Spitzer-MIPS survey of the young stellar content in the Vela Molecular Cloud-D
T. Giannini, D. Lorenzetti, M. De Luca, B. Nisini, M. Marengo, L. Allen, H. A. Smith, G. Fazio, F. Massi, D. Elia, F. Strafella
aa r X i v : . [ a s t r o - ph ] S e p To appear in Ap. J.
Spitzer-MIPS survey of the young stellar content in the VelaMolecular Cloud-D
T.Giannini , D.Lorenzetti , M. De Luca , , B.Nisini , M.Marengo , L.Allen , H.A.Smith ,G.Fazio , F.Massi , D.Elia , F.Strafella ABSTRACT
A new, unbiased Spitzer-MIPS imaging survey ( ∼ µ m and 70 µ m, respectively. Atotal of 849 sources are detected at 24 µ m and 52 of them also have a 70 µ mcounterpart. The VMR-D region is one that we have already partially mappedin dust and gas millimeter emission, and we discuss the correlation between theSpitzer compact sources and the mm contours. About half of the 24 µ m sourcesare located inside the region delimited by the CO(1-0) contours, correspondingto only one third of the full area mapped with MIPS. Therefore the 24 µ m sourcedensity increases by about 100% moving from outside to inside the CO contours.For the 70 µ m sources, the corresponding density increase is four times. About400 sources of these have a 2MASS counterpart, and we have used this to con-struct a K s vs. K s -[24] diagram and to identify the protostellar population insidethe cloud.We find an excess of Class I sources in VMR-D in comparison with other starforming regions. This result is reasonably biased by the sensitivity limits at 2.2and 24 µ m, or, alternatively, may reflect a very short lifetime ( . yr) of theprotostellar content in this molecular cloud. The MIPS images have identified INAF - Osservatorio Astronomico di Roma, via Frascati 33, 00040 Monte Porzio, Italy, giannini, deluca,dloren, [email protected] Dipartimento di Fisica - Universit`a di Roma ‘Tor Vergata’, via della Ricerca Scientifica 1, 00133 Roma,Italy Harvard-Smithsonian Center for Astrophysics, Cambridge, MA, USA INAF - Osservatorio Astrofisico di Arcetri, Largo E. Fermi 5, 50125 Firenze, Italy Dipartimento di Fisica - Universit`a del Salento, CP 193, 73100, Lecce, Italy
Subject headings:
Stars: formation – surveys –ISM: individual (Vela MolecularRidge) – ISM: clouds – ISM: jets and outflows – infrared:stars
1. Introduction
Infrared maps of star forming Giant Molecular Clouds (GMCs) are an essential toolin the modern study of star formation. When radio and millimeter maps also exist, therelationships between the regions of infrared, millimeter and radio activity provide one ofthe key new tools for clarifying the varieties of star formation that can occur. The sensitivityof infrared techniques means that even shallow surveys can in principle reveal the processesof both low and high mass star formation in clouds that are not too far away. There are,however, very few nearby GMCs with which to take full advantage of these techniques.Among these available targets, one is the Vela Molecular Ridge (VMR), a complex of fouradjoining GMCs (Murphy & May 1991; Yamaguchi et al. 1999), located in the galactic plane( b = ± ◦ ) outside the solar circle ( l ∼ ◦ - 275 ◦ ); most of the gas (clouds named A, C andD) is located at a distance of about 700 pc (Liseau et al. 1992).This team has studied the star formation activity in the VMR for many years: theconcentration of red and young sources (Liseau et al. 1992, Lorenzetti et al. 1993); thepresence of embedded clusters (Massi et al. 2000, 2003); the occurrence of protostellar jets(Lorenzetti et al. 2002; Giannini et al. 2001, 2005, De Luca et al. 2007, hereinafter D07).Recently we mapped with the SIMBA bolometer array at SEST a ∼ area of thecloud D in the 1.2 mm continuum of dust emission, and in the CO(1–0) and CO(2–1)transitions (Massi et al. 2007, hereinafter M07; Elia et al. 2007, hereinafter E07).The advent of the Spitzer Space Telescope (SST, Werner et al. 2004) and the imagingphotometric facilities on board, i.e. the Multiband Imaging Photometer for Spitzer (MIPS,24, 70, 160 µ m; Rieke et al. 2004) and the InfraRed Array Camera (IRAC, Fazio et al. 2004)has enabled us to obtain maps of the VMR from 3.8 to 70 µ m across the same area alreadysurveyed in the millimeter emission of dust and gas. The primary goal of this survey is toobtain a census of the embedded young stellar population of VMR-D and to correlate it withits gas and dust cores. 3 –This paper describes our MIPS observations of the VMR-D; it is the first of a series ofpapers we are preparing dealing with VMR-D as seen by Spitzer; the IRAC data of the sameregion will be presented in a separate paper. Spitzer surveys of several other star formingregions have been already published, most of them in the framework of the cores-to-disk(c2d) legacy project (Evans et al. 2003). Most of these regions, however, are located outsidethe Galactic plane ( | b | > ◦ ) in regions that were originally selected in part to avoid strongconfusion and extinction problems. A huge amount of observational material has been sofar accumulated and published on those clouds. For the VMR, our current multi-frequencydatabase, when combined with the increased sensitivity of current instruments, has allowedus to overcome many of the problems associated with observations of GMCs in the galacticplane. Since the plane is where most of the material currently forming stars is located, itis both a natural and critical region to understand, and will also help with the comparisonbetween the derived properties of our Galaxy with those of external galaxies, whose planesare the unique zones we are able to sample.Spitzer has surveyed many different types of star formation regions. Therefore, legiti-mate comparisons between them all would benefit from a standard analysis and presentation,although some problems could arise because their numerous differing parameters, as well asthe various details of the observations. In this paper we therefore adopt as much as possiblethe methods that have already been used successfully in the c2d program.Our paper is organized as follows: in Sect. 2 we give the details of the observations anddata reduction procedure; the results are presented in Sect. 3 and discussed in Sect. 4 and 5.Concluding remarks are given in Sect.6.
2. Observations
VMR-D cloud was observed with MIPS on board the Spitzer Space Telescope withinthe Guaranteed Time Observation program (PID 30335). The observations covered ∼ × α , δ (J2000) = 8 h m s , -43 ◦ ′ ′′ and8 h m s , -43 ◦ ′ ′′ at 24 µ m and 70 µ m, respectively (orientation: 145 ◦ W of N).Data were collected on 14 Jun 2006 in scan mode, medium speed, with 5 scan legs and160 ′′ cross-scan step, resulting in a total integration time of 40 seconds (for both 24 and70 µ m) per pixel. The mapping parameters were optimized for the 24 and 70 µ m bands: asa consequence, the 160 µ m map suffers from coverage gaps and saturation and will be notconsidered in the following.The SSC-pipeline, version S14.4.0, produced basic calibrated data (BCDs) that we have 4 –used to obtain mosaiced, pointing refined images by means of the MOPEX package providedby the Spitzer Science Center (Makovoz & Marleau 2005).The main instrumental artifacts have been removed from the mosaiced images by meansof the MOPEX package. Minor problems of residuals jailbars (expecially at 70 µ m) and back-ground matching between adjacent frames (at 24 µ m) are still visible close to the brightestobjects, but they do not affect significantly the point source photometry discussed in thispaper.The final 24 µ m map global properties can be summarized as follows: pixel scale of2.45 ′′ /pixel, background r.m.s. of 0.3 µ Jy/arcsec , within the regions of high level of diffuseemission. The brightest sources saturate at the emission peak: for these we estimate a lowerlimit to the integrated flux of 4 Jy. In the 70 µ m map the pixel scale is 4.0 ′′ /pixel andthe background r.m.s. ranges between 23 and 94 µ Jy/arcsec . None of the detected sourcesappears saturated at this wavelength.
3. Results
Figure 1 shows the two-color final mosaic of VMR-D (24 µ m in blue, 70 µ m in red), whileon Figures 2 and 3, images in each of the two filters are shown separately. In these latter,the 1.2 mm dust map (adapted from M07) and the CO intensity map integrated in thevelocity range -2 ÷
20 km s − (adapted from E07) have been superimposed for comparison.We define as ‘on-cloud’ all objects inside these latter contours. Such a definition is perforcejust the first level effort of delimiting the sources belonging to the molecular cloud; indeed,it is clear from Figures 2 and 3 that the CO emission remains well above the 3 σ level atthe north and west borders of the gas map, and thus sources belonging to VMR-D couldexist toward these directions. Considering such sources as ’off-cloud’ will have the effect ofreducing the distinctions between the ’on’ and ’off’ cloud populations; these sources shouldthen be considered on a case-by-case basis (see sect.5.1). We have also considered as ’off-cloud’ those regions where the CO peak velocity is faster than 20 km s − , since they arelikely to be more distant and unassociated with VMR-D (see Figure 1 in Lorenzetti et al.1993).The point-source extraction and photometry processes were performed by using the DAOPHOT task of the astronomical data analysis package
IRAF . Given the size of the IRAF, the Image Reduction and Analysis Facility, is a general purpose software written and supportedby the IRAF programming group at the National Optical Astronomy Observatories (NOAO) in Tucson, µ m in blue, 70 µ m in red) of VMR-D. 6 –Fig. 2.— Mosaic of VMR-D map at 24 µ m, with superposed the CO intensity map (whoselimits are depicted in yellow), where the contours (in green) are in the range -2 - 20 km s − (adapted by E07). Also overlaid is the 1.2 mm dust emission map (red contours, adaptedfrom M07). CO contour levels start from 5 K km s − and are in steps of 25 K km s − , whiledust contours start from 50 mJy/beam and are in steps of 50 mJy/beam. 7 –Fig. 3.— As Figure 2 for the 70 µ m map. 8 –MIPS mosaic it was impossible to apply any automatic procedure for finding sources downto the sensitivity limits without being affected by a locally varying background level; wetherefore applied a searching algorithm as deep as possible, but still compatible with anautomatic procedure. The search algorithm was applied to a differential image we producedbetween the final mosaic and a ’sky’ image, the latter obtained by applying to the mosaica median filter over boxes of 5 × σ has been imposed on the sky-subtracted image, which corresponds at least to 5 σ (depending on the local background) inthe unsubtracted image.The automated methods just described lead to the detection of 838 and 61 point sourcesat 24 µ m and 70 µ m, respectively. A further 12 detections have been added to the 24 µ mlist by applying local sky values in selected areas (see the discussion below). The sourcedistribution as a function of the measured flux is depicted in Figures 4 and 5, where thecompleteness limits can be evaluated as the flux bin corresponding to the maximum countsbefore the decline at lower fluxes due to the instrumental sensitivity. We determine that oursample is complete down to 5 and 250 mJy at 24 and 70 µ m, respectively.A statistical summary of the detected sources is presented in Table 1. About 45% of the24 µ m sources are spatially located inside the region delimited by the CO contours ( ∼ ), even though this latter is about one half the size of the remaining mapped area ( ∼ ). This result gives an initial indication of how the IR source density increases byabout 100% moving from outside to inside the CO contours, and the pattern becomes evenmore significant when considering the 70 µ m sources, whose source density increase is fourtimes. The 24 µ m counts per deg are represented in Figure 6, left panel, where differentsymbols indicate those sources located respectively within and outside the CO contours.Also in this plot, where the differential number density is shown, there is a drop for F ν <
35 mJy. For greater F ν values the number of objects inside the gas contours (i.e. thosemore likely associated to the cloud) systematically exceeds the number of objects outsidethe cloud, giving reasonable support to the empirical significance of this crude classification.In addition, Figure 6 is a comparison between the on- and off-cloud samples and the SpitzerWide-area Infrared Extragalactic Survey (SWIRE, Lonsdale et al. 2003) legacy program.A significant amount of contamination from the extragalactic background is predicted (at24 µ m) for flux densities <
10 mJy down to the completeness limit, so that ’on’ and ’off’source populations at this level become undistinguishable.The counts per deg at 70 µ m are depicted in Figure 6, right panel: here again thepossible extragalactic contamination appears just at (or even below) the completeness limit. Arizona (http://iraf.noao.edu). µ m. The completeness limit is around 5mJy, as indicated by the vertical line.Fig. 5.— Histogram of the sources detected at 70 µ m. The completeness limit is around 250mJy, as indicated by the vertical line. 10 – Log F (24) (mJy) -4-202 L ogd N / d F ( m J y - d e g - ) SWIRE galaxies countsVMR-D within CO contoursVMR-D out of CO contours
Log F (70) (mJy) -4-202 L ogd N / d F ( m J y - d e g - ) SWIRE galaxies countsVMR-D within CO contoursVMR-D out of CO contours
Fig. 6.— Left panel: differential number counts at 24 µ m. Thick and thin lines refer to sourcein VMR-D within and outside the CO contours, respectively. Extragalactic backgroundsources from the SWIRE ELAIS N1 field are shown for comparison (these latter have beentaken from Figs.6 and 7 in Rebull et al. 2007, hereinafter R07). Right panel: as in left panelat 70 µ m. 11 –In the same Figure 6, we also note that the 70 µ m counts confirm the on- and off-clouddistributions already found at 24 µ m.The complete catalog of the detected sources is given in electronic form (a short sampleversion is printed in Table 2). In Table 3 we show the list of the 70 µ m detections: of the61 sources, 52 of them are coincident with a 24 µ m source (i.e. the distance in both rightascension and declination is less than the 20 ′′ PSF radius at 70 µ m, see the summary ofTable 1). In Table 3, we list the 24 µ m coordinates (which are more accurate than the70 µ m ones because of the smaller PSF at 24 µ m), the distance from the 70 µ m coordinates,(∆ α /∆ δ ) , the measured flux at 24 and 70 µ m along with the relative uncertainties, a flagindicating whether or not the source is located inside the region delimited by CO emissioncontours, and the association with a dust core, if any. This latter is based on the distancebetween the 24 µ m and mm coordinates, (∆ α /∆ δ ) mm , which must be within the SIMBAHPBW of 24 ′′ . All the dust cores associated with a 24 µ m source are also associated withits 70 µ m counterpart.Nine 70 µ m sources have no 24 µ m counterpart. Four of these were not imaged at 24 µ mbecause of the shift between the two maps, four appear as diffuse or with a filamentarystructure at 24 µ m, and one has F < σ upper limit): this source ( µ m sources with dust coresfound in VMR-D by M07. The detailed study of these sources will be addressed in a futurepaper; here we give some preliminary results and point out some statistical aspects. In theregion mapped in the dust emission at 1.2 mm (see Figures 2 and 3), a robust sample of 29cores has been revealed, along with 26 cores whose size is below the map spatial resolution(24 ′′ ). D07 have associated 12 of these cores (8 resolved and 4 under-resolved) with anIRAS or MSX point source, while the remaining 43 cores are not associated with any FIRcounterpart, so that they appear to be either cold Class 0 sources/starless cores (in case ofresolved cores) or possibly data artifacts (in case of under-resolved cores). As stated in D07,such a high fraction of starless cores as compared to protostellar cores is most likely a resultof the poor sensitivity of the IRAS/MSX facilities. Our significantly more sensitive MIPSdata offers the opportunity to check whether or not such a bias exists, and to eventually findweak counterparts of the dust cores. In order to resolve this issue we closely reexamined ourmaps, performing photometry on the mm peaks coordinates using local rather than globalthresholds for the background level. This technique turned up 12 new objects at a fluxdensity as low as 0.7 mJy at 24 µ m, fainter than the completeness limit by more than afactor of 7.This procedure, together with automatic finding described above, when applied overall 12 –led to the association of 23 resolved and 20 under-resolved cores with 58 sources at 24 µ m,19 sources at 70 µ m (in some cases we found multiple associations), thereby dramaticallyincreasing the percentage of cores associated with an embedded protostar from 22% (D07)to 78%. This result is in general agreement with recent MIPS findings in other GMCs thathave substantially modified the percentage of active vs. inactive cores in favor of the former(e.g. Young et al. 2004). We also note that the existence of a MIPS counterpart to 20 outof 26 under-resolved cores significantly reduces the possibility that these objects are simplydata artifacts. The lack, even at the MIPS sensitivity, of a FIR counterpart to five resolveddust peaks (namely MMS 6, 13, 15, 20, 24 in the list by M07) makes these objects a veryrobust sample of genuine starless cores.
4. Comparison with IRAS sources
The similarity of the MIPS 24 and 70 µ m bandpasses to the 25 and 60 µ m filters on-board IRAS offers us the opportunity to evaluate directly the reliability of the IRAS pointsource catalogue (IRAS-PSC) fluxes in crowded and diffuse clouds like VMR-D, objects thatare commonly found in the galactic plane. A similar study has already been performedby R07 in the Perseus molecular cloud; although in this case the geographic location makesextended emission and source confusion less critical, only 61% (at 25 µ m) and 32% (at 60 µ m)of the objects of the IRAS-PSC are recovered by MIPS as point-like sources, while all theothers, although detected, remain confused by nebulosity. Higher rates of coincidence arefound, at least at 25 µ m, if the Faint Source Catalogue (FSC) - produced by point-sourcefiltering the individual detector data streams - is used.Unfortunately, the FSC does not cover the galactic plane, so that we cannot confirmthis result on VMR-D. Here, a total of 57 high (f qual =3) or moderate (f qual =2) qualitydetections are listed in the IRAS-PSC catalogue at 25 µ m; 46 of them (80%) are also seenby MIPS and recovered with our algorithm, while the remaining 11 IRAS objects appearas diffuse emission at 24 µ m and are thus undetected as point-sources. The matching ratefor VMR-D is thus higher than in Perseus. The same trend is seen at 60 µ m, where out of48 IRAS-PSC items, the retrieval rate is of about 50%. In Table 4 we give the list of theIRAS-PSC (with any f qual ) not recovered by MIPS. The IRAS sources in the table markedas ‘off-edge’ in one MIPS bandpass are necessarily ‘on-edge’ in the other, because of thespatial shift between the two focal plane arrays. Along with the f qual flag, we also give inTable 4 the IRAS correlation coefficient flag (cc) which provides an indication of the point-likeness confidence of the detected source. This flag is coded as alphabetical character andsubsequent letters correspond to decreasing accuracy (i.e. A > > qual =2,3), show, on average, ‘cc’ flag equal to ”A” or”B”: such an occurrence can thus be translated into a suitable tool to broadly distinguishbetween genuine point-source and diffuse emissions, if MIPS maps (and FSC detections) areunavailable.
5. Color-Magnitude diagrams5.1. K s vs. K s -[24] About half of the 24 µ m detections have identifiable 2MASS counterparts at K s (limitingmagnitude of 15.3) within a radius of 5 ′′ . These 2MASS fluxes have been used to constructthe K s vs. K s -[24] color-magnitude diagram given in Figure 7, where MIPS sources inside andoutside the CO contours are shown with different colors. Also reported as hatched areas arethe loci of the extragalactic sources in the SWIRE survey. As expected for a molecular cloudin the galactic plane, there are very few extragalactic sources seen. A remarkable numberof objects fall at K s < s -[24] ∼
0, which, given our completeness limit at 24 µ mof 5 mJy, delimits the region of normal photospheres in VMR-D. Noticeably, in this part ofthe diagram, the number density (per deg ) of the ’off-cloud’ sources is larger than that ofthe ’on-cloud’ ones (100 vs. 69, see Table 4): in principle, all the unreddened photospheresdetected in VMR-D could be indeed foreground/background stars. More reasonably, wecan affirm that no increase of main-sequence stars (with respect to the adjacent field) isregistered in VMR-D, as expected because of the youth of the region.The thick squares indicate the effects of an extinction of A V = 10 and 50 mag, respec-tively, on the data. The quantitative A V map of the overall region by Dobashi et al. (2005)does not provide values in excess to 5-10 mag (below the saturation limit of the catalog of15 mag), while toward the dust cores A V can increase up to ∼
20 mag (M07 and E07). Wethus conclude that sources with K s -[24] > µ m (Greene et al. 1994), according to which differentevolutionary stages, from the accretion phase (Class I) to the beginning of the main-sequence(Class III), are manifested. The same authors have found that the 2 -10 µ m spectral indexdoes not change substantially when computed using fluxes up to 20 µ m (by using photometryin the Q band); this result allows one to extrapolate the 24 µ m flux for different spectralindexes and accordingly to compute the expected value of the K s -[24] color. The result of 14 – Fig. 7.— Color-magnitude diagram for the 2MASS K s -band and the MIPS 24 µ m sources.Of the 849 24 µ m sources in the MIPS map, 401 have a K s detection within a radius of 5arcsec. These are shown by red dots if located inside the CO contour map (180 sources) andby black dots if outside (221 sources). Large dots denote sources with 70 µ m detections,while arrows refer to sources saturated at 24 µ m. Hatched areas are the loci of the sourcesin the SWIRE survey (taken from R07). The thick line indicates the effect of the extinctionfor different values of A V (open squares refer to A V =10 and 50 mag). 15 –this procedure is given in R07, who furthermore requested that, to select Class III sourcesfrom normal photospheres and foreground/background stars, K s -[24] >
2. The spectral clas-sification derived for VMR-D is depicted in Fig. 7 and also reported in Table 5. The ratioof ‘on-cloud’ over ‘off-cloud’ objects within the same class increases with increasing K s -[24]values; moreover, most of the younger ’on-cloud’ objects are also detected in the 70 µ m band(large, red dots in Fig. 7). Noticeably, 70% of the ‘off-cloud’ objects showing the characteris-tics of the youngest and coldest sources (black dots at K s -[24] ≃
7) are located just outside theNorth and West borders of the CO gas map, therefore they are reasonably genuine membersof VMR-D; the remaining sources (30%) with the same colours, if not belonging to VMR-D,could represent star forming regions at larger distances.In summary, we find a definitely many more young sources associated with the cloud,but confirming that active star formation behind and/or in the close neighbourhood of ourcloud is going on as well.The relative percentages of sources attributed to different evolutionary stages (see Table 4)can be compared with those of other well studied star forming regions. Schmeja et al.(2005), in particular, have investigated number ratios of sources in different evolutionaryclasses in several star forming regions ( ρ Ophiuchi, Serpens, Taurus, Chamaleon I, IC348)basing on data obtained before than the Spitzer advent. They find, on average, that Class Isources are ∼ i) VMR-D is significantlyyounger than either Perseus or Serpens. Such an hypothesis is supported by the age estimatesof 1-2 Myr derived in the Perseus cloud (Palla & Stahler 2000, R07) and of 2 Myr derivedin Serpens (Djupvik et al. 2006) as compared with an age of 10 -10 yr towards the clustersof VMR-D (Massi et al. 2000); ii) our K s vs. K s -[24] diagram suffers from missing twoimportant categories of sources. One category is represented by the ∼
450 objects detected 16 –at 24 µ m, but without a K s -2MASS counterpart (see Table 1). The sensitivity limits ofour survey in terms of power density at a given wavelength are λ F λ (2MASS) ≃ λ F λ (24-MIPS) ∼ − W m − . This implies that these 450 sources are objects whose SED isrising with wavelength and thus they could be additional young objects that tend even toincrease the already anomalous percentage of Class I sources. The second category, however,is represented by the about 5 10 µ m counterpart.Their SEDs are allowed to decrease with increasing wavelength, therefore, although manyof them could be foreground or background objects unrelated with the VMR population,they undoubtly represent a potential reservoir of Class II and III objects. It should besufficient that a very small fraction of them ( ∼ µ m up toan order of magnitude fainter than in Vela, therefore allowing to trace the SED also for faintK s sources that decline going from the near- to the far-infrared. In any case, we expectto provide a more certain answer to this issue in the next future, by means of forthcomingIRAC images covering the relevant spectral bands at more adequate sensitivity. In Figure 8 the color-magnitude diagram based on MIPS fluxes alone is shown. Herethe sources detected in both bands are plotted; the large majority of them are on-cloud,although there is no clear difference between sources associated or not associated with dustcores (large dots). Remarkably, all sources (except 3) are located to the right of [24]-[70]=2.This value pertains to SED’s that increase with wavelength in such a way that ( λ F λ ) =2 × ( λ F λ ) . These red objects are much more numerous than the 70 µ m detections depictedin Figure 7, since the majority of them lack a 2MASS counterpart. Forthcoming IRACdata will help us to reconstruct their SED’s more adequately, giving constraints on theirluminosity and evolutionary stage. A few sources (6) lie in the locus corresponding to black-body temperatures ranging between 40 and 50 K. These are values theoretically predicted(Shu, Adams & Lizano, 1987) for a collapsing isothermal sphere, identified as Class 0 objects.These 6 sources are all located inside the CO cloud; two of them lie within a mm core andone ( jet (see Sect. 6). Although at the moment aquantitative evaluation of their sub-mm vs. bolometric luminosity cannot be given, theystill represent the best candidates for members of the youngest population within the cloud. 17 – Fig. 8.— Color-magnitude diagram [24] vs. [24]-[70], where only sources not saturatedat 24 µ m are plotted. Red/black dots refer to sources inside/outside the CO contour map(41/12 sources). Large dots denote sources associated with a dust core, while numberedsources are the candidates exciting sources of the jets discussed in Sect. 6. The hatched areashows the locus of the SWIRE survey (taken from Fig.11 in R07). 18 –
6. MIPS associations with H protostellar jets MIPS offers a chance to identify, for the first time, very embedded compact excitingsources of molecular (H ) jets, found by D07, that have so far remained undetected evenat the longest IRAS wavelengths. We recall once again that our completeness limits aresignificantly higher than the sensitivity limits, so that by scrutinizing our data-base for theweakest MIPS sources in selected areas, we have been able to discover new objects downto 0.7 mJy at 24 µ m. We used this technique to search for the sources driving a numberof molecular jets that were previously found by narrow band imaging centered at the H (1-0)S(1) line (2.12 µ m, see D07 for details). Although our H driving source survey is stillincomplete, we here identify out some interesting cases in the VMR-D cloud.The results of the correlation between H maps with our MIPS maps are given in Table 6;here we list each jet, the length and the dynamical time of the jet itself (having adoptedd=700 pc); the total flux detected at 2.12 µ m and the corresponding H luminosity; theidentification of the exciting source in our MIPS catalogue; the coincidence with a dust mm-peak, and finally an estimate of the source bolometric luminosity obtained by summing upall the contributions from the near-IR (if any) to the 1.2 mm flux, as derived by M07 SIMBAmap. We identify several different morphologies among our sources, including discoveringthe driving sources for three of the jets (namely Jet 1 -
The jet center lies towards a millimeter peak (MMS2), where no infrared sourceis detected down to K=17 mag. In the MIPS 24 µ m band, an emission peak is found,although not aligned with the jet axis. The lack of any aligned source suggests two possiblealternative scenarios: (i) we are observing just one jet lobe or (ii), more reasonably, theexciting source is too faint to be detected even by MIPS (F(24) < Jets 2 and 3 -
Two small jets have been detected that correspond to faint dust peaks.The exciting sources, although not detected in the infrared (NIR bands, IRAS, MSX), areclearly recognizable in the MIPS images and one of them ( shock = 50 km s − ) and inclination angle ( i = 45 ◦ ) areassumed (Figures 10 and 11). Jet 4 -
A parsec scale jet emerges from a young near-infrared cluster centered onIRAS08476-4306. The proposed exciting source, detected in the near-IR bands is the IRS20-
Jet 5 -
A point-like 24/70 µ m source aligned with the jet and corresponding to a dust 19 –peak (umms19) is found about 2 arcmin away towards the NE. If this source is indeed drivingthe jet then we are observing just one jet lobe, being the counter-jet located outside the H investigated field. A NIR cluster is also found at the MIPS source position (Figure 13). Jet 6 -
A chain of H knots emerges from a MIPS source (not visible in the H band)centered at the dust emission peak MMS16. We do not observe a counter jet (Figure 14). 20 –Fig. 9.— H contours of jet 1 (green) superposed on the MIPS 24 µ m (left) and 70 µ m(right) images. Dust contours (from a 3 σ level in steps of 3 σ ) are shown in yellow. Dustcore MMS2 is located at the jet center.Fig. 10.— The same as Fig.9 for jet 2. Peak umms16 (under-resolved at the SIMBA spatialresolution) is found near the jet center. The proposed exciting source is image.Fig. 14.— The same as Fig.9 for jet 6. The candidate exciting source is the 24 and 70 µ msource associated with mm peak MMS16 (
7. Conclusions
MIPS maps covering 1.8 square degrees across the Vela Molecular Cloud D at 24 and70 µ m are presented. The data allowed us to derive the following results:- A total of 849 and 61 point sources at 24 and 70 µ m, respectively, have been detectedat completeness limits of 5 and 250 mJy.- About half of the 24 µ m sources and two thirds of the 70 µ m ones are spatially locatedinside a region delimited by the CO contours (0.6 deg ). The implication is that theIR source density doubles (and is four times when considering sources at 70 µ m) insidethe CO contours as compared to outside the molecular cloud. A quantitative analysisof the 24 and 70 µ m counts per deg confirms this result.- The maps allow us to correlate MIPS sources with the distribution of the dust coresfound within VMR-D, and, when we extend the search of MIPS sources down to theinstrumental sensitivity limit, we find that most of these cores result associated withred and cold objects.- The MIPS sensitivity has enabled us to identify many new starless cores; the resultwill prompt a revision of the relative percentages of young objects known so far inVMR-D.- IRAS-PSC detections of good quality (f qual =3,2) are also seen by MIPS, but only whenthe IRAS point-likeness confidence is high (correlation coefficient, cc, equal to A or B).This result may be adopted as a broad confidence prescription for finding genuine pointsources in the IRAS catalogue.- About 400 MIPS sources have 2MASS K s counterparts. Color-magnitude plots con-structed with magnitudes at 2.2, 24 and 70 µ m in VMR-D show an excess of ClassI objects in comparison with other well studied star formation regions. This excesscould be biased by the sensitivity limits of the 2MASS and MIPS surveys, otherwise,it could reflect the short time elapsed since the first collapse of the cloud. From theMIPS colors, 6 objects appear as potential candidates Class 0 objects.- We have detected the driving source in five out of six H protostellar jets in VMR-D, four of them embedded in mm-cores. Such circumstance, along with the very lowdynamical time estimated for the jets, indicates ages of 10 -10 yr for these sources.- We note that, given the southern location of VMR-D, many of the newly detectedMIPS sources will be excellent candidate targets for ALMA. 24 –
8. Acknowledgements
This paper is based on observations made with the Spitzer Space Telescope, which isoperated by the Jet Propulsion Laboratory, California Institute of Techonology under acontract with NASA. HAS acknowledges partial support from NASA grant NAG5-10659.
REFERENCES
De Luca, M., Giannini, T., Lorenzetti, D. et al. 2007 A&A, in press (arXiv:0704.1228, D07)Djupvik, A. A., Andr`e, P., Bontemps, S. et al. 2006, 458, 789Dobashi,K., Uehara, H., Kandori, R. et al. 2005 PASJ, 57, S1Elia, D., Massi, F., Strafella, F. et al. 2007 ApJ, 655, 316 (E07)Evans N.J.,II, Allen, L.E., Blake, G.A. et al. 2003 PASP, 115, 965Fazio, G.G., Hora, J.L., Allen, L.E. et al. 2004 ApJS, 154, 10Giannini, T., Massi, F., Podio, L., et al. 2005 A&A, 433, 941Giannini, T., Nisini, B., Vitali, F. & Lorenzetti, D. 2001 A&A, 379, L17Greene, T.P., Wilking, B.A., Andr`e, P., Young, E.T., Lada, C.J. 1994 ApJ, 434,614Harvey, P.M., Rebull, L.M., Brooke, T. et al. 2007, ApJ, 663, 1139Liseau, R., Lorenzetti, D., Nisini, B. Spinoglio, L. & Moneti, A. 1992 A&A, 265, 577Lonsdale, C.J., Smith, H.E, Rowan-Robinson, M. et al. 2003 PASP, 115, 897Lorenzetti, D., Giannini, T., Vitali, F., Massi, F. & Nisini B. 2002 ApJ, 564, 839Lorenzetti, D., Spinoglio, L., & Liseau, R. 1993 A&A, 275, 489Makovoz,D., & Marleau, F. 2005 PASP, 117, 1113Massi, F., De Luca, M., Elia, D. et al. 2007 A&A, 466, 1013 (M07)Massi, F., Lorenzetti, D. & Giannini, T. 2003 A&A, 399, 147Massi, F., Lorenzetti, D., Giannini, T. & Vitali, F. 2000 A&A, 353, 598 25 –Massi, F., Giannini, T., Lorenzetti, D., Liseau, R., Moneti, A., Andreani, P. 1999 A&AS,136, 471 1025Murphy, D.C. & May, J. 1991 A&A, 247, 202Muzerolle, J., Megeath, S.T., Gutermuth, R.A. et al. 2004 ApJS, 154, 379Muzerolle, J., Young, E., Megeath, S.T., Allen, L. 2005, on
Star formation in the Era ofThree Great Observatories , conference abstracts, p.41Palla, F. & Stahler, S.W. 2000 ApJ, 540, 255Reach, W.T., Rho, J., Young, E. et al. 2004 ApJS, 154, 385Rebull, L.M., Stapelfeldt, K.R., Evans, N.J,II et al. 2007 ApJ, 171, 447 (R07)Rieke, G.H., Young, E.T., Engelbracht, C.W. et al. 2004 ApJS, 154, 24Schmeja, S., Klessen, R.S, & Froebrich D. 2005 A&A, 437, 911Shu, F.H., Adams, F.C. & Lizano, S. 1987 ARAA, 25, 23Werner, M.W., Roellig, T.L., Low, F.J. et al. 2004 ApJS, 154, 1Yamaguchi, N., Mizuno, A., Saito H. et al. 1999 PASJ, 51, 775Young, C.H., Jorgensen,J.K., Shirley, Y.L. et al. 2004 ApJS, 154, 396Young, K.E., Harvey, P.M, Brooke, T.Y. et al. 2005 ApJ, 628, 283
This preprint was prepared with the AAS L A TEX macros v5.2.
26 –Table 1. Statistics of MIPS point sources. item overall inside CO contours a outside CO contours b µ m 849 378 47170 µ m 61 41 2024 µ m & 70 µ m c
52 40 12only 70 µ m 9 1 824 µ m & 2MASS-K sd
401 180 22124 µ m & 70 µ m & 2MASS-K sd c
23 524 µ m & dust peak e
58 55 370 µ m & dust peak e
19 19 0 a size=0.61 deg ; b size=1.23deg ; c coordinates coincident within 20 ′′ ; d coordinates coin-cident within 5 ′′ ; e coordinates coincident within 24 ′′ (SIMBA HPBW). A total of 58 (19)sources at 24 (70) µ m result associated with 22 (resolved) and 20 (under-resolved) mm peaks,as listed by M07.Note that the area mapped in dust continuum at 1.2 mm is slighthy largerthan that covered with CO(1-0) observations (see Figs. 2, 3). Table 2: MIPS sources in the Vela Molecular Cloud-D.
Spitzer name α (2000.0) δ (2000.0) F ∆F F ∆F CO contours dust core † (h m s) ( ◦ ′ ′′ ) (mJy) (mJy) (mJy) (mJy)SSTVMRD J084310.3-440052.7 8 43 10.3 -44 00 52.7 4.5 0.2 - - N -SSTVMRD J084332.8-440301.0 8 43 32.8 -44 03 01.0 31.6 0.2 - - N -SSTVMRD J084335.4-435539.0 8 43 35.4 -43 55 39.0 8.2 0.1 - - N -SSTVMRD J084342.4-440134.5 8 43 42.4 -44 01 34.5 2.3 0.1 - - N -SSTVMRD J084347.4-435946.3 8 43 47.4 -43 59 46.3 42.0 0.1 - - N -[The complete version of this table is in the electronic edition of the Journal. The printed edition contains only a sample.]Notes to the table: † : following the nomenclature by M07 and E07, dust peaks are called MMS Table 3: MIPS sources detected at 70 µ m. ID † Spitzer name α (2000.0) δ (2000.0) (∆ α /∆ δ ) (F ± ∆F) (F ± ∆F) CO contours dust core †† (∆ α /∆ δ ) mm (h m s) ( ◦ ′ ′′ ) ( ′′ / ′′ ) (mJy) (mJy) ( ′′ / ′′ )1 a SSTVMRD J084409.5-440018.3 08 44 09.5 -44 00 18.3 · · · < ± ± ± ± ± ± ± ± ± ± ±
263 Y MMS1 0.9/27.27 c SSTVMRD J084535.5-435107.2 08 45 35.5 -43 51 07.2 d · · · > b ±
449 Y8 e SSTVMRD J084536.7-435155.4 08 45 36.7 -43 51 55.4 0.0/ 16.5 3097 ± ±
448 Y9 e SSTVMRD J084537.0-435134.0 08 45 37.0 -43 51 34.0 4.5/ 4.9 2740 ± ±
448 Y10 SSTVMRD J084541.1-435146.9 08 45 41.1 -43 51 46.9 3.0/ 2.4 1611 ±
69 36680 ±
331 Y MMS3 15.3/14.911 SSTVMRD J084544.0-432710.7 08 45 44.0 -43 27 10.7 1.5/0.2 405.6 ± ± ± ± ± ±
34 N14 a SSTVMRD J084624.2-433415.6 08 46 24.2 -43 34 15.6 · · · diffuse 217 ± e SSTVMRD J084626.4-434247.5 08 46 26.4 -43 42 47.5 18.0/ 3.2 245.6 ± ± ± ± e SSTVMRD J084627.3-434239.5 08 46 27.3 -43 42 39.5 4.5/ 4.8 234.6 ± ± ±
10 16790 ±
91 Y19 e SSTVMRD J084631.6-435532.2 08 46 31.6 -43 55 32.2 7.5/ 7.2 367 ± ±
68 Y20 e SSTVMRD J084633.1-435539.6 08 46 33.1 -43 55 39.6 15.0/ 0.2 349 ±
11 14939 ±
68 Y21 SSTVMRD J084634.3-432115.1 08 46 34.3 -43 21 15.1 3.0/ 0.4 18.9 ± ± f SSTVMRD J084634.9-435437.1 08 46 34.9 -43 54 37.1 d · · · > b ±
173 Y MMS4 5.1/1.123 SSTVMRD J084637.4-435217.0 08 46 37.4 -43 52 17.0 1.5/ 6.7 4292 ± ±
45 Y24 SSTVMRD J084637.5-435357.1 08 46 37.5 -43 53 57.1 6.0/ 0.8 4758 ± ±
180 Y25 e SSTVMRD J084639.2-435254.1 08 46 39.2 -43 52 54.1 3.0/ 12.4 341 ± ±
110 Y26 e SSTVMRD J084639.5-435314.1 08 46 39.5 -43 53 14.1 7.5/ 7.6 215 ± ±
110 Y27 SSTVMRD J084712.1-432250.6 08 47 12.1 -43 22 50.6 1.5/ 1.9 102.1 ± ± ± ± ± ± ± ± ± ±
10 Y32 a SSTVMRD J084738.0-434255.3 08 47 38.0 -43 42 55.3 · · · diffuse 1653 ±
16 Y33 SSTVMRD J084742.8-434352.4 08 47 42.8 -43 43 52.4 13.5/ 1.7 1533.5 ± ±
12 Y umms11 4.8/13.4
Table 3: MIPS sources detected at 70 µ m ( continued). ID † Spitzer name α (2000.0) δ (2000.0) (∆ α /∆ δ ) (F ± ∆F) (F ± ∆F) CO contours dust core †† (∆ α /∆ δ ) mm (h m s) ( ◦ ′ ′′ ) ( ′′ / ′′ ) (mJy) (mJy) ( ′′ / ′′ )34 SSTVMRD J084748.4-432536.4 08 47 48.4 -43 25 36.4 9.0/ 5.9 142 ± ±
11 Y umms12 24/1935 SSTVMRD J084751.7-432523.4 08 47 51.7 -43 25 23.4 10.5/13.5 137 ± ± ± ± ± ± e SSTVMRD J084815.8-434715.8 08 48 15.8 -43 47 15.8 1.5/ 0.1 32 ± ± e SSTVMRD J084816.7-434719.4 08 48 16.7 -43 47 19.4 12.0/6.4 27 ± ± ± ± a,g SSTVMRD J084828.0-423630.8 08 48 28.0 -42 36 30.8 · · · · · · ±
49 N42 a,g
SSTVMRD J084829.3-423733.2 08 48 29.3 -42 37 33.2 · · · · · · ±
46 N43 a,g
SSTVMRD J084830.6-423559.9 08 48 30.6 -42 35 59.9 · · · · · · ±
63 N44 SSTVMRD J084834.0-433051.3 08 48 34.0 -43 30 51.3 1.5/ 2.9 228 ± ±
17 Y umms19/20 13.6/7.4;22.1/8.545 SSTVMRD J084841.6-433149.8 08 48 41.6 -43 31 49.8 1.5/ 2.0 116 ± ± ± ±
10 Y47 SSTVMRD J084846.4-425055.6 08 48 46.4 -42 50 55.6 12.0 13.6 45 ± ±
10 N48 h SSTVMRD J084848.2-425420.2 08 48 48.2 -42 54 20.2 d · · · > b ±
270 N49 i SSTVMRD J084848.7-433230.7 08 48 48.7 -43 32 30.7 d · · · > b ±
43 Y MMS12 3.9/2.750 SSTVMRD J084853.2-433057.1 08 48 53.2 -43 30 57.1 0.0/ 2.6 1163 ± ±
30 Y MMS16 1.8/1.051 SSTVMRD J084858.8-433825.1 08 48 58.8 -43 38 25.1 10.5/ 2.0 157 ± ±
18 Y MMS17 14.4/2.952 SSTVMRD J084904.3-433805.0 08 49 04.3 -43 38 05.0 3.0/ 4.9 158 ± ±
28 Y MMS18 10.5/7.153 SSTVMRD J084912.2-441636.5 08 49 12.2 -44 16 36.5 4.5/ 4.4 28 ± ±
10 N54 a SSTVMRD J084912.5-432953.3 08 49 12.5 -43 29 53.3 · · · diffuse 650 ±
10 N55 SSTVMRD J084913.1-433628.5 08 49 13.1 -43 36 28.5 0.0/ 0.9 685 ± ±
20 Y MMS21 1.5/0.456 SSTVMRD J084914.3-430019.8 08 49 14.3 -43 00 19.8 10.5/0.1 153.5 ± ± ± ±
19 Y58 SSTVMRD J084917.0-435600.3 08 49 17.0 -43 56 00.3 3.0/ 3.5 19.5 ± ± ± ± j SSTVMRD J084926.2-431710.2 08 49 26.2 -43 17 10.2 d · · · > b ±
86 Y MMS22 17.4/2.061 SSTVMRD J084928.6-440429.2 08 49 28.6 -44 04 29.2 0.0/ 0.0 1.9 ± ±
10 Y MMS23 13.8/6.962 k SSTVMRD J084932.8-441050.0 08 49 32.8 -44 10 50.0 d · · · > b ±
93 Y MMS26 15.6/9.963 SSTVMRD J084936.1-441200.2 08 49 36.1 -44 12 0.2 7.5/ 2.5 1420 ± ±
84 Y MMS27 23.1/4.264 SSTVMRD J084959.5-432300.7 08 49 59.5 -43 23 0.7 0.0/ 0.9 97 ± ±
13 Y umms26 6.9/5.165 a SSTVMRD J085038.9-434948.8 08 50 38.9 -43 49 48.8 · · · diffuse 356 ± ± ± a,g SSTVMRD J085149.4-430540.2 08 51 49.4 -43 05 40.2 · · · · · · ±
33 NNotes to the table: † : this ID is used for simplicity throughout the paper †† :following the nomenclature by M07 and E07, dust peaks are called MMS a :detected only at 70 µ m; b saturated; c IRS16 in the Liseau et al. (1992) list . ; d µ m coordinate; e associated with the same 70 µ m source; f IRS17; g source outside the 24 µ m map; h,i,j,k IRS18, IRS19, IRS20, IRS21 in the Liseau et al. (1992) list, respectively.
29 –Table 4. IRAS PSC detections not recovered by MIPS
PSC name MIPS 24 µ m IRAS 25 µ m MIPS 70 µ m IRAS 60 µ mfqual a CC b fqual CC08441-4357 diffuse 1 null missing 3 E08475-4255 diffuse 3 D diffuse 2 D08475-4311 diffuse 3 E diffuse 2 C08478-4303 intense knot 3 A diffuse 1 null08479-4311 diffuse 3 D diffuse 1 L08487-4250 diffuse (map edge) 1 C diffuse 3 B08459-4338 diffuse 2 D missing 3 C08489-4241 off edge 3 C diffuse 1 D08457-4229 off edge 1 E diffuse 2 D08460-4223 off edge 2 D intense knot (map edge) 2 C08462-4235 diffuse (map edge) 1 J diffuse? 3 B08465-4230 off edge 2 B diffuse? (map edge) 3 C08471-4228 off edge 3 G intense knot (map edge) 2 D08473-4235 off edge 3 E diffuse (map edge) 1 null08437-4323 diffuse 1 D missing 3 B08468-4330 diffuse 3 D missing 2 D08488-4308 diffuse 2 C diffuse 2 E08490-4319 diffuse 1 F missing 3 C08491-4310 diffuse 3 B diffuse 2 C08491-4257 diffuse? 1 E missing 3 B08493-4331 diffuse 1 H missing 3 C08495-4306 diffuse 1 C diffuse 3 C08477-4329 diffuse 1 H diffuse 3 D08462-4400 diffuse 2 D diffuse 1 D08463-4343 diffuse 2 B diffuse 3 D08478-4403 diffuse 3 C off edge 1 J08478-4353 diffuse 1 I diffuse 3 D a Flux density quality, encoded as 3: high quality, 2: moderate quality, 1: upper limit; b pointsource correlation coefficient encoded as alphabetic character (A=100%, B=99%, ....N=87%). Table 5. Classification of the MIPS sources. item inside CO map outside CO mapphotospheres 42(23%) 124(56%)Class III 5 (3%) 11 (5%)Class II 51(28%) 44 (20%)flat spectrum 40(22%) 31 (14%)Class I 42(23%) 11 (5%)Class 0 6 1? a Starless cores 5 0 a this source is observed only at 70 µ m, so that justupper limits appear in the 24-[24-70] plot.
30 –Table 6. Protostellar jets associated with MIPS sources in VMR-D.
Jet length T dyna F − S (1) L H b Exc. source mm peak c L bol yr) (erg s − cm − ) (10 − L ⊙ ) ⊙ )1 0.30 4.3 2.8 10 −
12 - MMS 2 -2 0.13 1.7 1.5 10 − − −
73 60 MMS 22 > d − d − a computed for i =45 ◦ and v shock =50 km s − ; b A V =10 mag and L(H )=10 × L(2.12 µ m) areassumed; c names from M07; dd