AstroparticlePhysics
Menu
Dishes of the ASKAP radio telescope in the Western Australian outback

Research area

Radio transients

Fast radio bursts and gamma-ray burst afterglows, observed in radio and followed up at very high energies.

Radio telescopes discover fast radio bursts and follow the afterglows of gamma-ray bursts from the first hour to years after the explosion. This page collects our work at that frontier.

Fast radio bursts (FRBs) are very brief yet extremely powerful flashes of radio emission. Their high luminosity, millisecond durations and large dispersion measures point to extreme, extragalactic sources, such as highly magnetised and rapidly spinning neutron stars known as magnetars. H.E.S.S. searches for their counterparts beyond the radio band through collaborations with major radio facilities, including the Parkes and UTMOST telescopes in Australia and MeerKAT in South Africa: the partners exchange alerts, and H.E.S.S. follows them up at very high energies.

Not every burst is cosmic, though. For years, Parkes also recorded strange bursts called “perytons”; the SUPERB team eventually traced them to a microwave oven in the observatory kitchen. Their paper is great fun to read; see also the story in The Conversation.

Fast radio bursts

FRB 150418 · Parkes, ATCA, H.E.S.S.

The first TeV limits on a fast radio burst

On 18 April 2015 the SUPERB team detected FRB 150418 at the Parkes telescope, in real time. The ATCA array then found a radio afterglow in the direction of the burst, lasting up to six days, and optical follow-up linked it to an elliptical galaxy at z = 0.492 (WISE J071634.59-190039.2). Had it held, this would have been the first identified host galaxy of a fast radio burst; renewed radio activity from the region almost a year later raised doubts about the connection.

H.E.S.S. observed the field about 14.5 hours after the burst, as soon as the sky was dark enough. The 1.4 hours of data show no gamma-ray afterglow; assuming a generic E⁻² spectrum, the upper limit above 350 GeV is 1.33 × 10⁻⁸ m⁻² s⁻¹ (99% C.L.). These were the first limits ever placed on the TeV afterglow of a fast radio burst, and I summarised them in the H.E.S.S. and SUPERB paper below.

Multi-wavelength light curve of the FRB 150418 field, with ATCA radio points, later VLA detections and the H.E.S.S. upper limit in red
The FRB 150418 field after the burst. Black squares are the ATCA radio afterglow, open circles the renewed activity seen by the VLA, and the red marker the H.E.S.S. upper limit.

FRB 150215 · Parkes, H.E.S.S., ANTARES

A polarised burst close to the Galactic plane

FRB 150215, recorded at Parkes on 15 February 2015, was 43 ± 5% linearly polarised, unlike most other bursts. Although it arrived about 25° from the Galactic Centre, its rotation measure is consistent with zero (−9 < RM < 12 rad m⁻² at 95% C.L.), so at least this burst was not produced in a dense, magnetised environment. Eleven telescopes followed it up, from radio to gamma rays and neutrinos, without finding any transient or variable counterpart.

I was responsible for the analysis of the H.E.S.S. data. ANTARES scanned its data in time windows of up to a day before and after the burst and found no neutrino candidate; a later ANTARES search covered most of the bursts detected between 2013 and 2017. A footnote: as a member of both H.E.S.S. and ANTARES, my name appeared twice in the joint author list, which the journal had not foreseen; I finally signed only for H.E.S.S.

H.E.S.S. significance map around the direction of FRB 150215, with the Parkes beam marked as a circle
Very-high-energy gamma-ray significance map around FRB 150215. The central circle (14.4′ across) marks the Parkes beam in which the burst was detected.

SGR 1935+2154 · April 2020

A magnetar behind a fast radio burst

On 28 April 2020 CHIME and STARE2 detected a fast radio burst from the Galactic magnetar SGR 1935+2154: the first observational link between fast radio bursts and magnetars. Swift-BAT alerts on the magnetar's outbursts had already triggered H.E.S.S., which observed the source for two hours a few hours before the radio burst, simultaneously with X-ray bursts recorded by INTEGRAL and Fermi-GBM. These were the first very-high-energy observations of a magnetar in a flaring state. No signal was found above 0.6 TeV, and the data set limits on both persistent and transient emission.

The LST-1 telescope of CTAO later observed SGR 1935+2154 for more than 25 hours during its 2021 and 2022 activity, including 0.1-second windows around known X-ray bursts, and set deep limits on tera-electronvolt emission from the bursts and from the source itself.

The full story News from April 2020

CHIME waterfall plot of the first radio burst from SGR J1935+2154, showing its two-component structure
The radio burst from SGR 1935+2154 as recorded by CHIME.

H.E.S.S. · 2015–2022

Seven years of gamma-ray follow-up

H.E.S.S. runs a dedicated programme to search for gamma-ray counterparts of fast radio bursts, through follow-up observations and simultaneous campaigns with radio and X-ray observatories. I led the paper giving the overview: it covers the sources observed from 2015 to 2022, nine of them localised with redshifts between 0.11 and 0.492. No very-high-energy emission was detected. The data limit the luminosity to between 10⁴⁴ and 10⁴⁸ erg s⁻¹, constraining both persistent emission from the burst regions and possible afterglows on timescales from hours to years.

An all-sky archival search compared the positions of fast radio bursts with gamma-ray bursts detected by Swift and Fermi. No significant counterpart was found: fewer than 4% of the radio bursts are associated with gamma-ray bursts in these samples.

MeerKAT · e-MERLIN

Persistent radio sources

Some repeating fast radio bursts come with a compact, persistent radio source, a clue to the engine that produces the bursts and to its surroundings. With MeerKAT we searched for persistent emission from the one-off FRB 20190714A and the repeaters FRB 20190711A and FRB 20171019A. For FRB 20171019A, H.E.S.S. observed simultaneously at very high energies, alongside searches in the ultraviolet, optical and X-ray bands.

A follow-up survey at sub-arcsecond resolution covered 25 well-localised bursts from ASKAP and MeerTRAP. It detected 14 radio sources, one of them variable, and set flux upper limits for 12 non-detections. None can yet be firmly classified as a persistent radio source; higher-resolution observations will tell. MeerTRAP also localised FRB 20210410D to sub-arcsecond precision, tying it to a galaxy at z = 0.1415 with no continuum radio emission at the burst position.

Gamma-ray bursts

PanRadio GRB · ATCA

Radio afterglows, from the first hour to years

The PanRadio GRB programme on the Australia Telescope Compact Array is a systematic, multi-year radio survey of all southern Swift gamma-ray bursts, following their afterglows from within an hour to years after the burst. Its first result traces GRB 230815A for 400 days. That burst would normally get no radio follow-up: it has no known redshift and strong extinction along the line of sight. Its early X-ray jet break, about 0.1 days after the burst, is at odds with how the radio light curves evolved.

Rapid-response triggering catches the earliest radio emission. ATCA automatically triggered a nine-hour observation of the short GRB 231117A (z = 0.257) after its Swift detection, seeing flaring, scintillating and plateau phases; the scintillation gives the earliest limit yet on the size of a GRB blast wave, below 10¹⁶ cm. For GRB 210702A, a five-hour radio flare about half a day after the burst is most likely due to interstellar scintillation.

Gamma-ray bursts at very high energies

GRB 221009A · the BOAT

The brightest burst of all time, in radio

The radio campaign on GRB 221009A spans the first 475 days after the burst and three orders of magnitude in frequency, from 0.15 to 230 GHz. Combined with published data, it is the most detailed radio data set of any gamma-ray burst in cadence and spectral coverage. Three separate synchrotron components are needed to explain the afterglow, and standard analytical afterglow models cannot reproduce its evolution.

At very high energies, bright moonlight held H.E.S.S. back for two days. No gamma rays were detected, but the limits still constrain models of this exceptional burst.

The full story

Chart comparing the prompt emission of GRB 221009A with five earlier record-holding long gamma-ray bursts

Looking ahead

SKA Observatory

Getting ready for the SKA Observatory

The SKA Observatory, with SKA-Low in Australia and SKA-Mid in South Africa, will bring a new level of sensitivity to radio transient science. Its community science book, Advancing Astrophysics with the SKA II (2026), sets out what the telescopes will make possible; we contributed to two of its chapters on transients.

Rapid-response triggering lets a telescope react automatically to a transient alert, from another facility or from its own data streams, and repoint to catch the earliest radio emission. Both SKA-Low and SKA-Mid will have this capability. The first chapter describes the science cases, from particle acceleration and transient central engines to coherent emission and outflow physics, and the system requirements behind them.

The second chapter covers commensal image-plane transient searches, which find transients in data taken for other projects. Building on experience with the SKA pathfinders, it sets out fast imaging strategies, automated detection pipelines, the filtering needed to reject artefacts, and triggered reprocessing tools, for all SKA-Mid and SKA-Low array deployments from AA* onward.

Advancing Astrophysics with the SKA II

Related news · Radio transients

From the news.