Journal of Undergraduate Research
Volume 3, Issue 5 - January 2002

Studying the formation of stars in the Milky Way: A Comparison between Molecular Cloud Structure and its Embedded Stellar Content

Helene Flohic

ABSTRACT


The purpose of this research is to investigate the conditions that give rise to the formation of stars, like our Sun, in the Milky Way. Using an infrared (IRAS) survey of young stars and a radio (FCRAO) survey of CO gas, we have compared the properties of molecular clouds with the stars that they contain. Clouds were outlined from images of molecular gas by eye using a cursor routine and then categorized according to shape and compactness. The mass and column density (the number of H2 molecules per cm2 along the line of sight) were determined and compared with the total luminosity of the IRAS sources within the cloud. We find that the luminosity of the IRAS sources increases with the mass of the cloud (bigger clouds form more stars). We also find that, for a given mass, compact clouds contain more stars than diffuse clouds but there is no statistical difference between the star forming capability of filamentary and circular clouds. We conclude that the conditions that give rise to star formation are determined on small scales and not by the large-scale features of the cloud.

Our Sun, and all the stars that we see in the night sky, were born in molecular clouds. So called because they are 90% molecular hydrogen, such clouds continue to form stars today. Radio telescopes can observe spectral lines from the molecules within the clouds and map out their shape, velocity, density, and composition.

One way to study the dependence of star formation on the molecular cloud properties is to compare the young stellar content of clouds with different physical conditions. In this paper, we compare a catalog of molecular clouds from the Five College Radio Astronomy Observatory (FCRAO) survey of the outer Galaxy with a catalog of young stars from the Infrared Astronomical Satellite (IRAS).

DATA

Radio Observations

The most abundant molecule in molecular clouds is H2 but this is generally not detectable at the cool temperatures (10-20 K) of the clouds. Rather, astronomers observe millimeter wavelength spectral lines from CO and use this as a tracer of H2.

The FCRAO surveyed the CO J=1-0 emission at 2.6mm wavelength from a 336 deg_ region (40° X 8°.4) of the second quadrant of the Galaxy. The survey consists of nearly 1.7X106 spectra of the region between 102.5° and 141.5° of galactic longitude and -3.0° and 5.5° of galactic latitude (Heyer et al. 1998).

The entire data cube of the FCRAO CO survey is available electronically from the NASA Astronomical Data Center (http://adc.gsfc.nasa.gov).

Infrared Observations

The youngest stars remain embedded in the clouds from which they formed. The dust in these clouds absorbs nearly all optical luminosity, and reradiates it at near-infrared to submillimeter wavelengths, 1-300 mm (Myers 1994).

Most of the far-infrared radiation (wavelengths longer than about 40 mm) does not penetrate to the ground and can only be detected from above the Earth's atmosphere from high altitude airplanes, balloons, rockets, or spacecraft. The Infrared Astronomical Satellite was launched in 1983 in order to survey the sky in four infrared bands at 12, 25, 60 and 100 um. It detected over 200,000 infrared sources. These form what is known as the IRAS Point Source Catalog and is also available at the Astronomical Data Center.

We selected those IRAS point sources with infrared colors characteristic of young stellar objects (Wouterloot & Walmsley 1986).

ANALYSIS

We first split the FCRAO survey into 44 boxes of 3.0° to 4.0° edges and outlined cloud boundaries at a consistent CO intensity by eye using a cursor-based routine.

In this study, we defined 57 clouds, varying in shapes, density and star formation efficiency. For most of the clouds the distance or the velocity could be found from previous articles. If neither the distance nor the velocity was known from references the velocity of the cloud was determined using spectra of the FCRAO survey (Heyer et al. 1998) the distance calculated using the flat rotation curve of the Galaxy.

Next cloud statistics were calculated: the mass of the cloud, the total luminosity of the embedded sources (Margulis et al. 1989), the average column density of H2 (the number of H2 molecules per cm2 along the line of sight) and the peak value of the column density.

Clouds were then classified according to different criteria. The first criterion was the shape: the cloud was either circular or filamentary. The distinction arose from the elongation of the cloud. Figure 1.a shows a filamentary cloud and figure 1.b shows a circular cloud. The second categorization of the clouds is based on the ratio of the peak column density to the mean value of the column density. If the ratio is less than 11.5 it means that the cloud has a fairly uniform density and is classified as "diffuse". In the other case, the cloud is classified as "compact". Figure 1.c shows a diffuse cloud while figure 1.d shows a compact cloud.

Figure 1a. A filamentary molecular cloud

Figure 1b. A circular molecular cloud

Figure 1b. A circular molecular cloud

Figure 1c. A diffuse molecular cloud

Figure 1d A compact molecular cloud

Figure 1. Images of different kinds of molecular clouds. The grayscale shows the CO emission, the white circles show the location of the IRAS sources and the solid white line materializes the cloud boundaries.

Lastly we are interested in the star formation efficiency of the clouds. This is measured by the ratio of the luminosity of embedded sources to the mass of the cloud. An efficient star-forming cloud has a larger number of stars within it compared to other clouds with similar mass and therefore a high luminosity to mass ratio. In this paper a cloud is considered efficient in star formation if its luminosity to mass ratio is greater than 0.3.

RESULTS

General Considerations

The luminosity of the young stellar objects increases with the peak value of the hydrogen column density. In figure 2 the luminosity of the IRAS sources is plotted against peak column density on a log-log plot. The data have been split into two bins depending on whether they have peak column densities greater than or less 1022. This threshold is marked by the dotted line on the plot. Average values and dispersions for each bin are indicated by the large solid points and show that the higher peak column density clouds have a higher luminosity than the lower peak column density clouds.

Figure 2. Luminosity of the IRAS sources compared to the peak column density of the molecular clouds.

Figure 2. Luminosity of the IRAS sources compared to the peak column density of the molecular clouds. The dashed line is at peak N H2 = 1022 (molecules per cm_). The two black dots are at the average peak N H2 and LIRAS values on both sides of the dashed line. The crosses show the dispersion of the data.

The luminosity of the young stars also increases with the mass of the molecular cloud. In Figures 3 and 4, we plot the luminosity versus mass on a log-log plot and make a least squares fit to the data. The slope of the fitted lines is almost equal to 1 in each case, which implies the luminosity of the IRAS sources within a cloud is, on average, directly proportional to its mass.

Figure 3. Luminosity of the IRAS sources compared to the mass of the molecular cloud for both compact and diffuse clouds.

Figure 3. Luminosity of the IRAS sources compared to the mass of the molecular cloud for both compact and diffuse clouds. The two sets of data are approximated by the least square fit shown by the solid and the dashed lines.

Figure 4. Luminosity of the IRAS sources compared to the mass of the molecular cloud for both filamentary and circular clouds.

Figure 4. Luminosity of the IRAS sources compared to the mass of the molecular cloud for both filamentary and circular clouds. The two sets of data are approximated by the least square fit shown by the solid and the dashed lines.

Compact Versus Diffuse

Compact clouds form stars more efficiently than diffuse clouds do. In figure 3 the logarithm of the luminosity is plotted as a function of the logarithm of the mass for both diffuse and compact clouds. The fitted line is obtained using the least square fit method. The slope of the fitted line for compact clouds is 0.95 and that for diffuse clouds is 0.91. The slopes for both cases are very similar and close to 1. This shows the luminosity to mass ratio is approximately constant, but the ratio is higher for compact clouds (L/M=1.37) than for diffuse clouds (L/M=0.26) (see table 1). We conclude that compact clouds have a higher star formation efficiency.

 

Table 1
Distribution of the molecular clouds into the different categories and the average luminosity to mass ratio within each one
Type Circular Filamentary Total
Compact N=9
<L/M> = 1.40
N=6
<L/M> = 1.33		 N=15
<L/M> = 1.37
Diffuse N=33
<L/M> = 0.29		 N=9
<L/M> = 0.16 		N=42
<L/M> = 0.26
Total N=42
<L/M> = 0.53 		N=15
<L/M> = 0.63 		N=57
<L/M> = 0.55

 

This result is confirmed by a numerical analysis of the data. Out of the 57 clouds studied, 15 are compact clouds with 13 of them having a luminosity to mass ration greater than 0.3. So 86.7% of the compact clouds are efficiently forming stars. Only 11 of the 42 diffuse clouds are forming stars, which corresponds to 26.1%. So star formation occurs in compact clouds rather than in diffuse clouds.

Round Versus Filamentary

We did not find any significant difference in the stellar content of circular and filamentary clouds. Figure 4 is the plot of the luminosity of the IRAS sources as a function of the mass of the cloud on a logarithmic scale for each type of cloud. The least square slopes are again similar to 1 indicating a constant luminosity to mass ratio but this time there is no clear offset between the two groups. That is, for a given cloud mass, the luminosity of the IRAS sources is about the same for a circular cloud and a filamentary cloud and we conclude that they form stars with approximately equal efficiency.

Again a numerical analysis confirms this. The 57 clouds consisted of 15 filamentary clouds and 42 circular clouds. 6 filamentary clouds are forming stars, which is 40% of the filamentary clouds. 18 circular clouds are forming stars, which is 42.8% of the circular clouds. So the star formation efficiency does not depend on the shape of the molecular cloud.

CONCLUSION

Our results show that the more material there is to form stars, the more stars are formed, on average. However, there are patterns within this broad picture: stars are preferentially found in clouds that have a high contrast between their peak and average column density, but the shape of the cloud, filamentary or circular, does not play a discernable role in determining a cloud's star forming capability. We conclude that the conditions that give rise to star formation are determined on small scales and not by the large scale features of a cloud.

To pursue this work further, I plan to use the travel budget of the USP grant to conduct observations of the denser regions of a star forming cloud using the FCRAO radio telescope in Amherst, MA this fall.

REFERENCES

  1. Heyer, M. H., Brunt, C., Snell, R. L., Howe, J. E., Schloerb, F. P., & Carpenter, J. M. 1998, ApJS, 115, 241
  2. IRAS Point Sources Catalog, Version 2.0, 1986, Joint IRAS Science Working Group (IPAC 1986)
  3. Margulis, M., Lada, C.J., Young, E.T. 1989, ApJ, 345, 906
  4. Myers, P. C. 1995, in "Molecular Clouds and Star Formation", Proceedings of the Seventh Guo-Shoujing Summer School on Astrophysics, Wuxi, China, ed. C. Yuan (Singapore: World Press)
  5. Wouterloot, J.G.A., & Walmsley, C.M. 1986, A&A, 168, 237
  6. This research has made use of the SIMBAD database, operated at CDS, Strasbourg, France

--top--

Back to the Journal of Undergraduate Research