Spectroscopic Assessment of WISE-based Young Stellar Object Selection
Xavier Koenig, Lynne Hillenbrand, Deborah Padgett, Daniel DeFelippis
SSpectroscopic Assessment of
WISE -based Young Stellar ObjectSelection
Xavier Koenig , Lynne A. Hillenbrand , Deborah L. Padgett , Daniel DeFelippis ABSTRACT
We have conducted a sensitive search down to the hydrogen burning limitfor unextincted stars over ∼
200 square degrees around Lambda Orionis and 20square degrees around Sigma Orionis using the methodology of Koenig & Lei-sawitz (2014). From
WISE and 2MASS data we identify 544 and 418 candidateYSOs in the vicinity of Lambda and Sigma respectively. Based on our followupspectroscopy for some candidates and the existing literature for others, we foundthat ∼
80% of the K14-selected candidates are probable or likely members ofthe Orion star forming region. The yield from the photometric selection criteriashows that
WISE sources with K S − w > . K S between 10–12 magare most likely to show spectroscopic signs of youth, while WISE sources with K S − w > K S >
12 were often AGNs when followed up spectro-scopically. The population of candidate YSOs traces known areas of active starformation, with a few new ‘hot spots’ of activity near Lynds 1588 and 1589 and amore dispersed population of YSOs in the northern half of the H ii region bubblearound σ and (cid:15) Ori. A minimal spanning tree analysis of the two regions toidentify stellar groupings finds that roughly two-thirds of the YSO candidatesin each region belong to groups of 5 or more members. The population of starsselected by
WISE outside the MST groupings also contains spectroscopicallyverified YSOs, with a local stellar density as low as 0.5 stars per square degree.
Subject headings: circumstellar matter — H ii regions — infrared: stars — stars:formation — stars: pre-main-sequence Department of Astronomy, Yale University, New Haven, CT 06511, USA Department of Astronomy, California Institute of Technology, Pasadena, CA 91125, USA NASA Goddard Space Flight Center, Greenbelt, MD 20771, USA a r X i v : . [ a s t r o - ph . GA ] J un
1. Introduction
While the data return from the
Spitzer Space Telescope vastly improved our picture ofstar formation and circumstellar disk evolution through study of a large number of molecularclouds at mid-infrared wavelengths,
Spitzer was a pointed mission with a small field of view.The more recent
WISE mission mapped the entire sky in 3.4, 4.6, 12, and 22 µ m filters toa specified 5 σ point source sensitivity of 0.08, 0.11, 1, and 6 mJy (or 16.5, 15.5, 11.3, and7.9 mag) at spatial resolution of 6–12 (cid:48)(cid:48) . With sky coverage better than IRAS and spatialresolution within a factor 2–3 as good as Spitzer , a clear advantage of
WISE is its coverageof many regions of interest that
Spitzer simply missed. Sensitivity with
WISE is comparableat all but the 22 / 24 µ m bands to Spitzer’s shallow surveys, e.g. GLIMPSE in the Galacticplane.Circumstellar disks surround young stars for the first several Myr of their lives. Thesedisks create infrared through millimeter wavelength excesses due to thermal emission fromdust that is distributed over distances from a few hundredths to several hundred AU fromthe star.
Spitzer and
WISE both probe the infrared excess emission that arises within a fewAU. As
Spitzer mapped primarily the young constituents of known molecular clouds, ourknowledge of disk evolution time scales based on
Spitzer studies is still biased and somewhatlimited.
WISE , however allows us to fill in many gaps in studies of nearby star formingregions. Particularly noteworthy wide-field studies that were enabled by
WISE include thoseby Rebull et al. (2011) on Taurus, Rizzuto et al. (2012) and Luhman & Mamajek (2012)on Upper Sco, and Koenig et al. (2012) and Koenig & Leisawitz (2014) on outer Galaxyregions.The Orion molecular cloud and young star complex is a benchmark star forming region(see Bally 2010 for a review). Orion’s prominence is due to its proximity ( ∼
400 pc) as wellas its representation of both high and low mass star formation, of clustered and isolatedstar formation, and of future, ongoing, and recent star formation. The σ Ori cluster, justwest of the Horsehead Nebula and below the easternmost “belt star” in the constellationOrion, is a 3–5 Myr old cluster associated with the central O9.5V star. Its stellar populationhas been studied by Wolk (1996), Sherry et al. (2008), Caballero et al. (2008), and Lodieuet al. (2009) and currently contains ∼
350 confirmed members and another ∼
300 photometriccandidates. There is relatively little molecular gas in the region (e.g. Lang et al. 2000); itis located between the Orion A and Orion B giant molecular clouds. The λ Ori cluster,near the famed Betelguese in Orion, is a 4–8 Myr old cluster associated with the central O8III star and contains ten additional B stars. Its low mass population has been studied byDolan & Mathieu (1999, 2001, 2002), Barrado y Navascu´es et al. (2004), and Sacco et al. 3 –(2008) though membership is vastly incomplete and knowledge of spectral types limited. Asurrounding ring includes dust and swept-up neutral and molecular gas (e.g. Lang et al.2000) and potentially younger stars than those immediately surrounding λ Ori. There isalso suggestion of a previous supernova explosion (Cunha & Smith 1996). For reviews ofthese two regions, see Walter et al. (2008) and Briceno (2008) respectively, for the σ Oricluster and the extended Orion OB1a/OB1b region included in the presentation below, andMathieu (2008) for the λ Ori region.Rich clusters in the age range of σ Ori and λ Ori are relatively rare. Complete surveyswith
WISE of these two clusters are thus important for census building, and for overall anduniform studies of circumstellar disk properties and for our understanding of disk evolution.Analysis of 2MASS data alone shows that these two clusters indeed have a lower fraction of near-infrared excess relative to many 1–3 Myr old clusters still assocatiated with moleculargas. A key question is whether the mid-infrared excess fraction is also proportionally loweror similar; the former would imply disk depletion at all radii relatively quickly whereas thelatter would suggest inside-out depletion.
2. The
WISE view of σ and λ Orionis
Figure 1 shows a 3-color composite of
WISE image data in the 3.4, 12, and 22 µ m( w , w , w
4) bands for the σ and λ Orionis fields. Emission at 12 µ m (green in the figure)is dominated by bright PAH emission that traces cloud surfaces and reveals the well known λ Orionis ring, as well as the ring around the extended σ Orionis field. The 3.4 µ m band(assigned to the blue channel in the figure) has some contribution from PAH emission lines,but mainly picks out stellar photospheric emission from foreground and background stars.The 22 µ m band (in red) captures thermal emission from dust grains which highlights cloudsurfaces in the same way as w
3, but also heated dust close to OB stars. A SIMBAD searchshows that in the λ Ori region this feature is seen around φ Orionis (B0.5III) and HR 1763(B1V). In the σ Ori region a bow-wave feature around σ Orionis, (see Ochsendorf et al. 2014)and halos around VV Ori (B1V) and HR 1861 (B1IV) are picked out by bright, extended22 µ m emission. These stars are labeled in the figure, as well as the three main belt stars ofthe Orion constellation and X Ori, a Mira variable evolved star.Note that the other red, evenly spaced, diffuse patches of emission aligned verticallywith the bright emission in the eastern part of the σ Ori region and also appearing in thelower, central part of the panel are image artifacts in the 22 µ m WISE images. 4 –Fig. 1.—
WISE σ (upper panel) and λ Orionis (right panel)regions. The image assigns the 3.4, 12, and 22 µ m channels to blue, green and red in theimage respectively. The large, diffuse red spots appearing at regular intervals northwardfrom NGC 2023/2024 and from 6 h m , − ◦ are bright source image latents from w
4. 5 –
3. Initial YSO Selection Criteria
The combination of 1.2, 1.6, and 2.2 µ m photometry from 2MASS with the 3.4, 4.6,12, and 22 µ m photometry from WISE is a sufficient lever arm on the stellar (e.g. from J − K colors) to circumstellar (e.g. from w − w w − w σ and λ Orionis from the
WISE
All Sky point source catalog. For the former we extracted a region bounded by84.0 < RA < − . < Dec < .
05 degrees (J2000.0), for the latter a polygonwith vertices at (RA, Dec) = (77.1 11.6; 87.7 15.4; 90.8 6.5; 80.5 2.8) to encompass the fullextent of the large ring of emission that surrounds the λ Orionis H ii region, as seen in the WISE band 3 atlas images at 12 µ m.The WISE
All Sky catalog also provides photometry in the near-infrared J , H and K S bands from 2MASS. The WISE
All Sky photometry pipeline uses a match radius of 3 (cid:48)(cid:48) toidentify the 2MASS counterparts of
WISE sources.We then applied an early version of the
WISE-2MASS
YSO classification and galaxyfiltration scheme of Koenig et al. (2012). As described by Koenig et al. the scheme first ap-plies
WISE color and magnitude cuts to remove background unresolved star-forming galaxiesand active galactic nuclei (AGNs). These objects are typically redder and fainter than themajority of young stars in Galactic star-forming regions. A further round of cuts is madeto remove objects likely to be shocked blobs of gas in young star outflows (bright at
WISE band 2) and spurious detections of bright nebular emission (objects with very red w − w WISE colors,matching the categories laid out by Greene et al. (1994) as best as possible. This schemedoes not attempt to find either Class III or Flat SED objects, however, only Class II andClass I. Additional young stars are selected based on combined
WISE and 2MASS colors,aiming to retrieve those objects missed due to their
WISE band 3 detection being obscuredby bright nebular background emission. The final steps of the scheme look for candidatetransition disk objects (Strom et al. 1989) with red w − w w − w w w < . WISE
T Tauri star locus. The region of
WISE color-color space roughly defined by0 . < w − w < . . < w − w < . WISE band 4 at 22 µ m by fake detections due to nebulosity produced a large number of candidatetransition disks in this initial YSO candidate list. We thus required a signal to noise at band4 of at least 4.5, and set an upper limit to the w − w σ Ori and λ Ori, we additionallyconstrained the source lists before each spectroscopic run at the telescope, as describedbelow.
4. Spectroscopic Target Selection and Observations
In total, 230 sources were observed spectroscopically at the Palomar Observatory 200”telescope. All data were collected with the Double Spectrograph (originally comissioned byOke & Gunn 1982 but with many upgrades since that time) using a dichroic at 4800 A anda 1200 l/mm red and a 300 l/mm blue grating. Exposures of FeAr and HeNeAr lamps wereobtained for the blue and red side wavelength calibration.The spectra were collected over several years, with evolving selection criteria.
As part of an unrelated program studying low mass stars in σ Ori (Cody & Hillenbrand2014), spectroscopic observations were taken in 2009 of known members that were laterrevealed to be of interest from the vantage of the current
WISE study. These objects wereselected for spectroscopy based on their presence in the fields photometrically monitored byCody & Hillenbrand (2010) and lack of a published spectral type at that time.As discussed by Cody (2012), observations took place on the nights of 18–21 Januaryand 19–20 December with 68 unique low mass stellar and brown dwarf objects in σ Oriobserved. 7 –
For the first run dedicated to
WISE follow-up in λ Ori and σ Ori, we required allstars to have non-null photometric error in
WISE bands 1 and 4 and to not suffer fromthe diffraction spike, scattered light halo, optical ghost or latent contaminant flags (SQLquery: cc flags not matches ’[DHOP]’). To minimize obvious contamination by backgroundgalaxies, we required the 2MASS Extended Source Catalog proximity column xscprox to beeither null or > (cid:48)(cid:48) . Finally, using the WISE band 4 image atlas we inspected all the YSOcandidates and rejected objects that were extended or whose 22 µ m emission appeared tobe offset from the shorter wavelength centroids by more than 2 (cid:48)(cid:48) . In both fields we noticeda concentration of objects with red colors that appears at magnitudes fainter than J = 16and K S = 14. The spatial distribution of these detections is roughly uniform; thus they arelikely mostly extragalactic, so we also cut them from further consideration. We obtainedoptical photometry for all sources where available from the NOMAD online database andrestricted our targets to objects with R magnitude less than 18.3. For σ Ori we were ableto use imaging data from the Sloan Digital Sky Survey to confirm that objects fainter than J = 16 are indeed dominated by galaxies.In σ Ori we obtained 29 objects with DBSP on 01 January, 2012, UT, selecting thoseYSO candidates with K S <
14 and without existing spectral types in the literature or inCody’s thesis work discussed above. Most objects having
J <
11 and w − w > . w − w > .
7. Among objects with
J >
11, the observations focused on those without previous designations in SIMBAD, i.e.
WISE is the first that attention has been called to them. A few objects meeting the abovecriteria had roughly equal brightness, close companions within a few arcsec in 2MASS or onthe telescope guide camera when slewing to the field. These were excluded from observationdue to concerns about contamination in the excess selection, since we didn’t have enoughinformation to attribute w w λ Ori we obtained 40 objects with DBSP on 02 January, 2012, UT, selecting YSOcandidates with
J <
16 and w <
14 and a K S <
14 cut during the observing. We alsolimited ourselves to targets south of Declination +09 so as to avoid the areas previouslystudied by others, namely, the 1 deg around λ Ori itself and the B30 and B35 regions, withthe goal being that our work with
WISE is unique in the lower Declination areas.The prioritization for observations was a combination of brightness (brighter objects at w − w w − w < .
15 was observed (i.e., noneof the Class III/transition disk candidates). We focused on the reddest w − w λ Ori ring towards the southeast was prioritized; this spatialcut also captured many of the reddest w − w λ Ori, almost every
WISE -selected YSO candidate with
J < . In 2013, all targets were in the λ Ori field, again with a focus on the western andsouthern parts of the region. We selected YSO candidates with 8 < w <
13 to avoid bothinfrared-bright asymptotic giant branch stars (AGBs) and faint AGNs, although we did notimpose a color criterion. In total, 80 sources were observed on 2, 4, 5 February 2013, UT.For this run, we did not apply the additional
WISE band 4 quality assessment, or therequirement on the cc flags parameter that were used for the 2012 run. One result potentiallyattributable to this modification is that the fraction of observed sources with emission lineswas lower than in the earlier spectra.
5. Spectroscopic Data Reduction and Analysis
Two-dimensional CCD images from both the blue and red channels of the spectrographwere corrected for detector bias and flat field and one-dimensional spectra were extracted,wavelength calibrated, and flux calibrated, all within the twodspec and onedspec packagesof IRAF. In 2012, Feige 34, Feige 110, Gl 38-31, HD 19445, and HD 199178 were used asflux standards while in 2013, Feige 34, HD 93521, Hiltner 600, and G 191B2B served thepurpose. For our setup, the full spectral range on the blue side was ∼ α , Li I II triplet at 8498, 8542 and 8662 ˚A, and related H 9 –and K lines at 3933 and 3968 ˚A. Other emission and absorption lines noted in some spectraincluded [O I ] 6300, 6363 ˚A, [S II ] 6717, 6732 ˚A, O I I I δ . Several objects with the Balmer jump clearly inemission were also identified.We found that objects with K S − w > . K S between 10–12 mag were mostlikely to show spectroscopic signs of youth such as strong H α or the Ca II triplet in emission.Extremely red objects with K S − w > K S >
12 were more likely to be AGNs.
6. Updated YSO Selection Criteria
The initial
WISE -based selection criteria and spectroscopic follow up observations de-scribed above served to obtain a sample of young stars with optical spectra in the λ and σ Orionis regions. Subsequent work by K14 showed that this original YSO finding schemegenerates a large number of spurious candidate young stars. We thus implemented for finalpresentation here the more robust YSO finding procedure detailed in K14. The K14 schemebegins in the same way as K12 by removing extragalactic contaminants (AGNs and star-forming galaxies). It does not include a step to remove detection of sources that appear reddue to nebular emission because it relies on the fact that YSO candidates occupy loci incolor space that naturally separate them from these detections. As with K12, the schemethen uses
WISE and then
WISE +2MASS color criteria to identify and classify Class I andII YSOs. Final steps select candidate transition disk objects with red w − w w
4. A newfeature of the K14 scheme is the addition of an explicit phase of AGB star removal that usescuts in the
WISE w − w w − w w w − w λ and σ Orionis is shown in Figure 2. To assess the new contribution made by our survey to thedistribution of YSO candidates in the two regions, we carried out a census of the relevantliterature. Both the λ and σ Orionis regions have been surveyed by many authors. Themain studies in λ Ori are those by Hern´andez et al. (2010) and Dolan & Mathieu (2002). In σ Ori we compile the Mayrit catalog of Caballero (2008) and the studies by Pe˜na Ram´ırezet al. (2012), Brice˜no et al. (2005), Megeath et al. (2012), B´ejar et al. (2011) and Luhmanet al. (2008). 10 –Figure 3 shows how the compiled list of previously known likely members of the tworegions relate to the newly identified
WISE
YSO candidates found in this paper. Figure 4shows the distribution of the targets followed up with optical spectroscopy in the two regions.The upper panels of Figures 2, 3 and 4 show that our extracted sample of YSO candidatesextends beyond the boundaries of the λ Ori ring. G192.16 − ii region at a distanceof 1.52 kpc (Shiozaki et al. 2011), with an associated young cluster (Carpenter et al. 1993).LDN 1617 is known for its association with Herbig Haro outflow sources and is at a distance ∼
400 pc (Kajdiˇc et al. 2012).
While the scheme of K14 provides an assessment of the distribution and a simple break-down of the evolutionary state of the YSO candidates that it selects, a more commonlyused means of dividing young objects with infrared excess emission is to use the slope oftheir spectral energy distributions (SEDs) in the near and mid-infrared. We follow themethodology of Greene et al. (1994) developed for the availability of K-band and 25 µ mIRAS measurements, and compute α , the slope of the SED, for each YSO candidate overthe longest possible wavelength baseline available from 2MASS and WISE (note: we do notperform a fit to the SEDs). Ideally we compute α from the slope between the 2MASS K S band to 22 µ m at WISE band 4, requiring non-null K S band photometric error, signal tonoise in w ≥ . < χ w < .
7. If a source does not meet the K S band requirement wederive α from the slope between w w
4, requiring signal to noise in w ≥
3. If the band4 photometry fails the above criteria we use
WISE band 3 with the same test on signal tonoise and χ at w
3, paired with either 2MASS K S or w H and WISE w K S , w w α is computed from the following equation: α = d log λF λ d log λ (1)Note that for all objects, we attempt to remove the effects of dust extinction by lookingup the value of A K in the nearest pixel in the K -band extinction map of Lombardi et al.(2011). We compute the extinction in the 2MASS bands using the extinction law of Rieke& Lebofsky (1985) and in the WISE bands using the interpolation presented in Koenig &Leisawitz (2014). Having derived a value of α for each YSO candidate, we divide them intothe categories of Greene et al. (1994) as Class I ( α ≥ . > α ≥ − . − . > α ≥ − .
6) and Class III ( α < − . σ and λ Orionis (upper and lower panelsrespectively) as identified by the scheme of Koenig & Leisawitz (2014). Red points andtriangles: candidate Class I’s from
WISE or WISE +2MASS. Yellow points and squares:candidate Class II’s from
WISE or WISE +2MASS. Blue points: candidate transition diskobjects. 12 –Fig. 3.— Distribution of literature YSOs from the papers cited in the text (cyan box points)and
WISE
YSO candidates (red-orange dot points). λ , (cid:15) and σ Orionis are plotted as whitediamonds. The locations of other notable dark clouds or clusters are also marked. 13 –Fig. 4.— Distribution of spectra acquired in this paper (green box points), the Cody et al.(2010) sample (blue box points) and two YSOs from Riaz et al. (2015) (light triangle points).YSO candidates are shown as red orange dot points. λ Ori, and in the lower panel (cid:15) , δ and σ Orionis are plotted as white diamonds and ζ Ori as a black diamond. 14 –signify ‘no infrared excess,’ but rather ‘weak excess.’ We tabulate the resulting breakdownof source types in Table 1. In both fields, some fraction of the K14 Class II sources arecategorized as Class III in the α prescription, and some are placed in the ‘Flat-SED’ class,although no more than 20% in either case. However, from the K14 Class I sources, 63%of the candidates in σ Ori move to the Flat SED class under the α calculation, comparedwith 25% in λ Ori. In general this suggests that the Class I sources in λ Ori extend to muchredder colors such that their slopes α are consistent with a Class I source in that prescription.However, it should be noted that the most embedded clusters in the σ Ori region, those nearto NGC 2023 and 2024, are poorly characterized in this survey, owing to the high degree ofsaturation in the
WISE band 3 and 4 images in that area. This issue limits the robustnessof the comparison of the α SED class distributions between λ and σ Ori.The Class III category in the α prescription is an overlapping set with K14’s Class IIand transition disk sources, because its only definition is a steeply negative SED slope. Sincetransition disk candidates are selected on the basis of weak excess in w − w w − w
4, most of the transition disk objects in our samples are actually countedas Class II YSOs by the α prescription (14/15 from the λ Ori sample, 11/13 from σ Ori).
7. Assessment of the WISE-Selected YSO Candidates7.1. Spectroscopic Success Rate in YSO Confirmation In λ Orionis, 75 of our spectra were YSO candidates from the K14 selection scheme.In σ Orionis, we obtained spectra of 48 K14 YSO candidates. We examined how many ofthese targets showed emission in either the H α (equivalent width < −
10 ˚A) or Ca II < α in emission. In λ Orionis, 50/75 objects (67%) met these criteria,while in σ Orionis 28/48 spectra (58%) met these requirements. While a more generalindicator of youth is provided by the Lithium Li I I equivalent width measurements. 15 – In Figure 6 we show color-color diagrams of the K14
WISE
YSO candidates in the twoOrion regions. We overlay the colors of the spectroscopic sample acquired in each field,separating those with and without H α or Ca II σ Ori, we also show the location ofthe two objects investigated by Riaz et al. (2015), both of which are candidate YSOs in ourselection scheme and both of which appear to be good spectroscopic YSO candidates.In both λ Ori and σ Ori a significant fraction (28/75 and 21/48 respectively) of spectro-scopically observed K14
WISE
YSOs do not show significant H α or Ca II J − K S = 1.3, a majority of WISE selected YSOs do exhibit emission inthese lines. Figure 6 shows that in
WISE color-space, a similar increase in the prevalence ofH α and/or Ca II w − w > . w − w > . α equivalentwidth between the K14 YSO candidates and those spectra not identified by these criteria(i.e., acquired under either the K12 criteria for infrared excess or as part of the Cody &Hillenbrand (2014) sample). Larger EW(H α ) in emission (i.e. more negative) are seen in theK14 selected YSO candidates. In Figure 5 we plot the equivalent width in the H α line versusobserved WISE w − w < − that the equivalent width distributions were drawn from the same parentdistribution. The anomalous red cross points with w − w > w WISE filters. The K14 selected YSOs are likely tohave greater amounts of circumstellar material and therefore perhaps higher accretion ratesthan those objects with only weak infrared excess.
WISE
Selection
Approximately half of the optical spectra are of objects not designated as YSO candi-dates by the K14 scheme. Specifically, in λ Ori, 55 of the 130 and in σ Ori, 52 of 100 objects 16 –Fig. 5.— Equivalent width in the H α line versus w − w λ and σ Orionis (black points, or green points if a spectrum was acquired). Additional plot symbolsshow spectroscopic targets, colored as in figure legend. 18 –for which we acquired spectra do not correspond to a K14
WISE
YSO candidate. As de-scribed above, this feature is partly because some of our spectroscopic targets were selectedwith earlier versions of our YSO selection criteria and partly because some came from theseparately selected targets of Cody et al. The K14 scheme aggessively selects against objectslikely to have fake detections in the
WISE catalog, in particular in
WISE band 4, whichhinders its ability to find objects that have infrared excess only in the longest wavelength
WISE band.As seen in Figure 5, at color w − w < . α emitters. At color w − w > . α emitters, and conversely some objects that are selected byK14 but do not exhibit strong H α emission (those with small equivalent widths).In Table 2 we present a summary of the properties of objects not selected under K14but for which we have spectra. We list first the total number of spectra that do not matchto a candidate WISE
YSO, then the number that exhibit H α or Ca II N ( young )). We then give the number thathave red infrared colors, either w − w > w − w > H − K S > − . × ( w − w
2) + 0 . N ( excess )). Some of these objects not selected under the K14scheme have weak or no excess in WISE bands 1 and 2: W − W < .
25 which makes themcandidate transition disk objects, that is, stars with cleared out or depleted inner regions.The number of these is given in Table 2 as N ( T D − like ). Of the full sample of spectra thatdo not match K14 WISE
YSO candidates, 24/107 (22%) show spectroscopic signs of youthin H α or Ca II Spectra marked in Figure 6 that appear to be those of AGNs or quasars in the λ Orifield are a result of the overlap in
WISE color space of these extragalactic sources withprotostars and the reddest Class II sources. These objects escape elimination because oftheir brightness in w λ Orionis field, of the K14 YSO candidates with spectral follow-up, four werefound to resemble those of extragalactic objects, whether AGNs or other galaxies. One K14Class I candidate object was found to be a known quasar in a SIMBAD cross-match. Oneobject appeared to have a featureless but blue spectrum. In Figure 7 we show a color-magnitude diagram summarizing the results of the overlap between the
WISE photometric 19 –Table 1.
W ISE
YSO Candidate Breakdown
Method N(Class I) N(Flat) N(Class II) N(Class III) N(Transition Disk) λ OriK12 405 · · · · · · · · · · · · α
58 80 353 53 · · · σ OriK12 159 · · · · · · · · · · · · α · · · Note. — K12 refers to the scheme of Koenig et al. (2012), specifically as described inSection 3. K14 refers to the scheme of Koenig & Leisawitz (2014). α refers to YSO classesbased on the slope of the infrared SED. Table 2. Spectroscopic targets not selected by
WISE scheme
Field Total N(Young) N(excess) N(TD-like) λ Ori 55 10 35 22 σ Ori 52 15 6 0Note. — TD-like means ‘transition-disk’ like, as discussedin the text.
20 –YSO selection scheme and our spectral follow-up program. In the panel for σ Ori, we alsoshow the location of the two YSOs studied by Riaz et al. (2015).The extragalactic spectra were found amongst the protostar candidates in the λ Orionisfield and in general were found in regions of low YSO surface density. In Figure 8 we showthe distribution of protostar candidate sources color-coded by their projected angular spacedensity σ , divided into quartiles of apparent space density. We compute σ by findingthe distance to each object’s 6th nearest neighbor in the complete YSO catalog (Class II,transition disk, as well as protostars) and following the prescription of Casertano & Hut(1985). The quartiles are as follows—the lowest 25% of protostars have σ < < σ < < σ < σ > WISE
YSO candidates above and below apparent density σ =0.75 stars per square degree. Wealso show for comparison a contour representation of the color-magnitude distribution ofstars within ∼ ◦ of the North and South Galactic poles (requiring w > χ < . K S uncertainty < . K S − w > K S > λ Ori may be extragalacticcontaminants. In σ Ori the numbers are 4 of 24 and 3 of 381 respectively. K14 estimatedthe Galactic contamination rate using the Galactic stellar populations model of Wainscoatet al. (1992) to predict an upper limit to the number density of non-YSOs that would appearin their test field in the same infrared magnitude and color ranges as YSOs in their scheme.We estimate that 28 of the Class I candidates and 93 of the Class II candidates in λ Ori 21 –Fig. 7.— Infrared color-magnitude diagrams of the K14-selected YSO candidates in σ and λ Orionis (black points, or green points if a spectrum was acquired). Additional plot symbolsshow spectroscopic targets, colored as in figure legend. 22 –Fig. 8.— Distribution of protostar candidates in the extended λ Orionis field color-coded bytheir local apparent space density. Red boxes: AGNs or quasars, orange box: spectra blueand featureless. 23 –may be Galactic contaminants and in σ Ori, 3 Class I and 9 Class II’s respectively may beGalactic contaminants. Our spectroscopic sample in λ Ori confirms none of the 6 Class Isources observed and 40 of the 64 Class II’s as young stars on the basis of H α or Ca II σ Ori (together with the samples ofCody et al. and Riaz et al.) confirms 3 of the 4 Class I sources observed and 26 of the 46Class II’s as young stars in the same way. The Class I success rates rate and K14 predictedcontamination rates appear to be consistent for the two fields. For the Class II sources, weconfirm only the smaller subset of our candidates that possess H α or Ca II ∼
480 square degree test field in the OuterGalaxy, or 0.36 sources deg − and 0.6 deg − respectively. These estimates predict 72 extra-galactic and 121 Galactic sources contaminating the λ Orionis
WISE
YSO list and 7 and12 sources in the σ Orionis list. The number of extragalactic contaminants was estimatedin K14 by randomly, repeatedly placing an appropriate number density of AllWISE catalogobjects, drawn from a patch of sky away from the Galactic Plane, behind the 2D extinctionmap of Schlegel et al. (1998) and finding the average number of these that would be selectedas YSO candidates by the K14 scheme. The Galactic contamination rate was estimated byusing the Galactic stellar populations model of Wainscoat et al. (1992) to predict an upperlimit to number density of non-YSOs that would appear in that part of the sky in the sameinfrared magnitude and color ranges as YSOs in the K14 scheme. Using the spectroscopicdata we have for the two fields combined (including the samples of Cody et al. and Riazet al.), for those sources selected by K14 as Class II YSOs, we find 66 out of 110 objectsfollowed up with spectra show either H α or Ca II α or Ca II ∼
22% for the two fields combined. Thisrate appears low for the Class I sources, albeit from a small sample. The success rate ofconfirmed Class II candidates in these fields appears low as well; however, the absence oflithium absorption equivalent width measurements for these objects means there may bemore young stars within this sample than we can ascertain from emission lines alone.We note that Assef et al. (2013) quote a true AGN (dusty quasar) surface density of62 per square degree, based on the simple selection criteria w − w ≥ . w < . σ Ori the median local spacedensity is 33 deg − , while in λ Ori it is 9.5 deg − . The AGN surface density of Assef et al.(2013) is also higher than the predictions of Koenig & Leisawitz (2014) for the extragalacticcontamination rate. K14 estimated that their YSO candidate list was likely contaminatedby approximately 173 extragalactic sources in their ∼
480 square degree test field in theOuter Galaxy, or 0.36 sources deg − . The high surface density of AGNs found by Assefet al. (2013) is largely due to their deeper survey in w
2. Our YSO candidate sample isentirely brighter than w .
5, resulting from a cut in w w w K magnitude upper limit is a critical criterionfor avoiding extragalactic objects in a spectroscopic survey. Tables 3 and 4 present the entire
WISE -2MASS derived catalogs of YSO candidates forthe σ and λ Orionis fields respectively. We give the AllWISE catalog designation, coordi-nates, 2MASS and
WISE photometry, source YSO class determined by the K14 scheme andthe computed SED slope α . We list the equivalent width measurements in the H α and Ca II for previous information in the literature regarding possiblestatus as a contaminant or a young star. Based on this information we have designatedsources in Tables 3 and 4 with a ”y” (yes) or ”m” (maybe) or ”n” (no) to indicate thelikelihood of membership in the Orion star forming region. Where spectral types wereavailable in the literature we have included these in the tables and noted the specific referencein the final column.In λ Ori, of 545 K14-selected candidates, 349 have no additional membership informa-tion from either the literature or our spectra. Of the 196 with supplemental informationbeyond just the 2
M ASS + W ISE photometry, we designate 149 as probable members (in-cluding 50 from our spectra) and 26 (including 17 from our spectra) as possible membersfor a confirmation rate of 76–89%. In σ Ori, of 418 K14-selected candidates, 147 have noadditional membership information from either the literature or our spectra. Of the 271 with http://simbad.u-strasbg.fr
25 –Fig. 9.— Color-magnitude diagrams of all λ Ori
WISE
YSO candidates above (left panel) andbelow (right panel) apparent space density σ =0.75 stars per square degree. The backgroundcontours shows a 2D histogram of objects drawn from an equivalent area of sky around theNorth and South Galactic poles in the AllWISE catalog (200 square degrees). Contour levelslog-spaced starting at 1, increasing in intervals of 0.7dex. Colored points are as in Fig. 7upper panel. 26 –supplemental information, 233 are probable members (including 38 from our spectra) and33 (including 6 from our spectra) are possible members, resulting in a confirmation rate of86–98%. Only 20 sources in λ Ori and 4 in σ Ori are ruled out as galaxies or evolved stars.
8. Final YSO Census8.1. Completeness
A summary of the number of YSO candidates found using the K14 scheme is presentedin Table 1. Assuming a 400 pc distance, the K s <
14 limit of the selected YSO candidatesillustrated in Figure 7 corresponds at an age of 1 Myr to a mass of 0.05 M (cid:12) , with the masslimit increasing as the stars get older (about 0.15 M (cid:12) at 10 Myr and 0.5 M (cid:12) at 1 Gyr). Inthe presence of even modest extinction, we should be sensitive to the vast majority of youngstar candidates having infrared excess.To assess the completeness of the K14-derived YSO sample, we take as a comparisonsample the updated list of young stars found in nearby star forming regions from the Coresto Disks (c2d) survey produced by Hsieh & Lai (2013). We take their Spitzer photometriccatalog, classify their young stars with the scheme of Gutermuth et al. (2009) and producelists of Class I, II and transition disk young stars. We then search for that subset of objectsin the AllWISE catalog and classify them following K14. At an apparent magnitude of K S =13.25 (10.3) the retrieval rate of Spitzer
Class I or II YSOs by
WISE is down to 50%(90%). An apparent magnitude of K S =13.25 converts to a mass limit of 0.1 M (cid:12) (1.4 M (cid:12) ) foran age between 3-5 Myr, using the stellar evolution models of Siess et al. (2000), assuminga distance of 400 pc and zero extinction. In Figure 2, we show the distribution on the sky of YSOs found using the scheme ofKoenig & Leisawitz (2014). Figure 4 shows the distribution of the spectra we acquired forthis paper.We note in Figure 9 that not all of the low density YSOs are contaminants. SeveralYSO candidates in the right panel have K S < . < K S − w <
2, a locus that isnot a feature of the reference field color-magnitude distribution. These are likely young starsthat either have drifted away from their birth clusters or that formed in relative isolation.We assess the clustering properties of the YSO candidates in Orion by constructing 27 –a minimal spanning tree from the source distribution. An MST is defined as the networkof lines, or branches, that connects a set of points together such that the total length ofthe branches is minimized and there are no closed loops. We follow the methodology ofGutermuth et al. (2009) to identify a branch length cut off that separates groups or clustersfrom the low density background.We fit two line segments to the cumulative distribution function of MST branch lengths:a steep-sloped segment at short spacings and a shallow-sloped segment at long spacings. Wethen adopt the MST branch length of the intersection point between the two lines as thecritical cutting length. For the purposes of this analysis, we define a cluster as a group offive or more points all with branch lengths shorter than the cutoff. We run this analysison both regions, using the complete YSO candidate lists and show the resulting clusters inFigure 10. In λ Ori, 65% of the YSO candidates belong to MST-identified groups of 5 ormore members, while in σ Ori the fraction is 69%.In the right panel, showing the λ Ori region, the MST analysis picks out several stellaraggregates away from the λ Ori ring. To the extreme north-east (RA > ◦ , Dec > ◦ ),it identifies a YSO cluster associated with the more distant G192.16 H ii region. Along thesouthern part of the panel, small groups near to L1617 are seen (RA > ◦ , Dec < ◦ ),as well as the northern edge of the larger 25 Ori/Orion OB1a cluster (81 < RA < ◦ , Dec < ◦ ). Within the λ Orionis ring our catalog of YSO candidates resides in clusters aroundthe head of the B35 pillar (85 < RA < ◦ , 8 < Dec < . ◦
5) and around B30 (RA ∼ . ◦
7, Dec ∼ . ◦ λ Ori cluster (purple, pink and green groups). The majority of the ringof emission seen in
WISE band 3 does not appear to possess a large quantity of young starswith infrared excess, however small clusters of Class II and I sources are coincident withL1588 and L1589 (magenta and red points, 79 < RA < . ◦
5, 5 < Dec < ◦ ).In the right panel showing the σ Ori region, the σ Ori cluster itself is identified bythe MST algorithm (red points, centered on RA ∼ . ◦
7, Dec ∼ − . ◦ > . ◦ − . < Dec ∼ − . ◦ < RA < ◦ , − . < Dec < − . ◦
2) is also picked out, but broken up by the MST algorithm into severalgroups. We are uncertain about the nature of the remaining, small groups in the right panelof the figure without further spectroscopic surveys to probe to lower masses and find andyoung stars without disk excess emission.Using our spectroscopic sample, we confirm 12 YSO candidates outside MST-definedgroups in λ Ori and 1 outside the MST-defined groups in σ Ori. The lowest local stellardensity around such objects, as characterized by σ (see Section 7.4), is 0.5 stars per square 28 –Fig. 10.— MST groups (colored points) and all other YSO candidates (black points) in σ and λ Ori, left and right panels respectively. Stars are colored according to their group forvisual identification. 29 –degree in λ Ori and 18.4 stars per square degree in σ Ori. This diffuse population of YSOscould be made up of stars that have dispersed from where they formed. As with the remainderof the young stars we identify though, a complete census of the mass function and the non-disked population is necessary to determine the exact nature of these objects.
9. Discussion9.1. The Wide-Field distribution of YSOs in σ and λ Ori
While
WISE lacks the sensitivity and resolution to match the depth of previous YSOsurveys in Orion, Figures 2, 3 and 4 demonstrate its ability to capture the basic distributionof star formation on large areal scales. σ Ori region
In and around σ Orionis itself, several extensive and deep surveys of YSOs have focusedon mapping out the low mass end of the initial mass function, resulting in the large numberof literature sources marked in Fig. 3. In the brightest region of nebular emission in the field,the
Spitzer survey of Megeath et al. (2012) also identifies a much greater number of YSOcandidates than our present
WISE survey, roughly 370 YSOs deg − , compared to 84 YSOsdeg − from WISE in that location, or a factor ∼ WISE
YSO candidates are mostly Class II or transition disk objects. The mismatch betweenthe literature YSOs and our sample in this part of the field is due to the ability of Brice˜noet al. to find weak-line T Tauri stars (WTTs) that lack significant infrared excess which ourcolor-criteria cannot easily pick out. We find only 2 of the 105 objects noted by Brice˜no et al.as WTT objects in this region, but 21 of their 33 ‘CTT’ objects in a 2 (cid:48)(cid:48) cross-match betweencatalogs. As discussed by Brice˜no et al., these northern YSOs are a part of the Orion OB1bassociation. We do not note an obvious preferential association of our YSO candidates inthis area with the cloud boundary/the bright rim traced by w λ Ori region In λ Ori our catalog of YSO candidates traces the previously observed distributionof YSOs that extends from the head of the B35 pillar across to B30, passing through thecentral λ Ori cluster. Our catalog captures a subset of the YSO distribution mapped outby the
Spitzer survey of Hern´andez et al. (2010) around the central λ Ori cluster. We alsomiss a fraction of the YSOs documented by the works of Dolan & Mathieu that lack stronginfrared excess. The small cluster associated with L1588 and L1589 appears to be new tothis study. Previous work in this area has focused on the several Herbig-Haro objects andoutflows (HH114, 115, 328 and 329) concentrated around IRAS 05155+0707 (see for exampleConnelley et al. 2007). Our catalog also finds a low density distribution of candidate ClassI sources to the east and north-east of λ Orionis, but still within the greyscale mosaic areain Figure 2. Two of these objects have already been found to be background AGNs (seeSection 7.4). As noted by Koenig & Leisawitz (2014), the
WISE
Class I candidate selectionis known to be the most highly contaminated by extragalactic sources. Because of their lackof obvious association with stellar groupings or bright nebular emission, we suspect thatmost of this group are likely background contaminants.
10. Conclusions
We have conducted a sensitive search down to the hydrogen burning limit for unextinctedstars using
WISE over ∼
200 square degrees around Lambda Orionis and 20 square degreesaround Sigma Orionis. We used the methodology of Koenig & Leisawitz (2014) that buildson the heritage of the work of Koenig et al. (2012), to identify a sample of 544 stars inLambda and 418 stars in Sigma that are candidate YSOs.We conducted optical spectroscopic followup of 14% of these candidates in Lambda and11% in Sigma Orionis. On the basis of strong emission in the H α or Ca II I WISE selected YSO candidates. Based on our followup spectroscopy for some candidatesand the existing literature for others, we found that ∼
80% of the K14-selected candidatesare probable or likely members of the Orion star forming region.We improved our understanding of the yield from the photometric selection criteria invarious ranges of color-color and color-magnitude space, in particular
WISE sources with K S − w > . K S between 10–12 mag were most likely to show spectroscopicsigns of youth. WISE sources with K S − w > K S >
12 were often AGNs when 31 –followed up spectroscopically.While we improved the census of young stars in the λ and σ Ori regions, the presenceof several strong Ca II and/or H α emitters among spectra obtained of stars not selectedvia the K14 methods suggests that the criteria are—though robust and reliable—not 100%complete in finding the young stars that are present.The population of newly identified and candidate YSOs roughly traces the known areasof active star formation, but we identify a few new ‘hot spots’ of activity near Lynds 1588and 1589. We confirm the more dispersed population of YSOs seen in the northern half ofthe H ii region bubble around σ and (cid:15) Ori surveyed in Brice˜no et al. (2005). External to the λ Orionis ring we note clusters of YSO candidates around the L1617 cloud and the moredistant G192.16 − ii region.We present a minimal spanning tree analysis of the two regions to identify stellar group-ings. We find that roughly two-thirds of the YSO candidates in each region belong to groupsof 5 or more members. Given the likely rate of contamination of the YSO candidate sampleby galaxies and older stars, as described in Section 7.4, we suspect the fraction of YSOs inclusters to be higher than this value. The population of stars selected by WISE outside theMST groupings does contain spectroscopically verified YSOs however, with a local stellardensity as low as 0.5 stars per square degree.
REFERENCES
Alecian, E., Wade, G. A., Catala, C., et al. 2013, MNRAS, 429, 1001Allen, L. E., & Strom, K. M. 1995, AJ, 109, 1379Assef, R. J., Stern, D., Kochanek, C. S., et al. 2013, ApJ, 772, 26Bally, J. 2010, Nature, 466, 928Barrado y Navascu´es, D., Stauffer, J. R., Bouvier, J., Jayawardhana, R., & Cuillandre, J.-C.2004, ApJ, 610, 1064Bayo, A., Barrado, D., Stauffer, J., et al. 2011, A&A, 536, A63B´ejar, V. J. S., Zapatero Osorio, M. R., Rebolo, R., et al. 2011, ApJ, 743, 64Brice˜no, C., Calvet, N., Hern´andez, J., et al. 2005, AJ, 129, 907Briceno, C. 2008, Handbook of Star Forming Regions, Volume I, 838 32 – T a b l e . σ O r i o n i s W I S E Y S O C a nd i d a t e s W I S E n a m e C oo r d i n a t e s J u J H u H K u K w u w w u w w u w u w w C l a ss α E W ( H α ) E W ( C a II ) M e m b e r S o u r ce T y p e R e f e r e n ce R A ( d e g ) D ec l. ( d e g )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag ) ˚A˚A J . − . . − . . . . . . . . . . . . . . . II − . ··············· J . + . . . . . . . . . . . . . . . . . II − . ······ y HA e B e ; B H D ; V i e i r a e t a l. ( ) J . − . . − . . . . . . . . . . . . . . . II − . ··············· J . + . . . . . . . . . . . . . . . . . II − . ··············· J . − . . − . . . . . . . . . . . . . . . II − . ······ y K B r i ce ˜ n o e t a l. ( ) N o t e . — T h i s T a b l e i s pub li s h e d i n i t s e n t i r e t y i n t h ee l ec tr o n i ce d i t i o n o f t h e A s t r oph y s i c a l J o u r n a l . A p o rt i o n i ss h o w nh e r e f o r g u i d a n ce r e ga r d i n g i t s f o r m a nd c o n t e n t . R i g h t a s ce n s i o n a nd D ec li n a t i o n c oo r d i n a t e s a r e J . .
33 – T a b l e . λ O r i o n i s W I S E Y S O C a nd i d a t e s W I S E n a m e C oo r d i n a t e s J u J H u H K u K w u w w u w w u w u w w C l a ss α E W ( H α ) E W ( C a II ) M e m b e r S o u r ce T y p e R e f e r e n ce R A ( d e g ) D ec l. ( d e g )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag )( m ag ) ˚A˚A J . + . . . . . . . . . . . . . . . . . I . ··············· J . + . . . . . . . . . . . . . . . . . II - . ··············· J . + . . . . . . . . . . . . . . . . . II - . ··············· J . + . . . . . . . . . . . . . . . . . I . ··············· J . + . . . . . . . . . . . . . . . . . II - . ··············· N o t e . — T h i s T a b l e i s pub li s h e d i n i t s e n t i r e t y i n t h ee l ec tr o n i ce d i t i o n o f t h e A s t r oph y s i c a l J o u r n a l . A p o rt i o n i ss h o w nh e r e f o r g u i d a n ce r e ga r d i n g i t s f o r m a nd c o n t e n t . R i g h t a s ce n s i o n a nd D ec li n a t i o n c oo r d i n a t e s a r e J . .
34 –Caballero, J. A. 2008, A&A, 478, 667Caballero, J. A., & Solano, E. 2008, A&A, 485, 931Caballero, J. A., Valdivielso, L., Mart´ın, E. L., et al. 2008, A&A, 491, 515Caballero, J. A., Cabrera-Lavers, A., Garc´ıa- ´Alvarez, D., & Pascual, S. 2012, A&A, 546,A59Carpenter, J. M., Snell, R. L., Schloerb, F. P., & Skrutskie, M. F. 1993, ApJ, 407, 657Carpenter, J. M., Mamajek, E. E., Hillenbrand, L. A., & Meyer, M. R. 2009, ApJ, 705, 1646Casertano, S., & Hut, P. 1985, ApJ, 298, 80Cieza, L. A., Schreiber, M. R., Romero, G. A., et al. 2012, ApJ, 750, 157Cody, A. M. 2012, Ph.D. Thesis, California Institute of TechnologyCody, A. M., & Hillenbrand, L. A. 2010, ApJS, 191, 389Cody, A. M., & Hillenbrand, L. A. 2014, ApJ, 796, 129Connelley, M. S., Reipurth, B., & Tokunaga, A. T. 2007, AJ, 133, 1528Connelley, M. S., & Greene, T. P. 2010, AJ, 140, 1214Cunha, K., & Smith, V. V. 1996, A&A, 309, 892Dolan, C. J., & Mathieu, R. D. 1999, AJ, 118, 2409Dolan, C. J., & Mathieu, R. D. 2001, AJ, 121, 2124Dolan, C. J., & Mathieu, R. D. 2002, AJ, 123, 387Gray, R. O., & Corbally, C. J. 1998, AJ, 116, 2530Greene, T. P., Wilking, B. A., Andr´e, P., Young, E. T., & Lada, C. J. 1994, ApJ, 434, 614Gutermuth, R. A., Megeath, S. T., Myers, P. C., et al. 2009, ApJS, 184, 18Herbig, G. H., & Bell, K. 1988, LicOB, 1111, 90Hern´andez, J., Morales-Calderon, M., Calvet, N., et al. 2010, ApJ, 722, 1226Hsieh, T.-H., & Lai, S.-P. 2013, ApJS, 205, 5 35 –Jacoby, G. H., Hunter, D. A., & Christian, C. A. 1984, ApJS, 56, 257Kajdiˇc, P., Reipurth, B., Raga, A. C., Bally, J., & Walawender, J. 2012, AJ, 143, 106Koenig, X. P., Leisawitz, D. T., Benford, D. J., et al. 2012, ApJ, 744, 130Koenig, X. P., & Leisawitz, D. T. 2014, ApJ, 791, 131Kwok, S., Volk, K., & Bidelman, W. P. 1997, ApJS, 112, 557Lang, W. J., Masheder, M. R. W., Dame, T. M., & Thaddeus, P. 2000, A&A, 357, 1001Lee, H.-T., & Chen, W. P. 2007, ApJ, 657, 884Lodieu, N., Zapatero Osorio, M. R., Rebolo, R., Mart´ın, E. L., & Hambly, N. C. 2009, A&A,505, 1115Lombardi, M., Alves, J., & Lada, C. J. 2011, A&A, 535, A16Luhman, K. L., Hern´andez, J., Downes, J. J., Hartmann, L., & Brice˜no, C. 2008, ApJ, 688,362Luhman, K. L., & Mamajek, E. E. 2012, ApJ, 758, 31Manoj, P., Bhatt, H. C., Maheswar, G., & Muneer, S. 2006, ApJ, 653, 657Mathieu, R. D. 2008, Handbook of Star Forming Regions, Volume I, 757Megeath, S. T., Gutermuth, R., Muzerolle, J., et al. 2012, AJ, 144, 192Nesterov, V. V., Kuzmin, A. V., Ashimbaeva, N. T., et al. 1995, A&AS, 110, 367Ochsendorf, B. B., Cox, N. L. J., Krijt, S., et al. 2014, A&A, 563, AA65Oke, J. B., & Gunn, J. E. 1982, PASP, 94, 586Pe˜na Ram´ırez, K., B´ejar, V. J. S., Zapatero Osorio, M. R., Petr-Gotzens, M. G., & Mart´ın,E. L. 2012, ApJ, 754, 30Rebull, L. M., Koenig, X. P., Padgett, D. L., et al. 2011, ApJS, 196, 4Reid, I. N., Cruz, K. L., & Allen, P. R. 2007, AJ, 133, 2825Riaz, B., Thompson, M., Whelan, E. T., & Lodieu, N. 2015, MNRAS, 446, 2550Rieke, G. H., & Lebofsky, M. J. 1985, ApJ, 288, 618 36 –Rigliaco, E., Natta, A., Testi, L., et al. 2012, A&A, 548, A56Rivinius, T., Carciofi, A. C., & Martayan, C. 2013, A&A Rev., 21, 69Rizzuto, A. C., Ireland, M. J., & Zucker, D. B. 2012, MNRAS, 421, L97Sacco, G. G., Franciosini, E., Randich, S., & Pallavicini, R. 2008, A&A, 488, 167Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500, 525Sherry, W. H., Walter, F. M., Wolk, S. J., & Adams, N. R. 2008, AJ, 135, 1616Shiozaki, S., Imai, H., Tafoya, D., et al. 2011, PASJ, 63, 1219Skiff, B. A. 2014, VizieR Online Data Catalog, 1, 2023Siess, L., Dufour, E., & Forestini, M. 2000, A&A, 358, 593Strom, K. M., Strom, S. E., Edwards, S., Cabrit, S., & Skrutskie, M. F. 1989, AJ, 97, 1451Su´arez, O., Garc´ıa-Lario, P., Manchado, A., et al. 2006, A&A, 458, 173Vieira, S. L. A., Corradi, W. J. B., Alencar, S. H. P., et al. 2003, AJ, 126, 2971Wainscoat, R. J., Cohen, M., Volk, K., Walker, H. J., & Schwartz, D. E. 1992, ApJS, 83,111Walter, F. M., Sherry, W. H., Wolk, S. J., & Adams, N. R. 2008, Handbook of Star FormingRegions, Volume I, 732West, A. A., Morgan, D. P., Bochanski, J. J., et al. 2011, AJ, 141, 97Zapatero Osorio, M. R., B´ejar, V. J. S., Pavlenko, Y., et al. 2002, A&A, 384, 937