MUSE-ALMA Haloes XIII. Molecular gas in $z \sim 0.5$ HI-selected galaxies

The MUSE-ALMA Haloes survey significantly expands the sample of HI-selected galaxies at $z \sim 0.5$ by doubling the number of CO detections, revealing a dual star formation efficiency where low-mass systems follow main-sequence relations while high-mass systems exhibit suppressed activity, thereby offering crucial insights into the baryon census and galaxy evolution at this redshift.

The Gist

The Cosmic Gas Hunt: Finding the "Invisible" Fuel of Galaxies

Imagine the universe as a giant, bustling city. In this city, galaxies are the neighborhoods, and stars are the people living there. But just like people need food to survive, stars need gas (specifically hydrogen) to be born.

For a long time, astronomers have been trying to map out where this gas is. They usually look for bright, glowing gas clouds that are actively making stars—like looking for a city by the bright lights of its busiest downtown. But this paper is about a different kind of search. The astronomers are looking for the "dark" or "quiet" neighborhoods where the gas is there, but it's not lighting up the night sky yet.

Here is the story of their new discovery, broken down simply:

1. The "Shadow" Detective Work

Most astronomers look for galaxies by seeing their light. But this team, led by the MUSE-ALMA Haloes project, used a clever trick. They looked at bright, distant quasars (which are like cosmic lighthouses) and watched for shadows.

When gas clouds pass in front of a lighthouse, they block a tiny bit of the light, creating a "shadow" or an absorption line. By studying these shadows, the team found 79 galaxies hiding around these lighthouses. These are the "shadow galaxies." They are often dim, quiet, and hard to see directly, but they are full of raw material (atomic hydrogen gas).

2. The Big Question: Is the Gas Ready to Cook?

Finding the raw gas (hydrogen) is like finding a warehouse full of flour. But to make bread (stars), you need to turn that flour into dough (molecular gas). In astronomy, we look for Carbon Monoxide (CO) to find this "dough."

The team used the ALMA telescope (a super-powerful radio telescope) to look at these 79 shadow galaxies and asked: "Is the flour ready to be baked? Is there molecular gas here?"

3. The Results: A Tale of Two Cities

They looked at 60 of these galaxies (adding 39 new ones to the list). Here is what they found:

• The Success Rate: They found molecular gas in 12 out of 60 galaxies (about 20%). This is a big deal because they didn't pick the "easy" targets; they picked them randomly based on the shadows.
• The "Efficient" Galaxies (The Bakers): Some of the galaxies with molecular gas were like efficient bakeries. They had just enough gas, and they were turning it into stars very quickly. These galaxies fit right in with the "normal" busy cities we already know.
• The "Hoarding" Galaxies (The Piled-Up Flour): This is the surprise! The other galaxies with molecular gas were sitting on massive piles of flour (huge amounts of gas) but were barely baking any bread (very few new stars).
  • Analogy: Imagine a bakery with a warehouse full of flour, but the ovens are cold. They have all the ingredients to make a million loaves, but they aren't making any.

4. Why Are Some Galaxies "Hoarding" Gas?

The paper suggests two main reasons for these "hoarding" galaxies:

1. They are new arrivals: They might have just recently pulled a huge truckload of gas from the space between galaxies (the cosmic web). They have the fuel, but the "engine" (star formation) hasn't started up yet.
2. They are in a bad neighborhood: They might be in a group of galaxies where the environment is chaotic. Maybe the gas is too turbulent, or the metal content (which helps gas clump together) is too low, making it hard to turn that gas into stars.

5. The "Invisible" Problem

The team also noticed that for the galaxies where they didn't find molecular gas, it doesn't necessarily mean the gas wasn't there. It might be "CO-dark."

• Analogy: Imagine trying to find a campfire in a thick fog. If the fire is small or the fog is too thick, you can't see the flames, even though the fire is there. Similarly, in these low-metallicity galaxies, the gas might be there, but the conditions are so "foggy" (low dust, low heavy elements) that the carbon monoxide signal is too faint for our telescopes to see.

The Big Picture

This paper is a major step in understanding the life cycle of galaxies.

• Before: We mostly studied the "loud" galaxies that were already making stars.
• Now: We are studying the "quiet" ones. We found that the universe is full of galaxies that are gathering fuel but haven't started the engine yet.

By finding these "shadow" galaxies and measuring their hidden gas, the astronomers are completing the census of the universe's baryons (normal matter). They are showing us that the story of galaxy evolution isn't just about the bright, star-filled cities; it's also about the quiet, gas-rich suburbs that are just getting ready to wake up.

In short: They found that many galaxies are sitting on massive gas reserves, waiting to turn into star factories, proving that the universe is full of potential that we are only just beginning to understand.

Technical Summary

Here is a detailed technical summary of the paper "Molecular gas in z ∼0.5 H i –selected galaxies" (MUSE-ALMA Haloes XIII) by Bollo et al. (2026).

1. Problem and Scientific Context

Galaxy evolution is regulated by the baryon cycle—the accretion, conversion to stars, and ejection of gas. While the atomic hydrogen (H i) and ionized gas phases of the circumgalactic medium (CGM) have been extensively studied, the molecular gas (H₂) content in H i-selected systems remains poorly understood.

• The Bias: Previous studies of molecular gas in absorbers often relied on pre-selection based on high metallicity or known H₂ absorption, leading to high detection rates (>60–80%) but potentially biased samples that miss the broader population of gas-rich, metal-poor systems.
• The Gap: There is a lack of systematic, unbiased surveys of molecular gas in galaxies selected solely via strong H i absorption (DLAs and sub-DLAs) at intermediate redshifts ($z \sim 0.5$). Understanding whether these systems follow standard star-forming scaling relations or exhibit distinct evolutionary phases (e.g., gas accumulation without efficient star formation) is critical for a complete baryon census.

2. Methodology

The study utilizes the MUSE-ALMA Haloes survey, a multi-wavelength project combining VLT/MUSE, HST, and ALMA data.

• Sample Selection: The target sample consists of 60 galaxies associated with strong H i absorbers ($\log N(\text{H i}) > 18$) at redshifts $0.3 < z < 1.2$.
  • New Observations: This paper presents new ALMA Cycle 10 Large Program observations (PI: C. Péroux) of 39 galaxies, expanding an initial pilot sample of 21.
  • Unbiased Approach: Crucially, the sample was selected based only on H i absorption properties, without pre-selection for metallicity or star formation activity.
• Observations:
  • Instrument: ALMA (Band 4 and Band 6).
  • Targets: CO(2–1), CO(3–2), and CO(4–3) transitions.
  • Sensitivity: Reaching an RMS of $\sim 0.16$ mJy beam$^{-1}$, probing molecular gas masses ($M_{\text{H2}}$) $\sim 1.2$ dex deeper than previous absorber studies.
• Analysis Techniques:
  • Emission Search: A dual approach using targeted aperture optimization (rotating elliptical apertures to maximize SNR) and a blind search using sofia-2.
  • Stacking: For the 30 non-detected sources, a spectral stacking analysis was performed on the CO(2–1) line to detect average emission, weighted by luminosity distance and sensitivity.
  • Physical Properties: Molecular gas masses ($M_{\text{H2}}$) were derived using CO luminosities and a metallicity-dependent $\alpha_{\text{CO}}$ conversion factor (following Genzel et al. 2015). Stellar masses and metallicities were derived from HST photometry and VLT/MUSE emission lines.

3. Key Contributions

• First Unbiased Survey: This is the first large-scale, non-metallicity-biased survey of molecular gas in H i-selected galaxies at $z \sim 0.5$.
• Deep Sensitivity: The survey pushes the detection limit of molecular gas in absorbers by more than an order of magnitude (factor of $\sim 1.2$ dex) compared to previous work, accessing the regime of low-mass, low-metallicity systems.
• Comprehensive Dataset: It combines 60 systems with multi-wavelength data (H i column densities, stellar masses, star formation rates, and gas-phase metallicities), allowing for a robust comparison with the main sequence of star-forming galaxies.

4. Key Results

A. Detection Rates and Statistics

• Detection Rate: CO emission was detected in 9 out of 39 newly observed galaxies (23%).
• Total Sample: Combining with previous data, the total detection rate is 12 out of 60 galaxies (20%).
• Stacked Signal: The stacked spectrum of non-detected sources revealed a significant signal at $\sim 4.2\sigma$, confirming the presence of molecular gas in the non-detected population, albeit below individual detection thresholds.
• No Simple Correlations: No clear correlation was found between CO detection and redshift, stellar mass, or impact parameter. However, a statistically significant difference ($p=0.004$) exists in the metallicity distributions: CO-detected galaxies tend to have higher metallicities ($\log Z \gtrsim 8.5$) compared to non-detections.

B. Dual Behavior in Star Formation Efficiency

The study reveals a bimodal behavior in the relationship between molecular gas mass ($M_{\text{H2}}$) and star formation:

1. Low-$M_{\text{H2}}$ Systems ($\log M_{\text{H2}} \lesssim 9.8$): These galaxies lie slightly above ($\sim 0.3$ dex) the star-forming main sequence. They exhibit efficient star formation, following standard scaling relations for depletion times and stellar mass growth.
2. High-$M_{\text{H2}}$ Systems ($\log M_{\text{H2}} > 9.8$): These galaxies lie significantly below ($\sim 1.5$ dex) the main sequence. They possess massive molecular reservoirs but exhibit suppressed star formation rates and lower-than-expected stellar masses.
   • Interpretation: These systems likely represent galaxies in an early evolutionary stage, actively accreting gas from the intergalactic medium (IGM) or interacting within group environments, where the conversion of gas to stars has not yet reached equilibrium.

C. The "CO-Dark" Phenomenon

The lack of CO detection in many metal-poor systems (despite the presence of H i) suggests the existence of "CO-dark" molecular gas. In low-metallicity environments, UV radiation penetrates deeper into clouds, dissociating CO while H₂ remains shielded. This implies that the true molecular gas content in these absorbers is likely underestimated by CO observations alone.

D. Environmental Context

• 11 of the 12 CO-detected galaxies are found in environments with other galaxies at similar redshifts (groups or filaments), suggesting that environmental interactions (mergers, tidal forces, group-scale accretion) play a key role in accumulating molecular gas.
• Kinematic analysis indicates that absorbing gas in these systems can originate from inflows, outflows, or rotating disks, highlighting the physical diversity of the CGM.

5. Significance and Implications

• Completing the Baryon Census: By probing molecular gas masses more than 1 dex below previous limits, this survey provides a more complete picture of the cold gas budget in the universe at $z \sim 0.5$, capturing systems that emission-selected surveys miss.
• Challenging Scaling Relations: The discovery of high-$M_{\text{H2}}$ systems with suppressed star formation challenges the universality of standard scaling relations, indicating that H i absorbers trace a distinct phase of galaxy evolution (gas buildup) not well-represented in flux-limited samples.
• Simulation Constraints: The observed diversity in star formation efficiencies and the prevalence of CO-dark gas provide critical constraints for cosmological simulations. Models must now account for non-equilibrium chemistry, metallicity-dependent CO-to-H₂ conversion, and the complex interplay between CGM accretion and star formation feedback.
• Future Directions: The results highlight the need for alternative tracers (e.g., [C I], dust continuum) to fully characterize the molecular content of metal-poor, H i-selected systems.

In summary, the MUSE-ALMA Haloes XIII survey demonstrates that H i-selected galaxies are a heterogeneous population comprising both evolved, efficient star-forming systems and young, gas-rich systems undergoing inefficient star formation, likely driven by ongoing gas accretion and environmental interactions.