In performing cosmological N-body simulations, it is widely appreciated that the growth of structure on the largest scales within a simulation box will be inhibited by the finite size of the simulation volume. Following ideas set forth in Seto (1999), this paper shows that standard (a.k.a. 1-loop) cosmological perturbation theory (SPT) can be used to predict, in an approximate way, the deleterious effect of the box scale on the power spectrum of density fluctuations in simulation volumes. Alternatively, this approach can be used to quickly estimate post facto the effect of the box scale on power spectrum results from existing simulations. In this way SPT can help determine whether larger box sizes or other more-sophisticated methods are needed to achieve a particular level of precision for a given application (e.g. simulations to measure the non-linear evolution of baryon acoustic oscillations). I focus on SPT in this note and show that its predictions differ only by about a factor of two or less from the measured suppression inferred from both powerlaw and $Lambda$CDM $N$-body simulations. It should be possible to improve the accuracy of these predictions through using more-sophisticated perturbation theory models. An appendix compares power spectrum measurements from the powerlaw simulations at outputs where box-scale effects are minimal to perturbation theory models and previously-published fitting functions. These power spectrum measurements are included with this paper to aid efforts to develop new perturbation theory models.
Download