pith. sign in

arxiv: 2509.22180 · v2 · submitted 2025-09-26 · 🌌 astro-ph.CO

Detection of HI filament: Pair Stacking vs. Filament Stacking

Yuxi Meng , Jie Wang , Yingjie Jing , Hongxiang Chen , Zerui Liu This is my paper

Pith reviewed 2026-05-18 12:44 UTC · model grok-4.3

classification 🌌 astro-ph.CO
keywords neutral hydrogencosmic filaments21 cm signalstacking techniquesEAGLE simulationIllustrisTNGcosmic webradio intensity mapping
0
0 comments X

The pith

Filament stacking reaches 10^16 to 10^17 cm^{-2} HI column density after masking haloes, unlike pair stacking.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper compares pair stacking and filament stacking for detecting faint 21 cm emission from neutral hydrogen in cosmic filaments, using EAGLE and IllustrisTNG simulations. Pair stacking connects massive knots but suffers heavy contamination from those structures, so the signal falls by orders of magnitude once haloes are masked. Filament stacking instead sums signals along filaments identified directly from galaxy positions and maintains column densities of roughly 10^16 to 10^17 per square centimeter even after every halo is removed. The authors conclude that filament stacking is the more viable route and will improve further with denser galaxy samples and higher-resolution radio maps.

Core claim

Our analysis indicates that, although pair stacking is convenient, it faces contamination from massive structures; after removing this contamination, the filament signal is significantly reduced. In contrast, HI detection via filament stacking appears more promising. The column density in filament stacking reaches ∼10^{16}--10^{17} cm^{-2} even when all haloes are masked, whereas pair stacking does not reach this level even without masking, and is further suppressed by several orders of magnitude once masking is applied.

What carries the argument

Filament stacking that aggregates the 21 cm signal along filaments identified from galaxy distributions, contrasted with pair stacking that connects pairs of massive knots.

If this is right

  • Filament stacking can recover detectable HI signals even after all halo contributions are removed.
  • Higher galaxy number density directly increases the effectiveness of filament stacking.
  • Improved spatial resolution in radio intensity mapping strengthens the filament-stacking signal.
  • Detection of neutral hydrogen in cosmic filaments becomes feasible with upcoming optical and radio surveys.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If filament stacking works in observations, it could provide a direct map of neutral hydrogen along the cosmic web.
  • The same stacking logic could be tested on other faint tracers such as diffuse radio emission or Lyman-alpha.
  • Comparing masked and unmasked signals in real data offers a practical way to validate which stacking method is less biased.

Load-bearing premise

Filaments identified from galaxy distributions in the simulations accurately trace real cosmic filaments without major biases from galaxy selection, identification algorithms, or simulation resolution limits.

What would settle it

A real-sky measurement that finds HI column density below 10^{16} cm^{-2} in masked filaments, or a signal no stronger than masked pair stacking, would falsify the claim that filament stacking is superior.

Figures

Figures reproduced from arXiv: 2509.22180 by Hongxiang Chen, Jie Wang, Yingjie Jing, Yuxi Meng, Zerui Liu.

Figure 1
Figure 1. Figure 1: HI and volume fraction in knots, filaments, and voids in EAGLE (dots) and TNG100 (stars) simulations. The blue points represent knot regions, which are within 5 R200,c of haloes Mh ≥ 103 M⊙/h. The orange points represent filament regions, which are within 1 Mpc/h vicin￾ity of the cosmic filaments that identified with galaxies of M∗ ≥ 108 M⊙/h using DisPerSE. The green points repre￾sent void regions, which … view at source ↗
Figure 2
Figure 2. Figure 2: A demonstration of galaxy pairs (left panel) and filaments identified with galaxy distribution (right panel) in EAGLE simulation. Orange points represent galaxies in both panels, and blue lines represent galaxy pairs (left panel) or cosmic filaments (right panel). For clarity, the galaxy pairs are selected from the most massive galaxies of M∗ ≥ 1011 M⊙/h. The filaments are identified with galaxies with M∗ … view at source ↗
Figure 3
Figure 3. Figure 3: Total, halo contamination and background den￾sity and their dependence on the low limit of halo mass for masking. Curves of different colors represents different pair selections. Blue curves are from all pairs. Orange curves rep￾resents massive pairs with M∗ ≥ 1011 M⊙/h. Green curves are obtained with pairs of separation shorter than 8 Mpc/h. The solid, dashed, dotted, and dotted dashed curves repre￾sent t… view at source ↗
Figure 5
Figure 5. Figure 5: Density profile obtained with filament stacking, with different galaxy number densities (corresponding to dif￾ferent stellar mass thresholds) in EAGLE simulation. Re￾gions within 2×R200,c around halos of Mh ≥ 1011 M⊙/h are masked. Curves from red to blue represent galaxy number densities increasing from the sparsest sample (all galaxies of M∗ ≥ 1011 M⊙/h) to the densest sample (all galaxies of M∗ ≥ 107 M⊙/… view at source ↗
Figure 6
Figure 6. Figure 6: Comparison of two stacking methods. The blue, orange and green curves represents results of filament stack￾ing, obtained with galaxy catalogues of different number den￾sities (set by stellar mass thresholds). In these three curves, solid ones are the peak of the density profile within 20 kpc/h around the spine of filaments, and dashed ones are mean density within 200 kpc/h. The red, purple and brown curves… view at source ↗
read the original abstract

The faint 21 cm signal emitted by neutral hydrogen in cosmic filaments is expected to be detectable. However, due to its weakness, stacking techniques are required. We assessed two stacking methods--pair stacking and filament stacking--using the EAGLE and IllustrisTNG simulations. Pair stacking leverages the fact that cosmic filaments connect massive structures (i.e., knots) in the cosmic web, while filament stacking directly aggregates filaments identified from galaxy distributions. Our analysis indicates that, although pair stacking is convenient, it faces contamination from massive structures; after removing this contamination, the filament signal is significantly reduced. In contrast, HI detection via filament stacking appears more promising. The column density in filament stacking reaches $\sim 10^{16}$--$10^{17}~\mathrm{cm}^{-2}$ even when all haloes are masked, whereas pair stacking does not reach this level even without masking, and is further suppressed by several orders of magnitude once masking is applied. The effectiveness of filament stacking can be further improved with higher galaxy number density and better spatial resolution in radio intensity mapping observations. With the advent of upcoming optical and radio data, the detection of HI in cosmic filaments remains promising.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript compares pair stacking and filament stacking techniques for detecting the faint 21 cm HI signal in cosmic filaments, using the EAGLE and IllustrisTNG hydrodynamical simulations. Pair stacking relies on connections between massive knots, while filament stacking aggregates HI along filaments identified directly from galaxy distributions. The central result is that filament stacking yields column densities of ∼10^{16}–10^{17} cm^{-2} even after masking all haloes, whereas pair stacking is contaminated by massive structures and drops by several orders of magnitude after masking; the authors conclude filament stacking is more promising and improves with higher galaxy density and observational resolution.

Significance. If the quantitative comparison holds after methodological details are supplied, the work would provide a practical guide for HI intensity-mapping analyses with upcoming optical and radio surveys, favoring direct filament identification over pair-based approaches. The deployment of two independent simulation suites offers a modest cross-check, though the absence of robustness tests limits the strength of the conclusions.

major comments (3)
  1. [§2–3 (Methods)] §2–3 (Methods): The filament identification procedure is described only at a high level in the abstract and introduction; no specifics are given on the algorithm (e.g., DisPerSE or equivalent), its parameters, galaxy magnitude or number-density cuts, or any tests against alternative finders or DM-only runs. This is load-bearing for the central claim that filament stacking remains robust after halo masking while pair stacking collapses.
  2. [§4 (Results)] §4 (Results): Quantitative details on the masking procedure, error estimation, and how column densities are computed (including any integration along the line of sight or beam convolution) are not supplied, despite the abstract stating precise numerical outcomes (∼10^{16}–10^{17} cm^{-2}). Without these, the reported difference between the two stacking methods cannot be verified or reproduced.
  3. [§4–5] §4–5: No robustness checks are reported against variations in galaxy selection, simulation resolution, or baryonic physics differences between EAGLE and IllustrisTNG; the abstract notes improvement with higher galaxy density but does not quantify how results change under these variations, which directly affects the claimed superiority of filament stacking.
minor comments (2)
  1. [Abstract] Abstract: The phrase 'all haloes are masked' should be clarified with a brief definition of the halo mass threshold or selection used.
  2. [Figures] Figures: Stacked profiles should explicitly label the masking status and include uncertainty bands or bootstrap errors to allow direct visual comparison of the two methods.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their careful reading of the manuscript and for the constructive comments, which have helped us identify areas where additional clarity is needed. We address each major comment below and will revise the manuscript accordingly to improve reproducibility and strengthen the presentation of our results.

read point-by-point responses

  1. Referee: §2–3 (Methods): The filament identification procedure is described only at a high level in the abstract and introduction; no specifics are given on the algorithm (e.g., DisPerSE or equivalent), its parameters, galaxy magnitude or number-density cuts, or any tests against alternative finders or DM-only runs. This is load-bearing for the central claim that filament stacking remains robust after halo masking while pair stacking collapses.

    Authors: We agree that the filament identification method requires more detailed description to support the central claims. In the revised manuscript we will expand Sections 2 and 3 to specify the exact algorithm (including whether DisPerSE or an equivalent was used), all relevant parameters, the galaxy magnitude and number-density selection cuts, and any validation tests performed against alternative finders or dark-matter-only runs. These additions will allow readers to assess the robustness of the filament-stacking results after halo masking. revision: yes

  2. Referee: §4 (Results): Quantitative details on the masking procedure, error estimation, and how column densities are computed (including any integration along the line of sight or beam convolution) are not supplied, despite the abstract stating precise numerical outcomes (∼10^{16}–10^{17} cm^{-2}). Without these, the reported difference between the two stacking methods cannot be verified or reproduced.

    Authors: We acknowledge that the current manuscript lacks the quantitative methodological details needed for verification. In the revised Section 4 we will provide a full description of the halo-masking procedure, the error estimation technique, and the precise computation of HI column densities, including line-of-sight integration and any beam convolution applied. These additions will substantiate the reported column-density values and the contrast between the two stacking approaches. revision: yes

  3. Referee: §4–5: No robustness checks are reported against variations in galaxy selection, simulation resolution, or baryonic physics differences between EAGLE and IllustrisTNG; the abstract notes improvement with higher galaxy density but does not quantify how results change under these variations, which directly affects the claimed superiority of filament stacking.

    Authors: The referee correctly identifies the absence of systematic robustness tests. While results from two independent simulations are presented, we did not quantify variations with galaxy density, resolution, or baryonic physics differences. In the revision we will add quantitative tests for galaxy number density (as noted qualitatively in the abstract) and discuss consistency between EAGLE and IllustrisTNG. Full resolution-convergence and exhaustive baryonic-physics variations lie beyond the scope of the present study and would require additional simulation suites. revision: partial

Circularity Check

0 steps flagged

No significant circularity in method comparison

full rationale

The paper performs an empirical comparison of pair stacking versus filament stacking on HI signals extracted from the independent EAGLE and IllustrisTNG hydrodynamical simulations. Column-density results are obtained by direct aggregation after applying halo masks; no equations, fitted parameters, or predictions are shown to reduce by construction to the input definitions or to prior self-citations. Filament identification is treated as an external input from galaxy catalogs rather than a self-referential step. The central claim therefore remains independent of the paper's own outputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim depends on the fidelity of two hydrodynamical simulations and the accuracy of filament identification from galaxy catalogs; no free parameters are explicitly fitted in the abstract.

axioms (1)
  • domain assumption EAGLE and IllustrisTNG simulations provide a sufficiently realistic model of neutral hydrogen distribution and filamentary structure in the cosmic web.
    All reported column-density differences rest on the outputs of these specific simulations.

pith-pipeline@v0.9.0 · 5746 in / 1213 out tokens · 48062 ms · 2026-05-18T12:44:07.865208+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

18 extracted references · 18 canonical work pages · 1 internal anchor

  1. [1]

    2005, Astronomy & Astrophysics, 441, 893 Astropy Collaboration, Robitaille, T

    Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013, A&A, 558, A33, doi: 10.1051/0004-6361/201322068 Astropy Collaboration, Price-Whelan, A. M., Sip˝ ocz, B. M., et al. 2018, AJ, 156, 123, doi: 10.3847/1538-3881/aabc4f

    work page doi:10.1051/0004-6361/201322068 2013

  2. [2]

    2016, MNRAS, 457, 2391, doi: 10.1093/mnras/stw142 de Graaff, A., Cai, Y.-C., Heymans, C., & Peacock, J

    Clampitt, J., Miyatake, H., Jain, B., & Takada, M. 2016, MNRAS, 457, 2391, doi: 10.1093/mnras/stw142 de Graaff, A., Cai, Y.-C., Heymans, C., & Peacock, J. A. 2019, A&A, 624, A48, doi: 10.1051/0004-6361/201935159

    work page doi:10.1093/mnras/stw142 2016

  3. [3]

    J., Ruiz-Macias, O., et al

    Hahn, C., Wilson, M. J., Ruiz-Macias, O., et al. 2023, AJ, 165, 253, doi: 10.3847/1538-3881/accff8

    work page doi:10.3847/1538-3881/accff8 2023

  4. [4]

    R., Millman, K

    Harris, C. R., Millman, K. J., van der Walt, S. J., et al. 2020, Nature, 585, 357, doi: 10.1038/s41586-020-2649-2

    work page doi:10.1038/s41586-020-2649-2 2020

  5. [5]

    Hunter, J. D. 2007, Computing in Science and Engineering, 9, 90, doi: 10.1109/MCSE.2007.55

    work page doi:10.1109/mcse.2007.55 2007

  6. [6]

    C., Schaller, M., et al

    McAlpine, S., Helly, J. C., Schaller, M., et al. 2016, Astronomy and Computing, 15, 72, doi: 10.1016/j.ascom.2016.02.004

    work page doi:10.1016/j.ascom.2016.02.004 2016

  7. [7]

    The IllustrisTNG simulations: public data release

    Nelson, D., Springel, V., Pillepich, A., et al. 2019, Computational Astrophysics and Cosmology, 6, 2, doi: 10.1186/s40668-019-0028-x

    work page doi:10.1186/s40668-019-0028-x 2019

  8. [8]

    H., Rai c evi\'c M., Schaye J., 2013, @doi [Monthly Notices of the Royal Astronomical Society] 10.1093/mnras/stt066 , 430, 2427–2445

    Rahmati, A., Pawlik, A. H., Raiˇ cevi´ c, M., & Schaye, J. 2013, MNRAS, 430, 2427, doi: 10.1093/mnras/stt066

    work page doi:10.1093/mnras/stt066 2013

  9. [9]

    2011, MNRAS, 417, 333, doi: 10.1111/j.1365-2966.2011.19266.x

    Sousbie, T. 2011, MNRAS, 414, 350, doi: 10.1111/j.1365-2966.2011.18394.x

    work page doi:10.1111/j.1365-2966.2011.18394.x 2011

  10. [10]

    2011, MNRAS, 417, 333, doi: 10.1111/j.1365-2966.2011.19266.x

    Sousbie, T., Pichon, C., & Kawahara, H. 2011, MNRAS, 414, 384, doi: 10.1111/j.1365-2966.2011.18395.x

    work page doi:10.1111/j.1365-2966.2011.18395.x 2011

  11. [11]

    2020a, A&A, 637, A41, doi: 10.1051/0004-6361/201937158 10

    Douspis, M. 2020a, A&A, 637, A41, doi: 10.1051/0004-6361/201937158 10

    work page doi:10.1051/0004-6361/201937158

  12. [12]

    2022, A&A, 667, A161, doi: 10.1051/0004-6361/202244158

    Tanimura, H., Aghanim, N., Douspis, M., & Malavasi, N. 2022, A&A, 667, A161, doi: 10.1051/0004-6361/202244158

    work page doi:10.1051/0004-6361/202244158 2022

  13. [13]

    The EAGLE simulations of galaxy formation: Public release of particle data

    Malavasi, N. 2020b, A&A, 643, L2, doi: 10.1051/0004-6361/202038521 The EAGLE team. 2017, arXiv e-prints, arXiv:1706.09899, doi: 10.48550/arXiv.1706.09899

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1051/0004-6361/202038521 2017

  14. [14]

    2019, MNRAS, 489, 385, doi: 10.1093/mnras/stz2146

    Tramonte, D., Ma, Y.-Z., Li, Y.-C., & Staveley-Smith, L. 2019, MNRAS, 489, 385, doi: 10.1093/mnras/stz2146

    work page doi:10.1093/mnras/stz2146 2019

  15. [15]

    2021, MNRAS, 505, 4178, doi: 10.1093/mnras/stab1301

    Vernstrom, T., Heald, G., Vazza, F., et al. 2021, MNRAS, 505, 4178, doi: 10.1093/mnras/stab1301

    work page doi:10.1093/mnras/stab1301 2021

  16. [16]

    2023, Science Advances, 9, eade7233, doi: 10.1126/sciadv.ade7233

    Vernstrom, T., West, J., Vazza, F., et al. 2023, Science Advances, 9, eade7233, doi: 10.1126/sciadv.ade7233

    work page doi:10.1126/sciadv.ade7233 2023

  17. [17]

    E., et al

    Virtanen, P., Gommers, R., Oliphant, T. E., et al. 2020, Nature Methods, 17, 261, doi: 10.1038/s41592-019-0686-2

    work page doi:10.1038/s41592-019-0686-2 2020

  18. [18]

    J., & Afshordi, N

    Yang, T., Hudson, M. J., & Afshordi, N. 2020, MNRAS, 498, 3158, doi: 10.1093/mnras/staa2547

    work page doi:10.1093/mnras/staa2547 2020