APEX SV projects

Molecular cores and excitation temperature in the Eagle Nebula's fingers

Coordinator: F. Schuller, K. M. Menten

Data:
Program is available and data products can be downloaded

Scientific justification:

The Eagle Nebula, M16, is a prominent Hii region located at a distance of 2.2 kpc from the Sun. It is associated with the NGC 6611 star cluster, which contains several hundred intermediate to high mass stars that formed over the last few million years. To the south of this cluster, the radiation from hot cluster stars has sculpted columns of dense obscuring material on a few arcmin scales, usually referred to as "fingers" (or, dramatically, "Pillars of God"), whose breath-taking HST images even entered popular culture (Hester et al. 1996).
Indications of present-day star formation near the tips of the fingers are seen e.g. in the infrared (Thompson et al. 2002), while the three fingers have been mapped in molecular lines and submm continuum radiation by White et al. (1999) using the JCMT 15 m and OSO 20 m radiotelescopes. They report CO(2-1) and CO(3-2) peak brightness temperatures above 40 K and 60 K, respectively, and a factor 2 - 3 lower temperature in the CO(1-0) line, which they interpreted as effects of depletion onto the coldest phase of the dusty material (although that appears difficult to understand to us). The CO(2-1) transition was also observed by Andersen et al. (2004) using the SEST 15 m telescope. Their map cover an area further to the south, at the base of the pillars, in the vicinity of the Herbig-Haro object HH216.
The dust and molecular gas inside the columns are dense enough to be opaque to optical emission, but may not have reached yet a density sufficient to produce significant star formation. Here we propose to map that region in the higher excitation lines CO(4-3) and CO(7-6) to better constrain its geometry, and to derive (using LVG modelling) temperatures and densities in the dense molecular cores. The improved spatial resolution provided by APEX at 807 GHz will allow us to precisely localise the warmer and denser regions of the molecular cores and to address the issue of cloud fragmentation. In addition, given that there is a boundary between the Hii region and the dense molecular gas, this source is probably a textbook example of a Photodissociation Region (PDR). We would like to probe this PDR with the 1-0 and 2-1 transitions of atomic carbon.

Proposed observations
We propose to perform raster mapping, covering 1'×1' in CO to map the tip of the northern most finger, and 1'×30" in [C I] lines at the top of this finger, to probe the interface between the molecular gas and the Hii region.
Integration time

Assuming CO(4-3) is as bright as CO(2-1), and CO(7-6) twice fainter, i.e. peak at 20 K, detecting 10% peak at 3- requires an rms of 0.7 K at 810 GHz. The FFTS backend with 16k channels provides a velocity resolution 23 m/s at 810 GHz, and can be smoothed to 0.5 km/s (20 channels) without any loss of structure according to the maps in White et al. (1999). Therefore an rms of 3 K per spectral element would give a sufficient sensitivity. With a system temperature of 2500 K at 810 GHz, an integration time of 20 s per point with standard raster mapping is well suited.
To map the proposed 3'×1' area with a sampling of half the beam at 807 GHz (i.e. 4") would then require 16×16 points, or 1.4 hours of integration, which translates to 4 hours with overheads.
Plume et al. (1999) observed four molecular complexes in [C I]3 P1 -3 P0 and CO(2-1) lines. They report peaks in integrated line intensities 5 to 15 times fainter in [C I] than in CO(2-1), with comparable linewidths. Using a ratio of ten, we can expect a peak brightness temperature of order 4 K in [C I]. Then, to reach 20% of the peak at 3- requires an rms below 0.3 K. This rms can be reached by smoothing to a resolution of 0.5 km/s so that an rms of 1.2 K per channel will be enough. An integration time of 140 s per point will provide this sensitivity at 809 GHz. Then, mapping 1 ×30 with 6 sampling (11×6 points) takes 2.6 hours of integration, or 7.7 hours with overheads. Thus, we are asking for a total of 12 hours of observing time to complete this project.

Possible extensions
This exploratory project can later be complemented by observing:
- a larger area (e.g. 6.5 ×4.5 ) covering the three fingers from their tips to their bases well inside the nebula in the same CO lines
- the same area in [C I] lines
- dust emission in a larger ( half degree) region using the Laboca and 350 ľm bolometer arrays

References

Andersen M., Knude J., Reipurth B., et al., 2004, A&A 414, 969
Hester J., Scowen P.A., Sankrit R., et al., 1996, AJ 111, 2349
Plume R., Jaffe D.T., Tatematsu K., Evans N.J., Keene J., 1999, ApJ 512, 768
Thompson R.I., Smith B.A., Hester J.J., 2002, ApJ 570, 749
White G.J., Nelson R.P., Holland W.S., et al., 1999, A&A 342, 233