Baryonic effects on weak-lensing two-point statistics and its cosmological implications

TL;DR

This paper develops an analytical model of baryonic effects on weak-lensing statistics, demonstrating that accounting for baryons is crucial for accurate cosmological parameter estimation at small scales.

Contribution

An extended Halo Model incorporating baryonic physics with a single free parameter, validated against simulations, and applied to cosmological constraints from weak lensing.

Findings

Baryonic effects are negligible up to B5=4000 but significant at B5=8000.

Neglecting baryons can bias cosmological parameters by up to 10C3.

Including the baryonic parameter reduces bias and allows precise measurement of baryonic effects.

Abstract

We develop an extension of \textit{the Halo Model} that describes analytically the corrections to the matter power spectrum due to the physics of baryons. We extend these corrections to the weak-lensing shear angular power spectrum. Within each halo, our baryonic model accounts for: 1) a central galaxy, the major stellar component whose properties are derived from abundance matching techniques; 2) a hot plasma in hydrostatic equilibrium and 3) an adiabatically-contracted dark matter component. This analytic approach allows us to compare our model to the dark-matter-only case. Our basic assumptions are tested against the hydrodynamical simulations of Martizzi et. al. (2014), with which a remarkable agreement is found. Our baryonic model has only one free parameter, $M_{\rm crit}$, the critical halo mass that marks the transition between feedback-dominated halos, mostly devoid of gas, and gas rich halos, in which AGN feedback effects become weaker. We explore the entire cosmological parameter space, using the angular power spectrum in three redshift bins as the observable, assuming a Euclid-like survey. We derive the corresponding constraints on the cosmological parameters, as well as the possible bias introduced by neglecting the effects of baryonic physics. We find that, up to $\ell_{max}$=4000, baryonic physics plays very little role in the cosmological parameters estimation. However, if one goes up to $\ell_{max}$=8000, the marginalized errors on the cosmological parameters can be significantly reduced, but neglecting baryonic physics can lead to bias in the recovered cosmological parameters up to 10$\sigma$. These biases are removed if one takes into account the main baryonic parameter, $M_{\rm crit}$, which can also be determined up to 1-2\%, along with the other cosmological parameters.