The flip side of galaxy formation: a combined model of galaxy formation and cluster heating

Only similar to 10 per cent of baryons in the Universe are in the form of stars, yet most models of luminous structure formation have concentrated on the properties of the luminous stellar matter. Such models are now largely successful at reproducing the observed properties of galaxies, including the galaxy luminosity function and the star formation history of the universe. In this paper we focus on the 'flip side' of galaxy formation and investigate the properties of the material that is not presently locked up in galaxies. This 'by-product' of galaxy formation can be observed as an X-ray emitting plasma [the intracluster medium (ICM)] in groups and clusters. Since much of this material has been processed through galaxies, observations of the ICM represent an orthogonal set of constraints on galaxy formation models. In this paper, we attempt to self-consistently model the formation of galaxies and the heating of the ICM. We set out the challenges for such a combined model and demonstrate a possible means of bringing the model into line with both sets of constraints. In this paper, we present a version of the Durham semi-analytic galaxy formation model galform that allows us to investigate the properties of the ICM. As we would expect on the basis of gravitational scaling arguments, the previous model fails to reproduce even the most basic observed properties of the ICM. We present a simple modification to the model to allow for heat input into the ICM from the active galactic nucleus (AGN) 'radio-mode' feedback. This heating acts to expel gas from the X-ray luminous central regions of the host halo. With this modification, the model reproduces the observed gas mass fractions and luminosity-temperature (L-T) relation of groups and clusters. In contrast to simple 'pre-heating' models of the ICM, the model predicts mildly positive evolution of the L-T relation, particularly at low temperatures. The model is energetically plausible, but seems to exceed the observed heating rates of intermediate-temperature clusters. Introducing the heating process into the model requires changes to a number of model parameters in order to retain a good match to the observed galaxy properties. With the revised parameters, the best-fitting luminosity function is comparable to that presented in Bower et al. The new model makes a fundamental step forward, providing a unified model of galaxy and cluster ICM formation. However, the detailed comparison with the data is not completely satisfactory, and we highlight key areas for improvement.