🔭 astrophysics

Catalogs of optically-selected clusters and photometric luminous red galaxies from the Hyper Suprime-Cam Subaru Strategic Program final year dataset

This work presents comprehensive catalogs of over 10,000 optically selected galaxy clusters and approximately 6 million luminous red galaxies derived from last year's Hyper Suprime-Cam Subaru Strategic Program dataset, and demonstrates precise photometric redshifts as well as the use of stacked weak lensing signals to refine mass-richness constraints.

Original authors: Masamune Oguri, Yen-Ting Lin, Nobuhiro Okabe, Naomi Ota, I-Non Chiu, Momoka Fujikawa, Hung-Yu Jian, Tadayuki Kodama, Lihwai Lin, Atsushi J. Nishizawa, Rhythm Shimakawa, Yoshiki Toba

Published 2026-05-06

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Masamune Oguri, Yen-Ting Lin, Nobuhiro Okabe, Naomi Ota, I-Non Chiu, Momoka Fujikawa, Hung-Yu Jian, Tadayuki Kodama, Lihwai Lin, Atsushi J. Nishizawa, Rhythm Shimakawa, Yoshiki Toba

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 vast, cosmic city. In this city, the most massive buildings are galaxy clusters – enormous groups of galaxies held together by gravity. Since these clusters consist largely of invisible "dark matter," we cannot see them directly. Instead, astronomers must find them by looking for concentrations of red galaxies, much like recognizing a lively party by a crowd wearing red shirts.

This article is essentially a massive address book and inventory for these cosmic parties, created using data from the Hyper Suprime-Cam (HSC), a powerful camera on the Subaru Telescope in Hawaii. The team led by Masamune Oguri has updated its list with last year's data, covering a huge patch of sky (approximately 1,200 square degrees, which is like viewing 6,000 full moons).

Here is a breakdown of what they found, using simple analogies:

1. The "Party Guest" List (The Cluster Catalog)

The researchers used an intelligent computer program called CAMIRA to scan the sky. Think of CAMIRA as a very sharp-eyed bouncer who knows exactly what a "red galaxy" looks like.

The Count: They found over 10,000 galaxy clusters.
The Range: These clusters are distributed across time (redshift), from relatively nearby to very distant (up to a redshift of 1.38). To put this in perspective: observing a redshift of 1.38 is like looking back into the past when the universe was much younger.
The "Richness": They measured how "rich" each cluster is. In this context, richness has nothing to do with money; it is a count of how many massive galaxies are in the group. They listed only clusters with at least 15 massive members.
The Accuracy: The team is very confident about where these clusters lie in time. Their distance estimates (redshifts) are incredibly precise, with an error margin of less than 1% for most clusters. It is like guessing someone's age within a few months, even from very far away.

2. The "Scale" (Weak Gravitational Lensing)

Since we cannot weigh these clusters on a scale, astronomers used a trick called weak gravitational lensing.

The Analogy: Imagine placing a heavy bowling ball (the galaxy cluster) on a trampoline. If you roll marbles (light from background galaxies) past the bowling ball, their paths will curve slightly. By measuring how much the background light is bent, you can determine how heavy the bowling ball is.
The Result: They used this method to weigh the clusters and update the relationship between a cluster's "richness" (how many members it has) and its actual "mass" (how heavy it is). They confirmed that richer clusters are indeed heavier and refined the mathematics connecting the two.

3. The "Party Center" Problem (Miscentering)

Sometimes the bouncer (CAMIRA) picks the wrong person as the "host" of the party. The brightest galaxy in the cluster (the host) is not always exactly at the center of the dark matter.

The Check: They compared their optical clusters with X-ray clusters (which act like thermal cameras, seeing the hot gas at the cluster's center).
The Finding: They discovered that in about 59% of cases, their optical method correctly identified the center. In the other 41% of cases, they were slightly off-center. This is important because when trying to weigh a cluster, you must know exactly where the center is to get an accurate measurement.

4. The "Red Galaxy" Directory (Photometric LRGs)

While searching for clusters, the computer also created a separate list of Luminous Red Galaxies (LRGs). These are the individual "red-shirted" galaxies that make up the clusters.

The Count: They cataloged about 6 million of these galaxies.
The Range: These extend up to a redshift of 1.25.
The Use: Just like the cluster list, the distance estimates for these individual galaxies are very accurate (with an error of about 2% for the average distance range). This list is a byproduct of the cluster-finding process but represents a valuable treasure for studying how these specific galaxy types evolve over time.

Summary

In short, this paper offers the astronomical community two massive, high-quality tools:

A list of over 10,000 galaxy clusters with precise distances and updated weight estimates.
A list of 6 million red galaxies with precise distances.

These tools are based on the deepest images ever taken of this specific patch of sky and enable scientists to study the structure of the universe and the nature of gravity with much greater precision than before. The data are now public, ready for other scientists to use for their own cosmic investigations.

Technical Summary: Catalogs of Optically Selected Galaxy Clusters and Photometric LRGs from the Final-Year Dataset of HSC-SSP

Problem Statement and Context
Galaxy clusters serve as critical testbeds for theories of gravity and cosmology, where their abundance and internal structure provide constraints on cosmological parameters and insights into galaxy evolution. While traditional identification methods rely on X-ray emission or the Sunyaev-Zel'dovich effect, optical selection based on the spatial concentration of red galaxies offers a powerful alternative, particularly when combined with deep, multi-wavelength imaging. Previous efforts using the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP) have established catalogs of galaxy clusters (e.g., Oguri et al. 2018a), yet the survey has since progressed to its final-year dataset (S23B), covering approximately 1200 deg². This contribution addresses the need to create and characterize updated, larger catalogs of optically selected galaxy clusters and photometric luminous red galaxies (LRGs) from this final dataset to enable more precise cosmological and astrophysical studies.

Methodology
The authors utilize the final-year dataset of HSC-SSP (S23B), which includes the Wide, Deep, and Ultradeep layers. The overarching galaxy sample is selected from the Wide layer (covering ~1220 deg² with masking of bright stars) and the Deep+UD layers, applying strict quality flags (e.g., removal of edge effects, interpolation, and cosmic rays) as well as magnitude limits ( $z < 24$ ).

Cluster Search (CAMIRA Algorithm): The study employs the CAMIRA algorithm (Oguri 2014), which fits photometric galaxies to a stellar population synthesis model by Bruzual & Charlot (2003). The model is calibrated using a large spectroscopic galaxy sample compiled from various surveys (e.g., DESI DR1, SDSS DR15, zCOSMOS) to refine the colors of the red sequence. The algorithm constructs a 3D richness map by summing the probabilities that galaxies are members of the red sequence with stellar masses $M_* > \sim 10^{10.2} M_\odot$ within a physical radius of $R < \sim 1 h^{-1}$ Mpc. Brightest cluster galaxies (BCGs) are identified near richness peaks, and richness ( $\hat{N}_{mem}$ ) as well as photometric redshifts ( $z_{cl}$ ) are recalculated relative to these centers.
Construction of Photometric LRGs: As a byproduct of the CAMIRA fitting process, a catalog of photometric LRGs is generated. All photometric galaxies are fitted with the same stellar population synthesis model, and those with a best-fit chi-squared $\chi^2 < 10$ are selected as LRGs.
Validation and Analysis:
- Redshift Accuracy: The photometric redshifts of clusters and LRGs are validated against spectroscopic redshifts of BCGs and matched spectroscopic galaxies, respectively.
- Miscentering: Positional offsets between CAMIRA cluster centers and X-ray centers (from the eRASS1 catalog) are analyzed to quantify miscentering effects.
- Mass-Richness Relation: Stacked weak gravitational lensing signals ( $\Delta\Sigma$ ) are measured using the three-year HSC-SSP shear catalog. The signals are modeled using a conditional probability distribution of richness given halo mass, incorporating miscentering effects (via a two-component model) and two-halo terms to constrain the mass-richness relation.

Main Contributions and Results

Cluster Catalogs: The authors present catalogs containing over 10,000 clusters with a richness of $\hat{N}_{mem} > 15$ in the redshift range $0.1 \le z_{cl} \le 1.38$ . Approximately 3,000 of these clusters are located at $z \ge 0.8$ . The total number of probable red member galaxies exceeds $6 \times 10^5$ .
- Redshift Accuracy: The photometric redshifts of the clusters prove to be precise, with a scatter of $\sigma_z \approx 0.01$ for $z < 1$ . A degradation in accuracy is observed at $z > 1$ , attributed to fewer spectroscopic calibrators and the absence of near-infrared bands.
- Miscentering: Comparison with X-ray clusters from eRASS1 shows that the fraction of miscentered clusters is $f_{cen} = 0.590 \pm 0.035$ , consistent with previous results.
- Mass-Richness Relation: Using stacked weak lensing, the study updates constraints on the mass-richness relation. The analysis confirms that a richness of $\hat{N}_{mem} = 20$ corresponds to a halo mass of $M_{200m} \sim 10^{14} h^{-1} M_\odot$ . The derived parameters largely agree with previous works (Murata et al. 2019) but, in this iteration, rely exclusively on lensing constraints.
Photometric LRG Catalogs: The study releases a catalog of approximately 6.7 million photometric LRGs from the Wide layer (plus ~120,000 from Deep+UD) in the redshift range $0.05 \le z_{LRG} \le 1.25$ and stellar mass range $M_* \ge 10^{10.3} M_\odot$ .
- Redshift Performance: For the redshift range $0.4 < z_{LRG} < 1.0$ , the scatter in photometric redshift is $\sigma_z < 0.02$ . The scatter is slightly larger than that of cluster redshifts due to the lack of averaging over multiple member galaxies.
- Completeness: The sample is nearly complete up to $z \sim 1.1$ , with incompleteness arising at higher redshifts due to magnitude limits affecting lower-mass LRGs.

Significance
The authors claim that these catalogs represent a significant advancement in studies of optical clusters and LRGs, offering approximately five times more objects than the previous first-year catalogs (S16A). The extended redshift reach (up to $z \sim 1.38$ for clusters and $z \sim 1.25$ for LRGs) surpasses the capabilities of earlier surveys such as DES, KiDS, and DESI Legacy Imaging Surveys, serving as a "springboard" for future Euclid cluster catalogs. The authors emphasize that the high accuracy of the photometric redshifts and the large sample size enable more precise studies of galaxy evolution and cosmology, including the calibration of redshifts for weak lensing sources and the investigation of density profiles of quiescent galaxies at the center. The catalogs are made publicly available to facilitate these applications.

1. The "Party Guest" List (The Cluster Catalog)

2. The "Scale" (Weak Gravitational Lensing)

3. The "Party Center" Problem (Miscentering)

4. The "Red Galaxy" Directory (Photometric LRGs)

Summary

More like this