AT 2019qyl in NGC 300: Early Outflow Collisions for a Very Fast Nova in a Symbiotic Binary
Jacob E. Jencson, Jennifer E. Andrews, Howard E. Bond, Viraj Karambelkar, David J. Sand, Schuyler D. van Dyk, Nadejda Blagorodnova, Martha L. Boyer, Mansi M. Kasliwal, Ryan M. Lau, Shazrene Mohamed, Robert Williams, Patricia A. Whitelock, Rachael C. Amaro, K. Azalee Bostroem, Yize Dong, Michael J. Lundquist, Stefano Valenti, Samuel D. Wyatt, Jamie Burke, Kishalay De, Saurabh W. Jha, Joel Johansson, César Rojas-Bravo, David A. Coulter, Ryan J. Foley, Robert D. Gehrz, Joshua Haislip, Daichi Hiramatsu, D. Andrew Howell, Charles D. Kilpatrick, Frank J. Masci, Curtis McCully, Chow-Choong Ngeow, Yen-Chen Pan, Craig Pellegrino, Anthony L. Piro, Vladimir Kouprianov, Daniel E. Reichart, Armin Rest, Sofia Rest
DDraft version February 24, 2021
Typeset using L A TEX twocolumn style in AASTeX63
AT 2019qyl in NGC 300: Early Outflow Collisions for a Very Fast Nova in a Symbiotic Binary ∗† Jacob E. Jencson, Jennifer E. Andrews, Howard E. Bond,
2, 3
Viraj Karambelkar, David J. Sand, Schuyler D. van Dyk, Nadejda Blagorodnova, Martha L. Boyer, Mansi M. Kasliwal, Ryan M. Lau, Shazrene Mohamed,
8, 9, 10
Robert Williams,
11, 3
Patricia A. Whitelock,
8, 9
Rachael C. Amaro, K. Azalee Bostroem, Yize Dong, Michael J. Lundquist, Stefano Valenti, Samuel D. Wyatt, Jamie Burke,
13, 14
Kishalay De, Saurabh W. Jha, Joel Johansson, C´esar Rojas-Bravo, David A. Coulter, Ryan J. Foley, Robert D. Gehrz, Joshua Haislip, Daichi Hiramatsu,
13, 14
D. Andrew Howell,
13, 14
Charles D. Kilpatrick, Frank J. Masci, Curtis McCully,
13, 14
Chow-Choong Ngeow, Yen-Chen Pan, Craig Pellegrino,
13, 14
Anthony L. Piro, Vladimir Kouprianov,
18, 21
Daniel E. Reichart, Armin Rest,
3, 22 andSofia Rest Steward Observatory, University of Arizona, 933 North Cherry Avenue, Tucson, AZ 85721-0065, USA Department of Astronomy & Astrophysics, Pennsylvania State University, University Park, PA 16802, USA Space Telescope Science Institute, 3700 San Martin Dr., Baltimore, MD 21218, USA Division of Physics, Mathematics and Astronomy, California Institute of Technology, Pasadena, CA 91125, USA IPAC, California Institute of Technology, 1200 E. California Blvd, Pasadena, CA 91125, USA Department of Astrophysics/IMAPP, Radboud University, Nijmegen, The Netherlands Institute of Space & Astronautical Science, Japan Aerospace Exploration Agency, 3-1-1 Yoshinodai, Chuo-ku, Sagamihara, Kanagawa252-5210, Japan South African Astronomical Observatory, PO Box 9, 7935 Observatory, South Africa Department of Astronomy, University of Cape Town, Private Bag X3, Rondebosch 7701, South Africa National Institute for Theoretical Physics (NITheP), KwaZulu-Natal, South Africa Department of Astronomy and Astrophysics, University of California, Santa Cruz, CA 95064, USA Department of Physics and Astronomy, University of California, 1 Shields Avenue, Davis, CA 95616-5270, USA Department of Physics, University of California, Santa Barbara, CA 93106-9530, USA Las Cumbres Observatory, 6740 Cortona Dr, Suite 102, Goleta, CA 93117-5575, USA Department of Physics and Astronomy, Rutgers, the State University of New Jersey, 136 Frelinghuysen Road, Piscataway, NJ 08854,USA The Oskar Klein Centre, Department of Physics, AlbaNova, Stockholm University, SE 10691 Stockholm, Sweden Minnesota Institute for Astrophysics, School of Physics and Astronomy, University of Minnesota, 116 Church Street SE, Minneapolis,MN 55455, USA Department of Physics and Astronomy, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA Graduate Institute of Astronomy, National Central University, 300 Jhongda Road, 32001 Jhongli, Taiwan The Observatories of the Carnegie Institution for Science, 813 Santa Barbara St., Pasadena, CA 91101, USA Central (Pulkovo) Observatory of the Russian Academy of Sciences, 196140, 65/1 Pulkovskoye Ave., Saint Petersburg, Russia The Johns Hopkins University, Baltimore, MD 21218, USA
ABSTRACTNova eruptions, thermonuclear explosions on the surfaces of white dwarfs (WDs), are now recognizedto be among the most common shock-powered transients. We present the early discovery and rapidultraviolet (UV), optical, and infrared (IR) temporal development of AT 2019qyl, a recent nova inNGC 300. The light curve shows a rapid rise lasting (cid:46) M V = − . M WD (cid:38) . M (cid:12) . We present an unprecedented view of the early spectroscopic evolution of suchan event. Three spectra prior to the peak reveal a complex, multi-component outflow giving rise to Corresponding author: Jacob E. [email protected] ∗ This paper includes data gathered with the 6.5 m Magellan Tele-scopes located at Las Campanas Observatory, Chile. † Based on observations made with the NASA/ESA
Hubble SpaceTelescope , obtained from the data archive at the Space TelescopeScience Institute. STScI is operated by the Association of Uni-versities for Research in Astronomy, Inc., under NASA contractNAS 5-26555. a r X i v : . [ a s t r o - ph . S R ] F e b Jencson et al. internal collisions and shocks in the ejecta of an He/N-class nova. We identify a coincident IR-variablecounterpart in the extensive pre-eruption coverage of the transient location, and infer the presence ofa symbiotic progenitor system with an O-rich asymptotic-giant-branch donor star, as well as evidencefor an earlier UV-bright outburst in 2014. We suggest that AT 2019qyl is analogous to the subset ofGalactic recurrent novae with red-giant companions such as RS Oph and other embedded nova systemslike V407 Cyg. Our observations provide new evidence that internal outflow collisions likely play animportant role in generating the shock-powered emission from such systems. INTRODUCTIONNovae are a class of cataclysmic variables (CVs) whoseeruptions are the result of a thermonuclear runaway(TNR) on the surface of a white dwarf (WD) accretinghydrogen-rich material from a non-degenerate compan-ion (Gallagher & Starrfield 1978). Novae are among themost common explosive thermonuclear transients (e.g.,Darnley et al. 2006; Shafter 2017; De et al. 2021), andgreatly contribute to the chemical enrichment of galax-ies (e.g., Gehrz 1988, and see Gehrz et al. 1998; Jos´eet al. 2006 for reviews). Nova eruptions are also uniqueprobes of the underlying CV population, which is key tounderstanding mass transfer in binaries across a range ofmasses and evolutionary stages (e.g., Townsley & Bild-sten 2005; Nelemans et al. 2016).While novae have been studied in earnest for morethan a century (e.g., Ritchey 1917; Payne-Gaposchkin1964), our understanding of their eruptions remains in-complete. The last decade has witnessed substantialrenewed interest in complex mass ejections and internalshocks in multi-component nova outflows (see Chomiuket al. 2020 for a recent review) beginning with the dis-covery of gamma-ray emission from novae by NASA’s
Fermi Gamma-Ray Space Telescope (Abdo et al. 2010;Ackermann et al. 2014; Cheung et al. 2016). Buildingon prior lines of evidence, recent observations, includ-ing correlations between gamma-ray and optical lightcurve peaks (Metzger et al. 2014; Li et al. 2017; Aydiet al. 2020a), hard X-rays ( (cid:38) <
100 yr(Schaefer 2010), although this is almost certainly a se-lection effect based on the time for which reliable astro-nomical records are available. Essentially all novae areexpected to recur, with the predicted recurrence timesfor some systems exceeding 10 years (Yaron et al. 2005;Wolf et al. 2013).The recurrence time of a nova is set by the time tobuild up sufficient accreted mass to trigger the TNR—primarily a function of the accretion rate and the WDmass ( M WD ). The shortest recurrence times occur onmassive WDs with lower critical envelope masses owingto their larger surface gravities and pressures. Indeed,the RNe for which WD masses have been reliably mea-sured are near the Chandrasekhar mass ( (cid:38) M (cid:12) ; Os-borne et al. 2011; Page et al. 2015). As such, RNe havelong been considered possible progenitors of Type Ia su-pernovae (SNe; e.g., della Valle & Livio 1996; Kato &Hachisu 2012; Maoz et al. 2014; Schaefer 2010; Starr-field et al. 1985), especially the extragalactic popula-tion of rapid recurrent novae discovered in recent years(Darnley & Henze 2020). Systems like the remarkableM31N 2008-12a (Henze et al. 2015a,b; Darnley et al.2014; Darnley 2017; Tang et al. 2014), with a recur-rence timescale (cid:46) yr (Kato et al. 2014).Owing to their low envelope masses on explosion, RNeare also among the fastest-evolving novae. Thus, ob-taining early observations of RNe that can reveal the T 2019qyl: Nova in NGC 300 DISCOVERY AND FOLLOW-UPOBSERVATIONS2.1.
DLT40 discovery in NGC 300
AT 2019qyl was discovered on UT 2019 Septem-ber 26.21 (MJD 58752.21) by the Distance <
40 Mpcsub-day cadence supernova (SN) search (DLT40; seeTartaglia et al. 2018 for survey details, and Bostroemet al. 2020 for recent improvements in our transient de-tection and triggering algorithms). The unfiltered dis-covery data were taken with the 0.4 m Panchromatic Robotic Optical Monitoring and Polarimetry Telescopes(PROMPT) PROMPT5 telescope at the Cerro TololoInter-American Observatory (CTIO) operated by theSkynet telescope network (Reichart et al. 2005). Itsdiscovery magnitude was 17 . , Hen-den & Munari 2014) r band. An earlier DLT40 non-detection at ≈ . r = 17 . Swift
Ob-servatory (hereafter
Swift ; Gehrels et al. 2004) in placeto acquire high-cadence early UV light curves of nearbytransients, with the first data arriving ≈ ≈ I . However, as described in moredetail in Section 3.5, our spectroscopic analysis confirmsthis source as a nova.At a position of 00 h m . s , − ◦ (cid:48) . (cid:48)(cid:48) . (cid:48)
54 from the galaxy’s center (see location in Fig-ure 1). Throughout this work, we assume a distancemodulus for NGC 300 of ( m − M ) = 26 . ± . D = 1 .
81 Mpc), based on the most recently avail-able Cepheid measurement by Bhardwaj et al. (2016).We adopt a value for the Galactic extinction toward Jencson et al.
NGC 300 of E ( B − V ) = 0 .
01 mag, based on the Schlafly& Finkbeiner (2011) recalibration of the Schlegel et al.(1998) dust maps, and assume a standard (Fitzpatrick1999) reddening law with R V = 3 . Imaging observations
We obtained a sequence of images with the DLT40PROMPT5 0.4-m telescope at CTIO. The PROMPT5telescope has no filter (‘Open’), which we calibrate to r band (see Tartaglia et al. 2018, for further reduc-tion details). Immediately after discovery, we beganan intense photometric campaign with the Las Cum-bres Observatory global telescope network (Brown et al.2013) in the U BV gri bands with the Sinistro camerason the 1-m telescopes at CTIO (Chile), Siding Spring(Australia), and Sutherland (South Africa). These datawere taken as part of the Global Supernova Project(GSP), as well as PI-led programs ( griz s ; OPTICON19B-053; PI N. Blagorodnova). The images were re-duced with the Beautiful Algorithms to Normalize Zil-lions of Astronomical Images (BANZAI) pipeline (Mc-Cully et al. 2018). For GSP data, we performed PSF-fitting photometry without template subtraction using lcogtsnpipe (Valenti et al. 2016), a PyRAF-based re-duction pipeline. BV and gri -band data were calibratedto Vega and AB magnitudes, respectively, using APASSDR9 (Henden et al. 2016) catalog stars in the images.For U -band data, we used Landolt standard-star ob-servations taken on the same night by the same tele-scope. For the additional Las Cumbres data in the griz s bands, aperture photometry was performed usinga custom Python-based pipeline and calibrated to theSkyMapper catalog (Wolf et al. 2018). As the sourceis relatively isolated in NGC 300, both methods of pho-tometry give very similar results.The Swope 1-m telescope at Las Campanas Observa-tory (LCO) was used for uBV gri observations with theDirect 4k ×
4k imager. All bias-subtraction, flat-fielding,image stitching, registration, and photometric calibra-tion were performed using photpipe (Rest et al. 2005)as described in Kilpatrick et al. (2018). Similarly,
BV gri imaging was taken with the Lulin One-meter telescopein Taiwan, and was reduced with standard IRAF tasks;final image calibration was done utilizing the Swope re-duction pipeline. For these images, aperture photome-try was performed for AT 2019qyl on the reduced imageswithout image differencing.UV and optical images were obtained during the earlyportion of the light curve with the Ultraviolet/Optical https://github.com/nblago/utils/tree/master/src/photometry telescope (UVOT; Roming et al. 2005) on board Swift .The data were downloaded from the NASA
Swift
DataArchive , and the images were reduced using standardsoftware distributed with HEAsoft . Photometry wasperformed for all the uvw uvm uvw U , B , and V -band images using a 3 . (cid:48)(cid:48) Spitzer Space Telescope (Werner et al.2004; Gehrz et al. 2007) in the 3.6 and 4.5 µ m imag-ing channels ([3.6] and [4.5]) between 2014 and the endof 2020 during regular monitoring of NGC 300 by theSPitzer Infrared Intensive Transients Survey (SPIRITS;PI: M. Kasliwal; PIDs 10136, 11063, 13053, 14089) andin observations targeting the ultraluminous X-ray sourceNGC 300 ULX1 (PI: R. Lau; PID 14270). The post-basic calibrated data level images were downloaded fromthe Spitzer
Heritage Archive and Spitzer
Early ReleaseData Service and processed through an automatedimage-subtraction pipeline (for survey and pipeline de-tails see Kasliwal et al. 2017; Jencson et al. 2019). Forreference images, we used the Super Mosaics consistingof stacks of images obtained between 2003 November 21and 2007 December 29. We performed aperture photom-etry on our difference images adopting the appropriateaperture corrections from the IRAC instrument hand-book and following the method for robust estimate ofthe photometric uncertainties as described in Jencson(2020). We converted our flux measurements to Vega-system magnitudes using the zero-magnitude fluxes pre-sented for each IRAC channel in the IRAC instrumenthandbook.We executed Target of Opportunity (ToO) observa-tions with the Hubble Space Telescope ( HST ) Wide FieldCamera 3 (WFC3) UVIS channel in subarray mode inF555W (23 frames, 690 s total exposure; PI S. Van Dyk;PID GO-15151) on 2020 January 26.06 with the pri-mary goal of obtaining a precise position for AT 2019qyl https://heasarc.gsfc.nasa.gov/cgi-bin/W3Browse/swift.pl https://heasarc.gsfc.nasa.gov/docs/software/heasoft/ https://sha.ipac.caltech.edu/applications/Spitzer/SHA/ http://ssc.spitzer.caltech.edu/warmmission/sus/mlist/archive/2015/msg007.txt Super Mosaics are available as
Spitzer
Enhanced ImagingProducts through the NASA/IPAC Infrared Science Archive:https://irsa.ipac.caltech.edu/data/SPITZER/Enhanced/SEIP/overview.html http://irsa.ipac.caltech.edu/data/SPITZER/docs/irac/iracinstrumenthandbook/ T 2019qyl: Nova in NGC 300 Figure 1.
Pre- and post-outburst imaging of the site of AT 2019qyl in NGC 300. The upper, leftmost panel shows an archival,color-composite image of the host ( B in blue, V in green, and R and H α in red; credit MPG/ESO). The location of AT 2019qylalong a northern spiral arm of the galaxy is shown in more detail in the 20 (cid:48)(cid:48) × (cid:48)(cid:48) middle, rightmost zoom-in panel based onBaade/FourStar NIR images of NGC 300 from 2011 October 6 ( J in blue, H in green, and K s in red). There is a relativelyisolated point source consistent with the position of AT 2019qyl detected in all three NIR filters. Above in the top right corner,we show the same region in the archival Spitzer /IRAC [4.5] Super Mosaic (2003–2007 stack), where we also identify a coincidentmid-IR (MIR) precursor source. The bottom rightmost panel shows a zoom-in view of the immediate 4 (cid:48)(cid:48) × (cid:48)(cid:48) region aroundthe transient in the archival HST
ACS/WFC images from 2002 July 19 in three filters (F435W in blue, F555W in green, andF814W in red). At the precise location of the transient, determined from our 2020 January 26 follow-up
HST /WFC3 UVISobservations in F555W (bottom, leftmost panel in blue outline), we identified a very red source in the archival
HST framesdetected in F814W, but not in F555W (bottom, center-right panel) or F435W. In the WFC3/UVIS F225W image from 2014December 1 (bottom, center-left panel) we also identify a UV point source consistent with the transient position, approximatelyfive years before the nova eruption.
Jencson et al. in comparison with archival imaging (see Section 4 fordetails). To obtain photometry of AT 2019qyl, we pro-cessed the data with
Dolphot (Dolphin 2000, 2016), firstrunning the individual frames corrected for charge trans-fer efficiency through
AstroDrizzle (Gonzaga et al.2012), to flag cosmic-ray hits. We adopt a weightedaverage of the
Dolphot measurements from 21 frames ofF555W = 22 . ± .
07 mag (Vega).We show our multi-band light curves of the transientin Figure 2, including only measurements with errors of < Spectroscopy
We obtained a sequence of five optical spectra be-tween 2019 September 26.29 and 2019 October 4.04,spanning from 1 . − grism (OPTICON 19B-053; PI N.Blagorodnova). The spectra were reduced using stan-dard techniques including wavelength calibration witharc-lamp spectra and flux calibration using spectropho-tometric standard stars. In particular: 1) we used stan-dard tasks in the Gemini IRAF package for the GMOSspectrum following procedures provided in the GMOSData Reduction Cookbook ; 2) we followed the proce-dures of Valenti et al. (2014) for the FLOYDS spectra,and 3) used the Python package PypeIt for the AL-FOSC spectrum (Prochaska et al. 2020a,b). Our opticalspectra are shown in Figure 3. http://ast.noao.edu/sites/default/files/GMOS Cookbook/ https://pypeit.readthedocs.io/en/latest/ A late-time NIR spectrum was obtained with theNear-Infrared Echellette Spectrometer (NIRES) onthe 10 m Keck 2 Telescope on Maunakea in Hawaii on2019 December 4.25, ≈
69 days post-discovery. NIRESuses a 0 . (cid:48)(cid:48)
55 slit and provides wavelength coverage from9500 to 24,600 ˚A across five spectral orders at a meanresolution of R = 2700. During the observations, wenodded the target along the slit between exposures in astandard ABBA pattern to allow for accurate subtrac-tion of the sky background. Observations of the A0 Vtelluric standard star HIP 4064 near the target posi-tion were also taken immediately preceding the sciencetarget observation for flux calibration and correction ofthe strong NIR telluric absorption features. The datawere reduced, including flat-fielding, wavelength cali-bration, background subtraction, and 1D spectral ex-tractions steps, using a version of the IDL-based datareduction package Spextool developed by Cushing et al.(2004), updated by M. Cushing specifically for NIRES.Telluric corrections and flux calibrations were performedwith the standard-star observations using the methoddeveloped by Vacca et al. (2003) implemented with theIDL tools xtellcor or xtellcor general developedby Cushing et al. (2004) as part of Spextool. The NIRspectrum is shown in Figure 4.A summary of all our spectroscopic observations isprovided in Table 2. THE TRANSIENTThis section presents our analysis of the host environ-ment of the transient and its post-eruption photometricand spectroscopic evolution.3.1.
Host extinction and environment
The transient is located in a spiral arm of NGC 300,as shown in Figure 1, but it does not appear to be as-sociated with an H II region or a region of particularlydense star formation. Furthermore, the galaxy is largelyface-on, and because there is no evidence of a prominentdust lane, any extinction from the host environment willbe small.Additional evidence from our analysis of the transientpoints to low or negligible host extinction. Shafter et al.(2009) suggested a mean color of Galactic novae at max-imum light of B − V = 0 . ± .
06 mag from van denBergh & Younger (1987). Correcting for Galactic extinc-tion only, we find B − V = 0 . ± .
03 mag at the timeof the light curve peak (see Section 3.3 and Table 3),implying a low value for the host contribution to the ex-tinction. Additionally, our optical spectra, discussed in T 2019qyl: Nova in NGC 300 − − Time since eruption [days]14.016.018.020.022.024.026.0 A pp a r e n t m ag n i t ud e t − . t − . s s s s s s (near-IR)[4 . − . . − . z s − . i or i − . r or r − . g or g + 0 . V + 1 . B + 1 . U + 3 . u or u + 3 . uvw . uvm . uvw . − . . . . . . . . t = 58752 .
123 (MJD)ATLAS-o( − . Figure 2.
The post-explosion, multi-band light curves of AT 2019qyl in the UV, optical, and IR. The times correspondingto our spectroscopic observations reported in Table 2 are indicated by the black ticks labeled with an ‘s’. Broken power-lawfits to flux measurements are shown as the dashed curves in the corresponding colors for each of the
UBV and ugri lightcurves. The best-fitting parameters defining these curves are reported in Table 3. A dotted curve between the last V -bandmeasurement and the late-time F555W measurement is shown to illustrate the steep drop-off in the light curve occurring around t ≈
80 days. Power-law declines corresponding to the free-free emission model light curves of Hachisu & Kato (2006, 2007) areshown for comparison as black solid curves. In the bottom-left inset we show the early-time DLT40 light curve of AT 2019qylduring the rise of the transient, with time t since t in days on the x -axis (linear scale), and the most constraining pre-eruptionnon-detection from ATLAS (ATLAS-o − . E ( B − V ) = 0 . Section 3.5, show no evidence for Na I D absorption orabsorption by the diffuse interstellar bands at the hostredshift, which are known to correlate with dust extinc-tion (e.g., Merrill & Wilson 1938; Hobbs 1974; Phillipset al. 2013). Finally, in our analysis in Section 4.2, wefind that the SED of the quiescent precursor source canbe well modelled without any additional extinction fromthe host. Therefore, we assume a very low or negligi-ble contribution from the host to the total extinction toAT 2019qyl, and correct only for Milky Way extinctionthroughout this work.We can estimate the metallicity in the environmentof AT 2019qyl using observed metallicity gradients forNGC 300. For the galactic orientation model of Bresolinet al. (2009), the galactocentric distance of AT 2019qylis ≈ . (cid:48) ≈ II regions in NGC 300, we estimate a subsolar metallicityat the location of AT 2019qyl of 12+log(O / H) ≈ .
4, ap-proximately that of the LMC. Using the measurementsof Gazak et al. (2015) based on spectral modeling of redsupergiant stars in NGC 300, we again find a subsolarmetallicity of log(
Z/Z (cid:12) ) ≈ − . Time of eruption
As described in Section 2.1, pre-discovery non-detections by DLT40 and ATLAS constrain the timeof the eruption to < Jencson et al.
Table 1.
Photometry from follow-up observationsMJD Phase a Tel/Inst. Band App. Magnitude b (days) (mag)58752.21 0.09 DLT40 Open 17 .
38 (0 . .
34 (0 . .
39 (0 . .
40 (0 . .
41 (0 . .
42 (0 . r (cid:48) .
52 (0 . u (cid:48) .
39 (0 . g (cid:48) .
34 (0 . i (cid:48) .
60 (0 . a Phase refers to time since t on MJD 58752.12. b σ uncertainties are given in parentheses. Ground-based magnitudesare given in their native system, Vega magnitudes for UBV and ABmagnitudes for ugri and u (cid:48) g (cid:48) r (cid:48) i (cid:48) z s . DLT40 instrumental magnitudesare calibrated to r -band in AB magnitudes. For space-based facilities( HST , Spitzer , and
Swift ), measurements are in the Vega system.
Note —Table 1 is published in its entirety in the machine-readableformat. A portion is shown here for guidance regarding its form andcontent.
Table 2.
Log of spectroscopic observationsUT Date MJD Phase Tel./Instr. Range Resolution H α Resolution(days) (˚A) ( λ/δλ ) (km s − )2019 Sep 26.29 58752.29 0.17 Gemini S/GMOS 3750–7000 560 5402019 Sep 26.39 58752.39 0.27 FTN/FLOYDS 3200–10000 240–420 9002019 Sep 27.09 58753.09 0.97 SALT/RSS 3200–9000 500–1700 2102019 Sep 29.49 58755.49 3.37 FTN/FLOYDS 3200–10000 240–420 9002019 Oct 04.04 58760.04 7.92 NOT/ALFOSC 3200–9600 360 8302019 Dec 04.25 58821.25 69.13 Keck 2/NIRES 9500–24600 2700 · · · the discovery magnitude of 17 . . f ν ∝ t α . The best-fitting power-lawindex to the early rise is α = 0 .
16, which sets the time oferuption, t , to 2019 September 26.12 (MJD 58752.12),just 2.1 hours before the first DLT40 detection. Throughthis work, we use t as our reference epoch for the phaseof the transient ( t ).3.3. Post-eruption photometric evolution
As shown in Figure 2, our comprehensive follow-upobservations of AT 2019qyl with several ground- andspace-based facilities (described in Section 2.2) track themulti-wavelength photometric evolution in the UV, op-tical, and IR for the first ≈
120 days. Following the peak,the light curves display a smooth, monotonic decline un-til t ≈
71 days, after which a steep drop-off is observed.In similar fashion to the analysis of the recent eruptionsof the M31 recurrent nova M31N 2008-12a by Darnleyet al. (2016), the evolution of AT 2019qyl can be dividedinto several distinct phases: the early rise; the fast, ini-
T 2019qyl: Nova in NGC 300 f λ + c o n s t a n t [ − e r g s − c m − ˚A − ] H α H β H γ H δ H e i H e i H e i H e i H e i H e i H e i N ii A l i Fe ii Fe ii Fe ii Fe ii (27,28) (38)(37) (42) (48)(49)(55) (74) He i Paschen seriesO i O i H e i H e i H e i H e i H e i H e i H e ii N ii N ii N iii F e ii Fe ii F e ii F e ii H α H β H γ H δ ⊕⊕ T = 13000 K T = 12000 K T = 9000 K t = 0 .
17 days, +2 . t = 0 .
27 days, +1 . t = 0 .
97 days, +1 . t = 3 .
37 days, +0 . t = 7 .
92 days, +0 . ≈ − . Figure 3.
Optical spectroscopic sequence of AT 2019qyl. The spectra have been dereddened for Galactic extinction with aFitzpatrick (1999) extinction law with R V = 3 . E ( B − V ) = 0 .
01. The phase of each spectrum and vertical offsetin flux (for clarity) are indicated along the right side of the figure. Prominent emission features identified primarily from the linelists of Williams (2012) are labeled. Blackbodies that approximate the continuum emission at t = 0 .
17, 0 .
27, and 0 .
97 days with T = 13000, 12000, and 9000 K, respectively, are shown as the thick, gray curves. For comparison, we show the 1979 August 1optical spectrum of V745 Sco, a recurrent nova in a symbiotic binary system, taken ≈ t = 3 .
37 days. tial decline; the slower decline or “plateau” phase; andthe subsequent drop-off. Here, we describe each of thesephases in detail.3.3.1.
The early rise to peak: t (cid:46) day In the ground-based g -band light curve, the observedpeak magnitude of the transient is g = 17 . ± .
01 mag(AB system, M g = − . t = 0 .
61 days. The ugri data together display a conspicuous trend, with theobserved light curve peaks occurring earlier in the bluer bands ( t = 0 .
13 days in u ) and later in the redder bands( t = 0 .
98 days in i ). A similar trend is apparent in the U , B , and V bands, successively peaking at t = 0 .
25, 0 . .
96 days. Despite our very high cadence imagingin the first hours of the eruption, it is still difficult toprecisely determine the timing of the peaks owing to theextremely fast early evolution of the transient. Possibledifferences in calibration of the ground-based
U BV andcorresponding
Swift light curves also add uncertainty inestimating the peak times for those bands. Regardless of0
Jencson et al. . . . . . f λ [ − e r g s − c m − ˚A − ] Pa γ Pa δ He i He i O i O i . . . . . f λ [ − e r g s − c m − ˚A − ] Pa β . . . . . f λ [ − e r g s − c m − ˚A − ] Br-10Br-11Br-12
Rest wavelength [˚A] . . . . f λ [ − e r g s − c m − ˚A − ] Br γ Br δ Figure 4.
The late-time NIR spectrum of AT 2019qyl from t = 69 .
14 days. We label prominent emission features of H I ,He I ( λ λ β fluoresced O I ( λ the above uncertainties, the general trend of a faster riseto peak for the bluer bands, with all bands peaking attimes (cid:46) (cid:46) Swift /UVOT uvw t = 0 .
25 and 0 .
71 days, fad-ing by 0 . . ± . u and Swift U lightcurves, the UV peak may have occurred near the timeof our earliest Swift observations between t = 0 .
13 and0 .
25 days. Darnley et al. (2016) noted a correlation be-tween times to peak and wavelength for M31N 2008-12a,with shorter times to peak for the bluer bands. Despitesimilar rise time in the visible, we find the UV lightcurves of AT 2019qyl peaked even faster compared to a rise time of ≈ uvw The initial decline: days ≈ Following the peaks in the optical bands by t =0 .
94 days, we observe a rapid initial decline until a dis-tinct break is observed in the optical
BV gr -bands be-tween ≈ BV gri light curves forphases of 1–80 days with broken power laws in flux, al-lowing the time of the break, t break , to be determined bythe fit. The results of these fits are shown in Figure 2,and the best-fitting power-law indices to the early andlate portions of the decline are given in Table 3 as α and α , respectively. The BV g -band light curves all behavesimilarly, showing an initial steep decline ( α ≈ − . − .
1) and light curve breaks between t break ≈ r -band light curve occursfirst at t = 4 . α emission that dominates the optical spectra (see Sec-tion 3.5). The break in the i -band light curve occurslater at t break = 30 . r band, possibly indicating thatstrong emission lines may also begin to dominate the i -band flux at this phase. As our u and U -band coveragelasts only until t = 11 days, we fit the light curves withonly a single power law and find similar decline rates tothe BV g -bands during the initial decline.The timescale of the optical light curve decline for no-vae are traditionally described by the time for the lightcurve to fall by 1, 2, and 3 mag from the peak, which werefer to as t , t , and t , respectively. While this sim-ple parameterization does not capture the full diversityof nova light curves (see, e.g., Strope et al. 2010), it isuseful for placing AT 2019qyl in the context of knownGalactic novae as well as the growing sample of extra-galactic events. We measure these using linear interpo-lations of the photometric measurements that bracket 1,2, and 3 mag drops from the peak, and list our resultsin Table 3. In the V band, we find t , t , and t to be1 .
0, 3 .
5, and 10 . t <
10 days; Payne-Gaposchkin1964). As discussed below in Section 5.1, this placesAT 2019qyl among the fastest known novae, which aregenerally believed to occur on massive WDs (cid:38) M (cid:12) (e.g., Yaron et al. 2005).3.3.3. The slow decline phase and drop-off: t (cid:38) days T 2019qyl: Nova in NGC 300 Table 3.
Properties derived from optical light curvesFilter t peak m peak a,b M peak a,b t c t c t c t break α d α d (days) (mag) (mag) (days) (days) (days) (days) U − .
86 1.3 3.9 11.6 · · · − . · · · B − .
94 1.2 4.0 17.4 5.2 − . − . V − .
22 1.0 3.5 10.3 6.5 − . − . u − .
95 2.2 6.1 · · · · · · − . · · · g − .
24 1.2 3.4 14.23 11.2 − . − . r − .
20 2.7 38.4 · · · − . − . i − .
17 2.7 8.9 23.5 30.0 − . − . a Vega magnitudes given for
UBV and AB magnitudes given for ugri . 1 σ uncertainties given inparentheses. b Corrected for Galactic extinction toward NGC 300 assuming E ( B − V ) = 0 . c t , t , and t are defined as the times for the light curves to decline by 1, 2, and 3 magnitudes,respectively, post peak. d Best-fitting power-law index to flux measurements between 1 ≤ t [days] ≤ t break for α and t break ≤ t [days] ≤
80 for α . Following the steep, initial decline, AT 2019qyl entersa phase of slower decline after t ≈
10 days, as indicatedby the shallower power-law indices found in this phasefor the
BV gri light curves ( α in Table 3). Beyondthe optical bands, a similar change in the decline rateis also apparent in the IR [3.6] and [4.5] bands from Spitzer /IRAC. This phase is sometimes referred to as a“plateau” as defined by Strope et al. (2010) for class“P” nova light curves, though it includes events likeAT 2019qyl that continue to decline through the plateauphase, though at a slower rate. This phase lasts untilat least t ≈
71 days in the optical light curves. Subse-quently, the r -band enters a phase of steeper decline be-tween t = 94 days and the end of our ground-based pho-tometric monitoring at 127 days. Similarly, the late timeWFC3/UVIS F555W observation with HST , in compar-ison with the last ground-based V -band measurement,also indicates a steep drop-off occurred by t = 122 days.This suggests the drop in observed flux is not only drivenby a drop in H α luminosity, but likely reflects a drop inthe optical continuum emission.Hachisu & Kato (2006, 2007) have suggested a “uni-versal decline law” for novae under the assumption thatfree-free emission from the optically thin, expandingnova shells dominates the continuum flux. Their modelconsists of a broken power law for the mid- and late-time light curve, with the time of the break correspond-ing to a drop in the wind mass-loss rate at the end ofsteady hydrogen-burning on the WD surface. The timeof the break in their model depends primarily on the WD mass (and weakly on composition), providing an obser-vationally convenient method to estimate WD massesfor well-observed novae. In this model, the initial tran-sition to the slow optical decline and subsequent drop-off are timed with the emergence and turn-off of super-soft X-ray emission powered by the continued nuclearburning, also used as a proxy for the WD mass (e.g.Henze et al. 2011; Schwarz et al. 2011; Wolf et al. 2013).We show their predicted power laws in comparison toour light curves in Figure 2, and, while the measuredpower-law indices for t >
10 days ( α in Table 3) do notprecisely match the predicted value, we note qualita-tive similarity between the predictions and the observedslow-decline phase and subsequent drop-off.This light curve evolution is very similar to the Galac-tic RN RS Oph. During its 2006 eruption, its lightcurve transitioned from an initial steep decline into aslow decline or plateau, lasting until day 83, and sub-sequently a steep drop-off (e.g., Schaefer 2010, andreferences therein). In the context of other very fast-evolving and Galactic RG novae for which WD masseshave been estimated via this method (e.g., ≈ M (cid:12) forRS Oph, Hachisu et al. 2006; Hachisu & Kato 2018; seeSection 5.1 for further discussion), we may infer a sim-ilarly massive WD for AT 2019qyl. We note, however,that the applicability of this simple model is limited byseveral factors than can substantially affect the broad-band lightcurves, such as the interaction of the novaejecta with the pre-existing wind of the RG compan-ion, internal shocks (Sections 3.5.1 and 5.2), and other2 Jencson et al. complicating factors such as dust formation or the con-tribution of strong emission lines.The decline in the IR [3.6] and [4.5] light curves withinthe first 67 days appears consistent with the one ob-served in the optical bands. We note here that our
Spitzer photometry of the active transient is based onreference-subtracted images, and hence the contributionof the underlying RG (Section 4.2.1) companion hasbeen removed. Some slow novae on CO WDs, partic-ularly those of the DQ Her class, are characterized bythe formation of optically thick dust shells in the ejecta.Notable examples are NQ Vul (Ney & Hatfield 1978),LW Ser (Gehrz et al. 1980a), and V5668 Sgr (Gehrzet al. 2018). These shells are observed to re-radiate themaximum luminosity of the nova as IR emission at ≈ ≈
70 days. As discussedbelow in Section 3.5, the NIR spectrum taken at a simi-lar phase of t = 69 .
14 days does not show evidence for ared continuum beyond ≈ µ m that would indicate ther-mal emission from newly formed dust.3.4. SED evolution
We constructed quasi-contemporaneous SEDs fromour photometry at several representative epochs in theevolution of AT 2019qyl between t = 0 .
13 and 63 days.We consider observations in different photometric bandsto be contemporaneous if the difference in time betweenthem is less than one tenth the age of the transientat that phase and adopt the average phase of the in-cluded measurements at each epoch. We converted ourextinction-corrected photometric measurements to bandluminosities ( λL λ ) at the assumed distance to NGC 300and for the appropriate zero-magnitude fluxes and nom-inal effective wavelengths for each filter. The resultingSEDs are shown in Figure 5.At the earliest phases ( t (cid:46) . (cid:46) . µ m. Following this, the peak ofthe SED quickly shifts into the optical by t = 0 .
93 days.This is reflected in the rapid decline of the UV lightcurves as the optical light curves rise to maximumlight as described above (Section 3.3.1). We fit black-body approximations to these early SEDs using a cus-tom Markov Chain Monte Carlo (MCMC) code
BBFit https://github.com/nblago/utils/blob/master/src/model/BBFit.py . . . . . µ m]10 λ L λ [ L (cid:12) ] .
13 days × .
25 days × .
73 days × .
93 days1 . . . Figure 5.
The SED evolution of AT 2019qyl from photom-etry is shown for several representative phases from t = 0 . based on emcee (Foreman-Mackey et al. 2013). The re-sults for the best-fitting models and 1- σ (68%) confi-dence intervals are given in Table 4. Initially, we in-fer high effective temperatures of the photosphere of T = 11690 and 13600 K at t = 0 .
13 and 0 .
25 days, re-spectively. The apparent rise in temperature betweenthese early epochs may not be reliable owing to the lackof UV data to constrain the fit at t = 0 .
13 days. Witha radius of ≈ cm (141 R (cid:12) ), we obtain a high bolo-metric luminosity of 2 . × erg s − at t = 0 .
25 days,assuming the observed UV-optical flux accounts for thebulk of the radiated emission. This is (cid:38)
10 times theelectron-scattering Eddington luminosity for a massiveWD of L Edd = 1 . × (cid:16) M . M (cid:12) (cid:17) erg s − . At the T 2019qyl: Nova in NGC 300 Table 4.
Black-body fits to early SEDsPhase
T R L (days) (K) (10 cm) ( R (cid:12) ) (10 erg s − ) (10 L (cid:12) )0.13 11690 +70 − . +0 . − . +1 − . +0 . − . . +0 . − . +200 − . +0 . − . +4 − . +0 . − . . +0 . − . +100 − . +0 . − . +6 − . +0 . − . . +0 . − . +300 − . +0 . − . +2 − . +0 . − . . +0 . − . time of the optical light curve maxima at t = 0 .
93 days,we find that the SED can now be approximated by acooler black body with T ≈ R ≈ R (cid:12) .We thus infer a photospheric velocity of ≈ − during the rise to peak. This is similar to that inferredfrom the velocity of the P Cygni absorption features ob-served for prominent H I and He I emission lines in theearly spectra of ≈ − (see Section 3.5.1).The inferred bolometric luminosity decreases somewhatto ≈ . × erg s − , but remains substantiallysuper-Eddington. Overall, the early SED evolution ofAT 2019qyl is consistent with the so-called “fireball”phase, consisting of an increasing photospheric radiusand decreasing effective temperature as the opticallythick nova envelope expands and cools (Gehrz 1988;Hauschildt et al. 1994).During the decline phase ( t (cid:38) r -band excess that appears after t (cid:38) . α .In the IR, as noted above in Section 3.3.3, we do notobserve the development of a strong IR excess betweenphases of t ≈ t (cid:38)
60 days, the IR fluxis only a factor of ≈ Spectroscopic evolution
Our optical spectroscopic sequence, shown in Figure 3,provides an unprecedented view of the early time evolu-tion during the rise of a rapid nova outburst, with ourearliest spectrum taken only 0 .
17 days ( ≈ t ≈ T = 13000 K black body and prominent emission features of H and He I with P Cygni absorp-tion components at v ≈ − . We also note nu-merous weaker emission features of neutral and singlyionized species, including N II , Fe II , and possibly Al I based on the line lists of Williams (2012). There are anumber of additional apparent emission features, simi-lar in strength to the Fe II lines, for which we have notdetermined secure identifications. These weak emissionfeatures also appear narrow, similar to the instrumentalresolution of our GMOS spectrum reported in Table 2( v (cid:46)
500 km s − ). These features, along with narrowcomponents of H and He, are expected from the slow-moving wind of the companion star, likely an O-richasymptotic-giant-branch (AGB) star (see Section 4.2.1),flash-ionized by the explosion on the WD.In analogy with the P Cygni absorption components ofthe dominant emission lines, we note additional absorp-tion features that may be associated with weaker linesin a high velocity outflow. For example, the absorptionfeature near 4600 ˚A may be due to blended absorptionof the nearby Fe II (37; λ I ( λ II (42; λ I ( λ II ( λ t = 1 . ≈ t = 0 .
27 day spec-trum to ≈ t = 0 .
97 day spectrum. Thisis similar to the temperature evolution of black-bodyapproximations to the early SEDs discussed above inSection 3.4. During this time, the H and He I P Cygnifeatures display a complex evolution and develop multi-ple higher velocity components ( v ≈ − ),which we examine in more detail below in Section 3.5.1.In a similar fashion, the unidentified absorption fea-ture near 5650 ˚A (in the rest frame of the host) appearsshifted to the blue, and three additional absorption fea-tures near 7400, 7715, and 8150 ˚A are now apparent.The velocity structure and evolution of these features isfurther discussed in Section 3.5.2.During the light curve decline phase at t (cid:38) ≈ − . We also notethe emergence of O I emission lines ( λλ t = 3 .
37 days. At this phase, the spectrum bearsa strong resemblance to that of the 1979 outburst ofV745 Sco, a well-studied recurrent nova in a symbiotic4
Jencson et al. binary system (Williams et al. 1991). Based on thiscomparison, we note that features of N II and weakerFe II are still present, and identify additional emissionfeatures of He II and N III . The most prominent emis-sion features at t = 3 .
37 days after H are those of Heand N, consistent with an He/N spectral classification.In the late-time, NIR spectrum at t = 69 .
14 days, wedetect strong nebular emission features of H, He I , andO I with narrower, symmetric profiles. In addition tothe strong He I λ λ D - F o transition of He I . We do not observe anyNIR lines of C I that are hallmarks of Fe II -class novae(strongest lines at 1.166, 1.175, 1.689 µ m, and severallines between 1.72 and 1.79 µ m Das et al. 2008; Baner-jee & Ashok 2012), consistent with our optical spectralclassification of AT 2019qyl as an He/N nova. The spec-trum is still dominated by the nebular emission of thenova. We do not see the cool continuum and absorptionfeatures of the underlying AGB companion, or a thermalcontinuum indicative of newly formed, warm dust.3.5.1. Emission line profile evolution
Our very early spectroscopic observations reveal acomplex velocity structure and evolution in the strong Hand He I lines (see Figure 6). We examine the line profileevolution in detail here, and our velocity measurementsof individual emission and absorption components of H α and H β are further illustrated in Figure 7. Of particu-lar interest is the evolution of the P Cygni absorptioncomponents within the first t (cid:46) t = 0 . t = 0 .
27 days, each of the profiles shows a nar-row emission core (FWHM v ≈
360 km s − ) consistentwith the rest frame zero velocity of NGC 300. The emis-sion profiles are approximately Lorentzian, with wingsextending to ≈ − . The narrow emission linecore, again similar to our instrumental resolution, arisesfrom recombination in the slow-moving, ionized wind ofthe companion. The broader wings are likely the resultof interaction as the nova shock wave propagates intothe ambient medium of the companion wind. This isanalogous to the early spectra of CSM-interacting TypeIIn SNe (e.g., Schlegel 1990; Filippenko 1997) and isconsistent with Galactic examples of novae in symbi-otic systems like RS Oph (e.g. Bode 1987; Evans et al.2008) and V407 Cyg (Munari et al. 2011). The linesalso show a blueshifted P Cygni absorption componentat ≈ − , where the deepest portion of the ab-sorption component appears to be spread over a rangeof velocities between ≈ − . . . . . . . f λ ( n o r m a li ze d ) H α t = 0 .
17 days t = 0 .
27 days t = 0 .
97 days t = 3 .
37 days t = 7 .
92 days − . . . . . .
08 H α . . . . . . f λ ( n o r m a li ze d ) H β − . − . . . . i ( λ − − ] − . . . . . . f λ ( n o r m a li ze d ) H γ − − ] − . . . . . . . β He i ( λ i ( λ t = 69 .
14 days
Figure 6.
The left column shows, from top to bottom,the line profile evolution of H α , H β , and H γ . In the rightcolumn, we show a zoom-in of the H α profiles to highlightthe evolution of the P Cygni absorption features (top), theHe I ( λ t = 69 .
14 days (bottom). All line profilesare in the rest frame of NGC 300 and have been continuumsubtracted and normalized by the peak flux in the line.
Near the time of the optical light curve peaks at t = 0 .
97 days, we note multiple changes in the profileof the lines. First, while the profiles retain their narrowemission peak from the ionized companion wind, thebase of the emission component has broadened. Theprofiles can now be well approximated by two separateGaussian components, a narrow and intermediate-widthcomponent having FWHM v ≈ − and v ≈ − , respectively. The intermediate-width component is similar in velocity to the early PCygni absorption component seen at t = 0 . .
27 days.Now, the P Cygni absorption has shifted to higher ve-locities between v ≈ − and 4500 km s − , andwe see clear evidence for distinct velocity components.In particular, we note separate absorption minima in- T 2019qyl: Nova in NGC 300 ≈ − and fastcomponent at 4000 km s − . For any individual line pro-file, our choice of where to fit the continuum and poten-tial contamination by other unidentified atomic specieswould make it difficult to confidently associate these fea-tures with real P Cygni absorption components; how-ever, we observe markedly similar profiles in each of theH α , H β , H γ , and He I ( λ ≈ − within the first day after eruption may cor-respond to acceleration of the ejected material or clumpsby radiation pressure from the underlying, continued nu-clear burning on the WD surface (e.g., Williams et al.2008).By t = 3 .
37 days, the line profiles have transitionedto pure emission features. The lines appear symmet-ric about their peaks and are notably broader, withwings extending to ≈ − . At this phase, theprofiles can again be well-approximated by separatecomponents consisting of a narrower core with FWHM v ≈
360 km s − and an even broader base with FWHM v ≈ − . We suggest that the broadened emis-sion line profiles arise from the shock interaction of phys-ically distinct outflows. As a consistency check, we con-sider a 4000 km s − shell launched at t = 1 day thatwould overtake a 2000 km s − shell launched at t by t ≈ t ≈ t = 7 .
92 days, the line pro-files may now be approximated by a single Gaussiancomponent of FWHM v ≈ − . Finally,in our late time NIR spectrum, the strongest emissionlines of Pa β , He I ( λ I ( λ v ≈
350 km s − , about a factor of two larger than theinstrumental resolution of NIRES. This may be inter-preted as progressive deceleration of the shock front asit continues to propagate through and sweep up massfrom the companion wind.3.5.2. Unidentified absorption features
As described above, we noted several absorption fea-tures in our early-time spectra ( t (cid:46) | V e l o c i t y | [ k m s − ] Narrow emissionIntermediate-widthemissionSlowejecta Fastejecta H α FWHMH α P Cygni | V e l o c i t y | [ k m s − ] Narrow emissionIntermediate-widthemissionSlowejecta Fastejecta H β FWHMH β P Cygni
Figure 7.
Velocity evolution of the components of theH α (top panel, red circles) and H β (bottom panel, bluesquares) line profiles. FWHM velocities of individual narrowand intermediate-width emission components are shown asfilled symbols, where velocities comparable to the instrumen-tal resolution are indicated as upper limits with downwardarrows. Absolute values of P Cygni absorption componentvelocities are shown as unfilled symbols. which we have not determined secure identifications.The feature near 5650 ˚A is detected in all three of ourearly spectra, and shows a distinct shift to the blue at t = 0 .
97 days. As demonstrated in Figure 8, this clearlymirrors the evolution of the H α and H β P Cygni ab-sorption if we assume a rest wavelength for the featureof ≈ α and H β profiles at t = 0 .
17 days. Furthermore, for our assumedrest wavelength, it shows a comparable shift in the deep-est portion of the absorption from ≈ − and extends across a similar range in velocities ( ≈ − ) at t = 0 .
97 days.The other three features toward the red end of theoptical range are visible in the t = 0 .
97 day SALTspectrum, but were not covered by our earliest GMOSspectrum. They are also not clearly detected in the t = 0 .
27 day FLOYDS spectrum owing to a lower SNR.Yet, for assumed rest wavelengths of ≈ Jencson et al. − − ]02468 f λ ( n o r m a l z i e d ) + c o n s t a n t t = 0 .
17 daysH α H β − k m s − − − ] t = 0 .
27 days − k m s − − − ] t = 0 .
97 days − k m s − Figure 8.
Unidentified absorption features are shown incomparison to the P Cygni absorption of H α and H β atphases of t = 0 .
17, 0 .
27, and 0 .
97 days from left to rightin each of the three panels. The obscured fluxes have beencontinuum subtracted as in Figure 6, normalized by thedepth of the absorption, and shifted by an arbitrary con-stant for clarity. We infer possible rest wavelengths for eachfeature (labeled along the left side of the figure) by assum-ing their velocity evolution matches that of the H features.The dashed vertical lines indicate outflow velocities of − − − characteristic ofthe inferred multi-component ejecta. α , H β , and λ IV ] λ II λ λ I λ λ I λ II λ λ ≈ − ), and Williams et al. (2008) andWilliams & Mason (2010) suggested they arise from apre-existing circumbinary reservoir of gas, possibly lost from the donor star or accretion disk. Alternatively, asargued in a revised hypothesis by Williams (2012, 2013),the THEA features may arise in material stripped fromthe donor star during the nova eruption. In a recentlypresented sample of CNe observed spectroscopically be-fore maximum light, Aydi et al. (2020b) found that theTHEA features in one object, V906 Car, tracked thevelocity evolution of the slow component of the novaejecta, supporting the picture that they are associatedwith the nova flow, rather than pre-existing circumbi-nary gas. They noted however, that these lines did notdisplay the higher velocity component associated withthe fast ejecta, possibly owing to differences in densityor abundances between the ejecta components. Thesefindings differ from the line profiles discussed here, whichappear to trace both components of the ejecta. ARCHIVAL IMAGING AND THEPRE-ERUPTION COUNTERPARTThe location of AT 2019qyl has been extensively cov-ered by multiple ground-based and space-based imag-ing data sets. The available archival
HST imaging in-cludes ACS/WFC imaging in the F435W, F555W, andF814W filters taken on 2002 July 19 (PID: GO-9492; PI:F. Bresolin), and WFC3/UVIS imaging in the F218Wand F225W filters taken on 2014 December 1 (PID: GO-13743; PI: D. Thilker).To determine the precise positionof the transient in the archival
HST frames, we regis-tered our new WFC3/UVIS F555W observations of theactive transient to the 2002 ACS/WFC F555W frames.Using centroid measurements of 22 stars in common be-tween the two frames, we achieved an astrometric rmsuncertainty of 0.03 WFC pixels (0 . (cid:48)(cid:48) HST images using DOLPHOT and report the results here onthe Vega magnitude scale. The red source was detectedby DOLPHOT as a “good star” at F814W = 23 . ± . . ± .
152 and F225W = 23 . ± .
059 mag. Given the pres-ence of a red source, the UV measurements will be con-taminated by the non-negligible sensitivity of the UVfilters beyond 1 µ m, referred to as “red leak” . Wedescribe our estimate of the level of this contamina- The level of red leak affecting WFC3/UVIS filters is describedin Section 6.5.2 of the WFC3 Instrument handbook: https://hst-docs.stsci.edu/wfc3ihb
T 2019qyl: Nova in NGC 300 σ limiting magnitudes based on the signal-to-noise ra-tio (SNR) reported by DOLPHOT of detected sourceswithin 30 pixels (0 . (cid:48)(cid:48)
05) of the transient location. Thisgives F435W > . > . Spitzer
Super Mosaics inthe IRAC [3.6], [4.5], [5.8], and [8.0] bands and the MIPS24 µ m channel, each made from stacked images taken be-tween 2003–2007. The [3.6] and [4.5] Super Mosaics wereused as reference images for subtraction in our process-ing of the post-eruption imaging as part of the SPIRITSprogram (see Section 2). We identified a clear, point-like source in both of the [3.6] and [4.5] reference imagesconsistent with the location of the transient in our differ-ence images (bottom center panel in Figure 1). There isno pre-eruption counterpart detected at the location inthe longer wavelength Spitzer images. To obtain accu-rate photometry of the source and reliable upper limitswe built source catalogs and performed PSF-fitting pho-tometry for each of the IRAC Super Mosaic images us-ing the DAOPHOT/ALLSTAR package (Stetson 1987),where a model of the PSF is constructed using isolatedstars in the image. Our PSF-fitting and photometryprocedure, including corrections for the finite radius ofthe model PSF using the method of Khan 2017, is de-scribed in detail in Karambelkar et al. (2019). This gives[3 .
6] = 18 . ± .
09 and [4 .
5] = 18 . ± .
07 mag (Vega)for the IR precursor source. For the longer wavelengthIRAC images we adopt 5 σ limiting magnitudes basedon the SNR of detected sources in our catalogs within40 (cid:48)(cid:48) of the transient position and obtain [5 . > . . > . σ limitingmagnitude of 11 . µ m MIPSimage.Finally, we examined pre-eruption JHK s images ofNGC 300 obtained with the FourStar IR camera (Pers-son et al. 2013) on the Magellan Baade Telescope atLCO at five separate epochs between 2011–2014. The2014 images were obtained as part of a concomitantmonitoring program of SPIRITS galaxies. The IR pre-cursor object is clearly detected in all three filters as arelatively isolated point source (see top center panel ofFigure 1). We performed simple aperture photometryof the source with the aperture size set by the seeingin each image. The photometric zero points were cali-brated using several isolated 2MASS stars in each im-age. Our pre-eruption, ground-based NIR photometryis provided in Table 5 and shown in Figure 9. Table 5.
Pre-eruption NIR photometry fromBaade/FourStarMJD Phase a Band App. Magnitude b (days) (mag)55813.38 -2938.74 H .
41 (0 . K s . . J .
07 (0 . K s .
06 (0 . K s .
96 (0 . H .
28 (0 . J .
00 (0 . J . . H . . K s . . K s . . J . . a Phase refers to time since t on MJD 58752.12. b Vega magnitudes on the 2MASS system. 1 σ un-certainties are given in parentheses. Pre-eruption variability
We searched for historical variability of the pre-eruption source using the extensive archival coveragewith
Spitzer /IRAC at [3.6] and [4.5] going back 16 yearsbefore the nova eruption and regular monitoring bySPIRITS since 2014. We obtained photometry on alldifference images as described for the post-eruption de-tections of the transients above in Section 2.2. We thencomputed the Vega-system magnitudes of the source ateach epoch by adding back in the reference flux mea-surements from our PSF-fitting photometry on the Su-per Mosaics. We further stacked individual measure-ments in bins of width ∆ t = 300 days to increase ourSNR for detecting variability. Our pre-eruption Spitzer photometry is provided in Table 6 and shown in Fig-ure 9 along with the multi-epoch NIR photometry fromBaade/FourStar.The [3.6] and [4.5] light curves appear to vary largelyin sync, keeping a relatively constant color of [3 . − [4 . ≈ . ≈ ≈ gatspy (Vanderplas 2015; Van-derPlas & Ivezi´c 2015) implementation of the Lomb–8 Jencson et al.
Table 6.
Pre-eruption IR photometry from
Spitzer /IRAC a MJD Phase b dF ν, [3 . c F ν, [3 . d [3 . dF ν, [4 . c F ν, [4 . d [4 . − . − . .
2) 8 . .
3) 18 . .
4) 2 . .
3) 10 . .
3) 18 . . − . − . .
1) 7 . .
3) 18 . . − . .
9) 7 . .
0) 18 . . − .
18 1 . .
5) 10 . .
7) 18 . . − . .
9) 8 . .
0) 18 . . − .
11 5 . .
5) 14 . .
6) 18 . . · · · · · · · · · − . − . .
5) 7 . .
6) 18 . . − . .
8) 6 . .
8) 18 . . − .
88 8 . .
8) 17 . .
0) 18 . .
1) 3 . .
0) 11 . .
0) 17 . . − .
71 7 . .
2) 16 . .
4) 18 .
09 (0 .
09) 3 . .
6) 11 . .
6) 18 . . − .
36 6 . .
6) 15 . .
8) 18 . .
1) 1 . .
8) 10 . .
8) 18 . . − .
69 5 . .
0) 14 . .
0) 18 . .
4) 5 . .
6) 13 . .
6) 17 . . − .
78 4 . .
9) 13 . .
0) 18 . .
2) 2 . .
8) 10 . .
9) 18 . . − .
41 7 . .
8) 15 . .
0) 18 . .
1) 1 . .
0) 9 . .
1) 18 . . − .
56 6 . .
9) 15 . .
1) 18 . .
1) 0 . .
0) 8 . .
0) 18 . . − .
02 2 . .
4) 11 . .
5) 18 . .
2) 0 . .
8) 8 . .
9) 18 . . − .
54 4 . .
3) 13 . .
4) 18 . .
2) 1 . .
0) 9 . .
1) 18 . . − .
18 4 . .
2) 13 . .
3) 18 . .
3) 1 . .
0) 10 . .
1) 18 . . − .
09 1 . .
6) 9 . .
7) 18 . . − . .
5) 6 . .
5) 18 . . − . − . .
9) 7 . .
0) 19 . . − . .
1) 3 . .
2) 19 . . − .
47 2 . .
0) 11 . .
1) 18 . . · · · · · · · · · − .
27 0 . .
4) 9 . .
5) 18 . . − . .
1) 5 . .
2) 18 . . − .
30 0 . .
9) 9 . .
0) 18 . . − . .
7) 5 . .
7) 18 . . − .
92 3 . .
4) 12 . .
5) 18 . .
2) 1 . .
3) 10 . .
4) 18 . . − .
91 2 . .
3) 11 . .
5) 18 . . − . .
7) 6 . .
7) 18 . . − .
29 6 . .
6) 15 . .
7) 18 . .
2) 2 . .
5) 11 . .
6) 18 . . − .
54 6 . .
7) 15 . .
8) 18 . .
2) 2 . .
5) 11 . .
6) 18 . . a σ uncertainties given in parentheses. b Phase refers to time since t on MJD 58752.12. c Measured flux in difference images. d Total flux, including that from PSF-fitting photometry on our reference images.
Scargle method (Lomb 1976; Scargle 1982). We re-stricted our period search to sinusoidal signals given thesparse sampling of the light curves. The best-fitting pe-riod is 1820 days with a reduced χ value of χ /ν = 0 . χ /ν = 2 .
1. Furthermore, the peak Lomb-Scargle power is only 0.66, equal to the power in a sec-ondary period of 330 days. Thus, we are unable to makea robust determination of any periodicity with the avail-able data.Our temporal coverage in the NIR
JHK s bands ismore limited. During our 2011 coverage, we note a risein all three bands over a period of 55 days. The largest amplitude of variability is observed in the K s band tobe 0.7 mag, however it is likely the JHK s amplitudesare larger than is represented in the available data. Thetiming of the 2014 UV imaging observations with HST is also indicated in Figure 9. Though we infer the pre-cursor source must be in an outburst state based on theUV flux at this epoch (see Section 4.2.2 below), we donot see a corresponding substantial increase in flux inthe closest [3.6] and [4.5] images taken 48 days earlier orthe
JHK s images taken 23 days after.The observed IR variability may suggest semiregu-lar pulsations of the AGB companion (Section 4.2.1).In this case, comparing the pre-eruption magnitudes ofAT 2019qyl to the [3.6] and [4.5] period-luminosity rela- T 2019qyl: Nova in NGC 300 .
5] 0100020003000400050006000201918 [3 .
6] 01000200030004000500060002019 A pp a r e n t m ag n i t ud e K s H J Quiescence(F435W, F555W,F814W) UV Outburst(F218W, F225W)
Figure 9.
Pre-explosion light curves of the progeni-tor of AT 2019qyl in the NIR J , H , and Ks bands fromBaade/FourStar and at [3.6] and [4.5] from Spitzer /IRAC.Time on the x -axis is relative to the inferred eruption epochof the nova at MJD 58752.12. As in Figure 2, all broad-band photometry is corrected for Milky Way extinction. Ourmeasurements of the IR counterpart source in the archival Spitzer /IRAC [3.6] and [4.5] Super Mosaic images from PSF-fitting photometry are indicated by unfilled symbols, wherethe dashed horizontal lines and shaded bars indicate therange of observation dates included in the image stack andthe uncertainty in the magnitude measurements, respec-tively. The individual epoch of NIR imaging at 55840.00 usedin the construction of the precursor SED (see Figure 10) areindicated by black-outlined symbols. The times of the
HST
ACS/WFC F453W, F555W, and F814W images (during qui-escence) and the WFC3/UVIS F218W and F225W images(during outburst) are shown as the vertical dashed lines aslabeled near the bottom of the figure. Colors used in thisfigure for different bands mirror those used in Figure 10. tions of Whitelock et al. (2017) and Karambelkar et al.(2019), we would expect a period of ≈ (Tonryet al. 2018; Smith et al. 2020) to constrain the opticalvariability of the progenitor during this time. In stackedmeasurements with 1-day binning, there are no prior de-tections of the source to typical 5- σ limiting magnitudesof (cid:38) The pre-eruption SED
Figure 10 shows the multi-epoch SED of the pre-eruption counterpart, constructed using the availablearchival imaging. We include UV, optical and MIR mea-surements taken between 2002 and 2014 (see Section 4for details). The photometric magnitudes were con-verted to band-luminosities, λL λ , using the zero pointflux densities and effective wavelengths compiled by theSpanish Virtual Observatory (SVO) Filter Profile Ser-vice for the appropriate filters (Rodrigo et al. 2012;Rodrigo & Solano 2020). The most striking feature ofthe pre-eruption SED is the presence of distinct red andblue components, peaking in the IR and UV, respec-tively. The following sections provide an analysis of eachindividual component.4.2.1. The IR component
The IR SED component peaks in the J band at aband-luminosity of λL λ = 3 . +1 . − . × L (cid:12) , where the ATLAS forced-photometry server: https://fallingstar-data.com/forcedphot/ Documentation for the SVO Filter Profile Service is availableat http://ivoa.net/documents/Notes/SVOFPSDAL/index.htmland http://ivoa.net/documents/Notes/SVOFPSDAL/index.html Jencson et al. . . . . . . . µ m]10 λ L λ [ L (cid:12) ] J H K s µ m AG DraV745 Sco V407 CygAG Dra 1979(quiescence)AG Dra 1994(outburst)
Figure 10.
The multi-epoch, pre-eruption SED of theAT 2019qyl precursor from archival imaging is shown as thelarge squares. The time of observation for each point is indi-cated by different colors as labeled along the top of the fig-ure. Error bars (not including the distance uncertainty) aresmaller than the plotting symbols. 5 σ upper limits from non-detections are indicated by unfilled squares with downwardarrows. The level of variability observed in the pre-eruptionlight curves for a given band is indicated by the vertical bars(see Figure 9). GRAMS O-rich AGB models that providegood fits to the optical and IR data (see main text) areshown as thin orange curves, and the best-fitting model fora star with L ∗ = 3 . × L (cid:12) and T eff = 2500 K is indicatedby the thick-lined, dashed curve. The maximum allowableflux for the Rayleigh-Jeans tail of a hot black body compo-nent (blue dotted curve) that is consistent with the visible HST upper limits from 2002 (in sum with the best-fittingIR component model; orange dotted curve) is a factor of ≈ HST in 2014, indicatinga UV outburst at that epoch. For comparison, we show qui-escent SEDs of the symbiotic binaries and novae V745 Sco(gray, thin diamonds; Schaefer 2010; Darnley et al. 2012) andV407 Cyg (dark gray, thick diamonds; Esipov et al. 1988),and that of the classical symbiotic star AG Dra from pho-tometry (light gray circles) along with IUE spectra from 1979September 25 during quiescence (light gray) and 1994 July28 during an outburst (blue; Skopal 2005). For V407 Cyg weuse an updated distance estimate of 3.6 kpc based on the Ita& Matsunaga (2011) period-luminosity relation for HBB Mi-ras with K = 3 . E ( B − V ) = 1 mag as in Hachisu & Kato (2018). uncertainty includes the observed variability. This sug-gests the presence of a cool giant, for which the broad-band IR colors can be used to discriminate betweenAGB and RGB stars and provide a photometric classi-fication (Whitelock 1987; Blum et al. 2006; Cioni et al.2006; Srinivasan et al. 2009; Whitelock et al. 2009). At M K s = − . J − [3 .
6] = 1 . J − [3 . > . − M (cid:12) yr − (Delfosse et al. 1997; Blumet al. 2006; Lagadec & Zijlstra 2008; Boyer et al. 2015).Thus, we infer that the IR component likely correspondsto a low-mass, O-rich AGB star in a pre-Mira phase ofmass loss. The source has an intermediate luminositybetween the symbiotic binary progenitors of the novaeV745 Sco and V407 Cyg.Based on the photometric classification, we fit the op-tical and IR SED of the precursor with the Grid ofRed supergiant and Asymptotic Giant Branch Mod-elS (GRAMS; Sargent et al. 2011; Srinivasan et al.2011) suite of radiative transfer models with O-rich(silicate) dust to estimate physical parameters of thestar. This consists of a base grid of 1225 models ofspherically symmetric shells of varying amounts of sili-cate dust (Ossenkopf et al. 1992) around stars of con-stant mass-loss rates computed using the dust radiativecode (Ueta & Meixner 2003). The grid usesPHOENIX model photospheres (Kuˇcinskas et al. 2005,2006) for 1 M (cid:12) stars with effective temperatures, T eff ,between 2100 and 4700 K at fixed sub-solar metallic-ity of log( Z/Z (cid:12) ) = − . g = − . R ∗ ≈ R (cid:12) . The basic in-put parameters of each model are T eff , the inner radiusof the dust shell, R in , and the optical depth of the dustshell at 10 µ m, τ . The luminosity of the star, L ∗ , iscomputed by integrating the output spectrum over allwavelengths, and can be scaled by a factor s with theflux density of the output spectrum to match stars ofvarying brightness for a given T eff . For fixed log g , theinferred stellar mass and radius then scale as M ∗ ∝ s and R ∗ ∝ s / . From τ , the models also include thedust mass-loss rate, ˙ M dust , assuming a dust wind-speedof v w = 10 km s − . T 2019qyl: Nova in NGC 300
JHK s SED points. For the IRAC points, we use the[3 . − [4 .
5] color as another constraint on the modelsowing to the large degree of variability observed in the[3.6] and [4.5] light curves at approximately constantcolor (see Section 4.1). The base grid includes mod-els with R in = [3 , , , R ∗ , but we found that thefits were largely insensitive to this parameter. Thus werestricted our fitting procedure to include only modelswith R in = 11 R ∗ in order to reduce the number of freeparameters. We compute the reduced- χ ( χ /ν ) statis-tic for each model, and consider acceptable fits to bethose within a factor of e of the best-fitting model thatminimizes χ /ν . We further rejected models that are in-consistent with any of the upper limits estimated fromthe F435W, F555W, [5.5], [8.0], and 24 µ m images. Ac-ceptable model fits are shown along with the data inFigure 10.The best-fitting model has T eff = 2500 K, L ∗ =3 . × L (cid:12) , and τ = 10 − , and gives χ /ν = 0 . ≤ T eff [K] ≤ . × ≤ L ∗ [ L (cid:12) ] ≤ . × ,and 10 − ≤ τ ≤ .
21, but we note that the allowedranges include values near their respective minima inthe grid for T eff and τ . Adopting the luminosity scal-ing relations for fixed log g described above, the im-plied best-fitting values and allowed ranges of the stellarmass and radius are M ∗ = 1 . . M (cid:12) and R ∗ = 320–410 R (cid:12) , with best-fitting values of M ∗ = 1 . M (cid:12) and R ∗ = 320 R (cid:12) , respectively. As is evident in Figure 10,we are limited in our ability to constrain the strengthof the 10 µ m silicate feature owing to our lack of goodphotometric constraints in this wavelength region, per-mitting values of τ up to 0 .
21. This corresponds todust mass-loss rates between ˙ M dust = 2 . × − and7 . × − M (cid:12) yr − , and total wind mass-loss rates be-tween ˙ M w ≈ × − and 2 × − M (cid:12) yr − for agas-to-dust ratio of 200 (Groenewegen et al. 2009).We note that the assumption of spherical symmetrymay be a poor approximation for the winds of symbi-otic stars, where interactions with a massive WD canfocus circumstellar material in the orbital plane of thebinary (e.g., Booth et al. 2016 for RS Oph). We discussthe possible implications of this for mass transfer in theprogenitor of AT 2019qyl further in Section 5.1.3.4.2.2. The UV component
Given the deep limits from
HST at F435W andF555W, the bright UV counterpart detected in F W and F W in 2014 indicates a separate, hot componentof the SED. These UV filters have non-negligible out- of-band transmission in the red and IR, referred to as“red leak”. The IR source at the location will thus con-taminate our UV flux measurements. We estimate thelevel of contamination assuming the best-fitting modelspectrum for the IR source described above and com-pute the expected observed flux for this source throughthe WFC3/UVIS F218W and F225W filters using theSTScI pysynphot package, which accounts for the fullthroughput calibration of HST including every opticalcomponent. We estimate the IR source accounts foronly 0.01% and 0.02% of the flux observed in F218Wand F225W, respectively, and subtract these contribu-tions from our estimate of the flux in each band. Thiscontamination is thus very small, and has a negligibleeffect on our analysis of the UV precursor source.At a UV band-luminosity of λL λ ≈ L (cid:12) , the 2014UV source is too luminous to correspond to a quiescentphase spectrum consistent with the visible HST
F435Wand F555W limits from 2002. We infer a limit on the UVluminosity of such a source by considering the maximumflux of the Rayleigh-Jeans tail of a hot black body ( f λ ∝ λ − ) that, when summed with our best-fitting spectrumfor the IR source, is consistent with the 2002 HST visiblelimits. As illustrated in Figure 10, the observed UVluminosity is a factor of ≈ ≈ Spitzer light curves (Figure 9), itis plausible such outbursts may account for some portionof the observed IR variability. DISCUSSIONIn this study, we presented a detailed analysis of theearly time evolution of a rapidly evolving nova, and an-alyzed its symbiotic progenitor system. In this section,we first compare AT 2019qyl with the sample of RS Oph-like Galactic RNe with RG companions and long orbitalperiods. We argue specifically that it is likely to bea short-recurrence-time system (Section 5.1). We then2
Jencson et al. examine the properties of the outflow inferred from ourunique data set for such an event, consisting of well-sampled early light curves and spectroscopy within thefirst day of the nova eruption. In particular, we dis-cuss implications for possible outflow mechanisms andthe importance of internal shocks as well as interactionsbetween the ejecta and companion wind (Section 5.2).5.1.
AT 2019qyl in the context of RG RNe novae
AT 2019qyl is most similar to the subgroup of RNewith RG companions and orbital periods (cid:38)
200 days, in-cluding the famed RS Oph, as well as T CrB, V745Sco, V3890 Sgr (e.g., Evans et al. 2008; Schaefer 2010;Darnley et al. 2012, and references therein). The 2010nova outburst of V407 Cyg, which hosts a O-rich Miracompanion is also similar to this class in many respects,though it has not been confirmed to recur (e.g., Munariet al. 2011; Shore et al. 2011; Hinkle et al. 2013, andreferences therein). Here we examine AT 2019qyl in thecontext of this class in terms of the properties of thenova eruption itself, and the binary configuration.5.1.1.
Ejecta velocity, ejecta mass, and WD mass
RG RNe are associated with very fast evolving lightcurves, usually expected to correspond to the most mas-sive WDs ( (cid:38) M (cid:12) ). Because of their larger surfacegravities, they require low critical masses of accreted ma-terial to ignite the TNR, resulting in low ejecta masses of ≈ − –10 − M (cid:12) and high velocities (cid:38) − (e.g.,Yaron et al. 2005; Hillman et al. 2015, 2016). This ap-pears to be borne out for RG RNe where ejecta and WDmasses have been estimated in connection with very fastevolution timescales, including RS Oph and V745 Sco(Osborne et al. 2011; Page et al. 2015). AT 2019qyl dis-plays such high ejecta velocities (Section 3.5.1), and with t ( V ) = 3 . t ( V ) = 10 . t (cid:46) (cid:18) M env M (cid:12) (cid:19) = 0 .
825 log( t ) − . , (1)from which we obtain M env ≈ × − M (cid:12) . As notedin Section 3.3.3, the steep light curve drop-off after ≈
80 days for AT 2019qyl may be related to the end ofsteady nuclear burning on the WD surface, a proxy for M WD . Using Table 3 in (Hachisu & Kato 2006), wethereby estimate M WD ≈ . . M (cid:12) . Hachisu & Kato(2019) have found similarly massive WDs between ≈ M (cid:12) from optical light curves for RS Oph, TCrB, V745 Sco, and V407 Cyg.5.1.2. Companion wind interaction
A fundamental difference between RG nova systemsand CNe with main-sequence donors is that the ex-plosion is embedded in the dense wind of the evolvedcompanion, leading to distinct observational signatures.In particular, during the rise to maximum light for t (cid:46) I ( (cid:46)
300 km s − ; Figure 6) as well asnumerous weaker narrow emission features, which likelyarise from the slow-moving wind of the RG, flash ion-ized by the explosion on the WD. These features arecommon among RG novae observed early (e.g., Bode2010 for RS Oph, Munari et al. 2011 for V407 Cyg). Asthe nova evolves, shock interactions between the ejectaand the pre-existing wind shape the spectra, producingbroader emission components whose velocities decreasewith time as the shock is decelerated by the dense wind,features seen clearly in the evolution of AT 2019qyl (Sec-tion 3.5.1).These external interactions also generate non-thermalsynchrotron as well as high-energy emission from hotplasma, observed in radio and X-rays for RS Oph forits 1985 and 2006 eruptions (e.g., Hjellming et al. 1986;O’Brien et al. 2006; Sokoloski et al. 2006), and more re-cently for other embedded novae (e.g., Nelson et al. 2012;Delgado & Hernanz 2019; Orio et al. 2020). V407 Cyg,in fact, was the first gamma-ray detected nova (e.g.,Abdo et al. 2010). The shocks were interpreted as aris-ing from collisions between the nova ejecta and the denseMira wind (Nelson et al. 2012; Martin & Dubus 2013),though Martin et al. (2018) argue that internal shockscould have given rise to the observed gamma-rays evenwithout any contribution from ejecta-wind interactions(see Section 5.2 below for a discussion of internal shocksfor AT 2019qyl). While secure radio or high-energy de-tections would be challenging to obtain at the distanceof AT 2019qyl in NGC 300, the observed optical spectralsignatures confirm the influence of the evolved, giantcompanion wind on the evolution of the eruption.5.1.3. The binary configuration
In Section 4.2.1, we estimated R ≈ R (cid:12) and M ≈ . M (cid:12) for the AGB companion based on modeling ofthe IR component of the precursor source SED. For abinary with a similar mass WD (mass ratio q ≈ R RL to the semimajor axis, a , is 0 .
38 (Eggleton 1983). Thus, we infer a lower limit
T 2019qyl: Nova in NGC 300 a (cid:38) R (cid:12) ( (cid:38) P (cid:38) (cid:38) ≈ ≈ R RL above. If theobserved IR variability is taken to correspond to or-bital modulations, we may thus expect the donor starto transfer mass onto the WD via Roche-lobe overflow.Alternatively, the IR variations could be attributed tosemiregular pulations of the low-mass AGB companion,allowing for a larger orbital separation and longer orbitalperiod more similar to V407 Cyg. In this case, masstransfer onto the WD could still occur via a processcalled wind Roche-lobe overflow (WRLOF; Mohamed& Podsiadlowski 2012, and see I(cid:32)lkiewicz et al. 2019 fora direct application to V407 Cyg), or standard Bondi-Hoyle-Lyttleton (BHL) wind accretion (Hoyle & Lyttle-ton 1939; Bondi & Hoyle 1944).In BHL scenario, assuming an orbital separation of16 AU similar to that inferred for V407 Cyg, we canestimate the wind-accretion rate of material onto theWD as it moves through the AGB wind following Livio& Warner (1984):˙ M acc =1 . × − (cid:18) M WD M (cid:12) (cid:19) (cid:18) v rel
10 km s − (cid:19) − × (cid:16) a cm (cid:17) − (cid:32) ˙ M w − M (cid:12) yr − (cid:33) M (cid:12) yr − , (2)where v rel is the relative velocity between the AGB windand the WD given by v = v w + (cid:0) πaP (cid:1) . Adopting v w =10 km s − and ˙ M w ≈ − M (cid:12) yr − from our analysisof the IR source in Section 4.2.1, we obtain ˙ M acc (cid:46) . × − M (cid:12) yr − . For a 1 . M (cid:12) WD, this would implya recurrence time of ∼
20 yr comparing to calculationsby Wolf et al. (2013) and those presented in Chomiuket al. (2020). We find, however, that even at this wide separation,accretion onto the WD is likely to proceed via WRLOF.In this scenario, the AGB wind is concentrated into theorbital plane through interactions with the WD, lead-ing to funneling of material and significantly enhancedaccretion onto the WD. This process has been exam-ined, for example, in both RS Oph (Booth et al. 2016)and V407 Cyg (I(cid:32)lkiewicz et al. 2019). Using 3D hy-drodynamical simulations, Mohamed & Podsiadlowski(2012) found that WRLOF led to more than an order-of-magnitude increase in the accretion rate for a symbi-otic Mira with a 20 AU binary separation and an 0.6 M (cid:12) WD. For a massive WD like that in AT 2019qyl, the in-teraction would be even stronger. Thus, the BHL accre-tion rate (and corresponding recurrence time) estimatedabove may only be a lower (upper) limit. We cautionthat there are substantial uncertainties in our knowledgeof the relevant binary parameters for this system thatcould alter this estimate. Still, this suggests AT 2019qylis a strong candidate for a short-recurrence-time systemsimilar to the Galactic RG RNe.5.2.
Outflow collisions and internal shocks inAT 2019qyl
In Section 3.5.1, we presented consistent evidenceacross several H and He lines for multiple, distinct ve-locity components in the nova outflow of AT 2019qyl.We observe an early ≈ − “slow” P Cygni ab-sorption component at t = 0 . .
27 days, which ac-celerates to ≈ − by the time of the opticalpeak at 0 .
97 days. Concurrently, we observe the appear-ance of a “fast” absorption component at ≈ − and an intermediate-width ( ≈ − ) emis-sion component. The broadened emission is indicativeof shock interactions at this phase, likely between theearly slow ejecta and the pre-existing circumbinary ma-terial of the companion wind, but internal interactionsbetween ejecta components may also already contribute.At t = 3 .
37 days, the line profiles are in emission onlyand the intermediate-width component velocity has in-creased to ≈ − suggesting continued and in-tensified shock interaction. While we cannot completelydisentangle the effects of ejecta-wind interactions andinternal collisions, the timing of the appearance of thebroadened component is consistent with expectations forcollisions within the observed multi-component outflow.This is similar to the general picture of early evolutionthat has been recognized for several decades in CNe, butat notably higher velocities (cf. 100–1000 km s − for theintial “slow” component; McLaughlin 1942; Gallagher& Starrfield 1978). Aydi et al. (2020b) recently showedwith a sample of pre-maximum spectra of twelve no-4 Jencson et al. vae that this scenario may be universal—though theirsample notably did not include any novae with declinetimes t < v esc =[2 GM/R w ] / ), we can estimate the outflow launchingradius, R w , for a given component. For the 2000 km s − “slow” component, we obtain R w ≈ × cm (0 . R (cid:12) ,8 R WD ). This is much smaller than the binary orbitestimated above, which excludes a common-envelope-like scenario in this case. An alternative scenario forthe early mass ejection in AT 2019qyl may be an im-pulsive ejection coincident with the TNR (Seaquist &Bode 2008; Shore 2014; Mason et al. 2018), or a short-duration wind powered by radioactive heating from β -decay (Starrfield et al. 2008). The opacity-driven, pro-longed wind model of Kato & Hachisu (1994) is likelystill important given the sustained luminosity of novaefor days after the TNR, as seen here for AT 2019qyl.This may account for secondary phases of mass loss,i.e., the “fast” ejecta, and may also contribute to theobserved acceleration of the slower ejecta (as in, e.g.,Friedjung 1987). SUMMARYWe have described the early discovery and promptfollow-up observations of AT 2019qyl, a very fast novawith an O-rich AGB donor in NGC 300. The DLT40supernova survey discovered AT 2019qyl within ≈ t ,V = 3 . V -band and placing AT 2019qyl among the fastest known novae. The light curves decline smoothly untilat least t = 71 days after which a steeper drop-off isobserved, possibly corresponding to the end of nuclearburning on the WD surface. In analogy with similar sys-tems, the rapid decline and timing of the drop-off pointto a massive WD of M WD (cid:38) . M (cid:12) . The early evolu-tion of the broadband SED is largely consistent with anexpanding “fireball” that cools from ≈ . . × erg s − (8–13 × L Edd ).We obtained three high-quality optical spectra dur-ing the ≈ (cid:46) few ×
100 km s − ), likely arising inthe dense, pre-existing wind of a RG companion star.The evolution of the P Cygni profiles of the strongestlines reveals multiple, distinct velocity components inthe ejecta, including an early 2000 km s − componentthat accelerates to 2500 km s − concurrently with theappearance of a superimposed higher velocity compo-nent at 4000 km s − near the time of the optical peakaround t = 1 day. By t = 3 .
37 days, the emission lineshave broadened, indicative of ongoing shock-interaction,the timing of which is consistent with internal collisionsbetween the previously ejected outflows. The widths ofthe emission lines are then observed to decline, provid-ing further evidence of the influence of the dense wind ofa RG companion that acts to decelerate the ejecta. Theearly line evolution is remarkably similar to that longrecognized in early optical spectra of CNe (McLaughlin1942; Gallagher & Starrfield 1978), and recently shownto be commmon, if not ubiquitous, in slower CN out-bursts (Aydi et al. 2020b). In contrast with proposedscenarios for CNe, the high velocities in AT 2019qyl re-quire the outflows to originate close to the WD, possiblyin a multi-phase, opacity-driven wind, and preclude acommon-envelope-like mass-loss mechanism.We also presented a detailed examination of the ex-tensive archival data set available for this event, whichreveals a red/IR- and UV-bright source at the nova lo-cation. Models of the pre-eruption SED suggest an O-rich AGB companion to the WD in a long period bi-nary ( (cid:38)
HST imaging also indicates a prior symbiotic-like out-
T 2019qyl: Nova in NGC 300
Software:
Facility:
CTIO:PROMPT, LCOGT (Sinistro),Swope (CCD), LO:1m, Keck:II (NIRES), Gemini:South(GMOS), SALT (RSS), FTN (FLOYDS), NOT (AL-FOSC), HST (WFC3), Spitzer (IRAC), Swift (UVOT),Magellan:Baade (FourStar)ACKNOWLEDGMENTSWe would like to thank Jorge Anais Vilchez, AbdoCampillay, Yilin Kong Riveros, and Natalie Ulloa fortheir help with Swope observations, and the Magellanobservers of Las Campanas Observatory and A. Monsonfor help with the Baade/FourStar data. We also thankElias Aydi for helpful discussions.Research by S.V. is supported by NSF grantsAST–1813176 and AST-2008108. Support for
HST program GO-15151 was provided by NASA through agrant from STScI. Research by D.J.S. is supported byNSF grants AST-1821967, AST-1821987, AST-1813708,AST-1813466, and AST-1908972, as well as by theHeising-Simons Foundation under grant
Jencson et al. tory was made possible by the generous financial sup-port of the W. M. Keck Foundation. The authors wishto recognize and acknowledge the very significant cul-tural role and reverence that the summit of Maunakeahas always had within the indigenous Hawaiian commu-nity. We are most fortunate to have the opportunity toconduct observations from this mountain.This research is based on observations made with theNASA/ESA
Hubble Space Telescope obtained from theSpace Telescope Science Institute, which is operated bythe Association of Universities for Research in Astron-omy, Inc., under NASA contract NAS 5–26555. Theseobservations are associated with program(s) GO-15151,GO-9492, and GO-13743.This work is based in part on archival data obtainedwith the
Spitzer Space Telescope , which is operated bythe Jet Propulsion Laboratory, California Institute ofTechnology, under a contract with NASA.This research was made possible through the useof the AAVSO Photometric All-Sky Survey (APASS),funded by the Robert Martin Ayers Sciences Fundand NSF AST-1412587.This research has made use ofthe NASA/IPAC Extragalactic Database (NED), whichis operated by the Jet Propulsion Laboratory, Cal-ifornia Institute of Technology, under contract withthe National Aeronautics and Space Administration. This research has made use of the SVO Filter ProfileService (http://svo2.cab.inta-csic.es/theory/fps/) sup-ported from the Spanish MINECO through grantAYA2017-84089.The national facility capability for SkyMapper hasbeen funded through ARC LIEF grant LE130100104from the Australian Research Council, awarded to theUniversity of Sydney, the Australian National Univer-sity, Swinburne University of Technology, the Univer-sity of Queensland, the University of Western Australia,the University of Melbourne, Curtin University of Tech-nology, Monash University and the Australian Astro-nomical Observatory. SkyMapper is owned and oper-ated by The Australian National University’s ResearchSchool of Astronomy and Astrophysics. The survey datawere processed and provided by the SkyMapper Teamat ANU. The SkyMapper node of the All-Sky VirtualObservatory (ASVO) is hosted at the National Compu-tational Infrastructure (NCI). Development and supportthe SkyMapper node of the ASVO has been funded inpart by Astronomy Australia Limited (AAL) and theAustralian Government through the Commonwealth’sEducation Investment Fund (EIF) and National Col-laborative Research Infrastructure Strategy (NCRIS),particularly the National eResearch Collaboration Toolsand Resources (NeCTAR) and the Australian NationalData Service Projects (ANDS).REFERENCES
Abdo, A. A., Ackermann, M., Ajello, M., et al. 2010,Science, 329, 817, doi: 10.1126/science.1192537Ackermann, M., Ajello, M., Albert, A., et al. 2014, Science,345, 554, doi: 10.1126/science.1253947Andrews, J., Valenti, S., Lundquist, M., et al. 2019,Transient Name Server Classification Report, 2019-1920,1Aydi, E., Page, K. L., Kuin, N. P. M., et al. 2018, MNRAS,474, 2679, doi: 10.1093/mnras/stx2678Aydi, E., Sokolovsky, K. V., Chomiuk, L., et al. 2020a,Nature Astronomy, 4, 776,doi: 10.1038/s41550-020-1070-yAydi, E., Chomiuk, L., Izzo, L., et al. 2020b, arXiv e-prints,arXiv:2010.07481. https://arxiv.org/abs/2010.07481Banerjee, D. P. K., & Ashok, N. M. 2012, Bulletin of theAstronomical Society of India, 40, 243.https://arxiv.org/abs/1306.0343Bath, G. T., & Shaviv, G. 1976, MNRAS, 175, 305,doi: 10.1093/mnras/175.2.305Bhardwaj, A., Kanbur, S. M., Macri, L. M., et al. 2016, AJ,151, 88, doi: 10.3847/0004-6256/151/4/88 Blum, R. D., Mould, J. R., Olsen, K. A., et al. 2006, AJ,132, 2034, doi: 10.1086/508227Bode, M. F. 1987, RS Ophiuchi (1985) and the recurrentnova phenomenon. Proceedings of the ManchesterConference, held at Manchester, UK, 16 - 18 December1985.—. 2010, Astronomische Nachrichten, 331, 160,doi: 10.1002/asna.200911319Bondi, H., & Hoyle, F. 1944, MNRAS, 104, 273,doi: 10.1093/mnras/104.5.273Booth, R. A., Mohamed, S., & Podsiadlowski, P. 2016,MNRAS, 457, 822, doi: 10.1093/mnras/stw001Bostroem, K. A., Valenti, S., Sand, D. J., et al. 2020, ApJ,895, 31, doi: 10.3847/1538-4357/ab8945Boyer, M. L., McQuinn, K. B. W., Barmby, P., et al. 2015,ApJ, 800, 51, doi: 10.1088/0004-637X/800/1/51Bresolin, F., Gieren, W., Kudritzki, R.-P., et al. 2009, ApJ,700, 309, doi: 10.1088/0004-637X/700/1/309Brown, T. M., Baliber, N., Bianco, F. B., et al. 2013,PASP, 125, 1031, doi: 10.1086/673168
T 2019qyl: Nova in NGC 300 Buckley, D. A. H., Swart, G. P., & Meiring, J. G. 2006, inSociety of Photo-Optical Instrumentation Engineers(SPIE) Conference Series, Vol. 6267, Proc. SPIE, 62670Z,doi: 10.1117/12.673750Burgh, E. B., Nordsieck, K. H., Kobulnicky, H. A., et al.2003, in Society of Photo-Optical InstrumentationEngineers (SPIE) Conference Series, Vol. 4841,Proc. SPIE, ed. M. Iye & A. F. M. Moorwood,1463–1471, doi: 10.1117/12.460312Cheung, C. C., Jean, P., Shore, S. N., et al. 2016, ApJ, 826,142, doi: 10.3847/0004-637X/826/2/142Chomiuk, L., Metzger, B. D., & Shen, K. J. 2020, arXive-prints, arXiv:2011.08751.https://arxiv.org/abs/2011.08751Chomiuk, L., Linford, J. D., Yang, J., et al. 2014, Nature,514, 339, doi: 10.1038/nature13773Cioni, M. R. L., Girardi, L., Marigo, P., & Habing, H. J.2006, A&A, 448, 77, doi: 10.1051/0004-6361:20053933Cushing, M. C., Vacca, W. D., & Rayner, J. T. 2004,PASP, 116, 362, doi: 10.1086/382907Darnley, M. J. 2017, in Astronomical Society of the PacificConference Series, Vol. 509, 20th European White DwarfWorkshop, ed. P. E. Tremblay, B. Gaensicke, &T. Marsh, 515. https://arxiv.org/abs/1611.01301Darnley, M. J., & Henze, M. 2020, Advances in SpaceResearch, 66, 1147, doi: 10.1016/j.asr.2019.09.044Darnley, M. J., Ribeiro, V. A. R. M., Bode, M. F.,Hounsell, R. A., & Williams, R. P. 2012, ApJ, 746, 61,doi: 10.1088/0004-637X/746/1/61Darnley, M. J., Williams, S. C., Bode, M. F., et al. 2014,A&A, 563, L9, doi: 10.1051/0004-6361/201423411Darnley, M. J., Bode, M. F., Kerins, E., et al. 2006,MNRAS, 369, 257, doi: 10.1111/j.1365-2966.2006.10297.xDarnley, M. J., Henze, M., Bode, M. F., et al. 2016, ApJ,833, 149, doi: 10.3847/1538-4357/833/2/149Das, R. K., Banerjee, D. P. K., Ashok, N. M., & Chesneau,O. 2008, MNRAS, 391, 1874,doi: 10.1111/j.1365-2966.2008.13998.xDe, K., Kasliwal, M. M., Hankins, M. J., et al. 2021, arXive-prints, arXiv:2101.04045.https://arxiv.org/abs/2101.04045Delfosse, X., Kahane, C., & Forveille, T. 1997, A&A, 320,249Delgado, L., & Hernanz, M. 2019, MNRAS, 490, 3691,doi: 10.1093/mnras/stz2765della Valle, M., & Livio, M. 1996, ApJ, 473, 240,doi: 10.1086/178139Dolphin, A. 2016, DOLPHOT: Stellar photometry,Astrophysics Source Code Library.http://ascl.net/1608.013 Dolphin, A. E. 2000, PASP, 112, 1383, doi: 10.1086/316630Eggleton, P. P. 1983, ApJ, 268, 368, doi: 10.1086/160960Esipov, V. F., Taranova, O. G., & Yudin, B. F. 1988,Astrophysics, 29, 582, doi: 10.1007/BF01005961Evans, A., Darnley, M. J., Bode, M., & O’Brien, T. J., eds.2008, Astronomical Society of the Pacific ConferenceSeries, Vol. 401, RS Ophiuchi (2006) and the RecurrentNova PhenomenonFazio, G. G., Hora, J. L., Allen, L. E., et al. 2004, ApJS,154, 10, doi: 10.1086/422843Filippenko, A. V. 1997, ARA&A, 35, 309,doi: 10.1146/annurev.astro.35.1.309Finzell, T., Chomiuk, L., Metzger, B. D., et al. 2018, ApJ,852, 108, doi: 10.3847/1538-4357/aaa12aFitzpatrick, E. L. 1999, PASP, 111, 63, doi: 10.1086/316293Foreman-Mackey, D., Hogg, D. W., Lang, D., & Goodman,J. 2013, PASP, 125, 306, doi: 10.1086/670067Friedjung, M. 1966, MNRAS, 132, 317,doi: 10.1093/mnras/132.2.317—. 1987, A&A, 180, 155—. 2011, A&A, 536, A97,doi: 10.1051/0004-6361/201016397Friedjung, M., & Duerbeck, H. W. 1993, Models of classicaland recurrent novae, Vol. 507, 371–412Gallagher, J. S., & Starrfield, S. 1978, ARA&A, 16, 171,doi: 10.1146/annurev.aa.16.090178.001131Gazak, J. Z., Kudritzki, R., Evans, C., et al. 2015, ApJ,805, 182, doi: 10.1088/0004-637X/805/2/182Gehrels, N., Chincarini, G., Giommi, P., et al. 2004, ApJ,611, 1005, doi: 10.1086/422091Gehrz, R. D. 1988, ARA&A, 26, 377,doi: 10.1146/annurev.aa.26.090188.002113Gehrz, R. D., Grasdalen, G. L., Hackwell, J. A., & Ney,E. P. 1980a, ApJ, 237, 855, doi: 10.1086/157934Gehrz, R. D., Hackwell, J. A., Grasdalen, G. I., et al.1980b, ApJ, 239, 570, doi: 10.1086/158143Gehrz, R. D., Truran, J. W., Williams, R. E., & Starrfield,S. 1998, PASP, 110, 3, doi: 10.1086/316107Gehrz, R. D., Roellig, T. L., Werner, M. W., et al. 2007,Review of Scientific Instruments, 78, 011302,doi: 10.1063/1.2431313Gehrz, R. D., Evans, A., Woodward, C. E., et al. 2018,ApJ, 858, 78, doi: 10.3847/1538-4357/aaba81Gimeno, G., Roth, K., Chiboucas, K., et al. 2016, inSociety of Photo-Optical Instrumentation Engineers(SPIE) Conference Series, Vol. 9908, Proc. SPIE, 99082S,doi: 10.1117/12.2233883Gonzaga, S., Hack, W., Fruchter, A., & Mack, J. 2012, TheDrizzlePac Handbook Jencson et al.
Gordon, A. C., Aydi, E., Page, K. L., et al. 2020, arXive-prints, arXiv:2010.15930.https://arxiv.org/abs/2010.15930Groenewegen, M. A. T., Sloan, G. C., Soszy´nski, I., &Petersen, E. A. 2009, A&A, 506, 1277,doi: 10.1051/0004-6361/200912678Hachisu, I., & Kato, M. 2006, ApJS, 167, 59,doi: 10.1086/508063—. 2007, ApJ, 662, 552, doi: 10.1086/517600—. 2018, ApJS, 237, 4, doi: 10.3847/1538-4365/aac833—. 2019, ApJS, 242, 18, doi: 10.3847/1538-4365/ab1b43Hachisu, I., Kato, M., Kiyota, S., et al. 2006, ApJL, 651,L141, doi: 10.1086/509615Hack, W. J., Dencheva, N., Fruchter, A. S., et al. 2012, inAmerican Astronomical Society Meeting Abstracts, Vol.220, American Astronomical Society Meeting Abstracts