CFHT MEGACAM Imaging Survey Proposal A Deep Optical Survey of the Taurus and Orion Molecular Clouds C. Dougados, Laboratoire d'Astrophysique, Obs. Grenoble, FRANCE co-Is: J.L. Monin, Laboratoire d'Astrophysique, Obs. Grenoble, FRANCE F. Menard, Canada-France-Hawaii Telescope Corp., Hawaii, USA E. Caux, Centre d'Etudes Spatiale des Rayonnements, Toulouse, FRANCE We propose to obtain with MEGACAM deep optical images of 2 of the closest prototypical star forming regions: the Taurus and Orion molecular clouds. Two modes of star formation seem to exist: a distributed mode associated mostly to diffuse, low-mass star forming regions and a clustered one observed in denser and more massive environments. The Taurus and Orion regions sample these two extreme conditions and therefore play a central role in our understanding of the star formation process. Their typical angular extent on the sky range from 30 to 100 square degrees. To date less than 1% of the total cloud surface areas have been imaged in the visible down to typical sensitivities of I=19. A significant fraction of the low luminosity stellar population is thus still missing in both regions. We propose to increase by 1 to 2 orders of magnitude the area covered and improve by a factor 3-5 the sensitivity of actual surveys, thus providing the most complete optical census to date (and for a long time) of the low luminosity embedded population on the scale of a whole molecular cloud. Many outstanding issues regarding the star formation process could be addressed with the help of these data: - Probe the low mass end of the IMF down to the H-burning limit and investigate wether differences occur in different cloud populations. Young star forming regions represent one of our best opportunity to determine the initial mass function as they probe co-eval star populations which still keep track of their initial conditions. - Investigate the spatial distribution of the low luminosity sources, especially their clustering properties and relation to the massive stars. Wether the vast majority of stars form in a cluster environment is still an open question, which has important bearing on the evolution of the protostellar environment as well as formation of multiple systems. Studying star formation regions with and without high mass stars is therefore important for comparison purposes. - Study the star formation history throughout the cloud: what initiates the star formation process ? Does star formation take place simultaneously and in bursts over the whole cloud or is it a continuous process ? Do low and high mass stars form simultaneously ? Large scale optical imaging surveys appear as a necessary complement to the on-going near-IR surveys like 2mass (JHK) and Denis (IHK) with typical completeness limits of I=18, J=16, H=15 and K=14.5. In the case of young stellar objects, the R and I wavelength domains are the least affected by circumstellar excess emission, thus allowing the most accurate determination of photospheric effective temperature and luminosity. A reliable estimate of these two parameters is essential for deriving individual ages and masses. In addition, while the near-infrared surveys will remain more sensitive in the most obscured regions of the cloud (Av > 5 typically), an optical survey reaching sensitivites of I=20-21 would allow to probe the very low mass population on the cloud surface more efficiently. The Taurus dark cloud, with total gaz mass of 10^4 msun, is among the closest star-forming region to the sun at a distance of 140 pc. Its extension on the sky is 100 deg^2. It mainly forms low-mass stars, distributed loosely across the cloud surface or gathering in small aggregates of 10-15 members. Our scenario for individual star formation is almost entirely based on the study of individual young stars associated with this cloud. To date, the census of the population is complete down to I=12.5 (Kenyon & Hartmann 1995), corresponding to a stellar mass of 0.3 msun at the age of the Taurus population (4 Myrs). A significant fraction of the very low mass objects is thus still missing. The average low extinction (Av=1) associated with this cloud, as well as its young age, combine to provide a high sensitivity to very low mass objects in the optical. According to current evolutionary track predictions (Baraffe et al. 1998), reaching sensitivities of I=20 would allow to detect stars with masses down to 0.02 msun, ie well into the brown dwarf regime. 2 degr^2 have been recently imaged by Briceno et al. (1998) down to I=19, R=21.5 and revealed 9 new very low-mass T Tauri candidates. A deep photometric survey of a region typically 30 deg^2 would allow to increase by at least an order of magnitude the numbers of very low-mass objects known in this cloud and to obtain the FIRST RELIABLE ESTIMATE OF THE IMF WELL INTO THE BROWN DWARF REGIME IN A STAR FORMING REGION. The Orion A and B molecular clouds are the closest giant molecular clouds (d =450 pc) and active massive star forming regions. They each extend over typically 30deg^2 with total gas mass of 10^5 msun. Their associated stellar population spans the whole spectrum of stellar masses up to 30 msun. They harbor rich dense clusters, the most prominent being the Orion Nebular Cluster (ONC) and most likely represent the typical environment in which a galactic star forms. 5 regions 1/2 degree each in extent (including the ONC) have been mapped at optical wavelengths in the Orion A cloud (Allen 1996, Hillenbrand 1997) down to sensitivities of I=19.3, R=20.2. The Orion B cloud has been significantly mapped only at near-IR wavelengths (Lada 1992). The average extinction across both clouds is typically < Av=5, although it can reach up to 20-100 mag in the most obscured regions (ONC and dense CS cores). Mapping down to I=21 would allow to detect all embedded stars down to the hydrogen burning limit, with ages < 10 Myrs (typical lifetime of a GMC) and Av < 5 typically, thus providing for the first time the opportunity to STUDY THE LOW MASS STAR FORMATION PROCESS ON THE GLOBAL SCALE OF A GIANT MOLECULAR CLOUD. One proposed observing strategy: --------------------------------- The proximity of these Star Formation Regions assures us adequate angular resolution to avoid overcrowding problems (except in the core of Orion). At least two red filters are required in order to isolate low mass pre-main sequence candidates in a color-magnitude diagram. R and I are very efficient probe of low mass objects and least contaminated by circumstellar excess emission. Alternatively z' could be used in place of I or maybe R. These options require further investigation and in particular examining predictions from current PMS tracks for low mass objects. A third filter could be observed: V or B for isolating high accretion candidates. We wish to reach completeness limits of the order of I=20-21 and (R-I)=2.5. Accurate photometry (errors < a few %) as well as astrometry are required. 30 MEGACAM field would be typically observed towards each cloud. With 2 filters and an average of 1 hour total on-source integration per field, 30 Hours on source integration would be required per cloud. Observing the 3 regions would require a total of 90 Hours, ie roughly 11 nights. This time would be increased if a third filter is used. Contamination by galactic field stars could be a potential problem: in the Taurus case, Briceno et al. (1998) have shown that contamination by field stars above the 10 Myr isochrone at similar sensitivities is 50 %. It will be significantly less in Orion. Nearby off-cloud fields should thus be observed and used in combination to predictions form current galactic population models to assess statistical contamination. Follow-up optical spectroscopic observations of suspected new low mass pre-main sequence candidates will also be necessary. Extrapolating from current population statistics, the estimated number of new low-mass candidates that such a survey would reveal is of the order of a few 100s in the Taurus case and a few 10000s for the Orion GMC's. The database provided should include astrometry and photometry for all the pre-main sequence candidates detected. Group expertise: ---------------- Our group has extensive experience in star formation studies and has developped in the recent years expertise in IMF studies of young clusters (see for example Carpenter et al. 1997, Giovannetti et al. 1998). We recently participated in a proposal for a european research network (TMR) centered on young cluster studies (PI: M.McCaughrean). In addition, we are currently getting familiar with the reduction and analysis process of large scale optical imaging data through an on-going project centered on observations of the NGC2264 region obtained with the CFH12k camera (F. Menard observer). The list of co-Is is still preliminary and could be significantly extended if the project developes. Potential candidates are: L. Allen (Cfa, Cambridge, USA), L. Hillenbrand (Caltech, USA). References: ---------- Allen L., 1996, PhD thesis, University of Massachusetts. Baraffe, I. et al., 1998, A&A, 337, 403 Bouvier J. et al. 1998, A&A, 336, 490 Briceno, C. et al, 1998, AJ, 115, 2074 Carpenter J. et al. 1997, AJ, 114, 198 Giovannetti et al., 1998, A&A, 330, 990 Hillenbrand L., 1997, AJ, 113, 1733 Hillenbrand L. et al. 1998, AJ, 116, 1816 Kenyon S. & Hartmann L., 1995, ApJS, 101, 117 Lada E., 1992, ApJL, 393, 25 Note: A related proposal centered on outflow detection via deep line imaging is proposed by Menard et al. . MEGACAM working group: --------------------- F. Menard and/or J.L. Monin would be interested in participating in this working group as a representant of the star formation community.