It will probably be awhile before NGC 3606 YC can be recommended as a candidate globular in the usual sense of a compact gravitationally bound stellar object. Presently 3603’s C-M diagram is a near linear track straight up the MS; few of its stars have moved off the main sequence. Its 130+ very massive O and somewhat less massive B stars have not reached the turn-off point. There are three Wolf-Reyets, though, the same number as the 5-times large R136 cluster in the Tarantula Nebula, but 3603’s are a full magnitude brighter. UV radiation from the O and B giants have cleaned out most residual star-forming gas; we can readily see this in visual-band astro images of the cluster. Equally important to the professional astronomers with 8m class equipment, the hot O and B winds smooth reddening across the core, or rather de-lumpify it, which makes distancing more accurate. Most of 3603’s <Msol stars appear to have been formed primordially out of a collapsing original gas cloud of 10,000+ Msol; the O and B stars then formed dynamically from cloud shocking during a rapid mass collapse in the core; the time scale between these two starform periods is uncertain but likely abrupt. The two nearby dust pillars we see in published images on the S and W are succumbing to the withering UV bombardment they’re getting from the cluster core. In half-million to a million years will come the shock and MHD turbulence of multiple core collapse SNs, probably a large number of them in close succession. The agonies of M82 show what happens to a galaxy-scale object after a large number of closely spaced SN spasms; now imagine what’s in store for this rather modest cluster. 3603’s overly bright future is to be expected given the near-instantaneous core formation documented by Brandner’s 2012 paper.* When SN shocking does occur, the combined supersonic shock wave and interstellar gas heating will considerably alter the gravitational gradient of 3603’s remaining sub-8 Msol stars; the likely scenario will be something akin to a core collapse globular on a modest 5000-star scale.
*Robert, were you citing Brandner's
2012 paper or the
2010 paper?
Presently 3603 lies between the Carina Arm and the denser Scutum-Centaurus Arm of the galaxy. As it transits to the next arm it will encounter collapsing new clusters and high-velocity molecular clouds whose younger cousins we see so readily as dark nebulae in wide-field astrophotos of 3603’s surroundings today. Presently there is a moderate overweighting of O versus B stars in the SW quadrant at sub half-parsec distances; the Bs are more evenly spread in the E and N. The O stars will SN first. While we don’t know how the SNe in 3603’s <2myr future will affect the cluster’s galactic orbit, a sector-specific SN light-up could nudge the cluster into a lissajous orbit above and below the disc like the Sun experiences over its 2.3 gyr cycle. This would add orbital crossing shocks to 3603’s already fulminous future.
Hence, as we observe it today, we are seeing a doomed cluster about to lose mass on a significant scale. (And I think MY diet is awful!) First will come core-collapse SNs and expulsion of regional star-forming gas; then core collapse of the low-mass stars remaining; and finally, in the really long term, PN gas expulsion as a population of WDs emerge. All these factors appear to preclude 3603 becoming a GC. It might end up a dense open cluster of the NGC 6719 type, with a negligible surface evaporation of old low-mass stars and WDs shocked by density wave and disc crossings.
I am not sure about the MPIA article’s interpretation, ‘a mass of more than 10,000 suns packed into a volume with a diameter of a mere 3 light-years’. Brandner’s paper isn’t about size or astrometry considerations, it’s about the
cluster’s primordial or dynamical segregation and timescale. For population, stellar class, and distance accuracy, I would stick with the
Melena, Massey et al 2008 paper, which uses spectroscopic parallax, main-sequence fitting, and kinematic distancing based on the MW rotation, to establish 3603’s DM at 18.7-19.1. Also see
Stolte and Brandner et al 2006, which directly compares 3603 with the R136 cluster in 30 Doradus in the LMC, a cluster pair that is five times as massive and >50 times more spread out. There are so many papers about 3603 it is entertaining to check out the chronology of all the et.als who have chanced on this visually unimpressive cluster. That’s how I discovered it, too: by accident.
The discussion about 3603’s classification serves to put Robert’s point in a larger picture: things are getting so complicated that old, easy paradigms are on shakier ground by the day. For example, the term “globular cluster” no longer means a disc or halo aggregation of metals-thin Pop II stars with a large C-M variance.
NGC 1818 in the LMC is classified as a GC even though it is less than 40 myr old and has a complex near-solar metals content. The matter of 3603’s brand-label classification category is modest compared with its origin, evolution till now, near-term future, and above all, what we can learn about it as we glimpse it through our modest backyard observer eyes, AP gear, and spectroscopes. Consider what 3603 and its associated giant H2 complex and the Tarantula Nebula in the LMC must look like to some home-brew astronomer with a 300 mm scope living on an inconspicuous (and hopefully not cloud-bound) planet somewhere in M33: Pretty much what NGC 604 and NGC 588 in M33 look like to us. There’s a little Herschel in all of us who look up into the night and never come back.