Star formation and AGN activity in a sample of local Luminous Infrared Galaxies through multi

20 Pages·2017·13.9 MB·English

MNRAS000,1–19(2015) Preprint4July2017 CompiledusingMNRASLATEXstylefilev3.0
Star formation and AGN activity in a sample of local
Luminous Infrared Galaxies through multi-wavelength
characterization
Rub´en Herrero-Illana,1,2(cid:63) Miguel A´. P´erez-Torres,1,3 Zara Randriamanakoto,4
Antxon Alberdi,1 Andreas Efstathiou,5 Petri V¨ais¨anen,6,7, Erkki Kankare,8
Erik Kool,9,10 Seppo Mattila,11,12 Rajin Ramphul,4,6 and Stuart Ryder.9
7 1Instituto de Astrof´ısica de Andaluc´ıa (IAA-CSIC). Glorieta de la Astronom´ıa s/n, 18008, Granada, Spain
1 2European Southern Observatory (ESO), Alonso de C´ordova 3107, Vitacura, Casilla 19001, Santiago de Chile, Chile
0 3Visiting Scientist: Departamento de F´ısica Teo´rica, Facultad de Ciencias, Universidad de Zaragoza, Spain
2 4University of Cape Town, Astronomy Department, Private Bag X3, Rondebosch 7701, South Africa
l 5School of Sciences, European University Cyprus, Diogenes Street, Engomi, 1516 Nicosia, Cyprus
u 6South African Astronomical Observatory, P.O. Box 9 7935, South Africa
J 7Southern African Large Telescope, PO Box 9, Observatory 7935, Cape Town, South Africa

3 8Astrophysics Research Centre, School of Mathematics and Physics, Queen’s University Belfast, Belfast BT7 1NN, UK
9Australian Astronomical Observatory, 105 Delhi Rd, North Ryde, NSW 2113, Australia

] 10Department of Physics and Astronomy, Macquarie University, Sydney, NSW 2109, Australia
A 11Tuorla Observatory, Department of Physics and Astronomy, University of Turku, V¨ais¨al¨antie 20, FI-21500 Piikki¨o, Finland
G 12Finnish Centre for Astronomy with ESO (FINCA), University of Turku, V¨ais¨al¨antie 20, FI-21500 Piikki¨o, Finland
.
h
p
-
AcceptedXXX.ReceivedYYY;inoriginalformZZZ
o
r
t
s
a ABSTRACT
[ Nuclear starbursts and AGN activity are the main heating processes in luminous in-

frared galaxies (LIRGs) and their relationship is fundamental to understand galaxy
2
evolution. In this paper, we study the star-formation and AGN activity of a sample
v
of 11 local LIRGs imaged with subarcsecond angular resolution at radio (8.4GHz)
3
6 andnear-infrared(2.2µm)wavelengths.Thisallowsustocharacterizethecentralkpc
6 of these galaxies with a spatial resolution of (cid:39) 100pc. In general, we find a good
9 spatial correlation between the radio and the near-IR emission, although radio emis-
0 sion tends to be more concentrated in the nuclear regions. Additionally, we use an
. MCMC code to model their multi-wavelength spectral energy distribution (SED) us-
5
ing template libraries of starburst, AGN and spheroidal/cirrus models, determining
0
7 the luminosity contribution of each component, and finding that all sources in our
1 sample are starburst-dominated, except for NGC6926 with an AGN contribution of
v: (cid:39) 64%. Our sources show high star formation rates (40 to 167M(cid:12)yr−1), supernova
rates (0.4 to 2.0SNyr−1), and similar starburst ages (13 to 29Myr), except for the
i
X young starburst (9Myr) in NGC6926. A comparison of our derived star-forming pa-
r rameters with estimates obtained from different IR and radio tracers shows an overall
a consistencyamongthedifferentstarformationtracers.AGNtracersbasedonmid-IR,
high-ionization line ratios also show an overall agreement with our SED model fit es-
timates for the AGN. Finally, we use our wide-band VLA observations to determine
pixel-by-pixel radio spectral indices for all galaxies in our sample, finding a typical
median value (α (cid:39)−0.8) for synchrotron-powered LIRGs.
Key words: galaxies: interactions – galaxies: nuclei – galaxies: starburst – infrared:
galaxies – radio continuum: galaxies
(cid:63) E-mail:[email protected]
©2015TheAuthors
2 Herrero-Illana et al.
1 INTRODUCTION astrongAGNcontribution(e.g.Roy&Norris1997;Donley
et al. 2005; Ivison et al. 2010; Del Moro et al. 2013) and a
Luminous Infrared Galaxies (LIRGs) are defined as those
FIRexcess(q>2.34)issuggestiveofintensestarformation,
galaxies with an infrared luminosity L [8–1000µm] >
IR this is not always the case (e.g., the obscured AGN in the
1011L(cid:12). The majority of these sources, discovered in the mostFIR-excessgalaxy,NGC1377;Costagliolaetal.2016).
early 1980s by the IRAS satellite, are actually galaxies un-
Ingeneral,the q-factorcannotbeusedasastandalonetool
dergoing a merging process.
toseparateAGNfromstarburstgalaxies(Mori´cetal.2010;
While most LIRGs are dominated by violent episodes
Padovani et al. 2011).
of nuclear star formation (starbursts; see e.g., Petric et al.
The underlying physics for the FIR-radio correlation is
2011;Stierwaltetal.2013),manyofthemalsocontainanac-
usually associated with a star formation origin: dust repro-
tivegalacticnucleus(AGN).Theexistenceandlinkbetween
cesses massive-star UV radiation into far infrared photons,
thesetwoprocesses(e.g.,Shaoetal.2010)makeLIRGsideal
whiletheexplosionsofthosesamestarsassupernovae(SNe)
laboratories in which to study the AGN-starburst connec-
acceleratethecosmicrayelectrons,responsiblefortheradio
tion.
non-thermal synchrotron radiation (e.g., Voelk 1989; Lisen-
The parameters describing the star formation in a
feldetal.1996;Lackietal.2010;Lacki&Thompson2010).
galaxy can be obtained by means of a number of indirect
Thermal bremsstrahlung arising from Hii regions (free-free
tracers and prescriptions based on different bands of the
emission) is also linked to this correlation (e.g., Murphy
electromagnetic spectrum, including UV and far-IR contin-
etal.2011).Therefore,boththermalandnon-thermalemis-
uum,aswellasseveralrecombinationorforbiddenlines(see
sion are correlated with the far-IR and are good tracers
review by Kennicutt 1998). Recently, near-IR observations
of star-formation. Correlations between radio emission and
have also proven useful to characterize the star formation
windows in the IR regime other than the far-IR exist (e.g.,
properties of LIRGs through the study of super star clus-
3.3µm and 8.7µm; as used for NGC1614 in Alonso-Herrero
ters (SSCs; Portegies Zwart et al. 2010; Randriamanakoto
etal.2001;V¨ais¨anenetal.2012;Herrero-Illanaetal.2014),
et al. 2013a,b), which are young ((cid:39) 10Myr) massive star
but are not as clear nor that well studied.
clusters that preferentially form whenever there is strong
A problem of many of the starburst indicators men-
ongoing starburst activity. Other tracers are used to infer
tioned above is the frequent contamination of the tracers
the AGN type, its luminosity and its relative contribution
by the effects of an AGN. Notable examples are the con-
tothebolometricluminosityofgalaxies,suchasX-rayemis-
tamination of the radio continuum by putative jets (Cleary
sion(Treisteretal.2009;Mullaneyetal.2011),mid-IRcon-
et al. 2007) and the significant contribution to the far-IR
tinuum (Asmus et al. 2014) or optical (e.g. Bassani et al.
through the heating of the narrow line region by the AGN
1999; Heckman et al. 2005) and IR spectral lines (Genzel
(Tadhunter et al. 2007). A way to overcome this problem is
et al. 1998; Pereira-Santaella et al. 2010; Diamond-Stanic
tofitamulti-wavelengthspectralenergydistribution(SED)
et al. 2009).
combiningtemplatesofbothstarburstsandAGN(e.g.,Efs-
However, the large amounts of dust present in LIRGs,
tathiouetal.2000;Netzeretal.2007;Mullaneyetal.2011;
whosereprocessingofultravioletphotonsfrommassivestars
Calistro Rivera et al. 2016).
is responsible for the high IR luminosities in these systems,
In this paper, we model the SED of a sample of 11
impose a limitation to optical and near-IR tracers due to
local LIRGs and compare our derived starburst and AGN
dust obscuration (Mattila & Meikle 2001). For this reason,
propertieswithothermodels.WealsopresentradioX-band
observationsatradiowavelengths,unaffectedbydustextinc-
(8.4GHz) and near-IR K -band (2.2µm; hereafter referred
tion,areanalternativeandpowerfultooltotracestarburst S
to as K-band) observations of this sample, comparing them
(SB) and AGN processes in the innermost regions of these
and analyzing possible near-IR/radio correlations. The pa-
systems(Condon1992;Parraetal.2007;P´erez-Torresetal.
perisstructuredasfollows:insection2wepresentthesam-
2009, 2010; Murphy et al. 2011; Romero-Can˜izales et al.
ple, together with a concise individual description of the
2012a).
sources, and in section 3 we present our observations and
There is a well-known tight linear correlation between
details on the data reduction. Through the discussion (sec-
the far-IR and radio (1.4GHz) luminosities (van der Kruit
tion 4) we describe our SED modeling in section 4.1, and
1973; de Jong et al. 1985; Helou et al. 1985; Condon 1992),
compare it with other tracers of star formation and AGN
withnoevidentdependencewithredshift(Ivisonetal.2010;
activity. We then compare our radio and near-IR observa-
Pannella et al. 2015). To quantify this correlation, the so
tions (section 4.2), analyze the special case of IRAS16516-
called q-factor (Helou et al. 1985) was defined as:
0948(section4.3)anddiscusstheradiospectralindexofour
(cid:18)FIR/3.72×1012Hz(cid:19) sources(section4.4).Wesummarizeourresultsinsection5.
q=log , (1) ThroughoutthispaperweadoptacosmologywithH =
S 0
1.49GHz 75kms−1Mpc−1, ΩΛ=0.7 and Ωm=0.3.
where S is the flux density at 1.49GHz in units of
1.49GHz
Wm−2Hz−1 and FIR is defined as
FIR=1.26×10−14(2.58S60µm+S100µm), (2) 2 THE SAMPLE
with S and S being the IRAS fluxes in Jy at 60 The sources in our LIRG sample are taken from the IRAS
60µm 100µm
and 100µm, respectively. The mean value of the q-factor Revised Bright Galaxy Sample (IRBGS; Sanders et al.
in the IRAS Bright Galaxy Sample (Condon & Yin 1990) 2003), and fulfill the following criteria: D < 110Mpc, lu-
is (cid:104)q(cid:105) = 2.34 (Condon & Broderick 1991; Yun et al. 2001). minosity log(LIR/L(cid:12))>11.20, and declination δ >−35◦. We
While a radio excess (q < 2.34) is typically associated with chosethosecriteriasothatour8.4GHzKarlG.JanskyVery
MNRAS000,1–19(2015)
Star formation and nuclear activity in local LIRGs 3
Large Array (VLA) observations in A-configuration (angu- MCG+08-11-002 Arp 299 ESO 440-IG058
lar resolution of (cid:39) 0.3(cid:48)(cid:48)) could image the central kpc re-
gionofallgalaxieswithspatialresolutionsof(cid:39)70−150pc,
whichwouldallowustodisentanglethecompactradioemis-
sion from a putative AGN from the diffuse, extended radio
emission linked to a starburst, as well as potentially allow-
ing the detection of other compact sources, e.g., individual
12 kpc 7 kpc 19 kpc
supernovaeorsupernovaremnants(SNR).Furthermore,we
IC 883 CGCG 049-057 NGC 6240
excludedwarm LIRGswithIRAS color f /f >0.2topre-
25 60
ventcontaminationfromobscuredAGNactivity(e.g., Far-
rahetal.2007),exceptforArp299(f /f =0.22),whichis
25 60
borderline,butwasincludedinthestudysinceitisoneofthe
mostluminouslocalLIRGs.Atotalof54outof629sources
intheIRBGSsatisfytheconditionsabove,fromwhichours
15 kpc 5 kpc 18 kpc
isarepresentativesample.Finally,wealsoexcludedgalaxies
with no nearby reference star to guide the adaptive optics IRAS 16516-0948 IRAS 17138-1017 IRAS 17578-0400
(AO) observations with the VLT or the Gemini telescopes.
In Table 1 we show our final sample, which consists of
11 LIRGs (although the two components of one of them,
Arp 299, are treated separately), all of them included in
theGOALSsample(Armusetal.2009),togetherwiththeir
IR luminosity and luminosity distance. We also include the 11 kpc 12 kpc 7 kpc
merger stage as derived from IRAC 3.6µm morphology IRAS 18293-3413 NGC 6926
(Stierwalt et al. 2013), and the q-factor values (Helou et al.
1985),obtainedusingtheIRAS fluxesfromtheIRBGSand
the1.4GHzfluxesfromtheNRAOVLASkySurvey(NVSS;
Condonetal.1998).Anopticalimageofeachsourceinour
sampleisshowninFigure1.InFigure2weshowthecorre-
lation between L and the radio luminosity at 1.4GHz for
IR 10 kpc 11 kpc
thesourcesinoursample(plottedwithastarsymbol)com-
pared with the resulting sources of a cross-match between
the IRBGS and the New VLA Sky Survey by Yun et al. Figure 1.Opticalcolorcompositeimagesofthegalaxiesinour
(2001). LIRG sample. Images are from the Sloan Digital Sky Survey
(SDSS)DR9(Ahnetal.2012)whenavailable,orfromtheSTScI
A short individual description of each source, some of
DigitizedSkySurvey(DSS)otherwise.
which have been barely studied, is shown below.
24
2.1 MCG+08-11-002 320
Thisgalaxy,alsonamedIRAS05368+4940,isabarredspiral 280
galaxy (type SBab) with a complex morphology, in a late )23¢
z
stage of merging (Stierwalt et al. 2013; Davies et al. 2016). H 240
/
Its mid-IR extended emission is clearly silicate dominated W
(D´ıaz-Santosetal.2011).Daviesetal.(2016)foundevidence /( 22 200pc)
ofapossibleprecedingstarburstepisode,likelylinkedtothe GHz 160(M
previous encounter of the galaxy nuclei.
1.421 D
L 120
¡
(
g
2.2 Arp299 o 80
l20
Arp299isoneofthemostluminousLIRGsinthelocalUni- 40
verse,andforthatreasononeofthemoststudiedsystems.It
is in an early merger stage according to Keel & Wu (1995) 19
8 9 10 11 12 13
or in a mid-stage according to Stierwalt et al. (2013) and log(L /L )
IR
Larson et al. (2016). Arp299 is formed by two galaxies and fl
exhibits two clear radio nuclei (Gehrz et al. 1983), A and
B, and two secondary components, C and C(cid:48). The remain- Figure 2. Radio (1.4GHz) vs. infrared (8−1000µm) luminos-
ing compact structure, D, is believed to be a background ity plotted for a sample of IRAS galaxies (dots) and our galaxy
quasar, unrelated to the system (Ulvestad 2009). Each ra- sample(stars),whichfallsintotheLIRGregime.Themostradio-
diosourceisidentifiedinFigure3.Severalsupernovaehave powerful galaxy in our sample, NGC6240, is also the one that
been recently discovered in the inter nuclear region (Ry- deviatesmostfromthecorrelation.
der et al. 2010; Mattila et al. 2010; Herrero-Illana et al.
2012b; Romero-Can˜izales et al. 2014; Kankare et al. 2014)
MNRAS000,1–19(2015)
4 Herrero-Illana et al.
Table 1.Galaxysampleproperties.
Name RA Dec. DL Mergerstage2 log(LIR) log(L1.4GHz) q-factor3 X-ray
(J2000) (J2000) (Mpc) (L(cid:12)) (ergs−1Hz−1) classification4
MCG+08-11-002 054043.7 +494141 77.2 d 11.41 29.56 2.61 ···
Arp2991 112829.8 +583343 47.7 c 11.88 30.27 2.30 AGN
ESO440-IG058 120651.9 −315654 100.5 b 11.36 29.80 2.32 ···
IC883 132035.3 +340822 100.0 d 11.67 30.10 2.34 AGN
CGCG049-057 151313.1 +071332 59.1 N 11.27 29.35 2.74 NoAGN
NGC6240 165258.9 +022403 103.9 d 11.85 30.74 1.83 AGN
IRAS16516-0948 165424.0 −095321 96.9 d 11.24 29.91 2.08 ···
IRAS17138-1017 171635.8 −102039 75.9 d 11.42 29.66 2.47 ···
IRAS17578-0400 180031.9 −040053 58.6 b 11.35 29.52 2.64 NoAGN
IRAS18293-3413 183241.1 −341127 77.8 c 11.81 30.21 2.33 NoAGN
NGC6926 203306.1 −020139 81.9 d 11.26 29.98 1.97 AGN
IRluminositiesandluminositydistanceswereobtainedfromSandersetal.(2003),whileradio1.4GHzluminositiescomefromCondon
etal.(1998).
1 IncludingbothNGC3690WandNGC3690E.
2 FromthevisualinspectionoftheSpitzer-IRAC3.6µmimages(Stierwaltetal.2013).Thecodeisasfollows:(N)nonmerger;(a)
pre-merger;(b)early-stagemerger;(c)mid-stagemerger;(d)latestagemerger.
3 Seeequation1forthedefinitionofq-factor(Helouetal.1985).
4 Whennotstated,X-raydataarenotavailableorinconclusive.Seeindividualdescriptionofthegalaxiesforthecorresponding
references.
2.3 ESO440-IG058
C’
ThisLIRG,alsoknownasIRAS12042-3140,consistsoftwo
A C merging galaxies separated by (cid:39) 6kpc. The northern com-
ponent is very compact and has been classified as a LINER
(Corbett et al. 2003). While the northern galaxy is domi-
nated by star formation, the southern emission appears to
bedominatedbyshocks(v=100−150kms−1;Monreal-Ibero
et al. 2010).
B
D Based on the IR luminosity, Miluzio et al. (2013) es-
timated that ESO440-IG058 has a star formation rate
of 36M(cid:12)yr−1 and an expected SN rate of 0.4SNyr−1.
Rodr´ıguez-Zaur´ın et al. (2011) derived an age of the stel-
lar population of t (cid:46)6.5Myr for the currently star forming
Figure 3.VLA8.4GHzimageshowingtheradiosourcesinthe stellar population.
Arp 299 system. Nuclei A and B are the main components of
NGC3690EastandNGC3690West,respectively.Furtherdetails
ontheimagecanbefoundinsection3.1
. 2.4 IC883
IC883, also known as UGC8387, is a late-merger LIRG at
adistanceof100Mpc,showingapeculiarmorphology,with
and in nucleus B (Ulvestad 2009; Romero-Can˜izales et al. extended perpendicular tidal tails visible in the optical and
2011,P´erez-Torresetal.,inprep.).However,itistheAnu- near-IR(Smithetal.1995;Scovilleetal.2000;Modicaetal.
cleus that hosts a very rich supernova factory (Neff et al. 2012). This source was classified as an AGN/SB composite
2004; P´erez-Torres et al. 2009; Ulvestad 2009; Bondi et al. (Yuan et al. 2010), which has been confirmed by the direct
2012; Herrero-Illana et al. 2012a), as well as a low lumi- detectionofanumberofradiocomponentesthatareconsis-
nosityAGN,suggestedfromX-rayobservations(DellaCeca tent with an AGN and with SNe/SNRs (Romero-Can˜izales
et al. 2002; Zezas et al. 2003; Ballo et al. 2004), and found et al. 2012b) and with the detection, in the near-IR, of two
by means of VLBI observations (P´erez-Torres et al. 2010). SNe (Kankare et al. 2012) within the innermost nuclear re-
SED modeling for this system yielded a star formation rate gion of the galaxy. Through SED model fitting, Romero-
(SFR) of 90 and 56M(cid:12)yr−1 for the east and west compo- Can˜izalesetal.(2012b)estimatedacorecollapsesupernova
nents, respectively (Mattila et al. 2012). (CCSN)rateof1.1SNyr−1andaSFRof185M(cid:12)yr−1.IC883
Whiletherearedifferentnomenclaturestonamethetwo alsopresentsstrongPAHemission,silicateabsorptionanda
galacticcomponentsofArp299,somehaveproducedconfu- steep spectrum beyond 20µm (Vega et al. 2008). Recently,
sion in the literature (Corwin 2004). To avoid that, we use using radio VLBI and X-ray data, Romero-Can˜izales et al.
theunequivocaldesignationsNGC3690EastandNGC3690 (2017)havereportedunequivocalevidenceofAGNactivity,
West for these components, which are treated individually with the nucleus showing a core-jet structure and the jet
throughout this paper. having subluminal proper-motion.
MNRAS000,1–19(2015)
Star formation and nuclear activity in local LIRGs 5
2.5 CGCG049-057 2.10 IRAS18293-3413
Also known as IRAS15107+0724, this is the only LIRG in This source was classified as an Hii galaxy based onits op-
our sample classified as isolated. However, despite its ap- ticalspectrum(Veilleuxetal.1995).Itwasdetectedinhard
parent isolation (Larson et al. 2016), it has a complex and X-ray data (2–10keV), obtained with ASCA, at a 5σ level
dusty nuclear morphology, so some form of past interaction byRisalitietal.(2000),findingnoevidenceofanyAGNcon-
cannot be ruled out. It hosts an OH megamaser (Bottinelli tributiontotheX-rayspectrum.Therearediscrepancieson
et al. 1986; Baan et al. 1987). It is optically classified as a themergerstagedetermination.Stierwaltetal.(2013)clas-
pure starburst, supported by Chandra X-ray observations sified it as a mid-stage merger from the visual inspection of
(Lehmer et al. 2010). However multi-band radio observa- Spitzer-IRAC3.6µmimages;ontheotherhand,Haanetal.
tions show evidence of a buried AGN within the SB (Baan (2011) classified it as a very early merger, with canonical
& Kl¨ockner 2006). disksandnotidaltailsbasedonitsHST morphology.How-
ever, V¨ais¨anen et al. (2008b), using high-resolution near-IR
K-band adaptive optics imaging with VLT/NACO, showed
2.6 NGC6240 the galaxy to have a very complex morphology, strongly
suggesting a late stage interaction. The system includes a
ThisbrightLIRGisawell-studiedlate-stagemerger(Stier-
rare un-evolved elliptical companion as well. Using SED
walt et al. 2013). It hosts one of the few binary AGN de-
modeling, Mattila et al. (2007a) estimated a CCSN rate of
tected so far using Chandra hard X-ray observations (Ko- 1SNyr−1forIRAS18293-3413.SupernovaeSN2004ip(Mat-
mossaetal.2003),withaprojecteddistanceof(cid:39)1kpc.This
tila et al. 2007b; P´erez-Torres et al. 2007) and AT2013if
was later supported by the detection of two compact unre-
(Kooletal.inprep.)werediscoveredwithinthenuclearre-
solved sources at radio wavelengths with inverted spectral
gions of this galaxy.
indices(α=+1.0andα=+3.6forthenorthandsouthcom-
ponentrespectively;seeGallimore&Beswick2004),finding
alsoathirdcomponentwithaspectralindexconsistentwith 2.11 NGC6926
a radio supernova. Close to the southern nucleus, a water-
This relatively low luminosity LIRG is a spiral galaxy in
vapor megamaser was found (Nakai et al. 2002; Sato et al.
a very early phase of interaction with the dwarf elliptical
2005).
NGC6929, located 4(cid:48) to the east. Optically identified as a
Seyfert 2 (Veilleux et al. 1995), NGC6926 has a powerful
water-vapor megamaser (Greenhill et al. 2003; Sato et al.
2.7 IRAS16516-0948
2005),typicallyfoundinheavilyobscuredAGN(e.g.Green-
This unexplored LIRG was optically identified as a star hilletal.2003;Masinietal.2016).ItsX-rayhardnessratio
forming galaxy (Mauch & Sadler 2007), and classified as (Terashima et al. 2015) is also suggestive of an AGN. This
a late merger from its infrared morphology (Stierwalt et al. is the only source for which we lack near-IR observations.
2013). IRAS16516-0948 was part of the COLA project, for
whichhigh-resolutionradioimagingdidnotdetectanycom-
pact core (Corbett et al. 2002). Despite the scarce informa-
3 OBSERVATIONS AND DATA REDUCTION
tionavailableforthisgalaxy,wefoundittobeaninteresting
source, as discussed in section 4.3. The observational data used in this study comes from (1)
multi-wavelengtharchivaldataobtainedthroughtheVizieR
photometrytool(usedfortheSEDmodelfits)and(2)from
2.8 IRAS17138-1017 ourownobservationsatbothradioandnear-IRwavelengths.
Table2showsasummaryoftheseobservations.Theimages
This LIRG is a highly obscured starburst galaxy (Depoy
are shown in Figure 4 and discussed in section 4.2.
et al. 1988) in a late stage of interaction. An extremely
extinguished supernova (AV = 15.7±0.8mag) was discov-
eredinIRAS17138-1017usinginfraredK-bandobservations 3.1 Radio
(SN2008cs;Kankareetal.2008b)Twomoresupernovaewere
The radio data used in this paper were X-band (8.4GHz,
also found: SN2002bw (Li 2002; Matheson et al. 2002) and
3.6cm) observations in full polarization mode and with a
SN2004iq (Kankare et al. 2008a).
total bandwidth of 2048MHz (project 12B-105; PI: M. A´.
P´erez-Torres) taken between 5 October and 26 December
2012 using the new VLA capabilities. The total on-source
2.9 IRAS17578-0400
time for each target was approximately 30min. For every
This is a galaxy pair in an early stage of merging (Stier- sourceweused3C286asfluxandbandpasscalibrator,while
walt et al. 2013). Ultra-hard X-ray (14–195keV) observa- we observed for each case a nearby bright point-like source
tionswiththeSwift BurstAlertTelescope(BAT)searching for phase calibration purposes. We used the Common As-
forAGNdidnotdetectanycompactemissioninthissource tronomy Software Applications (CASA, McMullin et al.
(Koss et al. 2013). There is abundant archival data in the 2007) package for data reduction purposes, which consisted
X-ray,optical,IR,andmillimeterbandsforthissource(see, ofstandardamplitudeandphasecalibration.Weimagedthe
e.g., Lu et al. 2014; Stierwalt et al. 2014; Rich et al. 2015), sourcesusingaBriggsweightingschemewithROBUST = 0.5
althoughnosignificantresultsforthepurposesofthispaper toobtainthebestcompromisebetweensensitivityandreso-
have been reported in the literature. lution.Table2showsthebeamsize,noiseachievedandpeak
MNRAS000,1–19(2015)
6 Herrero-Illana et al.
44.0" MCG+08-11-002 4.00
43.0"
2.00
42.0"
1.00
41.0"
0.50
+49°41'40.0"
418 pc
0.25
43.86s 43.68s 5h40m43.50s 43.86s 43.68s 5h40m43.50s 43.86s 43.68s 5h40m43.50s
4.00
NGC3690 W
51.0"
2.00
48.0"
45.0" 1.00
42.0" 0.50
+58°33'39.0" 237 pc
0.25
31.80s 31.44s 31.08s 30.72s11h28m30.36s31.80s 31.44s 31.08s 30.72s11h28m30.36s31.80s 31.44s 31.08s 30.72s11h28m30.36s
4.00
NGC3690 E
50.0"
2.00
48.0"
1.00
46.0"
44.0" 0.50
+58°33'42.0" 237 pc
0.25
34.26s 34.00s 33.74s 33.48s11h28m33.22s34.26s 34.00s 33.74s 33.48s11h28m33.22s34.26s 34.00s 33.74s 33.48s11h28m33.22s
4.00
ESO 440-IG058
-31°56'45.0"
2.00
50.0"
1.00
55.0"
57'00.0" 0.50
05.0" 534 pc
0.25
52.60s 52.22s 51.84s 51.46s12h06m51.08s 52.60s 52.22s 51.84s 51.46s12h06m51.08s 52.60s 52.22s 51.84s 51.46s12h06m51.08s
4.00
26.0" IC883
2.00
24.0"
1.00
22.0"
0.50
+34°08'20.0"
524 pc
0.25
35.52s 35.34s 13h20m35.16s 35.52s 35.34s 13h20m35.16s 35.52s 35.34s 13h20m35.16s
Figure 4.Radioandnear-IRcomparisonofoursample.Theimageshowsthe8.4GHz(3.6cm)radiomaps(left),the2.2µmIRmaps
(middle) and the ratio between radio and IR (right), where redder colors imply radio dominated regions in contrast with bluer colors.
(cid:16)√ (cid:17)n
Contoursaredrawnevery3 3 ×rmslevel,withn=0,1,2,3...Adirectcomparisoninasingleplotbetweenradioandnear-IRisshown
inFigureA1. MNRAS000,1–19(2015)
Star formation and nuclear activity in local LIRGs 7
4.00
CGCG 049-057
36.0"
2.00
34.0"
32.0" 1.00
30.0"
0.50
+7°13'28.0"
295 pc
0.25
13.26s 13.08s 15h13m12.90s 13.26s 13.08s 15h13m12.90s 13.26s 13.08s 15h13m12.90s
4.00
NGC6240
06.0"
2.00
05.0"
04.0" 1.00
03.0"
0.50
02.0"
503 pc
+2°24'01.0" 0.25
59.04s 58.90s 16h52m58.76s 59.04s 58.90s 16h52m58.76s 59.04s 58.90s 16h52m58.76s
4.00
IRAS 16516-0948
-9°53'14.0"
16.0" 2.00
18.0"
1.00
20.0"
22.0" 0.50
24.0" 496 pc
0.25
24.32s 24.10s 23.88s 16h54m23.66s 24.32s 24.10s 23.88s 16h54m23.66s 24.32s 24.10s 23.88s 16h54m23.66s
4.00
IRAS 17138-1017
-10°20'36.0"
38.0" 2.00
40.0"
1.00
42.0"
0.50
44.0"
378 pc
0.25
36.12s 35.94s 35.76s 17h16m35.58s36.12s 35.94s 35.76s 17h16m35.58s 36.12s 35.94s 35.76s 17h16m35.58s
4.00
IRAS17578-0400
-4°00'50.0"
2.00
52.0"
1.00
54.0"
56.0" 0.50
58.0" 300 pc
0.25
32.04s 31.86s 18h00m31.68s 32.04s 31.86s 18h00m31.68s 32.04s 31.86s 18h00m31.68s
Figure 4–continued
MNRAS000,1–19(2015)
8 Herrero-Illana et al.
4.00
IRAS18293-3413
-34°11'18.0"
2.00
24.0"
1.00
30.0"
0.50
36.0"
391 pc
42.0" 0.25
42.24s 41.76s 41.28s 40.80s18h32m40.32s42.24s 41.76s 41.28s 40.80s18h32m40.32s 42.24s 41.76s 41.28s 40.80s18h32m40.32s
NGC6926
-2°01'36.0"
38.0"
40.0"
42.0" 429 pc
06.24s 06.08s 20h33m05.92s
Figure 4–continued
Table 2.Observationssummary.
Radio(8.4GHzor3.6cm) Near-IR(2.2µm)
Name FWHM rms Peak Int.flux Pixelsize rms Peak Int.flux
(arcsec) (µJybeam−1) (mJybeam−1) (mJy) (arcsec) (µJy) (mJy) (mJy)
MCG+08-11-002 0.26×0.21 8.64 0.73 16.68 0.022 0.06 0.006 21.27
NGC3690W 0.24×0.20 12.59 8.85 82.20 0.022 0.01 0.404 31.61
NGC3690E 0.24×0.20 16.91 24.85 22.13 0.022 0.03 0.017 93.90
ESO440-IG058 0.60×0.18 8.12 1.07 6.39 0.02 0.007 0.006 21.07
IC883 0.27×0.21 13.52 10.51 34.06 0.022 0.02 0.006 17.09
CGCG049-057 0.24×0.21 23.77 19.73 26.94 0.022 0.02 0.002 17.89
NGC6240 0.27×0.20 29.26 19.70 52.86 0.027 0.008 0.078 36.67
IRAS16516-0948 0.33×0.20 8.28 0.09 0.27 0.022 0.006 0.005 10.84
IRAS17138-1017 0.33×0.20 7.61 0.68 8.79 0.022 0.03 0.004 47.42
IRAS17578-0400 0.55×0.21 9.66 17.21 23.24 0.022 0.005 0.002 22.17
IRAS18293-3413 0.60×0.17 11.64 0.83 25.30 0.027 0.001 0.039 101.27
NGC6926 0.25×0.19 7.26 3.99 6.00 ··· ··· ··· ···
Theangularresolutionforthenear-IRimagesisbetween0.07(cid:48)(cid:48) (diffraction-limitedFWHMatK-band)and0.10(cid:48)(cid:48).ESO440-IG058was
observedwithGeMS/GSAOI,NGC6240andIRAS18293-3413withNACO,andtheremainingsourceswithALTAIR/NIRI.Integrated
fluxesarequotedabove5σ.
flux density for each image. To compare these images with (nterms = 2) of the function:
our near-IR data, we converted the flux density units from
Jybeam−1 to Jypx−1. Isky=Isky(cid:18) ν (cid:19)α, (3)
ν ν0 ν
0
Although our data were taken in a single band, the
where Isky is the multiscale image, ν is the reference fre-
large bandwidth (2GHz) VLA X-band receivers allowed us 0
to estimate the spectral index, α (Sν ∝ να), through this quency (in our case ν0 = 8.459GHz), and α is the spectral
index.
radio band. To this end, we imaged our sources using mul-
The two output images are the first two coefficients of
tiscalemultifrequencysynthesis(mode = mfswithinCASA;
the expansion, i.e.:
seeRau&Cornwell2011),whichmodelsthewide-bandsky
bwrhigohsetnaemsspalistaudlienseaforllcoowmbainTaatyiolonr-opfoGlyanuosmsiaianl-liinkefrfeuqnucetinocnys. I0=Iν0 I1=Iν0α. (4)
The task performs a Taylor expansion to the second order Wethenobtainedtheradiospectralindexmapofeach
MNRAS000,1–19(2015)
Star formation and nuclear activity in local LIRGs 9
source by simply using the ratio I /I . These maps are dis- often depend on the individual environment and star form-
1 0
cussed in section 4.4. ing history of each galaxy. A more reliable method is to
takeadvantageofmulti-wavelengthobservationsandfitthe
SED to combined templates of starbursts and AGN to de-
3.2 Near-IR data rive the relevant parameters (AGN contribution, starburst
age,starformationrateandsupernovarate).Inthissection,
We used ALTAIR/NIRI and GeMS/GSAOI on the Gem-
we obtain these parameters and compare them with those
ini North and South telescopes, respectively (PI: S. Ry-
obtained with traditional tracers.
der), and NACO on the VLT (PIs: S. Mattila) to ob-
We modeled the multi-wavelength SED for the sources
tain near-IR K-band (2.2µm) laser guide star adaptive op-
in our sample. To this end, we combined libraries of star-
tics (AO) images. These instruments have a pixel scale of
0.022(cid:48)(cid:48)px−1 (ALTAIR/NIRI), 0.02(cid:48)(cid:48)px−1 (GeMS/GSAOI), burst models (Efstathiou et al. 2000; Efstathiou & Sieben-
and 0.027(cid:48)(cid:48)px−1 (NACO, with camera S27). IRAS18293- morgen 2009), AGN torus models (Efstathiou & Rowan-
Robinson 1995; Efstathiou et al. 2013) and models of the
3413 and NGC6240 were observed using NACO, ESO440-
spheroidal/cirrus component (Efstathiou et al., in prep.).
IG058usingGeMS/GSAOI,andtheremainingtargetswere
ThelatterisbasedoncirrusmodelsbyEfstathiou&Rowan-
observed with ALTAIR/NIRI. Data from ALTAIR/NIRI
Robinson (2003) that rely on calculations of the radiative
come from a multi-epoch survey intended for the detection
transfer problem in a medium where dust and stars are
andstudyofnuclearcore-collapsesupernovae(CCSNe).The
mixedinaspheroidaldistribution.WeusedaMarkovChain
ALTAIR/NIRI images used in this paper are a combina-
MonteCarlo(MCMC)fittingcode(SATMC; Johnsonetal.
tion of data from 2008 to 2012. NACO observations were
2013)toobtainrealisticuncertaintiesonthefittedparame-
performed on 13 September 2004 (IRAS18293-3414) and
ters. SATMC is based on Bayesian statistics, resulting in
31 May 2011 (NGC6240), and GeMS/GSAOI observations
more reliable results and error determination than tradi-
for ESO440-IG058 were performed on 05 March 2015. The
tionalleast-squarefitting.FordetailsontheSATMCfitting
total integration time spans from 990s (for NGC6240) to
process, we refer the reader to Johnson et al. (2013).
2192s (for NGC3690W). No near-IR data are available for
NGC6926.Spatialresolutionwastypically(cid:39)0.1(cid:48)(cid:48)intheAO The SFR of the starburst and its errors were derived
corrected images. self-consistentlyfromtheradiativetransfermodelswhichin-
We reduced the near-IR data using standard IRAF- corporate the stellar population synthesis model of Bruzual
based tasks, including flat-fielding and sky subtraction, & Charlot (1993, 2003). The libraries for a Salpeter initial
and created the final images by average-combining indi- massfunction(IMF),solarmetallicityandstarsintherange
vidual frames from different epochs after having shifted 0.1–125M(cid:12) were used. The models of Efstathiou et al. pre-
them to a common reference. A detailed description of the dicttheSEDofastarburstatdifferentagest∗ assumingthe
reduction process of the ALTAIR/NIRI and NACO/VLT star formation rate declines exponentially with an e-folding
data is presented in Randriamanakoto et al. (2013b), while time τ.FromtheSATMCfitweknowboththebestfitval-
GeMS/GSAOIdataarepartoftheSUNBIRDsurvey(Kool ues of t∗ and τ and their 1σ uncertainties. Additionally, we
et al. 2016, in prep.). In multi-epoch data, we found no ev- knowthefittedstarburstluminosity LSB andtheassociated
idence of variability among the different individual epochs uncertainties. The combination of LSB, t∗, τ and the tables
above 2-σ. ofBruzual&Charlot(1993,2003)uniquelydeterminedthe
total mass of stars formed in the starburst episode for each
combination. The stellar mass is divided by 50Myr to give
3.3 Astrometry calibration and image convolution the mean SFR that we provide.
We obtained the SED data points from public data
Since the Field of View (FOV) of the near-IR images was
throughtheVizieRphotometrytool,whichincludednear-to
small (between 22 and 54arcsec on a side), we first cali-
far-IRphotometry(2MASS,IRAC,WISE,AKARI,IRAS),
brated larger FOV archival data from NOT or HST/ACS
as well as UV (GALEX), optical (SDSS), and (sub-)mm
usingstarsfromGuideStarCatalogue-2or2MASSK -band
S (SCUBA)dataforsomeofthesources(seephotometricdata
catalogue.WethenaddedtheWCSinformationintheFITS
points in Figure 5). All the photometric data we are using
header of our targets using these intermediate images (see
refer to the whole galaxy, either because the PSF was big
Randriamanakotoetal.2013b,fordetails).Weestimatethe
enough (far-IR data) or because we used integrated fluxes.
astrometry calibration uncertainty to be (cid:39)0.15(cid:48)(cid:48).
Inaddition,weincludedtheSpitzer IRSspectrawhenavail-
Given the significantly different resolution of the radio
able. For these, and to minimize aperture effects, we scaled
andnear-IRimages(seeTable2),weconvolvedthenear-IR
uptheirfluxesinsuchawaythattheedgesofeachspectrum
images with a Gaussian with the size of the radio beam for
matchthefluxoftheadjacentphotometricdatapoints.The
eachcase,andthenrebinnedthenear-IRimagestothepixel
spectral resolution of the IRS data is reduced so that they
size used in our radio images (0.04(cid:48)(cid:48)).
arebettermatchedtotheresolutionoftheradiativetransfer
models. The IRS data included in the fitting with SATMC
have a wavelength grid which is separated in steps of 0.05
4 RESULTS AND DISCUSSION in the log of rest wavelength. We additionally added more
points around the 9.7µm silicate feature and the PAH fea-
4.1 SED modeling
tures. There was no spectrum available for CGCG049-057,
There is a wide variety of methods, based on data at dif- andwedidnotuseIRSdataforthefittingofArp299either,
ferent wavelengths, to determine the starburst and AGN since we are fitting both components together.
properties of galaxies. These are not always consistent and WeshowinTable3thebestfitparameters,andinFig-
MNRAS000,1–19(2015)
10 Herrero-Illana et al.
ure 5 show the fitted SED for ten sources. The CCSN rates inagreementwiththeabove,whileitrangesbetween42and
are estimated by convolving the SN rate at a given time 89% for the rest of the sources.
with the star formation history of the starburst (see details In the following sections we compare the results from
inMattilaetal.2012).ForconsistencywemodeledArp299 ourSEDmodelsdescribedabovewithotherproxiesandin-
(NGC3690E+NGC3690W)asauniquegalaxybutwenote dicators of both SFR and AGN.
that a similar, yet not as complete model was published
in Mattila et al. (2012) for the individual components. We
4.1.1 Comparison with SFR indicators
note a discrepancy in the SB ages of NGC3690E (45Myr)
andNGC3690W(55Myr)obtainedbyMattilaetal.(2012) Table3quotestheSFRobtainedfromthemulti-wavelength
andthosederivedhereandalsointhemodelingbyAlonso- SEDfitofoursources,averagedoverthepast50Myr.How-
Herrero et al. (2000) using the evolutionary synthesis mod- ever, there is a large number of methods to indirectly esti-
els by Rieke et al. (1993) and Engelbracht et al. (1996), mate SFR in galaxies. From these prescriptions, only those
whichyieldsignificantlylowerages(6−8Myrand4.5−7Myr thatarenotstronglyaffectedbyextinction(suchasIRand
forNGC3690EandNGC3690W,respectively).However,we radio tracers) are of real use in the dusty environments of
find the models not comparable since they are considering LIRGs.InFigure6weshowacomparisonbetweenourSED
differentIMF,differentstarformationhistory,and,specially, modeling and three of these methods, described below:
different apertures: the whole system in this study, versus
each component in Mattila et al. (2012) or the innermost (i) The outcome of massive stars heating the interstel-
< 5arcsec region in Alonso-Herrero et al. (2000) We also lar dust is an intense infrared flux re-emitted by the dust
note that the higher SFRs quoted in Mattila et al. (2012) through black-body radiation. The following relation (Ken-
wereaveragedoverthewholedurationofthestarburst,while nicutt1998)isusedtoconnecttheinfraredluminositywith
in this study we are averaging it over 50Myr. Nonetheless, thestarformationrateincaseofstarformation<100Myrof
the SN rate between both studies are compatible. age,whichissatisfiedinthecaseswepresentinthispaper:
SFR=4.5×10−44L , (5)
Our modeling allowed us to quantify the luminosity IR
contribution of each component: the starburst, the AGN, with SFR in M(cid:12)yr−1 and LIR, measured between 8 and
and the spheroid, i.e., the underlying host galaxy stellar 1000µm, in ergs−1, where we use the values quoted in Ta-
population minus the current starburst. We find that the ble 1.
SED of all sources in our sample is starburst-dominated (ii) ThereisatightcorrelationbetweentheIRandradio
except for NGC6926 (AGN contribution of (cid:39) 64%), with luminosities (see section 1). Thus, radio luminosity is ex-
two additional sources having a significant AGN contribu- pectedtotraceSFRinasimilarwayasL does.Acommon
IR
tion((cid:38)20%).OfthesevensourceswithanX-rayclassifica- prescription used to derive SFR from radio data (typically
tion (see Table 1), four of them are catalogued as AGN: NVSS)isgivenbyMurphyetal.(2011),andadaptingitfor
NGC6926 and NGC6240 show an important AGN con- a Salpeter IMF (as for the rest of the SFR indicators), we
tribution in our analysis, although Arp299 and IC883 do have:
not. However, we note that these two sources are known to
host an AGN. We also derive star formation rates (40 to SFR=1.02×10−28L1.4GHz, (6)
167M(cid:12)yr−1), supernova rates (0.3 to 2.0SNyr−1), and star- where SFR is given in M(cid:12)yr−1 and L1.4GHz in ergs−1Hz−1,
burstages(23to55Myr)thatareconsistentwiththeirLIRG
obtained from NVSS (Condon et al. 1998). The luminosity
nature, with the exception of NGC6926, a young starburst
atthisfrequencyisessentiallyduetonon-thermalemission.
(9Myr)wheresupernovaeventshavenotyetbeentriggered.
(iii) Randriamanakoto et al. (2013a) established a rela-
TheoldeststarburstofoursamplecorrespondstoMCG+08-
tionbetweenthenear-IRK-bandmagnitudeofthebrightest
11-002 at (cid:39)29Myr, while the youngest one is NGC6926 at
SSCinagalaxyanditsglobalSFR.ForthelocalUniverse,
just (cid:39)9Myr.
this relation can be expressed as:
Three of our sources show a remarkably high CCSN Mbrightest=−2.56log(SFR)−13.39. (7)
rate (which depends on both SFR and age) above k
1.5SNyr−1:IRAS18293-3413,with1.97SNyr−1,NGC6240 TherangebetweenhighestandlowestestimatesofSFR
with 1.80SNyr−1, and IC883 with 1.78SNyr−1, the latter (i.e. SFRmax−SFRmin) of the different tracers shown in Fig-
consistent with previous studies (ν = 1.1+1.3, Romero- ure 6 is approximately similar, with a median scatter of
CCSN −0.6
Can˜izales et al. 2012b). The extreme CCSN rate and SFR 78.1M(cid:12)yr−1,withonlyonesource,NGC6240,deviatingbe-
of IRAS18293-3413, together with the controversy on its yond the standard deviation, for which the SFR derivation
mergerstageanditsextremelysteepspectrum(α=−1.73± throughitsradioluminosityishighlyoverestimated.Thera-
0.70,seesection4.4),turnsitintoaninterestingcase,likely dio luminosity tracer is indeed known to over-estimate the
hostingaveryrichstarburst.Thisissupportedbythedetec- SFRinthepresenceofstrongAGNactivity(DelMoroetal.
tionofhundredsofsuperstarclusters(SSCs)inthefieldof 2013; Bonzini et al. 2015), which is also supported by the
IRAS18293-3413 (Randriamanakoto et al. 2013b). We note large SFR derived from the radio luminosity in NGC6926,
however, the lower SN rate derived with a previous simpler withanAGNcontributionof(cid:39)64%.Inthissense,Figure7
model(1SNyr−1;Mattilaetal.2007a).Bycontrast,ourfit showstheabsoluteandrelativescatteroftheSFRdetermi-
forNGC6926ismodeledbysuchayoungstarburst(9Myr) nation with respect to the relative AGN contribution. We
thatnoSNeventhasyetbeenobservedfromitscurrentstar suggest the existence of a weak trend, where more AGN
formation episode. The ratio of the SB luminosity to L dominated sources present a larger scatter in the SFR de-
IR
fromTable1(Sandersetal.2003)isonly5%forNGC6926, termination from different proxies.
MNRAS000,1–19(2015)