A disturbing FABLE of mergers, feedback, turbulence, and mass biases in simulated galaxy clusters

Abstract: The use of galaxy clusters as cosmological probes often relies on understanding the properties and evolution of the intracluster medium (ICM). However, the ICM is a complex plasma, regularly stirred by mergers and feedback, with non-negligible bulk and turbulent motions and a non-thermal pressure component, making it difficult to construct a coherent and comprehensive picture. To this end, we use the FABLE simulations to investigate how the hydrostatic mass bias is affected by mergers, turbulence, and feedback. Following in detail a single, massive cluster we find the bias varies significantly over cosmic time, rarely staying at the average value found at a particular epoch. Variations of the bias at a given radius are contemporaneous with periods where outflows dominate the mass flux, either due to mergers or interestingly, at high redshift, AGN feedback. The $z=0$ ensemble median mass bias in FABLE is $\sim!13$ per cent at $R_\mathrm{500}$ and $\sim!15$ per cent at $R_\mathrm{200}$, but with a large scatter in individual values. In halo central regions, we see an increase in temperature and a decrease in non-thermal pressure support with cosmic time as turbulence thermalises, leading to a reduction in the mass bias within $\sim!0.2 \, R_\mathrm{200}$. When using a fitted pressure profile, instead of the simulation data, to estimate the bias, we find there can be significant differences, particularly at larger radii and higher redshift. We therefore caution over the use of such fits in future work when comparing with the next generation of X-ray and SZ observations.