Skip to main content
Sandbox Physics

Fast Radio Burst Dispersion and the Cosmic Baryon Census

How can the delay of a radio burst count the ordinary matter between galaxies? Free electrons delay the low frequencies of a radio burst more than the high ones, so the sweep counts the electrons along the path. Bursts from galaxies with known redshifts gain electrons with distance at the rate expected if the ordinary matter missing from galaxies is thin ionised gas between them. Each sightline crosses different halos, so one burst’s delay is not a precise distance.

Drag a sightline through halos and intergalactic gas, watch a radio pulse disperse, and compare simulated dispersion with real localized bursts.

Interactive modelFast Radio Burst Dispersion and the Cosmic Baryon Census
Cosmic dispersion on this sightline516 pc cm−3516\,\mathrm{pc\,cm^{-3}}
Average over all sightlines449 pc cm−3449\,\mathrm{pc\,cm^{-3}}
Gas-filled halos crossed66
Total a telescope measures663 pc cm−3663\,\mathrm{pc\,cm^{-3}}
Redshift if dispersion were distancez=0.64z=0.64

Dispersion against redshift

Dispersion collected along this sightline

How far sightlines scatter at this redshift

What each mark is
  • Pink diamonds: real observations, the localized ASKAP bursts of Macquart et al. (2020), table 1, with 80 removed for the Milky Way. The open diamond is FRB 190611, whose host is uncertain.
  • Teal band and blue dots: simulation. The band holds 90 percent of 1200 random sightlines through the teaching model; each dot adds a random host galaxy.
  • Dashed curves: the ideal mean relation, computed separately from the simulation and never fitted to it.
  • Gold star and gold lines: the sightline on the stage. It is one sightline in one fixed random universe, not a measured burst.

A radio pulse counts the electrons it passes

BackgroundStars, cold gas and the hot gas seen in X-rays add up to only part of the ordinary matter that the early universe implies. The rest was long suspected to be thin ionised gas around and between galaxies, too faint to see directly.

Why it mattersA fast radio burst crosses all of that gas. Free electrons slow low frequencies more than high ones, and the size of that sweep, the dispersion measure, is the number of electrons per unit area along the path. Once the host galaxy’s redshift is known, the bursts become a census.

Focus question
Why do bursts at the same redshift show different dispersion, and why does that not spoil the census?
One-sentence intuition
The average dispersion at each redshift fixes how much ionised gas there is; the scatter around it records how clumpy that gas is.

Core mathematical model

The delay grows as the inverse square of frequency

Δt≃4.15 ms(DMpc cm−3)[(νloGHz)−2−(νhiGHz)−2],DM=∫ne dl\Delta t\simeq 4.15\,\mathrm{ms}\left(\frac{\mathrm{DM}}{\mathrm{pc\,cm^{-3}}}\right)\left[\left(\frac{\nu_{\mathrm{lo}}}{\mathrm{GHz}}\right)^{-2}-\left(\frac{\nu_{\mathrm{hi}}}{\mathrm{GHz}}\right)^{-2}\right],\qquad \mathrm{DM}=\int n_e\,\mathrm{d}l

The pulse’s colours on the stage separate by an amount set by the dispersion collected so far. Real delays are seconds across a radio band; the stage only shows the order: low frequencies arrive last.

The mean relation

⟨DMcosmic⟩(z)=∫0zc nˉe(z′)H0(1+z′)2E(z′) dz′,nˉe=fd ρb(z)mp(1−Y2)\langle\mathrm{DM}_{\mathrm{cosmic}}\rangle(z)=\int_0^{z}\frac{c\,\bar n_e(z')}{H_0(1+z')^2E(z')}\,\mathrm{d}z',\qquad \bar n_e=\frac{f_{\mathrm d}\,\rho_b(z)}{m_p}\left(1-\frac{Y}{2}\right)

Here ρb∝(1+z)3\rho_b\propto(1+z)^3 is the cosmic baryon density, Y=0.25Y=0.25 the helium mass fraction and fdf_{\mathrm d} the share of baryons in diffuse ionised gas. It is equation 2 of Macquart et al. (2020); the dashed curves in every chart are this formula.

What a telescope measures

DMFRB=DMMW,ISM+DMMW,halo+DMcosmic+DMhost1+z\mathrm{DM}_{\mathrm{FRB}}=\mathrm{DM}_{\mathrm{MW,ISM}}+\mathrm{DM}_{\mathrm{MW,halo}}+\mathrm{DM}_{\mathrm{cosmic}}+\frac{\mathrm{DM}_{\mathrm{host}}}{1+z}

The host term is divided by 1+z1+z because the burst’s frequencies were higher when it crossed its own galaxy, and the delay scales as ν−2\nu^{-2}. The Milky Way terms are fixed at 30 and 50 in the Lab, as in the paper.

The scatter shrinks slowly with distance

Δ=DMcosmic⟨DMcosmic⟩,σΔ≈F z−1/2\Delta=\frac{\mathrm{DM}_{\mathrm{cosmic}}}{\langle\mathrm{DM}_{\mathrm{cosmic}}\rangle},\qquad \sigma_{\Delta}\approx F\,z^{-1/2}

A longer path averages over more independent structures, so the relative spread falls roughly as one over the square root of redshift. FF grows when more gas stays in halos, which is why feedback can be read from the scatter.

Common difficulties

Dispersion is not a distance

Typical misconceptionA larger dispersion measure always means a farther burst.

Better mental modelThe host galaxy and any halo on the path add electrons that have nothing to do with distance. In the paper’s sample, FRB 190608 at redshift 0.118 carries as much dispersion as the mean relation gives at about 0.3, because its host contributes a lot. The redshift readout shows the error you make by ignoring this.

One burst cannot weigh the universe

Typical misconceptionIf a burst sits above the mean curve, the universe must have more baryons.

Better mental modelA single sightline can cross a cluster and sit far above the mean, or thread voids and sit below it. Only the average over many bursts, fitted together with the scatter and the host term, constrains the baryon density; that is why Macquart et al. fit all bursts jointly.

Run the experiment

  1. 01

    Watch one pulse

    Leave the default running and follow the pulse from the host to the Milky Way while watching the middle chart.

    What to observe: The teal area climbs steadily through the intergalactic gas; each gas-filled halo the pulse crosses adds an orange step. The pulse’s low-frequency end falls further behind.
  2. 02

    Steer the sightline

    Drag the host up and down, then try Straight through a cluster and Threading the voids.

    What to observe: The same redshift gives about 0.8 to 1.6 times the average. The gold line in the histogram jumps across the spread of all simulated sightlines.
  3. 03

    Change the baryon budget

    Choose Half the baryons missing and compare the dashed curve with the pink ASKAP points; then choose Feedback empties the halos.

    What to observe: With half the diffuse gas, the mean falls well below the real bursts. Emptying halos keeps the mean but narrows the band, because the gas spreads more evenly.

Sources and model scope

The mean relation, the Milky Way terms, the real bursts and the scatter form follow Macquart et al. (2020), equation 2, table 1 and methods. The halo gas profile and cluster abundance follow Prochaska and Zheng (2019), sections 2 and 6. The larger DSA-110 census of Connor et al., 69 localized bursts in its arXiv version 2, is discussed in the reading guide. An independent Python script recomputes the mean, the scatter and the published fit and saves its inputs, outputs and failure cases.

Study the FRB baryon-census course: from one delayed pulse to the missing matter

Related experiments and articles

What this page links to directly

See every article series on the content map →

Fast radio bursts and the cosmic baryon census · Reading path

From one delayed pulse to the missing baryons

Articles on this path that use this experiment, followed by the other experiments they suggest.

  1. ArticleHow do fast radio bursts count the ordinary matter between galaxies?
  2. ArticleHow free electrons delay a radio burst
  3. ArticleAdding up electrons across an expanding universe
  4. ArticleWhy every sightline gives a different dispersion
  5. ArticleRecompute the localized-burst baryon census
  6. ArticleReading the FRB baryon census papers
Something broken, confusing, or missing on this page?Tell us
What kind of feedback?

Sent with this page’s address, its Lab ID, the site language and your device type. Please leave out personal details. Feedback is kept for up to 12 months.Privacy policy