🔭 astrophysics

Deciphering Baryonic Feedback from ACT tSZ Galaxy Clusters

This paper utilizes a flexible Dark Matter + Baryon (DMB) model to analyze thermal Sunyaev-Zeldovich selected galaxy clusters from the Atacama Cosmology Telescope, demonstrating that the tSZ Y-M relation can effectively constrain baryonic feedback parameters at the percent level and favor intermediate to strong feedback scenarios consistent with hydrodynamical simulations.

Original authors: Nihar Dalal, Chun-Hao To, Chris Hirata, Tae Hyeon-Shin, Matt Hilton, Shivam Pandey, J. Richard Bond

Published 2026-03-27

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Nihar Dalal, Chun-Hao To, Chris Hirata, Tae Hyeon-Shin, Matt Hilton, Shivam Pandey, J. Richard Bond

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

The Big Picture: The Universe's "Invisible" Problem

Imagine you are trying to build a model of a city using only a satellite photo. You can see the roads and the big buildings (these are the Dark Matter halos), but you can't see the people, the cars, or the smoke coming from the factories (these are the Baryons—normal matter like gas and stars).

For a long time, cosmologists thought the city's layout was determined entirely by gravity (the roads). But recently, they realized that the "people" (gas and stars) are actually causing a lot of traffic jams and moving things around. In the universe, this is called Baryonic Feedback.

When massive clusters of galaxies form, the super-hot gas inside them gets blasted by powerful explosions from black holes and supernovae. This "feedback" pushes the gas outward, making the cluster less dense than gravity alone would predict. This changes the shape of the universe's structure, and if we don't account for it, our calculations for the age and makeup of the universe will be wrong.

The Detective Work: Using "Heat Signatures"

The authors of this paper are like detectives trying to figure out exactly how much the "people" are messing with the "city layout." They use two main tools:

The ACT (Atacama Cosmology Telescope): This telescope looks at the Cosmic Microwave Background (the afterglow of the Big Bang). When this light passes through a galaxy cluster, the hot gas leaves a specific "heat signature" called the thermal Sunyaev-Zeldovich (tSZ) effect. Think of it like seeing a steam cloud rising from a hot pot; the telescope sees the steam (the gas) even if it can't see the pot itself clearly.
DES (Dark Energy Survey): This survey uses Weak Gravitational Lensing. Imagine looking at a funhouse mirror. The mass of the galaxy cluster bends the light from galaxies behind it, distorting their shapes. By measuring how much the light is bent, scientists can weigh the cluster.

The Problem: The "Filter" Mismatch

Here is the tricky part. The ACT telescope doesn't just take a picture; it uses a mathematical "filter" to find the clusters, kind of like using a sieve to find specific sized pebbles in a bucket of sand.

The Old Way: Scientists used a standard sieve (a fixed filter) that assumed all galaxy clusters looked exactly the same.
The Reality: Because of the "feedback" (the explosions), some clusters are puffy and spread out, while others are tight and dense. If you use a standard sieve on a puffy cloud, you might miss it or measure it wrong.

The Innovation: The authors built a "Filter Mismatch Emulator."
Think of this as a virtual reality simulator. They created thousands of fake galaxy clusters with different shapes (some puffy, some tight) and ran them through their telescope's "sieve" to see how the measurements changed. This allowed them to create a custom correction for every single cluster, ensuring they weren't measuring the wrong size just because the cluster looked different than expected.

The Experiment: Weighing the Clouds

The team took 4,195 galaxy clusters found by the ACT telescope.

They measured the "heat signature" (tSZ) of each cluster.
They used the DES data to get the "true weight" (mass) of these clusters.
They compared the two.

The Logic:

If feedback is weak, the gas stays tight. The cluster looks heavy for its heat signature.
If feedback is strong, the gas is blown far away. The cluster looks "puffy" and less dense. To get the same heat signature, the cluster must actually be heavier than it looks.

By comparing the "heat" to the "true weight," they could reverse-engineer how strong the feedback explosions were.

The Results: The Universe is "Puffier" Than We Thought

The team found that the feedback is intermediate to strong.

Analogy: Imagine a campfire. If the wind (feedback) is strong, the smoke (gas) spreads out over a huge area. If the wind is weak, the smoke stays in a tight column.
Finding: The universe's galaxy clusters are like campfires in a strong wind. The gas is spread out much more than simple gravity models predicted.

This "puffiness" suppresses (lowers) the power of the universe's structure on small scales by about 5% to 30%. This is a huge deal because future telescopes (like the ones measuring the "shear" or distortion of the universe) need to know this number to avoid errors.

Why This Matters

Better Maps: By understanding how the "smoke" (gas) moves, we can make much more accurate maps of the universe's dark matter skeleton.
Checking the Simulations: The authors compared their findings to supercomputer simulations (like IllustrisTNG and Flamingo). They found their results matched some simulations well but disagreed with others (like IllustrisTNG), suggesting those simulations might need to be tweaked to account for stronger explosions.
Future Proofing: This method is a "gold standard" for future surveys. It shows that we can use galaxy clusters not just to count them, but to understand the violent physics happening inside them.

In a Nutshell

The universe is a messy place where giant explosions push gas around, changing the shape of galaxy clusters. This paper developed a new, super-smart way to measure those shapes using a "virtual reality" correction for telescope filters. They found that the explosions are stronger than some models thought, which means the universe is slightly more "puffy" than we previously calculated. This helps us build a more accurate model of how our universe grew up.

1. Problem Statement

Precision cosmology aims to constrain cosmological parameters using weak gravitational lensing (cosmic shear) on small scales ( $k > 1 \, h\text{Mpc}^{-1}$ ). However, baryonic feedback (energy injection from Active Galactic Nuclei and supernovae) significantly alters the distribution of matter in galaxy halos, suppressing the matter power spectrum by up to 30% on these scales.

The Challenge: Current analyses often discard small-scale data to avoid modeling uncertainties or rely on computationally expensive hydrodynamic simulations that are sensitive to subgrid physics prescriptions.
The Gap: There is a need for a flexible, semi-empirical model that can be directly constrained by observational data to accurately predict baryonic suppression without discarding valuable signal-to-noise.

2. Methodology

The authors developed an end-to-end pipeline to constrain baryonic feedback using a sample of galaxy clusters detected via the Thermal Sunyaev-Zeldovich (tSZ) effect.

A. Theoretical Framework: The DMB Halo Model

The study employs the Dark Matter + Baryon (DMB) halo model, a flexible semi-empirical model that decomposes the halo density profile into three components:

Intracluster Medium (Gas): Modeled as a cored double power law.
Central Galaxy (Stars): Modeled as a truncated Gaussian.
Collisionless Matter (Dark Matter + Satellite Stars): Modeled as an adiabatically contracted NFW profile.
The model includes 10 free parameters governing gas ejection, slope, stellar fractions, and non-thermal pressure. These parameters determine the gas distribution and, consequently, the suppression of the matter power spectrum.

B. Data Sources

tSZ Clusters: 4,195 galaxy clusters from the Atacama Cosmology Telescope (ACT) DR5 catalog, selected with signal-to-noise $>4$ .
Mass Calibration: The cluster masses are calibrated using Weak Lensing (WL) data from the Dark Energy Survey (DES) Year 3. This corrects for the hydrostatic mass bias (where tSZ-derived masses are typically $\sim30\%$ higher than true masses).
Data Vector: The analysis uses the mean tSZ mass ( $M_{\text{tSZ}}$ ) as a function of the fixed-filter Compton- $y$ parameter ( $y_{\text{fixed}}$ ), binned into 5 logarithmic bins.

C. Key Technical Innovation: Filter Mismatch Emulator

A critical hurdle is that ACT clusters are detected using matched filters assuming a Universal Pressure Profile (UPP). However, the DMB model predicts different density profiles.

Solution: The authors developed a Chaos Polynomial Emulator to characterize the Filter Mismatch Function ( $Q$ ).
Process: They generated mock catalogs using the DMB model, simulated the ACT observation process (including noise, beam convolution, and CMB foregrounds), and measured the ratio between the "true" integrated Compton- $y$ and the "observed" $y$ derived from the fixed filter.
Result: This allows the pipeline to forward-model the observable $y$ - $M$ relation directly from the DMB parameters, correcting for the bias introduced by using a fixed filter on varying cluster profiles.

D. Statistical Analysis

Likelihood: A Gaussian likelihood is constructed comparing the model-predicted mean $M_{\text{tSZ}}$ in each $y$ -bin to the observed data.
Mass Bias: The hydrostatic mass bias is modeled as a function of mass and redshift, constrained by external WL data (Shin et al., in prep.) and treated as informative priors.
Inference: Markov Chain Monte Carlo (MCMC) is used to sample the posterior distributions of the DMB parameters.

3. Key Contributions

End-to-End Forward Modeling: The paper presents a novel pipeline that directly models the tSZ observables (rather than just the underlying physics), explicitly accounting for the instrument's matched filter response via an emulator.
Joint Constraints: It successfully constrains the DMB halo model parameters using tSZ clusters with WL-calibrated masses, bridging the gap between cluster physics and cosmological suppression.
Robustness to Systematics: By using a fixed filter scale and an emulator for the mismatch, the method mitigates noise biases and selection function complexities that plague traditional cluster analyses.

4. Results

Feedback Strength: The analysis finds a preference for intermediate to strong levels of baryonic feedback.
Power Spectrum Suppression: The constrained suppression of the matter power spectrum at $k=1 \, h\text{Mpc}^{-1}$ is approximately 5%.
Parameter Constraints:
- The parameter $\mu_\beta$ (mass dependence of the gas profile slope) is the most tightly constrained.
- Parameters describing the gas extent ( $\theta_{ej}$ ) and pivot mass ( $M_{c0}$ ) are also well-constrained.
- While individual parameters show some degeneracy, the principal component analysis (PCA) reveals a well-constrained combination of parameters that directly dictates the power spectrum suppression.
Comparison with Simulations:
- The results are consistent with hydrodynamic simulations like Flamingo, cosmoOWLS, and Bahamas.
- There is a $3\sigma$ tension with IllustrisTNG and EAGLE simulations, which predict weaker feedback (less suppression).
Cross-Probe Consistency: The derived suppression is in impressive agreement with independent constraints from:
- WL + kSZ measurements (Bigwood et al. 2024).
- Shear $\times$ tSZ joint analyses (Pandey et al. 2025).

5. Significance and Implications

Cosmological Impact: The study demonstrates that tSZ clusters, when combined with WL mass calibration, provide competitive constraints on baryonic feedback. This allows future cosmic shear surveys (like LSST and Euclid) to utilize small-scale data without discarding signal or relying solely on simulations.
Model Validation: The DMB model proves capable of capturing the essential physics of baryonic feedback across different mass scales, validating its use as a flexible tool for cosmological inference.
Future Outlook: The authors highlight that upcoming data (ACT DR6, SPT-3G) will double cluster counts and lower mass thresholds, enabling even tighter constraints on the redshift and mass evolution of feedback. This paves the way for "self-calibrating" baryonic systematics in next-generation cosmological surveys.

In summary, this paper provides a rigorous, observationally motivated framework for decoding baryonic feedback, resolving a major systematic uncertainty in precision cosmology.