Taggart, K and Perley, DA
Core-collapse, superluminous, and gamma-ray burst supernova host galaxy
populations at low redshift: the importance of dwarf and starbursting galaxies
http://researchonline.ljmu.ac.uk/id/eprint/14945/
Article
LJMU has developed LJMU Research Online for users to access the research output of the
University more effectively. Copyright © and Moral Rights for the papers on this site are retained by
the individual authors and/or other copyright owners. Users may download and/or print one copy of
any article(s) in LJMU Research Online to facilitate their private study or for non-commercial research.
You may not engage in further distribution of the material or use it for any profit-making activities or
any commercial gain.
The version presented here may differ from the published version or from the version of the record.
Please see the repository URL above for details on accessing the published version and note that
access may require a subscription.
For more information please contact researchonline@ljmu.ac.uk
http://researchonline.ljmu.ac.uk/
Citation (please note it is advisable to refer to the publisher’s version if you
intend to cite from this work)
Taggart, K and Perley, DA (2021) Core-collapse, superluminous, and
gamma-ray burst supernova host galaxy populations at low redshift: the
importance of dwarf and starbursting galaxies. Monthly Notices of the
Royal Astronomical Society, 503 (3). pp. 3931-3952. ISSN 0035-8711
LJMU Research Online
MNRAS 503, 3931–3952 (2021) doi:10.1093/mnras/stab174
Advance Access publication 2021 March 02
Core-collapse, superluminous, and gamma-ray burst supernova host
galaxy populations at low redshift: the importance of dwarf and
starbursting galaxies
K. Taggart
1,2‹
andD.A.Perley
1
1
Astrophysics Research Institute, Liverpool John Moores University, 146 Brownlow Hill, Liverpool L3 5RF, UK
2
Department of Astronomy and Astrophysics, University of California, Santa Cruz, CA 95064, USA
Accepted 2021 January 18. Received 2021 January 18; in original form 2020 July 4
ABSTRACT
We present a comprehensive study of an unbiased sample of 150 nearby (median redshift, z = 0.014) core-collapse supernova
(CCSN) host galaxies drawn from the All-Sky Automated Survey for Supernovae (ASAS-SN) for direct comparison to the
nearest long-duration gamma-ray burst (LGRB) and superluminous supernova (SLSN) hosts. We use public imaging surveys to
gather multiwavelength photometry for all CCSN host galaxies and fit their spectral energy distributions (SEDs) to derive stellar
masses and integrated star formation rates (SFRs). CCSNe populate galaxies across a wide range of stellar masses, from blue
and compact dwarf galaxies to large spiral galaxies. We find 33
+4
−4
per cent of CCSNe are in dwarf galaxies (M
∗
< 10
9
M
)
and 2
+2
−1
per cent are in dwarf starburst galaxies [specific star formation rate (sSFR) > 10
−8
yr
−1
]. We reanalyse low-redshift
SLSN and LGRB hosts from the literature (out to z<0.3) in a homogeneous way and compare against the CCSN host sample.
The relative SLSN to CCSN supernova rate is increased in low-mass galaxies and at high sSFRs. These parameters are strongly
covariant and we cannot break the degeneracy between them with our current sample, although there is some evidence that both
factors may play a role. Larger unbiased samples of CCSNe from projects such as ZTF and LSST will be needed to determine
whether host-galaxy mass (a proxy for metallicity) or sSFR (a proxy for star formation intensity and potential IMF variation) is
more fundamental in driving the preference for SLSNe and LGRBs in unusual galaxy environments.
Key words: transients: supernovae – transients: gamma-ray bursts – galaxies: dwarf – galaxies: photometry – galaxies: star
formation.
1 INTRODUCTION
Massive stars (>8M
) evolve rapidly, and after a short life (up to a
fewFor image calibration, we used catalogues of stars (PS1 tens of
million years), they die in violent core-collapse supernova (CCSN)
explosions. CCSNe have a profound influence on their environment:
they produce heavy elements and deposit large amounts of energy
into their environments, driving feedback and chemical evolution
in galaxies (e.g. Chevalier 1977). In addition, because of the short
progenitor lifetime, the volumetric CCSN rate is a direct tracer of
star formation. Thus, CCSNe can be used to quantify the contribution
to cosmic star formation from distinct galaxy sub-classes and to
pinpoint rare individual star-forming galaxies, especially at low
stellar mass, where galaxy catalogues are incomplete (e.g. Sedgwick
et al. 2019).
Candidate CCSN progenitors are diverse, as are the explosion
properties they produce. Observations of CCSN explosions and their
progenitors provide a means to test theories of stellar evolution and
the explosion channels of very massive stars. However, despite the
importance of CCSNe to many areas of astrophysics, mapping a star’s
evolution (accounting for complicating factors such as metallicity,
binarity, and rotation) from its beginning to end is acomplex problem.
E-mail: k.taggart@ucsc.edu
Observationally, CCSNe are classified into types I and II based on
the presence (II) or absence (I) of hydrogen emission lines in their
spectra at maximum light (Filippenko 1997). Some CCSN progen-
itors lose part/all of their hydrogen stellar envelope prior to their
explosion due to stellar winds (Maeder & Meynet 2000) or binary
mass transfer (Podsiadlowski, Joss & Hsu 1992) and are observed
as a helium-rich (Ib and IIb) or helium-poor (Ic) stripped-envelope
SNe (Smartt 2009). In recent years, due to a new generation of
all-sky surveys and ever-increasing observational capabilities, many
new types of stellar explosion have emerged beyond this classical
picture. One example is the class of superluminous supernovae
(SLSNe) which are also classified into types I and II, but whose
extreme luminosities exceed ordinary CCSNe by a factor of 10–
100 (Quimby et al. 2011; Gal-Yam 2012; see Moriya, Sorokina &
Chevalier 2018; Gal-Yam 2019 for more recent reviews) and likely
require an additional power source.
SLSNe-II are most likely powered by SN interaction with a
dense circumstellar shell of hydrogen created by an ultra-massive
progenitor star before the explosion (Chevalier & Irwin 2011;
Ginzburg & Balberg 2012;Moriyaetal.2013) or episodic mass-
loss in a pulsational pair-instability explosion (PPISNe; Woosley,
Blinnikov & Heger 2007; Chatzopoulos & Wheeler 2012). However,
the mechanism that powers SLSN-I is still puzzling. In theory, an
extremely massive stellar core (Moriya et al. 2010; Young et al.
2010) could produce enough
56
Ni to power an SLSN via radioactive
C
2021 The Author(s)
Published by Oxford University Press on behalf of Royal Astronomical Society
Downloaded from https://academic.oup.com/mnras/article/503/3/3931/6156623 by guest on 05 May 2021
3932 K. Taggart and D. A. Perley
decay, but mass-loss during a star’s lifetime makes it difficult to
retain such a massive core. Several other theoretical mechanisms
have been proposed to explain SLSN-I, including interaction with
non-hydrogen circumstellar-material (Chatzopoulos & Wheeler
2012; Sorokina et al. 2016; Vreeswijk et al. 2017), a Pair-Instability
SN (PISN; Barkat, Rakavy & Sack 1967; Rakavy & Shaviv 1967)
from a very massive and metal-poor star (∼0.2 Z
; Yusof et al.
2013) or an engine-driven scenario (similar to that invoked for
long-duration gamma-ray bursts) which would provide a long-lived
energy source behind the SN ejecta (e.g. Kasen & Bildsten 2010;
Metzger et al. 2015).
Long-duration gamma-ray bursts (LGRBs) are brief, but ex-
tremely luminous flashes of high-energy radiation associated with
the formation of a relativistic jet from a ‘central engine’ (a fast-
spinning neutron star or black hole) at the centre of a collapsing
and rapidly rotating massive stellar core. While most LGRBs occur
at high redshifts, events that occur sufficiently nearby are typically
observed in association with CCSNe (Galama et al. 1998;Hjorthetal.
2003; Woosley & Bloom 2006); these SNe are universally luminous,
helium-poor stripped-envelope SNe with broad spectral features (Ic-
BL) indicating large ejecta velocities (Cano et al. 2017b).
However, despite this association, the nature of LGRB progenitors
is uncertain, including whether the progenitor is a single star (Yoon,
Langer & Norman 2006) or a binary system (Cantiello et al. 2007)
and it is not yet firmly established whether all LGRBs occur in
association with SN Ic-BL, and vice versa. Two LGRBs from
2006 have no reported SN association to deep limits (Della Valle
et al. 2006; Fynbo et al. 2006; Gal-Yam et al. 2006;Gehrels
et al. 2006), although it has been suggested that some SN-less
LGRBs are not associated with the death of massive stars, but
may be compact binary mergers with unusually long duration
(e.g. Ofek et al. 2007; Kann et al. 2011). In addition, most
known SN Ic-BL are found in optical surveys with no observed
association with a LGRB. Some of these may represent LGRBs
observed off-axis, but they could also represent events in which
the jet fails to break out of the star or is not produced to begin
with.
The physical powering mechanisms and progenitors of SLSNe
and LGRBs are still under debate. However, it is highly unlikely that
pre-explosion imaging will ever uncover the progenitor properties
of SLSN or LGRBs due to a combination of their low volumetric
rate (∼1 in 1000 CCSNe; Quimby et al. 2013; Prajs et al. 2017)
and their high-redshift nature: the closest SLSN discovered to date
is at a distance of ∼110 Mpc (SN 2018bsz; Anderson et al. 2018)
and the closest LGRB-SN is at ∼40 Mpc (SN 1998bw; Galama et al.
1998). This motivates the use of indirect methods to probe SLSN and
LGRB progenitor properties and to constrain their poorly understood
explosion mechanisms. One method is to analyse the properties of
the galaxies they inhabit, to search for trends in morphology, colour,
chemical composition, and star formation, which can be tied to the SN
progenitor models themselves. For example, a PISN likely requires
a low-metallicity, star-forming environment to produce a star with
sufficient initial mass and to avoid losing its mass in line-driven
winds. Single-star progenitor mechanisms for central-engine models
of LGRBs also likely require a low metallicity, since line-driven
winds would otherwise quickly sap the progenitor of its rotational
energy. More exotically, some models postulate that LGRBs and/or
SLSNe may arise as the result of runaway collisions in young and
dense star clusters (van den Heuvel & Portegies Zwart 2013). In
this scenario, one may expect to find SLSNe more frequently in
galaxies undergoing an exceptionally high rate of star formation,
even after accounting for the fact that any CCSN is proportionally
more likely to occur in a galaxy with a high star formation rate
(SFR).
There is ample evidence that LGRB and SLSN-I host galaxies
differ from the bulk of the star-forming galaxy population. For
example, both LGRBs and SLSNe-I seem to occur preferentially in
faint, low-mass galaxies with irregular structure (Fruchter et al. 2006;
Neill et al. 2011; Lunnan et al. 2014; Angus et al. 2016
). Japelj et al.
(2016b) found the B-band luminosity, stellar mass, SFR and sSFR
of SLSNe-I and LGRBs are statistically similar between a redshift
range of 0.3 <z<0.7; and Lunnan et al. (2014) found that SLSN-I
host galaxies at 0.1 <z<1.6 (discovered in the PS medium deep
survey) are statistically indistinguishable from LGRB host galaxies.
There is also good evidence in particular that metallicity affects
SLSN and LGRB production: high-metallicity environments rarely
produce LGRBs (Kr
¨
uhler et al. 2015; Vergani et al. 2015;Japeljetal.
2016a;Perleyetal.2016b;Palmerioetal.2019)orSLSNe(Perley
et al. 2016a;Chenetal.2017a; Schulze et al. 2018).
1
However, population studies with larger sample sizes show that
there may also be some subtle differences between the SLSNe and
LGRBs populations themselves. For example, the median half-light
radius of LGRB host galaxies is ∼1700 pc (Lyman et al. 2017),
and for SLSNe it is ∼900 pc (Lunnan et al. 2015). In addition,
Lunnan et al. (2014) bolstered the PS medium deep survey SLSNe-I
with SLSNe-I from the literature (typically at lower redshift) and
found that SLSNe-I were statistically distinct from LGRBs, with a
fainter B-band luminosity and lower stellar mass. Leloudas et al.
(2015) suggested that on average, SLSNe-I explode in lower mass
and higher sSFR than the hosts of LGRBs (0.1 <z<1.6). These
findings were further supported by Schulze et al. (2018), who used
the largest sample size of LGRBs and SLSNe (in comparison to
previous studies) and found that the B-band luminosity, stellar mass,
and sSFR of SLSNe-I and LGRBs are statistically distinct over a
redshift range of 0.3 <z<1.
In contrast, CCSNe have typically been found in massive spiral
galaxies. In part, this was a reflection of the fact that CCSN samples
(prior to untargeted all-sky surveys) were found via targeted surveys
of pre-selected nearby galaxies. Therefore, CCSNe were always
found in massive, nearby galaxies (most of which were massive
spirals), but about half of high-redshift (0.28 <z<1.2) CCSNe
found blindly in deep surveys (covering small) fields of view also
explode in spiral galaxies (Svensson et al. 2010), in contrast to only
∼10 per cent of LGRB hosts.
Graur et al. (2017a, b) found that the relative rate of Ib/c
stripped-envelope SNe versus non-stripped CCSNe declines in low-
mass (<10
10
M
) galaxies; they are underrepresented by a factor
of ∼3. In addition, Graur et al. (2017a, b) also note that there
appears to be a strong metallicity bias, with the relative rate of
Ib/c to II SNe increasing with metallicity. However, this is not
interpreted as evidence for the single-star scenario: the single-star
stellar evolution models underpredict the observed absolute numbers
of SE–SN; therefore, the binary scenario could be important and
there could be multiple channels at play. In addition, the binary
scenario can also show a strong metallicity dependence, although
binary star channels are much more uncertain than the single-star
channel.
1
However, this is not the entire picture since over the past few years, as the
statistical sizes of nearby SLSN and LGRBs have increased, there have been
a handful of cases of large spiral galaxies with high-metallicities hosting
SLSN-I (MLS121104, PTF10uhf, SN 2017egm, Lunnan et al. 2014; Perley
et al. 2016c; Dong et al. 2017) and nearby LGRBs (e.g. Izzo et al. 2019).
MNRAS 503, 3931–3952 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/3/3931/6156623 by guest on 05 May 2021
CCSN host galaxies from ASAS-SN 3933
Nevertheless, there is some disagreement in the literature; Arcavi
et al. (2010) found that while the relative proportion of Ic SNe versus
non-stripped CCSNe decreased in low-mass galaxies, the relative
rates of all other stripped-envelope SNe (Ic-BL, Ib, IIb) versus non-
stripped CCSNe increased in low-mass, low-metallicity galaxies,
which may be a result of a reduced metallicity-driven mass-loss
causing some massive stars that would have exploded as a Ic SN
in a metal-rich galaxy to retain some H and He and explode as a
Ib/IIb event instead. There are also differences in the environments
of stripped-envelope CCSNe themselves. Ordinary Ic CCSNe are
found in more metal-rich galaxies with lower sSFRs than their
more energetic Ic-BL cousins (with and without LGRB associations)
that may suggest that Ic-BL harbour LGRB jets from a compact
central engine, which in turn requires a low-metallicity environment,
whereas ordinary Ic SNe may not require such an environment (Japelj
et al. 2018; Modjaz et al. 2020).
Additionally, there are also some indications that metallicity alone
may not fully explain the unusual properties of the host galaxies of
SLSNe and LGRBs. In particular, many SLSN-I hosts show very
high specific star formation rates (sSFR = SFR/M
∗
)aswellas
low metallicities, evidenced by their very high equivalent widths
(Leloudas et al. 2015): as many as ∼50 per cent of SLSNe-I are
found in extreme emission-line galaxies (EELGs; Leloudas et al.
2015). While sometimes attributed to a very young progenitor that
simply explodes earlier than other types of SNe (Leloudas et al. 2015;
Th
¨
one et al. 2015), it could also point towards an intrinsic preference
in starbursting galaxies that favours the production of SLSNe, such as
a top-heavy IMF (e.g. Dabringhausen, Kroupa & Baumgardt 2009)
or the collisional model of van den Heuvel & Portegies Zwart (2013).
A complicating factor is that all key galaxy observational parame-
ters we may want to use to diagnose the nature of the progenitor (e.g.
stellar mass, metallicity and sSFR) correlate across the star-forming
galaxy population (e.g. Tremonti et al. 2004; Salim et al. 2007). For
example, a low-mass and low-metallicity galaxy tends to have a star
formation history with short bursts of concentrated star formation and
therefore is more likely to be observed as a starburst than a high-mass
and high-metallicity galaxy. Thus, it is still unclear to what extent
the environmental properties of SLSNe and LGRBs (low-mass, low-
metallicity, and high sSFR) reflect their specific physical influences
(progenitor and explosion mechanism).
In order to disentangle the role of metallicity and SFR and to
determine if both properties are equally important in governing
SLSN and LGRB production, we need an unbiased and representative
sample of star-forming galaxies to provide testable predictions for
where we might expect SLSNe and LGRBs to occur under various
hypotheses about their formation preferences. Ideally, the sample
of star-forming galaxies should be selected in the same manner
as an SLSN or an LGRB – via the explosion of a massive star
as detected in a time-domain imaging survey – to minimize the
large methodological differences between selecting via SNe versus
selecting via galaxy counts in flux-limited surveys. In other words,
we require a high-quality sample of ‘ordinary’ CCSNe.
This sample must have several properties. First, it must enclose
a sufficiently large volume to be representative of the average dis-
tribution of galaxies, since large-scale structure can potentially bias
the galaxy population seen within smaller volumes. Secondly, the
SNe must be discovered in an unbiased way (not via galaxy-targeted
surveys). Thirdly, the sample must be able to securely distinguish
CCSNe from Ia SNe for all transients, ideally via spectroscopy.
Finally, it must have multiwavelength galaxy data from UV to NIR
in order to derive physical parameters for the hosts. Few existing
SN samples have these properties, and until recently, none of these
samples havebeen at low redshift where detailed host studies are most
practical. Examples of other large, untargeted SN samples include
SDSS (Frieman et al. 2008;Sakoetal.2008) and SNLS (Bazin et al.
2009) but these surveys are not spectroscopically complete, and this
leads to ambiguities in the classifications.
In this paper, we address this need by compiling a large, unbiased,
representative sample of CCSN host galaxies (which we assume
sample the explosions of ‘typical’ massive stars, unlike SLSNe and
LGRBs). We provide photometry of this sample with integrated UV-
through-NIR SEDs and stellar masses and SFRs derived from these
measurements. We investigate star formation within the CCSN host
galaxy sample and compare to a sample of SLSN and LGRBs.
The paper is organized as follows. Section 2 describes how the
transient host galaxies are selected to form our CCSN, SLSN, and
LGRB samples. In Section 3, we describe our photometry method
and show all other archival photometry that has been used in this
paper. In Section 4, we present the methodology used to measure the
SFRs and stellar masses of each host galaxy based on UV through
NIR colours. In Section 5, we show our results, and in Section 6,
we summarize our findings and present our conclusions. Throughout
this paper, we adopt CDM cosmology, with
0
= 0.27,
= 0.73,
and H
0
= 70 km s
−1
Mpc
−1
(Komatsu et al. 2011).
2 HOST GALAXY SAMPLES
2.1 Core collapse supernovae
A variety of galaxy-untargeted SN catalogues exist, including the
Dark Energy Survey (Flaugher 2005), Catalina Real-Time Survey
(Drake et al. 2009), the Palomar Transient Factory (Law et al.
2009), SuperNova Legacy Survey (Bazin et al. 2009), Pan-STARRS
(Kaiser et al. 2002), La Silla Quest (Hadjiyska et al. 2012), the
Gaia transient survey (Hodgkin et al. 2013), SkyMapper (Keller
et al. 2007), SDSS Supernova Survey (Frieman et al. 2008), and
the All-Sky Automated Survey for Supernovae (ASAS-SN; Shappee
et al. 2014). We drew our CCSN sample from ASAS-SN, since it
is shallow (m
V,limit
∼17 mag) but is all-sky, so the SNe it finds
are bright and generally very nearby. This means that excellent
photometric galaxy information exists in public catalogues and that
almost all SNe are bright enough (even with small telescopes) for
the global SN community to follow-up, spectroscopically classify
and derive a redshift estimate. Therefore, the ASAS-SN sample is
spectroscopically complete for SNe with peak V-band light-curve
magnitudes m
V
< 15.8 and is roughly 50 per cent complete at m
V
=
∼ 17 (Holoien et al. 2017a). This was important since we required an
unambiguous sample of CCSN selected host galaxies and a reliable
SN redshift estimate for our host analysis.
We compiled all spectroscopically confirmed CCSNe discovered
by ASAS-SN (2013–2014, 2015, 2016, 2017; Holoien et al. 2017a,
b, c, 2019), and adopted any SN classifications and redshift estimates
that were updated since the initial classification was made. We also
included any SNe that were not discovered by ASAS-SN, but were
‘recovered’ in their data and therefore do not have an ASAS-SN
name designation. We refer to these SNe in the paper text using the
designated IAU name, or the discovery group name (6 SNe) when
there is no IAU name to our knowledge. For the sake of brevity, we
shortened any possible supernova (PSN) object names to the first
eight digits.
There were some ambiguous classifications that we removed from
the sample. We removed two claimed SLSNe: ASAS-SN 15lh was
classified as an SLSN-I (Dong et al. 2016), but was omitted since
it was unclear whether this event was an SLSN or a tidal disruption
MNRAS 503, 3931–3952 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/3/3931/6156623 by guest on 05 May 2021
3934 K. Taggart and D. A. Perley
event (Leloudas et al. 2016; Margutti et al. 2017) and ASAS-SN 17jz
was re-classified as an SLSN-II, but its classification is ambiguous; it
could be a very luminous SN-II (Xhakaj et al. 2017) or alternatively
itcouldbeanAGN(Arcavietal.2017). In addition, we removed
SN 2015bh since the classification was ambiguous. Despite having
a data set spanning a 21-yr time period, it was unclear whether SN
2015bh was the terminal explosion of the star resulting in a CCSN
or if it was a precursor LBV hyper-eruption (Elias-Rosa et al. 2016;
Th
¨
one et al. 2017).
We limited our sample to a declination greater than −30
◦
because
uniform, public, deep optical survey data is not available across
the entire Southern hemisphere. Two supernovae (SN 2016afa and
2017ivu) have the same host (NGC 5962) and this galaxy is included
twice in the host galaxy analysis. We also imposed a galactic latitude
cut (|b|> 15
◦
) in order to eliminate the galaxies where stellar crowd-
ing significantly affects the photometry and thus remove SN 2015an,
2015W, 2016bpq, 2016G, 2017eaw, 2017gpn, ASAS-SN 17ny, 17kr,
and PSNJ1828 from the sample. In addition, we imposed a minimum
distance cut out to 10 Mpc, meaning that one supernova (AT 2014ge)
was removed from our sample. Primarily, we made this cut since
performing consistent photometry for very extended galaxies within
this volume using the same methods used for more distant galaxies
is difficult. Also, making this cut avoided sample overlap with the
comparison sample used in this analysis [the Local Volume Legacy
(LVL) survey] which is volume-complete to within ∼11 Mpc.
Our sample is comprised of 150 SNe discovered from 2013 to the
end of 2017. The redshift distribution covers the range of 0.00198–
0.08, with a median value of 0.014. Table1 details the division of
transient types within our samples and a table with details of these
host galaxies can be found in Appendix B1. A mosaic showing our
ASAS-SN CCSN host galaxy sample is provided in Figs 1 and 2.
We used methods detailed by Lupton et al. (2004) to convert PS1 gri
images into a colour composite image. Each cutout has a constant
physical size scale in the rest frame of the SN host of 21 kpc on each
side and a scale bar showing an angular size of 10 arcsec is shown
on each cutout.
2.2 Superluminous supernovae
We collated our initial SLSN sample based on archival SLSNe in
the literature. We included SLSN hosts from Neill et al. (2011),
2
SUSHIES (Schulze et al. 2018), and PTF (Perley et al. 2016c). In
addition, we included five candidates identified by Quimby et al.
(2018) following their reanalysis of archival PTF spectra: two likely
SLSNe-I (PTF12gty and PTF12hni) and three possible SLSNe-I
(PTF09q, PTF10gvb, and PTF11mnb) at slightly lower luminosities
(M > −21 mag) than the PTF sample of SLSN host galaxies by
Perley et al. (2016c). These SLSN candidates and their properties are
summarized in Table 2. Rest frame g-band magnitudes for PTF12gty
and PTF12hni were taken from De Cia et al. (2018) and PTF09q,
PTF10gvb, and PTF11mnb were taken from Quimby et al. (2018).
Thumbnail images of each host are shown in the bottom row of Fig. 1;
the physical scale is the same as for the CCSN hosts, with a yellow
scale bar of 2 arcsec.
We restricted our analysis to SLSNe with a redshift of z<0.3
for two main reasons. First, including distant SLSNe could have
caused incompleteness in the sample due to the increased difficulty
2
We did not include SN1995av and SN1997cy since their classifications are
unclear: SN1997cy could be a SN Ia or IIn and SN1995av may have been
associated with a LGRB.
in spectroscopically confirming members of this class without a
bright associated host galaxy. Secondly, we wanted to reduce cosmic
evolution effects when comparing to the z ∼ 0.014 CCSN sample.
After making this cut and excluding PTF09q, PTF10gvb, and
PTF11mnb, our final statistical sample consisted of 29 SLSNe-I
and 21 SLSNe-II in total.
2.3 LGRBs
Our LGRB sample consists of all z<0.3 LGRBs discovered prior to
the end of 2017 with an associated optical counterpart: a supernova,
an optical afterglow, or both. The requirement for an optical afterglow
was imposed to better match the optical selection of SNe used for
comparison and to ensure a high degree of confidence in the host-
galaxy association: while many additional low-z LGRBs have been
reported based on X-ray associations alone, it is difficult to rule out
the possibility that these are higher-z events seen in coincidence with
a foreground galaxy. This sample was comprised of 17 LGRBs; 12
of which had confirmed SN associations and 5 without any reported
SN (see Table 3).
Of the five LGRBs without reported SNe, two were highly
publicized events from 2006 (LGRBs 060505 and 060614) for which
an SN was ruled out to deep limits (Fynbo et al. 2006; Gal-Yam et al.
2006; Della Valle et al. 2006; Gehrels et al. 2006). These appear
to have genuinely different progenitors (such as compact binary
mergers) and/or explosion mechanisms from ordinary SN-associated
long-duration GRBs, a possibility that makes scrutiny of their
host properties particularly relevant. The remaining events, LGRBs
050826, 080517, and 111225A, have relatively poor constraints on
the extinction column towards the LGRB and/or on the presence of
an SN peaking 1–3 weeks after the event (e.g. Stanway et al. 2015).
3 PHOTOMETRY
3.1 CCSN host multiwavelength data
The galaxies in our CCSN sample are nearby (z<0.08), so most
were detectable in all-sky multiwavelength surveys. Therefore, our
primary image and source catalogues were public surveys. We used
images from the Galaxy Evolution Explorer (GALEX; Martin et al.
2005), the Panoramic Survey Telescope and Rapid Response System
(PS1; Kaiser et al. 2010), the Sloan Digital Sky Survey (SDSS; York
et al. 2000),andtheTwoMicronAll-SkySurvey(2MASS;Huchra
et al. 2012).
Our aim was to derive consistent mass and star formation estimates
for our host galaxy sample, thus we matched aperture sizes across
the optical and NIR wavelengths. This was particularly important
for nearby and massive galaxies, since the aperture size can signif-
icantly increase or decrease the flux measurements. In addition, the
automated pipeline of GALEX, 2MASS, and WISE often incorrectly
deblends galaxies with a large angular diameter on the sky and does
not capture the lowsurface-brightness parts of the galaxy. If available,
we used SDSS ugriz and GALEX FUV and NUV photometry from the
NASA Sloan Atlas (NSA; Blanton et al. 2011). The NSA is a unified
catalogue of galaxies out to z ∼ 0.05, optimized for nearby extended
objects since the flux measurements are derived from reprocessed
SDSS images with a better background subtraction (Blanton et al.
2011). We used the elliptical petrosian aperture photometry, with an
elliptical aperture radius defined by the shape of the light profile of
the galaxy as in Blanton et al. (2011) and Yasuda et al. (2001). The
NSA flux measurements were available for about half of the Northern
hemisphere sample. Otherwise, we performed the photometry using
MNRAS 503, 3931–3952 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/3/3931/6156623 by guest on 05 May 2021
