Early-time searches for coherent radio emission from short GRBs with the Murchison Widefield Array

Abstract

Many short gamma-ray bursts (GRBs) originate from binary neutron star mergers, and there are several theories that predict the production of coherent, prompt radio signals either prior, during, or shortly following the merger, as well as persistent pulsar-like emission from the spin-down of a magnetar remnant. Here we present a low frequency (170–200 MHz) search for coherent radio emission associated with nine short GRBs detected by the Swift and/or Fermi satellites using the Murchison Widefield Array (MWA) rapid-response observing mode. The MWA began observing these events within 30–60 s of their high-energy detection, enabling us to capture any dispersion delayed signals emitted by short GRBs for a typical range of redshifts. We conducted transient searches at the GRB positions on timescales of 5 s, 30 s, and 2 min, resulting in the most constraining flux density limits on any associated transient of 0.42, 0.29, and 0.084 Jy, respectively. We also searched for dispersed signals at a temporal and spectral resolution of 0.5 s and 1.28 MHz, but none were detected. However, the fluence limit of 80–100 Jy ms derived for GRB 190627A is the most stringent to date for a short GRB. Assuming the formation of a stable magnetar for this GRB, we compared the fluence and persistent emission limits to short GRB coherent emission models, placing constraints on key parameters including the radio emission efficiency of the nearly merged neutron stars ( $\epsilon_r\lesssim10^{-4}$ ), the fraction of magnetic energy in the GRB jet ( $\epsilon_B\lesssim2\times10^{-4}$ ), and the radio emission efficiency of the magnetar remnant ( $\epsilon_r\lesssim10^{-3}$ ). Comparing the limits derived for our full GRB sample (along with those in the literature) to the same emission models, we demonstrate that our fluence limits only place weak constraints on the prompt emission predicted from the interaction between the relativistic GRB jet and the interstellar medium for a subset of magnetar parameters. However, the 30-min flux density limits were sensitive enough to theoretically detect the persistent radio emission from magnetar remnants up to a redshift of $z\sim0.6$ . Our non-detection of this emission could imply that some GRBs in the sample were not genuinely short or did not result from a binary neutron star merger, the GRBs were at high redshifts, these mergers formed atypical magnetars, the radiation beams of the magnetar remnants were pointing away from Earth, or the majority did not form magnetars but rather collapse directly into black holes.

High time resolution search for prompt radio emission from the long GRB 210419A with the Murchison Widefield Array

“…This allows for the capture of high temporal and frequency resolution raw voltage data (100 µs/10 kHz). Therefore, compared to previous works that use the MWA standard mode to trigger on GRBs with a temporal resolution of 0.5 s (Kaplan et al 2015;Anderson et al 2021a;Tian et al 2022), the triggered VCS observations are almost one order of magnitude more sensitive to millisecond duration transient signals (for a comparison of sensitivity between the standard MWA correlator and the VCS see figure 8 in Tian et al 2022). The new system triggers only on Swift GRBs with ∼arcsec localisations.…”

“…While the high time resolution data are most suitable for searching for the prompt radio emission, snapshot images on ∼ min timescales can be used for detecting dispersed long duration signals (see Section 4.2.2). Compared to MWA imaging in the standard correlator mode (e.g., Tian et al 2022), there is a prerequisite step for imaging the VCS data, i.e. offline correlation for creating visibilities (Sett et al 2022).…”

“…Following this analysis, identified 143 trials with a SNR> 6, with a maximum SNR of 7.1 (for the distribution of the SNRs see Figure A2 in Appendix A). As a further test on the sample of > 6𝜎 trials and the likelihood of some being real, we performed another single pulse search by creating a set of time series on the same dataset using (unphysical) negative DM trials (see Tian et al 2022, for further details). We found 119 candidates and a maximum SNR of 6.7 for the negative DM trials.…”

“…Anderson et al (2021a) reported the first short GRB trigger with the upgraded MWA triggering system, and performed a search for dispersed signals using images with a temporal and spectral resolution of 0.5 s/1.28 MHz, obtaining a fluence upper-limit range from 570 Jy ms at a dispersion measure (DM) of 3000 pc cm −3 (𝑧 ∼ 2.5) to 1750 Jy ms at a DM of 200 pc cm −3 (𝑧 ∼ 0.1), corresponding to the known redshift range of short GRBs (Rowlinson et al 2013). Finally, Tian et al (2022) presented a similar search in the image domain for coherent radio emission from a sample of nine short GRBs between 2017 and 2020 using the MWA rapid-response observing mode and obtained the most stringent prompt emission fluence limit of 100 Jy ms from GRB 190627A.…”

“…There have been searches for prompt radio emission from GRBs, which could be associated with the central engine or the relativistic jet, but so far none have yielded a detection (e.g. Dessenne et al 1996;Obenberger et al 2014;Bannister et al 2012;Palaniswamy et al 2014;Kaplan et al 2015;Anderson et al 2018;Rowlinson et al 2019;Anderson et al 2021a;Bouwhuis et al 2020;Rowlinson et al 2021;Tian et al 2022). These non-detections could be attributed to the small sample size (not all GRBs are predicted to produce detectable radio emission), the limited sensitivity of allsky instruments, or long delays (up to several minutes) between the transient event and the target acquisition of pointed telescopes.…”