2MASS Observations of the Perseus, Orion A, Orion B, and Monoceros R2 Molecular Clouds


John M. Carpenter
(jmc@astro.caltech.edu)

Department of Astronomy
California Institute of Technology
MS 105-24
Pasadena, CA 91125




Abstract

We use the 2MASS Second Incremental Release Point Source Catalog to investigate the spatial distribution of young stars in the Perseus, Orion A, Orion B, and MonR2 molecular clouds. After subtracting a semi-empirical model of the field star contamination from the observed star counts, stellar surface density maps are used to identify compact clusters and any stellar population found more uniformly distributed over the molecular cloud. Each cloud contains between 2 to 7 clusters, with at least half of the cluster population found in a single, rich cluster. In addition, a distributed stellar population is inferred in the Orion A and MonR2 molecular clouds within the uncertainties of the field star subtraction with a surface density between 0.013 - 0.083 arcmin**-2. Sensitivity calculations suggest, however, that the number of stars in the distributed population may be underestimated by a factor of 2 or more if stars have been forming with a Miller-Scalo IMF at a constant star formation rate for longer than 10 Myr. After considering the possible evolutionary status of the distributed population, the global star formation efficiency implied by the sum of the distributed and cluster populations ranges between 1-9% among the four clouds. The fraction of the total stellar population contained in clusters for the nominal extinction model ranges from ~50-100% if the distributed population is relatively young (< 10 Myr), to ~25%-70% if it is relatively old (~100 Myr). The relatively high fraction of stars contained in clusters regardless of the age of the distributed population, in conjunction with the young ages generally inferred for embedded clusters in nearby molecular clouds, indicates that a substantial fraction of the total stellar population in these regions has formed within the past few million years in dense clusters. This suggests that either the star formation rate in each these clouds has recently peaked if one assumes clouds have ages > 10 Myr, or molecular clouds are younger than typically thought if one assumes that the star formation rate has been approximately constant in time.

[paper at ADS] [ reprint]



Figures
(click on thumbnails to see enlarged versions)


Figure 1:
A Ks band stellar surface density map for point sources in the 2MASS Second Incremental Release for the area between Galactic longitudes of 130 deg and 250 deg and Galactic latitudes of -40 deg and +40 deg. The map was produced by binning stars with magnitudes 6.0 <= m(Ks) <= 14.3 in 5' x 5' pixels. The map is presented in the Hammer-Aitoff projection with a square-root image stretch. Darker gray scales denote higher stellar surface densities, with the surface densities ranging from ~2 arcmin**-2 in the Galactic Plane to ~0.2 arcmin**-2 at higher Galactic latitudes. The white regions represent tiles not included in the Second Incremental Release, or in less than 1% of the instances, tiles not meeting the sensitivity criteria adopted for this paper (see text). The location of several well known molecular clouds are indicated.


Figure 2:
Image of the average J-Ks stellar color over the same region shown in Figure 1. The darker gray scales represent larger J-Ks colors and indicate regions where background field stars and embedded young stars have been reddened by dust in molecular clouds. The Taurus, Perseus, Orion A, Orion B, and MonR2 molecular clouds are clearly visible, as well as numerous molecular clouds within the Galactic Plane.


Figure 3:
The observed mean stellar surface density as a function of Galactic latitude (top panels), the mean residuals after subtracting a polynomial fit to the observed star counts (middle panels), and the RMS of the residuals (bottom panels) at J, H, and Ks band. Only regions between Galactic longitudes of 130 deg and 250 deg and latitudes of -35 deg and -7 deg that did not contain a stellar cluster or a molecular cloud are shown here and included in the polynomial fit. The fit was performed as a function of both Galactic longitude and latitude, but are averaged over longitude for presentation in this figure. For reference, the Perseus, Orion A, Orion B, and MonR2 molecular clouds analyzed in this paper are located between latitudes of -20 deg and -10 deg.


Figure 4:
Upper left An adaptive kernel, Ks band stellar surface density map of the Perseus molecular cloud for stars with magnitudes of 6.0 <= m(Ks) <= 14.3. Upper right The IRAS 60 micron image displayed in a logarithmic stretch. Lower left An image of the average J-Ks color for stars observed by 2MASS. Lower right A map of the integrated 13CO(J=1-0) intensity map (Padoan etal. 1999). In each panel, darker halftones represent the higher intensities. In the Ks density map and the average J-K color image, the white strips are tiles not included in the 2MASS Second Incremental Release, and the white ``crosses'' are regions around bright stars that were masked out when generating the Second Incremental Release Point Source Catalog. The labels indicate the location of either prominent star forming regions or stellar clusters identified in the \KB\ surface density image (see Table 2).


Figure 5:
Same as in Figure 4, except for the Orion A (L1641) molecular cloud. The 13CO image is from Bally etal. (1987).


Figure 6:
Same as in Figure 4, except for the Orion B (L1630) molecular cloud. The 13CO image is from Miesch & Bally (1994).


Figure 7:
Same as in Figure 4, except for the MonR2 molecular cloud. The 13CO image is from Miesch & Bally (1994). The sources labeled VDB are reflection nebula cataloged by van den Bergh (1966), and sources labeled GGD are from the list of Herbig-Haro objects noted by Gyullbudaghian, Glushkov, & Denisyuk (1978).


Figure 8:
Contour maps of the Ks stellar surface density for each of the clusters summarized in Table 2 and labeled in Figures 4-7. Each contour map is centered on the cluster coordinates listed in Table 2. The angular size of the contour maps varies depending on the dimensions of the cluster. The contour levels in each map begin at 2 sigma above the field star stellar surface density, where sigma is the noise per pixel in the field surface density map estimated assuming Poisson statistics. The contour intervals are 10sigma for IC348, ONC, NGC2024, GGD12-15, and MonR2, and 3sigma for the remaining clusters. The approximate noise level in field star surface density is ~0.1 arcmin**-2 for IC348, ~0.15 arcmin**-2 for clusters in Orion, and ~0.3 arcmin**-2 for clusters in MonR2. Note that part of the cluster area in IC348, L1641C, NGC2024, and the ONC has been masked in the 2MASS Second Incremental Release Point Source Catalog to eliminate artifacts from bright stars. Also, the western portion of the NGC2068 cluster has not been imaged with 2MASS at the time of this study.


Figure 9:
J-H vs H-Ks color-color diagram for each cluster identified in the Ks band stellar density maps. Only stars that have magnitudes of 6.0 <= m(Ks) <= 14.3 and photometric uncertainties less than 0.1 mag in all three bands are shown. The solid curves show the locus of unreddened main sequence and giant stars in the CIT system (Bessell & Brett 1988), and the dashed lines show the reddening vectors from Cohen etal. (1981). Each cluster contains a number of red objects, supporting the notion that the clusters are indeed embedded within the molecular clouds.


Figure 10:
Histogram of the J, H, and Ks stellar surface densities observed toward the Perseus molecular cloud after subtracting the nominal field star model. The open histograms are for all lines of sight toward the Perseus molecular cloud as defined by 13CO(J=1-0) emission. The hatched regions represent lines of sight outside the cluster boundaries but within the cloud area, and represent the surface density distribution of the distributed population. The mean surface density of the distributed population as inferred from the hatched histogram is summarized in Table 3.


Figure 11:
Same as Figure 10, except for the Orion A (L1641) molecular cloud.


Figure 12:
Same as Figure 10, except for the Orion B (L1630) molecular cloud.


Figure 13:
Same as Figure 10, except for the MonR2 molecular cloud.


Figure 14:
The visual extinction (top panel) and J, H, and Ks band stellar surface density (bottom panels) as a function of Galactic longitude for the Perseus molecular cloud. The field stars have been subtracted from the star counts using the nominal extinction model, and the results averaged over Galactic latitude. The solid horizontal line shows the average surface density of the distributed population for the nominal extinction model, and the dashed horizontal lines show the inferred surface density assuming the low and high extinction models. This figure shows that the tightest constraints on the surface density of the distributed population is provided at Ks band where the field star subtraction is less sensitive to the assumed extinction model.


Figure 15:
Same as in Figure 15, but for the Orion A molecular cloud.


Figure 16:
Same as in Figure 14, but for the Orion B molecular cloud.


Figure 17:
Same as in Figure 14, but for the MonR2 molecular cloud.


Figure 18:
Ks band iso-magnitude contours as a function of stellar age and mass for stars at the distance of Perseus (320 pc), Orion A and Orion B (480 pc), and MonR2 (830 pc). The magnitudes were computed using the D'Antona \& Mazzitelli (1997, 1998) pre-main-sequence evolutionary tracks assuming Av = 0 and that no near-infrared excess emission is present. The shaded area highlights the parameter space that is probed for the adopted Ks magnitude thresholds (6.0 <= m(Ks) <= 14.3).


Figure 19:
The fraction of a model stellar population with magnitudes of 6.0 <= m(K) <= 14.3 as a function of the molecular cloud age for the Perseus, Orion A/B, and MonR2 molecular clouds. The model assumes that stars have been forming at a constant at rate in time with a Miller-Scalo Initial Mass Function over the mass range of 0.08-10 Mo and that the visual extinction is the average value inferred from the 13CO maps (Av~3). The magnitudes were computed using the D'Antona \& Mazzitelli (1997, 1998) pre-main-sequence evolutionary tracks for objects with masses up to 3 Mo, and assuming main sequence magnitudes for the more massive stars. The results indicate that the fraction of the model stellar population within the adopted Ks magnitude thresholds for ages < 100 Myr ranges from > 67% at the distance of Perseus to > 26% at the distance of MonR2.


Figure 20:
The predicted surface density of stars with apparent magnitudes of 6.0 <= m(K) <= 14.3 as a function of the molecular cloud age for the Perseus, Orion A, Orion B, and MonR2 molecular clouds for the model calculations described in Figure 19. The solid curves are the predicted stellar surface density for an assumed star formation efficiency of 1% (bottom curves in each panel), 2%, 5%, and 8% (top curves in each panel) using the cloud masses implied by the nominal extinction model. The horizontal dashed line shows the inferred surface density for the distributed population (if a positive value; see Table 3), and the horizontal dotted line shows surface density for the clusters averaged over the entire cloud area. The older ages are intended to apply only to the distributed population. The global star formation efficiency implied by the sum of the cluster and distributed population ranges between ~2% and 9% for the four clouds depending on the age assumed for the distributed population.


Figure 21:
The fraction of the total stellar population currently contained in clusters as a function of the cloud age assuming that only part of the distributed population is detected at a given age as indicated by the model calculations shown in Figure 19. The solid lines show the fraction of stars in clusters for the nominal extinction model, and the dotted lines represent the low and high extinction models.