iPTF16fnl: a faint and fast tidal disruption event in an E+A galaxy
N. Blagorodnova, S. Gezari, T. Hung, S. R. Kulkarni, S. B. Cenko, D. R. Pasham, L. Yan, I. Arcavi, S. Ben-Ami, B. D. Bue, T. Cantwell, Y. Cao, A. J. Castro-Tirado, R. Fender, C. Fremling, A. Gal-Yam, A. Y. Q. Ho, A. Horesh, G. Hosseinzadeh, M. M. Kasliwal, A. K. H. Kong, R. R. Laher, G. Leloudas, R. Lunnan, F. J. Masci, K. Mooley, J. D. Neill, P. Nugent, M. Powell, A. F. Valeev, P. M. Vreeswijk, R. Walters, P. Wozniak
DDraft version March 29, 2018
Preprint typeset using L A TEX style AASTeX6 v. 1.0 iPTF16fnl: A FAINT AND FAST TIDAL DISRUPTION EVENT IN AN E+A GALAXY
N. Blagorodnova , S. Gezari , T. Hung , S. R. Kulkarni , S. B. Cenko , D. R. Pasham † , L. Yan ,I. Arcavi † , S. Ben-Ami , B. D. Bue , T. Cantwell , Y. Cao , A. J. Castro-Tirado , R. Fender ,C. Fremling , A. Gal-Yam , A. Y. Q. Ho , A. Horesh , G. Hosseinzadeh , M. M. Kasliwal ,A. K. H. Kong , R. R. Laher , G. Leloudas , R. Lunnan , F. J. Masci , K. Mooley , J. D. Neill ,P. Nugent , M. Powell , A. F. Valeev , P. M. Vreeswijk , R. Walters , P. Wozniak (Dated: Received XX XX XXXX; accepted XX XX, XXXX) Cahill Center for Astrophysics, California Institute of Technology, Pasadena, CA 91125, USA; [email protected] Department of Astronomy, University of Maryland, Stadium Drive, College Park, MD 20742-2421, USA Joint Space-Science Institute, University of Maryland, College Park, MD 20742, USA NASA Goddard Space Flight Center, Mail Code 661, Greenbelt, MD 20771, USA Center for Space Research, Massachusetts Institute of Technology, Cambridge, MA 02139 Infrared Processing and Analysis Center, California Institute of Technology, Pasadena, CA 91125, USA Department of Physics, University of California, Santa Barbara, CA 93106-9530, USA Las Cumbres Observatory, 6740 Cortona Dr Ste 102, Goleta, CA 93117-5575, USA Smithsonian Astrophysical Observatory, Harvard-Smithsonian Center for Astrophysics, 60 Garden St., Cambridge, MA 02138, USA Department of Particle Physics and Astrophysics, Weizmann Institute of Science, Rehovot 7610001, Israel Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91125, USA Jodrell Bank Centre for Astrophysics, Alan Turing Building, Oxford Road, Manchester M13 9PL, UK Department of Astronomy, University of Washington, Box 351580, Seattle, WA 98195-1580, USA Instituto de Astrof´ısica de Andaluc´ıa (IAA-CSIC), P.O. Box 03004, E-18080 Granada, Spain Unidad Asociada Departamento de Ingeniera de Sistemas y Automtica, Univ. de M´alaga, Spain Astrophysics, Department of Physics, University of Oxford, Keble Road, Oxford OX1 3RH, UK Department of Astronomy, The Oskar Klein Center, Stockholm University, AlbaNova, 10691 Stockholm, Sweden Racah Institute of Physics, Hebrew University, Jerusalem, 91904, Israel Institute of Astronomy and Department of Physics, National Tsing Hua University, Hsinchu 30013, Taiwan Dark Cosmology centre, Niels Bohr Institute, University of Copenhagen, Juliane Maries vej 30, 2100 Copenhagen, Denmark Lawrence Berkeley National Laboratory, Berkeley, California 94720, USA Department of Astronomy, University of California, Berkeley, CA 94720-3411, USA Department of Physics and Yale Center for Astronomy & Astrophysics, Yale University PO Box 208120, New Haven, CT 06520-8120, USA Special Astrophysical Observatory, Nizhnij Arkhyz, Karachai-Cherkessian Republic, 369167 Russia Los Alamos National Laboratory, MS-D466, Los Alamos, NM 87545, USA † Einstein Fellow
ABSTRACTWe present ground-based and
Swift observations of iPTF16fnl, a likely tidal disruption event (TDE)discovered by the intermediate Palomar Transient Factory (iPTF) survey at 66.6 Mpc. The lightcurveof the object peaked at absolute M g = − . L p (cid:39) (1 . ± . × erg s − , an order of magnitude fainter than any other opticalTDE discovered so far. The luminosity in the first 60 days is consistent with an exponential decay, with L ∝ e − ( t − t ) /τ , where t = 57631.0 (MJD) and τ (cid:39)
15 days. The X-ray shows a marginal detectionat L X = 2 . . − . × erg s − ( Swift
X-ray Telescope). No radio counterpart was detected downto 3 σ , providing upper limits for monochromatic radio luminosity of νL ν < . × erg s − and νL ν < . × erg s − (VLA, 6.1 and 22 GHz). The blackbody temperature, obtained from combined Swift
UV and optical photometry, shows a constant value of 19,000 K. The transient spectrum at peakis characterized by broad He II and H α emission lines, with an FWHM of about 14,000 km s − and10,000 km s − respectively. He I lines are also detected at λλ ∼
650 Myr and solar metallicity. The characteristics of iPTF16fnl make it an a r X i v : . [ a s t r o - ph . H E ] M a y outlier on both luminosity and decay timescales, as compared to other optically selected TDEs. Thediscovery of such a faint optical event suggests a higher rate of tidal disruptions, as low luminosityevents may have gone unnoticed in previous searches. Keywords: accretion, accretion discs – black hole physics – stars: individual (iPTF16fnl), galaxies:nuclei INTRODUCTIONA tidal disruption event (TDE) is the phenomenonobserved when a star is torn apart by the tidal forcesof a supermassive black hole (SMBH), usually lurkingin the core of its galaxy. As a consequence, a brightflare is expected when some of the bound material ac-cretes onto the SMBH. Although such events were the-oretically predicted a few decades ago (Hills 1975; Lacyet al. 1982; Rees 1988; Evans & Kochanek 1989; Phinney1989), observational signatures are more recent. Thefirst detection of TDEs were made in the soft X-raydata. The flares, consistent with the proposed stellardisruption scenario, were identified in ROentgen SATel-lite (ROSAT) all-sky survey (Bade et al. 1996; Komossa& Bade 1999; Saxton et al. 2012). Detection in gamma-ray data of
Swift events Swift J1644+75 (Bloom et al.2011; Levan et al. 2011; Burrows et al. 2011; Zaudereret al. 2011; Cenko et al. 2012) and Swift J1112.2-8238(Brown et al. 2015) were attributed to relativistic out-bursts caused by jetted emission. We refer the readerto Komossa (2015); Auchettl et al. (2016) for a broaderreview of the status of observations in different wave-lengths.Ultraviolet detections of nuclear flares were reportedfrom the
GALEX survey (Gezari et al. 2006, 2008,2009). TDEs are now being discovered by optical sur-veys such as the Sloan Digital Sky Survey (SDSS; vanVelzen et al. 2011), PanSTARRS-1 (PS1; Gezari et al.2012; Chornock et al. 2014), Palomar Transient Factory(PTF; Arcavi et al. 2014), All Sky Automated Surveyfor Supernovae (ASAS-SN; Holoien et al. 2014, 2016b,a),Robotic Optical Transient Search Experiment (ROTSEVink´o et al. 2015), Optical Gravitational Lensing Exper-iment (OGLE; Wyrzykowski et al. 2017) and the inter-mediate Palomar Transient Factory (iPTF; Hung et al.(2017), Duggan G. et. al. in prep and the work pre-sented here). The optical sample has revealed that animportant fraction of the TDEs appear to be found inE+A (“quiescent Balmer-strong”) galaxies (Arcavi et al.2014; French et al. 2016), which can be interpreted asmiddle-aged ( < DISCOVERY AND HOST GALAXY2.1.
Discovery and classification iPTF16fnl was discovered on UT 2016 August 29.4 inan image obtained during the g +R experiment (Milleret al. 2017): during the night, one image each wasobtained in g and R band; the images are separatedby at least an hour in order to filter out asteroids.The event was identified by two real-time differenceimaging pipelines (Cao et al. 2016; Masci et al. 2017),shown in Figure 1. The discovery magnitudes were g = 17 . ± .
09 and R = 17 . ± .
09 (see Ofek et al.(2012) for photometric calibration of PTF). Given anRMS of 0.5 (cid:48)(cid:48) , the coordinates of the source, α J =00 h m s . δ J = +32 h m s . g − R = − .
28 mag) andcentral location of the transient in its host galaxy, madeit a prime candidate for prompt follow-up observation.On the night after discovery, we observed the sourcewith the FLOYDS spectrograph (Sand et al. 2011) onthe Las Cumbres Observatory (LCO; Brown et al. 2013)2-m telescope and the Spectral Energy Distribution Ma-chine (SEDM) on the Palomar 60-inch (P60) telescope.FLOYDS are a pair of robotic low-resolution (R ∼ ∼ (cid:48)(cid:48) , allows for robotic spectroscopy. Their data re-duction pipeline allow rapid reduction with minimal in-tervention from the user (aperture placing). Figure 2shows the classification spectra, displaying a blue con-tinuum and broad He II and H α emission lines, char-acteristic of previously observed optical TDE spectra(Arcavi et al. 2014). The fast spectral identification ofiPTF16fnl as a TDE candidate allowed us to rapidly in-form the astronomical community (ATel Host galaxy
The host galaxy of iPTF16fnl is Markarian 950 (Mrk950) located at z =0.016328 (Cabanela & Aldering 1998;Petrosian et al. 2007). Given the edge-on host in- RA (J2000) +32°52'40.0"53'00.0"20.0"40.0"54'00.0"20.0" D e c ( J ) PTFP48MJD 56900.3
RA (J2000)PTFP48MJD 57629.4
RA (J2000)PTFP48MJD 57629.4 g=17.11 ± Figure 1 . P48 cutouts of 2 (cid:48) diameter sky region centered in the position of the transient. The coordinates of the transientare α J = 00 h m s . δ J = +32 h m s .
48. The cutouts show the host galaxy approximately one year before thediscovery, on the day of the discovery, and the difference image showing the transient location in the core of the galaxy. Å ]0.00.20.40.60.81.0 F l u x F λ ( e r g / Å / c m / s ) + c o n s t a n t
1e 14 SEDM 900sFTN 2700s H η H δ H γ H e II H β H α H e I C a II Figure 2 . Two classification spectra obtained 2 days beforethe peak in g -band. The top black thick line shows the900 s exposure obtained with SEDM, on the Palomar 60-inch(1.5 m). The bottom blue line shows a 2700 s exposure ob-tained with the FLOYDS spectrograph on the Las CumbresObservatory (LCO) 2-m telescope. The blue continuum andprominent He II line are clearly identified in both spectra. clination and the existence of a peculiar bar and nu-cleus, the galaxy is classified as Sp ( spindle ). The lu-minosity distance is D L =66.6 Mpc (distance modulus µ = 34.12 mag) using H = 69.6 km s − Mpc − , Ω M =0.29, Ω Λ = 0.71 in the reference frame of the 3 K cosmicmicrowave background (CMB; Fixsen et al. 1996).The estimated Galactic colour excess at the positionof the transient is E (B − V) = 0 . ± .
001 mag (fromNED ), after adopting the extinction law of Fitzpatrick(1999) with corrections from Schlafly & Finkbeiner(2011). Assuming R V = 3 .
1, the Galactic visual ex-tinction is A V = 0.192 mag.The archival host magnitudes are shown in Table1. The derived K -band luminosity of the galaxy is The NASA/IPAC Extragalactic Database (NED) is operatedby the Jet Propulsion Laboratory, California Institute of Tech-nology, under contract with the National Aeronautics and SpaceAdministration.
Table 1 . Archival photometry of Mrk 950.Survey Band Magnitude Reference(mag)GALEX FUV AB ± a [1]GALEX NUV AB ± a [1]SDSSDR12 u ± b [2]SDSSDR12 g ± b [2]SDSSDR12 r ± b [2]SDSSDR12 i ± b [2]2MASS J ± H ± K ± ± ± ± > < < × − erg s − [6] a Measured within 7.5 (cid:48)(cid:48) diameter aperture. b Model mag-nitude. References: [1] Bianchi et al. (2011), [2] Alamet al. (2015), [3] Jarrett et al. (2000), [4] Wright et al.(2010), [5] Condon et al. (1998), [6] Voges et al. (1999) L K = 1 . × L (cid:12) . Using the galaxy color u − g =1 .
5, we compute a mass to light ratio for the K -band,log ( M/L ) = − . M ∗ (cid:39) . × M (cid:12) . Provided thatsize of the PSF in SDSS DR13 is an upper limit forthe angular size of an unresolved bulge, we assume thatan upper limit for the bulge-to-total ratio (B/T) canbe estimated with an average (across all bands) valueof psfFlux / cModelFlux ∼ . ± .
05. We use this ra-tio to scale the total galaxy mass and derive M BH ≤ . ± (0 . . M (cid:12) , according to the M BH − M bulge re-lation (McConnell & Ma 2013), including the 1 σ scatterof 0.34 dex.The age and metallicity of the host galaxy were deter-mined by fitting a grid of galaxy models from Bruzual &Charlot (2003), using stellar evolutionary models from(Chabrier 2003). The fit was done using the pPXF code(Cappellari 2016). The best model was in agreementwith a single burst of star formation with an age of650 ±
300 Myr and a metallicity of Z =0.18 (here, Z = 0 . v host = 89 ± − , fromthe Ca II λλ − σ relation (McConnell & Ma 2013), this corre-sponds to a M BH = 10 . ± . M (cid:12) , consistent with ourprevious estimate.The field of iPTF16fnl was extensively observed forthe last six years by PTF/iPTF. No prior activity in thehost is detected with upper limits of ∼ ± FOLLOW-UP OBSERVATIONS3.1.
Photometric observations
Following spectroscopic identification of iPTF16fnl asa TDE candidate, the source was monitored at Palo-mar and the Ultraviolet and Optical Telescope (UVOT;Roming et al. 2005) on board the
Swift observatory(Gehrels et al. 2004). The UVOT observations weretaken in
UVW2 , UVM2 , UVW1 , U , B and V ; seeTable 2. The data were reduced using the software UVOTSOURCE using the calibrations described in Pooleet al. (2008) and updated calibrations from Breeveldet al. (2010). We use a 7.5 (cid:48)(cid:48) aperture centered on theposition of the transient.At Palomar, photometry in the g and Mould- R bandswere obtained with the iPTF mosaic wide-field cam-era on the Palomar 48-inch telescope (P48; Rahmeret al. 2008). Difference-image photometric measure-ments were provided by the IPAC Image Subtractionand Discovery Pipeline developed for the iPTF sur-vey (PTFIDE; Masci et al. 2017). Difference-imagingphotometry in the u’g’r’i’ bands, obtained with theSEDM, were computed using the FPipe software (Frem-ling et al. 2016). The zeropoints were calibrated usingstars in the SDSS footprint. Table 2 reports the mea-
Table 2 . 3 σ upper limits on radio emission for iPTF16fnl.Date Telescope Frequency Flux(UT) (GHz) (Jy)2016 Aug 31 VLA 6.1, 22 < × − < × − < × − < × − < . × − < × − < × − sured Swift aperture photometry magnitudes and thedifference imaging photometry for the Palomar data.The multi-band
Swift and optical lightcurve, correctedfor Galactic extinction, is shown in Figure 3. To correctthe UV bands for host light contamination, we used theaverage of the last four epochs of
Swift data, from MJD57712.7 to 57724.7 ( >
80 days), to subtract from theearly part of the lightcurve. The measurements wereconstant with an RMS of ∼ Swift data were excludedfrom the analysis, as they were dominated by the host,not by the central point source.In order to estimate the extinction in the host galaxy,we use all available
Swift
UV data and difference imag-ing photometry in ugri bands to fit blackbody emissioncurves, as detailed in Section 4. For each epoch, thephotometry is corrected for both Galactic (fixed) andadditional host extinction E( B − V ) from 0 to 0.25 magin 0.05 steps. The likelihood between the de-reddenedphotometry and a blackbody model is computed foreach epoch. In a final step, we marginalize over allepochs to derive the final value. We find that the bestfit corresponds to E( B − V )=0, with an upper limit ofE( B − V )=0.05. The selection of different extinctionlaws for host extinction (Calzetti et al. 2000; Fitzpatrick1999) does not change our conclusions. Given the as-sumption that the emission from iPTF16fnl in fact fol-lows a distribution, from now on we will assume thereddening in the host to be negligible.3.2. Radio observations
Radio follow-up observations of iPTF16fnl were takenwith the Jansky Very Large Array (VLA; PI A. Horesh),the Arcminute Microkelvin Imager (AMI; PI K. Moo-ley) and the James Clerk Maxwell Telescope andthe Submillimetre Common-User Bolometer Array 2(JCMT/SCUBA-2; PI A. K. H. Kong). The upper limitscorresponding to the observations are shown in Table 2.The limits for the monochromatic radio luminosities cor-responding to the first VLA epoch are νL ν < . × erg s − and νL ν < . × erg s − at 6.1 and 22 GHz.
10 0 10 20 30 40 50 60MJD - 57632.1161718192021 M a g n i t u d e SSSSSS SS SSS SSS S S SS SS
P60+SED-Machine Swift+UVOT PTFP48
UVW2UVM2UVW1u gri+0.5 A b s o l u t e m a g n i t u d e Figure 3 . Observed lightcurve for iPTF16fnl. The green solid line shows the best fit spline to the g -band data, which wascorrected for Galactic extinction. The errors for epochs later than 30 days are likely underestimated, as the bulge of the host is ∼ g -band, MJD 57632.1 is used as the reference epoch. The smallsymbol “S” on top shows the epochs when spectra were taken. (A color version of this figure is available in the online journal.) These limits are respectively two and one order of mag-nitude deeper than the VLA detection of ASASSN-14liat peak for a similar frequency range (van Velzen et al.2016; Alexander et al. 2016). t [yr]10 L [ e r g s ] E j =10 ergE j =10 ergE j =10 ergE j =10 erg Figure 4 . Upper limits for our 6 GHz (small triangle) and15 GHz (big triangles) observations. The lines show ana-lytic lightcurves for different TDE on-axis jet energies (colourcoded) from Generozov et al. (2017). Solid lines representthe lightcurves for 15 GHz and dashed lines for 6 GHz. Weassume n = 11 and the fiducial values provided in theirmodels for an optically thick case. The results are also con-sistent with an optically thin case. The X -axis is computedrelative to our lower limit of 11 days for the time to peaklight. We can use the limits reported here to argue againstthe presence of an on-axis relativistic outflow, or at leastconstrain the energy of the jet, E j . We compare ourlimits in 6 and 15 GHz with the analytical lightcurvesfor on-axis TDE radio emission from Generozov et al.(2017). Figure 4 shows that our early time observationssuggest jet energies with E j < erg s − .Additionally, we contrast our measurements, scaled tothe redshift of PS1-11af, with the GRB afterglow modelsof van Eerten et al. (2012) presented in Chornock et al.(2014) (their figure 13). Based on our non-detections, wecan rule out the existence of a relativistic jet, as viewed30 ◦ off-axis. For larger angles, the radio emission is ex-pected to arise at later times ( > X-ray observations
We observed the location of iPTF16fnl with the X-Ray Telescope (XRT; Burrows et al. 2005) on-board the
Swift satellite beginning at 19:32 UT on 30 August 2016.Regular monitoring of the field in photon counting (PC)mode continued over the course of the next four months(PIs T. Holoien and B. Cenko).No significant emission is detected in individualepochs (typical exposure times of ≈ . . × − counts s − in the 0.3–10.0 keV bandpassover this time period.Stacking all the XRT data obtained over this periodtogether (58 ks of total exposure time), we find evidencefor a weak ( ≈ σ significance) X-ray source at this loca-tion with a 0.3–10.0 keV count rate of (2 . ± . × − counts s − .With only 15 source counts we have limited abilityto discriminate between spectral models; however, withseveral photon energies detected above 1 keV. We deriveresponse matrices for the stacked XRT observations us-ing standard Swift tools. Adopting a power-law modelfor the spectrum with a photon index of 2 and account-ing for line-of-sight absorption in the Milky Way (Will-ingale et al. 2013), we find the measured count rate cor-responds to an unabsorbed flux of 0.3–10.0 keV flux of4 . +3 . − . × − erg cm − s − .At the distance of Mrk 950, this corresponds to an X-ray luminosity of L X = 2 . +1 . − . × erg s − . Withoutadditional information (e.g., variability and/or spectra),we cannot determine conclusively if this X-ray emissionis associated with the transient iPTF16fnl, or if this isunrelated X-ray emission from the host nucleus (e.g., anunderlying active galactic nucleus) or even a populationof X-ray binaries or ultra-luminous X-ray sources. How-ever, the lack of evidence for ongoing star formation orAGN-like emission lines in the late-time optical spec-tra of Mrk 950 ( §
5) suggest an association with the tidaldisruption event.If this is indeed the case, the implied X-ray emissionwould be extremely faint, both in an absolute and arelative sense. We can contrast, for example, with the X-ray emission observed from ASASSN-14li (Holoien et al.2016b; van Velzen et al. 2016), with a peak luminosityapproximately four orders of magnitude above that seenfor iPTF16fnl. Even sources with much fainter X-rayemission, such as ASASSN-15oi (Holoien et al. 2016a),still outshine iPTF16fnl by more than a factor of 100.3.4.
Spectroscopic Observations
Spectroscopic follow-up observations of iPTF16fnlhave been carried out with numerous telescopes and in-struments, summarized in the spectroscopic log in Table1. Figure 5 shows the spectral sequence for iPTF16fnl,spanning three months. The spectroscopic data aremade public via WISeREP (Yaron & Gal-Yam 2012).Given the brightness of the galactic bulge relative tothe TDE, the interpretation of the TDE spectrum posessome challenges. A noticeable feature is the strong hostcomponent in all the available spectra. Different slitwidths, the variable seeing, different orientations of theslit during the acquisition of the data (generally takenat the parallactic angle) and the highly elongated ge-ometry of the host galaxy, contribute to create a strongvariation in the contribution from the host component, which appears to vary from one instrument to another.Early epoch spectra ( <
50 days), the most prominentemission lines correspond to broad He II and H α , shownin Figure 8. These lines have an average FWHM of ∼ ,
000 km s − and ∼ ,
000 km s − respectively. Theanalysis and evolution of their profiles is further exploredin Section 4.2.Several narrow absorption lines, associated with thehost galaxy, were identified. The region around H α con-tains an emission line that can be associated with [N II ]at λ II at λλ I D doubletat λλ I λλ I λλ II is detected at λλ λλ ANALYSIS4.1.
SED and bolometric lightcurveSwift host subtracted
U V W U V M U V W
Python package emcee (Foreman-Mackey et al. 2013). The blackbody bolometric luminos-ity, along with the best fit for the temperature and theradius are shown in Figure 6. Because only g -band mea-surements were available for our first detection epoch,we assumed that the luminosity follows a blackbodyemission with a temperature of 21,000 K (average for thefirst 2 weeks) and scaled the flux to match the g -bandmagnitude.The bolometric luminosity at peak is L p (cid:39) (1 . ± . × erg s − . This is one order of magnitude lowerthan most of the optical TDEs (PS1-10jh, ASASSN-14ae, ASASSN-14li and ASASSN-15oi; see Figure 6).If compared to ASASSN-15lh, a TDE candidate froma rotating high-mass SMBH (Leloudas et al. 2016) ,the peak is two order of magnitude fainter. Based onour M BH estimate, we derive its Eddington luminosity L Edd = 2 . +3 . − . × erg s − , implying that at peak,iPTF16fnl shines only 2 −
10% of L Edd . Assuming aradiative efficiency of η = 0 .
1, this translates into apeak accretion rate of ˙ M peak ∼ . × − M (cid:12) yr − ( L = η ˙ M c ).Integrating the available bolometric luminosity (seeFigure 6), we find the radiated energy to be E R = (2 . ± . × erg. The accreted mass for this interval is M acc ∼ . × − M (cid:12) .The rest-frame blackbody bolometric luminosity (Fig- This object is somehow controversial and has been initiallyinterpreted as a superluminous supernova (Dong et al. 2016) Å ]012345678 n o r m a li z e d f l u x F λ + c o n s t a n t -1.9 d -1.7 d-0.9 d-0.8 d-0.1 d 0.0 d1.4 d2.1 d11.3 d 29.3 d29.3 d44.2 d51.3 d 55.2 d 88.2 d119.2 dmodel H η H δ H γ H e II H β M g F e I F e I N a I B a II H α H e I P a I C a II P P P a P P Figure 5 . Spectral sequence of iPTF16fnl. The spectra are color coded by instrument: SEDM/P60 - black, DBSP/P200 - brown,2 m LOYDS/LCO - blue, LRIS/Keck - purple, Deveny/DCT - green, GMOS-N/Gemini North - red and GTC/OSIRIS-olive.Telluric bands are marked with blue shaded areas. The labels on the top of the panel correspond to the main identified linesboth from the TDE and the host galaxy. The spectra have been binned using the average of 3 pixels and low S/N areas wereexcluded from the plot. The best-fit galaxy model is shown in the bottom with gray colour. ure 6) is fit with the characteristic power law L ∝ (( t − t ) /τ ) − / and an empirically motivated exponen-tial profile, L ∝ e − ( t − t ) /τ , where t and τ are free pa-rameters. For the decaying part of the lightcurve, theexponential model fits the data better ( χ = 17 vs. 110),and best fit parameters are τ (cid:39)
15 and t (cid:39) −
6. Forcomparison, the decay for other optical TDEs is slower: ASASSN-14ae had τ = 30 days, whereas ASASSN-15oiand ASASSN-14li faded on timescales of 46.5 and 60days respectively. Continued, high quality photometricmonitoring would be required to draw conclusive resultson long-term evolution, beyond the initial fading stage.We estimate a lower limit for the time from disrup-tion to peak light t peak ≥
11 days from the bolometriclightcurve. We select the measurements at ±
10 day fromthe peak in g -band and fit the luminosity with a 2 de-gree polynomial. For the raising part of the lightcurve,we use our only available g -band measurement. Whilethis approach is widely used in estimating the explosiontime for supernovae, the emission mechanism for TDEsis different and therefore it only yields to lower limits,as seen when applied to PS1-10jh.Assuming that our bolometric lightcurve traces therate of mass falling into the black hole, ˙ M ( t ), we usethe lightcurve models from Guillochon & Ramirez-Ruiz(2013) (hereafter GR13), scaling them to the peak ac-cretion mass rate and time to peak. The lightcurvesare defined for a range of impact parameters 0 . ≤ β ≤ .
5, where β is defined as the depth of the encounter β = R T /R p , R T is the tidal disruption radius and R p the pericentre radius. We impose a M BH = 2 × M (cid:12) , but we leave the mass and radius of the disruptedstar as free parameters. Our best fit corresponds toa star with polytropic index γ = 4 / M ∗ ∼ .
03 M (cid:12) , R ∗ ∼ . (cid:12) and a depth of theencounter comparable to the disruption radius ( β (cid:39) t peak ∼
11 days, comparable withour previous naive estimate. If we impose that lowmass stars are fully convective and fit for an object with γ = 5 /
3, we find a relatively good fit for a partial dis-ruption ( β ∼ .
6) and a similar value of M ∗ ∼ .
06 M (cid:12) ,although in this case t peak is shorter than our obser-vations suggest. Although these values illustrate thatthe disrupted object was likely a low-mass star, detailedmodeling of the event would be required to draw quan-titative results.The blackbody model has an average temperature ofT BB = 19 , ± BB . Given the uncertainty of theextinction in the host, these values can be assumed aslower limits. The model blackbody radius, R BB starts at ∼ × cm, linearly declines for the first twenty daysand then flattens to 5 × cm. These radii are muchlarger than the Schwarzschild radius r Sch ∼ × cmof the nuclear black hole. In comparison, we note thatthe tidal disruption radius of such SMBH for a mainsequence Solar-like star is R T (cid:39) × cm. Photo-spheric emission at radii larger than R T is commonlyobserved for the optical sample of TDEs. This has beenattributed to the existence of a reprocessing layer atlarger radii, which re-emits the X-ray and UV in opticalbands (Loeb & Ulmer 1997; Guillochon et al. 2014). Al-ternatively, the emission mechanism may originate fromthe energy liberated by shocks between streams in theapocenter, during the formation of the accretion disk (Piran et al. 2015). Such an optically thick layer, mainlyformed of stellar debris, is associated with the origin ofthe emission line signature for optical TDEs (Roth et al.2016; Metzger & Stone 2016). l o g L b o l ( e r g s − ) ∝ ( t − t ) − / ∝ e ( t − t /τ GR13 model [ γ = 4 / ] T BB ( K )
20 10 0 10 20 30 40 50 60 70MJD - 57632.1 (restframe days)0123 R BB ( c m ) Figure 6 . Top: Bolometric blackbody lightcurve foriPTF16fnl. Blue circles represent the fits with Galactic ex-tinction correction only. The first (empty) data point wascomputed assuming an average blackbody temperature of21,000 K (average for the first 2 weeks) and scaling the fluxto match the g -band magnitude. The dashed line shows thebest fit to a power law of the form L ∝ e − ( t − t ) /τ . Thedotted line shows the best fit to a L ∝ ( t − t ) − / . A solidline shows the best fit to GR13 models. Thick lines repre-sent a sample of fast-fading TDEs for comparison: PS1–11af:dot-dashed magenta (Chornock et al. 2014), ASASSN-14ae:solid gray (Holoien et al. 2014), ASASSN-14li: dashed brown(Holoien et al. 2016b) and ASASSN-15oi: dotted orange(Holoien et al. 2016a). The reference MJD for the objectsis the discovery date or epoch of peak luminosity (wheneveravailable). Middle: Temperature evolution. Bottom: Evolu-tion of the blackbody radius. (A color version of this figureis available in the online journal.) Spectroscopic analysis
The early time spectrum of iPTF16fnl is dominated byblue continuum radiation and the characteristic broad F [ e r g c m s Å ] iPTF16fnl +29.3 daysHostBlack bodyHost + black body Å ] R e s i d u a l s H He IHe II He IH HH He II
Figure 7 . Example of host subtraction. The original spec-trum, taken a +29.3 days (black thick line) was fit with acombination of host spectrum (blue solid line) and a black-body fit (magenta dashed line). The best fit is shown witha red line. The residuals (corrected with a 2 degree polyno-mial) are shown in the lower panel. We mark the relevantemission lines. He II lines are clearly identified at λλ λλ He II λ α , although the He I λ β inemission. However the strong host contribution makesits identification challenging at late times. In our anal-ysis, we use the late time host spectrum (+119.2 days)as our template. We select the highest signal-to-noise(S/N) spectra, and fit them with a combination of hostand a blackbody continuum, as show in Figure 7. On theresidual spectrum, we fit a line model using the python package lmfit (Non-Linear Least-Squares Minimizationand Curve-Fitting for Python). After masking the re-gions affected by telluric absorption, He II + H β andH α + He I lines are fit using two component Lorentzianmodel, in order to derive the width (FWHM) and cen-tral location of the emission. The results, plotted asinsets, are shown in Figure 8. The fluxes for each lineare derived from the best fit model and shown in Table3. As discussed in Brown et al. (2017), if all the fluxin the He II line would be attributed to recombinationproduced by black body photoionizing radiation, the ob-served flux of > erg s − would require black bodyradiation with temperature ∼ × K, which is higherthan our fit, requiring an additional energy source topower this line.Around peak, the He II lines appear to have highervelocity, showing an average value of FWHM HeII (cid:39) , ± ,
000 km s − , in contrast to the H α line,with FWHM Hα (cid:39) , ±
500 km s − . At +30 and+45 days after peak, the FWHM narrows down to Table 3 . Flux values for H α and He II 4868˚Afor the linesshown in Figure 8. The values were derived from the bestmodel fit line profile. From the fit uncertainties, we estimateerrors of 40% and 30% of the total flux for H α and He IIlines respectively.MJD Phase H α He II(d) (d) (erg s − ) (erg s − )57630.4 − × × − × × × × × × × × × × × , ± − for He II and 6 , ±
600 km s − for H α . The center of the lines appears constantwithin the scatter for the first 90 days: for He II ,the lines appear marginally blueshifted with velocity of − ±
700 km s − , while the H α lines appear to be con-sistent with the reference wavelength, with a shift invelocity of − ± − . DISCUSSIONiPTF16fnl is the faintest and fastest event in the cur-rent sample of optically discovered TDEs. Assumingour extinction estimation method is accurate, its lumi-nosity at peak is one order of magnitude lower than anyother optical/UV TDE discovered so far. Its timescale,as shown in Figure 10, also makes it as an outlier amongthe existing sample.The host of iPTF16fnl is another example of a TDE ina post-starburst galaxy, further linking the propensity ofTDEs to such galaxies (Arcavi et al. 2014; French et al.2016). Moreover, E+A galaxy hosts seem to be exclusivefor the lowest redshift TDEs ( z < .
05) (see Figure 10).The origin of the burst could be associated with a mergerepisode, as discussed in the case of ASASSN-14li (Prietoet al. 2016). The violent relaxation in the stellar orbitscould enhance the rate of captures, as stars can undergoencounters that will scatter them towards the SMBH.Lower SMBH masses ( < M (cid:12) ) can increase thenumber of deeper encounters (Kochanek 2016), allow-ing for disruptions with smaller pericenter radius, R p .However, theoretical works (Guillochon & Ramirez-Ruiz2013; Stone et al. 2013) only show a weak correla-tion between the impact parameter β , and the peakof the flare. Therefore, the low luminosity and fasttimescales shall be attributed to a lower mass black holeand/or lower mass for the disrupted star. Using ourfit to TDE lightcurves from Guillochon & Ramirez-Ruiz(2013) with our estimated M BH ∼ × M (cid:12) , we ob-0 -15 0 150 -1.7d H α He I He II H β -1.7d -15 0 15 -15 0 15 -0.8d H α He I He II H β -0.8d -15 0 15 -15 0 15 H α He I He II H β -15 0 15 -15 0 15 H α He I He II H β -15 0 15 -15 0 15 H α He I He II H β -15 0 15 -15 0 15 H α He I He II H β -15 0 15 -15 0 15 H α He I He II H β -15 0 15 -15 0 15 H α He I He II H β -15 0 15 H vel [ km/s]He II vel [ km/s] Figure 8 . Residual normalized spectrum showing the line region around He II λ α (bottom row) lines forthe higher S/N spectra of iPTF16fnl. In addition, the location of H β (top) and He I (bottom) lines are shown. Telluric regions,shown with shaded areas, were excluded from the fit. The first three epochs of H α were fit using a Gaussian model, as theLorentzian provided a worse fit. All the other lines were fit using a linear background and a Lorentzian line profile. We couldnot find a good fit for the last three epochs of H α , but the spectrum is included for completeness. (A color version of this figureis available in the online journal.) F W H M ( k m s − ) Figure 9 . Top: Evolution of FWHM for He II λλ α (red circles) vs. phase of the spec-trum. The last three epochs of H α do not have a reliablemeasurement, and therefore are excluded form the figure. (Acolor version of this figure is available in the online journal.) tain the mass of the disrupted star to be M ∗ ∼ .
03 for t peak value of 11 days.iPTF16fnl has clearly faster decay timescales thanother TDEs, but also lower M BH . Figure 11 shows acomparison of the e-folding timescale for iPTF16fnl andother optical TDEs, computed from exponential decaymodels, and the galaxy M BH . There seems to be atrend between these two values for the optical/UV TDEpopulation, in general agreement with the theoreticalscaling between fallback timescale and black hole mass, t ∝ M B H / . As a cautionary note, while literaturegenerally reports M BH based on bulge mass/luminosity,our best measurement is based on the M − σ relation,although the bulge luminosity method yielded to sim-ilar results. Figure 10 shows that the most luminousflares ( L bol > erg s − ) tend to fade on intermedi- ate timescales, ∼
50 days. However, there does not seemto be an evident correlation between the peak luminos-ity and the black hole mass, as discussed in Hung et al.(2017) (see their figure 15).The tension between theoretical prediction of TDErates and the ones inferred from observations is an activefield of research. While it is difficult to explain the differ-ences in terms of host galaxy properties (Stone & Met-zger 2016), an observational bias towards the brighterevents seems to offer a more plausible explanation. Thediscovery of iPTF16fnl has consequences for previousoptical searches for nuclear tidal disruptions. In fact, itspeak absolute magnitude M g = − . M g ∼ −
20 mag) nuclear flares (Arcavi et al. 2014).Systematic searches using the colour (including UV)and location of the transient, rather than its absolutemagnitude, will increase our sensitivity to fainter flares.Consistent candidate selection using future surveys suchas ZTF or LSST will allow us to explore the full lumi-nosity function of tidal disruption flares. Spectroscopicconfirmation of the candidates will be essential to iden-tify this faint population. Dedicated instruments fortransient classification such as SEDM will become thebig players in this new era.1
Figure 10 . Comparison of the peak luminosity and decaytime of iPTF16fnl with a sample of optical TDE from litera-ture. The dot size encodes the redshift of the host galaxy. Anexternal circle symbolizes the classification of the host galaxyas a post-starburst E+A galaxy. The optical TDE sampleis based on published data: Gezari et al. (2008); van Velzenet al. (2011); Gezari et al. (2012); Chornock et al. (2014);Arcavi et al. (2014); Holoien et al. (2016b, 2014, 2016a) andHung et al. (2017). M BH [ M fl ]020406080100120 e - f o l d i n g t i m e [ d a y s ] iPTF16fnlPS1-10jhiPTF16axa09geASASSN14aeASASSN14liASASSN15oiD3-13TDE1 D1-9 l o g L p e a k [ e r g s − ] Figure 11 . Mass of the host galaxy SMBH compared to thee-folding timescale for a sample of optical TDE. The dot colorencodes the TDE peak luminosity. The peak luminositieswere derived from literature: D1-9, D3-13 Gezari et al. (2008,2009), PS1 (Gezari et al. 2012), PS1-11af (Chornock et al.2014), ASASSN-14ae, ASASSN-14li, ASASSN-15oi (Holoienet al. 2016b, 2014, 2016a). The bolometric luminosities forTDE1 van Velzen et al. (2011) and 09ge (Arcavi et al. 2014)were derived by scaling the reported blackbody tempera-ture emission to match the reported M g . We assumed thestandard dispersion in the McConnell & Ma (2013) relationwhenever uncertainties for M BH were not reported. A ten-tative correlation t ∝ ( M BH ) / is provided to guide the eye(Guillochon & Ramirez-Ruiz 2013). (A color version of thisfigure is available in the online journal.) 6. CONCLUSIONSWe have presented the discovery and follow-up datafor iPTF16fnl, a TDE candidate discovered by the iPTFsurvey on 2016 August 29th. The real-time image-subtraction pipeline and rapid spectroscopic classifi-cation allowed us to initiate a timely follow-up cam-paign. The photometric and spectroscopic signatures ofiPTF16fnl are consistent with the sample of previous op-tically selected TDEs. As observed in other TDEs, theobject shows very strong emission in UV wavelengths,with a T BB (cid:39) ,
000 K. The temperature does notshow strong evolution and the decrease in luminosityis best explained as a decrease in the size of the radi-ating region. In agreement with previous work, the sizeof this region, defined by its photospheric radius, is alsoabout an order of magnitude larger than R T . The earlytimes, the spectroscopic signature of iPTF16fnl is dom-inated by He II and hydrogen lines, although we alsodetect emission from He I. After two months after peaklight, most of the lines have faded. The exception isHe II , which can be identified with a relatively constantFWHM of ∼ − .iPTF16fnl is remarkable in three ways: it is the near-est well studied optical/UV TDE (66.6 Mpc), and ithas one of the shortest exponential decay timescales(about 15 days) and one of the lowest peak luminosi-ties, L p (cid:39) (1 . ± . × erg s − . Also, its hostgalaxy has the lowest M BH among the optical sample ofTDEs. Although this could justify the fast decay, its lowluminosity may be related to the disruption of a lowermass star, or even a partial disruption.Our work demonstrates that TDEs cover a wide rangeof luminosities and timescales. Current and future sur-veys, such as ATLAS, PS-1, ZTF and LSST, will providelarge numbers of events with exquisite temporal cov-erage. Multifrequency follow-up of the candidates willlead to a better understanding of the underlying emis-sion mechanism, hopefully leading to a unified multi-frequency emission model. Large samples obtained withwell understood selection criteria will be key to the studyof TDE demographics, providing a unique link betweentheoretical studies of SMBH astrophysics and observa-tions.2 REFERENCES Alam, S., Albareti, F. D., Allende Prieto, C., et al. 2015, ApJS,219, 12Alexander, K. D., Berger, E., Guillochon, J., Zauderer, B. A., &Williams, P. K. G. 2016, ApJL, 819, L25Arcavi, I., Gal-Yam, A., Sullivan, M., et al. 2014, ApJ, 793, 38Auchettl, K., Guillochon, J., & Ramirez-Ruiz, E. 2016, ArXive-prints, arXiv:1611.02291Bade, N., Komossa, S., & Dahlem, M. 1996, A&A, 309, L35Bell, E. F., McIntosh, D. H., Katz, N., & Weinberg, M. D. 2003,ApJS, 149, 289Bianchi, L., Herald, J., Efremova, B., et al. 2011, Ap&SS, 335,161Bloom, J. S., Giannios, D., Metzger, B. D., et al. 2011, Science,333, 203Breeveld, A. A., Curran, P. A., Hoversten, E. A., et al. 2010,MNRAS, 406, 1687Brown, G. C., Levan, A. J., Stanway, E. R., et al. 2015,MNRAS, 452, 4297Brown, J. S., Kochanek, C. S., Holoien, T. W.-S., et al. 2017,ArXiv e-prints, arXiv:1704.02321Brown, T. M., Baliber, N., Bianco, F. B., et al. 2013, PASP, 125,1031Bruzual, G., & Charlot, S. 2003, MNRAS, 344, 1000Burrows, D. N., Hill, J. E., Nousek, J. A., et al. 2005, SSRv, 120,165Burrows, D. N., Kennea, J. A., Ghisellini, G., et al. 2011,Nature, 476, 421Cabanela, J. E., & Aldering, G. 1998, AJ, 116, 1094Calzetti, D., Armus, L., Bohlin, R. C., et al. 2000, ApJ, 533, 682Cao, Y., Nugent, P. E., & Kasliwal, M. M. 2016, PASP, 128,114502Cappellari, M. 2016, ArXiv e-prints, arXiv:1607.08538Cenko, S. B., Krimm, H. A., Horesh, A., et al. 2012, ApJ, 753, 77Chabrier, G. 2003, PASP, 115, 763Chornock, R., Berger, E., Gezari, S., et al. 2014, ApJ, 780, 44Condon, J. J., Cotton, W. D., Greisen, E. W., et al. 1998, AJ,115, 1693Dong, S., Shappee, B. J., Prieto, J. L., et al. 2016, Science, 351,257Drake, A. J., Djorgovski, S. G., Mahabal, A., et al. 2009, ApJ,696, 870Dressler, A., Smail, I., Poggianti, B. M., et al. 1999, ApJS, 122,51Evans, C. R., & Kochanek, C. S. 1989, ApJL, 346, L13Evans, P. A., Beardmore, A. P., Page, K. L., et al. 2009,MNRAS, 397, 1177Fitzpatrick, E. L. 1999, PASP, 111, 63Fixsen, D. J., Cheng, E. S., Gales, J. M., et al. 1996, ApJ, 473,576Foreman-Mackey, D., Hogg, D. W., Lang, D., & Goodman, J.2013, PASP, 125, 306Fremling, C., Sollerman, J., Taddia, F., et al. 2016, A&A, 593,A68French, K. D., Arcavi, I., & Zabludoff, A. 2016, ApJL, 818, L21Gehrels, N., Chincarini, G., Giommi, P., et al. 2004, ApJ, 611,1005Generozov, A., Mimica, P., Metzger, B. D., et al. 2017, MNRAS,464, 2481Gezari, S., Martin, D. C., Milliard, B., et al. 2006, ApJL, 653,L25Gezari, S., Basa, S., Martin, D. C., et al. 2008, ApJ, 676, 944Gezari, S., Heckman, T., Cenko, S. B., et al. 2009, ApJ, 698, 1367Gezari, S., Chornock, R., Rest, A., et al. 2012, Nature, 485, 217Gezari, S., Hung, T., Blagorodnova, N., et al. 2016, TheAstronomer’s Telegram, 9433 Guillochon, J., Manukian, H., & Ramirez-Ruiz, E. 2014, ApJ,783, 23Guillochon, J., & Ramirez-Ruiz, E. 2013, ApJ, 767, 25Hills, J. G. 1975, Nature, 254, 295Holoien, T. W.-S., Prieto, J. L., Bersier, D., et al. 2014,MNRAS, 445, 3263Holoien, T. W.-S., Kochanek, C. S., Prieto, J. L., et al. 2016a,MNRAS, 463, 3813—. 2016b, MNRAS, 455, 2918Hung, T., Gezari, S., Blagorodnova, N., et al. 2017, ArXive-prints, arXiv:1703.01299Jarrett, T. H., Chester, T., Cutri, R., et al. 2000, AJ, 119, 2498Kochanek, C. S. 2016, MNRAS, 461, 371Komossa, S. 2015, Journal of High Energy Astrophysics, 7, 148Komossa, S., & Bade, N. 1999, A&A, 343, 775Lacy, J. H., Townes, C. H., & Hollenbach, D. J. 1982, ApJ, 262,120Leloudas, G., Fraser, M., Stone, N. C., et al. 2016, NatureAstronomy, 1, 0002Levan, A. J., Tanvir, N. R., Cenko, S. B., et al. 2011, Science,333, 199Loeb, A., & Ulmer, A. 1997, ApJ, 489, 573Masci, F. J., Laher, R. R., Rebbapragada, U. D., et al. 2017,PASP, 129, 014002McConnell, N. J., & Ma, C.-P. 2013, ApJ, 764, 184Metzger, B. D., & Stone, N. C. 2016, MNRAS, 461, 948Miller, A. A., Kasliwal, M. M., Cao, Y., et al. 2017, ArXive-prints, arXiv:1703.07449Ofek, E. O., Laher, R., Surace, J., et al. 2012, PASP, 124, 854Petrosian, A., McLean, B., Allen, R. J., & MacKenty, J. W.2007, ApJS, 170, 33Phinney, E. S. 1989, in IAU Symposium, Vol. 136, The Center ofthe Galaxy, ed. M. Morris, 543Piran, T., Svirski, G., Krolik, J., Cheng, R. M., & Shiokawa, H.2015, ApJ, 806, 164Poole, T. S., Breeveld, A. A., Page, M. J., et al. 2008, MNRAS,383, 627Prieto, J. L., Kr¨uhler, T., Anderson, J. P., et al. 2016, ApJL,830, L32Rahmer, G., Smith, R., Velur, V., et al. 2008, in Proc. SPIE,Vol. 7014, Ground-based and Airborne Instrumentation forAstronomy II, 70144YRees, M. J. 1988, Nature, 333, 523Roming, P. W. A., Kennedy, T. E., Mason, K. O., et al. 2005,SSRv, 120, 95Roth, N., Kasen, D., Guillochon, J., & Ramirez-Ruiz, E. 2016,ApJ, 827, 3Sand, D. J., Brown, T., Haynes, R., & Dubberley, M. 2011, inBulletin of the American Astronomical Society, Vol. 43,American Astronomical Society Meeting Abstracts Table 1 . Log of spectroscopic observations of iPTF16fnl.MJD Slit a Telescope+Instrument Grism / Grating Dispersion Resolution b Exposure(d) (arcsec) (˚A/pix) ˚A km/s) (s)57630.2 3.0 P60+SEDM none 25.5 540 (2900 km s − ) 90057630.4 2.0 LCO 2-m+FLOYDS – 1.7 15.3 (820 km s − ) 270057631.0 1.0 GTC+Osiris 1000B+2500I 2.12+1.36 7.1 (380 km s − ) 30057631.0 1.0 GTC+Osiris 2500R 1.04 7.1(380 km s − ) 30057631.2 2.2 P60+SEDM none 25.5 504 (2700 km s − ) 90057631.3 1.5 P200+DBSP 600/4000 1.5 10.5 (560 km s − ) 60057633.5 3.6 P60+SEDM lenslet arr. 25.5 672 (3600 km s − ) 180057634.2 3.0 P60+SEDM lenslet arr. 25.5 613 (3300 km s − ) 120057636.2 3.0 P60+SEDM lenslet arr. 25.5 659 (3500 km s − ) 120057637.2 3.6 P60+SEDM lenslet arr. 25.5 589 (3200 km s − ) 120057638.2 3.0 P60+SEDM lenslet arr. 25.5 556 (3000 km s − ) 120057643.22 1.0 VLT Kueyen+UVES 437+860 0.030+0.050 0.15 (8 km s − ) 2x150057643.25 1.0 VLT Kueyen+UVES 346+580 0.024+0.034 0.15 (8 km s − ) 2x150057643.4 1.5 DCT+Deveny 300 2.17 6.9 (370 km s − ) 60057661.4 1.0 Keck I+LRIS 400/3400+400/8500 2.0 6.5 (350 km s − ) 60057661.4 1.0 Keck I+LRIS 400/3400+400/8500 2.0 6.5 (350 km s − ) 60057676.3 1.0 GeminiN+GMOS – 1.35 8.1 (435 km s − ) 60057683.4 1.0 GeminiN+GMOS – 1.35 8.1 (435 km s − ) 120057687.3 1.5 P200+DBSP 600/4000 1.5 10.5 (560 km s − ) 60057694.4 1.0 Keck I+LRIS 400/3400+400/8500 2.0 7.2 (390 km s − ) 135057720.3 1.0 Keck I+LRIS 400/3400+400/8500 2.0 6.5 (350 km s − ) 180057751.3 1.0 Keck I+LRIS 400/3400+400/8500 2.0 6.5 (350 km s − ) 1800 a For the IFU, the extraction radius (in arcsec) is indicated. a Measured using the FWHM of λ Wright, E. L., Eisenhardt, P. R. M., Mainzer, A. K., et al. 2010,AJ, 140, 1868Wyrzykowski, (cid:32)L., Zieli´nski, M., Kostrzewa-Rutkowska, Z., et al.2017, MNRAS, 465, L114 Yaron, O., & Gal-Yam, A. 2012, PASP, 124, 668Zabludoff, A. I., Zaritsky, D., Lin, H., et al. 1996, ApJ, 466, 104Zauderer, B. A., Berger, E., Soderberg, A. M., et al. 2011,Nature, 476, 425
Facility:
DCT, Gemini North, GTC, Hale, JMCT, Keck:I,Keck:II, LCOGT, PO:1.2 m, PO:1.5 m, PS1, VLT
Software:
AstroPy, EMCEE (Foreman-Mackey et al. 2013),IRAF, LMFIT, pPXF (Cappellari 2016), PTFIDE (Masci et al. 2017),PYRAF, UVOTSOURCE.
ACKNOWLEDGMENTS
We thank M. Urry for the use of her programme time to obtain target of oppor-tunity observations of iPTF16fnl with DBSP on 200-inch telescope on Palomar.We also thank Nathaniel Roth, Sjoert Van Velzen and Nicholas Stone for usefulcomments and discussions, and James Guillochon for his support with
TDEFit .This work was supported by the GROWTH project funded by the National Sci-ence Foundation under Grant No 1545949. We thank the Gemini Fast Turnaroundprogram (PI: T. Hung). S.G. is supported in party by NSF CAREER grant1454816 and NASA Swift Cycle 12 grant NNX16AN85G. A.Y.Q.H. was supportedby a National Science Foundation Graduate Research Fellowship under Grant No.DGE1144469. A.J.C.T. acknowledges support from the Spanish Ministry ProjectAYA 2015-71718-R. Support for I.A. was provided by NASA through the Ein-stein Fellowship Program, grant PF6-170148. G.H. is supported by the NationalScience Foundation (NSF) under Grant No. 1313484. This work makes use ofobservations from the LCO network. G.H. is supported by the National ScienceFoundation (NSF) under Grant No. 1313484. This work makes use of observations from the LCO network.We acknowledge the use of public data from the Swift data archive. The CSSsurvey is funded by the National Aeronautics and Space Administration underGrant No. NNG05GF22G issued through the Science Mission Directorate Near-Earth Objects Observations Program. The CRTS survey is supported by theU.S. National Science Foundation under grants AST-0909182. LANL participa-tion in iPTF was funded by the US Department of Energy as part of the Lab-oratory Directed Research and Development program. Part of this research wascarried out at the Jet Propulsion Laboratory, California Institute of Technology,under a contract with the National Aeronautics and Space Administration. Partof this work is based on observations obtained at the Gemini observatory underthe Fast Turnaround program. The Gran Telescopio Canarias (GTC) operatedon the island of La Palma at the Spanish Observatorio del Roque de los Mucha-chos of the Instituto de Astrofisica de Canarias. This publication makes useof data products from the Wide-field Infrared Survey Explorer, which is a jointproject of the University of California, Los Angeles, and the Jet Propulsion Lab-oratory/California Institute of Technology, funded by the National Aeronauticsand Space Administration. The James Clerk Maxwell Telescope is operated bythe East Asian Observatory on behalf of The National Astronomical Observatoryof Japan, Academia Sinica Institute of Astronomy and Astrophysics, the KoreaAstronomy and Space Science Institute, the National Astronomical Observatoriesof China and the Chinese Academy of Sciences (Grant No. XDB09000000), withadditional funding support from the Science and Technology Facilities Council ofthe United Kingdom and participating universities in the United Kingdom andCanada. VAF is supported by the Russian Science Foundation under grant 14-50-00043.
APPENDIX4
Table 2 . Optical and UV photometry of iPTF16fnl in AB magnitude system.These are the originally measured magnitudes with
Swift and difference imaging piplines. For P48+CFH12k, the r -band column contains measure-ments in Mould- R filter system.These measurements are not corrected for Galactic extinction. Table 2 is published in its entirety in the machine-readable format. A portion isshown here for guidance regarding its form and content. MJD Telescope
UV W UV M UV W U B V u g r i (days) + Instrument (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag)57626.4 P48+CFH12k – – – – – – – 17.50 ± ± ± ± ± ± ± ± ± ± ± ±±