Michael E. Anderson

Hot gaseous halos are predicted around all large galaxies and are critically important for our understanding of galaxy formation, but they have never been detected at distances beyond a few kpc around a spiral galaxy. We used the Chandra ACIS-I instrument to search for diffuse X-ray emission around an ideal candidate galaxy: the isolated giant spiral NGC 1961. We observed four quadrants around the galaxy for 30 ks each, carefully subtracting background and point source emission, and found diffuse emission that appears to extend to 40-50 kpc. We fit $\beta$-models to the emission, and estimate a hot halo mass within 50 kpc of $5\times10^9 M_{\odot}$. When this profile is extrapolated to 500 kpc (the approximate virial radius), the implied hot halo mass is $1-3\times10^{11} M_{\odot}$. These mass estimates assume a gas metallicity of $Z = 0.5 Z_{\odot}$. This galaxy's hot halo is a large reservoir of gas, but falls significantly below observational upper limits set by pervious searches, and suggests that NGC 1961 is missing 75% of its baryons relative to the cosmic mean, which would tentatively place it below an extrapolation of the baryon Tully-Fisher relationship of less massive galaxies. The cooling rate of the gas is no more than 0.4 $M_{\odot}$/year, more than an order of magnitude below the gas consumption rate through star formation. We discuss the implications of this halo for galaxy formation models.

Galaxies are missing most of their baryons, and many models predict these baryons lie in a hot halo around galaxies. We establish observationally motivated constraints on the mass and radii of these haloes using a variety of independent arguments. First, the observed dispersion measure of pulsars in the Large Magellanic Cloud allows us to constrain the hot halo around the Milky Way: if it obeys the standard NFW profile, it must contain less than 4-5% of the missing baryons from the Galaxy. This is similar to other upper limits on the Galactic hot halo, such as the soft X-ray background and the pressure around high velocity clouds. Second, we note that the X-ray surface brightness of hot haloes with NFW profiles around large isolated galaxies is high enough that such emission should be observed, unless their haloes contain less than 10-25% of their missing baryons. Third, we place constraints on the column density of hot haloes using nondetections of OVII absorption along AGN sightlines: in general they must contain less than 70% of the missing baryons or extend to no more than 40 kpc. Flattening the density profile of galactic hot haloes weakens the surface brightness constraint so that a typical L$_*$ galaxy may hold half its missing baryons in its halo, but the OVII constraint remains unchanged, and around the Milky Way a flattened profile may only hold $6-13%$ of the missing baryons from the Galaxy ($2-4 \times 10^{10} M_{\odot}$). We also show that AGN and supernovae at low to moderate redshift - the theoretical sources of winds responsible for driving out the missing baryons - do not produce the expected correlations with the baryonic Tully-Fisher relationship and so are insufficient to explain the missing baryons from galaxies. We conclude that most of missing baryons from galaxies do not lie in hot haloes around the galaxies, and that the missing baryons never fell into the potential wells of protogalaxies in the first place. They may have been expelled from the galaxies as part of the process of galaxy formation.