No Arabic abstract
We report on NICER observations of the Magnetar SGR~1935+2154, covering its 2020 burst storm and long-term persistent emission evolution up to $sim90$ days post outburst. During the first 1120~seconds taken on April 28 00:40:58 UTC we detect over 217 bursts, corresponding to a burst rate of $>0.2$ bursts s$^{-1}$. Three hours later the rate is at 0.008 bursts s$^{-1}$, remaining at a comparatively low level thereafter. The $T_{90}$ burst duration distribution peaks at 840~ms; the distribution of waiting times to the next burst is fit with a log-normal with an average of 2.1 s. The 1-10 keV burst spectra are well fit by a blackbody, with an average temperature and area of $kT=1.7$ keV and $R^2=53$ km$^2$. The differential burst fluence distribution over $sim3$ orders of magnitude is well modeled with a power-law form $dN/dFpropto F^{-1.5pm0.1}$. The source persistent emission pulse profile is double-peaked hours after the burst storm. We find that the bursts peak arrival times follow a uniform distribution in pulse phase, though the fast radio burst associated with the source aligns in phase with the brighter peak. We measure the source spin-down from heavy-cadence observations covering days 21 to 39 post-outburst, $dot u=-3.72(3)times10^{-12}$ Hz s$^{-1}$; a factor 2.7 larger than the value measured after the 2014 outburst. Finally, the persistent emission flux and blackbody temperature decrease rapidly in the early stages of the outburst, reaching quiescence 40 days later, while the size of the emitting area remains unchanged.
We report on INTEGRAL observations of the soft $gamma$-ray repeater SGR 1935+2154 performed between 2020 April 28 and May 3. Several short bursts with fluence of $sim10^{-7}-10^{-6}$ erg cm$^{-2}$ were detected by the IBIS instrument in the 20-200 keV range. The burst with the hardest spectrum, discovered and localized in real time by the INTEGRAL Burst Alert System, was spatially and temporally coincident with a short and very bright radio burst detected by the CHIME and STARE2 radio telescopes at 400-800 MHz and 1.4 GHz, respectively. Its lightcurve shows three narrow peaks separated by $sim$29 ms time intervals, superimposed on a broad pulse lasting $sim$0.6 s. The brightest peak had a delay of 6.5$pm$1.0 ms with respect to the 1.4 GHz radio pulse (that coincides with the second and brightest component seen at lower frequencies). The burst spectrum, an exponentially cut-off power law with photon index $Gamma=0.7_{-0.2}^{+0.4}$ and peak energy $E_p=65pm5$ keV, is harder than those of the bursts usually observed from this and other magnetars. By the analysis of an expanding dust scattering ring seen in X-rays with the {it Neil Gehrels Swift Observatory} XRT instrument, we derived a distance of 4.4$_{-1.3}^{+2.8}$ kpc for SGR 1935+2154, independent of its possible association with the supernova remnant G57.2+0.8. At this distance, the burst 20-200 keV fluence of $(6.1pm 0.3)times10^{-7}$ erg cm$^{-2}$ corresponds to an isotropic emitted energy of $sim1.4times10^{39}$ erg. This is the first burst with a radio counterpart observed from a soft $gamma$-ray repeater and it strongly supports models based on magnetars that have been proposed for extragalactic fast radio bursts.
A few years after its discovery as a magnetar, SGR J1935+2154 started a new burst-active phase on 2020 April 27, accompanied by a large enhancement of its X-ray persistent emission. Radio single bursts were detected during this activation, strengthening the connection between magnetars and fast radio bursts. We report on the X-ray monitoring of SGR J1935+2154 from ~3 days prior to ~3 weeks after its reactivation, using Swift, NuSTAR, and NICER. We detected X-ray pulsations in the NICER and NuSTAR observations, and constrained the spin period derivative to |Pdot| < 3e-11 s/s (3 sigma c.l.). The pulse profile showed a variable shape switching between single and double-peaked as a function of time and energy. The pulsed fraction decreased from ~34% to ~11% (5-10 keV) over ~10 days. The X-ray spectrum was well fit by an absorbed blackbody model with temperature decreasing from kT ~ 1.6 to 0.45-0.6 keV, plus a non-thermal component (Gamma ~ 1.2) observed up to ~25 keV with NuSTAR. The 0.3-10 keV X-ray luminosity (at 6.6 kpc) increased in less than four days from ~ 6e33 erg/s to about 3e35 erg/s and then decreased again to 2.5e34 erg/s over the following three weeks of the outburst. We also detected several X-ray bursts, with properties typical of short magnetar bursts.
Fast radio bursts (FRBs) are millisecond-duration, bright radio signals (fluence $mathrm{0.1 - 100,Jy,ms}$) emitted from extragalactic sources of unknown physical origin. The recent CHIME/FRB and STARE2 detection of an extremely bright (fluence $sim$MJy$,$ms) radio burst from the Galactic magnetar SGR~1935$+$2154 supports the hypothesis that (at least some) FRBs are emitted by magnetars at cosmological distances. In follow-up observations totalling 522.7$,$hrs on source, we detect two bright radio bursts with fluences of $112pm22mathrm{,Jy,ms}$ and $24pm5mathrm{,Jy,ms}$, respectively. Both bursts appear affected by interstellar scattering and we measure significant linear and circular polarisation for the fainter burst. The bursts are separated in time by $sim$1.4$,$s, suggesting a non-Poissonian, clustered emission process -- similar to what has been seen in some repeating FRBs. Together with the burst reported by CHIME/FRB and STARE2, as well as a much fainter burst seen by FAST (fluence 60$mathrm{,mJy,ms}$), our observations demonstrate that SGR 1935+2154 can produce bursts with apparent energies spanning roughly seven orders of magnitude, and that the burst rate is comparable across this range. This raises the question of whether these four bursts arise from similar physical processes, and whether the FRB population distribution extends to very low energies ($sim10^{30},$erg, isotropic equivalent).
During April and May 2020, SGR J1935+2154 emitted hundreds of short bursts and became one of the most prolific transient magnetars. At the onset of the active bursting period, a 130-s burst ``forest, which included some bursts with peculiar time profiles, were observed with the $Fermi$/Gamma-ray Burst Monitor. In this paper, we present the results of time-resolved spectral analysis of this burst ``forest episode, which occurred on April 27, 2020. We identify thermal spectral components prevalent during the entire 130-s episode; high-energy maxima appear during the photon flux peaks, which are modulated by the spin period of the source. Moreover, the evolution of the $ u F_{ u}$ spectral hardness (represented by $E_{rm peak}$ or blackbody temperature) within the lightcurve peaks is anti-correlated with the pulse phases extrapolated from the pulsation observed within the persistent soft X-ray emission of the source six hours later. Throughout the episode, the emitting area of the high-energy (hotter) component is 1--2 orders of magnitude smaller than that for the low-energy component. We interpret this with a geometrical viewing angle scenario, inferring that the high-energy component likely originates from a low-altitude hotspot located within closed toroidal magnetic field lines.
Using numerical simulations we show that low-amplitude Alfven waves from a magnetar quake propagate to the outer magnetosphere and convert to plasmoids (closed magnetic loops) which accelerate from the star, driving blast waves into the magnetar wind. Quickly after its formation, the plasmoid becomes a thin relativistic pancake. It pushes out the magnetospheric field lines, and they gradually reconnect behind the pancake, generating a variable wind far stronger than the normal spindown wind of the magnetar. Repeating ejections drive blast waves in the amplified wind. We suggest that these ejections generate the simultaneous X-ray and radio bursts detected from SGR 1935+2154. A modest energy budget of the magnetospheric perturbation $sim 10^{40}$ erg is sufficient to produce the observed bursts. Our simulation predicts a narrow (a few ms) X-ray spike from the magnetosphere, arriving almost simultaneously with the radio burst emitted far outside the magnetosphere. This timing is caused by the extreme relativistic motion of the ejecta.