An analytical model for the non-linear redshift-space power spectrum

We use N-body simulations to test the predictions of redshift distortion in the power spectrum given by the halo model in which the clustering of dark matter particles is considered as a result both of the clustering of dark haloes in space and of the distribution of dark matter particles in each individual dark halo. The predicted redshift distortion depends sensitively on several model parameters in a different way from the real-space power spectrum. An accurate model of the redshift distortion can be constructed if the following properties of the halo population are modelled accurately: the mass function of dark haloes, the velocity dispersion among dark haloes, and the non-linear nature of halo bias on small scales. The model can be readily applied to interpreting the clustering properties and velocity dispersion of different populations of galaxies once a cluster-weighted bias (or equivalently an halo occupation number model) is specified for the galaxies. Some non-trivial bias features observed from redshift surveys of optical galaxies and of IRAS galaxies relative to the standard low-density cold dark matter model can be easily explained in the cluster weighted bias model. The halo model further indicates that a linear bias can be a good approximation only for k less than or equal to 0.1 h Mpc(-1).