3 History

3.1 Long-term variability

$\omega$ Ori has a long history of variability in Balmer emission lines, oscillating between B and Be phases (Hubert-Delplace & Hubert 1979). The long-term increase and decrease of the H$\alpha$ emission intensity for the last 25 years are seen in Fig. 1. Note that the maximum of H$\alpha$ observed during the early nineties coincides with a maximum in brightness according to the Hipparcos database (Perryman et al. 1997). Hayes & Guinan (1984) discovered polarization episodes, which were found to be nicely correlated with the presence of continuum flux and color change, weakly correlated with H$\alpha$ emission strength variations, but not correlated with enhanced wind variability (Sonneborn et al. 1988). The star showed recurrent light bursts every 11 months (long-term period) detected from the ground (Bergin et al. 1989; McDavid et al. 1996) which are compatible with the timescales given by Hipparcos (Hubert & Floquet 1998).

At the time of the MuSiCoS observations, the star was about to reach a minimum of the long-term photometric period and was close to a minimum of a Be phase. However, emission was still present in the wings of some lines during the campaign (e.g. H$\alpha$, H$\beta$, He I 5876, 6678, 7065, Si II 6347, C II 6578, 6583). The H$\alpha$ line had a double-peaked emission ($R \sim V$) profile with $I_{\max}/I_{\rm c} = 1.4$. Note in Fig. 1 that the emission has risen again since the end of the year 2000.

3.2 Short-term variability

Light modulation has been detected with a period of 1.96 days in optical flux (Balona et al. 1992), whereas the stellar wind seen in the UV lines is modulated by a shorter period: about 1 d in September 1987 and 1.2 d in February 1996 (Peters & Gies 2000). The period derived from a CCF (Cross Correlation Function) method on the photospheric lines from these UV spectra is 1.28 d. These results were interpreted as cyclic variability in the wind of the star, but the phenomenon could be caused by a localized region of mass loss sweeping past the line-of-sight as the star rotates. Recently Balona et al. (2001, hereafter B01) found an averaged period of 0.97 d using spectroscopic and photometric measurements, similar to the one found by Neiner et al. (2002, see also Henrichs et al. 2000b). B01 could not discriminate between NRPs and rotation as an explanation for this periodicity, but chose to describe it in terms of co-rotating circumstellar clouds.