Measuring the distance to the black hole candidate X-ray binary MAXI J1348-630 using HI absorption
J. Chauhan, J. C. A. Miller-Jones, W. Raja, J. R. Allison, P. F. L. Jacob, G. E. Anderson, F. Carotenuto, S. Corbel, R. Fender, A. Hotan, M. Whiting, P. A. Woudt, B. Koribalski, E. Mahony
MMNRAS , 1–5 (2020) Preprint 7 December 2020 Compiled using MNRAS L A TEX style file v3.0
Measuring the distance to the black hole candidate X-ray binaryMAXI J1348–630 using H i absorption J. Chauhan (cid:63) , J. C. A. Miller-Jones † , W. Raja , J. R. Allison , P. F. L. Jacob ,G. E. Anderson , F. Carotenuto , S. Corbel , , R. Fender , A. Hotan , M. Whiting ,P. A. Woudt , B. Koribalski , E. Mahony International Centre for Radio Astronomy Research – Curtin University, GPO Box U1987, Perth, WA 6845, Australia CSIRO Astronomy and Space Science, Australia Telescope National Facility, PO Box 76, Epping NSW 1710, Australia Sub-Dept. of Astrophysics, Department of Physics, University of Oxford, Denys Wilkinson Building, Keble Rd., Oxford, OX1 3RH, UK AIM, CEA, CNRS, Universite´ı Paris Diderot, Sorbonne Paris Cite´ı, Universite´ı Paris-Saclay, F-91191 Gif-sur-Yvette, France Station de Radioastronomie de Nan¸cay, Observatoire de Paris, PSL Research University, CNRS, Univ. Orl´eans, 18330 Nan¸cay, France Department of Astronomy, University of Cape Town, Private Bag X3, Rondebosch 7701, South Africa
Accepted XXX. Received YYY; in original form ZZZ
ABSTRACT
We present H i absorption spectra of the black hole candidate X-ray binary (XRB)MAXI J1348–630 using the Australian Square Kilometre Array Pathfinder (ASKAP)and MeerKAT. The ASKAP H i spectrum shows a maximum negative radial velocity(with respect to the local standard of rest) of − ± − for MAXI J1348–630,as compared to − ± − for a stacked spectrum of several nearby extragalacticsources. This implies a most probable distance of 2 . +0 . − . kpc for MAXI J1348–630,and a strong upper limit of the tangent point distance at 5 . ± . ±
10 % of the Eddington luminosityat the peak of its outburst, and that the source transited from the soft to the hardX-ray spectral state at 2 . ± . i spectrum of MAXI J1348–630 (obtained from the older, low-resolution 4k mode) isconsistent with the re-binned ASKAP spectrum, highlighting the potential of theeventual capabilities of MeerKAT for XRB spectral line studies. Key words: black hole physics – ISM: jets and outflows – radio continuum: transients– X-rays: binaries – X-rays: individual: MAXI J1348–630
The distance to an astrophysical object is a key physicalparameter, as it provides us with a way to determine phys-ical quantities from observables. For X-ray binaries (XRBs)in our Galaxy, the distance can be directly determined bymeasuring the parallax to the source, with either very longbaseline interferometry (VLBI; e.g. Miller-Jones et al. 2009;Reid et al. 2011; Atri et al. 2020) or
Gaia (Gaia Collabo-ration et al. 2018). XRB distances can also be determinedby studying the line features in infrared/optical spectra ofthe companion star (Jonker & Nelemans 2004), by usingthe proper motions of the jet ejecta (Mirabel & Rodr´ıguez1994) or by studying the interstellar extinction (Zdziarskiet al. 1998). However, these distances are often poorly con-strained, and suffer large uncertainties that in some casescan exceed 50% (Jonker & Nelemans 2004). If the donor staris too faint, or a parallax measurement is not possible, analternative technique is to use the 21-cm line of neutral hy- (cid:63)
E-mail: [email protected] (JC) † [email protected] (JCAM-J) drogen (H i , e.g. Goss & Mebold 1977; Dickey 1983; Dhawanet al. 2000; Chauhan et al. 2019a). The Galactic H i cloudsalong the line of sight to an XRB are rotating about the cen-tre of the Milky Way, and their Doppler-shifted absorptionfeatures allow us to determine the kinematic distance to thesource (Gathier et al. 1986; Kuchar & Bania 1990). Whilethe kinematic distance estimates obtained from this methodare fairly accurate (at least for circular rotation) when thesource is located beyond the Solar orbit, they suffer ambi-guities if the source is located within the Solar circle. Sincethe rotation curves are then dual-valued, this leads to twodistance estimates (near and far) for a single velocity mea-surement (e.g. Reid et al. 2014; Wenger et al. 2018).On 2019 January 26, the Monitor of All-sky X-ray Im-age ( MAXI ; Matsuoka et al. 2009) discovered an uncata-logued XRB at a position RA (J2000) = 13 h m . s ± . − d (cid:48) . (cid:48)(cid:48) ± .
04 (Galactic coordinates l = 309 . ◦ b = − . ◦ http://maxi.riken.jp/top/index.html © a r X i v : . [ a s t r o - ph . H E ] D ec J. Chauhan et al.
Table 1.
Co-ordinates and 1.42-GHz ASKAP flux densities ofMAXI J1348–630 and our extragalactic comparison sources. Posi-tions taken from [1] Russell et al. (2019); [2] Murphy et al. (2007).
Source RA (J2000) Dec (J2000) Flux Density a Name (hh:mm:ss) (dd:mm:ss) (mJy)MAXI J1348–630 [1] 13:48:12.79 –63:16:28.48 155 ± ± ± ± ± ± ± ± ± a σ errors are quoted, calculated by adding in quadrature theerror on the Gaussian fit and the rms noise in the image. counterpart of the source was detected by Denisenko et al.(2019), and the outburst was subsequently observed acrossthe electromagnetic spectrum (e.g. Carotenuto et al. 2019;Chauhan et al. 2019b; Kennea & Negoro 2019). The sys-tem is believed to harbour a stellar-mass black hole (Russellet al. 2019; Zhang et al. 2020), although the key systemparameters are poorly constrained.In this investigation, we present H i absorption spec-tra from both ASKAP and MeerKAT observations, and usethe Doppler-shifted 21-cm absorption line to constrain thedistance to MAXI J1348–630. We further use our distanceconstraints to estimate the peak X-ray luminosity and thespectral state transition luminosity. ASKAP (Hotan et al. 2014) observed MAXI J1348–630 on2019 February 13 for 9.91 hours (on source exposure 8.39hours) from 13:22–23:16 UTC, using the full array of 36dishes, with 36 overlapping beams. The large field of view( ≈
30 deg ) and high angular resolution ( ∼ (cid:48)(cid:48) ) of ASKAPallow us to simultaneously detect the H i absorption towardsboth MAXI J1348–630 and a set of nearby extragalacticsources (Table 1), enabling a discrimination between nearand far kinematic distances. Our observation was performedat a central frequency of 1.34 GHz, with a total bandwidthof 288 MHz divided into 15368 fine channels, each of whichhas a frequency resolution of 18.519 kHz (velocity resolution3.9 km s − ).For reducing our multiple-beam full array data onMAXI J1348–630, we used the ASKAP data analysis soft-ware, ASKAPsoft . Although our MAXI J1348–630 data havemore antennas and beams, we follow a similar calibrationprocedure to that described in Chauhan et al. (2019a). How-ever, for generating the spectral cube from the calibrated,continuum-subtracted measurement set created by ASKAP-soft , we used the
TCLEAN task in the Common AstronomySoftware Application (
CASA v5.1.2-4: McMullin et al. 2007)to ensure we could use the same procedure to extract bothMeerKAT and ASKAP spectra, and to allow quick optimisa-tion of our imaging parameters (e.g. uv -range, deconvolutiondepth, deconvolver), given the demand on supercomputingtime to run ASKAPsoft . We produced a spectral sub-cube of378 channels centered at the rest frequency (1420.4 MHz) of the H i line using a Briggs weighting parameter (robustness)of 0.5, and adopting a minimum baseline length of 700 m.From the ASKAP spectral cube, we extracted the H i absorption spectrum for MAXI J1348–630 and the eight ex-tragalactic sources listed in Table 1, by measuring the bright-ness in each frequency channel at the position correspondingto the peak flux density (determined from the continuum im-age). We used the IMFIT task in
CASA to measure source fluxdensities and 1 σ uncertainties from the continuum images. MAXI J1348–630 was monitored as part of the MeerKATLarge Survey Project for slow transients (ThunderKAT;Fender et al. 2016). Here we use data from 2019 February09 between 05:08 and 05:23 (UTC), when the source wasbrightest in the radio (486 ± i absorp-tion. MeerKAT provides a field of view of 0.86 deg , and aspatial resolution of ∼ (cid:48)(cid:48) at 1.42 GHz. Our observationswere carried out using 60 MeerKAT antennas, at a centralfrequency of 1284 MHz, with 860 MHz of bandwidth. Onlythe 4k correlator mode was available, which gave a spectralresolution of 209 kHz, equal to 44 km s − at 1420.4 MHz.Spectral line data reduction was carried out using astandard procedure that implemented tasks from MIRIAD (Sault et al. 1995). The data were first converted to FITS for-mat using
CASA , selecting only channels in the range 10 MHzeither side of 1420.4 MHz (equivalent to radial velocities of ± − ). Calibration of the bandpass and flux scalewas carried out using PKS B1934–638 (Reynolds 1994),and the time varying antenna gains using PKS B1421–490. To avoid corrupting the calibration solutions withGalactic H i emission and absorption, the central 20 chan-nels ( ±
400 km s − ) were flagged and then interpolated fromneighbouring channels. Further self-calibration of the MAXIJ1348–630 data was carried out to correct the time-varyinggain phase. After subtraction of the continuum flux density,a spectral cube was formed within 5 arcmin of MAXI J1348–630 using a robustness of 0.5, and a minimum baseline lengthof 700 m. The final spectrum was extracted from the cube,adopting a similar procedure to that described for ASKAPin Section 2.1. In the left panel of Fig. 1 we present an ASKAP contin-uum image of the MAXI J1348–630 field created by mosaic-ing beams 14, 15, 20 and 21. Residual uncertainties remainaround the bright ( > ± MNRAS , 1–5 (2020) he distance to MAXI J1348–630 centre of the beam) in the right panel of Fig. 1. This showsa short-duration flare, peaking at 252 ±
13 mJy at 15:51:32(UTC), and then gradually decreasing to 111 ± i absorption spectra In Fig. 2, we show the spectrum for MAXI J1348–630, to-gether with a stacked spectrum for all eight extragalacticsources, and the 3 σ noise levels measured from nearby re-gions. We detect significant ( > σ ) H i absorption complexesout to maximum negative velocities (with respect to thelocal standard of rest, or LSR) of − ± − and − ± − for MAXI J1348–630 and the extragalacticsources, respectively.To determine the distance to MAXI J1348–630, we com-puted the Galactic rotation curve for the Galactic longitudeof MAXI J1348–630, and determined the variation of the ra-dial velocity ( V LSR ) of the LSR with distance from the Sun( d ). For simplicity we assumed that the Galactic rotationcurve is flat, that the Milky Way is rotating with a circularvelocity ( V ) of 240 ± − (Reid et al. 2014), and thatthe Sun is located at a distance R = 8 . ± .
16 kpc from theGalactic centre (Reid et al. 2014). Fig. 3 shows that the pre-dicted LSR radial velocities are negative within a few kpc ofthe Sun, and by comparison with our observed H i spectrumallows us to determine the near distance, the far distance andthe tangent point distance ( R T = R cos l ≡ . ± . for infer-ring kinematic distances and associated uncertainties, usingthe rotation curve provided by Reid et al. (2014). Usingthis approach, we determined the near and far distancesof MAXI J1348–630 to be 2 . +0 . − . kpc and 8 . +0 . − . kpc, re-spectively. The maximum negative absorption velocity (withrespect to the LSR) observed for the extragalactic sources( − ± − ) is in good agreement with the tangent pointvelocity ( − ± − ) for this line of sight (within errorbars, and calculated using Reid et al. 2014 and Wenger et al.2018). H i absorption at the tangent point velocity is not ob-served towards MAXI J1348–630, implying that it must becloser than the tangent point distance of 5 . ± . . +0 . − . kpc. MeerKAT detected MAXI J1348–630 as a bright pointsource, of flux density 486 ± − , sowe rebinned our ASKAP data (with velocity resolution3.9 km s − ) to match the MeerKAT resolution. The uncer-tainties on the MeerKAT spectra are smaller, both becausethe instrument has a lower system temperature, and becauseMAXI J1348–630 was brighter at the time of the MeerKATobservations than it was during the ASKAP observations (155 ± Tominaga et al. (2020) studied the complete outburst ofMAXI J1348–630 using
MAXI /GSC data in the 2–20 keVenergy range. They found the peak of the outburst to haveoccurred on 2019 February 09, and the soft-to-hard spectralstate transition to have occurred on 2019 April 27 (Tominagaet al. 2020). We used the High Energy Astrophysics ScienceArchive Research Center (HEASARC) tool WebPIMMS tocalculate the bolometric X-ray flux in the energy range 0.01–100 keV from the X-ray flux (2–20 keV) and spectral modeldetermined by Tominaga et al. (2020). We derived a peakunabsorbed X-ray flux of 2 . ± . × − erg cm − s − , andan unabsorbed flux of 4 . ± . × − erg cm − s − for thesoft-to-hard X-ray spectral state transition, correspondingto luminosities of 1 . ± . × and 2 . ± . × erg s − ,respectively, at our preferred distance.If we consider the compact object to be a black hole(Russell et al. 2019; Zhang et al. 2020) of typical mass8 ± M (cid:12) (Kreidberg et al. 2012), the peak unabsorbed lumi-nosity corresponds to 0 . ± . L Edd , where L Edd is the Ed-dington luminosity. This is in reasonable agreement with therange of 0 . . L Edd found for canonical black hole XRBs(McClintock & Remillard 2009). We further found that thesystem transitioned from the soft to the hard X-ray spec-tral state at 0 . ± . L Edd , consistent with the range of0 . − . L Edd determined by Maccarone (2003), Kalemciet al. (2013) and Vahdat Motlagh et al. (2019) for typicalblack hole XRBs.Using the soft-to-hard X-ray spectral state transitionluminosity and the measured column density towards thesource, Tominaga et al. (2020) placed it in front of theScutum-Centaurus arm, at < i dis-tance is consistent with (albeit more precise than) this esti-mate, and implies a state transition luminosity towards thelower end of their assumed range. Tominaga et al. (2020) fur-ther used their measured inner disk radius in the soft X-rayspectral state to constrain the black hole mass as a functionof distance, inclination angle and black hole spin. Fitting theX-ray data of Tominaga et al. (2020) with the same kerrbb model, and leaving the distance free to vary within our 1 σ uncertainty range, we find that the lower limit on the in-ferred black hole mass (for the case of a non-rotating blackhole with an inclination angle of 0 ◦ ) reduces to 3 . M (cid:12) , ris-ing to 5 . M (cid:12) for an inclination angle of 60 ◦ . If the systemis at low inclination and slowly rotating, this weakens theevidence for a particularly massive black hole in the system. Over the past year, MeerKAT has conducted weekly XRBmonitoring (e.g. Bright et al. 2020; Tremou et al. 2020;Williams et al. 2020) under the large survey project Thun-derKAT (Fender et al. 2016). However, the low spectral res-olution limited the potential for H i studies. The recent cor-relator upgrade (providing 32K spectral channels with a ve-locity resolution of 6.1 km s − ) enables spectral resolution https://heasarc.gsfc.nasa.gov/cgi-bin/Tools/w3pimms/w3pimms.pl MNRAS000
MAXI /GSC data in the 2–20 keVenergy range. They found the peak of the outburst to haveoccurred on 2019 February 09, and the soft-to-hard spectralstate transition to have occurred on 2019 April 27 (Tominagaet al. 2020). We used the High Energy Astrophysics ScienceArchive Research Center (HEASARC) tool WebPIMMS tocalculate the bolometric X-ray flux in the energy range 0.01–100 keV from the X-ray flux (2–20 keV) and spectral modeldetermined by Tominaga et al. (2020). We derived a peakunabsorbed X-ray flux of 2 . ± . × − erg cm − s − , andan unabsorbed flux of 4 . ± . × − erg cm − s − for thesoft-to-hard X-ray spectral state transition, correspondingto luminosities of 1 . ± . × and 2 . ± . × erg s − ,respectively, at our preferred distance.If we consider the compact object to be a black hole(Russell et al. 2019; Zhang et al. 2020) of typical mass8 ± M (cid:12) (Kreidberg et al. 2012), the peak unabsorbed lumi-nosity corresponds to 0 . ± . L Edd , where L Edd is the Ed-dington luminosity. This is in reasonable agreement with therange of 0 . . L Edd found for canonical black hole XRBs(McClintock & Remillard 2009). We further found that thesystem transitioned from the soft to the hard X-ray spec-tral state at 0 . ± . L Edd , consistent with the range of0 . − . L Edd determined by Maccarone (2003), Kalemciet al. (2013) and Vahdat Motlagh et al. (2019) for typicalblack hole XRBs.Using the soft-to-hard X-ray spectral state transitionluminosity and the measured column density towards thesource, Tominaga et al. (2020) placed it in front of theScutum-Centaurus arm, at < i dis-tance is consistent with (albeit more precise than) this esti-mate, and implies a state transition luminosity towards thelower end of their assumed range. Tominaga et al. (2020) fur-ther used their measured inner disk radius in the soft X-rayspectral state to constrain the black hole mass as a functionof distance, inclination angle and black hole spin. Fitting theX-ray data of Tominaga et al. (2020) with the same kerrbb model, and leaving the distance free to vary within our 1 σ uncertainty range, we find that the lower limit on the in-ferred black hole mass (for the case of a non-rotating blackhole with an inclination angle of 0 ◦ ) reduces to 3 . M (cid:12) , ris-ing to 5 . M (cid:12) for an inclination angle of 60 ◦ . If the systemis at low inclination and slowly rotating, this weakens theevidence for a particularly massive black hole in the system. Over the past year, MeerKAT has conducted weekly XRBmonitoring (e.g. Bright et al. 2020; Tremou et al. 2020;Williams et al. 2020) under the large survey project Thun-derKAT (Fender et al. 2016). However, the low spectral res-olution limited the potential for H i studies. The recent cor-relator upgrade (providing 32K spectral channels with a ve-locity resolution of 6.1 km s − ) enables spectral resolution https://heasarc.gsfc.nasa.gov/cgi-bin/Tools/w3pimms/w3pimms.pl MNRAS000 , 1–5 (2020)
J. Chauhan et al. h m m m m − ◦ − ◦ − ◦ RA (J2000) D ec ( J ) MGPS J134353MGPS J134551MGPS J134625MGPS J135145MGPS J135236MGPS J135401MGPS J135546 MGPS J134559 F l uxd e n s it y ( m J y / b ea m ) Figure 1.
Left panel: A continuum mosaic of the field surrounding MAXI J1348–630, from our 1.34-GHz ASKAP observation on 2019February 13. The image has a size of 1.7 ◦ × ◦ , centered at RA = 13 h m . s
78, DEC = − d (cid:48) . (cid:48)(cid:48)
93. The location of MAXI J1348–630 is highlighted with the red circle, and the comparison extragalactic sources (listed in Table 1) are indicated by the blue squares.
Right panel: The time-resolved 1.34-GHz ASKAP light curve of MAXI J1348–630 (black stars) and the extragalactic sources, for beam20 only. The flux density of MAXI J1348–630 varies on a ∼ − − −
50 0 50 100 150
Radial velocity [km/s] − − − S ν / S [ % ] MAXI J1348–630 (S = 155 ± Figure 2.
The H i absorption complex observed in the directionof MAXI J1348–630 using our ASKAP observation from 2019February 13. S ν is calculated from the spectral cube, whereas S is measured from the continuum image. The red and blue curvesshow the H i absorption against MAXI J1348–630 and the stackof the eight extragalactic sources (Table 1), respectively. The cor-responding dotted lines represent the respective per-channel 3 σ noise levels (normalized to the source continuum flux density),taken from nearby source-free regions. The black dashed verti-cal line represents the rest frequency of the H i line. The dotted( −
31 km s − ) and dot-dashed vertical ( −
50 km s − ) lines high-light the ( > σ significant) maximum negative radial velocities(with respect to the LSR) observed from the spectra of MAXIJ1348–630, and the merged extragalactic sources, respectively.The maximum negative velocity of − ± − towards MAXIJ1348–630 determines the most probable distance as 2 . +0 . − . kpc,whereas the non-detection of more negative velocities sets a strin-gent upper limit of the tangent point at 5 . ± . comparable to that of ASKAP. Our data demonstrate thefuture potential of MeerKAT, which with its high sensitivitywill be well placed to routinely provide H i distance estimatesfor all bright, outbursting XRBs.The combined temporal coverage of our ASKAP andMeerKAT observations shows that the flare detected byASKAP (Fig. 1) is a secondary re-brightening, following thepeak detected during the MeerKAT observation (Carotenuto Distance [kpc] − − − − − R a d i a l v e l o c it y [ k m / s ] Figure 3.
The expected variation of the radial velocity of the localstandard of rest in the direction of MAXI J1348–630 with dis-tance from the Sun, calculated using the Monte Carlo approachdescribed by Wenger et al. (2018). The black line represents theexpected curve, and the blue shaded region shows the effect ofincorporating the 1 σ uncertainties on the Galactic rotation pa-rameters from Reid et al. (2014). The horizontal solid line showsthe maximum negative radial velocity (with respect to the LSR)measured from the H i absorption spectrum. The solid and dashedvertical lines show the most likely distance and the tangent pointdistance, respectively. The grey shaded regions show the 1 σ un-certainties. The most probable distance of MAXI J1348–630 isthe near kinematic distance of 2 . +0 . − . kpc. et al. 2019). This is not unusual, as multiple jet ejectionshave been observed in several previous black hole XRB out-bursts (e.g. Mirabel & Rodr´ıguez 1994; Brocksopp et al.2013). In future, the combination of MeerKAT and ASKAPobservations, together with higher frequency facilities, willprovide high-cadence light curves that can help constrainkey parameters such as jet speed, energetics and geometry(Tetarenko et al. 2017). We have used ASKAP to detect H i absorption towardsMAXI J1348–630 out to a maximum negative radial velocity(with respect to the LSR) of − ± − , implying amost probable kinematic distance of 2 . +0 . − . kpc. By compar- MNRAS , 1–5 (2020) he distance to MAXI J1348–630 − −
50 0 50 100 150 200
Radial velocity [km/s] − − − S ν / S [ % ] ASKAP MAXI J1348–630 [S = 155 ± = 486 ± Figure 4.
The MeerKAT (blue) and rebinned ASKAP (red) H i absorption spectra towards MAXI J1348–630. The 3 σ rms noiselevels for both instruments are shown as dotted lines. The twospectra match well within uncertainties. ison with the absorption towards a stack of the extragalacticsources in the field of view, we place a robust upper limit of5 . ± . ± M (cid:12) , we found that MAXI J1348–630 was accreting at 17 ±
10 % of the Eddington luminosityover the peak of the outburst, and the soft-to-hard X-rayspectral state transition happened at 2 . ± . i distance measurements for future black hole X-raybinaries in outburst. ACKNOWLEDGEMENTS
We thank the referee for their valuable comments. TheAustralian SKA Pathfinder is part of the Australia Tele-scope National Facility which is managed by CSIRO. Op-eration of ASKAP is funded by the Australian Govern-ment with support from the National Collaborative Re-search Infrastructure Strategy. ASKAP uses the resourcesof the Pawsey Supercomputing Centre. Establishment ofASKAP, the Murchison Radio-astronomy Observatory andthe Pawsey Supercomputing Centre are initiatives of theAustralian Government, with support from the Governmentof Western Australia and the Science and Industry En-dowment Fund. We acknowledge the Wajarri Yamatji peo-ple as the traditional owners of the Observatory site. TheMeerKAT telescope is operated by the South African Ra-dio Astronomy Observatory, which is a facility of the Na-tional Research Foundation, an agency of the Departmentof Science and Innovation. JCAM-J and GEA are the recip-ients of an Australian Research Council Future Fellowship(FT140101082) and a Discovery Early Career ResearcherAward (DE180100346), respectively, funded by the Aus-tralian Government.
DATA AVAILABILITY
The processed ASKAP data are available in the CASDAarchive at https://data.csiro.au/collections/ . The un-calibrated MeerKAT visibility data are publicly available,and stored at the South African Radio Astronomy Observa-tory (SARAO) Archive at https://[email protected] . REFERENCES
Atri P., et al., 2020, MNRAS, 493, L81Bright J. S., et al., 2020, Nature Astronomy, 4, 697Brocksopp C., Corbel S., Tzioumis A., Broderick J. W., RodriguezJ., Yang J., Fender R. P., Paragi Z., 2013, MNRAS, 432, 931Carotenuto F., Tremou E., Corbel S., Fender R., Woudt P., Miller-Jones J., 2019, ATel, 12497, 1Chauhan J., et al., 2019a, MNRAS, 488, L129Chauhan J., Miller-Jones J., Anderson G., Russell T., HancockP., Bahramian A., Duchesne S., Williams A., 2019b, ATel,12520, 1Denisenko D., et al., 2019, ATel, 12430, 1Dhawan V., Goss W. M., Rodr´ıguez L. F., 2000, ApJ, 540, 863Dickey J. M., 1983, ApJ, 273, L71Fender R. P., Belloni T. M., Gallo E., 2004, MNRAS, 355, 1105Fender R., et al., 2016, in MeerKAT Science: On the Pathway tothe SKA. p. 13 ( arXiv:1711.04132 )Gaia Collaboration et al., 2018, A&A, 616, A1Gathier R., Pottasch S. R., Goss W. M., 1986, A&A, 157, 191Goss W. M., Mebold U., 1977, MNRAS, 181, 255Hotan A. W., et al., 2014, Publ. Astron. Soc. Australia, 31, e041Jonker P. G., Nelemans G., 2004, MNRAS, 354, 355Kalemci E., Din¸cer T., Tomsick J. A., Buxton M. M., BailynC. D., Chun Y. Y., 2013, ApJ, 779, 95Kennea J. A., Negoro H., 2019, ATel, 12434, 1Kreidberg L., Bailyn C. D., Farr W. M., Kalogera V., 2012, ApJ,757, 36Kuchar T. A., Bania T. M., 1990, ApJ, 352, 192Maccarone T. J., 2003, A&A, 409, 697Matsuoka M., et al., 2009, PASJ, 61, 999McClintock J. E., Remillard R. A., 2009, in Lewin W., van derKlis M., eds, , Compact Stellar X-ray Sources. CambridgeUniv. Press, pp 157–214, doi:10.1017/cbo9780511536281.005McMullin J. P., Waters B., Schiebel D., Young W., Golap K.,2007, in Shaw R. A., Hill F., Bell D. J., eds, Astron. Soc. Pac.Conf. Ser. Vol. 376, Astronomical Data Analysis Software andSystems XVI. p. 127Miller-Jones J. C. A., Jonker P. G., Dhawan V., Brisken W., Ru-pen M. P., Nelemans G., Gallo E., 2009, ApJ, 706, L230Mirabel I. F., Rodr´ıguez L. F., 1994, Nature, 371, 46Murphy T., Mauch T., Green A., Hunstead R. W., PiestrzynskaB., Kels A. P., Sztajer P., 2007, MNRAS, 382, 382Reid M. J., McClintock J. E., Narayan R., Gou L., RemillardR. A., Orosz J. A., 2011, ApJ, 742, 83Reid M. J., et al., 2014, ApJ, 783, 130Reynolds J., 1994, AT Technical Document AT/39.3/040Russell T., Anderson G., Miller-Jones J., Degenaar N., EijndenJ. v. d., Sivakoff G. R., Tetarenko A., 2019, ATel, 12456, 1Sault R. J., Teuben P. J., Wright M. C. H., 1995, in Shaw R. A.,Payne H. E., Hayes J. J. E., eds, Astron. Soc. Pac. Conf. Ser.Vol. 77, Astronomical Data Analysis Software and SystemsIV. p. 433 ( arXiv:astro-ph/0612759 )Tetarenko A. J., et al., 2017, MNRAS, 469, 3141Tominaga M., et al., 2020, ApJ, 899, L20Tremou E., et al., 2020, MNRAS, 493, L132Vahdat Motlagh A., Kalemci E., Maccarone T. J., 2019, MNRAS,485, 2744Wenger T. V., Balser D. S., Anderson L. D., Bania T. M., 2018,ApJ, 856, 52Williams D. R. A., et al., 2020, MNRAS, 491, L29Yatabe F., et al., 2019, ATel, 12425, 1Zdziarski A. A., Poutanen J., Mikolajewska J., Gierlinski M., Ebi-sawa K., Johnson W. N., 1998, MNRAS, 301, 435Zhang L., et al., 2020, arXiv e-prints, p. arXiv:2009.07749This paper has been typeset from a TEX/L A TEX file prepared bythe author.MNRAS000