Observing relativistic features in large-scale structure surveys -- II: Doppler magnification in an ensemble of relativistic simulations

Autor: Coates, Louis, Adamek, Julian, Bull, Philip, Guandalin, Caroline, Clarkson, Chris
Rok vydání: 2020
Předmět:
Zdroj: MNRAS,504(3),2021,3534-3543
Druh dokumentu: Working Paper
DOI: 10.1093/mnras/stab1076
Popis: The standard cosmological model is inherently relativistic, and yet a wide range of cosmological observations can be predicted accurately from essentially Newtonian theory. This is not the case on `ultra-large' distance scales, around the cosmic horizon size, however, where relativistic effects can no longer be neglected. In this paper, we present a novel suite of 53 fully relativistic simulations generated using the gevolution code, each covering the full sky out to $z \approx$ 0.85, and approximately 1930 square degrees out to $z \approx$ 3.55. These include a relativistic treatment of massive neutrinos, as well as the gravitational potential that can be used to exactly calculate observables on the past light cone. The simulations are divided into two sets, the first being a set of 39 simulations of the same fiducial cosmology (based on the Euclid Flagship 2 cosmology) with different realisations of the initial conditions, and the second which fixes the initial conditions, but varies each of seven cosmological parameters in turn. Taken together, these simulations allow us to perform statistical studies and calculate derivatives of any relativistic observable with respect to cosmological parameters. As an example application, we compute the cross-correlation between the Doppler magnification term in the convergence, $\kappa_v$, and the CDM+baryon density contrast, $\delta_{\rm cb}$, which arises only in a (special) relativistic treatment. We are able to accurately recover this term as predicted by relativistic perturbation theory, and study its sample variance and derivatives with respect to cosmological parameters.
Comment: 9 pages, 6 figures, corresponds to published version
Databáze: arXiv