Modeling the Lyman-alpha Forest in Collisionless Simulations

TL;DR

The paper introduces IMS, a new method that uses small high-resolution hydrodynamic simulations to accurately model the Lyman-alpha forest in large collisionless N-body simulations, enabling realistic large-volume IGM studies.

Contribution

IMS is a novel iterative mapping technique that accurately reproduces Lyman-alpha forest statistics in collisionless simulations, overcoming resolution limitations of hydrodynamic methods.

Findings

IMS perfectly reproduces the PDF and power spectrum of the Lyman-alpha forest.

IMS achieves 5-20% accuracy in the 3D flux power spectrum.

Gaussian smoothing is less accurate (20-80%) compared to IMS for large-volume simulations.

Abstract

Cosmological hydrodynamic simulations can accurately predict the properties of the intergalactic medium (IGM), but only under the condition of retaining high spatial resolution necessary to resolve density fluctuations in the IGM. This resolution constraint prohibits simulating large volumes, such as those probed by BOSS and future surveys, like DESI and 4MOST. To overcome this limitation, we present Iteratively Matched Statistics (IMS), a novel method to accurately model the Lyman-alpha forest with collisionless N-body simulations, where the relevant density fluctuations are unresolved. We use a small-box, high-resolution hydrodynamic simulation to obtain the probability distribution function (PDF) and the power spectrum of the real-space Lyman-alpha forest flux. These two statistics are iteratively mapped onto a pseudo-flux field of an N-body simulation, which we construct from the matter density. We demonstrate that our method can perfectly reproduce line-of-sight observables, such as the PDF and power spectrum, and accurately reproduce the 3D flux power spectrum (5-20%). We quantify the performance of the commonly used Gaussian smoothing technique and show that it has significantly lower accuracy (20-80%), especially for N-body simulations with achievable mean inter-particle separations in large-volume simulations. In addition, we show that IMS produces reasonable and smooth spectra, making it a powerful tool for modeling the IGM in large cosmological volumes and for producing realistic "mock" skies for Lyman-alpha forest surveys.