A Wide and Deep Exploration of Radio-detected Active Galactic Nuclei with Subaru HSC (WERGS). XIII. High-z Radio Quasar Selection from HSC--VLASS over about1200 deg squared

arXiv:2603.28044 · astro-ph.GA · Submitted 2026-03-30 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.

Vera: Next we'll be talking about the paper "A Wide and Deep Exploration of Radio-detected Active Galactic Nuclei with Subaru HSC (WERGS). XIII. High-z Radio Quasar Selection from HSC--VLASS over about1200 deg squared ".

Jocelyn: The paper was written by Youwen Kong, Kohei Ichikawa, Hisakazu Uchiyama, Yuxing Zhong, Xiaoyang Chen et al. from Institute of Astronomy, Graduate School of Science, The University of Tokyo and Frontier Research Institute for Interdisciplinary Sciences, Tohoku University and Astronomical Institute, Tohoku University and Department of Advanced Sciences, Faculty of Science and Engineering, Hosei University and National Astronomical Observatory of Japan and Department of Physics, School of Advanced Science and Engineering, Faculty of Science and Engineering, Waseda University and Research Center for the Early Universe, Graduate School of Science, The University of Tokyo and ILANCE, CNRS – University of Tokyo International Research Laboratory and Research Center for Space and Cosmic Evolution, Ehime University and Department of Physics, Graduate School of Science, Nagoya University and Academia Sinica Institute of Astronomy and Astrophysics and National Taiwan University and Department of Physical Sciences, Ritsumeikan University and Astronomy Research Group, Institut Teknologi Bandung and Ecole Polytechnique.

Vera: Stay tuned as we take you through the paper and discuss its implications.

Title: Vera: We're starting today with a massive new paper that's been hitting the arXiv, titled "A Wide and Deep Exploration of Radio-detected Active Galactic Nuclei with Subaru HSC (WERGS). XIII. High-z Radio Quasar Selection from HSC--VLASS over about1200 deg two".

Jocelyn: That is quite a long title, Vera, but I see the WERGS project is moving into its thirteenth installment.

Vera: It really is, and the team behind it—led by Youwen Kong and Kohei Ichikawa—has been working on this for quite a while.

Jocelyn: I noticed they're covering about one thousand two hundred square degrees of the sky, which seems like a huge area to scan for something as rare as high-redshift quasars.

Vera: It’s an enormous footprint, especially when you consider they're combining the Subaru Hyper Suprime-Cam data with the VLASS radio survey.

Jocelyn: Does that mean they're looking for things that are visible in both optical and radio wavelengths?

Vera: Exactly, and that's how they can pinpoint these active galactic nuclei even when they're incredibly far away.

Subrahmanyan: This scale is exactly what we need because if you only look at tiny patches of the sky, you'll never find enough of these high-redshift objects to understand how the early universe actually functioned.

Jocelyn: So, Subrahmanyan, why is it so vital that we find these specific radio-detected quasars from that early epoch?

Subrahmanyan: We're trying to figure out how supermassive black holes grew so fast and how their energy output influenced the galaxies around them during the first few billion years.

Vera: And this paper seems to be setting up a much bigger search than anyone has managed before.

Jocelyn: It sounds like they're building a massive catalog to help us see that era more clearly.

Vera: They really are, and we should look at what they actually found in those one thousand two hundred degrees.

Summary: Vera: Moving into the actual findings, this team identified about four hundred high-redshift radio AGN candidates sitting out there at redshifts greater than four.

Jocelyn: That's a significant number of candidates for such a distant part of space.

Vera: It is, and most of them are quite faint in the optical bands, clustering around an i-band magnitude of twenty-four to twenty-six.

Jocelyn: If they're that faint, I'm guessing older surveys like SDSS just couldn't see them at all?

Vera: You hit the nail on the head; those objects are far too dim for shallower surveys to detect reliably.

Jocelyn: What about their radio signatures, though? Are they showing the usual steep spectra we expect from high-redshift sources?

Vera: That's actually one of the most surprising parts, because most of them show flat or only moderately steep radio spectra instead.

Subrahmanyan: This is a huge deal because it tells us that our previous understanding of these objects was biased toward a specific type.

Jocelyn: How does that bias affect the way we calculate things like their total energy output?

Subrahmanyan: Well, when they did the SED fitting, they found these AGNs cluster around bolometric luminosities of log L equal to forty-six or forty-seven.

Vera: And they also noticed a sharp decline in how many of these objects exist once you hit the i-dropout regime around redshift six.

Jocelyn: Does that drop-off suggest that these luminous radio AGNs were actually disappearing as we look further back toward the epoch of reionization?

Subrahmanyan: It's definitely a possibility, though we have to consider if it's a real physical decline or just our instruments struggling to see them.

Vera: That leads us perfectly into how they actually managed to find them without falling into that same trap.

Improvements: Jocelyn: I want to go back to what Vera said about the radio spectra not being as steep as we thought, because that implies the old way of finding these things was broken.

Vera: It wasn't necessarily broken, but it was definitely incomplete.

Jocelyn: Can you explain how they actually changed their selection process to fix this?

Vera: Traditionally, astronomers used "ultra-steep spectrum" selection, looking for sources with a spectral index less than negative one point three.

Jocelyn: So if a source had a flatter spectrum, it would just be ignored by the old searches?

Vera: Exactly, and this paper shows that most of the population they found doesn't even meet that steepness criteria.

Subrahmanyan: Instead of relying on the radio shape alone, they used this "dropout" technique with the Subaru optical data to find them based on their colors.

Jocelyn: So they're using the fact that certain light wavelengths get blocked by gas in the early universe to identify them?

Vera: Yes, they look for these g-, r-, or i-dropouts, which tells them the object is likely at a very high redshift.

Subrahmanyan: By combining that optical dropout method with the VLASS radio detections, they've created a much more complete census of the population.

Jocelyn: It sounds like this methodology allows us to catch all those "flat-spectrum" quasars that were previously invisible to us.

Vera: It really does, and it means we're finally seeing a more diverse group of early black holes.

Subrahmanyan: This shift in methodology is going to force theorists like me to rethink how we model the radio emission from these early active galaxies.

Jocelyn: We've covered a lot of ground, so let's wrap this up and see what the big picture looks like now.

Conclusion: Vera: We've reached the end of our look at "A Wide and Deep Exploration of Radio-detected Active Galactic Nuclei with Subaru HSC (WERGS). XIII. High-z Radio Quasar Selection from HSC--VLASS over about1200 deg two".

Jocelyn: This paper really changes the game for how we hunt for the most distant radio sources in the sky.

Vera: It provides a massive, clean catalog of hundreds of candidates that are perfect targets for future telescopes.

Jocelyn: I'm curious to see when we get the spectroscopic confirmation to prove exactly how far away these four hundred candidates really are.

Subrahmanyan: From my perspective, this is about correcting our cosmic census so we can finally understand the true growth rate of black holes in the early universe.

Vera: We'll definitely be watching for those follow-up studies to see if that decline at redshift six holds up.

Jocelyn: Thanks for joining us, everyone; we'll see you next time when we have another fascinating paper to dissect.

Subrahmanyan: It’s been a pleasure, looking forward to the next one.

Vera: Goodbye for now!

Youwen Kong, Kohei Ichikawa, Hisakazu Uchiyama, Yuxing Zhong, Xiaoyang Chen, Kotaro Kohno, Tohru Nagao, Kianhong Lee, Bovornpratch Vijarnwannaluk, Yoshiki Matsuoka, Yoshiki Toba, Itsna Khoirul Fitriana, Sakiko Obuchi, Yuta Ishikawa, Victor Kadri

Institute of Astronomy, Graduate School of Science, The University of Tokyo · Frontier Research Institute for Interdisciplinary Sciences, Tohoku University · Astronomical Institute, Tohoku University · Department of Advanced Sciences, Faculty of Science and Engineering, Hosei University · National Astronomical Observatory of Japan · Department of Physics, School of Advanced Science and Engineering, Faculty of Science and Engineering, Waseda University · Research Center for the Early Universe, Graduate School of Science, The University of Tokyo · ILANCE, CNRS – University of Tokyo International Research Laboratory · Research Center for Space and Cosmic Evolution, Ehime University · Department of Physics, Graduate School of Science, Nagoya University · Academia Sinica Institute of Astronomy and Astrophysics · National Taiwan University · Department of Physical Sciences, Ritsumeikan University · Astronomy Research Group, Institut Teknologi Bandung · Ecole Polytechnique

astro-ph.GA

Submitted: 2026-03-30

Updated: 2026-09-11

Comments: Accepted for publication in ApJS

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 3/100

The gist: This paper presents a large-area survey of high-redshift radio AGN candidates using the Subaru Hyper Suprime-Cam (HSC) and the Very Large Array Sky Survey (VLASS).

Key concepts

High-redshift quasars
These are extremely distant and luminous objects powered by supermassive black holes. Studying them helps astronomers understand how these black holes grew so quickly and how their energy output influenced the early universe during its first few billion years.
Dropout technique
This method identifies distant objects by looking for specific light wavelengths blocked by gas in the early universe. By observing "dropouts" in certain optical bands, such as g, r, or i, astronomers can identify objects likely located at very high redshifts.
Radio spectrum
This describes how an object's brightness changes across different radio frequencies. While previous searches focused on "ultra-steep" spectra, this study found many high-redshift candidates have flatter or only moderately steep spectra, revealing a more diverse population than previously understood.

Terminology

Summary

This paper presents a large-area survey of high-redshift radio AGN candidates using the Subaru Hyper Suprime-Cam (HSC) and the Very Large Array Sky Survey (VLASS). By moving beyond traditional selection methods, the study provides a more complete census of the high-redshift RLAGN population, which is essential for understanding jet-driven feedback and supermassive black hole growth in the early Universe.

How it works

The researchers performed positional crossmatching within 1.′′5 between the VLASS Epoch 2 catalog and the HSC–SSP Wide-layer catalog (i ≲ 26) over an area of approximately 1200 deg2. To construct a clean parent sample, they applied several quality cuts to the radio sources, including selecting only compact radio sources and requiring specific astrometric accuracy. The criteria for the VLASS catalog included:

** E RA and E DEC < 2.′′5; 2. Duplicate flag 30′′; and 6. P sidelobe < 0.1 to exclude sidelobe artifact[s]. **

The final candidates were identified using the Lyman-break (dropout) technique based on grizy band colors to target specific redshift regimes, resulting in a final catalog of 305 g-dropouts (z ∼ 4), 42 r-dropouts (z ∼ 5), 3 i-dropouts (z ∼ 6), and 2 z-dropouts (z ∼ 7). This selection step ensures reliable optical identifications for sources with larger positional offsets.

Radio and optical properties

The study finds that the selected candidates are predominantly HSC-level sources, meaning they are optically faint and cluster at iAB ≃ 24–26. This characteristic makes them largely inaccessible to shallower surveys such as SDSS. Regarding radio emission, the spectral analysis reveals that most sources exhibit flat to moderately steep radio spectra (−1 ≲ α ≲ 0), with some showing turnover radio spectra. This finding demonstrates that "conventional ultra-steep-spectrum (USS; α < −1.3) selection would miss most of the population selected in this study," as many high-redshift sources do not follow the canonical USS criterion. Furthermore, nearly all candidates are strongly radio-loud, with typical values of log Robs ≳ 3.

AGN and host galaxy characterization

Using CIGALE SED fitting with a Type 1 AGN-dominated SED scenario, the researchers derived bolometric luminosities for the subset of sources with sufficient multiwavelength photometry. The results show that these objects populate the luminous quasar regime, with typical values of log(Lbol /erg s−1) ∼ 46–47. These findings indicate that the sample is consistent with tracing the luminous end of the quasar luminosity function and represents a significant population of powerful AGNs at z ≳ 4.

Cosmological implications

By examining the comoving number density, the study identifies a sharp decline around the i-dropout regime (z ∼ 6). This trend suggests a possible disappearance of luminous radio AGNs toward the epoch of reionization. The authors note that this decline could be driven by intrinsic evolution or by enhanced inverse-Compton cooling off the cosmic microwave background (“CMB quenching”), which suppresses synchrotron emission from extended radio structures.

Improvements for AI systems

  1. Improvement: Multi-Modal Cross-Domain Feature Fusion with Resolution Sensitivity

Capability: The improved AI system can reliably associate entities across heterogeneous datasets (e.g., radio, optical, and near-infrared) by learning joint representations that account for vastly different angular resolutions, signal-to-noise ratios, and positional uncertainties.

  1. Improvement: Physics-Informed Anomaly Detection via Manifold/Color Constraints

Capability: The system can identify rare needle in a haystack objects within massive datasets by moving beyond simple statistical outlier detection to identifying entities that satisfy specific domain-specific feature manifolds (e.g., dropout selection logic), significantly reducing false positives from noise.

  1. Improvement: Artifact-Aware Data Cleaning and Validation Layers

Capability: The system can automatically filter complex signal contamination (such as sidelobes, duplicates, or transient artifacts in coadded data) by implementing morphological checks and quality-flagging protocols, ensuring high-integrity training sets for high-stakes classification.

  1. Improvement: Probabilistic Regression with Censored Data Handling

Capability: The system can perform highly accurate regression to estimate latent physical properties (e.g., redshift or luminosity) from multi-wavelength inputs while explicitly incorporating non-detections as censored data (upper limits) rather than treating them as zeros or missing values.

Abstract

We report the results of g-, r-, i-, and z- dropout selections based on optical identifications of Very Large Array Sky Survey (VLASS) radio sources using the Hyper Suprime-Cam Subaru Strategic Program survey (HSC--SSP). By positional crossmatching within 1.''5 between the VLASS Epoch 2 catalog and the HSC--SSP Wide-layer catalog (i 26), we obtain about 400 high-redshift radio AGN candidates at z 4 over a about1200 deg squared survey footprint, extending optimistically to z about 7. Their i-band AB magnitudes range from i AB 24 --26, placing most below the SDSS detection limit. By further crossmatching the HSC--VLASS dropout catalog with VLA Faint Images of the Radio Sky at Twenty-centimeters (FIRST) at 1.4 GHz, the LOFAR Two-metre Sky Survey (LoTSS) at 144 MHz, and the TIFR GMRT Sky Survey (TGSS) at 150 MHz, we obtain multi-frequency radio measurements for the majority of the sample. Approximately 85% of the sources have three radio detections. The majority of the high- z candidates show flat to moderately steep radio spectra (-1 α 0, with f ν proportional to ν α), while some also exhibit turnover radio spectra, demonstrating a broad diversity of radio spectral properties among the dropout-selected candidates. SED fitting yields typical AGN bolometric luminosities of (L bol/ erg,s-1) about46.5 --47.5. We also find that the comoving number density declines from the g-dropout to the z-dropout regimes, with substantial decreases toward higher redshift, suggesting fewer luminous radio AGN candidates toward reionization.

Sources

Related papers