An intermediate-luminosity-optical-transient (ILOT) model for the young stellar object ASASSN-15qi
aa r X i v : . [ a s t r o - ph . S R ] M a r MNRAS , 1–7 (2016) Preprint 23 February 2018 Compiled using MNRAS L A TEX style file v3.0
An intermediate-luminosity-optical-transient (ILOT)model for the young stellar ob ject ASASSN-15qi
Amit Kashi ⋆ and Noam Soker † Physics Department, Ariel University, Ariel, POB 3, 40700, Israel Deparment of Physics, Technion, Haifa 3200003, Israel
Accepted XXX. Received YYY; in original form ZZZ
ABSTRACT
We construct a scenario where the outburst of the young-stellar-object ASASSN-15qiis an intermediate luminosity optical transient (ILOT). In this scenario a sub-Jupiteryoung planet was tidally destructed on to a young main-sequence star. The system isyoung, therefore the radius of the planet is larger than its final value, and consequentlyits density is smaller. The lower density allows the tidal destruction of the youngSaturn-like planet on to the main-sequence star of mass ≈ . ⊙ , resulting in aformation of a disc and a gravitationally-powered ILOT. Unlike the case of the moreenergetic ILOT V838 Mon, the mass of the destructed planet is too low to inflatea giant envelope, and hence the merger remnant stays hot. If our suggested modelholds, this ILOT possesses two interesting properties: (1) its luminosity and totalenergy are below those of novae, and (2) it is not as red as other ILOTs. The unusualoutburst of ASASSN-15qi, if indeed is an ILOT, further increases the diversity ofthe already heterogeneous group of ILOTs. We mark the region on the energy-timediagram occupied by such young ILOTs. Key words: planet-star interactions — accretion, accretion discs — stars: pre-main-sequence — stars: flare
As in recent years the quality and quantity of sky surveysincrease, more attention is given to rare explosions andoutbursts in the energy gap between novae and supernovae(e.g. Mould et al. 1990; Rau et al. 2007; Ofek et al. 2008,2016; Prieto et al. 2009; Botticella et al. 2009; Smith et al.2009; Berger et al. 2009a,b; Kulkarni & Kasliwal 2009;Mason et al. 2010; Pastorello et al. 2010; Kasliwal et al.2011; Tylenda et al. 2013; Kasliwal et al. 2011;Kurtenkov et al. 2015; Tartaglia et al. 2016; Villar et al.2016; Blagorodnova et al. 2017). Those outbursts, knownas intermediate luminosity optical transients (ILOTs)form an extended family that has a number of subgroups(see Kashi & Soker 2016 for a detailed nomenclature:Intermediate-Luminous Red Transients, LBV giant erup-tions and SN Impostors, and Luminous Red Novae or RedTransients or Merger-bursts). ⋆ E-mail: [email protected] † E-mail: [email protected]
Researchers have been modeling ILOTs, or sub-groups of ILOTs, either as single-star phenomena (e.g.,Thompson et al. 2009; Kochanek 2011 for eruptivered giants and Ofek et al. 2013 for a SN impostors),or as interacting binary systems (Kashi et al. 2010;Kashi & Soker 2010b; Soker & Kashi 2011, 2012, 2013;Mcley & Soker 2014; Nandez et al. 2014; Goranskij et al.2016; Pejcha et al. 2016b; Soker 2016), including acommon envelope evolution (Retter & Marom 2003;Retter et al. 2006; Tylenda et al. 2011; Ivanova et al. 2013;Ivanova & Nandez 2016; Tylenda et al. 2013; Nandez et al.2014; Kami´nski et al. 2015b; Soker 2015; MacLeod et al.2016; Blagorodnova et al. 2017).There are two main diagrams to characterize ILOTs.One is the peak luminosity versus eruption duration (timescale; Rau et al. 2009; Kasliwal 2013), and the second di-agram is that of the total eruption energy versus eruptiontime. The latter is called the energy-time diagram , and wepresent it in Fig. 1.In the present study we deal with a subgroup of ILOTs An updated version of the energy time diagram is available at http://phsites.technion.ac.il/soker/ilot-club/ c (cid:13) A. Kashi and N. Soker timescale [days] T o t a l E n e r gy [ e r g s ] Optical Transient Stripe (OTS)
ASSASN 15qi
Novae Exploding Massive StarsSN Ia
NGC 300 OTV838 Mon M85 OT2006SN 2008SM31 RVV1309 ScoPTF10fqs NGC 3432 OT2008−9NGC 3432 OT2000SN 2009ip 2012bSN 2009ip 2012a SN 2009ip 2011 SN 2009ip TotSN 2010mc OGLE−2002−BLG−360CK Vul 1670SN 2007svM31LRN2015SN Hunt275 (PTF 13efv) NGC 4490−OT2011V838 Mon Eta Car GEEta Car LEP Cyg 1600ADV1309 Sco R71SN Hunt248
BD−Planet MergerburstsPlanet−Planet MergerburstsStar−Planet Mergerbursts M p =10M J M p =1M J M p =0.1M J NGC 6302OH231.8+4.2M1−92IRAS 22036+5306
YoungSystems
Figure 1.
Observed transient events on the energy time diagram. Blue empty circles represent the total (radiated plus kinetic) energyof the observed transients as a function of the duration of their eruptions, i.e., usually the time for the visible luminosity to decrease by 3magnitudes. The Optical Transient Stripe is populated by ILOT events that we suggest are powered by gravitational energy of completemerger events or vigorous mass transfer events (Kashi & Soker 2010b, 2016; Soker & Kashi 2016). For ILOTs that had sufficient datato create a model to calculate their total available energy, we mark it by a black asterisk above, or overlapping with, the blue circle.The total energy does not include the energy that goes to lifting the envelope and does not escape from the star. Novae models aremarked with a green line (della Valle & Livio 1995), with red crosses (Yaron et al. 2005), or with diamonds (Shara et al. 2010). The fourhorizontal lines represent planetary nebulae (PNe) and pre-PNe that might have been formed by ILOT events (Soker & Kashi 2012).Merger models of a planet with a planet/BD/star (Bear et al. 2011) are shown on the left hand side, together with models we addedfor smaller merging planet with mass 0 . J . The lower-left part (hatched in green) is our new extension for younger objects, includingASASSN-15qi (red square), where the planets are of lower density and can more easily undergo tidal destruction. that are powered by gravitational energy which is releasedfrom a complete merger process of two stars (termed Lu-minous Red Novae, or Red Transients, or Merger-bursts),such as V838 Mon (Soker & Tylenda 2003; Tylenda & Soker2006) and V1309 Sco (Tylenda et al. 2011; Nandez et al.2014; Kami´nski et al. 2015a). Differently from the objectsabove, in this study we discuss the destruction of a planeton a star, rather than a star on a star. Bear et al. (2011) propose that a V838 Mon like merger-burst can happen on smaller scales and low energies – be-tween a planet and a low mass main-sequence (MS) star,between a planet and a brown dwarf (BD) or between twoplanets. In this process the planet is tidally shredded intoa disc, and the accretion of the gas in the disc onto thestar, onto a brown dwarf, or onto another planet, leads toan outburst. According to this model, the destruction of theplanet occurs before it touches the more massive object, be-
MNRAS , 1–7 (2016) n ILOT model for the YSO ASASSN-15qi cause the density of the planet is lower than the densityof the more massive object. For a typical mass of the de-stroyed planet of ≈ J , these outbursts populate the lowerleft part of the optical transient stripe, with timescales of afew days and total energies of 10 . –10 . erg (Bear et al.2011). We note that in the triple-planet scenario proposedby Retter & Marom (2003) and Retter et al. (2006) for theoutburst of V838 Mon, the planets enter the envelope of thestar intact, and hence their scenario is different than thescenario we propose here.The thorough study by Metzger et al. (2012) further es-tablished the star-planet merger process as member of theILOT heterogeneous group. They study the interaction be-tween a Sun-like star and planets of masses of 1–10 M J , andfind that the ratio of the mean densities of the planet and thestar determines the outcome. For low enough mean densityof the planet, the interaction can lead to tidal-dissipationevent where the planet transfers mass at about steady rate tothe star. For density ratio in the range ≈ ≈ ergand the time scale to be in the order of a few weeks. The All-Sky Automated Survey for Supernovae (ASAS-SN)variability survey (Shappee et al. 2014) discovered an out-burst designated ASASSN-15qi (also referred to as 2MASSJ22560882+5831040) on JD 2,457,298 (2 October 2015).Herczeg et al. (2016) report the observational properties ofASASSN-15qi in detail. We present the light curve in V inFig. 2. The following make the ASASSN-15qi outburst in-teresting. (1) Its location among young objects, suggestingit is associated with a young stellar object (YSO). (2) Itshowed a fast brightening of 3.5 mag in the optical bands inless than 23 hours. (3) It blew a fast wind that faded as theoutburst decayed over 4-5 months.Herczeg et al. (2016) calculate the outburst radiativeenergy to be E rad ≈ × erg over a duration of 6 months.The kinetic energy might be much larger than the radiatedenergy, and it might accounts for most of the energy of theoutburst. Herczeg et al. (2016) mention that other outburstsfrom young stellar objects, such as V899 Mon and Z CMa,share some similar spectroscopic features with ASASSN-15qi. They speculated that ASASSN-15qi might be eithera mass transfer event, connected to interactions between astar and a planet on an eccentric orbit, formation of an ex-cretion disc, or be some kind of a magnetic reconnection andoutflow event.The energy and timescale of ASASSN-15qi place thisevent on the energy-time diagram just below the regionwhere Bear et al. (2011) predict the location of events wherea merger of a planet with a low-mass main-sequence startake place. This close location motivates us to propose aplanet-destruction model for the intermediate luminosity op-tical transient ASASSN-15qi.In addition to the event in 2015, Herczeg et al. (2016)mention an earlier outburst in 1976. We here concentrateon the event of 2015. If our scenario holds, then two planetshave been destructed on the star, one in 1976 and one in 2015. Surely two such events make our proposed scenariomuch rarer. Yet, there are now many known planetary sys-tems with a large number of planets and with rich variety ofproperties. Most pronounced is the detection of 7 earth-likeplanets around a very low mass star (Gillon et al. 2017). Ourproposed scenario requires a planetary system with severalplanets, as one or more planets should perturb the orbit ofthe planet to be destroyed. We note that a scenario for anILOT as a result of triple-planet collisions have been pro-posed by Retter et al. (2006) for the outburst of V838 Mon. We propose that the young stellar object outburst ASASSN-15qi is an ILOT event, and construct a planet-destructionscenario where a sub-Jupiter young planet was tidally de-stroyed onto a young main sequence star. If our proposedscenario holds, this ILOT is unusual in two aspects. Firstit is not ‘intermediate’ between the typical luminosities ofnovae and luminosities of supernovae, because its luminosityis like those of novae. Second, it does not have a red pho-tosphere as is the case for other ILOTs that are not LBVmajor outbursts, though it will appear red as a result ofextinction. Below we explain that this non-red photosphereis the result of not enough mass being available to inflate agiant envelope.The density ratio of the planet to that of the star de-termines the outcome of the merger (Metzger et al. 2012).We here study a case where the planet is destroyed outsidethe star, hence the planet density should be lower than thedensity of the star. This in turn requires the planet to beyoung and be larger than its final equilibrium size.A Saturn like planet ( M p ≃ . J ), for example,has a radius of R p ≃ .
95 R J at an age of 300 Myr(e.g., Fortney et al. 2007), compared to its final radius of ≃ .
84 R J . The radii of young planets depend on many pa-rameters. The important ones are the planet composition,the size of the core, the accretion rate onto the core, theplanet pressure profile, and the efficiency of cooling withthe presence of stellar irradiation and tidal heating (e.g.,Guillot 2005 and references therein). In a recent reviewBaruteau et al. (2016) present some modern calculations ofyoung planets, with a wide range of parameters that deter-mine their radii. At a very young age of 30 Myr these radiiare in the range R p ≃ (1 . . R pf , (1)where R pf is the final radius of the planet. We shall usethe most pessimistic value to our proposed scenario of R p =1 . J .Solar type stars contract to their final zero-age main se-quence radius in about their thermal timescale of ≈
30 Myr.More massive stars contract on a shorter time scale (e.g.,Bernasconi 1996). A star with a mass of M ∗ = 2 . M ⊙ con-tracts within a time scale of about only ≈ MNRAS , 1–7 (2016)
A. Kashi and N. Soker −40 −20 0 20 40 60 80 100 120 140 160 180 200 220 240 260131415161718 JD − 2,457,298 V ( m a g ) steep decline for ~5 daysmoderate slopered arrows: breaks in the lightcurve Figure 2.
The light curve of ASASSN-15qi from the observations in Herczeg et al. (2016). A change from a sharp decline to a moderateslope occurs about 5–6 days after the peak. The red arrows indicate times at the light curve where a break in the slope of the light curveappears. our proposed planet-destruction event, the central star hadenough time to reach the main sequence. The stellar densityis larger than the planet density at the considered time.The orbital separation at which planets are shreddedby tidal forces is given by (e.g. Nordhaus et al. 2010) R s ≈ R p (cid:18) M ∗ M p (cid:19) / = 3 (cid:18) R p . J (cid:19) (cid:18) M ∗ . ⊙ (cid:19) / (cid:18) M p . J (cid:19) − / R ⊙ , (2)where M ∗ is the mass of the star taken from Herczeg et al.(2016). This tidal-destruction radius is larger than the ra-dius of the star, and hence equation (2) implies that theyoung planet is tidally destroyed outside the star. This or-bital separation at tidal destruction is larger than the tidaldestruction separation at older ages.At the tidal destruction orbital separation given byequation (2) the Keplerian time is t K ≃ . R s , then the tidal destruction processcontinues as the planet moves further in along its orbit. Asa result of that the accretion process will be shorter even. An eccentric orbit might result from a perturbation by one ormore larger planets. In general, if we are to explain also the1976 outburst, we require that the systems contains severalplanets in unstable orbits.The mass accreted onto the star releases a gravitationalenergy that amounts to E acc ≃ . GM ∗ M acc R ∗ = 2 . × (cid:18) M ∗ . ⊙ (cid:19) (cid:18) M acc − M ⊙ (cid:19) (cid:18) R ast ⊙ (cid:19) − erg . (3)In scaling equation (3) we used the well studied ILOT V838Mon. Soker & Tylenda (2006) estimate the kinetic energy ofthe ejected mass in V838 Mon to be E kin ≈ erg, aboutan order of magnitude larger than the radiated energy. Theejected mass in ASASSN-15qi could have carried a similaramount of kinetic energy. For example, an ejected mass of M ej ≈ . M acc ≈ − M ⊙ that is ejected at the observedvelocity of v ≈ − (Herczeg et al. 2016) has akinetic energy of E kin = 10 erg.As suggested by Bear et al. (2011), the accreted massmay form an accretion disc or an accretion belt around thestar. The disc that is formed by such a violent process is not MNRAS , 1–7 (2016) n ILOT model for the YSO ASASSN-15qi expected to be flat. We will therefore scale with H/R a = 0 . H is the thickness of the disc at a distance of R a fromof the center of the star. The accretion time t acc of mass fromthe accretion disc should be longer than the viscous timescale for the gas in the disc to lose its angular momentum t acc > ∼ t visc ≃ R a ν ≃ (cid:16) α . (cid:17) − (cid:18) H/R a . (cid:19) − × (cid:18) C s /v φ . (cid:19) − (cid:18) R a ⊙ (cid:19) / (cid:18) M ∗ . ⊙ (cid:19) − / days , (4)where in the equation above C s is the sound speed, α is thedisc viscosity parameter, v φ is the Keplerian velocity, and ν = α C s H is the viscosity of the disc.The light curve of ASASSN-15qi shows that for aboutfive days the decline in luminosity is very steep. Only afterfive days the light curve decline slope becomes moderate.This might be related to the viscosity time scale of the discas we derived in equation (4).Most ILOTs have a red photosphere after the outburstbecause the envelope expands to large dimensions. In thosecases the envelope of the merger product has a structureof an asymptotic-giant-branch (AGB) star. In V838 Mon,for example, the gas of the destroyed low mass main se-quence star inflated a giant envelope with an envelope massof M env > . M ⊙ (Tylenda & Soker 2006).The ASASSN-15qi event was not red. In our proposedmodel of tidally destroyed planet, the bluer event is ex-plained by the small mass of the destroyed planet. Only avery small mass was potentially available to inflate an enve-lope and make the star a giant. Although the luminosity ofASASSN-15qi is similar to an AGB or an upper red-giant-branch (RGB) star, the envelope mass was too low to inflatean extended envelope. When the envelope mass of AGB starsdecreases to M env < ∼ . ⊙ they cease to expand. Whentheir envelope mass further decreases to several × − M ⊙ the envelope radius decreases to only several R ⊙ (e.g.,Soker 1992). As mass is unavailable to form an envelope,the merger product stayed small and hot, and not muchdust was formed.To summarize this point, this ILOT, although might becategorized as a luminous red nova by the properties of themerger process (see Kashi & Soker 2016), did not becomered since a low mass planet was the source of the accretedmass.An even more speculative scenario that might lead tosimilar ILOT properties can be constructed by using a moremassive young planet, such as a young Jupiter-like planet.In this scenario the young Jupiter-like planet has an eccen-tric orbit that brings it very close to the star at periastronpassages. Yet, this close distance is far enough that only thelow-density outer layers of the planet are removed. Only amass of ≈ − M ⊙ is removed from the planet and accretedon to the star. The planet itself survives the encounter. Thisspeculative suggestion should be studied in the future witha three-dimensional hydrodynamical code of binary interac-tion. In this scenario, the 1976 event was a previous peri-astron passage of the Jupiter-like planet, and this scenariopredicts another outburst in 2054. We now compare ASASSN-15qi with other ILOT events byplacing it on the energy time diagram (Fig. 1).For the total energy of the 2015 event we take E tot =10 –10 erg, as we explained in section 2. The energy ofthe ASASSN-15qi 2015 event is well below the energy of low-energy novae and below the optical transient stripe (markedin blue on Fig. 1). This goes also to the mean ratio of theluminosity to the Eddington luminosity of this eventΓ Edd = ( E tot /t ) κ πGM ∗ c ≃ . . , (5)where κ is the opacity, and c and G have their usual mean-ings. This ratio is much smaller that the typical values ofΓ Edd , OTS ≃ . ≈ ≈ , ,
60 and 90 days after the peakthat are not seen in other ILOTs (see Fig. 2.We notice the following interesting property of thelightcurve. There is an exponential decay of 1 magnitudein 6 days, e.g., before the first break. There is a constantdecay slope in the lightcurve after 90 days post-maximum.If we extrapolate this part backward in time, we find that itintersects the original lightcurve at 20 days post-maximum,at the location of one of the breaks. This might indicate thatthe decline has two main phases, with a transition around20 days post-maximum. This complicated lightcurve ham-pers us from determining an accurate value for the declinetime of this ILOT. We simply take this time scale to be inthe range of t = 6–90 days, and mark ASASSN-15qi at atimescale of 20 days in the energy time diagram (Fig. 1).Following our proposed planet-star merger scenario forthe 2015 event of ASASSN-15qi, we now extend the planet-merger region on the energy time diagram beyond the regionpresented originally by Bear et al. (2011). This new region(hatched-green region in Fig. 1) requires young systems withages of less than about 0 . We examined the possibility that the unusual outburst ofthe YSO ASASSN-15qi is an ILOT event, similar in manyrespects to V838 Mon, but much fainter and of lower totalenergy. As in the model of V838 Mon, the erupting systemwas young, but unlike the model for V838 Mon, we here
MNRAS , 1–7 (2016)
A. Kashi and N. Soker suggested that the secondary object that was tidally de-stroyed onto the primary main-sequence star was a Saturn-like planet rather than another low mass main-sequence star.The young age in both systems is crucial. The reason isthat along most of the main-sequence the density decreaseswith increasing stellar mass. Therefore, low mass stars onthe main-sequence will not suffer a tidal destruction by moremassive primary main-sequence stars. This holds for sub-stellar secondary objects as well. In young systems, the timescale on which the object shrinks to its final radius is longerfor smaller mass. Therefore, young low-mass objects havelower density than their final density, while the much moremassive star can already be very close to its zero-age mainsequence density. The density of the low mass secondaryobject can be low enough to allow tidal destruction if itcomes close enough to the primary star.Very low-mass main-sequence stars and brown dwarfs inold systems have higher densities than more massive mainsequence stars, and they can tidally destroy old planets, aswell as young ones. This is indicated by the three lines inthe lower left of Fig. 1.The destruction of young planets on stars results inweak ILOTs, as we suggest here for ASASSN-15qi. Suchevents occupy the lower left part of the energy time diagram(hatched-green region in Fig. 1). The typical luminosity ofthese events is not necessarily above novae, and they do nothave a red atmosphere. So they are not really intermediatebetween the luminosities of novae and supernovae, and theyare not internally red.The difference in the mass of the destroyed object be-tween V838 Mon and ASASSN-15qi brings another signifi-cant difference in the properties of the merger product. Theluminosity of the outburst in both cases is in the range ofgiant stars. In the case of V838 Mon, part of the mass ofthe destroyed secondary star inflated a huge envelope. Inour proposed scenario for the 2015 outburst of ASASSN-15qi, the mass that is available to inflate an envelope is < .
001 M ⊙ . As is known from the evolution of post-AGBstars, this mass is too low to build a large envelope. Thisis the reason that ASASSN-15qi had a relative small radiusduring the event, and hence it was relatively hot and notred. ILOTs constitute an heterogeneous group of objectsthat covers a large area in the energy-time diagram or theluminosity-time diagram. We here suggested that this group,that is powered by gravitational energy in binary systems,is more heterogeneous and covers a larger area in these dia-grams than the conventional view until now was. REFERENCES
Baruteau, C., Bai, X., Mordasini, C., & Molli`ere, P. 2016, SpaceSci. Rev.Bernasconi, P. A. 1996, A&AS, 120, 57Bear, E., Kashi, A., & Soker, N. 2011, MNRAS, 416, 1965Berger, E., Foley, R., & Ivans, I. 2009a, Astron., Tel., 2184, 1Berger, E., Soderberg, A. M., Chevalier, R. A., et al. 2009b, ApJ,699, 1850Blagorodnova, N., Kotak, R., Polshaw, J., et al. 2017, ApJ, 834,107Botticella, M. T., Pastorello, A., Smartt, S. J., et al. 2009, MN-RAS, 398, 1041 della Valle, M., & Livio, M. 1995, ApJ, 452, 704Fortney, J. J., Marley, M. S., & Barnes, J. W. 2007, ApJ, 659,1661Gillon, M., Triaud, A. H. M. J., Demory, B-O., et al. Nature 542,456Goranskij, V. P., Barsukova, E. A., Valeev, A. F., et al. 2016,arXiv:1609.00731Guillot, T. 2005, Annual Review of Earth and Planetary Sciences,33, 493Herczeg, G. J., Dong, S., Shappee, B. J., et al. 2016, ApJ, 831,133Ivanova, N., Justham, S., Avendano Nandez, J. L., & Lombardi,J. C. 2013, Science, 339, 433Ivanova, N., & Nandez, J. L. A. 2016, MNRAS, 462, 362Kami´nski, T., Menten, K. M., Tylenda, R., Hajduk, M., Patel,N. A., & Kraus, A. 2015, Nature, 520, 322Kami´nski, T., Mason, E., Tylenda, R., & Schmidt, M. R. 2015,A&A, 580, A34Kankare E., Kotak, R., Pastorello, A. et al. 2015 A&A Letters,submittedKashi, A. 2010, MNRAS, 405, 1924Kashi, A., Frankowski, A., & Soker, N. 2010, ApJ, 709, L11Kashi, A., Soker, N., & Moskovitz, N. 2013, MNRAS, 436, 2484Kashi, A., & Soker, N. 2010b, arXiv:1011.1222Kashi, A., & Soker, N. 2016, Research in Astronomy and Astro-physics, 16, 014Kasliwal, M. M. 2013, Binary Paths to Type Ia Supernovae Ex-plosions, 281, 9Kasliwal, M. M., et al. 2011, ApJ, 730, 134Kochanek, C. S. 2011, ApJ, 741, 37Kulkarni, S. R., & Kasliwal, M. M. 2009, astro2010: The Astron-omy and Astrophysics Decadal Survey, 2010, 165Kurtenkov, A. A., Pessev, P., Tomov, T., et al. 2015, A&A, 578,L10Mason, E., Diaz, M., Williams, R. E., Preston, G., & Bensby, T.2010, A&A, 516, A108Mcley, L., & Soker, N. 2014, MNRAS, 440, 582MacLeod, M., Macias, P., Ramirez-Ruiz, E., Grindlay, J., Batta,A., & Montes, G. 2016, arXiv:1605.01493Metzger, B. D., Giannios, D., & Spiegel, D. S. 2012, MNRAS,425, 2778Mould, J., Cohen, J., Graham, J. R., et al. 1990, ApJ, 353, L35Nandez, J. L. A., Ivanova, N., & Lombardi, J. C., Jr. 2014, ApJ,786, 39Nordhaus, J., Spiegel, D. S., Ibgui, L., Goodman, J., & Burrows,A. 2010, MNRAS, 408, 631Ofek, E. O., Kulkarni, S. R., Rau, A., et al. 2008, ApJ, 674, 447Ofek, E. O., Sullivan, M., Cenko, S. B, et al. 2013, Nature494, 65Ofek, E. O., Cenko, S. B., Shaviv, N. J., et al. 2016, ApJ, 824, 6Pastorello, A., et al. 2010, MNRAS, 408, 181Pejcha, O., Metzger, B. D., & Tomida, K. 2016a, MNRAS, 455,4351Pejcha, O., Metzger, B. D., & Tomida, K. 2016b, MNRAS, 461,2527Prieto, J. L., Sellgren, K., Thompson, T. A., & Kochanek, C. S.2009, ApJ, 705, 1425Rau, A., Kulkarni, S. R., Ofek, E. O., & Yan, L. 2007, ApJ, 659,1536Rau, A., Kulkarni, S. R., Law, N. M., et al. 2009, PASP, 121,1334Retter, A., & Marom, A. 2003, MNRAS, 345, L25Retter, A., Zhang, B., Siess, L., & Levinson, A. 2006, MNRAS,370, 1573Shara, M. M., Yaron, O., Prialnik, D., Kovetz, A., & Zurek, D.2010, ApJ, 725, 831Shappee, B., Prieto, J., Stanek, K. Z., et al. 2014, American As-tronomical Society Meeting Abstracts , 1–7 (2016) n ILOT model for the YSO ASASSN-15qi Smith, N., Ganeshalingam, M., Chornock, R., et al. 2009, ApJ,697, L49Soker, N. 1992, ApJ, 389, 628Soker, N. 2015, ApJ, 800, 114Soker, N. 2016, New Astron., 47, 16Soker, N., & Kashi, A. 2011, arXiv:1107.3454Soker, N., & Kashi, A. 2012, ApJ, 746, 100Soker, N., & Kashi, A. 2013, ApJ, 764, L6Soker, N., & Kashi, A. 2016, MNRAS, 462, 217Soker, N., & Tylenda, R. 2003, ApJ, 582, L105Soker, N., & Tylenda, R. 2006, MNRAS, 373, 733Thompson, T. A., Prieto, J. L., Stanek, K. Z., Kistler, M. D.,Beacom, J. F., Kochanek, C. S. 2009, ApJ, 705, 1364Tartaglia, L., Pastorello, A., Sullivan, M., et al. 2016, MNRAS,459, 1039Tylenda, R., Hajduk, M., Kami´nski, T., et al. 2011, A&A, 528,A114Tylenda, R., Kami´nski, T., Udalski, A., et al. 2013, A&A, 555,A16Tylenda, R., & Soker, N. 2006, A&A, 451, 223Villar, V. A., Berger, E., Chornock, R., et al. 2016, ApJ, 830, 11Yaron, O., Prialnik, D., Shara, M. M., & Kovetz, A. 2005, ApJ,623, 398MNRAS000