The three galaxies discussed in this paper are of quite different morphological types. However, the results that we have presented suggest that we have a warm, dense and widely distributed CO gas in all the three observed objects. The J=3 transition is 33 K above ground level. If the CO(3-2) lines were to be optically thick, temperatures of K are expected to prevail in the emission regions. In the framework of a simple LVG analysis we expect the ratio if the CO gas has K, , and . For a more detailed computation of the possible values of the parameters for various velocity gradients we refer to Mauersberger et al. (1999). In the case that molecular clouds are sufficiently warm and dense, the opacity of the CO(3-2) line will be larger than that of the CO(1-0) line. In this case and if the excitation temperature of clump surfaces is larger than that in clump interiors, the CO(3-2) emission traces regions definitely warmer gas than the CO(1-0) line. This effect may bring the line ratio to be significantly larger than unity even when the emissions are opaque. Obviously, further supporting observations of other species (13CO, C18O etc.) are required to further the study of this interesting discovery. Additional observations of the CO(4-3) line at 460 GHz would also be very helpful. Also more accurate line ratios must be determined to differentiate between the various possible scenarios.
Although we have presented data for only three galaxies, we argue that surprisingly warm and dense CO gas is likely to be found in all nearby galaxies. The fact that the line intensity, line ratios and line widths are very similar in our three morphologically different galaxies indicates that similar heating mechanisms are at work. This would mean that quite high temperatures and densities must be present in the molecular clouds that fill only a small portion of our beam. In the recent paper by Mauersberger et al. (1999) this trend was shown to be present for the nuclei of many nearby galaxies.
The present observations of galaxies in the CO lines should also be considered in light of the measurements of dust temperatures using mm and sub-mm wavelength bolometers. In numerous studies of nearby galaxies the dust temperatures were found to have distinct components at two temperatures. The 1.2 mm continuum emission of NGC 4631, observed with a very similar telescope beam (Braine et al. 1995), gave temperature components of 20.5 K and 55 K. The decomposition of dust temperatures for M51 (Guélin et al. 1995) gave temperature components of 18 K and 50 K. The data from the ISO satellite (Hippelein et al. 1996) also suggest relatively high dust temperatures for the central region of M51. Our observations however suggest hotter CO gas to be present away from the nucleus of M51 in the outer spiral arms, where the dust temperatures are in fact lowest. Obviously detailed and more accurate line ratio determinations are required for the whole galaxies (as opposed to studies of nuclear regions only) to understand the situation in the galaxies.
© European Southern Observatory (ESO) 1999
Online publication: June 30, 1999