The isotropy of the Universe as seen through galaxy clusters

Abstract

The cosmological principle (CP) postulates that the Universe is isotropic and homogeneous. This implies that the Universe has the same expansion rate toward every direction and holds similar amounts of matter when averaged over sufficiently large scales. The CP is a fundamental building block of the standard cosmological model and is broadly taken as an axiom in extragalactic astronomy. However, its observational foundation remains uncertain. Multiple studies that tested the CP have been conducted in the past, frequently returning controversial results. Since each methodology suffers from certain limitations and potential biases, it is crucial to develop several independent tests to scrutinize the CP further. Galaxy clusters can be an excellent probe of cosmic isotropy. They are the largest, gravitationally bound systems in the Universe, and they contain up to thousands of galaxies and vast amounts of dark matter. Their intracluster medium (ICM) is filled with hot ionized plasma that strongly emits X-ray radiation. This work aims to apply a novel, powerful test for the CP using galaxy clusters. Specifically, I use their so-called scaling relations (SCs) to probe the isotropy of the local Universe. This work consists of two projects. <br /> In the first project, I focus on the tight correlation between the X-ray luminosity L_X and the temperature T of galaxy clusters. To estimate L_X from the measured X-ray flux, one needs to assume the values of the cosmological parameters, such as the expansion rate of the Universe, H₀. On the other hand, the T measurement is nearly free of cosmological assumptions. I exploit this property to draw conclusions about cosmic isotropy by investigating the directional behavior of the L_X - T relation. If H₀ is spatially constant and no previously undetected, X-ray absorbing gas and dust clouds exist, then L_X - T should be statistically similar toward all directions. To constrain the L_X - T relation I use a homogeneously selected galaxy cluster sample with 313 objects. Their temperatures, metallicities, and redshifts were measured, using observations from the Chandra and XMM-Newton telescopes. Further corrections in the publicly available L_X values were also performed to better account for the X-ray absorption, point source contamination, and merging clusters. By scanning the extragalactic sky, it is shown that L_X - T manifests a strongly anisotropic behavior at a &#8819; 4&sigma; confidence level. The apparently preferred axis lies toward the Galactic coordinates (l, b) ~ (280°, -20°). A large variety of potential systematic biases that could result in artificially observed anisotropies is inspected, including X-ray absorption effects. None of these tests reveal any significant bias in the anisotropy detection. To further check the consistency of the results, I repeat the analysis with two other, completely independent cluster samples. Both samples exhibit a similar L_X - T anisotropy. When all three samples are combined, they result in 842 individual clusters. Their joint analysis further amplifies the statistical significance of the observed anisotropy, boosting it to a ~ 5&sigma; level. The most anisotropic direction is (l, b) ~ (303°, -27°) with a H₀ variation of ~ 15% across the sky, in agreement with several past studies, conducted with different probes. A conclusive answer on the origin of this anisotropy, however, could not be given yet. <br /> In the second project, I attempt to decisively identify the origin of the apparent anisotropies. To do so, I use three more cluster properties; the total integrated Compton parameter Y_sz, the infrared luminosity of the brightest cluster galaxy, and the X-ray half-light radius of the clusters. All three are nearly free of absorption issues. This allows for the construction of 10 multiwavelength SCs, through which essential information on the nature of the cluster anisotropies can be obtained. Five of these SCs are presented in the literature for the first time. I make use of nearly 570 individual clusters and I employ advanced statistical tools to robustly assess the statistical significance of the results. Low-scatter SCs that are sensitive only to unknown X-ray absorption issues (e.g., L_X - Y_sz) and not to H₀ spatial variations or bulk flows, do not show any significant anisotropies. Therefore, the scenario of excess X-ray absorption biasing the results is strongly disfavored. Next, I exploit SCs that only trace cosmological phenomena and do not suffer from absorption issues (e.g., Ysz - T). Nearly the same anisotropies are detected as in the L_X- T relation, with a higher statistical significance due to the reduced scatter of the new SCs. This result supports the notion that the observed anisotropies are of cosmological origin. When all available information is combined, a 9% anisotropy in H₀ is robustly detected toward (l, b) ~ (280°, -15°), rejecting the CP at a 5.4&sigma; confidence level. Comparisons with isotropic Monte Carlo simulations confirm the high statistical significance of the detected anisotropy. Additional tests show that numerous, generally known biases, do not alleviate the observed tension. An alternative explanation to the H₀ anisotropy would be the existence of a ~ 900 km/s bulk flow (coherent peculiar motion of clusters toward a specific direction), extending out to ?? 500 Mpc. This would also contradict the CP, since such a motion would require a large-scale matter inhomogeneity, inconsistent with the concordance cosmology predictions. <br /> Since a cornerstone of concordance cosmology is in doubt, more tests are needed to provide a definite answer. Forthcoming morphological parameters of the used clusters will help reduce the scatter of the relations and provide more precise results. The upcoming eROSITA All-Sky Survey will also allow for anisotropy studies with many more clusters at greater distances. This is a necessity in order to distinguish between a cosmological anisotropy and a bulk flow.