MYS0CVCM6RE3JKJ·Mystery · signal

Fast radio bursts and the origin of the one-off bursts

Psr1909 · CC BY-SA 4.0

First found: The Lorimer burst, found in archival Parkes pulsar-survey data and reported in 2007⁠[1]
Duration: Milliseconds or less⁠[1]
Distance: Extragalactic; localised hosts range from a globular cluster of the nearby galaxy M81 to redshift 1⁠[26]
Catalogued: 4,539 bursts from 3,641 sources in CHIME/FRB Catalog 2 (2018–2023), 83 of them repeating⁠[3]
First repeater: FRB 121102, 2016⁠[4]
Galactic burst: FRB 200428 from the magnetar SGR 1935+2154, 28 April 2020⁠[6]
Latest: FRB 20250316A, the brightest nearby one-off, localised to NGC 4141 with no persistent radio source⁠[12]
Status: Partly explained
Questions: 0 of 5 answered

People Duncan R. LorimerAstronomer, West Virginia University; reported the first burst Laura G. SpitlerLead author of the discovery of the first repeater

Catalog keyMYS0CVCM6RE3JKJ

A fast radio burst is a pulse of radio waves lasting a few milliseconds or less. The first was reported in 2007 by Duncan Lorimer and colleagues, who found it in archival pulsar-survey data from the Parkes telescope: a single 30 Jy burst near the Small Magellanic Cloud, whose low frequencies arrived later than its high frequencies by an amount that implied it had crossed far more intergalactic plasma than our Galaxy contains⁠[1]. The delay, the dispersion measure, is the signature of an extragalactic origin. Four more bursts in 2013 showed a population at cosmological distances⁠[2]. Since 2018 the CHIME telescope in British Columbia has found them by the thousand: its second catalogue lists 4,539 bursts from 3,641 sources⁠[3].

Two discoveries narrowed the question. In 2016 one source, FRB 121102, was seen to repeat, so whatever made it survived the event⁠[4]. In April 2020 CHIME and the STARE2 array detected a burst from SGR 1935+2154, a magnetar inside the Milky Way: about a thousand times more energetic than any earlier radio burst from a magnetar and close enough in energy to the weakest extragalactic bursts that, seen from another galaxy, it would have counted as one⁠[5]⁠[6]. At least some fast radio bursts are made by magnetars, neutron stars with surface magnetic fields above 10¹⁴ gauss⁠[6].

Most sources have been seen only once, and for these the origin is not settled. Repeater bursts are wider and narrower in frequency than one-off bursts, which may mean different sources or a different viewing geometry⁠[7]. A repeater in an old globular cluster and another in a dead elliptical galaxy do not fit magnetars born recently in supernovae⁠[8]⁠[9]. In 2025 two studies of a one-off burst found its emitting region no more than about 30,000 km across and its polarisation swinging as a pulsar's does, both pointing to emission from a neutron star's magnetosphere⁠[10]⁠[11]. The brightest nearby one-off, FRB 20250316A, has no persistent radio source of the kind seen around active repeaters⁠[12]. Of the 5 questions on this site, 0 are answered. The most likely answer to “What produces most of the one-off fast radio bursts?”, not yet confirmed: magnetars or other neutron stars formed late, by mergers of neutron stars or white dwarfs or by the collapse….

Key questionsset by the editors

Are repeating and one-off sources one population or several?No answer ahead yet

What produces most of the one-off fast radio bursts?Most likely: magnetars or other neutron stars formed late, by mergers of neutron stars or white dwarfs or by the collapse…

All 5 questions

What happened

The Lorimer burst was found in archival Parkes survey data: a burst shorter than 5 ms at 30 Jy, three degrees from the Small Magellanic Cloud, with no repeat in 90 hours of later observations⁠[1]. In 2013 Thornton and colleagues reported four more from the High Time Resolution Universe survey, all far from the Galactic plane, with dispersion measures implying redshifts of 0.5 to 1⁠[2].

Gemini North image of the host galaxy of FRB 121102Gemini Observatory, 2016. CC BY 4.0Artifact

In 2016 Arecibo observations found ten further bursts from the position of FRB 121102, the first known repeater⁠[4]. In 2017 the Very Large Array localised it to a faint galaxy, beside a compact, variable persistent radio source⁠[13]. In 2019 ASKAP and the Deep Synoptic Array prototype localised one-off bursts to massive galaxies⁠[14]⁠[15]. In 2020 CHIME found that the repeater FRB 180916.J0158+65 is active in a 5-day window of a 16.35-day cycle⁠[16], and on 28 April 2020 the magnetar SGR 1935+2154 produced FRB 200428, detected by CHIME at 400–800 MHz and by STARE2 at 1.3–1.5 GHz⁠[5]⁠[6].

Map of the worldOpen the full map

+−

CHIME at the Dominion Radio Astrophysical Observatory, near Penticton, British ColumbiaRadio telescope

Mapped

Karl G. Jansky Very Large Array, New MexicoRadio telescope array

Within 1 km

Gemini North telescope, Maunakea, HawaiiOptical telescope

Mapped

Arecibo Observatory, Puerto RicoRadio telescope (305 m dish collapsed 2020)

Within 1 km

Parkes radio telescope (Murriyang), New South WalesRadio telescope

Within 1 km

MappedWithin 1 kmNatural Earth

Timeline

Timeline

Squeezed where little happened. Each mark is a published claim, by how it stands:AcceptedMost likelyStill possibleDisputedRuled out

The accounts

There are no witnesses; the record is the published detections and their catalogues. The CHIME/FRB Collaboration's two catalogues are the largest uniform accounts of the population, the first with 536 bursts from its first year and the second with 4,539 from five years⁠[17]⁠[3]. Reviews by Petroff, Hessels and Lorimer, and by Zhang, set out what the observations established by the early 2020s and how the theories divide⁠[18]⁠[19]. A catalogue of theories assembled in 2019 records how many progenitor models were in play before the magnetar detection⁠[20].

What was recorded

A burst is recorded as a dynamic spectrum: intensity against time and frequency. From it come the dispersion measure, the width, the scattering time, the fluence and, for well-recorded bursts, polarisation. CHIME records every burst at 400–800 MHz with 0.983 ms resolution, a position to about 10 arcminutes, and the exposure and sensitivity needed to turn counts into rates⁠[3]. Its first catalogue calibrated selection effects by injecting simulated bursts into the live pipeline⁠[17]. Both catalogues are public as tables, with waterfall data, the beam model and injection data for Catalog 1⁠[21]⁠[22]⁠[23].

Localisation to host galaxies is the second record. Interferometers (ASKAP, the Deep Synoptic Array, the VLA, European VLBI and, by 2024, the CHIME Outriggers) place bursts to arcseconds or better, which gives a host galaxy and a redshift⁠[14]⁠[24]⁠[9]. With redshifts, the dispersion measure becomes a measure of the ionised gas between galaxies⁠[25]. The most distant localised burst so far is FRB 20220610A at redshift 1.016, with an energy of 2 × 10⁴² erg⁠[26].

Gemini North image of the host galaxy of FRB 180916.J0158+65Gemini Observatory, 2020. CC BY 4.0Artifact

The record for the one-off bursts includes three recent kinds of measurement. Scintillation in FRB 20221022A limited the size of its emitting region⁠[10]; its polarisation angle swung by 130 degrees in 2.5 ms in the pattern pulsars show⁠[11]. Deep radio imaging of the brightest nearby one-off, FRB 20250316A in NGC 4141 at about 40 Mpc, found no persistent radio source to a limit three orders of magnitude below those around active repeaters⁠[12], while hydrogen imaging found a recent starburst in the burst's part of the galaxy⁠[27].

The explanations

Magnetars are the leading explanation. The Galactic burst of 2020 showed that a magnetar can produce a burst of the right duration and close to the right energy⁠[5]⁠[6]. The 2025 scintillation and polarisation results place the emission of one one-off burst inside or just outside a neutron-star magnetosphere and disfavour models in which emission arises in shocks far from the source⁠[10]⁠[11]. Hosts are mostly star-forming galaxies, and massive ones are over-represented⁠[24].

How the magnetars form is the open part. A magnetar born in a core-collapse supernova is young and should sit among young stars. FRB 20200120E sits in a globular cluster of M81, whose stars are billions of years old⁠[8]; FRB 20240209A sits in a quiescent elliptical galaxy with a stellar age of about 11 billion years⁠[9]. The population as a whole appears delayed relative to cosmic star formation⁠[28]. These point to a second channel: magnetars made by neutron-star or white-dwarf mergers, or by the collapse of a white dwarf, or sources that are not magnetars at all, such as accreting binaries⁠[9].

Whether repeaters and one-offs are one population is the other open part. Repeater bursts are wider and narrower in bandwidth on average⁠[7], but the two groups share sky distribution and dispersion measures⁠[17]. If every source repeats given time, the difference could be selection; machine-learning classifiers flag dozens of apparent one-offs as likely repeaters⁠[29]. If some bursts are truly single, cataclysmic models (mergers, collapses) remain possible for them, as the earliest papers suggested⁠[1]⁠[30]. The brightest one-offs may also be a distinct class: an extreme-value analysis of Catalog 2 finds FRB 20250316A outside the bounded distribution of the rest⁠[31].

Published claims

Published claims and their status

Year	Who	Claimed	Status	Reason
2007	Duncan Lorimer and colleagues⁠[1]	A single 30 Jy millisecond burst in Parkes survey data came from outside the Galaxy, at a distance of under 1 Gpc.	Accepted	Bursts have since been localised to host galaxies out to redshift 1⁠[13]⁠[26].
2011	Sarah Burke-Spolaor and colleagues⁠[33]	Terrestrial pulses at Parkes with the same frequency sweep as the Lorimer burst cast doubt on its extragalactic origin.	Ruled out	The pulses were traced to microwave ovens that could not have made the bursts, and bursts have since been localised to other galaxies⁠[32]⁠[13].
2013	Dan Thornton and colleagues⁠[2]	Four more bursts far from the Galactic plane show a population of sources at cosmological distances.	Accepted	CHIME has since catalogued 3,641 sources, and localised hosts reach redshift 1⁠[3]⁠[26].
2015	Emily Petroff and colleagues⁠[32]	The peryton pulses at Parkes come from microwave ovens opened early, which could not have produced the fast radio bursts.	Accepted	The cause was reproduced on site, and later bursts have been localised to host galaxies⁠[32]⁠[13].
2016	Laura Spitler and colleagues⁠[4]	FRB 121102 repeats, so at least one source survives its bursts and cannot be a one-time cataclysm.	Accepted	The source has repeated many times since and was localised to a host galaxy⁠[13].
2017	Manasvi Lingam and Abraham Loeb⁠[34]	Fast radio bursts could be leakage from beams that extragalactic civilisations use to drive light sails.	Disputed	No observation tests it directly, but a magnetar was seen to make a burst in 2020 and one-off bursts have since been traced to neutron-star magnetospheres⁠[6]⁠[10].
2017	Shami Chatterjee and colleagues⁠[13]	The repeater FRB 121102 lies in a faint galaxy beside a compact, variable radio source, which settles its extragalactic origin.	Accepted	Dozens of bursts have since been placed in host galaxies by several arrays⁠[14]⁠[9].
2019	Brian Metzger, Ben Margalit and Lorenzo Sironi⁠[35]	Bursts are synchrotron maser emission from shocks where flare ejecta from a young magnetar hit material from earlier flares.	Disputed	The small emitting region and the pulsar-like polarisation swing of FRB 20221022A disfavour emission far from the star, though the model may still apply to some bursts⁠[10]⁠[11].
2020	Christopher Bochenek and colleagues⁠[6]	The burst FRB 200428 from the Galactic magnetar SGR 1935+2154 shows that magnetars can make fast radio bursts and may make the extragalactic ones.	Most likely	The Galactic burst is accepted, but sources in a globular cluster and a dead galaxy do not fit magnetars born recently in supernovae⁠[8]⁠[9].
2021	Ziggy Pleunis and colleagues⁠[7]	Bursts from repeaters are wider in time and narrower in frequency than one-off bursts, which may mean different sources.	Still possible	The difference is measured, but the two groups share sky and dispersion distributions and could differ by viewing angle or selection⁠[17]⁠[29].
2022	Franz Kirsten and colleagues⁠[8]	The repeater FRB 20200120E sits in an old globular cluster of M81, so some sources formed late, by a merger or the collapse of a white dwarf.	Most likely	The localisation holds, and a second repeater in an old elliptical galaxy points the same way⁠[9].
2025	Kenzie Nimmo and colleagues⁠[10]	Scintillation limits the emitting region of the one-off FRB 20221022A to about 30,000 km, inside or just outside a neutron star's magnetosphere.	Most likely	An independent polarisation study of the same burst reached the same conclusion, but it is one burst⁠[11].
2025	Ryan Mckinven and colleagues⁠[11]	The polarisation angle of FRB 20221022A swings by 130 degrees in the pattern of a rotating pulsar, so the emission comes from a magnetosphere.	Most likely	It agrees with the scintillation result for the same burst, though no wider sample of one-offs has been measured⁠[10].
2025	Tarraneh Eftekhari and colleagues⁠[9]	The repeater FRB 20240209A lies in a quiescent elliptical galaxy about 11 billion years old, which points to a source channel other than the collapse of massive stars.	Most likely	With the globular-cluster repeater it makes two old hosts, and the burst rate as a whole lags star formation⁠[8]⁠[28].
2025	N. Mankatwit and colleagues⁠[29]	A machine-learning classifier flags 36 apparent one-off sources in CHIME data as likely repeaters.	Still possible	The candidates have not yet been shown to repeat by follow-up observation⁠[29].
2026	Wen-Long Zhang and Jun-Jie Wei⁠[31]	FRB 20250316A lies far beyond the upper bound of the fluence distribution of the other Catalog 2 bursts, so the brightest one-offs may be a separate class.	Still possible	It rests on one outlier, which also lacks the persistent radio source seen around active repeaters⁠[12].

What would settle it

For the one-off population the decisive data are localisations by the hundred with host properties and offsets, which CHIME's Outriggers now provide, combined with long monitoring of each position to measure the fraction that repeat. A second Galactic or very nearby burst with simultaneous X-ray coverage would test the magnetar mechanism directly. A one-off burst caught with a gravitational-wave signal or a supernova would establish a cataclysmic channel; the absence of any such coincidence over many nearby bursts would weaken it. Polarisation and scintillation measurements for many one-off bursts would show whether a magnetospheric origin is the rule.