🔭 astrophysics

4MOST Cosmology Redshift Survey (CRS): Clustering Properties of CRS Bright Galaxy and Luminous Red Galaxy Target Catalogues

This paper validates the target selection, photometry, masking strategies, and redshift distributions for the 4MOST Cosmology Redshift Survey's Bright Galaxy and Luminous Red Galaxy catalogues by demonstrating that their angular clustering properties are robust, uniform across survey depths and hemispheres, and consistent with independent spectroscopic and photometric data, thereby confirming their suitability for precision large-scale structure analyses.

Original authors: Behnood Bandi, Antoine Rocher, Aurélien Verdier, Jon Loveday, Zhuo Chen, Johan Richard, Jean-Paul Kneib, Tom Shanks, Michael J. I. Brown

Published 2026-03-03

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Behnood Bandi, Antoine Rocher, Aurélien Verdier, Jon Loveday, Zhuo Chen, Johan Richard, Jean-Paul Kneib, Tom Shanks, Michael J. I. Brown

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, three-dimensional city made of stars and galaxies. Astronomers want to build a perfect map of this city to understand how it was built, how it's growing, and what invisible forces (like dark energy) are pushing it apart.

To do this, they need a massive survey called 4MOST CRS. Think of 4MOST as a super-powered camera on a telescope that can take "ID photos" (spectra) of 5.4 million galaxies at once. But before the camera can take these photos, the astronomers need a guest list. They need to know exactly which galaxies to invite to the party.

This paper is all about checking that guest list. The authors are asking: "Is our list of guests (the Bright Galaxies and Luminous Red Galaxies) accurate? Are we accidentally inviting streetlights (stars) or excluding important guests? And is the list consistent no matter which part of the sky we look at?"

Here is a breakdown of their work using simple analogies:

1. The Guest List (Target Selection)

The astronomers are looking for two types of "guests":

Bright Galaxies (BG): The loud, energetic party-goers nearby.
Luminous Red Galaxies (LRG): The older, reddish, heavy-set guests that are further away.

They created a list using images from a previous survey (the "Legacy Survey"). But, just like a party planner, they had to worry about obstacles.

The Problem: Sometimes, a bright streetlamp (a nearby star) or a smudge on the camera lens (an artifact) looks like a galaxy. If they don't remove these, the map gets messed up.
The Solution: They created "Veto Masks." Imagine drawing a circle around every streetlamp and saying, "No galaxies allowed inside this circle." They tested different sizes of these circles to make sure they weren't accidentally throwing out real guests while keeping the fake ones out.
- Result: They found the perfect size for the circles. It's like finding the Goldilocks zone: not too big (wasting space) and not too small (leaving in the noise).

2. The "Clustering" Test (Are the guests grouping correctly?)

In the universe, galaxies don't stand alone; they hang out in groups, like friends at a party. This is called clustering.

The Test: The authors measured how often these galaxies appeared near each other.
The Analogy: Imagine you are trying to figure out if a crowd of people is just randomly scattered in a park, or if they are actually forming tight circles of friends.
The Masking Check: They checked if their "Veto Masks" (the circles around streetlamps) were messing up the friendship groups. They found that with the right masks, the "friendship patterns" looked natural and consistent, whether they were looking at the North or South part of the sky.

3. The "Limber Scaling" Test (The Depth Check)

This is a clever trick to check if the survey is fair.

The Analogy: Imagine you have a bucket of marbles. Some are bright red (close galaxies), and some are dim red (far galaxies). If you look at the bucket from the top, the dim ones look fainter.
The Test: The astronomers took their list of galaxies and split them into "slices" based on how bright they looked. They asked: "If we account for the fact that the dim ones are further away, do the 'friendship patterns' of the bright ones and the dim ones look exactly the same?"
The Result: Yes! When they adjusted for distance, all the slices collapsed onto the same pattern. This proved that their list is uniform. They aren't accidentally picking different types of galaxies just because they are brighter or dimmer. The "guest list" is consistent from start to finish.

4. The "Cross-Reference" Check (The ID Check)

Sometimes, you can't tell how far away a galaxy is just by looking at a photo (photometry). You need a "spectroscopic ID" (like a passport check) to know the exact distance.

The Problem: The 4MOST survey hasn't finished taking all the ID photos yet.
The Solution: They used a "helper" survey called DESI (which has already taken ID photos of millions of galaxies). They compared their "guest list" against DESI's "ID database."
The Result: The distribution of distances in their list matched the real ID data almost perfectly. This confirmed that their method for guessing distances (based on color and brightness) is reliable.

5. The "Halo" Model (The Invisible Skeleton)

For the older, red galaxies (LRGs), the authors used a model called HOD (Halo Occupation Distribution).

The Analogy: Imagine galaxies are like furniture in a house. The "Dark Matter Halo" is the invisible house itself. The HOD model asks: "How many pieces of furniture (galaxies) fit inside a house of a certain size?"
The Result: They found that their red galaxies fit into these invisible houses in a way that matches what we expect from physics. It's like checking that the furniture fits the room dimensions.

The Big Picture Conclusion

The authors spent this paper doing quality control. They are essentially saying:

"We have built a massive, high-tech guest list for the universe's biggest party. We have checked for fake IDs, we have made sure the list is fair across the whole sky, and we have confirmed that the guests are grouping together exactly as physics predicts. The list is ready. Now, we can use it to map the universe and solve the mysteries of dark energy."

This paper is the "quality assurance" report that gives scientists the confidence to use this data for major discoveries about how our universe works.

1. Problem Statement

The 4MOST Cosmology Redshift Survey (CRS) aims to obtain 5.4 million spectroscopic redshifts over $\sim5700$ deg $^2$ to map large-scale structure (LSS) and constrain cosmological parameters (dark energy, gravity). To achieve this, it requires a robust, spatially uniform target selection for Bright Galaxies (BG) and Luminous Red Galaxies (LRG) based on photometric data from the DESI Legacy Survey (LS) DR10.1.

The core challenge addressed in this paper is validating the target catalogues. Specifically, the authors must ensure that:

The target selection criteria and masking strategies (removing bright stars, artifacts, etc.) do not introduce artificial clustering or suppress real signals.
The photometric uniformity is consistent across different survey regions (North vs. South Galactic Caps) and depth variations (DECaLS vs. DES).
The derived redshift distributions ( $N(z)$ ) and clustering statistics are reliable for precision cosmology, particularly for cross-correlations with weak lensing surveys (e.g., LSST, Euclid).

2. Methodology

The authors employed a multi-faceted approach combining angular clustering analysis, cross-correlation techniques, and halo occupation modeling:

Data Sources:
- Target Catalogues: BG and LRG targets selected from DESI Legacy Survey DR10.1 (combining DECaLS, BASS, MzLS, and DES).
- Reference Data: DESI Data Release 1 (DR1) spectroscopic data used to validate redshift distributions and model spatial correlations.
- Random Catalogues: Legacy Survey imaging randoms to trace angular selection and masking.
Masking Strategy Evaluation:
- Tested various combinations of Legacy Survey MASKBITS (for bright stars, large galaxies, clusters) and unWISE W1 band artifact masks (diffraction spikes, ghosts, halos).
- Used the Landy-Szalay estimator to compute the angular two-point correlation function, $w(\theta)$ , for different masking configurations.
- Defined stability as the point where further masking changes $\Delta w(\theta)$ are statistically insignificant ( $\lesssim 1\sigma$ ).
Limber Scaling Test (for BG):
- Applied Limber's equation to test the consistency of angular clustering across different $r$ -band magnitude slices.
- Assumed a broken power-law spatial correlation function, $\xi(r)$ , derived from projected correlation measurements, $w_p(r_p)$ , from DESI DR1.
- Scaled $w(\theta)$ curves from different magnitude slices to a reference slice to check if they collapse onto a single power law, verifying photometric uniformity.
Clustering Redshifts (Cluster- $z$ ):
- Estimated the redshift distribution $N(z)$ of photometric BG targets by measuring their angular cross-correlation with the spectroscopic DESI DR1 reference sample.
- Compared these estimates against the direct spectroscopic $N(z)$ from DESI DR1.
LRG Analysis & HOD Modeling:
- Measured angular clustering in photometric redshift slices ( $0.4 \le z < 1.0$ ) for LRGs in both DECaLS and DES footprints.
- Fitted the projected correlation function, $w_p(r_p)$ , using a Halo Occupation Distribution (HOD) model.
- Accounted for photometric redshift errors by applying a "smearing" effect to mock galaxies along the line of sight.

3. Key Contributions

Optimized Masking Protocols: Established the definitive masking configuration for 4MOST-CRS.
- BG: Uses LS MASKBITS 11, 12, and 13 (covering Gaia stars $G < 16$ , large galaxies, and globular clusters).
- LRG: Uses LS MASKBITS 11, 12, 13 plus the full set of unWISE W1 artifact masks to suppress small-scale excess power caused by diffraction spikes and halos.
Validation of Photometric Uniformity: Demonstrated that the BG target selection is uniform across the North and South Galactic Caps and consistent with depth variations via the Limber scaling test.
Redshift Distribution Validation: Validated the BG redshift distributions using clustering-redshift techniques, showing strong agreement with spectroscopic data at fainter magnitudes.
LRG Clustering Characterization: Provided the first detailed angular clustering and HOD analysis for the 4MOST-CRS LRG sample, confirming consistency with recent DESI LRG studies.

4. Key Results

Masking Stability:
- For LRGs, stricter masking (including unWISE artifacts) was essential. Looser masks resulted in negative $\Delta w(\theta)/\sigma$ at small scales, indicating contamination from misclassified stars/artifacts. The final mask reduced small-scale power to a clean power law.
- For BGs, the clustering signal was more stable, but extending masks to Gaia stars with $G < 16$ (Maskbit 11) was necessary to remove residual over-densities around bright stars.
Limber Scaling Test (BG):
- After applying scaling factors derived from Limber's equation, the $w(\theta)$ curves for five different magnitude slices collapsed onto a near-common power law in both NGC and SGC.
- Cross-cap residuals were within $\sim 1\sigma$ , confirming that observed differences were due to redshift distribution ( $N(z)$ ) variations rather than systematic selection effects.
Clustering Redshifts (BG):
- The cluster- $z$ derived $N(z)$ matched the spectroscopic DESI DR1 $N(z)$ shape well for fainter magnitudes ( $r \gtrsim 17.5$ ).
- At bright magnitudes ( $r \lesssim 16.75$ ), higher shot noise led to larger fluctuations, but the overall trend remained consistent.
LRG Angular Clustering:
- Angular clustering in photo- $z$ slices ( $0.4 \le z < 1.0$ ) was well-described by a power law.
- The slope $\gamma$ increased slightly with redshift, while amplitude decreased.
- One bin ( $0.8 < z < 0.9$ ) showed elevated $\chi^2$ at small scales, suggesting minor contamination, but large-scale agreement between DECaLS and DES was excellent ( $\lesssim 1\sigma$ ).
HOD Modeling (LRG):
- Fits yielded a satellite fraction of $\sim 10\%$ and a minimum halo mass for central galaxies ( $M_{cen}$ ) consistent with previous LRG studies.
- The linear bias evolution ( $b_{lin}$ ) followed the expected trend with the growth factor $D(z)$ , though the CRS-LRG sample showed slightly lower bias at high redshift compared to DESI-LRG, consistent with CRS selecting brighter objects.

5. Significance

This paper provides the foundational validation for the 4MOST-CRS Bright Galaxy and Luminous Red Galaxy catalogues. By demonstrating that the target selections and masking strategies yield:

Stable and unbiased clustering statistics ( $w(\theta)$ ),
Uniform photometric properties across the survey footprint, and
Reliable redshift distributions validated against spectroscopy,

the authors confirm that the 4MOST-CRS is ready for precision cosmological analyses. These results are critical for the survey's ability to:

Measure Baryon Acoustic Oscillations (BAO) and Redshift-Space Distortions (RSD).
Perform high-precision cross-correlations with weak lensing surveys (LSST, Euclid) to calibrate photometric redshifts and control galaxy bias systematics.
Constrain dark energy and modified gravity models using the southern sky.

The work establishes a robust pipeline for future 4MOST cosmology papers, ensuring that the massive dataset (5.4 million spectra) will yield scientifically rigorous results.