Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 333, 125-140 (1998)

Previous Section Next Section Title Page Table of Contents

8. Conclusions

8.1. Summary of the results for µ Cen

The circumstellar variability largely consists of discrete events which can be characterized as outbursts during which matter is ejected by the star and possibly even (partly) transferred to the disk. The relative contributions of outbursts and stellar wind to the total mass of the disk are currently unconstrained. In any event, µ Cen is building up a new persistent disk which was not detected between 1977 and 1989.

An outburst cycle appears to be composed of a sequence of relative quiescence, a short pre-outburst phase (precursor), the outburst, and an extended relaxation phase. However, outbursts may differ substantially not only in amplitude and duration, but also in the relative prominence of the above noted constituents. This depends on the absolute strength of the disk emission already present, i.e., on the mass and optical thickness of the disk. Especially the burst and the relaxation phases can be missing completely in outbursts during which the disk emission is strong. By contrast, the precursor phase is the stronger the higher the disk emission is.

Even at low emission levels a new outburst may appear before the relaxation to the pre-outburst level is completed. Also, relaxation may proceed more slowly with increasing disk mass, so that relative quiescence as defined in this paper may eventually never be reached any more once a strong disk has been built up.

The relative quiescence is a phase of merely minor variability in the hydrogen lines, the smallest separation between double emission peaks of all phases, and the absence of Fe ii emission lines. As a precursor to an outburst, a sudden decrease in emission peak height occurs. At the same time, broad emission wings appear. The outburst is defined by a rapid increase of the emission strength, rapid cyclic [FORMULA] -variability, high velocity absorptions, and a rapid increase of the emission peak separation. In major outbursts, also Fe ii emission lines appear. The relaxation phase can be described briefly as the development from outburst into quiescence phase.

For the first time in a Be star the time scale of rapid cyclical [FORMULA] -variability was found to be quite different from the dominant period of the stellar line-profile variability of 0.5 day (cf. Paper II). However, this [FORMULA] variability seems to be linked to the outbursts. In fact, the differences of [FORMULA] 5% of the former between different outbursts and the mean value of about 0.6 day are consistent with an ejected cloud of gas which orbits the star a few times at a small radius until it is dispersed or merges with the disk or falls back to the star or some combination of these processes.

Especially at times of enhanced [FORMULA] variability, short-lived discrete absorption components of the stronger He i lines may appear which are blueshifted by about 1.5 times the stellar v sin i. They are, therefore, unlikely to arise from the photosphere. The opposite is true of other sharp absorption features which appear at both positive and negative velocities but are restricted to about [FORMULA] 0.7 v sin i. Because of their probable photospheric origin and association with the photospheric line-profile variability, they are described only in Paper II.

Some of the generalizations that have been attempted in this study of the outbursts of µ Cen may also apply to other Be stars. However, with no second case known in nearly as much detail, this is arbitrarily uncertain. Especially the geometry of the disk and its orientation with respect to the observer may lead to some modifications.

8.2. Comparison with other Be stars

Outbursts do not seem to have been observed at the same level of detail in other classical Be stars. But those (less complete) observations that are available suggest a number of similarities which are worthwhile recalling:

  1. Rapid de- and/or increases in equivalent width of emission lines have been seen in stars like FY CMa (Peters, 1988), [FORMULA] Eri (Smith, 1989), [FORMULA] Ori (Guinan & Hayes, 1984; Hayes & Guinan, 1984), or HD 76 534 (Oudmaijer & Drew, 1997).
  2. Enhanced wings of the H [FORMULA] emission during an outburst have also been detected in [FORMULA] Ori (Baade, 1986). Also the anticorrelation between [FORMULA] and Fe ii [FORMULA] 5317 that has been observed in a few profiles of HR2825 and HR5223 by Hanuschik et al. 1996 (their Figs. 34d and 56d) would fit into the description of an outburst phase. While the [FORMULA] peak height is reduced in these spectra, Fe ii [FORMULA] 5317 and the [FORMULA] wings are enhanced. Also the peculiar shape of Br [FORMULA] in 59 Cyg observed by Waters & Marlborough (1994) is remarkably similar to the Pa15 profile of µ Cen during an outburst. But the example of [FORMULA] Cas (McDavid & Gies, 1988) may be indicative of this not being a uniform behaviour.
  3. Rapid cyclic, or possibly even periodic, [FORMULA] variations have been reported for various other Be stars, e.g. 28 CMa (1.37 days, Baade 1982), [FORMULA] Eri (0.701 days, Bolton 1982; 0.714 days, Smith 1989) and EW Lac (0.7 days, Pavlovski & Schneider 1989). However, in these cases the [FORMULA] time scales seem to be indistinguishable from the periods of the photometric or low-order photospheric line profile variability while our observations of µ Cen exhibit a significant difference of the order of 20%.

It would, therefore, appear not only a necessary but perhaps even reasonably promising undertaking to search other Be stars for a similar behaviour so that it becomes clearer to what extent µ Cen is a representative case.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1998

Online publication: April 15, 1998