Mon.Not.R.Astron.Soc.000,000–000 (0000) Printed30January2015 (MNLATEXstylefilev2.2) Supernova 2013by: A Type IIL Supernova with a IIP-like ⋆ light curve drop 5 S. Valenti1,2†, D. Sand3, M. Stritzinger4, D. A. Howell1,2, I. Arcavi1,5, C. McCully1,2, 1 0 M. J. Childress6,7, E.Y. Hsiao4,8, C. Contreras4,8, N. Morrell8, M. M. Phillips8, 2 M. Gromadzki9,10, R. P. Kirshner11, G. H. Marion11,12 n a 1 Las Cumbres Observatory Global Telescope Network, 6740 Cortona Dr., Suite 102, Goleta, CA 93117, USA J 2 Department of Physics, University of California, Santa Barbara, Broida Hall, Mail Code 9530, Santa Barbara, CA 93106-9530, USA 3 Physics Department, TexasTechUniversity, Lubbock, TX , 79409, USA 9 4 Department of Physics and Astronomy, Aarhus University,Ny Munkegade 120, DK-8000 Aarhus C, Denmark 2 5 Kavli Institute for Theoretical Physics, University of California, Santa Barbara, CA 93106, USA 6 Research School of Astronomy and Astrophysics, Australian National University,Canberra, ACT 2611, Australia ] E 7 ARC Centre of Excellence for All-sky Astrophysics (CAASTRO); Australian National University; Canberra, ACT 2611, Australia 8 Carnegie Observatories, Las Campanas Observatory, Colina El Pino, Casilla 601, Chile H 9 Millennium Institute of Astrophysics, Sotero Sanz 100, Oficina 104, Providencia, Santiago h. 10 Instituto de F´ısica y Astronom´ıa, Universidad de Valpara´ıso, Av. Gran Bretan˜a 1111, Playa Ancha, Casilla, 5030, Chile p 11 Harvard-Smithsonian Centerfor Astrophysics, 60 Garden Street, Cambridge, MA 02138, USA ) - 12 Universityof Texasat Austin, 1University Station C1400, Austin, TX, 78712-0259, USA o r t s a Accepted .....;Received....;inoriginalform.... [ 2 v ABSTRACT 1 Wepresentmulti-bandultravioletandopticallightcurves,aswellasvisual-wavelength 9 andnear-infraredspectroscopyoftheTypeII linear(IIL)supernova(SN)2013by.We 4 show that SN 2013by and other SNe IIL in the literature, after their linear decline 6 phase that start after maximum, have a sharp light curve decline similar to that 0 seen in Type II plateau (IIP) supernovae. This light curve feature has rarely been . 1 observedinotherSNeIILduetotheirrelativerarityandtheintrinsicfaintnessofthis 0 particular phase of the light curve. We suggest that the presence of this drop could 5 be used as a physical parameter to distinguish between subclasses of SNe II, rather 1 thantheirlightcurvedeclinerateshortlyafterpeak.Closeinspectionofthespectraof : v SN2013byindicate asymmetricline profilesandsignaturesofhigh-velocityhydrogen. i Late (∼ 90 days after explosion) near-infrared spectra of SN 2013by exhibit oxygen X lines, indicating significant mixing within the ejecta. From the late-time light curve, ar we estimate that 0.029 M⊙ of 56Ni was synthesized during the explosion. It is also shown that the V-band light curve slope is responsible for part of the scatter in the luminosity (V magnitude 50 days after explosion) vs. 56Ni relation. Our observations of SN 2013byandother SNe IIL throughthe onsetof the nebular phase indicate that their progenitors are similar to those of SNe IIP. Key words: supernovae: general – supernovae: SN 2013by,– galaxies: 1 INTRODUCTION TypeIIsupernovae(SNeII)havehistoricallybeendividedintotheTypeIIL(linear)andTypeIIP(plateau)subclassesbased ontheshapeoftheirlightcurvesintheweeksfollowingexplosion1 (Barbon, Ciatti & Rosino1979).SNeIIPlightcurveshave ⋆ Thispaperisbasedonobservations gatheredwith:theLCOGTnetworkoftelescopes,the6.5meterMagellanTelescopes,theSwope 1metertelescope, andtheSwifttelescope. † e–mail:[email protected] 1 Note that we are not considering Type IIn nor Type IIb inthis work, even though there are stillseveral open issues on where these objectsaresituatedintermsofSNIIdiversity. (cid:13)c 0000RAS 2 Valenti et al. aclearplateauphase,wheretheSNbrightnessstaysnearlyconstant(intheoptical)forroughly100days.Theplateauphase is mainly powered byamovinghydrogen recombination front that travels throughthehydrogen-richmaterial that is ejected and ionized during the explosion (Woosley et al. 1987). A prototypical example of a Type IIP SN is SN 1999em (Hamuy 2003). SNe IIL exhibit a linear decay that starts soon after peak brightness, They are more rare than SNe IIP and, because light curvesof Type IIL SNe decline faster than light curves of Type IIP SNe, only a handful of SNe IIL havebeen followed for more that three months after peak brightness. Prototypical example of Type IIL SNe are SN 1979C and SN 1980K (see Filippenko 1997). SNe IIL are on average more luminous than SNe IIP by ∼1.5 mag (Patat et al. 1993, 1994; Anderson et al. 2014; Sanderset al. 2014; Faran et al. 2014a). Spectroscopically, SNeIIL haveon average redder continuaand have higher oxygen to hydrogen ratio as compared to ordinary SNe IIP (Faran et al. 2014a). SNe IIL also exhibit higher expansion velocities at early times (Faran et al. 2014a), and less pronounced P-Cygni Hαprofiles (Guti´errez et al. 2014). Giventhesedifferences,thereisageneralconsensusthattheprogenitorstarsofSNeIILhavearelativelysmallamountof hydrogen in theirenvelopes, while SNeIIPmore likely originate from hydrogen-richstars. It is not currentlyknown whether theprogenitorsofSNeIILgraduallylosetheirhydrogenlayer,creatingacontinuousclassextendingfromSNeIILtoSNIIP,or ifthereisaspecificmechanismthatcreatesdistinctclassificationsashasbeenpreviouslysuggested(Barbon, Ciatti & Rosino 1979;Patat et al. 1993). This traditional classification scheme has been supported in recent compilations of SN II light curves by Arcavi et al. (2012) and Faran et al. (2014a). However, Anderson et al. (2014) and Sanderset al. (2014) have shown that the historical distinction between the two classes could be due to the small number of well observed SNe IIL. They have also suggested that the historical distinction between IIP and IIL, based on the presence of a plateau or a linear decay in the light curve, is insufficient for a complete mapping of SN II diversity. Anderson et al. (2014) suggested that if all SNe IIL are followed for a long enough time, they will exhibit, after the linear decay, a significant drop in their light curves. This would provide furtherevidencethatIIP’sandIIL’ssharethesameunderlyingphysics.However,themajority ofSNeIILrecentlypresented byAnderson et al.(2014)andSanderset al.(2014)haveonlybeenfollowed foralimited amountoftime(∼<70-80 daysfrom discovery). Herewepresentdetailedultravioletandopticalbroad-bandphotometryoftheTypeIILSN2013by,whichcoverstheflux evolutionofthisobjectforover150days.Alsopresentedarefourvisual-wavelengthspectraandthreenear-IR(NIR)spectra. The datapresented havebeen obtained by theLas CumbresObservatory Global Telescope (LCOGT) network (Brown et al. 2013), the Carnegie Supernova Project (CSP; Hamuyet al. 2006), and with the UVOT camera aboard the Swift X-ray telescope (Burrows et al. 2005; Roming et al. 2005). The organisation of this paper is as follows. In Section 2 we present the spectroscopic and photometric observations, and briefly characterise the data reduction process. Presented in Section 3 are the photometric data of SN 2013by, while in Section 4 we analyse the optical and NIR spectra of SN 2013by. In Section 5 we discuss the amount of 56Ni produced in SN 2013by, and compare our results with the previous works of Hamuy (2003) and Spiroet al. (2014). Our results are summarized in Section 7. 2 OBSERVATIONS SN2013bywasdiscoveredbytheBackyardObservatorySupernovaSearch(BOSS)on2013April23.542(UT),withcoordinates at α=16h59m02s.43, δ=−60d11’41”.8 (Parker et al. 2013). The SN is located 3′′ West and 76′′ North of the nucleus of the galaxy ESO 138−G10. ThetheNASAExtragalactic Database (NED)2 distance, corrected forLocal-group infall towards the Virgo cluster and assuming H0 =73±5 kms−1Mpc−1, is D=14.8±1 Mpc (distance modulus = 30.84 ±0.15). This value is adopted throughout this work. The closest available pre-discovery limit is not stringent (on April 1.554 UT, Parker et al. 2013), but early spectroscopic and photometric follow-up are consistent with a supernova discovered a few days or less after explosion. In what follows we adopt April 21.5 (UT), i.e., JD = 2456404±2 days, as the explosion epoch. Spectroscopic confirmation as a young SN II came from both optical and NIR observations (Parkeret al. 2013), along with a tentative classification as a Type IIL/IIn. Based on the light curve, SN 2013by, in the traditional schema, is a typical SN IIL (see Section 3). However X-ray emission has also been detected for SN 2013by with Swift (Margutti, Soderberg & Milisavljevic 2013), supporting the idea that SN 2013by may have experienced moderate interaction with circumstellar material (CSM) duringearly phases (see Section 4). Photometric monitoring in BVgriof SN2013by with theLCOGT 1 m telescope network began on 2013 April 24 (UT), andcontinuedevery2-3nights(52epochsofdatawerecollected)formorethan150days,wellafterthelightcurvesettledonto the 56Co decay tail. The LCOGT science images were reduced using a custom pipeline that performs point spread function (PSF) fitting and a low order polynomial fit to remove any background contamination (see Valenti et al. 2014). SN 2013by exploded close (but not coincident) to two point-likesources that are > 2 magnitudes fainter than ourlast detection. AdditionalimagingwasobtainedfromSwiftandtheCSP.TheSwiftdata(15epochs)wasreducedfollowingthestandard procedures described by Brown et al. (2009). While the PSF fitting technique gives a good result for ground-based data of SN 2013by, Swift magnitudes are computed using aperture photometry. By comparing initial Swift results with ground 2 http://ned.ipac.caltech.edu/ (cid:13)c 0000RAS,MNRAS000,000–000 SN 2013by 3 -22 10 -20 12 es -18es d d u14 u gnit -16gnit a a ent M16 uuvvwm22 11..50 -14ute M ar uvw1 0.5 ol pp18 U -0 bs A u -1.5 -12A B -2.0 20 g -2.5 V -3.0 -10 r -3.5 22 i -4.2 -8 0 50 100 150 Phase from maximum [days] Figure 1. Ultraviolet and optical light curves of SN 2013by from observations obtained by LCOGT, CSP and Swift. From maximum lightto∼65daysaftermaximum,SN2013byexperiencesanearlylineardecline,withadeclinerateper50daysofs50V =1.46mag.The V-band lightcurve then abruptly drops until a new, slow decline is established at ∼ +90 days from maximum, presumably associated withtheradioactivedecaytailof56Co→56Fe. basedmeasurements,hostgalaxycontaminationwasevidentintheSwiftdata.ThismotivatedaSwiftTarget ofOpportunity program (PI S.Valenti) to re-image the field of SN 2013by one year after discovery in order to properly remove background host-galaxy contamination as prescribed byBrown et al. (2009). TheCSPobtained 17epochsofscienceimages usingtheSITe3CCD cameraalong with asetof ugriBV filtersattached totheSwope1mtelescopelocatedatLasCampanasObservatory(LCO).Theseimageswerereducedusingthemethodology described in Contreras et al. (2010) and Stritzinger et al. (2011). The CSP, LCOGT and Swift photometry are in good agreementexcepttheSwiftV bandthatissystematically0.1magnitudefainterthattheLCOGTandCSPphotometry.Given the good photometric coverage in V band with CSP and LCOGT data, we did not investigate this systematic difference further. Fourepochsofvisual-wavelength spectraofSN2013by wereobtained withWifeS (Dopita et al. 2007)ontheANU2.3m Telescope, and three epochs of NIR spectra with FIRE (Simcoe et al. 2013) on the Magellan 6.5 m Baade telescope (see Table A1). The optical spectra were reduced with PyWiFes as described by Childress et al. (2013), while the FIRE spectra werereducedusingtheIDLpipelineFirehose.TheFirehosepipelineperformedthefollowing steps:flatfielding,wavelength calibration, sky subtraction, spectral tracing and extraction and flux calibration. 3 LIGHT CURVES The multi-bandlight curvesof SN2013by are shown in Figure 1, while the corresponding photometric data are tabulated in TableA5. Afterinspectingthelight curves,wesee familiar features includinga short rise(∼ 10 days)tomaximum,followed by a linear phase lasting 65 days (from maximum). After the linear phase, the light curve of SN 2013by shows a clear drop untilit sits on theradioactive decay tail of 56Co to 56Fe. In order to determine which type of SN is SN 2013by, we compare its V-band light curve with templates presented by Faran et al. (2014a) (see left panel of Figure 2). SN 2013by lies in the middle of the SN IIL templates of Faran et al. (2014a). For completeness we also compare SN 2013by with the CSP sample of SNe II published by Anderson et al. (2014) (see right panel of Figure 2). The s2 parameter, used by Anderson et al. (2014) to quantify the slope of the plateau, is the V-bandmagnitude decline per 100 days measured in thesecond part of the plateau (see black line in left panel of Figure 2). Faran et al. (2014a) use a different parameter. Specifically, they use the magnitude decline in 50 days computed between maximum light and 50 daysafter explosion (s50V)3.In ordertoavoid theproliferation of parameters tocharacterize SNeII, we adopt thes50V in thispaper.Faran et al. (2014a) defineall SNeIIwith s50V >0.5 mag as TypeIIL events.The decline rate for SN 2013by is s50V =1.46±0.06 mag. 3 ThereadershouldbeawarethatsincetherisetimeofSNeIIisoftennotwelldefinedandcantakes severaldaysfromexplosion,the s50V parameterisusuallycomputedwithinarangeof∼40–45days. (cid:13)c 0000RAS,MNRAS000,000–000 4 Valenti et al. −1 2013by 4 4 IIL (Faran 2014) 02ew 03ej IIP (Faran 2014) 0 ht g 3 3 m li mu 1 xi sV50 a 2 2 m om 2 s2 s2 e fr d 1 1 u nit 3 g a m Anderson 2014 V 4 0 2013by 0 2007od 2009bw 2008fq 5 −1 −1 −20 0 20 40 60 80 100 120 140 160 −14 −15 −16 −17 −18 −19 Days from maximum Mmax Figure 2. left panel: SN 2013by compared with template light curves from Faranetal. (2014b). The s2 parameter is the V-band magnitudes per 100 days of the second, shallower slope observed in the light curve as defined by Andersonetal. (2014). right panel: AbsoluteV-bandmagnitudeofSN2013byvs.s2,comparedtoobjectsfromAndersonetal.(2014).SN2009bw,SN2007odandSN2008fq havebeenalsoaddedforcomparison. In this schema SN 2013by should be labeled as a SN IIL. However, the presence of a drop in magnitude at the end of the hydrogen recombination phase is usually considered the defining feature of a SN IIP. Is the light curve drop-off in SN 2013by atypical for a SN IIL? To answer this question, we have compiled a large sample of SNe IIL from the literature, and systematically measured from their V-band light curves the parameter s50V. Among these, we show a subset with s50V > 1.0 (typically classified as SN IIL, as they are the fastest decliners) in Figure 3. We also show a handful of objects with 0.5 < s50V < 1.0 (orange symbols; SNe 2007od, 2007pk, 2009bw, 2009dd). If we follow the definition of Faran et al. (2014a),whereSNeIILdeclinewiths50V >0.5, SN2007od, SN2007pk andSN2009bw should also beclassified as SNeIIL. These SNe II are as luminous as SNe IIL (MIIL = −17.44 mag Li et al. 2011) (MV07od = −17.4 mag and MV09bw = −17.2 mag, Inserraet al. 2013). For comparison SNe IIP are fainter (MIIP =−15.66 mag), (Li et al. 2011). These SNe have been studiedindetailbyInserra et al.(2011)andInserra et al.(2012).TheyidentifyhighvelocityHαabsorptioninthespectra(∼ 13,000-15,000 kms−1) and suggest this is evidence for moderate interaction with CSM. Chugai, Chevalier & Utrobin (2007) haveshownthatthepresenceofhigh-velocityfeaturesinTypeIISNecanbeindeedinterpretedasinteractionbetweenrapidly expandingSN ejecta and circumstellar material (CSM). TheSNeinFigure3arepredominantlytakenfromAnderson et al.(2014)andFaran et al.(2014b),butalsoincludedare thehistoricalTypeIILSN1979C(Barbon, Ciatti & Rosino1982)(s50V =1.5mag)andSN1980K(Barbon, Ciatti & Rosino 1982) (s50V =2.1 mag). Each of theSNeIIL plotted in Figure 3 show a linear decay up until∼80–120 daysafter explosion, followed by a steep and rapid decline prior to reaching a secondary linear decline phase powered by radioactive decay. All SNe IIL that have been followed for more that 80 days from discovery show the drop in magnitude that is characteristic of SNe IIP. Based on observations of several SNe IIL, Anderson et al. (2014) noted a similar luminosity drop, however, their photometric coverage is not as dense as that presented here (particularly at late phases), so a drop could not be robustly demonstrated for most of their events. The light curve coverage of SN 2013by is such that it affords the best coverage of a late-time luminosity drop for a SN IIL. We also performed an extensive literature search for SNe IIL that do not show this light curve drop, and found that only SN 1979C may have been one such object. Actually also SN 1979C, as suggested by Anderson et al. (2014), may show a light curve drop around 50 days after explosion, though, if there, the drop would have occurred quite early and less pronounced than all the other cases. It is also worth mentioning (see Figure 3) that the light curve drop to the radioactive decay tail in SNe IIL occurs at ∼80–100 days (versus ∼ 100-140 days for SNe IIP, confirming thecorrelation between theslope of the plateau and plateau length previously reported by Anderson et al. (2014). 4 SPECTROSCOPY The visual-wavelength spectra of SN 2013by are plotted in Figure 4, while the NIR spectra are shown in Figure 5. Also includedinFigure4arecomparison spectraofSN2009bwandSN2007od,whicharefoundbythespectralclassification tool GELATO (Harutyunyanet al. 2008) to best match the spectra of SN 2013by at 16 and 34 days after shock breakout. For the first spectrum of SN 2013by, GELATO find as best fit SN 1998S and several other SNe IIn. However the fit are poor, (cid:13)c 0000RAS,MNRAS000,000–000 SN 2013by 5 1986L 1.03 2001cy 1.04 0 2005me 1.18 on 2013ej 1.20 si 2005an 1.29 o expl1 12909038Sif 11..3331 after s50V=0.5 22000069bkrl 11..3366 ays 2 22000089aawu 11..3377 10 d 22000086Kai 11..3493 o 2008fq 1.62 ve t3 s50V=1 2003ej 1.69 elati 22000013fhaf 11..7778 e r4 2008gi 1.82 d 1999em 0.25 u gnit 22000097bodw 00..7952 Ma 2009dd 0.88 V 5 2007pk 0.82 1980K 2.11 1979C 1.50 2013by 1.46 6 50 100 150 200 250 Days from explosion Figure 3. V-band light curves of a compilation of SNe II from the literature, along with SN 2013by. Each object is listed and color coded in the right panel along with its measured s50V parameter. We have chosen objects primarilywith s50V >1.0 to highlight the fastestdeclining, SN IIL-likeevents. We havealsoincludedseveral objects with0.5<s50V <1.0,whichareplotted inorange andthe prototype of Type IIP SN 1999em. Note that all of the high s50V objects have a steep drop-off from their linear ‘plateau’ phase at around∼80-120days,beforetheradioactivedecaytailpowersthelightcurve. probably for the lack of SNe IIL spectra obtained at early phases. The spectrum of SN 2009bw obtained 4 days after shock breakout shows several similarities to SN 2013by, although the latter exhibitsconspicuous Hei λ5876 (see Fig. 4). Thefirst opticalspectrum exhibitsabluecontinuumwiththeBalmer andthePaschen series clearly detected.Both Hei andOiiiaredetectedinthefirstopticalspectrumofSN2013by.AstheSNevolvestheBalmerlinesbecomemoreprominent andtypicalfeaturesofSNeIIemergeincluding:Caii(H&Kλλ3934,3968˚A),FeII(especiallylinesλλ4924,5018and5169˚A), Tiii(with many multipletsbelow 5400˚A), Oi λ7774 ˚A and theCaiiinfrared triplet λλ8498, 8542 and 8662 ˚A). The spectra at 16 and 34 days after explosion show several absorption features blueshifted with respect to photospheric Hα (marked as A, B and C in Figure 4b). While absorption feature A is consistent with Siii, absorption features B and C are probably due to Hα at 15000 kms−1 and 8000 kms−1, respectively. As mentioned in Section 3, these high-velocity features have also been identified in SN 2009bw and SN 2007od (Inserra et al. 2011, 2012), and interpreted as having an origin related to the interaction between rapidly expandingSN ejecta and circumstellar material (CSM). Besides the overall similarity with SN 2007od, SN 2013by does show one clear difference: the Oi λ7774 absorption line in thespectrum of SN 2013by at 34 daysafter the explosion is quite prominent,while it is almost absent in the spectrum of SN 2007od. Faran et al. (2014a) have recently shown that oxygen seems to be more pronounced in SNeIIL than in SNeIIP and it can be interpreted as a sign of a more massive progenitor. However, oxygen should be used carefully as a progenitor tracer for SNe IIP. Maguire et al. (2012) and Jerkstrand et al. (2012) have shown that part of the oxygen visible in Type IIP SNeis synthesized soon after the explosion, while the rest is primordial oxygen that is mixed in the envelope during the evolution of theprogenitor. Disentangling these contributionsis not easy. A comparison of the observed visual-wavelength spectrum at 34 days after the estimate shock breakout to our synow fit in Figure 6. The spectrum has been reproduced with a black body temperature of 7200 K and a photospheric velocity of 7000kms−1.Thefollowingionshasbeenusedinthesynowspectrum:HI,Srii,Scii,Caii,Oi,Feii,Nai,Tiii,andSiii.The synow spectrum well reproduces theobserved spectrum, except for the ratio of the Hα / Hβ. This is a well known problem relatedtothefactthatsynowisbasedontheunderlyingassumptionoflocalthermodynamicalequilibrium(Dessart & Hillier 2010). Plotted in Figure 5 are the 3 FIRE NIR spectra of SN 2013by compared to the spectra of other SNe II, including SNe 2002hh (Pozzo et al. 2006), 2012A (Tomasella et al. 2013) and 2012aw (Dall’Ora et al. 2014). Beside the asymmetric profile ofHydrogenlines,alsovisibleintheopticalspectra,theNIRspectraofSN2013byaresimilar tothoseofotherTypeIISNe. Of particular interest is the last (+90d) FIRE spectrum of SN 2013by, which, to our knowledge, is one of only a handful of NIR late-time spectra published to date for a Type II SN, and the first of a Type IIL SN. Pozzo et al. (2006) published a large set of infrared spectra of SN 2002hh at late time. SN 2002hh with a s50V = 0.27 mag/50 days is a Type IIP SN. The first of their spectra (+137d) is shown in the comparison,andseveraloftheirlineidentificationshaveprovideduswithguidelinesforouranalysisofSN2013by.Besidethe Brackett series which is clearly visible in thespectrum of SN 2002hh in therange 16000-18000 ˚A,most of theother lines are also visible in thelast spectrum of SN2013by. This includes, e.g., Feii,Oi, Mgi, Hei and Sii. (cid:13)c 0000RAS,MNRAS000,000–000 6 Valenti et al. velocity [km/s] OIII? 20000 10000 0 (b) (a) Hδ ] 1−Å Hγ NIII A B C 2− | | | m c Hβ 1−1.0 6100 6400 6700 6400 6700 s g Hα r HeI e 13− 0 1 [ st. 2d(04Apr) n o c + 0.5 4d2009bw x u 16d(08May) ed fl OI CaII al | | | 34d(26May) c S 35dSN2007od 0.0 119d(19Aug) 4000 5000 6000 7000 8000 9000 Rest wavelength [Å] Figure 4. Visual-wavelengths spectra of SN 2013by. Inset (a) and (b) show the region blue-wards of Hα, and highlight absorption featureslabeledasA,B,andC. +2 +2 +30 Pβ Pδ Pβ .95 1.0 1.25 1.3 1.25 1.3 2013by +2 t . s on 2012aw +14 c + ) 2013by +30 x u l F( 2012aw +23 g10 o 2012A +67 L OI SiI Brγ Fe II HeI Paschen series MgI CI + HeI 2013by +90 2002hh +137 .8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 Wavelength [µm] Figure 5. Infrared spectra of SN 2013by and several Type IIP SNe from the literature. The three inset panels show the asymmetric profileofPaschenlinesinthespectraat2days (Pβ andPδ)and30days(Pβ)aftertheshockbreakout. (cid:13)c 0000RAS,MNRAS000,000–000 SN 2013by 7 1e−14 3.5 1]− 2013by Å synow 2m−3.0 SrII FeII 1sc− TiII ScII NaID g er2.5 13− 0 1 st. [2.0 n o c + 1.5 x u Balmer serie d fl1.0 e al OI Sc CaII CaII 4000 5000 6000 7000 8000 9000 Rest wavelength [Å] Figure 6.SynowfitofthespectrumofSN2013byat34daysaftertheestimateshockbreakout. ThepresenceofOiatλ11290˚AconfirmstheOiidentificationatopticalwavelengths.AsreportedbyPozzo et al.(2006), this line is a Bowen resonance fluorescence line which requires the presence of microscopic mixing of hydrogen and oxygen in the ejecta to be excited. Hei is also clearly visible at λ10830 ˚A and λ20580 ˚A. The presence of Hei at this phase can be explained only if iron-group elements are mixed in the ejecta. The presence of Oi and Hei confirm that significant mixing has occurred within theejecta. Mgi λ15030 ˚A, is also clearly detected,while nosign of [Sii] λ16068 ˚A is apparent. 5 56NI IN SNE IIP/IIL Recent studies of SNe II have shown that the Type IIL objects are on average more luminous at peak brightness than Type IIP objects (MIIL =−17.44 and MIIP =−15.66, Li et al. 2011; see also Anderson et al. 2014; Faran et al. 2014a and Sanderset al. 2014). This is consistent with a scenario where the envelope of SNe IIL progenitors retains a much smaller amountofhydrogenthantheirSNeIIPcounterparts.Amassivehydrogen-richenvelopecausesaslowerreleaseofenergyand afaintermaximumluminosityduetoitsionizationandexpansionduringtheSNexplosion(Patat et al.1994).Acontribution to theextra luminosity of SNeIIL may also come from CSM interaction (see Section 6). An alternative explanation for the higher luminosity observed in SNe IIL is a larger amount of 56Ni synthesized during the explosion. The most appropriate way to estimate the 56Ni content is by observing the supernova after ∼100 days from explosion, when the56Codecay becomes thedominatesource of energy thatpowers thebroad-bandemission. Unfortunately SNe IIL decline 4–5 magnitudes from peak and are often too faint to be observed at this phase. However, we were able to follow SN 2013by, and a few other SNe IIL for more than 100 days, recovering both the drop from the plateau typical of SNeIIPand thesubsequentfall of thelight curveonto theradioactive tail. Using SN 1987A as reference, we estimated the amount of 56Ni produced during the explosion using the method of Spiro et al. (2014). This method consists of comparing the pseudo-bolometric light curve of these objects with the pseudo- bolometric light curve of SN 1987A (integrated in the same bands) as soon as the SN fall onto the radioactive tail, such that L MSN(Ni)=0.075× LSNM⊙. (1) 87A Armed with this methodology, we measure the 56Ni mass for 8 SNe II (see Table 1, recently published or with work in preparation).Hamuy(2003)andSpiroet al.(2014)presentedasimilaranalysisforasampleof27and17SNeII,respectively. Figure7showstherelation between56NimassandtheabsoluteV-bandmagnitudeat50daysfrom explosion forthe44SNe (Hamuy2003andSpiroet al.2014)andthe8SNefromTable1.Forasub-sampleoftheseSNe,wewereabletomeasurethe s50V parameter. These SNehavebeen plotted in Figure 7 with different colors dependingon theirs50V values (bluerpoints for larger slopes, red points for lower slopes). Hamuy (2003) have shown a clear relation between 56Ni mass and the absolute V-band magnitude (at 50 days from explosion).Spiroet al.(2014)confirmedthatfaintSNeIIPalsofollowthesamerelation.EventhoughthenumberofSNeIIL with a 56Ni estimate from theradioactive tail is small, these objects seem to follow the same relations. However, part of the scatterinthisrelationseemstoberelatedtothes50V parameter.Bluepoints(SNeIIL)sitslightlybrighterthanthegeneral trend for a given 56Ni mass. (cid:13)c 0000RAS,MNRAS000,000–000 8 Valenti et al. 0.0 0.2ays) d 0 5 g/ 0.4a m e ( 0.6e rat n cli e 0.8d e d u 1.0gnit a m V- 1.2 1.4 Figure7.AbsoluteV-bandmagnitudeat50daysfromexplosionversus56NimassforasampleofSNeII.SNeIIL(bluepoints)cluster onthetop-leftsideoftherelation.ThedataarefromHamuy(2003),Spiroetal.(2014)andthiswork. 6 SNE IIL AND CSM INTERACTION We have shown in section 3 and section 4 that, SN 2013by is similar to the class of moderately interacting Type II SNe. X-rayemission has also been detected for SN 2013by with Swift (Margutti, Soderberg & Milisavljevic 2013).They measure a 0.3-1.0 keV count-rate of 2.1 ± 0.7 cps, that assuming a simple power-law spectra model with photon index Gamma =2, translate in a flux of 1.1e-13 erg/s/cm2 (.3-10 keV). This raises several questions: Are all Type IIL SNe interacting with CSM? Do all moderately interacting Type II SNe decline like SNe IIL? It is not straight forward to generalize, but if SNe IIL are coming from progenitors that lost most of their hydrogen envelopeduringpre-SN evolution, it is more likely that these SNe(more than TypeIIP SNe)will show CSM interaction. Unfortunately,formost oftheobjects wedonot haveenoughinformation toanswerthesequestions.Itis indeed difficult toclearly separate between interacting and not interactive SNeII. At very early phases SNe IIP/IIL may be very similar to SNe IIn since the SN ejecta has not yet had time to reach and shock the CSM. SN 2008fq is a clear example of a SN II that shows sign of interaction at very early phases. It has been consideredtobeaSNIInbyTaddia et al.(2013)becauseofthecleardetectionofnarrowlinesinitsspectrabeforemaximum, while Faran et al. (2014a) include this SN in their sample of SNe IIL since no narrow lines are visible after maximum light. Photometrically SN 2008fq and SN 2013by have similar slopes after maximum even though SN 2008fq is one magnitude brighterthanSN2013by(seerightpanelofFigure2andFigure8).UnfortunatelybothlightcurvesfromFaran et al.(2014a) and Taddia et al. (2013) stop at ∼ 60 days, before the drop from the plateau would have occurred. Plotted in Figure 8 (left panel) are theabsolute V-bandmagnitudelight curveof SN2008fq andthreeother SNethat (right, top panel) show signs of interaction with CSM at early phases. 7 CONCLUSIONS WehavepresentedUVandopticalbroad-bandphotometryfortheTypeIILSN2013byrangingupto150daysafterexplosion. Our extended and dense photometric coverage confirms that all SNe IIL show a similar drop in the light curve down to the radioactive tail as is seen in SNe IIP, but at an earlier stage. If SNe IIL are followed for more than 80-100 days, they show thatSNeIIPandIILsharesimilarunderlingphysics,supportingtheideathattheseparationintwoclassesispurelynominal. Only a handful of objects that decline as fast as SNe IIL do not show the drop from the plateau, suggesting that their light curves may be powered (also) by a different source of energy (different than recombination). We suggest that the drop from theplateau (instead of theslope) should beuse as a more physicalparameter to distinguish different typesof SNeII. Wehavealsopresentedvisual-wavelengthandNIRspectraofSN2013by,andhavemadeadetailedcomparisontosimilar dataofotherSNeII.Thevisual-wavelengthspectrasuggestthatSN2013byhasexperiencedamoderateamountofinteraction between its rapidly expandingejecta and its CSM for more than onemonth after explosion. SeveralSNeII show evidenceof interaction with theCSMat early phases.Most oftheseobjects aremoderately luminousand theyshow afast V light curve decline after maximum similar to SNe IIL. A late (+90d relative to peak) NIR spectrum of SN 2013by exhibits similarities to a NIR spectrum of the Type IIP SN 2002hh. This comparison strengthens the similarity between Type IIP and IIL SNe suggesting that strong mixingoccurs in the progenitors of both varieties. We also investigate whether or not SNe IIL are on average more luminous than SNe IIP, and if so, if this is related to the amount of 56Ni synthesized during the explosion. We use the magnitude vs. 56Ni relations introduced by Hamuy (2003) (cid:13)c 0000RAS,MNRAS000,000–000 SN 2013by 9 −20 2013by 2008fq 1998S 2007pk 2008fq 1998S 2007pk −19 d) n ba−18 V e ( d−17 u nit g Ma−16 e ut ol−15 s b A −14 −13 20 40 60 80 100 120 140 640067006400670064006700 Days from explosion Hα Hα Hα Figure 8.(leftpanel)V-bandabsolutemagnitudelightcurveofSNeIIthatshowatearlyphasessignsofinteractionwithCSM;(right panel)Hαprofilessoonafterdiscovery(top) andattheendoftheplateauphase(bottom) forthreeSNeIIP/L-IIn. Table 1.MainparametersforTypeIILandIIP SNe(addedtothepreviousworks) Supernova Nickel MVa MVb50 Distance E(B−V)host E(B−V)dMW Explosion Refe. modulusc epoch SN2003hn 0.038(002) −17.40(14) −16.78(03) 32.25(12) 0.173 0.014 2452870.0(4) 1 SN2009kr 0.009(004) −16.82(30) −15.74(08) 32.09(50) 0.0 0.077 2455140.5(2) 2 SN2013by 0.029(005) −18.21(14) −17.12(11) 30.85(15) 0.0 0.195 2456404.0(2) 3 SN2013ej 0.018(006) −17.27(13) −16.61(10) 29.79(02) 0.0 0.061 2456497.4(2) 4 SN2013fs 0.057(006) −17.71(15) −16.82(04) 33.5(15) 0.0 0.035 2456571.2(1) 5 SN2014G 0.019(003) −17.46(15) −16.32(08) 31.83(02) 0.0 0.012 2456668.3(1) 6 SN2012A 0.011(001) −16.28(16) −15.63(08) 29.96(15) 0.012 0.024 2455932.5(2) 7 SN2012aw 0.056(013) −16.92(10) −16.72(10) 29.96(02) 0.028 0.058 2456002.5(1) 8 a Absolutemagnitudeatmaximum.b Absolutemagnitudeat50daysafterexplosion.c FromNED,correctedforLocal-Groupinfall ontotheVirgoclusterandassumingH0=73±5kms−1Mpc−1.d Schlegel,Finkbeiner&Davis(1998).e References: 1=Krisciunasetal.(2009);2=Fraseretal.(2009),Elias-Rosaetal.(2011);3=thiswork,4=Valentietal.(2014);5=Trematerraetal inprep;6=Yaronetalinprep.;7=Tomasellaetal.(2013);8=Dall’Oraetal.(2014). and Spiro et al. (2014), adding as extra information the slope in V-band light curve (s50V). We find that SNe IIL broadly follow these relations, but for a similar amount of 56Ni produced during the explosion, SNeIIL are on average brighter than SNe IIP. This is in agreement with the idea that SNe II with a larger ejected mass, have a slower release of energy and a fainter maximum luminosity dueto its ionization and expansion duringtheSN explosion. ACKNOWLEDGEMENTS This material is based upon work supported by NSF under grants AST–0306969, AST–0908886, AST–0607438, and AST– 1008343. M.D.S. and the CSP gratefully acknowledge generous support provided by the Danish Agency for Science and TechnologyandInnovationrealized throughaSapereAudeLevel2grant.MGacknowledgessupportfromJoinedCommittee ESO and Government of Chile 2014 and the Ministry for the Economy, Development, and Tourisms Programa Inicativa CientificaMileniothroughgrantIC12009,awardedtoTheMillenniumInstituteofAstrophysics(MAS)andFondecytRegular No. 1120601. We are grateful to Rubina Kotak who provided us spectra of SN 2002hh. This paper includes data gathered with the 6.5 meter Magellan Telescopes located at Las Campanas Observatory, Chile. This paper is based on observations made with thefollowing facilities: Swift, LCOGT. REFERENCES Anderson J. P. et al., 2014, The Astrophysical Journal, 786, 67 Arcavi I.et al., 2012, TheAstrophysical Journal, 756, L30 Barbon ., Ciatti ., Rosino ., 1979, Astronomy and Astrophysics,72, 287 Barbon, 1982, Astronomy and Astrophysics, 116, 35 (cid:13)c 0000RAS,MNRAS000,000–000 10 Valenti et al. Brown P. J. et al., 2009, The Astronomical Journal, 137, 4517 Brown T. M. et al., 2013, Publications of theAstronomical Society of the Pacific, 125, 1031 Burrows D. N.et al., 2005, Space Science Reviews, 120, 165 Childress M. J., Vogt F. P.A., Nielsen J., Sharp R. G., 2013, Astrophysicsand SpaceScience, 349, 617 Chugai N.N., 2001, Monthly Notices of theRoyal Astronomical Society,326, 1448 Chugai N.N., Chevalier R. A., Utrobin V.P., 2007, The Astrophysical Journal, 662, 1136 Contreras C. et al., 2010, The Astronomical Journal, 139, 519 Dall’Ora M. et al., 2014, TheAstrophysical Journal, 787, 139 Dessart L., Hillier D.J., 2010, Monthly Notices of theRoyalAstronomical Society, 410, 1739 Dopita M., Hart J., McGregor P., Oates P., Bloxham G., Jones D., 2007, Astrophysics and Space Science, 310, 255 Elias-Rosa N.et al., 2011, The Astrophysical Journal, 742, 6 Faran, T., Poznanski, D., Filippenko, A.V., et al. 2014, Monthly Notices of theRoyal Astronomical Society,445, 554 Faran, T., Poznanski, D., Filippenko, A.V., et al. 2014, Monthly Notices of theRoyal Astronomical Society,442, 844 Fassia A. et al., 2000, Monthly Notices of theRoyal Astronomical Society, 318, 1093 Filippenko A. V.,1997, AnnualReview of Astronomy and Astrophysics,35, 309 Fraser M. et al., 2009, The Astrophysical Journal Letters, 714, L280 Guti´errez C. P. et al., 2014, The Astrophysical Journal, 786, L15 Hamuy M., 2003, The Astrophysical Journal, 582, 905 Hamuy M. et al., 2006, Publications of theAstronomical Society of the Pacific, 118, 2 HarutyunyanA.H. et al., 2008, Astronomy and Astrophysics, 488, 383 Inserra C. et al., 2013, Astronomy & Astrophysics, 555, A142 Inserra C. et al., 2011, Monthly Notices of theRoyal Astronomical Society,417, 261 Inserra C. et al., 2012, Monthly Notices of theRoyal Astronomical Society,422, 1122 Jerkstrand A., Fransson C., Maguire K., Smartt S., Ergon M., Spyromilio J., 2012, Astronomy and Astrophysics, 546, 28 Krisciunas K.et al., 2009, TheAstronomical Journal, 137, 34 Li W. et al., 2011, Monthly Notices of the RoyalAstronomical Society, 412, 1441 Maguire K. et al., 2012, Monthly Notices of theRoyal Astronomical Society, 420, 3451 Margutti ., Soderberg ., Milisavljevic ., 2013, The Astronomer’s Telegram Parker . et al., 2013, Central Bureau Electronic Telegrams Patat F., Barbon R., Cappellaro E., Turatto M., 1994, Astronomy and Astrophysics(ISSN0004-6361), 282, 731 Patat ., Barbon ., Cappellaro ., Turatto ., 1993, Astronomy and Astrophysics SupplementSeries (ISSN0365-0138), 98, 443 Pozzo M. et al., 2006, Monthly Notices of theRoyalAstronomical Society,368, 1169 Roming P. W. A.et al., 2005, SpaceScience Reviews, 120, 95 Sanders . et al., 2014, eprint arXiv:1404.2004 Schlegel D.J., FinkbeinerD. P., Davis M., 1998, The Astrophysical Journal, 500, 525 Simcoe R.A. et al., 2013, Publications of the Astronomical Society of thePacific, 125, 270 Spiro S. et al., 2014, Monthly Notices of theRoyal Astronomical Society,439, 2873 Stritzinger M. D.et al., 2011, The Astronomical Journal, 142, 156 Taddia F. et al., 2013, Astronomy & Astrophysics, 555, A10 Tomasella L. et al., 2013, Monthly Notices of the RoyalAstronomical Society, -1, 25 Valenti S.et al., 2014, Monthly Notices of theRoyalAstronomical Society: Letters, 438, L101 Woosley S.E., Pinto P. A.,Martin P.G., WeaverT. A.,1987, The Astrophysical Journal, 318, 664 APPENDIX A: TABLES (cid:13)c 0000RAS,MNRAS000,000–000