The origin of wide-angle tailed radio galaxies

Michael R. Merrifield 0 0 School of Physics and Astronomy, University of Nottingham, University Park , Nottingham NG7 2RD 1 Mullard Space Science Laboratory, University College London , Holmbury St Mary, Dorking, Surrey RH5 6NT A B S T R A C T To investigate the origins of wide-angle tailed radio sources (WATs), we have compiled a sample of these systems in Abell clusters for which X-ray data exist. Contrary to conventional wisdom, the WATs are found to be significantly displaced from the X-ray centroids of their host clusters. The bends in the radio jets of WATs are found to be oriented preferentially such that they point directly away from or toward the cluster centre, with more of the former than the latter. If this morphology is attributed to ram pressure, then the WATs are on primarily radial orbits, with more approaching the X-ray centroid than receding. There is also some evidence that the incoming WATs are on average further from the X-ray centroid than the outgoing ones. All of these observations strongly support a scenario in which WATs are created in cluster mergers. I N T R O D U C T I O N Wide-angle tailed radio galaxies (WATs) form a class of radio galaxies, usually found in clusters, whose radio-emitting jets have been bent into a wide C shape. This structure gives the immediate impression that the jets are being swept back by the dynamic pressure resulting from the motion of the associated galaxy through the surrounding intracluster medium (ICM). This ram pressure model was first developed by Begelman, Rees & Blandford (1979), and studied in more detail by Vallee, Bridle & Wilson (1981) and Baan & McKee (1985). Unfortunately, there is a piece of evidence that seems to contradict this intuitively appealing model. WATs are usually associated with the brightest clusters ellipticals (D or cD galaxies), and these galaxies are generally found at rest, close to the centres of clusters (Quintana & Lawrie 1982; Bird 1994; Pinkney 1995). From a theoretical point of view, this finding can be understood, since models of cluster formation imply that large galaxies form close to their cluster centres (Bode et al. 1994; Garijo, Athanassoula & Garca-Gomez 1997). Even if a massive galaxy were initially placed on a high-velocity orbit that carried it far out in its cluster, dynamical friction would rapidly drag it down to rest at the centre of the system (Ostriker & Tremaine 1975). Since it seems that the D/cD galaxies that host WATs should lie at rest in the centres of their clusters, they should not possess the motion required to produce the observed bends in their radio jets by ram pressure. It has therefore been thought necessary to invoke alternative mechanisms to explain the observed bends in jets of WATs. One candidate for this mechanism is an electromagnetic force arising from the interaction between a jet that carries a net electrical current and the magnetic field in the ICM (Eilek et al. 1984). Given our poor understanding of currents in jets and magnetic fields in clusters, this model has not been extensively explored. One problem with it is that it requires a highly and favourably ordered magnetic field in order to produce the symmetric shape of WATs. Alternatively, jets could be deflected by collisions with dense clouds in the ICM. Although this process may be at work in some radio galaxies whose jets are deflected and disrupt abruptly (Burns et al. 1986), again it has difficulty reproducing the large-scale symmetric structure of WATs. Thus, neither of the suggested alternative jet-bending mechanisms is entirely satisfactory. A possible solution to this dilemma has come from the realization that clusters are dynamically young, and merge frequently. Theoretical and observational studies have forced us to discard the idealized picture of a spherical relaxed cluster that is isolated and does not interact with its surroundings. Instead, structure in the Universe is now viewed as evolving hierarchically, with large feature such as clusters forming through the repeated mergers of smaller groups (e.g. Evrard 1990; Jing et al. 1995; Frenk et al. 1996). It has therefore been suggested that the galaxy motions required to bend WATs by ram pressure are a by-product of collisions between clusters (Pinkney, Burns & Hill 1994; Loken et al. 1995; Go mez et al. 1997a,b). Consider a radio galaxy located at the centre of a cluster. If this cluster collides with a second comparable system, then the collisional nature of the ICM means that the kinetic energy of the gas will rapidly dissipate, and the two separate gaseous components will merge into a single structure. The radio galaxy, on the other hand, is an essentially 00 35 17.19 01 10 20.45 01 23 27.55 01 41 19.20 01 46 30.66 02 55 02.99 02 55 03.08 03 27 32.25 06 47 54.58 06 58 52.48 07 05 21.39 08 03 05.14 08 36 13.50 08 38 06.74 09 08 32.61 09 09 48.50 10 11 26.68 10 25 47.94 11 08 54.20 11 31 16.25 11 59 30.41 12 00 34.13 12 21 07.38 12 25 33.20 12 27 (...truncated)