Abstract
In this paper, we introduce Physically motivated, Internally Consistent Analysis of Cluster Scaling (PICACS), a detailed model of scaling relations between galaxy cluster masses and their observable properties. This model can be used to constrain simultaneously the form, scatter (including its covariance) and evolution of the scaling relations, as well as the masses of the individual clusters. In this framework, scaling relations between observables (such as that between X-ray luminosity and temperature) are modelled explicitly in terms of the fundamental mass-observable scaling relations, and so are fully constrained without being fit directly. We apply the PICACS model to two observational data sets, and show that it performs as well as traditional regression methods for simply measuring individual scaling relation parameters, but reveals additional information on the processes that shape the relations while providing self-consistent mass constraints. Our analysis suggests that the observed combination of slopes of the scaling relations can be described by a deficit of gas in low-mass clusters that is compensated for by elevated gas temperatures, such that the total thermal energy of the gas in a cluster of given mass remains close to self-similar expectations. This is interpreted as the result of AGN feedback removing low entropy gas from low-mass systems, while heating the remaining gas. We deconstruct the luminosity-temperature (L-T) relation and show that its steepening compared to self-similar expectations can be explained solely by this combination of gas depletion and heating in low-mass systems, without any additional contribution from a mass dependence of the gas structure. Finally, we demonstrate that a self-consistent analysis of the scaling relations leads to an expectation of self-similar evolution of the L-T relation that is significantly weaker than is commonly assumed.
Author
Maughan, B. J.
Journal
Monthly Notices of the Royal Astronomical Society
Paper Publication Date
January 2014
Paper Type
Astrostatistics