extended blob. Future research involving deeper spectroscopic searches for other lines (like He II or C IV) and higher-resolution mapping of dust geometry will be essential to understanding the escape of Ly$α$ radiation in such extreme environments.
Summary updated