Harvard Astronomy 201b

The Hot ISM (Evidence and Observations)

In Uncategorized on March 3, 2011 at 6:59 pm

The hot ISM, also called coronal gas, refers to very hot, low density gas which has been shock-heated by fast stellar winds and blast waves from novae and supernovae. Its temperature and density are T \begin{smallmatrix} > \\ \sim \end{smallmatrix} 10^{5.5} K and n_H \sim 0.004 cm^{-3}, and it is believed to fill about half of the volume of the galactic disk, as well as much of the volume above and below the disk through chimney flows. Beyond the galaxy, much of the IGM is believed to be at T \begin{smallmatrix} > \\ \sim \end{smallmatrix} 10^6 K.

Unlike gas in typical HII regions, which is photoionized, coronal gas is collisionally ionized and contains very highly ionized species, such as OVI. Therefore, absorption lines from OVI and other ions is currently the most effective way of detecting hot ISM. When the gas temperature exceeds 10^6 K, it can also be detected via x-ray bremsstrahlung emission. Finally, radio synchrotron emission from rarefied coronas around galaxies, including our own, can be detected as well.

X-ray emission from hot gas

Left: Diffuse x-ray emission in the LMC observed with ROSAT, from Sasaki et al. (2002). Right: Surface brightness (0.1 - 2.4 keV) of the Vela SNR obtained with ROSAT, from Aschenbach et al. (1995).

Initial evidence

The existence of hot ISM in the galaxy was first postulated by Lyman Spitzer in 1956, with the publication of On a possible interstellar galactic corona. In this article, Spitzer presents the four main arguments in favor of the existence of a galactic corona:

a) Clouds at high latitude: Possibly the strongest piece of evidence at the time originated from analyzing the stability of normal clouds at great distances from the galactic plane, such as 500 pc. The argument is the following: if the rms speed of interstellar atoms, in the direction perpendicular to the galactic plane, is assumed to be around 10 km/s (which corresponds to T = 10000 K for H atoms), then the gas density at 500 pc should be 5% of the density in the galactic plane. So, if a normal galactic cloud with n_H = 10 cm^{-3} is placed in such a medium, it would expand at the speed of sound, about 1 km/s for T = 100 K. Hence a cloud 5 pc in radius would double its radius in about 5 \times 10^6 years and its pressure would drop to that of the surrounding medium in about 10^7 years. On the other hand, the period of oscillation back and forth across the galactic plane for clouds at 200 pc is about 8 \times 10^7 years, and the period increases for greater amplitudes. Therefore, these clouds must be in pressure equilibrium with the medium surrounding it. This equilibrium is explained by the existence of a tenuous gas at a high kinetic temperature, between 10^5 and 3 \times 10^6 K.

b) Equilibrium of spiral arms: This argument is similar to (a), but now applied to the equilibrium of the gas in the spiral arms of the Galaxy. This issue had already been considered by Chandrasekhar and Fermi in 1953, when they determined the magnetic field strength from the condition that the gravitational pressure equals the sum of the kinetic and magnetic pressures. However, the rms speed of 5 km/s used in their work seems low. Spitzer adopted a more accurate value of 12 km/s, which increases the kinetic pressure by a factor of 6 and makes it nearly equal to the gravitational pressure. The magnetic field pressure also increases and turns out to be significantly larger than the gravitational pressure. This suggests that, unless B is the same both inside and outside of the spiral arm, the gravitational pressure is not enough to account for the stability of a spiral arm. The situation would be greatly eased if the spiral arm is embedded in a rarefied gas with a pressure equal to the gas pressure in spiral arms.

c) Radio noise: As suggested by Shklovsky in 1952, much of the cosmic radio noise received at Earth might originate in an extended spherical corona surrounding the Galaxy. This suggestion was supported by observations on M31 at a wavelength of 3.7 m (Baldwin, 1954). These measurements revealed a spherical corona around M31, of radius of about 10 kpc, which accounts for about 2/3 of its radiation at these wavelengths. Baldwin performed similar measurements for our Galaxy, with positive results, but the results were not conclusive at the time.

d) Related suggestion by Pickelner: He proposed a somewhat different kind of corona. Although this view was not supported by Spitzer, he considered it important enough to be included in the publication. Pickelner claimed that the presence of wide H and K absorption lines, which Spitzer, Epstein, and Li Hen (1950) observed in supergiant B stars and which they attributed to stellar absorption, were in fact interstellar in origin.

Further evidence

In 1968, a rocket experiment by Bowyer et al. revealed the existence of diffuse x-ray emission in the energy range below 1 keV, which could be separated into an extragalactic and an “anomalous” component which appeared to be of Galactic origin, as predicted by Spitzer in 1956.

Soft x-ray background

Diffuse x-ray emission from Bowyer et al. (1968)

A definitive moment in our knowledge about the ISM came with the launch of COPERNICUS in 1972, which was inspired and promoted by Lyman Spitzer. This satellite detected a pervasive OVI absorption line, which was then analyzed in order to rule out a possible circumstellar origin, leading to the establishment of the hot ISM phase. As expressed by Donald G. York in 1974,

If the OVI lines are not circumstellar, they must therefore be formed in another phase of the interstellar medium, separate from normal HI and HII regions (Spitzer 1956; Cox and Smith 1974). It seems likely […] that there exists a hot plasma at temperatures ranging from 5 \times 10^5 to 2 \times 10^6 K.

However, it became clear that the hot medium observed with COPERNICUS could not reproduce the spectral characteristics of the soft x-ray background detected by Bowyer et al., so it seemed as if both observations were sampling different gas. This issue could only be solved by a unifying model of the ISM.

In 1977, McKee and Ostriker combined everything that was known at the time, including the newly discovered coronal gas, into a consistent picture of the ISM. The result was a three-component ISM in the galactic disk, composed of a cold neutral medium (CNM) coated by warm neutral clouds (WNM), with their ionized surfaces representing the warm ionized medium (WIM), and a hot ionized medium (HIM) with a fractional volume as large as f_V = 0.8, occupying most of the Milky Way disk. This number has been reduced to about f_V = 0.5, but this is still a matter of debate, with different authors claiming values between 0.3 and 0.7. However, the McKee and Ostriker model itself has outlived its competitors. Modern versions of this model, which include dust and additional ISM phases and their interactions, are in widespread use.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: