Seventeen Tidal Disruption Events from the First Half of ZTF Survey Observations:
Entering a New Era of Population Studies
Sjoert van Velzen
1,2
, Suvi Gezari
1,3
, Erica Hammerstein
1
, Nathaniel Roth
1,3
, Sara Frederick
1
, Charlotte Ward
1
,
Tiara Hung
4
, S. Bradley Cenko
3,5
, Robert Stein
6
, Daniel A. Perley
7
, Kirsty Taggart
7
, Ryan J. Foley
4
, Jesper Sollerman
8
,
Nadejda Blagorodnova
9
, Igor Andreoni
10
, Eric C. Bellm
11
, Valery Brinnel
12
, Kishalay De
10
, Richard Dekany
13
,
Michael Feeney
13
, Christoffer Fremling
10
, Matteo Giomi
14
, V. Zach Golkhou
11,15
, Matthew J. Graham
10
,
Anna. Y. Q. Ho
10
, Mansi M. Kasliwal
10
, Charles D. Kilpatrick
4
, Shrinivas R. Kulkarni
10
, Thomas Kupfer
15
,
Russ R. Laher
16
, Ashish Mahabal
10,17
, Frank J. Masci
16
, Adam A. Miller
18,19
, Jakob Nordin
12
, Reed Riddle
13
,
Ben Rusholme
16
, Jakob van Santen
6
, Yashvi Sharma
10
, David L. Shupe
16
, and Maayane T. Soumagnac
20,21
1
Department of Astronomy, University of Maryland, College Park, MD 20742, USA; sjoert@nyu.edu
2
Center for Cosmology and Particle Physics, New York University, NY 10003, USA
3
Joint Space-Science Institute, University of Maryland, College Park, MD 20742, USA
4
Department of Astronomy and Astrophysics, University of California, Santa Cruz, CA 95064, USA
5
Astrophysics Science Division, NASA Goddard Space Flight Center, MC 661, Greenbelt, MD 20771, USA
6
Deutsches Elektronensynchrotron, Platanenallee 6, D-15738, Zeuthen, Germany
7
Astrophysics Research Institute, Liverpool John Moores University, 146 Brownlow Hill, Liverpool L3 5RF, UK
8
The Oskar Klein Centre & Department of Astronomy, Stockholm University, AlbaNova, SE-106 91 Stockholm, Sweden
9
Department of Astrophysics/IMAPP, Radboud University Nijmegen, P.O. Box 9010, 6500 GL Nijmegen, The Netherlands
10
Division of Physics, Mathematics, and Astronomy, California Institute of Technology, Pasadena, CA 91125, USA
11
DIRAC Institute, Department of Astronomy, University of Washington, 3910 15th Avenue NE, Seattle, WA 98195, USA
12
Institute of Physics, Humboldt-Universität zu Berlin, Newtonstr. 15, D-12489 Berlin, Germany
13
Caltech Optical Observatories, California Institute of Technology, Pasadena, CA 91125, USA
14
The eScience Institute, University of Washington, Seattle, WA 98195, USA
15
Kavli Institute for Theoretical Physics, University of California, Santa Barbara, CA 93106, USA
16
IPAC, California Institute of Technology, 1200 E California Boulevard, Pasadena, CA 91125, USA
17
Center for Data Driven Discovery, California Institute of Technology, Pasadena, CA 91125, USA
18
Center for Interdisciplinary Exploration and Research in Astrophysics (CIERA) and Department of Physics and Astronomy, Northwestern University, 2145
Sheridan Road, Evanston, IL 60208, USA
19
The Adler Planetarium, Chicago, IL 60605, USA
20
Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA 94720, USA
21
Department of Particle Physics and Astrophysics, Weizmann Institute of Science, Rehovot 76100, Israel
Received 2020 January 13; revised 2020 October 6; accepted 2020 October 9; published 2021 February 8
Abstract
While tidal disruption events (TDEs) have long been heralded as laboratories for the study of quiescent black
holes, the small number of known TDEs and uncertainties in their emission mechanism have hindered progress
toward this promise. Here we present 17 new TDEs that have been detected recently by the Zwicky Transient
Facility along with Swift UV and X-ray follow-up observations. Our homogeneous analysis of the optical/UV
light curves, including 22 previously known TDEs from the literature, reveals a clean separation of light-curve
properties with spectroscopic class. The TDEs with Bowen fluorescence features in their optical spectra have
smaller blackbody radii, lower optical luminosities, and higher disruption rates compared to the rest of the sample.
The small subset of TDEs that show only helium emission lines in their spectra have the longest rise times, the
highest luminosities, and the lowest rates. A high detection rate of Bowen lines in TDEs with small photometric
radii could be explained by the high density that is required for this fluorescence mechanism. The stellar debris can
provide a source for this dense material. Diffusion of photons through this debris may explain why the rise and fade
timescale of the TDEs in our sample are not correlated. We also report, for the first time, the detection of soft X-ray
flares from a TDE on ∼day timescales. Based on the fact that the X-ray flares peak at a luminosity similar to the
optical/UV blackbody luminosity, we attribute them to brief glimpses through a reprocessing layer that otherwise
obscures the inner accretion flow.
Unified Astronomy Thesaurus concepts: Astrophysical black holes (98); Tidal disruption (1696); Galaxy
nuclei (609)
Supporting material: data behind figure, tar.gz file
1. Introduction
The occasional (∼10
−4
yr
−1
) luminous flare of radiation from
a galaxy nucleus due to the tidal disruption of a star by an
otherwise dormant central massive black hole originated as a
theoretical concept (Lidskii & Ozernoi 1979; Rees 1988),but
thanks to the rapid increase in wide-field survey capabilities
across the electromagnetic spectrum, it is now a well-established
class of transients. While the first candidates were detected as
soft X-ray outbursts in previously quiescent galaxy nuclei by the
ROSAT All-Sky Survey (Donley et al. 2002), these tidal
disruption events (TDEs) have more recently emerged as a
unique class of nuclear transients in optical surveys with
common photometric properties: persistent blue colors, a
relatively long rise time compared to most supernovae (SNe),
The Astrophysical Journal, 908:4 (26pp), 2021 February 10 https://doi.org/10.3847/1538-4357/abc258
© 2021. The American Astronomical Society. All rights reserved.
1
and a smooth, power-law decline from peak (van Velzen et al.
2011, 2019e;Hungetal.2017). The spectroscopic features of
TDEs are characterized by a hot, blue thermal continuum and
very broad ((5–15)×10
3
km s
−1
; Arcavi et al. 2014;Hung
et al. 2017) emission lines, which are distinct from nearly all
SNe (when observed postpeak) and active galactic nuclei
(AGNs). The inferred volumetric rate of photometric and
spectroscopic TDEs class falls off steeply above the “Hills
mass,” for which a star can be disrupted before disappearing
behind the black hole event horizon (Hills 1975),further
strengthening the association of this class of transients as bona
fide stellar disruptions (van Velzen 2018).
However, while discoveries of TDEs are becoming increasingly
more common in wide-field optical surveys such as iPTF
(Blagorodnova et al. 2017, 2019;Hungetal.2017),ZTF(van
Velzen et al. 2019e), ASAS-SN (Holoien et al. 2014, 2016a,
2016b, 2019a;Weversetal.2019b), and Pan-STARRS (Gezari
et al. 2012; Chorno ck et al. 2014; Holoien et al. 2019b; Nicholl
et al. 2019a), the nature of what is powering their relatively
uniform optical light curves is uncertain. Unlike the soft X-ray
component detected in some optically selected TDEs, which is
consistent with thermal emission from the inner radii of an
accretion disk (Komossa 2015; Miller et al. 2015;Gezarietal.
2017; van Velzen et al. 2019e;Weversetal.2019b), the inferred
blackbody radius of the UV/optical thermal component is a factor
of 10
–100 larger than expected for the size of the nascent debris
disk expected to form from the circularization of the stellar debris
streams. This implies the existence of an unknown, larger
structure, potentially produced as a result of an outflow or wind
(Miller 2015; Metzger & Stone 2017; Dai et al. 2018),orthe
intersecting debris streams themselves (Piran et al. 2015;Jiang
et al. 2016;Bonnerotetal.2017). Indeed, there are now several
examples of outflow signatures from optical (Hung et al. 2019),
UV (Cenko et al. 2016; Blagorodnova et al. 2018; Brown et al.
2018),andX-ray(Miller et al. 2015;Karaetal.2017)
spectroscopy, plus potentially also in the radio (Alexander et al.
2016, 2017)—however, see van Velzen et al. (2016 ),Pasham&
van Velzen (2018) for a different explanation of the radio
emission from optical TDEs.
There has been a recent expansion of spectroscopic subclasses
for TDEs, from the first optical TDE spectra (van Velzen et al.
2011;Gezarietal.2012), one of which surprisingly showed only
broad He
II lines and no hydrogen emission (Gezari et al. 2012),to
the He-rich to H-rich sequence proposed by Arcavi et al. (2014),to
including classes with Bowen fluorescence emission-line features
(Blagorodnova et al. 2018; Leloudas et al. 2019), low-ionization
Fe
II lines (Wevers et al. 2019b), as well as a TDE that showed the
gradual disappearance of broad H lines, while the broad He
II λ
4686 line remained strong (Nicholletal.2019a).TheUVspectra
of TDEs are also unique, characterized by strong N
III]λ1750
emission but weak Mg
IIλλ2896, 2803 and C III]λ1909 (Cenko
et al. 2016). The nature of this spectral diversity has been attributed
to the chemical composition of the star (Gezari et al. 2012;
Kochanek 2016), ionization state of the debris (Guillochon et al.
2014), radiative transfer effects in an optically thick envelope (Roth
et al. 2016), and reprocessing of X-ray emission through dense,
optically thick gas (Leloudas et al. 2019; Wevers et al. 2019b).
In this paper, we present the largest spectroscopic TDE sample
to date. We discovered correlations between the spectroscopic
subclass of the TDE and the host-galaxy and flare properties.
These correlation provide new insights into the origin of the
spectral diversity in TDEs.
We were able to discover these correlations thanks to a
homogeneous treatment of well-sampled optical/UV light
curves of 31 spectroscopic TDEs. This factor of ≈2 increase
in sample size of known TDEs can, for a large part, be
attributed to the start of the Zwicky Transient Facility (ZTF;
Bellm et al. 2019b) in 2018 March. We searched the ZTF data
for new TDEs using a combination of photometric selection
and spectroscopic and multiwavelength follow-up (van Velzen
et al. 2019e). While ZTF is not always the first survey to report
these events to the Transient Name Server (TNS) and thus
claim discovery credit (see Table 1), for most sources, ZTF
provides the deepest difference-imaging light curves that are
publicly available (Masci et al. 2019; Patterson et al. 2019).
Besides the origin of optical emission, a second important
(and unexpected) observation of optically selected TDEs is
their X-ray faintness. The most common explanation is that the
soft X-rays from accretion in the inner disk are absorbed and
reprocessed into optical photons (e.g., Guillochon et al.
2014;
Auchettl et al. 2017; Dai et al. 2018). In this scenario, the
X-rays can only break out after the obscuring gas has expanded
enough to become transparent to X-rays (Metzger &
Stone 2016; Lu & Kumar 2018). However, intrinsically faint
soft X-ray TDEs have also been proposed as a result of delayed
accretion due to the timescale required for the circularization of
the debris into an appreciable accretion disk (Piran et al. 2015;
Krolik et al. 2016; Gezari et al. 2017). Discriminating between
these models, and thus determining if the optical emission is
powered by accretion or the stream kinetic energy, is possible
by looking at the relative timing and response of the optical
flare to the soft X-ray emission from TDEs (Pasham et al.
2017). Signi ficant soft X-ray variability has recently been
observed, including a late-time brightening that is antic-
orrelated with the smooth decline of the optical component
(Gezari et al. 2017; van Velzen et al. 2019e; Wevers et al.
2019b). In this paper, we present four more optically selected
TDEs with soft X-ray detections, including both flaring and
late-time X-ray brightening, which provide new constraints on
the emission mechanisms.
In Section 2, we present the selection of TDE candidates from
the ZTF stream and spectroscopic follow-up, as well as our
naming scheme for three spectroscopic classes. In Section 3,we
investigate the host galaxies of our TDEs, obtaining estimates of
their mass and star formation histories, followed by Section 4,
which contains the details of our multiwavelength follow-up
observations. In Section 5, we present our light-curve model that
is applied to 39 spectroscopic+photometric TDEs. In Section 6,
we present correlations between features extracted from our light-
curve model, plus a discovery of differences in the photometric
features between the TDEs of each spectroscopic class.
We adopt a flat cosmology with Ω
Λ
=0.7 and
=
H
0
--
7
0kms Mpc
11
. All magnitudes are reported in the AB
system (Oke 1974).
2. Candidate Selection and Classification
2.1. Zwicky Transient Facility
Our search for new TDEs is done exclusively using ZTF data.
The strength of ZTF (Graham et al. 2019) is a combination of
depth (limiting magnitude of m≈20.5 per visit) and area
(47 deg
2
field of view). Most of our sources originate from the
public MSIP survey, which aims (Bellm et al. 2019a) to visit the
entire visible northern sky every three nights in both the g and r
2
The Astrophysical Journal, 908:4 (26pp), 2021 February 10 van Velzen et al.
filters. The use of two filters is an essential ingredient to our TDE
selection pipeline, as it allows for efficient photometric filtering
(Figure 1) to narrow down the number of targets for spectroscopic
follow-up observations.
2.2. ZTF Alert Filtering
We use the information from the data stream (Patterson et al.
2019) of ZTF alerts, which is produced at IPAC and contains
the difference-imaging photometry and astrometry of transients
and variable sources (Masci et al. 2019).
We place no requirement on the host-galaxy type or color.
However, we reject galaxies that can be spectroscopically
classified as broad-line AGNs. This rejection step implies that
our sample will not contain TDE candidates in broad-line
AGNs, such as PS16dtm (Blanchard et al. 2017). For AGN
identification we use the Million Quasars Catalog (Flesch 2015,
v5.2.). In addition, we construct a light curve from the
NeoWISE (Mainzer et al. 2011) photometry and reject any
galaxies with significant variability (
c >dof 10
2
) or a mean
W1 − W2 color that exceeds the AGN threshold of Stern et al.
(2012). Our filter is executed by Ampel (Nordin et al. 2019),
which includes fast catalog matching by catsHTM (Soumagnac
& Ofek 2018), and we use the GROWTH marshal ( Kasliwal
et al. 2019) to coordinate our follow-up observations and
spectroscopic classifications.
Compared to our TDE search in ZTF commissioning data
(van Velzen et al. 2019e), we use a more liberal cut on the star–
galaxy score (Tachibana & Miller 2018) of <0.8. This
increases the galaxy sample at the cost of a much higher
background due to bright variable stars (these often have a
score equal to 0.5 due to issues with the PS1 photometry for
bright and variable objects). We therefore veto the star–galaxy
score if the source has a detected parallax in Gaia DR2 (Gaia
Collaboration et al. 2016, 2018) or if the ratio of the Gaia G-
band flux to the PS1 PSF g, r, i flux (converted to the G band)
is consistent with a point source. Because we require a match to
a known source in the ZTF reference image, we can use a
relatively liberal cut on the real–bogus score (Mahabal et al.
2019) of 0.3.
As demonstrated in Figure 1, TDEs can be discriminated
from SNe and AGNs based on their rise/fade timescale, g – r
color, and lack of color evolution. We rank photometric TDE
candidates for spectroscopic follow-up based on their distance
from the locus of photometric properties of SNe. In general,
we rejected transients that are significantly off-center (mean
offset >0
4), or have significant g−r color evolution
(
()
D
->grdt0.015
day
−1
), or show only a modest flux
increase when comparing the difference flux to the PSF flux in
the ZTF reference image (
->mm1.5
diff ref
). We also rejected
all objects that can be classified as SNe or broad-line AGNs in
our spectroscopic follow-up observations. The details of our
photometric selection, including estimates for the completeness
and selection effects which are required to compute event rates,
will be presented in a forthcoming publication.
2.3. Discovery and Classification History
In Table 1, we list the IAU name, the ZTF name, our internal
nickname,
22
the name given by other optical transient surveys,
and reference to the first public spectroscopic classification of
this transient as a TDE. The table is sorted by the date of the
first ZTF detection and credit for discovery of the transient,
based on the first report to the Transient Name Server (TNS),is
indicated using boldface.
Our TDE discovery pipeline does not use the TNS as input;
we read and filter the ZTF alerts directly from their source
(Patterson et al. 2019). The TNS reporting of ZTF alerts is
mainly provided by AMPEL (Nordin et al. 2019) and by the
Redshift Completeness Factor project (Fremling et al. 2020),
Table 1
Names and Discovery Name (in Boldface)
IAU Name ZTF Name GOT Name Other/Discovery Name First TDE Classification Report
AT2018zr ZTF18aabtxvd Ned PS18kh ATel#11444 (Tucker et al. 2018)
AT2018bsi ZTF18aahqkbt Jon ATel#12035 (Gezari et al. 2018)
AT2018hco ZTF18abxftqm Sansa ATLAS18way ATel#12263 (van Velzen et al. 2018)
AT2018iih ZTF18acaqdaa Jorah ATLAS18yzs, Gaia18dpo This paper
AT2018hyz ZTF18acpdvos Gendry ASASSN-18zj, ATLAS18bafs ATel#12198 (Dong et al. 2018)
AT2018lni ZTF18actaqdw Arya This paper
AT2018lna ZTF19aabbnzo Cersei ATel#12509 (van Velzen et al. 2019d)
AT2019cho ZTF19aakiwze Petyr This paper
AT2019bhf ZTF19aakswrb Varys This paper
AT2019azh ZTF17aaazdb
a
Jaime ASASSN-19dj, Gaia19bvo ATel#12568 (van Velzen et al. 2019a)
b
AT2019dsg ZTF19aapreis Bran ATLAS19kl ATel#12752 (Nicholl et al. 2019b)
AT2019ehz ZTF19aarioci Brienne Gaia19bpt ATel#12789 (Gezari et al. 2019)
AT2019eve ZTF19aatylnl Catelyn Gaia19bti, ATLAS19kfv This paper
AT2019mha ZTF19abhejal Bronn ATLAS19qqu This paper
AT2019meg ZTF19abhhjcc Margaery Gaia19dhd AN-2019-88 (van Velzen et al. 2019b)
c
AT2019lwu ZTF19abidbya Robb ATLAS19rnz, PS19ega This paper
AT2019qiz ZTF19abzrhgq Melisandre ATLAS19vfr, Gaia19eks, PS19gdd ATel#13131 (Siebert et al. 2019)
Notes. Names in boldface indicate the discovery name, i.e., the first survey to report photometry of the transient detection to the TNS.
a
The year 2017 in the ZTF name of this transient is due to a detection by the IPAC alert photometry pipeline at the end of the ZTF commission phase, although after
visual inspection of the difference image, we flag this measurement as spurious.
b
First spectrum obtained by Heikkila et al. (2019) on 2019 February 21 but classification not yet conclusive.
c
First spectrum published by Nicholl et al. (2019c) on 2019 August 1 but classification not yet conclusive.
22
Given the seven-character length of the ZTF names required by the large
volume of ZTF transient alerts, for ease of communication, we chose an
internal naming scheme for our TDE candidates based on characters from the
HBO TV show, Game of Thrones (GOT).
3
The Astrophysical Journal, 908:4 (26pp), 2021 February 10 van Velzen et al.
plus more recently by a filter implemented in the ALeRCE
broker. For 10 of the 17 sources in our sample, ZTF was the
first survey to report a detection to TNS. As listed in Table 1,
ATLAS provided three discoveries, ASAS-SN two discoveries,
and Gaia and PS1 each claim one more discovery.
2.4. Spectroscopic Classification
In order to classify the TDEs into spectroscopic subclasses,
we use the “best” spectrum, namely high signal-to-noise and
prominent line features, for each of our TDEs from our various
follow-up programs with the 4.3 m Discovery Channel
Telescope De Veny Spectrograph (DCT/De Veny, PI: Gezari),
the 200 inch Palomar Telescope Double Spectrograph (P200/
DBSP, PI: Kulkarni), the 10 m Keck Low Resolution Imaging
Spectrograph (Keck/LRIS, PI: Kulkarni), and the 3 m Lick
Kast Double Spectrograph (Lick/Kast, PI: Foley). Spectra
were reduced with PyRAF using standard long-slit spectrosc-
opy data reduction procedures. For those spectra not corrected
for telluric absorption, we show the spectra in Figure 2 with
those wavelength regions masked out. In three cases, we use
publicly available spectra from the TNS. In Table 2,we
indicate the IAU name, date, phase in days since peak, and
telescope and instrument of the spectrum we use for
determining the spectroscopic subclassification shown in
Figure 2, the TDE class, and the redshift. In all cases, the
redshift is measured from host-galaxy absorption features.
We find that our ZTF TDE sample can be divided into three
spectroscopic classes:
1. TDE-H: broad Hα and Hβ emission lines.
2. TDE-H+He: broad Hα and Hβ emission lines and a
broad complex of emission lines around He
II λ4686. The
majority of the sources in this class also show N
IIIλ4640
and emission at λ4100 (identified as N III λ4100 instead
of Hδ), plus and in some cases also O
IIIλ3760.
3. TDE-He: no broad Balmer emission lines, a broad
emission line near He
IIλ4686 only.
In our flux-limited sample of TDEs with ZTF observations,
the relative ratios of the classes are H:H+He:He=8:7:1. In
Section 7, we will elaborate on how the rarity of the TDE-He
class might be an important clue to understand what conditions
are needed to provide the spectroscopic properties of TDEs.
Two of the TDEs (AT2019meg and AT2019dsg) also have
strong narrow emission lines from star formation in their host
galaxies.
These classifications are based on a single spectral epoch
obtained near the peak of the flare. There is at least one case in
which a TDE showed the late-time disappearance of Hα
emission (Nicholl et al. 2019a), which according to our
classification scheme, would result in a change of spectral class
from TDE-H+He to TDE-He.
A detailed analysis of the spectroscopic evolution of the ZTF
TDEs and their line features with time will be presented in a
future paper (T. Hung et al. 2021, in preparation). For one of
our TDEs, AT2019eve, a follow-up spectrum obtained 6
months after peak, demonstrated the appearance of strong,
narrow He
I lines, potentially a signature of a Type Ibn SN
(although the presence of strong Balmer lines is inconsistent
with this classification). We therefore label the spectral class of
AT2019eve as “Unknown” until more careful analysis of the
host-galaxy contribution to the emission-line spectrum. We
note also that the rise time of the flare and the blackbody
temperature are also outliers compared to the rest of the TDEs
in our sample, the blackbody temperature of AT2019eve is
below the threshold for TDE identification established by van
Velzen et al. (2020).
Figure 1. Yield of nuclear transients after 1.5 yr of ZTF observations. Contours
enclose two-thirds of all spectroscopically classified nuclear supernovae (SNe)
in our sample and two-thirds of the AGNs. The latter are classified based on
archival data or prior variability. In the top panel, we see that that TDEs have
both longer rise times and a longer fading timescale compared to the majority
of SNe. The middle panel demonstrates that color evolution provides further
separation of TDEs from SNe. Here we display the mean g−r color and the
color change (Δ(g−r)/t) , both measured using all detections of the light
curve. Tidal disruption flares show an almost constant optical color, while in
postpeak observations most SNe show cooling (i.e., an increase of the color).
For photometric selection of TDEs detected before maximum light, their blue
color and slow rise time can be used (bottom panel), although this metric yields
a larger background of SNe.
4
The Astrophysical Journal, 908:4 (26pp), 2021 February 10 van Velzen et al.
Figure 2. Spectroscopic classifications of our ZTF TDE sample from medium-resolution spectroscopy. Left: TDEs with Balmer line features only (TDE-H, in red).
Right: TDEs with Balmer lines and a broad emission feature at He
II (TDE-H+He, in green), and TDEs with only He II emission (TDE-He, in blue). Spectra have not
been host galaxy subtracted.
Table 2
Spectroscopic Observations and TDE Classifi cation
IAU Name Date Phase Telescope/Inst. TDE Class Redshift ID
AT2018zr 2018 Mar 28 25 WHT/ISIS
a
TDE-H 0.071 1
AT2018bsi 2018 May 13 34 DCT/De Veny TDE-H+He 0.051 2
AT2018hco 2018 Nov 10 29 Keck/LRIS TDE-H 0.088 3
AT2018iih 2019 Mar 10 102 DCT/De Veny TDE-He 0.212 4
AT2018hyz 2018 Nov 12 6 FTN/Floyds-N
b
TDE-H 0.0458 5
AT2018lni 2019 Mar 1 81 DCT/De Veny TDE-H+He 0.138 6
AT2018lna 2019 Jan 26 0 Palomar/DBSP TDE-H+He 0.091 7
AT2019cho 2019 May 2 58 DCT/De Veny TDE-H+He 0.193 8
AT2019bhf 2019 May 29 90 DCT/De Veny TDE-H 0.1206 9
AT2019azh 2019 May 1 46 Keck/LRIS TDE-H+He 0.0222 10
AT2019dsg 2019 May 13 13 NTT/EFOSC2
c
TDE-H+He 0.0512 11
AT2019ehz 2019 Jun 14 35 Lick/Kast TDE-H 0.074 12
AT2019eve 2019 Jun 29 50 DCT/De Veny Unknown 0.0813 13
AT2019mha 2019 Aug 27 17 Palomar/DBSP TDE-H 0.148 14
AT2019meg 2019 Aug 10 8 Palomar/DBSP TDE-H 0.152 15
AT2019lwu 2019 Aug 27 31 DCT/De Veny TDE-H 0.117 16
AT2019qiz 2019 Nov 5 29 DCT/De Veny TDE-H+He 0.0151 17
Notes.
a
Spectrum published in Hung et al. (2019).
b
Publicly available spectrum on TNS posted by Dong et al. (2018).
c
Publicly available spectrum on TNS posted by Nicholl et al. (2019b).
5
The Astrophysical Journal, 908:4 (26pp), 2021 February 10 van Velzen et al.
