X-ray mass proxies from hydrodynamic simulations of galaxy clusters - I

Using extended sets of cosmological hydrodynamical simulations of galaxy clusters, we present a detailed study of scaling relations between total cluster mass and three mass proxies based on X-ray observable quantities: temperature of the intracluster medium (ICM), gas mass and the product of the two, Y-X = MgasT. Our analysis is based on two sets of high-resolution hydrodynamical simulations performed with the TreePM-SPH GADGET code. The first set includes about 140 clusters with masses above 5 x 10(13) h(-1) M-circle dot, with 30 such clusters having mass above 10(15) h(-1) M-circle dot. All such clusters have been simulated in two flavours, both with non-radiative physics and including cooling, star formation, chemical enrichment and the effect of supernova feedback triggering galactic ejecta. The extensive statistics offered by this set of simulated clusters is used to quantify the robustness of the scaling relations between mass proxies and total mass, to determine their redshift evolution and to calibrate their intrinsic scatter and its distribution. Furthermore, we use a smaller set of clusters including 18 haloes with masses above 5 x 10(13) h(-1) M-circle dot, four of which are more massive than 10(15) h(-1) M-circle dot, to test the robustness of mass proxies against change in the physical processes that are included in the simulations to describe the evolution of the intracluster medium. Each cluster is simulated in seven different flavours to study the effects of (i) thermal conduction, (ii) artificial viscosity, (iii) cooling and star formation, (iv) galactic winds and (v) active galactic nucleus (AGN) feedback. As a general result, we find the M-Y-X scaling relation to be the least sensitive to variations in the ICM physics, its slope and redshift evolution always being very close to the predictions of the self-similar model. As regards the scatter around the best-fitting relations, its distribution is always close to a log-normal one. M-gas is the mass proxy with the smallest scatter in mass, with values of sigma(1n) (M) similar or equal to 0.04-0.06 depending on the physics included in the simulation and with a mild dependence on redshift. In terms of the mass-temperature relation, it is the one with the largest scatter, with sigma(1n) (M) greater than or similar to 0.1 at z = 0 increasing to greater than or similar to 0.15 at z = 1. The intrinsic scatter in the M-Y-X relation is slightly larger than that in the M-M-gas relation, with sigma(1n) (M) similar or equal to 0.06 at z = 0 and 0.08 at z = 1. These results confirm that both Y-X and M-gas mass proxies are well suited for cosmological applications in future large X-ray surveys. As a word of caution, we point out that the analysis presented in this paper does not include the observational effects expected when measuring temperature by fitting X-ray spectra and gas mass from X-ray surface-brightness profiles. A detailed assessment of such effects will be the subject of a forthcoming paper.