🔭 astrophysics

The FAST Hundred-Deg $^2$ HI Deep (HD $^2$ ) Survey: Early Results from the Pilot Survey

This paper presents early results from the pilot phase of the FAST HD $^2$ survey, demonstrating its enhanced detection rate and high matching accuracy with optical counterparts to validate the feasibility of a future deep HI census for studying galaxy cold gas scaling relations.

Original authors: Chen Xu, Yingjie Jing, Jie Wang, Hu Zou, Wei Du, Niankun Yu, Hong Guo, Zerui Liu, Qingze Chen, Tiantian Liang, Zhipeng Hou, Yiwei Xu, Xiao Li, Hui-Jie Hu, Ziming Liu, Pinsong Zhao, Taotao Fang, Liang

Published 2026-05-25

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Chen Xu, Yingjie Jing, Jie Wang, Hu Zou, Wei Du, Niankun Yu, Hong Guo, Zerui Liu, Qingze Chen, Tiantian Liang, Zhipeng Hou, Yiwei Xu, Xiao Li, Hui-Jie Hu, Ziming Liu, Pinsong Zhao, Taotao Fang, Liang Gao, Qi Guo, Qiusheng Gu, Zhen Jiang, Xi Kang, Xu Kong, Cheng Li, Jun Pan, Tao Wang, Yanrui Zhou, Wenting Wang, Jing Wang

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

Imagine the universe as a giant, dark ocean. For a long time, astronomers have been trying to map the "islands" in this ocean—galaxies. But there's a problem: many of these islands are made of invisible gas (mostly hydrogen) that fuels the birth of stars. Without seeing this gas, we only see the "land" (stars) and miss the "water" (gas) that makes the island grow.

This paper is a report on a pilot test for a massive new project called HD2, which uses the world's largest radio telescope, FAST, to finally map this invisible gas in high definition.

Here is the story of what they did, how they did it, and what they found, explained simply:

1. The Tool: A Giant Ear in the Sky

Think of the FAST telescope (located in China) as a giant, super-sensitive ear. It's so big (500 meters wide) that it can hear the faint "whispers" of hydrogen gas from distant galaxies that smaller telescopes miss.

The Pilot Survey: Before building a massive map of the whole sky, the team tested the waters on a small patch of sky (about 10 square degrees, roughly the size of 40 full moons).
The Goal: They wanted to see how many "whispers" they could catch in this small area and check if their new methods worked.

2. The Challenge: Finding a Needle in a Haystack

The team was looking for hydrogen gas in galaxies. But there are two big problems:

The Gas is Invisible: You can't see it with a regular camera. You need a radio telescope to "hear" it.
The Haystack is Crowded: When the telescope hears a signal, it doesn't know which galaxy it belongs to. It's like hearing a voice in a crowded stadium and trying to guess who is speaking.

The Solution: The "DESI" Matchmaker
To solve the "who is speaking" problem, the team used data from a different project called DESI (Dark Energy Spectroscopic Instrument).

The Old Way (SDSS): Previous surveys used a "matchmaker" (SDSS) that could only identify the loudest, brightest voices. It missed the quiet ones.
The New Way (DESI): The new matchmaker (DESI) is like a super-powered detective. It can identify the voices of much fainter, dimmer galaxies.
The Result: By combining FAST's "ears" with DESI's "eyes," the team found 339 hydrogen sources in their small test patch. That's six times more than previous wide-sky surveys found in the same area!

3. The Findings: What They Discovered

Once they found the gas, they started asking questions about the galaxies that held it.

The "Blue" vs. "Red" Galaxies:
- Think of galaxies as having two "personalities." Blue galaxies are young, active, and full of gas (like a busy construction site). Red galaxies are old, quiet, and have used up their gas (like a retired neighborhood).
- The team found that while most of the gas was in the "blue" active galaxies, they also found gas in some "red" quiet galaxies. This is surprising because we thought those old galaxies had run out of fuel. It's like finding a full gas tank in a car that hasn't moved in years.
The "Gas-to-Star" Ratio:
- They measured how much gas a galaxy has compared to how many stars it has. They found that this relationship holds true whether you look at a small galaxy or a giant one. It's like finding that every house, from a tiny cottage to a mansion, follows the same rule for how much water its pipes can hold.
The "Confusion" Problem:
- Because FAST's "beam" (its field of view) is wide, sometimes it hears two galaxies talking at once. The team had to figure out how often this "confusion" happened.
- They found that about 20% of the time, the signal was a mix of two galaxies. This is actually good news for science because it means the new DESI data is so deep it's revealing galaxies that older surveys missed, even if it makes the data a bit messier to sort through.

4. Why This Matters (The "So What?")

This pilot survey is like a successful dress rehearsal before the main show.

Proof of Concept: It proved that using FAST with DESI data works perfectly. They can find faint, gas-rich galaxies that were previously invisible.
The Future: The team is now ready to expand this from a small 10-degree patch to a massive 100-degree map of the sky.
The Big Picture: By mapping this gas across such a large area, they will finally be able to understand how galaxies grow, how they run out of fuel, and how the "cosmic web" of gas connects everything in the universe.

In a nutshell: The team built a super-sensitive radio ear, used a high-tech detective to identify who was speaking, and successfully mapped a hidden layer of the universe that was previously invisible. They found that the universe is much "gas-ier" and more complex than we thought, and they are now ready to map the whole thing.

Technical Summary: The FAST Hundred-Deg2 H I Deep (HD2) Survey: Early Results from the Pilot Survey

Problem and Motivation
Blind extragalactic H I surveys have historically faced a trade-off between sky coverage and sensitivity. Wide-field interferometric surveys (e.g., ALFALFA, WALLABY) cover large areas but lack the sensitivity to detect low-mass, gas-poor systems or faint galaxies at high redshifts. Conversely, deep interferometric surveys are limited by small sky areas, making them susceptible to cosmic variance. Furthermore, blind H I surveys often suffer from biases, missing H I-poor galaxies (particularly red, early-type systems), which skews the derivation of H I mass relations. While spectral stacking and targeted observations (e.g., xGASS) have mitigated some of these issues, a high-completeness census of cold gas in the local Universe requires a survey that combines the sensitivity of a single-dish telescope with the deep spectroscopic coverage of modern optical surveys. The Dark Energy Spectroscopic Instrument (DESI) provides an unprecedented opportunity to identify optical counterparts and perform spectral stacking with high completeness, yet its performance in the context of radio surveys requires evaluation against established benchmarks like the Sloan Digital Sky Survey (SDSS).

Methodology
The authors present early results from a 10-deg² pilot survey of the Hundred-deg2 H I Deep (HD2) project, conducted using the Five-hundred-meter Aperture Spherical Radio Telescope (FAST).

Observations: The pilot survey targeted a contiguous region (RA 15h36m–16h32m, Dec 42°48′–43°48′) overlapping with the DESI Early Data Release (EDR) Survey Validation 3 (SV3), HectoMAP, and Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP). Observations utilized the 19-beam receiver in Multibeam On-The-Fly (OTF) mapping mode with a scan speed of 5″ s⁻¹. The survey achieved an effective integration time of 7.3 minutes per beam, resulting in a median map RMS of 0.45 mJy beam⁻¹ at 4.8 km s⁻¹ resolution.
Data Reduction: Raw spectra were processed using HiFAST, a dedicated pipeline for FAST H I data. This pipeline includes noise diode calibration, baseline fitting, standing-wave removal (via FFT-based filtering), flux density calibration, and gridding.
Source Finding: H I sources were identified using SoFiA 2 with low detection thresholds (4.0σ) to minimize missed detections, followed by rigorous visual inspection. The process involved checking morphology across two polarizations and three independent scans. To recover faint sources missed by the automated linking algorithm, the authors performed targeted searches at the positions of known DESI redshifts.
Optical Counterpart (OC) Matching: A probabilistic matching method was employed to associate H I sources with optical galaxies. The probability ( $P_{OC}$ ) was calculated as the product of four terms: spatial offset ( $P_{coord}$ ), velocity separation ( $P_{vel}$ ), magnitude ( $P_{mag}$ ), and color ( $P_{color}$ ). This method was tested against DESI spectroscopic redshifts, DESI photometric redshifts, and SDSS data to quantify matching and correctness rates.
Parameter Measurement: Fluxes were measured using a curve-of-growth (CoG) method to define source apertures, ensuring full flux enclosure. Velocity widths ( $W_{50}$ , $V_{85}$ ) were derived using both standard peak-fraction methods and an automated spectral CoG technique.
Stacking: To probe scaling relations for non-detected sources, the authors performed spectral stacking on isolated galaxies from the DESI BGS sample and on galaxy groups from the DESI Legacy Survey group catalog.

Key Contributions and Results

Source Catalog: The pilot survey detected 339 H I sources at $z < 0.09$ , corresponding to a source density of 33.9 deg⁻². This is nearly six times higher than the detection rate of previous wide-field surveys in similar areas. Of these, 259 were identified via SoFiA, and 80 were recovered through targeted searches at DESI redshift positions, highlighting the necessity of optical priors for faint source recovery.
Optical Counterpart Matching: The study demonstrates that DESI spectroscopic redshifts significantly improve OC identification. For galaxies with $r < 19.5$ mag, both the matching rate and correctness rate exceed 90%, a substantial improvement over SDSS. The authors quantify the "confusion" rate (where multiple optical counterparts exist within the beam) at approximately 20% for DESI-selected samples, noting that SDSS-based estimates underestimate this rate due to missing faint galaxies.
Detection Rates: Within the DESI Bright Galaxy Survey (BGS) sample, constrained to $r < 17.8$ mag and $0.01 < z < 0.05$ , the pilot survey achieved an H I detection rate of nearly 50%. For massive galaxies ( $\log M_\star > 10$ ), the rate approaches 70%.
Scaling Relations: The paper presents H I gas fraction scaling relations versus stellar mass, stellar mass surface density, $NUV-r$, and specific star formation rate (sSFR). These relations, derived from both direct detections and spectral stacking, are consistent with previous surveys (ALFALFA, DINGO, xGASS). The results confirm that blue, disk-dominated galaxies possess larger gas reservoirs.
H I – Halo Mass Relation: Using group stacking, the authors derived the H I – halo mass relation. The results for isolated galaxies ( $N_g=1$ ) agree well with ALFALFA group stacking. However, due to the limited survey volume, the pilot data does not yet resolve the potential flattening or dip in the relation at high halo masses ( $10^{11.8} - 10^{13} M_\odot$ ) predicted by some simulations.

Significance
The paper claims that the pilot survey validates the feasibility of the full HD2 survey. It demonstrates that the combination of FAST's single-dish sensitivity and DESI's deep spectroscopic coverage enables a high-completeness H I census that overcomes the biases of blind surveys. The results justify the full survey's design to map a contiguous ~100 deg² area with a sensitivity of 0.28 mJy beam⁻¹. This capability will allow for:

A more accurate determination of the faint end of the H I mass function in the local Universe.
The study of scaling relations for low-mass and gas-poor systems.
The identification of rare populations, such as low surface brightness galaxies and dark H I clouds (RELHICs).
Probing the dependence of H I content on large-scale environment and cosmic web structure, complementing interferometric surveys with significantly higher single-dish sensitivity.

The authors conclude that while the pilot survey is subject to cosmic variance and selection effects, it confirms that technical challenges have been addressed and establishes the scientific potential of the full HD2 project to provide insights into the connection between atomic gas reservoirs and galaxy properties.

1. The Tool: A Giant Ear in the Sky

2. The Challenge: Finding a Needle in a Haystack

3. The Findings: What They Discovered

4. Why This Matters (The "So What?")

More like this