The Milky Way Galactic Ring Survey (GRS), a Boston University (BU) and Five College Radio Astronomy Observatory (FCRAO) collaboration, is a new large-scale molecular line survey of the Inner Galaxy. Recent advances in mm wave receiver technology, provided by the 16 (soon to be 32) element array receiver SEQUOIA on the FCRAO 14m telescope, enable us to probe the structure and kinematics of molecular clouds in the First Galactic Quadrant with unprecedented sensitivity (0.2-0.4 K), angular resolution and sampling (45", 22"), and spectral resolution (0.2 km s^-1).

Moreover, this improved sensitivity allows us to use both a column density (¹³CO 1-0) and a volume density tracer (CS 2-1) to image molecular clouds and cloud cores in the Inner Galaxy, especially in the 5 kpc Ring, the Milky Way's dominant star-forming structure. Due to smaller line widths of ¹³CO, we will be able to avoid velocity crowding and to more accurately establish kinematic distances to molecular clouds and dense gas associated with embedded infrared sources.

In our first of two GRS posters, we present results for the Galactic Ring Survey's first two square degrees, from
l = 44.3  to 46.3 and b =-0.5 to 0.5 degrees. We also compare the GRS to previous molecular line surveys.

• Survey with higher sensitivity, spectral and angular resolution, and better sampling than previous efforts.
• Observe two species, a column density tracer, ¹³CO, and a volume density tracer, CS; each has smaller line widths and so avoids velocity crowding.
• Identify clouds and clumps on smaller scales than ever before (~50").
• Solve the distance ambiguity toward these objects.
• Associate the clumps with infrared sources and assign distances to them.
• Determine the luminosity function of infrared sources in the First Galactic Quadrant.

About 70% of the molecular gas inside the Solar circle is concentrated in an annulus of dense gas with a galactocentric radius of about 5 kpc. This 5 kpc Ring (see Fig.1) dominates the star formation activity in the Milky Way and plays a crucial role in its dynamics, structure, and chemical evolution. Most of the Galactic giant HII regions, far infrared luminosity, diffuse ionized gas, and supernova remnants appear to be associated with the Ring (cf. Burton 1976, ARAA, 14, 275; Robinson et al. 1984, ApJ, 283, L31).
 

Fig.1:Galactic face-on  distribution of molecular hydrogen  (as traced by CO) for the Northern  Galactic Disk (adapted from  Clemens, Sanders & Scoville 1998,  ApJ, 327, 139).

 

Although the 5 kpc Galactic Ring was discovered over 20 years ago (Burton et al. 1976, ApJ, 202, 30; Scoville & Solomon 1975, ApJ, 199, L105), it remains largely unexplored.

Faced with the obstacle of obtaining high angular and spectral resolution observations of tens of square degrees of the sky, previous molecular line surveys of the 5 kpc Ring could not resolve small scale cloud structure due to a combination of undersampling with a small beam, low angular resolution, and velocity crowding of optically thick lines. Moreover, solving the kinematic distance ambiguity was not possible for many sources due to a lack of high quality complementary surveys (e.g. HI or H₂CO).

Using the BU-Arecibo Galactic HI Survey (Bania 1999, private communication), which provides the highest possible angular resolution for 21 cm H I emission observed with a single dish telescope (4'), we can resolve the distance ambiguity for the majority of the clumps we identify.

Fig.2 shows first moment maps for ¹³CO 1-0 and CS 2-1 obtained for the first two square degrees of the GRS emission in the velocity range 0-80 km s^-1. The maps consist of ~62,000 positions on a Nyquist sampled grid for both molecular species. 
 

Fig.2: GRS 13CO 1-0 and CS 2-1 first moment maps covering the velocity range  0-80 km s-1. The plot range was chosen to start above the average 3 sigma level  for both data sets. The beam size is given in the lower left corner of each panel.

The ¹³CO emission shows a combination of molecular clumps, arcs and filaments. Since the data were Nyquist-sampled, we detect structure on all scales down to the resolution limit. Moreover, because the plot range is well above the 3sigma sensitivity level, every feature in the map corresponds to a real cloud fragment. We can, therefore, decompose the clouds into much smaller clumps and cores than previous studies and associate those with a large number of infrared sources. The GRS substantially improves not only the sensitivity for the detection of molecular clumps, but also the ability to associate clouds with infrared sources, an important prerequisite to determine the infrared luminosity function of Young Stellar Objects. Note that although many clumps may overlap spatially in the moment map, they are well distinguished in velocity space (see channel maps in Fig.1 on the second poster).

Although the CS emission shows the same basic appearance as the ¹³CO emission, it is clearly enhanced toward the higher density cores. The intensity ratio of CS to ¹³CO is largest toward the compact cores. Low level CS emission, however, is detected towards more than 50% of the positions with detectable ¹³CO emission (averaged over a few square arcminutes).

 

Fig.3: A comparison of 13CO 1-0 emission from the GRS (45" resolution, 22" grid, top) with 12CO 1-0 emission from the  same region obtained for the UMass Stony-Brook survey (45" resolution, 3'grid, middle) and the Columbia survey (8' resolution, 8' grid, bottom).  In all three cases, emission has been integrated over a velocity range of 50-70 km s-1.

 A comparison of the new ¹³CO GRS data with the ¹²CO UMASS Stony-Brook (UMSB) and Columbia surveys is shown in Fig.3. This figure reveals the vast amount of hitherto unresolved details provided by the GRS which exploits the full resolution of the FCRAO 14m telescope. Although the UMSB survey was observed with the same telescope, and hence the same angular resolution, many compact sources escape detection because this survey was severely undersampled (3' spacing). Table 1 compares the GRS to previous molecular line surveys of the First Galactic Quadrant. Notice that the GRS achieves better angular resolution, spectral resolution, and spatial sampling.

¹The Bell Labs survey has not yet been published.

In addition to the higher angular resolution and better sampling, the use of the more optically thin, and hence column density tracing, ¹³CO for the GRS results in smaller line widths and less velocity crowding. Fig.4 shows how the use of ¹³CO allows a cleaner separation of velocity components and a better measurement of column density.
 
 

Fig.4:Sample spectra of 13CO 1-0 (blue) from the GRS, 12CO 1-0 (red) from  the UMSB survey (both 45" resolution), on HI (filled histogram) as observed  with the Arecibo telescope (BU-Arecibo Galactic HI survey, 4' resolution).

In Fig.4, the atomic hydrogen shows a broad and complex intensity distribution over the complete velocity range, whereas the ¹²CO intensities indicate distinct velocity components. Due to the higher spectral resolution and the smaller line widths for ¹³CO, these velocity components are clearly separated. Moreover, due to the lower optical depth in the ¹³CO lines, a cloud at v_LSR= 25 km s^-1 is seen in emission, as opposed to ¹²CO and HI where it is seen in self-absorption. The observed ¹³CO emission and HI self-absorption at the same velocity indicate HI self-absorption due to cool atomic hydrogen associated with molecular emission.

A full size version of the poster can be obtained here (gzipped postscript file)

The GRS is supported by the NSF via grant AST-9800334 and AST-0098562