🔭 astrophysics

CHORD HI-Galaxy Survey Forecasts: Searching for nearby dark galaxies and high redshift giants

This paper forecasts that the next-generation CHORD radio telescope will revolutionize the census of neutral hydrogen galaxies by detecting roughly ten times more sources than current catalogs, extending the survey to significantly lower-mass galaxies ( $M_{\mathrm{HI}} \sim 10^{5.5} M_{\odot}$ ) and uncovering thousands of massive high-redshift giants to better understand the evolution of galactic gas reservoirs.

Original authors: Akanksha Bij (Department of Physics, Engineering Physics and Astronomy, Queen's University, Kingston, Canada), Kristine Spekkens (Department of Physics, Engineering Physics and Astronomy, Queen's Univ

Published 2026-07-29✓ Author reviewed ⓘ

📖 8 min read🧠 Deep dive

CC BY 4.0

Original authors: Akanksha Bij (Department of Physics, Engineering Physics and Astronomy, Queen's University, Kingston, Canada), Kristine Spekkens (Department of Physics, Engineering Physics and Astronomy, Queen's University, Kingston, Canada), Hans S. Hopkins (Perimeter Institute for Theoretical Physics, Waterloo, Canada, Waterloo Center for Astrophysics, University of Waterloo, Waterloo, Canada), Michael G. Jones (IPAC, Caltech, Pasadena, USA), Arnab Chakraborty (Department of Physics, McGill University, Montreal, Canada, Trottier Space Institute, Montreal, Canada), Simon Foreman (Department of Physics, Arizona State University, Tempe, USA), Alex S. Hill (Department of Math, Physics, & Statistics, University of British Columbia, Okanagan Campus, Kelowna, Canada, Dominion Radio Astrophysical Observatory, Herzberg Research Centre for Astronomy and Astrophysics, National Research Council, Penticton, Canada), Dustin Lang (Perimeter Institute for Theoretical Physics, Waterloo, Canada, Waterloo Center for Astrophysics, University of Waterloo, Waterloo, Canada), Adrian Liu (Department of Physics, McGill University, Montreal, Canada, Trottier Space Institute, Montreal, Canada)

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 by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Here is the revised summary:

Imagine the universe as a giant, cosmic ocean. Most of the time, when we look at this ocean, we see the bright, glittering islands of galaxies—massive collections of stars that light up the darkness. But astronomers have long suspected that beneath the waves, there are hidden islands made entirely of invisible gas, floating in the dark without a single star to give them away. These are called "dark galaxies." To find them, and to understand how the visible islands formed in the first place, scientists need to map the distribution of a specific type of gas called Neutral Hydrogen (H I). Think of Neutral Hydrogen as the raw building material of the universe; it's the fuel that stars and galaxies drink to grow. By measuring how much of this fuel exists in galaxies of different sizes, from tiny dwarfs to massive giants, scientists can build a "mass function"—a census that tells us how common small gas clouds are compared to huge ones. This census is crucial because it helps us test our theories about how the universe works, specifically whether our current models of gravity and dark matter are missing something fundamental about how galaxies are born.

Enter CHORD, a brand-new radio telescope currently being built in Canada, which is about to take a massive, high-definition snapshot of this cosmic gas. This paper is a set of forecasts, essentially a "what-if" simulation, predicting exactly what CHORD will see once it starts its full survey. The authors used computer models to generate millions of fake galaxies based on what we already know, then simulated how CHORD would detect them. They found that in just one year, CHORD will spot hundreds of thousands of galaxies, and in five years, it will find a million. Most excitingly, the telescope is so sensitive that it will finally be able to see the tiny, faint gas clouds that have been hiding in the dark, potentially finding dozens of the elusive "dark galaxies" that have only been theoretical until now. At the other end of the scale, it will also find thousands of giant, gas-rich galaxies far away in the early universe, helping us understand how these massive monsters evolved over time.

The Cosmic Census: What CHORD Will See

The paper focuses on the Canadian Hydrogen Observatory and Radio-transient Detector (CHORD), a next-generation radio telescope being constructed at the Dominion Radio Astrophysical Observatory. Unlike traditional telescopes that use a single giant dish or a few scattered ones, CHORD is a "redundant interferometer." Imagine a choir where every singer is singing the exact same note in perfect unison; this redundancy makes the sound (or in this case, the radio signal) incredibly loud and clear, allowing the telescope to see very faint signals. CHORD will consist of 512 small dishes arranged in a grid, scanning the northern sky as the Earth rotates.

The authors created a "mock" universe, a computer simulation filled with galaxies drawn from our best current estimates of how common they are. They then ran a simulation of CHORD's 1-year and 5-year surveys to see how many galaxies it would detect. The results are staggering. A 1-year survey is predicted to find about 300,000 galaxies, while a full 5-year survey could detect roughly 1 million galaxies. This would increase the number of known gas-rich galaxies by about ten times compared to current catalogs.

Hunting the Invisible: The Low-Mass Frontier

One of the main goals of this survey is to find the "dark" or "starless" galaxies. These are clouds of gas that are massive enough to be galaxies but haven't started making stars yet, or perhaps never will. The paper suggests that CHORD will be able to push the census of galaxies down to a mass of about $10^{5.5} M_{\odot}$ (solar masses). To put that in perspective, previous surveys could only see galaxies down to about $10^7 M_{\odot}$ . This means CHORD will be looking at objects that are roughly 30 times smaller and fainter than anything we've reliably mapped before.

The authors predict that in a 5-year survey, CHORD will find thousands of these tiny gas clouds. Specifically, they estimate finding about 70 sources similar to "Leo T," a known dwarf galaxy, and potentially around 30 "dark galaxies" (specifically a type called HIDEs) that are completely devoid of stars. These discoveries would be a huge deal because they would provide the first real observational proof of a theoretical limit: a mass threshold below which dark matter halos are too small to hold onto gas and form stars. If CHORD finds these objects, it confirms our theories; if it finds a sudden drop-off in numbers, it tells us there's a hard limit to how small a galaxy can be.

The Giants of the Deep: High-Redshift Forecasts

On the other end of the spectrum, CHORD will also look for the giants. The paper forecasts that the telescope will detect about 1,000 massive, gas-rich galaxies at distances corresponding to redshifts between 0.3 and 0.5. These are galaxies that are much larger and more massive than our own Milky Way, and finding them so far away helps scientists understand how these giants evolved over cosmic time. Currently, we don't have a good count of these massive objects at these distances because they are rare, and building a statistical sample means surveying a very large volume of the universe, out to great distances. CHORD's wide field of view and sensitivity will allow it to spot these giants in numbers that were previously impossible, helping to fill in the "high-mass end" of the galaxy census.

The Challenges: Noise and Confusion

Of course, looking at the universe isn't as simple as pointing a telescope and waiting. The paper carefully considers two major hurdles: Radio Frequency Interference (RFI) and source confusion.

RFI is like static on a radio caused by human technology, such as GPS satellites. The authors note that a specific range of frequencies (corresponding to redshifts between 0.1 and 0.3) is heavily contaminated by these signals. While CHORD has advanced technology to try to filter this out, the authors suggest that this "noise zone" might limit the telescope's ability to see clearly in that specific distance range.

Source confusion is another issue. Because CHORD has a relatively large "beam" (the area of sky it looks at at once), it might accidentally see two or more galaxies overlapping in the same spot. The paper estimates that in the middle distance range (between 40 million and 100 million light-years), about 25% to 60% of the detections might be "blends" of multiple galaxies. However, the authors argue that this won't ruin the science. For the tiny, nearby galaxies, confusion is rare. For the distant, massive giants, even if they are blended with smaller neighbors, the massive galaxy usually dominates the signal, so we can still tell it's there. The authors suggest this "blended" zone could actually be useful, acting as a bridge to help test other methods of measuring the universe.

The Cosmic Lottery: Cosmic Variance

The paper also discusses "cosmic variance," which is essentially the luck of the draw regarding where our telescope is pointing. If CHORD points at a region of space that happens to be a giant cluster of galaxies (like the Virgo Cluster), it might find way more galaxies than average. If it points at a void, it might find very few. The authors used a sophisticated simulation called SIBELIUS-DARK to model this. They found that for the smallest, faintest galaxies, the results could vary significantly depending on the local neighborhood because these tiny galaxies can only be seen in our "backyard" (within about 40 million light-years). This means the final count of tiny galaxies will have some uncertainty, but the huge number of galaxies CHORD will find overall should still give us a very clear picture of the universe.

The Bottom Line

In summary, this paper is a roadmap for what CHORD will achieve. It suggests that with a 5-year survey, we will move from a blurry, low-resolution picture of the universe's gas content to a high-definition census. We will likely find thousands of tiny, faint galaxies that have never been seen before, potentially solving the mystery of "dark galaxies." We will also map out the distribution of massive giants in the early universe. While challenges like radio noise and overlapping signals exist, the authors are confident that CHORD will revolutionize our understanding of how galaxies form, grow, and evolve, providing the first solid observational constraints on the very edges of the galaxy population.

Technical Summary: CHORD H I-Galaxy Survey Forecasts

Problem Statement
Population studies of gas-rich galaxies across the full range of Neutral Hydrogen (H I) masses ( $10^5 \lesssim M_{\text{H I}} \lesssim 10^{11} M_\odot$ ) are currently limited by the sensitivity and sky coverage of existing surveys. The H I mass function (HIMF) is poorly constrained at both extremes: the low-mass end suffers from sensitivity limitations and small survey volumes, while the high-mass end is constrained by the rarity of massive sources and the exponential cutoff in the local universe. Furthermore, the redshift evolution of the HIMF remains uncertain, particularly for $z > 0.1$ , due to radio frequency interference (RFI) and limited survey volumes. There is a specific need to identify "dark galaxies" (starless H I reservoirs) predicted by $\Lambda$ CDM simulations and to characterize the evolution of massive, gas-rich giants.

Methodology
The authors forecast the performance of the Canadian Hydrogen Observatory and Radio-transient Detector (CHORD), a next-generation, highly redundant interferometer located at the Dominion Radio Astrophysical Observatory. The study employs a simulation-based approach to predict survey outcomes for fiducial 1-year and 5-year drift-scan surveys covering declinations $+20^\circ < \delta < +80^\circ$ .

Mock Catalog Generation: Galaxy catalogs are drawn from the ALFALFA HIMF (Jones et al. 2018), extrapolated to masses between $10^5$ and $10^{12} M_\odot$ . Sources are assigned random sky locations, redshifts, and inclinations. Rotational velocities ( $V_{\text{H I}}$ ) are derived from the H I mass using a polynomial fit (Lewis 2019).
Spectral Modeling: Mock H I spectra are generated using the generalized "busy function" (Westmeier et al. 2014) to model diverse profile shapes (Gaussian, boxy, double-horned). Profiles are convolved with a Gaussian kernel to account for thermal and turbulent broadening.
Survey Simulation: The authors calculate the noise per spectral channel using the radiometer equation, incorporating CHORD's specific parameters (512 dishes, 6m diameter, $T_{\text{sys}} = 30$ K, 14,400 m $^2$ collecting area). Integration times are scaled by declination to achieve uniform sensitivity across the footprint.
Detection Criteria: Sources are considered detected if they achieve an integrated signal-to-noise ratio ( $S/N_{\text{int}}$ ) of 6.
Confusion and Variance Analysis: The study estimates spectroscopic confusion rates using the model from Jones et al. (2015) and assesses cosmic variance using a constrained N-body simulation (SIBELIUS-DARK) to evaluate large-scale structure biases.

Key Contributions and Results

Survey Sensitivity and Source Counts:
- A 1-year survey (8,700 deg $^2$ ) is forecast to achieve a sensitivity of 0.6 mJy (5 km s $^{-1}$ resolution), detecting $\sim 3 \times 10^5$ sources.
- A 5-year survey (13,000 deg $^2$ ) is forecast to reach 0.3 mJy sensitivity, detecting $\sim 10^6$ sources. This represents an increase of one to two orders of magnitude over current catalogs like ALFALFA.
Low-Mass End Constraints:
- The 5-year survey is expected to push the H I mass census down to $M_{\text{H I}} \sim 10^{5.5} M_\odot$ , an order of magnitude deeper than previous surveys.
- This regime allows for the detection of thousands of dwarf galaxies ( $10^6 < M_{\text{H I}} < 10^7 M_\odot$ ) and roughly 70 Leo T-type sources ( $10^5 < M_{\text{H I}} < 10^6 M_\odot$ ).
- If the low-mass end of the HIMF turns over (as predicted by some models), CHORD will provide the first observational constraints on the mass threshold of this turnover.
High-Mass and High-Redshift Forecasts:
- CHORD is predicted to detect $\sim 10^3$ massive gas-rich giants ( $M_{\text{H I}} \gtrsim 10^{10.5} M_\odot$ ) in the redshift range $0.3 \lesssim z \lesssim 0.5$ .
- These detections will explore the evolution of the high-mass end of the HIMF relative to local universe estimates.
Dark Galaxy Candidates:
- Based on number density estimates from simulations (Zheng et al. 2025), the authors forecast the detection of approximately 30 dark galaxies (HIDEs) in the 5-year survey within the Local Group vicinity ( $5 < D < 10$ Mpc).
- These candidates are expected to be isolated, distinguishing them from tidal or ram-pressure stripped clouds.
Impact of Confusion and RFI:
- Confusion: Spectroscopic confusion is negligible for nearby sources ( $D < 40$ Mpc) but affects $\sim 25\%$ of detections at intermediate distances ( $40 < D < 0.2$ Gpc). At high redshifts ( $z > 0.2$ ), while the confusion rate for individual massive sources is low ( $<3\%$ ), the contribution from stacked, sub-threshold sources is estimated at $\sim 20\%$ of the detection mass limit.
- RFI: The redshift range $0.1 \lesssim z \lesssim 0.3$ (1090–1290 MHz) is heavily contaminated by GNSS RFI. While the 5-year survey may mitigate this via millisecond time-resolution data, the 1-year survey's redshift reach is significantly limited by these effects.

Significance
The paper claims that CHORD will be the first highly-redundant interferometer to conduct an extragalactic H I survey, offering a unique capability to probe the extremes of the galaxy population. By expanding the census of H I-rich galaxies by an order of magnitude, CHORD will:

Provide the first observational constraints on the low-mass turnover of the HIMF, testing theories of galaxy formation in low-mass dark matter halos.
Deliver a census of rare, massive gas-rich giants at $z \sim 0.3\text{--}0.5$ , constraining the high-mass slope and its evolution.
Enable the discovery of a population of dark galaxies, potentially resolving the "missing satellites" problem by identifying starless H I reservoirs.
Serve as a bridge between individual galaxy detections and intensity mapping, particularly in the intermediate distance regime where source blending is significant but informative.

The authors conclude that while cosmic variance and source confusion introduce uncertainties, the sheer volume and depth of the CHORD survey will significantly improve the understanding of neutral gas reservoirs at both the low-mass and high-mass extremes of the galaxy population.