An Extragalactic 12CO J=3-2 survey with the Heinrich-Hertz-Telescope
Rui-Qing Mao, Andreas Schulz, Christian Henkel, Rainer Mauersberger, Dirk Muders, Dihn-V-Trung
aa r X i v : . [ a s t r o - ph . C O ] S e p Accepted for publication in The Astrophysical Journal
Preprint typeset using L A TEX style emulateapj v. 08/22/09
AN EXTRAGALACTIC CO J = 3–2 SURVEY WITH THE HEINRICH-HERTZ-TELESCOPE Rui-Qing Mao , Andreas Schulz , Christian Henkel , Rainer Mauersberger , Dirk Muders andDihn-V-Trung Accepted for publication in The Astrophysical Journal
ABSTRACTWe present results of a CO J = 3–2 survey of 125 nearby galaxies obtained with the 10-m Heinrich-Hertz-Telescope, with the aim to characterize the properties of warm and dense molecular gas in alarge variety of environments. With an angular resolution of 22 ′′ , CO 3–2 emission was detected in114 targets. Based on 61 galaxies observed with equal beam sizes the CO 3–2/1–0 integrated lineintensity ratio R is found to vary from 0.2 to 1.9, with an average value of 0.81. No correlationsare found for R to Hubble type and far infrared luminosity. Possible indications for a correlationwith inclination angle and the 60 µ m/100 µ m color temperature of the dust are not significant. Higher R ratios than in “normal” galaxies, hinting at enhanced molecular excitation, may be found ingalaxies hosting active galactic nuclei. Even higher average values are determined for galaxies withbars or starbursts, the latter being identified by the ratio of infrared luminosity versus isophotal area,log [( L FIR /L ⊙ )/( D /kpc )] > R value. Thismay be a consequence of particularly vigorous star formation activity, triggered by galaxy interactionand merger events. The nuclear CO luminosities are slightly sublinearly correlated with the globalFIR luminosity in both the CO J = 3–2 and the 1–0 lines. The slope of the log-log plots riseswith compactness of the respective galaxy subsample, indicating a higher average density and a largerfraction of thermalized gas in distant luminous galaxies. While linear or sublinear correlations for the CO J = 3–2 line can be explained, if the bulk of the observed J = 3–2 emission originates frommolecular gas with densities below the critical one, the case of the CO J = 1–0 line with its smallcritical density remains a puzzle. Subject headings: galaxies: ISM – galaxies: starburst – galaxies: active – radio lines: galaxies – ISM:molecules – surveys INTRODUCTION
Low lying rotational transitions of CO are widely usedas tracers of molecular hydrogen and are essential to de-termine dynamical properties and total molecular massesof galaxies. The widespread use of CO J = 1–0 and2–1 (hereafter CO(1–0) and CO(2–1)) spectroscopy (e.g.Braine et al. 1993; Young et al. 1995; Chini et al. 1996;Elfhag et al. 1996; Albrecht et al. 2004, 2007) is, how-ever, not sufficiently complemented by systematic sur-veys in higher rotational CO transitions to constrain theexcitation conditions of the dense interstellar medium(ISM). While the J = 1 and 2 states of CO are only5.5 and 17 K above the ground level, the J = 3 stateis at 33 K and traces a component of higher excitation.The “critical densities”, at which collisional de-excitationmatches spontaneous decay in the optically thin limit, is ∼ cm − for CO J = 3–2 (hereafter CO(3–2)) in Purple Mountain Observatory, Chinese Academy of Sciences,210008 Nanjing, PR China; [email protected] Max-Planck-Institut f¨ur Radioastronomie, Auf dem H¨ugel 69,D-53121 Bonn, Germany Argelander-Institut f¨ur Astronomie, Universit¨at Bonn, Aufdem H¨ugel 71, D-53121 Bonn, Germany Institut f¨ur Physik und ihre Didaktik, Universit¨at zu K¨oln,Gronewaldstr. 22, D-50931 K¨oln, Germany Joint ALMA Observatory, Av. Alonso de C´ordova 3107 , Vi-tacura, Santiago, Chile Institute of Astronomy and Astrophysics, Academia Sinica,Taipei, Taiwan Center for Quantum Electronics, Institute of Physics, Viet-namese Academy of Science and Technology, 10 DaoTan, BaDinh,Hanoi, Vietnam contrast to 10 . and 10 . cm − for the two lower rota-tional CO transitions. Therefore, the CO(3–2) line is aparticularly useful tracer of the molecular gas propertiesin the central regions of galaxies, where the moleculargas is generally believed to be warmer and denser thanin typical galactic disk clouds (e.g., G¨usten et al. 1981;Mauersberger & Henkel 1993). The CO(3–2) to (1–0)line intensity ratio is better suited to constrain the gastemperature and density than the ratio of CO(2–1) to(1–0).In the local universe, most of the evidence for a higherexcited gas phase comes from species other than CO.In many cases, however, such rare molecular species aredifficult to detect, particularly in higher excited tran-sitions. To investigate properties (e.g. spatial density,column density, kinetic temperature) of the bulk of thegas for a large sample of galaxies, observations of stronglines are needed. Thus, CO transitions of higher exci-tation have to be observed. This has been proved tobe very successful in the first extragalactic CO(3–2) sur-vey encompassing a significant (29) number of galaxies(Mauersberger et al. 1999), in which CO(3–2) was de-tected in all of the targets studied.Encouraged by this result (see also Devereux et al.1994), we have used the Heinrich Hertz Telescope (HHT)on Mt. Graham (Baars & Martin 1996) to observe theCO(3–2) line in an extended sample of galaxies. Afterwe started this extended project, additional extragalacticCO(3–2) surveys have been carried out, aiming at differ-ent types of galaxies. These include Virgo cluster galax- Mao, Schulz, Henkel et al.ies (Hafok & Stutzki 2003), infrared luminous galax-ies (Yao et al. 2003; Narayanan et al. 2005), early typegalaxies (Vila-Vilar´o et al. 2003), compact and dwarfgalaxies (Meier et al. 2001; Israel 2005), double barredgalaxies (Petitpas & Wilson 2003, 2004), nearby galax-ies of various types (Bayet et al. 2006), and most re-cently some nearby spiral and elliptical galaxies with twonewly mounted submillimeter telescopes in Chile, APEXand ASTE (Nakanishi et al. 2007; Komugi et al. 2007;Galaz et al. 2008). Some extended CO(3–2) maps werealso reported toward a handful of nearby galaxies (e.g.,Dumke et al. 2001; Muraoka et al. 2009; Warren et al.2010). Interferometric CO(3–2) maps of individualgalaxies are also feasible thanks to the Submillimeter Ar-ray (hereafter SMA; see, e.g., Wilson et al. 2008, 2009).However, the detailed study of warm gas in nearby galax-ies is still in its infancy. This is regrettable, in particularbecause of its crucial role in highly redshifted targets,where not CO(1–0) but higher excited CO lines are com-monly observed (e.g., Solomon & Vanden Bout 2005).In this paper, we present the results of our survey,which covers the by far largest sample of galaxies mea-sured so far in the CO(3–2) line. The data were ob-tained in a “coherent” way, making use of a specifictelescope/receiver/backend combination. With comple-mentary information from CO(1–0), taken from the lit-erature, we thus present a data base of unprecedentedsize, providing a suitable basis to check the quality ofour data (as well as those reported earlier) and allow-ing us to tackle a number of astrophysical questions.The data are used for the following main purposes: 1)to provide a large and homogeneous data set of CO(3–2) spectra, which will form an essential basis for fu-ture studies, either aiming at higher angular resolutionor searching for higher excited CO transitions see, e.g.,(Van der Werf et al. 2010) for a pioneering study), 2) tosystematically trace the global properties of the warmand dense molecular gas in various galaxies, 3) to testwhether there are any correlations between the molecu-lar gas excitation (given by the CO(3–2)/CO(1–0) inten-sity ratio) and galaxy properties such as Hubble type,nuclear activity, far infrared (FIR:40–400 µ m) luminos-ity, 60 µ m/100 µ color temperature of the dust, and in-clination, 4) to evaluate the effect of galaxy interactionson the molecular gas properties, 5) to test whether theCO(3–2) line is a better tracer of star formation than theCO(1–0) line, and 6) to evaluate the Schmidt-Kennicuttlaw in the light of the new data. We present the sampleselection in §
2, the observations in §
3, the basic resultsin §
4, a systematic analysis of correlations in §
5, and thesummary in § THE SAMPLES
Sample selection
Our sample selected for the CO(3–2) survey consists of125 galaxies which are part of five major partially over-lapping sub-samples. Table 1 lists all the sample galaxiesalong with some basic properties mostly drawn from the NED and HyperLEDA (Paturel et al. 2003).The first sub-sample consists of 58 nearby galax-ies of various types. It contains 22 reobservedsources that were already part of our initial survey(Mauersberger et al. 1999), and is complemented by theremaining IRAS point sources with S µm >
50 Jy and δ > − ◦ (Henkel, Wouterloot & Bally 1986) as well assources observed by Braine et al. (1993) in the CO (1–0)and (2–1) transitions (with the IRAM 30-m telescope) ifintegrated intensities are ≥
10 K km s − .The second sub-sample consists of 32 galaxies from avolume limited sample ( V < − ) of all Seyfertgalaxies and low-ionization nuclear emission-line regions(LINERs) in Huchra’s catalog of AGN (Huchra 1993) orin the V´eron-Cetty & V´eron (1991) catalog that are alsoincluded in the Revised Shapley Ames Catalog (RSA) andthat are accessible with the HHT (74 in total).The third sub-sample consists of 25 early type galax-ies from Henkel & Wiklind (1997) with CO(1–0) and/orCO(2–1) lines detected. Observations at radio, optical,and X-ray wavelengths have shown that early-type galax-ies contain an interstellar medium (ISM) comprising thesame components as found in spiral galaxies, but withdifferent mass fractions of the gas components (see alsoHenkel & Wiklind 1997 for a review).The fourth sub-sample consists of clearly iden-tified 11 interacting or merging systems, whichare mainly luminous infrared galaxies (LIRGs,10 L ⊙ ≤ L FIR < L ⊙ ). The selection is basedon their relatively high single dish and/or interfer-ometer CO(1–0) fluxes (see, e.g., Sanders et al. 1991;Gao et al. 1999; Lo et al. 2000, and references therein).These galaxy systems are thought to be at differentmerging/interaction phases, i.e. at the early (pre-sumably pre-starburst: Arp 303N/S, UGC 8335A/B,NGC 5257/8, Arp 302N/S, Arp 293 and NGC 6670A/B),intermediate (Arp 55, Mrk 848 and NGC 4038/9), or latestages of interaction (NGC 1614, NGC 5256), accordingto the spatial separation of the respective galaxy pair ineach system. The two core positions of five early mergers(not Arp 293) as well as NGC 4038/9 (the Antennae)were measured separately because they are spatiallyresolved by our 22 ′′ beam (see § § A.2.The members of the fifth sub-sample are prominentOH megamaser galaxies, including 4 ultraluminous in-frared galaxies (ULIRGs, L FIR ≥ L ⊙ : IRAS 17208-0014, Mrk 231, Mrk 273, and Arp 220) and 2 LIRGs(III ZW 35 and NGC 3690B), all of them being late merg-ers (following the classification outlined above), exceptNGC 3690B, which is “early”. Sample properties as a whole
The IRAS fluxes
98 of our sample galaxies are part of the
IRAS
Re-vised Bright Galaxy Sample (RGBS) by Sanders et al. The NASA/IPAC Extragalactic Database (NED) is operatedby the Jet Propulsion Laboratory, California Institute of Tech-nology, under contract with the National Aeronautics and SpaceAdministration. HyperLEDA database: http://leda.univ-lyon1.fr n extragalactic CO(3–2) survey 3(2003), which provides revised IRAS fluxes. 23 of thesewere determined with particularly high precision, alsoprofiting from the HIRES imaging reconstruction tech-nique (Surace et al. 2004). With the higher spatial res-olution obtained by this technique, four of our galaxypairs (i.e., Arp 303N/S, NGC 5257/8, UGC 8335A/B,and Arp 302N/S) are resolved and fluxes for each indi-vidual galaxy are available. The mid- to far-IR emis-sion of the NGC 4038/9 system (the Antennae) orig-inates predominantly from the overlap region wherethe disks of two galaxies interact (e.g., Schulz et al.2007). Flux densities directly obtained from IRAScatalogs, i.e. the IRAS Point Source Catalog (PSC;Joint IRAS Science Working Group 1988), the IRAS Ex-planatory Supplement (Beichman et al. 1988), and theIRAS Faint Source Catalog (FSC; Moshir et al. 1992)were taken for the rest of the sample. We then appliedthe flux densities to calculate the FIR luminosity ( L FIR = L (40–400 µ m), following Moshir et al. (1992), and the60 µ m/100 µ m color temperature ( T dust ), assuming anemissivity that is proportional to the frequency ν . Twogalaxies, IC 750 and NGC 4138, were not observed byIRAS. Galaxy classifications
With improved observations, galaxy classificationsmay have to be modified in some cases. We have usedthe NED classifications from August 2008 as standardthroughout the paper. While Seyferts/LINERs can bedirectly recognized from NED, starbursts are not explic-itly indicated. There exists a variety of definitions of thestarburst phenomenon in the literature, which have beenreviewed by, e.g., Heckman (2005) and Kennicutt et al.(2005). A starburst can be defined in terms of its ab-solute star forming rate (SFR), its SFR surface density(the SFR per unit area), or if its SFR exceeds an averagevalue from the past by a fixed amount. The situation isfurther complicated by the choice of the respective SFRtracer like, e.g., the ultraviolet emission, the far infraredemission, or the radio continuum. Spectroscopic tracerslike recombination lines (e.g., H α ) have also been fre-quently used.Here we select L FIR as the measure of the SFR and de-fine a starburst galaxy in terms of its SFR surface den-sity. Lacking high resolution information, we use withthe isophotal diameter D the ratio L FIR / D to deter-mine the SFR surface density and classify targets withlog ( L FIR / D ) ≥ ⊙ kpc − as starburst galaxies.This parameter is plotted as a function of log L FIR inFig. 1, where the boundary between starburst and non-starburst galaxies is marked by a dashed horizontal line.The borderline was chosen to ensure that most of the wellknown starburst galaxies are properly classified. Galax-ies, which were classified as starbursts in the literature(regardless of the details of the definition), are marked asstars. While all (U)LIRGs and most of the well knownstarburst galaxies are well above the borderline, there are19 galaxies that were “misclassified” (following our defi-nition) as non-starbursts and 7 galaxies that were “mis-classified” as starbursts. NGC 253, a typical starburstgalaxy, and IC 342, a galaxy similar to our Milky Waygalaxy, are part of Fig. 1 to ensure that the classificationmethod is correct. Both are located in the expected zone.The starburst sub-sample, selected as such, includes 24 classical starbursts, 16 Seyfert composites, 13 starburstsupported LINERs, 3 dwarf starburst galaxies ( M B > –18; e.g., NGC 1569 ), and all 28 (U)LIRGs, or in total77 galaxies (some of these galaxies have more than twoassignments).Characterized by FIR luminosity and nuclear activ-ity, the entire sample consists of 4 ULIRGs, 24 LIRGs,45 Seyferts, 45 LINERs, 49 starbursts neither beingULIRGs nor LIRGs, and 11 “normal” galaxies. Note thatone object may be part of more than one sub-sample.The sample classification is presented in more detail inTable 2.The sub-sample of Seyferts is severely biased to Seyfert2 galaxies, with only 7 galaxies classified as Seyfert 1.Although individually not satisfying the LIRG criterion10 L ⊙ ≤ L FIR < L ⊙ , Arp 303 S and N are bothclassified as LIRGs since the system as a whole meetsthe LIRG criterion. There are 16 Seyferts that are alsoclassified as LINERs. Hence there is a total of 74 AGN inour sample. Excluding those overlapping with the star-burst and (U)LIRG sub-samples, there remain 35 galax-ies which show “pure” AGN activity.The sample can also be broken down by Hubble types.We observed 42 early-type (including 19 lenticulars, 2ellipticals, 1 cD, and 20 early-type spirals) and 54 late-type galaxies (5 irregulars and 49 late type spirals),with a Hubble type index of t = 3 (or Sb in theRC3, de Vaucouleurs et al. 1991; Paturel et al. 2003) be-ing used as the dividing line ( t <
3: early type, t ≥ M B > –18). The sample distribution
In Fig. 2 we present some basic properties of the entiresample of 125 observed galaxies, i.e. the distribution ofHubble type, FIR luminosity ( L FIR ), 60 µ m/100 µ m dustcolor temperature ( T dust ), distance ( d p ), optical angularsize ( D in arcmin), optical linear size ( D in kpc),inclination angle ( i ), and absolute B-band magnitude( M B ). Our sub-samples cover almost all types of galax-ies, with most of them belonging to Hubble types 3 – 5(see de Vaucouleurs et al. 1991), corresponding to the re-vised (de Vaucouleurs) morphological types Sb–Sbc–Sc.The far-infrared luminosity of our sample spans almost5 orders of magnitude, log( L FIR /L ⊙ ) ∼ L FIR /L ⊙ ) ∼ T dust , ob-tained by assuming an emissivity propertional to ν (seefootnote to Table 1), varies between 24 K and 50 K, witha peak at about 35 K. The distance distribution shows astrong peak at 10–20 Mpc, where our beam size of 22 ′′ (see §
3) corresponds to a linear scale of about 1–2 kpc.These are typical sizes for circumnuclear starbursts. Theoptical diameter is in a range between 0.4 to 18.6 arcminon an angular and 1.3 to 78 kpc on a linear scale, withvalues of 25–40 kpc being most typical. About 70% of oursample galaxies have an optical diameter ( D ) smallerthan 5 arcmin, and the strong peak at 1–2 arcmin is dueto the merging sequence sub-sample (see § ∼ ◦ and90 ◦ . There are only few galaxies with inclinations below i ∼ ◦ . Our sample spans a B-band absolute magnituderange from –16.4 to –22.6, the majority having with M B < –20 a high luminosity. OBSERVATIONS
All the CO(3–2) observations were conducted withthe 10-m Heinrich-Hertz Telescope (HHT) on Mt. Gra-ham/Arizona with a beamwidth of 22 ′′ . Most ofour galaxies were observed during Feb., Apr. and Nov.1999, Jan. and Mar. 2000. The majority of the merg-ing/interacting galaxies was observed in Mar. 2003,Mar. 2004 and Mar. 2005. In all cases, the samedual channel 345 GHz SIS (Superconductor-Insulator-Superconductor) receiver was employed. Spectral pro-files were obtained with two acousto-optical spectrome-ters (AOSs), each with 2048 channels (channel spacing ∼
480 kHz, frequency resolution ∼
930 kHz, correspond-ing to a velocity resolution of ∼ − ) and a totalbandwidth of 1 GHz.Spectra were taken using a wobbling (2 Hz) secondarymirror with beam throws of ± ′′ to ± ′′ in azimuth.Scans obtained with reference positions on either side ofthe source were coadded to ensure flat baselines. Re-ceiver temperatures were of order of 170 K and systemtemperatures were ∼
900 K on a T ∗ A scale, respectively.Calibration at submillimeter wavelengths is often diffi-cult, especially for extragalactic observations, and needsto be carefully checked. The receivers were sensitive toboth sidebands. Any imbalance in the gains of the lowerand upper sideband would thus lead to calibration er-rors. To account for this, galactic calibration sources(e.g. Orion-KL, IRC+10216, Sgr B2, G34.3, and W51,depending on availability at the time of observation) wereobserved prior to the target source with the same re-ceiver tuning setup. Published spectral line survey datain the 345 GHz band were used for intensity calibrations,e.g. Schilke et al. (1997) for Orion-KL, Groesbeck et al.(1994) for IRC+10216, Sutton et al. (1991) for Sgr B2,Hatchell et al. (1998) for G34.3, and Wang et al. (1994)for W51. Pointing and focus were carefully checked be-fore the calibration spectra were taken. Nevertheless, theabsolute calibration error could be as large as ±
30% (see § T ∗ A wasconverted to main beam brightness temperature T mb via T mb = T ∗ A ( F eff / B eff ) (see, Downes 1989). The mainbeam efficiency, B eff , was 0.5 at 345 GHz, as obtainedfrom measurements of Saturn, and the forward hemi-sphere efficiencies, F eff , was 0.9 (see also Mao et al.2002).To reduce as much as possible the number of receivertunings, we used the same tuning setup to observe asmany galaxies as possible with similar velocities. There-fore, in some cases the line is detected well outside thecenter of the spectrum and sometimes even reaches theband edge of the backend, in which case only a zero or-der baseline subtraction was performed. Additionally, in The HHT was operated by the Submillimeter Telescope Ob-servatory on behalf of Steward Observatory and the Max-Planck-Institut f¨ur Radioastronomie. a few galaxies like Mrk 273, NGC 6240, IRAS 17208-0014, Arp 220, and Arp 302N, the full width to zeropower of the line is as wide as ∼ − (as shown inwide band interferometer data; e.g., Scoville et al. 1997),which exceeds the bandwidth of the backend used forour observations. The intensities in such cases can onlybe considered as lower limits, unless some concatenatedspectra were obtained, as in the case of Arp 220 andArp 302N. RESULTS
Spectra and line parameters
Figure 3 shows the CO(3–2) spectra (on a T mb scale)towards all detected galaxies. The line parameters orthe upper limits in case of non-detections are given inTable 3. The spectra have been smoothed to a velocityresolution of ∼ − in order to show the emissionfeatures more prominently. Spectra with the velocity in-tegrated intensity I = I CO(3 − = R T mb d v larger thanthree times the r.m.s noise are considered to be detected.We have determined the integrated line intensity, the ra-dial velocity and the line width using either Gaussian fitsto the lines, or the moments of the spectra in the caseof non-Gaussian line profiles. Spectra of Arp 220 andArp 302N were concatenated from two different velocitysetups to cover the full velocity ranges that exceed thebandwidth of the backend. Of the observed 125 galax-ies, 114 were detected, among which CO(3–2) data of 65galaxies are reported here for the first time. For spectrawith a signal to noise ratio of less than 3, an upper limitis derived using I < σ (∆ V δv ) / , where σ is ther.m.s noise in T mb for a single channel, and ∆ V repre-sents the full linewidth taken from the FCRAO CO(1–0)survey results (Young et al. 1995; Kenney et al. 1988) orarbitrarily set to 400 km s − if there was no CO(1–0)data available. δv denotes the channel spacing.The CO(3–2) luminosity, L CO(3 − in units ofK km s − pc , is calculated within our 22 ′′ beam by L CO(3 − = [ π/ (4 ln I d (1 + z ) − , (1)where Θ mb = 22 ′′ = 1.067 × − rad is the full widthto half maximum (FWHM) main beam size of the HHTat 345 GHz, d L = d c (1+ z ) is the luminosity distance inpc ( d p : proper distance in pc, see footnote to Table),and z = v hel /c denotes the redshift. The CO(1–0) lu-minosity is derived similarly (see footnotes in Table 3).Because of identical beam sizes (22 ′′ ), the averaged in-tensity ratio between the J =3–2 and 1–0 CO lines, R = I CO(3 − / I CO(1 − , is calculated for galaxies with avail-able IRAM-30m CO(1–0). For galaxies with CO(1–0)data from other telescopes, upper or lower limits aregiven for R , depending on the CO(1–0) beam size. Detection rates and non-detections
Our CO(3–2) detection rates are 91% (10/11), 86%(64/74), 100% (49/49), and 100% (28/28) in normal,Seyfert/LINER, starburst galaxies, and (U)LIRGs, re-spectively, or about 90% in total. Fortuitously, both thetotal number and the detection rate, 89% (39/45), are thesame for the Seyferts and LINERs. Those sample galax-ies that are known to host 22 GHz H O and/or 18 cmOH masers are all detected in CO(3–2). For the Virgon extragalactic CO(3–2) survey 5Cluster galaxies, dwarf galaxies, early type and late-typegalaxies (see § <
20K km s − ), with the only exception of NGC 4438where a relatively strong ( ∼
70 K km s − ) CO(1–0) inten-sity was reported. NGC 4438, together with NGC 2841and NGC 5866 were, however, observed at poor weatherconditions. NGC 7077 is considered to be a tentativedetection, since the central velocity differs by about140 km s − from that of the CO(1–0) line, while itsintegrated intensity marginally satisfies the detectioncriteria with an S/N ratio of 4 σ . One of our non-detections, NGC 855, a dwarf elliptical, was detected byNakanishi et al. (2007) after a deep integration. The re-ported intensity is below our 2 σ level and therefore wellbelow our detection limit.It is interesting to note that, except for one dwarf el-liptical (NGC 855), all the rest of the non-detections areAGN hosts (either Seyferts or LINERs), and are mostlyearly type galaxies (7 lenticulars and 3 spirals). Thisis suggestive of a possible destruction of molecular gasreservoirs by AGN feedback, and consequently a sup-pression of star formation in early type galaxies (see, e.g.,Schawinski et al. 2007). Except for 3 SAB galaxies, thenon-detections were all obtained from unbarred galaxies. Consistency of the observed CO intensities
As already mentioned in §
3, calibration uncertaintiesmay rise up to ± ′′ beam,any shift & ′′ could yield significant discrepancies inboth line shape and intensity. This could further in-crease the uncertainties of measured absolute intensitiesand requires a detailed comparison with data from previ-ous surveys with respect to both intensity and line shape.The large sample analyzed here brings us into the uniqueposition to test not only the quality of our own data butalso that of previously studied samples. The compari-son of spectroscopic results is given in the Appendix andstarts with previous measurements also obtained with a10-m sized telescope ( § A.1) and continues with the in-clusion of data from the James Clerk Maxwell 15-m tele-scope (JCMT, § A.2).In general, our results are consistent with publisheddata. The inconsistencies found for a few individualsources can be attributed to errors of pointing, calibra-tion and baseline subtraction (especially for broad spec-tra) which are difficult to quantify. A typical error of ∼
30% is not unusual even for millimeter observations.Therefore, the inconsistencies shown in the Appendixare within expected ranges, still leaving space for signif-icant improvements, possibly obtained by mapping thegalaxy cores. Among the sources with large discrepan-cies in intensities ( > § DISCUSSION
The CO(3–2)/(1–0) Line Intensity Ratio
The beam averaged integrated intensity ratio, R = I CO(3 − / I CO(1 − , can serve as an indicator of themolecular gas excitation since the ratio is sensitive tothe temperature and density of the molecular gas (e.g.,Mauersberger et al. 1999). Although the excitation sta-tus of the molecular gas cannot be accurately determinedwith only two optically thick transitions, i.e. CO(3–2)and (1–0), we can use R to constrain the molecular gastemperature and density with either Large Velocity Gra-dient (LVG) or Photon-Dominated Region (PDR) mod-els. While it is likely that R is varying within theregion observed (see, e.g., Dumke et al. 2001, who findthat CO(3–2) is more centrally concentrated), our R values provide a representative average over the size ofthe beam. Molecular Gas Excitation Traced by R Considering Λ = [ n (CO)/ n (H )]/(d v /d r ) =10 − and 10 − (km s − /pc) − , consistent withMauersberger et al. (1999), one-component LVG cal-culations (c.f. Scoville & Solomon 1974; Henkel et al.1980; Mao et al. 2000) with the recent collision ratesof Flower (2001) show that R ≥ T k &
60 K and an H density of n (H ) & . cm − .Both of these values are larger than those given inMauersberger et al. (1999), where old collision rateswere used. For the extreme case of R = 1.9, a gastemperature of at least ∼
200 K is required. A ratio of R = 0.2 would instead indicate n (H ) <
300 cm − forΛ = 10 − (km s − /pc) − , or n (H ) < − for Λ= 10 − (km s − /pc) − if T k = 20 – 60 K.The thermal budget of the interstellar molecular gasin starburst regions can be described predominantly interms of a PDR scenario (see, e.g., Hollenbach & Tielens1997; Mao et al. 2000; Schulz et al. 2007). If we take atypical strength of the incident far-ultraviolet (FUV) ra-diation field, G ∼ . − . (in units of the local galacticflux, 1.6 × − erg s − cm − , c.f. Mao et al. 2000, andreferences therein), as for the starburst in M82, the stan-dard PDR model (Kaufman et al. 1999) results in n (H )= 10 . − . cm − and cloud surface temperatures of 300– 600 K for R = 1.0 – 1.6.Among our 114 galaxies with detected CO(3–2) emis-sion, 68 have published IRAM-30m CO(1–0) data. Theseare the best candidates for a comparison (c.f. § ′′ ). Seven of thesegalaxies (marked with a superscript † in Col.(10) of Ta-ble 3) have been excluded because of a positional dis-crepancy by ≥ ′′ (for the coordinates used by us, see Mao, Schulz, Henkel et al.Table 1). Therefore, there remain 61 sources for ouranalysis. For those galaxies with no available IRAM-30m CO(1–0) data but with measurements from othertelescopes (i.e. NRAO-12m, FCRAO-14m, SEST-15m,Onsala-20m and NRO-45m), upper or lower limits wereestimated. The results are listed in Table 3.Figure 4 shows the distribution of the R ratio ofthe 61 sources. The distribution has a prominent peakaround 0.5, which is the typical value in the spiral armsof the Galactic disk (e.g., Oka et al. 2007), and an ad-ditional minor excess at about 1.5, which is a sign ofhighly excited gas as found in the Central MolecularZone (CMZ) of the Galaxy (for the Sgr A region see, e.g.,Oka et al. 2007). The values of R range widely from 0.2to 1.9 with a mean of 0.81 ± R distribution is deviating froma normal one, so that (following Chebyshev’s inequality)probabilities within a given range of standard deviationsare expected to be moderately lower than those in thecase of a normal distribution.There are 18 sources with R >
1, indicative of veryhigh excitation combined with low optical depth. As ex-pected, these are mostly starbursts or (U)LIRGs, sincesuch high R ratio gas may arise predominantly fromUV-irradiated surfaces of molecular clouds or shockedregions, possibly generated by the interaction with su-pernova expansion waves (Oka et al. 2007, and referencestherein), which are fairly common in starburst regions.NGC 3310, a starburst galaxy with an exceptionallyhigh CO(2–1)/(1–0) intensity ratio (2.6 ± R ratio (1.9 ± R ratio does not always need high excitation and has also been found in theGalactic interarm regions where low density gas domi-nates (Oka et al. 2007). However, such interarm regionsshould not dominate the overall CO emission of a spiralgalaxy. Alternatively, a warm opaque cloud veiled by acool foreground layer of low density, which absorbs theCO(1–0) but not the CO(3–2) emission, could also raisethe R ratio to an exceptional level without participa-tion of highly excited gas of low opacity (see below).Table 4 summarizes the results related to R for thedifferent galaxy types outlined in § R ratios. Correlations between R and galaxy properties In Fig. 5, R is shown as a function of Hubble type,distance (or linear beam size), inclination, FIR luminos-ity, 60 µ m/100 µ m dust color temperature, and opticalsize ( D ).1) Hubble type – There is no correlation between R and Hubble type (Fig. 5a, but see Nakanishi et al. 2007for elliptical galaxies). The bulk of the CO emission inthe majority of galaxies arises from the central region,which is largely decoupled from the Hubble type and theshape of the large scale disks (Kennicutt 1998). 2) Projected beam size – Fig. 5b shows no correlation,agreeing with Yao et al. (2003) on a similar analysis fortheir sample. Such a lack of correlation also holds withina given Hubble type of galaxies. It may imply that to-ward the nearby galaxies we see exclusively the nuclearregion, while in the more distant more luminous galaxiesthe central regions become so dominant that the largerprojected beam size is not important any more.3)
Inclination – Inclination may affect the observed R , since we tend to include more low excitation gasfrom the outer disk into the observing beam for galax-ies seen more edge-on, thus lowering R . Nevertheless,Fig. 5c shows no strong trend between R and the cosineof the inclination. However, the upper envelope of the R distribution as well as the number of sources withhigh R increase with decreasing inclination. While thisagrees with the expected trend, the correlation is not sig-nificant.4) FIR luminosity and color temperature of the dust –The FIR luminosity and the temperature of the dust areexpected to be correlated with the molecular gas excita-tion, if dust and gas are coupled. In Figs. 5d and e wetherefore correlate R with L FIR and the 60 µ m/100 u mdust color temperature ( T dust ). Although there may bea weak trend with T dust , similar to that mentioned abovefor the inclination, no significant correlation is evident.This could be partially attributed to the fact that weplot the global dust properties against the rather local-ized line ratio R emphasizing the nuclear regions. InFig. 5f we also plot R as a function of L FIR / D , ameasure of the SFR per unit area, but again there is noconvincing correlation (see also Yao et al. 2003). Thiscan be interpreted in the sense that star formation is alocally confined activity.5) Nuclear activity – In addition to star formation,AGN may also provide a source of heating for thesurrounding molecular gas (see, e.g., Matsushita et al.2004). As summarized in Table 4, the average R ratiosare 0.65 ± ± ± ± ± R ratios (0.89 ± ± R in excessof unity, 14 belong to starbursts. This indicates that thepresence of a nuclear starburst is a major reason for ahigh, beam averaged R value.6) Bars – Bars are expected to enhance the gas flowtoward the center of galaxies, building up the nucleargas reservoir to maintain nuclear activity and affect-ing the molecular gas excitation. We find that R ishigher in barred SB and SAB galaxies (0.88 ± ± ± The merging sequence – Most of the galaxies inthe merging sequence sub-sample have no correspondingn extragalactic CO(3–2) survey 7IRAM CO(1–0) data. However, in one of the interme-diate mergers, the Antennae, we do not see a significantenhancement of CO(3–2), with R being 0.7, 0.3 and 0.8for NGC 4038, NGC 4039 and the overlap region, whilethe mean ratio of the “late” mergers NGC 1614, Mrk 231,Arp 220, and IRAS 17208–0014 is 1.0 ± R values than “normal”galaxies, while R ratios are highest in starburst andbarred galaxies. Surprisingly, significant correlations of R with other galaxy properties are not found. FIR-to-CO luminosity correlations and starformation laws
The FIR luminosity (as a measure of the SFR in galax-ies) is correlated with the CO luminosity (a measure ofthe total molecular content). Based on the scaling re-lation SFR ∝ n Ngas , this can be expressed in terms ofthe so-called Kennicutt-Schmidt law (hereafter KS law;Schmidt 1959; Kennicutt 1998), which connects the SFRper area with the total gas surface density, i.e. Σ
SFR ∼ Σ N gas , with N = 1.4 ± s< s =1), or super-linear ( s>
1) correlation indices, whichdiffer from N (e.g, Kennicutt 1998; Gao & Solomon2004a,b; Baan et al. 2008; Bussmann et al. 2008).Recent models to interpret the various observed corre-lation indices in a uniform way (Krumholz & Thompson2007; Narayanan et al. 2008) assume that s dependson the cloud’s gas mass above the critical density ofthe observed molecular transition. The super-linearSFR– L CO(1 − correlation is caused by the fact thatthe gas density is on average higher than the CO(1–0)line’s critical density, thermalizing most of the CO(1–0) emission. The sub-linear SFR– L HCN(3 − correlation(Bussmann et al. 2008) is a result of a small fraction ofthermal emission as the critical density of the HCN J =3–2 transition is high ( n (H ) ∼ cm − ). Observationalsupport for such models is, however, still fragile giventhe large uncertainties in (sub)millimeter molecular linecalibration ( § A.1 and A.2) and additional problems men-tioned below. In the following, we will first address someof these problems and will then derive the correspondingcorrelations from our data.
Possible problems in the correlation analysis
For pointed CO observations, as in our case, a directpower law correlation of the nuclear CO luminosity withthe global FIR luminosity could be misleading becausethese two quantities refer to different spatial scales. Toevade such a situation, one should ideally obtain ex-tended CO maps to measure the entire molecular gascontent for a large sample of galaxies with various Hub-ble types and FIR luminosity ranges. Full maps of sucha large sample of galaxies in CO(3–2) are, however, notyet available.Yao et al. (2003) employed an alternative way in scal-ing down the total FIR luminosity as obtained by the IRAS data to the volume marked by the angular size ofthe CO observations. The scaling factor is determinedby the peak-to-total flux density ratio derived from thecorresponding 850 µ m submillimeter continuum images.The basic assumption for such a technique to be appli-cable is that the FIR brightness distribution is similarto that of the 850 µ m submillimeter continuum. This is,however, most likely not the case since the FIR emissionmeasured by the IRAS satellite is only sensitive to warmdust ( T dust &
30 K), while the submillimeter continuumalso traces dust at cooler color temperatures. In mostgalaxies, this cooler component dominates. Therefore,the FIR emission is expected to be much more centrallyconcentrated than the submillimeter emission, especiallyin the case of nuclear starbursts. Indeed, even the 200 µ mcontinuum emission has already shown significant colddust at large galactocentric radii (e.g., Alton et al. 1998;Kramer et al. 2010).Another important aspect is the linear regression fititself. Published correlation studies sometimes use L CO - L FIR and sometimes L FIR - L CO . Caution has to be exer-cised, however, in comparing these two approaches. Sim-ply taking an inverse slope (i.e., 1/ s instead of s ) is notappropriate because the standard linear regression fit as-sumes that X values are exactly correct, and that errorsor variability only affect the Y values. Hence the regres-sion of X on Y is different from the regression of Y on X . To determine the line dependent parameter s in anal-ogy to N ( § The FIR-to-CO correlation
In Fig. 6, we present the correlation between the nu-clear CO line luminosity and the global FIR luminosity L FIR . We performed the fit in both ways, with and with-out the method introduced by Kelly (2007), consideringuncertainties along both axes in the latter case. However,slopes derived with the Kelly packages are (within a per-cent) the same as those obtained with the unweightedlinear regression fits. This is perhaps due to the largenumber and small intrinsic scatter of our data along bothaxes in the log-log plots. In the following we will there-fore use the unweighted linear regression fits.Intriguingly, both nuclear CO(3–2) and CO(1–0) lineluminosities are tightly correlated to the global FIR lu-minosity, and our unweighted linear regression fits re-sult in almost identical slopes slightly below unity, s =0.87 ± ± § L CO - L FIR results in slopes of 0.96 ± ± L CO(3 − and L FIR values, is notincluded in the fits of Fig. 6. Fits including IC 10 give, Mao, Schulz, Henkel et al.however, similar slopes of 0.88 ± ± § A.1). The newfit gives the same slope, s = 0.87 ± L CO (y-axis)- L FIR (x-axis) slopes with scaled FIR luminosities ( § On the angular size dependence of the correlation
Instead of following Yao et al. (2003) and scaling downthe FIR fluxes to the area of the CO observations, we canalso test our results by defining different D ranges, as-suming that galaxies with similar optical angular sizeshave similar central (22 ′′ ) to total integrated CO in-tensity ratios. This assumption is based on the factthat the diameter of the CO(1–0)-emitting region D CO is found to be correlated with the optical diameter D as D CO /D ≈ . D ≤ ′ (filled circles), 2) 2 ′ < D ≤ ′ (empty triangles) and 3) 4 ′ < D ≤ ′ (crosses). Cor-responding slopes ( s ) and correlation coefficients ( r ) oflinear regression fits are also given. For a constant D CO /D ratio the resulting slopes should be close tothe case of FIR and CO emission arising from the sameregion.The three groups of galaxies show different correla-tions, irrespective of the CO transition studied. Thecompact galaxies show slopes of order 1.0–1.1, the inter-mediate sample is characterized by s ∼ ′′ ) integrated intensities is expected to besimilar.In summary, and without having to assume equal spa-tial distributions of the CO(3–2) and (1–0) emission,we obtain rising slopes s with increasing compactnessof the observed targets. This can be interpreted interms of the relationship between molecular line emis-sion and gas density, anchored by the underlying KS law(Gao & Solomon 2004b; Krumholz & Thompson 2007;Narayanan et al. 2008; Bussmann et al. 2008) as out-lined in § ∼ cm − , as the bulk of thenuclear CO emission should arise from a more diffusemedium ( n (H ) ∼ − cm − ; e.g., Mauersberger et al.1999; Mao et al. 2000). It is, however, difficult to in-terpret the correlation in the same way with CO(1–0)because the density of the bulk of the nuclear molec-ular gas may not be lower than ∼ . cm − . Clearly,this deserves further study. More beam matching CO(1–0) data, more maps providing a measure of the entireCO(3–2) emission of a galaxy, data from more than twoCO transitions, and infrared data of high angular reso-lution would thus be helpful. CONCLUSIONS
1. CO(3–2) spectra from the central region of a sampleof 125 galaxies are presented. With an angular resolu-tion of 22 ′′ , CO(3–2) emission is detected in 114 targets.Our survey significantly increases the number of avail-able CO(3–2) data from galaxies and provides a reliabledata base for future surveys with higher angular resolu-tion, establishing a bridge to the high J lines observedtoward redshifted targets.2. The CO(3–2)/(1–0) integrated line intensity ratio R varies widely from 0.2 to 1.9. The line ratio ap-pears to be independent of galaxy properties such asHubble type and FIR luminosity and only shows tenta-tive, not significant correlations with 60 µ m/100 µ m dustcolor temperature and inclination angle.3. To be consistent with common designations butto use at the same time a clear definition, we havespecified the term “starburst galaxy” by the conditionlog [( L FIR /L ⊙ )/( D /kpc )] > R ratios are found to be largerin galaxies with nuclear activity (AGN and starbursts)or with bars than in those without. Apparently, theseare the galaxies showing enhanced molecular excitation.Most galaxies with a line ratio of R ≥ R value, which maybe caused by particularly vigorous activity triggered bygalaxy interaction and merging.5. The nuclear CO luminosities show a slightly sub-linear correlation with the global FIR luminosity in boththe CO(3–2) and the (1–0) lines. Subdividing our sam-ple into several bins with different angular sizes to com-pensate for the different size of the regions from whereCO and FIR emission have been measured reveals signif-icant differences. Compact and thus mostly distant lu-minous galaxies show the largest slopes, possibly a con-sequence of relatively high overall molecular densities,yielding larger fractions of thermalized gas. A similartrend for CO(1–0) is more difficult to explain, becausethis would require densities below 10 . cm − . ACKNOWLEDGEMENT n extragalactic CO(3–2) survey 9We wish to thank an anonymous referee for criticallyreading the manuscript. We also wish to thank the HHTstaff for their enthusiastic support of the project and fortheir flexibility in changing schedules according to vari-able weather conditions. We also thank Dr. JiangshuiZhang for his help in preparing some of the tables andDr. B. Kelly for a latest version of his software package.We acknowledge useful discussions with Drs. M. Dumkeand A. Weiß. RQM is partly supported by NSFC undergrants 10373025 and 10733030.
APPENDIX
A.1. COMPARISON OF CO(3–2) SPECTRA TAKENWITH THE SAME ANGULAR RESOLUTION
1) A comparison of 22 galaxies with the data byMauersberger et al. (1999, HHT-M99) shows goodagreement within the errors. Eight sources werere-observed at different positions, four of them atpositions displaced by more than 10 ′′ (NGC 3627,NGC 3628, NGC 6946, and NGC 7541), with thenew data showing more symmetric profiles andstronger intensities. The significant difference inobserved line shapes toward NGC 2146 is likely dueto a position offset of 6 ′′ , and the flat-topped pro-file of Mauersberger et al. (1999) looks more like aline from the center of the galaxy than our sharplypeaked profile. For most of the other galaxies(NGC 3227, NGC 3351, NGC 3368, NGC 4414,NGC 4818), our new data show higher quality pro-files although the integrated intensities are quiteconsistent.2) Dumke et al. (2001, HHT-D01) presented extendedCO(3–2) maps toward nine of our sample galaxieswhere we can check the pointing by comparing ourline profile with their individual spectra. For Maf-fei 2, M 82, M 51, and NGC 6946, Dumke et al.(2001) present spectra, which are consistent withours with respect to both lineshape and intensity.For NGC 3628 and M 83, our profiles and intensi-ties resemble theirs at the (10 ′′ ,0 ′′ ) and (–10 ′′ ,0 ′′ )offsets, respectively, where the line intensities areabout 30% weaker than the peak intensities at their(0 ′′ ,0 ′′ ) positions. Relatively large discrepancies ex-ist for three sources. The line shape of NGC 4631resembles that given by Dumke et al. (2001) buttheir intensity is about twice as high. The othertwo are observed at positions slightly different fromthe nominal position used by Dumke et al. (2001).Our position of NGC 891 corresponds to their (–6 ′′ ,–7 ′′ ) offset position, where their integrated in-tensity of about 50 K km s − is tripling our value(17.3 K km s − ). The position we observed forNGC 2146 is 5 ′′ east of their reference position, butour spectrum looks more like theirs at the (10 ′′ ,–10 ′′ ) offset position. At this position, their inten-sity is about twice as strong as ours. The spec-trum by Mauersberger et al. (1999) is consistentwith Dumke et al. (2001) with respect to both lineshape and intensity, although the nominal positionsdiffer by about 10 ′′ . Our observations of all thesethree source seem to suffer from large calibrationerrors, and NGC 2146 may suffer from an addi-tional pointing error. 3) Vila-Vilar´o et al. (2003, HHT-V02) covered fiveearly type galaxies of our sample, with NGC 404and NGC 4691 being slightly stronger (by 25% and5%, respectively than our spectra. NGC 855 andNGC 5666 are non-detections in both data sets(NGC 5666 was a tentative detection in their pa-per). Our spectrum of NGC 3593 looks more likethat of their (10 ′′ ,0 ′′ ) position which gives an in-tensity twice as strong as ours.4) Narayanan et al. (2005, HHT-N05) observed threeof our sample galaxies. While their integrated in-tensities for NGC 3079 and Arp 220 are about twiceas large as our values, IRAS 17208-0014 shows lessthan half of the strength we got. A detailed checkis not possible, however, since their given positions(their Table 1) are erroneous.5) For the six common sources also observed byBayet et al. (2006, CSO-B06) with the CSO-10mtelescope, we find consistent results for NGC 3079,NGC 6946 and the Antennae system. Our intensi-ties of Mrk 231, M83, and Arp 220 are, however,all about twice as large as theirs. Their CO(3–2)spectrum of NGC 4736 looks more symmetric thanours and has an intensity, which is 40% higher.6) Among the five common sources also observedby Komugi et al. (2007, ASTE-K07) with theASTE-10m telescope, four galaxies (NGC 1068,NGC 1084, NGC 1087 and NGC 7479) were ob-served at similar (offsets ≤ ′′ ) positions and showquite consistent results. The large discrepancy inthe case of NGC 157 is due to a position offset ofabout one telescope beam (22 ′′ ). Our spectra showgenerally better baselines thanks to the backend,which is twice as wide as theirs.To evaluate consistencies on a quantitative basis, wedefine a relative intensity deviation as log( I ′ /I ), where I ′ is the integrated CO(3–2) intensity obtained from theliterature and I is from this work. Figure 8 shows therelative intensity deviations for galaxies with CO(3–2)data available in the articles mentioned above. About80% of the data points, excluding those observed at nom-inal position offsets & ′′ (open squares in Fig. 8), arefalling into the ± A.2. COMPARISON OF CO(3–2) SPECTRA TAKENWITH DIFFERENT ANGULAR RESOLUTION
Comparisons with observations at different angularresolution are not straightforward and need to be treatedwith utmost caution, since the molecular gas is rarelysmoothly distributed in galaxies, and any simple scalingcould easily become artificial. For the galaxies with moreor less known structure, comparisons of line profiles andintensities can, however, still be helpful to check consis-tency.In Fig. 9, we compare our results with publishedCO(3–2) data taken with the JCMT-15m telescope byYao et al. (2003, JCMT-Y03) and Wilson et al. (2008,JCMT-W08). Two straight lines denote the theoreti-cal relationship of intensities obtained with the JCMT0 Mao, Schulz, Henkel et al.and HHT assuming point-like (dashed) and uniformly-extended (dotted) structures with respect to the observ-ing beams. Most of the sources are located betweenthese two lines, as expected in case of well calibratedintensities. In general, the JCMT-15m CO(3–2) datatend to yield higher intensities, as expected, given theirhigher angular resolution. This also holds for M 83(Muraoka et al. 2009) and NGC 3521 and NGC 3627(Warren et al. 2010), with the former two galaxies re-vealing integrated CO(3–2) intensity compatible with ourvalues, while their CO(3–2) emission peak toward thelatter is stronger by a factor of 2.5.Toward the interacting Arp 302 N/S system, Yao et al.(2003) observed a position in between the pair of nuclei,where the CO emission is weak and where we only ob- tained an upper limit (indicated by the arrow pointingtowards the left). This refers to “Arp 302 center” inTables 1 and 3. We also observed this system at thepositions of its two nuclei (i.e., Arp 302 N/S), whereemission is stronger. Sources, where JCMT CO(3–2)intensities were derived from maps, are labeled by ar-rows pointing downwards. Our CO(3–2) intensity ofNGC 3690 (or Arp 299 in Wilson et al. 2008) is a sumof NGC 3690 A and B. Differences between the JCMTand our HHT integrated intensities are most pronouncedtoward Mrk 848 and NGC 5258. While in the case ofMrk 848 this may be due to differences in calibration,the results from NGC 5258 may be caused by an unusualgas morphology (see also Wilson et al. 2008).
REFERENCESAlbrecht, M., Chini, R., Kr¨ugel, E., M¨uller, S. A. H., & Lemke,R. 2004, A&A, 414, 141Albrecht, M., Kr¨ugel, E., & Chini, R. 2007, A&A, 462, 575Alton, P. B., et al. 1998, A&A, 335, 807Baan, W. A., Henkel, C., Loenen, A. F., Baudry, A., & Wiklind,T. 2008, A&A, 477, 747Baars, J. W. M., & Martin, R. N. 1996, Rev. Mod. Astro., 9, 111Balick, B., & Heckman, T. 1981, A&A, 96, 271Bayet, E., Gerin, M., Phillips, T. G., & Contursi, A. 2006, A&A,460, 467Beichman, C. A. 1987, ARA&A, 25, 521Beichman, C. A., Neugebauer, G., Habing, H. J., Clegg, P. E., &Chester, T. J., eds. 1988, IRAS Catalogs andAtlasesExplanatory Supplement (NASA RP-1190)(Washington, DC: GPO)Benedict, G. F., Howell, D. A., Jørgensen, I., Kenney, J. D. P., &Smith, B. J. 2002, AJ, 123, 1411Braine, J., & Combes, F. 1992, A&A, 264, 433Braine, J., Combes, F., Casoli, F., Dupraz, C., Gerin, M., Klein,U., Wielebinski, R., & Brouillet, N. 1993, A&AS, 97, 887Brunthaler, A., Castangia, P., Tarchi, A., Henkel, C., Reid, M. J.,Falcke, H., & Menten, K. M. 2009, A&A, 497, 103Bussmann, R. S., et al. 2008, ApJ, 681, L73Casoli, F., Combes, F., Augarde, R., Figon, P., & Martin, J. M.1989, A&A, 224, 31Chini, R., Kr¨ugel, E., & Lemke, R. 1996, A&AS, 118, 47Chini, R., Kr¨ugel, E., & Steppe, H. 1992, A&A, 255, 87Combes, F., Casoli, F., Encrenaz, P., Gerin, M., & Laurent, C.1991, A&A, 248, 607Combes, F., Young, L. M. & Bureau M. 2007, MNRAS, 377, 1795Dahlem, M., Heckman, T. M., Fabbiano, G., Lehnert, M. D., &Gilmore, D. 1996, ApJ, 461, 724de Vaucouleurs, G., de Vaucouleurs, A., Corwin, H. G., Jr., Buta,R. J., Paturel, G., & Fouque, P. 1991, Third ReferenceCatalogue of Bright Galaxies (New York: Springer)Devereux, N., Taniguchi, Y., Sanders, D. B., Nakai, N., & Young,J. S. 1994, AJ, 107, 2006Downes D. 1989, Evolution of Galaxies: AstronomicalObservations, Lecture Notes in Physics 333, eds. I. Appenzeller,H. Habing, P. L´ena, Springer Verlag, Berlin, p353Downes, D., Radford, S. J. E., Guilloteau, S., Guelin, M., Greve,A., & Morris, D. 1992, A&A, 262, 424Dumke, M., Nieten, C., Thuma, G., Wielebinski, R., & Walsh, W.2001, A&A, 373, 853Elfhag, T., Booth, R. S., Hoeglund, B., Johansson, L. E. B., &Sandqvist, A. 1996, A&AS, 115, 439Flower, D. R. 2001, Journal of Physics B Atomic MolecularPhysics, 34, 2731Galaz, G., Cort´es, P., Bronfman, L., & Rubio, M. 2008, ApJ,677, L13Gao, Y., & Solomon, P. M. 1999, ApJ, 512, L99Gao, Y., & Solomon, P. M., 2004a, ApJ, 606, 271Gao, Y., & Solomon, P. M., 2004b, ApJS, 152, 63.Gerin, M., & Phillips, T. G. 2000, ApJ, 537, 644Golla, G., & Wielebinski, R. 1994, A&A, 286, 733 Graci´a-Carpio, J., Garc´ıa-Burillo, S., Planesas, P., & Colina, L.2006, ApJ, 640, L135Greve, A., Becker, R., Johansson, L. E. B., & McKeith, C. D.1996, A&A, 312, 391Groesbeck, T. D., Phillips, T. G., & Blake, G. A. 1994, ApJS,94, 147G¨usten, R., Walmsley, C. M., & Pauls, T. 1981, A&A, 103, 197Hafok, H., & Stutzki, J. 2003, A&A, 398, 959Handa, T., Nakai, N., Sofue, Y., Hayashi, M., & Fujimoto, M.1990, PASJ, 42, 1Hatchell, J., Thompson, M. A., Millar, T. J., & MacDonald,G. H. 1998, A&AS, 133, 29Heckman, T. M., Blitz, L., Wilson, A. S., Armus, L., & Miley,G. K., 1989, ApJ, 342, 735Heckman, T. M. 2005, in Astrophys. Space Sci. Library 329,Starbursts: From 30 Doradus to Lyman Break Galaxies, eds. R.de Grijs & R. M. Gonz´alez Delgado (Dordrecht: Springer), 3Henkel, C., Walmsley, C. M., & Wilson, T. L. 1980, A&A, 82, 41Henkel, C., & Wiklind, T. 1997, Space Science Reviews, 81, 1Henkel, C., Wouterloot, J. G. A., & Bally, J. 1986, A&A, 155,193Hollenbach, D. J., & Tielens, A. G. G. M. 1997, ARA&A, 35, 179Huchra, J. 1993, electronic version of “Catalogue of SeyfertGalaxies and Other Bright AGN”Iono, D. et al. 2009, ApJ, 695, 1537Israel, F. P. 2005, A&A, 438, 855Joint IRAS Science Working Group. 1988. IRAS Catalogs andAtlases: The Point Source Catalog, Version 2.0 (NASARP-1190) (Washington, DC: GPO) (PSC)Kaufman, M. J., Wolfire, M. G., Hollenbach, D. J., & Luhman,M. L. 1999, ApJ, 527, 795Kelly, B. C. 2007, ApJ, 665, 1489Kenney, J. D. & Young, J. S. 1988, ApJS, 66, 261Kennicutt, R. C., Jr. 1998, ARA&A, 36, 189Kennicutt, R. C., Jr., Lee, J. C., Funes, J. G., Sakai, S., &Akiyama, S. 2005, in Astrophys. Space Sci. Library 329,Starbursts: From 30 Doradus to Lyman Break Galaxies, ed. R.de Grijs & R. M. Gonz´alez Delgado (Dordrecht: Springer), 187Komugi, S., Kohno, K., Tosaki, T., Nakanishi, H., Onodera, S.,Egusa, F., & Sofue, Y. 2007, PASJ, 59, 55Komugi, S., Sofue Y., Kohno K., Nakanishi H., Onodera S.,Egusa F., & Muraoka K. 2008, ApJS, 178, 225Kramer, C. et al. 2010, A&A, 518, L67Kregel, M., & Sancisi, R. 2001, A&A, 376, 59Krumholz, M. R., & Thompson, T. A. 2007, ApJ, 669, 289Kuno, N., et al. 2007, PASJ, 59, 117Leech, J., Isaak, K. G., Papadopoulos, P. P., Gao, Y., & Davis,G. R. 2010,MNRAS, 406, 1364Lo, K. Y., Hwang, C. Y., Lee, S. W., Kim, D.-C., Wang, W. H.,Lee, T. H., Gruendl, R., & Gao, Y. 2000, in ASP Conf. Ser.197, Dynamics of Galaxies: from the Early Universe to thePresent, ed. F. Combes, G. A. Mamon & V. Charmandaris(San Francisco, CA:ASP), 279Mao, R. Q., Henkel, C., Schulz, A., Zielinsky, M., Mauersberger,R., St¨orzer, H., Wilson, T. L., & Gensheimer, P. 2000, A&A,358, 433 n extragalactic CO(3–2) survey 11
Mao, R. Q., Yang, J., Henkel, C., & Jiang, Z. B. 2002, A&A,389, 589Matsushita, S., et al. 2004, ApJ, 616, L55Mauersberger, R., & Henkel, C. 1993, Rev. M.A., 6, 69Mauersberger, R., Henkel, C., Walsh, W., & Schulz, A. 1999,A&A, 341, 256Mei, S., et al. 2007, ApJ, 655, 144Meier, D. S., Turner, J. L., Crosthwaite, L. P., & Beck, S. C.2001, AJ, 121, 740Meier, D. S., Turner, J. L., & Beck, S. C. 2001, AJ, 122, 1770Moshir, M., Kopman, G., & Conrow, T. A. O. 1992, IRAS FaintSource Survey, Explanatory Supplement, version 2 (Pasadena,CA: Infrared Processing and Analysis Center, Cal. Tech.)Muroaka, K. et al. 2009, ApJ, 706, 1213Nakanishi, H., Tosaki, T., Kohno, K., Sofue, Y., & Kuno, N.2007, PASJ, 59, 61Narayanan, D., Cox, T. J., Shirley, Y., Dav´e, R., Hernquist, L., &Walker, C. K. 2008, ApJ, 684, 996Narayanan, D., Groppi, C. E., Kulesa, C. A., & Walker, C. K.2005, ApJ, 630, 269Nishiyama, K., & Nakai, N. 2001, PASJ, 53, 713Oka, T., Nagai, M., Kamegai, K., Tanaka, K., & Kuboi, N. 2007,PASJ, 59, 15Paturel, G., Petit, C., Prugniel, P., Theureau, G., Rousseau, J.,Brouty, M., Dubois, P., & Cambr´esy, L. 2003, A&A, 412, 45Petitpas, G. R., & Wilson, C. D. 2003, ApJ, 587, 649Petitpas, G. R., & Wilson, C. D. 2004, ApJ, 603, 495Radford, S. J. E., Downes, D., & Solomon, P. M. 1991, ApJ, 368,L15Reuter, H. P., Pohl, M., Lesch, H., & Sievers, A. W. 1993, A&A,277, 21Reuter, H.-P., Sievers, A. W., Pohl, M., Lesch, H., & Wielebinski,R. 1996, A&A, 306, 721Sage, L. J., Salzer, J. J., Loose, H.-H., & Henkel, C. 1992, A&A,265, 19Salom´e, P., et al. 2006, A&A, 454, 437Sanders, D. B., Mazzarella, J. M., Kim, D.-C., Surace, J. A., &Soifer, B. T. 2003, AJ, 126, 1607Sanders, D. B., Scoville, N. Z., & Soifer, B. T. 1991, ApJ, 370,158Sanders, D. B., Scoville, N. Z., Young, J. S., Soifer, B. T.,Schloerb, F. P., Rice, W. L., & Danielson, G. E. 1986, ApJ,305, L45 Sarzi, M., Allard, E. L., Knapen, J. H., & Mazzuca, L. M. 2007,MNRAS, 380, 949Sawada, T. et al. 2001, ApJS, 136, 189Schawinski, K., Thomas, D., Sarzi, M., Maraston, C., Kaviraj, S.,Joo, S.-J., Yi, S. K., & Silk, J. 2007, MNRAS, 382, 1415Schilke, P., Groesbeck, T. D., Blake, G. A., & Phillips, T. G.1997, ApJS, 108, 301Schmidt, M. 1959, ApJ, 129, 243Schulz, A., Henkel, C., Muders, D., Mao, R. Q., R¨ollig, M., &Mauersberger, R. 2007, A&A, 466, 467Scoville, N. Z., & Solomon, P. M. 1974, ApJ, 187, L67Scoville, N. Z., Yun, M. S., & Bryant, P. M. 1997, ApJ, 484, 702Sil’chenko, O. K., & Afanasiev, V. L. 2002, A&A, 385, 1Solomon, P. M., Downes, D., & Radford, S. J. E. 1992, ApJ, 387,L55Solomon, P. M., Downes, D., Radford, S. J. E., & Barrett, J. W.1997, ApJ, 478, 144Solomon, P. M., & Vanden Bout, P. A. 2005, ARA&A, 43, 677Surace, J. A., Sanders, D. B., & Mazzarella, J. M. 2004, AJ, 127,3235Sutton, E. C., Jaminet, P. A., Danchi, W. C., & Blake, G. A.1991, ApJS, 77, 255Van der Werf, P. et al. 2010, A&A, 518, L42V´eron-Cetty, M. P., & V´eron, P. 1991, electronic version of “ACatalog of Quasars and Active Nuclei” (5th ed.; Garching:ESO Sci. Rep. No. 10)Vila-Vilar´o, B., Cepa, J., & Butner, H. M. 2003, ApJ, 594, 232Vila-Vilar´o, B., Taniguchi, Y., & Nakai, N. 1998, AJ, 116, 1553Wang, Y., Jaffe, D. T., Graf, U. U., & Evans, N. J., II 1994,ApJS, 95, 503Warren, B. E. et al. 2010, ApJ, 714, 571Wiklind, T., Combes, F., & Henkel, C. 1995, A&A, 297, 643Wiklind, T., & Henkel, C. 1989, A&A, 225, 1Wilson, C. D., et al. 2008, ApJS, 178, 189Wilson, C. D., et al. 2009, ApJ, 693, 1736Yao, L., Seaquist, E. R., Kuno, N., & Dunne, L. 2003, ApJ, 588,771Young, J. S., et al. 1995, ApJS, 98, 219
TABLE 1
The HHT extragalactic CO(3–2) survey sample and galaxy propertiesNo. SOURCE R.A.
DEC v hel d p D i M B log L FIR T dust Classification( h m s ) ( ◦ ′ ′′ ) (km s − ) (Mpc) ( ′ ) (deg) (mag) (L ⊙ ) (K)(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14)1 IC 10 00 20 27.4 +59 17 15 –348 0.7 a · · · c S 9.9 dIrr IV/BCD2 NGC 157 00 34 45.3 –08 23 51 1652 22.6 4.2 62 –21.3 10.4 33.1 S 4.0 SAB(rs)bc HII3 NGC 404 01 09 27.7 +35 43 08 –48 3.1 a c –2.8 SA(s)0-: LINER4 NGC 660 01 43 02.4 +13 38 45 850 11.6 8.3 78 –19.1 10.4 36.9 S 1.3 SB(s)a pec;HII LINER5 III ZW 35 01 44 30.7 +17 06 09 8225 112.0 0.4 · · · · · · · · · LIRG Sy26 NGC 855 02 14 03.8 +27 52 38 595 9.7 a c N –4.6 E7 NGC 891 02 22 32.5 +42 20 48 528 9.8 a a · · · c S 4.0 SAB(rs)bc10 NGC 1055 02 41 45.2 +00 26 39 994 13.6 7.6 63 –19.6 10.1 31.3 3.2 SBb: sp LINER211 NGC 1068 02 42 40.8 –00 00 47 1137 15.6 7.1 21 –21.3 11.0 41.5 L 3.0 (R)SA(rs)b;Sy1 Sy212 NGC 1084 02 46 00.1 –07 34 38 1407 19.3 3.2 46 –20.5 10.5 35.1 S 4.9 SA(s)c HII13 NGC 1087 02 46 25.1 –00 29 53 1517 20.8 3.12 33 –20.5 10.2 33.4 S 5.2 SAB(rs)c14 NGC 1275 03 19 48.2 +41 30 42 5264 71.9 2.2 58 –22.5 10.9 46.3 S –2.2 cD;pec;NLRG Sy215 NGC 1530 04 23 26.6 +75 17 43 2461 33.7 4.6 58 –21.4 10.5 32.0 N 3.1 SB(rs)b16 NGC 1569 04 30 46.5 +64 51 01 –104 2.0 a p
04 34 00.0 –08 34 44 4778 65.2 1.3 42 –21.3 11.5 45.5 L 4.9 SB(s)c pec;HII:Sy218 NGC 1637 04 41 28.3 –02 51 28 717 12.0 a c S 4.5 SB(s)bc pec:21 NGC 2655 08 55 38.8 +78 13 28 1400 19.2 4.9 66 –21.1 9.3 29.6 c c –4.3 S0 1/2 LINER24 Arp 55 p
09 15 54.7 +44 19 49 11782 160.1 1 59 –21.6 11.6 37.3 L 5.2 LINER;LIRG HII25 NGC 2782 09 17 15.7 +39 54 14 2543 34.8 3.5 45 –20.8 10.4 39.2 S 1.1 SAB(rs)a;Sy1 Sbrst26 NGC 2841 09 22 02.7 +50 58 36 638 8.7 8.1 68 –20.7 9.0 25.3 c p
09 46 20.3 +03 02 44 6002 81.9 1.06 73 –20.5 10.7 b b L 2.0 SB(r)ab:pec29 Arp 303N p
09 46 21.1 +03 04 17 5990 81.7 1.92 77 –21.6 10.8 b b L 6.8 SA(s)cd? pec VLIRG30 NGC 2985 09 50 20.9 +72 16 44 1322 18.1 4.6 38 –20.7 9.9 29.5 2.3 (R’)SA(rs)ab LINER31 NGC 3032 09 52 08.2 +29 14 29 1533 21.0 2 26 –18.8 9.4 32.9 c N –1.8 SAB(r)0 ∧ ∧ HII32 NGC 3034 09 55 52.6 +69 40 47 203 2.8 11.2 79 –18.0 10.4 48.8 S 8.0 I0;Sbrst HII33 NGC 3079 10 01 58.2 +55 40 43 1116 15.3 7.9 83 –21.4 10.5 34.7 S 6.5 SB(s)c;LINER Sy234 NGC 3077 10 03 21.1 +68 44 02 14 3.8 a c S 9.9 Im pec HII41 NGC 3310 10 38 46.1 +53 30 08 993 13.6 3.1 31 –20.0 10.2 41.9 S 4.0 SAB(r)bc pec HII42 NGC 3351 10 43 57.3 +11 42 16 778 10.5 a a a · · · –21.1 · · · · · ·
49 NGC 3628 11 20 17.0 +13 35 20 843 11.5 14.8 79 –21.3 10.3 35.5 3.1 SAb pec sp;HII LINER50 NGC 3642 11 22 18.4 +59 04 34 1588 21.7 1.76 32 –20.5 9.4 30.6 c S 4.0 SA(r)bc: LINER Sy351 NGC 3682 11 27 42.7 +66 35 25 1515 20.7 1.7 52 –18.8 9.6 33.6 c S 0 SA(s)0/a:?52 NGC 3690A p
11 28 30.9 +58 33 44 3064 41.9 2.9 44 –20.1 11.6 47.3 L 8.7 IBm pec HII53 NGC 3690B p
11 28 33.6 +58 33 46 3121 42.7 2.9 44 –20.1 11.6 47.3 L 8.7 SBm? pec HII54 NGC 3810 11 40 58.9 +11 28 20 993 13.6 1.69 48 –20.0 9.9 32.1 S 5.2 SA(rs)c HII55 NGC 3982 11 56 28.1 +55 07 30 1109 22.0 a · · · · · · c N 2.2 Sab: sp57 NGC 4038 p
12 01 53.0 –18 52 03 1642 22.0 a · · · · · · S 8.9 SB(s)m pecNGC 4038/9 12 01 55.1 –18 53 00 22.0 a · · · · · · · · · b b · · · · · · (the overlap region)58 NGC 4039 p
12 01 53.6 –18 53 11 1641 22.0 a · · · · · · S 8.9 SA(s)m pecLINERSbrst59 NGC 4102 12 06 22.6 +52 42 39 846 11.6 2.7 58 –19.2 10.2 39.2 S 3.0 SAB(s)b?;HII LINER60 NGC 4138 12 09 30.7 +43 41 16 888 12.2 2.6 64 –18.7 · · · · · · –0.8 SA(r)0+ Sy1.961 NGC 4192 v
12 13 48.3 +14 54 01 –142 16.5 a v
12 18 49.5 +14 25 03 2407 32.9 5.4 32 –22.5 10.5 32.6 S 5.2 SA(s)c63 NGC 4258 12 18 57.5 +47 18 14 448 8.4 a c v
12 21 12.9 +18 22 58 893 12.2 5.6 59 –20.0 9.6 33.4 c v
12 21 54.7 +04 28 20 1566 16.5 a c v
12 22 55.2 +15 49 22 1571 16.5 a a c a v
12 27 45.8 +13 00 30 71 16.5 a v
12 28 59.3 +03 34 16 882 16.5 a a v
12 34 08.7 +02 39 11 1736 16.5 a c n extragalactic CO(3–2) survey 13 TABLE 1 (continued)No. SOURCE R.A.
DEC v hel d p D i M B log L FIR T dust Classification( h m s ) ( ◦ ′ ′′ ) (km s − ) (Mpc) ( ′ ) (deg) (mag) (L ⊙ ) (K)(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14)77 NGC 4631 12 42 07.6 +32 32 28 606 8.3 15.5 85 –22.1 10.2 35.9 N 6.6 SB(s)d78 NGC 4639 v
12 42 52.4 +13 15 26 1018 13.9 2.8 52 –19.1 9.0 29.9 c v
12 43 56.1 +13 07 43 1046 16.5 a v
12 49 39.0 +15 09 55 1125 16.5 a a a
10 60 –20.5 9.8 33.8 2.4 (R)SA(rs)ab;HII Sy287 NGC 4941 13 04 12.9 –05 33 07 1108 15.1 3.6 36 –19.3 9.0 30.4 c a p
13 15 30.8 +62 07 45 9230 125.6 0.85 · · · · · · · · · · · · L · · · SC; LINER HII91 UGC 8335B p
13 15 35.0 +62 07 29 9313 126.8 1.47 · · · · · · b b L · · · SC HII92 Arp 193 p
13 20 35.3 +34 08 25 6985 95.2 1.5 62 –20.5 11.6 40.0 L 9.9 Im: pec;HII LINER93 NGC 5194 13 29 52.5 +47 11 53 463 7.1 a a · · · p
13 38 17.5 +48 16 37 8353 113.8 1.22 · · · · · ·
Pec;Sy2;LIRG Sbrst96 NGC 5257 p
13 39 52.9 +00 50 24 6798 92.7 1.8 62 –21.8 11.3 b b L 3.1 SAB(s)b pec;HII LIRG97 NGC 5258 p
13 39 57.7 +00 49 51 6757 92.1 1.7 34 –21.3 11.0 b b L 3.1 SA(s)b pec;HII LINER98 NGC 5273 13 42 08.3 +35 39 15 1064 14.6 2.8 57 –18.8 8.8 40.4 c –1.9 SA(s)0 ∧ ∧ Sy1.999 MRK 273 13 42 51.6 +56 08 13 11326 154.6 0.72 64 –20.8 12.1 47.0 U · · ·
Ring galaxy;Sy2 LINER100 NGC 5347 13 53 17.8 +33 29 27 2335 31.9 1.7 45 –19.6 9.6 35.9 c c S 6.4 S?102 Arp 302S p
14 57 00.3 +24 36 25 10029 136.4 0.6 · · · · · · < b · · · L · · · SC HIIArp 302 center 14 57 00.5 +24 36 44 10103 137.4 · · · · · · · · · · · · · · · · · · · · ·
DBL SYS103 Arp 302N p
14 57 00.7 +24 37 03 10094 137.3 0.9 · · · · · · b b L · · · (Sb);HII LINER104 NGC 5866 15 06 30.2 +55 45 46 672 9.2 4.7 86 –19.9 9.2 29.9 –1.2 S0 3 HII/LINER105 NGC 5907 15 15 52.9 +56 19 33 667 9.1 12.77 87 –20.9 9.5 27.8 N 5.4 SA(s)c: sp HII:106 Mrk 848 15 18 05.9 +42 44 53 12049 163.7 0.9 90 –20.4 11.8 44.7 L –1.7 S0? pec HII107 NGC 5953 15 34 32.3 +15 11 42 1965 26.9 1.6 44 –19.6 10.3 37.4 S 0.2 SAa: pec;LINER;Sy2108 Arp 220 15 34 57.2 +23 30 12 5434 74.2 1.5 57 –21.0 12.1 44.7 U 8.4 S?;LINER;HII Sy2109 NGC 6240 16 52 58.8 +02 24 04 7339 100.0 2.1 82 –21.5 11.7 43.9 L –0.2 I0: pec;LINER Sy2110 17208-0014 17 23 22.3 –00 17 02 12834 174.3 0.4 50 –20.2 12.4 44.5 U 3.8 Sbrst HII111 Arp 293 p
16 58 30.6 +58 56 19 5600 76.4 · · · · · · · · · · · ·
GPair112 NGC 6524 17 59 14.9 +45 53 17 5698 77.8 1.3 69 –20.9 10.8 34.4 c S –2.8 S0:113 NGC 6670B p
18 33 34.1 +59 53 21 8428 114.8 · · · · · · –20.1 11.5 38.4 L –1.1 HIINGC 6670 p
18 33 35.1 +59 53 21 8650 117.8 1 78 · · · p
18 33 37.7 +59 53 22 8719 118.7 · · · · · · · · · a c c S –3.9 BCD/E HII119 NGC 7217 22 07 52.2 +31 21 35 952 13.0 3.9 36 –20.4 9.6 29.3 2.5 (R)SA(r)ab;Sy LINER120 NGC 7331 22 37 03.5 +34 24 43 816 15.1 a · · · · · · · · · –21.5 · · · · · · · · · · · ·
121 NGC 7465 23 02 00.8 +15 57 56 1968 26.9 1.2 64 –19.2 10.0 39.3 S –1.9 (R’)SB(s)0 ∧ ∧ : Sy2122 NGC 7469 23 03 15.6 +08 52 26 4892 66.8 1.5 30 –21.7 11.5 41.8 L 1.1 (R’)SAB(rs)a Sy1.2123 NGC 7479 23 04 56.7 +12 19 23 2381 32.6 4.1 36 –21.6 10.6 36.6 S 4.3 SB(s)c;LINER Sy2124 NGC 7541 23 14 43.7 +04 32 02 2689 31.3 a The columns contain the following information: Col.(1): The sequence number of the specific source. Col.(2): Galaxy name; p : galaxy pair; v: Virgo cluster galaxy.Cols.(3) and (4): Right ascension (R.A.) and declination (DEC) in J2000.0 coordinates. Col.(5): Heliocentric velocity ( v hel) from NED. Col.(6): Galaxy proper distancecalculated from v hel using H − − a : recently measured distances drawn mostly from acrosslink in NED (c.f. NED 1D for references). The distance to Virgo cluster galaxies is set to 16.5 Mpc (Mei et al. 2007). Col.(7): Optical diameter ( D
25) from NED.Col.(8): Galaxy inclination angle ( i ) from HyperLEDA. Col.(9): B-band absolute magnitude (MB) from HyperLEDA. Col.(10): FIR luminosity ( L FIR = L (40–400 µ m))calculated following the prescription of Moshir et al. (1992) (see Sec.2.2.1). b : calculated with infrared fluxes taken from Surace et al. (2004) where the HIRES processingis adopted allowing for a deconvolution of close galaxy pairs. c : sources not included in the RBGS (Sanders et al. 2003); Col.(11): Dust temperature ( T dust) derived fromthe IRAS 60 µ m/100 µ m color assuming an emissivity that is proportional to the frequency ν . Col.(12): Galaxy classification from this paper; N: normal; S: starburst; L:LIRG; U: ULIRG. All the rest are “pure” AGN (see Sect. 2.2.2). Col.(13): Galaxy type code from HyperLEDA. Col.(14): Galaxy classification from NED. TABLE 2Galaxy classification of the sample
LINER Seyfert Starburst LIRG ULIRGLINER 45 16 20 5 2Seyfert 16 45 27 8 3Starburst 20 27 77 24 4LIRG 5 8 24 24 –ULIRG 2 3 4 – 4
Note . — The diagonal gives the total number of sourcesof a specific class (e.g., there are 45 LINERs). Nondiagonalcoefficients show the number of targets belonging to at leasttwo specific classes (e.g., there are 20 galaxies which have beenclassified both as LINERS and as starburst galaxies). For de-tails of the classification, see § n extragalactic CO(3–2) survey 15 TABLE 3
Observed Quantities of the HHT extragalactic CO J = 3–2 surveyNo. SOURCE I v ∆ v T mb log L CO32 log L CO10 R Ref.(K km s − ) (km s − ) (km s − ) (mK) (K km s − pc ) (K km s − pc )(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)1 IC 10 7.1 ± ± ± ±
30 4.6 ± ± ± a, † ± ∗ ±
10 7.6 ± ± ± a ± ± ± ±
22 5.8 ± a ± ± ± ±
21 8.1 ± ± ± a ± ∗ ± ± < < c < · · · ( 100 ) ( 26 ) < < a ± ± ± ±
40 7.3 ± ± ± a, † ± ± ± ±
13 8.3 ± < < b ± ± ± ±
49 7.2 ± ± ± a
10 NGC 1055 19.6 ± ± ±
10 105 ±
18 7.7 ± ± ± a
11 NGC 1068 116.0 ± ± ± ±
24 8.6 ± ± ± a
12 NGC 1084 14.9 ± ± ± ±
14 7.9 ± ± ± a
13 NGC 1087 7.7 ± ± ± ±
11 7.6 ± ± ± a
14 NGC 1275 9.0 ± ±
12 248 ±
23 34 ±
14 8.8 ± a
15 NGC 1530 30.0 ± ± ± ±
34 8.6 ± a
16 NGC 1569 2.9 ± ± ±
10 64 ±
20 5.2 ± ± ± a
17 NGC 1614 71.1 ± ± ±
19 303 ±
73 9.6 ± ± ± a
18 NGC 1637 4.9 ± ± ± ± ± ± ± a
19 NGC 2146 66.9 ± ± ± ±
36 8.1 ± ± ± a, †
20 NGC 2559 37.3 ± ± ± ±
33 8.3 ± < < b
21 NGC 2655 < · · · ( 400 ) ( 17 ) < < < b
22 NGC 2681 17.1 ± ± ±
14 117 ±
40 7.3 ± ± ± a
23 NGC 2768 < · · · ( 200 ) ( 14 ) < < a
24 Arp 55 11.1 ± ∗ ± ± < < b
25 NGC 2782 31.2 ± ∗ ±
25 8.7 ± < < b
26 NGC 2841 < · · · ( 200 ) ( 39 ) < < a
27 NGC 2903 59.6 ± ∗
525 143 390 ±
69 7.6 ± a
28 Arp 303 S 18.6 ± ∗ ±
12 9.2 ± · · · · · · · · ·
29 Arp 303 N 37.8 ± ∗ ±
18 9.5 ± · · · · · · · · ·
30 NGC 2985 8.0 ± ±
15 183 ±
38 41 ±
17 7.5 ± ± ± a
31 NGC 3032 4.1 ± ±
11 130 ±
31 30 ±
12 7.4 ± a
32 NGC 3034 1056.0 ± ± ± ±
166 8.0 ± a
33 NGC 3079 93.4 ± ∗ ±
57 8.4 ± ± ± a
34 NGC 3077 8.0 ± ± ± ±
12 6.2 ± ± ± a
35 NGC 3110 24.5 ± ± ±
11 71 ± ± b,d
36 NGC 3166 < · · · ( 300 ) ( 141 ) < a
37 NGC 3169 11.0 ± ∗ ±
15 7.6 ± < < b
38 NGC 3147 19.8 ± ±
12 466 ±
28 40 ± ± a, †
39 NGC 3227 18.1 ± ± ±
16 70 ±
12 7.8 ± ± ± a
40 HARO 2 2.5 ± ± ± ± ± ± ± a
41 NGC 3310 6.9 ± ± ±
13 48 ±
15 7.2 ± ± ± a
42 NGC 3351 26.0 ± ± ±
11 122 ±
27 7.6 ± ± ± a
43 NGC 3367 7.5 ± ± ±
15 82 ±
22 8.2 ± > > d
44 NGC 3368 24.6 ± ± ±
10 119 ±
19 7.6 ± ± ± a
45 NGC 3521 14.5 ± ± ±
19 76 ±
20 7.4 ± < > b ,23 d
46 NGC 3556 14.4 ± ± ±
17 143 ±
31 7.2 ± b ,31 d
47 NGC 3593 17.3 ± ∗
643 128 99 ±
17 7.2 ± a
48 NGC 3627 33.4 ± ± ± ±
25 7.8 ± a NGC 3627A 21.9 ± ± ± ±
19 7.7 ± a
49 NGC 3628 140.7 ± ± ± ±
25 8.4 ± ± ± a
50 NGC 3642 7.4 ± ±
19 235 ±
39 29 ±
12 7.6 ± · · · · · · · · ·
51 NGC 3682 12.2 ± ±
16 240 ±
35 48 ±
16 7.8 ± < < e
52 NGC 3690A 36.4 ± ± ±
12 273 ±
46 8.9 ± a
53 NGC 3690B 48.7 ± ± ±
10 191 ±
33 9.1 ± a
54 NGC 3810 20.5 ± ± ±
13 133 ±
24 7.7 ± ± ± a
55 NGC 3982 6.7 ± ± ± ±
16 7.6 ± > > d
56 IC 750 37.4 ± ± ±
14 172 ±
41 7.6 ± ± ± a
57 NGC 4038 44.0 ± ± ± ±
19 8.4 ± ± ± a NGC 4038/9 82.6 ± ∗ ±
25 8.7 ± ± ± a
58 NGC 4039 15.2 ± ∗ ±
30 8.0 ± ± ± a
59 NGC 4102 49.0 ± ∗
823 220 172 ±
39 7.9 ± a
60 NGC 4138 < · · · ( 400 ) ( 34 ) < < < b
61 NGC 4192 7.0 ± ± ±
19 93 ±
14 7.4 ± b ,26 c
62 NGC 4254 24.0 ± ∗ ±
20 7.9 ± b ,23 c
63 NGC 4258 64.4 ± ± ± ±
17 7.8 ± a
64 NGC 4293 20.4 ± ± ± ±
11 7.9 ± a
65 NGC 4303 33.7 ± ± ±
13 250 ±
52 8.1 ± b ,23 c
66 NGC 4314 37.6 ± ∗ ±
53 7.9 ± ± ± a
67 NGC 4321 56.9 ± ± ±
14 302 ±
84 8.3 ± ± ± a
68 NGC 4369 8.2 ± ± ± ±
18 7.3 ± a
69 NGC 4395 < · · · ( 400 ) ( 18 ) < · · · · · · · · ·
70 NGC 4414 31.2 ± ± ± ± ± ± ± a
71 NGC 4438 < · · · ( 300 ) ( 84 ) < < a
72 NGC 4457 20.3 ± ± ± ±
14 7.9 ± < < b
73 NGC 4490 24.4 ± ±
32 411 ±
68 56 ±
28 7.5 ± < < b
74 NGC 4527 37.4 ± ∗ ±
48 8.1 ± < < b ,31 d TABLE 3 (continued)No. SOURCE I v ∆ v T mb log L CO32 log L CO10 R Ref.(K km s − ) (km s − ) (km s − ) (mK) (K km s − pc ) (K km s − pc )(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)75 NGC 4565 8.3 ± ± ±
21 86 ±
28 7.5 ± ± ± a
76 NGC 4594 6.8 ± ±
17 292 ±
41 22 ± ± < b
77 NGC 4631 17.7 ± ± ± ±
22 7.2 ± a
78 NGC 4639 < · · · ( 400 ) ( 29 ) < < · · · b
79 NGC 4654 18.9 ± ± ± ±
40 7.8 ± ± ± a
80 NGC 4666 36.7 ± ±
13 242 ±
27 142 ±
45 8.3 ± b ,31 d
81 NGC 4691 12.9 ± ± ± ±
50 7.6 ± b ,31 d
82 NGC 4710 18.7 ± ∗ ±
27 7.8 ± b ,31 d
83 NGC 4736 21.3 ± ∗
279 122 133 ±
19 6.8 ± ± ± a, †
84 MRK 231 8.1 ± ±
10 215 ±
24 35 ± ± ± ± a
85 NGC 4818 42.4 ± ± ± ±
26 8.1 ± a
86 NGC 4826 86.8 ± ∗
430 214 340 ±
61 7.8 ± b ,23 d
87 NGC 4941 6.7 ± ± ± ±
19 7.3 ± > > d
88 NGC 5033 16.7 ± ± ±
16 66 ±
13 7.5 ± ± ± a
89 NGC 5055 25.8 ± ∗
595 190 99 ±
46 7.5 ± a
90 UGC 8335 3.6 ± ∗ ± ± < < c
91 UGC 8335B 8.0 ± ∗ ±
14 9.2 ± < < c
92 Arp 193 20.5 ± ∗ ±
14 9.4 ± ± ± a
93 NGC 5194 44.4 ± ± ± ±
87 7.5 ± ± ± a, †
94 M 83 153.9 ± ± ± ±
76 7.6 ± ± a
95 NGC 5256 5.7 ± ∗ ± ± < < c
96 NGC 5257 18.2 ± ±
13 312 ±
29 55 ±
16 9.3 ± > > d
97 NGC 5258 27.0 ± ± ±
16 73 ±
12 9.5 ± > > d
98 NGC 5273 < · · · ( 100 ) ( 11 ) < > · · · d
99 MRK 273 20.1 ± ∗ ± ± ± ± a
100 NGC 5347 12.0 ± ± ±
10 144 ±
28 8.2 ± c ,11 d
101 NGC 5666 9.2 ± ±
29 276 ±
51 31 ±
17 8.0 ± ± ± a
102 Arp 302S 8.5 ± ∗ ±
11 9.3 ± < < c Arp 302 center < · · · ( 400 ) ( 22 ) < · · · · · · · · ·
103 Arp 302N 18.1 ± ∗ ± ± < < c
104 NGC 5866 < · · · ( 200 ) ( 78 ) < < a
105 NGC 5907 8.4 ± ± ± ±
19 7.0 ± ± ± a
106 Mrk 848 4.2 ± ± ±
11 44 ± ± a
107 NGC 5953 21.5 ± ∗ ±
40 8.3 ± d
108 Arp 220 58.5 ± ∗ ± ± ± ± a
109 NGC 6240 74.9 ± ∗ ±
21 10.0 ± ± ± a
110 17208-0014 24.6 ± ±
11 386 ±
24 60 ± ± ± ± a
111 Arp 293 13.3 ± ∗ ±
14 9.0 ± < < b
112 NGC 6524 6.6 ± ±
10 125 ±
24 49 ±
14 8.7 ± · · · · · · · · ·
113 NGC 6670B 19.8 ± ∗ ±
20 9.5 ± < < c NGC 6670 8.3 ± ∗ ±
19 9.2 ± < < c
114 NGC 6670A 16.2 ± ∗ ± ± < < a c
115 NGC 6814 2.1 ± ± ±
13 17 ± ± b ,11 d
116 NGC 6946 132.3 ± ± ± ±
56 7.7 ± ± ± a
117 NGC 7013 10.5 ± ±
13 352 ±
30 28 ± ± a
118 NGC 7077 0.8 ± ±
16 222 ±
31 19 ± ± ± a
119 NGC 7217 8.1 ± ±
12 269 ±
25 28 ± ± ± ± a
120 NGC 7331 7.2 ± ∗
884 50 74 ±
15 7.3 ± ± ± a NGC 7331A 12.6 ± ∗
903 145 86 ±
25 7.6 ± ± ± a
121 NGC 7465 6.3 ± ±
10 127 ±
17 47 ±
16 7.8 ± < < b
122 NGC 7469 35.2 ± ± ±
10 133 ±
22 9.3 ± b ,11 d
123 NGC 7479 20.1 ± ± ±
19 79 ±
15 8.4 ± < < b
124 NGC 7541 17.8 ± ± ± ±
20 8.3 ± a, †
125 NGC 7679 17.2 ± ∗ ±
29 9.0 ± < e The columns contain the following information: Col.(1): Sequence number. Col.(2): Galaxy name. Cols.(3)–(6): The CO(3–2) line intensity ( I
32 = R T mb d v ), LSRvelocity, width (FWHM), and main beam brightness temperature, respectively, with corresponding standard errors from Gaussian fits or from ∗ moments in the case ofnon-Gaussian line shapes. Non-detections are listed with their rms noise level σ (Col.(6) values in brackets) at a channel spacing of δv ∼ −
1. In these cases the fullwidth of the CO(1–0) line (∆ V
10, Col.(5) values in brackets) was estimated from spectra taken from the literature or was set to 400 km s −
1; corresponding upper limitsfor I
32 were obtained using I < σ × (∆ V × δ v )1 / § ′′ . Col.(8):The CO(1–0) luminosity calculated with an equation similar to that of L CO(3 −
2) for galaxies with IRAM-30m CO(1–0) data available in the literature (listed in Col(10));otherwise either upper or lower limits are given, depending on the telescope used for CO(1–0). Col.(9): The CO(3–2)/CO(1–0) line intensity ratio R
31 and its standarddeviation for galaxies with IRAM-30m CO(1–0) data available in the literature (listed in Col(10)); otherwise either upper or lower limits are given, depending on thetelescope used for the CO(1–0) data. Col.(10): References for the CO(1–0) data: 1. Young et al. (1995); 2. Albrecht et al. (2004); 3.Braine et al. (1993); 4. Combes et al.(1991); 5. Gao & Solomon (2004b); 6. Gerin et al. (2000); 7. Heckman et al. (1989); 8. Mauersberger et al. (1999); 9. Radford et al. (1991); 10. Sanders et al. (1991);11. Vila-Vilar´o et al. (1998); 12. Wiklind & Henkel (1989); 13. Yao et al. (2003); 14. Kenney et al. (1988); 15. Solomon et al. (1997); 16. Reuter et al. (1996); 17.Schulz et al. (2007); 18. Casoli et al. (1989); 19. Chini et al. (1992); 20. Wiklind et al. (1995); 21. Sage et al. (1992); 22. Reuter et al. (1993); 23. Nishiyama et al.(2001); 24. Golla & Wielebinski (1993); 25. Handa et al. (1990); 26. Kuno et al. (2007); 27. Combes et al. (2007); 28. Baan et al. (2008); 29. Solomon et al. (1992); 30.Greve et al. (1996); 31. Komugi et al. (2008). Telescopes used are a IRAM-30m; b FCRAO-14m; c NRAO–12m; d NRO-45m; e OSO-20m. † : The CO(1–0) data was takenat an offset position > ′′ relative to our CO(3–2) position. n extragalactic CO(3–2) survey 17 TABLE 4 CO(3–2)/(1–0) integrated line intensity ratios. galaxy R statisticstype N mean d σ e median min max(activity)normal 7 0.61 ± a
20 0.65 ± a
12 0.82 ± b
18 0.78 ± c
25 0.89 ± ± ± ± ± ± a Including starbursts but not (U)LIRG overlaps. b See § . 2.2.2 for details. c Excluding (U)LIRGs. d The mean value and its standard error. e Standard deviation of an individual target. N G C N G C NGC 7469NGC 1055 N G C N G C N G C N G C N G C N G C N G C N G C N G C N G C N G C N G C N G C N G C N G C N G C N G C N G C N G C M N G C Arp 303 NNGC 6240Arp 55Mrk 848 N G C / / N G C N G C A r p S N G C N G C M a ff e i IC 10 NGC 4321NGC 2782NGC 1275 H A R O N G C N G C N G C NGC 3521NGC 891NGC 7331NGC 1530 N G C N G C NGC 3147 N G C NGC 1068 N G C N G C NGC 4395 N G C N G C N G C N G C N G C N G C N G C N G C Arp 220Mrk 231Mrk 273 N G C N G C N G C N G C N G C N G C l og L F I R / D ( L k p c - ) log L FIR (L ) N G C N G C N G C NGC 4414NGC 2559 N G C N G C N G C NGC 6524 N G C III ZW 35 M N G C NGC 404 NGC 3077 N G C Fig. 1.—
Starburst definition: Red open stars denote galaxies classified as starbursts in the literature. Blue triangles mark galaxies notbeing classified as such. All our CO J = 3–2 non-detections arise from the latter sample and are marked by open triangles. Also markedare NGC 253 (by a green star), a typical starburst galaxy (e.g., Brunthaler et al. 2009), and IC 342 (by a green triangle), a galaxy similar tothe Milky Way (Downes et al. 1992). The dashed horizontal line, with log( L FIR /D ) = 7.25 L ⊙ kpc − , is used as the borderline betweenstarburst and non-starburst galaxies throughout this paper. -4 -2 0 2 4 6 8 1005101520 a) N u m be r Hubble Type (U)LIRG starburst ’pure’ AGN normal 7 8 9 10 11 12 13051015202530 b) N u m be r log L FIR (L )
24 28 32 36 40 44 48051015202530 c)f) N u m be r T dust (K) d) N u m be r log d p (Mpc) e) nu m be r D (arcmin) f) N u m be r log D (kpc) g) N u m be r Inclination (degree) -24 -23 -22 -21 -20 -19 -18 -17 -16 -15051015202530354045 h) N u m be r M B (mag) Fig. 2.—
Number distributions of the observed galaxy sample: a) Hubble type (see de Vaucouleurs et al. 1991), b) far-infrared luminosity,c) 60 µ m/100 µ m dust color temperature, d) distance, e) optical angular size, f) linear size, g) inclination, and h) B-band magnitude of theobserved sample. n extragalactic CO(3–2) survey 19 Fig. 3.— CO J = 3–2 spectra. The velocity scale corresponds to Local Standard of Rest in units of km s − . The intensity is displayedin units of main beam brightness temperature (K). Fig. 3.— ( Continued ) n extragalactic CO(3–2) survey 21 Fig. 3.— ( Continued ) N u m be r R (U)LIRG starburst ’pure’ AGN normal Fig. 4.—
Number distribution of the matching beam line intensity ratio R for 61 galaxies from our sample (see § f) R L FIR /D (L kpc -2 ) c) R cos i
24 28 32 36 40 44 480.00.40.81.21.62.02.4 e) R T dust (K) R log L FIR (L ) d) b) R Distance (Mpc) a) normal starburst Seyfert LINER (U)LIRG R Hubble type
Fig. 5.— R versus a) Hubble type, b) distance (the lower X-axis) or projected beam size (the upper X-axis in kpc) c) cosine of theinclination, d) FIR luminosity, e) dust temperature, and f) FIR luminosity per unit area ( L FIR / D ) of sample galaxies with galaxy typesbeing indicated in a). The error bars are removed for clarity in the panels b) – f). n extragalactic CO(3–2) survey 23 s r0.87(0.05), 0.90 normal starburst LINER/Seyfert (U)LIRG not from IRAM log L CO(1-0) (K km s -1 pc ) l og L F I R ( L ) l og L F I R ( L ) s r0.87(0.03), 0.92 normal starburst LINER/Seyfert (U)LIRG non-detection log L CO(3-2) (K km s -1 pc ) Fig. 6.—
Log-log plot of the correlation between nuclear CO and global FIR luminosity for CO(3–2) (upper panel) and CO(1–0) (lowerpanel). Both CO line luminosities cover the central 22 ′′ region, except for the crosses representing CO(1–0) data with the larger ( ∼ ′′ )beams of the FCRAO-14m or NRAO-12m telescopes. Straight lines show linear regression fits to the unweighted data. Slopes (s) andcorrelation coefficients (r) are given at the lower right corner of each panel. Only our CO(3–2) detections and CO(1–0) data from theIRAM-30 m are included in the fits. Non-detections (upper limits) and data from the FCRAO and NRAO were not considered. IC 10, theisolated dot at the lower left corner of each panel, is also not part of the fits (see § log L CO(1-0) (K km s -1 pc ) c) FCRAO-14m/NRAO-12m 0.88(0.09), 0.92 0.70(0.07), 0.84 s r 1.04(0.09), 0.93 log L CO(1-0) (K km s -1 pc ) s r 1.06(0.07), 0.98 log L CO(3-2) (K km s -1 pc ) D
2’ s r 1.02(0.06), 0.95 l og L F I R ( L ) a) 2’< D
4’ 0.82(0.08), 0.89 D > 4’ 0.62(0.09), 0.69 Fig. 7.—
Log-log correlation of a) L CO(3 − and L FIR (this work), b) L CO(1 − and L FIR (CO from the IRAM-30m telescope), andc) L CO(1 − and L FIR (CO from the FCRAO-14m and NRAO-12m antennas), with our sample galaxies being divided into three groupscharacterized by their optical angular sizes ( D ): 1) D ≤ ′ (filled circles, solid lines denoting the corresponding linear regression fit),2) 2 ′ < D ≤ ′ (empty triangles, dashed lines), and 3) 4 ′ < D ≤ ′ (crosses, dash-dotted lines). The corresponding slopes ( s ) andcorrelation coefficients ( r ) are also given. n extragalactic CO(3–2) survey 25 IC 10NGC 404NGC 891Maffei 2NGC 1068NGC 1084NGC 1087NGC 2146NGC 2903M 82NGC 3079NGC 3227NGC 3351NGC 3368NGC 3593NGC 3627NGC 3628NGC 4102NGC 4414NGC 4631NGC 4691NGC 4736NGC 4818NGC 5055Mrk 231M 51M 83NGC 5907Arp 22017208-0014NGC 6946NGC 7331NGC 7479NGC 7541 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0
HHT-M99 HHT-D01 HHT-V02 HHT-N05 CSO-B06 ASTE-K07 log(I’/I ) Fig. 8.—
A comparison of our integrated CO(3–2) intensities with previously published results. The horizontal axis gives the logarithmicintensity deviation, where I ′ denotes the integrated CO(3–2) intensity taken from the literature, while I is from this work. Data pointsmeasured at nominal position offsets & ′′ are represented as open squares. A pair of vertical lines marks the ± § A.1.
10 10010100 I C O ( - ) _ J C M T ( K k m s - ) I CO(3-2)_this paper (K km s -1 ) JCMT-Y03 JCMT-W08Arp 302 NGC 5258Mrk 273Mrk 848 NGC 3690Arp 193Mrk 231 NGC 1614M 83Arp 220NGC 624017208-0014 NGC 3110NGC 5953 NGC 5257NGC 3367NGC 5256Arp 55