2003A&A...403...83T

We estimate the time dependent emission of the bright knots detected in large scale (tens to hundreds kpc) jets and discuss how the existing radio-to-X-ray observations, coupled with the correspondingly long cooling timescales, pose strong constraints on the structure and dynamics of knots. We suggest that the phenomenology of the jet in 3C 273 as observed by HST, Chandra and VLA, and in particular the fast decrease of the flux outside knots, can be satisfactorily accounted for if the emitting plasma is clumped. Clumping of the knots can also explain the month-year variability timescale reported for the large scale emission in M87. We also point out that emission between the bright knots should be detected with the next generation of instruments (e.g. NGST). In fact the relatively intense X-ray emission discovered with Chandra suggests that jets (at least in powerful quasars) are still relativistic at large distances from the active nucleus. In such a case Compton scattering off the CMB photons is efficient even when electrons are cold or mildly relativistic in the comoving frame. This provides a powerful tool for testing the continuous vs discrete nature of the jet flow.