MNRAS 000, 1–23 (2020) Preprint 9 November 2020 Compiled using MNRAS LATEX style file v3.0
SN 2019muj – a well-observed Type Iax supernova that bridges the
luminosity gap of the class
Barnabás Barna,1,2★ Tamás Szalai,2,3 Saurabh W. Jha,4 Yssavo Camacho-Neves,4
Lindsey Kwok,4 Ryan J. Foley,5 Charles D. Kilpatrick,5 David A. Coulter,5
Georgios Dimitriadis,5 Armin Rest,6,7 César Rojas-Bravo,5 Matthew R. Siebert,5
Peter J. Brown,8,9 Jamison Burke,10,11 Estefania Padilla Gonzalez,10,11
Daichi Hiramatsu,10,11 D. Andrew Howell,10,11 Curtis McCully,10 Craig Pellegrino,10,11
Matthew Dobson,12 Stephen J. Smartt,12 Jonathan J. Swift,13 Holland Stacey,13
Mohammed Rahman,13 David J. Sand,14 Jennifer Andrews,14 Samuel Wyatt,14
Eric Y. Hsiao,15 Joseph P. Anderson,16 Ting-Wan Chen,17 Massimo Della Valle,18
Lluís Galbany,19 Mariusz Gromadzki,20 Cosimo Inserra,21 Joe Lyman,22
Mark Magee,23 Kate Maguire,23 Tomás E. Müller-Bravo,24 Matt Nicholl,25,26
Shubham Srivastav,12 Steven C. Williams27,28
1Astronomical Institute, Czech Academy of Science, Bocni II 1401, 141 00 Prague, Czech Republic
2Department of Optics and Quantum Electronics, University of Szeged, Dóm tér 9, Szeged, 6720 Hungary
3Konkoly Observatory, Research Centre for Astronomy and Earth Sciences, H-1121 Budapest, Konkoly Thege Miklós út 15-17, Hungary
4Department of Physics and Astronomy, Rutgers, the State University of New Jersey, 136 Frelinghuysen Road, Piscataway, NJ 08854, USA
5Department of Astronomy and Astrophysics, University of California, Santa Cruz, CA 95064, USA
6Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218, USA
7Department of Physics and Astronomy, The Johns Hopkins University, Baltimore, MD 21218, USA
8Department of Physics and Astronomy, Texas A&M University, 4242 TAMU, College Station, TX 77843, USA
9George P. and Cynthia Woods Mitchell Institute for Fundamental Physics & Astronomy, USA
10Las Cumbres Observatory, 6740 Cortona Drive, Suite 102, Goleta, CA 93117-5575, USA
11Department of Physics, University of California, Santa Barbara, CA 93106-9530, USA
12Astrophysics Research Centre, School of Mathematics and Physics, Queen’s University Belfast, Belfast, BT7 1NN, UK
13Thacher Observatory, Thacher School, 5025 Thacher Rd. Ojai, CA 93023, USA
14Steward Observatory, University of Arizona, 933 North Cherry Avenue, Tucson, AZ 85721-0065, US
15Department of Physics, Florida State University, 77 Chieftan Way, Tallahassee, FL 32306, USA
16European Southern Observatory, Alonso de Córdova 3107, Casilla 19 Santiago, Chile
17The Oskar Klein Centre, Department of Astronomy, Stockholm University, AlbaNova, SE-10691 Stockholm, Sweden
18INAF- Capodimonte, Naples, Salita Moiariello 16, 80131, Naples, Italy
19Departamento de Física Teórica y del Cosmos, Universidad de Granada, E-18071 Granada, Spain
20Astronomical Observatory, University of Warsaw, Al. Ujazdowskie 4, 00-478 Warszawa, Poland
21School of Physics & Astronomy, Cardiff University, Queens Buildings, The Parade, Cardiff, CF24 3AA, UK
22Department of Physics, University of Warwick, Coventry CV4 7AL, UK
23School of Physics, Trinity College Dublin, The University of Dublin, Dublin 2, Ireland
24School of Physics and Astronomy, University of Southampton, Southampton, Hampshire, SO17 1BJ, UK
25Birmingham Institute for Gravitational Wave Astronomy and School of Physics and Astronomy, University of Birmingham,
Birmingham B15 2TT, UK
26Institute for Astronomy, University of Edinburgh, Royal Observatory, Blackford Hill, EH9 3HJ, UK
27Finnish Centre for Astronomy with ESO (FINCA), Quantum, Vesilinnantie 5, University of Turku, 20014 Turku, Finland
28Department of Physics and Astronomy, University of Turku, 20014 Turku, Finland
Accepted XXX. Received YYY; in original form ZZZ
© 2020 The Authors
ar
X
iv
:2
01
1.
03
06
8v
1 
 [a
str
o-
ph
.H
E]
  5
 N
ov
 20
20
2 B. Barna et al.
ABSTRACT
We present early-time (𝑡 < +50 days) observations of SN 2019muj (= ASASSN-19tr), one
of the best-observed members of the peculiar SN Iax class. Ultraviolet and optical photometric
and optical and near-infrared spectroscopic follow-up started from ∼5 days before maximum
light (𝑡max (𝐵) on 58 707.8 MJD) and covers the photospheric phase. The early observations
allow us to estimate the physical properties of the ejecta and characterize the possible diver-
gence from a uniform chemical abundance structure. The estimated bolometric light curve
peaks at 1.05 × 1042 erg s−1 and indicates that only 0.031 𝑀 of 56Ni was produced, making
SN 2019muj a moderate luminosity object in the Iax class with peak absolute magnitude of
𝑀V = −16.4 mag. The estimated date of explosion is 𝑡0 = 58 698.2 MJD and implies a short
rise time of 𝑡rise = 9.6 days in 𝐵-band. We fit of the spectroscopic data by synthetic spectra,
calculated via the radiative transfer code TARDIS. Adopting the partially stratified abundance
template based on brighter SNe Iax provides a good match with SN 2019muj. However, with-
out earlier spectra, the need for stratification cannot be stated in most of the elements, except
carbon, which is allowed to appear in the outer layers only. SN 2019muj provides a unique
opportunity to link extremely low-luminosity SNe Iax to well-studied, brighter SNe Iax.
Key words: supernovae: general – supernovae: individual: SN 2019muj (ASASSN-19tr)
1 INTRODUCTION
Althoughmost normal Type Ia supernovae (SNe Ia) form a homoge-
neous class (often referred to as ‘normal’ or ‘Branch-normal’ SNe,
Branch et al. 2006), a growing number of peculiar thermonuclear ex-
plosions are being discovered (Taubenberger 2017). These objects
are also assumed to originate from a carbon-oxygen white dwarf
(C/O WD), but do not follow the correlation between the shape
of their light curve (LC) and their peak luminosity (Phillips et al.
1993) controlled by the synthesized amount of 56Ni (Arnett 1982).
These peculiar thermonuclear SNe also show unusual observables
compared to those of normal SNe Ia.
One of these subclasses is the group of Type Iax SNe (SNe
Iax; Foley et al. 2013), or, as named after the first discovered object,
‘2002cx-like’ SNe (Li et al. 2003; Jha et al. 2006). Jha (2017)
present a collection of ∼60 SNe Iax, making the group probably
the most numerous of the Ia subclasses. However, the exact rate
of SNe Iax is highly uncertain as Foley et al. (2013) found it to
be 31+17−13% of the normal SN Ia rate based on a volume-limited
sample, consistent with results of Miller et al. (2017). The peak
absolute magnitudes of SN Iax cover a wide range between -14.0 –
-18.4 mag, from the extremely faint SNe 2008ha (Foley et al. 2009;
Valenti et al. 2009) and 2019gsc (Srivastav et al. 2020; Tomasella
et al. 2020) to the nearly normal Ia-bright SNe 2011ay (Szalai et
al. 2015; Barna et al. 2017) and 2012Z (Stritzinger et al. 2015).
The distribution is probably dominated by the faint objects and the
number of the brighter (MV < −17.5 mag) ones is only 2-20% of
the total SNe Iax population (Li et al. 2011; Graur et al. 2017).
The expansion velocities also have a very diverse nature, showing a
photospheric velocity (𝑣phot) of 2,000–9,000 km s−1 at the moment
of maximum light, which is significantly lower than that of SNe
Ia (typically > 10 000 km s−1, Silverman et al. 2015). A general
correlation between the peak luminosity and the expansion velocity
(practically, 𝑣phot at the moment of maximum light) of SNe Iax has
been proposed by McClelland et al. (2010). However, the level of
such correlation is under debate, mainly because of reported outliers
★ E-mail: bbarna@titan.physx.u-szeged.hu
like SNe 2009ku (Narayan et al. 2011) and 2014ck (Tomasella et al.
2016).
The spectral analysis of SNe Iax explored a similar set of lines
in the optical regime as in the case of normal SNe Ia (Phillips et al.
2007; Foley et al. 2010a). However, the early phases are dominated
mainly by the features of iron-group elements (IGEs), especially
the absorption features of Fe II and Fe III. While Si II and Ca II
are always present, their characteristic lines are not optically thick,
and high-velocity features (Silverman et al. 2015) have not been
observed in SNe Iax spectra.
McCully et al. (2014) reported the discovery of a luminous
blue point source coincident with the location of SN 2012Z in the
pre-explosion HST images. The observed source was similar to a
helium nova system, leading to its interpretation as the donor star for
the exploding WD. SN 2012Z could be the first ever thermonuclear
SN with a detected progenitor system, and thus linked to the single
degenerate (SD) scenario by direct observational evidence. The
model of SNe Iax originating fromWD-He star systems is supported
by the generally young ages of SNe Iax environments (Lyman et al.
2018; Takaro et al. 2020), though at least one SN Iax exploded in
an elliptical galaxy (Foley et al. 2010b). The detection of helium
is also reported in the spectra of SNe 2004cs and 2007J (Foley et
al. 2013; Jacobson-Galán et al. 2019; Magee et al. 2019), but there
is a debate on the real nature of these two objects (see White et
al. 2015; Foley et al. 2016). Moreover, the apparent lack of helium
in the majority of known SNe Iax (Magee et al. 2019) shows the
necessity of further investigation of this point.
However, one of the most critical questions regarding the Iax
class is whether these SNe with widely varying luminosities could
originate from the same explosion scenario? Among the existing
theories, the most popular scenario (Jha 2017) in the literature is
the pure deflagration of a Chandrasekhar-mass C/O WD leaving a
bound remnant behind, which is supported by the predictions of
hydrodynamic explosion models (Jordan et al. 2012b; Kromer et al.
2013; Fink et al. 2014; Kromer et al. 2015). A modified version of
this theory, in which the progenitor is a hybrid CONeWD, has been
also published in several papers to explain the origin of the faintest
members of the class (see e.g. Denissenkov et al. 2015; Kromer et
al. 2015; Bravo et al. 2016).
MNRAS 000, 1–23 (2020)
Type Iax SN 2019muj 3
As a further discrepancy compared to normal SN Ia is that SNe
Iax never enter into a truly nebular phase like normal SNe Ia, instead
showing both P Cygni profiles of permitted lines and emission of
forbidden transitions∼200 days after the explosion (Jha et al. 2006).
A possible explanation of the late-time evolution comes from Foley
et al. (2014), who proposed a super-Eddington wind from the bound
remnant as the source of the late-time photosphere. The bound
remnant was originally predicted by hydrodynamic simulations of
pure deflagration of a Chandrasekhar-mass WD (Fink et al. 2014).
Such a remnant may have been seen in late-time observations of SN
2008ha (Foley et al. 2014). Understanding the dual nature of these
spectra requires further investigation and observations of extremely
late epochs.
The deflagration scenarios that fail to completely unbind the
progenitor WD are able to reproduce the extremely diverse lumi-
nosities and kinetic energies of the Iax class. On the other hand, the
synthetic light curves from the pure deflagration models of Kromer
et al. (2013) and Fink et al. (2014) seem to evolve too fast after their
maxima, indicating excessively low optical depths, i.e. insufficient
ejecta masses. Nevertheless, it has to be noted that only a small
range of initial parameters of deflagration models leaving a bound
remnant has been sampled in these studies; therefore, it cannot be
excluded that with different initial conditions (ignition geometry,
density, composition) larger ejecta masses (for a given 56Ni mass)
are possible in the deflagration scenario.
Apart from the pure deflagration, other single-degenerate (SD)
explosion scenarios, which may partially explain the observables
of SNe Iax, are the deflagration-to-detonation transition (DDT;
Khokhlov et al. 1991a) scenarios. In classical DDT models, the
WD becomes unbound during the deflagration phase; in a varia-
tion, the pulsational delayed detonation (PDD) scenario, the WD
remains bound at the end of the deflagration phase, then undergoes
a pulsation followed by a delayed detonation (Ivanova, Imshennik
& Chechetkin 1974; Khokhlov et al. 1991b; Höflich, Khokhlov &
Wheeler 1995). The detonation can happen via a sudden energy
release in a confined fluid volume; this group of models includes
gravitationally confined detonations (GCD, see e.g. Plewa, Calder&
Lamb 2004; Jordan et al. 2008), ‘pulsationally assisted’ GCD mod-
els (Jordan et al. 2012a) and pulsating reverse detonation (PRD)
models (Bravo & García-Senz 2006; Bravo et al. 2009). From all
of these delayed detonation scenarios, only PDD models have been
used for direct comparison with the data of an SN Iax (SN 2012Z,
Stritzinger et al. 2015).
A common feature of all the DDT scenarios is the stratified
ejecta structure caused by the mechanism of the transition from
deflagration-to-detonation. On the other hand, one of the character-
istic ejecta properties of the pure deflagration scenarios, is the well-
mixed ejecta, resulting in nearly constant chemical abundances.
Indications for the mixed ejecta structure based on the near-infrared
light curves have been reported since the discovery of SNe Iax (Li
et al. 2003; Jha 2017). As it was shown by Kasen (2006), the IJHK
light curves provide a diagnostic tool for the mixing of IGEs, which
elements drive the formation of a secondary maximum by drastic
opacity change due to their recombination (see e.g. Pinto&Eastman
2000; Höflich, Khokhlov &Wheeler 1995; Höflich et al. 2017). Al-
though several parameters (e.g. progenitor metallicity, abundance
of IMEs) have also a significant impact on its formation, the exis-
tence of the NIR secondary peak is mainly a consequence of the
concentration of IGEs in the central region of SN Ia ejecta. Phillips
et al. (2007) showed, that the NIR LCs of SN 2005hk, where the first
and second maxima are indistinguishable, is similar to that of the
SN Ia model containing 0.6 𝑀 of 56Ni in a completely homoge-
Figure 1. The 𝑔𝑟𝑖 composite image of the field with the VV 525 galaxy and
SN2019muj obtained by Swope telescope at the LasCampanasObservatory.
nized composition structure (Kasen 2006). The authors argued that
despite the different model, the lack of a separated secondary peak
in the NIR LCs is a direct evidence for mixed ejecta structure of
SNe Iax.
At the same time, the pure deflagration producing a bound rem-
nant picture might contradict the bright SN Iax 2012Z (Stritzinger
et al. 2015), in which the velocity distribution of intermediate mass
elements (IMEs) do not indicate significant mixing in the outer
ejecta. Barna et al. (2018) performed abundance tomography for a
small sample of SNe Iax and found a stratified structure as a feasible
solution for the outer regions of SNe Iax, however, the impact of the
outermost layers are strongly affected by the choice of the density
profile.
Despite the fact that the former results in this question are
contradictory, constraining the chemical structure could be the key
to either confirm the assumption of the pure deflagration scenario
or reject it. Note that with the term (pure) deflagration we refer only
on the weak deflagration scenarios hereafter, which leave behind
a bound remnant (Fink et al. 2014; Kromer et al. 2015), and thus,
provide a promising explanation for the SN Iax class.
A promising way to test the theory whether SN Iax share a
similar origin (i.e. the pure deflagration of a CO/CONe 𝑀Ch WD)
is the investigation of the link between the two extremes of the
Iax class, the relatively luminous and extremely faint objects. A
continuous distribution of physical and chemical properties, as well
as their possible correlations with the peak luminosity, favor the idea
that SNe Iax may form a one (or maybe a few) parameter family. So
far, these efforts are limited by the lack ofwell-observed,moderately
luminous SNe Iax in between the extremes. The subject of our
current study, SN 2019muj, now provides a good opportunity to
carry out a detailed analysis of an SN Iax belongs to this luminosity
gap.
The paper is structured in the following order. In Section 2, we
introduce SN 2019muj and give an overview of the collected data.
The light curves, spectra, and the estimated ultraviolet (UV)-optical
MNRAS 000, 1–23 (2020)
4 B. Barna et al.
spectral energy distributions (SEDs) are shown in Section 3. We
summarize our conclusions in Section 4.
2 OBSERVATIONS AND DATA REDUCTION
2.1 SN 2019muj
SN 2019muj (ASASSN-19tr) was discovered (Brimacombe et al.
2019) by theAll SkyAutomated Survey for SuperNovae (ASAS-SN;
Shappee et al. 2014) on 7 August 2019, UT 09:36 (MJD 58 702.4).
The location in the sky (R.A. 02:26:18.472, Dec. −09:50:09.92,
2000.0) is associated with VV 525, a blue star-forming spiral galaxy
(see Fig. 1) with morphological type of SAB(s)dm: (de Vaucouleurs
et al. 1991) at redshift of 𝑧 = 0.007035. Considering the low redshift,
the K-correction is assumed negligible (Hamuy et al. 1993; Phillips
et al. 2007). We adopted 𝑑 = 34.1 ± 2.9Mpc for the distance of the
galaxy from Tully et al. (2013, 2016) (using the lower uncertainty
value from the two studies) calculated by the Tully-Fisher method,
assuming 𝐻0 = 70 km s−1 Mpc−1. This value is slightly higher than
the distance of 𝑑z = 30.2Mpc estimated from themeasured redshift.
Hereafter, we use the former distance for spectral modelling and the
calculation of the quasi-bolometric LC.
The SN is in the outskirts of the galaxy at a 48” projected
separation (8 kpc) from its center (see Fig. 1). Because of the
lack of significant Na I D line absorption at the redshift of the
galaxy, we assume the host galaxy reddening at the SN position as
𝐸 (𝐵 − 𝑉)host = 0.0, while the Galactic contribution is adopted as
𝐸 (𝐵 − 𝑉)Gal = 0.023 from Schlafly & Finkbeiner (2011). More-
over, because the SN is well-separated from its host galaxy, image
subtraction was not required during the data reduction.
At discovery, the apparent magnitude of SN 2019muj was
17.4 mag in the g-band. Based on the first spectrum, obtained at
MJD 58702.79, the Superfit classification algorithm (Howell et al.
2005) found SN 2019muj similar to SN 2005hk around a week
before maximum, thus, categorized as a SN Iax (Hiramatsu et al.
2019). These properties indicated SN 2019muj to be the brightest
SN Iax with pre-maximum discovery since the discovery of SN
2014dt and intensive follow-up campaigns were started by several
collaborations and observatories.
2.2 Photometry
The field of SN 2019muj was observed by the Asteroid Terrestrial-
impact Last Alert System program (ATLAS; Tonry et al. 2018)
before its discovery, providing a deep (∼21 mag in orange-band)
non-detection limit before MJD 58 700 (see in Fig. 2). The moni-
toring in 𝑜- and 𝑐-bands continued through the whole observable
time range.
The ATLAS data provide a strong constraint on the explosion
epoch. ATLAS detected SN2019muj on MJD 58 702.53, just a few
hours after the ASAS-SN discovery. As with all ATLAS transients,
forced photometrywas performed (as described in Smith et al. 2020)
around the time of discovery, with the fluxes presented in Table A1
and plotted in Fig. 2. There is a marginal 1.8𝜎 detection on MJD
58 700.52, two days before discovery. This can either be interpreted
as a detection at 𝑜 = 20.5 ± 0.6 mag or a 3𝜎 upper limit at 𝑜 < 20
mag. Our analysis is not affected by either choice, as discussed
further in Sec. 3.1.
SN2019mujwas also observedwith theNeilGehrels SwiftOb-
servatory Ultraviolet/Optical Telescope (UVOT) (hereafter Swift;
Burrows et al. 2005; Roming et al. 2005) on 13 epochs between
Figure 2.The pre-maximumATLAS photometry in cyan- and orange-filters
at the location of SN 2019muj. The dashed lines represent the fit of the early
c-, o- and g-band LCs (latter observations are also plotted ) with power-law
index n=1.3 (blue) and n=2.0 (red). Before MJD 58 700, the arrows show
the 3𝜎 upper limits. Note that the first detection has only 2𝜎 significance.
For further details see Sec. 3.1.
Figure 3. Swift photometry of SN 2019muj.
MJD 58 702.8 and 58 735.7. The Swift data were reduced using
the pipeline of the Swift Optical Ultraviolet Supernova Archive
(SOUSA; Brown et al. 2014), without subtraction of the host galaxy
flux. Swift light curves (see Fig. 3) provide unique UV information
in UVW2, UVM2 and UVW1 filters and supplementary informa-
tion to the ground-based data in 𝑈𝐵𝑉 (see Table A2). Note that
flux conversion factors are spectrum dependent and the UVW2 and
UVW1 filters are heavily contaminated by the optical fluxes (Brown
et al. 2010, 2016). This red leak may cause high uncertainties for
the magnitudes of the wide bands.
MNRAS 000, 1–23 (2020)
Type Iax SN 2019muj 5
Ground-based photometric follow-up was obtained with the
Sinistro cameras on Las Cumbres Observatory network (LCO) of
1-m telescopes at Sutherland (South Africa), CTIO (Chile), Sid-
ing Spring (Australia), and McDonald (USA; Brown et al. 2013),
through the Global Supernova Project (GSP). Data were reduced us-
inglcogtsnpipe (Valenti et al. 2016), a PyRAF-based photometric
reduction pipeline, by performing PSF-fitting photometry. Images
in Landolt filters were calibrated to Vega magnitudes, and zero-
points were calculated nightly from Landolt standard fields taken
on the same night by the telescope, the corresponding zeropoints
and color terms are also calculated in the natural system. Since the
target is in the Sloan field, zeropoints for images in Sloan filters
were calculated by using Sloan AB magnitudes of stars in the same
field as the object. The estimated uncertainties take into account
zero points noise, Poisson noise, and the read out noise.
We imaged 2019muj from MJD 58 702.5 to 58 908.0 with the
Direct 4k×4k imager on the Swope 1-m telescope at Las Campanas
Observatory, Chile. We performed observations in Sloan 𝑢𝑔𝑟𝑖 fil-
ters and JohnsonBV filters. All bias-subtraction, flat-fielding, image
stitching, registration, and photometric calibration were performed
using photpipe (Rest et al. 2005) as described in Kilpatrick et al.
(2018). No host-galaxy subtraction was performed, which would
cause no significant differences, as SN 2019muj is appeared at the
edge of the galaxy. We calibrated our photometry in the Swope
natural system (Krisciunas et al. 2017) using photometry of stars
in the same field as 2019muj from the Pan-STARRS DR1 catalog
(Flewelling et al. 2016) and transformed using the Supercal method
(Scolnic et al. 2015). Final photometry was obtained using DoPhot
(Schechter, Mateo & Saha 1993) on the reduced images. For better
comparison, the BV magnitudes are then transformed to the Vega
magnitudes (Blanton & Roweis 2007). The uncertainties are com-
puted by combining the statistical uncertainties in the photometry,
and the systematic uncertainties due to the calibration and transfor-
mations used.
Additional optical imaging was collected by Thacher Observa-
tory (Swift & Vyhnal 2018) in Ojai, CA from 9 August 2019 to 31
August 2019. The observations were taken in Johnson 𝑉 and Sloan
𝑔𝑟𝑖𝑧 filters. All reductions were performed in photpipe (Rest et al.
2005) following the same procedures as the Swope data but with
photometric calibration in the PS1 system for 𝑔𝑟𝑖𝑧 and using PS1
magnitudes transformed to Johnson 𝑉 band using the equations
in Jester et al. (2005). In addition, the Thacher telescope camera
is non-cryogenic, and so we performed dark current subtractions
using 60 second dark frames obtained in the same instrumental
configuration and on the same night as our science frames.
The ground-based photometric data including the 𝐵𝑉𝑅 and
𝑢𝑔𝑟𝑖𝑧 magnitudes can be found in Tables A3 and A4 respectively,
while the LCs are shown in Fig. 4. We compare the absolute𝑉-band
LC of SN 2019muj to a selection of other SNe Iax that populate the
observed range of absolute magnitudes of this class in Fig. 5.
2.3 Spectroscopy
Optical spectroscopy of SN 2019muj was obtained by the 9.2-m
Southern African Large Telescope (SALT) with the Robert Stobie
Spectrograph (RSS; Smith et al. 2006) through Rutgers University
program 2019-1-MLT-004 (PI: SWJ). The observations were made
with the PG0900 grating and 1.5” wide longslit with a typical spec-
tral resolution 𝑅 = 𝜆/Δ𝜆 ≈ 1000. Exposures were taken in four
grating tilt positions to cover the optical spectrum from 3500 to
9300 Å. The data were reduced using a custom pipeline based on
Figure 4.Ground-based photometry of SN 2019muj. The filled circles, open
circles, and triangles represent the observations of LCO, Swope and Thacher
observatories, respectively.
standard Pyraf spectral reduction routines and the PySALT package
(Crawford et al. 2010).
During the first 50 days after explosion, spectra were also
obtained five times with the ESO Faint Object Spectrograph and
Camera version 2 (EFOSC2, Buzzoni et al. 1984) at the 3.6-m New
Technology Telescope (NTT) of European Southern Observatory
as part of the ePESSTO+ collaboration (Smartt et al. 2015). These
epochs were observed with grisms #11 (covering 3300-7500 Å and
#16 (6000-9900 Å), or #13 (3500-9300 Å) (see Table A5). The data
reduction of the NTT spectra was performed using the PESSTO
pipeline1 (Smartt et al. 2015).
LCO optical spectra were taken with the FLOYDS spectro-
graphs mounted on the 2-m Faulkes Telescope North and South
at Haleakala (USA) and Siding Spring (Australia), respectively,
1 https://github.com/svalenti/pessto
MNRAS 000, 1–23 (2020)
6 B. Barna et al.
Figure 5. Comparisons of absolute 𝑉 -band light curves of SNe Iax. References to the listed SNe Iax can be found in Szalai et al. (2015) for SN 2011ay,
Yamanaka et al. (2015) and Stritzinger et al. (2015) for SN 2012Z, Phillips et al. (2007) for SN 2005hk, Li et al. (2003) for SN 2002cx, Tomasella et al. (2016)
for SN 2014ck, Magee et al. (2016) for SN 2015H, Stritzinger et al. (2014) for SNe 2008ha and 2010ae. Note that in the case of SN 2019gsc (Srivastav et al.
2020; Tomasella et al. 2020), the 𝑉 -band LC is estimated from 𝑔- and 𝑟 -band magnitudes adopting the transformation presented in Tonry et al. (2012).
MNRAS 000, 1–23 (2020)
Type Iax SN 2019muj 7
through the Global Supernova Project. A 2” slit was placed on
the target at the parallactic angle (Filippenko et al. 1982). One-
dimensional spectra were extracted, reduced, and calibrated follow-
ing standard procedures using the FLOYDS pipeline 2 (Valenti et
al. 2014).
Observations using the Kast spectrograph on the Shane-3m
telescope of the Lick Observatory (Miller & Stone 1993) were
made using the 2”-wide slit, the 452/3306 grism (blue side), the
300/7500 grating (red side) and the D57 dichroic. These data were
reduced using a custom PyRAF-based KAST pipeline3, which ac-
counts for bias-subtracting, optical flat-fielding, amplifier crosstalk,
background and sky subtraction, flux calibration and telluric correc-
tions using a standard star observed on the same night (procedure
described in Silverman, Kong & Filippenko 2020) and at a similar
airmass.
Finally, two NIR spectra were taken with the FLAMINGOS-2
spectrograph (Eikenberry et al. 2008) on Gemini-South. The data
acquisition and reduction are similar to that described in Sand et
al. (2018). The JH grism and 0.72” width longslit were employed,
yielding a wavelength range of ∼1.0 – 1.8 𝜇m and 𝑅 ∼ 1000. All
data was taken at the parallactic angle with a standard ABBA pat-
tern, and an A0V telluric standard was observed near in airmass and
adjacent in time to the science exposures. The data were reduced
in a standard way using the F2 PyRAF package provided by Gem-
ini Observatory, with image detrending, sky subtraction of the AB
pairs, spectral extraction, wavelength calibration and spectral com-
bination. The telluric corrections and simultaneous flux calibrations
were determined using the XTELLCOR package (Vacca, Cushing &
Rayner 2020).
The optical spectra of SN 2019muj are plotted in Fig. 8, while
the log of the spectroscopic observations can be found in Table A5.
Our spectroscopic data were not always taken under spec-
trophotometric conditions; clouds and slit losses can lead to errors
in the overall flux normalization. As an example, the pupil illu-
mination of SALT, moreover, changes during the observation, so
even relative flux calibration from different grating angles can be
difficult. Synthetic photometry of the spectra can differ from the
measured photometry by a few tenths of a magnitude. For these
reasons, we choose to color match all of our spectroscopic data to
match the observed photometry. After this color-matching, the syn-
thetic photometry of the spectra reproduces the photometric data to
better than 0.05 mag across optical filters.
All the spectra are scaled and color-matched to the observed
broadband photometry. Subsequently, the spectra were corrected
for redshift and reddening according to the extinction function of
Fitzpatrick & Massa (2007). All data will publicly released via
WIseREP4.
3 DISCUSSION AND RESULTS
3.1 Photometric analysis
The ATLAS photometry provides a reliable and deep non-detection
limit (𝑜 > 19.7 mag, using 3𝜎 upper limit) on MJD 58 696.58
and a marginal (2𝜎 significance) detection on MJD 58 700.52 (as
shown in Fig. 2 and Tab. A1). Assuming the so-called expanding
fireball model, the observed flux increases as a quadratic function of
2 https://github.com/svalenti/FLOYDS_pipeline
3 https://github.com/msiebert1/UCSC_spectral_pipeline
4 https://wiserep.weizmann.ac.il/
Figure 6. The color evolution of SN 2019muj compared to that of SNe Iax
that provide examples of more and less luminous events with respect to SN
2019muj. All magnitudes are corrected for galactic and host extinction with
data taken from Stritzinger et al. (2014) for SN 2008ha and Phillips et al.
(2007) for SN 2005hk.
Figure 7. The observed UVW2-, UVM2- and UVW1-band light curves of
SNe 2011ay (Szalai et al. 2015), 2012Z (Stritzinger et al. 2015), 2005hk
(Phillips et al. 2007) and 2019muj. For better comparison, the light curves
are shifted along both the horizontal and vertical axes tomatch at themoment
of their V-band maximum.
time (𝐹 ∼ 𝑡2exp) before the epoch of maximum light (Arnett 1982;
Riess et al. 1999; Nugent et al. 2011). Note that because of the
lack of coverage in the first few days after explosion, the fitting of
different LCs may deliver a different 𝑡exp. To determine the time
of the explosion more accurately, we simultaneously fit the g-, c-
and o-band LCs including the earliest observation and the best pre-
MNRAS 000, 1–23 (2020)
8 B. Barna et al.
Figure 8.Spectroscopic follow-up of SN2019muj obtainedwith SALT/RSS,
Lick/Kast, NTT/EFOSC2 and LCO/FLOYDS spectrographs. The epochs
show the days with respect to B-maximum. The observation log of spectra
can be found in Tab. A5. The positions of strong telluric lines are marked
with ⊕.
Figure 9. Comparison between spectra of Type Iax SNe 2011ay, 2005hk,
2010ae, and 2019muj obtained at similar epochs. References can be found
in Szalai et al. (2015), Phillips et al. (2007) and Stritzinger et al. (2014),
respectively. The orange and black vertical lines show the absorptionminima
of the Si II 𝜆6355 and C II 𝜆6580 lines in the SN 2019muj spectra. With
no visible absorption of the Si II 𝜆6355, the orange line show its position
blueshifted with 5600 km s −1 (the 𝑣phot of the first TARDIS model, see
below).
MNRAS 000, 1–23 (2020)
Type Iax SN 2019muj 9
Figure 10. Zoomed insets of the spectra of SNe 2005hk, 2010ae, and 2019muj plotted in Fig. 9. For better comparison, the spectra of SNe 2005hk and 2010ha
are scaled to those of SN 2019muj, and also shifted to match the photospheric velocity of SN 2019muj estimated from the Doppler-shift of the C II 𝜆6580
absorption. The black labels show the ions forming the prominent absorption features. In some cases, where the emission peaks provide clear delimiters for
the spectral feature, the pseudo-equivalent widths of the lines are also estimated and shown with the corresponding color codes. Examples for the different
pseudo-continuum are indicated with dashed lines in the plot of the Si II 𝜆6355 feature in the middle row.
maximum time resolution. The fitting function is:
𝐹g = 𝑎 ·
(
𝑡 − 𝑡exp
)𝑛
, (1)
where 𝑎 is a fitting parameter, while 𝑛 = 2 is kept fixed at first.
The resulting date of explosion is found to be t0 = 58 694.4 MJD,
which seems too early regarding to the constraints provided by
the o-band LC. However, another critical aspect of this analysis is
the choice of power-law index. Detailed studies reported different
power-law indices for samples of normal SNe Ia with mean values
of n = 2.20 – 2.44 (Firth et al. 2015; Ganeshalingam, Li & Filipenko
2011). However, our pre-maximum dataset for SN 2019muj is not
sufficient to handle 𝑛 as a free parameter, which would lead to highly
unlikely results (𝑛 = 0.7 with constrained date of MJD 58 700.5).
As another approach, a fixed value for 𝑛 < 2 can be chosen.
In the case of SN 2015H, Magee et al. (2016) presented the most
detailed pre-maximum LC for a SN Iax and found 𝑛 = 1.3 as the
best-fit power-law index. If we assume similarity to SN 2015H, and
adopt the same vaue of 𝑛 for Eq. 1 constraining the date of explosion
asMJD 58 698.1, which is a more realistic date and compatible with
the ATLAS forced photometry. Note that using 𝑛 = 1.3 is still an
arbitrary choice and no meaningful uncertainty can be estimated.
Thus, we do not claim MJD 58 698.1 as the explosion date of SN
2019muj, instead, we use it only for comparison with the results of
other methods hereafter (see in Sec. 3.2 and 3.3).
We fit the LCs with low-order polynomial functions from the
first observation until two weeks after their maxima. The resulting
LC parameters can be found in Table A6. According to the peak
absolute magnitude in V-band, SN 2019muj is less luminous than
the majority of studied SNe Iax having peak M𝑉 greater than −17
mag. Between these more luminous members of the class and the
few extremely faint objects (M𝑉 > −15mag), there is a gap of well-
studied SNe Iax. According to the peak absolute magnitude (𝑀V =
−16.42 mag), the closest relatives of SN 2019muj are SNe 2004cs
and 2009J (Foley et al. 2013). However, SN 2009J has only post-
maximum spectral epochs, which are insufficient to characterize the
chemical properties via abundance tomography. SN 2004cs is also
categorized as an SN Iax based on its only spectrum obtained at
MNRAS 000, 1–23 (2020)
10 B. Barna et al.
day +42 day, and White et al. (2015) argue that the object does not
even belong to the Iax class. Thus, SN 2019muj provides a unique
opportunity to link the moderately bright SNe Iax to the brighter
ones.
The estimated value of the decline rate is Δ𝑚15 = 2.4 mag
in 𝐵-band and 1.0 mag in 𝑟-band. These values indicate a faster
dimming compared to SN 2005hk (1.56 mag in 𝐵-band, 0.7 mag in
𝑟-band Phillips et al. 2007) and all the more luminous SNe Iax with
detailed photometric analysis (Foley et al. 2013). At the same time,
the decline rates of SN 2019muj are comparable with those of SNe
2010ae, 2008ha (2.43 and 2.03 mag in 𝐵-band; 1.01 and 1.11 mag
in 𝑟-band, respectively, Stritzinger et al. 2014) and 2019gsc (1.14
in 𝑟-band Tomasella et al. 2020), the less energetic explosions in
the class. As a conclusion, the decline rate is not well correlated to
luminosity over the full range of SNe Iax.
The 𝐵 − 𝑉 intrinsic colour evolution of SN 2019muj is com-
pared again to a more luminous (SN 2005hk, Phillips et al. 2007)
and a fainter (SN 2008ha, Stritzinger et al. 2014) object. All SNe
Iax show similar colour evolution in general (Foley et al. 2013):
the originally blue colour curve gets redder with time until 10–15
days after 𝐵-maximum, then a nearly constant color phase starts.
As can be seen in Fig. 6, the 𝐵 − 𝑉 evolution of SN 2019muj is
almost identical to that of SN 2005hk over the first ∼60 days. The
only minor difference is the slightly bluer colour of SN 2019muj
during the constant phase starting at 15 days after 𝐵-maximum.
However, the actual extent of the discrepancy is uncertain because
of the highly scattered photometry at this phase. On the other hand,
SN 2008ha is redder with ∼0.6–0.7 mag at every epoch before the
peak. It shows a similar 𝐵 − 𝑉 value of ∼1.25 mag as SN 2019muj
during the constant phase, though we note the lack of observations
for SN 2008ha between 25 and 40 days after maximum.
SN 2019muj is only the fourth SN Iax in the literature with
UV-photometry obtained by Swift. In Fig. 7, the UVW2, UVM2
and UVW1 LCs are compared to those of SNe 2011ay, 2012Z and
2005hk. All three SNe are more luminous than SN 2019muj. As
one can observe, the post-maximum LCs of SN 2019muj show a
faster fading, which could be a sign of a steep density profile of the
outer ejecta.
3.2 Spectroscopic analysis
Observed spectra of SN 2019muj are scaled and color-matched to
the observed broad-band photometry. The resulting series of spectra
are plotted in Fig. 8, while the log of each observation is listed in
Tab. A5.
In Fig. 9, the spectra of three objects are compared to those
of other SNe Iax that have spectra close in time to SN 2019muj
and that provide examples of more and less luminous events with
respect to SN 2019muj. Although the same spectral lines appear in
all objects, their relative optical depths as well as the shape of the
continuum vary from object to object. For better comparison, some
spectral features at each epochs are displayed in the insets of Fig.
10.
At 6 days before the maximum, SN 2019muj shows only weak
signs of IMEs. The spectrum over 5500 Å is nearly featureless,
with C II 𝜆6580 as the only easily identifiable line. The shorter
wavelengths are dominated by wide Fe II and Fe III features, which,
together with the blue continuum, suggest the presence of a hot
medium (𝑇 > 12 000 K) close to the assumed photosphere. The
appearance of the prominent IGE lines (e.g features at ∼4300 and
∼5000 Å), as well as their pseudo-equivalent widths (pEW) are
strikingly similar to those of SN 2005hk (Phillips et al. 2007) at this
epoch. Note that the pEW-measurements (for a brief description
see e.g. Childress et al. 2013) adopted mainly for normal SNe Ia
are questionable in the case of SNe Iax, because of the the severe
overlaps of the wide absorption features.
At the same time, SN 2010ae (Stritzinger et al. 2014), the
extremely faint SN Iax involved in this comparison, shows strong Si
II 𝜆6355 line and “W” feature of S II, but restrained IGE absorption
lines. These features, which typically require lower photospheric
temperatures (𝑇 < 11 000K), are not identifiable in the spectrum of
SN 2019muj. All these attributes make probable that SN 2019muj
tends to be more similar to the more luminous SNe Iax compared to
the subluminous objects of the Iax class at the pre-maximumepochs.
The only common feature in both spectra is the significant C II
𝜆6580 line, which indicates relatively high carbon abundance in the
outermost region of SN 2019muj. However, the limited wavelength
range of the spectrum of SN 2010ae (and the lack of spectra of
other extremely faint SNe Iax at this epoch) makes this comparison
incomplete.
The match changes around the maximum when the structures
and the estimated pEWs of the IME features of SN 2019muj re-
semble more closely those of SN 2010ae, than those of SN 2015hk.
The strength of the C II 𝜆6580 feature still provide a good match
with that of SN 2010ae, while this feature is barely identifiable
in the spectrum of SN 2005hk. At the same time, the IGE dom-
inated absorption feattures below 6000 Å show a diverse nature.
The pseudo-emission peaks between 4300 and 4700 Å are in better
agreement with the same features of SN 2005hk, while the more
narrow Fe II absorption lines between 4700 and 5150 Å are more
similar to those of SN 2010ha. At ∼12 days after the maximum, the
whole range spectral range of SN 2019muj is nearly a perfect match
to that of SN 2010ae, while the SN 2005hk spectrum shows several
more dominant pseudo-emission peaks (e.g. at 4700 and 5700 Å.)
The different nature of SN 2019muj before and after maximum
light might be the result of the steeper density structure. The thin
outer region allows insight to the deeper, thus, hotter layers, which
changes as the photosphere starts forming in themore dense regions.
Such density structure may resemble more with that of the more
luminous objects in the outer region with a cut-off at the highest
velocities (see the N5def and N20def deflagration models and the
abundance tomographies of SNe Iax, Fink et al. 2014; Barna et al.
2018), and the profile quickly increase toward the innermost regions
(see the N1def and the hybrid N5def deflagration models, Fink et al.
2014; Kromer et al. 2015). It can not be stated whether this behavior
is regular among the moderately luminous SNe Iax, or not, without
studying more objects with approximately the same luminosity.
The best candidates for this purpose are SNe 2014ck (Tomasella et
al. 2016) and 2015H (Magee et al. 2016) so far, both SNe had a
relatively moderate luminosity (still a magnitude brighter than SN
2019muj) and detailed observational data. However, constraining
the density profile (see below) of multiple objects is beyond the
scope of this study.
In order to test the physical structure of the ejecta, we fit the
spectral time series with synthetic spectra calculated via the 1D
Monte Carlo radiative transfer code TARDIS (Kerzendorf & Sim
2014; Kerzendorf et al. 2019; Vogl et al. 2019) in multiple ways.
First, we use the abundance template proposed by Barna et al.
(2018) for the sample of more luminous SNe Iax. The template (see
Fig. 11) is paramterized by a velocity shift of the abundance struc-
ture, measured by the velocity where the 56Ni abundance becomes
dominant. In the case of 2019muj, we use this velocity shift of the
abundance template as a free parameter for fitting the spectra. The
assumed density function is the same as those used by Barna et al.
MNRAS 000, 1–23 (2020)
Type Iax SN 2019muj 11
Figure 11. The adopted abundance template from Barna et al. (2018) shifted with 6500 km s−1 (i.e. the transition velocity between the 56Ni dominated inner
and the O dominated outer regions). The green, cyan and magenta dashed lines represent the density profiles of the N1def (Fink et al. 2014), the N5def_hybrid
(Kromer et al. 2015), and the feasible fit TARDISmodel (this paper) scaled to texp = 100 s, respectively. Note that the lack of earlier spectra and the steep cut-off
in the adopted density profile above ∼6500 km s−1 makes the constraints on abundance structure uncertain in this outer region (highlighted with purple area).
Moreover, the latest TARDIS model has photospheric velocity of 3600 km s−1, thus, we cannot test the chemical abundances below this limit (highlighted with
grey area).
Figure 12. The chemical abundances of the N1def pure deflagration model (Fink et al. 2014) as comparison to Fig. 11. The density profiles of N1def and
our TARDIS model are also shown with dashed lines. Similarly to Fig. 11, the uncertain (above ∼6500 km s−1) and unseen (below 3600 km s−1) regions are
highlighted with purple and grey colors, respectively.
MNRAS 000, 1–23 (2020)
12 B. Barna et al.
Figure 13. Four of the observed spectra of SN 2019muj obtained around maximum (grey), corrected for reddening and redshift, the TARDIS synthetic spectra
assuming the abundance template from Barna et al. (2018) (blue) and the impact of assuming uniform abundances based on the N1def model from Fink et al.
(2014) (red). The purple residuals show the difference between the two synthetic spectra in each panel. Vertical lines indicate the absorption minima of the C
II lines 𝜆4268, 𝜆4746, 𝜆6580 and 𝜆7234 in the residual plots. Key spectral lines formed by other ions in the template model are indicated by black arrows.
(2018, note that the related Eq. 1 in the cited paper is incorrect)
for reproducing the density structure of various deflagration models
of Fink et al. (2014). The density function contains an exponential
inner part, which continues with a shallow cut-off past the velocity
𝑣cut:
𝜌(𝑣, 𝑡exp) =

𝜌0 ·
(
𝑡exp
𝑡0
)−3 · exp (− 𝑣𝑣0 ) for 𝑣 ≤ 𝑣cut
𝜌0 ·
(
𝑡exp
𝑡0
)−3 · exp (− 𝑣𝑣0 ) · 8− 8(𝑣−𝑣cut )2𝑣2cut for 𝑣 > 𝑣cut
(2)
where 𝜌0 is the reference density at the reference time of 𝑡0 = 100
s, 𝑣 is the velocity coordinate and 𝑣0 is reference velocity, which
defines the slope of the density profile. In our fitting strategy, we use
𝜌0, 𝑣0 and 𝑣cut as free parameters, while the date of the explosion
is adopted from the LC analysis. Luminosities and photospheric
velocities are constrained individually for each spectrum.
Note that our approach of spectral fitting is only a simpli-
fied version of the technique called abundance tomography (Stehle
et al. 2005), because the chemical abundances are controlled via
the template. Moreover, our fitting strategy does not result in a
‘best-fit’ model, because the whole parameter-space cannot be fully
explored manually. Instead, we aim to find a feasible solution for the
main characteristics of SN 2019muj assuming physical and chemi-
cal structure which reproduced the main spectral features of several
epoch in the case of more luminous SNe Iax.
We find a feasible solution assuming explosion date of MJD
58 697.5 ± 0.5, which is in good agreement with the previously
estimated value from the early LC fitting (MJD 58 698.1). The core
density is assumed to be 𝜌0 = 0.55 g cm−3, while the slope of
the function in the inner region is 𝑣0 = 1800 km s−1. The density
MNRAS 000, 1–23 (2020)
Type Iax SN 2019muj 13
Figure 14. The two infrared spectra obtained with Gemini-South/FLAMINGOS-2 spectrograph, compared to to Type Iax SNe 2005hk (Kromer et al. 2013)
and 2010ae (Stritzinger et al. 2014) obtained at similar epochs. The TARDIS synthetic spectrum fitted to the optical spectrum and extended to the longer
wavelengths is also plotted in the first panel. The epochs show the days with respect to 𝐵-band maximum. The blue, orange and brown vertical lines show the
absorption minima of the Co II 𝜆15795, 𝜆16064 and 𝜆16361 lines for the second epoch. The wavelength range heavily affected by major telluric absorption
lines are marked with grey stripe.
profile starts to deviate from the pure exponential function above
𝑣cut = 5500 km s−1. In accord with the expectations based on the
loose correlation between the ejecta structure and the luminosity,
these parameters are lower for SN 2019muj compared to those of
previously analyzed SNe Iax (Barna et al. 2018). The constructed
density profile also falls short of reaching the density function of
the N1def model (Fink et al. 2014), but stays above the density
function of the N5def_hybrid model (Kromer et al. 2015), directly
developed for a possible explanation of the extremely faint SNe Iax
(Fig. 11). Note that the photospheric velocity of the latest TARDIS
model of this analysis limits the part of the ejecta we can study.
Thus, we cannot gain any knowledge neither about the density or
the chemical abundances below 3600 km s−1.
As it was indicated in Sec. 1, the impact of the outer regions
on the synthetic spectrum strongly depend on the choice of density
at high velocities. The adopted density profile has a steep cut-off
above ∼6500 km s−1 (see in Fig. 11), which makes the abundance
structure highly uncertain in this region. Thus, one should test,
whether the abundances of specific chemical elements are indeed
sensitive to the outermost velocity region. After constraining the
set of fitting parameters for the abundance template, we calculate
synthetic spectra with constant abundances using the same physical
properties. The mass fractions of these models are adopted from
the N1def model of Fink et al. (2014) (see in Fig. 12), averaging
and normalizing the abundances of the chemical elements listed in
Fig. 11. The comparison between the two sets of synthetic spectra
is plotted in Fig. 13.
The template models show a relatively good agreement with
the observed spectra, especially a week after maximum light, where
the synthetic spectra reproduce almost all the absorption features.
At the earliest epochs, an excess of carbon is found despite the
mass fraction decrease inward of the photosphere. This suggests
that stronger limitation of carbon abundance is required to produce
the spectral lines correctly. Around maximum light, the absorption
features of IMEs do not or just slightly appear, which may require a
more complex treatment of the temperature and density profile.
At the same time, the spectra calculated from the constant
abundance model match similarly well to the observed data, apart
from the spectral features of C II (see Fig. 13). Due to the steep cut-
off in the adopted density profile, testing the ejecta structure far over
the photosphere of the first epoch (𝑣phot = 6 000 km s−1) is highly
uncertain as it is indicated in Figs. 11 and 12. In the regions probed
in our work we cannot tell the difference between the stratified and
constant abundance model as our spectra do not go early enough,
except in the case of carbon. As it is shown in Fig. 13, carbon forms
too strong absorptions in the case of the constant abundance model
around the maximum light. This exceed in the uniform abundance
models would result from the relatively high 𝑋 (𝐶) ≈ 0.13 mass
fraction below 6000 km s−1 in the constant profile. This confirms
the assumption of Barna et al. (2018) that a significant amount of
unburned material in the ejecta of SNe Iax is likely found only in
the outermost layers.
As a result, our analysis seems to favour the implied a stratified
structure over the constant abundancemodel, as stratified abundance
profile is required to reproduce the main spectral features of car-
bon. In the case of other chemical elements, the differences between
the two models can be probed only with earlier spectral observa-
tions. Apart from carbon, the uniform abundance profiles of the
N1def model describe well the chemical structure of SN 2019muj
in general. Considering the possible explosion scenarios (see Sec.
1), which can explain some of the observables of SNe Iax, the
constrained abundance profiles of SN 2019muj may not contradict
strongly with the predictions of the pure deflagration scenario, but
definitely require some further tuning in the hydrodynamic models
at least in the case of unburned material.
The self-consistent TARDIS models also provide us an estima-
tion of the photospheric velocity based on the simultaneous fits of
several spectral lines and the flux continuum. The evolution of vphot
is monotonically decreasing, with a steeper descent before the mo-
ment of themaximum light. The derived behaviour of vphot (see Fig.
15) shows good agreement with the average velocity estimated from
the blueshift of the Si II 𝜆6355 absorption minima. Moreover, the
MNRAS 000, 1–23 (2020)
14 B. Barna et al.
Figure 15. The evolution of photospheric velocity of SN 2019muj (𝑀V =
−16.42 mag) and five relatively bright SNe Iax (Barna et al. 2018) with
peak absolute brightness covering a range from −17.3 (SN 2015H) to −18.4
mag (SN 2011ay). The velocities (illustrated with filled circles) were con-
strained via abundance tomography analysis performed with TARDIS. The
photospheric velocities of the extremely faint SN 2019gsc (𝑀V ∼ −13.8
mag) constrained via spectral fitting with TARDIS are also plotted (Srivastav
et al. 2020). The crosses (between 5 and 221 days) and the star (at 34.4 days)
show the photospheric velocity estimated from the shift of the absorption
minima of Si II 𝜆6355 and the NIR Co II lines, respectively.
vphot of the Co II feature in the NIR regime at +26.5 days also sup-
ports this velocity evolution. The velocity profile of SN 2019muj
fits into the trend of other SNe Iax, which showed a relation be-
tween the peak luminosity and the value of vphot at V-maximum
(McClelland et al. 2010; Foley et al. 2013). This result suggests that
this correlation is tight not only for the most luminous SNe Iax,
but probably for the whole class. However, to achieve more general
conclusions regarding to the luminosity-velocity relation, more SNe
Iax (including the reported outliers, e.g. SNe 2009ku and 2014ck,
Narayan et al. 2011; Tomasella et al. 2016) have to be studied with
the same technique.
The velocity estimation could be further improved by analyz-
ing the near-infrared (NIR) spectral features. The two spectra of SN
2019muj obtained in the NIR regime are shown in Fig. 14. There
are only a few SNe Iax with near-infrared (NIR) spectra: such obser-
vations have been presented in the case of SNe 2012Z (Stritzinger
et al. 2015), 2005hk (Kromer et al. 2013), 2008ge (Stritzinger et al.
2014), 2014ck (Tomasella et al. 2016), 2015H (Magee et al. 2016)
and 2010ae (Stritzinger et al. 2014). Although TARDIS modeling
is unreliable for fitting spectra at infrared wavelengths due to the
black body approximation, the synthetic spectra could be used for
identifying spectral features. Thus, we shift the flux of the TARDIS
synthetic spectrum created for the -1.7 day epoch to match with
the -2.4 day NIR spectrum. The shape of the continuum is of the
synthetic and observed spectra are surprisingly similar, despite that
real flux of the the synthetic spectrum is scaled with approximately
an order of magnitude. However, the synthetic spectrum shows very
weak spectral lines, which mostly overlap each other, making the
precise line identification unfeasible. Note that the epoch of the lat-
est optical spectrum fit with our TARDIS is 10.9 days after B-band
maximum, thus, the comparison for the second NIR spectrum is not
possible.
In Fig. 14,we also compare the twoNIR spectra of SN2019muj
Figure 16.The SED of SN 2019muj calculated fromUV/optical photometry
at three different epochs, compared to the spectra obtained at the closest
epochs (grey and black). The epochs indicate the time with respect to 𝐵-
band maximum (MJD 58 707.8). The flux values estimated from the Swift
UVW2 and UVW1 bands are represented by two green points on each panel.
to those of SNe 2005hk and 2010ae obtained at similar epochs. Al-
though, even the two comparative spectra are very similar at few
days before the maximum light (it was also shown by Stritzinger et
al. 2014), there are some differences between 10 000 and 12 000 Å
dominated by Mg II and Fe II lines (Tomasella et al. 2016), where
the absorption features are stronger, especially those at ∼11 300 Å,
which barely noticeable in the spectrum of SN 2019muj. These
properties make SN 2019muj resembling to the more luminous SN
2005hk at the earlier epoch. Similarly to the optical spectral evolu-
MNRAS 000, 1–23 (2020)
Type Iax SN 2019muj 15
Figure 17. The bolometric LC of SN 2019muj and the best-fit Arnett-model
describedwithEq. 3. The light curve peaks atMJD58707.2. For comparison,
the UVOIR LC of the N1def Fink et al. (green, 2014) and N5def_hybdrid
Kromer et al. (cyan, 2015) deflagration models are also plotted.
tion (Fig. 9), SN 2019muj at NIR wavelengths also turns to be more
similar to the fainter SN 2010ae, as their spectra obtained at few
weeks later (+26.5 and +18 days, respectively) are almost identical.
At these epochs, the most prominent NIR features are the set of
Co II lines, which appear with significantly lower relative strength
in the spectrum of SN 2005hk. The Co II features do not overlap
and allow us to measure the photospheric velocity with at post-
maximum epochs (Stritzinger et al. 2014; Tomasella et al. 2016).
Among the reported lines, Co II 𝜆15795, 𝜆16064 and 𝜆16361 can
be easily identified in the later NIR spectrum of SN 2019muj. Based
on the blueshifts of the absorption minima, the photospheric veloc-
ity is 𝑣phot = 2500 km s−1 at +26.5 days after 𝐵-band maximum
light. We compare this value with other methods to investigate the
evolution of the photosphere below.
3.3 Bolometric light curve and SED
We calculated F𝜆 flux densities from the Swift,𝑈𝐵𝑉 and 𝑔𝑟𝑖 dered-
dened magnitudes based on Bessel, Castelli & Plez (1998) and
Fukugita et al. (1996), respectively, and created SEDs for several
epochs. Note that Swift wide band magnitudes are not used for this
calculation, because of the potentially high uncertainties (see Sec.
2.2). For a given date, if SN 2019muj was not observed in some
of the filters, we use the corresponding value from the fitting of
a low-order polynomial function to the observed magnitudes. The
SEDs are directly compared to the spectra (corrected for reddening)
in Fig. 16. Integrating the SED functions to the total flux and scaling
it according to the assumed distance, a quasi-bolometric LC is es-
timated (Fig. 17). The derived luminosities are fit with the analytic
LC model introduced by Chatzopoulos, Wheeler & Vinkó (2012),
which is based on the radioactive decay diffusion model of Arnett
(1982) and Valenti et al. (2008). The function has the following
form:
𝐿(𝑡) = 𝑀Ni · exp
(
−𝑥2
)
·
[
1 − exp
(
−𝐴/𝑡2exp
)]
·
[
(𝜖Ni − 𝜖Co)
∫ 𝑥
0
2𝑧 · exp
(
𝑧2 − 2𝑧𝑦
)
𝑑𝑧
+ 𝜖Co
∫ 𝑥
0
2𝑧 · exp
(
𝑧2 − 2𝑧𝑦 + 2𝑧𝑠
)
𝑑𝑧
] (3)
where 𝑡exp is the time since the explosion, MNi is the radioactive
nickel mass produced in the explosion, while 𝜖Co and 𝜖Ni are the
energy generation rates by the decay of the radioactive isotopes of
cobalt and nickel. Gamma-ray leakage from the ejecta is considered
in the last term of the equation, where 𝐴 = (3𝜅𝛾𝑀ej) / (4𝜋𝑣2)
includes the gamma-ray opacity 𝜅𝛾 , the expansion velocity 𝑣 and
the mass of the ejecta 𝑀ej. The additional parameters are 𝑥 = 𝑡 / 𝑡d,
𝑦 = 𝑡d / (2𝑡Ni), 𝑧 = 1 / 2𝑡d, and 𝑠 = 𝑡d · (𝑡Co− 𝑡Ni)/(2𝑡Co𝑡Ni), where
td is the effective diffusion time.
The fit of the bolometric LC is plotted in Fig. 17, while the
best-fit parameters are listed in Table A7. The date of explosion is
found as MJD 58 698.2 ± 0.5. This is in good agreement with the
date estimated from the spectral fitting (MJD 58 697.5 ± 0.5) and
almost perfectly matches with the estimated value from the early
LCs (MJD 58 698.1). Accepting MJD 58 698.2 as opposed to the
one provided by the fireball model allows us to adjust the power-
law fit of the early LCs, yielding 𝑛 ≈ 1.26. The rise times of SNe
Iax are significantly shorter than those of normal SNe Ia, and this
trend is confirmed in SN 2019muj, with 𝑡rise = 9.6 ± 0.6 days in
the 𝐵-band (see Tab. A6). SN 2019muj also fits well to the trend of
SNe Iax showing a loose correlation between their rise times and
peak absolute magnitudes (see e.g. Tomasella et al. 2016; Magee et
al. 2016, for V- and r- band comparisons).
Based on the bolometric LC fit, the amount of 56Ni produced
during the explosion (MNi = 0.031±0.005M) is low, even within
the Iax class. It is approximately half of the estimated nickel mass
of SN 2015H (0.06 M , Magee et al. 2016), ∼ 30% of that of SN
2014ck (0.1 M , Tomasella et al. 2016) and only ∼ 14% of that of
the brightest type Iax SN 2011ay (0.225 M , Szalai et al. 2015).
At the same time, this nickel mass is an order of magnitude higher
than those of the extremely faint SNe 2008ha and 2019gsc (0.003
𝑀 , Foley et al. 2013; Srivastav et al. 2020), which confirms the
placement of SN 2019muj in the gap of the luminosity distribution
of SNe Iax. Nevertheless, the initial 56Ni mass of SN 2019muj
matches reltively well with the value of MNi = 0.0345 M of the
N1def pure deflagration model calculated by Fink et al. (2014) via
hydrodynamical simulations. The similarity toN1def occurs not just
on the peak brightness, but also on the post-maximum evolution of
the LC (Fig. 17). This match further supports the idea that despite
the possible discrepancies in the chemical abundances, SN 2019muj
is the product of the pure deflagration scenario.
Assuming a correlation between 𝑀ej and 𝑀Ni as in the model
grid of Fink et al. (2014), we may set the 𝑀ej = 0.0843 M of
the N1def model as an upper limit for SN 2019muj. Note however
that the synthetic LCs of these pure deflagration models decline too
rapidly compared to the SNe Iax LCs. The increased transparency
of the ejecta is probably the simple consequence of the insufficient
ejecta masses in the hydrodynamic simulations. Another method
was proposed by Foley et al. (2009) taking advantage that both the
rise time and the expansion velocity are proportional to the ejecta
mass (Arnett 1982), which leads to:
𝑀ej ∝ 𝑣exp × 𝑡2rise. (4)
Assuming that the mean opacity of the ejecta is uniform among
thermonuclear SNe, this simplification allows to estimate the ejecta
mass by comparing the bolometric rise time (𝑡rise = 9.0 days, see
in Fig. 17) and expansion velocity (here we adopt 𝑣exp = 5000 km
s−1) to those of normal SNe Ia. Following the formula presented by
Ganeshalingam et al. (2012), we estimate𝑀ej = 0.17M , while the
kinetic energy of the ejecta is 𝐸kin = 4.4×1049 erg. The discrepancy
between ejecta masses calculated by this method and that of N1def
deflagrationmodel assumed as upper limit for SN2019muj is similar
MNRAS 000, 1–23 (2020)
16 B. Barna et al.
to the mismatch estimated by Magee et al. (2016). While the ejecta
mass of SN 2019muj is comparable to the values estimated for
the less energetic Type Iax SN 2019gsc (𝑀ej ≈ 0.13 − 0.20 M;
Tomasella et al. 2020; Srivastav et al. 2020, respectively), the kinetic
energy is significantly higher than that of SN 2019gsc (𝐸kin ≈
1.0 − 1.2 × 1049 erg; Srivastav et al. 2020; Tomasella et al. 2020,
respectively).
4 CONCLUSIONS
We present the observations of SN 2019muj, a SN Iax that exploded
in the nearby galaxy VV 525. After discovery, the object was fol-
lowed by both Swift and ground-based optical telescopes, providing
LCs in the UV regime, as well as in 𝑈𝐵𝑉𝑅 and 𝑔𝑟𝑖 bands. The
spectral series was delivered by a wide collaboration of observato-
ries starting at 0.5 days after the discovery (-5 days with respect to
𝐵-band maximum) and covering a period of 60 days. The estimated
peak luminosity places SN 2019muj between the relatively bright
and the extremely faint objects in the diverse Iax class of thermonu-
clear explosions. Such a transitional SN Iax has not previously been
a subject of detailed follow-up and analysis.
The rise times constrained from fitting a power-law function
of 𝑛 = 1.3 (similarly to the case of SN 2015H, Magee et al.
2016) fit well to the correlations with the peak absolute magnitudes
(Tomasella et al. 2016; Magee et al. 2016). However, the estimated
decline rates deviate from the previously reported loose correlation
and show more similarity with those of extremely faint objects like
SNe 2008ha and 2010ae. The intrinsic B-V colour evolution of SN
2019muj is almost identical to that of SN 2005hk. The blue contin-
uum rapidly changes after B-maximum and shows a nearly constant
B-V colour in the next weeks. Compared to SN 2008ha, the intrinsic
colour of SN 2019muj is significantly bluer in the first three weeks
after the explosion.
The transitional nature of SN 2019muj can also be observed in
the evolution of the optical spectra. According to the comparisons
of the spectra obtained at the same epochs, SN 2019muj showsmore
similarity with the more luminous SNe 2005hk and 2011ay a few
days before the date of B-maximum. At +12 days, both the lines of
IMEs and IGEs match well with that of SN 2010ae.
Based on the comparisons of the observables, SN 2019muj
shows a better match with the more luminous SNe Iax before the
moment of maximum light, and with the extremely faint ones at the
post-maximum epochs. This kind of change in the evolution of SN
2019muj might be the impact of its steep density profile. In order
to examine, whether this behavior is regular among the moderately
luminous SNe Iax, the study of more objects with approximately
the same luminosity is required, which is out of the scope of this
paper.
The spectrawere also analyzed via a simplified version of abun-
dance tomography, where a stratified abundance template (Barna et
al. 2018) is fit by shifting it in the velocity space. Physical pa-
rameters, like luminosity and photospheric velocity, as well as the
exponential function of the density profile, have also been free pa-
rameters in the fitting process. The density structure of the model
shows a slope of 𝑣0 = 1800 km s−1 with a steep cut-off above 5500
km s−1. The inferred feasible synthetic spectra were compared to
spectra calculatedwith constant abundances.We found that the strat-
ified abundance profile does not improve significantly the goodness
of the fit in the case of most of the elements. However, without
earlier spectra we are unable to accurately test abundances and den-
sities in the outermost layers, where the deflagration and stratified
models diverse. In the regions probed in this work the stratified and
deflagration models are very similar, with the exception of carbon,
which element is not allowed below 6500 km s−1 in the stratified
model. Our results demonstrate the need for carbon stratification,
which is not inline with the state-of-the-art deflagration models.
We constructed a quasi-bolometric LC of SN 2019muj based
on the UV and optical photometry. The estimated luminosities were
fit with an analytic model, which allowed us to constrain the mass
of radioactive nickel produced in the explosion as 𝑀Ni = 0.031M .
This value is slightly lower than the𝑀Ni predicted by theN1def pure
deflagration model reported by Fink et al. (2014). However, even
weaker explosionswere simulated assuming pure deflagration under
special circumstances (Kromer et al. 2015), thus, the bolometric
LC of SN 2019muj does not contradict with deflagration scenario
leaving behind a bound remnant. The ejecta mass was calculated
assuming a constant opacity, as in the case of normal SNe Ia, with
a value of 𝑀ej = 0.17 M . The analysis also delivered another
constraint on the date of explosion as MJD 58 698.2 ± 0.5. This
shows a good agreement with the values obtained from early LC
fitting (MJD 58 698.1) and abundance tomography (MJD 58 697.5±
0.5).
SN 2019muj is not the first moderately luminous SN Iax, as
SNe 2015H and 2014ck from the brighter side have also started to
bridge the luminosity gap in the Iax class. However, the even lower
peak luminosity, the detailed pre-maximum observations, the long-
time follow-up, make SN 2019muj a good candidate to represent
the intermediate population of the class. Moreover, all the esti-
mated physical and chemical properties suggest that SN 2019muj is
a transitional object in the SN Iax class. Based on the preliminary
comparisons with more luminous and extremely faint objects, it is
very likely that certain physical properties have a continuous dis-
tribution among SNe Iax. This could be the source of several tight
correlations, e.g. the previously proposed connection between the
peak luminosity and expansion velocity. Nevertheless, this conclu-
sion does not contradict with the idea that all SNe Iax share the
similar explosion scenario, the pure deflagration of a WD leaving
behind a bound remnant.
The verification of these assumptions require the re-
investigation of the previously reported outliers (e.g. SN 2014ck,
Tomasella et al. 2016) to understand the origin of the discrepancies.
Moreover, further SNe Iax with moderate peak luminosities have to
be studied to prove that these objects indeed bridge the gap between
the more luminous and the extremely faint SNe Iax.
ACKNOWLEDGEMENTS
We are grateful for M. Stritzinger and S. Valenti for providing
the NIR spectra of SN 2005hk. Based on observations collected
at the European Organisation for Astronomical Research in the
Southern Hemisphere, Chile as part of PESSTO, (the Public ESO
Spectroscopic Survey for Transient Objects Survey) ESO program
ID 1103.D-0328. This work makes use of observations from the
LCO network. This work has made use of data from the Asteroid
Terrestrial-impact Last Alert System (ATLAS) project. ATLAS is
primarily funded to search for near earth asteroids through NASA
grants NN12AR55G, 80NSSC18K0284, and 80NSSC18K1575; by
products of the NEO search include images and catalogs from the
survey area. The ATLAS science products have been made possible
through the contributions of the University of Hawaii Institute for
Astronomy, theQueen’sUniversity Belfast, and the Space Telescope
Science Institute.
MNRAS 000, 1–23 (2020)
Type Iax SN 2019muj 17
BB was funded by the Czech Science Foundation Grant no.
19-15480S and the institutional project RVO:67985815. This work
is part of the project Transient Astrophysical Objects GINOP-2-3-2-
15-2016-00033 of the National Research, Development and Inno-
vation Office (NKFIH), Hungary, funded by the European Union.
Research on type-Iax supernovae at Rutgers University (SWJ, YC-
N, LK) is supported by NSF award AST-1615455. L.G. was funded
by the European Union’s Horizon 2020 research and innovation
programme under the Marie Skłodowska-Curie grant agreement
No. 839090. DAH, JB, DH, and CP are supported by NSF AST-
1911225 and NASA grant 80NSSC19K1639. Research by DJS
is supported by NSF grants AST-1821967, 1821987, 1813708,
1813466, and 1908972, and by the Heising Simons Foundation
under grant #2020-1864. MN is supported by a Royal Astronomical
Society Research Fellowship. TWC acknowledges the EU Funding
under Marie Skłodowska-Curie grant agreement No 842471. KM
acknowledges support from ERC Starting Grant grant no. 758638.
STFC 2018 - 2020 : SJS and SS acknowledges funding from STFC
Grant Ref: ST/P000312/1. This work has been partially supported
by the Spanish grant PGC2018-095317-B-C21 within the Euro-
pean Funds for Regional Development (FEDER). TMB was funded
by the CONICYT PFCHA / DOCTORADOBECAS CHILE/2017-
72180113. MG is supported by the Polish NCN MAESTRO grant
2014/14/A/ST9/00121. JDL acknowledges support from STFCvia
grant ST/P000495/1. M.R.S. is supported by the National Science
Foundation Graduate Research Fellowship Program Under Grant
No. 1842400. D.A.C. acknowledges support from the National
Science Foundation Graduate Research Fellowship under Grant
DGE1339067.
The UCSC team is supported in part by NSF grant AST-
1518052,NASA/Swift grant 80NSSC19K1386, theGordon&Betty
Moore Foundation, the Heising-Simons Foundation, and by a fel-
lowship from the David and Lucile Packard Foundation to R.J.F.
This work includes data obtained with the Swope Telescope at
Las Campanas Observatory, Chile, as part of the Swope Time Do-
main Key Project (PI: Piro, Co-Is: Drout, Phillips, Holoien, French,
Cowperthwaite, Burns, Madore, Foley, Kilpatrick, Rojas-Bravo,
Dimitriadis, Hsiao). We wish to thank Swope Telescope observers
Jorge Anais Vilchez, Abdo Campillay, Yilin Kong Riveros, Nahir
Munoz Elgueta and Natalie Ulloa for collecting data presented in
this paper.
We are grateful to the SALT Astronomers and support staff
who obtained the SALT data presented here.
Based on observations obtained at the international Gem-
ini Observatory, under program GS-2019B-Q-236, which is man-
aged by the Association of Universities for Research in Astron-
omy (AURA) under a cooperative agreement with the National
Science Foundation. On behalf of the Gemini Observatory part-
nership: the National Science Foundation (United States), National
Research Council (Canada), Agencia Nacional de Investigación y
Desarrollo (Chile), Ministerio de Ciencia, Tecnología e Innovación
(Argentina), Ministério da Ciência, Tecnologia, Inovações e Comu-
nicações (Brazil), andKoreaAstronomy and Space Science Institute
(Republic of Korea).
Research at Lick Observatory is partially supported by a gen-
erous gift from Google.
This research made use of Tardis, a community-developed
software package for spectral synthesis in supernovae (Kerzendorf
& Sim 2014; Kerzendorf et al. 2019). The development of Tardis
received support from the Google Summer of Code initiative and
from ESA’s Summer of Code in Space program. Tardis makes
extensive use of Astropy and PyNE.
DATA AVAILABILITY
The data underlying this article will be available on WISeREP, at
https://wiserep.weizmann.ac.il/object/12935.
REFERENCES
Arnett W. D., 1982, ApJ, 253, 785
Barna B., Szalai T., Kromer M., Kerzendorf W., Vinkó J., Silverman J. M.,
Marion G. H., Wheelere J. C., 2017, MNRAS, 471, 4865
Barna B., Szalai T., Kerzendorf W. E., Kromer M., Sim S. A., Magee M. R.,
Leibundgut B., 2018, MNRAS, 480, 3609
Bessell M. S., Castelli F., Plez B., 1998, A&A, 333, 231
Blanton M. R., Roweis S., 2007, AJ, 133, 734
Branch D. et al., 2006, PASP, 118, 560
Brimacombe J. et al., 2019, ATel #13004
Bravo E., García-Senz D., 2006, ApJ, 642, L157
Bravo E., Garcí-Senz D., Cabezón R. M., Domínguez I., 2009, ApJ, 695,
1257
Bravo E., Gil-Pons P., Gutiérrez J. L., Doherty C. L., 2016, A&A, 589, A38
Brown T. M. et al., 2013, PASP, 125, 1031
Brown P. J. et al., 2010, ApJ, 721, 1608
Brown P. J., Breeveld A. A., Holland S., Kuin P., Pritchard T., 2014, Ap&SS,
354, 89
Brown P. J., Breeveld A., Roming P. W. A., Siegel M., 2016, AJ, 152, 102
Burrows D. N. et al., 2005, Space Sci. Rev., 120, 165
Buzzoni B. et al., 1984, Msngr, 38, 9
Chatzopoulos E., Wheeler J. C., Vinkó J., 2012, ApJ, 746, 121
Childress M. et al., 2013, ApJ, 770, 29
Crawford S. M. et al, 2010, Proc. SPIE Conf. Ser., 7737, 25
Denissenkov P. A., Truran J. W., Herwig F., Jones S., Paxton B., Nomoto
K., Suzuki T., Toki H., 2015, MNRAS, 447, 2696
de Vaucouleurs G., de Vaucouleurs A., Corwin H. G. Jr., Buta R. J., Paturel
G., Fouque P., 1991, Third Reference Catalogue of Bright Galaxies,
Springer, New York, 2091
Eikenberry S. et al., 2008, SPIE conf. ser., 7014, 70140V
Flewelling H. A. et al., 2016, arXiv e-prints, arXiv:1612.05243
Filippenko A. V., 1982, PASP, 94, 715
Fink M. et al., 2014, MNRAS, 438, 1762
Firth, R. E. et al., 2015, MNRAS, 446, 3895
Foley R. J. et al., 2009, AJ, 138, 376
Foley R. J., Brown P. J., Rest A., Challis P. J., Kirshner R. P.; Wood-Vasey
W. Michael, 2010, ApJ, 708, 61
Foley R. J. et al., 2010, AJ, 140, 1321
Foley R. J. et al., 2013, ApJ, 767, 57
Foley R. J. et al., 2014, ApJ, 792, 29
Foley R. J., Jha S. W., Pan Y.-C., Zhen, W. K., Bildsten L., Filippenko A.
V., Kasen D., 2016, ApJ, 792, 29
Fitzpatrick E. L., Massa D., 2007, ApJ, 663, 320
Fukugita M., Ichikawa T., Gunn J. E., Doi M., Shimasaku K., Schneider D.
P. et al., 1996, AJ, 111, 1748
Graur O., Bianco F. B., Modjaz M., Shivvers I., Filippenko A. V., Li W.,
Smith N., 2017, ApJ, 837, 121
Ganeshalingam M., Li W., Filippenko A. V., 2011, MNRAS, 416, 2607
Ganeshalingam M. et al., 2012, ApJ, 751, 142
HamuyMario, Phillips M.M., Wells Lisa A., Maza Jose, 1993, Publications
of the Astronomical Society of the Pacific, 105, 787
Hamuy M., et al. 2006, Publications of the Astronomical Society of the
Pacific, 118, 2
Hiramatsu D., Arcavi I., Burke J., Howell D. A., McCully C., Pellegrino
C., Valenti, S., 2019, Transient Name Server Classification Report, No.
2019-1442
Howell D. A. et al., 2005, ApJ, 634, 1190
Höflich P., Khokhlov A. M., Wheeler J. C., 1995, Apj, 444, 831
Höflich P. et al., 2017, Apj, 846, 58
Ivanova L. N., Imshennik V. S., Chechetkin V. M., 1974, Ap&SS, 31, 497
Jacobson-Galán W. V. et al., 2019, MNRAS, 487, 2538
MNRAS 000, 1–23 (2020)
18 B. Barna et al.
Jester S., et al. 2005, AJ, 130, 873
Jha S. W. et al., 2006, AJ, 132, 189
Jha S.W. 2017, Handbook of Supernovae, Springer International Publishing
AG, 375
Jordan G. C., IV et al., 2008, ApJ, 681, 1448
Jordan G. C., IV et al., 2012a, ApJ, 759, 53
Jordan G. C., IV, Perets H. B., Fisher R. T., van Rossum D. R., 2012b, ApJ,
761, L23
Kasen D., 2006, ApJ, 649, 939
Kerzendorf W. E., Sim S. A., 2014, MNRAS, 440, 387
Kerzendorf W. E. et al., 2018, tardis-sn/tardis: TARDIS v2.0.2 release,
Zenodo Software Release
Kilpatrick C. D et al., 1998, MNRAS, 473, 4805
Khokhlov A. M., 1991a, A&A, 245, 114
Khokhlov A. M., 1991b, A&A, 246, 383
Krisciunas K. et al., 2017, AJ, 154, 211
Kromer M. et al., 2015, MNRAS, 450, 3045
Kromer M. et al., 2015, MNRAS, 450, 3045
Li W. et al., 2003, PASP, 115, 453
Li W. et al., 2011, MNRAS, 412, 1441
Lyman J. D. et al., 2018, MNRAS, 473, 1359
Magee M. R. et al., 2016, A&A, 589, 89
Magee M. R., Sim S. A., Kotak R., Maguire K., Boyle A., 2019, A&A, 622,
102
McClelland C. M., et al., 2010, ApJ, 720, 704
McCully C., et al., 2014, Natur, 512, 54
Miller J. S., Stone R. P. S., 1993, Lick Obs. Tech. Rep. 66., Lick Obs., Santa
Cruz
Miller A. A. et al., 2017, ApJ, 848, 59
Narayan G. et al., 2011, ApJ, 731, L11
Nugent P. E., Sullivan M., Cenko S. B., et al. 2011, Nature, 480, 344
Phillips M. M., 1993, ApJ, 413, 105
Phillips M. M. et al., 2007, PASP, 119, 360
Pinto P. A., Eastman R. G., 2000, ApJ, 530, 757
Plewa T., Calder A. C., Lamb D. Q., 2004, ApJ, 612, L37
Rest A. et al., 2005, ApJ, 634, 1103
Riess A. G., Filippenko A. V., Li W., Schmidt B. P. 1999, AJ, 118, 2668
Roming P. W. A., et al. 2005, Space Sci. Rev., 120, 95
Sand D. J. et al, 2018, ApJ, 863, 24
Schechter P. L., Mateo M. & Saha A., 1993, PASP, 105, 1342
Schlafly E. F., Finkbeiner D. P. 2011, ApJ, 737, 103
Scolnic D. et al., 2015, ApJ, 815, 117
Shappee B. J. et al, 2014, ApJ, 788, 48
Silverman J. M. et al., 2015, MNRAS, 451, 1973
Silverman J. M., Kong J. J., Filippenko A. V., 2020, MNRAS, 425, 1819
Smartt S. J. et al., 2015, A&A, 579, A40
Smith M. P. et al, 2006, Proc. SPIE Conf. Ser., 6269, 2
Smith K. W. et al., 2020, arXiv e-prints, arXiv:2003.09052
Srivastav S. et al., 2020, ApJ, 892, 24
Stritzinger M. D. et al., 2014, A&A, 561, 146
Stritzinger M. D. et al., 2015, A&A, 573, 2
Stehle M., Mazzali P. A., Benetti S., Hillebrandt W., 2005, MNRAS, 360,
1231
Swift J., Vyhnal Ch. 2018, RTSREP, 1, 281
Szalai T. et al., 2015, MNRAS, 453, 2103
Takaro T. et al., 2020, MNRAS, 493, 986
Taubenberger, S. 2017, Handbook of Supernovae, Springer International
Publishing AG, 317
Tonry J. L. et al., 2012, ApJ, 750, 99
Tonry J. L. et al., 2018, PASP, 130, 064505
Tomasella L. et al., 2016, MNRAS, 459, 1018
Tomasella L. et al., 2020, arXiv e-prints, arXiv:2002.00393
Tully R. B. et al., 2013, AJ, 146, 86
Tully R. B., Courtois H. M., Sorce J. G., 2016, AJ, 152, 50
Vacca W. D., Cushing M. C., Rayner J. T., 2003, PASP, 115, 389
Valenti S. et al., 2008, MNRAS, 383, 1485
Valenti S. et al., 2009, Nature, 459, 674
Valenti S. et al., 2014, MNRAS, 438, 101
Valenti S. et al., 2016, MNRAS, 459, 3939
Vogl C., Sim S. A., Noebauer U. M., Kerzendorf W. E., Hillebrandt W.,
2016, MNRAS, 459, 3939
Yamanaka M. et al, 2015, ApJ, 806, 191
White Ch. et al., 2015, ApJ, 799, 52
MNRAS 000, 1–23 (2020)
Type Iax SN 2019muj 19
APPENDIX A: OBSERVATIONAL DATA
Table A1: Log of ATLAS photometry obtained in cyan- and orange-filters. The
fluxes are weighted means of the individual 30 second exposures on each night,
and the errors are the standard deviations of those fluxes. All mags are in AB
system. Before MJD 58 700, we adopt Flux [mag] = -2.5 log10 (3 × Flux_error)
+ 23.9 as 3𝜎 upper limit.
MJD cyan-filter orange-filter
Flux [𝜇Jy] Flux [mag] Flux [𝜇Jy] Flux [mag]
58686.54 - - 13.92 (101.61) > 17.7
58688.60 - - 6.89 (19.47) > 19.5
58690.59 - - -0.54 (16.26) > 19.7
58692.56 - - -1.51 (15.38) > 19.7
58694.51 1.68 (4.51) > 21.0 - -
58696.58 - - 1.87 (19.79) > 19.5
58700.52 - - 22.52 (12.56) 20.52 (0.60)
58702.53 392.09 (6.36) 17.42 (0.02) - -
58704.53 - - 589.18 (14.71) 16.97 (0.03)
58706.54 937.08 (7.96) 16.47 (0.01) - -
58708.47 - - 820.62 (38.08) 16.61 (0.05)
58712.52 - - 912.76 (72.30) 16.50 (0.09)
Table A2: Log of Swift photometry.
MJD UVW2 UVM2 UVW1 U B V
58702.85 17.78 (0.11) 17.48 (0.08) 16.77 (0.09) 16.21 (0.08) 17.26 (0.08) 17.19 (0.13)
58703.85 17.49 (0.11) 17.58 (0.08) 16.63 (0.08) 15.89 (0.07) 16.92 (0.08) 16.86 (0.11)
58704.72 17.62 (0.10) 17.71 (0.09) 16.70 (0.08) 15.76 (0.07) 16.75 (0.07) 16.78 (0.10)
58705.62 17.92 (0.11) 18.04 (0.10) 16.79 (0.08) 15.87 (0.07) 16.70 (0.07) 16.53 (0.09)
58706.83 17.89 (0.11) 18.13 (0.11) 17.00 (0.09) 15.77 (0.07) 16.51 (0.07) 16.49 (0.09)
58710.69 18.79 (0.16) 19.45 (0.24) 17.80 (0.12) 16.51 (0.08) 16.67 (0.07) 16.27 (0.09)
58712.94 19.77 (0.28) 20.20 (0.34) 18.36 (0.16) 17.17 (0.10) 16.98 (0.08) 16.58 (0.10)
58717.80 - - - 18.73 (0.24) 18.07 (0.12) 17.04 (0.13)
58720.12 - - - - 18.54 (0.16) 17.10 (0.14)
58724.58 - - - - 18.85 (0.16) 17.50 (0.20)
58726.10 - - - - 18.69 (0.24) 17.77 (0.28)
58729.72 - - - - 19.19 (0.21) 17.72 (0.17)
58730.31 - - - 19.12 (0.25) 19.18 (0.19) 17.97 (0.18)
58735.66 - - - - 19.34 (0.32) 17.99 (0.28)
MNRAS 000, 1–23 (2020)
20 B. Barna et al.
Table A3: Log of ground-based photometry obtained in UBV filters.
MJD U B V Telescope
58702.70 16.56 (0.04) 17.32 (0.02) 17.42 (0.03) LCO
58702.70 16.51 (0.03) 17.34 (0.02) 17.40 (0.02) LCO
58703.40 - 17.17 (0.01) 17.08 (0.01) Swope
58703.40 16.45 (0.04) 17.11 (0.02) 17.16 (0.02) LCO
58703.40 16.37 (0.04) 17.12 (0.01) 17.13 (0.02) LCO
58703.80 16.19 (0.03) - - LCO
58704.38 - 16.85 (0.01) 16.76 (0.01) Swope
58704.40 16.08 (0.04) 16.79 (0.02) 16.84 (0.02) LCO
58704.40 16.12 (0.05) 16.79 (0.02) 16.82 (0.02) LCO
58704.50 - - 16.81 (0.01) Thacher
58705.10 16.02 (0.02) - - LCO
58705.10 16.01 (0.02) - - LCO
58705.50 - - 16.61 (0.02) Thacher
58706.00 15.88 (0.03) 16.49 (0.02) 16.52 (0.02) LCO
58706.00 15.87 (0.02) 16.49 (0.02) 16.49 (0.02) LCO
58706.40 15.88 (0.04) 16.52 (0.01) 16.50 (0.01) LCO
58706.40 - 16.49 (0.01) 16.50 (0.02) LCO
58706.50 - - 16.49 (0.03) Thacher
58707.30 15.90 (0.05) 16.46 (0.02) 16.41 (0.02) LCO
58707.30 15.90 (0.05) 16.48 (0.02) 16.43 (0.01) LCO
58707.70 15.93 (0.02) 16.44 (0.01) 16.38 (0.01) LCO
58707.70 15.93 (0.03) 16.44 (0.01) 16.40 (0.01) LCO
58708.20 16.07 (0.02) - - LCO
58709.10 16.01 (0.03) 16.44 (0.02) 16.34 (0.02) LCO
58709.10 16.06 (0.03) 16.45 (0.02) 16.36 (0.02) LCO
58710.30 16.35 (0.08) 16.63 (0.03) 16.37 (0.03) LCO
58710.30 16.17 (0.10) 16.60 (0.03) 16.37 (0.02) LCO
58710.40 16.18 (0.03) 16.50 (0.01) 16.35 (0.02) LCO
58710.40 16.27 (0.02) 16.51 (0.02) 16.36 (0.02) LCO
58712.41 - 16.96 (0.01) 16.42 (0.01) Swope
58713.00 16.99 (0.05) 16.95 (0.03) 16.53 (0.03) LCO
58713.00 17.03 (0.06) 16.97 (0.03) 16.49 (0.02) LCO
58714.10 17.20 (0.10) 17.31 (0.02) 16.73 (0.02) LCO
58714.10 17.33 (0.03) 17.34 (0.02) 16.73 (0.02) LCO
58714.10 17.28 (0.03) - - LCO
58715.29 - 17.58 (0.01) 16.71 (0.01) Swope
58715.30 17.59 (0.04) 17.53 (0.02) 16.70 (0.02) LCO
58715.30 17.53 (0.03) 17.55 (0.02) 16.71 (0.01) LCO
58717.30 - 17.96 (0.02) 16.90 (0.02) Swope
58717.40 - 17.97 (0.02) 16.81 (0.02) LCO
58717.40 18.46 (0.05) 17.93 (0.03) 16.91 (0.02) LCO
58717.51 - - 16.87 (0.01) Thacher
58718.30 18.33 (0.04) 18.11 (0.02) 17.04 (0.02) LCO
58718.30 18.29 (0.03) 18.13 (0.02) 17.01 (0.02) LCO
58719.39 - 18.31 (0.01) 17.12 (0.01) Swope
58720.43 - 18.54 (0.05) 17.21 (0.02) Swope
58724.30 19.14 (0.04) 18.88 (0.02) 17.59 (0.02) LCO
58724.30 19.16 (0.04) 18.87 (0.02) 17.56 (0.01) LCO
58725.39 - 18.92 (0.01) 17.62 (0.01) Swope
58726.38 - 18.99 (0.02) 17.66 (0.01) Swope
58727.33 - 19.03 (0.01) 17.74 (0.01) Swope
58727.50 - - 17.72 (0.01) Thacher
58728.10 19.69 (0.10) 19.06 (0.03) 17.85 (0.02) LCO
58728.10 19.48 (0.08) 19.16 (0.03) 17.83 (0.02) LCO
58729.33 - 19.16 (0.01) 17.82 (0.01) Swope
58732.40 19.58 (0.17) - - LCO
58732.40 19.61 (0.16) 19.30 (0.03) 17.98 (0.02) LCO
Continued on next page
MNRAS 000, 1–23 (2020)
Type Iax SN 2019muj 21
Table A3 – continued from previous page
MJD U B V Telescope
58732.50 - 19.33 (0.03) 17.96 (0.02) LCO
58736.20 20.45 (0.29) 19.55 (0.05) - LCO
58736.20 20.47 (0.31) 19.51 (0.05) - LCO
58736.34 - 19.38 (0.02) 18.05 (0.02) Swope
58736.60 19.89 (0.23) 19.35 (0.06) 18.16 (0.03) LCO
58736.60 19.48 (0.23) 19.38 (0.05) 18.20 (0.03) LCO
58737.40 - 19.42 (0.02) 18.14 (0.01) Swope
58740.20 - 19.24 (0.20) 18.11 (0.12) LCO
58740.20 - 19.76 (0.34) 17.98 (0.10) LCO
58744.30 - 19.66 (0.10) 18.23 (0.04) LCO
58744.30 - 19.49 (0.08) - LCO
58745.33 - 19.56 (0.04) 18.42 (0.03) Swope
58746.34 - 19.62 (0.02) 18.31 (0.02) Swope
58748.40 - 19.69 (0.05) 18.46 (0.02) Swope
58752.10 - 19.67 (0.03) 18.65 (0.03) LCO
58752.10 - 19.70 (0.03) 18.60 (0.03) LCO
58752.33 - 19.74 (0.01) 18.49 (0.01) Swope
58753.15 - 19.78 (0.02) 18.62 (0.02) Swope
58754.27 - 19.80 (0.02) 18.59 (0.01) Swope
58755.27 - 19.73 (0.01) 18.58 (0.01) Swope
58758.40 - 19.80 (0.05) 18.74 (0.03) LCO
58758.40 - 19.86 (0.05) 18.64 (0.03) LCO
58759.28 - 19.80 (0.02) 18.66 (0.02) Swope
Table A4: Log of ground-based photometry obtained in ugriz filters.
MJD u g r i z Telescope
58702.40 - 17.40 (0.00) - - - Cassius
58702.70 - 17.21 (0.01) 17.47 (0.02) 17.71 (0.05) - LCO
58702.70 - 17.20 (0.01) 17.47 (0.02) 17.75 (0.05) - LCO
58703.39 16.88 (0.01) 16.96 (0.01) 17.13 (0.01) 17.42 (0.01) - Swope
58703.40 - - 17.23 (0.01) 17.45 (0.02) - LCO
58703.40 - - - 17.45 (0.02) - LCO
58704.37 16.70 (0.01) 16.69 (0.01) 16.84 (0.01) - - Swope
58704.40 - 16.64 (0.01) 16.82 (0.01) - - LCO
58704.40 - 16.62 (0.01) 16.82 (0.01) - - LCO
58704.48 - 16.72 (0.01) 16.87 (0.02) 17.11 (0.04) - Thacher
58705.49 - 16.53 (0.01) 16.72 (0.01) 16.93 (0.04) 17.06 (0.02) Thacher
58706.10 - 16.53 (0.01) - 16.83 (0.01) - LCO
58706.10 - 16.55 (0.01) 16.76 (0.01) 17.03 (0.01) - LCO
58706.40 - 16.36 (0.01) 16.56 (0.01) 16.81 (0.01) - LCO
58706.40 - 16.36 (0.01) 16.54 (0.01) 16.79 (0.01) - LCO
58706.48 - 16.42 (0.01) 16.57 (0.01) 16.82 (0.02) 16.98 (0.02) Thacher
58707.30 - 16.36 (0.01) 16.42 (0.01) 16.71 (0.01) - LCO
58707.30 - 16.32 (0.01) 16.44 (0.01) 16.73 (0.01) - LCO
58707.70 - 16.30 (0.01) 16.45 (0.01) - - LCO
58707.70 - 16.31 (0.01) 16.45 (0.01) 16.73 (0.01) - LCO
58710.30 - 16.40 (0.01) - - - LCO
58710.30 - 16.34 (0.02) - - - LCO
58710.40 - 16.30 (0.01) 16.44 (0.01) 16.70 (0.02) - LCO
58710.40 - 16.31 (0.01) 16.42 (0.01) - - LCO
58710.48 - 16.35 (0.02) - - - Thacher
58712.40 - 16.60 (0.01) 16.43 (0.01) 16.68 (0.01) - Swope
58713.00 - 16.77 (0.02) 16.65 (0.02) 16.68 (0.02) - LCO
58713.00 - 16.75 (0.02) 16.59 (0.02) 16.84 (0.02) - LCO
58714.10 - 16.95 (0.01) - - - LCO
Continued on next page
MNRAS 000, 1–23 (2020)
22 B. Barna et al.
Table A4 – continued from previous page
MJD u g r i z Telescope
58714.10 - 16.94 (0.01) 16.64 (0.01) - - LCO
58714.20 - - 16.67 (0.01) 16.72 (0.02) - LCO
58714.20 - - - 16.70 (0.01) - LCO
58715.28 18.34 (0.03) 17.11 (0.01) 16.59 (0.01) 16.74 (0.01) - Swope
58715.30 - 17.08 (0.01) 16.55 (0.01) 16.68 (0.01) - LCO
58715.30 - 17.09 (0.01) 16.56 (0.01) 16.68 (0.01) - LCO
58717.29 18.91 (0.05) 17.45 (0.02) 16.70 (0.01) 16.74 (0.01) - Swope
58717.40 - 17.48 (0.01) 16.72 (0.01) 16.78 (0.01) - LCO
58717.40 - 17.49 (0.01) 16.71 (0.01) 16.77 (0.01) - LCO
58717.49 - 17.40 (0.02) 16.72 (0.02) 16.82 (0.03) 16.89 (0.05) Thacher
58718.40 - 17.69 (0.01) 16.74 (0.01) 16.79 (0.01) - LCO
58718.40 - 17.70 (0.01) 16.75 (0.01) 16.79 (0.01) - LCO
58719.38 19.32 (0.04) 17.69 (0.01) 16.77 (0.01) 16.79 (0.01) - Swope
58720.43 - 17.82 (0.02) 16.88 (0.01) 16.85 (0.01) - Swope
58724.40 - 18.43 (0.01) 17.22 (0.01) 17.18 (0.01) - LCO
58724.40 - 18.44 (0.01) 17.20 (0.01) 17.14 (0.01) - LCO
58725.38 20.11 (0.04) 18.35 (0.02) 17.20 (0.01) 17.20 (0.01) - Swope
58726.36 20.17 (0.04) 18.37 (0.01) 17.31 (0.01) 17.21 (0.01) - Swope
58727.32 20.18 (0.04) 18.44 (0.01) 17.34 (0.01) 17.30 (0.01) - Swope
58727.50 - 18.44 (0.03) 17.45 (0.02) 17.36 (0.02) 17.40 (0.03) Thacher
58728.10 - 18.66 (0.02) - - - LCO
58728.10 - 18.67 (0.02) 17.50 (0.01) - - LCO
58728.20 - - 17.53 (0.01) 17.49 (0.02) - LCO
58728.20 - - - 17.44 (0.02) - LCO
58729.32 20.32 (0.04) 18.50 (0.01) 17.43 (0.01) 17.40 (0.01) - Swope
58732.50 - 18.76 (0.02) 17.68 (0.02) - - LCO
58732.50 - 18.76 (0.02) 17.68 (0.02) 17.60 (0.02) - LCO
58733.38 - 18.78 (0.02) 17.78 (0.01) 17.70 (0.01) - Swope
58736.34 - 18.78 (0.02) 17.75 (0.01) 17.68 (0.01) - Swope
58736.60 - 18.92 (0.03) 17.89 (0.02) 17.60 (0.02) - LCO
58736.60 - 18.88 (0.03) 17.89 (0.02) 17.83 (0.02) - LCO
58737.40 - 18.81 (0.01) 17.81 (0.01) 17.78 (0.01) - Swope
58740.20 - 18.76 (0.11) 17.92 (0.05) 17.84 (0.06) - LCO
58740.20 - 19.02 (0.11) 18.05 (0.06) 17.91 (0.06) - LCO
58744.30 - 18.99 (0.04) - - - LCO
58745.32 - 18.99 (0.03) 18.09 (0.02) 18.06 (0.02) - Swope
58746.33 - 19.02 (0.03) 18.16 (0.01) 18.13 (0.01) - Swope
58748.39 - 19.08 (0.01) 18.18 (0.01) 18.14 (0.01) - Swope
58752.10 - 19.15 (0.02) 18.29 (0.02) 18.24 (0.03) - LCO
58752.10 - 19.16 (0.01) 18.29 (0.02) 18.19 (0.03) - LCO
58752.32 21.20 (0.07) 19.14 (0.01) 18.28 (0.01) 18.19 (0.01) - Swope
58753.13 - 19.18 (0.01) 13.39 (0.01) 18.30 (0.01) - Swope
58754.29 21.20 (0.08) 19.17 (0.01) 18.31 (0.01) 18.24 (0.01) - Swope
58755.25 21.09 (0.06) 19.09 (0.01) 18.28 (0.01) 18.11 (0.01) - Swope
MNRAS 000, 1–23 (2020)
Type Iax SN 2019muj 23
Table A5: Log of the spectra; 𝑡exp shows the time since the date of explosion
constrained in the abundance tomography (MJD 58 697.5); while the phases are
given relative to the maximum in B-band (MJD 58 707.8).
MJD texp [d] Phase [d] Telescope/Instrument Wavelength range [Å]
58702.8 5.3 -5.0 LCO/FLOYDS 3800 - 10000
58703.6 6.1 -4.2 LCO/FLOYDS 3500 - 10000
58704.3 6.8 -3.5 NTT/EFOSC2 3300 - 9900
58705.1 7.6 -2.7 SALT/RSS 3500 - 9300
58705.4 7.9 -3.4 Gemini-S/FLAMINGOS-2 9900 - 18000
58705.6 8.1 -2.2 LCO/FLOYDS 3500 - 9900
58706.1 8.6 -1.7 SALT/RSS 3500 - 9300
58707.6 10.1 -0.2 LCO/FLOYDS 3500 - 9900
58708.8 11.3 +1.0 LCO/FLOYDS 3500 - 9900
58709.8 11.3 +2.0 LCO/FLOYDS 3500 - 9900
58717.3 19.8 +9.5 NTT/EFOSC2 3300 - 9900
58718.2 20.7 +10.4 NTT/EFOSC2 3600 - 9200
58718.8 21.3 +11.0 LCO/FLOYDS 3800 - 9900
58721.5 24.0 +13.7 Lick/Kast 3500 - 10500
58724.2 26.7 +16.6 NTT/EFOSC2 3300 - 7400
58725.0 27.5 +17.2 SALT/RSS 3500 - 9300
58727.5 30.00 +19.7 Lick/Kast 3500 - 10500
58727.6 30.1 +19.8 LCO/FLOYDS 3800 - 9900
58730.2 32.7 +22.4 NTT/EFOSC2 6000 - 9900
58734.3 36.8 +26.5 Gemini-S/FLAMINGOS-2 9900 - 18000
58735.5 38.0 +27.7 Lick/Kast 3500 - 10500
58736.6 39.1 +28.8 LCO/FLOYDS 3800 - 9900
58752.9 55.4 +45.1 SALT/RSS 3500 - 9300
58757.9 60.4 +50.1 SALT/RSS 3500 - 9300
Table A6: Optical light curve parameters of SN 2019muj. The rising times (𝑡rise)
are based on the date of explosion estimated from the quasi-bolometric light curve
fitting.
Filter U B g V r i
tmax 58706.7(0.5) 58707.8 (0.5) 58708.2(0.6) 58709.3(0.7) 58709.7(0.9) 58710.5(1.2)
Peak Obs. [mag] 15.88(0.08) 16.42(0.05) 16.29(0.06) 16.33(0.05) 16.39(0.07) 16.61(0.07)
Peak Abs. [mag] −16.92(0.09) −16.36(0.06) −16.48(0.07) −16.42(0.06) −16.35(0.08) −16.12(0.09)
Δm15 [mag] 3.1 2.4 2.0 1.2 1.0 0.7
trise [days] 8.5(0.7) 9.6(0.7) 10.0(0.8) 10.5(0.8) 11.5(1.1) 12.3(1.3)
Table A7: Fitting parameters of the quasi-bolometric light curve of SN 2019muj
described in Sec. 3.3.
M56𝑁𝑖 [ M] texp [𝑀𝐽𝐷] td [days] A𝛾
0.031 ± 0.005 58698.2 ± 0.5 7.1 ± 0.5 5.5 ± 1.5
This paper has been typeset from a TEX/LATEX file prepared by the author.
MNRAS 000, 1–23 (2020)