SIMBAD references

An atlas of time series of ultraviolet spectra is presented for 10 bright O stars. The spectra were obtained with the International Ultraviolet Explorer during seven observing campaigns lasting several days over a period of 6years. The UV P Cygni lines in 9 out of the 10 studied stars exhibit a characteristic pattern of variability in the form of discrete absorption components (DACs) migrating through the absorption troughs on a timescale of a day to a week. This pattern is significantly different for each star, but remains relatively constant during the time span of our observations for a given star. A quantitative evaluation of the statistical significance of the variability is given. The winds of a number of stars appear to vary over the full range of wind velocities: from 0km/s up to velocities exceeding the terminal velocity v_∞ of the wind as measured by the asymptotic velocity reached by DACs. The amplitude of variability reaches a maximum at about 0.75v_∞ in the unsaturated resonance lines of stars showing DACs. In saturated resonance lines we find distinct changes in the steep blue edge. This edge variability is also found, although with smaller amplitude, in unsaturated resonance lines. The subordinate line of N IV at 1718 A in ξ shows weak absorption enhancements at low velocities in the blue-shifted absorption that are clearly associated with the DACs in the UV resonance lines. We interpret these three manifestations of variation as reflecting a single phenomenon. The DACs are the most conspicuous form of the variability. The changes at the edge can often be interpreted as DACs, but superposed on a saturated underlying wind profile; in many cases, however, at the same time two or more absorption events in different stages of their evolution can be identified in the unsaturated profiles, hampering a detailed interpretation of the edge variability. The low velocity absorption enhancements in the subordinate lines are the precursors of DACs when they are formed close to the star. The constancy of the pattern of variability over the years and the (quasi-)periodic recurrence of DACs strongly suggest that rotation of the star is an essential ingredient for controlling wind variability. The observation of low-velocity variations in subordinate lines, which are supposedly formed at the base of the stellar wind, indicate an origin of wind variability close to or at the photosphere of the star.