We report on a 67 ks High-Energy Transmission Grating observation of the optically brightest early O star z Puppis (O4 f). Many resolved X-ray lines are seen in the spectra over a wavelength range of 5-25 Å . Chandra has sufficient spectral resolution to study the velocity structure of isolated X-ray line profiles and to distinguish the individual forbidden, intercombination, and resonance (fir) emission lines in several He-like ions, even where the individual components are strongly Doppler-broadened. In contrast to X-ray line profiles in other hot stars, z Pup shows blueshifted and skewed line profiles, providing the clearest and most direct evidence that the Xray sources are embedded in the stellar wind. The broader the line, the greater the blueward centroid shift tends to be. The N vii line at 24.78 Å is a special case, showing a flat-topped profile. This indicates that it is formed in regions beyond most of the wind attenuation. The sensitivity of the He-like ion fir lines to a strong UV radiation field is used to derive the radial distances at which lines of S xv, Si xiii, Mg xi, Ne ix, and O vii originate. The formation radii correspond well with a continuum optical depth of unity at the wavelength of each line complex, indicating that the X-ray line emission is distributed throughout the stellar wind. However, the S xv emission lines form deeper in the wind than expected from standard wind-shock models.

The Great Nebula in Carina provides an exceptional view into the violent massive star formation and feedback that typifies giant HII regions and starburst galaxies. We have mapped the Carina star-forming complex in X-rays, using archival Chandra data and a mosaic of 20 new 60-ks pointings using the Chandra X-ray Observatory's Advanced CCD Imaging Spectrometer, as a testbed for understanding recent and ongoing -2star formation and to probe Carina's regions of bright diffuse X-ray emission. This study has yielded a catalog of properties of >14,000 X-ray point sources; >9800 of them have multiwavelength counterparts. Using Chandra's unsurpassed X-ray spatial resolution, we have separated these point sources from the extensive, spatiallycomplex diffuse emission that pervades the region; X-ray properties of this diffuse emission suggest that it traces feedback from Carina's massive stars. In this introductory paper, we motivate the survey design, describe the Chandra observations, and present some simple results, providing a foundation for the 15 papers that follow in this Special Issue and that present detailed catalogs, methods, and science results.

The Chandra spectrum of Ori A shows emission lines from hydrogen-and helium-like states of Si, Mg, Ne, and O, along with N vii Ly and lines from ions in the range Fe xvii-Fe xxi. In contrast to the broad lines seen in Pup and Ori (850 AE 40 and 1000 AE 240 km s À1 half-width at half-maximum [HWHM], respectively), these lines are broadened to only 430 AE 60 km s À1 HWHM. This is much lower than the measured wind terminal velocity of 2000 km s À1. The forbidden, intercombination, and resonance (fir) lines from He-like ions indicate that the majority of the X-ray line emission does not originate at the base of the wind, in agreement with the standard wind shock models for these objects. However, in that model the X-ray emission is distributed throughout an expanding, X-ray-absorbing wind, and it is therefore surprising that the emission lines appear relatively narrow, unshifted, and symmetric. We compare the observed line profiles to recent detailed models for X-ray line profile generation in hot stars, but none of them offers a fully satisfactory explanation for the observed line profiles.

We report on a Chandra line spectrum observation of the O supergiant, ζ Orionis (O9.7 Ib). A 73.4 ks HETGS observation shows a wide range of ionization stages and line strengths over the wavelength range of 5 to 26Å. The observed emission lines indicate a range in temperature of 2 to 10 MK which is consistent with earlier X-ray observations of ζ Ori. Many lines are spectrally resolved showing Doppler broadening of 900 ± 200 km s −1 . The observed He-like ions (O VII, Ne IX, Mg XI, and Si XIII) provide information about the spatial distribution of the X-ray emission. Although the observations support a wind distribution of X-ray sources, we find three conflicting results. First, line diagnostics for Si xiii indicate that this line emission forms very close to the stellar surface, where the density is of order 10 12 cm −3 , but the velocity there is too small to produce the shock jump required for the observed ionization level. Second, the strong X-ray line profiles are symmetric and do not show any evidence of Doppler blue-shifted line centroids which are expected to accompany an outwardly moving source in a high density wind. Third, the observed velocity dispersions do not appear to correlate with the associated X-ray source radii velocities, contrary to expectations of wind distributed source models. A composite source model involving wind shocks and some magnetic confinement of turbulent hot plasma in a highly non-symmetric wind, appears to be needed to explain the line diagnostic anomalies.