2019A&A...621A..39E

Aims. We present the reconstruction of hydrostatic mass profiles in 13 X-ray luminous galaxy clusters that have been mapped in their X-ray and Sunyaev-Zeldovich (SZ) signals out to R₂₀₀ for the XMM-Newton Cluster Outskirts Project (X-COP).
Methods. Using profiles of the gas temperature, density, and pressure that have been spatially resolved out to median values of 0.9R₅₀₀, 1.8R₅₀₀, and 2.3R₅₀₀, respectively, we are able to recover the hydrostatic gravitating mass profile with several methods and using different mass models.
Results. The hydrostatic masses are recovered with a relative (statistical) median error of 3% at R₅₀₀ and 6% at R₂₀₀. By using several different methods to solve the equation of the hydrostatic equilibrium, we evaluate some of the systematic uncertainties to be of the order of 5% at both R₅₀₀ and R₂₀₀. A Navarro-Frenk-White profile provides the best-fit in 9 cases out of 13; the remaining 4 cases do not show a statistically significant tension with it. The distribution of the mass concentration follows the correlations with the total mass predicted from numerical simulations with a scatter of 0.18 dex, with an intrinsic scatter on the hydrostatic masses of 0.15 dex. We compare them with the estimates of the total gravitational mass obtained through X-ray scaling relations applied to Y_X, gas fraction, and Y_SZ, and from weak lensing and galaxy dynamics techniques, and measure a substantial agreement with the results from scaling laws, from WL at both R₅₀₀ and R₂₀₀ (with differences below 15%), from cluster velocity dispersions. Instead, we find a significant tension with the caustic masses that tend to underestimate the hydrostatic masses by 40% at R₂₀₀. We also compare these measurements with predictions from alternative models to the cold dark matter, like the emergent gravity and MOND scenarios, confirming that the latter underestimates hydrostatic masses by 40% at R₁₀₀₀, with a decreasing tension as the radius increases, and reaches ∼15% at R₂₀₀, whereas the former reproduces M₅₀₀ within 10%, but overestimates M₂₀₀ by about 20%.
Conclusions. The unprecedented accuracy of these hydrostatic mass profiles out to R₂₀₀ allows us to assess the level of systematic errors in the hydrostatic mass reconstruction method, to evaluate the intrinsic scatter in the NFW c-M relation, and to robustly quantify differences among different mass models, different mass proxies, and different gravity scenarios.