NGC 5128 is a massive elliptical galaxy at the heart of a moderately rich group of galaxies (Sect. 1.4); basic data are summarized in Table 3. Although most of its properties are fairly normal for a luminous triaxial elliptical galaxy, it is remarkable in two aspects: it hosts a very large radio source (Sect. 2) and its inner parts harbour a relatively massive disk of dust, gas and young stars (Sect. 4). Both have been proposed as the consequence of past merger activity. The location of the galaxy amidst several dwarf galaxies lends plausibility to such a suggestion. Indeed, although no direct evidence of a merger has been found, the appearance of the galaxy and in particular the properties of its embedded disk, such as gas mass, kinematics, warp and polar orientation along the photometric minor axis (cf. Bertola et al. 1988) as well as the bimodality of its globular cluster system (Zepf & Ashman 1993; Perelmuter 1995), the existence of luminous optical and and HI shells as well as the outcome of various dynamical models and scenarios (Sect. 6.2) all provide strong indirect evidence that at least one major merger event occurred some 108-109 years ago. The gas mass of the dusty disk, a few times 109 M, point at capture of a fairly sized late-type spiral galaxy rather than a small irregular. The shell structures in particular suggest that NGC 5128 experienced more than just one merger (Weil & Hernquist 1996).
The radio source Centaurus A, associated with NGC 5128, is a very near example of a large class of radio galaxies of moderate luminosity known as FR-I galaxies (Fanaroff & Riley 1974). Radio sources of this class are generally presumed to have moderately active nuclei with relativistic outflows on a subparsec scale not aligned with our line of sight. The observations of Centaurus A at radio and X-ray / ray wavelengths are consistent with this interpretation. NGC 5128 contains a very compact nucleus of size 1200 ± 500 A.U. from which subluminal relativistic jets emanate (sect. 5.4) that become subrelativistic within 1.5 pc. The jets appear to propagate at a large angle to our line of sight. The nucleus itself is strongly obscured by a small (radius 150 pc) circumnuclear disk (Sect. 5.2) and is quite variable at radio and X-ray wavelengths (Sect. 5.6). The polarization of the central region, the ionization of the optical filaments and the apparent similarity of the high-energy spectrum (but not the luminosity) of Centaurus A (in particular at -ray energies) to that of blazars and quasars such as 3C273, have been used to argue that the galaxy indeed harbours a misaligned BL Lac/blazar nucleus (Bailey et al. 1986; Morganti et al. 1991; Dermer & Schlickeiser 1993; Kinzer et al. 1995; see also Steinle et al. 1998, and references therein). The substantially lower luminosity of Centaurus A is then explained by our viewing the galaxy from the side, and not down the jet axis. However, some caution to this conclusion has been expressed by Antonucci & Barvainis (1990) and Kellerman et al. (1997).
|(1950)0||13:22:31.6 ± 0.2||Giles 1986|
|(1950)0||-42:45:30.3 ± 0.4||Giles 1986|
|Galactic Longitude l||309.5||degrees|
|Galactic Latitude b||+19.4||degrees|
|Systemic Velocity VHel||543 ± 2||km s-1||Table 1|
|Galaxy Size D25||18 × 14||arcmin||RC2|
|Radio Source Size||8 × 4||degrees||Section 2.1|
|Distance||3.4 ± 0.15||Mpc||Section 1.2|
|Apparent Magnitude B||7.96||mag||RC2|
|Colour (B - V)T||0.98||mag||RC2|
|Foreground Reddening E(B - V)||0.11||mag||Section 3.4|
|Total Galaxy Mass||4 ± 1 × 1011||M||Mathieu et al. 1996|
|Total HI Mass||8.3 ± 2.5 × 108||M||Section 4.2|
|Gas Mass Dusty Disk||1.3 ± 0.4 × 109||M||Section 4.2|
|Gas Mass Circumnuclear Disk||1.1 ± 0.3 × 107||M||Section 5.2|
|Outer Radio Lobe||250||kpc||Section 2.1|
|Middle Radio Lobe||30||kpc||Section 2.2|
|Inner Radio Lobe||5||kpc||Section 2.3|
|Inner Radio Jet||1.35||kpc||Section 2.3|
|Relativistic Nuclear Jet||1.65||pc||Jones et al. 1996|
|Radio Core||0.008||pc||Kellerman et al. 1997|
|Radius Dusty Disk||7||kpc||Section 4.1|
|Radius Circumnuclear Disk||150(-40, +130)||pc||Section 5.2|
Nuclear activity must have been going on for a considerable amount of time, given the size of the outer radio lobes. The bulk speeds of 5000 km s-1 estimated for the inner jets (Sect. 2.4) and the outer radius of 250 kpc of the giant lobes of radio emission suggest a lower limit of 50 million years. As the inner jets appear to dissolve into plumes ("inner lobes") at about 5 kpc from the nucleus (Sect. 2.3), and as the position angle of the outer radio features is much different from that of the inner features, it is reasonable to conclude that significant deceleration occurs over most of the radio source, leading to a substantially higher age. Indeed, the age of the inner lobes alone was already estimated at 6 × 108 years (Slee et al. 1983), although this may be too high. The jets appear to lose much of their energy within a few parsec from the nucleus, presumably by interaction with ambient material. The peculiar radio brightness evolution of component C1 in the nuclear jet may provide a clue to this process (Tingay et al. 1998) underscoring the need for further VLBI monitoring of Centaurus A, as well as the desirability of filling the resolution gap in the 0.1-0.3" range. The inner jets dissolve in the more extended inner lobe plumes, which exhibit a profound clockwise bending (decreasing position angle). Again, ambient material and its movement in the galaxy, may explain the observed morphology (see e.g. Sparke 1982; Gopal-Krishna & Saripalli 1984; Heckman et al. 1985), but hard evidence is lacking. Moreover, this is unlikely to also explain the similarly profound clockwise bending of the giant outer lobes, well outside the optical galaxy. Noting a continuous decrease of position angle (i.e. clockwise bending) of various features at increasing distance to the nucleus, Haynes et al. (1983) have proposed that the central collimating source precesses at a rate of the order of 10-5 degrees per year. The discovery of a circumnuclear disk perpendicular to the nuclear and inner jet, yet inclined to the minor axis of the elliptical galaxy, supports the idea of precession. If the rate of procession is correctly estimated, the structure of the radio source should exhibit the effect of several precessional periods. It would be interesting to see whether the run of position angles with radius can indeed be modelled by such a precession of the collimating agent. Alternatively, a combination of precession and ambient gas dynamics may be required, while the structure of the outer lobes, in addition, may be influenced by tumbling and orbital motion of the galaxy as a whole (Burns et al. 1983).
The putative age of the merger (Sect. 6.3) suggests a link to the origin of the radio source. Although the presence of an active nuclear source predating a merger cannot be excluded, it is tempting to associate its origin with the accumulation of matter in the centre caused by transfer of angular momentum through viscous damping after such an event. An intriguing indication that the origin of the radio source is connected to merger activity is provided by a morphological argument. The bended appearance of the giant radio lobes (Fig. 4) is very similar to that of the tilted rings forming the dust band (cf. Fig. 11b in Nicholson et al. 1992) rotated by 90°. For instance, the position angle of 0° characterizing the outer radio contours corresponds to the position angle of 90° of the outer rings. Because the dynamical time scale of the outer rings is much longer than that of the more strongly tilted inner rings, their present position angle should more closely resemble the original orientation of the inner disk structure at the time that the matter now forming the outer lobes was ejected. As the dust disk originated in a merger event, and the morphology of the radio lobes appear to follow its subsequent evolution, it seems likely that the activity creating these lobes is also a consequence of the merger event.
If the nuclear source is in fact a black hole, its estimated bolometric luminosity (half of it at high energies) of about 1043 erg s-1 implies a lower limit to the black hole mass of about 5 × 104 M whereas the total luminosity of the radio source suggests a mass 107 M (Terrell 1986 and references therein; Kinzer et al. 1995). As the dynamical mass within 40 pc is about 4 × 108 M, and there is no obvious sign of Keplerian rotation (see Fig. 10), its upper limit must be a few times 107 M. Although the high obscuration of the centre of NGC 5128 and the lack of H2O masers (Braatz et al. 1996) precludes, at the moment, a more accurate mass determination, the actual mass of the putative black hole nevertheless is fairly well-constrained and is comparable to that of the circumnuclear disk (Sect. 5.2). This mass is not very high and infalling molecular clouds, especially dense cores, may penetrate deeply before being tidally disrupted. The variability of the nucleus may represent the accretion of individual stellar or cloud remnants onto the black hole triggering renewed jet activity (Sect. 5.4 through 5.6) and fueling the radio source. Details of these processes are not clear yet, but careful and frequent monitoring of Centaurus A at radio, X-ray and -ray wavelengths may provide important information. For instance, how does the nucleus drive the nuclear jets, and how are the relativistic nuclear jets transformed into the nonrelativistic inner jets? The circumnuclear disk (Sect. 5.2) does not seem capable of controlling the collimation of the nuclear jets, but its orientation exactly perpendicular to these jets, suggests that it is somehow connected with the collimating agent. Comparison of Centaurus A features with very-high resolution observations (HST, VLBA) of other active elliptical galaxies suffering less nuclear extinction, such as the ten times more distant NGC 4261 (e.g. Jones & Wehrle 1997, and references therein) may prove particularly fruitful.
Acknowledgement. It is a pleasure to thank David Malin, Do Kester, Norbert Junkes, Thijs van der Hulst, Stéphanie Côté, Steven Tingay, Jack Burns and Paul van der Werf for kindly supplying the illustrations in this review. I also would like to thank Paul van der Werf, Hans Bloemen, George Miley and in particular Tim de Zeeuw for critical comments on an earlier version of this work. The burden of literature searches was greatly relieved by the use of the NASA Astrophysics Data System (ADS) Astronomy Abstract Service.