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

4. Discussion

4.1. Tomography and the nature of the steep-spectrum `sheath'

It has recently been suggested (e.g. Katz-Stone & Rudnick 1997) that the jets in some FRI radio galaxies have a two-component structure, consisting of a flat-spectrum `core jet' and steep-spectrum surrounding `sheath'. Katz-Stone et al. (1999) show that the same picture may apply to two WAT sources from the sample of O'Donoghue et al. (1990). The observed spectral steepening with distance from the core in FRI sources might therefore be unrelated to spectral ageing and expansion, as is frequently assumed; it might simply be a consequence of the increasing dominance of the sheath component.

To test whether such a picture is viable in 3C 130, I constructed a spectral tomography gallery as discussed by Katz-Stone & Rudnick; this involves generating a set of maps by subtracting a scaled version of the high-frequency map from the low-frequency map, so that for each pixel of the tomography map () we have

where is varied. Features of a given spectral index vanish on the tomography map corresponding to that spectral index; if the apparent steepening in 3C 130 is due to varying blends of a flat- and steep-spectrum component, and the steep-spectrum component is relatively smooth, the plumes should appear more uniform in a tomography map with a spectral index corresponding to that of the flat-spectrum component, as the flat-spectrum component should then have vanished, leaving only a (scaled) version of the steep-spectrum component. If there is no single, uniform flat-spectrum component, the plumes will still show structure for any value of .

The full gallery of tomography images is not shown, but Fig. 5 shows a representative example, made with the L and X-band maps taking . It will be seen that the jets and N hotspot are oversubtracted, giving rise to negative flux densities on the tomography map - this is as expected, since their spectral index is about 0.5 (Paper I). In the N plume, there is still considerable structure in this image, but the S plume has a much more uniform surface brightness after subtracting the flat-spectrum component, suggesting that a two-component model of the source is close to being adequate here. This is further illustrated in Fig. 6, which shows the results of spectral tomography on slices across the S plume; these show that, at least within 1.5 arcmin of the core, the plume can be modelled as a superposition of a flat-spectrum component with and a broader steep-spectrum component with , with the flat-spectrum component becoming progressively fainter with distance along the plume; this is consistent with the results of Katz-Stone et al. (1999). The spectrum of the flat-spectrum component, as estimated from the spectral index at which it disappears on tomography slices, appears to have steepened by 105 arcsec from the core; this is true even after a rough correction is applied for the effects of the undersampling of the X-band data on large spatial scales (as assessed in Sect. 3.1).

Fig. 5. A tomography slice between 1.4 and 8.4 GHz with

Fig. 6. Tomography slices across the S plume taken at (from top) 40 arcsec, 75 arcsec and 105 arcsec from the core. The solid lines show the surface brightness as a function of distance across the plume (integrated along 8-arcsec strips) at 1.4 GHz (upper line) and 8.4 GHz (lower line); the dashed lines show the results of tomographic subtraction, starting at the top with and proceeding in steps of 0.05 in spectral index. Note that for the nearer two slices the most uniform surface brightness after subtraction is given by - 0.6. For the furthest slice the best tomographic spectral index appears steeper.

The situation is certainly more complicated in the N plume, where there is in any case less evidence for a steep-spectrum sheath in the spectral index maps of Paper I; if a two-component model is to be viable there, it must allow for some spatial variation in the spectrum of the flat-spectrum component. But this would not be surprising, since there is much stronger evidence for ongoing particle acceleration in the N plume. I return to this point below.

If there are two spectral components, what is the origin of the steep-spectrum material? Katz-Stone & Rudnick identify several possibilities for the sheath in 3C 449. There may be a two-component jet, with the steep-spectrum material only becoming visible at a flare point; or the steep-spectrum material may have evolved from the flatter-spectrum component through ageing, adiabatic expansion, diffusion into a region of lower magnetic field or a combination of these. Without additional low-frequency observations it is impossible to say whether the injection spectral indices of the two components are the same, so we cannot rule out a two-component plume in 3C 130. But it is certainly also possible that the sheath has evolved from the flatter-spectrum component. Modelling of the synchrotron spectrum does not allow me to rule out any of the possibilities; the sheath may be substantially older than the flat-spectrum jet, or it may be of comparable age and in a weaker magnetic field, or a combination of the two. It is possible to say that the two regions cannot simultaneously be in local energy equipartition and be the same age if they have aged in the same B-field.

In any case, it is clear that the steepening of the overall spectrum of the plumes with distance from the source, as discussed in Paper I, is better modelled in terms of a two-component spectral model than in terms of spectral ageing along the jet.

4.2. The high-frequency spectra of the plumes

The striking difference between the high-frequency spectra of the N and S plumes (Fig. 4) is unusual in radio galaxies, particularly in a source as symmetrical at low frequencies as 3C 130. It is, of course, possible that the symmetry is illusory and that for some reason the electrons in the S plume are moving much more slowly, and therefore appear to be ageing much more rapidly, than those in the N plume. However, it seems more likely that the spectral difference is related to particle acceleration in the plumes.

In the S plume, there is no clear evidence in any single-frequency map or in the polarization maps for a compact hotspot like the one seen in the N of the source. The two-frequency spectral index maps presented in Paper I show the flat-spectrum S jet penetrating the S plume for some distance, but do not show any particularly flat-spectrum termination region; the best candidate region was in the area of maximal surface brightness at arcsec from the core. From those data it seemed possible that there was a hidden compact hotspot, perhaps suppressed by Doppler beaming, and that the particle-acceleration situations in the two plumes were nevertheless symmetrical. But the 15-GHz data taken together with the absence of a hotspot suggest a model in which there is currently little or no shock-related particle acceleration in the S plume, and consequently no shock-related termination of the jet. The steep 8.4-15-GHz spectrum is inconsistent with continuous injection models for the electron spectrum. If we assume for the ageing B-field the equipartition field of 0.46 nT used in Paper I, and (as in that paper) use a Jaffe & Perola (1973) aged electron spectrum then we can estimate the time for which particle acceleration must have been turned off to produce the observed spectrum of the southern plume (Table 1) from an initially power-law spectrum with , as observed in the northern hotspot; it is of order years. This is an appreciable fraction of the commonly assumed lifetime of a radio source, but it is strongly dependent on the assumed ageing B-field. (Note that, because the region of flux measurement is defined on 15-GHz maps, the plume spectrum used here is essentially that of the flat-spectrum component discussed above, and does not include a contribution from the steep-spectrum sheath.)

© European Southern Observatory (ESO) 1999

Online publication: September 2, 1999
helpdesk.link@springer.de