iPTF Discovery of the Rapid "Turn On" of a Luminous Quasar
S. Gezari, T. Hung, S.B. Cenko, N. Blagorodnova, Lin Yan, S.R. Kulkarni, K. Mooley, A.K.H. Kong, T.M. Cantwell, P.C. Yu, Y. Cao, C. Fremling, J.D. Neill, C.C. Ngeow, P.E. Nugent, P. Wozniak
aa r X i v : . [ a s t r o - ph . H E ] D ec Accepted for Publication in ApJ
Preprint typeset using L A TEX style emulateapj v. 5/2/11 iPTF DISCOVERY OF THE RAPID ”TURN ON” OF A LUMINOUS QUASAR
S. Gezari , T. Hung , S. B. Cenko , N. Blagorodnova , Lin Yan , S. R. Kulkarni , K. Mooley ,A. K. H. Kong , T. M. Cantwell , P. C. Yu , Y. Cao , C. Fremling , J. D. Neill , C.-C. Ngeow ,P. E. Nugent , and P. Wozniak Accepted for Publication in ApJ
ABSTRACTWe present a radio-quiet quasar at z = 0 .
237 discovered “turning on” by the intermediate PalomarTransient Factory (iPTF). The transient, iPTF 16bco, was detected by iPTF in the nucleus of a galaxywith an archival SDSS spectrum with weak narrow-line emission characteristic of a low-ionizationemission line region (LINER). Our follow-up spectra show the dramatic appearance of broad Balmerlines and a power-law continuum characteristic of a luminous ( L bol ≈ ergs s − ) type 1 quasar12 years later. Our photometric monitoring with PTF from 2009-2012, and serendipitous X-rayobservations from the XMM-Newton Slew Survey in 2011 and 2015, constrain the change of stateto have occurred less than 500 days before the iPTF detection. An enhanced broad H α to [O III] λ >
10 increase in Eddingtonratio inferred from the brightening in UV and X-ray continuum flux is more likely due to an intrinsicchange in the accretion rate of a pre-existing accretion disk, than an external mechanism such asvariable obscuration, microlensing, or the tidal disruption of a star. However, further monitoring willbe helpful in better constraining the mechanism driving this change of state. The rapid “turn on” ofthe quasar is much shorter than the viscous infall timescale of an accretion disk, and requires a diskinstability that can develop around a ∼ M ⊙ black hole on timescales less than a year. Subject headings: galaxies:active – accretion, accretion disks – black hole physics – surveys INTRODUCTION
Variability is a ubiquitous property of quasars ontimescales of hours to years, and likely attributed to pro-cesses in the accretion disk fueling the central supermas-sive black hole (SMBH) (Pereyra et al. 2006; Kelly et al.2009). More dramatic changes in accretion activity arealso expected to occur on much longer timescales. Hy- Department of Astronomy, University of Maryland, StadiumDrive, College Park, MD 20742-2421, USA [email protected] Joint Space-Science Institute, University of Maryland, Col-lege Park, MD 20742, USA NASA Goddard Space Flight Center, Mail Code 661, Green-belt, MD 20771, USA Department of Astronomy, California Institute of Technol-ogy, Pasadena, CA 91125, USA Caltech Optical Observatories, Cahill Center for Astronomyand Astrophysics, California Institute of Technology, Pasadena,CA 91125, USA Infrared Processing and Analysis Center, California Instituteof Technology, Pasadena, CA 91125, USA Astrophysics, Department of Physics, University of Oxford,Keble Road, Oxford OX1 3RH, UK Institute of Astronomy, National Tsing Hua University,Hsinchu 30013, Taiwan Jodrell Bank Centre for Astrophysics, Alan Turing Building,Oxford Road, Manchester M13 9PL, UK Graduate Institute of Astronomy, National Central Univer-sity, Taoyuan City 32001, Taiwan eScience Institute and Astronomy Department, Universityof Washington, Seattle, WA 98195, USA Department of Astronomy, The Oskar Klein Center, Stock-holm University, AlbaNova, 10691, Stockholm, Sweden Department of Astronomy, University of California, Berke-ley, CA 94720-3411, USA Lawrence Berkeley National Laboratory, 1 Cyclotron Road,MS 50B-4206, Berkeley, CA 94720, USA Los Alamos National Laboratory, MS D436, Los Alamos,NM, 87545, USA drodynamic simulations of quasar fueling and feedbackreveal a “duty cycle” of accretion activity on timescalesof a Myr (Novak et al. 2011), and may explain the lackof clear observational evidence for causal connections be-tween AGN activity and star-formation in galaxy studies(Hickox et al. 2014). Observations of ionization nebulaeon the outskirts of galaxies have indicated the shut offof quasar engines on timescales of tens of thousands ofyears (e.g., Schawinski et al. 2010), and our own Galacticcenter shows signs of X-ray reflection from enhanced nu-clear activity on a timescale of only hundreds of years ago(Ponti et al. 2010). However, as we explore the behaviorof active galactic nuclei (AGN) more systematically inthe time domain with optical imaging and spectroscopicsurveys, we are finding evidence for significant accretionstate changes on even shorter timescales.In the AGN unification model, type 1 (spectra withnarrow and broad lines) and type 2 (spectra with narrowlines only) classifications are explained as a viewing angleeffect due to nuclear obscuration of the broad-line regionand accretion disk continuum (Antonucci 1993). A chal-lenge to this paradigm is the rare class of “changing-look”AGN, who change their spectral class with the appear-ance and/or disappearance of broad Balmer lines, ac-companied by large-amplitude changes in the AGN con-tinuum. Most of the changing-look AGN cases reportedto date have been spectroscopically known Seyferts whohave shown a dramatic appearance or disappearance oftheir broad Balmer lines in follow-up spectra resulting ina “change of state” between a type 1 and a type 1.8-2spectrum. A type 1.8 or 1.9 Seyfert classification de-pends on the presence of weak broad H β or broad H α ,respectively (Osterbrock 1981). Gezari et al.Notable examples of changing-look AGN include theappearance of broad, double-peaked H α and H β linesin LINER galaxy NGC 1097 (Storchi-Bergmann et al.1993), and the complete disappearance of the broad H β line in Seyfert 1 galaxy Mrk 590 (Denney et al. 2014).More recently the ASAS-SN optical time domain surveydiscovered an outburst from NGC 2617 at z = 0 . ∼
10 yr (Shappee et al. 2014). And recent follow-up ob-servations of the changing-look AGN Mrk 1018, whichhad changed from a type 1.9 to type 1 Seyfert in 1984,revealed a change back to a type 1.9 Seyfert 30 yearslater (McElroy et al. 2016). The fact that the vari-able UV/optical continuum in Mrk 1018 followed the L ∼ T relation expected for thermal emission from adisk, and that there was no evidence for neutral hydro-gen absorption in its X-ray spectrum, favored intrinsicchanges in the accretion flow instead of an obscurationevent (Husemann et al. 2016), which has been inferredto be the cause for rapid drops in the UV continuum(Guo et al. 2016) and X-ray flux (Risaliti et al. 2009;Marchese et al. 2012) in some AGN.The first case of a changing-look AGN with the lumi-nosity of a quasar (which we define as L bol > ergss − ) was SDSS J0159+0033 at z = 0 .
312 (LaMassa et al.2015), which was discovered to change from a type 1 totype 1.9 spectrum between its SDSS DR1 spectrum in2001 and its SDSS-III BOSS spectrum in 2010. Sincethen, there was a systematic search by Ruan et al. (2016)of SDSS quasars with multiple epochs of spectra forwhich the spectroscopic pipeline classification changedbetween a ”QSO” to a ”GALAXY”, or vice versa, thatrecovered SDSS J0159+0033, and revealed two morecases of changing-look quasars (at z = 0 .
198 and z =0 . β line on a timescale of 5 − and large amplitudes ofvariability (∆ g > z = 0 . − .
6, 5 of which show the appear-ance of broad H β between the SDSS and BOSS spectralepochs. New epochs of spectra from the Time-DomainSpectroscopic Survey (TDSS; Morganson et al. (2015))have also revealed one new case of a transition froma type 1 to type 1.9 quasar at z=0.246 (Runnoe et al.2016).Here we report the rapid ( < z = 0 .
237 with weaknarrow-line emission in its pre-event spectrum character-istic of a LINER nucleus that does not require an AGNto power the line emission. Throughout this paper, werefer to quasars as radio quiet or radio loud AGN abovea bolometric luminosity of 10 ergs s − . We also adopta cosmology where H = 70 km s − and Ω M = 0 . Λ = 0 .
7, yielding a luminosity distance for iPTF 16bcoof d L = 1186 Mpc. OBSERVATIONS
SDSS Archival Imaging and Spectroscopy
The source SDSS J155440.25+362952.0 was imaged bythe SDSS survey on UT 2003 April 29 (all days hereafter are in the UT system), and morphologically classified asa galaxy with r = 18 . ± .
01 mag. It was targeted inthe SDSS spectroscopic legacy survey as a ugri -selectedquasar, selected for lying more than 4 σ from the stel-lar locus (Richards et al. 2002) at high Galactic latitude(QSO CAP). It turns out that while the colors measuredin the SDSS survey are outside the stellar locus, with u − g = +0 . ± .
04 mag, g − r = +0 . ± .
01 mag, and r − i = +0 . ± .
01 mag, given the known redshift ofthe galaxy, they are inconsistent with the quasar color-redshift relation measured for the SDSS spectroscopicsample (Schneider et al. 2007).The SDSS legacy spectrum, obtained on 2004 June16, was determined by the spectroscopic pipeline tohave a z = 0 . σ ⋆ = 176 ±
14 km/s. The galaxy classification was duethe presence of strong galaxy absorption features (CaH&K, G band, Mg I, Na D), with only weak [O III]and [N II] emission lines detected. According to the M BH − σ ⋆ scaling relation, this velocity dispersion corre-sponds to a central black hole mass of (1 +2 − . ) × M ⊙ (McConnell & Ma 2013).We find no evidence for significant flux variationsbetween the SDSS survey image in 2003 and legacyspectrum in 2004. The synthetic r -band magnitudeof the 2004 SDSS spectrum (measured by projectingthe best-fit spectral template onto the r -band filter) is spectroSynFlux r = 18 . ± .
02 mag. The correspond-ing fiber magnitude for the SDSS imaging in 2003 (mea-sured with an aperture equal to the 3 ′′ diameter spectro-scopic fiber) is r = 18 .
95 mag. Since the SDSS spectrumin 2004 is dominated by host galaxy starlight in the wave-length range of the r -band ( λ eff = 6231 ˚A), we concludethat the SDSS image in 2003 is also dominated by hostgalaxy starlight. GALEX
Archival Imaging
The source was observed by the
GALEX
All-Sky Imag-ing Survey on 2004 May 15 with an 6 ′′ (4 pixel) radiusaperture magnitude corrected for the total energy en-closed (Morrissey et al. 2007) of N U V = 21 . ± . σ point-source upper limit of F U V > . t exp = 112 s, and a background of 2 . × − cts s − pix − . Note that the color measured by GALEX and SDSS of
N U V − r = +3 . ± .
35 mag is entirely con-sistent with normal galaxies with a similar luminosity onthe blue sequence (Wyder et al. 2007). iPTF Detection iPTF 16bco was discovered as a transient detection bythe Palomar 48-in telescope (P48) on 2016 Jun 1 dur-ing an (intermediate) Palomar Transient Factory (iPTF) g + r band experiment by the real-time difference imagingpipeline run at LBNL (Cao et al. 2016), with g = 19 . r = 19 . .
44 arcsecfrom its host galaxy; within our centroiding accuracy of0 . Follow-up Spectroscopy
PTF 16bco 3On the next day after the discovery (2016 Jun 2), thetransient iPTF 16bco was followed-up with the roboticlow-resolution (R = λ/ ∆ λ ∼ . ′′ ∼ . ′′ ∼ z = 0 . Palomar 60 inch Imaging
We also monitored the source in 3 filters ( g, r, i ) withthe SEDM on the P60 telescope. The data were host-subtracted using
FPipe (Fremling et al. 2016). Oneepoch of the P60 data points on MJD 57576 are fromthe GRBCam. We do not plot the i -band GRBCamdata from this night due to the large difference in theshape of its filter transmission curve in this band. Thelight curve of the transient iPTF 16bco is presented inFigure 2, and the photometry is given in Table 1. We ad-just the g -band P48 photometry by +0.25 mag in orderto match the P60 photometry. This offset is attributedto the difference in filter curves in the g -band, and thestrong blue continuum. Note that since its discovery byiPTF on 2016 Jun 1 in the g and r bands, iPTF 16bcohas retained a blue color, g − r ≈ − . ∼ ∼ PTF Historical Light Curve
The source was observed in the PTF survey in 2009 −
151 epochs of r < . r < .
10 mag, or a ∆ r < .
08 mag during this timeperiod. The P48 observations constrain the onset of thenuclear transient to be after the last non-detection on2012 May 28, 4 years before the iPTF discovery.
Swift Observations
The source was observed with our Swift Key Projectprogram for UV follow-up of iPTF nuclear transients (PI:Gezari) on 2016 June 21 and July 5. We extracted thesource from a 5 . ′′ . ′′ uvotsource in HEASoft whichincludes a correction for the enclosed energy in the aper-ture. The source was observed in the uvw ∼ . ± .
05 magand 19 . ± .
05 mag in the AB system, respectively.The corresponding UV-optical color of iPTF 16bco (witha negligible contribution of UV flux from the host) is
N U V − r ∼ ‘ − . N U V − r colors of low-redshift quasars measured by GALEX andSDSS (
N U V − r = 0 . − . Swift
XRT observations were processedwith the UK Swift Data Science Centre pipeline thattakes into account dead columns and vignetting to ex-tract counts from the source in the energy range of 0.3-10 keV. The X-ray count rate on June 21 and July 5is 0 . ± . . ± . − , respec-tively. We further obtained a 1.7 ks Swift
XRT expo-sure on 2016 October 21 and the source was detectedwith 0 . ± .
004 counts s − . This confirms the lackof significant X-ray variability between the Swift obser-vations. Furthermore, the combined spectrum of all thedata can be modeled with an absorbed power-law witha spectral index of Γ = 2 . ± . N H fixed at theGalactic value of 1 . × cm − (Dickey & Lockman1990). The average unabsorbed 0.2-10 keV flux of thesource is 9 × − ergs s − cm − , corresponding to a lu-minosity of 1 . × ergs s − by assuming an absorbedpower-law with N H = 1 . × cm − and Γ = 2 . z = 0 . Archival X-ray Observations
The ROSAT upper limit in the 0.1-2.4 keV band fromAll-Sky Survey in 1990-1991 (Voges et al. 1999) is 0.1 ctss − which corresponds to an unabsorbed flux of ∼ × − ergs s − cm − . There are even more constraining2 σ upper limits from the XMM Slew Survey of < . < .
817 cts s − in the 0 . − . < . × − and < . × − ergs s − cm − , respectively. The latestXMM Slew Survey upper limit implies a factor of > Swift
XRT detection on a timescaleof < . Radio Observations
We observed iPTF 16bco with the AMI-LA at 15.5GHz on 2016 Oct 16.63. The source is not detected, http://xmm.esac.esa.int/UpperLimitsServer/ Gezari et al.
Fig. 1.—
Series of follow-up spectra of iPTF 16bco. Spectra are normalized, and have been offset vertically for clarity. Dotted lines showthe wavelengths of broad Balmer lines at z = 0 . ∼ with the 3 σ upper limit of 68 µ Jy. We can also convertthis to a 1.4 GHz upper limit of 370 µ Jy using a spectralindex of − .
7. This is consistent with the non-detectionin the VLA FIRST survey (Becker et al. 1995) from 1999which gives an independent 1.4 GHz upper limit of 500 µ Jy. ANALYSIS
Host Galaxy Classification
The archival SDSS spectrum from 2004 was fittedby the automated spectroscopic pipeline (Bolton et al.2012) with a combination of stellar, galaxy, and quasartemplates plus emission lines. After visual inspection ofthe pipeline fit, we found a poor fit to the the H α +[N II]and [O III] emission line complexes, and refitted the spec-trum with host galaxy template and emission-line gascomponents using ppxf (Cappellari & Emsellem 2004;Cappellari 2016) which uses the MILES stellar tem-plate library (Vazdekis et al. 2010). The emission linefits are shown in Figure 3, and are fitted with a nar-row Gaussian with a σ = 420 km s − , with no evi-dence for a broad H α or H β components. The narrow-line ratios of log([O III] λ β ) = 0 . ± .
13 andlog([N II] λ / H α ) = 0 . ± .
07, together with L ([OIII] λ . ± . × ergs s − , classify theSDSS spectrum as a type 2 AGN in the LINER region(Kewley et al. 2006) in the diagnostic narrow-line dia-grams (Baldwin et al. 1981; Veilleux & Osterbrock 1987;Kauffmann et al. 2003) shown in Figure 4. Note thatthe archival WISE colors from the all-sky survey in 2010(Cutri et al. 2011) of W1-W2 = 0.48 ± .
04 mag and W2- W3 = 1.6 ± . µ m, respectively, place the host in the regionof Seyfert and star-forming galaxies (Yan et al. 2013).However, in the WHAN diagram (shown in Figure 4) foremission-line galaxies (Cid Fernandes et al. 2010, 2011),the weak equivalent-width of H α ( W H α = 1 . ± . not an AGN. Change of State
The transient iPTF 16bco shows two remarkablechanges: a factor of 10 increase in UV flux, and a trans-formation from a LINER galaxy to a luminous type 1quasar. Figure 5 shows the dramatic change of state be-tween the SDSS spectrum in 2004 and the follow-up Keckspectrum in 2016.
Continuum Variability
According to the empirical correlation between hardX-ray emission and [O III] luminosity for AGNs(Heckman et al. 2005; Ueda et al. 2015), from the [OIII] λ L (2 − ∼ ergs s − . This could be even lower if the narrow line emis-sion in iPTF 16bco is powered by stars, and infers a pre-event X-ray luminosity at least an order of magnitude be-low the X-ray luminosity of 1 . × ergs s − measuredby Swift XRT during the type 1 state of iPTF 16bco in2016. The change in N U V flux between the
GALEX
AISPTF 16bco 5
Fig. 2.—
Light curve of iPTF 16bco. Optical g , r , and i band difference-imaging photometry are from the Palomar 48-in (solid-circles)and 60-in telescopes (solid-squares: SEDM, open squares: GBMCam), while ultraviolet aperture photometry in the uvw Swift telescope (stars), with a negligible contribution from host galaxy light. Dashed red line shows the mean r -band upper limit measuredduring PTF observations between 2009 − Fig. 3.— H β (left) and H α (right) regions of iPTF 16bco during its pre-event spectrum from SDSS. The red line shows our galaxytemplate fit, and blue line shows the emission line component fit. The residual from the galaxy template fit is also shown, along with theemission line component fit. The individual Gaussian components of the [NII]+H α complex are plotted with dashed lines. The emissionlines are all fitted with a narrow Gaussian with a σ = 420 km s − , with no evidence for a broad H α or H β line. measurement in 2004 and the Swift
UVOT measurementin 2016 also indicates a brightening by ∆ m = − . ± . N U V flux in the
GALEX measurement to be from star-formation in thehost galaxy, this yields a lower-limit of a factor of ∼ r -band magni-tude measured by SDSS (see § . . lower limit to the true amplitude of the continuum increase inthe optical.Thus we conclude that the continuum in iPTF 16bcohad an increase in X-ray/UV/optical flux by a factor of >
10 between 2004 and 2016. It is not entirely surprisingthat such a large increase in the photoionizing continuumwas also accompanied by dramatic spectral changes.
Spectral Variability
Gezari et al.
Fig. 4.—
Diagnostic narrow-line ratio diagrams for iPTF 16bco during its pre-event spectrum from SDSS.
Left:
The BPT diagramdiagram with the lines demarcating star-forming galaxies from AGN. The solid line is the theoretical curve from Kewley et al. (2001), anddashed line is the empirical curve based on the SDSS spectroscopic sample from Kauffmann et al. (2003). The line demarcating Seyfertsfrom LINER galaxies from Cid Fernandes et al. (2010) is plotted with a dotted line.
Right:
The diagnostic WHAN diagram defined byCid Fernandes et al. (2010) with the regions demarcating star-forming galaxies, Seyferts, weak AGN (wAGN) and “fake AGN” poweredby stars from Cid Fernandes et al. (2011).
PTF 16bco 7The spectra of iPTF 16bco demonstrate a strong bluecontinuum and broad Balmer-line, Fe II, and He I fea-tures characteristic of a type 1 quasar. Figure 5 alsoshows the difference between the Keck spectrum in 2016and the archival SDSS spectrum from 2004. After cor-recting for Galactic extinction of E(B − V) = 0.021 magfrom the Schlegel et al. (1998) dust extinction map andusing the extinction curve of Cardelli et al. (1989), thecontinuum in the difference spectrum (after masking thebroad emission lines) is reasonably fitted with a singlepower law of f λ ∝ λ α λ , where α λ = − .
45. This isshallower than the standard theoretical thin accretiondisc spectrum with f ν ∝ ν α ν , where α ν = 1 /
3, and α λ = − ( α ν + 2) = − .
33, which is well fitted to differ-ence spectra (Wilhite et al. 2005; MacLeod et al. 2016)and difference spectral energy distributions (Hung et al.2016) of quasars. However, it is close to the power-lawobserved in averaged optical quasar spectra ( α λ = − . β (Vanden Berk et al. 2001; Wilhite et al.2005). A power-law index similar to the average quasarspectrum in the difference spectrum is another indicationthat the low-state spectrum has very little contributionfrom a non-stellar continuum.The [O III] λ . ± . × ergs s − , consistent within the errors of the [O III] lu-minosity measured in SDSS. The broad H α flux, incontrast, dramatically appears, with L (H α, broad) =(8 . ± . × ergs s − with a full-width at half-maximum (FWHM) of 4048 ±
36 km s − . This differencein behavior between the broad and narrow lines can beexplained by the fact the [O III] line luminosity traces theaverage AGN continuum over a much longer timescalethat the flaring event detected by iPTF. While the emis-sivity decay time (dominated by recombination chargetransfer) for [O III] λ < ∼ λL λ (5100˚A) = λf λ πd L (1 + z ) = 1 . × − ergs s − ˚A − πd L (1 + z ) = 1 . × erg s − . This is in excellent agreement with the nearlylinear correlation between broad H α luminosity and opti-cal continuum luminosity in AGN (Greene & Ho 2005),which for this broad H α luminosity, one would expect λL λ (5100 ˚A) ∼ . × ergs s − .The FWHM of the broad H β line, 4770 ±
200 kms − , and the monochromatic luminosity at 5100 ˚A canbe used to estimate the central black hole mass to be2 +4 . − . × M ⊙ (Vestergaard & Peterson 2006), in goodagreement with M BH inferred from the host galaxy stel-lar velocity dispersion. We adopt a bolometric correc-tion factor for the monochromatic optical luminosity of8.1 (Runnoe et al. 2012) to get L bol = 1 . × ergss − . We then derive an Eddington ratio of λ Edd = L bol /L Edd = 0 .
05 during the type 1 quasar state.
Nature of the Variability
Variable Obscuration or Microlensing
The Eddington ratio inferred for iPTF 16fnl from itsnearly constant narrow [O III] λ λ Edd ∼ . α line in iPTF 16bco. In the case of an obscuringcloud, the distance between the nucleus and the cloudmust be larger than the radius of the broad-line region itis obscuring. Using the radius-luminosity relation mea-sured for AGN from reverberation mapping studies ofbroad lines (Bentz et al. 2013), the luminosity of iPTF16bco would have H β broad-line emission with a charac-teristic radius of R BLR ∼
45 days. Following the argu-ment of LaMassa et al. (2015) (Equation 4), this trans-lates to a crossing time on a circular, Keplerian orbit, t cross = ∆ φ/ω K , where ∆ φ =arcsin (cid:16) r src r orb (cid:17) is the angularlength of the arc, and ω K = √ GM π r − / is the Keplerianfrequency, which for r orb > R BLR yields t cross > Tidal Disruption Event
One mechanism to rapidly increase the mass accre-tion rate onto a SMBH is to get a new supply of gasfrom a star that wanders close enough to the SMBHto be torn apart by tidal forces. In a tidal disruptionevent (TDE), roughly half of the disrupted stellar de-bris remains bound to the black hole, falls back ontothe SMBH, circularizes through shocks, and is accreted(Rees 1988). The characteristic timescale of a TDEis the orbital period of the most tightly bound debris,which is given by ∆ t = 0 . M / m − ⋆ r / ⋆ yr, where M = M BH / M ⊙ , m ⋆ = M ⋆ /M ⊙ and r ⋆ = R ⋆ /R ⊙ (Lodato & Rossi 2011). The peak mass accretion rate isgiven by ˙ M acc = (1 / M ⋆ / ∆ t ), and can exceed the Ed-dington rate ( ˙ M Edd = 0 . M ( η/ . − M ⊙ yr − , where η is the radiative efficiency) for black holes < M ⊙ .Following the peak, a TDE has a characteristic t − / power-law decay determined by the fallback rate of thestellar debris to pericenter (Rees 1988; Phinney 1989;Evans & Kochanek 1989).However, we point out several issues with interpretingthe flaring state of iPTF 16bco with a TDE. 1) For a M BH > ∼ M ⊙ , the tidal disruption radius of a solar-type star ( R T = R ⋆ ( M BH /M ⋆ ) / = 3 . × M / cm) is smaller than the Schwarzschild radius ( R S =2 GM BH /c = 2 . × M cm), and the star crossesthe event horizon before being disrupted. 2) The X-raypower-law (Γ = 2 .
1) continuum of iPTF-16bco is unlikethe extremely soft, thermal X-ray spectra observed inTDEs (Komossa 2002; Miller et al. 2015). 3) The light Gezari et al.
Fig. 5.—
Left : Dramatic change in spectrum between the archival SDSS legacy spectrum obtained on 2004 June 16, and the follow-upspectrum obtained by Keck2+DEIMOS on 2016 June 4.
Right : Difference spectrum corrected for Galactic extinction. Tick marks showthe broad Balmer lines (H α , H β , and H δ ) as well as the broad Fe II complexes, and He I λ f λ ∝ λ α λ , with α λ = − . curve of iPTF 16bco shows a complex shape uncharacter-istic of a TDE, with month-long plateau, followed by a 2week rise to another plateau. 4) The broad Balmer linesin iPTF 16bco are narrower and stronger than have beenobserved in TDEs, and iPTF 16bco does not have strongbroad He II λ β line vs. the ratio of the equivalent widths ofthe Fe II λ β strength ( R FeII )(Sulentic et al. 2000), favors a change in ˙ M acc of a pre-existing accretion disk, instead of a newly formed debrisdisk from a TDE. Continued photometric monitoring candetermine if the light curve of iPTF 16bco eventuallyevolves into a power-law decline as expected for a TDE. Accretion Disk Instabilities
We now investigate the scenario that iPTF 16bco wasthe result of a change of state in a pre-existing quasaraccretion disk. Interestingly however, iPTF 16bco putsstringent limits on the timescale over which such accre-tion rate changes must occur. In the rest-frame, iPTF16bco demonstrates a dramatic change in continuum fluxover a timescale of ∆ t < / (1 + z ) = 3 .
23 yr or in < . t infl . Furthermore, this timescale is ex-pected to be longer for a quasar “turning on” insteadof “turning off”, since it scales as λ − (LaMassa et al.2015). The t infl corresponding to the Eddington ratioand black hole mass estimated for iPTF 16bco in its dimstate, and assuming a radiation-pressure dominated in-ner region of a Shakura-Sunyaev disk, is, t infl = 1300yr h α . i − (cid:20) λ Edd . (cid:21) − h η . i (cid:20) r r g (cid:21) / (cid:20) M . (cid:21) , a much longer timescale than the observed rapid changein continuum flux in iPTF 16bco.One possibility could be an accretion disk eruption asthe result of thermal-viscous instabilities in a partial ion- ization zone, analogous to the outbursts observed in cat-aclysmic variables and X-ray novae (Siemiginowska et al.1996). However, while such instabilities can produceamplitudes of a factor of 10 , the expected durationsscale with the central mass, and so while cataclysmicvariables (CVs) and X-ray novae show large-amplitudeoutbursts on the timescale of weeks to months, for anaccretion disk around a 10 M ⊙ black hole, this corre-sponds to timescales of ∼ yr. However, state changeson ∼ minute timescales have been observed in some X-ray binaries (Fender 2001; Fender & Belloni 2004), whichwould imply similar changes in a SMBH on the timescaleof only ∼
10 yr.In the accretion-disk instability model, the more nar-row the unstable zone, the more rapid the timescaleof variability. The size of the unstable zone dependson the accretion rate: the lower the accretion rate,the closer in the unstable region is to the inner edgeof the disk. However, the smaller the unstable zone,the smaller the expected amplitude of variability. Formodels that demonstrate a factor of ∼
10 variability( α = 0 . , ˙ M = 3 × − M ⊙ yr − ) in Siemiginowska et al.(1996), they have a “turn-on” timescale of ∼ yr. Thisis still 3 orders of magnitude longer than we require forthe “turn-on” timescale of iPTF 16bco.Interestingly, the thermal timescale itself, t th ∼ /α Ω K , is much shorter, adopting Equation 8 from(Siemiginowska et al. 1996): t th ∼ . α/ . − M − . ( r/ cm) . yr . A disk with local thermal fluctuations, potentially drivenby the magneto-rotational instability, could be consis-tent with the rapid timescale of the continuum variabil-ity in iPTF 16bco (Dexter & Agol 2011). While an in-homogeneous disk has been demonstrated to fit com-posite difference spectra (Ruan et al. 2014) and colorvariability of quasars (Schmidt et al. 2012), Hung et al.(2016) find a simple disk model adequate to fit difference-flux UV/optical spectral energy distributions of individ-ual quasars. Moreover, Kokubo (2015) argue that thetight inter-band correlations observed in SDSS quasarlight curves are inconsistent with the inhomogeneous diskmodel. However, whether or not these thermal fluc-tuations can be coherent enough to produce a large-PTF 16bco 9amplitude outburst, is still to be determined.An instability that arises in a radiation-pressure domi-nated disk (Lightman & Eardley 1974) has been used tomodel recurrent flares in accreting Galactic X-ray bina-ries (Belloni et al. 1997) and has recently been applied bySaxton et al. (2015) and Grupe et al. (2015) to explainthe large-amplitude soft X-ray flares in Seyfert galaxiesNGC 3599 and IC 3599, respectively. In this scenario,when the internal radiation pressure of the disk becomegreater than the gas pressure, a heating wave propagatesthrough the disk. This results in a enhanced local viscos-ity, scale height, and accretion rate, which rapidly drainsthe disk. The instability recurs when the inner disk fillsback in. The rise-time and recurrence timescale can beas short as a year to hundreds of years for a 10 M ⊙ blackhole, respectively.Another mechanism for driving large amplitude vari-ability in a quasar accretion disk could be related to thepresence of a binary SMBH. Hydrodynamical simulationsshow that in a circumbinary disk, streams penetrate thedisk cavity to feed the primary and secondary black holeat a periodic rate, and that at close to equal mass ra-tios, the perturbed circumbinary disk has an enhancedaccretion rate that can be quite bursty on a timescaleof ∼ ∼ . ∼ . − . α in-stability models of Siemiginowska et al. (1996) could bepromising. A potential testable prediction, is that inthese models, the accretion disks spend the majority oftheir time in the low-state. The variability during theenhanced accretion state in iPTF 16bco could also be a signature of clumpy accretion in an advection-dominatedaccretion flow (ADAF), for which cold clumps form inthe accretion flow due to instabilities in the radiation-dominated regions of the disk (Wang et al. 2012) Disk-Jet Connection
The dramatic change in accretion rate from λ Edd < ∼ .
005 to λ Edd ∼ .
05 inferred for iPTF16bco could be accompanied by a structural changein the accretion flow if the quasar accretion disk istransitioning from a radiatively inefficient to radiativelyefficient mode. Note that such changes in X-ray binariesare often accompanied by changes in jet activity. Theimplied high-state radio-to-optical flux density ratiofor iPTF 16bco of R = log( S . /S opt ) < . L R < × W Hz − , is typical ofradio-quiet AGN (e.g., Padovani et al. 2011). Whilethese values are consistent with the LINER classificationfrom the optical spectrum, it is surprising that if theaccretion event in iPTF 16bco were triggered by a diskinstability, that there is no evidence for a jet or outflowduring its high-state in the radio. This is in contrastwith X-ray binaries and CVs, which generally showflaring at radio, optical and X-ray wavelengths alongsidestrong Balmer emission lines. Similarly, the fundamentalplane of black hole activity, e.g. (Plotkin et al. 2012;Saikia et al. 2015) predicts L R significantly greater than ∼ W Hz − when L X = 1 . × erg s − . Comparison to Other Changing-Look Quasars iPTF 16bco is one of only a dozen other changing-lookquasars (here we define as M i < −
22 mag, L ([O III]) > ergs s − ), roughly half of which have been caught inthe act of “turning on” by demonstrating the sudden ap-pearance of broad-lines. Figure 6 shows the redshift and[O III] luminosity of all the changing-look quasars in theliterature that pass our [O III] luminosity cut (thus weexclude SDSS J0126-0839 and SDSS J2336+0017 fromRuan et al. (2016)), color-coded by whether they showappearing broad lines, or disappearing broad lines. Wealso do not include three changing-look quasars from theMacLeod et al. (2016) sample that do not have good cov-erage of the broad H α line in its high state (appearingSDSS J214613 at z = 0 .
62, disappearing SDSS J022562at z = 0 .
63, and both appearing and disappearing SDSSJ022556 at z = 0 . α line in the BOSS spectrum is possibleat higher-redshift than finding a disappearing broad H α line in the SDSS spectrum.Figure 6 also shows the [O III] λ α lu-minosity in all the changing-look quasars in their type1 state, in comparison to the full SDSS quasar sam-ple. We determine the [O III] λ SpecLine ta-ble and the DR12 interactive spectrum line measurementtable. We use the broad H α fluxes from the Shen et al.(2011) catalog for broad-line quasars, or from the liter-ature when available. All luminosities are calculated for0 Gezari et al. Fig. 6.—
Comparison to other changing-look quasars that have disappearing (dots) and appearing (circled dots) broad-line emission.
Left : [O III] λ Right : Broad H α luminosity during the high-state of thequasar vs. the [O III] λ α )vs. L([O III]) ratio for the DR7 SDSS quasar sample from Shen et al. (2011). iPTF 16bco is an outlier of this distribution, with a highbroad H α luminosity relative to its [O III] λ our adopted cosmology. Note that iPTF 16bco is on theedge of the normal quasar distribution, while the otherchanging-look quasars reported in the literature appearto lie squarely in the distribution of normal quasars inthis parameter space.The enhanced broad H α luminosity observed in iPTF16bco relative to [O III] in comparison to normalquasars, as well as the previously discovered changing-look quasars, is likely a signature of its rapid transitionto a type 1 state. As discussed in § ∼ α in iPTF 16bcocompared to these objects, this would imply an evenshorter “turn-on” timescale, in agreement with the in-ferred “turn-on” timescale for the continuum in iPTF16bco of < ∼ SUMMARY
We present the rapid “turn on” of a luminousbroad-line quasar at z = 0 .
237 discovered from itsnuclear optical variability in the iPTF survey (iPTF16bco), and identified as a newly emerged quasar fromcomparison of follow-up spectroscopy with an archivalSDSS spectrum from over a decade earlier which showsLINER narrow-line emission potentially powered bystars. Pre-event optical, UV, and X-ray imaging indicatethat the quasar continuum increased by a factor of > < ∼ λ α to narrow [O III] λ REFERENCESAg¨ueros, M. A., et al. 2005, AJ, 130, 1022Antonucci, R. 1993, ARA&A, 31, 473Arcavi, I., et al. 2014, ApJ, 793, 38Baldwin, J. A., Phillips, M. M., & Terlevich, R. 1981, PASP, 93, 5Becker, R. H., White, R. L., & Helfand, D. J. 1995, ApJ, 450, 559Belloni, T., M´endez, M., King, A. R., van der Klis, M., & vanParadijs, J. 1997, ApJ, 479, L145Bentz, M. C., et al. 2013, ApJ, 767, 149Bianchi, L., et al. 2005, ApJ, 619, L27Bolton, A. S., et al. 2012, AJ, 144, 144Cackett, E. M., G¨ultekin, K., Bentz, M. C., Fausnaugh, M. M.,Peterson, B. M., Troyer, J., & Vestergaard, M. 2015, ApJ, 810,86Cao, Y., Nugent, P. E., & Kasliwal, M. M. 2016, PASP, 128,114502Cappellari, M. 2016, ArXiv e-prints, 1607.08538Cappellari, M., & Emsellem, E. 2004, PASP, 116, 138Cardelli, J. A., Clayton, G. C., & Mathis, J. S. 1989, ApJ, 345,245Cid Fernandes, R., Stasi´nska, G., Mateus, A., & Vale Asari, N.2011, MNRAS, 413, 1687Cid Fernandes, R., Stasi´nska, G., Schlickmann, M. S., Mateus, A.,Vale Asari, N., Schoenell, W., & Sodr´e, L. 2010, MNRAS, 403,1036Cooper, M. C., et al. 2012, MNRAS, 419, 3018Cutri, R. M., et al. 2011, Explanatory Supplement to the WISEPreliminary Data Release Products, Tech. rep.Denney, K. D., et al. 2014, ApJ, 796, 134Dexter, J., & Agol, E. 2011, ApJ, 727, L24Dickey, J. M., & Lockman, F. J. 1990, ARA&A, 28, 215Eracleous, M., Livio, M., & Binette, L. 1995, ApJ, 445, L1Evans, C. R., & Kochanek, C. S. 1989, ApJ, 346, L13Faber, S. M., et al. 2003, in Proc. SPIE, Vol. 4841, InstrumentDesign and Performance for Optical/Infrared Ground-basedTelescopes, ed. M. Iye & A. F. M. Moorwood, 1657–1669Farris, B. D., Duffell, P., MacFadyen, A. I., & Haiman, Z. 2014,ApJ, 783, 134Fender, R., & Belloni, T. 2004, ARA&A, 42, 317Fender, R. P. 2001, MNRAS, 322, 31Fremling, C., et al. 2016, A&A, 593, A68Gezari, S., et al. 2008, ApJ, 676, 944——. 2012, Nature, 485, 217Greene, J. E., & Ho, L. C. 2005, ApJ, 630, 122Grupe, D., Komossa, S., & Saxton, R. 2015, ApJ, 803, L28Guo, H., et al. 2016, ApJ, 826, 186Heckman, T. M., Ptak, A., Hornschemeier, A., & Kauffmann, G.2005, ApJ, 634, 161Hickox, R. C., Mullaney, J. R., Alexander, D. M., Chen, C.-T. J.,Civano, F. M., Goulding, A. D., & Hainline, K. N. 2014, ApJ,782, 9Holoien, T. W.-S., et al. 2016a, MNRAS, 463, 3813——. 2016b, MNRAS, 455, 2918——. 2014, MNRAS, 445, 3263Hung, T., et al. 2016, ArXiv e-prints, 1609.06307Husemann, B., et al. 2016, A&A, 593, L9Kauffmann, G., et al. 2003, MNRAS, 346, 1055Kelly, B. C., Bechtold, J., & Siemiginowska, A. 2009, ApJ, 698,895 Kewley, L. J., Dopita, M. A., Sutherland, R. S., Heisler, C. A., &Trevena, J. 2001, ApJ, 556, 121Kewley, L. J., Groves, B., Kauffmann, G., & Heckman, T. 2006,MNRAS, 372, 961Kokubo, M. 2015, MNRAS, 449, 94Komossa, S. 2002, in Lighthouses of the Universe: The MostLuminous Celestial Objects and Their Use for Cosmology, ed.M. Gilfanov, R. Sunyeav, & E. Churazov, 436–+Korista, K. T., & Goad, M. R. 2004, ApJ, 606, 749LaMassa, S. M., et al. 2015, ApJ, 800, 144Lawrence, A., et al. 2016, ArXiv e-prints, 1605.07842Lightman, A. P., & Eardley, D. M. 1974, ApJ, 187, L1Lodato, G., & Rossi, E. M. 2011, MNRAS, 410, 359MacLeod, C. L., et al. 2012, ApJ, 753, 106——. 2016, MNRAS, 457, 389Marchese, E., Braito, V., Della Ceca, R., Caccianiga, A., &Severgnini, P. 2012, MNRAS, 421, 1803Masci, F., et al. 2016, ArXiv e-prints, 1608.01733McConnell, N. J., & Ma, C.-P. 2013, ApJ, 764, 184McElroy, R. E., et al. 2016, A&A, 593, L8Merloni, A., et al. 2015, MNRAS, 452, 69Miller, J. M., et al. 2015, Nature, 526, 542Morganson, E., et al. 2015, ApJ, 806, 244Morrissey, P., et al. 2007, ApJS, 173, 682Newman, J. A., et al. 2013, ApJS, 208, 5Novak, G. S., Ostriker, J. P., & Ciotti, L. 2011, ApJ, 737, 26Osterbrock, D. E. 1981, ApJ, 249, 462Padovani, P., Miller, N., Kellermann, K. I., Mainieri, V., Rosati,P., & Tozzi, P. 2011, ApJ, 740, 20Pereyra, N. A., Vanden Berk, D. E., Turnshek, D. A., Hillier,D. J., Wilhite, B. C., Kron, R. G., Schneider, D. P., &Brinkmann, J. 2006, ApJ, 642, 87Phinney, E. S. 1989, in IAU Symposium, Vol. 136, The Center ofthe Galaxy, ed. M. Morris, 543–+Plotkin, R. M., Markoff, S., Kelly, B. C., K¨ording, E., &Anderson, S. F. 2012, MNRAS, 419, 267Ponti, G., Terrier, R., Goldwurm, A., Belanger, G., & Trap, G.2010, ApJ, 714, 732Rees, M. J. 1988, Nature, 333, 523Richards, G. T., et al. 2002, AJ, 123, 2945Risaliti, G., et al. 2009, ApJ, 696, 160Ruan, J. J., et al. 2016, ApJ, 826, 188Ruan, J. J., Anderson, S. F., Dexter, J., & Agol, E. 2014, ApJ,783, 105Runnoe, J. C., Brotherton, M. S., & Shang, Z. 2012, MNRAS,422, 478Runnoe, J. C., et al. 2016, MNRAS, 455, 1691Saikia, P., K¨ording, E., & Falcke, H. 2015, MNRAS, 450, 2317Saxton, R. D., Motta, S. E., Komossa, S., & Read, A. M. 2015,MNRAS, 454, 2798Schawinski, K., et al. 2010, ApJ, 724, L30Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500,525Schmidt, K. B., Rix, H.-W., Shields, J. C., Knecht, M., Hogg,D. W., Maoz, D., & Bovy, J. 2012, ApJ, 744, 147Schneider, D. P., et al. 2007, AJ, 134, 102Shappee, B. J., et al. 2014, ApJ, 788, 48Shen, Y., et al. 2011, ApJS, 194, 45
Siemiginowska, A., Czerny, B., & Kostyunin, V. 1996, ApJ, 458,491Storchi-Bergmann, T., Baldwin, J. A., & Wilson, A. S. 1993,ApJ, 410, L11Sulentic, J. W., Zwitter, T., Marziani, P., & Dultzin-Hacyan, D.2000, ApJ, 536, L5Ueda, Y., et al. 2015, ApJ, 815, 1van Velzen, S., et al. 2011, ApJ, 741, 73Vanden Berk, D. E., et al. 2001, AJ, 122, 549Vazdekis, A., S´anchez-Bl´azquez, P., Falc´on-Barroso, J., Cenarro,A. J., Beasley, M. A., Cardiel, N., Gorgas, J., & Peletier, R. F.2010, MNRAS, 404, 1639 Veilleux, S., & Osterbrock, D. E. 1987, ApJS, 63, 295Vestergaard, M., & Peterson, B. M. 2006, ApJ, 641, 689Voges, W., et al. 1999, A&A, 349, 389Wang, J.-M., Cheng, C., & Li, Y.-R. 2012, ApJ, 748, 147Wilhite, B. C., Vanden Berk, D. E., Kron, R. G., Schneider,D. P., Pereyra, N., Brunner, R. J., Richards, G. T., &Brinkmann, J. V. 2005, ApJ, 633, 638Wyder, T. K., et al. 2007, ApJS, 173, 293Yan, L., et al. 2013, AJ, 145, 55
PTF 16bco 13
TABLE 1iPTF Photometry
Telescope+Camera Filter MJD Magnitude ErrorP48+CFH12k g g g g g g g r r r g g g g g g g g g g g g g g g g g g g g g g g g g g g r r r r r r r r r r r r r r r r r r r r r r r r r r r r i i i i i i i i i i i TABLE 1 — Continued
Telescope+Camera Filter MJD Magnitude ErrorP60+SEDM i i i i i i i i i i i i i i i i g rr