A&A, 703, A81 (2025)
https://doi.org/10.1051/0004-6361/202555626
c© The Authors 2025
Astronomy
&Astrophysics
Polarimetric diversity in tidal disruption events: Comparative study
of low-polarised sources with AT2020mot
A. Floris1,2,3,? , I. Liodakis1,4, K. I. I. Koljonen5 , E. Lindfors6,7, B. Agís-Gonzalez1 , A. Paggi1,2 , D. Blinov1,2,
K. Nilsson7, I. Agudo8 , P. Charalampopoulos6 , M. A. Díaz Teodori6,9 , J. Escudero Pedrosa8,10 ,
J. Otero-Santos8,11, V. Piirola6 , M. Newsome12,13 , and S. Van Velzen14
1 Institute of Astrophysics, FORTH, N.Plastira 100, Vassilika Vouton, 70013 Heraklion, Greece
2 Department of Physics University of Crete, Voutes University Campus, 70013 Heraklion, Greece
3 National Institute for Astrophysics (INAF), Astronomical Observatory of Padova, IT-35122 Padova, Italy
4 NASA Marshall Space Flight Center, Huntsville, AL 35812, USA
5 Institutt for Fysikk, Norwegian University of Science and Technology, Høgskloreringen 5, Trondheim 7491, Norway
6 Department of Physics and Astronomy, 20014 University of Turku, Turku, Finland
7 Finnish Centre for Astronomy with ESO (FINCA), Quantum, Vesilinnantie 5, 20014 University of Turku, Turku, Finland
8 Instituto de Astrofísica de Andalucía, IAA-CSIC, Glorieta de la Astronomía s/n, 18008 Granada, Spain
9 Nordic Optical Telescope, Rambla José Ana Fernández, Pérez 7, E-38711 Breña Baja, Spain
10 Center for Astrophysics | Harvard & Smithsonian, 60 Garden Street, Cambridge, MA 02138, USA
11 Istituto Nazionale di Fisica Nucleare, Sezione di Padova, 35131 Padova, Italy
12 Las Cumbres Observatory, 6740 Cortona Drive, Suite 102, Goleta, CA 93117-5575, USA
13 Department of Astronomy, The University of Texas at Austin, 2515 Speedway, Stop C1400, Austin, TX 78712, USA
14 Leiden Observatory, Leiden University, PO Box 9513, 2300 RA Leiden, The Netherlands
Received 22 May 2025 / Accepted 5 October 2025
ABSTRACT
Context. Tidal disruption events (TDEs) occur when a star is disrupted by the tidal forces of a supermassive black hole (SMBH),
which produces bright multi-wavelength flares. Among these events, AT2020mot has so far exhibited the highest recorded optical
polarisation, with tidal shocks proposed as the primary source of its polarised emission.
Aims. We present a comprehensive analysis of 13 TDEs with available polarimetric observations, aiming to determine whether the
unusually high polarisation of AT2020mot stems from unique physical processes or arises from mechanisms shared by other TDEs.
Methods. We present new optical polarisation measurements of TDEs obtained from multiple ground-based telescopes, combining
them with optical, UV, and X-ray light curves from the Zwicky Transient Facility and the Swift observatory. We derived intrinsic TDE
properties – such as SMBH and stellar masses – using MOSFiT and TDEMass, and compared them with those of the sample popula-
tion.
Results. Our population study reveals that AT2020mot aligns with the broader TDE sample in terms of most physical properties,
including blackbody temperature, luminosity, and rise timescales. However, its optical polarisation degree is exceptionally high com-
pared to the low or undetected polarisation observed in other events. Additionally, according to our MOSFiT fit, AT2020mot has an
elevated column density, which suggests a more complex environment than is typically assumed.
Conclusions. We conclude that although AT2020mot fits well within the general TDE population in terms of global characteris-
tics, its extraordinarily high polarisation and higher column density challenge current models based purely on shock or reprocessing
mechanisms. More extensive, time-resolved polarimetric monitoring of newly discovered TDEs will be critical to determine whether
AT2020mot represents an outlier or the extreme end of a continuum of TDE properties.
Key words. techniques: polarimetric – galaxies: active – galaxies: nuclei
1. Introduction
Tidal disruption events (TDEs) are transient astrophysical phe-
nomena that occur when a star is scattered onto an orbit pass-
ing sufficiently close to the supermassive black hole (SMBH)
in the central region of its host galaxy (Rees 1988; Komossa
2015). Once the black hole’s tidal forces overcome the star’s
self-gravity, stripping the star of its gas (Hills 1975), a bright
flare is produced that typically emits across the X-ray, ultravio-
let (UV), optical, and infrared (IR) wavelengths.
UV-optical TDEs are identified through pronounced flares
in these bands and by their broad optical and UV spec-
tral lines (van Velzen et al. 2020). In most cases, however,
? Corresponding author: afloris@ia.forth.gr
the optical peak is not accompanied by contemporaneous X-
ray emission (Gezari et al. 2008, 2012; Holoien et al. 2016;
Auchettl et al. 2018; Hinkle et al. 2021; van Velzen et al. 2021).
The origin of the optical luminosity and the reason for the
apparent absence of contemporaneous X-rays remain debated.
Two main scenarios have been proposed to explain these obser-
vations: (1) an accretion disc forms rapidly after disruption
and produces X-rays, which are then reprocessed into UV-
optical emission by an optically thick layer of gas at radii much
larger than the tidal radius (Metzger & Stone 2016; Dai et al.
2018), and (2) a predominantly shock-powered mechanism
whereby stellar debris streams collide in the outer regions
of a highly eccentric disc, rather than power generating pri-
marily from accretion onto the black hole (Piran et al. 2015;
Open Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0),
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
This article is published in open access under the Subscribe to Open model. Subscribe to A&A to support open access publication.
A81, page 1 of 11
Floris, A., et al.: A&A, 703, A81 (2025)
Shiokawa et al. 2015). Different events appear to favour dif-
ferent physical mechanisms. For example, accretion disc mod-
els are often invoked to explain Bowen fluorescence lines
(Leloudas et al. 2019; Blagorodnova et al. 2019) and coronal
emission lines (Trakhtenbrot et al. 2019; Koljonen et al. 2024),
presumably formed by excitation from high-energy photons,
even in TDEs where X-rays are partially or completely obscured.
On the other hand, the variable polarisation reported in several
TDEs (Leloudas et al. 2022; Patra et al. 2022; Liodakis et al.
2023; Koljonen et al. 2024) and, in particular, the extraordi-
narily high polarisation degree observed in AT2020mot (the
highest measured to date in the absence of a jet, Π ∼
25±4%; Liodakis et al. 2023) strongly suggests that at least
some TDEs involve mechanisms other than electron-scattering.
Pure electron-scattering reprocessing scenarios generally predict
a maximum polarisation degree <14% (Leloudas et al. 2022;
Charalampopoulos et al. 2023).
Polarimetry offers a sensitive probe of TDE geometry and
emission physics. Even in the absence of intrinsically polarised
sources (e.g. relativistic jets, which appear to be rare among
optical TDEs; Wiersema et al. 2012, 2020), moderate to high
polarisation can arise through electron scattering whenever the
reprocessing medium or shock fronts are aspherical. Measure-
ments of the polarisation degree (Π) and polarisation angle (Θ)
thus constrain the shape and orientation of the scattering region
(Chornock et al. 2014; Patra et al. 2022) and can be used to dis-
criminate between reprocessing-dominated and shock-powered
scenarios.
In this context, AT2020mot stands out as a unique source:
all other TDEs with polarimetric measurements exhibit sig-
nificantly lower Π, even before accounting for its host-galaxy
contamination (Liodakis et al. 2023). Since the integrated host
galaxy emission is unpolarised, it effectively depolarises the
TDE emission measured within a circular aperture. Therefore,
the host galaxy correction renders a higher level of polarisation.
Late-time polarimetry, obtained after the flare has faded, is there-
fore indispensable for isolating the intrinsic polarisation of the
event (Andruchow et al. 2008).
Here we present new optical polarimetric observations
obtained within the Black hOle Optical polarization TimE-
domain Survey (BOOTES) programme, a campaign designed
to systematically monitor TDE polarisation. We report our first
results, including late-time measurements of AT2020mot, and
present a homogeneous analysis that combines these data with
photometry for a sample of well-observed TDEs. Our goal is to
compare the multi-wavelength properties of AT2020mot to those
of other TDEs in order to determine the physical reasons for why
it stands out from the rest of the TDE population.
We describe our sample selection and observations in Sect. 2,
including new polarisation measurements for the sources in our
sample. Section 3 outlines the methods used to estimate TDE
properties and the results of our population study, and we discuss
the interpretation of our findings in Sect. 4. Finally, we provide
a summary of our conclusions in Sect. 5.
Throughout this paper we adopt a flat Λ cold dark matter
cosmology with H0 = 67.4 km s−1 Mpc−1, ΩM = 0.315, and
ΩΛ = 0.685 (Planck Collaboration VI 2020).
2. New polarisation observations
2.1. Sample
The sample used in this work consists of TDEs that were moni-
tored as part of the BOOTES programme of polarimetric obser-
vation, prior to mid-2024. The data were gathered as part of an
ongoing campaign to track the polarisation evolution of TDEs
with several ground-based facilities; each source was observed
at multiple epochs throughout its flare and subsequent decline.
The polarimetric observations were obtained with the fol-
lowing instruments:
– The RoboPol polarimeter (Ramaprakash et al. 2019) on the
1.3 m telescope at the Skinakas Observatory, Crete.
– The ALFOSC polarimeter (see Nilsson et al. 2018, for
details on data reduction and calibration) on the Nordic Opti-
cal Telescope (NOT).
– The DIPOL-1 polarimeter (Otero-Santos et al. 2024) on the
90 cm telescope (T90) at the Observatorio de Sierra Nevada
(OSN).
– The CAFOS polarimeter1 (see Escudero Pedrosa et al. 2024,
for details on data reduction and calibration using CAFOS
and DIPOL-1) on the 2.2 m telescope at the Centro
Astronómico Hispano en Andalucía (CAHA).
Standard polarimetric calibration procedures were applied to
all data. We observed both unpolarised and polarised standard
stars to verify the accuracy of the instrumental polarisation
and to calibrate the instrumental zero-point for the polarisa-
tion angle. Uncertainties in the polarisation measurements are
derived through error propagation including photon noise of the
detector, background subtraction and instrumental corrections
for both Π and Θ from standard stars. Typical exposure times per
target ranged from a few minutes up to over an hour, depending
on the source magnitude and the specific telescope-instrument
configuration, in order to obtain a sufficiently high signal-to-
noise ratio (S/N) that ensured reliable polarisation measure-
ments or 3σ upper limits. The new measurements reported here
are listed in Table A.1. Previously published polarimetry for
AT2020mot (Liodakis et al. 2023), AT2022fpx (Koljonen et al.
2024), and AT2023clx (Koljonen et al. 2025) are also included.
The final sample comprises 13 TDEs. Among these,
AT2020mot is the only source displaying a high polarisation
degree (Π ≈ 25%), posing a challenge to current models. By
contrast, the remaining TDEs exhibit polarisation degrees of
Π < 6% or yield only non-detections (Π − 3σΠ < 0%, where σΠ
is the associated uncertainty). One of the TDEs in our sample,
AT2022dbl, has undergone two flares: the first in February 2022
(Stanek 2022) and the second in late 2023 (Yao et al. 2024), which
have been interpreted as arising from partial disruption of the
same star (Lin et al. 2024). In this work we treated the two flares
separately and refer to the second flare as ‘AT2022dbl2’. The
basic properties of the full sample, together with the maximum
observed polarisation degree (Πmax), are summarised in Table 1.
2.2. AT2020mot follow-up observations
As discussed in Sect. 1, it is critical to measure the TDE’s intrin-
sic polarisation, which requires subtracting any host contribution.
In Liodakis et al. (2023), the polarisation degree of AT2020mot
was corrected under the assumption that its host galaxy was
unpolarised. To verify this assumption, we obtained late-time
polarisation observations of the host galaxy, once the flare had
faded. We confirmed through multiple observations that the host
galaxy is unpolarised (Π < 0.75% at 3σ confidence interval).
This confirms that AT2020mot, which was the most polarised
TDE to date even in absence of host-correction (Π ∼ 8.3%
uncorrected), had a maximum intrinsic polarisation degree∼25%
(Liodakis et al. 2023). Figure 1 illustrates the polarisation evolu-
tion of AT2020mot alongside its optical and UV light curves.
1 https://www.caha.es/telescope-2-2m/cafos
A81, page 2 of 11
Floris, A., et al.: A&A, 703, A81 (2025)
Table 1. TDE sample properties.
Name Spectral type z Πmax
[%]
(1) (2) (3) (4)
AT2020mot TDE H+He 0.07 25 ± 4
AT2020afhd TDE H+He 0.027 <2.8
AT2022dbl TDE H+He 0.0284 <2.1
AT2022fpx TDE H+He + ECLE 0.073 2.4 ± 0.5
AT2022gri TDE featureless 0.028 1.29 ± 0.39
AT2022hvp TDE He 0.12 6.86 ± 1.14
AT2022upj TDE He+ECLE 0.054 6.40 ± 1.20
AT2022wtn TDE H+He 0.049 <0.72
AT2023clx TDE H+He 0.01107 4.8 ± 0.6
AT2023ugy TDE H 0.106 <1.2
AT2023lli TDE H+He 0.036 0.62 ± 0.20
AT2024bgz TDE H+He 0.0585 <8.1
AT2024gre TDE featureless 0.12 4.1 ± 1.1
Notes. (1) TNS name of the source. (2) Spectral type of the TDE.
(3) Redshift from the TNS. (4) Maximum host-corrected polarisation
degree of the source detected. Upper limits are 3-σ non-detections.
2.3. Host-galaxy correction
For sources with at least one significant polarisation detection
and a non-negligible host contribution to the photometry, we cor-
rected for host-galaxy dilution following Liodakis et al. (2023),
assuming that the host-galaxy is unpolarised:
Πcorr = Πobs × II − Ihost · (1)
Here, Πobs is the observed polarisation degree, I is the observed
total flux density of the source at the time of the polarimet-
ric observation, and Ihost the host-galaxy flux measured as the
median flux obtained from pre-flare observations in the g and
r bands from the Zwicky Transient Facility (ZTF). Photome-
try in the g and r bands was converted to photometry in the
filter used for polarimetric observations using the transforma-
tions described in Tonry et al. (2012). AT2023lli, despite exhibit-
ing detections of Π, lacks contemporaneous ZTF photometry,
preventing a reliable host correction. All host-corrected mea-
surements are listed in Table A.2. Additionally, we estimated
the maximum galactic interstellar polarisation in the various
adopted bands using the Serkowski et al. (1975) law and the
Schlafly & Finkbeiner (2011) reddening maps. All sightlines
yield a maximum expected interstellar polarisation (Πmax) of less
than a fraction of a percent, well below our statistical errors.
Thus, no interstellar polarisation correction was applied, since
it is also negligible with respect to the host correction.
2.4. Spectral classification
Spectral observations are essential both for distinguishing TDEs
from other types of transients and for tracing their physical
properties over time (van Velzen et al. 2020; Holoien et al. 2020;
Charalampopoulos et al. 2024). We report the spectral type of
the sources as observed in their classification spectra available
on the Transient Name Server (TNS) website2 and classified
according to the convention detailed in van Velzen et al. (2020)
and Langis et al. (2025):
2 https://www.wis-tns.org/
– H TDEs exhibit only broad Balmer lines in their optical spec-
trum.
– He TDEs exhibit only broad Heiiλ4686 lines in their optical
spectrum.
– H+He TDEs exhibit both Balmer and Heii broad lines in
their optical spectrum.
– Featureless TDEs exhibit no broad lines in their optical spec-
trum.
When high-ionisation coronal lines are present, we consider
the source to be an extreme coronal-line emitter (ECLE;
Trakhtenbrot et al. 2019). Approximately half of our sample
are H+He TDEs; the remainder consists of one pure H TDE
(AT2023ugy), two He TDEs, and two featureless events. Two
objects in the sample are classified as ECLEs.
3. Population study
Our goal is to identify what makes AT2020mot uniquely highly
polarised, whereas the majority of optical TDEs exhibit little or
no polarisation. To this end, we compared the UV, optical, and X-
ray light curves of the 13 events in our sample and searched for
correlations between their photometric and polarimetric proper-
ties.
We used photometry from the Swift Ultraviolet/Optical Tele-
scope (UVOT; Roming et al. 2005), which provides three near-
UV bands (UVW2, UVM2, and UVW1) and three optical
bands (U, B, and V). Optical light curves in the ZTF g and
r bands were obtained from the BHTOM3 archive and pro-
cessed as described by Langis et al. (2025). X-ray data were
collected from all available Swift X-ray telescope and XMM-
Newton observations; the reduction followed the procedure of
Langis et al. (2025). Source-by-source X-ray properties are sum-
marised in Appendix B.
Figure 1 displays the multi-band light curve of AT2020mot
alongside its measured Π and Θ. UVOT magnitudes were con-
verted from the Vega system to the AB system using the con-
version factors from Blanton & Roweis (2007). Figure 2 dis-
plays the distributions of the Stokes parameters q and u, high-
lighting that AT2020mot is the most highly polarised TDE in
our sample. By contrast, the other TDEs are mostly consistent
with low polarisation, even after the host-correction. Both near
the peak (approximately 23 days after the peak and then again
approximately 76 days after peak luminosity) and at late times,
AT2020mot also appears unpolarised, matching the general TDE
population. This raises some fundamental questions, suggesting
that the other TDEs might have been missed at epochs of poten-
tially higher Π. This prompted us to perform a broader popula-
tion study, exploring the physical parameters derived from our
multi-wavelength dataset.
3.1. Blackbody spectral energy distribution fitting
We modelled the spectral energy distribution of each TDE at
peak brightness with a single black body, using an emcee-
based algorithm (Foreman-Mackey et al. 2013). The fits com-
bined host-subtracted Swift/UVOT and ZTF photometry: for
UVOT we adopted the latest available post-flare observations as
a host template, while for ZTF we subtracted the median pre-
flare magnitude. All fluxes were shifted to the rest frame and cor-
rected for Galactic extinction using the Schlafly & Finkbeiner
(2011) reddening maps.
3 https://bh-tom2.astrolabs.pl/
A81, page 3 of 11
Floris, A., et al.: A&A, 703, A81 (2025)
14
16
18
20A
B 
M
ag
ni
tu
de
 +
 c
on
st
an
t
g +2.5
r +2.5
UVW2 -1
UVM2 -1.5
UVW1 -2
U -2.5
B -3
V -3.5
0
10
20
 [%
]
I
R
V
58750 59000 59250 59500 59750 60000 60250 60500
MJD
0
50
100
150
 [
]
I
R
V
Fig. 1. Top panel: AB-magnitude light curve of AT2020mot, rescaled for clarity. Middle panel: Intrinsic Π measurements of AT2020mot over time.
Non-detections, defined as Π − 3σΠ < 0%, are plotted as upper limits. Bottom panel: Θ measurements of AT2020mot over time. Non-detections
are omitted, as Θ is undefined in such cases.
10 5 0 5 10
q  [%]
10
5
0
5
10
u 
 [%
]
Other TDEs
AT2020mot
20 10 0 10 20
q  [%]
20
10
0
10
20
Other TDEs
AT2020mot
Fig. 2. Left panel: Observed Stokes q and u parameters for the sample used in this work. Measurements associated with AT2020mot are highlighted
in red. The dashed grey circle represents a value of Π = 6%. Right panel: Same, but for the host-corrected Stokes q and u parameters for the sample
used in this work.
A81, page 4 of 11
Floris, A., et al.: A&A, 703, A81 (2025)
Table 2. Peak bolometric luminosities and blackbody temperatures.
Name log Lbol Tbb
[erg s−1] [K]
(1) (2) (3)
AT2020mot 44.10+0.03−0.03 31706
+2861
−2319
AT2020afhd 44.03+0.01−0.02 47698
+1641
−2599
AT2022dbl 43.81+0.02−0.02 35413
+2055
−1812
AT2022dbl2 43.36+0.03−0.04 36158
+4013
−3194
AT2022fpx 43.77+0.01−0.01 15187
+341
−323
AT2022gri 43.42+0.02−0.02 22209
+1331
−1135
AT2022hvp 45.29+0.01−0.01 39734
+1051
−1013
AT2022upj 43.73+0.03−0.03 29632
+2610
−2114
AT2022wtn 43.35+0.01−0.01 14912
+503
−459
AT2023clx 42.66+0.01−0.01 15280
+427
−400
AT2023ugy 44.54+0.02−0.02 47987
+1473
−2800
AT2023lli 43.67+0.04−0.04 37242
+6097
−4854
AT2024bgz 43.89+0.01−0.01 16880
+421
−396
AT2024gre 44.13+0.01−0.01 17125
+356
−341
Notes. (1) TDE name. (2) Logarithm of the bolometric luminosity at
the peak. (3) Blackbody temperature fit to the photometric data of the
source at the peak.
From the best-fit blackbody temperature and distance (d) of
each TDE, we calculated the peak bolometric luminosity (Lbol),
adopting the cosmology defined in Sect. 1. Table 2 lists the
resulting bolometric luminosities and blackbody temperatures.
3.2. Black hole and stellar mass estimates
To constrain the black hole mass (MBH) and the mass of the dis-
rupted star (Mstar), we employed two approaches based on dif-
ferent publicly available codes:
– TDEMass (Ryu et al. 2020) uses an outer shock model to
describe the TDE emission, requiring peak Lbol and Tbb as
inputs. This method assumes shocks during debris circular-
isation are the main power source of the flare. The code is cal-
ibrated for 105 <∼ MBH/M <∼ 5 × 107 and 0.1 <∼ Mstar/M <∼
40; two of our events (AT2022hvp and AT2023clx) fall out-
side this range and are thus omitted from the corresponding
columns.
– MOSFiT (Guillochon et al. 2018) employs the fast circularisa-
tion model (Mockler et al. 2019) and assumes that the optical
and UV emission originates from X-ray photons produced by
an optically thick gas layer surrounding the rapidly formed
accretion disc. We fitted the multi-band light curves with ten
free parameters (fixing the source z) and derived posterior dis-
tributions for key physical properties, including MBH, Mstar,
the column density of the gas (NH) in the central region of the
host galaxy, and the scaled impact parameter (b), which is a
measure of how close to the black hole the star is disrupted.
Both approaches used host-subtracted and de-reddened data
as described in Sect. 3.1. The results are listed in Table 3.
Following Mockler et al. (2019), we included a systematic
uncertainty of ±0.20 dex to the MBH estimate and a system-
atic uncertainty of ±0.66 dex on the Mstar estimate. In general,
MOSFiT returns larger MBH values than TDEMass, while the
stellar-mass estimates are broadly consistent. AT2022fpx and
AT2022gri are especially striking: their MOSFiT-derived MBH
values are significantly larger than those of the rest of the sam-
ple, which also translates into correspondingly large Mstar. In
contrast, TDEMass places both AT2022fpx and AT2022gri at
MBH ∼ 106 M and subsolar Mstar. Meanwhile, AT2020mot is
consistent in the two models within a 2σ uncertainty and aligns
reasonably well with the general population, despite having the
third-largest MOSFiT-derived MBH value in our sample.
3.3. Light curve fitting
To estimate the rise time (trise) of each flare, we fitted the ZTF g
and r light curves with an asymmetric Gaussian profile using a
minimum-χ2 approach. The apparent magnitude, m(t), was mod-
elled as
m(t) =
mhost − A exp
(
− t−tpeak2σ
)2
, t ≤ tpeak,
mhost − B − A exp
(
− t−tpeak2τ
)2
, t > tpeak,
where mhost is the median pre-flare magnitude, A is the flare
amplitude, and σ and τ characterise the widths of the rise and
decay phases, respectively. The parameter B accounts for any
plateau-like emission above the host after peak brightness (e.g.
Mummery & van Velzen 2025).
We defined trise as the difference between the times at which
the flare amplitude first reaches 1% and then 99% of its peak,
based on σ. Although this approach could underestimate the
total rise if observations are sparse, it treats all sources uniformly
and thus remains suitable for our population study. The tpeak and
trise parameters determined from the light curve fitting are dis-
played in Table 4. Given the sparse sampling of the ZTF data for
AT2023ugy and AT2023lli, the light curve fitting for those two
sources has been performed using Asteroid Terrestrial-impact
Last Alert System (ATLAS) c- and o-band data.
3.4. X-ray luminosity
We derived the peak X-ray luminosity (LX) in the 0.3–10 keV
range for each event from its X-ray flux FX, and only one source
(AT2020afhd) showed detectable X-ray emission near maximum
light. For the rest, we report 3σ upper limits based on non-
detections.
Table 5 compares the key physical parameters of AT2020mot
with those of the rest of the sample. For each parameter, we list
the median value of the sample (excluding AT2020mot) along
with its standard deviation, then contrast these with the cor-
responding values of AT2020mot. We find that AT2020mot is
consistent with the rest of the sample for most of the measured
parameters. However, it shows a marked deviation in both its
maximum polarisation degree (Πmax) and the NH inferred from
MOSFiT.
4. Discussion
4.1. Polarisation trends
We have presented new polarimetric observations of a sam-
ple of TDEs, most of which exhibit Π < 5% or remain
below our detection threshold. Such low values are consistent
with expectations from both rapid-disc-formation models, in
which X-rays are reprocessed by an optically thick layer (e.g.
A81, page 5 of 11
Floris, A., et al.: A&A, 703, A81 (2025)
Table 3. TDEMass and MOSFiT mass results.
Name TDEMass MBH TDEMass M∗ MOSFiT MBH MOSFiT M∗
[106 M] [M] [106 M] [M]
(1) (2) (3) (4) (5)
AT2020mot 2.80+0.96−0.78 0.97
+0.08
−0.07 18.20
+15.68
−7.88 0.63
+2.40
−0.56
AT2020afhd 0.75+0.17−0.10 0.80
+0.02
−0.02 2.95
+3.22
−1.57 1.21
+4.65
−0.48
AT2022dbl 1.00+0.24−0.20 0.67
+0.03
−0.03 1.20
+0.94
−0.54 1.00
+3.62
−0.81
AT2022dbl2 0.36+0.12−0.09 0.23
+0.05
−0.05 5.13
+5.34
−2.00 0.67
+2.55
−0.66
AT2022fpx 10.00+0.93−0.89 0.92
+0.03
−0.03 213.80
+132.94
−81.97 14.25
+51.55
−12.42
AT2022gri 1.50+0.31−0.27 0.43
+0.04
−0.04 281.84
+196.79
−112.02 15.07
+54.95
−13.90
AT2022hvp ... ... 14.13+14.05−6.89 0.99
+3.59
−0.82
AT2022upj 1.30+0.45−0.35 0.62
+0.05
−0.04 4.57
+5.90
−2.53 0.05
+0.19
−0.04
AT2022wtn 3.90+0.48−0.45 0.50
+0.02
−0.03 6.03
+5.72
−3.08 0.14
+0.59
−0.13
AT2023clx ... ... 0.47+5.29−0.41 0.05
+0.30
−0.04
AT2023ugy 1.90+0.28−0.13 2.2
+0.45
−0.31 9.33
+12.55
−5.61 0.52
+3.18
−0.51
AT2023lli 0.58+0.40−0.24 0.51
+0.07
−0.07 3.47
+3.29
−1.69 0.40
+1.64
−0.39
AT2024bgz 10.00+0.93−0.89 1.00
+0.03
−0.03 2.63
+3.26
−1.48 0.22
+1.08
−0.21
AT2024gre 14.00+0.51−0.54 2.10
+0.22
−0.20 2.51
+2.86
−1.22 0.57
+2.31
−0.56
Notes. (1) TNS name of the source. (2) Black hole mass estimated using the TDEMass code. (3) Mass of the disrupted star inferred using the
TDEMass code. (4) Black hole mass estimated using the MOSFiT code. (5) Mass of the disrupted star inferred using the MOSFiT code.
Table 4. Light curve fitting results.
Name tpeak trise
[d] [d]
(1) (2) (3)
AT2020mot 59057.87 71.5 ± 2.5
AT2020afhd 60351.53 68.6 ± 1.8
AT2022dbl 59636.28 24.0 ± 1.2
AT2022dbl2 60342.72 52.3 ± 3.4
AT2022fpx 59785.53 136.4 ± 1.3
AT2022gri 59759.52 129.4 ± 8.4
AT2022hvp 59692.11 9.2 ± 0.8
AT2022upj 59887.48 104.0 ± 4.2
AT2022wtn 59874.31 35.3 ± 1.4
AT2023clx 59987.50 2.9 ± 10.1
AT2023ugy 60233.40 42.2 ± 3.5
AT2023lli 60169.69 67.7 ± 1.9
AT2024bgz 60349.57 29.0 ± 2.1
AT2024gre 60441.22 56.4 ± 2.2
Notes. (1) TDE name. (2) Peak time MJD. (3) Duration of the rise phase
of the TDE flare.
Leloudas et al. 2022; Charalampopoulos et al. 2023), and shock-
powered scenarios (e.g. Piran et al. 2015; Shiokawa et al. 2015;
Charalampopoulos et al. 2023), although the latter allow for a
wider range of Π and a time-variable polarisation angle that
depends on the geometry of the colliding debris streams. At
present these measurements cannot discriminate between the
two pictures. Additionally, low polarisation detection is consis-
tent with the trends observed in other works, where polarimet-
ric observations of TDEs reported low observed polarisation in
the absence of a jet for ASASSN-18pg (Holoien et al. 2020),
AT2019qiz (Patra et al. 2022), AT2018dyb, AT2019azh, and
AT2019dsg (Leloudas et al. 2022), AT2022fpx (Koljonen et al.
2024), and AT2023clx (Koljonen et al. 2025). AT2020mot
stands out in this context, displaying remarkably high polarisa-
tion even before any subtraction of the host flux (Liodakis et al.
2023). After accounting for host contamination, Π remains
exceptionally large, despite the absence of any jet signature
(Liodakis et al. 2023) and the lack of a concurrent X-ray detec-
tion at peak luminosity. This behaviour contrasts with the major-
ity of our TDE sample, where polarisation remains low through-
out the flare.
Once host flux is removed, AT2022hvp and AT2022upj each
show a single measurement with Π > 6%, exceeding the pre-
dictions of electron scattering models (Leloudas et al. 2022).
Although these detections are less significant than those of
AT2020mot, they demonstrate that host corrections can unveil
higher polarisation states in other events.
The two polarimetric measurements of AT2022upj obtained
on the same night in the B and R bands hint at a strong Π wave-
length dependence. However, this is likely the result of contin-
uum depolarisation due to the prominent Hα line, which lies in
the R band (see Appendix B).
Several TDEs in our sample also show modest variabil-
ity in Π or Θ. For instance, AT2022hvp and AT2022fpx
(Koljonen et al. 2024), along with AT2023clx (Koljonen et al.
2025), exhibit changes in Π during the flare. Another intriguing
case is AT2024gre, which displays (within 3σ) nearly constant
values of both Π and Θ, albeit only over 9 days of observation.
Given its featureless spectrum, one might speculate that a stable
polarisation signature could be linked to geometric or optical-
depth effects in the reprocessing region.
As previously mentioned, it is possible that high-Π states
have simply been missed in most TDEs. Flares often last a year
or longer; AT2020mot, for example, persisted for ∼300 days, yet
A81, page 6 of 11
Floris, A., et al.: A&A, 703, A81 (2025)
Table 5. Comparison of global properties.
Parameter Median σ AT2020mot Agreement
(1) (2) (3) (4) (5)
T [K] 29 600 12 100 31 700 X
log Lbol [erg s−1] 43.77 0.61 44.10 X
log LX [erg s−1] 42.45 0.69 <42.69 X
trise [d] 56.4 40.9 71.5 X
z 0.049 0.036 0.07 X
Πmax [%] 4.1 2.3 25 X
log MBH,TDEMass [M] 6.18 0.52 6.45 X
Mstar,TDEMass [M] 0.67 0.62 0.97 X
log MBH,MOSFiT [M] 6.66 0.76 7.26 X
Mstar,MOSFiT [M] 0.57 5.11 0.63 X
b 0.96 0.39 0.91 X
logNH [cm−2] 18.29 1.14 20.80 X
Notes. (1) Parameter. (2) Median value across the sample (excluding AT2020mot). (3) Standard deviation of that parameter (excluding
AT2020mot). (4) Value of AT2020mot. (5) Whether AT2020mot matches the population within 1σ interval.
Liodakis et al. (2023) obtained only four polarimetric epochs.
Assuming that each epoch samples a single night, the coverage
is merely ∼1% of the flare, and the extreme polarisation appears
in just one night. Since other TDEs have similar fractional cover-
age, it is plausible that similarly strong, brief polarisation peaks
have gone undetected. The physical mechanism capable of pro-
ducing such extreme Π, however, remains unclear.
4.2. Shock versus reprocessing
When we compare AT2020mot with other TDEs (Table 5), most
of its global parameters – for example TBB, Lbol, and trise – fall
within the typical range of these parameters for the observed
TDEs. However, AT2020mot departs substantially from the
norm in two key aspects: it exhibits a much higher polarisation
degree than any other event in our sample, and it also appears to
have an unusually large NH in our MOSFiT fits.
Intriguingly, Newsome et al. (2024a) report a strong IR flare
associated with this event, which may point to a thick layer of
dust or gas in the vicinity of the disruption. Normally, one might
expect a substantial reprocessing layer to suppress polarisation,
as scattering in optically thick material tends to reduce net polar-
isation signals (Charalampopoulos et al. 2023). The high polar-
isation of AT2020mot thus appears incompatible with this sce-
nario unless other factors – such as tidal shocks or geometric
asymmetries – are at play. Another possibility is that certain
modelling assumptions (for instance, those in MOSFiT) do not
hold in this particular system, leading to an overestimated NH or
an erroneous description of the reprocessing environment.
It is important to note that high NH and high polarisation
can indeed coexist: in Seyfert 2 nuclei high polarisation (Π ∼
20–30%) has been observed in the X-rays, produced from the
back-scattering off material above the torus despite the high NH
that characterises these sources (Ursini et al. 2023; Marin et al.
2024). In the studied Seyfert 2 galaxies, however, no variability
in the polarisation angle was observed, contrary to AT2020mot.
The physical conditions in the TDE environment are indeed
expected to be very different from Seyfert 2 galaxies, and repro-
ducing similar polarisation behaviour would require a clumpy
medium, under specific physical conditions, orbiting the SMBH.
We caution that the elevated NH inferred for AT2020mot is
model-dependent and could arise from features of the light curve
used during the MOSFiT fit rather than the intrinsic properties of
the event. Independent diagnostics, such as the Balmer decre-
ment, could provide an additional estimate of NH. Although the
AT2020mot flare has now concluded and the available optical
spectra are not suitable for similar measurements due to the low
S/N, this test could possibly be applied to future TDEs via opti-
cal spectroscopic observations.
Taken together, the extreme optical polarisation, elevated
NH, and strong IR emission render AT2020mot an outlier among
TDEs. Although various models can accommodate parts of the
observed behaviour, no single mechanism fully explains the
data. Polarisation variability, however, significantly disfavours
reprocessing models in general, and the observed polarisation
behaviour favours shock models (Koljonen et al. 2025). Future
theoretical investigations and more extensive polarimetric and
multi-wavelength monitoring of sources that exhibit similar
properties will be essential to clarify whether AT2020mot rep-
resents an extreme end of typical TDE physics or if it is gov-
erned by additional processes that are not captured by standard
reprocessing or shock-powered models.
5. Conclusions
We have analysed 14 flares from 13 TDEs, presenting new
polarimetric measurements obtained with multiple ground-based
telescopes. We draw several key conclusions.
– AT2020mot exhibits an extraordinarily high polarisation
degree (Π ∼ 25%), significantly exceeding that of all other
TDEs in our sample, which mostly show Π < 6% or lie
below detection thresholds, a behaviour that is consistent
with observations from other works.
– The polarisation degree of AT2020mot evolves from low to
high and again to low values, suggesting that existing obser-
vations of other TDEs might have missed epochs of poten-
tially higher Π.
– Several TDEs display varying Π or Θ (e.g. AT2022hvp,
AT2022fpx, and AT2023clx), hinting at evolving geometries
or optical depths.
– Current polarimetric data cannot firmly distinguish between
shock-powered and reprocessing-dominated models,
although high polarisation and variability disfavour repro-
cessing models. Additional data, especially time-resolved
polarimetry near peak brightness, will be critical.
A81, page 7 of 11
Floris, A., et al.: A&A, 703, A81 (2025)
– Aside from its extraordinary polarisation and inferred
column density (the latter is a model-dependent result),
AT2020mot appears to be a typical optical TDE.
These findings call for polarimetric monitoring that begins
soon after discovery and continues throughout the flare’s evo-
lution. Upcoming surveys such as the Vera Rubin Observatory
(Ivezic´ et al. 2019) will enlarge the TDE sample by orders of
magnitude, enabling systematic polarimetric campaigns and pro-
viding the statistical leverage needed to map the diversity of
TDE geometries. We are confident that larger, homogeneous
datasets, combined with improved theoretical models, will ulti-
mately clarify the physical conditions that govern TDE emission
and its polarisation signature.
Acknowledgements. AF, IL, AP and BAG were funded by the European Union
ERC-2022-STG – BOOTES – 101076343. Views and opinions expressed are
however those of the author(s) only and do not necessarily reflect those of the
European Union or the European Research Council Executive Agency. Nei-
ther the European Union nor the granting authority can be held responsible
for them. KIIK has received funding from the European Research Council
(ERC) under the European Union’s Horizon 2020 research and innovation pro-
gramme (grant agreement No. 101002352, PI: M. Linares). The IAA-CSIC co-
authors acknowledge financial support from the Spanish “Ministerio de Cien-
cia e Innovación” (MCIN/AEI/10.13039/501100011033) through the Center of
Excellence Severo Ochoa award for the Instituto de Astrofísica de Andalucía-
CSIC (CEX2021-001131-S), and through grants PID2019-107847RB-C44 and
PID2022-139117NB-C44. PC acknowledges support via Research Council of
Finland (grant 340613). MADT acknowledges support from the EDUFI Fellow-
ship and the Johannes Andersen Student Programme at the Nordic Optical Tele-
scope. The data in this study include observations made with the Nordic Optical
Telescope, owned in collaboration by the University of Turku and Aarhus Uni-
versity, and operated jointly by Aarhus University, the University of Turku and
the University of Oslo, representing Denmark, Finland and Norway, the Uni-
versity of Iceland and Stockholm University at the Observatorio del Roque de
los Muchachos, La Palma, Spain, of the Instituto de Astrofisica de Canarias.
The data presented here were obtained in part with ALFOSC, which is pro-
vided by the Instituto de Astrofísica de Andalucía (IAA) under a joint agreement
with the University of Copenhagen and NOT. Some of the data are based on
observations collected at the Observatorio de Sierra Nevada; which is owned
and operated by the Instituto de Astrofísica de Andalucía (IAA-CSIC); and
at the Centro Astronómico Hispano en Andalucía (CAHA); which is oper-
ated jointly by Junta de Andalucía and Consejo Superior de Investigaciones
Científicas (IAA-CSIC). We acknowledge funding to support our NOT obser-
vations from the Finnish Centre for Astronomy with ESO (FINCA), Univer-
sity of Turku, Finland (Academy of Finland grant nr 306531). E.L. was sup-
ported by Academy of Finland projects 317636 and 320045. This research has
made use of data from the RoboPol programme, a collaboration between Cal-
tech, the University of Crete, IA-FORTH, IUCAA, the MPIfR, and the Nicolaus
Copernicus University, which was conducted at Skinakas Observatory in Crete,
Greece.
References
Andruchow, I., Cellone, S. A., & Romero, G. E. 2008, MNRAS, 388, 1766
Arcavi, I., Faris, S., Newsome, M., Sniegowska, M., & Trakhtenbrot, B. 2024,
TNS Classif. Rep., 2024-466, 1
Auchettl, K., Ramirez-Ruiz, E., & Guillochon, J. 2018, ApJ, 852, 37
Blagorodnova, N., Cenko, S. B., Kulkarni, S. R., et al. 2019, ApJ, 873, 92
Blanton, M. R., & Roweis, S. 2007, AJ, 133, 734
Charalampopoulos, P., Bulla, M., Bonnerot, C., & Leloudas, G. 2023, A&A, 670,
A150
Charalampopoulos, P., Kotak, R., Wevers, T., et al. 2024, A&A, 689, A350
Chornock, R., Berger, E., Gezari, S., et al. 2014, ApJ, 780, 44
Christy, C. T., Alexander, K. D., Franz, N., et al. 2024, TNSAN, 56, 1
Dai, L., McKinney, J. C., Roth, N., Ramirez-Ruiz, E., & Miller, M. C. 2018, ApJ,
859, L20
Escudero Pedrosa, J., Agudo, I., Morcuende, D., et al. 2024, AJ, 168, 84
Foreman-Mackey, D., Hogg, D. W., Lang, D., & Goodman, J. 2013, PASP, 125,
306
Fulton, M., Srivastav, S., Smartt, S. J., et al. 2022a, TNSAN, 106, 1
Fulton, M., Srivastav, S., Smith, K. W., et al. 2022b, TNS Classif. Rep., 2022-
3389, 1
Gezari, S., Basa, S., Martin, D. C., et al. 2008, ApJ, 676, 944
Gezari, S., Chornock, R., Rest, A., et al. 2012, Nature, 485, 217
Godson, B., Warwick, B., Pursiainen, M., et al. 2024, TNS Classif. Rep., 2024-
401, 1
Golay, W. W., Alexander, K. D., Berger, E., et al. 2024, TNSAN, 70, 1
Guillochon, J., Nicholl, M., Villar, V. A., et al. 2018, ApJS, 236, 6
Hammerstein, E., Gezari, S., van Velzen, S., et al. 2021, ApJ, 908, L20
Hammerstein, E., Chornock, R., Gezari, S., & Yao, Y. 2024, TNSAN, 37, 1
Hills, J. G. 1975, Nature, 254, 295
Hinkle, J. 2023, TNS Classif. Rep., 2023-1521, 1
Hinkle, J. T., Holoien, T. W. S., Auchettl, K., et al. 2021, MNRAS, 500, 1673
Hinkle, J. T., Auchettl, K., Hoogendam, W. B., et al. 2024, OJAp, submitted
[arXiv:2412.15326]
Holoien, T. W. S., Kochanek, C. S., Prieto, J. L., et al. 2016, MNRAS, 455, 2918
Holoien, T. W. S., Auchettl, K., Tucker, M. A., et al. 2020, ApJ, 898, 161
Hoogendam, W. B., Hinkle, J. T., Shappee, B. J., et al. 2024, MNRAS, 530, 4501
Hosseinzadeh, G., Hiramatsu, D., DeMarchi, L., et al. 2020, TNS Classif. Rep.,
2020-2478, 1
Huang, S., Jiang, N., Zhu, J., et al. 2024, ApJ, 964, L22
Ivezic´, Ž., Kahn, S. M., Tyson, J. A., et al. 2019, ApJ, 873, 111
Koljonen, K. I. I., Liodakis, I., Lindfors, E., et al. 2024, MNRAS, 532, 112
Koljonen, K. I. I., Nilsson, K., Liodakis, I., & Lindfors, E. 2025, MNRAS, 542,
2238
Komossa, S. 2015, J. High Energy Astrophys., 7, 148
Langis, D. A., Liodakis, I., Koljonen, K. I. I., et al. 2025, arXiv e-prints
[arXiv:2506.05476]
Leloudas, G., Dai, L., Arcavi, I., et al. 2019, ApJ, 887, 218
Leloudas, G., Bulla, M., Cikota, A., et al. 2022, Nat. Astron., 6, 1193
Lin, Z., Jiang, N., Wang, T., et al. 2024, ApJ, 971, L26
Liodakis, I., Koljonen, K. I. I., Blinov, D., et al. 2023, Science, 380, 656
Marin, F., Marinucci, A., Laurenti, M., et al. 2024, A&A, 689, A238
Metzger, B. D., & Stone, N. C. 2016, MNRAS, 461, 948
Mockler, B., Guillochon, J., & Ramirez-Ruiz, E. 2019, ApJ, 872, 151
Mummery, A., & van Velzen, S. 2025, MNRAS, 541, 429
Newsome, M., Dgany, Y., Arcavi, I., et al. 2022, TNS Classif. Rep., 2022-3231,
1
Newsome, M., Arcavi, I., Howell, D. A., et al. 2024a, ApJ, 961, 239
Newsome, M., Arcavi, I., Howell, D. A., et al. 2024b, ApJ, 977, 258
Nilsson, K., Lindfors, E., Takalo, L. O., et al. 2018, A&A, 620, A185
Onori, F., Nicholl, M., Ramsden, P., et al. 2025, MNRAS, 540, 498
Otero-Santos, J., Piirola, V., Escudero Pedrosa, J., et al. 2024, AJ, 167, 137
Patra, K. C., Lu, W., Brink, T. G., et al. 2022, MNRAS, 515, 138
Piran, T., Svirski, G., Krolik, J., Cheng, R. M., & Shiokawa, H. 2015, ApJ, 806,
164
Planck Collaboration VI. 2020, A&A, 641, A6
Ramaprakash, A. N., Rajarshi, C. V., Das, H. K., et al. 2019, MNRAS, 485, 2355
Rees, M. J. 1988, Nature, 333, 523
Roming, P. W. A., Kennedy, T. E., Mason, K. O., et al. 2005, Space Sci. Rev.,
120, 95
Ryu, T., Krolik, J., & Piran, T. 2020, ApJ, 904, 73
Schlafly, E. F., & Finkbeiner, D. P. 2011, ApJ, 737, 103
Serkowski, K., Mathewson, D. S., & Ford, V. L. 1975, ApJ, 196, 261
Sfaradi, I., Horesh, A., & Fender, R. 2022, TNSAN, 57, 1
Sfaradi, I., Horesh, A., Bright, J., et al. 2023, TNSAN, 51, 1
Shiokawa, H., Krolik, J. H., Cheng, R. M., Piran, T., & Noble, S. C. 2015, ApJ,
804, 85
Somalwar, J., Chornock, R., Ravi, V., et al. 2024, TNS Classif. Rep., 2024-1840,
1
Stanek, K. Z. 2022, TNSTR, 2022-1433, 1
Taguchi, K., Uno, K., Nagao, T., & Maeda, K. 2023, TNS Classif. Rep., 2023-
438, 1
Tonry, J. L., Stubbs, C. W., Lykke, K. R., et al. 2012, ApJ, 750, 99
Trakhtenbrot, B., Arcavi, I., Ricci, C., et al. 2019, Nat. Astron., 3, 242
Ursini, F., Marinucci, A., Matt, G., et al. 2023, MNRAS, 519, 50
van Velzen, S., Holoien, T. W. S., Onori, F., Hung, T., & Arcavi, I. 2020, Space
Sci. Rev., 216, 124
van Velzen, S., Gezari, S., Hammerstein, E., et al. 2021, ApJ, 908, 4
Wiersema, K., van der Horst, A. J., Levan, A. J., et al. 2012, MNRAS, 421, 1942
Wiersema, K., Higgins, A. B., Levan, A. J., et al. 2020, MNRAS, 491, 1771
Yao, Y., & Qin, Y. 2023, TNS Classif. Rep., 2023-2663, 1
Yao, Y., Gezari, S., Velzen, S. V., Hammerstein, E., & Somalwar, J. 2022, TNS
Classif. Rep., 2022-1156, 1
Yao, Y., Chornock, R., LeBaron, N., et al. 2024, TNSAN, 43, 1
A81, page 8 of 11
Floris, A., et al.: A&A, 703, A81 (2025)
Appendix A: New observations
In this section we present the results from the polarimetric observations of the TDEs employed in this work in Table A.1. In Table
A.2 we present the host-corrected polarisation observations, performed following the criteria described in Sect. 2.3.
Table A.1. New polarisation observations.
Name MJD ∆tpeak Π Θ q u Telescope Filter
[d] [d] [%] [◦] [%] [%]
(1) (2) (3) (4) (5) (6) (7) (8) (9)
AT2020afhd 60345.351 -6.18 < 1.8 ... 1.44 ± 0.60 −0.40 ± 0.60 OSN T90 R
60345.384 -6.15 < 1.8 ... −0.73 ± 0.61 1.08 ± 0.61 OSN T90 R
60347.384 -4.15 < 2.1 ... 1.56 ± 0.70 0.66 ± 0.70 OSN T90 R
60348.324 -3.21 < 1.8 ... −0.10 ± 0.61 −0.03 ± 0.72 OSN T90 R
60371.319 +19.79 < 2.1 ... 0.16 ± 0.70 0.78 ± 0.70 OSN T90 R
60384.333 +32.80 < 2.7 ... −0.05 ± 0.99 −0.40 ± 0.90 OSN T90 R
AT2020mot 59867.567 +809.70 < 0.75 ... 0.53 ± 0.24 0.59 ± 0.25 NOT R
59896.531 +838.66 < 0.7 ... −0.19 ± 0.29 −0.35 ± 0.30 NOT R
59929.465 +871.60 < 0.45 ... −0.17 ± 0.15 0.22 ± 0.15 NOT R
59936.472 +878.60 < 0.42 ... 0.04 ± 0.14 0.24 ± 0.14 NOT R
60180.649 +1122.78 < 0.44 ... −0.14 ± 0.27 −0.11 ± 0.26 NOT R
60207.492 +1149.62 < 0.33 ... −0.05 ± 0.11 0.34 ± 0.11 NOT V
60207.583 +1149.71 < 0.35 ... −0.20 ± 0.15 −0.03 ± 0.15 NOT R
AT2022dbl 59647.578 +11.30 < 0.51 ... −0.01 ± 0.20 0.29 ± 0.16 NOT R
AT2022dbl2 60357.720 +12.93 < 2.1 ... −0.09 ± 0.66 −0.01 ± 0.63 OSN T90 R
60385.660 +40.87 < 1.8 ... −0.22 ± 0.59 0.03 ± 0.60 OSN T90 R
60402.488 +57.70 < 1.8 ... −0.01 ± 0.65 −0.24 ± 0.63 OSN T90 R
60411.462 +66.67 < 2.1 ... 0.03 ± 0.71 0.74 ± 0.70 OSN T90 R
60426.308 +81.52 < 1.8 ... 1.10 ± 0.67 −0.70 ± 0.37 Skinakas 1.3m r
AT2022gri 59720.403 -39.12 0.65 ± 0.15 162.6 ± 6.8 0.55 ± 0.16 −0.38 ± 0.17 NOT B
AT2022hvp 59713.327 +21.22 4.39 ± 0.71 47.5 ± 4.2 −0.39 ± 0.64 4.37 ± 0.71 Skinakas 1.3m R
59738.336 +46.23 < 2.79 ... −0.22 ± 0.94 0.30 ± 0.93 Skinakas 1.3m R
AT2022upj 59909.480 +22.00 < 0.45 ... −0.26 ± 0.17 −0.07 ± 0.17 NOT R
59909.496 +22.02 2.92 ± 0.26 48.4 ± 2.5 −0.37 ± 0.27 3.11 ± 0.24 NOT B
AT2022wtn 59930.427 +56.12 < 0.72 ... −0.58 ± 0.28 0.40 ± 0.28 NOT B
59930.442 +56.13 < 0.48 ... −0.40 ± 0.18 −0.22 ± 0.18 NOT R
AT2023ugy 60240.324 +6.92 < 1.9 ... 0.82 ± 0.61 −0.60 ± 0.71 Skinakas 1.3m r
60309.348 +75.95 < 1.2 ... 0.82 ± 0.48 0.85 ± 0.48 NOT B
60309.362 +75.96 < 0.75 ... 0.24 ± 0.29 0.51 ± 0.29 NOT R
AT2023lli 60207.620 +37.93 0.62 ± 0.20 115.1 ± 10.9 −0.42 ± 0.23 −0.51 ± 0.23 NOT B
60207.635 +37.95 < 0.48 ... −0.03 ± 0.16 0.02 ± 0.16 NOT R
60208.490 +38.80 < 0.42 ... 0.26 ± 0.14 0.07 ± 0.13 Skinakas 1.3m r
60237.344 +67.65 0.50 ± 0.13 45.5 ± 10.3 −0.01 ± 0.18 0.50 ± 0.13 Skinakas 1.3m r
60250.469 +80.78 < 0.57 ... −0.19 ± 0.22 −0.44 ± 0.22 NOT R
AT2024bgz 60371.411 +21.84 < 5.4 ... 1.00 ± 1.85 −2.29 ± 1.81 OSN T90 R
60385.561 +35.99 < 8.1 ... 2.57 ± 2.69 −0.44 ± 2.47 OSN T90 R
60404.416 +54.85 < 7.8 ... 4.38 ± 2.60 −0.44 ± 2.47 OSN T90 R
AT2024gre 60470.416 +29.20 2.43 ± 0.37 75.3 ± 4.4 −2.12 ± 0.38 1.20 ± 0.36 NOT B
60470.431 +29.21 1.32 ± 0.29 83.1 ± 6.5 −1.31 ± 0.32 0.32 ± 0.32 NOT R
60478.404 +37.18 2.05 ± 0.63 64.7 ± 9.9 −1.38 ± 0.71 1.68 ± 0.74 NOT B
60478.419 +37.20 1.48 ± 0.42 84.1 ± 9.0 −1.53 ± 0.48 0.32 ± 0.49 NOT R
60479.381 +38.16 2.5 ± 0.6 79.7 ± 6.2 −2.34 ± 0.59 0.88 ± 0.55 CAHA 2.2m R
60479.426 +38.21 2.0 ± 0.6 65.3 ± 8.4 −1.30 ± 0.59 1.52 ± 0.59 CAHA 2.2m R
Notes. (1) TDE name. (2) MJD of the observation. (3) Time from peak. (4) Measured polarisation degree, with the associated uncertainty. (5)
Measured polarisation angle, with the associated uncertainty. The measured polarisation angle is only reported when the polarisation degree is
detected within 3σ. (6) Measured q Stokes parameter, with the associated uncertainty. (7) Measured u Stokes parameter, with the associated
uncertainty. (8) Telescope with which the observation was conducted. (9) Observing filter.
A81, page 9 of 11
Floris, A., et al.: A&A, 703, A81 (2025)
Table A.2. Host-corrected observations.
Name MJD ∆tpeak Πcorr Θ Telescope Filter
[d] [d] [%] [◦]
(1) (2) (3) (4) (5) (6) (7)
AT2022gri 59720.403 -39.12 1.29 ± 0.39 162.6 ± 6.8 NOT B
AT2022hvp 59713.327 +21.22 6.86 ± 1.14 47.5 ± 4.2 Skinakas 1.3m R
59738.336 +46.23 < 6.34 ... Skinakas 1.3m R
AT2022upj 59909.480 +22.00 < 2.34 ... NOT R
59909.496 +22.02 6.40 ± 1.20 48.4 ± 2.5 NOT B
AT2024gre 60470.416 +29.20 2.74 ± 0.43 75.3 ± 4.4 NOT B
60470.431 +29.21 1.97 ± 0.48 83.1 ± 6.5 NOT R
60478.404 +37.18 2.32 ± 0.72 64.7 ± 9.9 NOT B
60478.419 +37.20 2.39 ± 0.75 84.1 ± 9.0 NOT R
60479.381 +38.16 4.1 ± 1.1 79.7 ± 6.2 CAHA 2.2m R
60479.426 +38.21 3.2 ± 1.1 65.3 ± 8.4 CAHA 2.2m R
Notes. (1) TDE name. (2) MJD of the observation. (3) Time from peak. (4) Host-corrected polarisation degree, with the associated uncertainty.
(5) Measured polarisation angle, with the associated uncertainty. The measured polarisation angle is only reported when the polarisation degree is
detected within 3σ. (6) Telescope with which the observation was conducted. (7) Observing filter.
Appendix B: Sample description
In this section we briefly describe each source in our sample, summarising relevant findings from previous studies and providing
details on their spectral classification. We also discuss their X-ray behaviour and host galaxy classification, where available.
AT2020mot was first discovered on 14 June 2020, and identified as a TDE H+He due to the presence of the broad Heiiλ4686 line
together with Hβ in its spectrum (Hosseinzadeh et al. 2020). RoboPol observations obtained with the 1.30m telescope at Skinakas
observatory revealed its unique nature: it exhibits the highest measured polarisation degree (Π ∼ 25%), consistent across the V , R,
and I bands after subtracting the unpolarised host flux contribution (Liodakis et al. 2023). Polarimetric observations started ∼ 25
days after the optical peak and continued for nearly three years. Liodakis et al. (2023) also reports radio observations with the Very
Large Array (VLA) at 15 GHz, yielding a 27 µJy upper limit non-detection. Newsome et al. (2024a) report a strong i-band excess,
interpretable as emission from two concentric dust rings. AT2020mot does not exhibit X-ray emission across the entire duration of
the TDE flare. Its host, WISEA,J003113.52+850031.8 (at z = 0.07), is classified as an E+A galaxy (Hosseinzadeh et al. 2020), a
type commonly associated with TDEs (Hammerstein et al. 2021).
AT2020afhd was initially discovered on 20 October 2020, but exhibited an optical flare in 2024 that classifies it as a TDE H+He due
to its strong Balmer and Heii emission (Hammerstein et al. 2024) and a Bowen fluorescence flare (Arcavi et al. 2024). Polarimetry
covers from ∼ 6 days before to ∼ 30 days after peak. AT2020afhd exhibited X-ray emission at peak luminosity and the emission
persisted during the decay phase of the TDE flare. VLA observations at 15 GHz detected a flux density of 253 ± 18 µJy at the
transient coordinates (Christy et al. 2024). Its host, LEDA 145386 (at z = 0.027), is classified as a Seyfert 2 galaxy.
AT2022gri was discovered on 3 April 2022 and identified as a featureless TDE, showing a blue continuum superimposed on the
host-galaxy absorption lines at z = 0.028 (Yao et al. 2022). A single polarimetric epoch with ALFOSC@NOT was obtained ∼ 40
days before peak. No X-ray emission has been detected.
AT2022hvp was discovered on 19 April 2022 and classified as a TDE He due to the broad Heii line on top of a blue continuum
(Fulton et al. 2022a). Polarimetry spans from ∼ 20 to ∼ 45 days after peak luminosity. AT2022hvp did not exhibit X-ray emission
throughout the duration of the optical flare. Its host galaxy, SDSS J095445.24+552625.2 (at z = 0.12), appears to be a red, quiescent
system lacking prominent absorption features (Fulton et al. 2022a).
AT2022fpx was discovered on 31 March 2022 and classified as a TDE H+He, given the presence of Balmer and Heii lines in
its spectra. The source also shows multiple high-ionisation lines, classifying it as an ECLE (Koljonen et al. 2024). Polarimetric
observations started ∼ 10 days after peak luminosity, and continued until ∼ 390 days after peak luminosity. Its host galaxy (at
z = 0.0735) is an E+A galaxy with a ∼ 56% probability of hosting an active galactic nucleus (Koljonen et al. 2024).
AT2022upj was first discovered on 18 September 2022 and identified as a TDE He based on the presence of the broad Heii line
(Newsome et al. 2022). AT2022upj also exhibited the [Fexiv]λ5303 and [Fex]λ6375 high ionisation lines, which are characteristic
of the ECLE class of TDEs (Newsome et al. 2024b). Although the source is X-ray bright 350 days after maximum, the preceding
A81, page 10 of 11
Floris, A., et al.: A&A, 703, A81 (2025)
observation was 130 days earlier, so the onset time is uncertain. Newsome et al. (2024b) reports X-ray detections at the time of
the peak. However, the discrepancy likely arises from the more conservative S/N cut that we adopted for the events. Polarimetry
was obtained ∼ 20 days after peak luminosity. Its host galaxy, LEDA 924801, is found at z = 0.054. The source exhibits a clear
discrepancy between the observed polarisation in the B and R bands, hinting at the possibility of a wavelength dependence. It is
likely, however, that the difference is a result of a dilution of the continuum polarisation from the Hα line, which was quite prominent
at the time of the classification spectrum4. Since no contemporary spectra were taken at the time of our polarisation measurements,
a quantitative correction is not feasible.
AT2022wtn was discovered on 2 October 2022 and classified as a TDE H+He (Fulton et al. 2022b; Onori et al. 2025). Polarimetric
observations of the source were conducted ∼ 60 days after peak luminosity. AT2022wtn does not exhibit X-ray emission across
the entire duration of the flare. The host galaxy of AT2022wtn, SDSS J232323.79+104107.7 (at z = 0.049), is considered to be
likely hosting star-formation, and is currently merging with the more massive neighbouring galaxy SDSS J232323.37+104101.7
(Onori et al. 2025).
AT2022dbl was first detected on 22 February 2022 and identified as a TDE H+He based on its nuclear location and the presence
of broad Balmer lines as well as Heii (Stanek 2022). A second flare occurred at the end of 2023 (Yao et al. 2024), attributed to
the partial disruption of the same star, making it a partial TDE (Lin et al. 2024; Hinkle et al. 2024). Polarimetric observations were
conducted ∼ 7 days after peak brightness of the first flare, and restarted ∼ 15 days after the second flare peaked, until ∼ 80
days after peak brightness of the second TDE flare. No X-ray emission has been detected throughout the duration of the flare.
AT2022dbl was also observed with the VLA at 15,GHz, yielding a flux density of 32 ± 7µJy (Sfaradi et al. 2022). Its host, WISEA
J122045.05+493304.7 (at z = 0.0284), is classified as an E+A galaxy (Stanek 2022).
AT2023clx was first detected on 22 February 2023 and classified as a TDE H+He given its nuclear position and its broad Balmer
and Heii lines lines superimposed on a blue continuum (Taguchi et al. 2023; Hoogendam et al. 2024). Charalampopoulos et al.
(2024) carried out a detailed spectral analysis, identifying multiple components in the emission lines. Polarimetric observations
were conducted ∼ 5 days after peak luminosity, and continued until ∼ 35 days after peak luminosity. The source did not exhibit
X-ray emission throughout the duration of the optical flare. AT2023clx was also observed with the Arcminute Microkelvin Imager
- Large Array (AMI-LA) at a central frequency of 15.5 GHz, and a flux of 0.40 ± 0.08 mJy was detected, although the transient
nature of the detected signal was not confirmed (Sfaradi et al. 2023). The host galaxy, NGC 3799 (at z = 0.01107), is an SAB low-
ionisation nuclear emission-line region system that is interacting with the neighbouring galaxy NGC 3800 (Charalampopoulos et al.
2024).
AT2023ugy was discovered on 6 October 2023 and identified as a TDE H from its broad Hα emission atop a blue continuum
(Yao & Qin 2023). Polarimetric observations were conducted at peak luminosity, until ∼ 70 days after. X-ray emission was not
detected from the source throughout the TDE flare. The host galaxy of AT2023ugy, SDSS J214044.01+210058.4 is found at z =
0.106.
AT2023lli was found on 23 June 2023 and classified as a TDE H+He characterised by a very broad Hα line (full width at half
maximum of ∼ 15000 km s−1) and the Heii line (Hinkle 2023). Polarimetric observations were conducted ∼ 50 days after peak
luminosity, and ended ∼ 45 days later. X-ray emission from the source was not detected at the time of the peak but was detected
starting from ∼ 110 days after optical peak luminosity. Huang et al. (2024) report u−r = 2.08 mag for the host galaxy (at z = 0.036),
placing it in the ‘green valley’ typical of TDE hosts (Hammerstein et al. 2021), consistent with observed strong Balmer absorption
features indicative of a post-starburst galaxy. AT2023lli has a predicted maximum interstellar polarisation higher than the observed
polarisation, making it the only source in our sample with a significant Πmax (Πmax ∼ 0.9%). However, the observed variation in Θ
(see Table A.1) points to an intrinsic origin to the TDE.
AT2024bgz was first discovered on 1 February 2024 and identified as a TDE H+He (Godson et al. 2024). VLA observations at 15
GHz did not detect any emission at the transient coordinates (Golay et al. 2024). Polarimetric observations started approximately at
peak luminosity, and ended ∼ 70 days after peak luminosity. X-ray emission was not detected from the source throughout the TDE
flare. The host galaxy of AT 2024bgz, 2MASX J09440480-0412051, is found at z = 0.0585.
AT2024gre was discovered on 16 April 2024 and identified as a featureless TDE, showing a blue continuum with no promi-
nent emission lines (Somalwar et al. 2024). Polarimetric observations started ∼ 20 days after peak luminosity, and ended 9
days later. X-ray emission was not detected from the source throughout the TDE flare. The host galaxy of AT 2024gre, SDSS
J103138.88+345430.0, is found at z = 0.12.
4 https://www.wis-tns.org/object/2022upj
A81, page 11 of 11