We would dramatically limit our understanding of bulges if we confined the analyses to their central regions. Substantial progress has been brought about by spatially resolved spectroscopy, which enables radial gradients of stellar populations to be measured. Investigations of such radial gradients using large surveys have until recently been addressed only to early-type galaxies, i.e. elliptical and lenticular galaxies, with only very modest and rare excursions into the case of later-type galaxies (e.g. Sansom et al. 1994; Proctor et al. 2000; Ganda et al. 2006).
Moorthy & Holtzman (2006) recently published a large study of 38 bulges, composed of about equal numbers of nearly face-on spirals and highly inclined ones. For most of their bulges they found steady decreases in metallicity-sensitive indices with radius and a positive increase in [a/Fe], with the exception that their small bulges have generally weak gradients or no gradient, sometimes positive ones. While age gradients are generally absent for their sample galaxies, some exhibit positive ones, the majority of those being barred spirals. Very interestingly, they found a correlation between line-strength gradients in the bulge and in the disk.
Most of these qualitative results have been confirmed by Jablonka et al. (2007), who chose a different strategy. In order to get rid of the disk population totally, they selected 32 genuinely (or nearly) edge-on spiral galaxies with Hubble types from S0 to Sc. They obtained spectra along the bulge minor axes, out to the bulge effective radius and often much beyond. Most of their bulges do present radial stellar-population gradients. The outer parts of bulges do have weaker metallic absorption lines than the inner regions. The distribution of the gradient amplitudes is generally well peaked, but they also display a real intrinsic dispersion, implying a variety of star-formation histories.
Mgb AC4668/Alog R
Figure 27.3. Characteristics of bulges, showing clear radial changes in their spectral indices. Colors code Hubble types from S0 to Sbc. Left panel: the arrows join the values of the indices (Fe> and Mgb (expressed in terms of magnitude) at the bulge effective radius and in a central (r = 2 arcsec) aperture. Thomas et al. (2003) grids of single-stellar-population models are shown at [a/Fe] = 0.0 and 0.3, for ages between 3 Gyr and 15 Gyr and metallicities from -0.33 to +0.35. Right panel: model lines of variations in pure metallicity (plain lines) and pure age (dotted lines) are derived. They directly predict the relation between pairs of index gradients as observed in the bulge spectra.
The left panel of Figure 27.3 illustrates the decrease in [a/Fe], from the bulge effective radii to their central regions, for galaxies exhibiting clear gradients. Just like in the case of the central spectral properties, the morphology of the parent galaxy is not a driving parameter of the gradient amplitudes. The right panel of Figure 27.3 shows the models of pure metallicity (plain lines) and pure age variation (dotted lines) from Thomas et al. (2003). Bulges populate the region close to the pure-metallicity lines, leaving some room, but of a much smaller magnitude, for [a/Fe] and age variations. A quantitative analysis indicates that radial gradients in luminosity-weighted mean metallicity are twice (on a logarithmic scale) the gradients in age. While [Fe/H] at the bulge effective radii is on average 0.4 dex lower than in the bulge central regions, the age difference is of the order of 1.5 Gyr, the inner regions being younger. The variations in [a/Fe] are small (of the order of 0.1 dex) and rather constant among bulges. These various points indicate that the outer regions of bulges reveal their earliest stages of star formation. Interestingly, the sensitivity of the gradients to the central velocity dispersion is very different from what has been reported for the bulge central indices. Literally, there is no correlation between the gradient amplitude and the bulge central-velocity dispersion. Instead, one sees that bulges with large velocity dispersions can exhibit either strong or negligible gradients. The probability of a strong gradient diminishes with decreasing velocity until it reaches zero. The same had been observed for elliptical galaxies.
This gradual build-up of the index versus a relation can be clearly observed only for indices with large dynamical ranges, such as Mg2 and Mg1.
Was this article helpful?