AGN luminosity and stellar age -- two missing ingredients for AGN unification as seen with iPTF supernovae

0.36 MB·

by  Beatriz Villarroel

AGN luminosity and stellar age – two missing ingredients for
AGN unification as seen with iPTF supernovae
7
Beatriz Villarroel1,2, Anders Nyholm3, Torgny Karlsson1, S´ebastien Comero´n4, Andreas J.
1
0 Korn1, Jesper Sollerman3, Erik Zackrisson1
2
1. Department of Physics and Astronomy, Uppsala University, SE-751 20, Uppsala, Sweden

n 2. Centre for Interdisciplinary Mathematics (CIM), Uppsala University, SE-751 06, Uppsala, Sweden
a
3. Department of Astronomy and The Oskar Klein Centre, Stockholm University, SE-106 91 Stockholm,
J
Sweden
0
4. University of Oulu, Astronomy Research Unit, 90014 Oulu, Finland
3

]
A
G
ABSTRACT
.
h
p Active Galactic Nuclei (AGN) are extremely powerful cosmic objects, driven by accretion of
- hot gas upon super-massive black holes. The zoo of AGN classes are divided into two major
o
groups, with Type-1 AGN displaying broad Balmer emission lines and Type-2 narrow ones.
r
t For a long time it was believed that a Type-2 AGN is a Type-1 AGN viewed through a dusty
s
a kiloparsec-sizetorus, but an emerging body of observations suggests more than just the viewing
[ angle matters. Here we report significant differences in supernova counts and classes in the first

study to date of supernovae near Type-1 and Type-2 AGN host galaxies, using data from the
1
v intermediatePalomarTransientFactory,theSloanDigitalSkySurveyDataRelease7andGalaxy
7 Zoo. We detect many more supernovae in Type-2 AGN hosts (size of effect 5.1σ) compared
4 to Type-1 hosts, which shows that the two classes of AGN are located inside∼host galaxies with
6
differentproperties. In addition,Type-1 andType-2AGN thatare dominatedby starformation
8
according to WISE colours m m < 0.5 and are matched in 22 µm absolute magnitude
0 W1 W2
−
. differbyafactorofteninL[OIII]λ5007luminosity,suggestingthatwhenresidinginsimilartype
1
ofhostgalaxiesType-1AGNaremuchmoreluminous. Ourresultsdemonstratetwomorefactors
0
7 thatplay animportant rolein completing the currentpicture: the ageof stellar populations and
1 theAGNluminosity. ThishasimmediateconsequencesforunderstandingthemanyAGNclasses
: and galaxy evolution.
v
i
X Subject headings: AGN — supernova— surveys– unification – active – galactic – nuclei
r
a 1. Introduction anisotropic sample selection or small sample sizes
(Antonucci 2012).
The AGN Unification theory (Antonucci 1993)
Recently,itwasshownthatevenwithisotropic
has been subject to many successes and, also,
selection criteria and large sample sizes of thou-
some controversies. In its simplest form only
sands of pairs, the galaxy neighbours to Type-1
the viewing angle towards the torus matters.
andType-2AGNdiffersignificantly(Villarroel & Korn
This basic picture has long been challenged
2014)withinafewhundredkiloparsec,includinga
by statistical tests (Laurikainen & Salo 1995;
difference inthe number ofclose neighbour galax-
Dultzin-Hacyan et al.1999;Koulouridis et al.2013).
ies (Jiang et al. submitted).
Many statistical tests are often overlookeddue to
This claim is not uncontroversial. Seemingly,
[email protected] Gordon et al. (2016) do not find any differ-
[email protected] ences in the galaxy neighbours. However, their
study suffers from poor statistics (using only tens
1
of pairs) and does not mention some significant spirals than in spirals hosting AGN. This fraction
colour differences they find in neighbours, see p- is connected to the earlier morphological type of
values from the Kolmogoroff-Smirnofftests in Ta- the AGN hosts (Hakobyan et al. 2014) and the
ble 4 of Gordon et al. (2016). stage of eventual merger in close pairs of galaxies
Recent works (Donoso et al. 2014; Koulouridis (Nazaryan et al. 2013).
2014; Trippe 2014; Bitsakis et al. 2015) support In AGN Unification theory the SN counts and
that factors beyond the viewing angle must play typesareexpectedto bethe sameforbothclasses
a role (Krongoldet al. 2002). Some believe the ofAGN.InthisstudyusingSNefromtheinterme-
torus is a nuclear stellar nurserydrivenby inflows diate Palomar Transient Factory (iPTF), we test
and outflows (David & Hickox 2012) or the outer whether Type-1 and Type-2 AGN host galaxies
parts of a disk-wind coming from the accretion havethe sameor differentoccurenceratesof SNe.
disk (Elitzur & Shlosman 2006), driving an evo- DifferentSNrateswouldindicatetheseAGNtypes
lutionary sequence from Type-1 Type 1.2/1.5 resideingalaxieswithdifferentstar-formationhis-
→
1.8/1.9 Type-2 (Elitzur et al. 2014). Pres- tories.
→ →
ence ofspectralfeaturesat10and18µm indicate The AGN and galaxy samples are taken from
thattheAGNhasdifferentclumpinessofthetorus the Sloan Digital Sky Survey (SDSS) (York et al.
depending on if it is a Type-1 or Type-2 AGN 2000)DataRelease7(Abazajian et al.2009). The
andifisolatedorinmerger(Mendoza et al.2015). SNearetakenfromthe iPTFcatalogue. Thissur-
Theabsenceofdetectedbroad-lineregionsinlow- veyiswellsuitedforthisworkduetoitscoherent,
luminosityAGN(Nicastro et al.2003),andlackof untargeted mode of detecting transient sources
toriatthehigh-luminosityend,furthercomplicate with one and the same telescope.
the picture.
InSection2wediscussthesampleselectionand
These findings open up for some important methods. In Section 3 the main results, in Sec-
questions: what are the true physical differences tion 4 different potential biases that can influence
between the Type-1 and Type-2 AGN popula- the results. In Section 5 we present the statisti-
tions? Differences could lie in the physics of calanalysis. Finally,wepresenttheconclusionsin
the central engines, in the structure of the tori Section 7.
(Ramos-Almeida et al. 2011; Ricci et al. 2011;
Elitzur 2012), but also in the host galaxies them- 2. Methods
selves, e.g. Heckman et al. (1989); Maiolino et al.
(1995). 2.1. The iPTF survey
The star-formation histories of the host galax-
The iPTF is an un-targeted wide-field sky sur-
ies can be readily compared with the help of su-
vey using the 1.2 m Samuel Oschin telescope
pernovae (SNe), requiring no assumptions about
(P48)atPalomarObservatorytodetectandfollow
the composition of the galaxy spectra. The
up transient astronomical sources. The scientific
luminosity-weighted age of the stellar popula-
scope of the iPTF spans from small solar system
tion is reflected in the occurrence of different
objects to extragalactic phenomena. The iPTF
SN types: progenitors of core-collapse (c-c) SNe
project has been running since 2013,and it had a
are massive stars (M 8 M ) with short life-
⊙ precursor, the Palomar Transient Factory (PTF),
≥
times (< 107 yrs) and indicate recent or ongo-
which was active during 2009-2012. The techni-
ing star formation. Thermonuclear SNe, whose
cal background of PTF is presented in Law et al.
progenitors are white dwarfs that take on aver-
(2009) and the scientific motivation in Rau et al.
age 109 years to form, are indicators of ear-
(2009). TheiPTFispresentedinKulkarni(2013).
∼
lier epochs of star formation. Early-type galaxies
For brevity, we will refer to our SN catalogue (in-
which are dominated by old stellar populations
cludingSNefoundduringthePTFperiod)simply
are not known to host core-collapse SNe, whereas
as the iPTF catalogue.
thermonuclearSNearefoundinalltypesofgalax-
In this study, we use the iPTF SN catalogue
ies (van den Bergh & Tammann 1991). Recent
fromthetimewindowbetween2009March2until
works indicate a larger fraction of core-collapse
2014June17. Thiscataloguecontains2190extra-
to thermonuclear SNe in non-active, star-forming
2
galacticSNewithknownrightascension(α),decli- The Asiago catalogue is extensive but has uneven
nation (δ), spectroscopic redshift (z ) and spec- quality – some of the SNe in it lack spectroscopic
SN
troscopic classification. This SN catalogue con- classification, some lack spectroscopic redshift.
tains 1494 thermonuclear SNe (i.e. Type Ia) and The SN catalogue compiled by Lennarz et al.
632 core-collapse SNe. In the core-collapse cate- (2012) contains data for 5526 extragalactic SNe,
gory, we include SNe Types Ib, Ic, Ib/c, Ibn, II, whereastheSternbergSNcatalogue,presentedby
IIP, IIL, IIn and IIb. The remaining 64 SNe are Tsvetkov et al.(2004),containslessthan3000ex-
either of unclear spectroscopic type or superlumi- tragalactic SNe. The circumstance that they are
nousSNe. Noneofthe11superluminousSNewith compiledfroma wide rangeofsourcesmake them
z < 0.2 in our catalogue was found in a galaxy less suitable compared to the cohesive iPTF cat-
with a spectrum in SDSS DR7. The superlumi- alogue. The same holds for The Open Supernova
nous SNe can therefore, for our purposes, be in- Catalogue (Guillochon et al. 2016), with 37000
≈
cluded in the remainder category. SNe(ofwhich12%havespectrainthecatalogue)
The SNe in this sample are located at declina- asof2016December,collectedfromdifferentpub-
tions 25◦ <δ <80◦, mostlyatgalacticlatitudes lic sources.
−
b > 20◦. The mean redshift of the sample is As discussed by Anderson & Soto (2013), ear-
| |
z 0.09, with 95 % of the SNe having z < 0.2. lierSNsearcheshaveprioritisedSNdetectionover
≈
Foramotivationofcurtailingthe SNcatalogueat completeness with respect to SN types or host
2014 June 17, see Sect. 4.3. galaxy types. This bias should affect such com-
An advantage of the iPTF catalogue is its co- pilations as the Asiago catalogue and other cata-
hesive nature. All the SNe have been discovered logues listing SNe found before the startof untar-
during an untargeted search with the same tele- geted SN searches.
scope, and spectroscopic classifications have been
2.3. Selection of AGN
made in a timely fashion. The compatibility with
SDSS in sky coverage makes the iPTF SNe sam- The samples of host galaxies were obtained
ple suitable for our investigation. Our SDSS sam- through the SDSS Data Release 7. We select
ples covers 10◦ <δ <70◦ and galactic latitudes objects classified as either ’Quasars’ or ’Galax-
−
b > 20◦, comparable to the distribution of the ies’, within redshift 0.03 < z < 0.2, unless
| |
SN locations. flagged for brightness (flags&0x2=0), saturation
(flags&0x40000=0), or blending (flags&0x8=0)
2.2. Comparison with other supernova
(Stoughton et al. 2002, their table 9).
catalogues
The emission lines are obtained from the
Another SN catalogue that comes to mind is SpecLine table in DR7. We require that the ob-
the SDSS SN catalogue (Sako et al. submitted). jectshaveHαinemissionandselectType-1AGN,
The 902 confirmed SNe in the SDSS SN survey1 Type-2 AGN and star-forming galaxies using op-
represents a SN sample of smaller scope than the tical emission line diagnostics. Our Type-1 AGN
iPTF sample. Not all SNe are spectroscopically areobjects withσ(Hα) >10˚A (orFWHM(Hα) >
classified,butthesurveyisdeepandhomogenous. 1000 km/s). The Type-2 AGN have narrow lines
Unfortunately,theSNearealllocatedintheSDSS σ(Hα) < 10 ˚A fulfilling the Kauffmann criterion
southernequatorialStripe82,whichisoutsidethe (Kauffmann et al. 2003):
area of the SDSS DR7 sample from the central
region used in this study. This renders the SDSS
log([Oiii]/Hβ)>0.61/(log([Nii]/Hα)) 0.05)+1.3
SNe unsuitable for our purposes. −
(1)
The Asiago SN catalogue (Barbon et al. 1999)
is a historically comprehensive compilation of ex-
The star-forminggalaxiesare defined as all the
tragalactic SNe. The catalogue encompasses 6530
other narrow-line objects.
SNe as of 2016 March 27, in both hemispheres.
In this way, we classify the objects into Type-
1Listedathttp://classic.sdss.org/supernova/snlist.dat 1s, Type-2s and star-forming galaxies, referred
to as “largest samples”, using optical emission
3
line diagnostics. For all objects we search for lect a galaxy from the second parent sample hav-
morphological classifications (’Spiral’, ’Ellipti- ingtheclosestvalueinredshiftandaspecificprop-
cal’, ’Uncertain’) from the project Galaxy Zoo 1 erty of interest. After the matching is done, we
(Lintott et al.2008, 2011), emissionline measure- firstthrowawayallmatchedpairsthatdiffermore
ments, redshifts and celestial coordinates, leaving than 20% in the property of interest.
“parent samples” of 11632 Type-1 AGN (1864 Fourtypes ofspecific propertiesandmatchings
spiral), 77708 Type-2 AGN (36720 spiral) and are explored:
137489 star-forming galaxies (49072 spiral). For
the vast majority of these objects we can also 1. Redshift only. This allows to remove biases
find Wide-field Infrared Survey Explorer (WISE) in Galaxy Zoo morphology classifications as
magnitudes. well as the Malmquist bias.
2. L[Oiii]5007 from the narrow-line region
2.3.1. Refined samples
(NLR). In the simplest AGN Unification,
As additional samples later used only for di- the NLR is believed to be isotropically dis-
rect comparison of host galaxy properties, we tributed outside the torus and to be equally
also create some ’refined samples’. Starting from strong for AGN of the same activity level
the parent samples, we select only face-on spi- irrespective of the viewing angle. Select-
ral hosts with high S/N in the emission lines ing on L[Oiii]5007 – meaning one selects
from Galaxy Zoo Data Release 2 (Willett et al. all Type-1 and Type-2 AGN above a se-
2013), minimizing dust extinction due to host lected certain line flux in a sample – should
galaxy inclination. We also require S/N > 3 give the same host galaxy properties un-
in Hα, minimum SDSS Gaussian line heights der the conditions of isotropy (Antonucci
h(Hα) > 10 * 10−17erg/s/cm2/˚A and h(Hβ) > 1993). Matching on L[Oiii]5007 is simi-
5 * 10−17erg/s/cm2/˚A in order to avoid effects lar to selecting on L[Oiii]5007 if the two
of stellar absorption affecting weak lines in our L[Oiii]5007 distributions are the same (as
classification. Out of these, we select only those predicted by the simplest Unification) but
having WISE colours. canbe problematicifthedistributionsdiffer
The refined samples will be used for compar- at the high-luminosity end. Moreover, we
ing star formation with WISE colours in Type-1 expect the same line width σ[Oiii]5007 in
andType-2AGNorthe[Oiii]5007inhostgalaxies matched samples Type-1 and Type-2 AGN
matched by amount of dust. samples.
3. WISEM (22µm)absolutemagnitude. We
2.4. Pairwise matching w4
use this match as our stellar-mass proxy as-
The numbers of coherently collected SNe are suming the dust emission traces the stel-
scarce (2190 in our sample as on 2014 June 17). lar mass. The 22µm magnitude is a good
Thus, we create pairwise matched subsamples of measure of heated-dust emission in the host
Type-1 AGN, Type-2 AGN and star-forming ob- galaxy,especiallyingalaxieswherethetorus
jectstocompareobjectsassimilartoeachotheras contribution to the total 22µm is negligi-
possibleinredshiftdistributionandselectedprop- ble in comparison, meaning galaxies having
erties e.g. the luminosity L[Oiii]5007. We com- m m < 0.5 (Wright et al. 2010). A
W1 W2
−
pare (i) Type-1 AGN to Type-2 AGN, and (ii) less favourableoption is to use dust redden-
Type-2 AGN to star-forming galaxies. The aim ing F(Hα/Hβ), but if dust reddening in the
with the latter test is to probe whether the ob- BLR and NLR differs (Gaskell 1984), the
served Type-2 AGN properties can be explained matching will be biased. We therefore nei-
by star formation alone. ther correctL[Oiii]5007 for dust reddening.
For two samples of intrinsically similar objects This matching is only done for the galaxies
the probability of detecting faint SNe is the same inthe parentsamples thathave WISE mag-
if they have similar redshift distributions. There- nitudes.
fore, for each galaxy in the parent sample, we se-
4. Exponential fit scale radius (r-band). As
4
the star-formation history depends on the than the redshift matching condition of
availablegasmass,matchingbyanapparent z z < 0.01 used in the study by
SN AGN
| − |
measure of the galaxy volume should min- Wang et al.(2010). In their study, however,
imize differences in star-formation histories most (97 %) of their sample of 620 cross-
undertheassumptionofamass-sizerelation. matches fulfills the z z < 0.003
SN AGN
| − |
condition.
We do a two-sample Kolmogoroff-Smirnofftest
for each matched property to ensure the sam- If both these conditions are fulfilled, the SN is
ple distributions are statistically similar. For the considered to be associated with the galaxy. The
M -matchedsampleswehadto doa finermatch SNe inside a galaxy are collected by the d < 10
w4
by throwing away matched pairs differing more kpc, and those inside a galaxy or in a close com-
than 5% in M . panion are found with the d < 100 kpc criterion.
w4
The pairwise matched subsamples for exam- If this is wrong and there is no association be-
ining SN counts are created by matching in tweenthe AGN and the SN, then the distribution
three different parameters (redshift z, luminosity ofdifferentSNewithdifferent ∆z = zSN zAGN
| | | − |
L[Oiii]5007 and m ) as described earlier. The ought to be a uniform distribution. This can be
W4
sizes of the pairwise matched subsamples can be checkedbyplotting a histogramwiththe ∆z be-
| |
found in Table 4. An example of the redshift and tween the AGN and the SNe, see Fig 4. It is
propertydistributionsinthe matchedType-1and clear that the distribution is bottom-heavy and
Type-2samplescanbeseeninFigure1and2. The that most of the SNe are associatedwith the host
similarity of all the matched samples is ensured galaxies.
through two-sample Kolmogoroff-Smirnoff where It is important to point out, that the SN red-
the null hypothesis (that two matched samples shift determination method brings in uncertainty
are similar) holds for the nominal value α=0.05. regarding the redshift uncertainties. Most of the
The procedure is repeated for pairwise matched timetheredshiftismeasuredfromthehostgalaxy
Type-2 AGN and star-forming galaxies. of the SN and is rather accurate. But some-
Finally, the refined samples of face-on hosts times, the SN redshift is measured directly from
withhighsignal-to-noisearealsomatchedwiththe the SN itself and influenced by Doppler broaden-
samemethod. Inaddition,therefinedsamplesare ing and expansion velocities, being slightly larger
matched by Balmer decrement F(Hα)/F(Hβ). (Blondin & Tonry 2007).
The degree of association between SNe and
2.5. Matching of supernovae galaxies can also be checked by visually inspect-
ing the SDSS DR7 images of the galaxies, with
ASNisconsideredtobematchedwithagalaxy
the SN positions overplotted. We note the sub-
if the following two conditions are both satisfied:
jectivity involved, e.g. in cases where a SN oc-
curs in an interacting pair of galaxies. For the
The projected distance on the plane of the
• largest sample of Type-2 AGNs (77708 galaxies),
sky between the SN and the galaxy is com-
and the 59 SNe matched to them, about 10 % of
puted(planeapproximation). Thesearchra-
the SNe visually appears to be located in a com-
dius d is set by converting the desired phys-
panion galaxy of a sample galaxy. For the largest
ical search radius into a angular radius at a
sample of starforming galaxies (137489 galaxies),
galaxy distance assumed to be z c/h .
AGN · 0 and the 152 SNe matched to them, a comparable
A Hubble constant h = 72 km s−1 Mpc−1
0 fraction (about 9 %) appears to be located in a
is used throughout this work. If the SN is
companion galaxy. These results from visual in-
found to lie inside the given search radius,
spection of SN positions complements the conclu-
the compliance with a redshift condition is
sions drawn from Fig. 4.
also checked.
Our redshift condition is z z <
SN AGN
• | − |
0.003 and accounts for redshifts due to
peculiar motions. This is more strict
5
z−matched [OIII] 5007−matched w4−matched
2000 1000 2000
N
G 800
1500 1500
A

1 600
−
e 1000 1000
p 400
y
T
500 500
N, 200
0 0 0
0 0.1 0.2 0 0.1 0.2 0 0.1 0.2
2000 1000 2000
N
G 800
1500 1500
A

2 600
−
e 1000 1000
p 400
y
T
500 500
N, 200
0 0 0
0 0.1 0.2 0 0.1 0.2 0 0.1 0.2
redshift z redshift z redshift z
Fig. 1.— The redshift distributions of matched samples. The redshift distributions of matched Type-1
and Type-2 AGN are demonstrated for three different types of matchings: redshift, L[Oiii]5007 and m .
w4
Two-sample Kolmogoroff-Smirnofftests confirm the pairwise matched Type-1 and Type-2 distributions are
the same.
6
w4−matched [OIII] 5007−matched
7000 6000
N 6000 5000
G
A 5000
4000

1
4000
−
e 3000
p 3000
y 2000
T 2000

,
N 1000 1000
0 0
4 6 8 10 12 0 1000 2000 3000 4000 5000
w4−magnitude F[OIII] 5007
w4−matched [OIII] 5007−matched
7000 6000
N 6000 5000
G
A 5000
4000

2
4000
−
e 3000
p 3000
y 2000
T 2000

,
N 1000 1000
0 0
4 6 8 10 12 0 1000 2000 3000 4000 5000
w4−magnitude F[OIII] 5007
Fig. 2.— The distributions of L[Oiii]5007 and m . The distributions of L[Oiii]5007 or m in matched
w4 w4
Type-1 and Type-2 AGN samples are demonstrated for two different types of matchings: L[Oiii]5007 and
m . In the L[Oiii]5007 histograms, four objects in each sample above F > 5000 are not plotted in the
w4
histogram due to visibility. Two-sample Kolmogoroff-Smirnoff tests confirm the pairwise matched Type-1
and Type-2 distributions are the same.
7
3. Results its. The difference in L[Oiii]5007-selectedType-1
andType-2AGNsamplesisnoteable(F[Oiii]5007
3.1. Supernova counts >10 case: size of effect 2.8 σ, F[Oiii]5007>30
∼
case: 1.8 σ) following the same trend of Table 3.
We begin our analysis by counting SNe within
This reflects purely changes in the sample sizes,
twodifferentprojecteddistancesfromthecenterof
thehostgalaxies: 10kpc(withinthetypicalradius unless the difference in matched vs selected sam-
ples originates in a break-down of Unification at
of a spiral galaxy), 100 kpc (within the galaxy or
a possible close companion). We start with the the higher luminosities.
largest samples, see Table 3. As early-type objects are well-known (Li et al.
2011) for having few SNe setting off, it would be
The first thing to notice is the lack of SNe
around Type-1 AGN at projected separations d idealtouseonlyspiralhosts. Wedonotfindasin-
gleSNaroundspiralType-1AGN.Butusingasta-
< 100 kpc. Only one SN is found. The number
of SNe around Type-2 AGN are clearly system- tisticalhypothesistest(Krishnamoorthy & Thomson
atically higher for the largest samples in Table 3 2002)(seeSection5.3)onlyaborderline-significant
difference in SN counts for Type-1 and Type-2
(size of effect 5.1σ). We verify that this is no
biasduetopot∼entialdifficultiesindetectingSNin AGN at d < 100 kpc (p-value 0.06) is found,
∼
showing the need of larger SN samples.
theimmediatevicinityofbrightortransientAGN
byexcludingallSNe within3arcsecondsfromthe For the pairwise matched samples the counts
host centres, see Section 4.3.2. This still yields 1 aretoosmalltoshowanysignificanceindividually.
SNnear 11632Type-1AGN (detectionfraction The L[Oiii]5007-matchedsamples we commented
∼
8.6 10−5) and 46 SN near 77708 Type-2 AGN upon earlier, while for the redshift-matched sam-
×
(detection fraction 5.9 10−4) – a significant ples a significant difference is found for d < 10
∼ ×
difference (size of effect 4.1 σ, see Section 5.1). kpc. The lack of significant difference for d< 100
∼
Visual inspection show that the majority of the kpc therefore stems from the poor statistics. The
SNe (> 90 %) come directly from the AGN hosts samples matched in redshift and apparent galaxy
(only a small fraction from the companion galax- size yield small, insignificant differences between
ies). Type-1andType-2AGN.Asthesizeofthegalaxy
depends on the star-formation history, no differ-
For the Type-1 and Type-2 samples matched
in redshift or M the differences are significant. ence in counts could mean that we have either
w4
But for the L[Oiii]5007-matchedsamples we only matchedsuccessfullyinthestar-formationhistory,
or more realistically,that the sample sizes are too
find one SN in eachsample, yielding no difference
at all. Does this mean the Simplest Unification is small. However, the collected results in Table 4
are visually presented in Figure 5 where the left
validandthereisnodifferenceingalaxyproperties
andSNcountsinsamplesselectedandmatchedin column reinforces our earlier conclusions: Type-
L[Oiii]5007? If so, also the host morphologies for 2 AGN hosts have higher SN rates than Type-1
AGN hosts, much higher than from the expec-
the matched samples should be the same. But
the fraction of Type-1 and Type-2 AGN in spiral tations of the simplest Unification theory (repre-
sented by the grey line). Also the insignificant
hosts is 20 % vs 44 %, in disagreement with the
differences fall into the same area of the plot.
Simplest Unification. This suggests that the lack
of significant difference in SN counts stems from The larger count of SNe could mean either a
poor statistics. difference in stellar age or stellar mass (or both).
Asanalternativetest,weselectonL[Oiii]5007. Butsamplesunmatchedinstellarmassleadtodif-
This should be suitable as in the Simplest Uni- ferences inclustering onMpc scale (Mendez et al.
fication no difference is expected in L[Oiii]5007 2016)while wefindnodifferencesinSNcountson
on the higher luminosity-end (while matching re- largescale,seeSections4.3.3and4.3.4. Moreover,
the matching in M – our proxy for stellar mass
movespotential differences at the high-luminosity w4
end). We do this twice for the largestsamples us- – still yields significant different SN counts. This
ingbothF[Oiii]5007>10 10−17erg/s/cm2and suggests that the discrepancy in SN counts is due
F[Oiii]5007 >30 10−17×erg/s/cm2 as flux lim- to differences in stellar age between Type-1 and
× Type-2 AGN.
8
Theconclusionissupportedbythemuchhigher properties. Wetrythisfortwoseparatelowerflux
w2 w3 colour indices in Type-2s indicating limits: F > 10 or 30 10−17 erg/s/cm2, yielding
− ×
stronger star formation (Coziol et al. 2015) in the following SN counts within d < 100 kpc from
samples consisting of face-on, spiral-host Type- the host galaxies in the largest samples:
1 and Type-2 AGN that are dominated by dust
emission from stars with m m < 0.5 1. F > 10 10−17 erg/s/cm2. We have 5101
W1 W2
− ×
(Wright et al. 2010) and pair-wise matched in Type-1 AGN and 47010 Type-2 AGN. The
L[Oiii]5007, see Section 4.1. number of SNe near Type-1 AGN is 1 (de-
Less striking differences are found between tection fraction 1.96 10−4) The number
∼ ∗
Type-2 AGN and star-forming galaxies. Star- of SNe near Type-2 AGN is 41 (detection
forming spirals show higher core-collapse SN fraction 8.72 10−4). Theestimateofthe
∼ ∗
counts at d < 10 kpc ( 3.3σ) and even in the size of effect is 2.8σ.
∼ ∼
redshift-matched samples a difference is present.
2. F > 30 10−17 erg/s/cm2: We have 3403
The rightcolumninFigure5showsarelationship ×
Type-1 AGN and 19976 Type-2 AGN. The
between the two classes, although offset in y-axis
numberofSNenearType-1AGNis1(detec-
following the shape of the grey line.
tionfraction 2.94 10−4). Thenumberof
∼ ∗
SNe near Type-2 AGN is 19 (detection frac-
3.2. Anisotropy?
tion 9.51 10−4). Theestimateofthesize
One may argue that the L[Oiii]5007 may be of eff∼ect is ∗ 1.8σ.
∼
anisotropic or even have different physical origins
in Type-1 and Type-2 AGN. If so, one expects The number of SNe in the largest Type-2 host
to find differences in the Gaussian line width of samples (77709 objects) within d < 100 kpc was
the [Oiii]5007 emission in Type-1s and Type-2s 59 SNe (see Table 3). Considering the smaller
and/or differences in SN counts. size of the flux-restricted Type-2 AGN samples,
the expected new SN counts are: (a) F > 10:
3.2.1. SNsamples: anisotropicsampleselection? 47010/77709*59SNe 35,(b)19976/77709*59
∼ ∼
15. Therefore, we can easily see that the loss of
As Type-1 AGN have a contribution of a non-
significanceintheF[Oiii]5007-selectedsamplere-
stellar, power-law continuum component to their
flectsuponthedecreasednumberofobjectsinthe
observed luminosity, this could influence the de-
samples.
tections of Type-1 and Type-2 AGN. However,
the emission lines are independent of the con-
3.2.2. NLR kinematics: anisotropic sample se-
tinuum emission and in the Simplest Unification
lection?
L[Oiii]5007 is expected to be an isotropic indica-
tor of AGN luminosity. To further probe the relevant NLR kinemat-
Type-1 and Type-2 samples selected on, or ics behind the L[Oiii]5007 emission, we analyze
matched in, L[Oiii]5007 should be the same in the AGNthemselves. Ifthe same physicalmecha-
allotherproperties: theyshouldhavesimilarhost nismsarecausingthe[Oiii]5007emissioninType-
galaxy types and similar NLR kinematics. How- 1sandType-2s–andisotropically–the[Oiii]5007
ever, in (Villarroel & Korn 2014) L[Oiii]5007- luminosity-normalized line widths σ([Oiii]5007)
matched hosts showed different colours of their must be the same.
galaxy neighbours. When we explore the Gaussian line widths
Using L[Oiii]5007-matching we see no differ- σ([Oiii]5007) of the various classes of objects,
ence, or cannotsee, in the SN counts near Type-1 we use refined samples matched in redshift and
and Type-2 AGN: we find one SN near Type-1 F[Oiii]5007.
hosts and one SN near Type-2 hosts. Thenormallydistributedlogoftheσ([Oiii]5007)
While matching is good,a better way is still to values for estimating means and errors are cal-
select on F[Oiii]5007. Selecting on F[Oiii]5007 culated for L[Oiii]5007-matched, face-on spiral
meansthatallType-1andType-2aboveacertain hosts. No significant difference in line width for
flux value in F[Oiii]5007 should have the same the Type-1s and Type-2 objects is found (p =
9
0.13). This disagrees with the earlier observation 0.8, i.e. have the WISE bands dominated by dust
that the NLR has a component of motion giving heated by star-formation,permitting the compar-
rise to geometric differences (Gaskell submitted). ison using w2 w3 colours. An example are the
AclearerlinewidthdifferencebetweentheType-2 L[Oiii]5007-m−atched samples where only 109 out
(log (σ)= 0.242 0.003) and star-forming ob- of 123 Type-1 AGN and 116 out of 123 Type-2
10 ±
jects (log (σ)=0.191 0.004). The slighty wider AGN have their dust dominantly heated by star-
[Oiii]500710line in the±Type-1 and Type-2 AGN formation.
(over star-forming galaxies) support that a sig- The best way to find out if this influenced our
nificant contribution to the [Oiii]5007 flux might conclusions, is by redoing the tests using only ob-
arise isotropically distributed in a region close to jects that fulfil the w1 w2 < 0.8 condition by
the AGN nucleus where the clouds rotate around Assef et al. (2012) and−, or the even the stricter
the center of the galaxy at higher velocities, caus- Wrightetal. (2010)conditionw1 w2<0.5. Us-
ing additional Doppler broadening. But also out- ingthesecriteriatheobjectsclearl−yaredominated
flows in AGN are known to cause broadening of bydustemissionnearthe starsandcomparisonof
the[Oiii]5007lineandcanathighredshiftz 2.5 the w2 w3 colour as well as the w4-matching
in extreme cases show broadening correspon∼ding is valid.−These objects we match by either w4
to 2600 5000km/s (Zakamska et al.2016). Per- or L[Oiii]5007. The resulting samples are clearly
haps, th−e slight differences in [Oiii]5007 between smaller.
the AGN and star-forminggalaxies therefore may The resulting, new w4- and L[Oiii]5007-
indicatethepresenceofoutflowsfromthenucleus.
matched samples show the same averagew2 w3
The results are displayed in Table 1. as before. Type-2 shows again higher −star-
formation than Type-1. Also here in the w4-
3.3. AGN luminosity
matched samples one can see that Type-1 are
If the obscuration is the dominant factor that much luminous than Type-2 AGN. An alterna-
tive view can be gained from comparing the M
separates Type-2 from Type-1 AGN, one may w4
expect that the refined samples of Type-1s and in L[Oiii]5007-matched samples: -30.2 +/- 0.1
Type-2smatchedintheheated-dustemissionfrom mag (Type-1 AGN) or -30.8 +/- 0.1 mag (Type-2
AGN), supporting there is more dust in Type-2
their host galaxy are more or less as luminous
in L[Oiii]5007. For the 137 paired Type-1s and AGN hosts than in Type-1 AGN hosts.
Type-2s in the refined samples matched in red- One may wonder if this particular result has
shiftandM ,wecomparethemeanL[Oiii]5007. any connection to the receding torus model
w4
The mean log L[Oiii]5007 [erg/s] is 40.934 (Lawrence et al.1991),wheretheopeningangleof
10 ±
0.0455 for Type-1 objects and 39.9738 0.0889 thetorusgetslargerwithincreasingAGNluminos-
±
for Type-2s. This demonstrates that Type-1s are ity. The increased ratio of Type-1/Type-2 AGN
much more luminous ( 9.6σ) than Type-2s in at larger luminosities (Simpson 2005; Lusso et al.
∼
host galaxies with similar dust distributions. The 2013) supports the idea of a receding torus. How-
effectisequallyconvincingifexploringandmatch- ever, as it appears that the age of the stellar pop-
ing objects with WISE bands dominated by star ulation differs between Type-1 and Type-2 AGN,
formation m m <0.5. itseemsmorereasonablethatthedifferenceinthe
W1 W2
−
dustisonhostgalaxyscale. While ourresultssay
3.3.1. Refined samples dominated by dust emis- thatType-1AGNaremoreluminousthanType-2
sion near stars AGN, they do not support a receding torus per
se.
With the simple criterion w1 w2 >=0.8 one
−
can easily identify sources dominated by AGN
4. Biases
(Assef et al. 2012). In general, WISE-selected
samplesarebiasedtowardsmoreSeyfert-likeAGN There are some potential selection biases that
with low luminosity (Mingo et al. 2016). We do can influence our samples. Weak lines used in the
thisforourType-1andType-2AGNandfindthat objectclassificationmightbe influencedby stellar
majority in the matched samples have w1 w2< absorption. Inthisstudy,demandingaS/N >3in
−
10