Halo-based models have been successful in predicting the clustering of matter. However, the validity of the postulate that the clustering is fully determined by matter inside haloes remains largely untested, and it is not clear a priori whether non-v
irialised matter might contribute significantly to the non-linear clustering signal. Here, we investigate the contribution of haloes to the matter power spectrum as a function of both scale and halo mass by combining a set of cosmological N-body simulations to calculate the contributions of different spherical overdensity regions, Friends-of-Friends (FoF) groups and matter outside haloes to the power spectrum. We find that matter inside spherical overdensity regions of size R200,mean cannot account for all power for 1<k<100 h/Mpc, regardless of the minimum halo mass. At most, it accounts for 95% of the power (k>20 h/Mpc). For 2<k<10 h/Mpc, haloes with mass M200,mean<10^11 Msun/h contribute negligibly to the power spectrum, and our results appear to be converged with decreasing halo mass. When haloes are taken to be regions of size R200,crit, the amount of power unaccounted for is larger on all scales. Accounting also for matter inside FoF groups but outside R200,mean increases the contribution of halo matter on most scales probed here by 5-15%. Matter inside FoF groups with M200,mean>10^9 Msun/h accounts for essentially all power for 3<k<100 h/Mpc. We therefore expect halo models that ignore the contribution of matter outside R200,mean to overestimate the contribution of haloes of any mass to the power on small scales (k>1 h/Mpc).
We use cosmological hydrodynamical simulations to investigate how the inclusion of physical processes relevant to galaxy formation (star formation, metal-line cooling, stellar winds, supernovae and feedback from Active Galactic Nuclei, AGN) change th
e properties of haloes, over four orders of magnitude in mass. We find that gas expulsion and the associated dark matter (DM) expansion induced by supernova-driven winds are important for haloes with masses M200 < 10^13 Msun, lowering their masses by up to 20% relative to a DM-only model. AGN feedback, which is required to prevent overcooling, has a significant impact on halo masses all the way up to cluster scales (M200 ~ 10^15 Msun). Baryonic physics changes the total mass profiles of haloes out to several times the virial radius, a modification that cannot be captured by a change in the halo concentration. The decrease in the total halo mass causes a decrease in the halo mass function of about 20%. This effect can have important consequences for abundance matching technique as well as for most semi-analytic models of galaxy formation. We provide analytic fitting formulae, derived from simulations that reproduce the observed baryon fractions, to correct halo masses and mass functions from DM-only simulations. The effect of baryonic physics (AGN feedback in particular) on cluster number counts is about as large as changing the cosmology from WMAP7 to Planck, even when a moderately high mass limit of M500 ~ 10^14 Msun is adopted. Thus, for precision cosmology the effects of baryons must be accounted for.
The observed clustering of galaxies and the cross-correlation of galaxies and mass provide important constraints on both cosmology and models of galaxy formation. Even though the dissipation and feedback processes associated with galaxy formation are
thought to affect the distribution of matter, essentially all models used to predict clustering data are based on collisionless simulations. Here, we use large hydrodynamical simulations to investigate how galaxy formation affects the autocorrelation functions of galaxies and subhaloes, as well as their cross-correlation with matter. We show that the changes due to the inclusion of baryons are not limited to small scales and are even present in samples selected by subhalo mass. Samples selected by subhalo mass cluster ~10% more strongly in a baryonic run on scales r > 1Mpc/h, and this difference increases for smaller separations. While the inclusion of baryons boosts the clustering at fixed subhalo mass on all scales, the sign of the effect on the cross-correlation of subhaloes with matter can vary with radius. We show that the large-scale effects are due to the change in subhalo mass caused by the strong feedback associated with galaxy formation and may therefore not affect samples selected by number density. However, on scales r < r_vir significant differences remain after accounting for the change in subhalo mass. We conclude that predictions for galaxy-galaxy and galaxy-mass clustering from models based on collisionless simulations will have errors greater than 10% on sub-Mpc scales, unless the simulation results are modified to correctly account for the effects of baryons on the distributions of mass and satellites.
Upcoming weak lensing surveys, such as LSST, EUCLID, and WFIRST, aim to measure the matter power spectrum with unprecedented accuracy. In order to fully exploit these observations, models are needed that, given a set of cosmological parameters, can p
redict the non-linear matter power spectrum at the level of 1% or better for scales corresponding to comoving wave numbers 0.1<k<10 h/Mpc. We have employed the large suite of simulations from the OWLS project to investigate the effects of various baryonic processes on the matter power spectrum. In addition, we have examined the distribution of power over different mass components, the back-reaction of the baryons on the CDM, and the evolution of the dominant effects on the matter power spectrum. We find that single baryonic processes are capable of changing the power spectrum by up to several tens of per cent. Our simulation that includes AGN feedback, which we consider to be our most realistic simulation as, unlike those used in previous studies, it has been shown to solve the overcooling problem and to reproduce optical and X-ray observations of groups of galaxies, predicts a decrease in power relative to a dark matter only simulation ranging, at z=0, from 1% at k~0.3 h/Mpc to 10% at k~1 h/Mpc and to 30% at k~10 h/Mpc. This contradicts the naive view that baryons raise the power through cooling, which is the dominant effect only for k>70 h/Mpc. Therefore, baryons, and particularly AGN feedback, cannot be ignored in theoretical power spectra for k>0.3 h/Mpc. It will thus be necessary to improve our understanding of feedback processes in galaxy formation, or at least to constrain them through auxiliary observations, before we can fulfil the goals of upcoming weak lensing surveys.
