Astron. Astrophys. 356, 913-928 (2000)

3. Long-term variability

For the majority of Be stars, their long-term variations are usually the most pronounced ones. Their further studies and especially the investigation of mutual possible correlation between spectral, photometric and polarimetric changes may be essential for explaining the Be phenomenon. Here, we present data which may shed new light on the long-term variations of H emission, brightness and colour of 60 Cyg and may also have important consequences for the ultimate understanding of the nature and character of rapid line-profile changes.

3.1. H line

Prior to this study, records of the behaviour of the H line of 60 Cyg were rather scarce. The only cases when quantitative measurements of H line strength were possible in published profiles are summarized in Table 7. The time variability of H is shown schematically in Fig. 1. This is because the main body of previous records rests on estimates from plates presented in Atlas or on our interpretation of the descriptions found in the literature. Some "normalization" with respect to more recent records is based on the comparison of a photographic spectrum of 60 Cyg from Atlas, taken in July 1976, with the published electronic profile of H (Fontaine et al. 1982) taken one month later.

Fig. 1. Schematic representation of H versus time. 1 denotes strong emission, 0.5 an intermediate state, and 0 a "pure" absorption.

Table 7. Published data on the H line profiles of 60 Cyg.
References in column "Ref.": 1... Fontaine et al. (1982); 2... Andrillat (1983); 3... Doazan et al. (1991);

The existing records including our new observations seem to indicate that the maximum strength of the H emission never exceeds a peak intensity of about 2.0 of the continuum level (see Fig. 20). In phase of an intermediate strength of the H emission a relatively narrow absorption is flanked by faint emission components (see Fig. 20, spectrum from JD 49555.5).

Fig. 1 shows that 60 Cyg underwent two long-lasting H emission episodes between 1953 and 1990, each of them some ten years long. However, a denser coverage in 1990's (typically only one or two spectra per year are available in Atlas) shows that the evolution of the emission episodes of 60 Cyg may be more complicated. Andrillat & Houziaux (1991) reported that a new emission episode in H started after August 1991; however, their conclusion is based only on the behaviour of Paschen lines and on a simultaneous observation of H and Paschen lines in 1965 when H was in emission while a shallow absorption was present in the Paschen lines (Andrillat & Houziaux 1967). Andrillat & Houziaux (1991) recorded a change of the profile of the P7 line from absorption to emission between August and the end of October 1991. This may imply that at least by the end of October 1991 H of 60 Cyg was in emission. However, ten months later "pure" H absorption was recorded at Ondejov - cf. spectrogram taken on JD 48840.5 in Fig. 20.

Examples of different appearance of the H profile after 1992 is depicted in more detail in Fig. 20. The comparison with the synthetic spectrum suggests that even in the case of "pure" absorption some traces of H emission were present.

The upper panels of Fig. 2 below show the time variation of the V/R ratio of the blue and red emission peaks and of the H peak intensity. It is seen that the emission has been more or less steadily increasing since about JD 2448800, with some cyclic variations overlapping. One can also note a good agreement between the data from all three observatories. On the other hand, it appears that the V/R ratio has varied more rapidly, all the time with about the same amplitude, independently of the changing strength of the H emission.

Fig. 2. The plots of the H emission intensity, ratio of the double peaks of the H emission and mean V and of 60 Cyg, averaged over 150 d, vs. time. Different photometric stations are denoted by different symbols as follows: diamonds for Hvar, empty triangles for Xinglong, open circles for Toronto, filled circles for Hipparcos, plus signs for OHP, and filled square for SPM.

3.2. Other lines

As can be inferred from Tables 1 and 2, the available observations are only suitable for long-term study of H and He I 6678 Å in the spectrum of 60 Cyg. The data for He I 4471 Å and H are very scanty. In Figs. 21, 22, and 23 we present profiles of H and He I 4471 Å and 6678 from phases of "pure" absorption and "full" emission in H. The most variable is the He I 6678 Å profile. The development of the emission causes decrease of the line width as can be seen in Fig. 11 presented in Sect. 5.2.2 below. Note, however, that even the width of the He I 4471 Å profile varies secularly with time (see Fig. 10 in the same section).

3.3. Light and colour changes

The existing rich series of differential UBV observations analysed below has only a limited overlap in time with the electronic spectra at our disposal. The lower panels of Fig. 2 show that a secular brightening was accompanied by the development of the Balmer emission in the spectra of 60 Cyg.

This is a behaviour characteristic for the so-called positive correlation between the emission strength and brightness of the object according to the classification devised by Harmanec (1983, 1994).

In Fig. 3, the long-term changes of 60 Cyg in the vs. colour diagram with respect to the standard colour sequences (adopted from Golay 1974) are shown. It is seen that if one takes into account the interstellar reddening, the variations are characterized by changes of the photometric luminosity class, corresponding always to the same spectral type of about B1. This is again typical for Be stars showing the positive correlation. If Harmanec's (1983, 1994) interpretation is sound, one could conclude that 60 Cyg is observed under an intermediate angle, not just equator-on.

Fig. 3. The vs. colour diagram in which the 150-d mean UBV values of 60 Cyg from the Xinglong, Hvar, and SPM observations (shown by empty triangles, diamonds, and a square, respectively) are compared with the standard main-sequence and supergiant-sequence UBV colours.

© European Southern Observatory (ESO) 2000

Online publication: April 17, 2000
helpdesk.link@springer.de