The mass function of IC4665 revisited by the UKIDSS Galactic Clusters Survey
N. Lodieu, W.-J. de Wit, G. Carraro, E. Moraux, J. Bouvier, N. C. Hambly
aa r X i v : . [ a s t r o - ph . E P ] J un Astronomy&Astrophysicsmanuscript no. 16883ms c (cid:13)
ESO 2018November 20, 2018
The mass function of IC 4665 revisited by the UKIDSS GalacticClusters Survey ⋆ ⋆⋆⋆⋆⋆
N. Lodieu , , W.-J. de Wit , G. Carraro , E. Moraux , J. Bouvier , and N. C. Hambly Instituto de Astrof´ısica de Canarias, C / V´ıa L´actea s / n, E-38200 La Laguna, Tenerife, Spaine-mail: [email protected] Departamento de Astrof´ısica, Universidad de La Laguna (ULL), E-38205 La Laguna, Tenerife, Spain European Southern Observatory, Alonso de Cordoba 3107, Casilla 19001 Santiago 19, Chilee-mail: [email protected], [email protected] UJF-Grenoble 1 / CNRS-INSU, Institut de Plan´etologie et d’Astrophysique de Grenoble (IPAG) UMR 5274, Grenoble, F-38041,Francee-mail:
[email protected], [email protected] Scottish Universities’ Physics Alliance (SUPA), Institute for Astronomy, School of Physics & Astronomy, University of Edinburgh,Royal Observatory, Blackford Hill, Edinburgh EH9 3HJ, UKe-mail: [email protected]
November 20, 2018; November 20, 2018
ABSTRACT
Context.
Knowledge of the mass function in open clusters constitutes one way to constrain the formation of low-mass stars and browndwarfs as does the knowledge of the frequency of multiple systems and the properties of disks.
Aims.
The aim of the project is to determine the shape of the mass function in the low-mass and substellar regimes in the pre-mainsequence (27 Myr) cluster IC 4665, which is located at 350 pc from the Sun.
Methods.
We have cross-matched the near-infrared photometric data from the Eighth Data Release (DR8) of the UKIRT InfraredDeep Sky Survey (UKIDSS) Galactic Clusters Survey (GCS) with previous optical data obtained with the Canada-France-Hawaii(CFH) wide-field camera to improve the determination of the luminosity and mass functions in the low-mass and substellar regimes.
Results.
The availability of i and z photometry taken with the CFH12K camera on the Canada France Hawaii Telescope added strongconstraints to the UKIDSS photometric selection in this cluster, which is located in a dense region of our Galaxy. We have derived theluminosity and mass functions of the cluster down to J = ∼ ⊙ at the distance and ageof IC 4665 according to theoretical models. In addition, we have extracted new candidate members down to ∼
20 Jupiter masses in apreviously unstudied region of the cluster.
Conclusions.
We have derived the mass function over the 0.6–0.04 M ⊙ mass range and found that it is best represented by a log-normalfunction with a peak at 0.25–0.16 M ⊙ , consistent with the determination in the Pleiades. Key words.
Techniques: photometric — stars: low-mass, brown dwarfs — stars: luminosity function, mass function — infrared: stars— galaxy: open clusters and associations: individual (IC 4665)
1. Introduction
The initial mass function (IMF), the number of stars formed inthe galactic disk and in clusters, was first defined by Salpeter(1955). Later, Miller & Scalo (1979) and Scalo (1986) con-ducted a more accurate study of the IMF. It represents an impor-tant link between stellar and galactic evolution and is one of thekey tools for our understanding of the theories of star formation
Send o ff print requests to : N. Lodieu ⋆ This work is based in part on data obtained as part of the UKIRTInfrared Deep Sky Survey. The United Kingdom Infrared Telescope isoperated by the Joint Astronomy Centre on behalf of the Science andTechnology Facilities Council of the U.K. ⋆⋆ This work is partly based on observations obtained at the Canada-France-Hawaii Telescope (CFHT), which is operated by the NationalResearch Council of Canada, the Institut National des Sciences del’Univers of the Centre National de la Recherche Scientifique of France,and the University of Hawaii. ⋆⋆⋆
Tables A.1, B.1, and C.1 are only available in electronic form atthe CDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or viahttp: // cdsweb.u-strasbg.fr / cgi-bin / qcat?J / A + A / in the high-mass, low-mass, and substellar regimes. The currentknowledge drawn from the latest sets of data available in the so-lar neighbourhood and in clusters points towards a power-law orlog-normal representation of the IMF (Kroupa 2002; Chabrier2003; Larson 2005).Over the past decade, deep optical surveys conductedin 50–600 Myr old open clusters have uncovered numer-ous young pre-main-sequence stars and brown dwarfs.New stellar and substellar members have been identi-fied in clusters, including the Pleiades (Bouvier et al.1998; Zapatero Osorio et al. 1999; Dobbie et al. 2002b;Moraux et al. 2003; Bihain et al. 2006; Lodieu et al. 2007a), α Per (Stau ff er et al. 1999; Barrado y Navascu´es et al. 2002;Lodieu et al. 2005), IC 2391 (Barrado y Navascu´es et al. 2001,2004; Spezzi et al. 2009), M 35 (Barrado y Navascu´es et al.2001), Blanco 1 (Moraux et al. 2007), IC 4665 (de Wit et al.2006; Cargile & James 2010), the Hyades (Dobbie et al.2002a; Bouvier et al. 2008), and NGC 2547 (Naylor et al.2002; Je ff ries et al. 2004; Je ff ries & Oliveira 2005). Clustermembership in these works is generally assessed by N. Lodieu et al.: The mass function of IC 4665 revisited by the UKIDSS GCS proper motion (Hambly et al. 1999; Adams et al. 2001;Moraux et al. 2001; Deacon & Hambly 2004; Lodieu et al.2007a), near-infrared photometry (Zapatero Osorio et al. 1997;Barrado y Navascu´es et al. 2002; Lodieu et al. 2007a), andoptical spectroscopy for spectral typing, chromospheric activity,and surface gravity (Steele & Jameson 1995; Mart´ın et al.1996; Lodieu et al. 2005; Bihain et al. 2010). The I − J and I − K optical-to-infrared colours have proven their e ffi ciencyto weed out background giants and field dwarfs as empha-sised by optical-infrared surveys carried out in the Pleiades(Zapatero Osorio et al. 1997), α Per (Barrado y Navascu´es et al.2002; Lodieu et al. 2005), and σ Orionis (B´ejar et al. 2001;Caballero et al. 2007).The Two Micron All-sky Survey (2MASS; Cutri et al. 2003)and Deep Near-Infrared Survey (DENIS; Epchtein et al. 1997)provide
JHK photometry for all sources down to K s ∼ , made of five sub-surveys, is designed toreach three to four magnitudes deeper than 2MASS and coverseveral thousand of square degrees at infrared wavelengths. TheUKIDSS project is defined in Lawrence et al. (2007) and usesthe Wide Field Camera (WFCAM; Casali et al. 2007) installedon the UK InfraRed Telescope (UKIRT) and the Mauna KeaObservatory (MKO; Tokunaga et al. 2002) photometric systemdescribed in Hewett et al. (2006). The pipeline processing isdescribed in Irwin et al. (2009, in prep) and the WFCAMScience Archive (WSA) in Hambly et al. (2008).The Galactic Clusters Survey (GCS), one of the UKIDSSsub-surveys, aims at probing young brown dwarfs in ten star-forming regions and open clusters over large areas in five pass-bands ( ZY JHK ) across the 1.0–2.5 micron wavelength rangewith a second epoch in K . The main goal is to measure theform of the IMF (Salpeter 1955; Miller & Scalo 1979; Scalo1986) in the substellar regime to tackle important issues includ-ing the formation and spatial distribution of low-mass stars andbrown dwarfs. Early results for Upper Scorpius, in the Pleiades, σ Orionis, and in Praesepe are presented in Lodieu et al. (2006,2007b, 2008, 2011), Lodieu et al. (2007a), Lodieu et al. (2009),and (Baker et al. 2010), respectively.Among the open clusters targeted by the GCS is IC 4665,whose central coordinates are R.A. = h
40 and dec = ◦ .This cluster is of special interest because it is among the fewpre-main-sequence open clusters with an age between 10 and50 Myr. It has su ff ered little dynamical evolution and the mea-sured mass function is close to the real IMF because star-formation processes have ended. The cluster is young witha lithium age of 27 Myr (Manzi et al. 2008), consistent withthe nuclear age estimate (36 Myr; Mermilliod 1981), activ-ity probes ( <
40 Myr; Montes & Martin 1998), and other in-dependent studies (30–40 Myr; Lynga 1995; Dambis 1999;Kharchenko et al. 2005a). Several distance estimates are avail-able for IC 4665. Abt & Cha ff ee (1967) found 324 pc fromisochrone fitting, Lynga (1995) reported 430 pc in his cata-logue of open clusters and Mermilliod (1981) derived ∼ Extensive details on the data reduction of the UKIDSS images isavailable at http: // casu.ast.cam.ac.uk / surveys-projects / wfcam / technical,see also Irwin et al. (2004) pc. Independent studies based on Hipparcos observations de-rived a mean distance close to 350 pc: Dambis (1999) obtained370 ±
100 pc, Hoogerwerf et al. (2001) reported 385 ±
40 pc, andKharchenko et al. (2005a) inferred 352 pc. In this paper, we willadopt an age of 30 Myr and a distance of 350 pc. Previous studiesof the cluster provided a list of low-mass members based on theirspectral types, radial velocities, and lithium abundances (for asummary, see Prosser 1993). A recent deep optical ( I , z ) surveycovering the central 3.82 square degrees in the cluster identifiednew candidate members down to 0.03 M ⊙ (de Wit et al. 2006).Near-infrared photometry was extracted from 2MASS with addi-tional photometry obtained for a subsample of brown dwarf can-didates with the infrared camera on the Canada-France-Hawaii(CFH) telescope. de c ( J ) Filled dots: confirmed optical candidatesCrosses: rejected optical candidatesTriangles: New infrared candidates from GCS
Fig. 1.
Schematic view of the UKIDSS GCS coverage in IC 4665is shown with thick solid lines. We included known highprobability photometric candidates (star symbols) from Prosser(1993), the optical ( I , z ) candidates de Wit et al. (2006) con-firmed (filled circles) and rejected (plus symbols) by the GCSphotometry, as well as new photometric candidates identifiedin the GCS outside the optical coverage (open triangles). Amore detailed view of the optical coverage is shown in Fig. 2of de Wit et al. (2006).We present ZY JHK photometry for a ∼ . Lodieu et al.: The mass function of IC 4665 revisited by the UKIDSS GCS 3
2. The optical survey
The optical survey was described in detail in de Wit et al. (2006).We provide here a brief summary to bring our infrared observa-tions (Sect. 3) into context. The open cluster IC 4665 was tar-geted within the framework of a deep, wide-field, photometricsurvey of open clusters and star-forming regions aimed at study-ing the influence of the environment on tthe mass function. Theobservations, conducted in June 2002, surveyed 3.82 square de-grees in the cluster down to a completeness limit of I , z = ∼ ⊙ , assuming a distance of 350pc and an age of 50–100 Myr (Prosser 1993). We adopt the fol-lowing notation throughout the paper: Z (in capital letter) refersto the WFCAM Z filter with a narrow bandpass (centred at 0.8 µ m) and z refers to the CFH12K optical z filter. The selectionof cluster member candidates in the ( I − z , I ) colour-magnitudediagram yielded a total of 691 low-mass and 94 brown dwarfcandidates, assuming the age and distance indicated above. The2MASS catalogue was queried to provide infrared photome-try for candidates brighter than K s = JHK ) photometry was obtained for a subset of 101 can-didates with the CFHT infrared camera (CFHTIR). The numberof candidates selected from the optical colour-magnitude dia-gram alone and later rejected on the basis of their I − K coloursis about 35 % at faint magnitudes, showing the e ffi ciency of thenear-infrared photometry. The contamination by field stars wasalso estimated from two control fields. The derived mass func-tion after correction for statistical contamination was best fit bya lognormal with a mean mass of 0.32 M ⊙ , comparable to re-sults in the Pleiades (Mart´ın et al. 1998; Dobbie et al. 2002b;Tej et al. 2002; Moraux et al. 2003; Lodieu et al. 2007a), and α Per (Barrado y Navascu´es et al. 2002).
3. The UKIDSS GCS infrared survey
We extracted all point sources from the UKIDSS GCS DR8(September 2010) that are located within a three degree radiusaround the centre of IC 4665 using a similar Structure QueryLanguage (SQL) query to our work in the Pleiades (Lodieu et al.2007a). In summary, we selected only point sources detected in
ZY JHK , as well as point sources with
JHK photometry, butundetected in Z and Y . The final catalogue contains 303,978sources distributed in the sky as illustrated in Fig. 1. The cor-responding ( Z − J , Z ) colour-magnitude diagram is presented inFig. 2 for one out of ten point source. All sources fainter than Z = J = K = J ∼ I − z , z )colour-magnitude diagram (Figure 7 in de Wit et al. 2006) whereonly objects fainter than I =
15 mag were considered, corre-sponding to a mass of ∼ ⊙ according to 30 Myr isochrones(Bara ff e et al. 1998; Chabrier et al. 2000) shifted to a d = J ∼ K ∼ J = ∼ − J20181614 Z UKIDSS GCS DR8: IC 4665
Fig. 2. ( Z − J , Z ) colour-magnitude diagram for 4.3 square de-grees surveyed in IC 4665 by the UKIDSS GCS. One out of 10sources is plotted. Table 1.
Log of the observations.
Tile a R.A. b dec. b Date1 17 h m + ◦ ′ h m + ◦ ′ h m + ◦ ′ h m + ◦ ′ h m + ◦ ′ h m + ◦ ′ h m + ◦ ′ h m + ◦ ′ a Each WFCAM tile is 54 by 54 arcmin aside and was observed in allfive passbands (
ZY JHK ). Average seeing measured on the imagesand in each individual filter ranged from 0.75 to 1.0 arcsec b Central coordinates (in J2000) of each tile from UKIDSS GCS DR8 tween the 2MASS (observations taken between 26 April and12 June 2000; Cutri et al. 2003) and GCS observations (May2006) for all sources brighter than J = Z -band photomery) out of the 393 optical candidates inthe I = K -band photometry for 101 faint candidates (i.e. ∼
4. Cross-matching with optical data
The depth and coverage of the GCS observations allow us to pro-vide infrared counterparts in five broad-band filters (
ZY JHK ) forthe large majority of optical candidates down to the complete-ness limit of the CFH12K survey ( ∼ ⊙ ). As mentioned inthe previous section, we have photometry for 313 out of the 393(77–78%) optical candidates with I = N. Lodieu et al.: The mass function of IC 4665 revisited by the UKIDSS GCS
Table 2.
Number of optical candidates with GCS photometry(opt) for di ff erent intervals of Z magnitude after the di ff erentselection steps. Range opt S1 a S2 b S3 c S4 d S5 e R1 f R2 g Z = Z = Z = Z = Z = Z = Z = Z ≤ Z ≥ Z = a Number of optically selected candidates kept after applying the Z − J selection (step 1; S1) b Number of candidates kept after the Z − K selection (step 2; S2) c Number of candidates kept after the Y − J selection (step 3; S3) d Number of candidates kept after the Y − K selection (step 4; S4) e Number of candidates kept after the J − K selection (step 5; S5) f The R1 percentage refers to the S5 / opt ratio g The R2 percentages refers to the S5 / S1 ratio
We have plotted various colour-magnitude diagrams usingoptical and infrared magnitudes to assess the membership of theoptically selected candidates published in de Wit et al. (2006).On the left-hand side of Fig. 3 we show the ( z − J , z ) and the( Z − J , Z ) colour-magnitude diagrams. The same diagrams aredisplayed for the z − K and Z − K colours on the right-hand sideof Fig. 3. The majority of optical candidates rejected by the z − J and z − K colours are also discarded by the Z − J and Z − K colours, showing the e ffi ciency of optical-to-infrared colours inremoving interlopers (Fig. 3).Additional colour-magnitude diagrams employed to rejectphotometric contaminants include the ( Y − J , Y ) and ( J − K , J )presented in Fig. 4. We also examined the ( H − K , J − H ) colour-colour diagram to assess the nature of the contaminants. Thisdiagram, not shown here, suggests that the contamination in theoptical selection originates from a mixture of reddened early-type dwarfs and field M dwarfs. To verify the membership of the 785 optical candidates fromde Wit et al. (2006) using our near-infrared photometry, wecross-matched that list with the full GCS catalogue with a match-ing radius of five arcsec and found 654 common objects. Wefound that 58 out 785 optical candidates lie in the region outsideof GCS coverage. Therefore, 785 − =
727 objects are within theGCS coverage, implying that 727 − =
73 sources are not re-covered by our GCS selection. The GCS detection of these 73sources, located at the position of the optical source, did notsatisfy the criteria set in the SQL query. We found that 19 ofthe 73 objects are missing photometry in (at least) one of the
JHK passbands, and the remaining sources do not satisfy thepoint sources criteria (
Class and
ClassStat parameters in atleast one filter). This problem is inherent to the SQL query andrepresents about 10% of the total number of candidates, a valueslightly higher than for σ Orionis (4.4–7.8%; Lodieu et al. 2009).We note that 35 out of the 54 ( ∼ ZY JHK mag-nitudes from GCS DR8 would remain as photometric member candidates after applying the colour selections described in thenext paragraph.We generated various colour-magnitude diagrams (Figs. 3 &4) to design selection cuts based on the position of the 654 opti-cal candidates with
ZY JHK photometry. Looking at these can-didates in the several colour-magnitude diagrams, we generateda subsample of candidates by extracting only those to the rightof lines defined in each diagram below: • ( Z − J , Z ) = (0.72,14.0) to (1.15,17.8) • ( Z − J , Z ) = (1.15,17.8) to (2.3,21.5) • ( Z − K , Z ) = (1.5,14.0) to (2.0,17.5) • ( Z − K , Z ) = (2.0,17.5) to (3.4,21.5) • ( Y − J , Y ) = (0.45,14.0) to (0.55,17.0) • ( Y − J , Y ) = (0.55,17.0) to (1.15,20.5) • ( Y − K , Y ) = (1.3,14.0) to (1.45,17.0) • ( Y − K , Y ) = (1.45,17.0) to (2.2,20.5) • ( J − K , J ) ≥ J = • ( J − K , J ) = (0.8,17.0) to (1.3,19.5)We are left with 493 optical candidates that satisfy those cri-teria, and accordingly we confirm their candidacy as photometricmembers on the basis of the five-band infrared photometry. Thissample will be refered to as “high-probability members” in therest of the paper. These members have a magnitude range thatspans Z = Iz ) and infrared ( ZY JHK ) magnitudes,as well as the proper motions in right ascension and declinationexpressed in mas / yr. We emphasise that proper motions for ob-jects fainter than J = ∼
20% to ∼ Z = ⊙ ; Table 2), in agreement with theconclusions drawn by de Wit et al. (2006) from a limited sampleof K -band observations and the study of two control fields (seealso Moraux et al. 2003) in this open cluster located at lowergalactic latitude and towards a dense region (see e.g. Prosser1993). The most powerful selection criterion clearly is the Z − J colour because the other colour cuts remove on average less than10% of the candidates (R2 percentage in Table 2).We looked further into the level of contamination expectedin the GCS data. We selected two control fields with galacticcoordinates similar to IC 4665. These are located about 0.2 de-grees outside the cluster radius determined by Kharchenko et al.(2005b). The two control fields contain 6610 and 7725 sourceswith ZY JHK photometry, respectively, compared to a total of7220 sources in the central 0.16 square degrees in IC 4665. Afterapplying the various selection cuts designed above, we are leftwith 89 candidates in the central region of the cluster comparedwith 57 and 50 candidates in the two control fields. The averagelevel of contamination is of the order of 55–60% and is compa-rable in the bright and faint magnitude ranges. This level of con-tamination is very likely an upper limit because the control fieldsare located very close to the cluster, much closer than the controlfields used in the optical study (0.2 vs 3 degrees). The estimated . Lodieu et al.: The mass function of IC 4665 revisited by the UKIDSS GCS 5
CFH12K − J20181614 z C F H K CFH12K − K20181614 z C F H K − J20181614 Z sun sun sun sun sun sun sun sun sun sun sun − K20181614 Z sun sun sun sun sun sun sun sun sun sun sun Fig. 3.
Various colour-magnitude diagrams for ∼ ff e et al. 1998) and DUSTY (dashed line; Chabrier et al. 2000)isochrones shifted to a distance of 350 pc. Masses are given on the right-hand side of the plot for the 30 Myr isochrone. Overplottedare the lines for photometric selection as described in Section 4.2. Filled circles represent the optically selected member candidatesfrom de Wit et al. (2006) confirmed by infrared photometry. Crosses are optically selected candidates rejected as cluster memberson the basis of their infrared colours. Top left: ( z CFH K − J , z CFH K ) diagram. Top right: ( z CFH K − K , z CFH K ) diagram. Bottomleft: ( Z − J , Z ) diagram. Bottom right: ( Z − K , Z ) diagram. Note that the 30 Myr NextGen and DUSTY isochrones are plotted on thetop hand side diagrams with the set of filters indicated on the axis.contamination is higher than the levels derived for the Pleiades(Moraux et al. 2001) and Alpha Per (Barrado y Navascu´es et al.2002), most likely because of the lower galactic latitude ofIC 4665. Considering the low-mass and substellar sequence defined bythe high-probability members, we have identified new candi-dates over the full area covered by the GCS. The same colourcuts as defined in Sect. 4.2 returned a total of 2236 sources with Z = N. Lodieu et al.: The mass function of IC 4665 revisited by the UKIDSS GCS − J20191817161514 Y sun sun sun sun sun sun sun sun sun sun sun sun − K19181716151413 J sun sun sun sun sun sun sun sun sun sun sun sun Fig. 4. ( Y − J , Y ) and ( J − K , J ) colour-magnitude diagrams for a ∼ ff e et al. 1998) and DUSTY (dashed line; Chabrier et al. 2000) isochrones shifted to a distanceof 350 pc. Masses are given on the right-hand side of each plot for the 30 Myr isochrones. Filled circles represent the opticallyselected candidates from de Wit et al. (2006) whose membership is confirmed by their infrared photometry. Crosses are opticallyselected candidates rejected as cluster members on the basis of their infrared colours. Overplotted are the lines for photometricselection as described in Section 4.2.optical coverage whereas 1372 are located in the common areabetween the optical and GCS surveys, including the 493 opti-cal candidates confirmed as photometric members by the GCS(Sect. 4.2). Hence, we have 1372 − =
879 potential new can-didates (643 are brighter than Z = Z − J , Z ) diagram, suggesting that thelarge majority are indeed non-members. This trend is also ob-served in the ( Y − J , Y ) diagram and clearly confirmed in the op-tical ( i − z CFHT , i CFHT ) diagram. Furthermore, 635 and 30 ( = i and z filters, respec-tively. Of these 214 candidates, only 26 satisfy the colour cri-teria set by de Wit et al. (2006): 23 were not selected by theoptical study because their quality flags were greater than 1,another one was excluded from the criteria set for the flux ra-dius parameter in SEXtractor, and the remaining two (UGCSJ174823.92 + + −
2) candidates selected from the GCS but unselected bythe CFH12K survey, the level of contamination of the GCS se-lected in IC 4665 would be of the order of 64%(879 / / − =
864 new candidates fromthe GCS located outside the optical coverage, including 664brighter than Z = To estimate the spatial distribution of the candidates selectedfrom the UKIDSS data alone, we calculate the number of objectsper square degree located within an annulus R , R + dR centred onthe cluster centre α = h m s , δ = d m s (de Wit et al.2006). The corresponding radial profile is shown in the left-handside of Figure 5. This distribution is then fitted by a King profile(King 1962) plus a constant assuming that the density of con-taminants n cont is uniform: n ( x ) = k " √ + x − √ + x t + n cont , (1)where k is a normalisation constant, x = ( r / r c ) and x t = ( r t / r c ) . The distance from the cluster centre is r , r c the coreradius and r t the tidal radius.To find the best fit minimizing χ , we left all these parame-ters free, except for the tidal radius that we forced to be between6.4 and 12.1 pc. Indeed, Piskunov et al. (2008) found the clustertidal radius to be r t = . ± . . ± . k = ±
189 sq.deg. − , r c = ± . r t = ± . n cont = ± ± ± k = ±
40 sq.deg. − , r c = ± . r t = ± . n cont = ± . Lodieu et al.: The mass function of IC 4665 revisited by the UKIDSS GCS 7 Fig. 5.
Radial distributions for the full GCS coverage (left) and the common area between the GCS and the CFH12K surveys (right).Again using that the common area is 3.475 square degrees, wefind a contamination level of 50 ± The mean proper motion of the IC 4665 cluster is estimatedto be ( µ α cos δ , µ δ ) = ( − ± − ± / yr from 30members included in the study of Kharchenko et al. (2005b),in agreement with the average proper motion of 13 membersmeasured by Hipparcos (Hoogerwerf et al. 2001). For more de-tails, we refer the reader to the discussion in Section 4.1 ofde Wit et al. (2006). Hence, the cluster’s motion is low and noteasy to separate from field stars and reddened background giants.However, astrometric information can be used to reject photo-metric candidates with large proper motions.We derived proper motions for all point sources identified inIC 4665 and brighter than J = ∼ / yr for J ≤ σ (corresponding to a total proper motion lessthan 45 mas / yr; Fig. 6) cut, which should optimize the selection(Lodieu et al. 2011).Assuming those parameters, we are left with 278 out the 316optical candidates brighter than J = J ≤ We attempted to identify fainter and lower mass candidates us-ing sources that are Z -band non-detections. We applied the samecolour criteria in Y − J , Y − K , and J − K colours as presentedin Sect. 4.2. However, we limited our selection to J = -100 -50 0 50 100pmRA (mas/yr)-100-50050100 p m D E C ( m a s / y r) Fig. 6.
Vector point diagram for all optical candidates confirmedas probable members by the GCS photometry (black dots) andphotometric non-members (crosses). Typical error bars on theproper motion are 15 mas / yr.6–8. The SQL query returned three potential candidates after re-moving those clearly detected on the Z images. However, two ofthem are actually detected on the deep CFH12K images (Table3), suggesting that they do not belong to the cluster. These threecandidates are listed at the top of Table 3.We repeated the same procedure for Z and Y non-detections,using the J − K colour as sole criterion. The query returned twogood candidates after removal of objects located at the edge ofthe detector and those clearly detected on the Z and Y images.These two candidates are listed at the bottom of Table 3.We repeated those selections in the cluster centre and the twocontrol fields, but the few objects returned in each area were all N. Lodieu et al.: The mass function of IC 4665 revisited by the UKIDSS GCS
Table 3.
Faint
Y JHK (top) and
JHK (bottom) candidates iden-tified in the UKIDSS GCS.
R.A. a dec a Y a J a H a K a + b + b + c + c + ca Coordinates and photometry from UKIDSS GCS DR8 b Detected on the deep CFH12K i and z images c Undetected on the deep CFH12K images detected in Z for the Y JHK candidates and in Z and Y for the JHK candidates.
5. The luminosity and mass functions
In this section, we derive the cluster luminosity and mass func-tions based on the sample of optical photometric candidates con-firmed by the GCS photometry. This sample contains the 493cluster member candidates discussed in Sect. 4.2. We did notcorrect the luminosity and mass function for the level of con-tamination estimated in section 4.5 because this sample repre-sents the highest quality sample extracted to date. We expect alow level of contamination in this photometric sample, whichcombines optical and near-infrared photometry: about 10% ofthem may be proper motion non-members (see Section 4.5) andanother 5–10% spectroscopic non-members according to ourspectroscopic study of Upper Sco (Lodieu et al. 2011). An in-dependent photometric and spectroscopic study of Blanco 1 byMoraux et al. (2007) suggests that the expected number of con-taminants is low after combining optical and infrared photomet-ric bands.
This sample of optical photometric candidates with infraredphotometry contains 493 sources in the J ∼ / substellar limit is at K = J = ff e et al. 1998)and DUSTY (Chabrier et al. 2000) models to be most consistentwith the age of the cluster. Thus, our sample contains ∼
91% ofstars and 8–9% (40–46 out of 493) substellar candidates.Figure 7 displays the luminosity function, i.e. the number ofstars as a function of J magnitude. The number of objects in-creases up to J ∼
15 mag, then decreases down to J ∼ + ( √ ( dN + . √ ( dN − .
25) assuming a confidence level ofone sigma.
To convert magnitudes into masses we used the NextGen andDUSTY models at 30 Myr from the Lyon group (Bara ff e et al.1998; Chabrier et al. 2000). Estimated masses at a given magni-tude and directly taken from the mass-magnitude relationship
13 14 15 16 17 18 19J magnitude020406080100120 N u m be r o f ob j e c t s Fig. 7.
Cluster luminosity function: number of objects as a func-tion of J magnitude (Table 4). Error bars are Gehrels errors. Table 4.
Numbers for the luminosity and mass functions. ∆ Mag a J bc ∆ Mass c dN d dN / dM e dN / dlogM e a Magnitude ranges in the J -band b Central magnitudes ( J c ) c Mass ranges ( ∆ Mass). The mass intervals were chosen identical tothose in de Wit et al. (2006) for direct comparison. We assumed anage of 30 Myr and a distance of 350 pc for IC 4665 d Numbers of objects per magnitude interval e Mass functions in linear (dN / dM) and logarithmic (dN / dlogM)scales extracted from the models. Samples of masses are shown onthe right-hand side of the colour-magnitude diagrams in Fig. 3.We used those relations to derive masses and construct the massfunction. The highest mass cluster member candidate has ∼ ⊙ and the lowest mass object has 0.021 M ⊙ , the latter beingoutside the magnitude and mass ranges shown in Table 4. We didnot attempt to correct the mass function for binaries. If IC 4665is indeed a pre-main-sequence cluster, the current system massfunction should reproduce the IMF fairly well (Salpeter 1955;Miller & Scalo 1979; Scalo 1986).Figure 8 shows the system mass function for IC 4665 overthe 0.66–0.02 M ⊙ mass range (filled circles and solid line). Themass function in logarithmic scale peaks around 0.25–0.16 M ⊙ and decreases down to our completeness limit at ∼ ⊙ . Lodieu et al.: The mass function of IC 4665 revisited by the UKIDSS GCS 9 l og ( d N / d l og M ) IC 4665 (30 Myr; 350 pc): GCS; this workIC 4665 (50 Myr; 350 pc): de Wit et al. (2006)Pleiades: Lodieu et al. (2007)
Fig. 8.
Cluster mass function assuming an age of 30 Myr(Manzi et al. 2008) and a distance of 350 pc. The conver-sion from magnitudes into masses made use of the NextGenand DUSTY models (Bara ff e et al. 1998; Chabrier et al. 2000).Values are given in Table 4. The vertical dot-dashed lines repre-sent the GCS completeness limits in IC 4665.(dot-dashed line in Fig. 8). The completeness in terms of mass issimilar to our study in the Pleiades (red open diamonds and solidline; Lodieu et al. 2007a) because the di ff erence in age betweenthe Pleiades and IC 4665 (125 Myr vs 27 Myr) is partly com-pensated by the larger distance of IC 4665 (Hoogerwerf et al.2001). The penultimate point is probably real and lies higherthan the point before at higher masses. The same trend is ob-served in the previous mass function in IC 4665 published byde Wit et al. (2006) as well as in the Pleiades mass function.Moraux et al. (2007) pointed out that this rise is not caused byan intrinsic rise of the mass function but rather reflects a dropin the mass-luminosity relationship owing to the onset of dust inthe atmospheres of low-mass stars (the M7 / M8 gap as definedby Dobbie et al. 2002c). The last point of our mass function,however, su ff ers from incompleteness and represents a lowerlimit because the mass range is below our completeness limit(Table 4). Similarly, the first point of the mass function is in-complete because of the saturation of the deep CFH12K images(de Wit et al. 2006).In Fig. 8 we overplotted the published mass function (blueopen triangles and solid line) of IC 4665 by de Wit et al. (2006).The main di ff erence lies in the choice of the age: we have usedthe recent age estimate of ∼
30 Myr (Manzi et al. 2008) basedon the lithium method (Rebolo et al. 1992; Basri et al. 1996),whereas de Wit et al. (2006) showed the mass function for 50and 100 Myr following the earlier work by Prosser (1993).We emphasise the large uncertainty on the distance of IC 4665(385 ±
40 pc; Hoogerwerf et al. 2001). This range in distancewould translate into a shift in mass with small di ff erences inthe substellar regime ( ∼ ⊙ ) but more significant at highermasses (several tens of Jupiter masses for masses larger than 0.5M ⊙ ). We observe a sharper decline of the mass function derived from the GCS data than the one derived by de Wit et al. (2006),suggesting that the level of contamination of the optical surveybelow 0.1 M ⊙ was higher than the one estimated by this studydespite the correction applied by de Wit et al. (2006).The mass function derived in our study is very similarto the Pleiades mass function drawn from the GCS dataset(Lodieu et al. 2007a) down to 0.05 M ⊙ . Both mass functionspeak around the same mass (between 0.16 M ⊙ and 0.25 M ⊙ )followed by a decrease down to the completeness limit of theGCS survey. The di ff erence in mass observed at the M7 / M8gap (0.05 M ⊙ for the Pleiades vs 0.035 M ⊙ for IC 4665) couldarise from the large uncertainty on the distance of IC 4665(40 pc; Hoogerwerf et al. 2001) compared to the 5 pc error onthe Pleiades’s distance (Johnson 1957; Gatewood et al. 2000;Southworth et al. 2005). Moreover, the Pleiades mass functionwas based on a photometric and proper motion sample, whereasthe IC 4665 dataset is purely photometric because we do nothave proper motion over the full magnitude range probed by theGCS.
6. Conclusions and future work
We have presented the outcome of a wide-field near-infraredsurvey of IC 4665 conducted by the UKIDSS Galactic ClustersSurvey. The cross-matching of the GCS survey with a previousoptical survey to a similar depth leads to a revision of the mem-bership of low-mass stars and brown dwarfs in IC 4665 based ontheir optical-to-infrared and infrared colours. The main resultsof this paper can be summarised as follows:1. We confirmed a total of 493 photometric candidates iden-tified in a pure optical survey as high-probability membercandidates using five infrared (
ZY JHK ) filters in addition tothe two optical ( Iz ) passbands.2. We identified new cluster member candidates in a previouslyunstudied region of the cluster.3. We derived the luminosity and mass functions in the J = ⊙ at an age of 30 Myr and a distance of 350 pc.4. We found that the mass function is similar in shape to thePleiades. It is best represented by a lognormal function peak-ing around 0.25–0.16 M ⊙ over the 0.66–0.04 M ⊙ mass range.The full area of the cluster will be observed again in K as partof the GCS, which will provide additional proper motion mea-surements over the full magnitude range to add proper motioninformation to the photometric selection. We are planning anoptical spectroscopic follow-up with the AAOmega multi-fibrespectrograph installed on the 3.9-m Australian AstronomicalTelescope (Lewis et al. 2002; Sharp et al. 2006) to confirm themembership of all optical and GCS candidates. This spectro-scopic follow-up will provide (i) a full spectroscopic mass func-tion from 0.65 to 0.04 M ⊙ , (ii) a better estimate of the level ofcontamination in the optical and GCS surveys, and (iii) a fullspectroscopic sequence of 30 Myr-old late-M dwarfs. Moreover,optical and near-infrared spectroscopy of IC 4665 members willbe extremely valuable to bridge the gap between star-formingregions like Upper Sco (5 Myr; Preibisch & Zinnecker 2002)and older open clusters like the Pleiades (125 Myr; Stau ff er et al.1998) to define a temperature scale at 30 Myr. For this reason,IC 4665 can serve as a benchmark to study the evolution of thebinary properties in the substellar regime, investigate the roleof gravity as a function of age, the evolution of disks, and testcurrent evolutionary models. Finally, deeper surveys of IC 4665 may reveal late-L and T dwarf photometric candidates of inter-mediate age that would be ideal targets for photometric and spec-troscopic studies with upcoming facilities such as the E-ELT andJames Webb Space Telescope. Acknowledgements.
NL acknowledges funding from the Spanish Ministry ofScience and Innovation through the Ram´on y Cajal fellowship number 08-303-01-02 and the national program AYA2010-19136. NL thanks ESO for the shortbut productive stay in May 2008 and the LAOG for his visit in December2010. We are grateful to Isabelle Bara ff e for providing us with the NextGen andDUSTY models for the CFHT / CFH12K and UKIRT / WFCAM filters. We thankDavid Pinfield for his comments that improved the content of the paper.This research has made use of the Simbad database, operated at the Centrede Donn´ees Astronomiques de Strasbourg (CDS), and of NASA’s AstrophysicsData System Bibliographic Services (ADS) and the 2MASS database.This work is based in part on data from UKIDSS project and the CFH12Kcamera. The United Kingdom Infrared Telescope is operated by the JointAstronomy Centre on behalf of the U.K. Science Technology and FacilityCouncil. The Canada-France-Hawaii Telescope (CFHT) is operated by theNational Research Council of Canada, the Institut National des Sciences del’Univers of the Centre National de la Recherche Scientifique of France, andthe University of Hawaii.
References
Abt, H. A. & Cha ff ee, F. H. 1967, ApJ, 148, 459Adams, J. D., Stau ff er, J. R., Monet, D. G., Skrutskie, M. F., & Beichman, C. A.2001, AJ, 121, 2053Baker, D. E. A., Jameson, R. F., Casewell, S. L., et al. 2010, MNRAS, 408, 2457Bara ff e, I., Chabrier, G., Allard, F., & Hauschildt, P. H. 1998, A&A, 337, 403Barrado y Navascu´es, D., Stau ff er, J. R., Brice˜no, C., et al. 2001, ApJS, 134, 103Barrado y Navascu´es, D., Stau ff er, J. R., & Jayawardhana, R. 2004, ApJ, 614,386Barrado y Navascu´es, D., Bouvier, J., Stau ff er, J. R., Lodieu, N., &McCaughrean, M. J. 2002, A&A, 395, 813Barrado y Navascu´es, D., Stau ff er, J. R., Bouvier, J., & Mart´ın, E. L. 2001, ApJ,546, 1006Basri, G., Marcy, G. W., & Graham, J. R. 1996, ApJ, 458, 600B´ejar, V. J. S., Mart´ın, E. L., Zapatero Osorio, M. R., et al. 2001, ApJ, 556, 830Bihain, G., Rebolo, R., B´ejar, V. J. S., et al. 2006, A&A, 458, 805Bihain, G., Rebolo, R., Zapatero Osorio, M. R., B´ejar, V. J. S., & Caballero, J. A.2010, A&A, 519, A93Bouvier, J., Kendall, T., Meeus, G., et al. 2008, A&A, 481, 661Bouvier, J., Stau ff er, J. R., Mart´ın, E. L., et al. 1998, A&A, 336, 490Caballero, J. A., B´ejar, V. J. S., Rebolo, R., et al. 2007, A&A, 470, 903Cargile, P. A. & James, D. J. 2010, ArXiv e-printsCasali, M., Adamson, A., Alves de Oliveira, C., et al. 2007, A&A, 467, 777Chabrier, G. 2003, PASP, 115, 763Chabrier, G., Bara ff e, I., Allard, F., & Hauschildt, P. 2000, ApJ, 542, 464Cutri, R. M., Skrutskie, M. F., van Dyk, S., et al. 2003, 2MASS All Sky Catalogof point sources, 2246Dambis, A. K. 1999, Astronomy Letters, 25, 7de Wit, W. J., Bouvier, J., Palla, F., et al. 2006, A&A, 448, 189Deacon, N. R. & Hambly, N. C. 2004, A&A, 416, 125Dobbie, P. D., Kenyon, F., Jameson, R. F., et al. 2002a, MNRAS, 329, 543Dobbie, P. D., Kenyon, F., Jameson, R. F., et al. 2002b, MNRAS, 335, 687Dobbie, P. D., Pinfield, D. J., Jameson, R. F., & Hodgkin, S. T. 2002c, MNRAS,335, L79Epchtein, N., de Batz, B., & Capoani, L. e. 1997, The Messenger, 87, 27Gatewood, G., de Jonge, J. K., & Han, I. 2000, ApJ, 533, 938Gehrels, N. 1986, ApJ, 303, 336Hambly, N. C., Collins, R. S., Cross, N. J. G., et al. 2008, MNRAS, 384, 637Hambly, N. C., Hodgkin, S. T., Cossburn, M. R., & Jameson, R. F. 1999,MNRAS, 303, 835Hewett, P. C., Warren, S. J., Leggett, S. K., & Hodgkin, S. T. 2006, MNRAS,367, 454Hoogerwerf, R., de Bruijne, J. H. J., & de Zeeuw, P. T. 2001, A&A, 365, 49Irwin, M. J., Lewis, J., Hodgkin, S., et al. 2004, in Optimizing Scientific Returnfor Astronomy through Information Technologies. Edited by Quinn, Peter J.;Bridger, Alan. Proceedings of the SPIE, Volume 5493, pp. 411-422 (2004).,ed. P. J. Quinn & A. Bridger, 411–422Je ff ries, R. D., Naylor, T., Devey, C. R., & Totten, E. J. 2004, MNRAS, 351,1401Je ff ries, R. D. & Oliveira, J. M. 2005, MNRAS, 358, 13Johnson, H. L. 1957, ApJ, 126, 121Kharchenko, N. V., Piskunov, A. E., R¨oser, S., Schilbach, E., & Scholz, R.-D.2005a, A&A, 438, 1163 Kharchenko, N. V., Piskunov, A. E., R¨oser, S., Schilbach, E., & Scholz, R.-D.2005b, A&A, 438, 1163King, I. 1962, AJ, 67, 471Kroupa, P. 2002, Science, 295, 82Larson, R. B. 2005, MNRAS, 359, 211Lawrence, A., Warren, S. J., Almaini, O., et al. 2007, MNRAS, 379, 1599Lewis, I. J., Cannon, R. D., Taylor, K., et al. 2002, MNRAS, 333, 279Lodieu, N., Dobbie, P. D., Deacon, N. R., et al. 2007a, MNRAS, 380, 712Lodieu, N., Dobbie, P. D., & Hambly, N. C. 2011, A&A, 527, A24Lodieu, N., Hambly, N. C., & Jameson, R. F. 2006, MNRAS, 373, 95Lodieu, N., Hambly, N. C., Jameson, R. F., & Hodgkin, S. T. 2008, MNRAS,383, 1385Lodieu, N., Hambly, N. C., Jameson, R. F., et al. 2007b, MNRAS, 374, 372Lodieu, N., McCaughrean, M. J., Barrado Y Navascu´es, D., Bouvier, J., &Stau ff er, J. R. 2005, A&A, 436, 853Lodieu, N., Zapatero Osorio, M. R., Rebolo, R., Mart´ın, E. L., & Hambly, N. C.2009, A&A, 505, 1115Lynga, G. 1995, VizieR Online Data Catalog, 7092, 0Manzi, S., Randich, S., de Wit, W. J., & Palla, F. 2008, A&A, 479, 141Mart´ın, E. L., Rebolo, R., & Zapatero Osorio, M. R. 1996, ApJ, 469, 706Mart´ın, E. L., Zapatero Osorio, M. R., & Rebolo, R. 1998, in ASP Conf. Ser.134: “Brown Dwarfs and Extrasolar Planets”, eds. R. Rebolo, E. L. Mart´ın,and M. R. Zapatero Osorio, p 507Mermilliod, J. C. 1981, A&A, 97, 235Miller, G. E. & Scalo, J. M. 1979, ApJS, 41, 513Montes, D. & Martin, E. L. 1998, A&AS, 128, 485Moraux, E., Bouvier, J., & Stau ff er, J. R. 2001, A&A, 367, 211Moraux, E., Bouvier, J., Stau ff er, J. R., Barrado y Navascu´es, D., & Cuillandre,J.-C. 2007, A&A, 471, 499Moraux, E., Bouvier, J., Stau ff er, J. R., & Cuillandre, J.-C. 2003, A&A, 400, 891Naylor, T., Totten, E. J., Je ff ries, R. D., et al. 2002, MNRAS, 335, 291Piskunov, A. E., Schilbach, E., Kharchenko, N. V., R¨oser, S., & Scholz, R. 2008,A&A, 477, 165Preibisch, T. & Zinnecker, H. 2002, AJ, 123, 1613Prosser, C. F. 1993, AJ, 105, 1441Rebolo, R., Mart´ın, E. L., & Magazz`u, A. 1992, ApJL, 389, L83Salpeter, E. E. 1955, ApJ, 121, 161Scalo, J. M. 1986, Fundamentals of Cosmic Physics, 11, 1Sharp, R., Saunders, W., Smith, G., et al. 2006, in Presented at the Societyof Photo-Optical Instrumentation Engineers (SPIE) Conference, Vol. 6269,Ground-based and Airborne Instrumentation for Astronomy. Edited byMcLean, Ian S.; Iye, Masanori. Proceedings of the SPIE, Volume 6269, pp.62690G (2006).Southworth, J., Maxted, P. F. L., & Smalley, B. 2005, A&A, 429, 645Spezzi, L., Pagano, I., Marino, G., et al. 2009, A&A, 499, 541Stau ff er, J. R., Barrado y Navascu´es, D., Bouvier, J., et al. 1999, ApJ, 527, 219Stau ff er, J. R., Schultz, G., & Kirkpatrick, J. D. 1998, ApJL, 499, 219Steele, I. A. & Jameson, R. F. 1995, MNRAS, 272, 630Tej, A., Sahu, K. C., Chandrasekhar, T., & Ashok, N. M. 2002, ApJ, 578, 523Tokunaga, A. T., Simons, D. A., & Vacca, W. D. 2002, PASP, 114, 180Zapatero Osorio, M. R., Mart´ın, E. L., & Rebolo, R. 1997, A&A, 323, 105Zapatero Osorio, M. R., Rebolo, R., Mart´ın, E. L., et al. 1999, A&AS, 134, 537 Appendix A: Optical candidates confirmed byinfrared photometryAppendix B: Optical candidates rejected byinfrared photometryAppendix C: New photometric candidates identifiedin the GCS . Lodieu et al.: The mass function of IC 4665 revisited by the UKIDSS GCS 11
Table A.1.
Coordinates, optical and near-infrared photometry, proper motions (in mas / yr), luminosities (in L / L ⊙ ), and mass (in M ⊙ )for the 493 optically selected cluster member candidates in IC 4665 confirmed as such by their infrared colours extracted from theEighth Data Release of the UKIDSS Galactic Clusters Survey UGCS J. . . a R.A. b Dec. b I c z c Z d Y d J d H d K d µ α cos δ e µ e δ L / L ⊙ Massh m s ◦ ’ ” mag mag mag mag mag mag mag mas / yr mas / yr17:42:05.93 + + − − − + + − − a Name following the IAU and UKIDSS nomenclatures (see http: // / archive / archive.html for more details). b Coordinates in J2000 from UKIDSS GCS DR8 c Optical ( I , z ) photometry from the CFH12K survey (de Wit et al. 2006) d Near-infrared (
ZY JHK ) photometry from UKIDSS GCS DR8 e Proper motions (mas / yr) in right ascension and declination from the GCS vs 2MASS cross-match only for sources brighter than J = Table B.1.
Coordinates, optical and near-infrared photometry, and proper motions for 161 optically selected candidates in IC 4665classified as photometric non-members based on their infrared colours extracted from the Eighth Data Release of the UKIDSSGalactic Clusters Survey
UGCS J. . . a R.A. b Dec. b I c z c Z d Y d J d H d K d µ α cos δ e µ e δ h m s ◦ ’ ” mag mag mag mag mag mag mag mas / yr mas / yr174220.48 + + − + + − a Name following the IAU and UKIDSS nomenclatures (see http: // / archive / archive.html for more details). b Coordinates in J2000 from UKIDSS GCS DR8 c Optical ( I , z ) photometry from the CFH12K survey (de Wit et al. 2006) d Near-infrared (
ZY JHK ) photometry from UKIDSS GCS DR8 e Proper motions (mas / yr) in right ascension and declination from the GCS vs 2MASS cross-match only for sources brighter than J = Table C.1.
Coordinates and near-infrared photometry for the new cluster member candidates in IC 4665 extracted from the UKIDSSGCS DR8
UGCS J. . . a R.A. b Dec. b Z b Y b J b H b K b µ α cos δ c µ c δ UGCS J17:42:01.72 + + + + a Name following the IAU and UKIDSS nomenclatures (see http: // / archive / archive.html for more details). b Coordinates and photometry from UKIDSS GCS DR8 c Proper motions (mas / yr) in right ascension and declination from the GCS vs 2MASS cross-match only for sources brighter than J ==