Mon. Not. R. Astron. Soc. 388, 444–456 (2008) doi:10.1111/j.1365-2966.2008.13413.x Galactic clusters with associated Cepheid variables – VII. Berkeley 58 and CG Cassiopeiae D. G. Turner, 1† , ‡ , § D. Forbes, 2 † D. English, 2 ‡ P. J. T. Leonard, 3 J. N. Scrimger, 4 A. W. Wehlau, 5 R. L. Phelps, 6 † L. N. Berdnikov 7 § and E. N. Pastukhova 8 1 Department of Astronomy and Physics, Saint Mary’s University, Halifax, Nova Scotia B3H 3C3, Canada 2 Department of Physics, Sir Wilfred Grenfell College, Memorial University of Newfoundland, Corner Brook, NF A2H 6P9, Canada 3 ADNET Systems, Inc., 7515 Mission Dr, Suite A1C1, Lanham, Maryland 20706, USA 4 Faculty of Computer Science, Dalhousie University, Halifax, Nova Scotia B3H 1W5, Canada 5 Department of Physics and Astronomy, The University of Western Ontario, London, Ontario N6A 3K7, Canada 6 Office of Integrative Activities, National Science Foundation, Division of Astronomical Sciences, 4201 Wilson Blvd, Arlington, Virginia 22230, USA 7 Sternberg Astronomical Institute, 13 Universitetskij prosp., Moscow 119992, Russia 8 Institute of Astronomy, Russian Academy of Sciences, 48 Pyatnitskaya ul, Moscow 109017, Russia Accepted 2008 May 1. Received 2008 May 1; in original form 2008 March 31 ABSTRACT Photoelectric, photographic and CCD UBV photometry, spectroscopic observations and star counts are presented for the open cluster Berkeley 58 to examine a possible association with the 4.37 d Cepheid CG Cas. The cluster is difficult to separate from the early-type stars belonging to the Perseus spiral arm, in which it is located, but has reasonably well-defined parameters: an evolutionary age of ∼10 8 yr, a mean reddening of E(B − V ) (B0) = 0.70 ± 0.03 s.e. and a distance of 3.03 ± 0.17 kpc (V 0 −M V = 12.40 ± 0.12 s.d.). CG Cas is a likely cluster coronal member on the basis of radial velocity, and its period increase of +0.170 ± 0.014 s yr −1 and large light amplitude describe a Cepheid in the third crossing of the instability strip lying slightly blueward of strip centre. Its inferred reddening and luminosity are E(B − V ) = 0.64 ± 0.02 s.e. and M V =−3.06 ± 0.12. A possible K supergiant may also be a cluster member. Key words: stars: evolution – Cepheids – open clusters and associations: individual: Berkeley 58. 1 INTRODUCTION After the rediscovery in the early 1950s of spatial coincidences be- tween Cepheids and open clusters by Irwin (1955, 1958), Eggen (see Sandage 1958) and Kholopov (1956), a number of searches for additional coincidences were made by Kraft (1957), van den Bergh (1957) and Tifft (1959), among others. Tifft’s search resulted in the discovery of a near-spatial coincidence between the 4.37 d Cepheid CG Cassiopeiae and an anonymous open cluster, subsequently cat- alogued as Berkeley 58 (Setteducati & Weaver 1962), which lies E-mail: [email protected]†Visiting Astronomer, Kitt Peak National Observatory, National Optical As- tronomy Observatories, which is operated by the Association of Universities for Research in Astronomy, Inc. (AURA) under cooperative agreement with the National Science Foundation. ‡Visiting Astronomer, Dominion Astrophysical Observatory, Herzberg In- stitute of Astrophysics, National Research Council of Canada. §Visiting Astronomer, Harvard College Observatory Photographic Plate Stacks. less than one cluster diameter to the west. The field is coincident with a portion of the Perseus spiral arm that is relatively rich in open clusters, and the cluster NGC 7790 with its three Cepheid members lies in close proximity. The possibility that CG Cas might be an outlying member of NGC 7790 was raised at one time by Efre- mov (1964a,b), and found some support in a star count analysis by Kovalenko (1968). More detailed star counts in the field (Turner 1985) indicate otherwise, as do the available proper motion data (Frolov 1974, 1977). The Cepheid does lie in the corona of Berkeley 58 (Turner 1985), although Frolov has argued that it is not a probable cluster member. Given a probable distance of 3 kpc to both CG Cas and Berkeley 58 (e.g. Frolov 1979; Phelps & Janes 1994), it is not clear that existing proper motion data are precise enough to provide conclusive evidence pertaining to the cluster membership of CG Cas. The present study was therefore initiated in order to examine the case in more detail. As demonstrated here, there is strong ev- idence that CG Cas is a likely member of Berkeley 58 and that it can serve as a calibrator for the Cepheid period–luminosity (PL) relation. C 2008 The Authors. Journal compilation C 2008 RAS
13
Embed
Galactic clusters with associated Cepheid variables – VII ... › 585 › 1 › galactic_clusters_associate… · 7Sternberg Astronomical Institute, 13 Universitetskij prosp., Moscow
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Mon. Not. R. Astron. Soc. 388, 444–456 (2008) doi:10.1111/j.1365-2966.2008.13413.x
Galactic clusters with associated Cepheid variables – VII. Berkeley 58
and CG Cassiopeiae
D. G. Turner,1�†,‡,§ D. Forbes,2† D. English,2‡ P. J. T. Leonard,3 J. N. Scrimger,4
A. W. Wehlau,5 R. L. Phelps,6† L. N. Berdnikov7§ and E. N. Pastukhova8
1Department of Astronomy and Physics, Saint Mary’s University, Halifax, Nova Scotia B3H 3C3, Canada2Department of Physics, Sir Wilfred Grenfell College, Memorial University of Newfoundland, Corner Brook, NF A2H 6P9, Canada3ADNET Systems, Inc., 7515 Mission Dr, Suite A1C1, Lanham, Maryland 20706, USA4Faculty of Computer Science, Dalhousie University, Halifax, Nova Scotia B3H 1W5, Canada5Department of Physics and Astronomy, The University of Western Ontario, London, Ontario N6A 3K7, Canada6Office of Integrative Activities, National Science Foundation, Division of Astronomical Sciences, 4201 Wilson Blvd, Arlington, Virginia 22230, USA7Sternberg Astronomical Institute, 13 Universitetskij prosp., Moscow 119992, Russia8Institute of Astronomy, Russian Academy of Sciences, 48 Pyatnitskaya ul, Moscow 109017, Russia
Accepted 2008 May 1. Received 2008 May 1; in original form 2008 March 31
ABSTRACT
Photoelectric, photographic and CCD UBV photometry, spectroscopic observations and starcounts are presented for the open cluster Berkeley 58 to examine a possible association with the4.37 d Cepheid CG Cas. The cluster is difficult to separate from the early-type stars belongingto the Perseus spiral arm, in which it is located, but has reasonably well-defined parameters:an evolutionary age of ∼108 yr, a mean reddening of E(B − V) (B0) = 0.70 ± 0.03 s.e. and adistance of 3.03 ± 0.17 kpc (V0−MV = 12.40 ± 0.12 s.d.). CG Cas is a likely cluster coronalmember on the basis of radial velocity, and its period increase of +0.170 ± 0.014 s yr−1
and large light amplitude describe a Cepheid in the third crossing of the instability striplying slightly blueward of strip centre. Its inferred reddening and luminosity are E(B − V) =0.64 ± 0.02 s.e. and 〈MV 〉 = −3.06 ± 0.12. A possible K supergiant may also be a clustermember.
Key words: stars: evolution – Cepheids – open clusters and associations: individual:Berkeley 58.
1 IN T RO D U C T I O N
After the rediscovery in the early 1950s of spatial coincidences be-tween Cepheids and open clusters by Irwin (1955, 1958), Eggen(see Sandage 1958) and Kholopov (1956), a number of searches foradditional coincidences were made by Kraft (1957), van den Bergh(1957) and Tifft (1959), among others. Tifft’s search resulted in thediscovery of a near-spatial coincidence between the 4.37 d CepheidCG Cassiopeiae and an anonymous open cluster, subsequently cat-alogued as Berkeley 58 (Setteducati & Weaver 1962), which lies
�E-mail: [email protected]†Visiting Astronomer, Kitt Peak National Observatory, National Optical As-tronomy Observatories, which is operated by the Association of Universitiesfor Research in Astronomy, Inc. (AURA) under cooperative agreement withthe National Science Foundation.‡Visiting Astronomer, Dominion Astrophysical Observatory, Herzberg In-stitute of Astrophysics, National Research Council of Canada.§Visiting Astronomer, Harvard College Observatory Photographic PlateStacks.
less than one cluster diameter to the west. The field is coincidentwith a portion of the Perseus spiral arm that is relatively rich in openclusters, and the cluster NGC 7790 with its three Cepheid memberslies in close proximity. The possibility that CG Cas might be anoutlying member of NGC 7790 was raised at one time by Efre-mov (1964a,b), and found some support in a star count analysis byKovalenko (1968). More detailed star counts in the field (Turner1985) indicate otherwise, as do the available proper motion data(Frolov 1974, 1977). The Cepheid does lie in the corona ofBerkeley 58 (Turner 1985), although Frolov has argued that it isnot a probable cluster member.
Given a probable distance of 3 kpc to both CG Cas andBerkeley 58 (e.g. Frolov 1979; Phelps & Janes 1994), it is notclear that existing proper motion data are precise enough to provideconclusive evidence pertaining to the cluster membership of CGCas. The present study was therefore initiated in order to examinethe case in more detail. As demonstrated here, there is strong ev-idence that CG Cas is a likely member of Berkeley 58 and that itcan serve as a calibrator for the Cepheid period–luminosity (PL)relation.
A variety of observations were obtained for the present investi-gation. Table 1 presents photoelectric UBV photometry for brightmembers of Berkeley 58, obtained during observing runs at KittPeak National Observatory in 1981 September, 1982 August and1984 August. The data, acquired using 1P21 photomultipliers andstandard UBV filter sets used in conjunction with pulse-countingphotometers on the No. 4 0.4-m, No. 2 0.9-m, and 1.3-m telescopesat Kitt Peak, have associated uncertainties typical of our previousinvestigations of Cepheid clusters (Turner 1992; Turner, Forbes &Pedreros 1992; Turner, Mandushev & Forbes 1994), namely stan-dard internal errors for a single observation of ±0.01 in V andB − V, and ±0.02 in U − B, for stars brighter than V = 13. Theestimated external errors for all but the faintest stars are similar inmagnitude. The stars are identified by their numbering in Fig. 1, aswell as by their 2000 coordinates in the Two Micron All-Sky Survey(2MASS) (Cutri et al. 2003); the number of individual observationsfor each star is given in Column 7 of Table 1.
Star 6 is the eclipsing system V654 Cas, for which Berdnikov(1993) cites photoelectric values of V and B − V outside of eclipsethat are close to the values given here. Star 30 is a close opti-cal double with components of nearly identical brightness. Thephotoelectric values apply to the combined light from both stars,whereas CCD observations provide uncontaminated data for thesouth-western star of the pair, as established by its CCD magni-tude being 0.75 mag fainter. By contrast, the CCD V magnitude forstar 35 is 0.21 mag brighter, which suggests possible variability inthe object. Individual photoelectric observations for CG Cas arepresented in Table 2.
Photographic UBV photometry was also obtained for stars inthe nuclear and coronal regions of Berkeley 58 from photographicplates of the cluster field obtained in 1984 September using the1.2-m Elginfield telescope of the University of Western Ontario. Thestar images were measured using the IRIS diaphragm photometer atSaint Mary’s University, and were reduced to the UBV system andcalibrated with reference to the photoelectric standards identified inTable 1 using the techniques discussed by Turner & Welch (1989).
Figure 1. A finder chart for the field of Berkeley 58 from the red image ofthe Palomar Observatory Sky Survey. The field of view measures 20 arcmin× 20 arcmin and is centred at 2000 coordinates: RA = 00h00m12.s9, Dec. =+60◦56′07′′. The top image depicts the location of CG Cas relative to thecluster core, the lower image identifies photoelectrically observed stars.[The National Geographic Society-Palomar Observatory Sky Atlas (POSS-I) was made by the California Institute of Technology with grants from theNational Geographic Society.]
The resulting data are presented in Table 3 in similar format to thedata of Table 1, and the stars are identified by their 2000 coordinates.The photographic values for cluster stars in common with the CCDsurvey (Phelps & Janes 1994) agree very closely with the CCDvalues, when the latter are adjusted to the present system. However,earlier photographic UBV photometry of cluster stars by Frolov(1979) displays systematic differences relative to the present data.Since the present survey samples a much larger number of cluster
Table 2. Photoelectric UBV observations for CG Cas-siopeiae.
stars, no attempt was made to combine Frolov’s data with the presentphotometry.
CCD UBV photometry for stars in the nuclear region of Berkeley58 was published previously by Phelps & Janes (1994), but for thisstudy was recalibrated using the Table 1 stars as standards. Therevised photometry for cluster stars is presented in Table 4, wherethe star numbers correspond to the scheme adopted by Phelps &Janes (1994), incremented by 1000. The stars are also identifiedby their 2000 coordinates. Since the U-band measurements have amuch brighter limit than the B and V measures, the CCD photometryis less useful for studying the reddening in the field. But it is valuablefor identifying the faint portion of the cluster main sequence.
Spectroscopic imaging of bright stars in Berkeley 58 wasmade in 1984 July and 1985 September using the Cassegrainspectrograph on the 1.8-m Plaskett telescope of the DominionAstrophysical Observatory. The observations, at a dispersionof 15 Å mm−1 and centred in the blue spectral region, wererecorded photographically and later scanned for radial velocitymeasurement with the PDS microdensitometer at the David DunlapObservatory of the University of Toronto (see Turner & Drilling1984). It was also possible to estimate spectral types for the starsfrom the photographic spectra, with results presented in Table 1.
The field of the Cepheid CG Cas was also examined on archivalimages in the collections of Harvard College Observatory andSternberg Astronomical Institute in order to obtain brightness esti-mates for the star and to construct seasonal light curves for com-parison with a standard light curve constructed from photoelectricobservations (Berdnikov 2007). The resulting data were used toestimate times of light maximum for the Cepheid and to track itsO−C changes, the differences between observed (O) and computed(C) times of light maximum. Rate of period change, in conjunctionwith light amplitude, is an excellent diagnostic of the location ofindividual Cepheids in the instability strip (Turner, Abdel-SabourAbdel-Latif & Berdnikov 2006a), such information providing anexcellent parameter for comparison with what can be gleaned frominformation on the age of the surrounding stars provided by thecluster Hertzsprung–Russell diagram.
3 STA R C O U N T S
The first step in studying Berkeley 58 involved star counts madeusing a photographic enlargement from a glass copy of the Palomar
Observatory Sky Survey (POSS) E plate for the field. Strip counts inseveral different orientations delineated the cluster centre, followedby ring counts illustrated in Fig. 2; the centre of symmetry is locatedat RA = 00h00m12.s9, Dec. = +60◦56′07′′ (2000). The upperportion of Fig. 2 illustrates ring counts for stars detected on the2MASS survey (Cutri et al. 2003) to the survey limit, whereas thelower portion shows star counts from the POSS-E plate to twodifferent magnitude limits.
The counts from the 2MASS survey were made without regardfor overlap with the star cluster NGC 7790, which lies 23 arcmin tothe north-west of Berkeley 58, whereas the counts from the POSS-Eplate were restricted beyond 11 arcmin from the cluster centre tosectors that avoided overlap with the outlying regions of NGC 7790.The effect of contamination from the coronal region of NGC 7790is detectable in the 2MASS star counts beyond roughly 12 arcminfrom the cluster centre, but because of restrictions imposed by thelocation of Berkeley 58 on the POSS, we were unable to establishuncontaminated star counts from the POSS-E plate beyond about15 arcmin from the cluster centre. Nevertheless, the two sets ofcounts appear to yield similar parameters for the inner regions ofthe cluster. Berkeley 58 is estimated to have a nuclear radius of rn �4.5 arcmin (4.0 pc) in the notation of Kholopov (1969), whereas thecoronal (or tidal) radius is estimated to be Rc � 11 arcmin (9.7 pc)from the trends in the 2MASS star densities as well as the apparentflattening of the POSS-E star densities in the outermost rings.
Star counts predict a total of 197 ± 27 members brighter thanthe limit of the 2MASS survey lying within 5 arcmin of the clustercentre, 487 ± 82 members within 11 arcmin of the cluster centre,field stars within the same regions being 715 and 4835, respectively.Field stars clearly outnumber cluster members in both regions. CGCas is located 5.8 arcmin from the centre of Berkeley 58, in the clus-ter coronal region just beyond its nuclear boundaries. Although notprojected on the core of Berkeley 58, CG Cas is spatially coincidentwith the cluster, which occupies most of the field of Fig. 1.
4 B E R K E L E Y 5 8
Fig. 3 is a UBV colour–colour diagram for the field of Berkeley 58surveyed in this study, as constructed from the data of Tables 1, 3and 4. The phase-averaged data for CG Cas are from Berdnikov(2007). A reddened sequence of B- and A-type cluster members
can be detected in the data, but a cluster reddening of E(B − V)� 0.7 places them in a section of the colour–colour diagram wherethey can be confused photometrically with unreddened, foreground,G-type stars. For that reason, it becomes essential to make theprocess of photometric identification of likely spectral classes forindividual stars as reliable as possible, through the use of a well-established interstellar extinction relation. The spectral types ob-tained for six of the B-type, photoelectrically observed, cluster starsimply a reddening law for Berkeley 58 described by E(U − B)/E(B − V) = 0.75, along with a small curvature term (Turner 1989),identical to the reddening slope found previously for star clustersspatially adjacent to Berkeley 58 (Turner 1976b). Berkeley 58 starswere therefore dereddened with such a relationship, except for late-type stars where a steeper relationship was adopted, dependent uponthe likely intrinsic colours of the stars.
Fig. 3 data indicate an absence of any unreddened O-, B- orA-type stars in the observed sample. That feature is confirmed byavailable 2MASS data for the observed stars (Cutri et al. 2003),which are depicted in the JHK colour–colour diagram of Fig. 4.An intrinsic relation for main-sequence stars in the 2MASS systemwas constructed from 2MASS observations of unreddened standardstars and stars in open clusters of uniform reddening (e.g. Turner1996b), adjusted with a reddening slope E(H − K)/E(J − H) =0.55, as derived from reddened stars of known spectral type. Thenumber of cluster stars with U-band observations is a small fractionof the total sample, so Fig. 4 contains many more stars than Fig. 3.The selection of 2MASS data was also not restricted according tothe magnitude of cited uncertainties in the data, so several pointsin Fig. 4 display unusually large scatter. It seems clear, however,that the sample of cluster stars surveyed consists mainly of starsreddened by E(J − H) ≥ 0.1, which corresponds to E(B − V) ≥0.36.
The correlation of reddening with distance towards Berkeley 58was established from the available UBV photometry by deredden-ing the colours for individual stars in conjunction with a copy ofthe POSS field on which derived colour excesses E(B − V) wererecorded as they were obtained, with multiple solutions resolved byreference to the reddenings for spatially adjacent stars as well asby the reddenings derived for the stars from their 2MASS colours(Fig. 4). In most cases, the smaller JHK reddening of stars rela-tive to those obtained from UBV colours was sufficient to resolve
questions about likely intrinsic colours for the stars, but there werea number of ambiguous cases where the data from the two surveysyielded disparate solutions, for example 2MASS colours implyingan early spectral type and UBV colours implying a late spectral type.Such cases were unimportant in the final analysis, but are curiousnevertheless.
Distance moduli were calculated for individual stars by adop-tion of zero-age main sequence (ZAMS) values of MV (Turner1976a, 1979), so the values systematically underestimate V − MV
for unresolved binaries and evolved stars. The resulting scatter inthe variable-extinction diagram of Fig. 5 therefore contains a sys-tematic component towards small values of V − MV . Within suchconstraints, it is possible to discern certain trends in the data, suchas the lack of any significant reddening out to distances of ∼600 pc(V0 − MV = 8.9), with a reddening of E(B − V) ≥ 0.4 beyondthat to distances of ∼2700 pc (V0 − MV = 12.16) or more. Atthe Galactic location of CG Cas (l = 116.◦845, b = −1.◦315), amore encompassing survey by Neckel & Klare (1980) implies asimilar trend, with the reddening beginning at distances of ∼400–900 pc. Apparently, the main extinction for stars in the direction
of Berkeley 58 occurs near the far side of the local spiral armfeature.
But the picture is not that simple. When the derived reddeningsare compared star-for-star in the field of Berkeley 58, there are noobvious trends with spatial location, and trends with distance aredifficult to establish without highly accurate luminosities for theobserved stars. It can be surmised that there is additional reddeningoccurring on the near side of the Perseus spiral arm, given the natureof the scatter in the colour excesses. Likely members of Berkeley58 generally have reddenings of E(B − V) � 0.70, with largervalues possibly arising from circumstellar extinction, particularlyfor late B-type stars where rapid rotation is common (e.g. Turner1993, 1996a). An identical feature is observed in the adjacent clusterNGC 7790 (Takala 1988). A lower envelope trend for the reddenedstars in Fig. 5 implies a ratio of total-to-selective extinction for thefield of R = AV /E(B − V) = 2.95 ± 0.30 from least squares and non-parametric analyses. The value is consistent with previous studiesof clusters in this region of the Galaxy (Turner 1976b), as well aswith a value of R � 2.95 expected for local extinction described by areddening slope of 0.75 (Turner 1996a). For subsequent calculations
Figure 2. Star densities for the field of Berkeley 58, as measured in ringsrelative to the adopted cluster centre. The upper diagram contains ring countsmade from the 2MASS survey, the lower two diagrams ring counts fromthe POSS-E plate of the field for a faint limit (middle) and a brighter limit(lower). The location of CG Cas relative to the cluster centre is indicated byan arrow.
a value of R = 2.95 was adopted, the exact choice affecting estimatesof distance but not the derived luminosity for CG Cas as a clustermember.
An observational colour–magnitude diagram for the sampledstars is presented in Fig. 6, with a ZAMS plotted for V − MV =14.29, the apparent distance modulus at E(B − V) = 0.70 for pointson the lower relation of Fig. 5. Such parameters provide a reasonablefit to the data, but there remain anomalies requiring further exami-nation. For example, Fig. 6 contains reddened B-type stars more lu-minous than the turnoff magnitude for a cluster containing CG Cas,a point also indicated in Fig. 3, where dashed relations indicate red-dening lines for B6.5 V and A2 V stars, the former corresponding tothe expected turnoff colour [(B − V)0 = −0.13] for stars associatedwith a 4.37 d Cepheid (Turner 1996c). Clearly, the field containsa number of stars younger than the expected evolutionary age ofCG Cas.
Such complications may be endemic to the field of both Berkeley58 and NGC 7790, where the line of sight crosses the interarmregion between the Sun and portions of the local spiral feature,then intercepts the Perseus spiral arm with a marked increase inspace density for young B-type stars and young-to-intermediateage star clusters. The separation of spiral arm stars from clustermembers is difficult but achievable, since the radial velocities forCG Cas and Berkeley 58 stars listed in Table 5 imply a conspicuous
Figure 3. A UBV colour–colour diagram for observed Berkeley 58 stars:photoelectric observations (filled circles), photographic observations sup-plemented by CCD observations (open circles), CCD observations (filledtriangles), and CG Cas (circled point). The intrinsic relation for main se-quence stars is plotted as a solid line, with the same relation reddened byE(B − V) = 0.38 and E(B − V) = 0.70 shown by dotted lines. The reddeningrelations for stars of spectral type B6.5 V and A2 V are shown as dashedlines.
Figure 4. A 2MASS colour–colour diagram, H − K versus J − H, for starsexamined in the field of Berkeley 58, without regard to the uncertainties inthe observations (Cutri et al. 2003). The intrinsic relation for main sequencestars is plotted as a solid line, as derived from the observed colours ofstandard stars and stars in clusters of uniform reddening. The direction ofreddening in the 2MASS system is indicated.
Figure 5. A variable-extinction diagram for observed Berkeley 58 stars,with symbols as in Fig. 3. Reddening relations of slope R = AV /E(B − V) =2.95 are shown corresponding to distances of d � 600 pc (V0 − MV = 8.9)and d � 2700 pc (V0 − MV = 12.16).
Figure 6. A colour–magnitude diagram for Berkeley 58 from all observa-tions: photoelectric (filled circles), photographic (open circles), and CCD(triangles) data. CG Cas is the circled point. The ZAMS is depicted forE(B − V) = 0.70 and V − MV = 14.29.
velocity difference between the cluster and spiral arm stars. Theanomalously young B stars noted above are objects like stars 6(V654 Cas), 7 and possibly 24, which have systematically morepositive velocities than likely cluster members: stars 11, 18 and23, which have radial velocities close to the systemic velocity ofCG Cas (see Fig. 7, which includes radial velocity measurementsfrom Joy 1937; Metzger et al. 1991; Gorynya et al. 1998). Exceptfor star 11, which may be anomalous, stars with radial velocities
Figure 7. The radial velocity variations of CG Cas, with cited uncertainties,as measured in this paper (open circles), from Metzger et al. (1991) (filledsquares) and Gorynya et al. (1998) (filled circles), and from Joy (1937)(open squares). The curve is a simple spectroscopic binary solution to thedata from the first three data sets, the data from Joy (1937) exhibitingsystematic deviations near velocity minimum.
Table 5. Radial velocity data for Berkeley 58 Stars.
close to that of CG Cas also have spectral types near the expectedB6.5 V turnoff. Unfortunately, it is not possible to identify faintercluster members by the same technique, given the bright limit forthe present radial velocity survey. Follow-up observations would beuseful in that regard.
The complications arising from contamination of the cluster fieldby young stars in the Perseus arm and likely circumstellar reddening
for late B-type members were addressed by identifying unaffectedcluster stars from their reddenings, which are close to E(B − V)(B0) = 0.70. The field of the CCD survey near the cluster cen-tre was found to exhibit a mean reddening of E(B − V) (B0) =0.697 ± 0.025, that for the region of CG Cas a mean reddening ofE(B − V) (B0) = 0.685 ± 0.022. Stars with full UBV data wereidentified as likely cluster members on the basis of reddenings com-parable to or larger than those values, while stars near the clustercentre lacking U-band data were assumed to have B0 star colourexcesses as above, but intrinsic colours adjusted for the spectral typedependence of reddening (see Fernie 1963). A-type dwarfs can suf-fer complications arising from the effects of rotation on their stellarcontinua and UBV colours (Turner, Usenko & Kovtyukh 2006b), sothe adoption of space reddenings for such stars may circumvent po-tential biases introduced by dereddening their colours to the intrinsicrelation for zero-age zero-rotation main-sequence stars. The result-ing reddening-corrected colour–magnitude diagram for the clusteris plotted in Fig. 8 for 145 likely members, along with CG Cas andits light variations and star 5, which is considered to be a potentialK giant member. The reddening for CG Cas corrected for its colouris E(B − V) = 0.64 ± 0.02. A photometric reddening could be ob-tained from the BVIc observations of Henden (1996) (see Laney &Caldwell 2007), but a field reddening was adopted as a precautionagainst potential bias towards large-amplitude Cepheids lying nearthe centre of the instability strip (unnecessary in the present case,as it turns out).
The distance to Berkeley 58 is established by 40 of its A-typeZAMS members, which yield a value of V0 − MV = 12.40 ± 0.12
Figure 8. A reddening free colour–magnitude diagram for Berkeley 58.The dashed curve represents the ZAMS for V0 − MV = 12.40, and the solidline with dotted lines on either side represents an isochrone from Meynetet al. (1993) for log τ = 8.0 ± 0.1. The range of light variations for CG Casis depicted, as are the observational boundaries for the Cepheid instabilitystrip. The red object on the evolved giant sequence is star 5.
s.d., corresponding to a distance of 3026 ± 166 pc. Except for star11, which is conceivably a rapid rotator observed nearly pole-on,the bluest cluster stars correspond to spectral type B6 with (B − V)0
= −0.16. A comparison with stellary evolutionary models (Meynet,Mermilliod & Maeder 1993) implies a cluster age of 10 ± 1 × 107 yr(log τ = 8.0 ± 0.05). The corresponding mass of cluster stars fallingat the tip of the main-sequence red turnoff is 5.4 M (Meynet et al.1993).
5 C G CASSIOPEIAE
The systemic radial velocity of CG Cas (Table 5) is a close matchto the mean velocity of Berkeley 58 derived from likely clustermembers 11, 18 and 23, and the evolutionary age of the clus-ter closely matches what is predicted for the pulsation periodof the Cepheid (Turner 1996c). The luminosity of CG Cas as alikely member of Berkeley 58 is 〈MV 〉 = −3.06 ± 0.12, whichmatches a value of 〈MV 〉 = −3.04 predicted with a Cepheid period–radius relation and the inferred effective temperature of CG Cas(log Teff = 3.775) from its derived intrinsic colour (Turner & Burke2002). The case for membership of CG Cas in Berkeley 58 is verystrong.
The exact evolutionary status of CG Cas can be established fromthe direction and rate of its period changes (Turner et al. 2006a),in conjunction with its large blue light amplitude of �B = 1.22(Berdnikov 2007). The period changes for CG Cas were estab-lished here from examination of archival photographic plates in theHarvard and Sternberg collections, as well as from an analysis ofnew and existing photometry for the star. A working ephemeris forCG Cas based upon the available data was
JDmax = 2432436.94 + 4.3656292 E,
where E is the number of elapsed cycles. An extensive analysis ofall available observations produced the data summarized in Table 6,which lists the results for different epochs, the type of data ana-lyzed (PG = photographic, VIS = visual telescopic observations,B = photoelectric B, and V = photoelectric V), the number of ob-servations used to establish the times of light maximum and thesource of the observations, in addition to the temporal parameters.The data are plotted in Fig. 9.
A regression analysis of the O−C data of Table 6 produced aparabolic solution for the ephemeris defined by
JDmax = 2432436.9493(±0.0080)
+ 4.3656289(±0.0000024) E + 1.1757(±0.0983) × 10−7 E2,
which is plotted in Fig. 9. The parabolic trend corresponds to aperiod increase of +0.170 ± 0.014 s yr−1 (log P = −0.770±0.036),a value typical of Cepheids lying slightly blueward of the centreof the instability strip and in the third crossing. The location ofCG Cas in Fig. 8 relative to the observational boundaries of theCepheid instability strip (Turner et al. 2006b) is consistent withthat conclusion, although the stellar evolutionary models seem torequire adjustments (metallicity, mixing of surface layers?) to matchthe observations.
6 D ISCUSSION
The case for potential membership of the Cepheid CG Cas in thesparse open cluster Berkeley 58 has been studied using photomet-ric (pe, pg, CCD) observations, spectroscopy (VR, spectral types),star counts and O−C data for the Cepheid. The cluster Berkeley58 is particularly difficult to separate from the young stars of the
HJDmax ±σ Band Epoch O−C Observations Reference(E) (phase) (n)
241 3407.3442 0.0292 PG −4359 +0.1714 55 This paper (Harvard)241 5144.8677 0.0428 PG −3961 +0.1746 7 This paper (SAI )241 6314.8382 0.0338 PG −3693 +0.1566 72 This paper (Harvard)241 7572.0492 0.1070 PG −3405 +0.0664 11 This paper (SAI)241 9794.1315 0.0336 PG −2896 +0.0436 63 This paper (Harvard)242 3788.6688 0.0271 PG −1981 +0.0304 98 This paper (Harvard)242 6102.4299 0.0196 PG −1451 +0.0082 128 This paper (Harvard)242 6940.6916 0.0567 VIS −1259 +0.0691 46 Lange (1933)242 8023.3553 0.0571 PG −1011 +0.0569 19 This paper (SAI)242 8455.5134 0.0271 PG −912 +0.0177 92 This paper (Harvard)242 9568.7382 0.0559 PG −657 +0.0071 28 This paper (SAI)243 0847.7814 0.0321 PG −364 −0.0790 81 This paper (Harvard)243 1576.8823 0.0854 PG −197 −0.0381 17 Erleksova (1961)243 3100.4046 0.0430 PG +152 −0.1203 59 This paper (Harvard)243 3183.4566 0.0308 PG +171 −0.0152 37 This paper (SAI)243 3371.0678 0.0848 PG +214 −0.1261 23 Erleksova (1961)243 4117.6643 0.0350 PG +385 −0.0521 25 This paper (SAI)243 5174.1804 0.0285 PG +627 −0.0182 74 This paper (SAI)243 5379.4291 0.0443 PG +674 +0.0459 10 Romano (1959)243 5619.5498 0.1388 PG +729 +0.0570 19 Erleksova (1961)243 5837.7841 0.0168 PG +779 +0.0099 18 Zonn & Semeniuk (1959)243 6802.5876 0.0070 B +1000 +0.0094 13 Oosterhoff (1960)243 6802.6183 0.0119 V +1000 +0.0401 15 Oosterhoff (1960)243 6933.5492 0.0054 B +1030 +0.0021 22 Bahner, Hiltner & Kraft (1962)243 6937.9440 0.0085 V +1031 +0.0313 23 Bahner et al. (1962)243 8666.6957 0.0174 PG +1427 −0.0061 41 This paper (SAI)243 9077.0406 0.0299 PG +1521 −0.0303 16 This paper (SAI)244 0268.8346 0.0241 PG +1794 −0.0530 24 This paper (SAI)244 1146.3548 0.0121 PG +1995 −0.0242 95 This paper (SAI)244 1866.7282 0.0142 PG +2160 +0.0204 55 This paper (SAI)244 2355.6761 0.0178 PG +2272 +0.0178 47 This paper (SAI)244 2862.0722 0.0159 PG +2388 +0.0010 74 This paper (SAI)244 3045.5091 0.0058 V +2430 +0.0815 71 Chekanikhina (1982)244 3957.9197 0.0206 PG +2639 +0.0756 25 This paper (SAI)244 4844.1310 0.0099 B +2842 +0.0643 9 Berdnikov (1986)244 4852.8817 0.0150 V +2844 +0.0837 11 Berdnikov (1986)244 5189.0177 0.0117 B +2921 +0.0663 8 Berdnikov (1986)244 5189.0509 0.0074 V +2921 +0.0995 8 Berdnikov (1986)244 5394.1872 0.0098 B +2968 +0.0512 14 This paper244 5429.1355 0.0115 V +2976 +0.0745 15 This paper244 5883.1690 0.0061 B +3080 +0.0826 8 Berdnikov (1986)244 5883.1870 0.0086 V +3080 +0.1006 8 Berdnikov (1986)244 7760.4530 0.0042 B +3510 +0.1461 39 Berdnikov (1992a)244 7760.4823 0.0059 V +3510 +0.1754 39 Berdnikov (1992a)244 8118.4162 0.0060 B +3592 +0.1277 18 Berdnikov (1992b)244 8118.4546 0.0085 V +3592 +0.1661 18 Berdnikov (1992b)244 8515.7127 0.0043 B +3683 +0.1520 20 Berdnikov (1992c)244 8515.7328 0.0052 V +3683 +0.1721 20 Berdnikov (1992c)245 1458.2287 0.0152 V +4357 +0.2341 27 Wozniak et al. (2004)
Perseus spiral arm, which raises concerns about future studies ofdistant open cluster calibrators for the Cepheid PL relation. Care-ful analysis of the available data leads to a cluster reddening ofE(B − V) (B0) = 0.70, a distance of 3.03 ± 0.17 kpc and an age of10 ± 1 × 107 yr. CG Cas is a likely member on the basis of radialvelocity, location outside the cluster nucleus within the cluster coro-nal region, evolutionary status indicated by its period changes andlight amplitude, and implied luminosity. It becomes an importantCepheid calibrator lying near the centre of the instability strip.
It may seem unusual that many potential Cepheid calibrators liein cluster coronae rather than in cluster nuclear regions (Turner
1985), but a possible explanation relates to two dynamical linesof evidence. First, massive cluster members lie preferentially inouter regions of clusters (Burki 1978), possibly because of howprotocluster interstellar clouds fragment into protostars. Secondly,as indicated by colour–magnitude diagrams for NGC 654 (Stone1980) and other young clusters (Turner 1996b), cluster nuclearregions tend to be dominated by rapidly rotating stars, possibly theresult of merged binary systems, and other close binaries, in whichcase potential Cepheid progenitors are less likely to evolve to thedimensions typical of pulsating variables because of restrictions ontheir dimensions engendered by potential physical companions. The
Figure 9. The differences between observed (O) and computed (C) timesof light maximum for CG Cas, computed in units of pulsation phase. Theupper diagram shows the actual O−C variations with their uncertainties,the lower diagram the residuals from the calculated parabolic evolutionarytrend.
case of CG Cas in Berkeley 58 appears to be yet another exampleof the effect.
AC K N OW L E D G M E N T S
The present study was supported by research funding awardedthrough the Natural Sciences and Engineering Research Councilof Canada (NSERC), through the Small Research Grants programof the American Astronomical Society, through the Russian Foun-dation for Basic Research (RFBR), and through the program ofSupport for Leading Scientific Schools of Russia. We are indebtedto Ron Lyons for scanning the radial velocity plates used in thisstudy, and to the director of Harvard College Observatory for ac-cess to the plate stacks.
REFERENCES
Bahner K., Hiltner W. A., Kraft R. P., 1962, ApJS, 6, 319Berdnikov L. N., 1986, Perem. Zvezdy, 22, 369Berdnikov L. N., 1992a, Astron. & Astrophys. Trans., 2, 107Berdnikov L. N., 1992b, Astron. & Astrophys. Trans., 2, 157Berdnikov L. N., 1992c, Astron. Lett., 18, 130Berdnikov L. N., 1993, Perem. Zvezdy, 23, 80Berdnikov L. N., 2007, http://www.sai.msu.ru/groups/cluster/cep/pheBurki G., 1978, A&AS, 62, 159Chekanikhina O. A., 1982, Bull. Astrophys. Inst. AN Tadj. SSR, 71, 25Cutri R. M. et al., 2003, NASA/IPAC Infrared Science Archive, The IRSA
2MASS All-Sky Point Source Catalog of Point SourcesEfremov Y. N., 1964a, Perem. Zvezdy, 15, 242Efremov Y. N., 1964b, Astr. Tsirk., 292, 3Erleksova G. E., 1961, Bull. Astrophys. Inst. AN Tadj. SSR, 30, 28Fernie J. D., 1963, AJ, 68, 780Frolov V. N., 1974, Astr. Tsirk., 848, 1Frolov V. N., 1977, Izv. Glavnaia Astr. Obs. Pulkovo, 195, 80Frolov V. N., 1979, Izv. Glavnaia Astr. Obs. Pulkovo, 196, 69Gorynya N. A., Samus N. N., Sackhov M. E., Rastorguev A. S., Glushkova
E. V., Antipin S. V., 1998, Pis’ma Azh, 24, 939Henden A. A., 1996, AJ, 112, 2757
Irwin J. B., 1955, Mon. Not. Astron. Soc. South Africa, 14, 38Irwin J. B., 1958, AJ, 63, 197Joy A. E., 1937, ApJ, 86, 363Kholopov P. N., 1956, Perem. Zvezdy, 11, 325Kholopov P. N., 1969, SvA, 12, 625Kovalenko V. M., 1968, Astr. Tsirk., 473, 2Kraft R. P., 1957, ApJ, 126, 225Laney C. D., Caldwell J. A. R., 2007, MNRAS, 377, 147Lange G., 1933, Bull Leningrad Univ. Astr. Obs., 2, 9Metzger M. R., Caldwell J. A. R., McCarthy J. K., Schechter P. L., 1991,
ApJS, 76, 803Meynet G., Mermilliod J.-C., Maeder A., 1993, A&AS, 98, 477Neckel Th., Klare G., 1980, A&AS, 42, 251Oosterhoff P. Th., 1960, BAN, 15, 199Phelps R. L., Janes K. A., 1994, ApJS, 90, 31Romano G., 1959, Pub. Osserv. Astr. Padova, 116, 3Sandage A., 1958, ApJ, 128, 150Setteducati A. F., Weaver H. F., 1962, Newly Found Star Clusters, Radio
Astron. Lab. Univ. California, BerkeleyStone R. C., 1980, PASP, 92, 426Takala J. M., 1988, MSc thesis, Saint Mary’s Univ.Tifft W. G., 1959, ApJ, 129, 241Turner D. G., 1976a, AJ, 81, 97Turner D. G., 1976b, AJ, 81, 1125Turner D. G., 1979, PASP, 91, 642Turner D. G., 1985, in Madore B. F., ed., IAU Colloq. 82, Cepheids: Theory
and Observations. Cambridge Univ. Press, Cambridge, p. 209Turner D. G., 1989, AJ, 98, 2300Turner D. G., 1992, AJ, 104, 1865Turner D. G., 1993, A&AS, 97, 755Turner D. G., 1996a, in Milone E. F., Mermilliod J.-C., eds., ASP Conf.
Series Vol. 90, The Origins, Evolution and Destinies of Binary Stars inClusters. Astron. Soc. Pac., San Francisco, p. 382
Turner D. G., 1996b, in Milone E. F., Mermilliod J.-C., eds, ASP Conf.Series Vol. 90, The Origins, Evolution and Destinies of Binary Stars inClusters. Astron. Soc. Pac., San Francisco, p. 443
Turner D. G., 1996c, JRASC, 90, 82Turner D. G., Drilling J. S., 1984, PASP, 96, 292Turner D. G., Welch G. A., 1989, PASP, 101, 1038Turner D. G., Burke J. F., 2002, AJ, 124, 2931Turner D. G., Forbes D., Pedreros M., 1992, AJ, 104, 1132Turner D. G., Mandushev G. I., Forbes D., 1994, AJ, 107, 1796Turner D. G., Abdel-Sabour Abdel-Latif M., Berdnikov L. N., 2006a, PASP,
118, 410Turner D. G., Usenko I. A., Kovtyukh V. V., 2006b, Observatory, 126, 207van den Bergh S., 1957, ApJ, 126, 323Wozniak P. R. et al., 2004, AJ, 127, 2436Zonn W., Semeniuk I., 1959, Acta Astr., 9, 141
SUPPLEMENTA RY MATERI AL
The following supplementary material is available for this article:
Table 3. Photographic UBV data for stars in Berkeley 58.
Table 4. CCD UBV data for stars in the nucleus of Berkeley 58.
This material is available as part of the online articlefrom: http://www.blackwell-synergy.com/doi/abs/10.1111/j.1365-2966.2007.13413.x(this link will take you to the article abstract).
Please note: Blackwell Publishing are not responsible for the con-tent or functionality of any supplementary materials supplied bythe authors. Any queries (other than missing material) should bedirected to the corresponding author for the article.
This paper has been typeset from a TEX/LATEX file prepared by the author.