The relationship between substructure in 2D X-ray surface brightness images and weak lensing mass maps of galaxy clusters: A simulation study

Recent X-ray and weak-lensing observations of galaxy clusters have revealed that the hot gas does not always directly trace the dark matter within these systems. Such configurations are extremely interesting. They offer a new vista onto the complex interplay between gravity and baryonic physics, and may even be used as indicators of the clusters’ dynamical state. In this paper, we undertake a study to determine what insight can be reliably gleaned from the comparison of the X-ray and the weak lensing mass maps of galaxy clusters. We do this by investigating the 2D substructure within three high-resolution cosmological simulations of galaxy clusters. Our main results focus on non-radiative gas dynamics, but we also consider the effects of radiative cooling at high redshift. For our analysis, we use a novel approach, based on unsharp-masking, to identify substructures in 2D surface mass density and X-ray surface brightness maps. At full resolution (∼ 15h kpc), this technique is capable of identifying almost all self-bound dark matter subhaloes with M > 10hM⊙. We also report a correlation between the mass of a subhalo and the area of its corresponding 2D detection; such a correlation, once calibrated, could provide a useful estimator for substructure mass. Comparing our 2D mass and X-ray substructures, we find a surprising number of cases where the matching fails: around one third of galaxy-sized substructures have no X-ray counterpart. Some interesting cases are also found at larger masses, in particular the cores of merging clusters where the situation can be complex. Finally, we degrade our mass maps to what is currently achievable with weak-lensing observations (∼ 100hkpc at z = 0.2). While the completeness mass limit increases by around an order of magnitude, a mass-area correlation remains. Our paper clearly demonstrates that the next generation of lensing surveys should start to reveal a wealth of information on cluster substructure.