Results and discussion

From the comparison between our model predictions (Figure 28.1) and the observed G-dwarf-like diagrams derived for various radii by Harris & Harris (2002, Figure 18) for the elliptical galaxy NGC 5128, we can derive some general considerations. The qualitative agreement is remarkable: we can explain the slow rise in the [Z/H] distribution as the effect of the infall, whereas the sharp truncation at high metallicities is the first direct evidence of a sudden and strong wind that stopped the star formation. The suggested outside-in formation process is reflected in a more-asymmetric

[Z/H]

Figure 28.1. 'G-dwarf' distributions for [Z/H] in luminosity (solid line) and mass (dotted line). Left panel: values at 0.1 Reff. Right panel: values at 1 Reff. The plots are presented to the same scale in order to allow the reader better to appreciate the differences among the distributions.

Figure 28.1. 'G-dwarf' distributions for [Z/H] in luminosity (solid line) and mass (dotted line). Left panel: values at 0.1 Reff. Right panel: values at 1 Reff. The plots are presented to the same scale in order to allow the reader better to appreciate the differences among the distributions.

shape of the (G-dwarf diagram at larger radii, where the galactic wind occurs earlier (i.e. closer to the peak of the star-formation rate), with respect to the galactic centre.

From a quantitative point of view, properties such as the stellar metallicity distribution of the combined-stellar population (CSP) inhabiting the galactic core, allow us to study the creation of a mass-metallicity relation (see Chapter 44), which is typically inferred from the spectra taken at ~ 0.1 times the effective radius. In Figure 28.2 we have plotted the time evolution of the mass-metallicity relation of the stars (which reflects the average chemical enrichment of the galactic core as seen at the present day; dashed line) and that in the gas (which, instead, is closer to the composition of the youngest SSPs, thus being more indicative of a high-redshift object; solid line). The mean Fe abundance in the stellar component can reach the Solar value in only 0.5 Gyr, making ellipticals among the most metal-rich objects of the Universe.

On the other hand, at variance with the G-dwarf-like diagrams as a function of [Z/H] (and [Fe/H]), abundance ratios such as [a/Fe] have narrow and almost-symmetric distributions. This means that, also from a mathematical point of view, the [(a/Fe)] ratios are representative of the whole CSP (PMC06). The robustness of the [a/Fe] ratios as constraints for the galactic formation history is testified to by the fact that [(a/Fe)] ~ [(a/Fe)y], with very similar distributions. In particular, we find that the skewness parameter is much larger for the [Z/H] and [Fe/H] distributions than for the case of the [a/Fe] one, by more than an order of magnitude. Moreover, the asymmetry increases on going to large radii (see Figure 28.1, right panel), by up to a factor of ~7 with respect to the inner regions. Therefore, it is not

Figure 28.2. The temporal evolution of the mass-metallicity relation for the two galactic components studied (stars and gas).

Figure 28.2. The temporal evolution of the mass-metallicity relation for the two galactic components studied (stars and gas).

surprising that the [(Z/H)] value does not represent the galaxy at large radii. Hence, we stress that care should be taken when one wants to infer the real abundances of the stellar components for a galaxy by comparing the observed indices (related to a CSP) with the theoretical ones (predicted for a SSP). Only the comparison based on the [(a/Fe)] ratios seems to be robust.

Another possible source of discrepancies is the fact that luminosity-weighted averages (which are more closely related to the observed indices) and mass-weighted averages (which represent the real distributions of the chemical elements in the stellar populations) might differ more in the most external zones of the galaxy (compare the panels in Figure 28.1). All these considerations result in the fact that the chemical-abundance pattern used by modellers to build their SSPs might not necessarily reflect the real trends. Therefore, the interpretation of line-strength indices in term of abundances can be seriously flawed (see PMC06 for further details).

The analysis of the radial variation in the CSPs inhabiting elliptical galaxies seems to be promising as a powerful tool to study ellipticals. Pipino, Puzia & Matteucci (in preparation) make use of the G-dwarf-like distributions predicted by PMC06 to explain the multimodality in the globular-cluster (GC) metallicity distribution as well as their high a enhancement (Puzia et al. 2006). In particular, preliminary results show that the GC distribution as a function of [Fe/H] for the whole galaxy can be constructed simply by combining distributions such as those of Figure 28.1 (typical of various radii), once they have been rescaled by means of a suitable function (of time and metallicity) that links the global star-formation rate to that of GC creation. Neither an enhanced rate of GC formation during mergers nor a strong role of the accretion of external objects seems to be required in order to explain the various features of the GC metallicity distributions.

Since GCs are the closest approximation to a SSP, we expect that this technique will be a very helpful one to probe the properties of the stellar populations in spheroids, thus avoiding the uncertainties typical of analysis based on their integrated spectra.

Was this article helpful?

0 0

Post a comment