2015ApJ...800...14M

The Milky Way hosts a hot (~2x10⁶ K), diffuse, gaseous halo based on detections of z = 0 O VII and O VIII absorption lines in quasar spectra and emission lines in blank-sky spectra. Here we improve constraints on the structure of the hot gas halo by fitting a radial model to a much larger sample of O VII and O VIII emission line measurements from XMM-Newton/EPIC-MOS spectra compared to previous studies (~650 sightlines). We assume a modified β-model for the halo density distribution and a constant-density Local Bubble from which we calculate emission to compare with the observations. We find an acceptable fit to the O VIII emission line observations with Ξ_red² (dof) = 1.08 (644) for best-fit parameters of n_or_c^3β = 1.35 ± 0.24/cm3 kpc^3β and β = 0.50±0.03 for the hot gas halo and negligible Local Bubble contribution. The O VII observations yield an unacceptable Ξ_red² (dof) = 4.69 (645) for similar best-fit parameters, which is likely due to temperature or density variations in the Local Bubble. The O VIII fitting results imply hot gas masses of M(<50 kpc) = 3.8_–0.3^+0.3 x 10⁹M_☉ and M(<250 kpc) = M (<250 kpc) = 4.3_–0.8^+0.9 10¹⁰M_☉, accounting for ≲ 50% of the Milky Way's missing baryons. We also explore our results in the context of optical depth effects in the halo gas, the halo gas cooling properties, temperature and entropy gradients in the halo gas, and the gas metallicity distribution. The combination of absorption and emission line analyses implies a sub-solar gas metallicity that decreases with radius, but that also must be ≥ 0.3 Z_☉ to be consistent with the pulsar dispersion measure toward the Large Magellanic Cloud.