E10 Computational Cosmology

Topic

The core of this model is a cosmological N-body simulation that tracks the trajectories of a vast number of dark matter particles under the influence of their own gravity.

To keep the simulation computationally manageable, it combines two complementary techniques based on the range of the force being calculated: a tree algorithm—suitable for short-range forces between nearby particles—and a grid-based method that solves for gravity using Fourier techniques—suitable for long-range forces. This combination allows the simulation to track both close-range and long-range interactions without excessive computational cost.

When this simulation incorporates hydrodynamics, it includes not only dark matter but also cosmic gas, accounting for radiative cooling, the star formation enabled by that cooling, and the feedback injected back into the gas by supernovae and active galactic nuclei. Adding these physical processes shifts the simulation from describing gravity alone to also describing the accompanying baryonic physics.

Accurately reproducing the statistics of baryon acoustic oscillations requires running the cosmological simulation within boxes on the order of a thousand megaparsecs in size, as smaller boxes fail to capture that characteristic scale. The synthetic universes resulting from this process serve, in turn, to calibrate observational estimators and to quantify the bias with which visible galaxies imperfectly trace the true distribution of matter. Thus, from the N-body cosmological simulation that opens this document to the tracer bias ultimately quantified—and encompassing everything from the combination of algorithms based on force range and the inclusion of baryonic physics to the minimum box size required by baryon statistics—the cosmological simulation enables an entire universe, reduced to particles and equations, to serve as a tool for calibrating how observations of the actual universe are interpreted.

ID:4112