🔭 astrophysics

A first measurement of baryonic feedback with Fast Radio Bursts

This paper presents the first measurement of baryonic feedback using Fast Radio Bursts, where analysis of approximately 100 localized FRBs combined with BAHAMAS hydrodynamic simulations robustly constrains the feedback parameter and strongly rejects no-feedback scenarios at over 3 $\sigma$ confidence.

Original authors: Robert Reischke, Steffen Hagstotz

Published 2026-06-29

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Robert Reischke, Steffen Hagstotz

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 "Fog" and the Cosmic Flashlights

Imagine the universe is a giant, dark room filled with invisible fog. This fog isn't made of water, but of baryons (ordinary matter like gas and plasma) that have been heated and pushed around by powerful cosmic forces, like the supermassive black holes at the centers of galaxies.

Astronomers have long struggled to map this fog because it's hard to see. However, this paper introduces a new way to measure it using Fast Radio Bursts (FRBs). Think of FRBs as incredibly bright, millisecond-long cosmic flashlights that flash from deep space. As these flashes travel to Earth, they pass through the cosmic fog.

The Core Concept: The "Scatter" of the Signal

When a radio wave travels through this fog, it gets slowed down. The more fog (electrons) it hits, the more it slows down. This delay is called the Dispersion Measure (DM).

The Expectation: If the fog were perfectly smooth and evenly spread out, the delay would be very predictable based on how far away the flashlight is.
The Reality: The fog is "lumpy." It clumps together in some places and is thin in others. This lumpiness causes the delay to vary (scatter) more than expected.

The Analogy of the Hiking Trail:
Imagine two hikers walking the same distance.

Hiker A (No Feedback): Walks through a forest where the trees (matter) are packed tightly together. The path is very bumpy and uneven.
Hiker B (Strong Feedback): Walks through a forest where a giant wind (energy from black holes) has blown the trees apart, spreading them out evenly. The path is much smoother.

This paper asks: Is the cosmic forest bumpy (clumpy) or smooth?

What the Scientists Did

The researchers took a "snapshot" of about 100 of these cosmic flashlights (FRBs) that had been located near their home galaxies. They looked at how much the "delay" (DM) varied for each one.

They used a sophisticated computer model (calibrated against massive supercomputer simulations called BAHAMAS) to predict what the delay should look like if the universe had different amounts of "wind" (feedback) blowing the matter around.

The Key Findings

The "No Feedback" Scenario is Ruled Out:
The data strongly rejects the idea that the cosmic fog is just sitting there in its natural, clumpy state (like a "Dark Matter Only" universe). The paper claims there is a 99.7% confidence (or even higher with more data) that the fog has been smoothed out.
Strong "Cosmic Wind" is Real:
The results show that the universe is experiencing strong feedback. The energy from active galactic nuclei (supermassive black holes) is acting like a powerful fan, blowing gas out of galaxies and smoothing out the distribution of matter.
- The Result: The authors measured a specific "strength" of this wind, finding it to be quite powerful, consistent with other recent measurements using different methods (like the Sunyaev-Zel'dovich effect, which is like looking at the heat of the fog).
It's Not Just the Milky Way:
A major concern was that the fog inside our own galaxy (the Milky Way) or inside the host galaxies of the FRBs might be messing up the measurements. The team tested this by treating these local contributions as variables. They found that even if they weren't 100% sure about the local fog, it didn't change the main conclusion: the cosmic wind is strong.
The "Lumpiness" is Less Than Expected:
Because the feedback is strong, the electrons are more evenly distributed than they would be otherwise. This means the "scatter" in the radio signals is smaller than it would be in a universe without this feedback. The data matches a model where the matter distribution is significantly smoothed out.

Why This Matters

This is the first time scientists have used Fast Radio Bursts to measure this specific type of cosmic feedback.

Before: We knew feedback existed, but it was hard to pin down exactly how strong it was using these specific radio signals.
Now: We have a direct measurement confirming that the universe is "smoother" than a simple gravity-only model would predict. This helps astronomers refine their maps of the universe, which is crucial for future experiments trying to understand dark energy and the expansion of the universe.

A Note on the "Fog" Models

The authors admit that their models for the local fog (inside our galaxy and the host galaxies) aren't perfect yet. They found that for very nearby FRBs, the data was a bit harder to fit perfectly. This suggests that while the main conclusion (strong feedback) is solid, we need better maps of our own cosmic neighborhood to get even more precise numbers in the future.

In summary: By listening to the echoes of cosmic flashlights, this team proved that the universe's invisible gas has been stirred up and smoothed out by powerful cosmic forces, ruling out a quiet, clumpy universe in favor of a dynamic, energetic one.

Technical Summary: A First Measurement of Baryonic Feedback with Fast Radio Bursts

Problem Statement
Contemporary cosmological analyses, particularly weak lensing surveys (e.g., Euclid, LSST, Roman), face significant uncertainties due to baryonic feedback. Dynamic processes driven by active galactic nuclei (AGN), supernovae, and stellar winds redistribute gas within and around halos, altering the total matter distribution on small to intermediate scales. Current approaches to mitigate this, such as scale cuts, diminish the constraining power of observations. While hydrodynamical simulations (e.g., BAHAMAS) offer insights, they require calibration against observations. Existing external tracers like the thermal and kinetic Sunyaev-Zel'dovich (tSZ/kSZ) effects are beginning to constrain feedback, but a novel, direct probe is needed to break degeneracies and validate models without relying solely on CMB or X-ray data.

Methodology
This paper utilizes Fast Radio Bursts (FRBs) as a direct probe of the integrated ionized electron density along the line of sight, quantified by the Dispersion Measure (DM). The authors focus on the scatter of the DM-redshift relation (Macquart relation), which is sensitive to the "lumpiness" of the electron distribution and, consequently, baryonic feedback processes.

Modeling Framework: The authors employ a forward-modeling approach calibrated to the BAHAMAS hydrodynamic simulation suite. They utilize a one-parameter family of feedback models (HMCode2020), where the feedback strength is governed by the parameter $\log T_{\text{AGN}}$ . This parameter modifies the concentration-mass relation of halo profiles.
Statistical Likelihood: The total DM ( $D_{\text{tot}}$ $D_{tot}$ ) is decomposed into contributions from the host galaxy ( $D_{\text{host}}$ $D_{host}$ ), the Milky Way ( $D_{\text{MW}}$ $D_{MW}$ ), and the Large Scale Structure ( $D_{\text{LSS}}$ $D_{LSS}$ ).
- The $D_{\text{LSS}}$ component is modeled using a log-normal probability density function (PDF), derived from the electron power spectrum $P_e(k, z)$ , which is computed using an emulator trained on BAHAMAS simulations.
- The host and Milky Way contributions are also modeled as log-normal (for hosts) and Gaussian (for MW ISM and halo) distributions, respectively.
- The likelihood is constructed by marginalizing over the host and MW contributions.
Data Selection: The analysis uses a fiducial sample of 69 localized FRBs (host galaxy identified with independent redshift). A "full set" of 97 FRBs is also analyzed, including additional low-redshift events. To address cosmic variance correlations between sightlines, pairs with a correlation coefficient $r_{ij} > 0.3$ are identified and removed (4 pairs excluded), ensuring the assumption of independence in the likelihood calculation remains valid.
Cosmological Priors: Cosmological parameters ( $\Omega_m, \Omega_b, \sigma_8, n_s, h$ ) are fixed using correlated priors from Planck 2020 results, allowing the analysis to focus on constraining the feedback parameter and host contributions.

Key Contributions

First FRB-based Feedback Measurement: This work presents the first measurement of baryonic feedback using the statistical scatter of FRB dispersion measures.
Forward Modeling of DM Scatter: The authors demonstrate how the scatter in the DM-redshift relation directly constrains the suppression of the matter power spectrum caused by baryonic feedback. Stronger feedback smooths the electron distribution, reducing the scatter.
Robustness Checks: The study rigorously tests the impact of sightline correlations, the choice of likelihood functional form (log-normal vs. Gaussian), and the modeling of Milky Way and host contributions. It confirms that accurate calibration of host and MW contributions is critical, though current uncertainties do not dominate the feedback constraints.

Results

Feedback Constraint: Applying the model to the fiducial sample, the authors constrain the feedback parameter to $\log T_{\text{AGN}} = 7.87^{+0.16}_{-0.22}$ (68% confidence) for the full set.
Rejection of No-Feedback Scenarios: The results robustly reject the "no-feedback" scenario (where the matter distribution follows dark matter only, corresponding to $\log T_{\text{AGN}} \approx 7.3$ $lo g T_{AGN} \approx 7.3$ ).
- For the fiducial sample, the no-feedback scenario is rejected at $>99.7\%$ confidence ( $3\sigma$ ).
- For the full sample (including low-redshift FRBs), the rejection reaches $5\sigma$ .
Power Spectrum Suppression: The constraints translate to a suppression of the matter power spectrum $S(k) = P_{\text{hydro}}/P_{\text{dmo}}$ . The measured suppression is consistent with strong feedback predictions from the FLAMINGO simulations and other recent measurements (e.g., kSZ, X-ray).
Goodness of Fit: Frequentist tests ( $p$ -values) indicate that while the fiducial sample fits both feedback and no-feedback models well, the inclusion of low-redshift FRBs in the full set makes the no-feedback model a poor fit ( $p < 0.05$ ). This suggests that low-redshift modeling (specifically host and MW contributions) may require refinement as data volumes increase.

Significance and Claims
The paper claims to provide the first direct evidence using FRBs that baryonic feedback is active and relatively strong, consistent with independent tracers like the thermal and kinetic Sunyaev-Zel'dovich effects. The authors emphasize that FRBs offer a unique, direct probe of the baryon distribution that complements weak lensing and CMB studies.

The results are presented as robust against current modeling assumptions and sightline correlations. However, the authors remain modest regarding future implications, noting that:

The current constraints are consistent with, but not yet precise enough to uniquely distinguish between, all strong-feedback simulation variants (e.g., FLAMINGO vs. BAHAMAS).
Future surveys will require improved modeling of the host galaxy and Milky Way contributions, particularly for low-redshift FRBs, to avoid systematic biases.
As sample sizes grow, the treatment of correlations between FRB sightlines will become increasingly critical for unbiased cosmological inference.

In conclusion, this work establishes FRBs as a viable and powerful cosmological probe for constraining baryonic physics, marking a significant step toward precision cosmology in the era of large-scale structure surveys.