Self-Correlation Studies of the Short-Term Photometric Variability of Be Stars

Presented at the meeting of the Working Group on Active B Stars during the 25th IAU General Assembly in Sydney, Australia on 2003 July 16

John R. Percy

Department of Astronomy & Astrophysics, University of Toronto, Toronto ON Canada M5S 3H8
email: jpercy@utm.utoronto.ca

Received: 2003 September 15; Accepted: 2003 September 17.

The nature and cause of the short-term (0.5 to 2 days) variability of Be stars are still controversial, though Rivinius, Baade & Stefl (2003) have made a very strong case for non-radial pulsation. The Hipparcos archive of epoch photometry can provide useful information, since many of the several hundred Be stars observed by Hipparcos were measured on time scales of tens of minutes for intervals of up to several days (but typically 0.2 to one day).

We have recently applied self-correlation analysis — a simple form of time-series analysis — to Hipparcos photometry of bright Be stars (Percy et al. 2002, 2003). Self-correlation analysis determines the cycle-to-cycle behaviour of the star, averaged over all the data. It is a useful adjunct to Fourier analysis, especially for Hipparcos photometry, in which there are gaps of typically 25 days between the groups of more closely-spaced measurements. Hubert & Floquet (1998) carried out a comprehensive time-series analysis of Hipparcos photometry of Be stars using Fourier and CLEAN analysis.

Percy et al. (2002) demonstrated the utility of self-correlation analysis of Hipparcos photometry of Be stars. They determined the average amplitude and characteristic time scale of the short-term variability of 82 bright Be stars which were known or suspected to have such variability. They also presented tantalizing evidence for a relationship between the amplitude of short-term variability, and the phase in the "outburst" of one Be star —  Cyg.

Percy et al. (2003) have used the same technique to analyze all of the Be stars observed by Hipparcos, but not analyzed by Percy et al. (2002). Over 100 stars had photometry which provided some information about the short-term variability. For the other stars, the time distribution of the measurements did not provide any useful information.

The principal results are as follows:

• The incidence and amplitude of short-term variability are greater among early-B stars than among late B stars; this result is consistent with the results of studies using other datasets or approaches (e.g. incidence of absorption-line profile variation: Baade 1989; general photometric variability: Stagg 1987).

• We found no evidence for periods of 0.1-0.3 day which might be due to radial ( Cephei type) pulsation in these stars.

• We identified the following new short-term variable Be stars: HD 7636 (V764 Cas), HD 11606 (V777 Cas), HD 13661 (V549 Per), HD 34921, HD 36408, HD 40978, HD 58343 (FW CMa), HD 63460 (o Pup), HD 88195 (17 Sex), HD 89353 (AG Ant, HR 4049), HD 129954 (CO Cir), HD 158220 (V862 Ara), HD 173219 (V947 Sct), and HD 187567 (V1339 Aql).

• We confirmed the following known short-term variable Be stars, using an independent form of analysis: HD 52918 (19 Mon, a  Cep star), HD 105382 (V863 Cen), HD 137387 (-1 Aps), HD 157832 (V750 Ara), and HD 163868 (V3984 Sgr).

Perhaps the most interesting tentative discoveries are short-term variability in the Vega-type star 17 Sex (SpT A1V), and in the prototype post-AGB star HR 4049 (SpT B9.5Ib-II). We also found an interesting form of self-correlation diagram in FW CMa, V3984 Sgr, and QR Vul (Percy et al. 2002). In these three stars, there is activity on time scales of less than four days (i.e. the self-correlation diagram rises significantly on this time scale), but this is part of the "long-term" (weeks to months) activity and variability which is associated with the appearance and disappearance of the gas disc.

We have now exhausted the supply of Hipparcos photometry of Be stars for self-correlation analysis. There are other photometric datasets, of course, such as our long-term monitoring program (Percy & Bakos 2001) but, for self-correlation analysis, there must be sets of measurements made over intervals of hours to a few days. Our next project, however, is to see whether a self-correlation analysis can provide a useful "variability profile" of the longer-term (weeks) variability of active Be stars such as CX Dra.

Acknowledgements. My co-authors (Percy et al. 2002, 2003) were undergraduate research students. We were supported by a research grant from NSERC Canada, and by the Ontario Work-Study Program.

References:

Baade, D. 1989, A&A, 222, 200

Hubert, A.M. & Floquet, M. 1998, A&A, 335, 565

Percy, J.R. & Bakos, A.G. 2001, PASP, 113, 748

Percy, J.R., Hosick, J., Kincaide, H. & Pang, C. 2002, PASP, 114, 551

Percy, J.R., Harlow, C. & Wu, A. 2003, PASP, submitted

Rivinius, Th., Baade, D. & Stefl, S. 2003, A&A, in press

Stagg, C.R. 1987, MNRAS, 227, 213

back to Meeting Summary