Weak lensing in the Horizon-AGN simulation lightcone

Small-scale baryonic effects

¹ Institut d’Astrophysique de Paris, UMR7095 CNRS & Sorbonne Université, 98bis Bd Arago, 75014 Paris, France
e-mail: celine.gouin@ias.u-psud.fr, gouin@iap.fr
² Institut d’Astrophysique Spatiale, CNRS/Université Paris-Sud, Université Paris-Saclay, Bâtiment 121, 91405 Orsay, France
³ Korea Institute of Advanced Studies (KIAS) 85 Hoegiro, Dongdaemun-gu, Seoul 02455, Republic of Korea
⁴ Institute for Astronomy, University of Edinburgh, Royal Observatory, Blackford Hill, Edinburgh EH9 3HJ, UK
⁵ Sub-department of Astrophysics, University of Oxford, Keble Road, Oxford OX1 3RH, UK
⁶ Université Côte d’Azur, Observatoire de la Côte d’Azur, CNRS, Laboratoire Lagrange, Bd de l’observatoire, 06304 Nice, France

Received: 5 September 2018
Accepted: 26 April 2019

Abstract

Context. Accurate model predictions including the physics of baryons are required to make the most of the upcoming large cosmological surveys devoted to gravitational lensing. The advent of hydrodynamical cosmological simulations enables such predictions on sufficiently sizeable volumes.

Aims. Lensing quantities (deflection, shear, convergence) and their statistics (convergence power spectrum, shear correlation functions, galaxy-galaxy lensing) are computed in the past lightcone built in the Horizon-AGN hydrodynamical cosmological simulation, which implements our best knowledge on baryonic physics at the galaxy scale in order to mimic galaxy populations over cosmic time.

Methods. Lensing quantities are generated over a one square degree field of view by performing multiple-lens plane ray-tracing through the lightcone, taking full advantage of the 1 kpc resolution and splitting the line of sight over 500 planes all the way to redshift z ∼ 7. Two methods are explored (standard projection of particles with adaptive smoothing, and integration of the acceleration field) to ensure a good implementation. The focus is on small scales where baryons matter most.

Results. Standard cosmic shear statistics are affected at the 10% level by the baryonic component for angular scales below a few arcminutes. The galaxy-galaxy lensing signal, or galaxy-shear correlation function, is consistent with measurements for the redshift z ∼ 0.5 massive galaxy population. At higher redshift z ≳ 1, the effect of magnification bias on this correlation is relevant for separations greater than 1 Mpc.

Conclusions. This work is pivotal for all current and upcoming weak-lensing surveys and represents a first step towards building a full end-to-end generation of lensed mock images from large cosmological hydrodynamical simulations.