Solar-like oscillations in cluster stars
D. Stello, S. Basu, T. R. Bedding, K. Brogaard, H. Bruntt, W. J. Chaplin, J. Christensen-Dalsgaard, P. Demarque, Y. P. Elsworth, R.A. García, R. L. Gilliland, S. Hekker, D. Huber, C. Karoff, H. Kjeldsen, Y. Lebreton, S. Mathur, S. Meibom, J. Molenda-Żakowicz, A. Noels, I. W. Roxburgh, V. S. Aguirre, C. Sterken, R. Szabó
aa r X i v : . [ a s t r o - ph . S R ] J un Astron. Nachr. / AN , No. 88, 789 – 793 (2006) /
DOI please set DOI!
Solar-like oscillations in cluster stars ⋆ D. Stello ,⋆⋆ , S. Basu , T. R. Bedding , K. Brogaard , H. Bruntt , W. J. Chaplin , J. Christensen-Dalsgaard , P. Demarque , Y. P. Elsworth , R. A. Garc´ıa , R. L. Gilliland , S. Hekker , D. Huber ,C. Karoff , H. Kjeldsen , Y. Lebreton , S. Mathur , S. Meibom , J. Molenda- ˙Zakowicz , A. Noels ,I. W. Roxburgh , V. S. Aguirre , C. Sterken , and R. Szab´o Sydney Institute for Astronomy (SIfA), School of Physics, University of Sydney, NSW 2006, Australia Department of Astronomy, Yale University, P.O. Box 208101, New Haven, CT 06520-8101 Department of Physics and Astronomy, Aarhus University, 8000 Aarhus C, Denmark LESIA, CNRS, Universit´e Pierre et Marie Curie, Universit´e Denis Diderot, Observatoire de Paris, 92195 Meudon,France School of Physics and Astronomy, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK Laboratoire AIM, CEA/DSM-CNRS, Universit´e Paris 7 Diderot, IRFU/SAp, Centre de Saclay, 91191, Gif-sur-Yvette,France Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, Maryland 21218, USA GEPI, Observatoire de Paris, CNRS, Universit´e Paris Diderot, 5 Place Jules Janssen, 92195 Meudon, France High Altitude Observatory, NCAR, P.O. Box 3000, Boulder, CO 80307, USA Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA, 02138, USA Instytut Astronomiczny Uniwersytetu Wrocławskiego, ul. Kopernika 11, 51-622 Wrocław, Poland Institut d’Astrophysique et de G´eophysique de l’Universit´e de Li`ege, 17 All´ee du 6 Aoˆut, B-4000 Li`ege, Belgium Queen Mary University of London, Mile End Road, London E1 4NS, UK Max Planck Institute for Astrophysics, Karl Schwarzschild Str. 1, Garching bei M¨unchen, D-85741, Germany Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium Konkoly Observatory, H-1525 Budapest, P.O. Box 67, HungaryReceived 25 May 2010, accepted 24 June 2010Published online later
Key words stars: fundamental parameters — stars: oscillations — stars: interiors — techniques: photometric — openclusters and associations: individual (NGC 6819)We present a brief overview of the history of attempts to obtain a clear detection of solar-like oscillations in cluster stars,and discuss the results on the first clear detection, which was made by the Kepler Asteroseismic Science Consortium(KASC) Working Group 2. c (cid:13) Star clusters are extremely important in stellar astrophysics.Most stars form in open clusters, many of which disperseinto the diversity of field stars in the interstellar medium.Understanding the formation and evolution of cluster starsis therefore important for achieving a comprehensive the-ory of stellar evolution. Stars in a cluster are thought tobe formed coevally, from the same interstellar cloud of gasand dust. Each cluster member is therefore expected to havesome properties in common (age, composition, distance),which strengthens our ability to constrain our stellar mod-els when tested against an ensemble of cluster stars, espe-cially for asteroseismic analyses (Gough & Novotny 1993).Asteroseismology has the capability to probe the interior ofstars and hence help us understand the fundamental physi- ⋆ Data from
Kepler ⋆⋆ Corresponding author: e-mail: [email protected] cal process that govern stellar structure and evolution (e.g.,Christensen-Dalsgaard 2002). In particular, the detection ofsolar-like oscillations provide many modes, which each car-rying unique information about the stellar interior. Stars thatpotentially exhibit solar-like oscillations, covering most starsthat we see, are cooler than the red edge of the classical in-stability strip, and have a convection zone near the surface(necessary for the excitation of the modes). Solar-like os-cillations are reasonably well described by current theory,giving us some confidence that we can use them as tools tounderstand stellar physics, and hopefully also to learn moreabout the more subtle aspects of the oscillations themselves.Combining asteroseismic analysis of solar-like oscillationswith the study of cluster stars has therefore been a long-sought goal. c (cid:13)
90 D. Stello et al.: Solar-like oscillations in cluster stars
Fig. 1
Amplitude spectrum (high-pass filtered) of one ofthe stars targeted by Gilliland et al. (1993). The horizontalline marks the expected location of the oscillations.
Fig. 2
Amplitude spectrum of red giant star observed byGilliland et al. (1993). Horizontal line marks the expectedlocation of the oscillations.
Kepler is certainly not the first attempt to detect solar-likeoscillations in cluster stars. A quick (and hence incomplete)perusal of the history of previous attempts to detect solar-like oscillations in open and globular clusters shows thatseveral attempts were made to detect oscillations since theearly 1990s. Among the most ambitious was that of Gilliland et al.(1993), who used 4-m class telescopes to target the starsin the open cluster M67 at the cluster turn-off in a multi-site campaign that lasted one week. While an impressivelylow noise level was obtained, the data did not reveal theclear detection of stellar oscillations (Figure 1). However, ared giant star that happened to be in the field did show in-triguing evidence of excess power in the expected frequencyrange (Figure 2). Unfortunately, the length of the time se-ries did not allow individual modes to be resolved for suchan evolved star with much smaller frequency separations be-tween modes. A clear detection remained elusive, as oscilla-tions could not be distinguished from the rising backgroundtowards low frequency.Inspired by Gilliand’s results, Stello et al. (2007) tar-geted specifically the red giants in M67 during a 6-weeklong multi-site campaign of 1–2m class telescopes. Strongevidence for excess power was found in a number of stars,but no unambiguous detection of the solar-like pattern ofequally spaced modes was claimed by the authors (Figure 3).In parallel, several attempts to detect oscillations in glob-ular clusters were carried out. From the ground, Frandsen et al.(2007) aimed at the red giants in M4, which delivered lowerlimits on amplitudes, indicating that the low metallicity ofM4 could have the effect of lowering the oscillation ampli-
Fig. 3
Power spectra of three red giant stars observed byStello et al. (2007). Black arrow marks the location of theoscillations expected from scaling the solar value.tudes. Again, detection was hindered by long-term stabilitynot being high enough and varying data quality resulting instrong aliasing in the weighted amplitude spectra. Slightlymore successful were the efforts using the
Hubble SpaceTelescope by Edmonds & Gilliland (1996); Stello & Gilliland(2009). In the former study, clear variation was found in alarge number of red giants in 47 Tuc, but the low frequencyresolution provided by the 40-hour time series did not al-low the authors to establish this as solar-like oscillations.The later study was aimed at the red giants in the extremelymetal poor NGC 6397, using archival data originally ob-tained to detect the cluster’s faint white dwarf population.The far from ideal data of highly saturated photometry ofthe red giants meant that only one star showed good evi-dence for oscillations, with excess power at the right fre-quency range and amplitude. Despite the 27-day long timeseries, this fell just short for an unambiguous detection ofequally spaced frequencies in this highly evolved asymp-totic giant branch star.The main conclusion from these previous efforts is thatdedicated space-based missions are required to achieve theultra-high precision photometry and long-term stability inorder to detect solar-like oscillations in clusters with suchaccuracy that they will be useful for asteroseismic analysis.We note that in addition to the previous marginal detec-tions, these campaigns resulted in firm detection of oscilla-tions in a number of classical pulsators that exhibit muchlarge amplitude than solar-like oscillations (see e.g. Bruntt2007 and references therein). c (cid:13) stron. Nachr. / AN (2006) 791 Fig. 4
HR-diagram of NGC 6819. Empty symbols markthose where a detection of solar-like oscillations was re-ported by Stello et al. (2010).
Kepler
Kepler has a unique capability to overcome the shortcom-ings that have limited previous efforts aimed at stellar clus-ters. Both quality and quantity of the
Kepler data outshinethat of early explorations by several orders of magnitude,and it will undoubtedly be the front runner for cluster seis-mology in the next 5–10 years.As reported by Stello et al. (2010), the first month of
Ke-pler data already revealed clear detection of solar-like oscil-lations in a large sample of red giant stars in the open clus-ter NGC 6819 (see also Gilliland et al. 2010). Based on thespacecrafts so called long-cadence mode, which provides atime averaged exposure every 29.4 minutes, detection wasreported in 47 red giant stars that range almost from thebottom to the tip of the red giant branch (Figure 4). Wesaw periodicity in the light curves that span about a fac-tor of 100, corresponding to a factor of ∼ in radius. Twosample light curves are shown in Figure 5. Power spectraof the stars marked with numbers in Figure 4 are shownin Figure 6. Panels are sorted according to apparent mag-nitude (brightest at the top), which for a cluster is indica-tive of luminosity. One noticeable result is that not all starswith high membership probability from radial velocity sur-veys (see Hole et al. 2009) follow the expected monotonictrend of increasing frequency of the oscillations (and de-creasing amplitude) for decreasing luminosity. We indicatethe expected frequency location with an arrow for stars thatseem to behave strangely compared to the classical scalingrelations for the amplitude and the frequency of maximumpower (e.g. Kjeldsen & Bedding 1995). Possible explana- Fig. 5
Kepler time series for two red giants in NGC 6819.Numbers refer to the numbering in Figure 4. Note the differ-ent time scale of the variation. Photometry and isochrone isof Hole et al. (2009) and Marigo et al. (2008), respectively.
Table 1
Open clusters in
Kepler field
Cluster Age [Fe/H] M turnoff
Gyr M ⊙ NGC 6866 ∼ ∼ − ∼ ∼ ∼ − ∼ ∼ ∼ − ∼ ∼ ∼ + ∼ Values are from Grundahl et al. (2008) (NGC 6791),Hole et al. (2009) (NGC 6819), Loktin & Matkin (1994) (NGC 6866)and unpublished work by Meibom. tions for this behaviour are that these “odd” stars are notmembers, or that they have unusual evolution histories.Stello et al. (2010) were further able to measure the am-plitudes of the modes using the method by Kjeldsen et al.(2009), assuming the relative amplitudes of the modes ofdifferent spherical degree was the same as for the Sun. Fromthis we could test the
L/M scaling relation (Kjeldsen & Bedding1995; Samadi et al. 2007), and found that ( L/M ) . /T provided the best match to the data.For further details on what is reported here, we refer tothe source paper of Stello et al. (2010). There are four open clusters in
Kepler’s field of view. Theyspan a range in metallicity and age, which brackets the solarvalues, and are therefore ideal for testing our current modelsof stellar evolution (Table 1). c (cid:13)
92 D. Stello et al.: Solar-like oscillations in cluster stars
Fig. 6
Power spectra of 11 stars marked in Figure 4,which are representative for the entire sample. ‘AM’ indi-cates that the star is an asteroseismic member (i.e. observa-tion agrees with scaling relations). Dashed lines show themeasured large frequency separation. For stars where thelarge separation could not be determined (no dashed lines),we localised the power excess from the hump of power inthe smoothed power spectrum (solid black curve). The ar-rows indicate the expected location of the excess power forstars where observations do not agree with expectation.In Figure 7 we show log( g ) vs T eff for a representativesample of the stars in Kepler’s field of view together withthe representative isochrones for the four open clusters thatare targets in our future asteroseismic analyses.For NGC 6819 we expect to achieve a signal-to-noiselevel for the turn-off stars that after 3.5 years of data matcheswhat we see in the bottom panels of Figure 6. This willprovide detection in up to 100 stars ranging stellar evolu-tion from the main sequence F stars to the asymptotic gi-ant branch including M giants, as well as a number of bluestragglers. This will potentially provide unprecedented testsof state-of-the-art stellar evolution models.In NGC 6791 we already see evidence for power in thered giants, and expect firm detections for all stars on thishighly populated red giant branch, with unique potential fortesting intrinsic variation among practically identical stars.The two younger clusters NGC 6811 and NGC 6866are less populated but provide the opportunity to investigate
Fig. 7 log( g ) vs T eff for stars in Kepler’s field of view.We represent the four open clusters by suitable isochrones.The order in which we have plotted the cluster names cor-responds to their turn-off stars, with NGC 6866 having thehottest (heaviest) turn-off stars and NGC 6791 the coolest(lightest). The dashed line indicates the red edge of the clas-sical instability strip.classical pulsators in great detail. NGC 6811 also containsa few He-core burning red giants.The combination of results from all four clusters promisesgreat prospects for testing asteroseismic scaling relations ondistinct stellar populations that span a large range in stellarage and brackets the solar metallicity.
Acknowledgements.
Funding of the Discovery mission is providedby NASA’s Science Mission Directorate. The authors thank the en-tire
Kepler team without whom this investigation would not havebeen possible. The authors also thank all funding councils andagencies that have supported the activities for Working Group 2of the KASC. In particular, DS would like to thank HELAS forsupport to attend the HELAS IV meeting in Lanzarote.
References
Bruntt, H. 2007, Communications in Asteroseismology, 150, 326Christensen-Dalsgaard, J. 2002, Reviews of Modern Physics, 74,1073Edmonds, P. D., & Gilliland, R. L. 1996, ApJ, 464, L157Frandsen, S., et al. 2007, A&A, 475, 991Gilliland, R. L., et al. 1993, AJ, 106, 2441—. 2010, PASP, 122, 131Gough, D. O., & Novotny, E. 1993, in ASP Conf. Ser. 42: GONG1992. Seismic Investigation of the Sun and Stars, ed. T. M.Brown, 355Grundahl, F., Clausen, J. V., Hardis, S., & Frandsen, S. 2008,A&A, 492, 171Hole, K. T., Geller, A. M., Mathieu, R. D., Platais, I., Meibom, S.,& Latham, D. W. 2009, AJ, 138, 159Kjeldsen, H., & Bedding, T. R. 1995, A&A, 293, 87Kjeldsen, H., Bedding, T. R., & Christensen-Dalsgaard, J. 2009, inIAU Symposium, Vol. 253, IAU Symposium, 309–317Loktin, A. V., & Matkin, N. V. 1994, Astronomical and Astrophys-ical Transactions, 4, 153Marigo, P., Girardi, L., Bressan, A., Groenewegen, M. A. T., Silva,L., & Granato, G. L. 2008, A&A, 482, 883 c (cid:13) stron. Nachr. / AN (2006) 793Samadi, R., Georgobiani, D., Trampedach, R., Goupil, M. J., Stein,R. F., & Nordlund, ˚A. 2007, A&A, 463, 297Stello, D., & Gilliland, R. L. 2009, ApJ, 700, 949Stello, D., et al. 2007, MNRAS, 377, 584—. 2010, ApJ, 713, L182 c (cid:13)(cid:13)