Redshift-space distortions of galaxies, clusters and AGN: testing how the accuracy of growth rate measurements depends on scales and sample selections

Redshift-space clustering anisotropies caused by cosmic peculiar velocities provide a powerful probe to test the gravity theory on large scales. However, to extract unbiased physical constraints, the clustering pattern has to be modelled accurately, taking into account the effects of non-linear dynamics at small scales, and properly describing the link between the selected cosmic tracers and the underlying dark matter field. We use a large hydrodynamic simulation to investigate how the systematic error on the linear growth rate, $f$, caused by model uncertainties, depends on sample selections and comoving scales. Specifically, we measure the redshift-space two-point correlation function of mock samples of galaxies, galaxy clusters and Active Galactic Nuclei, extracted from the Magneticum simulation, in the redshift range 0.2 < z < 2, and adopting different sample selections. We estimate $f\sigma_8$ by modelling both the monopole and the full two-dimensional anisotropic clustering, using the dispersion model. We find that the systematic error on $f\sigma_8$ depends significantly on the range of scales considered for the fit. If the latter is kept fixed, the error depends on both redshift and sample selection, due to the scale-dependent impact of non-linearities, if not properly modelled. On the other hand, we show that it is possible to get unbiased constraints on $f\sigma_8$ provided that the analysis is restricted to a proper range of scales, that depends non trivially on the properties of the sample. This can have a strong impact on multiple tracers analyses, and when combining catalogues selected at different redshifts.