Modeling features in the redshift-space halo power spectrum with perturbation theory

We study the ability of perturbative models with effective field theory contributions and infra-red resummation to model the redshift space clustering of biased tracers in models where the linear power spectrum has "features" -- either imprinted during inflation or induced by non-standard expansion histories. We show that both Eulerian and Lagrangian perturbation theory are capable of reproducing the Fourier space two-point functions of halos up to the non-linear scale from a suite of $4096^3$ particle N-body simulations. This is the first demonstration that perturbative models can accurately fit the redshift-space clustering of biased tracers in N-body simulations of such theories. By comparing different theoretical models and IR resummation schemes we assess the current theoretical uncertainty in predicting power spectra for models with features. Our results suggest that future surveys will be able to detect or tightly constrain features in the primordial spectrum below the one percent level across a wide range of scales.