Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 358, 521-534 (2000)

Previous Section Next Section Title Page Table of Contents

1. Introduction

How many regions of massive star formation are there in the galactic disk and how are they distributed? Do all these regions have similar luminosities? Dust extinction prevents optical observations of OB associations farther than 3 kpc from the Sun. An alternative way to study the population of recently formed OB stars in the Galaxy is to use observations at longer wavelenghts in the far infrared (FIR), where the ISM is more transparent. Ultraviolet light from OB stars, born in the dense dusty cores of giant molecular clouds, ionizes the closeby gas creating ultracompact (UC) H II regions and heating the surrounding dust, which reradiates the energy mostly in the FIR. The star-forming region becomes thus detectable by IRAS, generally as a point-like source. But which of the many IRAS point-like sources in the galactic plane are heated by embedded massive stars? The FIR spectral characteristics of UC H II regions identified with the VLA have been studied by Wood & Churchwell (1989a), who calibrated their FIR colours as a tool for their identification. Applying their criterion to the IRAS point-like sources in the galactic plane, they found 1646 sources which, according to their IRAS 60 µ/12 µ and 25 µ/12 µ flux ratios, are candidates to be sites of OB star formation containing UC H II regions (Wood & Churchwell 1989b; WC89).

Millimeter-wave emission lines from dense gas cores, where OB stars form, can be used to obtain their kinematic distances and derive the FIR luminosity of their associated dust, heated by the embedded stellar sources. The [FORMULA] line has been surveyed toward IRAS point-like sources in the outer Galaxy (Wouterloot & Brand 1996; and references therein), using less restrictive FIR colors criteria. Within the solar circle, however, where the majority of massive star formation takes place, there are several molecular clouds at different distances for most lines of sight, yielding complex CO velocity profiles. The [FORMULA] line, affected by large optical depths, will tell us little about the dense interior of molecular clouds. The [FORMULA] and [FORMULA] lines, of intermediate and low opacity in the galactic plane, are still sensitive to molecular gas column density, which can be enhanced not only toward dense cores but also toward extended regions of intermediate density.

One of the best specific tracers of dense gas is the CS molecule. Its [FORMULA] rotational line, at a frequency of 99 GHz, has an excitation density threshold of [FORMULA]-[FORMULA] cm-3 and is ubiquitous toward regions of massive star formation but seldom elsewhere in the galactic disk - with the exception of the galactic center region (Bally et al. 1987). A complete survey of the [FORMULA] line toward IRAS point-like sources with FIR colors of UC H II regions in the galactic plane (Fig. 1) was presented by Bronfman et al. (1996; BNM). Nearly in all cases the [FORMULA] profiles obtained in such survey are singular; the dense gas cores are associated one-to-one with the heated dust which produces the FIR emission detected by IRAS as a point-like source. The [FORMULA] line, as well as other higher frequency rotational lines of CS, have been used successfully to map dense molecular gas cores and to derive their physical parameters (Plume et al. 1997; and references therein).

[FIGURE] Fig. 1a and b. Distribution of massive star forming regions in the whole Galaxy: a  Longitude-velocity; and b  Longitude-latitude. Sources with FIR fluxes higher than [FORMULA], corresponding to an O8 star with a luminosity of [FORMULA] at a distance of 4.25 kpc, are represented with open circles

The main goal of the present paper is to derive the mean radial distribution of the face-on FIR surface luminosity produced by embedded massive stars in the whole Galaxy, and to compare it with the mean radial distribution of H2 surface density as derived from CO surveys of the Milky Way. For a direct comparison, thus, we analyze our sample of OB star formation regions in a manner as close as possible as that used for the derivation of the H2 surface density (Bronfman et al. 1988; Paper I).

As an initial step in the analysis of the BNM CS[FORMULA] survey, we present the azimuthally averaged distribution of regions of massive star formation along the whole galactic disk, [FORMULA]. We also perform separate analyses of the northern (I and II quadrants) and southern (III and IV quadrants) Galaxy in order to search for large scale deviations from axial symmetry. Using the kinematic distances derived from the CS[FORMULA] line profiles and the FIR fluxes from the IRAS Point Source Catalog (PSC), we evaluate the FIR face-on surface luminosity, averaged over galactocentric rings [FORMULA] (0.85 kpc) wide, for the northern, southern, and complete galactic disk. The results obtained are compared with the molecular hydrogen surface density distribution as derived from large scale [FORMULA] surveys of the Galaxy.

In Sect. 2 we summarize the observations and the results of the BNM [FORMULA] survey. In Sect. 3 we describe our axisymmetric analysis and derive the mean radial distribution of OB star formation in the Galaxy, the azimuthally averaged face-on FIR surface luminosity these stars produce, and the mean FIR luminosity per OB star forming region. We evaluate the shortcomings in our analysis which are due to the spiral structure of the Galaxy and to the incompleteness of our sample. In Sect. 4 we compare these results with the surface density of cold molecular gas as traced by CO, discuss the large scale asymmetries of OB star formation in the Galaxy, and estimate the contribution of embedded OB stars to the FIR luminosity of Giant Molecular Clouds (GMCs). In Sect. 5 we summarize our conclusions.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: June 8, 2000