A large fraction of ordinary gas in the Universe shines so faintly that telescopes can barely detect it directly. A study published in the journal Nature Astronomy therefore takes an indirect approach: it measures how this gas delays brief radio signals from distant galaxies on their journey to Earth. Astrophysicist Kritti Sharma, a doctoral student at the California Institute of Technology (Caltech), and her team evaluated 114 so-called fast radio bursts. Within their modeling framework, the data suggest that gas movements moderately smooth the distribution of matter. Several scenarios in which galaxies expel their gas particularly far outward are less consistent with the measurements.

The study, titled “Signatures of suppressed matter clustering revealed by fast radio bursts,” was published on September 8, 2026. Science writer Igor Danilov reported on it on September 30, 2026, in the online magazine ALLATRA Media. This article is based on his reporting.


What Fast Radio Bursts Are and Why Their Unexplained Origin Does Not Hinder Gas Measurements

Fast radio bursts, commonly abbreviated as FRBs, are brief pulses of radio emission. The bursts examined in the study originate in distant galaxies. Such signals often last only milliseconds. Where exactly they come from remains not fully understood to this day.

An important clue came from a radio burst recorded in the Milky Way in 2020. It originated from the magnetar SGR 1935+2154. Magnetars are neutron stars with extremely powerful magnetic fields. This observation is considered strong evidence that at least some radio bursts can be produced by such objects. In addition, researchers are exploring hypothetical scenarios in which compact objects merge, such as two neutron stars or a neutron star and a black hole.

The fact that the exact origin of the bursts remains open does not hinder their use as a measuring tool. The reason: how radio waves propagate through plasma is physically far better understood than the processes that trigger the bursts themselves. Plasma is a gas in which electrons have detached from their atoms. To measure how this gas has altered the signal, a complete theory of the burst’s origin is not required. An unexplained phenomenon thus serves as a tool to investigate another, difficult-to-access one.


Why Cosmologists Need to Know How Ordinary Gas Is Distributed in Space to Test Dark Matter

The question of gas is more than just a detail for cosmology, the science of the structure and evolution of the Universe. It touches upon how researchers test their standard model of the Universe.

Galaxies are not evenly distributed in space. They form groups and clusters strung along long filaments. In between lie vast regions containing little matter. This structure is called the cosmic web. For cosmologists, what matters is not only how filaments and nodes are arranged, but also how strongly the density of matter fluctuates across different spatial scales—that is, in small as well as very large regions of space.

According to today's understanding, supported in part by measurements of the cosmic microwave background—a faint remnant radiation from the early Universe—matter in the early Universe was distributed far more evenly. Gravity amplified small differences over billions of years: denser regions attracted more matter, while sparser regions became even emptier. How strongly matter is clumped today therefore allows inferences about what happened in the past.

In the standard cosmological model, dark matter accounts for most of the matter. Its nature is unknown, but its gravity helps explain how galaxies move, how light is deflected, and how cosmic structures grow. Dark energy explains in this model why the expansion of the Universe is accelerating. The history of this expansion in turn influences how quickly concentrations of matter grow. Researchers compute in computer simulations how matter evolves under these assumptions and compare the results with observations.

Since most matter does not emit light, astronomers use weak gravitational lensing. Mass along the path of light slightly distorts the images of distant galaxies, as described by Einstein's general theory of relativity. From correlated distortions of many galaxies, one can statistically infer how mass is distributed in between. Cosmological surveys with the Euclid space telescope and the ground-based Vera C. Rubin Observatory build upon such measurements.

This is where ordinary gas comes into play. If it moves, it alters the mass distribution captured by lensing measurements. The same observed clustering can then be interpreted differently: as a consequence of specific cosmological properties or as a result of specific gas behavior. To cleanly determine the effects of dark matter and dark energy, researchers therefore need to know how strongly gas movements alter matter clustering. This is precisely the quantity that Sharma's team sought to constrain.


How Stars and Active Galactic Nuclei Expel Gas from Galaxies and Smooth the Distribution of Matter

Gas is held by the gravity of a galaxy and its halo. The halo is an extended envelope surrounding the galaxy, consisting primarily of dark matter. When gas flows into a galaxy, it can cool, condense, and form stars.

Massive stars then affect their surroundings: through radiation, stellar winds (streams of particles from their surfaces), and supernova explosions. A second source of energy lies in the center of many galaxies. Matter falling toward the supermassive black hole located there heats up, radiates, and can drive powerful gas outflows. Such a center is called an active galactic nucleus.

Both processes can push gas far out of the dense central regions. The gas does not necessarily have to leave the galaxy. Some of it falls back later, while another part travels beyond the halo. Astronomers call this influence of stars and active nuclei on gas “feedback.”

For the galaxy itself, this means that heated or displaced gas is less available for new stars. For cosmology, another effect is what matters: when gas is dispersed over a larger volume, the density contrast between the center and its surroundings decreases. Because gas accounts for part of the total mass and is gravitationally coupled to the rest of the matter, matter on the affected scales becomes distributed more smoothly overall.

How pronounced this effect is depends on how a model describes the energy transfer to the gas. The simulations IllustrisTNG and SIMBA, for example, handle processes around active galactic nuclei differently. To make the effect visible, researchers compare calculations including gas physics with calculations where only gravity operates, under otherwise identical initial conditions.


How Free Electrons in Gas Delay Lower Frequencies of Fast Radio Bursts

A radio burst consists of waves of different frequencies—that is, different numbers of oscillations per second. When it traverses ionized gas—gas in which electrons have detached from atoms—the free electrons slow down lower-frequency components more than higher-frequency ones. The lower frequencies therefore arrive at Earth later. The more free electrons lie along the path, the greater the time difference under otherwise identical conditions.

From this difference, astronomers determine the dispersion measure, abbreviated as DM. It indicates how many free electrons the signal has traversed in total, taking the expansion of the Universe into account. Where exactly along the path each gas cloud was located is not revealed by this value.

To infer the amount of gas from the electrons, researchers take into account the composition of the gas and its ionization state—that is, how many electrons the atoms have lost. The advantage of this method: even very diffuse gas leaves a measurable trace without needing to shine brightly itself. Two limitations remain: the temperature of the gas cannot be determined this way alone, and neutral matter without free electrons is not directly detected by this method.

The delay becomes even more informative when the distance to the source is known, since a longer path generally passes through more gas. To this end, researchers identify the galaxy in which the burst originated and measure its redshift. The expansion of the Universe stretches light waves. Within the adopted cosmological model, the extent of this stretching yields a distance and an epoch in cosmic history.

That this relationship can be utilized was already demonstrated in 2020 by a team led by radio astronomer Jean-Pierre Macquart. Using the dispersion measure and redshift, they determined the average density of ordinary matter in the Universe, known technically as baryonic matter. The value agreed with independent cosmological measurements.


Why the Scatter in Delays Helps Assess How Strongly Gas Clusters Around Galaxies

Sharma's team went a step further. They were interested not only in how much gas exists overall, but in how it is distributed in space. To this end, the researchers compared signals arriving from similar distances but from different directions.

The reasoning behind this: if gas clusters into isolated dense clumps, some signals encounter a large amount of gas while others encounter very little. The delays then scatter widely. If gas is distributed more evenly across a large volume, the differences are smaller under otherwise identical conditions. Researchers therefore pay attention not only to the average delay, but also to how much it fluctuates from burst to burst. Comparing this scatter with model calculations allows them to assess how concentrated the gas is.

Astrophysicist Matthew McQuinn described the potential of this approach back in 2014. In 2025, Sharma and colleagues used simulations to investigate how the gas distribution relates to the statistics of the delays. In the new study, the team applied this relationship to real observations.


How the Team Led by Kritti Sharma Compared Fast Radio Burst Signals with Models of Gas Distribution

The data come from several radio telescopes and observing systems, including DSA-110, ASKAP, CHIME/FRB, and MeerKAT. Only bursts that could be reliably associated with a host galaxy and whose redshift had been measured spectroscopically—via precise dispersion of light—were included in the sample. This selection shapes the result: which bursts a telescope detects at all, and for which of them the origin can be identified, influences the analysis.

Not all of the delay arises in the space between galaxies. Part of it is due to gas in the Milky Way, and another part to gas in the host galaxy and the immediate environment of the source. The researchers estimated these contributions using physical models. For the host galaxies, they fitted a statistical distribution of contributions. How accurate these corrections are remains one of the method's weaknesses.

For the gas around galaxies and galaxy groups, the researchers assumed various density profiles. These describe how gas density changes with distance from the center. One profile description was calibrated to match hydrodynamic simulations—computer models simulating how gas flows. Another allowed the shape of the gaseous envelopes to vary freely. For each variant, they calculated how delays should be distributed for sources at different distances and compared this with measurements.

The procedure had two stages. First, the team determined which gas distributions were consistent with the measured delays. Then, they calculated how strongly these gas distributions reduce the clustering of all matter across various scales. A calculation using the same cosmological parameters in which only gravity operates served as a benchmark. Only this second stage connects the radio measurements to the cosmological research question.

Additional assumptions were required for this. In the analysis of the preprint, researchers held cosmological parameters fixed and adopted external estimates of stellar mass. These help determine how much ordinary matter remains in the form of gas.


Radio Burst Study: Gas Movements Moderately Smooth Matter Distribution Under Model Assumptions

An initial version of the study appeared in April 2026 as a preprint—a scientific pre-publication. It included 109 radio bursts. In the journal version published in September, the sample grew to 114. The detailed results reported by Danilov stem from the analysis of the preprint.

According to this, under the adopted model assumptions, the observations point to a moderate smoothing of the matter distribution. This effect was examined for density fluctuations of various sizes, from large-scale to smaller structures.

Several simulations in which gas is dispersed particularly strongly are less consistent with the measurements. For the most extreme scenarios, the discrepancy was on the order of two standard deviations. The standard deviation is a statistical measure of the dispersion of values. A discrepancy of this magnitude, according to Danilov's account, is not sufficient to definitively rule out these scenarios.

In the most flexible description of the gas profiles, the data primarily constrain a combination of multiple parameters rather than each property of the gas individually. The study thus narrows the range of possible gas distributions. It does not yet provide a detailed map of individual gas clouds.

The scope is also limited. Today's sample is primarily sensitive to gas in galaxy groups and clusters in the comparatively nearby Universe, even though the bursts themselves can originate from great distances. The result therefore cannot be generalized to all galaxies and all epochs.


Radio Bursts Yield More Gas in Galaxy Groups and Clusters Than Some eROSITA X-ray Estimates

A comparison with other measurement techniques reveals a discrepancy. In the preprint analysis, a higher gas fraction was found for groups and clusters than in some estimates based on data from the eROSITA X-ray telescope.

Part of the explanation could lie in the differing sensitivities of the methods. The X-ray signal depends particularly strongly on hot, dense gas. In contrast, the delay of radio waves also detects diffuse ionized gas. X-ray observations might therefore not fully capture cooler, more diffuse gas. This remains a hypothesis so far: the radio method alone does not measure temperature, and the discrepancy can also depend on the models and the selection of objects. Systematic measurement errors are also possible.

Precisely because the methods have different sensitivities and are subject to different sources of error, they can cross-check each other. The team demonstrated that under the chosen model assumptions, just around a hundred localized radio bursts achieve statistical precision comparable to several constraints from X-ray and microwave data. Radio bursts thus become a complementary tool to examine the distribution of gas around galaxies.


Result Passes Robustness Test; A Second Radio Burst Study Finds Stronger Gas Displacement

In the preprint, the authors tested whether a single unusual radio burst was driving the result. They removed each source one by one and repeated the calculation. The conclusion remained stable. In addition, they tested a scenario in which the contribution of host galaxies changes with cosmic epoch.

Another study, however, arrived at a different conclusion. Cosmologists Robert Reischke and Steffen Hagstotz also estimated the strength of feedback from a sample of roughly 100 localized radio bursts. Their paper was published in June 2026 in the Open Journal of Astrophysics. They used a model tuned to the computer simulations named BAHAMAS. Within this model, the observations favored fairly strong feedback.

What accounts for the difference remains open. A fair comparison must take into account the samples, the limitations of the models, and the assumptions regarding gas near the sources and in the Milky Way. From this, Danilov formulates an open research question: Which parts of the results are determined by the observations, and which depend on the assumptions made?


How Around 10,000 Fast Radio Bursts Will Measure Galactic Gas Envelopes More Accurately

In their forecasts, Sharma and colleagues anticipate around 10,000 well-localized radio bursts that can be matched against galaxy catalogs. Such a sample requires not only more observations with radio telescopes, but also the measurement of the redshift of each host galaxy. The smaller the statistical uncertainty becomes, the more precisely researchers must also account for which signals their telescopes miss and how the properties of the sources change.

With sufficient data, astronomers will, according to Danilov, be able to compare signals passing at different distances from galaxies of similar mass. A widely extended gaseous envelope should contribute to the delay even at large distances from the center. A compact envelope should yield a different pattern. Repeating this comparison across various cosmic epochs will allow testing whether gaseous envelopes have changed over time. This would give the finding a spatial specificity that today is derived primarily from the overall scatter of the delays.

The next step would be combining this with maps of weak gravitational lensing. The radio signals provide information about free electrons, while lensing measurements reveal the overall gravitational effect of matter. A joint model must explain both observations. If the description of the gas passes this test, cosmologists will, in Danilov's assessment, be able to use the relevant scales with greater confidence to investigate the structure of the Universe.


Sources:

- Igor Danilov: “How Fast Radio Bursts Help Us Explore the Structure of the Universe”, ALLATRA Media, September 30, 2026
https://allatra.media/space/fast-radio-bursts-cosmic-gas

- Sharma K. et al. (2026): “Signatures of suppressed matter clustering revealed by fast radio bursts”. Nature Astronomy. DOI: 10.1038/s41550-026-02957-9 (doi(.)org/10.1038/s41550-026-02957-9)
- Sharma K. et al. (2026): “Signatures of Suppressed Matter Clustering revealed by Fast Radio Bursts”. Preprint with 109 sources, arXiv:2604.17162v1. DOI: 10.48550/arXiv.2604.17162 (doi(.)org/10.48550/arXiv.2604.17162)
- Sharma K. et al. (2025): “A Hydrodynamical Simulations-based Model that Connects the FRB DM–Redshift Relation to Suppression of the Matter Power Spectrum via Feedback”. The Astrophysical Journal 989, 81. DOI: 10.3847/1538-4357/adeca4 (doi(.)org/10.3847/1538-4357/adeca4)
- Reischke R., Hagstotz S. (2026): “A first measurement of baryonic feedback with Fast Radio Bursts”. The Open Journal of Astrophysics 9. DOI: 10.33232/001c.164244 (doi(.)org/10.33232/001c.164244)
- Macquart J.-P. et al. (2020): “A census of baryons in the Universe from localized fast radio bursts”. Nature 581, 391-395. DOI: 10.1038/s41586-020-2300-2 (doi(.)org/10.1038/s41586-020-2300-2)
- McQuinn M. (2014): “Locating the ‚Missing‘ Baryons with Extragalactic Dispersion Measure Estimates”. The Astrophysical Journal Letters 780, L33. DOI: 10.1088/2041-8205/780/2/L33 (doi(.)org/10.1088/2041-8205/780/2/L33)
- CHIME/FRB Collaboration et al. (2020): “A bright millisecond-duration radio burst from a Galactic magnetar”. Nature 587, 54-58. DOI: 10.1038/s41586-020-2863-y (doi(.)org/10.1038/s41586-020-2863-y)
- Gourdji K. et al. (2020): “Constraining a neutron star merger origin for localized fast radio bursts”. Monthly Notices of the Royal Astronomical Society 497, 3131-3141. DOI: 10.1093/mnras/staa2128 (doi(.)org/10.1093/mnras/staa2128)
- Pillepich A. et al. (2018): “Simulating Galaxy Formation with the IllustrisTNG Model”. MNRAS 473, 4077-4106. DOI: 10.1093/mnras/stx2656 (doi(.)org/10.1093/mnras/stx2656)
- Davé R. et al. (2019): “Simba: Cosmological Simulations with Black Hole Growth and Feedback”. MNRAS 486, 2827-2849. DOI: 10.1093/mnras/stz937 (doi(.)org/10.1093/mnras/stz937)
- Euclid Collaboration (Mellier Y. et al.) (2025): “Euclid. I. Overview of the Euclid mission”. Astronomy & Astrophysics 697, A1. DOI: 10.1051/0004-6361/202450810 (doi(.)org/10.1051/0004-6361/202450810)
- Ivezić Ž. et al. (2019): “LSST: From Science Drivers to Reference Design and Anticipated Data Products”. The Astrophysical Journal 873, 111. DOI: 10.3847/1538-4357/ab042c (doi(.)org/10.3847/1538-4357/ab042c)
- Bond J. R., Kofman L., Pogosyan D. (1996): “How filaments of galaxies are woven into the cosmic web”. Nature 380, 603-606. DOI: 10.1038/380603a0 (doi(.)org/10.1038/380603a0)