Baryonification II: Constraining feedback with X-ray and kinematic Sunyaev-Zel'dovich observations

arXiv:2507.07991 · astro-ph.CO · Submitted 2025-07-10 · 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: We're looking at the paper: "Baryonification II: Constraining feedback with X-ray and kinematic Sunyaev-Zel'dovich observations".

Jocelyn: Here's the gist of the paper. Baryonic feedback alters the matter distribution on small and intermediate scales, posing a challenge for precision cosmology The new, component-wise baryonification (BFC) approach introduced in Ref.

1: provides a self-consistent framework to model feedback effects for different observables In this paper we use this framework to fit kinematic SunyaevZel’dovich (kSZ) observations from the Atacama Cosmology Telescope (ACT) alongside halo X-ray gas fractions from eROSITA, investigating baryonic feedback in a cosmological context We first show that the kSZ data from ACT is consistent with the gas fractions from eROSITA, both suggesting a feedback model that is stronger than what is assumed in most hydrodynamical simulations Based on our combined fit to data from ACT and eROSITA, we find a power spectrum suppression that exceeds the percent-level at modes above k = 0.3 − 0.6 h Mpc−1, growing to 2-8 percent at k = 1 h Mpc−1, and to 20-25 percent at k = 5 h Mpc−1, consistent with strong-feedback hydrodynamical simulations Finally, we compare our best-fitting model to the observed gas density and pressure profiles of massive galaxy clusters from the X-COP sample, finding excellent agreement.

Jocelyn: Theoretical Framework The baryonification model is based on post-processing gravity-only N-body simulations to incorporate baryonic feedback effects due to gas and stars By displacing particles around halo centers according to empirical halo profiles, the model reproduces the impact of baryonic physics on the large-scale structure The hot gas component is described by a smoothly declining density profile to account for the extended distribution of the hot gas The total halo pressure can be defined as Ptot = Pnth +Pth, where Pnth represents the non-thermal and Pth the thermal component of the pressure.

Vera: Observational Constraints The work utilizes complementary observational probes to constrain baryonic physics The kinematic Sunyaev-Zel’dovich effect provides a direct probe of the line-of-sight momentum of ionized gas X-ray measurements trace its thermal state and spatial distribution in galaxy groups and clusters The ACT kSZ measurements probe the line-of-sight momentum of ionized gas, providing direct sensitivity to the gas density distribution when combined with reconstructed galaxy velocities X-ray gas fractions are used to constrain the hot gas mass fraction across a broad range of halo masses.

Title and authors: Vera: So this paper, "Baryonification II: Constraining feedback with X-ray and kinematic Sunyaev-Zel'dovich observations," it’s looking at how we can use these two different types of data to really nail down what’s happening in the universe on small and intermediate scales.

Jocelyn: It focuses on using the kinematic Sunyaev-Zel'dovich effect from ACT and X-ray gas fractions from eROSITA to test this idea.

Subrahmanyan: Essentially, they are taking these two probes, which look at different aspects of the gas in galaxy clusters, and combining them through a new modeling approach to see if the physics we assume about feedback is actually right.

Vera: Exactly. It’s about moving beyond just looking at one dataset and seeing how those constraints line up when you put them together.

Jocelyn: And the title hints that they are constraining feedback, which is that process where baryonic physics changes how matter is distributed on these smaller scales in the universe.

Subrahmanyan: And what this paper suggests is that this combined approach gives us a self-consistent framework to model those feedback effects for different things we want to measure.

The paper's summary: Vera: Now, looking at the summary, the main point is that they first show the kinematic Sunyaev-Zel'dovich data from ACT is actually consistent with the gas fractions measured by eROSITA.

Jocelyn: That consistency suggests that we need a feedback model that’s stronger than what most of our current hydrodynamical simulations assume.

Subrahmanyan: That’s significant because it contradicts older results from pre-eROSITA gas fraction measurements, which pointed toward weaker feedback in tension with the kinematic Sunyaev-Zel'dovich results.

Vera: The authors suspect these differences are just selection bias in the older data or maybe just different ways of estimating halo mass between the two datasets.

Jocelyn: And then they take this combined fit to use their model to predict how much the matter power spectrum gets suppressed because of this baryonic feedback.

Subrahmanyan: That suppression is quite substantial, exceeding the percent level at modes where k is between zero point three and zero point six per h megaparsec, growing up to twenty percent or twenty-five percent at a wavenumber of five per h megaparsec.

The paper's improvements: Vera: What they suggest as an improvement is using this component-wise baryonification approach, which is introduced in a reference we call the BFC framework, to model feedback effects for different observables in a self-consistent way.

Jocelyn: It’s not just fitting the data; it’s building a framework that lets you model these feedback effects across different measurements—gas fractions, SZ effects, and matter clustering.

Subrahmanyan: They use this BFC model to show that when you combine ACT kSZ observations with eROSITA gas fractions, you get robust and self-consistent constraints on how strong the feedback is.

Vera: It’s interesting because they manage to reproduce the observed gas density and pressure profiles of massive galaxy clusters from the X-COP sample remarkably well, staying within one sigma across the whole radial range for both density and pressure.

Jocelyn: So they aren't just getting a power spectrum suppression; they are also nailing the physical structure of these clusters.

Conclusion: Vera: To wrap up, the main implication is that combining ACT kSZ and eROSITA data gives us self-consistent constraints on baryonic feedback strength.

Jocelyn: It confirms that eROSITA gas fractions are more consistent with a strong feedback scenario, while the older pre-eROSITA results align better with the simulation called m8 in FLAMINGO.

Subrahmanyan: From my side, this result points toward strong baryonic feedback effects being necessary to jointly describe these observations across different scales.

Vera: We’re seeing a predicted suppression of power that matches the fgas-eight sigma FLAMINGO simulation across a wide range of scales, while disfavoring the m8 model based on multiple independent observables.

Jocelyn: It’s a result that clearly deviates from the fiducial FLAMINGO m8 simulation and points toward strong feedback effects required to jointly describe the combination of gas fractions from eROSITA and kSZ observations.

Argelander-Institut für Astronomie, University of Bonn · Jodrell Bank Centre for Astrophysics, The University of Manchester · Institute for Particle Physics and Astrophysics, ETH Zurich · Department of Astrophysics, University of Zurich · Van Swinderen Institute for Particle Physics and Gravity, University of Groningen · Department of Astronomy and Oskar Klein Centre, Stockholm University · Department of Astrophysical Sciences, Princeton University · Leiden Observatory, Leiden University · Lorentz Institute for Theoretical Physics, Leiden University

astro-ph.CO

Submitted: 2025-07-10

Updated: 2026-11-16

Comments: 40 pages, 12 figures, 2 tables, published in JCAP

Journal ref: JCAP 11 (2025) 046

DOI: 10.1088/1475-7516/2025/11/046

Code: https://github.com/LSSTDESC/CCL

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 89/100

The gist: Baryonic feedback alters the matter distribution on small and intermediate scales, posing a challenge for precision cosmology The new, component-wise baryonification (BFC) approach introduced in Ref.

Key concepts

Component-wise Baryonification (BFC)
This is a self-consistent framework used to model feedback effects by displacing particles in gravity simulations based on empirical halo profiles. It allows researchers to test how different physical processes, like gas and star formation, change the matter distribution across various scales.
Kinematic Sunyaev-Zel'dovich (kSZ) effect
This is a direct probe of the line-of-sight momentum of ionized gas. By measuring this signal from ACT, scientists can gain direct sensitivity to the gas density distribution when combined with reconstructed galaxy velocities, providing a unique constraint on baryonic physics.
X-ray Gas Fractions
These measurements trace the thermal state and spatial distribution of hot gas in galaxy groups and clusters using X-ray observations. They are used to constrain the total hot gas mass fraction across a wide range of halo masses, helping to determine how much baryonic matter is present.
Power Spectrum Suppression
This refers to the reduction in the clustering of matter on small scales, measured by analyzing the matter power spectrum. The model predicts this suppression exceeds 20% at scales corresponding to k = 5 h Mpc-1, indicating strong baryonic feedback effects are necessary.

Terminology

Summary

Baryonic feedback alters the matter distribution on small and intermediate scales, posing a challenge for precision cosmology The new, component-wise baryonification (BFC) approach introduced in Ref. p. 1

Where the paper says it

Baryonic feedback alters the matter distribution on small and intermediate scales, posing a challenge for precision cosmology. The new, component-wise baryonification (BFC) approach introduced in Ref.

[1] provides a self-consistent framework to model feedback effects for different observables In this paper we use this framework to fit kinematic SunyaevZel’dovich (kSZ) observations from the Atacama Cosmology Telescope (ACT) alongside halo X-ray gas fractions from eROSITA, investigating baryonic feedback in a cosmological context We first show that the kSZ data from ACT is consistent with the gas fractions from eROSITA, both suggesting a feedback model that is stronger than what is assumed in most hydrodynamical simulations Based on our combined fit to data from ACT and eROSITA, we find a power spectrum suppression that exceeds the percent-level at modes above k = 0. p. 1
Where the paper says it

In this paper we use this framework to fit kinematic SunyaevZel’dovich (kSZ) observations from the Atacama Cosmology Telescope (ACT) alongside halo X-ray gas fractions from eROSITA, investigating baryonic feedback in a cosmological context. We first show that the kSZ data from ACT is consistent with the gas fractions from eROSITA, both suggesting a feedback model that is stronger than what is assumed in most hydrodynamical simulations.

3 − 0.6 h Mpc−1, growing to 2-8 percent at k = 1 h Mpc−1, and to 20-25 percent at k = 5 h Mpc−1, consistent with strong-feedback hydrodynamical simulations Finally, we compare our best-fitting model to the observed gas density and pressure profiles of massive galaxy clusters from the X-COP sample, finding excellent agreement. p. 1
Where the paper says it

6 h Mpc−1 , growing to 2-8 percent at k = 1 h Mpc−1 , and to 20-25 percent at k = 5 h Mpc−1 , consistent with strong-feedback hydrodynamical simulations. Finally, we compare our best-fitting model to the observed gas density and pressure profiles of massive galaxy clusters from the X-COP sample, finding excellent agreement.

Theoretical Framework

The baryonification model is based on post-processing gravity-only N-body simulations to incorporate baryonic feedback effects due to gas and stars By displacing particles around halo centers according to empirical halo profiles, the model reproduces the impact of baryonic physics on the large-scale structure The hot gas component is described by a smoothly declining density profile to account for the extended distribution of the hot gas The total halo pressure can be defined as Ptot = Pnth +Pth, where Pnth represents the non-thermal and Pth the thermal component of the pressure.

Observational Constraints

The work utilizes complementary observational probes to constrain baryonic physics The kinematic Sunyaev-Zel’dovich effect provides a direct probe of the line-of-sight momentum of ionized gas X-ray measurements trace its thermal state and spatial distribution in galaxy groups and clusters The ACT kSZ measurements probe the line-of-sight momentum of ionized gas, providing direct sensitivity to the gas density distribution when combined with reconstructed galaxy velocities X-ray gas fractions are used to constrain the hot gas mass fraction across a broad range of halo masses.

Joint Fitting and Consistency

The joint fit of the kSZ signal and the eROSITA gas fractions yields self-consistent constraints on baryonic feedback strength The combination of these two observables jointly favors a strong-feedback scenario, despite probing different mass, redshift, and radial regimes The results show that the combination of kSZ observations from ACT and X-ray gas fractions from eROSITA yields robust and self-consistent constraints on baryonic feedback when modeled with the updated component-wise BFC framework.

Implications for Structure Formation

The predicted matter power spectrum suppression exceeds the percent level at k > 0.3 h/Mpc, reaching up to 25% at k = 5 h/Mpc, in line with strong-feedback hydrodynamical simulations The model reproduces the gas density and pressure profiles of massive clusters in the X-COP sample remarkably well, staying within 1σ across the full radial range for both gas density and pressure. p. 4

Where the paper says it

3 h/Mpc, reaching up to 25% at k = 5 h/Mpc, in line with strong-feedback hydrodynamical simulations, such as the fgas − 8σ FLAMINGO simulation. Furthermore, the same model successfully reproduces the gas density and pressure profiles of massive clusters in the X-COP sample, reinforcing its applicability across mass scales and observables.

Conclusion on Feedback Strength

The combination of kSZ and eROSITA data enables self-consistent constraints on baryonic feedback strength The results demonstrate that the addition of eROSITA gas fractions enables consistency with X-ray constraints as well. The suppression of power predicted by our model matches the fgas–8σ FLAMINGO simulation across a wide range of scales, while disfavoring the m8 model based on multiple independent observables. p. 4

Where the paper says it

3 h/Mpc, reaching up to 25% at k = 5 h/Mpc, in line with strong-feedback hydrodynamical simulations, such as the fgas − 8σ FLAMINGO simulation. Furthermore, the same model successfully reproduces the gas density and pressure profiles of massive clusters in the X-COP sample, reinforcing its applicability across mass scales and observables.

Future Applications

The BFC framework provides a powerful, selfconsistent model of baryonic feedback across mass scales and observables, including gas fractions, SZ effects, and matter clustering Its predictive accuracy and physical interpretability make it highly promising for application to additional cosmological observables such as thermal SZ or X-ray observations. The results show that the BFC framework provides a powerful, selfconsistent model of baryonic feedback across mass scales and observables, including gas fractions, SZ effects, and matter clustering. p. 1

Where the paper says it

Finally, we compare our best-fitting model to the observed gas density and pressure profiles of massive galaxy clusters from the X-COP sample, finding excellent agreement. These results show that BFC provides a self-consistent picture of feedback across mass- and length scales as well as different cosmological observables, thus making it promising for applications to multiwavelength studies to jointly constrain cosmology and baryonic effects.

Parameter Sensitivity

The analysis shows that the kSZ data imposes a preference for lower gas fractions at the group scale, creating tension with the trend implied by the pre-eROSITA gas fractions alone The discrepancy between individual datasets suggests limited consistency in the underlying constraints The parameter reduction procedure leaves four free parameters (BFC3 + M200,kSZ): Mc, µ, and δ, which shape the gas profile, and the effective halo mass M200,kSZ.

Satellite Contribution

The omission of satellites does not significantly affect the level of suppression inferred from current data The posterior distributions shift only slightly, remaining consistent with the fiducial constraints within statistical uncertainties. The modeling choice to neglect satellites is justified for the current signal-to-noise level. p. 6

Where the paper says it

We focus here on the hot gas component, which dominates the baryonic budget in group- and cluster-scale halos and is particularly sensitive to AGN feedback. Its accurate modeling is therefore essential for interpreting current observations.

Power Spectrum Suppression

The resulting suppression exceeds the percent level at k ≳ 0.3 h Mpc−1, reaching 20–25% at k = 5 h Mpc−1, in line with strong-feedback predictions from the FLAMINGO fgas–8σ run This result clearly deviates from the fiducial FLAMINGO m8 simulation, and points towards strong baryonic feedback effects required to jointly describe the combination of gas fractions from eROSITA and kSZ observations. p. 4

Where the paper says it

3 h/Mpc, reaching up to 25% at k = 5 h/Mpc, in line with strong-feedback hydrodynamical simulations, such as the fgas − 8σ FLAMINGO simulation. Furthermore, the same model successfully reproduces the gas density and pressure profiles of massive clusters in the X-COP sample, reinforcing its applicability across mass scales and observables.

Tension with Older Data

A joint fit using pre-eROSITA gas fractions leads to significant tension with the kSZ signal, favoring parameter values that are incompatible with the suppression seen at low masses This discrepancy may be driven by selection biases in the older X-ray datasets, which predominantly include gas-rich systems, though differences in halo mass estimation may also play a role. The pre-eROSITA joint fit overshoots the gas fractions from the fgas–8σ model across most mass scales, while aligning more closely with the m8 simulation in the mass range covered by the X-ray data

Parameter Dependence

The parameter dependence of Mc shows that a lower Mc value leads to steeper gas profiles at fixed halo mass A sharper transition can lead to more pronounced suppression or enhancement of gas at certain masses A small δ results in a flatter profile that extends to large radii, redistributing more gas to the outskirts and lowering the central density A larger δ values steepen the outer profile, concentrating more gas inward and increasing both the kSZ signal and the gas fraction.

Final Constraint Summary

The joint fit to the kSZ signal and X-ray gas fractions constrains both the amplitude and shape of the gas distribution, with the kSZ data playing a crucial role in limiting the gas content at low masses The joint fit closely traces the gas fraction data and remains within 1σ of all measurements, though it tends to lie on the lower edge at low masses. The inclusion of kSZ data shifts the gas fraction fit downward, particularly in the mass range around M200,ksz ≈ 3×1013M⊙ (M500 = 2. p. 4

Where the paper says it

In this work, we build upon the updated BFC model introduced and validated in Paper I [1], where it was shown to reproduce the matter power spectrum across a variety of feedback scenarios and redshifts by fitting to the ratios of gas and stellar mass profiles in the hydrodynamical simulations FLAMINGO [6, 39] and IllustrisTNG [10]. Here, we apply the updated BFC model to jointly fit the kSZ signal from the Atacama Cosmology Telescope (ACT) [40] and X-ray gas fraction measurements from both the recent eROSITA observations [41] and a compilation of pre-eROSITA datasets [42–44].

1×1013M⊙). The pre-eROSITA joint fit exhibits a clear downturn in the gas fraction, while the individual X-ray-only fit continues to predict comparatively high values. p. 4
Where the paper says it

In this work, we build upon the updated BFC model introduced and validated in Paper I [1], where it was shown to reproduce the matter power spectrum across a variety of feedback scenarios and redshifts by fitting to the ratios of gas and stellar mass profiles in the hydrodynamical simulations FLAMINGO [6, 39] and IllustrisTNG [10]. Here, we apply the updated BFC model to jointly fit the kSZ signal from the Atacama Cosmology Telescope (ACT) [40] and X-ray gas fraction measurements from both the recent eROSITA observations [41] and a compilation of pre-eROSITA datasets [42–44].

The inclusion of kSZ data induces a downturn in fgas,500, reflecting the suppression of gas content required by the kSZ signal. p. 2
Where the paper says it

1 Comparison of 3D Profiles to FLAMINGO and X-COP 20 6. 2 Power spectrum suppression 23 7 Conclusions 24 A Two-halo term and halo exclusion 27 B Parameter dependence of kSZ and fgas,500 27 C Parameter reduction 29 D Posterior Distributions from Joint Fits to kSZ and X-ray Gas Fractions 30 E pre-eROSITA power spectrum suppression 32 – 1 –

The results confirm that the eROSITA gas fractions are more consistent with a strong feedback scenario, while the pre-eROSITA results align better with the m8 feedback strength in FLAMINGO. p. 4
Where the paper says it

Our results show that the combination of kSZ observations from ACT and X-ray gas fractions from eROSITA yields robust and self-consistent constraints on baryonic feedback when modeled with the updated component-wise BFC framework. We find that both datasets are mutually consistent and jointly favor a strong-feedback scenario, despite probing different mass, redshift, and radial regimes (e.

p. 4
Where the paper says it

Our results show that the combination of kSZ observations from ACT and X-ray gas fractions from eROSITA yields robust and self-consistent constraints on baryonic feedback when modeled with the updated component-wise BFC framework. We find that both datasets are mutually consistent and jointly favor a strong-feedback scenario, despite probing different mass, redshift, and radial regimes (e.

Model Robustness

The BFC framework provides a powerful, selfconsistent model of baryonic feedback across mass scales and observables, including gas fractions, SZ effects, and matter clustering Its predictive accuracy and physical interpretability make it highly promising for application to additional cosmological observables such as thermal SZ or X-ray observations. The model reproduces both the low-mass and high-mass gas density profiles without being explicitly fitted to the X-COP density data, highlighting its physical consistency and predictive power across halo masses The results show that the BFC framework provides a powerful, selfconsistent model of baryonic feedback across mass scales and observables, including gas fractions, SZ effects, and matter clustering. p. 1

Where the paper says it

Finally, we compare our best-fitting model to the observed gas density and pressure profiles of massive galaxy clusters from the X-COP sample, finding excellent agreement. These results show that BFC provides a self-consistent picture of feedback across mass- and length scales as well as different cosmological observables, thus making it promising for applications to multiwavelength studies to jointly constrain cosmology and baryonic effects.

Power Spectrum Prediction

The suppression of power predicted by our model matches the fgas–8σ FLAMINGO simulation across a wide range of scales, while disfavoring the m8 model based on multiple independent observables This result clearly deviates from the fiducial FLAMINGO m8 simulation, and points towards strong baryonic feedback effects required to jointly describe the combination of gas fractions from eROSITA and kSZ observations

Parameter Reduction Summary

The parameter reduction procedure leaves four free parameters (BFC3 + M200,kSZ): Mc, µ, and δ, which shape the gas profile, and the effective halo mass M200,kSZ These remaining parameters are well constrained by the joint dataset and are not significantly affected by fixing the others. The final model (dark blue, BFC3 +M200,kSZ) retains only Mc, µ, δ, and the effective halo mass. p. 5

Where the paper says it

1) dbar(ri,bar|M200, c200, θbfc) = rf,bar(Mbar) − ri,bar(Mbar), (2. 2) where ri and rf are the inverted cumulative mass profiles of the initial (gravity-only) and final (baryonified) states, M200 is the halo mass4 , c200 is the halo concentration, and θbfc represents BFC model parameters.

Parameter Dependence Summary

The parameter dependence of Mc shows that a lower Mc value leads to steeper gas profiles at fixed halo mass A sharper transition can lead to more pronounced suppression or enhancement of gas at certain masses A small δ results in a flatter profile that extends to large radii, redistributing more gas to the outskirts and lowering the central density A larger δ values steepen the outer profile, concentrating more gas inward and increasing both the kSZ signal and the gas fraction.

Final Synthesis

Taken together, our results show that the BFC framework provides a powerful, selfconsistent model of baryonic feedback across mass scales and observables, including gas fractions, SZ effects, and matter clustering This is in line with previous work (e.g. [36, 38]) and it will be interesting to compare it with recent and upcoming weak lensing results The BFC framework provides a powerful, selfconsistent model of baryonic feedback across mass scales and observables, including gas fractions, SZ effects, and matter clustering. p. 1

Where the paper says it

Finally, we compare our best-fitting model to the observed gas density and pressure profiles of massive galaxy clusters from the X-COP sample, finding excellent agreement. These results show that BFC provides a self-consistent picture of feedback across mass- and length scales as well as different cosmological observables, thus making it promising for applications to multiwavelength studies to jointly constrain cosmology and baryonic effects.

Power Spectrum Suppression Summary

The suppression of power predicted by our model matches the fgas–8σ FLAMINGO simulation across a wide range of scales, while disfavoring the m8 model based on multiple independent observables This result clearly deviates from the fiducial FLAMINGO m8 simulation, and points towards strong baryonic feedback effects required to jointly describe the combination of gas fractions from eROSITA and kSZ observations. The suppression of power predicted by our model matches the fgas–8σ FLAMINGO simulation across a wide range of scales, while disfavoring the m8 model based on multiple independent observables

Model Validation

The results show that the BFC framework provides a powerful, selfconsistent model of baryonic feedback across mass scales and observables, including gas fractions, SZ effects, and matter clustering The model reproduces both the low-mass and high-mass gas density profiles without being explicitly fitted to the X-COP density data The results show that the BFC framework provides a powerful, selfconsistent model of baryonic feedback across mass scales and observables, including gas fractions, SZ effects, and matter clustering.

Parameter Constraints

The joint fit to the kSZ signal and X-ray gas fractions constrains both the amplitude and shape of the gas distribution The joint fit closely traces the gas fraction data and remains within 1σ of all measurements, though it tends to lie on the lower edge at low masses The joint fit closely traces the gas fraction data and remains within 1σ of all measurements, though it tends to lie on the lower edge at low masses. The joint fit closely traces the gas fraction data and remains within 1σ of all measurements, though it tends to lie on the lower edge at low masses.

Final Conclusion

The combination of kSZ and eROSITA data yields self-consistent constraints on baryonic feedback strength The joint fit simultaneously recovers the amplitude and shape of the gas fraction–mass relation and constrains the radial gas distribution, particularly at low halo masses where the kSZ signal provides unique sensitivity The results confirm that the eROSITA gas fractions are more consistent with a strong feedback scenario, while the pre-eROSITA results align better with the m8 feedback strength in FLAMINGO.

Power Spectrum Prediction Summary

The suppression of power predicted by our model matches the fgas–8σ FLAMINGO simulation across a wide range of scales, while disfavoring the m8 model based on multiple independent observables This result clearly deviates from the fiducial FLAMINGO m8 simulation, and points towards strong baryonic feedback effects required to jointly describe the combination of gas fractions from eROSITA and kSZ observations The suppression of power predicted by our model matches the fgas–8σ FLAMINGO simulation across a wide range of scales, while disfavoring the m8 model based on multiple independent observables The suppression of power predicted by our model matches the fgas–8σ FLAMINGO simulation across a wide range of scales, while disfavoring the m8 model based on multiple independent observables. The suppression of power predicted by our model matches the fgas–8σ FLAMINGO simulation across a wide range of scales, while disfavoring the m8 model based on multiple independent observables

Parameter Constraints Summary

The joint fit to the kSZ signal and X-ray gas fractions constrains both the amplitude and shape of the gas distribution The joint fit closely traces the gas fraction data and remains within 1σ of all measurements, though it tends to lie on the lower edge at low masses. The joint fit closely traces the gas fraction data and remains within 1σ of all measurements, though it tends to lie on the lower edge at low masses. The joint fit closely traces the gas fraction data and remains within 1σ of all measurements, though it tends to lie on the lower edge at low masses.

The combination of kSZ and eROSITA data yields self-consistent constraints on baryonic feedback strength The joint fit simultaneously recovers the amplitude and shape of the gas fraction–mass relation and constrains the radial gas distribution, particularly at low halo masses where the kSZ signal provides unique sensitivity. The results confirm that the eROSITA gas fractions are more consistent with a strong feedback scenario, while the pre-eROSITA results align better with the m8 feedback strength in FLAMINGO.

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. Improved AI can perform self-consistent modeling of Dark Matter and baryons by displacing particles in gravity-only simulations according to analytic, physically motivated gas and stellar profiles, enabling fast predictions of baryonic effects across different cosmologies and feedback scenarios.

  2. The improved system can jointly constrain cosmology and baryonic effects using a component-wise baryonification (BFC) approach to fit kinematic Sunyaev-Zel’dovich (kSZ) observations from ACT alongside halo X-ray gas fractions from eROSITA, which suggests a feedback model stronger than assumed in many hydrodynamical simulations.

  3. The AI system can predict the matter power spectrum suppression based on the best-fit BFC parameters, finding that this suppression exceeds the percent level at k equals 0.3 to 0.6 h Mpc-1 and reaches 20 to 25 percent at k equals 5 h Mpc-1, consistent with strong-feedback hydrodynamical simulations.

  4. The system can provide robust constraints on the gas distribution by jointly fitting kSZ signal and X-ray gas fractions, recovering a smooth step function in the gas fraction mass relation that reflects mass-dependent efficiency of AGN feedback.

  5. The AI can accurately reproduce 3D gas density and pressure profiles for massive galaxy clusters across different scales, showing consistency with observations from the X-COP sample, which supports the validity of the model across halo masses and observables.

  6. The system can identify tensions between observational datasets by comparing joint fits using different gas fraction samples; for instance, it notes that a joint fit using pre-eROSITA gas fractions leads to significant tension with kSZ signal, favoring parameter values incompatible with the suppression seen at low masses.

Abstract

Baryonic feedback alters the matter distribution on small and intermediate scales, posing a challenge for precision cosmology. The new, component-wise baryonification (BFC) approach provides a self-consistent framework to model feedback effects for different observables. In this paper we use this framework to fit kinematic Sunyaev-Zel'dovich (kSZ) observations from the Atacama Cosmology Telescope (ACT) alongside halo X-ray gas fractions from eROSITA, investigating baryonic feedback in a cosmological context. We first show that the kSZ data from ACT is consistent with the gas fractions from eROSITA, both suggesting a feedback model that is stronger than what is assumed in most hydrodynamical simulations. This finding is in contrast to older, pre-eROSITA gas fraction measurements that point towards weaker feedback in tension with the kSZ results. We suspect these discrepancies to be due to selection bias in the pre-eROSITA sample, or differences in halo mass estimation between the two data sets. In a further step, we use the BFC model to predict the baryonic suppression of the matter power spectrum. Based on our combined fit to data from ACT and eROSITA, we find a power spectrum suppression that exceeds the percent-level at modes above k=0.3-0.6,h, Mpc-1, growing to 2-8 percent at k=1,h, Mpc-1, and to 20-25 percent at k=5,h, Mpc-1, consistent with strong-feedback hydrodynamical simulations. Finally, we compare our best-fitting model to the observed gas density and pressure profiles of massive galaxy clusters from the X-COP sample, finding excellent agreement. These results show that BFC provides a self-consistent picture of feedback across mass- and length scales as well as different cosmological observables, thus making it promising for applications to multiwavelength studies to jointly constrain cosmology and baryonic effects.

Sources

Related papers