Astron. Astrophys. 358, 65-71 (2000)

4. Discussion

4.1. A dust halo?

The 2 mm map (Kuno & Matsuo 1997) shows continuum emission with a box-like shape, indicating a dust halo around M 82. Although the emission at 1.2 mm and 2 mm should be similar, we do not see any evidence for a significant halo contribution to the cold dust. This difference could be due to the error beam of the Nobeyama 45-m telescope.

Effects of the error beam and the antenna pattern have been investigated and largely removed from our map (see Sect. 2.3), while this is not mentioned in the case of the Nobeyama measurements. The total flux density in the Nobeyama map is 1.7 Jy, which is much more than expected at this wavelength (see Fig. 4). This extra flux might be due to emission entering through sidelobes and the error beam.

The 850 µm map made by Alton et al. (1999) with the JCMT shows continuum emission at the 50 mJy level up to 40" above the galactic plane. This translates to 15 mJy at 1.2 mm, corresponding to the second contour (3) in our map. Considering only the overall extent of the emission, the JCMT map is consistent with ours. The different shapes might be due to problems with the baseline subtraction, because the area mapped by Alton et al. is only slightly larger than the extent of the continuum emission.

4.2. Comparison of dust and CO

In non-active galaxies like NGC 4565 (Neininger et al. 1996) and NGC 5907 (Dumke et al. 1997) the molecular line emission is more concentrated towards the central region than the thermal dust emission. As can be seen from Figs. 3 and 5 the galaxy M 82 exhibits the opposite behaviour. The CO emission appears to be much more extended than the continuum. Since the maps in Fig. 5 are made with the same telescope at the same frequency, the difference can not be due to different beam patterns and error beams.

Fig. 5. Comparison of the 1.2 mm continuum and the CO(2-1) line emission. The upper image shows the CO-corrected 1.2 mm continuum map of M 82. Contours start at 2 (10 mJy/beam). The lower image shows the integrated CO(2-1) line intensity. Contours start at 3 (20 K km s-1/beam). The beam size of the IRAM 30-m telescope is given in the lower left corner of each image.

The dust is heated by the UV radiation field, which is concentrated towards the dense central region, because the UV photons can hardly escape without being absorbed. Only a small fraction of the UV photons makes its way out to kpc distances from the galactic plane, where it might contribute to the observed polarized radiation.

The molecular gas can be excited by the radiation field, but also by low-energy cosmic rays and by soft X-ray photons (see e.g. Glassgold & Langer 1973), which have a much larger scale height than the UV photons (Shopbell & Bland-Hawthorn 1998).

Fig. 3 shows that the synchrotron intensity at 20 cm drops faster with distance from the galaxy plane than the CO and even the dust emission. However, at low radio frequencies there may be a significant radio halo (low-energy cosmic rays have longer lifetimes), which is already indicated by the 327 MHz map of Reuter et al. (1992). Hence, there may be abundant cosmic rays for heating. Such cosmic rays (E 100 MeV) remain, however, invisible in the radio window, since even in the strong (G) magnetic field of M 82 they produce synchrotron radiation at frequencies below about 10 MHz. Soft X-rays are also seen far out of the plane of M 82 (e.g. Bregman et al. 1995). These circumstances altogether provide the likely heating sources for any molecular material transported out of the disk of M 82 into the halo, rendering the CO "overluminous". This is also indicated by the low conversion ratio of CO line intensity to molecular gas column density, derived by Smith et al. (1991). The existence of a soft X-ray and a low-frequency radio halo around M 82 readily explains the larger extent of the CO emission as compared to the cold dust.

© European Southern Observatory (ESO) 2000

Online publication: June 26, 2000
helpdesk.link@springer.de