 |  |
Astron. Astrophys. 357, 61-65 (2000)
4. Available data and SSP templates
Stellar clusters in the Large Magellanic Cloud provide a convenient
and, in practice, unique set of templates for young and intermediate
age stellar populations spanning a wide range of metallicities
(between 1/100 and solar, e.g. Sagar
& Pandey 1989). The available [CO] data are summarized in
Fig. 3. The large scatter of the points at a given age reflects
variations of metallicities, statistical effects related to the
intrinsic small number of luminous red stars in the clusters and,
possibly, field contamination (see e.g. Chiosi et al. 1986,
Santos et al. 1997). A further complication occurs beyond
600 Myr, when carbon stars
appear. These often display a very red spectrum with strong continuum
emission from the envelope which dilutes the CO bands and produce an
anti-correlation between J-K colours and [CO] index (Persson et al.
1983). Note in particular that the mild anti-correlation between CO
index and age visible in Fig. 3 also reflects the fact that the
older clusters are, on average, less metallic than the youngest
one's.
![[FIGURE]](img58.gif) |
Fig. 3. Observed [CO] in different samples of stellar clusters and galaxies. The first panel includes old Galactic globular clusters (Origlia et al. 1997, equivalent widths scaled to using Eqs. (1) and (2)), note the quite tight correlation between CO index and metallicity. The second plot includes younger clusters in the LMC, open circles are from the photometric data of Persson et al. (1983) while the filled symbols are from the spectroscopic equivalent widths of Oliva & Origlia (1998), scaled to using Eqs. (1) and (2). The large scatter of the points at a given age reflects variations of metallicities, statistical effects related to the intrinsic small number of luminous red stars in the clusters and, possibly, field contamination (see also Sect. 4). The third panel shows the distribution of the CO indices observed in starburst galaxies. The data are from Oliva et al. (1995, 1999, equivalent widths transformed into using Eqs. (1),(2)) and from Shier et al. (1996), Goldader et al. (1997) and Doyon et al. (1994) whose "broad spectroscopic index" is translated into using the prescriptions in the latter reference. Note that virtually all of the starbursters are within the range covered by - yr old LMC clusters.
|
For the purposes of this paper, the most important fact is that
several of the LMC clusters with ages
100 Myr display [CO] indices much
larger than those predicted by the models based on the Geneva's
tracks. In other words, a highly metallic stellar population of
100-1000 Myr could have a [CO] similar to that of a 10-100 Myr cluster
of the same metallicity, as indeed predicted by models including the
whole AGB evolution (see Sect. 3).
Therefore, finding a galaxy with a very deep CO index does not
necessarily imply that its stellar population must be younger than 100
Myr, as sometimes assumed in literature.
Fig. 3 also shows results from a wider set of data, including
old stellar systems (Galactic globular clusters and ellipticals) and
starburst galaxies. In general, the only clear observational result is
that objects with [CO] significantly larger than 0.18 cannot be old
stellar systems of (sub)solar metallicities, but require younger
stellar populations or, alternatively, old stellar populations much
more metallic than ellipticals. Encouragingly, several starburst
galaxies do indeed display CO indices significantly larger than the
above threshold but, in most cases, within the range covered by LMC
clusters of ages 109 yr
(see Fig. 3).
On the other hand, however, other well studied starbursters have
values of [CO] lower than 0.18 and more similar to ellipticals and
bulges. This probably reflects metallicity variations, i.e. weaker CO
features can be associated with young stellar systems of lower
metallicities.
© European Southern Observatory (ESO) 2000
Online publication: May 3, 2000
helpdesk.link@springer.de  |