Cosmological inference with halo clustering reconstructed from the redshift-space galaxy distribution

arXiv:2604.11694 · astro-ph.CO · Submitted 2026-04-13 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.

Vera: I'm Vera.

Jocelyn: And I'm Jocelyn, here with Subrahmanyan.

Vera: We're looking at the paper: "Cosmological inference with halo clustering reconstructed from the redshift-space galaxy distribution".

Jocelyn: Here's the gist of the paper. Accurate modeling of small-scale redshift-space clustering is crucial for full shape RSD analyses, where satellite galaxies contribute to 1-halo terms and Finger-of-God distortions Using DESI-like luminous red galaxy mock catalogs from the AbacusSummit simulations at z = 1.1, we perform effective field theory (EFT)-based full-shape modeling of the power spectrum of the reconstructed-halo sample We show that the dominant reconstruction-induced systematics can be described and incorporated within the standard EFT framework The reconstructed-halo sample yields unbiased constraints on cosmological parameters, including the growth rate fσ8 and Alcock-Paczynski parameters Compared to the galaxy sample, it enables both improved robustness and increased statistical precision.

Vera: Halo Reconstruction Methodology The Cylinder-grouping (CG) method provides a practical approach for reconstructing a halo-like field directly from the observed galaxy distribution The primary goal of the method is to mitigate small-scale nonlinear effects in redshift space, in particular the FoG distortions produced by the virial motions of satellite galaxies within dark matter halos.

Jocelyn: A key feature of the CG method is that the grouping window is anisotropic, taking the form of a cylinder aligned with the LOS The algorithm ranks galaxies by a halo-mass proxy and iteratively identifies central–satellite systems by removing galaxies within a cylinder aligned with the LOS, chosen to encompass FoG displacements.

Vera: Systematic Decomposition of Effects The CG procedure yields a reconstructed halo field, consisting of the galaxies identified as centrals The reconstructed autocorrelation function can be expanded as a sum involving the baseline term and terms encoding reconstruction imperfections.

Title and authors: Vera: So we're talking about the title of this work, "Cosmological inference with halo clustering reconstructed from the redshift-space galaxy distribution," and who wrote it. The authors are Ryuichiro Hada and Teppei Okumura.

Jocelyn: They’re coming from different orbits, which is interesting because they’re linking observational clustering techniques to theoretical modeling within the effective field theory framework.

Subrahmanyan: It suggests that the way we reconstruct structure from galaxies isn't just a data processing step; it has a direct and quantifiable impact on the cosmological parameters we can extract.

Vera: The paper is essentially showing how you can take what you see in galaxy distributions, use a technique like Cylinder Grouping to pull out halo information, and then model the resulting power spectrum using EFT to get cleaner cosmology.

Jocelyn: It’s about turning observational noise into a structured field that we can use for cosmological inference while keeping the modeling framework standard.

Subrahmanyan: The implication here is that we don't have to invent entirely new ways of modeling these nonlinearities; we just need to be smart about how we incorporate the reconstruction effects into existing tools like EFT.

Vera: Right. It’s about making the connection between the messy reality of galaxy distributions and clean cosmological parameters more direct and less biased by small-scale noise.

The paper's summary: Jocelyn: So, what they summarize is that they used the Cylinder Grouping method to reconstruct a halo field from galaxy data, specifically designed to tackle those small-scale redshift space distortions like the Finger-of-God effects.

Vera: They show that by modeling this reconstructed sample's power spectrum using EFT, they can account for the dominant reconstruction systematics within that same EFT framework.

Subrahmanyan: The key result they highlight is that this approach leads to unbiased constraints on parameters such as the growth rate f sigma eight and the Alcock-Paczynski parameters.

Jocelyn: They also found that when you compare their reconstructed sample to just the galaxy sample, it provides both improved robustness and higher statistical precision for parameter estimation.

Vera: Specifically, they managed to get a better constraint on the growth rate f sigma eight by extending the fitting range beyond k max equal point zero two h Mpc minus one.

Subrahmanyan: That’s significant because in galaxy-based analyses, that growth rate starts showing a systematic shift at that scale. So, this reconstruction method helps keep things stable where we need them to be.

The paper's improvements: Vera: Now for the improvements they suggest. They break down the reconstruction imperfections into two main channels: a small-scale exclusion effect related to the cylinder grouping itself, and a large-scale directional masking effect.

Jocelyn: For that small scale exclusion, they model it as an effective hard-core window acting on the baseline power term, which separates into a pure exclusion part and something describing how that window removes power from pairs inside the excluded region.

Subrahmanyan: And for the large scale directional masking effect, they treat that as an apparent anisotropy caused by misidentification in the grouping, and they correct it by applying large scale ratios measured from their mocks as a multiplicative rescaling of the deterministic part of their model.

Vera: So, essentially they’re saying you can decompose these complex reconstruction effects into something manageable: a small scale exclusion and a large scale directional masking that you can correct explicitly.

Jocelyn: And that lets them keep the EFT framework intact for the stochastic and counterterm sectors, which is really tidy for modeling.

Subrahmanyan: That’s an important point because it shows how these reconstruction systematics map directly onto the ingredients of the EFT model itself, suggesting they are naturally complementary tools.

Conclusion: Vera: To wrap up this paper on "Cosmological inference with halo clustering reconstructed from the redshift-space galaxy distribution," the main message is that using CG-based halo reconstruction provides a practical way to mitigate those satellite-driven nonlinearities and Finger-of-God effects at the field level.

Jocelyn: They demonstrate that this method allows for more robust cosmological inference from quasi-nonlinear scales while staying within standard EFT pipelines without adding extra reconstruction parameters.

Subrahmanyan: The results are compelling because they achieve a reduction in the statistical uncertainty of f sigma eight by more than twenty percent at high k max compared to galaxy samples, which is a direct improvement over what host halo samples can do alone.

Vera: It shows that this reconstructed-halo sample yields unbiased cosmological constraints in the compressed parameter space, and it’s consistent with our fiducial cosmology.

Jocelyn: So, the paper suggests we should look at how these reconstruction systematics fit into EFT to see if other analyses can benefit from this kind of structured modeling approach.

Subrahmanyan: It confirms that when you have a good halo reconstruction method, you can get tighter constraints on the growth rate f sigma eight by extending the analysis to higher k max values where galaxy data usually breaks down.

Vera: That’s what we're looking at today with this paper, and it shows that combining observational reconstruction with EFT modeling is a very natural fit for getting cleaner cosmology from these complex structures.

Academia Sinica Institute of Astronomy and Astrophysics · Kavli Institute for Particle Physics at the University of Tokyo

astro-ph.CO

Submitted: 2026-04-13

Updated: 2026-11-13

Comments: 22 pages, 11 figures, 2 tables; published in PRD

Journal ref: Phys. Rev. D 114, 103516 (2026)

DOI: 10.1103/4fr5-9l8j

Code: https://github.com/sfschen/velocileptors

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 95/100

The gist: Accurate modeling of small-scale redshift-space clustering is crucial for full shape RSD analyses, where satellite galaxies contribute to 1-halo terms and Finger-of-God distortions Using DESI-like…

Key concepts

Cylinder-grouping (CG)
A practical method used to reconstruct a halo field directly from observed galaxies. It works by grouping galaxies into anisotropic cylinder shapes aligned with the line of sight. This helps in mitigating small-scale nonlinear effects, specifically the Finger-of-God distortions caused by satellite motions within dark matter halos.
Effective Field Theory (EFT)
A theoretical framework used to model cosmological power spectra at different scales. The paper incorporates reconstruction systematics—like exclusion and directional masking effects—into this standard EFT structure, allowing for unbiased parameter inference without needing new, reconstruction-specific parameters.
Exclusion Effect
This describes the small-scale systematic error arising from the cylinder grouping process. It is modeled as an effective hard-core window acting on the baseline clustering term. This effect is treated within the EFT stochastic and counterterm sector, allowing it to be absorbed into standard modeling.
Directional Masking Effect
This refers to a large-scale systematic where misidentification in grouping causes apparent anisotropic clustering. This is corrected by applying multiplicative renormalization using large-scale ratios measured from mock catalogs, treating it as a multipole-dependent rescaling of the deterministic model.

Terminology

Summary

Accurate modeling of small-scale redshift-space clustering is crucial for full shape RSD analyses, where satellite galaxies contribute to 1-halo terms and Finger-of-God distortions Using DESI-like luminous red galaxy mock catalogs from the AbacusSummit simulations at z = 1. p. 01, we perform effective field theory (EFT)-based full-shape modeling of the power spectrum of the reconstructed-halo sample We show that the dominant reconstruction-induced systematics can be described and incorporated within the standard EFT framework The reconstructed-halo sample yields unbiased constraints on cosmological parameters, including the growth rate fσ8 and Alcock-Paczynski parameters Compared to the galaxy sample, it enables both improved robustness and increased statistical precision. p. 0

Halo Reconstruction Methodology

The Cylinder-grouping (CG) method provides a practical approach for reconstructing a halo-like field directly from the observed galaxy distribution The primary goal of the method is to mitigate small-scale nonlinear effects in redshift space, in particular the FoG distortions produced by the virial motions of satellite galaxies within dark matter halos. A key feature of the CG method is that the grouping window is anisotropic, taking the form of a cylinder aligned with the LOS The algorithm ranks galaxies by a halo-mass proxy and iteratively identifies central–satellite systems by removing galaxies within a cylinder aligned with the LOS, chosen to encompass FoG displacements. p. 1

Systematic Decomposition of Effects

The CG procedure yields a reconstructed halo field, consisting of the galaxies identified as centrals The reconstructed autocorrelation function can be expanded as a sum involving the baseline term and terms encoding reconstruction imperfections. These imperfections are decomposed into two qualitative channels: (i) a small-scale exclusion effect associated with cylinder grouping, whose residual impact is treated within the EFT stochastic and counterterm sector, and (ii) a large-scale “directional masking” effect that primarily rescales the clustering amplitude in a multipole-dependent manner. p. 1

Modeling Reconstruction Systematics

The small-scale exclusion contribution, denoted as the “exclusion” effect, is modeled by an effective hard-core window acting on the dominant baseline term This correction naturally separates into a “pure” exclusion contribution and a term describing how the exclusion window removes the part of the baseline power sourced by pairs within the excluded region. The large-scale directional masking effect is interpreted as an apparent anisotropic clustering induced by misidentification in the CG grouping, which is corrected by applying large-scale ratios measured from the mocks as a multiplicative renormalization of the large-scale deterministic part of the model. p. 1

Cosmological Inference Outcomes

The reconstructed-halo sample yields unbiased cosmological constraints in the compressed parameter space (q∥, q⊥, dm, fσ8), consistent with the fiducial cosmology. Compared to galaxy-based analyses, the halo reconstruction leads to a clear improvement in both robustness and precision In particular, the inferred growth rate fσ8 remains stable as the fitting range is extended beyond kmax ≃ 0. p. 02 h Mpc−1, where the galaxy-based result begins to exhibit a systematic shift. p. 6

Parameter Constraint Stability

The reconstructed-halo sample achieves a reduction in the statistical uncertainty of fσ8 by more than ∼ 20% at high kmax. This improvement is traced to reduced degeneracies in the nuisance sector, specifically a substantially reduced SN0–SN2 degeneracy, which weakens the correlations with fσ8 and enables tighter constraints. p. 0 The host-halo sample exhibits an intermediate behavior, with partial mitigation of these degeneracies but a less pronounced improvement compared to the reconstructed-halo sample. p. 1

Conclusion

CG-based halo reconstruction provides a practical route to mitigating satellite-driven nonlinearities and FoG effects at the field level, thereby enabling more robust cosmological inference from quasi-nonlinear scales within standard EFT pipelines. The treatment adopted here also clarifies how reconstruction systematics map onto EFT ingredients and suggests that halo reconstruction and EFT modeling are naturally complementary. p. 7

References

[1] N. Kaiser, MNRAS 227, 1 (1987)

[2] A. J. S. Hamilton, Astrophys. J. 385, L5 (1992)

[3] S. Cole, K. B. Fisher, and D. H. Weinberg, MNRAS 267, 785 (1994), astro-ph/9308003

[4] A. J. S. Hamilton, Linear Redshift Distortions: a Review (1998), vol. p. 7 231 of Astrophysics and Space Science Library, p. p. 0 185

[5] L. Guzzo, M. Pierleoni, B. Meneux, E. Branchini, O. Le Fevre, C. Marinoni, B. Garilli, J. Blaizot, G. De Lucia, A. Pollo, et al., Nature (London) 451, 541 (2008), 0802. p. 41944

[6] T. Okumura, C. p. 0 Hikage, T. Totani, M. Tonegawa, H. Okada, K. Glazebrook, C. Blake, P. G. Ferreira, S. More, A. Taruya, et al., Publ. Astron Soc Japan 68 38 (2016), 1511-08083

[7] S. Alam, M. Ata, S. Bailey, F. Beutler, D. Bizyaev, J. A. Blazek, A. S. Bolton, J. R. Brownstein, A Burden, C.-H Chuang, et al., MNRAS 470 2617 (2017), 1607-03155

[8] T. Okumura, T. p. 0 Matsubara, D. J Eisenstein, I. Kayo, C. Hikage, A S Szalay, and D P Schneider, Astrophys J 676 889-898 (2008), 0711-3640

[9] G. d’Amico, J. Gleyzes, N Kokron, K Markovic, L Senatore, P Zhang, F Beutler, and H Gil-Mar´ın, JCAP 05 005 (2020), 1909-05271

[10] M. M. Ivanov, M Simonovi´c, and M Zaldarriaga, JCAP 05 042 (2020), 1909-05277

[11] O. H. E. Philcox, M M Ivanov, M Simonovi´c, and M Zaldarriaga, JCAP 05 032 (2020), 2002-04035

[12] Y. Kobayashi, T Nishimichi, M Takada, and H Miyatake, Phys Rev D 105 83517 (2022), 2110-06969

[13] A. G Adame, J Aguilar, S Ahlen, S Alam, D M Alexander, C Allende Prieto, M Alvarez, O Alves, A Anand, U Andrade et al., JCAP 07 028 (2025), 2411-12022

[14] J C Jackson, MNRAS 156 1P (1972), 0810.3908

[15] T Okumura, N Hand, U Seljak, Z Vlah, and V Desjacques, Phys Rev D 92 103516 (2015), 1506-05814

[16] J J M Carrasco, M P Hertzberg, and L Senatore, Journal of High Energy Physics 2012 82 (2012), 1206-2926

[17] R A Porto, L Senatore, and M Zaldarriaga, JCAP 05 022 (2014), 1311-2168

[18] Z Vlah, M White, and A Aviles, JCAP 09 014 (2015), 1506-05264

[19] W Handley and P Lemos, Phys Rev D 100 043504 (2019), 1902-0435

[20] P Lemos, M Raveri, A Campos, Y Park, C Chang, N Weaverdyck, D Huterer, A R Liddle, J Blazek R Cawthon et al., MNRAS 505 6179 (2021), 2012-09554

[21] A G´omez-Valent, Phys Rev D 106 063506 (2022), 2385-16285

[23] F Simpson, A F Heavens, and C Heymans, Phys Rev D 88 083510 (2013), 1306-6349

[24] T Okumura, M Takada, S More, and S Masaki, MNRAS 469 459 (2017), 1611-04165

[25] DESI Collaboration, A Aghamousa, J Aguilar, S Ahlen, S Alam, L E Allen, C Allende Prieto, J Annis, S Bailey, C Balland et al., arXiv e-prints arXiv:1611. p. 000036 (2016), 1611-00036

[26] M Takada, R S Ellis, M Chiba, J E Greene, H Aihara, N Arimoto, K Bundy, J Cohen O Dor´e G Graves et al., Publ Astron Soc Japan 66 R1 (2014), 1206-0737

[27] T Baldauf, U Seljak, R E Smith, N Hamaus, and V Desjacques, Phys Rev D 88 083507 (2013), 1305-2917

[28] F C van den Bosch, S More, M Cacciato, H Mo, and X Yang, MNRAS 430 725 (2013), 1206-6890

[29] N A Maksimova, L H Garrison, D J Eisenstein, B Hadzhiyska, S Bose, and T P Satterthwaite, MNRAS 508 4017 (2021), 2110-11398

[30] S Yuan, B Hadzhiyska, S Bose, and D J Eisenstein, MNRAS 512 5793 (2022), 2202-12911

[31] Z Zheng, A L Coil, and I Zehavi, Astrophys J 667 760 (2007), astro-ph/0703457

[32] M Maus, S Chen, M White, J Aguilar, S Ahlen, A Aviles, S Brieden, D Brooks T Claybaugh S Cole et al., JCAP 01 138 (2025), 2404-07312

[33] S-F Chen, Z Vlah, and M White, JCAP 07 062 (2020), 2005-0523

[34] S-F Chen, Z Vlah, E Castorina, and M White, JCAP 03 100 (2021), 2012-04636

[35] S Brieden, H Gil-Mar´ın, and L Verde, JCAP 12 054 (2021), 2106-07641

[36] C Alcock and B Paczynski, Nature (London) 281, 358 (1979)

[37] W E Ballinger, J A Peacock, and A F Heavens, MNRAS 282 877 (1996), arXiv:astro-ph/9605017

[38] U Seljak, JCAP 03 004 (2012), 1201-594

Appendix A

Figure 5 shows the full posterior distributions in the complete parameter space for galaxies (red), halos (blue), and reconstructed halos (green). The diagonal panels show marginalized 1D posteriors, while the off-diagonal panels show the corresponding 2D confidence regions. p. 9 Figure 5 is intended as a compact reference for assessing parameter degeneracies beyond the compressed parameter set discussed in the main text. p. 9 Figure 6 summarizes the kmax dependence of the AP parameters (q∥ and q⊥) and the shape parameter dm. p. 9 Consistent with the behavior of fσ8 discussed in Section V, the reconstructed-halo sample yields stable best-fit values for q∥, q⊥, and dm as kmax is increased, while achieving reduced uncertainties compared to galaxies. p. 0

Figure 1

Figure 1 shows the redshift-space monopole and quadrupole of the galaxy, host-halo, and reconstructed halo samples at z = 1.1. The galaxy sample shows the expected signature of satellite contamination. p. 9 In particular, the quadrupole shows a clear suppression of power on small scales relative to the host-halo sample, reflecting the FoG damping induced by virial motions of satellite velocities. p. 9 The reconstructed-halo sample closely follows the host-halo clustering over a wide range of scales. p. 9 In particular, the small-scale suppression in the quadrupole is largely eliminated, demonstrating that the CG procedure successfully removes most satellite contributions. p. 9

Figure 2

Figure 2 shows the posterior distributions in the compressed parameter space (q∥, q⊥, dm, fσ8) at z = 1.1 for a representative choice of kmax = 0. p. 9 p. 92 h Mpc−1. p. 0 p. 0 All three tracer samples yield constraints that are consistent with the fiducial cosmology, indicating that no significant bias is introduced by the halo reconstruction within our modeling framework. p. 9 Quantitatively, the reconstructed-halo sample achieves a constraint on the growth parameter of fσ8 = 0. p. 9412 ± 0.018, which is tighter than that obtained from galaxies (0. p. 9420 ± 0.023) and marginally tighter than that from host halos (0. p. 9416 ± 0.021).

Figure 3

Figure 3 shows the joint posterior distributions of fσ8, the linear bias parameter b1, and the stochastic parameters SN0 and SN2 at z = 1.1 for a representative choice of kmax = 0.2 h Mpc−1. The best-fitting value of b1 for the reconstructed halos (b1 = 1. p. 9218 ± 0.022) is closer to that for galaxies than to that for the true host halos because of how we correct for the large-scale masking effect in Sec. p. 1 IV D. The reconstructed sample exhibits a substantially reduced degeneracy within the stochastic sector, particularly between SN0 and SN2. p. 2

Figure 4

Figure 4 shows the inferred constraints on the growth-rate parameter fσ8 as a function of the maximum wavenumber kmax obtained for galaxy (red), host-halos (blue), and reconstructed-halo (green) samples. For kmax ≳ 0. p. 015 h Mpc−1, the reconstructed sample yields more than a 20% reduction in the 1σ uncertainty on fσ8 relative to the galaxy case, exceeding the improvement obtained from the host-halo sample. p. 9

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. System can perform unbiased cosmological parameter inference on reconstructed-halo samples by incorporating reconstruction-induced systematics within a standard EFT framework without introducing additional reconstruction-specific parameters.

  2. The improved system can model large-scale directional masking as a multipole dependent rescaling of the clustering amplitude, corrected by applying large scale ratios measured from the mocks as a multiplicative renormalization of the deterministic part of the model.

  3. The system can achieve more robust and precise constraints on cosmological parameters like growth rate fsigma8, with uncertainty reductions exceeding 20% at high kmax compared to galaxy samples.

  4. The AI can identify that residual small-scale changes are absorbed by the standard counterterm and stochastic sector, without introducing additional reconstruction-specific parameters.

  5. The system can provide a clear separation of reconstruction effects into an exclusion effect (small scale) and a directional masking effect (large scale), allowing for transparent interpretation within the EFT template.

Sources

Related papers