Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 344, 779-786 (1999)

Previous Section Next Section Title Page Table of Contents

4. A look forward

In this paper we have presented a full spectral scan from 43 to 189 microns of what could reasonably be considered a normal galactic disk. It is likely that this is what an outside observer would obtain when looking at the molecular ring in our galaxy, without all the problems (being blinded by local material and unable to derive a face-on view of our disk) we have when observing our galaxy from within its thin disk.

The problem in obtaining the global spectrum of the Galaxy can be illustrated as follows. Two roughly co-eval CO(1-0) surveys of the Milky Way (Sanders et al. 1984; Dame & Thaddeus 1985) obtained quite different results for the H2 mass of the Galaxy. Most of the difference is not due to different [FORMULA] factors (Bronfman et al. 1988) but rather to derived CO luminosities. Using the COBE FIRAS instrument, the luminosities in the CO lines estimated by Wright et al. (1991) are much lower than either previous estimate. Bennett et al. (1994) point out that Wright et al. underestimate the CO luminosities because they assumed that the CO/FIR ratio was constant. In fact, the CO is more centrally peaked and this effect increases with J. Velocity information is not available to help determine more accurate luminosities for this unique survey of CO lines. If the Galactic CII/CO(2-1) luminosity ratio is really [FORMULA], as in NGC 4414, then Wright et al. underestimated the CO(2-1) luminosity by a factor 2 (if [FORMULA] is correct) due to the uncertainty in the distribution of the emission.

It is important to address the issue of high-J CO lines because of the increasing number of IR-luminous objects, usually gravitationally lensed, detected in high-J CO lines redshifted to mm wavelengths. Will we be able to detect normal galaxies, if they exist, at high redshift? Solomon et al. (1992) pointed out that the CO line intensity varies as [FORMULA] where [FORMULA] is the luminosity distance and [FORMULA] is the CO luminosity in brightness units (K km s-1 pc2). At a given telescope and frequency, the beam size [FORMULA] stays constant and [FORMULA] should not decrease with increasing redshift because [FORMULA] increases slowly with z, making it much easier than expected to detect CO in distant objects by observing successively higher transitions as one goes to higher redshift objects. An important caveat, however, is the constant [FORMULA] hypothesis. The Wright et al. data indicate that for the Milky Way [FORMULA] for [FORMULA] (constant luminosity in [FORMULA]). Such a steep decrease appears unreasonable but underlines how little we know about [FORMULA] CO transitions in normal galaxies. The answer conditions whether we will be able to detect high-redshift "normal" galaxies or prove they do not exist.

Our data (Fig. 4) should be considered a template for the large-scale emission of a disk. As such, they are complementary to Galactic observations where good linear resolution is available, enabling individual molecular clouds and HII regions to be studied, but where we are too sensitive to local material to be confident of the overall picture. The continuing improvements in models and observations and the ever-increasing wavelength coverage will yield more precise images of the emission of the various components of the ISM in our galaxy - PDRs, CNM, ELDWIM... The NGC 4414 data can be viewed as the sum of all these over large areas and the respective masses in each phase of the ISM will become clearer as our knowledge of the individual components improves.

After the LWS observations, the main gap in our knowledge of the emission of the ISM is the submm from 1 mm to 200 µm (see Fig. 4). CI (492 GHz) is the next most important coolant of the neutral ISM and necessary in any template. High-J CO observations would show whether an observable amount of very dense warm gas is present, much like Harris et al. (1991) for starburst nuclei. With ground-based sub-mm telescopes offering beam-sizes of [FORMULA] at 492 GHz and nearby galaxies like NGC 4414, it is possible to measure the CI/CO ratio as a function of radius and follow up earlier observations of the CO([FORMULA] line ratio in NGC 4414 (Paper II). The CI line is optically thin so it provides an estimate of the neutral carbon column density which can be compared with our existing CO, 13CO, HI, and dust continuum measurements - does the fraction of atomic carbon increase as the gas temperature, abundance, and radiation field decrease with galactocentric distance? The high-J CO lines come from the same warm dense regions as the OI 63 µm emission (HII/PDR/molecular cloud interfaces) and the CI from the more diffuse photo-dissociated gas which emits strongly in the CII line. The relative strengths of the CI and high-J CO lines provide one of the clearest means of separating the dense and diffuse PDR components both as a function of radius and integrated over the disk.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1999

Online publication: March 29, 1999