A search for High Mass Stars Forming in Isolation using CORNISH & ATLASGAL
Chenoa D. Tremblay, Andrew J. Walsh, Steven N. Longmore, James S. Urquhart, Carsten König
PPublications of the Astronomical Society of Australia (PASA)c (cid:13)
Astronomical Society of Australia 2018; published by Cambridge University Press.doi: 10.1017/pas.2018.xxx.
A search for High Mass Stars Forming in Isolation usingCORNISH & ATLASGAL
Chenoa D. Tremblay ∗ , Andrew J. Walsh , Steven N. Longmore , James S. Urquhart , , and Carsten K¨onig International Centre for Radio Astronomy Research, Curtin University, GPO Box U1987, Perth WA 6845, Australia Astrophysics Research Institute, Liverpool John Moores University, Twelve Quays House, Egerton Wharf, Birkenhead CH411LD, UK Max-Planck-Institut f¨ur Radioastronomie, Auf dem H¨ugel 69, Bonn, Germany Centre for Astrophysics and Planetary Science, University of Kent, Canterbury, CT2 7NH
Abstract
Theoretical models of high mass star formation lie between two extreme scenarios. At one extreme, all themass comes from an initially gravitationally-bound core. At the other extreme, the majority of the masscomes from cluster scale gas, which lies far outside the initial core boundary. One way to unambiguously showhigh mass stars can assemble their gas through the former route would be to find a high mass star formingin isolation. Making use of recently available CORNISH and ATLASGAL Galactic plane survey data, wedevelop sample selection criteria to try and find such an object. From an initial list of approximately 200sources, we identify the high mass star forming region G13.384+0.064 as the most promising candidate. Theregion contains a strong radio continuum source, that is powered by an early B-type star. The bolometricluminosity, derived from infrared measurements, is consistent with this. However, sub-millimetre continuumemission, measured in ATLASGAL, as well as dense gas tracers, such as HCO + (3-2) and N H + (3-2) indicatethat there is less than ∼
100 M (cid:12) of material surrounding this star. We conclude that this region is indeeda promising candidate for a high mass star forming in isolation, but that deeper near-IR observations arerequired to put a stronger constraint on the upper mass limit of young, lower mass stars in the region.Finally, we discuss the challenges facing future studies in proving a given high mass star is forming inisolation.
Keywords: isolated – stars: formation – ISM:
High mass stars – O or early B type star of sufficientmass to produce a Type II supernova (Zinnecker et al.2007) or > (cid:12) (Miettinen 2012) – dominate the energycycles and chemical enrichment of galaxies. However,understanding the formation of high mass stars remainsa challenge, and several different theoretical formationscenarios have been proposed (Zinnecker et al. 2007;Tan et al. 2014). The general observational phases (i.e.formation of cold dense cloud, gravitational collapse ofa hot core, accretion, and formation of ultracompactH II regions) are typically agreed upon, but the dom-inant physical processes and their relevant time scalesare still under debate. The problems associated withour understanding are that it is difficult to observe theearly stages of formation due to high dust extinction,the theoretical problem is complex and high mass stars ∗ E-mail: [email protected] are seldom, if ever, formed in isolation (Zinnecker et al.2007).Over the last few decades, high mass star formationtheories have been discussed in the context of two ex-treme scenarios: that of the turbulent core (McKee &Tan 2003) and the competitive accretion (Bonnell etal. 2001; Bonnell et al. 2006) scenarios. In the former,all the mass comes from an initially gravitationally-bound core. In the latter, the majority of the masscomes from cluster scale gas that is far outside the ini-tial core boundary. Early debate suggested that stellarcollisions (Bonnell et al. 1998) could be a potential cre-ation mechanism, but this has largely been discounteddue to the extremely high stellar densities required .More recent theoretical and simulation work, addingmore physics (e.g. radiation pressure, ionisation) and Although see Izumi et al. ( 2014) who suggest a star system inthe extreme outer galaxy ( >
18 kpc) may have been formed bylarge scale collision. a r X i v : . [ a s t r o - ph . GA ] N ov Tremblay et al. overcoming previous limitations in numerical methods,have found that increasing the feedback, initial densityfluctuations and turbulence leads to an increase in thefraction of the final stellar mass which comes directlyfrom an initial gravitationally-bound core (Krumholz etal. 2007; Krumholz et al. 2012; Hennebelle et al. 2011;Commercon et al. 2011; Myers et al. 2012; Bonnell et al.2004; Peters et al. 2010a). The apparent dichotomy be-tween the competitive and core accretion models seemsto be less extreme, or at least the reasons for the di-chotomy are now better understood.Nevertheless, understanding whether the two extremeformation scenarios are viable routes for high mass starsto assemble their mass has important consequences forthe host galaxy. If high mass stars can only form in thepresence of an attendant cluster, the stellar initial massfunction (IMF) will be sampled very differently thanif high mass stars can (albeit rarely) form in isolation(Bastian et al. 2010). When averaged on galactic scales,this can make a dramatic difference in the number of(very) high mass stars, and hence the level of feedback,chemical enrichment etc.Given the wider importance of how the IMF is sam-pled, many observational studies have tried to find evi-dence of high mass stars forming in isolation (Bressertet al. 2012; Tout et al. 1997; Weidner et al. 2009; Wei-dner & Kroupa 2005; Parker et al. 2007; de Wit et al.2005; Schilbach & R¨oser 2008). These previous searcheshave primarily focused on optical and infrared data tofind young high mass stars with no lower mass youngstars around them. However, as these high mass starsmust already have cleared their natal gas cloud in orderto be optically visible, it is very difficult to determineif they formed at their present location, as opposed tohaving been ejected from their parent stellar nursery oflower mass stars.In this paper we try a different approach, aiming tofind very young high mass stars while they are still em-bedded in their natal gas cloud. While extinction makesit impossible to find these objects in the optical andnear-IR, their prodigious luminosity and Lyman con-tinuum flux means they should be conspicuous at far-IR wavelengths and have bright cm-continuum, free-freeemission. Even in the most optimistic scenarios, highmass stars forming in isolation are expected to be veryrare, requiring large-area surveys to identify candidates.Thanks to an enormous effort from the Galactic obser-vational community, Galactic plane surveys now existacross much of the electromagnetic spectrum at suffi-cient sensitivity and resolution to identify the major-ity of young high-mass star formation regions in theGalaxy. With the data now in hand, we aim to use asimple selection criterion to pick out the best candidatesfor young, high mass stars forming in isolation.
For this first attempt to try and find examples of highmass stars forming in isolation, we used a series ofdata summarised by Urquhart et al. (2013) for ATLAS-GAL (Schuller et al. 2009) and CORNISH (Purcell etal. 2013). Urquhart et al. (2013) noted that targetedsurveys of compact and ultra-compact (UC) H II re-gions identified by infrared colours can be contaminatedwith intermediate mass young stellar objects (YSOs)and planetary nebulae (PNe). However, by incorporat-ing radio astronomy data mixed into the identificationprocess, this is no longer a concern. Intermediate massstars do not show radio continuum emission, so wouldbe eliminated in the cross match. Planetary nebulae arenot usually associated with dust that is bright enoughto be detected by ATLASGAL. We use their selection criteria to select approximately200 bona fide YSOs, associated with compact and UCH II regions. Figure 1 compares the clump gas massand Lyman continuum flux of these regions. The sourceG13.384+0.064 stands out in Figure 1 as having a verylow gas mass for its Lyman continuum flux. This Figureis closely matched to the upper panel from Figure 26 inUrquhart et al. (2013). The aim of this paper is to betterconstrain the luminosity, mass and lyman photon fluxbased on a thorough literature search of the region andusing GLIMPSE (Churchwell et al. 2009; Benjamin etal. 2003) and MIPSGAL (Carey et al. 2009) data to totest for evidence of a surrounding cluster. We selectedthis as the most promising candidate in our initial sam-ple for a high mass star forming in isolation. Below wedescribe our efforts to use data in the literature to de-termine whether or not we can confirm or rule out thisstatus. The GLIMPSE (Churchwell et al. 2009; Benjamin etal. 2003) and MIPSGAL (Carey et al. 2009) images ofthe star forming region G13.384+0.064 show a brightcore surrounded by gas with an empty area or bubblebetween the two (Figure 2). The bubbles are similarto those seen in other high mass star forming regionsand they are likely to be created by stellar feedback(e.g. Weaver et al. 1977). The CORNISH radio contin-uum image has sufficient resolution to show an extendedshell like source in the centre of the region (Figure 2).The region inside the radio emission also contains lowlevels of diffuse emission in the UKIRT Infrared Deep There were some instances noted where PNe were identified inATLASGAL by their mid-infrared emission. However, they wereremoved from the source list and monte-carlo simulations agreedwith their identified numbers.PASA (2018)doi:10.1017/pas.2018.xxx n Isolated Star? Figure 2.
GLIMPSE 3 colour image with blue as 3.6 µm , green as 4.5 µm and and red as 8.0 µm . Contour image on the left showsCORNISH 5 GHz radio continuum emission from 2 mJy/beam to 10 mJy/beam in steps of 2 mJy/beam. The white cross on the imageon the left represents the centre of the ATLASGAL contours and the image on the right shows the GLIMPSE 3 colour image withATLASGAL contours. The contour levels are from 0.25 Jy/beam and increase in steps of 0.25 Jy/beam up to 1.0 Jy/beam. We identifyan evolved star, seen at G13.380+0.050 and discussed in section 3.8 that is unrelated to the source. Figure 1.
Plot of clump mass versus Lyman photon flux for aseries of ATLASGAL and CORNISH sources, similar to the up-per panel of Figure 26 in Urquhart et al. (2013). When the clumpmass is plotted against the bolometric luminosity, the source atG13.384+0.064 does not stand out as different compared to otherstar forming regions. This discrepancy is investigated in this pa-per. The error bars represent the best available data reported inthis paper.
Sky Survey (UKIDSS) K band images, as shown in Fig-ure 4. It is likely this emission corresponds to BrackettGamma (B γ ) emission associated with the H II region,as the morphology of the diffuse K band emission resem-bles the radio continuum contours (Beck et al. 2010).We used data from UKIDSS (Lawrence et al. 2007),2MASS (Skrutskie et al. 2006) and GLIMPSE (Church-well et al. 2009; Benjamin et al. 2003) to investigatethe near and mid-IR source populations surroundingG13.384+0.064 and search for an embedded young stel-lar population. First, we downloaded all the sourcesin the UKIDSS point source catalogue within a 10 arcminute radius of G13.384+0.064 and looked for an in-crease in the surface density of infrared objects to-wards the source. There is no statistical difference inthe surface density of infrared sources at the locationof G13.384+0.064 compared to similar regions close to,but offset from the ATLASGAL contours. We concludethere is no evidence of a cluster based on source density.The 3-colour diagram of UKIDSS data, Figure 4,shows some red stars within the contours from COR-NISH. To determine if these are field stars behind thedust cloud or if they are within the cloud, we use theUKIDSS, 2MASS (for saturated UKIDSS sources) andGLIMPSE photometry data to plot a colour-colour di-agram (CCD) and colour-magnitude diagrams (CMDs)of the infrared sources on and off G13.384+0.064. Fig-ure 3 shows a representative colour-colour diagram us-ing H and K-band data from UKIDDS (Lawrence et al.2007) and 2MASS (Skrutskie et al. 2006) and L (3.6 µ m) PASA (2018)doi:10.1017/pas.2018.xxx
Tremblay et al. -0.2 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2H-K / mag00.51.01.52.0 K - . / m ag Off SourceOn Source<10^5 Years
Figure 3.
Colour-Colour diagram plotting sources 30 arc sec-ond around G13.384+0.064 (on source;blue) and 30 arc secondsaround G13.36+0.075 (off source;red). The IMF curve is datarepresenting a cluster < years old from Lejeune & Schaerer(2001). The vector represents the extinction vector plotted as perNishiyama et al. (2009). There are no clear signs there are a clus-ter of main sequence stars. band data from GLIMPSE (Churchwell et al. 2009; Ben-jamin et al. 2003) for a 30 arc second radius aroundthe source at G13.384+0.064 (matching with the con-tours from ATLASGAL shown in Figure 2) and off thesource at G13.36+0.075, as shown in Figure 3. The red-dening vector was determined from Nishiyama et al.(2009) and the IMF data was plotted from Lejeune &Schaerer (2001) for a cluster less than 10 years old.While there are clearly reddened sources in the field,the CCD does not show an obvious excess of intrinsi-cally red stars towards G13.384+0.064. The reddeningvalue for each star was measured and a KS test wascompleted to see if the two populations of data (on andoff source) were similar. The results suggest that twopopulations could not be differentiated, making the redstars (from K band UKIDSS data) in Figure 4 likely tobe field stars and not representative of a cluster. Theonly possible exceptions are the two blue dots at [K] − [3.6] ∼ § . . . Galactic longitude G a l a c t i c l a t i t ude Figure 4.
Three-colour image with red at K band, green at Hband and blue at J band of UKIDSS data overlaid with contoursfrom CORNISH 5 GHz radio continuum emission with contoursat 0.19 and 0.32 Jy/beam.
The near kinematic distance of 1.9 kpc and far distanceof 14.1 kpc was measured by Schlingman et al. (2011)using HCO + (3-2) and N H + (3-2) observed by the Hein-rich Hertz Submillimeter Telescope at 1.1 mm. The ve-locity observed for these two molecules are 15.1 km s − for HCO + and 14.1 km s − for N H + . Lockman (1989)reported 3 cm radio recombination line observationscompleted using the Greenbank telescope with an ob-served V LSR of 18.3 ± − . However, Lockman(1989) commented that the FWHM line for this sourcewas so broad that the results could not be easily inter-preted.H CO was observed at 4.83 GHz, using the Nanshanstation (Du et al. 2011). The results for the sourceG13.384 +0.064 show two molecular clouds. The firstcloud has a velocity of 10.71 ± − and the sec-ond has a velocity of 51.34 ± − . The velocityof the first cloud agrees well with the HCO + , N H + and radio recombination observations. Data from theJames Clark Maxwell Telescope (JCMT) reported threevelocity components for CO(J=3-2) at 10, 15 and50 km s − (Dempsey et al. 2013). It is likely, then,that the star formation is associated with the gas at15 km s − , rather than the 50 km s − , since the gas at50 km s − only appears in the relatively low density gastracers of H CO and CO and the star formation ismost likely associated with the denser gas.The distance to this region has been determined bySchlingman et al. (2011) from the radial velocity and
PASA (2018)doi:10.1017/pas.2018.xxx n Isolated Star? near and far distances.These distance ambiguities can be resolved using H I data by comparing the velocity of absorption dips seenin the spectra with the source velocity as measuredfrom thermally excited molecular lines. The two mostcommonly used methods are H I self-absorption (HISA;e.g., (Jackson et al. 2002; Roman et al. 2009)) and H I emission-absorption (HIEA; e.g., Kolpak et al. ( 2003),Andersen & Bania (2009), Urquhart et al. (2012)).In Figure 5 we present the HISA profile (continuumsubtracted) and HIEA (continuum included) H I profilesseen towards G13.384+0.064 and its associated H II re-gion. The source velocity is approximately coincidentwith the velocity of a broad absorption feature seen inthe H I profile; this is consistent with the source beinglocated at the near distance due to the fact that thereis too much intervening warm H I gas at the same ve-locity as the source for any absorption to be present forsources located at the far distance.The HIEA method is based on the principle that forany strong emission source located at the far distancewe would expect to observe absorption features at allvelocities up to and including the tangent velocity. Thisis due to the high density of cold H I clouds along anyline of sight through the inner Galaxy. The lack of anyabsorption features between 50 and 150 km s − wouldsuggest that the source is again located at the near dis-tance. The absorption seen at 50 km s − may suggestthat the source velocity may be incorrectly assigned tothis source and may in fact be associated with anotherobject within the line of sight. However, the lack of anyemission from high-density molecular tracers (HCO + and N H + ) would rule out this possibility. Both meth-ods therefore suggest a near distance is more likely. Forfurther discussion see § ±
10 km s − when the streaming motions andpeculiar velocities are considered, is ± The near distance of 1.9 ± ±
47 and105 ±
73 M (cid:12) , respectively. Urquhart et al. (2013) usedthe integrated flux measured in ATLASGAL (Contreraset al. 2013) at 870 µ m and a temperature of 20 K. Miet-tinen (2012) used 870 µ m observations from LABOCAand assumed a temperature of 35 K. Since both reportedmasses are from the same instrument the differencesare from the different temperatures and calibration er-rors. By using the Hildebrand (1983) equation and mak- Figure 5.
The H I continuum subtracted (top) and H I continuum(bottom) profiles seen towards G13.384+0.064 and its associatedH II . In both of these panels, the source and tangent velocities areindicated by the red and blue vertical lines and the grey regionshows a region 10 km s − either side of the source velocity. In thelower panel the green line indicates the maximum velocity foundof the absorption features and the magenta line shows the 5 σ rmsnoise for the H I data (see Urquhart et al. (2012) for more details).The presence of an absorption feature at a similar velocity as thesource in the upper panel and the lack of absorption features upto the tangent velocity in the lower panel both strongly supportand near kinematic distance for this source. ing the same assumption as stated in equation 1 ofUrquhart et al. (2013), we determine the cloud massto a greater accuracy from the derived values specificto this source. M clump M (cid:12) = ( Dkpc ) ( S ν mJy ) RB ν ( T dust ) κ ν (1)where S ν is the integrated flux at 870 µ m of603.94 mJy (Schuller et al. 2009), D is the heliocentricdistance to the source (1.9 ± § B ν is the Planckfunction for a dust temperature T dust (calculated in § . ± . κ ν is the dust absorption coeffi-cient taken as 1.85 cm g − (as used by Urquhart et al.2013). This yields a value for the clump mass aroundthe star as 57 ±
35 M (cid:12) . PASA (2018)doi:10.1017/pas.2018.xxx
Tremblay et al.
An estimation of the luminosity was determined fromthe IRAS fluxes (Neugebauer et al. 1984) using theequation initially published by Casoli et al. (1986) butusing the same assumptions as equation 3 of Walsh etal. (1997). F tot = ( f δν + f δν + f δν + f δν ) / . . × L (cid:12) . This value is simi-lar to that reported by Miettinen (2012) of 7 . × L (cid:12) ,which also uses the IRAS flux, but using the same as-sumptions as Casoli et al. (1986). The bolometric lumi-nosity reported by Urquhart et al. (2013), calculated byscaling the MSX 21 µm flux, was 3 . × L (cid:12) which isalmost half the value calculated from the IRAS fluxes.We note that the luminosity values reported by Miet-tinen (2012), Urquhart et al. (2013) and contained inthis work, all assume a distance of 1.9 kpc.The IRAS measurements can be considered as astrong upper limit on the bolometric luminosity. Thelarge IRAS beam encapsulates all emission and there isno loss of flux due to extended emission that is maskedout with other observations employing nodding or jit-tering method, such as BOLOCAM or ATLASGAL.A more accurate value of the bolometric luminosityis measured using two component fitting of the spectralenergy distribution (SED). The integrated flux data wascompiled from a combination of reported catalogues val-ues ranging from 4 µ m to 21 cm as well as measured val-ues through aperture photometry from the MSX (Priceet al. 2001), Hi-GAL (Molinari et al. 2010) and AT-LASGAL (Schuller et al. 2009) maps. A plot of the in-tegrated fluxes versus wavelength shows a curve thatpeaks in the IR and is flat at radio wavelengths asshown in Figure 6. This curve is typical of the spectralenergy distribution of an embedded forming high massstar. The peak in the infrared emission is characteristicof the short wavelength stellar light being reprocessedto longer wavelengths by dust in the surrounding gascloud, and the flat region is from free-free radio emis-sion characteristic of an H II region.The flux density from the MSX, Hi-GAL and ATLAS-GAL maps, Fapp, was measured within an aperture ra-dius of 35.7, corresponding to 3 σ of a Gaussian fitted tothe source and centred on the peak flux pixel positionof the 250 µ m image. The background flux density, Fbg,was obtained as the median pixel value from an annu-lus with r inner =47.6 arcsec to r outer =59.5 arcsec aroundthe aperture. Subtracting the background flux densityfrom the aperture flux density yields the backgroundcorrected source flux F = F app − F bg , reconstructingthe SED in 10 bands from 8 µ m to 870 µ m. The errorsof the flux densities are calculated by adding the ab-solute calibration uncertainty to the intrinsic measure- ment error in quadrature. We assume a measurementuncertainty of 20 per cent for all bands except for the500 µ m band, where we take into account the large pixelsize of 15 arcseconds, hence assuming a measurementuncertainty of 50 per cent.The SED was then fitted with a two-componentmodel consisting of a greybody and blackbody. Thegreybody (i.e. a modified blackbody) models the colddust envelope’s emission, taking into account the wave-length dependence of the dust in the far-infrared tosubmm wavelength regime, whereas the blackbodymodels a hot, optically thick, deeply embedded com-ponent: F λ ( T d , β, τ , T h , Ω h ) = F λ, hot ( T h , Ω h ) + F λ, dust ( T d , τ ) (3)where F λ, hot is the hot component given by a blackbodyscaled with the effective solid angle of the hot compo-nent and F λ, dust is the greybody emission from the dustenvelope given by: F λ, dust ( T d , τ ) =Ω d · B λ ( T d ) · (cid:16) − e − τ ( µ m λ ) β (cid:17) (4)where Ω d is the solid angle subtended by the source, B λ ( T d ) the blackbody intensity at the dust tempera-ture T d , τ the dust optical depth at the referencewavelength of 870 µ m and β the dust spectral index. Weleave the dust spectral index β fixed to a value of 1.75,as computed as the mean value from the dust opacitiesof Ossenkopf & Henning (1994) for the submm regime.As a result the dust temperature was determined to be33 . ± . ± × L (cid:12) which is comparable to that reported by Urquhart etal. (2013) and is consistent with the upper limit fromthe IRAS measurements . A number of observations of the region were completedin the radio using the NRAO Very Large Array (VLA)(Zoonematkermani et al. 1990; Becker et al. 1994; Gar-wood et al. 1988; Purcell et al. 2013), Nanshan Ra-dio Telescope (Du et al. 2011), NRAO Green Bank(GBT), (Lockman 1989) and the Effelsberg 100’m Tele-scope (Alltenhoff et al.1978). The integrated flux fromtwo 6 cm observations, CORNISH (Purcell et al. 2013)and VLA 5 GHz Survey (Becker et al. 1994), were usedto calculate the Lyman-continuum flux based on equa-tion 1 and 3 in Kurtz et al. (1994), which representmodified equations presented by Mezger & Henderson(1967). Both of these surveys were completed using theNRAO VLA but CORNISH used B configuration,witha restoring beam of 1.5 (cid:48)(cid:48) , which is not as sensitive to
PASA (2018)doi:10.1017/pas.2018.xxx n Isolated Star? (cid:48)(cid:48) . Therefore, we expectthe C-array data to recover more emission. The loga-rithm of the number of Lyman continuum photons persecond (Log N c ) using the CORNISH integrated fluxof 603 mJy and an assumed gas to dust ratio of 100,was determined to be 47.3 photon s − , as reported byUrquhart et al. (2013). Using the integrated flux fromBecker et al. (1994) of 891.8 ± c ) =47.3 +0 . − . photon s − .The flux derived by Altenhoff et al. (1978) for the100 m Effelsberg telescope at 4.9 GHz is 0.9 Jy, which issimilar to the value reported by Becker et al. (1994) of0.891 Jy. This suggests that all the extended emissionwas accounted for in the C configuration observations. The Lyman-continuum flux and bolometric luminosityare compared to Table 1 in Davies et al. (2011) to de-termine a mass of the star powering the H II region.The bolometric luminosity as determined by the SEDfit is Log(L (cid:63) /L (cid:12) ) = 3.61 ± (cid:12) . The Log of the Lyman photon fluxis 47.3 +0 . − . photon s − . This relates to a star between15-20 M (cid:12) .The mass measured using the SED fit is 13 ± (cid:12) .A main sequence star type B, is a star with a mass of2.1-16 M (cid:12) which is consistent with the values measuredfor this source. As B stars are known to have an ex-cess of Lyman flux (see § (cid:12) as suggested by the bolometric lumi-nosity measurement (as suggested by Table 1 in Davieset al. (2011) and SED curve in Figure 6). Figure 2 shows a bright infrared star at G13.380+0.050,approximately one arcminute to the south. This star co-incides with a 1612 MHz OH maser, detected by Seven-ster et al. (2001), indicating that the star is evolved andunlikely to be part of the same star formation process.Given the offset between this star and G13.384+0.064,the projected distance is 0.5 pc. The image in Figure2 shows the structure of the cloud in the infrared ex-tends toward the star at G13.380+0.050 but does notoverlap with this star. Furthermore, there is no edgebrightening in the extended infrared emission observed.This suggests that the bright star is unlikely to have adirect physical influence on the star forming region andis most likely an unrelated star projected along the lineof sight, but not necessarily physically close to the starforming region. Wavelength λ [ µ m] F l u x den s i t y S ν [ J y ] T d = 33 . ± . AGAL013.384+00.064
Figure 6.
Spectral energy distribution created by data of dif-ferent surveys plotted on a log scale. To obtain the dust tem-perature and luminosity, a two-component model was fitted tothe flux densities measured through aperture photometry fromthe MSX (Price et al. 2001), Hi-GAL (Molinari et al. 2010) andATLASGAL (Schuller et al. 2009) maps (blue fit). Additionally,catalogued data of the extended emission from the radio as wellas some select infrared and submilimetre data. Open Triangle-MSX (Egan et al. 2003), Diamond-BOLOCAM (Schlingman etal. 2011), Triangle-Nobeyama (Handa et al. 1987), Open Circle-VLA 5 GHz (Becker et al. 1994), and Triangle Right-VLA 1.5 GHz(Garwood et al. 1988).
The radio continuum-derived luminosity of3 . × L (cid:12) derived in § . × L (cid:12) .We must reconcile this difference.If we assume that G13.384+0.064 is at the far kine-matic distance of 14.1 kpc, then the radio-derived anddust-derived luminosities are 3 . × L (cid:12) and 2 . × L (cid:12) , respectively. These two luminosities are moreconsistent than those when the near distance is as-sumed. However, even though this luminosity argumentfavours the far kinematic distance, we still favour thenear kinematic distance for the following reasons.Firstly, as discussed in § I observationsstrongly favour the near distance. The H I observationsanalysed by Urquhart et al. (2013) and shown in Fig-ure 5 shows little evidence of high velocity H I self-absorption, which would be expected from interveninggas, if it was at the far kinematic distance. The only highvelocity absorption is seen around 50 km s − . This cor-responds to the velocity of the cloud seen in the H COand CO observations reported previously and so theabsorption is most likely not related to G13.384+0.064.
PASA (2018)doi:10.1017/pas.2018.xxx
Tremblay et al.
Secondly, there are uncertainties in the expected Ly-man continuum flux from young high mass stars. Smith(2014) concludes that the Lyman continuum flux ofearly B type stars may be in excess of previous models,caused by the accretion of cold gas from the circum-stellar disk onto hot-spots on the surface of the youngstar. Comparing the measurements for G13.384+0.064,we find that it sits on the upper edge of their modelsuch that the excess radio continuum flux is consistentwith the bolometric luminosity. If the radio continuumflux were much larger, this source would show an unac-counted for excess of radio continuum, compared to thedust-derived luminosity.A final consideration is the expected number of lowermass stars surrounding the high-mass star if it were atthe near vs far kinematic distance. If G13.384+0.064was at the far distance of 14.1 kpc, then the mass of thestar would be approximately 38 M (cid:12) , based on the massto luminosity relationship. The mass of the surroundinggas cloud would increase to 5800 M (cid:12) , making it verylikely that a cluster is present. However, as discussed in § We now seek to use the properties derived above to de-termine whether G13.384+0.064 is indeed a high-massstar forming in isolation.Parker et al. (2007) defines an isolated B type star asa star 10 M (cid:12) < M ∗ < . (cid:12) in which the cluster massis <
100 M (cid:12) and there are no O-stars present. The es-timated mass of the star based on the SED curve is13 ± (cid:12) and the estimated remaining clump mass is57 ±
35 M (cid:12) . Both of these are consistent with the defini-tion described by Parker et al. (2007) for isolated highmass star formation.Based solely on the detection of a strong radio contin-uum source, we can conclude that there must be at leastone embedded high mass star. Incorporating a model bySmith (2014), we find that such a star must account fornearly all (if not, all) of the infrared flux, leaving littleroom for a cluster. This is because lower mass stars willnot contribute significantly to the radio-derived lumi-nosity, but will significantly contribute to the infraredor dust-derived luminosity.We can model what a cluster might look like, byconstraining the quantities of dust-derived luminosityand Lyman continuum flux. In order to do this, we usea Monte-Carlo simulation of the initial mass function(IMF) to generate a cluster. Based on previous workby Walsh et al. (2001), we found that commonly-usedfunctions of the IMF do not greatly affect the make-upof simulated clusters, so we choose the IMF model ofKroupa et al. (1993). In our simulation, we randomly generate stars between masses of 0.1 and 100 M (cid:12) , ac-cording to this IMF and then calculate the cluster phys-ical parameters, such as total mass, luminosity and Ly-man continuum flux. In order to measure the relation-ship between Lyman continuum flux and luminosity, weuse the values given in Table 1 of Thompson (1984) butwe note that previous studies have found an excess ofLyman continuum photons from early B-type stars (eg.Urquhart et al. 2013). Therefore, in order to take intoaccount the work of Smith (2014) we apply their mostextreme case for ratio of Lyman flux, compared to pre-vious models, where the Lyman flux may be reducedby up to an order of magnitude for the same star withthe same bolometric luminosity (ie. reduction of 10 . to 10 . photon s − ). It is important to note that intaking such an extreme reduction in Lyman flux, oursimulation will favour the formation of a cluster, ratherthan an isolated star. This allows us to use the data ofThompson (1984) and correct it for more recent mod-elling by Smith (2014).In our Monte-Carlo simulation, we continue to addmembers to the cluster until the total cluster luminos-ity is greater than 4 . × L (cid:12) . We generated 68,877clusters with sufficient luminosity to meet this criterion.However, we note that the majority of generated clus-ters have total luminosities far in excess of this value.This is because the last star added to the cluster istypically a high mass star with very high luminosity.Thus, we exclude those generated clusters that haveluminosities in excess of the IRAS-derived luminosity(7 . × L (cid:12) ), leaving 30,601 clusters. We choose theIRAS-derived luminosity here because it is a strong up-per limit on the bolometric luminosity, given that IRASwill likely overestimate the total infrared flux, but notunderestimate it.Of our remaining clusters, we find that the medianluminosity for the highest mass star in each cluster is2 . × L (cid:12) which means that for most clusters, the lu-minosity is dominated by one star. We also find that thehighest mass star generated in any cluster has a lumi-nosity of 7 . × L (cid:12) , which we calculate has a corre-sponding Lyman continuum flux of 10 . photons s − .This Lyman flux is lower than we expect (10 . pho-tons s − ) by about an order of magnitude. In summary,our simulations indicate that it is very difficult to ran-domly generate a cluster with the properties that weobserve for G13.384+0.064. The only way to generate agood match is for the first star selected from the IMF tobe a high mass star with the right luminosity and Ly-man flux properties. However, classifying such a singlestar as a cluster is questionable. In this paper we compared the Lyman-continuum pho-ton flux and clump mass of approximately 200 star
PASA (2018)doi:10.1017/pas.2018.xxx n Isolated Star?
Acknowledgements
The authors would like to thank the referee for thecomments and advice. SNL would like to thank AndyLongmore for very helpful discussions about the analy-sis of the infrared data. This research has made use ofthe SIMBAD database, operated at CDS, Strasbourg,France. This research has also made use of the VizieRcatalogue access tool, CDS, Strasbourg, France. Theoriginal description of the VizieR service was publishedin A&AS 143, 23.
REFERENCES
Altenhoff, W.J, Downs, D. & Pauls. T, 1978, AAS, 35, 23Andersen L. & Bania T., 2009, ApJ, 690, 706-719Bastian N. et al., 2010, ARA&A, 48, 339Bate M. & Bonnell I., 1997, MNRAS, 285, 33-48Beck T. et al., 2010, ApJ, 722, 1360-1372Becker R.H. et al, 1994, A&AS, 91, 347, 387Benjamin, R. et al., 2003, PASP, 115, 953Bessell & Brett, 1988, PASA, 100, 1134Bressert E. et al., 2012, A&A, 542, 14Bonnell I, Bate M, & Zinnecker H. 1998, MNRAS, 298, 93Bonnell I. et. al., 2001, MNRAS, 324, 573Bonnell I. et. al., 2004, MNRAS, 349, 735Bonnell I. & Bate M, 2006, MNRAS, 370, 488-494Brand J. & Blitz L., 1993, AAP, 275, 67Carey S.J. et. al., 2009, ASP, 121, 875, 76-97Casoli et. al., 1986, AA, 169, 281-297Churchwell, E. et al., 2009, PASP, 121, 213Commercon B., Hennebelle P., & Henning T., 2011, ApJL,742, 1, 6Contreras Y. et al., 2013, JA&A, 549, 45Cutri R.M. et al., 2012, WISE Catalogues, 2311Davies B. et al., 2011, MNRAS, 416, 972-990Deguchi S. et al., 2005, ASJ, 57, 933-949Dempsey J.T. et al., 2013, ApJS, 209, 8 de Wit, W. J. et al., 2005, A&A, 437, 247Douguet N., Kokoouline V., & Greene C., 2009, PRA, 80Du et al., 2011, A&A, 532A, 127-127(2003), Edgar R. & Clarke C., 2004, MNRAS, 349, 678Egan et al., 2003, AFRLT, 1589Feldt et al., 1998, A&A, 339, 759Garwood R et al.,1988, ApJ, 96, 1655-1670Handa, J. et al., 1987, PASJ, 39, 709Hennebelle P. et al., 2011, A&A, 528, 17Hewett P. et al., 2006, MNRAS, 367, 2, 454Hildebrand R. H., 1983, QJRAS, 24, 267Indebetouw et al., 2005, ApJ, 619, 2, 931Izumi N et al., 2014, ApJ, 795, 1, 66Jackson et al., 2002, ApJL, 566, L81-L84Joshi G.C. et al., 2015, NA, 40, 68Kennicutt RC et al, 2012, ARA&A, 50,531-608Knudsen K.K. et al., 2007, ApJ, 666, 1Kolpak et al., 2003, ApJ, 582, 756-769Kroupa P. et al., 1993, MNRAS, 262, 545Krumholz et al. ,2007, ApJ, 656, 2, 59-979Krumholz et al. ,2009, AAAS, 323, 5915, 754-757Krumholz et al. ,2012, ApJ, 760, 2, 155Kurtz S. et al., 1994, ApJS, 91, 659-712Kuchar T.A. and Clark F.O., 1997, ApJ, 488, 224-233Kuiper R. et al., 2010, ApJ, 722, 1556-1576Larson R.B., 1982, MNRAS, 200,159-174Lawrence et al., 2007, MNRAS, 379,1599-1617Lockman F.J, 1989, ApJS, 71, 469-479Lejeune & Schaerer, 2001, A&A, 366, 538Mart´ın S. et al., 2006, ApJS, 164, 2, 450-476McClure-Griffiths et al., 2012, ApJS, 199, 12McKee C. and Tan J., 2003, AJ, 585, 2, 850-871McKee C. and Ostriker E., 2007, ARA&A, 45, 1, 565-687Mezger P.G. and Henderson A.P., 1967, ApJ, 147, 471Miettinen O. 2012, A&A, miettinen,Molinari et al, 2010, PASP, 122, 314-325Myers A. et al, 2013, ApJ, 766, 2, 18Neugebauer, G et al., 1984, ApJL, 278, L1-L6Nishiyama et al., 2096, ApJ, 696, 2, 1407Myers A. et al, 2013, ApJ, 766, 2, 18Ossenkopf V. & Henning T., 1994, AAP, 291, 943-959Panagia N., 1973, AJ, 78, 929-934Parker R. and Goodwin S. , 2007, MNRAS, 380, 3, 1271-1275Peters T. et al, 2010, ApJ, 711, 2, 1017Pillai T. et al, 2011, A&A, 530, 118Purcell C.R. et al., 2013, ApJS, 205, 1Price S. et al., 2001, AJ, 121, 2819Roman et al, 2009, ApJ, 699, 1153-1170Rosolowsky E. et al., 2010, ApJS, 188, 123-138Schilbach, E. & R¨oser, S. 2008, A&A, 489, 105Schlingman W. et al., 2011, ApJS, 195, 2, 14
PASA (2018)doi:10.1017/pas.2018.xxx Tremblay et al.