Testing the Collisionless Nature of Dark Matter with the Radial Acceleration Relation in Galaxy Clusters

Abstract

The radial acceleration relation (RAR) represents a tight empirical relation between the inferred total and baryonic centripetal accelerations, g tot = GM tot (< r)/r 2 and g bar = GM bar (< r)/r 2 , observed in galaxies and galaxy clusters. The tight correlation between these two quantities can provide insight into the nature of dark matter. Here we use BAHAMAS, a state-of-the-art suite of cosmological hydrodynamical simulations, to characterize the RAR in cluster-scale halos for both cold and collisionless dark matter (CDM) and selfinteracting dark matter (SIDM) models. SIDM halos generally have reduced central dark matter densities, which reduces the total acceleration in the central region when compared with CDM. We compare the RARs in galaxy clusters simulated with different dark matter models to the RAR inferred from CLASH observations. Our comparison shows that the cluster-scale RAR in the CDM model provides an excellent match to the CLASH RAR obtained by Tian et al. including the high-acceleration regime probed by the brightest cluster galaxies (BCGs). By contrast, models with a larger SIDM cross-section yield increasingly poorer matches to the CLASH RAR. Excluding the BCG regions results in a weaker but still competitive constraint on the SIDM cross-section. Using the RAR data outside the central r < 100 kpc region, an SIDM model with σ/m = 0.3 cm 2 g −1 is disfavored at the 3.8σ level with respect to the CDM model. This study demonstrates the power of the clusterscale RAR for testing the collisionless nature of dark matter.

X-ray analysis of JWST’s first galaxy cluster lens SMACS J0723.3−7327

“…The discrepancy is so large that it cannot be explained by either the possible biases or systematics in our analysis, such as the uncertainty in stellar mass and hydrostatic bias, or the 0.12 dex scatter in the McGaugh et al (2016) relation. This is in broad agreement with the results for X-COP clusters (Eckert et al 2022) and BAHAMAS simulation (Tam et al 2022). Due to the removal of the core, we have no constraints on g bar at high acceleration, where Eckert et al (2022) observed that the RAR of X-COP clusters eventually catched up with the McGaugh et al (2016) relation for galaxies.…”

“…Not only is the normalization of acceleration higher than what is found for galaxies (e.g., Tian et al 2020), but also the overall shape of the observed cluster RAR clearly deviates from the model in Eq. 4, implying a completely different RAR between clusters and galaxies (e.g., Eckert et al 2022;Tam et al 2022).…”