Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

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

Previous Section Next Section Title Page Table of Contents

4. Medium-term changes

As Fig. 2 shows, the spectrum of 60 Cyg varies also on the time scale of hundreds days or shorter. The light might vary in the similar way as can be seen from Fig. 6.

4.1. Spectroscopy

In some cases (e.g. KX And or [FORMULA] Dra) the medium-term spectral changes of Be stars were found to be related to the binary nature of the stars in question - cf., e.g. Harmanec (1983). This is the case even if the system is not a simple analog of a mass-exchanging binary with a secondary star filling its Roche lobe, as it can be demonstrated on 4 Her (HD 142926) - see Koubský et al. (1997).

Therefore, the RV measurements could help to identify the time scales of possible medium-term changes in the spectra of Be stars. In the case of 60 Cyg, the determination of reliable radial velocities is not an easy task. All the profiles of the helium lines are strongly affected by the rapid line-profile changes and also by the long-term variability. To overcome this problem, we therefore measured the wings of the H[FORMULA] emission and also the central absorption core. The RVs were measured using the method of alignment of direct and mirror profiles on the computer screen where the digitized profiles were displayed. Keeping in mind that the position of the emission line wings could be influenced by the rectification of the spectra, we made all efforts to reduce all spectra in a homogeneous way. We indeed found real RV variations. This is demonstrated by Fig. 4 where the emission and absorption profiles, corresponding to extreme measured values of emission and absorption RVs, are shown. Using Stellingwerf 's (1978) PDM period search technique, we carried out a period search in several observed quantities: emission and absorption RV of H[FORMULA], mean bisector RV of He I 6678 Å and the V/R ratio of the emission peaks of H[FORMULA]. The search was carried out over the range of periods from 0[FORMULA]2 to 200[FORMULA]0. The best fit was found in all cases for periods near about 150 d. After inspecting the PDM periodograms and sinusoidal least-squares fits, we concluded that the most probable period is close to 146[FORMULA]6. We suggest the following linear ephemeris:


[FIGURE] Fig. 4. Extremes of the radial velocity of emission (upper part, E) and absorption (lower part, A) components of H[FORMULA]. The particular radial velocities are displayed in the left part of the figure.

Corresponding phase diagrams for this period are shown in Fig. 5. One can see that the emission and absorption RVs of H[FORMULA] vary in antiphase. The semi-amplitude of the H[FORMULA] absorption is less than 3 km s-1, therefore the reality of this variation should be further tested. The RV of He I 6678 Å shows a great deal of scatter which is mainly due to sub-features moving rapidly across the line profiles - see below. Yet, the fit shows that both, the systemic velocity and the amplitude of the RV changes of the H[FORMULA] emission and He I 6678 Å absorption are very similar, probably identical within the limits of their errors. The series of He I 4471 Å and He I 4388 Å line profiles are not numerous enough to be used but their RV changes do not contradict the 147-d period. Notably, the V/R variation of H[FORMULA] calls for a significantly longer period of about 151-154 d.

[FIGURE] Fig. 5. Periodic RV variations of the H[FORMULA] and He I 6678 Å lines with the 146[FORMULA]6 period. See the text for details.

4.2. Photometry

In order to search for medium-term changes in the light of 60 Cyg the long-term trend was first removed. The prewhitening with a spline function is depicted in Fig. 6. A period search on these O-C light changes gives a very complicated power spectrum and no dominant frequencies are found. Also a plot of prewhitened V magnitudes vs. phase of the 146[FORMULA]6 period does not show any compelling evidence of variations on this time scale.

[FIGURE] Fig. 6. Prewhitening of the observed V magnitudes of 60 Cyg with a spline function after Vondrák (1969, 1977).

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: April 17, 2000