Christopher M. Hayden (eMail)
Washington College of Chestertown, MD

REU Program-Summer 2004
Univ. of Wisconsin - Madison
Madison, WI 53706


A Mid-Infrared Study of Protostar

Development in the Massive HII Region RCW79

Introduction

    For years, scientists have been studying regions of dense ionized hydrogen gas (HII regions) throughout the interstellar medium. One of the most intriguing features of these massive star forming objects (SFOs) is the presence of many possible protostars in several different evolutionary stages of development. The life cycle of stars, while understood conceptually, is not very well understood theoretically, particularly the period of time early in the life of a young star during which it is accreting matter. Studying protostars in HII regions promises to provide the scientific community with a much better understanding of the dynamics of star formation, as well as increased comprehension of the galactic structure as a whole.

    The ongoing Galactic Legacy Infrared Mid-Plane Survey Extraordinaire (GLIMPSE)1 project is a survey of two-thirds of the southern galactic plane in four mid-infrared wavelength bands utilizing the Infrared Array Camera (IRAC)2 onboard the Spitzer Space Telescope (SST).3 With a resolution of less than two seconds of arc, and very high sensitivities, the data garnered by this survey promises to be some of the best available to the scientific community to date, perhaps equaled only by the Two Micron All Sky Survey (2MASS) project that occurred a few years ago. Utilizing various combinations of 2MASS data along with GLIMPSE data allows for viewing of HII regions in many different wavelength combinations, thus making the identification of protostars simpler and more accurate in many stages of their development. It is very conceivable that this data could provide the scientific community with a much more complete understanding of star formation in the coming years.

    An HII region of particular interest is that of RCW79, located at galactic coordinates 308.647 +0.579.4 It is located approximately 5.31kpc away5, assuming that our solar system has a galactocentric radius of approximately 8.5kpc, and possesses a calculated radial velocity of approximately -50km/s4, a temperature of approximately 103.14K4, and a size of approximately 0.25x0.39 degrees. The optical visibility of RCW79 indicates that is is extremely luminous, and the GLIMPSE survey has indicated that there is a very dense ring of dust and ionized hydrogen gas surrounding a dense central cluster of stars. Excitation of the cloud may be related to radiation emitted by the central cluster of stars, and it appears that a portion of the northern section of the ring has been blown out. Preliminary indications are that there are a large number of reddened stars with infrared excesses in this HII region, many of which may be hidden within the gas itself, suggesting the possibility of ongoing star formation. Point source lists for the RCW79 region have been compiled, and the goal of this project is to perform appropriate point source analyses. These analyses include the use of color-color and color-magnitude diagrams, a study of the relationship between radius from the central cluster and the density of reddened stars, a study of the diffuse emission, and studies of the spectral energy distributions (SEDs) of some of the reddened stars. A discussion of the indications of the results regarding current star formation concludes.

Fig. 1a: RCW79 colored image in the 3.6, 5.8, and 8.0 micron bands.

Fig. 1b: RCW79 colored image in the 3.6, 4.5, and 8.0 micron bands.

Fig. 1c: RCW79 rainbow image in the 8.0 micron band showing the diffuse emission. Emission scales from low in blue to high in red.

Fig. 1d: RCW79 in the 3.6 micron band. Note the bubble-like object at the lower center and the apparent burst of star formation in it.

Fig. 1: Various mosaic images of RCW79 and surrounding region. Click each image to see a larger version.

Observations

    The region of space containing the RCW79 HII complex was imaged two times with 1.2s exposures in each of the four IRAC bands. The data was first processed by the Spitzer Science Center, and point sources were then extracted at the University of Wisconsin-Madison. Positional accuracies are better than on arc-second3, and point source full-width-half-max resolutions are less than two seconds of arc in each of the four bands. A draft of the GLIMPSE Point Source Archive was compiled in the region of RCW79, and from this archive only those point sources within a 0.421 degree radius that were detected at least twice in one band and at least once in an adjacent band were considered for data analysis. Additionally, the standards of B.A. Whitney et al.6 were adopted to further reduce the data by considering only those point sources brighter than at least two of the following magnitudes: 13.0, 12.5, 11.7, and 11.7 magnitude in bands one, two, three, and four, respectively. To be included in any of the color-color and color-magnitude diagrams, a point source must have satisfied the magnitude criteria for all bands utilized in generating the diagram. The criteria used to define a reddened source were also adopted from B.A. Whitney et al.6 such that a source must have satisfied at least one of the following criteria to be considered reddened: [3.6]-[4.5]> 0.2 mag, [4.5]-[5.8]> 0.35 mag, or [5.8]-[8.0]> 0.3 mag. After beginning with roughly 235,000 total point sources, the data was reduced to 15,239 point sources satisfying the magnitude cuts and 981 point sources satisfying the definition of a reddened object. No flux calibrations were done.

Results

Fig. 2a: Point source plot for RCW79 and surrounding region.

Fig. 2b: Point source plot for field region.

Fig. 2: Sky plots of RCW79 and the field region. The black dots represent normal point sources while the red dots represent the reddened point sources. The central cluster is visible in the RCW79 image here approximately 308.70 +0.60. Click the images for larger versions.

    Figure 2 displays plots of the point source positions within 0.421 degrees of radius of the central position of RCW79 and the surrounding region as well as for a field region of equal size centered directly south of RCW79 at galactic coordinates 308.647 -0.579. Looking at the images, it is easy to pinpoint the central cluster of RCW79 as a dense region of red point sources located at approximately 308.70 +0.60. It is also quite obvious that the density of red point sources decreases to some degree with increasing radius from this central position, as would be expected for an HII region with ongoing star formation. The distribution of red stars in the field region appears significantly more uniform, having no dense and centralized HII region, though there does appear to be an increase in the frequency of red point sources in the northwestern portion of the image. One disturbing issue is that there are an excessive number of red point sources in the field region compared to what would be expected. In fact, of the approximately 18,000 point sources analyzed in the field region, 1,400 appear reddened. This number is approximately three halves the number of red point sources in the region of RCW79 despite the fact that there was approximately only a twenty percent increase in the number of analyzed point sources. Possible reasons for this will be discussed later.

Fig. 3a1: Color-color plot of RCW79 [4.5]-[5.8] vs. [3.6]-[4.5]

Fig. 3b1: Color-color plot of RCW79 [5.8]-[8.0] vs. [4.5]-[5.8]

Fig. 3c1: Color-color plot of RCW79 [3.6]-[4.5] vs. [5.8]-[8.0]

Fig. 3a2: Color-color plot of field region [4.5]-[5.8] vs. [3.6]-[4.5]

Fig. 3b2: Color-color plot of field region [5.8]-[8.0] vs. [4.5]-[5.8]

Fig. 3c2: Color-color plot of field region [3.6]-[4.5] vs. [5.8]-[8.0]

Fig. 3: Color-color plots of the stars in the region of RCW79 with corresponding color-color diagrams from the field region below. Black dots in these images represent normal main sequence stars while red dots represent reddened stars. Click each image to see a larger plot.

    Figure three displays color-color diagrams for RCW79 at the top and the corresponding color-color diagrams for the field region to the bottom. The black dots are representative of regular point sources while the red dots indicate those that appear to have infrared excesses. Comparing these diagrams to those of B.A. Whitney et al.6 reveals a very similar structure in the plots for the RCW49 and RCW79 regions. The direction or reddening is in general quite obvious, and as such it has been omitted from the diagrams in favor of adopting the vectors from B.A. Whitney's analysis of RCW49. Comparing the diagrams of point sources in RCW79 to those of point sources in the field region, no perceptible difference can be seen, though data analysis has not been carried out to verify this; however, it would be somewhat consistent considering the shear number of red point sources being observed in the field region.

Fig. 4a1: Color-magnitude diagram of RCW79 [3.6] vs [3.6]-[4.5]

Fig. 4b1: Color-magnitude diagram of RCW79 [4.5] vs [4.5]-[5.8]

Fig. 4c1: Color-magnitude diagram of RCW79 [5.8] vs [5.8]-[8.0]

Fig. 4d1: Color-magnitude diagram of RCW79 [8.0] vs [3.6]-[8.0]

Fig. 4a2: Color-magnitude diagram of field region [3.6] vs [3.6]-[4.5]

Fig. 4b2: Color-magnitude diagram of field region [4.5] vs [4.5]-[5.8]

Fig. 4c2: Color-magnitude diagram of field region [5.8] vs [5.8]-[8.0]

Fig. 4d2: Color-magnitude diagram of field region [8.0] vs [3.6]-[8.0]

Fig. 4: Color-magnitude diagrams for RCW79 and the surrounding region with corresponding plots from the field region below. The dots follow the same description as in figure 3. Click each image to see a larger plot.

    Figure 4 displays color-magnitude diagrams for the RCW79 region at the top and corresponding color-magnitude diagrams for the field region below. As with the color-magnitude diagrams above, there is no perceptible difference between the point sources in the region of RCW79 and the field region, but, as before, data analysis has not been done to verify this. Comparing these diagrams to those produced by B.A. Whitney et al.6 for RCW49 there is again a striking similarity, suggesting at least a certain degree of precision in the data.

Fig. 5a1: Logarithmic plot of the fraction of red point sources per unit area as a function of radius.

Fig. 5b1: Logarithmic plot of the fraction of point sources that are red as a funtion of radius.

Fig. 5a2: Linear plot of the fraction of red point sources per unit area as a function of radius.

Fig. 5b2: Linear plot of the fraction of point sources that are red as a funtion of radius.

Fig. 5: Plots of reddened point sources as a function of radius from the central cluster. The equations of the best fit curve for each plot, along with one sigma errors, are shown below. Also plotted on each plot is the same analysis of red point sources from the field region. The lines corresponding to 1/x, 1/x2, and 1/x3 are plotted on the logarithmic plots for reference as well. Click each plot to see a larger version.

Eq. 5a: y=10-3.666350.0286773*x-0.6478700.0315674 Eq. 5b: y=10-1.527970.0346948*x-0.5191950.0381913

    Figure 5 displays two different plots, each displayed in logarithmic and linear form, that show the density of red stars as a function of radius from the central position of the RCW79 object. The first plot displays the fraction of red point sources that lie in a given annulus per unit area in the annulus. The second plot displays the fraction of point sources lying in a given annulus that appear reddened. In each plot there is a very obvious decrease in the density of red point sources as the radius from the central position increases. This is in accordance with what would be expected if star formation was occurring in a dense and circular HII region such as RCW79, and is also consistent with the results of the analysis performed on RCW49 by B.A. Whitney et al.6 Also shown by the dotted lines on the logarithmic plots is the corresponding distribution of red point sources in the field region. While the color-color and color-magnitude diagrams indicate almost no perceptible difference between RCW79 and the field region, here is the first real evidence that there is in fact a significant difference. Unlike in the RCW79 region, the distribution of red point sources in the field region is almost perfectly uniform, which is precisely what would be expected. This has been mathematically verified, as the best fit line for corresponding logarithmic plots of the distribution of red point sources in the field region has a slope of less than 0.01 in each case with a one sigma error that is only a fraction of that. Despite the fact that the distribution is nearly uniform, it is still seen that the density of red point sources in the field region is still higher than that of the RCW79 region. This is consistent with what would be expected considering that there are significantly more red point sources in the field region, and will be discussed in further detail in the conclusion.

Fig 6a1: Logarithmic plot of diffuse emission in channel 1 in RCW79 and field region.

Fig 6b1: Logarithmic plot of diffuse emission in channel 2 in RCW79 and field region.

Fig 6c1: Logarithmic plot of diffuse emission in channel 3 in RCW79 and field region.

Fig 6d1: Logarithmic plot of diffuse emission in channel 4 in RCW79 and field region.

Fig 6a2: Linear plot of diffuse emission in channel 1 in RCW79 and field region.

Fig 6b2: Linear plot of diffuse emission in channel 2 in RCW79 and field region.

Fig 6c2: Linear plot of diffuse emission in channel 3 in RCW79 and field region.

Fig 6d2: Linear plot of diffuse emission in channel 4 in RCW79 and field region.

Fig. 6:  Plots of the diffuse emission as a function of radius from the central cluster. There are two plots for each wavelength band, one logarithmic above and one normal below, each of which has best fit curves for the portion of the plot during which the emission is falling, as well as an overplot of the diffuse emission in the field region for the corresponding channel. The equations to these best fit power equations are shown on the plots and with one sigma errors below. The logarithmic plots also have the lines for 1/x, 1/x2, and 1/x3 plotted for reference. Click each image to see a larger plot.

Eq. 6a: y=1010.61452.74576 * x-3.686830.979979 + 4.35720 Eq. 6c: y=1010.74310.398605 * x-3.383130.134315 + 9.71742
Eq. 6b: y=109.389212.05353 * x-3.267750.732919 + 2.44167 Eq. 6d: y=1010.93170.161297 * x-3.281430.0543511 + 21.3914

More results here.

Fig. 7a: Mid-IR spectral energy distribution for a red star from the RCW79 region.

Fig. 7b: Contour plot produced by model fitter for a red star in the RCW79 region.

Fig. 7c: Close-up of the star analyzed in the model.

Fig. 7d: Relative position of the star analyzed in the model.

Fig. 7: Output from the model fitter for a red star in the RCW79 region. The left plot shows a spectral energy distribution normalized to the channel 4 flux along with the best fit model curve. The right plot is a contour plot showing the most likely evolutionary class and temperature of the star. Darker regions indicate a better fit. The bottom two plots show the star being analyzed. The left image shows a close-up image of the star, which is enclosed in a green circle. The left image shows the position of the star relative to the whole RCW79 complex. Click each plot to see a larger version.

Conclusion

References

  1. Benjamin, R.A., et al. 2003, PASP, 115, 953
  2. Fazio, G., et al. 2004, ApJS, unpublished at the time of this work.
  3. Werner, et al. 2004, ApJS, unpublished at the time of this work.
  4. Caswell, J.L. and Haynes, R.F. 2004, AA, 171, 261
  5. Derived using a calculation based on the Brandt-Blitz model of galactic rotation.
  6. Whitney, B.A., et al. 2004, ApJS, unpublished at the time of this work.