Low-metallicity Young Clusters in the Outer Galaxy. III. Sh 2-127
Chikako Yasui, Naoto Kobayashi, Masao Saito, Natsuko Izumi, Warren Skidmore
aa r X i v : . [ a s t r o - ph . GA ] F e b Draft version February 11, 2021
Preprint typeset using L A TEX style emulateapj v. 5/2/11
LOW-METALLICITY YOUNG CLUSTERS IN THE OUTER GALAXY. III. Sh 2-127
Chikako Yasui , Naoto Kobayashi
2, 3, 4 , Masao Saito
5, 6 , Natsuko Izumi
7, 8 , and Warren Skidmore National Astronomical Observatory of Japan, California Office, 100 W. Walnut St., Suite 300, Pasadena, CA 91124, [email protected] Institute of Astronomy, School of Science, University of Tokyo, 2-21-1 Osawa, Mitaka, Tokyo 181-0015, Japan Kiso Observatory, Institute of Astronomy, School of Science, University of Tokyo, 10762-30 Mitake, Kiso-machi, Kiso-gun, Nagano397-0101, Japan Laboratory of Infrared High-resolution spectroscopy (LIH), Koyama Astronomical Observatory, Kyoto Sangyo University, Motoyama,Kamigamo, Kita-ku, Kyoto 603-8555, Japan National Astronomical Observatory of Japan, 2-21-1 Osawa, Mitaka, Tokyo 181-8588, Japan The Graduate University of Advanced Studies (SOKENDAI), 2-21-1 Osawa, Mitaka, Tokyo 181-8588, Japan College of Science, Ibaraki University, 2-1-1 Bunkyo, Mito, Ibaraki 310-8512, Japan Institute of Astronomy and Astrophysics, Academia Sinica, No. 1, Section 4, Roosevelt Road, Taipei 10617, Taiwan Thirty Meter Telescope International Observatory, 137 W Walnut Ave, Monrovia, CA 91016, USA
Draft version February 11, 2021
ABSTRACTIn deep near-infrared imaging of the low-metallicity ([O / H] = − . II region Sh 2-127(S127) with Subaru/MOIRCS, we detected two young clusters with 413 members (S127A) in a slightlyextended H II region and another with 338 members (S127B) in a compact H II region. The limitingmagnitude was K = 21 . σ ), corresponding to a mass detection limit of ∼ M ⊙ . Theseclusters are an order of magnitude larger than previously studied young low-metallicity clusters andlarger than the majority of solar neighborhood young clusters. Fits to the K -band luminosity functionsindicate very young cluster ages of 0.5 Myr for S127A and 0.1–0.5 Myr for S127B, consistent withthe large extinction (up to A V ≃
20 mag) from thick molecular clouds and the presence of a compactH II region and class I source candidates, and suggest that the initial mass function (IMF) of thelow-metallicity clusters is indistinguishable from typical solar neighborhood IMFs. Disk fractions of28% ±
3% for S127A and 40% ±
4% for S127B are significantly lower than those of similarly aged solarneighborhood clusters ( ∼ <
30 %), probably due to S127B’s age. This suggests that a largefraction of very young stars in low-metallicity environments have disks, but the disks are lost on a veryshort timescale. These results are consistent with our previous studies of low-metallicity star-formingregions, suggesting that a solar neighborhood IMF and low disk fraction are typical characteristics forlow-metallicity regions, regardless of cluster scales.
Subject headings:
Galaxy: abundances — infrared: stars — open clusters and associations: general— planetary systems: protoplanetary disks — stars: formation — stars: pre-main-sequence — ISM: H II regions INTRODUCTION
Considering that the only elements in existence at thebeginning of our universe were H and He (and an in-significant amount of Li), and that the present chemicalcomposition of the universe is the result of metal pollu-tion due to elemental synthesis inside stars and the influ-ence of supernovae explosions, exploring the physical pro-cesses involved with star formation under different metal-licity environments is of great interest. Although even atpresent, metals only account for 2 % of the mass in oursolar system and the local universe, they can be criti-cal factors for star and planet formation; metals stronglyaffect heating and cooling related to radiative transferin star formation processes. Also in planet formationprocesses, dust forms planet cores despite being only avery small mass fraction in disks (only ∼ ≥
1, we seethe early universe, where metallicity is lower than at present. Even in our Galaxy, there is a wide range ofmetallicity, from − ∼ M ⊙ ). This enables us to compare results obtainedfor low-metallicity environments with those in the solarneighborhood on the same basis. Therefore, we are fo-cusing on our Galaxy, the only region where observationsdown to the substellar mass regime can be conductedwith existing telescopes and instruments. Although theLarge Magellanic Cloud (LMC; D ∼
50 kpc) and SmallMagellanic Cloud (SMC; D ∼
620 kpc) are well known asextragalactic objects that are relatively close to the Sun,the mass detection limit can only reach about 1 M ⊙ withcurrent facilities due to the larger distances compared towithin our Galaxy ( D ∼ ∼ ∼− ∼− z ∼ ∼ M ⊙ ):Digel Cloud 2 (Yasui et al. 2006, 2008b), S207 (Yasui etal. 2016b), and S208 (Yasui et al. 2016a). The first ex-ample is the Cloud 2 clusters, which are associated witha molecular cloud (Digel et al. 1994) located in the ex-treme outer Galaxy with a Galactrocentric distance ( R G )of ≃
19 kpc in the direction of ( l, b ) = (137 . , − . ∼− ∼ M ⊙ and estimated the ages as ∼ α -selected bright H II regions; Sharpless 1959), ii) the region associated withclusters having a significant number of cluster members,and iii) an oxygen metallicity [O / H] ≤ − . / H) = 8 .
73 (As-plund et al. 2009). Among ∼
10 selected regions, we firstpresented the results of S207 (Yasui et al. 2016b, here-after Paper I) and S208 (Yasui et al. 2016a, hereafterPaper II), which are two of the lowest-metallicity H II regions (each with [O / H] ≃ − l, b ) = (151 . , .
13) in Galactic coordinates. With NIRdeep imaging, we identified one cluster in each region,having 73 and 89 cluster members in S207 and S208, re-spectively, with a mass detection limit of . M ⊙ forS207 and ∼ M ⊙ for S208. From the K -band lumi-nosity function (KLF) fitting of the clusters, S207 andS208 are likely to be located at D = 4 kpc from theSun, which suggests that two regions are located in theinterarm region between the Cygnus and Perseus arms(Vall´ee 2020). The fitting also suggested that S207 isat the end of the embedded cluster phase ( ∼ ∼ N stars ) < N stars > N stars > and more common clusters with N stars ∼ (see Sec-tion 5.2), and the disk fraction in more massive or denserclusters is suggested to be low (see Section 5.3).In this paper, we present the results for our third tar-get, Sh 2-127 (S127), which is a low-matallicity star-forming region in the Galaxy with [O / H] ≃ − . ∼ N stars ), more than an order of magnitude higherthan in previous targets. This paper is organized as fol-lows. Section 2 describes previous studies of S127, fo-cusing on studies of star-forming activities in S127 us-ing multiwavelength data, e.g., mid-infrared (MIR) datafrom the Wide-field Infrared Survey Explorer ( WISE ),NIR data from the Two Micron All Sky Survey (2MASS),H α data from the Isaac Newton Telescope PhotometricH-Alpha Survey (IPHAS), and radio continuum emissionfrom the NRAO VLA Sky Survey (NVSS). Section 3 de-scribes our Subaru Multi-Object InfraRed Camera andSpectrograph (MOIRCS) deep JHK S images and datareduction. Section 4 describes the results for the star-forming clusters in S127. In Section 5, we discuss thebasic cluster parameters, such as cluster scale, age, IMF,and disk fraction, of the S127 clusters. Finally, in Sec-tion 6, we discuss the IMF the low-metallicity environ-ment. S127
In this section, the properties of the target star-formingregion, S127, are summarized. In Table 1, we summa-rize the properties from previous works, including co-ordinates, distance, oxygen abundance, and metallicity.We also show the large-scale NIR and MIR pseudocolorand H α images of S127 in Figure 1. Basic Properties from the Literature
The region S127 is located at ( l, b ) =(96 . ◦ , +2 . ◦ ) on the Galactic plane with coor-dinates of ( α . , δ . ) = (21 h m . s , +54 ◦ ′ ′′ )from the SIMBAD database (Wenger et al. 2000). Ithas an extended H II region traced by H α (Sharpless1959) and radio continuum (Fich 1993) emission. Strong This research has made use of the SIMBAD database, op-erated at the Centre de Donn´ees Astronomiques de Strasbourg,France.
MIR emission is detected with IRAS, IRAS 21270+5423in the IRAS Point Source Catalog (Beichman et al.1988; see the large red plus sign in Figure 1) andIRAS X2127+544 in the IRAS Small-Scale StructureCatalog (Helou & Walker 1988). CO emission isreported in e.g., Blitz et al. (1982) and Wouterloot& Brand (1989), and the results of high-resolutionCO observations are shown in Brand et al. (2001). Astar-forming cluster is identified by Bica et al. (2003)as [BDS2003]24 using 2MASS images, with a centerof ( α . , δ . ) = (21 h m s , +54 ◦ ′ ′′ ) andangular dimensions of 1 . ′ × . ′
1. The photometricdistance, which is determined from spectroscopic andphotometric observations, is estimated to be 9.7 kpcfor the O8V-type star (Chini & Wink 1984), ALS18695 (see the small red plus sign in Figure 1). Thekinematic distance is also estimated at ≃
10 kpc usingthe radial velocities of V LSR ∼ −
95 km s − derived invarious observations: V LSR = − . − for COobservations by Blitz et al. (1982), V LSR = − . − for the H α Fabry–Perot observation by Fich et al.(1990), V LSR = − .
09 km s − for CO observationsby Wouterloot & Brand (1989), V LSR = − . − for Fabry–Perot observations of the H II region byCaplan et al. (2000), and V LSR = − . − − for CO line data by FCRAO and H I data byCGPS, respectively (Foster & Brunt 2015). Accordingto Foster & Brunt (2015), the most recent derivation,the estimated distance is 9 . ± .
73 kpc, which isconsistent with the kinematic distance. Assuming thatthe Galactocentric distance of the Sun is R ⊙ = 8 . R G ≃ . II regions by measuring optical emission linefluxes in spectroscopic observations, while Caplan et al.(2000) measured optical emission line fluxes for 36 H II regions based on Fabry–Perot observations. Rudolphet al. (1997) estimated the abundance for five H II re-gions by measuring FIR emission line fluxes with theKuiper Airborne Observatory, while Peeters et al. (2002)measured line fluxes between 2.3 and 196 µ m mainly forIRAS point sources in 45 (compact) H II regions basedon Infrared Space Observatory (ISO) spectroscopy. Rudolph et al. (2006) reanalyzed the elemental abun-dances of 117 H II regions with updated physical param-eters. Among them, the oxygen abundances of S127 areestimated to be 12 + log(O / H) = 7 . . +0 . − . ,8 . +0 . − . , 8 . +0 . − . , 7 . +0 . − . , and 8 . +0 . − . , using thedata by Vilchez & Esteban 1996, Rudolph et al. 1997, Deharveng et al. (2000) subsequently derived the abundancesusing data presented by Caplan et al. (2000). Mart´ın-Hern´andez et al. (2002) subsequently derived the oxy-gen abundance using data by Peeters et al. (2002).
Caplan et al. 2000, Peeters et al. 2002, and Rudolph etal. 2006, respectively). This corresponds to a metal-licity of [O / H] ≃ − . / H) = 8 .
73 (Asplund et al. 2009).The electron temperatures ( T e ) are also sensitive indi-cators of the abundances, with higher temperatures forlower abundances (Shaver et al. 1983). The estimatedtemperatures are very high for S127, ∼ ±
820 K from Scaife et al. (2008) and 11428 ±
305 Kfrom Balser et al. (2011). They are some of the highesttemperatures among H II regions in our Galaxy, sug-gesting that S127 is a very low-metallicity region. Ac-cording to the relationship between the electron temper-atures and oxygen abundances by Shaver et al. (1983)of 12 + log (O / H) = 9 . − . T e / , the temperatureof S127 ( ∼ Star-forming Activities
The top panel of Figure 1 shows an NIR and MIR pseu-docolor image of S127 with a wide field of view (5 ′ × ′ )centered at IRAS 21270+5423, shown with the largered plus sign. The figure is produced by combining the2MASS (Skrutskie et al. 2006) K S -band (2.16 µ m; blue), WISE (Wright et al. 2010) band 1 (3.4 µ m; green), and WISE band 3 (12 µ m; red) images. Two components canbe seen to the north and south of the IRAS source (largered plus sign) in WISE band 3, whose emission is mainlyfrom polycyclic aromatic hydrocarbon (PAH) emission,tracing photodissociation regions around H II regions.The bottom panel of Figure 1 shows an H α image fromIPHAS (Drew et al. 2005) in gray scale and the 1.4 GHzradio continuum from NVSS (Condon et al. 1998) in bluecontours. The overall distribution of H α and radio con-tinuum emissions that trace the photoionized H II regionis consistent with that of 12 µ m features. There are twoNVSS radio sources in this field, NVSS 212841+543634and NVSS 212843+543728, indicated by blue diamonds,which are located in the centers of the northern andsouthern components, but the distribution of the radiocontinuum is not divided into two components at thenorth and south of the IRAS source due to the relativelylow spatial resolution of NVSS (45 ′′ ). The images of ra-dio continuum emissions with higher spatial resolution(5 . ′′
5) by Rudolph et al. (1996) show the two compo-nents, and there are two NVSS radio sources in this field,NVSS 212841+543634 and NVSS 212843+543728, indi-cated by blue diamonds, which are located in the centerof the northern and southern component. Rudolph et al.(1996) refer to northern and southern regions as S127Aand S127B, respectively. They estimated the propertiesof the H II regions (S127A and S127B) separately. Theirresults showed that S127A is an extended H II region(1.7 pc diameter), while S127B is a compact H II region(0.5 pc diameter). Rudolph et al. (1996) also pointed outthat the two regions have cometary shapes, which maybe due to pressure confinement of the expanding ionized Rudolph et al. (2006) collected some estimations from severalindependent studies for various positions: 8 . +0 . − . and 8 . +0 . − . both within S127, 8 . +0 . − . for S127A, 8 . +0 . − . for S127B, and7 . +0 . − . for IRAS 21270+5423. Yasui et al.gas, the so-called champagne flow. Brand et al. (2001)presented high-resolution CO observations. They showednorthern and southern CO complexes around the S127Aand S127B H II regions. The distribution of CO emis-sions correlates fairly well with that of radio continuumemissions. The northern CO complex around S127A ob-scures most of the optical emission, suggesting that theCO complex is located in the foreground. This is alsoseen as reduced emission (the gray area within S127A)in the H α image (bottom panel of Figure 1) around thenorthern NVSS source.The compact H II region (S127B) is located at theeastern edge of a patch of obscuration, which correspondsto a peak of the southern CO complex. This suggeststhat the H II region lies on the near side of the molecularcomplex. This is also seen in the H α image (bottompanel of Figure 1) to the west of the southern NVSSsource marked by the abrupt edge. Because ALS 18695is located around the center of the southern H II region inthe vicinity of the southern NVSS source, the O-type starshould be the exciting source of the southern H II region.The spectral type of ALS 18695 (O8V) is consistent withthe results of Rudolph et al. (1996) that the spectraltype of a single zero-age main-sequence (ZAMS) star thatcould provide a flux of ionizing photons of the S127B H II region would be O8.5. OBSERVATIONS AND DATA REDUCTION
Subaru MOIRCS JHK Imaging
Using the same instrumental setup described in PapersI and Paper II, MOIRCS (Ichikawa et al. 2006; Suzuki etal. 2008) was used on the 8.2 m Subaru telescope to ob-tain deep
JHK S images with the Mauna Kea Observato-ries (MKO) NIR filters (Simons & Tokunaga 2002; Toku-naga et al. 2002) over a 3 . ′ × ′ field at 0 . ′′
117 pixel − .The long-exposure observations described in this pa-per were performed on 2006 September 2 UT. Observingconditions were photometric, and the seeing was excel-lent ( ∼ . ′′ . ′′
45) throughout the night. For the long-exposure images, individual exposure times for the J , H ,and K S bands were 150, 20, and 30 s, respectively, whilethe total integration times are 1350, 1080, and 1080 sfor the J , H , and K S bands, respectively. For countsover 20,000 ADU, the detector output linearity is notguaranteed. To ensure accurate flux calibration for thebrightest targets in the cluster, we also obtained short-exposure images on 2007 November 22 UT. The expo-sure time for individual short-exposure images is 13 s,and the total integration time is 52 s for all bands. Thewhole H II region described in Section 2.2 is covered byone chip ( ∼ . ′ × . ′
0, hereafter the “S127 frame”; seethe white and black boxes in Figure 1), whose center isat α = 21 h m . s , δ = +54 ◦ ′ . ′′
2. For thebackground subtraction, to avoid the nebulosity of S127,the telescope was nodded by 3 . ′ ′′ dithering, the two chips being alternately directedto the field, continuously observing the field. After thelong-exposure images were obtained, one of the MOIRCSscience grade detectors was exchanged for an engineeringgrade detector. Although this engineering grade detectorwas used while gathering short-exposure images, we onlyincluded observations obtained with the science grade de- tector in our analysis. The 3 . ′ Data Reduction and Photometry
All data were reduced with IRAF using the proce-dure described in Papers I and II: flat fielding, bad-pixel correction, median-sky subtraction, image shiftswith dithering offsets, and image combination. For theflat fielding, sky flats were used that were made withMOIRCS data in the closest run obtained from theSMOKA data archive. Before any image combination,the MOIRCS image reduction package MCSRED wasused to correct for image distortion using the process de-scribed in Papers I and II. Figure 3 shows a pseudocolorimage of S127 constructed by combining the J (1.26 µ m;blue), H (1.64 µ m; green), and K S (2.15 µ m; red) long-exposure images.For the long-exposure images, photometry with pointspread function (PSF) fitting using IRAF/DAOPHOTwas performed. For deriving PSFs, we selected unsatu-rated bright stars where the highest pixel count was be-low the nonlinear sensitivity regime (20,000 ADU) thatwere not close to the edge of the frame and do not haveany nearby stars with magnitude differences of more than5 mag. The PSF photometry was performed using theALLSTAR routine with two iterations, once using theoriginal images and a second time using the images withsources from the first iteration subtracted. We used PSFfit radii of 3.5, 3.75, and 3.75 pixels for the J , H , and K bands, which are the PSF FWHM values, and set the in-ner radii and width of the sky annulus as four and threetimes as large as the PSF fit radii, respectively. Pers-son 9166 (GSPC P330-E; J = 11 . H = 11 . K = 11 .
419 mag), which is an MKO standard (Leggettet al. 2006), was used for the photometric calibration.The limiting magnitudes (10 σ ) based on the pixel-to-pixel noise of long-exposure images for the S127 frameare J = 22 . H = 21 .
2, and K S = 21 . J = 22 . H = 21 .
2, and K S = 21 . J . H . .
5, and K S .
15 mag are saturated in long-exposure images. Forthe photometry of such bright stars, the short-exposureimages are used. The photometry was performed withthe same procedure as for long-exposure images but usinga PSF fit radii of 7 pixels, which is the PSF FWHM value,and setting the inner radii and width of the sky annulusto four and three times as large as the PSF fit radii, re-spectively. For the photometric calibration, stars whosemagnitudes can be estimated in both short- and long- IRAF is distributed by the National Optical Astronomy Ob-servatories, which are operated by the Association of Universitiesfor Research in Astronomy, Inc., under cooperative agreement withthe National Science Foundation. SMOKA is the Subaru–Mitaka–Okayama–Kiso Archive Sys-tem operated by the Astronomy Data Center, National Astronom-ical Observatory of Japan. exposure images with small uncertainties (magnitudes of J < . H <
17, and K S <
17 mag, and magnitudeuncertainties of < J . H .
14, and K S . . TWO YOUNG EMBEDDED CLUSTERS IN S127
Identification of Young Clusters in S127
Using the pseudocolor image (Figure 3), enhancementsof the stellar density compared to the surrounding areawere identified in the center of the field observed withMOIRCS. The enhancements are located near the re-gions where the emission of
WISE band 3 (12 µ m) isvery strong (Figure 1), which is often the case for youngclusters (see Koenig et al. 2012).We determined the contour map of stellar density inthe frame by counting the number of stellar sources de-tected in the K S band that are included within circleswithin a 50 pixel ( ∼ ′′ ) radius. The circles are set all overthe frame with 25 pixel steps. Because the stellar densityis very high in the center of the frame, we derived thebackground level from outside of the cluster area, shownwith a cyan ellipse in Figure 4. The average number ofstars in each of the circles is 13 . ± .
1. Figure 4 showsthe distribution of detected sources in the MOIRCS K S -band image with contour levels of 3 σ , 4 σ , ..., 20 σ . Fromthe map, there are two large components that are locatedin the north and south of the IRAS source. The peak co-ordinates are α = 21 h m . s , δ = +54 ◦ ′ ′′ and α = 21 h m . s , δ = +54 ◦ ′ ′′ with anaccuracy of ∼ ′′ . Both stellar enhancement peaks arelocated very close to the peaks of the NVSS radio sources,shown with blue diamonds, and the distributions of theclusters are consistent with the distribution of H II re-gions observed by Rudolph et al. (1996) for both S127Aand S127B (Section 2.2). The obscurations by molecularcomplex, pointed out by Brand et al. (2001), were seenin the northern half of the S127A cluster and the westof the S127B cluster. We defined the cluster regions, en-closed with green polygons in the figure, as regions withstellar densities 3 σ larger compared to that of the entireframe. The sizes of the two clusters are ∼ . ′ × . ′ ∼ . ′ × . ′ ∼ ′ × ′ , whichis consistent with the cluster size of angular dimensionsof 1 . ′ × . ′ ∼ × D = 10 kpc. We used the full sky frame as a controlfield. The control field is used for subtracting the con-tamination of field objects in the following discussion. Color–Magnitude Diagram
For our S127 measurements, we followed the same in-vestigation process as described in Paper II for S208.Figure 5 shows the J − K S versus K S color–magnitudediagrams for all detected point sources in the S127A(left panel) and S127B (right panel) cluster regions. Wealso plotted detected point sources in the control fieldin Figure 6. The isochrone models for the ages of 1 Myr are shown as blue lines. The models are by Leje-une & Schaerer (2001) for the mass of M/M ⊙ ≥ < M/M ⊙ ≤ . ≤ M/M ⊙ ≤
3, and a distance of 10 kpc is assumed.Arrows show the reddening vector of A V = 5 mag. In thecolor–magnitude diagram, the extinction A V of each starwas estimated from the distance between its location andthe A V = 0 isochrone models along the reddening vec-tor. Figure 7 shows the distributions of the extinctionof stars in the cluster region (thick lines) and controlfield (thin lines) for the S127A and S127B clusters in theleft and right panels, respectively. The distribution forthe control field is normalized to match with the totalarea of the cluster regions. The resultant distributionfor the control field shows a peak of A V = 1–3 mag,and then the number count decreases with larger A V ,whereas that for the cluster region shows a peak at themuch larger extinction of A V ∼ A V ∼
20 mag. This suggests that stars with A V ≥ . A V < . A V , cluster members can be distinguished from con-taminating noncluster stars that appear in the cluster re-gion, as is the case with the S208 clusters (Paper II). Thefollowing criteria are applied to identify members of theS127 clusters: the stars (1) are distributed in the clusterregions and (2) have large A V excess compared with nor-mal field stars (extinction of A V ≥ N cl ) areidentified as S127A and S127B cluster members, respec-tively. The average A V value of the cluster members isestimated at A V = 8 . ± . A V = 6 . ± . R g of the S127 clus-ters ( R g = 13 . K = 21 . A V distributions of all the sources in the clus-ter regions and the field objects in the control field (Fig-ure 7). Because the number of field objects in the controlfield decreases significantly at A V ≥ A V ≥ A V ≥ A V < N ′ cl − N ′ fi ) / ( N cl +( N ′ cl − N ′ fi )),where N ′ fi is the normalized number of field objects with A V < N ′ cl is the number of stars in the clusterregion with A V < N cl is the number of identified clustermembers. The fractions of cluster members missed forthe S127A and S127B clusters are estimated at 3 % and Yasui et al.11 %, respectively.On the isochrone models in Figure 5, the positions of0.1, 1, 3, 5, 10, 20, 40, and 60 M ⊙ are shown with hor-izontal lines. With the average A V for all S127 clustermembers of ∼ K -band limiting magnitude of21.3 mag (10 σ ) for an age of 1 Myr corresponds to amass of 0.2 M ⊙ assuming a distance of D = 10 kpc.The mass detection limit is sufficiently low, close to thesubstellar mass limit, which enables by KLF fitting toderive parameters describing the IMF down to aroundthe IMF peak (Section 5.2) and to derive the disk frac-tion with the same criteria as in the solar neighborhood(Section 5.3). Because the most likely age of the S127cluster is estimated at ∼ ∼ M ⊙ with the average A V of ∼ Color–Color diagram
We constructed J − H versus H − K S color–color di-agrams for stars in the S127 cluster regions (Figure 8).Cluster members identified in Section 4.2 are shown inred, while sources in the cluster regions but not iden-tified as cluster members are shown in black. We alsoconstructed the color–color diagram for stars in the con-trol field (Figure 9). All sources that are detected atmore than 10 σ in all JHK bands are plotted. The dwarfstar track for spectral types from late B to M6 in theMKO system by Yasui et al. (2008b) is shown with theblue curve. The classical T Tauri star (CTTS) locus,originally derived by Meyer et al. (1997) in the CIT sys-tem, is shown as a cyan line in the MKO system (Yasuiet al. 2008b). The arrow shows the reddening vector of A V = 5 mag.Stars in star-forming regions sometimes show large H − K color excesses due to their circumstellar dust disks,in addition to large extinctions (e.g., Lada & Adams1992). We estimated the color excesses for each sourceusing the procedure described in Papers I and II. First,the intrinsic ( H − K ) colors (( H − K ) ) were estimated bydereddening along the reddening vector to the young starlocus in the color–color diagram (see Figure 8), whichwas approximated by the extension of the CTTS locus.Only stars that are above the CTTS locus were used.The obtained ( H − K ) distributions for the S127 clus-ter members and those in the control field are shown inFigure 10. After normalizing the distribution for the con-trol field to the total area of the cluster regions, there isa larger number of red stars with ( H − K ) > . H − K ) valuesfor cluster members are estimated at 0.39 mag for boththe S127A and S127B clusters, whereas that in the con-trol field is estimated at 0.14 mag. The difference in theaverage ( H − K ) between the stars in the cluster regionand the field stars ( ≃ K band, ∆ K disk , as0.25 mag, assuming that disk emissions appear in the K but not in the H band. KLF of the S127 clusters
The KLF for the S127 cluster members is shown in Fig-ure 11 as a black line up to the K = 20 . K = 13 . K = 18 . σ detection mag-nitude for the S127 frame is K = 21 . ∼ K = 18 . A V dispersion of the S127 clus-ters ( A V ∼ A V values in Figure 11, A V = 4 . A V distribution of the cluster members,8 . ± . . ± . A V samplesare vertically shifted by +0.1 mag for both clusters. Asa result, the discrepancy between the KLFs for all clus-ter members and those for stars with limited A V valuesis found to be within the uncertainty range, suggestingthat the selection of stars with different limited A V val-ues causes a negligible influence on the obtained KLF.Therefore, we used the original KLF (the KLF from allS127 cluster members) in the following discussion. DISCUSSION
Scale of the clusters
Adams et al. (2006) found a clear correlation betweencluster size and the number of cluster members for youngclusters in the solar neighborhood from their embeddedstage up to ages of ∼
10 Myr. Figure 12 shows the num-ber of stars in a cluster versus cluster radius by opensquares from the compilation of clusters in Lada & Lada(2003) and Carpenter (2000). The figure shows thatmost clusters have ∼ N stars ) andradii ( R ) of ∼ R ( N stars ) = R p N stars /
100 with R = √ √ R , shown with dot-ted lines.In Section 4, we identified two clusters in S127, theS127A and S127B clusters. The S127A cluster has 413cluster members in a region of ∼ . ′ × . ′
8, correspondingto a cluster radius of ∼ . ′
6, while the S127B cluster has338 cluster members in ∼ . ′ × . ′
6, corresponding to acluster radius of ∼ . ′
4. Because 1 ′ corresponds to 3 pc atthe 10 kpc distance of S127, the cluster radii of the S127Aand S127B clusters correspond to 1.8 and 1.2 pc, respec-tively. We plot the values in Figure 12 with red filledcircles. The plots show that the density of the S127Bcluster is a little higher than that of the S127A cluster.Both clusters have both R and N stars that are within therange found for clusters in the solar neighborhood but atthe upper end of that range.We also plotted properties for other young low-metallicity clusters (Section 1) with red open circles:S207, S208, and the Cloud 2-N and -S clusters (Papers I,II; Yasui et al. 2008b). They are relatively small clusterswith N stars of less than 100, even though the mass detec-tion limits are similar to the limit for S127 of . M ⊙ .The N stars is 73, 89, 52, and 59 for S207, S208, Cloud 2-N, and Cloud 2-S, respectively, while the detection limitsare . M ⊙ for S207, ∼ M ⊙ for S208, and ∼ M ⊙ for Cloud 2-N and -S. The cluster radii for all clustersare ∼ R ) of the S127 clusters are comparable to the radiiof those clusters. However, the numbers of cluster mem-bers ( N stars ) of the S127 clusters are larger by aboutan order of magnitude, compared to other young low-metallicity clusters.In the following sections, we estimate the underlyingIMF and disk fractions of the S127 clusters. Becausethe S127 clusters are the first large-scale targets in low-metallicity environments detected down to the substellarmass regime, they are very appropriate targets to exam-ine whether results from previous young low-metallicityclusters hold even in such large clusters. If distinguishedbetween metallicity dependence and dependence on clus-ter scales (size and number of members), genuine metal-licity dependencies can be derived. In fact, the possibilitythat the IMF and disk fraction depend on cluster scalesis suggested for clusters in the solar neighborhood, i.e.a dependence on cluster mass and cluster density (seeSections 5.2 and 5.3). The S127 clusters are also usefulfor examining whether suggested dependencies on clusterscales seen for clusters in the solar neighborhood hold forclusters in low-metallicity environments. Implication for the IMF and age
Although our final goal is to derive the IMF in low-metallicity environments, information about the age ofthe cluster is necessary for the derivation. Here we esti-mate ages by assuming the canonical IMF observed in thesolar neighborhood as a first step, examine the adequacyof the estimated age, and finally develop constraints onthe underlying IMF. We performed fitting of the KLF,which is known to strongly depend on age and IMF, inthe same way as Papers I and II. We note that we useobserved KLFs in Section 4.4 for the fitting. Althoughit would be ideal if all of the detected sources were cor-rected for extinctions derived in Section 4.2, this cannotbe possible because not all stars are detected at morethan one band, and at least two bands of data are nec-essary for the derivation. Moreover, especially for clus-ters with large extinctions, such as S208, the longer NIRwavelength can detect lower mass stars due to the smallerinfluence of extinctions, and that observed KLF is mostappropriate for deriving the IMF only from photometricdata (Muench et al. 2002). Therefore, we consider A V and ∆ K excess values by inputting them into model KLFsinstead. The method was originally developed in Muenchet al. (2002) and simplified in Yasui et al. (2006). We con-structed model KLFs with ages from 0.1 to 3 Myr, whichare shown with colored lines in Figure 13. A distance of10 kpc is assumed, and A V and ∆ K excess , which are es-timated in Sections 4.2 and 4.3, are applied ( A V = 8 . K excess = 0 .
25 mag). The model KLF for 0.1 Myr has a shallower slope of bright magnitudes be-fore the peak than the KLF for 0.5 Myr. The KLFs for0.1–0.5 Myr have brighter peak magnitudes ( K = 18 . K = 19 . K = 20 . K = 18 . K = 13 . ∼ A V values for S127A and S127B cluster members estimatedin Section 4.2, up to ∼
20 mag with A V distributionsof the cluster members of 8 . ± . . ± . CO at the peak positionin S127B corresponds to A V ≈ ∼ II region (Section 2.2). Compact H II regions arethought to be a very early phase of H II regions, the nextstage after the ultracompact H II phase (Habing & Israel1979). Because the lifetime of ultracompact H II regionsis estimated to be . II regions is estimated to be ∼ II region with a diameter of 1.7 pc, is olderthan the estimated age of the S127B cluster located ina compact H II region with a diameter of 0.5 pc. Thisis consistent with older H II regions having larger radii(Dyson & Williams 1980). In addition, this is consistentwith the results for S207 and S208 in Papers I and II:the age of the S207 cluster located in an H II region witha diameter of 2.6 pc is estimated to be ∼ II region with adiameter of 1.4 pc is estimated to be ∼ ∼ column density and the size of theH II regions. This suggests that the IMF of the S127clusters, which are in a low-metallicity environment, isconsistent with the typical IMF in solar metallicity re-gions for masses ≥ M ⊙ . Because the KLF slope forbright stellar magnitudes before the peak strongly de-pends on the slope of the higher-mass region of the IMF(Muench et al. 2000, 2002), the very good fit to the KLF Yasui et al.peak also suggests that the higher-mass IMF slope in theS127 clusters is consistent with the typical IMF. As forthe KLF peaks, their magnitudes of K = 18 . ≃ M ⊙ .This is also consistent with the canonical IMF within themargin of error, log M c /M ⊙ ∼ . ± . ∼ ∼ M ⊙ ),the cluster mass is roughly estimated on the very sim-ple assumption that all stars have a mass of 1 M ⊙ (e.g.,Yasui et al. 2008b). With this assumption, the clustermasses of the S127 clusters are estimated as ∼ M ⊙ , while those for the other young low-metalicity clus-ters are < M ⊙ . Here we suggested that the IMFsfor clusters with low-metallicity environments are con-sistent with the typical IMFs observed in clusters withsolar metallicity regardless of cluster mass between < ∼ M ⊙ . This is in the same manner as for clus-ters with solar metallicity environments; there seem to beno clear indications that the IMF for clusters with ≤ M ⊙ depends on their cluster scale, while the possibilityhas been pointed out that the IMF for starburst clus-ters, with > M ⊙ , can change from the typical IMF(Bastian et al. 2010). Disk Fraction
The disk fraction is the fraction of cluster member starswith protoplanetary disks out of all cluster members. Itis often used for estimating disk lifetime, which is thoughtto be directly connected to the duration of planet forma-tion (Haisch et al. 2001; Lada & Lada 2003). Disk life-time is estimated using the age–disk fraction plot for var-ious star-forming clusters (Figure 14; Lada 1999, Hillen-brand 2005). In Figure 14, we plot data for clusters in thesolar neighborhood as black squares (Yasui et al. 2009,2010). The fit shown with a black curve is from Yasui etal. (2014). Disk fractions derived with NIR
JHK -bandobservations show very high values for very young clus-ters ( ∼
60 %) that decrease with increasing age. The disklifetime is often defined as the time when the disk frac-tion reaches ∼ L -band observa-tions and space MIR observations, the characteristics arequite similar (Lada 1999, Yasui et al. 2009; see the redline in the right panel of Figure 5 in Yasui et al. 2014).We estimated the disk fraction for the S127 clusters us-ing the NIR color–color diagram (Figure 8) and the samemethod as described in our previous papers. In Figure 8,we used the dotted–dashed line parallel to the reddeningvector that passes through the point at the end of thedwarf main-sequence star curve (blue line) at the pointwhere the H − K S value for the curve is maximum (theM6 point on the curve) as the border between stars withand without circumstellar disks (see details in Yasui etal. 2009). Assuming that disk emission is only evident in the K band, we classed stars on the lower right sideof the borderline as disk excess sources and calculatedthe ratio of the number of cluster members with disk ex-cesses to that of all cluster members. As a result, diskfractions for the S127A and S127B clusters are estimatedto be 28% ±
3% (108 / ±
4% (128 / f stars )in each intrinsic ( H − K ) color bin ( H − K ) for theS127A cluster in red and the S127B cluster in blue (alsoshown as black lines in Figure 10), as well as thosefor other young clusters in low-metallicity environments:S207 (thick solid line), S208 (thin solid line), Cloud 2-N(dashed line), and Cloud 2-S (dotted line). The verti-cal dashed line represents the borderline for estimatingthe disk fraction in the MKO system, i.e. the dotted–dashed line in Figure 8. The distribution becomes bluerand sharper with lower disk fractions for nearby youngclusters (see the bottom panel of Figure 7 in Yasui et al.2009), which is also the case for clusters in low-metallicityenvironments (Yasui et al. 2009). The peak ( H − K ) ofthe S127 cluster is relatively red, ( H − K ) ∼ . H − K ) of ∼ ∼
30% rather than those ofthe S207 and Cloud 2-N clusters with disk fractions of < ∼ ∼ t = 0 as for clusters in the solar neighborhoodof 64%. Yasui et al. (2010) pointed out the tendency forlow-metallicity clusters to have lower disk fractions thansolar metallicity clusters for a given age and suggestedthat the disk lifetime in low-metallicity environments isquite short, as discussed in Yasui et al. (2009). However,note that the estimated disk fraction for the S127B clus-ter is higher than those for other low-metallicity clusters.It may be possible that the disk fraction for the S127Bcluster is elevated due to factors other than metallicityand young age (e.g., position in the Galaxy, cluster scale,etc). However, the disk fraction for the S127A clusteris comparable to those for clusters with the same age,S208 and Cloud 2-S, despite the fact that S127A is inthe same location in the Galaxy as S127B and both haveidentical cluster scales. Therefore, the high value of thedisk fraction for the S127B cluster is probably due to thevery young age, between 0.1 and 0.5 Myr, which is theyoungest among the low-metallicity cluster sample. Theresults that quite young clusters in low-metallicity en-vironments have relatively high disk fractions, althoughdisk fractions of only .
30 % had been obtained for theclusters previously observed, suggest that a large fractionof stars have disks even in low-metallicity environmentsin the very early phases and that they lose their disks ona very short timescale.In the solar neighborhood, it is suggested that diskfraction, and thus disk lifetime, depends on cluster scales,such as cluster mass and stellar density (e.g., Fang et al.2013; Stolte et al. 2010). Fang et al. (2013) suggestedthat the disks in sparse stellar associations are dissipatedmore slowly than those in denser (cluster) environments,while Stolte et al. (2010) suggested that disk depletionis significantly more rapid in compact starburst clustersthan in moderate star-forming environments. In the pre-vious studies of young clusters in low-metallicity environ-ments, observed clusters are relatively small, with num-bers of identified cluster members of less than 100, whilethe S127 clusters have ∼ J − K colors of larger than 3 (equal to the sum of J − H (y-axis) and H − K (x-axis) colors on the NIR color–color diagram). This suggests that young stellar objectsin low-metallicity environments are initially surroundedby thick circumstellar materials, as is the case for the so-lar neighborhood, but they disperse very quickly, as alsodiscussed in Paper II. This also supports that the veryyoung age estimated for S127 in Section 5.2 (0.5 Myr) isreasonable, considering the estimated age of S208 is alsoestimated as ∼ REFERENCESAdams, F. C., Proszkow, E. M., Fatuzzo, M., et al. 2006, ApJ,641, 504Asplund, M., Grevesse, N., Sauval, A. J., & Scott, P. 2009,ARA&A, 47, 481Balser, D. S., Rood, R. T., Bania, T. M., & Anderson, L. D.2011, ApJ, 738, 27Bastian, N., Covey, K. R., & Meyer, M. R. 2010, ARA&A, 48, 339Beichman, C. A., Neugebauer, G., Habing, H. J., Clegg, P. E., &Chester, T. J. 1988, Infrared Astronomical Satellite (IRAS)Catalogs and Atlases, Vol. 1: Explanatory Supplement, 1Bessell, M. S., & Brett, J. M. 1988, PASP, 100, 1134Bica, E., Dutra, C. M., Soares, J., & Barbuy, B. 2003, A&A, 404,223Blitz, L., Fich, M., & Stark, A. A. 1982, ApJS, 49, 183Brand, J. & Wouterloot, J. G. A. 2007, A&A, 464, 909Brand, J., Wouterloot, J. G. A., Rudolph, A. L., et al. 2001,A&A, 377, 644Brewer, M.-M. & Carney, B. W. 2006, AJ, 131, 431Caplan, J., Deharveng, L., Pe˜na, M., Costero, R., & Blondel, C.2000, MNRAS, 311, 317Carpenter, J. M. 2000, AJ, 120, 3139Chini, R., & Wink, J. E. 1984, A&A, 139, L5Chru´sli´nska, M., Jeˇr´abkov´a, T., Nelemans, G., et al. 2020, A&A,636, A10Condon, J. J., Cotton, W. D., Greisen, E. W., et al. 1998, AJ,115, 1693D’Antona, F., & Mazzitelli, I. 1997, MmSAI, 68, 807D’Antona, F., & Mazzitelli, I. 1998, in ASP Conf. Ser. 134:Brown Dwarfs and Extrasolar Planets, 442Deharveng, L., Pe˜na, M., Caplan, J., & Costero, R. 2000,MNRAS, 311, 329Digel, S., de Geus, E., & Thaddeus, P. 1994, ApJ, 422, 92Drew, J. E., Greimel, R., Irwin, M. J., et al. 2005, MNRAS, 362,753Dyson, J. E., & Williams, D. A. 1980, (New York: Halsted Press),204Elmegreen, B. G., Klessen, R. S., & Wilson, C. D. 2008, ApJ,681, 365 Ercolano, B. & Clarke, C. J. 2010, MNRAS, 402, 2735Fang, M., van Boekel, R., Bouwman, J., et al. 2013, A&A, 549,A15Fich, M. 1993, ApJS, 86, 475Fich, M., Dahl, G. P., & Treffers, R. R. 1990, AJ, 99, 622Foster, T., & Brunt, C. M. 2015, AJ, 150, 147Habing, H. J. & Israel, F. P. 1979, ARA&A, 17, 345Haisch, K. E., Jr., Lada, E. A., & Lada, C. J. 2000, AJ, 120, 1396Haisch, K. E., Jr., Lada, E. A., & Lada, C. J. 2001, ApJL, 553,L153Hartmann, L., Ballesteros-Paredes, J., & Bergin, E. A. 2001, ApJ,562, 852Helou, G., & Walker, D. W. 1988, NASA RP-1190, Vol. 7, 0Hillenbrand, L. A. 2005, arXiv:astro-ph/0511083Hoare, M. G., Kurtz, S. E., Lizano, S., et al. 2007, Protostars andPlanets V, 181Ichikawa, T., Suzuki, R., Tokoku, C., et al. 2006, Proc. SPIE,6269, 626916Kobayashi, N. & Tokunaga, A. T. 2000, ApJ, 532, 423Kobayashi, N., Yasui, C., Tokunaga, A. T., & Saito, M. 2008,ApJ, 683, 178Koenig, X. P., Leisawitz, D. T., Benford, D. J., et al. 2012, ApJ,744, 130Lada, C. J., & Adams, F. C. 1992, ApJ, 393, 278Lada, C. J., & Lada, E. A. 2003, ARA&A, 41, 57Lada, E. A. 1999, in The Origin of Stars and Planetary Systems,ed. C. J. Lada & N. D. Kylafis (Dordrecht: Kluwer), 441Leggett, S. K., Currie, M. J., Varricatt, W. P., et al. 2006,MNRAS, 373, 781Lejeune, T., & Schaerer, D. 2001, A&A, 366, 538Leschinski, K. & Alves, J. 2020, A&A, 639, A120Lubowich, D. A., Brammer, G., Roberts, H., et al. 2004, Originand Evolution of the Elements, 37Mart´ın-Hern´andez, N. L., Peeters, E., Morisset, C., et al. 2002,A&A, 381, 606Meyer, M. R., Calvet, N., & Hillenbrand, L. A. 1997, AJ, 114, 288Minowa, Y., Kobayashi, N., Yoshii, Y., et al. 2005, ApJ, 629, 29
Mottram, J. C., Hoare, M. G., Davies, B., et al. 2011, ApJL, 730,L33Muench, A. A., Lada, E. A., & Lada, C. J. 2000, ApJ, 533, 358Muench, A. A., Lada, E. A., Lada, C. J., & Alves, J. 2002, ApJ,573, 366Peeters, E., Mart´ın-Hern´andez, N. L., Damour, F., et al. 2002,A&A, 381, 571Pettini, M., Smith, L. J., King, D. L., et al. 1997, ApJ, 486, 665Rolleston, W. R. J., Smartt, S. J., Dufton, P. L., et al. 2000,A&A, 363, 537Rudolph, A. L., Brand, J., de Geus, E. J., et al. 1996, ApJ, 458,653Rudolph, A. L., Fich, M., Bell, G. R., Norsen, T., Simpson, J. P.,Haas, M. R., & Erickson, E. F. 2006, ApJS, 162, 346Rudolph, A. L., Simpson, J. P., Haas, M. R., Erickson, E. F., &Fich, M. 1997, ApJ, 489, 94Santos, C. A., Yun, J. L., Clemens, D. P., et al. 2000, ApJL, 540,L87Santos, N. C., Melo, C., James, D. J., et al. 2008, A&A, 480, 889Scaife, A. M. M., Hurley-Walker, N., Davies, M. L., et al. 2008,MNRAS, 385, 809Sharpless, S. 1959, ApJS, 4, 257Shaver, P. A., McGee, R. X., Newton, L. M., et al. 1983,MNRAS, 204, 53Siess, L., Dufour, E., & Forestini, M. 2000, A&A, 358, 593Simons, D. A., & Tokunaga, A. 2002, PASP, 114, 169Skrutskie, M. F., Cutri, R. M., Stiening, R., et al. 2006, AJ, 131,1163Smartt, S. J., Dufton, P. L., & Rolleston, W. R. J. 1996, A&A,305, 164Snell, R. L., Carpenter, J. M., & Heyer, M. H. 2002, ApJ, 578, 229 Stolte, A., Morris, M. R., Ghez, A. M., et al. 2010, ApJ, 718, 810Sun, Y., Xu, Y., Yang, J., et al. 2015, ApJL, 798, L27Suzuki, R., Tokoku, C., Ichikawa, T., et al. 2008, PASJ, 60, 1347Tokunaga, A. T., Simons, D. A., & Vacca, W. D. 2002, PASP,114, 180Vall´ee, J. P. 2005, AJ, 130, 569Vall´ee, J. P. 2020, MNRAS, 494, 1134Vilchez, J. M., & Esteban, C. 1996, MNRAS, 280, 720Wenger, M., Ochsenbein, F., Egret, D., et al. 2000, A&AS, 143, 9Wouterloot, J. G. A., & Brand, J. 1989, A&AS, 80, 149Wright, E. L., Eisenhardt, P. R. M., Mainzer, A. K., et al. 2010,AJ, 140, 1868Yasui, C. 2020, arXiv:2010.00209Yasui, C., Izumi, N., Saito, M., et al. 2017, in IAU Symp. 321,Formation and Evolution of Galaxy Outskirts, 34Yasui, C., Kobayashi, N., Saito, M., et al. 2016a, AJ, 151, 115Yasui, C., Kobayashi, N., Tokunaga, A. T., & Saito, M. 2014,MNRAS, 442, 2543Yasui, C., Kobayashi, N., Tokunaga, A. T., Saito, M., & Izumi,N. 2016b, AJ, 151, 50Yasui, C., Kobayashi, N., Tokunaga, A. T., Saito, M., & Tokoku,C. 2008a, in ASP Conf. Ser. 396, Formation and Evolution ofGalaxy Disks, 225Yasui, C., Kobayashi, N., Tokunaga, A. T., Saito, M., & Tokoku,C. 2009, ApJ, 705, 54Yasui, C., Kobayashi, N., Tokunaga, A. T., Saito, M., & Tokoku,C. 2010, ApJL, 723, L113Yasui, C., Kobayashi, N., Tokunaga, A. T., Terada, H., & Saito,M. 2006, ApJ, 649, 753Yasui, C., Kobayashi, N., Tokunaga, A. T., Terada, H., & Saito,M. 2008b, ApJ, 674, 443
TABLE 1Properties of S127.
Name S127Galactic longitude (deg) 96.287 (1)Galactic latitude (deg) +2.594 (1)R.A. (J2000.0) 21 28 41.6 (1)Decl. (J2000.0) +54 37 00 (1)Photometric heliocentric distance (kpc) 9.7 (2)Kinematic heliocentric distance (kpc) 9.97 (3)Adopted distance (kpc) ≃ a (kpc) ≃ / H) 8 . +0 . − . (4, 5), 8 . +0 . − . (5, 6), 8 . +0 . − . (5, 7), 7 . +0 . − . (5, 8)Metallicity [O/H] (dex) b ≃− ±
820 (9), 11428 ±
305 (10)
Notes.
References are shown in parentheses. a Assuming a solar Galactocentric distance of R ⊙ = 8 . b Assuming a solar abundance of 12 + log (O / H) = 8 .
73 (Asplund et al. 2009).
References. (1) SIMBAD (Wenger et al. 2000), (2) Chini & Wink (1984), (3) Foster & Brunt (2015), (4) Vilchez & Esteban(1996), (5) Rudolph et al. (2006), (6) Rudolph et al. (1997), (7) Caplan et al. (2000), (8) Peeters et al. (2002), (9) Scaife et al.(2008), (10) Balser et al. (2011). TABLE 2Summary of MOIRCS Observations.
Modes Date Band t total t Coadds N total Seeing Sky Condition(1) (2) (3) (4) (5) (6) (7) (8) (9) J -long 2006 Sep 2 J . ′′ H -long 2006 Sep 2 H . ′′ K S -long 2006 Sep 2 K S . ′′ J -short 2007 Nov 22 J
52 13 1 4 (3) 0 . ′′ H -short 2007 Nov 22 H
52 13 1 4 (3) 0 . ′′ K S -short 2007 Nov 22 K S
52 13 1 4 (3) 0 . ′′ Notes.
Column (4): total exposure time (s). Column (5): single-exposure time (s). Column (6): number of coadds. Column(7): total number of frames. Column (9): P: photometric, and H: high humidity. The values for the sky frames are shown inparentheses.
TABLE 3Limiting magnitudes (10 σ ) of long-exposure images for MOIRCS observations. Frame J Band H Band K S BandCluster 22.0 21.2 21.3Sky 22.2 21.2 21.0
Fig. 1.—
Pseudocolor (top) and H α (bottom) images of S127 with a wide field of view of 5 ′ × ′ centered at ( α . , δ . ) =(21 h m s , +54 ◦ ′ ′′ ) in equatorial coordinates and ( l, b ) = (96 . ◦ , +2 . ◦ ) in Galactic coordinates, which is the coordinate of IRAS21270+5423. North is up, and east is to the left. The 1 ′ corresponds to 2.9 pc for the distances of S127. Top: the image is produced bycombining the 2MASS K S -band (2.16 µ m; blue), WISE band 1 (3.4 µ m; green), and WISE band 3 (12 µ m; red). The large red plus signshows the IRAS point source, while the small red plus sign shows the bright stars in the optical bands, ALS 18695. The white box showsthe location and size of the MOIRCS field of view. Bottom: IPHAS H α image of S127 shown in gray scale. The 1.4 GHz radio continuumemission by NVSS is also shown with blue contours. The contours are plotted at 1 mJy beam − × , − / , , ... . The blue diamondsshow the NVSS radio point sources, NVSS 212841+543634 and NVSS 212843+543728. The red plus symbols are the same as those in thetop panel, while the black box is the same as the white box in the top panel. Fig. 2.—
Top view of the Milky Way galaxy, showing S127 in relation to the spiral arms. The filled circle shows S127 at a distance of D = 9 . ± .
73 kpc (Foster & Brunt 2015) from the sun. The Sun is shown at a Galactocentric distance of 8 kpc by a circled dot. Spiralarms from Vall´ee (2005) are shown with different colors (red, yellow, green, and cyan for the Norma-Cygnus, Perseus, Sagittarius-Carina,and Scutum-Crux arms, respectively).
Right Ascension (2000) D ec li n a t i o n ( ) Fig. 3.—
Pseudocolor image of S127 produced by combining the J - (1.26 µ m), H - (1.64 µ m), and K S -band (2.15 µ m) MOIRCS imagesfrom 2006 September. The equatorial coordinates of the center of the image are α = 21 h m . s , δ = +54 ◦ ′ . ′′ . The field ofview of ∼ . ′ × ′ is shown with white and black boxes in Figure 1, and the symbols are the same as in Figure 1. Right Ascension (2000) D ec li n a t i o n ( ) S127B S127A Fig. 4.—
Stellar density of detected sources in the MOIRCS K S -band image is shown with yellow contours, superposed on the MOIRCS K S -band image, whose field of view is the same as Figure 3. The contour levels represent stellar densities of 3 σ , 4 σ , 5 σ , ..., and 20 σ higherthan the average stellar density in the field outside of the cyan ellipse. The green polygons show two identified star-forming clusters, S127Aand S127B clusters. −1 0 1 2 3 4 J − K S K S
60 M ⊙
40 M ⊙
20 M ⊙
10 M ⊙ ⊙ ⊙ ⊙ ⊙ A V = m a g S127A −1 0 1 2 3 4
J − K S K S
60 M ⊙
40 M ⊙
20 M ⊙
10 M ⊙ ⊙ ⊙ ⊙ ⊙ A V = m a g S127B
Fig. 5.— ( J − K S ) vs. K S color–magnitude diagram of the S127 clusters, the S127A (left) and S127B (right) cluster. Identified clustermembers in the cluster region ( A V ≥ A V = 5 mag. The dashed lines mark the limiting magnitudes (10 σ ). The black lines show the dwarf tracks byBessell & Brett (1988) in spectral types O9–M6 (corresponding mass of ∼ M ⊙ ). The blue lines denote the isochrone models forthe age of 1 Myr by D’Antona & Mazzitelli (1997, 1998; 0 . ≤ M/M ⊙ ≤ < M/M ⊙ ≤ M/M ⊙ ≥ A V = 0 and 3 mag, respectively. A distance of 10 kpcis assumed. The short horizontal lines are placed on the isochrone models and shown with the same colors as the isochrone tracks, whichshow the positions of 0.1, 1, 3, 5, 10, 20, 40, and 60 M ⊙ . −1 0 1 2 3 4 J − K S K S
60 M ⊙
40 M ⊙
20 M ⊙
10 M ⊙ ⊙ ⊙ ⊙ ⊙ A V = m a g Fig. 6.—
Same as Figure 5 but for the control field. A V N S127A A V N S127B
Fig. 7.— A V distributions for the sources in the S127 clusters (thick lines) and those in the control field (thin lines). Left and rightpanels show the S127A and S127B clusters, respectively. The distributions for the control field are normalized to match the area of eachcluster region. −0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 (H − K S ) MKO −0.50.00.51.01.52.02.53.03.5 ( J − H ) M K O A V = m a g S127A −0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 (H − K S ) MKO −0.50.00.51.01.52.02.53.03.5 ( J − H ) M K O A V = m a g S127B
Fig. 8.— ( H − K S ) vs. ( J − H ) color–color diagram of S127. Identified cluster members are shown in red, while sources in the clusterregion but not identified as cluster members are shown in black. The blue curve in the lower left portion is the locus of points correspondingto the unreddened main-sequence stars. The dotted–dashed line, which intersects the main-sequence curve at the maximum H − K S values(M6 point on the curve) and is parallel to the reddening vector, is the border between stars with and without circumstellar disks. TheCTTS locus is shown with the cyan line. −0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 (H − K S ) MKO −0.50.00.51.01.52.02.53.03.54.0 ( J − H ) M K O A V = m a g Fig. 9.—
Same as Figure 8 but for the control field. −0.5 0.0 0.5 1.0 1.5 2.0 (H − K) N S127A −0.5 0.0 0.5 1.0 1.5 2.0 (H − K) N S127B
Fig. 10.— ( H − K ) distributions for the S127 cluster members (thick line) and stars in the control field (thin line). Left and right panelsshow the S127A and S127B clusters, respectively. The distribution for the control field is normalized to match the total area of the clusterregion. The vertical black and gray dashed lines show average ( H − K ) values for the S127 cluster members and stars in the control field,respectively.
12 14 16 18 20 22 K l o g N S127A
12 14 16 18 20 22 K l o g N S127B
Fig. 11.—
Raw KLFs for the S127A cluster members (left) and S127B (right). The KLF for all cluster members with A V ≥ A V samples are shown by gray lines for clustermembers with A V of 4.2–12.4 and A V of 3.5–9.1 mag for S127A and S127B, respectively (see text for more detail). For clarity, the KLFsfor limited A V samples are vertically shifted by +0.1 for both clusters. The vertical dotted–dashed line shows the limiting magnitudes ofthe 10 σ detection (21.3 mag). −1 R (pc) N s t a r s AB Fig. 12.—
Correlation between the number of stars in a cluster ( N stars ) and the radius of the cluster ( R ). The red filled circles show S127clusters, while red open circles show other young low-metallicity cluster samples: S207, S208, and Cloud 2-N and -S from other papers inour series. The open squares show clusters in the solar neighborhood whose data are from Lada & Lada (2003) and Carpenter (2000). Thesolid line shows a rough fit to the data for clusters in the solar neighborhood; most points are scattered within a factor of √ R , shownwith dotted lines. The solid and dotted lines represent lines of constant cluster density. Fig. 13.—
Comparison of the S127 KLFs (black lines) with model KLFs of various ages (colored lines). Error bars are the uncertaintiesfrom Poisson statistics. The cyan, red, blue, magenta, and green lines represent model KLFs of 0.1, 0.5, 1, 2, and 3 Myr, respectively. Thevertical dotted–dashed lines show the limiting magnitudes of the 10 σ detection (18.5 mag). Fig. 14.—
Disk fraction as a function of cluster age.
JHK disk fractions of the young clusters in low-metallicity environments are shownwith red circles. The S127 clusters are shown with red filled circles, while other clusters are shown with red open circles.
JHK disk fractionsof young clusters with solar metallicity are shown by black filled squares. The black and red lines show the disk fraction evolution undersolar metallicity and in low-metallicity environments, respectively. Fig. 15.—
Comparison of intrinsic H − K color distributions. The fractions of stars ( f stars ) per each intrinsic color bin ( H − K )0