The signature of major mergers on the hydrostatic mass bias of galaxy clusters

TL;DR

The paper addresses how cluster mergers affect the hydrostatic mass bias, using a high-resolution AMR cosmological simulation to track $b=(M_{HE}-M_{true})/M_{true}$ across radii up to $R_{vir}$ and through major mergers in $0\le z \le 1.5$. It demonstrates a characteristic dip–peak–relaxation pattern in $b$ during mergers, dominated by ICM density-field distortions rather than thermodynamics, and finds only a weak, but present, link to the dynamical state at $z=0$. The authors propose simple time-dependent (sinusoidal) and stage-based prescriptions to calibrate observational mass biases in dynamically active clusters, and show that outer regions experience stronger, longer-lasting perturbations while inner regions relax earlier, leaving a residual non-thermal-pressure-driven bias. These results imply that accounting for the merger stage and the dynamical history is crucial for accurate cluster mass estimates in cosmological analyses and motivate incorporating assembly-state diagnostics into mass-calibration pipelines. The work highlights that non-thermal motions and density-morphology changes in the ICM persist long after mergers, challenging the assumption of hydrostatic equilibrium in precision cosmology.

Abstract

While galaxy cluster masses are fundamental cosmological observables, estimates based on intra-cluster medium observations rely on hydrostatic equilibrium, introducing a systematic bias. We investigate how mergers drive the time evolution of this hydrostatic mass bias, identifying the dominant physical mechanisms and their dependence on dynamical state and merger history. Using a high-resolution AMR Eulerian+$N$-body cosmological simulation, we analyse a sample of cluster mergers within $1.5 \leq z \leq 0$, comparing true and hydrostatic masses derived from gas density and temperature profiles, and tracing their evolution. At $z=0$, the hydrostatic mass bias shows a mild correlation with dynamical state. During major mergers, the bias follows a characteristic trend: a sharp negative dip around the merger time, a transient positive peak, and a gradual return to pre-merger levels. This behaviour is primarily driven by morphological and dynamical reconfigurations of the gas density within the ICM, while thermodynamical processes play a secondary role. The pattern shows no strong dependence on secondary parameters, such as mass ratio or impact parameter, but it can be fitted to a simple time-dependent functional form. This trend is present at radii $r\le R_{\mathrm{vir}}$, although with reduced amplitude and shorter timescales as the radius decreases. Hydrostatic mass bias is closely linked, albeit in a non-trivial way, with the merging history of galaxy clusters. We find that the bias values are weakly correlated with the dynamical state of clusters. Nevertheless, our results give a robust estimation of the hydrostatic mass bias values in the pre-merger, merging, and post-merger phases. These findings highlight the importance of delving deeper into the observational assessment of cluster assembly state in order to improve mass estimations for cosmological analyses.

Figures (11)

• Figure 1: Outcome of varying the orbital parameters on the merger mock simulations. Top panel: Time evolution of the hydrostatic mass bias, evaluated at $R_{\mathrm{vir}}$, for different values of the mass ratio $q$. The impact parameter is fixed to $p=0$, and the horizontal axis represents the time elapsed since the beginning of the test. The vertical dashed line represents the moment at which the infalling halo centre enters the virial volume of the host. Bottom panel: Same evolution as the top figure but with a fixed mass ratio of $q=0.3$ and a varying impact parameter.
• Figure 2: Hydrostatic mass bias evaluated at $R_\mathrm{vir}$ as a function of the combined dynamical state indicator, $\chi$, for all the clusters in our sample at $z=0$. Since the indicator has a continuous range of values, we set decile bins (each of a different color) with respect to the relaxation of galaxy clusters.
• Figure 3: Average hydrostatic mass bias per dynamical state ($\chi$) bin, corresponding to the subsamples depicted in Fig \ref{['relax bands z0']}. The error bars show the standard deviation within each bin, while the shadowed region corresponds to $1\sigma$ error on the linear fit over the average mass bias values (dashed line).
• Figure 4: Prototypical examples of hydrostatic mass bias evolution in relation to cluster assembly, for three particular clusters from our sample. Left panels: Hydrostatic mass bias evaluated at the virial radius, $b(R_\mathrm{vir})$ (purple), and of the dynamical state indicator, $\chi$ (orange). Right panels: Logarithmic mass growth history for the same period of time for the three clusters shown in the left panel. The vertical dashed lines indicate the time at which a major merger took place, in case there was any.
• Figure 5: Standardised evolution of the hydrostatic mass bias, evaluated at several radii: $0.25R_{\mathrm{vir}}$, $0.5R_{\mathrm{vir}}$, $0.75R_{\mathrm{vir}}$ and $R_{\mathrm{vir}}$