ON THE ORIGIN OF WIND LINE VARIABILITY IN O STARS

Abstract

We analyze 10 UV time series for 5 stars which fulfill specific sampling and spectral criteria to constrain the origin of large-scale wind structure in O stars. We argue that excited state lines must arise close to the stellar surface and are an excellent diagnostic complement to resonance lines which, due to radiative transfer effects, rarely show variability at low velocity. Consequently, we splice dynamic spectra of the excited state line, N IV1718, at low velocity to those of 1393 component of the Si IV 1400 doublet at high velocity in order to examine the temporal evolution of wind line features. These spliced time series reveal that nearly all of the features observed in the time series originate at or very near the stellar surface. Further, we positively identify the observational signature of equatorial co-rotating interaction regions in two of the five stars and possibly two others. In addition, we see no evidence for features originating further out in the wind. We use our results to consolidate the fact that the features seen in dynamic spectra must be huge, in order to remain in the line of sight for days, persisting to very large velocity and that the photospheric footprint of the features must also be quite large, ~ 15 - 20% of the stellar diameter.