🔭 astrophysics

MIGHTEE-HI: The direct detection of neutral hydrogen in galaxies at $z>0.25$

This paper presents the first direct interferometric detection of neutral hydrogen in 11 individual galaxies at redshifts greater than 0.25 using the MIGHTEE survey, revealing that these high-redshift galaxies possess significantly larger H{\sc i} masses than their local counterparts and establishing the first baryonic Tully-Fisher relation at these distances.

Original authors: Matt J. Jarvis, Madalina N. Tudorache, I. Heywood, Anastasia A. Ponomareva, M. Baes, Natasha Maddox, Kristine Spekkens, Andreea Varasteanu, C. L. Hale, Mario G. Santos, R. G. Varadaraj, Elizabeth A. K

Published 2026-03-30

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Matt J. Jarvis, Madalina N. Tudorache, I. Heywood, Anastasia A. Ponomareva, M. Baes, Natasha Maddox, Kristine Spekkens, Andreea Varasteanu, C. L. Hale, Mario G. Santos, R. G. Varadaraj, Elizabeth A. K. Adams, Alessandro Bianchetti, Barbara Catinella, Jacinta Delhaize, M. Maksymowicz-Maciata, Pavel E. Mancera Piña, Hengxing Pan, Amélie Saintonge, Gauri Sharma, O. Ivy Wong

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

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

Imagine the universe as a giant, bustling city. In this city, galaxies are the buildings, and stars are the lights inside them. But what makes the lights turn on? What fuels the construction of new stars? The answer is neutral hydrogen gas. Think of this gas as the "raw lumber" or "construction fuel" waiting to be turned into stars.

For decades, astronomers could only see this fuel in the "neighborhood" of our own cosmic city (the local universe). Looking further away was like trying to spot a single candle in a stadium from a mile away; the signal is just too faint.

This paper, titled "MIGHTEE-Hi," is a report on a successful expedition to find that faint candlelight in the distant past. Here is how they did it and what they found, explained simply.

1. The Challenge: The Faintest Whisper

Hydrogen gas emits a very specific, very weak radio signal (the "21-cm line"). It's so quiet that standard radio telescopes usually can't hear it from far away.

The Old Way: Previously, astronomers had to look at huge, random patches of sky and hope to get lucky, or they could only look at very close galaxies.
The New Strategy: Instead of shouting into the void and hoping for an answer, the team decided to knock on specific doors. They used a powerful new telescope called MeerKAT (located in South Africa) and a massive catalog of known galaxies called DESI.

The Analogy: Imagine trying to find a specific person in a crowded, noisy stadium.

Untargeted Search: You shout "Hello!" and hope someone answers. You might hear nothing or get confused by the crowd noise.
Targeted Search (This Paper): You have a list of 900 specific people's names and seat numbers. You walk directly to their seats and whisper, "Is that you?" Because you know exactly where to look, you can hear them even if they are whispering.

2. The Discovery: Finding 11 "Fuel Tanks"

Using this targeted approach, the team found 11 galaxies that are rich in hydrogen gas, located at a distance where the light has been traveling for billions of years (redshift $z > 0.25$ ).

The Record: The farthest one they found is at a distance corresponding to $z = 0.38$ . This is like looking back in time to when the universe was about 11 billion years old (roughly 3 billion years younger than it is today).
The Significance: This is the first time a group of these distant, gas-rich galaxies has been confirmed using an interferometer (a telescope that combines signals from many dishes to get a sharp image), rather than a single large dish.

3. The Surprise: The "Gas Giants"

When the team measured how much fuel these distant galaxies had, they found something interesting.

The Expectation: They thought distant galaxies might have less gas because they are younger and haven't had time to build up reserves.
The Reality: These distant galaxies were huge gas tanks. They had much more hydrogen than similar-sized galaxies we see nearby today.

The Analogy: Imagine you are looking at two cars. One is a brand-new model from 2026, and the other is a classic from 1990. You expect the new car to be more efficient. But instead, you find the 1990 car has a massive, oversized gas tank, while the 2026 car has a tiny one.

Why? It's not that the gas changed. It's that the telescope is looking at a much larger volume of space in the past. Just as a wide net catches more fish than a small one, looking at a huge volume of the early universe means we are catching the rare, "gas-rich" giants that are hard to find in our smaller, local neighborhood.

4. The Physics Check: The "Speed vs. Weight" Test

The team also tested a famous rule in astronomy called the Tully-Fisher Relation.

The Rule: There is a tight link between how heavy a galaxy is (its mass) and how fast it spins (its rotation speed). Heavier galaxies spin faster. It's like a figure skater: if they pull their arms in (adding mass to the center), they spin faster.
The Test: The team measured the spin of these 11 distant galaxies and compared it to their weight.
The Result: They fit the rule perfectly! The distant galaxies spin exactly as fast as their weight predicts, just like their local cousins.
A Tiny Twist: The heaviest, most massive galaxies seemed to spin slightly slower than the rule predicts.
- Possibility A: The rule changes as the universe ages (evolution).
- Possibility B: These heavy galaxies have a lot of molecular gas (the super-dense fuel for stars) that we couldn't see. If we added that invisible weight to the scale, the rule would work perfectly again.

5. Why This Matters

This paper is a "proof of concept." It proves that we can now reliably find and study the fuel of the universe from billions of years ago.

The Future: This is just the beginning. The team used only a fraction of their data. With more observations, they expect to find hundreds more of these "gas giants."
The Goal: By mapping out how much gas exists at different times in history, astronomers can finally write the full biography of how galaxies are born, grow, and eventually run out of fuel.

Summary

In short, the MIGHTEE team used a "smart search" strategy to find 11 distant galaxies that are swimming in a sea of hydrogen gas. They discovered that these ancient galaxies are surprisingly fuel-rich and spin in a way that mostly follows the rules of the local universe, though the heaviest ones might be hiding some extra weight. This study opens the door to a new era of understanding how the universe's "construction sites" have worked over the last 10 billion years.

1. Problem Statement

Neutral hydrogen (Hi) is the primary reservoir for molecular gas and star formation in galaxies. However, detecting the 21-cm emission line from individual galaxies at high redshift ( $z > 0.25$ ) has historically been extremely difficult due to the faintness of the line and the limited sensitivity of previous radio telescopes.

Limitations of Previous Methods: Untargeted (blind) surveys have generally been restricted to $z \lesssim 0.1$ . Targeted searches have been attempted but often lacked the angular resolution to distinguish individual systems or suffered from confusion. Single-dish telescopes (like FAST) have detected high-redshift Hi but suffer from poor angular resolution, leading to potential mass overestimates due to source confusion.
The Gap: There is a lack of direct, interferometric detections of Hi in individual galaxies beyond the local Universe ( $z > 0.25$ ), hindering the study of gas reservoir evolution, the baryonic Tully-Fisher relation (bTFr), and the Hi mass function at cosmic timescales corresponding to the peak of star formation.

2. Methodology

The authors utilized the MeerKAT International Tiered Extragalactic Exploration (MIGHTEE) Survey, specifically the MIGHTEE-Hi Data Release 1 (DR1) spectral line cube covering the COSMOS field.

Data Sources:
- Radio: MIGHTEE-L1 spectral cube (960–1150 MHz, $\sim$ 5 deg $^2$ , $\sim$ 15 arcsec resolution, 104.5 kHz channels).
- Optical/NIR: Dark Energy Spectroscopic Instrument (DESI) Early Data Release, providing spectroscopic redshifts for emission-line galaxies.
Target Selection:
- The team cross-matched the MIGHTEE-Hi cube with the DESI catalogue, restricting the search to the redshift range $0.25 < z < 0.5$ (corresponding to the 960–1150 MHz window).
- This resulted in an initial sample of 915 galaxies with known optical spectroscopic redshifts.
Detection Strategy:
- Targeted Search: Instead of a blind search, the team performed a targeted search for 21-cm emission centered on the exact position and redshift of the DESI galaxies.
- Visual Inspection: 3-D data cubes were collapsed into 2-D moment-0 maps and 1-D spectra. Candidates were identified by eye based on:
  1. A prominent signal in the 2-D image within 4 arcsec of the optical position.
  2. Significant emission in the 1-D spectrum spanning at least 5 channels ( $\sim$ 150 km s $^{-1}$ ).
- Noise Estimation: To quantify robustness, the team measured the noise distribution using $\sim$ 500 random spatial regions near each source. They calculated the Signal-to-Noise Ratio (SNR) using the 16th and 84th percentiles of the flux density distribution (assuming Gaussian noise, verified via Anderson-Darling test).
- False Positive Mitigation:
  - Offset Check: The search was repeated at positions offset by 50 arcmin from the optical galaxies to check for spurious noise spikes.
  - Continuum Check: Radio continuum maps were checked to ensure detections were not artifacts of imperfect continuum subtraction.
  - Systematic Uncertainties: Systematic errors were estimated by comparing aperture flux measurements (10 $\times$ 10 arcsec $^2$ ) with single-pixel peak flux measurements to account for marginally resolved sources.

3. Key Contributions

First Interferometric Sample at $z > 0.25$ : This study presents the first sample of 11 individually detected Hi galaxies at $z > 0.25$ using an interferometer (MeerKAT), with the highest redshift detection at $z = 0.3841$ .
Methodological Advancement: It demonstrates the efficacy of combining deep interferometric Hi data with large spectroscopic redshift surveys (DESI) to overcome the sensitivity limits of blind searches.
Baryonic Tully-Fisher Relation (bTFr): It provides the first-ever measurement of the bTFr using Hi at $z > 0.25$ , utilizing the Hi line width ( $W_{50}$ ) as a proxy for rotation velocity.

4. Key Results

Sample Properties:
- 11 Confirmed Detections: All have SNR $> 4$ . The sample spans $0.25 \lesssim z \lesssim 0.39$ .
- Hi Masses: The detected galaxies have Hi masses ranging from $\log(M_{HI}/M_\odot) \approx 10.2$ to $10.9$.
- Stellar Properties: Using SED fitting (Bagpipes) with multi-wavelength data (UV to FIR), the authors derived stellar masses ( $M_*$ ) and Star Formation Rates (SFR). The galaxies generally lie on or above the star-forming main sequence.
Hi Mass vs. Stellar Mass:
- The detected galaxies exhibit significantly larger Hi masses for a given stellar mass compared to low-redshift ( $z < 0.1$ ) Hi-selected samples and stacking results at similar redshifts.
- Interpretation: This is attributed to the Malmquist bias inherent in direct detections (only the most gas-rich galaxies are detectable at these distances) and the larger cosmological volume probed at higher redshifts. The authors argue this does not require an evolution in the Hi mass function; the volume effect explains the detection of rare, massive Hi systems.
Gas Depletion Times:
- The gas depletion times ( $t_{dep} = M_{HI}/SFR$ ) for this sample are generally lower (mean $\sim$ 3.1 Gyr) than low-redshift samples, driven by higher SFRs in these high-redshift, gas-rich systems.
Baryonic Tully-Fisher Relation (bTFr):
- The sample follows the local bTFr but probes much higher baryonic masses and rotation velocities.
- Flattening: There is tentative evidence of a flattening of the bTFr at high masses/redshifts (high-mass objects lie below the local relation).
- Explanations:
  1. Evolution: Consistent with hydrodynamic simulations (e.g., simba) predicting evolution in the bTFr due to merger histories.
  2. Molecular Gas: The flattening may be an artifact of unaccounted molecular gas. If the molecular gas mass is comparable to the Hi mass in these massive galaxies, including it would restore the galaxies to the local bTFr.
  3. Kinematics: At high masses, the global Hi line width ( $W_{50}$ ) may trace $V_{max}$ rather than the flat rotation velocity ( $V_{flat}$ ), leading to an overestimation of the true rotational velocity.

5. Significance

Proof of Concept: This study proves that direct detection of Hi in individual galaxies at $z > 0.25$ is feasible with current-generation interferometers like MeerKAT when utilizing targeted approaches.
Future Prospects: It paves the way for future studies of gas evolution in the Universe. The authors predict that with the full MIGHTEE survey and the deeper LADUMA survey, many more detections will be possible, allowing for robust constraints on the evolution of the Hi mass function and the bTFr up to $z \sim 0.5$ .
Pre-SKA Era: These results serve as a critical precursor to the Square Kilometre Array (SKA), demonstrating the science potential of high-redshift Hi studies before the SKA comes online.

In summary, the paper successfully bridges the gap between local Hi studies and high-redshift cosmology, revealing a population of gas-rich, star-forming galaxies at $z \sim 0.35$ and providing new constraints on the scaling relations that govern galaxy evolution.