Bipolar hypershell models of the extended hot interstellar medium in spiral galaxies

¹ Institute of Astronomy, University of Tokyo, Mitaka, Tokyo 181-0015, Japan
² CEA/Saclay, DAPNIA, Service d'Astrophysique, L'Orme des Merisiers, Bât. 709, 91191 Gif-sur-Yvette, France

Corresponding author: Y. Sofue, sofue@ioa.s.u-tokyo.ac.jp

Received: 23 August 2000
Accepted: 15 January 2001

Abstract

We simulated the million degree interstellar medium and its soft X-ray images in the disk and halo of spiral galaxies using the bipolar hypershell (BHS) model. In this model dumbbell- or hourglass-shaped expanding shells of several kpc radii are produced by a sudden energy release in the central region. We then applied our model to a mini-sample of starburst galaxies seen under different inclinations, namely the nearly edge-on galaxies NGC 253, NGC 3079 and M 82, the highly inclined galaxies NGC 4258 and NGC 1808 as well as the nearly face-on galaxy M 83. For all galaxies, our results reproduce the X-ray characteristics observed in the 0.1-2.4 keV ROSAT energy band: the bipolar hypershell morphology, the spectral energy distribution of the diffuse disk and halo emission as well as absorption gaps in the diffuse X-ray emission caused by a shadowing of soft X-rays due to cold intervening gas in the disks of the galaxies. In general, the required total energy for the starburst is estimated to be of the order of 10⁵⁵ ergs, corresponding to the overall kinetic energy generated in ~10⁴ type-II supernova explosions. The expansion velocity of the shells is estimated to be ~200 km s^-1, which is necessary to heat the gas to ~0.2 keV (2.3 million K), and the age to be of the order of 3 10⁷ years. In the case of the very nearby, nearly edge-on galaxy NGC 253 all characteristics of the BHS model can be studied with high spatial resolution. Using the property that the shell morphology is sensitive to the ambient density distribution, we propose using soft X-ray data to probe the gas distributions in the disk, halo and intergalactic space in general. The application of our model to images at higher spatial and spectral resolution, as provided by Chandra and XMM, will help us to further disentangle the ISM density distributions and will lead to a better understanding of the disk halo interface.