Transient Nature of J195754+353513 3

arXiv:1707.03878v1 [astro-ph.HE] 12 Jul 2017

htpeiu iiain ntedtcino ai tran- ⋆ exam- radio For astrophysical. of not detection expectation and the the instrumental were on with sients limitations consistent previous ra- are potential that which some transients revealed observations dio archival and rect follow-up emitters. through transient found suspected or been known have of most see, transients observations So, 2009)). radio (e.g., al. the efficiently et of Matsumura for sources 1986; searched Taylor radio which & (Gregory surveys transient radio re- and few (e.g Until variable only review). variability a are of for there (2004) levels cently McLaughlin high & Lazio show masers, Cordes, nuclei and stars, galactic bursts flaring active gamma-ray dwarfs, supernovae, vari- micro-quasars, brown time are pulsars, different example, There at variability For etc. show which scales. resolution constraints, sources time of observational tele- types poor of view, ous lack and of including width field reasons band more large studied (1995); stud- or not with is one Helfand well scopes sky to & radio due White considerably dynamic Becker, detail but in is (1998)) (e.g. al. sky et modeled Condon radio and state ied steady The INTRODUCTION 1 o.Nt .Ato.Soc. Astron. R. Not. Mon. aysciPal Sabyasachi J195754+353513 of nature Transient c 1 4 3 2 ninCnr o pc hsc,4 hlnia ai Stati Garia Chalantika, 43 Physics, Space for Centre Indian .N oeNtoa etefrBscSine,BokJ,Sec Block-JD, Sciences, Basic for NS Centre Kensington, National Parade, Bose Anzac N. Wales, S. South Edinbur Avenue, New West of Group, University Technology and Science Defense -al [email protected](SP) [email protected] E-mail: 07RAS 2017 eetporm osac o ai rninsfo di- from transients radio for search to programs Recent 1 ⋆ 000 umnaPatra Dusmanta – 21)Pitd1 ue22 M L (MN 2021 June 14 Printed (2017) 1–9 , uigtesa forsuyadi aidbtenls hn03myt mJy 0.3 than occa less one between varied In variability. it intra-day det and from high is rose study It showed our X-1. also Cygnus of re J195754+353513 from not span far was arcminutes the nature 23.8 transient during located its but is catalog source WENSS The sourc and The NVSS 1983 behavior. in between transient earlier data showing of J195754+353513 Kar years source the twenty th from a used data in We using (JVLA). sources band) Array (L radio Large GHz variable 1.4 in and/or X-1 Cygnus transient micro-quasar for searched have We ABSTRACT iia rpris h aueo h orei tl nnw.W spe We unknown. still is words: Key source the of tran nature. nature Galactic possible other The with properties. source the similar of count properties near-infrared the the compared be We may J19575420+3535152 2MASS 0.25. u:sas–sas ae–sas aibe:gnrl–sas ind J1745–3009 stars: GCRT – general J1742–3001, variables: GCRT stars: X-1, – Cygnus flare stars: – stars uum: oaie msincudb eetdfo h orewith source the from detected be could emission polarized ∼ 0myto mJy 20 ai otnu:gnrl–rdocnium rnins–rdocon radio – transients continuum: radio – general continuum: radio 1 oiu Hollick Monique nRa,708,India 700084, Road, on ∼ . o-I,Sl ae okt,709,India 700098, Kolkata, Lake, Salt tor-III, h ot utai,51,Australia 5111, Australia, South gh, 8 J 0 eod o iie ubro ae circularly cases of number limited For second. 700 mJy 180 ,25,Australia 2052, W, rnin ai ore r urnl o eycmo,es- common, of very detections not MWA that currently the show are see (2016)) sources also radio al. transient et (2014); Rowlinson al. by et (Bell study (MWA) Array field incoet h ot eeta oe(twr ta.2016) al. et (Stewart Pole deg Celestial 1925 North re covering the of to monitoring from close source gion transient one and only 2016) of al. detection et (Polisensky and (VLITE) Ionosphere Experiment Low-band Transient Array Large Very usi Jansky pointings the 2800 with deg of 12000 comprising recent search However, from transient sources systemic (2014)). transient any Macquart of radi detection (e.g. variable no and effectively transient more study sources to helps frequencies, dio (2007)). al. et source Lorimer rela- these (e.g. to believe origin due extra-galactic but some have unclear measures, al still dispersion et is high Ravi FRBs (2007); tively of al. only origin et lasted The Lorimer Radio which (2016)). Fast (e.g. far milliseconds of few so tens a detected few for been A co- have 2007). ultra- (FRBs) periodic, al. an Bursts et in A 2007; (Hallinan found 2008). was dwarf al. burst cool et al. polarized (Niinuma et circularly Kida and latitudes found herent 2007; were Galactic bursts al. low et radio Matsumura and Jy object 1–3 stellar high young Many at a 2003). from al. detected et was (Bower burst giant A ple, in orefo 40deg 1430 from source sient ueipoeet nfil fve,epcal tlwra- low at especially view, of field in improvements Huge A T E tl l v2.2) file style X 2 , 3 2 adpK Chakrabarti K. Sandip sn OA n odtcino tran- of detection no and LOFAR using 2 erhuigMrhsnWide- Murchison using search V/I vda:J195754+353513, ividual: aybten01 to 0.15 between vary a lomentioned also was e rato h source. the of erpart n 03 efound We 2003. and in ore having sources sient edo Galactic of field e ce aytimes many ected .Jnk Very Jansky G. l in h source the sion, otdearlier. ported uaeo its on culate 0 mJy. 201 o 4 , 1 tin- ng o 2 s - . 2 Pal et al. pecially at low radio frequencies. Future deep, wide-field significance of various findings in Section 4. Finally, we make searches may potentially be far more fruitful (e.g. recent concluding remarks in Section 5. transient rate calculations in Metzger, Williams & Berger (2015); Mooley et al. (2016)). Large archival data from various telescopes are im- 2 OBSERVATION AND DATA ANALYSIS portant resource to look for transient and variable radio WITH THE KARL G. JANSKY VERY sources. Earlier, Bannister et al. (2011) reported 15 tran- LARGE ARRAY sient sources from the study of 22 yr archival data of the Mo- 2 longlo Observatory Synthesis Telescope covering 2776 deg A blind search for new variable radio sources was conducted survey area. Recently, Murphy et al. (2017) found a candi- using archival JVLA data1 at the L-band frequency of 1400 date transient source at low frequency from comparison of MHz. L band is good for this kind of search with JVLA as it TIFR GMRT Sky Survey Alternative Data Release 1 (TGSS provides right balance between field of view and sensitivity. ADR1, see Intema et al. (2017)) and the GaLactic and Ex- Though the field of view in 74 and 325 MHz band with JVLA tragalactic All-sky Murchison Widefield Array (GLEAM, will be higher, these bands have relatively poor sensitivity. see Hurley-Walker et al. (2017)) survey catalogues. Many Availability of archival data is also less in frequencies less variable and transient radio sources were reported from than 1400 MHz. Also, for some sub-classes, the transient archival data search of the NRAO VLA Sky Survey (NVSS) detection rate is higher at 1400 MHz, as discussed in Section (Condon et al. 1998) and FIRST survey (Levinson et al. 1. 2002; Thyagarajan et al. 2011). Ten milli Jansky level tran- JVLA comprises of twenty seven fully steerable anten- sients were detected from 22 years of archival the Karl G. nas each with 25-meter diameter in a Y-shaped array. It is Jansky Very Large Array (JVLA) observations of a single located around eighty km west of Socorro, New Mexico. An- field-of-view at 5 and 8.4 GHz (Bower et al. 2007). Though it tennas are periodically moved in different configuration to was shown later that more than half of the transient sources achieve different scientific goals where the most expanded reported in Bower et al. (2007) were due to data artifacts antenna configuration is known as A configuration and the and rest of the sources had low signal-to-noise ratio (S/N) most compact one is D configuration. B and C configura- than Bower et al. (2007) to conclusively find transient na- tions are intermediate between A and D. Occasionally an- ture of these sources (Frail et al. 2012). tennas are placed in a hybrid configurations, like AB or BC Earlier detection of two transient radio sources GCRT when some of the antennas are in one configuration and J1745–3009 (Hyman et al. 2005, 2007; Roy et al. 2010) and some of them in other. The maximum size of the baselines GCRT J1742–3001 (Hyman et al. 2009) were made from Bmax in A, B, C and D configurations are 36.4, 11.1, 3.4 systematic search near Galactic Center region using The and 1.03 km respectively which corresponds to synthesized Karl G. Jansky Very Large Array (JVLA) and Giant Me- beam-width (θHPBW ) 1.4, 4.6, 15 and 49 arcsec respectively terwave Radio Telescope (GMRT). GCRT J1745–3009 is a at 1400 MHz. unique source which was detected only three times in 2002 The data used for the present work was taken from dif- (Hyman et al. 2005), 2003 and 2004 (Hyman et al. 2007). In ferent configurations of the JVLA between 12 October 1983 2002, this source exhibited ∼10 minute long, ∼1 Jy peaked (MJD 45619) and 4 June 2003 (MJD 52794). In total, we bursts with a ∼77 minute period. The emission from the have used 262 different epochs of observations with various source was coherent (Hyman et al. 2005) with extremely intervals ranging from days to years between successive ob- steep spectral index (α = −13.5 ± 3.0) and high circular servations. The antennas were in configuration A, AB, AD, polarization (Roy et al. 2010). All the three detections of B, BC, C, CD, and D for 53, 16, 5, 46, 15, 45, 26, and 56 the source were in 330 MHz. The characteristics of GCRT days, respectively. J1742–3001 did not match with any known mechanisms of We studied a field centered on Cygnus X-1, a radio emit- emission in transient compact sources. As a result, the source ting X-ray binary (Bowyer et al. 1965; Giacconi et al. 1972) ◦ ′ ′′ seems to represent a member of a new class of coherently with co-ordinates 19h58m21.676s +35 12 05.78 (J2000). emitting objects. GCRT J1742–3001 was active for a few This area of the sky was chosen because the field of Cygnus months with ∼150 mJy flux density at the peak and showed X-1 is one of the best studied Galactic black hole binaries no periodicity in emission (Hyman et al. 2009). The source using JVLA due to the fact that it is the first system widely was detected in 235 MHz and exhibited steep spectral index accepted to contain a black hole (Murdin & Webster 1971; (α < −2). For both of these sources, no counterpart was Tananbaum et al. 1972; Gies & Bolton 1986). This make the discovered in high energy, making them impossible to be field ideal for looking variable/transient sources. detected by conventional follow-up radio observations from There are big data gap between April 1986 to April high energy observations. 1988, May 1991 to October 1996 as well as February 2001 In this paper, we look for new variable/transient sources to June 2003. The ‘on source’ observation time in each epoch from a single well observed field centered at micro-quasar was between 2 to 15 minutes. Observing bandwidth for most Cygnus X-1. Though we could not detect a new source, we of the days was 50 MHz. discovered hitherto un-reported transient behavior of one Analysis and imaging of the data was carried out with NVSS source, namely, NVSS J195754+353513. This source Astronomical Image Processing System (AIPS)2. Bad data is located approximately 23.8 arcminutes far from the micro- were flagged. Perley & Butler (2013) flux density scale was quasar Cygnus X-1 at J2000 co-ordinates 19h57m54.3s (± ◦ ′ ′′ ′′ 0.7s) +35 34 59.6 (±0.6 ). In Section 2, we summarize observational details and data analysis procedure. In Section 1 https://science.nrao.edu/facilities/vla/archive/index 3, we summarize various results on the source. We discuss 2 http://www.aips.nrao.edu

c 2017 RAS, MNRAS 000, 1–9 Transient nature of J195754+353513 3

0 5 10 -2 0 2 4 6 8

35 36 30 35 36 30

00 00

35 30 35 30

00 00 Declination (J2000) Declination (J2000)

34 30 34 30

00 00

19 58 00 57 58 56 54 52 50 48 19 58 00 57 58 56 54 52 50 48 Right ascension (J2000) Right ascension (J2000)

Levs = 2.014E-03 * (1, 2, 3, 4, 5, 6, 7, 8, 9, 10) Jy/beam Levs = 1.710E-03 * (1, 2, 3, 4, 5, 6, 7, 8, 9, 10) Jy/beam

-10 0 10 20

35 36 30

35 45

00 Declination (J2000)

34 45

19 58 00 57 58 56 54 52 50 48 Right ascension (J2000)

Levs = 9.544E-04 * (1, 2, 3, 4, 5, 6, 7, 8, 9, 10) Jy/beam

0 20 40

35 36 30

35 30

00 Declination (J2000)

34 30

19 58 02 00 57 58 56 54 52 50 48 Right ascension (J2000)

Levs = 7.422E-03 * (1, 2, 3, 4, 5, 6, 7, 8, 9, 10) Jy/beam

Figure 1. Images of J195754+353513 in different configurations of JVLA at 1490 MHz. In the upper left and right panels, we have shown images of the source in B configuration on 7th November 1999 and 11th February 2000 with flux density 56.5 mJy and 26.0 mJy respectively. In the middle left, middle right and lower left panels, we have shown examples of detection of the source in A, C and D configurations of JVLA. In the lower right panel, we have shown an example of no detection when JVLA was in C configuration. We have included the synthesized beam at the left-bottom corner of each panel. The location of the NVSS position is shown by a cross in all the images where size of the cross is error in NVSS location multiplied by 10 for easy visualization. For details, see text and Table 1.

c 2017 RAS, MNRAS 000, 1–9 4 Pal et al.

Table 1. Details of observations corresponding to Figure 1

Image Date Date Conﬁguration Flux Density RMS Synthesized beam On source time Location (UT) (MJD) (mJy) (mJy beam−1) (sec)

Upper Left 07/11/1999 51489 B 56.5 0.7 4.60′′× 4.22′′ 170 Upper Right 11/02/2000 51585 B 26.0 0.4 4.44′′× 4.14′′ 120 Middle Left 04/03/1990 47954 A 56.2 0.3 1.28′′× 1.17′′ 13820 Middle Right 06/05/1984 45826 C 53.0 1.6 15.56′′× 12.48′′ 390 Lower Left 18/03/1991 48333 D 80.7 2.0 19.72′′× 19.72′′ 1140 Lower Right 31/08/1997 50691 C — 5.5 14.70′′× 14.48′′ 100 used. For eighteen epoch of observations, the quality of data 3 RESULTS was not good and we could not make reasonably good images. We did not use data for these days. While using data 3.1 Radio light curve of J195754+353513 for CD and D configuration, we used lower uv cut-off to avoid strong background extended radiation from the Galac- We have detected a transient radio source with co-ordinate ± ◦ ′ ′′ ± ′′ tic plane. We have made correction for the primary beam 19h57m54.3s ( 0.7s) +35 34 59.6 ( 0.6 ) (J2000). This using AIPS task PBCOR. position is the average of result from fitting an elliptical Gaussian around the peak of the source, along with a background level to the source and the error is 1σ uncertainty Amplitude calibration was conducted in reference to 3C in the fitting. The corresponding Galactic co-ordinates are ◦ ◦ 286 and phase calibration was based on observations of the l = 71.61 , b = 3.34 . After cross-correlation with NVSS nearby source J2007+404 or J2015+371. We have not per- sources, we found the transient source detected by us is formed self calibration due to lack of any strong point source NVSS J195754+353513. in the field. The integration time used for solving amplitude In the Figure 1, examples of images of J195754+353513 and phase during calibration was 10 sec. The images of dif- are shown during observations in different JVLA configura- ferent days have variable noise levels. Since the source is tions and time. The image of the source in B configuration bit far from the field centre, the noise level near the source on 7th November 1999 and 11th February 2000 with respec- position is high. Also small on-source time for individual ob- tive flux density 56.5 mJy and 26.0 mJy is shown in the servation resulted relatively high noise level. The noise level upper left and upper right panels of the Figure. In the mid- −1 near the source position varied between 0.08 mJy beam dle left, middle right and lower left panels, we have shown −1 and 18.7 mJy beam . The median value of noise was 1.8 examples of detection of the source in A, C and D config- −1 mJy beam . urations of JVLA. In the lower right panel, an example of no detection is shown when JVLA was in C configuration. We have included the synthesized beam at the left-bottom Two models with Intermediate Frequencies (IFs) 1.3851 corner of each panel. The observation details and different GHz (model 1) and 1.4649 GHz (model 2) were created image parameters corresponding to Figure 1 is summarized combing data of 94 and 18 days respectively. Model with in Table 2. similar frequency configuration was subtracted with indi- The location of NVSS J195754+353513 is shown in all vidual single epoch observations using the AIPS task UVSUB images of Figure 1. The source is within extended emission to look for variable sources. Apart from Cygnus X-1, we region of J195754+353513. NVSS survey is carried out us- found that only one source J195754+353513 is present in ing D configuration of JVLA and the source is unresolved many of the subtracted images with significantly fluctuat- in NVSS with flux density 50.8 ± 1.6 mJy at 1400 MHz ing flux. The subtracted flux density for J195754+353513 (Condon et al. 1998). It was not resolved during our study was up to ∼120 mJy. The noise levels close to the location also when it was observed in D configuration. of J195754+353513 in model 1 and model 2 were 0.19 mJy − − beam 1 and 0.46 mJy beam 1 respectively. The elongation of the source visible in the Figure 1 in north-south direction is most probably not due to intrinsic source structure but an effect of bandwidth smearing due to To be sure that the variation of J195754+353513 is high channel width of 50 MHz. The direction of elongation not due to any kind of systematic effects or error in ampli- of the source, as visible in Figure 1, is towards the point- tude calibration, we measured flux density of another source ing center which is a signature of bandwidth smearing. In NVSS J195823+345728 present in our field. The recorded all the images, the source is consistently elongated in same flux density of the source in NVSS catalog is 52.3 ± 1.6 direction, when detected. mJy (Condon et al. 1998). We found that during our study, Light curves displaying variation of the source’s flux the mean and median flux density of the source were 50.8 densities between 1983 to 2003 is shown in Figure 2. The and 51.7 mJy respectively. The standard deviation in mea- flux density is calculated using AIPS task JMFIT using the surements of flux-densities of the source was 5.2 mJy which results of fitting a Gaussian, along with a background level. means the percentage of deviation was 9.9%. This shows er- The non-detections are reflected by the triangular points, ror in amplitude calibration play little role in large variation which correspond to the 3σ upper limits. The error bars rep- of J195754+353513, which will be discussed in more detail resent the rms noise levels of the images near the location in coming sections. of J195754+353513 and a 10% uncertainty in the absolute

c 2017 RAS, MNRAS 000, 1–9 Transient nature of J195754+353513 5

200 200

150 150

100 100 Flux Density (mJy) Flux Density (mJy) 50 50

0 0 01/83 01/84 01/85 01/86 01/87 01/88 01/89 01/90 01/91 01/92 01/96 01/97 01/98 01/99 01/00 01/01 01/02 01/03 01/04 Date Date

Figure 2. Light curves of J195754+353513 at 1490 MHz. In left panel the observations between 1983 and 1991 are plotted and in the right panel the observations between 1996 and 2003 are plotted. The observations are done by The JVLA. Here the triangles show days of non-detection with 3σ upper limits. calibration of each data set added in quadrature. There are ern Sky Survey; Rengelink et al. (1997)) catalog which is only nineteen observations of the field between MJD 46612 5.11 arcsec far from the NVSS position of the source. Since (July, 1986) and MJD 50326 (August, 1996). The source was the positional uncertainty in WENSS for faint sources are ∼ detected in 161 occasion and not detected in 83 occasions. 5 arcsec, this is most probably the same source. The source α The data was not good enough to make reasonably good has a flat spectral index of α = −0.19 (assuming Sν ∝ ν ) images for 18 days. So, the source was successfully detected between NVSS and WENSS measurements (Massaro et al. for 66.0% days amongst the all observations with good data 2014). Since the source is highly variable and there is gap of the field. The source showed multiple short bursts and between measurements between NVSS and WENSS (obser- high variation. It varied from less than 0.3 mJy to 201 mJy. vations of WENSS took place between 1991–1993 and ob- The median value of flux density was 38.9 mJy. On 25th servations of NVSS took place between 1993–1996), the flat January 2002 (MJD 52299) J195754+353513 reached maxi- spectral index measured between NVSS and WENSS is mis- mum value within the observational period reported in this leading. paper with 201 mJy flux density. The source showed signature of various flares and the peaks reached more than 150 mJy on 23rd July 1998 (186 mJy), 24th March 1999 (175 3.2 Intra-day flux density variation mJy) and 16th June 2000 (160 mJy). Based on previous studies of the bursting transient GCRT Due to gaps in the observations, we can study a com- J1745–3009 (Hyman et al. 2005, 2007; Roy et al. 2010) and plete flare only in a few cases. In Figure 3 we have shown two GCRT J1742–3001 (Hyman et al. 2009), we searched for such flares in March and April-May of 1991 when the peak flux density variations of J195754+353513 between differ- ∼ ∼ reached 85 mJy and 75 mJy respectively. Antennas were ent scans of the observations of same day. We also imaged in D configuration during these times. Though there were each minute separately, when the source is detected, to look not enough data points to precisely determine life-span and for minute-scale variation of J195754+353513. Since the flux rise/decay rate of these flares, the approximate life span of density of the source was not adequate to image it all the these flares were in the range of 10–18 days. Both of these time with smaller time scale, we could make a study of the flares show sharp rise and relatively slow decay. intra-day variation only for limited amount of time. In Figure 4 we have shown the field of J195754+353513 Significant scan to scan and intra-scan flux density vari- combining 9 days of observations. All observations used in ation is detected when it was possible to image separate this Figure was made in 1991 when antennas were in D scans. In the right panel of Figure 3 we have plotted an configuration. The RMS noise of the image was 0.17 mJy example of intra-day variation of J195754+353513 within −1 ′′ × ′′ beam and the resolution of the image was 43.23 41.71 . a scan. The observation took place on 3rd February 2000 We can clearly see J195754+353513 with 60 mJy flux density (MJD 51577). Antennas were in B configuration during this along with Cygnus X-1 and many other background sources. observation. The source rose from 20 to 180 mJy within 700 We looked for possible periodicity in J195754+353513 seconds and then came back to ∼20 mJy level within ∼100 as was found in some other radio transients. The light- seconds. The rise time constant, resulting from the exponen- curve data was systematically folded with different period tial fit over the points in rising phase, is τ = 342.8 ± 89.0 β to search for signature of any periodicity in the emission. seconds. A power-law fit (Sν ∝ t ) over points in rising No periodicity was found in the source light curve. phase of the flare yields β = 1.5 ± 0.8. There are no available JVLA archival data of the source In Table 2, we have summarized intra-day flux density in any other band except L band. Thus a study of the spec- variations for different days. The error in flux density men- tral index of the source was not possible using JVLA data. tioned in the table is RMS noise near the location of the There is a source in 325 MHz WENSS (Westerbork North- source. For five different days we found more than 20% dif-

c 2017 RAS, MNRAS 000, 1–9 6 Pal et al.

200

90 180 f(t)=aet/b 80 160 a=27.3 ± 1.1 b=342.8 ± 89.0 70 140 χ2=7.4

60 120

50 100

40 80

30 60 Flux Density (mJy) Flux Density (mJy)

20 40

10 20

0 0 16/03/91 23/03/91 30/03/91 06/04/91 13/04/91 20/04/91 27/04/91 04/05/91 0 100 200 300 400 500 600 700 800 Date Time (Sec)

Figure 3. Left (Inter-day variation): Example of two flares of J195754+353513 observed at L band. These flares were observed in March and April-May 1991. Antennas were in D configuration during these time. Right (Intra-day variation): An example of intra-day variation of J195754+353513 within a scan. The observation took place on 3rd February 2000. Antennas were in B configuration during this observation. Exponential fitting over the points in rising phase is also shown in the figure.

Table 2. Intra-day Flux Density variation of J195754+353513

Date 1st Scan 1st Scan 2nd Scan 2nd Scan Time gap (UT) Time Span Flux Density Time Span Flux Density between scans (min) (mJy) (min) (mJy) (min)

03/07/84 6.5 50.9 ± 8.4 6.5 216 ± 7.9 52.5 04/03/90 29.16 194 ± 0.7 29 31.4 ± 0.7 204.5 16/03/91 19.0 68.1 ± 2.4 13.5 55.4 ± 2.1 5.5 25/03/91 15.5 88.1 ± 3.6 14.5 37.9 ± 2.6 5.5 16/03/97 0.92 89.8 ± 1.2 0.92 127 ± 2.8 0b 25/01/99 1.33 111 ± 5.0 1.33 48.4 ± 3.7 0b 24/10/02 2.83 37.6 ± 2.6 3.17 76.7 ± 2.8 57.5

bNote: Zero difference between two time-spans mean we are looking for intra-scan variations. ference in flux densities between two successive scans of ob- December 1999 (MJD 51513) with flux density 8.3 mJy and − servations with time difference 5.5 minute to 204.5 minute. RMS noise 0.6 mJy beam 1. The corresponding V/I was On 4th March 1990 (MJD 47954), the flux decayed from 194 0.15. More than 5σ Stokes V was not detected from data of mJy to 31.4 mJy with just 204.5 minute separation and on any other days. 3rd July 1984 (MJD 45884), the flux density rose from 50.9 mJy to 216 mJy when the gap between two successive scans were just 5.5 minutes. For two days, we found significant variation within a scan. 3.4 Optical/IR and X-ray counterpart of the The inter and intra scan variation, often more than source twice in flux density, suggests the radio emission from J195754+353513 consists of many small bursts. No known sources are reported to emit X-ray emission from the nearby position of J195754+353513 making it impossible to detect radio emission from follow-up observation of X-ray emission at the time of flare. 3.3 Circularly polarized emission We have searched for optical/infra-red counterpart of We looked for the presence of circularly polarized emission J195754+353513. There are two sources within 10 arc- from J195754+353513 as was found for the case of GCRT sec from the source location of J195754+353513 in the J1745–3009 (Roy et al. 2010). The measurement of Stokes V Two Micron All-Sky Survey (2MASS) point source catalog polarization on 19th March 1991 (MJD 48334) was 12.6 mJy (Cutri et al. 2003; Skrutskie et al. 2006). The nearest source with RMS noise of 1.19 mJy beam−1 and the measurement in 2MASS catalog is J19575420+3535152 whose location is of Stokes V polarization on 2nd May 1991 (MJD 48378) 1.82 arcsec away from J195754+353513 and may be the near − was 11.5 mJy with RMS noise of 1.36 mJy beam 1. The infra-red counterpart of J195754+353513. The other source corresponding value of V/I on 19th March and 2nd May in 2MASS catalog within 10 arcsec from J195754+353513 is 1991 were 0.25 and 0.24 respectively with 3% uncertainty in J19575378+3535221 whose position is 9.37 arcsec away from error. Relatively weak Stokes V detection was done on 1st J195754+353513. The brightnesses of J19575420+3535152

c 2017 RAS, MNRAS 000, 1–9 Transient nature of J195754+353513 7 in J, H and K bands are 15.45, 14.90 and 14.55 mags re- from minutes to months, even if interstellar scattering play spectively. some role, it is unlikely that all this variations are due to scattering effect. For the most of black-hole X-ray binaries, a universal correlation between radio and X-ray luminosities has been 4 DISCUSSION reported (Corbel et al. 2003; Gallo, Fender & Pooley 2003). Assuming the relation given in Gallo, Fender & Pooley 4.1 The nature of the source J195754+353513 (2003), the ∼ 200 mJy peak should have corresponding X- We have searched the environment of J195754+353513 for ray flux of ∼ 0.25 Crab in its hard state. Such a strong flare associated discrete sources or extended structures. The in X-ray would not go un-noticed by all sky X-ray moni- source is not close to any known supernova remnant. We tors, given that the source is in flaring stage quite regularly. did not find trace of any extended structure close to the Though some of the exceptions are found for X-ray binaries source, either. which do not follow this universal correlation, all the ex- We have found one 2MASS source within 1σ positional amples have lower radio luminosity than the universal value error of J195754+353513 and it is possible that the radio but none of them have significant lower X-ray luminosity. So, emission of J195754+353513 arises from activities in a fore- it is unlikely that the system is a black-hole X-ray binary ground flaring star. Many flaring stars are known to ex- system. hibit activity in both radio and X-ray wavelengths such as On the other hand, if we consider a two component the giant outburst from a young stellar object reported by accretion flow (TCAF, Chakrabarti & Titarchuk (1995)) Bower et al. (2003), but the detection of radio flares hav- where the Keplerian disk is flanked by a transonic flow in ing no apparent associated X-ray emission is not uncom- hard states when jets are stronger, it is well known that the mon. For example, radio flares from UV Ceti stars with sec- Keplerian disk does not have to penetrate distances close to onds to minutes time-scale were detected at low frequencies the black hole and the X-ray could be faint. We postulate by Spangler, Shawhan & Rankin (1974) and Karpen et al. that the disk with very low accretion rate could be located (1977) with YZ Canis Minoris where no corresponding X-ray at a large distance with inner edge ∼ 20000 Schwarzschild emission was found. At higher frequencies (4.9 and 8.4 GHz), radii or more so that the source is presently having a qui- Osten et al. (2005) reported short duration radio flares from escence stage, as far as X-rays is concerned. If this is the the dMe flare star EV Lacertae which was not clearly re- case, occasional outbursts every few to few tens of years is lated to the star’s X-ray flares. The radio flares reported possible and one could look for them. It is also possible that in these stars range from a few milli Jansky to a few tens the disk is shrouded by copious winds from the companion, of milli Jansky, with rise and decay times of ∼ 1 min and much like SS 433 (Margon 1984; Clark 1984), where disk ∼ 1 hr respectively. We have also found small flares from X-ray is completely blocked. In fact, the circularly polarized J195754+353513 of ∼ 700 s duration as featured in Figure emission is an indication of aligned magnetic fields. If there 3. is an outflow with a small inclination angle with the line Richards et al. (2003) presented results on five years of of sight, a polarization fraction similar to those reported in continuous monitoring of radio flares of Algol-type and RS Section 3.3 can be produced from its radio emission. How- CVn systems; many of the flares reached hundreds of milli ever, in that case, a profusely mass-losing star star would Jansky with a few days to a month duration. Numerous be expected in the vicinity. As discussed in Section 3.4, it short bursts within each flare were also visible. Strong pe- is possible that 2MASS J19575420+3535152 is the IR coun- riodic activities are also found in these systems where the terpart. The time-scale of rising and fading for every micro- shortest periodicity was found in β Per with 48.9±1.7 days. flare would be expected to be of much shorter duration for Though we could not detect any periodicity in emission from a compact binary. J195754+353513, the source could be RS CVn system due to similarity in time scale of flaring episodes. 4.2 Comparison with Galactic Center Transients Pulsars can produce highly circularly polarized emission in single pulse (eg. Kazbegi et al. (1991)). There are some There are many similarities between the temporal evolu- pulsars which show non-periodic flaring events like Crab tion of J195754+353513 to that of GCRT J1742–3001 and a pulsar (e.g. Buehler et al. (2012)). No known radio pulsar transient radio source detected close to Galactic Center re- is reported from the nearby location of J195754+353513. gion (GCT). While GCRT J1742–3001 was detected as part Also we have not seen any sign of nearby supernova rem- of transient search program near Galactic centre region in nant or nebula. Though majority of pulsars are associ- March 2006 to May 2007 at 235 MHz (Hyman et al. 2009), ated with supernova remnants (Frail et al. 1994), some be- GCT was detected in monitoring observations of Sgr A* lief that this association are by chance and actually false from December 1990 to September 1991 at different radio (Gaensler, & Johnston 1995a,b,c). One should look for pos- wavelengths from 1.3 to 22 cm (Zhao et al. 1992). There are sible pulsar emission in the location of J195754+353513. also some similarity with another transient GCRT J1745– Variations in the light curve of J195754+353513 could 3009 (Hyman et al. 2005, 2007; Roy et al. 2010) located also be due to some kind of extreme scattering of the inci- close to Galactic centre at 330 MHz. The flare of both GCRT dent radiation in the interstellar medium of our Galaxy (e.g. J1742–3001 and GCT took about a month to rise while it Rickett (1990); Dennett-Thorpe & de Bruyn (2002)). Vari- decayed in about three months. For J195754+353513 we ation up to few day time-scale can occur due to interstel- could not catch total rising and fading profile of many lar scintillation in GHz frequencies (Peng et al. 2000). Since flares due to inadequate sampling rate and data gap, typ- J195754+353513 showed variation of different time-scales, ical rising time of flares were ∼5 days and typical fading

c 2017 RAS, MNRAS 000, 1–9 8 Pal et al.

X-ray binary (Zhao et al. 1992), which is yet to be estab- lished.

5 CONCLUSIONS We found transient nature of a radio source J195754+353513, approximately 24 arcminutes far from Galactic micro-quasar Cygnus X-1. The source showed high variability with different time-scales. J195754+353513 showed evidence of variable circular polarized emission. The source has no known X-ray counter part of the system. 2MASS J19575420+3535152 may be an NIR counterpart of the source. The nature of the source is still unknown. It is not unlikely that the system is a black-hole X-ray binary with the disk covered by significant amount of matter from the companion. On the other hand, it is also possible that the emission is coming from a foreground flare star. Clearly more multi-frequency monitoring is required to come to a definite conclusion.

ACKNOWLEDGMENTS We acknowledge the anonymous referee whose detail and Figure 4. 1490 MHz JVLA image of the field of J195754+353513 combining 9 days of observations. All observations used in this productive suggestions helped to improve the manuscript image were conducted in 1991 when antennas were in D configu- significantly. DP and SP acknowledge support of MOES ration. The RMS noise of the image is 0.17 mJy beam−1. Cygnus fund to carry out this project. We have used data from JVLA X-1 and J195754+353513 are indicated by the arrow. The syn- which is run by The National Radio Astronomy Observatory thesized beam of the image is 43.24′′ × 41.71′′. (NRAO). NRAO is a facility of the National Science Foun- dation operated under cooperative agreement by Associated Universities, Inc. time was ∼10 days. Though the active time of individual flares of J195754+353513 were less than GCRT J1742–3001 REFERENCES and GCT, J195754+353513 exhibited higher frequency of such flares unlike other two sources mentioned above. GCRT Bannister K. W., Murphy T., Gaensler B. M., Hunstead R. J1742–3001 also showed fewer small bursts before the main W., Chatterjee S., 2011, MNRAS, 412, 634 flare and the GCT showed the presence of a significantly in- Becker R. H., White R. L., Helfand D. J., 1995, ApJ, 450, tense secondary burst in the 18–22 cm observations about six 559 months after the primary burst. Though GCRT J1745–3009 Bell M. E. et al., 2014, MNRAS, 438, 352 was detected only three times, it showed intra-day variability Bower G. C., Plambeck R. L., Bolatto A., McCrady N., with time-scale of hundreds of second as J195754+353513. Graham, J. R., De Pater I., Liu M. C., Baganoff F. K., GCRT J1742–3001 peak flux density was ∼100 mJy in 2003, ApJ, 598, 1140 235 MHz and peak flux density of GCT was ∼1 Jy in the Bower G. C., Saul D., Bloom J. S., Bolatto A., Filippenko wavelength range 18–22 cm (∼1.5 GHz). The peak flux den- A. V., Foley R. J., Perley D., 2007, ApJ, 666, 346 sity of individual flares of J195754+353513 varied between Bowyer S., Byram E. T., Chubb T. A., Friedman H., 1965, 30 to 200 mJy. Since GCRT J1742–3001 had a very steep Science, 147, 394 α spectrum (α . −2, Sν ∝ ν ), its peak flux density should Buehler, R. et al., 2012, ApJ, 749, 26 be fainter than J195754+353513 in 1400 MHz. Assuming Chakrabarti S., Titarchuk L. G., 1995, ApJ, 455, 623 α = −2, we calculate peak flux density of GCRT J1742– Clark D. H., 1984, The quest for SS 433, New York: Viking 3001 at 1400 MHz to be .21 mJy. The peak flux-densities Adult, ISBN 9780140089967. for three detections of GCRT J1745–3009 were ∼ 1 Jy, ∼ 0.5 Condon J. J., Cotton W. D., Greisen E. W., Yin Q. F., Jy and ∼ 0.06 Jy. The peak intensity may indicate strength Perley R. A., Taylor G. B., Broderick J. J., 1998, AJ, 115, of magnetic fields. 1693 J195754+353513 showed presence of variable circular Corbel S., Nowak M. A., Fender R. P., Tzioumis A. K., polarization as GCRT J1745–3009 (Roy et al. 2010). Cir- Markoff S., 2003, A&A, 400, 1007 cular polarization was not reported for the case of GCRT Cordes J. M., Lazio T. J. W., McLaughlin M. A., 2004, J1742–3001 and GCT. New Astronomy Reviews, 48, 1459 No X-ray counter-parts were confirmed for the case Cutri R. M. et al., 2003, VizieR Online Data Catalog, 2246 of GCT, GCRT J1742–3001 and GCRT J1745–3009. Dennett-Thorpe J., de Bruyn A. G., 2002, Nature, 415, 57 J195754+353513 also has no known X-ray counter part. Frail D. A., Goss W. M., Whiteoak J. B. Z., 1994, ApJ, There is some suggestion that GCT may be associated with 437, 781

c 2017 RAS, MNRAS 000, 1–9 Transient nature of J195754+353513 9

Frail D. A., Kulkarni S. R., Ofek E. O., Bower G. C., Nakar Roy S., Hyman S. D., Pal S., Lazio T. J. W., Ray P. S., E., 2012, ApJ, 747, 70 Kassim N. E., 2010, ApJ, 712L, 5R Gaensler B. M., Johnston S., 1995a, MNRAS, 275, 73 Skrutskie M. F. et al., 2006, AJ, 131, 1163 Gaensler B. M., Johnston S., 1995b, MNRAS, 277, 1243 Spangler S. R., Shawhan S. D., Rankin J. M., 1974, ApJ, Gaensler B. M., Johnston S., 1995c, PASA, 12, 76 190, L129 Gallo E., Fender R. P., Pooley G. G., 2003, MNRAS, 344, Stewart A. J. et al., 2016, MNRAS, 456, 2321S 60 Tananbaum H., Gursky H., Kellogg E. M., Levinson R., Giacconi R., Murray S., Gursky H., Kellogg E., Schreier Schreier E., Giacconi R., 1972, ApJ, 174, L143 E., Tananbaum H., 1972, ApJ, 178, 281 Thyagarajan N., Helfand D. J., White R. L., Becker R. H., Gies D. R., Bolton C. T., 1986, ApJ, 304, 371 2011, ApJ, 742, 49 Gregory P. C., Taylor A. R, 1986, AJ, 92, 371 Zhao J. H. et al., 1992, Science, 255, 1538 Hallinan G. et al., 2007, ApJ, 663, L25 A Hartman J.D., Bakos G., Stanek K.Z., Noyes R.W., 2004, This paper has been typeset from a TEX/ LTEX ﬁle prepared AJ, 128, 1761 by the author. Hurley-Walker et al., 2017, MNRAS, 464, 1146 Hyman S. D., Lazio T. J. W., Kassim N. E., Ray P. S., Markwardt C. B., Yusef-Zadeh F., 2005, Nature, 434, 50H Hyman S. D., Lazio T. J. W., Roy S., Ray P. S., Kassim N. E., Neureuther J. L., 2006, ApJ, 639, 348H Hyman S. D., Roy S., Pal S., Lazio T. J. W., Ray P. S., Kassim N. E., Bhatnagar S., 2007, ApJ, 660L, 121H Hyman S. D., Wijnands R., Lazio T. J. W., Pal S., Starling R., Kassim N. E., Ray P. S., 2009, ApJ, 696, 280 Intema H. T., Jagannathan P., Mooley K. P., Frail D. A., 2017, A&A, 598, 78 Karpen J. T. et al., 1977, ApJ, 216, 479 Kazbegi A. Z., Machabeli G. Z., Melikidze G. I., 1991, MN- RAS, 253, 377 Kida S. et al., 2008, New Astron., 13, 519 Levinson A., Ofek E. O., Waxman E., Gal-Yam A., 2002, ApJ, 576, 923 Lorimer D. R., Bailes M., McLaughlin M. A., Narkevic D. J., Crawford F., 2007, Science, 318, 777 Macquart J., 2014, PASA, 31, 31 Margon B., 1984, ARA&A, 22, 507 Massaro F., Giroletti M., D’Abrusco R., Masetti N., Paggi A., Cowperthwaite Philip S., Tosti G., Funk S., 1984, ApJS, 2014, 213, 3 Matsumura N. et al., 2007, AJ, 133, 1441 Matsumura N. et al., 2009, AJ, 138, 787 McLaughlin M. A. et al., 2006, Nature, 439, 817 Metzger B. D., Williams P. K. G., Berger E., 2015, ApJ, 806, 224 Mooley K. P. et al., 2016, ApJ, 818, 105 Murdin P., Webster B. L., 1971, Nature, 233, 110 Murphy T. et al., 2017, MNRAS, 466, 1944 Niinuma K. et al., 2007, ApJ, 657, L37 Osten R. A., Hawley S. L., Allred J. C., Johns-Krull C. M., Roark C., 2005, ApJ, 621, 398 Peng B. et al., 2000, A&A, 353, 937 Perley R. A., Butler B. J., 2013, ApJS, 204, 19 Polisensky E. et al., 2016, ApJ, 832, 60 Ravi V. et al., 2016, Science, 354, 1249 Rengelink R. B., Tang Y., de Bruyn A. G., Miley G. K., Bremer M. N., Roettgering H. J. A., Bremer M. A. R., 1997, A&AS, 124, 259 Richards M. T., Waltman E. B., Ghigo F. D., Richards D. St. P., 2003, ApJS, 147, 337 Rickett B. J., 1990, ARA&A, 28, 561 Roeser S., Bastian U., 1988, A&AS, 74, 449R Rowlinson A., 2016, MNRAS, 458, 3506

c 2017 RAS, MNRAS 000, 1–9