A Massive Molecular Gas Reservoir in the z=2.221 Type-2 Quasar Host Galaxy SMM J0939+8315 Lensed by the Radio Galaxy 3C220.3

1.2 MB·

by  T. K. Daisy Leung

Accepted to the ApJ
PreprinttypesetusingLATEXstyleemulateapjv.5/2/11
A MASSIVE MOLECULAR GAS RESERVOIR IN THE Z=2.221 TYPE-2 QUASAR HOST GALAXY
SMMJ0939+8315 LENSED BY THE RADIO GALAXY 3C220.3
T. K. Daisy Leung and Dominik A. Riechers
DepartmentofAstronomy,SpaceSciencesBuilding,CornellUniversity,Ithaca,NY14853,USA;[email protected]
Accepted to the ApJ
ABSTRACT
We report the detection of CO(J=3→2) line emission in the strongly-lensed submillimeter galaxy
(SMG) SMMJ0939+8315 at z = 2.221, using the Combined Array for Research in Millimeter-wave
6
Astronomy. SMMJ0939+8315 hosts a type-2 quasar, and is gravitationally lensed by the radio
1
galaxy 3C220.3 and its companion galaxy at z = 0.685. The 104 GHz continuum emission un-
0
derlying the CO line is detected toward 3C220.3 with an integrated flux density of S = 7.4±1.4
2 cont
mJy. Using the CO(J=3→2) line intensity of ICO(3-2) = (12.6±2.0)Jykms−1, we derive a lensing-
n
and excitation-corrected CO line luminosity of L(cid:48) = (3.4±0.7)×1010(10.1/µ )K km s−1 pc2
a CO(1-0) L
J for the SMG, where µL is the lensing magnification factor inferred from our lens modeling. This
translates to a molecular gas mass of M = (2.7±0.6)×1010(10.1/µ )M . Fitting spectral en-
2 gas L (cid:12)
2 ergy distribution models to the (sub)-millimeter data of this SMG yields a dust temperature of
T =63.1+1.1K, a dust mass of M = (5.2±2.1)×108(10.1/µ )M , and a total infrared lumi-
−1.3 dust L (cid:12)
] nosityofL =(9.1±1.2)×1012(10.1/µ )L . Wefindthatthepropertiesoftheinterstellarmedium
A IR L (cid:12)
of SMM J0939+8315 overlap with both SMGs and type-2 quasars. Hence, SMM J0939+8315 may
G be transitioning from a star-bursting phase to an unobscured quasar phase as described by the “evo-
. lutionary link” model, according to which this system may represent an intermediate stage in the
h
evolution of present-day galaxies at an earlier epoch.
p
Subjectheadings: cosmology: observations—galaxies: evolution—galaxies: high-redshift—galaxies:
-
o starburst — submillimeter: galaxies
r
t
s
a 1. INTRODUCTION SMGs that are gravitationally lensed, as lensing ampli-
[ fies the intrinsic fluxes of these sources, making them
Submillimeter-selected galaxies (SMGs) are predomi-
the brightest unveiled in large sky surveys (Negrello
2 nantlyfoundatredshiftsz∼1–3(Chapmanetal.2005),
et al. 2010; Vieira et al. 2010; Oliver et al. 2012), and
v during the epoch of stellar mass and galaxy assembly,
making follow-up studies considerably less time consum-
2 with a tail out to z > 6 (Riechers et al. 2013). Previous
ing. A particularly interesting and peculiar lensing sys-
7 works have shown that SMGs are extremely luminous in
1 the infrared wavelengths (L ∼ 1012 L ) with high star tem was discovered serendipitously in a study carried
IR (cid:12) out with the Herschel Space Observatory, in which a
4 formationrates(SFR(cid:38)500M yr−1;seee.g.,reviewsby
0 Blainetal.2002;Lagacheetal(cid:12).2005;Caseyetal.2014). type-2 quasar host SMG — SMMJ0939+8315 (here-
. after SMMJ0939) is being lensed by the double-lobed
1 Following the pioneering works in the discovery of this
Fanaroff-Riley Class II (FR-II; Fanaroff & Riley 1974)
0 population(Smailetal.1997;Hughesetal.1998;Barger
radio galaxy 3C220.3 at z=0.685, which has a compan-
6 etal.1998),considerableamountsofefforthavebeenin-
iongalaxy“B”asdetectedinKeck2.2µmandtheHubble
1 vestedintoobtaininglargesamplesofSMGsbycarrying
Space Telescope 702nm images (Haas et al. 2014, here-
: out large sky surveys with (sub)-mm facilities such as
v after H14). SMMJ0939 is currently one of the brightest
the Herschel Space Observatory (e.g., H-ATLAS, SPT,
i known lensed SMGs, with a lensing-magnified flux den-
X HerMES; Ealesetal.2010;Carlstrometal.2011;Oliver
et al. 2012). sity of S250µm=440±15 mJy. Detections of C IV 1549˚A
ar Tocharacterizethephysicalpropertiesofthegasreser- and He II 1640˚A line emission toward SMMJ0939 place
voirs in the interstellar medium (ISM) where active star the redshift of this galaxy at z=2.221. Based on the
formation takes place, carbon monoxide (12CO) rota- spectrallinefluxesandlinewidths,H14suggestthepres-
enceofanobscuredactivegalacticnucleus(AGN)inthe
tional lines have been commonly used as tracers due to
form of a type-2 quasar in this SMG.
itshighabundanceintheISMaswellasitslowexcitation
In this paper, we present the detection of
energy; the ground state transition line thereby directly
CO(J=3→2) line emission toward the background
probesthecoolgasthatisessentialtofuelstarformation
SMGobtainedwiththeCombinedArrayforResearchin
(seee.g.,reviewsbySolomon&VandenBout2005;Car-
Millimeter Astronomy (CARMA), which confirms and
illi & Walter 2013). Observations of CO in SMGs have
refines the redshift, and permits a study of the physical
demonstrated that these galaxies have large gas reser-
voirs typical of >1010M (e.g., Frayer et al. 1998; Neri conditions in the ISM of SMMJ0939 in great detail.
(cid:12)
We report the detection of the continuum emission
et al. 2003; Riechers et al. 2011a,b; Ivison et al. 2011;
underlying the CO line and place constraints on the
Bothwell et al. 2013).
spectral energy distribution (SED) of the foreground
Many recent detailed studies have been carried out on
FR-II galaxy at millimeter (mm) wavelengths (∼104
2 Leung & Riechers
GHz). Based on the magnification factor derived from flux density is 7.39±1.42 mJy. An overlay image of the
our lens model, we infer various intrinsic properties of 104GHzcontinuumemissionwiththe9GHzcontinuum
SMMJ0939. We conclude this paper by comparing our emission(H14)isshowninFigure1, demonstratingthat
findingstoother similarlybright, strongly-lensedSMGs, thecontinuumemissionismarginallyresolvedattheres-
as well as other type-2 quasars at z∼2−3. olution of our observations. It is therefore plausible that
WeadoptaflatΛCDMcosmologicalmodelthroughout non-thermal emission from the radio lobes and core of
this paper, with H = 69.32 km Mpc−1 s−1, Ω =0.286, the foreground galaxy dominate the integrated flux den-
0 M
Ω =0.713, based on the WMAP9 results (Hinshaw sity of the measured continuum. We discuss this further
Λ
et al. 2013). The luminosity distances at z=0.685 and in Section 4.2.1.
z=2.221 are 4214 Mpc and 18052 Mpc, respectively; 1(cid:48)(cid:48) The frequency range of our observations cov-
corresponds to 7.169 kpc at z=0.685, and 8.406 kpc at ers the HCO+(J=2→1), HNC(J=2→1), and
z=2.221. H O(3 →2 ) transition lines in the foreground
2 13 20
galaxy, at the redshifted frequencies of 105.86, 107.71,
2. OBSERVATIONS and 108.79 GHz, respectively. We establish 3σ up-
2.1. CARMA per limits employing a typical FWHM line width of
∼300 km s−1, based on the CO(J=1→0) line mea-
Observations of the CO(J=3→2) rotational transi-
surements in a sample of local radio galaxies (z < 0.1;
tion (ν =345.7959899 GHz) toward the background
rest Smolˇci´c & Riechers 2011). This results in upper limits
galaxy SMM J0939 (z=2.221) were carried out using
of < 2.66 Jy km s−1 on the integrated emission line
CARMAataredshiftedfrequencyofν =107.357 GHz
obs strengths.
(2.79 mm; program ID: cf0142; PI: Riechers). The
3mm receivers were used to cover the redshifted
3.2. Background Galaxy: SMMJ0939
CO(J=3→2) line and the nearby observed-frame
2.88mm continuum emission. The correlator was con- We detect CO(J=3→2) line emission at ∼8σ sig-
figured to provide an effective bandwidth of 3.708 GHz nificance toward the background SMG SMMJ0939 at
ineachsideband,andaspectralresolutionof5.208MHz z=2.221. The lensing-magnified spatial extent of this
(∼14.5 km s−1). The line was placed in the upper side- SMG is ∼5(cid:48)(cid:48), as shown in the Submillimeter Array
band, with the local oscillator tuned to ν ∼104.2609 (SMA) 1mm dust continuum image in Figure 2 (H14);
LO
GHz. Observationswerecarriedoutundergoodweather assuch,thedetectedCO(J=3→2) lineemissionisspa-
conditions in the E array configuration on 2014 July 12. tially unresolved. We therefore extract the line pro-
This resulted in 1.56 hours of 15 antenna-equivalent on- file (Figure 2) at the peak position of the unresolved
sourcetimeafterdiscardingunusablevisibilitydata. The CO emission. Fitting a four-parameter single Gaus-
nearby source J1039+811 (0.65 Jy) was observed every sian to the spectrum yields a peak flux density of
20 minutes for pointing, amplitude, and phase calibra- 21.61±2.66 mJy, superimposed on a continuum level of
tion. Mars was observed as the primary absolute flux 4.15±0.48 mJy beam−1, and a line full width at half-
calibrator, and the quasar 3C273 was observed as the maximum (FWHM) of 546±36 km s−1.
secondary flux calibrator. J0927+390 was observed for We construct a velocity-integrated (0th moment) map
bandpass calibration, yielding ∼15% calibration accu- of the CO(J=3→2) line emission after subtracting
racy. continuum emission in the visibility plane. This re-
We use the miriad package to calibrate and ana- sults in a velocity-integrated CO(J=3→2) line flux
lyze the visibility data, which are deconvolved and im- of I =12.6±2.0 Jy km s−1 over a velocity range of
CO
aged with “natural” weighting. This yields a synthe- ∆v∼1420 km s−1, the uncertainty does not include
sized cleanbeamsize of11(cid:48).(cid:48)5×6(cid:48).(cid:48)2, −56.1◦ east ofnorth ∼15% calibration uncertainty. Our CO(J=3→2) line
for the upper sideband image cube and an rms noise measurement confirms the redshift of SMMJ0939, yield-
of σ =9.49 mJy beam−1 per channel of width ∼29 ing z=2.2212±0.0010.
ch
km s−1. The continuum image is created by averaging
over all line-free channels; this yields a synthesized clean 4. ANALYSIS
beam of 12(cid:48).(cid:48)0×6(cid:48).(cid:48)5, −55.9◦ east of north, and an rms 4.1. Lens Modelling
noise of σ = 0.50 mJy beam−1 over 6.8 GHz.
cont To study the intrinsic properties of the background
galaxy, we determine the magnification factor and the
3. RESULTS
half-light radius of the dust region by performing lens
3.1. Foreground Galaxy: 3C220.3
modeling on the SMA 1mm continuum data presented
Averaging over all line-free channels, we detect con- by H14 of this system. Lens modeling is carried out in
tinuum emission at ∼9σ significance at an averaged the visibility (uv-) plane using an updated version of the
frequency of ν =104.2106 GHz (∼2.9 mm) in the publiclyavailablesoftwareuvmcmcfit(Bussmannetal.
cont
observed-frame,correspondingto175.6GHz(∼1.7mm) 2015a),detailsoftheparametriclensmodelcanbefound
at z=0.685. In this lensing system, the foreground in Bussmann et al. (2015b). The surface mass densities
galaxy (3C220.3) is radio-loud, we thus expect it to be of the two lensing galaxies, 3C220.3 and its companion
thedominantcontributortothecontinuumemission(see galaxy B, are described by singular isothermal ellipsoid
§4.2.1 for details). The task imfit is used to estimate profiles, and the source is assumed to have an elliptical
the peak position of the continuum emission, where the Gaussian profile.
flux density is S =4.93±0.31 mJy beam−1. From the The resulting best-fit model as shown in Figure 3
ν
continuum measurement, the deconvolved source size is shows no significant bowls in the residual image, and
(8(cid:48).(cid:48)4±1(cid:48).(cid:48)1)×(4(cid:48).(cid:48)9±0(cid:48).(cid:48)6) at −53.8◦, and the integrated the knots (lensed emission) in the observed SMA data
Study of a strongly-lensed type-2 quasar host SMG at z=2.221 3
Figure 1. Left: Contourmapofthe104GHzcontinuumemissionintheforegroundradiogalaxy3C220.3. Thebeamsizeis12(cid:48).(cid:48)0×6(cid:48).(cid:48)5,
atP.A.= −56◦,asindicatedinthebottomleftcorner. Right: CARMA104GHzcontinuumemission(redcontours)overlaidontheVLA
9GHzcontinuumemission(greencontoursandgrayscale;H14). ThesynthesizedbeamsizeoftheVLAobservationsis0(cid:48).(cid:48)6×0(cid:48).(cid:48)2,atP.A.
76◦. Thecontourlevelsofthe104GHzcontinuumemissionstartat±2σ,incrementingatstepsof±2σ,whereσ=0.5mJybeam−1. The
contourlevelsofthe9GHzcontinuumemissionstartat±4σ,whereσ=0.064mJybeam−1,andincrementatstepsof±2nσ,wherenisa
positiveinteger. ThebluecrossescorrespondtothecentroidlocationsofthelensingknotsdetectedintheSMA1mmcontinuumemission
(seeFigure2). Thecentralcrossoneachpanelindicatesthepositionoftheradiocoreof3C220.3.
Figure 2. Top Left: Continuum-subtracted moment-0 map of CO(J=3→2)line emission toward the background SMG with
σ=1.03Jykms−1 beam−1 over a velocity range of ∆v∼514kms−1. The beam size is 11(cid:48).(cid:48)5×6(cid:48).(cid:48)2, at P.A.pdf.=−56◦, as indicated
in the bottom left corner. Top Right: Velocity-integrated CO(J=3→2)line emission (red contours) overlaid on the SMA 1mm dust
continuum(greencontoursandgrayscale;H14),withanrmsnoiseofσ1mm=0.84mJybeam−1. ThebeamsizeoftheSMAobservations
is 1(cid:48).(cid:48)4×1(cid:48).(cid:48)2, P.A. −34◦, as shown in the bottom left corner. The central cross on each image corresponds to the same coordinates as in
Figure1. Thecontourlevelsinbothimagesstartat±3σ,incrementingatstepsof±1σ. Bottom: Spectrumextractedatthepeakposition
ofCOlineemission,withaspectralresolutionof∆v ∼29kms−1,andanrmsofσ =9.5mJybeam−1 perchannel. Thesolidblackline
ch
showsaGaussianfittotheCO(J=3→2)lineprofile,wherethevelocityscaleisrelativetoz=2.221.
4 Leung & Riechers
spectral index α is (cid:38) 0.5. While the contribution from
extended components decreases, studies using samples
of radio galaxies have suggested that the flat/inverted-
spectrumofthecompactradiocorecomponentrisesand
dominates the flux density at higher frequencies (Keller-
mann&Pauliny-Toth1981;Begelmanetal.1984). This
has been observed in a FR-II galaxy at similar redshift
— 3C220.1 at z = 0.610, where observations were car-
ried out at the observed-frame frequency of ∼90 GHz
(Hardcastle & Looney 2008).
Previously, an upper limit of < 0.17 mJy at 4.6 GHz
has been established by Mullin et al. (2006) on the core
Figure 3. Double-lensmodelingofSMMJ0939usinguvmcmcfit
component of 3C220.3, and an unambiguous detection
on the SMA 1mm continuum data. The contours start at ±2σ,
√
incrementingatstepsof±2 2σ inbothpanels. Left: SMA1mm of 0.8 mJy at 9 GHz has been reported by H14, sug-
continuum(redcontours)overlaidonthebest-fitmodel(grayscale gesting a substantially inverted spectrum of the core
image),assuminganellipticalGaussianprofileforthebackground (Figure 4). Consequently, we may naively expect the
SMG. The lenses are represented by the black dots, the half-light
integrated flux density in our continuum detection of
area of the background source is represented by the magenta el-
lipse,andthecriticalcurvesarerepresentedbythegreenellipses. S104GHz=7.39±1.42 mJy to be dominated by the un-
Right: ResidualcontoursandimageobtainedbytakingtheFourier resolvedcorecomponentoftheforegroundFR-IIgalaxy,
transformofthedifferencebetweentheSMAdataandthebest-fit which is at z=0.685. However, the deconvolved spatial
model in the visibility plane. Solid (dashed) contours show the
size of the source matching that in the resolved image
positive(negative)residuals.
(see Figure 1) is suggestive of a marginally resolved de-
tection of the extended lobe components. This is plau-
sible given that the orientation of the synthesized beam
Table 1
Lensmodelingparametersandresults in our observations is in alignment with the axis along
the lobes of the radio galaxy, as shown in Figure 1. We
Parameters Best-FitValues investigate this disparity by fitting models to existing
SED measurements as listed in Table 2, and extrapolat-
Lens0(3C220.3) ingthefittoestimatethefluxdensityofthelobesatthe
OffsetinRA ∆α ((cid:48)(cid:48)) 0.403±0.026 frequency of our continuum measurement.
lens0
OffsetinDec ∆δlens0 ((cid:48)(cid:48)) -0.181±0.027 Following Equation (1) in Cleary et al. (2007), the fit
AxialRatio qlens0 0.446±0.063 tothelobeemissioncanbeexpressedasaparabolicfunc-
Positionangle φ (deg) 31.56±4.15
Einsteinradius θEle0ns0 ((cid:48)(cid:48)) 1.218±0.010 tion:
ν
logFlobe(ν)∝−β(log ν−logν )2+log(exp( )) (1)
Lens1(CompaniongalaxyB) ν t νlobe
c
OffsetinRA ∆α ((cid:48)(cid:48)) -0.804±0.034
lens1
OffsetinDec ∆δ ((cid:48)(cid:48)) -1.243±0.017 whereFlobe isthefluxdensityofthelobes,β isaparam-
lens1 ν
Axialratio qlens1 0.608±0.138 eter representing the bending of the parabola, νt is the
Positionangle φ (deg) 14.2±15.7
lens1 frequency at which the optical depth of the synchrotron
Einsteinradius θE1 ((cid:48)(cid:48)) 0.745±0.015 emittingplasmareachesunity,andνlobe isthefrequency
c
Source(SMMJ0939) corresponding to the cutoff energy of the lobe plasma
OffsetinRA ∆αs ((cid:48)(cid:48)) -0.163±0.035 energy distribution. The extrapolated flux density at
OffsetinDec ∆δs ((cid:48)(cid:48)) -0.193±0.048 104 GHz is consistent with the peak flux density of our
Axialratio qs 0.424±0.237 continuum measurement (Figure 4). The 9σ detection
Positionangle φs (deg) 174.34±8.89 of the continuum thereby suggests a dominant contribu-
Effectiveradius rs ((cid:48)(cid:48)) 0.106±0.033 tion from the lobes, and that the peak flux density is
Magnificationfactor µL 10.13±1.38
not dominated by emission toward the core. Moreover,
Note. — All angular offsets are with respect to α= the peak position of the 104 GHz continuum is centered
9h39m23s.54, δ=83◦15(cid:48)26(cid:48).(cid:48)10 (J2000). The corresponding
masseswithintheEinsteinradiiofthegalaxies3C220.3andits toward the brighter northern lobe (Figure 1), which fur-
companiongalaxyBareM(θ<θE)=(4.86±0.08)×1011 M(cid:12) ther supports our argument. Consequently, a conserva-
andM(θ<θE)=(1.82±0.07)×1011 M(cid:12),respectively. tive upper limit of Sν<4.93 mJy on the core emission
can be established using the measured peak flux den-
are reproduced well by the best-fit model. Our best-fit
sity. Yet, by considering the difference between the inte-
model yields a magnification factor of µ = 10.13±1.38
and a half-light radius of r =0(cid:48).(cid:48)11 ± 0(cid:48).L(cid:48)03, correspond- grated flux density from our measurement and the flux
s densityfromanextrapolationofthemodel(seeFigure4;
ing to ∼0.9 kpc at z=2.221. All best-fit parameters are
S = 5.10 mJy), we establish a more stringent
listed in Table 1. 104GHz,fit
constrain on this upper limit of S <2.29 mJy. We did
ν
not extrapolate the core measurements to the frequency
4.2. SED Fitting
of our continuum, as previous measurements of the core
4.2.1. 3C220.3
are taken across different epochs, and the core may be
Synchrotron continuum emission from extended com- time-variable.
ponentsofaradiogalaxydecreaseswithincreasingradio Studies by Meisenheimer et al. (1989) and Hardcas-
frequencies, and the spectrum is commonly character- tle & Looney (2008) have suggested that spectra of
ized by a power law distribution S ∝ ν−α, where the hotspots are flat up to optical frequencies, where some
Study of a strongly-lensed type-2 quasar host SMG at z=2.221 5
Table 2
Continuumdataofthelensinggalaxy3C220.3and
backgroundSMGSMMJ0939
Wavelength FluxDensity Instrument
SMMJ0939
70 µm 29.5±5 mJy PACS
100 µm 102±7 mJy PACS
160 µm 289±9 mJy PACS
250 µm 440±15 mJy SPIRE
350 µm 403±20 mJy SPIRE
500 µm 268±30 mJy SPIRE
1000 µm 51±14a mJy SMA
Frequency FluxDensity Reference
3C220.3Integrated(Core&Lobes)
104.2 GHz 7.39±1.42b mJy LR16
10.7 GHz 270±30 mJy KP73
10.7 GHz 253±28 mJy L80
5.0 GHz 640±100 mJy K69 Figure 4. SEDs of 3C220.3 (solid purple line) and SMMJ0939
5.0 GHz 636±50 mJy L80 (dashedpurplelineandsolidcyanline)includingthenewmeasure-
2.7 GHz 1.33±0.07 Jy K69 ments presented in this paper. The solid purple line corresponds
2.7 GHz 1.34±0.10 Jy L80 to the parabolic function we fit to the existing data associated
1.4 GHz 2.95±0.09 Jy C98 with 3C220.3 (black dots; see Table 2). The red dots at 104 GHz
1.4 GHz 2.99±0.06 Jy P66 correspondtoourcontinuummeasurements(integratedandpeak,
1.4 GHz 2.80±0.14 Jy K69 respectively),andtheredtrianglescorrespondtotheupperlimits
1.4 GHz 2.89±0.09 Jy L80 on the radio core. The dashed purple line and the solid cyan line
0.75 GHz 5.94±0.28 Jy L80 correspondtothebest-fitopticallythickandopticallythinmodels
0.75 GHz 5.94±0.21 Jy P66 ofSMMJ0939,respectively,usingthephotometricdatafromH14.
0.75 GHz 5.60±0.84 Jy K69
352 MHz 11.3±0.453 Jy WENSS
352 MHz 11.6±0.464 Jy WENSS
metric data obtained with Herschel/PACS and SPIRE,
178 MHz 15.7±2.35 Jy K69
178 MHz 17.1±1.71 Jy L80 atwavelengthsbetweenobserved-frame70µm−500µm,
152 MHz 22.6±0.08 Jy B85 and the interferometric data obtained with the SMA
152 MHz 22.5±0.04 Jy B85 at 1mm (H14). We use the publicly available software
86 MHz 51.6±9.90 Jy L80
73.8 MHz 37.5±3.82 Jy C07 mbb emcee1 to perform the SED fitting; the code uses
38 MHz 49.6±4.96 Jy L80 an affine-invariant Markov chain Monte Carlo (MCMC)
38 MHz 40.2±6.30 Jy K69 approach, and further details of the code are given by
37.8 MHz 60.7±6.07 Jy H95
Riechers et al. (2013) and Dowell et al. (2014).
17.8 MHz 64.9±6.49 Jy H95
The functional form of the fit comprises a single-
3C220.3(CoreOnly) temperature, modified blackbody function joined to a
104.2 GHz <2.29c mJy LR16 Bλ ∝ λα power law on the blue side of the SED. We
9.0 GHz 0.80±0.06 mJy H14 fit both optically thick and optically thin models. In the
4.86 GHz <0.17 mJy M06 opticallythickcase, thewavelengthλ =c/ν isanaddi-
0 0
tional parameter representing the rest-frame wavelength
References. — B85=Baldwinetal.(1985);C98=Con- at which the optical depth τ = (ν/ν )β reaches unity.
ν 0
don et al. (1998); C07 = Cohen et al. (2007); H95 = Hales Thus, the functional form of the modified blackbody in
et al. (1995); H14 = Haas et al. (2014); K69 = Kellermann
the optically thick regime is as follows:
etal.(1969);KP73=Kellermann&Pauliny-Toth(1973);L80
=Laing&Peacock(1980);LR16=thiswork;M06=Mullin
et al. (2006); P66 = Pauliny-Toth et al. (1966); WENSS = (1−exp−(λ0(λ1+z))β)(c)3
Rengelinketal.(1997)† B ∝ λ (2)
λ hc
Note. — Photometric data of SMMJ0939 are from Haas expλkT/(1+z) −1
etal.(2014).
a Errorsincludecalibrationuncertainties and in the optically thin regime, the functional form re-
b Integratedfluxdensity. Peakfluxdensityofthecontinuum duces to:
emissionis4.93±0.31mJybeam−1 (c)β+3
c ConstraintfromSEDmodeling B ∝ λ (3)
† www.astron.nl/wow/testcode.php?survey=1 λ expλkT/h(c1+z) −1
exhibit spectral steepening in cm and mm wavelengths whereT istherest-framecolddusttemperature,β isthe
(e.g.,3C123). Attheresolutionofourobservations,itre- dust emissivity index , and α is the mid-infrared power
mainsunclearwhetherthemeasuredfluxdensityisdom- lawspectralindex. Theoverallfitisnormalizedusingthe
inated by emission from the compact hotspots or that observed-frame 500 µm flux density, hence this becomes
from the surrounding diffuse lobe components. an additional parameter (f ) in the fit. For
norm, 500µm
both models, we impose an upper limit of 60 K on the
4.2.2. SMMJ0939+8315 observed-frame dust temperature (T/(1 + z)), and an
To constrain the dust and gas properties in the ISM
of SMMJ0939, we perform SED fitting to the photo- 1 https://github.com/aconley/mbb_emcee
6 Leung & Riechers
Penzias 1970; Downes & Solomon 1998), transition lines
Table 3 of higher rotational states (J > 1) are frequently ob-
SEDfittingresults
served in high-redshift sources as the ground state tran-
sitionlineisredshiftedtolowerfrequenciesthatcanonly
Parameters OpticallyThick OpticallyThin
be observed with traditional radio telescopes (Carilli &
χ2 2.25 5.31 Walter 2013). Consequently, assumptions on the CO
D.O.F 2 3 excitation conditions are required to derive the molec-
Tβα (K) 6123...991+−+−+−000011.....63.5413 5022...780+−+−+−000011.....22.3232 u(αlaRCrOegc)aenwstmhoeanbssseexurvtsrainatpigoontlhsaetiinMngh(ifHgrho2m-)r-ethod-isgLhh(cid:48)CifeOtr-qcJuoanCsvOaerrslhiinooensst.fsascutogr-
λ0a (µm) 248.7+−8162.30.8 — gestthattheratioisR31∼1(Riechersetal.2006,2011c).
λpeak b (µm) 254.7+−66..21 301.4+−2390..01 Inthecaseofhigh-redshifttype-2quasars,Riechersetal.
fnorm,500µmc (mJy) 267.4+−1166..73 244.3+−1155..33 (2011c) report a brightness temperature ratio of R31 =
LIRd (1012 L(cid:12)) 88.5+−22..66 89.2+2.25.5 1.00±0.10 for IRAS F10214+4724 (hereafter F10214),
Mduste (108 M(cid:12)) 50.5+−2200..42 25.7+−35..95 which is currently the only known type-2 quasar with
both CO(J=3→2) and CO(J=1→0) line measure-
Note. —Errorsreportedhereare±1σ. LIRandMdarereported
priortolensingcorrection. ments. Here,wederivethemoleculargasmassassuming
a Rest-framewavelengthwhereτν=1 thermalized excitation of CO, as SMMJ0939 is postu-
b Observed-framewavelengthoftheSEDpeak lated to be hosting a type-2 quasar (H14).
c Observed-framefluxdensityat500µm
d Rest-frame8-1000µmluminosity We calculate the CO(J=1→0) line luminosity us-
e Derived assuming a standard absorption mass coefficient κ=2.64 ing a standard relation (e.g., Solomon & Van-
m2 kg−1 atλ=125.0µm(Dunneetal.2003) den Bout 2005) and assuming a conversion factor
of α = 0.8M (K km s−1 pc2)−1 based on empir-
upperlimitof2.2onβ. Fortheopticallythickmodel,we CO (cid:12)
ical relations from local ULIRGs, which is typi-
imposeanadditionalupperlimitof3000µmonλ (1+z).
0
cally adopted for SMGs (e.g., Tacconi et al. 2006,
Thebest-fitvaluesinbothregimesarelistedinTable3,
2008; Bothwell et al. 2013). This corresponds to
and the correlation plots are available in the Appendix.
Thebest-fitsolutionofopticallythinmodelscorresponds L(cid:48)CO(1-0)= (3.42±0.71)×1010(10.1/µL) K km s−1 pc2;
to χ2 = 5.31 with 3 degrees of freedom, whereas that of hence the inferred total molecular gas mass is M
gas
optically thick models corresponds to χ2 = 2.25 with 2 = (2.74±0.57)× 1010M after correcting for lensing
(cid:12)
degreesoffreedom,suggestingabetterfitthanintheop- magnification. This results in a gas-to-dust ratio of
tically thin case. In the subsequent analysis, we employ f =M /M =55±24. This is in good agree-
gas-dust gas dust
the inferred values from the optically thick model. The mentwiththevaluesfoundforotherSMGs(Coppinetal.
best-fit solution yields a far-infrared luminosity (rest- 2008; Michal(cid:32)owski et al. 2010; Riechers et al. 2011a).
frame 42.5−122.5µm) of L = 53.3+1.1×1012L , and
FIR −1.1 (cid:12)
a total infrared (IR; rest-frame 8−1000 µm) luminos- 4.3.2. Star Formation Rate & Star Formation Efficiency
ity of L = 88.5+2.6×1012L 2. Assuming a dust ab- We derive the SFR using the lensing-corrected far-
IR −2.6 (cid:12)
sorption coefficient of κ = 2.64 m2 kg−1 at 125.0 µm infrared luminosity assuming that the dominant heat-
ν
(Dunne et al. 2003), we find a dust mass of M ing source of cold-dust is young and massive stars, and
dust
= 50.520.4 ×108 M ; the uncertainties do not include that a contribution from the dust-enshrouded AGN is
−20.2 (cid:12) negligible. Thisassumptionstemsfromtheresultsofre-
thoseinthedustabsorptioncoefficient(κ ). Theseprop-
ν cent studies using various approaches, such as spectral
ertiesarederivedbasedontheSEDfittingtothephoto-
decomposition techniques and correlation between far-
metricdata,i.e.,priortolensingcorrection. Wenotethat
infrared luminosity and other tracers of star-formation,
the dust mass is weakly constrained owing to the dearth
suggesting that far-infrared emission dominantly origi-
of data in the rest-frame FIR waveband. As such, we in-
nates from star-formation in host galaxies, even in the
vestigate how the dust mass would be affected by fitting
most energetic QSOs (e.g., Netzer et al. 2007; Mullaney
additional optically thick models with an upper limit of
et al. 2011; Harrison et al. 2015).
β adjusted from 2.2 to 3.0. While the difference in each
Using the Kennicutt (1998) relation and adopting a
best-fitparameterbetweenthisscenarioandtheprevious
Chabrier (2003) stellar initial mass (IMF) function, we
models(withanupperlimitofβ =2.2)iswithin3%,we
find that the dust mass inferred from this best-fit SED find a SFRFIR=526±73 M(cid:12) yr−1. The starburst in
model is boosted by a factor of ∼2. SMMJ0939 can be maintained at its current rate for
a time that can be approximated by the gas deple-
4.3. Physical Properties of the ISM in SMMJ0939 tion timescale, τdepl=Mgas/SFR, which assumes no re-
plenishment of gas and feedbacks. This corresponds to
4.3.1. Molecular Gas Mass
τ =52±8Myr,whichisingoodagreementwiththose
depl
While the ground state CO transition line traces the found in other SMGs (e.g., Greve et al. 2005).
cold molecular gas in the ISM (e.g., Wilson, Jefferts, & The SFR per unit mass of molecular gas is commonly
taken as a measure of the star formation efficiency. We
2 Owning to the positive K-correction blue-ward of the dust compute this ratio using the far-infrared and CO lumi-
peak, in which the foreground radio galaxy contributes a non- nosities. The derived SFE is therefore independent of
negligibleamounttotheMIRluminosity,wedonotfitforasepa-
the magnification factor, the CO luminosity to gas mass
rate AGN component. Instead, we adopt a power-law to account
fortheMIRexcess,whichallowsustoestimatetheIRluminosity conversion factor (αCO), and the IMF. This, however,
(e.g., Casey2012;Riechersetal.2013;Kirkpatricketal.2015). assumes that differential lensing between the CO and
Study of a strongly-lensed type-2 quasar host SMG at z=2.221 7
far-infrared emission is negligible. The resulting ratio thestarformationactivitytakingplaceatarateof∼526
is SFE =154±25 L (K km s−1 pc2)−1, this is com- M yr−1. If the star forming activity continues at the
FIR (cid:12) (cid:12)
parable to those found in “typical” SMGs (Greve et al. current rate, the gas reservoir will be depleted within
2005; Tacconi et al. 2006; Riechers et al. 2011a). (cid:46) 52 Myr, which is consistent with the short timescales
We compute the surface densities by dividing half the found in other SMGs (Greve et al. 2005). The derived
SFR and gas mass by the area subtended by the half- intrinsicpropertiesofSMMJ0939areevidentofongoing
light radius (e.g., Genzel et al. 2010; Harrison et al. rapid star formation; this is in good agreement with the
2015), yielding Σ = 106 M yr−1 kpc−2 and Σ = current conjecture that SMGs are a population of high-
SF (cid:12) gas
5.48×109 M kpc−2, respectively. These results are redshift galaxies that build up the bulk of stellar mass
(cid:12)
in good agreement with values typical for SMGs (Tac- in present-day galaxies, thus play an important role in
coni et al. 2006; Hodge et al. 2015). The inferred sur- galaxy formation and evolution (e.g., Dickinson et al.
face densities of SMMJ0939 follow a universal Schmidt- 2003).
Kennicutt relation between the star formation rate sur- We compare our findings for SMMJ0939 with a sam-
face density and the molecular gas surface density: Σ ple of typically unlensed or only weakly magnified,
SF
= 9.3(±2)×10−5 (M /2πR2 )1.71(±0.05), which was 850 µm−selected SMGs (Bothwell et al. 2013, hereafter
gas 1/2 B13). Their properties are listed in Table 4, showing
derived using a sample consisting of local star-forming
that SMMJ0939 has properties similar to other SMGs
galaxies and high-redshift galaxies out to z∼2.5, and
studied to date. The gas mass in SMMJ0939 is slightly
assuming a Chabrier IMF (Bouch´e et al. 2007).
lower than the median in the B13 sample, but we can-
not rule out the possibility that this difference is due
4.3.3. Physical Size and Dynamical Mass
to the different assumptions made for the gas excitation
Ourlensmodelsuggestsahalf-lightradiusofr ∼1kpc
s conditions. The gas properties (CO luminosity and gas
forthedust-emittingregioninSMMJ0939. Thisiscom-
mass) of the SMGs in the B13 sample are derived based
parable to the half-light radii found in other SMGs with
on the assumption of typical excitation conditions found
highresolutionimaging. Similarsizeshavebeenreported
from CO spectral line energy distribution (SLED) mod-
by Bussmann et al. (2013), who find typical radii of 1.5
elling of the sample average, which the authors find to
kpcforasampleofHerschel−selectedlensedSMGswith
be very similar to those of the cosmic Eyelash. Thus,
S500µm >100mJy. Also, Simpsonetal.(2015)reporta we additionally compare SMMJ0939 in more detail to
radial extent of 1.2 kpc for a sample of un-lensed SMGs
two other well-studied, strongly-lensed SMGs with com-
with S850µm = 8−16 mJy. parablyhighapparentsubmillimeterfluxesfoundatsim-
We estimate the dynamical mass of SMMJ0939 us-
ilar redshifts — HLSW-01 and the cosmic Eyelash. The
ing our CO(J=3→2) line measurement and assuming
properties of these sources are derived using similar ap-
that the molecular gas is virialized. With this assump-
proaches to those employed in this paper.
tion, we use an isotropic virial estimator (e.g., Engel
As shown in Table 4, while the cosmic Eyelash has
et al. 2010), with the FWHM of the CO(J=3→2) line
the least amount of molecular gas, as well as the longest
profile and the half-light radius from our lens model for
gas depletion timescale, the overall gas and dust proper-
R , assuming that the dust emission traces the same
eff ties of SMMJ0939 fall between those of HLSW-01 and
emitting region as the CO. We find a dynamical mass of
the cosmic Eyelash. Such distinction is likely a result
M =(7.84±2.84)×1010M ,andagas-to-dynamical
dyn (cid:12) of our selection bias: while these sources appear simi-
massfractionoff =0.35±0.14,consistentwith
gas-to-dyn larly bright at 250 µm, the lensing magnification varies
those of other SMGs (Tacconi et al. 2006). We note
by a factor of ∼3. In particular, with the cosmic Eye-
thataderiveddynamicalmassbasedonthisassumption
lash having the highest lensing magnification among the
is likely to be biased towards low values, as the CO-
three,thisintrinsicallyfainter,andlessgas-richSMGap-
emitting region can be apparently more extended than
pearsnotablybrightat250µm,whileitsCOlineandIR
thedustemittingregionduetothelowdustopticaldepth
luminosities are lower than those of most SMGs studied
at larger radii. This is supported by recent studies, in
to date. While lensing can probe sources of various in-
which CO source sizes ranging from ∼4−20 kpc have
trinsic properties, we find that SMMJ0939 is consistent
been found, which are larger than typical dust contin-
withthe“typical”SMGpopulation,withitsintrinsicCO
uum sizes (Tacconi et al. 2006; Riechers et al. 2011a;
lineluminosity,IRluminosity,dustmass,SFR,SFE,de-
Ivison et al. 2011; Hodge et al. 2013, 2015).
pletion timescale, and gas mass fraction comparable to
those found in “typical” SMGs studied to date.
5. DISCUSSIONANDCONCLUSIONS
Since SMMJ0939 also hosts a type-2 quasar, we com-
We present the detection of CO(J=3→2) line emis-
pare its properties against those of eight CO-detected
sion toward SMMJ0939+8315, a strongly-lensed SMG
obscured AGNs at z = 1.6−2.8 (Polletta et al. 2011,
that is hosting a type-2 quasar, refining the redshift to
and references therein). Among these obscured quasars,
z=2.2212±0.0010. The underlying continuum is de-
F10214hasthelowestmoleculargasmassaswellasSFR.
tected at ∼9σ significance, where the flux density is
ThefactthatthegasmassofF10214intheircompilation
likelydominatedbyemissionfromthelobesandhotspots
wasderivedusingCO(J=3→2) lineemission(Solomon
of the foreground radio galaxy 3C220.3.
& Vanden Bout 2005) has a minor effect on the result-
The detection of CO in SMMJ0939 implies a CO
ing low gas mass; Riechers et al. (2011c) report a sim-
luminosity of L(cid:48) = (3.4±0.7)×1010(10.1/µ )
CO(1-0) L ilarly low gas mass derived using their CO(J=1→0)
K km s−1 pc2, corresponding to a gas mass of line emission. With F10214 being the most strongly-
M =(2.7±0.6×1010(10.1/µ M ). This suggests the lensed high-redshift type-2 quasar (µ = 17; Solomon
gas L (cid:12) L
presence of a massive molecular gas reservoir that fuels
8 Leung & Riechers
& Vanden Bout 2005)3, its exceptionally low molecular Casey,C.M.2012,MNRAS,425,3094
gas mass and SFR is evident that this source lies on the Casey,C.M.,Narayanan,D.,&Cooray,A.2014,Phys.Rep.,
lowendoftheCOandIRluminositydistributionsofthe 541,45
Chabrier,G.2003,PASP,115,763
population. In contrast to what was found for F10214,
Chapman,S.C.,Blain,A.W.,Smail,I.,&Ivison,R.J.2005,
we find that the properties (e.g., FWHM of the CO line
ApJ,622,772
profile, Mgas, and SFE) of SMMJ0939 are similar to the Cleary,K.,Lawrence,C.R.,Marshall,J.A.,Hao,L.,&Meier,
statistical means, except for the SFR, which is lower by D.2007,ApJ,660,117
a factor of ∼1.5, but is nevertheless consistent within Cohen,A.S.,Lane,W.M.,Cotton,W.D.,etal.2007,AJ,134,
1245
the measurement uncertainties. We thus find that the
Condon,J.J.,Cotton,W.D.,Greisen,E.W.,etal.1998,AJ,
gas mass in SMMJ0939 is consistent with other type-2 115,1693
quasars. Conley,A.,Cooray,A.,Vieira,J.D.,etal.2011,ApJ,732,L35
WhilethegasmassinSMMJ0939isslightlylowerthan Coppin,K.E.K.,Swinbank,A.M.,Neri,R.,etal.2008,
the median of SMGs in the B13 sample, we find it to be MNRAS,389,45
Danielson,A.L.R.,Swinbank,A.M.,Smail,I.,etal.2011,
consistent with other type-2 quasars, suggesting a possi-
MNRAS,410,1687
ble scenario in which a significant fraction of the gas has Deane,R.P.,Heywood,I.,Rawlings,S.,&Marshall,P.J.2013,
already been converted into stars and used for fueling MNRAS,434,23
the quasar. As such, given the presence of an obscured Dickinson,M.,Papovich,C.,Ferguson,H.C.,&Budav´ari,T.
quasar and the overlap of properties with both type-2 2003,ApJ,587,25
Dowell,C.D.,Conley,A.,Glenn,J.,etal.2014,ApJ,780,75
quasar and SMG populations, a physical interpretation
Downes,D.,&Solomon,P.M.1998,ApJ,507,615
might be that SMMJ0939 is transitioning from a short, Dunne,L.,Eales,S.A.,&Edmunds,M.G.2003,MNRAS,341,
star-bursting phase to an unobscured quasar phase, con- 589
sistent with the proposed evolutionary link model be- Eales,S.,Dunne,L.,Clements,D.,etal.2010,PASP,122,499
Engel,H.,Tacconi,L.J.,Davies,R.I.,etal.2010,ApJ,724,233
tween dusty starbursts and quasars (e.g., Sanders et al.
Fanaroff,B.L.,&Riley,J.M.1974,MNRAS,167,31P
1988; Coppin et al. 2008; Simpson et al. 2012).
Frayer,D.T.,Ivison,R.J.,Scoville,N.Z.,etal.1998,ApJ,506,
L7
Gavazzi,R.,Cooray,A.,Conley,A.,etal.2011,ApJ,738,125
We thank the referee for providing constructive com-
(G11)
ments to improve this manuscript. We thank Shane Genzel,R.,Tacconi,L.J.,Gracia-Carpio,J.,etal.2010,
Bussmann for providing the code uvmcmcfit for lens MNRAS,407,2091
modeling, sharing the SMA and VLA data, and for Greve,T.R.,Bertoldi,F.,Smail,I.,etal.2005,MNRAS,359,
1165
helpful discussions; Alex Conley for providing the code
Haas,M.,Leipski,C.,Barthel,P.,etal.2014,ApJ,790,46(H14)
mbb emcee for SED fitting. Support for CARMA con-
Hales,S.E.G.,Waldram,E.M.,Rees,N.,&Warner,P.J.1995,
structionwasderivedfromtheGordonandBettyMoore MNRAS,274,447
Foundation, the Kenneth T. and Eileen L. Norris Foun- Hardcastle,M.J.,&Looney,L.W.2008,MNRAS,388,176
dation, the James S. McDonnell Foundation, the Asso- Harrison,C.M.,Simpson,J.M.,Stanley,F.,etal.2015,ArXiv
e-prints,arXiv:1510.06740
ciates of the California Institute of Technology, the Uni-
Hinshaw,G.,Larson,D.,Komatsu,E.,etal.2013,ApJS,208,19
versity of Chicago, the states of Illinois, California, and
Hodge,J.A.,Carilli,C.L.,Walter,F.,Daddi,E.,&Riechers,D.
Maryland, and the National Science Foundation. Ongo- 2013,ApJ,776,22
ing CARMA development and operations are supported Hodge,J.A.,Riechers,D.,Decarli,R.,etal.2015,ApJ,798,L18
by the National Science Foundation under a cooperative Hughes,D.H.,Serjeant,S.,Dunlop,J.,etal.1998,Nature,394,
241
agreement, and by the CARMA consortium universities.
Ivison,R.J.,Papadopoulos,P.P.,Smail,I.,etal.2011,MNRAS,
Facilities: CARMA 412,1913
Ivison,R.J.,Swinbank,A.M.,Swinyard,B.,etal.2010,A&A,
REFERENCES 518,L35
Kellermann,K.I.,&Pauliny-Toth,I.I.K.1973,AJ,78,828
—.1981,ARA&A,19,373
Baldwin,J.E.,Boysen,R.C.,Hales,S.E.G.,etal.1985, Kellermann,K.I.,Pauliny-Toth,I.I.K.,&Williams,P.J.S.
MNRAS,217,717 1969,ApJ,157,1
Barger,A.J.,Cowie,L.L.,Sanders,D.B.,etal.1998,Nature, Kennicutt,Jr.,R.C.1998,ARA&A,36,189
394,248 Kirkpatrick,A.,Pope,A.,Sajina,A.,etal.2015,ApJ,814,9
Begelman,M.C.,Blandford,R.D.,&Rees,M.J.1984,Reviews Lagache,G.,Puget,J.-L.,&Dole,H.2005,ARA&A,43,727
ofModernPhysics,56,255 Laing,R.A.,&Peacock,J.A.1980,MNRAS,190,903
Blain,A.W.,Smail,I.,Ivison,R.J.,Kneib,J.-P.,&Frayer, Meisenheimer,K.,Roser,H.-J.,Hiltner,P.R.,etal.1989,A&A,
D.T.2002,Phys.Rep.,369,111 219,63
Bothwell,M.S.,Smail,I.,Chapman,S.C.,etal.2013,MNRAS, Michal(cid:32)owski,M.J.,Watson,D.,&Hjorth,J.2010,ApJ,712,942
429,3047(B13) Mullaney,J.R.,Alexander,D.M.,Goulding,A.D.,&Hickox,
Bouch´e,N.,Cresci,G.,Davies,R.,etal.2007,ApJ,671,303 R.C.2011,MNRAS,414,1082
Bussmann,R.S.,P´erez-Fournon,I.,Amber,S.,etal.2013,ApJ, Mullin,L.M.,Hardcastle,M.J.,&Riley,J.M.2006,MNRAS,
779,25 372,113
Bussmann,S.,Leung,T.K.D.,&Conley,A.2015a,uvmcmcfit, Negrello,M.,Hopwood,R.,DeZotti,G.,etal.2010,Science,330,
doi:10.5281/zenodo.27357 800
Bussmann,R.S.,Riechers,D.,Fialkov,A.,etal.2015b,ApJ, Neri,R.,Genzel,R.,Ivison,R.J.,etal.2003,ApJ,597,L113
812,43 Netzer,H.,Lutz,D.,Schweitzer,M.,etal.2007,ApJ,666,806
Carilli,C.L.,&Walter,F.2013,ARA&A,51,105 Oliver,S.J.,Bock,J.,Altieri,B.,etal.2012,MNRAS,424,1614
Carlstrom,J.E.,Ade,P.A.R.,Aird,K.A.,etal.2011,PASP, Pauliny-Toth,I.I.K.,Wade,C.M.,&Heeschen,D.S.1966,
123,568 ApJS,13,65
Polletta,M.,Nesvadba,N.P.H.,Neri,R.,etal.2011,A&A,533,
3 Note that Deane et al. (2013) suggest a magnification factor A20
ofµL =6±1.5fortheCOemissioninF10214.
Study of a strongly-lensed type-2 quasar host SMG at z=2.221 9
Rengelink,R.B.,Tang,Y.,deBruyn,A.G.,etal.1997,A&AS, —.2015,ApJ,799,81
124,259 Smail,I.,Ivison,R.J.,&Blain,A.W.1997,ApJ,490,L5
Riechers,D.A.,Walter,F.,Carilli,C.L.,etal.2006,ApJ,650, Smolˇci´c,V.,&Riechers,D.A.2011,ApJ,730,64
604 Solomon,P.M.,&VandenBout,P.A.2005,ARA&A,43,677
Riechers,D.A.,Hodge,J.,Walter,F.,Carilli,C.L.,&Bertoldi, Swinbank,A.M.,Smail,I.,Longmore,S.,etal.2010,Nature,
F.2011a,ApJ,739,L31 464,733
Riechers,D.A.,Carilli,L.C.,Walter,F.,etal.2011b,ApJ,733, Swinbank,A.M.,Papadopoulos,P.P.,Cox,P.,etal.2011,ApJ,
L11 742,11
Riechers,D.A.,Carilli,C.L.,Maddalena,R.J.,etal.2011c, Tacconi,L.J.,Neri,R.,Chapman,S.C.,etal.2006,ApJ,640,
ApJ,739,L32 228
Riechers,D.A.,Cooray,A.,Omont,A.,etal.2011d,ApJ,733, Tacconi,L.J.,Genzel,R.,Smail,I.,etal.2008,ApJ,680,246
L12 Vieira,J.D.,Crawford,T.M.,Switzer,E.R.,etal.2010,ApJ,
Riechers,D.A.,Bradford,C.M.,Clements,D.L.,etal.2013, 719,763
Nature,496,329 Wilson,R.W.,Jefferts,K.B.,&Penzias,A.A.1970,ApJ,161,
Sanders,D.B.,Soifer,B.T.,Elias,J.H.,etal.1988,ApJ,325,74 L43
Simpson,J.M.,Smail,I.,Swinbank,A.M.,etal.2012,MNRAS,
426,3201
10 Leung & Riechers
s s Og
Type-2QSO a2.27±0.32–––f450±180e3.0 2.4±1.4i2.8···855±480i35···347±268······ R11=Riecher helowest-JCSLEDmodelin
rk; ontCO
Table4ComparisonofSMMJ0939withSMGsandtype-2QSOsatz∼2. SMMJ0939HLSW-01CosmicEyelashSMGs QuantityUnitReferenceReference az2.2212.957R112.326S102.2µ10.1±1.410.9±0.7G1137.5±4.5S11–LbSmJy440±15425±10C11366±55I10–250−1IJykms12.6±2.09.7±0.5R1113.2±0.1D11–CO(3-2)(cid:46)−1cccde∆vkms546±36350±25R11800D11550±90FWHM(cid:48)10−12L10Kkmspc3.4±0.74.2±0.4R111.7±0.2D115.2±1.0CO(1-0)g1010M2.7±0.63.3±0.3R111.6±0.1S104.2±0.8Mgas(cid:12)12hL10L5.3±0.711.0±0.9C111.8±0.2I106.0±0.6(cid:12)FIR8ijM10M5.2±2.11−5.2R11∼4.0I105.4±1.5(cid:12)dustk−1iSFRMyr526±731430±160C11∼235I10600±60(cid:12)FIRiiτMyr52±823±3R1168LR1670±15deplif55±2460−330R11∼40I1078±26gas−dust−12−1iiSFEL(Kkmspc)256±41340±40R11135±20LR16182±38(cid:12)10ilM10M7.8±2.83.7±1.8LR166.0±0.5S117.2±1.3(cid:12)dynif0.4±0.10.9LR160.6±0.1S110.6±0.2gas−dyn References.—C11=Conleyetal.(2011);D11=Danielsonetal.(2011);G11=Gavazzietal.(2011);I10=Ivisonetal.(2010);LR16=thiswoetal.(2011d);S11=Swinbanketal.(2011);S10=Swinbanketal.(2010) Note.—PropertiesofSMGsandtype-2QSOsarebasedontheresultsfromB13andPollettaetal.(2011),respectively.Gasmassisestimatedbasedlinemeasurementsavailable.Thermalizedexcitation(i.e.,R=1)hasbeenassumedfortype-2QSOsandtheexcitationconditionsforSMGsarebasedon31(seeBothwelletal.2013).Valueslistedfromrow6onwardsarelensing-corrected,andtheerrorsquotedforSMMJ0939includesuncertaintiesinµ.LaStatisticalaverageinthesamplebH14cCO(J=3→2)dEstimatedfromFigure1inD11eBasedonCO(J=3→2)andCO(J=4→3)lineobservationsfBasedonCO(J=2→1),CO(J=3→2),andCO(J=4→3)lineobservationsg−12−1α=0.8M(Kkmspc)CO(cid:12)hInferredfromradiocontinuummeasurementsviatheFIR-radiocorrelation(B13)iDerivedfromthereportedvaluesjUsingSandopticallythin,Rayleigh-Jeansapproximation(?)µm850kChabrierIMFlUsingthephysicalsizeofCO(J=5→4)emission(G11)