The 21-cm H I lines have three related observational aspects: good spin temperature correlation at high latitudes with optical depth, the suggestion of a ubiquitous intercloud component that is incompatible with observations of other species, and emission lines that are broader than their associated absorption profiles. The first of these can be reproduced by introducing a column density-spin temperature correlation into uniform cloud models. The intercloud medium can be reproduced by putting hot skins around absorbing cloud cores, in the mode of the McKee-Ostriker (1977) model of the interstellar medium. The disparate line widths of emission and absorption features require cloud models in which the gas temperature varies substantially and continuously across a given cloud.
No takes yet. Share an insight, caveat, or question.
H. S. Liszt (1983) studied this question.