DHC 30June2004

Draft text for Banff poster:

Figure XX: The strongest line in the MEG spectrum of component AB is the oxygen Lyman alpha line at 18.97 Angstroms.  We fit this line with a Gaussian lineshape model (plus a polynomial to account for the contribution of the weak continuum), allowing the position and the width of the model to be free parameters, along with the amplitude.  The best-fit model is shown as a red dotted line, superimposed on the coadded -1 and +1 MEG spectral order data.  This model is slightly, but significantly, broadened.  For comparison, we also show a totally narrow line model (green dashed) and a model with a half-width of 1000 km/sec (blue dot-dashed), as might be expected from a wind-shock source like the O4 supergiant, zeta Pup.  All of these models have been convolved with the instrument response function and their amplitudes are normalized to the total line flux in the data.  The residuals shown in the lower panel are for the best-fit, slightly broadened, model. 


Text about line widths: 

The strongest emission lines in the MEG spectrum are quite narrow, but the ones with the highest signal-to-noise do show a small amount of Doppler broadening.  In Fig. XX we show the primary example of this, the oxygen Lyman-alpha line.  The best-fit Gaussian line profile model has a half-width of 280 +/- 30 km/sec.  This is roughly five times greater than both the expected thermal broadening and the wavelength split of the two lines that compose the doublet.  The wind terminal velocity of sigma Ori A is more than 1000 km/sec (Howarth & Prinja, 1989), and CAK theory predicts a terminal velocity more than twice that value. It has generally been found that wind-shock O stars, like zeta Pup (Cassinelli et al. 2001), have X-ray emission line half-widths of roughly half their wind terminal velocities.  That the X-ray lines' widths tend not the be the full wind terminal velocity is an indication that much of the shock-heated plasma forms in the acceleration region of the wind (Kramer, Cohen, & Owocki 2003).  However, for the standard wind-shock picture to account for the only slightly broadened lines seen in sigma Ori A, the hot plasma would have to form quite near the photosphere, in the base of the wind.  Similar phenomenology was also found in the unusual, young B0 V star, tau Sco (Cohen et al. 2003), where line broadening was very small, but non-zero.  It is possible that these very late O and very early B stars, with their weak winds, have a different type of shock-heated plasma distribution than early O stars with much more robust winds.  While magnetic fields may play a role in modulating the wind dynamics, the MCWS model predicts much harder X-rays than are seen in component A (one need only look at component E for a comparison).  Finally, no late-type coronal X-ray emission has ever shown detectable broadening in Chandra grating spectra, so it seems unlikely that the emission from sigma Ori A can be explained by the type of coronal mechanism that exists on late-type stars. 

Text about line ratios: 

The helium-like complexes of O, Ne, and Mg are detected in the MEG spectrum of component AB.  In all cases, the forbidden line strength is consistent with zero.  In hot stars, weakened He-like forbidden lines (1s2s ^3S - 1s^2 ^1S) are thought to be due to photoexcitation out of the upper level (1s2s ^3S - 1s2p ^3P) by UV radiation from the photosphere, rather than by collisional excitation, as is the case in late-type coronae.  The ratio of the forbidden to intercombination (1s2p ^3P - 1s^2 ^1S) line is thus a diagnostic of the mean intensity of the UV field, and thus of the distance of the X-ray emitting plasma from the photosphere.  We have performed preliminary modeling of the Ne IX emission line complex using a suite of detailed atomic models and non-LTE statistical equilibrium and radiation transport codes (http://www.prism-cs.com/Software/Software_overview.htm).  The non-detection of the forbidden line requires that the X-ray emitting plasma be within about 2 stellar radii of the photosphere.  This result is consistent with the inferences from the line widths that the X-ray emitting plasma is near the base of the wind.