The historical record of massive star formation in Cygnus
AAstronomy & Astrophysics manuscript no. cyg˙msg˙wide c (cid:13)
ESO 2020September 30, 2020
The historical record of massive star formation in Cygnus (cid:63)
F. Comer´on , A.A. Djupvik , N. Schneider , and A. Pasquali European Southern Observatory, Karl-Schwarzschild-Strasse 2, D-85748 Garching bei M¨unchen, Germanye-mail: [email protected] Nordic Optical Telescope, Aarhus Universitet, Rambla Jos´e Ana Fern´andez P´erez 7, E-38711 Bre˜na Baja, Spain I. Physik Institut, University of Cologne, D-50937 Cologne, Germany Astronomisches Rechen-Institut, Zentrum f¨ur Astronomie der Universit¨at Heidelberg, M¨onchhofstr. 12-14, D-69120 Heidelberg,GermanyReceived; accepted
ABSTRACT
Context.
The Cygnus region, which dominates the local spiral arm of the Galaxy, is one of the nearest complexes of massive starformation, extending over several hundred parsecs. Its massive stellar content, regions of ongoing star formation, and molecular gashave been studied in detail at virtually all wavelengths. However, little is known of the history of the region beyond the past 10 Myr.
Aims.
We use the correlations between age, mass and luminosity of red supergiants to explore the history of star formation in Cygnusprevious to the formation of the present-day associations. The brightness and spectroscopic characteristics of red supergiants makeit easy to identify them and build up a virtually complete sample of such stars at the distance of the Cygnus region, thus providinga record of massive star formation extending several tens of Myr into the past, a period inaccessible through the O and early B starsobservable at present.
Methods.
We have made a selection based on the 2MASS colors of a sample of bright, red stars in an area of 84 square degreescovering the whole present extension of the Cygnus association in the Local Arm. We have obtained spectroscopy in the red visiblerange allowing an accurate, homogeneous spectral classification as well as a reliable separation between supergiants and other coolstars. Our data are complemented with Gaia Data Release 2 astrometric data.
Results.
We have identified 29 red supergiants in the area, 17 of which had not been previously classified as supergiants. Twenty-four of the 29 most likely belong to the Cygnus region and four of the remaining to the Perseus arm. We have used their derivedluminosities and masses to infer the star formation history of the region. Intense massive star formation activity is found to havestarted approximately 15 Myr ago, and we find evidence for two other episodes, one taking place between 20 and 30 Myr ago andanother one having ended approximately 40 Myr ago. There are small but significant di ff erences between the kinematic properties ofred supergiants younger or older then 20 Myr, hinting that stars of the older group were formed outside the precursor of the presentCygnus complex, possibly in the Sagittarius-Carina arm. Key words. stars: supergiants — stars: kinematics and dynamics — Galaxy: open clusters and associations — Galaxy: structure
1. Introduction
The Milky Way in the direction of Cygnus harbors one of thenearest giant molecular cloud complexes, containing OB associ-ations, massive star forming regions, recent supernova remnants,interstellar bubbles blown by the combined energetic action ofmassive stars, and an abundant sampler of virtually all the stagesof stellar evolution near the upper end of the initial mass func-tion (Odenwald 1989; Odenwald & Schwartz 1993; Bochkarev& Sitnik 1985; Reipurth & Schneider 2008). The Cygnus regionis the name generally used to refer to this physically coherentcomplex, a definition that excludes both foreground and back-ground structures physically unrelated to the complex, such asthe portions of the Cygnus Rift or the Perseus arm that are alsoprojected in the general direction of the constellation.The recent massive star forming activity of the regionhas been thoroughly studied through the O- and early B-typestars of its OB associations, particularly in the rich associationCygnus OB2 and its surroundings (e.g., Hanson 2003; Drewet al. 2008; Negueruela et al. 2008; Comer´on & Pasquali 2012; (cid:63)
Based on observations collected at the Centro Astron´omicoHispano Alem´an (CAHA) at Calar Alto, operated jointly by the Juntade Andaluc´ıa and the Instituto de Astrof´ısica de Andaluc´ıa (CSIC)
Wright et al. 2015; Berlanas et al. 2018, 2019) and the Cygnus Xcomplex of molecular and ionized gas (Wendker et al. 1991;Schneider et al. 2006, 2016). While those stars provide a goodpicture of the most recent star forming activity, the fast decreasein luminosity with decreasing mass, combined with the consider-able extinction toward many areas of the region at visible wave-lengths and the general crowdedness at its low galactic latitude,make the stellar census of the region severely incomplete formain sequence spectral types beyond O. Furthermore, the shortlifetimes of such massive stars limit their usefulness as probes ofthe star formation history of the host giant molecular complex tothe last ∼
10 Myr.Open clusters are accessible tracers of past star formationover much longer periods. However, many if not most of thestellar aggregates that appear as clusters at early ages becomeunbound after removal of their parental gas, their members dis-perse also on timescales of a few Myr, and the identification ofclusters in very crowded regions like Cygnus is di ffi cult. For thatreason, whereas the presence of a few clusters with ages of tensof Myr in Cygnus shows that star forming activity in the regionhas taken place over such time spans (Costado et al. 2017), thecomparison of their number with that of younger aggregates does a r X i v : . [ a s t r o - ph . GA ] S e p . Comer´on et al.: The historical record of massive star formation in Cygnus not provide a suitable representation of the star formation his-tory.Red supergiants of spectral types K and M provide an alter-native approach to probe the history of massive star formationover longer timescales. They are the descendants of stars withmasses above ∼ (cid:12) undergoing the helium-burning phase attheir cores, during which their radii grow to several AU, theirphotospheric temperatures drop to near or below 4,000 K, andtheir bolometric luminosities reach values comparable to thoseof the brightest O stars (Levesque 2010; Ekstr¨om et al. 2013).Their temperatures make their spectral energy distributions peakin the near infrared, where they are among the most luminousstars, making their detection easy from large distances and mit-igating the e ff ects of interstellar extinction when observing atthose wavelengths. The luminosity, as well as the beginning andthe duration of the red supergiant phase, are primarily a functionof the initial mass of the star, with rotation of the precursor play-ing a secondary role (Ekstr¨om et al. 2012). For a star of 7 M (cid:12) with an initial rotational velocity of 40% of the critical veloc-ity (at which the centrifugal force counterbalances the gravita-tional acceleration), the red supergiant phase starts at ∼
50 Myrand lasts almost 2 Myr, whereas a star with an initial mass of20 M (cid:12) rotating at the same fraction of the critical velocity be-comes a red supergiant at 9.7 Myr, remaining in that phase for0.3 Myr. Therefore, red supergiants are bright signposts of paststar formation potentially covering the entire lifetime of a giantmolecular complex, appearing in that phase for a limited but notnegligible time span, which makes them moderately abundant.However, the study of the past massive star formation record ofthe entire Cygnus region through its red supergiant content ishampered by the lack of a complete, homogeneous census ofsuch stars.In this paper we present the results of our study of the red su-pergiant content in the entire Cygnus region. We present a sam-ple of red supergiant candidates selected in an area of 84 squaredegrees on the basis of their near-infrared colors, which we es-timate to be complete down to initial masses M =
10 M (cid:12) , withstill a substantial completeness fraction down to M = (cid:12) .Visible spectroscopy in the red of all the candidates is presentedin order to discard evolved, less massive stars having the samephotometric and very similar spectroscopic characteristics. Theremaining set, refined with the use of Gaia DR2 data, constitutesan essentially complete sample of red supergiants in Cygnus.We use this sample to investigate the history of massive star for-mation in the region, and to obtain a crude estimate of its pastcontent.
2. Target selection
The Cygnus complex, which is one of the main structures ofthe Local Arm of our Galaxy (Xu et al. 2013), extends over avast region covering the interval 71 ◦ < l < ◦ , − ◦ < b < + ◦ , which encompasses the associations Cygnus OB1, OB2,OB3, OB8, and OB9. This excludes the doubtful, and possiblyforeground association Cygnus OB4 (de Zeeuw et al. 1999); thealso doubtful associations Cygnus OB5 and OB6 (Uyanıker et al.2001); and the foreground association Cygnus OB7 (de Zeeuwet al. 1999).Red supergiants at distances comparable to those of the OBassociations are expected to appear very bright at near-infraredwavelengths, and also have very red colors due both to theirlow photospheric temperatures and to the foreground extinction, mainly associated with the Cygnus Rift (Straiˇzys et al. 2015).We made an initial selection of all the stars in the 2MASS PointSource Catalog in the area defined above having K S < . J − K S > .
1. Using the T e ff versus intrinsic color in the Johnson-Cousins system from Kuˇcinskas et al. (2005), and transform-ing to the 2MASS filter system using the transformations fromCarpenter (2001), we obtain that the lower limit in J − K S colorselects the supergiants near 4,000 K (having an intrinsic color( J − K S ) = .
88) if reddened by A V > .
7, where we adoptthe extinction law of Cardelli et al. (1989) with total-to-selectiveextinction ratio R = .
1. Cooler stars meet the color selectioncriteria with even lower values of the extinction. The locationof the region beyond the Cygnus Rift implies a moderate levelof foreground extinction across the whole region of interest, asconfirmed by Comer´on & Pasquali (2012) and Berlanas et al.(2018) who find almost no stars with A V < A V >
3, and those of Cygnus OB3, which is theleast obscured of the associations, have A V > J − K S > . (cid:12) would have abolometric magnitude M bol = − . M K = − . K − band bolo-metric correction scale for red supergiants from Levesque et al.(2007). At the distance modulus DM = . A V < . A V = − A V (cid:39)
5, isfound by Schneider et al. (2007) in the neighborhood of the HIIregion S106 in Cygnus X, including parts of Cygnus OB1. Wetherefore expect most red supergiants with that initial mass to beincluded in our sample. Foreground extinction can reach higherlocal values in some specific areas in Cygnus, but those are gen-erally associated with sites of ongoing star formation such asthe lines of sight toward the highest column density areas of theCygnus X molecular clouds (Schneider et al. 2006), often asso-ciated with thermal HII regions (Downes & Rinehart 1966). Wedo not expect the presence of evolved objects such as red super-giants in such areas.Completeness increases rapidly with increasing mass, andfor a mass M =
10 M (cid:12) our limit K < . A V <
12, su ffi cient to cover the whole extinctiondepth in the region. Rotation has a small e ff ect on the luminos-ity at the He-burning stage for red supergiants in the 8 −
10 M (cid:12) range, di ff ering by ∆ log L (cid:39) .
03 in the nonrotating case withrespect to the rotating case adopted in the sense of nonrotatingsupergiants being fainter. This translates into A V = . M = (cid:12) would be included in our sample, and A V = . M =
10 M (cid:12) . The latter is still well abovethe bulk of extinction values found in Cygnus, therefore makingour conclusions on completeness only weakly dependent on theassumed initial rotation velocity.
2. Comer´on et al.: The historical record of massive star formation in Cygnus
The infrared color-magnitude selection criterion that we usemay be expected to include in the sample of candidate super-giants a high, possibly dominating fraction of evolved red gi-ant branch and asymptotic giant branch stars, descending fromlower-mass stars with large ages. Many of these have been spec-troscopically observed by previous works and have a spectralclassification placing them at spectral types M6 or later, at arange of cool temperatures unreachable even by the coolest redsupergiants, whereas others have been recognized as long-periodMira variables from their light curves. We have removed bothtypes of stars from our sample, as they are unrelated to the youngpopulation of Cygnus.
The distances to associations, clusters and other structures inCygnus have been traditionally di ffi cult to determine, due to acombination of reasons that include di ff erential extinction andpossible extinction anomalies (Terranegra et al. 1994), the dif-ficulties in estimating distances based on the observed spectraltypes and magnitudes of O stars (Mahy et al. 2015), the shallowdependency of radial velocities on distance in the directions ofCygnus foreground to the Perseus arm, or contamination of themembership of open clusters due to field crowdedness (Wang& Hu 2000; Costado et al. 2017). Furthermore, the distinctionamong the OB associations included in the area under consid-eration is mostly historical rather than physical. Some of them,like Cygnus OB1, OB8 and OB9 have been proposed to be partof a single structure (Mel’Nik & Efremov 1995), and it has beensuggested that the di ff erences between Cygnus OB2 and OB9are due to the progression of star formation from lower to highergalactic longitudes rather than to distinct, independent episodesof star formation (Comer´on et al. 2008; Comer´on & Pasquali2012). Despite large uncertainties and conflicting results, some-times yielding discrepant distances by factors up to 2 for thesame cluster (Straiˇzys et al. 2014), most studies yield values inthe 1.3-1.8 kpc range for the associations in the area under con-sideration (Garmany & Stencel 1992; Hanson 2003; Kharchenkoet al. 2005; Comer´on & Pasquali 2012; Straiˇzys et al. 2014,2019; Sitnik et al. 2015, e.g.,). Distances to structures in the in-terestellar medium have been established through evidence ofphysical links with OB associations, as illustrated by Schneideret al. (2007).Many of the di ffi culties encountered by more traditionalmethods have been overcome in recent years through the deter-mination of trigonometric parallaxes to masers using Very LargeBaseline Interferometry (VLBI). This has led to the determina-tion of precise distances to star forming regions in the Cygnus Xmolecular complex (Rygl et al. 2012), which is closely linkedto Cygnus OB2 (Schneider et al. 2006, 2016), as well as toother star forming regions outside it (Xu et al. 2013; Nagayamaet al. 2015). These latter determinations yield distances around1.4 kpc, broadly consistent with most determinations based onthe stellar component of clusters and associations.The trigonometric parallaxes measured by Gaia appear nev-ertheless to raise some tension with those previous determina-tions. In their analysis of the stellar population of Cygnus OB2Berlanas et al. (2019) report evidence for a bimodal distributionof stellar distances, with two groups separated by ∼
400 pc alongthe line of sight, but without any obvious angular separation onthe sky. A distance of ∼ ,
760 pc ( DM = .
15) is proposed forthe main group which, taken at face value, would call into ques-tion the physical link, supported by many observations, betweenmost of Cygnus OB2 and Cygnus X if the VLBI distances to its star forming regions are adopted for the latter. A second, lessnumerous foreground population appears at ∼ ff ect of the known zero point o ff set in Gaia DR2 paral-laxes begins to be noticeable. The distance estimates to clustersprovided by Cantat-Gaudin et al. (2018) are corrected for a sys-tematic o ff set of 0.029 mas following Lindegren et al. (2018),but Sch¨onrich et al. (2019) have argued that the average o ff setcan be as much as 0.054 mas. Furthermore, the systematic o ff setmay reach up to 0.1 mas locally (Lindegren et al. 2018; Luri et al.2018). If the Gaia DR2 parallaxes were systematically o ff set by0 . ff set) could be reconciled with the more nearby VLBI distancesand with other pre-Gaia determinations that favor the shorter dis-tance. Our sample of red supergiants is of little help to place thedistance to the Cygnus region on a firmer standing, as not onlyare their Gaia DR2 parallaxes a ff ected like other stars by thesame uncertainties and systematic e ff ects, but their very red col-ors may introduce additional systematics (Drimmel et al. 2019).Besides this, red supergiants present specific astrometric di ffi -culties due to their large diameters and large, evolving convec-tive cells (L´opez Ariste et al. 2018), which can displace theirphotocenters by angular distances greater than their parallaxes,therefore introducing a virtually random astrometric noise.Given the uncertainties outlined above, we have given pref-erence in this study to the distances derived through VLBI and tothose derived through other methods based on the stellar compo-nent. We therefore adopt a distance modulus DM = . ×
160 pc − . By adopting a single distance weneglect both depth e ff ects along the line of sight, which areexpected to be significant especially for the older stars in oursample (see Section 4.3), as well as possible distance gradi-ents across the plane of the sky. Regarding the latter we notethat there is no evidence for trends in the parallax as a func-tion of galactic longitude among the nine clusters in the catalogof Cantat-Gaudin et al. (2018) that we assign to the Cygnus re-gion, which suggests that its associated structures are spatiallyarranged in a direction that runs roughly perpendicular to ourline of sight.
3. Observations and spectral classification
The brightness of red supergiants in Cygnus and the range ofinterstellar extinction in the area make them generally accessi-ble to spectroscopy at red and far-red visible wavelengths usingshort exposure times with 2m-class telescopes. This spectral re-gion o ff ers abundant atomic and molecular features sensitive totemperature and surface gravity, enabling the establishment ofprecise classification criteria in both spectral subtype and lumi-nosity class (Torres-Dodgen & Weaver 1993).We observed all the stars fulfilling the location, magnitude,and color criteria described in Section 2, with the exceptions of
3. Comer´on et al.: The historical record of massive star formation in Cygnus those previously classified as M6 or later or as Mira variablesin the literature as noted there. This amounts to a sample of 78stars, which includes all the stars for which no spectral classi-fication has been published thus far, as well as those that havebeen previously classified in other works as M5 or earlier regard-less of the published luminosity class. Indeed, a literature searchshows that published classifications are based on di ff ering setsof criteria, having been obtained with a variety of instruments,spectral resolutions, wavelength coverage, and signal-to-noiseratio. We found a number of cases in which significantly di ff er-ent spectral types are assigned to the same star in di ff erent works.Furthermore, given the subtle spectroscopic criteria that distin-guish young red supergiants from older red giant branch stars,we preferred to observe all the stars to ensure that our sampleof red supergiants is based on the uniform application of classi-fication criteria to spectra obtained with the same instrumentalsetup.Our observations took place on two nights from August 2-4, 2016, using CAFOS, the facility imager and low-resolutionspectrograph at the 2.2m telescope on Calar Alto Observatoryin Southern Spain. A grism was used covering the wavelengthrange 6000 < λ (Å) < λ/ ∆ λ (cid:39) (cid:48)(cid:48) R magnitude estimatedfrom the cataloged JHK S magnitudes. We also observed withthe same setup a grid of MK standard stars in the K2-M4 spec-tral range, with luminosity classes Ia to III, from the catalog ofGarcia (1989). The spectrophotometric standard BD + ◦ ff ects ofthe di ff erent amounts of foreground extinction toward each ob-ject. Spectral classification was carried out using the normalizedspectra.Spectral subtypes were determined by carefully comparingthe strength of the TiO and VO features to those of the grid ofstandards. With only one exception discussed below, all our tar-gets have spectral types in the K-M range. At the resolution ofour spectra in the wavelength range covered di ff erences couldbe discerned between types K2, K5, K7 and M0, but no finerdistinction could be made among the spectral subtypes in theK2-M0 range. The greater sensitivity of spectral type with tem-perature for spectral types later than M0 allowed us to performa more detailed subtype classification in the M range, with anestimated accuracy of ± . ff erences inspectral features displaying a dependency on surface gravity.Useful features for spectral classification in the red / far-red rangeacross the entire spectral sequence are presented in detail in theatlas of Torres-Dodgen & Weaver (1993), which forms the basisof our luminosity classification. The most prominent luminosity-sensitive features in the K-M range are the cluster of CN bandsnear 7970 Å and the CaII infrared triplet at 8498, 8542, and8662 Å, all of which have a positive luminosity e ff ect withtheir equivalent widths increasing with luminosity. Other fea-tures that we used are the blend of metallic lines at 6497 Å, theatomic lines of TiI in the 7345-7364 Å interval, and the KI lineat 7699 Å , also having a positive luminosity class e ff ect. Tothese we added other luminosity-sensitive features discussed by F e I − T i I C a II F e I C a II C a II F e I Fig. 1.
Luminosity e ff ects at spectral type M2 in the regionaround the calcium infrared triplet. The solid line corresponds tothe M2Ib standard HD 10465, and the dashed line to the M2IIIstandard HD 219734 (Garcia 1989). Both spectra have been ra-tioed to their best-fitting 7th-degree polynomial (fit to the whole6000 < λ (Å) < ff ect (seeFig. 1). We measured equivalent widths of each of these features,both for the MK standards and for our target stars. For the stan-dards, the equivalent width of any given feature as a functionof spectral type traces a well-defined curve generally increasingfrom early to late K-types, and then reaching a plateau at lateK- or early M-types, as shown in Fig. 2. The behavior is lesssystematic among the supergiants of luminosity classes Ia to II,for which the equivalent widths do not follow such a strict rela-tionship. However, they always fall above those of giants of thesame spectral type. Therefore, even though our spectra do not al-low us to discern among the supergiants of luminosity classes Ia,Iab, Ib or II, the sequence of equivalent widths as a function ofspectral type for luminosity class III giants forms a clear lowerboundary, with supergiants generally having equivalent widthsdistinctly above that boundary.We used the luminosity-sensitive features noted above to in-dependently classify each star as giant or supergiant on the ba-sis of each of them, and then we compared the results obtainedacross those various luminosity indicators. Most of the stars inthe sample are consistently classified as giants or supergiants ac-cording to each of the luminosity-sensitive features used for clas-sification. For some stars we found one or two of them for whichthe classification di ff ers from the one obtained from the others,or for which the measured equivalent width is not su ffi cientlyseparated from the giant sequence locus to clearly classify themas giants or supergiants. We assigned those stars to the luminos-ity class obtained from the majority of the other lines. In a fewcases in which the number of features favoring either classifica-tion was similar, we adopted the luminosity class derived fromthe CN and CaII features. Table 3 lists our spectral classificationsfor the 29 supergiants identified in our sample.Figure 3 shows the distinct kinematical properties of thered supergiants, which display a low velocity dispersion, ascompared to the proper motion distribution of all the other
4. Comer´on et al.: The historical record of massive star formation in Cygnus
K2 K5 K7 M4M2M0 E qu i v a l e n t w i d t h ( A ) Spectral type o Fig. 2.
Added equivalent widths of all the features used as lumi-nosity class discriminants as indicated in Sect. 3 as a function ofspectral type. Open circles indicate the stars from Garcia (1989)used as standards of luminosity classes I and II, and asteriskscorrespond to standards of luminosity class III. The location ofall the stars in Cygnus observed by us are represented by dots.
Fig. 3.
Gaia DR2 proper motions of the targets selected accord-ing to the criteria described in Section 2. The 29 stars spectro-scopically classified as supergiants are represented with filledcircles, and the rest of the sample is represented by the open cir-cles.stars in our sample, which is characteristic of an evolved, kine-matically hot population. The renormalized unit weight error(RUWE) quantifying the goodness of the astrometric solution(Lindegren et al. 2018) is well below the threshold of 1.4 forall the red supergiants, implying that their astrometric solutionsprovide good fits to the data, and only one, the M4 supergiantJ202308.60 + + Fig. 4.
Distribution of the parallax errors listed in the Gaia DR2catalog for the targets spectroscopically classified as supergiantsin this work (upper panel) and those classified as giants coveringthe same range of spectral types. The e ff ect of the intrinsic pho-tospheric properties of supergiants on the accuracy with whichparallaxes can be determined is apparent.Most of the stars in our sample are identified as supergiantsfor the first time. Twelve of the 29 supergiants had no previousspectral classification, and five had spectral types published inthe literature but no luminosity class determined. Eleven havebeen classified as supergiants in previous works (see Notes toTable 3), with spectral types very close to the ones that we de-termine. As noted in Table 3 one of the stars that we classify asM3.5 supergiant, J204924.95 + + + K S magnitudes are among the faintest in the sample, all ofwhich is more consistent with membership in the backgroundPerseus arm. This is also the case for J201126.21 + + + + +
5. Comer´on et al.: The historical record of massive star formation in Cygnus
Table 1.
Late-type giants
Star Type Star TypeJ195955.80 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + J202736.78 + ff erence in spatial veloc-ity between J201919.97 + − , which is on thelow end of typical runaway velocities, and the di ff erence iseven lower for J202736.78 + + + + + = J202114.07 + + + K S = .
48 in 2MASS). From the candidate su-pergiants near Cygnus OB2 in Comer´on et al. (2016) weconfirm J203323.91 + + K − band spectroscopy. We do notconfirm as supergiants RAFGL 2600 (J203128.70 + + = J202643.03 + + ff erent from K or M is J205443.1 + α emission (Kohoutek & Wehmeyer 1999).Its spectrum is dominated by bands of ZrO, clearly identifyingJ205443.1 + ω = . ± . +
4. Results
We adopt the spectral type versus e ff ective temperature ( T e ff )calibration for galactic supergiants from Levesque et al. (2007)to obtain T e ff for the stars in our sample. We estimate then the K S -band extinction using the T e ff versus intrinsic ( J − K ) colorrelationship for supergiants from Kuˇcinskas et al. (2005), and thecolor transformation from the Johnson-Glass photometric sys-tem used by Kuˇcinskas et al. (2005) to the 2MASS photomet-ric system using the transformation equations from Carpenter(2001):( J − K S ) = . J − K ) JC − . . (1)The K S -band foreground extinction A K is obtained using theextinction law of Cardelli et al. (1989) with a total-to-selectiveextinction ratio R V = . A K = . J − K S ) − ( J − K S ) ] . (2)Finally the luminosity L is obtained using the T e ff vs. bolo-metric correction BC K for galactic supergiants from Levesqueet al. (2005) and the adopted distance modulus DM = . L / L (cid:12) ) = − . K S − DM − A K + BC K − . . (3)Once T e ff and L are derived as described, we estimate theinitial mass and the age of each star using the Geneva evolu-tionary models (Ekstr¨om et al. 2012). We base our estimates onthe rotating models with solar metallicity, and an initial rotationvelocity that is 40% of the critical value, which we take as be-ing representative of the ensemble. In the mass interval relevantfor our sample, rotation at that initial velocity increases log L by ∼ . ∼
20% with respect tothe nonrotating case. We adopt the duration of the red supergiantphase as the duration of the helium-burning stage or, for starswith mass below ∼
12 M (cid:12) undergoing a blue loop during thatstage, the time span between the beginning of the helium burn-ing and the beginning of the blue loop, since the post-blue loophelium-burning stage is much shorter than the pre-blue loop one.The duration of the red supergiant phase defined in this way inthe rotating case increases from 1.8 Myr for an initial mass of
6. Comer´on et al.: The historical record of massive star formation in Cygnus
Table 2.
Derived intrinsic properties of supergiants in the Cygnus associations
Star type T e ff ( J − K S ) BC K A K log LL (cid:12) M t RSG (K) (M (cid:12) ) (Myr)J201842.17 + + + + + + + + + + + + + + + + + + + + + + + +
20 101215
Fig. 5.
Derived temperatures and luminosities of the stars spec-troscopically classified as supergiants and assigned to theCygnus region, assuming a common distance module DM = .
9, plotted together with the evolutionary tracks of Ekstr¨omet al. (2012) for stars with various initial masses. The solid linesdescribe the evolution of nonrotating stars, and the dot-dashedlines correspond to stars with an initial rotation of 40% of thecritical value. Adopting a distance of 1.7 kpc, seemingly favoredby Gaia, would shift the location of the stars upward by 0.1 dexin luminosity.7 M (cid:12) to 2.2 Myr for an initial mass of 10 M (cid:12) , then decreasing to0.3 Myr at 20 M (cid:12) .Table 2 shows the intrinsic properties of the 24 red super-giants that we assign to the Cygnus region under those assump- tions. The mass is derived from the average value of the lumi-nosity during the helium burning phase in the rotating mod-els, and t RSG is the age at which the star enters the red super-giant phase. Figure 5 shows their location in the temperature-luminosity diagram, with the evolutionary tracks for rotatingmodels of various initial masses. Most of the stars are locatedwithin the steep, narrow band defining the helium-burning se-quence, with the latest spectral types generally near the up-per part of that sequence. There are a few exceptions neverthe-less, most notably J205425.72 + + ff erence inderived luminosity with respect to other members of the Cygnusregion having similar spectral types reaches nearly one orderof magnitude, which would place them in the Perseus arm ata distance of ∼ . σ for both stars. Furthermore, their radial velocities dif-fer by ∼
40 km s − from the typical radial velocities foundin the Perseus arm members (like J204636.65 + + T e ff versus log L diagram.
7. Comer´on et al.: The historical record of massive star formation in Cygnus
The age range covered by red supergiants makes them poor trac-ers of the location of their parent associations. Taking the ve-locity dispersion inside a typical OB association as 4.5 km s − (Melnik & Dambis 2020), red supergiants can be expected tohave moved typically by ∼
100 pc from their birthplaces in20 Myr, a distance exceeding by a factor of a few the typical sizeof an OB association. Therefore, we should not expect to findour red supergiants confined to the boundaries of the identifiedOB associations.Figure 6 shows the position of the red supergiants in theregion under discussion, superimposed on a map obtained byMSX, the Midcourse Space Experiment satellite (Price 1995) inthe infrared band A (6 . − . µ m) that highlights the inter-stellar medium as traced by warm dust. Red supergiants show avery mild concentration toward the central regions of the map,broadly overlapping the area covered by the associations CygnusOB1, OB2, OB8, and OB9 although some stars appear outsidethose boundaries. The three stars with the highest galactic lon-gitude (near the left border of Figure 6) are within the bound-aries of Cygnus OB7, which extends further toward higher galac-tic longitude and is at roughly half of the distance to the otherOB associations. However, there is no clear evidence support-ing membership of these stars to Cygnus OB7, as both theirGaia DR2 parallaxes, proper motions and radial velocities arewithin the range of the other stars that we consider members ofthe Cygnus region.Figure 6 also shows the relative positions of red supergiantswith estimated ages above and below 20 Myr. No obvious trendis found, ruling out the existence of a large-scale age gradienttraced by them. Indications of a progression in star formationwith time from lower to higher latitudes has been reported inthe surroundings of Cygnus OB2 (Drew et al. 2008; Comer´onet al. 2008; Comer´on & Pasquali 2012) based on younger stars.The lack of evidence for a large-scale gradient among the redsupergiant population suggests that the progression reportedis probably a localized and more recent feature, where mas-sive star formation initiated in Cygnus OB9 propagated towardCygnus OB2, and currently continues at even higher longitudein Cygnus X North, where the DR21 complex is an active siteof ongoing massive star formation (Motte et al. 2007; Comer´onet al. 2008; Csengeri et al. 2011).There is an absence of red supergiants at the lowest galacticlongitudes, with no such stars appearing within or anywhere nearthe boundaries of the relatively rich association Cygnus OB3(Rao et al. 2020), with the only exception of the M4 supergiant201126.21 + ω = . ± . . ± . / OB2 / OB9complex (Straiˇzys et al. 2019; Rao et al. 2020). However, ourcriteria used to select red supergiant candidates should have se-lected them as well, as the modest increase of 0.6 mag in dis-tance modulus should be partly o ff set by a lighter extinction(Straiˇzys et al. 2019) in its direction. The lack of young red su-pergiants might be due to massive star formation in the regionhaving started only recently, implying that only the most massivestars, with very short lifetimes as red supergiants, have enteredthat phase. The catalog of Kharchenko et al. (2016) lists sevenpossible open clusters in the area, five of them with estimatedages above 10 Myr, and their distances are consistent with those of the Cygnus region (Cantat-Gaudin et al. 2018). However, themembership to most of them, and even their physical existenceas clusters, is highly uncertain due to the crowdedness of thearea. The most outstanding cluster in Cygnus OB3, NGC 6871,is listed by Kharchenko et al. (2016) as having an age just shortof 10 Myr, still consistent with the youth of the area. At the adopted distance of 1.51 kpc, the typical proper motionuncertainty of 0 . − in the Gaia DR2 catalog translatesinto a tangential velocity uncertainty of only 1 . − , wellbelow the internal velocity dispersion of individual associations(Melnik & Dambis 2020). In turn, the standard deviation of theproper motions of the stars that we allocate to the Cygnus regiontranslates into a standard deviation of the tangential velocity of9.4 km s − , which is of the same order as the standard devia-tion of the radial velocities, 7.1 km s − . The internal velocitystructure of OB associations and the relative velocities amonggroups is therefore well resolved by Gaia, and the fact that thedispersion in space velocity among the stars in our sample is sig-nificantly larger than the internal velocity dispersion in associa-tions suggests that they had their origin in kinematically distinctgroups across the region.The kinematic accuracy, combined with a galactic potentialmodel, allows us to investigate the regions of origin of our sam-ple of red supergiants. Age is nevertheless the dominant sourceof uncertainty, followed by the distance. The spread of distancesalong the line of sight is likely to be relevant as well since, evenif the stars might have been born within a narrow range of dis-tances to the Sun, the velocity dispersion along the line of sightintroduces a scatter reaching ±
300 pc in distance for the agesof the oldest stars in our sample, of the same order as the un-certainty in the distance derived from the trigonometric parallax.Taken together these factors severely limit our ability to deter-mine the location of origin of the red supergiants, especially ofthe oldest among them. However, some intriguing trends do ap-pear.We traced back in time the trajectories of the 23 red super-giants of the Cygnus region for which full kinematical informa-tion is available using an axisymmetric galactic potential thatyields a flat rotation curve within the range of galacticentric dis-tances covered by the orbits of the stars in our sample. The radialforce f R toward the galactic center per unit mass is then f R = . × − km s − yr − V [km s − ] R − [kpc − ] , where V is the circular velocity of the local standard of restvelocity around the galactic center and R is the distance to thegalactic center. We adopt the parameters obtained by McMillan(2017): V = . − , distance of the Sun to the galac-tic center R = .
20 kpc, and components of the velocity of theSun with respect to the local standard of rest ( U (cid:12) , V (cid:12) , W (cid:12) ) = (11 . , . , .
25) km s − , where the three components of thevelocity are directed toward the galactic center, the direction ofgalactic rotation, and the north galactic pole, respectively. We as-sume the force per unit mass perpendicular to the galactic plane f z to be proportional to the distance z to it, and use the parame-ters that reproduce the force at 1.1 kpc from Kuijken & Gilmore(1991), as also adopted by McMillan (2017), obtaining f z = . × − km s − yr − z [kpc] . We integrated the orbits of the 23 stars backward in time, alsocomputing the evolution of their galactic coordinates as would
8. Comer´on et al.: The historical record of massive star formation in Cygnus galactic longitude g a l ac ti c l a tit ud e Fig. 6.
Spatial distribution of the red supergiants belonging to the Cygnus region, overplotted on a mosaic of infrared imagesobtained by the Midcourse Space Experiment (MSX) satellite in the infrared band A (6 . − . µ m) that outlines the distribution ofwarm dust. Blue triangles indicate the positions of stars younger than 20 Myr, and red circles those of older stars. The approximateboundaries of present-day associations as listed by Uyanıker et al. (2001) are indicated. The pervading nebulosity between l (cid:39) ◦ and l (cid:39) ◦ outlines the Cygnus X molecular complex.be measured by an observer moving with the circular velocityalong the solar circle at a position coinciding with that of the Sunat the present time. Figure 7 shows the present position of eachstar projected on the sky as compared with its position 15 Myrago, which roughly gives the locations of the young stars (de-fined as those having ages estimated at less than 20 Myr fromthe evolutionary models with rotation) near the time when theyformed.As it can be seen in Figure 7, although old and young starsoccupy similar distribution on the sky at present, both groups arekinematically di ff erent on the average. Most of the stars in theyoung group would have appeared at somewhat higher galacticlongitude for an observer 15 Myr ago, with only 3 out of the 9stars in that group having been located at somewhat lower galac-tic longitude, di ff ering by 2 ◦ ff ering by more than 2 ◦ from their current position. That is, the young and old groups,which at present overlap in the plane of the sky, would havebeen spatially separated 15 Myr ago, and the average galacticlongitude of the young group would have been larger than thatof most members of the old group. The distribution of initial masses of present-day red supergiantsrepresents an historical record of the star formation history ofthe region, which we can reconstruct thanks to their relationshipwith their ages. We use the cumulative mass distribution of the24 red supergiants that we identify in the Cygnus region, fol-lowing a treatment similar to that presented by Comer´on et al.(2016) and Comer´on & Torra (2018). The number N ( M ) of red Fig. 7.
Positions at present (circles) and 15 Myr ago (end of thelines having their origin at the present position of each star) forthe 23 red supergiants in Cygnus having measured proper mo-tions and radial velocity. The sample is divided as in Figure 6,with the younger and older samples separated in two panels forclarity. The galactic longitude and latitude are defined as theywould be measured by an observer moving with the circular ve-locity along the solar circle and located at the present time at theposition of the Sun.supergiants observed at the present time t = M can be written as N ( M ) ∝ (cid:90) t = −∞ (cid:90) ∞ M (cid:48) = M S FR ( t ) V ( M (cid:48) , − t ) Ψ ( M (cid:48) )d M (cid:48) d t , (4)
9. Comer´on et al.: The historical record of massive star formation in Cygnus where
S FR ( t ) is the star formation rate as a function of time, Ψ ( M (cid:48) ) is the initial mass function that we assume to remain con-stant with time, and V ( M (cid:48) , − t ) is a hat function that has unit valueif a star of mass M and age − t is undergoing the red supergiantphase at present, and zero otherwise. We note that we define t as negative toward the past, hence the negative sign in front of t to denote the present age of the star. Also, it may be noted that V ( M , t ) should also include a dependency on the initial rotationvelocity of the massive precursor, and the expression for N ( M )should therefore include an additional integral over the initialrotation velocity and its distribution function. The latter has asgeneral features a peak at small velocities, followed by a de-crease and then a high rotation velocity tail (Huang et al. 2010).Since the initial rotation velocities of our stars are not known,we ignore such details in our treatment, as we did when com-puting the estimated physical properties of the stars in Table 2,assuming instead a typical rotation velocity for all of them. Fig. 8.
Reconstruction of the star formation rate normalized toits value 10 Myr ago, fit to reproduce the currently derived dis-tribution of masses of red supergiants in Cygnus, assuming evo-lutionary models of Ekstr¨om et al. (2012) with initial rotation40% of the critical value. Error bars reflect uncertainties due tothe actual initial rotation velocities and to the limited number ofstars in each mass bin.Using a power law with Salpeter slope − .
35 in linear massunits for the initial mass function and V from the evolutionarymodels of Ekstr¨om et al. (2012), we can numerically invert theintegral. In this way we obtain the best fitting shape of the func-tion S FR ( t ) describing the overall star formation history tracedby our sample.The star formation history S FR ( t ) that we obtain in this wayis shown in Figure 8. A piecewise S FR ( t ) function is adjusted byfitting its value at 1 Myr intervals so as to obtain the best possi-ble match to the cumulative histogram of masses obtained fromTable 2 through Eq. 4. To estimate the uncertainty on S FR ( t ) wehave repeated the fit multiple times by adding Poissonian noiseto the various mass bins to account for the uncertainties intro-duced by small number statistics, as well as by increasing or re-ducing at random the time of entrance of each star in the heliumburning phase by up to 20% to mimic the e ff ects of the unknowninitial rotation velocity of each individual star. The cumulativehistogram is presented in Figure 9. In this way, a star formationhistory in the region emerges in which the star formation ratestarted rising steeply only about 15 Myr ago. This is consistentwith the estimated ages of the most prominent clusters of the re-gion such as NGC 6871, NGC 6910, and NGC 6913, as well as Fig. 9.
Cumulative histogram of masses of the 24 red supergiantsin Cygnus identified in this work. The solid line shows the cor-responding best-fit cumulative histogram derived from the starformation history depicted in Figure 8. For reference, the pre-dicted cumulative histograms obtained assuming a constant starformation rate (dotted line) and a star formation rate increasinglinearly with time (dashed line) are shown as well.with the beginning of the star formation in the present-day as-sociations Cygnus OB1, OB8, and OB9. Such a recent increasein the star formation rate is required to account for the abundantpresence of high-mass red supergiants and the rather shallow tailof the cumulative histogram of masses in the high-mass end. Thestar formation rate dropped to a very low level between approx-imately 15 Myr and 20 Myr ago, with no significant levels ofhigh-mass star formation traced by the present-day red super-giants content.Some sustained period of high-mass star formation tookplace over a span of ∼
10 Myr, starting about 30 Myr ago. O-type stars formed during this episode have completed their life-cycles and are not visible anymore, and the red supergiants thattrace it come from B-type precursors currently observed as lateK-type supergiants. Some open clusters with ages estimated tolie in this range are listed by Kharchenko et al. (2016), and areprobably part of this episode. Despite being an active period ofstar formation, we estimate its intensity to have reached no morethan 20 −
40% of the level of the peak period 10 Myr ago. Thisperiod was preceded by another period of inactivity, roughly 10Myr long.The lowest-mass red supergiants visible at present are thedescendants of B1-B2 stars formed roughly 40 Myr ago. Thedispersion in space of the stars formed at this time, the likelyincompleteness of our sample at the lowest luminosities and thelarger uncertainties in ages make it di ffi cult to give any details onthe star formation activity in that period, but the increase in thenumber of stars with initial masses near or below 10 M (cid:12) requiresit to have been rather intense, perhaps at the level of ∼
50% ofthe peak level of 10 Myr ago, and more extended in time.For the sake of comparison, Figure 9 also shows the his-togram of masses with the histogram predicted by two simplestar formation histories, in both cases normalized to the totalnumber of stars. A constant star formation rate (dotted line inFigure 9) consistently underpredicts the number of stars at allmasses above the lowest-mass bin, or alternatively greatly over-predicts the number of low-mass red supergiants if the normal-ization condition is lifted. On the other hand, a star formationrate having started from zero 40 Myr ago and monotonically ris-ing with a constant slope until 9 Myr ago, which is the estimated
10. Comer´on et al.: The historical record of massive star formation in Cygnus age of the youngest stars in our sample, yields a somewhat bettermatch at higher masses (dashed line) but predicts a much highernumber of the stars in the lowest mass bins, even at those wherecompleteness of our magnitude-limited sample is not a concern.Clearly both simplified shapes of the star formation rate fail toreproduce the distribution of properties of the red supergiants inCygnus.
The fossil record of massive star formation in Cygnus nowadaysrepresented by red supergiants shows that the most intense ac-tivity started in the recent past (10-15 Myr ago), although olderdescendants from massive stars exist in the area at present. Whenthe kinematics of the stars is considered jointly with the starformation history of the region we find that the latter cannotbe interpreted simply in terms of a sequence of bursts in a gi-ant molecular cloud complex. One of our most remarkable find-ings is the separation between most of the stars with ages below20 Myr and most of the stars with ages above that value whentheir spatial locations are traced back 15 Myr or more in the past.This suggests that, although the giant molecular complex out ofwhich the newest generations of stars have formed also harborsolder stars, most of those stars formed elsewhere.In a qualitative way, our results suggest that the giant molec-ular complex precursor of the present OB associations and activestar forming sites was permeated over 10 Myr by a slow streamof massive stars coming from the general direction currently de-fined by its lower galactic longitude edge, and having formedover an extended period ranging from approximately 20 to morethan 40 Myr. Those stars are now undergoing the relatively short-lived (less than 2 Myr) red supergiant phase, but they were earlyB-type stars when they encountered the Cygnus complex. At thatpoint they were most likely accompanied by somewhat higher-mass stars that have already ended their lifecycles as supernovaein the meantime, whereas even higher-mass stars having formedtogether with them went through their entire evolution beforeencountering the Cygnus complex.Since our results suggest that most of the stars currently ob-served as old red supergiants did not form within the presentCygnus molecular complex, it is interesting to speculate abouttheir possible place of origin, perhaps linking it to the large-scalestructure of our Galaxy and the history of the local environmentin the last ∼
50 Myr. Thanks to the recent improvements in thedetermination of distances to tracers of spiral structure, espe-cially through VLBI parallaxes to massive star forming regions(Xu et al. 2016; Reid et al. 2019) and to young stars and ag-gregates with Gaia (Chen et al. 2019; Cantat-Gaudin & Anders2020; Khoperskov et al. 2020), the existence of four major armsthat can be followed from the inner regions of the Galaxy all theway to beyond the solar circle appears to be well established (seehowever Grosbøl & Carraro (2018) for a di ff erent view). In thispicture the solar neighborhood is roughly halfway between twomajor arms, and the Cygnus region is the most prominent struc-ture of the so-called Local Arm (Xu et al. 2013), which mayinclude the Orion complex as its nearest point to the Sun. Theconnection of the Local Arm with the grand design spiral struc-ture and the underlying density wave is not clearly establishedthough.The objects that delineate the spiral structure are generallyrelated to ongoing or very recent massive star formation, andtherefore provide a snapshot of the present appearance of ourGalaxy. Its evolution can nevertheless be investigated throughthe use of other tracers that can be followed further back in time, leaving the trail of the passage of the spiral structure throughthe galactic disk. A review of such studies has been presentedby Vall´ee (2020), who discusses the fundamental parameters ofspiral structure of 24 nearby galaxies, including ours. In particu-lar, the results of Vall´ee (2018, 2020) derive an angular velocityof the spiral pattern Ω sp =
19 km s − kpc − for the Milky Way,slower than the circular angular velocity in the solar circle.With the distance of the Sun to the galactic center of 8 .
20 kpcadopted in Section 4.3, an average galactic longitude l = ◦ andan average distance of 1 .
51 kpc for the bulk of the population ofold red supergiants in Cygnus, their galactocentric distance is R = . V = . − , theiraverage angular rotation velocity is Ω = . − kpc − , im-plying a drift with respect to the overall spiral structure with theangular velocity Ω − Ω sp = . − kpc − . When comparingto the present location of the major spiral arms, this indicates thatthe old red supergiants are moving relative to them coming fromthe direction of the Sagittarius-Carina arm and going toward thePerseus arm.At present, the Sagittarius-Carina arm intersects the galac-tocentric circle of radius 8 . . l (cid:39) ◦ (Mel’Niket al. 1998; Reid et al. 2019). From this we obtain an angu-lar distance λ from the Sagittarius-Carina arm in the proxim-ities of Carina OB1 to the Cygnus region measured from thegalactic center λ (cid:39) ◦
5, which translates into an arc of length D = (2 π R λ ) / ◦ (cid:39) . . − kpc − with re-spect to the spiral structure, the old red supergiants would havecovered the length of that arc in a time t = D / [( Ω − Ω sp ) R ] (cid:39)
43 Myr, which is similar to the age of the oldest red supergiantsof our sample.We must recall that the ages of the stars in our sample aresubjected to important uncertainties, and that the reconstruc-tion of their trajectories also becomes more uncertain as wetrace them further into the past. Moreover, our estimates neglectstreaming motions induced by the potential well associated withthe spiral density wave and velocity jumps caused by large-scalespiral shocks that should reflect as deviations from the circu-lar velocity (e.g., Ram´on-Fox & Bonnell 2018; Comeron et al.1997). However, those estimates make the origin of at least theoldest of our red supergiants in the Sagittarius-Carina arm appearas a plausible hypothesis. In this scenario, the present massivestar formation in Cygnus may be regarded as an indirect byprod-uct of star formation in the Sagittarius-Carina arm, when a groupof massive stars formed there and, having drifted away from itas the density wave lagged behind the local galactic rotation, en-countered the precursor of the present Cygnus giant molecularcloud complex. Although this is admittedly highly speculative,the apparent misfit between the Local Arm and the four mainarms of our Galaxy might be explained if the Local Arm con-sisted of molecular material in the interarm region that survivedas such the passage through the spiral density wave, and onlybecame unstable against star formation when massive stars thathad formed in a main arm entered it.
5. Conclusions
Our search for red supergiants has more than duplicated thenumber of such objects known in the direction of Cygnus, pro-ducing a homogeneously classified sample ranging from K2 toM4 spectral types. Of the 29 stars firmly established as super-
11. Comer´on et al.: The historical record of massive star formation in Cygnus giants in that interval, we assign 24 of them to the Cygnus regionassociated with the Local Arm. Four of the remaining five super-giants are most likely associated with the background Perseusarm, while the fifth is possibly a foreground star.The presence of 24 red supergiants in the region, covering awide range of masses, luminosities and ages, demonstrates thatmassive star formation in Cygnus started long before the OBassociations that dominate the stellar component at present wereborn. We find no evidence of a spatial segregation between theyounger and the older supergiants of our sample and, as expectedfrom the internal velocity dispersion in OB associations, there isno obvious clustering among them and no particular proximityto the known OB associations.Using the results of models of massive star evolution, we es-timate the initial mass of each red supergiant based on its lumi-nosity at present, and also the age based on the time at which redsupergiants of various masses enter the helium-burning phase.These relationships contain uncertainties intrinsic to the models,and also due to their dependency on the initial rotation velocity,which is not known for our stars. With these caveats in mind, wehave derived the star formation history of our sample. We findthat the intense episode of star formation that has continued withthe formation of the present-day OB associations started approx-imately 15 Myr ago, but we also find evidence for previous starformation episodes of less intensity having taken place between20 and 30 Myr ago, and more than about 40 Myr ago. Theseepisodes account for the existence of the less massive, cold andluminous red supergiants in the region. We find slightly di ff er-ent average kinematical properties between the younger and theolder red supergiants: For an observer moving along the solarcircle with the local standard of rest, the populations of young( <
20 Myr) and old ( >
20 Myr) red supergiants would haveappeared more separated in the past, with the older supergiantsgroup moving on the average from lower to higher galactic lon-gitudes, and the younger group moving in the opposite direction.We take this result as an indication that the red supergiants maynot be simply tracing the star formation history of the Cygnusgiant molecular complex. Instead, we favor an interpretation inwhich most of the older supergiants formed elsewhere outsidethe complex, entering it later from lower galactic longitudes.We propose that the oldest red supergiants currently observed inCygnus actually started their lives as moderately massive starsin the Sagittarius-Carina main spiral arm of our Galaxy, drift-ing away from it as the density wave responsible for the spiralpattern made its progress across the stellar and gaseous disk. Wepresent some crude estimates based on the parameters of the spi-ral structure of the Milky Way that support the plausibility of thisscenario.The exploitation of the Gaia astrometric legacy has juststarted, and the release of the final catalog of the mission inthe coming years will make it possible to extend studies like theone presented here and investigate the properties of less massivemembers of Cygnus still on the main sequence, once nearly com-plete and nearly uncontaminated samples can be built. It may beexpected that a much more detailed picture will emerge fromsuch future studies, perhaps confirming our findings that hint toa longer and more complicated history of this complex than onemay have suspected.
Acknowledgements.
We are very pleased to thank the excellent support pro-vided by the sta ff at the Calar Alto Observatory, especially on this occasion byGilles Bergond, Ana Guijarro, and David Galad´ı. The constructive review of thepaper by the referee, Richard Boyle, is gratefully acknowledged. NS acknowl-edges support by the French ANR and the German DFG through the project”GENESIS” (ANR-16-CE92-0035-01 / DFG1591 / / California Institute of Technology,funded by the National Aeronautics and Space Administration and the NationalScience Foundation. This work has made use of data from the EuropeanSpace Agency (ESA) mission
Gaia ( ),processed by the Gaia
Data Processing and Analysis Consortium (DPAC, ). Fundingfor the DPAC has been provided by national institutions, in particular theinstitutions participating in the
Gaia
Multilateral Agreement. This researchhas made use of the SIMBAD database, operated at CDS, Strasbourg, France,and has made use of data products from the Midcourse Space Experiment.Processing of the data was funded by the Ballistic Missile Defense Organizationwith additional support from NASA O ffi ce of Space Science. References
Berlanas, S. R., Herrero, A., Comer´on, F., et al. 2018, A&A, 612, A50Berlanas, S. R., Wright, N. J., Herrero, A., Drew, J. E., & Lennon, D. J. 2019,MNRAS, 484, 1838Bidelman, W. P. 1957, PASP, 69, 326Bochkarev, N. G. & Sitnik, T. G. 1985, Ap&SS, 108, 237Cameron, D. & Nassau, J. J. 1956, ApJ, 124, 346Cantat-Gaudin, T. & Anders, F. 2020, A&A, 633, A99Cantat-Gaudin, T., Jordi, C., Vallenari, A., et al. 2018, A&A, 618, A93Cardelli, J. A., Clayton, G. C., & Mathis, J. S. 1989, ApJ, 345, 245Carpenter, J. M. 2001, AJ, 121, 2851Chen, B. Q., Huang, Y., Hou, L. G., et al. 2019, MNRAS, 487, 1400Chen, P. S., Xiong, G. Z., & Wang, X. H. 1997, Ap&SS, 250, 227Comer´on, F., Djupvik, A. A., Schneider, N., & Pasquali, A. 2016, A&A, 586,A46Comer´on, F. & Pasquali, A. 2012, A&A, 543, A101Comer´on, F., Pasquali, A., Figueras, F., & Torra, J. 2008, A&A, 486, 453Comer´on, F. & Torra, J. 2018, A&A, 618, A67Comeron, F., Torra, J., & Figueras, F. 1997, A&A, 325, 149Costado, M. T., Alfaro, E. J., Gonz´alez, M., & Sampedro, L. 2017, MNRAS,465, 3879Csengeri, T., Bontemps, S., Schneider, N., et al. 2011, ApJ, 740, L5de Zeeuw, P. T., Hoogerwerf, R., de Bruijne, J. H. J., Brown, A. G. A., & Blaauw,A. 1999, AJ, 117, 354Downes, D. & Rinehart, R. 1966, ApJ, 144, 937Drew, J. E., Greimel, R., Irwin, M. J., & Sale, S. E. 2008, MNRAS, 386, 1761Drimmel, R., Bucciarelli, B., & Inno, L. 2019, Research Notes of the AmericanAstronomical Society, 3, 79Ekstr¨om, S., Georgy, C., Eggenberger, P., et al. 2012, A&A, 537, A146Ekstr¨om, S., Georgy, C., Meynet, G., Groh, J., & Granada, A. 2013, in EASPublications Series, Vol. 60, EAS Publications Series, ed. P. Kervella, T. LeBertre, & G. Perrin, 31–41Fluks, M. A., Plez, B., The, P. S., et al. 1994, A&AS, 105, 311Gahm, G. F. & Hultqvist, L. 1972, A&A, 16, 329Garcia, B. 1989, Bulletin d’Information du Centre de Donnees Stellaires, 36, 27Garmany, C. D. & Stencel, R. E. 1992, A&AS, 94, 211Grasdalen, G. L. & Sneden, C. 1979, PASP, 91, 337Gri ffi n, R. F. & Redman, R. O. 1960, MNRAS, 120, 287Groenewegen, M. A. T. 1993, A&A, 271, 180Grosbøl, P. & Carraro, G. 2018, A&A, 619, A50Hanson, M. M. 2003, ApJ, 597, 957Huang, W., Gies, D. R., & McSwain, M. V. 2010, ApJ, 722, 605Jorissen, A. & Mayor, M. 1992, A&A, 260, 115Keenan, P. C. & Hynek, J. A. 1945, ApJ, 101, 265Keenan, P. C. & McNeil, R. C. 1989, ApJS, 71, 245Kharchenko, N. V., Piskunov, A. E., R¨oser, S., Schilbach, E., & Scholz, R. D.2005, A&A, 438, 1163Kharchenko, N. V., Piskunov, A. E., Schilbach, E., R¨oser, S., & Scholz, R. D.2016, A&A, 585, A101Khoperskov, S., Gerhard, O., Di Matteo, P., et al. 2020, A&A, 634, L8Kohoutek, L. & Wehmeyer, R. 1999, A&AS, 134, 255Kuijken, K. & Gilmore, G. 1991, ApJ, 367, L9Kuˇcinskas, A., Hauschildt, P. H., Ludwig, H. G., et al. 2005, A&A, 442, 281Levesque, E. M. 2010, New A Rev., 54, 1Levesque, E. M., Massey, P., Olsen, K. A. G., & Plez, B. 2007, ApJ, 667, 202Levesque, E. M., Massey, P., Olsen, K. A. G., et al. 2005, ApJ, 628, 973Lindegren, L., Hern´andez, J., Bombrun, A., et al. 2018, A&A, 616, A2L´opez Ariste, A., Mathias, P., Tessore, B., et al. 2018, A&A, 620, A199Luri, X., Brown, A. G. A., Sarro, L. M., et al. 2018, A&A, 616, A9Mahy, L., Rauw, G., De Becker, M., Eenens, P., & Flores, C. A. 2015, A&A,577, A23McMillan, P. J. 2017, MNRAS, 465, 76
12. Comer´on et al.: The historical record of massive star formation in Cygnus
Melnik, A. M. & Dambis, A. K. 2020, MNRAS, 493, 2339Mel’Nik, A. M. & Efremov, Y. N. 1995, Astronomy Letters, 21, 10Mel’Nik, A. M., Sitnik, T. G., Dambis, A. K., Efremov, Y. N., & Rastorguev,A. S. 1998, Astronomy Letters, 24, 594Motte, F., Bontemps, S., Schilke, P., et al. 2007, A&A, 476, 1243Nagayama, T., Omodaka, T., Handa, T., et al. 2015, PASJ, 67, 66Nassau, J. J., Blanco, V. M., & Morgan, W. W. 1954, ApJ, 120, 478Negueruela, I., Marco, A., Herrero, A., & Clark, J. S. 2008, A&A, 487, 575Odenwald, S. F. 1989, AJ, 97, 801Odenwald, S. F. & Schwartz, P. R. 1993, ApJ, 405, 706Oke, J. B. 1990, AJ, 99, 1621Price, S. D. 1995, Space Sci. Rev., 74, 81Ram´on-Fox, F. G. & Bonnell, I. A. 2018, MNRAS, 474, 2028Rao, A., Gandhi, P., Knigge, C., et al. 2020, MNRAS, 495, 1491Reid, M. J., Menten, K. M., Brunthaler, A., et al. 2019, ApJ, 885, 131Reipurth, B. & Schneider, N. 2008, Star Formation and Young Clusters inCygnus, ed. B. Reipurth, 36Rygl, K. L. J., Brunthaler, A., Sanna, A., et al. 2012, A&A, 539, A79Schneider, N., Bontemps, S., Motte, F., et al. 2016, A&A, 591, A40Schneider, N., Bontemps, S., Simon, R., et al. 2006, A&A, 458, 855Schneider, N., Simon, R., Bontemps, S., Comer´on, F., & Motte, F. 2007, A&A,474, 873Sch¨onrich, R., McMillan, P., & Eyer, L. 2019, MNRAS, 487, 3568Sitnik, T. G., Egorov, O. V., Lozinskaya, T. A., et al. 2015, MNRAS, 454, 2486Stephenson, C. B. 1989, Publications of the Warner & Swasey Observatory, 3,53Straiˇzys, V., Boyle, R. P., Milaˇsius, K., et al. 2019, A&A, 623, A22Straiˇzys, V., Milaˇsius, K., Boyle, R. P., et al. 2014, AJ, 148, 89Straiˇzys, V., Vrba, F. J., Boyle, R. P., et al. 2015, AJ, 149, 161Terranegra, L., Chavarria-K., C., Diaz, S., & Gonzalez-Patino, D. 1994, A&AS,104, 557Torres-Dodgen, A. V. & Weaver, W. B. 1993, PASP, 105, 693Uyanıker, B., F¨urst, E., Reich, W., Aschenbach, B., & Wielebinski, R. 2001,A&A, 371, 675Vall´ee, J. P. 2018, Ap&SS, 363, 243Vall´ee, J. P. 2020, New A, 76, 101337Wang, J. J. & Hu, J. Y. 2000, A&A, 356, 118Wendker, H. J., Higgs, L. A., & Landecker, T. L. 1991, A&A, 241, 551Wo´zniak, P. R., Vestrand, W. T., Akerlof, C. W., et al. 2004, AJ, 127, 2436Wright, N. J., Drew, J. E., & Mohr-Smith, M. 2015, MNRAS, 449, 741Xu, Y., Li, J. J., Reid, M. J., et al. 2013, ApJ, 769, 15Xu, Y., Reid, M., Dame, T., et al. 2016, Science Advances, 2, e1600878
13. Comer´on et al.: The historical record of massive star formation in Cygnus T a b l e . C oo l s up e r g i a n t s i n t h e d i r ec ti ono f C ygnu s S t a r t yp e J K S l b ˜ ω µ α c o s δ µ δ v r a d ( ◦ )( ◦ )( m a s )( m a s y r − )( m a s y r − )( k m s − ) S up e r g i a n t s i n t h e C ygnu s a ss o c i a ti on s J . + . K . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . K . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . K . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . K . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . K . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . K . ± . . ± . . − . . ± . . ± . . ± . − . ± . J . + . M . ± . . ± . . − . . ± . − . ± . − . ± . − . ± . J . + . M . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . K . ± . . ± . . − . . ± . − . ± . − . ± . − . ± . J . + . M . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . K . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . M . ± . . ± . . − . . ± . − . ± . − . ± . − . ± . J . + . M . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . M . ± . . ± . . . . ± . − . ± . − . ± . J . + . M . . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . K . ± . . ± . . − . . ± . − . ± . − . ± . − . ± . J . + . K . ± . . ± . . . . ± . . ± . − . ± . − . ± . J . + . M . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . K . ± . . ± . . . − . ± . − . ± . − . ± . − . ± . J . + . M . . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . M . ± . . ± . . − . . ± . − . ± . − . ± . − . ± . J . + . M . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . K . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . M . ± . . ± . . − . . ± . − . ± . − . ± . − . ± . S u s p ec t e d f o r e g r ound a ndb ac kg r ound s up e r g i a n t s S t a r t yp e J K S l b ˜ ω µ α c o s δ µ δ v r a d ( ◦ )( ◦ )( m a s )( m a s y r − )( m a s y r − )( k m s − ) J . + . M . ± . . ± . . . . ± . − . ± . − . ± . J . + . M . ± . . ± . . . . ± . − . ± . − . ± . − . ± . J . + . K . ± . . ± . . . . ± . . ± . − . ± . − . ± . J . + . M . ± . . ± . . − . . ± . − . ± . − . ± . − . ± . J . + . M . . ± . . ± . . . . ± . − . ± . − . ± . − . ± . : B D + ◦ : HD , c l a ss i fi e d a s K I b ( B i d e l m a n1957 ; G r i ffi n & R e d m a n1960 ) : HD , c l a ss i fi e d a s K i n t h e H e n r y D r a p e r ca t a l og4 : HD , c l a ss i fi e d a s K i n t h e H e n r y D r a p e r ca t a l og5 : V C yg , c l a ss i fi e d a s M . I i n L e v e s qu ee t a l . ( ) : B I C yg , c l a ss i fi e d a s M I a b i n G a h m& H u lt qv i s t ( ) : BCC yg , c l a ss i fi e d a s M I( L e v e s qu ee t a l . ) , M I a ( G a h m& H u lt qv i s t ) : C l a ss i fi e d a s M I i n G r a s d a l e n & S n e d e n ( ) : KY C yg , c l a ss i fi e d a s M I- M I i n ( L e v e s qu ee t a l . )
14. Comer´on et al.: The historical record of massive star formation in Cygnus : C l a ss i fi e d a s M i n N a ss a u e t a l . ( ) : V C yg12 : H a r o - C h a v i r a , c l a ss i fi e d a s M I i n G r a s d a l e n & S n e d e n ( ) : H a r o - C h a v i r a , c l a ss i fi e d a s M I i n G r a s d a l e n & S n e d e n ( ) : C l a ss i fi e d a s M I i n G r a s d a l e n & S n e d e n ( ) : V C yg , c l a ss i fi e d a s M I i n G r a s d a l e n & S n e d e n ( ) : RRC yg , c l a ss i fi e d a s M i n C a m e r on & N a ss a u ( ) : V C yg , c l a ss i fi e d a s M i n N a ss a u e t a l . ( ) : HD , c l a ss i fi e d a s K . II b i n K ee n a n & M c N e il ( ) : L i s t e d a s ca r bon s t a r ca nd i d a t e i n S t e ph e n s on ( ) ; C h e n e t a l . ( ))