The Comings and Goings of Be Stars

Presented at the meeting of the Working Group on Active B Stars during the 26th IAU General Assembly in Prague, Czech Republic on 2006 August 18

M. Virginia McSwain^1,2

¹ Department of Astronomy, Yale University, New Haven, CT 06520-8101; mcswain@astro.yale.edu
²NSF Astronomy and Astrophysics Postdoctoral Fellow

Received: 2006 September 1; Revised: 2007 March 9; Accepted: 2007 March 12.

Abstract: I present preliminary findings from a spectroscopic monitoring program of Be stars in 19 open clusters. From a 3- or 4-year time baseline of photometric and spectroscopic data, I have identified 26 Be stars that have undergone disk outbursts or whose disks have disappeared. The rate of disk variability does not seem to be constant among clusters, and I have begun an investigation of the stellar and cluster properties that may affect their long term variability.

In 2002, we performed a photometric search for Be stars in 55 open clusters using the CTIO 0.9-m telescope with Strömgren b, y, and narrow-band Hα filters (McSwain & Gies 2005a, 2005b). Among the 48 clusters with reliable results, this investigation resulted in 52 definite Be star detections, many previously unidentified as Be stars, and an additional 129 possible Be star candidates. Since completing that study, we have been performing a spectroscopic follow-up to confirm the candidate Be stars. We obtained spectra of 20 stars in NGC 3766 during 2003 March with the CTIO 1.5-m telescope and Cassegrain spectrograph. Additionally, we observed NGC 3766 and 20 other clusters with the CTIO 4-m telescope and Hydra spectrograph during 2005 February and 2006 May. The results from those data have increased the total number of definite Be stars to 83, with 47 remaining candidates. Based on these updates, the mean Be star fraction in open clusters is between 3.3--5.1% of the B-type population.

The 3- and 4-year time baseline separating our photometry and spectroscopy has also revealed another intriguing result: we have identified 26 Be stars that have experienced a disk appearance or disappearance during this time. Three examples from NGC 3766 are shown in Figure 1. Although we do not have the same time sampling for every cluster, our preliminary findings suggest that the fraction of long-term variable Be stars is not constant among different clusters. Thus we have begun an investigation of the stellar and cluster properties that might influence the disk variability.

While the long term variability of Be stars has often been noted in the literature (e.g. Slettebak 1988), only one study has attempted long term monitoring of a large sample of Be stars to quantify their variability. Hubert & Floquet (1998) investigated 273 bright Be stars that were observed by Hipparcos between August 1989 to August 1993. They found that early type Be stars exhibit a very high degree of variability while most late Be stars maintained a constant magnitude during the duration of the Hipparcos mission, suggesting that the variability is temperature dependent. However, we compare the absolute magnitudes and ages of variable and non-variable Be stars in Figure 2. Long term variable Be stars appear distributed across a large range of both M_y and ages, which indicates that the disk variability may not be dependent on temperature or evolutionary status.

Metallicity may be another important factor in the disk state. Meilland et al. (2006) discuss observational characteristics that may determine whether Be star disks form by successive outbursts from the central star or by a more continuous change in mass loss. In this second scenario, changes in the opacity may result in disk state transitions, and we should observe a variability dependence on metallicity. Our preliminary results indicate that the cluster fraction of disk variable Be stars increases toward the Galactic center, where the metallicity is greater, suggesting that metallicity may be a factor in the origins of the disks.

We have recently obtained high quality, blue optical spectra of three open clusters to measure the physical parameters of the normal B-type and Be populations. Hubert & Floquet (1998) propose that disk outbursts may be also related to V sin i, and we will investigate this relationship in our new spectra. We will also obtain accurate measurements of T_eff, log g, and metallicity to investigate other possible sources of disk variability.

References

Hubert, A. M. & Floquet, M. 1998, A&A, 335, 565
McSwain, M. V., & Gies, D. R. 2005a, ApJ, 622, 1052
McSwain, M. V., & Gies, D. R. 2005b, ApJS, 161, 118
Meilland, A., Stee, Ph., Zorec, J., & Kanaan, S. 2006, A&A, 455, 953 Slettebak, A. 1988, PASP, 100, 770

Acknowledgments

M. V. M. is supported by an NSF Astronomy and Astrophysics Postdoctoral Fellowship under award AST-0401460.