HOT GAS IN THE INTERSTELLAR-MEDIUM - A REANALYSIS OF A O VI ABSORPTION DATA

The Copernicus O+5 column densities toward 72 stars provide a rare and valuable tracer of 10(5.5) K gas in the interstellar medium. The original analysis of the data by Jenkins provided important clues about the distribution of interstellar O+5 ions, but our understanding of the local interstellar medium has since grown substantially. We revisit that work, including the possibility that local hot gas may contribute a significant O+5 column density to most lines of sight. Our reanalysis also includes slight improvements in the statistics and was found to be reliable when tested on simulated data sets. In the end, we come to conclusions about the distribution of interstellar O+5 ions that differ considerably from those of the original analysis. 1. Good fits can be found by including the entire data set. 2. The combined mean column density of the Local Bubble and its interface with the Local Cloud is roughly 1.6 x 10(13) cm-2. 3. We find a smaller value for the midplane volume density of O+5 ions outside the Local Bubble (1.3-2.1 x 10(-8) cm-3) than Jenkin's 2.8 x 10(-8) cm-3 found assuming no Local Bubble. 4. The O+5-bearing material beyond the Local Bubble is probably clumped into features of much larger column density than estimated by Jenkins (2-7 x 10(13) O+5 ions cm-2 vs. 1 x 10(13) cm-2). 5. The mean line-of-sight distance between features beyond the Local Bubble is thus several times that of Jenkins (450-1300 pc vs. 165 pc). These changes strongly influence our understanding of the phase distribution in the interstellar medium, in particular the volume occupation of the hot and warm components, and mechanisms responsible for them. Stellar wind bubbles, cloud interfaces, fountains, and supernova remnants are discussed. With our reanalysis, some theoretical models now show promise. For example, our Local Bubble column density compares favorably with the estimated quantity of O+5 within the remnant of an ancient local explosion. Similarly, our mean O+5 column density per feature in more distant regions is like that found in models of hot interstellar bubbles from either stellar winds or ancient supernova explosions in a warm diffuse interstellar environment, suggesting that the hot gas in interstellar space may exist primarily within discrete regions of modest volume occupation rather than in a continuous and pervasive phase.