Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 349, 381-388 (1999)

Previous Section Next Section Title Page Table of Contents

5. Speculations on source models

If there is no explicit jet termination, how does this fit in with models for WAT formation? We can clearly see a well-collimated jet entering the southern plume. By analogy with the jets in FRIIs we believe this jet to be supersonic, and numerical modelling (e.g. Norman et al. 1988, Loken et al. 1995) has suggested that to make a WAT a shock should form at the end of the jet, giving rise to the characteristic flaring at the base of the plumes; in any case, we should see some evidence for a transition between the supersonic jet and the diffuse, trans-sonic plume. But there is no evidence for a jet-termination shock either in the form of a hotspot as in the northern plume, as discussed in Paper I, or in particle acceleration, as discussed above. How, then, does the southern jet terminate?

Perhaps the most attractive model is one in which the southern jet currently does not terminate, while the northern jet in 3C 130 is currently impinging on the edge of the source, causing a shock (Fig. 7); the terminations of both jets are dynamic and move about in the base of the plume, and it is only when one intersects with the plume edge that we see a shock and associated particle acceleration. Simply from the small-scale deviations from linearity seen in high-resolution maps of the jets, we know that the point at which the jet enters the plume must vary with time. In FRII radio galaxies, numerical simulations have shown (e.g. Norman 1996) that the working surface can move about as a result of turbulence in the cocoon. In WATs, bulk motions in the X-ray emitting medium on scales comparable to those of the jets may provide another source of jet buffeting.

[FIGURE] Fig. 7. Schematic of the possible current situation in 3C 130.

This sort of model for 3C 130 seems incompatible with a picture in which the flaring of WAT plumes is caused by propagation across a shock in the external medium (Norman et al. 1988) or into a crosswind (Loken et al. 1995), because in these models we would expect to see hotspots on both sides at all times. The latter model is in any case hard to reconcile with the straight plumes seen in 3C 130 and in some other sources which are morphologically WATs (Paper I). Instead, it may be the case that WATs of 3C 130's type are the natural result of the action of the external medium on a low-power FRII source.

The pressure in the lobes of an FRII is expected to fall with time (e.g. Kaiser & Alexander 1997, Eq. 20) and if it falls below the thermal pressure of the external medium, the lobes can no longer be supported and will begin to collapse. The natural result is a breaking of the source self-similarity and a slow crushing of the cocoon, which begins with the region closest to the centre, where the thermal pressure is highest (cf. Williams 1991). 1 In many FRII sources, X-ray observations show that the thermal pressure from the external medium is greater than the minimum pressure in the radio lobes; only the very smallest sources seem to be unambiguously overpressured. Cocoon crushing is therefore a viable process. Once the radio lobes have been squeezed away from the nucleus, asymmetries in the thermal atmosphere, together with buoyancy effects, can account for the large-scale distortions in the lobes seen in many low-power FRII galaxies (Williams 1991). But, if the environment is suitable, there seems to be no reason why buoyancy cannot drive this process further in particular sources, pushing the lobes further and further away from the centre. The detailed shapes of the sources this would produce would depend on the advance speed of the front of the lobe, but if the lobes were pushed out far enough we would start to see WAT-like objects, provided that the jets continued to terminate at the end of the lobe nearest the centre. At intermediate stages we would see objects like NGC 326 (Worrall et al. 1995), which differs from a WAT only in that its tails are slightly recessed from the termination of its jets. The evolutionary sequence from FRII to WAT is represented in Fig. 8. The bending of WAT tails on large scales can still, of course, be understood in terms of bulk motions of cluster gas.

[FIGURE] Fig. 8. An `evolutionary sequence' of double objects in clusters. Top left (3C 438): cocoon crushing has driven the radio emission away from the axis at the centre, backflow is deflected. Middle (NGC 326): outflow is in `tails' flowing sideways, perhaps bent by buoyant forces. Bottom (3C 465): the lobes have merged with the tails to form plumes; bulk motion in the cluster gas bends them. All maps are VLA images at [FORMULA] GHz. 3C 438 and 3C 465 are taken from the 3CRR atlas (Leahy et al. 1998); the image of NGC 326 was supplied by Mark Birkinshaw.

If true, this model would imply that we do not expect to see very young (small) WATs; they all evolve from FRIIs and need a certain amount of time (dependent on jet power and properties of the external medium) to do so. It is certainly the case that the WATs studied by O'Donoghue et al. (1993) show a range of core-`hotspot' distances that begins around 20 kpc and is much smaller than the range of core-hotspot distances seen in classical doubles. What is not clear is whether the timescales of the processes needed to push the lobes of radio galaxies out beyond the jet termination are short enough to be active here. More detailed information on the environments of WATs will become available with the launch of Chandra and XMM ; it will need to be coupled with detailed, fully three-dimensional simulations of jets in realistic atmospheres to answer all the outstanding questions on the dynamics of these sources.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1999

Online publication: September 2, 1999