🔭 astrophysics

The impact of cosmic filaments on the abundance of satellite galaxies

Using the IllustrisTNG simulation, this study reveals that while satellite galaxies are more abundant around central galaxies in cosmic filaments compared to the field, this excess is primarily driven by higher host halo masses and is significantly diminished when accounting for halo mass differences and biases introduced by the choice of filament tracers.

Original authors: Yuxi Meng, Haonan Zheng, Shihong Liao, Lizhi Xie, Lan Wang, Hongxiang Chen, Liang Gao, Quan Guo, Yingjie Jing, Jie Wang, Hang Yang, Guangquan Zeng

Published 2026-04-16

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Yuxi Meng, Haonan Zheng, Shihong Liao, Lizhi Xie, Lan Wang, Hongxiang Chen, Liang Gao, Quan Guo, Yingjie Jing, Jie Wang, Hang Yang, Guangquan Zeng

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 not as a random scattering of stars, but as a giant, three-dimensional spiderweb made of invisible dark matter. This "Cosmic Web" has three main parts:

Knots: The dense, busy intersections where massive galaxy clusters live (like a bustling city center).
Voids: The empty, quiet spaces between the threads (like the open ocean).
Filaments: The long, thin strands connecting the knots (like the roads or bridges connecting cities).

For a long time, astronomers have wondered: Does living on one of these "roads" (a filament) change how many "satellite" galaxies a big central galaxy has?

Think of a central galaxy as a large family living in a big house. The "satellite galaxies" are the smaller, younger family members living in the backyard or nearby sheds. The question is: Do families living on the main highway have bigger backyards with more kids than families living in the quiet countryside?

The Big Discovery

Using a super-computer simulation called IllustrisTNG (which is like a "Universe in a Box" that plays out billions of years of cosmic history in a few weeks), the researchers found that yes, families on the highway do seem to have more kids.

When they looked at galaxies using real star maps (like a census), they found that central galaxies sitting on filaments had roughly 2 to 3 times more satellite galaxies than those in the empty fields. This matched what we see in real life from the Sloan Digital Sky Survey.

The Twist: It's Not Just the Location

Here is where the story gets interesting. The researchers asked: "Is it the road itself that makes the family grow, or is it something else?"

They discovered two main "hidden factors" that explain most of this difference:

1. The Size of the House (Halo Mass)
In the universe, every galaxy sits inside a giant, invisible bubble of dark matter called a "halo." Think of this as the size of the family's estate.

The Finding: Galaxies on filaments tend to live in much bigger estates than galaxies in the field.
The Analogy: It's not that the road makes you have more kids; it's that the rich neighborhoods (filaments) happen to have bigger houses. Big houses naturally have more space for more family members.
The Result: When the researchers mathematically "moved" the field galaxies into big houses to match the filament galaxies, the difference in the number of satellite galaxies dropped by about 80%. The "road" wasn't the main cause; the "house size" was.

2. The Map Maker's Bias (Tracer Choice)
This is the most clever part of the study. To find the "roads" (filaments) in the simulation, you need to use something as a marker.

Method A (Galaxies as Markers): You look at where the stars are and draw lines between them.
Method B (Dark Matter as Markers): You look at the invisible dark matter itself to draw the lines.

The Problem: Galaxies are like "bouncy balls" that tend to clump together. If a galaxy has a huge backyard full of satellites, it looks "brighter" and "denser" to our map-making algorithm. The algorithm gets confused and thinks, "Wow, this must be a major highway!" and draws a road right through it.

The Result: By using galaxies to find the roads, we accidentally over-count the roads in places where there are already lots of satellites. It's like using a crowd of people to find a highway; the algorithm will draw the highway right through the crowd, making it look like the highway caused the crowd.
The Fix: When they used the invisible dark matter (which doesn't care about how many satellites a galaxy has) to draw the roads, the difference between "road" and "field" families vanished by over 70%.

The Simple Takeaway

Yes, galaxies on filaments have more satellites.
But, it's mostly because galaxies on filaments live in bigger dark matter "houses" (halos), and bigger houses naturally hold more satellites.
Also, our maps are a bit biased. When we use galaxies to find the cosmic roads, we accidentally make the roads look like they are full of satellites just because the satellites helped us draw the map in the first place.

In a nutshell: The universe isn't necessarily "magical" about roads making families grow. It's mostly that the rich neighborhoods (filaments) have bigger houses, and our way of drawing the map makes those rich neighborhoods look even richer than they actually are.

1. Problem Statement

Understanding how the large-scale cosmic web environment influences galaxy formation is a fundamental question in astrophysics. While the impact of filaments on host halos and central galaxies (e.g., spin alignment, star formation quenching) is well-studied, the effect on satellite galaxy abundance remains debated.

Observational Tension: Some studies (e.g., Guo et al. 2015, using SDSS data) suggest that central galaxies in filaments host significantly more satellites (by a factor of $\sim$ 2) than those in the field. Conversely, other simulations (e.g., GIMIC) suggest a weak or negligible environmental dependence.
The Core Question: Is the observed excess of satellites in filaments a genuine physical effect driven by the environment (e.g., higher merger rates), or is it an artifact of confounding variables such as host halo mass distributions, central galaxy luminosity distributions, or biases introduced by the methods used to identify filaments?

2. Methodology

The authors utilize the IllustrisTNG100-1 cosmological hydrodynamical simulation (z=0) to investigate these questions with high statistical power and precise physical definitions.

Data Sample:
- Central galaxies are selected based on SDSS $r$ -band absolute magnitude ( $M_{r,cen}$ ) in four bins ($-23$ to $-20$).
- Satellites are defined as subhalos within the virial radius ( $R_{200c}$ ) of the host, containing $>100$ star particles ( $M_{star} \gtrsim 10^8 M_\odot$ ).
- "Knot" environments (clusters, $M_{200c} \ge 10^{13.5} M_\odot$ ) are excluded to isolate filament vs. field effects.
Filament Identification:
- Algorithm: The DisPerSE (Discrete Persistent Structures Extractor) algorithm is used to identify filament spines based on critical points in the density field.
- Tracer Comparison: The study compares two tracer populations:
  1. Galaxy Tracers: Galaxies with $M_{star} \ge 10^9 M_\odot$ (mimicking observational constraints).
  2. Dark Matter (DM) Tracers: A smoothed DM density field (mimicking the true underlying mass distribution).
- Environment Definition: Galaxies within a cylinder of radius $R_{filament} = 1$ Mpc around the filament spine are classified as "filament" galaxies; others are "field" galaxies.
Analysis Techniques:
- Satellite Luminosity Function (LF): Calculated as the number of satellites per central galaxy as a function of magnitude difference ( $M_{diff} = M_{sat} - M_{cen}$ ).
- Resampling (Matching): To isolate specific variables, the authors perform controlled resampling:
  1. Match $M_{r,cen}$ distributions between filament and field samples.
  2. Match host halo mass ( $M_{200c}$ ) distributions in addition to $M_{r,cen}$ .
- Bias Quantification: A controlled test compares "satellite-rich" vs. "satellite-poor" central galaxies to determine if filament-finding algorithms preferentially select one type.

3. Key Contributions

This paper provides a rigorous dissection of the environmental dependence of satellite abundance, distinguishing between physical drivers and methodological biases. Its primary contributions are:

Quantifying the Halo Mass Effect: Demonstrating that the majority of the observed satellite excess in filaments is driven by the fact that filament central galaxies reside in more massive halos than their field counterparts of similar luminosity.
Revealing Tracer Bias: Uncovering a significant systematic bias where using galaxies (rather than dark matter) as tracers for filament identification artificially enhances the measured environmental dependence.
Reconciling Simulations and Observations: Providing a physical explanation for why simulations often show a stronger environmental signal than some hydrodynamical models, and how observational limitations (tracer choice) amplify this signal.

4. Key Results

A. Environmental Dependence (Galaxy-Traced)

When filaments are identified using galaxies (mimicking observations):

Central galaxies in filaments host significantly more satellites than those in the field.
Abundance Ratios ( $\phi_{fil}/\phi_{field}$ ):
- $M_{r,cen} \sim -22$ : 3.49
- $M_{r,cen} \sim -21$ : 2.61
- $M_{r,cen} \sim -20$ : 1.90
This result qualitatively agrees with SDSS observations (Guo et al. 2015).

B. Impact of Distributions (Resampling Analysis)

Magnitude Matching: Resampling to match $M_{r,cen}$ distributions reduces the abundance ratio by only ~5–17%. This indicates that the difference in central galaxy brightness distributions is a minor factor.
Halo Mass Matching: Further resampling to match host halo masses ( $M_{200c}$ $M_{200 c}$ ) drastically reduces the environmental difference.
- The abundance ratios drop to 1.55, 1.34, and 1.43 for the respective bins.
- This represents a reduction of 53–79% of the original excess.
- Conclusion: Differences in host halo mass distributions account for the majority of the observed satellite excess in filaments.

C. Impact of Tracer Choice

Dark Matter Tracers: When filaments are identified using the DM density field, the environmental dependence is significantly weaker.
- Original ratios drop to 1.67, 1.48, and 1.26.
- The difference between galaxy-traced and DM-traced results is a reduction of >70% in the measured environmental effect.
Mechanism of Bias: The study reveals that galaxy-traced algorithms systematically bias filament classification toward "satellite-rich" central galaxies. Because satellite-rich galaxies contribute more to the local galaxy density field, they are more likely to be identified as lying on a filament spine.
- In the galaxy-traced case, the ratio of satellite-rich to satellite-poor centrals in filaments is $\sim 1.3$ .
- In the DM-traced case, this ratio is $\sim 1.0$ (no preference).

5. Significance and Implications

Interpretation of Observations: The study suggests that the strong environmental dependence of satellite abundance seen in observational surveys (like SDSS) is likely an overestimate caused by two factors:
1. The natural correlation between filament environments and higher host halo masses.
2. A methodological bias where galaxy-based filament finders preferentially select galaxies that already have more satellites.
Physical Understanding: Once halo mass and tracer biases are accounted for, the intrinsic environmental effect of filaments on satellite abundance is much smaller (ratios $\sim 1.0-1.5$ ). This implies that while filaments do influence satellite populations (likely via higher merger rates), the effect is not as dominant as raw observational comparisons suggest.
Methodological Guidance: Future studies of the cosmic web must carefully consider the choice of tracers. Using galaxies to define the cosmic web introduces a circular bias when studying galaxy properties within those structures.

In summary, Meng et al. demonstrate that while filaments do host more satellites, the magnitude of this effect is largely driven by the fact that filaments host more massive halos and that the method of identifying filaments using galaxies preferentially selects satellite-rich systems.