Cosmological implications of the anisotropy of ten galaxy cluster scaling relations

¹ Argelander-Institut für Astronomie, Universität Bonn, Auf dem Hügel 71, 53121 Bonn, Germany
e-mail: kmigkas@astro.uni-bonn.de
² Center for Astrophysics | Harvard & Smithsonian, 60 Garden Street, Cambridge, MA 02138, USA
³ Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR) Projektträger, Joseph-Beuys-Allee 4, 53113 Bonn, Germany
⁴ Institut für Astronomie und Astrophysik, Sand 1, 72076 Tübingen, Germany
⁵ Max Planck Institute for Extraterrestrial Physics, Gießenbachstraße 1, 85748 Garching bei München, Germany
⁶ INAF – Osservatorio Astronomico di Bologna, Via Piero Gobetti, 93/3, 40129 Bologna, BO, Italy

Received: 7 January 2021
Accepted: 22 March 2021

Abstract

The hypothesis that the late Universe is isotropic and homogeneous is adopted by most cosmological studies, including studies of galaxy clusters. The cosmic expansion rate H₀ is thought to be spatially constant, while bulk flows are often presumed to be negligible compared to the Hubble expansion, even at local scales. The effects of bulk flows on the redshift–distance conversion are hence usually ignored. Any deviation from this consensus can strongly bias the results of such studies, and thus the importance of testing these assumptions cannot be understated. Scaling relations of galaxy clusters can be effectively used for this testing. In previous works, we observed strong anisotropies in cluster scaling relations, whose origins remain ambiguous. By measuring many different cluster properties, several scaling relations with different sensitivities can be built. Nearly independent tests of cosmic isotropy and large bulk flows are then feasible. In this work, we make use of up to 570 clusters with measured properties at X-ray, microwave, and infrared wavelengths to construct ten different cluster scaling relations and test the isotropy of the local Universe; to our knowedge, we present five of these scaling relations for the first time. Through rigorous and robust tests, we ensure that our analysis is not prone to generally known systematic biases and X-ray absorption issues. By combining all available information, we detect an apparent 9% spatial variation in the local H₀ between (l, b)∼(280°_−35°^+35°, −15°_−20°^+20°) and the rest of the sky. The observed anisotropy has a nearly dipole form. Using isotropic Monte Carlo simulations, we assess the statistical significance of the anisotropy to be > 5σ. This result could also be attributed to a ∼900 km s⁻¹ bulk flow, which seems to extend out to at least ∼500 Mpc. These two effects will be indistinguishable until more high-z clusters are observed by future all-sky surveys such as eROSITA.