Abstract
Context: Analyzing the large-scale structure (LSS) with galaxy surveys
demands accurate structure formation models. Such models should ideally be fast
and have a clear theoretical framework to rapidly scan a variety of
cosmological parameter spaces without requiring large training data sets.
Aims: This study aims to extend Lagrangian perturbation theory (LPT),
including viscosity and vorticity, to reproduce the cosmic evolution from dark
matter N-body calculations at the field level.
Methods: We extend Augmented LPT (ALPT) to an Eulerian framework, dubbed
eALPT. This enables modelling the stress tensor, with this introducing
vorticity. To compensate that ALPT assumes curl-free fields, a fraction of the
vorticity, emerging after each Eulerian transformation, is added to the
subsequent timestep. The model has three free parameters apart from the choice
of cosmology, redshift snapshots, cosmic volume, and the number of
particles-cells.
Results: We find that the cross-correlation of the dark matter distribution
as compared to N-body solvers increases at k = 1 h Mpc$^-1$ from ~55% with
the Zel'dovich approximation (~70% with ALPT); to ~96 and 97% with eALPT, and
power spectra within percentage accuracy up to k~ 0.3 and 0.7 h Mpc$^-1$,
using three and five steps, respectively.
Users
Please
log in to take part in the discussion (add own reviews or comments).