Matter power spectra in dynamical-Dark Energy cosmologies

We used a suite of numerical cosmological simulations in order to investigate the effect of gas cooling and star formation on the large-scale matter distribution. The simulations follow the formation of cosmic structures in five different dark energy models: the fiducial Λcold dark matter (ΛCDM) cosmology and four models where the dark energy density is allowed to have a non-trivial redshift evolution. Each simulation includes a variety of gas physics, ranging from radiative cooling to UV heating and supernova feedback [although the active galactic nuclei (AGN) feedback is not incorporated]. Moreover, for each cosmology we have a control run with dark matter only, in order to allow a direct assessment of the effect of baryonic processes. We found that the power spectra of gas and stars, as well as the total matter power spectrum, are in qualitative agreement with the results of previous works not including the AGN effects in the framework of the fiducial model, although several quantitative differences exist. We used the physically motivated halo model in order to investigate the backreaction of gas and stars on the dark matter distribution, finding that it is very well reproduced by simply increasing the average dark matter halo concentration by 17 per cent, irrespective of the mass. This is in agreement with the cooling of gas dragging dark matter in the very centre of haloes, as well as adiabatic contraction steepening the relative potential wells. Moving to model universes dominated by dynamical dark energy, it turns out that they introduce a specific signature on the power spectra of the various matter components, which is qualitatively independent of the exact cosmology considered. This generic shape is well captured by the halo model if we blindly consider the cosmology dependences of the halo mass function, bias and concentration. However, the details of the dark matter power spectrum can be precisely captured only at the cost of a few slight modifications to the ingredients entering in the halo model. The backreaction of baryons on to the dark matter distribution works pretty much in the same way as in the reference ΛCDM model, in the sense that it is very well described by an increment in the average halo concentration. None the less, this increment is less pronounced than in the fiducial model (only ˜10 per cent), in agreement with a series of other clues pointing towards the fact that star formation is less efficient when dark energy displays a dynamical evolution.