ASKAP bags 20 FRBs, opening way to ‘weighing’ Universe

ASKAP found its first fast radio burst (FRB) in early 2017. In Nature on 10 October 2018, a team led by Ryan Shannon (Swinburne University of Technology) reported another 19 FRBs detected by ASKAP over an eight-month period. This was a big step up from the 34 FRBs that the world’s radio telescopes had collectively found before the ASKAP survey began in 2017. (And ASKAP didn’t stop there: it has continued finding FRBs, which will be reported in due course.)

With so many FRBs in hand, Shannon et al. were able to robustly compare them with fainter ones found previously with the Parkes telescope. They concluded that a key characteristic of FRBs, the dispersion measure, is a reliable indicator of distance. This opens the way to using FRBs to locate and measure the baryonic matter – perhaps 30–80% of the Universe’s total baryonic matter – that lurks in the intergalactic medium.



Pulse profiles and dynamic spectra of ASKAP FRBs (Shannon et al. 2018)

Non-standard observing mode gives huge field of view

ASKAP’s cracking pace of discovery came from using the non-standard “fly’s eye” mode of observing. As described by Keith Bannister (CSIRO), in this mode ASKAP’s antennas are pointed in different directions. ASKAP’s phased array feeds give each antenna a 30° field of view, and fly’s eye mode multiplies this many-fold. Shannon et al. observed with eight ASKAP antennas, a subset of the full array.

Comparison shows IGM’s contribution to dispersion measure

FRB pulses arrive on Earth ‘dispersed’, with their different frequency components showing a spread of arrival times. The effect is caused by the pulse interacting with ionised plasma along its line of travel: the delay is greater for long wavelengths (it varies quadratically with wavelength) and is proportional to the column density of ionised matter between the FRB source and Earth.

Shannon et al.’s batch of ASKAP bursts solves a long-standing question about the location of this ionised matter. The matter is found in four regions: our Milky Way Galaxy, the intergalactic medium, the general FRB host galaxy and the environment immediately around the FRB source. The question has been, how much does the matter in each region contribute to the dispersion measure (DM)? We’ve had a handle only on the Milky Way’s contribution: the halo is thought to account for ~15–50 cm -3 pc of the DM, and the interstellar medium ~30 cm-3 pc at high Galactic latitudes.

Because the fly’s eye technique is relatively insensitive, it finds bright bursts. Shannon et al. compared their 20 ASKAP bursts with ~30 fainter ones found previously with the Parkes telescope. The ASKAP FRBs have an average DM half that of the Parkes ones while their limiting fluence (i.e. brightness) is ~20 times greater. The comparison shows that there is a relationship between brightness and dispersion measure, (albeit one with scatter). This strongly implies that a large component of the DM is related to its distance, and hence originates from intergalactic baryons. This is a key step to using FRBs to locate baryonic matter in the intergalactic medium (IGM). But a further element is needed: the redshifts of the FRB host galaxies.

Follow-up of FRB 171020

One event in the ASKAP sample, FRB 171020, has a DM of 114 pc cm−3, the lowest DM of any FRB detected to date, suggesting that its host galaxy must be relatively nearby.

Elizabeth Mahoney (CSIRO) led a follow-up search for the host. The DM implies that the host has a redshift of no more than 0.08, corresponding to a distance of 350 Mpc. The ASKAP localisation was poor (50 × 34 arcmin) but the low distance meant that the total search volume was still relatively small (a maximum co-moving volume of 1620 Mpc3). Mahoney et al. conducted catalogue searches and identified ESO 601–G036 as the most promising candidate. They then made follow-up observations, both optical (with the VLT and Gemini South) and radio continuum (ATCA).

ESO 601–G036 shares some characteristics with the host galaxy of the ‘repeater’ FRB 121102, the only FRB for which a host has been identified: the two are similar in size, metallicity and star-formation rate. But ESO 601–G036 lacks the compact, persistent radio continuum source found in the repeater host. This suggests that not all FRBs may be associated with a persistent radio source.

The repeater is an outlier in other ways. Despite more than 12,000 hours of follow-up observations, Shannon et al. saw none of the ASKAP FRBs repeat. It’s now clear that there are at least two classes of FRBs.

Interferometry for localisation

“The first facility that can routinely localize ‘classical’ [non-repeating] FRBs will become the BeppoSAX of the FRB field!” So writes Sri Kulkarni (Caltech), in looking at how the exploration of FRBs compares with that of GRBs (gamma-ray bursts).

ASKAP has taken up the challenge. Under Keith Bannister’s leadership, ASKAP has switched from fly’s eye mode to standard interferometry for FRB observations. When an FRB is detected, observers will be able to obtain an accurate position for it by using a few seconds of data stored in circular buffers in the beamformers. In this observing mode, it will be possible to localise FRBs to within a few arcseconds and so determine the likely host galaxies. ASKAP has taken a big step in linking DMs to distances: perhaps it will also take the next big step the field is waiting for.