Joint Astronomy Centre
Show document only
JAC Home
JCMT
UKIRT
Contact info
JAC Divisions
OMP
Outreach
Seminars
Staff-only Wiki
Weather
Web Cameras
____________________

JCMT home
Observing at JCMT
OMP Observation Manager
Telescope
Spectral Line Observing
Continuum Observing
Schedule
Data Archive
Future Developments
Legacy Surveys
Newsletter & Publications
JAC Style guide: template
JACH | JCMT | UKIRT | Computer Services | Local server

Physical Conditions in Molecular Clouds in M33

We have used the JCMT to observe the 12CO J=2-1, 13CO J=2-1, and 12CO J=3-2 lines in a sample of seven giant molecular clouds in the Local Group spiral galaxy M33. The clouds were chosen to cover a wide variety of star formation conditions, from clouds with no optical H II regions to a cloud located in the brightest giant HII region in the galaxy. Determining the physical conditions inside molecular clouds is important for understanding the link between the properties of the molecular gas and the types and amount of stars that are formed. Cloud properties that could affect the star formation process include the temperature and density of the molecular gas, as well as the mass fraction in high density gas. In return, star formation, particularly massive star formation, can affect conditions inside molecular clouds by compressing the gas at the boundaries of stellar wind or supernova shocks and heating the gas by increasing the ultraviolet radiation field.

The 12CO/13CO J=2-1 line ratios are very uniform, with an average value of 7.3 and an rms dispersion of 1.3 (18%). This average line ratio compares quite well with values measured in other spiral galaxies, and is slightly smaller than the value measured in the starburst galaxies (13 ± 5, Aalto et al. 1995). The 12CO J=3-2/J=2-1 line ratios show more scatter. In particular, the line ratio obtained for NGC 604-2 (1.07) is significantly higher than the value for any other cloud. Excluding NGC 604-2, the average 12CO J=3-2/J=2-1 line ratio is 0.69 with an rms dispersion of 0.15 (21%). Thus we conclude that the 12CO J=3-2/J=2-1 line ratio is uniform across six of the seven clouds in our sample, but is significantly enhanced in NGC 604-2. This average line ratio for the six clouds in M33 is in good agreement with values measured in other starburst and normal galaxies, while higher line ratios similar to the value for the cloud NGC 604-2 are seen in the nuclei of both NGC 253 (Wall et al. 1991) and M51 (Garcia-Burillo et al. 1993).

If we compare the CO line ratios for the three clouds without H II regions (MC19, MC32, and NGC 604-4) with those for the clouds with normal H II regions (MC 1, MC 13, and MC 20), the average line ratios for the two sets of clouds agree very well. The one cloud with a large 12CO J=3-2/J=2-1 line ratio, NGC 604-2, is located in the very intense star formation environment of the giant H II region NGC 604. The difference between the two clouds (NGC 604-2 and NGC 604-4) in and around this giant H II region is striking. The projected separation of the two clouds is only 120 pc (30"), and yet their 12CO J=3-2/J=2-1 line ratios are very different. Only the cloud in the immediate vicinity of the giant H II region and its powering OB association has an unusual line ratio.

The high 12CO J=3-2/J=2-1 line ratio in NGC 604-2 may be a clue to unusual physical conditions in the molecular clouds from which the H II region NGC 604 formed, i.e. a pre-star formation difference in the physical conditions that may have caused the formation of the giant H II region. Alternatively, the high line ratio may be due to heating of the gas by the massive stars, i.e. a post-star formation change in the physical conditions in the molecular cloud. This interpretation implies that the giant H II region has a relatively small sphere of influence over which its intense radiation field can change the properties of the dense molecular gas. It appears that both an unusually intense radiation field and a cloud in close proximity to the source of the ionizing radiation are required to produce a significant change in the CO line ratios.

We used a large velocity gradient code to estimate the physical conditions in the molecular clouds in M33. To provide better constraints for the models, we include the 12CO/13CO J=1-0 line ratio observed for these clouds in a 55" beam (Wilson & Walker 1994). We performed separate fits for the average line ratios for the six clouds and for NGC 604-2. The results from the LVG analysis show that there is a significant difference in the kinetic temperature of NGC 604-2 (TK =< 100 K) compared to the other clouds (TK =< 30 K). Thus the large 12CO J=3-2/J=2-1 line ratio in NGC 604-2 is likely the result of the higher kinetic temperature of this cloud. In addition, the average solution suggests that NGC 604-2 has a higher column density by about an order of magnitude compared to the six other clouds. The molecular gas in the six clouds is quite dense (10(3) - 3x10(4) cm(-3)). The volume-averaged densities for the six normal clouds are 40 - 290 cm(-3), significantly lower than the densities derived from the LVG analysis. Thus the volume filling factor of the dense gas within the clouds is <10%. For NGC 604-2 the volume-averaged density is somewhat higher, 500 cm(-3), which combined with the somewhat lower density from the LVG model of 1 - 3x10(3) cm(-3) gives a filling factor for the dense gas of 17 - 50%.

center

Figure 1: 12CO J=2-1, 13CO J=2-1, and 12CO J=3-2 spectra for three of the seven giant molecular clouds in M33. The spectra are binned to 1 km/s resolution and scaled to the main beam temperature scale. The 13CO J=2-1 spectrum has been scaled up by a factor of 3. Note in particular the varying strength of the 12CO J=3-2 line relative to the J=2-1 line.

Thus NGC 604-2 is distinguished from the other clouds by a larger column density, volume filling factor of dense gas, and kinetic temperature. The kinetic temperature of the gas is probably the physical parameter most easily affected by the presence of the intense burst of star formation in the giant H II region. The gas column density could be increased through shock-initiated merging of two or more molecular clouds or through a partial collapse of the molecular cloud due to an increase in the external gas pressure. However, the molecular column density could also be decreased through photo-dissociation of the molecular gas and thus it is difficult to predict the net effect on the column density from the formation of the giant H II region. It is tempting to speculate that the higher column density and filling factor of the dense gas in NGC 604-2 (or clouds like it that have since been destroyed) could have played a role in initiating the burst of star formation that formed the giant H II region.

The uniformity of the line ratios of the six molecular clouds observed in the normal disk of M33 suggests that similar observations of ensembles of molecular clouds in more distant galaxies are likely to produce meaningful measurements of the average physical conditions of the molecular gas. The relatively small sphere of influence (100 pc) of the giant H II region NGC 604 suggests that in normal galaxies only the most intense star forming regions may produce significant changes in the molecular gas. The change in the line ratios is likely to be measurable only in relatively nearby galaxies (< 10 Mpc), where the warm molecular gas in the H II region is not diluted by emission from cooler gas included in the beam. It would be interesting to test whether similarly uniform line ratios are observed in individual clouds in the intense ultraviolet field of a starburst galaxy, but such observations must await the construction of an imaging submillimeter interferometer, such as the Smithsonian Submillimeter Array.

References

Aalto, S., Booth, R. S., Black, J. H., & Johansson, L. E. B., 1995, A&A, 300, 369

Garcia-Burillo,S., Guelin, M., & Cernicharo, J., 1993, A&A, 274, 123

Wall, W. F., Jaffe, D. T., Bash, F. N., & Israel, F. P., 1991, ApJ, 380, 384

Wilson, C. D., & Walker, C. E., 1994, ApJ, 432, 148

C. D. Wilson, McMaster University

C. E. Walker, University of Arizona

M. D. Thornley, University of Maryland


JACH | JCMT | UKIRT | Computer Services | Local server

Last Modification Date 1996/08/13 - Last Modification Author: gdw
Graeme Watt (gdw)
Contact: Jonathan Kemp. Updated: Tue Aug 17 17:32:19 HST 2004

Return to top ^