Constraining Type Ia supernovae through their heights in edge-on galaxies Lilit V . Barkhudaryan

2025-09-29 0 0 476.99KB 7 页 10玖币

MNRAS 000,1–6(2022) Preprint 14 December 2022 Compiled using MNRAS L

X style ﬁle v3.0

Constraining Type Ia supernovae through their heights in edge-on galaxies

Lilit V. Barkhudaryan?

Center for Cosmology and Astrophysics, Alikhanian National Science Laboratory, 2 Alikhanian Brothers Str., 0036 Yerevan, Armenia

Accepted 2022 December 13. Received 2022 November 22; in original form 2022 September 2

ABSTRACT

In this Letter, using classiﬁed 197 supernovae (SNe) Ia, we perform an analyses of their height distributions from the disc in

edge-on spirals and investigate their light-curve (LC) decline rates (∆m15). We demonstrate, for the ﬁrst time, that 91T- and

91bg-like subclasses of SNe Ia are distributed differently toward the plane of their host disc. The average height from the disc

and its comparison with scales of thin/thick disc components gives a possibility to roughly estimate the SNe Ia progenitor

ages: 91T-like events, being at the smallest heights, originate from relatively younger progenitors with ages of about several

100 Myr, 91bg-like SNe, having the highest distribution, arise from progenitors with signiﬁcantly older ages ∼10 Gyr, and

normal SNe Ia, which distributed between those of the two others, are from progenitors of about one up to ∼10 Gyr. We ﬁnd a

correlation between LC decline rates and SN Ia heights, which is explained by the vertical age gradient of stellar population in

discs and a sub-Chandrasekhar mass white dwarf explosion models, where the ∆m15 parameter is a progenitor age indicator.

Key words: supernovae: individual: Type Ia – galaxies: disc – galaxies: stellar content – galaxies: structure.

1 INTRODUCTION

Type Ia supernovae (SNe Ia) are known to arise from carbon–oxygen

(CO) white dwarfs (WDs) in interacting close binaries. About one-

third of SNe Ia contain unusual properties and divided into follow-

ing main subclasses: 91T-like SNe are ∼0.6 mag overluminous

than normal SNe Ia at the B-band maximum; 91bg-like events are

∼2 mag subluminous than normal ones (e.g. Taubenberger 2017).

SNe Ia luminosities at B-band maximum and their light curve (LC)

decline rates (∆m15 - difference between magnitudes at the maxi-

mum light and those of 15 days) are correlated (Phillips 1993): more

luminous SNe Ia have slower declining LCs.

Many studies demonstrated that the progenitor population age of

SN Ia subclasses is increasing in the sequence of 91T-, normal, and

91bg-like events (e.g. Howell 2001;Ashall et al. 2016). Theoreti-

cally, the progenitor age distribution at the current epoch should have

a bimodal shape, with the ﬁrst peak being below or close to 1 Gyr

and corresponding to the young/prompt SNe Ia, and the second peak

being at about several Gyr and including old/delayed events (e.g.

Childress et al. 2014).

In Hakobyan et al. (2017, hereafter H17), taking into account that

the height from the disc plane is an indicator of stellar population

age (e.g. Seth et al. 2005;Yoachim & Dalcanton 2006;Ciucˇ

a et al.

2018), we showed that the majority of SNe Ia are localized in the

discs of edge-on galaxies and they have about two times larger scale

height than core-collapse (CC) SNe, whose progenitors’ ages are up

to ∼100 Myr. Also, we showed that the scale height of SNe Ia is

compatible with that of the older thick disc population of the Milky

Way (MW) galaxy. Nevertheless, we did not investigate different

subclasses of SNe Ia separately. In this Letter, for the ﬁrst time, we

?E-mail: l.barkhudaryan@yerphi.am

attempt to accomplish this by studying the distributions of heights

of various SN Ia subclasses from the host discs.

Recently, in Hakobyan et al. (2020, hereafter H20), we veriﬁed an

earlier ﬁnding on the correlation between LC decline rates of SNe Ia

and the global ages of their host galaxies: SNe Ia from older and

younger stellar populations, respectively, have larger and smaller

∆m15 values. This result can be interpreted within the frameworks

of the sub-Chandrasekhar mass (MCh ≈1.4M) WD explosion

models. The explosion mechanism is realized in the double detona-

tion of a sub-MCh WD, in which accreted helium shell detonation

initiates second detonation in the core of CO WD (e.g. Sim et al.

2010;Blondin et al. 2017;Shen et al. 2017). More luminous SNe Ia

that have slower declining LCs (smaller ∆m15 values) are produced

by the explosion of more massive sub-MCh WD, because the lumi-

nosity of SN Ia is related to the mass of 56Ni synthesized during

the WD explosion (e.g. Stritzinger et al. 2006), which in turn is re-

lated to the mass of the WD (see e.g. Piro et al. 2014;Shen et al.

2018, for a variety of speciﬁc explosion models). On the other hand,

more massive WD would come from more massive main-sequences

stars, which have shorter lifetime than the progenitors of less mas-

sive WDs. In addition, due to the gravitational wave emission, mas-

sive WDs in the binary system would interact in a shorter timescale.

Thus, it should follow that the LC decline rate ∆m15 of SN Ia is

correlated with the age of the SN progenitor system (e.g. Shen et al.

2017,2021). Given this, in our study we simply check the potential

correlation between the SN Ia heights from host discs and their LC

decline rates, which may provide an indication that both parameters

are appropriate stellar population age indicators.

2 SAMPLE SELECTION AND REDUCTION

In this study, to ensure a sufﬁcient number of SNe and to appro-

priately measure the SN heights from their host galactic discs, we

arXiv:2210.13249v2 [astro-ph.GA] 13 Dec 2022

L2 L. V. Barkhudaryan

selected the spectroscopically classiﬁed SN Ia subclasses (normal,

91T- and 91bg-like) with distances 6200 Mpc from the Open Su-

pernova Catalog (Guillochon et al. 2017). In order to have high con-

ﬁdence on the SN subclasses, the information is additionally ver-

iﬁed utilizing data from the Weizmann Interactive Supernova data

REPository (Yaron & Gal-Yam 2012), Astronomer’s Telegram, web-

site of the Central Bureau for Astronomical Telegrams, etc.1

Since we are interested in SNe Ia that exploded in highly inclined

spiral galaxies, we need to roughly classify the morphology and esti-

mate the inclination of hosts. To perform this we employed the Sloan

Digital Sky Survey Data Release 16 (DR16; Ahumada et al. 2020),

the SkyMapper DR2 (Onken et al. 2019), and the Pan-STARRS DR2

(Chambers et al. 2016), which together cover the whole sky and pro-

vide the gri bands composed images for each host galaxy. Hosts

with visible low inclinations (i∼

<60◦)and obviously elliptical,

lenticular, or irregular morphology were excluded from the study.

Following H20, we further morphologically classiﬁed the hosts and

created the 25 mag arcsec−2elliptical apertures for each galaxy on

the surveys’ g-band images enabling exact measurements of the SN

hosts’ inclinations, semi-major (R25)and semi-minor (Z25)axes.

The next step was to use the estimated elongations (R25/Z25)and

morphological types of galaxies to calculate inclinations, following

the approach of Paturel et al. (1997). It is worth noting that the cal-

culated inclinations for galaxies with prominent bulges are inaccu-

rate, as the isophotes of bulges in highly inclined galaxies reduce the

real galaxy disc inclinations. Such scenarios got special attention for

exact inclination calculation, with only the isophotes of discs be-

ing taken into account (see H17). Finally, we limited the sample of

host galaxies to those with an inclination of 80◦6i690◦. As

a result, we sampled 196 S0/a–Sdm galaxies with a nearly edge-on

view, where a total of 197 SNe Ia were discovered (Table A1).

It is important to test the representativeness of our edge-on SN

host sample compared to a sample of galaxies arbitrarily aligned

along line-of-sight. Using the two-sample Kolmogorov–Smirnov

(KS) and Anderson–Darling (AD) tests,2we compared the distri-

butions of the sampled SN Ia subclasses and the morphological

types of their hosts with the same distributions of nearly complete

volume-limited (680 Mpc) sample of the Lick Observatory Super-

nova Search (LOSS; Li et al. 2011). In our and LOSS samples, the

representations of SN Ia subclasses are not statistically different (the

probabilities that the distributions are drawn from the same parent

sample are >0.6). The frequencies of morphologies in our and

LOSS samples are also consistent between each other (probabilities

are >0.1). Therefore, any artiﬁcial loss or excess of SN subclasses

and/or host’s morphologies should be not signiﬁcant in our sample.

We used the methods described in our earlier study on edge-on

SN hosts (H17) to determine the height (V)of a SN from the plane

of its host disc (i.e. the vertical distance of a SN from the major

axis of host), as well as the projected radius (U)along the plane.

The g-band images were used for these measurements. The values

of Uand Vare given in arcsec units. In this study, we used the R25

normalization (in arcsec) to bring the galaxies to relatively the same

size (normalized height is V/R25, normalized projected radius is

U/R25 ). For more details on the measurement techniques, the reader

is referred to Hakobyan et al. (2016), H17. In the Appendix, we also

applied Z25 normalization to the Vparameter (V/Z25).

Recall that the galaxies in our sample have an inclination of

80◦−90◦, which can introduce discrepancies in (projected) height

1The sources provide also the equatorial coordinates of the selected SNe Ia.

2We used a 5 per cent signiﬁcance level as the threshold for the tests.

Table 1. Comparison of the positional distributions of the SN Ia subclasses

along major Uand minor Vaxes.

SN NSN h|U|/R25iversus h|V|/R25iPMC

KS PMC

Normal 144 0.28+0.05

−0.04 versus 0.07+0.01

−0.01 <0.001 <0.001

91T 30 0.25+0.12

−0.07 versus 0.05+0.03

−0.02 <0.001 <0.001

91bg 23 0.25+0.15

−0.08 versus 0.14+0.08

−0.04 0.022 0.145

All 197 0.27+0.04

−0.03 versus 0.07+0.01

−0.01 <0.001 <0.001

Notes. The PKS and PAD probabilities that the distributions are drawn

from the same parent sample are calculated using a MC simulation with 105

iterations. Each subsample’s mean values with their 95 per cent conﬁdence

intervals (CIs) are presented. The P-values are bolded when differences

between the distributions are statistically signiﬁcant (P60.05).

measurements when compared to physical heights in the same galax-

ies with inclination of 90◦. To check the impact of this effect, using

a Monte Carlo (MC) simulation, we generated 1000 SN heights in

i= 90◦disc adopting a generalized vertical distribution (f(˜z) =

sech2(˜z/˜z0), where ˜z=V/R25)and its scale heights for SNe Ia in

spiral galaxies (˜z0= 0.083;H17). Then, we randomly assigned an

inclination within 80◦−90◦to the disc for each SN and estimated

the projected SN height from the major axis of the host. Eventu-

ally, the comparison of the generated and projected heights showed

that the differences between them is not signiﬁcant (PKS = 0.130,

PAD = 0.200). Thus, the mentioned effect has a minor impact on

the real height measurements, which accounts, on average, for about

10 per cent of the measured value (±0.01 in absolute units, typically

within the range of measurement errors).

Because we aim to investigate the possible relationships between

photometric features like ∆m15 and the heights of SNe Ia from the

disc, following H20, we conducted a thorough literature search to

assemble the B-band LC decline rates for our 197 SNe. Only 69 of

the SNe Ia in our sample have available ∆m15 values.

Table A5 provides our database of 197 individual SNe Ia (SN

designation, Uand V, spectroscopic subclass and ∆m15 with their

sources) and their 196 hosts (galaxy designation, distance, morpho-

logical type, R25, and Z25 ).

3 RESULTS AND DISCUSSION

3.1 Directional distributions of SNe Ia in edge-on spiral hosts

Following Hakobyan et al. (2021), for the SNe Ia in edge-on spi-

rals of the current study, we perform the two-sample KS and AD

tests comparing the |V|/R25 and |U|/R25 distributions between

each other. Table 1shows that the bulk of SNe Ia in all of the SN

subclasses are localized in the host galaxies’ discs. For 91bg-like

SNe only, the AD test shows barely signiﬁcance, unlike the KS test,

which is probably due to the statistics with the smallest sample size.

We then compare the projected and normalized radii |U|/R25 and

the heights |V|/R25 between different SN Ia subclasses. In Table 2,

the KS and AD tests show that the radial distributions of normal,

91T- and 91bg-like SNe are consistent with one another. In addition,

the height distributions of normal and 91T-like SNe are consistent

between each other. At the same time, the height distributions of

91T- and 91bg-like SNe are signiﬁcantly different. The same is hap-

pens for the distributions of normal and 91bg-like SNe (with barely

KS test signiﬁcance). Fig. 1shows a scatterplot of |V|/R25 versus

|U|/R25, and the cumulative distributions of |V|/R25 values for dif-

ferent SN Ia subclasses. The 91T-like SNe have the smallest height

distributions, closest to the disc plane, whereas the 91bg-like SNe

MNRAS 000,1–6(2022)

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

MNRAS000,16(2022)Preprint14December2022CompiledusingMNRASLATEXstylelev3.0ConstrainingTypeIasupernovaethroughtheirheightsinedge-ongalaxiesLilitV.Barkhudaryan?CenterforCosmologyandAstrophysics,AlikhanianNationalScienceLaboratory,2AlikhanianBrothersStr.,0036Yerevan,ArmeniaAccepted2022December13.Recei...

展开>> 收起<<

Constraining Type Ia supernovae through their heights in edge-on galaxies Lilit V . Barkhudaryan.pdf

共7页,预览2页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

Constraining Type Ia supernovae through their heights in edge-on galaxies Lilit V . Barkhudaryan

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: