The Galaxy Bias Profile of Cosmic Voids:A Comparison of Void Finders

This study compares five distinct void-finding algorithms applied to the IllustrisTNG simulation to demonstrate that while the radial gradient of individual galaxy bias within cosmic voids is a robust feature, the specific selection of anti-biased galaxies and contamination by high-bias boundary galaxies depend significantly on the adopted void definition and density thresholds.

The Gist

Imagine the Universe not as a smooth, empty space, but as a giant, three-dimensional sponge. Most of the sponge is made of "stuff" (galaxies, gas, and dark matter), but there are huge, hollow holes running through it. These holes are called cosmic voids. They are the emptiest places in the Universe.

This paper is like a team of detectives trying to figure out how to measure the "emptiness" of these holes and how the few galaxies living inside them behave.

The Big Question: How Do We Measure a Hole?

The problem is that there is no single, universal rule for what counts as a "void." Different scientists use different tools (algorithms) to find them:

• The "Spherical" Tool: This assumes every hole is a perfect ball, like a marble.
• The "Watershed" Tool: This treats the Universe like a landscape of hills and valleys. A void is a valley where water would pool. This method finds weird, jagged shapes, not just balls.
• The "Popcorn" Tool: This is a hybrid. It starts with balls but merges them together if they overlap, creating a free-form shape that looks more like a popped piece of popcorn than a perfect sphere.

The authors wanted to know: Does it matter which tool we use? If we use a "ball" finder versus a "jagged valley" finder, do we get different results about the galaxies living inside?

The Main Character: "Bias"

To understand the galaxies, the authors looked at something called "bias."
Think of bias as a measure of how much a galaxy "likes" to hang out with other galaxies.

• High Bias: A galaxy that loves crowds. It only shows up where there are already lots of other galaxies (like a party animal at a concert).
• Low (or Negative) Bias: A galaxy that is a loner. It actually prefers to be alone and avoids the crowds. In the deepest, emptiest parts of a void, galaxies can have "negative bias," meaning they are less clustered than the invisible dark matter holding the universe together. They are the ultimate introverts of the cosmos.

The Experiment

The researchers used a super-computer simulation of the Universe (called IllustrisTNG) to create a fake universe with billions of galaxies. They then ran five different "void-finding" tools on this same fake universe to see what they found.

What they found:

1. The "Introvert" Trend is Real: No matter which tool they used, they found that galaxies deep inside the voids tend to be "introverts" (they have negative bias). They are less clustered than the average galaxy.
2. The Shape of the Trend: As you move from the very center of a void out toward its edge, the galaxies change.
   • In the center: The galaxies are the biggest "loners" (most negative bias).
   • At the edge: As you get closer to the wall of the void (where the "stuff" of the universe begins), the galaxies become less lonely and start acting more like normal galaxies (bias goes up).
   • Analogy: Imagine walking out of a deep, silent cave. In the very center, it's so quiet you can hear a pin drop (extreme negative bias). As you walk toward the exit, you start hearing more people talking and noise increases (bias rises).

The Twist: The Tools Matter

While the general trend (loners in the center, less loners at the edge) was the same for all tools, the tools themselves picked up different groups of galaxies.

• The "Strict" Tools (Sparkling & Popcorn): These tools are very picky. They only find the deepest, most empty parts of the voids. Because they are so strict, they mostly find the "super-introvert" galaxies (those with the most negative bias). They are like a bouncer who only lets the quietest people into the VIP section.
• The "Loose" Tools (Zobov & Revolver): These tools are more relaxed. They find the valleys, but they also include the hillsides and the edges of the valleys. Because they are less strict, they accidentally include many galaxies that live near the "walls" of the void. These wall-galaxies are less lonely (higher bias).
  • Result: The "loose" tools made the voids look like they had more "social" galaxies than they actually did deep inside. They diluted the "loner" signal with "party" galaxies from the edges.

The "Popcorn" Winner

The authors found that the Popcorn method was the best at isolating the true "loner" galaxies. Because it merges overlapping spheres, it creates a cleaner boundary that keeps out the "wall" galaxies better than the other methods. It gave the purest picture of the galaxies living in the deepest emptiness.

The Bottom Line

The paper concludes that:

1. Galaxies in voids are unique: They are fundamentally different from galaxies in crowded areas, acting as "anti-clusters."
2. The trend is real: The pattern of galaxies becoming "less lonely" as you move from the center of a void to the edge is a real physical feature, not just a trick of the math.
3. Methodology matters: If you want to study the galaxies in the deepest emptiness, you must use a tool that strictly defines the void's boundaries (like Popcorn or Sparkling). If you use a tool that is too loose, you will mix in galaxies from the edges and miss the true nature of the void's interior.

In short, the Universe has deep, quiet caves where galaxies are very shy. How we choose to draw the map of those caves changes which shy galaxies we see, but the shyness itself is a real, universal trait of the void.

Technical Summary

Technical Summary: The Galaxy Bias Profile of Cosmic Voids: A Comparison of Void Finders

Problem Statement
Cosmic voids serve as unique laboratories for studying galaxy formation in extreme low-density environments and as probes of cosmology. Recent work (Montero-Dorta et al. 2025) established that individual galaxy bias ($b_i$), which quantifies how galaxies trace the underlying dark matter field, exhibits a characteristic radial dependence within spherical voids: galaxies near void centers display systematically lower (often negative) bias values compared to those at the boundaries. However, the extent to which this "void bias profile" depends on the specific methodology used to identify voids has remained unexplored. Given that different algorithms (spherical, watershed, free-form) yield catalogs with varying geometrical and dynamical characteristics, it is critical to determine if the observed bias trends are intrinsic to the physical environment of voids or contingent on the identification method.

Methodology
The authors utilize the $z=0$ snapshot of the IllustrisTNG300-1 cosmological magneto-hydrodynamical simulation. They construct a galaxy sample of central galaxies with stellar masses $\log_{10}(M_*/h^{-1}M_\odot) > 8.33$. Individual galaxy bias ($b_i$) is computed for each galaxy using a Fourier-space method that correlates the galaxy's position with the dark matter density contrast on linear scales ($k \leq 0.2 \, h \, \text{Mpc}^{-1}$).

The core of the study involves applying five distinct void-finding algorithms to this identical galaxy sample to generate comparative catalogs:

1. Sparkling: A spherical void finder based on integrated density contrast thresholds ($\Delta_{\text{lim}} = -0.9$).
2. ZOBOV: A parameter-free, watershed-based method using Voronoi Tessellation Field Estimators (VTFE) without explicit density constraints.
3. Revolver-ZOBOV: A watershed method using VTFE for density estimation, converted to effective spherical representations.
4. Revolver-Voxel: A watershed method using grid-based (particle-mesh) density estimation, also converted to effective spheres.
5. Popcorn: A free-form method that merges overlapping spheres to satisfy a global integrated underdensity threshold ($\Delta_{\text{lim}} = -0.9$), preserving non-spherical topology.

The analysis involves computing the mean individual bias $\langle b_i \rangle$ as a function of physical distance ($R$) and normalized distance ($R/R_{\text{max}}$) from void centers. Additionally, the authors investigate the correlation between $b_i$ values and void membership across different integrated underdensity thresholds ($\Delta_{\text{lim}} = -0.9, -0.8, -0.7$) for density-based methods.

Key Contributions and Results

• Robustness of the Bias Profile: The study confirms that the radial gradient of individual bias within voids—increasing from negative values at the centers to higher values at the boundaries—is a robust feature across most void definitions. This trend holds even when distances are normalized by the characteristic size of the void ($R/R_{\text{max}}$).
• Methodological Dependence on Bias Purity:
  • Density-Threshold Methods (Sparkling, Popcorn): These methods preferentially select galaxies with negative bias ($b_i < 0$). They produce distributions shifted toward lower bias values relative to the full sample, effectively isolating "anti-biased" tracers.
  • Watershed Methods (ZOBOV, Revolver): These methods, particularly ZOBOV without density constraints, include substantial contamination from high-bias galaxies located at void boundaries or in filamentary structures. Consequently, ZOBOV voids show a bias distribution much closer to the global average, with some profiles even showing positive bias near the edges.
• Impact of Underdensity Thresholds: When the integrated underdensity threshold is relaxed (e.g., from $\Delta_{\text{lim}} = -0.9$ to $-0.7$), the fraction of strongly anti-biased galaxies ($b_i \ll 0$) contained within the void catalogs increases. The Popcorn algorithm demonstrates the highest purity in isolating these anti-biased populations, even under relaxed thresholds, due to its ability to reconstruct void boundaries more accurately than spherical approximations.
• Geometric Effects: The Revolver-Voxel catalog exhibits a distinct negative slope in its normalized bias profile. The authors attribute this to a geometric projection effect: because Revolver-Voxel voids are systematically smaller, galaxies near their "walls" often lie deep within the interiors of larger structures identified by other methods, where bias values are still rising.

Significance
The paper claims that the environmental modulation of galaxy bias is a physical reality that is robust across void definitions, provided the methodology is understood. The findings clarify previous ambiguities in the literature regarding environmental dependence: studies using void definitions that do not strictly isolate deep underdensities (like standard watershed methods) are likely to detect weaker or ambiguous trends due to the inclusion of boundary galaxies.

The results suggest that:

1. Individual galaxy bias is a sensitive diagnostic for galaxy-matter coupling in low-density regions.
2. Normalizing distances by void size ($R/R_{\text{max}}$) reveals a universal shape to the bias profile, supporting the use of void size as a fundamental scale for characterizing environmental dependencies.
3. Density-threshold methods (specifically Popcorn) are superior for isolating the deepest underdensities and the associated anti-biased galaxy populations, which is crucial for cosmological tests relying on void statistics and void-galaxy correlations.

The authors conclude that while the radial trend is universal, the amplitude and purity of the signal depend heavily on the void-finder's ability to exclude high-density boundary material. This distinction is essential for interpreting observational data and for improving cosmological parameter constraints derived from large-scale structure surveys.