Hoinga: A supernova remnant discovered in the SRG/eROSITA All-Sky Survey eRASS1
W. Becker, N. Hurley-Walker, Ch. Weinberger, L. Nicastro, M.G.F.Mayer, A. Merloni, J. Sanders
AAstronomy & Astrophysics manuscript no. main © ESO 2021March 1, 2021
Hoinga: A supernova remnant discovered in the SRG/eROSITAAll-Sky Survey eRASS1
W. Becker , (cid:63) , N. Hurley-Walker , Ch. Weinberger , L. Nicastro , M. G. F. Mayer , A. Merloni , J. Sanders Max-Planck-Institut für extraterrestrische Physik, Giessenbachstraße, 85748 Garching, Germany Max-Planck-Institut für Radioastronomie, Auf dem Hügel 69, 53121 Bonn, Germany International Centre for Radio Astronomy Research, Curtin University, Bentley WA 6102, Australia INAF – Osservatorio di Astrofisica e Scienza dello Spazio di Bologna, Via Piero Gobetti 93 /
3, I-40129 Bologna, ItalyReceived 17.12.2020; accepted 12.02.2021
ABSTRACT
Supernova remnants (SNRs) are observable for about (6 − × years before they fade into the Galactic interstellar medium. Witha Galactic supernova rate of approximately two per century, we can expect to have of the order of 1200 SNRs in our Galaxy. However,only about 300 of them are known to date, with the majority having been discovered in Galactic plane radio surveys. Given thatthese SNRs represent the brightest tail of the distribution and are mostly located close to the plane, they are not representative of thecomplete sample. The launch of the Russian-German observatory SRG / eROSITA in July 2019 brought a promising new opportunityto explore the universe. Here we report findings from the search for new SNRs in the eROSITA all-sky survey data which led tothe detection of one of the largest SNRs discovered at wavelengths other than the radio: G249.5 + ◦ .
4’ and shows a circular shaped morphology with di ff use X-ray emission filling almost the entire remnant. Spectral analysis of theremnant emission reveals that an APEC spectrum from collisionally ionised di ff use gas and a plane-parallel shock plasma model withnon-equilibrium ionisation are both able to provide an adequate description of the data, suggesting a gas temperature of the order ofkT = . + . − . keV and an absorbing column density of N H = . + . − . × cm − . Various X-ray point sources are found to be locatedwithin the remnant boundary but none seem to be associated with the remnant itself. Subsequent searches for a radio counterpartof the Hoinga remnant identified its radio emission in archival data from the Continuum HI Parkes All-Sky Survey (CHIPASS) andthe 408-MHz ‘Haslam’ all-sky survey. The radio spectral index α = − . ± .
08 obtained from these data definitely confirms theSNR nature of Hoinga. We also analysed INTEGRAL SPI data for fingerprints of Ti emission, which is an ideal candidate withwhich to study nucleosynthesis imprinting in young SNRs. Although no Ti emission from Hoinga was detected, we were able to seta 3 σ upper flux limit of 9 . × − ph cm − s − . From its size and X-ray and radio spectral properties we conclude that Hoinga is amiddle-aged Vela-like SNR located at a distance of about twice that of the Vela SNR, i.e. at ∼
500 pc.
Key words.
Stars: supernovae: general – Stars: supernovae: individual: G249.5 +
1. Introduction
A long series of observations have taught astronomers that thereare many di ff erent types of stars. Findings in atomic and nu-clear physics have made it possible to understand the develop-ment of these stars over the past few decades. According to this,the fate of a star at the end of its thermonuclear evolution essen-tially depends on only one parameter: the mass of a star decideswhether its death is gentle or violent. More massive stars with M ≥ (cid:12) end their lives with a supernova (SN) explosion, which isnot only often associated with the formation of other exotic startypes such as neutron stars (NSs) or black holes, but also rep-resents a new beginning of stellar evolution by enrichment anddecompression of the surrounding interstellar medium. A promi-nent example for this is the Solar System itself which shows im-prints in metal abundance of a past SN which took place 4.567Gyr ago (Gritschneder et al. 2012).Supernovae are considered to be rare events which happenin our Milky Way on average every 30 −
50 years (e.g. Keane & (cid:63)
E-mail: [email protected]
Kramer 2008), though no SN event has been directly observed inour Galaxy in the past 400 years. Indeed, in the past two millen-nia, only seven Galactic SN are the subject of historical records:SN 185 (RCW 86), SN 386 (G11.2–0.3), SN 1006, SN 1054(Crab), SN 1181 (3C58), SN 1572 (Tycho), and SN 1604 (Ke-pler); see also Stephenson (2017) and references therein. How-ever, there are additional promising candidates discussed in theliterature, such as for example CAS A (Green & Stephenson2017) and Vela-Jr(Aschenbach 1998).Certainly, visible-band extinction of the SN emission and itsdistance to earth plays a crucial role when it comes to recog-nising a SN with the naked eye. A prominent example of thise ff ect is demonstrated by the missing reports of the CAS A SNevent which is believed to have taken place about 300 years ago.No widespread reports of CAS A exist in the literature of the17th century (cf. Hartmann et al. 1997). A more recent exam-ple of an unrecognised SN is that of the youngest SN known inour Galaxy, G1.9 + Article number, page 1 of 12 a r X i v : . [ a s t r o - ph . H E ] F e b & A proofs: manuscript no. main
In contrast to SNe which are only observable on a timescaleof months to years, their remnants (SNRs) are detectable over alarge range of the electromagnetic bands for more than 60 000 to150 000 years. However, today only about 300 SNRe are known(cf. Green 2019), most of which were discovered in Galacticplane radio surveys. Assuming that the radio lifetime of a SNRbright enough to be detected with current radio telescopes is atleast about 60 kyr (Frail et al. 1994), there is a discrepancy bya factor of between four and six between the observed and ex-pected number of SNRs. Even if one takes into account the factthat very massive stars may form a black hole without a lumi-nous SN (e.g. Kochanek et al. 2008; Adams et al. 2017) thereis still a significant mismatch between the expected and knownnumber of SNRs. The discrepancy is possibly explained by thefact that the radio sample of SNRs is not complete. Reasons thatmay prevent a radio bright remnant from being detected in radiosurveys are various: – A SN shock wave may expand within the hot phase of theISM and reach a very large diameter until it has swept up suf-ficient mass from the low-density gas to form a radio shell.Density inhomogeneities in such a large volume will causedistortions in the shell and can make the identification as aSNR rather di ffi cult, in particular in the presence of confus-ing unrelated emission from other nearby sources in the sameregion of the sky. – A SN shock wave may expand in a very dense medium, mak-ing the SNR lifetime rather short, because material is quicklyswept up and decelerated. Such an environment is likely tobe relevant for example for massive star members of OB-associations that are surrounded by dense molecular cloudsand warm gas. Even during their short lifetime, such eventsare di ffi cult to identify within the strong thermal radio emis-sion from those regions. – There is a strong bias towards bright resolved objects in ob-servations towards the inner Galaxy. – Low-surface-brightness SNRs are easily missed in radio sur-veys if they are below the sensitivity limits of the surveys orif they are confused with other objects in the same area. – Old SNRs which are in the phase of dissolving into the ISMmay have incomplete radio shells that may then prevent thesesources from being identified as SNRs. – SNRs located away from the Galactic plane are easily missedin radio surveys, as this area is where these events are typi-cally targeted.Given these selection e ff ects in radio surveys and the de-tection of unknown SNRs in previous X-ray surveys (e.g. Pf-e ff ermann & Aschenbach 1996; Busser et al. 1996; Asaoka &Aschenbach 1994; Asaoka et al. 1996; Egger et al. 1996; Fol-gheraiter et al. 1996), as well as the detection of more than 70highly significant SNR candidates in our analysis of the ROSATAll-Sky-Survey data, it was deemed worthy to start searching forundiscovered SNRs in the first eROSITA All-Sky Survey RASS1(Predehl et al. 2020a).In this paper we report the discovery of the SNRG249.5 + ◦ . Ti decay radiation associated with the remnant. In §5we summarise and discuss our results.
2. X-ray observations and data analysis
The German-built X-ray telescope eROSITA (extended Rönt-gen Survey Imaging Telescope Array) is one of two instrumentson the Russian-German observatory SRG (Spectrum Röntgen-Gamma; Sunyaev et al., 2020). eROSITA consists of sevenaligned X-ray telescopes (TM1 − TM7), each nested with 54gold-coated mirror shells which have a focal length of 1600 mm.All telescopes observe the same sky-region simultaneously in the0.2–8 keV band-pass though each focuses the collected X-rayson its own pn-CCD camera (Meidinger et al. 2014). The latteris an improved version of the pn-CCD camera aboard XMM-Newton (Strüder et al. 2001). eROSITA has a spectral resolutionof ∼
70 eV at 1 keV and a temporal resolution of 50 ms. Itsfield of view (FOV) is 1 ◦ . The on-axis e ff ective area of all seventelescopes combined is slightly higher than that of the XMM-Newton pn + MOS cameras in the key 0.5-2.0 keV band-pass.In pointing mode (on axis) the angular resolution of eROSITA is18 (cid:48)(cid:48) (HEW) whereas in survey mode it is 26 (cid:48)(cid:48) (FOV averaged).Source location accuracy is of the order of 4 (cid:48)(cid:48) . σ ). The sec-ond instrument onboard SRG is the Russian X-ray concentratorM ikhail P avlinsky ART-XC (Astronomical Röntgen Telescope– X-ray Concentrator) (Pavlinsky et al. 2018), which is sensitivein the hard X-ray band from 4 up to 30 keV, making it comple-mentary to the eROSITA soft band.The SRG was launched into an L2 orbit on July 13, 2019,with a Russian Proton-M launch vehicle. After a three-monthcalibration and science verification phase it started its first all-sky survey on December 13, 2019. With a scan rate of 0 . / s, a spacecraft revolution duration of 4 h and a central FOVpassage time of about 40 s (Predehl et al. 2020a), each surveytakes 6 months to complete. eROSITA is supposed to take eightall-sky surveys over a time period of 4 years.The X-ray data we report here were taken during the firsteROSITA all-sky survey eRASS1, completed on June 12, 2020.As the main science driver of the SGR mission is to explore thenature of dark energy, its orbit was chosen so that the Eclipticpoles get the deepest exposure, leading to an exposure of theGalactic plane which is of the order of ∼ −
300 seconds persurvey.First results of eRASS1, including a fascinating, detail-rich,three-energy-band colour-coded image of the 0.3-8.0 keV X-raysky, were recently released. The survey represents the deepestview of the whole X-ray sky today and led to the discovery ofthe large-scale symmetric hot-gas structures in the Milky Wayhalo, called ‘eROSITA Bubbles’ (Predehl et al. 2020b), amongmany other exciting results.Searching this survey map for unknown extended sourcesrevealed the existence of a new SNR at Galactic coordinates l = ◦ . b = ◦ .
5, labelled G249.5 + a . Figure 1 depicts a colour-coded image of the relevantsky region which shows Hoinga with its neighbours the AntliaLoop and the Vela SNR.The data we use in our analysis were processed by the eSASS(eROSITA Standard Analysis Software) pipeline and have the a In honor of the first author’s hometown Bad Hönningen am Rhein:Hoinga was its medieval name.Article number, page 2 of 12.Becker, et al.: Hoinga - A SNR discovered in eRASS1
Fig. 1.
Cutout of the SRG / eROSITA all-sky survey image from eRASS1data. The image shows, among many other sources, the extended X-ray emission from the 24 ◦ diameter large Antlia Loop in its upper leftquadrant and the emission from the Vela SNR in its lower right. Theemission from the Hoinga SNR in the upper right quadrant of the imageis indicated. The image is an Aito ff projection of photons that have beencolour-coded according to their energy (red for energies 0.3–0.6 keV,green for 0.6–1 keV, blue for 1–2.3 keV). The image was smoothed witha 10 (cid:48) FWHM Gaussian filter and is publicly available on the internet. processing number b . Within theeSASS pipeline, X-ray data of the eRASS sky are divided into4700 partly overlapping sky tiles of 3 ◦ . × ◦ . (cid:48)(cid:48) whichreflects eROSITA’s FOV averaged angular resolution in surveymode. Data from all seven telescopes were used as we did notnotice a significant impact of the light leak in TM5 and TM7.In order to enhance the visibility of Hoinga’s di ff use emissionin these images whilst leaving point sources unsmoothed to thegreatest possible extent we applied the adaptive kernel smooth-ing algorithm of Ebeling et al. (2006) with a Gaussian kernel of4 . σ .The image analysis clearly reveals that Hoinga’s X-ray emis-sion is very soft. The majority of its emission is detected in the b cf. https: // erosita.mpe.mpg.de / Fig. 2.
Hoinga SNR as seen in the eROSITA all-sky survey eRASS1.Photons to produce this 7 ◦ . × ◦ . = = − After the discovery of Hoinga in eRASS1 data we went backto the archival ROSAT all-sky survey (RASS) to check whetherthe remnant was detected. The ROSAT RASS was performedbetween June 1990 and August 1991, almost exactly 30 yearsbefore eRASS1. The ROSAT PSPC (position-sensitive propor-tional counter), which was in the focal plane during the survey,was sensitive in the 0.1-2.4 keV energy range (Pfe ff ermann et al.2003). The angular resolution in the survey was 45 (cid:48)(cid:48) . RASS dataare divided into 1378 partly overlapping sky tiles, each cover-ing 6 ◦ . × ◦ . Article number, page 3 of 12 & A proofs: manuscript no. main
Fig. 3.
The Hoinga SNR as seen in the ROSAT all-sky survey. Pho-tons to produce this image have been selected from within the 0.1–0.7keV energy band. A Gaussian smoothing filter with x,y- σ = ff use emis-sion. The gray scale colors are distributed so that white corresponds toa pixel intensity value of 0.09 and black to 0.45 cts / pixel. The image isvignetting and deadtime corrected though no exposure correction wasapplied. The inset in the lower right corner shows a 40 (cid:48) × (cid:48) zoom tothe region of the X-ray sources located slightly to east of the remnant’sgeometrical centre. hard bands, the soft-band image clearly shows a hint of circu-lar shaped soft X-ray emission. As in the eROSITA data, its softX-ray emission is brighter toward the south. Figure 3 shows theRASS soft-band image of the relevant sky region. The e ff ectivesurvey exposure in the image varies from about 480 s at the east-ern side of the remnant to about 474 s near to its central regionand 380 s at its western side. ROSAT’s scan direction imprintin that sky region is clearly visible in the image by the slightlyinhomogeneous exposure, from approximately the southwesternto the northeastern direction. In order to identify a possible compact remnant associatedwith Hoinga we applied a source detection to the ROSAT andeROSITA survey data. The point sources detected in both sur-veys along with their properties are summarised in Table 1. TheeROSITA 68% position uncertainty for point sources detected inthe all-sky survey eRASS1 is 4 (cid:48)(cid:48) .
5; for the ROSAT survey it is13 (cid:48)(cid:48) (Voges et al. 1999). In Figures 2 and 3, three point sourcescan be seen slightly to the east of the remnant geometrical cen-tre, though rather centred with respect to the di ff use X-ray emis-sion. While the positions of sources ff set towards the southeast of almost 20 (cid:48)(cid:48) . Forthe purpose of a further source identification, we correlated theeROSITA positions with various radio and optical catalogues,for example NVSS (Condon et al. 1998) and Gaia DR2 (GaiaCollaboration et al. 2018). Table 1.
X-ray sources detected within the Hoinga SNR in eROSITAeRASS1 and ROSAT RASS data. The detection significance of thelisted sources is ≥ σ . The position uncertainty of eROSITA pointsources is 4 (cid:48)(cid:48) . σ confidence). < MJD > is the Modified Julian Dateof the observation in eRASS1 and RASS, respectively. The numberingfor the centrally located sources − Source RA (J2000) Dec (J2000) Obs. Timeh:m:s d:m:s < MJD > eROSITA eRASS11 09:33:41.096 -17:09:18.932 58987.402822 09:33:18.088 -17:14:41.741 58987.319463 09:34:30.071 -17:21:21.224 58987.652744 09:37:57.489 -17:10:14.453 58988.236305 09:27:29.469 -18:06:20.653 58986.485506 09:27:50.522 -16:40:01.672 58985.986147 09:36:25.814 -18:21:05.829 58988.402578 09:28:45.469 -15:24:10.805 58985.736679 09:28:38.097 -15:21:08.361 58985.7368210 09:40:02.199 -17:09:55.614 58988.6531011 09:26:58.572 -16:30:06.584 58985.73626ROSAT RASS1 09:33:41.421 -17:08:59.602 48210.593652 09:33:18.151 -17:14:39.386 48210.526873 09:34:30.177 -17:21:17.739 48210.860654 09:37:57.671 -17:10:07.991 48211.727997 09:36:26.231 -18:21:05.791 48211.8279310 09:40:02.494 -17:09:58.469 48212.2621511 09:26:58.256 -16:30:01.930 48208.55845For the eROSITA sources , , , c . For all sources, we find a close overlap with opticalsources from the Gaia DR2 catalogue d . From the proper motionand parallax information for the potential counterparts, it seemslikely that sources , , , , and , , , , , and (cid:48)(cid:48) found on average(68% confidence) for ROSAT RASS sources (Voges et al. 1999).Assuming a real o ff set for source ∼ (cid:48)(cid:48) /
30 years, which seems unlikely to us as we did notfind a nearby bright star as optical counterpart. Indeed, of the 11X-ray sources detected within the Hoinga SNR, none have an op-tical counterpart fainter than the twentieth magnitude in the GaiaG-band. Similarly, in the infrared band where the fainter objectis found, H (cid:39)
16 and W2 (4 . µ m) (cid:39)
14 in 2MASS and WISEcatalogues, respectively. We therefore conclude that all 11 X-raysources are either foreground or background objects which arenot associated with Hoinga. c https: // / nvss d https: // gea.esac.esa.int / archiveArticle number, page 4 of 12.Becker, et al.: Hoinga - A SNR discovered in eRASS1 In order to properly correct the source spectrum and energy fluxfor contributions from the instrument- and sky-background, weanalysed a sky field of about 8 ◦ × ◦ centred on the remnant.Hoinga’s energy spectrum was extracted from the eROSITAeRASS1 data by selecting all events recorded within an ellip-tical region of semi-minor and major axis of 2 ◦ . ◦ .
35, re-spectively. The elliptical region was centred at the position RA = = − ◦ .
5. SAOI m - age ds ff erence between its inner and outer region of 0 ◦ . ff erlight leaks related to the sun–satellite angle, making their soft-response calibration quite uncertain at that early stage of themission. Model spectra were simultaneously fitted to Hoinga’ssource and background spectra. We used Xspec 12.10.1f (Dor-man et al. 2003) and applied the C-statistics to the fits inwhich we modelled the source and background spectra indepen-dently. Of the fitted model spectra, the APEC spectrum fromcollisionally ionised di ff use gas (Foster et al. 2012) and thePSHOCK model (Plane-parallel SHOCK plasma model withnon-equilibrium ionisation) (Borkowski et al. 2001) were foundto provide fits of equal goodness and with similar spectral pa-rameters to the observed spectrum. We used the abundance tableand the TBabs absorption model from Wilms et al. (2000). Forthe meaning of the fitted spectral parameters, we refer the readerto the Xspec manual e and references therein.Figure 4 depicts the best-fit APEC model. The model spec-trum folded through the detector response is shown with ablack solid line. Table 2 lists the best-fit spectral parameters ofboth models. Due to the preliminary calibration status of theeROSITA instruments at the time of writing, we refrain from giv-ing absolute energy fluxes as obtained from the best-fit models.The contour plot shown in Figure 5 gives the parameter depen-dence of the temperature versus the column absorption for theAPEC model.
3. Radio observations and data analysis
The Murchison Widefield Array (MWA; Tingay et al. 2013;Wayth et al. 2018) is a low-frequency radio telescope operat-ing in Western Australia, and is a precursor to the low-frequencycomponent of the Square Kilometre Array. The GaLactic andExtragalactic All-sky MWA (GLEAM; Wayth et al. 2015) sur-vey observed the whole sky south of declination (Dec) + ◦ from 2013 to 2015 between 72 and 231 MHz. A major datarelease covering 24 402 square degrees of extragalactic skywas published by Hurley-Walker et al. (2017), while individualstudies have published smaller regions such as the Magellanic e https: // heasarc.gsfc.nasa.gov / xanadu / xspec / XspecManual.pdf
Fig. 4.
Energy spectrum of the Hoinga SNR as observed with theeROSITA TM1, 2, 3, 4, and 6 telescope and detector units and simulta-neously fitted to an absorbed APEC spectral model ( upper panel ). Thespectra have been binned for visual clarity and plotting purposes. Thesignal-to-noise ratio in each bin is 15 σ . The folded best-fit APEC spec-tral model is plotted as a solid black line. Fit residuals are shown in thelower panel. Fig. 5.
Contour plot showing the relative parameter dependence of thefitted spectral parameters kT (temperature) vs. N H (column absorption)for the APEC model fit to the energy spectrum of Hoinga. The threecontours represent the 1 σ , 2 σ, and 3 σ confidence levels for two param-eters of interest. The small red dot marks the best-fit position.Article number, page 5 of 12 & A proofs: manuscript no. main
Fig. 6. ∼ ◦ × ◦ of the region surrounding Hoinga as seen by GLEAM at 103-134 MHz (R), 139-170 MHz (G), and 170-200 MHz (B). Theleft panel shows the image from the data release of Hurley-Walker et al. (2017), and the right panel shows the region after reprocessing to subtractsources and highlight large-scale structure (see Section 3.1). Hoinga is visible as an ellipse in the centre of the image; steep-spectrum Galacticcirrus becomes a strong contaminant at these low frequencies and is visible as large-scale filaments around the remnant. The bright source in thenorthwest is Hydra A. Table 2.
Best-fit parameters from the fit of APEC and PSHOCK modelsto the spectrum of Hoinga. Errors represent the 68% confidence range.
Parameter APEC PSHOCK N H (10 cm − ) 3 . + . − . . + . − . kT (keV) 0 . + . − . . + . − . τ (10 s cm − ) ... > . a Normalization 0 . + . − . . + . − . C Statistic / d.o.f. 7625 . / . / Notes. ( a ) The ionisation timescale τ is only weakly constrained by thefitted spectrum, which is why we only give a 95% lower limit. Clouds (For et al. 2018) and parts of the Galactic plane (Hurley-Walker et al. 2019c). An important feature of this radio survey isits sensitivity to large-scale (1 ◦ − ◦ ) features, which has enabledstudies of SNRs and H ii regions across a wide range of sizes andthe full range of frequencies, independent of resolution biases(see e.g. Hindson et al. 2016; Su et al. 2018; Hurley-Walker et al.2019a).Hoinga is visible in the public GLEAM images f but is con-taminated by the presence of hundreds of radio sources, the ma-jority of which are likely unrelated radio galaxies (left panelof Fig. 6). To accurately measure the radio flux density ofHoinga, we reprocessed 13 two-minute observations spanning103-231 MHz from a drift scan centred at Dec − ◦ taken on2014-03-04, with three or four observations in each 30.72-MHzband, yielding integration times of ≈
10 minutes per band. For f http: // gleam-vo.icrar.org / gleam_postage / q / form each observation, we performed the following steps, in each caseattenuating the brightness of modelled sources using the MWAprimary beam model of Sokolowski et al. (2017): – Download the data from the All-Sky Virtual Observatory g instandard measurement set format, averaged to 40 kHz and 2 sfrequency and time resolution; – calculate a first-pass amplitude and phase calibration foreach antenna using a sky model comprised of the brightnearby source Hydra A and the GLEAM catalogue, via thesoftware calibrate , an implementation of the M itch C al al-gorithm (O ff ringa et al. 2016); – apply the derived calibration solutions; – use the peel software to remove Hydra A from the visibili-ties, with a solution interval of 4 s; – directly subtract the GLEAM sources from the visibilitiesusing subtrmodel ; – use the widefield radio imaging package WSC lean (O ff ringaet al. 2014) to image the data using natural weighting andmulti-scale multi-frequency synthesis over the full 30.72-MHz band down to a threshold of three times the local imagenoise, and then clean the data down to the local image noisein regions found to contain brightness.The ionosphere was found to be in a relatively quies-cent state, with minor ( ≈ arcsec) position shifts imparted tothe radio sources; the images were corrected using fits _ warp (Hurley-Walker & Hancock 2018). For each 30.72-MHz band,the primary-beam-corrected images were then mosaicked using swarp (Bertin et al. 2002). The resulting image is shown in theright panel of Fig. 6. Hoinga is visible as a pair of arcs of width g https: // asvo.mwatelescope.org / Article number, page 6 of 12.Becker, et al.: Hoinga - A SNR discovered in eRASS1 ≈ ◦ , 5 ◦ apart from one another. The local di ff use Galactic syn-chrotron is also visible as a fainter series of filaments with asimilar colour (i.e. spectral index).We used the software poly _ flux (Hurley-Walker et al.2019b) to measure the total flux densities of Hoinga in eachband, estimating and subtracting a mean background level. Asthe selection of the boundaries of the SNR is somewhat sub-jective, we used the tool ten times and recorded the average re-sult. The results are shown in Table 3. The uncertainties are es-timated at 20 %, dominated by the di ffi culty in selecting the truebounds of the SNR and calculating the true background level ofthe Galactic cirrus. The all-sky 408-MHz ‘Haslam’ survey was performed with theGreen Bank and Parkes Radio telescopes and remains the lowest-frequency total-power measurement of the full sky (Haslam et al.1982). The Hoinga SNR is visible in the Haslam images (Fig. 7)but the scanning pattern of Parkes is visible as a series of verti-cal lines of varying brightness throughout the image. As this isa total power measurement, the largest scale Galactic cirrus fea-tures are much brighter than Hoinga, leading to a large increasein brightness between the east and west parts of the image. Theimages are also invisibly contaminated by the same radio sourcesresolved in the GLEAM data (Section 3.1). To mitigate these is-sues, we used the following steps: – model and subtract the GLEAM extragalactic catalogue forthis region, extrapolating the source spectra to 408 MHz, ei-ther via their spectral index α as measured by GLEAM or forthe fainter sources, by an assumed value of − . – determine the average brightness profile over the lower por-tion of the image (south of Hoinga) as a function of rightascension, and subtract this profile from the full image.This resulted in the right-hand panel of Fig. 7, where theartefacts and contaminating sources have largely been removed.Similarly to the GLEAM data, we ran poly _ flux and found thatthe uncertainty on the final results was dominated by the di ffi -culty in subtracting the background, which still has large scanartefacts. We therefore conservatively estimate the error at 20 %.We also attempted to use the ‘de-striped’ ‘de-sourced’ ver-sion of the Haslam image produced by Remazeilles et al. (2015),but Hoinga was invisible in this version, possibly because it hassimilar angular scale to the scanning artefacts, and so was re-moved by the clean-up algorithms employed. The continuum map of the HI Parkes All-Sky Survey (CHIPASS;Calabretta et al. 2014) maps the radio sky at 1.4 GHz south ofDec + ◦ . We downloaded the data h , and cropped and regriddedit to match the MWA mosaics (left panel of Fig. 8). We selectedsources within 15 ◦ of Hoinga from the NRAO VLA Sky Sur-vey (NVSS; Condon et al. 1998), convolved them to match theCHIPASS resolution, and produced an output FITS image in thesame sky frame as the regridded CHIPASS data. We subtractedthe NVSS model from the CHIPASS image, producing the rightpanel of Fig. 8. We used poly _ flux to measure the flux densityof Hoinga, shown in Table 3. The errors are dominated by theselection of the region for subtraction, and after repeated mea-surements, we estimate this at about 5 %, which is 1 Jy. h https: // / people / mcalabre / CHIPASS / index.html Table 3.
Integrated flux densities of Hoinga measured from the radiodata described in Section 3. Measurements were made on images wherecontaminating sources and background had been removed using thesoftware poly _ flux . Survey Frequency / Resolution / Flux density / MHz (cid:48)
JyGLEAM 118 7 . × . ± . × . ± . × . ± . × . ± ± . ± . . ± . The S-Band Polarization All Sky Survey (SPASS ; Carretti et al.2019) is a survey of polarized radio emission over the southernsky at Dec < − ◦ using the Parkes radio telescope at 2.3 GHz.Unlike for CHIPASS (Section 3.3) there is no independent cat-alogue of extragalactic radio sources at 2.3 GHz. Meyers et al.(2017) derived a catalogue of radio sources from a version of theS-PASS images where the large-scale emission had been filteredout, with slightly worse resolution (10 (cid:48) .
75) than the publishedimages (8 (cid:48) . α = − .
75. Subtracting this model from the S-PASS data resultsin the right-hand panel of Fig. 9. Running poly _ flux repeatedlywe find more consistent results than for CHIPASS; the uncer-tainty is most likely dominated by the less clean source subtrac-tion. The residual RMS after source subtraction in a given beamis ≈
20 mJy beam − ; Hoinga subtends 256 SPASS beams; theerror is therefore estimated as 0.32 Jy.As S-PASS is a polarisation survey, we can also examine theStokes Q and U images of the region, which indicate the degreeof linear polarisation at angles of ± ◦ and ± ◦ , respectively.Figure 9 shows that the brightest parts of the shell (left and right‘limbs’) show clear linear polarisation, which is what would beexpected from a middle-aged SNR shell with a large shock com-pression ratio. These also correspond to flatter parts of the SNRshell, perhaps indicating a local increase in gas density.
4. Constraints on Ti emission from INTEGRAL Explosive nucleosynthesis in SNe is considered the main driverof Galactic, chemical evolution. Its imprints can be readily in-vestigated by observing the γ -rays emitted in the decay fromfreshly synthesized, radioactive nuclei. With a half-life of 58.9years, the abundantly produced Ti is an ideal candidate withwhich to study nucleosynthesis imprinting in young SNRs.In core collapse supernovae (ccSN) Ti is mainly producedduring the α -rich freeze-out (Woosley et al. 1973) deep in thecentral region, where the nucleosynthesis yields are strongly de-pendent on the thermodynamic conditions (Magkotsios et al.2010; Hermansen et al. 2020). While models of ccSN fail to ro-bustly produce explosions in a wide stellar mass range so far,it appears safe to assume that asymmetries are required to drivesuccessful explosions. Depending on the applied, simplified ex- Article number, page 7 of 12 & A proofs: manuscript no. main
Fig. 7.
100 deg of the region surrounding the Hoinga SNR as seen at 408 MHz by the survey by Haslam et al. (1982), after conversion from Kto Jy beam − . The left panel shows the original image, and the right panel shows the image after source-subtraction and backgrounding, discussedin Section 3.2. Hoinga is visible as an ellipse in the centre of the image, while Galactic cirrus and scan line artefacts from the Parkes observingstrategy dominate the surroundings. The bright source in the northwest is Hydra A, and subtraction of this source has not been performed. Fig. 8.
100 deg of the region surrounding Hoinga as seen at 1.4 GHz by CHIPASS, after conversion from K to Jy beam − . The left panel shows theoriginal image, and the right panel shows the image after source subtraction, discussed in Section 3.3. Hoinga is clearly visible as a crescent-moonin the centre of the image, while Galactic cirrus and residuals around poorly subtracted di ff use sources are visible in the surroundings. The brightsource in the northwest is Hydra A. Faint scan lines are visible from the Parkes observing strategy. plosion scheme, the predicted Ti ejecta yield can vary in therange 10 − − − M (cid:12) , depending also on the initial mass of theexploding star (Timmes et al. 1996; Wanajo et al. 2018; Limongi& Chie ffi Ti ejecta masses. Multiple scenarios lead-ing to the disruption of a white dwarf star are considered viable,as the progenitors of these explosions have not yet been unam-biguously identified. For the standard model, involving a cen-trally ignited Chandrasekhar-mass white dwarf star, Ti ejecta masses range between 10 − and 10 − M (cid:12) (Maeda et al. 2010;Seitenzahl et al. 2013; Fink et al. 2014). However, in the double-detonation scenario, ejecta masses of 10 − − − M (cid:12) are possi-ble (Fink et al. 2010; Woosley & Kasen 2011; Moll & Woosley2013), where some exotic models even predict Ti masses of upto 0.1 M (cid:12) (Perets et al. 2010; Waldman et al. 2011).Evidence for the production of Ti can be obtained by mea-suring the decay radiation in the decay chain of Ti → Sc → Ca. The dominant decay lines are emitted at 68 and 78 keVduring the Ti decay with a half life of 58.9 years (Ahmad et al.
Article number, page 8 of 12.Becker, et al.: Hoinga - A SNR discovered in eRASS1
Fig. 9.
100 deg of the region surrounding Hoinga as seen at 2.3 GHz by SPASS. The top-left panel shows the Stokes I image, and the top-rightpanel shows the image after source subtraction, discussed in Section 3.4. Hoinga is clearly visible as a filled ellipse in the centre of the image,while Galactic cirrus and residuals around poorly subtracted di ff use sources are visible in the surroundings. The bottom left and bottom rightpanels show the Stokes Q and U images, respectively. Sc decay with a halflife of 4 hours (Audi et al. 2003). Photons are emitted with aprobability (branching ratio) of 93.0 %, 96.4 % , and 99.9 % perdecay, respectively (Chen et al. 2011)Here, the spectrometer SPI (Vedrenne et al. 2003) onINTEGRAL (Winkler et al. 2003) is used to search for thedecay radiation in both subsequent decay steps in the HoingaSNR. We use the spimodfit analysis tool (Strong et al. 2005;Halloin 2009) to extract the spectrum in the relevant energyranges 50–100 keV and 1100–1200 keV from the raw SPIdata. The spectrum is extracted assuming an extended sourceof emission modelled by a circular region of 2 ◦ . LS ( E ; E , F , σ ) = F √ πσ · exp (cid:32) ( E − E ) σ (cid:33) + A · (cid:32) EE C (cid:33) α , (1)where F is the measured line flux, E is the energy of theDoppler-shifted line centroid, and σ is the line width. As we ex-pect a low signal-to-noise ratio for the decay lines, we searchfor a combined signal in all lines simultaneously, that is we as-sume that the branching ratio corrected fluxes, Doppler shifts,and broadening are identical in all lines. Due to the presence ofa complex of strong background lines between 50 and 65 keVinduced by germanium, we excluded the 68 keV line in the anal-ysis.We find no significant flux excess in the vicinity of the 78or 1157 keV line or in the combined line analysis. As the broad-ening of the 78 or 1157 keV lines is related to the expansionvelocity of the Ti-containing ejecta and determines the size of
Article number, page 9 of 12 & A proofs: manuscript no. main the selected background region, we deduce a 3 σ upper flux limitof 9 . × − ph cm − s − by assuming an expansion velocity of4000 km s − (Nagataki et al. 1998; Diehl et al. 2015). This ex-pansion velocity translates into a line broadening of ≈ ≈
20 keV FWHM at 1157 keV, respec-tively.
5. Summary and Discussion
Using data from the first SRG / eROSITA observatory all-sky sur-vey we discovered one of the largest SNRs in the sky. Despite95 % of SNR discoveries being made at radio wavelengths, andits clear existence in multiple radio surveys, we conclude thatHoinga was missed by previous searches for several reasons.Firstly, its location at high Galactic latitudes; most radio searcheshave focused on low latitudes, where the density of SNRs is ex-pected to be highest. Another reason for not noticing it in pre-vious X-ray and radio surveys is its total flux density. Althoughit is large, its surface brightness is relatively low. As it has verylittle fine-scale structure, it also does not appear at all in most in-terferometric maps. In single-dish radio images, it is visibly con-taminated by about 100 extragalactic radio sources, with manymore below the sensitivity and confusion limits, meaning thatits di ff use radio emission remained uncovered. Hoinga is nearlythe largest SNR ever detected at radio wavelengths, subtending ≈ (cid:48) × (cid:48) , and comparable in size to the largest detectedobject, G 65.3 + ff use Galactic syn-chrotron makes it less obvious than smaller and brighter sources.The clear shell structure, particularly evident in Figs. 1 and 8,indicates it is likely to be a classic shell-type SNR that is not cen-trally powered, and its highly circular nature indicates that it isexpanding into a region of relatively uniform density. Figure 10shows the radio flux densities plotted as a function of frequency,with a fitted spectral index of α = − . ± .
08, for S ν ∝ ν α .This radio spectral energy distribution indicates that non-thermalsynchrotron emission dominates the radio spectrum, again con-sistent with a shell-type SNR.A distance to the SNR would enable transformation ofour measurements into physical properties. Dubner & Giacani(2015) discuss the challenge of estimating the distance of radio-detected SNRs; a method that does not rely on additional ob-servations is to search for nearby neutron stars that appear aspulsars and may have formed at the same time as the SNR, andusing their dispersion measure in combination with electron den-sity models of the Galaxy to determine their distance.We used the Australia Telescope National Facility pulsar cat-alogue v1.59 (Manchester et al. 2005) i to search for known ra-dio pulsars within 20 ◦ of Hoinga’s geometrical centre, but foundnone with attributes that would indicate a clear association. Fromthe group of pulsars located in the region of interest we excludedpossible matches on the basis of: – period P <
10 ms, indicating a recycled origin; – characteristic ages ( P P ) >
45 Myr, which would be extremelyinconsistent with a SNR age of < . – measured proper motion inconsistent with having a commoncentre of origin; – measured dispersion measure inconsistent with a nearby lo-cation. i atnf.csiro.au / research / pulsar / psrcat / Fig. 10.
Radio SED of the total flux density of Hoinga as measured bythe surveys discussed in Section 3. Black points show the data fromTable 3; the blue line shows a least-squares weighted fit to the data,yielding S = . ± .
03 Jy and α = − . ± .
08 for S ν ∝ ν α . This avenue is therefore unpromising, but because the cover-age of pulsar surveys is denser at low Galactic latitude, a pulsarcould have been missed by existing observations, and follow-up observations within the SNR shell may yet reveal a counter-part. Assuming a distance of ∼
500 pc, a remnant NS with atransversal speed of the order of 1000 km s − would have bynow reached the SN shell if the explosion happened ∼ × < L . < W Hz − (e.g. Case & Bhattacharya 1998). As-suming that Hoinga is more luminous than 5 × W Hz − , wecan obtain a limit on its distance from Earth by (cid:113) L . π S . , i.e. D >
450 pc. Additionally, radio SNRs do not typically have di-ameters greater than 100 pc (Badenes et al. 2010). If we assumethat Hoinga has a diameter <
100 pc, by geometry its distancefrom Earth must be D < . L . < . × W Hz − , which puts Hoingaon the lower end of the SNR luminosity distribution. We notethat other high-latitude SNRs have also been found to have un-usually low brightness compared to those at low latitudes; seee.g. G181.1 + . − . ◦ . Article number, page 10 of 12.Becker, et al.: Hoinga - A SNR discovered in eRASS1 × cm − (Dickey & Lockman 1990). The values found fromour X-ray spectral fits are of the order of N H = . + . − . × cm − which gives another indication for Hoinga being a nearby SNR.If we assume that the column density derived in Section 2.4is representative along the entire line of sight, we can derive arange of local ISM densities by dividing by the distance limits.For a column density of N H = × cm − , and distances of0.45–1.2 kpc, the resulting local density n H = . .
16 cm − .Inputting these into the SNR evolutionary model calculator pro-vided by Leahy & Williams (2017), with otherwise standardmodel and input values, we calculate the range of possible agesas 21–150 kyr. However, the morphology of the SNR suggests amuch lower age, and therefore we suggest the SNR is likely tobe at the closer, younger, and higher n H ends of the allowableranges.Taking into account the fact that no pulsar has been asso-ciated with the object so far, it is highly possible that Hoingais the remnant of a type Ia SN. This would also be consistentwith the high latitude of the SNR, as the massive star progeni-tors of core-collapse SNe are expected to be more concentratedin the Galactic plane (Taylor et al. 1993; Cordes & Lazio 2002;Faucher-Giguere & Kaspi 2006).eROSITA will perform a total of eight all-sky surveys. Withfurther surveys completed, more data from the Hoinga remnantwill become available in the next few years. This will allow usto study the remnants fine structure and spectral properties inmore detail, hopefully allowing us to further constrain its dis-tance, age, chemical composition, and SN type. The findings ofHoinga represent a highlight of the beginning of a wider programsetup by the authors WB and NHW as part of an eROSITA-Australian-based joint-venture collaboration defined to explorethe X-ray-radio-sky in order to uncover further exciting surprisesin the SNR sphere. Acknowledgements.
We thank Bernd Aschenbach and Nicholas Pingel for fruit-ful discussions and the anonymous referee for valuable comments.eROSITA is the primary instrument aboard SRG, a joint Russian-German sci-ence mission supported by the Russian Space Agency (Roskosmos), in the in-terests of the Russian Academy of Sciences represented by its Space ResearchInstitute (IKI), and the Deutsches Zentrum für Luft- und Raumfahrt (DLR). TheSRG spacecraft was built by Lavochkin Association (NPOL) and its subcon-tractors, and is operated by NPOL with support from IKI and the Max PlanckInstitute for Extraterrestrial Physics (MPE). The development and constructionof the eROSITA X-ray instrument was led by MPE, with contributions from theDr. Karl Remeis Observatory Bamberg & ECAP (FAU Erlangen-Nürnberg), theUniversity of Hamburg Observatory, the Leibniz Institute for Astrophysics Pots-dam (AIP), and the Institute for Astronomy and Astrophysics of the University ofTübingen, with the support of DLR and the Max Planck Society. The ArgelanderInstitute for Astronomy of the University of Bonn and the Ludwig MaximiliansUniversität Munich also participated in the science preparation for eROSITA.The eROSITA data shown here were processed using the eSASS / NRTA soft-ware system developed by the German eROSITA consortium.NHW is supported by an Australian Research Council Future Fellowship (projectnumber FT190100231) funded by the Australian Government. This scientificwork makes use of the Murchison Radio-astronomy Observatory, operated byCSIRO. We acknowledge the Wajarri Yamatji people as the traditional owners ofthe Observatory site. Support for the operation of the MWA is provided by theAustralian Government (NCRIS), under a contract to Curtin University adminis-tered by Astronomy Australia Limited. Establishment of the Murchison Radio-astronomy Observatory and the Pawsey Supercomputing Centre are initiativesof the Australian Government, with support from the Government of WesternAustralia and the Science and Industry Endowment Fund. We acknowledge thePawsey Supercomputing Centre which is supported by the Western Australianand Australian Governments. Access to Pawsey Data Storage Services is gov-erned by a Data Storage and Management Policy (DSMP). ASVO has receivedfunding from the Australian Commonwealth Government through the NationaleResearch Collaboration Tools and Resources (NeCTAR) Project, the AustralianNational Data Service (ANDS), and the National Collaborative Research Infras-tructure Strategy. This research has made use of NASA’s Astrophysics Data Sys-tem Bibliographic Services.MGFM acknowledges support by the International Max-Planck Research School on Astrophysics at the Ludwig-Maximilians University, IMPRS.This work has made use of data from the European Space Agency (ESA)mission
Gaia ( ), processed by the Gaia
Data Processing and Analysis Consortium (DPAC, ). Funding for the DPAC has been pro-vided by national institutions, in particular the institutions participating in the
Gaia
Multilateral Agreement.The following 3rd-party software was used in this work: aoflagger and cot - ter (O ff ringa et al. 2015); WSC lean (O ff ringa et al. 2014; O ff ringa & Smirnov2017); A egean (Hancock et al. 2018); miriad (Sault et al. 1995); T op C at (Taylor 2005) N um P y v1.11.3 (Dubois et al. 1996; Harris et al. 2020); A s - tro P y v2.0.6 (Astropy Collaboration et al. 2013); S ci P y v0.17.0 (Oliphant 2007),M atplotlib v1.5.3 (Hunter 2007). The manuscript was prepared on the web-based L A TEX editor, Overleaf.
References
Adams, S. M., Kochanek, C. S., Gerke, J. R., Stanek, K. Z., & Dai, X. 2017,MNRAS, 468, 4968Ahmad, I., Greene, J. P., Moore, E. F., et al. 2006, Physical Review C, 74Asaoka, I. & Aschenbach, B. 1994, A&A, 284, 573Asaoka, I., Egger, R., & Aschenbach, B. 1996, in Röntgenstrahlung from theUniverse, ed. H. U. Zimmermann, J. Trümper, & H. Yorke, 233–234Aschenbach, B. 1998, Nature, 396, 141Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013, A&A, 558,A33Audi, G., Bersillon, O., Blachot, J., & Wapstra, A. H. 2003, Nuclear Physics A,729, 3Badenes, C., Maoz, D., & Draine, B. T. 2010, MNRAS, 407, 1301Becker, W. 2009, in Astrophysics and Space Science Library, ed. W. Becker, Vol.357, 91Bertin, E., Mellier, Y., Radovich, M., et al. 2002, in Astronomical Society of thePacific Conference Series, Vol. 281, Astronomical Data Analysis Softwareand Systems XI, ed. D. A. Bohlender, D. Durand, & T. H. Handley, 228Borkowski, K. J., Lyerly, W. J., & Reynolds, S. P. 2001, ApJ, 548, 820Busser, J.-U., Egger, R., & Aschenbach, B. 1996, in Röntgenstrahlung from theUniverse, ed. H. U. Zimmermann, J. Trümper, & H. Yorke, 239–240Calabretta, M. R., Staveley-Smith, L., & Barnes, D. G. 2014, PASA, 31, e007Carretti, E., Haverkorn, M., Staveley-Smith, L., et al. 2019, MNRAS, 489, 2330Case, G. L. & Bhattacharya, D. 1998, ApJ, 504, 761Chatterjee, S., Vlemmings, W. H. T., Brisken, W. F., et al. 2005, ApJ, 630, L61Chen, J., Singh, B., & Cameron, J. A. 2011, Nuclear Data Sheets, 112, 2357Condon, J. J., Cotton, W. D., Greisen, E. W., et al. 1998, AJ, 115, 1693Cordes, J. M. & Lazio, T. J. W. 2002 [ astro-ph/0207156v3 ]Cordes, J. M., Romani, R. W., & Lundgren, S. C. 1993, Nature, 362, 133de Gasperin, F., Intema, H. T., & Frail, D. A. 2018, MNRAS, 474, 5008Dickey, J. M. & Lockman, F. J. 1990, ARA&A, 28, 215Diehl, R., Siegert, T., Greiner, J., et al. 2018, Astronomy & Astrophysics, 611,A12Diehl, R., Siegert, T., Hillebrandt, W., et al. 2015, A&A, 574, A72Dorman, B., Arnaud, K. A., & Gordon, C. A. 2003, in AAS / High Energy Astro-physics Division, Vol. 7, AAS / High Energy Astrophysics Division | spimodfit | Explanatory Guide and Users Manual, version 2.9edn., Max Planck Institut für extraterrestrische Physik, Giessenbachstraße 1,85748 Garching, GermanyHancock, P. J., Trott, C. M., & Hurley-Walker, N. 2018, PASA, 35, e011
Article number, page 11 of 12 & A proofs: manuscript no. main
Harris, C. R., Millman, K. J., van der Walt, S. J., et al. 2020, Nature, 585, 357Hartmann, D. H., Predehl, P., Greiner, J., et al. 1997, Nuclear Physics A, 621, 83Haslam, C. G. T., Salter, C. J., Sto ff el, H., & Wilson, W. E. 1982, A&AS, 47, 1Hermansen, K., Couch, S. M., Roberts, L. F., Schatz, H., & Warren, M. L. 2020,The Astrophysical Journal, 901, 77Hindson, L., Johnston-Hollitt, M., Hurley-Walker, N., et al. 2016, PASA, 33,e020Hunter, J. D. 2007, Comput. Sci. Eng., 9, 90Hurley-Walker, N., Callingham, J. R., Hancock, P. J., et al. 2017, MNRAS, 464,1146Hurley-Walker, N., Filipovi´c, M. D., Gaensler, B. M., et al. 2019a, PASA, 36,e045Hurley-Walker, N., Gaensler, B. M., Leahy, D. A., et al. 2019b, PASA, 36, e048Hurley-Walker, N. & Hancock, P. J. 2018, Astronomy and Computing, 25, 94Hurley-Walker, N., Hancock, P. J., Franzen, T. M. O., et al. 2019c, PASA, 36,e047Joye, W. A. & Mandel, E. 2003, in Astronomical Society of the Pacific Con-ference Series, Vol. 295, Astronomical Data Analysis Software and SystemsXII, ed. H. E. Payne, R. I. Jedrzejewski, & R. N. Hook, 489Keane, E. F. & Kramer, M. 2008, MNRAS, 391, 2009Kochanek, C. S., Beacom, J. F., Kistler, M. D., et al. 2008, ApJ, 684, 1336Kothes, R., Reich, P., Foster, T. J., & Reich, W. 2017, A&A, 597, A116Leahy, D. A. & Williams, J. E. 2017, AJ, 153, 239Limongi, M. & Chie ffi , A. 2018, ApJS, 237, 13Maeda, K., Röpke, F. K., Fink, M., et al. 2010, ApJ, 712, 624Magkotsios, G., Timmes, F. X., Hungerford, A. L., et al. 2010, ApJS, 191, 66Manchester, R. N., Hobbs, G. B., Teoh, A., & Hobbs, M. 2005, AJ, 129, 1993Meidinger, N., Andritschke, R., Bornemann, W., et al. 2014, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 9144,Space Telescopes and Instrumentation 2014: Ultraviolet to Gamma Ray,91441WMeyers, B. W., Hurley-Walker, N., Hancock, P. J., et al. 2017, PASA, 34, e013Moll, R. & Woosley, S. E. 2013, ApJ, 774, 137Nagataki, S., Shimizu, T. M., & Sato, K. 1998, ApJ, 495, 413O ff ringa, A. R., McKinley, B., Hurley-Walker, et al. 2014, MNRAS, 444, 606O ff ringa, A. R., McKinley, B., Hurley-Walker, N., et al. 2014, MNRAS, 444,606O ff ringa, A. R. & Smirnov, O. 2017, MNRAS, 471, 301O ff ringa, A. R., Trott, C. M., Hurley-Walker, N., et al. 2016, MNRAS, 458, 1057O ff ringa, A. R., Wayth, R. B., Hurley-Walker, N., et al. 2015, PASA, 32, e008Oliphant, T. E. 2007, Comput. Sci. Eng., 9, 10Pavlinsky, M., Levin, V., Akimov, V., et al. 2018, in Society of Photo-Optical In-strumentation Engineers (SPIE) Conference Series, Vol. 10699, Space Tele-scopes and Instrumentation 2018: Ultraviolet to Gamma Ray, 106991YPerets, H. B., Gal-Yam, A., Mazzali, P. A., et al. 2010, Nature, 465, 322Pfe ff ermann, E. & Aschenbach, B. 1996, in Röntgenstrahlung from the Universe,ed. H. U. Zimmermann, J. Trümper, & H. Yorke, 267–268Pfe ff ermann, E., Briel, U. G., & Freyberg, M. J. 2003, Nuclear Instruments andMethods in Physics Research A, 515, 65Predehl, P., Andritschke, R., Arefiev, V., et al. 2020a, A&A, submittedPredehl, P., Sunyaev, R., Becker, W., et al. 2020b, Nature, submittedRemazeilles, M., Dickinson, C., Banday, A. J., Bigot-Sazy, M. A., & Ghosh, T.2015, MNRAS, 451, 4311Reynolds, S. P., Borkowski, K. J., Green, D. A., et al. 2008, ApJ, 680, L41Sault, R. J., Teuben, P. J., & Wright, M. C. H. 1995, in ASP Conference Se-ries, Vol. 77, Astronomical Data Analysis Software and Systems IV, ed. R. A.Shaw, H. E. Payne, & J. J. E. Hayes, 433Seitenzahl, I. R., Ciaraldi-Schoolmann, F., Röpke, F. K., et al. 2013, MNRAS,429, 1156Siegert, T., Diehl, R., Weinberger, C., et al. 2019, Astronomy & Astrophysics,626, A73Sokolowski, M., Colegate, T., Sutinjo, A. T., et al. 2017, PASA, 34, e062Stephenson, F. R. 2017, Historical Records of Supernovae, ed. A. W. Alsabti &P. Murdin, 49Strong, A. W., Diehl, R., Halloin, H., et al. 2005, Astronomy & Astrophysics,444, 495Strüder, L., Briel, U., Dennerl, K., et al. 2001, A&A, 365, L18Su, H., Macquart, J. P., Hurley-Walker, N., et al. 2018, MNRAS, 479, 4041Taylor, J. H., Manchester, R. N., & Lyne, A. G. 1993, The Astrophysical JournalSupplement Series, 88, 529Taylor, M. B. 2005, in ASP Conference Series, Vol. 347, Astronomical DataAnalysis Software and Systems XIV, ed. P. Shopbell, M. Britton, & R. Ebert,29Timmes, F. X., Woosley, S. E., Hartmann, D. H., & Ho ff man, R. D. 1996, ApJ,464, 332Tingay, S. J., Goeke, R., Bowman, J. D., et al. 2013, PASA, 30, 7Vedrenne, G., Roques, J.-P., Schönfelder, V., et al. 2003, A&A, 411, L63Voges, W., Aschenbach, B., Boller, T., et al. 1999, A&A, 349, 389Waldman, R., Sauer, D., Livne, E., et al. 2011, The Astrophysical Journal, 738,21Wanajo, S., Müller, B., Janka, H.-T., & Heger, A. 2018, ApJ, 852, 40Wayth, R. B., Lenc, E., Bell, M. E., et al. 2015, PASA, 32, e025Wayth, R. B., Tingay, S. J., Trott, C. M., et al. 2018, PASA, 35[ arXiv:1809.06466 ]Weinberger, C., Diehl, R., Pleintinger, M. M. M., Siegert, T., & Greiner, J. 2020,Astronomy & Astrophysics, 638, A83Wilms, J., Allen, A., & McCray, R. 2000, ApJ, 542, 914Winkler, C., Courvoisier, T. J.-L., Di Cocco, G., et al. 2003, A&A, 411, L1Woosley, S. E., Arnett, W. D., & Clayton, D. D. 1973, ApJS, 26, 231Woosley, S. E. & Kasen, D. 2011, ApJ, 734, 38Zimmermann, H. U., Becker, W., Belloni, T., et al. 1994, EXSAS User’s Guide]Weinberger, C., Diehl, R., Pleintinger, M. M. M., Siegert, T., & Greiner, J. 2020,Astronomy & Astrophysics, 638, A83Wilms, J., Allen, A., & McCray, R. 2000, ApJ, 542, 914Winkler, C., Courvoisier, T. J.-L., Di Cocco, G., et al. 2003, A&A, 411, L1Woosley, S. E., Arnett, W. D., & Clayton, D. D. 1973, ApJS, 26, 231Woosley, S. E. & Kasen, D. 2011, ApJ, 734, 38Zimmermann, H. U., Becker, W., Belloni, T., et al. 1994, EXSAS User’s Guide