Welcome to the first issue of the Galactic plane survey newsletter! The present edition differs somewhat from what should be the norm in that it contains summaries of the Birds-of-a-Feather discussions from the recent consortium meeting and many instances of general considerations rather than specifics about work actually being done. In future, as we will already know what everyone's overall plans are, there should be more articles discussing details of work under way and interesting disocveries concerning this object or that. As well, in this first volume, authors have generallly not made use of the possibilities of the chosen medium: not a single gif, tiff, or jpeg file appears; there are few links to other documents; and only a handful of postscript files are presented. Hopefully this will change as people have more and more interesting results to show off!
There was some talk of creating an exchange forum to complement the newsletter. The following ideas have been put forth:
If you have comments about the format of the newsletter and such things, please let me know. All suggestions are welcome but those accompanied by useful information are more likely to be heeded. Translating the newsletter into both official languages would be extremely time consuming and, I suspect, not really worth the effort. However you are always welcome to write your contribution in English or French or both if you so desire.
This choice has three points in its favour.
It is particularly crucial for all of us to be informed of graduate student projects. This issue was brought home at the meeting by three presentations related to the prominent supernova remnant G127.1+0.5 visible in the "A" mosaics. The study of this object is, in fact, part of the Ph.D. research of Richard Christie. To avoid encroachment on thesis research project we have instituted a separate graduate program project registry on the CGPS web pages. We ask all graduate students and their supervisors to send a thesis title and a short description of the research project to Steve. If there appears to be a conflict between a general consortium science project and a graduate project, we ask the PI's to contact the student and supervisor involved.
We will plan to have our usual science meeting in the spring as well. The site for this is open. Perhaps somewhere in the east?
Here's the scoop on how the observing is getting along.
First, the overall picture on what is being attempted: CPGS observing began in April '95 and is scheduled to be complete in March '00, a total of 190 fields covering the galactic longitude range 75 to 146 degrees. The actual observing is accompanied by a lot of inhouse processing (in contrast to the outhouse processing which is left to the user) which will not be completed until some time after March '00.
The various stages of observing and processing at DRAO, for the synthesis telescope, are
The short-spacing data has not yet been added in. For spectral-line data this requires that the data first be collected with the 26m. The production run for the 26m has an anticipated time scale of ~4 months and should begin any day now - literally: there have been a couple of false starts but it really is that close - looked after by co-op student Diana, overall coordination of the effort in the care of Ken who could provide better information of exactly where it's at.
Things that are stopping the "preliminary" mosaic from becoming "final" mosaics (list is not complete):
Calibration of the 21cm continuum and line data from the ``difficult'' end of the survey region, the Cygnus region, has been completed. The dynamic range of both the continuum and line data from these fields is typically limited by the presence of Cygnus A that lies only 1 degree of galactic latitude above the survey region. However, we have managed to successfully model Cyg A in all but one field, and have attained a dynamic range of at least 10,000:1 in the final mosaic (see figure below). The peak intensity in the mosaic is 6.2Jy/beam (in S106) and the rms background is approx 0.6 mJy/beam.
In the remaining Cygnus field (``O4''), Cyg A is only 70 arcminutes from the phase-centre of the image, and we have not succeeded in reducing image artifacts to a satisfactory level. This is also the limitation in our attempts to calibrate the ``A1'' field containing the SNR 3C58. These two fields present a challenge to calibration that must be overcome before we arrive at the fields near Cass A.
Calibration of data from the high longitude end of the survey has been started. Between longitude 140 and 147 degrees, there is little extended continuum emission and the bulk of the unresolved sources have peak intensities of around a few hundred mJy/beam. This presents few calibration difficulties compared to the Cygnus region. However, these fields are not without problems, in particular those presented by short term (an hour or less) variations in the continuum emission from the bright pulsar 0329+54 in the field V0. Modifications of MODCAL have been made to address these problems and by the next newsletter we'll know how successful these are! In addition, the bright source 3C86 (6 Jy/beam peak) is in the V2 field. Once V0 and V2 have been dealt with, a VW mosaic of 14 fields will be available from the GPS ftp site.
Once the VW mosaic is complete, and I have identified a scheme to deal with calibration of fields that include bright sources away from the phase centre (A1/1A, O4 etc) calibration of the 408 MHz data from the V and W fields will be started.
The major millimeter wave imaging project at FCRAO during the last 3 years has been the CO Survey of the Outer Galaxy. The survey covers galactic longitudes between 102° and 142° and latitudes of -3° to +5.4° at 50'' sampling and VLSR between -150 to +40 km s-1 with resolution of 1 km s-1. It is designed to address several outstanding questions concerning the molecular interstellar medium:
While final data reductions and analyses are underway, preliminary images and scientific results have already been extracted and published which illustrate the power of the survey. Images of CO J=1-0 integrated intensity over the local and Perseus spiral arm velocity intervals are shown in Figure 1. Also displayed are the locations and mean sizes of optical HII regions and positions of OB stars within the observed field. While the molecular emission is sparsely distributed within the outer Galaxy, the image shows bright molecular interfaces at the edges of HII regions, coherent features over scales of 200 pc centered on rich OB associations, and many isolated, compact clouds.
The survey is the most sensitive, unbiased census of molecular gas in the far outer Galaxy. In Figure 2, we show the mass surface density as a function of galactocentric radius derived from the survey~ data cube and assuming kinematic distances and a CO-H2 conversion of 6.2 M(sun)/K-km s-1-pc2. Also shown in Figure 2 is the variation of z distance of the molecular midplane as a function of galactocentric radius. Due to limited latitude coverage, the survey does not completely sample the distribution of molecular emission with R<10 kpc. The peak of the radial profile at R=11 kpc is due to the concentration of molecular material within the Perseus spiral arm. Beyond 14 kpc, the survey~ has successfully identified over 100 individual clouds. Nevertheless, the surface brightness of the emission from gas at these large radii is extremely low, and this material does not comprise a significant amount of mass or star forming capability. The distribution of material detected at large radii is also observed to be skewed toward higher galactic latitudes and follows the general warp of the outer Galaxy known from previous HI 21cm line surveys.
To describe the relationship of the molecular gas to global spiral
structure, we have identified 886 objects with
One of the primary motivations for the survey is to provide structural and kinematic descriptions of the molecular emission in order to diagnose the underlying physics. Spectroscopic imaging studies such as the survey demand statistical techniques to fully exploit the vast information. To this end, we have investigated the use of the multivariate analyses: applying Principal Component Analysis (hereafter, PCA) and hierarchical clustering to astronomical data cubes as well as multifractal descriptions of molecular line images. Each technique has advantages and disadvantages for the description of molecular clouds, and we plan to apply each of them to the survey data base.
The survey, made with the Cambridge Low-Frequency Synthesis Telescope (CLFST) at 151 MHz, covers approximately the region 80° < l < 180°, |b|<5°, with a resolution of about 1.2 × 1.2 cosec(declination) arcmin2 (EW×NS). The survey consists of 24 fields, spaced 3°.5 in l, and alternating between b=+3° and -3°, but with a gap near l=110°. There is some coverage to |b|<9°.
These data are, at present, only available before publication to members of the GPS consortium.
As the CLFST has a wide field-of-view, and operates at a low-frequency, ionospheric fluctuations are significant. These means that the absolute positions of synthesised images are not always well known (which requires considerable careful calibration and analysis on a field-by-field basis), and also that the quality of images synthesised near bright sources is poor. Consequently the sensitivity of the survey varies with l. The main product from this survey, so far, is a compact source list that is complete to typically 0.2 Jy (with poorer sensitivity towards: l=80°, i.e. near Cyg A; in the gap at l=110°, i.e. near Cas A; or l=185°, i.e. near Tau A, the Crab Nebula).
This survey was the basis of a PhD research project recently completed by Simon Vessey (who is no longer working in Astronomy). A paper, Vessey & Green, containing the compact source list has recently been submitted to MNRAS.
Images from the survey will also be made available, again initially to GPS consortium members only, although producing such images in a format so that they can be easily compared with other data is not straightforward. The CLFST is not exactly east-west, so the synthesised beam varies across the images, so that production of ``mosaics'' is complicated. These beam distortion effects are taken into account in the production of the compact source list. Also, the images made by the CLFST are not stored on a grid projected at a standard epoch (e.g. B1950 or J2000). Instead a grid at the epoch of the observations is used, as for these large field images any other projection is not strictly correct.
For further information, including the preliminary version of the compact source list, see the pages for GPS consortium members on the MRAO WWW server (use the same username and password as for the main GPS private pages).
Mike Fich, Dave Routledge and I are studying a bi-polar HI feature in the Cygnus region. The feature is located at RA(J2000) = 20h 20m 30s, DEC(J2000) = 38d 7', v(LSR) = 35 -> 47 km/s. The morphology of the feature varies with velocity, suggesting kinematics not unlike those seen in molecular outflows. IRAS data reveals faint IR emission coincident with the HI feature, but there is no signature of a central engine.
The HI feature lies adjacent, in both position and velocity, to an apparent wind-blown bubble. This bubble is seen in both the HI and in the POSS plates. It is not clear the relationship, if any, between the bubble and the bi-polar feature.
It is tempting to hypothesize that the bi-polar HI is associated with a molecular outflow. However, low sensitivity CO data from the CfA show no evidence of molecular material associated with the HI. Service observing time has been granted on the JCMT to look for CO in the 2->1 transition at selected positions along the HI feature, but the observations have not yet been made. We hope that these observations will provide much-needed clues as to the nature of this feature.
Spurred by several discussions on the last GPS science meeting I summarize my basic knowledge on the Cygnus X region. Keep in mind that some statements are not accepted as universal truth by the community.
References are not too numerous in the moment. But you may want to look into the CX-XVIII paper before starting work on the region (Wendker,H.J., Higgs,L.A., Landecker,T.L. 1991, A&A 241, 551). Therein is a list of holes which some people prematurely call already bubbles (but beware) and a list of ridges which some people call already shock fronts but which are probably star forming regions of intermediate mass.
|Name 1||Name 2||comment||distance [kpc]||remarks|
|G74.9+1.2:||CTB 87||certain||12||at edge of O5 field|
|G76.9+1.0:||-||some doubts||>10||interesting evolutionary status;
|G78.2+2.1:||W66||certain||1.5||don't call it 'Gamma Cyg SNR', the star has nothing to do with it;
LAH: it could be closer (1.3 kpc);
HJW: even further (up to 2 kpc).
|G82.2+5.3:||W63||certain||2 ?||only partly covered in GPS,|
|G84.2-0.8:||-||certain||3||maybe an interarm SNR.|
|G84.9+0.5:||-||doubtful||-||I told BW already that I see baseline problems and then it is thermal radiation.|
|G89.0+4.7:||HB 21||certain||0.8||Cyg OB7 association?|
A nasty remark:
I regard the Aschenbach and Routledge list of possible SNR as a not very helpful one (at least in the Cyg X area) because the most prominent candidate (W66) is missing and the certain thermal area around WR 136 (NGC 6888) is contained.
|Name 1||Name 2||distance [kpc]||remarks|
|NGC 6888||G75.4+2.5||1.8||ring nebula around WR 136|
|ON2||G75.8+0.3||0.6||star forming volume|
|S 106||G76.4-0.7||0.6||star forming volume|
|IC1318a||G79.2+3.5||1.5||evolved HII region|
|IC1318b/c||G78.7+0.9||1.5||evolved HII region|
|DR 15||G79.3+0.2||1.0||young HII region|
|DRAO Bubble||G79.3+0.3||2.0||ring nebula around an LBV|
|DR 6||G78.1+0.6||1.5||young HII region|
|DR 9||G78.0+0.1||1.5||young HII region|
|DR 13||G78.2-0.4||1.5||evolved HII region|
|DR 7||G79.4+1.3||7.||giant HII region|
|DR 20||G80.9+0.4||2.||evolved HII region|
|DR 22||G80.9-0.1||0.8||evolved HII region|
|DR 17||G81.3+1.2||0.8||evolved HII region|
|DR 21||G81.7+0.6||3.||star forming volume|
|DR 23||G81.6+0.0||3.||evolved HII region|
|W 80||G84.9-1.0||0.5||very evolved HII region|
There are several HII region which must be situated in the Perseus arm according
to their radial velocity (in a recombination line). The best example is
DR7. Another is G85.2+0.0.
18P61 (G80.4+0.7) most probably is even an outer arm object (RV approx. -65 km/s).
How many weak HII regions without a measured RV fall into these categories is unknown.
|distance in kpc||object|
|<0.5||Pretty dull volume. Local bubble and beyond. Thin hot/warm ISM.|
|0.5||North America - Pelican nebulae complex (W80, NGC 7000)|
|0.6-0.8||The Cygnus Rift. Huge dark cloud. Produces large foreground extinction. Side facing the sun seems quiet. A few objects on the backside are star forming complexes like ON2, S106.|
|0.7||Cyg OB7, covers a large area, contains probably HB21.|
|1||DR 15 plus ridges|
|1.5||IC1318a/b complex, probably includes DR6 and W66.|
|1.8||Cyg OB1, includes probably NGC 6888 and the star P Cyg.|
|2||Cyg OB2, huge association, suspected supershell in X-rays, some speculate that all (!) features of the Cyg X region belong to this association. But it is even unkown which radial velocity (RV) is appropriate for the gas in order to belong to OB2. There are suggestions for a host of outlying objects like W63, some optically visible filaments, a large cometry nebula etc., the DRAO bubble G79.29, 4 WR stars.|
|3||the DR21 and DR23 complex. Lanie Dickel still keeps DR21 at 2 kpc.|
|4||The realm of the local arm. RV about -20 km/s.|
|6 - 8||Crossing the Perseus arm. RV about -40 to -60 km/s. Contains DR7.|
|10 - 12|| Crossing the outer arm, if it is really an arm and
not an outlying ensemble of some clouds. RV about -80 to -100 km/s.
Contains Cyg X-3.
Before the recent meeting the participants in the Disk-Halo Interaction science team were Neb Duric, Jayanne English, Carl Heiles, Judith Irwin, Magdalen Normandeau, and Russ Taylor. Lloyd Higgs had been in communication with this group about an area of overlap (see item 2 below). At the meeting Michael Fich, Doug Johnstone, and Shantanu Basu presented work which also resides in the disk-halo interaction category.
The motivation to examine the region close to the galaxy's mid-plane for evidence of the exchange of energy and material between the disk and the halo has been described in one of the CGPS/RCPG NSERC proposals:
"The dramatic contrast between the disk and halo implies that rapid and extreme changes in physical conditions are occurring over a relatively short distance above and below the Galactic plane. The discovery of discrete vertical structures of gas at the disk-halo interface, for ex. neutral hydrogen "supershells" and "worms" ..., also suggests that the disk and halo are intimately connected through dynamical processes."
So far we have discussed 3 broad strategies which we enumerate below. Although not explicit in the following descriptions, we intend to examine all available wavelength regimes and velocity as well as polarization data. Nor can we be limited to specific patches of the sky for this analysis. We are also mentioning work which we expect to do because we hope the newsletter exposes areas of overlap with other teams and will allow us to set up the appropriate collaborations.
The search for dissociating star has begun. This project has the immmense advantage that it is not hampered by the lack of shortspacing information (continuum or HI) as we are looking for structures which are only a few arcminutes across. Two attempts have been made over the past year to test the feasibility of detecting 'dissociating stars' from the Galactic plane survey data. These are stars which do not have enough UV radiation to create easily identifiable HII regions, but can be recognized by their association with HI which has been formed from dissociated molecular hydrogen. The spectral ranges are expected to lie in the range B2-B5. Neither attempt has so far proved to be a roaring success, but these projects are in preliminary stages.
In the first attempt 24 IRAS sources were selected (at FCRAO) from the pilot survey data, according to flux values at 60 and 100 microns, and the CO, HI, and 21 cm continuum emission images were examined. Mark Heyer assembled the four 1x1 deg. images, centred on the IRAS positions. A candidate is expected to be associated with a CO cloud, and have no (or at most weak) continuum emission. Several IRAS sources fit nicely into this category, but identifying any associated HI emission is the difficult part. None of the candidates stands out as having definite HI emission associated with it. Further analyses will have to consider different velocity ranges over which to integrate the CO and HI, and perhaps involve differencing between on and off source to try to detect HI.
The prototype for this proposed class of objects, IRAS 23545+6508 (Dewdney et al 1991 ApJ 370,243) is a strong infrared source with IRAS colours falling within the Hughes & MacLoed (1989) criteria for HII regions. It had no continuum counterpart on the DRAO images and most importantly it had an associated smudge of HI emission. With this in mind, Magdalen undertook a search starting with the IRAS point source catalogue, selecting those sources with the colours of HII regions and sufficiently high 100 micron flux. This is followed by looking for counterparts in the 1420 MHz continuum images (if it has a corresponding compact source then it's not a dissociating star; there's too much ionized gas). For those IR sources without continuum counterparts the HI cube is searched for an HI smudge at that position and, when available, CO maps are also consulted. So far, not so good. The A mosaics have been attacked in this manner to no avail. There are 11 sources with appropriate IRAS colours, 7 of which have no associated radio continuum compact source, and none of these has an associated HI smudge. A similar approach applied to the pilot data had equally disappointing results. How sad. Was the estimate for the number of dissociating stars put forth by the authors of the prototype paper excessive? We shall find out: the search continues.
Once we have a few good candidates, further progress will presumably involve observations with the VLA in order to obtain more sensitive values for emissions from continuum and HI.
The main method under investigation at the moment is the characterisation of HI ``objects'' by using their density distribution (statisticians lingo for a histogramme of the brightness distribution). Sonia Jean, an M.Sc. student at université Laval, has done most of the work, under the supervision of Nadia Ghazzali and Gilles Joncas.
HI objects are defined as all the connected pixels above a certain threshold within a certain area. The thresholds were chosen quite subjectively to conserve whatever we deemed to be an object. This obviously has the effect of truncating the density distribution at the low brightness temperature end. Nonetheless a set of objects were considered and cluster analysis based on the difference in area between the various density distribution curves was performed.
Ideally, one would want to use the true brightness distribution of the objects, unmarred by noise. The plan was therefore to deconvolve the noise distribution from the density distributions of the objects. This has not been entirely successful. For this to work, one needs to know the noise distribution and, while it is somewhat akin to the usually assumed gaussian, it's not quite a normal distribution. Nor is it the same everywhere on the mosaic. Basically, the noise for the synthesis telescope data images before correction for the primary beam is Gaussian. But from there, the individual fields get corrected for the primary beam, there is u-v plane filtering, adding of the shortspacing information, and finally mosaicing. Not surprisingly it isn't Gaussian anymore after all that. Be that as it may, the initial attempts at noise deconvolution were done using a normal distribution. Work is now under using a more realistic noise distribution though it is, of course, still a single noise distribution for the entire mosaic.
Results both with deconvolution (histogrammes, clustering tree) and without (histogrammes, clustering tree) are presented for an initial set of eight objects (greyscales of objects 1,2,8,6 and greyscales of objects 7,5,3,4 are available). The cluster analysis results for both cases are quite similar, indicating that the noise had little effect on the objects studied (the theresholds were between 3 and 8 sigma, where sigma was calculated from an empty channel map). With so few objects it would not be wise to start thinking about trends in classification which could be tied to physical processes; for now, we are mostly working on developing the analysis method. An additional 34 objects are now under consideration. Among these are instances of the same HI feature at different velocities to examine whether or not they will be classified together.
Send all comments and questions to Nadia Ghazzali
HI line and 21cm continuum data towards six pulsars and two neutron star candidates have been extracted from the Canadian Galactic Plane Survey. Limits on the 21cm continuum emission from surrounding extended nebulae have been determined for five of the pulsars and one of the neutron star candidates; extended emission towards one pulsar and one neutron star candidate have been analyzed and found to be positional coincidences. HI spectra and maps reveal compact emission possibly associated with the pulsar PSR J2029+3744.
The objects currently analyzed are pulsars J0141+6009, J0147+5922, J0157+6212, J2013+3845, J2029+3744, J2037+3621, and two old neutron star candidates which are believed to be accreting material due to their motion inside a molecular cloud. More objects will be analyzed as the CGPS coverage is expanded.
The continuum data is visually inspected to look for evidence of emission, both compact and extended, associated with the neutron stars. HI spectra are extracted, sampling emission on size scales up to 5' in diameter, to look for compact HI associated with the neutron stars. HI cubes one degree in size are then visually inspected to both confirm the structures in the HI spectra and to look for HI structure which may be associated with the neutron stars but which was missed by the spectra.
In a manner similar to the search for HI associated with neutron stars, a search is being made for HI associated with planetary nebulae (PNe). Unlike the search for HI associated with neutron stars however, we know that HI is associated with some planetary nebulae (e.g. Taylor et al., 1989, 1990; Gussie and Taylor, 1995).
Approximately 70 PNe lie in the CGPS area. Analysis has begun on five objects, with no obvious associated HI having been detected. This may be a real effect, or it may simply be due to the large beamsize of the the DRAO data compared to the size of the PNe.
The results of the pilot project polarization study, which revealed Faraday rotation effects in the interstellar medium acting on the diffuse Galactic polarization, are soon to be submitted for publication in two papers. The first, to be submitted to Nature (Gray, Landecker, Dewdney, Taylor), discusses the enigmatic lenticular feature (ELF) coincident with the W5 HII region (Figure 1), which we argue to be due to a foreground "bubble" of excess electron density of as yet unknown origin. The second paper will be submitted to Astrophysical Journal (Gray, Landecker, Dewdney, Taylor, Willis, Normandeau), and will cover the widespread mottled background and the depolarizing effect of the HII regions W3/W4/W5 (Figure 2), effects which are attributed to Faraday rotation in the interstellar medium in and around the HII regions. These phenomena were discussed in Andrew Gray's presentation to the Science Meeting in April. Preprints of these papers will be available upon request.
Two recent additions to the polarization group are Jo-Anne Brown (PhD student) and Marta Peracaula (post-doc), both working with Taylor at U.Calgary. Jo-Anne visited DRAO in May to reduce polarization data from the N5 field (in the Cygnus X region), part of which shows clearly the impact of an ionized continuum filament on the diffuse Galactic polarization. By comparing the change in observed polarization through the filament, the electron density of which can be found from the total intensity emission, it will be possible to measure the magnetic field strength. Jo-Anne will present this work at the up-coming CASCA meeting in Edmonton. Marta is expected to arrive later this summer and work on theoretical aspects related to studies of the polarization from the Galaxy.
In addition to science goals, we are also investigating some technical issues that have an impact on the polarization data. In particular, the unknown ellipticity of the reference antenna's nominally circular feeds limits polarization image fidelity. Andrew has shown that, for the ideal case of noiseless data and perfectly orthogonal feeds, it is possible to use a source of known polarization (such as 3C286) to deduce and correct for the ellipticity parameters. Jo-Anne will continue with this analysis to investigate the practical issues of implementing the corrections for real data. A second issue that needs to be addressed is correcting the polarization data for off-axis effects. Some direct measurements were made at DRAO in summer '96 (by Andrew's summer student Nathalie Robitaille), and Tom Landecker has also been studying the relevant antenna theory. Finally, Faraday rotation effects in the ionosphere are known to introduce a small but significant day/night variation in observed polarization angle. A means of accounting for this effect is being investigated at DRAO.
As reported at the last CGPS meeting at the DRAO, I have begun exploring ways of determining spectral indices over large areas of the CGPS data. A first approach has been to produce spectral index maps using the ``convolution differential-spectral-index '' technique (Zhang et al. 1997), based itself on the use of so-called TT-plots. Tests on the A mosaic appear promising, even though at this stage the absence of short-spacing information limits the applicability of the method.
I have also experimented with a ``more dynamic'' technique developped by Katz-Stone (1997, Ph.D. thesis, Univ.~of Minnesota) and called ``spectral tomography'' whereby, from the observed map at 1420 MHz, a family of ``theoretical'' 408 MHz maps is generated by continuously varying the spectral index. The observed 408 MHz map is then subtracted from each member of this family. The spectral index of a given feature corresponds to the difference map at which this feature is absent. One interesting way of looking at the data is by loading the family of difference maps in ``newcube'' and scanning through the family, looking for features which flash out of view. This technique was successfully tested on the observations of the SNR CTA1. Contrary to the TT-plot technique however, this method requires prior removal of the background level.
Work on particular WR stars and SNRs awaits the inclusion of short-spacing data. However a preliminary examination of the continuum maps of the field including the WR star WR 143 shows no evidence of enhanced emission near the position of the star.
Star Formation was the title of a science team, one that was though to be a driver of the Galactic Plane Survey. The topic is broad in scope, but no specific project was listed for this on the registry. However, some ideas are being discussed on how the survey data can contribute to this topic.
After some discussion on whether the GPS data can be used to initiate new projects in the area of star formation, it was decided that they are probably best used as follow up. Two examples are: HI halos around molecular clouds, and interfaces next to star forming regions. The former has been studied to some degree by Wannier and Andersson, but aspects such as the contribution to gravitational binding by HI have yet to be explored. For the latter, CO data will be required and they are not yet publicly available.
Massive star formation is difficult to study. Since they comprise a minority of the total stellar population of our Galaxy, are formed preferentially in massive molecular clouds, and are short-lived, massive star forming sites are located quite far away and mostly near the Galactic Plane. Events move quickly when massive stars form - they run through their pre-main sequence phases very rapidly, and then destroy the evidence by rapidly dispersing the dense molecular regions in which they formed. The ionising radiation and strong winds which do this means that the environment in which massive stars form is complex and multi-phase (molecular, atomic, ionised). In contrast, low mass star formation is much more widespread, occurs in nearby clouds which can be found well away from the confusion of the Galactic Plane, proceeds at a rather more leisurely rate, and does not disrupt greatly the natal environment.
The formation of massive stars is vitally important to understand far beyond what their few numbers would suggest, however, since they make an enormous impact on the structure, dynamics, evolution, and chemistry of the interstellar medium in our own and in other galaxies.
How can the Canadian Galactic Plane Survey be used to study star formation? As far as low mass stars are concerned, it is a difficult proposition: the relatively inconspicuous low mass star formation activity is extremely difficult to study at large distances in the Galactic Plane, with confusing backgrounds, and with the large scale devastation wrought by the formation of their high mass cousins. For low mass stars, the most important data product would undoubtedly be the HI. The formation of low mass stars within relatively isolated molecular globules could plausibly be tackled by the CGPS: in the case of such objects we have a better hope of picking out the star formation related HI features from the general confusion. However, the HI would probably only serve as an adjunct for the study of these objects, which would be identified and studied primarily by other means (infrared and (sub)millimetre wavelengths). Amplifying on Bill McCutcheon's article, it is hard to see the CGPS HI database acting as a primary driving force for the study of low mass star formation. [Editor's note: this ties in with the topic of dissociating stars discussed elsewhere in this newsletter.]
For high mass star formation, many more possibilities exist, although we need to make a distinction between what we could call small scale and large scale phenomena. For small scale studies, by which we mean the study of the formation and early evolution of individual high mass stars, we have a situation similar to that as exists for low mass star formation: the CGPS most likely would find itself in a supporting role to studies primarily done at other wavelengths. This would apply to study of star forming molecular rims and cores on the edges of HII regions, for example.
Study of large scale phenomena, such as those of entire high mass star forming regions (i.e. W3, W4, W5) is probably where CGPS data will find greater prominence, although again it is critically important that molecular line data be folded into the mixture. Again echoing Bill, the relative lack of large scale molecular line data means that the regions that can be effectively studied as well as the size scale available, will be rather limited. The study of star formation on large scales is (at least at present) somewhat outside my research expertise.
A joint meeting of the Stellar Wind and SNR groups was held at the Penticton consortium gathering. The following people were present: Shantanu Basu, Sean Dougherty, Neb Duric, Mike Fich, Lloyd Higgs, Doug Johnstone, Magdalen Normandeau, Serge Pineault, Dave Routledge, Brad Wallace, and Heinz Wendker. And this is what they had to say:
Two specific approaches were discussed. The first one would consist in producing spectral index maps using the ``convolution differential-spectral- index '' technique (discussed in Zhang et al.~(AA 1997, in press)). Examples of this technique were presented at the meeting by Lloyd (SNR G78.2+2.1) and Serge (A mosaic). The second one would involve using the ratio of IR to radio emission, shown by Furst et al.~(AAS, 71, 63 (1987)) to be a good discriminator between thermal and non-thermal emission. As pointed out, this last method is the only one allowing one to separate flat-spectrum SNRs from thermal emission features. It was pointed out that the low-order-spacing information which will be included in the CGPS continuum data (Effelsberg and Haslam Survey data) will be exactly the same as that already used by others to study spectral index variations over wide angular regions on the sky. The fine angular structure available from the all-spacings CGPS data set, however, will be new.
Of the people present, those who expressed an interest in producing these new maps were (with varying degrees of interest, the following order more or less reflects this fact): Serge, Neb, Heinz and Lloyd. It was suggested that the next step would be to select interesting features (candidates for deeper study) from the spectral-index maps, and to list them. Consortium members coud then select features from the list, and investigate them fully; when a member selected a feature from the list he or she would announce to the consortium at large that the feature was being studied.
We have formed a de facto theory group -- another complementary science team -- devoted to theoretical understanding of various ISM properties or processes whose study is motivated by the CGPS. It is hoped that this group will operate in collaboration with and as a resource for the other science teams.
Our group's interests include: Massive Star - ISM connection (Basu, Johnstone, Kerton, Martin), bulk kinematic model of the Galaxy (Johnstone, with Fich), dynamical MHD processes in the ISM (Basu), theory of star formation (Basu, Johnstone, Martin), and radio polarization properties (Johnstone). A third-person description of work in the first four areas is given below.
Ballantyne (U Vic co-op student), Kerton, and Martin have been exploring the merits of multifrequency analysis of H II regions and their environs using the data base being assembled in the Galactic plane survey. They have made a study of KR 140 to the SW of W3, which appears to be a region of spontaneous star formation. In many ways it is a textbook example of the development of an H II region. It has a shell structure, and the parent CO cloud has a clearly developed cavity. This project also illustrates the interesting new information on the distribution of hot dust (25 micron) and PAH molecules (12 micron) that can be obtained using HIRES processing at these shorter wavelengths. Consequently, they are contemplating adding another survey product, HIRES at these shorter wavelengths, to complement what is already available from the IGA.
Martin is a part of an HST collaboration studying the physical structure and composition of the Orion nebula.
Johnstone has investigated the last stages of star formation, particularly the destruction of the circumstellar disk, with a recent focus on objects in the Orion nebula. While primarily oriented to low-mass stars, models for ultracompact H II regions have also been developed.
There are no results yet concerning the ISM environs of Wolf-Rayet stars. For a description of the project (scientific justification, outline, list of participants and list of stars), please consult the Wolf-Rayet Web page put together by Nicole St-Louis.