Virgo is the nearest rich cluster to our galaxy. The cluster is quite irregular and spiral rich (de Vaucouleurs, 1961; Abell, 1975; Bautz and Morgan, 1970), although the X-ray emission comes from an elliptical-rich core surrounding the galaxy M87 (Figure 1a).
M87 is classified as a peculiar elliptical galaxy. It has fairly extended optical emission (de Vaucouleurs and Nieto, 1978), but is not a cD. This galaxy is one of the brightest radio sources in the sky. Nonthermal radio emission is observed from the nucleus and from the prominent optical jet (see Figure 23), and both the nucleus and jet also produce nonthermal X-ray emission (Schreier et al., 1982; Figure 25). There is also a larger scale radio halo source with a size comparable to that of the entire galaxy (Andernach et al., 1979; Hanisch and Erickson, 1980).
The X-ray emission from this cluster is strongly concentrated to the region around M87, and is much less broadly distributed than the galaxies in the cluster (Malina et al., 1976; Gorenstein et al., 1977; Fabricant and Gorenstein, 1983). Figure 24 shows the Einstein IPC X-ray image of this cluster; Figure 25 is the HRI image. The top panel shows the outer contours, while the bottom panel is an expanded view of the center, showing X-ray emission from the jet. The emission is also considerably cooler (Tg 2.5 keV) than is usually associated with cluster emission (Lea et al., 1982). Nearly all of the emission comes from a 1° region around M87, in which more than 95% of the optical emission comes from M87. As a result, it seems reasonable to assume that the bulk of the emission is associated with M87 itself, rather than the cluster as a whole.
Figure 24. The X-ray surface brightness of the M87/Virgo cluster of galaxies as observed by Fabricant and Gorenstein (1983) with the IPC on the Einstein satellite. The lines are contours of constant X-ray surface brightness. The X-ray emission is centered on M87.
Similar arguments led Bahcall and Sarazin (1977) and Mathews (1978b) to suggest that the X-ray emitting gas is gravitationally bound to M87 itself. Bahcall and Sarazin showed that M87 could only bind the gas if it had a very massive halo, with a total mass of 1 - 6 × 1013 M. This estimate was based on early low resolution X-ray observations. More recent Einstein observations (Fabricant et al., 1980; Fabricant and Gorenstein, 1983; Stewart et al., 1984a) appear to support this model and provide a more accurate estimate of the mass of 3 - 6 × 1013 M. These observations measure, at least approximately, the temperature gradient in M87 and allow a direct determination of the mass profile of the galaxy. The massive halo must extend out to roughly 1° from M87. The optical surface brightness of the galaxy is very low in this region, and thus the mass-to-light ratio of the material making up this halo must be rather large. These observations suggest that at least this one elliptical galaxy possesses a massive, dark missing mass halo of the type discussed in Section 2.8. Whether this is typical of giant ellipticals or whether it is a consequence of M87's position at the center of the Virgo cluster is uncertain.
Figure 25. A higher resolution X-ray image of the M87/Virgo cluster emission centered on the galaxy M87, from Schreier et al. (1982). The lines are contours of constant X-ray surface brightness. (Upper), a lower spatial resolution version, showing the peaking of the emission on the center of M87, and its asymmetrical structure. (Lower), a blowup of the center of the galaxy, showing X-ray emission along the optical jet.
Alternatively, Binney and Cowie (1981) have suggested that the mass of M87 might actually be rather small. They argue that the cooler (Tg 2.5 keV) gas providing the bulk of the X-ray emission is confined by the pressure of a hotter (Tg 8 keV), lower density, intracluster medium.
A key prediction of the Binney-Cowie model is the existence of a significant amount of hot gas (Tg 8 keV) surrounding M87. The density of this gas is determined by the requirement of pressure equilibrium with the cooler gas in M87. Early observations (Davison, 1978; Lawrence, 1978) suggested that there was extended, hard X-ray emission in this cluster. More recent observations (Mushotzky et al., 1977; Ulmer et al., 1980a; Lea et al., 1981, 1982) have not confirmed its existence, and it has been suggested that the previously observed hard X-ray emission is entirely from the nucleus of M87. The observed temperature gradient in M87 is apparently not consistent with the Binney-Cowie model (Fabricant et al., 1980; Fabricant and Gorenstein, 1983; Stewart et al., 1984a).
A host of X-ray lines have been detected from M87 (see Section 4.3). Both Fe L and K lines have been observed (Serlemitsos et al., 1977; Lea et al., 1979), as well as lines from lighter elements. The observations indicate that the abundance of these elements is reasonably uniform within M87 (Fabricant et al., 1978; Lea et al., 1982). The gas temperature appears to be roughly constant at projected radii of 5 arc min, and decreases rapidly within this radius. In the inner regions, large amounts of emission by quite cool gas are observed in the higher resolution spectra (Canizares et al., 1979, 1982; Lea et al., 1982). The X-ray surface brightness is also strongly peaked in this region.
The presence of a range of temperatures of cool gas and the central peak in the X-ray surface brightness both suggest that the gas in M87 is radiatively cooling and being accreted onto the center of the galaxy (Gorenstein et al., 1977; Mathews, 1978b). Comparisons between models for the accretion (Mathews and Bregman, 1978; Binney and Cowie, 1981) and the observations suggest that the accretion rate is 3 - 20 M / yr. Like NGC1275 in Perseus, M87 has optical line emitting filaments in its inner regions (Ford and Butcher, 1979; Stauffer and Spinrad, 1979), which may be produced by thermal instabilities in the cooling gas (Fabian and Nulsen, 1977; Mathews and Bregman, 1978). As discussed above for NGC1275, the radio source may be powered by further accretion of a small fraction of the gas onto the nucleus (Mathews, 1978a), while the bulk of the accreted material may form low mass stars (Sarazin and O'Connell, 1983). Both the line emission and halo radio emission are concentrated to the north side of M87. De Young et al. (1980) suggest that M87 is moving at 200 km/s relative to the intracluster gas it is accreting, but Dones and White (1985) show that the thermodynamic structure of the gas is inconsistent with this motion.
Recently, Tucker and Rosner (1983) have suggested a hydrostatic (no accretion) model for M87. The gas in the outer portions of the galaxy is heated by the nonthermal electrons that produce the halo radio emission. Heat is conducted from these hotter outer regions into the cooler central regions, where it is radiated away; unlike Binney and Cowie (1981), Tucker and Rosner assume full thermal conductivity. The heating by nonthermal electrons balances the cooling, and the gas is in thermal equilibrium and hydrostatic. The radio source is itself powered by accretion; thus they argue that the behavior of the system is episodic, with alternating periods of accretion and nonthermal heating.
Many other galaxies in the central regions of the Virgo cluster have been detected as X-ray sources with luminosities in the range Lx 1039-41 ergs / s (Forman et al., 1979). Although no spectra are available for these galaxies, it is likely that the X-rays are due to thermal emission from hot gas. Figure 26 shows the X-ray emission from the very optically luminous galaxy M84 and the galaxy M86, which has an X-ray plume extending away from M87. Fabian et al. (1980) suggest that M86 contains hot gas, which is currently being stripped by ram pressure as it moves into the core of the cluster (see Sections 2.10.2 and 5.9). The interpretation of the X-ray emission from M87 and the other galaxies in the Virgo cluster is discussed in more detail in Section 5.8.
Figure 26. The X-ray emission from the Virgo cluster galaxies M86 (east or left) and M84 (west or right) from Forman and Jones (1982). Contours of constant X-ray surface brightness are shown superimposed on an optical image of the galaxies. Note the plume of X-ray emission extending to the north of M86, which may indicate that this galaxy is currently being stripped of its gas.