EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 542 (June 2012) A&A, 542 (2012) A66 Full HTML
Free Access
Issue
A&A
Volume 542, June 2012
Article Number A66
Number of page(s) 22
Section Galactic structure, stellar clusters and populations
DOI https://doi.org/10.1051/0004-6361/201118627
Published online 08 June 2012

© ESO, 2012

1. Introduction

The nearby Large Magellanic Cloud (LMC) is an excellent laboratory for the study of the life-cycle of matter from relatively quiescent interstellar medium (ISM) through star formation, main sequence (MS) stellar existence, and eventual feedback of material as stars evolve and die. It is well situated for observations, far from the confusion and the extinction of the Galactic plane and near enough for high-resolution observations. At a distance of  ~50 kpc (e.g., Panagia 2005; Schaefer 2008), the resolution is  ~14.5 pc/arcmin. The metallicity falls within the range Z = 0.3−0.5   Z⊙ (cf. Dufour et al. 1982; Westerlund 1997), similar to that of galaxies at redshift z ~ 1.5 (Pei et al. 1999), a time of peak galaxy formation. Morphologically, the LMC has an evident bar structure and a disk component, seen from a relatively face-on viewing-angle (~24−35°; Weinberg & Nikolaev 2001; van der Marel & Cioni 2001) such that the distance to each source is effectively identical.

Lists of young stellar object (YSO) candidates covering the entire 7°  ×  7° area imaged by the Spitzer Space Telescope (Spitzer) legacy program “Surveying the Agents of Galaxy Evolution” (SAGE; Meixner et al. 2006) have been created by Whitney et al. (2008) and Gruendl & Chu (2009), hereafter W08 and GC09, using different strategies. These initial surveys focus on the most massive and embedded YSOs which are the easiest to identify and suffer the least contamination in terms of color selection. The study of W08 employs a set of selection criteria which are simple in philosophy but complex in implementation. They use several different color − magnitude diagrams (CMDs) to select as many YSO candidates as possible while minimizing the selection of background galaxies and evolved stars. Selection is guided by YSO-model color − magnitude space contrasted with that of observed “galaxies” and evolved star models. They then rely upon spectral energy distribution (SED) fitting to determine the likelihood that a given source is a YSO and remove some galaxies identified by eye in Spitzer InfraRed Array Camera (IRAC; Fazio et al. 2004) or 24 μm Multiwavelength Imaging Photometer for Spitzer (MIPS; Rieke et al. 2004) images. This approach is rigorous and uniform but conservative due to the use of the strictly-defined SAGE Epoch 1 Point Source Catalog (Meixner et al. 2006)1 of point spread function (PSF) fit sources with high signal-to-noise ratios and the use of conservative color − magnitude cuts. GC09 use one, simple color selection (which excludes candidates less massive than  ~4   M⊙ and all sources without measured flux in either 4.5 or 8.0 μm) and their own less-stringent catalog of aperture photometry. From the color-selected list, they then identify probable YSOs via visual inspection of SEDs and images. The two catalogs complement one another, with the  ~1000 W08 sources including many fainter sources that do not fall within the GC09 color-selection, and the 1172 GC09 candidates including more extended sources and sources in more complicated environments. Both of these initial surveys are, however, incomplete in at least two ways. They miss less embedded massive YSOs (e.g., Stage ii and Stage iii) and low luminosity (lower mass) YSOs.

We present a streamlined set of YSO candidate selection criteria, using five color − magnitude combinations which include all four IRAC bands and MIPS 24 μm. These data are described in Sect. 2. After initial color selection (Sect. 3.1), we fit SEDs to estimate physical parameters, using the model grid and fitter of Robitaille et al. (2006, 2007). We require better fits for sources fulfilling fewer color criteria, as described in Sect. 3.2. We apply our selection criteria to optical through Mid-infrared (Mid-IR) photometry for nine Hii regions in the LMC: LHA 120-N 11, LHA 120-N 44, LHA 120-N 51, LHA 120-N 105, LHA 120-N 112, LHA 120-N 120, LHA 120-N 144, LHA 120-N 160, and LHA 120-N 206 (hereafter N11, etc.; Henize 1956), detailed in Sect. 4. We show the efficacy of our new YSO candidate selection scheme, provide a comparison to other YSO candidate lists, and examine discrepancies in Sect. 5.1. In Sect. 5.2, we provide star-formation rate (SFR) estimates for each region. Section 6 encompasses some final remarks.

2. Data

Our data are from SAGE in the Spitzer/IRAC 3.6, 4.5, 5.8, and 8.0 μm bands and MIPS 24 μm which we use for color-selection. We then fit the SEDs to select the strongest YSO candidates. We add photometry in the near-IR bands J, H, and Ks from the Magellanic Clouds Point Source Catalog created with the Simultaneous 3-color IR Imager for Unbiased Surveys (SIRIUS) aboard the InfraRed Survey Facility (IRSF) by Kato et al. (2007). We convert these to the Two-Micron All Sky Survey (2MASS; Skrutskie et al. 2006) photometric system for compatibility with the SED fitter, following the discussion in Kato et al. (2007). We use optical photometry from the Magellanic Clouds Photometric Survey (MCPS; Zaritsky et al. 1997) in the U, B, V, and I bands. SAGE data processing is described by Meixner et al. (2006), and public catalogs are available through the Spitzer Science Center2. The IRAC angular resolution is  ~2″. Photometric completeness is expected to be  >99% down to 16, 16, 14, and 13.5 mag in [3.6], [4.5], [5.8], and [8.0], respectively, and 50% down to 18, 17, 15.5, and 15 mag. We provide photometry for all YSO candidates in our 9 regions in Tables 1 and 2, with mosaicked IRAC photometry matched to MCPS, IRSF, and MIPS 24 μm within 1″.

Table 1

Flux table.

Table 2

Magnitude table.

Table 3

Matches table.

3. Identification of YSOs

We present a simplified procedure (compared to that of W08) for the color-selection of YSO candidates. We use only five infrared color − magnitude combinations, making sure to use Spitzer/IRAC 3.6, 4.5, 5.8, and 8.0 μm bands and MIPS 24 μm. In each CMD, we select an area in which there is a high density of model YSOs (from Robitaille et al. 2006) and a low density of other types of sources; these are our α color selections and are described in detail in Sewiło et al. (in prep.). A secondary (β) set of criteria complement the α selections, sampling CMD space with more “contamination” from galaxies and evolved stars but a high density of YSO models, particularly those of low mass. With the high degree of contamination, we consider β criteria appropriate only for known star-forming regions with high background emission, where background galaxies are largely blocked by the ISM, and YSOs are more abundant than asymptotic giant branch stars (AGBs). We apply the SED fitter of Robitaille et al. (2007) to the color-selected sources and select our final list of “well-fit” YSO candidates according to the number of color-criteria met and the quantitative goodness (χ2/pt) of the best fits.

3.1. Color selection

Our YSO candidate color selections are plotted in Figs. 1 and 2 and described by Eqs. (3)−(10). The α criteria are appropriate for galaxy-wide surveys; we expect very low contamination from non-YSO sources. We designed these α criteria for the Small Magellanic Cloud (Sewiło et al., in prep.) and scale here for the LMC distance. These yield YSO candidate lists similar to those of the other surveys, identifying only luminous, high-mass YSO candidates. The β set of criteria is to be used in star-forming regions where color-selected sources are significantly more likely to be YSOs than background galaxies or evolved stars, choosing color − magnitude space in which we still expect most sources to be YSOs but in which (galaxy-wide) there is significant contamination from non-YSO sources. Our β-selections are appropriate only in these dusty, star-forming regions in which YSO candidates are concentrated along dust lanes, where we can reasonably expect only a small percentage of contaminants.

Table 4

LMC “well-fit” YSO candidate fitting requirements.

Figures 1 and 2 shows the five CMDs with α and β selection criteria indicated. The full list of LMC photometric sources is shown in a density plot in the background in Fig. 1, and the Robitaille et al. (2006) YSO models are the background in Fig. 2. Our selections avoid the primary “vertical” population of MS and giant stars. YSO models from Robitaille et al. (2006) have little overlap with this “vertical” population. YSO models do overlap significantly with the color − magnitude locations of evolved sources (candidate AGB stars from Srinivasan et al. 2009, are marked in Fig. 1) and galaxy candidates (here “blank” fields from the edges of the SAGE images, as in W08, also in Fig. 1). We mark previously identified LMC YSO candidates (“List YSO Candidates” from W08 and GC09) for the full LMC in Fig. 1 and for the examined regions only in Fig. 2, noting their distinct CMD locations, the result of using different color-criteria. Our final list of Well-Fit YSO Candidates from these nine star-forming regions is also plotted in Fig. 2. Planetary nebulae from the Reid & Parker (2006) Hα study, identified by Hora et al. (2008) in the SAGE data, do contaminate the full color − magnitude space and are unavoidable in IR color-selection. However, like AGB stars, planetary nebulae are not expected to be abundant in young star formation regions.

thumbnail Fig. 1

CMDs showing α (red lines) and β (blue lines) selection criteria on full LMC data. To the right (redward) of the red line is the α section of the CMD, where we expect to find massive YSOs and little contamination from non-YSO sources. The blue lines mark the boundary of the β-selection, in which we expect more contamination, particularly from background galaxies. List YSO candidates plotted are the full lists of W08 and GC09.

thumbnail Fig. 2

CMDs showing α (red lines) and β (blue lines) selection criteria with YSO candidates from these regions. The background is from the YSO models of Robitaille et al. (2006). List YSO candidates are from W08, GC09, Chen et al. (2009), Romita et al. (2010), and Chen et al. (2010) for these regions only. Well-fit YSO candidates are our final list of well-fit sources from this paper.

(1)(2)(3)(4)(5)(6)(7)(8)(9)(10)}

thumbnail Fig. 3

Example SED for each selection Set detailed in Table 4. Error bars on fluxes are shown but usually less than the size of the dot. The best fit is shown in black with other fits with shown in grey. The dashed line shows the photospheric emission from the central source. These are the same fits used to determine approximate physical and evolutionary properties. SEDs shown in a), c), d), and e) are classified as Stage i, i) as Stage i/ii, f) and h) as Stage ii, and b) and g) as Stage iii. The characteristic dip in SED c) at 4.5 μm, compared to the other three IRAC bands, indicates the presence of strong PAH emission, and error bars have been increased in 3.6, 5.8, and 8.0 μm, as described in the text.

3.2. YSO candidate selection and classification

We use the YSO model grid and fitter of Robitaille et al. (2006, 2007) to determine which sources are well-fit by YSO models. The grid incorporates 20 000 model YSOs with a wide range of physical parameters, each with SEDs generated for 10 viewing angles, for a total of 200 000 SEDs. The fitter allows us to specify included photometric bands and applied apertures. The fitter calculates χ2 and χ2/pt to estimate the goodness of fits, and for a source to be considered well-fit, we require  as listed in Table 4. We also discard sources best fit by AGB, galaxy, or MS templates.

We apply the SED fitter to 9 Sets of sources, according to how many selection criteria are met. Requirements for each Set are shown in Table 4. Sources falling within at least one α-selection area are considered quite reliable. The brightest, most massive sources are extended or saturated in multiple bands, and we do not expect them to appear in all of the selection CMDs. For fitting, we require sources meeting at least 1α criterion to have at least 4 bands of photometry defining the SED and consider a relatively large  well-fit (i.e. reduced χ2; see Carlson et al. 2011, for a detailed explanation). We consider sources with at least 1α selection criterion and falling within either an α or β selection region in all five CMDs to be the most reliable. Given the higher likelihood of contamination from non-YSO sources in our β selections, we require better quality SED fits for β-selected sources. In order to be considered in our initial color-selected list, we also require that a source meets at least 2β criteria if it meets no α criteria. For these, we require 5 bands of photometry for fitting and smaller  values to be called well-fit. Example SEDs are shown in Fig. 3.

We use the calculated error flux bars in SED fitting, typically on the order of 10%, but caveats should be noted. First, at the distance of the LMC all of our effective apertures are larger than the largest apertures calculated for the model YSOs from Robitaille et al. (2007). Also, the photometric apertures differ between the surveys, so each is encompassing a slightly different area. We match sources in the various catalogs to IRAC sources within a search radius of 1″ and do expect a some false matches; visual inspection of automated matches indicates  ~2% contamination by false matches to IRAC. The conversion from IRSF to 2MASS magnitudes introduces an additional error, but this is estimated included in the conversion as given by Kato et al. (2007). Studies such as Carlson et al. (2011) and Chen et al. (2009) have shown that the large majority (~70% or more) of IRAC YSO candidates at these distances are in fact (IR) dominant members of (optical) protoclusters. The YSO model set does not include multiple sources. However, we showed in Carlson et al. (2011) that the SED fitter can be used in the identification of YSOs even when they are members of protoclusters. This is in part because we expect one or two sources to dominate the luminosity of a protocluster: LYSO  ∝   (i.e., Bernasconi & Maeder 1996), though different sources may dominate in different wavelength regimes. Estimating how much each of these factors might contribute to the overall error budget for SED fitting is difficult. Our χ2 / pt values for sources to be considered well-fit (Table 4) were chosen empirically with a knowledge of the data and of these issues.

Polycyclic-aromatic hydrocarbons (PAHs), considered star formation indicators, can contribute significantly to the measured fluxes in the 3.6, 5.8, and 8.0 μm bands but are not yet incorporated into the YSO models (Robitaille et al. 2006). We are able to estimate their potential contribution to SEDs. The [4.5] IRAC band is unaffected by PAHs, such that sources with significant PAH emission display a characteristic dip in their SEDs around 4.5 μm. In Carlson et al. (2011), we found that these sources fell within the color range ([3.6]  −  [4.5]) < 0.5 and 1.5 < ([4.5]  −  [5.8]). For sources meeting these color-criteria, we adjust our treatment of IRAC fluxes in SED fitting. Where no 24 μm measurement is available, we increase flux error bars from the  ~10% calculated photometric errors to 20%, 30%, and 40% in 3.6, 5.8, and 8.0 μm, respectively. Where a 24 μm measurement is available, we increase the error bar on 3.6 μm and set 5.8 μm and 8.0 μm fluxes as upper-limits. Notice that PAH contributions will move sources redward in and brighter in [8.0] v. [4.5] − [8.0] and in [3.6] v. [3.6] − [5.8] and redward in [4.5] v. [4.5] − [5.8]. We expect no effect on [4.5] v. [4.5] − [24]. Only in [8.0] v. [8.0] − [24] do we expect PAH sources to appear more blue. Indeed, many of our final well-fit and list YSO candidates lie redward of the color − magnitude area covered by the YSO models in Figs. 2a and c.

We classify our well-fit YSO candidates into the evolutionary Stages i − iii defined by Robitaille et al. (2006), based on the fractional disk mass and the envelope accretion rate:

  • 1.

    Stage i: embedded Source (Ṁenv > 10-6   M⋆ yr-1);

  • 2.

    Stage ii: source with Disk (Ṁenv < 10-6   M⋆ yr-1 and Mdisk / M⋆ > 10-6);

  • 3.

    Stage iii: source with Optically Thin Disk or No Disk (Ṁenv < 10-6   M⋆ yr-1 and Mdisk / M⋆ < 10-6).

Stage i models are the least evolved, most embedded sources with significant ongoing accretion from their circumstellar envelopes. Stage ii sources are surrounded by optically thick disks, and Stage iii sources have optically thin disks or no disks. We list sources that can be well-fit by both Stage i and Stage ii models as Stage i/ii and sources fit by both Stage ii and Stage iii as Stage ii/iii. Some sources are well-fit by models covering all three stages; we consider these “Unclassified”. This study is grounded in deep mid-IR (IRAC) data, and we are most sensitive to the more embedded Stage i sources. This is also likely a physical effect in part resulting from faster evolution of later stages of protostellar evolution (e.g., W08 and Carlson et al. 2011). Evolutionary stages as used here are reasonably well-correlated with the Classes i, ii, and iii of Lada (1987) and Types i, ii, and iii from Chen et al. (2009).

4. Application to nine star-forming regions

We apply the YSO selection criteria described in Sect. 3 to LMC star formation regions N11, N44, N51, N105, N113, N120, N144, N160, and N206. These nine regions are selected to cover a range of sizes, morphologies, and levels of star formation activity. Figure 4 shows their distribution within the LMC and with the exact areas we study. Of more than 66 000 SAGE sources within these regions,  ~5300 meet our color-selection criteria for sets 1−9, and we ultimately consider 1045 to be well-fit YSO candidates.

thumbnail Fig. 4

Examined star-forming regions on MCELS Hα (blue) + SAGE 8 μm (green) + 24 μm (red). Hα image from the Magellanic Cloud Emission Line Survey (MCELS; Smith & MCELS Team 1998; Smith et al. 2005).

We report 918 entirely new candidate YSOs here. We consider 1045 sources to be well-fit YSOs, 127 of which are also selected by W08, GC09, Chen et al. (2009), Romita et al. (2010), Chen et al. (2010), Fleener et al. (2010), or some combination. For completeness, our images and photometric catalog include another 358 YSO candidates from those lists, most of which do meet our color-selection criteria but are not well fit by YSO models. We do assign approximate evolutionary stages from the fitter for use in images; we make no cuts on the basis of fit quality and make no claims about their physical characteristics. Combining our list of well-fit sources with the candidate YSOs from the other 5 lists, we consider 1403 sources. A detailed comparison of the source lists is presented in Sect. 5.1 and Table 7. Visually, we find that our new candidates align well with previously identified candidates, spectroscopically “confirmed” YSOs, and the physical structure of the ISM.

We determine approximate physical parameters for each “well-fit” source, averaging parameters of all YSO models with SEDs in the range . Resulting parameters are listed in Table 5. For each source, we provide the range of χ2 / pt values used, the number of models fit within that range, the derived evolutionary stage, and the physical parameters of the YSO models: stellar mass, stellar luminosity, envelope accretion rate, disk mass, and envelope mass.

Table 5

Physical parameters of “well-fit” YSO candidates.

Table 6 shows the breakdown of the evolutionary stages of our well-fit sources in each region. Approximately 56% of all of our YSOs are classified as Stage i, 12% as Stage i/ii, 23% as Stage ii, and only  ~9% as Stage ii/iii, Stage iii, or Unclassified. This is at least partially an effect of our sensitivity. Stage i sources are brightest in the mid-IR. Interestingly, 81% of N160 sources are Stage i or Stage i/ii, compared to just over 70% in N105 and 62 − 66% in the other 7 regions.

4.1. N11

Located in the north-west corner of the LMC, N11 is one of the most well-studied star-forming regions in the LMC. The nebula is second in size and Hα luminosity only to the famous mini-starburst 30 Doradus on the other side of the galaxy (Walborn & Parker 1992). It displays a central cavity, relatively devoid of IR emission, surrounded by a complex structure of Hii regions of radius  ~100 pc, described as a “ring of Hii regions” by Rosado et al. (1996). The young stellar association LH 9 (Lucke & Hodge 1970) has cleared the central cavity with winds of its massive stars, creating diffuse soft X-ray emission (Nazé et al. 2004), and triggering star formation around the bordering regions. In addition to star formation immediately around the central bubble (notably, abundant Herbig Ae/Be stars encircling the LH 9 cavity, following the Hα ridge; Hatano et al. 2006, from IRSF observations), star formation is propagating outward through the surrounding nebula. N11B is one of the secondary nebulae (the 24 μm-bright region on the north-east edge of the central cavity in Fig. 5) and contains the embedded stellar association LH 10 and the star-formation signpost of a 6.7 GHz methanol maser MC18 (Ellingsen et al. 1994; Barbá et al. 2003; Green et al. 2008; Ellingsen et al. 2010), which is also a spectroscopically confirmed YSO (Seale et al. 2009). Three X-ray point sources and excess X-ray emission, probably indicating the presence of young T-Tauri stars (Nazé et al. 2004). Vallenari et al. (2010) find further evidence of sequential star formation from the distribution of their YSO candidates. Using the same “Types” as Chen et al. (2009), they find that Type II are preferentially close to LH 10, and Type I are even farther from LH 9.

Although age estimates vary, several authors report a marked difference in the ages of LH 9 and LH 10, suggesting that the LH 9 has triggered the current star formation in LH 10 and in other surrounding nebular bubbles. Walborn et al. (1999) estimate the triggering difference to be δt  ~  2 Myr and estimate ages of 3.5 and 1 Myr for LH 9 and LH 10, respectively, based on HST optical imaging. Massey et al. (2000) examine the most massive stars both photometrically and spectroscopically and report that all stars in LH 9 with masses greater than 20 M⊙ are approximately coeval with an age of  ~4 Myr and a turn-off mass of  ~45 M⊙. Age estimates from Mokiem et al. (2007) are significantly older at 7  ±  1 Myr and 3  ±  1 Myr but still indicate sequential formation. Vallenari et al. (2010) examine both HST/ACS and Spitzer data to identify a significant optical PMS population and 30 color-selected YSO candidates with HST PMS stars and protoclusters physically mixed. They find that the PMS stars intermixed with MS stars in the central cavity, with LH 9 populations in the 3 − 10 Myr age range and the main PMS population  ~5 Myr old. They estimate the LH 10 association age as 3  ±  1 Myr. Heydari-Malayeri et al. (2000) and Heydari-Malayeri et al. (2001) find similar results based on spectroscopy of blue stars in N11B.

We consider a circular area of radius (nearly 300 pc or an area of 0.28 kpc2 at the distance of the LMC), centered at 04h56m48s,  − 66°24′00″. Seventy-four candidate YSOs were identified within this area by W08 and/or GC09. Seale et al. (2009) confirmed the YSO nature of 32 sources using Spitzer/IRS spectroscopy, and Woods et al. (2011) confirmed one more. Of the Seale et al. (2009) sources, thirty-one were chosen from the GC09 list, and the last corresponds to the methanol maser. We identify 242 well-fit YSO candidates in the region, 218 of which are on no previous list. We find that star formation is continuing throughout the region with particular concentration along filaments and near the edges of photodissociation regions.

4.2. N44

Along with 30 Doradus and N11, N44 is one of the most active star-forming regions in the LMC, injecting vast amounts of energy into the ISM. Located in north of the LMC bar, N44 is characterized by interacting shells of star formation. The most prominent feature is Shell 1 (Chen et al. 2009), the central bubble in our Fig. 6, which encircles the LH 47 cluster of massive, ionizing stars (Stasińska et al. 1986; Meaburn & Laspias 1991). Will et al. (1997) estimate the age of LH 47 at 5 − 6 Myr, based on a combination of photometry and spectroscopy. Oey & Massey (1995) note two populations, one  ≳ 10 Myr inside Shell 1 and one outside, resulting from a burst of star formation in the last  ~5 Myr. Shell 1 is expanding  ~  spherically (Meaburn & Laspias 1991; Chu et al. 1993) with a dynamical age greater than 1 Myr, evidently triggering star formation along the edges, as can be seen in Fig. 6. Hi maps suggest similar scenarios for the other Hi shells in the area; compression from interacting shells expanding into one another is triggering star formation (Kim et al. 1998). Massey et al. (2000) classify one Nitrogen rich W-R star in LH 47 and estimate an age of  ~2.95 Myr for all LH 47 stars with M > 20   M⊙ and an age of  ~2.14 Myr for the 3 most massive stars. Exterior to (i.e., around the rim of) the superbubble, they estimate a turn-off mass of  ~55   M⊙.

Chen et al. (2009) present a detailed study of the massive YSO population in N44 and note evidence of triggered star formation. They identify 99 candidate YSOs, using the same color-selection criteria as GC09, then narrow their list to 60 probable YSOs via visual inspection of SEDs and imaging. They use deep IRAC and MIPS imaging and add NIR and optical imaging and photometry. MCELS Hα and sundry optical archival HST images are used for image examination. They fit SEDs of 36 YSO candidates judged to be unambiguously single or dominant sources. Using the grid of Robitaille et al. (2006) as a guide, they then classify each of these sources into Types I − III by visual examination rather than relying upon the physical characteristics suggested by the YSO model grid. They also report the range of Robitaille et al. (2006) stages found and find that stage and type are equivalent for most sources.

The area we study, a circle of radius centered at RA = 05h22m07s, Dec =  −67°56′24″ and covering 0.11 kpc2 in the LMC, includes 32 spectroscopically confirmed YSOs from Seale et al. (2009), many occurring in clumps with several photometric YSO candidates. Star formation is highly concentrated in shell-like lanes or ridges. We find a total of 139 well-fit YSO candidates, 24 of which appear in at least one previous list. Figure 6 shows the distribution of candidates, including 74 previously identified by W08, GC09, and/or Chen et al. (2009) and 115 new candidates.

thumbnail Fig. 5

Three-color image of N11 with YSO candidates marked. Full region in IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked in red for Stage i, magenta for Stage i/ii, yellow for Stage ii, green for Stage ii/iii, cyan for Stage iii, and blue for unclassified. Circles are newly identified candidates. Crosses are from W08 and/or GC09. Diamonds indicate sources on our list that are also in another list, and blue Xs are spectroscopically confirmed YSOs from Seale et al. (2009) or Woods et al. (2011).

Table 6

LMC YSO candidate evolutionary stages.

Table 7

“Well-fit” YSO candidates compared to YSO candidate lists from W08, GC09, Chen et al. (2009, 2010), and Romita et al. (2010).

4.3. N51

In the northern disk, the N51 star-forming region is characterized by lanes of active star formation along the boundaries between expanding superbubbles. Superbubble N51D surrounds OB associations LH51 and LH54 (Lucke & Hodge 1970; Oey & Smedley 1998). Meaburn & Terrett (1980) use [O ii] emission line profiles to show that N51D is expanding at  ~35   km   s-1 toward the observer and at  ~25   km   s-1 away from the observer, possibly into a denser medium. Expansion is likely driven by a supernova remnant or possibly energetic winds of enclosed massive stars. Chu et al. (2005) identify 3 YSO candidates in N51D. HST emission line images show that 2 of the YSO candidates are associated with Bok globules, and the morphology and the Hα/[Sii] ratio of the third suggests that it may be a Herbig-Haro object. From the models of Whitney et al. (2003a), Chu et al. (2005) estimate all three sources to have masses in the B star range. Breysacher et al. (1999) report HD 36502 in LH54 as a binary or trinary system WC4(+O)+O8I (spectroscopy from Moffat et al. 1990), again supporting the young age of the association.

The area considered here is a circle of radius  (167.5 pc; area 0.087 kpc2) centered at RA = 05h26m53s, Dec =  −67°28′48″. There are 38 YSO candidates from W08 and/or GC09, including seven spectroscopically confirmed by Seale et al. (2009). We consider seven of these to be well fit and identify 96 new candidates tracing the same ridges of star formation.

4.4. N105

N105 is an average-sized Hii region, situated in a quiescent part of the LMC at the western side of the bar. The molecular cloud is still being evaporated, creating an inhomogeneous region of dust globules and pillars surrounded by hot gas. At the center of N105 sits the OB association LH 31 (corresponding to the strongest X-ray emission Dunne et al. 2001), which includes 2 W-R stars. Massey et al. (2000) find that the young (optical) stellar population is not co-eval; the 3 most massive are the youngest (~1.5 Myr), but the population of stars with M > 20   M⊙ are  ~3.5 Myr in age. They estimate the turn-off mass of these massive stars to be  ~65   M⊙, but note that there are some evolved stars with masses  ~15−40   M⊙, indicating either ongoing or somewhat episodic star formation. The densest central region, N105A, shows limited evidence for massive star feedback; no shell structure is seen in Oiii images (Ambrocio-Cruz et al. 1998), in spite of its proximity to LH 31. Star formation is ongoing deep in its core with a massive spectroscopic YSO identified by Epchtein et al. (1984) and Oliveira et al. (2006) and several maser sources (water, OH, and methanol, Sinclair et al. 1992; Lazendic et al. 2002; Oliveira et al. 2006; Green et al. 2008; Ellingsen et al. 2010). A cluster of four compact radio continuum sources has also been detected (Indebetouw et al. 2004) in this region, signposts of young embedded ultra-compact Hii regions.

We cover both the dense N105A region and its diffuse surroundings, looking at a circle of radius  (181 pc; 0.10 kpc2 in area) centered at RA = 05h09m53s, Dec =  −68°52′48″. We find 91 well-fit YSO candidates, eighty-four of which are new and seven of which are also in the GC09 list or both the W08 and GC09 lists. In Fig. 8, we also mark another 20 sources from the W08 and/or GC09 lists, six spectroscopic YSOs from Seale et al. (2009), and the masers mentioned above.

thumbnail Fig. 6

Three-color image of N44 with YSO candidates marked. Full region in IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Circles are newly identified candidates. Crosses are from W08, GC09, and/or Chen et al. (2009). Diamonds indicate sources on our list that are also in another list, and blue Xs are spectroscopically confirmed YSOs from Seale et al. (2009).

thumbnail Fig. 7

Full N51 region in IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5. Circles are newly identified candidates. Crosses are from W08 and/or GC09. Diamonds indicate sources on our list that are also in another list, and blue Xs are spectroscopically confirmed YSOs from Seale et al. (2009).

4.5. N113

The H ii region N113 falls in the middle of the LMC bar. It shows a complex structure of Hα bubbles with a rich molecular gas and dust morphology. Several young clusters (<10 Myr; NGC 1874, NGC 1876, and NGC 1877) are associated with N113 (Bica et al. 1996) and massive stars in these clusters are responsible for the several ionized gas bubbles that compress the remnant molecular material (Oliveira et al. 2006). In this compressed dense lane of neutral gas and dust star formation is occurring, pinpointed by maser sources in its earliest stages and by the continuum emission as the massive stars evolve (Brooks & Whiteoak 1997; Wong et al. 2006). This region also hosts the most intense H2O maser3 of the Magellanic Clouds (Whiteoak & Gardner 1986; Lazendic et al. 2002; Oliveira et al. 2006) as well as an OH maser4 (Brooks & Whiteoak 1997), both signposts of the early stages of star formation. N113 therefore presents a clear example of star formation triggered by the winds from massive stars (Oliveira et al. 2006). We highlighted this region in our Herschel YSO demonstration paper (Sewiło et al. 2010), showing that there are at least 9 far-IR YSO candidates, even more embedded than Stage i sources presented here.

We center our study at RA = 05h13m36s, Dec =  −69°21′00″, covering an area of radius 11′ (160 pc in radius or 0.080 kpc2 in area). We find 95 new YSO candidates and find six previously-identified candidates to be well-fit. In Fig. 9, we also mark 25 candidates from W08 and GC09 that do not meet our selection criteria. There are 9 spectroscopically “confirmed” YSOs (Seale et al. 2009) in this region. Two are in our list and in the GC09 list, while six are in the GC09 list and one in both W08 and GC09.

4.6. N120

The third of our medium-sized star-forming regions in the LMC’s stellar bar is N120. It is an Hii region of deceptively simple ring morphology. The ionization front is associated with interior stars, such as those in OB association LH 42 (Lucke & Hodge 1970). It also includes a supernova remnant (identified via strong [Sii] and [Oi] lines; Danziger & Dennefeld 1974). Two wind-blown bubbles join together to form the incomplete ring. The hottest stars are not centrally located, and their winds and ionizing radiation serve to drive violent motion at the ring’s boundaries (Laval et al. 1992).

We consider an area of radius 11′ (~160 pc) centered at RA = 05h18m48s, Dec =  −69°43′48″ (J2000). At the distance of the LMC, this is an area of 0.080 kpc2. We identify 51 YSO candidates, 50 of which are newly identified, following the dust ridges as seen in Spitzer images (Fig. 10). We also mark 13 further candidates from GC09 and 3 from W08. Five spectroscopic YSOs from Seale et al. (2009) are in this region, four of which are in the GC09 list and one in both the GC09 list and ours. One W08 source was classified as a non-embedded source, possibly an O-rich AGB, by Seale et al. (2009).

thumbnail Fig. 8

Three-color image of N105 with YSO candidates marked. Full region in IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5: circles are new candidates, crosses are from W08 and/or GC09, diamonds are in our list and at least one other. Blue Xs are spectroscopically confirmed YSOs from Seale et al. (2009). Large squares show masers from Sinclair et al. (1992); Lazendic et al. (2002); Oliveira et al. (2006); Ellingsen et al. (2010), as described above.

thumbnail Fig. 9

Three-color image of N113 with YSO candidates marked. Full region in IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5. Red squares indicate the locations of Herschel YSO candidates from Sewiło et al. (2010). Black squares show masers from Brooks & Whiteoak (1997); Whiteoak & Gardner (1986); Lazendic et al. (2002); Oliveira et al. (2006), as described above.

thumbnail Fig. 10

Three-color image of N120 with YSO candidates marked on IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5: circles – new candidates, crosses – W08/GC09 candidates, diamonds – previously identified candidates on our list. Blue Xs are spectroscopic YSOs from Seale et al. (2009).

4.7. N144

The little-studied N144 Hii region is located North of the LMC bar and West of the massive 30 Dor complex, on the edge of the super giant shell LMC 3 and is notable for its strong 24 μm emission. Massey et al. (2000) find three W-R stars in the central region (LH 58) and estimate an optical turn-off mass of  ~50   M⊙. For the stars of mass greater than 20   M⊙, they give  ≲ 3 Myr, and find the most massive stars to be coeval, though Lortet & Testor (1988) note that the UV/optical stellar population seems to be somewhat older in the south-east than the north.

We inspect an area of radius centered at RA = 05h26m47s, Dec =  −68°48′48″ (J2000). At the distance of the LMC, this is an area of 0.037 kpc2. Fourteen YSO candidates were identified in this region by W08 and/or GC09. One was spectroscopically confirmed by Seale et al. (2009) and another by Woods et al. (2011). We identify 31 YSO candidates as well-fit, 29 of them newly identified (Fig. 11).

thumbnail Fig. 11

Three-color image of N144 with YSO candidates marked on IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5.

4.8. N160

The region for which we use the shorthand N160 here actually encompasses a large section of the molecular ridge south of 30 Dor, including N155, N159 (subject of recent YSO discovery paper Chen et al. 2010), N172, N173, and part of N158. Hii regions in this molecular ridge have a significantly higher CO content than that of 30 Dor itself (Johansson et al. 1998), indicative of early stages of star formation. Fariña et al. (2009) spectroscopically classify over 150 O and early B-type stars in the region. Part of the extended 30 Doradus complex, various sub-regions have been extensively studied (cf. Israel et al. 1996; Nakajima et al. 2005; Nakajima 2005, and many others). Chen et al. (2010) present a detailed examination of the subregion N159, beginning with the YSO selection techniques of GC09 and adding high resolution NIR imaging. They find that many Spitzer YSO candidates are actually compact clusters or multiple sources, and some of their color-selected candidates are probably MS stars in front of bright dust. Similarly, Fleener et al. (2010) study high mass star formation in NGC 2074 and find 5 of the 10 GC09 YSO candidates show signs of multiplicity. Three 22-GHz water masers have also been observed. Two of these are in the NGC 2080 region, one in combination with 6.035-GHz OH maser and one in combination with 6.7-GHz methanol Ellingsen et al. (2010).

This is the second largest region we inspect, covering an area of 0.15 kpc2 (radius 15′; 217.5 pc) centered at RA = 05h39m44s, Dec =  −69°38′00″ (J2000). We find 214 well-fit YSO candidates and mark another 63 from W08, GC09, and Chen et al. (2010) in Fig. 12. Twenty-six of our well-fit sources are also on at least one other list. Also in the region covered here are 28 spectroscopically “confirmed” YSOs, twenty-seven from Seale et al. (2009) and one from Shimonishi et al. (2008). One candidate from W08 and one from GC09 were found to be non-embedded sources by Seale et al. (2009).

thumbnail Fig. 12

Three-color images of N160 with YSO candidates marked. a) Full region in IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5. Crosses represent sources from W08, GC09, Chen et al. (2010), and Fleener et al. (2010). Large black squares are water masers from Ellingsen et al. (2010) as discussed in text. The three brightest and most active regions are, from north to south NGC 2074, NGC 2080, and N159.

4.9. N206

Southwest of 30 Dor and N160, N206 is relatively isolated in the disk of the galaxy. It is characterized by a dense molecular ridge surrounding an evacuated shell. Kim et al. (1998) list two expanding Hi shells (GS 69 and GS 70), the edges of which correspond to the dusty ridges. Active star formation is continuing throughout the region, concentrated along the ridges as shown by Romita et al. (2010) in a detailed study of the region using the same imaging and photometry used here. They identified more than 100 candidate YSOs and estimated a SFR of 2.7  ×  10-2   M⊙   yr-1   kpc-2.

We consider the same area as Romita et al. (2010), a circle of radius  (167.5 pc) centered at RA = 05h30m48s, Dec =  −71°03′00″. We find 73 well-fit YSO candidates, 43 of which are newly identified. In Fig. 13, we also mark 93 candidates from W08, GC09, and Romita et al. (2010) that we do not consider well-fit. Seventy-five candidates are only in the Romita et al. (2010) list, which was constructed via a similar technique to that used here; see Sect. 5.1 for details. Four Seale et al. (2009) spectroscopic sources are in the GC09 list.

thumbnail Fig. 13

Three-color image of N206 with YSO candidates marked on IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5. Squares are candidates from Romita et al. (2010).

5. Results

5.1. Comparison to previous lists

Our YSO candidate selection reproduces previous lists well. While only 127 of the 485 YSO candidates identified by W08, GC09, Chen et al. (2009), Romita et al. (2010), and Chen et al. (2010) meet our final fitting criteria, only two sources do not fall within our color-cuts, both from Romita et al. (2010). Alternative names and classifications for all sources from W08, GC09, Seale et al. (2009), Chen et al. (2009), Romita et al. (2010), Fleener et al. (2010), and Chen et al. (2010) are presented in Table 3.

One-hundred-nine candidates from GC09 and/or Chen et al. (2009, 2010) are not in our SAGE catalog at all or are but have insufficient photometry for our color-selection process. These are preferentially bright or extended sources that do not fulfill our strict PSF photometry. Based on their native photometry, all would fit our color criteria: forty-nine in Set 1, four in Set 2, forty-eight in Set 3, two in Set 4, three in Sets 5, one in Set 6, and two in 8. We do not perform careful SED fitting on these sources but do a cursory treatment to assign evolutionary stages for plotting on images.

Sixty-one previously identified candidates meet only one β criterion: four from W08 and fifty-seven in Romita et al. (2010). This is to be expected, as Romita et al. (2010) require only one of their “faint” criteria to be met. These are listed as Set 10 in Table 2. Again, fitting is only performed for plotting purposes.

The remaining 176 candidates from other lists that we do not replicate are color-selected but do not meet our χ2 criteria for good SED fits. These include three sources in N120 and N160 which Seale et al. (2009) found to be unembedded sources (see Sects. 4.6 and 4.8). Another 34 of these color-selected candidates can be well-fit both as YSOs and by templates for galaxies, AGBs, and/or naked stars. We consider these fits ambiguous and note them in Table 3. Note that W08 and GC09 candidates were chosen via color-selection, and SEDs and images were visually inspected. Neither applied a quantitative SED fitting requirement, i.e. there were no χ2 requirements for matching YSO model SEDs.

Romita et al. (2010) identified and classified  ~100 YSO candidates in N206, using the same photometry and YSO model fitter we use here. Our new examination reveals 73 well-fit sources, only 23 of which fall in the final Romita et al. (2010) catalog. We find that 57 of the Romita et al. (2010) sources fit only one of our β selection criteria and zero α criteria; two more were not selected in any of our CMD cuts, accounting for most of the discrepancy in the number of identified candidates. Romita et al. (2010) used more complicated selection criteria of W08 with the addition of faint color-selections. They also examined the SED and multiple wavelength images of every source, allowing quality cuts to include higher values of χ2 / pt and allowing the inclusion of objects meeting only one “faint” color-selection. Where appropriate, they added upper-limits to 24 μm fluxes by hand, allowing more sources to be fit with higher accuracy. Fourteen candidates were removed from the Romita et al. (2010) list based on the double-peaked morphology of their SEDs suggesting multiple sources or “transition objects”, with significant gaps between the central stellar object and the inner edge of the disk. None of these sources satisfy our new selection criteria; ten are selected by only one β-cut, and the four selected in Sets 4 − 9 do not fulfill our χ2 criteria.

thumbnail Fig. 14

Mass function of the YSO candidates in each cluster. The full well-fit lists are under the thick black line. Top lines show the Kroupa IMF. The shaded portion indicates α-selected sources. a) N11: total mass of all 242 well-fit sources is 1391  ±  26   M⊙ with mass from integrated Kroupa IMF  ~ 4905   M⊙. b) N44: mass of 139 well-fit sources: 900  ±  25   M⊙; from integrated Kroupa IMF  ~ 3537   M⊙. c) N51: mass of 103 well-fit sources: 557  ±  17   M⊙; from integrated Kroupa IMF  ~ 2339   M⊙. d) N105: mass of 91 well-fit sources: 526  ±  14   M⊙; from integrated Kroupa IMF  ~ 2539   M⊙. e) N113: mass of 101 well-fit sources: 586  ±  15   M⊙; from integrated Kroupa IMF  ~ 3157   M⊙. f) N120: mass of 51 well-fit sources: 291  ±  10   M⊙; from integrated Kroupa IMF  ~ 998   M⊙. g) N144: mass of 31 well-fit sources: 163  ±  8   M⊙; from integrated Kroupa IMF  ~ 879   M⊙. h) N160: mass of 214 well-fit sources: 1424  ±  24   M⊙; from integrated Kroupa IMF  ~ 6841   M⊙. i) N206: mass of 73 well-fit sources: 431  ±  12   M⊙; from integrated Kroupa IMF  ~ 2088   M⊙.

Table 8

Star formation rates.

5.2. Mass functions and star formation rates

We also use the derived masses to construct a mass function (MF; Fig. 14) and derive a star formation rate (SFR) for each region. In constructing the mass function histograms, we include only Stage i sources, to treat a population of roughly equivalent age. We know that we are incomplete at the low-mass end because of sensitivity and at the high-mass end due to saturation, crowding, and extension. We therefore fit the peak of the observed mass distribution where we expect approximate completeness and assume a standard two-part IMF (Kroupa 2001) from 0.08 to 50 M⊙, with for Mstar < 0.5   M⊙ and for Mstar > 0.5   M⊙. We report the (Stage i) integrated mass and SFRs for each region in Table 8.

We calculate SFR for each region using a “bottoms-up” approach. We divide the total integrated mass from the MF by two possible timescales for the Stage i formation, 1 Myr and 0.2 Myr as adopted by W08 for Stage i sources in the LMC. For comparison, we also report SFRs using a top-down approach, in which the integrated 70 μm flux is used as a proxy for the total infrared (as in Lawton et al. 2010). While this integrated 70 μm flux is more likely to be contaminated by diffuse interstellar radiation heating the cold dust, these SFRs are calculated for the exact regions we examine here, known star-forming regions where we expect the IR to be dominated by forming stars. The Lawton et al. (2010) method rooted on Kennicutt (1998) and subsequent work, which is calibrated for regions of hundreds of parsecs to galaxy scales. We are at the lower end of the applicable scale. Our “bottoms-up” approach yields similar results to the top-down approach if we assume a star-formation timescale of 1 Myr. For N11, N44, and N144, the SFR per area calculated from our sources is within  ~5% the SFR from the Lawton et al. (2010) method. Our N105 value is  ~35% higher than that obtained from 70 μm, and our N206 value is  ~50% higher than the 70 μm would suggest. N113 and N51 SFRs obtained from the bottoms-up method are a factor of 1.9 and 2.25 greater than the 70 μm flux indicates. Regions N120 and N160 show bottoms-up SFRs significantly lower than the 70 μm SFR, at 66% in N120 and less than 59% in N160. Paired with the discrepancy between the unusually high percentage of Stage i candidates, this may indicate that SFR is increasing in N160.

6. Summary and ending remarks

Our new color-selection criteria are can be used in combination with SED fitting to select highly reliable candidates. We increase the total number of YSO candidates reported in these 9 star-forming regions from 485 (in Whitney et al. 2008; Gruendl & Chu 2009; Chen et al. 2009, 2010; Romita et al. 2010) to 1403, a factor of 3 increase. Without fitting SEDs, the color-selection includes  >99% of previous candidates. Our selections have the strength of using all four IRAC bands and MIPS 24 μm without losing sources with one or two missing bands. In particular, our β criteria allow the selection of lower mass YSO candidates in star-forming regions. Coupled with SED fitting, our full selection criteria largely avoids the selection of background galaxies and known non-YSOs.

We present new YSO candidate lists for each of the 9 star-forming regions covering a wide variety of physical conditions. As expected, our well-fit candidates are concentrated along dust ridges and near previously known YSOs. We take physical properties for each YSO candidate from the SED fits and derive SFRs for each region. These SFRs agree well with SFRs from the Lawton et al. (2010) 70 μm TIR proxy for the SFR calculations of Kennicutt (1998) if we assume a 1 Myr timescale for star formation. As with other studies, the majority (~56%) of our YSO candidates are classified as Stage i. Another  ~ 22% are classified as Stage ii and  ~12% as Stage i/ii, such that  ~90% of our identified YSO candidates have stages earlier than Stage ii/iii. This is largely a question of sensitivity at the wavelengths analyzed, as slightly older PMS sources, having lost most of their circumstellar matter are fainter in the infrared.

1

SAGE data and documentation are available from http://irsa.ipac.caltech.edu/data/SPITZER/SAGE/

2

We use the list “Single Frame  +  Mosaic Photometry Archive” as described in the documentation on http://irsa.ipac.caltech.edu/data/SPITZER/SAGE/

3

H2O maser: 22 GHz.

4

OH maser: 1665-MHz.

Acknowledgments

This work has been supported by grants and contracts from NASA NAG5-12595, NASA ADP-NNX11AG50G, and JPL-Spitzer contract #1275598. The authors would like to thank Barb Whitney, Remy Indebetouw, Joana Oliveira, and Jonathan Seale for helpful discussion and Sean Points and the MCELS team for providing the beautiful Hα image.

References

  1. Ambrocio-Cruz, P., Laval, A., Marcelin, M., Amram, P., & Comeron, F. 1998, A&A, 339, 173 [NASA ADS] [Google Scholar]
  2. Barbá, R. H., Rubio, M., Roth, M. R., & García, J. 2003, AJ, 125, 1940 [NASA ADS] [CrossRef] [Google Scholar]
  3. Bernasconi, P. A., & Maeder, A. 1996, A&A, 307, 829 [NASA ADS] [Google Scholar]
  4. Bica, E., Claria, J. J., Dottori, H., Santos, Jr., J. F. C., & Piatti, A. E. 1996, ApJS, 102, 57 [NASA ADS] [CrossRef] [Google Scholar]
  5. Breysacher, J., Azzopardi, M., & Testor, G. 1999, A&AS, 137, 117 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  6. Brooks, K. J., & Whiteoak, J. B. 1997, MNRAS, 291, 395 [NASA ADS] [CrossRef] [Google Scholar]
  7. Carlson, L. R., Sewiło, M., Meixner, M., et al. 2011, ApJ, 730, 78 [NASA ADS] [CrossRef] [Google Scholar]
  8. Chen, C., Chu, Y., Gruendl, R. A., Gordon, K. D., & Heitsch, F. 2009, ApJ, 695, 511 [NASA ADS] [CrossRef] [Google Scholar]
  9. Chen, C., Indebetouw, R., Chu, Y., et al. 2010, ApJ, 721, 1206 [NASA ADS] [CrossRef] [Google Scholar]
  10. Chu, Y., Mac Low, M., Garcia-Segura, G., Wakker, B., & Kennicutt, Jr., R. C. 1993, ApJ, 414, 213 [NASA ADS] [CrossRef] [Google Scholar]
  11. Chu, Y., Gruendl, R. A., Chen, C., et al. 2005, ApJ, 634, L189 [NASA ADS] [CrossRef] [Google Scholar]
  12. Danziger, I. J., & Dennefeld, M. 1974, A&A, 36, 149 [NASA ADS] [Google Scholar]
  13. Dufour, R. J., Shields, G. A., & Talbot, Jr., R. J. 1982, ApJ, 252, 461 [NASA ADS] [CrossRef] [Google Scholar]
  14. Dunne, B. C., Points, S. D., & Chu, Y. 2001, ApJS, 136, 119 [NASA ADS] [CrossRef] [Google Scholar]
  15. Ellingsen, S. P., Whiteoak, J. B., Norris, R. P., Caswell, J. L., & Vaile, R. A. 1994, MNRAS, 269, 1019 [NASA ADS] [CrossRef] [Google Scholar]
  16. Ellingsen, S. P., Breen, S. L., Caswell, J. L., Quinn, L. J., & Fuller, G. A. 2010, MNRAS, 404, 779 [NASA ADS] [CrossRef] [Google Scholar]
  17. Epchtein, N., Braz, M. A., & Sevre, F. 1984, A&A, 140, 67 [NASA ADS] [Google Scholar]
  18. Fariña, C., Bosch, G. L., Morrell, N. I., Barbá, R. H., & Walborn, N. R. 2009, AJ, 138, 510 [NASA ADS] [CrossRef] [Google Scholar]
  19. Fazio, G. G., Hora, J. L., Allen, L. E., et al. 2004, ApJS, 154, 10 [NASA ADS] [CrossRef] [Google Scholar]
  20. Fleener, C. E., Payne, J. T., Chu, Y., Chen, C., & Gruendl, R. A. 2010, AJ, 139, 158 [NASA ADS] [CrossRef] [Google Scholar]
  21. Green, J. A., Caswell, J. L., Fuller, G. A., et al. 2008, MNRAS, 385, 948 [NASA ADS] [CrossRef] [Google Scholar]
  22. Gruendl, R. A., & Chu, Y. 2009, ApJS, 184, 172 [Google Scholar]
  23. Hatano, H., Kadowaki, R., Nakajima, Y., et al. 2006, AJ, 132, 2653 [NASA ADS] [CrossRef] [Google Scholar]
  24. Henize, K. G. 1956, ApJS, 2, 315 [NASA ADS] [CrossRef] [Google Scholar]
  25. Heydari-Malayeri, M., Royer, P., Rauw, G., & Walborn, N. R. 2000, A&A, 361, 877 [NASA ADS] [Google Scholar]
  26. Heydari-Malayeri, M., Charmandaris, V., Deharveng, L., et al. 2001, A&A, 372, 527 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  27. Hora, J. L., Cohen, M., Ellis, R. G., et al. 2008, AJ, 135, 726 [NASA ADS] [CrossRef] [Google Scholar]
  28. Indebetouw, R., Johnson, K. E., & Conti, P. 2004, AJ, 128, 2206 [NASA ADS] [CrossRef] [Google Scholar]
  29. Israel, F. P., Maloney, P. R., Geis, N., et al. 1996, ApJ, 465, 738 [NASA ADS] [CrossRef] [Google Scholar]
  30. Johansson, L. E. B., Greve, A., Booth, R. S., et al. 1998, A&A, 331, 857 [NASA ADS] [Google Scholar]
  31. Kato, D., Nagashima, C., Nagayama, T., et al. 2007, PASJ, 59, 615 [NASA ADS] [Google Scholar]
  32. Kennicutt, Jr., R. C. 1998, ARA&A, 36, 189 [Google Scholar]
  33. Kim, S., Chu, Y., Staveley-Smith, L., & Smith, R. C. 1998, ApJ, 503, 729 [Google Scholar]
  34. Kroupa, P. 2001, MNRAS, 322, 231 [NASA ADS] [CrossRef] [Google Scholar]
  35. Lada, C. J. 1987, in Star Forming Regions, ed. M. Peimbert, & J. Jugaku, IAU Symp., 115, 1 [Google Scholar]
  36. Laval, A., Rosado, M., Boulesteix, J., et al. 1992, A&A, 253, 213 [NASA ADS] [Google Scholar]
  37. Lawton, B., Gordon, K. D., Babler, B., et al. 2010, ApJ, 716, 453 [NASA ADS] [CrossRef] [Google Scholar]
  38. Lazendic, J. S., Whiteoak, J. B., Klamer, I., Harbison, P. D., & Kuiper, T. B. H. 2002, MNRAS, 331, 969 [NASA ADS] [CrossRef] [Google Scholar]
  39. Lortet, M., & Testor, G. 1988, A&A, 194, 11 [NASA ADS] [Google Scholar]
  40. Lucke, P. B., & Hodge, P. W. 1970, AJ, 75, 171 [NASA ADS] [CrossRef] [Google Scholar]
  41. Massey, P., Waterhouse, E., & DeGioia-Eastwood, K. 2000, AJ, 119, 2214 [NASA ADS] [CrossRef] [Google Scholar]
  42. Meaburn, J., & Laspias, V. N. 1991, A&A, 245, 635 [NASA ADS] [Google Scholar]
  43. Meaburn, J., & Terrett, D. L. 1980, A&A, 89, 126 [NASA ADS] [Google Scholar]
  44. Meixner, M., Gordon, K. D., Indebetouw, R., et al. 2006, AJ, 132, 2268 [NASA ADS] [CrossRef] [Google Scholar]
  45. Moffat, A. F. J., Niemela, V. S., & Marraco, H. G. 1990, ApJ, 348, 232 [NASA ADS] [CrossRef] [Google Scholar]
  46. Mokiem, M. R., de Koter, A., Evans, C. J., et al. 2007, A&A, 465, 1003 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  47. Nakajima, Y. 2005, J. Korean Astron. Soc., 38, 173 [NASA ADS] [CrossRef] [Google Scholar]
  48. Nakajima, Y., Kato, D., Nagata, T., et al. 2005, AJ, 129, 776 [NASA ADS] [CrossRef] [Google Scholar]
  49. Nazé, Y., Antokhin, I. I., Rauw, G., et al. 2004, A&A, 418, 841 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  50. Oey, M. S., & Massey, P. 1995, ApJ, 452, 210 [NASA ADS] [CrossRef] [Google Scholar]
  51. Oey, M. S., & Smedley, S. A. 1998, AJ, 116, 1263 [NASA ADS] [CrossRef] [Google Scholar]
  52. Oliveira, J. M., van Loon, J. T., Stanimirović, S., & Zijlstra, A. A. 2006, MNRAS, 372, 1509 [NASA ADS] [CrossRef] [Google Scholar]
  53. Panagia, N. 2005, in Cosmic Explosions, On the 10th Anniversary of SN1993J, ed. J.-M. Marcaide, & K. W. Weiler, IAU Colloq., 192, 585 [Google Scholar]
  54. Pei, Y. C., Fall, S. M., & Hauser, M. G. 1999, ApJ, 522, 604 [NASA ADS] [CrossRef] [Google Scholar]
  55. Reid, W. A., & Parker, Q. A. 2006, MNRAS, 373, 521 [NASA ADS] [CrossRef] [Google Scholar]
  56. Rieke, G. H., Young, E. T., Engelbracht, C. W., et al. 2004, ApJS, 154, 25 [NASA ADS] [CrossRef] [Google Scholar]
  57. Robitaille, T. P., Whitney, B. A., Indebetouw, R., Wood, K., & Denzmore, P. 2006, ApJS, 167, 256 [NASA ADS] [CrossRef] [Google Scholar]
  58. Robitaille, T. P., Whitney, B. A., Indebetouw, R., & Wood, K. 2007, ApJS, 169, 328 [NASA ADS] [CrossRef] [Google Scholar]
  59. Romita, K. A., Carlson, L. R., Sewilo, M., et al. 2010, ApJ, 721, 357 [NASA ADS] [CrossRef] [Google Scholar]
  60. Rosado, M., Laval, A., Le Coarer, E., et al. 1996, A&A, 308, 588 [NASA ADS] [Google Scholar]
  61. Schaefer, B. E. 2008, AJ, 135, 112 [NASA ADS] [CrossRef] [Google Scholar]
  62. Seale, J. P., Looney, L. W., Chu, Y.-H., et al. 2009, ApJ, 699, 150 [NASA ADS] [CrossRef] [Google Scholar]
  63. Sewiło, M., Indebetouw, R., Carlson, L. R., et al. 2010, A&A, 518, L73 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  64. Shimonishi, T., Onaka, T., Kato, D., et al. 2008, ApJ, 686, L99 [NASA ADS] [CrossRef] [Google Scholar]
  65. Sinclair, M. W., Carrad, G. J., Caswell, J. L., Norris, R. P., & Whiteoak, J. B. 1992, MNRAS, 256, 33 [NASA ADS] [CrossRef] [Google Scholar]
  66. Skrutskie, M. F., Cutri, R. M., Stiening, R., et al. 2006, AJ, 131, 1163 [NASA ADS] [CrossRef] [Google Scholar]
  67. Smith, R. C., & MCELS Team 1998, PASA, 15, 163 [Google Scholar]
  68. Smith, R. C., Points, S. D., Chu, Y.-H., et al. 2005, in BAAS, 37, AAS Meeting Abstracts, 1200 [Google Scholar]
  69. Srinivasan, S., Meixner, M., Leitherer, C., et al. 2009, AJ, 137, 4810 [NASA ADS] [CrossRef] [Google Scholar]
  70. Stasińska, G., Testor, G., & Heydari-Malayeri, M. 1986, A&A, 170, L4 [NASA ADS] [Google Scholar]
  71. Vallenari, A., Chiosi, E., & Sordo, R. 2010, A&A, 511, A79 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  72. van der Marel, R. P., & Cioni, M. 2001, AJ, 122, 1807 [NASA ADS] [CrossRef] [Google Scholar]
  73. Walborn, N. R., & Parker, J. W. 1992, ApJ, 399, L87 [NASA ADS] [CrossRef] [Google Scholar]
  74. Walborn, N. R., Drissen, L., Parker, J. W., et al. 1999, AJ, 118, 1684 [NASA ADS] [CrossRef] [Google Scholar]
  75. Weinberg, M. D., & Nikolaev, S. 2001, ApJ, 548, 712 [NASA ADS] [CrossRef] [Google Scholar]
  76. Westerlund, B. E. 1997, The Magellanic Clouds (Cambridge University Press) [Google Scholar]
  77. Whiteoak, J. B., & Gardner, F. F. 1986, MNRAS, 222, 513 [NASA ADS] [CrossRef] [Google Scholar]
  78. Whitney, B. A., Wood, K., Bjorkman, J. E., & Wolff, M. J. 2003a, ApJ, 591, 1049 [Google Scholar]
  79. Whitney, B. A., Sewiło, M., Indebetouw, R., et al. 2008, AJ, 136, 18 [NASA ADS] [CrossRef] [Google Scholar]
  80. Will, J., Bomans, D. J., & Dieball, A. 1997, A&AS, 123, 455 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  81. Wong, T., Whiteoak, J. B., Ott, J., Chin, Y., & Cunningham, M. R. 2006, ApJ, 649, 224 [NASA ADS] [CrossRef] [Google Scholar]
  82. Woods, P. M., Oliveira, J. M., Kemper, F., et al. 2011, MNRAS, 411, 1597 [NASA ADS] [CrossRef] [Google Scholar]
  83. Zaritsky, D., Harris, J., & Thompson, I. 1997, AJ, 114, 1002 [NASA ADS] [CrossRef] [Google Scholar]

All Tables

Table 1

Flux table.

In the text
Table 2

Magnitude table.

In the text
Table 3

Matches table.

In the text
Table 4

LMC “well-fit” YSO candidate fitting requirements.

In the text
Table 5

Physical parameters of “well-fit” YSO candidates.

In the text
Table 6

LMC YSO candidate evolutionary stages.

In the text
Table 7

“Well-fit” YSO candidates compared to YSO candidate lists from W08, GC09, Chen et al. (2009, 2010), and Romita et al. (2010).

In the text
Table 8

Star formation rates.

In the text

All Figures

thumbnail Fig. 1

CMDs showing α (red lines) and β (blue lines) selection criteria on full LMC data. To the right (redward) of the red line is the α section of the CMD, where we expect to find massive YSOs and little contamination from non-YSO sources. The blue lines mark the boundary of the β-selection, in which we expect more contamination, particularly from background galaxies. List YSO candidates plotted are the full lists of W08 and GC09.
In the text
thumbnail Fig. 2

CMDs showing α (red lines) and β (blue lines) selection criteria with YSO candidates from these regions. The background is from the YSO models of Robitaille et al. (2006). List YSO candidates are from W08, GC09, Chen et al. (2009), Romita et al. (2010), and Chen et al. (2010) for these regions only. Well-fit YSO candidates are our final list of well-fit sources from this paper.
In the text
thumbnail Fig. 3

Example SED for each selection Set detailed in Table 4. Error bars on fluxes are shown but usually less than the size of the dot. The best fit is shown in black with other fits with shown in grey. The dashed line shows the photospheric emission from the central source. These are the same fits used to determine approximate physical and evolutionary properties. SEDs shown in a), c), d), and e) are classified as Stage i, i) as Stage i/ii, f) and h) as Stage ii, and b) and g) as Stage iii. The characteristic dip in SED c) at 4.5 μm, compared to the other three IRAC bands, indicates the presence of strong PAH emission, and error bars have been increased in 3.6, 5.8, and 8.0 μm, as described in the text.
In the text
thumbnail Fig. 4

Examined star-forming regions on MCELS Hα (blue) + SAGE 8 μm (green) + 24 μm (red). Hα image from the Magellanic Cloud Emission Line Survey (MCELS; Smith & MCELS Team 1998; Smith et al. 2005).
In the text
thumbnail Fig. 5

Three-color image of N11 with YSO candidates marked. Full region in IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked in red for Stage i, magenta for Stage i/ii, yellow for Stage ii, green for Stage ii/iii, cyan for Stage iii, and blue for unclassified. Circles are newly identified candidates. Crosses are from W08 and/or GC09. Diamonds indicate sources on our list that are also in another list, and blue Xs are spectroscopically confirmed YSOs from Seale et al. (2009) or Woods et al. (2011).
In the text
thumbnail Fig. 6

Three-color image of N44 with YSO candidates marked. Full region in IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Circles are newly identified candidates. Crosses are from W08, GC09, and/or Chen et al. (2009). Diamonds indicate sources on our list that are also in another list, and blue Xs are spectroscopically confirmed YSOs from Seale et al. (2009).
In the text
thumbnail Fig. 7

Full N51 region in IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5. Circles are newly identified candidates. Crosses are from W08 and/or GC09. Diamonds indicate sources on our list that are also in another list, and blue Xs are spectroscopically confirmed YSOs from Seale et al. (2009).
In the text
thumbnail Fig. 8

Three-color image of N105 with YSO candidates marked. Full region in IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5: circles are new candidates, crosses are from W08 and/or GC09, diamonds are in our list and at least one other. Blue Xs are spectroscopically confirmed YSOs from Seale et al. (2009). Large squares show masers from Sinclair et al. (1992); Lazendic et al. (2002); Oliveira et al. (2006); Ellingsen et al. (2010), as described above.

In the text
thumbnail Fig. 9

Three-color image of N113 with YSO candidates marked. Full region in IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5. Red squares indicate the locations of Herschel YSO candidates from Sewiło et al. (2010). Black squares show masers from Brooks & Whiteoak (1997); Whiteoak & Gardner (1986); Lazendic et al. (2002); Oliveira et al. (2006), as described above.
In the text
thumbnail Fig. 10

Three-color image of N120 with YSO candidates marked on IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5: circles – new candidates, crosses – W08/GC09 candidates, diamonds – previously identified candidates on our list. Blue Xs are spectroscopic YSOs from Seale et al. (2009).
In the text
thumbnail Fig. 11

Three-color image of N144 with YSO candidates marked on IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5.

In the text
thumbnail Fig. 12

Three-color images of N160 with YSO candidates marked. a) Full region in IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5. Crosses represent sources from W08, GC09, Chen et al. (2010), and Fleener et al. (2010). Large black squares are water masers from Ellingsen et al. (2010) as discussed in text. The three brightest and most active regions are, from north to south NGC 2074, NGC 2080, and N159.
In the text
thumbnail Fig. 13

Three-color image of N206 with YSO candidates marked on IRAC 3.6 μm (blue) + 8.0 μm (green) + MIPS 24 μm (red). Candidate YSOs are marked as in Fig. 5. Squares are candidates from Romita et al. (2010).
In the text
thumbnail Fig. 14

Mass function of the YSO candidates in each cluster. The full well-fit lists are under the thick black line. Top lines show the Kroupa IMF. The shaded portion indicates α-selected sources. a) N11: total mass of all 242 well-fit sources is 1391  ±  26   M⊙ with mass from integrated Kroupa IMF  ~ 4905   M⊙. b) N44: mass of 139 well-fit sources: 900  ±  25   M⊙; from integrated Kroupa IMF  ~ 3537   M⊙. c) N51: mass of 103 well-fit sources: 557  ±  17   M⊙; from integrated Kroupa IMF  ~ 2339   M⊙. d) N105: mass of 91 well-fit sources: 526  ±  14   M⊙; from integrated Kroupa IMF  ~ 2539   M⊙. e) N113: mass of 101 well-fit sources: 586  ±  15   M⊙; from integrated Kroupa IMF  ~ 3157   M⊙. f) N120: mass of 51 well-fit sources: 291  ±  10   M⊙; from integrated Kroupa IMF  ~ 998   M⊙. g) N144: mass of 31 well-fit sources: 163  ±  8   M⊙; from integrated Kroupa IMF  ~ 879   M⊙. h) N160: mass of 214 well-fit sources: 1424  ±  24   M⊙; from integrated Kroupa IMF  ~ 6841   M⊙. i) N206: mass of 73 well-fit sources: 431  ±  12   M⊙; from integrated Kroupa IMF  ~ 2088   M⊙.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.