ICE-COLA: Towards fast and accurate synthetic galaxy catalogues optimizing a quasi-N-body method

Abstract

Next generation galaxy surveys demand the development of massive ensembles of galaxy mocks to model the observables and their covariances, what is computationally prohibitive using N-body simulations. COmoving Lagrangian Acceleration (COLA) is a novel method designed to make this feasible by following an approximate dynamics but with up to three orders of magnitude speed-ups when compared to an exact N-body. In this paper, we investigate the optimization of the code parameters in the compromise between computational cost and recovered accuracy in observables such as two-point clustering and halo abundance. We benchmark those observables with a state-of-the-art N-body run, the MICE Grand Challenge simulation. We find that using 40 time-steps linearly spaced since zi ~ 20, and a force mesh resolution three times finer than that of the number of particles, yields a matter power spectrum within 1 per cent for k ≲ 1 h Mpc-1 and a halo mass function within 5 per cent of those in the N-body. In turn, the halo bias is accurate within 2 per cent for k ≲ 0.7 h Mpc-1 whereas, in redshift space, the halo monopole and quadrupole are within 4 per cent for k ≲ 0.4 h Mpc-1. These results hold for a broad range in redshift (0 < z < 1) and for all halo mass bins investigated (M > 1012.5 h-1 M⊙). To bring accuracy in clustering to one per cent level we study various methods that re-calibrate halo masses and/or velocities. We thus propose an optimized choice of COLA code parameters as a powerful tool to optimally exploit future galaxy surveys.