Centaurus A: multiple outbursts or bursting bubble?

We present new radio observations of the brighter region of the northern lobe (the northern middle lobe, NML) of Centaurus A obtained at 20 cm with the Australia Telescope Compact Array. The angular resolutions are ∼50 and ∼130 arcsec, therefore much higher than for the previously available radio images of this region. The most interesting feature detected in our images is a large-scale jet that connects the inner radio lobe and the NML, and which is imaged for the first time. The NML itself appears as diffuse emission with a relatively bright ridge on the eastern side. The radio morphology of Centaurus A and, in particular, its NML could be the result of a precessing jet that has undergone a strong interaction with the environment at least on the northern side. The very big drop in intensity between the inner jet and the large-scale jet can be explained with a sequence of bursts of activity at different epochs in the life of the source. Alternatively (or additionally) a ‘bursting bubble’ model is proposed to explain this big drop in intensity, which could also explain the good collimation of the large-scale jet. In this model, the plasma accumulated in the inner lobe would be able to ‘burst’ out only through one nozzle, which would be the region where the large-scale jet forms. The location of the nozzle would represent a region where the pressure gradient is more favourable. From the comparison between the radio emission and the regions of ionized gas discovered by Graham & Price (the so-called optical filaments) we find that the inner optical filament (∼8 kpc from the centre) falls about 2 arcmin (∼2 kpc) away from the large-scale radio jet. Thus, the inner filament does not seem to have experienced a direct interaction with the radio plasma. The complex velocity field observed in this filament could therefore be the result of strong instabilities produced by the ‘bursting bubble’. The outer filaments appear to be, in projection, closer to and aligned with the radio emission in the transition region between the jet and the lobe, suggesting a direct interaction with the radio jet. However, also in this case a more complicated interaction than assumed so far has to be occurring because of the relative position of the ionized and neutral gas regions compared with the radio jet as well as the kinematics of the ionized gas.