LoCuSS: A Comparison of Cluster Mass Measurements from XMM-Newton and Subaru - Testing Deviation from Hydrostatic Equilibrium and Non-Thermal Pressure Support
We compare X-ray hydrostatic and weak-lensing mass estimates for a sample of 12 clusters that have been observed with both XMM-Newton and Subaru. At an over-density of Δ=500, we obtain 1-M^{X}/M^{WL}=0.01+/-0.07 for the whole sample. We also divided the sample into undisturbed and disturbed sub-samples based on quantitative X-ray morphologies using asymmetry and fluctuation parameters, obtaining 1-M^{X}/M^{WL}=0.09+/-0.06 and -0.06+/-0.12 for the undisturbed and disturbed clusters, respectively. In addition to non-thermal pressure support, there may be a competing effect associated with adiabatic compression and/or shock heating which leads to overestimate of X-ray hydrostatic masses for disturbed clusters, for example, in the famous merging cluster A1914. Despite the modest statistical significance of the mass discrepancy, on average, in the undisturbed clusters, we detect a clear trend of improving agreement between M^{X} and M^{WL} as a function of increasing over-density, M^{X}/M^{WL}=(0.908+/-0.004)+(0.187+/-0.010) \log_{10}(Δ/500). We also examine the gas mass fractions, f_{gas}=M^{gas}/M^{WL}, finding that they are an increasing function of cluster radius, with no dependence on dynamical state, in agreement with predictions from numerical simulations. Overall, our results demonstrate that XMM-Newton and Subaru are a powerful combination for calibrating systematic uncertainties in cluster mass measurements.