Astron. Astrophys. 345, 172-180 (1999)

1. Introduction

Pulsating stars are found in nearly every part of the HR diagram. Among O stars, however, only very few pulsators are known. Fullerton et al. (1996) listed 3 confirmed and 6 suspected pulsating O stars and noticed that all these massive stars are located in the instability strip predicted by Kiriakidis et al. (1993). Besides its asteroseismological potential, the search for non-radial pulsations (NRP) in early-type stars is largely motivated by the unknown origin of the widely observed cyclic variability in their winds, notably in the absorption parts of the ultraviolet P Cygni profiles. Most prominent are the migrating discrete absorption components (DACs) with a recurrence time scale that can be interpreted as (an integer fraction of) the stellar rotation period. The cause of the cyclical wind variability is an unsolved issue. It is not known whether the variability is strictly periodic. For example, a comparison of four different datasets obtained in subsequent years of the O7.5III star Persei (Kaper et al. 1999) shows that the dominant period remains 2 (or 4) days, but detailed changes in the variability pattern occur from year to year.

The wind structures can be traced in radial velocity back down to the vsini value of the star (see for  Per Henrichs et al. 1994, Kaper et al. 1996). This argues in favor of a model with corotating wind structures similar to the Corotating Interaction Regions in the solar wind (CIRs, in the context of hot-star winds first proposed by Mullan 1984). The CIR model invokes fast and slow wind streams that originate at different locations at the stellar surface. Due to the rotation of the star, the wind streams are curved, so that fast wind material catches up with slow material in front forming a shock at the interaction region. The shock pattern in the wind is determined by the boundary conditions at the base of the wind and corotates with the star. In the radiative hydrodynamical computations by Cranmer & Owocki (1996) these spiral-like regions indeed emerge, giving rise to accelerating DACs in wind lines, very similar to what is observed. The key point is, however, that the physical origin of the fast and slow streams is not specified in the calculation: either magnetic fields or non-radial pulsations could equally provide the required differentiation of the emerging wind. In the first case the number of wind structures is determined by the number of magnetic footpoints and the modulation comes directly from the stellar rotation, whereas in the case of a single NRP mode the value of the azimuthal number m determines the azimuthal distribution of the wind structures and the modulation is caused by the traveling speed of the pulsation superposed on the stellar rotation in the observers frame. A third case could also be considered, in which coadding amplitudes of multiple modes may give rise to wind perturbations with a periodic nature depending on the specific modes.

The timescales of pulsation, rotation and wind flow are all on the order of one day, which makes it particularly difficult to disentangle these effects, and which forces a ground-based multi-site approach, preferably simultaneously with UV spectroscopy from space. A different approach has been followed by Howarth et al. (1998), who were able to recover pulsation periods of HD 64760 B0.5Ib and HD 93521 O9.5V derived from UV data by applying cross-correlation techniques. Their paper is concerned with the same problem as addressed here. In our study we decided to concentrate on the O stars Persei O7.5III(n)((f)) and Cephei O6I(n)fp because of their brightness, conveniently close relative location in the sky, excellent record of their UV resonance line behavior, and suitable recurrence period of DACs (1-2 days). The strategy was to probe simultaneously the outer part of the stellar wind (using UV resonance lines), the inner part of the wind (using N IV 1718 and H) and the stellar photosphere (using optical lines). Lines formed deep in the photosphere are used to study pulsation behavior by means of Doppler imaging techniques.

This paper reports the results of a new analysis of the behavior of such a photospheric line HeI 4713 of such a campaign in 1989 during 5 days. Whereas previous analyses of this dataset did not yield convincing results, the advent of new methods of line-profile analysis allowed us to detect the pulsation modes described below.

© European Southern Observatory (ESO) 1999

Online publication: April 12, 1999
helpdesk.link@springer.de