Evolving cooling flows

Time-dependent models of cooling flows in elliptical galaxies are presented. In these models distributed mass drop-out is not permitted and we follow Ciotti et al. in assuming that the rate of supernova heating declines faster than the rate at which dying stars inject mass into the ISM. Consequently, the models move from outflow to inflow and ultimately a central cooling catastrophe. We investigate the dependence of this evolution upon the current supernova rate, upon its extrapolation into the past, and on the intergalactic pressure. We find that the evolution is very sensitive to intergalactic pressures of the expected order, in support of the conjecture of Ciotti et al. that differences in the asymptotic pressures on galaxies of similar luminosity Lopt account for the large scatter in the (Lx, Lopt) plane. The X-ray brightness profiles of the models are compared with observations of NGC 4472, NGC 4649 and NGC 4636. Rough fits are obtained for the first two galaxies, but NGC 4636 cannot be even roughly fitted, probably because it is well past its first cooling catastrophe. We model mass and energy input by a central black hole. We assume that the hole powers a subrelativistic jet when excited by catastrophic cooling at the centre of the flow. Radio observations of NGC 4472 and of the Milky Way indicate that most of the jet’s power goes to heating the ISM rather than to producing synchrotron radiation. Short sharp bursts of power from the jet transform the centre of the cooling flow while injecting only a small fraction of the energy injected by stars. Given that the synchrotron luminosity of the jet is probably orders of magnitude smaller than the jet’s mechanical luminosity, the efficiency with which even a small mechanical output by the jet modifies the X-ray emission suggests that the X-ray emission of most giant elliptical galaxies has been significantly affected by nuclear activity. Heating by the jet changes the cusped X-ray brightness profile characteristic of a cooling catastrophe into a cored profile. Once the central hole has switched off, the core again cools catastrophically within 0.5 Gyr. If the jet does not induce star formation, each successive cooling catastrophe is more violent than the last. In reality it seems likely that a quasi-steady state is reached as a result of mass drop-out and star formation at the interface of jet-heated and inflowing material. In this state, short bursts of nuclear activity alternate with periods of quiescent cooling. The duration of the quiescent periods is set by the distance to which the jet penetrates. This is in turn set by the jet’s power.