EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 664 (August 2022) A&A, 664 (2022) A36 Full HTML
Open Access
Issue
A&A
Volume 664, August 2022
Article Number A36
Number of page(s) 11
Section Stellar atmospheres
DOI https://doi.org/10.1051/0004-6361/202142919
Published online 04 August 2022

© S. Giarratana et al. 2022

Licence Creative CommonsOpen Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This article is published in open access under the Subscribe-to-Open model. Subscribe to A&A to support open access publication.

1 Introduction

Long-duration gamma-ray bursts (GRBs) are extremely powerful flashes that generally last more than 2 s (Mazets et al. 1981; Norris et al. 1984; Kouveliotou et al. 1993) and whose prompt emission is detected mainly in the γ-ray and X-ray domains. They are thought to signpost the catastrophic death of a massive star (see e.g. Galama et al. 1998; Piran 2004, Kumar & Zhang 2015) that has previously expelled its hydrogen envelope into the surrounding medium (Woosley & Heger 2006), and the subsequent formation of a spinning stellar mass black hole (BH; Woosley 1993; Paczyński 1998) or neutron star (NS; Usov 1992). This newborn central engine may power and launch relativistic jets of ejected matter within which internal shocks (Rees & Mészáros 1994) or magnetic reconnections (Drenkhahn & Spruit 2002) can convert a fraction of the bulk kinetic energy into the observed short-lived γ-ray radiation (Mészáros 2002). These expanding jets interact with the circumburst medium (Rees & Mészáros 1992; Mészáros & Rees 1993), triggering both a forward shock and a reverse shock (RS; Mészáros 2002). The electrons at the shock fronts are accelerated to relativistic energies, producing a long-lived afterglow through synchrotron emission, which can be observed from high energies (GeV) through X-rays, optical, and near-infrared down to the radio bands (see e.g. Gehrels et al. 2009; Kouveliotou et al. 2012).

The radio light curve is fundamental for the afterglow modelling; together with frequent optical and X-ray observations, it helps us to better constrain the multi-dimensional parameter space and to distinguish between different scenarios, providing relevant information to understand the progenitor’s nature and the GRB origin. Nevertheless, the detection rate of GRBs observed in the radio band is only ~30%, and an even smaller number of events have multi-epoch observations (Chandra & Frail 2012). In events where radio emission has been detected, it can be observed for months or even years after the burst (Piran 2004). It provides a powerful tool to better constrain not just the internal jet physics, but also the geometry and physical evolution of the jet. Evidence of scintillation has helped study the expansion velocity of the outflow (Frail et al. 1997), achromatic light curve behaviour can inform on jet opening angles (‘jet breaks’) and constrain the transition from relativistic to non-relativistic expansion (Frail et al. 2004), which can be used to infer the total kinetic energy by performing radio calorimetry (see e.g. Berger et al. 2004; Frail et al. 2005). For the nearest events the high angular resolution provided by Very Long Baseline Interferometry (VLBI) has proved to be complementary to studying the afterglow; it is a unique tool to measure the expansion (Taylor et al. 2004) and the centroid displacement (Mooley et al. 2018) of the outflow, to constrain its size (Ghirlanda et al. 2019), and to distinguish the proper compact afterglow emission from contaminating components within the host galaxy.

At the opposite end of the spectrum, only four GRBs have a bona fide detection in the very high-energy (VHE; ≥ 100 GeV) range at either early epochs (e.g. GRB 190114C, 300GeV–1 TeV; MAGIC Collaboration 2019 and GRB 201216C; Blanch et al. 2020b) or at later times deep in the afterglow phase (e.g. GRB 180720B, 100–400 GeV; Abdalla et al. 2019 and GRB 190829A, 180 GeV–3.3 TeV;H.E.S.S. Collaboration 2021). Studying this emission component allows the physical properties of the emitting region and/or of the shocked accelerated particles to be constrained, and the most natural interpretation for this VHE emission is the inverse-Compton scattering of synchrotron photons, known as synchrotron self-Compton (SSC) emission. Based on the very few events detected so far, it seems that the VHE emission characterises both very energetic events, such as GRB 180720B and GRB 190114C, and low-energy events, such as 190829A, but any possible peculiarities of VHE detected bursts will become clearer as the sample of events increases. However, multi-wavelength follow-up of these events has proved a fundamental tool to test the afterglow emission model; for example, for GRB 190829A the VHE emission detected by the H.E.S.S. telescopes was first interpreted as synchrotron emission (H.E.S.S. Collaboration 2021), while multi-wavelength follow-up studies agree on an SSC emission origin (Salafia et al. 2022; Zhang et al. 2021; Fraija et al. 2021).

GRB 201015A was discovered on 2020 October 15 at 22:50:13 UT as a multi-peaked 10 s GRB by the Neil Gehrels Swift Burst Alert Telescope (hereafter Swift/BAT) (D’Elia et al. 2020). Subsequent observations reported the presence of an associated transient in the optical (Lipunov et al. 2020a,b; Malesani et al. 2020; Ackley et al. 2020; Hu et al. 2020; de Ugarte Postigo et al. 2020; Zhu et al. 2020a,b; Belkin et al. 2020a,b; Jelinek et al. 2020; Grossan et al. 2020; Rastinejad et al. 2020; Kumar et al. 2020a; Moskvitin et al. 2020; Pozanenko et al. 2020), X-rays (Kennea et al. 2020; Fletcher et al. 2020; Gompertz et al. 2020; D’Elia & Swift Team 2020), UV (Marshall et al. 2020), and radio (Fong et al. 2020) bands. Remarkably, GRB 201015A was observed by the Major Atmospheric Gamma-ray Imaging Cherenkov (MAGIC) telescopes about 40 s after the Swift trigger, and a hint of a VHE counterpart with a significance ≥3.5σ was reported from preliminary analyses (Blanch et al. 2020a; Suda et al. 2021). With the Fermi Gamma-ray Burst Monitor (GBM) spectrum, Minaev & Pozanenko (2020) suggested that this burst is consistent with the Epeak − Eiso Amati relation (Amati et al. 2002) for long-duration GRBs, with an isotropic equivalent energy of Eiso − (1.1 ± 0.2) × 1050 erg. If confirmed, this would be the fifth and least luminous GRB ever detected in this band.

Optical spectroscopy in the 3700−7800 Å range revealed a redshift for the source of ~0.426 (de Ugarte Postigo et al. 2020; Izzo et al. 2020). To date, all the GRBs that have been detected at VHE have relatively low redshifts: 0.654, 0.425, 0.0785, and 1.1 for GRB 180720B, GRB 190114C, GRB 190829A, and GRB 201216C, respectively (Vreeswijk et al. 2018; Selsing et al. 2019; Valeev et al. 2019; Vielfaure et al. 2020); their isotropic equivalent energies span three orders of magnitude (Rhodes et al. 2020a).

In this paper, we present a multi-wavelength follow-up campaign of GRB 201015A performed with the Karl G. Jansky Very Large Array (VLA), the enhanced Multi Element Remotely Linked Interferometer Network (e-MERLIN), the European VLBI Network (EVN), the Multiple Mirror Telescope (MMT), and the Chandra X-ray Observatory (Chandra). The observations are presented in Sect. 2, while the results are shown in Sect. 3. We exploit the standard model for GRB afterglows in Sect. 4 to explain the multi-wavelength observations, and we compare our results for GRB 201015A with previous GRBs in Sect. 5. We conclude with a brief summary in Sect. 6. Throughout the paper we assume a standard Λ-CDM cosmology with H0 = 69.32 km Mpc−1 s−1, Ωm = 0.286, and Ωλ = 0.714 (Hinshaw et al. 2013). With this cosmology, 1″ corresponds to roughly 5.6 kpc at z = 0.426.

2 Observations

2.1 VLA Observations at 6 GHz

Observations with the VLA were performed 1.41 days post-burst (PI: Fong; project code: 19B-217) at a central frequency of 5.7 GHz with a bandwidth of 1.6 GHz (C band). The target and the phase calibrator J2355+4950 were observed in eight-minute cycles, with seven minutes on the former and one minute on the latter. The distance between the target and the phase calibrator is about 4.5°. Finally, 3C147 was used as bandpass and flux calibrator. The data were calibrated using the casa pipeline, and they were subsequently imaged with the tclean task in casa (Version 5.1.1., McMullin et al. 2007).

2.2 E-MERLIN Observations at 1.5 GHz

We started observing at 1.5 GHz with e-MERLIN 20 days post-burst (2020 November 4; PI: Rhodes, project code: DD10003) with two further observations 23 (2020 November 7) and 101 (2021 January 24) days post-burst. The observations were made at a central frequency of 1.51 GHz with a bandwidth of 512MHz (L band). For each epoch the target and phase calibrator, J2353+5518, were observed in ten-minute cycles, with seven minutes on the former and three on the latter. The distance between the phase calibrator and the target is about 3°. Each observation ended with scans of the flux (J1331+3030) and bandpass calibrators (1407+2827). The data were reduced using the custom e-MERLIN pipeline1. The calibrated measurement sets were imaged in casa (Version 4.7).

2.3 E-MERLIN Observations at 5 GHz

Observations at 5 GHz with e-MERLIN were performed 21 (2020 November 5), 24 (November 8), 60 (December 14), 85 (2021 January 8), and 100 (January 23) days post-burst (PI: Giroletti; project code: DD10004). All epochs but December 14 were centred at 4.50-5.01 GHz (C band) with a bandwidth of 512 MHz divided into four spectral windows of 128 MHz each. For December 14, the frequency range was within 6.55−7.06 GHz (C band). The data were first pre-processed with the casa e-MERLIN pipeline using J1407+2827 as bandpass calibrator and J1331+3030 as flux calibrator. Two phase calibrators were used: J2353+5518, a fainter one on a rapid cycle, and J2322+5057, a brighter one used less frequently (once per hour) to correct for both short- and long-term atmospheric effects. All epochs were observed in eight-minute cycles, with six minutes on the target and two minutes on J2353+5518.

On November 5, an electronic problem occurred and the Defford antenna missed the bandpass and flux calibrators; consequently, the pipeline automatically flagged out this antenna, and there was a considerable data loss. To recover it we performed a further calibration of this epoch. We built a model for J0319+4130 using the pipeline results first, and we subsequently calibrated the data manually using the J0319+4130 model as bandpass and flux calibrator, improving the final image output. After the calibration, we cleaned the dirty image with the tclean task in casa (Version 5.1.1.).

On November 8, the Knockin antenna lost one polarisation channel, and an improved image was achieved using only J2322+5057 for the phase calibration, which is about 3.3° from the target source.

2.4 EVN Observations at 5 GHz

Observations at 5 GHz with EVN were performed 25 (2020 November 9), 47 (December 1), and 117 (2021 February 9) days post-burst (PI: Marcote; project code: RM016). The first epoch (2020 November 9) was conducted at a maximum bitrate of 4 Gbps per station, dividing the full band upon correlation into 16 spectral windows of 32 MHz and 64 frequency channels each, covering the frequency range of 4.57−5.11 GHz (C band). The other two following epochs were conducted at a lower rate of 2 Gbps, resulting in eight spectral windows of 32 MHz and 64 frequency channels each, covering the frequency range of 4.77−5.05 GHz. All observations were correlated in real time (e-EVN operational mode) at JIVE (the Netherlands) using the SFXC software correlator (Keimpema et al. 2015).

The following sources were used as fringe finders and/or bandpass calibrators among the different epochs: BL LAC, J0854+2006, 3C 84, J0555+3948, and J0102+5824. The same phase calibrator as in the e-MERLIN observations was used: J2353+5518, in a phase-referencing cycle of 4.5 min on the target source and 1.5 min on the phase calibrator. The source J2347+5142 was observed as a check source to account for possible phase-referencing losses.

The EVN data were reduced using AIPS2 (Greisen 2003) and Difmap (Shepherd et al. 1994) following standard procedures. An a priori amplitude calibration method was performed using the known gain curves and system temperature measurements recorded individually on each station during the observation. We manually flagged data affected by radio frequency interference (RFI) and then we fringe-fitted and bandpass-calibrated the data using the fringe finders and the phase calibrator. We imaged and self-calibrated the phase calibrator in Difmap to improve the final calibration of the data. We used the same model of the phase calibrator, obtained from the 2020 December 1 epoch, to improve the calibration of all epochs. We note that we chose this epoch because it produced the most reliable image of J2353+5518 in terms of amplitude scales at all baseline lengths (including the short spacing given by the e-MERLIN stations). No apparent changes in the calibrator were observed among these three observations. The obtained solutions were then transferred to the target scans, which were subsequently imaged for each epoch. The check source J2347+5142 was also imaged and self-calibrated, confirming that no significant losses (≲ 10−20%) were present in the obtained amplitudes due to the phase-referencing technique. We note that the Shanghai 65 m Radio Telescope (Tianma) and the Nanshan 25 m Radio Telescope (Urumqi) only participated in the first observation, and since they provided the longest baselines the resolution for the other two epochs decreases significantly (see Table 1).

2.5 Optical Observations and Public Data

At 1.4, 2.2, and 4.3 days post-burst, we observed the position of the afterglow in the i and z bands with the Binospec instrument mounted on the 6.5 m MMT (PI: Fong; project code: 2020c-UAO-G204-20B). We reduced our images using a custom Python pipeline3 and registered the images to the USNO-B1 catalogue (Monet et al. 2003) using standard IRAF tasks (Tody 1993). In the first two epochs we clearly detected an uncatalogued source in both bands that did not appear in our deep image at 4.3 days post-burst. To remove any contamination from the nearby galaxy, we performed image subtractions between the first two epochs and the final epoch using HOTPANTS (Becker 2015). We then calibrated the images to the PanSTARRS Data Release 2 catalogue (Chambers et al. 2016) and performed aperture photometry on the image subtractions with the IRAF/phot task.

We gathered additional optical information from the public GCN Circulars Archive, and the detected emission was de-absorbed with the dust_extinction Python package4, using a Galactic extinction Av = 0.93 (Schlafly & Finkbeiner 2011).

Table 1

Radio observations performed with the VLA, e-MERLIN, and EVN in the L and C bands.

2.6 X-ray Observations and Public Data

We obtained the Swift X-ray Telescope (XRT) unabsorbed flux light curve integrated in the 0.3−10keV energy range from the Swift Burst Analyzer5 provided by the UK Swift Science Data Centre at the University of Leicester (UKSSDC, Evans et al. 2007, 2009).

Moreover, we obtained two epochs of Chandra observations with the Advanced CCD Imaging Spectrometer (ACIS) in very faint mode (PI: Gompertz; project code: 22400511). Exposures were centred around 8.4 and 13.6 days after trigger, with exposure times of 30 ks and 45 ks, respectively. The data were analysed using ciao v4.14 and xspec v12.11.1, following the Chandra X-ray Observatory science threads6.

3 Results

3.1 Radio

A point-like source was clearly visible with the VLA 1.4 days post-burst with a peak brightness of 132 ± 8 μJy beam−1, where the uncertainty includes the r.m.s. noise and a 5% calibration error added in quadrature. The r.m.s. noise uncertainty is 5 μJy beam−1, and therefore the detection has a significance of 26σ confidence. The source was found at a position (J2000) α = 23h37m 16.403s, δ = 53°24′56.39″, with an uncertainty of 0.14″ (1/10 of the beam size, Taylor et al. 1999). Thanks to the wide bandwidth and high signal-to-noise ratio, we were able to split the data in four spectral windows in order to estimate the spectral index β where the flux density is F ∝ vβ. We found β ≃ 2.5. To further improve this estimate, we produced a spectral map with the tclean task in casa by setting nterms = 2 and deconvolver = ‘mtmfs’. We found β = 2.3 ± 0.1 at the peak of the target emission. We attribute the emission to the afterglow of GRB 201015A. Finally, we divided the one-hour observation into two intervals of equal duration and determined the peak brightness in each one, which turned out to be 126 ± 9 μJy beam−1 and 144 ± 10 μJy beam−1 (see Fig. 2, blue stars).

The resulting images from the first and second e-MERLIN epoch at 1.5 GHz showed a point source with a peak brightness of 213 ± 40 μJy beam−1 and 261 ± 48 μJy beam−1, where the quoted uncertainty includes the r.m.s. noise and a 10% calibration error added in quadrature, at the position (J2000) α = 23h37m 16.423s, δ = +53°24′56.43″. The r.m.s. noise uncertainties are 34 μJybeam−1 and 40 μJy beam−1, hence the detections have a significance of 6.2 and 6.5σ confidence, respectively. The uncertainty on the position, which was computed as the ratio between the beam size and the signal-to-noise ratio (Taylor et al. 1999), is 0.03″. Unfortunately, the observation at 101 days was heavily affected by RFI and as a result we obtained a 5σ upper limit of 285 μJybeam−1. The data are shown in Fig. 2 as gold squares.

At 5 GHz a point-like transient was clearly detected with e-MERLIN on November 5 (Fig. 1) at the position (J2000) of α = 23h37m16.422s, δ = 53°24′56.44″. The uncertainty on the position is 0.01″. The point-like source was also detected on November 8 at the position (J2000) α = 23h37m 16.419s, δ = 53°24′56.33″. The uncertainty on the position is 0.02″. Although both positions are in agreement with the coordinates provided by the VLA, we note that they are not consistent with each other at 3σ confidence level. We ascribe the offset in the position to the phase calibration of the second epoch: if the phase calibrator is observed less frequently (i.e. once per hour), it may not be able to trace perfectly the short-term atmospheric effects, and therefore correct for them. Nevertheless, we were not able to improve the phase calibration further. The measured peak brightness is 107 ± 20 μJy beam−1 and 116 ± 28 μJy beam−1 for November 5 and 8, respectively, where the quoted uncertainty includes the r.m.s. noise uncertainty and a 10% calibration error added in quadrature. The r.m.s. noise uncertainties are 17 μJy beam−1 and 26 μJy beam−1, hence the detections have a significance of 6.3 and 4.5σ confidence, respectively. On December 14, January 8, and January 23 no source was detected; the r.m.s. noise is 43, 19, and 16 μJybeam−1, respectively. The data are shown in Fig. 2 as blue dots.

GRB 201015A was detected as a point-like source also in the first two epochs with EVN at 5 GHz (25 and 47 days after the burst) at a consistent (J2000) position of α = 23h37m16.42232s ± 0.2mas, δ = 53°24′56.4392″ ± 0.3 mas. The quoted uncertainties include the statistical uncertainties (0.05 and 0.12 mas for α and δ, respectively), the uncertainties in the absolute International Celestial Reference Frame position of the phase calibrator (0.11 mas), and check source (0.15 mas; Beasley et al. 2002; Gordon et al. 2016), and the estimated uncertainties from the phase-referencing technique (0.13 and 0.2 mas; Pradel, et al. 2006) added in quadrature.

The derived peak brightness measurements are 85 ± 13 μJybeam−1 and 73 ± 12 μJy beam−1, respectively, where the errors comprise both the r.m.s. noise uncertainty and a 10% calibration error, added in quadrature. The r.m.s. noise uncertainties are 9 μJy beam−1 and 10 μJybeam−1, hence the detections have a significance of 9.4 and 7.3σ confidence, respectively. No significant emission above the 3σ r.m.s. level (σ =13 μJybeam−1) was reported in the third epoch. The data are shown in Fig. 2 as blue squares.

The upper limits for the flux densities in the radio band were taken with 3σ confidence level. The full list of radio observations is given in Table 1.

thumbnail Fig. 1

e-MERLIN detection on 2020 November 5. The synthesised beam is shown in the lower left corner.

3.2 Optical

At 1.4 and 2.2 days post-burst, we clearly detected the optical afterglow in both i and z bands at α = 23h37m16.43s, δ = +53°24′56.6″ (J2000; uncertainty = 0.2″). In addition, we detected the host galaxy at α = 23h37m16.48s, δ = +53°24′54.6″ (J2000; uncertainty = 0.2″).

The optical light curve is shown in Fig. 2: g-band data from Belkin et al. (2020b, green hexagons), Grossan et al. (2020, green dots) and Ackley et al. (2020, green circles); r-band data from Belkin et al. (2020a, red pentagons), Belkin et al. (2020b, red stars), Zhu et al. (2020a,b, red hexagons), Moskvitin et al. (2020, red diamonds), Grossan et al. (2020, thin red diamonds), Kumar et al. (2020b, red plus), Pozanenko et al. (2020, red circles); i-band data from Grossan et al. (2020, purple squares) and our MMT/Binospec observations (purple circles); our z-band MMT/Binospec observations (brown circles).

The emission peaked between 200 and 300 s after the GRB trigger, reaching a maximum of R~16.5 mag (Jelinek et al. 2020; Zhu et al. 2020a). Between 0.1 and 3 days our light curve follows a power law F (t) ∝ t−0.84±0.06, which is consistent with previous results in the GCNs (Pozanenko et al. 2020). Remarkably, a type Ic-BL supernova (SN) contribution can be seen between 3 and 20 days after the burst (Pozanenko et al. 2020; Rossi et al. 2021), which corroborates the long-duration nature of this burst.

3.3 X-rays

The Swift/XRT light curve was further analysed by splitting the last two observations in four time intervals. We retrieved the XRT spectral files from the online archive7 and analyse them with the public software xspec v12.10.1f, assuming a simple power-law model. The tbabs model for the Galactic absorption and the ztbabs model for the host galaxy absorption, adopting the source redshift z = 0.426, are used in the fitting procedure. The absorption parameters are fixed to the values reported by the Swift website for this burst, namely NH,gal = 3.6 × 1021 atoms cm−2 (Kalberla et al. 2005; Willingale et al. 2013) and NH,intr = 5 × 1021 atoms cm−2. Leaving the normalisation and the photon index of the power-law free to vary, we find integrated fluxes consistent with those reported on the Swift website.

From our two epochs of Chandra observations we find 0.5−7 keV source count rates of (4.07 ± 0.38) × 10−3 cts s−1 and (3.11 ± 0.29) × 10−3 cts s−1. In a combined spectral fit of both Chandra epochs and the late XRT observations (>10 days), the data are well modelled (cstat/dof=600/1808) by an absorbed power law of the form powerlaw*tbabs*ztbabs (Wilms etal. 2000) with aphoton index of Г = 2.10 ± 0.13. The intrinsic absorption column is fixed to NH,intr = 5 × 1021 atoms cm−2 at z = 0.426 over the Galactic value of NH,gal = 3.6 × 1021 atoms cm−2 (Kalberla et al. 2005; Willingale et al. 2013) to match those reported on the UKSSDC. From this, we derived unabsorbed 0.3−10 keV fluxes of (1.26 ± 0.05) × 10−13 erg cm−2 s−1 at 8.4 days and (1.10 ± 0.04) × 10−13 erg cm−2 s−1 at 13.6 days.

The X-ray light curve is shown in Fig. 2 for the Swift/XRT public data (dark blue circles) and our Chandra observations (dark blue squares). For the Swift/XRT light curve, we included the results from the Swift Burst Analyzer up to ~0.12 days, and from that epoch on we used our re-analysis of the last two observations. Our XRT analysis suggests that the light curve can be fitted with a power law with index F−1.1±0.3 between 0.04 and 0.71 days post-burst, which is shallower but still consistent with the previous analysis from D’Ai et al. (2020). However, the subsequent detections at 8.4 and 13.6 days with Chandra show a respective flux ~6 and 8 times higher than expected from extrapolating the earlier XRT light curve, and the increased flux is further confirmed by the late time (~20 days after the burst) Swift/XRT follow-up (D’Elia & Swift Team 2020).

4 Broadband Modelling

The multi-wavelength afterglow synchrotron emission of a GRB seen on-axis can be studied through a standard model (see e.g. Granot & Sari 2002; Zhang & Mészáros 2004). First, assuming that the flux density can be parametrised as F ∝ vβtα, the spectrum can be fitted with several power law segments, which join at specific break frequencies: (i) the self-absorption frequency vsa, (ii) the maximum frequency vm, and (iii) the cooling frequency vc. The other parameters needed to build the spectrum are (iv) the maximum flux density Fm and (v) the electron distribution index p. Once we have determined these quantities and their temporal evolution, the multi-wavelength light curves are constrained. For this work we use the relations provided by Granot & Sari (2002), and throughout the paper we consider two possible density profiles for the circumburst medium: a wind-like profile ρ = Ar−2, which is naturally expected if the progenitor is a massive star collapsing into a BH or a NS, and a homogeneous surrounding medium ρ = const, which can be ascribed either to the canonical ISM or to a wind bubble shocked against the ISM (Aksulu et al. 2022). Hereafter we use the term ISM for a homogeneous profile indiscriminately.

We note that in our modelling we do not include the description of the coasting phase, the contribution from the RS, or the late time SN emission. A more sophisticated modelling that comprises the RS contribution would introduce more parameters; if frequent observations are available around the epoch at which the RS is supposed to prevail (at about 1 day in the radio band; see e.g. Rhodes et al. 2020a) these parameters can be constrained. With only one detection in the С band before 20 days post-burst, we could not constrain the parameters. In the optical, the emission before 0.01 days shows a bump that could be due to a possible RS contribution, while after 3 days the SN emission becomes dominant (Pozanenko et al. 2020; Rossi et al. 2021), hence the prediction of the modelling should be considered only from about 0.01 to 3 days post-burst in this band.

To derive the modelling light curves, we performed a comparison of the simplified afterglow prescription with the available data, changing the above-mentioned parameters to get as close as possible to the observed multi-wavelength light curves and to reproduce the afterglow spectrum at three sampling epochs, namely 0.12, 1.41, and 23 days after the GRB trigger (see Fig. 3).

thumbnail Fig. 2

Multi-wavelength afterglow light curves (see Sect. 3). For each band, the light curves predicted by the standard model with vsa = 13 GHz, vm = 6 GHz, vc = 2 × 107 GHz, Fm = 800 μJy, and p = 2.05 at 1 day for a homogeneous surrounding medium are shown: 1.5 GHz (orange); 5 GHz (blue); r, g, i, and z bands (red, lime, violet, and brown, respectively); integrated X-ray light curve (dark blue). The green, orange, and brown vertical lines pinpoint the epochs of the spectra at 0.12, 1.41, and 23 days, respectively (see also Fig. 3). The dashed line shows a simple model for the SN contribution in the r band (see Sect. 5.1).

4.1 ISM Profile

For the ISM profile we built the spectrum at 0.12 days, with the optical r band from Pozanenko et al. (2020) and the XRT detections at 1.41 days with the VLA detection (see Sect. 3) and our optical i- and z-band observations, and the spectrum at 23 days with our radio detection at 1.5 and 5 GHz, the optical r band from Pozanenko et al. (2020), and the last XRT detection (Fig. 3). From the spectra and the multi-wavelength light curves we constrain the parameter space as follows. First, from the spectral index β = 2.3 ± 0.1 derived with the VLA data we cannot discern whether the emission at 6 GHz lies in the v2 or v5/2 portion of the spectrum at 1.41 days, and therefore we consider three different cases. At this epoch it could be that (i) 6 GHz < vsa < vm, (ii) vm < 6 GHz < vsa, or (iii) 6 GHz < vm < vsa. Moreover, at 23 days the spectral slope between 1.5 GHz and 5 GHz is reversed, meaning that the flux density is decreasing with the frequency, and hence we expect that vm < vsa < 1.5 GHz. Finally, at 23 days the optical emission is dominated by the SN, hence we consider the optical detections as upper limits. To build the modelling light curves and spectra we derive the break frequencies, the p value, and the maximum flux density Fm at 1 day, in order to simplify the equations from Granot & Sari (2002).

  • (i)

    If 6 GHz < vsa< vm, since vm > 6 GHz at 1.41 days, vm ∝ t−3/2 and vsa is constant in time, to avoid vm crossing vsa before 1.41 days we impose vm > 15 GHz and vsa > 9 GHz at 1 day. However, once vm crosses vsa, vsa ∝ t−(3p+2)/2(p+4). fore at 1 day vsa < 13 GHz, otherwise at 23 days vsa > 1.5 GHz, and consequently vm < 24 GHz (otherwise it does not cross vsa before 23 days). At 1 day the flux density at vm is found to be 500 μJy < Fm < 600 μJy. With a lower Fm we underestimate the emission at 5 GHz observed with EVN, while with a higher flux we overestimate the e-MERLIN detections at the same frequency. With the slope of the optical light curve we can constrain the p value: since the light curve shows a clear slope that can be described by a single power law between 0.01 and 3 days, vm < optical < vc and F ∝ t3(1−p)/4 in this regime. Finally, the X-ray integrated light curve allows us to further constrain p and determine vc: for v < vc we have F ∝ t3(1−p)/4, while for v > vc we have F ∝ t(2−3p)/4; hence, the sooner vc crosses the X-ray band, the fainter the detected emission will be. In summary, to reproduce both the spectra and the light curves we find that 9 GHz < vsa < 13 GHz, 15 GHz < vm < 24 GHz, 5×106 GHz < vc < 108 GHz, 500 μJy < Fm < 600 μJy and 2.01 <p< 2.10 at 1 day.

  • (ii)

    If vm < 6 GHz < vsa at 1.41 days, since vsa ∝ t(2−3p)/2(р+4); we impose that vsa > 10 GHz at 1 day; moreover, vsa < 18 GHz at 1 day, otherwise at 23 days vsa > 2 GHz and our detections at 1.5 GHz would lie in the v5/2 portion of the spectrum and the emission at 5 GHz would be overestimated. To reproduce the spectra and the light curves we find that the range for vsa is further constrained to 13 GHz < vsa < 16 GHz. Since at 1.41 days vm ≤ 4 GHz (otherwise the lowest end of the bandwidth of the VLA detection would be underestimated), at 1 day vm ≤ 7 GHz. Finally, with the same argument presented in case (i), we find that at 1 day 6 × 106 GHz < vc < 108 GHz, 800 μJy < Fm < 1 mJy and 2.01 <p<2.20. We note that in this case Fm refers to the flux density at vsa.

  • (iii)

    If 6GHz < vm < vsa at 1.41 days, we can have both 6 GHz < vm < vsa and 6 GHz < vsa < vm at 1 day. Considering both these sub-cases, since vm ∝ t−3/2, at 1 day vm> 13 GHz, otherwise at 1.41 days vm < 8 GHz and it would lie too close to the highest end of the bandwidth of the VLA detection to reproduce the spectrum; conversely, if at 1 day vm > 18 GHz, we cannot reproduce the light curve in the С band because the detections at 6 GHz with the VLA are underestimated, while e-MERLIN and EVN observations are overestimated. Since at 1.41 days vsa > vm, we find that 13 GHz < vsa < 18 GHz (for larger values we cannot reproduce the С band light curve). Once again, with the same argument presented in case (i), we derived 5 × 106 GHz < vc < 2 × 108GHz, 630 μJy < F < l mJy and 2.01 < p < 2.20 at 1 day. In this case Fm refers to the flux density of vsa or vm for the two sub-cases. We note that these ranges for the parameters are the superposition of the ranges derived for the two sub-cases.

In Table 2, we report our results for the parameter space at 1 day. The model light curves for the ISM profile are shown in Fig. 2 for vsa = 13 GHz, vm = 6 GHz, vc = 2×107 GHz, Fm = 800 μJy at 1 day, and an electron distribution index p = 2.05. The 1.5 GHz and the 5 GHz light curve are displayed in orange and blue, respectively; the r, g, i, and z bands are in red, lime, violet, and brown, respectively; the X-ray light curve is displayed in dark blue. Although this modelling provides a satisfactory description of the multi-wavelength light curves, the optical light curve contains the already discussed features in addition to the forward shock emission: before 0.01 days there is a bump which could be due to a possible RS contribution, while after three days the SN emission becomes dominant (Pozanenko et al. 2020; Rossi et al. 2021).

thumbnail Fig. 3

Spectra at 0.12 (green), 1.41 (orange), and 23 (brown) days after the GRB onset for a homogeneous surrounding medium with vsa = 13 GHz, vm = 6 GHz, vc = 2 × 107 GHz, Fm = 800 μJy, and p = 2.05 at 1 day. Spectrum at 0.12 days: optical observations from Pozanenko et al. (2020) and XRT butterfly plot. Spectrum at 1.41 days: our VLA and MMT detections. Spectrum at 23 days: our 1.5 and 5 GHz observations, optical data from Pozanenko et al. (2020) and the XRT butterfly plot; the optical emission is dominated by the SN component.
Table 2

Constraints on the model parameters at 1 day for a homogeneous circumburst medium.

4.2 Wind-like Profile

For the wind-like profile we first tried to reproduce the optical and X-ray data, finding that vsa = 1 GHz, vm = 30 GHz, vc = 2 × 107 GHz, Fm = 200 μJy at 1 day, and the electron distribution index p = 2.01. Since this model conspicuously failed to reproduce the radio detections and the optical slope, we tried to reproduce the radio light curve at 5 GHz first, and we found that vsa = 4 GHz, vm = 103 GHz, vc = 2 × 107 GHz, Fm = 600 μJy at 1 day, and the electron distribution index p = 2.01. Neither of these models reproduces the optical slope, and the second model fails to reproduce the X-ray emission. Different choices of the parameters in the wind-like scenario provide even poorer fits. We can therefore conclude that the modelling provided by the ISM provides the best agreement with the data, and we consider it hereafter. We note that this further corroborates the need of X-ray, optical, and radio observations in order to break the degeneracy in the afterglow modelling, as with only two of them data can be misinterpreted.

4.3 Intrinsic Host Galaxy Extinction

As pointed out by Kann et al. (2006), the intrinsic host galaxy extinction can be relevant in the optical/NIR. By changing the model parameters, we tried overestimating the optical emission and, from the discrepancy between the observed and the modelled optical flux densities, the contribution due to the intrinsic host galaxy absorption can be estimated. However, our modelling light curves and spectra cannot predict values for the flux density that are larger than those observed in the optical data. Moreover, by changing the maximum flux density and the p-value, we cannot reproduce the observed light curves in the radio band. As our modelling light curve already underestimates the afterglow optical emission (see Fig. 3), by adding the intrinsic host galaxy extinction the discrepancy would increase. Therefore the only constraint we can put on the intrinsic host galaxy absorption is that it is negligible, if we assume that the model is correct. Although more sophisticated models could take into account this further correction, this is beyond the goals of this work.

5 Discussion

Once the free parameters vsa, vm, vc, Fm, and p are constrained, we can exploit the relations provided by Granot & Sari (2002) to derive the global and microphysical parameters of the jet: the isotropic kinetic energy E, the density of the medium that surrounds the progenitor n, the fraction of internal energy retained by the magnetic field ϵB, and the fraction of internal energy retained by the electrons ϵe. From the conservation of energy, we know that ϵe ≤ 1, ϵB ≤ 1, and ϵe + ϵB ≤ 1. A further constraint is given by the VHE emission; if we consider the sub-TeV emission to be due to the SSC from the relativistic electrons, then ϵe ≥ ϵb (Sari & Esin 2001; Zhang & Mészáros 2001). If we try to solve the equations from Granot & Sari (2002), the inferred parameters violate the conservation of energy (i.e. ϵe + ϵB ≥ 1); however, these values are determined under the implicit assumption that all the electrons that are swept up by the forward shocks are accelerated, while this is expected to be true only for a fraction f of them. As shown by Eichler & Waxman (2005), if me/mp ≤ f ≤ 1 the observed emission does not change when scaling the parameters as follows: E → E/f, ϵe → ϵef, ϵB → ϵBf, n → n/f (van der Horst et al. 2014). In order to find the solutions, we make E and ϵe vary within physically reasonable ranges (i.e. 1050 erg ≤ E ≤ 1055 erg and 10−4 ≤ ϵe ≤ 1), and we subsequently calculate ϵB and n using the inferred break frequencies, Fm and p. Finally, we apply the constraints given by the conservation of energy and the sub-TeV emission. The final solutions are listed in the second column of Table 3.

Furthermore, since we expect ϵe to be of the order of 0.1 from numerical simulations (Sironi et al. 2013, and references therein), we provide the full set of inferred values for the 0.05 ≤ ϵe ≤ 0.15 case in the second column of Table 4. We find that the isotropic kinetic energy goes from 3 × 1050 to 1055 erg. If we consider the isotropic-equivalent energy derived by Minaev & Pozanenko (2020) from the prompt emission, we can roughly estimate the efficiency of the prompt emission as η = Eiso/(E + Eiso). We estimate that η ≃ 10−3−27%.

To discuss these values in a broader context we consider a recent work by Aksulu et al. (2022), who examined 26 GRBs with well-sampled broadband data sets. The authors found that ϵb ranges from ≈2.6 × 10−6 (GRB 030329) to ≈0.91 (GRB 130907A) for those GRBs that can be described with an ISM profile (hereafter ISM Sample), and 3 out of 13 GRBs have ϵB ≥ 0.5; for ϵe they found a range between ≈0.14 (GRB 090328) and ≈0.89 (GRB 010222); finally, n goes from ≈5 × 10−3 (GRB 010222) to ≈390 cm−3 (GRB 030329).

We then consider long GRBs with a claimed RS detection (in X-rays, optical, and/or radio) whose multi-wavelength light curves can be aptly described with an ISM profile (hereafter RS Sample): GRB 990123, 021004, 021211, 060908, 061126, 080319B, 090102, and 090424 (Japelj et al. 2014); GRB 130427A (Perley et al. 2014); GRB 160509A (Laskar et al. 2016); GRB 160625B (Alexander et al. 2017); GRB 161219B (Laskar et al. 2018); GRB 180720B (Wang et al. 2019); GRB 190829A (Rhodes et al. 2020a). The circumburst density for the GRBs of the RS Sample goes from ≈5 × 10−5 cm−3 for GRB 160625B to ≈360 cm−3 for GRB 090201, while ϵe ranges from ≈4 × 10−4 for GRB 090102 to ≈0.93 for 161219B and ϵB goes from ≈2 × 10−5 for GRB 090102 to ≈0.11 for GRB 160509A. The values we infer for GRB 201015A are therefore consistent with those found in the ISM and RS samples, even though the surrounding density is generally higher.

Finally, we consider three GRBs that have been detected at VHE: GRB 180720B, GRB 190114C, and GRB 190829A. For these bursts ϵe goes from 0.02 (GRB 190114C; Misra et al. 2021) to 0.1 (GRB 180720B; Wang et al. 2019); ϵB goes from 4.7 × 10−5 (GRB 190114C; Misra et al. 2021) to 10−4 (GRB 180720B; Wang et al. 2019); and the surrounding medium density n goes from 0.1 (GRB 180720B; Wang et al. 2019) to 23 (GRB 190114C; Misra et al. 2021). These values are consistent with those we derive for GRB 201015A in this work.

From the maximum flux density Fm at 8.5 GHz we calculate the luminosity L of the afterglow with (Chandra & Frail 2012), where dl is the luminosity distance in cm, Fm is expressed in erg s−1 cm−2 Hz−1, z is the redshift, and α = β = 0 since the peak in the light curve is also a peak in the spectrum. We find that L ≃ 3.5 × 1030 erg s−1 Hz−1 at 1.9 days, which is slightly below the average value for radio-detected GRB afterglows (Chandra & Frail 2012). Finally, the maximum luminosity L ≃ 5.4 × 1030 erg s−1 Hz−1 at 15.7 GHz at 0.8 days is consistent with the radio luminosity previously found for the other GRBs detected at VHE (Rhodes et al. 2020a).

We note that the allowed ranges inferred for the microphysical and global parameters of GRB 201015A are too large to pinpoint any possible deviation of this burst from the samples we used, and hence to derive important information on the production of VHE photons in GRBs. Moreover, a population study is still hindered by the paucity of GRBs detected at VHE and their proximity (z < 1.1), which could lead to a strong bias. A larger and more complete sample is therefore needed. On the other hand, the fact that we cannot flag any possible deviation from the mentioned samples could be consistent with the VHE GRBs being drawn from the same parent population as the other radio-detected long GRBs (Rhodes et al. 2020a).

Table 3

Global and microphysical parameters for GRB 201015A in the ISM scenario.

Table 4

Global and microphysical parameters for GRB 201015A in the ISM scenario if 0.05 ≤ ϵe ≤ 0.15.

5.1 Additional Emission Components

It is worth noting that a refined model could possibly be obtained by including the RS component, whose prescription could explain the bump and the observed excess in the optical emission before 0.01 days. All the GRBs with a confirmed VHE emission were in fact successfully modelled once a RS component was included: GRB 180720B (Fraija et al. 2019; Wang et al. 2019), GRB 190114C (Laskar et al. 2019b), GRB 190829A (Rhodes et al. 2020a).

Concerning the SN emission, if we take the emission of SN1998bw in the r band (Galama et al. 1998), de-absorb the flux density using AV = 0.2 (Galama et al. 1998), and move the SN to z = 0.426 and seven days earlier, we find that its light curve is consistent with that observed for GRB 201015A after three days from the burst (see Fig. 2, dashed line). This further strengthens the SN origin of the bump observed around ten days post-burst.

Finally, we suggest that a transition between the wind-like profile and the ISM profile at around 0.1−0.2 days could possibly explain the change in slope observed in the X-ray light curve after ~0.2 days (see e.g. Kamble et al. 2007; Veres et al. 2015). The optical slope between 0.03 and 0.2 days follows a power law F ∝ t−1.1±0.2, which is consistent with the prediction from a model with a wind-like profile, namely F ∝ t−1.3, if the optical lies between vm and vc. The prediction for the fireball model with a homogeneous circumburst medium is F ∝ t−0.8, which is still consistent but shallower.

5.2 High-Resolution Observations

To measure the expansion or the proper motion of the outflow, a high signal-to-noise ratio is required as it allows both a follow-up of the afterglow up to later times and a smaller uncertainty on the position of the detected source (Taylor et al. 1999). While we achieved a milliarcsecond angular resolution with EVN, we could not pinpoint any displacement of the centroid (off-axis GRB, Mooley et al. 2018; Ghirlanda et al. 2019) nor an expansion of the source (on-axis GRB, Taylor et al. 2004).

The position of the afterglow in the two detections with EVN is consistent within the uncertainties (i.e. ∆α = 0.2 mas and ∆δ = 0.3 mas). At z = 0.426, the centroid displacement before 47 days post-burst is therefore smaller than 1.1 pc in right ascension and 1.7 pc in declination; assuming that the burst is observed at the viewing angle θ that maximises the apparent velocity βapp = Г (i.e. ), we derive a Lorentz factor upper limit of Гα ≤ 40 in right ascension and Гδ ≤ 61 in declination. Considering the previous outstanding burst for which a proper motion was observed (i.e. GRB 170817A at z = 0.0093), a displacement of the same magnitude as that of GRB 170817A would have been seen as 0.08 mas at z = 0.426 after ~207 days post-burst.

On the other hand, if the GRB is seen on-axis, by taking the minor axis of the beam we constrain the size of the afterglow to be ≤5pc and ≤16pc at 25 and 47 days, respectively. Considering the only case for which the expansion was confirmed (i.e. GRB 030329 at z = 0.1685), an expansion of the same magnitude as that of GRB 030329 would have been seen as 0.09 mas at z = 0.426 after ~80 days post-burst.

Since our best resolution with EVN is 1.8 mas × 0.9 mas, we would have detected such an expansion or displacement if (i) the size of the beam had not changed in later observations; (ii) the afterglow had been observable and detectable with a signal-to-noise ratio higher than 10 for about 200 days and 80 days in the case of displacement and expansion, respectively; and (iii) the displacement or expansion had occurred along the coordinate corresponding to the minor axis of the beam.

Conversely, considering the worst resolution reached with our VLBI observations, 3.1 mas × 3.6mas, we would have pinpointed these effects if the afterglow had been detectable for about 800 days or 320 days in the case of proper motion and expansion, respectively, so that the measurements to be performed would have been of the order of 0.3 mas.

5.3 Host Galaxy

The host galaxy was first pinpointed by Belkin et al. (2020a) and subsequently confirmed by Rastinejad et al. (2020) and Rossi et al. (2021), who found a magnitude r = 22.9 ± 0.2.

With the MMT observations, we derive the position of the host of α = 23h37m16.4757s, δ = +53°24′54.626″ (J2000; uncertainty = 0.235″); this is found to be 1.86″ from the source observed at 1.5 GHz, which corresponds to roughly 10kpc at z = 0.426. The uncertainty in the radio position at 1.5 GHz is 0.03″, which is ~170pc, and therefore we can state that the emission observed at 1.5 GHz is consistent with being generated by the afterglow. Moreover, as the beam size at 1.5 GHz is roughly 0.18″× 0.12″, the emitting region should be of the order of 1 kpc× 0.7 kpc; if the detected emission were caused by a very active star-forming region, we would have observed a stable emission in the optical at the same position instead of a transient event.

A safe discrimination between the galactic contamination and the proper afterglow emission at 1.5 GHz could also be achieved with a higher resolution and an improved sensitivity in late epochs in order to obtain better constraints on the light curve. While the former requirement is provided by VLBI observations, the latter is reached with the Pathfinders of the Square Kilometre Array (SKA): the Meer Karoo Array Telescope (MeerKAT; see e.g. Rhodes et al. 2020a) and the Australian Square Kilometre Array Pathfinder (ASKAP). Moreover, abetter sensitivity allows the detection of possible late time jet breaks, and therefore the measurement of the jet opening angle.

6 Conclusions

GRB 201015A was a nearby (z = 0.426; de Ugarte Postigo et al. 2020; Izzo et al. 2020) long-duration GRB discovered on 2020 October 15 by Swift/BAT (D’Elia et al. 2020). Its long-lasting afterglow emission has been observed from γ rays down to radio bands; it is claimed to be the fifth GRB ever detected at VHE energies (Blanch et al. 2020a; Suda et al. 2021).

We performed a radio follow-up with the VLA, e-MERLIN, and EVN at 1.5 and 5 GHz over 12 epochs from 1.4 to 117 days after the GRB onset. At 5 GHz we detected a point-like source consistent with the afterglow position on 2020 October 17; 2020 November 5, 8, and 9; and 2020 December 1; conversely, on 2020 December 14, 2021 January 8 and 23, and 2021 February 9 no source was detected. At 1.5 GHz we detected a point-like source on 2020 November 4 and 7, while on 2021 January 24 no source was detected.

We observed and detected the afterglow of GRB 201015A also in X-rays with Chandra (8.4 and 13.6 days post-burst) and in the optical with MMT (1.4 and 2.2 days post-burst). Finally, we collected public X-ray data from Swift/XRT and optical data from the GCN Circulars Archive. We built multi-wavelength light curves and three spectra at 0.12, 1.41, and 23 days post-burst, and we exploited the standard model provided by Granot & Sari (2002) for a sharp-edged jet seen on-axis to constrain the global and microphysical parameters of the outflow. We find that the observed light curves can be reproduced with a homogeneous circumburst medium profile, and that the parameters we derived for GRB 201015A are consistent with those previously found in the literature for other GRBs, even though we caution that a fully reliable modelling will require a proper characterisation of the VHE detection, which is unavailable at present.

Despite the high angular resolution we achieved with the EVN observations, we could not pinpoint any change in the afterglow position. If the GRB is seen slightly off-axis, we constrain the proper motion of the outflow to be smaller than 1.1 pc in right ascension and 1.7 pc in declination before 47 days post-burst. This proper motion corresponds to a Lorentz factor upper limit of Гα ≤ 40 in right ascension and Гδ ≤ 61 in declination, if we assume that the GRB is seen at the viewing angle θ which maximises the apparent velocity βapp (i.e. ). Conversely, if the GRB is seen on-axis, we find that the size of the afterglow is ≤5 pc and ≤16 pc at 25 and 47 days, respectively.

We note that the bump before 0.01 days post-burst in the optical light curve could be explained by an RS component. On the other hand, we find that the Chandra and the last Swift/XRT detections are brighter than expected from the model and from the extrapolation of the previous data points. Even though further observations are needed, a late time central engine activity or a transition from a wind-like profile to a homogeneous surrounding medium at early times could possibly explain the change in the slope of the X-ray light curve.

Acknowledgements

The authors would like to thank the anonymous referee for their helpful comments. e-MERLIN is a National Facility operated by the University of Manchester at Jodrell Bank Observatory on behalf of STFC, part of UK Research and Innovation. The European VLBI Network is a joint facility of independent European, African, Asian, and North American radio astronomy institutes. Scientific results from data presented in this publication are derived from the following EVN project code: RM016. We thank the directors and staff of all the EVN telescopes for making this target of opportunity observation possible. The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc. MMT Observatory access was supported by Northwestern University and the Center for Interdisciplinary Exploration and Research in Astrophysics (CIERA). The scientific results reported in this article are based in part on observations made by the Chandra X-ray Observatory (PI: Gompertz; project code: 22400511). This research has made use of software provided by the Chandra X-ray Center (CXC) in the application packages CIAO and Sherpa. BM and JMP acknowledge financial support from the State Agency for Research of the Spanish Ministry of Science and Innovation under grant PID2019-105510GB-C31 and through the Unit of Excellence María de Maeztu 2020-2023 award to the Institute of Cosmos Sciences (CEX2019-000918-M). AJL has received funding from the European Research Council (ERC) under the European Union’s Seventh Framework Programme (FP7-2007-2013) (Grant agreement no. 725246). BPG acknowledges funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 948381, PI: Nicholl).

References

  1. Abdalla, H., Adam, R., Aharonian, F., et al. 2019, Nature, 575, 464 [Google Scholar]
  2. Ackley, K., Galloway, D. K., Mong, Y.-L., et al. 2020, GRB Coordinates Network, Circular Service, 28639 [Google Scholar]
  3. Aksulu, M. D., Wijers, R. A. M. J., van Eerten, H. J., et al. 2022, MNRAS, 511, 2848 [NASA ADS] [CrossRef] [Google Scholar]
  4. Alexander, K. D., Laskar, T., Berger, E., et al. 2017, ApJ, 848, 69 [NASA ADS] [CrossRef] [Google Scholar]
  5. Amati, L., Frontera, F., Tavani, M., et al. 2002, A&A, 390, 81 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  6. Beasley, A. J., Gordon, D., Peck, A. B., et al. 2002, ApJS, 141, 13 [NASA ADS] [CrossRef] [Google Scholar]
  7. Becker, A. 2015, Astrophysics Source Code Library [record ascl:1504.004] [Google Scholar]
  8. Belkin, S., Pankov, N., Pozanenko, A., et al. 2020a, GRB Coordinates Network, Circular Service, 28656 [Google Scholar]
  9. Belkin, S., Kim, V., Pozanenko, A., et al. 2020b, GRB Coordinates Network, Circular Service, 28673 [Google Scholar]
  10. Berger, E. 2014, ARA&A, 52, 43 [CrossRef] [Google Scholar]
  11. Berger, E., Kulkarni, S. R., & Frail, D. A. 2004, ApJ, 612, 966 [NASA ADS] [CrossRef] [Google Scholar]
  12. Blanch, O., Gaug, M., Noda, K., et al. 2020a, GRB Coordinates Network, Circular Service, 28659 [Google Scholar]
  13. Blanch, O., Longo, F., Berti, A., et al. 2020b, GRB Coordinates Network, Circular Service, 29075 [Google Scholar]
  14. Chambers, K. C., Magnier, E. A., Metcalfe, N., et al. 2016, ArXiv e-prints [arXiv:1612.0556] [Google Scholar]
  15. Chandra, P., & Frail, D. A. 2012, ApJ, 746, 156 [NASA ADS] [CrossRef] [Google Scholar]
  16. Condon, J. J. 1992, ARA&A, 30, 575 [Google Scholar]
  17. D’Ai, A., Gropp, J. D., Kennea, J. A., et al. 2020, GRB Coordinates Network, Circular Service, 28660 [Google Scholar]
  18. D’Elia, V., & Swift Team 2020, GRB Coordinates Network, Circular Service, 28857 [Google Scholar]
  19. D’Elia, V., Ambrosi, E., Barthelmy, S. D., et al. 2020, GRB Coordinates Network, Circular Service, 28632 [Google Scholar]
  20. de Ugarte Postigo, A., Kann, D. A., Blazek, M., et al. 2020, GRB Coordinates Network, Circular Service, 28649 [Google Scholar]
  21. Drenkhahn, G., & Spruit, H. C. 2002, A&A, 391, 1141 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  22. Eichler, D., & Waxman, E. 2005, ApJ, 627, 861 [NASA ADS] [CrossRef] [Google Scholar]
  23. Evans, P. A., Beardmore, A. P., Page, K. L., et al. 2007, A&A, 469, 379 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  24. Evans, P. A., Beardmore, A. P., Page, K. L., et al. 2009, MNRAS, 397, 1177 [Google Scholar]
  25. Fletcher, C., Veres, P., & Fermi-GBM Team 2020, GRB Coordinates Network, Circular Service, 28663 [Google Scholar]
  26. Fong, W., Schroeder, G., Rastinejad, J., et al. 2020, GRB Coordinates Network, Circular Service, 28688 [Google Scholar]
  27. Fraija, N., Dichiara, S., Pedreira, A. C. C. E. S., et al. 2019, ApJ, 885, 29 [NASA ADS] [CrossRef] [Google Scholar]
  28. Fraija, N., Veres, P., Beniamini, P., et al. 2021, ApJ, 918, 12 [NASA ADS] [CrossRef] [Google Scholar]
  29. Frail, D. A., Kulkarni, S. R., Nicastro, L., et al. 1997, Nature, 389, 261 [NASA ADS] [CrossRef] [Google Scholar]
  30. Frail, D. A., Metzger, B. D., Berger, E., et al. 2004, ApJ, 600, 828 [NASA ADS] [CrossRef] [Google Scholar]
  31. Frail, D. A., Soderberg, A. M., Kulkarni, S. R., et al. 2005, ApJ, 619, 994 [NASA ADS] [CrossRef] [Google Scholar]
  32. Galama, T. J., Vreeswijk, P. M., van Paradijs, J., et al. 1998, Nature, 395, 670 [Google Scholar]
  33. Gehrels, N., Ramirez-Ruiz, E., & Fox, D. B. 2009, ARA&A, 47, 567 [NASA ADS] [CrossRef] [Google Scholar]
  34. Geng, J.-J., Dai, Z.-G., Huang, Y.-F., et al. 2018, ApJ, 856, L33 [NASA ADS] [CrossRef] [Google Scholar]
  35. Ghirlanda, G., Salafia, O. S., Paragi, Z., et al. 2019, Science, 363, 968 [NASA ADS] [CrossRef] [Google Scholar]
  36. Giarratana, S., Giroletti, M., Marcote, B., et al. 2020, GRB Coordinates Network, Circular Service, 28939 [Google Scholar]
  37. Gompertz, B., Levan, A., Tanvir, N., et al. 2020, GRB Coordinates Network, Circular Service, 28822 [Google Scholar]
  38. Gordon, D., Jacobs, C., Beasley, A., et al. 2016, AJ, 151, 154 [NASA ADS] [CrossRef] [Google Scholar]
  39. Granot, J., & Sari, R. 2002, ApJ, 568, 820 [NASA ADS] [CrossRef] [Google Scholar]
  40. Greisen, E. W. 2003, Information Handling in Astronomy - Historical Vistas, 109 [CrossRef] [Google Scholar]
  41. Grossan, B., Maksut, Z., Kim, A., et al. 2020, GRB Coordinates Network, Circular Service, 28674 [Google Scholar]
  42. H.E.S.S. Collaboration (Abdalla, H., et al.) 2021, Science, 372, 1081 [NASA ADS] [CrossRef] [Google Scholar]
  43. Hinshaw, G., Larson, D., Komatsu, E., et al. 2013, ApJS, 208, 19 [Google Scholar]
  44. Hu, Y.-D., Fernandez-Garcia, E., Castro-Tirado, A. J., et al. 2020, GRB Coordinates Network, Circular Service, 28645 [Google Scholar]
  45. Izzo, L., Malesani, D. B., Zhu, Z. P., et al. 2020, GRB Coordinates Network, Circular Service, 28661 [Google Scholar]
  46. Japelj, J., Kopac, D., Kobayashi, S., et al. 2014, ApJ, 785, 84 [NASA ADS] [CrossRef] [Google Scholar]
  47. Jelinek, M., Strobl, J., Karpov, S., et al. 2020, GRB Coordinates Network, Circular Service, 28664 [Google Scholar]
  48. Jin, Z.-P., Covino, S., Della Valle, M., et al. 2013, ApJ, 774, 114 [NASA ADS] [CrossRef] [Google Scholar]
  49. Kalberla, P. M. W., Burton, W. B., Hartmann, D., et al. 2005, A&A, 440, 775 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  50. Kamble, A., Resmi, L., & Misra, K. 2007, ApJ, 664, L5 [NASA ADS] [CrossRef] [Google Scholar]
  51. Kann, D. A., Klose, S., & Zeh, A. 2006, ApJ, 641, 993 [NASA ADS] [CrossRef] [Google Scholar]
  52. Keimpema, A., Kettenis, M. M., Pogrebenko, S. V., et al. 2015, Exper. Astron., 39, 259 [NASA ADS] [CrossRef] [Google Scholar]
  53. Kennea, J. A., Tagliaferri, G., Campana, S., et al. 2020, GRB Coordinates Network, Circular Service, 28635 [Google Scholar]
  54. Kouveliotou, C., Meegan, C. A., Fishman, G. J., et al. 1993, ApJ, 413, L101 [NASA ADS] [CrossRef] [Google Scholar]
  55. Kouveliotou, C., Wijers, R. A. M. J., & Woosley, S. 2012, Gammaray Bursts, eds. Chryssa Kouveliotou, Ralph A. M. J. Wijers, Stan Woosley, (Cambridge, UK: Cambridge University Press) [CrossRef] [Google Scholar]
  56. Kumar, P., & Zhang, B. 2015, Phys. Rep., 561, 1 [Google Scholar]
  57. Kumar, H., Sahu, D. K., Gupta, R., et al. 2020a, GRB Coordinates Network, Circular Service, 28680 [Google Scholar]
  58. Kumar, H., Stanzin, U., Bhalerao, V., et al. 2020b, GRB Coordinates Network, Circular Service, 28681 [Google Scholar]
  59. Laskar, T., Alexander, K. D., Berger, E., et al. 2016, ApJ, 833, 88 [CrossRef] [Google Scholar]
  60. Laskar, T., Alexander, K. D., Berger, E., et al. 2018, ApJ, 862, 94 [NASA ADS] [CrossRef] [Google Scholar]
  61. Laskar, T., Alexander, K. D., Gill, R., et al. 2019, ApJ, 878, L26 [NASA ADS] [CrossRef] [Google Scholar]
  62. Lipunov, V., Gorbovskoy, E., Kornilov, V., et al. 2020a, GRB Coordinates Network, Circular Service, 28633 [Google Scholar]
  63. Lipunov, V., Gorbovskoy, E., Kornilov, V., et al. 2020b, GRB Coordinates Network, Circular Service, 28634 [Google Scholar]
  64. MAGIC Collaboration, Acciari, V. A., Ansoldi, S., et al. 2019, Nature, 575, 455 [CrossRef] [Google Scholar]
  65. Malesani, D. B., de Ugarte Postigo, A., & Pursimo, T. 2020, GRB Coordinates Network, Circular Service, 28637 [Google Scholar]
  66. Marcote, B., Ribó, M., Paredes, J. M., et al. 2020, GRB Coordinates Network, Circular Service, 29028 [Google Scholar]
  67. Markwardt, C. B., Barthelmy, S. D., Cummings, J. R., et al. 2020, GRB Coordinates Network, Circular Service, 28658 [Google Scholar]
  68. Marshall, F. E., D’Elia, V., & Swift/UVOT Team 2020, GRB Coordinates Network, Circular Service, 28662 [Google Scholar]
  69. Mazets, E. P., Golenetskii, S. V., Ilyinskii, V. N., et al. 1981, Ap&SS, 80, 119 [NASA ADS] [CrossRef] [Google Scholar]
  70. McMullin, J. P., Waters, B., Schiebel, D., et al. 2007, Astron. Data Anal. Softw. Syst. XVI, 376, 127 [Google Scholar]
  71. Mészáros, P. 2002, ARA&A, 40, 137 [CrossRef] [Google Scholar]
  72. Mészáros, P., & Rees, M. J. 1993, ApJ, 405, 278 [CrossRef] [Google Scholar]
  73. Mészáros, P., & Rees, M. J. 1994, MNRAS, 269, L41 [CrossRef] [Google Scholar]
  74. Mészáros, P., Rees, M. J., & Papathanassiou, H. 1994, ApJ, 432, 181 [CrossRef] [Google Scholar]
  75. Minaev, P., & Pozanenko, A. 2020, GRB Coordinates Network, Circular Service, 28668 [Google Scholar]
  76. Misra, K., Resmi, L., Kann, D. A., et al. 2021, MNRAS, 504, 5685 [NASA ADS] [Google Scholar]
  77. Monet, D. G., Levine, S. E., Canzian, B., et al. 2003, AJ, 125, 984 [Google Scholar]
  78. Mooley, K. P., Deller, A. T., Gottlieb, O., et al. 2018, Nature, 561, 355 [Google Scholar]
  79. Moskvitin, A. S., Aitov, V. N., & GRB follow-up Team 2020, GRB Coordinates Network, Circular Service, 28721 [Google Scholar]
  80. Norris, J. P., Cline, T. L., Desai, U. D., et al. 1984, Nature, 308, 434 [NASA ADS] [CrossRef] [Google Scholar]
  81. Paczyński, B. 1998, ApJ, 494, L45 [Google Scholar]
  82. Perley, D. A., Cenko, S. B., Corsi, A., et al. 2014, ApJ, 781, 37 [NASA ADS] [CrossRef] [Google Scholar]
  83. Piran, T. 2004, Rev. Mod. Phys., 76, 1143 [Google Scholar]
  84. Pozanenko, A., Belkin, S., Volnova, A., et al. 2020, GRB Coordinates Network, Circular Service, 29033 [Google Scholar]
  85. Pradel, N., Charlot, P., Lestrade, J.-F. 2006, A&A, 452, 1099 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  86. Rastinejad, J., Paterson, K., Kilpatrick, C. D., et al. 2020, GRB Coordinates Network, Circular Service, 28676 [Google Scholar]
  87. Rees, M. J., & Mészáros, P. 1992, MNRAS, 258, 41 [Google Scholar]
  88. Rees, M. J., & Mészáros, P. 1994, ApJ, 430, L93 [Google Scholar]
  89. Rhodes, L., Fender, R., Bray, J., et al. 2020, GRB Coordinates Network, Circular Service, 28945 [Google Scholar]
  90. Rhodes, L., van der Horst, A. J., Fender, R., et al. 2020, MNRAS, 496, 3326 [Google Scholar]
  91. Rossi, A., Benetti, S., Palazzi, E., et al. 2021, GRB Coordinates Network, Circular Service, 29306 [Google Scholar]
  92. Salafia, O. S., Ravasio, M. E., Yang, J., et al. 2022, ApJL, 931, L19 [NASA ADS] [CrossRef] [Google Scholar]
  93. Sari, R., & Esin, A. A. 2001, ApJ, 548, 787 [NASA ADS] [CrossRef] [Google Scholar]
  94. Schlafly, E. F., & Finkbeiner, D. P. 2011, ApJ, 737, 103 [Google Scholar]
  95. Selsing, J., Fynbo, J. P. U., Heintz, K. E., et al. 2019, GRB Coordinates Network, Circular Service, 23695 [Google Scholar]
  96. Shepherd, M. C., Pearson, T. J., & Taylor, G. B. 1994, BAAS, 26, 987 [NASA ADS] [Google Scholar]
  97. Sironi, L., Spitkovsky, A., & Arons, J. 2013, ApJ, 771, 54 [Google Scholar]
  98. Suda, Y., Artero, M., Asano, K., et al. 2021, Proceedings of 37th International Cosmic Ray Conference - PoS(ICRC2021), 395, 797 [CrossRef] [Google Scholar]
  99. Taylor, G. B., Carilli, C. L., & Perley, R. A. 1999, Synthesis Imaging in Radio Astronomy II, 180 [Google Scholar]
  100. Taylor, G. B., Frail, D. A., Berger, E., et al. 2004, ApJ, 609, L1 [NASA ADS] [CrossRef] [Google Scholar]
  101. Tody, D. 1993, Astronomical Data Analysis Software and Systems II, 52, 173 [Google Scholar]
  102. Usov, V. V. 1992, Nature, 357, 472 [NASA ADS] [CrossRef] [Google Scholar]
  103. Valeev, A. F., Castro-Tirado, A. J., Hu, Y.-D., et al. 2019, GRB Coordinates Network, Circular Service, 25565 [Google Scholar]
  104. van der Horst, A. J., Paragi, Z., de Bruyn, A. G., et al. 2014, MNRAS, 444, 3151 [NASA ADS] [CrossRef] [Google Scholar]
  105. Veres, P., Corsi, A., Frail, D. A., et al. 2015, ApJ, 810, 31 [NASA ADS] [CrossRef] [Google Scholar]
  106. Vielfaure, J.-B., Izzo, L., Xu, D., et al. 2020, GRB Coordinates Network, Circular Service, 29077 [Google Scholar]
  107. Vreeswijk, P. M., Kann, D. A., Heintz, K. E., et al. 2018, GRB Coordinates Network, Circular Service, 22996 [Google Scholar]
  108. Wang, X.-Y., Liu, R.-Y., Zhang, H.-M., et al. 2019, ApJ, 884, 117 [NASA ADS] [CrossRef] [Google Scholar]
  109. Woosley, S. E. 1993, ApJ, 405, 273 [Google Scholar]
  110. Willingale, R., Starling, R. L. C., Beardmore, A. P., et al. 2013, MNRAS, 431, 394 [NASA ADS] [CrossRef] [Google Scholar]
  111. Wilms, J., Allen, A., & McCray, R. 2000, ApJ, 542, 914 [Google Scholar]
  112. Woosley, S. E., & Heger, A. 2006, ApJ, 637, 914 [CrossRef] [Google Scholar]
  113. Zhang, B., & Mészáros, P. 2001, ApJ, 559, 110 [NASA ADS] [CrossRef] [Google Scholar]
  114. Zhang, B., & Mészáros, P. 2004, Int. J. Mod. Phys. A, 19, 2385 [NASA ADS] [CrossRef] [Google Scholar]
  115. Zhang, W., & MacFadyen, A. 2009, ApJ, 698, 1261 [NASA ADS] [CrossRef] [Google Scholar]
  116. Zhang, L.-L., Ren, J., Huang, X.-L., et al. 2021, ApJ, 917, 95 [NASA ADS] [CrossRef] [Google Scholar]
  117. Zhu, Z. P., Liu, X., Fu, S. Y., et al. 2020a, GRB Coordinates Network, Circular Service, 28653 [Google Scholar]
  118. Zhu, Z. P., Liu, X., Fu, S. Y., et al. 2020b, GRB Coordinates Network, Circular Service, 28677 [Google Scholar]
2

The Astronomical Image Processing System (AIPS) is a software package produced and maintained by the National Radio Astronomy Observatory (NRAO).

All Tables

Table 1

Radio observations performed with the VLA, e-MERLIN, and EVN in the L and C bands.

In the text
Table 2

Constraints on the model parameters at 1 day for a homogeneous circumburst medium.

In the text
Table 3

Global and microphysical parameters for GRB 201015A in the ISM scenario.

In the text
Table 4

Global and microphysical parameters for GRB 201015A in the ISM scenario if 0.05 ≤ ϵe ≤ 0.15.

In the text

All Figures

thumbnail Fig. 1

e-MERLIN detection on 2020 November 5. The synthesised beam is shown in the lower left corner.
In the text
thumbnail Fig. 2

Multi-wavelength afterglow light curves (see Sect. 3). For each band, the light curves predicted by the standard model with vsa = 13 GHz, vm = 6 GHz, vc = 2 × 107 GHz, Fm = 800 μJy, and p = 2.05 at 1 day for a homogeneous surrounding medium are shown: 1.5 GHz (orange); 5 GHz (blue); r, g, i, and z bands (red, lime, violet, and brown, respectively); integrated X-ray light curve (dark blue). The green, orange, and brown vertical lines pinpoint the epochs of the spectra at 0.12, 1.41, and 23 days, respectively (see also Fig. 3). The dashed line shows a simple model for the SN contribution in the r band (see Sect. 5.1).
In the text
thumbnail Fig. 3

Spectra at 0.12 (green), 1.41 (orange), and 23 (brown) days after the GRB onset for a homogeneous surrounding medium with vsa = 13 GHz, vm = 6 GHz, vc = 2 × 107 GHz, Fm = 800 μJy, and p = 2.05 at 1 day. Spectrum at 0.12 days: optical observations from Pozanenko et al. (2020) and XRT butterfly plot. Spectrum at 1.41 days: our VLA and MMT detections. Spectrum at 23 days: our 1.5 and 5 GHz observations, optical data from Pozanenko et al. (2020) and the XRT butterfly plot; the optical emission is dominated by the SN component.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.