3. BULGE TYPES

The early allusion by Hubble to ellipticals originated the concept of bulges as scaled down ellipticals, or simply ellipticals surrounded by disks. I will show that this concept is erroneous, at least for massive bulges. Nevertheless, some bulges share properties with ellipticals, and these define the classical concept of galaxy bulges. In the current literature, one can find three different stellar systems referred to as bulges. (In fact, they are all photometric bulges.) Let us briefly discuss them, starting with the classical connotation.

Classical bulges. These systems are not as flat as disks, i.e. they stick out of the disk plane when seen at sufficient inclinations. They are somewhat spheroidal (which is hard to see at low inclinations), featureless (no spiral arms, bars, rings etc.), contain mostly old stars (not much dust or star-forming regions), and are kinematically hot, i.e. dynamically supported by the velocity dispersion of their stars, . In the current framework, they are thought to form via mergers (i.e. accretion of usually smaller external units) in violent events, inducing fast bursts of star formation if gas is available. This depends on the orbit configuration of the merger event. In many cases, the accreted material does not reach the galaxy center, but stays in the outer halo. Figure 5 shows an example of a classical bulge.

Figure 5. Classical bulge in M81. [Credit: NASA, ESA and the Hubble Heritage Team (STScI/AURA).]

Disk-like bulges. These systems are also referred to as pseudo-bulges. They are as flat (or almost as flat) as disks, which might be difficult to see in very inclined galaxies. They may contain sub-structures such as nuclear bars, spiral arms or rings. They usually show signs of dust obscuration, younger stellar populations or ongoing star formation, and, finally, they are kinematically cold, i.e. dynamically supported by the rotation velocity of their stars, V_rot. These systems seem to be built mostly via disk instabilities, such as bars (but also possibly spiral arms, ovals or lenses), in a relatively slow, continuous and smooth process. Essentially, such instabilities induce a redistribution of angular momentum along the galaxy, and, as a result, mostly gas but also stars are driven to the disk center (Athanassoula 2003, Sheth et al. 2005). Recent work has shown that the current star formation is enhanced in the centers of barred galaxies (e.g. Ellison et al. 2011, Oh et al. 2012), and that the distribution of mean stellar ages in bulges of barred galaxies has a peak at low ages, absent for unbarred galaxies (Coelho & Gadotti 2011, see also Pérez & Sánchez-Blázquez 2011) in agreement with this scenario. Figure 6 shows an example.

Figure 6. Disk-like bulge in NGC 6782. [Credit: NASA, ESA and the Hubble Heritage Team (STScI/AURA).]

Box/Peanuts. These systems stick out of the disk plane and show a boxy or peanut-like morphology. They are usually featureless and show no signs of dust obscuration, young stellar populations or star-forming regions. They are also kinematically cold and usually referred to as pseudo-bulges. A number of studies have shown that these structures are just the inner parts of bars that grow vertically thick due to dynamical instabilities (e.g. Combes & Sanders 1981, de Souza & Dos Anjos 1987, Kuijken & Merrifield 1995, Bureau & Freeman 1999, Merrifield & Kuijken 1999, Lütticke et al. 2000, Chung & Bureau 2004, Bureau & Athanassoula 2005). Figure 7 shows an example. Although box/peanuts are photometric bulges, they are just the inner parts of bars, and not a distinct physical component. They have basically the same dynamics and stellar content as bars, just their geometry is somewhat different. As such, the term `box/peanut bulge' is a misnomer. Note that box/peanuts are not seen if the galaxy is not inclined enough. In a face-on galaxy, if it has a box/peanut, it will be seen as part of the bar. Therefore, in bulge/bar/disk decompositions such as in Gadotti (2009), box/peanuts are accommodated in the bar model. It is worthy to point out that the Milky Way shows a box/peanut, a fact known since the 1990's when the COBE satellite flew (see Fig. 8), a clear indication that the Galaxy has a bar. Another remarkable case is that of M31, known to have a bar, with its box/peanut inner part (see Athanassoula & Beaton 2006, see Fig. 9). One should also be aware of the rare thick boxy bulges (see Fig. 10, Lütticke et al. 2004). These seem to be present in only 2 per cent of disk galaxies. They are too big to be parts of bars and are thought to be built via mergers, possibly still at an ongoing stage.

Figure 7. Box/peanut in ESO 597-G036. [Credit: NASA, ESA and the Hubble Heritage Team (STScI/AURA).]

Figure 8. The COBE image of the Milky Way. (Credit: COBE Project, DIRBE, NASA.)

Figure 9. Top: Spitzer 3.6 µm image of M31. Bottom: residual image after subtraction of a 2D bulge/bar/disk model derived with budda (de Souza et al. 2004, Gadotti 2008). The X-shape in the residual image is the typical signature of a boxy/peanut-like vertically thickened inner part of a bar.

Figure 10. Thick boxy bulge in ESO 510-G13. Compare it with the box/peanut in ESO 597-G036 (Fig. 7) and that in the Milky Way, shown in Fig. 8. [Credit: NASA, ESA and the Hubble Heritage Team (STScI/AURA).]

Concerning the dynamical support of bulges, it is known for long that, although classical bulges have little rotational support, they do rotate more significantly than ellipticals (e.g. Kormendy & Illingworth 1982). In addition, box/peanuts rotate even more significantly, as one would expect from the fact that these are actually bars (e.g. Kormendy 1993). Plotting V_rot / as a function of the ellipticity of the system, , has proven in these works to be a powerful way to assess dynamical support. More recently, the SAURON team (see e.g. Emsellem et al. 2004, Falcón-Barroso et al. 2006, Ganda et al. 2006) performed powerful 2D kinematical analysis of bulges and ellipticals. Although their results are evidence that the central regions of galaxies are far more complex than understood before, they generally corroborate such previous conclusions. In addition, the SAURON team (see also Williams et al. 2011) found that box/peanuts rotate cylindrically, as predicted from theoretical studies on bars (e.g. Athanassoula & Misiriotis 2002).