CGPS News: Volume 6

CGPS News Volume 6
Nouvelles du RCPG Volume 6
September / septembre 1998


Word from the editor

Magdalen Normandeau

Of the fifty-six people listed on the CGPS membership directory, so far only ~50% have contributed to the newsletter. What is everyone else doing?


From the management committee

Russ Taylor

Management committee news

It was a tight finish for the election to the management committee this summer. Newly elected members are Lloyd Higgs, Dave Routledge and Nicole St-Louis. Lloyd returns after serving last year. Departing members are Judith Irwin and Gilles Joncas. Thanks Judith and Gilles for lending your wisdom to the oversight of the project.

In November we will achieve the three-year point in the NSERC funding of the CGPS. This is the critical review point and the management committee must prepare a report to NSERC by the end of September. Funding for the 4th and 5th year of the project is contingent upon successful review of our report. Needless to say we are hard at work drafting a report that highlights our accomplishments to date. If we are successful, new funding will be available on December 1. Our research officer at NSERC, Chantal Abou Debs, has promised to let us know as soon as a decision has been taken.

Second CGPS Postdoctoral Fellowship Competition

The current round of CGPS PDF awards comes to an end in November. If funding continues, we will be able to fund a second round. The new PDF positions can begin anytime between December 1998 and November 1999. The positions are for two years and the CSP grant provides fund for half of a salary. In the first round thas has been $14,250. We have submitted a request to NSERC for a budget increase for PDF stipends. If successful, the contribution will rise to $16,000. The PDF positions are allocated based on proposals submitted to the management committee. The proposals should indicate

  1. the supervisor(s) and the location of tenure of the position,
  2. a description of the CGPS-related project that the PDF will work on,
  3. the source of funds for the remaining half of the salary.

CGPS consortium faculty at Canadian institutions are eligible for the awards. A joint submission from members of a project team, and pooling contributions from NSERC individual grant might be an easy way to generate the funds. Applications, should be no more than a page or two and should be sent to Russ Taylor (russ@ras.ucalgary.ca) before October 15. The management committee will review the proposals shortly thereafter and announce the results. There are three PDF awards to be given out.

JCMT Long-Wavelength Filler Program Proposal

Henry Matthews and Gerald Moriarity-Scheiven have pointed out that there is a dearth of JCMT back-up programs for marginal weather conditions on Mauna Kea. This is particularly true for Canadian back-up programs. They have suggested that the CGPS consortium may wish to submit a proposal to the JCMT to carry out CGPS follow-up studies as back-up programs. A back-up program does not require that an observer travel to the JCMT. The observations are carried out by the JCMT staff when weather conditions are not acceptable for a scheduled high-frequency observation. Low-frequency observations suitable for back-up programs are CO 2-1 observations or SCUBA imaging of dust at about 0.8 mm wavelength.

The deadline for submitting proposals is two weeks away. We will likely put together a proposal. Anyone who has an interest in pursuing this or has an idea for a JCMT follow-up study that would be suitable for back-up please let Russ Taylor or another member of the management committee know. Purely technical questions should be addressed to Henry Matthews (with copy to the management committee). We will begin to draft the proposal in about a week, so please send in your input as soon as possible. Before the end of the month we will need a short paragraph on science justification, a list of objects (or at least a definition of a class of objects), and a short technical description of the required observations.


Observing Status

Andrew Gray

At the end of September this year observing at DRAO for the Canadian Galactic Plane Survey will have been underway for 3.5 years. We will have completed 128 fields with the Synthesis Telescope, representing 67% of the total (190) in 70% of the allotted survey time. In the same period we will also have completed 29 non-CGPS fields, including the equivalent of about 2 fields for solar observations, plus 2 repeats of CGPS fields, a total of 159 fields at an average rate of 45.3 fields per year.

Current observing rates exceed this average by at least one field per year, due to a decline in development work and an increase in system reliability. It is a testament to the hard work of the DRAO staff that problems causing significant data loss are now few and far between. Indeed, the only significant cause of lost data in recent times is the main observing computer, which has been exhibited some unreliability. Needless to say, this problem is being worked on and a solution may be at hand. In the meantime, it is only necessary to observe 40 CGPS fields per year to complete the survey observations by March 2000, the end of the allotted period, so we are able to offer 6 fields per year for external proposals.


CGPS Infrared Data Set Completed

Charles Kerton

Last month I completed the HIRES processing of mid-infrared images for the CGPS survey region along with a high-latitude extension to the existing far-infrared data set.

The mid-infrared images have been dubbed MIGA, for Mid-Infrared Galaxy Atlas. MIGA is essentially the mid-infrared equivalent to the Infrared Galaxy Atlas (IGA) consisting of 1st and 20th iteration HIRES images along with a slew of ancillary maps detailing beam shape, photometric noise, and IRAS detector coverage and scan patterns. If you would like a bit more information on the details of MIGA, you can obtain a copy of a couple of related papers.

Production of the mid-infrared CGPS mosaics is currently underway at CITA. The pre-processing steps taken as part of the MIGA construction allow large, high-quality mosaics to be easily constructed. For example, the MW1 and MW2 12 micron mosaics are shown below. I expect that production of all of the mosaics should be finished by early November.

MW1 12 Micron Mosaic MW2 12 Micron Mosaic

The extension of the IGA (EIGA, for Extended-IGA) to beyond +4.7 degrees for the CGPS region also is complete. The EIGA was constructed in order to match the IR coverage with the radio coverage of the CGPS. Shown below is the original 100 micron MV2 mosaic along with the MV2 mosaic with the EIGA strip added in. Agreement between the EIGA and the existing IGA images is excellent. Steve Gibson has updated all of the CGPS M mosaics so that they now include the EIGA images.

Original 100 Micron MV2 Mosaic Extended 100 Micron MV2 Mosaic


Tyler Foster: new grad student at UofA

Dave Routledge

After completing a B.A. in English Literature and Philosphy (double major) in 1992, Tyler returned to school and completed his B.Sc. (Hon) Physics in April 1998. He has deep roots (nearly fifteen years) in amateur astronomy. He chanced to get involved with optical astronomical research using first the King's University College Observatory, and then the Devon Astronomical Observatory, which is operated by the University of Alberta. From work on these two systems, he has been fortunate enough to co-author three publications on CCD photometry and CCD systems with various colleagues.

Currently, Tyler is heavily involved in the development of a wide-field high-efficiency CCD imaging system for use on the Devon 0.5 m reflector. This system will allow for both high precision photometric studies of variables, and detection of optical supernova remnants in broad and narrow band filters. For his M.Sc. thesis, he plans to do both multi-color variable star photometry and a study of one or more supernova remnants in as many spectral regions (radio, optical, IR, X-ray, gamma-ray) as possible.

For five years, Tyler has been an astronomy instructor at the Edmonton Space and Sciences Center. He also serves to co-direct the center's public observatory, and has served as Public Education Director for the Royal Astronomical Society of Canada, Edmonton Center. Continuing with his amateur roots, he enjoys observing, telescope making, and teaching others to do both.


On the Atomic-Molecular Gas Relationship

Chris Brunt

This is a short note on the CO-HI relationship. Clearly, given the nature of the two data sets of HI 21cm emission and 12CO(J=1-0) emission, the relationship can be investigated in a much more complete way than ever before. ie What (statistically) is the spatial relationship between atomic and molecular gas? To what extent are molecular gas and atomic gas mixed? What is the nature (geometry, physical properties) of the structures that display CO-HI anticorrelation? How is this related to energetic events occuring in the vicinity? etc But : how to proceed?

A search for absorption features in HI that can be unambiguously associated with the presence of molecular gas is underway (Gibson et al). One can imagine inspecting velocity-integrated maps (over some specified and perhaps optimized) velocity interval for detailed (anti)correlations between HI and CO. What is the optimum interval? Is there one? This is also of relevance to Lloyd Higgs' differencing method discussed here. ie What is the optimum velocity interval over which differences should be formed? Additionally, we would like to be able to extend this type of study to regions in which there is no immediately obvious correspondance between HI and CO. How should we begin to decompose the complex 3D information (of both fields) in such a way that progress on understanding their relationship can be made?

It is suggested that the use of Principal Component Analysis or "PCA" (Heyer & Schloerb 1997) may provide the groundwork for a large study of the HI-CO relationship. Briefly, given a data set I(l,b,v) (2D-spatial;1D-spectroscopic), PCA (as used by Heyer and Schloerb) decomposes the data onto a variance-ordered series of (orthogonal, spectroscopic) eigenvectors. The eigenvectors are designed to identify the appropriately-ordered sources of variance within the data. PCA is not limited to "object" identification, and utilizes the full 3-dimensional information. After identifying the eigenvectors, a sequence of "eigenimages" can be constructed - which are simply the integration of the spectra weighted by the appropriate eigenvector. PCA effectively defines spectroscopic windows that are specifically tied to the structure of the data set in question.

Typically, PCA is carried out on a single data set. (An example of this can be seen at link given below.) Here, the CO emission at W3 has been decomposed by PCA. The first eigenimage is always very similar to the integrated intensity of the whole data set. The second eigenimage, in this case, identifies the two "arms of the horseshoe" as the second most significant source of variance. The third eigenimage shows a "ring-bar" structure that is the next major source of variance - the significance of which at this stage is not clear. The eigenvectors are seen to define the appropriate velocity intervals that trace these sources of variance (ie "structure").

The same type of decomposition could be carried out on the HI. However, to make the link between HI and CO, the following method is proposed :

The results of such a procedure for the W3 region are shown here : CO-HI PCA, and appear promising. In particular, the 3rd eigenimage referred to above, shows a reasonably-well detailed anticorrelation between the CO emission and the HI emission, that is not obvious from a straightforward channel-by-channel or fully-integrated comparison. If unweighted but restricted integrations are desired, then it is suggested that for this particular example (W3), the "half-wavelength" (or similar measure) of this 3rd eigenvector should define the appropriate (optimized?) velocity interval that maximizes the CO-HI anticorrelation. This interval should be appropriate either for integration or differencing schemes. It should be noted however that all eigenimages can potentially show such (anti)correlated structure, to a greater or lesser extent. The same procedure carried out for an ensemble of fields will be needed to more completely define the characteristics of this type of analysis. The HI emission could also be used for the first decomposition (subsequently projecting the CO with the HI eigenvectors), but initial tests of this procedure are less successful.

It should be realized that the form of the eigenvectors in the above example effectively produce differencing operations as the emission is integrated (ie since the eigenvectors contain negative numbers). For this reason, some interpretation of the eigenimages must be made. A variant of PCA (Postive Matrix Factorization; Juvela et al) requires the "eigenvectors" to remain positive-definite and so may be easier to interpret. This is untested (for the particular goal discussed here) at the present time. We note that there is the likely possibility, however, that a differencing scheme may be more suited to finding anticorrelated structures against a background.

It should also be pointed out that the eigenvectors are purely spectroscopic - ie are not a function of l and b for the field under study, and so contain no detailed information about the CO spatial structure. They simply provide a sequence of variance-ordered global spectroscopic windows that are "self-defined" by the structure of the CO data set.

This work is clearly at a very preliminary stage, but I feel that the initial results are encouraging enough to be reported here. In the near future (probably in the early part of 1999) I hope to begin a systematic investigation of this technique and to apply it to an ensemble of fields in the Outer Galaxy. Anyone interested in being part of this study, or anyone interested in finding out more about this for their own purposes should contact Chris Brunt at the email address below.

References :

Heyer M. and Schloerb, F.P. ApJ 475, 173, 1997.

Gibson S. : CGPS Newsletter vol.5, and see also the HI Self-Absorption Web Page.

Juvela M., Lehtinen K. and Pentti P. : (hyperlink above)


The Interstellar Environment of the Wolf-Rayet Star WR 143

François Cazzolato & Serge Pineault

The Wolf-Rayet star WR 143 is located in the general direction towards the Cygnus OB1/OB3 infrared superbubble [discussed by Lozinskaya & Repin (1990) and Saken et al. (1992)], near its northeast boundary.

A first quick-look at the CGPS 1420 MHz radio continuum emission (thanks, Brad!) shows no significant extended emission associated with the star, a fact confirmed by other previous radio surveys (e.g., Taylor et al. 1996, Condon et al. 1994), and also by the IRAS/HIRES infrared data at 60 and 100 microns. At these IR wavelengths, no point source is seen coincident with the WR star, although a point source is present some 3' from the stellar position. This nearby source is also seen on the high-resolution 12 and 25 micron maps of the area (thanks, Charles K.), but again no trace of the WR star.

A preliminary survey of the HI data cube shows the possibility of HI deficiency at the stellar position at two different velocities, none of which is consistent with the galactic kinematical distance in that direction.

The next steps will be to assess the statistical significance of the nearby IR point source (physically associated to WR 143, line-of-sight superposition of external galaxy ?...) and of the possibly associated lack of HI at the WR star position. If the HI feature is indeed associated with the WR star, then the star cannot be inside the Cygnus OB1/OB3 superbubble (at 1.5 kpc, and hence cannot contribute its wind luminosity to it). It would however be compatible with the quoted distance of 0.9 kpc (van der Hucht 1988) placing the WR star in front of the superbubble (although the observed velocity remains to be explained).

[Editor's note: François Cazzolato is a M.Sc. student at l'université Laval, working on this study of WR 143 for his thesis. Bienvenue au consortium, François!]


Commentaires ou suggestions? Comments or suggestions?
Ecrivez à / Write to mnormandeau@astro.berkeley.edu