5.2. Intermediate-age/Old Populations and Chemical Inhomogeneity

Whereas young populations in dIrrs tend to be fairly homogeneous, old and intermediate-age stellar populations show considerable metallicity spreads. In part this may be due to the large age range sampled here, and the difficulty of assigning ages to individual field stars. Metallicity spreads have mainly been derived based on the color width of the red giant branch in color-magnitude diagrams, or via metallicity-sensitive photometric systems (e.g., Cole, Smecker-Hane, & Gallagher 2000; Davidge 2003). These methods have the drawback that they are affected by the age-metallicity degeneracy. Near-infrared CaII triplet spectroscopy is now increasingly being employed for general [Fe/H] derivations instead. In the LMC, red giants in different parts of the galaxy show significantly different mean abundances (Cole et al. 2000, 2004). Cole et al. conclude that in the LMC azimuthal metallicity variations may in part be due to different fractions of bar and disk stars sampled at different positions (with the bar stars being younger and more metal-rich). With regard to field populations, star clusters have the advantage of consisting of well-datable, single-age populations. Old globular clusters in the LMC may differ substantially in metallicity (Olszewski et al. 1991). Interestingly, there is also evidence for a radial abundance gradient in the LMC old cluster population, i.e., a trend for old clusters to be more metal-rich closer to the LMC's center (Da Costa 1999). In the SMC, there are indications that intermediate-age star clusters of a given age may occasionally differ by a few tenths of dex in [Fe/H] (Da Costa & Hatzidimitriou 1998; Da Costa 2002), which would indicate considerable chemical differences in the enrichment of their birth clouds - possibly due to infall. However, refined age determinations and more spectroscopic abundance determinations are needed to verify the SMC intermediate-age metallicity spread.

In NGC6822 and IC1613, abundance spreads among the field red giants have been confirmed spectroscopically (Tolstoy et al. 2001; Zucker & Wyder 2004); again, age uncertainties remain. Although the present-day field star metallicity in NGC6822 lies between those of the LMC and the SMC, the cluster metallicities tend to lie below those of the SMC (Chandar, Bianchi, & Ford 2000; Strader et al. 2003; see Fig. 3). One needs to caution that the cluster measurements are based on a number of different studies and methods. Also, the present-day metallicity of NGC6822 may have been enhanced by the recent interaction-triggered star formation episode (see Section 3.2).

Figure 3. Age versus metallicity for clusters and field stars in the SMC (circles) and in NGC 6822 (squares). The diagram for the SMC was adopted from Da Costa (2002) and comprises both spectroscopic and photometric abundances. The data points for NGC 6822 are based on a variety of different measurements and methods (clusters: Cohen & Blakeslee 1998; Chandar, Bianchi, & Ford 2000; Strader, Brodie, & Huchra 2003; field: Muschielok et al. 1999; Tolstoy et al. 2001; Venn et al. 2001) and should not be used to derive a quantitative age-metallicity relation. The solid-line box denoting the mean metallicity of NGC6822's red giant field population is shown for an assumed age of 10 Gyr, while the much larger dashed box indicates the spread in metallicity and gives a rough idea of the possible age range.