Million degree gas in the halos of spiral galaxies

Today is the deadline for the latest round of Chandra observing proposals, and it looks like I'm not going to finish my epic new proposal on time. Grrr.

But it reminds me that there really aren't many good pictures, on the popular astronomy side of the web, of what galaxies look like, as seen using the X-ray's the galaxies emit. AGN, star clusters, galaxy clusters. Tons of those, but non-active galaxies are much less represented.

The images above shown below are false-color composities of X-ray emission (blue), optical H-alpha line emission (red) and "normal" optical starlight (green), of nearby roughly-edge on spiral galaxies.

Note: blogger is doing something odd with displaying the table, its leaving a huge space, so scroll down a bit and the images will appear.

X-ray and optical images of edge-on spiral galaxies

M82 (starburst)	NGC 1482 (starburst)
NGC 253 (starburst w/LLAGN)	NGC 3628 (starburst)
NGC 3079 (starburst w/LLAGN)	NGC 4945 (Sy 2 w/starburst)
NGC 4631 (starburst)	NGC 6503 (normal)
NGC 891 (normal)	NGC 4244 (normal)

Our own galaxy (the Milky Way) is a "normal" spiral galaxy, as is its neighbours M31 (Andromeda) and M33. Of the galaxies shown above, the Milky Way has long been thought to be closest in type, size, total mass and star formation activity to NGC 891. However more modern observations, particularly with Spizter, indicate NGC 891 is significantly more actively star-forming than the Milky Way.

These images are all shown in the same scale (the size of the image is roughly equivalent to 65 thousand light years on side at the distance of the galaxy in question), and they're all taken from one of my papers on Chandra observations of a sample of nearby edge-on spirals.

As you can see spiral galaxies are more than two-dimensional disks, and the more actively star-forming a galaxy is, the more likely it is to have hot gas (2 to 5 million degree plasma) in its halo, which is what we see in the X-ray data. The galaxies are shown in terms of their infra-red "warmth", a ratio of the brightness in two wavelengths measured with the old IRAS satellite. The IR warmth is physically a measure of the temperature of interstellar dust (more like soot really), which itself is determined by the average number of stars forming per unit area in the disk of the galaxy.

Galaxies essentially have atmospheres, and those atmospheres are important for a number of reasons I might get around to mentioning in a future blog. In some of the images you can also see that the H-alpha emission (from ionized gas at only T ~ 10000 degrees) also reaches into the halo. Other galaxies appear to show now emission in their halos, but at present it is not clear whether this really means they don't have gaseous halos, or whether there is just so little gas that its too
faint to detect - both hypothesises are consistent with present data.

For the brighter examples, the starburst galaxies, the halo emission is almost certainly not a static atmosphere but gas lit-up by, and incorporated into, a galaxy-sized wind flowing out of the galaxy, a phenomen otherwise known as a superwind.

More about superwinds some other time. The following informational is more background information on the images themselves.

The intensity scale is square-root in all images and in all bands (so in the images the galaxies all appear roughly equally bright), but in reality the absolute intensities differ from image to image.

Blue is diffuse soft (0.3-2.0 keV energy band) X-ray emission from the Chandra observations, which is basically the light emitted by a gas heated to several million degrees. These images have been rather heavily smoothed, so much of the genuine smaller scale structure in the X-ray data is not visible in these images.

Green is optical R-band, i.e. emission from stars, primarily old stars. In true-color galaxies are pretty close to looking white while R-band is somewhat red, but this can't be shown in the form of color composite image I'm using.

Red is continuum subtracted H-alpha, or H-alpha+[N II] emission, i.e. emission from warm ionized gas with T~10^4 K. Within the disk of these galaxies you're seeing HII regions (gas photo-ionized by the UV radiation of hot, massive, young stars), but the extra-planar H-alpha in the starbursts is also collisionally-ionized (i.e. heated by shock-waves) warm ionized gas being dragged out of these galaxies by the superwind at velocities between 200 and 1000 km/s, depending which superwind we're interested in.

Note:We did not detect any extra-planar (|z|> 2kpc) soft X-ray emitting gas around either NGC 6503 and NGC 4244 (normal, not starburst, galaxies). There is only weak extra-planar diffuse X-ray emission detected above the nuclear regions of NGC 4945. The X-ray emission apparent in the top left and bottom right of the NGC 4945 image, and that outside the optical disk in the image of NGC 6503, is at the noise level of the X-ray data (we believe it is noise).

[Update: 17/03/2007 - corrected some of the typos]