🔭 astrophysics

How well is the local Large Scale Structure of the Universe known? CosmicFlows vs. Biteau's Galaxy Catalog with Cloning

This paper compares the CosmicFlows density field models with J. Biteau's cloned galaxy catalog, finding that while Biteau's approach is preferred for the Local Volume and non-obscured regions, it relies on fictitious data within the Zone of Avoidance and suffers from radial uncertainties, whereas CosmicFlows reveals significant large-scale structures missed by catalogs but occasionally misaligns angular positions.

Original authors: Yifei Li, Glennys R. Farrar

Published 2026-02-04

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Yifei Li, Glennys R. Farrar

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

The Big Picture: Mapping the Neighborhood

Imagine the universe as a giant, dark city. We live in one specific apartment (the Milky Way galaxy), and we want to draw a complete map of the entire neighborhood to understand how the city is built. This map is crucial for two main reasons mentioned in the paper:

Cosmic Rays: These are like high-speed "bullets" (particles) flying through space. To know where they are coming from, we need to know where the "buildings" (galaxies) are that might be shooting them.
Dark Matter: We want to know where the invisible "ghost matter" is hiding, which holds the city together.

The paper compares two different teams trying to draw this map of our local universe (within about 300 million light-years).

The Two Map-Makers

Team 1: The "Motion Detectives" (CosmicFlows)

How they work: They don't just look at where galaxies are; they look at how they are moving. Imagine you are in a crowd of people. Even if you can't see everyone clearly, if you see a group of people rushing toward a specific spot, you can guess there is a heavy object (like a magnet) pulling them there.
The Tool: They use a sophisticated computer model that calculates gravity and movement to fill in the blanks.
The Catch: Because they rely on movement, their map is a bit "smoothed out." It's like a weather map showing a general storm front rather than every single raindrop. They are also unsure about the exact location of things because the "wind" (peculiar velocities) can push galaxies off their true paths.

Team 2: The "Direct Photographers" (Biteau's Catalog)

How they work: They take a direct inventory. They look at a catalog of galaxies, measure how bright they are, and calculate their mass based on that. It's like counting every car in a parking lot and weighing them based on their size.
The Tool: A massive list of 489,000 galaxies.
The Catch: There are two big problems with this method:
1. The "Foggy Window" (Zone of Avoidance): Our view of the universe is blocked by the dusty center of our own galaxy (the Milky Way). It's like trying to take a photo of a city through a dirty, foggy window. You can't see the buildings behind the glass.
2. The "Speed Trap" (Distance Errors): To know how far away a galaxy is, this team mostly assumes it is moving away from us at a steady speed (the Hubble flow). But galaxies also have their own "side-to-side" movements. If a galaxy is rushing toward us, the team might think it's much farther away than it really is, or vice versa.

The "Cloning" Trick and Why It's Risky

The biggest issue with Team 2 (Biteau) is how they handle the "Foggy Window" (the Zone of Avoidance). Since they can't see the galaxies behind the dust, they use a trick called "Cloning."

The Analogy: Imagine you are looking at a wall with a hole in it. You can see the wallpaper pattern on the top and bottom of the hole, but not the middle. To fill the hole, you take a piece of the wallpaper from the top, flip it upside down, and paste it into the hole. You do the same with the bottom piece.
The Result: You have a complete wall, but the pattern in the middle is fake. It's a mirror image of what's above and below.
The Paper's Finding: The authors found that this "cloning" creates entirely fake structures in the map. In some areas, it doubles the amount of matter that isn't actually there. It's like drawing a fake building in the middle of a park just because you couldn't see the real one behind the trees.

What They Found When Comparing the Maps

1. The "Local" Neighborhood (Within 11 Million Light-Years)

Winner: Team 2 (Biteau).
Why: In our immediate backyard, we have very accurate measurements of galaxy distances (using techniques like Cepheid stars). Here, the "Direct Photographers" have a better, more detailed map than the "Motion Detectives." They found that our local neighborhood is actually denser (has more stuff) than the Motion Detectives' smooth model suggested.

2. The "Foggy" Center (The Zone of Avoidance)

Winner: Team 1 (CosmicFlows).
Why: Because Team 2's "cloning" creates fake structures, their map is unreliable in the dusty center of our galaxy. Team 1's method of using gravity and motion is the only way to get a decent guess of what's hiding behind the dust, even if it's not perfect.

3. The "Far" Neighborhood (Beyond 11 Million Light-Years)

The Conflict:
- Team 2 sees massive clumps of galaxies (like the Pavo-Indus group) that look very dense.
- Team 1 sees these same areas as much more spread out and less dense.
The Reason: Team 2 is likely making distance errors. Because they assume galaxies are moving in a straight line away from us, the "side-to-side" movements of galaxies in clusters make them look like they are spread out over a huge distance range (a "finger" pointing at us). Team 1's model corrects for this, showing the galaxies are actually closer together.

4. The "South Pole Wall" Surprise

Team 1 (CosmicFlows) discovered a massive wall of galaxies called the "South Pole Wall."
Team 2 (Biteau) does not see this wall in their catalog.
Why? The wall is likely hidden behind the dusty "Foggy Window" of our galaxy. This proves that even outside the central fog, dust can hide huge structures that a simple galaxy catalog will miss.

The Bottom Line

If you want to know what's right next door: Trust the "Direct Photographers" (Biteau). Their map is more detailed and accurate for the immediate neighborhood.
If you want to know what's hiding behind the dust: Trust the "Motion Detectives" (CosmicFlows). The "cloning" trick used by the other team creates fake buildings that don't exist.
If you want to know the true shape of distant clusters: The "Motion Detectives" are likely more accurate because they account for the messy, side-to-side movements of galaxies that confuse the "Direct Photographers."

The paper concludes that neither map is perfect. The best approach for the future is to combine the two: use the detailed local data from the catalogs, but use the motion-based physics to fill in the gaps where dust hides the view and where distance measurements get tricky.

Technical Summary: How well is the local Large Scale Structure of the Universe known? CosmicFlows vs. Biteau's Galaxy Catalog with Cloning

Problem Statement
Accurate knowledge of the local matter density distribution is critical for various cosmological applications, including modeling ultrahigh energy cosmic ray (UHECR) sources and predicting diffuse Dark Matter annihilation signals. Two primary methodologies exist for reconstructing this local Large Scale Structure (LSS): (1) inferring density fields from observed galaxy peculiar velocities and redshift-independent positions (the CosmicFlows approach), and (2) constructing catalogs of galaxies with distance estimates and filling in unobserved regions (the Biteau approach). While recent works have utilized these models to explain UHECR dipole anisotropies, a quantitative comparison of their fidelity, particularly regarding the "Zone of Avoidance" (ZoA) obscured by the Milky Way and the radial distribution of mass, has been lacking. This paper addresses the discrepancies between these models to determine which provides a more reliable baseline for local mass distribution.

Methodology
The authors perform a quantitative comparison between three density models:

CosmicFlows2 (CF2): A quasi-linear density field derived from ~8,000 galaxies using the Wiener Filter and constrained realizations (Hoffman et al. 2018).
CosmicFlows4 (CF4): A state-of-the-art reconstruction using the HAMLET algorithm and Hamiltonian Monte Carlo sampling on a sample of 55,877 galaxies (Valade 2026, in preparation).
Biteau's Catalog (B21): A semi-empirical catalog of ~489,000 galaxies within 350 Mpc (Biteau 2021). B21 assigns masses based on luminosity functions and employs a "cloning" technique to fill the ZoA ( $|b| < 20^\circ$ for $|\ell| < 30^\circ$ ) by mirroring galaxies across the ZoA boundary.

To enable comparison, the authors voxelize 3D space into spherical shells ( $\Delta d \approx 1.43$ Mpc) and Healpix pixels. They apply smoothing to the Biteau catalog to match the resolution of the CosmicFlows fields, utilizing Gaussian profiles for radial smoothing ( $R_s = 5$ and $10$ Mpc) and specific angular smoothing schemes. For the local volume ( $<10$ Mpc), a 2D nested pixel smoothing is used to avoid unphysical mass redistribution to negative distances. The comparison involves analyzing mean over-densities as a function of distance and generating skymaps of density fluctuations in various radial shells.

Key Contributions and Results

Local Volume ( $<11$ Mpc): Within 11 Mpc, the Biteau model (derived directly from the Local Volume catalog of Karachentsev et al. 2018) is found to be superior to CosmicFlows. It reveals a local overdensity approximately twice as large as that inferred by CF4. The authors attribute the CF4 underestimation to the lack of a prior placing the observer inside a Milky Way-like galaxy; CF4 reconstructs mass based on relative velocities without biasing for the observer's location in a dense local environment.
The Zone of Avoidance (ZoA) and Cloning Artifacts: The study identifies that the "cloning" procedure used in B21 to fill the ZoA creates entirely fictitious large-scale structures. In regions toward the Galactic center, cloning mirrors structures from $|b| \approx 20^\circ$ into the obscured zone, artificially doubling the overdensity in certain shells (e.g., at 85 Mpc). Consequently, the B21 density field in the ZoA is considered unreliable except within the 11 Mpc Local Volume where direct observations exist.
Radial Distribution and "Finger-of-God" Effects: At larger distances, the B21 model exhibits a less robust radial mass distribution compared to CF4. Because B21 largely assigns distances based on redshift ( $d = cz/H_0$ ), it fails to account for significant peculiar velocities (infall velocities in massive clusters can exceed 1000 km/s). This leads to "finger-of-God" artifacts where mass is spread radially over incorrect distance shells, whereas CF4's self-consistent modeling of peculiar velocities provides a more concentrated and physically motivated radial distribution.
Structural Discrepancies: While CF2 and CF4 show similar large-scale structures, CF4 reveals more detailed sub-structures (e.g., Antlia, NGC 5846) and larger variances. Notably, the authors find unexpected misalignments in the angular positions of structures in CF4 compared to the galaxy catalog in non-ZoA regions (e.g., the Pavo-Indus region), where B21 shows a significant overdensity that CF4 does not identify. The authors suggest this may be due to incorrect redshift-based distance assignments in B21 or a failure of CF4 to capture specific mass concentrations, though the exact cause remains a mystery.
The South Pole Wall: The paper highlights that the South Pole Wall, discovered by CosmicFlows, is absent in galaxy catalogs, demonstrating that obscuration can be a problem even outside the strict ZoA boundaries.

Significance and Conclusions
The paper concludes that the choice of model depends on the specific application and distance scale:

For the Local Volume ( $<11$ Mpc): The direct catalog approach (B21/LV) is preferred due to its robust distance measurements and ability to capture the true local overdensity.
For the Zone of Avoidance: The CosmicFlows approach (CF4) is currently the most reliable, as it relies on the self-consistency of large-scale motions rather than fictitious cloning.
For General Large-Scale Structure: While B21 provides good angular information where obscuration is minimal, its radial distribution is compromised by peculiar velocity effects. Conversely, CF4 provides a physically consistent mass distribution but may occasionally misalign angular structures or underestimate local overdensities due to modeling priors.

The authors emphasize that for UHECR studies, the sensitivity to source distance and the Galactic magnetic field means that the radial distribution errors in B21 and the artificial structures from cloning could significantly impact predictions of the UHECR dipole magnitude. They suggest that future models should ideally combine the physical priors of CosmicFlows with robust distance measurements from galaxy catalogs, and that the "cloning" method for the ZoA should be replaced by techniques that statistically reweight observed galaxies based on dust and gas modeling.

The Big Picture: Mapping the Neighborhood

The Two Map-Makers

The "Cloning" Trick and Why It's Risky

What They Found When Comparing the Maps

The Bottom Line

More like this