A HIGH SPATIAL RESOLUTION X-RAY AND Hα STUDY OF HOT GAS IN THE HALOS OF STAR-FORMING DISK GALAXIES. II. QUANTIFYING SUPERNOVA FEEDBACK

摘要

We investigate how the empirical properties of hot X-ray-emitting gas in a sample of seven starburst and three normal edge-on spiral galaxies (a sample that covers the full range of star formation intensity found in disk galaxies) correlate with the size, mass, star formation rate, and star formation intensity in the host galaxies. From this analysis we investigate various aspects of mechanical energy "feedback"- the return of energy to the interstellar medium from massive star supernovae and stellar winds- on galactic scales. The X-ray observations make use of the unprecedented spatial resolution of the Chandra X-Ray Observatory to remove X-ray emission from point sources more accurately than in any previous study and hence obtain the X-ray properties of the diffuse thermal emission alone. Intriguingly, the diffuse X-ray properties of the normal spirals (in both their disks and halos) fall where extrapolation of the trends from the starburst galaxies with superwinds would predict. We demonstrate, using a variety of multiwavelength star formation rate and intensity indicators, that the luminosity of diffuse X-ray emission in the disk (and, where detected, in the halo) is directly proportional to the rate of mechanical energy feedback from massive stars in the host galaxies. Accretion of gas from the intergalactic medium (IGM) does not appear to be a significant contributor to the diffuse X-ray emission in this sample. Nevertheless, with only three nonstarburst normal spiral galaxies it is hard to exclude an accretion-based origin for extraplanar diffuse X-ray emission around normal star-forming galaxies. Larger galaxies tend to have more extended X-ray-emitting halos, but galaxy mass appears to play no role in determining the properties of the disk or extraplanar X-ray-emitting plasma. The combination of these luminosity and size correlations leads to a correlation between the surface brightness of the diffuse X-ray emission and the mean star formation rate per unit area in the disk (calculated from the far-infrared luminosity and the optical size of the galaxy, L_(FIR)/D_(25)~2). Further observational work of this form will allow empirical constraints to be made on the critical star formation rate per unit disk area necessary to blow hot gas out of the disk into the halo. We show that a minor generalization of standard superbubble theory directly predicts a critical star formation rate per unit area for superbubble blowout from the disk and by extension for superwinds to blow out of the gaseous halos of their host galaxy. At present there are a variety of poorly known parameters in this theory that complicate comparison between observation and theory, making it impossible to assess the quantitative accuracy of standard superbubble blowout theory. We argue that the crucial spatial region around a galaxy that controls whether gas in starburst-driven super-winds will escape into the IGM is not the outer halo ～100 kpc from the host galaxy, but the inner few halo scale heights, within ～20 kpc of the galaxy plane. Given the properties of the gaseous halos we observe, superwind outflows from disk galaxies of mass M～10~(10)-10~(11) solar mass should still eject some fraction of their material into the IGM.