Comparing different mass estimators for a large subsample of the Planck-ESZ clusters

Context. Total mass is arguably the most fundamental property for cosmological studies with galaxy clusters. The individual cluster masses can be obtained with different methods, each with its own biases and limitations. Systematic differences in mass measurements can strongly impact the determination of the hydrostatic bias and of the mass-observable relations, key requirements of many cluster abundance studies.Aims. We investigate the present differences in the mass estimates obtained through independent X-ray, weak-lensing, and dynamical studies using a large subsample of the Planck-ESZ clusters. We also discuss the implications for mass bias analyses.Methods. After assessing the systematic differences in the X-ray-derived masses reported by distinct groups, we examine the mass estimates obtained with independent methods and quantify the differences as the mean ratio 1-b=M-HE/M-WL,M- dyn, where HE refers to hydrostatic masses obtained from X-ray observations, WL refers to the results of weak-lensing measurements, and dyn refers to the mass estimates either from velocity dispersion or from the caustic technique. So defined, the 1-b parameter includes all possible astrophysical, observational, and methodological biases in one single value.Results. Recent X-ray masses reported by independent groups show average differences smaller than similar to 10%, posing a strong limit on the systematics that can be ascribed to the differences in the X-ray analysis when studying the hydrostatic bias. The mean ratio between our X-ray masses and the weak-lensing masses in the LC2-single catalog is 1-b=0.740.06, which corresponds to a mass bias of 26 +/- 6%, a value insufficient to reconcile the Planck cluster abundance and cosmic microwave background results. However, the mean mass ratios inferred from the WL masses of different projects vary by a large amount, with APEX-SZ showing a bias consistent with zero (1-b=1.02 +/- 0.12), LoCuSS and CCCP/MENeaCS showing a significant difference (1-b=0.76 +/- 0.09 and 1-b=0.77 +/- 0.10, respectively), and WtG pointing to the largest deviation (1-b=0.61 +/- 0.12), which would substantially reduce the tension between the Planck results. Because of small differences between our M-Y-X relation and the one used by the Planck collaboration, our X-ray masses are on average 7% lower (4% at the same physical radius) than the Planck masses and can further reduce the required bias. At odds with the WL results, the dynamical mass measurements show better agreement with the X-ray hydrostatic masses, although there are significant differences when relaxed or disturbed clusters are used. However, the comparison is currently limited by the small sample sizes.Conclisions. The systematic differences between total masses obtained with recent independent X-ray analyses are smaller than those found in previous studies. This shifts the focus to WL and dynamical studies for a better convergence of the level of mass bias. However, the ratios obtained using different mass estimators suggest that there are still systematics that are not accounted for in all the techniques used to recover cluster masses. This prevents the determination of firm constraints on the level of hydrostatic mass bias in galaxy clusters.