MMon. Not. R. Astron. Soc. , 1– ?? (2007) Printed 18 November 2018 (MN L A TEX style file v2.2)
About the nature of Mercer 14
D. Froebrich (cid:63) , G. Ioannidis † Centre for Astrophysics and Planetary Science, University of Kent, Canterbury, CT2 7NH, UK
Received sooner; accepted later
ABSTRACT
We used UKIRT near infrared (NIR) broad band
JHK photometry, narrow band imag-ing of the 1-0 S(1) molecular hydrogen emission line and mid infrared
Spitzer
IRAC data toinvestigate the nature of the young cluster Mercer 14. Foreground star counts in decontami-nated NIR photometry and a comparison with the Besancon Galaxy Model are performed toestimate the cluster distance. This method yields a distance of 2.5 kpc with an uncertainty ofabout 10 % and can be applied to other young and embedded clusters. Mercer 14 shows clearsigns of ongoing star formation with several detected molecular hydrogen outflows, a highfraction of infrared excess sources and an association to a small gas and dust cloud. Hence,the cluster is less than 4 Myrs old and has a line of sight extinction of A K = 0.8 mag. Basedon the most massive cluster members we find that Mercer 14 is an intermediate mass clusterwith about 500 M (cid:12) . Key words:
Galaxy: open clusters, individual; ISM: jets and outflows; stars: formation
Understanding the formation of stars is one of the key topics of cur-rent astrophysical research. It occurs in embedded clusters withingiant molecular clouds. A large fraction of all young stars areformed in such clusters (e.g. Lada & Lada (2003)). Low star for-mation efficiency, as well as feedback (radiation, winds, outflows)from young and massive stars leads to the disruption of many ofthose clusters in the first 10 Myrs of their life and generates the fieldstar population. Thus, in order to understand the formation of stellarclusters one has to investigate the youngest (up to a few Myrs old)objects. Preferably a large number of objects should be investigatedto account for environmental influences, but detailed investigationsof individual objects are vital to accurately determine their physicalparameters such as age, distance and line of sight reddening.Investigations of these young embedded clusters are hamperedby the high column density material in their vicinity. Recent yearshave seen advances of large scale near and mid-infrared surveys (2Micron All Sky Survey (2MASS, Skrutskie et al. (2006)); UK In-frared Deep Sky Survey (UKIDSS, Lawrence et al. (2007)) Galac-tic Plane Survey (GPS, Lucas et al. (2008)); GLIMPSE, Benjaminet al. (2003)) which allow us to discover and characterise the prop-erties of a variety of these young embedded clusters at wavelengthsless influenced by interstellar extinction. A large number of so farunknown stellar clusters in the Galaxy have been found in thosesurveys (e.g. Bica et al. (2003). Mercer et al. (2005), Froebrich etal. (2007), to name just a few).In particular the list by Mercer et al. (2005), based on mid- (cid:63)
E-mail: [email protected] † E-mail: [email protected] infrared
Spitzer data, contains a number of embedded young clus-ters hidden from view in shorter wavelengths surveys. One of these,Mercer 14, has so far not attracted any attention besides the discov-ery paper. We aim to determined the nature of this cluster mainly bymeans of near infrared
JHK photometry obtained from the UKIDSSGPS and the UK Infrared Telescope (UKIRT) Widefield InfraredSurvey for H (UWISH2, Froebrich et al. (2011)) which is using anarrow band filter centred on the 1-0 S(1) emission line of molecu-lar hydrogen at 2.122 µ m.We briefly discuss the data used in our paper in Sect. 2. Ouranalysis, in particular the results of the decontamination of the pho-tometry, the cluster distance determination and the identification ofmolecular outflows from young stars in the cluster are discussed inSect. 3. We then put forward our conclusions in Sect. 4. We obtained near infrared imaging data in the 1-0 S(1) line ofmolecular hydrogen using the Wide Field Camera (WFCAM,Casali et al. (2007)) at UKIRT. The data are part of the UWISH2survey (Froebrich et al. (2011)) and the H images are taken witha per pixel integration time of 720 s and under very good seeingconditions. The full width half maximum of the stellar point spreadfunction is 0.7 (cid:48)(cid:48) and the 5 σ point source detection limit is about18 mag. Our narrow band data were taken on the 30th of July in2009.In order to perform the continuum subtraction of our narrowband images we further utilised the UKIDSS data in the near in- c (cid:13) a r X i v : . [ a s t r o - ph . GA ] A ug Froebrich & Ioannidis frared
JHK bands taken with the same telescope and instrumentalsetup as part of the GPS. This broad band data was taken on the1st of June in 2006 in all three filters. All GPS data used was takenfrom data release 7.All NIR (UWISH2 and GPS) data reduction and photometryare performed by the Cambridge Astronomical Survey Unit. Re-duced images and photometry tables are available via the WideField Astronomy Unit and are downloaded from the WFCAM Sci-ence Archive (Hambly et al. (2008)). The basic data reductionprocedures applied are described in Dye et al. (2006). Calibra-tion (photometric as well astrometric) is performed using 2MASS(Skrutskie et al. (2006)) and the details are described in Hodgkinet al. (2009). All NIR magnitudes used in this paper are quoted for
AperMag3 , i.e. a 2 (cid:48)(cid:48) aperture, the standard for GPS point sources(see Lucas et al. (2008) for details).
Spitzer
IRAC data
We also downloaded IRAC images for all filters and point sourcephotometric catalogues of the region that have been obtained by
Spitzer as part of the GLIMPSE Survey (Benjamin et al. (2003)).All data is part of the final release of enhanced data products forGLIMPSE. Except two, all of the potential outflow driving sourcesfor which we require photometry are extended. Hence, they are notpart of the GLIMPSE point source catalogue. We thus performedaperture photometry on the IRAC images.We defined apertures encompassing all the flux from the ex-tended sources, as well as apertures around some bright pointsources (e.g. obj. C, see Table 1). Since the cluster is embeddedin partly dense material, the background flux for each source hasbeen determined in an aperture as close to the source as possible.Fluxes above the background were then converted into magnitudesand calibrated using the bright source obj. C, whose magnitudesare taken from the GLIMPSE point source catalogue. Uncertaintiesintroduced due to background estimation and the choise of aper-ture are about 0.1 mag for the two short wavelengths filter. For thelonger wavelengths we estimate a larger error of about 0.2 mag,since the background emission is stronger (in particular in the 8 µ mfilter) and spatially variable on small scales. In conjunction with theGLIMPSE calibration uncertainty of 0.1 mag to 0.2 mag our pho-tometric uncertainties are 0.2 – 0.3 mag with the larger value validfor the longer wavelengths filters. The cluster Mercer 14 has been discovered by Mercer et al. (2005)during their search for embedded star clusters utilising GLIMPSEdata. The cluster is embedded in a small, mid-infrared brightmolecular cloud. It has, except in the discovery paper, not at-tracted any attention nor have its parameters be determined. Thenear infrared
JHK images give the impression of a sphericallysymmetric accumulation of stars, clearly visible at near infraredwavelengths. The central coordinates, determined visually are at18:58:05.8 + flows and fea-tures, mostly just to the North-North-East of the cluster (see Fig. 1). If they are associated with the cluster itself, then this would indicatethat Mercer 14 is a young stellar cluster, with partly still ongoingstar formation. At the northern edge of the cluster there is a veryred, fuzzy nebulosity visible in the JHK colour images. It coincideswith one of the IRAC detected Extended Green Objects (EGOs;EGO G035.13-0.74; Cyganowski et al. (2008); Obj. A in Table 1).Only 4 (cid:48)(cid:48) to the north-east of this, we find a smaller red reflectionnebula, a fainer EGO not included in the list by Cyganowski et al.(2008) (Obj. B). About 75” to the North-East-East of the clustercentre we detect a bright star (K = 9.5 mag in 2MASS) in the centreof a small HII region. This object is actually listed as a PlanetaryNebula (PN G035.1-00.7) in the SIMBAD database. The NIR im-ages, however, clearly suggest that this object is an HII region (seea more detailed discussion in Sect. 3.6).
In order to analyse the cluster properly, we need to decontaminatethe NIR photometry from fore and background objects. The de-contamination is based on the method established by Bonatto &Bica (2007). But we use the improvements introduced by Froebrichet al. (2010). We download the near infrared photometry from theGPS within 3 (cid:48) around the cluster centre. Only stars with a detectionabove the local completeness limit (J = 18.5 mag, H = 17.5 mag,K = 17.0 mag) in each filter, photometric uncertainties of less than0.1 mag, and with a pstar value of more than 0.9 are considered(i.e. only star like objects, see Lucas et al. (2008)).We define as ’cluster area’ everything within 0.9 (cid:48) from thecluster centre. As ’control region’ we use all stars between 2 (cid:48) and 3 (cid:48) from the central coordinates. The calculation of the P ccm member-ship probabilities for the individual stars in the cluster area is basedon the 15 th nearest neighbour in the colour-colour-magnitude space(see Froebrich et al. (2010) for details).The result of the procedure is shown in the decontaminatedcolour-magnitude (CMD) and colour-colour (CCD) diagrams inFig. 2 together with a cluster isochrone as a guide to the cluster se-quence (see Sect. 3.5). There is a large fraction of stars with P ccm membership probabilities of more than 60 %, as well as possiblefore or background objects not related to the cluster. There are a variety of ways to determine the distance to Mercer 14using the available data set. These are: i) Using an object associatedto the cluster or its parental molecular cloud which has a known dis-tance; ii) Perform isochrone fitting to the cluster’s decontaminatedCMD and CCD; iii) Estimate the number of foreground stars tothe cluster or its parental molecular cloud and compare this to theBesancon Galaxy model by Robin et al. (2003).i) A search in the vicinity of Mercer 14 for objects of knowndistance did not reveal any usable object. The object nearest to thecluster with a reliable distance estimate is the massive star form-ing object G35.2N. Uruquart et al. (2011) place it at a distanceof 2.3 kpc. Considering the projected separation of about 4.3 (cid:48) (orabout 3 pc at this distance), it is not entirely clear if the cluster andG35.2N are physically related. The 8 µ m GLIMPSE image of theregion also reveals that Mercer 14 is just surrounded by a smallcloud with no obvious connection to G35.2N.ii) Given the low age and large number of pre-main sequencestars in the cluster (indicated by the presence of molecular outflows,as well as the isochrone fit, see below) it is extremely difficult to si-multaneously fit the distance and age of the cluster to satisfactory c (cid:13) , 1– ?? bout the nature of Mercer 14 Figure 1.
Gray scale representation of the H -K difference image of the region containing outflows near Mercer 14. The image is centred at RA = 18:58:06.6and DEC = +01:37:03 (J2000). North is up, East to the left. Circles and ellipses mark the MHOs discussed in the text. Their numbers are added as labels.Squares indicate the five high probability driving sources and the solid lines mark some of the discussed flows. North is to the top and East to the left. accuracy. Rather the distance needs to be known in order to deter-mine the approximate age and reddening to the cluster from pho-tometry alone. Thus, an estimate of the number of stars foregroundto the cluster or the parental molecular cloud can be determinedand compared to the Besancon Galaxy model (Robin et al. (2003))in order to estimate the distance. It has been shown e.g. by Scholzet al. (2010) that such an approach can give a reasonably accuratedistance estimate.iii) As a first possibility we count the number of blue (J -K < (cid:48) – 4 (cid:48) ). Fields too close to the clustermight contain cluster members, contaminating the analysis. On theother hand, fields too far away might show a too low extinctionto successfully distinguish stars foreground or background to thecloud. One might even measure the distance to a cloud not relatedto the cluster at all, given its position in the Galactic Plane and thepossibility of overlapping clouds along the line of sight. The areawhere foreground stars are counted should be as large as possible tominimise statistical errors. We choose one square arcminute sizedfields. The number of foreground stars in them varies between 13and 22, with an average of 18 stars per square arcminute. We then estimated the photometric properties of the near in-frared data set obtained from the GPS (completeness limit and pho-tometric errors as a function of magnitude). A model photometriccatalogue for 10 square arcminutes (we chose a larger area to im-prove the accuracy of the model) towards the cluster position andwith the above determined restrictions for the JHK photometry wasdownloaded from the Besancon galaxy model webpage . A com-parison of the measured number of foreground stars with the modelpredictions leads to a distance estimate. The model stars are sortedby distance. We find the star in the list for which all stars closerthan it would give a model star density equal to our measured fore-ground star density. The distance to this model star is then used asthe distance to the cloud. By considering the variations of the for-ground star density we obtain a range of distances for the cloud of2.1 kpc to 2.8 kpc with an average of 2.5 kpc.The second possibility is to determine the number of stars perunit area that are foreground to the cluster itself. This is based onthe decontaminated CMD of the cluster (see Sect. 3.2). The colourof most of the high probability cluster members is J − K = http://model.obs-besancon.fr/c (cid:13) , 1– ?? Froebrich & Ioannidis
Figure 2.
Decontaminated colour-magnitude (left panel) and colour-colour (right panel) diagram of the cluster Mercer 14. Only objects in the GPS with pstar greater than 0.9 are included. Cluster membership probabilities P ccm are indicated by different symbols: above 80 % red square; 60–80 % green triangle;40–60 % blue plus sign; 20–40% magenta cross; below 20 % black dot. The P ccm probabilities are based on the 15th nearest neighbour in the colour-colour-magnitude space. The over plotted isochrone (solid line) has an age of 4 Myrs, a distance of 2.5 kpc and a K-band extinction of 0.8 mag. The dashed isochronein the right panel has the same properties but is un-reddened. The dotted line in the left panel indicates the detection limits in the cluster area. The soliddiagonal lines show the reddening band for photospheres based on our adopted extinction law. stars. The total number of stars bluer than J − K = (cid:48) around the cluster centre. If we weight thecontribution of each star by the probability that it is not a clustermember ( − P ccm ) then there are 17 stars per square arcminuteforeground to the cluster. This corresponds to a distance of 2.5 kpc,in perfect agreement to the star counting estimates using the fieldsin the parental molecular cloud. This distance has been determinedusing the nearest 15 neighbours in colour-colour-magnitude spacefor the decontamination. When using the nearest 10 neighbours thedistance estimate is about 2.0 kpc, for the 20th nearest neighbourswe obtain 2.8 kpc. Thus, including the uncertainties, the value alsoagrees with the distance of G35.2N. Hence, Mercer 14 could indeedbe part of the same star forming complex as the massive YSOs inG35.2N.We use 2.5 kpc as distance to Mercer 14 throughout the paper. As can be seen in Fig. 1, there are a number of molecular hy-drogen outflow features just to the North-East of the cluster (la-beled MHO 2423 – MHO 2428). In the following we will describetheir properties and discuss possible driving sources. This discus-sion will be focused on the most obvious driving source candidateslisted in Table 1 and labeled by letters A – E in Fig. 1. There are alarger number of other objects in the cluster region that show near infrared excess. About one in three of the high probability clustermembers show K-band excess emission (objects to the right of thereddening band, see right panel of Fig. 2). Each of these could inprinciple drive an outflow and some of them might be the sourcefor H emission features. Due to their large numbers and withoutfurther information such as proper motions of the emission knotswe cannot include them into our considerations.The numbering and naming of the molecular hydrogen emis-sion line objects (MHOs) has been done in accordance with theconvention set out by Davis et al. (2010). In the Appendix (on-line version only) we present two higher resolution continuum sub-tracted H -K images (Figs. A1 and A2) with detailed labels for eachidentified MHO. None of the outflow features is visible in the H α images taken by the IPHAS survey (Drew et al. (2005)). This ismost likely caused by the high optical extinction towards the clus-ter (see below).A summary of the MHO positions, fluxes and possible drivingsources is given in Table 2. MHO 2423
Objects MHO 2423 A and MHO 2423 B seem to form two sidesof a symmetric bipolar outflow with a position angle of about150 ◦ . It has a length of 1 (cid:48) , which corresponds to about 0.75 pc atthe adopted distance. Given the symmetry of the flow, the driving c (cid:13) , 1– ?? bout the nature of Mercer 14 source should be situated on the axis defined by the two featuresand be located roughly half way between the two ends. There is noobvious bright source detectable in the NIR data. One of the twored and extended sources (obj. B in Table 1 – the faint EGO) is sit-uated just off-axis. There is, however, a relatively bright and red Spitzer source (obj. D, undetected in the NIR images) closer to theaxis. Based on its position and red ([3.6-4.5] = 3.1 mag) colour it isthe most likely source for the flow.
MHO 2424
The identification of the source(s) of MHO 2424 A andMHO 2424 B is not so straight forward. Both could be drivenby the bright EGO (obj. A). However, the small chain of threeknots seems not to be aligned with this extended source. It alsocould be driven by the bright IRAC source (obj. C) which iscompletely invisible in the NIR images. In this case the flow wouldbe one-sided and 23 (cid:48)(cid:48) (0.3 pc) long. Given that we do not detectany counter flow and that there is plenty of material in the vicinityof the cluster, this does not seem a likely explanation.Another possibility is that MHO 2424 A and MHO 2424 Bform one flow which is driven by the red source obj. E. This objectis faint in
JHK but clearly detected in IRAC. If this is the drivingsource then the flow would have an s-shape, indicating the sourceis a binary. In particular the two knots which are symmetric andvery close (4 (cid:48)(cid:48) ) to source, seem to favour this explanation. The totallength of the flow would then be about 0.8 (cid:48) (0.6 pc). However, wecannot rule out that some of the H emission, in particular fromMHO 2424 A, is driven by the bright EGO (obj. A). MHO 2425
MHO 2425 is the a bright bow-shaped emission line object. It ap-pears to form an open bow pointing towards the NE. As there isno chain of H emission (or other indications such as a counterflow), it is not possible to trace its source. It could be any of theyoung sources (obj. A – D). In each case the flow length betweenthe shock and the source is about 0.5 pc. MHO 2426
The object MHO 2426 A consists of two faint knots pointing awayfrom the main region of outflow driving candidates. In particularthe faint EGO (obj. B) seems to be aligned with the flow axis. Fur-thermore, MHO 2426 B could be part of this flow, as it is alignedwith the axis (but see below in the next subsection). In any case, wewould just detect one side of the flow, which would have a lengthof 0.5 pc (0.25 pc without MHO 2426 B).
MHOs 2426 B and MHO 2427
These three H knots are all situated about half a parsec to thenorth-north-east of the cluster. If MHO 2426 B is not part of a flowwith MHO 2426 A, then these three objects could form an indepen-dent outflow, as they are aligned on a common axis with a lengthof 0.6 pc. However, there is no obvious source candidate (red GPSor Spitzer source) along this axis. Furthermore, for each H objectindividually, there is no source candidate. There are other H emis-sion features to the West of the cluster. These are all associated withthe HII region and apparent cloud edges in the 8 µ m Spitzer image.Hence all these features are very likely to be fluorescently excited by either the more massive young stars in the cluster, the centralstar of the HII region, or the general UV radiation field.
MHO 2428
MHO 2428 is a single H emission knot near MHO 2423 B. Basedon the position and the clear flow structure of MHO 2423, this ob-ject seems unrelated. The most likely source might be the faintEGO (obj. B), which is 0.15 pc away. However, obj. B is also thebest candidate source for MHO 2426 A (and MHO 2426 B). Using the decontaminated photometry of the cluster region we plotcolour-colour and colour-magnitude diagrams for the high proba-bility cluster members (see Fig. 2). We over plot isochrones to de-termine the cluster parameters. Given the large distance of 2.5 kpc,we expect to detect relatively massive young stars (above about1 M (cid:12) ) and no lower mass stars in our NIR data. We thus useisochrones from Girardi et al. (2002), which provide data for moremassive stars.The distribution of high probability members in the CMD andin particular the CCD shows that the cluster suffers from a highamount of extinction, which to a large extent is intrinsic to thecluster. Hence, the differential reddening is very high, caused bythe youth of Mercer 14. This is also indicated by the distribution ofthe 8 µ m emission as seen by Spitzer (right panel of Fig. 3). Thereone can clearly see that the cluster coincides with a 8 µ m emissionpeak. There is no correlation of a stars colour with distance to theemission peak. This indicates that the feature is not foreground tothe cluster.We use the CCD and an isochrone for 4 Myrs (the lowest ageavailable from Girardi et al. (2002)) to estimate the minimum ex-tinction shown by a high probability cluster member. This mini-mum is likely to be the foreground extinction. Using an extinc-tion law of A K /A J = 0.382 and A K /A H = 0.612 (based on Mathis(1990) and Froebrich et al. (2010)) we find a minimum foregroundextinction of A K = 0.8 mag, or A V = 7.4 mag. This is in good agree-ment to the 2MASS based extinction maps by Rowles & Froebrich(2009). Their highest resolution map has 0.9 (cid:48) resolution at the clus-ter position and determined an optical extinction of 5.9 mag. Thedifferential reddening, indicated by the scatter of sources alongthe reddening band is at least of the same order of magnitude(A K = 0.8 mag) as the foreground extinction.All high probability members appear in the bottom half of thereddening band in the CCD, in agreement with the fact that weonly detect early spectral type sources and no low-mass stars in thecluster. A fraction of about one third of cluster members is foundbelow the reddening band, indicating K-band excess emission froma disk. This fraction of objects with disk is likely to be a lower limitand indicates an age of less than 4 Myrs (Lada & Lada (2003)). Thisis also in agreement with the detected outflow activity and indicatesa very low age for the cluster, potentially much less than the 4 Myrsused for the isochrone in Fig. 2.The three brightest high probability cluster members ( P ccm > %) have an apparent K-band brightness of about 9.5 mag. Thiscorresponds to roughly a 20 M (cid:12) mass star. According to Weidner& Kroupa (2006) this maximum stellar mass indicates a total massof the cluster of about 500 M (cid:12) . c (cid:13) , 1– ?? Froebrich & Ioannidis
Table 1.
IRAC photometry of the possible outflow driving sources in the Mercer 14 region. Their coordinates as well as IRAC magnitudes are listed.Magnitudes for obj. C and obj. E are taken directly from the GLIMPSE point source catalogue. Values for the other objects are determined by us (see Sect. 2.2).Note that typical uncertainties are about 0.2 mag. We do not list the K-band excess sources identified in the NIR data which could in principle drive outflowsas well. obj. RA DEC I1 I2 I3 I4 Notes(J2000) (J2000) [mag] [mag] [mag] [mag]A 18:58:06.5 +01:36:52 9.5 8.4 8.0 10.3 bright EGO G035.13-0.74B 18:58:06.4 +01:37:02 10.6 9.7 9.0 11.3 faint EGOC 18:58:05.6 +01:37:01 12.1 9.0 7.4 6.8 bright IRAC source, no GPSD 18:58:06.4 +01:37:07 13.2 10.1 8.7 9.2 faint IRAC source, no GPSE 18:58:07.5 +01:36:41 11.2 9.8 8.9 8.4 IRAC and GPS detection
Table 2.
Molecular hydrogen photometry of the MHOs in the Mercer 14 region. We list the MHO number, the coordinates in (J2000), the apparent fluxesin the 1-0 S(1) line, the flow H luminosity (dereddened with A K = 0.8 mag and assuming 10 % of the H flux is in the 1-0 S(1) line), possible flow drivingsources (from Table 1) and length in pc. A ’?’ in the last column indicates that the source identification is unclear.MHO RA DEC F[1-0 S(1)] L[H ] Possible sources(J2000) (J2000) [W/m ] [L (cid:12) ] and length2423 A 18:58:05.9 +01:37:20 35 × − × − × − × − × − × − × − × − × − × − Figure 3.
Gray scale representation of the GPS K-band continuum (left) and the
Spitzer
IRAC 8 µ m image (right) of the area around Mercer 14. The largecircle indicates the ’cluster area’ centred on RA = 18:58:06.6 and DEC = +01:37:03 (J2000) with a radius of 0.9 (cid:48) . North is to the top and East to the left in theimage. c (cid:13) , 1– ?? bout the nature of Mercer 14 About 1.2 (cid:48) (0.9 pc) East of the cluster one can identify a bright star(18:58:10.5, +01:36:57 (J2000)) within an almost semi-circularlyshaped emission region (see Fig. 1 and left panel of Fig. 3). Theobject is listed as planetary nebula PN G035.1-00.7 in the SIM-BAD database. Judging by the appearance this seems to be a mis-classification and the object appears to be an HII region. With anapparent radius of 14 (cid:48)(cid:48) (0.17 pc) this corresponds to a compact HIIregion. The western edge of the region does not show any H emis-sion, possibly because there is no material present to be excited bythe central star, as it has been cleared to low density by the brightcluster stars in the vicinity.The central star of the HII region is partly saturatedin the UKIDSS GPS data, we hence use the 2MASS fluxes(J = 10.51 mag, H = 9.88 mag, K = 9.50 mag) to investigate its na-ture. We de-redden the star to place it onto the main sequencein the model isochrones by Girardi et al. (2002). With our ap-plied extinction law we require 0.78 mag of K-band extinction. Thisagrees well with the minimum reddening found for the cluster stars(A K = 0.8 mag) and thus places the HII region at the same distanceas the cluster. Using the cluster distance, reddening and apparentmagnitudes, we thus find that the central star of the HII region hasabout 20 solar masses, similar to the brightest cluster members.Hence its spectral type is about O9 or B0.Given the properties of the central star, we can clearly inferthat the object is miss-classified as planetary nebula and indeed anHII region. We investigate the nature of the young embedded cluster Mercer 14utilising near infrared
JHK photometry from the UKIDSS GPS andnarrow-band imaging data obtained as part of the UWISH2 survey.The cluster shows clear signs of currently ongoing star formation,indicated by several detected molecular hydrogen outflows. Mid-infrared
Spitzer data reveals that there are still substantial amountsof gas and dust within, or in very close proximity to the cluster.We identify at least five molecular outflows driven by young, partlydeeply embedded sources in the northern half of the cluster.We estimate the distance of Mercer 14 by star counts of blueforeground objects to its parental molecular cloud and a compari-son to the Besancon Galaxy model. We furthermore use the clusterphotometric decontamination algorithm to determine the numberof possible foreground stars per unit area towards the cluster. Bothmethods result in a distance estimate of 2.5 ± K = 0.8 mag, with a differential extinction of the same order ofmagnitude. The colours and magnitudes of the brightest clustermembers correspond to 20 solar mass stars. The nearby compactHII region also contains a star of this mass. This suggests a to-tal cluster mass of around 500 M (cid:12) . The content of young mas-sive stars, the detection of currently ongoing star formation indi-cated by jets and outflows, a large fraction of sources with disks, as well as the high differential reddening caused by remaining gasand dust within the cluster suggests that Mercer 14 is a young (lessthat 4 Myrs) intermediate mass cluster. ACKNOWLEDGMENTS
The United Kingdom Infrared Telescope is operated by the JointAstronomy Centre on behalf of the Science and Technology Facil-ities Council of the U.K. The data reported here were obtained aspart of the UKIRT Service Program. This research has made use ofthe WEBDA database, operated at the Institute for Astronomy ofthe University of Vienna.
REFERENCES
Benjamin, R.A., Churchwell, E., Babler, B.L., Bania, T.M.,Clemens, D.P., Cohen, M., Dickey, J.M., et al., 2003, PASP, 115,953Bica, E., Dutra, C.M., Soares, J., Barbuy, B., 2003, A&A, 404,223Bonatto, C.; Bica, E.; 2007, MNRAS, 377, 1301Casali, M., Adamson, A., Alves de Oliveira, C., Almaini, O.,Burch, K., Chuter, T., Elliot, J., et al., 2007, A&A, 467, 777Cyganowski, C.J.; Whitney, B.A.; Holden, E.; Braden, E.; Bro-gan, C.L.; Churchwell, E.; Indebetouw, R.; Watson, D.F.; et al.;2008, AJ, 136, 2391Davis, C. J.; Gell, R.; Khanzadyan, T.; Smith, M. D.; Jenness, T.,2010, A&A, 511, 24Drew, J.E., Greimel, R., Irwin, M.J., Aungwerojwit, A., Barlow,M.J., Corradi, R.L.M., Drake, J.J., et al., 2005, MNRAS, 362,752Dye, S., Warren, S.J., Hambly, N.C., Cross, N.J.G., Hodgkin, S.T.,Irwin, M.J., Lawrence, A., et al., 2006, MNRAS, 372, 1227Froebrich, D.; Davis, C. J.; Ioannidis, G.; Gledhill, T. M.; Takami,M.; Chrysostomou, A.; Drew, J.; Eisl¨offel, J.; et al.; 2011, MN-RAS, 413, 480Froebrich, D.; Schmeja, S.; Samuel, D.; Lucas, P. W.; 2010, MN-RAS, 409, 1281Froebrich, D., Scholz, A., Raftery, C.L. 2007, MNRAS, 374, 399Girardi, L., Bertelli, G., Bressan, A., et al. 2002, A&A, 391, 195Hambly, N.C., Collins, R.S., Cross, N.J.G., Mann, R.G., Read,M.A., Sutorius, E.T.W., Bond, I., et al., 2008, MNRAS, 384, 637Hodgkin, S.T., Irwin, M.J., Hewett, P.C., Warren, S.J., 2009, MN-RAS, 394, 675Lada, C.J., Lada, E.A., 2003, ARA&A, 41, 57Lawrence A., Warren, S.J., Almaini, O., Edge, A.C., Hambly,N.C., Jameson, R.F., Lucas, P.W., et al., 2007, MNRAS, 379,1599Lucas, P. W.; Hoare, M. G.; Longmore, A.; Schr¨oder, A. C.; Davis,C. J.; Adamson, A.; Bandyopadhyay, R. M.; de Grijs, R.; et al.;2008, MNRAS, 391, 1281Mathis, J.S. 1990, ARA&A, 28, 37Mercer, E.P.; Clemens, D.P.; Meade, M.R.; Babler, B.L.; Indebe-touw, R.; Whitney, B.A.; Watson, C.; Wolfire, M.G.; et al.; 2005,ApJ, 635, 560Robin, A.C.; Reyl´e, C.; Derri`ere, S.; Picaud, S.; 2003, A&A, 409,523Rowles, J. & Froebrich, D. 2009, MNRAS, 395, 1640Scholz, A.; Froebrich, D.; Davis, C.J.; Meusinger, H.; 2010, MN-RAS, 406, 505 c (cid:13) , 1–, 1–
Benjamin, R.A., Churchwell, E., Babler, B.L., Bania, T.M.,Clemens, D.P., Cohen, M., Dickey, J.M., et al., 2003, PASP, 115,953Bica, E., Dutra, C.M., Soares, J., Barbuy, B., 2003, A&A, 404,223Bonatto, C.; Bica, E.; 2007, MNRAS, 377, 1301Casali, M., Adamson, A., Alves de Oliveira, C., Almaini, O.,Burch, K., Chuter, T., Elliot, J., et al., 2007, A&A, 467, 777Cyganowski, C.J.; Whitney, B.A.; Holden, E.; Braden, E.; Bro-gan, C.L.; Churchwell, E.; Indebetouw, R.; Watson, D.F.; et al.;2008, AJ, 136, 2391Davis, C. J.; Gell, R.; Khanzadyan, T.; Smith, M. D.; Jenness, T.,2010, A&A, 511, 24Drew, J.E., Greimel, R., Irwin, M.J., Aungwerojwit, A., Barlow,M.J., Corradi, R.L.M., Drake, J.J., et al., 2005, MNRAS, 362,752Dye, S., Warren, S.J., Hambly, N.C., Cross, N.J.G., Hodgkin, S.T.,Irwin, M.J., Lawrence, A., et al., 2006, MNRAS, 372, 1227Froebrich, D.; Davis, C. J.; Ioannidis, G.; Gledhill, T. M.; Takami,M.; Chrysostomou, A.; Drew, J.; Eisl¨offel, J.; et al.; 2011, MN-RAS, 413, 480Froebrich, D.; Schmeja, S.; Samuel, D.; Lucas, P. W.; 2010, MN-RAS, 409, 1281Froebrich, D., Scholz, A., Raftery, C.L. 2007, MNRAS, 374, 399Girardi, L., Bertelli, G., Bressan, A., et al. 2002, A&A, 391, 195Hambly, N.C., Collins, R.S., Cross, N.J.G., Mann, R.G., Read,M.A., Sutorius, E.T.W., Bond, I., et al., 2008, MNRAS, 384, 637Hodgkin, S.T., Irwin, M.J., Hewett, P.C., Warren, S.J., 2009, MN-RAS, 394, 675Lada, C.J., Lada, E.A., 2003, ARA&A, 41, 57Lawrence A., Warren, S.J., Almaini, O., Edge, A.C., Hambly,N.C., Jameson, R.F., Lucas, P.W., et al., 2007, MNRAS, 379,1599Lucas, P. W.; Hoare, M. G.; Longmore, A.; Schr¨oder, A. C.; Davis,C. J.; Adamson, A.; Bandyopadhyay, R. M.; de Grijs, R.; et al.;2008, MNRAS, 391, 1281Mathis, J.S. 1990, ARA&A, 28, 37Mercer, E.P.; Clemens, D.P.; Meade, M.R.; Babler, B.L.; Indebe-touw, R.; Whitney, B.A.; Watson, C.; Wolfire, M.G.; et al.; 2005,ApJ, 635, 560Robin, A.C.; Reyl´e, C.; Derri`ere, S.; Picaud, S.; 2003, A&A, 409,523Rowles, J. & Froebrich, D. 2009, MNRAS, 395, 1640Scholz, A.; Froebrich, D.; Davis, C.J.; Meusinger, H.; 2010, MN-RAS, 406, 505 c (cid:13) , 1–, 1– ?? Froebrich & Ioannidis
Skrutskie, M.F., Cutri, R.M., Stiening, R., Weinberg, M.D.,Schneider, S., Carpenter, J.M., Beichman, C., et al., 2006, AJ,131, 1163Urquhart, J. S.; Moore, T. J. T.; Hoare, M. G.; Lumsden, S. L.;Oudmaijer, R. D.; Rathborne, J. M.; Mottram, J. C.; Davies, B.;et al.; 2011, MNRAS, 410, 1237Weidner, C.; Kroupa, P.; 2006, MNRAS, 365, 1333 c (cid:13) , 1– ?? bout the nature of Mercer 14 APPENDIX A: DETAILED IMAGES OF H FLOWS c (cid:13) , 1–, 1–
Skrutskie, M.F., Cutri, R.M., Stiening, R., Weinberg, M.D.,Schneider, S., Carpenter, J.M., Beichman, C., et al., 2006, AJ,131, 1163Urquhart, J. S.; Moore, T. J. T.; Hoare, M. G.; Lumsden, S. L.;Oudmaijer, R. D.; Rathborne, J. M.; Mottram, J. C.; Davies, B.;et al.; 2011, MNRAS, 410, 1237Weidner, C.; Kroupa, P.; 2006, MNRAS, 365, 1333 c (cid:13) , 1– ?? bout the nature of Mercer 14 APPENDIX A: DETAILED IMAGES OF H FLOWS c (cid:13) , 1–, 1– ?? Froebrich & Ioannidis
Figure A1.
Gray scale representation of the continuum subtracted H image of the outflows near Mercer 14. Here we show the northern part of the region.North is up, East to the left. c (cid:13) , 1– ?? bout the nature of Mercer 14 Figure A2.
Gray scale representation of the continuum subtracted H image of the outflows near Mercer 14. Here we show the southern part of the region.North is up, East to the left.c (cid:13) , 1–, 1–